Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
listlengths 1
91
| keywords
listlengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
Many disease-susceptible SNPs exhibit significant disparity in ancestral and derived allele frequencies across worldwide populations . While previous studies have examined population differentiation of alleles at specific SNPs , global ethnic patterns of ensembles of disease risk alleles across human diseases are unexamined . To examine these patterns , we manually curated ethnic disease association data from 5 , 065 papers on human genetic studies representing 1 , 495 diseases , recording the precise risk alleles and their measured population frequencies and estimated effect sizes . We systematically compared the population frequencies of cross-ethnic risk alleles for each disease across 1 , 397 individuals from 11 HapMap populations , 1 , 064 individuals from 53 HGDP populations , and 49 individuals with whole-genome sequences from 10 populations . Type 2 diabetes ( T2D ) demonstrated extreme directional differentiation of risk allele frequencies across human populations , compared with null distributions of European-frequency matched control genomic alleles and risk alleles for other diseases . Most T2D risk alleles share a consistent pattern of decreasing frequencies along human migration into East Asia . Furthermore , we show that these patterns contribute to disparities in predicted genetic risk across 1 , 397 HapMap individuals , T2D genetic risk being consistently higher for individuals in the African populations and lower in the Asian populations , irrespective of the ethnicity considered in the initial discovery of risk alleles . We observed a similar pattern in the distribution of T2D Genetic Risk Scores , which are associated with an increased risk of developing diabetes in the Diabetes Prevention Program cohort , for the same individuals . This disparity may be attributable to the promotion of energy storage and usage appropriate to environments and inconsistent energy intake . Our results indicate that the differential frequencies of T2D risk alleles may contribute to the observed disparity in T2D incidence rates across ethnic populations .
The global rise in incidence of type 2 diabetes ( T2D ) has been called a pandemic , and is now commonly identified as a major international health concern [1] . Although environmental factors play a substantial role in the etiology of T2D , genetic susceptibility has been established as a key component of risk [2] , [3] . However , the potential disparities in genetic risks for T2D and other major diseases across different ethnic groups and subpopulations are poorly characterized . In this study , we evaluate the hypothesis that disparities in global patterns of T2D risk allele frequencies contribute to disparities in genetic risk of T2D across diverse global populations , compared with genomic background and other diseases . We evaluate this hypothesis using three catalogs of genetic variation sampled from diverse ethnic populations [4] , [5] and published associations between genetic variants and disease traits [6] . Specifically , we include 1 , 397 individuals from 11 populations represented in the HapMap project [4] and to follow-up findings , we added 1 , 064 individuals from 53 indigenous populations across the world in the Human Genome Diversity Panel ( HGDP ) [7] . Additionally , we recently sequenced and publically released the whole genomes of 49 individuals from 10 diverse populations . Disease associated variants were collated from the thousands of genetic variants associated with increased disease risk through thousands of genome-wide and candidate association studies [8] , leveraging our previous efforts to catalog , summarize , and integrate these risk variants across hundreds of human diseases [6] . Enabled by the catalogs described above , a recent analysis demonstrated wide variations in the frequencies of some disease-susceptible risk alleles across 11 HapMap subpopulations [9] , but with the caveat that many of the well-known disease SNPs were discovered in population of European ancestry [10] . To further explore the association between individual genetic risks in subpopulations , we propose conducting a systematic evaluation of the differentiation of the risk allele frequencies ( RAFs ) and individual genetic risks across population groups . The results aim to further elucidate the role of ethnicity in clinical diagnosis of genetic disease risk and also provide guidance towards discovering additional disease-susceptible genetic variants from diverse population groups . We are focusing on T2D and its comparison with other diseases because of its pandemic prevalence and a striking feature of extreme population differentiation we find in this study . Inconsistent results have been previously reported on the population differentiation of RAF and genetic risk of T2D [11]–[15] . However , to our knowledge , most of the studies have been focused on individual SNPs , and none of them considered the consistency of directions of population differentiation across all T2D-susceptible risk alleles . Furthermore , none of these previous studies have systematically compared the population differentiation between T2D and other diseases . In this study , we systematically calculate the directional population differentiation of RAFs across cross-ethnic T2D risk alleles using genotyping data from diverse population groups in both HapMap and HGDP , and evaluate the statistical significance of differentiation against genomic background and other diseases . We then calculate the predicted genetic risk ( PGR ) of 1 , 397 individuals from 11 HapMap populations , and identify diseases with significant differentiation across diverse population groups .
We manually curated thousands of papers covering human disease genetics and identified independent risk alleles increasing the risk of Type 2 diabetes ( T2D ) across diverse population groups using the following procedure . Specifically , we curated 50 , 730 SNPs associating with 1 , 495 human diseases from 5 , 065 papers , and built a quantitative human disease-SNP association database , called VARiant-INforming MEDicine ( Varimed ) [6] . Of these , 8 , 377 SNPs for 437 diseases had been initially reported with a p-value<10−6 , and only these SNPs were used in this study . From 120 T2D SNPs reported in 295 studies in 132 papers with p-value<10−6 , we identified cross-ethnic SNPs as being replicated in five or more different populations . We then removed SNPs that are in linkage disequilibrium ( R2>0 . 7 ) in Caucasian population in HapMap ( merged HapMap 2+3 ) [4] . Unless otherwise specified , HapMap were referred to this merged version throughout this study . In total , we identified 12 independent cross-ethnic risk alleles increasing T2D risk , including one in SLC30A8 , IGF2BP2 , KCNJ11 , FTO , and two in TCF7L2 ( R2 = 0 . 512 ) , two in CDKAL1 ( R2 = 0 . 677 ) , two in KCNQ1 ( R2 = 0 . 425 ) , and two outside gene regions . All 12 SNPs had been validated to associate with T2D with p<5×10−8 in two or more diverse populations , except rs11196205 ( Table S1 ) . Previous independent reviews and meta-analysis also support the notion that these 12 SNPs are likely to be generally relevant to T2D across populations [16]–[19] . For each of these 12 SNPs , we found that the same allele had been consistently identified as the risk allele with similar odds ratios across 34 studied populations ( Figure S1 , Table S2 ) , though three SNPs show high heterogeneity of allelic odds ratios between studies ( I2>75% ) and three SNPs show moderate heterogeneity ( I2>50% ) ( Table S3 ) . To evaluate whether the shared risk allele and effect sizes were caused by the bias towards cross-ethnic SNPs , we identified a second set of replicated T2D SNPs as those that were replicated in two independent papers with p<5×10−8 without the requirement of five or more studied populations . We identified 11 independent replicated T2D SNPs , including 10 cross-ethnic SNPs and rs864745 in JAZF1 . Each of these 11 replicated T2D SNPs shared the same risk allele and similar effect sizes across all studied populations , suggesting that they are the best representatives of the causal alleles based on the current data . We then plotted the RAF across 11 HapMap populations at independent cross-ethnic T2D SNPs and evaluated their significance of population differentiation against frequency-matched control genomic alleles . Two risk alleles , including rs5219 in KCNJ11 and rs2074196 in KCNQ1 , were not analyzed due to the lack of frequency data . Most of remaining ten cross-ethnic T2D risk alleles shared a similar pattern showing the highest RAF in African populations and the lowest RAF in Asian populations ( Figure 1 ) . To evaluate the significance of differentiation , we retrieved all genomic alleles that shared the same European frequencies and calculated how many of those had both higher African frequencies and lower Asian frequencies than what we observed at each risk allele . Seven out of ten SNPs showed larger differentiation of RAF than 95% European frequency-matched control genomic alleles ( Figure 1 ) . Interestingly , none of the protective alleles of these ten SNPs exhibited significantly higher frequencies in the African population groups . We then calculated T2D RAF in 53 indigenous populations from 1 , 064 individuals in the Human Genome Diversity Panel ( HGDP ) [20] . One risk allele , rs5219 in KCNJ11 was not analyzed due to the lack of frequency data . Nine out of the remaining 11 T2D SNPs exhibited larger differentiation of RAF than 95% European frequency-matched control genomic alleles , except for two SNPs in CDKAL1 ( Six SNPs in Figure 2 , five SNPs in Figure S2 ) . There is a clear pattern of gradually decreasing RAF from the Sub-Saharan Africa to the East Asia through the Europe regions , which was shared among all T2D SNPs . To evaluate whether the observed allele frequency differences among continental groups for T2D SNPs are unusually high , we calculated FST as a measure of population differentiation [21] . Specifically , for each of ten cross-ethnic SNPs ( rs5219 and rs2074196 were not analyzed due to the lack of frequency data ) , we calculated a global as well as three pairwise FST values by pooling populations for each of the three major geographic regions: Africa , Europe , and East Asia . We then compared them to the genome-wide distribution for allele frequency matched SNPs ( defined as within the same 5% minor allele frequency bin in the pooled European samples ) . Consistent with previous results [12] , [13] , we found that all T2D SNPs showed elevated FST values , with five out of ten T2D SNPs among the top 10% , and one of them ( rs11196205 in TCF7L2 ) among the top 1% of the empirical distribution for at least one of the four population comparisons ( Figure S3 ) . Overall , FST values were significantly higher in T2D SNPs than the frequency matched genomic SNPs in global FST ( p = 0 . 0057 ) , using a Mann-Whitney U test . Of the pairwise comparisons , FST of both African vs . East Asian ( p = 0 . 0041 ) and East Asian vs . European ( p = 0 . 025 ) were also highly significant , whereas African vs . European was not ( p = 0 . 3 ) . Thus , FST values also support our findings of extreme population differentiation at T2D SNPs . FST has been widely used to evaluate population differentiation; however , it does not take into account whether risk alleles share the same direction of population differentiation . Here , we developed a novel directional population differentiation method by comparing the average increased frequencies across all T2D risk alleles against the null distribution of European frequency-matched control genomic alleles . We randomly drew 10 genomic alleles that share similar frequencies with the T2D risk alleles in the European populations . Then we calculated their average increased frequencies in the African compared with European populations , which is defined as RAFAfrican−RAFEuropean . Repeating the process 10 , 000 times , we got a null distribution of increased frequencies between African and European populations . T2D risk alleles as an ensemble demonstrated significantly higher RAF in the HapMap African populations against the null distribution of control genomic alleles with a two-side p value of 1 . 4×10−2 ( Figure 3B ) . We adopted the two-side p value to make our method applicable to any new disease without any prior knowledge on whether RAFAfrican−RAFEuropean is larger or smaller than 0 . T2D risk alleles as an ensemble showed significantly higher RAF in the populations in the Sub-Saharan Africa regions ( p = 4 . 9×10−2 , Figure 3D ) , and significantly lower RAF in the East Asia regions ( p = 1 . 0×10−4 , Figure 3C ) , compared to control genomic alleles . We also observed very similar results when analyzing the directional population differentiation using a single T2D SNP from each gene . We then evaluated whether this increased RAF in the Sub-Saharan Africa regions was significant against risk alleles from hundreds of other diseases . As a background , we extracted 15 , 649 risk alleles that had previously been reported as significantly associated with 975 human diseases from Varimed [6] . We randomly drew 11 risk alleles which share similar frequencies with T2D risk alleles in the European populations and calculated a null distribution of increased frequencies . We found that T2D risk alleles showed significantly higher RAF in the Sub-Saharan Africa region than risk alleles from other diseases ( Figure 3D , p = 0 . 05 ) . Similarly , we found that T2D risk alleles showed significantly lower RAF in the HapMap Asia ( p = 8 . 5×10−3 , Figure 3A ) and HGDP East Asia regions ( p = 4 . 0×10−4 , Figure 3C ) . Therefore , our novel method demonstrated that T2D risk alleles as an ensemble showed significantly larger population differentiation than frequency-matched control genomic alleles and risk alleles from hundreds of other diseases . We used this method to systematically evaluate the directional differentiation of RAF of cross-ethnic risk alleles for all diseases in Varimed . We identified 12 common diseases that contain five or more independent cross-ethnic risk alleles ( CEU R2<0 . 7 ) ; each of which had been known to associate with the disease with p<1×10−6 and be replicated in five or more different populations ( Materials and Methods ) . T2D was the only disease that showed significantly decreased RAF in the East Asian populations ( p = 1×10−4 , Figure 4 ) and significantly increased RAF in the Sub-Saharan African populations ( p = 4 . 9×10−2 , Figure 5 ) , compared with frequencies in the European populations . Prostate cancer showed relatively increased RAF in the Sub-Saharan African populations ( p = 6 . 4×10−2 , Figure 5 ) , without decreased RAF in the East Asian populations ( ( p = 0 . 61 , Figure 4 ) . We repeated the study using disease-susceptible risk alleles that had been replicated in two independent papers with p<5×10−8 , and got very similar results ( Figures S4 , S5 , S6 ) . One small difference is that prostate cancer showed significantly higher RAF in the Sub-Saharan African populations at replicated risk alleles ( p = 4×10−3 , Figure S6 ) . Therefore , T2D risk alleles showed the most extreme differentiation of RAF across human populations , compared with other diseases . Interestingly , ensembles of obesity-susceptible risk alleles did not show significant differentiation of RAF in either East Asian ( p = 0 . 37 , Figure 4 ) or Sub-Saharan African populations ( p = 0 . 79 , Figure 5 ) . Eight out of 12 independent cross-ethnic obesity-susceptible risk alleles did not show significantly larger differentiation of RAF than control genomic alleles , individually ( Figure S7 ) . Among the remaining four risk alleles , two are located within FTO and known to increase the risk of T2D as well . Therefore , we did not observe any consistent pattern of RAF across obesity-susceptible risk alleles . The decoupling between obesity- and T2D-susceptible risk alleles indicates that their evolutionary histories are likely different . Having shown the extreme differentiation of frequencies across T2D risk alleles , we further combined the effect size from all independent risk variants to systematically evaluate the differentiation of genetic risk of T2D and 39 other diseases . Using a method we had previously developed [22] , [23] , we calculated a Predicted Genetic Risk ( PGR ) of 40 diseases for each of 1 , 397 individuals using 1 . 46 million SNPs genotyped in the HapMap release 3 ( HapMap3 ) . We only used the HapMap3 data to calculate the PGR so that the same set of SNPs was used for each individual . For each disease , we identified risk SNPs that have been validated in five or more populations , estimated their increased Likelihood Ratio ( LR ) using the genotype frequencies in the case/control groups reported in each previous study , and finally combined the LRs from multiple SNPs using ethnic-specific linkage disequilibrium R2 data to report a summarized risk score for each disease . The summarized score estimates the Predicted Genetic Risk ( PGR ) for an individual given their genotypes at independent risk SNPs . We calculated the PGR for each of 1 , 397 individuals on 40 diseases ( Table S5 ) , and found that seven diseases showed significantly larger population differentiation than global frequency-matched genomic SNPs ( Figure 6 ) . After Benjamini-Hochberg multi-test correction , only T2D and colorectal cancer demonstrated significantly larger population differentiation than random genomic SNPs . After accounting for the genomic SNPs , T2D showed significantly increased PGR in the African populations relative to others ( p = 4 . 7×10−3 ) , and significantly decreased PGR in the Asian populations relative to others ( p = 3 . 0×10−5 ) . Colorectal cancer showed a similar pattern , but the PGR differences among population groups were much smaller because there were only two cross-ethnic SNPs . The statistical significance of observed PGR differences was estimated using a genomic background by randomly selecting global frequency-matched genomic SNPs and re-calculating the PGR on each disease using the random SNPs . We then calculated the likelihood of observing a larger PGR difference between each of three population groups ( Asian , European , African ) and others . Overall , T2D demonstrated the most significant population differentiation on PGR compared to frequency-matched genomic SNPs . Finally , we evaluated the robustness of the observed ethnic disparity of T2D PGR in three conditions . To evaluate the effect of SNPs discovered and validated in specific subpopulations , we compared the distributions of PGR using only the ethnicity-specific T2D SNPs , their genotype frequencies in the case and control groups , and LR separately from studies specifically on each of the following ethnicities: Caucasian , African , Chinese , Japanese , and Indian Asian ( Figure 7 ) . T2D PGR distributions demonstrated a strikingly similar pattern no matter which original studies were used to retrieve the previously discovered significant T2D SNPs , genotype frequencies , and LRs . African populations always have the highest PGR and Asian populations always have the lowest PGR . Recently , a T2D genetic risk score had been validated to associate with the increased risk of developing diabetes in a prospective random trial on 2 , 843 Diabetes Prevention Program participants from five ethnic groups representative of the U . S . population [19] . The T2D genetic risk score was calculated by multiplying the number of risk alleles by the natural log of the odds ratio at each SNP , and summing over 34 SNPs . Twenty SNPs were measured in the HapMap3 individuals . We calculated the T2D risk score for each of 1 , 397 HapMap3 individuals , and found the same differential T2D genetic risks , with significantly increased risk in the African populations ( p = 0 . 01 ) and significantly decreased risk in the Asian populations ( p = 3 . 5×10−3 , Figure 8 ) . Third , in each of the studies above , we were constrained in the number of SNPs usable for calculating PGR for T2D , based on intersecting the set of SNPs known to be associated with T2D with the set of SNPs genotyped in individuals in HapMap3 and HGDP . To ensure our findings are not an artifact of bias in the measured SNPs , we repeated our analysis using a larger set of T2D associated SNPs in 49 individuals for which whole genome sequencing is obtained , including 4 JPT , 4 CHB , 5 MEX , 3 Puerto Rican , 2 TSI , 14 CEU , 5 ASW , 1 MKK , 4 LWK , and 7 YRI . The subpopulation distributions of PGR were very similar between whole genome sequencing and genotyping technologies ( HapMap3 ) for T2D ( Figure 9 ) . This equality did not hold for all diseases , as very different distributions are observed for melanoma . This suggests that current genotyping technology captures enough signals to reproduce the ethnic disparity for some common diseases , such as T2D , while whole genome sequencing will likely capture many more genetic variants influencing melanoma and perhaps other diseases .
We developed a novel method to systematically evaluate the directional differentiation of Risk Allele Frequencies ( RAF ) of ensembles of cross-ethnic SNPs for 12 common diseases across 11 populations from the HapMap project and 53 indigenous populations from the HGDP project . We found that type 2 diabetes ( T2D ) demonstrated the significant differentiation of RAF among diverse populations , compared with the European frequency-matched control genomic alleles and risk alleles for other diseases ( Figure 3 , Figure S4 ) . T2D showed the most extreme differentiation among 12 common diseases , no matter whether we used cross-ethnic SNPs that had been replicated in five different populations ( Figure 4 , Figure 5 ) or SNPs that had been replicated in two studies ( Figures S5 , S6 ) . This extreme differentiation is caused by the phenomenon that all T2D risk alleles share a consistent pattern of gradually decreasing population frequencies from Sub-Saharan Africa through Europe to East Asia regions ( Figure 1 , Figure 2 , Figure S2 ) . This phenomenon , that T2D risk alleles decrease frequencies when humans migrate [24] , suggests many potential explanations . One likely cause is the adaptation to the disparities of agriculture development across continents . It has been previously reported that some T2D SNPs have higher risk allele frequencies in populations where cereals are the main dietary component , and observed risk allele frequency might be related to historical events , such as the dispersal out of Sub-Saharan Africa to regions with different climates and the adoption of more specialized-often less diverse-diets ( i . e . farming and animal husbandry vs . foraging ) [25] . There were three major events in human evolution , including early migration from 200KYA to 10KYA , agriculture revolution and population expansion from 10KYA to 4KYA , and new world discovery and associated mass-migration and admixture after 4KYA [26] . The significantly decreased frequencies of T2D risk alleles in the East Asia might be caused by the agriculture revolution , including the cultivation of white rice and pork in China . A related explanation stems from the thrifty genotype [27] hypothesis , which asserts that a predisposition to insulin resistance may have protected individuals during periods of food deprivation by reducing muscle utilization of glucose and favoring glucose utilization in organs , such as brain , that operates through an insulin independent mechanism [27] . Combining these two related explanations together , we speculate that the decreasing T2D risk allele frequencies are caused by the promotion of energy storage and usage appropriate to environments and insistent energy intakes . Another speculation is that T2D is known to find roots in the mismatch between our genetics and environment , as food contributes a significant environmental impact . When humans migrate , environmental change may have led to a mismatch between genetics and available diet , and put a positive evolutionary pressure on the frequencies of T2D protective alleles . Therefore , the decreasing T2D risk alleles are expected , while the other diseases are unusual given the underlying demographic history . Future evolutionary analysis on these T2D SNPs may provide some insight on the origin of this pandemic disease , as will more population-specific genetic studies . Having shown the extreme differentiation of T2D RAF , we further combined the effect sizes from all independent risk variants and calculated a Predicted Genetic Risk ( PGR ) for each of 1 , 397 individuals in the HapMap3 project . T2D showed the most significant population differentiation among 40 diseases , after correcting for control genomic genotypes ( Figure 6 ) . We identified a consistent pattern of high PGR in the African and low PGR in the Asian regardless whether we used ethnic-specific SNPs ( Figure 7 ) , validated risk scores ( Figure 8 ) , or different genotyping/sequencing technologies ( Figure 9 ) . Our results indicate that there is indeed a differential T2D genetic risk across different populations across continents . The distributions we have found are very similar to a recent report measuring 19 common variants on five continent populations [15] , with the highest risk in the African populations , and lowest risk in the East Asian populations . The populations examined by this study are distributed broadly around the world , representing a wide range of environmental exposures and lifestyles . Hence , it is challenging to associate the increased prevalence of risk-associated alleles with actual manifestations of T2D , which we know to be heavily influenced by environmental factors . However , studies in England and the United States have consistently shown that individuals with African ancestry have increased diabetes rates relative to their neighbors of European or East Asian ancestry [28]–[31] , while those with Chinese ancestry had lower incidence compared to others in a recent 10-year Canadian study [32] . At the same time , citizens in China have higher prevalence of T2D within their own country [33] , [34] . Disparities in T2D rates may be attributed to social , cultural , and economic differences or possible genetic confounders such as admixing of ancestral ethnicities , though our results suggest that differential genetics may indeed play some role in these differences in incidence rates . We also found that African had higher PGR on prostate cancer than other populations . Epidemiology data from Center for Disease Control and Prevention from 1999 to 2007 show that incidence of prostate cancer is 1 . 56 times higher in the African American than white American . Further investigation on the genetic reasons behind the observed ethnic disparity of disease incidence rates across ethnic/racial groups might identify personalized medicine to improve the health disparity . Many challenges to evaluate the population differentiation of RAF and PGR remain . Foremost , many of the SNPs identified from genome-wide association studies ( GWASs ) are tag SNPs and are therefore not assumed to be causal [35] . However , each of the 12 cross-ethnic T2D SNP share the same risk allele and similar effect sizes across 34 different studied populations ( Figure S1 , Table S2 ) , suggesting that they are the best representatives of the causal alleles based on the current data . The consistent observation of differential T2D genetic risk with different SNPs , risk scores , and technologies suggests validity as new causal variants are identified , but this remains a hypothesis that needs to be tested in the future . Second , we acknowledge that we adopted a relaxed p value cutoff of p<1×10−6 to identify cross-ethnic SNPs for a wide-variety of diseases for comparisons . With more GWAS in diverse population groups , a more rigorous cutoff and ethnicity-specific effects should be used . Third , there may be some ethnic-specific gene-environment interaction . Forth , our observed disparity of PGR between population groups might be related to the disparities in the application of modern genetic tools to study diseases across ethnicities . Finally , a large component of heritable risk is still missing for most common diseases , and consequently missing in our analysis here [36] . Future GWAS and sequencing studies on different ethnic groups under diversified environmental conditions will likely further reveal and illustrate the origins of complex diseases . In conclusion , we found that T2D risk alleles demonstrated extreme differentiation compared to other diseases , with population frequencies decreasing from Sub-Saharan Africa and through Europe to East Asia . These patterns may contribute to the observed disparity of T2D incidence rates across worldwide ethnic populations .
As described previously [6] , we have been manually curating a quantitative human disease-SNP association database from literature . First , we downloaded all abstracts from MEDLINE , and identified human genetic papers using a list of Medical Subject Headings [37] , such as “Genome-Wide Association Study” , “Genetic Variation” , “Polymorphism , Genetic” , “Genome , Human” , “Polymorphism , Single Nucleotide” , “Genotype” , “Genetic Predisposition to Disease” , “Case-Control Studies” , “Alleles” , “Cohort Studies” . We then filtered these papers through abstracts , titles , and keywords . We also downloaded papers from curated disease-SNP databases , such as GWAS catalog from National Human Genome Research Institute [38] . Combined together , we identified 5 , 065 human genetic papers representing 1 , 495 diseases . Second , four curators manually extracted data from the full text , figures , tables , and supplemental materials of 5 , 065 human genetics papers , and recorded more than 100 features from each paper . We recorded many aspects of the associations , including the disease name ( e . g . coronary artery disease ) , specific phenotype ( e . g . acute coronary syndrome in coronary artery disease ) , study population ( e . g . Finnish individuals ) , case and control population ( e . g . 2 , 508 patients with coronary artery disease proven by angiography ) , gender distribution , genotyping technology , major/minor/risk alleles , odds ratio , 95% confidence interval of the odds ratio , published p value , and genetic model . Studies on similar diseases were categorized and mapped to the Concept Unique Identifiers ( CUI ) from the Unified Medical Language System ( UMLS ) [39] . For each study , the frequencies of each genotype and allele in the case and control populations were recorded , and used to estimate the effect size [22] , [23] . We identified all disease-susceptible alleles from Varimed that were reported to increase the risk of human disease with p-value<10−6 . We focused on alleles that were directly associated with the increased risk of human disease by removing the following SNPs: SNPs that were specified as non-significant in the original papers , SNPs that were identified from studies with diseased patients in the control groups , SNPs that were associated with non-disease traits , SNPs that were associated with disease through haplotype blocks or interaction terms . Risk alleles on the negative strands were translated into alleles in the positive strands . Negative strands were identified by comparing the major/minor alleles in the study with the major/minor alleles in the similar population in the HapMap3 . Many risk alleles had been reported in multiple studies , and we integrated all studies and ranked the risk alleles by the strength of replication , including the number of replicated populations , studies , and total sample sizes . We identified cross-ethnic SNPs as SNPs being associated with a disease with a p-value<1×10−6 in at least one populations and reported as significant in five or more different subpopulations with a p-value<1×10−6 in GWAS studies or p-value<0 . 01 in small candidate studies . We removed SNPs that were specified as non-significant in the original papers . For each disease , we ranked all cross-ethnic SNPs by the number of replication studies , the total sample sizes and the number of populations where the associations being replicated . Starting from risk alleles with the strongest evidence , we identified risk alleles with the linkage disequilibrium R2≥0 . 7 in the CEU population in the HapMap project , and removed ones with less evidence . We identified 12 diseases with 5 or more independent cross-ethnic risk alleles , each of them being validated in five or more populations . We found 12 independent cross-ethnic risk alleles for Type 2 diabetes ( T2D ) , and plotted their association p values across 34 populations using the levelplot function in R . We then evaluated the between-study heterogeneity of allelic odds ratios on each of 12 cross-ethnic T2D SNPs using the meta R package . Similarly , we identified disease SNPs that have been replicated in two or more papers with p<5×10−8 . We retrieved the allele frequencies at 2 . 3 million SNPs in the 11 populations from HapMap 2+3 , which were released in August 2010 . We plotted the RAFs across the 11 populations as bar graphs at each independent cross-ethnic risk allele for each disease using a barplot function in R . We then evaluated the statistical significance of the population differentiation of RAF at each disease SNP by calculating the percentage of European frequency-matched control genomic alleles that show the RAF difference larger than the observed . For each risk allele , we retrieved all control genomic alleles sharing similar average frequencies within ±0 . 01 in the European ( TSI , CEU ) populations . Then , we calculated the percentage of matched genomic alleles that have both African frequencies higher than the observed and Asian frequencies lower than the observed . We recorded the percentage as a p value which is the likelihood of finding similar or larger differentiation of RAF from frequency-matched genomic alleles . All risk alleles with p<0 . 05 were considered as showing significantly larger population differentiation than control genomic alleles . We received the measured and imputed allele frequencies at 3 . 1 million SNPs across 53 populations from 1 , 064 individuals from Human Genome Diversity Panel ( HGDP ) from Joseph Pickrell from University of Chicago [7] , [13] . We modified a script from Joseph Pickrell using Generic Mapping Tools [40] and plotted the worldwide map showing the distribution of RAF across the 53 HGDP populations [13] . Similar with the analysis on HapMap , we calculated the percentages of European frequency-matched genome alleles that have both Sub-Saharan African frequencies higher than the observed and East Asian frequencies lower than the observed . We categorized populations into three major regions . Sub-Saharan Africa region includes BantuSouthAfrica , Biaka Pygmies , Mandenka , Mbuti Pygmies , San , Yoruba , and BantuKenya . East Asia region includes Cambodian , Dai , Daur , Han , Han-NChina , Hezhen , Japanese , Lahu , Miao , Mongola , Naxi , Oroqen , She , Tu , Tujia , Uygur , Xibo , Yakut , and Yi . Europe region includes Adygei , Basque , French , Italian , Orcadian , Russian , Sardinian , and Tuscan . Global FST , as well as three pairwise FST ( African vs . East Asian , African vs . European , East Asian vs . European ) values were calculated for all SNPs from HapMap3 , using a custom script implementing the method in the PopGen module of BioPerl [41] . For the calculations , all populations for the respective geographic regions were pooled , so that the global FST reflected the overall differentiation of those major geographic regions . All SNPs were grouped into 10 bins according to their average Minor Allele Frequencies ( MAF ) across the European HapMap3 populations ( CEU , TSI ) , and T2D SNPs were compared to all genomic SNPs within the same MAF bin . For the Mann-Whitney U test , we used normalized FST values , which were obtained by subtracting the mean and dividing by the standard deviation within each MAF bin for each SNP . We evaluated the population differentiation of an ensemble of 10 independent cross-ethnic T2D risk alleles by comparing their average increased RAF in the African populations against the null distribution of control genomic alleles and 15 , 649 risk alleles from 975 diseases . First , we randomly retrieved 10 control genomic alleles that share similar European frequencies ( ±0 . 05 ) with T2D risk alleles , and calculated their average increased frequencies in the African populations ( MKK , LWK , YRI , ASW ) . Repeating the above process 10 , 000 times , we drew a null distribution of increased frequencies between African and European populations from control genomic alleles . The average increased frequency of 10 T2D risk alleles in the African vs . European populations was then compared with the null distribution to calculate the likelihood of observing consistently increased African frequencies from matched genomic alleles . Similarly , we randomly retrieved 10 European frequency-matched alleles ( ±0 . 05 ) from 15 , 649 risk alleles from other diseases , and calculated a null distribution of increased African frequencies from disease risk alleles . We also calculated the null distribution of Asian frequencies from HapMap , East Asian and Sub-Saharan African frequencies from HGDP using European frequency-matched genomic alleles and risk alleles from other diseases . Using a method we described previously [22] , [23] , we calculated the Predicted Genetic Risk ( PGR ) of 40 diseases for 1 , 397 individuals from HapMap release 3 . Each individual was genotyped at 1 . 46 million SNPs . We calculated the PGR of 40 diseases using independent cross-ethnic SNPs , each of them had been validated in five or more different populations . We estimated a genetic risk using a likelihood ratio for each SNP defined by the relative frequency of an individual's genotype in the diseased vs . healthy control populations ( e . g . , given a genotype “AA” , LR = Pr ( AA|diseased ) /Pr ( AA|control ) ) . The LR incorporated both the sensitivity and specificity of the test and provided a direct estimate of how much a test result would change the odds of having a disease [23] . We excluded studies with diseased patients in the control group . For each allele , we averaged the LRs from multiple studies with a weight of the square root of the sample size to give higher confidence to studies with larger sample size . For SNP pairs in linkage disequilibrium ( R2≥0 . 3 in the corresponding population group ) , we removed the SNP with weaker evidence according to the number of replication studies , sample sizes and validated populations . We considered remaining SNPs as independent genetic test and multiplied their LRs to report the summarized score as the PGR . We then plotted the distribution of PGR across 11 HapMap populations using a kernel density function in the R package . To evaluate the statistical significance of the population differentiation of PGR after correcting the difference of RAF , we randomly replaced disease genotypes with global frequency-matched non-associated genotypes , and re-calculated the PGR for each of 1 , 397 HapMap individuals . We calculated the PGR using the random control genomic genotypes along with the original LR , averaged the log ( PGR ) values for each of 11 population groups , and repeated the process 100 , 000 times . Then we calculated the p_Afr as the percentage of obtaining a value of log ( PGR , African ) -log ( PGR , Other ) from random genotypes larger than the observed values . Similarly , we calculated p_Asi and P_Eur as the likelihood of obtaining more extreme values of log ( PGR , Asian ) -log ( PGR , Other ) and log ( PGR , European ) -log ( PGR , Other ) from random genomic genotypes . Each p value was calculated as a two-side p value . We recalculated the distribution of PGR across the 11 HapMap populations using SNPs and LRs that were associated with T2D with p<1×10−6 in the original studies on each of the following populations , including Caucasian , African , Chinese , Japanese , and Indian Asian . Whole-genome sequences were produced using published methods [42] for 49 cell-line derived DNA samples obtained from the Coriell Institute . Each genome was sequenced to over 55× coverage and calls were produced using a local de novo assembly based pipeline . Comparison of these data to dbSNP was performed using alignment-based methods which account for SNPs which may be contained within more complex variant sequences . We pooled the 49 samples into three groups: Asian , African , and Others . Then we calculated the distribution of PGR across these three population groups on T2D and Melanoma . All disease SNPs with p<1×10−6 in the original studies on any ethnicity were used . | We identified 12 risk alleles that had been validated to increase the risk of type 2 diabetes ( T2D ) in five or more different subpopulations . These risk alleles share a consistent pattern of decreasing frequencies in the human genomes from Sub-Saharan Africa and through Europe to East Asia regions . These differential frequencies are statistically significant , compared with European frequency-matched control genomic alleles and risk alleles for other diseases . The differential frequencies of T2D risk alleles further caused the significant differentiation of genetic risks of T2D , with higher risk in the African and lower risk in the Asian populations . The differences might be caused by the promotion of energy storage and usage appropriate to environments and inconsistent energy intake . Future evolutionary and environmental analysis of this unique pattern may provide further insight on the origin of T2D and modern disparities in T2D incidence . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"public",
"health",
"and",
"epidemiology",
"statistics",
"population",
"genetics",
"algorithms",
"genome",
"analysis",
"tools",
"genome",
"databases",
"mathematics",
"epidemiology",
"biology",
"metabolic",
"disorders",
"computer",
"science",
"genetics",
"genomics",
"computational",
"biology",
"genetics",
"and",
"genomics",
"human",
"genetics"
]
| 2012 | Type 2 Diabetes Risk Alleles Demonstrate Extreme Directional Differentiation among Human Populations, Compared to Other Diseases |
CD4+ and CD8+ T cells are central players in immunity to helminth infections . However , the role of T cell subsets in human helminth infections is not well understood . In addition , the common γc cytokines , IL-2 , IL-4 , IL-7 , IL-9 and IL-15 play an important role in the maintenance of these CD4+ and CD8+ T cell subsets . To examine the major T cell subsets and their association with the common γc cytokines , the absolute numbers of CD4+ and CD8+ naïve , central memory , effector memory and effector cells and the plasma levels of IL-2 , IL-4 , IL-7 , IL-9 and IL-15 were measured in Strongyloides stercoralis ( Ss ) infected ( INF , n = 60 ) , helminth—uninfected ( UN , n = 58 ) and in post treatment INF individuals . Ss infection is characterized by significantly increased absolute numbers of naïve and decreased absolute numbers of central and effector memory CD4+ T cells in comparison to UN individuals . No significant difference in the numbers of CD8+ T cell subsets was observed between the groups . The numbers of naïve cells and central memory CD4+ T cells were significantly reversed after anthelmintic treatment . Circulating levels of IL-2 , IL-7 and IL-15 were significantly diminished , whereas the levels of IL-4 and IL-9 were significantly increased in INF compared to UN individuals . Following anthelminthic treatment , IL-2 , IL-7 and IL-15 levels were significantly increased , while IL-4 and IL-9 levels were significantly decreased . Our data also showed a significant positive correlation between the levels of IL-7 and the numbers of central and effector memory CD4+ T cells . Ss infection is characterized by alterations in the absolute numbers of CD4+ T cell subsets and altered levels of common γc cytokines IL-2 , IL-4 , IL-7 , IL-9 and IL-15; alterations which are partially reversed after anthelmintic treatment .
Strongyloides stercoralis ( Ss ) , a soil transmitted nematode that resides in the small intestine of humans , infects approximately 30–100 million people worldwide [1] . The clinical manifestations of Ss infection can range from the clinically asymptomatic to , at its most severe , the potentially fatal hyperinfection syndrome . Ss infection is associated with down modulation of Th1 and Th17 responses and up-regulation of Th2 and Th9 CD4+ T cell responses [2 , 3] . How Ss infection influences CD8+ T cell responses has not been studied in detail . In addition , very little is known about CD4+ or CD8+ memory T cell subset distribution in Ss infection . Common cytokine receptor γ-chain family ( γc cytokines ) are associated with the process of memory T cell generation [4–6] . The sharing of the γ chain by their receptors , common downstream signalling pathways , link members of this cytokine family functionally . Data reveal that IL-2 , IL-4 , IL-7 , IL-9 and IL-15 participate in the initiation of T cell responses and that some of these cytokines are vital for the development or maintenance of memory T cells [7] . Murine studies have shown that different cell types produce the major γc cytokines IL-7 and IL-15 , that play important roles in the maintenance of CD4+ [8] and CD8+ T cells [9 , 10] . Human studies also have shown that T cells proliferate in response to common γc dependent cytokine signaling [11 , 12] , but the association between memory T cell subsets and these common γc cytokines in helminth infections has not been examined . The common γc cytokines , IL-2 , IL-7 and IL-15 play an important role in peripheral T cell growth and survival [4–6] . However , the effects of helminth infection on common γc cytokine—IL-2 , IL-4 , IL-7 , IL-9 and IL-15- levels have not been explored in Ss infection . We hypothesized that Ss infection would be associated with alterations in memory T cell subset distribution , alterations that could be reflective of changes in IL-2 , IL-4 , IL-7 , IL-9 and IL-15 . We , therefore , examined the ex vivo phenotypic profile of CD4+ and CD8+ memory T cell subsets and the circulating levels of common γc cytokines ( IL-2 , IL-4 , IL-7 , IL-9 and IL-15 ) in Ss-infected ( INF ) and -uninfected ( UN ) individuals . We also examined the effect of anthelmintic treatment on the distribution of these memory T cell subsets and the cytokines .
All individuals ( age between 18–65 years ) were examined as part of a natural history study protocol approved by Institutional Review Boards of the National Institute of Allergy and Infectious Diseases ( USA ) and the National Institute for Research in Tuberculosis ( India ) , and informed written consent was obtained from all participants . We studied 118 individuals comprised of 60 clinically asymptomatic , Ss-infected ( hereafter INF ) individuals and 58 Ss-uninfected , endemic healthy ( hereafter UN ) individuals in Sirukalathur village , Kanchipuram District , Tamil Nadu , South India ( Table 1 ) . These individuals were all recruited from a rural population by screening of individuals for helminth infection by stool microscopy and serology as described previously [13–15] . None had previous anthelmintic treatment or a history of prior helminth infection . Follow up among the INF individuals was performed at 6 months following treatment . These individuals were different from our previous studies on serum cytokines in Ss individuals [16] . Ss infection was diagnosed by the presence of IgG antibodies to the recombinant NIE antigen as described previously [14 , 15] . This was further confirmed by stool microscopy . A single stool sample was obtained and examined for intestinal helminth infection by Kato-Katz technique . Stool samples found to be negative for other intestinal helminths by stool microscopy and positive for Ss infection by serology were then subjected to specialized stool examination with nutrient agar plate cultures . Only the individuals who were positive for Ss infection by both serology and nutrient agar culture plate technique were selected for this study . Filarial infection was excluded in all study participants by virtue of being negative in tests for circulating filarial antigen . All INF individuals were treated with single dose of ivermectin ( 12mg ) and albendazole ( 400 mg ) and follow—up blood draws were obtained six months later . Follow up examination was done by stool microscopy of a single stool sample using Kato-Katz and nutrient agar plate cultures , which was negative for Ss as well as other intestinal helminth infection . In addition , serology showed a significant decrease in the IgG titers to NIE antigen . Leukocyte counts and differentials were performed on all individuals using an AcT5 Diff hematology analyzer ( Beckman Coulter ) . All antibodies used in the study were from BD Biosciences ( San Jose , CA ) , BD Pharmingen ( San Diego , CA ) , eBioscience ( San Diego , CA ) , or R&D Systems ( Minneapolis , MN ) . Whole blood was used for ex vivo phenotyping and it was performed on all 118 individuals . Briefly , 250ul aliquot of whole blood was added to a cocktail of monoclonal antibodies specific for various immune cell types . T cell phenotyping was performed using antibodies directed against CD45-Peridinin chlorophyll protein ( PerCP; clone 2D1 , BD ) , CD3-AmCyan ( clone SK7; BD ) , CD4-phycoerythrin ( PE ) Cy7 ( clone SK3; BD ) , CD8-allophycocyanin ( APC ) H7 ( clone SK1; BD ) , CD45RA-Pacific Blue ( clone H1100; Biolegend , Cambridge , UK ) , and CCR7-FITC ( clone 3D12; eBioscience ) ( S1 Table ) . Naive cells were classified as CD45RA+ CCR7+ , central memory cells as CD45RA- CCR7+ , effector memory cells as CD45RA-CCR7- and effector cells as CD45RA+ CCR7- ( S2 Table ) . Following 30 min of incubation at room temperature , erythrocytes were lysed using 2 ml of FACS lysing solution ( BD Biosciences Pharmingen ) , cells were washed twice with 2 ml of PBS and suspended in 200 ul of PBS ( Lonza , Walkersville , MD ) . Eight- color flow cytometry was performed on a FACS Canto II flow cytometer with FACSDIVA software , version 6 ( Becton Dickinson ) . The gating was set by forward and side scatter , and 1 , 00000 gated events were acquired . Gating strategy for memory T cell subsets were shown in S1 Fig . Data were collected and analyzed using FLOW JO software ( TreeStar , Ashland , OR ) . Leukocytes were gated using CD45 expression versus side scatter . Total lymphocyte counts were obtained from the hematology profile and the percentage of gated lymphocytes by flow cytometry was used to calculate the absolute numbers of T cell subsets . Circulating levels of IL-2 , IL-4 , IL-7 and IL-15 were measured using the Quantikine ELISA kit ( R&D Systems ) and IL-9 ( eBiosciences ) were measured by enzyme-linked immunosorbent assay ( ELISA ) , according to the manufacturer’s instructions . The lowest detection limits were as follows: IL-2 , 31 . 2 pg/mL; IL-4 , 31 . 2 pg/mL; IL-7 , 7 . 813 pg/mL; IL-15 , 16 . 625 pg/mL; IL-9 , 3 . 1 pg/mL . Data analyses were performed using GraphPad PRISM ( GraphPad Software , Inc . , San Diego , CA , USA ) . Geometric means ( GM ) were used for measurements of central tendency . Statistically significant differences were analyzed using the nonparametric Mann-Whitney U test used to compare INF versus UN and Wilcoxon signed rank test was used to compare memory T cell panel and common γ-chain cytokines levels before and after treatment . Multiple comparisons were corrected using the Holm’s correction . Correlations were calculated by the Spearman rank correlation test . Analyses were performed using Graph-Pad PRISM Version 6 . 0 ( GraphPad , San Diego , CA ) or R . JMP 13 ( SAS ) software was used to perform Spearman rank correlation matrix . Logistic regression analysis was used to identify factors that influenced by Ss infection . P ≤ 0 . 05 was considered statistically significant . STATA 15 . 0 ( StataCorp , College Station , Texas , USA ) was used for Logistic regression analysis .
The baseline characteristics and demographics of the study population are shown in Table 1 . No significant differences in age , gender , socioeconomic status , or geographical location were observed between the two groups . The baseline hematological features of the study population are also shown in Table 1 . As can be seen , INF individuals had few differences in any of the hematological parameters measured with the exception of the RBC and absolute eosinophil counts ( AECs ) that were higher in INF individuals ( p<0 . 0001 ) . To determine if Ss infection altered absolute numbers of T cells , we measured the absolute leukocyte count and absolute CD4+ and CD8+ T cell counts in INF and UN individuals . There were no significant differences in absolute leukocyte counts or absolute CD4+ and CD8+ T cell counts between INF and UN individuals ( Fig 1A ) . To study the distribution of CD4+ memory T cell subsets in Ss infection , we examined the counts of four different CD4+ T cell subsets ( naive , central memory , effector memory and effector ) in INF and UN individuals at baseline . As shown in Fig 1B , INF had significantly increased numbers of naïve CD4+ T cells ( GM of 339 . 1 in INF versus 246 . 1 in UN ( p = 0 . 0289 ) ) in comparison with UN individuals . In contrast , central memory ( GM of 241 in INF versus 297 . 9 in UN ( p = 0 . 0027 ) ) and effector memory ( GM of 213 . 7 in INF versus 253 . 2 in UN ( p = 0 . 0483 ) ) CD4+ T cell numbers were significantly decreased in INF individuals when compared with UN individuals . There were no significant differences in the counts of naive , central , effector memory , and effector cells in CD8+ T cell compartment between INF and UN individuals ( Fig 1C ) . Therefore , Ss infection appears to be associated with alterations in the memory subset distribution of CD4+ T cells . To determine the effect of treatment on absolute leukocyte counts and absolute numbers of CD4+ and CD8+ T cells , we measured these parameters in INF individuals 6 months following anthelmintic treatment . There were no significant differences in absolute leukocyte counts or absolute CD4+ and CD8+ T cell counts ( Fig 2A ) . In contrast , as shown in Fig 2B , the absolute numbers of naïve CD4+ T cells ( GM of 339 . 1 in pre-treatment ( Pre-Tx ) compared to 248 . 6 in post-treatment ( Post-Tx ) p = 0 . 0247 ) were significantly decreased and central memory T cell ( GM of 241 in Pre-Tx versus 314 . 2 in Post-Tx p<0 . 0001 ) counts were significantly increased following anthelmintic treatment . As shown in Fig 2C , CD8+ T cell central memory counts ( GM of 201 . 8 in Pre-Tx versus 227 . 5 in Post-Tx , p = 0 . 0087 ) were also significantly increased after treatment . Therefore , treatment of Ss infection is associated with partial but significant reversal of the alterations seen in naïve and central memory T cells prior to treatment . Moreover , following treatment , there was an increase in the CD8+ central memory T cell compartment . Because common γc cytokines play an essential role in homeostasis and expansion of memory T cells , we wanted to understand the association of γc cytokines with Ss infection and following treatment . We measured the circulating plasma levels of IL-2 , IL-4 , IL-7 , IL-9 and IL-15 in INF and UN individuals . As shown in Fig 3A , INF had significantly lower serum levels of IL-2 ( GM of 111 . 6 pg/ml in INF versus 156 . 8 pg/ml in UN , p = 0 . 0005 ) ; IL-7 ( GM of 64 . 72 pg/ml in INF versus 172 . 5 pg/ml in UN , p = 0 . 0003 ) and IL-15 ( GM of 12 . 79 pg/ml in INF versus 36 . 28 pg/ml in UN , p = 0 . 0001 ) in comparison to UN individuals . In contrast , INF individuals had significantly enhanced serum levels of IL-4 ( GM of 2196 pg/ml in INF versus 1396 pg/ml in UN , p = 0 . 0004 ) and IL-9 ( GM of 245 . 7 pg/ml in INF versus 183 . 2 pg/ml in UN , p = 0 . 0002 ) when compared with UN individuals . Following treatment , IL-2 ( fold change 1 . 183 ) , IL-7 ( fold change 1 . 130 ) , and IL-15 levels ( fold change 1 . 513 ) levels exhibited an increase from pretreatment levels in INF individuals . In contrast , levels of IL-4 ( fold change 0 . 727 ) and IL-9 ( fold change 0 . 799 ) were diminished from pretreatment levels when compared to post treatment levels ( Fig 3B ) Thus , Ss infection is associated with altered levels of common γc cytokine levels and partial reversal following treatment . The relationship between the levels of common γc cytokines , IL-2 , IL-4 , IL-7 , IL-9 and IL-15 and CD4+ memory T cell subsets ( naïve cells , central memory cells , effector memory cells and effector cells ) were next assessed by Spearman correlation . As shown in Fig 4A and 4D , CD4+ naïve and effector cells did not reveal any correlation with common γc cytokines ( IL-2 , IL-4 , IL-9 and IL-15 ) . As shown in Fig 4B , the levels of IL-7 exhibited a significant positive correlation with the absolute numbers of CD4+ central memory T cells ( r = 0 . 2366; p = 0 . 0153 ) , whereas other common γc cytokines ( IL-2 , IL-4 , IL-9 and IL-15 ) did not show any significant correlation with central memory T cell subsets . Similarly , IL-7 ( but not other common γc cytokine ) levels exhibit a significant positive correlation with CD4+ effector memory T cells ( r = 0 . 2413; p = 0 . 0041 ) ( Fig 4C ) . Logistic regression analysis was done to examine the influence of age , sex or haematological parameters on CD4 and CD8 memory subsets—naïve , central memory , effector memory and effector cells and the γc cytokines IL-2 , IL-4 , IL-7 , IL-9 and IL-15 . As shown in Table 2 , logistic regression analysis did not reveal any effect of on the CD4+ memory subsets between the two groups . Among the γc cytokines , the adjusted odds ratio for IL-15 alone was significant , other cytokines did not show any difference . Similarly , CD8+ memory subsets did not show any difference . Thus , Ss infected individuals associated with alterations in the T cell subset distribution and altered plasma levels of IL-2 , IL-4 , IL-7 , IL-9 and IL-15 .
The major subsets of memory CD4+ and CD8+ T cells can be defined by the expression of CD45RA and CCR7 [17] , and these cells can be subdivided into naive , central memory , effector memory and effector T cells in the circulation based on the expression pattern of the above markers . CCR7+ memory T cells are termed central memory T cells and are able to home to secondary lymphoid organs and produce high levels of IL-2 but low levels of other cytokines , whereas CCR7– memory T cells are termed effector memory T cells and are able to produce high levels of effector cytokines , exert rapid effector functions , and home to peripheral tissues [18] . CD4+ memory T cells have been shown to mediate protection against re-infection in experimental helminth infection [19] . Central memory and effector memory T cells have been shown to play important roles in protective immune responses in animal models of vaccination or protective immunity with central memory T cells dominating the antigen-specific immune response in vaccination experiments [20–22] . Alteration of memory T cell responses may be involved in the modulation of T cell responses in individuals with patent filarial infection , another tissue-invasive helminth parasite [23] . Indeed , effector memory and central memory CD4+ T cells are associated with protective immunity in some parasitic infections [24 , 25] . In patients with schistosomiasis , proportion of CD4+ memory T cells was significantly lower than in uninfected people [26] . Similarly , the present study demonstrated diminished number of effector memory and central memory CD4 cells in Ss-infected individuals , with a corresponding increase in naïve CD4 cells . Because central memory T cells have a high proliferative potential required to mediate protection against a number of pathogens [20 , 27 , 28] , the fact that central memory T cells are decreased in Ss and other helminth infections ( e . g . filarial infections ) and fail to proliferate to antigen [29 , 30] suggests that central memory T cells could play a role in the immune response to helminth infections in humans . It is possible that the decreased numbers of effector memory cells in Ss infected individuals could be due to increased migration of memory cells from the circulation to mucosal sites . Previous studies on filarial infection have shown that alterations in effector and memory cell population ex vivo could contribute to the antigen specific T cell hypo-responsiveness , commonly seen in these infections [23 , 31] . In addition , the differences in the T memory cell distributions might be due to alterations in antigen presentation by antigen presenting cells . CD4+ effector and memory T cells require more abundant presentation of antigens by antigen presenting cells than do CD8+ memory T cells [32] . In the present study , Ss infected individuals exhibited increased naïve CD4+ T cell counts and decreased central memory and effector memory CD4+ T cell counts . The naïve CD4+ T cell counts and the central memory CD4+ T cell counts were significantly but partially reversed at 6 months following anthelmintic treatment . This is similar to the findings in another helminth infection , wherein schistosome infected individuals showed reversal of memory CD4+ T cells after anthelmintic treatment [33] . The increased levels of naïve cells and decreased levels of central memory and effector CD4+ T cell numbers could potentially reflect an important component of the chronic immune response to Ss infection , especially as the effector compartment also undergoes expansion in size following chemotherapy . Moreover , protective immunity to S . stercoralis in mice requires CD4+ T cells but not CD8+ T cells [24] . Hence , it is not surprising to observe changes related to CD4+ T cell memory subsets in Ss infection . The common γc cytokines play an essential role in peripheral T cell expansion , function , and survival [34] and are also vital growth factors for T cells [11] . IL-2 is essential for the induction of TH2 cell differentiation [35 , 36] and is mainly expressed by T cells , primarily the CD4+ Th1 subsets and also by stimulated CD8+ T cells and dendritic cells ( DCs ) . IL-2 can induce the proliferation and survival of TCR-activated human and mouse T cells [37] and is required for sustained expansion of T cell populations [38] . In our previous study , we have shown that IL-2 levels have been significantly diminished in Ss infection when compared to uninfected individuals [16] . Our current study extends and corroborates these findings . IL-4 is the main cytokine necessary for the induction of TH2 cells . IL-4 also has an important role in allergy and immunoglobulin class switching [39] . Protective immunity to Ss larvae in mice is dependent on CD4+ T cells , and these cells typically produce IL-4 [24] . IL-9 is produced by a subset of activated CD4 T cells [40] and it provokes the activation of epithelial cells , B cells , eosinophils and mast cells [41] . IL-9 plays the role as T cell growth factor during the late phase of an immune response [42] . Animal studies revealed that Th9 cells have been associated in resistance against intestinal helminth infection [43] . We have previously examined the role of Th2 and Th9 cells in Ss infection [2 , 3] . In this study , we extend these findings and demonstrate that IL-4 and IL-9 levels were significantly enhanced in INF individuals . This confirms our previous data demonstrating elevated IL-4 and IL-9 responses in Ss infection and its reversal following anthelmintic therapy [16] . IL-7 is an important cytokine , an essential survival factor for T cells , plays a major role in T cell homeostasis , expansion of memory CD4+ T cells and proliferation of naïve and CD8+ T cells [10 , 44] . Our data showed diminished levels of IL-7 in Ss infected individuals when compared to uninfected individuals at baseline . Our data also revealed that IL-7 levels exhibited significant positive correlation with CD4+ central memory and effector memory subsets . IL-15 is a pleiotropic cytokine , which has different roles in the innate and adaptive immune system , including the development , activation , homing and survival of immune effector cells and antigen-independent expansion of naive and memory CD4+ and CD8+ T cells [45 , 46] . In the present study , IL-15 levels were also diminished in Ss infected individuals in comparison with uninfected individuals . There was a significant reversal of IL-7 and IL-15 levels following anthelmintic treatment . However , IL-15 did not exhibit any relationship with memory CD4+ T cell subsets . Thus , our data provides evidence that changes in the plasma levels of the common γc cytokines , IL-2 , IL-4 , IL-7 , IL-9 and IL-15 are associated with the differential memory T cell compartment alterations seen in Ss infected individuals . However , only IL-7 appears to exhibit a significant correlation . Our study also clearly depicts the alteration of the cytokine profile with treatment . IL-2 , IL-7 and IL-15 could possibly serve as lymphoid growth factors and could underlie novel strategies for immune recovery and the optimization of immune therapies in helminth infections . Our study has limitations in that we have not explored the functional significance of these changes in cellular subsets . However , it does provide impetus to further examine the function of these T cell subsets in Ss as well as other parasitic infections . Nevertheless , our work highlights the growing importance of these subsets and the role of common γc cytokines to parasitic infections . | Strongyloides stercoralis ( Ss ) , an intestinal nematode , the causative agent of strongyloidiasis , infects 30–100 million people worldwide . Ss infection is often clinically asymptomatic and long lasting due , in large part , to the parasites auto-infective life cycle and their ability to modulate the host immune system . Th1 cells are down modulated and Th2 cells are essential for fighting against helminth infections . T cells proliferate in response to common γc dependent cytokine signaling . The role of CD4+ and CD8+ T cell subset distribution and the association between memory T cell subsets and the common γc cytokines ( IL-2 , IL-4 , IL-7 , IL-9 and IL-15 ) in helminth infections has not been explored well . We examined the ex-vivo phenotypic profile of CD4+ and CD8+ T cell subsets and the circulating levels of common γc cytokines in Ss-infected , uninfected and post treatment individuals . Ss infected individuals showed alterations in the T cell subset distribution and these alterations were partially reversed following anthelminthic treatment . This was associated with altered plasma levels of IL-2 , IL-4 , IL-7 , IL-9 and IL-15 and partial reversal following anthelminthic treatment . IL-7 exhibited significant positive association with central and effector memory CD4+ T cells . Our study would provide stimulus to examine further about the function of T cell subset distribution and the role and association of common γc cytokines with parasitic infections . | [
"Abstract",
"Introduction",
"Material",
"and",
"methods",
"Results",
"Discussion"
]
| [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"immunology",
"parasitic",
"diseases",
"nematode",
"infections",
"developmental",
"biology",
"molecular",
"development",
"cytotoxic",
"t",
"cells",
"cell",
"enumeration",
"techniques",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"memory",
"t",
"cells",
"animal",
"cells",
"t",
"cells",
"immune",
"system",
"helminth",
"infections",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types"
]
| 2018 | Altered levels of memory T cell subsets and common γc cytokines in Strongyloides stercoralis infection and partial reversal following anthelmintic treatment |
Estrogen receptors ( ER ) are important regulators of metabolic diseases such as obesity and insulin resistance ( IR ) . While ERα seems to have a protective role in such diseases , the function of ERβ is not clear . To characterize the metabolic function of ERβ , we investigated its molecular interaction with a master regulator of insulin signaling/glucose metabolism , the PPARγ , in vitro and in high-fat diet ( HFD ) -fed ERβ -/- mice ( βERKO ) mice . Our in vitro experiments showed that ERβ inhibits ligand-mediated PPARγ-transcriptional activity . That resulted in a blockade of PPARγ-induced adipocytic gene expression and in decreased adipogenesis . Overexpression of nuclear coactivators such as SRC1 and TIF2 prevented the ERβ-mediated inhibition of PPARγ activity . Consistent with the in vitro data , we observed increased PPARγ activity in gonadal fat from HFD-fed βERKO mice . In consonance with enhanced PPARγ activation , HFD-fed βERKO mice showed increased body weight gain and fat mass in the presence of improved insulin sensitivity . To directly demonstrate the role of PPARγ in HFD-fed βERKO mice , PPARγ signaling was disrupted by PPARγ antisense oligonucleotide ( ASO ) . Blockade of adipose PPARγ by ASO reversed the phenotype of βERKO mice with an impairment of insulin sensitization and glucose tolerance . Finally , binding of SRC1 and TIF2 to the PPARγ-regulated adiponectin promoter was enhanced in gonadal fat from βERKO mice indicating that the absence of ERβ in adipose tissue results in exaggerated coactivator binding to a PPARγ target promoter . Collectively , our data provide the first evidence that ERβ-deficiency protects against diet-induced IR and glucose intolerance which involves an augmented PPARγ signaling in adipose tissue . Moreover , our data suggest that the coactivators SRC1 and TIF2 are involved in this interaction . Impairment of insulin and glucose metabolism by ERβ may have significant implications for our understanding of hormone receptor-dependent pathophysiology of metabolic diseases , and may be essential for the development of new ERβ-selective agonists .
The estrogen receptors ( ERs ) are members of the nuclear hormone receptor family ( NHR ) which act as eukaryotic ligand-dependent transcription factors . ERs are involved in the regulation of embryonic development , homeostasis and reproduction . Two major estrogen receptors , alpha and beta ( ERα and ERβ ) , convey the physiological signaling of estrogens ( 17β-estradiol , E2 ) [1] . Additionally , ERs are activated by specific synthetic ligands such as raloxifene , tamoxifen , the ERβ-specific ligand diarylpropionitrile ( DPN ) , and the ERβ-specific agonist propylpyrazole-triol ( PPT ) , which belong to the group of selective estrogen receptor modulators ( SERMS ) [2]–[4] . The prevalence of metabolic diseases such as obesity , insulin resistance and type 2 diabetes has increased dramatically during the recent ten years [5] . Gender differences in the pathophysiology of obesity and metabolic disorders are well established [6]–[8] . However , the molecular mechanisms of sexual dimorphism in metabolic diseases are largely unknown . In addition , lack of ER activation has been implicated in postmenopausal impairment of glucose and lipid metabolism , resulting in visceral fat distribution , insulin resistance and increased cardiovascular risk after menopause [9] . In this context the investigation of ER-signaling and its role in metabolic disorders has gained increasing attention [4] , [8] . To identify the ER subtype involved in the regulation of metabolic disorders , studies have been carried out in ER-deficient mice . ERα-deficient ( αERKO ) mice have profound insulin resistance and impaired glucose tolerance [10]–[13] . These studies indicate that ERα has a protective role in metabolic disorders by improving insulin sensitivity and glucose tolerance . The metabolic function of ERβ is not clear . ERβ knockout mice ( βERKO ) have a similar body weight and equal fat distribution in comparison to wild type littermates . Additionally , βERKO and wild-type ( wt ) mice exhibit similar insulin and lipid levels [14] . However , previous studies in βERKO mice were only carried out under low fat diet , which may have concealed a phenotype relevant for human obesity normally induced by high-energy/fat diet . The peroxisome proliferator-activated receptor gamma ( PPARγ ) belongs to the NHR family and is a major regulator of glucose and lipid metabolism by modulating energy homeostasis in adipose tissue , skeletal muscle and liver [15]–[17] . Glitazones or thiazolidinediones ( TZDs ) are high-affinity PPARγ agonists , and act as insulin sensitizers . TZDs induce adipogenesis and adipose tissue remodeling followed by an improvement of glucose tolerance [18] . The role of PPARγ in the control of glucose homeostasis expands beyond its primary action in adipose tissue , and involves the regulation of adipocytokine production such as adiponectin , leptin , and resistin [19]–[21] . Consistently , reduced PPARγ activity has important metabolic and cardiovascular pathophysiological consequences leading to insulin resistance , diabetes and end organ damage [15] . The molecular mechanisms underlying PPARγ function are similar to those of ER-signaling . In a basal state , PPARγ , similar to ERs , is bound to corepressor proteins such as nuclear receptor corepressor ( NCoR ) or silencing mediator of retinoic acid and thyroid hormone receptor ( SMRT ) [22] . After binding within the ligand binding domain ( LBD ) , PPARγ ligands induce its heterodimerization with retinoid x receptor alpha ( RXRα ) , and its subsequent interaction with co-activators like steroid receptor coactivators ( SRCs ) followed by binding to PPARγ response elements ( PPREs ) within target gene promoters [23] . Importantly , PPARγ is sharing a similar pool of cofactors with ERβ which provides a platform for mutual interactions between these two NHRs [23] , [24] . To study the crosstalk between ERβ and PPARγ , we investigated the regulation of PPARγ-mediated transcriptional activity by ERβ . Our in-vitro experiments in 3T3-L1 preadipocytes showed that ERβ inhibits ligand-mediated PPARγ-transcriptional activity . That resulted in the blockade of PPARγ-induced adipocytic gene expression and in decreased adipogenesis . Overexpression of nuclear coactivators such as steroid receptor coactivator 1 ( SRC1 ) and transcriptional intermediary factor 2 ( TIF2 ) prevented the ERβ-mediated inhibition of PPARγ activity , whereas the presence of vitamin D receptor ( VDR ) -interacting protein 205 ( DRIP205 ) or PPARγ coactivator-1alpha ( PGC1α ) had no effect indicating a role for distinct nuclear coactivators for ERβ-PPARγ interaction in-vitro . High fat diet ( HFD ) -fed βERKO mice showed increased body weight and fat mass . In contrast , triglyceride content in liver and muscle was decreased in βERKO mice , which was associated with a marked improvement of hepatic and muscular insulin signaling . Compared to wt , βERKO mice demonstrated improved systemic insulin sensitivity and glucose tolerance . In consonance with the metabolic phenotype and with the in-vitro data , βERKO mice exhibited augmented PPARγ signaling in adipose tissue corresponding to increased food efficiency and significantly elevated RQ ( respiratory quotient ) . Blockade of adipose PPARγ signaling in βERKO mice by PPARγ antisense oligonucleotide injection resulted in a reversal of the βERKO phenotype including body weight reduction and impairment of insulin sensitivity . In summary , the present data demonstrate that ERβ impairs insulin and glucose metabolism which may , at least in part , result from a negative cross-talk with adipose PPARγ .
In order to demonstrate a molecular interaction between PPARγ and ERβ in a metabolically relevant cell system , we first investigated ligand-dependent PPARγ activity in the presence of ERβ in 3T3-L1 preadipocytes . Cells were treated with the PPARγ-agonist pioglitazone ( 10 µM ) , with or without additional E2 stimulation , and PPARγ activation was measured using pGal4-hPPARγDEF/pG5TkGL3 luciferase assay [25] . Upon pioglitazone stimulation , 3T3-L1 preadipocytes showed pronounced PPARγ activation ( bar 1+2 , Figure 1A ) . This activation was not affected by co-treatment with ligands for ERβ such as E2 ( bar 2 vs . 3 , Figure 1A ) or DPN ( data not shown ) . Overexpression of ERβ led to a marked inhibition of ligand-dependent PPARγ activity ( bar 2 vs . 4+6+8 , Figure 1A ) which was also corroborated in a PPARγ response element ( PPRE ) luciferase assay ( Figure S1 ) . This inhibition was E2 ( bar 4+6+8 vs . 5+7+9 , Figure 1A ) and DPN independent ( data not shown ) . The inhibitory effect of ERβ seemed to be isoform specific , since ERα overexpression resulted in no inhibition of PPARγ activity ( bar 11 , Figure 1A and Figure S2 ) . To further explore the regulation of PPARγ by ERβ , we performed additional experiments coexpressing an activation function 1 domain ( AF-1 ) deleted-ERβ construct in 3T3-L1 cells . Overexpression of this truncated form of ERβ which still contains a functional ligand binding domain ( LBD ) did not reduce PPARγ activity indicating that ERβ AF-1 is necessary for regulation of PPARγ by ERβ ( bar 10 , Figure 1A ) . To assure adequate overexpression and function of ERβ in our system , 3T3-L1 preadipocytes were transiently transfected with ERβ followed by Western blot analysis and transactivation assays using ER response elements ( ERE ) -luciferase system ( Figure 1B , C ) . Both assays confirmed adequate expression and function of ERβ . While our data implicated a negative regulation of ligand-mediated PPARγ transcription by ERβ , we next investigated the regulation of PPARγ-dependent gene expression during 3T3-L1 preadipocyte differentiation . The preadipocytes were transfected with indicated plasmids and differentiated for 3 days using standard differentiation medium [25] . As the full differentiation procedure requires 7-10 days of treatment , the observed effect on fat droplet accumulation and expression pattern are typical for early phase of adipocyte differentiation . The 3T3-L1 cells transfected with ERβ and differentiated for 3 days showed reduced adipogenesis visualized by fat droplet accumulation in comparison to control cells ( Figure 2A ) . Low levels of ERβ could also be detected in untransfected 3T3-L1 cells and its expression was slightly elevated during differentiation ( data not shown ) underlining the physiological importance of our findings . Overexpression of the ERα isoform in these cells did not show any inhibitory effect on preadipocyte differentiation ( Figure 2A ) . The adipocyte protein 2 ( aP2 ) gene belongs to the classical PPARγ-regulated genes involved in the early phase of adipogenesis [26] . The expression level of aP2 measured by real-time PCR was significantly elevated in the differentiated control cells ( bar 2 vs . 1 , Figure 2B ) . Overexpression of ERβ-but not ERα- in these cells led to a significant reduction of aP2 expression ( bar 2 vs . 4 and 6 , Figure 2B ) indicating that endogenous PPARγ activation in 3T3-L1 cells was inhibited by ERβ . Furthermore pioglitazone ( 10 µM ) treatment of 3T3-L1 cells overexpressing PPARγ/RXRα showed increased adipogenesis , an effect that was markedly inhibited by coexpression of ERβ ( Figure 2C ) . aP2 expression level was also significantly reduced in cells co-expressing ERβ together with PPARγ/RXRα ( bar 2 vs . 3 , Figure 2D ) . These data indicate that ERβ inhibits PPARγ-transcriptional activity resulting in the blockade of PPARγ-induced adipocytic target gene expression and amelioration of adipogenesis . To investigate ERβ's action on PPARγ in vivo , we studied PPARγ activity and PPARγ target genes in HFD-fed βERKO and wt mice . βERKO mice and their wt littermates were fed HFD containing 60% calories from fat for 12 weeks followed by the analysis of PPARγ-dependent gene expression in gonadal fat tissue . Adipose mRNA expression of PPARγ target genes involved in triglycerides ( TG ) synthesis such as lipoprotein lipase ( Lpl ) , phosphoenolpyruvate carboxykinase ( PEPCK ) and CD36 was significantly upregulated in βERKO mice ( Figure 3 A–C ) . Key mediators of insulin and glucose metabolism such as the retinol-binding protein 4 ( RBP4 ) were also regulated in βERKO mice ( Figure 3D ) . Consistently with these findings , adiponectin mRNA expression and adiponectin serum levels were elevated in βERKO mice ( Figure 3E , F ) . No difference of PPARγ target gene regulation between βERKO and wt mice was observed in liver ( data not shown ) . Positive regulation of a series of adipose PPARγ target genes in βERKO mice suggested a general induction of PPARγ transcription in βERKO mice . To prove this , we performed EMSA assays in gonadal fat from βERKO and wt mice after 12 weeks on HFD . Nuclear fractions isolated from adipose tissues from βERKO mice showed an increased binding/activation of endogenous PPARγ in comparison to wt mice ( line 4–7 vs . 1–3 , Figure 3G ) in the presence of similar PPARγ expression levels , as shown by real-time RT-PCR analysis and Western Blot ( Figure 3G ) . Increased adipose PPARγ target gene expression and PPARγ-DNA binding confirmed an augmented PPARγ signaling in adipose tissue from βERKO mice . To exclude the possibility that the augmented expression of PPARγ target genes measured in HFD-fed βERKO is the result of increased adipose tissue mass , we performed experiments using ex-vivo fat pads isolated from wt and βERKO mice , treated for 24h with 10 µM pioglitazone or vehicle-control , followed by analysis of PPARγ target gene expression using real-time RT-PCR . In this system augmented ligand-induced PPARγ target gene expression mainly results from enhanced PPARγ transcriptional activity and not from increased fat mass . The expression level of PEPCK and Lpl was significantly increased in both wt and βERKO fat pads under pioglitazone treatment ( bar 1 vs . 2 and bar 3 vs . 4 Figure 4 A and B ) . However , pioglitazone-induced PPARγ target gene expression was markedly elevated in βERKO mice compared to wt mice , indicating an augmented PPARγ signaling in the absence of ERβ ( bar 2 vs . 4 , Figure 4 A and B ) . To further characterize ERβ ligand dependency for its interaction with PPARγ in the mouse model , additional in-vivo studies were performed in estrogen-depleted , ovariectomized wt mice treated with the ERβ-ligand DPN . Analysis of PPARγ target genes ( Lpl , PEPCK , CD36 and adiponectin ) in gonadal fat isolated from these mice revealed no significant differences in the expression level between vehicle and DPN-treated rodents indicating ligand independency ( Figure 4C ) . These data are consistent with the in-vitro study in 3T3-L1 preadipocytes , where PPARγ activation was not affected by co-treatment with ligands for ERβ such as E2 ( bar 2 vs . 3 , Figure 1A ) or DPN ( data not shown ) . Given the central role of PPARγ in insulin and glucose metabolism , the metabolic phenotype of βERKO mice was assessed . No difference in fasting/fed blood glucose food intake , and mean arterial blood pressure was observed between βERKO and wt mice under HFD ( Table 2 ) . Body weight gain was significantly enhanced in βERKO mice , compared to wt mice ( mean BW difference βERKO vs . wt mice after 12 week HFD: 3+/−0 . 4 g , p<0 . 05 , Figure 5A ) . Increased body weight in βERKO mice resulted from increased adipose tissue mass . MRI-analysis of body composition demonstrated significantly higher fat mass in βERKO mice compared to wt littermates ( Figure 5B ) , and fat pad weight from gonadal and perirenal depots was increased ( Table 1 ) . In contrast , liver weight was significantly reduced in βERKO mice in comparison to wt control littermates ( Table 1 ) . Reduced hepatic weight likely resulted from decreased TG-accumulation assessed by H/E-staining of liver tissue sections ( Figure 5C ) , and by TG quantification in dried liver tissue ( Figure 5D ) . In accordance with reduced hepatic TG-content , hepatic insulin signaling was improved . After injection of insulin in the portal vein , liver tissue was dissected and proteins were isolated for Western blot analysis . Insulin-stimulated Akt phosphorylation was enhanced in βERKO mice ( Figure 5E and Figure S3 ) . In parallel to decreased TG levels in liver , βERKO mice had decreased muscular TG-accumulation under HFD and improved insulin signaling ( Figure 5F , G , and Figure S3 ) . Skeletal muscle and liver are the major insulin responsive tissues , and important sites of glucose metabolism in-vivo . An important mechanism of PPARγ-mediated insulin sensitization involves adipose tissue remodeling and trapping of circulating triglycerides ( TG ) which protects the liver and skeletal muscle against TG overload . Increased adipose tissue mass in βERKO mice may protect these animals against TG-overload in liver and skeletal muscle resulting in an improvement of hepatic and muscular insulin sensitivity . Next we investigated insulin and glucose metabolism in βERKO and wt mice . Whole body glucose disposal was assessed using an oral glucose tolerance test ( OGTT ) ( Figure 6A ) . Following an oral glucose challenge βERKO mice on HFD had moderately but significantly improved glucose tolerance compared to HFD-fed wt mice ( Figure 6A , B ) . In addition insulin sensitivity measured by an insulin tolerance test ( ITT ) was improved in comparison to wt mice ( Figure 6C , D ) . No difference in fasting and fed blood glucose was observed between βERKO and wt mice under HFD ( Table 2 ) . Despite an increased fat mass in βERKO mice , systemic insulin sensitivity and glucose tolerance were significantly improved under HFD when compared to wt-control . To further examine the enhanced weight gain and fat deposition in βERKO mice , we performed indirect calorimetry and monitored food consumption . Food intake did not differ between wt-control and βERKO mice ( Table 2 ) . However , deletion of ERβ resulted in a marked increase of food efficiency ( ratio of weight gain and food intake , Figure 6E ) . No significant difference in O2 consumption ( Figure 6F ) , energy expenditure ( Table 2 ) , or locomotor activity ( Table 2 ) was detected between βERKO and wt mice . Low RQ values have previously been described for rodents under HFD and in diabetes [27] . Both wt and βERKO mice exhibited low RQ values . βERKO mice had a significantly higher RQ when compared to wt-controls which may be indicative for attenuated fatty acid ( FA ) oxidation promoting fat accumulation ( Figure 6G ) . These data show that βERKO mice are partially protected against HFD induced insulin resistance . Increased fat mass may likely result from increased food efficiency based on reduced oxidative utilization of fat and increased fat storage . The metabolic phenotype of βERKO mice including increased fat mass , reduced hepatic/muscular TG and improved systemic insulin sensitivity exhibits high similarity to augmented PPARγ activation e . g . under thiazolidinedione ( TZD ) treatment [28] , [29] . To directly demonstrate the role of PPARγ in HFD-fed βERKO mice , PPARγ signaling was disrupted by intraperitoneal ( i . p . ) injection of PPARγ antisense oligonucleotide ( ASO ) . HFD-fed βERKO mice were injected twice a week for 6 weeks with either PPARγ ASO or control oligonucleotides . PPARγ expression was significantly reduced in liver of ASO-treated βERKO mice , similar to previously reported results in apoB/BATless mice ( data not shown ) [30] . However , suppression of hepatic PPARγ by ASO injection is unlikely to play an important role in our model , since hepatic PPARγ signaling did not differ between wt and βERKO mice , respectively . More importantly , i . p . application of PPARγ ASO in βERKO mice resulted in 63±4 . 8% ( p<0 . 05 ) reduction of PPARγ expression in gonadal adipose tissue compared to βERKO mice injected with control oligonucleotides ( Figure 7A ) . Accordingly , expression of the PPARγ target genes Lpl , PEPCK , CD36 , and adiponectin was markedly decreased in adipose tissue from PPARγ ASO-injected βERKO mice , and adipocyte diameters were increased ( Figure 7A , G ) . These data corroborate a relevant reduction of adipose PPARγ signaling by ASO intervention . Body weight gain and gonadal fat accumulation in HFD-fed-βERKO mice were significantly attenuated by PPARγ-ASO injection ( Figure 7B , C ) . Finally , blockade of adipose PPARγ by ASO led to reversal of the improved insulin response observed in βERKO mice , and to an impairment of insulin sensitivity and glucose tolerance ( Figure 7D–F ) . Together these data underline the importance of adipose PPARγ signaling for the metabolic phenotype observed in βERKO mice . Nuclear coactivators such as SRC1 and TIF2 are important mediators of ERβ and PPARγ-induced transcriptional activation . It has previously been shown that competition of distinct nuclear receptor ( NR ) for coactivator binding results in a negative cross-talk between NRs [31] . To prove whether common coactivators are involved in ERβ-PPARγ interactions , SRC1 , TIF2 , DRIP205 or PGC1α were co-expressed together with ERβ and ligand induced PPARγ activation was measured . Overexpression of SRC1 and TIF2 prevented the ERβ-mediated inhibition of PPARγ activity ( Figure 8A , B ) whereas the presence of DRIP205 ( Figure 8C ) and PGC1α ( Figure S4 ) had no effect . To demonstrate that SRC1 and TIF2 are also involved in ERβ-PPARγ interaction in-vivo , we performed ChIP experiments with gonadal fat from HFD-fed βERKO and wt mice . The adiponectin promoter was selected as a PPARγ-target promoter . Binding of SRC1 and TIF2 to the adiponectin promoter was enhanced in gonadal fat from βERKO mice ( Figure 8D ) , indicating that the absence of ERβ in adipose tissue results in exaggerated coactivator binding to a PPARγ target promoter . Together these data suggest that the coactivators SRC1 and TIF2 are involved in the negative regulation of PPARγ by ERβ in vitro and in vivo .
The present study demonstrates that ERβ is a negative regulator of ligand-induced PPARγ activity in-vitro . Consequently , data from βERKO mice suggest that ERβ negatively regulates insulin and glucose metabolism which may , at least in part , result from an impairment of regular adipose tissue function based on a negative cross-talk between ERβ and PPARγ . Loss of ERβ resulted in enhanced body weight gain and fat accumulation in HFD-fed mice . However , absence of ERβ prevented hepatic/ muscular triglyceride overload , preserved regular insulin signaling in liver/ skeletal muscle , and improved whole-body insulin sensitivity and glucose tolerance under HFD . This metabolic phenotype strongly suggested augmented PPARγ signaling in mice lacking ERβ . And indeed , PPARγ target genes and PPARγ-DNA binding were markedly induced in gonadal fat from βERKO mice . Along this line , blockade of adipose PPARγ signaling by PPARγ ASO injection reversed the metabolic changes in βERKO mice . A mutual signaling cross-talk between ERs and PPARγ has been described previously . PPARγ together with its heterodimeric partner RXRα has been shown to suppress ER-induced target gene expression through competitive binding to an ERE site in the vitellogenin A2 promoter [32] . In accordance with a bidirectional interaction , Wang and colleagues demonstrated that ERs are capable of inhibiting ligand-induced PPARγ activation in two different breast cancer cell lines [33] . In contrast to our results , these authors show that basal and agonist-stimulated PPRE-activity is also blocked by ERα . Transcriptional activity of PPARγ differs markedly depending on the cell system and tissues . The highest level of PPARγ-mediated transcription has been described in adipocytes and adipocytic cell lines , where molecular conditions such as cofactor availability seemed to be optimized [34] . Compared to adipocytes , breast cancer cells exhibit low PPARγ expression and activity reflected by a less than 2-fold induction of PPRE-activity after ligand stimulation [33] . The presence of PPARγ suppression by ERα in breast cancer cells might be a result of weak basal PPARγ transcriptional activity in these cells . In contrast , the pronounced activation of the exogenous PPARγ LBD in 3T3-L1 preadipocytes may require more potent inhibitory stimuli which could not be achieved by ERα overexpression in our system . Suppression of PPARγ-LBD activation by ERβ did not depend on ERβ ligands which is consistent with previous reports [33] . Also our in vivo studies in estrogen-depleted , ovariectomized wt mice treated with the ERβ-ligand DPN indicate that PPARγ-ERβ interaction is ligand independent . More importantly , overexpression of a truncated form of ERβ containing solely the ERβ-LBD/ AF2 domain did not induce any inhibitory effect on PPARγ suggesting an important role of ERβ's NH2-terminal AF1 domain for ERβ-PPARγ interactions . Consistently , activity of the ER-AF1 domain is usually not dependent on ligand activation [35] . Furthermore , Tremblay and coworkers demonstrated that ERβ-AF1 activation involves ligand-independent recruitment of SRC-1 , a cofactor involved in ERβ-PPARγ interactions in our study [36] . These data corroborate our observation that PPARγ suppression by ERβ involves the AF1 domain and ligand-independent interactions with the coactivators SRC1 and TIF2 . Repression of PPARγ activity through ERβ was reversed by titration of the p160 coactivators , SRC1 and TIF2 , suggesting that the suppressive action of ERβ is a result of p160 coactivator interaction with ERβ thereby preventing the binding of PPARγ to the same coactivators . Similar interactions have been described previously for ER interaction with the thyroid receptor [31] . The present study demonstrates for the first time that ERβ impairs insulin sensitivity and glucose tolerance under HFD implicating pro-diabetogenic actions of this receptor . In consonance , we could recently demonstrate that ERβ has a suppressive role on glucose transporter 4 ( GLUT4 ) expression in skeletal muscle [8] , [37] . GLUT4 has been identified as the major mediator of insulin-induced glucose uptake in fat and skeletal muscle . In addition , removal of the E2-ERβ signaling by ovariectomy in ERα-deficient mice improved glucose and insulin metabolism supporting the diabetogenic effect of ERβ [12] . Loss of ERβ resulted in a marked augmentation of adipose PPARγ activity in our model indicating that ERβ mediates its metabolic actions by a negative interaction with PPARγ in adipose tissue . This concept is corroborated by a number of observations . HFD-fed βERKO mice exhibited increased adipose tissue mass in the presence of improved insulin sensitivity and glucose tolerance . These metabolic changes are usually observed under chronic PPARγ stimulation [17] . PPARγ has been identified as an essential regulator of whole-body insulin sensitivity . Two major mechanisms have been described: ( 1 ) Adipose PPARγ protects non-adipose tissue against excessive lipid overload and maintains normal organ function and insulin responses ( liver , skeletal muscle ) by preserving regular adipose tissue function , and ( 2 ) Adipose PPARγ guarantees a balanced and adequate production of adipocytokine secretion such as adiponectin from adipose tissue , factors which are important mediators of insulin action in peripheral tissues [38]–[40] . Both processes could be observed in βERKO mice . Further support of this notion comes from clinical actions of anti-diabetic PPARγ agonists ( TZD ) [28] , [29] . Activation of PPARγ by TZDs in diabetic patients resembles the phenotype of βERKO mice including improved insulin sensitization and glucose tolerance in the presence of weight gain . We also observed increased food efficiency and changes in nutrient partitioning reflected by an increased RQ in βERKO mice . Loss of ERβ appears to result in attenuated fatty acid ( FA ) oxidation which may favor the storage of TGs in adipose tissue and increased fat accumulation , and may provide a possible explanation for the enhanced weight gain . Interestingly , treatment of obese mice with a synthetic PPARγ agonist has been shown to mediate similar changes including an increase in food efficiency and higher RQ values [41] . Finally , blockade of PPARγ signaling in adipose tissue of βERKO mice resulted in a reversal of the metabolic phenotype corroborating the importance of adipose PPARγ in the present model . The observed suppression of hepatic PPARγ activity by ASO injection is unlikely to play a major role since the initial metabolic characterization of untreated βERKO mice under HFD did not reveal any dysregulation of hepatic PPARγ signaling . In summary , the metabolic phenotype of βERKO mice is mediated by an augmented adipose PPARγ action , which implies that in the presence of ERβ , PPARγ activity might be partially suppressed . The notion , that ERβ-PPARγ crosstalk requires receptor-p160 interaction , was underlined by our observations in WAT from βERKO mice . Binding of SRC1 and TIF2 to the PPARγ-regulated adiponectin promoter in WAT was enhanced in the absence of ERβ . It has recently been demonstrated that p160 coactivators are important regulators of PPARγ transcriptional activity in WAT [42] . In particular , TIF2 has been identified as a nuclear coactivator involved in the adipogenic actions of PPARγ . Future experiments are required to define the functional relevance of TIF2 and SRC1 in our model . So far one may conclude that the metabolic phenotype of HFD-fed βERKO mice is , at least in part , explained by increased adipose PPARγ activity as a result of exaggerated binding of p160 coactivators to PPARγ-regulated target gene promoters . Diabetogenic actions of ERβ are of major significance for the pharmaceutical development of new ERβ-selective agonists intended for use against a multitude of diseases such as rheumatoid arthritis or postmenopausal osteoporosis [43] , [44] . Despite the high tissue selectivity of such compounds , and despite the fact that the actions observed in our study were ligand-independent , one has to be aware of the potentially deleterious actions of ERβ on insulin- and glucose metabolism . As a precautionary measure metabolic profiling of new ERβ agonist should be performed . Collectively , our data provide first evidence that ERβ negatively regulates insulin signaling and glucose metabolism that involves an impairment of regular adipose PPARγ function . Moreover our data suggest that the coactivators SRC1 and TIF2 are involved in this inhibition . In consonance , impairment of insulin and glucose metabolism by ERβ has significant implications for our understanding of hormone receptor-dependent pathophysiology of metabolic diseases , and is essential for the development of new ERβ-selective agonists .
Female estrogen receptor β -/- mice ( βERKO ) received from J . -A . Gustafsson ( Karolinska Institutet , Huddinge , Sweden ) and their wt littermates were housed in a temperature controlled ( 25°C ) facility with a 12-h light/dark cycle and genotyped using genomic DNA isolation kit ( Invitek ) and PCR primers described elsewhere [45] . 4–5 week old mice were fed ad libitum with a high-fat diet ( 60% kcal from fat , [25] ) for 12 weeks . Body weight and food intake were determined throughout the experiment . At start and end of treatment , body composition was determined by nuclear magnetic resonance imaging ( Bruker's Minispec MQ10 ) . After 12 weeks' treatment , blood samples were collected from overnight-fasted animals by retroorbital venous puncture under isoflurane anesthesia for analysis of serum adiponectin ( mouse-adiponectin ELISA; Linco Research ) and glucose ( colorimetric glucose test; Cypress Diagnostics ) . An OGTT using a dose of 2 g/kg body weight ( BW ) glucose and ITT with intraperitoneally injected 0 . 5 units/kg BW insulin ( Actrapid; Novo Nordisk ) were performed . Tail vein blood was used for glucose quantification with a glucometer ( Precision Xtra; Abbott ) . Blood pressure was measured invasively in the abdominal aorta using a solid-state pressure transducer catheter ( Micro-Tip 3F; Millar Instruments ) under isoflurane anesthesia . Afterwards animals were killed and organs were dissected . For immunohistochemical studies organs were fixed in 4% formalin , embedded in paraffin and stained with Haematoxylin/Eosin ( H&E ) ; for RNA , Western blot analysis and measurement of TG content isolated organs were frozen in liquid nitrogen; for EMSA and Chromatin IP assays abdominal fat was stored in ice-cold PBS with proteinase inhibitors ( Complete Mini , Roche ) , and immediately proceeded as described below . For DPN- treatment , 10 week old female C57BL/6J mice were ovariectomized , and after 1 week recovery set on soy-free diet . Subsequently mice were treated for 21 days with DPN ( 8 mg/kg ) or vehicle administered using subcutaneous pellets ( Innovative Research of America ) . Afterwards animals were killed under isoflurane anesthesia and organs were dissected . All animal procedures were in accordance with institutional guidelines and were approved . ASO complementary to murine PPARγ ( Gen-BankTM accession number U09138 . 1 ) , ISIS 141941 , 5′-AGTGGTCTTCCATCACGGAG-3′ , and ASO control , ISIS 141923 , 5′-CCTTCCTGAAGGTTCCTCC-3′ was generously provided by ISIS Pharmaceuticals ( Carlsbad , CA , U . S . A . ) . Both ASO's were injected intraperitoneally twice a week into 6 week-old female βERKO mice ( n = 7 per group ) . Injections were continued over 6 weeks at a dose of 100 mg/kg/week as described previously [30] . At the end of the experiment animals were metabolically phenotyped as described above . After HFD feeding , βERKO mice and their wt littermates were analyzed for energy expenditure , RQ , and locomotor activity using a custom-made 4-cage calorimetry system ( LabMaster; TSE Systems ) . The instrument consists of a combination of highly sensitive feeding and drinking sensors for automated online measurement . The calorimetry system is an open-circuit system that determines O2 consumption , CO2 production , and RQ . A photobeam-based activity monitoring system detects and records every ambulatory movement , including rearing and climbing movements , in every cage . All the parameters can be measured continuously . Mice ( n = 7 per group ) were placed in the calorimetry system cages for 24h . Tissue samples from gonadal fat were prepared from female wt and βERKO mice . Explanted gonadal fat samples were washed 3 times with ice-cold Hanks Balanced Salt Solution ( HBSS ) and treated for 24h with 10 µM pioglitazone or vehicle in Dulbecco's modified Eagle's medium F2 ( DMEM:F12 , Invitrogen ) . Afterwards tissue samples were washed with ice-cold PBS and RNA extraction was performed using trizol ( Invitrogen ) . 3T3-L1 preadipocytes were purchased from the American Type Culture Collection . Preadipocytes were cultured in Dulbecco's modified Eagle's medium with 10% Fetal Bovine Serum ( FBS ) and 1% Pen-Strep ( Invitrogen ) . For differentiation experiments preadipocytes were grown to confluence and after 12h culture medium was supplemented with methylisobutylxanthine ( 0 . 5 mM ) , dexamethasone ( 0 . 25 µM ) , and insulin ( 1 µg/ml ) in DMEM containing 10% FBS for 72h [25] . Afterwards cells were washed with ice-cold PBS and RNA extraction was performed using trizol ( Invitrogen ) according to the manufacturer's instructions . For the staining procedure differentiated cells were washed twice with ice-cold PBS , fixed with 4% PFA , and stained for 1h at room temperature with Oil-red-O solution . Transient transfection and luciferase assays were performed as previously described [25] . Briefly 3T3-L1 cells were plated in 12-well plates and transfected using lipofectamine 2000 and OptiMEM ( Invitrogen ) with 100 ng pGal4-hPPARγDEF; 400 ng pG5TkGL3 , TIF2-pSG5 , DRIP205-pSG5 ( kindly provided by B . Staels , Institut Pasteur de Lille , France ) , 5 ng pRL-CMV , a renilla luciferase reporter vector ( Promega ) , hPPARγ2-pSG5 and hRXRα-pCDNA , pSG5 ( Stratagene ) , hSRC1-pSG5 ( kindly provided by M . Parker , Institute of Reproductive and Developmental Biology , Imperial College London , United Kingdom ) , pERE-TkGL3 ( kindly provided by P . J . Kushner , Metabolic Research Unit and Diabetes Center , University of California , San Francisco , USA ) , hERα-pSG5 and ERβ-pSG5 ( kindly provided by P . Chambon , Institut Clinique de la Souris , Illkirch Cedex , France ) , and PGC1α kindly provided by Addgene , USA . Delta AF1-ERβ-pSG5 was cloned from full length ERβ-pSG5 . After 3h of transfection cells were washed , left for 12h in serum deprived medium ( 0 . 5% FCS , 1% Pen-Strep ) , and afterwards treated for 24h with 10 µM pioglitazone ( Takeda Pharmaceutical Co . Ltd , Japan ) or vehicle ( DMSO ) . When treated with E2 or specific ERβ agonist diarylpropionitrile ( DPN ) , cells were cultivated in phenol red free DMEM and coal-striped FCS . Luciferase activity was measured 36 h after transfection using the dual-luciferase reporter assay system ( Promega ) . Transfection experiments were performed in triplicate and repeated at least three times . Total RNA from cultured preadipocytes , abdominal fat tissue and skeletal muscle was isolated using trizol ( Invitrogen ) according to the manufacturer's instructions . For real-time PCR analysis RNA samples were DNAse digested ( Invitrogen ) , reverse transcribed using Superscript ( Promega ) , RNasin ( Promega ) , dNTPs ( Invitrogen ) , according to the manufacturer's instructions , and used in quantitative PCR reactions in the presence of a fluorescent dye ( Sybrgreen , BioRad ) . Relative abundance of mRNA was calculated after normalization to 18S ribosomal RNA . Primer sequences are provided in Table S1 . For Western blot detection of ERβ cells were grown on Φ10 cm plates and transfected with increasing amount of ERβ plasmid or empty vector control . After 24h cells were harvested and WB analysis was performed as following: cells ( and tissues for Akt analysis ) were lysed in RIPA buffer ( 50 mM Tris pH 7 . 5 , 150 mM NaCl , 5 mM MgCl2 , 1% Nonidet P-40 , 2 . 5% glycerol , 1 mM EGTA , 50 mM NaF , 1 mM Na3VO4 , 10 mM Na4P2O7 , 100 µM phenylmethylsulfonyl fluoride with proteinase inhibitors ( Complete Mini , Roche ) . Lysates ( tissues ( 30 µg ) and cells ( 20 µg ) ) were analyzed by immunoblotting using antibody raised against ERβ ( H-150 , Santa Cruz ) , antibody raised against pS473- Akt and total-Akt ( Cell Signalling ) , and secondary horseradish-conjugated antibodies ( Amersham ) . For PPARγ immunoblotting , 20 µg of nuclear fractions used for EMSA were analyzed using antibody raised against PPARγ ( E-8 , Santa Cruz ) . For detection , enhanced chemiluminescent substrate kit ( Amersham ) was used . Nuclear extracts were prepared by using a nonionic detergent method as described previously [46] . The inputs were normalized for protein contents , as ERβ-deficient mice have increased fat tissue mass . Detection of PPARγ was performed with a [32P] γATP-labeled PPRE oligo ( 5′-CAAAACTAGGTCAAAGGTCA-3′ 5′- TGACCTTTGACCTAGTTTTG-3′ ) . The DNA binding reactions were performed with 40 µl of binding buffer ( 20 µg nuclear extracts , 2 µg of poly ( dI-dC ) , 1 µg of bovine serum albumin ( BSA ) , 5 mM dithiothreitol ( DTT ) , 20 mM HEPES , pH 8 . 4 , 60 mM KCl , and 10% glycerol ) for 30 min at 37°C . For competition experiments , a cold oligonucleotide probe was used . The reaction products were analyzed via 5% polyacrylamide gel electrophoresis using 12 . 5 mM Tris , 12 . 5 mM boric acid , and 0 . 25 mM EDTA , pH 8 . 3 . Gels were dried and exposed to Amersham TM film ( Amersham Pharmacia Biotech ) at −80°C using an intensifying screen . Abdominal fat tissue ( gonadal fat ) isolated from wt and βERKO mice was washed in ice-cold PBS with proteinase inhibitors ( Complete Mini , Roche ) , cut into small pieces , and incubated for 12h in 1% formaldehyde , PBS and proteinase inhibitors ( Complete Mini , Roche ) with rotation at 4°C . Formaldehyde was removed by intensive washing in ice-cold PBS and centrifugation . Samples were lysed in RIPA ( with proteinase inhibitors , Complete Mini , Roche ) , sonicated on ice ( Sonopuls HD 2070 , 4 times 10s , 100% ) , and centrifuged . Samples from each group were pooled and protein content of clear phase lysates was measured using a Bradford assay ( Amersham ) . For each immunoprecipitation ( IP ) 1 . 5 mg of protein was taken . The volume of the samples was kept constant by using dilution buffer ( prepared according to Upstate protocol ) . For preclearance 90 µl of Protein A Sepharose slurry ( Amersham ) was added , and the samples were rotated for 1h in 4°C . After centrifugation beads were discarded , and 1% of supernatant volume per aliquot was used as an input control . The residual volume was incubated with 6 µg of appropriate antibodies ( anti-Pol II ( C-18 , Santa Cruz ) , anti-Flag ( Sigma ) , anti-SRC1 ( M-20 , Santa Cruz ) , anti-TIF2 ( C-20 , Santa Cruz ) ) . The antibody-bound proteins were then precipitated using 300 µl Protein A Sepharose slurry ( Amersham ) , washed and further processed according to the Upstate protocol . Triglyceride-content in skeletal muscle and liver was measured as described previously [47] . Briefly , tissues were homogenized in liquid nitrogen and treated with ice-cold chloroform/methanol/water mixture ( 2:1:0 . 8 ) for 2 min . After centrifugation the aqueous layer was removed and the chloroform layer was decanted . The mixture was incubated at 70°C for chloroform clearance , and the residues were dissolved in isopropanol , and assessed for the triglyceride content using an enzymatic-calorimetric test ( Cypress diagnostics ) according to the manufacturer's instructions . Results from real-time PCR of cell lines , transfections , and animal experiments were analyzed by ANOVA followed by multiple comparison testing or with paired/unpaired t tests , as appropriate . Data are expressed as mean±SEM or as indicated . Results were considered to be statistically significant at p<0 . 05 . | In the present study , we demonstrate for the first time a pro-diabetogenic function of the ERβ . Our experiments indicate that ERβ impairs insulin sensitivity and glucose tolerance in mice challenged with a high fat diet ( HFD ) . Loss of ERβ , studied in ERβ -/- mice ( βERKO mice ) , results in increased body weight gain and fat deposition under HFD-treatment . Conversely , absence of ERβ averted accumulation of triglycerides and preserved regular insulin signaling in liver and skeletal muscle . This observation was associated with improved whole-body insulin sensitivity and glucose tolerance . Increased adipose tissue mass in the presence of improved insulin sensitivity and glucose tolerance is usually observed under chronic stimulation of the nuclear hormone receptor PPARγ . In consonance , we show that activation of PPARγ was markedly induced in gonadal fat from βERKO mice and blockade of adipose PPARγ signaling by antisense oligonucleotide injection reversed the metabolic phenotype . Moreover , our cell culture experiments indicate that ERβ is a negative regulator of ligand-induced PPARγ activity in vitro . Finally , we identify SRC1 and TIF2 as key players in the ERβ-PPARγ interaction . In summary , the present study demonstrates that ERβ impairs insulin and glucose metabolism , which may , at least in part , result from a negative cross-talk with adipose PPARγ . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"cell",
"biology/cell",
"signaling",
"physiology",
"diabetes",
"and",
"endocrinology/obesity",
"physiology/endocrinology",
"pharmacology",
"diabetes",
"and",
"endocrinology",
"biochemistry/transcription",
"and",
"translation",
"molecular",
"biology",
"cell",
"biology/gene",
"expression",
"diabetes",
"and",
"endocrinology/type",
"2",
"diabetes"
]
| 2008 | Metabolic Actions of Estrogen Receptor Beta (ERβ) are Mediated by a Negative Cross-Talk with PPARγ |
Signaling pathways mediate the effect of external stimuli on gene expression in cells . The signaling proteins in these pathways interact with each other and their phosphorylation levels often serve as indicators for the activity of signaling pathways . Several signaling pathways have been identified in mammalian cells but the crosstalk between them is not well understood . Alliance for Cellular Signaling ( AfCS ) has measured time-course data in RAW 264 . 7 macrophage cells on important phosphoproteins , such as the mitogen-activated protein kinases ( MAPKs ) and signal transducer and activator of transcription ( STATs ) , in single- and double-ligand stimulation experiments for 22 ligands . In the present work , we have used a data-driven approach to analyze the AfCS data to decipher the interactions and crosstalk between signaling pathways in stimulated macrophage cells . We have used dynamic mapping to develop a predictive model using a partial least squares approach . Significant interactions were selected through statistical hypothesis testing and were used to reconstruct the phosphoprotein signaling network . The proposed data-driven approach is able to identify most of the known signaling interactions such as protein kinase B ( Akt ) → glycogen synthase kinase 3α/β ( GSKα/β ) etc . , and predicts potential novel interactions such as P38 → RSK and GSK → ezrin/radixin/moesin . We have also shown that the model has good predictive power for extrapolation . Our novel approach captures the temporal causality and directionality in intracellular signaling pathways . Further , case specific analysis of the phosphoproteins in the network has led us to propose hypothesis about inhibition ( phosphorylation ) of GSKα/β via P38 .
Cells regulate their function through a complex circuitry that involves myriad interacting networks from intracellular signaling and metabolic pathways to genetic regulatory pathways . Intracellular signaling is the first step in translating the environmental cue in regulating various processes e . g . cell growth , differentiation and apoptosis . Activation of proteins through phosphorylation is an important event in intracellular signaling and serves as a metric for the flux in the signaling pathway . Understanding of the regulation of protein phosphorylation is the key to identifying cellular mechanisms which interpret the environmental cues . The knowledge of the regulation and interaction of various phosphoproteins is sparse . The goal of this paper is to develop an approach for data-driven reconstruction of phosphoprotein signaling networks and to test them using the large-scale phosphoprotein data available from the Alliance for Cellular Signaling ( AfCS ) . Among many types of posttranslational modifications of proteins , protein phosphorylation is the most studied and has substantial impact on biological function [1] , [2] . Phosphorylation occurs on serine , threonine or tyrosine residues . Phosphoproteins ( PPs ) are considered as markers of signaling pathways because the levels of phosphorylation generally indicate the level of signaling activity in the pathway . The gamut of cellular processes affected by phosphoproteins varies from signal transduction , gene-expression , post-translational modifications of other proteins , cell differentiation , and development to cell cycle control . For example , the extracellular signal-regulated kinase ( ERK ) and P38 pathways are involved in cellular differentiation/proliferation [3] . c-Jun N-terminal kinase pathway is involved in cytokine induced apoptosis ( tumor necrosis factor ( TNF ) signaling ) . In a similar manner , the other signaling pathways mediate important processes such as inflammation , the most prominent one is the P50–P65 nuclear-factor kappa beta ( NF-KB ) pathway [4] , [5] Some molecules in these pathways , such as P50–P65 NF-KB and P38 translocate to the nucleus and act as transcription factors or regulate gene-expression through other mechanisms . In short , the ability to measure phosphoproteins has equipped the biologists to study the role of cytosolic intracellular signaling in virtually all aspects of biological processes . Many of the signaling pathways function as a cascade . In a cascade , subsequent proteins are activated via a previous ( phosphorylated/activated ) protein . For example , in the P38 MAPK pathway , the cascade is composed of MAP or ERK kinase kinase 4 ( MEKK4 ) and TGF-beta activated kinase 1 ( TAK1 ) ( a MAPK kinase kinase or MAPKKK ) , MKK3 and MKK6 ( a MAP kinase kinase or MAPKK ) , and p38 MAPK [3] , [6] . This pathway is activated by stress signals . Upon activation of MAPKKK , it phosphorylates and activates MAPKK which in turn phosphorylates and activates P38 MAPK . The information flows downstream with time , thus these interactions are causal in nature and can only be captured by dynamic mapping . Some pathways are activated in a non-cascade manner . One example is the cAMP signaling pathway in which upon production of cAMP within the cell through activation of adenylate cyclase via G-protein Gsα , cAMP activates protein kinase A ( PKA ) which in turn stimulates degradation of cAMP itself through phosphorylation of a phosphodiesterase ( an example of a negative feedback ) [7] . Even in such pathways , information flow is generally downstream with time , e . g . Gsα → adenylate cyclase → cAMP → PKA . The availability of the temporally resolved measurements of many phosphoproteins allows us to study signaling pathways and cross-talk between them . The development of high-throughput technologies have made these studies of pathways possible by allowing distinct types of simultaneous quantitative measurements of the cellular components such as mRNA levels , protein phosphorylations and metabolites . While lack of large datasets is one limiting factor in detailed and quantitative studies of regulation of signaling and metabolic pathways , such studies have also been impeded by the unavailability of suitable mathematical approaches to integrate diverse types of data and knowledge . Further , the complexity of intracellular signaling arising from feedback and feed-forward loops and cross-talk between different signaling pathways has exacerbated the problems associated with developing reliable mathematical approaches [8] , [9] . This complexity is manifested by the presence of multiple time-scales ranging from few seconds to several hours across various biochemical processes . Data measured at accordingly appropriate time intervals are required to reconstruct causal networks for such processes . The differences in the time-scales and lack of knowledge about time-lags in various processes make it difficult to decipher their interactions . However , systems biology approaches open avenues to decipher the interactions between the components and aid partial reconstruction of the underlying cellular network . Computational systems biology has seen tremendous advances during this decade . In the past few years , research in computational systems biology has moved beyond simple clustering and correlation based interaction networks . Major efforts on data-driven network reconstruction and model development have been centered on input/output-based and probabilistic graphical models . Input/output-based approaches are less tedious compared to probabilistic graphical models . Some of the contributors to input/output-based modeling include Bonneau et al . [10] , Janes et al . [11] and Pradervand et al . [12] . Among the probabilistic graphical models , Bayesian networks are most popular [13] , [14] . Contributions in Bayesian network-based modeling include the work of Sachs et al . [15] , Hartemink et al . [16] and Yu et al . [17] . As discussed in a recent review by Camacho et al . [18] , many other approaches such as partial correlation analysis and other statistical and systems engineering methods have been developed [19]–[25] . These recent efforts emphasis the importance of applying systems approaches to decipher and reconstruct cellular networks using high-throughput data . The time lag between the interactions because of localization , membrane barrier and transportation has motivated researchers to employ dynamic modeling techniques . Dynamic models and temporally causal networks are derived by mapping the input data at a previous ( current ) time-point to the output data at the current ( future ) time-point to capture the temporal causal effects explicitly [26]–[28] . Recently , many approaches have been developed for reconstructing networks using dynamic ( time series ) data . These include ( 1 ) state-space representation based techniques [10] , [29] , [30] and ( 2 ) dynamic Bayesian networks [31] , [32] . Such networks have been more efficient in predicting significant connections reported in the available literature . Most approaches intended for utilizing dynamic data for network identification can also handle steady-state data by setting the time-rate of change of the output or the state nodes to zero [10] . In the present work , we have applied the linear regression approach ( input/output modeling ) and statistical hypothesis testing to infer the important connections amongst signaling pathways using phosphoprotein time-series data . Partial least squares ( PLS ) method is used for input/output mapping ( linear regression ) . The advantage of using PLS is that it calculates principal components ( PCs ) in the direction of output and captures sufficient variation in the output data with relatively lesser number of PCs or latent variables [33]–[35] . F-test and t-test are used for statistical hypothesis testing . This manuscript is organized as follows . In the next section , we present the results and validation of the model followed by discussion . The last section briefly discusses the experimental data preprocessing and the methodology used for the network reconstruction .
The potential relationships were inferred using the correlation between the input ( predictor ) variables and the output ( response ) variables . The correlation matrix between the input and output data is visualized using heat-map in Figure 1 . The rows and the columns represent the inputs ( PPs at tk−1 ) and outputs ( PPs at tk ) , respectively . Please see Table 1 for the names of the PPs . High correlation was observed along the diagonal in Figure 1 with the indication that most of the phosphoproteins were highly self-regulated . The high self-correlation of PPs can be explained from the fact that most of the chosen PPs are from independent signaling pathways in this study and majorly activated by their upstream signaling molecule rather than by interaction/cross-talk between pathways . Thus in our modeling approach , we have also allowed this possibility via self activation of its phosphorylation . PPs in the same pathways showed high correlation . For example , ERK1/2 and RSK are the part of classical map kinase pathway and showed high correlation with each other . Variants of the same PP ( i . e . GSKα/β and ST1A/B ) and the member of the same family ( EZR and MOE: part of ERM family ) also show high correlation . The PPs belonging to independent pathway ( e . g . P40 and ST1A/B ) showed no correlation with most of other PPs . In this dynamic correlation matrix , high correlation was also observed from P38 to ERK1/2 and its downstream target RSK . We did not observe good correlation from ERK1/2 to P38 , which suggested that there is a directed edge from P38 to ERK1/2 but not vice versa . The isoforms of protein kinase C ( PKCD and PKCM ) also did not show good correlation with each other indicating that they are regulated differently . The PLS models were developed using Eq . 2 . Quality control of the model was performed at two steps: ( 1 ) during the PLS model development ( at least 50% variance in output data captured ) and ( 2 ) during the K-fold cross-validation . Overall , only 14 phosphoprotein models were selected out of 17 phosphoproteins ( S6 , P40 and P65 were rejected ) . Table S1 in Supporting Information lists the number of PCs used and the percentage variance captured in each PLS model . Significant interactions were deciphered using the statistical method discussed in the Materials and Methods section . The method of PLS is preferred for input/output modeling because of its ability to handle noise and dependency in the data and to reduce the dimension of the linear equations . Figure 2 shows the plot of the ratio ( r ) of the coefficient-values in actual model to the standard deviation of the corresponding coefficient-values in random models for each input for selected outputs . We have omitted those output variables whose model was not valid under the selection criteria described in the model development and cross-validation . The significant interactions are identified by applying the threshold on r . Only a few of interactions are discovered as being significant per model and capture the significant variation in the output variable . Names of PPs with significant coefficients and possible significant coefficients ( 90% of the threshold ) are shown in Figure 2 . The minimal model was constructed using only the significant connections identified from Figure 2 . For comparison with the PLS models , the percentage variance captured by the minimal models is listed in Table S1 in Supporting Information . To check the validity of the minimal model , we also developed the full model containing all variables as inputs . The root mean square error was calculated for the minimal model ( σr ) and the full model ( σf ) . Figure 3 shows a scatter-plot between experimental measurements versus the predicted outputs for the minimal models at a future time-point . Solid lines ( y = x ) , the dashed line ( y = x±σf ) and dotted line ( y = x±2σf ) were drawn to show the fit to the experimental data . Most of the data points lie within the ±2σf band and hence , the reduced model is a good predictive model . F-test was conducted to verify the similarity in the prediction-error between the reduced model ( σr ) and the full model ( σf ) . It was successful with 95% confidence level for all valid models . Figure 4A represents the combined network obtained from the models . The thickness of the interaction lines is proportional to ( the square root of the ) confidence obtained from statistical analysis . Strong similarity was observed between correlation coefficients and the significant PLS regression coefficients for most of the phosphoproteins . Examples include AKT → GSKα and β , ERK1 and 2 ↔ RSK and PKCD ↔ SMD2 . Variants of the same PP ( i . e . ERK1 and 2 , GSKα and β , and ST1A and B ) and the member of the same family ( EZR and MOE: part of ERM family ) showed strong significant interaction with each other and similar interaction with other PPs . Thus for the clarity of presentation , we have combined these nodes into one node , e . g . ERK1 and 2 were combined into ERK1/2 , GSKα and β in GSKα/β , ST1A and B in ST1A/B and EZR and MOE in ERM ( Figure 4B ) . It is important to note that since dynamic mapping is used to reconstruct the signaling network , the resulting network inherently captures the temporally causal connections . The reconstructed network was characterized with respect to commonly used graph-theoretic metrics defined in the Materials and Methods section . For all pairs of distinct nodes in the network of Figure 4B , all the directed paths between them were identified using a depth-first search approach [36] ( for full list of all the paths , refer Table S2 in Supporting Information ) . As an example , Table 2 lists all the paths from P38 to GSKα/β . From this , the total number of paths , the shortest path length and the average path length for the pairs of nodes are computed ( Table 3 ) . Given that there are some false-positive connections ( correlation can result in false reversible connection ) in the network , these measures are slightly biased . However , they are still useful because they provide a lower-bound ( for the minimum path-length ) and upper-bound ( for the average path-length ) and usually exclude totally unrelated interactions . In the example shown in Table 2 , P38 affects GSKα/β directly and indirectly via AKT . It is difficult to predict the true paths but they do assist in formulating hypotheses for further investigation through specific knockout or inhibition experiments . P38 has the highest degree and acts as a hub . The implication is that any perturbation to P38 will affect many downstream targets/pathways .
The experimental data was generated by the Alliance for Cellular Signaling ( AfCS ) laboratories using the RAW 264 . 7 cells . Single- and double-ligand screening for 22 ligands ( Table 1 ) in 251 experiments were performed . In each experiment , time-dependent changes in the phosphorylation-level of 21 signaling proteins were measured at 1 , 3 , 10 , 30 minutes after stimulation using phosphoprotein-specific antibodies ( western blot analysis , AfCS protocols #PP00000177 and #PP00000181 [61] . Most experiments were done in duplicates and triplicates . For complete details of the data and the experimental protocols , please refer the AfCS website [62] . The PP data is available from the AfCS Data Center ( “RAW 264 . 7 ligand screen” and “RAW 264 . 7 two-ligand screen” ) . The 22 ligands used and the phosphoproteins measured in experiments are listed in Table 1 . Time scale for most of the signaling events lies in minutes . Response at early times , such as five minutes after the stimulation with the ligand ( s ) , shows the primary effect of ligands stimulation and signaling from upstream ( phospho ) protein to the expected pathways . In most cases , the strength of an early response is much higher than the degradation rate of the signaling molecule . Contrary to this , during the later phase , the desensitization/internalization of the cell surface receptors decreases/stops the effect of ligand stimulation . The changes in the signaling molecular concentrations observed at those times are primarily due to its self degradation . Here , in the case of phosphoprotein data , degradation mostly represents dephosphorylation of the phosphoprotein . Thus , to avoid such false positive results in linear modeling approach , we have used the early time point data at only 1 and 3 minutes and ignored the later time point data at 10 and 30 minutes . However , the data at 10 min has been used for accessing the predictive power of the model . The data at t = 0 minute was not included as it was always equal to 1 ( available as fold-change from the AfCS data center ) . For the purposes of mathematical treatment , replicates are considered as separate conditions . Explicit use of replicates has two advantages: ( 1 ) different experiments were repeated different number of times , so separate treatment of replicates resulted in more weighting to the experiments with more replicates and ( 2 ) the effect of experimental variation was automatically accounted in the statistical analysis during model development and cross validation . Data was log2 transformed prior to the modeling to give equal weighting to up- and down-regulation . Both input and output data were normalized with respect to the mean and standard deviation of the respective variables . | Cellular systems are dynamic in nature , perform various biological functions and can adapt to their environment through compositional and structural remodeling . This remodeling is initiated by the binding of external ligands to receptor proteins on the cell surface or other stimuli resulting in the activation of various signaling pathways . The activation of signaling pathways in turn results in altered gene-expression which leads to changes in the molecular composition both inside and outside the cell . Thus , a thorough study of the complex interconnectivity of signaling pathways using dynamic data is necessary to understand the biological function and environmental adaptation . Protein phosphorylation is a key event in the activation of signaling pathways . We have developed an advanced computational approach to decipher the connections between different signaling pathways using time-resolved phosphoprotein data provided by the Alliance for Cellular Signaling . Important signaling events such as activation of glycogen synthase kinase 3 by protein kinase B ( Akt ) are captured by our reconstructed network . Novel links as well as testable hypotheses are also generated by our analysis approach . | [
"Abstract",
"Introduction",
"Results",
"Materials",
"and",
"Methods"
]
| [
"molecular",
"biology/bioinformatics",
"mathematics/statistics",
"computational",
"biology/systems",
"biology",
"computational",
"biology/signaling",
"networks"
]
| 2010 | Identification of Crosstalk between Phosphoprotein Signaling Pathways in RAW 264.7 Macrophage Cells |
Single nucleotide polymorphisms ( SNPs ) on chromosome 9p21 are associated with coronary artery disease , diabetes , and multiple cancers . Risk SNPs are mainly non-coding , suggesting that they influence expression and may act in cis . We examined the association between 56 SNPs in this region and peripheral blood expression of the three nearest genes CDKN2A , CDKN2B , and ANRIL using total and allelic expression in two populations of healthy volunteers: 177 British Caucasians and 310 mixed-ancestry South Africans . Total expression of the three genes was correlated ( P<0 . 05 ) , suggesting that they are co-regulated . SNP associations mapped by allelic and total expression were similar ( r = 0 . 97 , P = 4 . 8×10−99 ) , but the power to detect effects was greater for allelic expression . The proportion of expression variance attributable to cis-acting effects was 8% for CDKN2A , 5% for CDKN2B , and 20% for ANRIL . SNP associations were similar in the two populations ( r = 0 . 94 , P = 10−72 ) . Multiple SNPs were independently associated with expression of each gene ( P<0 . 05 after correction for multiple testing ) , suggesting that several sites may modulate disease susceptibility . Individual SNPs correlated with changes in expression up to 1 . 4-fold for CDKN2A , 1 . 3-fold for CDKN2B , and 2-fold for ANRIL . Risk SNPs for coronary disease , stroke , diabetes , melanoma , and glioma were all associated with allelic expression of ANRIL ( all P<0 . 05 after correction for multiple testing ) , while association with the other two genes was only detectable for some risk SNPs . SNPs had an inverse effect on ANRIL and CDKN2B expression , supporting a role of antisense transcription in CDKN2B regulation . Our study suggests that modulation of ANRIL expression mediates susceptibility to several important human diseases .
The chromosome 9p21 . 3 region adjacent to the loci encoding the cyclin-dependent kinase inhibitors CDKN2A ( ENSG00000147889 ) and CDKN2B ( ENSG00000147883 ) is an important susceptibility locus for several diseases with a complex genetic background . Recent genome-wide association ( GWA ) studies have shown that single nucleotide polymorphisms ( SNPs ) in this region are associated with coronary artery disease ( CAD ) [1]–[4] , ischaemic stroke [5] , [6] , aortic aneurysm [7] , type II diabetes [8] , [9] , glioma [10] , [11] , and malignant melanoma [12] . Candidate gene approaches have also reported SNPs in this region to be associated with breast [13] , [14] , ovarian [15] , and pancreatic carcinoma [16] , melanoma [17] , and acute lymphoblastic leukaemia [18] , as well as with poor physical function in the elderly [19] . Variants associated with these diseases are represented in Figure 1 . Most of the risk variants in the chromosome 9p21 region identified by GWA studies are in non-coding regions , suggesting that their effects are likely to be mediated by influences on gene expression . Sequence variation can influence expression by cis or trans mechanisms . Trans-acting elements influence transcript levels of both alleles via diffusible factors and are usually located distant to the genes they regulate , whereas cis-acting elements act on genes on the same chromosome in an allele-specific manner and are usually located close to the genes they regulate . Since most reported risk variants in the 9p21 region do not appear in mature transcripts , and there are no known or predicted microRNAs mapping to this region [20]–[23] , these variants are unlikely to produce diffusible trans-acting factors and are therefore likely to influence expression of nearby genes in cis . Genes in the region include the cyclin-dependent kinase inhibitors CDKN2A ( p16INK4a ) including its alternative reading frame ( ARF ) transcript variant ( p19ARF ) , CDKN2B ( p15INK4b ) , and a recently-discovered non-coding RNA , designated ANRIL ( CDKN2BAS , ENSG00000240498 ) , that undergoes splicing and is transcribed from the opposite strand to CDKN2A/B . The ARF/CDKN2A/B proteins are established tumour suppressors deleted in a range of cancers including familial cutaneous malignant melanoma [24]; they block cell cycle progression and influence key physiological processes such as replicative senescence , apoptosis , and stem-cell self-renewal [25] . Cis-acting regulatory elements for these genes have been identified in vitro using reporter assays [26]–[30] , but expression levels are also influenced by factors such as age , chemotherapeutic agents , DNA damage by ultraviolet or ionizing radiation , and levels of transcriptional regulators [31] , all of which are likely to act in trans . The function of ANRIL is unknown , but other processed non-coding RNAs are involved in the regulation of gene expression through transcriptional and translational control mechanisms [32] . Genetic effects on expression can be assessed by comparing total expression levels in individuals with different genotypes at a putative regulatory locus . This is termed expression quantitative trait locus ( eQTL ) mapping [33] . This approach utilises information from all members of the population , but reflects the net effect of both cis and trans-acting influences; the sensitivity to detect cis-acting effects is therefore reduced in the presence of significant variation in trans-acting influences such as the environmental factors outlined above . An alternative approach that is specific for mapping cis-acting influences is to measure allelic expression ( aeQTL mapping ) . An unequal amount of transcript arising from each allele in an individual heterozygous for a transcribed polymorphism indicates the presence of cis-acting effects on expression . While traditional eQTL analysis assesses the influence of polymorphisms by comparing expression between samples , allelic expression analysis compares the expression levels of alleles within individual samples , making it much more robust to trans-acting influences that affect both alleles such as age , gender , population stratification , or experimental variability . This maximises the sensitivity for detecting cis-acting effects [34] . Variants associated with CAD span a region greater than 100kb , but the association is accounted for by SNPs within a 53kb interval that define a core risk haplotype [35] . Lead SNPs for CAD and diabetes are in separate LD blocks in Caucasians and are independently associated with the two separate diseases [35] . To date , CAD risk SNPs have shown inconsistent association with CDKN2A , CDKN2B and ANRIL by eQTL mapping . One CAD risk SNP was associated with altered ANRIL expression in blood , but not with CDKN2A or CDKN2B expression [36] , whilst a different CAD risk SNP has been associated with reduced expression of all three genes in peripheral blood T-cells [37] . However , the latter study found no association with expression for other CAD risk SNPs [37] , and another report also found no association of a lead CAD risk SNP with these genes or with global gene expression in primary vascular tissue and lymphoblastoid cells [38] . Based on evolutionary conservation and effects on expression , individual SNPs ( rs10757278 and rs1333045 ) have been highlighted as potential causal variants for the association with CAD [36] , [37] . However , if multiple cis-acting effects are present at a locus , resolving a disease association by fine-mapping may not be possible . Examining gene expression rather than disease phenotype increases the power to map cis-acting effects , and we used this approach to determine whether multiple sites independently influence expression . Caucasian populations have strong linkage disequilibrium ( LD ) in the chromosome 9p21 region which limits the ability to separate the effects of individual SNPs on expression [35] . Populations of African ancestry have less LD [39] , [40] , which can be exploited to improve the fine-mapping of functional polymorphisms associated with quantitative traits [41] , [42] . We therefore used eQTL and aeQTL mapping to perform detailed fine-mapping of the association of SNPs at the 9p21 . 3 locus with expression of CDKN2A , CDKN2B and ANRIL using a mixed-ancestry South African ( SA ) population , as well as a British Caucasian cohort . We identified multiple SNPs independently associated with expression of each gene , suggesting that several sites may modulate disease susceptibility . The markers identified in GWA studies were all associated with allelic expression of ANRIL , but association with the other two genes was only detectable for some of them . Our study suggests that modulation of ANRIL expression mediates susceptibility to a range of important human diseases .
Total expression levels showed substantial inter-individual variation for each of the three genes , up to 13 . 9-fold for CDKN2A , 36 . 1-fold for CDKN2B , and 25 . 5-fold for ANRIL . Allelic expression ratios at individual transcribed markers also showed considerable inter-individual variation , up to 5 . 6-fold for CDKN2A , 2 . 4-fold for CDKN2B , and 6 . 8-fold for ANRIL . Plots of the allelic expression ratios at each transcribed SNP in the SA and Caucasian cohorts are shown in Figure S1 and Figure S2 and plots of the normalised total expression Ct values are shown in Figure S3 . Standard errors for ANRIL were higher than for the other two genes in both the allelic and total expression assays , which is likely to be due to the fact that peripheral blood expression of ANRIL was lower than for CDKN2A and CDKN2B . We estimated the proportion of the variance in total expression that can be attributed to cis-acting effects for each transcribed SNP in the three genes , as described in the Methods section . For CDKNA this proportion was 8% when rs3088440 was used to estimate the variance in cis acting effects , and 4% when rs11515 was used . For CDKN2B the corresponding values were 5% ( using rs3217992 ) , 5% ( using rs1063192 ) and for ANRIL 20% ( using rs10965215 ) , and 19% ( using rs564398 ) . Total expression levels of CDKN2A , CDKN2B and ANRIL were positively correlated ( r = 0 . 24 to 0 . 30 , all P<4×10−5 ) as shown in Figure S4 , suggesting that expression of these genes is co-regulated . Allelic expression ratios ( AER ) measured at the two transcribed SNPs in each gene were highly correlated ( CDKN2A r = 0 . 68 P = 1 . 7×10−3; CDKN2B r = 0 . 80 P = 1 . 7×10−12; ANRIL r = 0 . 90 P = 1 . 0×10−26; all genes combined r = 0 . 96 P = 3×10−61 ) as shown in Figure S5 . This was expected since the two transcribed SNPs selected to assess AER in each gene are located in the same exon and the same transcripts . We therefore used the AERs from both transcribed markers in each gene ( as described in the Methods section ) for the aeQTL analysis . This increased the number of informative heterozygotes at which allelic expression could be assessed for each gene and increased the power to detect significant effects , as shown in Table 1 . Unlike allelic expression ratios , total expression data may be influenced by covariates that influence expression in trans . We therefore corrected total expression values for covariates ( age , sex , and ethnicity ) and excluded outlying individuals as described in the Methods section . These corrections did not significantly alter the results of the eQTL analysis , as shown in Figure S6 . All subsequent analyses are presented using the covariate-corrected eQTL data . We compared cis-acting effects assessed by eQTL and aeQTL mapping , as shown in Figure 2 . There was a strong correlation both for the effect size ( r = 0 . 87 , P = 4 . 7×10−51 ) and significance of association ( r = 0 . 97 , P = 4 . 8×10−99 ) at each mapping SNP between the two techniques . However , the associations were more significant for allelic expression than for total expression analysis , indicating that allelic expression had greater power for detecting cis-acting effects . This suggests that trans-acting effects make a substantial contribution to the overall variance of expression in these genes , which is consistent with our estimates that cis-acting effects account for only between 4 and 20% of the overall variance in expression of these genes . We compared aeQTL analysis between the SA and British Caucasian samples . Results of aeQTL mapping were highly correlated between the two populations , both for the significance of the detected association ( r = 0 . 94 , P = 10−72 ) and the estimated magnitude of the effect on expression for each SNP ( r = 0 . 82 , P = 2×10−38 ) , as shown in Figure 3 . Patterns of LD in the two populations are shown in Figure S7 . Minor allele frequency in the SA population was higher ( which increases the proportion of informative heterozygotes for allelic expression analysis ) for 33 of the 53 SNPs typed in both populations . In view of the similarity of the effects in the two cohorts , we combined the data in subsequent analyses , increasing the power to detect cis-acting effects of smaller magnitude and enabling us to adjust for the effects of individual SNPs . The significance of associations for individual SNPs in the combined cohort is shown in Figure 4 . Subsequent results refer to the combined dataset , with specific discussion of differences between the populations where relevant . As described in the Methods section , we defined significance thresholds for all SNP associations using the family wise error rate ( FWER ) where multiple testing was taken into account by using a Bonferroni correction for the 56 SNPs tested . Associations with a FWER threshold of 0 . 05 ( corresponding to a nominal P-value of 8 . 9×10−4 , −log10P of 3 . 05 , and −log10 FWER of 1 . 3 ) were regarded as significant . Table S2 shows the −log10 of the nominal P-values and FWER for all SNP associations , and nominal P-values are reported in the text . The effect of each SNP on AER is also shown in Table S2 . The maximum change in allelic expression associated with any SNP was 1 . 4-fold for CDKN2A , 1 . 33-fold for CDKN2B , and 1 . 97-fold for ANRIL . Due to the power of our combined dataset we were able to detect SNP effects on allelic expression as small as 1 . 05-fold that were significant . As shown in Figure 4 , multiple SNPs were associated with cis-acting influences on expression of CDKN2A , CDKN2B and ANRIL . This could be the result of multiple independent loci influencing expression of each gene , but could also be a reflection of strong LD in the region since associations might be observed for ‘non-functional’ SNPs ( that do not directly influence expression ) which are in LD with other ‘functional’ polymorphisms . Adjusting for the effect of individual SNPs was used to assess whether multiple SNPs were independently correlated with expression of the three genes , as shown in Figure 5 . For each gene stepwise adjustments were made for the effect of the SNP which showed the most significant association with expression , until independent effects could no longer be detected . Associations remained significant after adjusting for the top SNP for CDKN2A and CDKN2B , and the top two SNPs for ANRIL . Our results indicate that even after adjusting for the effects of the most significant marker , some of the remaining SNPs still showed significant association with ANRIL expression . This could be explained by the presence of more than one functional polymorphism affecting expression , but could also reflect the presence of a functional polymorphism that is in disequilibrium with both markers . However , examination of the allelic expression patterns provides additional support for the presence of multiple sites affecting expression . For example , Figure 6 shows the allelic expression ratios observed at the transcribed SNP rs564398 in ANRIL , grouped according to the genotype at rs10965215 . These two SNPs are in strong LD ( D′ = 0 . 98 ) , hence the absence of individuals homozygous for the A allele at rs10965215 that are heterozygous at rs564398 . We observe that the G allele of the transcribed SNP ( rs564398 ) is overexpressed ( G/A AER values greater than 1 ) , however overexpression is stronger ( P = 10−15 using the Mann-Whitney test ) for individuals that are also heterozygous at the second polymorphism ( rs10965215 ) . This pattern is not consistent with allelic expression being determined by a single biallelic polymorphism acting in cis and suggests that there is more than one functional polymorphism or that this polymorphism is multiallelic . Such patterns were common in our data . The direction of cis-acting effects on expression was compared between genes for SNPs showing significant associations with expression of each gene , as shown in Table 2 . SNP effects for CDKN2A and ANRIL were in the same direction for all 10 SNPs , meaning that alleles associated with overexpression of CDKN2A were also associated with overexpression of ANRIL . By contrast , for all 8 SNPs that were significantly associated with allelic expression of both CDKN2A and CDKN2B , the alleles associated with CDKN2A overexpression were associated with CDKN2B underexpression . Similarly for all 3 SNPs significantly associated with allelic expression of both CDKN2B and ANRIL , alleles associated with overexpression of CDKN2B were associated with ANRIL underexpression . The total expression analysis had insufficient power for similar analyses to be performed . We investigated whether SNPs within regulatory regions previously identified by in vitro reporter assays were associated with cis-acting effects on expression in vivo . The effect on gene expression and significance of the association for each SNP is summarised in Table S2 . CDKN2A expression was significantly correlated with SNPs in its promoter and the ARF transcript promoter [26]–[29] , and with SNPs close to the RDINK4/ARF domain that has been shown to regulate expression of CDKN2A , ARF and CDKN2B in vitro [30] . CDKN2B expression was also significantly correlated with SNPs in the CDKN2A and ARF promoter regions , suggesting that these elements influence expression of both genes . CDKN2B expression was not significantly correlated with the single SNP typed in its promoter ( rs2069418 ) prior to adjustment , but this became significant after adjustment for the most significant SNP in the ARF promoter ( rs3218018 ) . ANRIL expression was strongly associated with SNPs in the CDKN2B promoter ( P = 10−72 ) , ARF promoter ( P up to10−53 ) and RDINK4/ARF domain ( P = 10−12 ) , as well as with SNPs adjacent to the CDKN2A promoter ( rs3731239 , P = 10−25 ) . These data validate in vivo the function of the regulatory elements identified by in vitro transfection studies , and confirm that shared cis-acting elements influence expression of CDKN2A , CDKN2B and ANRIL . We examined the correlation of allelic expression of CDKN2A , CDKN2B and ANRIL with SNPs reported to confer disease susceptibility . The effect on gene expression and significance of the association for each SNP is summarised in Table 3 .
This is the most detailed study to date of cis-acting influences on expression at the chromosome 9p21 locus . We have shown that multiple sites in the 9p21 region independently influence CDKN2A , CDKN2B and ANRIL expression , and demonstrated that SNPs associated with diseases including CAD , diabetes , and cancers are all highly associated with ANRIL expression , suggesting that modulation of ANRIL expression may mediate disease susceptibility . We also report novel methodology for allelic expression analysis that allowed us to combine data from multiple transcribed polymorphisms and to adjust for the effects of particular SNPs . We have demonstrated that this approach has greater power than total expression analysis for mapping cis-acting effects . Total expression levels of CDKN2A , CDKN2B and ANRIL , which reflect the combined influence of cis and trans-acting factors , were positively correlated . This corroborates other recent data [37] , and suggests that expression of these genes is co-regulated . We have shown that trans-acting influences account for the majority of the observed variance in expression of these genes ( 80–96% ) , and the correlation in total expression levels is likely to reflect co-regulation of the genes through trans-acting factors . In addition , our allelic expression analysis demonstrated that expression is also influenced by shared cis-acting elements in the region . Despite the positive correlation in total expression levels , cis-acting effects associated with individual SNP alleles may act in opposite directions; the effect of individual SNPs on CDKN2B expression were opposite to effects on CDKN2A and ANRIL expression ( which were concordant ) in our study . Because cis-acting effects represent only a small proportion of the overall variance in expression of these genes , the effects acting in trans are likely to account for the positive correlation seen in total expression , but this does not diminish the potential biological significance of the cis-acting effects . ANRIL overlaps and is transcribed in antisense with respect to CDKN2B [49] . It is modestly conserved across species [36] and its function is not known , but recent work has demonstrated that antisense transcription from CDKN2B downregulates CDKN2B expression in cis through heterochromatin formation [50] . This is consistent with our observation of an inverse effect of SNPs on ANRIL and CDKN2B expression . By contrast , CDKN2A and ANRIL showed positive correlations for both allelic and total expression in our study . CDKN2A and ANRIL do not overlap , but are transcribed divergently from transcription start sites separated by just 300 base pairs . Although the ANRIL promoter is currently not characterised , it may share promoter elements with CDKN2A and the resulting co-regulation could account for the positive correlation in expression we observed for these genes , similar to that described at other sites [51] . In this context , inhibition of CDKN2B expression by ANRIL would enable a level of crosstalk between CDKN2A and CDKN2B expression , which would be consistent with the inverse cis-acting effect of SNPs on CDKN2A and CDKN2B that we observed . The observation that cis-acting genetic effects played a greater role in expression of ANRIL compared to CDKN2A and CDKN2B ( 20% compared to less than 8% and 5% respectively ) makes it a good candidate for genetic causation mediated through influences on expression . We compared total expression and allelic expression for the investigation of cis-acting influences on expression . While traditional eQTL analysis assesses the influences of polymorphisms by comparing expression between samples , allelic expression analysis compares the expression levels of alleles within individual samples , making it more robust to influences that affect both alleles such as age , gender or population stratification . This offers an important advantage for dissecting such cis-acting influences on expression , which although of lesser magnitude than trans-acting influences , may be of biological importance and possibly account for the genetic susceptibility observed in recent GWA studies . For aeQTL mapping we used a novel adaptation of our previously reported methodology [52] to combine multiple transcribed SNPs per gene , which increased the number of informative individuals and the power for detecting cis-acting effects . We demonstrated this approach using two transcribed polymorphisms per gene , but our methodology offers the potential for the inclusion of multiple additional transcribed variants . The results obtained by eQTL and aeQTL mapping were similar , consistent with previous work suggesting that the two approaches identify the same cis-acting loci [42] . However , we demonstrated that aeQTL analysis had substantially greater power than the eQTL approach . Adjusting for trans-acting covariates including age , sex and ethnicity in our eQTL analysis did not substantially alter the results . An influence of age on CDNK2A has been reported [53] , but there was little variability in the age of our SA cohort ( 90% of whom were between the ages of 18 and 30 years ) . The fact that allelic expression is a more efficient way to identify cis-acting influences on expression has implications for future studies investigating the effects of SNPs on expression at other loci , for example for the hundreds of non-coding SNPs correlated with different diseases by recent GWA studies [54] . Allelic expression quantifies the relative contributions of each allele to the mRNA pool irrespective of the absolute mRNA levels , and therefore provides information about transcriptional effects and polymorphisms within the transcript influencing RNA degradation in cis . By contrast , total expression analyses that quantify absolute mRNA levels are also sensitive to post-transcriptional regulatory effects , such as mRNA degradation by microRNAs . In extreme cases tight post-transcriptional regulation could keep total mRNA levels constant irrespective of the contributions of each allele to the total mRNA pool . The fact that the results of eQTL and aeQTL mapping were so similar in our study suggests that the effect of regulation at the post-transcriptional level is limited , although regulation of CDKN2A expression by a microRNA has been described [55] . In general , although allelic expression is a robust method for mapping sites influencing expression in cis , investigation of total expression and other intermediate phenotypes such as protein levels or protein activity will provide complementary information that contributes to fully understanding the phenotypic effects of cis-acting polymorphisms . It would be desirable to determine whether the significant associations with mRNA expression observed for CDKN2A and CDKN2B are confirmed at the protein level . Although we had hoped to use trans-ethnic fine-mapping to refine the associations with expression , the results of aeQTL mapping were in fact very similar in the SA and Caucasian populations . This replication in a separate cohort strongly supports the validity of our findings and enabled us to perform a combined analysis of the two cohorts . This approach of pooling data from ethnically-divergent populations has been previously shown to increase the power to detect influences on expression that are shared across populations [42] , [56] . The principal difference we identified between the two populations was for the SNPs associated with type II diabetes . The lead diabetes SNP rs10811661 was correlated with ANRIL underexpression in the Caucasian cohort , but not in the SA population , despite greater power to detect effects in that cohort . This may reflect differences in LD between the populations , but suggests that rs10811661 may not itself be the causal variant influencing diabetes susceptibility through effects on ANRIL expression . Studies to determine whether this SNP is associated with diabetes in populations of African origin would be of interest . The power of our analyses to detect differences in expression enabled us to adjust for the effects of individual SNPs . Using this we were able to demonstrate that expression , and therefore probably disease predisposition , is independently influenced by multiple sites and that the observed effects cannot be explained by a single polymorphic site . From our analysis we cannot exclude the existence of rare variants with large effects , but previous resequencing studies in this region did not find rare variants associated with disease phenotypes [2] , [3] . We are unable to say whether the individual SNPs for which we found associations are the actual ‘causal’ variants responsible for the effects on expression , or if the association simply reflects linkage disequilibrium between these SNPs and the causative polymorphisms . Although fine mapping studies often purport to identify causal variants , in the context of complex diseases characterising the pathways involved in disease predisposition may be more important . This is of particular interest for these genes where variation in expression is mostly due to trans effects which may be substantially influenced by non-genetic factors , raising the prospect that it may be amenable to therapeutic modulation . The putative causal variants rs10757278 and rs1333045 previously associated with altered ANRIL expression [36] , [37] were significantly associated with reduced ANRIL expression in vivo in our analysis , but their effects were relatively modest compared to other SNPs in the region and adjustment for the effect of these SNPs accounted for only a small proportion of the effect observed at other SNPs . The maximum changes in expression associated with individual SNPs were substantial , up to 2-fold for ANRIL , but we were also able to detect effects of much smaller magnitude; the minimum significant effect was associated with just a 1 . 05-fold change in expression . Although the associations of SNPs with expression that we observed were statistically highly significant , we cannot say what impact such effects on expression have on disease risk . However , even small differences in gene expression due to genetic factors that are present throughout an individual's lifetime could contribute to differences in common late-onset phenotypes such as CAD and diabetes , and the effects may be even greater in tissues related to disease . We examined in vivo expression in primary cells rather than in transformed cell lines . Although cell lines have been extensively used to investigate cis-acting influences on expression [56] , [57] , patterns of expression may be altered in immortalised cells , particularly for genes such as these that are associated with senescence and cell-cycle regulation . Furthermore , widely used cell lines are pauciclonal or monoclonal [58] , [59] and since a significant proportion of human genes exhibit random patterns of monoallelic expression within single clones of cell lines [60] , cis-acting effects in these cells are unlikely to be representative of polyclonal cell populations in vivo . Previous studies have delineated the promoters and other elements regulating CDKN2A/ARF and CDKN2B expression using reporter assays [26]–[30] . Such studies are valuable to identify causative polymorphisms , but since they examine the effects on expression outside of the normal haplotype , chromatin and cellular context their findings require confirmation by in vivo studies [34] , [61] . Our analysis confirmed that polymorphisms in upstream regulatory elements identified by in vitro assays were significantly associated with cis-acting effects on expression in vivo , but we also demonstrated that other loci located up and downstream were associated with effects on expression of similar or even larger magnitude . These data highlight the complexity and multiplicity of sites influencing expression in the region . The assays we used to investigate CDKN2A expression also included the ARF transcript variant . This gave the possibility to detect sites influencing expression of both transcripts , and we were able to detect effects of SNPs in both the CDKN2A and ARF promoter regions , although differential effects of loci on individual transcripts cannot be distinguished using this approach . All of the SNPs in the region associated with disease in GWA studies were associated with influences on ANRIL expression , suggesting that modulation of ANRIL expression may mediate susceptibility to these phenotypes . SNPs in the CAD core risk haplotype region [35] that are most strongly associated with CAD in GWA studies were associated with reduced ANRIL expression , but other SNPs associated with CAD which lie outside of the core risk haplotype showed independent and stronger associations with ANRIL underexpression . This may reflect differences in the relative importance of particular sites in the tissues responsible for the association with CAD . Indeed , the patterns of association we have observed in peripheral blood in healthy individuals may differ from those in primary disease tissues . Similarly , differences in the relative contribution of each SNP to modulation of expression in the tissues crucial for the pathogenesis of the different conditions could explain why particular diseases are associated with different subsets of SNPs that influence ANRIL expression . Recent work also suggests that ANRIL has multiple transcripts , which may be differentially expressed between tissues [36] , [38] . Confirmation of our findings in tissues relevant to each disease and for different ANRIL transcripts would therefore be desirable , although for CAD and other complex diseases the cell populations responsible for mediating disease susceptibility are unknown and may be inaccessible . Although tissue specificity of cis-acting influences is well documented , variation in cis-acting effects is primarily explained by genetic variation , with allele-specific expression at most SNPs being the same between tissues in the same individual [62] . Analysis of expression in blood is therefore likely to give biologically relevant information despite the fact that this may not be the tissue in which influences on expression actually mediate disease susceptibility . Previous genomewide expression analyses using microarrays and immortalised cell lines did not identify association of CDKN2A and CDKN2B expression with markers in this region , although they did not examine ANRIL expression [56] , [57] . However , two recent studies specifically examining expression in the chromosome 9p21 region in primary cells reported associations between CAD risk SNPs and gene expression in blood [36] , [37] . Jarinova et al found significant association of CAD risk variant rs1333045-C with ANRIL expression , but not with CDKN2A or CDKN2B expression [36] . Liu et al reported that a different CAD risk allele rs10757278-G was associated with reduced expression levels of CDKN2A , CDKN2B , and ANRIL , but in the same study found no correlation for five other SNPs tested , including two additional SNPs associated with CAD ( rs518394 and rs564398 ) . They also found no association for two SNPs associated with diabetes ( rs10811661 and rs564398 ) , the frailty risk SNP rs2811712 , and a melanoma risk SNP ( rs11515 ) [37] . We demonstrated that CAD risk SNPs rs1333045-C and rs10757278-G both correlate with ANRIL underexpression , but found no correlation of these SNPs with CDKN2A or CDKN2B expression . However , we identified highly significant influences on expression associated with other SNPs for which Liu et al found no association ( rs10811661 with CDKN2A and ANRIL; rs564398 with ANRIL; rs2811712 with CDKN2B; rs11515 with CDKN2A and CDKN2B ) . These findings are likely to reflect the greater power of our analysis for detection of cis-acting effects due to the larger sample size and increased sensitivity of our aeQTL mapping approach . The finding that disease associated SNPs are all associated with ANRIL expression suggests that ANRIL plays a role in influencing disease susceptibility . Although little is known about the targets of ANRIL , its effects may be mediated through antisense transcription regulation of CDKN2B in the tissues critical for the pathogenesis of the different diseases . The observation that the effects of sequence variants acting in cis were stronger for ANRIL than for CDKN2B may reflect selection pressure against variants that have substantial direct effects on the expression of critical genes . CDKN2A , ARF and CDKN2B are cell cycle regulators and are plausible candidates for involvement in the pathogenesis of the diseases for which we found SNP associations with ANRIL . Mutations involving these genes are well documented in glioma [63] , [64] and melanoma [49] , [65] , [66] . Overexpression of CDKN2A and CDKN2B in murine models is associated with pancreatic islet hypoplasia and diabetes [67] , [68] , and there is also emerging evidence that vascular cell senescence involving these pathways is involved in the pathogenesis of atherosclerosis [69] , [70] . Our data show that multiple independent sites in the chromosome 9p21 region influence CDKN2A , CDKN2B and ANRIL expression . SNPs associated with disease in GWA studies are all associated with ANRIL expression , indicating that modulation of ANRIL expression mediates susceptibility to a variety of conditions .
Peripheral blood for DNA and RNA analysis was collected from anonymous adult volunteers in two cohorts: 310 SA blood donors and 177 British Caucasians from north-east England . The self-reported ethnicity of the SA cohort was: 200 Cape mixed-ancestry; 67 African black; 19 Indian; 10 white; 4 other/unknown . 42% were male , with median age 20 years ( range 17–60 , lower quartile 19 , upper quartile 23 ) . In the Caucasian cohort , 50% were male , with median age 63 years ( range 25–101 , lower quartile 51 , upper quartile 69 ) . The study complies with the principles of the Declaration of Helsinki . Informed consent was obtained from all participants and the study was approved by the Newcastle and North Tyneside Local Research Ethics Committee and the University of Cape Town Faculty of Health Sciences Research Ethics Committee . For the South African samples DNA was extracted using a phenol/chloroform method from 4ml of peripheral blood in EDTA collected at the time of the RNA sample . For the British samples , DNA was obtained from the RNA solution prior to DNase treatment . RNA was extracted from 2 . 5ml of peripheral blood collected using the PAXgene system ( Qiagen ) following the manufacturer's protocol and was DNase treated using RQ1 RNase-Free DNase ( Promega ) . For AEI measurements , approximately 2µg of total RNA was reverse transcribed and eluted in 20µl , using SuperScript VILO cDNA Synthesis Kit ( Invitrogen ) for the SA samples and SuperScript III First-Strand Synthesis System for RT-PCR ( Invitrogen ) for the British samples . For real-time PCR measurements , 500ng of total RNA was reverse transcribed using High Capacity RNA-to-cDNA Master Mix ( Applied Biosystems ) and eluted in 20µl . Using the NCBI Entrez Gene database ( http://www . ncbi . nlm . nih . gov/ , 28/01/08 ) , transcribed SNPs with expected heterozygosity >0 . 2 in the HapMap CEU population were selected as suitable candidates for assessment of allelic expression . Transcribed polymorphisms in ANRIL , which was not annotated in the databases at the time of the design , were identified by comparing the reported mRNA sequence [49] with NCBI dbSNP . Transcribed SNPs selected using these criteria were: rs3088440 and rs11515 in exon 3 of CDKN2A; rs3217992 and rs1063192 in exon 2 of CDKN2B; rs10965215 and rs564398 in exon 2 of ANRIL . The two CDKN2A SNPs are also present in ARF , allowing the assessment of cis-acting influences on both of these transcripts . Another SNP rs10738605 in exon 3 of ANRIL also satisfied these criteria but was subsequently excluded due to poor performance of the assay . SNPs previously reported to be associated with disease phenotypes were selected for mapping effects on expression [6] , [8] , [9] , [13]–[19] , [43] , [45]–[48] , [71]–[74] . Additional tag SNPs required to capture common variation in a core region of interest ( Chr9:21958155–22115505 ) based on HapMap CEU data were also selected using HaploView 4 . 0 Tagger software using the following parameters: minimum minor allele frequency 0 . 01 , pairwise tagging , r2 threshold >0 . 8 . SNPs within other functionally important elements such as CDKN2A and CDKN2B promoters [26] , [29] , [30] , [75] or a putative ANRIL promoter region ( which we arbitrarily defined as 1kb up and downstream of the transcription start site ) were selected if they were reported more than once in NCBI dbSNP , had expected heterozygosity >5% , and were associated with alteration of transcription factor binding sites ( using PROMO v . 3 . 0 . 2 ) [76] , [77] . Details of included SNPs are shown in Table S1 . Multiplex SNP genotyping was performed by primer extension and MALDI-TOF mass spectrometry using iPLEX Gold technology from Sequenom ( Sequenom Inc , San Diego , USA ) . SNP assays were designed using Sequenom's RealSNP ( www . RealSNP . com ) and MassARRAY Assay Design v3 . 0 Software ( multiplex details and primer sequences available in Table S4 ) . PCR was performed using 20ng of DNA in a 10µl reaction volume for 35 cycles using standard iPLEX methodology . Spectra were analysed using MassARRAY Typer v3 . 4 Software ( Sequenom ) . Spectra and plots were manually reviewed and auto-calls were adjusted if required . Positive and negative controls were included . Individual samples with low genotype call rates ( <80% ) and SNP assays with poor quality spectra/cluster plots were excluded . Correspondence to Hardy-Weinberg proportions was checked for each SNP . PCR primers for the selected transcribed SNPs were designed using Primer3 ( v . 0 . 4 . 0 ) software [78] . CDKN2A primers span exons 3–4 and include both transcribed SNPs ( rs3088440 and rs11515 ) in the same amplicon . ANRIL primers span exons 1–2 and include both transcribed SNPs ( rs10965215 and rs564398 ) in the same amplicon . For CDKN2B , separate primer pairs for transcribed SNPs rs1063192 and rs3217992 were designed entirely within exon 2 ( due to the distance of transcribed SNPs from the exon boundary ) . Quantification of the allelic expression ratio was performed by primer extension and MALDI-TOF mass spectrometry using iPLEX Gold with similar parameters to the genotyping assay . Spectra were analysed using MassARRAY Typer v3 . 4 Software ( Sequenom ) and allelic ratios were estimated as the ratios of the area under the peak representing allele 1 to that representing allele 2 . Measurements were performed in four replicates using 50ng of cDNA template . Results from amplification of genomic DNA were used as an equimolar reference to normalise the cDNA values . Genomic normalisation reactions for CDKN2B used the same PCR primers as used for cDNA , but for CDKN2A and ANRIL ( where primers were cDNA-specific ) separate assays designed to be as close as possible in size and location to the cDNA primers were used . Primer sequences are shown in Table S3 . For some assays the allelic ratios measured in gDNA ratios did deviate from a 1∶1 ratio , as shown in Table S4 , confirming that allelic ratios in cDNA required correction for assay bias . However , as expected the gDNA ratios for each assay were relatively homogeneous with little inter-individual variability compared to cDNA ratios ( Figure S1 and Figure S2 ) . We compared the results of expression mapping using two different normalisation strategies in the SA cohort: normalising to a mean population normalisation factor versus normalising each individual's cDNA to their own gDNA ratio . There was no difference in the results obtained using these two normalisation strategies , as shown in Figure S8 . The mean gDNA ratios for each assay were the same in the SA cohort and a sample of Caucasian individuals ( no significant difference using a two sample t-test ) , and we therefore used the mean gDNA ratios for normalisation of all samples . The appropriateness of genomic normalisation ratios and linearity of the AER response were checked by mixing PCR products from individuals homozygous for the minor and major alleles in varying ratios ( 8∶1 , 4∶1 , 1∶1 , 1∶4 , 1∶8 ) and using these as template for the allelic expression assays . These experiments confirmed that allelic expression showed a linear response and that normalisation ratios obtained using allelic expression assays on a 1∶1 mixture of alleles for each SNP correspond to normalisation ratios obtained from genomic DNA ( Table S4 and Figure S9 ) . Allelic expression ratios for the two transcribed markers in each gene were highly correlated ( CDKN2A r = 0 . 68 , p = 1 . 7×10−3; CDKN2B r = 0 . 80 , p = 1 . 7×10−12; ANRIL r = 0 . 90 , p = 1 . 0×10−26; all genes combined r = 0 . 96 , p = 3×10−61 ) as shown in Figure S5; we therefore used a novel approach of combining allelic ratios from the two transcribed markers in each gene to increase the number of informative heterozygotes . Real-time PCR reactions were performed using TaqMan gene expression gene expression probes and reagents ( Applied Biosystems ) and run on a 7900HT Real-Time PCR System ( Applied Biosystems ) . Commercially available FAM-labelled TaqMan assays were used for CDKN2A exons 2–3 ( Hs00923894_m1 ) and ANRIL exons 1–2 ( Hs01390879_m1 ) . A custom FAM-labelled assay was used for exon 2 of CDKN2B . Commercially available VIC-labelled TaqMan assays were used for three reference genes shown to be suitable for normalisation of expression in peripheral blood [79] , [80]: B2M ( 4326319E ) , GAPD ( 4326317E ) , and HPRT1 ( 4326321E ) . TaqMan assays are validated by the manufacturer to have close to 100% amplification efficiency and assays were selected to quantify the same transcripts as the allelic expression assays . PCR was performed according to the manufacturer's protocol using four replicates , 25ng cDNA template per reaction , and the following multiplex combinations: CDKN2A/B2M , CDKN2B/GAPD , and ANRIL/HPRT1 . Relative total expression was analysed using the comparative cycle threshold ( Ct ) method . Ct values for each target gene were normalised to the mean Ct value of the three reference genes [79] . Standard errors and variances of measurements for allelic and total expression analyses in the SA population are shown in Table S5 . The association between total expression , as measured by real time PCR , and each of the SNPs was assessed using linear regression of the log transformed normalized expression values on the genotype assuming no dominance or interactions between the effects of different SNPs . The effect of including age , sex , and ethnicity as covariates , as well as excluding outlying individuals as determined by visual inspection ( highlighted in Figure S4 ) were investigated . Self reported ethnicity was included as a categorical variable ( categorised as “Cape mixed-ancestry” , “black African” , “white” , “Indian” , and “other” ) . These corrections made no significant difference to the results of eQTL mapping ( Figure S6 ) . All analyses were performed using the corrected data . Plots illustrating the associations between genotype and total expression for selected SNPs are shown in Figure S10 . We analysed allelic expression ratios using an extension of the approach we published previously [52] . We restrict ourselves to biallelic markers , and code one arbitrarily chosen allele as 0 and the other as 1 . We designate with g the phase-known and with T the phase-unknown genotype of an individual . The latter can be ascertained through genotyping . We assume that the amount of mRNA originating from a single allele follows a lognormal distribution where the variance does not vary between different alleles . The log of the ratio between the expression levels of both alleles , I , can therefore be assumed to be normally distributed . For an individual that is heterozygous for m transcribed polymorphisms , m ratios can be determined . We designate the vector of the logarithms of these ratios as . Under the assumptions above , the components of I are normally distributed with where the means depend on the genotype but the variance is genotype independent but may depend on the site used to measure the allelic expression ratio . We model the expected value as a linear combination of the influences of the typed polymorphisms:where represents the effect of the ith cis acting markers; and characterizes the phase between transcribed and putative cis acting markers:In order to assess the association between a specific SNP and allelic expression , let us consider a set of individuals . For an individual ( ) we can measure the unphased genotype and a vector representing the log of the allelic expression ratios . Up to a multiplicative constant the likelihood of observing a certain pattern of imbalance in this set of individuals given their genotyping results is:withwhere designates the probability of the phased genotypes g given the genotyping results and describes the density of the distribution of given the genotype g . was estimated using the hap procedure from the R-package gap ( as deposited in the CRAN archive http://cran . r-project . org/ ) to phase the genotypes of the two populations separately . We assume that the allelic expression ratios measured at different sites are conditionally independent given the genotype . Therefore:where and denotes the density of a normal distribution with the individual expression ratio as variate , a genotype dependent mean and a variance . Therefore depends on ( i = 1 , … , n ) and ( k = 1 , … , m ) , and maximisation of this likelihood allows assessment of the effects of single or groups of SNPs and to adjust for the effects of other markers by comparing nested models using likelihood ratio tests . For both total and allelic expression multiple testing was taken into account by calculating the family wise error rate using a Bonferroni correction for the 56 SNPs tested . Associations with family wise error rate below a threshold of 0 . 05 ( corresponding to a nominal P-value of 8 . 9×10−4 , −log10P of 3 . 05 , and −log10FWER of 1 . 3 ) were called significant . From our allelic and total expression data we also estimated the proportion of total expression variance that is due to cis-acting effects . This assumes that cis and trans-acting factors act in an additive manner , do not interact , are independent and that there is random mating , no segregation distortion , and the locus is not subject to imprinting . Given these assumptions , we estimate the variance due to cis acting effects , , as , where is the allelic expression ratio for individual i , and the proportion of the total variance due to cis acting effects can be estimated as , where is the estimated total variance , i . e . with and represent the total expression level for individual i as determined by real time PCR . | Genetic variants on chromosome 9p21 have been associated with several important diseases including coronary artery disease , diabetes , and multiple cancers . Most of the risk variants in this region do not alter any protein sequence and are therefore likely to act by influencing the expression of nearby genes . We investigated whether chromosome 9p21 variants are correlated with expression of the three nearest genes ( CDKN2A , CDKN2B , and ANRIL ) which might mediate the association with disease . Using two different techniques to study effects on expression in blood from two separate populations of healthy volunteers , we show that variants associated with disease are all correlated with ANRIL expression , but associations with the other two genes are weaker and less consistent . Multiple genetic variants are independently associated with expression of all three genes . Although total expression levels of CDKN2A , CDKN2B , and ANRIL are positively correlated , individual genetic variants influence ANRIL and CDKN2B expression in opposite directions , suggesting a possible role of ANRIL in CDKN2B regulation . Our study suggests that modulation of ANRIL expression mediates susceptibility to several important human diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"cardiovascular",
"disorders/coronary",
"artery",
"disease",
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/cancer",
"genetics",
"genetics",
"and",
"genomics/gene",
"expression"
]
| 2010 | Chromosome 9p21 SNPs Associated with Multiple Disease Phenotypes Correlate with ANRIL Expression |
Outbreaks of H5N1 in poultry in Vietnam continue to threaten the livelihoods of those reliant on poultry production whilst simultaneously posing a severe public health risk given the high mortality associated with human infection . Authorities have invested significant resources in order to control these outbreaks . Of particular interest is the decision , following a second wave of outbreaks , to move from a “stamping out” approach to the implementation of a nationwide mass vaccination campaign . Outbreaks which occurred around this shift in policy provide a unique opportunity to evaluate the relative effectiveness of these approaches and to help other countries make informed judgements when developing control strategies . Here we use Bayesian Markov Chain Monte Carlo ( MCMC ) data augmentation techniques to derive the first quantitative estimates of the impact of the vaccination campaign on the spread of outbreaks of H5N1 in northern Vietnam . We find a substantial decrease in the transmissibility of infection between communes following vaccination . This was coupled with a significant increase in the time from infection to detection of the outbreak . Using a cladistic approach we estimated that , according to the posterior mean effect of pruning the reconstructed epidemic tree , two thirds of the outbreaks in 2007 could be attributed to this decrease in the rate of reporting . The net impact of these two effects was a less intense but longer-lasting wave and , whilst not sufficient to prevent the sustained spread of outbreaks , an overall reduction in the likelihood of the transmission of infection between communes . These findings highlight the need for more effectively targeted surveillance in order to help ensure that the effective coverage achieved by mass vaccination is converted into a reduction in the likelihood of outbreaks occurring which is sufficient to control the spread of H5N1 in Vietnam .
Highly pathogenic avian influenza subtype H5N1 was first identified in Vietnam in December 2003 [1] and was followed by a major wave of outbreaks dispersed widely throughout the country around the 2004 festival of Tê′t , a New Year celebration representing a peak in annual poultry production and consumption . Measures employed to control both this and a second wave of outbreaks , occurring a year later , included a ‘stamping out’ policy with compulsory mass ring culling of all poultry around an outbreak carried out . This led to over 40 million poultry being culled during the first wave alone , contributing to an estimated direct national loss of US$ 200 million [2] , a sharp decline in demand for poultry products [3] and an estimated loss to affected individual stakeholders of between US$ 69 and US$108 [4] , an amount exceeding the average monthly wage in Vietnam . The resources and scale of destruction necessary to maintain these measures and the repeated incursions of H5N1 which occurred despite them resulted in a shift in policy in 2005 [5] . Following a field trial and pilot campaign in two test provinces , a nationwide mass vaccination campaign began , using an H5N2 vaccine to inoculate chickens and a recombinant H5N1 vaccine for ducks [6] , with priority given to recently affected areas and those with high human and poultry densities [7] . Approximately 160 million doses were administered twice a year costing close to $US 21 million [8] . Although a third wave of outbreaks coincided with the beginning of this campaign , a sero-survey of poultry following vaccination estimated that a 60% level of protective coverage had been achieved [6] . A nationwide ban on the hatching of waterfowl was also enforced from February 2005 [7] . Following this third wave , Vietnam experienced a year during which no outbreaks were reported . Despite follow-up vaccination campaigns from August to November 2006 and March to June 2007 , a fourth wave of outbreaks occurred in the South of Vietnam in December 2006 and a fifth , beginning in late April 2007 , in the North , coinciding with the lifting of the ban on duck hatching . During both of the latter waves , in order to minimise losses to farmers , ring vaccination was carried out around each reported outbreak and only flocks within which infection had been identified were culled [2] . Detailed records were not kept during the first two weeks of the first wave [5] . However , following the implementation of systematic community-based and veterinary surveillance and reporting networks as part of a detailed National Action Plan [9] , temporal and spatial outbreak data were collected at a commune-level resolution for all subsequent waves , providing a unique opportunity to assess the effects of the interventions , and in particular country-wide vaccination . Here we focus upon northern Vietnam , where data were available for the sizeable waves occurring before , during and after vaccination was implemented , with outbreaks generally concentrated around the high risk Red River Delta ( RRD ) region ( Fig . 1 ) . However the analysis of this data is challenging because of the inherent dependency between infection events ( i . e . the risk of infection of one commune depends on the infection status of all neighbouring communes ) and missing data issues ( the time of first infection of a commune is not known as only reporting dates have been recorded ) . We therefore developed an inter-commune transmission model , characterised by a spatial kernel describing how infectivity scales by distance and a parametric infection-to-report distribution , in order to capture the spatial and temporal dynamics of the spread of infection in the context of Vietnam where the smallholding of poultry is commonplace . Bayesian MCMC data augmentation methods which treat infection times as nuisance parameters [10]–[14] were then used in order to impute the missing infection times and fit this model to the observed commune-level outbreak reports . This allowed us to explore the changes which occurred between waves , both in terms of the transmissibility between communes and the rate at which outbreaks were detected . Next , through an analysis of the possible chains of transmission ( the “epidemic tree” ) , we investigated how changes in the rate of reporting outbreaks affected the size and duration of the wave of outbreaks which occurred following the implementation of vaccination .
Our results suggest that the expected daily number of secondary outbreaks generated by one infected commune varied little between the waves occurring before and during the vaccination campaign . However , during the 2007 wave , following vaccination , this measure of spread reduced substantially as a result of a significant reduction in the daily probability of transmission between communes ( Fig . 2a–c ) , with a joint posterior mean estimate suggesting that , per head of poultry , infectivity was only 55% of that during the 2004/5 wave ( Fig . 2d ) . Making the assumption that changes in infectivity are caused by vaccination and not influenced by other factors including changes in the effectiveness of other control measures such as bio-security or movement restriction , this can be compared to the effects of a vaccination campaign achieving 55% uniform effective coverage ( where effective coverage refers to the overall vaccination coverage multiplied by the protective efficacy of the vaccine ) . In contrast , the infectious period of communes increased significantly during the 2007 wave ( Table 1 and Fig . 3b ) . Consequently , despite the coincident reduction in the daily transmission probability , the RRD region continued to sustain infection ( Fig . 3a and c ) , resulting in a wave which was lower in intensity but longer lasting than the preceding two waves ( Fig . 4a ) . Overall , accounting for changes in the distribution of poultry between the 2004/5 and 2007 waves , there was an 11% reduction in the number of communes with estimated local reproductive numbers ( see Methods and Text S1 ) above unity The posterior mean commune-level infectious period decreased from an estimated 5 . 9 days during the 2004/5 wave to just 4 . 7 days during the 2005 wave ( Fig . 3b ) , coinciding with an increase in the level of compensation provided to farmers with infected or depopulated flocks [15] . As a result , despite the fact infection spread more rapidly in the initial stages , the wave was brought under control more rapidly than during the other two waves ( Fig . 4a ) . There was also a high proportion , relative to the other waves , of transmission taking place over distances of less than 20km ( Fig . 4b ) . We assessed the robustness of our qualitative conclusions about the changes in the dynamics of infection following vaccination to different assumptions about the infectious period and effects of control measures with detailed sensitivity analyses ( see Text S1 ) . We were unable to explain the reduction in infectivity and increase in the infectious period following vaccination by allowing outbreaks to remain infectious for a longer time following a report or by modelling non-constant infectivity throughout the infectious period . We also found that the estimated reduction in the daily probability of transmission ( Fig . 2b ) could not be attributed to a decrease in the proportion of outbreaks detected . Treating each outbreak as the root of a separate sub-epidemic or “clade” [16] , where a clade is defined to consist of all future generations of outbreaks arising from an individual infection ( see Methods and Text S1 ) , we explored the impact more effective surveillance would have had upon the 2007 wave of outbreaks . We made the assumption that this would have primarily resulted in the earlier reporting of outbreaks estimated to have remained unreported for the longest time . . We found that , had all outbreaks been identified and successfully removed within two weeks of the initial infection , the expected eventual size of the wave would have been halved and the wave would have been eliminated twice as fast , with an 18% probability that the wave would have become extinct within the first 10 outbreaks . This probability rises to 20% , with a 67% reduction in the size of the wave , if the rate of detection had been maintained at that estimated for the 2004/5 wave ( Fig . 4c ) , prior to the implementation of vaccination .
Our results demonstrate the application of data augmentation techniques to existing livestock disease modelling methodologies in order to quantify the spatial and temporal spread of disease in a setting where the times of infection are unobserved . We found that , following the implementation of vaccination , the day-to-day probability of infection spreading between communes has been significantly reduced , with our estimate of a 55% effective vaccination coverage agreeing with post-campaign sero-surveillance [6] . However , we also found that the duration of time taken to report outbreaks had also increased significantly , allowing infection to spread . This result may support the hypothesis that vaccination within a flock can contribute to the “silent spread” of infection whereby a low-level of flock mortality or asymptomatic infections can make outbreaks more difficult to detect [17] , [18] and this was identified locally as an exacerbating factor in the spread of infection [19] . There does , however , also appear to have been a shift in the distribution of host species involved in outbreaks . During the 2004/5 wave , only 30% of outbreaks were identified in ducks whereas in 2007 this figure rose to 77% [7] . This change has previously been attributed to the lifting of a duck-hatching ban in February 2007 . Scavenging ducks traditionally play an important role in the poultry production system and they have previously been identified as being associated with outbreaks of HPAI [7] . Outbreaks also coincided with the end of rice-harvest season when ducks are allowed to graze freely on rice paddies [20] . H5N1 infection in ducks has been shown to be less pathogenic than in chickens , with clinical signs being less apparent and slower to develop [21] , [22] . As a result , the concentration of infection within duck flocks may have contributed to an elongated commune-level infectious period . Changes in the transmissibility or reporting rates of outbreaks may also be linked with other factors such as differential vaccine efficacy or observed changes in the genotype of the virus [23] . We found that a large proportion of transmission occurring concurrently with the first round of vaccination took place over relatively short distances . Such an increase could be due to the movement of vaccinators who , it has been suggested , may have acted as carriers of infection from one commune to another [5] . It is also interesting that our estimates didn't suggest a reduction in the daily per-capita poultry transmissibility 2005 , despite the concurrent vaccination campaign which was taking place . This may suggest that outbreaks predominantly occurred within large regions yet to be vaccinated or in areas where vaccinated poultry had yet to develop effective immunity . We demonstrated how imputed infection times can be used to explore all possible chains of infection and to assess the range of possible effects which can be achieved by ‘pruning’ an outbreak from the epidemic tree . Applying this to outbreaks occurring during the 2007 wave , we found that if the reporting rates had been maintained to those achieved during the previous waves , the size and the duration of wave would have been considerably diminished . This highlights that , in order to ensure that these reductions in infectivity achieved through vaccination ( Fig . 2a , c , d ) are translated into a noticeable impact on outbreak size , it is essential that the infectious period of communes is shortened ( Fig . 3b ) via improved surveillance and reporting . Measures towards this have already been attempted in Vietnam with the trial of a pro-active surveillance programme whereby resources are allocated according to known risk factors such as the presence of live bird markets and high semi-commercial farm density and a prior history of outbreaks [24] . The programme also includes an awareness campaign informing stakeholders about how outbreak detection criteria should change following the vaccination of a flock . An alternative approach is to use unvaccinated birds , or those vaccinated with an inactivated heterologous strain of avian influenza , as sentinels for the detection of the spread of infection . This has been implemented under the Differentiating Infected from Vaccinated Animals ( DIVA ) guidelines in developed countries such as Italy and the U . S . and has been successful in helping to control outbreaks of low pathogenic avian influenza ( LPAI ) [25] , [26] . Given the size of , and low-levels of biosecurity within , the poultry industry and the ubiquity of poultry-keeping in Vietnam [27] , developing such detection capacity will involve overcoming a large variety of logistical , economic and social issues [3] . However , in the absence of such measures , sporadic outbreaks of the kind currently being experienced in Vietnam , and the public health risks they entail [28] , seem set to re-occur for the foreseeable future .
By drawing samples of infection times from the MCMC output it is possible to calculate the marginal probability that any given outbreak is infected by any other outbreak: It is then possible to calculate both the expected number of infections arising from each individual outbreak commune and the expected distance over which infection was transmitted ( see Text S1 ) . Moreover , in order to assess the effects of earlier detection , realisations of the potential epidemic tree ( i . e . the unambiguous chain of transmission where each outbreak is assigned a source infector ) can be repeatedly sampled for each set of infection times by randomly selecting a source outbreak for each outbreak according to these probabilities . From this , secondary infections which arose between the “real-life” removal time of an outbreak and a putative earlier removal time , occurring as a result of improved surveillance , were identified , according to the assumed impact of this surveillance . Then , by pruning clades of outbreaks attributed to these secondary infections [16] , the epidemic tree which would have occurred as a result of the earlier removal time is obtained ( assuming pruned outbreaks would not have been infected from alternative sources at a latter stage of the wave ) . This form of analysis was used to assess the impact that the surveillance scenarios listed in Fig . 4 would have had upon the 2007 wave of outbreaks ( see Text S1 ) . | Highly pathogenic avian influenza H5N1 continues to spread rapidly between flocks of poultry in many parts of the world including areas in Southeast Asia and Africa where infection has become endemic . Meanwhile the number of human cases and fatalities are steadily accumulating . As a result , the control of outbreaks in poultry remains both a key public and animal health priority . In Vietnam control policies have evolved from a policy of reliance upon drastic “stamping out” measures to regular mass vaccination campaigns . Using Bayesian data augmentation techniques in order to take into account the unobserved infection times , we found that this has led to a significant reduction in the daily probability of transmission between communes but that the time taken to detect outbreaks has increased . As a result it is still possible for sustained transmission to occur , albeit at a slower rate . Through an analysis of the reconstructed epidemic tree , we found that any measures that have the effect of restoring detection capacity to that estimated for the “stamping out” policy would have a large effect upon the size and scale of any future outbreak wave and may be effective at preventing sustained transmission between communes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"public",
"health",
"and",
"epidemiology/global",
"health",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
]
| 2010 | A Bayesian Approach to Quantifying the Effects of Mass Poultry Vaccination upon the Spatial and Temporal Dynamics of H5N1 in Northern Vietnam |
Triggering receptor expressed on myeloid cells ( TREM ) -1 and TREM-2 are key regulators of the inflammatory response that are involved in the clearance of invading pathogens . Melioidosis , caused by the "Tier 1" biothreat agent Burkholderia pseudomallei , is a common form of community-acquired sepsis in Southeast-Asia . TREM-1 has been suggested as a biomarker for sepsis and melioidosis . We aimed to characterize the expression and function of TREM-1 and TREM-2 in melioidosis . Wild-type , TREM-1/3 ( Trem-1/3-/- ) and TREM-2 ( Trem-2-/- ) deficient mice were intranasally infected with live B . pseudomallei and killed after 24 , and/or 72 h for the harvesting of lungs , liver , spleen , and blood . Additionally , survival studies were performed . Cellular functions were further analyzed by stimulation and/or infection of isolated cells . TREM-1 and TREM-2 expression was increased both in the lung and liver of B . pseudomallei-infected mice . Strikingly , Trem-2-/- , but not Trem-1/3-/- , mice displayed a markedly improved host defense as reflected by a strong survival advantage together with decreased bacterial loads , less inflammation and reduced organ injury . Cellular responsiveness of TREM-2 , but not TREM-1 , deficient blood and bone-marrow derived macrophages ( BMDM ) was diminished upon exposure to B . pseudomallei . Phagocytosis and intracellular killing of B . pseudomallei by BMDM and alveolar macrophages were TREM-1 and TREM-2-independent . We found that TREM-2 , and to a lesser extent TREM-1 , plays a remarkable detrimental role in the host defense against a clinically relevant Gram-negative pathogen in mice: TREM-2 deficiency restricts the inflammatory response , thereby decreasing organ damage and mortality .
In sepsis , defined as a deregulated host response to a life-threatening infection , a careful balance between inflammatory and anti-inflammatory responses is vital [1–3] . Pathogen- or danger-associated molecular patterns are recognized by intracellular sensory complexes and cell surface receptors expressed on innate immune cells that can initiate the inflammatory and anti-microbial response . Well-known examples of these pattern recognition receptors ( PRRs ) are the Toll-like receptor ( TLR ) , nucleotide-oligomerization domain-like receptor ( NLR ) and C-type lectin receptor ( CLR ) families [4] . A more recently discovered group of innate immune receptors are the membrane-bound triggering receptors expressed on myeloid cells ( TREMs ) , which act as key modulators , rather than as initiators , of the inflammatory response [5–7] . TREM-1 and TREM-2 are the most studied members of the TREM-family , however their exact role in the pathogenesis of sepsis remains ill-defined . Upon recognition of partially still unspecified ligands , both receptors phosphorylate the adaptor molecule DNAX adaptor protein 12 ( DAP12 ) after which the cellular response is initiated [8 , 9] . Only recently , binding of TREM-1 to a complex of peptidoglycan recognition protein 1 ( PGLYRP1 ) and bacterially derived peptidoglycan has been demonstrated [10] . TREM-1 is expressed on neutrophils and monocyte subsets [11] and amplifies pro-inflammatory TLR-mediated responses in vitro [12] . There are conflicting reports on the role of TREM-1 in in vivo infection models . TREM-1 deficiency impaired bacterial clearance in a model of Klebsiella pneumonia-induced liver abscess formation [13] , pneumococcal [14] and Pseudomonas ( P . ) aeruginosa pneumonia [15] . However , blocking TREM-1 with an analogue synthetic peptide derived from the extracellular moiety of TREM-1 ( LP17 ) actually improved survival during gram-negative sepsis [16] and endotoxaemia [17] . Interestingly , in a murine pneumonia model of Legionella pneumonia no impact of TREM-1 deficiency was found on bacterial clearance or neutrophil influx towards the primary site of infection [18] . TREM-2 is primarily expressed on macrophages , dendritic cells , microglia and osteoclasts [19–22] and has been suggested to bind to bacterial lipopolysaccharide ( LPS ) and lipotechoic acid [23] . In contrast to TREM-1 , TREM-2 acts as a negative regulator of inflammatory responses in macrophages and dendritic cells [19 , 21] . In addition , TREM-2 is involved in phagocytosis [24 , 25] and killing of bacteria by macrophages [26] . Blocking TREM-2 in vivo by a recombinant protein in a polymicrobial sepsis model revealed that TREM-2 is required for bacterial clearance and improves survival [27] . In contrast , TREM-2 plays a detrimental role during pneumococcal pneumonia [25] . Melioidosis , considered to be an illustrative model for Gram-negative sepsis , is caused by the Tier 1 biological treat agent Burkholderia pseudomallei [28 , 29] . Melioidosis is characterized by pneumonia and abscess formation and an important cause of community-acquired sepsis in Southeast Asia and Northern Australia [28] . The high mortality rate , that can approach 40% , and the emerging antibiotic resistance of B . pseudomallei [30] emphasize the need to better understand the pathogenesis of melioidosis , which could ultimately lead to novel treatment strategies . We previously found increased soluble ( s ) TREM-1 plasma levels and TREM-1 surface expression on monocytes of patients with melioidosis [31] , suggesting an important role for TREM-1 in the host defense against B . pseudomallei . Treatment with a peptide mimicking a conserved-domain of sTREM-1 partially protected mice from B . pseudomallei induced lethality [31] . In this study we now examine the role of TREM-1 and TREM-2 during experimental melioidosis , utilizing recently generated Trem-1/3-deficient ( Trem-1/3-/- ) [15] and Trem-2-deficient ( Trem-2-/- ) mice [19] to determine their contribution to the host response against B . pseudomallei . We hypothesized that TREM-1 deficiency would decrease inflammation and improve survival during murine melioidosis while TREM-2 deficiency would instead lead towards increased inflammation and a worsened survival . Unexpectedly however , we found that TREM-2 , but not TREM-1 , plays an important detrimental role during melioidosis . TREM-2 deficiency improves survival of B . pseudomallei infected mice , by limiting inflammation and organ damage . These data identify TREM-2 as a potential treatment target for sepsis caused by B . pseudomallei .
The Animal Care and Use of Committee of the University of Amsterdam approved all experiments ( DIX102273 ) , which adhered to European legislation ( Directive 2010/63/EU ) . Pathogen-free 8- to 10-week-old male wild-type ( WT ) C57BL/6 mice were purchased from Charles River ( Leiden , The Netherlands ) . Trem-1/3-/- [6 , 14] and Trem-2-/- [19] mice were backcrossed >97% to a C57BL/6 genetic background . B . pseudomallei , derived from our aliquoted frozen stock , was grown to log-phase and further diluted in sterile PBS ( 1x ) . Experimental melioidosis was induced by intranasal inoculation with 5 × 102 colony forming units ( CFU ) of B . pseudomallei strain 1026b ( a clinical isolate ) as described [32–34] . For survival experiments mice were observed 4–6 times daily , up to 14 days post-infection . Sample harvesting , processing , and determination of bacterial growth were performed as described in detail in the S1 Appendix[33 , 34] . All work concerning live B . pseudomallei was performed in a ( A ) BSL III facility . Chemo- and cytokine levels were determined in plasma , lung and liver . Distant organ damage was more closely assessed by plasma transaminases , lactate dehydrogenase ( LDH ) and blood urea nitrogen ( BUN ) levels . Total RNA was isolated using the Isolate II RNA mini kit ( Bioline , Taunton , MA , USA ) , treated with DNase ( Bioline ) and reverse transcribed using an oligo ( dT ) primer and Moloney murine leukemia virus RT ( Promega , Madison , WI , USA ) . Primers and RT-PCR conditions can be found in the supplemental data . Data were analyzed using the comparative Ct method . Paraffin-embedded 4-μm lung , liver and spleen sections were stained with haematoxylin and eosin and analyzed for inflammation and tissue damage , as described previously [14 , 34] . Granulocyte ( Ly6G ) staining was done exactly as described previously [35] . Whole blood , alveolar macrophages ( AM ) and bone-marrow derived macrophages ( BMDM ) were harvested from naïve WT and Trem1/3-/- and Trem-2-/- mice as described [34 , 36 , 37] and stimulated overnight with either medium , ultrapure LPS ( Invivogen , San Diego , CA , USA ) or B . pseudomallei ( 107 CFU/ml or MOI of 50 ) , after which supernatant was harvested and stored at -20°C until assayed for TNFα . Phagocytosis was determined as described previously [38] . In brief , AM and BMDM ( 5x 104 cells/well ) were incubated with or without heat-inactivated FITC-labelled B . pseudomallei ( MOI 50 ) for 60 and/or 120 minutes at 37°C and 5% CO2 air and internalization was measured directly after collection by flow cytometry . Bacterial killing was evaluated as described [36 , 39] . In short , BMDM were incubated with to log-phase grown B . pseudomallei ( MOI 30 ) for 20 minutes at 37°C in 5% CO2 air , after which they were washed and incubated with kanamycin 250 μg/ml for 30 minutes at 37°C in 5% CO2 air ( this point was taken as time zero ) [36] . At designated time points the BMDM were washed and lysed and appropriate dilutions of these lysates were plated onto blood-agar plates and incubated at 37°C for 24–48 h before CFU were counted . Values are expressed as mean ± standard error of the mean ( SEM ) . Differences between groups were analyzed by Mann-Whitney U test . For survival analysis , Kaplan-Meier analysis followed by log-rank test was performed . These analyses were performed using GraphPad Prism version 5 . 01 . Values of P< 0 . 05 were considered statistically significant .
Septic melioidosis patients present with pneumonia and bacterial dissemination to distant body sites [28 , 40] . Since it is not feasible to study TREM-1 and TREM-2 mRNA expression at tissue level in these patients , we used our well-established murine model of pneumonia-derived melioidosis in which mice are intranasally infected with a lethal dose of B . pseudomallei [33 , 34] . Mice were killed at 0 , 24 , and 72h after infection ( i . e . , directly before the first predicted death ) , and TREM-1/-2 mRNA expression was determined in lungs and livers . At baseline , TREM-1 and TREM-2 expression was low , corresponding with our previous data on sTREM-1 levels in melioidosis patients [31] , TREM-1 was strongly up-regulated in lung and liver tissue ( P<0 . 05 lung at 24h , P<0 . 01 liver at 72h; Fig 1A and 1B ) . TREM-2 mRNA expression was increased in experimental melioidosis as well ( P<0 . 05 in both lung and liver; Fig 1C and 1D ) . The increase in both TREM-1 and TREM-2 expression was much more pronounced at the primary site of infection , the pulmonary compartment , when compared to the hepatic compartment . Having established that both TREM-1 and TREM-2 are highly up-regulated during melioidosis , we further investigated the involvement of these receptors in the outcome of melioidosis . Therefore , we infected WT , Trem-1/3-/- and Trem-2-/- mice intranasally with a lethal dose of B . pseudomallei and observed them for 14 days ( Fig 2A and 2B ) . There was no significant difference in survival between Trem-1/3 and WT mice following a lethal B . pseudomallei challenge: 95% of Trem-1/3-/- and WT mice died within 6 days after inoculation ( Fig 2A ) . Strikingly however , Trem-2-/- mice were significantly protected: 70% of Trem-2-/- survived until the end of the 14-day observation period while all WT mice died within 6 days ( P< 0 . 001; Fig 2B ) . To substantiate the finding that Trem-2-/- mice are protected during melioidosis , we determined bacterial loads in lung and BALF as well as in blood , liver and spleen 72h post-infection . Relative to WT mice , Trem2-/- mice displayed strongly reduced bacterial loads both at the primary site of infection ( P<0 . 01 for lung and BALF; Fig 2C and 2D ) as well as in distant organs and the systemic compartment ( P<0 . 01 for blood and spleen; Fig 2E–2G ) . 72h post-infection 100% of WT but only 20% of Trem2-/- mice had become bacteraemic . These findings indicate that TREM-2 plays a key deleterious role during experimental melioidosis by antagonizing bacterial clearance leading to increased dissemination of infection . Since TREM-2 has been described as a negative regulator of inflammation [19 , 20] , we next assessed the inflammatory response in the pulmonary compartment . Therefore we studied the extent of inflammation in lung homogenates and BALF . We observed markedly decreased levels of pro-inflammatory cytokines TNF-α , IL-6 , IL-1β and the chemokine KC in both lung homogenates and BALF of TREM-2 deficient mice compared to controls ( P<0 . 01–0 . 05; Table 1 ) . To further obtain insight into the involvement of TREM-2 in the inflammatory response following B . pseudomallei infection , we semi-quantitatively scored lung histology slides generated from Trem-2-/- and WT mice . However , all mice displayed severe pulmonary inflammation and no differences were observed between the mouse strains ( Fig 3A–3C ) . Neutrophil recruitment to the lung is an essential part of the inflammatory host response to melioidosis . Therefore , we determined the granulocyte influx into the pulmonary compartment by Ly6G-immunostaining in WT and Trem-2-/- mice 72h post-infection with B . pseudomallei ( Fig 3D–3F ) . This immunostaining recognizes Gr-1 , that is granulocyte-specific , Corresponding to the diminished bacterial loads and decreased levels of cyto- and chemokines in lung tissue , a reduced influx of granulocytes in lungs of Trem-2-/- mice was found ( P<0 . 05 , Fig 3D ) . To evaluate the role of TREM-2 in the systemic inflammatory response , we determined plasma cytokine levels 72h post-infection with B . pseudomallei . Consistent with the lower pulmonary cytokine levels and bacterial loads , we found that the plasma levels of TNF-α , IL-6 , IL-1β , MCP-1 , IL-10 , IFN-γ and KC were all significantly reduced in Trem-2-/- mice compared to WTs ( P<0 . 01–0 . 05 , Table 1 ) . Furthermore , we obtained spleen pathology scores and performed routine clinical chemistry tests to evaluate hepatic , renal and systemic injury . In line with the observed decreased splenic bacterial loads , Trem-2-/- mice showed less inflammation compared to WT mice 72h after inoculation with B . pseudomallei ( P<0 . 05; Fig 4A ) . Plasma AST levels of Trem-2-/- mice were decreased when compared to controls 72h post-infection , reflecting decreased hepatocellular injury in these animals ( P<0 . 05; Fig 4B ) . Consistently , we observed a trend towards lower ALT , BUN and LDH levels in Trem-2-/- mice compared to controls suggesting less organ damage respectively ( Fig 4C–4E ) . Having established that TREM-2 plays an important deleterious role during experimental melioidosis and is involved in the inflammatory response , we next assessed what cells are responsible for these effects . It is known that blood monocytes , alveolar macrophages ( AM ) and BMDM express TREM-2 [25] , therefore we harvested these cells and first stimulated them overnight with the TLR4-ligand LPS and B . pseudomallei . We found a clear trend towards lower TNF-α levels when whole blood , AM or BMDM of Trem-2-/- mice were stimulated with LPS ( Fig 5A–5C ) . This effect was even more pronounced after stimulation with B . pseudomallei: the TNF-α response of whole blood and BMDM derived from TREM-2 deficient mice was significantly reduced compared to controls ( P<0 . 05; Fig 5A and 5B ) . Considering TREM-2’s known phagocytic properties [24 , 25] and the observed lower local and systemic bacterial loads in TREM-2-deficient mice , we determined the phagocytic capacity of AM and BMDM harvested from WT and Trem-2-/- mice . Despite a trend towards enhanced phagocytosis of FITC-labelled B . pseudomallei by TREM-2 deficient macrophages , no significant differences were found ( Fig 5D and 5E ) . In line , TREM-2 did not impact on the intracellular killing of B . pseudomallei by BMDM ( S1 Fig ) . In a final set of experiments we studied the role of TREM-1 in the host defense against B . pseudomallei using Trem-1/3-/- mice . In contrast to the data derived from Trem-2-/- mice , no differences in bacterial counts in lung or BALF were observed between B . pseudomallei-challenged Trem-1/3-/- and WT mice ( Fig 6A and 6B ) . In line , TREM-1 deficiency did not impact on lung pathology and cytokine levels , except for decreased KC levels , which did not influence pulmonary neutrophilic content as determined by Ly6-stainings ( Fig 6E and 6F , Table 2 ) . However , TREM-1 did influence bacterial dissemination as bacterial loads in blood and liver were significantly decreased in Trem-1/3-/- mice compared to WTs 72h after infection ( P<0 . 01; Fig 6C and 6D ) . We next evaluated TREM-1’s role in systemic inflammation and end organ damage . At 72h post-infection , the levels of key regulatory cytokines in the systemic compartment ( TNF-α , IL-6 , IL-10 , MCP-1 and IFN-γ ) did not differ between Trem-1/3-/- mice and WT ( Table 2 ) . Induced pathology of the spleen ( Fig 6G ) was similar in Trem-1/3-/- and WT mice . In correspondence with the lower hepatic bacterial counts at 72h , we found lower levels of the hepatocellular injury markers AST and ALT levels in Trem-1/3-/- mice compared to WT mice ( Fig 6H and 6I ) . LDH levels , reflecting general organ injury , were elevated in Trem-1/3-/- mice at 24 h , while they were reduced compared to their WT counterparts at 72h post-infection ( P<0 . 05; Fig 6J ) . No difference in plasma BUN levels was observed between mice strains ( Fig 6K ) . TREM-1 is abundantly expressed on monocytes and macrophages following exposure to B . pseudomallei [31] . In line with previous findings [11] , Trem-1/3-/- BMDM produced less TNF-α in response to LPS stimulation ( P<0 . 05; Fig 7B ) . Surprisingly , no differences in cellular responsiveness were found between AM and whole blood derived from WT and Trem-1/3-/-mice ( Fig 7A–7C ) . Lastly , we wished to determine whether TREM-1 contributes to phagocytosis and/or killing of B . pseudomallei . No differences in phagocytic and killing capacities between WT and TREM-1 deficient BMDM were observed ( Fig 7D and 7E ) .
TREM-1 and TREM-2 are innate immune receptors that have demonstrated to either amplify or regulate TLR and NLR signaling after recognition of pathogen-associated molecular patterns . Our study is the first to examine the role of both TREM-1 and TREM-2 during experimental melioidosis . We observed increased TREM-1 and TREM-2 expression during experimental melioidosis , both at the local site of infection and systemically . Subsequently , we found that TREM-2 impairs the host defense against murine B . pseudomallei-induced sepsis , as demonstrated by an improved survival of infected Trem-2-/- mice as a direct result of diminished bacterial dissemination , decreased inflammation and less organ damage . Our ex vivo studies suggest that the protective effect of TREM-2 deficiency in part results from the diminished capacity of TREM-2-deficient macrophages to elicit a pro-inflammatory response which is an important contributor to organ injury in the event of sepsis . TREM-1 was also found to play a detrimental role during B . pseudomallei infection , which is in line with our earlier finding that blocking TREM-1 could improve survival during melioidosis [31] . However when compared to TREM-2 the role of TREM-1 in the host response against B . pseudomallei seems to be limited . Previous studies have demonstrated that soluble TREM-1 levels are up-regulated in plasma of patients with sepsis , pneumonia and melioidosis [31 , 41 , 42] . In addition , it is known that surface TREM-1 expression is increased on monocytes of melioidosis patients [31] . However , soluble TREM-1 levels in septic patients do not always correlate to the expression of membrane-bound TREM-1 on different myeloid cell types [31 , 43] . Less is known about the kinetics of TREM-2 expression during infection . A recent study demonstrated that during sepsis TREM-2 expression on ascites-retrieved cells of patients with abdominal sepsis was increased [27] . Correspondingly , TREM-2 was up-regulated on AM of mice infected with S . pneumoniae [25] . In line with these earlier studies , we now show that both TREM-1 and TREM-2 mRNA expression is elevated in lung and liver tissue of mice infected with B . pseudomallei . Further research however is warranted to study the cell surface protein expression of TREM-2 on neutrophils and macrophages during melioidosis . The in vivo role of TREM-2 in infectious diseases remains ill defined . In a model studying P . aeruginosa keratitis TREM-2 deficiency increased corneal bacterial loads [44] . More recently , Chen et al . demonstrated that TREM-2 is required for efficient bacterial clearance in a murine polymicrobial sepsis model using a TREM-2 blocking recombinant protein [27] . In the same study it was shown that administration of TREM-2 overexpressing bone marrow derived myeloid cells improved survival during polymicrobial sepsis , but not endotoxaemia [27] . In sharp contrast , Gawish et al . demonstrated a beneficial effect of TREM-2 deficiency during endotoxaemia [45] . The same group also observed a survival benefit of Trem-2-/- mice during S . pneumoniae pneumonia [25] , while no effect on mortality of TREM-2 deficiency was seen during E . coli sepsis [45] . To evaluate how TREM-2 deficiency led to increased clearance of B . pseudomallei , we assessed the functional roles of macrophages that express TREM-2 [19 , 25] . TREM-2 is known to be involved in direct killing [27 , 44] and phagocytosis of bacteria by macrophages [24 , 25] . Interestingly , we did not find impaired bacterial killing or phagocytosis of B . pseudomallei by BMDM or AM of Trem-2-/- mice . Several characteristics of this facultative intracellular bacterium when compared to other bacteria might in part explain these discrepancies; B . pseudomallei is capable of invading both phagocytic and non-phagocytic cells [46] and circumvents intracellular defense mechanisms efficiently in order to replicate and spread to adjacent cells [47 , 48] . TREM-2 is traditionally regarded as a negative regulator of the in vitro inflammatory response in response to TLR-ligands [19 , 21 , 45] In contrast , our study now demonstrates that TREM-2 deficiency leads to a reduced inflammatory response to B . pseudomallei both ex vivo and in vivo suggesting that TREM-2’s role during inflammation may not be that upfront . This is in line with recent studies investigating the role of TREM-2 in models of pneumococcal pneumonia [25] , post-stroke inflammation [49] and DSS-induced colitis [50] . Different elements , can explain these inconsistencies: differences in mice strains used ( BALB/C versus C57Bl/6 ) , different experimental murine models ( e . g . caecal ligation and puncture ( CLP ) - model versus a intranasal inhalation model for sepsis ) , differences in TREM-2 blockade ( e . g . by using TREM-2 deficient mice or TREM-2 antibodies ) and lastly the difference of an in vitro approach in contrast to our ex vivo cellular challenge model . Interestingly , a recent study showed augmented inflammation by TREM-2 deficient peritoneal macrophages in response to LPS [45] , while the same group observed the reversed phenotype in alveolar macrophages [25] , underlining possible cell-specific functions of TREM-2 . Of importance , neutrophil recruitment to the lung , an important defense mechanism during melioidosis [32 , 51] , was reduced in Trem-2-/- mice during experimental melioidosis as determined by Ly6-staining . This may be a potential result of the decreased inflammatory response and production of chemokines following infection . In this respect , it is noteworthy , that IL-1β–which we and others have shown to be involved in excessive deleterious neutrophil influx during experimental melioidosis [37 , 52]—was also significantly reduced in Trem-2-/- mice . No differences were observed in the influx of macrophages ( S2 Fig ) . Excessive inflammation and neutrophil influx and activation can lead towards multi-organ failure [53] , which is almost universally seen in lethal cases of melioidosis . Distant organ injury was significantly reduced in Trem-2-/- mice , potentially as a result of a reduced influx of inflammatory cells . Trem-2-/- mice displayed an evidently reduced inflammatory response , which resulted in a strong survival benefit . In addition , it is well known that B . pseudomallei can replicate intracellularly [28] , and neutrophils may act as its permissive host cell [52] . We could therefore hypothesize that the anti-inflammatory phenotype and the reduced bacterial loads seen in TREM-2 deficient mice are a result of decreased intracellular bacterial replication at the infection site , due to reduced neutrophilic influx . Taken together , during melioidosis , TREM-2 deficiency resulted in a restricted inflammatory response , thereby decreasing organ damage and mortality . Future research should focus on the potential of anti-TREM-2 treatment of B . pseudomallei-infected mice . TREM-1 amplifies TLR-responses and therefore might dangerously enhance the inflammatory response to bacterial infection [18] . Controversial results have been found on the role of TREM-1 during bacterial infection . TREM-1 deficiency has shown to be detrimental during endotoxaemia [17] and polymicrobial sepsis [12 , 54] , while in contrast , moderate levels of TREM-1 can improve survival during polymicrobial sepsis , but not endotoxaemia [55] . Blockade of TREM-1 with a peptide called LP17 could partially protect mice from B . pseudomallei induced lethality [31] . In this study however , we observed , using the same infection model , that survival of B . pseudomallei-infected TREM-1-deficient mice was similar to WTs . This might be explained by the fact that these mice were completely TREM-1-deficient and in addition lacked TREM-3 , a DAP12-coupled activating receptor on murine macrophages , which supposedly acts as an activating receptor [56] . In contrast , in humans TREM-3 is a pseudogene [56] . However , since DAP12 is known to both potentiate and attenuate TLR-signaling , it is perhaps not surprising that the net-effect on bacterial clearance of B . pseudomallei is not affected . TREM-1 has other functions next to TLR-signaling enhancement , such as phagocytosis and the production of reactive oxygen species [57] . Furthermore , TREM-1 has been recently linked to trans-epithelial migration of neutrophils after infection with P . aeruginosa [15] . Blocking TREM-1 completely could therefore interfere with these important antibacterial mechanisms . We did not find a role for TREM-1 in the killing or phagocytosis of B . pseudomallei , which is in line with the fact that TREM-1/3 deficiency in neutrophils neither impacts on bacterial killing , phagocytosis and chemotaxis of P . aeruginosa [15] . This suggests that other phagocytic receptors on leukocytes are more important for the efficient eradication of B . pseudomallei [38 , 58 , 59] . Murine models like the one used here , which make use of relatively young mice exposed to an intranasal bacterial inoculum , do show inter-experiment variation , as reflected by differences in bacterial dissemination and as a result inflammation at the latter time-points before mice will succumb to infection . In addition , caution is needed when extrapolating data from murine experiments to human disease . ” Taking these precautions into mind , we here demonstrate that murine melioidosis is associated with increased TREM-1 and -2 expression . TREM-2 deficiency is beneficial during experimental Gram-negative sepsis caused by a clinical relevant pathogen , resulting in lower bacterial loads , reduced organ damage , decreased inflammation and improved survival . When compared to TREM-2 , TREM-1 plays a limited detrimental role during experimental melioidosis . These results provide new information on the expression and function of TREM-2 during melioidosis and may demonstrate its potential therapeutic usefulness . | Triggering receptor expressed on myeloid cells ( TREM ) -1 and -2 are receptors on immune cells that act as mediators of the innate immune response . It is thought that TREM-1 amplifies the immune response , while TREM-2 acts as a negative regulator . Previously , we found that TREM-1 is upregulated in melioidosis patients . In contrast , nothing is known on TREM-2 expression and its role in melioidosis . In this study we examined the expression and functional role of both TREM-1 and -2 in a murine melioidosis model . We found that TREM-1 and-2 expression was upregulated during melioidosis . Using our experimental melioidosis model , we observed that Trem-2-/- mice were protected against B . pseudomallei-induced lethality . Trem-2-/- mice demonstrated reduced bacterial loads , inflammation and organ damage compared to wild-type mice in experimental melioidosis . Despite reduced bacterial dissemination of B . pseudomallei to distant organs in Trem-1/3-/ mice- , no differences in survival were found between Trem-1/3-/- and wild-type mice during melioidosis . Lastly , we investigated cellular functions of TREM-1 and TREM-2 and found that TREM-2 deficiency led to decreased cellular responsiveness to B . pseudomallei infection . In conclusion , we found that TREM-2 plays an important role during experimental murine melioidosis . TREM-2-deficiency reduces inflammation and organ damage , thereby improving survival . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"melioidosis",
"immunology",
"cell",
"processes",
"animal",
"models",
"bacterial",
"diseases",
"sepsis",
"model",
"organisms",
"signs",
"and",
"symptoms",
"neutrophils",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"white",
"blood",
"cells",
"inflammation",
"animal",
"cells",
"alveolar",
"macrophages",
"mouse",
"models",
"phagocytosis",
"hematology",
"immune",
"response",
"diagnostic",
"medicine",
"blood",
"anatomy",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types"
]
| 2016 | Triggering Receptor Expressed on Myeloid Cells (TREM)-2 Impairs Host Defense in Experimental Melioidosis |
In the human neocortex , single excitatory pyramidal cells can elicit very large glutamatergic EPSPs ( VLEs ) in inhibitory GABAergic interneurons capable of triggering their firing with short ( 3–5 ms ) delay . Similar strong excitatory connections between two individual neurons have not been found in nonhuman cortices , suggesting that these synapses are specific to human interneurons . The VLEs are crucial for generating neocortical complex events , observed as single pyramidal cell spike-evoked discharge of cell assemblies in the frontal and temporal cortices . However , long-term plasticity of the VLE connections and how the plasticity modulates neocortical complex events has not been studied . Using triple and dual whole-cell recordings from synaptically connected human neocortical layers 2–3 neurons , we show that VLEs in fast-spiking GABAergic interneurons exhibit robust activity-induced long-term depression ( LTD ) . The LTD by single pyramidal cell 40 Hz spike bursts is specific to connections with VLEs , requires group I metabotropic glutamate receptors , and has a presynaptic mechanism . The LTD of VLE connections alters suprathreshold activation of interneurons in the complex events suppressing the discharge of fast-spiking GABAergic cells . The VLEs triggering the complex events may contribute to cognitive processes in the human neocortex , and their long-term plasticity can alter the discharging cortical cell assemblies by learning .
Evolution has shaped the human neocortex producing microcircuit features that are specific to our species [1] . Neuronal density , ultrastructural features , and functional properties of neurons [2–4] reflect specific adaptations in the human neocortex to perform complex and fast signal processing [5–13] . A remarkable feature in the human neocortex is that single pyramidal cell ( PC ) action potentials ( APs ) are able to generate di- and polysynaptic GABAergic interneuron discharge known as complex events [10 , 11] . The events emerge from the activity of a small subset of excitatory connections forming very large glutamatergic excitatory postsynaptic potentials ( VLEs ) , specifically to GABAergic interneurons in supragranular layers of the frontal , the temporal , and the prefrontal cortices [10 , 11] . Similar strong excitatory connections between individual neocortical neurons have not been found in nonhuman brains . Therefore , it has been proposed that the VLEs and the complex events participate in cortical information encoding in high order cognitive processes [10] . However , this would predict that these events are dynamically modulated by learning [14 , 15] . Yet , it is unknown whether the VLEs show use-dependent long-term plasticity , and if their specific modulation indeed affects the complex events . We hypothesize that the immense strength of VLEs is generated and regulated by common activity-driven synaptic long-term plasticity processes , and that they may occur in various different inhibitory interneuron types [16 , 17] . Alternatively , these connections could be hard-wired selectively in a specific , yet unknown , subset of postsynaptic GABAergic interneurons without prominent lasting plasticity in the adult neocortex [18 , 19] . We asked whether VLEs show activity-induced long-term plasticity , and if their selective modulation had impact on the local complex events . By performing triple and dual whole-cell recordings of synaptically connected identified neurons , we found that VLEs exhibit metabotropic glutamate receptor ( mGluR ) -dependent long-term depression ( LTD ) that converts them to common weak excitatory postsynaptic potential ( EPSP ) connections . In addition , this alters the neocortical complex events suppressing the cell assemblies activated by the PC . Thus , the VLEs occur in various interneuron types , and their occurrence is regulated by the synapse’s activity history . To the best of our knowledge , this is the first study reporting synaptic plasticity in human neocortical interneurons .
By performing triple and dual whole-cell recordings from identified L2–3 human neocortical neurons , we found that some fast-spiking GABAergic interneurons ( FSINs ) receive glutamatergic input from individual afferent PCs showing VLEs ( Fig 1A and 1B ) . Simultaneous recording from three neurons demonstrated that a fast-spiking GABAergic cell can receive VLEs ( average amplitude 9 . 60 ± 0 . 20 mV , showing no failures ) from one L2–3 PC and small amplitude glutamatergic EPSPs ( average 3 . 29 ± 0 . 12 mV , showing no failures ) similar to EPSPs between PCs from another PC [10] . Recordings from 21 synaptically connected PC–FSIN pairs revealed vast differences in the single AP-evoked EPSP amplitudes between the pairs ( Fig 1B , S1 Table , S1 Data ) . In FSINs , the EPSP averages showed a range from 0 . 62 mV to 16 . 49 mV , with nonparametric distribution ( failures excluded , evoked with 10 s interval at Em −68 . 5 ± 1 . 4 mV , n = 21 , Shapiro-Wilk test ) . Despite the amplitude difference , the excitatory postsynaptic currents ( EPSCs ) in FSINs similarly exhibited fast time-to-peak kinetics ( 0 . 59 ± 0 . 04 ms , n = 18 ) ( Fig 1B , S1 Data ) , indicating that the amplitude variability is unlikely to derive from different electrotonic filtering of the glutamatergic synaptic inputs . Likewise , distribution of the average EPSP amplitudes in PCs to non–fast-spiking interneuron ( non-FSIN ) pairs showed nonparametric distribution with a range from 0 . 7 mV to 6 . 9 mV ( failures excluded , at Em −70 . 3 ± 1 . 5 mV , n = 9 , Shapiro-Wilk test ) ( S1 Table ) . Thus , VLEs are not occurring solely in FSINs , but are exhibited in various types of GABAergic neurons including fast- and non–fast-spiking cells . On the contrary , PC–PC connections showed parametric distribution of average EPSPs with small amplitude ( 2 . 01 ± 0 . 02 mV at Em −69 . 4 ± 1 . 8 mV , failures excluded , Shapiro-Wilk test , n = 16 ) ( Fig 1B , S1 Table , S1 Data ) . The interneuron EPSPs were defined as VLEs when their average ( failures excluded ) was larger than mean + 2 x standard deviation ( SD ) of the EPSPs in PC–PC connections ( 4 . 21 mV , failures excluded ) in baseline conditions ( mean ± SD = 2 . 01 ± 1 . 10 mV , n = 480 in 16 cells ) . The postsynaptic interneurons in the triple and paired recordings were immunohistochemically confirmed positive for vesicular GABA transporter ( vgat+ ) ( n = 31 ) . Cells that in addition were immunopositive for parvalbumin ( pv+ ) showed rapid axon currents ( spike inward current width [SW] of 0 . 43 ± 0 . 02 ms , n = 11 ) characteristic of the FSINs [10 , 11] . The pv+ cells , together with vgat+ interneurons showing similar fast spike kinetics ( SW 0 . 49 ± 0 . 02 ms , n = 11 ) , but with nonconclusive or untested pv reaction , were considered FSINs ( n = 22 ) . Ten FSINs were further identified as putative basket cells by their axon morphology [10] . The non–fast-spiking vgat+ interneurons and the PCs had significantly longer spike kinetics with SW of 0 . 96 ± 0 . 04 ms ( n = 9 ) and 1 . 18 ± 0 . 06 ms ( n = 16 ) , respectively ( p < 0 . 01 between all groups , ANOVA with Tukey’s posthoc test ) [13] . The non-FSINs with intact soma were immunohistochemically tested for somatostatin ( sst ) for further identification of the cells [20] . Detailed results on the EPSPs excluding and including failures , the EPSCs , and the immunohistochemical reaction analyses with cell type identification are shown in S1 Table . In all potential connections tested between neurons ( n = 1 , 056 ) , we found ( including connections lost during baseline ) a monosynaptic response in 11 . 0% of cases . Success rate for identified PC–FSIN pairs was 4 . 0% , PC–non-FSINs connections: 1 . 9% , PC–PC pairs: 1 . 1% , and FSIN–PC connections: 3 . 8% , showing similar or slightly lower connectivity rates than reported in the rodent neocortex L2–3 [21–23] . We asked whether glutamatergic connections to interneurons showed long-term plasticity akin to that reported in the rodent cortex [24 , 25] . To test this , we performed experiments applying high frequency bursts of APs ( 5 APs at 40 Hz , x 40 with 0 . 5 s interval ) in the presynaptic PC after a baseline of EPSPs ( at least 5 min , but less than 10 min , analyzed including failures ) [26 , 27] . Postsynaptic FSINs ( SW 0 . 40 ± 0 . 02 ms ) were held in resting membrane potential in current clamp ( –67 . 6 ± 2 . 2 mV , n = 9 ) ( Fig 2 ) . First , we tested VLEs ( average in baseline 5 . 85 ± 0 . 59 mV , did not show failures , n = 5 ) in control conditions and found that the afferent PC bursts firing generated an LTD of the EPSPs ( amplitude to 0 . 52 ± 0 . 02 of baseline at 20–25 min after 40 Hz bursts , n = 5 cells , p < 0 . 01 , Wilcoxon test ) ( Fig 2A , S1 Fig , S2 Data , S7 Data ) . The LTD was associated with a reduced paired-pulse EPSP ratio ( 1st/2nd EPSP amplitude with 50 ms interval ) to 0 . 74 ± 0 . 06 of baseline ( p < 0 . 05 , Mann-Whitney test , baseline mean 1 . 41 ± 0 . 22 ) and a decrease in the EPSP amplitude CV−2 ( 1/squared coefficient of variation ) value ( to 0 . 44 ± 0 . 12% from baseline , p < 0 . 05 , Mann-Whitney test ) ( S3 Data ) , indicating presynaptic site of depression [28] . The results on the paired-pulse ratio ( PPR ) and altered EPSP amplitude variation by LTD are summarized in histograms in Fig 2B . Likewise , corresponding experiments in occasional PC–non-FSIN pairs with VLEs ( n = 2 , averages in the baseline including failures 5 . 81 mV and 6 . 89 mV , failure rates 0% and 13% , respectively ) showed that single fiber burst firing can also generate LTD in some non-FSINs ( S2 Fig ) . We next tested whether LTD in VLEs requires group I mGluRs as various forms of long-term plasticity in glutamatergic synapses to FSINs depend on these receptors in the rodent cortex [24 , 29–32] . We studied four PC–FSIN pairs with VLEs ( average in baseline 8 . 42 ± 2 . 83 mV at Em −69 . 5 ± 2 . 1 mV , did not show failures , n = 4 ) as above , but in the presence of LY367385 ( 100 μM ) and 2-Methyl-6- ( phenylethynyl ) pyridine hydrochloride ( MPEP , 25 μM ) . Thus , LTD was blocked in the VLE connections ( Fig 2A ) with no significant change in the amplitude ( 1 . 03 ± 0 . 04 of baseline at 20–25 min , n = 4 cells , Wilcoxon test ) , PPR ( 1 . 21 ± 0 . 11 of baseline , baseline mean 1 . 07 ± 0 . 17 ) or 1/CV2 ( 1 . 63 ± 0 . 39 of baseline ) ( Mann-Whitney test ) ( Fig 2B , S2 and S3 Data ) . To conclude , PC–FSIN connections with large EPSPs show activity-driven LTD , which requires group I mGluRs . In contrast to VLEs , the PC–FSIN pairs with small EPSPs ( average in baseline with failures 1 . 89 ± 0 . 43 mV at Em −69 . 2 ± 3 . 5 mV , failure rate 11 . 2 ± 9 . 5% , SW 0 . 48 ± 0 . 04 ms , n = 5 ) failed to show lasting plasticity following the 40 Hz bursts ( amplitude 1 . 07 ± 0 . 06 of baseline at 20–25 min ) ( Fig 3A , S1 Fig , S4 Data ) . Likewise , no lasting plasticity was seen in any ( paired t-test ) of the three vgat+ non-FSINs ( SW 0 . 91 ± 0 . 07 , n = 3 ) with small amplitude EPSP ( 2 . 08 ± 0 . 58 mV , n = 3 ) in similar experiments . Given that LTD in the VLEs was accompanied by stronger postsynaptic depolarization during the presynaptic spike bursts , we studied whether small amplitude EPSP connections would show plasticity if the postsynaptic FSIN was depolarized during the PC bursts . We reproduced experiments above with a separate set of PC–FSIN ( SW 0 . 47 ± 0 . 03 ms , n = 5 ) pairs with small amplitude EPSP ( average with failures 1 . 44 ± 0 . 22 mV , failure rate 11 . 0 ± 3 . 3% , n = 5 ) , and paired presynaptic PC spike bursts with postsynaptic cell depolarization ( 20–30 mV , 250 ms steps from Em , see Methods for details ) beyond the firing threshold ( Fig 3A ) . This protocol also failed to generate long-lasting change in EPSPs in the PC–FSIN pairs ( 0 . 94 ± 0 . 04 of baseline at 20–25 min , n = 5 , Wilcoxon test ) ( Fig 3A , S4 Data ) . Interestingly , a small but significant LTD was observed with this configuration in two ( 0 . 71 ± 0 . 10 and 0 . 83 ± 0 . 05 at 20–25 min compared to baseline , p < 0 . 05 for both cells , paired t-test ) of three individual PC–non-FSIN ( vgat+ ) pairs tested . Finally , we studied the synaptic connections between L2–3 PCs applying presynaptic 40 Hz bursts while the postsynaptic cell was at resting membrane potential . The PC–PC pairs were connected with small amplitude EPSPs ( average with failures 1 . 40 ± 0 . 30 mV at Em −65 . 9 ± 5 . 4 mV , failure rate 4 . 2 ± 3 . 2% , n = 4 ) , and the 40 Hz bursts failed to generate lasting plasticity in the EPSP ( 1 . 01 ± 0 . 05 of baseline at 20–25 min , n = 4 cells , Wilcoxon test ) ( Fig 3B , S4 Data ) , possibly because LTP and LTD in human PCs require strong postsynaptic depolarization for either activation of glutamate NMDA receptors or L-type voltage-gated calcium channels [7] . Input resistance in the plasticity recordings showed small increase to 1 . 09 ± 0 . 01 ( baseline-normalized ) at 20–25 min from baseline ( n = 26 , p < 0 . 01 , t-test ) [33] . The results show that following just a single PC burst firing , LTD specifically occurs in large EPSPs between PCs and interneurons , and not in other investigated synaptic connections . Because studies in rodents have reported LTD in glutamatergic synapses to cortical interneurons either by chemical or strong synaptic activation of group I mGluRs , we studied whether FSINs with weak excitatory inputs showed the LTD when multiple glutamatergic fibers were simultaneously activated ( Fig 3C ) [27 , 29 , 34] . Evoking compound EPSCs from many small glutamatergic inputs with extracellular electrical stimulation ( see Methods ) , we applied 40 Hz bursts to the glutamatergic pathway as above after baseline ( at least 5 min , but less than 10 min ) . Focusing the EPSC analysis in the FSINs on the monosynaptic component of the current ( see Fig 3C ) [35 , 36] we found that the 40 Hz burst stimulation resulted in LTD ( EPSC 0 . 72 ± 0 . 02 from baseline at 20 min , n = 7 cells , p < 0 . 001 , t-test ) . Blockers for glutamate N-methyl-D-aspartatereceptor ( NMDARs ) ( DL-2-Amino-5-phosphonopentanoic acid [DL-APV] , 100 μM ) and GABAARs ( PiTX , 100 μM ) were present in the experiments . The EPSC amplitude LTD was accompanied by decreased CV−2 ( baseline-normalized to 0 . 72 ± 0 . 16% , n = 7 , t-test ) . This LTD was blocked in experiments with group I mGluR antagonists LY367385 ( 100 μM ) and MPEP ( 25 μM ) ( n = 7 , p < 0 . 05 , t-test ) ( Fig 3C , S5 Data ) . The FSINs in these extracellular stimulation experiments showed narrow SW 0 . 62 ± 0 . 04 ms , n = 14 . Accordingly , we reproduced these experiments with rat glutamatergic fibers to FSINs in L2–3 ( SW 0 . 64 ± 0 . 06 ms , n = 10 ) and confirmed LTD ( EPSC 0 . 64 ± 0 . 04 from baseline at >15 min , n = 5 , p < 0 . 01 , Wilcoxon test ) and its blockade with the group I mGluR antagonists ( EPSC from baseline 0 . 95 ± 0 . 5 , n = 5 , Wilcoxon test ) ( Fig 3C , S5 Data ) . The EPSC amplitude in LTD showed reduced CV−2 ( baseline-normalized to 0 . 41 ± 0 . 09% , n = 5 , Mann-Whitney test ) , but not when LTD was blocked with the mGluR antagonists ( 1 . 04 ± 0 . 24% of baseline at >15 min ) ( p < 0 . 05 between the groups at >15 min , Mann-Whitney test ) ( Fig 3C , S5 Data ) . We tested three of the recorded rat FSINs for pv immunoreaction and found them all positive . Thus , in both human and rat cortex , weak glutamatergic connections to L2–3 FSINs exhibit group I mGluR-dependent LTD if multiple glutamatergic inputs are activated simultaneously . Given that interneuron–PC connections with VLEs have been proposed to be essential in generation of the neocortical complex events , we studied whether the LTD in these connections would selectively modify network activity . First , we confirmed that single PC AP-evoked VLEs in the FSIN as well as in the non-FSIN elicited firing of these interneurons from the resting membrane potential [11] . We found that VLE-evoked postsynaptic spikes in a FSIN ( Em −69 mV ) typically followed with a short 3–5 ms delay ( Fig 4A ) [10] , whereas in a non-FSIN , ( Em −69 mV ) the spikes showed long delay with large jitter ( S2 Fig ) . Similarly , whole-cell recordings between identified PCs revealed disynaptic GABAAR-mediated inhibitory currents ( dIPSCs ) in complex events elicited by a single AP ( interval 10 s ) ( Fig 4B ) [10 , 11] . The dIPSCs occurred with short delay ( 6 . 23 ± 0 . 72 ms , n = 16 pairs ) and high probability ( 0 . 70 ± 0 . 05 , n = 16 pairs in baseline conditions ) ( S6 Data ) . The dIPSCs showed longer and more variable delay to the presynaptic spike than monosynaptic GABAAR-mediated inhibitory currents ( monIPSCs ) from FSINs ( 0 . 96 ± 0 . 10 ms , n = 9 , p < 0 . 001 , t-test ) ( SW 0 . 48 ± 0 . 03 ms , n = 9 ) ( S3 Fig , S9 Data ) . In addition , dIPSCs were blocked by the glutamate AMPAR blocker GYKI53655 ( 25 μM ) ( n = 3 , p < 0 . 001 , Chi-square test ) ( Fig 4C , S4 Fig , S6 Data ) . The evoked dIPSC amplitudes ( averages excluding failures in all plasticity recordings 32 . 1 ± 3 . 7 pA , n = 12 ) were similar to monIPSCs ( 35 . 2 ± 5 . 6 pA , n = 9 , t-test ) ( S5 Fig , S10 Data ) , indicating that these early complex event inhibitory currents ( IPSCs ) were generated by a single FSIN . The dIPSCs and the monIPSCs were recorded at −55 mV . The dIPSCs were detected in 3 . 0% of all potential connections tested ( n = 1 , 056 ) . To determine whether plasticity modified this network , we performed long recordings from PC pairs showing that the probability and delay of the dIPSCs were stable for at least 30 min ( n = 3 ) in normal conditions ( Fig 4D , 4E and S6 Data ) . However , if the 40 Hz burst firing was delivered in the presynaptic PC ( similar to the VLE LTD experiments in Fig 2 ) , the dIPSC occurrence ( total in 25 ms from PC spike ) rapidly and permanently attenuated after a baseline showing strong LTD ( from 0 . 71 ± 0 . 04 in baseline to 0 . 14 ± 0 . 09 at 15–20 min , n = 3 , p < 0 . 05 , Chi-square test ) ( Fig 4F , 4G and S6 Data ) . Interleaved experiments in the presence of group I mGluR antagonists LY367385 ( 100 μM ) and MPEP ( 25 μM ) showed that LTD of dIPSCs was blocked in the presence of the group I mGluR antagonists ( 0 . 81 ± 0 . 10 in baseline and 0 . 88 ± 0 . 12 at 15–20 min , n = 3 , Chi-square test ) ( Fig 4H , 4I and S6 Data ) . In conclusion , the results demonstrate that a single PC burst firing at 40 Hz elicits robust LTD in VLE connections to FSINs and causes suppression of phase-locked early dIPSCs between PCs . These two LTDs both require group I mGluRs . Thus , the results show that plasticity of VLEs in FSINs changes the activation pattern of the neurons discharging in supragranular layers during complex events .
Almost a decade ago , Molnar et al . [10] first reported that a subset of excitatory PC synapses in the human neocortex form VLEs in local GABAergic interneurons in supragranular layers . These strong connections specifically from PCs to inhibitory interneurons have since been reported in the frontal , prefrontal , parietal , and temporal cortices , where the VLEs often are suprathreshold , driving assemblies of inhibitory interneurons to fire after a single PC spike [11 , 37] . Similar strong connections between two individual neocortical neurons have not been found in nonhuman species [37–39] . Analyses of large datasets from rodent visual and somatosensory cortices have revealed that neurons in local networks are not randomly connected , but specific local connectivity patterns exist between neuron types , and the strongest excitatory synapses control local network activity [40–42] . This suggests that there is a skeleton of strong connections in the network that dominates the activity [40] . Therefore , it has been proposed that the VLE connections in the human might be important in generating neocortical cell assemblies and be involved in higher cognitive functions [11] . However , until now it had remained unknown how neuronal plasticity regulates these connections and whether their selective modulation indeed alters discharge of neuronal assemblies in the human neocortical network activity . Our result that the VLEs occur specifically in human interneurons , and not in between PCs , is consistent with previous studies [10 , 11] . In addition , we demonstrate that VLEs are generated in different interneuron subpopulations , including fast-spiking ( such as basket cells ) and non–fast-spiking supragranular vgat+ neurons . Furthermore , single GABAergic interneurons receive both VLEs and more common small EPSP connections from layer 2–3 PCs . Indeed , anatomically identified L2–3 basket cells show huge variability in strength of PC inputs . It is likely , although we do not directly demonstrate it here , that a single presynaptic L2–3 PC cell evokes VLEs and small EPSPs in different L2–3 postsynaptic interneurons showing synapse specificity . A comprehensive recent study in rodent demonstrated that certain connectivity patterns between neurons are repeated in the neocortex across different regions [39] . The strong VLE connection , occurring between PCs and L2–3 inhibitory interneurons in various neocortical regions is one such feature , and probably specific to the human neocortical microcircuits . We show that strength of VLE connections is controlled by their activity history . A common form of mGluR-dependent LTD characterized in the extracellular stimulation experiments of glutamatergic synapses in the rodent cortex [24] converts the VLEs to small amplitude EPSPs . This LTD suppresses VLE synapses through a presynaptic mechanism most likely controlling the vesicular transmitter release as indicated by the changes in the EPSP amplitude PPR and the coefficient of variation in the depressed synapses [37] . The LTD of VLEs in FSINs , many of which are also pv+ , involves the group I mGluRs . In the rodent cortex , group I mGluRs have been shown to play a central role in long-term plasticity processes including presynaptic LTD and LTP [24 , 29–31 , 34 , 43–46] . Our results from the group I mGluR-dependent LTD in FSINs in the human and the rodent neocortex indicate that this is an evolutionarily conserved plasticity mechanism for controlling the fast-spiking interneuron activity in the mammalian brain . Interestingly , we report that in a non–fast-spiking cell , LTD was not blocked by the group I mGluR antagonists ( see S2 Fig ) . In addition , we found significant LTD in two of three PC–non-FSIN connections with small EPSP when the postsynaptic cell was depolarized during PC bursts . These suggest that there may be various LTD forms in human cortical interneuron synapses [24 , 25] , and that non-FSINs may exhibit different LTD mechanism than the FSINs , possibly depending on postsynaptic depolarization . Indeed , many synapse-specific properties including long-term plasticity have been reported in glutamatergic fibers in the rodent cortex [21 , 32 , 47–50] . Correspondingly , synapses originating from the same PC in the human neocortex may exhibit distinct long-term plasticity , depending upon the postsynaptic target cell; and different activity patterns may be required for plasticity in the synapses [24 , 50] . In this study , we have used elevated extracellular calcium ( 3 mM ) to increase the stability of disynaptic IPSCs in baseline conditions . However , compared to recordings with 2 mM extracellular calcium , this modification is unlikely to strongly affect the high-frequency firing-evoked synaptic release and the long-term plasticity in FSIN synapses: the probability of synaptic release in VLEs in FSINs is already very high at 2 mM Ca2+ , and only slightly modulated by further increase of calcium [37] . Yet , as demonstrated in the recent study by Molnar et al . [37] , the VLE release probability can markedly decrease when extracellular calcium is reduced from 3 mM to 1 . 5 mM , which is considered lower range of the cerebrospinal fluid total calcium level in physiological conditions [51] . Therefore , it is also possible that in calcium concentrations close to 1 . 5 mM , the PC firing pattern used in this study may not be sufficient for such a robust LTD as reported here in the FSINs . Importantly , some non-FSINs show VLEs with low presynaptic release probability even in 3 mM calcium , as indicated by the large EPSP amplitude coefficient of variation . In these synapses , extracellular calcium modulations may have even stronger effects on the short- and long-term plasticities than in the FSINs . Strikingly , in the human neocortex , the activity of a single PC is sufficient to trigger mGluR-dependent LTD in the VLE connections , but not in weak glutamatergic pathways . The level of the postsynaptic depolarization does not explain the failure of LTD in small EPSP connections to fast-spiking cells , and a potential explanation is that there is insufficient activation of the postsynaptic group I mGluRs [34 , 43] . The connections with VLEs are likely to release more glutamate and activate the critical mGluRs [52–54] . The hypothesis on strong glutamate release is supported by our finding that the small EPSP connections were unable to generate LTD by single fiber activity , but they showed the mGluR-dependent plasticity when multiple fibers were activated simultaneously with local extracellular stimulation [55] . Indeed , a recent study revealed that human neocortical PC–FSIN synapses with VLEs have more transmitter vesicle release sites , although the glutamate release quantal size is similar compared to synapses in rat neocortex [37] . This indicates multivesicular release in synapses with VLEs , and it is interesting to speculate that the conversion to common small EPSPs via the presynaptic LTD might reflect their transformation from a multivesicular release site to a single vesicle-releasing synapse . Although a link between very large excitatory synapses and human cortical complex events has been suggested earlier [10 , 11] , a relation between their selective modulation and complex events had not been directly demonstrated until this study . The LTD triggered by a single PC firing in the current conditions is specific to connections with VLEs and therefore provides a useful tool to test the relation between VLEs and neocortical network activity . Results here show that the VLE connections indeed trigger the complex events , and the LTD changes their temporal structure . The activation of individual fast-spiking interneurons was commonly observed in dual PC recording as disynaptic GABAergic currents with timing corresponding to the APs in fast spiking interneurons [11] . LTD of the VLE in FSINs and the suppression of dIPSCs in complex events were both induced by single PC burst firing , and they both required group I mGluRs . Some non-FSINs also exhibit VLEs and can show LTD , although these cells were unlikely to contribute to the GABAergic disynaptic currents investigated here: the dIPSCs occurred with short delay and high precision , whereas the non-FSINs show long and variable delay in their response to fire APs ( see S2 Fig ) . However , the results of non-FSINs are based on small sample sizes , and should therefore be interpreted with caution . In conclusion , the human neocortex is unique in many aspects , since its microcircuits show differences at the molecular , ultrastructural , and physiological levels compared to other mammalian species [1 , 3 , 56] . The capacity of the human neocortex to perform extraordinary and highly complex tasks may at least partly result from these microcircuit level specializations . We propose that VLEs with robust activity-induced plasticity and their contribution to neocortical cell assemblies may be crucial for higher cognitive functions and abstract mental abilities of the human brain . In addition , evidence in animal models suggests the involvement of group I mGluR-mediated plasticity in neocortical learning processes , and perturbation of the mGluR-dependent plasticity has been reported with mental decline [43] . Therefore , the human-specific microcircuit features may also be substrates for pathological processes resulting in cognitive decline and other neurological and neuropsychiatric dysfunctions that we as a species are vulnerable to [57–61] .
All procedures were performed according to the Declaration of Helsinki with the approval of the University of Szeged Ethical Committee and Regional Human Investigation Review Board ( ref . 75/2014 ) . Human neocortical slices were derived from material that had to be removed to gain access to the surgical treatment of deep-brain tumors from the left and right frontal , temporal , and parietal regions with written informed consent of the patients prior to surgery . The patients were 10–85 y of age ( mean ± SD = 50 ± 4 y ) , including 17 males and 14 females . The tissue obtained from underage patients was provided with agreement from a parent or guardian . The resected samples were cut from the frontal and temporal lobes of left or right hemisphere . Anesthesia was induced with intravenous midazolam and fentanyl ( 0 . 03 mg/kg , 1–2 lg/kg , respectively ) . A bolus dose of propofol ( 1–2 mg/kg ) was administered intravenously . The patients received 0 . 5 mg/kg rocuronium to facilitate endotracheal intubation . After 2 min , the trachea was intubated and the patient was ventilated with O2/N2O mixture ( a ratio of 1:2 ) . Anesthesia was maintained with sevoflurane at monitored anesthesia care volume of 1 . 2–1 . 5 . After surgical removal , the resected tissue blocks were immediately immersed in ice-cold standard solution containing ( in mM ) : 130 NaCl , 3 . 5 KCl , 1 NaH2PO4 , 24 NaHCO3 , 1 CaCl2 , 3 MgSO4 , 10 D ( + ) -glucose , and saturated with 95% O2 and 5% CO2 . Slices were cut perpendicular to cortical layers at a thickness of 350 μm with a microtome ( Microm HM 650 V ) and were incubated at room temperature ( 20–24°C ) for 1 h in the same solution . Rat neocortical slices were prepared as described before [62] . Male Wistar rats were anaesthetized using halothane , and following decapitation ( 320 μm thick ) , coronal slices were prepared from the somatosensory cortex . The solution used during electrophysiology experiments was identical to the slicing solution , except it contained 3 mM CaCl2 and 1 . 5 mM MgSO4 . Recordings were performed in a submerged chamber ( perfused 8 ml/min ) at approximately 36–37°C . Cells were patched using water-immersion 20× objective with additional zoom ( up to 4x ) and infrared differential interference contrast video microscopy . Micropipettes ( 5–8 MOhm ) were filled with intracellular solution for whole-cell patch-clamp recording ( in mM ) : 126 K-gluconate , 8 KCl , 4 ATP-Mg , 0 . 3 Na2–GTP , 10 HEPES , 10 phosphocreatine ( pH 7 . 20; 300 mOsm ) with 0 . 3% ( w/v ) biocytin . Current and voltage clamp recordings were performed with Mutliclamp 2B amplifier ( Axon Instruments ) , low-pass filtered at 6 kHz ( Bessel filter ) . Series resistance ( Rs ) and pipette capacitance were compensated in current clamp mode and pipette capacitance in voltage clamp mode . Cell capacitance compensation was not applied . Rs was monitored and recorded continuously during the experiments . The recording in voltage clamp mode was discarded if the Rs was higher than 25 ΩM or changed more than 20% . In paired cell recordings , APs were generated in the presynaptic cell with brief ( 2–3 ms ) suprathreshold depolarizing ( 60–70 mV ) paired pulses ( 50 ms interval ) in voltage clamp delivered every 10 s from −60 mV . Postsynaptic cells were at resting membrane potential in current clamp mode . In some cells with VLEs , the postsynaptic cell was hyperpolarized ( up to −10 mV ) with constant current to prevent the VLE from triggering an AP . The 40 Hz firing protocol was similarly applied in voltage clamp mode with series of 2–3 ms depolarizing pulses ( 5 pulses at 40 Hz , delivered every 0 . 5 sec 40 times ) , while the postsynaptic cell was held in current clamp resting membrane potential . In some experiments ( Fig 3A , 4–5 ) , the postsynaptic cell was depolarized in voltage clamp during presynaptic 40 Hz firing with a continuous step ( 20–30 mV , 250 ms ) . This elicited on average 2 . 2 postsynaptic spikes for 1st presynaptic spike ( in following 25 ms , n = 200 in 5 cells ) of the 40 Hz train , and on average 0 . 80 postsynaptic spike probability for the 2nd–5th PC AP . Extracellular stimulation was applied with a concentric bipolar electrode ( 125 μm tip diameter , FHC Inc . , US ) positioned on L2–3 . Paired pulse stimuli ( 50 μs , with 50 ms interval , intensity range from 20 to 300 μA ) were delivered every 15 s with current isolator stimulator ( Model DS3 , Digitimer , UK ) . Compound EPSCs in Fig 3 were confirmed by observing less than 100 pA increases in the evoked EPSC amplitude when gradually increasing stimulation intensity . Data were acquired with Clampex software ( Axon Instruments , US ) at 20 kHz . EPSC/P , IPSC , action current duration , and the cell input resistance were analyzed off-line with p-Clamp software ( Axon Instruments , US ) and Spike2 ( version 7 . 0 , Cambridge Electronic Design , UK ) . Liquid junction potential was not corrected . EPSC amplitude and kinetics analysis in voltage clamp mode ( time-to-peak from onset ) was omitted when access resistance was higher than 25 MΩ . SW was calculated from the onset of inward action current till recovery to baseline holding level . Data for SWs were collected in the beginning of experiments when synaptic connections from all cells were briefly tested in voltage clamp mode . All data are presented as mean ± s . e . m . and when showing baseline-normalized EPSPs of many cells , the values were calculated from binned ( 30 s bin ) data . In rare cases when cell-spiked and accurate EPSP amplitude data was not available , bin includes two instead of three data points . For statistical analysis , ANOVA with posthoc Tukey’s test and t-test were used for data with normal distribution ( Shapiro-Wilk test ) and sample sizes larger than n = 6 . Chi-square test was used for categorical variables ( occurrence of dIPSC in 25 ms time window from PC spike ) . Otherwise , Mann-Whitney U-test ( unpaired ) and Wilcoxon Signed Rank Test ( paired ) were used . EPSP amplitude in individual plasticity experiments was tested with paired t-test comparing data points in 5 min baseline and an equal time window at 20–25 min following the presynaptic bursts , unless stated otherwise ( in some shorter experiments in the last 5 min of the recording ) . Correlation was determined with Pearson’s r-test . Differences were accepted as significant if p < 0 . 05 . Failures were included in the EPSP mean values ( in binned data ) in plasticity analysis . DL-APV , GYKI53655 , LY367385 , MPEP , and picrotoxin ( PiTX ) were applied via bath and purchased from Sigma Aldrich ( Hungary ) . After electrophysiological recording , slices were immediately fixed in a fixative containing 4% paraformaldehyde and 15% picric acid in 0 . 1 M phosphate buffer ( PB; pH = 7 . 4 ) at 4°C for at least 12 h , then stored in 0 . 1 M PB with 0 . 05% sodium azide as a preservative at 4°C . Slices were embedded in 10% gelatin and further sectioned into slices of 50 μm thickness in cold PB using a vibratome VT1000S ( Leica Microsystems , UK ) . After sectioning , the slices were rinsed in 0 . 1 M PB ( 3 x 10 min ) and cryoprotected: first step in 10% ( 30 min ) , and later in 20% sucrose ( 1 h ) dissolved in PB and then permeabilized using a freeze and thaw procedure . Finally , they were incubated in fluorophore ( Cy3 ) -conjugated streptavidin ( 1:400 , Jackson ImmunoResearch Lab . Inc . US ) in 0 . 1 M Tris-buffered saline ( TBS , pH 7 . 4 ) for 2 . 5 h ( at 22–24°C ) . After washing with 0 . 1 M PB ( 3 x 10 min ) , the sections were covered in Vectashield mounting medium ( Vector Laboratories Inc , US ) , put under cover slips , and examined under epifluorescence microscope ( Leica DM 5000 B , UK ) . Sections selected for immunohistochemistry and cell reconstruction were dismounted and processed as explained below . Some sections for cell structure illustrations were further incubated in a solution of conjugated avidin-biotin horseradish peroxidase ( ABC; 1:300; Vector Labs , UK ) in Tris-buffered saline ( TBS , pH = 7 . 4 ) at 4°C overnight . The enzyme reaction was revealed by the glucose oxidase-DAB-nickel method using 3’3-diaminobenzidine tetrahydrochloride ( 0 . 05% ) as chromogen and 0 . 01% H2O2 as oxidant . Sections were postfixed with 1% OsO4 in 0 . 1M PB . After several washes in distilled water , sections were stained in 1% uranyl acetate and dehydrated in ascending series of ethanol . Sections were infiltrated with epoxy resin ( Durcupan ) overnight and embedded on glass slices . Three-dimensional light microscopic reconstructions from sections were carried out using the Neurolucida system with 100 x objective ( Olympus BX51 , Olympus UPlanFI , Hungary ) . Images were collapsed in z-axis for illustration . Cells in Fig 1 were reconstructed from confocal microscope z-stack images of streptavidin fluorophore signal using Image-J software as described previously [32] . Vgat immunoreaction analysis was used in parallel to confirm the interneuron axon . For immunohistological reactions , free-floating sections were washed 3 times in TBS-TX 0 . 3% ( 15 min ) at 22–24°C , then moved in 20% blocking solution with horse serum in TBS-TX 0 . 3% . The sections were incubated in primary antibodies diluted in 1% serum in TBS-TX 0 . 3% over three nights at 4°C , then put in relevant fluorochrome-conjugated secondary antibodies in 1% of blocking serum in TBS-TX 0 . 3% overnight at 4°C . Sections were washed at first step in TBS-TX 0 . 3% ( 3 x 20 min ) and later in 0 . 1 M PB ( 3 x 20 min ) and mounted on glass slides with Vectashield mounting medium ( Vector Lab . Inc . , UK ) . The characterizations of antibodies: pv ( goat anti-pv , 1:500 , Swant , Switzerland , www . swant . com ) , sst ( rat anti-sst , 1:50 , Merck Millipore , Germany , www . merckmillipore . com ) and vgat ( rabbit anti-vgat , 1:500 , Synaptic Systems , Germany , www . sysy . com ) . Fluorophore-labelled secondary antibodies were: DyLight 488 ( Donkey anti goat , 1:400 , Jackson ImmunoResearch Lab . Inc . , www . jacksonimmuno . com , US ) , Alexa488 ( Donkey anti rat , 1:400 , Jackson ImmunoResearch Lab . Inc . ) and Cy5 ( Donkey anti rabbit , 1:500 , Jackson ImmunoResearch Lab . Inc . ) . Labelling of neurons by neurobiotin and immunoreactions were evaluated using first epifluorescence ( Leica DM 5000 B , UK ) and then laser scanning confocal microscopy ( Olympus FV1000 , Hungary ) . Immunoreaction was considered to be negative when fluorescence was not detected in relevant neurobiotin-labelled cell , but immunopositivity was detected in the same area in unlabelled cells . Immunoreactions were studied in axon boutons ( vgat and pv ) and soma and dendrites ( pv , sst ) . | Many microscale features in the human neocortex—a part of the brain involved in higher functions such as sensory perception , generation of motor commands , spatial reasoning , and language—are closely similar to those reported in experimental animals commonly used in neuroscience , like mice . However , the human neocortical neurons also exhibit specializations only reported in our species . One such feature is the capacity of excitatory principal cells to elicit firing in local inhibitory interneurons with a single action potential via very strong excitatory synapses . It has been suggested that this feature has specifically evolved to enhance coordinated firing of neuronal ensembles in higher brain functions . However , it is unknown how these circuits are modified by learning . Therefore , we investigated how these very strong excitatory synapses are changed , and if their impact on the firing of local inhibitory neurons is altered by repetitive action potentials mimicking learning-related activity . By recording in human neocortical slices , we report that the strong excitatory synapses on interneurons exhibit robust activity-dependent long-term plasticity . The plasticity also regulates the discharge of local interneurons driven by these synapses . Although these specialized synapses have only been reported in the human neocortex , their plasticity mechanism is evolutionarily conserved . We suggest that the strong synapses with robust plasticity have evolved to enhance complex brain functions and learning . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"nervous",
"system",
"membrane",
"potential",
"brain",
"electrophysiology",
"neuroscience",
"synaptic",
"plasticity",
"nerve",
"fibers",
"interneurons",
"excitatory",
"postsynaptic",
"potentials",
"developmental",
"neuroscience",
"animal",
"cells",
"axons",
"neocortex",
"cellular",
"neuroscience",
"cell",
"biology",
"anatomy",
"synapses",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"neurophysiology"
]
| 2016 | Plasticity in Single Axon Glutamatergic Connection to GABAergic Interneurons Regulates Complex Events in the Human Neocortex |
Red light promotes germination after activating phytochrome phyB , which destabilizes the germination repressor PIF1 . Early upon seed imbibition , canopy light , unfavorable for photosynthesis , represses germination by stabilizing PIF1 after inactivating phyB . Paradoxically , later upon imbibition , canopy light stimulates germination after activating phytochrome phyA . phyA-mediated germination is poorly understood and , intriguingly , is inefficient , compared to phyB-mediated germination , raising the question of its physiological significance . A genetic screen identified polyamine uptake transporter 2 ( put2 ) mutants that overaccumulate polyamines , a class of antioxidant polycations implicated in numerous cellular functions , which we found promote phyA-mediated germination . In WT seeds , our data suggest that canopy light represses polyamines accumulation through PIF1 while red light promotes polyamines accumulation . We show that canopy light also downregulates PIF1 levels , through phyA; however , PIF1 reaccumulates rapidly , which limits phyA-mediated germination . High polyamines levels in decaying seeds bypass PIF1 repression of germination and stimulate phyA-mediated germination , suggesting an adaptive mechanism promoting survival when viability is compromised .
Seeds are capsules maintaining the plant embryo in a resistant state and promoting plant dispersal . To successfully produce a seedling , seeds are able to remain viable over time and block their germination under potentially fatal conditions for the seedling . However , over time seeds irremediably accumulate oxidative events , which eventually will compromise their viability [1 , 2] . Seeds limit oxidative damage through antioxidants or physical barriers limiting oxygen access to the embryo [3–5] . Polyamines ( PAs ) are small polycations ubiquitously present in all living organisms where they regulate numerous fundamental cellular processes , including DNA replication , transcription , translation and post-translational modification , cell proliferation , cell cycle regulation and programmed cell death [6–8] . However , how PAs perform these functions is poorly understood . In response to oxidative stress PA levels rise and protect cells by scavenging reactive oxygen species ( ROS ) and by increasing antioxidant enzymes activity in leaves [9–13] . However , whether PAs accumulate in Arabidopsis seeds after oxidative stress is unknown . Arabidopsis seeds control their germination after perceiving abiotic parameters in their environment . This triggers signaling responses leading to opposite level changes of abscisic acid ( ABA ) and gibberellic acid ( GA ) , two hormones that repress and promote germination , respectively [2] . The quality of light perceived by the light receptor phytochromes phyB and phyA exerts a profound influence on seed germination . Phytochromes control the abundance of the transcription factor PHYTOCHROME-INTERACTING FACTOR 1 ( PIF1 ) , a key germination repressor regulating ABA and GA levels in seeds [14] . Red ( R ) light , favorable for photosynthesis , promotes germination while canopy light , enriched in far red ( FR ) light , represses germination . Early upon seed imbibition , a R pulse photoconverts phyB into its active PfrB form that interacts with PIF1 and promotes its destruction , thus promoting germination . In contrast , a pulse of FR light photoconverts phyB into its inactive PrB form , which leads to PIF1 stabilization , thus preventing germination [15] . Paradoxically , a second FR light pulse applied later on upon imbibition promotes germination by activating phyA [16–18] . Intriguingly , unlike R light , FR light promotes germination inefficiently and erratically among seed batches . This low efficiency was linked to limiting phyA levels and strong ABA-dependent responses early upon seed imbibition [16 , 17] . However , regulation of endogenous PIF1 levels by FR light through phyA has not been reported . phyA-mediated germination was interpreted as a last chance to form a seedling despite the presence of unfavorable canopy light [17 , 19] . Yet , its low efficiency suggests that its physiological significance is not fully understood . Here we found that recessive mutations in POLYAMINE UPTAKE TRANSPORTER 2 ( PUT2 ) enhance phyA-mediated germination . put2 seeds overaccumulate PAs and we show that PAs stimulate phyA-mediated germination . In WT seeds , our data suggest that upon phyB inactivation after an early FR pulse , PIF1 represses PAs accumulation . phyB activation by R light promotes PIF1 downregulation and PAs accumulation . Upon phyA activation by a second FR pulse , PIF1 is downregulated but , surprisingly , increase in PAs accumulation does not take place . We propose that this differential regulation of PAs levels arises from the duration of PIF1 extinction time , which is longer after R than after FR light irradiation . Accelerated aging procedures stimulated PAs accumulation and markedly enhanced phyA-mediated germination without downregulating PIF1 levels . Our results suggest that decaying seeds bypass PIF1-dependent repression of PAs accumulation to enhance phyA-mediated germination even under unfavorable canopy light conditions .
To better understand why phyA-mediated germination is limited in seeds , we sought to identify negative regulators of phyA signaling by screening for mutants displaying enhanced phyA-dependent seed germination responses . We generated a transgenic line carrying a firefly LUCIFERASE ( LUC ) reporter gene under the control of GA3ox1 promoter sequences ( pGA3ox1::LUC ) [20] . High GA3ox1 expression is characteristic of seeds undergoing germination [21] . pGA3ox1::LUC seeds were mutagenized using ethyl methanesulfonate ( EMS ) . FR/12h/FR assays lead to lower phyA-dependent germination relative to FR/48h/FR assays [16 , 17] . Thus , to identify negative regulators of phyA-dependent germination , we screened for pGA3ox1::LUC mutant seeds displaying high LUC bioluminescence and germination in a FR/12h/FR assay ( S1A Fig; see Materials and methods for details ) . Mutants identified in this manner were propagated and the resulting seed progeny was further studied . This led to identify 5 recessive and independent mutants , named fr/fr germination 1–5 ( ffg1-ffg5 ) having enhanced germination in a FR/12h/FR assay relative to the parental non-mutagenized pGA3ox1::LUC ( Parent . ) line ( Fig 1B–1D and S1B–S1D Fig ) . Unsurprisingly , among these mutants , we found two mutants ( ffg4 and ffg5 ) having high germination percentage in a FR assay ( S1B Fig ) . ffg4 had a G-to-A transition at nucleotide 370 in PIF1/PIL5 ( At2g20180 ) , which resulted in the substitution of Trp-94 with a Stop codon ( S1E Fig ) . ffg5 had a mutation at the splicing site of the third intron of the ABA biosynthetic gene ABA1 ( At5g67030 ) ( S1F Fig ) . PIF1/PIL5 and ABA were previously shown to negatively regulate phyB signaling and were therefore not further studied [17 , 22 , 23] . We also identified three independent recessive mutants that did not germinate in a FR assay but showed enhanced phyA-mediated germination ( Fig 1B–1D ) . We selected ffg1 for further study ( Fig 1E–1G ) . The ffg1 locus was mapped to a 200 kbp interval on chromosome 1 ( 11 . 3 ~ 11 . 5 Mbp . ) . Sequencing analysis revealed that ffg1 mutants had a G-to-A substitution at nucleotide 376 in the Arabidopsis locus At1g31830 ( PUT2 / LAT4 ( L-AMINO ACID TRANSPORTER 4 ) / PAR1 ( PARAQUAT RESISTANT 1 ) / PQR2 ( PARAQUAT-RESISTANT 2 ) ) , which converts Gly-126 to Arg . This G126R transition is hereafter referred as a put2-1 mutant allele ( Fig 2A ) . PUT2 encodes an amino acid permease family protein and put2 mutants were reported to display resistance to paraquat ( PQ ) , a methyl viologen widely used as a herbicide [24 , 25] . We found that put2-1 mutants were resistant to PQ , indicating that PUT2 activity is defective in put2-1 mutants ( S2A Fig ) . This suggested that PUT2 negatively regulates phyA-mediated germination . We observed enhanced phyA-mediated germination in two previously reported independent put2 mutant alleles , par1-1 [24] and par1-5 [24] ( hereafter put2-3 ) , and in a new one , named put2-2 , identified in this study after screening for PQ-insensitive mutant plants ( Fig 2B–2D and S2A Fig; see Materials and methods ) . Expectedly , phyAput2 double mutant seeds did not germinate in a FR/48h/FR assay , showing that the high percentage germination of put2-3 mutants in response to FR light is mediated by phyA ( Fig 2E–2G and S2B Fig ) . We also assessed germination percentages using different FR light fluences for the second FR pulse in a FR/48h/FR assay . put2-3 seed germination was enhanced relative to that of WT seeds under all FR fluences considered ( Fig 2H ) . Altogether , these results confirm that PUT2 encodes a negative regulator of phyA-mediated seed germination . We next sought to understand how PUT2 represses phyA-mediated responses in seeds . PUT2 belongs to the amino acid/polyamine/organocation ( APC ) transporter superfamily , which has four homologs in Arabidopsis: PUT1 , PUT3 , PUT4 and PUT5 [26–28] . The highest homology is found with PUT1 ( 75% identities , 83% positives , 3% gaps ) , followed by PUT3 ( 67% , 82% , 3% ) , PUT5 ( 53% , 69% , 7% ) and PUT4 ( 42% , 63% , 4% ) ( S3 Fig ) . Publicly available data show that PUT2 , PUT3 and PUT4 are expressed in developing , mature and imbibed seeds; however , PUT2 expression in developing seeds or upon seed imbibition is markedly higher relative to that of its homologs ( S4 Fig ) . This suggested that PUT2 is unique among its homologs in negatively regulating phyA-mediated seed germination . Indeed , the germination percentage of put1 , put3 , put4 and put5 mutant seeds exposed to a FR/48h/FR assay was low and similar to that of WT seeds ( Fig 3A–3C and S5 Fig ) . PUT2 , PUT1 and PUT3 were shown to transport polyamines ( PAs ) in yeast and plants [27 , 29 , 30] . This suggested that enhanced phyA-mediated seed germination of put2 seeds could result from defects in PAs cellular distribution or metabolism in seeds . The most abundant PAs in plants are putrescine ( Put ) , spermidine ( Spd ) and spermine ( Spm ) . Di-amine Put is made from arginine and is essential for plant survival [31] . Two spermidine synthases ( SPDS1 and 2 ) are responsible for tri-amine Spd synthesis from Put , and spermine synthase ( SPMS ) is responsible for tetra-amine Spm synthesis from Spd . Spd , but not Spm , is essential for plant survival [32 , 33] . We measured free Put , Spd and Spm accumulation in seeds exposed to a FR assay 24h after FR light irradiation . In WT seeds , overall PAs levels , i . e . the sum of Put , Spd and Spm levels , were variable between seed batches ranging from about 600 nmol per gram of seed dry weight ( nmol/g DW ) to 1 , 800 nmol/g DW ( S6A Fig ) . In every WT seed batch , Spd was the most abundant PA followed by Put and Spm . Overall PAs levels and the relative amounts of Put , Spd and Spm in put1 , put3 , put4 and put5 mutant seeds were similar to WT seeds ( Fig 3D ) . In contrast , overall PAs levels in put2-3 mutant seeds were systematically markedly higher relative to WT seeds among different seed batches , mostly due to higher Spd and Spm levels ( Fig 3D and S6B Fig ) . In dry seeds , higher PAs levels were also specifically observed in put2-3 mutants ( S6C Fig ) . Similar results were obtained with put2-1 , put2-2 and par1-1 alleles ( S6D Fig ) . These observations therefore suggest that high PAs levels could be responsible for the enhanced phyA-mediated seed germination in put2 mutant seeds , i . e . that PAs are positive regulators of phyA signaling in seeds . To further study the role of PAs in regulating phyA signaling in seeds , we considered assessing phyA-mediated responses in absence of endogenous PAs . However , PAs regulate fundamental cellular processes and mutations preventing PA biosynthesis in Arabidopsis are lethal , which makes this task challenging [8 , 31–33] . Instead , we considered identifying seeds with higher PAs levels . Five polyamine oxidases in Arabidopsis ( PAO1-PAO5 ) catabolize PAs by converting Spm to Spd and Spd to Put [34] . Previous reports showed that pao1—pao5 single mutants accumulate higher individual or overall PAs levels [35] . However , in some cases , as with pao4 mutants , lower levels in individual PAs were reported [36] . PAs levels in pao1—pao5 single mutant seeds were not previously reported . We observed moderately higher , i . e . less than twofold , overall PAs levels in pao1—pao5 seeds relative to WT ( Fig 3E ) . PAO1 and PAO5 are localized in the cytosol whereas PAO2 , PAO3 and PAO4 are localized in the peroxisome [34] . We measured PAs levels in pao1pao5 and pao2pao4 double mutants , defective in cytoplasmic and peroxisomal PAO activity , respectively [37] . We found more than twofold higher PAs levels in pao2pao4 double mutants whereas pao1pao5 double mutants had moderately higher , i . e . less than twofold , PAs levels ( Fig 3E ) . Strikingly , only pao2pao4 double mutant seed germination was markedly enhanced in a FR/48h/FR assay ( Fig 3F–3H ) . Furthermore , we observed that addition of individual PAs in the germination plates enhanced phyA-mediated germination ( Fig 3I ) . Altogether , these observations support the hypothesis that PAs are positive regulators of phyA-mediated seed germination . We sought to further evaluate this notion by identifying physiological conditions that could enhance PAs levels in seeds and whether they were associated with more efficient phyA-mediated germination . PAs act as antioxidants in plants where they accumulate in vegetative tissues [12 , 13] in response to oxidative stress . Seeds irremediably accumulate oxidative events as they age and oxidative stress is a major factor affecting seed viability [38–40] . Interestingly , we observed that the percentage of phyA-mediated seed germination markedly increased with old seed batches , reaching as much as 60% with five-year old seeds ( Fig 4A ) . This experiment was performed with seeds produced at different times , which could lead to differences in germination among seed batches . Nevertheless , these observations are consistent with the notion that phyA-mediated germination increases with oxidative stress . Whether oxidative stress promotes PAs accumulation in seeds was not previously investigated . To address this question , we subjected WT seeds to a controlled deterioration treatment ( CDT ) , which promotes oxidative stress and artificially accelerates seed aging [4 , 40] . Increased oxidative stress upon seed exposure to CDT was verified by measuring superoxide O2- levels in seeds ( S7A Fig ) . Next , seeds that had undergone CDT were exposed to a FR/48h/FR assay and PAs levels were measured 24h after the second FR light pulse . As anticipated , PAs levels increased after exposure to 6 days of CDT ( Fig 4B and S7B Fig ) . The increase in PAs levels was not as pronounced as in put2 mutants ( Fig 3D ) . Consistent with previous results , seeds exposed to white light or a FR/R assay decreased their seed germination percentage from 100% to 90% after exposure to CDT , indicating that seeds that had been exposed to CDT started to lose their viability [4 , 41] ( Fig 4C and S8A , S8B and S8C Fig ) . In contrast , the germination percentage in a FR/48h/FR assay markedly increased from 12% to 67% after exposure to CDT , i . e . nearer to the germination percentage observed in a FR/R assay ( Fig 4C ) . A similar trend was obtained with independent seed batches exposed to CDT ( S8B and S8C Fig ) . Altogether , these data further strengthen the notion that high endogenous PAs levels in seeds enhance phyA-mediated germination . We next sought to better understand 1 ) what makes phyA-mediated germination less efficient than phyB-mediated germination and 2 ) whether this low efficiency reflects how PA levels are regulated by light in seeds . These questions were addressed by studying the role of PIF1 , a key light regulated germination repressor , in regulating endogenous PAs levels . We first monitored endogenous PIF1 accumulation in seeds exposed to light treatments conducive of phyA-mediated germination , which was not previously reported . In a FR assay , which blocks WT seed germination , PIF1 levels rapidly increased between 1h and 6h ( Fig 5A and S9 Fig ) , remained high between 6h and 24h and slowly decreased thereafter , consistent with previous reports [42] . In contrast , phyA levels slowly increased between 1h and 12h and remained roughly similar between 24h and 48h , consistent with previous reports [17] , and further slowly increased between 48h and 96h ( Fig 5A ) . The phyA-mediated germination percentage of WT seeds exposed to a FR/12h/FR , and FR/48h/FR was 0% and 2% , respectively , whereas in seeds exposed to a FR/96h/FR assay it jumped to 37% despite the modest increase in phyA levels between 48h and 96h ( Fig 5B ) . This is consistent with previous results showing that phyA levels and the percentage of phyA-mediated germination are not linearly correlated [17] . In contrast , and as expected , close to 100% phyB-mediated germination was observed in WT seeds exposed to a FR/R assay ( Fig 5B ) . We monitored endogenous PIF1 accumulation in seeds exposed to the different above FR/Nh/FR assays ( N = 12 , 48 , 96 ) , conducive of phyA-mediated germination . Irrespective of the time of its application , the second FR pulse did not affect markedly phyA protein levels ( Fig 5C ) . Unexpectedly , however , the second FR pulse triggered , within one hour , rapid PIF1 downregulation in all assays ( Fig 5C ) . As expected , endogenous PIF1 downregulation was not observed in phyA mutant seeds but observed in phyBCDE mutant seeds , showing that PIF1 downregulation is driven by phyA [43] ( Fig 5C ) . Thus , and surprisingly , FR and R light are similarly able to downregulate PIF1 levels even though they do not stimulate germination with the same efficiency . We therefore wondered whether the duration of PIF1 extinction time could be different after R and later FR light irradiation . Indeed , the duration of PIF1 extinction time was about 12h longer after a R pulse than after a second FR pulse in both a FR48hFR assay or a FR96hFR assay ( Fig 5D ) . Altogether these observations show that phyA levels in seeds are not limiting to promote PIF1 downregulation in response to a second FR light pulse applied at different times upon imbibition . They rather suggest that phyA-mediated germination is inefficient , at least in part , due to the short duration of PIF1 extinction following FR irradiation . Interestingly , publicly available data indicate that expression of four polyamine biosynthesis genes is higher in pif1 mutant seeds relative to WT seeds exposed to a FR assay [44] ( S10A–S10D Fig ) . In contrast , expression of PUT2 is low in pif1 seeds relative to WT seeds exposed to a FR assay [44] ( S10E Fig ) . Furthermore , expression of these genes was similar in WT and pif1 seeds exposed to a FR/R assay ( S10A–S10E Fig ) . This suggested that PIF1 represses PAs accumulation in seeds exposed to a FR assay . To evaluate this possibility , we measured PAs levels in WT and pif1 seeds exposed to a FR assay . Consistent with this hypothesis , PAs levels were higher in pif1 mutant seeds relative to WT seeds exposed to FR assay ( Fig 6A ) . These data are consistent with the view that PIF1 represses PAs accumulation in seeds after an early FR light pulse . Next , we investigated whether PAs levels increase under conditions conducive of phyA-mediated germination . Unlike WT seeds exposed to a R pulse , PAs levels did not change after a second FR pulse relative to seeds exposed to a single FR pulse ( Fig 6B and S11 Fig ) . Altogether , these results suggest that the short duration of PIF1 extinction upon a second FR light pulse irradiation does not permit the elevation of PAs levels in seeds . As a result , low PAs levels would contribute to the low efficiency of phyA-mediated germination ( see model below ) . We next investigated whether increased phyA-mediated germination in put2 mutants or in WT seeds that had undergone CDT was linked to changes in phyA and PIF1 levels in seeds . In a FR assay , PIF1 levels in put2-3 mutant seeds were similar to those in WT seeds ( Fig 7A ) . put2-3 mutant seeds accumulated normal phyA levels up to 12h after imbibition ( Fig 7A ) . At 48h phyA levels were higher in put2-3 seeds relative to WT seeds; we did not further investigate this matter . Strikingly , a FR/12h/FR assay markedly stimulated germination of put2-3 seeds relative to WT seeds ( Fig 7B ) , even though they accumulated similar phyA levels ( Fig 7C; 0 h and S12A and S12B Fig ) . Higher germination percentage of put2-3 seeds was not associated with obvious differences in PIF1 levels , as assessed by data quantification in experiments with biological replicates , including in the duration of PIF1 extinction time after the second FR pulse ( Fig 7C and S12A and S12B Fig ) . Furthermore , in a FR/48h/FR assay the duration of PIF1 extinction time after the second FR pulse was similar between WT and put2-3 seeds even though the percentage of put2-3 seed germination was 100% whereas that of WT seeds was only 2% ( Fig 7B and 7D and S13A–S13C Fig ) . After an early FR pulse , WT seeds that had undergone CDT had no obvious changes in phyA or PIF1 levels relative to untreated WT seeds up to 48h after FR light irradiation , as assessed by data quantification in experiments with biological replicates ( Fig 7E and S14A Fig ) . After a second FR pulse ( FR/48h/FR assay ) , the percentage of germination of WT seeds exposed to CDT was enhanced compared to unexposed WT seeds without obvious changes in phyA or PIF1 levels , as assessed by data quantification in experiments with biological replicates ( Fig 7E and 7F and S14B–S14D Fig ) . These observations show that enhanced phyA-mediated germination in put2 seeds or WT seeds exposed to CDT can take place without marked changes in PIF1 levels . They therefore suggest that oxidative stress promotes PA accumulation , which in turn promotes phyA-mediated germination downstream of PIF1 or else independently of PIF1 ( Fig 8 ) .
Here we sought to better understand how seed germination is promoted by canopy light through phyA . We focused on understanding why this process is poorly efficient and its physiological relevance . Downstream phyA signaling components in seeds are poorly characterized . Their identification and study is rendered difficult by the fact that observing phyA-mediated germination requires first blocking germination through phyB inactivation . In the case of pif1 mutants , which fully germinate after an early FR pulse , germination mediated by phyA cannot be observed . Nevertheless , it is generally assumed that PIF1 represses germination downstream of phyA [23 , 43] . Here we found that put2 mutants are specifically enhanced in phyA-mediated germination ( Figs 1 and 2 ) , which was not previously reported . PUT2 encodes a PAs transporter and we showed that among other mutants deficient in homologous transporters only put2 mutants accumulate high PAs levels in seeds ( Fig 3 ) . Similarly , mutants deficient in PAs catabolism or WT seeds exposed to CDT increased PAs levels in seeds and enhanced phyA-mediated germination ( Figs 3 and 4 ) . Thus , our study provides correlative evidence suggesting that increased PAs levels in seeds positively regulate phyA-mediated germination . High PAs accumulation in put2 seeds could result from inappropriate distribution of PAs in cells . PUT2 is localized in the Golgi apparatus and the chloroplast [24 , 45] . A put2 mutant cell sensing low PAs accumulation in the Golgi or chloroplast might respond with increased PAs synthesis as a compensatory mechanism . put2 had increased Spd and Spm levels whereas WT seeds exposed to CDT and pao2pao4 mutants had mainly higher Spd levels . Furthermore , PUT2 preferentially transports Spd in yeast cells [29] . This could suggest that Spd is important to enhance phyA-mediated germination . However , exogenous application of Put , Spd and Spm similarly promoted phyA-mediated germination ( Fig 3 ) . Thus , our study does not pinpoint a particular individual PA specifically promoting phyA-mediated germination . Understanding the role of an individual PA in phyA signaling using genetic approaches is difficult since , to our best knowledge , there are no reported PA biosynthesis mutants accumulating a single PA . It remains to be understood how PAs promote phyA signaling in seeds . L-arginine is a precursor of Put and S-adenosylmethionine is a precursor of Spd and Spm . However , they are also precursors of nitric oxide ( NO ) and ethylene , respectively . Indeed , enhanced PAs accumulation was linked to an increase in NO levels and associated with both an increase and decrease in ethylene signaling [46–48] . NO and ethylene repress ABA responses in seeds and therefore PAs could enhance germination by repressing ABA signaling through NO or ethylene [49] . Against this possibility , we found that put2 mutants have normal ABA responses in seeds , consistent with a previous report showing that par1 mutants have normal ABA responses [24] ( S15 Fig ) . PAs are involved in a wide range of fundamental cellular processes including DNA replication , transcription , translation and post-translational modification [7] . PAs were also speculated to participate in abiotic stress signaling in plants [50] . In all cases , the mechanism by which PAs act is poorly understood as they are difficult to study due to their ubiquitous presence in cells and their essential function for survival [6] . Our data suggest that endogenous PAs accumulation is repressed by PIF1 in WT seeds exposed to an early FR pulse . This indicates that PAs biosynthesis is regulated downstream of PIF1 . The biological significance of the regulation of PAs biosynthesis gene expression and PAs levels in seeds by light remains to be understood . On the other hand , PAs levels were higher in put2 mutants or in WT seeds exposed to CDT and the resulting increase in phyA-mediated germination took place without changes in PIF1 levels . Therefore , this suggests that PAs levels can be regulated independently of PIF1 to regulate phyA-mediated germination . PAs could promote phyA-mediated germination downstream of PIF1 . However , transcriptomic studies have suggested that phyA activation by FR also triggers gene expression changes independently of PIF1 [18] . Thus , PAs could promote phyA-mediated germination in a PIF1-independent manner ( Fig 8; see putative “X” pathway in the model ) . PAs levels in seeds increased in response to R light but not in response to a second FR light pulse ( Fig 6 ) . We propose that this is due to the short PIF1 extinction time following FR irradiation . This shorter time was not due to limiting phyA levels since it remained unchanged in WT seeds exposed to a FR/96h/FR assay or in put2 seeds exposed to a FR/48h/FR assay , which had higher phyA levels ( Figs 5 and 7 ) . Therefore , PIF1 reaccumulation is differently regulated after R and FR light irradiation , respectively . The underlying mechanism accounting for this differential regulation remains to be identified . It is also unknown why put2 mutants accumulate higher phyA levels at later time points upon seed imbibition . It is generally accepted that PAs act as antioxidants in plants [51] . Seed oxidation is an unavoidable process compromising seed viability in the dry seed state . It is therefore expected that seeds have evolved adaptive mechanisms to sense oxidative damage and adapt their behavior such as their control of seed germination . In newly produced seeds , the first level of germination control is that of primary seed dormancy , a trait whereby germination is blocked even under favorable conditions . Dormancy prevents germination out of season . Seeds lose dormancy during a period of dry storage called dry after-ripening [2] . Seed oxidation is known to accelerate the release of seed dormancy during after-ripening [1] . Dormancy was shown to inhibit R- and phyB-mediated germination [52] and therefore it is expected that oxidation promotes phyB-mediated germination . However , the oxidation events that release dormancy are not sufficient to promote germination mediated by phyA: 18-month-old WT seeds , which have fully lost dormancy and fully germinated after a R pulse , germinated at less than 5% in a FR/48h/FR assay ( Fig 2 ) . Our results therefore suggest that there are two levels of seed germination regulation through oxidative stress . Younger and still viable seeds , with moderate levels of oxidative damage , have an advantage to repress their germination under canopy light , which is unfavorable for photosynthesis . As seeds age , however they decay as a result of continuous accumulation of oxidative damage , which compromises their capacity to form a viable seedling . It would then become advantageous to germinate under a broader range of light wavelengths , including canopy light [19] . In this context , increased PAs levels upon seed imbibition could serve a dual function: protect the decaying seed from the oxidative damage that accumulated during dry after-ripening while promoting in parallel phyA-mediated germination under unfavorable light cues such as canopy light . This would represent a mechanism providing a last chance for plant survival ( Fig 8 ) .
Arabidopsis T-DNA insertion lines , all in the Col-0 background , were obtained from the Nottingham Arabidopsis Stock Centre with the following accession numbers: put2-3; SALK_119707 , put1; SAIL_270_G10 , put3; SALK_206472 , put4; SAIL_1275_C06 , put5; SALK_122097 , pao1-2; SAIL_882_A11 , pao2-4; SALK_046281 , pao3-1; GABI_209F07 , pao4-1; SALK_133599 , pao5-2; SALK_053110 and phyA-211; N6223 . phyA-211put2-3 double mutants were generated after crossing phyA-211 and put2-3 plants . par1-1 [24] , pao1pao5 [37] , pao2pao4 [37] and phyBCDE [53] seeds were kindly provided by Jianru Zuo ( Chinese Academy of Sciences , China ) , Tomonobu Kusano ( Tohoku University , Japan ) and Pablo D . Cerdán ( Fundación Instituto Leloir , Argentina ) , respectively . All genotypes tested in each experiment were grown together under the same conditions and seeds were harvested the same day and allowed to after-ripen at room temperature for at least one month . For the germination assays , seeds were surface sterilized and 50–60 seeds of each genotype were sown on MS medium ( Sigma ) containing 0 . 8% ( w/v ) agar without seed stratification . For the germination assays in presence of polyamines , individual polyamines ( Sigma ) were added to the germination medium . In a FR assay , seeds were irradiated with a FR pulse ( 3 . 69 μmol m-2 s-1 ) for 5 min after 2 h seed imbibition under white light . In a FR/Nh/FR assay , seeds were irradiated with a first FR pulse ( 3 . 69 μmol m-2 s-1 ) for 5 min and further irradiated with a second FR pulse ( 3 . 69 μmol m-2 s-1 or as indicated in each experiment ) for 5 min after N ( e . g . 12 , 48 and 96 ) hours of dark incubation . In a FR/R assay , seeds were irradiated with a red ( R ) pulse ( 14 . 92 μmol m-2 s-1 ) for 5 min followed by a FR pulse . In all assays , light irradiated plates were kept in the dark for the indicated times . Thereafter a seed that had undergone endosperm rupture , i . e . radicle protrusion , was scored as a germination event . All the germination assays were performed with three technical replicates and the results were confirmed with at least two or three independent biological seed samples . Data value distribution among biological samples is shown by scatterplots as described in Weissgerber et al . ( 2015 ) [54] . Approximately 3kb of GA3ox1 promoter region [20] was amplified with primers ( 5’-CGCGGATCCCACCAGAGTGTGTGCTACATGC-3’ and 5’-CCGCTCGAGAACACAGCAGGCAGCTTGCTC-3’ ) . BamHI and XhoI restriction sites were used for cloning into binary vector pGPTVII [55] . WT ( Col-0 ) plants were transformed with a firefly luciferase ( LUC ) reporter gene under the control of GA3ox1 promoter sequences ( WT/pGA3ox1::LUC ) . With the aim of identifying mutants displaying enhanced phyA-mediated seed germination responses , a population of 20 , 000 WT/pGA3ox1::LUC seeds ( M0 ) was chemically mutagenized using 0 . 3% ethyl methanesulfonate ( EMS ) as previously described [56] . In M2 populations , mutants able to germinate in a FR/12h/FR assay or displaying high LUC bioluminescence were selected for further analysis ( Fig 1 and S1 Fig ) . LUC bioluminescence was performed as previously described [55] . Briefly , plants exposed to FR/12h/FR assay were sprayed with a luciferin ( Biosynth ) solution ( 315 μg/ml ) , under green safety light , 24h after the second FR pulse and examined using an Aequoria dark box with a mounted ORCAII CCD camera ( Hamamatsu ) . This led to identify three recessive and independent mutants ( ffg1—ffg3 ) having enhanced bioluminescence and germination in a FR/12h/FR assay relative to the parental non-mutagenized WT/pGA3ox1::LUC line . The same mutagenized seed population was used to identify put2-2 as previously described [24 , 25] . Briefly , the same mutagenized seed population was sown on a germination medium with 0 and 10 μM of PQ and cultured under WLc for 6 days to reveal PQ-insensitive mutants , which led to the identification of the put2-2 allele . For map-based cloning and whole genome sequencing , the ffg1 mutant was outcrossed to Ler . The ffg1 locus was mapped as previously described [57] . Briefly , a combination of cleaved amplified polymorphic sequences ( CAPS ) markers and simple sequence length polymorphisms ( SSLPs ) markers was used for fine mapping . The ffg1 locus was mapped to a 200 kbp interval on chromosome 1 ( 11 . 3 ~ 11 . 5 Mbp ) . Identification of mutations in this interval was done after sequencing the ffg1 mutant genome . Genomic library preparation was performed using TrueSeq® DNA Library Prep Kit ( Illumina ) according to manufacturer’s instructions . Sequencing was performed using HiSeq2000 ( Illumina ) . This led to the identification the put2-1 allele in the 200 kbp interval . To avoid detecting differences in PAs levels arising from plants at different development stages , we measured endogenous PAs levels in dry or non-germinated seeds that are harvested under green safety light . Standards of polyamines as well as the other chemicals and reagents were purchased from Sigma Aldrich chemical company ( St . Louis , MO , USA ) . Free PAs were isolated and derivatized by slightly modified method of previous report [58] . A 250 mL of 5% trichloroacetic acid ( TCA ) was added to a 5 mg of lyophilized seeds and homogenized using ZrO2 beads ( 3 mm ) in mixer-mill for 5 min at 27 Hz . Sample was then sonicated for 10 min at 25 °C , and after centrifugation at 12 , 400 × g for 5 min , supernatant was quantitatively transferred into another vial . A 500 μL of 2M NaOH was added following with 2 . 5 μL of benzoyl chloride ( in methanol 50:50 , v:v ) , and after vortexing for 5 sec reaction mixtures are left for 40 min at 25 °C . A 500 μL of saturated NaCl was added and benzoylated polyamines were extracted with 2 × 500 μL of diethyl ether . Ether was evaporated and dry samples were stored at -80 °C until analysis . All samples were dissolved in 50 μL of mobile phase ( 45% methanol in 15 mM formic acid , pH 3 . 0 ) , sonicated for 15 min , and centrifuged for 5 min at 12 , 400 × g prior to the analysis . Diaminohexane ( DAH ) was used as internal standard . Ultra-high performance liquid chromatography-tandem mass spectrometry ( UHPLC-MS/MS ) was performed on UltiMate™ 3000 liquid chromatographic system consisting of binary pumps , an autosampler and a column thermostat coupled to a TSQ Quantum Access Max triple quadrupole mass spectrometer ( Thermo Fisher Scientific , Waltham , MA , USA ) . Chromatographic separation was performed on an Acquity UPLC BEH C18 ( 50 × 2 . 1 mm; 1 . 7 μm particle size ) column ( Waters , Milford , MA , USA ) with appropriate pre-column kept at 40 °C . The mobile phase consisted of a mixture of aqueous solutions of 15 mM formic acid adjusted pH 3 . 0 with ammonium hydroxide ( Solvent A ) and methanol ( Solvent B ) . The analytes were separated using a binary gradient starting at 45% of B for 2 . 7 min , then increase to 57% for 0 . 3 min , isocratic at 57% for next 2 . 5 min , increase to 100% B for next 0 . 1 min , isocratic at 100% B for next 1 min , and decrease to 45% B for next 0 . 1 min . Finally , the equilibration to the initial conditions took 2 . 3 min . The flow rate was 0 . 4 mL/min and the injection volume 5 μL . Benzoylated PAs were detected in positive ionization mode electrospray ionization ( ESI+ ) . The selected reaction monitoring ( SRM ) transitions for benzoylated putrescine ( Put ) were 297 > 105 , and 297 > 176 at 20 eV collision energy ( CE ) , for spermidine ( Spd ) 458 > 175 , and 458 > 233 at 25 eV CE , and for spermine ( Spm ) 619 > 162 , 619 > 337 , and 619 > 497 at 30 eV CE . The spray voltage was set to 3 kV , the vaporizer temperature to 350 °C , and the ion transfer tube temperature to 320 °C , respectively . CDT was performed as described in a previous report [4] . Briefly , Col-0 dry seeds were stored in a closed container during the time period indicated in each experiment . The container was maintained at 37 °C and its interior had around 82% of relative humidity imposed by the presence of a saturated salt ( KCl ) . Then , seeds were dried back at 30% relative humidity ( room temperature ) and stored at -80 °C until they were used for superoxide O2- levels measurements , germination assays or protein gel blots analysis . The levels of superoxide O2- were determined based on its ability to reduce nitro blue tetrazolium ( NBT ) to blue formazan as previously described [59 , 60] . Dry seeds ( 0 . 035g ) were grinded in 1 . 3 mL of incubation solution ( 10 mM K-phosphate buffer pH 7 . 8 , 10 mM NaN3 , 0 . 05% NBT ) . After 30 min of incubation at room temperature , grinded tissue was collected at the bottom of the tube by centrifugation ( 8000 x g , 5 min , room temperature ) . Supernatant was diluted 10 x in incubation buffer , heated 85 °C for 15 min and then cooled on ice . Absorbance at 580 nm was measured by spectrophotometer to quantify NBT levels . PIF1 recombinant proteins were prepared using PIF1-his DNA ( pET21a ) provided by Enamul Huq ( University of Texas at Austin , USA ) , and induced and purified using a commercial kit according to manufacturer’s instructions ( Amersham ) . Polyclonal anti-PIF1 was obtained from rabbits immunized with PIF1 recombinant protein . PIF1 antibodies were further affinity-purified using PIF1 recombinant protein immobilized on nitrocellulose filters as described [61] . FR or R light irradiated WT and mutant seeds were harvested under green safety light at the time indicated in each experiment . 20 seeds were homogenized with homogenization buffer ( 0 . 0625 M Tris-HCl at pH 6 . 8 , 1% [w/v] SDS , 10% [v/v] glycerol , 0 . 01% [v/v] 2-mercaptoethanol ) , and total proteins were separated by SDS-PAGE gel and transferred to a PVDF membrane ( Amersham ) . PIF1 and UGPase proteins were detected using 1:500 dilution of anti-PIF1 or 1:10 , 000 dilution of anti-UGPase ( Agrisera ) , and anti-rabbit IgG HRP-linked whole antibody ( GE healthcare ) in a 1:10 , 000 dilution was used as a secondary antibody . PHYA proteins were detected using 1:5 , 000 dilution of anti-phyA ( kindly provided by Akira Nagatani; Kyoto University , Japan ) and the anti-mouse IgG HRP-linked whole antibody ( GE healthcare ) in a 1:10 , 000 dilution was used as a secondary antibody [17] . Quantification of band intensity was performed using imageJ . | Canopy light , unfavorable for photosynthesis , elicits paradoxical responses in seeds . Depending on the timing of irradiation upon seed imbibition , canopy light can either block or promote germination . The promotion effect is mediated by the light sensor phyA and , intriguingly , it is poorly efficient when compared to that of red light , which is favorable for photosynthesis . Why would germination be stimulated by canopy light , which is not favorable for young seedling photosynthesis ? We show that phyA-mediated germination is greatly enhanced in decaying seeds , which overaccumulate polyamines , a class of antioxidant molecules that promote germination . We propose an adaptive rationale for phyA-mediated germination: in decaying seeds it becomes advantageous to germinate even under canopy light as a last chance for seedling survival . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"plant",
"anatomy",
"oxidative",
"stress",
"brassica",
"light",
"plant",
"physiology",
"electromagnetic",
"radiation",
"plant",
"science",
"model",
"organisms",
"experimental",
"organism",
"systems",
"seedlings",
"plants",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"plant",
"reproduction",
"animal",
"studies",
"light",
"pulses",
"seeds",
"antioxidants",
"physics",
"biochemistry",
"eukaryota",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"seed",
"germination",
"organisms"
]
| 2019 | polyamine uptake transporter 2 (put2) and decaying seeds enhance phyA-mediated germination by overcoming PIF1 repression of germination |
Mating-type switching in Schizosaccharomyces pombe entails programmed gene conversion events regulated by DNA replication , heterochromatin , and the HP1-like chromodomain protein Swi6 . The whole mechanism remains to be fully understood . Using a gene deletion library , we screened ~ 3400 mutants for defects in the donor selection step where a heterochromatic locus , mat2-P or mat3-M , is chosen to convert the expressed mat1 locus . By measuring the biases in mat1 content that result from faulty directionality , we identified in total 20 factors required for donor selection . Unexpectedly , these included the histone H3 lysine 4 ( H3K4 ) methyltransferase complex subunits Set1 , Swd1 , Swd2 , Swd3 , Spf1 and Ash2 , the BRE1-like ubiquitin ligase Brl2 and the Elongator complex subunit Elp6 . The mutant defects were investigated in strains with reversed donor loci ( mat2-M mat3-P ) or when the SRE2 and SRE3 recombination enhancers , adjacent to the donors , were deleted or transposed . Mutants in Set1C , Brl2 or Elp6 altered balanced donor usage away from mat2 and the SRE2 enhancer , towards mat3 and the SRE3 enhancer . The defects in these mutants were qualitatively similar to heterochromatin mutants lacking Swi6 , the NAD+-dependent histone deacetylase Sir2 , or the Clr4 , Raf1 or Rik1 subunits of the histone H3 lysine 9 ( H3K9 ) methyltransferase complex , albeit not as extreme . Other mutants showed clonal biases in switching . This was the case for mutants in the NAD+-independent deacetylase complex subunits Clr1 , Clr2 and Clr3 , the casein kinase CK2 subunit Ckb1 , the ubiquitin ligase component Pof3 , and the CENP-B homologue Cbp1 , as well as for double mutants lacking Swi6 and Brl2 , Pof3 , or Cbp1 . Thus , we propose that Set1C cooperates with Swi6 and heterochromatin to direct donor choice to mat2-P in M cells , perhaps by inhibiting the SRE3 recombination enhancer , and that in the absence of Swi6 other factors are still capable of imposing biases to donor choice .
The fission yeast S . pombe exists as two haploid cell types , plus ( P ) and minus ( M ) , that differ at the mat1 locus . When starved for nitrogen , haploid cells undergo sexual differentiation , mate with the opposite cell type and sporulate . These events are driven by master regulators expressed from the mat1-P and mat1-M alleles [1] . The regulators first drive sexual differentiation and mating and , when co-expressed in the zygote , meiosis and sporulation . Homothallic ( h90 ) colonies sporulate very efficiently because they contain equal proportions of P and M cells due to frequent gene conversions at mat1 . The genetic information at mat1 is replaced with genetic information copied from one of two silent loci , mat2-P or mat3-M [2] . The organization of the ~35 kb region of chromosome 2 that comprises mat1 , mat2-P and mat3-M in h90 strains is depicted in Fig 1A . All three loci are flanked by short regions of sequence identity , the centromere-distal H1 box and the centromere-proximal H2 box . In addition , H3 homology boxes immediately adjacent to H2 are found exclusively at mat2-P and mat3-M . An alternative arrangement , known as h09 , has mat2-M and mat3-P cassettes [3] . Mating-type switching follows the so-called ‘Miyata’s rules’ inferred from pedigree analyses of dividing cells [4] . A single h90 cell produces both ‘unswitchable’ and ‘switchable’ cells . According to the ‘one-in-four-rule’ , illustrated in Fig 1B , one of two sister cells originating from an ‘unswitchable’ cell becomes ‘switchable’ . That cell produces one switched daughter upon cell division and one unswitched , but switchable , daughter . Thus , only one cell out of four cousins displays a switched mating-type ( the ‘one-in-four’ rule [4] ) and lineages of ‘switchable’ cells are observed ( the ‘recurrent switching’ rule [5] ) . The known mechanisms of mating-type switching provide an explanation for these rules . Mating-type switching is initiated by an imprint at mat1 introduced during DNA replication [6 , 7] . Replication stalls within mat1 at the MPS1 site and a nick or two ribonucleotides are incorporated into the lagging strand at the imprinting site situated nearby , at the junction of the mat1 cell-type specific information and H1 box [6–15] . This imprint creates a ‘switchable’ cell . During the next round of DNA synthesis , the imprint is converted into a double-strand break ( DSB ) that triggers homologous recombination and mating-type switching [10 , 14] . At least seven factors ( Swi1 , Swi3 , Pol1 ( Swi7 ) , Sap1 , Lsd1 , Lsd2 and Mrc1 ) are required for efficient DSB formation . The Swi1-Swi3 complex and Mrc1 are necessary for imprinting by pausing replication forks at MPS1 [7 , 16] . The DNA primase Pol1 and a DNA element targeted by the essential DNA-binding protein Sap1 are not required for replication fork stalling at MPS1; hence Pol1 and Sap1 are believed to catalyze imprint formation downstream of the Swi1-Swi3 complex [7] , while Lsd1 and Lsd2 are required upstream of Swi1-Swi3 [17 , 18] . In addition , Swi1 , Swi3 and Mrc1 block replication from the mat1 distal side at the replication termination site RTS1 to optimize mating-type switching [7 , 16] . Thus , replication proceeds unidirectionally when leading-strand synthesis reaches the imprint and the nick is converted into a one-ended DSB . The DSB end initiates repair by recombining with one of the heterochromatic and transcriptionally silent cassettes mat2-P and mat3-M [10 , 19 , 20] . The free DNA end can invade either mat2-P or mat3-M , however mating-type information opposite to the information present at mat1 is chosen with a ~90% probability , leading to Miyata’s observation that switchable cells nearly always switch to the opposite mating-type [4 , 9] . Surprisingly , this strongly biased donor selection relies on the heterochromatic state of the mat2-mat3 region [3 , 21–24] . In the wild-type , histone H3K9 methylation deposited between the inverted repeat boundaries IR-L and IR-R permits the binding of the key switching factor Swi6 , an HP1 homolog [25] . Defective donor choice in swi6 mutants biases h90 cell populations towards the M mating-type due to preferred use of a mat3-M adjacent recombination enhancer over a mat2-P adjacent enhancer in the absence of Swi6 [22 , 23 , 26] . Also essential at this step of switching is the Swi2-Swi5 complex , capable of interacting with Swi6 , and whose molecular role is inferred from the related Sfr1-Swi5 complex [27 , 28] . Sfr1 shares sequence homology with the C-terminus of Swi2 , in a domain that permits the interaction of either Swi2 or Sfr1 with the recombination mediator Swi5 and with the strand-exchange factor Rad51 . The Sfr1-Swi5 complex stabilizes Rad51 filaments in vivo and promotes Rad51-mediated strand exchange in vitro [28–30] . Consistent with similar functions for the Swi2-Swi5 complex , Swi2-Swi5 interacts with Rad51 in two-hybrid assays [27] . Thus , mechanistically , the ability of Swi2-Swi5 to interact with Swi6 suggests that the complex participates in donor choice by biasing strand invasion [27] . This idea is supported by cell-type specific associations of Swi2 and Swi5 with the mating-type region [22] . Swi2 localizes to the mat2-P and mat3-M adjacent enhancers SRE2 and SRE3 ( Swi2-dependent recombination enhancer element 2 and 3 ) in M cells , but only to SRE3 in P cells [22 , 23] . In swi6 mutant cells , Swi2 localizes only to SRE3 , as in P cells [22] . These observations indicate that the heterochromatin-mediated localization of Swi6 regulates Swi2-Swi5 localization to SRE2 to choose mat2-P [22 , 23 , 31] . After strand invasion into the homologous sequence at the H1 homology box , the donor locus information is copied by polymerase extension until it reaches the H3 homology box , then H3 is believed to form a hairpin loop structure [32] . The mismatch repair Msh2 ( Swi8 ) -Msh3 ( Swi4 ) complex recognizes this conformation and DNA synthesis stops at the donor cassette . The Rad51 mediator Rad55-Rad57 has been suggested to work together with Msh2 ( Swi8 ) , because mutants in these factors tend to form h+N rearrangements containing a duplication of the entire mat2-3 region at mat1 [33–36] . In addition to Rad55-Rad57 , the homologous recombination factor Rad52 is required for DSB repair at mat1 [37 , 38] . Presumably , Rad55-Rad57 and Rad52 are involved in annealing between two H2 boxes in mat1 and the donor cassette . The endonuclease Rad16 ( Swi9 ) -Swi10 and its activator Pxd1 cleave the intermediate between the H2 and H3 boxes [37 , 39] . This is followed by new DNA synthesis from H2 of mat1 to H1 to complete replication of this region and to thus switch mating-type . This switching system has been utilized to study multiple aspects of replication , histone modification and recombination . Historically , the mating-type switching related genes were classified functionally by Southern blotting analysis of the effect of mutants on mat1 switching [40] . Class Ia genes ( swi1 , 3 , and 7 ) are required for the imprinting step that leads to the DSB formation , hence mutants in these genes show no DSB in a Southern blot . Class Ib genes ( swi2 , 5 and 6 ) are not necessary for DSB formation , but required for efficient switching . The third group , Class II , ( swi4 , 8 , 9 , and 10 ) resolves the recombination intermediate , mutants in these genes contain frequent rearrangements of the mating-type region , in particular the h+N duplication . Subsequent screens identified additional factors ( Table 1 ) and suggested that yet more might exist . In particular , aspects of imprint formation and donor choice are still not understood . To identify yet unknown regulators of mating-type switching , we combined the deletion of 3420 nonessential genes ( Bioneer gene deletion library version 5 ) with an h90 strain background . The strain also included a dual reporter system with CFP under the control of a P-specific promoter and YFP under the control of an M-specific promoter , so that the ratio of P-to-M cells was determined by comparing CFP and YFP fluorescence . As a secondary screening strategy , the genetic content at mat1 was quantified with multiplex PCR using genomic DNA isolated from the candidates that passed the initial screen . These extensive screens identified several new mating-type switching genes whose deletion results in a bias toward M cells within h90 populations instead of the balanced P:M ratio . In addition , analysis of h09 strains revealed that some strains showed clonal biases in independent colonies . As mentioned above , Swi6 is an essential directionality factor [3 , 22 , 23] . Epistasis analysis with swi6Δ suggests that Clr4 , Sir2 , Swd1 , Clr3 work in the same pathway as Swi6 , whereas Brl2 , Pof3 and Cbp1 act via Swi6-dependent and -independent mechanisms . These observations provide new clues to understand the molecular mechanisms of mating-type switching .
We conducted a genome-wide screen for factors required for mating-type switching . The screen used an S . pombe gene deletion library ( Bioneer ) consisting of 3420 haploid strains , each of which lacks a non-essential gene . Mating-type switching occurs in h90 strains , yet the Bioneer library strains are heterothallic h+N strains for which a large duplication in the mating-type region abrogates mating-type switching . To construct h90 derivatives of the entire Bioneer collection , strain PG4045 was mated to the library . The h90 mating-type region of PG4045 could be selected in the progeny due to the linked LEU2 gene . In addition , PG4045 contains two fluorescent reporters specific for the P ( CFP controlled by the map2 promoter ) and M ( YFP controlled by the mfm3 promoter ) cell types , respectively . Both reporters are expressed from the leu1 locus where they were integrated together with the selectable ura4+ gene . Thus , we selected h90 LEU2 ura4+ segregants ( S1 Fig ) . We obtained 3298 h90 deletion strains in which we monitored expression of the fluorescent reporters ( Fig 2A ) . Efficient mating-type switching results in rapid homogenization of h90 cell populations to equal proportions of P and M cells ( Fig 1B ) . Here , screening specifically for mutants that displayed biased cell-type ratios , differing by > 3 standard deviations from the mean , we isolated 105 candidates with skewed proportions ( Fig 2B , S2 Fig , S2 Table ) . In several deletion strains , we detected co-expression of CFP and YFP in a cell . This phenotype was most likely caused by derepression of the mating-type information at mat2-P and mat3-M in these mutants [42] . In addition , 568 strains that could not be evaluated due to low fluorescence intensity or poor growth were examined by iodine staining , a stain for S . pombe spores that can be used as diagnostic for mat1 switching ( Fig 2C , S3 Table ) . Wild-type h90 colonies are stained darkly by iodine vapors because of their high spore content while mutants with altered mating or sporulation are stained less . Here , 124 deletion strains among the strains tested showed a staining different from wild-type . They were analyzed by quantitative multiplex PCR for mat1 content alongside the 108 candidates that had passed the fluorescence microscopy screening . Analyzing the content of mat1 permitted us to pinpoint mutants for which biased cell-type expression or poor sporulation is likely to result from switching defects . A number of candidates failed to show a biased mat1 content by multiplex PCR according to the chosen thresholds of P band intensity ( Fig 2A and 2D ) . Indeed , many mutations might result in biased reporter gene expression or altered sporulation without affecting mating-type switching , for example genes located in the L region of the mat locus were eliminated from this screening . In addition , 35 deletion mutants that were diploid and/or heterothallic ( h+N ) were eliminated from the list of candidates ( S4 Table ) . A bias in mat1 content was detected for 32 mutants , in all cases towards mat1-M , suggesting an increased use of mat3-M . The identity of the deleted gene in these mutants was confirmed using published barcode sequences or gene specific primers . Surprisingly , 9 genes encoding ribosomal proteins are in the list , possibly as a consequence of protein synthesis defects . We did not pursue the investigation of these mutants , but focused instead on the remaining 23 mutants . These included nearly all known non-essential switch-related genes , 16 of which were identified in total . A few switch-related genes ( swi1 , 5 or 10 ) were not deleted in the Bioneer library and were therefore not tested by the screen . A few other mutants might have escaped detection due to clonal variation , alternative switching phenotype with low frequency as for msc1Δ [49] or due to a switching bias close to the set thresholds , as for mrc1Δ [16] . Globally though the screening strategy was strongly validated by the identification of known switch-related factors alongside novel factors . The 7 newly identified factors include the F box protein Pof3 , the CK2 family regulatory subunit Ckb1 , the elongator complex subunit Elp6 , the E3 ubiquitin protein ligase Brl2 and three subunits of the Set1/compass complex ( Set1C ) , Swd1 , Swd2 and Spf1 ( Fig 2D , S2 Fig , Table 1 ) . A protein interaction network analysis regrouping novel and previously known factors showed a high degree of connectivity ( see below ) . The fluorescence microscopy and multiplex PCR analyses for known and newly identified factors ( S2 Fig ) were confirmed by Southern blot analysis of mat1 content with the DdeI restriction enzyme ( S3 Fig ) . Southern blots can be used to detect the imprint at mat1 and to determine whether rearrangements have occurred in the mating-type region . A DSB results from breakage at the fragile imprint site during DNA preparation [40] . As mentioned in the Introduction , the mating-type switching factors can be subdivided into three groups by Southern blot analysis of mutants , reflecting the molecular function of each factor . Class Ia is required for DSB formation at mat1 , Class Ib is involved in donor selection for mating-type switching or other steps in the use of the break , and Class II is required for processing the gene conversion intermediates [40] . The 23 strains selected for analysis were assayed by Southern blot ( Fig 3 ) . The analysis confirmed previous conclusions in the case of known factors , for instance swi3Δ abolished DSB , and rad57Δ , msh2Δ , msh3Δ , rad16Δ , and pxd1Δ caused high frequencies of rearrangements of the h+N type as expected for resolution-defective mutants [40] . The newly identified mating-type switching genes were assigned to Class Ib; elp6Δ , swd1Δ , spf1Δ , brl2Δ , pof3Δ and ckb1Δ strains were in that category together with swi2Δ , cbp1Δ , swi6Δ , sir2Δ and deletions of the Clr4 methyltransferase complex ( CLRC ) , clr4Δ , rik1Δ and raf1Δ , or Snf/Hdac-containing repressor complex ( SHREC ) , clr1Δ , clr2Δ and clr3Δ , subunits . As previously noticed for the swi6Δ mutant [51] a rearrangement producing an 8 . 2 kb HindIII fragment could be detected in several Class Ib mutants . The rearrangement could be a mat3:1 circle or a duplication of the mating-type region creating a mat3:1 cassette: mat1-L-mat2-P-K-mat3:1-L-mat2-P-K-mat3-M . The mat3:1 cassette would not be amplified by the primers used to detect mat1 content by PCR . The swd2Δ strain differed from the other mutants by showing a 9 . 9 kb HindIII fragment hybridizing to the mat1 probe , however reconstruction of the strain produced a Class Ib mutant lacking this additional band and the reconstructed deletion allele was used in further analyses . In the h09 mating-type region , the contents of the silent cassettes are swapped to mat2-M mat3-P ( Fig 4 , S4 Fig ) . This arrangement results in inefficient heterologous switching and in a mat1 content biased towards mat1-M [3] . How mutations affect this bias provides insights into the directionality of mating-type switching . For example , deletion of swi6 biases the mat1 content towards mat1-P in h09 cells , and towards mat1-M in h90 cells , consistent in both cases with preferred selection of mat3 as a donor [3] . This likely reflects a preferential use of the SRE3 recombination enhancer in swi6Δ cells [23] . We created h09 strains to test whether the newly identified Class Ib factors contribute to mating-type switching in the same or similar way as Swi6 . Each deletion mutant was crossed with the h09 strain PG4048 . After the selection of recombinants , we analyzed four independent colonies of each h09 deletion strain by multiplex PCR for mat1 content . As a control , colonies originating from spores of self-mating PG4048 cells were analyzed . As expected , PG4048 contained a greater proportion of M cells than P cells ( a mean of 86% M cells; Fig 4 ) . All mutations tested affected this ratio . The effects varied . During these analyses , we created a fresh deletion of swd2 to eliminate the rearrangement detected in the Bioneer mutant ( 9 . 9 kb band in Fig 3 ) , as mentioned above , and confirmed that the observed directionality defects were not a result of the rearrangement . A first group of factors comprised Swi6 , CLRC subunits ( Clr4 , Rik1 and Raf1 ) and the histone deacetylase Sir2 ( Fig 4 ) . Mutations affecting CLRC or Sir2 produced nearly identical values ( around 70% P cells ) , similar to the loss of Swi6 ( 77% P cells ) . These mutants were also very similar to each other in the h90 background ( around 20% P cells; Fig 2D ) . In the mating-type switching process , CLRC and its catalytic Clr4 subunit are believed to work by catalyzing the methylation of H3K9 in the mat2-mat3 heterochromatic domain and creating binding sites for Swi6 [24 , 52–55] . The role of Sir2 in the process might be to remove acetyl groups from H3K9 [46] , thus facilitating heterochromatin formation [46 , 56–58] . Indeed , the methylation of H3K9 and Swi6 association are reduced several fold in the mating-type region of sir2Δ cells [46] . This is consistent with Sir2 acting upstream of CLRC and Swi6 through H3K9 deacetylation , without excluding that other actions of Sir2 , e . g H3K4 deacetylation [59] , might also be relevant to directionality . The second group of mutants affecting mating-type switching in h09 cells were deletions of the genes for the ubiquitin E3 ligase Brl2 , the Set1C subunits Swd1 , Swd2 and Spf1 , and the Elongator subunit Elp6 ( Fig 4 ) . These mutants resulted in a consistent increase in mat1-P content in h09 cells , from ~20% in wild-type background to ~35% in each of the three mutants . While not as pronounced as for Swi6 or CLRC mutants , the increase occurred to the same degree in all four isolates examined in each case . These mutants were unexpected because it has been reported that set1Δ has no effect on mating-type switching [60] . We examined each component of Set1C ( Set1 , Swd1 , Swd2 , Swd3 , Spf1 , Ash2 , Shg1 and Sdc1 [61] ) by iodine staining and multiplex PCR of mutants ( Fig 5A–5D ) . Consistent with the previous report [60] , deletion of set1 or other subunit genes showed little effect on iodine staining of either h90 or h09 colonies ( Fig 5A and 5B ) . However , monitoring mat1 content clearly showed that individual gene deletions biased switching toward mat3-P in h09 cells and resulted in a correlated increased use of mat3-M in h90 cells for six Set1C subunits ( Fig 5C and 5D ) . This trend is similar to mutations in the H3K9 methylation pathway , although not to the same amplitude ( Fig 4 ) . Interestingly , the iodine staining level of set1Δ colonies differed between h90 and h09 in spite of similar cell-type ratios ( 39% P cells in h90 and 36% in h09 ) . Due to a different switching pattern , P and M cells might be less evenly mixed in h09 colonies compared with h90 . This would lead to less efficient mating and spore formation in h09 even though the cell-type bias is only a little more pronounced than in h90 . To investigate the functionality of each SRE element in set1Δ cells , we analyzed directionality in SRE element mutants ( Fig 5E–5H ) . Deletion of set1+ did not significantly affect mat1 content in 2×SRE2 cells ( where the SRE3 element is replaced with SRE2 ) ( Fig 5E ) or in SRE3Δ cells ( Fig 5G ) . However , in populations of 2×SRE3 cells ( where the SRE2 element is replaced with SRE3 ) deletion of set1 caused a small increase in M cells ( 31% P cells in 2×SRE3 set1Δ compared with 36% P cells in 2×SRE3 set1+ ) ( Fig 5F ) . In the case of SRE2Δ , donor choice was more strongly biased towards mat3-M in set1Δ cells ( 6% P cells ) than in set1+ cells ( 13% P cells ) ( Fig 5H ) . This suggests that Set1C normally inhibits the choice of mat3-M-SRE3 in M cells . It has similarly been observed that deletion of swi6 causes virtually no change in donor choice in the 2×SRE2 and SRE3Δ backgrounds , where SRE2 keeps being used , but decreases use of SRE3 in SRE2Δ cells ( 5% P cells in SRE2Δ swi6Δ compared with 16% P cells in SRE2Δ swi6+ ) [23] . Loss of the Brl2 ubiquitin ligase resulted in phenotypes similar to mutations compromising Set1C ( Figs 4 and 5 ) . Thus , like Swi6 and CLRC , Set1C and Brl2 appear to favor use of the SRE2 recombination enhancer over SRE3 when both enhancers are present , perhaps by inhibiting the use of SRE3 , to result in balanced switching . A third group of mutants displayed a variegated phenotype ( clr1Δ , clr2Δ , clr3Δ , ckb1Δ , pof3Δ and cbp1Δ; Fig 4 ) , with the proportion of P cells varying between independent cultures . In some isolates , the proportion of P cells was nearly wild-type whereas in others it was similar to the swi6Δ mutant . These phenotypes were not caused by rearrangements in the mating-type region ( S4 Fig ) . Clr1 , 2 and 3 are subunits of SHREC , the Snf2-histone deacetylase repressor complex [62] . Clr3 participates in the recruitment of Clr4 to the mating-type region , but in its absence a Clr3-independent , RNAi-dependent pathway accomplishes this function to some extent [63] . Clr3 localizes to three regions in the mating-type region , which are close to mat2-P ( REII ) , cenH and mat3-M ( REIII ) , respectively [62 , 64] . In the cenH region , the heterochromatin platform is likely established by RNAi-mechanisms; on the other hand , at the REIII site , it is established by an RNAi-independent mechanism [63 , 65 , 66] . The clonal variations observed in clr1 , clr2 and clr3 h09 mutants might reflect these distinct pathways of heterochromatin establishment similar to position effect variegation [67] . Heterochromatin would be partially formed and inherited in some clonal populations of the mutant strains but not in others , leading to populations differentially proficient for switching . It is also known that the CK2-dependent phosphorylation of Swi6 mediates Clr3 recruitment to centromeric regions [68] , possibly accounting for the Ckb1 defects observed here . To further address the mechanisms of mating-type switching , we analyzed the protein-protein interaction network linking newly identified and previously known mating-type switching factors in the STRING database ( Table 1 ) [69] . The obtained interaction network correlates strongly with categories established by Southern blotting and phenotypic classification ( Fig 6A ) . Among the newly identified factors , six subunits of Set1C and Brl2 are connected to Class Ib factors . While Pof3 , Ckb1 and Elp6 show no direct interaction with Class Ib factors in the STRING analysis ( Fig 6A ) , several studies have reported that Pof3 plays a role in heterochromatin silencing [70–72] and Ckb1 phosphorylates Swi6 [68] . Elp6 is an orthologue of a part of the six-subunit Elongator complex ( Elp1-6 ) in S . cerevisiae [73 , 74] . Elp3 also passed the initial screening ( S2 Table ) , however the other subunits of Elongator complex did not . The main cellular function of Elongator is thought to be in tRNA modification , but Elongator has also been proposed to acetylate histones [75–77] . We tested the hypothesis that some factors might act through Swi6 by performing an epistasis analysis . Double mutants combining swi6Δ with each candidate gene deletion were constructed and four independent colonies were analyzed by multiplex PCR for each of them ( Fig 6B and 6C ) . The mating-type switching phenotypes of the swi6Δ clr4Δ and swi6Δ sir2Δ double mutants were quite similar to the single deletions of swi6 , clr4 or sir2 ( Figs 2D , 4 , 6B and 6C ) . This is consistent with CLRC and Sir2 being required for heterochromatin establishment and Swi6 recruitment at the mat locus ( Fig 7 ) . Double mutants combining swi6Δ with swd1Δ , lacking a Set1C subunit , or clr3Δ , lacking a SHREC subunit , also displayed a phenotype quite similar to the swi6Δ single mutant ( Fig 6B and 6C ) . This indicates that Set1C and SHREC work in the Swi6 pathway . A previous study has shown that set1Δ does not affect Swi6 localization or silencing of a ura4+ marker gene at cenH in the mating-type region [60] but other assays have found that Set1C subunits participate in the repression of heterochromatic loci including the mating-type region [78] . This latter effect might be related to the occurrence of switching defects in Set1C mutants . Set1C may control Swi6 localization in a site-specific manner such as at SRE3 ( Fig 5 ) . Donor preference in swi6Δ ckb1Δ cells indicated that swi6Δ is also epistatic to ckb1Δ even though phenotypes could only be assigned in the h90 background due to apparently high rates of rearrangement in the h09 background . Nevertheless , the data suggest that Swi6 phosphorylation by Ckb1 [68] is important for switching directionality controlled by Swi6 . More complex epistatic relationships were observed for the remaining mutants , brl2Δ , pof3Δ , cbp1Δ , elp6Δ , and swi2Δ , when these mutations were combined with swi6Δ ( Fig 6B and 6C ) . The predominant mating-type had a tendency to vary between isolates , particularly with the h09 mating-type region , and rearrangements occurred . Double mutants combining brl2Δ and swi6Δ showed a mat1-M bias apparently even more pronounced than for the swi6Δ single mutant in h90 , while in h09 two swi6Δ brl2Δ isolates were similar to swi6Δ and two had more balanced mat1 contents . Populations of h90 swi6Δ pof3Δ cells showed biases similar to the h90 swi6Δ single mutant for three isolates , whereas the fourth isolate was biased towards P cells rather than M cells . Populations of h09 swi6Δ pof3Δ cells had varied ratios of P and M cells , and two isolates were rearranged . The switching bias for swi6Δ cbp1Δ was similar to swi6Δ with the h90 mating-type region ( Fig 6B ) , but two strains in four differed from swi6Δ with the h09 mating-type region ( Fig 6C ) . This phenotype may be caused by Swi2 expression level , which is controlled by Cbp1 [26 , 31] . Rearrangements in h90 swi6Δ elp6Δ and h09 swi6Δ elp6Δ mutants precluded analysis . These observations suggest that both Swi6-dependent and -independent pathways control switching directionality by Brl2 , Pof3 and Cbp1 . Finally , swi2Δ swi6Δ double mutants differed from swi6Δ in both h90 and h09 . It has been reported that Swi2 can localize to SRE3 in the absence of Swi6 [22 , 23 , 27] . The more balanced cell populations in swi6Δ swi2Δ mutants are probably caused by loss of Swi6-independent function of Swi2 . In summary , our observations suggest that the factors , Clr4 , Sir2 , Swd1 and Clr3 probably work in the same pathway as Swi6 , but Brl2 , Pof3 , Cbp1 and Swi2 have an effect on donor selection through Swi6-dependent and -independent mechanisms . Frequent DNA rearrangements in Class Ib mutants ( Fig 3 ) and in double mutants with swi6Δ indicate that histone modifications not only direct donor choice , but also facilitate resolution steps or prevent unequal sister chromatid exchanges between cassettes . We propose a model summarizing how each factor identified in this study might participate in the donor selection mechanism ( Fig 7 ) . In this model , the heterochromatin structure in M cells favors the cassette adjacent to SRE2 as a donor while structural changes in mutants and in P cells favor SRE3 [23] . In addition , chromatin structure prevents selection of the cassette adjacent to SRE3 in M cell ( Fig 7A ) . It has been reported that SRE2 can facilitate donor choice efficiently not only in M cells but also in P cells whereas SRE3 is more active in P cells than in M cells [23] . These data indicate that the inhibition of donor choice does not affect SRE2 . Mating-type switching is initiated by a site-specific imprint during replication . In the replisome , Swi1-Swi3 , Pol1 and Mrc1 are required for the imprint [16 , 40] . One of the novel switch factors identified here is the ubiquitin ligase component Pof3 . Pof3 interacts with the replisome , in an Mrc1- and Mcl1-dependent manner [71 , 79] . However , rather than participating in imprint formation , we found that Pof3 affects donor selection . Pof3 is also required for heterochromatic silencing near mat3 [70] . We speculate that both effects are brought about by the Pof3-mediated degradation of replisome components [80] or of Ams2 , a cell cycle-regulated transcription factor for histone genes [81] that also mediates long range chromosomal interactions [82] and interacts with Raf1 , a component of CLRC [83] . Thus , Pof3 would couple the deposition of new histones and their modification by CLRC . During S phase , partly as a result of new histones deposition onto replicated DNA , Swi6 and H3K9me2 levels decrease at silenced loci [84 , 85] . A wave of H3K9 acetylation , observed in other organisms in front of the replication fork [86] , might further weaken heterochromatin . The current search expands on previous work to show that enzymatic complexes required for the restoration of heterochromatin , both NAD+-dependent and -independent HDACs and CLRC , are necessary for donor selection . Remarkably , lack of Sir2 or of a CLRC subunit phenocopied the swi6 deletion in both h90 and h09 cells ( Figs 2D and 4 ) while the loss of SHREC components resulted in variegated phenotypes . We take these differences as reflecting the different substrate specificities and recruitment mechanisms of the two HDACs to heterochromatic regions [57 , 87] . In both cases , our epistasis analysis with the Swi6 mutant points to defects in Swi6 recruitment . Our search also identified Cbp1 and CK2 , both of which are thought to recruit Clr3 to heterochromatin regions [64 , 68] . Cbp1 also controls expression of the swi2 gene and the cell-type specific protein isoform it produces , together with the M-specific protein Mc [26 , 31] . A novel and intriguing outcome of our study is that multiple subunits of Set1C and the E3 ubiquitin ligase Brl2 are required for accurate donor selection . Set1C catalyzes the methylation of H3K4 . A role in heterochromatin appears paradoxical , given that methylated H3K4 is strongly associated with expressed genes . One possibility is that Set1C regulates the expression of Swi2 , central effector of switching directionality , or of other switching factors . RNA profiling analysis has revealed that switching genes are expressed to similar levels in wild type and individual Set1C mutants [87] , however more subtle effects such as shifts in transcription initiation have not been ruled out . In addition , S . pombe Set1C appears directly required for silencing at various locations [78] . In the mating-type region , all subunits except for Shg1 are required for silencing of the cenH repeat . Evidence has also emerged for roles in meiotic recombination [88 , 89] even though the effects of H3K4 methylation on recombination have been hard to unravel due to the prevalence of that modification genome-wide [90] . Here , we favor a simple model where Set1C affects donor choice directly . This could be through local , possibly temporally restricted methylation of H3K4 at SRE3 that would either act as such or by preventing H3K4 acetylation . Brl2 is part of HULC that catalyzes the ubiquitylation of histone 2B ( H2Bub ) [61 , 91 , 92] . H2Bub stabilizes the interaction of Set1 with chromatin in vitro [93] . The single brl2Δ deletion affected mat1 content similar to the deletion of Set1C components ( Figs 2 , 4 and 5 ) . These connections and phenotypic similarities indicated that Set1C and HULC might co-operate to choose a correct donor . ( Figs 4 and 6B ) . However , the phenotypes of two in four independent colonies of h09 swi6Δ brl2Δ strain differed from h09 swi6Δ swd1Δ ( Fig 6C ) . Brl2 is also known to interact with Nse5 , which is a part of the structural maintenance of chromosome 5/6 ( Smc5-6 ) holocomplex [94 , 95] . It may have multiple functions in mating-type switching . Following the cell-type specific deposition of Swi6 , Swi2-Swi5 localizes to SRE2 by interaction with Swi6 in M cells [22 , 23 , 27] . Together with the inhibited use of SRE3 , this regulated recruitment of Swi2-Swi5 effectively directs the Rad51 strand-exchange protein to initiate homologous recombination at the proper cassette , providing an increasingly well understood model for the effects of chromatin structure on recombination .
S . pombe strains were generated and propagated according to standard protocols [96] . They were manipulated with a Singer RoToR plate-handling robot ( Singer Instruments ) for high throughput screens . To test for mating-type switching defects by fluorescence analysis , a query strain ( PG4045: h90 ( Blp1 ) ::LEU2 leu1::ura4+-[mfm3p-YFP]-[map2p-CFP] ura4-D18 ade6-M216 was mated to the Bioneer gene deletion library ( h+ leu1-32 ura4-D18 ade6-210 or 216 , ORFΔ::kanMX4 ) . Mating was performed on SPA plates supplemented with 200 mg/l leucine , 100 mg/l uracil , and 100 mg/l adenine . Cells were allowed to mate and sporulate at 30°C for two days . The mating plates were then moved to 42°C for three days to eliminate vegetative cells . Following heat treatment , spores were transferred onto YES plates with 100 mg/l G418 and allowed to germinate and divide for three days at 30°C . To select for h90 progeny , cells were then transferred from the YES plates to MSA plates with 100 mg/l G418 and 100 mg/l adenine and grown for a further three days at 30°C . This scheme selects for ( Blp1 ) ::LEU2 , tightly linked to the h90 region , for ura4+ , tightly linked to the fluorescent reporters , and for kanMX4 that marks each ORF deletion . Single colonies were isolated from bulk recombinants by streaking cells onto MSA plates containing G418 and adenine , incubated for three days at 30°C . The sdc1 deletion strain ( h+ sdc1::Kanr ade6-M21 ? leu1-32 ura4-D18 ) was obtained from the National BioResource Project ( NBRP ID: FY23769 , strain name: P1-1G ) . Single colonies of recombinants obtained as described above were inoculated into 50 μL MSA medium supplemented with 100 mg/l adenine and grown for two days at 30°C in 96 well plates . Cells were then diluted 24 times into MSA medium with adenine and grown for ~20 hr at 30°C . Sixty μl cell suspensions diluted 15 times with MSA medium with adenine were transferred to 384 well microplates with clear bottom ( CellCarrier-384 ultra , Perkin Elmer ) . The fluorescence of cells was measured using an Opera high-content screening microscope ( Perkin Elmer ) . The following settings were used [Filter sets: Camera 1: 475/50 for CFP signals , Camera 2: 540/75 for YFP signals , Camera 3: 690/50 for bright field , Light source: 405/488/635] . Twelve images were taken for each well , for a total of ~200 cells per strain . The proportion of P cells was then calculated using the Acapella software program ( PerkinElmer ) . Selected strains were imaged again using a Delta Vision Elite microscope ( GE Healthcare ) . Cells were plated onto MSA medium supplemented with 100 mg/l adenine , allowed to form single colonies , and exposed to iodine vapors . S . pombe cells were propagated in 2 mL liquid YES cultures at 30°C to saturation . Genomic DNA was prepared from wild-type and mutant cells as described [97] . The genomic DNA concentration was measured using QuantiFluor One dsDNA Dye System ( Promega ) and 4 μL genomic DNA ( 1 . 25–5 . 0 ng/μL ) in TE was added to 16 μL PCR reaction reagent ( total 20 μL ) to perform multiplex PCR to determine the genetic content of the mat1 or mat3 locus . The primers used were FAM-MT1 ( 5’-AAATAGTGGGTTAGCCGTGAAAGG-3’ ) at 400 nM , MP1 ( 5’-ATCTATCAGGAGATTGGGCAGGTG-3’ ) at 200 nM and MM1 ( 5’-GGGAACCCGCTGATAATTCTTGG-3’ ) at 200 nM ( S3 Fig ) . The 5’ end of FAM-MT1 and FAM-MT3 were modified with 6-carboxyfluorescein ( FAM ) . To reduce non-specific PCR products , 400 nM heat-stable RecA protein from a thermophilic bacterium , Thermus thermophiles , and 400 μM ATP were included in the PCR reaction buffer ( 10 mM Tris-HCl pH 8 . 3 , 50 mM KCl , 2 . 5 mM MgCl2 ) [98] . The following amplification program was used: 2 min at 94°C—27 x [30 s at 94°C—30 s at 55°C—1 min at 72°C]—5 min at 72°C . PCR fragments corresponding to mat1-P and mat1-M alleles were resolved on 5% polyacrylamide gels . Fluorescence was detected and quantified using Typhoon FLA9500 ( GE Healthcare ) and ImageQuant ( GE Healthcare ) . Each gene knockout in the Bioneer collection contains ‘up-tag’ and ‘down-tag’ sequences that provide a unique barcode for each knockout . To confirm the identity of the mutants identified in our screen , we amplified the KanMX4 region using the U1 primer ( 5’-CGCTCCCGCCTTACTTCGCA-3’ ) and D1 primer ( 5’- TTGCGTTGCGTAGGGGGGAT ) . The PCR products were then sequenced using cpn1 ( 5'-CGTCTGTGAGGGGAGCGTTT-3' ) to read the up-tag and cpc300 ( 5'-AGACCGATACCAGGATCTTGCC-3' ) to read the down-tag . The results were compared with the barcode list . S . pombe cells were propagated in 16 mL liquid cultures ( YES ) at 30°C to pre-saturation and genomic DNA for Southern blots was prepared as described above . Genomic DNA was digested with HindIII to classify the mating-type switching defective genes according to DSB formation or presence of rearrangements in the mating-type region ( Fig 3 ) , or with DdeI to assay mat1 content ( S3 Fig ) . The digested samples were electrophoresed in 0 . 7% agarose gels . The probe to analyze mat1 content was a PCR product made with GTO-1369 ( 5’- GAGCCTACTGTTAATATAATAACATTATG-3’ ) and GTO-1370 ( 5’-CCTTCAACTACTCTCTCTTCTTTTCCTACCC-3’ ) , corresponding to the centromere-proximal DdeI-NsiI fragment [23] . The probes to classify the mutants were 10 . 4 kb HindIII fragments containing mat1-P or mat1-M . | Effects of chromatin structure on recombination can be studied in the fission yeast S . pombe where two heterochromatic loci , mat2 and mat3 , are chosen in a cell-type specific manner to convert the expressed mat1 locus and switch the yeast mating-type . The system has previously revealed the determining role of heterochromatin , histone H3K9 methylation and HP1 family protein Swi6 , in donor selection . Here , we find that other chromatin modifiers and protein complexes , including components of the histone H3K4 methyltransferase complex Set1C , the histone H2B ubiquitin ligase HULC and Elongator , also participate in donor selection . Our findings open up new research paths to study mating-type switching in fission yeast and the roles of these complexes in recombination . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
]
| [
"molecular",
"probe",
"techniques",
"dna-binding",
"proteins",
"organisms",
"fungi",
"model",
"organisms",
"experimental",
"organism",
"systems",
"molecular",
"biology",
"techniques",
"epigenetics",
"gel",
"electrophoresis",
"chromatin",
"schizosaccharomyces",
"heterochromatin",
"research",
"and",
"analysis",
"methods",
"electrophoretic",
"techniques",
"electrophoretic",
"blotting",
"artificial",
"gene",
"amplification",
"and",
"extension",
"chromosome",
"biology",
"proteins",
"gene",
"expression",
"schizosaccharomyces",
"pombe",
"chemistry",
"histones",
"molecular",
"biology",
"genetic",
"loci",
"yeast",
"biochemistry",
"southern",
"blot",
"chemical",
"elements",
"iodine",
"eukaryota",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"physical",
"sciences",
"polymerase",
"chain",
"reaction"
]
| 2018 | New insights into donor directionality of mating-type switching in Schizosaccharomyces pombe |
Countries have traditionally been split into two major groups: developed or industrialized ( “the North” ) and developing or underdeveloped ( “the South” ) . Several authors and organizations have challenged this classification to recognize countries that have reached an intermediate stage of social and economic development . As proposed by Morel and collaborators in 2005 , the concept of Innovative Developing Countries ( IDCs ) defines a group of nations with impactful scientific programs . Here , IDCs are reexamined by a variety of metrics to highlight their role in health innovation through research and development ( R&D ) programs on neglected tropical diseases ( NTDs ) that also positively impact epidemic preparedness . To address the global changes due to expanding globalization we updated the original indicator of the number of USPTO patents deposited by individual countries per GDP and per capita to the number of international patents applications , related to applicant residence and deposited under the Patent Cooperation Treaty ( PCT ) per GNI ( or GDP ) and per capita . A comparison of the originally described ranking of top innovative countries to those in the present study revealed new members that updated the list of IDCs and showed a prominent role now played by China . Analyzing scientific publications in international journals since the introduction of the IDC concept in 2005 we found that IDCs do prioritize Neglected Tropical Diseases ( NTDs ) as an area of research . Finally we investigated the role of IDCs in two major public health emergencies between 2012 and 2016 , the outbreaks of Ebola in West Africa and Zika in South America . An analysis of the co-authorship country networks demonstrated an important role for IDC infrastructure and personnel in the prevention and control of these epidemics . Different techniques can be used to evaluate and measure innovative performance of countries . Country rankings published by traditional indexes , such as the Bloomberg Innovation Index ( BII ) and the Global Innovation Index ( GII ) , only include high income economies among the top 20 performers . This is in sharp contrast to our approach , which identified 8-9 IDCs among the first 25 with China occupying the top position . Through an analysis of the pros and cons of the different methodologies , the IDC concept challenges more conventional approaches to address and estimate the innovative capacity of countries .
Countries have traditionally been divided into two broad categories according to their capacity to innovate: leaders , which have infrastructure along with the human and financial resources for the production and management of innovation; and followers , those that do not have the capacity for innovation and face the challenge to reproduce and absorb technologies from the leading countries through technology transfer processes . From analyzing this issue the Director General of the Council for Scientific and Industrial Research of India , R . A . Mashelkar ( http://www . mashelkar . com/ ) , proposed in 2003 an organization of countries in a 2x2 table that distributes them according to relative economic strength and autochthonous ( or “indigenous” as he called it ) science and technology ( S&T ) capacity Fig 1 [1]: Mashelkar pointed out that the positions of the countries in these four quadrants should not be seen as static , citing the example of South Korea which in 1996 left the lower-right quadrant to become an OECD country . To place the framework presented by Mashelkar into a quantitative basis , Morel and collaborators used the number of patents granted in the United States as a measure of the innovative capacity of a country when at least one inventor was from that country , which was correlated to the economic strength of the country based on economic and demographic criteria [2] . They designed a new indicator ( number of patents normalized per GDP and per capita for each country ) that made it possible to expand from Mashelkar’s vision of quadrants into rankings . Through this analysis , several countries that were allocated to the lower-right quadrant , which were not considered to have high income economies according to World Bank definitions , appeared among the top performers . The findings became the basis for the development of the concept of Innovative Developing Countries ( IDCs ) by Morel and collaborators that included Mashelkar [2 , 3] . After a brainstorming meeting at the Bellagio Center , Rockefeller Foundation , 10-13 May 2004 , it was agreed that: This perception that the term “developing countries” could not account for the diversity of countries based on innovation capacity was pioneering and visionary . For example , the 2016 edition of the World Development Indicators ( WDI ) published by the World Bank [4] , no longer uses the terms “developed” and “developing” countries as analytical categories since recent evidence shows that dividing countries into just two groups does not reflect reality [5 , 6]; instead , it adopts four categories ( high income , upper-middle , lower-middle and low income economies ) . Vollmer and collaborators have recognized that today there is an emergence of “three human development clubs” , which is in contrast to the conceptual viewpoint of the 60s , when the world was clearly divided into industrialized and developing nations [7] . In the past , it was straightforward to split the countries into two groups when using indicators such as child mortality and fertility rate; today these indicators spread them along a continuum . Therefore , allocating a country into a single category does not necessarily contribute to understanding the present global realities . In the area of health , it is particularly important to investigate how the research , development and innovation infrastructure built over the years by IDCs has been used to address neglected tropical diseases; furthermore whether this base could be mobilized for coping with new health challenges , such as specific epidemic situations . This paper aims to revisit the concept of IDCs , thirteen years after its proposed use to define a new category of countries , and its relevance to health innovaton . Here , we address three topics: ( i ) review and update of the original country innovation ranking; ( ii ) relevance of IDCs in health research , development and innovation , particularly in relation to NTDs; ( iii ) role of IDCs that had invested in NTDs during the Ebola and Zika epidemics control and preparedness .
The innovation index proposed by Morel et al [2 , 3] was originally obtained dividing the total number of patents filed in the United States by each country by their respective GDP per capita ( S1 Table , Supporting Information ) . In the present study we refined this approach by dividing the total number of patents filed by each country under the Patent Cooperation Treaty ( PCT ) by their respective GNI or GDP per capita . Countries that were not considered high income economies by the World Bank ( GNI per capita higher than US$ 12 , 476 in 2015 ) but in our analysis ranked among the top 25 innovative nations , fit the IDC category . The search for patents filed in 2015 under the Patent Cooperation Treaty ( PCT ) was performed in July 2017 using Patentscope from the World International Property Organization ( WIPO ) . Using the “field combination” tool , the query included the following fields: “WIPO publication number” ( WO* ) , “Applicant residence” ( XX* , where XX denoted country codes ) and “Publication date” ( 01 . 01 . 2015 to 31 . 12 . 2015 ) . To retrieve patents related to medicines , the additional field “International Class” was used to specify the patent subclass A61K* which according to International Patent Classification ( IPC ) refers to preparations for medical , dental or toilet purposes [8] . Population , Gross National Income ( GNI ) per capita and Gross Domestic Product ( GDP ) per capita of 2015 were obtained from the World Development Indicators DataBank [9] . Scientific publications addressing at least one of the 17 NTDs listed by the World Health Organization were used as a representation of country focus on IDCs common health burdens . Publications on NTDs were retrieved as raw data files from the Web of Science Core Collection ( WoS ) database . The total number of articles and articles on NTDs published by a given country during the 2005-2017 period were retrieved using the following profiles in “Advanced Search Mode”: ( i ) Total articles: cu = “name of country”; ( ii ) Articles mentioning at least one of of the NTDs in the abstract: cu = “name of country” AND ts = ( “buruli ulcer” OR “Chagas disease” OR “trypanosoma cruzi” OR dengue OR chikungunya OR dracunculiasis OR echinococcosis OR “food borne trematobiases” OR “human african trypanosomiasis” OR “sleeping sickness” OR leishman* OR leprosy OR filariasis OR onchocerciasis OR “river blindness” OR rabies OR schistosomiasis OR helminthiasis OR taeniasis OR cysticercosis OR trachoma OR yaws OR “endemic treponomatoses” ) . Countries were ranked according the decreasing percentage of articles mentioning at least one NTD in relation to the total number of articles published during the considered period . Social network analysis ( SNA ) of scientific collaborations was pioneered by Newman [10 , 11] . We used his approach as previously described [12–14] to investigate the role of IDCs in recent global epidemics . Network analysis is a theoretical approach that employs a set of techniques used to understand and quantify the relationship between members of a network ( nodes ) , which can be individuals , institutions , countries etc . [15] . By analyzing and quantifying the social structure of a network that is embedded in its nodes and connections , it is possible to assess different perspectives on the importance of individual nodes . In this work , we used the ‘betweeness centrality’ indicator to identify key countries that are frequently on the shortest paths between other countries , acting as intermediaries of information [16] . Articles published during the peak of the Ebola ( n = 1 , 461 articles in 2015 ) and Zika ( n = 1 , 477 articles in 2016 ) epidemics were retrieved from the Web of Science Core Collection ( WoS ) database as described above . The search query was directed to the title of the papers using the terms “Zika” or “Ebola” , accordingly . The unit of analysis ( the nodes in the network ) consisted of the countries where the authors were based at the time of publication , according to their affiliation data registered in each article . Only published or in press articles were included in the analysis . Publication data was imported into the data/text mining software VantagePoint ( Search Technology Inc . ) . After processing , data was formatted into adjacency matrixes [15] by VantagePoint to map co-authorship relationships between countries . Matrixes were imported into the open-source software Gephi [17] for network visualization and calculation of centrality metrics . For the spatial visualization of the international collaboration networks , country affiliation data were manually geocoded and processed using the “GeoLayout” and “Map of Countries” plugins available within Gephi . In these networks , nodes represent countries , and two or more countries were connected if their members shared the authorship of one or more papers . As co-authorship requires reciprocal cooperation among the participants , all connections have been considered as non-directional .
In our present calculations we substituted the number of patent applications deposited at the USPTO , used in the original study [2] , for the number of international patent applications deposited under the Patent Cooperation Treaty ( PCT ) , related to applicant residence . This change addressed the global changes due to the continual progression of globalization [18] . In addition , the survey of the international markets instead of just the US market provided a broader view of the intent of patent protection of technologies . We also explored the characteristics of our metrics in two additional ways . Firstly , using both the Gross National Income ( GNI ) per capita and the Gross Domestic Product ( GDP ) per capita . GDP measures total output produced , based on location , focused on domestic production and represents the strength of a country’s economy , while GNI measures the total income received , based on ownership , focused on income generated by citizens and represents economic strength of country’s nationals . We found that interchanging GDP for GNI did not significantly alter the final results and rankings ( Supporting Information S2 Table ) . As we are particularly interested in the country residence of the patent applicant and in the income generated by the residents of that country , we adopted GNI per capita as the default . Secondly , using specific patent classes in the calculations , therefore selecting and delineating areas of technology or products of interest , allowed the detection of those countries that are more active in a given “product space” [19] . Table 1 displays the comparison of the 25 top innovative countries ranked as described in the original 2005 paper ( first country column ) and in this study ( third country column ) . The second country column is an update of the first one using USPTO 2015 patent information . To demonstrate the potential and flexibility of our approach , the last column displays the country ranking and the leadership of India when only patents related to medicines are used in the calculations ( patent subclass A61K see Methods ) [20] ) . Comparing the 2005 rank ( first country column ) with the results of the present study ( third country column ) , some points are worth mentioning: In their original article Morel and collaborators addressed the innovative capacity of IDCs in health by analyzing patents that included the words “drug” , “vaccine” , or “pharmaceutical” in the abstract . It was observed that the rate of patenting was relatively constant during the first half of the 1990s , but accelerated dramatically after 1996 [2] . One important question however was not addressed at that stage: is the innovative capacity of IDCs being used to address health issues that are particularly relevant for their own populations or is it ‘market-driven’ , and aimed at competing in the global markets ? Or , as one participant said at the Bellagio meeting mentioned above: “Are the IDCs investing in their own health priorities or trying to develop the next blockbuster drug ? ” Mahoney , Morel and collaborators [2 , 21–24] identified six determinants , or components , of health technology innovation: ( i ) Development and expansion of national health delivery systems , including an attractive , domestic , private-sector market for health products; ( ii ) Development of manufacturing capability for health products; ( iii ) The drug and vaccine regulatory system; ( iv ) The IP regulatory system; ( v ) Development of R&D capability by the public and private sectors; ( vi ) Development of international trade systems for health products , including global procurement funds . Because these innovation components are dynamically linked , successfully developing and introducing new technologies requires concerted attention to each of the six components [22 , 23] . Using this framework , Morel et al analyzed the progress of developing countries in innovation capability and identified three intermediate stages before reaching a similar development level of industrialized nations [2] . NTDs represent an enormous health burden for developing countries which only recently have been recognized as a global challenge meriting global efforts and commitments from public and private sectors [25 , 26] . In order to investigate whether NTDs represents a priority for the research and technological development agenda of the top innovative countries ranked by our approach we focused on countries’ performances in relation to the fifth determinant listed above , the development of R&D capability by the public and private sector . For this purpose we analyzed the proportion of publications addressing at least one of the 17 NTDs listed by the World Health Organization [27] . Fig 2 ranks the top 30 innovative countries according to the proportion of their scientific publications addressing at least one NTD in the abstract . On the average , 0 . 53% of their publications addresses at least one NTD . The following countries publish above this average ( IDCs in italics ) : Brazil ( 2 . 17% ) , Thailand ( 1 . 57% ) , Argentina ( 1 . 48% ) , Indonesia ( 1 . 24% ) , India ( 0 . 92% ) , Mexico ( 0 . 91% ) , Singapore ( 0 . 70% ) , Switzerland ( 0 . 68% ) and Malaysia ( 0 . 61% ) . Two serious epidemics hit the developing world in 2014-2016 , leading the World Health Organization to issue alerts of Public Health Emergency of International Concern ( PHEIC ) on two occasions: Ebola in West Africa in 2014 [28] and Zika in South America in 2016 [29] . Fig 3 shows the evolution of publications of scientific articles having “Ebola” or “Zika” in the title from January 2012 to December 2016 . We investigated the role of IDCs in these major public health emergencies analyzing coauthorship networks of scientific publications as previously described [12–14] . The analysis of co-authorship networks through social network analysis ( SNA ) has been applied previously to understand scientific collaboration in NTDs in Brazil [12–14] , Canada [30] and Germany [31] . Co-authorship network analysis allows better understanding of the markedly cooperative context in which scientific knowledge is generated by identifying , among other information , key leading members ( countries , organizations or individuals ) that could act as “bridges” in the scientific community . Network analysis includes quantitative metrics addressing properties of network members , estimating the importance of a node relative to all other nodes in a given network , taking into account the different ways in which it interacts and communicates with the rest of the network . Centrality measures are the most commonly used to identify the nodes that have strategic significance in the network [16] . Particularly useful are the betweenness centrality values of individual nodes , indicating whether they are connecting parts of a network that would only be poorly connected or not connected at all . Nodes with high betweenness centrality are called bridges , brokers or boundary spanners for their ability to facilitate access to novel information , or resources , facilitate transfer of knowledge , and co-ordinate effort across the network [32] . They are considered key players in that their loss from a network would greatly affect its function and viability , and can be regarded as innovation hubs within networks [33 , 34] . Fig 4 displays the coauthorship networks addressing the epidemics of Ebola ( 2015 ) and Zika ( 2016 ) while Table 2 lists the countries and organizations that are network cutpoints , the number of papers they published on the epidemics and their betweenness centrality measures . In the Ebola 2015 network only industrialized countries and their organizations played a relevant role . In contrast , the Zika 2016 network showed two Brazilian institutions , the Oswaldo Cruz Foundation ( Fiocruz ) and the University of São Paulo ( USP ) among the top 5 . Brazil as a country—an IDC—had the 2nd strongest betweenness centrality , behind the US but ahead of the three OECD countries , France , UK and Italy .
The introduction of the IDC concept a decade ago suggested that some developing countries were mobilizing their scientific and technological workforce to address the main health problems affecting their populations . In some of these countries this investment in health innovation was accompanied by a strong growth of their GNI per capita from 2003 to 2015—7x for China , 4x for Brazil and 3x for India . It is interesting to note that the leading IDCs on the right side of Table 1 ( present study ) , include the BRICS , a group which has been improving its scientific excellence [35] , but now Turkey is also among them , a country that together with Russia , did not show up among the top-25 in the original study [2] . Different approaches have been used to evaluate and measure the innovative performance , capacity and potential of countries . The Global Innovation Index ( GII ) “was launched in 2007 with the simple goal of determining how to find metrics and approaches that better capture the richness of innovation in society and go beyond such traditional measures of innovation as the number of research articles and the level of research and development ( R&D ) expenditures” [36] . The Bloomberg Innovation Index ( BII ) “rates countries on seven factors that when used together , are a representation of innovation levels” [37] . Table 3 lists the top 25 most innovative countries according to these two approaches and to the methodology adopted in this study . It is worth noting that in the BII and GII ranks is the presence of two , or just one country , respectively , that do not belong to the high income economy category—China and Malaysia ( 21st and 23rd positions ) on the BII and China ( 22nd position ) on the GII . This is in stark contrast not only to this study’s rankings , which include eight IDCs among the top 25 countries , with China topping the very first position , but also to recent economic analyses indicating the prominent economic and technological role of China today [38 , 39] . Why such large discrepancies ? One explanation resides on the conceptual framework behind each methodology: While BII and GII use multiple activities , parameters , indicators and weights to estimate innovative capacity and performance , we focused on the number of total international patent applications ( PCT ) , normalized for each country’s economic strength and population , as a proxy to assess its innovation capacity . The prominent innovative role of China today is not yet fully recognized mainly due to the small penetration of the Chinese language in international technological and scientific databases . This language barrier , however , is becoming less important due to the technological evolution of machine translation systems . A partnership involving the European Patent Office and Google , for example , led to the creation of a new technology called Neural Machine Translation ( NMT ) [42] . NMT has allowed a large quantity of patent documents previously restricted to a country’s own patent office and language to become freely available in international databases . In 2015 China became the first country office to receive over a million patent applications in a single year , receiving almost as many applications as United States , Japan and Republic of Korea combined [43] ( Fig 5 ) . In this way technologies developed in China and patented in Chinese became freely available for consultation in other languages , disclosing to the rest of the world the recent technological progress of that country . Each ranking approach carries its own pros and cons , advantages and disadvantages and will therefore generate a different type of classification . Our approach relies on patent statistics , an indicator for science and technology directly relevant to innovation measurements , easily available in an entrepreneurial environment and whose importance is recognized by the OECD [44] . Although using patents as a proxy to innovation is an ongoing debate [45 , 46] , our results strongly suggest that innovation should not be regarded as a privilege of high income , industrialized economies and that IDCs , particularly China and India , have become now serious players among the big actors . The origin of the “neglected diseases” concept may be traced to the Rockefeller Foundation’s Program “The Great Neglected Diseases of Mankind” founded in 1977 by Kenneth Warren , the Foundation’s Director of Medical Sciences . In his view diseases such as schistosomiasis , malaria and others were neglected by funding agencies such as the US National Institutes of Health which invested mostly in other diseases such as cancer [47 , 48] . During the late 90s’ “neglected diseases” and “most neglected diseases” were regarded as neglected by the pharmaceutical companies as they had no interest in developing drugs or medicines for patients suffering from them [49 , 50] . Several international initiatives were created and implemented in order to compensate for this “neglect” from funding agencies or pharmaceutical companies . One of the first was the UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical diseases ( TDR ) , launched in 1975 and hosted at the World Health Organization in Geneva . In 2000 the United Nations Millennium Declaration and the establishment of the Millennium Development Goals was accompanied by the creation of several Partnerships for Product Development ( PDPs ) such as Medicines for Malaria Venture ( MMV ) , the Global Alliance for Tuberculosis Drug Development ( TB Alliance ) , the Foundation for Innovative New Diagnostics ( FIND ) and the Drugs for Neglected Diseases initiative ( DNDi ) . An important outcome of this global mobilization was the recognition that NTDs have poverty-promoting features and other socioeconomic consequences; in other words they not only occur in the setting of poverty , they actually promote poverty [51] . This shift from a passive definition ( neglected by someone else ) to an active one ( poverty promotion ) brought the NTDs to the center of the developing countries’ social and economic development agendas , forcing the health and science and technology systems of IDCs to play a more endogenous , autonomous and active role in NTD control and prevention [3] , instead of just waiting for medical solutions developed abroad , which was the main paradigm of the last century [52] . Fig 2 shows that IDCs invest above the average on NTDs in a critical component , or determinant , of health innovation: research and development [2 , 22 , 23]: Seven of the nine countries prioritizing NTDs R&D , are IDCs: Brazil , Thailand , Argentina , Indonesia , India , Mexico and Malaysia . The two industrialized countries that this study demonstrated that are also part of this top group , Singapore and Switzerland , have built and run two institutions fully dedicated to R&D on NTDs: the Novartis Institute for Tropical Diseases ( NITD ) in Singapore and the Swiss Tropical and Public Health Institute ( Swiss TPH ) in Basel . Another important conceptual paradigm shift that continues to evolve is the focus “from neglected diseases to neglected populations” [53] . We used this point of view to analyze the role of IDCs and their institutions in two recent sanitary crises that impacted neglected populations in Africa and South America between 2014-2016: the Ebola epidemics in West Africa and the spread of Zika virus in South America . Fig 4 and Table 2 illustrate that most of the work by researchers , epidemiologists and public health decision makers , as reflected by their publications in international peer reviewed journals that had the words “Ebola” or “Zika” in the title , had quite different profiles in terms of the geographic location of the authors and the relevance of the affected countries in coauthorship networks . During an epidemics , the position occupied by countries , institutions and authors in a network that is generating knowledge about a disease is an important parameter for influencing response , decision-making , preparedness and empowerment . In co-authorship networks the betweenness centrality of a country or institution can be an excellent indicator of this reality . A node with a higher betweenness centrality would have more control over the network due to the volume of information that will depend on that entity for the pass through of knowledge [54] . As a point of control in the communication network , betweenness centrality measures the degree to which a node can function and does not necessarily correlate with volume of publications . In the Ebola network , for example , Australia had higher betweenness centrality than France , but it was behind it in the number of publications ( Table 2 ) . In the Zika network , the UK , despite publishing more papers than France on this specific subject ( 104 and 81 papers , respectively ) , ranked lower in betweenness centrality . The majority of the work conducted in West Africa to detect , diagnose and control the Ebola epidemics was carried out by teams brought from abroad in response to a dramatic appeal from the Director General of the World Health Organization when she declared the Ebola epidemics a Public Health Emergency of International Concern ( PHEIC ) . On that occasion it was emphasized that West African countries’ health systems needed international help to manage infection [55] . The Zika epidemics in the Americas that started in the northeast of Brazil , on the other hand , was detected and characterized by physicians and researchers working at local health services , hospitals or universities . Brazilian scientists were responsible for seminal work on outbreak characterization [56–59] , clinical case definition [60] , sexual transmission [61] . Furthermore , their research was critical to document the anomalous high incidence of microcephaly and other newborn malformations that were associated them with Zika virus infecting pregnant women [62–66] , and precipitated studies on antiviral treatment [67 , 68] and vector biology [69 , 70] . When a network of international scientists issued an alarming statement to propose that the “Rio de Janeiro’s 2016 Olympic Games must not proceed…because Brazil’s Zika problem is inconveniently not ending” [71] , it was a report from Brazilian scientists [72] that quickly brought evidence that the epidemics had already receded . Based on facts , the WHO issued a formal press release stating that “there is no public health justification for postponing or cancelling the games” [73] . This decision proved correct: the Rio Olympics , which represented an investment of above 10 billion US dollars [74] , proceeded as expected and no Zika cases being reported during the event [75] . It is known that patents are not strictly direct indicators of the innovation process . Nevertheless , as patents are a legal property right over an invention that provides to its owner an exclusive right for a limited period , patents are often issued along the route leading to innovation . In accordance with the OSLO manual ( 2005 ) patent statistics never ceased to be an especially relevant science and technology indicator to measure of technologies of product and process innovation . We also recognize the limitation of using coauthorship data as a proxy of scientific collaboration , acknowledging that not all cooperative efforts result in publications , and not all co-authored papers necessarily infer collaboration and knowledge exchange . Even so , it is assumed that , in most cases , coauthorship indicates an active collaboration that goes beyond mere data sharing . Since its introduction in 2005 , the IDC concept has positively contributed to the analysis of the roles that different countries play in innovation . Our present analysis , based on the recent Ebola and Zika epidemics , demonstrated the importance of the preexisting healthcare infrastructure and research networks in an IDC to mount an effective response against an emerging health threat . The overall response to the Ebola epidemic , which only affected non-IDC countries , was primarily driven by outside experts who were severely constrained by local customs and societal norms . This observation has significant global policy implications for future responses: the global community clearly needs to prioritize long-term support to strengthen local leadership in countries that encompass geographic hotspots of disease emergence [76] . Expanding the science , technology and innovation base in these countries and regions will improve the response to emerging disease outbreaks and is well aligned with reaching the United Nations Sustainable Development Goals ( https://sustainabledevelopment . un . org/ ) . We anticipate that the application of the IDC concept to areas beyond healthcare will uncover the participation of these countries in social and economic development , which traditional analytic tools have underestimated or not considered . | Splitting countries into two groups—rich and poor; developed ( the “North” ) and developing ( the “South” ) ; leaders and followers—appears to us to be progressively more simplistic , unrealistic and a heritage from colonial times . Triggered by the first wave of globalization , the share of world income going to today’s wealthy nations soared from twenty to almost seventy percent between 1820 and 1990 , a fact that supported and strengthened this dichotomic vision; however , the new globalization driven by information technology has propelled the rapid industrialization of several developing nations and simultaneous deindustrialization of developed nations , a phenomenon that has not yet been fully understood nor reflected in traditional economic indexes and analyses . In this article we revisit the 2005 concept of Innovative Developing Countries ( IDCs ) that points to the underrepresentation of IDCs in well-known innovation indexes and country ranks . Our analysis clearly shows a prominent role for IDCs in health innovation , research and development on NTDs and in epidemics preparedness , prevention and control . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"china",
"economic",
"analysis",
"pathogens",
"tropical",
"diseases",
"geographical",
"locations",
"microbiology",
"social",
"sciences",
"viruses",
"rna",
"viruses",
"network",
"analysis",
"global",
"health",
"neglected",
"tropical",
"diseases",
"public",
"and",
"occupational",
"health",
"computer",
"and",
"information",
"sciences",
"south",
"america",
"medical",
"microbiology",
"microbial",
"pathogens",
"economics",
"brazil",
"centrality",
"people",
"and",
"places",
"asia",
"flaviviruses",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"organisms",
"zika",
"virus"
]
| 2018 | Revisiting the concept of Innovative Developing Countries (IDCs) for its relevance to health innovation and neglected tropical diseases and for the prevention and control of epidemics |
Chronic human immunodeficiency virus-1 ( HIV-1 ) infection in patients leads to multi-lineage hematopoietic abnormalities or pancytopenia . The deficiency in hematopoietic progenitor cells ( HPCs ) induced by HIV-1 infection has been proposed , but the relevant mechanisms are poorly understood . We report here that both human CD34+CD38- early and CD34+CD38+ intermediate HPCs were maintained in the bone marrow ( BM ) of humanized mice . Chronic HIV-1 infection preferentially depleted CD34+CD38- early HPCs in the BM and reduced their proliferation potential in vivo in both HIV-1-infected patients and humanized mice , while CD34+CD38+ intermediate HSCs were relatively unaffected . Strikingly , depletion of plasmacytoid dendritic cells ( pDCs ) prevented human CD34+CD38- early HPCs from HIV-1 infection-induced depletion and functional impairment and restored the gene expression profile of purified CD34+ HPCs in humanized mice . These findings suggest that pDCs contribute to the early hematopoietic suppression induced by chronic HIV-1 infection and provide a novel therapeutic target for the hematopoiesis suppression in HIV-1 patients .
Human immunodeficiency virus-1 ( HIV-1 ) infection in patients leads to multi-lineage hematopoietic abnormalities , including anemia , granulocytopenia and thrombocytopenia [1 , 2] . Abnormalities in fetal hematopoiesis have also been reported in aborted fetuses from HIV-1 seropositive women [3] . The defect in hematopoietic progenitor cells ( HPCs ) or hematopoiesis induced by HIV-1 has been proposed [1 , 4 , 5] . In addition , the degree of the hematopoietic pathology correlates with the stage of disease progression [6] , and end-stage disease is characterized by pancytopenia [1] . Long-term bone marrow ( BM ) cultures from HIV-1-infected patients exhibit low CD34+ progenitor cell growth and differentiation [7 , 8] , indicating functional impairment of early hematopoietic progenitors . Although the successful highly active antiretroviral therapy ( HAART ) clearly ameliorates HIV-1-associated hemato-suppression , it does not completely restore blood cell development [9] . These observations indicate that hematopoietic failure is an important aspect of HIV-1 infection-induced pathogenesis [10] . HPCs are comprised of diverse populations , including both early and intermediate progenitors . Each subpopulation expresses distinct sets of cell surface antigens , although they all express the cell surface antigen CD34 [11 , 12] . Early and intermediate populations can be distinguished by the expression of CD38 , with the former being negative for CD38 and the latter being positive for this antigen . Functionally , intermediate progenitors include common myeloid progenitors that can give rise to all myeloid , erythroid and megakaryocyte lineages . Due to limited access to the BM in humans , properties of human HPC subsets and their alterations in healthy and HIV-1 disease states have been difficult to characterize . The mechanisms underlying abnormal hematopoiesis in HIV-1 infection remains unclear due to the paucity of robust animal models that mimic human hemato-suppression in vivo . Although previous studies failed to detect HIV-1 infection of HPCs , recent reports indicated that HIV-1 could directly infect HPC subsets and lead to their impairment [13–16] . In addition , HIV-1 proteins such as Nef [17] and prolonged treatment with antiretroviral drugs could also compromise hematopoietic progenitors [18] . Although these studies investigated a litany of direct and indirect causes of HIV-1-associated hemato-suppression , how HIV-1 affects hematopoiesis in vivo remains unclear . There is emerging evidence that certain cytokines induced during inflammation have significant effects on HPCs in the BM . Type I and II interferon ( IFN ) [19–23] , tumor necrosis factor ( TNF ) [24–26] and lipopolysaccharide ( LPS ) [27 , 28] directly stimulate HPC proliferation and differentiation , thereby increasing the short-term output of mature effector leukocytes . However , chronic inflammatory cytokine signaling can lead to functional exhaustion of HPCs [19 , 22 , 28] . Our previous study demonstrated that plasmacytoid dendritic cells ( pDCs ) , the major type I interferon ( IFN-I ) -producing cells during acute or chronic HIV-1 infection , could inhibit viral replication while significantly contributing to HIV-1 infection-induced immune-pathogenesis , including increased immune cell death and reduced immune reconstitution of human CD45+ cells in humanized mice in vivo [29] . These findings suggest that pDCs play a pivotal role in the hemato-suppression induced by chronic HIV-1 infection . In this study , we sought to understand the role of pDCs in HIV-1-associated hemato-suppression in a humanized mouse model in vivo . We discovered that HIV-1 infection depleted CD34+CD38- early HPCs and functionally impaired human CD34+ HPCs in the BM of patients and humanized mice with HIV-1 infection . This phenomenon was further found to be dependent on pDCs , as depletion of pDCs significantly recovered HPC cell numbers and multi-lineage colony-forming functions . Our present study therefore reveals a novel mechanism for hematopoiesis suppression induced by chronic HIV-1 infection and provides a new strategy to rescue HPC function and halt HIV-1 disease progression .
By gating on live human CD45+ cells ( VLD-mCD45- ) with a lymphoid morphology that lacked common markers ( Lineage− ) for T cells ( CD3 ) , B cells ( CD19 and CD20 ) or NK cells ( CD56 and CD16 ) , we identified human BM-derived HPCs as CD34+ cells , which included early CD34+CD38- and intermediate CD34+CD38+ subpopulations ( S1 Fig ) . We then analyzed the human HPCs from the BM of humanized mice at various time points after human CD34+ cell transplantation . The early and intermediate HPCs could be detected significantly at both 16 weeks and 50 weeks after CD34+ cell transplantation with relatively stable levels ( Fig 1A ) . To determine if human HPCs derived from humanized mice were functional , human CD34+ cells were isolated from the BM of humanized mice and human fetal livers , and colony-forming unit ( CFU ) assays were performed by culturing CD34+ HPCs in a complete methylcellulose medium system . Two weeks later , HPCs derived from the BM of humanized mice produced 50 colonies on average for every 500 CD34+ cells plated , similar to that of human fetal liver-derived CD34+ cells ( Fig 1B ) . All hematopoietic lineages were generated in cultures from the BM of humanized mice . In addition , HPCs from humanized mice could proliferate and differentiate into various blood lineage cells in vitro at a similar frequency to that of CD34+ cells from human fetal livers , including colony-forming unit-granulocyte and macrophage ( CFU-GM ) , colony-forming unit-erythroid ( CFU-E ) and colony-forming unit-granulocyte , erythroid , macrophage , megakaryocyte ( CFU-GEMM ) ( Fig 1B ) . We then measured the proliferation capacity of HPCs by BrdU labeling in vivo and found that 8 . 9% of human CD34+ cells showed proliferation ( Fig 1C ) . Notably , the CD34+CD38- early HPCs were much more proliferative , with an average of nearly 25% of cells being BrdU positive , which was significantly higher than the relatively quiescent CD34+CD38+ intermediate HPCs with 1 . 6% BrdU labeling ( Fig 1C and 1D ) . These data suggest that the human CD34+CD38- early HPCs and CD34+CD38+ intermediate HPCs were both functionally developed and maintained in the BM of humanized mice . Utilizing the robust animal model , we were able to investigate whether chronic HIV-1 infection affected human HPCs . HIV-1 infection was established in humanized mice , as measured by plasma HIV-1 RNA ( copies/mL , S2 Fig ) . On termination , we also measured HIV-1 gag p24 expression in both T cells and CD34+ HPCs by flow cytometry . Although a previous study suggested that HIV-1 has the potential to infect intermediate CD34+CD38+ HPCs [13] , we found that p24 expression was absent in BM CD34+ HPCs from humanized mice with HIV-1 infection; in contrast , CD3+ T cells showed high levels of p24 expression ( 10 . 5% ) ( Fig 2A ) . Further analysis indicated that the frequency of CD34+CD38- early HPCs was largely decreased in humanized mice with chronic HIV-1 infection , while the proportion of intermediate CD34+CD38+ HSCs was relatively expanded ( Fig 2B ) . Summarized data further demonstrated that chronic HIV-1 infection significantly reduced CD34+CD38- early HPCs by nearly 8-fold as compared to the non-infected animals; meanwhile , the proportion of CD34+CD38+ intermediate HPCs was increased from 76% to nearly 100% as shown in the stacked bar graph ( Fig 2C ) . When the absolute cell counts of early and intermediate HPCs were calculated , the number of CD34+CD38- early HPCs was dramatically reduced in the BM of chronically infected animals ( Fig 2D ) , while CD34+CD38+ intermediate HPC counts were affected mildly ( Fig 2E ) . These results suggest a depletion of early HPCs during chronic HIV-1 infection . Importantly , we observed a similar depletion of BM CD34+CD38- early HPCs in HIV-1-infected patients . As shown in Fig 2F , the percentage of CD34+CD38- early HPCs was significantly decreased within total CD34+ HPCs in an HIV-1-infected patient when compared to a healthy control ( HC ) . The depletion of CD34+CD38- early HPCs in human BM from HIV-1 infection was strengthened by the addition of more patients ( n = 5 ) ( Fig 2G ) . In contrast , the proportion of intermediate CD34+CD38+ HSCs was increased within total HPCs in HIV-1-infected patients relative to those of HC subjects ( Fig 2H ) . These data indicated that early CD34+CD38- HSCs were preferentially depleted by chronic HIV-1 infection , and the humanized mouse is a highly relevant animal model that mimics HIV-1-induced hemato-suppression conditions in patients . We next analyzed the effect of chronic HIV-1 infection on the homeostatic proliferation of human HPCs in humanized mice . The results indicate that proliferation of human CD34+ cells in the BM was inhibited by approximately 3-fold in chronic HIV-1 infection compared to the mock animals ( Fig 3A ) . Consistent with the preferential reduction of CD34+CD38- HPCs , BrdU-positive CD34+CD38- early HPCs were significantly decreased by chronic HIV-1 infection; meanwhile , the proliferation of CD34+CD38+ intermediate HPCs was only mildly reduced by chronic HIV-1 infection ( Fig 3B and 3C ) . In terms of cell numbers , the early HPC counts were significantly decreased by chronic HIV-1 infection compared with mock treatment in mice ( Fig 3D ) , while intermediate HPC counts were only slightly reduced ( Fig 3E ) . Thus , while both cell types were less proliferative in the presence of HIV-1 , the more marked difference was observed in CD34+CD38- cells . Therefore , chronic HIV-1 infection appeared to suppress human hematopoiesis by inhibiting homeostatic proliferation of human early HPCs in the BM . In order to assess the quality of in vivo human HPCs during chronic HIV-1 infection , we measured CFU activity of purified human Lin-CD34+ HPCs from mock- or HIV-1-infected humanized mice , including GM , E and GEMM ( S3 Fig ) . As shown in Fig 4 , HPCs isolated from uninfected mice consistently produced over 50 colonies per 500 CD34+ cells on average , whereas HPCs derived from HIV-1-infected mice produced less than 30 colonies per 500 CD34+ HPCs . Moreover , the ability of HPCs to generate all lineages , including GM , E and GEMM , was also suppressed to some extent in chronically infected mice . Therefore , the results indicated that chronic HIV-1 infection leads to the impaired differentiation of HPCs in vivo . Increasing reports have demonstrated that chronic inflammation could lead to the functional exhaustion of BM HPCs [19 , 22 , 28] . Our recent study indicated that depletion of pDCs efficiently rescued human CD45 cell reconstitution in humanized mice with chronic HIV-1 infection [29] . Thus , we hypothesized that pDCs may be responsible for the depletion of CD34+CD38- early HPCs and their functional impairment during chronic HIV-1 infection . To address the role of pDCs in the impairment of CD34+CD38- HPCs in HIV-1 infection , humanized mice with chronic HIV-1 infection were treated with a pDC-depleting antibody ( 15B ) as in our previous report [29] . Similarly , the plasma viral load was increased upon the depletion of pDCs and maintained at a higher level until termination ( S4A Fig ) . Notably , the depletion of pDCs significantly changed the percentage of CD34+CD38- early HPCs in the BM from humanized mice with chronic HIV-1 infection ( Fig 5A ) . Pooled data further confirmed that both the percentages and cell counts of CD34+CD38- early HPCs were restored by depletion of pDCs in HIV-1-infected mice ( Fig 5B and 5C ) . In contrast , pDC depletion did not influence the percentages of CD34+CD38- early HPCs ( CD38 expression ) and their proliferation in vivo indicated by BrdU expression in the BM in the absence of HIV-1 infection ( S4B Fig ) . In addition , the cell counts of CD34+CD38+ intermediate HPCs showed only minor recovery by the depletion of pDCs during chronic HIV-1 infection , which is consistent with the slight decrease in proportion of CD34+CD38+ intermediate HPCs ( Fig 5B and 5C ) . Notably , only half of the animals showed a recovery in the proportion of CD34+CD38- early HPCs after pDC depletion in Fig 5B , which prompted us to investigate whether the extent of pDC depletion affected the effectiveness of rescue of CD34+CD38- early HPCs in humanized mice with chronic HIV-1 infection . We therefore divided the 13 animals into two groups; animals with less than the median percentage of CD34+CD38- HPCs were placed in the "non-rescued" group ( n = 6 ) , while the others were included in the "rescued" group ( n = 7 ) . The rescued group was found to have significantly more CD34+CD38- early HPCs and fewer CD34+CD38+ intermediate HPCs than the non-rescued group ( S4C and S4D Fig ) . Importantly , the rescued mice showed a marked lack of pDCs in the BM and lower levels of IFN-α in plasma compared with non-rescued mice ( S4E and S4F Fig ) . Accordingly , the rescued mice were also characterized by a higher level of HIV-1 replication than that of non-rescued mice ( S4G Fig ) . Thus , we observed that the CD34+CD38- early HPCs was negatively correlated with pDC percentages within CD45+ cells in BM of these HIV-1 infected humanized mice with pDC depletion ( S4H Fig ) . These data indicated that the extent of rescue of CD34+CD38- early HPCs was closely linked to in vivo pDC depletion rather than other potential causes such as HIV viral load in humanized mice with chronic HIV-1 infection . To further qualify HPCs after pDC depletion , Lin-CD34+ cells were purified for colony-forming assays ex vivo . Cell colonies including GM , E and GEMM were found in culture ( S3 Fig ) . The results demonstrated that pDC depletion could dramatically enhance CFU activity of the Lin-CD34+ cell population as well as increase the quantity of each colony type individually as compared with HIV-1-infected mice ( Fig 5D ) . To understand how pDCs contribute to the impairment of HPCs during chronic HIV-1 infection , we analyzed gene expression of human HPCs in BM from HIV-1 chronically infected humanized mice . Human Lin-CD34+ cells from BM of humanized mice were isolated and submitted for gene expression analysis by cDNA array ( Fig 6A ) , as described in a previous study [29] . A total of 3114 genes were significantly up-regulated , and 2994 genes were down-regulated spontaneously in CD34+ HPCs from HIV-1-infected humanized mice as compared to mock-treated mice ( fold change > 2 , Fig 6B and S5 Fig ) . Astonishingly , pDCs depletion during chronic HIV-1 infection in mice restored most of the interferon-stimulating genes ( ISGs ) to levels found in non-infected animals ( S6A and S6B Fig ) . Along with the recovery of ISG gene expression , only 924 genes were significantly up-regulated , and 364 genes were down-regulated in BM CD34+ HPCs with pDCs depletion as compared to mock-treated mice ( fold change > 2 , Fig 6B and S5 Fig ) . Thus , pDCs depletion resulted in a restoration of a total of 5664 genes among 6108 genes ( > 92 . 7% ) that changed during HIV-1 infection in humanized mice , drawing back the whole gene expression profile to a pattern quite similar to that in mock-infected mice ( Fig 6B and S5 Fig ) . We then searched the GeneGo database to identify potentially relevant pathways in the genes influenced by chronic HIV-1 infection . The hematopoietic cell lineage was the most affected pathway induced by chronic HIV-1 infection among the top 15 pathways , whereas those dysregulated genes were also significantly restored to mock levels in pDC-depleted BM CD34+ HPCs of humanized mice chronically infected with HIV-1 ( Fig 6C ) . We comprehensively analyzed the 88 genes in the hematopoietic cell lineage pathway ( S7A Fig ) and found that HIV-1 infection induced a significant up-regulation of 27 genes and down-regulation of 28 genes in humanized mice relative to mock controls ( S7B Fig ) . However , depletion of pDCs during HIV-1 infection only induced a significant change in expression of one gene in relation to mock controls ( S7B Fig ) . Summarized data further indicated that although HIV-1 infection led to a significant change of most of the genes with more than a 2-fold change , pDC depletion could attenuate the up-regulation of genes and restored the down-regulated genes to normal levels during chronic HIV-1 infection ( Fig 6D and S7C Fig ) . In particular , some genes related to the HPC quiescent state ( DTX3L and CXCR4 ) [30] , colony forming capacity ( CD34 , CD38 , FLT3 and TGFBR1-3 ) [31] , self-renewal and expansion capacity ( Notch1-2 , Jagged1-2 and Hes-1 ) [32 , 33] , as well as several important cell death genes were significantly altered by chronic HIV-1 infection , while pDCs depletion in HIV-1-infected humanized mice largely restored the abnormal expression of these genes to a similar pattern as seen in mock-infected mice ( Fig 6E and 6F , S1 Table ) . Simultaneously , we also performed mRNA expression analysis using spleen-derived CD45+ cells ( S8A Fig ) . A total of 4757 genes were significantly up-regulated or down-regulated in splenic CD45+ cells from HIV-1-infected humanized mice as compared to mock-infected mice ( fold change > 2 , S8B Fig ) . The depletion of pDCs resulted in a restoration of about 44 . 0% of genes ( 2092/4757 ) of splenic cells changed by HIV-1 infection in humanized mice to a pattern quite similar to that in mock-infected mice ( S8B Fig ) . The pathway analysis indicated that systemic lupus erythematosus ( SLE ) was the most affected pathway induced by chronic HIV-1 infection ( S8C Fig ) , whereas the dysregulated genes expressed by splenic CD45+ cells were not significantly restored to mock levels in pDC-depleted humanized mice chronically infected with HIV-1 ( S8D Fig ) . We also analyzed 127 genes in the SLE pathway changed by HIV-1 infection ( S8E Fig ) and found that HIV-1 infection induced significant changes of 48 genes relative to mock controls . However , depletion of pDCs during HIV-1 infection also induced significant changes in expression of 115 genes in relation to mock controls ( S8E Fig ) , indicating that pDCs depletion failed to restore the changes in gene expression by spleen CD45+ cells to normal levels during chronic HIV-1 infection ( S8F Fig ) . These data strongly suggest that pDCs has relatively unique effects on the numerical reduction and functional impairment of BM CD34+ HPCs in HIV-1 chronically infected humanized mice , contributing to the suppression of hematopoiesis characterized by dysregulation of gene expression profiles . We finally tested whether IFN-I directly up-regulates CD38 expression on HPCs in vitro . As shown in S9A and S9B Fig , neither IFN-α nor IFN-β showed any significant effect on CD38 expression on CD34+ cells in vitro . In addition , IFN-I culture did not affect CD34+ HPC expansion ( S9C Fig ) . These data indicate that IFN-I did not directly affect CD38 expression and expansion of CD34+ HPC cells in vitro .
The present study demonstrates , for the first time , that human CD34+CD38- early HSCs are subject to preferential depletion and functional impairment in vivo in the BM of humanized mice with chronic HIV-1 infection , in a pDC-dependent fashion . This study thus reveals a new target for the development of novel drugs targeting pDC activity to treat hematopoietic disorders during chronic HIV-1 infection and also demonstrates the utility of humanized mice to investigate important questions on HIV-1-mediated hematological abnormalities in the BM in vivo . Previous studies have attempted to delineate the mechanism by which HIV-1 infection induces the impairment of HPCs , but it remains an intractable question due to the lack of a suitable experimental animal model that closely mimics human hematopoiesis during an ongoing HIV-1 infection in vivo . Here , we provide evidence that both early and intermediate HPCs are functionally developed and maintained for long-term self-renewal in the BM of humanized mice in vivo . This animal model has been demonstrated to be persistently infected by various strains of HIV-1 and develop major immune pathogenesis induced by acute or chronic HIV-1 infection as observed in HIV-1 patients [29 , 34–36] . Importantly , Nixon et al . made initial advances toward adopting a humanized mouse model to investigate an HIV-1 infection-induced hematopoiesis disorder [13] . Our present study underscores the rationale for utilizing a humanized mouse model to study the impact of HIV-1 infection on hematopoiesis in vivo . Taken together , these studies support the humanized mouse model as an experimentally amenable in vivo system for investigating HIV-1-associated pathogenesis . Accumulating evidence has demonstrated that patients with long-term HIV-1 infection exhibit a deficiency in hematopoiesis [1 , 4 , 5] , although the stage at which HPCs are impaired by HIV-1 infection is unclear . Nixon et al . showed that HPCs were susceptible to HIV-1 infection in vitro and in vivo in humanized mice and concluded that direct infection of intermediate CD34+CD38+ HPCs by HIV-1 adversely affected their hematopoietic potential and correlated with the observed pancytopenia in HIV-1 infected patients [13] . In this study , we found that CD34+CD38- early HPCs were preferentially depleted during chronic HIV-1 infection , which correlated with the depression of hematopoiesis development and dysregulated gene expression in bulk Lin-CD34+ HPCs . However , our study failed to detect significant productive infection of Lin-CD34+ HPCs in the BM in vivo even with pDC depletion , as the depletion of pDCs led to dramatically increased viral replication . One possible explanation for this finding is that different HIV-1 strains may exhibit discrepancies during infection of HPCs in vivo . Future studies should examine the effect on HPC subsets by infection with other HIV-1 strains with different tropisms . The mechanisms leading to abnormal hematopoiesis have not been clearly addressed in HIV-1 infection . Aside from the direct infection of HPC subsets , pDCs are possibly critical factors leading to hematopoietic suppression during HIV-1 infection , as the depletion of pDCs rescued early HPCs and their hematopoiesis . Available lines of evidence have also demonstrated that pDCs substantially mediate detrimental effects during chronic HIV-1 infection in vivo , even while they inhibit viral replication [29 , 37–39] . In addition , pDCs can secrete other pro-inflammatory cytokines , including TNF-α and IL-6 . These chronic inflammatory cytokines can lead to exhaustion of hematopoiesis [19 , 22 , 28] . Most importantly , pDCs are the major IFN-I-producing cells during HIV-1 infection [29 , 40 , 41] . Currently , IFN-I is perhaps the primary contributor , since depletion of pDCs completely abolished IFN-I responses in humanized mice with chronic HIV-1 infection [29] . IFN-I has been recently demonstrated to be actively involved in immune pathogenesis of chronic virus infection [39 , 42–45] . Our data indicate that IFN-I did not directly affect CD38 expression on HPCs in vitro during short-term culture although it possibly promote the maturation of embryonic hematopoietic stem cells [46] . Other factors associated with chronic HIV-1 infection may contribute to IFN-induced HPC depletion and will be investigated in future study . Of course , we could not exclude the possibility that HIV-1 products are involved in the dysregulation of hematopoietic development . For example , HIV-1 Nef has been found to be responsible for hematopoietic defects of the BM in HIV-1 infection , dependent on the presence and activation of the PPARγ signaling pathway [17] . Thus , pDCs may contribute to abnormal hematopoiesis during chronic HIV-1 infection directly through viral infection or indirectly via diverse cytokines . Taking these studies into consideration , we also propose that HIV-1 infection may affect HPC function through multiple mechanisms ( S10 Fig ) . Future studies should investigate in detail the individual factors responsible for compromising hematopoietic activity . In summary , pDCs play a pivotal role in the immune-pathogenesis and hematopoiesis depression induced by chronic HIV-1 infection . This study , therefore , provides new insight into HIV-1-induced dysregulation of hematopoiesis and provides a novel strategy for treating abnormal hematopoiesis during chronic HIV-1 infection .
Approval for animal work was obtained from the University of North Carolina Institutional Animal Care and Use Committee ( IACUC ID: 14–100 ) . The study protocol on human samples was approved by the Institutional Review Board and the Ethics Committee of Beijing 302 Hospital in China . The written informed consent was obtained from each subject . Human BM samples were obtained from adult donors with liver transplantation as healthy controls and from HIV-1-infected adult patients for pathological diagnosis . Human fetal livers and thymuses ( gestational age 16 to 20 weeks ) were obtained from medically indicated or elective termination of pregnancies through a non-profit intermediary working with outpatient clinics ( Advanced Bioscience Resources , Alameda , CA ) . Written informed consent from the maternal donor was obtained in all cases under regulations governing the clinic . All animal studies were conducted following NIH guidelines for housing and care of laboratory animals . The project was reviewed by the University’s Office of Human Research Ethics , which determined that this submission does not constitute human subjects research as defined under federal regulations [45 CFR 46 . 102 ( d or f ) and 21 CFR 56 . 102 ( c ) ( e ) ( l ) ] . We constructed humanized Balb/c rag2-γc ( DKO ) mice and Nod-rag1-γc ( NRG ) mice ( The Jackson Laboratory ) in a similar manner as previously reported [36] . Briefly , human CD34+ cells were isolated from 16- to 20-week-old fetal liver tissues ( Advanced Bioscience Resources , Alameda , CA ) . Tissues were digested with liver digest medium ( Invitrogen , Frederick , MD ) . The suspension was filtered through a 70-μm cell strainer ( BD Falcon , Lincoln Park , NJ ) and centrifuged at 150 × g for 5 minutes to isolate mononuclear cells by Ficoll . After selection with the CD34+ magnetic-activated cell sorting ( MACS ) kit , CD34+ HPCs ( 0 . 5 × 106 ) were injected into the liver of each 2- to 6-day-old DKO or NRG mice , which had been previously irradiated at 300 rad . More than 95% of the humanized mice were stably reconstituted with human leukocytes in the blood ( 60–90% at 12–14 weeks ) . Each cohort had similar levels of engraftment . All mice were housed at the University of North Carolina at Chapel Hill . An R5-tropic strain of HIV-1 , JR-CSF , was used for chronic HIV-1 infection . All viruses were generated by transfection of 293 T cells ( SIGMA-ALORICH , Cat#12022001-1VL ) with pYK-JRCSF ( NIH AIDS reagents program , Cat# 2708 ) . Humanized mice with stable human leukocyte reconstitution were infected with JR-CSF at a dose of 10 ng p24/mouse , through intravenous ( i . v . ) injection . Humanized mice infected with 293 T mock supernatant were used in control groups ( S2 Fig ) . Viral genomic RNA in plasma was extracted using the QIAamp Viral RNA Mini Kit ( QIAGEN , Cat# 52904 ) according to the manufacturer’s instruction . HIV-1 replication ( genome copies/ml plasma ) was measured by real-time PCR ( ABI Applied Biosystem ) or by p24-FACS detection of productively infected human T cells . A monoclonal antibody specific to blood dendritic cell antigen-2 ( BDCA2 ) , 15B , was used to treat humanized mice through intra-peritoneal ( i . p . ) injection ( 4 mg/kg ) as previously reported [29] . Briefly , 15B was applied to mice at 7 weeks post-infection by injecting twice every week for another 4 weeks . For chronic JR-CSF infection , mice were terminated at 12 week post-infection . On termination , total leukocytes were isolated from mouse lymphoid organs as previously described [29 , 34–36] . Lymphoid tissues , including peripheral blood ( PB ) , peripheral lymph nodes ( pLN ) , mesenteric lymph nodes ( mLN ) , spleen and BM were harvested for analysis . Red blood cells were lysed with ACK buffer , and the remaining cells were stained and fixed with 1% ( wt/vol ) formaldehyde before FACS analysis . The total cell number was quantified by using Guava Easycytes with Guava Express software ( Guava ) . Human BM cells were isolated by ficoll-hypaque density gradient centrifugation and collected for further analysis . Surface and intracellular fluorochrome-conjugated antibodies or reagents from Biolegend , BD Bioscience , eBioscience and R&D Systems were used in this study . For humanized mice , live human leukocytes ( Y7-mCD45-hCD45+ ) were analyzed for HPC subsets or phenotypic expression by using the CyAn FACS instrument ( Dako ) . Live/dead fixable violet dead cell dye ( LD7 ) was purchased from Molecular Probes ( Eugene , OR ) . For intracellular p24 staining , freshly isolated cells were collected for surface staining , followed by cell permeabilization using a Cytofix/Cytoperm kit ( BD Bioscience ) and intracellular staining and washing . The data were analyzed using Summit Software . 5-Bromo-2’-deoxyuridine ( BrdU , Cat#: B5002 , Sigma-Aldrich , St . Louis , MO ) was first dissolved in water at a concentration of 10 mg/mL for stock in -20°C . The BrdU stock was then diluted in 200 μL PBS and injected i . p . at 100 mg/kg body weight . Four hours later , the mice were terminated , and BM cells were collected . BrdU staining was performed according to the manufacturer’s instructions . In brief , cells were first stained for surface markers and then incubated with a working solution of the BrdU staining buffer for 15 minutes , followed by incubation with DNase I ( BIO-RAD , Cat#: 7326828 ) for 1 hour at 37°C in the dark . Thereafter , the cells were stained with the FITC-conjugated anti-BrdU antibody for 30 minutes at room temperature in the dark and subsequently washed . The data were analyzed using Summit Software . The EasySep human CD34+ selection kit ( Cat#:18056 , StemCell Tech , Canada ) was used to isolate CD34+ cells from frozen BM cells . The purity of CD34+ cells was greater than 90% . The CD34+ cells were then counted and seeded in complete methylcellulose ( Methocult H04034; Stem Cell Technologies ) at a concentration of 500 cells/mL and plated in 35-mm grid plates , 1 mL/plate , in triplicate per mouse according to the manufacturer’s instructions . Colonies were counted 2 weeks later in a blinded fashion using a QImaging Micropublisher 3 . 3 CCD digital camera and QCapture software version 3 . 0 ( QImaging , Surrey , BC ) . BM cells were pooled by mouse groups for human CD45+ cells sorting . Cells were stained with human CD45 , mouse CD45 and 7-Aminoactinomycin D ( 7-AAD ) . For human CD34+ HSC sorting , anti-lineage ( anti-CD3 , anti-CD14 , anti-CD16 , anti-CD19 , anti-CD20 and anti-CD56 ) and anti-CD34 antibodies were added to the antibody mix . Cell sorting was performed by the UNC Flow Cytometry Core . CD34+ cells were isolated from human fetal liver tissues . Then the cells were cultured in StemSpan SFEM medium ( Stem Cell Technologies ) with heparin ( 10 μg/ml , Sigma ) , recombinant human SCF ( 20 ng/ml , R&D ) , thrombopoietin ( 40 ng/ml , Cell Sciences ) and CHIR99021 ( GSK3 inhibitor , 250 nM , STEMGENT ) for 48 hours in the presence of IFN-α or IFN-β at the dose of 20 IU/ml and 200 IU/ml , respectively . The cells were counted and collected for the detection of CD38 expression on HPCs . RNA purification was carried out using the RNeasy Plus Mini Kit ( Cat# 74134 , QIAGEN , Venlo , Limburg , Netherlands ) according to the manufacturer’s instructions . DNase ( QIAGEN ) treatment was added to the column to eliminate any potential DNA contamination during RNA preparations . Total RNA was checked for quantity , purity and integrity by capillary electrophoresis . RNA was amplified with Cy3- and Cy5-labeled CTP in separate reactions to produce differentially labeled samples and reference cDNAs . Total RNA ( 200 ng to 400 ng ) was used as the starting material to prepare cDNA . Both samples were hybridized to the same microarray ( UNC Genomic and Bioinformatics Core ) using SurePrint G3 Human Gene Expression 8 ◊ 60K Microarray Kit ( Agilent ) . Agilent Feature Extraction v18 software was used to analyze all images . Gene expression values were quantified by the log2 ratio of the red channel intensity ( mean ) vs . green channel intensity ( mean ) , followed by LOWESS normalization to remove the intensity-dependent dye bias . Data were analyzed using GraphPad Prism software version 5 . 0 ( GraphPad software , San Diego , CA ) . Data from different cohorts of mice were compared using a 2-tailed unpaired T test . All results were considered significant for P values < 0 . 05 . | Multi-lineage hematopoietic abnormalities generally occur during chronic infection which results in a disorder of human leukocyte development and differentiation , contributing to human immunodeficiency virus-1 ( HIV-1 ) -infection induced immune-pathogenesis in AIDS patients . Although successful antiretroviral therapy can reduce plasma viral loads to undetectable levels and ameliorate HIV-1-associated hemato-suppression , immune cell development is only partially restored . The mechanism for the abnormal hematopoiesis occurring during chronic HIV-1 infection remains unclear . HIV-1 infection may directly or indirectly functionally impair hematopoietic progenitors by either viral products or induction of persistent inflammatory responses , leading to hematopoiesis obstacles . Here , we show that HIV-1 infection significantly depleted and functionally impaired human hematopoietic progenitors in the bone marrow of both HIV-1-infected patients and humanized mice through a plasmacytoid dendritic cell ( pDC ) -dependent mechanism , as depletion of pDCs significantly recovered cell numbers and functions and gene expression profiles of hematopoietic progenitor cells in humanized mice in vivo . Our study clarifies a novel mechanism underlying hemato-suppression induced by chronic HIV-1 infection and provides a novel strategy to halt HIV-1 disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"blood",
"cells",
"flow",
"cytometry",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"animal",
"models",
"of",
"disease",
"immunology",
"microbiology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"animal",
"models",
"physiological",
"processes",
"model",
"organisms",
"rna",
"viruses",
"experimental",
"organism",
"systems",
"research",
"and",
"analysis",
"methods",
"animal",
"models",
"of",
"infection",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"spectrum",
"analysis",
"techniques",
"white",
"blood",
"cells",
"animal",
"studies",
"animal",
"cells",
"medical",
"microbiology",
"hiv",
"gene",
"expression",
"microbial",
"pathogens",
"mouse",
"models",
"hiv-1",
"spectrophotometry",
"cytophotometry",
"cell",
"staining",
"cell",
"biology",
"viral",
"pathogens",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"lentivirus",
"hematopoiesis",
"organisms"
]
| 2017 | HIV-1 infection depletes human CD34+CD38- hematopoietic progenitor cells via pDC-dependent mechanisms |
Transformation is an important mechanism of microbial evolution through which bacteria have been observed to rapidly adapt in response to clinical interventions; examples include facilitating vaccine evasion and the development of penicillin resistance in the major respiratory pathogen Streptococcus pneumoniae . To characterise the process in detail , the genomes of 124 S . pneumoniae isolates produced through in vitro transformation were sequenced and recombination events detected . Those recombinations importing the selected marker were independent of unselected events elsewhere in the genome , the positions of which were not significantly affected by local sequence similarity between donor and recipient or mismatch repair processes . However , both types of recombinations were sometimes mosaic , with multiple non-contiguous segments originating from the same molecule of donor DNA . The lengths of the unselected events were exponentially distributed with a mean of 2 . 3 kb , implying that recombinations are stochastically resolved with a fixed per base probability of 4 . 4×10−4 bp−1 . This distribution of recombination sizes , coupled with an observed under representation of large insertions within transferred sequence , suggests transformation has the potential to reduce the size of bacterial genomes , and is unlikely to act as an efficient mechanism for the uptake of accessory genomic loci .
Genetic transformation is the process by which cells take up DNA from the environment and integrate the sequence into their genome . Over sixty bacterial species , found in both Gram positive and negative phyla , have been shown to be naturally transformable in the laboratory [1] . These include important human pathogens such as Streptococcus pneumoniae ( the pneumococcus ) , the bacterium in which this phenomenon was first observed [2] . The discovery that the ‘transforming principle’ involved was DNA proved crucial in establishing the chemical nature of the hereditary material [3] . Pneumococci have a dedicated system for the acquisition of DNA from the environment . The double-stranded DNA ( dsDNA ) first contacts a pseudopilus [4] , [5] , then progresses to the uptake pore complex [6] . One strand is degraded [7] , [8] while the other is cleaved into fragments with a median size of ∼6 . 6 kb as it enters the cell [8] , [9] . Proteins such as RecA and DprA are loaded onto the ssDNA once it is in the cytosol , forming a nucleoprotein filament capable of invading the host chromosome at regions of similar sequence [10] , [11] . This permits the integration of the acquired DNA , and hence the horizontal transfer of polymorphisms . In vitro studies of this process using single polymorphism markers have the identified transversion mutations A•T↔C•G and C•G↔G•C as markers transferred via transformation with a high efficiency , the transversion A•T↔T•A as having an intermediate efficiency , while transitions acted as low efficiency markers [12]–[14] , although some inconsistencies , ascribed to neighbouring sequence context , were identified . These differences represent the efficiency with which these mutations are processed by the mismatch repair ( MMR ) system [15]–[21] . Deletions of 3 bp or shorter also act as low efficiency markers that are subject to MMR [14] , [22] , [23] , but those of 5 bp or longer tend to be transferred at a relatively higher rate due to less effective repair by MMR [12]–[14] , [23]–[25] . However , the effects of MMR are abrogated by high numbers ( more than ∼150 ) of imported polymorphisms saturating the repair mechanism [26]: were this not the case , many of the recombinations observed to occur both in vitro and would be impossible [27] . Nevertheless , an inverse linear in vivorelationship is observed between the mean level of sequence divergence and the logarithm of the frequency of recombination events , implying higher densities of polymorphisms also constitute a significant barrier to the exchange of sequence between bacteria [27]–[29] . This has been suggested to be the consequence of the requirement for a minimum threshold length of perfect sequence identity between the donor strand and recipient duplex ( a ‘Minimal Efficiently Processed Segment’ , or MEPS ) at each end of a recombination to allow a crossover to occur [30] . Based on the changing frequency of transfer with donor sequences of different levels of divergence from the recipient , the minimum summed length of the two MEPS flanking S . pneumoniae recombinations was estimated to be 27 bp [27] . The lengths of recombination events themselves have been estimated using several methods . The transfer of multiple polymorphic markers simultaneously found the mean size of recombinations to be around 2 kb [24] . Subsequent experiments investigating the integration of isotopically labeled DNA suggested a mean size between 3 and 6 kb [31] , [32] , while a mean of ∼4 . 4 kb was inferred using species-wide multilocus sequence typing data [33] . Similar sized events were observed through sequencing pbp genes transformed using plasmid DNA in vitro [34] . More recent estimates have been produced by sequencing the genomes of closely related S . pneumoniae isolates: a recent study of the PMEN1 lineage suggested a mean of 6 . 3 kb [35] , while putative transfers between co-colonising strains indicated a mean of between 6 . 9 and 28 kb [36] . Here we describe work focussed on recombinations occurring at the S . pneumoniae capsule biosynthesis ( cps ) locus . Recombinations affecting this region that alter the capsule type of S . pneumoniae have been observed to lead to evasion of anti-pneumococcal polysaccharide conjugate vaccines , which target only a subset of capsule types [35] , [37] . Furthermore , pbp2X and pbp1A , two genes encoding penicillin-binding proteins crucial in determining pneumococcal β lactam susceptibility , closely flank the cps locus . Hence long recombination events can simultaneously lead to both vaccine escape and the acquisition of penicillin resistance [37] , [38] . However , such large events may be disfavoured when cps loci are transferred from a penicillin-sensitive donor to a resistant recipient [35] . In order to study the nature of these events , and other recombinations around the chromosome , we applied next generation sequencing to selected transformants produced under controlled laboratory conditions .
The first experiment involved the transformation of an isolate of S . pneumoniae ATCC 700669 , a serotype 23F strain of the penicillin-resistant PMEN1 lineage [39] ( henceforth referred to as 23F-R ) , with genomic DNA from an acapsular derivative of the penicillin-sensitive S . pneumoniae TIGR4 strain , which has a kanamycin resistance marker in place of a capsule biosynthesis ( cps ) locus [40] ( henceforth referred to as TIGR4Δcps ) ( Figure 1 ) . Multiple transformations were performed using a concentration of either 5 ng mL−1 or 500 ng mL−1 of donor genomic DNA . Recombinations affecting the cps locus were detected either through selection with kanamycin alone , or kanamycin supplemented with ampicillin . This latter condition was used to test whether the transfer of capsule loci from the β lactam-sensitive donor to the resistant recipient may be inhibited by selection against any co-transfer of β lactam-sensitive PBPs [35] . With selection on just kanamycin , the transformation with the lower concentration of DNA , relative to the higher concentration , produced 38-fold fewer transformants ( Wilcoxon test , p = 0 . 024; Table S1 ) , suggesting that the availability of the marker was limiting the frequency of transformation . However , dual selection with ampicillin caused a small , but non-significant , decrease in the number of observed transformants . Therefore , selection for β lactam resistance does not appear to significantly inhibit exchange with penicillin-sensitive lineages at the cps locus , instead suggesting a strong limitation on the size of recombination events reducing the impact of linkage between genes . To test this hypothesis , 21 isolates from each of the four examined conditions ( low and high DNA concentration , and with and without ampicillin selection ) and a sample of the donor DNA were sequenced using the Illumina platform . Alignment of the complete genome sequences of the donor and recipient strains identified 21 , 541 base substitutions ( a mean density of one per 96 bp of sequence aligned between the donor and recipient ) , 579 insertions relative to 23F-R ( 1 bp–14 , 153 bp in size ) and 477 insertions relative to TIGR4Δcps ( 1 bp–76 , 847 bp in size ) . The apparent absence of a strong discernable population structure among pneumococci [41] , other than the association of very closely related isolates , suggests that this pairing should be broadly representative of interactions between different genotypes across the species . Using the criteria described by Harris et al . [42] , Illumina sequence reads generated from the donor DNA identified 17 , 534 SNPs when aligned to the recipient sequence , of which 385 appeared to be false positives that did not correspond to polymorphisms identified through the whole genome alignment . These positions were excluded from subsequent analyses . Similarly , sequence data from the 84 transformants identified 2 , 312 polymorphic sites , of which just 59 did not correspond to polymorphisms transferred from the donor . Six of these sites were false positives also identified using reads from the donor DNA , whereas many of the others appear likely to be the consequence of de novo point mutations or intragenomic recombinations ( examples involving an IS element and the repetitive protein pclA were observed ) . It is possible that this latter phenomenon may have been promoted by the upregulation of the recombination machinery , such as recA , during competence [43] , but we have no control data to support this hypothesis . Recombinant sequence segments ( RSSs ) were detectable as loci containing donor alleles at polymorphic sites , defining the minimum size of the recombination , bounded by recipient alleles at the flanking polymorphic sites , demarcating the maximum size of the exchange ( Figure 2A ) . The actual length may be estimated as being the median ( L50 ) between these two limits , positioning the flanks half way through each boundary region ( BR ) . As with the separation between any two loci , this length can be expressed either as a distance relative to the donor ( L50D ) or recipient ( L50R ) genome , which differ where there are sequence insertions distinguishing the strains . The selected recombination at the cps locus was detected in all transformants . In order to distinguish those events driven by selection from those elsewhere in the chromosome , a ‘primary locus’ , encompassing the cps region , was defined between the genomic loci 294 , 383 bp and 340 , 516 bp in strain 23F-R: all bases between these coordinates had undergone recombination in at least one of the 84 sequenced isolates ( Figure 2B ) . Furthermore , 107 unselected , ‘secondary’ recombinations were observed outside the primary locus ( Figure 2C ) , with one strain alone having a genome-wide total of 15 RSSs . The mean proportion of the recipient genome found to have undergone recombination was 1 . 4% , ranging up to a maximum of 2 . 5%; however , no significant correlation between in vitro growth characteristics and the extent of these secondary transfers could be found ( Figure S1 ) , although the selected transfer at the cps locus did appear to detrimentally affect the growth of all transformants consistently . Secondary recombinations were significantly more common in the strains transformed at a high concentration of DNA ( mean of 2 . 29 secondary events per strain ) than at a low concentration ( mean of 0 . 26 secondary events per strain; Wilcoxon test , p = 1 . 39×10−8 ) . Hence the effective concentration of DNA available for recombination inside the cell can vary . This implies that recombination events involving separate DNA molecules can occur within the same cell concurrently and independently , as has been previously observed in vitro [38] , [44] and inferred from the sequences of clinical isolates [45] , rather than all arising from the import of a single large molecule of DNA , as has been observed in S . agalactiae [46] . The high density of SNPs within the primary locus allows a high-resolution view of the boundaries of the selected RSSs that span the cps locus ( Figure 3 ) . By ordering the selected recombinations by either their upstream or downstream boundaries , the manner in which the primary RSSs end relative to their distance from the selected marker can be observed . This is best modelled as an exponential decay , with similar estimates of the decay constant on both sides: 3 . 43×10−4 bp−1 ( 95% confidence interval , 3 . 42–3 . 44×10−4 ) on the left flank , and 3 . 41×10−4 bp−1 ( 95% confidence interval , 3 . 40–3 . 42×10−4 ) on the right . The symmetrical nature of the decays on both flanks does not reflect spatial patterns of sequence similarity between donor and recipient , which are poorly correlated to distance from the cps locus ( Pearson correlation , R2 = 2 . 19×10−4 ) . Nor is the decay upstream of the locus strongly affected by an IS element insertion in the recipient sequence . Hence there does not appear to be sufficient local sequence heterogeneity in this region to strongly disrupt the overall patterns of sequence integration . The exponential declines suggest that the boundaries of RSSs are formed through a Poisson process , occurring with a low , constant probability per base . The assumption of such a mechanism in eukaryotic systems predicted such a distribution should be observed [47] . In the absence of sequence identity affecting the exponential declines , the number of isolates in the recombinant state should halve over each ∼2 . 4 kb stretch of sequence . On the basis of this rate , of the recombinations that directly affect the cps locus , 2 . 6% will replace pbp1A and 3 . 5% will replace pbp2X . However , the only transformant to have one of these genes ( pbp1A ) completely replaced was still able to grow , albeit at a greatly reduced rate , on ampicillin . Less than 0 . 1% of selected recombinations encompassing the cps locus would be expected to replace both pbp2X and pbp1A in their entirety , appearing to explain the lack of a readily observable inhibition of cps transfer by ampicillin selection . However , this is an oversimplification , as recombinant sequences can be found at high density around the selected recombination , where they are likely to affect the pbp1A and pbp2X genes . Thirty-six further RSSs occur in the primary locus but do not span the cps gene cluster ( Figure 2B ) . This concentration of unselected primary recombinations suggests that they are associated with that spanning the cps locus , forming a non-contiguous recombination ( NCR ) as has previously been observed on a smaller scale in the generation of penicillin-resistant pbp alleles in vitro [34] . This may represent a consequence of the recipient genome's properties , with restricted regions of the chromosome capable of undergoing recombination at high rates perhaps due to local chromatin arrangements or supercoiling properties . Under such circumstances , the flanking RSSs may arise from different donor molecules to the selected RSS . Alternatively , it may result from a single donor molecule being integrated in a non-contiguous manner , in which case these flanking events will have arisen from the same imported DNA as the acquired selected marker . The latter hypothesis is supported by the observation that strains transformed with the lower concentration of DNA actually have a larger mean number of primary RSSs ( 1 . 52 per strain ) than those exposed to the higher concentration ( 1 . 33 per strain ) , although this difference is not significant ( Wilcoxon test , p = 0 . 33 ) . This higher density of recombinations in the cells exposed to a lower concentration of donor DNA is the opposite of what would be expected if donor molecules could independently contribute to the genetic exchange within this locus , indicating that a single piece of DNA acts as the origin for multiple RSSs . While none of the selected RSSs affected the pbp genes to a large extent , two and four unselected primary RSSs overlapped with pbp2X and pbp1A , respectively , demonstrating how mosaicism complicates the issue of linkage between loci and increases the likelihood of capsule switches being associated with changes in penicillin resistance . A histogram of the L50R values for recombinations within the primary locus reveals the selected and unselected RSSs have very different size distributions ( Figure 4A ) . The unselected RSSs are typically less than 5 kb in length , and follow an approximately exponential distribution with a rate constant , λR , of 4 . 96×10−4 bp−1 ( 95% confidence interval 3 . 60–7 . 23×10−4 bp−1 ) . Those spanning the cps locus are longer ( and therefore estimate λR as being smaller ) , due to the necessity of spanning the 21 , 373 bp gene cluster , and follow a different size distribution , as the probability of transmitting the selectable marker to the recipient rises as the length of the recombination increases [47] . The size distribution of the imported strand , which can be calculated on the basis that all primary RSSs originate from the same DNA molecule , is similar ( Figure 4B ) . The median L50D of these values was found to be 5 . 9 kb , comparable to the median imported strand length of ∼6 . 6 kb [9] despite the size constraint of importing the kanamycin resistance gene . The size distribution of the secondary RSSs was also found to resemble an exponential distribution ( Figure 4C ) ; the larger number of datapoints in this set allowed this assessment to be confirmed quantitatively ( Figure S2 ) . The implied λR of 4 . 40×10−4 bp−1 ( 95% confidence interval 3 . 71–5 . 38×10−4 bp−1 ) is similar to that deduced from the unselected primary recombinations and reflects the mean length of these RSSs , 2 . 27 kb . To test whether they also exhibited the same mosaicism as observed in the primary locus , a bootstrapping analysis ( see Methods ) was used to compare distances between RSSs within the same isolate against the distribution of distances between RSSs in different isolates . This test was designed to find RSSs in an isolate that were significantly closer together than would be expected by chance , indicating that they are likely to have originated from the same important molecule of DNA , and link them together into NCRs . The outcome grouped the 107 secondary RSSs into 87 NCRs , which contained up to three segments of imported sequence; like the component RSSs , NCRs were significantly more frequent in strains transformed at a high concentration of DNA ( Wilcoxon test , p = 5 . 68×10−9 ) . Those secondary NCRs composed of more than one RSS were estimated to be composed of approximately half donor sequence ( mean proportion of 55% ) , with the remainder unmodified recipient sequence . However , the L50D lengths of the NCRs did not appear to fit an exponential distribution as well as the RSS lengths ( Figure 4D ) . If they had , it might have suggested each NCR was formed through a Poisson process , with the component RSSs produced through post-processing of this larger structure; as it is , the data indicate the RSSs are formed through a Poisson process , with the composite NCRs more irregular in their form . Being distributed around the chromosome , the secondary RSSs provided an opportunity to test whether sequence similarity between the donor and recipient affects the positioning of genetic exchanges in this system . Dividing the entire recipient genome into 1 , 000 equally-sized non-overlapping windows produced 890 outside the primary locus that were found to contain at least one marker SNP , and therefore had the potential to contain a detectable recombination . Comparing the distribution of mean SNP sequence identities ( calculated using 50 bp of aligned sequence on each side ) of all such windows against the subset of 168 that were found to overlap with secondary RSSs ( Figure 5 ) provided no evidence for the enrichment of imported sequence in regions of the genome with greater sequence identity between donor and recipient . A ‘walking hypergeometric test’ performed at different sequence identity thresholds confirmed the absence of a statistically significant relationship , which concurred with a more detailed sliding window analysis ( see Text S1 ) . The previously reported effects of sequence identity may reflect the necessity for MEPS extending beyond a threshold length , estimated at a total across both ends of 27 bp for S . pneumoniae [27] . The results of this experiment are almost entirely consistent with this constraint , except for a single RSS that transfers one SNP and has BRs totalling 26 bp in length . However , the sliding window analysis failed to find a significant enrichment of RSSs with BRs that were longer than expected by chance alone ( see Text S1 ) . This is likely to reflect the low density of markers across the recipient genome relative to the MEPS threshold , suggesting that the frequency of polymorphisms across the pneumococcal genome in this experiment is not high enough to strongly inhibit most transformation events . One other possible characteristic of the genome that might influence the positioning of recombinations is the presence of small interspersed repeats . However , comparing the density of BOX , RUP and SPRITE sequences [48] within and outside secondary RSSs failed to uncover any significant associations ( Table 1 ) . Past studies focussing on individual loci have found that MMR strongly influences the transfer of small number of polymorphisms [12] , [14] , [30] . However , as just a few hundred SNPs appear to be sufficient to overwhelm the pneumococcal MMR system [26] , it is not clear whether this this form of repair is likely to have extensively impacted on recombinations in this system . To test this , the number of each type of SNP outside the primary locus was compared with the frequencies of these mutations in the secondary recombinations , resulting in the observation of a tight correlation ( Pearson correlation R2 = 0 . 99 , p = 2 . 83×10−12; Figure 6 ) . Furthermore , the frequency of each SNP on the outermost position of each RSS is proportional to its prevalence in the nearest flanking unchanged position ( the two positions defining BRs; Pearson correlation R2 = 0 . 92 , p = 7 . 39×10−7 ) . Hence there is no evidence that the low efficiency markers , acted on most effectively by MMR , lead to entire recombinations being lost at a higher frequency , or that they trigger localised repair , which might have been a mechanism for the formation of NCRs . These graphs do show , however , that the most frequent mutations found distinguishing the donor and recipient strains , independent of the observed transformations , are the transversions , which are targeted most efficiently by mismatch repair . Contingent upon the majority of these polymorphisms having arisen through single step mutations , this supports the hypothesis that MMR has evolved to repair the most frequent mutations most efficiently [49] . By contrast , insertions and deletions ( indels ) appear to have a stronger effect on the positioning of RSSs . Within secondary RSSs , insertions ( relative to the recipient ) are underrepresented , although not to a statistically significant extent ( Table 1 ) . However , these insertions within RSSs are significantly smaller than the rest of the identified insertions outside the primary locus that are not transferred in this experiment ( Wilcoxon test , p = 0 . 015 ) . However , deletions relative to the recipient within secondary RSSs show no such significant difference in sizes compared to those outside transformed sequences ( Wilcoxon test , p = 0 . 86 ) . This is not likely to be an artefact of the overall distribution of deletion and insertion sizes , which do not differ significantly ( Wilcoxon test , p = 0 . 32 ) . Conversely , deletions were found to be significantly underrepresented in BRs ( Table 1 ) , and both insertions and deletions in BRs were significantly longer than those outside recombinations ( Wilcoxon tests , p = 0 . 0010 for insertions and p = 0 . 027 for deletions ) . Manual inspection of the mapped sequence read data found that the majority of these indels remained of the recipient allele and therefore had not been cotransfered with the neighbouring RSS . Combining these results concerning BRs suggests that small differences in length between the invading strand and recipient genome may strongly inhibit the formation of recombination edges , implying that indels can only be maintained in a heteroduplex if stabilised on either side by loci of matching lengths in the donor and recipient genotypes . In order to experimentally verify the apparent lack of a role for MMR in moderating the exchange of sequence in this in vitro system , and ascertain whether it may be involved in any localised repair mechanism responsible for the formation of NCRs , a transformation experiment was performed using the same donor DNA , at 500 ng mL−1 , and selection on kanamycin alone , with an MMR-deficient recipient ( S . pneumoniae 23F-RΔhexB ) alongside the wild type strain . Twenty isolates of each genotype were analysed as described for the first experiment ( Figure 7 ) , resulting in the identification of 1 , 463 polymorphic sites . Of these , 92 did not correspond to marker SNPs; the majority of these were found in the MMR-deficient strains , suggesting the loss of the repair functionality resulted in an increased rate of point mutation accumulation during culturing . Identification of recombinant sequence demonstrated that there was no significant difference between the number of primary RSSs in the wild type ( mean of 1 . 50 per strain ) and MMR-deficient ( mean of 1 . 65 per strain; Wilcoxon test , p = 0 . 90 ) strains . This indicates that MMR does not appear to have a role in generating the mosaic structure of recombinations . This is congruent with both the previous observations that the low efficiency markers are found at equal frequencies on either side of recombination boundaries , indicating these polymorphisms are not targeted by localised MMR , and that the distributions of secondary RSS lengths fit an exponential distribution more closely than those of the secondary NCR lengths , implying the RSSs form independently rather than emerging through partial repair of NCRs . Furthermore , there was no difference in the mean number of secondary RSSs between the wild type ( mean of 0 . 85 per strain ) and MMR-deficient ( mean of 1 . 00 per strain; Wilcoxon test , p = 0 . 94 ) isolates or in the mean lengths of these secondary recombinations ( Wilcoxon test , p = 0 . 71 ) . This confirms that under these conditions , with the genotypes used in this experiment , MMR does not appear to be moderating transfer of sequence between donor and recipient .
Many previous studies of transformation have necessarily focused on the rates of transfer of a small number of selectable mutations . The advantage of high-throughput genome sequencing , coupled with a controlled in vitro system , is the ability to characterise the full extent of multiple recombinations occurring throughout the chromosome . Using these data , it can be observed that multiple RSSs can be generated from a single donor molecule of DNA , thereby forming NCRs . Furthermore , multiple NCRs , each from a different donor molecule , can be generated during a single period of competence . Recent work on a smaller sample of Haemophilus influenzae transformant genomes [50] concurs with this conclusion , as well as the lack of effect of sequence diversity on the distribution of recombinations , although unfortunately the study could not evaluate these points statistically due to sample size constraints . However , the recombination lengths in the H . influenzae transformants appear to follow a different distribution to those described here in S . pneumoniae . The exponential length distribution apparent from this work suggests the boundary of the event is determined by random termination with a fixed per-base probability of λR . The relatively weak observed linkage between the cps locus and the pbp1A and pbp2X genes is a consequence of this size constraint . It is not clear what mechanism is responsible for this Poisson process: it seems likely to be the resolution of the heteroduplex intermediate , but could alternatively represent a process such as cleavage of the imported strand . The consequent exponential distribution has potentially interesting implications for bacterial evolution , as it suggests that transformation is optimised for transferring short lengths of similar sequence , rather than replacing large genetic features , such as operons , with homologous sequence in a single event . The calculated value for λR suggests over a third of RSSs will be less than 1 kb , the approximate size of a typical bacterial gene . This is likely to have been crucial in facilitating the generation of mosaic penicillin-resistant pbp genes through incorporating fragments of penicillin-sensitive S . mitis and S . oralis orthologues into the penicillin-sensitive S . pneumoniae versions [51] , [52] . Longer imports would have simply exchanged one sensitive allele for another . It will be interesting to observe whether proteinaceous antigens in S . pneumoniae display a similar mosaicism as a result of transformation . Does this mean large events , such as serotype switching , will occur only rarely in the population ? This depends , to a great extent , on the important consideration of whether the λR per base probability applies to the nucleotides of the incoming strand , the recipient genome , both , or only those bases that are paired between the incoming and resident DNA ( of course , there may be other effects of indels on heteroduplex stability that are independent of λR ) . This will largely determine how recombination behaves with regard to indels . If λR applies to the incoming DNA , then recombinations would presumably abort when terminated within an insertion in the donor relative to the recipient , due to the absence of a homologous locus in the recipient chromosome . This would lead to insertions being underrepresented in RSSs at a level related to their length . Conversely , if λR applies to the host DNA bases , then deletions in the donor relative to the recipient will be underrepresented in an analogous fashion . However , if λR applies either to only bases that pair in the formation of recombinant segments , or equally to insertions in both genotypes , then the Poisson process would not cause any asymmetry between insertions and deletions in the donor . Yet there would be an important difference at sites where one gene cassette replaces another ( e . g . at the cps locus ) , as in the former case the presence of the indels is potentially negligible , whilst in the latter case the lengths of the incoming and resident alleles would both contribute to inhibiting the transfer . However , it would be expected that an overall bias against insertions in the donor , relative to deletions , would be observed in both cases , as the physical size of the insert on the donor DNA makes it susceptible to cleavage upon import , whereas deletions cannot be degraded upon entry in the same way [53] . Previous work in S . pneumoniae has found , in reciprocal crosses , that length polymorphisms are imported more efficiently as deletions rather than as insertions [24] , [25] , [54] , although a more recent study did not replicate this result [53] . This is congruent with any of the outlined scenarios , except that where the Poisson process involves bases of insertions in the recipient but not the donor . In this context , it may be informative to note that the longest predicted homologous recombination found in the sequencing of over 200 PMEN1 strains was an event that deleted the ∼38 kb genomic island encoding the psrP antigen [35] , suggesting its extreme length may be explained through the bases of the insertion in the recipient not being taken into account by the Poisson process that determines the length of RSSs . The data described in this work found that RSSs were less likely to contain long insertions than deletions , in agreement with the aforementioned studies that suggest a bias against the acquisition of insertions . This suggests that the net effect of imported DNA cleavage , heteroduplex formation and resolution of the recombination is a tendency towards importing shorter alleles , rather than gain of longer sequences . If correct , when combined with the exponential distribution of recombination sizes , this would mean a ‘molecular drive’ would exist that would lead to the fixation of the shortest allele at a locus in the absence of selection for the sequences being deleted . This may provide an explanation for the observed , but not fully explained , deletional bias among prokaryotes [55] , and suggests that the maintenance of multiple alleles that are polymorphic in length , as observed with the different capsule biosynthesis gene clusters at the cps locus , should be rare in a transformable bacterial population . However , exactly how infrequent they are expected to be , in the absence of selection , cannot be determined until more precise details of the transformation mechanism are understood . The lengths of the recombinations changing capsule type in the PMEN1 clinical isolates can be compared with those selected in this work through applying the algorithm used to detect sequence imports in the PMEN1 strains [35] to the in vitro transformants . In these comparable datasets , the distribution of cps-spanning recombination sizes is very similar in both ( Figure 8A ) . Elsewhere in the genome , the distributions of recombination sizes are alike ( Figure 8B ) , but the set detected in the PMEN1 population has fewer short recombinations . This is likely to be due to a reduced power to differentiate such events , which import few SNPs , from the background of point mutations in the clinical isolates , as compared to the near-complete absence of variation outside of transformation events from the isolates in this study . Hence the exponential decline is less clear from the data derived from clinical isolates , and the rate parameter λR is estimated as being artefactually low ( 1 . 58×10−4 bp−1; 95% confidence intervals 1 . 47–1 . 72×10−4 bp−1 ) . Overall , recombinations were found to affect almost three-quarters of the reference genome length in the PMEN1 sample , which is congruent with the apparently random positioning of RSSs seen in this experiment . This would appear to be a consequence of the divergence between donor and recipient ( a mean density of one per 96 bp of sequence aligned between the donor and recipient ) . The import of sizeable DNA fragments from a genotype of such diversity into the studied recipient may be expected to include sufficient SNPs to overwhelm the MMR system [26] , while not containing a high enough density of polymorphisms to commonly restrict the resolution of a heteroduplex [27] . Hence both these potentially restrictive processes seem likely to be relatively uninfluential at this level of sequence divergence , with no evidence of either observed in this experimental setup . Some factors need be considered before applying these results to all pneumococcal transformations . For instance , it may be that in the absence of selection for such a large recombination at the cps locus , MMR may be more of a barrier to sequence transfer at low concentrations of DNA , when the cell imports less genetic material and therefore fewer polymorphisms . The disruptive effects of the selected transformation may also have been accentuated by the large different in length between the incumbent cps locus and small selected marker . Additionally , it is possible the experimental conditions used may not have induced expression of the MMR system in the recipient strain; it has been observed that pneumococci vary in the rate at which they import sequence [56] , which could potentially reflect differences in the overall activity or regulation of MMR in different strains . The use of other genotypes may have lead to stronger signals of MMR being identified in this experiment . The level of diversity found distinguishing the donor and recipient will also affect the pattern of observed transformation events , although the mean divergence between the strains in this experiment is typical of that observed when comparing pneumococci [41] . Aligning all available complete pneumococcal genomes to the recipient reveals a minimum and maximum SNP density of one SNP per 150 bp ( S . pneumoniae JJA ) and one SNP per 81 bp ( S . pneumoniae Hungary 19A-6 ) , respectively . By contrast , a comparison with S . mitis B6 [57] with the recipient sequence reveals a SNP density of one per 23 bp , enough to frequently interfere with the formation of a recombination edge , according to the MEPS model . Hence while the constraints of MMR and sequence diversity will impact on within-clone and between species transfers respectively , the positioning and selective advantage of accessory genome loci are likely to have the greatest mechanistic impact on homologous recombination between lineages of the pneumococcal species .
All statistical analyses were performed using R . The “fitdistrplus” package was used to find distributions to experimental data and establish 95% confidence intervals through bootstrapping analysis . The “grofit” package was used to extract statistics from growth curves . As data analysed in this study tended not to be normally distributed , non-parametric statistical tests were used throughout . The TIGR4Δcps strain was previously created through replacing the cps locus of S . pneumoniae TIGR4 with a kanamycin resistance marker [40] . A draft genome assembly for strain TIGR4Δcps was generated by using the sequence of the disrupted capsule biosynthesis locus reported following the construction of this strain ( EMBL accession code AF160759 ) to replace the serotype 4 capsule locus of the TIGR4 genome ( multilocus sequence type 205; EMBL accession code AE005672 ) [58] . This hybrid sequence served as an initial draft of the TIGR4Δcps genome , which was then corrected with ICORN [59] using Illumina sequence data generated from the donor DNA used in the transformation experiments . This resulted in the correction of 56 base substitutions and 20 single base indels . The sequence of S . pneumoniae 23F-R was derived by correcting the reference sequence of S . pneumoniae ATCC700669 ( multilocus sequence type 81; EMBL accession code FM211187 ) [39] with ICORN using resequencing data [35]; this revealed the presence of eight substitutions , one single base indel and the loss of prophage ΦMM1-2008 . The finalized genomes were aligned using MUGSY [60] resulting in 21 , 541 marker polymorphisms being identified . To avoid the false positive identification of transformation events , polymorphisms within the four rRNA operons , the hypervariable hsdS locus [58] and highly repetitive psrP gene [39] were excluded from all analyses . This left 20 , 773 marker SNPs for use in detecting RSSs . In order to remove any of these sites that would be liable to lead to the false positive inference of recombinations when analysing Illumina sequence reads , the data generated from the donor DNA ( used to correct the sequence of strain TIGR4Δcps ) were mapped to the corrected sequence of the recipient , strain 23F-R . Using the criteria described in Harris et al . [42] , 17 , 534 SNPs could be identified , all of which , except 385 false positives , corresponded to polymorphisms identified through the whole genome alignment . The false positive positions were excluded from subsequent analyses to avoid erroneous inference of transformation events . An evaluation of the SNP calling parameters was performed using these data; this found that the identification of polymorphisms did not alter greatly except when using the most extreme choices of parameter selection ( Figure S3 ) . Agarose gel electrophoresis was used to check that donor DNA used in these experiments was not heavily degraded . Overnight cultures of S . pneumoniae 23F-R were diluted 1∶100 into BHI ( Oxoid ) and grown statically at 37°C to an optical density of OD600 between 0 . 20–0 . 25 . A 1 mL sample of the culture was then added to 10 ng competence stimulating peptide 2 ( CSP-2; Sigma ) , 5 µl 500 mM calcium chloride and 5 µl of an aqueous solution of donor DNA , at a concentration of either 1 µg mL−1 or 100 µg mL−1 . These reactions were incubated at 37°C for 2 h , and then three 50 µL volumes each spread onto 5% horse blood agar plates supplemented with 200 g L−1 kanamycin ( Gibco ) , and three further 50 µL volumes each spread onto 5% horse blood agar plates supplemented with 200 g L−1 kanamycin and 100 mg L−1 ampicillin ( Sigma ) . Colonies were counted and picked after 16 h incubation at 37°C . The first experiment involved six transformations at the lower concentration of DNA and three transformations at the higher concentration , along with a negative control containing no CSP-2 that generated no kanamycin-resistant colonies . DNA was prepared from randomly-picked colonies , and sequenced as multiplexed libraries of 12 tags on an Illumina Genome Analyzer II as described in Croucher et al . [35] to give 76 nt paired end reads . The second experiment involved three transformations of the wild type and mutant strains as described above . DNA was prepared using an Xtractor Gene system ( Qiagen ) and sequenced as multiplexed libraries of 24 tags on an Illumina HiSeq , according to manufacturer's instructions , to give 76 nt paired end reads . Four datasets each from the wild type and mutant groups were discarded as they showed potential signs of contamination . For each of the 84 sequenced transformed strains from the first experiment , along with the donor and recipient strains , an overnight culture growing in 10 mL BHI ( Oxoid ) was diluted 1∶100 into 5 mL sterile BHI . A 300 µL sample of each freshly inoculated culture was then immediately transferred into a 96 well plate , and growth was followed for 12 h at 37°C by measuring the OD600 of each well using a FLUOstar Omega microplate reader . Three independent biological replicates of this experiment were performed . Regions around 500 bp in length on either side of hexB , both including BmgBI cut sites encoded by BOX elements , were amplified by PCR ( all primer sequences are listed in Table S3 ) and cloned into pGEM-T Easy , according to manufacturer's instructions , to give plasmids pLeft and pRight . The ermB erythromycin resistance gene was then amplified from strain S . pneumoniae 11930 [35] using the primers ermBL and ermBR ( each with an EcoRV cut site on the 5′ end ) , and cloned into pGEM-T Easy as described above to give plasmid pErm . All three plasmids were then transformed into E . coli TOP10 cells ( Invitrogen ) through electroporation with a 2 . 5 kV pulse , followed by blue-white selection on 100 µg mL−1 ampicillin . The plasmid carrying the ermB gene was then extracted using a QIAprep Spin Miniprep Kit ( Qiagen ) and digested using EcoRV ( New England Biolabs ) , releasing the insert with blunt ends . This fragment was purified through agarose gel electrophoresis using a QIAquick Gel Extraction kit ( Qiagen ) . Plasmid pRight was similarly extracted , then digested with BmgBI ( New England Biolabs ) , which also cuts to give blunt ends . The ermB fragment was cloned into this blunt cut site using T4 ligase ( Promega ) overnight at 4°C according to manufacturer's instructions . A 1 µL sample of this reaction was then used to transform electrocompetent E . coli TOP10 cells as described above . A 50 µL sample of this culture was then spread on LB agar plates supplemented with 100 µg mL−1 erythromycin . The plasmid pErmR was isolated from a colony , and its composite insert and the adjacent multiple cloning site ( MCS ) amplified using primers ermBL ( which has an EcoRV cut site on the 5′ end ) and T7 . This amplicon was purified through agarose gel electrophoresis as described above and digested with EcoRV and NcoI ( New England Biolabs ) , the latter of which cuts within the MCS of pGEM-T Easy . This was ligated into pLeft following its digestion with BmgBI and NcoI as described above . This ligation reaction was used to transform E . coli TOP10 cells as described above , with strains carrying the plasmids again selected on LB agar supplemented with 100 µg mL−1 erythromycin . This gave pErmLR , which was used to transform S . pneumoniae 23F-R . Pneumococcal transformants with disrupted hexB genes were selected on 5% blood agar plates supplemented with 0 . 1 µg mL−1 erythromycin . One of these colonies was picked and the insert checked through PCR amplification with primers hexBL and hexBR and capillary sequencing . Further confirmation was provided through de novo assembly of Illumina sequence reads generated from 23F-RΔhexB transformant DNA . This scheme is summarised in Figure S4 . Illumina data from transformants were independently mapped to the corrected genomes of 23F-R and TIGR4Δcps using SMALT [61] . The de novo identification of SNPs , to check for mutations not derived from imports of the donor sequence , was performed by identifying polymorphisms relative to the recipient genome as described in Harris et al . [42] . All strains had a mean sequence read coverage of at least 20 fold , with the mean coverage across the dataset of 123 fold , for strains sequenced using the Genome Analyzer II , and 506 fold , for strains sequenced using the HiSeq . To identify RSSs , only marker SNP sites were analysed . The alignment produced by SMALT was processed using samtools [62] and vcftools [63] to produce Variant Call Format ( VCF ) outputs for each strain . Within these VCFs , all homozygous marker sites with a base quality greater than 50 were used to identify sequence characteristic of the donor or recipient in the transformation experiment . Recombinations were initially defined as regions containing donor alleles at polymorphic sites with no intervening recipient alleles . To exclude false positives , putative recombinations were rejected unless there was evidence of the transfer of at least one of the supporting marker sites from the reciprocal mapping to the donor sequence , similar to the method described by Mell et al . [50] . The ambiguous boundary regions around each recombination extended between the outermost donor allele SNPs identified in the transformant and the nearest flanking sites that were found to have recipient allele SNPs . All sequenced transformants were found to have a unique pattern of transformation events , and would therefore appear to represent independent samples of the variation produced the transformation experiments . Many secondary RSSs in the same strain were positioned very close to one another , suggesting they may have arisen from the same molecule of donor DNA . A bootstrapping approach was used to test whether such apparent spatial associations were significant . To produce a population of test values , the shortest distance ( in terms of the donor genome ) between all secondary RSSs from all strains in the first experiment was calculated . Then , for each strain with more than one secondary RSS , the shortest distance between the two closest secondary RSSs in the strain was calculated , in terms of the donor sequence , to give the test value ( dtest ) . A bootstrapped sample , equal in size to the test population , was then obtained from the test population , and this distribution used to test the hypothesis that dtest was significantly shorter than expected under the null hypothesis ( H0 ) of recombinant sequences being positioned at random relative to one another . Multiple testing was accounted for by using a Holm-Bonferroni correction to alter the one-tailed threshold p value of 0 . 05 according to the number of secondary RSSs in the strain . One hundred such bootstrap tests were performed for each for each dtest , with distances rejecting H0 on 95 or more of these trials considered to be significantly close to one another and therefore likely to have arisen from the mosaic incorporation of the same molecule of donor DNA . If the first dtest in a strain was accepted , then the second smallest distance was tested with the appropriate corrected p value threshold; this process continued until dtest was accepted under H0 . Where secondary RSSs were significantly closer than expected under H0 , they were grouped into NCRs . | Transformation is the process by which cells take up DNA from the environment and integrate it into their genome . It was first observed in the bacterium Streptococcus pneumoniae , a common cause of pneumonia and meningitis . This ability has allowed S . pneumoniae to evolve resistance to penicillin and to change its surface antigens to evade vaccines . To characterise this process in detail , we transformed an S . pneumoniae strain in the laboratory and sequenced the genomes of the resultant mutants . This showed that multiple recombinations , arising from different molecules of imported DNA , could occur around the genome at the same time . Some individual imported molecules donated multiple segments of sequence into the recipient . The positions of the recombinations were not significantly affected by the level of sequence similarity between donor and recipient , as had previously been observed for transfers between species , or by the mismatch repair process , which had previously been found to inhibit the transfer of small numbers of mutations . The recombinations' lengths were exponentially distributed with a mean length of 2 . 3 kb . This implies that the smallest version of a genetic locus will become the most common in the bacterial population in the absence of selection for a longer alternative . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"genome",
"sequencing",
"genomics",
"genome",
"evolution",
"chromosome",
"biology",
"genetics",
"population",
"genetics",
"biology",
"evolutionary",
"biology",
"genomic",
"evolution",
"evolutionary",
"genetics",
"genetics",
"and",
"genomics"
]
| 2012 | A High-Resolution View of Genome-Wide Pneumococcal Transformation |
Glucose is the main energy substrate in the adult brain under normal conditions . Accumulating evidence , however , indicates that lactate produced in astrocytes ( a type of glial cell ) can also fuel neuronal activity . The quantitative aspects of this so-called astrocyte-neuron lactate shuttle ( ANLS ) are still debated . To address this question , we developed a detailed biophysical model of the brain’s metabolic interactions . Our model integrates three modeling approaches , the Buxton-Wang model of vascular dynamics , the Hodgkin-Huxley formulation of neuronal membrane excitability and a biophysical model of metabolic pathways . This approach provides a template for large-scale simulations of the neuron-glia-vasculature ( NGV ) ensemble , and for the first time integrates the respective timescales at which energy metabolism and neuronal excitability occur . The model is constrained by relative neuronal and astrocytic oxygen and glucose utilization , by the concentration of metabolites at rest and by the temporal dynamics of NADH upon activation . These constraints produced four observations . First , a transfer of lactate from astrocytes to neurons emerged in response to activity . Second , constrained by activity-dependent NADH transients , neuronal oxidative metabolism increased first upon activation with a subsequent delayed astrocytic glycolysis increase . Third , the model correctly predicted the dynamics of extracellular lactate and oxygen as observed in vivo in rats . Fourth , the model correctly predicted the temporal dynamics of tissue lactate , of tissue glucose and oxygen consumption , and of the BOLD signal as reported in human studies . These findings not only support the ANLS hypothesis but also provide a quantitative mathematical description of the metabolic activation in neurons and glial cells , as well as of the macroscopic measurements obtained during brain imaging .
The mammalian brain exhibits remarkable processing power . It is at the same time energy efficient . The design features that allow such efficient computation are mapped in cellular and molecular components and their roles in information processing . Concurrently , these features are anchored in , and constrained by , the universal metabolic chains that provide energy to cells . Deciphering the metabolic code and the neural code are thus tandem requirements for a comprehensive understanding of brain function . Understanding the metabolic underpinnings of information processing is also of added value to understanding the etiology and progression of neuropsychiatric and neurodegenerative disorders [1 , 2] . The picture that emerges from this dynamical system will reflect the cooperative function of neurons , glia and the vascular system . Glutamate , the brain’s major neurotransmitter , effects numerous cascades and processes in brain cells [3 , 4] . Among them , astrocytes couple synaptic activity to energy metabolism via a sodium-dependent uptake of glutamate [5] . The ensuing cascade of molecular events leads to the glycolytic processing of glucose and the release of lactate by astrocytes . A comprehensive model of brain energy metabolism must consider oxidative and non-oxidative glucose consumption , intracellular and extracellular compartmentalization and transport of choke-point metabolic intermediates such as lactate and pyruvate , as well as feedback mechanisms that report local synaptic and intrinsic neuronal activity [6 , 7] . These pathways are in turn complicit in the molecular and cellular mechanisms that contribute to the still poorly understood read-out of functional brain imaging [8] . The role of astrocytes and how they metabolically interact with neurons is well supported experimentally; some mostly theoretical considerations , however , have challenged this view . Magistretti and colleagues proposed that clearance of glutamate from the synaptic cleft by astrocytes could be coupled to glycolysis and subsequent lactate production [5] . Lactate produced in this way would then be transported to the extracellular space . Controversy remains surrounding directionality and timing of lactate flow in the brain; while a neuron-to-astrocyte lactate system ( NALS ) is proposed by some [9 , 10] , an astrocyte-to-neuron direction ( ANLS ) is supported by a large set of experimental evidence [11] . Biophysical models also weigh-in on the conditions and sequences of events required for lactate production and consumption [12–15] . The existence of an extracellular pool of lactate likely used as an energy reservoir at the onset of stimulation has been observed in rats and humans [16 , 17] . The distribution of monocarboxylate transporters at the membrane of neurons and astrocytes supports the hypothesis of a net transfer of lactate from astrocytes to neurons through the extracellular space [18] . Glial cells have been observed to take up most glucose [19 , 20] , while neurons are responsible for the largest part of brain oxygen consumption [21 , 22] . Additional evidence comes from the direct measure of NADH transients in brain slices , showing that neurons display early oxidative metabolism following presynaptic activity , while astrocytes display a delayed activation of glycolysis but no detectable oxidative response [23] . Nevertheless , the ANLS-hypothesis is still debated and challenged with arguments focusing now on the exact interpretation of the above observations [24 , 25] . Nicotine adenine dinucleotide , either oxidized or reduced ( NAD+ , or NADH ) , is a workhorse cofactor that acts as a central electron broker for metabolic redox cycles including glycolysis , the citric acid cycle ( Krebs , TCA ) and oxidative phosphorylation . Owing to its high UV wavelength absorption , it is also responsible for cellular auto-florescence . This coincidence makes it a useful indicator of metabolic activity . NADH is an important metabolic signal because it is produced or used during both mitochondrial activity and activation of the glycolytic pathway , and because it cannot diffuse freely through the mitochondrial membrane but needs to be transported by appropriate shuttles . Fluctuations of the NADH concentration measured in the appropriate cellular compartments can then indicate increased or decreased oxidative and glycolytic metabolism . A critical previous finding in this regard was the observation of early and late activity-dependent phases of metabolic activity with the early phase taking the form of a NADH “dip” and the late phase appearing as a NADH “overshoot” with a longer time constant of decay [23] . Interestingly , these phases also correlate with the fluctuations of the extracellular lactate concentration as determined in animals [16] and humans [26 , 27] . The emerging consensus is that the early phase represents NADH depletion in the dendrites of active neurons and that the overshoot represents glycolytic activity that results in the accumulation of NADH . This activity results in the high production of lactate in astrocytes as rapid glycolysis overtakes the subsequent consumption by oxidative pathways [23 , 26 , 27] . The accumulation and transportation of lactate between glial cells and neurons may in turn serve as an activity-dependent buffer that is informed by the neuronal release and glial uptake of glutamate [5] . It might also act as a signaling molecule to the vasculature [28] or to brain cells via binding to the G-protein coupled receptor GPR81 [29] . The preference for lactate over glucose as an energy substrate in neurons has been demonstrated in vivo as well as in vitro [30 , 31] , as has a neuroprotective role for lactate in the case of insulin-dependent hypoglycemia [32] and other conditions [33 , 34] . The role of the ANLS in homeostatic maintenance involves the regulation of blood glucose [35] and sodium [36] . While the ANLS hypothesis , since its initial formulation [5] , does not preclude the use of glucose by neurons as an energy substrate , it has been challenged by some studies defending the view that glucose , rather than lactate , is the sole energy substrate for oxidative metabolism in neurons [37 , 38] . Previous modeling efforts have advanced our knowledge of this functional metabolic network by demonstrating that lactate consumption by neurons occurs early in the stimulus regimen and that the early and late lactate transients correspond to the activity of two distinct populations of cells , neurons and glia . The current study builds on and complements those and other models [9 , 10 , 12–15 , 39–42] and addresses unresolved mechanisms of neuron-glial metabolic and vascular coupling . The model is based on several previous studies [13 , 40] with five significant improvements: 1 ) the compartmentalization of NADH between cytosolic and mitochondrial compartments; 2 ) the linking of metabolic and Hodgkin-Huxley formalisms; 3 ) the input to the neuronal and astrocytic compartments formulated as a presynaptic glutamatergic stimulation; 4 ) the model explicitly and continuously updates reversal potentials; and 5 ) the model was constrained using in vitro data and correctly predicts in vivo results without the need of invoking glycogen ( which is deliberately excluded from the model ) . In this paper , we will show that a biophysical model of astrocyte-neuron metabolic interactions designed following these principles leads to the presence of an activity-dependent lactate shuttle from astrocytes to neurons and that this model can reproduce the evoked response of NADH in its various compartments as reported by Kasischke and colleagues [23] . We will subsequently show that our biophysical model correctly predicts qualitatively—and to some extent quantitatively—the evoked responses of tissue lactate and tissue oxygen as observed in the rat brain in vivo [16] . Finally , our biophysical model predicts the evoked responses of tissue lactate , of the BOLD signal and the glucose and oxygen consumption as observed in the human brain in vivo .
ADPx is given as a function of the ATP concentration ( x stands for n or g ) . It reads: A D P x = A T P x 2 [ − q A K + q A K 2 + 4 q A K ( A / A T P x − 1 ) ] ( 1 ) with A = AMPx+ADPx+ATPx = 2 . 212 mM the total adenine nucleotide concentration and qAK = 0 . 92 the adenylate kinase equilibrium constant [40 , 45] . As a consequence: d A M P x d A T P x = − 1 + q A K 2 − 1 2 u x + q A K * A A T P x u x ( 2 ) with ux = q2AK+4·qAK· ( A/ATPx-1 ) . The model receives input from a presynaptic excitatory population . Glutamate released by excitatory presynaptic neurons drives the intracellular sodium concentration in neurons and astrocytes and activates AMPA receptors on neurons , thus inducing a synaptic current Isyn . The presynaptic population contains Nexc excitatory neurons discharging at frequency fexc ( t ) . This presynaptic population thus generates an excitatory conductance gexc ( t ) given by: g e x c ( t ) = N e x c g ¯ f e x c ( t ) ( 3 ) with g = 7 . 8·10-6 mS·cm-2·sec the total surface under the conductance evoked by one excitatory event [47 , 48] . The corresponding synaptic current is then given by: I s y n ( t ) = g e x c ( t ) ( ψ n − E A M P A ) ( 4 ) with ψn the neuronal membrane voltage and EAMPA = 0 mV the reversal potential of AMPA ionotropic receptors . It is estimated that about two thirds of the current generated at AMPA receptors is due to a flow of sodium ions [49] . Sodium also flows through voltage-dependent sodium channels when the neuron is active ( INa ) . As a consequence , the sodium drive to the neuron is approximated by: J s t i m n = S m V n F ( 2 3 I s y n − I N a ) ( 5 ) with Sm·Vn = 2 . 5·104 cm-1 the ratio between neuronal membrane surface and neuronal volume and F = 9 . 64853·104 C·mol-1 the Faraday constant . Finally , the glutamate is cleared from the synaptic cleft by excitatory amino acid transporters located on the astrocyte membrane . Those transporters use the electrochemical sodium gradient to transport glutamate with a stoichiometry of three sodium ions for one glutamate molecule . We thus write the sodium drive to the astrocyte as follows: J s t i m g = 3 Δ g l u t N e x c f e x c ( t ) ( 6 ) with ∆glut = 2 . 25·10-5 mM a constant , which corresponds to the total amount of glutamate released in the synaptic cleft by each presynaptic action potential multiplied by the ratio between synaptic and astrocytic fractional volumes . For the sake of simplicity , we assume that fexc ( t ) always follows the same temporal dynamics exponentially decaying from f0 = 3 . 2 Hz to f∞ = 0 . 5 Hz with a time constant tf = 2 . 5 sec and Nexc = 1500 . Following in vivo measurements in rodents [50 , 51] , the cerebral blood flow is modeled as a piecewise double exponential function delayed in time by t1 relatively to the onset of stimulation t0 . It reads: F ( t ) ={F0{F01 . 1+1 . 5[exp ( −t−t15 ) −exp ( t−t12 ) ]F0+[F ( tend ) −F0]exp ( −t−tend5 ) }ift<t1t1≤t≤tendt<tend} ( 7 ) with F0 = 0 . 012 sec-1 [46] . Typical values are t0 = 0 sec and t0 = 1 sec , tend being the time at which stimulation ends . Two distinct simulation scenarios are considered to mimic in vitro and in vivo conditions . In the in vivo scenario , Equation ( 7 ) is used while in the in vitro scenario , the capillary state variables remain constant at their steady-state while the rest of the variables are left free to vary . Our simulations have shown that this is almost equivalent to taking a constant blood flow F ( t ) = F0 . The blood-oxygen-level-dependent ( BOLD ) signal is computed following [52] . It is written as a function of the deoxyhemoglobin concentration ( dHb ) and of the venous volume ( Vv ) : B O L D ( t ) = V V , 0 [ ( k 1 + k 2 ) ( 1 − d H b d H b 0 ) − ( k 2 + k 3 ) ( 1 − V V V V , 0 ) ] ( 8 ) with dimensionless parameters k1 = 2 . 22 , k2 = 0 . 46 and k3 = 0 . 43 [40] . The steady-state values of deoxyhemoglobin ( dHb0 ) and venous volume ( Vv , 0 ) are given in Table A1 . As noted already by Aubert and colleagues [53] , most models of energy metabolism concentrate on erythrocytes , muscles or other organs such as the liver . It is also not clear whether or not parameters drawn from experiments could be directly injected as such into a model without spatial dimensions and without diffusion processes like ours . To circumvent this problem , we proceeded as follows: First , we chose target steady-state values for the concentration of metabolites following measures reported in the literature . Specifically , we chose the concentration of intracellular sodium following [54] , the concentration of intracellular glucose , phosphoenolpyruvate , pyruvate , adenosine triphosphate and phosphocreatine following [55] , and glyceraldehyde-3-phosphate following [56] . Finally , the NADH concentration in all four compartments where it appears in the model was chosen following [56] and calculations based on results by Kasischke and colleagues [23] . We then optimized a subset of model parameters ( see Table 3 ) by fitting its predictions to the temporal dynamics of NADH fluorescence as measured by Kasischke et al . [23] . Namely , the dynamics of the NADH concentration in various compartments was extracted empirically from Fig . 4D in [23] . Data points were then fitted with sums of exponentials in order to obtain continuous curves . We then optimized the model by minimizing the distance between the temporal dynamics of NADH in the model and the one in the smoothed curve obtained from [23] using least-square distance as the error measure and using the downhill simplex algorithm . After the optimization converged , we rounded the value of the optimal parameter set and recomputed the steady-state value . All along optimization , we checked that the steady-state was stable by computing its Jacobian matrix ( first order approximation ) [45] . The parameter set in Table 3 is the set resulting from this procedure . The simulations were run in MATLAB ( The Mathworks , Natick MA , USA ) . The model was integrated with the ordinary differential equation solver with fixed and optimized parameters ( ode15s ) that is adapted to stiff systems . We used a time step ∂t = 10-4 sec when the neuron is spiking and ∂t = 1 sec starting one second after the end of presynaptic stimulation . ∂t = 10-4 sec is smaller than the fastest time constant appearing in the Hodgkin-Huxley equations [tauh ( -80 mV ) = 6 . 4·10-4 sec] . The second time step ( ∂t = 1 sec ) is small enough for the slow metabolic processes and maintains simulation time and memory usage to reasonable values for an average desktop PC . Simulations take a couple of minutes to execute on a recent laptop . We developed a model of the coupling between neuronal activity and metabolic response in neurons and astrocytes . The model employed to simulate the neural-glial-vascular ( NGV ) functional system is composed of four distinct computational units representing a neuron , an astrocyte , a capillary and the extracellular space ( Fig . 1 ) . The core of our model is composed of the compartmentalized model of brain energy metabolism recently proposed by Aubert and Costalat [13 , 40] . This model connects a model of erythrocyte glycolytic metabolism [45 , 56] together with the so-called “Balloon model” of blood flow dynamics [46] . From this starting point , we added a precise description of neuronal membrane excitability formulated within the Hodgkin-Huxley framework [57] . Channels dynamics is drawn from a model proposed by Wang [44] . It includes all the standard Hodgkin-Huxley currents plus a high-threshold calcium current and a calcium-gated potassium current inducing spike-frequency adaptation . The Hodgkin-Huxley model is connected to the metabolic pathways through the electrogenic Na , K-ATPase pump which is responsible for a net outward current and concomitant ATP consumption . We modified the metabolic pathways to include compartmentalization of NADH between the cytosol and mitochondria . To do so , we developed a very simple model of mitochondrial respiration and added NADH malate-aspartate shuttles between the cytosol and mitochondria , drawing inspiration from a model by [58] . Finally , the model is driven by external input modeled as a global excitatory presynaptic activity and coordinated increase of the cerebral blood flow . The presynaptic population is coarsely described through a time-dependent excitatory conductance . This conductance drives sodium flow in neurons through AMPA receptors and action potential-generating voltage-gated sodium channels , and in astrocytes through excitatory amino acid transporters which co-transport glutamate using the sodium gradient . The model is illustrated in Fig . 1 and extensively described in the Methods section . We first tested the model for its responsiveness to an excitatory stimulus ( Fig . 2A ) and recorded its voltage response ( Fig . 2B ) . In response to this stimulus , the model generated action potentials within the initial 7 sec of the stimulation . Because of spike-frequency adaptation and of the time course of the excitatory stimulus , the frequency of elicited action potentials quickly decreased until the neuron eventually ceased to fire ( Fig . 2B inset ) . Response trajectories of intracellular sodium , in both the astrocyte ( red ) and neuron ( blue ) , showed significant differences in both amplitude and duration , with astrocytes exhibiting a smaller but more sustained response and a delayed recovery ( Fig . 2C ) . We then examined the time course of critical intermediates in energy metabolism in response to the same excitatory stimulus . We first focused on the concentration changes of adenosine triphosphate ( ATP ) and phosphocreatine ( PCr ) in the glial and neuronal compartments ( Fig . 3 ) . In both cases , the response to the excitatory stimulation , evidenced as a consumption of these energy rich metabolites , was slower in the astrocytic ( red ) than in the neuronal compartment ( blue ) ( Fig . 3A ) . And while the decrease in glial ATP surpassed that seen in the neuron ( Fig . 3C ) , the consumption of PCr predominated in the neuron ( Fig . 3B ) . In both compartments , the resulting decrease in ATP concentration was very limited . Next , we examined the trajectories of nicotinamide adenine dinucleotide ( NADH ) in response to the stimulation in the astrocytic , neuronal and mitochondrial compartments ( Fig . 4A ) . The dashed lines represent experimental data from Kasischke et al . [23] . Fig . 4A shows the temporal evolution of the concentration of NADH in the astrocytic cytosol , in the neuronal mitochondria and averaged over the whole tissue as evoked by a 20 sec stimulation episode ( see Methods ) . All three curves are in excellent quantitative agreement with the results reported by Kasischke et al . [23] . In particular , the NADH concentration in the neuronal mitochondria displays an initial dip of about-10% indicating a strong increase of the oxidative metabolism in neurons ( Fig . 4B ) . It then returns towards its baseline before the presynaptic bombardment has finished and finally slightly overshoots in the poststimulus period . On the contrary , the NADH concentration in the astrocytic cytosol increases significantly only about 10 sec after the onset of the stimulation and displays a long-lasting monophasic behavior . This corresponds to a strong and sustained increase of the glycolysis in this compartment ( Fig . 4B ) . The initial dip in the neuronal mitochondria is the result of consumption of NADH to produce ATP . A recovery and rebound results when NADH is produced from the consumption of lactate imported into the neuronal cytosol from the extracellular space ( Fig . 4C ) . In both Fig . 4B and C , it can be seen that the astrocytic response is slower than the neuronal response . In particular , both oxygen and glucose consumption increase immediately at the beginning of the stimulation in the neuronal compartment . Partially supporting this metabolic activity , neurons immediately start to import lactate from the extracellular space ( Fig . 4C ) . On the contrary , the increase in glucose consumption by astrocytes ( Fig . 4B ) is more gradual and the increase in lactate export by astrocytes to the extracellular space is slightly delayed ( Fig . 4C ) . The initial release of presynaptic glutamate with subsequent neuronal activity and reuptake into astrocytes lead to the increase in intracellular sodium concentration and activation of the Na , K-ATPase imposing , along with the conversion of glutamate to glutamine in astrocytes , an increased metabolic demand . However , as can be seen in Fig . 2C , the increase in intracellular sodium is slower and more gradual in the astrocytic compartment leading to the 10 second delay in the glial cell metabolic response to the stimulation . Finally , the dynamics of tissue NADH ( Fig . 4A ) is mirrored in the predicted tissue and extracellular lactate concentrations ( Fig . 4D ) . The utilization of glucose and oxygen by neurons and astrocytes during this 20 sec stimulation episode is shown in Fig . 4B . For model optimization , we imposed that the largest fraction of glucose goes to astrocytes while the largest fraction of oxygen goes to neurons [42] . We observed that this bias is further increased during stimulation . The neuronal oxygen utilization immediately increased at the onset of stimulation in register with the initial dip of the NADH in the neuronal mitochondria . This is consistent with reports that the astrocytic fraction of glucose utilization increases during stimulation [19] . One of the hallmarks of a successful model is its ability to reproduce and explain empirical observations . We thus now turn to an in vivo situation and compare the predictions of the mathematical model we designed in the precedent sections to two experiments carried out in rats . Tissue oxygen and lactate during stimulus ( Fig . 5B and C ) , and lactate transfer between compartments were compared ( Fig . 5D ) . Upon stimulation , CBF increases after a delay of ~1 sec ( see Equation 7 ) , quickly peaks before relaxing to an elevated plateau . This pattern matches neurovascular responses observed in rodents in response to sustained sensory stimulation ( for instance by mechanical activation of the whiskers [50 , 51] ( Fig . 5A ) ) . The 1 sec delay seen in panel A is hardcoded in the model ( see Equation 7 ) . This matches our own observations that CBF only starts to increase above its baseline ~0 . 5–1 sec after the onset of stimulation [51 , 59] . Because neuronal activity slightly precedes functional hyperemia , the oxygen concentration initially dips below its resting value before rising as cerebral blood flow finally increases after the 1 sec delay ( Fig . 5B inset ) . The dip in oxygen below baseline levels after stimulation has ceased reflects the rapid decrease in the replenishment rate by blood at a time when oxygen is still consumed to replenish the ATP that is used to fuel the Na , K-ATPase pump . Extracellular lactate is consumed throughout the stimulation . Its concentration initially dips until the cerebral blood flow increases and leads to a sustained overproduction of lactate ( Fig . 5C ) . At rest , the astrocytic compartment exports lactate to the extracellular space , part of which is taken up by the neuronal compartment for energy production ( Fig . 5D ) , the rest being exported to the circulation . Upon stimulation , export of extracellular lactate to the circulation is reduced while import into neurons is increased . Export of lactate to the extracellular space by the astrocytic compartment is also increased but with a delay and explains the initial dip in concentration . In the recovery period , all transports slowly return back to their baseline values explaining the long lasting overshoot of extracellular and tissue lactate . Export of extracellular lactate to the circulation is durably increased in the recovery period ( Fig . 5D; pink line ) . Fig . 5 shows results of simulations independent from the simulations that yielded Fig . 2 to 4 where the model was constrained to reproduce the experimental results from ref . [23] . In these new simulations , not only is the temporal course of tissue oxygen qualitatively predicted , but the amplitude of fluctuations is , to some extent , quantitatively predicted as well . Our model predicts that the oxygen pressure first drops by -1 . 7% ( inset ) , then overshoots at +17 . 7% before stabilizing 2 . 4% above its baseline in the last 20 sec of the stimulation . Finally , it undershoots to -2 . 6% in the post-stimulus period . These figures are to be compared with the values reported by [50] , namely , an initial drop at -1 . 8% , an overshoot at +19 . 9% , a stabilization 1 . 7% above the baseline and a final undershoot at -3 . 1% . The consumption of lactate closely tracks the stimulation dependent oxygen consumption but lacks the inflections corresponding to CBF changes as it is less affected by the blood flow ( Fig . 5D ) . Lactate is exported by the astrocyte to the extracellular space ( thick red line ) and imported by the neuron from the extracellular space ( thick blue line; net import is negative by convention in this model ) . A small amount of lactate is exported from the extracellular space to the capillary at baseline and this export increases by 69% after the end of the stimulation ( pink line ) . The thin red line denotes the activity of the lactate dehydrogenase converting pyruvate into lactate in the astrocytic cytosol while the thin blue line denotes the activity of the lactate dehydrogenase converting lactate into pyruvate in the neuronal cytosol ( again , the negative sign is a convention ) . These net transfers all contribute to the evolution of the tissue and extracellular lactate concentrations ( Fig . 5C ) . This prediction closely matches the experimental results of Hu and Wilson [16] ( their Fig . 1 ) . We then compared new simulations to measurements from human subjects [26 , 52 , 60] . As in Fig . 5 , the neurovascular response was adapted from experimental measurements ( Fig . 6A ) . Simulated levels of lactate ( Fig . 6B ) , the cerebral metabolic rate for glucose ( Fig . 6C ) , the cerebral metabolic rate for oxygen ( Fig . 6D ) , the ratio of cerebral uptake of O2 to cerebral uptake of glucose ( Oxygen-Glucose Index or OGI ) ( Fig . 6E ) , as well as the blood-oxygen-level dependent signal ( Fig . 6F ) all supported a validation of the model . The model qualitatively and to some extent quantitatively predicted well-known features of these various macroscopic observables , including the BOLD signal [52 , 60] , despite extrapolation from multiple sources . For instance , after an initial dip [17] , the tissue lactate concentration reached a plateau that last until the end of the stimulation [61] . The initial dip in extracellular lactate ( Fig . 6B inset ) is representative of a surge in lactate consumption at the beginning of neural activity , and a symptom of the intrinsic latency in the start-up of the ANLS and of functional hyperaemia . The time lag between the onset and offset of neuronal activation and the onset and offset of CBF also explain other transients such as the slow recovery of the lactate concentration after the cessation of neural activity [61] . Relative increase in glucose and oxygen consumption also matched experimental results and led to a decreased OGI during stimulation [26] .
We present here the first temporal multi-scale model of the NGV that accurately reflects experimental observations in multiple settings and organisms . These findings not only support the ANLS hypothesis but also provide a quantitative mathematical description of the metabolic activation in neurons and astrocytes , as well as of the macroscopic measurements obtained with functional brain imaging techniques . | The brain has remarkable information processing capacity , yet is also very energy efficient . How this metabolic efficiency is achieved given the spatial and metabolic constraints inherent to the designs and energy requirements of brain cells is a fundamental question in neurobiology . The major cell classes in mammalian nervous systems include neurons , glia and the microvasculature that supplies the molecular substrates of energy and metabolism . Together , this neuron-glia-vasculature ( NGV ) ensemble constitutes the functional unit that underlies the cost infrastructure of computation . In spite of its importance , a comprehensive understanding of this dynamic system remains elusive . While it is well established that glucose feeds the brain , few of the details regarding the destiny of glucose intermediates in metabolic pathways are known . Controversy remains regarding the degree of cooperativity between glia and neurons in sharing lactate , the product of aerobic glycolysis ( Warburg effect ) and one of the substrates for further energy extraction by oxidative processes . Specifically , while experimental data support the occurrence of a flow of lactate from glia to neurons , the astrocyte-neuron lactate shuttle ( ANLS ) , some theoretical considerations have been proposed to support the occurrence of lactate transport in the other direction ( NALS ) . Our computational model is the first to integrate multiple timescales of the NGV unit . It provides a quantitative mathematical description of metabolic activation in neurons and astrocytes , and of the macroscopic measurements obtained during brain imaging that uses metabolism as a proxy for neuronal activity . | [
"Abstract",
"Introduction",
"Methods",
"Discussion"
]
| []
| 2015 | Multi-timescale Modeling of Activity-Dependent Metabolic Coupling in the Neuron-Glia-Vasculature Ensemble |
Members of the TRIpartite interaction Motif ( TRIM ) family of E3 ligases have been shown to exhibit antiviral activities . Here we report a near comprehensive screen for antiretroviral activities of 55 TRIM proteins ( 36 human , 19 mouse ) . We identified ∼20 TRIM proteins that , when transiently expressed in HEK293 cells , affect the entry or release of human immunodeficiency virus 1 ( HIV ) , murine leukemia virus ( MLV ) , or avian leukosis virus ( ALV ) . While TRIM11 and 31 inhibited HIV entry , TRIM11 enhanced N-MLV entry by interfering with Ref1 restriction . Strikingly , many TRIM proteins affected late stages of the viral life cycle . Gene silencing of endogenously expressed TRIM 25 , 31 , and 62 inhibited viral release indicating that they play an important role at late stages of the viral life cycle . In contrast , downregulation of TRIM11 and 15 enhanced virus release suggesting that these proteins contribute to the endogenous restriction of retroviruses in cells .
Host cells express specific proteins to interfere with the replication of retroviruses . These proteins are referred to as restriction factors and are considered to be a part of an innate or intrinsic immune system [1–5] . The interferon inducible cytidine deaminase APOBEC3G is packaged into retroviruses and exerts its antiviral effect during reverse transcription . TRIM5 and murine Fv1 belong to a class of restriction factors that interfere with virus replication before and after reverse transcription , respectively . The Fv1 gene encodes an endogenous retroviral Gag found in the mouse genome and has two main alleles [6] . Fv1n , found in NIH Swiss mice , restricts infection by B-tropic MLV ( B-MLV ) but not N-tropic MLV ( N-MLV ) . In contrast , Fv1b , found in BALB/c mice , restricts N-MLV and not B-MLV [7] . NB-tropic MLV replicates in both mouse strains . The residues critical for the N-B tropism of MLV map to the retroviral capsid protein . TRIM5 was identified as a protein responsible for the species-specific restriction of HIV entry [8 , 9] . Moreover , TRIM5 also mediates the Ref1 restriction of specific mouse retroviruses such as N-MLV in mammalian cells [10–12] . TRIM5 binds to incoming retroviral capsids via its C-terminal B30 . 2 or the SPRY ( SPla/RYanodine receptor ) domain causing premature capsid disassembly [13–15] . TRIM5 belongs to the large family of TRIM/RBCC proteins with over 70 members . TRIM proteins display elements of a conserved modular tripartite motif structure consisting of an N-terminal E3 ubiquitin ligase RING ( Really Interesting New Gene ) domain followed by one or two zinc binding motifs named B-box and a predicted coiled coil ( CC ) region ( see Table 1 ) . The C-terminus is highly variable and contains specific domains such as the B30 . 2/PRY-SPRY domain ( Table 1 ) . The presence of a RING domain suggests that these proteins function as E3 ubiquitin ligases . The associated B-box and coiled coil are believed to participate in protein-protein interactions and formation of macromolecular complexes [16 , 17] . TRIM proteins localize to various regions within the cells and many define specific nuclear ( TRIM19/PML ) [18] or cytoplasmic compartments ( TRIM5 ) [8 , 19] . Others such as TRIM1 , 9 and 18 have been shown to associate with microtubules [20–23] . Proposed physiological roles for TRIM proteins include fundamental cellular processes such as apoptosis , transcription , differentiation , and regulation of cell cycle progression [17] . Moreover , mutations in several TRIM proteins have been linked to human disease [17] . A number of TRIM proteins besides TRIM5 and its close relatives have been shown to possess antiviral activities [2 , 3 , 17 , 24] . For example , TRIM1 has been shown to restrict N-MLV [12] . Broad antiviral activities have been described for TRIM19 , the defining component of PML bodies in the nucleus . The list of viruses inhibited by TRIM19 includes vesicular stomatitis virus , influenza A virus , human cytomegalovirus , herpes simplex type 1 , Ebola virus , Lassa fever virus , lymphocytic choriomeningitis virus , human foamy virus and HIV [2 , 18] . TRIM28 restricts MLV in cells of germline origin by inhibiting LTR-driven transcription [24] . TRIM22 and TRIM32 were reported to attenuate transcription of the HIV LTR [25 , 26] . The identification of TRIM25 as a K63 specific ubiquitin E3 ligase activating RIG-I presents direct evidence that TRIM proteins regulate innate immunity to viral infection [27] . The recognition and suppression of Sendai virus , New Castle disease virus and vesicular stomatis virus ( VSV ) replication by RIG-I depends on functional TRIM25 [27] . Thus , the association of several family members with antiviral activities coupled with the fact that many of them are induced by interferons [2] has led to the hypothesis that members of TRIM family proteins are a part of innate immune system to counter intracellular pathogens [2 , 17 , 28] . To systematically test antiretroviral activities of TRIM proteins , we investigated the ability of 55 TRIM proteins ( 36 human , 19 mouse ) to interfere with early and/or late stages of the retroviral life cycle .
We screened for potential antiviral activities of 36 human and 19 mouse TRIM proteins ( Table 1 ) by transient expression in HEK293 cells . These cells are highly permissive for most retroviruses and are easily transfectable . We first performed control experiments to verify that the transfection of 50 ng plasmids encoding TRIM proteins in a 24-well format minimally induced apoptosis and had little effect on cell viability or gene expression ( Figure S1A–S1C ) . In the case of human and mouse TRIM11 , transfected DNA levels were reduced to 10 ng ( Figure S1C ) . We then analysed the ability of the TRIM proteins to interfere with viral entry defined here as all early events in the retroviral life cycle leading up to the establishment of viral gene expression ( Figure 1A ) . To identify activities directed specifically against incoming retroviral capsids , all viruses carried the same subgroup A envelope glycoprotein of ALV ( ALV-A ) and target cells expressed the cognate receptor Tva950 [29] . To guarantee a ∼90% probability of Tva950 co-expression with each TRIM protein , plasmids encoding for both proteins were co-transfected 36 h prior to initiating infection . Infection levels were determined by measuring the expression of cytoplamic GFP from integrated viral genomes . To perform an entry screen for HIV , reporter viruses were generated in HEK293 cells by transfecting HXB2ΔEnv-GFP , an HXB2 derivative lacking envelope and encoding GFP instead of the nef gene , together with a plasmid encoding ALV-A Env . Culture supernatants were harvested 48 h later and tested for their ability to infect HEK293 cells expressing Tva950 in the presence or absence of individual TRIM protein . To evaluate the significance of the observed inhibitory and enhancing effects standard statistical analysis was employed to arrive at a cut-off of 2 . 5 ( by adding the maximum variability between control samples and two times the standard deviations ) . HIV entry was potently blocked by rhesus TRIM5 ( 38-fold ) and to a lesser extent by human TRIM5 ( 5-fold ) confirming previous results [8 , 10–12] ( Figure 1B ) . Interestingly , no other TRIM protein affected HIV entry as potently as rhesus TRIM5 . Mouse TRIM8 inhibited HIV entry about 6-fold . Moderate inhibitory effects were observed for human TRIM11 , 26 , 31 and mouse 10 , 11 and 56 . Expression of human TRIM38 and mouse TRIM21 enhanced HIV entry . To perform a similar experiment for N-tropic MLV ( N-MLV ) , reporter viruses were generated in HEK293 cells by transfecting plasmids encoding for N-tropic MLV GagPol , MLV LTR-GFP and ALV-A Env . Viruses were harvested as above and the susceptibility of HEK293 cells expressing Tva receptor and individual TRIM proteins to N-MLV was tested . N-MLV was strongly inhibited by TRIM1 ( 15-fold ) and TRIM5 ( 18-fold ) ( Figure 1C ) . A number of additional TRIM proteins moderately affected N-MLV entry ( human TRIM25 , 26 and 62; mouse TRIM8 , 25 , 31 and 56 ) . Interestingly , human and mouse TRIM11 as well as mouse TRIM30 enhanced N-MLV entry ( 4 . 5- , 3- and 4 . 5-fold , respectively ) . The inhibitory pattern observed for N-MLV was largely distinct from HIV ( Figure 1B and 1C ) . Notable exceptions included human TRIM26 , mouse TRIM8 and 56 that affected both HIV and MLV . Opposite effects were observed for the TRIM11 proteins . While they inhibited HIV entry , enhancing effects were observed for MLV . TRIM proteins specifically affecting N-MLV were human and mouse TRIM25 , human TRIM62 , mouse TRIM31 and mouse TRIM30 . In contrast , human TRIM proteins 31 , 38 and the mouse proteins 10 and 21 specifically affected HIV entry . A scatter plot depicting fold inhibition in infectivity for HIV versus N-MLV summarizes these results ( Figure 1F ) . The interference of TRIM 1 and 5 with MLV entry is specific for capsid determinants of N-MLV but not of B-MLV [10–12] . To test which TRIM proteins are specific for N-MLV , we performed an identical entry experiment for B-MLV . As previously reported , human TRIM1 and TRIM5 exhibited no effect on B-MLV entry ( Figure 1D ) . In contrast , the remaining TRIM proteins exhibited an inhibitory profile that resembled that observed for N-MLV ( compare Figure 1D to 1C ) . Interestingly , the enhancing properties of TRIM11 ( human and mouse ) and mouse TRIM30 proteins were specific for N-MLV and not observed for B-MLV . The ratio of fold inhibition for N versus B-MLV as well as the scatter plot analysis illustrate this finding ( Figure 1E and 1G ) . Thus , the inhibitory effects of TRIM1 and 5 as well as the enhancing effects of TRIM11 ( human and mouse ) and TRIM30 ( mouse ) strongly correlate with N-tropism . To validate critical results gained in our transient expression screen , we downregulated endogenously expressed human TRIM proteins 11 , 25 , 31 and 62 in HeLa cells using RNA interference ( RNAi ) ( Figure 1H ) . Consistent with the inhibitory effects of TRIM 11 and 31 on HIV entry , downregulation of both proteins using siRNA facilitated HIV entry ( 2–3-fold ) . While modest , these enhancements in viral entry suggest that TRIM11 and 31 contribute to the restriction of HIV in HeLa cells . Likewise , the enhancing effect of TRIM11 expression on N-MLV entry led to a corresponding inhibition following gene downregulation of the endogenous protein ( 3-fold ) . In contrast , TRIM25 and 62 exhibited inhibitory effects against MLV viruses when overexpressed or silenced using RNAi . Thus , the transient expression screen for TRIM proteins identified human TRIM11 and TRIM31 as factors modulating retroviral entry . The specific enhancement of N-MLV , but not B-MLV entry upon expression of TRIM11 proteins as well as mouse TRIM30 was unanticipated ( Figure 1C–1E ) . These enhancing properties of TRIM11 were observed over a wide range of expression levels ( Figure 2A ) . Silencing of endogenous TRIM11 by siRNA in HeLa or HEK293 cells led to an increase in restriction of N-MLV , but not B or NB-MLV ( Figure 1H and data not shown ) . Together , these data suggest that TRIM11 regulates the Ref1 restriction in human cells . HEK293 cells endogenously express low levels of human TRIM5 that restricts N-MLV entry [10–12] . To test if the effects of TRIM11 and TRIM30 on N-MLV entry are due to interference with the Ref1 restriction , we silenced TRIM5 in HEK293 cells by RNAi . Indeed , the enhancing effects of TRIM11 and TRIM30 on N-MLV entry were dependent on the presence of TRIM5 and lost in response to TRIM5 silencing ( Figure 2B ) . Potentially , these proteins affect TRIM5 protein levels . To test this hypothesis for TRIM11 , a plasmid encoding HA or GFP-tagged TRIM5 was co-transfected together with either empty vector or increasing amounts of a plasmid encoding mouse and or human TRIM11 . TRIM5 protein levels were then examined by western blot and fluorescence microscopy ( Figure 2C and 2D ) . Strikingly , expression of low amounts of TRIM11 led to the disappearance of TRIM5 , an effect that could be delayed by treating cells with the proteasome inhibitor MG132 ( Figure 2C and 2D ) . The reduction of TRIM5 levels as determined by western blotting corresponded with a loss of cytoplasmic bodies ( Figure 2D , lower panel ) . The protein levels of another TRIM protein , human TRIM15 , were largely unaffected by the expression of human TRIM11 ( Figure 2D ) . Because both proteins are expressed from the same promoter , the observed reduction in TRIM5 levels is likely not explained by effects of TRIM11 on transcription . Deleting the RING domain of TRIM11 did not affect TRIM5 protein levels , indicating a functional dependence on E3 ligase activity ( Figure 2D ) . Correspondingly , entry of N-MLV in HeLa cells was enhanced 16-fold by the expression of wild-type TRIM11 , whereas TRIM11 lacking the RING domain exhibited reduced activity ( Figure 2E ) . Together these results suggest that TRIM11 regulates the turnover of TRIM5 thereby regulating the level of Ref1 restriction in mammalian cells . The observed co-localization of TRIM11 and TRIM5 proteins to cytoplasmic bodies is consistent with such a model ( Figure 2F ) . In contrast to TRIM11 proteins , expression of mouse TRIM30 did not affect the protein levels of TRIM5 ( Figure 2D ) . How mouse TRIM30 interferes with the Ref1 restriction remains to be determined . Interestingly , TRIM30 is the closest homologue of human TRIM5 in the mouse genome , but carries a deletion in the variable region 1 within the B30 . 2 domain that is critical for interaction with capsid ( Figure S1D ) [30–32] . Mouse TRIM30 may function as a dominant-negative protein not unlike TRIM5 proteins lacking the B30 . 2 domain [8 , 31] . After studying the role of TRIM proteins during early events in viral replication , we next investigated if they exhibited antiviral effects at late stages of viral replication ( Figure 3A ) . To determine effects specific for HIV release , we bypassed entry by directly transfecting plasmids that encoded for TRIM proteins along with a HIV variant HXB2ΔEnv-GFP lacking Env and expressing cytoplasmic GFP . Infectious virions were generated by co-expression of VSVG . After 48 h , the culture supernatants were harvested and the level of GFP expression in producer cells was determined by flow cytometry . The viral infectivity in the harvested culture supernatant was determined by infecting susceptible target cells and measuring GFP-positive cells after an additional 36 h . In addition , the Gag protein released into the supernatant was determined by western blot using antibodies to the HIV capsid protein p24 . The results of such an experiment for HIV release is shown in Figure 3B in fold inhibition . HIV capsid released into the supernatant is presented in Figure S2A . Our analysis identified the human TRIM proteins 15 , 26 , 32 , the mouse proteins 11 , 25 , 27 , and 56 as factors that specifically affected HIV release from cells , but not viral gene expression ( Figure 3B ) . A number of TRIM proteins ( human TRIM19 , 21 , 25 and mouse TRIM8 ) were close to the statistically determined cut-off value of 6 . At the transfection level ( 50ng ) used in the assembly assay , human TRIM11 affected both viral gene expression and virus release . A scatter plot analysis depicting the effects of TRIM protein expression on virus release over effects on LTR expression summarizes these results ( Figure 3D ) . We then performed an identical experiment for NB-tropic MLV ( Figure 3C ) using a plasmid encoding for the Friend57 MLV genome carrying a GFP insertion into the Env protein [33 , 34] . MLV capsid ( p30 ) released into the supernatant is presented below the graph ( Figure 3C ) . Our analysis revealed a striking sensitivity of MLV to the expression of TRIM proteins , in particular human TRIM proteins . Overall , 21 TRIM proteins and rhesus TRIM5 inhibited MLV release at least 10-fold , 9 of which inhibited MLV release by more than 100-fold ( Figure 3C ) . A dose-response experiment revealed that transfecting small amounts of plasmids encoding for human TRIM15 , 25 , 31 , 62 and mouse 11 and 25 resulted in potent antiviral activity ( Figure S2B ) . In contrast , higher DNA amounts were required for the other TRIM proteins , particularly TRIM5 . TRIM22 , while removed from the screen due to varying results , did exhibit antiviral activity in a dose-dependent manner . Generally , we observed two distinct phenotypic groups for MLV restriction . The first group , including the TRIM proteins 8 , 15 , 19 , 25 , 26 , 28 and 35 specifically interfered with the release of infectious MLV into the culture supernatant without major effects on viral gene expression ( Figure 3C ) . For the second group , consisting of human TRIM proteins 1 , 11 , 13 , 14 , 21 , 27 , 31 , 32 , 62 , the mouse proteins 8 , 11 , 27 and rhesus 5 , the inhibitory effects on viral gene expression were close to the cut-off of 6 or higher implying that a suppression of viral gene expression contributed to the observed reduction in the release of infectivity . A scatter plot depicting effects of TRIM proteins on infectivity versus gene expression summarizes these results ( Figure 3E ) . For a number of TRIM proteins , the reduction in infectivity did not correspond with a proportionate reduction in p30 release . Western blot analysis of released virus for Env and Gag revealed that human TRIM13 , 21 and mouse TRIM19 preferentially affected Env incorporation ( Figure S2C ) . A reported role for TRIM13 in protein degradation at the endoplasmic reticulum is consistent with such a phenotype [35] . Finally , to measure ALV release , we studied the effects of TRIM expression in chicken fibroblasts because the release of this virus is restricted in mammalian cells [36] . Among the few potent antiviral proteins were human and mouse TRIM25 , human 32 and mouse TRIM11 ( Figure S2D ) . Human TRIM11 was not tested for ALV . The inhibitory effects observed above could be the result of overexpression of these proteins in HEK293 cells . To determine their role in the viral life cycle , we targeted endogenously expressed TRIM proteins for downregulation using RNAi . We concentrated on TRIM 11 , 15 , 25 , 31 and 62 because they were very effective at low transfection levels and were endogenously expressed in HEK293 and HeLa cells ( Figure S3A and S3B ) . Downregulation of TRIM proteins 25 , 31 and 62 inhibited HIV and MLV release in HEK293 cells , suggesting that these proteins play a role in efficient virus release ( Figure 4A and 4B ) . Notably , silencing of TRIM62 also strongly interfered ( 7-fold ) with HIV release in HeLa cells ( Figure 4B ) . Correspondingly , expression of low amounts of TRIM62 enhanced HIV gene expression and release ( Figure 4C ) . To gain further insights into the antiviral activities of these proteins we tested mutant proteins impaired in their E3 ligase activity . Interestingly , the E3 mutant of TRIM62 inhibited HIV and MLV release more potently than the wild-type protein , likely by exhibiting a pronounced dominant-negative effect ( Figure 4C and 4D ) . Together , these data suggest that TRIM 25 , 31 and 62 play an important role in virus release . In contrast to TRIM25 , 31 and 62 , downregulation of endogenous TRIM11 enhanced HIV release ∼4-fold in HEK293 and moderately ( 2×fold ) in HeLa cells ( Figure 4A and 4B ) . MLV release was enhanced 5-fold in response to silencing of TRIM15 in HeLa cells ( Figure 4B ) . The enhancement of virus release observed in response to gene silencing is consistent with the hypothesis that both TRIM proteins contribute to the endogenous restriction of HIV and MLV in mammalian cells . We next tested the contribution of E3 ligase function to the antiviral activity of both proteins . Interestingly , the antiviral activity of TRIM11 was critically dependent on a functional E3 ligase domain ( Figure 4D ) implying the involvement of the ubiquitin-dependent degradative pathway . In contrast , the E3 mutant of TRIM15 largely retained its inhibitory activity indicating that it interferes with viral replication via a different mechanism ( Figure 4D ) . To understand how TRIM15 can interfere with viral release in an E3 ligase independent manner , we performed a domain analysis for human TRIM15 . TRIM15 YFP fusion protein was as active as its untagged version . Hence this analysis was performed using YFP fusion proteins ( Figure 5A ) . Interestingly , TRIM15 lacking the B-box , but not RING or SPRY domains , lost all of its antiviral activity ( Figure 5A ) . In fact , the B-box alone exhibited antiviral activity . TRIM5 specifically interferes with retroviral entry by binding to incoming mature capsids [13–15] and it had recently been suggested that rhesus TRIM5 can also bind and degrade immature capsids interfering with virus production [37] . To test if TRIM15 can bind to the immature Gag precursor protein of MLV , we performed co-immunoprecipitations . Interestingly , antibodies against TRIM15 specifically co-immunoprecipitated MLV Gag and vice versa ( Figure 5B and 5C ) . Importantly , TRIM15 fragments containing the B-box interacted with MLV Gag , while TRIM15 mutants lacking the B-box did not . In fact , the B-box alone was capable of interacting with the MLV Gag precursor protein . Thus , TRIM15 interferes with MLV release by directly or indirectly binding the MLV Gag precursor protein via its B-box . To understand how TRIM15 binding to Gag alters the cellular fate of retroviral capsids , we transfected plasmids encoding for TRIM15-YFP together with MLV Gag-CFP and replication competent MLV into HEK293 . Visualization using fluorescence microscopy revealed on average a reduction of Gag fluorescence at the plasma membrane in cells containing cytoplasmic TRIM15 bodies ( Figure 5D ) . When TRIM15-CFP was expressed together with HIVGag-YFP , a similar phenotype was observed . Less HIV Gag reached the plasma membrane , but rather accumulated intracellularly ( Figure 5E ) .
Using a transient expression screen in HEK293 cells we have performed the first near comprehensive screen for antiviral activities of members of the TRIM family of proteins . Our screen identifies ∼20 TRIM proteins with antiviral activity demonstrating that multiple TRIM proteins can exhibit antiviral activities . Because many TRIM proteins are upregulated in response to interferons [2] , a potential role of these proteins in the establishment of an antiviral state should be investigated . The specific effects of TRIM proteins on the replication cycle of each retrovirus are summarized in Figure 6A using red/green color-coding for inhibitory and enhancing activities , respectively . This presentation readily displays the specificity of TRIM1 , 5 , 11 and 30 for N-MLV entry . This analysis also allows a comparison of the entry results with the viral gene expression data obtained in the screen for virus release ( Figure 6A–6C ) . For example , the inhibitory activities of TRIM1 and 62 on N-MLV entry likely include effects on viral gene expression ( Figure 6A and 6C ) . This analysis also reveals that an overexpression of human and mouse TRIM11 proteins affects MLV gene expression ( Figure 6A and 6C ) . Despite these inhibitory effects , TRIM11 expression specifically enhanced N-MLV , but not B-MLV entry . Under these conditions , the protein levels of transiently expressed TRIM5 were reduced . TRIM11 may contribute to the turnover of endogenous TRIM5 , because silencing of endogenous TRIM11 enhanced the Ref1 restriction . These two observed activities of TRIM11 , degradation of cytoplasmic proteins as well as the regulation of transcription , are similar to previous reports for a role of TRIM11 in the turnover of humanin and ARC105 [38 , 39] . Clearly , several cellular targets exist for TRIM11 . Its potential role in the turn over of TRIM5 , a protein that potently restricts retrovirus entry , could be of therapeutic importance . Strikingly , most of the antiviral TRIM proteins exhibited strong inhibitory effects against late stages of the viral life cycle ( Figure 6A ) . MLV was highly sensitive to the expression of TRIM proteins , particularly of human origin . Of the 14 TRIM proteins specifically interfering with MLV release , only two were of mouse origin ( Figure 6D ) . A similar cross species effect was observed for HIV . 21% of all mouse , but only 11% of all human TRIM proteins interfered with HIV release . Cross species effects are consistent with the hypothesis that TRIM proteins contribute to the innate control of retroviruses and that over time , viruses can adapt to inhibitory effects . In our further analysis we concentrated on TRIM proteins that were highly effective even at very low transfection levels . Gene silencing of TRIM 25 , 31 and 62 inhibited virus release suggesting that these E3 ligases play a role in cellular pathways critical for virus release . They were identified in our expression screen likely because overexpression of the wild-type protein exhibited a dominant-negative effect . In contrast , downregulation of TRIM11 and TRIM15 enhanced virus release suggesting that these proteins contribute to the restriction of MLV and HIV even in highly susceptible HEK293 and HeLa cells . A detailed understanding of host restriction may lead to antiviral therapies aimed at strengthening the innate immunity to retroviruses at the cellular level . The interaction of TRIM15 with retroviral Gag suggests that TRIM proteins , apart from entry , can recognize retroviral Gag proteins during assembly and budding and thereby inhibit viral release . A preferential targeting of late stages of the retroviral life cycle may be more consistent with a role for TRIM proteins in the establishment of an antiviral state .
HEK293 , HeLa , DFJ8 and DF-1 were described previously [40] . TZM-bl cells were a gift from Vineet KewalRamani ( NCI Frederick , MD ) . TRIM constructs presented in Table 1 were confirmed by sequencing to be authentic and in the correct reading frame . The reference for human TRIM22 is NM_006074 . YFP fusion proteins of human TRIM 11 , 15 , 25 , 31 and 62 were generated by insertion of PCR amplified genes into the EcoRI/XhoI sites of pEYFP-N1 ( Clontech , Palo Alto , CA ) . E3 mutants were created by substituting two active site cysteines to alanine using site-directed mutagenesis ( QuikChange , Stratagene , La Jolla , CA ) . TRIM15-YFP mutants were generated by PCR; BCPS , CPS , RBC , R-CPS-YFP fusions lacked the amino acids 1–64 , 1–119 , 346–465 and 81–119 , respectively . The B-box-YFP corresponds to amino acids 64–129 . N , B and NB-tropic MLV were prepared by transfecting 4 μg of a plasmid encoding a viral RNA ( pLZRS-GFP ) [40] , 4 μg plasmid encoding the envelope glycoprotein of subgroup A of ALV ( EnvA ) [41] and 4 μg of plasmids expressing either N- , B-tropic ( pCIG3-N or B , gifts from Greg Towers and Jonathan Stoye , University College London , UK ) [42] or NB GagPol ( pMDGag-Pol ) [41] into a 10 cm plate of HEK293 cells using FuGene 6 ( Roche , Indianapolis , IN , USA ) and serum-free OPTIMEM media ( Invitrogen Corporation , California ) . HIV-1 reporter viruses were generated by transfecting 4 μg of HXB2Env−-GFP , a HXB2 derivative ( lacking envelope and encoding GFP instead of nef gene; gift from Heinrich Gottlinger , Worcester , UMass , MA ) , and 4 μg plasmid encoding ALV EnvA [41] . ALV reporter viruses were from supernatants of DF-1 cells chronically infected with RCASBP ( A ) -GFP [41] . For siRNA experiments , viruses were generated carrying the Vesicular stomatitis virus G ( VSVG ) envelope protein instead of ALV-A Env . The culture medium was harvested 48 h after transfection , filtered through 0 . 45 μm , aliquoted and stored at −80 C . To determine the titer , serial dilutions of virus stocks were titrated onto DFJ8 cells in the presence of 5 μg/ml of polybrene followed by flow cytometry of GFP- positive cells ( FACS , Becton Dickinson ) 36–48 h later . HEK293 cells were co-transfected in 24 wells with 50 ng each of plasmids encoding the TRIM protein and 25 ng of ALV receptor Tva950 . 30 h after transfection , cells were seeded into 48-well plates at a density of 1 . 5 × 104 target cells/well . After an additional 6 h , cells were challenged with N , B-MLV , HIV or ALV carrying a GFP reporter genome . GFP-positive cells were quantified by flow cytometry after 36 h post infection . To perform the screen initially with a dynamic range that allows reliable detection of both inhibiting and enhancing effects , an amount of virus was used that resulted in infection levels of 5% . Strong inhibitory or enhancing TRIM proteins were characterized in a second round with adjusted infection levels . All experiments were at least performed four times on separate days . These data sets were combined for the final analysis shown in Figure 1 . Fold inhibitions in virus entry represent the ratio of percent GFP-positive cells of cells transfected with empty control vector versus those expressing TRIM proteins . The maximum variability between control samples in the absence of TRIM proteins was ∼1 . 5-fold . The cut-off value of 2 . 5 applied in Figure 1 was the derived from 1 . 5 plus two times the standard deviations ( confidence value of 95% ) of 0 . 5 . 2 . 5 × 105 HEK293 cells ( producer cells ) in 24-wells were transfected with 50 ng TRIM expressing construct with either 150 ng each of the HXB2Env−-GFP and 50 ng plasmid expressing VSVG . For MLV release assays , 200 ng of plasmid MLVEnv-GFP encoding full length Friend 57 MLV genome with a GFP insertion into the envelope protein [33 , 34] was co-transfected with TRIM expressing construct as above . ALV release assays were conducted by transfecting 3 × 104 DF-1 cells in 24-well plate with 50 ng TRIM expressing plasmids , 200 ng of ALV vector lacking the envelope protein ( plasmid DASBP-GFP , a gift from Stephen Hughes , NCI Frederick , MD ) and 50 ng of plasmid expressing ecotropic Friend MLV envelope protein ( pcDNA3-FrEnv ) [33] . 48 h after transfection , the released virus infectivity was measured by applying two dilutions of the culture supernatants differing by 10-fold onto target cells ( DFJ8 for MLV and ALV; HEK293 or TZMbl cells for HIV ) in presence of 5 μg/ml of polybrene . GFP-positive cells were enumerated after additional 36–48 h as above . For measuring viral gene expression , the mean fluorescence intensity ( MFI ) of GFP in the transfected producer cells ( 48 h after transfection ) was estimated using FACS ( Becton Dickinson ) . Fold inhibition in viral gene expression was calculated using the ratio of GFP MFI in cells transfected with vector to those expressing TRIM . The maximum variability between control samples in the absence of TRIM proteins was ∼2-fold . The cut-off value of 6 applied in Figure 3 was the derived from 2 plus two times the standard deviations ( confidence value of 95% ) of 2 . To measure the release of Gag ( p30 for MLV and p24 for HIV ) a parallel experiment as described above was conducted in triplicates . 48 h after transfection the culture supernatants from triplicate wells were combined and viruses sedimented at 12 , 000 × g in a microcentrifuge for 2 h . The resulting 12 , 000 g pellet was solubilized in SDS gel loading buffer and resolved on SDS-10% PAGE followed by western blot using antibodies to MLV capsid and Env ( p30 & gp70; Quality Biotech , Camden , NJ ) or HIV capsid ( p24; obtained through the AIDS Research and Reference Reagent Program , NIH from Dr . Michael H . Malim . ) Cell viability and Caspase 3/7 activity was measured sequentially for the same samples 48 h after transfection of 50 ng TRIM-expressing construct in HEK 293 cells using CellTiter-Blue™ Cell Viability and Caspase-Glo 3/7 Assay System ( Promega Corp . ) according to manufacturer's recommendations . To measure the effect on cellular gene expression due to transient expression of TRIM proteins , 10 ng of plasmid expressing GFP under a CMV promoter was co-transfected with 50 ng of TRIM encoding plasmids in HEK293 cells . 48 h later the MFI of GFP and fold inhibitions were calculated as above . 2 × 105 HeLa or HEK293 cells in 48 well plates were transfected with 80 nM TRIM-specific siRNA smartpool ( Dharmacon Inc ) or control siRNA ( Dharmacon Inc , #D-001210–01 ) using Lipofectamine 2000 ( Invitrogen , CA ) . After 4 h the medium was changed and after additional 20 h the cells were split 1:4 into 4 wells of a 48 well plate . For entry assays , 36 h post-transfection , the cells were challenged as above with VSV-G pseudotyped reporter N- , NB-MLV and HIV at two concentrations of virus differing by 2-fold . After additional 24 h the cells were harvested , fixed and analyzed by FACS . For virus release assays , 30 h after first siRNA transfection , cells were transfected again with 80 nM TRIM-specific siRNA smartpool or control siRNA together with either 200 ng plasmid encoding full length Friend MLV genome carrying a GFP insertion into the envelope protein or 100ng of full length replication competent HIV-1 plasmid pNL4–3 ( AIDS Research and Reference Reagent Program ) . After an additional 48 h , the culture supernatants were harvested and applied onto DFJ8 for MLV and TZMbl cells for HIV titration . For MLV , DFJ8 cells were harvested 48 h after infection and analyzed by FACS as before . For HIV , 48 h after infection , TZMBL cells were lysed and luciferase activity measured using a Luminometer ( Turner Biosystems ) . Fold increase in entry or release was calculated using a ratio of the percent GFP positive cells or luciferase activity ( measured as relative light units ) from experimental samples transfected with TRIM specific siRNA and those transfected with control siRNA . Silencing of endogenous human TRIM proteins was carried out using ON-TARGETplus siRNA smart pools ( a mix of 4 siRNAs ) from Dharmacon , Inc . pre-designed to reduce off-target effects by up to 90% . We routinely obtained 70 to 90% knock down of specific TRIM proteins as assessed by monitoring the levels of transiently expressed TRIM-GFP 36 h post-transfection . The sense sequences of siRNAs targeting ( 1 ) TRIM11 were #1: AG GCGAAGCUGGAGAAGUCUU , #2: GAGCUGAUC CUGUCUGAAGUU , #3: UCACUGCUA UUCAUCUUUCUU , #4: GGACAGCCCAGAGCGCU UUUU; ( 2 ) TRIM15 were #1: GGGAGAAACUUACUGCGAGUU , #2: GCGAGAACGAUGCCG AGUUUU , #3: CCCUGAAGGUGGUCCAUGAUU , #4: GCAGAACCACAGACGGCUUUU; ( 3 ) TRIM25 were #1: CGGAACA GUUAGUGGAUUUUU , #2: CAACAAGAAUACACGGAA AUU , #3: GCGGAUGACUGCAAACAGAUU , #4: GGGAUGAGUUCGAGUUUCUUU; ( 4 ) TRIM31 were #1: GGAGAAGAAUU UCCUGCUAUU , #2: GGAAGAACGCAAUCAGGUU UU , #3: AAUUUGAACUCCUGCAUCAUU , #4: CCACAAAUCCCAUAAUGUCUU; ( 5 ) TRIM62 were #1: CUACAAUGCUGAUGACAUGUU , #2: GCGAGAAGUUCCCUGGCAAUU , #3: AGACCAACCUCACAUAUGAUU , #4: GACCAAGUCUUCCACCAAGUU . For human TRIM5 , the sequences of the siRNA smart pool from Dharmacon were as previously reported [8] . TRIM5 silencing using this smart pool resulted in a ∼4-fold and ∼50-fold enhancement of N-MLV entry in HEK293 and HeLa cells , respectively . For easy interpretation of the screen we used scatter plots generated using Excel . Fold inhibitions for the two parameters compared were plotted against each other in log scale . Java TREEVIEW was used to represent data in color codes [43] . The input fold inhibition values obtained as described in previous sections were log2 transformed to obtain positive ( inhibition , shades of red ) and negative ( enhancement , shades of green ) values prior to data analysis using TREEVIEW . HEK293 cells were transfected as above with 50 ng of plasmids encoding TRIM15 derivatives with C-terminal YFP fusions or untagged TRIM15 and 200 ng of plasmid encoding full length Friend MLV genome carrying a GFP insertion into the envelope protein or 100 ng of MLV Gag-GFP . 48 h post-transfection , the cells were lysed using triple detergent lysis buffer ( TDLB , 100 mM Tris [pH 8 . 0] , 1% Triton-X-100 , 0 . 5% sodium deoxycholate , 0 . 2% sodium dodecyl sulphate , 150 mM NaCl ) . The nuclei and undissolved cellular components were removed by centrifugation at 12 , 000 × g for 30 minutes in a microcentrifuge . The clarified 12 , 000 × g supernatant was used for immunoprecipitation using protein-G beads prebound with antibodies to MLV capsid ( Quality Biotech , Camden , NJ ) or TRIM15 ( Abcam , Boston , MA ) raised in goat or isotype specific antibodies . The immunoprecipitates were washed three times with TDLB and analyzed using SDS-10%-PAGE followed by western blot using antibodies to GFP . The generation of fluorescently labeled MLV and HIV virions using Gag-GFP proteins was previously described [33] . To visualize MLV Gag and TRIM15 in HEK293 cells , plasmids encoding for MLV Gag-CFP ( 50 ng ) , replication competent MoMLV ( 200 ng ) and TRIM15-YFP ( 10 ng ) were co-transfected . To perform a similar experiment for HIV , 50 ng HIV Gag-YFP was co-transfected with 10 ng TRIM15-CFP . 24 h later , cells were fixed and the CFP and YFP channels monitored using the 60x oil objective ( NA 1 . 4 ) of a Nikon TE2000 inverted wide-field microscope . To monitor TRIM11 and TRIM5 , HEK293 cells were transfected with 5 ng TRIM11-YFP together with 50 ng TRIM5-CFP and cells imaged 24 h post-transfection using the 60× oil Nikon objective ( NA 1 . 4 ) and an Improvision spinning disc confocal microscope . Total RNA was extracted from HEK293 and HeLa cells using PrepEase RNA extraction kit ( USB ) , which has an on-column DNAse treatment step . The RNA was reverse transcribed with anchored oligodT using Reverse-iT first strand synthesis kit ( ABgene ) . The cDNA was then used to check presence TRIM-specific sequences by PCR using appropriate primer pairs . Control reactions which used plain RNA for PCR amplification did not yield any products ( data not shown ) . Specific primer pairs used for amplifications were obtained from Primer Bank database and sequences can be found at http://pga . mgh . harvard . edu/primerbank/index . html . | A lot of excitement in the field of innate immunity to retroviruses such as HIV has come from the discovery of TRIM5 as a key player in cross species restriction . TRIM5 belongs to a family of E3 ligases with over 70 members , a number of which have exhibited antiviral activity . These findings have led to the hypothesis that several TRIM proteins may contribute to the innate immunity to retroviruses . In this manuscript , we systematically test the antiviral activities of 55 human and mouse TRIM proteins . The results are astonishingly complex with activities affecting both early and late stages of the retroviral life cycle . Importantly , a number of TRIM proteins that affect HIV or MLV replication upon overexpression , enhance virus entry or release when downregulated by gene silencing . These experiments suggest that additional TRIM proteins contribute to the endogenous restriction of retroviruses . Future work should focus on the identification of TRIM proteins that are upregulated specifically in response to interferons as well as the mechanisms by which the identified proteins interfere with retroviral replication . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"viruses",
"mus",
"virology",
"in",
"vitro",
"microbiology",
"homo"
]
| 2008 | TRIM E3 Ligases Interfere with Early and Late Stages of the Retroviral Life Cycle |
Computational prediction of nucleotide binding specificity for transcription factors remains a fundamental and largely unsolved problem . Determination of binding positions is a prerequisite for research in gene regulation , a major mechanism controlling phenotypic diversity . Furthermore , an accurate determination of binding specificities from high-throughput data sources is necessary to realize the full potential of systems biology . Unfortunately , recently performed independent evaluation showed that more than half the predictions from most widely used algorithms are false . We introduce a graph-theoretical framework to describe local sequence similarity as the pair-wise distances between nucleotides in promoter sequences , and hypothesize that densely connected subgraphs are indicative of transcription factor binding sites . Using a well-established sampling algorithm coupled with simple clustering and scoring schemes , we identify sets of closely related nucleotides and test those for known TF binding activity . Using an independent benchmark , we find our algorithm predicts yeast binding motifs considerably better than currently available techniques and without manual curation . Importantly , we reduce the number of false positive predictions in yeast to less than 30% . We also develop a framework to evaluate the statistical significance of our motif predictions . We show that our approach is robust to the choice of input promoters , and thus can be used in the context of predicting binding positions from noisy experimental data . We apply our method to identify binding sites using data from genome scale ChIP–chip experiments . Results from these experiments are publicly available at http://cagt10 . bu . edu/BSG . The graphical framework developed here may be useful when combining predictions from numerous computational and experimental measures . Finally , we discuss how our algorithm can be used to improve the sensitivity of computational predictions of transcription factor binding specificities .
Transcription factors ( TFs ) bind short stretches ( usually 6–18 bp ) of DNA near the gene's transcription start site . This event is thought to facilitate regulation of expression of the downstream gene through TF interaction with the RNA polymerase and other factors in the pre-initiation complex [1] . Computational identification of transcription factor binding sites ( TFBS ) remains one of the most challenging and important problems at the interface of computational and experimental research . In general , research in a diverse array of fields from biophysics to systems biology often depends on the ability to accurately identify TF binding propensities and positions . For example , several models of promoter architecture require knowledge of binding locations to identify transcriptional logic gates [2] defined , in part by the relative binding positions of TFs [3 , 4] . The main in vivo approaches to TF binding site determination are variants of ChIP–chip assays , and DNA footprinting . The former , which is essentially a high-throughput version of the latter , can identify approximate location of binding , usually accurate enough to within the length of a promoter [5 , 6] . Footprinting can provide exact binding positions , but it is not easily generalized to high-throughput studies [7] . Thus , to identify binding positions at the genomic scale , researchers often combine high-throughput ChIP–chip experiments with computational algorithms to predict TF binding sites and nucleotide affinities . Developing TF:DNA binding models from first principles , however , is complicated by limited understanding of mechanisms governing transcription factor binding and subsequent transduction of the polymerase assembly . Instead , several empirically derived models have been proposed to identify biologically relevant stretches of promoter regions [8–12] . Most computational algorithms depend on experimental assays to identify sets of co-regulated genes and work by recognizing over-represented , short stretches of DNA . A recent evaluation of some of these algorithms shows that computational treatment of TFBSs is a largely unsolved problem , with the majority of tested algorithms predicting less than 50% of binding sites correctly [13] . Several well-known pitfalls intrinsic to both the biological and computational sides of this problem plague algorithmic identification of binding positions . First , the possible space of solutions is very large , while heuristic approaches often identify local optima [14] . Even if the algorithms could reliably identify global optima , empirically derived scoring functions do not reliably select biologically significant binding sites . Furthermore , the number of bound sites is close to that which could occur by random chance given the length of most eukaryotic promoters [15] , making identification by statistical over-representation challenging . The variability in DNA sequence that retains TF function and allows regulation of the expression of the downstream gene is unknown . While distance relative to the transcription start site was recently shown to be important [16] , this observation is not specific enough to apply in an algorithmic sense to TFBS identification . Finally , while the range of widths that TFs bind is largely accepted to be between 6 and 18 bp , an unbiased estimation for the width of the sequence specific to individual TFs has proven especially difficult [17 , 18] . Recent innovations in computational TF motif prediction have attempted to incorporate orthogonal information to improve predictions . Position-specific mutation models [19 , 20] , co-occurrence of binding sites for multiple TFs [21] , and phylogenetic conservation [20 , 22 , 23] , among other approaches [11 , 24] , have been proposed as additional measures . While all these measures can be shown to improve either accuracy or coverage of computational predictions , most introduce biases that may narrow their applicability . For example , requiring strict phylogenetic conservation automatically excludes identification of evolutionary changes of transcriptional regulation [16] , and those that rely on co-occurrence of different sites do not help with identifying binding of a single TF of interest . Several researchers have also outlined a strategy utilizing the consensus from a variety of programs to improve accuracy [5] . However , the improvement in accuracy of predictions from adopting this approach has not been rigorously quantified and consequently not well understood . Here , we build on existing computational approaches to improve prediction of TF binding positions without adding additional biases that may narrow the scope of application . We recently showed [25] that extensive repetition of Gibbs sampling on the same set of upstream promoters , termed ensemble Gibbs sampling ( see Methods ) , yields a power-law distribution of hits per nucleotide in each promoter: few nucleotides are selected very frequently , while the majority of nucleotides ( also representing the majority of the Gibbs sampling results ) are identified infrequently and do not correspond to biologically relevant results . Simple positional clustering to select the most frequently recurring nucleotides improves accuracy of TFBS identification [25] . Here we show positional clustering can be substantially improved by considering joint occurrences of nucleotides in the same motif . These joint probabilities can be represented as a binding site graph ( BSG ) . Using a well-established benchmark , we compare the predictive power of BSGs to 13 other TFBS prediction algorithms [13] . On yeast datasets , BSGs significantly outperform all previously evaluated algorithms in nearly every measure . In particular , the high percentage of correct predictions ( PPV , positive predictive value ) indicates that the approach is useful for directing downstream experimental research . Performance on non-yeast benchmarks , however , is dramatically worse , signifying that more research is required to reliably predict fly and mammalian regulatory motifs . We also find that BSG predictions are robust to the choice and length of input promoters , and thus more likely to succeed with limited or noisy experimental data . Encouraged by performance on yeast benchmarks , we use BSGs to predict the condition-specific nucleotide specificity for most known TFs in the Saccharomyces cerevisiae genome . Predictions for previously characterized TFs closely agree with previous experimental and computational studies . In addition , we predict 53 novel binding specificities , 16 at high statistical significance .
Given an input set of promoters , we construct a BSG from ensemble Gibbs sampling , as shown in Figure 1 . Briefly , for each dataset , we run the sampler until stability 512 times for each motif width 6–18 bp , producing a total of 6 , 656 predictions per dataset . We consider the sampler to reach stability when results do not change over 1 , 250 updates . For each prediction , we add an edge between all pairs of nucleotide positions in the same column of the returned multiple sequence alignment ( Figure 1B ) . Edges recurring in multiple Gibbs sampling results are represented by a single edge with weight w equal to the number of times the edge occurs normalized by the total number of Gibbs sampling predictions . Thus , w ∈ ( 0 , P] , where the maximum edge weight , P ≤ 1 , is the number of times the most frequently recurring edge is observed divided by the total number of edges in the graph . Once the BSG is built , it still remains to predict positions corresponding to functional TFBS . First , analogous to the frequency with which Gibbs sampling identifies a given nucleotide [25] , edge weights are power law distributed ( unpublished data ) , and nucleotides connected with high edge weights are predictive of TFBS ( Figure S1 ) . Second , we hypothesize that transitively connected nucleotides are closely related in sequence space . For example , if Gibbs sampling identified sites 1 and 2 in one run and sites 2 and 3 in another , those sites will have related , but not identical , sequences . We are interested in differentiating the case where the Gibbs sampler identifies random sets of k-mers from the case where the sampler repeatedly predicts the same set of sites . We hypothesize that the latter case corresponds to functional TFBS . This information is represented in BSG by dense clusters of nucleotides connected by high edge weights . The clustering coefficient of a nucleotide k in a BSG measures connectivity within the local neighborhood of k [28] . As the neighborhood of k approaches a clique , where all neighbors are connected , the clustering coefficient approaches 1 . As the neighborhood of k becomes sparse , where no neighbors of k co-occur in Gibbs sampling predictions , the clustering coefficient approaches 0 . The standard definition of the clustering coefficient is limited to unweighted graphs . However , because edge weights are predictive of functional TFBS , we use a modified version of the clustering coefficient that rewards higher edge weights [27] ( Figure 2 ) . To predict TFBS from a BSG , we can use the weighted clustering coefficient to find sets of nucleotides that are densely connected with high-weight edges . We will use a threshold 0 < ρ ≤ P to filter out all edges with inconsequential edge weight ( Figure 3 ) . Since dense clusters are more likely to occur at random in graphs with fewer vertices , simply maximizing on the weighted clustering coefficient is biased toward graphs with the fewest nodes ( Figure S3 ) . To account for this , we include a ( 1 − ρ/ P ) term in our BSGscore to reward larger but perhaps less densely connected subgraphs . Thus , for a BSG G and frequency threshold ρ , we define the BSGscore: where is mean weighted clustering coefficient over all nucleotides in graph G at threshold ρ . We want to select the edge weight threshold ρˆ that maximizes the BSGscore . To turn the resulting graph into predicted binding positions in the promoters , we extract all nucleotides in the BSG connected by an edge with w > ρˆ ( Figure 3B ) . Nucleotides adjacent in the original input promoters are grouped together into seed sequences ( see Methods , Figure 3C ) . Often , we find the seed sequences contain only the most conserved core of the TFBS . To capture important nucleotides at the edges , we extend the seed sequences to include neighboring nucleotides that are also frequently identified by Gibbs sampling , but perhaps do not pass the stringent cutoff of the core positions ( see Methods ) . The extended sequences are then aligned to create a position weight matrix ( PWM ) representing the binding motif . It should be noted that , at this point in the process , the remaining sequences are generally of similar length , and well-conserved . Hence , the primary motivation for using a sampling procedure here is not to define the end points of the alignment . Instead , we use sampling to solve the problem that we do not know the strand orientation of each binding site . The space of all possible permutations of strand orientations is exponential , and it is unfeasible to explore exhaustively for even a moderate number of predictions . Thus , we use sampling to heuristically predict the relative strand orientation of the predictions . Indeed , other procedures such as expectation maximization would also be appropriate here , and Gibbs sampling was chosen as a matter of convenience . To evaluate the statistical significance of a BSG prediction , it is important to understand the behavior of BSGscores under the null hypothesis that no motifs are present in the input set . We estimated the null distribution of BSGscores in yeast by calculating the maximal BSGscore from 429 sets of 7–30 randomly chosen ( without replacement ) yeast promoters . The resulting scores follow a generalized extreme value distribution ( Figure 4; p = 0 . 997 , KS test ) . We found no significant correlation between the size of the random input set and the BSGscore . Using this distribution , we can evaluate the p-value of a BSGscore from a dataset enriched in a TF binding motif ( see Methods ) . However , dramatic differences in promoter architecture between species may mean that an empirically derived background distribution is needed on a per-species basis . Additionally , we leave for future study a statistical evaluation of input sets with multiple motifs . We used BSGs to predict binding sites for the majority of TFs in the S . cerevisiae genome using the latest data from ChIP–chip experiments [5] in a number of experimental conditions . In total , BSGs predict significant binding motifs for 118 TF–condition pairs , representing 93 different TFs . These results compare favorably with the compendium of 124 TF motif predictions presented in MacIsaac et al . [29]: of the 77 TFs with predictions in each set , 59 ( 77% ) are similar ( see Methods ) . In addition , we predict a different motif for 25 of our significant TF–condition motifs , representing 22 TFs ( Figure S5 ) . It should be noted that the number of similar and different motifs combine to be more than the total number of TFs with predictions in both sets . This discrepancy is explained by four TFs ( MOT3 , SFP1 , MSN2 , and MSN4 ) for which we predict condition-dependent motifs . These results are summarized in Table 1 . Several possible reasons for differences in motif predictions include co-regulation of the same set of genes by different TFs , identification of statistically significant , but biologically inert , motifs , as well as false positive predictions . At the same time , the comparison numbers depend on arbitrary motif similarity thresholds ( see Methods ) . Since allowed degeneracy may be TF-specific , using a single cutoff may not be the optimal approach . However , the agreement provides a rough estimate of the consistency between BSG and previously reported experimental and computational results . Finally , BSGs predict motifs for 18 TFs with previously unknown affinities ( Figure S6 ) and fail to make a significant prediction for 47 TFs . Combining BSG predictions with those in MacIsaac et al . [29] gives a total of 142 TFs with significant motif predictions . However , experimental validation may be needed to confirm novel and revised predictions . To better understand the reasons behind improved performance of BSGs over traditional Gibbs sampling , we manually examine select significant ( p < 0 . 1 ) BSG predictions that do not agree with the best scoring Gibbs sampling prediction ( Figure 5 ) . For Gibbs sampling predictions , we chose the motif width that gave the largest MAP score [17] . Here , we only consider predictions that agree with those previously published . We observed two distinct mechanisms by which BSGs improve on Gibbs sampling . In some cases , such as the HSF1 and LEU3 predictions , the best-scoring result obtained from Gibbs sampling represents only a fraction of the final motif . These represent cases where , through integration of several motif widths , the BSGscore correctly identifies the motif width better than the Gibbs sampling MAP score [17] . Indeed , when we disregard the MAP score and manually choose a motif width according to previous predictions [29] , Gibbs sampling identifies the correct motif for HSF1 and LEU3 . In other cases , such as SIP4 , we find manually choosing the correct motif width does not result in prediction of the correct motif by traditional Gibbs sampling . Similarly , the Gibbs sampling width for the prediction of the RDS1 binding motif matches the width of the previously reported prediction , yet the binding motif does not match . Thus , we conclude that the BSGscore contains additional information about correct binding sites not necessarily present in the MAP score used by Gibbs sampling . In particular , the BSGscore considers both the positional information for each motif and the uniqueness of the nucleotides with respect to the rest of the similarly scoring predictions from the same set of upstream regions . By carefully studying the dynamics of BSG building , it may be possible to incorporate these characteristics into an improved Gibbs sampling procedure and score . In cases where multiple TFs act together to coordinately regulate a set of genes , numerous motifs may be enriched in a set of promoters . Preliminary evidence suggests these motifs arise as independent connected components in the filtered BSG . For example , BSGs predict two connected components for STE12 in YPD: the first component corresponds to the known STE12 binding motif; the second component is the binding motif for TEC1 . STE12 and TEC1 are known to act cooperatively to regulate haploid invasive and diploid pseudohyphal growth . Thus , clustering results into disjoint connected components allowed identification of two different TFs in the same input set . This procedure can be used as a predictor of sets of collaborating TFs in cis-regulatory modules . That , in turn , can be used to elucidate major regulatory switches and sets of genes functional in common pathways [30–32] . We benchmarked performance of BSGs against numerous other motif detection algorithms . We used the same datasets described in Tompa et al . and evaluated BSG predictions according to the statistical framework detailed therein [13] . On yeast-specific benchmark sets , high-confidence ( p < 0 . 1 ) BSG predictions significantly outperform all tested methods according to nearly every statistical measure ( Figure 6 ) . In terms of nucleotide correlation coefficient ( nCC ) , an overall measure of correctness , BSG predictions with p < 0 . 1 improve upon the second-best predictions by 19% . The only exception where BSG predictions do not outperform existing techniques is site-level sensitivity ( sSn ) [11] , where Weeder outperforms by 13% . Weeder's sensitivity , however , comes at the expense of many false positive results , as shown by BSGs' significant improvement in sPPV over Weeder ( 72% versus 55% , a 31% improvement ) . Moreover , BSG seems to be the only method that predicts many more true positives than false positives , corresponding to a site and nucleotide PPV ≫ 0 . 50 . Unlike other methods evaluated , BSG performance does not require any manual curation or custom post processing . Benchmarking with the mouse and human datasets , we found that the BSG performed among the best six algorithms in every category except nucleotide specificity , for which BSGs performed poorly; while performance was good , we did not observe the broad improvements obtained in yeast ( Figure S4 ) . We believe the performance drop in non-yeast sets indicates the need to develop species-specific binding site detection strategies . For example , in the human and mouse tests , the TF binding sites have a positional bias toward the transcription start site; the fly examples , however , tend to contain closely spaced clusters of binding sites . Better understanding of these differences in promoter architecture and usage between species will be critical in developing species-specific BSGscores . Another mechanism to assess the efficacy of a TFBS algorithm is to evaluate the effect of added decoy promoters on the stability and accuracy of TFBS predictions [33] . Decoy promoters are intergenic nucleotide sequences that contain no instances of the TFBS of interest , and may arise through false positives in prediction of the input set . The effect of decoy promoters is reduction in the concentration of TF binding sequences in the input set . For example , 20 instances of a 10-bp binding site in 20 upstream regions , each 1 , 000 bp long , results in about 1/100 signal:noise . If we add 20 more upstream regions without instances of the same TFBS , signal:noise would be closer to 1/200 . Decreasing the signal-to-noise ratio confounds identification of binding sites . Robustness to decoy sequences is necessary to make binding site predictions from noisy datasets such as high-throughput microarray experiments . Addition of decoy promoters also effectively simulates longer upstream regions encountered in higher eukaryotes . To evaluate BSG robustness to increasing noise , we first predict TFBS in a core set of promoters that ChIP–chip experiments predict are coregulated by a common TF . We then predict binding sites in versions of the core set augmented by increasing numbers of randomly selected intergenic sequences from the S . cerevisiae genome . We then plot PPV with respect to the relative amount of added noise . We evaluated robustness of BSG predictions to addition of decoy promoters for four input sets ( HAP4 , STE12 , YDR026C , and YAP1 ) . We then compare our results with those that could be expected given an equal amount of independent Gibbs sampling runs without BSG identification of TFBS . We find , for all TFs evaluated , the PPV of using BSG predictions is uniformly superior to Gibbs sampling alone , even for some predictions with p > 0 . 1 . Moreover , the difference between the PPV of BSG and Gibbs sampling alone increases with addition of decoy sequences ( Figure 7 ) . We also use the same framework to compare BSGs with positional clustering of frequently recurring Gibbs sampling results [25] . As shown in Figure 7 , while positional clustering is useful in improving binding predictions using ensemble Gibbs sampling , BSGs allow further improvement . These results indicate BSGs perform better with noisy input sets that could result from long eukaryotic upstream regions or inaccurate predictions of co-regulation . We found that BSG predictions for the PHO4 , CBF1 , and TYE7 TFs are particularly interesting . Despite regulating biologically different processes , all three are Helix–loop–helix proteins that bind the hexameric E-box motif CACGTG . In the case of PHO4 and CBF1 , a high-throughput microfluidics platform able to precisely measure low-affinity TF:DNA interactions [34] revealed differences in the specificity for E-box flanking nucleotides for PHO4 and CBF1 . Previous computational studies [5 , 29] , however , have struggled to identify significant differences in binding affinities . In agreement with the experimentally derived specificities , the BSG is the first high-throughput computational approach able to correctly resolve the differences in specificity of flanking nucleotides for both PHO4 and CBF1 ( gCACGTGG and gTCACGTG , respectively , Table S1 ) . Additionally , BSGs predict an extended TYE7 binding 10mer ( cATCACGTGa , Table S1 ) that differs from both the PHO4 and CBF1 binding motifs in the flanking nucleotides . We searched all yeast promoters for exact matches to the expanded binding motifs . As expected , promoters containing the PHO4 motif were significantly enriched in phosphate transport processes . Exact matches to the revised CBF1 and TYE7 motifs were both significantly enriched in amino acid metabolism; CBF1 motifs , however , were limited to metabolism of nitrogen R-groups , whereas TYE7 motifs were limited to metabolism of cysteine . We take this as preliminary evidence that the newly discovered flanking nucleotides may play a major role in allowing each E-box binding TF to regulate a subset of functionally specific proteins .
Here , we present a novel approach for determining the positions and binding affinities to TFs using putatively bound upstream promoter sequences . BSGs are a departure from traditional sequence alignment techniques such as Gibbs sampling primarily because BSGs capture global properties of promoter input sets that seem to be unique only to sets that share TFBS . This results in several important advantages in predicting TFBS using BSGs . First , according to most independent validation criteria , BSGs are more accurate than existing techniques . Additionally , we find BSGs are more robust to noisy decoy sequences than Gibbs sampling with and without positional clustering . Importantly , positional clustering provides an intermediate level of improvement over Gibbs sampling alone . This result suggests the improved performance of BSG is due to a combination of ensemble sampling and analysis of graph-theoretical properties of BSGs [25] . Robustness to decoy sequences may allow BSGs to better predict binding sites from co-expression data , which is more prone to false positive predictions than ChIP–chip , and does not necessarily result in gene sets co-regulated by a single TF . Second , BSG construction and cluster extraction algorithms provide an unbiased estimation of motif width that is better than those based on currently available scoring functions . This can be seen from examples with HSF1 and LEU3 ( Figure 5 ) . Finally , comparing BSGscores to a background distribution from graphs constructed for random sets of promoters enables calculations of statistical significance and identification of promoter sets lacking significant motif enrichment or alternatively those that have a high level of noise . In agreement with earlier observations in synthetic data [35] , our results suggest that unlike random promoter sequence sets , input promoter sets enriched in binding by a common TF have densely connected clusters in sequence space . While we chose to use a simple formulation of the weighted clustering coefficient to identify these clusters , other graph clustering approaches can be used to improve binding specificity predictions from BSGs . In their previous work , Pevzner and coworkers suggested using graphical models to predict TFBS [35] . In that work , the authors dissected a simple formulation limited to exactly one binding site per promoter , a fixed-motif width , and a maximum number of mutations per binding site . The authors proposed using graphs that form cliques to identify TFBS . While useful formulations from a theoretical perspective , constraints presented in that paper are limiting from a practical point of view . Our BSG approach does not make any of the above assumptions on motif structure or occurrence . Thus , we were able to apply BSGs to real datasets and successfully identify binding positions with superior accuracy . The graph-construction technique described here uses ensemble Gibbs sampling across a range of motif widths . We observed that sampling at widths close to the biologically relevant motif width will contribute higher-edge weights to the final graph than sampling far from the biological motif width , which mostly contributes nucleotides at the edges . Combining predictions for each motif width , we can predict the width of the biological motif . According to case studies of LEU3 and HSF1 , this strategy results in more accurate identification of motif widths as compared with existing scores . Alternatively , we could evaluate a graph for each possible motif width , and select predictions from the best-scoring graph . Constructing a graph for each motif width , however , would require the ensemble sampling procedure to be repeated many times ( once for each width of interest ) . Doing so is computationally infeasible with available technology; we leave a comprehensive analysis of this strategy for a future study . Ultimately , using ensemble Gibbs sampling to build BSGs is limited by the sensitivity of Gibbs sampling; thus , constructing BSGs using sampling from more sensitive , faster , or a combination of algorithms [25] may improve performance . BSGs can also aid in integrating ensemble sampling with diverse biological data such as distance from transcription start site; histone localization; free radical cleavage; DNA bending; and phylogenetic conservation into a coherent , unified framework for identifying TFBS [36] . Finally , we used BSGs to predict nucleotide specificity for the majority of TFs in the S . cerevisiae genome using input sets generated from the recently performed whole-genome ChIP–chip experiments . We found some interesting patterns that may be used to control the quality of the data or further our understanding of the interactions between coordinately acting TFs . For example , numerous sets of TFs , based both on our predictions and those of other independent studies , have very similar nucleotide specificity ( for example: STE12 and DIG1; PHO4 , CBF1 , and TYE7 ) . In the case of STE12 and DIG1 , protein domain analysis indicates the lack of a known DNA binding domain in one of the proteins ( DIG1 ) , and experimental evidence shows that STE12 and DIG1 physically interact [37] . As such , it is likely that DIG1 does not directly bind DNA , but instead co-precipitates with STE12 through formaldehyde crosslinking of protein–protein interactions in ChIP–chip experiments [38 , 39] . Motif similarity may also stem from cooperative or competitive binding between the factors . Importantly , the increased accuracy of BSG predictions allowed us to predict specificity-determining nucleotides for the E-box TFs PHO4 , CBF1 , and TYE7 [34] . Two out of the three extended predictions were independently confirmed by recently published experimental results . The third is awaiting further validation .
We use a threshold Gibbs sampling strategy similar to BioProspector [17] . Briefly , the threshold sampling strategy uses a high threshold to allow inclusion of multiple sequences per promoter , and a low threshold to allow reporting of no sequences in a promoter . The high threshold is set proportional to the product of the average promoter length and the window width , while the low threshold is initialized to 0 , and increased linearly to an upper bound . For a complete description of threshold sampling , see Liu et al . [17] . Additionally , we include a third-order background model from genomic promoters , and a modified motif score ( instead of pi , j ) , which we found better emphasized conservation within motif predictions: where N is the number of aligned segments in the motif , pi , j is the probability of nucleotide j at position i in the motif , and qj is the probability of nucleotide j in the third-order background [17] . We sample until stability ( predictions do not change over 1 , 250 updates ) 60 iterations and select the single best-scoring motif observed as a single prediction . We found running the sampler for more stable updates did not significantly alter results . To evaluate the ensemble Gibbs sampling predictions for a set of input promoters , we mask low-complexity sequences , and proceed to collect 512 Gibbs sampling predictions at each motif width from 6–18 bp . In total , 6 , 656 binding site predictions are collected from 60 × 6656 = 399 , 360 Gibbs sampling iterations . The number of predictions used was selected to ensure stability in graph construction , and we found performance deteriorated significantly when constructing graphs from fewer sampling predictions . High-performance computing was utilized to perform the ensemble sampling , requiring between 30 min and 5 h of running time on 1024 × 700 Mhz PowerPC 440 processors . While BSG construction currently requires access to high-performance computing , advancements in the algorithm , Gibbs sampling , and computer technology may all help to make the approach more accessible . A BSG is a weighted , undirected graph G: = ( V , E ) where each vertex v ∈ V corresponds to a nucleotide in the input set of promoters , and each edge e ∈ E indicates the alignment of a pair of nucleotides in a binding site for the same TF . Each edge e has weight we that measures the similarity between nucleotides as estimated using Gibbs sampling . We also introduce a threshold BSG , constructed by removing all edges with weight less than threshold ρ ∈ [0 , 1] from graph G . Formally , Gρ: = ( Vρ , Eρ ) where Eρ ⊆ E and e ∈ Eρ ↔ we ≥ ρ; and Vρ ⊆ V and v ∈ Vρ ↔ {v has at least one edge in Eρ} . For a set of input promoters , we initialize the BSG with one vertex for each nucleotide in the input set , and with no edges ( Figure 1A ) . We evaluate the ensemble behavior of Gibbs sampling over the input . For each pair of aligned nucleotides in each sampling result , we add a unit weight edge between the corresponding vertices in the BSG ( Figure 1B ) . Therefore , if a binding site prediction aligns N segments , each w nucleotides long , we add w · n ( n − 1 ) / 2 edges to the BSG . If an edge already exists , we instead increase the edge weight by 1 . Thus , we weigh each edge by the number of times ensemble Gibbs sampling predictions align the corresponding nucleotides ( Figure 1C ) . After collecting edges from all predictions , we normalize edge weights to [0 , 1] through division by the maximal possible edge weight ( i . e . , the number of Gibbs sampling results collected ) . For a vertex k , let v be the number of vertices adjacent to k , and let t be the number of triangles containing k . The clustering coefficient [28] , is defined as: Intuitively , the clustering coefficient is the probability that any two vertices adjacent to k have an edge between them . In the dense extreme , when k resides in a clique , all vertices adjacent to k have an edge between them and the clustering coefficient is 1 . In the sparse extreme , when k resides in a tree , no edge exists between any two neighbors of k and the clustering coefficient is 0 . The clustering coefficient is undefined when k has less than two adjacent vertices; in such cases , we let C ( k ) = 0 . Numerous generalizations of the clustering coefficient to weighted graphs have been proposed [40 , 41] . We use a definition that weights each triangle by its intensity [27]: where wij is the weight of the edge connecting vertices i and j . The weighted clustering coefficient of a graph G is the average weighted clustering coefficient over the vertices in G [28]: We predict binding sites from a BSG G using a two-step process . First , we select a threshold ρˆ that maximizes the BSGscore , where , P is the maximal edge weighted observed in G , and is the mean weighted clustering coefficient of the BSG filtered at ρ ( see above ) . A final BSG is then created by discarding all edges with weight < ρˆ , and the remaining nucleotides ( V ρˆ ) are collected . To convert the collected nucleotides into binding sites , nucleotides adjacent in the original input promoters are joined together into contiguous segments . The segments then serve as seeds for TF binding site predictions . We dust-filter single nucleotide segments from the seeds , and expand the remaining seeds according to a seed extension threshold τ . To do so , we evaluate the strength of each nucleotide n in the unfiltered BSG , where wn , i is the edge weight between nucleotides n and i , and an , i is a delta function equal to 1 when an edge exists between n and i , and equal to 0 otherwise [41] . The strength of each nucleotide is normalized to si ∈ [0 , 1] . Nucleotides adjacent in the original input promoters are grouped together into seed sequences . Initial seeds are then extended to include adjacent 5′ and 3′ nucleotides with s > τ . At the most sensitive extreme ( τ = 0 ) , initial seeds are maximally extended by including all neighboring nucleotides identified by Gibbs sampling . At the most specific extreme ( τ = 1 ) the initial seeds are returned , without extension , as predictions . Between these extremes , a tradeoff between sensitivity and PPV is made ( Figure S2 ) . For the purposes of this study , we used a seed extension threshold of τ = 0 . 7 , but other values may be more appropriate for different research needs . Lastly , the extended seeds are dust-filtered to remove predictions 6 bp or shorter . We evaluate BSG performance using the datasets and statistical measures described in Tompa et al . [13] . Briefly , Tompa et al . create a number of synthetic and real input promoter sets with known binding sites . While the study evaluates input sets from four species ( Homo sapiens , Mus musculus , Drosophila melanogaster , and S . cerevisiae ) , performance of each tool on non-yeast datasets was dramatically worse and for poorly understood reasons . Thus we limit our evaluation to the better understood S . cerevisiae input sets . We predict binding sites for each input set , and evaluate results according to a number of statistical measures . We compare performance in each measure with the published evaluations of 13 existing methods . We calculate statistical measures as follows . At the nucleotide level , true and false positives and negatives ( nTP , nTN , nFP , nFN ) are counted through comparison with nucleotides in the known binding sites in each input set . At the site level , true positives , false positives , and false negatives ( sTP , sFP , sFN ) are counted . A true positive ( sTP ) is defined as a predicted site that overlaps a known site for at least 25% of the known site . Based on these counts , we calculate the following nucleotide ( x = n ) and/or site ( x = s ) level measures: Sensitivity: xSN: = xTP/{xTP + xFN} Positive Predictive Value: xPPV: = xTP/{xTP + xFP} Specificity: nSP: = nTN/{nTN + nFP} Correlation coefficient [42]: Performance coefficient [35]: nPC: = nTP/{nTP + nFN + nFP} Average Site Performance: sASP: = {sSn + sPPV}/2 For a more detailed discussion of statistical measures used , see [13] . We use a PWM to represent TF binding motif predictions [43 , 44] . To align PWMs , we use a dynamic programming implementation of a modified ungapped local sequence alignment [45] similar to that of Pietrokovski [46] . Similarity between positions in two motifs was measured using Pearson's correlation coefficient: Where Xi is the distribution of nucleotides at position i in motif X; σXi is the variance at position i in motif X; and cov ( Xi , Yj ) is the covariance of nucleotides at position i in motif X with nucleotides at position j in motif Y . Alignment scores range from 0 , representing no positions aligned , to the length of the shorter of the two motifs , representing a perfect match between the two PWMs . A BSG PWM was considered to match to a previously published PWM if the ratio of the above alignment score ( i . e . , the optimal local Pearson's correlation coefficient ) to the information content of the previously published PWM [47] is greater than 0 . 375 . It is important to note that the correlation coefficient is at most w , with width of the alignment , whereas the mutual information is at most 2w . We use ChIP–chip assays [38] to identify sets of S . cerevisiae promoters bound with high confidence ( p < 0 . 001 ) by the TFs STE12 , HAP4 , and YDR026C in YPD growth media; and YAP1 in low hydrogen-peroxide conditions [5] . For each set , we collect promoter sequences ( up to 1 kb upstream ) to serve as a seed input for binding site predictions . We construct noisy input sets by supplementing each seed set with increasing numbers of randomly chosen S . cerevisiae promoters . We predict TF binding sites in the seed and noisy sets using both BSGs and Gibbs sampling . We label results as true or false positive ( TP , FP ) according to motif–motif alignment scores between the predicted and the known TF binding motif [25] , and calculate the PPV: i . e . , the percentage of predictions similar to the known binding motif . We use ChIP–chip data [5 , 6] to create input sets for all TFs under every condition studied , as described previously . We use BSGs to predict TF binding motifs for each set containing more than four bound probes . The results of the genomic study are available online at http://cagt10 . bu . edu/BSG . | A historically difficult problem in computational biology is the identification of transcription factor binding sites ( TFBS ) in the promoters of co-regulated genes . With increasing emphasis on research in transcriptional regulation , this problem is also uniquely relevant to emerging results from recent experiments in high-throughput and systems biology . Despite extensive research in the area , recent evaluations of previously published techniques show much room for improvement . In this paper , we introduce a fundamentally new approach to the identification of TFBS . First , we start by representing nucleotides in promoters as an undirected , weighted graph . Given this representation of a binding site graph ( BSG ) , we employ relatively simple graph clustering techniques to identify functional TFBS . We show that BSG predictions significantly outperform all previously evaluated methods in nearly every performance measure using a standardized assessment benchmark . We also find that this approach is more robust than traditional Gibbs sampling to selection of input promoters , and thus more likely to perform well under noisy experimental conditions . Finally , BSGs are very good at predicting specificity determining nucleotides . Using BSG predictions , we were able to confirm recent experimental results on binding specificity of E-box TFs CBF1 and PHO4 and predict novel specificity determining nucleotides for TYE7 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"mathematics",
"saccharomyces",
"genetics",
"and",
"genomics",
"computational",
"biology"
]
| 2007 | Binding Site Graphs: A New Graph Theoretical Framework for Prediction of Transcription Factor Binding Sites |
Human hepatitis B virus ( HBV ) causes chronic hepatitis and is associated with the development of hepatocellular carcinoma . HBV infection alters mitochondrial metabolism . The selective removal of damaged mitochondria is essential for the maintenance of mitochondrial and cellular homeostasis . Here , we report that HBV shifts the balance of mitochondrial dynamics toward fission and mitophagy to attenuate the virus-induced apoptosis . HBV induced perinuclear clustering of mitochondria and triggered mitochondrial translocation of the dynamin-related protein ( Drp1 ) by stimulating its phosphorylation at Ser616 , leading to mitochondrial fission . HBV also stimulated the gene expression of Parkin , PINK1 , and LC3B and induced Parkin recruitment to the mitochondria . Upon translocation to mitochondria , Parkin , an E3 ubiquitin ligase , underwent self-ubiquitination and facilitated the ubiquitination and degradation of its substrate Mitofusin 2 ( Mfn2 ) , a mediator of mitochondrial fusion . In addition to conventional immunofluorescence , a sensitive dual fluorescence reporter expressing mito-mRFP-EGFP fused in-frame to a mitochondrial targeting sequence was employed to observe the completion of the mitophagic process by delivery of the engulfed mitochondria to lysosomes for degradation . Furthermore , we demonstrate that viral HBx protein plays a central role in promoting aberrant mitochondrial dynamics either when expressed alone or in the context of viral genome . Perturbing mitophagy by silencing Parkin led to enhanced apoptotic signaling , suggesting that HBV-induced mitochondrial fission and mitophagy promote cell survival and possibly viral persistence . Altered mitochondrial dynamics associated with HBV infection may contribute to mitochondrial injury and liver disease pathogenesis .
Hepatitis B virus ( HBV ) infection affects nearly 350 million people worldwide and leads to chronic liver disease , liver failure , and hepatocellular carcinoma ( HCC ) [1] , [2] . HBV is an enveloped DNA virus that belongs to the Hepadnavirus family . HBV DNA genome encodes four overlapping open reading frames designated as pre-S/S ( the hepatitis B surface antigen , HBsAg ) , C ( core/e antigen , HBc/eAg ) , P ( polymerase , reverse transcriptase ) , and X ( HBx ) [1] , [2] . HBx is a regulatory protein with multiple functions involved in various cellular and physiological processes including a key role in the maintenance of viral replication [3] . It is predominantly localized to the cytoplasm and also associates with mitochondria via its interaction with voltage-dependent anion-selective channel 3 ( VDAC3 ) [4]–[7] . This association leads to a decrease in mitochondrial transmembrane potential ( ΔΨm ) and depolarization of mitochondria [3] , [4] , [8] . HBx also participates in activating transcription of whole host of cellular genes via protein-protein interactions both in the nucleus and cytoplasm [3] , [7] , [9]–[11] . HBx is not directly oncogenic but participates substantially in the process of liver oncogenesis [9] , [12] . HBx is a regulatory protein with pleiotropic activities and has been shown to promote endoplasmic reticulum ( ER ) stress , oxidative stress , deregulation of cellular calcium homeostasis , and mitochondrial dysfunction [3] . HBx also modulates the activation of several latent transcription factors such as nuclear factor-kappa B ( NF-κB ) , signal transducer and activator of transcription 3 ( STAT-3 ) , with resultant activation of cytoprotective genes [8] , [13] . The multiple effects of HBx protein may be a consequence of the trigger of the ER-mitochondria-nuclear nexus of signal transduction pathways . Mitochondrial injury and oxidative stress are prominent features of chronic Hepatitis B and C [14] , [15] . Histological manifestation of swollen mitochondria and mitochondria lacking cristae directly implicates mitochondrial injury in HBV-associated liver disease pathogenesis . HBV infection is associated with deregulated cellular Ca2+ signaling , mitochondrial depolarization and dysfunction and reactive oxygen species ( ROS ) generation [3] , [8] , [16] . HBV-induced elevated cellular ROS levels can also promote mitochondrial dysfunction [15] . Dysfunctional or damaged mitochondria trigger a vicious cycle of mitochondrial damage and ROS generation , which is detrimental for cell survival and can be confounded by rapid turnover of damaged mitochondria [17] . The removal of dysfunctional mitochondria is orchestrated by asymmetric mitochondrial fission to eliminate the damaged mitochondria by subsequent mitophagy ( selective autophagy of mitochondria ) [18] . Mitochondria subjected to physiological stress usually undergo perinuclear clustering , which precedes both mitochondrial fission and mitophagy [19] . HBV and in particular , HBx have been shown to induce bulk autophagy [20]–[24] . In this study , we investigated HBV-induced aberrant mitochondria dynamics , and mitophagy . Our data revealed that HBV shifts the balance of mitochondrial dynamics towards enhanced fission and promotes selective autophagic degradation of damaged mitochondria via mitophagy . HBV triggered mitochondrial fission by promoting mitochondrial translocation of Drp1 via upregulation of Drp1 Ser616 phosphorylation . HBV upregulated proteins that mediate mitophagy and induced the elimination of dysfunctional mitochondria via mitophagy . More specifically , mitochondrial translocation of Parkin a cytosolic E3 ubiquitin ligase was observed in HBV/HBx expressing cells resulting in its self-ubiquitination and of its substrate , mitofusin 2 ( Mfn2 ) . In addition to confocal microscopy , using a novel dual fluorescence reporter Mito-mRFP-EGFP , we demonstrated that HBV/HBx induces complete mitophagy evident by fusion of mitophagosome with lysosome . Our studies also showed that HBx protein alone or in the context of HBV full genome , is a critical activator of HBV-induced aberrant mitochondrial dynamics . Further , we demonstrated that inhibition of mitophagy by silencing Parkin results in enhanced mitochondrial apoptotic signaling in HBV-infected cells , suggesting that induction of mitochondrial fission and subsequent mitophagy subvert apoptosis impending due to accrued mitochondrial injury in HBV-infected cells . In summary , our results suggest that HBV-mediated modulation of mitochondrial dynamics may promote cell viability of infected cells . We envisage that the altered mitochondrial dynamics and induction of mitophagy possibly contribute to the persistence of HBV-infected hepatocytes . However a careful and rigorous examination of HBV-induced mitochondrial regulation and its relevance to persistent phenotype of infected hepatocytes is required to confirm this finding in in vivo conditions .
We investigated the HBV-induced morphological changes of mitochondria in the human hepatoma Huh7 cells transiently expressing wild-type 1 . 3mer HBV genome ( hereafter referred to as HBV ) . As shown in Figure 1A , distinct fragmented mitochondrial morphology was observed in HBV-expressing cells , compared to the typical tubular mitochondria in untransfected cells . HBx-expressing cells also displayed similar mitochondrial fragmentation ( Figure 1B ) . HBV/HBx-expressing cells displayed prominent mitochondrial clustering in the perinuclear regions ( Figure 1A and 1B ) , consistent with a previous report [19] . We then determined whether HBV infection triggered Drp1-mediated mitochondrial fission . As shown in Figure 1C , HBV stimulated both the expression and phosphorylation ( Ser616 ) of Drp1 . Mitochondrial translocation of Drp1 is modulated by phosphorylation at Ser616 by cyclin B/cyclin-dependent kinase 1 ( Cdk1 ) [25] . HBV gene expression has been shown to stimulate cyclin B/Cdk1 [26]–[30] . Similar results were obtained in HBx-expressing cells ( Figure S1C ) . Next , we examined if HBV induces Drp1 translocation to mitochondria by confocal microscopy . HBV-expressing cells displayed enhanced mitochondrial translocation of Drp1 , compared to uninfected cells . ( Figure 1D , see merged yellow spots ) . Using Drp1 antibody that specifically recognizes phosphorylated Ser616 residue , we demonstrated that most of Drp1 recruited to the mitochondria in HBV-infected cells is phosphorylated at Ser616 residue ( Figure 1E ) . Similar results were obtained in HBx-expressing cells ( Figure 1F ) . In support of these results , we demonstrated the accumulation of phosphorylated Drp1 in purified mitochondrial fraction from HepAD38 cells that stably express whole HBV genome under tetracycline-repressible promoter ( Figure 1C , bottom panel ) [31] . Together , these results indicate that HBV/HBx promotes mitochondrial fission via Drp1 translocation . When mitochondria are depolarized , PTEN-induced putative kinase 1 ( PINK1 ) , a mitochondrial Ser/Thr kinase , accumulates on the outer mitochondrial membrane and recruits Parkin to the mitochondria [18] . Parkin translocation to depolarized mitochondria is a hallmark of mitophagy [18] , [32] , [33] . Thus , we examined mitochondrial translocation of Parkin in HBV-expressing cells by confocal microscopy . Significant Parkin translocation to mitochondria was observed in HBV-expressing cells , compared to untransfected cells ( Figure 2A , see merged yellow spots ) . Similar results were also observed in HBx-expressing cells ( Figures S1A and B ) . Parkin is an E3 ubiquitin ligase which ubiquitinates itself and its mitochondrial substrates , Mfn2 and voltage-dependent anion-selective channel 1 ( VDAC1 ) [18] . Mfn2 functions to promote mitochondrial fusion and its degradation by HBV will favor fission activities [34] . Using purified mitochondrial and cytosolic fractions of HBV-expressing cells , we further analyzed mitochondrial translocation of Parkin . As shown in Figure 2B , Parkin is accumulated and mostly ubiquitinated in the mitochondrial fraction . Parkin ubiquitination was further analyzed by immunoprecipitation with anti-Parkin antibody , followed by subsequent Western blotting with anti-ubiquitin antibody ( Figure 2C ) . We also observed a moderate decrease in the expression levels of VDAC1 ( Figure 2B , lanes 2 and 6 ) , whereas Mfn2 is significantly degraded with a concomitant increase in its ubiquitinated form ( Figure 2D , first and third panels , lane 3 ) . Parkin-dependent ubiquitination of Mfn2 in HBV-expressing cells was verified by immunoprecipitation assay using HBV-expressing cells with Parkin knockdown ( Figure 2D , upper panel , lane 4 ) . These results indicate that HBV stimulates Parkin translocation to mitochondria and Parkin-dependent degradation of Mfn2 by ubiquitination . We then investigated whether HBV stimulates the expression of mitophagy-related genes . As shown in Figures 2E and F , HBV gene expression modestly stimulated the expression of Parkin , PINK1 , and microtubule-associated protein 1 light chain 3B ( LC3B ) at both the mRNA and protein levels in HepAD38 cells , but not in HepG2 cells , the parental cell line of HepAD38 . An increase in LC3B-I and LC3B-II was also observed ( Figure 2F ) . These results were further confirmed in Huh7 cells transiently expressing HBV genome ( Figure 2G ) . Activating transcription factor 4 ( ATF4 ) was also stimulated by HBV . ATF4 is a transcriptional factor that stimulates Parkin gene expression via unfolded protein response ( UPR ) [35] ( Figure 2F ) . HBx-expressing cells showed a similar stimulation of Parkin , LC3B-I , and LC3B-II ( Figure S1C ) . By coimmunoprecipitation assay using whole cell lysates extracted from Huh7 cells co-expressing Flag-HBx and mCherry-Parkin protein , physical interaction between HBx and Parkin is shown in Figure S1D . HBx-VDAC3 interaction has been previously reported [4] . Since Parkin binds VDAC [36] , it is tempting to speculate that Parkin , VDAC , and HBx may form a ternary complex to expedite the process of Parkin recruitment to mitochondria . In summary , these results demonstrate that HBV/HBx stimulated the expression of mitophagy-related genes . Next , we analyzed the formation of mitophagosome in Huh7 cells co-expressing HBV and GFP-LC3 protein by confocal microscopy , as we demonstrated the HBV-induced mitochondrial fission and stimulation of Parkin and LC3B expression in Huh7 cells ( Figures 1 and 2G ) . As shown in Figure 3A , we observed that Parkin-containing mitochondria are associated with GFP-LC3 puncta in HBV-expressing cells ( see white puncta in the zoomed images ) . These GFP-LC3 puncta were abrogated upon treatment of cells with autophagy inhibitor 3-methyladenine ( 3-MA ) , but not Bafilomycin A1 ( BafA1 ) ( Figure 3A ) . 3-MA inhibits autophagosome formation , whereas BafA1 inhibits fusion of autophagosomes with lysosomes [37] , [38] . Quantitative analysis of these results is presented in Figures 3B , C , and D . HBx alone was also capable of forming Parkin-associated mitophagosome ( Figure S2A ) . In contrast , cells expressing whole HBV genome defective in HBx expression ( HBV-ΔX ) failed to show any GFP-LC3 puncta containing mitochondria , ( Figure S2 ) . Quantitative analysis of these images is also presented in Figures S2C , D , and E . Together , these results indicate that HBV/HBx induces Parkin-dependent mitophagosome formation . The final step of mitophagy is the fusion of mitophagosomes with lysosomes where the cargo is delivered , degraded and recycled [18] . To analyze the progression of mitophagy from mitophagosomes to lysosomes , we made use of a tandem-tagged RFP-EGFP chimeric plasmid pAT016 encoding a mitochondrial targeting signal sequences fused in-frame with RFP and EGFP genes ( Figure 4A ) , which exploits the differential stabilities of RFP and GFP [39] . GFP signal is quenched in lower pH , while RFP can be visualized in both mitophagosomes and acidic mitophagolysosomes thus the prevalence of RFP fluorescence in the lysosomes indicates completion of mitophagic process . In contrast , under normal conditions , yellow structures indicating merged images of GFP and RFP that localize to mitochondria are observed ( Figure 4A ) . Cells cotransfected with plasmid pAT016 ( mito-mRFP-EGFP ) and whole HBV genome or , HBV-ΔX , or HBx-flag expressing plasmid , respectively , were analyzed by confocal microscopy . In control cells not expressing HBV or expressing HBV-ΔX , the yellow merged fluorescence that indicates the presence of both EGFP and mRFP in mitochondria was observed ( Figure 4B ) . In contrast , the cells expressing whole HBV genome or HBx displayed distinct red puncta , and fewer green puncta because mRFP is more stable in the lysosomes [39] . When EGFP and mRFP signals are merged , only red puncta were observed in mitophagolysosomes as EGFP signal is quenched in lower pH ( Figure 4B ) . Quantitation of these puncta is presented in Figure 4C . We continued to examine HBV/HBx-induced mitophagolysosome formation by conventional confocal microscopy . As shown in Figures 4D and S3A , GFP-LC3 puncta-containing mitochondria associated with lysosomes were observed in both HBV- and HBx-expressing cells ( see white puncta indicating merged images of GFP-LC3 , TOM20 , and LysoTracker ) . Treatment with 3-MA and BafA1 abrogated these puncta ( Figures 4D and S3A ) . Quantitative analysis of these results is shown in Figures 4E and S3B , respectively . Collectively , these results demonstrate that HBV and HBx , either expressed alone or in the context of whole HBV genome , respectively , induce complete mitophagy . As damaged mitochondria are eliminated via mitophagy , we observed a decline in mitochondrial number in HBV-expressing cells , but not in those treated with BafA1 or untransfected cells ( Figures S3C and D ) . Parkin knockdown in HBV-expressing cells also abrogated the decline in mitochondrial number ( Figures S3E and F ) . It should be emphasized that not all mitochondria in HBV/HBx-expressing cells engage in mitophagy . The fraction that undergoes mitophagy and fission is likely to impact the disease process . Mitochondrial dynamics is integrally linked to apoptosis [40] , [41] . We investigated if HBV enhances mitochondrial fission and mitophagy to modulate apoptotic cell death associated with mitochondrial injury . HepAD38 cells were transfected with Parkin-specific siRNA pool and analyzed for changes in mitochondrial apoptotic signaling pathway . Parkin silencing induced massive cytochrome C release from mitochondria to cytosol and promoted cleavage of poly ( ADP-Ribose ) polymerase ( PARP ) and caspase-3 , activation of caspase-3/7 , and prompted apoptosis as seen by TUNEL assay ( Figures 5A–C ) . These results strongly suggest that HBV-induced mitochondrial dynamics protects virus-infected hepatocytes from apoptotic cell death to facilitate the persistent virus infection ( Figure 5D ) .
Mitochondria are dynamic organelles that undergo fission ( fragmented mitochondria ) , fusion ( tubular mitochondrial network ) , and trafficking [18] . The rapid modulation of mitochondrial dynamics occurs in response to physiological stress , apoptotic stimuli , metabolic demands , and infections [42] . Perturbation in mitochondrial dynamics is involved in many diseases such as neurodegenerative disorders and cardiovascular diseases , underscoring the pivotal importance of this process in maintenance of mitochondrial and cellular homeostasis [42] . Clearance of damaged mitochondria is proposed is orchestrated by asymmetric mitochondrial fission and subsequent elimination of damaged mitochondrial pool by selective mitochondria autophagy ( mitophagy ) [18] . Fragmented mitochondria are better substrates for removal by mitophagy and elongated or fused mitochondria resist mitophagic degradation , suggesting that mitochondrial dynamics and mitophagy are two critical arms required for maintenance of mitochondrial homeostasis [18] , [43] . HBV gene expression is associated with physiological aberrations such as perturbed calcium homeostasis and elevated ROS levels which promote mitochondrial dysfunction and damage [3] , [6] , [8] . We observed that HBV/HBx expression triggers sequential cascade of events with mitochondria , initiating with their perinuclear accumulation followed by mitochondrial fission and ultimately the clearance of damaged mitochondria by mitophagy . Clearance of damaged mitochondria via mitophagy is crucial for establishing mitochondrial homeostasis and cell survival [18] , [43] . Impairment of this process by mutations in mitophagy-mediating genes such PINK1 or Parkin is linked to hereditary forms of Parkinson's disease , a neurodegenerative disorder , emphasizing the vital role of mitochondrial homeostasis in cell survival [44] , [45] . Many DNA and RNA viruses have been shown to tightly modulate the autophagy process to prevent their clearance by autophagy , to inhibit host immune response , or to favor viral replication and maturation events [46] . Hepatitis C virus , a positive-sense single-stranded RNA virus , induces Parkin-mediated selective mitophagy which may benefit viral replication [32] . Recently , several reports have described that HBV promotes bulk autophagy to favor its own replication [20]–[24] . Although the molecular mechanisms responsible for induction of autophagy during HBV infection are unclear , it seems that autophagy can either enhance HBV DNA replication or favor HBV envelopment [20]–[24] . HBV may promote autophagy either directly via viral factors that trigger autophagy , or indirectly by mechanisms mediated by virus-induced physiological aberrations and stress . Here , our data suggests a functional crosstalk between virus and autophagy pathways providing evidence that selective , rather than bulk autophagy is probably involved in promoting the viability of HBV-infected hepatocytes from imminent cell death due to the virus-mediated mitochondrial injury . This is in line with the emerging concept that selective autophagy of organelles contributes to tight control of host-pathogen interactions [46] . Although the role of HBV and HBx in regulating apoptotic signaling has long been debated [47] , [48] , based on the in vivo studies , it is inferred that HBV-infected hepatocytes maintain a persistent phenotype and that HBV replicates within infected hepatocytes noncytopathically [49] . The noncytopathic viruses have evolved mechanisms to mitigate the adverse effects of the pathophysiological perturbations manifested in the host cells due to recurrent infection . Here , we demonstrate that HBV modulates mitochondrial dynamics to promote mitochondrial fission and subsequent clearance of damaged mitochondria by mitophagy . Damaged mitochondria are a major source of ROS and can trigger a continuous and vicious cycle of subsequent damage to healthy mitochondria followed by ROS generation , ultimately leading to cell death [17] . Hence , a rapid turnover and clearance of damaged mitochondria is needed to confound imminent cell death due to mitochondrial injury accrued during HBV infection and sustain the viability of the infected cells . In agreement with this assumption , we observed a surge in mitochondrial-apoptotic signaling and resultant death of the HBV-infected hepatocytes upon inhibition of the mitophagy pathway . Hepatocyte damage commonly observed during chronic hepatitis is widely believed to be immune-mediated [49] and the failure to mount an efficient immune response to eliminate the infected hepatocytes is considered the primary cause for persistence of chronic infections , including HBV . However the intracellular mechanism ( s ) , which prevents cell death and support the viability of infected cells in conditions of virus-induced adverse intracellular physiology contributes to the lack of virus-induced cytopathology in non-lytic viruses and probably play a significant role in persistence of non-lytic chronic infections . However , further studies are required to confirm the significance of our findings in in vivo conditions . Currently , the most convenient way to test our hypothesis in vivo settings is the use of animal models of HBV infection such as Woodchucks , Ducks , or Tupaia to determine if abrogation of Parkin-mediated mitophagy pathway promotes specific death of HBV-infected hepatocytes and alleviates persistent HBV infection in these animals . It is believed that HBx sensitizes and indirectly participates in the onset of liver oncogenesis [3] , [47] . Its pleiotropic functions in activating host gene expression , interaction with numerous cellular targets , and its imminent role in altering mitochondrial physiology , promoting oxidative stress , and affecting the epigenetic changes in the chromatin collectively influence the early steps of liver neoplasia . Although not directly linked , HBx-mediated mitochondrial damage and aberrant mitochondrial dynamic may also contribute in pathogenesis of liver disease and HCC [50] , [51] . In summary , this study provides a unique insight into the probable involvement of HBV-induced altered mitochondrial dynamics and mitophagy in possibly facilitating the persistence of infection and pathogenesis of liver disease associated with infection thus unraveling potential newer avenues for design of novel therapeutics against chronic HBV infection .
Human hepatoma cell line Huh7 was grown in high-glucose DMEM ( Gibco ) supplemented with 10% fetal bovine serum ( Hyclone ) , 1% MEM non-essential amino acid ( Gibco ) , 100 units/ml penicillin ( Gibco ) , and 100 µg/ml streptomycin ( Gibco ) . Human hepatoma HepG2 and HepAD38 cells harboring HBV full-length genome were maintained in RPMI 1640 ( Gibco ) supplemented with 20% fetal bovine serum , 1% MEM non-essential amino acid , 100 units/ml penicillin , and 100 µg/ml streptomycin . In addition , HepAD38 cells were grown in the presence of 0 . 5 mg/ml G418 ( Invitrogen ) and 1 µg/ml tetracycline . NT-KD ( expressing non-target shRNA ) and P-KD ( expressing Parkin shRNA ) cells used in this study were maintained in the presence of 2 . 5 µg/ml of puromycin , as described previously and knockdown level of Parkin gene was shown in the previous study [32] . The pHBV1 . 3mer and pHBV-ΔX plasmid DNAs encoding wild-type HBV genome and HBx-deficient HBV genome , respectively , were a kind gift from Dr . Jing-hsiung James Ou ( University of Southern California ) . The pHBx-flag plasmid DNA was described previously [4] . The pEGFP-LC3 plasmid DNA was a kind gift from Dr . Tamotsu Yoshimori ( National Institute of Genetics , Japan ) . The mCherry-Parkin plasmid DNA ( plasmid #23956 ) was obtained from Addgene ( a generous gift of Dr . Richard Youle , National Institute of Health , Bethesda , MD ) . HepAd38 cells were a kind gift of Dr . Christoph Seeger ( Fox Chase Cancer Center , Philadelphia , PA ) . To create plasmid pAT016 ( p-mito-mRFP-EGFP ) , plasmid ptfLC3 ( Addgene plasmid #2174 , a generous gift of Dr . Tamotsu Yoshimori ) was double digested with Bgl2 and BamHI to remove LC3 coding sequences , resulting in plasmid p-mRFP-EGFP production and then , polymerase chain reaction product for mitochondrial targeting signal sequences of human cytochrome c oxidase subunit VIII amplified from pEYFP-mito ( Clontech ) was inserted N-terminally in frame into p-mRFP-EGFP . To conduct laser scanning confocal microscopy , the cells grown on coverslips were transfected with the indicated plasmid DNAs followed by immunofluorescence assay , as described previously [32] . Images were visualized under a 60× or 100× oil objectives using an Olympus FluoView 1000 confocal microscope . Quantification of images ( at least 10 cells per each sample ) was conducted with ImageJ and MBF ImageJ softwares . Chemical reagents used in this study were Bafilomycin A1 ( Enzo Life Sciences ) and 3-Methyladenine ( Sigma ) . Primary antibodies used in this study include the following: rabbit monoclonal anti-Drp1 ( Cell Signaling ) ; rabbit monoclonal anti-phospho-Drp1 ( S616 ) ( Cell Signaling ) ; rabbit monoclonal anti-ATF4 ( Cell Signaling ) ; rabbit polyclonal anti-Parkin ( Abcam ) ; rabbit monoclonal anti-LC3B ( Cell Signaling ) ; rabbit polyclonal anti-PINK1 ( Abcam ) ; rabbit polyclonal anti-VDAC1 ( Cell Signaling ) ; mouse monoclonal anti-Mfn2 ( Abcam ) ; goat polyclonal anti-VDAC3 ( Santa Cruz ) ; rabbit polyclonal anti-GAPDH ( Santa Cruz ) ; goat polyclonal anti-β-actin ( Santa Cruz ) ; mouse monoclonal anti-TOM20 ( BD ) ; rabbit polyclonal anti-TOM20 ( Abcam ) ; mouse monoclonal anti-Flag M2 ( Sigma ) ; rabbit polyclonal anti-DTKDDDDK-tag ( GenScript ) ; goat polyclonal anti-DDDDK ( Abcam ) ; mouse monoclonal anti-HBsAg ( Thermo Scientific ) ; mouse monoclonal anti-Ubiquitin ( Cell Signaling ) ; rabbit monoclonal anti-cleaved PARP ( Cell Signaling ) ; rabbit monoclonal anti-cleaved caspase-3 ( Cell Signaling ) ; rabbit polyclonal anti-cytochrome c ( Cell Signaling ) ; normal rabbit IgG ( Cell Signaling ) ; normal mouse IgG ( Santa Cruz ) . The secondary antibodies used for immunofluorescence were Alexa Fluor 350 , 488 , 594 , or 647 donkey anti-mouse , rabbit , or goat IgG ( Molecular Probe ) . The secondary antibodies used for Western blot analysis were HRP-conjugated anti-mouse IgG ( Cell Signaling ) , HRP-conjugated anti-rabbit IgG ( Cell Signaling ) , and HRP-conjugated anti-goat IgG ( Jackson Laboratories ) . Small interfering RNA ( siRNA ) pools used in this study were siGENOME SMARTpool for Parkin ( NM_004562 ) and non-targeting #1 control ( NT ) ( Dharmacon ) . The cells were transfected with siRNA ( 50 nM ) for the indicated times using DharmaFECT 4 transfection reagent according to the manufacturer's instructions ( Dharmacon ) . To analyze the expression levels of Parkin , PINK1 , and LC3B genes , total cellular RNA and subsequent complementary DNAs were prepared , as described previously [32] . The RNA levels of Parkin , PINK1 , and LC3B were quantified by real-time qRT-PCR using DyNAmo HS SYBR Green qPCR kit ( Finnzymes ) . The following primer sets were used for RT-PCR: Parkin forward , 5′-TACGTGCACAGACGTCAGGAG; Parkin reverse , 5′-GACAGCCAGCCACACAAGGC; PINK1 forward , 5′-GGGGAGTATGGAGCAGTCAC; PINK1 reverse , 5′-CATCAGGGTAGTCGACCAGG; LC3B forward , 5′- GAGAAGACCTTCAAGCAGCG; LC3B reverse , 5′- AAGCTGCTTCTCACCCTTGT; GAPDH forward , 5′-GCCATCAATGACCCCTTCATT; and GAPDH reverse , 5′-TTGACGGTGCCATGGAATTT . Real-time qPCR was conducted by using an ABI PRISM 7000 Sequence Detection System ( Applied Biosystems ) . For Western blot analysis , whole cell lysates ( WCL ) were extracted from cells , subjected to SDS-PAGE , transferred to nitrocellulose membrane ( Thermo Scientific ) , and Western blot analyzed with antibodies against the indicated proteins , as described previously [32] . For analysis of ubiquitinated Parkin and Mfn2 in WCL , immunoprecipitates were prepared followed by Western blotting with anti-ubiquitin antibody , as described previously [32] . For co-immunoprecipitation , Huh7 cells co-transfected with HBx-flag and mCherry-Parkin were suspended in 0 . 1 ml of RIPA buffer . The suspended cells were incubated for 20 min on ice and clarified by centrifugation at 15 , 000×g at 4°C for 20 min . The supernatant was mixed with 1 . 9 ml of RIPA buffer without SDS and immunoprecipitated with anti-flag antibody and protein-G Sepharose . The immunoprecipitates were Western blot analyzed with anti-Parkin , flag , and VDAC3 antibodies , respectively . The intensity of protein expression was quantified by ImageJ software . To isolate pure cytosolic and mitochondrial fraction , HepAD38 cells were homogenized and isolated , as described previously [32] , [52] . The activity of caspase-3/7 in HepG2 and HepAD38 cells transfected with siRNA were measured by using Caspase-Glo 3/7 assay kit according to the manufacturer's instructions ( Promega ) . Apoptotic cells death in HepG2 and HepAD38 cells transfected with siRNA were measured by using Click-iT TUNEL Alexa Fluor 488 imaging assay kit according to the manufacturer's instructions ( Invitrogen ) . For quantitative analysis , at least 1 , 000 cells on immunofluorescence image were counted . Statistical analyses using Student's t-test were performed by using Sigma Plot software ( Systat Software Inc . , San Jose , CA , USA ) . | Hepatitis B virus ( HBV ) chronic infections represent the common cause for the development of hepatocellular carcinoma . Mitochondrial liver injury has been long recognized as one of the consequences of HBV infection during chronic hepatitis . Mitochondria are dynamic organelles that undergo fission , fusion , and selective-autophagic removal ( mitophagy ) , in their pursuit to maintain mitochondrial homeostasis and meet cellular energy requirements . The clearance of damaged mitochondria is essential for the maintenance of mitochondrial and cellular homeostasis . We observed that HBV and its encoded HBx protein promoted mitochondrial fragmentation ( fission ) and mitophagy . HBV/HBx induced the expression and Ser616 phosphorylation of dynamin-related protein 1 ( Drp1 ) and its subsequent translocation to the mitochondria , resulting in enhanced mitochondrial fragmentation . HBV also promoted the mitochondrial translocation of Parkin , a cytosolic E3 ubiquitin ligase , and subsequent mitophagy . Perturbation of mitophagy in HBV-infected cells resulted in enhanced mitochondrial apoptotic signaling . This shift of the mitochondrial dynamics towards enhanced fission and mitophagy is essential for the clearance of damaged mitochondria and serves to prevent apoptotic cell death of HBV-infected cells to facilitate persistent infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | Hepatitis B Virus Disrupts Mitochondrial Dynamics: Induces Fission and Mitophagy to Attenuate Apoptosis |
Leptospirosis is a zoonosis caused by highly motile , helically shaped bacteria that penetrate the skin and mucous membranes through lesions or abrasions , and rapidly disseminate throughout the body . Although the intraperitoneal route of infection is widely used to experimentally inoculate hamsters , this challenge route does not represent a natural route of infection . Here we describe the kinetics of disease and infection in hamster model of leptospirosis after subcutaneous and intradermal inoculation of Leptospira interrogans serovar Copenhageni , strain Fiocruz L1-130 . Histopathologic changes in and around the kidney , including glomerular and tubular damage and interstitial inflammatory changes , began on day 5 , and preceded deterioration in renal function as measured by serum creatinine . Weight loss , hemoconcentration , increased absolute neutrophil counts ( ANC ) in the blood and hepatic dysfunction were first noted on day 6 . Vascular endothelial growth factor , a serum marker of sepsis severity , became elevated during the later stages of infection . The burden of infection , as measured by quantitative PCR , was highest in the kidney and peaked on day 5 after intradermal challenge and on day 6 after subcutaneous challenge . Compared to subcutaneous challenge , intradermal challenge resulted in a lower burden of infection in both the kidney and liver on day 6 , lower ANC and less weight loss on day 7 . The intradermal and subcutaneous challenge routes result in significant differences in the kinetics of dissemination and disease after challenge with L . interrogans serovar Copenhageni strain Fiocruz L1-130 at an experimental dose of 2×106 leptospires . These results provide new information regarding infection kinetics in the hamster model of leptospirosis .
Leptospirosis is a zoonosis with worldwide distribution caused by spirochetes that are spread by reservoir animals to humans and other accidental hosts . Up to 500 , 000 severe human infections are documented annually in tropical areas with an incidence of >10 cases per 100 , 000 population [1]–[3] . Clinical severity ranges from asymptomatic to life-threatening disease characterized by hepatorenal failure with or without pulmonary hemorrhage . The mortality rate for severe leptospirosis ranges from 5–40% [4] , [5] . Leptospirosis typically occurs in persons living in socioeconomically deprived conditions after exposure to flooding in humid , subtropical regions where conditions are favorable for environmental survival of the bacteria [6]–[8] . Contamination of water or soil occurs through urinary shedding by reservoir host animals . Although rodents are frequently the source of organisms causing human infections , many mammalian species have been found to harbor infection in their kidneys [9] . Acquisition of infection by a new host can occur through several possible routes , with cutaneous lesions as a major portal of entry [10] . The morphology and motility of leptospires and other spirochetes enhances their ability to penetrate across the skin and other tissue barriers , enter the bloodstream and disseminate to various organs via the endothelium [11]–[13] . Leptospirosis occurs as a dual-phase disease with a leptospiremic phase in the first week characterized by fever , myalgias , and other flu-like symptoms [14] . The immune phase begins as antibodies are produced and organisms are cleared from the bloodstream , associated with aseptic meningitis and recurrence of fever . In severe leptospirosis , the two phases may be obscured by the rapid progression of illness from its onset , leading to liver and kidney dysfunction characterized by jaundice and azotemia , respectively [4] , [15] . Animal models have been shown to recapitulate the two stages of disease by examining the time to appearance of organisms in various tissues depending on the challenge route [16]–[19] . These initial studies primarily involved microscopic observations and isolation of leptospires from experimentally infected guinea pigs and hamsters , which are highly susceptible to infection with virulent leptospires . More recently , sensitive and quantitative PCR methods have been used to measure the leptospiral burden in various organs after intraperitoneal ( IP ) challenge [20] , [21] . IP inoculation is the most common method employed in animal models of leptospirosis including hamsters and guinea pigs . Although this challenge route has certain technical advantages , it is not a natural route of infection as it bypasses relevant mucosal and cutaneous defense mechanisms . Other routes of inoculation such as epicutaneous , conjunctival , subcutaneous , intradermal , oral , intracardiac and intracranial have been reported [16] , [20] , [22]–[24] , but quantification of leptospiral organ burden has only been described for the intraperitoneal route [20] , [21] . Development of animal models that reproduce natural routes of transmission is critical to understanding host-pathogen interactions occurring early in leptospiral infection . In this study , we compared the kinetics of leptospiral infection after inoculation of hamsters via subcutaneous vs . intradermal routes of infection and compared changes in leptospiral burden in target organs with progression of disease as measured by body weight , changes in blood chemistry and vascular endothelial growth factor levels , histopathology , and antibody response .
Leptospira interrogans serovar Copenhageni strain Fiocruz L1-130 was cultivated in Ellinghausen-McCullough-Johnson-Harris ( EMJH ) medium [25] , supplemented with 1% rabbit serum ( Rockland Immunochemicals , Gilbertsville , PA ) and 100 µg/ml 5-fluorouracil at 30°C in a shaker incubator ( 150 rpm ) . Leptospiral cultures ( passage 2 ) in log phase of growth were centrifuged at 2 , 000×g for 5 minutes and resuspended in fresh EMJH prior to hamster inoculation . Female Syrian hamsters , 5 to 6 weeks of age ( Harlan Bioscience , Indianapolis , IN ) , were inoculated either subcutaneously ( 0 . 5 mL ) or intradermally ( 0 . 05 mL ) using a 25 gauge needle with 2×106 leptospires or EMJH alone on day 0 . For intradermal inoculation , fur was removed with a pair of electric clippers to better visualize the injection site ( Figure S1 ) . From 3 to 4 animals were selected randomly for euthanasia from 1–9 days post-infection unless they presented with clinical signs of leptospirosis , such as loss of appetite , gait difficulty , dyspnea , prostration , ruffled fur , or weight loss of ≥10% of the animal's maximum weight . Kidney and liver were collected in formalin for histopathology or incubated overnight at 4°C in RNAlater ( Ambion , Austin , TX ) , then RNAlater was decanted and tissue samples stored at −80°C . Paraffin embedded tissues were sectioned and stained with hematoxylin and eosin ( H&E ) or periodic acid-Schiff ( PAS ) in a Dako automated slide processor . Kidney sections were scored on a scale of 0 ( normal tissue ) to 5 ( severe renal histopathology ) , based on the severity of glomerular injury , tubular cell damage , intratubular cast formation , interstitial inflammation , and capsular depression ( up to 1 point for each ) . Blood was collected for serology , cell count , and chemical analyses ( Antech Diagnostics , Irvine , CA ) . Vascular endothelial growth factor levels were measured using the Rodent MAP version 2 . 0 of Rules Based Medicine ( Austin , Texas ) . All animal procedures were approved by the Veterans Affairs Greater Los Angeles Healthcare System , Institutional Animal Care and Use Committee ( IACUC #04044-02 ) and adhered to the United States Health Research Extension Act of 1985 ( Public Law 99–158 , November 20 , 1985 , “Animals in Research” ) , the National Institutes of Health's Plan for Use of Animals in Research ( Public Law 103-43 , June 10 , 1993 ) , U . S . Government Principles for the Utilization and Care of Vertebrate Animals Used in Testing , Research , and Training , Public Health Service Policy on Humane Care and Use of Laboratory Animals , the United States Department of Agriculture's Animal Welfare Act & Regulations , and Veterans Health Administration Handbook 1200 . 7 . Tissue DNA was extracted using either the FastDNA SPIN Kit ( MP Biomedicals , Santa Ana , CA ) , accordingly to the manufacturer instructions or with the DNeasy Blood and Tissue kit ( Qiagen , Valencia , CA ) , with modifications as previously described [26] . DNA quality was confirmed by measuring absorbance ratios at 260 nm vs 280 nm and also at 260 nm vs 230 nm . The purified DNA was stored at −80°C until use . DNA was tested by qPCR using the Bio-Rad iQ5 Real-time System ( Bio-Rad , Hercules , CA ) . One hundred nanograms of total DNA were combined with 1 µM of each primer and 12 . 5 µL iQ SYBR Green Supermix ( Bio-Rad ) and brought to a final volume of 25 µL with nuclease-free water ( Ambion , Austin , TX ) . Each sample was run in triplicate . qPCR primer pairs were LipL32-f , 5′-CGCTTGTGGTGCTTTCGGTG-3′ , and LipL32-r , 5′- GCGCTTGTCCTGGCTTTACG-3′ . The resulting amplicon was 152 bp . The PCR protocol consisted of an initial incubation step at 95°C for 12 . 5 min followed by 40 cycles of amplification ( 95°C for 15 s , 62°C for 30 s and 72°C for 30 s ) . Standard curves were generated ranging from 10 up to 1 . 6×106 copies of Leptospira ( 20-fold dilutions ) . DNA ( 100 ng ) from 4 uninfected hamsters were used as negative controls . 96-well ELISA microtiter plates ( Immulon 4HBX , Thermo Fisher , Waltham , MA ) were coated with 1×109 heat-inactivated leptospires/mL diluted in PBS , pH 7 . 2 ( Invitrogen , Carlsbad , CA ) , by overnight incubation at 4°C as described previously [27] . Briefly , the plates were blocked with Protein-Free Blocking Buffer ( PFBB , Thermo Fisher , Rockford , IL ) for 1 to 2 h at room temperature ( RT ) . Sera were tested in triplicate after 1∶6400 dilution with PFBB , to wells in a volume of 100 µL , and plates were incubated for 1 h at 37°C . After three washes in PFBB , wells were incubated with a 1∶5 , 000 dilution of peroxidase-conjugated anti-Syrian hamster IgG secondary antibody ( Cat . #307-036-003 , Jackson ImmunoResearch , West Grove , PA ) for 30 min at RT . 100 µL of 1-Step Turbo Ultra TMB HRP substrate ( Thermo Fisher ) were added to the wells and incubated for 30 min at RT with shaking . The reaction was stopped by the addition of 50 µL of 2 M H2SO4 and plates were immediately read at 450 nm in a Bio-Rad 550 Microplate Reader . One-way analysis of variance ( ANOVA ) with post-hoc Tukey and Šídák tests was used to test for differences between multiple ( ≥3 ) groups .
We compared uninfected hamsters with animals challenged with 2×106 leptospires via either the subcutaneous ( SQ ) or intradermal ( ID ) routes . Hamster body weight was included in the daily clinical examination . Uninfected hamsters gained an average of 3 . 6% of body weight per day during the course of the study . As shown in Figure 1 , infected hamsters continued to gain weight until day 6 after challenge . A decrease in body weight was the earliest clinical sign of leptospirosis and , as in our previous study [26] , the endpoint criterion of ≥10% weight loss prevented occurrence of spontaneous death . By day 7 after SQ challenge , 10/13 ( 77% ) of the remaining hamsters had lost ≥10% of peak weight , and no hamsters survived to day 9 after challenge . By comparison , weight loss among ID challenged hamsters was not as severe; by day 7 after challenge , only 6/13 ( 46% ) remaining animals had lost ≥10% of peak weight . The difference in weight loss on day 7 between the hamsters challenged ID vs . SQ was significant ( P<0 . 01 ) . At the same time that hamsters began losing weight , total serum protein became elevated ( Figure 1 ) . The observed hemoconcentration is consistent with impairment in renal concentrating ability and nonoliguric renal insufficiency typically seen in the early stages of kidney disease due to leptospirosis [28] . Additional indications of the onset of impaired renal function on day 6 was the elevation in the serum creatinine ( Figure 1 ) , blood urea nitrogen ( BUN ) ( Table S1 ) and blood phosphate levels ( Table S1 ) . Following the temporal pattern of weight loss and renal dysfunction , the absolute neutrophil count ( ANC ) and liver chemistries became elevated and peaked on day 6 . As shown in Figure 2 , the absolute neutrophil count ( ANC ) peaked at 5–6 fold over normal levels . After peaking on day 6 , the ANC declined more rapidly after intradermal challenge compared to subcutaneous challenge ( P<0 . 05 ) . We found that uninfected hamsters had vascular endothelial growth factor ( VEGF ) levels of 169±39 pg/mL , which is similar to normal levels previously reported in mice [29] , [30] . VEGF levels became elevated on day 6 and peaked on days 7–8 . Like the ANC , the liver enzymes alkaline phosphatase ( AP ) and serum glutamic pyruvic transaminase ( SGPT , alanine transaminase ) peaked on day 6 ( Figure 2 ) . The total bilirubin level increased more slowly and was more variable among animals than the liver enzyme elevations ( Table S1 ) . The leptospiral burden in kidneys and liver after SQ and ID challenge as measured by qPCR , in terms of DNA copies per microgram of tissue DNA , is presented in Figure 3 . In animals infected by either route , the organ with the highest burden of organisms was the kidney . Significant numbers of organisms were found in the kidney as early as day 1 after SQ challenge and day 3 after ID challenge ( t test , P<0 . 05 ) . In animals challenged ID , the leptospiral burden in the kidney increased considerably after day 4 and reached a peak on day 5 of 8 . 5×103 copies/µg of tissue DNA , after which the burden decreased . The leptospiral burden in the kidney after SQ challenge was similar to that after ID challenge through day 5 . However , the leptospiral burden in the kidney continued to increase on day 6 after SQ challenge , peaking at a level six-fold higher ( 5 . 0×104 copies/µg of tissue DNA ) , whereas it decreased on day 6 after ID challenge and was significantly lower than after SQ challenge ( P<0 . 01 ) . The leptospiral burden in the liver had a similar pattern to that in the kidney , reaching a peak on days 5 and 6 after ID and SQ challenge , respectively . However , in both challenge models , the peak leptospiral burden in the liver was two logs lower than it was in the kidney . The histopathology findings were similar in the animals inoculated subcutaneously and intradermally . Kidney histology appeared normal until day 4 after challenge when mild focal tubular damage was noted in some animals . This early renal inflammation was mirrored by inflammatory changes in the perinephric ( Fig . 4A ) and periureteral ( Fig . 4B ) fat and in perihilar lymph nodes ( Fig . 4D ) . On day 5 , subpelvic inflammation was noted between the renal cortex and the transitional epithelium ( Fig . 4C ) . By day 5 , kidneys of all animals showed tubular damage with hyaline casts ( Fig . 4E ) , interstitial infiltration of lymphocytes ( Fig 4F ) and glomerular damage with collapse of the Bowman's capsule ( Fig . 4G ) . By day 6 and afterwards , these renal changes became both severe and diffuse ( Fig . 4H ) resulting in areas of capsular depression . The progression of disease was reflected by the increase in the histopathology score on day 5 ( Fig . 1 ) , which presaged the onset of renal dysfunction and weight loss by one day . Although occasional areas of hemorrhage were observed in the gross appearance of the liver , histologic changes were not observed . Animals challenged by the SQ route did not show an increase in antibody level during the 8 days of the experiment . As shown in Figure 5 , animals challenged by the ID route were found to form significant antibody levels on days 8 and 9 after challenge .
The hamster is one of most widely used animal models of acute leptospirosis infection because of its reproducibility and the susceptibility of hamsters to a wide variety of pathogenic Leptospira strains [16] , [19] , [31] , [32] . However , little is known about the relative burden of organisms in key organs such as the kidney and liver , and the role of leptospiral burden in disease [20] . For this reason , we were interested in examining the time course of infection relative to an array of disease parameters including body weight , histopathology , inflammatory markers , and markers of renal and hepatic function . Organisms were detected in the kidney by qPCR at levels significantly above background by days 1 and 3 after infection by the SQ and ID routes , respectively . Beginning on day 4 or 5 after infection by either the ID or SQ routes , we observed a major increase in leptospiral burden in the kidney , with the appearance of renal and perinephric histopathology changes on day 5 . These histopathology changes were followed one day later ( day 6 ) by deterioration in kidney function , as demonstrated by an elevation in serum creatinine , and the onset of weight loss . The rapid progression from infection to disease that we observed in hamsters mimics the lack of a biphasic pattern of illness seen in severe human leptospirosis . Daily measurement of hamster weight provides a sensitive and objective measure of health , and as in our previous study [26] , utilization of ≥10% weight loss from maximum weight as an endpoint criterion completely prevented spontaneous death . Weight loss occurred simultaneously with elevation of total serum proteins , indicating that these changes were due to leptospirosis-induced dehydration and hemoconcentration through decreased fluid intake and impairment of sodium reabsorption in the renal proximal tubules [28] . The peak in the burden of infection on days 5 and 6 , followed closely by or concurrent with the onset of weight loss , elevation of inflammatory markers , and changes in blood analytical chemistry results indicative of renal and hepatic dysfunction , implies a causal relationship . Elevation in the absolute neutrophil count ( ANC ) is a typical response to systemic infection that is also observed in human leptospirosis [4] . In our study , the ANC was found to peak at 5- to 6-fold over the normal values for uninfected hamsters ( Figure 2 ) . It is unclear what role neutrophils had in the subsequent decrease in the burden of organisms in the kidney and liver . In vitro studies suggest that neutrophils are unable to kill pathogenic leptospires unless specific antibody is present [33] , [34] . However , we were not able to detect circulating leptospiral antibodies until day 8 after infection , and only in animals challenged intradermally ( Figure 5 ) . It is possible that leptospiral antibodies were not detectable in the serum at earlier timepoints because antibodies were complexed with organisms due to a state of antigen excess . The association between detectable anti-leptospiral antibody levels and lower tissue burdens of organisms in intradermally challenged hamsters is consistent with a known protective role for the humoral immune response in leptospirosis [35] . Levels of vascular endothelial growth factor ( VEGF ) were examined as an additional marker of inflammation . Elevated VEGF levels , as documented in our study , have been found to be an important determinant of morbidity and mortality in the mouse model of sepsis [36] . Kidney and liver dysfunction are hallmarks of the severe form of leptospirosis known as Weil's syndrome . As shown in Figures 1 and 2 , deterioration in renal function and elevation in liver enzymes mirrored the peak in leptospiral burden in the kidneys and liver . We have previously shown high numbers of leptospires in the glomeruli and hepatic sinusoids during acute leptospirosis [37] and others have shown that leptospiral infection leads to hepatocyte apoptosis as a mechanism of liver dysfunction in the guinea pig model of leptospirosis [38] . The qPCR approach allowed us to document striking 8- and 13-fold increases in leptospiral burden in the kidney between days 3 and 4 after ID challenge and between days 4 and 5 after SQ challenge , respectively . As shown in Figure 6 , these dramatic increases in burden of infection in the kidney are in stark contrast to relatively low rates of change on other days suggesting delivery of organisms to the kidney from the bloodstream and/or multiplication within the kidney . A similar increase in the burden of infection in the kidney between days 5 and 6 after ip challenge has also been reported by Lourdault et al . [20] , although in that guinea pig study , qPCR levels were significantly higher in the liver than the kidney . In contrast , we observed a higher burden of organisms in the kidney than the liver . These differences in organ predilection could be due to a variety of factors including the virulence of the leptospiral strain , the host animal species , the route of infection and/or immune-mediated delivery mechanisms . A limitation of this study is that leptospiremia levels were not measured in the blood . However , bacterial burdens were measured in the spleen , and given its role in blood filtration , the spleen can be considered as a surrogate for the blood . Another limitation is that some of the bacterial burden found in organs was due to residual blood , given that animals were not perfused with saline prior to organ harvesting . Nevertheless , bacterial burdens were higher in the kidney than in either the spleen or liver by more than an order of magnitude . These data indicate that leptospires selectively target the kidneys during infection and substantiate the contribution of the higher bacterial burdens to the pathogenesis of kidney dysfunction . The predilection for kidney infection was observed at early time points . Organisms were detected in the kidney by qPCR at levels significantly above background by day 1 and 3 after infection by the SQ and ID routes , respectively . Homing to the kidney by leptospires may be mediated either by leptospiral factors such as kidney-specific adhesins or by production of anti-leptospiral antibodies . Colonization of the lumen of proximal renal tubules has been documented by immunohistochemistry as early as 4 days after intraperitoneal challenge [39] . A novel aspect of our study was the comparison of the SQ and ID challenge routes with the finding of significant differences in the kinetics of infection . The differences in leptospiral tissue burden were most pronounced on day 6 , when we observed 31- and 36-fold higher levels of leptospiral DNA in kidney and liver tissues respectively , after SQ compared to ID inoculation . The lower leptospiral burden in kidney and liver was reflected in less weight loss on day 7 and better survival in ID challenged animals . The ID route is of interest because it is a biologically relevant route of infection in individuals with abrasions and cutaneous exposure . An ID challenge model incorporates interactions between leptospires and defense mechanisms of the skin , such as the dendritic cells , that may function as an early warning system for the immune response . Prompt immune activation during the earliest stages of the transmission process is to the advantage of the host if it results in early clearance of organisms and a lower leptospiral burden . The intradermal route may also be of interest as a more reliable way to evaluate transmission-blocking vaccines using surface antigens that are either expressed in the environmental phase or early in infection . | Leptospirosis is the most widespread bacterial infection transmitted from animals to man . Humans are exposed to infection when host animals that harbor the bacteria in their kidneys shed them in their urine . Human infections , caused by the bacterium Leptospira interrogans , frequently result in a life-threatening illness characterized by liver and kidney failure . In the hamster model of leptospirosis , signs of hepatic and renal dysfunction developed on days 6 and 7 , respectively , after intradermal and subcutaneous inoculation of L . interrogans . Renal dysfunction was preceded by the development of inflammatory changes and the appearance of large numbers of leptospires in the kidney on day 5 . On day 6 , animals began to lose weight , became dehydrated , and had elevated numbers of neutrophils circulating in their bloodstream . Importantly , animals inoculated intradermally had lower numbers of bacteria in their liver and kidneys on day 6 than animals inoculated subcutaneously and lower weight loss and circulating neutrophil levels on day 7 . These studies show that the hamster model of leptospirosis is similar to human infection and indicate that the route of infection has significant effects on the course of the illness . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"leptospira",
"pathology",
"and",
"laboratory",
"medicine",
"medical",
"microbiology",
"microbial",
"pathogens",
"pathogens",
"biology",
"and",
"life",
"sciences",
"leptospira",
"interrogans",
"microbiology",
"bacteria",
"bacterial",
"pathogens",
"organisms"
]
| 2014 | Kinetics of Leptospira interrogans Infection in Hamsters after Intradermal and Subcutaneous Challenge |
Wolbachia pipientis is a ubiquitous , maternally transmitted bacterium that infects the germline of insect hosts . Estimates are that Wolbachia infect nearly 40% of insect species on the planet , making it the most prevalent infection on Earth . The bacterium , infamous for the reproductive phenotypes it induces in arthropod hosts , has risen to recent prominence due to its use in vector control . Wolbachia infection prevents the colonization of vectors by RNA viruses , including Drosophila C virus and important human pathogens such as Dengue and Chikungunya . Here we present data indicating that Wolbachia utilize the host actin cytoskeleton during oogenesis for persistence within and transmission between Drosophila melanogaster generations . We show that phenotypically wild type flies heterozygous for cytoskeletal mutations in Drosophila profilin ( chic221/+ and chic1320/+ ) or villin ( qua6-396/+ ) either clear a Wolbachia infection , or result in significantly reduced infection levels . This reduction of Wolbachia is supported by PCR evidence , Western blot results and cytological examination . This phenotype is unlikely to be the result of maternal loading defects , defects in oocyte polarization , or germline stem cell proliferation , as the flies are phenotypically wild type in egg size , shape , and number . Importantly , however , heterozygous mutant flies exhibit decreased total G-actin in the ovary , compared to control flies and chic221 heterozygous mutants exhibit decreased expression of profilin . Additionally , RNAi knockdown of profilin during development decreases Wolbachia titers . We analyze evidence in support of alternative theories to explain this Wolbachia phenotype and conclude that our results support the hypothesis that Wolbachia utilize the actin skeleton for efficient transmission and maintenance within Drosophila .
Wolbachia pipientis is an intracellular α-proteobacterium that forms symbioses with an extremely broad array of hosts , including isopods , nematodes , and insects [1] . Wolbachia were first noted in the tissues of the mosquito , Culex pipiens , by Hertig and Wolbach in 1924 , but subsequently , many more insects were found to harbor Wolbachia . Current estimates suggest that upwards of 40% of insect species may be infected by the parasite , making Wolbachia one of the most common intracellular bacteria on the planet [2] . Wolbachia are well known for the reproductive effects induced in the host , which range from the exotic ( male killing ) to the most common of reproductive effects , cytoplasmic incompatibility ( CI ) [1] . This recalcitrant , obligate symbiont has received much attention recently due to medical relevance . Wolbachia are heavily studied as potential drug targets for filarial nematode infection [3 , 4] and are currently being implemented to prevent transmission of Dengue fever from mosquitoes to humans [5 , 6] . Wolbachia may be one answer to controlling some vector borne human diseases—indeed mosquitoes harboring a virus-blocking strain of Wolbachia are presently being released in underdeveloped parts of the world with this hope in mind [6–8] . Given the ubiquity of Wolbachia in the insect world , and its relevance to human health , it is essential to understand the biological basis of transmission of the symbiont between host generations . Wolbachia are maternally transmitted bacteria that infect the germline of their hosts such that their transmission fidelity in wild populations is extraordinarily high . Although physiologically stressful conditions are known to induce the loss of superinfections [9] , perfect transmission has been measured in control laboratory Drosophila populations as well as in insects harboring transferred Wolbachia infections [10–12] . Localization in the germline , and in the developing oocyte , is critical to Wolbachia’s maternal transmission and in addition , densities in the embryo , and posterior localization , are correlated with reproductive phenotype ( e . g . CI ) [13 , 14] . Previous studies have provided some support for Wolbachia interactions with host cytoskeletal elements . Specifically , in Drosophila , Wolbachia require host microtubules and the motors Dynein and Dynactin for anterior localization early in development and Kinesin-1 for posterior localization in mid oogenesis , positioning them for inclusion in the germline [15 , 16] . This localization is thought to be crucial to the bacterium’s faithful transmission to subsequent generations at the appropriate densities . Additionally , Wolbachia use astral microtubules during asymmetric divisions in the developing embryo , leading to the widespread , but uneven , pattern of localization of the bacteria in adult tissues [17] . In both worms and flies , Wolbachia undergo somatic cell to germline transmission , suggesting an ability for the bacterium to alter the host actin cytoskeleton to facilitate uptake by germ cells [18 , 19] . More recently , work has suggested interactions between Wolbachia proteins from the Brugia malayi symbiont and host actin [20] , although Wolbachia ultrastructure in Brugia does not reveal any obvious mechanism ( such as actin comet tails produced during infection in other Rickettsiales ) [21] . These previous studies have relied on microscopy and in vitro biochemistry and until now , no genetic evidence of interaction between Wolbachia and actin has been reported . Here we present data showing that Wolbachia persistence and transmission within Drosophila melanogaster is sensitive to mutations affecting the actin cytoskeleton . The importance of actin during Wolbachia infection was investigated by acquiring Drosophila mutants in actin binding proteins , both involved in the regulation of F-actin filaments: the homologs of profilin ( chickadee ) , which regulates the formation of filamentous actin , and villin ( quail ) , which bundles actin filaments . We show that flies heterozygous for mutations in profilin ( chic221/+ and chic1320/+ ) or villin ( qua6-396/+ ) lose Wolbachia infection after only a few generations . Importantly , the effect is due to both an inability of Wolbachia to efficiently colonize germaria in heterozygous mutant hosts and by a reduction in titer when the host is infected . Importantly , both the less severe chic allele ( chic1320 ) , known to decrease an oocyte specific isoform of Drosophila profilin chickadee [22] , as well as the null chic allele ( chic221 ) produced a Wolbachia titer phenotype . We identified two different actin binding proteins ( profilin and villin ) that affect Wolbachia transmission and maintenance , supporting the conclusion that Wolbachia persistence within the host is sensitive to actin .
Standard methods were used for all crosses and culturing . The following stocks were obtained from the Bloomington Drosophila Stock Center ( BDSC ) at Indiana University ( http://flystocks . bio . indiana . edu/ ) : stock number 145 , which carries w1 was used as the Wolbachia infected control line . Two chickadee mutant fly stocks were used in this study . The chic221 cn1/CyO; ry506 flies carry a null recessive allele resulting from the deletion of 5’ non-coding and some chic-coding sequences [22] . The P{PZ}chic01320 cn1/CyO; ry506 flies carry a strong homozygous infertile loss-of-function allele in chickadee , generated by P-element insertion [23] . The quail mutant flies , qua6-396/SM1 , carry a female sterile , recessive mutation induced by ethyl methanesulfonate [24] . We also utilized two chromosomal deficiency stocks: #9507 , w1118; Df ( 2L ) BSC148/CyO , is a chromosomal deletion of segments 36C8-36E3 , covering the region containing the quail locus . The second of these stocks #24377 , w1118; Df ( 2L ) BSC353/CyO , covers segments 26A3-26B3 , the region containing the chic locus . Both of these chromosomal deletions are part of the aberration stock collection and were created by FLP-mediated recombination between FRT-bearing transposon insertions [25] . Wolbachia were introduced into the heterozygous mutant backgrounds through crosses between w1 infected females ( stock 145 ) and uninfected heterozygous males ( mutant/CyO ) . In order to control for genetic background , we also created isogenized lines by backcrossing stock 145 and each mutant line to an uninfected w; Sco/Cyo stock for three generations ( as per [26] , S1 Fig ) . We used sibling controls to identify Wolbachia titer differences related to genotype . In addition to these isogenized lines , and to examine the effect on Wolbachia titer of profilin knockdown during development , we utilized a fly stock carrying a UAS inducible profilin-specific short hairpin silencing trigger ( RNAi; stock #34523 , genotype y1 sc* v1; P{TRiP . HMS00550}attP2 ) [27] . In order to test the effect of induction on fly development ( to recapitulate the developmental lethality of the profilin null ) we crossed homozygous females from this line to w; P{w+ , Act GAL4} /TM3 males . In order to knock down profilin , we then crossed homozygous females from this line to a homozygous Hsp70:Gal4 driver ( a generous gift from Brian Calvi ) . An additional control for expression from the Hsp70:Gal4 driver included a UAS:GFP stock ( also a gift from Brian Calvi ) . Flies were shocked at 37C for 10 minutes to induce the short hairpin . Wobachia infection status for stocks acquired from the BDSC was determined via PCR and Western blot targeting the gene wsp or its product ( see methods below ) . All flies examined for Wolbachia infection in the experiments below were age matched in order to avoid confounding correlations between fly age and Wolbachia titer . Flies were ground in 1 . 5ml centrifuge tubes using an electric hand drill and disposable pestle in lysis buffer: 150mM NaCl , 1% Triton X-100 , 50mM TrisHCl ( pH8 ) containing HALT protease inhibitor cocktail ( Thermo Scientific ) and 5 mM EDTA . The lysates were centrifuged for 1 minute at 8000 X g to pellet debris . Samples were heated for 5 minutes at 95°C in Laemmli sample buffer containing 5% β-mercaptoethanol ( Bio-Rad ) prior to SDS-PAGE electrophoresis . Proteins were separated on 4–20% Tris-Glycine NB precast gels ( NuSep ) in 1X Tris/Glycine/SDS running buffer ( Bio-Rad ) and transferred to PVDF membrane in Tris-Glycine transfer buffer with 15% methanol at 40v on ice for 3–4 hours . The membrane was blocked for 5 minutes in Starting Block T20 ( TBS ) Blocking Buffer ( Thermo Scientific ) , followed by incubation in primary antibody ( for 1 hour at RT or O/N at 4°C ) according to standard protocols . SuperSignal West Pico Chemiluminescent Substrate ( Thermo Scientific ) was used according to the manufacturer’s instructions to detect HRP ( after incubation with secondary antibodies ) on the immunoblots . Blots were re-probed after stripping in 100mM Glycine , 0 . 15 ND-40 , 1% SDS , pH 2 for 1 hour at RT , then overnight at 4°C . PageRuler prestained protein ladder ( Thermo Scientific ) was used as a molecular mass marker . The following antibody was obtained through BEI Resources , NIAID , NIH: Monoclonal Anti-Wolbachia Surface Protein ( WSP ) , NR-31029 , and used at a dilution of 1:1000 . Additionally , we used anti-actin monoclonal at 1:10 , 000 ( Seven Hills Bioreagents ) as a loading control as well as secondary antibodies: HRP enzyme conjugates ( Invitrogen ) at 1:5000 . Densitometry measures were made in ImageJ using scanned film with same exposure times across multiple experiments . Control and experimental flies were included on the same blot in order to ensure consistencies in measured ratios . Immunohistochemistry was performed as follows: ovaries for immunolocalization were dissected in Ringer’s solution 3–5 days after fly eclosion , then fixed as previously described [28] with following modification: 6% formaldehyde devitellinizing buffer was replaced with 5 . 3% paraformaldehyde in same ( Electron Microscopy Sciences ) . After a series of washes in PBS buffer , ovaries were blocked with 0 . 5% BSA in PBST for 10 min . The monoclonal anti-Heat Shock Protein 60 ( HSP60 ) , clone LK2 , H 3524 ( Sigma ) was diluted 1:150 in PBST with 1% BSA or a custom antibody created against full length Wolbachia FtsZ was diluted 1:150 in PBST with 1% BSA . Cy3 conjugated to goat anti-mouse secondary antibody ( Jackson Immunoresearch ) or rabbit secondary antibody ( Jackson Immunoresearch ) diluted 1:250 in PBST + BSA was used to detect the primary antibody . For F-actin detection we used Acti-stain 488 Fluorescent Phalloidin ( Cytoskeleton , Inc ) . Tissues were mounted in Slow Fade “Gold” antifade reagent ( Invitrogen ) and stored at 4°C . To confirm staining by immunohistochemistry , we also used fluorescent in situ hybridization , following published protocols [18] with the following modifications: post-fixation in 4% paraformaldehyde in DEPC treated PBS , ovaries were dehydrated in methanol and stored overnight at -20°C . In the morning , washes in DEPC-PBST preceded a 5 minute proteinase K treatment ( 0 . 05 mg/mL ) at 37C before prehybridization in hyb buffer ( 50% formamide , 5X SSC , 250 mg/L SS DNA , 0 . 5x Denhardts , 20 mM Tris-HCl and 0 . 1% SDS ) . Universal bacterial probe EUB338 conjugated to Alexa488 ( Molecular Probes ) was used to detect Wolbachia in the ovarioles . Hybridized ovaries were mounted in Slow Fade “Gold” antifade reagent ( Invitrogen ) . Images were taken as Z-series stacks at 1 . 5 um intervals using a Nikon E800 fluorescent microscope with 40x oil objective and processed using Metamorph imaging software ( Molecular Devices ) . Care was taken such that exposure times were normalized across all experiments . For quantification of Wolbachia and F-actin within the germarium z-sections maximum projections were used and regions of the germarium demarcated using masks ( S2 Fig ) . We were careful to exclude the peritoneal sheath for F-actin quantification and for Z-stacks where the sheath was difficult to exclude ( due to placement of the sections ) , the images were not included in the F-actin quantification . Germaria showing aggregates of Wolbachia were scored based on a striking pixel intensity in the presumed somatic stem cell niche . DNA was extracted from flies utilizing the Qiagen DNeasy Blood and Tissue Kit ( Qiagen ) according to directions with the following modification . Flies were ground in a 1 . 5ml centrifuge tube using a disposable pestle and an electric hand drill in 180 ul PBS , 200 ul ALT buffer , and 20 ul Proteinase K solution . The samples were incubated at 56°C for 10 minutes with vigorous shaking and then centrifuged briefly to pellet debris before continuing with the ethanol precipitation in the kit protocol . DNAs were quantified by measuring absorbance at 260nm using an Epoch spectrophotometer ( Biotek ) . Semi-quantitative PCR was performed by standardizing the amount of DNA in each reaction . We utilized Phusion High Fidelity PCR Master Mix with HF buffer ( New England Biolabs ) . The protocol for amplification was: 98°C for 3 minutes , followed by 25 cycles of 98°C for 10 seconds , 56°C for 45 seconds , 72°C for 1 minute 30 seconds with a final 10 minute extension at 72°C . Primers were as follows: wsp F1 5’-GTC CAA TAR STG ATG ARG AAA C—3’ and wsp R1 5’- CYG CAC CAA YAG YRC TRT AAA -3’ [29] . RNA and DNA were extracted from individual flies or pupae using a modified Trizol extraction protocol . Briefly , 500 uL of Trizol was added to flies and samples homogenized using a pestle . After a 5 minute incubation at room temperature , a 12 , 000 rcf centrifugation ( at 4C for 10 min ) was followed by a chloroform extraction . Aqueous phase containing RNA was extracted a second time with phenol:chloroform before isopropanol precipitation of RNA . This RNA pellet was washed and resuspended in The RNA Storage Solution ( Ambion ) . DNA extraction from the same flies or pupae was performed using ethanol precipitation of the organic phase during the first chloroform extraction . Quantitative PCR was performed on the DNA to detect the Wolbachia titer ( with reference to the host ) using an Applied Biosystems StepOne Real-time PCR system and SybrGreen chemistry ( Applied Biosystems ) . We used wsp primers for Wolbachia ( Forward: CATTGGTGTTGGTGTTGGTG; Reverse: ACCGAAATAACGAGCTCCAG ) and Rpl32 primers for the host ( Forward: CCGCTTCAAGGGACAGTATC; Reverse: CAATCTCCTTGCGCTTCTTG ) at the following temperatures: 95°C for 10 min , then 40 cycles of 95°C for 15 seconds and 60°C for 1 minute . To detect number of profilin transcripts we utilized the RNA extracted from these flies and the SensiFAST SYBER Hi-ROX One-step RT mix ( Bioline ) and the following primer set: chicF: TGCACTGCATGAAGACAACA , chicR: GTTTCTCTACCACGGAAGCG ( FlyPrimerBank , DRSC ) . Reactions were performed in a 96-well plate and calibration standards were used in every run to calculate primer efficiencies . These efficiencies , along with the CT values generated by the machine , were used to calculate the relative amounts of Wolbachia using the ΔΔ Ct ( Livak ) and Pfaffl methods [30] . In order to identify the ratio of filamentous to globular actin in ovaries from age matched flies , we used ultracentrifugation coupled to SDS-PAGE and Western blots using an in vivo F/G actin assay kit ( Cytoskeleton , Inc ) . Age-matched , virgin female flies from chic221/Cyo or control ( stock #145 ) were dissected in LAS2 buffer at 37°C and incubated for 10 minutes at 37°C . A brief 300 g centrifugation step ( 5 minutes ) was followed by a 1 hour ultracentrifugation at 100 , 000 g at 37°C . Supernatants containing globular actin were removed and pellets resuspended in actin depolymerization buffer on ice , by pipetting up and down every 15 minutes for 1 hour . Pellets containing F-actin fractions and supernatants containing G-actin fractions were run on an SDS-PAGE gel and Western blots performed ( as above ) using a primary mouse monoclonal anti-actin antibody . Bands were quantified using densitometric analysis in ImageJ ( as above ) .
All three actin binding protein mutant fly stocks used in this study were uninfected with Wolbachia upon receipt from the Bloomington Drosophila Stock Center . In order to establish an infection in the flies , infected control females were crossed with mutant uninfected males to generate F1 progeny , half of which carried the mutation , and half of which carried the Cyo balancer ( a second chromosome containing inversion breakpoints and a dominant visible mutation of curly wings ) . F1 heterozygous mutants for the actin binding protein alleles were then back-crossed to the paternal mutant line ( mutant/Cyo ) and F2 progeny from that cross , carrying the mutation and harboring straight wings , were collected . We screened both the F1 and F2 progeny for Wolbachia infection using PCR against the Wolbachia surface protein gene ( wsp ) ( Fig 1A ) . We observed a trend where Wolbachia transmission was not complete in these crosses . For example , the bacterium could be introduced into some heterozygous mutant backgrounds; F1 progeny were infected if they resulted from crosses between control females and chic1320/CyO as well as qua6-396/CyO fathers , but the bacterium failed to colonize chic221/+ F1 progeny efficiently . We were unable to detect Wolbachia in many of the F2 progeny ( Fig 1A ) . In order to quantify this reduction in titer , we performed qPCR on DNA extracts from F1 progeny from each of five individuals from the heterozygous mutants and compared these results to the quantified Wolbachia loads found in control flies ( Fig 1B ) . Progeny from each F1 cross have a statistically significant reduction in Wolbachia titer ( as quantified through qPCR ) compared to the control lines ( p < 0 . 01 for all pairwise comparisons , using a Bonferroni correction for df = 8 ) . As additional support for the importance of the chic and qua loci in the Wolbachia titer defects we observed , we also quantified the amount of Wolbachia within two chromosomal deficiency stocks ( deletions in the same region as either chic or qua in isogenic backgrounds ) [25] . These deficiencies showed the same phenotype as our chic and qua mutants , supporting our observation that these genomic loci are responsible for the Wolbachia titer defect ( Fig 1B ) . In addition to reductions in the F1 progeny , we also quantified a reduction in F2 progeny for the three actin mutant lines . For flies in which we can detect Wolbachia , the F2 progeny are further reduced in titer compared to the F1 lines ( ratio of expression F1 versus F2: min = 0 . 56 , max = 0 . 78 ) . In order to control for effects of host genetic background on Wolbachia titer , we created isogenized lines from the control stock ( 145 ) and each of the mutant stocks by backcrossing to an uninfected w; Sco/Cyo line for three generations . We then crossed these Wolbachia infected F3 females ( w; Sco/Cyo ) to Wolbachia uninfected w; mutant/Cyo males ( S1 Fig ) . In the F5 generation , we observed a significant effect of genotype on Wolbachia titer . Specifically , and regardless of mutant allele , mutant/Cyo progeny were reduced in Wolbachia titer by 1/3 compared to their w; Sco/Cyo siblings ( mean relative ratio wsp/rpl32; t = -4 . 514; df = 9; p = 0 . 001 ) . This result suggested to us that the reduction in titer was at least partially due to a result of a developmental defect in Wolbachia maintenance and persistence within the heterozygous mutant hosts . As an additional control for host genetic background and to explore direct effects on profilin knockdown during development , we took advantage of an infected fly stock carrying a UAS inducible profilin-specific short hairpin silencing trigger ( RNAi; stock #34523 , genotype y1 sc* v1; P{TRiP . HMS00550}attP2 ) [27] . In order to test the effect of induction on fly development ( to recapitulate the developmental lethality of the profilin null ) we crossed homozygous females from this line to w; P{w+ , Act GAL4} /TM3 males . From this cross we only recovered stubble progeny , suggesting that this particular RNAi line , which hadn’t previously been utilized in a publication to knock down profilin expression , is effective . In order to test the effect of induction on fly development we crossed homozygous females ( y1 sc* v1; P{TRiP . HMS00550}attP2 ) to a homozygous Hsp70:Gal4 driver [2–5] . Third instar larvae were shocked at 37C for 10 minutes to induce the short hairpin and late pupae collected for RNA and DNA extraction ( N = 8 for each treatment and genotype; y1 sc* v1; P{TRiP . HMS00550}attP2 with or without Hsp70:Gal4 and with or without heat shock ) . In the maternal y1 sc* v1; P{TRiP . HMS00550}attP2 background , heat shock did not affect either Wolbachia titers ( t = 1 . 207 , df = 2 , p = 0 . 351 ) nor profilin expression ( t = -1 . 144 , df = 2 , p = 0 . 371 ) . In contrast , profilin expression was statistically significantly reduced in flies expressing the RNAi construct compared to non-heat shocked siblings ( the mean expression ratio chic/rpl32 = 0 . 57; t = -6 . 240; df = 2; p = 0 . 025 ) . In addition , knockdown of profilin did have a significant and measurable effect on Wolbachia titers in these same flies; the fly Wolbachia titers were reduced by 1/3 compared to their non-heat shocked siblings ( mean relative ratio wsp/rpl32 = 0 . 66 , t = -8 . 593; df = 2; p = 0 . 013 ) . To provide additional support for the reduction in titer observed via PCR , we probed Western blots of pooled or individual fly lysates produced from the F1 and F2 progeny and their parents for Wsp ( Fig 1C , 1D and 1E ) . Results corroborated our previous finding that Wolbachia transmission was imperfect in the mutant flies ( Fig 1A and 1B ) . Specifically , infected F1 progeny , especially in the chic mutant backgrounds , appeared to carry a reduced titer of Wolbachia when compared to the maternal , infected line ( Fig 1 ) . Indeed , flies from control crosses are consistently higher titer in Wolbachia , as based on densitometric quantitation of Western blot bands ( Average +/- STERR over 5 experiments for Control = 13 , 106 +/- 3 , 294; chic1320/+ = 6 , 418 +/- 4 , 890; chic221/+ = 6 , 545 +/- 1 , 576; qua6-396/+ = 6 , 179 +/- 645; t-test; p = 0 . 036 , 0 . 001 , 0 . 002 for each heterozygous mutant compared to control ) . Additionally , we could detect a statistically significant reduction between the F1 and F2 heterozygous mutant flies ( p = 0 . 012 ) . As observed in our results based on PCR , Wolbachia titer ( based on quantity of protein on a Western blot ) is also reduced , with some variability , in the F2 progeny ( Fig 1D ) . We hypothesized that the loss of Wolbachia in some F2 progeny was a result of a reduction in Wolbachia titer in F1 females during oogenesis . We therefore visualized the Wolbachia infection in the germarium in F1 females ( mutant/+; below ) . To colonize the oocyte , and therefore complete maternal transmission , Wolbachia occupy the germline and somatic stem cell niches ( SSCN ) in their hosts [18 , 26 , 31] . Wolbachia can achieve this localization after injection into the fly abdomen , suggesting that the stem cell niche targets are essential for Wolbachia infection [31] . The Drosophila ovariole provides an opportunity to view oocyte development and Wolbachia localization within each progressive stage . Wolbachia concentrate preferentially in the somatic stem cell niche , which is thought to serve as a source of infection for the germline . As germline development progresses from regions 2a to 2b , Wolbachia are thought to infect via the somatic stem cell niche , increasing the numbers of bacteria found within the germline after association with the SSCN [31] . We utilized immunohistochemistry to detect Wolbachia in the germarium of our flies , producing localizations expected based on previous publications [15 , 18 , 26 , 31] ( S1 Table and S2 and S3 Figs ) . Wolbachia infection within the entire germarium is significantly reduced in heterozygous mutant flies ( when comparing the amount of fluorescence observed in control flies to that found in either chic221/+ , chic1320/+ or qua6-396/+ , respectively; Mann-Whitney U = 171 . 5 , Z = -3 . 995 , p < 0 . 001; Mann-Whitney U = 98 , Z = -5 . 295 , p < 0 . 001; Mann-Whitney U = 55 . 5 , Z = -5 . 496 , p < 0 . 001; Figs 2 and 3 ) . When either chic1320/+ or qua6-396/+ heterozygous mutant flies are infected , the Wolbachia titer in region 2 ( as quantified by anti-Hsp60 staining ) is also significantly reduced , compared to the control maternal line ( Mann-Whitney U = 194 , Z = -3 . 78 , p < 0 . 001; Mann-Whitney U = 134; Z = -4 . 097 , p < 0 . 001 , pairwise comparison between control and chic1320/+ or qua6-396/+ heterozygous mutants , respectively; Fig 3 ) . Additionally , Wolbachia infection within early egg chambers ( stage 1 ) is significantly reduced in all heterozygous mutant flies ( when comparing the amount of fluorescence observed in control flies to that found in either chic221/+ , chic1320/+ or qua6-396/+ , respectively; Mann-Whitney U = 74 , Z = -5 . 74 , p < 0 . 001; Mann-Whitney U = 58 , Z = -5 . 872 , p < 0 . 001; Mann-Whitney U = 39 , Z = -5 . 767 , p < 0 . 001; Figs 2 and 3 ) . In order to quantify this reduction , for each germarium , we calculated the ratio of fluorescence intensity in the earliest egg chamber over that found in region 2 ( as quantified by anti-Hsp60 staining ) . Each of the three mutant lines showed a statistically significant reduction in this ratio when compared to control germaria ( average ratios for control flies: 1 . 72; chic221/+: 0 . 52; chic1320/+: 0 . 64; qua6-396/+: 0 . 80 , t-test; p < 0 . 0001 ) . The reductions in infection in the germaria suggest two things: ( 1 ) that Wolbachia has difficulties in transiting or maintenance in a population within the germarium during development in the heterozygous mutant flies and ( 2 ) even when region two , the location of the SSCN , is occupied by Wolbachia , the bacteria are deficient in colonization of the early egg chamber in the heterozygous mutant flies ( Fig 3 ) . We did not quantify differences in staining of the presumed germline stem cell niche due to variability in staining in this region within the control flies . Within heterozygous mutant flies , we observed that the Wolbachia that successfully manage to colonize the germarium do so with a distinctive localization; these Wolbachia appear as aggregates , in sharp contrast to the more even distribution of Wolbachia within control germaria ( Table 1 and Fig 2 ) . Under high magnification ( 100x ) , the Wolbachia aggregates within the heterozygous mutant flies appear to be multiple Wolbachia forming micro-colonies within the tissue , based on shape and size and consistent localization within the genotypes . Both Drosophila profilin ( chickadee ) and villin ( quail ) are important in the regulation of F-actin during oogenesis . Because profilin promotes the polymerization of F-actin filaments and villin stabilizes these filaments through bundling , we were curious to know whether or not the heterozygous mutant flies differed in the quantity of F-actin found in the germarium , when compared to control flies . In addition , visualization of the F-actin cytoskeleton allowed us to examine the actin ring canals in the heterozygous mutant flies at all stages of oocyte development . At no point were ring canals occluded by nuclei , supporting our finding that cytoplasmic streaming and maternal dumping are unaffected in heterozygous mutant flies ( N = 300 , scored by eye ) . Using quantified fluorescence of F-actin in the images we were unable to detect a statistically significant difference between median levels of phalloidin staining in control flies compared to the heterozygous mutants ( Kruskal Wallis test: χ2 = 4 . 005 , df = 3 , p = 0 . 261; S4A Fig ) . Because F-actin levels in the germarium ( observed through phalloidin staining ) did not correlate with Wolbachia intensity ( as quantified by anti-Hsp60 staining ) , the Wolbachia titer phenotype observed in these flies may not be directly related to the F-actin network in the germarium . We therefore examined the in vivo amounts of filamentous and globular actin in ovaries from control flies and compared this to that seen in heterozygous mutant chic221 flies . Using ultracentrifugation coupled to Western blot , we found that we could consistently detect globular actin in the ovaries of control flies . In contrast , we found a statistically significant decrease in total amount of globular actin detected in the heterozygous mutant lines ( Kruskal Wallis: χ2 = 4 . 192 , df = 1 , p = 0 . 041; S4B Fig ) . The difference in G actin between control and heterozygous chic221 flies prompted us to investigate expression of profilin in the chic221/+ F1 mutants and control flies . The rationale was that although these flies are phenotypically wild type , the dosage effect of a single , wild type chromosome in the chic221/+ F1 mutants might be significant and correlate with Wolbachia absence . We extracted both RNA and DNA from individual F1 chic221/+ female flies as well as age-matched control flies and used quantitative RT-PCR to detect profilin transcript levels ( in total RNA ) and Wolbachia surface protein ( in total DNA ) relative to Rpl32 . Control flies express , on average , 2x as much profilin as heterozygous mutant F1 progeny ( means control μ = 4 . 03; mutant μ = 2 . 28; t = 2 . 590 , df = 11 . 31 , p = 0 . 025 ) . Additionally , although we could detect Wolbachia in each of the wild type flies included ( N = 10 ) , we were only able to detect a Wolbachia infection in three of the heterozygous mutant F1 flies ( S4C Fig ) . Wolbachia may have been present in these flies , but at titers below the limit of detection for this method . Because Wolbachia target the germline , and within the germarium , the stem cell niche [18 , 31] , the number of egg chambers produced by the host may affect Wolbachia’s ability to be transmitted between generations . Flies that are homozygous mutants in chickadee show defects in germline stem cell proliferation as well as enclosure by somatic cyst cells [32 , 33] so it was therefore important to confirm that the heterozygous mutant flies do not display similar defects . We counted the number of viable progeny resulting from individual crosses within mutant fly lines and compared the number of resulting offspring to those from control crosses . Heterozygous villin or profilin mutant flies do not show a defect in fertility when compared to control flies ( S5 Fig ) . Additionally , we observed over 300 eggs for each of the fly mutant stocks and did not see any morphological abnormalities when compared to the control stock ( N = 300 , scored by eye ) .
The developing oocyte is loaded with maternal determinants ( e . g . mRNA and protein ) , a process which begins early ( stage 1 ) , and continues until about stage 10 when maternal nurse cells dump their remaining cytoplasmic contents into the oocyte [35] . The actin cytoskeleton is critical to this process , as mutations in actin binding proteins have been known to cause severe defects . Specifically , cytoplasmic actin bundles are required to restrain the nurse cell nuclei during transport; mutations in quail , which regulates bundling of cytoplasmic actin , cause a dumpless phenotype [39 , 40] . In quail mutant flies , nurse cell nuclei can be observed extending through the actin ring canals [39] . We reasoned that although heterozygous mutant flies ( chic221/+ , chic1320/+ and qua6-396/+ ) produce viable progeny , and we found no occluded ring canals in any of these backgrounds , a subtle defect in maternal cytoplasmic dumping could alter the ability of Wolbachia to be transmitted faithfully to the oocyte . Wolbachia has been suggested to utilize cytoplasmic dumping to increase titer in the oocyte ( as compared to the nurse cells ) [15] . In addition to regulating the bundling of microtubules and therefore cytoplasmic streaming , profilin is also required for posterior patterning in the oocyte as chic mutants fail to localize STAUFEN and oskar mRNA [41] . Wolbachia utilizes these posterior determinants to localize in the oocyte , as disruption of osk and stau results in mislocalization of Wolbachia in D . melanogaster [16] . If heterozygous mutant flies are defective in cytoplasmic dumping or polarization , we should observe both egg size and morphology defects . Over 300 eggs were scored for each of the mutant lines , as well as control flies , without any phenotypic differences detected . Importantly , however , the primary loss of Wolbachia in these heterozygous mutants occurs in the germarium , before defects would begin to affect Wolbachia titers . Therefore , although our fly mutants could conceivably exhibit subtle polarization defects , these defects alone would not entirely explain the observed phenotype . In addition to serving important roles during maternal loading in the late stage oocyte , profilin functions in germline stem cell ( GSC ) maintenance and germ cell enclosure by somatic cyst cells [32 , 33]—homozygous chickadee mutants fail to maintain germline stem cell number . However , chic221/+ flies are equivalent to wild type [32]; that is to say , heterozygous mutant flies do not have a GSC deficiency . Importantly , although Wolbachia are known to alter germline stem cell proliferation [26] and some Wolbachia colonize the germline stem cell niche [18] , wMel colonizes the somatic stem cell niche in Drosophila melanogaster ( Fig 2 ) . Regardless , a defect in fertility , resulting from defects in GSC maintenance might affect Wolbachia proliferation in these mutant flies . We therefore counted the number of viable progeny ( a measure of fertility ) for each of the mutant lines . No statistically significant difference was observed for any of the heterozygous , mutant flies , when compared to the control ( S5 Fig ) . We therefore did not find support for this hypothesis to explain the Wolbachia clearing phenotype of profilin and villin heterozygous mutants . There is significant evidence that Wolbachia colonize the primordial germ cells and the posterior pole of developing embryos in numerous insect hosts . In D . melanogaster , for example , strain wMel concentrates at the posterior pole in a poleplasm dependent fashion [16 , 42 , 43] . However , this posterior concentration of Wolbachia is not universal in insects nor in Drosophila . Wolbachia strain wRi infects the entire embryo uniformly while B group Wolbachia actually show exhibit anterior localization [14] . Similarly , in other Drosophila species there are different patterns of Wolbachia colonization: although wWil infects primordial germ cells , wAu infects the entire embryo [44] . This posterior localization is clearly important—the extent of CI is correlated with the number of Wolbachia in the posterior of the embryo [14] . However , this posterior localization is not necessarily correlated with maternal transmission , which is near 100% for some Drosophila species and quite low for others [45–47] . This result suggests that high titer localization to primordial germ cells and the posterior pole does not guarantee maternal transmission . However , if our heterozygous mutant flies induce defects in these early localization patterns ( to the posterior pole or to the developing germ line ) , we might expect the inefficient transmission phenotype observed . What other ways might Wolbachia use to eventually colonize the germline ? Wolbachia colonization of somatic tissues has been known for some time [48] but recently , it has been suggested that Wolbachia infection of the soma may serve as a reservoir for germline infection . In the terrestrial isopod , Armadillidium vulgare , Wolbachia is absent from many early oocytes and infects the older oocytes late in development , an enrichment that is thought to come from a somatic reservoir ( the follicle cells ) [34] . In nematodes , Wolbachia initially are concentrated in the posterior of the P2 blastomere , the precursor of the adult germ line . However , Wolbachia are subsequently excluded from the germ line in the next cell division and instead , invade the germ cells later , from the surrounding somatic gonadal cells [19] . This soma to germ cell invasion in Brugia is correlated with a disruption in polymerized actin at those foci [19] . Because we observed a reduction in anti-Hsp60 staining in stage 1 egg chambers of heterozygous mutant flies as well as transmission defects , one interpretation of our data is that Wolbachia require actin for soma to germline transmission . Importantly , however , we did not observe actin disruptions ( similar to those seen in Brugia ) within Drosophila germaria . Our data suggest that Wolbachia rely on the actin cytoskeleton to achieve adequate titer in the Drosophila host during development . First , we observe reductions in titer of Wolbachia in heterozygous mutants compared to both their non-mutant sibling controls as well as parental controls ( Fig 1 ) . Second , knockdown of profilin in third instar larvae reduces Wolbachia titer in pupae , suggesting that the regulation of actin is important to the maintenance of a Wolbachia infection during development . Additionally , passage of Wolbachia through heterozygous mutant lines for multiple generations results in the enrichment for mutant Wolbachia; the heterozygous mutant flies bottleneck the Wolbachia infection , increasing the stochastic segregation of variants [49] . This decrease in titer may explain the inefficient transmission of Wolbachia observed in the mutant flies . Actin may be used by Wolbachia to properly localize during development , or may support the infection via other unknown mechanisms . Both of the proteins investigated here ( profilin and villin ) , are known to increase the amount and stability of F-actin in the Drosophila egg chamber . Profilin promotes F- actin in the follicular epithelium while villin bundles and binds to filamentous actin [37 , 50] . One potential cause of the Wolbachia phenotype in these backgrounds is a mis-regulation in F-actin content . Interestingly , chic mutants have been previously observed to exhibit decreased F-actin levels in the follicle cells [50] . Both the somatic stem cell niche and the follicular epithelium have been suggested to be a source of Wolbachia during oogenesis [18 , 34] . Because Wolbachia densely colonize the follicular epithelium tissue , and because it surrounds the oocyte throughout development , this tissue may be a candidate for the source of the infection . We detected a significant reduction in the amount of actin in heterozygous mutant chic221 flies compared to controls , which corresponded to a decrease in profilin transcripts and a decrease in detected Wolbachia ( S4 Fig ) . These data are suggestive of a role for actin in Wolbachia maintenance and transmission but do not elucidate an exact mechanism . We have shown that the host actin cytoskeleton is clearly important for the maintenance of a Wolbachia infection . Perhaps this reproductive parasite secretes proteins that interact directly with eukaryotic actin or host actin binding proteins . Indeed , other members of the Rickettsiales are known for their striking coopting of host actin in the production of comet tails [51] . However , when intracellular , Wolbachia persist within membrane-bound compartments and no such comet-like structures have been observed to be associated with the vacuole [21] . That said , our results here and the work of others strongly suggest that Wolbachia is able to enter and exit eukaryotic cells; Wolbachia transit to the germline from the fly abdomen and are loaded into the germ cells from surrounding somatic cells [18 , 26 , 31] . Wolbachia’s success likely depends upon an ability to secrete proteins that modify host actin to promote internalization by non-phagocytic cells . Recently , in vitro biochemical associations between the filarial nematode Wolbachia ( wBm ) PAL-like protein wBm0152 and actin have been observed , although results do not conclusively implicate this particular protein in interactions with host actin during infection [20] . Regardless , as is clear from our work , a Wolbachia infection depends on the actin cytoskeleton . Therefore , future work to identify and characterize Wolbachia proteins that bind to or alter host actin dynamics will be important for understanding the molecular basis of the interaction between the host and the symbiont . In order for intracellular , maternally transmitted symbionts to successfully infect the next generation , the bacteria must target the oocyte . Wolbachia achieves this through a specific infection of the somatic stem cell niche in the germarium of Drosophila melanogaster [18] . Here we show that Wolbachia is extraordinarily sensitive to the regulation of actin , such that phenotypically wild type heterozygous mutant flies cannot faithfully transmit the bacterium to their progeny . Our results , particularly that titer is significantly reduced in the germaria of chic221/+ , chic1320/+ , and qua6-396/+ flies , suggest that Wolbachia utilize host actin to enter and persist within host tissues during Drosophila development . Additionally , our finding that these heterozygous mutant flies cannot transmit the infection suggests that Wolbachia titers within a host are reduced when actin regulation is disrupted , impacting transmission efficiency . | The world’s most common intracellular infection , Wolbachia pipientis , infects 40% of insect species and is currently used to prevent transmission of Dengue by mosquitoes . The bacterium targets the germline of insects , where it is faithfully transmitted to the developing oocyte and the next generation . Here we identify host cytoskeletal proteins required by Wolbachia in order to be efficiently transmitted between Drosophila melanogaster generations . We show that after only two generations in a phenotypically wild type , heterozygous mutant fly , Wolbachia infections are cleared or reduced in titer . Characterization of the mutants suggests that Wolbachia is sensitive to the regulation of actin in the ovary and that actin may be used by Wolbachia to both target and proliferate within host tissues and to be faithfully , maternally transmitted . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | Wolbachia Utilize Host Actin for Efficient Maternal Transmission in Drosophila melanogaster |
Facioscapulohumeral dystrophy ( FSHD ) is caused by the mis-expression of DUX4 in skeletal muscle cells . DUX4 is a transcription factor that activates genes normally associated with stem cell biology and its mis-expression in FSHD cells results in apoptosis . To identify genes and pathways necessary for DUX4-mediated apoptosis , we performed an siRNA screen in an RD rhabdomyosarcoma cell line with an inducible DUX4 transgene . Our screen identified components of the MYC-mediated apoptotic pathway and the double-stranded RNA ( dsRNA ) innate immune response pathway as mediators of DUX4-induced apoptosis . Further investigation revealed that DUX4 expression led to increased MYC mRNA , accumulation of nuclear dsRNA foci , and activation of the dsRNA response pathway in both RD cells and human myoblasts . Nuclear dsRNA foci were associated with aggregation of the exon junction complex component EIF4A3 . The elevation of MYC mRNA , dsRNA accumulation , and EIF4A3 nuclear aggregates in FSHD muscle cells suggest that these processes might contribute to FSHD pathophysiology .
Facioscapulohumeral dystrophy ( FSHD ) is a progressive muscular dystrophy caused by mis-expression of the double-homeobox transcription factor DUX4 in skeletal muscle [1] . Normally , DUX4 is not expressed in skeletal muscle nor in most somatic tissues examined [2 , 3] . Ectopic expression of DUX4 in human and mouse cell lines as well as in vivo injection of DUX4 adenovirus into mouse muscle leads to rapid cellular apoptosis [4 , 5] . This cell death is dependent on the transcriptional activity of DUX4 because expression of DUX4 with mutations in the DNA binding domain or trans-activation domain do not exhibit toxicity [5 , 6] . More recently , it was demonstrated that endogenous levels of DUX4 produced in FSHD muscle cells similarly causes cellular death [7] . Apoptosis is known to be a critical cellular process for both tissue homeostasis as well as vertebrate ontogeny where cells in developing organs follow the general guidelines of “proliferation , differentiation and demolition” [8] . It is also appreciated that cellular apoptosis , outside of normal homeostatic or developmental contexts , is involved in autoimmune and neurological diseases [9 , 10] . As evidenced by numerous mouse knockout lines , programmed cell death is required for productive sperm development where excess or abnormal germ cells are constantly culled to ensure adequate space and nutrients [11] . Previously , immunodetection has identified DUX4 expression in cells in the seminiferous tubule , morphologically resembling spermatogonia or primary spermatocytes [2] and in the thymus [3] , both tissues with high rates of apoptosis . Thus , it is possible that expression of DUX4 in skeletal muscle inappropriately activates a program of apoptosis that might otherwise be a 'normal' consequence of DUX4 expression during developmental processes . As a transcription factor , DUX4 activates many genes that are expressed in stem cells and in the germline [12] . The long terminal repeat ( LTR ) of a subset of human endogenous retroviruses ( ERVs ) contain the DUX4 binding motif , and DUX4 binds and activates their transcription , occasionally creating a novel transcription start site for adjacent genes [13] . DUX4 expression also represses the innate immune response [12] and the nonsense mediated decay ( NMD ) pathway [14] , leading to an accumulation of normally degraded RNAs . However , it is not currently understood whether the changes in RNA stabilization following DUX4 expression lead to apoptosis . It has been demonstrated that FSHD muscle biopsies exhibit oxidative stress and mitochondrial dysfunction compared to control biopsies [15] and small molecule screens have demonstrated that compounds which protect cells from oxidative damage also protect against DUX4 toxicity [6 , 16] . It was also previously shown that Tp53 knockout mice were protected from the effects of DUX4 delivered to skeletal muscle by AAV transduction and that a P53 inhibitor decreased DUX4 toxicity in human HEK293 cells [5] . However , the DUX4-induced apoptotic pathways relevant to human skeletal muscle and FSHD remain poorly understood . We conducted a small interfering RNA ( siRNA ) screen to identify genes and pathways necessary for DUX4 toxicity . This screen identified the MYC-mediated apoptotic pathway and components of the double-stranded RNA ( dsRNA ) innate immune response as necessary for DUX4-induced apoptosis . We found that DUX4 expression resulted in the stabilization of several mRNAs , including MYC , and the accumulation of nuclear dsRNAs . Stabilization of MYC mRNA was associated with a dramatic increase in MYC protein levels and the activation of genes in the MYC-mediated apoptosis pathway , including BCL2L11 and EGR1; whereas the accumulation of dsRNA was associated with nuclear aggregation of EIF4A3 and phosphorylation of the kinase EIF2AK2/PKR and its downstream target , the eukaryotic translational initiation factor subunit EIF2S1/eIF-2α . The similar elevation of MYC mRNA , dsRNA accumulation , and EIF4A3 nuclear foci in FSHD muscle cells suggest that these processes might contribute to FSHD pathophysiology .
Although a prior study had identified TP53 as necessary for DUX4-induced cell death in some cells [5] , we found that human myoblasts with CRISPR mutated TP53 and the RD rhabdomyosarcoma cell line that does not contain a functional TP53 allele [17 , 18] both succumbed to DUX4-induced cell death as efficiently as primary human myoblasts ( S1A–S1D Fig ) . To establish a screen for genes necessary for DUX4-induced apoptosis , we transduced the RD cell line with a lentiviral vector encoding a doxycycline-inducible DUX4 coding sequence and puromycin selectable marker ( see S2A Fig for schematic ) and generated a clonal rhabdomyosarcoma cell line ( RD-DUX4i ) with robust doxycycline-inducible expression of DUX4 that showed cellular toxicity at 24 hours after DUX4 induction ( S2B Fig ) and more than 95% cell death by 48 hours , as determined by the ATP-based CellTiter-Glo assay ( S2C Fig ) . The dying cells exhibited an increase in activated caspase 3/7 ( S1A Fig ) indicating that DUX4 induction leads to apoptosis in the P53 deficient RD cells . To identify genes and pathways necessary for DUX4-induced toxicity in RD cells we measured cell survival following transfection of an siRNA library targeting 6 , 961 genes in the human “druggable” genome with a pool of four siRNAs per target gene . The parameters for the screen are summarized in S2D Fig and were optimized as in S3A Fig . As positive controls , we used two unique siRNAs targeting DUX4 which had different efficacies of knockdown ( S3B Fig ) and , consequently , different effects on cell viability after DUX4 induction ( S2E Fig ) . Transfections were performed in triplicate and cell survival following DUX4 induction was measured by CellTiter-Glo . Using a stringent mean Z-score threshold of 3 . 0 , 69 siRNA pools significantly increased cell viability in response to DUX4 . Targets that exceeded this threshold included some genes previously implicated in cellular apoptosis , for example: Death effector domain containing 2 ( DEDD2 ) , Cell death-inducing DFFA-like effector a ( CIDEA ) , MutS homolog 2 ( MSH2 ) , C-MYC ( MYC ) and its dimerization partner MAX , and RNASEL , a mediator of cellular antiviral defense . Using a more traditional Z-score cutoff of 2 . 0 , 353 siRNA pools enhanced cell viability after DUX4 induction . In addition , 30 siRNAs had a Z-score of less than -2 . 0 , suggesting that these genes might protect cells from DUX4-induced death . Among these targets were the DUX4 target genes , TRIM51 ( also known as SPRYD5 ) and TRIM43 as well as the anti-apoptotic BCL2L1 ( also known as BCL-X ) that can suppress MYC-mediated cell death [19] . As expected , siRNAs targeting TP53 had no effect on cell viabilty ( Z-score = -0 . 03 ) . Network analysis placed MYC/MAX as a 'hub' of protein-protein interactions among the candidate genes with an absolute value Z-score cutoff of 2 . 0 ( S3C Fig ) . The general results of the screen are summarized in S2E Fig , and the full ranked list is available in S1 Table . We proceeded to validate candidate siRNA pools identified in the initial screen using a strategy outlined in Fig 1A . As a first validation step we transfected RD-DUX4i cells with the ten siRNA pools with the highest Z-scores plus 15 additional pools with Z-scores ranging from 1 . 93 to 3 . 61 . All 25 pools increased viability over the non-silencing control siRNA , indicating that pools identified in the primary screen reproducibly enhanced cell survival after induction of DUX4 ( S4A Fig ) . As a second validation step we de-convoluted these 25 pools of four individual siRNAs to determine whether multiple siRNAs within each pool contributed to the increased survival of RD cells rather than a single siRNA which might indicate a nonspecific , off-target effect . Of the 25 pools retested , 21 had more than one siRNA that enhanced cell survival by 2-fold or greater ( Fig 1B and S4B Fig ) . For the four pools with a single siRNA that rescued viability , we tested whether the individual non-rescuing siRNAs from that pool might rescue if pooled together , reasoning that synergy among the individually non-rescuing siRNAs might indicate on-target activity . Re-transfection of these non-rescuing siRNAs as pools of three revealed that more than one siRNA against two genes ( EAF1 and TLR5 ) rescued from DUX4 toxicity ( Fig 1C and S4C Fig ) , whereas for the other two genes ( TLL1 and RGS3 ) , only a single siRNA had rescuing activity . Further analysis showed that the rescue conferred by siRNAs in pools where there was a single rescuing siRNA , such as siRGS3 , was correlated with the inhibition of doxycycline induction of DUX4 and of a doxycyline inducible luciferase transgene in RD cells ( RD-LUCi cells; S4D Fig ) . This indicated that some of the siRNA pools in the original screen likely achieved RD-DUX4i cell survival and a high Z-score because a single siRNA in the pool inhibited the doxycycline induction of DUX4 through unknown mechanisms . Therefore , as a third validation step we tested for the inhibition of the doxycycline induction pathway on the entire set of siRNA pools with an original Z-score ≥ 2 . 0 using the RD-LUCi cells . 110 of our initial 353 pools suppressed luciferase induction based on a cutoff Z-score of ≤ -0 . 5 ( Fig 1D and S2 Table ) , the minimum score of the 16 scrambled control siRNAs used in this secondary screen . For example , the RGS3 siRNA pool , which we had already determined had a single siRNA that suppressed doxycycline induction of luciferase , had a mean Z-score of -3 . 09 . Overall , there was a modest but significant inverse correlation between the RD-DUX4i screen Z-scores and the RD-LUCi screen Z-scores ( Spearman's rho = -0 . 396 ) , indicating that some of the pools identified in the original screen were likely secondary to inhibiting the doxycycline induction pathway . Therefore , we eliminated the 110 genes targeted by pools that depress luciferase induction by more than the non-silencing control . As a final filter , we performed RNA-sequencing on RD-DUX4i cells and required that the targeted mRNA must be expressed at greater than an average FPKM ( fragments per kilobase of transcript per million mapped reads ) of 0 . 5 to filter out low or non-expressed genes . For example , although TLR5 apparently rescued based on initial validation criteria ( see above ) , it was expressed below the threshold level and was therefore eliminated . Of the original set of 6 , 961 genes targeted by the library , 16 genes with a Z-score ≥ 3 . 0 passed our validation and filtering steps ( Table 1 ) . The final list of filtered targets with Z-score ≥ 2 . 0 is summarized in S3 Table . The filtered targets meeting a Z-score threshold of 3 . 0 included several genes that broadly regulate RNA transcription ( MYC , FOSB , EAF1 , PHF20 , CDC20 ) , protein translation ( TGS1 ) [20] , or mitochondrial function ( FXN ) [21] . MAX , the obligate heterodimer of MYC , also appeared as a candidate target in the initial screen , with a Z-score exceeding 3 . 0 , although MAX was subsequently eliminated from the final candidate list because the pool of siRNAs to MAX demonstrated a modest reduction in doxycycline induction of the luciferase transgene ( Z-score = -0 . 93 ) . It is interesting to note that many of the aforementioned genes were upregulated in our RNA-seq data in RD cells overexpressing DUX4: FOSB ( ~92 fold ) , MYC ( ~56 fold ) , EAF1 ( ~8 fold ) and CDC20 ( ~11 fold ) . Although siRNAs against MYC had a modest effect on transgene expression ( luciferase Z-score = -0 . 29 ) , the siRNAs also rescued viability following lentiviral transduction of a constitutively expressed DUX4 in cells with DUX4 protein levels and nuclear localization equivalent to the controls ( S4E–S4G Fig ) . Therefore , we decided to investigate the MYC-mediated apoptosis pathway because DUX4 expression resulted in a dramatically increased expression of MYC as well as a set of genes that facilitate a program of enhanced cellular growth or metabolism . We first verified that MYC protein was upregulated following DUX4 expression . Western blotting and RT-qPCR confirmed that MYC protein and RNA levels increased dramatically following DUX4 expression ( Fig 2A , S5A and S5B Fig ) , tracking closely with a direct target of the DUX4 transcription factor , MBD3L2 ( Fig 2A ) . The increase in MYC did not appear to correspond to a large change in MYC protein stability as determined by western blotting after cycloheximide ( CHX ) treatment to block de novo translation ( S5C Fig ) . There was no clear evidence of direct transcriptional activation of MYC by DUX4 based on prior chromatin immunoprecipitation sequencing ( ChIP-seq ) data [13] which did not identify a DUX4 binding site near the MYC gene ( S5A Fig ) , although PolII ChIP showed a modest increase in promoter-associated and elongating PolII occupancy at the MYC promoter ( S5D Fig ) , about 2-fold and 4-fold , respectively . Most dramatic , however , was a clear enhancement of MYC mRNA stability following DUX4 induction as evidenced by the increased half-life from roughly 44 minutes to 360 minutes in DUX4-expressing cells ( Fig 2B ) . To determine whether this increase in mRNA stability was specific to MYC or a more general effect on mRNA stability , we assessed the stability of two other labile mRNAs , JUN and CITED2 , following DUX4 expression . Although these results were less dramatic , the trend indicated an increase in mRNA stability of both genes following DUX4 expression ( CITED2 from approximately 45 to 182 minutes and JUN from 52 to 140 minutes; S5E Fig ) . We conclude that , although the effect might not be specific to MYC , a major contribution to the increased MYC levels was likely through inhibition of mRNA degradation . These results are in agreement with a more general inhibition of mRNA degradation pathways in DUX4-expressing cells , similar to the previously demonstrated inhibition of NMD which leads to the stabilization of numerous labile mRNAs [14] . Studies in other cells have shown that the MYC-mediated pathway of apoptosis involves MYC activation of EGR1 , and then MYC and EGR1 together activate expression of two BH3-only factors BCL2L11 and PMAIP1 ( also known as BIM and NOXA , respectively ) which antagonize the anti-apoptotic BCL2 proteins and lead to the loss of mitochondrial membrane potential [22] . Analysis of our RNA-seq data revealed that DUX4 induced the expression of EGR1 ( moderated fold-change of 48 ) , induced the expression of BIMγ , a specific isoform BCL2L11 with an alternative BH3-like domain [23] , and increased total BCL2L11 transcripts roughly 2-fold ( Fig 2C and S4 Table ) . The siRNA library used in our screen contained two separate siRNA pools against different splice isoforms of BCL2L11 . One of the two pools rescued RD-DUX4i viability ( Z-score = 2 . 01 ) and contained three siRNAs that targeted only the BIMγ isoform ( 3 out of 4 siRNAs ) , whereas the second pool did not rescue ( Z-score = -0 . 59 ) and did not contain siRNAs that targeted only this isoform ( S5F Fig ) . Together our data indicate that DUX4 expression resulted in decreased degradation of the MYC mRNA and increased abundance of the MYC protein with subsequent activation of components of the MYC-mediated apoptotic pathway , which has been shown to sensitize cells to apoptotic stimuli [24] . As noted above , knockdown of RNASEL , a gene involved in innate immunity to viral infection , rescued RD cells from DUX4 lethality ( DUX4 Z-Score = 3 . 71 ) and had minimal effect on the doxycyline induction of the luciferase transgene ( Luciferase Z-Score = -0 . 07 ) . This rescue was validated as on-target based on our deconvolution strategy ( S4B Fig ) and prompted us to look for other mediators of the cellular antiviral response in our screen . EIF2AK2/PKR is similarly involved in antiviral response and its knockdown marginally rescued from DUX4 toxicity in our original screen ( Z-score 1 . 93 ) , but it passed all other criteria of our validation and filtering process ( S4A and S4B Fig ) . These results led us to further postulate that at least part of DUX4 toxicity might be mediated via triggering an innate ( antiviral-like ) immune response in expressing cells . Both EIF2AK2/PKR and RNASEL are primary responders of the double stranded RNA innate immune response . Because DUX4 leads to the accumulation of normally degraded RNAs , including those destined for nonsense mediated decay [14] , activates the expression of RNAs from repetitive regions and retrotransposons [13] , and stabilizes some mRNAs ( this study ) , we considered the possibility that DUX4-expressing cells could increase the abundance of aberrant , endogenously formed dsRNAs . In order to test whether DUX4-expressing cells accumulate dsRNAs , we performed immunofluorescence using the J2 antibody , which recognizes dsRNAs , irrespective of the sequence [25] . These experiments revealed that DUX4-expressing cells had strong , nuclear dsRNA foci which were not present in uninduced cells ( Fig 3A ) . The signal observed after transfection of the dsRNA surrogate , poly ( I:C ) , was similar in intensity to DUX4 expressing cells , though occupying a more cytoplasmic compartment ( Fig 3A ) . This signal is not a non-specific property of the J2 antibody as we detected similar focal staining in DUX4-expressing cells using the separate monoclonal dsRNA-recognizing antibody K1 ( S6A Fig ) , nor is it an artifact of the doxycycline induction of transgene mRNA expression as our RD-LUCi line did not have similar nuclear staining after doxycycline treatment ( S6B Fig ) . Immunofluorescence of DUX4-induced RD cells using either the J2 or K1 antibodies showed that approximately 5–10% of the nuclei had obvious dsRNA aggregation at 19 hours post induction , whereas the uninduced cells did not exhibit any detectable nuclear staining . Accumulation of dsRNAs activates EIF2AK2/PKR via auto-phosphorylation . EIF2AK2/PKR , in turn , phosphorylates eukaryotic initiation factor 2 alpha ( EIF2S1/eIF-2α ) , which is thought to generally inhibit both host and viral gene translation , culminating in cellular apoptosis [26] . To determine whether dsRNA accumulation leads to EIF2AK2/PKR activation in DUX4-expressing cells , we measured phosphorylated EIF2AK2/PKR in our DUX4-expressing RD cells and observed that , following DUX4 expression , EIF2AK2/PKR was phosphorylated to a level that is comparable to activation via transfection of poly ( I:C ) , whereas total EIF2AK2/PKR levels remained unchanged ( Fig 3B ) . To determine whether the EIF2AK2/PKR activation as a result of DUX4 expression has a functional consequence in our cells , we also blotted for EIF2S1/eIF-2α phosphorylation in a time course following doxycycline induction . This experiment showed that both EIF2AK2/PKR and EIF2S1/eIF-2α became phosphorylated approximately 15 hours post doxycycline addition ( Fig 3C ) . The profound inhibition of the NMD pathway by DUX4 [14] might be the cause of both the increased MYC mRNA and the accumulation of dsRNA because the NMD endonuclease SMG6 has been shown to degrade MYC as a non-canonical NMD target [27] and NMD inhibition leads to dsRNA accumulation in yeast [28] . Therefore , we used siRNAs to knockdown two components of the NMD pathway , UPF1 and SMG6 . Knockdown of either NMD component alone or in combination showed a mild , non-significant ~1 . 1–1 . 6 fold increase in MYC mRNA and a ~1 . 6–2 . 3 fold increase in the inclusion of an NMD-targeted exon in SRSF3 , compared to the more than 24-fold increase in these RNAs following DUX4 induction ( S7 Fig ) . Therefore , NMD inhibition through these mechanisms was not sufficient to stabilize the MYC mRNA . To determine whether DUX4 induces the same MYC-mediated apoptotic pathways and dsRNA toxicity in human myoblasts , we assessed the expression of MYC , components of the MYC-mediated apoptotic pathway , and the dsRNA innate immune response in an immortalized human myoblast cell line with a doxcycycline-inducible DUX4 ( MB135-DUX4i ) . Our previous RNA-seq data generated using MB135-DUX4i cells fourteen hours after DUX4 induction [29] showed an approximate 3 . 3-fold increase ( based on an EdgeR differential expression analysis ) in MYC mRNA abundance and , interestingly , a preferential accumulation of transcripts from the P1 promoter ( Fig 4A ) . Similar to RD cells , DUX4 increased the abundance of EGR1 ( 5 . 4-fold ) and the BH3-only proteins PMAIP1/NOXA ( 11-fold ) and BCL2L11 ( 4 . 7-fold ) ( see Supplemental Table 1 in reference [29] ) , and induced expression of the BIMγ isoform of BCL2L11 ( Fig 4B ) . Therefore , the same MYC-mediated apoptotic pathways induced by DUX4 in RD cells were similarly induced by DUX4 in human myoblasts . To determine whether DUX4 expression in human myoblasts activated the same dsRNA pathway as in RD cells , we stained DUX4-expressing myoblasts with the J2 antibody and found strong focal nuclear dsRNA signal , similar to the observed pattern in RD cells ( Fig 4C ) . Expression of DUX4 also led to phosphorylation of EIF2AK2/PKR as well as EIF2S1/eIF-2α in DUX4 expressing myoblast cells , comparable to levels seen after poly ( I:C ) transfection ( Fig 4D ) . Similar to RD cells , knockdown of either MYC or RNASEL rescued the MB135-DUX4i human myoblasts from DUX4-induced cell death , which was measured by counting viable cells at two and four days after induction of DUX4 expression ( S8A Fig ) . However , knock-down of either MYC or RNASEL in human myoblasts also caused a delay in the accumulation of DUX4 protein and the activation of a DUX4 target gene , MBD3L2 ( S8B and S8C Fig ) . Although DUX4 expression was robust at four days following induction ( S8B and S8C Fig ) , a time when the MYC knock-down continued to show strong rescue , these rescue experiments should be interpreted cautiously regarding the individual necessity of MYC or RNASEL for DUX4-induced apoptosis in human myoblasts . In summary , expression of DUX4 in myoblasts , as in RD cells , resulted in increased MYC mRNA and components of the MYC-mediated apoptotic pathways as well as accumulation of dsRNA and activation of the pro-apoptotic dsRNA-sensing innate immune response . Next , we determined whether the dsRNA aggregates corresponded to known nuclear subdomains or RNA binding proteins by performing immunofluorescence using the K1 dsRNA antibody and costaining with antibodies to promyelocytic leukemia bodies ( PML ) , paraspeckles ( NONO ) , the exon junction complex ( EJC ) components EIF4A3 and RBM8A/Y14 , and TARDBP/TDP-43 , which has been shown to aggregate in DUX4-expressing nuclei [30] and may regulate cellular dsRNA accumulation [31] . Components of PML bodies and paraspeckles were not strictly associated with dsRNA staining ( S9A Fig ) , though NONO did form condensations in the nucleus following DUX4 induction , some of which overlapped with dsRNA staining ( S9B Fig ) . TARDBP/TDP-43 did not show a clear association with the dsRNA foci ( S9D Fig ) . In contrast , DUX4 strongly induced redistribution of EIF4A3 into aggregates that were almost entirely associated with dsRNA foci as determined by double-staining with either the J2 or K1 antibodies ( Fig 5A ) . In many cases DUX4-induced cells with EIF4A3 aggregates appeared to have reduced nuclear staining beyond the aggregates ( Fig 5A ) suggesting that the majority of EIF4A3 was redistributed to the dsRNA foci , similar to the depletion of MBNL proteins by the mutant nuclear RNA in myotonic dystrophy [32 , 33] . The other EJC component tested , RBM8A/Y14 , showed some association with the dsRNA foci but was not strictly associated with dsRNA aggregates , nor was it depleted from the rest of the nucleus ( S9C Fig ) . A time course staining cells at 0 , 16 , or 24 hours following DUX4 induction with antibodies to either dsRNA or EIF4A3 ( Fig 5B ) showed the initial accumulation of cytoplasmic dsRNA at 16 hours , a time point where EIF2AK2/PKR and EIF2S1/eIF-2α phosphorylation was becoming evident ( see Fig 4D ) . The formation of nuclear dsRNA foci initiated at 16 hours and increased through 24 hours . EIF4A3 nuclear foci were present at 16 hrs and more abundant at 24 hours . DUX4 expression inhibits NMD and at least part of this inhibition is secondary to decreased UPF1 protein levels [14] , and knockdown of UPF1 showed a modest increase in NMD-targeted SRSF3 isoform in RD cells ( S7 Fig ) . Similarly , knockdown of EIF4A3 or RBM8A/Y14 in MB135 cells resulted in a substantial increase ( ~2 . 5–3 . 8 fold ) in the NMD-targeted SRSF3 isoform ( Fig 5C ) , suggesting that the nuclear sequestration of EIF4A3 might contribute to NMD inhibition in DUX4-expressing cells . MYC mRNA levels were not affected by EIF4A3 of RBM8A/Y14 knockdown ( Fig 5C ) . Endogenous DUX4 is expressed in only a small percentage of cultured FSHD muscle cells [2] . Therefore , to determine whether DUX4 expression in FSHD cells correlates with higher levels of MYC mRNA , we used an RNA-seq dataset from FSHD cells FACS sorted based on the expression of a DUX4-reporter gene [7] . Based on our previous analysis of this dataset ( see Supplemental Table 1 in Jagannathan , et al [29] ) , muscle cells expressing the DUX4-responsive promoter showed an almost 2-fold increase in the level of MYC mRNA ( log2 fold-change ~0 . 9 and adjusted p-value = ~0 . 003 ) ; however , EGR1 , PMAIP1/NOXA and BCL2L11 were not significantly upregulated . Therefore , DUX4 expression is associated with higher levels of MYC mRNA in FSHD muscle cells , but the role in activating an apoptotic pathway requires further study . Immunofluorescence for dsRNAs in differentiated FSHD muscle cells showed either cytoplasmic or nuclear dsRNA staining in some of the cells with DUX4-positive nuclei but not in a control cell line ( Fig 6A ) . In addition , the nuclear dsRNA foci in FSHD muscle cells were associated with EIF4A3 aggregates that were not present in control muscle cells ( Fig 6B ) . Therefore , DUX4-expressing FSHD muscle cells showed increased levels of MYC mRNA and formation of nuclear foci with dsRNA and EIF4A3 accumulation , indicating that the initial screen identified pathways relevant to FSHD biology , and perhaps pathophysiology .
In this study , an siRNA screen identified that the MYC-mediated apoptoic pathway as well as components of the dsRNA response pathways were involved in DUX4-induced apoptosis in RD cells , and led to the demonstration that DUX4 expression resulted in the stabilization and accumulation of MYC mRNA , the accumulation of nuclear dsRNA , and formation of nuclear EIF4A3 foci . The accumulation of MYC mRNA correlated with higher MYC protein levels and the transcriptional activation of components of the MYC-mediated pro-apoptotic pathway , including EGR1 , BCL2L11 and its BIMγ isoform . The accumulation of dsRNA was correlated with the activation of the EIF2AK2/PKR innate immune response and the phosphorylation of EIF2S1/eIF-2α . Inhibiting the expression of individual components of each of these pathways with siRNAs diminished apoptosis in RD cells in response to DUX4 . These pathways were also induced by DUX4 in an immortalized human skeletal muscle cell line , with increased MYC RNA and the downstream effectors of MYC-mediated apoptosis ( EGR1 , PMAIP1/NOXA , and the BIMγ isoform of BCL2L11 ) , as well as the accumulation of dsRNA , EIF2AK2/PKR and EIF2S1/eIF-2α phosphorylation , and nuclear foci consisting of EIF4A3 and dsRNA . Finally , the elevation of MYC mRNA , dsRNA accumulation , and EIF4A3 nuclear foci in FSHD muscle cells suggests that these processes might contribute to FSHD pathophysiology . Although DUX4 expression caused apoptosis in a P53-dependent manner in some cells [5] , we found that P53 was not necessary for apoptosis in either RD cells or human myoblasts . It is possible that different requirements for P53 might be due to differences in the cell type or the mechanism of DUX4 delivery . This would be consistent with the fact that MYC causes apoptosis in a manner that is either P53-dependent or P53-independent , depending on the cellular context [34] . EIF4A3 is a component of the EJC complex that binds the exon-junction of spliced RNAs [35] . Although the mechanism of sequestering EIF4A3 to nuclear foci remains to be determined , the co-localization of the EIF4A3 foci with dsRNA foci suggests that DUX4 induction and/or stabilization of RNAs with dsRNA secondary structure might overwhelm normal nuclear RNA processing and transport , leading to their accumulation as foci with associated RNA binding proteins . The formation of EIF4A3 nuclear aggregates might contribute to the inhibition of NMD following DUX4 expression , because the formation of nuclear aggregates appears to deplete EIF4A3 from the rest of the nucleus and knockdown of EIF4A3 inhibits NMD . In this regard , FSHD might have some parallels with myotonic dystrophy , where the MBNL RNA binding proteins are sequestered in nuclear foci by the repeat expansion in the mutant RNA [33] . However , our prior demonstration that DUX4 expression leads to degradation of the UPF1 protein indicates that the possible sequestration of EIF4A3 is not the only mechanism by which DUX4 might inhibit NMD , and that simply re-expressing EIF4A3 might be unlikely to fully rescue this defect . Further studies will be necessary to determine the identity of the DUX4-induced dsRNAs and the RNAs that mediate EIF4A3 aggregation . Although it is currently difficult to diagram a clear epistatic model from our results , there are multiple known interactions between MYC , the RNA-mediated innate immune response , and NMD . It has been shown that MYC overexpression can inhibit NMD , leading to ROS-mediated stress and EIF2S1/eIF-2α phosphorylation [36] , both of which are associated with DUX4 expression ( [6] and this study ) . Thus the elevated MYC in DUX4 expressing cells might also contribute to NMD inhibition , feeding forward to further enhance levels of DUX4 , which is also an NMD target [14] . Inhibition of NMD might also lead to the accumulation of dsRNAs . In yeast , NMD controls the accumulation of antisense long non-coding derived dsRNAs [28] . It is therefore possible that the profound inhibition of NMD by DUX4 might similarly result in dsRNA accumulation , either from constitutively expressed RNAs or the many repetitive RNA families and LTR-containing ERVs transcriptionally activated by DUX4 [13 , 37] . Elevated levels of dsRNAs would then lead to an innate immune response , as has been observed in cancer cells that reactivate LTR-containing ERVs after DNA methylation inhibitor treatment [38 , 39] . While we observed phosphorylation of EIF2AK2/PKR , which occurs upon binding dsRNA ligand [40] , we did not see a robust interferon response ( based on our RNA-seq data ) . However , DUX4 has been shown to inhibit components of the innate immune response pathway , at least in part through upregulation of DEFB103 [12] , suggesting that the activation of this pathway by DUX4 might have a normal developmental role . The apoptotic pathways induced by DUX4 in RD cells and human myoblasts might give insight into its normal role in development . DUX4 is normally expressed in the testis , likely in the early germline cells such as the spermatogonia and primary spermatocytes , and has recently been reported to be expressed in the thymus , both areas of active developmental apoptosis [2 , 3] . It is possible that one normal function for DUX4 is to elicit cellular apoptosis , and indeed the BIMγ isoform of BCL2L11 that we showed is induced by DUX4 in skeletal muscle is also expressed in the testis [23] . However , the broad transcriptional program activated by DUX4 suggests a role in stem cell function beyond apoptosis , and apoptosis might occur in a cell-context dependent manner . In other words , cells that normally express DUX4 might be primed for apoptosis but also resistant in the correct developmental context . As mentioned above , DUX4 expression blocks the RNA-induced innate immune response to viral infection [12] , which might constitute a mechanism for protecting DUX4-expressing cells from toxicity in some normal developmental contexts compared to the expression of DUX4 in skeletal muscle cells . We speculate that DUX4 expression , which leads to elevated MYC levels , might afford an accompanying growth or fitness advantage in the developmental context where cells are competing for resources and survival . This notion is further supported by the fact that the MYC-mediated competitive advantage depends on the innate immune response pathway [41] . This normal process of cell competition is believed to control developmental size and eliminate cells with weaker attributes . Furthermore , the competitive advantage conferred by MYC might also be used in cancers for clonal expansion [42] . In this regard , it is interesting to note that DUX4 expression has recently been shown to be a causal factor in a subset of B-cell leukemias [43 , 44] . Further work will be necessary to determine how the DUX4-mediated modulation of MYC levels and RNA accumulation might contribute to aspects of development and the pathophysiology of FSHD .
This study used pre-existing de-identified human cell lines from approved repositories and was determined to not be Human Subjects Research by the FHCRC Institutional Review Board . The commonly used RD rhabdomyosarcoma cell line was obtained from the ATCC ( www . atcc . org ) . Myoblast samples were obtained from the Fields center at the University of Rochester ( https://www . urmc . rochester . edu/fields-center . aspx ) and patients gave informed written consent prior to sample use . For the use of third party data , no additional ethics approval was required and original ethics approval can be viewed in Jagannathan , et al [29] and Rickard , et al [7] . RD-DUX4i cells were grown in DMEM ( Gibco ) supplemented with 10% FBS ( Hyclone ) , 1% penicillin/streptomycin ( Gibco ) and 2 . 0 μg/ml puromycin ( Sigma ) . The DUX4 transgene was induced with 1 . 0 μg/ml of doxycycline hyclate ( Sigma ) . Unaffected ( MB135 ) and FSHD ( MB200 ) immortalized human myoblasts were cultured in F10 medium ( Gibco/Life Technologies ) supplemented with 20% FBS and 1% penicillin/streptomycin as well as 10 ng/ml recombinant human FGF ( Promega ) and 1 μM dexamethasone ( Sigma ) . Human myoblasts were differentiated by culturing in knockout serum replacement medium ( Gibco/Life Technologies ) . The DUX4 and firefly luciferase genes were subcloned into the NheI and SalI sites of the pCW57 . 1 vector ( Addgene plasmid #41393 ) . Lentiviral particles were produced by co-transfecting 293T cells with the pCW57 . 1 vector , pMD2 . G ( Addgene plasmid #12259 ) and psPAX2 ( Addgene plasmid #12260 ) using Lipofectamine 2000 ( ThermoFisher ) and following the manufacturer's instructions . Human RD cells were transduced and selected using 2 . 0 μg/ml puromycin ( Sigma ) at a sufficiently low multiplicity of infection to allow for clonal lines to be isolated using cloning cylinders . The MB135-DUX4i monoclonal cell line was produced as described in [29] . RD-DUX4i cells were seeded in a 96-well plate , induced with doxycycline for 48 hours , then assayed for caspase 3/7 activity using Caspase-Glo Assay Technology ( Promega ) according to the manufacturer’s instructions . RD-DUX4i cells were reverse transfected using LipofectamineRNAiMAX ( Invitrogen ) reagent in 96-well plates with 6 , 000 cells/well and 1 . 25 pmol/well of siRNAs ( flexitube , Qiagen ) . Cells were induced with doxycycline at 1 . 0 μg/ml for 48–96 hours prior to assaying , depending on the experiment . Viability was assessed using CellTiter-Glo assay ( Promega ) per the manufacturer’s instructions . Luciferase activity was assessed using ONE-Glo EX Luciferase Assay System ( Promega ) , per the manufacturer’s instructions . Duplexed oligonucleotide containing the guide RNA sequence against TP53 was cloned into lentiCRISPRv2 plasmid ( Addgene plasmid #52961 ) , essentially using the Zhang lab protocol [45] . We used the sequence: CGCTATCTGAGCAGCGCTCA for targeting purposes . Cells were transduced with lentiviral particles produced as above , selected with 2 . 0 μg/ml puromycin and then clonally reseeded such that isolated colonies could be picked . Mutations were subsequently identified by Sanger sequencing Topo-cloned ( ThermoFisher ) PCR amplicons which flanked the targeted region . Cells were treated with 10 . 0 μM of actinomycin D ( Sigma ) in DMSO to test induction of P53 . The Human Druggable Genome siRNA Set V4 . 0 was purchased from QIAGEN and contains 27 , 844 unique siRNAs multiplexed to target 6 , 961 genes in “druggable” functional groups such as receptors , nucleic acid binders , kinases , transcription factors , and signaling molecules . The master library was diluted with nuclease-free water ( HyClone ) into a 96-well parent library at a concentration of 0 . 5 pmol/μl and then converted to a 384-well child library set . All library manipulations were performed with a Beckman Biomek FX liquid handling robot using filtered sterile tips ( ART BioRobitix ) . 2 . 5 μl of 0 . 125 pmol/μl siRNA solution was added to each of three replicate 384-well assay plates ( ThermoFisher Scientific ) . Opti-MEM ( Gibco ) and RNAiMAX transfection reagent ( ThermoFisher Scientific ) were combined at 3 . 0 μl/ml RNAiMAX to Opti-MEM and subsequently distributed at 12 μl/well . 25 . 0 μl of cells at a concentration of 60 , 000 cells/ml of growth medium was then added to each well and plates were incubated for 24 hours at 37°C . Medium was then aspirated using a BioTek ELx405 Select cell washer and doxycycline-containing growth medium was added at 25 . 0 μl/well using the BioTek Micro Flo Select and incubated for an additional 72 hours . CellTiter-Glo reagent was added at 20 . 0 μl/well and luminescence was read on a microplate reader ( Perkin Elmer EnVision 2104 ) . Analysis of data was performed using the CellHTS2 Bioconductor package [46] and was normalized by using the median method . Protein interaction network analysis was performed using ConsensusPathDB software considering only high-confidence protein interactions [47] . RD-DUX4i cells were pre-incubated in the presence of doxycycline for 8 hours prior to treatment with DRB ( Sigma ) at 75 μM for up to two hours and levels were quantified using real time qPCR . Cells were seeded at 66 x103 cells/well in 12-well plates . RNA was isolated using TRIzol reagent ( ThermoFisher Scientific ) according to manufacturer’s instructions . Isolated RNA was treated with DNaseI ( ThermoFisher Scientific ) , heat inactivated and reverse transcribed into cDNA using Superscript III ( ThermoFisherScientific ) following the manufacturer’s protocol . Real time qPCR was performed using SYBR green reagent ( ThermoFisher Scientific ) for quantification . RNA-seq of RD-DUX4i cells was performed on RNA samples collected from 3 experimental replicates induced for 16–18 hours with doxycycline . Illumina TruSeq libraries were prepared with 500 ng total RNA per sample using the standard library preparation protocol . 100 bp single-end sequencing was performed using the HiSeq 2500 platform by the FHCRC Genomics facility . Base calling was performed with Real Time Analysis software version 1 . 18 . 66 . 3 . Raw reads were aligned to UCSC hg38 using Bowtie2 version 2 . 2 . 6 and Tophat version 2 . 1 . 0 [48] . Gene counts were calculated using the Gencode version 22 annotation file ( obtained from the UCSC genome browser ) and the GenomicAlignments version 3 . 3 Bioconductor package [49] in R version 3 . 4 . 0 using the intersection-strict mode . Differential expression analysis was performed using DESeq2 version 1 . 12 . 4 [50] . FPKM reads were calculated using a custom R script . Processed data were visualized using IGV software [51] . Raw and processed data have been deposited onto the NCBI Gene Expression Omnibus under the accession number GSE87495 . Reduced and boiled samples ( typically 10–20 μg total protein per assay ) were run on NuPage 4–12% precast polyacrylamide gels ( Life Technologies ) and transferred to PVDF membrane ( Life Technologies ) . After blocking in 5% milk in PBST , the membrane was incubated with appropriate antibodies ( described below ) in block solution overnight at 4°C . Membranes were then incubated with appropriate HPRT-conjugated secondary antibodies in block solution for one hour at room temperature and chemiluminescent substrate ( ThermoFisher Scientific ) was used for detection . Densitometry , when performed , was achieved using ImageJ software [52] . For all experiments involving RD cells , the cells were fixed with 4% paraformaldehyde ( Electron Microscopy Sciences ) in PBS for seven minutes at room temperature prior to permeabilization with 0 . 5% Triton X-100 ( Sigma ) in PBS . Cells were then incubated with 1% BSA in PBS block solution for 30 minutes at room temperature and then incubated overnight with J2 or K1 antibody at a concentration of 2 μg/ml at 4°C . Appropriate FITC- or TRITC-conjugated secondary antibodies ( Jackson ImmunoResearch ) and DAPI stain ( Sigma ) were then used prior to visualization . For the human myoblast experiment , conditions were essentially the same , except that cells were fixed with 2% paraformaldehyde and co-incubated overnight with J2 or K1 and anti-DUX4 ( E5-5 ) antibody . Cells were imaged on a Zeiss AxioPhot or , if indicated , a Leica TCS SP5 II confocal microscope and channel merging was performed using ImageJ software . Cross-linked ChIP coupled with MNase digestion was performed in triplicate on RD-DUX4i cells post doxycycline induction as described in [53] with two minor modifications: decreased MNase concentration ( 15 units of Worthington MNase incubated at 37°C for 12 min ) and greater sonication intensity ( 4 pulses at 30% amplitude , 15 seconds per pulse with 1 min rest between pulses ) . Briefly , 10 million RD-DUX4i cells were harvested per sample , 12 . 5 hours after the induction of DUX4 . After crosslinking with 1% formaldehyde for 10 min , samples were treated with MNase followed by sonication . 10 percent of the soluble chromatin from each sample was reserved as input and the remainder of each sample was divided into three equal aliquots for the mock , 8WG16 , and Ser2P ChIPs . Pol II antibodies were pre-bound to Protein G dynabeads ( ThermoFisher Scientific ) for 4 hours at room temperature , according to the manufacturer's directions and incubated with chromatin overnight . Histone H3 , Abcam ab1791; MBD3L2 , Abcam ab107999 , lot GR126890-2; MYC ( 9E10 ) , Santa Cruz Biotech SC-40; dsRNA , SciCons J2 , batch J2-1505; dsRNA , SciCons K1 , batch K1-1502; PKR , Cell Signaling D7F7 , lot 1; phosphorylated PKR , Abcam ab32036 , lot GR155191-6; eIF-2α , Santa Cruz Biotech SC-11386; phosphorylated eIF-2α , Abcam ab32157; 8WG16 UnphosphorylatedPolII , Abcam ab817 , lot GR153063-402; Serine2 phosphorylated PolII , Abcam ab5095 , lot GR271493-1; PML , Santa Cruz Biotech SC-5621 , lot B1216; eIF4A3 , Abcam ab180573 , lot GR148643-3; nmt55/p54nrb ( NONO ) , Abcam ab133574 , lot GR97976-7; Y14 ( RBM8A ) , Abcam ab181038 lot GR152694-1; TDP43 , Proteintech 1078-2-AP; p53 ( DO-1 ) , Santa Cruz Biotech SC-126 , lot l2316; rabbit monoclonal antibodies against DUX4 ( E14-3 and E5-5 ) were produced in collaboration with Epitomics and are described elsewhere [54] . | Facioscapulohumeral dystrophy ( FSHD ) is a common form of muscular dystrophy which is currently untreatable . It is caused by the inappropriate expression in skeletal muscle of the gene DUX4 that encodes a transcription factor normally expressed in some stem cells . When DUX4 is expressed in cultured human or mouse skeletal muscle cells , it activates a program of cell death . Knowing the molecular basis for the cell death induced by DUX4 is important to determine the mechanism of muscle damage in FSHD . We used a molecular screening approach to identify genes and pathways necessary for DUX4 to induce the cell death program . We found that DUX4 activated a known MYC-induced cell death pathway , at least in part through stabilization of MYC mRNA . We also found that DUX4 expression led to an accumulation of double stranded RNAs ( dsRNAs ) that induced a cell death pathway evolved to protect against viral infections . This dsRNA accumulation was accompanied by aggregation of the EIF4A3 protein , a factor involved in mRNA surveillance and decay , which may provide a partial mechanism for how DUX4 can inhibit RNA quality control pathways in cells . Because FSHD muscle cells have increased MYC mRNA , dsRNA accumulation , and EIF4A3 nuclear aggregates , we conclude that these processes might contribute to FSHD pathophysiology . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"phosphorylation",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"muscle",
"tissue",
"gene",
"regulation",
"cell",
"processes",
"population",
"genetics",
"immunology",
"gene",
"pool",
"non-coding",
"rna",
"stem",
"cells",
"population",
"biology",
"small",
"interfering",
"rnas",
"animal",
"cells",
"proteins",
"gene",
"expression",
"biological",
"tissue",
"myoblasts",
"immune",
"response",
"muscle",
"cells",
"biochemistry",
"rna",
"double",
"stranded",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"post-translational",
"modification",
"anatomy",
"apoptosis",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"evolutionary",
"biology"
]
| 2017 | DUX4-induced dsRNA and MYC mRNA stabilization activate apoptotic pathways in human cell models of facioscapulohumeral dystrophy |
Using robust , integrated analysis of multiple genomic datasets , we show that genes depleted for non-synonymous de novo mutations form a subnetwork of 72 members under strong selective constraint . We further show this subnetwork is preferentially expressed in the early development of the human hippocampus and is enriched for genes mutated in neurological Mendelian disorders . We thus conclude that carefully orchestrated developmental processes are under strong constraint in early brain development , and perturbations caused by mutation have adverse outcomes subject to strong purifying selection . Our findings demonstrate that selective forces can act on groups of genes involved in the same process , supporting the notion that purifying selection can act coordinately on multiple genes . Our approach provides a statistically robust , interpretable way to identify the tissues and developmental times where groups of disease genes are active .
Genetic variation is introduced into the human genome by spontaneously arising de novo mutations in the germline . The majority of these mutations have , at most , modest effects on phenotype; they are thus subject to nearly neutral drift and can be transmitted through the population , with some increasing in frequency to become common variants . Conversely , de novo mutations with large effects on phenotype may be subject to many different selective forces , both positive and negative , with the latter resulting in either the variant being completely lost from the population or maintained at very low frequencies [1] . Large-scale DNA sequencing can now be used to comprehensively assess de novo mutations , with many current applications focusing on the protein-coding portion of the genome ( the exome ) . This approach has been used to identify causal genes and variants in rare Mendelian diseases: for example , exome sequencing of ten affected individuals with Kabuki syndrome identified the methyl transferase KMT2D ( formerly MLL2 ) as causal , after substantial post hoc data filtering [2] . In complex traits , this approach has successfully identified pathogenic genes harboring de novo mutations in autism spectrum disorders [3] , intellectual disability [4] and two epileptic encephalopathies [5]; notably , all these studies sequenced the exomes of parent-affected offspring trios and quantified the background rate of de novo mutations in each gene using formal analytical approaches . They were thus able to identify genes harboring a statistically significant number of mutations , which are likely to be causal for disease [5 , 6] . These large-scale exome sequencing studies have demonstrated that the rate of non-synonymous de novo mutations is markedly depleted in some genes , and that these genes are more likely to harbor disease-causing mutations [6] . As synonymous de novo mutations occur at expected frequencies , this depletion is not driven by variation in the local overall mutation rate; instead , these genes appear to be intolerant of changes to amino acid sequence and are thus under selective constraint , with non-synonymous mutations removed by purifying selection . These genes represent a limited number of fundamental biological roles , which suggests that entire processes , rather than single genes , are under selective constraint . This is consistent with the extreme polygenicity of most human traits , where hundreds of genes play a causal role in determining organismal phenotype [7 , 8] . These genes must participate in the same cellular processes , but uncovering the relevant connections and the cell populations and developmental stages in which they occur remains a challenge . We and others have described statistical frameworks to test connectivity within a nominated set of genes [9–11] by considering how genes interact either in annotated pathways or in networks derived from protein interactions or gene co-expression across tissues , and these approaches have been successfully applied to detecting networks of genes underlying neurodevelopmental disease [12] . These studies have demonstrated that genes underlying complex diseases tend to aggregate in networks; we hypothesize that the same is true of constrained genes . However , unlike disease traits where the relevant organ system is known and hypotheses about pathogenesis can by formulated , the phenotypic targets of selective forces are usually unknown . Thus , systematic genome-wide approaches to assessing connectivity between a set of genes of interest and to identify relevant tissues are required to investigate how selective constraint acts on groups of genes and uncover the relevant physiology . To address these issues we have developed a robust , unbiased framework and applied it to genome-wide selective constraint data derived from exome sequences of 6 , 503 individuals [6] . We identified a single , statistically significant subnetwork of 72 interacting genes highly intolerant of non-synonymous variation , with no other interacting groups of genes showing evidence of such coordinate constraint . To establish biological context for this subnetwork , we developed a robust approach to test for preferential expression of the module as a whole , rather than the individual constituent genes . Using gene expression data from the cosmopolitan atlas of tissues in the Roadmap Epigenome Project [13 , 14] , we found that this subnetwork is preferentially expressed in several early-stage tissues , with the strongest enrichment in fetal brain . To more carefully dissect the role of this subnetwork in the central nervous system , we analyzed expression data from BrainSpan [15] , an atlas of the developing human brain , and found that the constrained gene subnetwork is preferentially expressed in the early development of the hippocampus . Consistent with this observation , this module is enriched for genes mutated in neurological , but not other , Mendelian disorders . We thus show that selective constraint acts on a set of interacting genes active in early brain development , and that these genes are in fact intolerant of mutation . Our Protein Interaction Network Tissue Search ( PINTS ) framework is publicly available at https://github . com/cotsapaslab/PINTSv1 .
We have previously described a framework to assess selective constraint across coding sequences in the genome [6] . Briefly , we calibrated an expectation for all possible conversions of one base to another by mutation from non-coding sequence . For each transition , we modeled the effect of the surrounding sequence and its conservation across species to correct for context effects . We then counted the number of synonymous and non-synonymous variants in the coding sequence of each gene in the genome and derived a statistic of constraint on each class of variation compared to this global expectation . We found that a number of genes show decreased rates of non-synonymous substitution but expected rates of synonymous substitution , consistent with purifying selection removing the non-synonymous alleles from the population . If constrained genes lie in biologically meaningful networks , we expect them to ( i ) interact and ( ii ) be expressed in the same tissues . We developed a robust , modular workflow ( PINTS–Protein Interaction Network Tissue Search; Fig 1 ) to test both of these hypotheses at a genome-wide level . To detect interactions between constrained genes we used a high-confidence protein-protein interaction network ( InWeb [16] ) , and employed a clustering algorithm previously validated on such networks [17] . We assessed significance empirically by randomly reassigning constraint scores to genes ( see Materials and Methods and S1 Text ) . We then tested any significant subnetworks for preferential expression in the diverse tissue atlas provided by the Roadmap Epigenome Project ( REP ) , which assays gene expression in 27 human primary samples across the developmental spectrum [14] . Our final dataset is comprised of 9729 genes both present in InWeb and detected in at least one REP tissue . Our workflow is both modular and flexible: clustering algorithms , gene-gene relationships and tissue atlases can be replaced as required , so that analyses can be tailored to suit specific biological problems . A flexible implementation , including all data described here , is freely available as an R package at https://github . com/cotsapaslab/PINTSv1 . We define highly constrained genes as those with evidence of constraint on non-synonymous de novo substitutions ( p < 5 x 10−6 , Bonferroni correction for the number of genes in our InWeb dataset ) but null synonymous constraint scores , indicating intolerance to functionally relevant mutation rather than fluctuations in the local mutation rate [6] . Of these , 107/9729 genes pass this stringent threshold ( binomial p < 2 . 2 x 10−16; S1 Table ) , and form the core of the analysis presented here . We found that 67/107 form a connected subnetwork ( Fig 2A; Table 1 ) . Five additional genes are included as our cluster detection algorithm by design looks for a backbone of null nodes connected to many signal nodes . To assess the significance of this observation , we randomly distribute constraint scores to InWeb nodes 1000 times and find that the constrained subnetwork is larger ( number of nodes: p < 0 . 001 ) and more densely connected ( number of edges: p < 0 . 001; clustering coefficient: p = 0 . 008 ) than expected by chance ( Fig 2B ) . As such , it also explains more total constraint in the genome than expected ( sum of constraint scores: p < 0 . 001 ) . After accounting for the genes forming this subnetwork , we found no evidence that the remaining 35 genes form statistically significant subnetworks by our criteria . The genes in the constrained subnetwork appear to represent several fundamental cell processes , most notably mitosis and cell proliferation ( SMC1A , SMC3 , CTNNB1 ) and transcriptional regulation ( CHD3 , CHD4 , SMARCA4 ) . We performed a formal pathway analysis to further test this and found enrichment of several annotated pathways reflecting these fundamental processes ( Table 2 ) . Encouraged that our detected subnetwork represents one or more biological processes under constraint , we sought to add cellular context to our observations . In particular , we wanted to determine if this group of genes is preferentially expressed in particular tissues , indicating a likely site of action . We thus developed an approach to estimate the joint probability of preferential expression of the genes in the subnetwork in each tissue of an atlas of expression data , while accounting for how frequently each gene is detected across the entire atlas . We applied our approach , which uses Markov random fields , to the expression data on 27 primary tissues and cell lines available from the Roadmap Epigenome Project . Using two conservative permutation-based significance tests , we find the constrained subnetwork is preferentially expressed in a number of fetal and immune tissues ( Fig 2C and Table 3 ) , including fetal brain ( permuted p < 0 . 001 ) , the immune cell subpopulations marked by CD34 ( permuted p < 0 . 001 ) and CD8 ( permuted p = 0 . 017 ) and fetal thymus ( permuted p = 0 . 048 ) . We note that , whilst only a subset of genes are expressed in any one tissue , the combinations of genes expressed in these tissues is highly statistically significant: each gene is only expressed in a small subset of the tissues interrogated , so the cumulative probability of seeing these genes coordinately expressed in any one tissue is small . As several tissues are enriched for subnetwork expression , we sought to understand whether we were capturing the same signature across multiple tissues reflecting a shared process . We assessed whether the same genes are preferentially expressed in each tissue , and found a distinct signature in the fetal brain and heart samples and the immune cell subpopulations ( CD34+ , CD8+ , CD3+ , thymus; pairwise p < 0 . 05 hypergeometric test; S2 Table ) . To ensure our tissue expression results are not an artifact of the threshold we set for preferential expression , we repeated the entire analysis with a range of threshold values and found consistent results across tissues; this is most notable in fetal brain ( Fig 2D and S3 Table ) , which remains significant irrespective of threshold used . Genes under selective constraint are more likely to harbor pathogenic mutations causing Mendelian diseases , consistent with intolerance of functional mutations [6] . Accordingly , we found that our subnetwork of 72 genes is significantly enriched for OMIM annotations ( Fisher’s exact p = 0 . 0013 ) . To further elucidate this observation , we mapped all OMIM entries to Medical Subject Headings ( MeSH ) disease categories and assessed enrichment per organ system category . We found that our subnetwork is significantly enriched for genes mutated in Mendelian diseases affecting the central nervous system ( Fisher’s exact p = 0 . 0017 , S5 Table ) , validating our observation of enrichment in fetal brain . We note that this enrichment is not in the inflammatory/immune neurological disease sub-category , suggesting no overlap with the discrete immune signature we found . Samocha et al [6] have previously reported that constrained genes are also enriched for de novo mutations associated with autism spectrum disorders , further strengthening our conclusion that this constrained subnetwork represents a brain-related biological process . To further elucidate the relevance of our constrained module to brain physiology , we interrogated expression data for multiple brain structures across developmental stages from the BrainSpan project [15] . We found a strong signature of preferential expression in very early stages of development , which declines rapidly and is absent by mid-gestation and remains inactive after birth into adulthood ( Fig 3A and Table 3 ) . Several transitional structures in the early brain exhibit significant preferential expression levels , including the ganglionic eminences that eventually form the ventral forebrain and the early structures of the hippocampus . The latter structure shows the most consistent signature across developmental time , with the module’s pattern of expression gradually weakening and becoming non-significant by mid gestation ( post-conception weeks 16–18; Fig 3B ) . These results , taken with the likely involvement of constrained genes in fundamental processes of mitosis and transcriptional regulation , suggest this gene module is relevant to developmental patterning at crucial time points in early brain development .
We have shown that selective constraint influences sets of interacting genes involved in core cellular control processes , and that these have elevated expression levels in early stages of central nervous system development . We found the strongest enrichment in the early hippocampal stages at post-conception weeks eight and nine , with additional signals in ventral forebrain structures and the parietal cortical wall . This stage of development involves neuronal proliferation through carefully orchestrated sequences of cell differentiation during developmental patterning across the brain . As the constrained subnetwork we have detected is enriched for genes involved in the control of mitosis and transcription , we speculate that it plays a fundamental role in these processes . Our finding that neurological Mendelian disease genes are over-represented , combined with previous reports of de novo mutations affecting autism spectrum disorders [6 , 18] , intellectual disability [6] and epileptic encephalopathy [5] , further support this notion , indicating that most perturbation leads to severe phenotype . This strong limitation in tolerance may also explain our observation of enrichment in immune cell populations , as precise control of developmental decisions is crucial to the correct differentiation of the lymphoid and myeloid lineages throughout life . As the selective constraint scores are by design corrected for both coding sequence length and GC bias [6] , constraint is more likely to be due to intolerance of changes to protein function rather than structural characteristics of the encoded proteins . Network analyses have been used to identify interacting groups of genes conserved across species [19] , and to identify groups of co-expressed genes in both healthy individuals [20] and groups of genes whose expression is coordinately altered in neurological disease [21] . In particular , network analyses of expression data across species suggest that co-regulated genes form stable interaction networks that evolve in a coordinate fashion [19] . These diverse analyses all suggest that functionally linked genes form stable networks and are targets of natural selection due to their group contribution to specific biological processes [22] . Our own results support this notion , demonstrating that interacting protein networks are under remarkable constraint within the human species , presumably because they underlie carefully orchestrated biological processes . More broadly , our results present a glimpse into how natural selection may affect entire groups of genes involved in central homeostatic functions . Most studies of selection aim to identify specific alleles inconsistent with the nearly neutral model of drift , with particular success in studies of recent positive selection [23 , 24] . We suggest that the majority of these effects represent near-Mendelian effects on relevant phenotypes , which are the actual targets of selective forces: for example , variability in lactase persistence is almost entirely explained by any one of handful of necessary and sufficient alleles [25] . However , the majority of human traits are polygenic , and selection would likely exert far weaker effects on risk alleles , most of which have been revealed by GWAS to only explain a fraction of phenotypic variance . Although such polygenic adaptation [26] has proven difficult to detect thus far , our data provide confirmation that selective forces can act on groups of genes involved in the same process , supporting the notion that purifying selection can act coordinately on multiple genes . We describe how selective constraint acts on groups of genes , suggesting such coordination , though we note that the constraint statistics contain no information about whether multiple genes are targets of the same pressure . We further note that the substantial preferential expression we see does not apply to the entire constrained subnetwork—this may be due either to imprecise specification of the network itself or limitations in detecting preferential expression in a limited tissue atlas . However , our results clearly support a coherent physiological role for this network in early fetal development . We have presented a robust approach to identifying sets of interacting genes under selective constraint and placing these into biological context , using the wealth of genome-scale data produced by large-scale public projects . Our approach builds on robust statistical frameworks to interrogate single variants or genes and thus provides previously lacking biological context from which further hypotheses can be drawn . The approach is flexible and not restricted to studies of constraint: per-gene measures derived from studies of other forms of natural selection , non-human hominid introgression , common and rare variant disease association can be analyzed in our framework . Further , as PINTS is modular , appropriate tissue atlases can be used to meaningfully interpret results . We believe our work represents a new class of approaches that can leverage multiple genome-scale datasets to gain new insight into biological activities responsible for health and disease .
We have used selective constraint scores as previously described [6] . Briefly , we used a mutation rate table—containing the probability of every trinucleotide XY1Z mutating to every other possible trinucleotide XY2Z—based on intergenic SNPs from the 1000 Genomes project and the sequence of a gene to determine that gene’s probability of mutation . These sequence context-based probabilities of mutation were additionally corrected for regional divergence between humans and macaques as well as the depth of coverage for each base in an exome sequencing study . Given the high correlation ( Pearson’s r = 0 . 94 ) between the probability of a synonymous mutation in a gene with the number of rare ( MAF < 0 . 01% ) synonymous variants in that gene seen in the NHLBI’s Exome Sequencing Project , we used a linear model to predict the number of rare missense variants expected per gene in the same dataset . The difference between observation and expectation was quantified as a signed Z score of the chi-squared deviation . The missense Z score was used as the basis for determining selective constraint . In this study , we took a conservative approach to assessing selective constraint , using the Bonferroni correction for number of InWeb genes to derive a significance threshold of pc < 5 x 10−6 . We used InWeb , a previously described comprehensive map of protein-protein interactions , containing 169 , 736 high-confidence interactions between 12 , 687 gene products , compiled from a variety of sources [16] . By mapping ENSEMBL IDs , we were able to identify 9729 genes with constraint scores from Samocha et al [6] also present in the REP expression data ( below ) , to which we restricted our analysis . To detect clusters of interacting constrained genes , we used a heuristic form of the prize-collecting Steiner tree ( PCST ) algorithm [27 , 28] , which has been previously applied to protein-protein interaction data[17] . The canonical form of the PCST algorithm takes a connected , undirected graph G ( V , E , w , u ) with V vertices and E edges , with vertex weights w and edge weights u; it then finds the connected subgraph T ( V’ , E’ ) with maximal profit ( T ) , which is some function of ( w’-u’ ) . By definition , T is a minimal spanning tree . The algorithm thus identifies the set of nodes with the strongest signal given the cost of their connecting edges . The classical PCST algorithm is , however , NP-hard , which makes it computationally intractable on the scale of InWeb [27] . Several heuristic simplifications have been proposed , including one previously validated as suitable for protein-protein interaction networks which we use here [17] . This approach partitions the set V into null ( with weights w < 0 ) and signal ( with weights w > 0 ) vertices ( genes ) and equal edge weights e before searching for T . Beisser et al have implemented this approach in the BioNet package for the R statistical language [29] . Here , we define signal genes as those with constraint scores passing the Bonferroni threshold of pc < 5 x 10−6 , and calculate the weights as w = -log ( pc ) + log ( 5 x 10−6 ) . The PCST algorithm returns a single , maximal T solution; to discover further independent subnetworks , we apply the method iteratively after we assign gene nodes in the previously discovered solution to be null . The algorithm always returns a solution for T , so we sought to assess the significance of our observations empirically . To understand if the observed solution is unlikely by chance , we permuted the constraint scores of genes 1000 times and for each iteration ran the heuristic PCST to generate 1000 random resampled subnetworks ( these are also used in the tissue-specificity analyses described below ) . We then quantified the following key parameters and assessed how many random subnetworks had values exceeding those of the true discovered subnetwork: size ( number of gene nodes ) ; density ( number of connections ) ; clustering coefficient and total amount of constraint explained ( sum of constraint scores ) . To address the possible contribution of degree bias to these results , we also performed biased permutations to select signal nodes with the same degree distribution as we had previously done for DAPPLE [9] . We found weak correlation between degree and significance ( S1 Fig ) and opted for random permutations where the number of combinations of random genes selected as signal nodes is much larger . We obtained gene expression data for a cosmopolitan set of tissues from the Roadmap Epigenome Project ( REP ) [14] . The REP data consists of 88 samples across 27 tissue types from diverse human organs , profiled on the Affymetrix HuEx-1_0-st-v2 exon array , which we downloaded on 9/25/2013 from http://www . genboree . org/EdaccData/Current-Release/experiment-sample/Expression_Array/ . We processed these data using standard methods available from the BioConductor project [30 , 31] . Briefly , we removed cross-hybridizing probesets , applied RMA background correction and quantile normalization and then summarized probesets to transcript-level intensities . We then mapped transcripts to genes using the current Gencode annotations for human genes ( version 12 ) . Transcripts with no match in Gencode were removed and the remaining transcripts we again quantile normalized . We then assigned transcript expression levels to their matching genes . Where multiple transcripts mapped to the same gene we used the transcript with maximum expression over all cell types . The Brainspan atlas [15] data are available as processed , gene-level expression levels from http://www . brainspan . org/static/download . html . We mapped these genes to the InWeb gene set using ENSEMBL IDs , and quantile normalized data for the overlapping genes . We then grouped replicate data by developmental stage and brain structure and calculated preferential expression as described above . We used a previously described approach to detect tissue-specific expression across each tissue atlas [32] . Briefly , we group together replicates from the same cell type and compute pairwise differential expression between all pairwise combinations of tissues , using an empirical Bayes approach to account for variance shrinkage [33] . Thus , for each gene there are 26 linear model coefficients and associated p values for each tissue , quantifying the comparison to all other tissues . For each gene in each tissue , we then capture the overall difference in expression from all other tissues as the sum of these coefficients . To reduce noise , only coefficients with p < 0 . 0019 ( p < 0 . 05 with Bonferroni correction for 26 tissues ) are considered . Rescaling all coefficient sums across all genes values to the range [−1 , 1] gives us a final preferential expression score . Intuitively , a gene highly expressed in only one tissue would get a high positive enrichment score in that tissue , as it is differentially expressed compared to all other tissues . The score is directional , strong negative values indicate very low expression in one tissue compared to all others . We partition the overall distribution into deciles and define preferential expression in a tissue if a gene has a score > 0 . 1 . To score the tissue specific expression of a subnetwork , we detect which genes in the subnetwork are preferentially expressed in each tissue of our expression atlas and assess the joint probability of this observation . Rather than ask if some nodes of the subnetwork are preferentially expressed in a given tissue , we developed an approach to account for the connections between genes; we thus assess whether the pattern of preferential expression across the whole subnetwork is unusual for a given tissue , suggesting the subnetwork is operational . Formally , we consider the subnetwork as a Markov random field with a particular configuration of preferentially expressed nodes in each atlas tissue . We compute a score for each configuration using a standard scoring function [34]: P ( x1 , … , xn ) =1Z∏ ( i , j ) ∈EdgesΦ ( xi , xj ) The partition function Z is defined as: Z=∑x1 , … , xn∏ ( i , j ) ∈EdgesΦ ( xi , xj ) where xi ( i = 1 , … , n ) represents a binary tissue specificity of the genes in the subnetwork for a given tissue with values either 1 ( expressed ) or 0 ( not expressed ) . The Φ ( xi , xj ) factor lists the co-occurrence of two connected nodes across tissues . This is calculated from the thresholded preferential expression data , and each pair of connected nodes is assigned exactly one configuration in each tissue , so that Φ ( xi=0 , xj=0 ) +Φ ( xi=1 , xj=0 ) + Φ ( xi=0 , xj=1 ) + Φ ( xi=1 , xj=1 ) =number of tissues We assess the significance of these scores using two conservative permutation approaches . First we assess how likely we are to see each observed configuration ( i . e . each pattern of detected/not detected nodes ) in each tissue of the atlas . We do this by permuting the preferential expression scores across tissues for each gene independently and rescoring the configuration found in each tissue . This alters the co-expression structure across genes and empirically assesses how likely we are to see a particular configuration of a specific subnetwork by chance . Second , we estimate the probability of observing the extent of tissue specificity in each tissue . We construct the null expectation by scoring the resampled subnetworks generated by permutation above in each tissue and compute the empirical significance from this distribution of scores . To ensure our results are not artifacts of a specific preferential expression threshold , we repeat this analysis across a spectrum of preferential expression thresholds ( See S3 Table ) . To test if any biological pathways are over represented in a subnetwork , we use the Gene Set Enrichment Analysis ( GSEA ) approach [35] . We obtained the full list of curated canonical pathways from the GSEA website ( http://www . broadinstitute . org/gsea/msigdb/collections . jsp ) and mapped the 9729 genes to each pathway using HUGO IDs . We then test for enrichment of subnetwork members over background using the hypergeometric test . To test if genes in the subnetwork are more likely to harbor pathogenic mutations causing Mendelian diseases than expected by chance , we retrieved OMIM records for all 9729 genes using the biomaRt package in BioConductor [31] . We then tested whether the proportion of 107 subnetwork genes with OMIM entries was higher than the background proportion of the full set of 9729 in our analysis using Fisher’s exact test ( S4 Table ) . We then mapped all OMIM entries to Medical Subject Headings ( MeSH ) disease categories using the Comparative Toxicogenomics Database ( CTD ) MEDIC disease vocabulary [36] and assessed enrichment in any disease category , again using Fisher's exact test ( S6 Table ) . | Some genes are extremely intolerant of mutations that alter their amino acid sequence . Such mutations are highly likely to drive disease , and previous reports have implicated these genes in multiple diseases . To better understand the function of these constrained genes and their place in cellular organization , we developed a framework to ask if these genes form biochemical networks expressed in specific tissues and developmental timepoints . Using clustering analysis over protein-protein interaction maps , we show that 72/107 such genes form a densely connected network . Using another new method , we found that these 72 genes are coordinately expressed in fetal brain and early blood cell precursors , but not other tissues , in the Roadmap Epigenomic Project , and then show that this gene module is active in very early developmental time points of the hippocampus included in the Brainspan Atlas . We also show that these genes , when mutated , tend to cause genetic diseases . Thus we demonstrate that evolution constrains mutation of key mechanisms that must therefore require careful control in both time and space for development to occur normally . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"genetic",
"networks",
"protein",
"interactions",
"protein",
"interaction",
"networks",
"brain",
"mutation",
"mathematics",
"network",
"analysis",
"genome",
"analysis",
"discrete",
"mathematics",
"combinatorics",
"computer",
"and",
"information",
"sciences",
"protein-protein",
"interactions",
"proteins",
"gene",
"expression",
"proteomics",
"biochemistry",
"hippocampus",
"permutation",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"computational",
"biology"
]
| 2016 | Network Analysis of Genome-Wide Selective Constraint Reveals a Gene Network Active in Early Fetal Brain Intolerant of Mutation |
The circadian regulatory network is organized in a hierarchical fashion , with a central oscillator in the suprachiasmatic nuclei ( SCN ) orchestrating circadian oscillations in peripheral tissues . The nature of the relationship between central and peripheral oscillators , however , is poorly understood . We used the tetOFF expression system to specifically restore Clock function in the brains of ClockΔ19 mice , which have compromised circadian clocks . Rescued mice showed normal locomotor rhythms in constant darkness , with activity period lengths approximating wildtype controls . We used microarray analysis to assess whether brain-specific rescue of circadian rhythmicity was sufficient to restore circadian transcriptional output in the liver . Compared to Clock mutants , Clock-rescue mice showed significantly larger numbers of cycling transcripts with appropriate phase and period lengths , including many components of the core circadian oscillator . This indicates that the SCN oscillator overcomes local circadian defects and signals directly to the molecular clock . Interestingly , the vast majority of core clock genes in liver were responsive to Clock expression in the SCN , suggesting that core clock genes in peripheral tissues are intrinsically sensitive to SCN cues . Nevertheless , most circadian output in the liver was absent or severely low-amplitude in Clock-rescue animals , demonstrating that the majority of peripheral transcriptional rhythms depend on a fully functional local circadian oscillator . We identified several new system-driven rhythmic genes in the liver , including Alas1 and Mfsd2 . Finally , we show that 12-hour transcriptional rhythms ( i . e . , circadian “harmonics" ) are disrupted by Clock loss-of-function . Brain-specific rescue of Clock converted 12-hour rhythms into 24-hour rhythms , suggesting that signaling via the central circadian oscillator is required to generate one of the two daily peaks of expression . Based on these data , we conclude that 12-hour rhythms are driven by interactions between central and peripheral circadian oscillators .
Circadian rhythms are daily oscillations of behavior and physiology that allow organisms to anticipate and respond to predictable daily changes in their environment [1]–[4] . In animals , these environmental variables include light , temperature , food availability , and predation . As a consequence , circadian rhythms regulate behaviors such as feeding rhythms and sleep/wake cycles [2] . At a tissue and cellular level , circadian rhythms compartmentalize the activity of biochemical pathways to appropriate times of day in tissues throughout the body [1] . Taken as a whole , the circadian regulatory network significantly influences normal organismal physiology , and contributes to the pathogenesis of clinically significant conditions including cancer , heart disease , and metabolic disorders [5]–[8] . Self-sustained circadian oscillations are generated at a molecular level via an elaborate transcriptional/translational feedback loop [9] . The positive arm of this feedback loop is mediated by bHLH-PAS transcription factors , BMAL1 and CLOCK/NPAS2 [10] , [11] , which heterodimerize and drive the expression of downstream target genes . Among these target genes are Period ( Per ) and Cryptochrome ( Cry ) , whose protein products accumulate in the cytoplasm , associate with each other , and ultimately translocate to the nucleus . Once in the nucleus , PER and CRY inhibit BMAL1/CLOCK activity , repressing their own transcription and thus forming the negative arm of the circadian oscillator [9] . In parallel , a second feedback loop is generated via RORE binding activators ( Rora , Rorb , Rorc ) and repressors ( Rev-erb-alpha , Rev-erb-beta ) , whose transcription is driven by BMAL1/CLOCK [12] . In conjunction with accessory genes that regulate the stability and activity of key circadian proteins [9] , these feedback loops comprise the circadian clock . Ultimately , these molecular oscillations drive 24-hour rhythms of transcription in downstream target genes . Termed “circadian output genes" , these transcriptional rhythms are not necessary for sustaining the core circadian oscillator , but are required for mediating circadian regulation of physiology and behavior [13] . Both core circadian oscillations and rhythmic output genes are found in tissues throughout the body [13] , [14] . However , not every tissue is equally important for maintaining proper circadian rhythmicity at an organismal level . Rhythms in peripheral tissues , such as liver and skeletal muscle , are self-sustaining in vitro , but require inputs from the central circadian oscillator in the suprachiasmatic nuclei ( SCN ) in the hypothalamus for proper coordination in intact animals [15] . The nature of the regulatory signals between the SCN and peripheral tissues ( as well as regulatory cues between peripheral tissues ) is poorly understood , but thought to involve neuronal circuitry , humoral factors ( e . g . glucocorticoids ) , and cascades of behavior ( i . e . the impact of the sleep wake cycle on eating and elimination ) . The general mechanism of circadian oscillations and the genes required for their maintenance is largely conserved between different tissues and species [3] , [13] , [16] , [17] . Nevertheless , circadian output genes are tissue specific , as one would expect given the diverse physiologies regulated by the clock [18]–[20] . Consequently , considerable efforts have been made to characterize circadian transcriptional output at a genome level using microarray technologies [18]–[33] . These studies have made considerable progress towards understanding how tissue-level circadian oscillations are translated into rhythms of organismal physiology . At the same time , this systems-level approach can be used in conjunction with tissue-specific manipulation of gene expression to dissect the relationship between central and peripheral oscillators . For example , a recent study used liver-specific over-expression of REV-ERB-alpha to knock-down BMAL1 expression in the liver , thereby ablating the local circadian oscillator in an otherwise wildtype animal [21] . They examined the impact of this on circadian regulation of liver gene expression by microarray analysis . This analysis showed that the vast majority of circadian output requires a functional circadian oscillator within the liver . Interestingly , there were several exceptions to this rule , indicating that some circadian genes are driven by systemic cues rather than the local circadian clock . Those genes , including Per2 , are therefore strong candidates to function as the relay between the SCN and peripheral tissues [21] . Thus , this study established that a functional liver oscillator is required for normal circadian regulation of liver gene expression . In this manuscript , we seek to extend this research by examining the contribution of central circadian clock function on peripheral physiology . To do this , we employed a tet-OFF expression system to specifically rescue wildtype CLOCK expression in the brains of Clock-mutant mice [21] , [34]–[36] . Brain-specific Clock rescue restored normal behavioral rhythmicity in constant conditions with approximately wildtype period lengths . We used genome-wide transcriptional profiling every two hours for two full days followed by JTK_CYCLE analysis to assess transcriptional circadian output in the mouse liver [37] . The majority of rhythmic genes in the liver required a fully functional liver circadian clock; however , 95 genes still oscillate with circadian periods in the livers of brain-rescued mice , albeit with diminished amplitudes in most cases . We observe that 12-hour transcriptional rhythms ( i . e . , circadian ‘harmonics’ [19] ) are entirely lost in Clock-mutant background . Interestingly , brain-specific rescue of Clock restores 24- , but not 12-hour rhythmicity to these genes , suggesting that systemic and locally-derived circadian cues are independently required for different peaks of these 12-hour rhythms .
Inducible , brain-specific expression of wildtype Clock was achieved using the tet-OFF system [34]–[36] , [38] . Wildtype Clock was tagged with an HA epitope and linked to a tTA-responsive tetO promoter ( tetO::Clock-HA ) . At the same time , tTA was expressed under the control of the Secretogranin II ( Scg2 ) promoter ( Scg2::tTA ) which drives expression specifically in the brain , pituitary , and adrenals , with especially high expression in the SCN [39] . When both transgenes were present in the same animal , wildtype Clock was expressed at constitutively high levels in the SCN [36] . When these mice were treated with low-doses of Doxycycline ( Dox ) via their drinking water , the binding of tTA to tetO promoters was inhibited , and Clock expression was abolished [36] . These mice were then crossed into a ClockΔ19 background . ClockΔ19 is an ENU-generated allele of Clock which results in the loss of exon 19 from mature Clock transcripts [40] , [41] . ClockΔ19 acts as a dominant-negative by binding to BMAL1 and inhibiting its activity [41] . Consequently , ClockΔ19 animals have severely disrupted circadian behavioral rhythms , with extremely long period lengths and arrhythmicity in prolonged constant conditions [40] , [42]–[44] . Wildtype animals showed robust circadian oscillations in 12-hour light/12-hour dark ( LD ) conditions with most locomotor activity restricted to the dark phase . The mice maintained these rhythms in constant darkness ( DD ) , with period lengths slightly shorter than 24-hours , in agreement with previous studies ( Figure 1A ) . When either Scg2::tTA or tetO::Clock-HA were expressed by themselves in a ClockΔ19 background , the mice showed normal LD activity rhythms , but quickly became arrhythmic in DD or had extremely long period lengths ( Figure 1B–1H ) . In all three control genetic backgrounds , the addition of Dox to the drinking water ( highlighted in yellow ) , did not change the circadian behavior of these mice ( Figure 1A–1H ) . Combining Scg2:tTA and tetO::Clock-HA in the ClockΔ19 background resulted in mice with normal LD rhythmicity . Unlike their littermate controls ( i . e . , Figure 1B–1H ) , however , these mice showed robust circadian rhythmicity in constant conditions ( Figure 1I–1L ) . Previous studies have shown [45] that over-expression of wildtype Clock is sufficient to rescue the behavioral phenotype of ClockΔ19 . In agreement with these studies , we detected a modest decrease in the average period length of these animals compared to wildtype ( Figure 1A and 1I–1L ) . When the rescued mice were treated with Dox ( thus inactivating the wildtype Clock transgene expression ) , normal circadian oscillations were quickly lost ( Figure 1I–1L ) . We observed significant animal-to-animal variability in the severity of the resulting phenotype . Generally , however , these mice were either arrhythmic or showed extremely long period lengths while being treated with Dox , consistent with the expected phenotype of ClockΔ19 mice ( Figure 1I–1L ) . This phenotype was completely reversible; when Dox was removed from the drinking water ( thus restoring Clock transgene expression ) , normal circadian rhythmicity was quickly reestablished ( Table 1 ) . Based on these behavioral data and the previously published expression pattern of Scg2::tTA ; tetO::Clock-HA mice [36] , we conclude that this system permits the brain-specific rescue of circadian rhythmicity in locomotor activity in a conditional and reversible manner . Although brain-specific rescue of Clock is sufficient to restore normal behavioral rhythms , it was unclear whether transcriptional and metabolic rhythms in peripheral tissues would be similarly rescued . To answer this , we collected liver samples from wildtype animals as well as Scg2:tTA ; tetO::Clock-HA mice that were treated with either normal drinking water or Dox ( hereafter referred to as tetO::Clock H2O and tetO::Clock DOX ) . Based on the behavioral data presented above , we expected the H2O-treated animals ( i . e . , Clock transgene-expressing ) to have normal brain rhythmicity and behavioral rhythms , while the Dox-treated animals ( i . e . , Clock-defective ) would have disrupted rhythms . It is important to note that in both cases the Scg2 promoter does not express in the liver , and thus , this peripheral clock is presumed to be Clock-defective . Liver samples were collected every two hours for 48 hours in constant darkness ( Figure 2A ) . Total RNA was extracted and global gene expression was profiled using Affymetrix Mouse Exon Arrays . Cycling genes were detected using JTK_Cycle with false-discovery rates ( FDRs ) based on the Benjamini-Hochberg procedure [37] . We found 576 cycling genes in wildtype mice at a FDR cutoff of q<0 . 05 ( corresponding to a p-value threshold of p<0 . 0011 ) , which is consistent with expected levels of transcriptional oscillations given the different sampling resolutions of these studies [18] , [19] , [46] . Over half of the oscillating genes in this study were previously identified as rhythmic [19] , which is an encouraging level of overlap given the typically low agreement between circadian microarray studies [47] , [48] . Known core clock genes – including Per2 , Bmal1 , and Rev-erb-alpha – showed high-amplitude oscillations with expected phase differences , as measured by microarray and confirmed by quantitative PCR ( qPCR ) ( Figure S1 ) . Taken together , these data indicate that this microarray study accurately reflects the underlying circadian transcriptome . At every statistical threshold we examined , wildtype livers showed significantly more rhythmic transcripts than Dox-treated tetO::Clock mice ( i . e . , Clock-mutant ) . Heatmaps of all cycling genes detected , as well as histograms of their amplitudes are shown in Figure S2 . H2O-treated ( i . e . , Clock-rescue ) mice had an intermediate , partially-rescued phenotype , with considerably more cycling genes detected than Dox-treated animals , though still less than wildtype ( Figure 2B and 2C ) . This pattern was also seen in core clock genes and high-amplitude circadian output genes . The strength of rhythmicity ( as measured by the statistical confidence of their detection ) consistently demonstrated that Clock-rescue ( tetO::Clock H2O ) mice had an intermediate phenotype between wildtype and Clock-mutant mice ( Figure 2D and Table S1 ) . Brain-specific expression of Clock did not rescue the amplitude of most high-amplitude transcriptional rhythms ( Figure 2E and Figures S1 and S2 ) . Even though Clock-rescue restored a considerable portion of the normal circadian output of the liver , these rhythms were frequently low-amplitude relative to wildtype , indicating that the local circadian oscillator – and in particular , wildtype Clock expression – is essential for generating high-amplitude rhythms . At the statistical threshold we have selected , there are 187 genes that cycle in one or both of the tetO::Clock samples without cycling in wildtype . Of these 187 genes , 151 were analyzed in Hughes et al . 2009 [19] , and 102 of them were found to oscillate ( Table S2 ) . Based on this , we conclude that the majority of non-wildtype cycling genes in the present study are actually bona fide cyclers that did not meet significance given the relatively stringent statistical threshold used . Consistent with this idea , the median p-value for these 102 genes in wildtype samples in the present study is ∼0 . 1 . Given the false-discovery rates for Clock-rescue and Clock-mutant animals ( q<0 . 10 and q<0 . 37 , respectively at p<0 . 0011 ) , we conclude that cycling genes specific to non-wildtype backgrounds are rare , in agreement with Kornmann et al . 2007 [21] . As expected , the average period length of transcriptional rhythms in wildtype mice was ∼24-hours ( Figure 3A ) . Consistent with the behavioral profiles discussed above ( Table 1 ) , Clock-rescue mice ( tetO::Clock H2O ) also showed average period lengths of ∼24-hours , while Clock-defective mice ( tetO::Clock DOX ) had considerably longer period lengths as would be expected in ClockΔ19 mutant mice ( Figure 3B and 3C ) . This phenotype is illustrated by the profiles of two core clock genes , Per2 and Rev-erb-beta ( Figure 3D and 3E , and Figure S1 ) . Both wildtype and Clock-rescue mice showed normal 24-hour rhythms with phases in agreement in both genotypes . In contrast , Per2 and Rev-erb-beta in Clock-defective mice had a longer-period phenotype with peak expression out-of-phase with wildtype . Since these animals were housed in constant darkness for 2 days before sample collection , we believe this apparent phase difference is a consequence of their free-running period length phenotype . These observations were seen in every core clock gene tested , as well as many key circadian output genes ( Figure 3F ) . In each case , these genes had ∼24-hour rhythms in wildtype and Clock-rescue mice , with considerably longer period length in Clock-defective mice . Overall , 95 genes were found to cycle in both wildtype and Clock-rescue animals ( p<0 . 0011 , corresponding to a q<0 . 05 in wildtype ) . Three of these genes had period lengths of ∼12-hours in wildtype animals ( i . e . circadian harmonics ) and are discussed in greater detail below . The remaining 92 genes all showed period lengths of ∼24-hours in both wildtype and rescue animals , and 77 of the 92 ( 84% ) have been previously seen to oscillate in mouse liver ( Table 2 ) [19] . The phase of circadian output rhythms was restored by brain-specific Clock-rescue . The heatmaps in Figure 4A and 4B demonstrate striking similarity between the profiles of the circadian transcriptomes in both sets of animals . Figure 4C shows the phase difference between wildtype and Clock-rescue as a scatter plot for each of the 92 rescued circadian genes . These data points were centered near zero , although there was a modest ( ∼1 . 5-hour ) phase-advance in Clock-rescue versus wildtype . This phase difference was less than one standard deviation from zero , so we do not consider it to be statistically significant , although we note that the slightly faster behavioral rhythms in Clock-rescue animals may account for this modest phase advance ( Figure 1 and Table 1 ) . Figure 4D shows the phases of all rescued ( blue circles ) and non-rescued genes ( red x's ) . Rescued genes were found with peak expression at every time of day , and there was no significant bias in the phase of rescued versus non-rescued genes ( chi-squared test ) . Overall , 76 of 92 rescued genes ( 82% ) had lower amplitudes in Clock-rescue animals versus wildtype , and the median amplitude in Clock-rescue was 20% lower than in wildtype . As would be expected given the un-rescued expression of ClockΔ19 in the liver , the most significantly affected genes were high-amplitude cyclers that are direct targets of BMAL1/CLOCK . We also compared this dataset to known Bmal1 and Reverb-alpha target genes ( Table 2 ) [49] , [50] . For example , Dbp , Bmal1 ( Arntl ) , Rever-alpha ( Nr1d1 ) , Reverb-beta ( Nr1d2 ) , Per3 , and Tef are among the cycling genes with the most significant amplitude defect in rescue animals ( Figure 4E ) . Likewise , Ubiquitin Specific Protease 2 ( Usp2 ) has significantly diminished amplitude in the rescue , and has also been shown to modify circadian rhythms [51] and is a direct target of CLOCK [52] . Taken as a whole , these data indicate that brain-specific rescue of Clock function is sufficient to restore normal period lengths and phases to a significant fraction of peripheral circadian output . However , the generation of robust circadian output ( as measured by the number of cycling genes and their amplitude ) depends on an intact peripheral oscillator . Core clock genes were preferentially rescued by Clock expression in the SCN . We performed DAVID analysis to determine whether rescued genes represented specific pathways or ontologies [53] . We found that core clock genes were the only ontological group significantly enriched in this data set ( n = 92 , enrichment = 0 . 79 , p<0 . 0005 , q<0 . 07 ) . Consistent with this observation , we found that core clock genes were preferentially rescued , even compared to other high-amplitude cycling genes ( amplitude >5 . 0 peak∶trough ) . Of the 11 high-amplitude core clock genes , 9 showed normal rhythms in Clock-rescue mice . In contrast , only 7 of 28 high-amplitude output genes were rescued in these samples . Taken together , we conclude that key components of the circadian clock are sensitive to either direct or indirect signals from the SCN , even in the absence of a functional local circadian clock . Taking this line of investigation one step further , a direct comparison between our 92 rescued circadian genes and the 31 system-driven genes identified by Kornmann et al . [21] is of obvious importance . Surprisingly , there was very little overlap between these two data sets ( Table 2 ) . Only three genes were common to both sets: Per2 ( Figure 3D ) , Nocturnin ( Ccrn41 ) ( Figure 5A ) , and Fbxo21 ( Figure 5B ) . We reasoned that the apparent disagreement between these two data sets may be a consequence of inconsistencies in the underlying statistical analyses . To address this , we re-analyzed Kornmann et al . 's data using JTK_Cycle ( Table S3 ) . At the same statistical threshold used in the present study ( p<0 . 0011 ) , we found only three genes that were systemically-driven ( Ccdc12 , Cry1 , and Dbp ) , none of which were previously identified . In the interest of identifying as many similarities as possible between the present study and Kornmann et al . 's data , we therefore loosened the statistical threshold to p<0 . 1 , which corresponds to a FDR of q<0 . 41 in the wildtype ( non-Reverb-alpha over-expresssing ) condition . At this confidence level , we found 47 unique genes that were systemically-driven ( Table S3 ) , including five that overlap with Kornmann et al . ( 1200016E24Rik , 4833417J20Rik , Fus , Hsap1b , Tuba4 ) and four that overlap with the present study ( Alas1 , Cabc1 , Dbp , and Nr1d2 ) . Therefore , the apparent disagreement between the present study and Kornmann et al . can be partially reconciled by standardizing the statistical analyses . For example , cold inducible RNA binding protein ( Cirbp ) was identified and validated by Kornmann et al . as a system-driven gene . Although we do not detect it as being rhythmic in the Clock-rescue condition , if we loosen our statistical threshold , it is clearly rhythmic in both wildtype and Clock-rescue ( Figure 5C ) . Further liberalizing statistical criteria can identify additional similarities between these studies , albeit at the expense of considerably more false-discoveries . Nevertheless , some genes are clearly divergent between these datasets , such as Hspca which is system-driven in Kornmann et al . but arrhythmic in our data ( Figure 5D ) . We speculate that these differences are due to the different genetic manipulations used in these studies . Kornmann et al . over-expressed Reverb-alpha to systematically inhibit Bmal1-mediated transcription , as well as every other Reverb-alpha target gene . Similarly , our study used ClockΔ19 as a dominant mutant to knock-down Bmal1/Clock activity . However , ClockΔ19 is not expected to dramatically affect Reverb-alpha target genes , which is supported by the presence of 10 Reverb-alpha targets among the 92 rescued circadian genes ( Table 2 ) [50] . Moreover , ClockΔ19 is insufficient to abolish all circadian molecular oscillations , as evidenced by the weak , long-period transcriptional rhythms seen in Clock-mutant mice . For both these reasons , we expected to see rescued transcriptional rhythms not previously seen in Kornmann et al . These rescued transcriptional rhythms could be bona fide system-driven genes , or downstream genes driven by the residual activity of the molecular oscillator in ClockΔ19 , or some combination thereof . Two candidate system-driven genes are shown in Figure 5E and 5F . Aminolevulinic acid synthase 1 ( Alas1 ) was not identified by Kornmann et al . , but was detected as system-driven in our re-analysis of their dataset ( Table S3 ) . It oscillates with normal period and phase in both wildtype and Clock-rescue , and has an amplitude approximately the same in both genotypes/treatments , as would be expected from a system-driven gene ( Figure 4E ) . Interestingly , Alas1 forms a junction between the circadian clock and heme bioactivity . It is the rate-limiting enzyme in heme biosynthesis and is regulated by Npas2 . Additionally , it regulates the activity of Bmal1/Npas2 , ultimately affecting the expression of Per1 and Per2 [54] , thereby making it a strong candidate for conveying system-driven cues into the peripheral circadian clock . Similarly , major facilitator superfamily domain containing 2A ( Mfsd2 ) is rhythmic in Clock-rescue animals ( Figure 5F ) . Like other potential system-driven genes ( e . g . Per2 and Nocturnin ) , Mfsd2 expression is largely unchanged between wildtype and Clock-rescue animals ( Figure 4E ) . Interestingly , Mfsd2 is highly induced in liver and brown fat by fasting and cold-induced thermogenesis [55] , consistent with ( and a possible molecular mechanism for ) Kornmann et al . 's hypothesis that temperature is a major entrainer of peripheral circadian clocks . At the least , Mfsd2 is an excellent candidate for conveying nutritional signals to the liver clock . In addition to 24-hour transcriptional rhythms , the liver and other tissues express ultradian rhythms with period lengths of 12- and 8- hours [19] . A recent study has demonstrated that at least some of these circadian ‘harmonics’ are disorganized in mice with genetically-disrupted circadian oscillations [56] . However , the extent to which these rhythms are driven by central versus peripheral oscillators is unclear . To address this , we examined the transcriptional profiles of circadian harmonics in wildtype , Clock-rescue , and Clock-defective mice . For example , Creld2 was previously identified as a 12-hour oscillator [19] , and re-confirmed by this study ( Figure 6A ) . This ultradian oscillator reverted to a 24-hour period length in the Clock-rescue mice ( Figure 6B ) , and became disorganized with lower overall expression in Clock-defective mice ( Figure 6C ) . This phenotype is consistent with the other 12-hour rhythms detected in these data ( N = 3 , Table 3 ) , as well as previously identified 12-hour cyclers [19] with some evidence of oscillatory behavior ( period = ∼12-hours AND p<0 . 1 ) in the present data set . In the every case , 12-hour oscillations revert to 24-hour period lengths in Clock-rescue mice ( Figure 6D and 6E , and Table 3 ) . Interestingly , the disorganization of these rhythms in Clock-defective mice is consistent between genes ( Figure 6F ) , strongly suggesting that their promoters share common transcriptional inputs . Moreover , the amplitudes of normal 12-hour rhythms and rescued 24-hour rhythms are largely indistinguishable ( Figure 7A ) , indicating that at least one of the two daily peaks of expression in wildtype is largely driven by systemic cues and not the local , peripheral oscillator . Likewise , the phases of the normal and rescued rhythms largely fall into a single cluster ( Figure 7B ) , consistent with the idea that they are responding to identical circulating cues . Given the unambiguous defect in generating 12-hour rhythms in Clock-rescue animals , we conclude that the local , peripheral oscillator is absolutely required for generating 12-hour transcriptional rhythms . However , the appearance of 24-hours in the absence of an intact liver clock suggests that half the peaks of these oscillations are derived from circulating cues downstream of the central oscillator in the SCN .
The circadian regulatory network is organized in a hierarchical fashion with signals originating in the SCN orchestrating rhythms in peripheral tissues . Understanding the nature of these signals and how they are communicated to peripheral tissues is a significant challenge for the field [57] . Here , we have used the tet-OFF expression system to rescue normal circadian function in the brains of otherwise circadian-defective mice . This approach allowed us to dissect aspects of peripheral rhythms that depend on central oscillator function from those that depend on local circadian oscillators . Using microarray analysis to characterize the circadian transcriptome of wildtype , Clock-rescue , and Clock-mutant mice , we found that brain-specific rescue of Clock partially restores circadian transcriptional output in the liver . The number of cycling genes was significantly higher in Clock-rescue mice compared to Clock-defective mice ( Figure 2B ) . Likewise , the confidence with which core clock genes were identified as cycling was greater in the rescue animals ( Figure 2D ) . Both period length ( Figure 3 ) and phase ( Figure 4 ) of transcriptional rhythms in the liver of Clock-rescue mice were consistent with their wildtype counterparts . In contrast , rhythms in Clock-defective mice were out-of-phase with wildtype and had significantly longer period lengths . These results agree with studies of transplanted fibroblasts that have been shown to synchronize their local circadian oscillators to the SCN of their host [58] . Taken as whole , these data indicate that signaling downstream of the central circadian oscillator is sufficient to reset the period and phase of many peripheral transcriptional rhythms in tissues that otherwise have disrupted circadian oscillations . Nevertheless , the total amount of circadian transcriptional output ( as measured by the number of cycling genes and their amplitudes ) was still significantly less than wildtype ( see Figure 2B and 2E , Figure 3D and 3E , and Figure 4 ) . This observation is consistent with a recent study by Kornmann et al . that profiled the liver circadian transcriptome in ad libitum fed mice with disrupted circadian oscillations in the liver due to the over-expression of Rev-erb-alpha [21] . In that study , ∼10% of cycling transcripts ( 31 of 351 ) continued to oscillate with normal periods and phases independently of the genetic manipulation to their peripheral clock , although the amplitudes of these rhythms were generally diminished . Similarly , we saw 16 . 5% ( 952 of 576 ) of genes had restored oscillations in Clock-rescue animals at a threshold of p<0 . 0011 , corresponding to a FDR of q<0 . 05 ( Figure 2C ) . Unlike the Kornmann et al . study , however , we detected 24-hour oscillations of many key circadian genes in Clock-rescue animals , including Nr1d1 , Arntl , Per1/Per2/Per3 , and Dbp/Tef/Hlf ( Table 2 ) . This disagreement between the present study and Kornmann et al . may not be as dramatic as it initially appears . When re-analzying the Kornmann et al . dataset with JTK_Cycle , we found Cry1 , Dbp , and Nr1d2 to all be system-driven . We therefore conclude that 24-hour periodicity in many components of the core circadian clock can be driven by oscillations in the SCN . The extent to which this rescue of peripheral clock periodicity is due to restored behavioral rhythmicity , humoral cues , or their interaction is unknown . We noticed that core clock genes were preferentially rescued by restoring CNS clock function . This held true even when compared with other high amplitude oscillating genes in the liver . We speculate that the promoters of clock genes may have evolved configurations of response elements that are particularly sensitive to humoral and behavioral cues . One can imagine that this property – sensitivity to the CNS clock – would be particularly advantageous for resetting to different light schedules . Core clock gene action , subsequently , would then synchronize tissue specific peripheral gene expression appropriately . At a broader level , we discovered 92 circadian genes ( including clock factors ) that have normal phases and periodicity in Clock-rescue animals . Since the rescue of Clock expression is largely confined to the brain , this restoration of normal phase and period is may be due to direct signaling from humoral cues emerging from the SCN , or alternatively due to indirect signaling , via the SCN's regulation of locomotor and feeding rhythms . In either case , a subset of rescued transcriptional rhythms is expected to be system-driven . Kornmann et al . have previously identified 31 system-driven circadian genes by over-expressing Reverb-alpha in the liver [21] . Surprisingly , there were only three genes in common between their dataset and the present study . Even when using standardized statistical methods , only 7 of the 92 rescued genes identified herein were found in common with Kornmann et al . This is not entirely unexpected given the difference in the genetic manipulations used . For example , over-expressing Reverb-alpha would be expected to inhibit the expression of any genes under the control of ROR elements . Consistent with this idea , we found 10 rescued genes not previously identified by Kornmann et al . that are known targets of Reverb-alpha [50] . The genetic lesion used in this study , ClockΔ19 , significantly diminishes the number of rhythmic genes as well as their amplitudes . However , some rhythmicity still persists , as evidenced by the weak , long-period rhythms still seen in Clock-mutant animals ( Figure 2 ) . At a mechanistic level , this implies that rescued genes in our dataset may be bona fide system-driven genes or peripheral clock-driven genes that are insensitive to CLOCK signaling . Consequently , strong candidates for system-driven genes such as Per2 and Nocturnin ( identified by both Kornmann et al . and the present study ) have amplitudes that are largely independent of genetic manipulation ( Figure 4E ) . Based on this criteria , we identified two novel system-driven candidates: Alas1 and Mfsd2 . Alas1 regulates and is regulated in turn by clock factors , while providing a link between the clock and heme biosynthesis [54] . Similarly , Mfsd2 is known to be driven by the clock , and is upregulated by fasting and cold-induced thermogenesis [55] . Both genes are thus potential nodes through which systemic cues synchronize and drive peripheral circadian rhythms . Both our study and Kornmann et al . agree that the number of system-driven circadian genes may be relatively high . Even accounting for the measurable false discovery rate of both studies , there are likely several dozen and potentially as many as a hundred cycling genes in the liver than can be driven in part by systemic cues . We speculate that this may be an evolutionary strategy to tightly link peripheral clocks with the physiological status of the animal . Unlike the SCN , peripheral clocks ( especially the liver ) are sensitive to a wide array of behavioral , environmental , and physiological stimuli . By having many different input pathways to synchronize liver rhythms , evolution may have built considerable redundancy into the system . This is analogous to the coupling of SCN neurons , which renders the entire timekeeping mechanism considerably more resistant to perturbation [59] , [60] . Similarly to 24-hour rhythms , circadian harmonics ( i . e . , 12- and 8-hour rhythms ) , are transcriptional oscillations found in tissues throughout the body that persist in constant darkness [19] , [56] . These rhythms have been confirmed to exist at the protein level , and may be a consequence of rhythms of lipid metabolism and ER stress [56] . Consistent with this observation , 12-hour transcriptional rhythms are disrupted in mice subjected to restricted feeding [19] . Nevertheless , the extent to which central and peripheral oscillators contribute to circadian harmonics is an open question . We found that disrupting Clock function throughout the body disorganized and diminished 12-hour rhythms , indicating that these rhythms derive ( at least in part ) from the conventional 24-hour circadian clock ( Figure 5 and Table 3 ) . Interestingly , brain-specific rescue of Clock function restored 24- but not 12-hour rhythmicity to these genes , with no discernible loss of amplitude ( Figure 6 , Figure 7A , and Table 3 ) . This observation is reminiscent of previous studies that demonstrated that both restricted feeding and Hlf/Dbp/Tef loss-of-function convert 12-hour rhythms to 24-hour period lengths [19] , [56] . Moreover , the phase of 12-hour transcription is precisely the same regardless of which tissues are examined , suggesting a common signaling origin [19] . Since 24-hour periodicity of most clock genes is restored in the livers of Clock-rescue mice ( albeit at consistently low amplitudes ) , we acknowledge that the rescue of 12-hour rhythms may be downstream of local oscillations in the liver . Nevertheless , we favor the explanation that circulating , tissue-non-autonomous signaling cues drive one of the two daily peaks of 12-hour transcriptional rhythms . Two observations support this interpretation . First , the amplitude of rescued circadian harmonics is largely the same as in wildtype ( Figure 7B ) . In contrast , most core clock genes and key circadian outputs have low-amplitude oscillations in the rescued livers ( Figure 4E and Table 2 ) . We consider it more likely that systemic cues downstream of fully rescued SCN and the resulting behavioral rhythms drive these rescued harmonic oscillations rather than low-amplitude oscillations of clock genes in the liver . Second , previous studies in dissociated cells with normal molecular oscillations have consistently failed to detect harmonic transcriptional rhythms [19] , [46] . Although these studies do not speak to the necessity of peripheral circadian clocks for generating 12-hour rhythms , they do indicate that cell-autonomous circadian oscillations are not sufficient for generating circadian harmonics . Regardless , based on the present study we can conclude for the first time that the conventional 24-hour circadian oscillator ( whether central or peripheral ) is the ultimate origin of harmonic transcriptional rhythms , and circulating cues downstream of the SCN are sufficient to restore one of the two daily peaks of expression .
Mice were housed in light-tight boxes and entrained to a 12 hour light , 12 hour dark schedule for one week before being switched to complete darkness . Wildtype mice ( C57BL/6J ) were acquired from Jackson Labs; experimental animals ( Scg2::tTA; tetO::Clockwt; ClockΔ19/Δ19 ) were generated as previously described [36] , [40] . The Scg2::tTA mice were congenic on a C57BL/6J background , and the tetO::Clockwt transgenic mice were co-isogenic on a C57BL/6J background . Mice were supplied with regular food and water ad libitum . Three days before the final lights off ( at ZT12 ) , experimental animals were treated with either water or doxycycline ( 30 ug/ml; Sigma-Aldrich ) supplied in the drinking water . Starting two days after the first day in DD ( i . e . , at CT48 ) , four wildtype mice ( two females and two males ) and Clock-rescue mice ( one female and one male ) were sacrificed in the dark every two hours . Liver samples were quickly excised and snap-frozen in liquid nitrogen . Liver samples were homogenized in Trizol ( Invitrogen ) and total RNA was purified using the manufacturer's protocol . All animal experiments were performed with the approval of the Committee on Animal Care and Use at Northwestern University . Mice were placed in individual running wheel cages and activity was recorded and analyzed using the ClockLab Data Collection System ( Actimetrics , Wilmette , IL ) . Mice were entrained to a 12-hour light/12-hour dark cycle ( LD ) for a minimum of 10 days before they were released into constant darkness ( DD ) . Doxycycline was supplied in the drinking water at a concentration of 10 µg/ml for the behavioral analysis as described previously [36] , [40] . Mice were supplied with regular food and water with or without doxycycline ad libitum . Dox-containing water was renewed every 2–3 days . RNA isolated from the liver were reverse transcribed with SuperScript First-Strand Synthesis System for RT-PCR ( Invitrogen ) . cDNA ( 1 . 25 µl ) was pre-amplified for specific target amplification ( STA ) using pooled 0 . 2× TaqMan Gene Expression Assays mix ( Applied Biosystems ) . The thermal cycling conditions used for STA were 95°C hold for 10 minutes followed by 14 cycles of 95°C for 15 seconds and 60°C for 4 minutes . TaqMan Gene Expression Assays ( Applied Biosystems ) used in this study are listed in the supplemental data ( Table S4 ) . Preamplified cDNA was diluted with TE buffer ( 1∶5 ) and qPCR assays were performed using the BioMark 48 . 48 Dynamic Array as specified ( Fluidigm ) . Assays were run in triplicate on each array and data were analyzed by the use of BioMark Real-Time PCR Analysis Software Version 2 . 0 ( Fluidigm ) to obtain Ct and ΔΔCt values . Quantitative PCR analysis ( Figure S1 ) was performed from the same RNA samples as those used for the microarray analysis ( see below ) . 5 µg total RNA per time point was submitted to the University of Pennsylvania School of Medicine Microarray Facility for labeling and hybridization to Affymetrix Mouse Exon 1 . 0 ST Arrays . Expression values were extracted using RMA implemented in Expression Console ( Affymetrix ) at the core gene level . JTK_Cycle implemented in R ( version 2 . 12 . 1 , 64-bit ) was used to detect cycling genes as previously described [37] , using a period length window of 10–40 hours . Due to the lower sampling resolution , re-analysis of the Kornmann et al . 2009 dataset was performed using a 24 hour period length window . Raw data and statistics were compiled into an Access database ( Microsoft ) . Heatmaps were generated using custom scripts implemented in MATLAB ( Mathworks , version R2010b ) . DAVID analysis was performed as previously described [53] , using all rhythmic genes in wildtype ( N = 576 , q<0 . 05 , p<0 . 0011 ) as a background list , and all 24-hour rescued genes ( n = 92 ) as the principal gene list . Amplitude estimates were made using JTK_Cycle , modified as previously described ( ( 2*JTK . AMP ) + ( percentile ( array , 0 . 1 ) / ( percentile ( array , 0 . 1 ) ) [61] All . CEL files are available from GEO ( accession number: GSE30411 ) ; custom scripts for MATLAB and R are available on demand . | Circadian rhythms confer adaptive advantage by allowing organisms to anticipate daily rhythms in their environment , such as light and temperature cycles . In mammals , the central circadian time-keeping mechanism resides within the hypothalamus , integrating information from the retina and synchronizing circadian oscillations in peripheral tissues , such as the liver . Nevertheless , the nature of this regulatory relationship is not completely understood . Here we use mice with disrupted circadian rhythmicity stemming from a mutation in Clock , a key component of the circadian time-keeping mechanism . Using recently developed transgenic technology , we expressed normal CLOCK protein in the brain , resulting in mice with normal brain rhythmicity , but defective circadian rhythmicity everywhere else . Using microarrays to detect transcriptional rhythms in the livers of these mice , we found that 95 genes showed normal oscillations ( albeit with generally lower amplitudes ) , even in the presence of a defective liver circadian clock . This observation suggests that many rhythmic genes in peripheral tissues are driven directly from the brain . Finally , we showed that harmonics of circadian oscillations , e . g . 12-hour rhythms , revert to 24-hour rhythms in these mice , indicating that both liver and brain-derived cues drive different aspects of these rhythms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"systems",
"biology",
"biology",
"genomics",
"neuroscience",
"genetics",
"and",
"genomics"
]
| 2012 | Brain-Specific Rescue of Clock Reveals System-Driven Transcriptional Rhythms in Peripheral Tissue |
There is increasing evidence for a connection between DNA replication and the expression of adjacent genes . Therefore , this study addressed the question of whether a herpesvirus origin of replication can be used to activate or increase the expression of adjacent genes . Cell lines carrying an episomal vector , in which reporter genes are linked to the murine cytomegalovirus ( MCMV ) origin of lytic replication ( oriLyt ) , were constructed . Reporter gene expression was silenced by a histone-deacetylase-dependent mechanism , but was resolved upon lytic infection with MCMV . Replication of the episome was observed subsequent to infection , leading to the induction of gene expression by more than 1000-fold . oriLyt-based regulation thus provided a unique opportunity for virus-induced conditional gene expression without the need for an additional induction mechanism . This principle was exploited to show effective late trans-complementation of the toxic viral protein M50 and the glycoprotein gO of MCMV . Moreover , the application of this principle for intracellular immunization against herpesvirus infection was demonstrated . The results of the present study show that viral infection specifically activated the expression of a dominant-negative transgene , which inhibited viral growth . This conditional system was operative in explant cultures of transgenic mice , but not in vivo . Several applications are discussed .
Herpesviruses are among the most complex DNA viruses , encoding up to 200 genes . The characteristic temporal control of viral gene expression ensures the appropriate regulation of the morphogenic process . Transcription factors , activated immediately upon virion entry into cells or after reactivation of latent viruses , regulate the expression of early genes and start a cascade of transcription of early-late , leaky-late and true-late genes [1] . Remarkably , in contrast to leaky-late genes , expression of true-late genes begins only after viral DNA replication , and can be blocked by inhibitors of the viral DNA polymerase such as phosphonoacetic acid ( PAA ) [2] . As the expression of true-late genes depends upon viral replication , the promoters of late genes were analyzed in the context of origin of lytic replication ( oriLyt ) sequences . Notably , isolated true-late promoters encoded by plasmids or integrated into the cellular genome respond like early promoters upon infection . Yet , the same promoters , together with an oriLyt provided in cis , restore the typical late expression patterns . This property was first demonstrated for HSV-1 [3] , but was confirmed for several other herpesviruses [4] , [5] . Six conserved replication proteins comprise the core components of lytic DNA replication and are shared by all herpesviruses [6] . To a certain degree , the core replication machinery can be complemented by another herpesvirus [7] . Yet , closely-related cytomegaloviruses ( CMV ) oriLyts are not exchangeable; the human CMV oriLyt cannot be activated by murine CMV or simian CMV [8] , [9] . Two models have been proposed to explain the mechanism of lytic DNA replication . One includes circularization of the parental linear viral genome in the nucleus followed by an initial bidirectional theta-replication followed by a rolling circle mechanism [10] . An alternative model suggests a biphasic process in which DNA replication is initially induced by an origin-binding protein , which then recruits the replication machinery to the oriLyt sequences [11] . Later , it switches to a recombination-dependent replication and/or rolling circle mechanism [12] , in which D-loop formation contributes to replication initiation [13] . This model is reflected by the highly branched concatemeric genomes observed in herpesvirus DNA replication . Lytic replication origins have been identified by transient plasmid replication assays [6] . In these studies , cells transfected with plasmids containing putative oriLyt sequences were infected with the respective viruses . Under these transient conditions , plasmids containing an oriLyt were replicated by the viral machinery provided in trans . Stable maintenance of oriLyt-containing plasmids has been described for herpesvirus amplicon vectors . Herpes viral amplicon vectors contain an oriLyt , a gene ( or genes ) and , importantly , packaging signals for vector DNA incorporation into herpesvirus virions [14] . Standard amplicon vectors are diluted in dividing cells , yet hybrid vectors containing additional elements such as EBNA1/oriP or surface/matrix attachment region ( S/MAR ) sequences enhance stability . While expression of transgenes delivered to target cells has been analyzed in several studies , the potential impact of the presence of the oriLyt on transgene expression has not been addressed . Conditional gene expression systems are important tools in virology and help to avoid toxicity and related problems . Inducible systems are usually based on additional elements that provide regulatory factors such as the Tet-system , or on recombination-dependent gene expression , such as the Cre/loxP mediated system . It is important to note that many viral promoters can also be trans-activated upon virus infection; however , their induction usually follows early or early-late kinetics and ranges only in the linear scale [15] and , thus , compares poorly with the viral expression levels and kinetics shown by late genes . Therefore , the aim of the present study was to construct an inducible expression system , which is stably maintained in the host cell and allows strong induction of late genes . Although replication of a DNA fragment containing a lytic origin of HSV-1 ( oriS ) was initiated by superinfection [16] , other studies show contradictory results [17] , suggesting effects due to the host chromatin environment . To avoid this , we used the pEPI-1 vector as a backbone . The pEPI vectors are non-viral episomes that are maintained by an S/MAR element and replicate semi-conservatively once per cell cycle . These vectors can be maintained in many cell types without selection [18] , [19] . Also , maintenance of the episome in mouse cell lines did not prevent silencing of the encoded transgene [20] . Here , we show that transcription units that are combined with an MCMV-oriLyt encoded on the pEPI background are indeed silenced . Yet , specific induction of DNA replication and gene expression both occurred after viral infection . Most notably , this system proved to be very useful for protein trans-complementation of defective viruses and studies of dominant-negative ( DN ) viral function .
A recurrent problem in transgenesis is the phenomenon of gene silencing [21] . Yet , gene silencing might be advantageous if constitutive gene expression is not favored . To obtain a gene expression system that is induced by infection by specific viruses , we relied on the fact that herpesviruses activate gene expression in trans , interfere with the host silencing machinery [22] , [23] , and will replicate constructs containing an oriLyt [9] , [24] . Therefore , we established a novel CMV-inducible gene expression system in which natural silencing of the gene of interest in the transgenic cell is obtained . DNA replication after CMV infection then liberates the transgene , leading to the powerful induction of expression ( Supp . Fig . S1C ) . MCMV was chosen because it is well characterized and facilitates in vivo investigation . Furthermore , the position of the minimal oriLyt of MCMV has already been defined [9] . It is located within a complex and highly structured locus containing several palindromes , inverted and direct repeats , and transcription factor binding sites . It has been shown for other herpesviruses that additional sequences flanking the minimal oriLyt can enhance replication; therefore , we decided to clone a 3 . 9-kb fragment containing the minimal 1 . 7-kb oriLyt fragment and adjacent sequences . Due to its size and the presence of various repeats , PCR amplification of the template was not feasible . So , we cloned the oriLyt fragment using a new BAC-based pick-up strategy ( Supp . Fig . S1A ) and inserted it into the episomal vector , pEpibo . To test the replicon vector , we used firefly luciferase ( FL ) as a reporter and measured the induction of gene expression in response to oriLyt activation by creating the vector , pEpibo-luc-ori ( Supp . Fig . S1B ) . Here , FL expression was driven by the minimal SV40 promoter , which is not induced upon infection with MCMV ( Supp . Fig . S2 ) . To characterize the gene expression driven by pEpibo-luc-ori , two independent pools of NIH3T3 transfectants were generated . In the absence of infection , FL expression decreased to the limit of detection after 16 weeks in both pools ( luc-ori-t1 and -t2 ) . However , infection with MCMV restored FL expression by 3 to 4 orders of magnitude in both transgenic cell pools , irrespectively of the treatment time ( Fig . 1A ) . Thus , loss of FL expression due to silencing was recovered upon virus infection . There are several different mechanisms of transgene silencing [21] . We wanted to determine whether the construct was inactivated by de novo methylation of the promoter sequences , or whether chromatin condensation was the main contributor . We added specific inhibitors to cell clones derived from the cell pool , luc-ori t1 , and measured FL expression after 36 h . Three sub-clones ( cl . 1 , cl . 2 , cl . 3 ) showed a basal level of FL expression and one clone was negative ( cl . 4 ) before treatment . Inhibition of CpG-methylation by 25 µM 5′-aza-cytidine [21] did not enhance expression , but treatment with 330 nM trichostatin A ( TSA ) resulted in about 100-fold induction in cl . 1 , cl . 2 and cl . 3 ( Fig . 1B ) . Efficiency of 5′ aza-cytidine treatment was controlled by recovery of methylation-dependent de-silencing of GFP fluorescence ( Supp . Fig S3A ) . Recovery of FL induction by TSA was dose-dependent , but achieved induction levels lower than those obtained by infection with MCMV ( Supp . Fig S3B ) . Furthermore , we found only limited co-operativity of TSA and MCMV infection on the induction of FL expression ( Supp . Fig S4B ) . Western Blot analysis of MCMV infected cells treated with TSA revealed an enhancement of immediate early gene expression as previously described [25] , but it disturbed gene expression at late time points ( Supp . Fig S4A ) . Accordingly , 330 nM TSA inhibited MCMV production by 100-fold in a multi-step growth curve analysis ( Supp . Fig S4B ) . Thus , MCMV infection could enhance FL signals induced by TSA , but was inhibited itself with increasing TSA concentrations . Consequently , the highest FL induction was achieved in infected cells without TSA treatment ( Supp . Fig S4C ) . Interestingly , FL expression by clone cl . 4 was not detectable under any conditions; this may have been due to the lack of an intact pEpibo-luc-ori episome or integration . As TSA is a potent inhibitor of histone deacetylases ( HDACs ) [26] , these results strongly suggest that pEpibo-luc-ori driven gene expression is constrained by histone modifications other than DNA methylation . We then asked whether virus specificity , namely infection with MCMV , is necessary to activate FL from the replicon vector . As the core replication machinery is conserved in herpesviruses , we determined whether a herpesvirus from another subfamily could induce gene expression by the MCMV oriLyt system . As NIH3T3 cells are also permissive for the MHV68 γ-herpesvirus , which induces lytic replication in almost all infected mouse cells , we examined infection of an isolated cell clone , luc-ori cl . 1 , with MHV68 and MCMV ( Fig . 2A ) . Whereas infection with MCMV resulted in typically high induction of FL expression ( by three orders of magnitude ) , infection with MHV68 resulted only in a marginal increase of ( ∼5–10-fold ) . Because only MCMV was able to fully activate expression of the MCMV-oriLyt-replicon , the induction of the replicon system appears to be specific for infection with the corresponding herpesvirus . In view of the fact that silencing of pEpibo-luc-ori was lifted upon MCMV infection , we analyzed the role of the oriLyt element . To this end we constructed pEpibo-luc , which lacks the oriLyt sequence but is otherwise identical to pEpibo-luc-ori . NIH3T3 cells transfected with the pEpibo-luc generated cell pool , luc t1 , and NIH3t3 cells transfected with pEpibo-luc-ori generated cell pool , luc-ori t3 ( Fig . 2B ) . Silencing of constitutive FL expression was observed in both cell lines over time ( data not shown ) . Yet , infection with MCMV enhanced the reporter signal selectively in the oriLyt-containing cell line , luc-ori t3 , but not in the oriLyt negative cell line , luc-t1 . It is important to note that oriLyt is located downstream of the multiple cloning sites into which the transgene/FL was inserted; thus , transgene transcription cannot be activated by means of cryptic promoter regions . This confirms the key role played by the oriLyt element in MCMV infection-dependent activation of FL expression in the luc-ori cell lines . Next , we examined whether induction of the luc gene in pEpibo-luc-ori requires DNA replication of the infecting MCMV . Cell clones , luc-ori cl . 1–cl . 4 were infected with MCMV in the presence or absence of PAA , a specific inhibitor of the viral polymerase [27] . In the absence of PAA , FL expression in cell clones luc-ori cl . 1–cl . 3 increased by 100–1000-fold upon infection with MCMV ( Fig . 2C ) . However , in presence of PAA , FL expression was similar to that in uninfected cells . PAA alone , in absence of infection , had no influence on FL expression . These data clearly show that induction of pEpibo-luc-ori expression upon MCMV infection was inhibited by PAA . Thus , MCMV-induced de-silencing and induction of FL was dependent on viral DNA replication . Cell line luc-ori cl . 4 did not respond to any treatment as already seen in the TSA experiments . Induction of FL in the controls began at the onset of DNA replication and increased exponentially over time ( Fig . 2D; dark hatched columns ) . Notably , PAA was only able to inhibit induction of the oriLyt system when added prior to the onset of DNA replication , which starts around 8–12 h p . i . in MCMV [28] , [29] ( Fig . 2D ) . If PAA was added after the initiation of replication , its effect on FL expression levels waned , further supporting the contribution of viral DNA replication . FL induction after infection proved to be a very robust property of the cell lines , as the level of induction was maintained in luc-ori cl . 1 cells even after more than 6 months of continuous culture ( data not shown ) . Because induction of reporter gene expression from the vector was dependent on the oriLyt sequence ( Fig . 2B ) , quantitative PCR ( qPCR ) experiments were performed to determine whether the episomal vector was amplified during infection . We measured the relative pEpibo-luc-ori copy numbers in infected and uninfected cells normalized to the endogenous murine lbr gene . Induction of FL expression ( Fig . 3A ) correlated with a significant increase in the pEpibo-luc-ori copy number ( about 50-fold; Fig . 3B ) . Notably , in the cell line , luc-ori cl . 4 , in which MCMV infection did not de-silence FL expression , there was no increase in the copy number of the replicon vector sequence ( Supp Fig . S5 ) . Phosphonoformic acid ( PF ) , like PAA , is a specific inhibitor of herpesvirus DNA polymerases . As expected , no amplification of the vector was detected after PAA or PF treatment of infected luc-ori cl . 1 cells ( Fig . 3B ) , whereas infection of luc-ori cl . 1 cells in the absence of inhibitors resulted in a ∼2500-fold increase in FL induction and a ∼50-fold increase in replicon vector amplification . By contrast , liberation of the vector from silencing by treatment with 300 nM TSA , without additional vector replication , resulted in only a moderate ∼25–30 fold induction of FL in luc-ori cl . 1 cells ( Fig . 3A , B ) . Trans-complementation of late gene defects by cell lines constitutively expressing the gene of interest is sometimes challenging . Incorrect timing , insufficient expression levels , aberrant intracellular distribution of isolated proteins , or toxicity of the constitutively expressed viral transgenes , can complicate cell based trans-complementation systems [30]–[32] . In particular , high constitutive expression of viral glycoproteins is often toxic to cells [33] , [34] . As the replicon vector was maintained with low basal , or even undetectable , levels of transgene expression , we hypothesized that this system may be suitable for trans-complementation of late gene defects . To examine this , we constructed two cell lines , one encoding a glycoprotein ( gO ) and the other encoding the protein M50 , which is known to be toxic after isolated expression [30] . This protein performs an essential function during export of the nascent viral capsid from the nucleus to the cytoplasm . We cloned the m74 gene , coding gO [35] , into the oriLyt vector to generate the responding cell line NIH3T3:gO-ori ( gO-ori ) . MCMV , like HCMV , lacking gO is restricted to focal spreading and the release of infectious virions into the supernatant is hampered [35] . In the cell line containing the gO-ori vector , however , infection with MCMV-ΔgO should lead to pEpibo-gO-ori replication and m74 gene expression; thereby reconstituting release of infectious progeny . MCMV-ΔgO released from NIH3T3 about ∼2–2 . 5 orders of magnitude less virus than MCMV-wt . By contrast , growth in the complementing cell line resulted in comparable titers for both MCMV-ΔgO and MCMV-wt ( Fig . 4A ) . Thus , the virus defect was rescued and the virus was no longer restricted to a focal growth pattern ( Fig . 4B ) . This confirms that the replicon expression system can be used to efficiently produce late viral proteins in trans . To determine whether the viral genome reverts to a wt-like virus due to recombination with the episomal replicon vector , PCR analysis of the m74 gene ( gO ) was performed using viruses harvested on day 5 from the supernatants used in the growth curve experiments . Supernatants were centrifuged to remove cells and the cellular DNA was digested with DNase to discriminate between the m74 gene in the replicon vectors in the cells and the viral genome . The fact that the supernatants were free from cellular debris was confirmed by the lack of the cellular gene lbr . Presence of viral genomic DNA was confirmed by amplifying the viral DNA polymerase , M54 . Whereas all supernatants were positive for the M54 gene and negative for the lbr gene , m74 was detected only in MCMV-wt , and not in MCMVΔgO , regardless of whether it originated from NIH3T3 or gO-ori cells ( Fig . 4C ) . Furthermore , we never observed any reversion of the phenotypic restriction of MCMVΔgO to cell-to-cell spreading ( data not shown ) . No recombination was detected between the replicon vector and the MCMVΔgO virus . To assess the suitability of the system for complementing a toxic protein , the M50 gene ( including a C-terminal HA-tag ) was cloned into the oriLyt vector to generate the cell line , NIH3T3:M50-ori ( M50-ori ) . Previous attempts at generating trans-complementing M50 cell lines using conventional methods failed due to the toxicity of the M50 protein [30] . To assess M50HA expression in two M50-ori cell pools ( M50-ori t1 and t2 ) , Western blot analysis was performed using an HA- specific antibody . Protein loading was controlled by actin detection , and infection was controlled by detecting the viral immediate-early proteins , IE1/IE3 . No M50HA protein was detectable in uninfected cells; however , infection of the cell pools resulted in de-silencing and strong expression of M50HA ( Fig . 4D ) . To analyze trans-complementation and recombination , virus reconstitution from MCMV-BAC DNA lacking the essential M50 gene was performed . To this end , M50-ori t1 and M50-ori t2 , as well as NIH3T3 cells and MEF cells , were transfected with the BAC pSM3fr-Δ1-16-ΔM50-F . This BAC harbors the deletion of most of the M50 ORF , however due to overlap with the M49 gene 75 aa of the C-terminus of M50 had to be left intact and is homologous to the sequence in the replicon vector . Plaque formation was detected 3 days post-transfection in both M50-ori pools ( data not shown ) , but not in NIH3T3 cells or MEF cells . Full cell lysis occurred 5 days post-transfection in the M50-ori cell pools , whereas no viral progeny were detected in the cells lacking the vector . To assess the efficiency of the reconstitution , serially diluted supernatants were used to infect the cloned cell line M50-ori cl . 2 . 1 , or NIH3T3 cells . Titration of the supernatant on M50-ori cl . 2 . 1 cells harboring the M50-ori vector revealed high titers ( ∼2×107 PFU/ml ) of the trans-complemented MCMV-ΔM50-F/M50HA . The reconstituted virus caused plaque formation and virus spread on M50-ori cl . 2 . 1 cells in a concentration-dependent manner , whereas only the signal from first-infected cells was seen in non-complementing NIH3T3 cells , without spread or plaque formation ( Fig . 4E ) . Thus , the replicon vector allowed efficient trans-complementation of the toxic essential late protein M50 . This complementation occurred at the protein level since the virus was unable to spread in non-complementing cells . However , we found that infection with MCMVΔM50-F/M50HA at a very high MOI resulted in formation of a few plaques , even in MEF cells , indicative of some genetic reversion . In the M50-ori cell pools , we calculated the reversion rate to be about 1 out of 104–105 PFU ( data not shown ) . Next , we studied whether virus replication can be inhibited by a DN transgene induced late in the viral cascade . Inhibition of viral spread by the use of DN viral proteins , called intracellular immunization , was proposed by Baltimore in 1988 [36] and was inspired by the work of Friedman and colleagues [37] . They provided the proof-of-principle experiment showing that a truncated VP16 protein could reduce the replication of HSV-1 when stably expressed by the host cell line . Since then , toxicity caused by DN transgenes has been a problem [37]–[40] . To test the usefulness of our system for studying dominant-negative effects , we cloned the DN gfpscp gene [41] , [42] , which codes for a fusion between GFP and the small capsid protein ( SCP ) of MCMV , into the oriLyt vector . During infection of the NIH3T3:gfpscp-ori ( gfpscp-ori ) cell line , MCMV should de-silence expression of the inhibitory protein GFPSCP . Strong expression of GFPSCP should , in turn , block the egress of viral capsids from the nucleus ( Fig . 5A ) . To examine the inhibitory potential of the gfpscp-ori cell line , we performed growth curve analysis of MCMV in the inhibitory cell line and NIH3T3 cells . In contrast to DN expression in cis in the viral genome ( which was about 1 million-fold reduced ) [42] , induced expression of the DN in trans resulted in inhibition of virus production by about 2 orders of magnitude only ( Fig . 5B ) . This was also true for other cell lines such as M210-B4 ( data not shown ) . However , complete inhibition could not be achieved , perhaps due to the fact that SCP represents the most abundant protein in viral capsids and , therefore , it is unlikely to be out-competed by the DN protein . We then considered heterogenous de-silencing of the transgene in individual cells , either along with the cell cycle status or due to other reasons , which may result in cell subpopulations in which DN expression is absent . To address this possibility , we tested recombinant viruses derived from pSM3fr-Δ1-16-SCPiChe-FRT ( MCMV-cherry ) . Here , the expression of mCherry served as a late infection marker as its expression starts concordant with that of the SCP gene . We observed a striking and specific correlation between infected cells and cells positive for GFPSCP fluorescence ( Fig . 5C ) . Therefore , limited amounts of the DN protein rather than heterogeneous de-silencing and subsequent lack of the inhibitory protein in individual cells , is probably the cause of the limited DN effect . To examine the transgene copy number prior to infection , fluorescence in situ hybridization ( FISH ) was performed . pEPI-1-based cell lines carry between 2–10 episomes per cell . FISH analysis of the gfpscp-ori cell line revealed approximately two episomal copies per cell ( Fig . 5D ) Integration events , as described for the original pEPI-EGFP [43] vector , were also detected . In 34 of the analyzed metaphase spreads , we identified 31 times episomal state of the pEpibo-gfpscp-ori vector . Additional integrated vectors were detected in five metaphase spreads . However , we did not find metaphase spreads associated with integrated vectors alone . As already observed for several attenuated MCMV-mutants , the reduction in virus titer of 99% observed in tissue culture may result in even stronger attenuation if the system is operative in vivo [44] . To test the oriLyt system in vivo , we transferred the conditional DN principle to the natural host of MCMV , the mouse . To our knowledge , no transgenic mouse has been created based on the pEPI-1 vector . Therefore , to assess episomal stability and to test de-silencing in vivo , we generated transgenic mice carrying the pEpibo-luc-ori construct . This should permit the general concept to be studied in vivo ( Fig . 6A ) . To this end , we transfected murine embryonic stem cells ( mES line E14 ) with pEpibo-luc-ori and isolated eight cell clones . Note that mES cells are non-permissive for MCMV infection . To identify cell clones suitable for blastocyst injection , several cell populations were differentiated for 3 weeks to enable productive MCMV infection [45] . To study virus specific de-silencing in differentiated mES , the bioluminescence signal was measured both prior to and after MCMV infection . In three of the eight clones , no FL expression was detected under any conditions ( A10 , B1 , B9 ) ; three other clones showed a weak increase in FL expression after infection ( A2 , A6 , B11 ) and two clones were induced 7- ( A3 ) and 30-fold ( B8 ) ( Fig . 6B ) . The latter two clones ( named line A and line B ) were used for blastocyst injection . This led to at least one chimera for each line , which transmitted the luciferase gene to the next generation ( lines were named VIOLA: virus inducible oriLyt-dependent luciferase animal ) . Basal FL expression and FL induction was assessed by measuring bioluminescence in lung explant cultures . Although the VIOLA-B line was positive for the luc gene , it did not express FL under any conditions ( data not shown ) . Lung explants from 2nd generation VIOLA-A animals expressed FL selectively after MCMV infection , and this property was maintained in subsequent generations ( Fig . 6C ) . Induction of FL expression after MCMV infection was detected in each organ , whereas basal FL expression in uninfected cells was undetectable . FL induction was also observed in ex vivo explant cultures of other organs such as bone marrow , heart , muscle , fat , spleen , and salivary glands of 4th generation animals ( Fig . 6D ) . It is important to note that FL expression in these ex vivo cultures was merely a semi-quantitative estimate , as the explant cultures represent a heterogeneous pool of cells both permissive and non-permissive for MCMV infection . Therefore , although virus load was calculated as MOI = 0 . 5 , viral infection could not be normalized . To test whether virus-induced gene expression also operates in vivo , we infected VIOLA-A mice with 1×106 PFU i . v . and measured the light signals using non-invasive imaging of whole animals over a period of 5 days . As a control , 129X1/SvJ mice were infected with 1×105 PFU i . v . MCMV-luc [46] . Surprisingly , although the luc gene was activated in the explant cultures , no bioluminescence signal was detectable in the VIOLA mice in vivo . This should not be due to limited detection of the signal because the FL signal generated by the virus was easily detectable after infection with MCMV-luc at a 10-fold lower virus dose ( Supp . Fig S6A ) . Furthermore , titration of the virus in organs of the infected VIOLA mice revealed a normal viral load . Therefore a failure to detect the FL signal in vivo is not due to reduced infectibility of VIOLA mice ( Supp . Fig S6B ) . Southern blot hybridization of the transgenic mouse genomic DNA confirmed the presence of pEpibo-luc-ori; however , our data also showed that pEpibo-luc-ori sequences were integrated into the mouse genome ( Supp . Fig S7 ) . Integration of the replicon vector was also suggested by the fact that the inheritance pattern followed classical Mendelian rules . Thus , pEpibo-luc-ori was transferred to the transgenic mice , but the episomal state was not faithfully maintained .
In this study , a herpesvirus lytic origin of replication was combined with the transcription unit from an episomal vector to generate a novel inducible expression system , which is induced by infection with wt virus . In this system , herein referred to as the “replicon vector” , basal background gene expression was negligible as it was silenced by a histone deacetylase-dependent mechanism . However , upon viral infection , de-silencing and activation of the replicon vector led to a >1000-fold increase in induced gene expression . Neither genetic manipulation of the viral genome nor administration of any chemical compound was necessary . This replicon vector system may resolve difficulties encountered when trans-complementing late proteins ( namely glycoproteins and toxic proteins ) , as shown by the trans-complementation of the viral glycoprotein O , and the toxic viral protein M50 . Also , its utility for inhibiting virus multiplication was demonstrated , as shown by the use of DN proteins , such as GFPSCP . Viral DNA replication was a prerequisite for strong reactivation of silenced transgenes as inhibition with PAA or PF abolished reporter gene expression . The presence of oriLyt in the replicon vector was , therefore , essential for induction of gene expression . Based on these data , we assume that oriLyt-mediated vector replication induced by MCMV is the key to gene expression . If there is further need for a true-late viral protein for activation ( whose production would be also inhibited by PAA ) , it must also act in the context of the oriLyt sequence . Induction of the MCMV replicon vector requires specific elements that are shared between the vector and the activating virus , since infection with MHV68 did not induce transgene expression . Although herpesviruses comprise a conserved core of replication proteins that can be exchanged between subfamilies [6] , the activation or origin binding protein/s in β- and γ-herpesviruses is highly diverse between subfamilies and even within subfamilies . Although the mechanism of replication initiation by origin binding proteins has been extensively studied in α-herpesviruses and roseoloviruses , this knowledge is not yet available for other herpesvirus subfamilies . Interestingly , in CMV and γ-herpesviruses , proteins that work as transcriptional activators , such as UL84 for HCMV [47] , Zta for EBV [48] , [49] and K8 and Rta for KSHV [50] , are also necessary for the initiation of viral DNA replication . In these cases , sequences have been identified that might act as promoters [51] , , and a number of transcription factor binding sites were identified in several oriLyts [52]–[54] . In the present study , we can only exclude the effect of an intrinsic oriLyt promoter , as the reporter gene was placed upstream of the replication origin . However , the contribution of an enhancer function within the oriLyt sequence remains the subject of debate . Since induction was strongly dependent on DNA replication , enhancer activity alone cannot explain the results . Therefore , we assume that oriLyt-induced gene expression is due to: ( a ) conformational changes in DNA structure or histone packaging making the promoter for the reporter transgene more accessible to the transcription machinery , and/or ( b ) a marked increase in the number of templates available for transcription . At first glance , the presence of both a viral oriLyt and transgenes appears similar to an amplicon vector construct , as the amplicon also contains both elements and , in addition , a pac signal . However , the oriLyt within amplicon vectors serves only to amplify the vector DNA for efficient packaging of a high number of vector copies into helper virus capsids via the pac signal [14] . Interestingly , the phenomenon of transgene silencing of amplicon vectors by HDAC-dependent mechanisms has also been described [55] . Whether or not the context of the oriLyt itself affects the level of gene expression has , to our knowledge , not been investigated . Neither has the question of oriLyt-specific de-silencing of the vector by super-infection of cells carrying the vector been addressed . Taking the results of our study into account , we expect that expression of transgenes provided by amplicons derived from human pathogens would be affected by natural super-infection . This prediction is testable using gene therapy studies incorporating herpesvirus amplicon vectors . The strength of this effect would probably depend on details such as transcription orientation and distance from the oriLyt sequence . The replicon vector system for inducible gene expression should be applicable to all herpesvirus subfamilies . Moreover , most DNA viruses regulate late gene expression upon DNA replication , although it is achieved via different mechanisms . Such replicon vector systems may be applicable for polyomaviruses [56] and adenoviruses [57] with minor modifications . Notably , there are suggestions of a connection between the origins of replication and enhanced gene expression in higher eukaryotic genomes [58] , [59] . Trans-complementation of late proteins involved in herpesvirus morphogenesis is still a difficult task . Incorrect timing , aberrant intracellular distribution due to missing viral interaction partners and incorrect expression levels of the viral protein may explain poor complementation results . For instance , the isolated expression of late herpesvirus promoters without providing an oriLyt in cis results in aberrant early expression of the transgene . Today , systems for inducible gene expression typically require the use of small chemical compounds such as tetracyclin or doxycyclin ( as in the case of Tet-on/Tet-off systems ) [60] or rapamycin ( for FKBP12-based systems ) [61] . In these systems , gene expression is activated synchronously and irrespective of the state of virus replication in all cells , whereas the oriLyt-based system uses viral DNA replication as the signal for induction . The expression of the gene increases in proportion with the amplification of the vector DNA and reflects the natural expression kinetics of late herpesvirus genes . As each cell is activated individually upon infection , the appropriate and correct timing of expression of the trans-complementing gene is determined by the infecting virus . Trans-complementation carries the risk of recombination of genes provided in trans with the respective defective viral genome . We did not observe phenotypic or genotypic reversion of the ΔgO-phenotype . By contrast , we observed viable viruses at low levels in the M50-ori cell lines after reconstitution of a non-viable M50 deficient BAC . The high selection pressure for the production of infectious progeny favors the replication of M50 recombined viruses . In the replication competent gO-ori system , recombination only results in virus release; a phenotype with limited selective advantages . However , the rare reversion events observed during M50 complementation did not result in detectable reversion of the phenotype in first generation progeny ( see Fig . 4E ) . Future studies are needed to identify the conditions that increase or decrease the risk of recombination using replicon vectors . We also used the system for antiviral intracellular immunization . Constitutive expression of DN viral proteins in earlier studies [40] resulted in major defects in mice , such as substantial weight loss , probably due to cross-reaction between the dominant-negative viral transcription activators and cellular proteins , thereby altering natural gene expression profiles [40] . The expected advantage of replicon vectors lies in the fact that the DN protein is silenced . Thus , adverse effects caused by constitutive expression of toxic proteins are minimized . Our results show that we have been able to demonstrate proof-of-principle . It is not surprising that we could not achieve complete inhibition of viral replication in cell culture . A simple explanation is that the high abundance of wt SCP protein produced during infection ( in terms of copy number per capsid ) [62] simply out-competed the DN protein . One option may be to isolate clones containing high initial copy numbers of the replicon vector , or to replace the DN with that of a less abundant target protein such as M53 [63] . Episomal maintenance of pEPI has been shown in transgenic pigs , in which the pEPI-1 vector was not inactivated [64] . The present study reports , for the first time , the generation of transgenic mice using the pEPI replicon vector system . Functionality was testable in differentiated mES cell clones prior to generation of the transgenic lines . Starting from two responsive cell clones , two lines of VIOLA mice were generated . However , integration of the pEpibo-luc-ori vector was apparent in both VIOLA lines; although we cannot exclude the possibility that some episomal copies remained . Stabilization of the episomal properties could perhaps be improved by using murine S/MAR sequences instead of the human sequences used so far . Remarkably , we found no evidence for gene induction in vivo , although reactivation of the reporter by MCMV was possible in ex vivo explant cultures . Controls indicated that the failure to detect the luciferase signal in living mice was not due to insufficient sensitivity . The failure of transgene induction in vivo perhaps reflects the chromatin conditions surrounding the integrated constructs . These conditions change under in vivo and in vitro conditions . Major epigenetic changes happen during ex vivo tissue culture [65] , which might impinge on the strength of replicon vector silencing . Perhaps only metastable silencing conditions are lifted upon infection . Virus induced replication and expression forms the basis of the here described replicon expression system . This represents a tool kit , which can be favorably utilized for studying herpesvirus DNA replication , trans-complementation , gene therapy vector production and genetic immunization in vivo .
All animal experiments were performed in strict accordance with German animal protection law ( TierSchG ) and approved by the responsible state office Regierung von Oberbayern ( ROB ) under protocol number 55 . 2-1-54-2531-195-09 . The mice were housed and handled in an SPF condition in accordance with good animal practice and all efforts were made to minimize suffering as defined by Federation of European Laboratory Animal Science Associations ( FELASA ) and the national animal welfare body Gesellschaft für Versuchstierkunde - Society for Laboratory Animal Science ( GV-SOLAS ) . The oriLyt of MCMV [9] was subcloned from an MCMV-bacterial artificial chromosome ( BAC ) using a pick-up-cloning strategy . A PCR fragment was amplified using primers H5′-MCMV-oriLyt-ori6kan-for ( 5′-GGCGGGAGCG ACGGGGGCGA GGCTGGAGAG ATCGTCGTCC GCCATGCTAG CACGCGTGCC AGTGTTACAA CCAATTAACC-3′ ) and H3′-MCMV-oriLyt-ori6kan-rev ( 5′-GAACGACCCC CGCTCCTGTA TAATTTCGAT GCCGGGGAGG TCGCCACGCG TCTGAAGATC AGCAGTTCAA CCTGTT-3′ ) containing homologies at H5′ to 91850–91894 bp and H3′ homologies at 91895–91939 bp of the MCMV-FRT BAC pSM3fr-FRT flanking a kanamycin resistance gene ( kanR ) and the bacterial oriR6K [66] . This PCR fragment was recombined by homologous recombination into pSM3fr-FRT as previously described [67] . Due to the insertion of a new NheI site ( at the most upstream position in the inserted fragment ) the oriLyt-containing genome segment could be excised together with the bacterial amplicon and the kanR marker by NheI digestion , as the next NheI site in the genome ( at 95767 bp ) was sited downstream of the oriLyt . The fragments generated by NheI digestion were recircularized by DNA ligation and used to transform PIR1 E . coli ( Invitrogen ) . In PIR1 E . coli , only the oriR6K ( and thereby the oriLyt-containing circularized NheI fragment ) was maintained after kanamycin selection . The oriLyt fragment was subcloned from the resulting pO6kan-oriLyt plasmid by subsequent digestion with MluI and ligated into the MluI site of a modified pEPI-1 vector containing a blasticidin resistance gene ( bsr ) and an mOrange fluorescent marker ( instead of the neomycin resistance marker and the gfp gene in the original vector ) [68] to yield pEpibo-oriLyt . An additional FL reporter gene , driven by the minimal SV40 promoter ( 48–250 bp ) was excised from the pGL3-control plasmid ( Promega , Acc# U47296 ) and inserted into the pEpibo-oriLyt vector upstream of the oriLyt via the KpnI and XbaI restriction sites . The control plasmid , pEpibo-luc , was constructed by inserting the same fragment into the pEpibo vector ( w/o oriLyt ) . To construct the pEpibo-GFPSCP-ori plasmid , the luciferase ORF was excised from pEpibo-luc-ori via the HindIII and XbaI sites and replaced with a PCR fragment containing the GFP-SCP ORF [41] flanked by the respective restriction sites . The vector pEpibo-gO-ori was generated by blunt-end cloning of the gO-ORF from pCR3-m74 [34] via Ecl126II into the HindIII and XbaI sites of pEpibo-luc-ori . To generate the pEpiNo-M50HA-ori vector , which contains a neomycin selection cassette instead of the blasticidin resistance gene , M50HA from pOriR6K-zeo-ie-M50HA [69] was cloned into the HindIII and XbaI sites of pEpiNo-luc-ori . Mouse embryonal fibroblasts ( MEFs ) and NIH3T3 cells ( ATCC CRL-1658 ) were maintained in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal calf serum ( FCS ) , 100 units/ml penicillin , and 100 units/ml streptomycin . To generate stable cell lines , NIH3T3 cells were transfected in 6-well plates with 2 . 5 µg of the pEpibo construct using the TransIT-3T3 ( Mirus ) transfection reagent according to the manufacturer's protocol . Transfected cells were selected with 10 µg/ml blasticidin S ( BS , Invivogen ) or with 200 µg/ml G418 ( Invivogen; for M50- ori ) and were either studied as cell pools or as clones isolated by limiting dilution . The MCMV-wt and MHV68-wt viruses were derived from BACs pSM3fr and Rγ HV68A98 . 02 , respectively [70] , [71] . Reconstitution of the viruses , preparation of virus stocks and titrations were performed as described previously [72] , [41] . To determine in vitro growth of MCMV in different cell lines , cells were plated at the same density and infected at an MOI of 0 . 1 ( unless otherwise indicated ) . After 1 h of virus infection , cells were washed and fresh medium supplied . Supernatants of the infected cells were taken at the indicated time points ( in triplicate ) and titrated against MEFs using a standard plaque assay , or against the respective replicon cell line according to TCID50 . Growth curve experiments were performed at least twice . MCMV-ΔgO virus production and trans-complementation were performed as described previously [34] . The M50-deficient virus was generated by modifying pSM3fr-Δ1-16-FRT [73] . First , to monitor late gene expression by MCMV , an expression marker was introduced into the 3′ UTR of the M48 . 2 gene , which encodes the small capsid protein ( SCP ) . The internal ribosome entry site ( IRES ) of the encephalomyocarditis virus was derived from pIREShyg3 ( Clontech , nt 1333–1924 ) and assembled upstream to the mCherry ORF from pmCherry-C1 ( Clontech nt 597–1448 ) . This IRES-mCherry cassette was then introduced into MCMV-BAC between nt 73570 and nt 73571 and directly after the M48 . 2 stop codon by recombineering [74] yielding pSM3fr-Δ1-16-FRT-SCPiChe . Next , the EGFP coding region from pEGFP-N1 ( Clontech nt 679–1398 ) and the ampicillin resistance gene from LITMUS 28 ( NEB nt 2764–1003 ) were assembled and introduced into pSM3frΔ1-16-FRT-SCPiChe between nt 75730–76451 via a second round of recombineering to yield pSM3fr-Δ1-16-ΔM50F . The newly-inserted cassette replaced most of the M50 ORF and allowed the EGFP ORF to be expressed instead of the deleted M50 . The C-terminal 75 aa of M50 , which overlap with the M49 gene , were left intact as described for the M50 deletion mutants [75] . To generate the VIOLA mouse line ( virus inducible oriLyt-dependent luciferase animal ) , the murine embryonic stem cell line , mES E14 , was transfected with pEpibo-luc-ori using an AMAXA nucleofector ( Lonza Cologne , Germany ) according to the manufacturer's instructions using program A-013 . Transfected mES cells were cultured in mES cell medium ( DMEM containing 15% FCS , 1% non-essential amino acids ( Gibco ) containing 1000 U/ml LIF ( Millipore , Temecula , USA ) ) on mitomycin C-treated NIH3T3-bsr fibroblasts and selected with 5 µg/ml BS for 3 days . To select the appropriate clones , a proportion of the mES cells was differentiated by removing the LIF and feeder cells to allow productive infection with MCMV [44] . Appropriate clones were injected into C57BL/6 blastocysts , which were then implanted into a NMRI foster mother . The resulting chimeras were backcrossed to 129X1/SvJ mice ( Jackson ) to obtain the VIOLA line . For the explant cultures , the lungs , heart , muscle , fat and salivary glands were excised from individual mice , washed with PBS and minced in dissociation buffer ( 12 . 5 mM HEPES , 200 U/ml DNaseI and 13 Wünsch U/ml Liberase TM in PBS ) . Tissues were incubated at 37°C until a smooth homogenate was obtained . The homogenates , as well as the spleens and bone marrow cells ( prepared from the femurs ) were passed through a 100-µm strainer and resuspended in DMEM . Lung cells were resuspended in LSGS medium ( DMEM containing , 15% FCS , 1% non-essential amino acids ( NEAA ) , 1 µg/ml hydrocortisone , 10 ng/ml human epidermal growth factor , 3 ng/ml basic fibroblast growth factor , 10 µg/ml heparin ) . Heart , muscle , salivary gland and fat cells were cultured in DMEM containing 15% FCS , 1% NEAA and 50 µM 2-mercaptoethanol . Spleen cells were cultured in RPMI1640 containing 10% FCS , 1% L-glutamine and 10 mM HEPES . All cells other than muscle were plated on gelatin-coated culture flasks . Muscle cells were grown on collagen type V-coated culture flasks . To quantify FL expression , 100 , 000 cells/well were seeded into 12-well plates ( in triplicate ) . After 4 h , the cells were either infected with MCMV or mock-treated . At 24 or 36 h after infection , the cells were carefully washed with PBS and lysed with 200 µl 1× Passive Lysis Buffer ( Promega ) . The luciferase activity in the lysates ( 10 µl ) was measured in a 96-well luminometer ( Berthold ) . Experiments were performed at least three times . For the analysis of the silencing mechanism , cells were treated with 330 nM trichostatin A ( TSA ) , or 25 µM 5′aza-cytidine for 36 h , and luciferase assays performed as described above . To block virus replication , 0 . 3 mg/ml PAA was added to the media upon infection . The vector DNA in uninfected and infected cells was determined by quantitative real-time PCR . Cells ( 200 , 000 ) were seeded into each well of a 12-well plate and infected with MCMV-wt at an MOI of 0 . 5 or mock-treated . Total genomic DNA was extracted from uninfected and infected cells 36 h post infection ( p . i . ) using the DNeasy blood and tissue kit ( Qiagen ) according to the manufacturer's instructions . To quantify the ratio of pEpibo-luc-ori per cell ( before and after infection ) , two PCRs were performed . One was specific for the pEpibo constructs , amplifying the blasticidin resistance gene ( bsr ) using the primers bsr-for-taqman 5′-CCTCATTGAA AGAGCAACGGCTAC-3′ and bsr-rev-taqman 5′-GCACCACGAGTTCTGCACAAGGT-3′ and the specific probe for bsr ( 5′-FAM- CATCTCTGAAGACTACAGCGTCGCCA-TAMRA-3′ ) . The second PCR was specific for the cellular lamin B receptor ( LBR ) gene and used the primers LBR-for ( 5′-GGAAGTTTGTTGAGGGTGAAGTGGT-3′ ) and LBR-rev ( 5′-CCAGTTCGGTGCCATCTTTGTATTT-3′ ) and a specific probe for LBR ( 5′-FAM-CTGAGCCACG ACAACAAATCCCAGCTCTAC-TAMRA-3′ ) . Vector DNA copy number was calculated by comparing the amplification with standard curves generated using the plasmids pEpibo-luc-ori or p06IET-LBRfl . DNA amplification was performed using the TaqMan 1000 RXN PCR core reagent kit ( 50 cycles of 95°C for 15 s and 60°C for 1 min ) ( Applied Biosystems ) . For the recombination assay , standard PCR was used to detect the m74 gene with the primers m74-for 5′-TCCGGACAACGTCTTTCCC-3′ and m74-rev 5′-CCTCAGTTCCACTTGCCAGC-3′ , which amplify 316 bp inside the m74 gene . The primers M54-for 5′-ATCATCCGTTGCATCTCGTTG-3 and M54-rev 5′-CGCCATCTGTATCCGTCCAT-3′ were used to detect the M54 gene . The immediate-early protein , ie1 ( a marker of infection ) was detected using an indirect immunofluorescence assay . Cells were fixed in 50% acetone-50% methanol and stained using the monoclonal antibody , Croma 101 ( kindly provided by Stipan Jonjic , University of Rijeka , Croatia ) followed by a Cy3-coupled goat anti-mouse antibody ( Dianova ) . Fluorescence and bright-field imaging were observed on an Axiovert 200 M system ( Zeiss ) . Metaphase spreads of proliferating NIH3T3:gfpscp-ori cells were obtained after 2–3 h incubation with demecolcemid ( 0 . 1 µg/ml , Sigma Aldrich ) and resuspension in 0 . 91% ( w/v ) tri-sodium citrate-dihydrate hypotonic solution . Cells were fixed in cold Carnoy's fixative ( methanol∶acetic acid; 3∶1 v/v ) and stored for 1 week before further processing . FISH was performed as previously described with minor modifications [76] . Briefly , three different probes were generated to detect the pEpibo-gfpscp-ori plasmid . The first probe , detecting the bsr gene , was PCR-labeled with biotin-ATP ( NEB , USA ) ( using the primers bsr-for-FISH: 5′-ATGGCCAAGCCTTTGTCTCA-3′ and bsr-rev-FISH 5′-AGATCGAGAAGCACCTGTCG-3′ ) and the second probe , detecting the gfpscp gene , was PCR-labeled with dig-UTP ( Roche ) ( using the primers P ( SV40 ) -for-FISH: 5′-TACCGAGCTCTTACGCGTGC-3′ and pA ( SV40 ) -rev-FISH 5′-TAAGATACATTGATGAGTTTGGA-3′ ) . The third probe , detecting the oriLyt fragment , was generated by nick translation using DEAC-UTP ( Perkin Elmer , USA ) . Immunolabeling of the probes was achieved by adding streptavidin-Cy3 . 5 ( Rockland , USA ) and an anti-dig-fluorescein antibody ( Roche Diagnostics , Germany ) . DNA was counterstained with DAPI . Images of the FISH slides were taken using an Axiovert 200 microscope ( Zeiss , Germany ) at 60× magnification . Western Blot analysis was performed as previously described [41] using rat anti-HA ( 3F10 , Roche ) , mouse anti-IE1/3 ( CHROMA 101 , kindly provided by Stipan Jonjic ) , mouse anti-gp48 ( CHROMA 221 , kindly provided by Stipan Jonjic ) and rabbit anti-actin primary antibodies ( 20–33 , Sigma ) . Horseradish peroxidase-coupled anti-mouse , anti-rat , and anti-rabbit immunoglobulin-specific antibodies ( Dianova ) were used for immunodetection followed by the ECL-Plus ( Amersham ) system . Statistical analysis was performed using GraphPad Prism4 . Two-way ANOVA was used to analyze differences unless otherwise indicated . Asterisks denote statistically significant differences ( *P<0 . 05; **P<0 . 01; ***P<0 . 001 ) . M50: GenBank ADD10432 . 1 m74: GenBank ADD10446 . 1 m48 . 2: GenBank ADD10430 . 1 MCMV Smith Strain: GenBank NC_004065 | All herpesviruses show a precisely regulated gene expression profile , including true-late genes , which are turned on only after the onset of DNA replication . We used this intrinsic viral mechanism to generate a versatile conditional gene expression system that exploits the activity of the murine cytomegalovirus ( MCMV ) viral origin of lytic replication ( oriLyt ) . Upon virus infection , replication of the viral genome also led to the replication and activation of the oriLyt-coupled episomal transgene . The oriLyt-based replicons were silenced in all stable cell lines and transgenic mice; however , virus infection liberated the plasmids from histone-deacetylase-induced inactivation . As maximum gene expression relied on relief from silencing via replication of the episomal constructs , very strong induction of the reporter gene was achieved . We showed that this system can be used for trans-complementation of late , toxic viral genes , to block virus production by activating dominant-negative ( DN ) transgenes , and to provide a new tool to study the principles of viral replication . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
]
| [
"biotechnology",
"medicine",
"infectious",
"diseases",
"model",
"organisms",
"genetics",
"biology",
"microbiology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
]
| 2012 | Cytomegalovirus Replicon-Based Regulation of Gene Expression In Vitro and In Vivo |
RNA silencing plays a critical role in plant resistance against viruses . To counteract host defense , plant viruses encode viral suppressors of RNA silencing ( VSRs ) that interfere with the cellular silencing machinery through various mechanisms not always well understood . We examined the role of Mungbean yellow mosaic virus ( MYMV ) AC4 and showed that it is essential for infectivity but not for virus replication . It acts as a determinant of pathogenicity and counteracts virus induced gene silencing by strongly suppressing the systemic phase of silencing whereas it does not interfere with local production of siRNA . We demonstrate the ability of AC4 to bind native 21–25 nt siRNAs in vitro by electrophoretic mobility shift assay . While most of the known VSRs have cytoplasmic localization , we observed that despite its hydrophilic nature and the absence of trans-membrane domain , MYMV AC4 specifically accumulates to the plasma membrane ( PM ) . We show that AC4 binds to PM via S-palmitoylation , a process of post-translational modification regulating membrane–protein interactions , not known for plant viral protein before . When localized to the PM , AC4 strongly suppresses systemic silencing whereas its delocalization impairs VSR activity of the protein . We also show that AC4 interacts with the receptor-like kinase ( RLK ) BARELY ANY MERISTEM 1 ( BAM1 ) , a positive regulator of the cell-to-cell movement of RNAi . The absolute requirement of PM localization for direct silencing suppression activity of AC4 is novel and intriguing . We discuss a possible model of action: palmitoylated AC4 anchors to the PM by means of palmitate to acquire the optimal conformation to bind siRNAs , hinder their systemic movement and hence suppress the spread of the PTGS signal in the plant .
Viruses are obligate intracellular parasites that exploit host machineries to propagate and spread in the host . Their presence and activity deploy diverse plant mechanisms to combat viral infections at both the cellular and the whole-organism levels . Double-stranded ( ds ) RNA forming during virus replication and self-complementary foldback RNAs from single-stranded viral RNAs or aberrant RNAs can trigger host defence responses via a mechanism of RNA interference ( RNAi ) that results in inhibition of target RNA expression [1 , 2] . The RNase III-type DICER enzymes process these viral RNAs into small-interfering ( si ) RNAs ( 21–24 nucleotides ) that accumulate in the infected cells and guide the RNA-induced silencing complex ( RISC ) to degradation of complementary viral RNA sequences [3 , 4] . RNA silencing is a non-cell autonomous process thus , silencing signals spread from the site of induction to neighbouring cells and systemically to confer silencing of homologous targets in distant tissues of the host plant [5 , 6] . However , the evidence that virus infection often induces symptom and damage in the host highlights the presence of a counter defence strategy that suppresses the host surveillance [7] . Viruses encode one or more proteins that can inhibit initiation ( viral RNA recognition and the subsequent degradation ) , maintenance , or systemic spreading of silencing thus allowing efficient viral replication in single cells and spread of the infection . These virulence factors , called viral suppressors of RNAi ( VSRs ) , share no obvious sequence homology with each other and follow distinct mechanisms of suppression by targeting different points of the RNA silencing pathway , such as viral RNA recognition , dicing , RISC assembly , RNA targeting , and amplification [2 , 8] . To overcome the host silencing machinery , several virus species have developed a siRNA sequestration strategy that the different VSR apply in various manners by preventing the assembly of the RISC effector [8] . As siRNA duplexes act as mobile silencing signals moving ahead of the virus to activate antiviral silencing in not yet infected cells , by sequestering and inactivating siRNA VSRs can counter react this defense strategy and allow spreading of the viral infection in the plant [9] . Members of the family Geminiviridae are small , circular , single-stranded DNA viruses composed of one or two genomic segments of 2500–3100 nucleotides which are encapsidated within small twinned icosahedral particle that replicate in the nucleus of an infected cell via double-stranded intermediates that also serve as templates for bidirectional transcription [10] . Geminivirus host different suppressor proteins encoded by open reading frame ( ORF ) AC2 , V2 , ORF β C1 , ORFs AC4 and AC5 [11–15] . The transcriptional activator protein ( TrAP ) encoded by the ORF AC2 of African cassava mosaic virus ( ACMV ) , Tomato golden mosaic virus and Mungbean yellow mosaic virus ( MYMV ) , the C2 of Tomato leaf curl virus and the β C1 of the Tomato yellow leaf curl China virus ( TYLCCNV ) share sequence nonspecific DNA binding activity and localization in the nucleus where they act by a mechanism depending on interaction with DNA and transcriptional activation or with key components of the RNA silencing pathway . On the other side , the V2 protein of Tomato yellow leaf curl virus ( TYLCV ) -Is has specific cytoplasmic localization and exerts its VSR activity by targeting a step after siRNA production thus representing a different example of VSR in geminivirus [16] . The MYMV AC5 , a protein encoded by some begomoviruses , suppresses post-transcriptional gene silencing ( PTGS ) and can reverse methylation-mediated TGS [14] . The AC4/C4 gene lies entirely within the Rep coding region , but in a different reading frame , and is one of the least conserved among members of the Geminiviridae family . Its function is very controversial: mutagenesis and/or transgenic expression of some AC4/C4 genes results in no phenotype or phenotypes consistent with movement protein or symptom determinant activity [17] . This puzzling information has been enriched with the discovery of a role of AC4/C4 in the suppression of RNA silencing in different strains of ACMV [13 , 18] , in MYMV [19] , in the monopartite TYLCV and in Bhendi yellow vein mosaic virus [20 , 21] . These proteins block cytoplasmic RNA silencing by a mechanism that involves binding of single-stranded siRNA and miRNA and possibly facilitates their degradation . This suggests that the severe developmental defects observed upon transgenic expression of some AC4/C4 might be due to suppression of overlapping steps in the siRNA and miRNA pathways [22 , 23] . Interestingly , AC2s and AC4s of cassava viruses behave differently in regulating silencing suppression exerting strong or weak activity depending on the viral strain , and apparently compensating each-other function [13] . Consequently , mixed-strain infections can be responsible of unusually severe cassava mosaic disease in the field [24] . S-acylation or palmitoylation , is a reversible posttranslational modification of a protein covalently attaching through a cysteine residue ( s ) to long chain fatty acid , usually the 16-carbon palmitate via a thioester bond . This modification increases protein membrane affinity and provides an important mechanism for regulating cellular functions including subcellular localization , stability , trafficking , stress response , disease resistance , hormone signaling , cell polarisation , cell expansion and cytoskeletal organization [25 , 26] . Unique among lipid modifications of proteins , this attachment is reversible , thus offering dynamic control over the cellular processes and protein function in response to stimuli . Our understanding of S-acylation function in plants is quite limited compared with other organisms and mainly comes from targeted studies on the functional characterization of individual proteins that happen to be S-acylated [27] . Several examples both in plant and animal systems describe palmitoylation as a modification used by proteins to switch subcellular localization between nucleus and plasma membrane ( PM ) and to accomplish their tasks . For example , specific functions regulated by transcription factor ( TF ) in the nucleus are triggered or hindered by palmitoylation-mediated protein localization to the nucleus or to the PM , respectively [28 , 29] . In plants , differential subcellular localization of TF induced upon palmitoylation , are associated to plant response to abiotic stresses such as salt and drought increase [28 , 30] . A large number of S-acylated proteins are also involved in plant–microbe interactions . Among them , a proteomic approach identified proteins involved in pathogen perception and response , mitogen-activated protein kinases ( MAPKs ) , leucine-rich repeat receptor-like kinases ( LRR-RLKs ) and RLK superfamily members , ATPases , integral membrane transporters , soluble N-ethylmaleimide-sensitive factor-activating protein receptors ( SNAREs ) and heterotrimeric G-proteins [31] . PM is a critical subcellular compartment for the actors of a pathosystem . In fact , plants use the covalent addition of fatty acids to target an array of sensor/receptor proteins to the PM and detect invading pathogens whereas pathogens ( except for viruses that do not penetrate plant cells actively ) secrete effector proteins into the plant cell , particularly the internal face of the host PM , to threaten this surveillance and induce plant susceptibility to infection [32] . Animal viral proteins such as glycoproteins from Vesicular stomatitis virus [33] , the Influenza virus hemagglutinin as well as the transmembrane Matrix-2 ( M2 ) [34 , 35] can also undergo S-acylation that is essential for virus replication or infection . However , palmitoylation of plant viral proteins has not been reported so far . Interestingly , C4 protein from Tomato yellow leaf curl virus ( TYLCV ) targets , through a yet unknown mechanism , PM and plasmodesmata ( PD ) where inhibits the intercellular spread of RNAi by interacting with receptor-like kinase ( RLK ) BARELY ANY MERISTEM 1 ( BAM1 ) [36] . In this study , we examined the role of MYMV AC4 in viral infectivity . We revealed that AC4 undergoes post-translational palmitoylation that mediates protein targeting to the PM . When localized to the PM , AC4 strongly suppresses systemic silencing whereas delocalization from such subcellular compartment impairs VSR activity .
As a first step to gain more insights into the function of MYMV AC4 , we determined whether AC4 is essential for successful MYMV infection . To this aim , we modified the three in frame start codons of the ORF AC4 within the infectious clone pGA1 . 3A [37] to obtain a MYMV-ΔAC4 DNA A mutant . Mutations were designed to be silent in the overlapping AC1 ORF and did not produce any change in the amino acid sequence of AC1 . Vigna mungo plants were biolistically inoculated with recombinant and wild type ( wt ) MYMV DNA A , each together with the infectious MYMV DNA B clone pGA1 . 3B [37] , and monitored for symptom appearance for two months . Typical yellow mosaic and leaf curling symptoms in the trifoliate leaves became evident 18 ( +/- 3 ) days post inoculation ( dpi ) on plants infected with wt MYMV whereas no symptom were observed on plants inoculated with a MYMV-ΔAC4 for the entire time of observation ( Fig 1A ) . PCR analysis of total DNA extract confirmed the absence of viral DNA in systemic leaves ( Fig 1B ) and highlighted the requirement of AC4 for systemic spread of MYMV in the host . The absence of systemic symptoms in plants inoculated with MYMVΔAC4 might be either consequence of a local event ( such as virus inability to replicate/accumulate in the initially inoculated cells or to move out of them ) or reflect long distance movement incompetence and , similar to homologue geminivirus VSR , MYMV AC4 could suppress very early host antiviral defence [13] . To investigate in more detail the role of AC4 in the early stage of infection , we conducted quantitative real-time ( qPCR ) experiments to analyse and monitor the accumulation of MYMV in V . mungo biolistically-inoculated leaves during the initial five days of infection . To this aim , we constructed two control virus mutants: MYMVΔAC1 and MYMVΔBC1 expressing null replicase and movement protein functions , respectively . To evaluate the genome replication and accumulation of MYMVΔAC4 , MYMVΔAC1 and MYMVΔBC1 virus mutants relatively to MYMV WT , we compared the quantification of the MYMV AC2 gene , unrelated to the mutated genes , at four time points ( 1-2-3-5 dpi ) and used contrast statistical analysis to compare the value recorded at 5 dpi with those obtained at the previous time points for each couple of virus constructs inoculated . Statistically equivalent trends of viral DNA accumulation were observed in plants inoculated with MYMV WT and MYMVΔAC4 , both of which reached the highest and statistically most distinct value at 5 dpi ( Fig 1C ) . Due to the lack of viral replicase , the concentration of MYMVΔAC1 dropped within the first two days and continued to decrease slowly but constantly accordingly to the continuous degradation of the input DNA ( Fig 1C , red line ) . Accumulation of the MYMVΔBC1 mutant , deprived of the movement protein function , also followed a negative trend statistically not different from MYMVΔAC1 ( Fig 1C ) . However , differently from MYMVΔAC1 , the concentration of MYMVΔBC1 decreased more slowly than MYMVΔAC1 , probably reflecting the occurrence of MYMVΔBC1 replication in single cells , and remained statistically not different from MYMVΔAC4 until 3dpi ( S1 Fig ) . From this time point , the inability to exit infected cells probably triggered MYMVΔBC1 DNA degradation or infected-cell death . Taken together , these results , obtained from three independent replications of the experiment , indicate that , in the absence of AC4 , MYMV can replicate and move from cell to cell , and rule out the absolute requirement of AC4 for virus replication and cell-to-cell movement . However , the lower efficiency of MYMVΔAC4 compared to wt ( S1 Fig ) , and the relatively low , albeit statistically significant difference with the movement-impaired MYMVΔBC1 mutant , suggest that a possible indirect contribution to the mechanism of virus transport in plant cannot be excluded . Many VSRs are pathogenic determinants of their virus host . They can interfere , independently or synergistically with other VSRs , to enhance the severity of symptoms caused by related or unrelated viruses [38 , 39] . As the knock out of AC4 in the viral genome hinders systemic plant infection , to investigate the involvement of this VSR in viral pathogenicity , we tested the effect of MYMV AC4 expression on the symptom onset induced by heterologous Potato virus X ( PVX ) [40] infection in Nicotiana benthamiana . All inoculated plants developed systemic leaf puckering and those expressing VSRs showed additional severe stunting ( Fig 2 ) . PVX-induced symptoms were significantly worsened by the simultaneous expression of AC4; however , the necrotic phenotype typically induced by Carnation Italian ringspot virus P19 , which we used as positive control [41 , 42] , was not observed in plants agroinfiltrated with PVX-AC4 ( Fig 2 ) . This result provides support for the notion that MYMV AC4 is a determinant of viral pathogenicity and suggests that , as for other VSR , the enhancement of PVX-induced symptoms might be related to the capacity of AC4 to interfere with components of the endogenous RNAi pathway . Aiming at understanding the molecular basis of the RNAi suppression activity of AC4 , we used GFP-transgenic N . benthamiana plants ( line 16c ) overexpressing GFP constitutively [39] . These plants show green fluorescence under UV light . Upon transient expression of GFP , inducing silencing of the transgenically expressed GFP , they display only chlorophyll autofluorescence and appear red under UV light . In the presence of a suppressor of RNA silencing GFP silencing is blocked and plants continue to exhibit green fluorescence . To induce silencing , we used a PVX-GFP plasmid [40] to agroinfiltrate N . benthamiana 16c plants and monitored the progress of silencing on the upper leaves for 30 days . Upon infiltration of PVX-GFP , ectopic GFP expression was observed in the inoculated leaves and in the veins of new leaves from 2 dpi under UV light ( Fig 3A ) . The intensity of the green fluorescence signal increased until 5 dpi but was followed by a rapid replacement of the green fluorescence with chlorophyll red autofluorescence , due to GFP silencing , at 6–7 dpi ( Fig 3A ) . By 12 dpi these plants appeared completely red fluorescent ( Fig 3A ) . Plants inoculated with equal amounts of PVX-GFP , PVX-AC4 ( PVX-GFP+PVX-AC4 ) and PVX-P19 ( PVX-GFP+PVX-P19 ) showed green fluorescence on newly emerging leaves starting from 5 dpi , simultaneously to the onset of silencing in PVX-GFP agroinfiltrated plants ( Fig 3A ) . At 15 dpi , when PVX-GFP plants were completely red fluorescent , plants infiltrated with PVX-GFP+PVX-AC4 , similar to the control and PVX-P19 ( PVX-GFP+PVX-P19 ) [42] , were still green fluorescent , and most of the leaves continued to show silencing suppression at 30 dpi , very strongly in the youngest emerging leaves ( Fig 3A ) . RNA gel blot analysis of GFP mRNA and 21–25 nt RNA confirmed these observations . 15 dpi , high levels of GFP mRNA were found in young leaves of plants inoculated with both PVX-GFP+PVX-AC4 and with PVX-GFP+PVX-P19 ( used as positive control ) whereas GFP mRNA was undetectable in plants inoculated with PVX-GFP ( Figs 3B and 6C ) . Conversely , GFP siRNA where detected only in plants inoculated with PVX-GFP alone ( Fig 3B ) . Collectively , these results demonstrate that co-delivery of GFP and MYMV AC4 onto GFP-expressing N . benthamiana strongly suppresses the onset of VIGS compared with the progress of gene silencing obtained with PVX-GFP alone . A Northern blot analysis was also conducted on the same samples with a PVX coat protein ( CP ) probe . Consistent with the extent of silencing suppression , very high levels of PVX chimera were observed in plants inoculated with PVX-GFP+PVX-AC4 and with the PVX-GFP+PVX-P19 control ( Fig 3C ) . The fact that the PVX CP probe detected much lower levels of viral RNA in plants inoculated with PVX-GFP alone ( Fig 3C ) suggests that the spread of GFP silencing observed in young leaves of plants inoculated also with PVX-GFP+PVX-AC4 was due more to the spread of the silencing signal than to a de novo silencing by PVX-GFP . These results demonstrate that MYMV AC4 can suppress silencing-related defence responses in transgenic N . benthamiana plants . Nevertheless , the loss of virus accumulation observed in systemic leaves of V . mungo infected with MYMV-ΔAC4 ( Fig 1B ) suggests that this may be the result of systemic VIGS also in the virus natural host . The VSR activity of AC4 was further investigated in PTGS experiments . To this aim , AC4 and P19 were cloned under the control of the 35S promoter and each co-infiltrated to line 16c plants , together with the silencing inducer ( full-length GFP ) under the control of the same promoter . Two dpi , infiltrated leaf patches appeared green fluorescent under UV light and , consistently with higher accumulation of the GFP transcript , those co-infiltrated with the P19 VSR control appeared brighter than the others ( Fig 4A and 4B ) . By 7 dpi , when red fluorescence had completely replaced green fluorescence in 35S-GFP infiltrated patches , GFP mRNA was still present in AC4 and P19 co-agroinfiltrated plants but almost undetectable in plants infiltrated with 35S-GFP alone ( Fig 4A and 4B ) . Interestingly , GFP siRNAs started to accumulate in 35S-AC4 co-infiltrated patches at 2 dpi , and at 7 dpi they were in a concentration similar to 35S-GFP but much higher than the 35S-P19 control ( Fig 4B ) . Remarkably , despite a red fluorescent front developed around the infiltrated area , systemic GFP silencing as well as accumulation of 21–25 nt RNA were not observed in the upper leaves of plants agro-infiltrated with 35S-GFP+35S-AC4 ( Fig 4A and 4B ) . The evidence that siRNAs accumulate in 35S-GFP+35S-AC4 patches at a concentration similar to 35S-GFP ( Fig 4B ) reveals that MYMV AC4 does not interfere with production of transgene-induced gene silencing whereas the absence of siRNAs in the upper leaves indicates a possible involvement in long-distance spreading of the silencing signal . To identify the major site ( s ) of subcellular localization of AC4 , we fused the recombinant AC4 with an influenza virus hemagglutinin epitope ( HA ) tag , and used it for protoplast transfection . Protoplasts transfected with pCKAC4HA were lysed at 24 h post transfection ( hpt ) and the lysate was submitted to differential centrifugations: low speed centrifugation ( 500 g ) to collect nuclei and residual intact cells , and high speed centrifugation ( 30000 g ) to separate the soluble membrane fraction from the crude part . Equivalent amounts of each fraction were analysed by immunoblotting with an anti-HA antibody . AC4 was detected in the pellets from both low- and high-speed centrifugations ( Fig 5A ) but was absent in the supernatants . To further investigate the association of AC4 with the membrane fraction , we treated the pellet obtained from high-speed centrifugation with Na2CO3 , urea or KCl that remove proteins weakly bound to membranes . After a second high-speed centrifugation , a band corresponding to AC4 was detected in every pellet regardless of the different treatment applied ( Fig 5A ) . The evidence that none of the treatments could dislodge AC4 from the membrane fraction indicates a very strong interaction of the protein with cellular membranes whereas the presence in the low-speed pellet suggests that AC4 could also be present in the cytosol and possibly in the nucleus . The subcellular localization of AC4 was further investigated by expressing the protein in fusion with GFP in N . benthamiana protoplasts . The fluorescence signal was monitored at different time points between 4 and 48 hpt . Between 4 and 6 hpt , a fluorescence signal localized to the PM was visible in most of the transfected protoplasts whereas only in few of them GFP-AC4 was also visible in the nucleus . Starting from 8 hpt the majority of fluorescent protoplasts showed a double localization to the PM and the nucleus ( Fig 5B ) that did not change in the course of the experiment . GFPAC4 was also expressed in V . mungo leaf mesophyll by means of biolistic particle delivery . Consistently with observation in protoplasts , GFPAC4 localizes in the nucleus and at the cell periphery of single mesophyll cells ( S2 Fig ) , and accumulates at PD ( S3 Fig ) . In silico analysis of the physical-chemical properties of AC4 , performed by using the web-interface SeqWeb of GCG Wisconsin Package ( version 2 ) [43] , predicted that AC4 is an hydrophilic protein except for a region comprised between aminoacids 6 to 12 ( S4 Fig ) . The core of this region is characterized by two hydrophobic phenylalanines flanking a polar cysteine ( position 11 ) , which , the CSS-Palm 4 . 0 software [44] predicted might be palmitoylated ( S4 Fig ) . The AC4 sequence following this hydrophobic part is expected to have a high surface probability ( S2 Fig ) . The pick of this region is occupied by the KRR amino acid sequence that was predicted to be a potential nuclear localization signal ( NLS ) by the NucPred software [45] . To gain more insight into the involvement of the two in silico-identified domains in the subcellular localization of AC4 , we replaced the amino acid C11 by an alanine to produce the GFPAC4 ( C-A ) mutant . The amino acids K19 , R20 and R21 were also replaced together by alanines to obtain the mutant GFPAC4 ( KRR-AAA ) . These single mutants and a double mutant comprising both mutations GFPAC4 ( C-A/KRR-AAA ) , were transiently expressed in N . benthamiana protoplasts and fluorescence signal was observed at 24 hpt . Upon alanine-substitution of C11 , GFPAC4 ( C-A ) accumulated only in the nucleus and didn’t show PM localization at any time ( Fig 5B ) . On the other side , mutation of the hypothetical NLS delocalized AC4 ( KRR-AAA ) from the nucleus and the protein accumulated only at PM ( Fig 5B ) . Mutation of both domains resulted in cytoplasmic diffusion of AC4 ( C-A/KRR-AAA ) with subcellular localization indistinguishable from free GFP ( Fig 5B ) . These results , further supported by similar results obtained upon expression in single mesophyll cells of bombarded V . mungo leaves ( S2 Fig ) , indicate that the in-silico predictions were correct and that the two predicted domains are indeed responsible for the subcellular localization AC4 . AC4 mutagenesis indicates that C11 is a critical amino acid for protein targeting to the PM and strongly supports the in silico prediction of post-translational palmitoylation of the protein . S-acylation ( palmitoylation ) is the reversible post-translational addition of a saturated fatty acids ( palmitate or stearate ) through thioester linkages to cysteine residues of proteins [46] . While no specific consensus domain exist for palmitoylation , the cysteine involved in the thioester bond should be localized inside the protein in a favourable context to allow insertion of the fatty acid and docking to the PM [47] . To confirm that C11 in AC4 can direct protein localization to the PM , we inserted the AC4 DNA sequence encoding aminoacids 1 through 12 upstream of egfp ( Fig 5C ) . The corresponding fusion protein , AC4 ( 1–12 ) GFP was transiently expressed in N . benthamiana protoplasts and observed at 24 hpt by video confocal microscopy . The addition of the N-terminal 12 amino acids of AC4 displaces GFP from te cytosol and AC4 ( 1–12 ) GFP is relocated to the PM ( Fig 5C ) . To get a definitive proof of AC4 palmitoylation , we performed a biotin-switch assay , a biochemical test using hydroxylamine for specific cleavage of thioester bonds . Therefore , this assay allows only palmitoylated proteins to be cleaved and biotinylated in a cellular lysate and , upon biotin affinity purification , their detection by immunoblotting . To this aim , we transfected protoplasts with the pCKAC4HA or pCKAC4 ( C-A ) HA mutant plasmids expressing recombinant proteins in fusion with the HA tag . The protoplast lysate was subjected to the biotin-switch assay and the presence of AC4-HA in the precipitate , detected by western blotting using an anti-HA antibody ( Fig 5D ) indicates that AC4 is S-acylated in planta . On the other hand , the evidence that AC4 ( C-A ) HA was not recovered from the neutravidin beads ( Fig 5D ) reveals that the mutant protein was not originally S-acylated and that C11 is essential for AC4 palmitoylation . Even though the site of accumulation of a protein does not necessarily correspond to its site of biological action , the evidence that AC4 is post-translationally modified to specifically target the cellular plasma membrane suggests that a protein function could be connected to palmitoylation . As AC4 acts as a VSR , we investigated the relation of silencing suppression function with PM localization . To this aim , we agroinfiltrated the PVX-GFP vector in combination with PVX-AC4 , PVX-AC4 ( C-A ) , PVX-AC4 ( KRR-AAA ) or PVX- AC4 ( C-A/KRR-AAA ) in N . benthamiana 16c plants . Starting from the onset of silencing , plants infiltrated with PVX-GFP in combination with the PVX-AC4 ( C-A ) and PVX- AC4 ( C-A/KRR-AAA ) showed the same systemic pattern as those infected with PVX-GFP alone ( Fig 6A , compare with Fig 3A ) . On the other hand , similarly to PVX-GFP/PVX-AC4 ( Fig 3A ) , plants inoculated with PVX-GFP/PVX-AC4 ( KRR-AAA ) appeared fluorescent consistently with absence of GFP silencing ( Fig 6A ) . From 12 dpi , only plants co-infected with PVX-AC4 or AC4 ( KRR-AAA ) showed suppression of systemic silencing whereas plants inoculated with the C-mutated proteins , appeared red-fluorescent under UV light ( Fig 6A ) . The GFP pattern remained unchanged even one month after agroinfiltration ( S5 Fig ) . RNA gel blot analysis confirmed that the persisting systemic expression of GFP in plants agroinfiltrated with PVX-GFP/PVX-AC4 ( KRR-AAA ) was the result of inhibition of VIGS , and in turns that the AC4 NLS is dispensable for silencing suppression ( Fig 6B and 6C ) . In fact , at 15 dpi the GFP siRNAs were detected in total RNA extracted from new leaves of plants agroinfiltrated with PVX-AC4 ( C-A ) and PVX- AC4 ( C-A/KRR-AAA ) but were absent in those infiltrated with PVX-AC4 ( KRR-AAA ) ( Fig 6B and 6C ) . Consistently , GFP mRNA levels observed in plants co-inoculated with PVX-AC4 and PVX-AC4 ( KRR-AAA ) were comparable with those in mock-inoculated plants ( Fig 6B and 6C ) confirming that the VSR function of AC4 is inhibited by mutation of C11 . These results strongly indicate that post-translational palmitoylation of AC4 and , consequent , PM localization are essential for efficient silencing suppression . Based on the evidence that the geminivirus TYLCV C4 targets PM and PD where interacts with BAM1 to inhibit intercellular spread of RNAi [36] , we investigated whether MYMV AC4 could also interact with BAM1 in N . benthamiana leaves . Indeed , we observed that BAM1 and AC4 co-localize at PM and in PD ( S6 Fig ) , and demonstrated their interaction by Bimolecular fluorescence complementation ( BiFC ) ( Fig 7A ) . The interaction between AC4 and BAM1 was further confirmed using Fӧrster resonance energy transfer–fluorescence lifetime imaging ( FRET-FLIM ) ( Fig 7B ) . These results convincingly support the evidence that AC4 requires PM localization for silencing suppression function . Collectively , our results indicate that AC4 undergoes a post-translational modification that mediates protein targeting to the PM . When localized to the PM , AC4 strongly suppresses systemic silencing whereas delocalization from such subcellular compartment impairs VSR activity . Furthermore , AC4 does not interfere with siRNA production and local PTGS and VIGS are not affected by the presence of the protein . To investigate whether MYMV VSR might interfere with the transport of silencing signal by sequestering siRNA , we tested the ability of MYMV AC4 to bind small RNAs in vitro by electrophoretic mobility shift assay . For this assay , we used purified viral protein expressed in fusion with the glutathione S-transferase ( GSTAC4 ) and gel purified GFP siRNAs produced upon PVX-GFP induced gene silencing in N . benthamiana 16c line . Upon combination with different concentration of GSTAC4 , we observed slower migration of siRNAs indicating the formation of a protein-siRNA complex which confirmed the ability of MYMV AC4 to bind native 21–25 nt siRNAs ( Fig 8 ) . Such ability is not lost upon mutation of the palmitoylated C in A . In fact , the GSTAC4 ( C-A ) mutant also binds siRNA , albeit probably less efficiently , as suggested by comparing the intensity of the siRNA fraction bound to equal amount of GSTAC4 and GSTAC4 ( C-A ) ( Fig 8 , lanes 3 and 5 second gel ) .
AC4 is among the least conserved proteins of all geminiviruses and appears to have divergent biological functions among species of the family , being mostly involved in virus-plant interactions and in pathogenesis [48–51] . It was shown to play a role in the regulation of cell division [52] whereas , in other species , mutagenesis and/or transgenic expression of AC4 has no consequence on infection of several host plant [53] . Aiming at gaining insights into the mechanism of pathogenesis of MYMV and , more in detail , into the way of action of AC4 , first and foremost , we have shown that infection of V . mungo with an AC4-deficient MYMV mutant develops asymptomatic phenotype , which reveals the essential role of AC4 for virus viability . Interestingly , this viral mutant replicates in inoculated leaves , albeit not at the same rate as the wild type MYMV and , opposite to MYMVΔBC1 , lacking movement function , it moves short distance cell–to-cell whereas systemic transport is fully hindered . Expression of MYMV AC4 increases severity of symptom induced by PVX in N . benthamiana , and suppresses systemic but not local GFP silencing in transgenic line 16c . Interestingly , we observed that AC4 targets the PM few hours post inoculation but shortly after , it starts to accumulate also into the nucleus via a typical NLS , and such specific double localization is maintained in the course of infection . The subcellular localization of AC4 is intriguing because normally , among plant viral proteins , only movement proteins localize to the PM . AC4 hosts no transmembrane domain but is post-translationally covalently modified by attachment of a lipid to the Cys11 that allows protein targeting and attachment to the PM . Such modification , known as S-acylation or palmitoylation and primarily meant to anchor otherwise soluble proteins to membranes , is now considered an important dynamic regulatory mechanism in signaling pathways in plants [25] . We found that the VSR activity of AC4 depends on protein binding to PM and is impaired upon mutation of Cys11 . MYMV hosts another strong silencing suppressor: the transactivator AC2 , which accumulates predominantly in the nucleus , excluding the nucleolus [12] . Opposite to AC2 that suppresses silencing in the nucleus , nuclear localization of AC4 is not related to this protein activity suggesting that the two MYMV VSR act in different cellular compartments and with different modalities . Therefore , the significance of AC4 targeting the nucleus remains to be investigated . Interestingly , we observed that PM is the first localization of MYMV AC4 while nuclear accumulation is visible later after transfection . This suggests that the protein could target the nucleus when accumulation to the PM attains saturation in a cycle of natural turnover for palmitoylated proteins [54 , 55] . Alternatively , activation/inhibition of palmitoylation could be a strategy to switch between different protein functions requiring distinct subcellular localization [30] . In fact , considering that some viral proteins evolved silencing suppression activity after or concomitantly with other functions essential for virus viability [9] , another role distinct from silencing suppression , which we proved to be uncoupled from nuclear localization , cannot be ruled out for AC4 . The absolute requirement of the Cys11 for silencing suppression activity reflects the need of such specific localization and justifies the highly stable association of AC4 to PM . The VSR function of East African cassava mosaic virus ( EACMV ) AC4 is also dependent on localization to the PM [18] . This protein is predicted to be N-myristoylated , and this modification is correlated to the VSR function . Proteins with the potential to become S-acylated often undergo myristoylation to interact with membranes and , subsequently , they become S-acylated and fixed to them [56] . Interestingly , EACMV AC4 hosts also a Cys in a favorable context for palmitoylation , whose mutation partially restricts the protein on perinuclear vesicles [18] . While palmitoylation of MYMV AC4 helps both binding and docking of the protein to the PM , in EACMV AC4 , the two actions could be mediated by palmitoylation and myristoylation , respectively . However , as the authors did not confirm experimentally the post-translational modification of the protein , it remains to be demonstrated the role of palmitoylation for the AC4 of this begomovirus . In Begomovirus infecting cassava , such as EACMV , African cassava mosaic virus ( ACMV ) and Indian cassava mosaic virus ( ICMV ) , both AC2 and AC4 VSR are functional and matched in a way that when AC2 is a strong suppressor its correspondent AC4 is a mild suppressor , and vice-versa [38] . Conversely , MYMV AC2 and AC4 are both strong VSR with distinct subcellular localization , and apparently both essential . The evidence that , despite the presence of functional AC2 , MYMV-ΔAC4 failed to establish systemic infection in V . mungo confirms that AC2 cannot compensate AC4 VSR function , and vice-versa [12] . While several examples of siRNA-sequestering VSRs as well as some movement proteins also acting as silencing suppressors are described in the literature [9] , to our best knowledge this is the first report of a siRNAs-binding VSR that absolutely requires PM localization to perform its function . AC4 is not a movement protein and the absolute requirement of PM localization for its silencing suppression activity is very interesting . TYLCV C4 also targets the PM and binds to BAM1 to hinder the spread of silencing signal triggered by this receptor-like kinase [36] . We demonstrate that MYMV AC4 also binds BAM1 and similar to C4 might hinder the silencing-related function of BAM1 . However , based on the experimental evidences collected in this study and particularly the capacity of MYMV AC4 ( unknown for TYLCV C4 ) to bind siRNAs and its much stronger VSR ability compared to C4 [57] , we hypothesize a different or additional mechanism of action for this protein . Upon targeting to the PM and particularly to PD , MYMV AC4 could bind siRNAs and stop their passage to the neighbouring cell thus suppressing the spread of the PTGS signal through the plant . However , further experimental in vivo evidence required to confirm this working hypothesis . The two nonpolar phenylalanines flanking Cys11 , might have the important role of creating the required hydrophobic environment to allow association of the hydrophilic AC4 to the PM and the insertion of palmitate into the double lipid layer of PM [47] . This specific amino acid context ( Phe-Cys-Phe ) suggests that AC4 could dock to the PM folded in shape of “V” where the bottom tip is occupied by the cysteine bond to palmitic acid inserted into PM . The two phenylalanines would provide hydrophobic stability to the bond whereas the hydrophilic tails of AC4 would be kept on the cytoplasmic side , away from the membrane and available for interactions with siRNAs . The evidence that the AC4 ( C-A ) mutant , missing the palmitoylated C , binds siRNA less stronger than AC4 WT , supports the hypothesis that the “Phe-Cys-Phe” hydrophobic domain might be important for conformational stability of the protein . Whether the AC4 interaction with BAM1 is functional in regulation of RNAi cell-to-cell spreading , and if this strategy is complementary or synergic with the siRNA sequestering capacity of AC4 is yet to be elucidated . Antiviral systemic signaling is a still unknown aspect of host defense and further validation is required to prove that plant immunity can be reached by systemic movement of vsiRNA . However , taken together the evidence here provided , it is tempting to speculate that MYMV AC4 would hijack the host lipidation machinery to target PD and , by binding vsiRNA , block the signal of “plant immunity” . The existence of different types of geminiviral VSRs , suggests that these proteins ( co ) -evolved to target different steps of the silencing pathway in a temporal and/or spatial manner . In the case of MYMV for example , AC4 might strategically be localized to PD to pose a physical barrier to the spread of silencing signal that could have escaped the suppression control of AC2 . As several suppressor proteins have multiple roles , including non-silencing functions critical for virus viability , and their synchronized action is essential in order to fulfill the multiple tasks , post translational modification could be an efficient strategy to reach this goal . In fact , S-acylation and/or N-myristoylation is predicted in AC4/C4 of the Geminiviridae species ( S2 Table ) , and phosphorylation , another mechanisms of post-translational modification , has been recently reported to regulate subcellular localization and in turn VSR activity of cucumber mosaic virus 2b protein [58] . Therefore , post-translational modification and its correlation to VSR function should be considered and investigated , particularly for those proteins with multifunctional behavior and potential localization to membrane compartment . In this study , we present the first report of palmitoylation and more in general of lipidation of a plant viral protein . The critical role of this post-translational modification on the function of MYMV AC4 suggests that lipidation is a very reliable way to target viral proteins to the membrane compartment and that more viral proteins might use these modifications for regulating their function at membranes .
All plasmids used in transient expression experiments are based on pCKGFP [59] , modified by replacing GFP with the EGFP coding sequence and by the addition of two restriction sites ( MluI and XbaI ) at the 3' end . The AC4 wt ORF was amplified by PCR from the pGA1 . 3A clone [37] with AC4-F and AC4-R primers ( S1 Table ) and the product was cloned into the MluI-XbaI sites of pCKEGFP in frame with EGFP . AC4 ( C11-A ) and AC4 ( KRR-AAA ) mutants were derived from pCKEGFPAC4wt using the QuickChange XL Site-Directed Mutagenesis Kit ( Agilent Technologies ) with AC4 ( C11-A ) F/AC4 ( C11-A ) R and AC4 ( KRR-AAA ) F/AC4 ( KRR-AAA ) R pairs of primers covering the aminoacids mutated , according to the manufacturer’s instructions . Plasmids containing mutated C and KRR were used as a template for site-directed mutagenesis to obtain the pCKEGFPAC4 ( C11-A /KRR-AAA ) . The AC4 sequence encoding aminoacids 1 through 12 was fused in frame with the 5’ of EGFP by amplifying EGFP with a forward primer containing the AC4 sequence fragment . Infectious clones based on the PVX genome , were obtained from the pGR107 plasmid . AC4 wt and mutant ORFs were amplified from the pCKEGFP clones with AC4 ( SmaI ) F and AC4 ( SalI ) R ( S1 Table ) whereas the p19 and the mGFP5 sequences were amplified from the pGA482p19 clone [41] with p19SmaI and p19SalI primers ( S1 Table ) and from DNA extracted from N . benthamiana 16c plants with mGFP5_F ( SmaI ) and mGFP5_R ( SalI ) primers ( S1 Table ) , respectively . PCR products were cloned in the SmaI-SalI sites of the multiple coning site of the pGR107 plasmid and the obtained plasmids introduced into Agrobacterium tumefaciens strain C58C1 by a freeze–thaw method . AC1 , AC4 and BC1 genes were knocked-out within the virus infectious clones pGA1 . 3A ( DNA A wt ) and pGA1 . 3B ( DNA B wt ) [37] using the QuickChange XL Site-Directed Mutagenesis Kit ( Agilent Technologies ) and the MYMV-AC1koF/ MYMV-AC1koR , MYMV-AC4koF/MYMV-AC4koR , and MYMV-BC1koF/MYMV-BC1koR pairs of primers ( S1 Table ) covering the aminoacids mutated , respectively . For expression and suppression of GFP RNA silencing , the binary plasmid pGA482p19 and a pCAMBIAAC4 plasmid based on pCAMBIA32 modified by cloning an expression cassette containing AC4 under the control of the 35S promoter in PstI restriction site , were introduced into Agrobacterium tumefaciens strain C58C1 by a freeze–thaw method . For the electrophoretic mobility shift assay , MYMV AC4 was amplified with the MYMVAC4 F/ MYMVAC4-HA R pair of primers ( S1 Table ) and cloned into the EcoRI/SalI sites of pGEX-6p-1 ( GE Healthcare ) in frame with the glutathione S-transferase ( GST ) coding sequence . The MYMVAC4-HA R reverse primer ( S1 Table ) contained the sequence encoding the 9 aminoacids of the HA epitope ( TAC CCA TAT GAC GTC CCA GAT TAC GCT encoding YPYDVPDYA ) . The third stop codon following the SalI site in the pGEX6p1 plasmid in frame with AC4 sequence was used for termination of translation . The HA ( human influenza hemagglutinin ) epitope tag was engineered onto the C- terminus of AC4 sequence so that the tagged protein could be analyzed and visualized using immunochemical methods . For BiFC and FRET-FLIM analysis , AC4 was cloned in pENTRD/TOPO ( Invitrogen ) using primers CACCATGAAGATGGAGAACCTCATCT and GTATATTGAGGGCCTGTAACTTG . Gateway cloning ( Invitrogen ) was used to fuse AC4 to GFP in pGWB505 [60] , to RFP in pB7RWG2 . 0 [61] , and nYFP/cYFP in pGTQL1211YN/pGTQL1221YC [62] . Construct to express C4-GFP , BAM1-RFP , PM-RFP ( Plasma Membrane protein NCBI number NP_564431 ) , C4-cYFP , BAM1-cYFP and BAM1-nYFP are described in Rosas-Diaz et al . , 2018 . Protoplasts were isolated from N . benthamiana and transfected as described [63] . V . mungo plants were biolistically transfected with clones to express AC4 WT and mutants and stained with DAPI . Fluorescent proteins were examined with a Nikon Eclipse 80i microscope equipped with video confocal technology ( VICO ) . For GFP , DsRed and DAPI images , the ET-GFP filter set ( Chroma 49002 , Nikon ) , the G-2A filter ( Nikon ) and DAPI filter were used , respectively . N . benthamiana plants were agroinfiltrated with clones to express BAM1-GFP and AC4-RFP and stained with aniline blue . Imaging was performed as described in Rosas-Diaz et al . , 2018 . Bimolecular fluorescent complementation ( BiFC ) assays were performed as described previously [36] . In brief , N . benthamiana plants were agroinfiltrated with clones to express the corresponding proteins , and samples were imaged two days later on a Leica TCS SMD FLCS confocal microscope , using the pre-set settings for YFP with Ex:514 nm , Em: 525–575 nm . FRET-FLIM experiments were performed as described previously [36] . N . benthamiana plants were grown at 25°C . At six-leaves stage plants were infiltrated with A . tumefaciens C58C1 harboring the appropriate constructs . A . tumefaciens carrying each construct was grown on selective media overnight , resuspended in the infiltration buffer ( 10 mM MES , 0 . 15 mM acetosyringone , 10 mM MgCl ) , kept at 25°C for 2-3h , and subsequently infiltrated into wt or 16c plant leaves at OD = 1 . In co-infiltration experiments , equal volumes/concentration of each suspension were mixed prior to infiltration . GFP fluorescence was observed under long-wavelength UV light ( Black Ray model B 100A , UV Products ) and photographed with a yellow filter . Leaf samples from MYMV bombarded V . mungo plants were collected and analysed in triplicates . DNA was extracted by Dellaporta method [64] with slight modification . 50 mg fresh leaf tissue was ground in liquid nitrogen , mixed with extraction buffer ( 50 mM Tris-HCl pH 8 . 0 , 20 mM EDTA pH 8 , 350 mM NaCl , 8 M Urea , 2% N-Lauril-Sarcosine ) and equal volume of phenol and incubated at 70°C for 5 min . DNA was extracted from the supernatant upon centrifugation by volume of phenol: chloroform ( 1:1 ) , isopropanol precipitation and RNase treatment . Two different couples of primers were designed: VrACtfor/ VrACtrev to amplify an endogenous actin gene ( Vigna radiata actin , accession number AF143208 ) and AC2-RT_F/ AC2-RT_R to amplify the target MYMV AC2 gene ( S1 Table ) . Relative qPCR was performed using C1000 thermal cycler ( Bio-Rad ) . The cycling profile consisted of 95°C for 20 s , 40 cycles of 3 s at 95°C and 30 s at 60°C , one cycle of 10 s at 95°C , as recommended by the manufacturer , using 2X Fast SYBR Green PCR Master Mix ( Applied Biosystems ) , 400 nM forward and reverse primers , 4 ng of V . mungo DNA and nuclease-free water in a total volume of 12 . 5 μL . Each DNA sample was amplified in duplicate for each primer pair and immediately after the final PCR cycle , a melting curve analysis was performed to determine the specificity of the reaction . Relative quantification was calculated using the comparative cycle threshold ( Ct ) method ( RQ = 2–ΔΔCt ) [65] , in which the change in the amount of the target viral RNA was normalized in relation to the endogenous control . Data were log-transformed [log ( x+1 ) ] before statistical analysis in order to fulfil the assumptions for parametric statistics . Transformed data were analyzed in a repeated measure factorial design using the MIXED procedure of SAS ( SAS/Stat Inc . ) , in which the variables ‘Experiment’ , ‘Treatment’ and ‘Time’ were considered as fixed effects . Contrasts were performed to test the experiment reproducibility , viz . the hypothesis of no difference between the two independent experiments made , and the difference among treatments . Total RNA was extracted from 100 mg of leaf tissue . Plant materials homogenized in liquid nitrogen was resuspended in 600 μl of extraction buffer ( 0 . 1 M Glycine-NaOH , pH 9 . 0 , 100 mM NaCl , 10 mM EDTA , 2% SDS ) and mixed with an equal volume of phenol . The aqueous phase was treated with equal volumes of phenol-chloroform , precipitated with ethanol , and finally resuspended in sterile water . RNA gel blot analysis of higher molecular weight RNAs was performed as previously described [41] . For analysis of siRNAs , low-molecular-weight RNAs ( LMW-RNAs ) were enriched from total RNAs extract by removing high-molecular-mass RNAs with 10 % polyethylene glycol ( PEG8000 ) and 1M NaCl . Approximately 5 μg of LMW RNAs were separated by 17 % PAGE with 7 M urea and then blotted onto Hybond-N+ membranes . After UV cross-linking , the membranes were hybridized at 42°C and the detection was carried out with DIG non-radioactive system ( Roche Applied Science ) according to the manufacturer’s instructions using a probe covering the entire mGFP5 sequence ( GenBank: U87973 ) . The blots were incubated in antibody solution , anti-DIG-AP Conjugate ( Roche ) and CDP-STAR ( Roche ) for chemiluminescence detection . Cell fractionation was performed as previously described with some modifications [66] [67] . Protoplasts transfected with pCKAC4HA plasmid were pelleted , resuspended in buffer ( 1X PBS ( pH 7 . 4 ) , protease inhibitors , 0 . 5% Tween ) and lysed by 5 freeze-and-thaw cycles . Cell debris ( P0 . 5 ) was isolated by 3 min centrifugation at 500 x g , 4°C . The supernatant ( S0 . 5 ) was further centrifuged at 30 , 000 x g and 4°C for 30 min to yield supernatant ( S30 ) and pellet fractions ( P30 ) . For analysis of the membrane part , the P30 fraction was incubated for 30 min on ice in the presence of one of the following reagents: 100 mM Na2CO3 ( pH 11 . 5 ) , 4 M urea , or 1 M KCl [68] . After centrifugation at 30 , 000 x g for 30 min at 4°C , pellets and supernatants were resolved on 12% SDS-PAGE , transferred to Hybond PVDF membrane ( Millipore ) and subjected to Western blot analysis . Proteins were detected with anti HA antibody ( 1:4000 , Santa Cruz Technology ) , visualized with SuperSignal West Pico Chemiluminescent Substrate ( Pierce ) according to the manufacturer’s instructions and scanned by Chemidoc Touch Imaging System ( BioRad ) . For biotin switch assay , collected protoplasts transfected with pCKAC4HA and pCKAC4 ( C-A ) HA were resuspended in 500 μL lysis buffer ( 1X PBS pH 7 . 4 , protease inhibitors , 1 mM EDTA , 1% Triton X-100 , 25 mM N-ethylmaleimide ) and treated following the method described [69] . Eluted proteins were analyzed by SDS-PAGE and Western blotting as described above . Recombinant GSTAC4 , GSTAC4 ( C-A ) , and GST proteins were produced by overexpression in Escherichia coli BL21 codon plus cells ( Agilent Technologies ) . Cells were grown to OD600 ≈ 0 . 7 and IPTG was added to a final concentration of 0 . 4 mM for AC4 and of 0 . 2 mM for GST for induction ( 3 h at 37°C ) . GST fusion protein supernatants obtained after bacterial lysis and centrifugation were purified with Glutathione Sepharose 4B beads ( GE Healthcare ) following the manufacturer’s instructions . For preparation of 21-25-nucleotide siRNA , 45 to 50 μg of total RNA were electrophoresed on 17% PAGE with 8 M urea followed by ethidium bromide staining in 1 × Tris-borate-EDTA . 21–25 nt siRNA fraction was cut and incubated in buffer 0 . 3 M NaCl , 0 . 1% SDS , overnight at 4°C with rocking . After a gentle centrifugation for 5 min at 2000g , the supernatant was transferred to a 50 mL tube . The crushed gel slice was incubated for a second elution in the same buffer with rocking . The gel residues were pelleted by centrifugation , and the two supernatants were precipitated together with ethanol . For binding assays , increasing amount of purified GSTAC4 and of GSTAC4 ( C-A ) , and siRNA ( 1 . 5 ug ) were mixed and incubated for 20 min at room temperature in binding buffer ( 20 mM Tris–HCl pH 8 , 5 mM MgCl2 , 50 mM KCl , 25 mM NaCl and 2 . 5 mM DTT , 0 . 02% Tween , 10% glycerol ) . Each sample contained 40U RNasin . The reaction was stopped by adding dyes , and loaded onto 8% native PAGE . The gel was transferred to Hybond-N+ membrane and after UV cross-linking , the membranes were hybridized as described above . | Plants have developed small RNA ( siRNA ) -mediated post-transcriptional gene silencing as a defense mechanism against viruses . In response , plant viruses encode viral suppressors of RNA silencing ( VSRs ) that can interfere with various steps of the silencing pathway . Mungbean yellow mosaic virus ( MYMV ) is a plant bipartite geminivirus responsible for a devastating disease in some areas of tropics and sub-tropics where its natural host , Vigna mungo , is a staple food crop . We discovered that the MYMV-encoded AC4 protein is a determinant of pathogenicity , binds native 21–25 nt siRNAs in vitro , and counteracts virus induced gene silencing by strongly suppressing the systemic phase of silencing but not the local production of siRNA . MYMV AC4 undergoes palmitoylation , a post-translational modification never reported before for plant viral proteins that mediates specific localization of the protein to the plasma membrane ( PM ) . Interestingly , palmitoylation and in turns PM localization is indispensable for direct VSR function of AC4 . AC4 also binds the PM-located BAM1 , a trigger of cell-to-cell spread of RNAi . Taken together our results suggest that AC4 has synergic mechanisms of action , based on the specific PM localization , to prevent spreading of antiviral RNAi silencing in not yet infected cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"plant",
"anatomy",
"rna",
"interference",
"gene",
"regulation",
"membrane",
"proteins",
"plant",
"science",
"plant",
"pathology",
"epigenetics",
"cellular",
"structures",
"and",
"organelles",
"small",
"interfering",
"rnas",
"genetic",
"interference",
"proteins",
"gene",
"expression",
"subcellular",
"localization",
"leaves",
"cell",
"membranes",
"biochemistry",
"rna",
"palmitoylation",
"plant",
"viral",
"pathogens",
"cell",
"biology",
"nucleic",
"acids",
"post-translational",
"modification",
"integral",
"membrane",
"proteins",
"plant",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna"
]
| 2018 | S‐acylation mediates Mungbean yellow mosaic virus AC4 localization to the plasma membrane and in turns gene silencing suppression |
It is widely accepted that transcriptional regulation of eukaryotic genes is intimately coupled to covalent modifications of the underlying chromatin template , and in certain cases the functional consequences of these modifications have been characterized . Here we present evidence that gene activation in the silent heterochromatin of the yeast Saccharomyces cerevisiae can occur in the context of little , if any , covalent histone modification . Using a SIR-regulated heat shock-inducible transgene , hsp82-2001 , and a natural drug-inducible subtelomeric gene , YFR057w , as models we demonstrate that substantial transcriptional induction ( >200-fold ) can occur in the context of restricted histone loss and negligible levels of H3K4 trimethylation , H3K36 trimethylation and H3K79 dimethylation , modifications commonly linked to transcription initiation and elongation . Heterochromatic gene activation can also occur with minimal H3 and H4 lysine acetylation and without replacement of H2A with the transcription-linked variant H2A . Z . Importantly , absence of histone modification does not stem from reduced transcriptional output , since hsp82-ΔTATA , a euchromatic promoter mutant lacking a TATA box and with threefold lower induced transcription than heterochromatic hsp82-2001 , is strongly hyperacetylated in response to heat shock . Consistent with negligible H3K79 dimethylation , dot1Δ cells lacking H3K79 methylase activity show unimpeded occupancy of RNA polymerase II within activated heterochromatic promoter and coding regions . Our results indicate that large increases in transcription can be observed in the virtual absence of histone modifications often thought necessary for gene activation .
Transcription in eukaryotes occurs in the context of chromatin . At most genes , transcriptional activation is accompanied by alterations to the chromatin template , exemplified by the enhanced DNase I sensitivity of coding regions and the presence of associated DNase I hypersensitive sites at linked regulatory elements [1] , [2] . DNase I hypersensitivity of promoter regions primarily arises from the partial or transient occupancy of nucleosomes ( i . e . , nucleosome depletion ) [3] , [4] , while the enhanced nuclease sensitivity of coding regions arises , in large part , from post-translational modification ( PTM ) of histones . Lysine acetylation of histones increases the accessibility of DNA wrapped around individual nucleosomes [5] , contributes to the decondensation of the 30 nm fiber [6] , [7] and facilitates nucleosomal displacement during elongation of RNA polymerase [8] . Covalent histone modifications can also create novel nucleosomal surfaces that serve as recognition sites for effector proteins . Histone PTMs , acting singly or in combination , may therefore control the ultimate expression state of a gene , or the ability of the underlying DNA to be repaired , recombined or replicated . This concept has been termed the histone code [9] , [10] . A prime example of this is the inducible INO1 promoter in the budding yeast S . cerevisiae where phosphorylation of the Ser10 residue of H3 ( H3S10 ) by the Snf1 kinase triggers acetylation of the Lys14 residue of H3 ( H3K14 ) by the Gcn5 acetyltransferase , and subsequent transcriptional activation [11] . Likewise , in mammalian cells , AMP-activated protein kinase ( AMPK ) -mediated phosphorylation of H2B S36 has been functionally linked to transcriptional activation of genes responsive to metabolic stress [12] . More recently , the combination of H3K4 trimethylation ( H3K4me3 ) and H4K16 acetylation ( H3K16ac ) was shown to serve as a high-affinity docking site for the human chromatin remodeling enzyme NURF [13] . Observations such as these have contributed to the widespread idea that histone PTMs play important roles in regulating gene transcription [14] , [15] . An alternative view is that histone modifications and the enzymes that impart them are not regulatory; that is , they do not play a causative role in transcription [16] . This arises from the fact that most enzymes that convey histone modifications have no specificity . Moreover , most histone PTMs are short-lived and do not persist in the absence of the proteins that recruit them [16] . Instead , according to this view , histone modifications play other roles . For example , they may assist in the fine-tuning of transcription or in its fidelity . They may also delineate one region of a gene ( e . g . , promoter ) from another ( e . g . , coding region ) , as well as from other genetic elements , thereby diversifying the chromatin landscape [17] . Therefore , histone modifications may occur as a consequence , rather than the cause , of dynamic processes such as transcription and nucleosome remodeling [17] . In addition to the precise role ( s ) played by covalent histone modifications , it is unclear whether the dynamic properties of euchromatin are shared with heterochromatin , the compartment of the nucleus that remains condensed throughout interphase , replicates late , is resistant to recombination , and contains relatively few transcribed genes . In multicellular organisms , heterochromatin can exist in both constitutive and facultative ( i . e . , regulated ) forms . Constitutive heterochromatin is enriched in HP-1 ( a structural protein ) and SU ( VAR ) 3–9 ( an H3K9 methyltransferase ) in organisms ranging from fission yeast to mammals , and is characteristic of telomeric and pericentric regions , repetitive DNA elements , and other DNA sequences critical to genomic stability . In contrast , facultative heterochromatin is characteristic of reversibly silenced genes such as X-linked genes in female mammals and genes encoding key developmental regulators , is enriched in H3K27me3- and ubiquitylated H2A-containing nucleosomes , and is under regulation of Polycomb chromatin modification complexes [18] , [19] , [20] . Budding yeast , although lacking HP-1 and Polycomb proteins , contains specialized chromatin structures that functionally resemble the Polycomb-regulated facultative heterochromatin of insects and vertebrates [21] . These heterochromatic domains are located at telomeres and the HML and HMR silent mating loci [reviewed in 22] , [23] . They are silenced via recruitment of a chromatin modification complex containing the Sir2 , Sir3 and Sir4 proteins that horizontally spreads over each telomere or HM locus . Absence of any one of the Sir proteins prevents the assembly of silent chromatin [24] , [25] , [26] . Sir3 and Sir4 are structural proteins [22] while Sir2 is a NAD+-dependent lysine deacetylase that deacetylates histones with a preference for H4K16 and H3K56 [27] . Sir2/Sir3/Sir4-mediated silent chromatin resembles the heterochromatin of other organisms in several ways: [i] it is repressive to gene transcription; [ii] it is organized into chromosomal domains that silence in a position-specific rather than sequence-specific fashion; [iii] its assembly involves entry sites that nucleate the formation and spread of repressor proteins; and [iv] the repressed expression state is mitotically inherited from mother to daughter cell [reviewed in 21] , [28] . While heterochromatin is generally repressive of transcription , hundreds of genes are localized within this nuclear compartment in protozoa , insects , plants and animals . The light gene , and at least eight others in Drosophila , depend on a heterochromatic location for normal expression [reviewed in 29] . Likewise , in placental mammals , expression of the Xist gene is 100-fold enhanced on the heterochromatic , inactive X chromosome relative to its euchromatic counterpart [reviewed in 30] . In trypanosomes , antigenic variation stems from variegated expression of telomere-silenced surface glycoprotein genes , a phenomenon underlying trypanosomiasis ( African sleeping sickness ) [reviewed in 31] . Despite the importance of heterochromatic genes , the mechanisms underlying their transcriptional activation remain largely unknown . To gain insight into this , we investigated chromatin alterations that accompany heterochromatic gene activation in S . cerevisiae . We found that large increases in the transcription of disparate heterochromatic genes occur in the absence ( or near absence ) of covalent histone modifications . Strikingly , when the same genes are placed in a euchromatic context , they heavily utilize such modifications .
To investigate changes in histone abundance and modification state that take place in activated heterochromatin , we took advantage of a previously described heat shock-inducible transgene system [32] , [33] . The system consists of the native HSP82 gene and chromosomal HSP82 alleles flanked by integrated copies of the HMRE mating-type silencer that differ in their dosage and arrangement ( illustrated in Figure 1A ) . As a consequence , the basal transcription of these transgenes is differentially silenced , from 3-fold for hsp82-201 bearing two upstream silencers to 30-fold for hsp82-2001 flanked by tandem silencers ( see Figure 1B , left ) . The transgene termed hsp82-1001 , flanked by single silencers , represents an intermediate case and is ∼6-fold silenced . As expected , Sir proteins occupy the promoter region of each transgene under non-heat-shock ( NHS ) conditions , and at levels that roughly correlate with the extent of silencing ( see Figure 2 , panels A and B for a Sir3 ChIP analysis; similar results were previously seen for Sir2 [24] ) . Sir3 was also observed within the coding regions of hsp82-1001 and hsp82-2001 but not within the ORF of the weakly silenced hsp82-201 gene . Indeed , the domain of silent chromatin at hsp82-1001 and hsp82-2001 spans at least 4 kb based on both ChIP and mRNA expression criteria ( Figure S1 ) , closely resembling that seen at the native HMR locus [26] , [34] . This observation is consistent with the idea that spread of the Sir2/3/4 complex from its site of recruitment is antagonized by the presence of enhancer and promoter sequences which serve as boundary elements [35] , [36] . In response to a 20 min heat shock ( HS ) , all three transgenes were strongly activated ( Figure 1B , right ) . Notably , despite >200-fold activation , Sir3 occupancy within the hsp82-2001 locus was only slightly reduced ( Figure 2 , panels A and B ) , contrasting with the less efficiently silenced transgenes where Sir3 localization was altered through either dispersal or dissociation ( Figures 2A and S1A , compare HS ( + ) samples with NHS ( − ) ) . Thus , transcriptional activation of the most efficiently silenced transgene occurred with minimal loss of the Sir2/3/4 complex at either promoter or coding region , consistent with earlier observations of the hsp82-2001 promoter and recent observations of an activated subtelomeric URA3 gene [24] , [37] . Disruption of silent chromatin through SIR4 deletion restored transcript levels of each transgene to WT levels ( Figure 1B , hsp82-1001 sir4Δ strain and data not shown ) . This indicates that cis-acting HMRE silencers ( and the sequence-specific proteins bound to them ) do not affect HSP82 regulation; assembly of the Sir protein complex is required . Consistent with this , the promoter chromatin structure of the three transgenes in a sir4Δ background is indistinguishable from the euchromatic HSP82 gene , based on nucleotide-resolution DNase I , MNase and dimethyl sulfate genomic footprint analyses [33] . We next investigated the effect of the Sir proteins on nucleosome density and stability . As previously seen for wild-type HSP82 [38] , histone H3 was rapidly depleted in response to heat shock of the euchromatic hsp82-2001 gene , exhibiting a 75% reduction within 60 sec and >90% reduction within 5 min ( Figure 2C , sir4Δ samples ) . Consistent with diminished transcriptional activity during chronic heat shock [39] , [40] , H3 levels were partially restored between 2 and 24 hr . Assembly of HSP82 into heterochromatin not only resulted in at least a twofold increase in nucleosomal density throughout the gene ( compare SIR+ with sir4Δ cells , 0 min ) , but also restricted the dynamic nature of the chromatin as the extent of nucleosomal disassembly was substantially reduced . Virtually identical results were seen when abundance of either myc-H4 or H2A was evaluated ( data not shown ) ; therefore , large ( >200-fold ) increases in expression can take place in the context of relatively modest changes in nucleosome occupancy , consistent with the retention of the Sir2/3/4 complex described above . Given the relatively static state of the chromatin , we next asked whether inducible occupancy of the Pol II transcriptional machinery could be detected at the heterochromatic hsp82-2001 transgene . Consistent with earlier observations of SIR-silenced genes [24] , [41] , [42] , [43] and reconstituted SIR-heterochromatin [44] , Pol II was detectable within the promoter region of non-induced hsp82-2001 , albeit at a reduced level ( Figure 3A , SIR+ , 0 min ) . Importantly , its occupancy was substantially enhanced ( 10- to 15-fold ) by heat shock . We additionally examined occupancy of the capping enzyme Cet1 . Previous analysis of the HMLα1/α2 and HMRa1 silent mating genes indicated that occupancy of Cet1 was strongly restricted under SIR-silencing conditions [42] , consistent with the notion that SIR elicits silencing , at least in part , by targeting steps downstream of PIC assembly . In support , we found that under non-inducing conditions Cet1 was at near-background levels at all three heterochromatic hsp82 transgenes ( Figure 3B ) . This restriction was dramatically overridden by heat shock , which resulted in a >30-fold increase in Cet1 occupancy of the 5′-end of hsp82-2001 and similar increases within the 5′-ends of hsp82-201 and hsp82-1001 . Therefore , at least two components of the basic transcriptional machinery , Pol II and Cet1 , dynamically occupy the hyperrepressed hsp82-2001 transgene in response to heat shock , in contrast to either histones or Sir proteins . Previous work has shown that euchromatic hsp82 alleles are dependent on the histone acetyltransferases Gcn5 and Esa1 for normal transcriptional activation [45] . Consistent with this , we found that HSP82+ activation was impaired in an H4 K16R mutant ( Figure 4A ) . We therefore asked whether acetylation of histones H3 and H4 , a hallmark of gene activation [14] , [46] , [47] , accompanies induction of the hsp82 transgenes . As shown in Figures 4B and 4C , nucleosomes occupying the euchromatic hsp82 transgene ( sir4Δ cells ) were both H3 di-acetylated and H4 tetra-acetylated under NHS ( − ) conditions , mimicking the wild-type HSP82 gene [38] . By contrast , nucleosomes within the SIR-silenced hsp82-2001 gene were negligibly acetylated under the same conditions . Thus , hsp82-2001 conforms to the general notion of a histone code , in which absence of N-tail histone acetylation codes for inactivation . The less efficiently silenced transgenes were also impoverished in both di-acetylated H3 and tetra-acetylated H4 , but to a lesser degree . Notably , upon heat shock ( + ) , the euchromatic transgene was enriched in both acetylated isoforms ( once again resembling HSP82+ ) yet there was no detectable enrichment of either at the hyperrepressed hsp82-2001 gene . Similarly , the moderately silenced hsp82-1001 transgene , despite >60-fold increase in expression in response to heat shock , showed no enrichment in acetylation . By contrast , the weakly silenced hsp82-201 gene was H3 hyperacetylated , resembling its euchromatic counterpart . The foregoing analysis indicates that the minimal acetylation state of heterochromatic hsp82-2001 remains unaltered following a 20 min heat shock . However , as transcript accumulation is evident within the first 60 sec ( see Figure 4D , SIR+ samples ) , it was possible that transient histone acetylation may have occurred . To address this , we assayed H3 and H4 acetylation of hsp82-2001 30 sec following an instantaneous heat shock , and at 15- to 30-sec intervals thereafter , to obtain ‘snapshots’ of the H3/H4 acetylation state within this gene . Despite a 30-fold increase in hsp82 transcript level within the first 60–90 sec of heat shock , neither di-acetylated H3 nor tetra-acetylated H4 was detectably increased ( Figures 4E and 4F; SIR+ background ) . This again contrasts with the euchromatic state where promoter-associated nucleosomes , already enriched in acetylated H3 and H4 molecules , showed a further ∼30% increase . We separately examined H4K16 acetylation , given that this modification is sufficient to inhibit formation of the compact 30 nm fiber in vitro [7] , and as previously mentioned , is a preferred Sir2 target . As above , there was no increase in H4K16 acetylation , even transiently , as this modification remained at background levels throughout the heat shock ( Figure 4G , SIR+ ) . This finding suggests that Sir2 may actively deacetylate H4K16 to suppress the extent of chromatin unfolding and , as a consequence , diminish gross transcriptional output . We also examined H4K12 acetylation , which has been implicated in telomeric heterochromatin formation and function in S . cerevisiae [48] . However , in contrast to euchromatic hsp82-2001 , the hyperrepressed hsp82-2001 transgene was assembled in nucleosomes containing only background levels of H4K12ac , with little if any enrichment upon heat shock ( Figure 5A ) . Thus , its role may be telomere-specific ( see below ) . Since the above results argue that heterochromatic transcription is uncoupled from histone N-terminal lysine acetylation , we asked whether acetylation of K56 , located within the globular core of H3 , or methylation of K36 , located at the junction between the H3 N-terminus and the globular domain , accompanies transcriptional activation . H3K56 acetylation is predicted to break a histone-DNA interaction , potentially destabilizing the nucleosome [49] and is a mark of Pol II elongation due to histone exchange [50] . Indeed , there exists a tight linkage between H3K56 acetylation and nucleosomal disassembly in vivo [51] . H3K56 acetylation is of additional interest , given that it is a target of Sir2 , and its deacetylation is important in the compaction of telomeric silent chromatin and concomitant repression of transcription [27] . SIR reduced H3K56ac levels at non-induced hsp82-2001 ∼2-fold ( Figure 6A ) . More strikingly , H3K56ac enrichment remained low following a 20 min heat shock time course , when transcript accumulation increased >200-fold . This static modification state contrasts with the euchromatic hsp82 gene where the already elevated H3K56ac levels exhibited a further 2 . 5 to 3-fold increase at both UAS and promoter regions ( Figure 6A , sir4Δ ) . H3K36me3 is a hallmark of Pol II elongation within gene coding regions [52] . Despite this , robust transcriptional activation of hsp82-2001 takes place in the context of minimal H3K36 trimethylation ( <5% of that seen in euchromatin; Figure 6B ) . Thus , similar to H3K56 acetylation , H3K36 methylation is largely suppressed during heterochromatic gene activation . We next addressed the role of H3 methylation at lysines 4 and 79 . H3K4 trimethylation is a broadly conserved covalent modification of active ( or potentially active ) gene promoters [52] where it can promote transcription by serving as a recognition site for a number of transcriptional co-regulators such as TFIID , SAGA and NuA3 [53] . Di-methylation of H3K79 , located within the globular core of H3 , is also linked with activation [54] and its presence correlates with enhanced access of transcription factors to chromatin [55] . Consistent with roles in expression , the euchromatic hsp82 gene is heavily modified with both methyl marks – H3K4me3 particularly within the promoter region and H3K79me2 particularly within the 3′-transcribed region ( Figure 6 , panels C and D ) . Strikingly , SIR strongly suppressed both H3K4me3 and H3K79me2 enrichment within the hsp82-2001 transgene throughout a heat shock time course . Moreover , deletion of DOT1 , which encodes the sole H3K79 methyltransferase in S . cerevisiae , had virtually no effect on either Pol II recruitment kinetics or occupancy levels throughout the hsp82-2001 gene ( Figure 6E ) . This indicates that H3K79 methylation is unnecessary for robust activation of heterochromatic hsp82-2001 , in contrast to observations of a telomeric URA3 gene which suggested that H3K79me1 and H3K79me2 are important in disrupting transcriptional silencing [37] ( see Discussion ) . These results , together with those described above , argue that a substantial increase in transcription can take place in the context of heightened nucleosomal density and minimal covalent histone modification . We next examined the role of the histone variant H2A . Z ( Htz1 in S . cerevisiae ) . Replacement of canonical H2A with H2A . Z at the +1 and or −1 nucleosome poises promoters for transcriptional activation as H2A . Z-containing nucleosomes are more susceptible to disassembly than canonical nucleosomes [56] , [57] . ChIP analysis revealed that H2A . Z is enriched within the euchromatic hsp82 promoter , consistent with prior findings [56] , and is preferentially evicted upon heat shock ( Figure 7A , hsp82-1001 sir4Δ , black bars ) . As expected , H2A . Z abundance was reduced at the SIR-repressed transgenes . At hsp82-201 , its abundance was reduced 40% at the promoter , while at hsp82-1001 , H2A . Z levels were altered only at the 3′-end . At both genes , promoter-associated H2A . Z was drastically reduced concomitant with transcriptional induction ( Figure 7A , ( + ) samples ) . Therefore , the promoter regions of these partially silenced transgenes , like their euchromatic counterparts , are assembled into H2A . Z-containing nucleosomes that are preferentially displaced in response to heat shock . In contrast , the hyperrepressed hsp82-2001 gene is nearly bereft of H2A . Z in non-induced cells ( ∼90% reduced within both promoter and coding region ) and this reduced level remained unchanged during activation . Consistent with this observation , deletion of the gene encoding H2A . Z did not impair hsp82-2001 mRNA induction , and may have even enhanced it ( Figure 7B , SIR+ htz1Δ samples ) . It could be argued that the near-absence of covalent histone modification at hsp82-2001 is a consequence of its lower transcriptional output relative to wild-type HSP82 . To address this possibility , we used an attenuated euchromatic hsp82 allele , hsp82-ΔTATA , that bears a 19 bp chromosomal substitution of the TATA box and surrounding region [38] . This mutation resulted in a 10-fold reduction in 20 min heat shock-induced transcript levels ( Figure 8A ) . By comparison , silent chromatin diminished the activated expression of hsp82-2001 only 2- to 3-fold ( e . g . , see Figures 1B , 7B ) . If absence of histone PTMs and other chromatin alterations stem from diminished gross transcriptional output , then it would be predicted that hsp82-ΔTATA chromatin would not be modified upon its activation . However , heat shock-activated hsp82-ΔTATA undergoes H4 displacement and substantial H3 acetylation . H4 depletion within the UAS/promoter of hsp82-ΔTATA was both rapid and extensive , closely resembling wild-type HSP82 , although there was little histone loss over the coding region ( Figure 8B ) , consistent with diminished expression . Moreover , acetylation of H3K18 , which is catalyzed by SAGA [58] and strongly correlates with transcriptional activation in euchromatin [59] , was particularly prominent within the UAS/promoter region of hsp82-ΔTATA ( Figure 8D ) . By contrast , the upstream region of heterochromatic hsp82-2001 was negligibly modified at H3K18 and with delayed kinetics ( Figure 8F; note difference in scale ) , whereas H3K18 acetylation at euchromatic hsp82-2001 resembled HSP82+ ( compare Figures 8C and 8E ) . These results argue that the Hsf1-activated heterochromatic hsp82-2001 gene is transcribed at a sufficiently high level to undergo nucleosomal disassembly and significant H3 acetylation over its promoter-proximal region , yet does not do so . The foregoing results indicate that transcriptional activation of heterochromatic hsp82 can occur largely in the absence of nucleosomal modifications that are characteristic of euchromatic genes . While a SIR-dependent chromatin structure is present at hsp82-2001 that , by several criteria , resembles the silent chromatin established at the natively silenced HM loci ( Figures 2 and S1A ) [24] , [33] , it is possible that the absence ( or near-absence ) of activating histone modifications is unique to Hsf1-regulated genes . Indeed , retention of epigenetic information at yeast heat shock genes may not be critical given the extent of histone loss that occurs within both regulatory and coding regions upon their activation ( Figure 2C ) [38] , [60] . To rule out an Hsf1-specific effect , we asked whether a natural subtelomeric gene could be activated in the absence of covalent histone modification and other chromatin alterations . Previous work has shown that the Sir2/3/4 complex extends ∼3 kb from the right telomere of chromosome VI [61] , and that expression of the subtelomeric gene , YFR057w , located ∼1 kb from the chromosomal tip , is under SIR regulation [62] . RT-qPCR demonstrates that this gene is efficiently silenced by SIR ( >100-fold; Figure 9A , 0 min ) . While the function of YFR057w is unknown , we reasoned that it might play a role in pleiotropic drug resistance due to the presence of a consensus DNA sequence for Stb5 ( www . yeastract . com ) , which forms a heterodimer with Pdr1 . Pdr1 is known to activate other genes involved in pleiotropic drug resistance [e . g . , 63] . Consistent with this idea , we found that heterochromatic YFR057w was strongly induced by exposure of cells to 200 µg/ml cycloheximide ( Figure 9A , black bars ) . Notably , euchromatic targets of Pdr1 such as PDR5 and SNQ2 were also induced by this novel regimen ( data not shown ) , one in which cells remained fully viable ( see Figure S2 ) . The substantial increase in YFR057w transcript abundance in the SIR+ strain ( ∼150-fold during first 50 min ) is unlikely to be a primary consequence of cycloheximide-mediated mRNA stabilization [64] given that its isogenic sir2Δ counterpart showed just a several-fold increase over the same time frame ( Figure 9A , white bars ) . Consistent with bona fide transcriptional induction , Pol II occupancy of the promoter and ORF markedly increased over the cycloheximide time course in both euchromatic and heterochromatic contexts ( Figure 9B ) . The presence of Pol II at the non-induced heterochromatic promoter is consistent with H3 ChIP-Seq data indicating the presence of a 60–80 bp nucleosome-free region upstream of the +1 nucleosome [65] . These striking observations permitted us to ask whether activation of YFR057w takes place in the presence of covalent histone modifications , as is the case for euchromatic genes , or in their absence , as is the case for the comparably repressed hsp82-2001 transgene . We found that induction of heterochromatic YFR057w occurred without histone displacement ( Figure S3 , A and B , black bars ) and in the context of minimal nucleosomal alterations including H3K18 and H4K16 acetylation ( <5% the level seen at the euchromatic gene ( sir2Δ; Figures 9C and S3C ) ; H3K4me2 , 3 and H3K79me2 ( <5% of sir2Δ context; Figure 9C ) ; H3K36me3 and H2A . Z deposition ( 20–40% of that seen in a sir2Δ context; Figure 9D ) ; and H3K56 acetylation ( 10–30% of that seen at the euchromatic gene with no detectable enrichment following induction; Figure 9E ) . Minimal H3K56ac at YFR057w is consistent with foregoing results with heterochromatic hsp82-2001 , which maintained a low level of H3K56ac and a comparatively high nucleosomal density during its activation , although it is unlike observations of a SIR-silenced HMRE-GAL10 transgene that suggested acetylation of this lysine is needed for Pol II elongation [43] . Finally , as mentioned above , H4K12 acetylation has been observed at telomeric heterochromatin in S . cerevisiae , where it has been implicated in several telomere-related processes including replication , recombination and basal transcription [48] . Indeed , we observed an enrichment of H4K12ac at heterochromatic YFR057w relative to other acetylation marks ( 20–40% of that seen at the euchromatic gene; Figure 5B ) . Nonetheless , H4K12ac levels remained low in response to cycloheximide induction , a circumstance in which YFR057w transcript levels increased several hundred-fold . This , coupled with the fact that heterochromatic hsp82-2001 contained only background levels of H4K12 acetylation ( Figure 5A ) , argues that this PTM is unlikely to play an important role in heterochromatic gene activation . We conclude that activation of the subtelomeric YFR057w gene , like that of hsp82-2001 , takes place in the context of minimal chromatin modification .
A widespread belief in the transcription/chromatin field is that covalent modification of nucleosomal histones is integral to the mechanism by which eukaryotes regulate gene expression [5] , [9]–[15] , [37] , [43] , [45] , [47]–[49] , [66] . Results presented here demonstrate that in contrast to this notion , histone modification of activated heterochromatic genes in the model eukaryote S . cerevisiae is minimal , and in certain instances undetectable , despite robust induction ( >200-fold over the basal silent state ) and dynamic occupancy of the transcriptional machinery . Activation with minimal chromatin modification is observed at genes regulated by disparate activators and is seen in distinct genetic backgrounds . It is not the consequence of insubstantial transcription , since the euchromatic gene , hsp82-ΔTATA , despite being expressed at a level of only 20–30% of heterochromatic hsp82-2001 , exhibits striking chromatin modification during its activation including promoter-proximal H3 hyperacetylation and histone displacement . Moreover , at least in the case of HSP82 , it is not due to a lack of a requirement for covalent histone modifications . Previous work has shown a clear dependence of euchromatic hsp82 alleles on the histone acetylases Gcn5 and Esa1 present in SAGA and NuA4 , respectively [45] . Consistent with this , both HSP82 and hsp82-ΔTATA are extensively acetylated upon heat shock activation and HSP82 expression is reduced in an H4 N-terminal tail mutant . A similar requirement for histone acetylation likely applies to the euchromatic YFR057w gene as its expression is reduced in an H4K16R mutant ( Figure S4 ) . The work presented here significantly extends earlier observations of the absence of tetra-acetylated H4 enrichment following heat shock activation of the hsp82-2001 transgene [24] in many ways . These include analysis of the coding region of this and other heat shock transgenes , analysis of multiple covalent histone modifications typically linked to gene activation , and most significantly extending the finding to a natural heterochromatic gene , YFR057w , whose inducibility was heretofore unknown and whose activation is unlikely to be under the control of Hsf1 . It also substantially extends recent observations of a telomere-linked URA3 gene that like HSP82 is able to activate in absence of H3 and H4 acetylation [37] ( see below ) . It is possible that absence of detectable covalent marks is a consequence of their transience . We examined histone PTMs at multiple time points during the transcriptional induction of two unrelated genes , including time points spaced as tightly as 15 sec apart , and could find no compelling evidence for their presence . Nonetheless , our experiments cannot rule out the presence of covalent histone modifications that are erased ( or occluded , for example , by “reader” molecules ) moments after they appear . However , our inability to detect PTMs is unlikely to be due to masking of stable modifications . We utilized antibodies specific to multiple epitopes within the histone octamer – globular domain of H3 , N-terminal epitope of Myc-H4 , acidic patch of H2A – and all were readily detectable in SIR-mediated heterochromatin . In addition to the absence of covalent histone modification , the heterochromatic hsp82-2001 gene is only partially and transiently depleted of the H3/H4 tetramer during heat shock . This is in contrast to euchromatic hsp82-2001 that sustains a >90% loss of H2A , H3 and H4 during the early stages ( 5–25 min ) of heat shock induction ( Figure 2 and data not shown ) . Interestingly , even under maximally inducing conditions , the density of H3 and H4 at heterochromatic hsp82-2001 equals or exceeds its density at the non-induced euchromatic transgene . The continued presence of the Sir complex during heat-induced activation is likely crucial for suppression of both histone loss and covalent histone modifications . Regarding PTMs , in the case of H3/H4 acetylation , the Sir2 deacetylase apparently “wins” the competition with the SAGA and NuA4 acetylases for the chromatin template . Why H3 methylation does not occur is less clear , and is the subject of ongoing investigation . It is possible that the minimal changes in histone modification state that are detected at silent hsp82-2001 and YFR057w are sufficient to serve as novel surfaces to which bromodomain- , chromodomain- and PHD-domain-containing regulatory complexes may bind . We believe that this is unlikely , even in the extreme case in which it is assumed that the euchromatic counterparts of these genes are saturated with acetylated or methylated histone isoforms at their maximum point of enrichment . Since in most cases there is a >10-fold difference in modification levels in euchromatic ( sir2Δ or sir4Δ ) versus heterochromatic ( SIR+ ) states , modification of the heterochromatic gene is maximally one PTM ( of a given type ) for every 5–10 nucleosomes . Since the upstream regulatory region of HSP82 spans only two nucleosomes [67] , [68] , this means that <1 PTM of a specific type is present within the heterochromatic hsp82-2001 promoter . This , combined with the fact that there is no evidence for epigenetic variegation of hsp82-2001 expression ( at least when tested under non-inducing conditions [24] ) , argues against the existence of a subpopulation of cells enriched for activating histone marks and disproportionately contributing to the HSP82 transcript measured in our assays . Instead , our results suggest a dominant role for gene-specific activators ( Hsf1 in the case of hsp82-2001 and Pdr1/Stb5 in the case of YFR057w ) in recruiting transcriptional co-regulators [69] that disrupt SIR-mediated silencing . Our findings also provide an interesting contrast to those of Grunstein , Carey and coworkers [37] . These workers examined the chromatin properties of a telomere-linked URA3 transgene under control of the Sir2/3/4 chromatin modification complex , both in cell populations in which it was expressed ( cells grown in synthetic medium lacking uracil ) , as well as those in which it was repressed ( cells grown in rich medium containing 5-FOA ) . As was the case here , the expressed state was characterized by substantial levels of Sir3 and deacetylated H3 and H4 . In contrast to our findings , however , other histone modifications – in particular , H3K4me3 , H3K36me3 and both mono- and di-methylated forms of H3K79 – were abundant [37] . Based on these and other lines of evidence , the authors speculated that the H3K79me2 mark , while not diminishing the abundance of Sir3 at the expressed URA3-TEL gene , disrupted its physiological interaction with the underlying nucleosomes , thereby accounting for the transcriptionally permissive template [37] . As our data indicate that both hsp82-2001 and YFR057w can be transcribed essentially in the absence of H3K4 , H3K36 and H3K79 methylation , an obligatory role for these histone modifications in promoting Pol II transcription of heterochromatic genes seems unlikely . Instead , our results are more consistent with a large scale mutational analysis in S . cerevisiae which showed that despite the widespread localization of H3K4me , H3K36me and H3K79me marks , deletion of the enzymes responsible for imparting these marks – Set1 , Set2 and Dot1 – had very specific effects with the expression of most genes unaffected [70] . Also congruent with our results are observations of Drosophila mutants in which essentially normal transcriptional regulation can occur in the complete absence of H3K4 methylation [71] . We have demonstrated that in yeast heterochromatin , substantial increases in transcription can take place in the absence ( or near-absence ) of chromatin alterations often thought necessary for activation . Nonetheless , gross transcriptional output of hsp82-2001 and YFR057w is reduced several-fold as a consequence of their location within SIR-heterochromatin . Therefore , at least in these two cases , histone modifications may play a role in increasing transcriptional output . Other explanations for how Sir2/3/4 proteins reduce transcription of these genes , including direct or indirect roles in enhancing nucleosome stability [44] , are also possible . Whatever the physiological role of histone modifications might be , it is notable that similar to the examples presented here , heat stress-induced activation of heterochromatic transgenes in Arabidopsis occurs without reversal of repressive chromatin marks such as DNA methylation and H3K9 and H3K27 methylation , and without the appearance of activating modifications such as H3 and H4 acetylation [72] . Therefore , absence of activating histone modifications may be a common feature of stress-induced heterochromatic transcription .
Strains used in this study are listed in Table 1 . The hsp82 transgenic strains ( SLY101 background ) were generated by integrating HMRE silencers ( 144 bp modules [separated by 6 bp spacers in the case of tandem integrants] ) both 5′ and 3′ of the chromosomal HSP82 gene using two-step gene transplacement methods [33] . To permit expression of Myc-tagged histone H4 , strains were transformed with an episomal myc-HHF2 gene borne on plasmid pNOY436 ( TRP1-CEN6-ARS4 ) as previously described [38] . Strains MSY529 and MSY541 were gifts of M . M . Smith ( University of Virginia ) . Yeast strains were cultivated at 30°C to early log phase ( A600 = 0 . 3 to 0 . 7 ) in either rich yeast extract-peptone-glucose broth supplemented with 0 . 03 mg/ml adenine or synthetic complete medium lacking tryptophan ( SDC-Trp ) . Heat shock induction was achieved by transferring the culture ( typically 50 ml ) to a vigorously shaking 39°C water bath; once the temperature reached 39°C , incubation was allowed to continue for an additional 20 min before addition of either sodium azide ( to a 20 mM final concentration ( RNA assays ) ) or formaldehyde ( to a 1% final concentration ( ChIP ) ) . For time course assays , instantaneous 30° to 39°C upshift was achieved by rapidly mixing equal volumes of 30°C culture and pre-warmed medium ( 55°C ) and then incubating with rapid shaking at 39°C for the times indicated . Induction was terminated through addition of sodium azide . For drug induction , 100 ml early log cultures ( A600 ∼0 . 6 ) were made 200 µg/ml in cycloheximide through a 1∶100 dilution of a 20 mg/ml stock solution prepared in DMSO . Cells were then incubated with agitation at 30°C for various times ( indicated in Figure legends ) prior to addition of 20 mM sodium azide . Cells were then harvested , washed in 20 mM azide , and stored at −80°C for subsequent chromatin or RNA isolation . End point ( gel-based ) ChIP-PCR analysis ( Figures 2A , 3B , 4B , 4C , 4E–G , 7A , S1A ) was conducted essentially as described [42] . Briefly , 50 ml of mid-log culture ( A600 ∼0 . 5 ) were cross-linked with 1% formaldehyde for 10 min , then converted to spheroplasts with lyticase ( 4 mg/ml; Sigma ) . Spheroplasts were lysed using one volume of 0 . 5 mm glass beads for 30 min at 4°C on an Eppendorf 5432 mixer . Chromatin was sheared to a mean size of ∼0 . 5 kb with a Branson 250 sonifier equipped with a microtip using three 25 s pulses at constant power and an output setting of 22 watts . The clarified supernatant ( final volume 3 . 0 ml ) was used in immunoprecipitations ( IPs ) that were typically achieved by adding 2–5 µl antiserum to 300 µl of chromatin lysate . Signal quantification was done using a Storm 860 PhosphorImager ( Molecular Dynamics ) and ImageQuant 5 . 2 software . To calculate the relative abundance of a given gene sequence present in an IP , we used the following formula: Qgene = IPgene/Inputgene . In most cases , the abundance of each test locus was expressed relative to that of either PHO5 or ARS504 , which served as internal recovery controls . In the case of Sir3 , abundance at a given locus was quantified relative to its abundance at HMRa1 . To reduce background , we subtracted the signal arising from a mock IP ( -Ab ) for the histone covalent modification , Htz1 and Cet1 ChIPs , and the signal arising from pre-immune serum for the Pol II ( Rpb1 ) ChIPs . For the nonspecific ARS504 IP signal , only gel background was subtracted . Real Time ( ChIP-qPCR ) analysis ( Figures 2B , 2C , 3A , 5 , 6 , 8[B–F] , 9[B–E] , S3 ) was conducted as follows . Briefly , 125 ml of mid-log cell culture were used and 25 ml aliquots were removed for each time point . Two ml crosslinked chromatin were obtained from each , and ∼10% of that ( 200–250 µl ) was employed for each IP . For all ChIP-qPCR assays , chromatin was isolated as above except cells were lysed with glass beads in the presence of 1% Triton X-100 and 0 . 1% sodium deoxycholate . Immunoprecipitations were conducted through addition of 40 µl of a 50% slurry of CL-4B Protein A Sepharose beads , with incubation at 4°C overnight . Following immunoprecipitation , DNA was purified and dissolved in 60 µl TE; 2 µl of immunoprecipitated DNA added to each 20 µl Real Time PCR reaction . This was performed on an Applied Biosystems 7900HT Real-Time PCR system using RT2 qPCR SYBR Green/ROX MasterMix ( SABiosciences; #330529 ) . Through use of a standard curve specific for each amplicon , the quantity of DNA present in each IP was determined , and background signal was subtracted . In the case of Myc ChIPs , the background was the signal arising from chromatin isolated in a parallel culture of an isogenic strain lacking Myc-tagged H4; for histone PTM ChIPs , background was the signal arising from a beads alone control . For Pol II ChIP , background was the signal arising from pre-immune serum . To normalize for variation in sample recovery , abundance of a non-transcribed region on chromosome V ( ARS504 ) was determined for each ChIP DNA sample , and HSP82/ARS504 and YFR057w/ARS504 quotients were derived . In certain cases ( see figure legends ) , normalization to the PHO5 promoter was done instead . Finally , to account for nucleosome loss , all PTM data are presented as ( histone PTM ) /Myc-H4 or ( histone PTM ) /H3 quotients . The following antibodies were used: Myc ( MAb 9E10 , Santa Cruz Biotechnology ) ; H3 globular domain ( ab1791 , Abcam ) ; H3 K9ac , K14ac ( 12-360; Millipore ) ; H3 K18ac ( ab1191; Abcam ) ; H3 K56ac ( gift of M . Grunstein , UCLA ) ; H4 K16ac ( 07-329 , Millipore ) ; H4 K12ac ( 07-595 , Millipore ) ; H4 tetra-acetylated ( K5 , K8 , K12 , K16 ) ( 06-866 , Millipore ) ; H2A acidic patch ( residues 88-97; 07-146 , Millipore ) ; H3 K4me3 ( ab8580; Abcam ) ; H3 K4me2 , 3 ( ab6000; Abcam ) ; H3 K36me3 ( ab9050; Abcam ) ; H3 K79me2 ( ab3594; Abcam ) ; Htz1 ( residues 1–100; ab4626; Abcam ) ; Cet1 ( gift of S . Buratowski , Harvard Medical School ) ; Sir3 ( gift of R . T . Kamakaka , University of California , Santa Cruz ) ; Pol II , rabbit antiserum raised in our laboratory against a recombinant GST-CTD polypeptide bearing 52 heptad repeats of the mouse large subunit [38] . Amplicons for endpoint PCR were as follows ( coordinates relative to ATG ) : HSP82 promoter , −401 to −34; HSP82 ORF , +1248 to +1444; HSP82 3′-UTR , +1883 to +2155; ARS504 , coordinates 9746 to 9817; PHO5 promoter , −507 to −33; CIN2 5′ ORF , +149 to +427; CIN2 3′-UTR , +523 to +797; YAR1 ORF , +3 to +200; YAR1 3′-UTR , +1109 to +1297 ) ; IQG1 5′ ORF , +1016 to +1211; IQG1 3′ ORF , +2007 to +2289; SUI3 , +451 to +748; YPL236c , +71 to +229; HMRa1 , −80 to +50 . For Real Time PCR , the amplicons were: HSP82 UAS , −227 to −140; HSP82 promoter , −157 to −88 or −227 to −88 ( Figures 2B , 2C , 6C–E ) ; HSP82 ORF , +1248 to +1444; HSP82 3′-UTR , +2134 to +2228; YFR057w promoter , −115 to −45; YFR057w ORF , +312 to +437; ARS504 , coordinates 9746 to 9817; PHO5 promoter , −197 to −124; PMA1 5′-coding region , +49 to +112 . For the expression analyses summarized in Figure 1B and illustrated in Figures 4D and S1[B-E] , RNA was isolated from 10 ml aliquots of cell culture employed for ChIP assays using the glass bead lysis method [73] , and blots were hybridized to a gene-specific probe , exposed to PhosphorImager , then re-hybridized to an ACT1 probe as done previously [42] . Probes used were as follows: HSP82 , +2167 to +2228; CIN2 , +149 to +427; YAR1 , +3 to +200; SUI3 , +451 to +748; IQG1 , +1016 to +1211; and ACT1 , +606 to +1000 . For the expression analyses illustrated in Figures 4A , 7B and 8A , cells were cultivated to an A600 of ∼0 . 5 in a 125 ml culture at 30°C and 15 ml aliquots were removed and subjected to an instantaneous 30 to 39°C upshift for the indicated times . Heat shock induction was terminated through addition of 20 mM sodium azide , and RNA was isolated as above . For induction of YFR057w ( Figures 9A , S4 ) , cycloheximide was added to a final concentration of 0 . 2 mg/ml , 15 ml aliquots were removed at the indicated times and RNA was isolated . Contaminating genomic DNA was removed from each RNA sample by digestion with RNase-free DNase I ( OMEGA Bio-tek Inc #E1091 ) , followed by phenol/chloroform extraction . 0 . 5 µg purified RNA was used in each cDNA synthesis with ProtoScript RT-PCR Kit ( NEB #E6400S ) . Oligo ( dT ) primers were used in cDNA synthesis for quantification of Pol II gene transcripts; random primers were used in cDNA synthesis for quantification of SCR1 RNA . 2% of the synthesized cDNA was used in each qPCR , which was performed as described above . Primers were designed to target the 3′UTR of HSP82 , YFR057w and PMA1 , or the body of SCR1 . Their coordinates are: HSP82 , +2134 to +2228; YFR057w , +312 to +437; PMA1 , +2998 to +3083 and SCR1 , +385 to +483 . For quantification , SCR1 was used to normalize HSP82 and YFR057w mRNA levels . | The proper regulation of gene expression is of fundamental importance in the maintenance of normal growth and development . Misregulation of genes can lead to such outcomes as cancer , diabetes and neurodegenerative disease . A key step in gene regulation occurs during the transcription of the chromosomal DNA into messenger RNA by the enzyme , RNA polymerase II . Histones are small , positively charged proteins that package genomic DNA into arrays of bead-like particles termed nucleosomes , the principal components of chromatin . Increasing evidence suggests that nucleosomal histones play an active role in regulating transcription , and that this is derived in part from reversible chemical ( “covalent” ) modifications that take place on their amino acids . These histone modifications create novel surfaces on nucleosomes that can serve as docking sites for other proteins that control a gene's expression state . In this study we present evidence that contrary to the general case , covalent modifications typically associated with transcription are minimally used by genes embedded in a specialized , condensed chromatin structure termed heterochromatin in the model organism baker's yeast . Our observations are significant , for they suggest that gene transcription can occur in a living cell in the virtual absence of covalent modification of the chromatin template . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"biology",
"chromosome",
"biology",
"gene",
"expression",
"genetics",
"biology",
"and",
"life",
"sciences",
"epigenetics",
"chromatin",
"dna",
"transcription"
]
| 2014 | Uncoupling Transcription from Covalent Histone Modification |
Many cellular processes pertinent for viral infection are regulated by the addition of small ubiquitin-like modifiers ( SUMO ) to key regulatory proteins , making SUMOylation an important mechanism by which viruses can commandeer cellular pathways . Epstein-Barr virus ( EBV ) is a master at manipulating of cellular processes , which enables life-long infection but can also lead to the induction of a variety of EBV-associated cancers . To identify new mechanisms by which EBV proteins alter cells , we screened a library of 51 EBV proteins for global effects on cellular SUMO1 and SUMO2 modifications ( SUMOylation ) , identifying several proteins not previously known to manipulate this pathway . One EBV protein ( BRLF1 ) globally induced the loss of SUMOylated proteins , in a proteasome-dependent manner , as well as the loss of promeylocytic leukemia nuclear bodies . However , unlike its homologue ( Rta ) in Kaposi’s sarcoma associated herpesvirus , it did not appear to have ubiquitin ligase activity . In addition we identified the EBV SM protein as globally upregulating SUMOylation and showed that this activity was conserved in its homologues in herpes simplex virus 1 ( HSV1 UL54/ICP27 ) and cytomegalovirus ( CMV UL69 ) . All three viral homologues were shown to bind SUMO and Ubc9 and to have E3 SUMO ligase activity in a purified system . These are the first SUMO E3 ligases discovered for EBV , HSV1 and CMV . Interestingly the homologues had different specificities for SUMO1 and SUMO2 , with SM and UL69 preferentially binding SUMO1 and inducing SUMO1 modifications , and UL54 preferentially binding SUMO2 and inducing SUMO2 modifications . The results provide new insights into the function of this family of conserved herpesvirus proteins , and the conservation of this SUMO E3 ligase activity across diverse herpesviruses suggests the importance of this activity for herpesvirus infections .
The functions of many cellular and viral proteins are controlled by the addition of small ubiquitin-like modifiers ( SUMO ) , in the form of SUMO1 or SUMO2 or SUMO3 chains ( referred to as SUMOylation ) . These modifications can affect protein stability or localization and can promote protein-protein interactions via binding of SUMO to SUMO-interacting sequences ( SIMs ) [1 , 2] . SUMOylation controls many nuclear processes , including genome stability , gene expression , cell cycle progression , senescence and stress and innate immune responses [1 , 3–7] . Not surprisingly based on these roles , SUMO signal transduction has been identified as a key factor in the development of several types of cancer; SUMOylation is highly upregulated in many cancers and some cancers have been shown to be dependent on a functioning SUMO system [1 , 4] DNA viruses manipulate many of the cellular processes regulated by SUMOylation and therefore SUMO pathways provide a mechanism to alter these processes . Several different mechanisms have been described by which viral proteins usurp the SUMO system . Cellular SUMOylation involves the SAE1/SAE2 E1 SUMO-activating enzyme , the Ubc9 E2 SUMO-conjugating enzyme and several different SUMO E3 ligases that mediate the interaction of charged Ubc9 with the target protein , facilitating the SUMO transfer [2] . Some viral proteins globally alter SUMOylation by hijacking Ubc9 [8] . For example , the E6 protein of human papillomavirus ( HPV ) type 16/18 lowers Ubc9 levels thereby globally decreasing SUMOylation [9] . Some viruses upregulate SUMOylation by encoding E3 SUMO ligases that function in conjunction with Ubc9 . Two adenovirus proteins have been reported to have SUMO ligase activity; adenovirus E1B-55K induces SUMO1 modification of p53 [10 , 11] , while adenovirus E4-ORF3 induces SUMO3 modification of the transcription intermediary factor 1γ ( TIF-1γ ) [12] . In addition , the K-bZIP protein of Kaposi’s sarcoma-associated herpesvirus ( KSHV ) was found to be a SUMO2/3-specific E3 ligase that modifies itself as well as p53 and Rb [13] . Viruses can also downregulate SUMOylation by encoding SUMO-specific proteases ( SENPs ) or SUMO-targeted ubiquitin ligases ( STUbLs ) ; the latter which ubiquitylates SUMO-modified proteins leading to their degradation . Both human adenoviruses and vaccinia virus encode a SENP [14] . In addition , two distinct viral STUbLs have been identified; the ICP0 protein of herpes simplex virus 1 ( HSV-1 ) and its homologue ( ORF61 ) in varicella zoster virus , and the unrelated Rta protein of KSHV and its homologue in murine gammaherpesvirus 68 [15–19] . Like the cellular STUbL RNF4 [20] , a major target of degradation of the viral STUbLs are promyelocytic leukemia ( PML ) and associated proteins since they are highly SUMO-modified [15 , 16 , 21] . Since PML nuclear bodies are part of the innate immune response that senses herpesvirus genomes and suppresses their lytic infection , these STUbLs have important roles in promoting lytic infection [16 , 22 , 23] . EBV is a wide-spread herpes virus that is a causative agent of several types of cancer , including Burkitt’s lymphoma , nasopharyngeal carcinoma and 10% of gastric carcinoma [24–26] . Life-long infection occurs due to the ability of EBV to alternate between latent modes of infection , with restricted viral gene expression , and lytic infection involving expression of an additional ~70 viral proteins . Together , these lytic proteins manipulate many cellular processes including innate immune responses , cell cycle progression and DNA damage responses , in order to promote cell survival and virion production [27 , 28] . However , the functions of many EBV lytic proteins are unknown and even proteins with an assigned function may have additional unidentified roles . Given the importance of SUMO pathways in oncogenesis and cellular processes manipulated by EBV , it seems likely that some EBV proteins may function by targeting SUMO pathways . Little is currently known about the interplay between EBV and SUMO systems . To date only one EBV protein , the latent membrane protein 1 ( LMP1 ) , has been shown to act directly on SUMOylation . LMP1 binds to Ubc9 and upregulates its activity , resulting in increased SUMOylation in EBV latent infection [29 , 30] . This includes SUMOylation of KRAB-associated protein 1 ( KAP1 ) , which promotes its repression of the lytic immediate early promoters and lytic replication origin , thereby promoting latency [31] . In lytic infection , SUMO2/3 conjugates have been found to accumulate late in infection , which in part may be due to expression of a viral miRNA that downregulates RNF4 [32] . Presently no SUMO ligases , SENPs or STUbLs have been identified for EBV . To investigate how EBV proteins impact cellular pathways , we previously generated an expression library for most of the EBV lytic proteins , and used the C-terminal FLAG tag on each protein to determine their subcellular localization [33] . These EBV proteins were then screened for the ability to disrupt or alter PML nuclear bodies [33] , contribute to cell cycle arrest at the G1/S interface [28] and inhibit the cellular DNA damage response [34]; characteristics typical of EBV lytic infection . These screens have led to new functions for several EBV proteins demonstrating the utility of the approach . Here we screened the library of EBV proteins for those that globally affect cellular SUMO1 and SUMO2 modifications and identified a few EBV proteins that upregulate SUMOylation and one that downregulates SUMOylation . The downregulator is the Rta homologue of the KSHV and gammaherpesvirus 68 ( γHV68 ) STUbLs , suggesting a conserved role for these protein in decreasing SUMOylation . We show that one of the SUMO upregulators ( SM ) has characteristics of a SUMO E3 ligase and that this activity is conserved in SM homologues in HSV1 ( UL54/ICP27 ) and human cytomegalovirus ( CMV; UL69 ) , identifying the first SUMO E3 ligase for any of these herpesviruses .
To identify EBV proteins that hijack cellular SUMOylation , we screened a library of 51 EBV lytic proteins for the ability to globally affect SUMO1 and SUMO2 modifications upon overexpression . Such a screen can identify proteins with intrinsic SUMOylation or SUMO degradation activities or that have prominent interactions with SUMO pathway proteins [15 , 35–37] . Initial screens were performed in both 293T cells transiently expressing His6-tagged SUMO1 or SUMO2 and in HeLa cells containing integrated copies of His6-SUMO1 or His6-SUMO2 that express these proteins at close to endogenous levels [38] . In both cases , FLAG-tagged EBV proteins were expressed by transient transfection and , 36 hrs later , His-tagged proteins were recovered from cell lysates on metal chelating resin under denaturing conditions . Western blots were then performed using SUMO1 or SUMO2 antibodies to detect all proteins covalently modified by His-SUMO . The SUMOylation profile in the presence of each viral protein was compared to that with empty FLAG-tagged plasmid . In addition , we used the EBV LMP1 protein as a positive control for global upregulation of SUMOylation and the HSV1 ICP0 protein as a positive control for global downregulation of SUMOylated proteins . Examples of these controls and selected EBV proteins are shown in Fig 1 ( see S1 Fig for confirmation of ICP0 and LMP1 expression ) . The up or down regulation of global SUMOylation was assessed by comparing the intensity of the ladder of SUMOylated proteins in the presence of the EBV protein to that of the empty plasmid , and results for all 51 EBV proteins in the two cell systems are summarized in Table 1 ( with the degree of the effect indicated by the number of + signs for upregulation or– signs for downregulation ) . For proteins that appeared to affect global SUMOylation in 293T and HeLa cells , assays with transient expression of His6-SUMO1 or His6-SUMO2 were repeated in a nasopharyngeal carcinoma cell line ( CNE2Z ) relevant for EBV infection ( Fig 1C and Table 1 ) . We then looked for EBV proteins that consistently up or down regulated SUMO1 or SUMO2 modifications in all three cell systems . Six EBV proteins met this criteria and had moderate to high effects in at least one cell system ( marked in grey in Table 1 ) . Five of these increased SUMOylation while one ( BRLF1 or Rta ) decreased SUMOylation . Quantification of these effects is shown in Table 2 . Ability to modulate SUMOylation did not correspond to the expression levels of the viral proteins , as BMLF1 and BRLF1 were expressed at relatively low levels and some highly expressed proteins ( such as BXLF1 ) did not affect SUMOylation . In addition , increases in SUMOylation was not simply due to SUMOylation of the viral protein itself , as even when SUMOylation of the viral protein could be detected , it resulted in only a few discreet bands and not the high molecular weight smears seen in the SUMOylation screen ( see examples in S2 Fig ) . Upregulation of SUMOylation by BGLF2 , BMRF1 and SM and down regulation of SUMOylation by Rta was also confirmed by detecting endogenous SUMO levels in 293T cells ( Fig 2 ) . BRLF1 was the only EBV protein that we found to consistently decrease the level of SUMOylated proteins with a moderate to high effect . This decrease was seen for both SUMO1- and SUMO2-modified proteins . This effect was further verified by expressing BRLF1 in 293T cells and blotting for endogenous SUMO ( Fig 2 ) . The observed decrease in SUMO1 and SUMO2 modified proteins was not due to effects on SUMO or Ubc9 transcripts , as the level of SUMO1 , SUMO2 or Ubc9 mRNA was not significantly changed by Rta expression ( S3 Fig ) . Since the KSHV homologue of BRLF1 ( Rta ) was shown to be a STUbL , we asked whether BRLF1 had characteristics consistent with a STUbL . Since STUbLs target SUMOylated proteins for proteasomal degradation , we asked whether the loss of SUMOylated proteins induced by BRLF1 could be restored by blocking the proteasome with MG132 . As shown in Fig 3A , MG132 at least partly countered the loss of both SUMO1 and SUMO2-modified proteins that is induced by BRLF1 , mirroring the results with the HSV1 STUbL , ICP0 . Since BRLF1 has been reported to associate with the cellular STUbL , RNF4 [39] , we also asked whether the loss of SUMOylated proteins caused by BRLF1 required RNF4 . However , much like ICP0 , BRLF1 was found to induce the loss of both SUMO1 and SUMO2-modified proteins in CNE2Z cells ( relative to empty vector control ) regardless of whether or not RNF4 was silenced ( Fig 3B ) . Rta proteins from KSHV and γHV68 have ubiquitin ligase activity that can be observed by upregulation of global cellular ubiquitylation upon their overexpression [15 , 18 , 40] . To determine if this was also true for BRLF1 , we expressed it along with His6-myc-ubiquitin in both 293T and CNE2Z cells , with and without MG132 treatment , then recovered ubiquitylated proteins on metal chelating resin and detected them by Western blotting with anti-myc antibody ( Fig 3C ) . While KSHV Rta and HSV1 ICP0 induced ubiquitylation in both cell lines both with and ( to a lesser degree ) without MG132 , no induction of ubiquitylation was detected with BRLF1 . This suggests that , unlike Rta and ICP0 , BRLF1 induces the loss of SUMOylated proteins without endogenous ubiquitin ligase activity . We next asked whether BRLF1 could bind SUMO . To this end , FLAG-BRFL1 was purified from 293T cells on anti-FLAG resin under high salt conditions ( to limit protein interactions ) and extensively washed ( Fig 4A , left panel; BRLF1 ) . Negative control resin was similarly generated using lysates from 293T cells with empty FLAG vector ( Fig 4A , left panel; EV ) . GST-SUMO1 , GST-SUMO2 or GST alone were generated in Ecoli ( ( Fig 4A , left panel ) and equal amounts were applied to negative control or BRLF1-containing resin followed by extensive washing . Resin containing BRFL1 , but not empty vector , retained GST-SUMO1 and GST-SUMO2 but not GST alone ( Fig 4A , right panel ) , showing that BRLF1 can bind SUMO1 and SUMO2 . Since PML proteins are highly SUMO-modified , they are often a target of proteins that bind SUMO and disrupt/degrade SUMOylated proteins [15 , 41–43] . Interestingly our previous screen of over 200 EBV , HSV1 and CMV proteins for ability to induce loss of PML NBs in U2OS cells identified BRLF1 as one of the top hits [33] . We further verified this property by examining the effect of BRLF1 on PML nuclear bodies and proteins in CNE2Z nasopharyngeal carcinoma cells ( Fig 4B and 4C ) . BRLF1 consistently decreased the number and intensity of PML nuclear bodies and the level of PML proteins ( although not as dramatically as ICP0 which is known to have multiple mechansims of targeting PML [44] ) . The results are consistent with the ability of BRLF1 to induce the loss of SUMO-modified proteins . Two of the proteins that upregulated SUMOylation in our screens , BMLF1 and SM ( also called EB2 ) share common sequences , as BMLF1 is the C-terminal part of SM ( Table 1 , Fig 2 , Fig 5 ) . We focused our studies on SM since it is the functional protein . In assays in 293T cells detecting endogenous SUMO levels , SM expression was found to induce SUMO1 modifications without a noticeable effect on SUMO2 modifications ( Fig 2 ) . This was consistent with what we observed in the initial screen , where SUMO1 modifications were more obviously upregulated than SUMO2 modifications ( Table 1 ) . Interestingly , SM itself was efficiently modified by SUMO2 and less obviously by SUMO1 ( S1 Fig ) . SM is known to have several roles in lytic EBV infection , including EBV mRNA export , splicing activation and transcriptional activation [45–49] . SM is conserved in all herpesviruses and its homologues in HSV1 ( UL54 or ICP27 ) and CMV ( UL69 ) have similar roles as SM in lytic infection [50 , 51] . We reasoned that if SUMO induction by SM was important for viral infection , then this property would be conserved in the SM homologues in other herpesviruses . Therefore we compared the abilities of SM , UL54 and UL69 to induce SUMO1 and SUMO2 modifications in all three cell systems used for the initial screen ( Fig 5 ) . SM and UL69 were consistently found to upregulate SUMO1 modifications in all three cell lines , with little to no effect on SUMO2 modifications . Conversely , UL54 consistently increased SUMO2 modifications with little to no effect on SUMO1 modifications . Quantification from three independent experiments in each cell lines is shown in Table 2 . These effects were not due to induction of SUMO or Ubc9 transcripts as neither SUMO1 , SUMO2 nor Ubc9 mRNA levels were significantly affected by SM , UL54 or UL69 ( S3 Fig ) . Since HSV1 , CMV and EBV represent the three different subfamilies of herpesviruses ( alpha , beta and gamma respectively ) , the results indicate that global induction of SUMOylation is an activity that is conserved in the SM family of proteins , although different family members have different specificities for SUMO1 vs SUMO2 . Proteins that directly affect SUMOylation typically bind to SUMO and/or the Ubc9 E2 SUMO-conjugating enzyme . Therefore to determine whether SM , UL54 and UL69 are acting directly on cellular SUMOylation , we first examined their association with SUMO1 , SUMO2 and Ubc9 in human cells . To this end , HeLa cells containing integrated copies of His6-SUMO1 or His6-SUMO2 were transfected with FLAG-tagged SM , UL54 or UL69 followed by recovery of the His6-SUMOs on metal chelating resin under native conditions ( Fig 6A and 6B ) . All three viral proteins were pulled down by SUMO1 and SUMO2 to varying degrees indicating that they interact with SUMO . However , SM and UL69 were more efficiently recovered with SUMO1 than UL54 ( Fig 6A ) , and UL54 was more efficiently recovered with SUMO2 than SM and UL69 ( Fig 6B ) . This specificity reflects the ability of these proteins to induce SUMO1 vs SUMO2 modifications , suggesting that the degree of SUMO binding influences their SUMOylation induction specificity . To further investigate the SUMO specificities of the viral proteins and determine if SUMOs are bound directly , we expressed and partially purified FLAG-tagged SM , UL54 and UL69 from E . coli ( Fig 7A ) and used it in GST pull down assays with GST-tagged SUMO1 or SUMO2 ( also generated in E . coli ) . All three viral proteins were retained on glutathione resin by GST-SUMO1 and GST-SUMO2 but not by GST alone ( Fig 7B ) . However , once again the SUMO specificities varied among the viral proteins , with SM and UL69 being more efficiently bound by SUMO1 than SUMO2 , and UL54 being more efficiently bound by SUMO2 than SUMO1 . The results suggest that the ability to directly bind SUMO is an important factor in SUMOylation induction in cells . We also asked whether SM , UL54 and UL69 interact with the E2 enzyme for SUMOylation , Ubc9 . Immunoprecipitation of endogenous Ubc9 from 293T cells expressing SM , UL54 or UL69 showed that all three viral proteins were associated with Ubc9 ( Fig 6C ) . We examined whether these associations were direct by using the E . coli purified viral proteins in GST pull down assays with GST-Ubc9 ( Fig 7B ) . SM , UL54 and UL69 were all retained on glutathione resin by GST-Ubc9 but not by GST alone , showing that each can directly bind Ubc9 . The properties of SM , UL54 and UL69 in inducing global SUMOylation , binding directly to SUMO and binding directly to Ubc9 are consistent with SUMO E3 ligases . Therefore we investigated whether these viral proteins have SUMO E3 ligase activity in a purified in vitro system with the SUMO E1 ( SAE ) and E2 ( Ubc9 ) enzymes , SUMO1 or SUMO2 and full length monomeric p53 as a substrate [52] . p53 was used as it is a well characterized substrate for SUMOylation that has previously been shown to require E3 ligases in vitro [10 , 11 , 13 , 53] . All of the proteins in this assay ( SAE , Ubc9 , SUMO1 , SUMO2 , p53 and the viral proteins ) were generated in E . coli to ensure that no contaminating SUMO E3 ligases are present . The addition of increasing amounts of purified SM ( Fig 8A ) , UL54 ( Fig 8B ) or UL69 ( Fig 8C ) resulted in titratable induction of a shifted form of p53 consistent with mono-SUMOylated p53 ( top panels ) . This shift was seen with both SUMO1 ( left panels ) and SUMO2 ( right panels ) . Western blots for SUMO1 or SUMO2 confirmed that the shifted band was SUMO-modified p53 ( middle panels ) . This band was not seen in assays using the highest amounts of the viral proteins in the absence of SAE , confirming that this SUMOylation was dependent on the E1 SUMO conjugating enzyme . Together the results indicate that SM , UL54 and UL69 can act as SUMO E3 ligases . Since we have shown that SM , UL54 and UL69 can all catalyze p53 SUMOylation in vitro , we investigated whether they also induced p53 SUMOylation in cells . To this end , 293T cells were co-transfected with constructs expression His6-SUMO1 or His6-SUMO2 and FLAG-tagged SM , UL54 or UL69 . His-tagged proteins were then recovered on metal chelating resin under denaturing conditions and analysed by Western blotting using anti-p53 antibody . All three viral proteins were found to induce mono-SUMOylation of endogenous p53 by SUMO1 , to a similar degree as the LMP1 positive control ( Fig 9A ) . UL54 induced less SUMO1-modification of p53 than SM and UL69 , consistent with the trend we observed for global SUMOylation . In SUMO2 experiments ( Fig 9B ) , while the mono-SUMO2 p53 band ( ~70 KDa ) was not obviously affected by SM , UL54 or UL69 as compared to the empty vector control , higher molecular weight SUMO2 products were evident with UL54 ( and the LMP1 positive control ) . Together the results suggest that SM , UL54 and UL69 can all affect p53 SUMOylation , but with different SUMO1 vs SUMO2 specificities that reflect their effects on global SUMOylation and SUMO binding . The fact that several EBV proteins can affect SUMOylation suggests a complicated interplay with SUMO pathways during EBV infection which would be best understood after identifying the specific targets of each of the EBV proteins . However , to begin to explore the effect of SM on SUMOylation in the context of EBV infection , we examined the cellular SUMO1 profile in EBV-positive gastric carcinoma cells ( AGS-EBV ) after lytic reactivation with and without SM depletion with SM-targeted siRNA . Quantification of SM transcripts showed that this gene was expressed by 16 hours post-reactivation and continued to increase in levels at 24 and 48 hour time points ( Fig 10A ) . Pre-treatment with SM siRNA significantly decreased SM levels , particularly at 24 and 48 hour time points ( Fig 10A ) . At the same time points , cell lysates were collected and analysed for SUMO1 profiles by Western blotting ( Fig 10B ) . Two SUMO1-containing bands migrating at ~62 and 68 Kda were observed during the early stages of lytic infection but not at 48 hrs post induction ( at which time greatly increased SUMO1 profiles suggest a stress response ) . These bands were reduced with SM depletion , most obviously at 24 hours post-induction when SM is accumulating in the cells . Note that these bands are too small to be any SUMO1-modified version of SM , which would migrate above the 72 Kda marker ( although S2 Fig suggests there would be little SUMO1-modified SM ) . The results support the ability of SM to affect SUMOylation of some proteins in the context of EBV infection .
EBV manipulates many cellular pathways in order to promote infection and avoid host immune responses , and under some circumstances such alterations can lead to oncogenesis . Here we show that some EBV proteins can dramatically affect the SUMOylation of host proteins , providing an additional mechanism by which EBV can manipulate cells . We identified one EBV protein ( BRLF1/Rta ) that decreases the level of SUMOylated proteins and four distinct proteins that increase SUMOylated proteins . One of the latter proteins ( SM ) has characteristics of a SUMO E3 ligase , and this activity is conserved in SM homologues in HSV1 ( UL54 ) and CMV ( UL69 ) suggesting the importance of this activity for herpesvirus infections . Our screen using overexpressed EBV proteins was designed so that viral proteins with the ability to affect SUMO pathways would manifest as a global SUMOylation phenotype , even though in the context of infection they may only affect a subset of SUMOylated proteins . Herpesviruses in lytic infection modulate a wide variety of cellular processes that involve SUMOylation , the best studied being the disruption of PML NBs by viral SUMO-targeted ubiquitin ligases that degrade PML proteins or SUMO-interacting proteins that disrupt PML protein interactions [15 , 19 , 54–57] . Indeed , the SUMO downregulator that we identified here ( BRLF1 ) was also shown to disrupt PML NBs . Additional SUMO-regulated processes controlled by herpesviruses include interference with cell cycle progression , resulting in G1/S arrest , and inhibition of DNA damage responses [58–64] . Our previous screens have identified BGLF2 and BMRF1 as contributing to G1/S arrest [28] and BMRF1 as an inhibitor of DNA damage response [65] . We do not yet know the mechanism of these effects , but the fact that both proteins were found to upregulate SUMOylation suggests that they may be affecting cell cycle progression and the DDR by increasing SUMOylation of some cellular proteins . The only EBV protein that we found to globally decrease SUMOylation was BRLF1 ( also called Rta ) ; an immediate early protein conserved in γ-herpesviruses and known to function as a transcriptional activator . Interestingly , the BRLF1 homologues ( Rta ) in KSHV and murine γHV68 were both shown to have ubiquitin ligase activity [18 , 40] and KSHV Rta was later shown to be a SUMO-targeted ubiquitin ligase ( STUbL ) [15] . Like these Rta proteins , the protein loss we observed with BRLF1 was proteasomal dependent , suggesting that induction of protein degradation is a conserved function of the Rta proteins . While the ability of γHV68 Rta to bind SUMO has not been reported , it is known to degrade the RelA subunit of NF-κB which is highly SUMO-modified by PIAS3 , raising the possibility that SUMOylation of RelA is part of the targeting mechanism by Rta [18 , 66] . KSHV Rta was shown to contain SUMO interacting motifs that bind SUMO and to induce degradation of the highly SUMOylated PML proteins [15] . In addition , the STUbL activity of KSHV Rta was shown to be required for transcriptional activation , suggesting that removal of suppressive SUMOylated proteins promotes transcription [15] . Similarly , we found that EBV Rta binds SUMO and induces loss of PML NBs and proteins ( [33] and Fig 4 ) . In addition , EBV Rta is SUMO modified and this modification has been found to increase its transactivation activity [33 , 67 , 68] . However , despite the many similarities between KSHV and EBV Rta , we could not find evidence of ubiquitin ligase activity in EBV Rta , even though this activity was readily apparent in the same assay with KSHV Rta ( Fig 3C ) . This is likely due to the low sequence conservation between these proteins , including the lack of catalytic cysteines which are conserved in KSHV and γHV68 Rta [18] . Interestingly , a KSHV Rta mutant lacking ubiquitin ligase activity still induces the degradation of PML proteins suggesting that it has two mechanisms of degrading SUMOylated proteins [15] . This second ( unknown ) mechanism of degradation of SUMOylated proteins may be conserved in EBV Rta . Our screen identified four distinct EBV proteins that globally upregulated SUMOylation . ( SM , BGLF2 , BMRF1 , BVRF2 ) . None of these proteins have been previously shown to impact SUMO pathways , although both BMRF1 and BVRF2 were shown to be SUMO-modified [32 , 69] . Since global SUMOylation patterns can be caused by stress responses , it is possible that some of these proteins are eliciting a stress response [70] . Little is known about BVRF2 other than its conserved role as a scaffold protease [71] . BGLF2 is a viral tegument protein that has been found to induce p21 and interfere with G1/S cell cycle progression [28] and also to induce BZLF1 expression and the AP-1 signalling pathway through p38 and c-Jun N-terminal kinases activation [72 , 73] . BMRF1 belongs to the family of DNA polymerase processivity factors that are conserved in all herpesviruses [74] . However , BMRF1 also has additional roles in transcriptional activation [75–77] , cell cycle progression [28] and in inhibiting the DNA damage response to double stranded DNA breaks [34] . It will be interesting to determine how the multiple roles of BGLF2 and BMRF1 relate to their ability to upregulate SUMOylation . One of the EBV proteins identified in our screens that globally upregulated SUMOylation was the SM ( or EB2 ) protein that is conserved in other herpesviruses . SM and its homologues in HSV1 ( UL54 or ICP27 ) and CMV ( UL69 ) have conserved functions in RNA binding , splicing modulation and the export and translation of viral mRNA [45 , 46 , 48–51 , 78] . In addition , several studies have identified roles for HSV1 UL54 in cell signalling and apoptosis [51 , 79–81] . Our findings that SM , UL54 and UL69 induce SUMOylation suggests that the cellular effects of these proteins may be more extensive than is currently known . In addition , SUMOylation might play roles in the known functions of these proteins . While the role of SUMOylation in RNA splicing and mRNA transport has not been extensively studied , heterogeneous nuclear RNA binding proteins ( hnRNPs ) , which have multiple roles in RNA splicing stabilization and export , are highly SUMOylated by both SUMO1 and SUMO2/3 [82–84] . In addition , the RanBP2/Nup358 component of the nuclear pore complex , which plays roles in mRNA export and translation , is a SUMO E3 ligase [36 , 85 , 86] . Our data suggest that SM , UL54 and UL69 are SUMO E3 ligases . These proteins not only upregulated SUMOylation in multiple cell systems , but bound directly to SUMO and Ubc9 , as expected of SUMO E3 ligases . In addition , in a purified system using E . coli generated proteins , SM , UL54 and UL69 all induced SUMOylation of p53 in conjunction with Ubc9 and SAE . In cells , the three viral proteins varied in their abilities to induce SUMO1 vs SUMO2 modifications , with SM and UL69 preferentially inducing SUMO1 modifications and UL54 preferentially inducing SUMO2 modifications . This preference was also reflected in SUMO binding assays , where both immunoprecipitations from cells and in vitro SUMO binding assays showed that SM and UL69 preferentially bound SUMO1 , while UL54 preferentially bound SUMO2 . Preferences for SUMO1 or SUMO2/3 have been previously reported for some other SUMO binding proteins . For example , the KSHV K-bZIP and cellular ZNF451 SUMO E3 ligases have been shown to have specificity for SUMO2/3 over SUMO1 , due to the presence of SIMs that preferentially bind SUMO2/3 over SUMO1 [13 , 87 , 88] . In contrast ORF61 of Varicella-Zoster virus ( orthologue of the HSV1 ICP0 STUbL ) preferentially binds SUMO1 [21] . SIMs can vary considerably in sequence , making them difficult to accurately predict by sequence analysis [89] , and hence the SIMS in SM , UL54 and UL69 have yet to be determined . SUMO E3 ligases are also difficult to predict since their sequence and structures vary considerably . The PIAS family of SUMO E3 ligases contain modified RING domains similar to RING-type ubiquitin ligases , however other SUMO E3 ligases , such as RANBP2 and K-bZIP , lack conserved catalytic domains [89 , 90] . Zinc fingers have been found to be an important component of some E3 SUMO ligases , for example RANBP2 contains eight tandem zinc fingers and the ZNF451 family of SUMO E3 ligases contain 12 C2H2 zinc-finger domains [88 , 91] . Interestingly , the structure of the C-terminal domain of ICP27 ( UL54 ) revealed a novel CHCC-type zinc finger in which the zinc coordinating residues are conserved in UL54 homologues in other herpesviruses , including UL69 and SM [92 , 93] . The conservation of this zinc-binding domain , and the presence of similar zinc binding domains in cellular SUMO E3 ligases , suggests that it might be an important component of the SUMO E3 ligase activity of SM , UL54 and UL69 . We have shown that p53 is one cellular protein that can be SUMOylated by SM , UL54 and UL69 . p53 is known to be modified by either SUMO1 or SUMO2/3 at K386 [94] . The consequences of these modifications are a matter of debate , as there are reports that these modifications increase transcriptional activation by p53 and other reports that they decrease p53 activity [13 , 94–97] . We showed that SM , UL54 and UL69 are all capable of inducing SUMO1 and SUMO2 modifications of p53 under in vitro conditions . In cells we also observed increased SUMOylation of p53 by these viral proteins , although SM and UL69 preferentially increased SUMO1 modification while UL54 increased SUMO2 modifications . This pattern parallels the global SUMOylation effects of these viral proteins . Whether or not p53 is an intended SUMOylation target of these viral proteins or whether other cellular proteins are preferential targets remains to be determined . Although SM , UL54 and UL69 globally upregulate SUMOylation under our assays conditions , we imagine that these proteins have preferred substrates during infection . Identifying these substrates will provide important insights into the functions and mechanisms of action of these proteins . Our results build on the growing body of literature that manipulation of host SUMO pathways is important for lytic infection by herpesviruses . KSHV is known to express both a STUbL ( Rta ) and a SUMO2/3-specific SUMO E3 ligase ( K-bZIP ) in lytic infection , and SUMO2/3 modifications in general have been shown to suppress KSHV reactivation and expression of KSHV lytic genes [13 , 15 , 98 , 99] . HSV1 is also known to encode a well-studied STUbL ( ICP0 ) [16] . Our study provides the first identification of SUMO E3 ligases in HSV1 , CMV and EBV , and also identifies EBV BRLF1 as a negative regulator of SUMOylation . To our knowledge , these are the first reports of EBV lytic proteins that globally affect SUMOylation . Previous studies on EBV and SUMOylation pathways have shown that LMP1 promotes latency by upregulating SUMOylation in EBV latent infection [30 , 31] , and that one of the EBV miRNAs expressed in lytic infection can promote SUMOylation by downregulating RNF4 [32] . In addition , several EBV proteins have been found to be SUMO modified , suggesting that their activities can be regulated by SUMOylation [32 , 33 , 39 , 69] . It will be interesting to determine how the interplay between viral proteins that increase and decrease SUMOylation contribute to herpesvirus infections and why UL54 preferentially induces SUMO2 modifications while its homologues in EBV and CMV induce SUMO1 modifications .
The EBV expression library was generated in pMZS3F for expression in mammalian cells , as previously described [28] , such that proteins are expressed fused to a C-terminal calmodulin binding peptide and triple FLAG epitope . EBV SM in pcDNA3 ( a generous gift from Sankar Swaminathan ) and EBV BRLF1 in pMZS3F were subcloned between the Sal I and Xba I sites in pCMV3FC , and HSV-1 UL54/ICP27 and CMV UL69 in pMZS3F were subcloned between the Xho I and Xba I sites in pCMV3FC , to generate proteins with C-terminal triple FLAG tags . Plasmids encoding EBV HA-tagged LMP1 ( a gift from Nancy Raab-Traub ) [100] , HSV-1 ICP0 ( pCI-110; a gift from Roger Everett ) [101] and human His6-SUMO1 and His6-SUMO2 ( in pcDNA3; gifts from Ronald T . Hay ) [38 , 102] for expression in mammalian cells have been previously described . For expression in E . coli , SM , UL54/ICP27 and UL69 coding sequences were excised from pCMV3FC using the restriction enzymes indicated above and were ligated into the corresponding sites of a modified pET15b with a multicloning site . To generate the modified pET15b , the Xba I site was mutated in pET15b then oligos ( 5'- CCA TGG GCA GCA GCC ATC ATC ATC ATC ATC ACA GCA GCG GCC TCG AGG CTA GCG TCG ACG GTA CCT CTA GAG ACG TAG CGG CCG CGG CGG ATC C -3' ) containing a multicloning sequence ( Xho I , Nhe I , Sal I , KpnI , Xba I , Not I ) were inserted between the Nco I and Bam HI sites of pET15b downstream of the hexahistidine tag . Oligos encoding a triple FLAG tag were then inserted between the Not I and Bam HI sites so that recombinant proteins contain N-terminal His6 and C-terminal 3FLAG tags . pGEX2T-SUMO1 , pGEX2T-SUMO2 and pGEX2T-Ubc9 were kindly provided by Ronald T . Hay [103] . pcDNA4/TO plasmid expressing KSHV Rta ( ORF50 ) with C-terminal Strep tag was a gift from Britt Glausinger and is described in Davis et al 2015[104] . The plasmid expressing His-Myc-ubiquitin ( pHis-Myc-Ub ) was a gift from Filippo Giancotti [105] HeLa cells ( cervical carcinoma ) containing integrated His6-SUMO1 or His6-SUMO2 ( kindly provided by Ronald T . Hay; [106] ) and 293T cells ( embryonic kidney; ATCC ) were cultured in DMEM with 10% FBS . CNE2Z cells ( EBV-negative nasopharyngeal carcinoma [107]; a gift from Fei-Fei Liu ) were cultured in α-MEM with 10% FBS . AGS-EBVcells ( EBV- positive gastric carcinoma [108]; a gift from Lindsay Hutt-Fletcher ) were grown in RPMI with 10% FBS . All cells were plated 24 hrs before transfection and transfected at a confluency of 70–80% using PolyJet ( FroggaBio ) or Lipofectamine 2000 ( Life Technologies ) or linear polyethylenimine ( PEI; Polyscience Inc . catalogue number 23966 ) , as suggested by the manufacturer . Protein fractions were separated by SDS-PAGE ( 8% to 15% depending on the experiment ) and transferred onto nitrocellulose . Membranes were blocked with 1% BSA in PBS-Tween 0 . 1% ( PBS-T ) for 1 h , followed by incubation with primary antibodies in blocking buffer overnight at 4°C . Membranes were washed three times with PBS-T for 10 min and then incubated with secondary antibodies conjugated to horseradish peroxidase in blocking buffer for 1 h . Membranes were washed four times with PBS-T for 15 min , and signals were detected by enhanced chemiluminescence ( Santa Cruz sc-2048 or Biorad Clarity 1705061 ) . Antibodies to SUMO1 ( FL-101 , 1:1000 dilution ) , SUMO2 ( FL-103 , 1:2000 dilution ) , Ubc9 ( H-81 , 1:500 dilution ) , His ( H-15 , 1:500 dilution ) , p53 ( DO-1 , 1:5000 dilution ) , myc ( sc-40; 1:2000 dilution ) and actin ( C-11 , 1:2000 dilution ) were from Santa Cruz . Antibodies to PML ( A301-167A , 1:2000 dilution ) , GST ( A190-122A; 1:10 , 000 ) and FLAG ( S190-102 , 1:10000 dilution ) were from Bethyl . Anti-FLAG ( F1804 , 1:10000 dilution ) was from SIGMA and anti-ICP0 ( H1A027 , 1:5000 dilution ) was from Virusys . RNF4 antibody ( a kind gift from Ronald T . Hay; [109] ) and was used at 1:5000 dilution . 293T and CNE2Z cells in 6 well dishes were co-transfected with plasmids ( 0 . 5 μg each ) expressing His6-SUMO1 or His6-SUMO2 and FLAG-tagged viral proteins ( in pMZS3F or pCMV3FC ) or LMP1 ( positive control for SUMO upregulation ) or ICP0 ( positive control for SUMO downregulation ) or empty plasmid negative control . HeLa cells containing integrated His6-SUMO1 or His6-SUMO2 were similarly transfected with 0 . 5 μg of plasmids expressing FLAG-tagged viral proteins , ICP0 or LMP1 or empty plasmid control . His6-tagged SUMO conjugates were purified under denaturing conditions essentially as previously described [106] . Briefly , approximately 2 x 106 cells were harvested 36 hrs post transfection . 10% of the cells were lysed in 2X SDS loading buffer ( 60 mM Tris . HCl pH 6 . 8 , 1% SDS , 100 mM DTT , 5% glycerol ) to provide the input sample . 90% of the cells were resuspended in 0 . 5 ml lysis buffer G ( 6 M guanidine hydrochloride , 10 mM Tris , 100 mM sodium phosphate , pH 8 . 0 ) and incubated on ice 20 min . Lysates were passed through a 30G needle five times . Then , lysates were added 50 μl of TALON Metal Affinity Resin ( Clontech ) previously equilibrated with lysis buffer , and incubated 3 hrs at room temperature with end-over-end rotation . The resin was washed four times with 1 ml of wash buffer U ( 8 M urea , 10 mM Tris , 100 mM sodium phosphate , pH 8 . 0 ) and proteins were eluted in 2X SDS loading buffer . Inputs and purified fractions were analyzed by Western blotting with SUMO1 , SUMO2 , FLAG and actin antibodies . For SUMOylation experiments involving RNF4 silencing , CNE2Z were plated in a 6-well dish at 10% confluency and transfected with either 40 pmoles of Stealth siRNF4 ( Invitrogen ) or Qiagen Allstars control siRNA with Lipofectamine 2000 ( ThermoFisher Scientific ) according to the manufacturer’s instructions . The transfection was repeated after 24 hours . 9 hours later the cells were transfected ( using PEI ) with1 μg of pCMV3FC , pCI-110 ( expressing ICP0 ) or pCMV3FC-BRLF1 and 1 μg of plasmids expressing His6-SUMO1 or His6-SUMO2 . Cells were harvested 39 hours later and lysed in 150 μl of 8 M urea buffer ( 8 M urea , 20 mM Tris pH 8 , 100 mM NaCl , protease inhibitor cocktail ( P8849; Sigma ) ) . Lysates were sonicated and clarified by centrifugation . 360 μg of clarified lysate was incubated with 50 μl equilibrated TALON resin for 2 hours at RT with mixing , followed by washing and elution as above . Elutions were analysed by 8% SDS-PAGE and Western blotting for SUMO1 or SUMO2 . 30 μg of each clarified lysate was analysed by 12% SDS-PAGE and Western blotting for RNF4 , ICP0 , FLAG ( BRLF1 ) and actin ( Thermofisher ) . 293T cells in 6 well dishes were transfected with 2 μg plasmid expressing ICP0 or FLAG-tagged BRLF1 using PEI according to the manufacturer’s protocol . For one set of samples , MG132 ( Sigma ) was added to 10 μM 24 hours post-transfection and 10 hours prior to harvesting . All samples were harvested 34 hours post-transfection and lysed in 9 M urea , 10 mM Tris pH 6 . 8 with sonication . 40 μg of clarified lysates were analysed on 8% SDS-PAGs followed by Western blotting with antibodies against SUMO1 , SUMO2 , FLAG , ICP0 and actin . CNE2Z or 293T in 6 well dishes were transfected with PEI with 1 μg pCMV-3FC , pCI-110 ( expressing ICP0 ) , pcDNA4/TO expressing Strep-tagged KSHV Rta or pCMV3FC-BRLF1 and 1 μg of plasmid expressing His-Myc-ubiquitin [105] . 24 hours later , MG132 ( Cell Signalling , 2194S ) was added to 10 μM for 10 hours . Cells were harvested and lysed in 150 μl ( CNE2Z ) or 300 μl ( 293T ) 8M urea buffer ( 8M urea , 20 mM Tris pH 8 , 100 mM NaCl , protease inhibitor cocktail ( Sigma P8849 ) ) . Lysates were sonicated and centrifuged to clarify . 50 μg ( CNE2Z ) or 160 μg ( 293T ) of lysate was added to 50 μl TALON resin ( prewashed with 8M urea buffer ) , followed by incubation , washing and elution as described above for SUMOylation screen . 30 μl of elutions and 30 μg of clarified lysates were analysed by Western blotting using anti-myc antibody . Input lysate were also probed with antibodies against FLAG ( BRLF1 ) , ICP0 and Strep ( KSHV Rta ) . Immunofluorescence microscopy analysis was performed as described previously [45] . Briefly , CNE2Z cells on cover slips were transfected with plasmids expressing ICP0 ( pCI-110 ) or FLAG-tagged BRLF1 ( in pCMV3FC ) , fixed 24 hrs post transfection and stained with antibodies against PML and ICP0 or FLAG . PML NBs were counted in 50 cells for each sample in two independent experiments . For PML Western blots , cells were transfected as described above and , 48 hrs post transfection , cells were lysed in 9M urea buffer ( 9M urea , 10 mM Tris pH 6 . 8 ) . 40 μg of clarified lysates were loaded onto 10% SDS-PAGE , transferred onto nitrocellulose and analyzed by Western blotting with PML , ICP0 and FLAG antibodies . 293T cells in 12 well plates were transfected with 0 . 5 μg plasmid expressing LMP1 , ICP0 or the indicated FLAG-tagged EBV protein . 36 hrs later , 1 x 106 293T cells were lysed in passive lysis buffer ( Promega E194A ) . 50 μg of clarified lysates were loaded onto 8% SDS-PAGE , transferred to nitrocellulose and analyzed by Western blotting with SUMO1 , SUMO2 , FLAG and actin antibodies . To evaluate the interaction of SM , UL54 , and UL69 with SUMO proteins , HeLa cells containing integrated His6-SUMO1 or His6-SUMO2 in 6 cm dishes were transfected with 2 . 5 μg of pCMV3FC plasmids expressing FLAG-tagged SM , UL54 or UL69 or with pCMV3FC alone . 36 hrs later , cells were harvested and lysed in RIPA buffer without EDTA [50 mM Tris pH 8 . 0 , 200 mM NaCl , 1 . 0% ( v/v ) NP-40 , 0 . 5% ( w/v ) sodium deoxycholate , protease inhibitor cocktail SIGMA P8340] . Lysates ( 1 mg ) were subjected to His-pull down using 100 μl of Ni-NTA agarose ( Qiagen ) for 2 hrs at 4°C with mixing , and beads were then washed four times with 1 ml of RIPA buffer . Proteins were eluted in 2X SDS loading buffer and analyzed by Western blotting with antibodies against FLAG and His . To evaluate the interaction of SM , UL54 , and UL69 with Ubc9 , 293T cells were transfected as described above , and 36 hrs post-transfection cells were lysed in RIPA buffer . 2 mg of lysate was incubated overnight at 4°C with 25 μg of agarose-conjugated Ubc9 antibody ( C-12; Santa Cruz ) , with mixing , then beads were washed four times with 1 ml of RIPA buffer . Proteins were eluted in 2X SDS loading buffer and analyzed by Western blotting with FLAG and Ubc9 antibodies . Recombinant His6-SM-3FLAG , His6-UL54-3FLAG and His6-UL69-3FLAG proteins were generated in E . coli for in vitro assays . BL21-pLysS E . coli containing pET15b-SM-3FLAG , pET15b-UL54-3FLAG or pET15b-UL69-3FLAG were grown in Luria broth ( LB ) to OD580 0 . 6 then protein expression was induced with 1 mM IPTG overnight at 18°C . Bacteria from 1 L of culture were resuspended in 20 ml of binding buffer ( 50 mM sodium phosphate , 300 mM NaCl , 5 mM imidazole , protease inhibitor cocktail , pH 7 . 8 ) and lysed with 3 rounds of sonication for 20 sec each . Lysates were clarified by centrifugation at 10000 x g for 20 min at 4°C and then incubated with 400 μl of Ni-NTA agarose ( Qiagen ) for 1 h at 4°C with mixing . Agarose was washed 4 times with 2 ml washing buffer ( 50 mM sodium phosphate , 300 mM NaCl , 20 mM imidazole , protease inhibitor cocktail , pH7 . 6 ) , then transferred into a gravity-flow column and washed once more with 2 ml washing buffer . Proteins were eluted with 1 ml of elution buffer ( 50 mM sodium phosphate , 300 mM NaCl , 200 mM imidazole , protease inhibitor cocktail , pH7 . 4 ) and collected in 200 μl fractions . Hexahistidine-tagged full length p53 with L344P point mutation ( rendering it monomeric ) was expressed and purified from E . coli as described previously [52] . GST , GST-SUMO1 , GST-SUMO2 and GST-Ubc9 were generated by standard methods . Briefly , DH5α E . coli containing pGEX-2T expression plasmids were grown in LB to OD580 0 . 6 , then protein expression was induced with 1 mM IPTG for 2 hrs at 37°C . Bacteria from 1 L of culture were resuspended in 20 ml of PBS supplemented with 1% Triton X-100 and 1 mM PMSF , and lysed with 3 rounds of sonication for 20 sec each . Lysates were clarified by centrifugation at 10000 x g for 20 min at 4°C and then incubated with 500 μl of Glutathione Sepharose 4B ( GE life sciences ) for 2 hrs at 4°C with mixing . Resin was washed 4 times with 2 ml PBS-TritonX100-PMSF , then resuspended in 0 . 5 ml PBS with 1 mM PMSF and stored at -80°C . GST-pull down assays were performed to evaluate the interaction of purified , recombinant SM , UL54 and UL69 with GST-tagged SUMO1 , SUMO2 and Ubc9 . To this end , GST-tagged proteins ( or GST alone ) bound to glutathione resin ( described above ) were blocked with 2% BSA in PBS for 2 hrs at 4°C then levels of the GST proteins were evaluated by SDS-PAGE and Coomassie blue staining . Equal amounts of GST-SUMO1 , GST-SUMO2 , GST-Ubc9 or GST control bound to resin ( corresponding to ~2 μg of full length protein ) were combined with equal amounts ( ~1 μg as estimated from Coomassie stained SDS-PAGs ) of full-length recombinant His6-SM-3FLAG , His6-UL54-3FLAG or His6-UL69-3FLAG ( partially purified from E . coli as described above ) in 150 μl binding buffer ( 20 mM Tris pH 8 . 0 , 200 mM NaCl , 0 . 2 mM EDTA , 10% glycerol , 0 . 1% Triton X-100 , protease inhibitor cocktail ) for 3 hrs at 4°C with mixing . The resin was then washed with 1 ml of binding buffer four times and proteins eluted in 2X SDS loading buffer . 10% of the elution was loaded onto 15% SDS-PAGE and subjected to Coomassie staining . 90% of the elution was analyzed by western blotting with anti-FLAG antibody . 10 cm dishes of 293T were transfected with 8 μg each pCMV3FC or pCMV3FC-BRLF1 using 24 μl of PEI according to the manufacturer’s protocol . Cells were harvested 48 hours later and lysed in 50 mM Tris pH 8 . 0 , 1 M NaCl , 0 . 1% sodium deoxycholate , 0 . 5% NP40 , 2 mM EDTA , protease inhibitor cocktail ( P8340; Sigma ) . Cells were lysed by sonication and lysates clarified by centrifugation . Protein concentrations were adjusted to 6 . 5 mg/ml and 600 μl of lysate was incubated with 5 μl equilibrated anti-FLAG M2 resin ( Sigma A2220 ) for 2 hours . The resin was washed twice with 1 ml lysis buffer , twice with 1 ml BC100 ( 20mM Tris pH7 . 9 , 100mM NaCl , 10% glycerol , 0 . 2mM EDTA , 0 . 2% TritonX100 , protease inhibitor cocktail ( Sigma P8849 ) ) then blocked by 2 hr incubation in 1 ml BC100 containing 2% BSA . Bacterial lysates containing equivalent amounts of GST , GST-SUMO1 or GST-SUMO2 ( generated as above except using BC100 lysis buffer ) diluted to 100 μl in BC100/2%BSA were then added to the FLAG resin . The resin was washed five times with 1 ml BC100 containing 1% TritonX-100 and eluted in 2X SDS loading buffer . 20% of the elutions and 20% of the equivalent GST inputs were run anaylsed by SDS-PAGE and Coomassie staining to compare levels of input proteins . The remaining elutions ( and 1% of the GST inputs ) were analysed by Western blotting using anti-GST antibody and goat-anti rabbit secondary antibody ( SAB3700878; 1:5000 ) . Assays for in vitro SUMOylation of p53 were performed using the SUMOylation Assay Kit from Abcam ( ab139470 ) , as suggested by the manufacturer . 10 μl reactions included 50 ng of recombinant p53 as substrate ( purified from E . coli as indicated above ) and varying amounts ( ~100 , 200 and 400 ng ) of recombinant His6-SM-3FLAG , His6-UL54-3FLAG or His6-UL69-3FLAG ( partially purified from E . coli as described above ) . Reactions with no viral proteins or with viral proteins but no SAE were included as negative controls . Reactions were incubated 1 . 5 hrs at 37°C and stopped by the addition of 2X SDS loading buffer . The presence of viral proteins and the SUMOylation of p53 were evaluated by Western blotting with FLAG , p53 , SUMO1 and SUMO2 antibodies . 293T cells were co-transfected with plasmids expressing His6-SUMO1 or His6-SUMO2 and pCMV3FC expressing FLAG-tagged SM , UL54 or UL69 or empty pCMV3FC or positive control LMP1 and ICP0 plasmids as describe above for the SUMO screen . 36 hrs post transfection , 10% of the cells were lysed in 2X SDS loading buffer ( input fraction ) and 90% of the cells were lysed in lysis buffer G followed by purification of His-tagged SUMO conjugates on metal chelating resin as described above . Inputs and purified fractions were analyzed by western blotting with p53 and FLAG antibodies . AGS-EBV cells in 6-well dishes were transfected with 100 pmoles of SM-specific siRNA ( 5’-GCUGCACCGAUGAAAGUUATT-3’ ) or AllStars negative-control siRNA ( Qiagen ) using 2 μl of Lipofectamine 2000 ( Thermo Fisher Scientific ) . Two additional rounds of silencing were performed after 24 and 48 hours . Twenty-four hours later , cells were treated with 3 mM sodium butyrate ( NaB ) and 20 ng/ml 12-O-tetradecanoylphorbol-13-acetate ( TPA ) to induce the lytic cycle and harvested at 0 , 8 , 16 , 24 and 48 hours post- treatment for reverse-transcriptase quantitative PCR ( qRT-PCR ) of SM transcripts and Western blot analysis . For qRT-PCR , total RNA was extracted using TRIzol ( Invitrogen ) according to the manufacturer’s instructions . One microgram of total RNA was treated with 0 . 5 units of DNase I ( New England BioLabs ) for 15 min and reverse transcribed in a 20 μl reaction mixture using the SuperScript IV reverse transcriptase ( Invitrogen ) with random hexamer primers according to the manufacturer’s instructions . qRT- PCR was performed with 1 μl of 1:10 dilution of the cDNA using Luna Universal qPCR Master Mix ( New England BioLabs ) with a total reaction volume of 10 μl in a Bio-Rad CFX384 Real-Time System ( Bio-Rad ) . Primers used were: SM forward 5’-CCTGCTTCCTTCCTAACACG-3’ , SM reverse 5’-CGTGCCAGGGTTGTAATTCT-3’ , β-actin forward 5′-GGACTTCGAGCAAGAGATGG-3′ and β-actin reverse 5′- AGCACTGTGTTGGCGTACAG-3′ . The relative mRNA expression level was derived from 2-ΔΔCT by use of the comparative threshold cycle ( CT ) method . The amount of mRNA in each sample was normalized to the amount of actin mRNA . For Western blot analyses , Cells were lysed in 9 M urea-10 mM Tris ( pH 6 . 8 ) followed by sonication and clarification by centrifugation . Twenty micrograms of clarified lysates were analysed by 8% SDS-PAGE and Western blotting using antibodies against SUMO1 ( rabbit , 1:1000 dilution , sc-9060 , Santa Cruz ) or β-actin ( mouse , 1:10 , 000 dilution , sc-47778 , Santa Cruz ) and secondary antibodies goat-anti-rabbit ( 1:5000 dilution , SAB3700878-1 , Sigma ) or goat anti-mouse ( 1:5000 dilution , sc-2005 , Santa Cruz ) . | The functions of many cellular proteins important for anti-viral responses and oncogenesis are controlled by modifications by small ubiquitin-like modifiers ( SUMOs ) . Here we present the first screen of Epstein-Barr virus ( EBV ) proteins for those that can globally alter SUMO modifications of cellular proteins . We identify four distinct EBV proteins that increase SUMO modifications and one that decreases them . One of the SUMO upregulating proteins ( SM ) is conserved in other herpesviruses and we show that this activity is conserved in homologues from herpes simplex virus 1 ( HSV1 ) and cytomegalovirus ( CMV ) . We also show that these three homologues have SUMO E3 ligase activity in in vitro assays and that they bind SUMO and Ubc9 , consistent with the expectation of SUMO E3 ligases . The results provide new insights into the functions and mechanisms of action of this family of herpesvirus proteins . Our study identifies the first SUMO E3 ligases for EBV , HSV1 and CMV and provides a new mechanism by which EBV can manipulate cellular processes , through global effects on cellular SUMOylation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"293t",
"cells",
"enzymes",
"pathogens",
"biological",
"cultures",
"enzymology",
"dna-binding",
"proteins",
"microbiology",
"viral",
"structure",
"plasmid",
"construction",
"viruses",
"sumoylation",
"dna",
"viruses",
"dna",
"construction",
"molecular",
"biology",
"techniques",
"ligases",
"herpesviruses",
"research",
"and",
"analysis",
"methods",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"kaposi's",
"sarcoma-associated",
"herpesvirus",
"cell",
"lines",
"molecular",
"biology",
"tegument",
"proteins",
"biochemistry",
"post-translational",
"modification",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"organisms"
]
| 2018 | A genome-wide screen of Epstein-Barr virus proteins that modulate host SUMOylation identifies a SUMO E3 ligase conserved in herpesviruses |
RIG-I-Like Receptors ( RLRs ) sense cytosolic viral RNA to transiently activate type I IFN production . Here , we report that a type I IFN inducible DExD/H helicase , DDX24 , exerts a negative-regulatory effect on RLR function . Expression of DDX24 specifically suppressed RLR activity , while DDX24 loss , which caused embryonic lethality , augmented cytosolic RNA-mediated innate signaling and facilitated RNA virus replication . DDX24 preferentially bound to RNA rather than DNA species and influenced signaling by associating with adaptor proteins FADD and RIP1 . These events preferentially impeded IRF7 activity , an essential transcription factor for type I IFN production . Our data provide a new function for DDX24 and help explain innate immune gene regulation , mechanisms that may additionally provide insight into the causes of inflammatory disease .
The principal purpose of the immediate innate immune response is the rapid synthesis and secretion of type I interferons ( IFNβ and IFNα ) as well as inducing other key host defense genes [1] , [2] . Innate immune signaling is initiated by pattern recognition receptors ( PRRs ) that specifically recognize pathogen-associated molecular patterns ( PAMPs ) , which are unique to microbes and rarely found in the host [3] . Upon virus infection , virus-associated PAMPs such as genomic DNA or RNA can be recognized by PRRs , which initiate signaling events leading to the synthesis of type I IFN and to the transcription of other IFN-inducible genes ( ISG's ) in a paracrine or autocrine manner [1] , [3] . Three major PRRs participating in the recognition of viral PAMPs have been identified as the Toll-like receptors ( TLRs ) , retinoic acid-inducible gene I-like ( RIG-I ) receptors ( RLRs ) and cytosolic DNA receptors [4]–[6] . TLR3 , 7 , 8 and 9 are TLRs responsible for recognizing viral nucleic acid species . TLR3 is known to recognize dsRNA , while TLR7 and TLR8 recognize ssRNA , and TLR9 senses unmethylated CpG DNA [3] . The RLR family comprises three receptors , RIG-I , melanoma differentiation-associated gene 5 ( MDA5 ) and laboratory of genetics and physiology 2 ( LGP2 ) [5] . RIG-I recognizes 5′-triphosphate RNA and short forms of the synthetic dsRNA analog poly I:C , whereas MDA5 is mainly responsible for recognizing longer dsRNA species [7]–[9] . LGP2 has also been shown to exert anti-viral properties [10] , [11] . The mechanisms of cytosolic viral DNA recognition is the least characterized pathway but is known to involve AIM2 which activates inflammatory response [12] , [13] and STING ( TMEM173/MITA/MPYS ) which has been shown to be critical for cytosolic DNA triggered type I IFN production as well as pro-inflammatory gene production [14]–[17] . Furthermore , NLRs ( nucleotide-binding domain , leucine-rich repeat containing ) , key modulators of the inflammasome , also recognize dsRNA . Rather than inducing IFN , they are important for triggering IL-1 beta production , and play important roles in the host response to a wide range of microbial pathogens , inflammatory diseases , and autoimmune disorders [18] . Upon recognition of cytosolic viral RNAs , RLRs are recruited to the adaptor protein IPS-1 ( MAVS , Cardif , VISA ) located on mitochondria and peroxisomes , through the mitochondrial-associated membrane ( MAM ) , a distinct membrane compartment that links the endoplasmic reticulum to mitochondria [19]–[24] . IPS-1 subsequently recruits TBK-1/IKKi , which phosphorylates IFN-regulatory factor 3 ( IRF3 ) and 7 ( IRF7 ) . Phosphorylated IRF3 and IRF7 dimerize and translocate into the nucleus to trigger the production of type I IFN and other primary innate immune genes . NF-κB and AP-1 are also activated in this pathway and required for the optimal type I interferon production [25] , [26] . RLR signaling is facilitated by co-regulators such as Fas associated death domain ( FADD ) and receptor interacting protein 1 ( RIP1 ) , originally identified as crucial players in apoptotic and inflammatory signaling pathways [27] , [28] . It has been demonstrated that RIP1 and FADD can form a complex with TRADD and IPS-1 following RNA virus infection to co-ordinate signaling [29] . Indeed , loss of FADD or RIP1 leads to defective type I IFN production and significantly increased susceptibility to RNA virus infection [28] , [30] . RIP1 may exert its effects with FADD by associating with IRF7 to promote its activation . IRF7 activity is known to be reduced in RIP1 deficient MEFs , again suggesting that RIP1 is a positive regulator in RLR dependent signaling [31] . However , the detailed mechanisms of RLR signaling that involves FADD and RIP1 remains to be fully clarified , including the mechanisms of negative regulation . Here , we report that an IFN-inducible helicase referred to as DDX24 is able to negatively regulate the RLR-signaling pathway . DDX24 attenuates the RLR signaling possibly through competing with RIG-I for binding of RNA . Moreover , it also disrupts the IRF7/RIP1 interaction , which is required for robust innate immune gene activation . Our data provides significant molecular insight into the control of innate immunity and may provide information into the causes of inflammatory disease .
Given that FADD is important in regulating innate immune signaling , we searched for proteins that may mediate FADD activity . We noted that FADD had been reported by proteomic analysis to be potentially associated with a helicase , referred to as DDX24 [32] , [33] . DDX24 , an 859 amino acid Asp-Glu-Ala-Asp ( DEAD ) -box ATP-dependent RNA helicase , lacks CARD domains typical of RIG-I and MDA5 though has an N-terminal region rich in glutamic acid and lysine residues ( Figure 1A and Figure S1A ) . Over 90 percent of homology at the amino acid level is shared between human DDX24 and mouse DDX24 ( Figure S1A ) . It is also of interest to note that unlike the majority of DEAD-helicases , DDX24 has several potential interferon-regulated transcription sites in its promoter region , similar to RIG-I , MDA5 and LGP2 , such as STAT1 and IRF7 binding sites ( Figure 1B ) . To further evaluate the possible association of DDX24 with FADD , we used FADD as bait in a yeast-two hybrid assay , with DDX24 as prey . This study confirmed that full length DDX24 could specifically associate with FADD in this system ( Figure 1C ) . To complement this approach , we carried out co-immunoprecipitation analysis by co-overexpressing DDX24 and FADD in 293T cells and found DDX24 and FADD could interact with each other ( Figure 1D ) . We next examined whether endogenous FADD and DDX24 could co-immunoprecipitate from primary human HUVEC ( human umbilical vein endothelial ) cells . First , we confirmed that the available antibody could recognize DDX24 in HUVECs . To accomplish this , we suppressed DDX24 expression in HUVECs by RNAi and then carried out an immunoblot analysis . We observed a single band of molecular weight of approximately 100 kDa in untreated cells that was not evident in the RNAi treated HUVECs , confirming DDX24 expression in this cell-type ( Figure 1E , left panel ) . Subsequent analysis using anti-DDX24 antibody similarly indicated co-association of DDX24 and FADD ( Figure 1E , right panel ) . An extended analysis suggested that FADD associated with the amino terminal region of DDX24 containing the DExD/H box helicase ATP binding domain ( Figure 1F ) . Expression studies suggested that DDX24 was expressed in a wide variety of cells ( Figure 1G and Figure S1B ) . We observed that DDX24 and FADD were localized both nucleus and cytoplasm by immunofluorescent and fractionation experiments ( Figure S2A and S2C ) . Interestingly , we observed elevated cytoplasmic DDX24 6 and 9 hours following poly I:C treatment ( Figure S2B ) . Consistently , increased levels of DDX24 co-precipitated with FADD after 6 hours of poly I:C treatment in MEF , suggesting a stronger DDX24/FADD association following RLR activation ( Figure S2D ) . Treatment of murine embryonic fibroblasts ( MEFs ) or HUVECs with type I IFN or poly I:C confirmed that DDX24 was IFN-inducible , similar to RIG-I , and the induction was STAT1 dependent as predicted ( Figure 1H–K and Figure S2E ) . Thus , DDX24 appears to be an interferon-inducible DEAD-box helicase that can associate with FADD . To evaluate the importance of DDX24 in possible innate immune signaling regulation , we overexpressed DDX24 in MEF cells and confirmed that the type I IFN inducer , poly I:C , could initiate the transcription of an IFNβ promoter driven luciferase construct ( Figure 2A ) . However , we noted that the overexpression of DDX24 could inhibit poly I:C's ability to activate the IFNβ promoter ( Figure 2A ) . Similarly , DDX24 was seen to inhibit the ability of vesicular stomatitis virus ( VSVdM , with a defective matrix protein that enables virus-mediated type I IFN production ) to activate IFNβ-luciferase production [34] ( Figure 2B ) . This was confirmed at the level of IFNβ mRNA expression and endogenous protein expression ( Figure 2C–2F ) . Thus , DDX24 may negatively regulate innate immune signaling processes that activate the type I IFN promoter . To further evaluate this possibility , we treated 293T cells with RNAi that targeted DDX24 . After confirming knockdown by immunoblot , we observed that loss of DDX24 led to an increase in polyI:C's ability to activate luciferase under control of the IFNβ promoter ( Figure 3A ) . A similar effect was observed following infection with VSVdM ( Figure 3B ) . We complemented this approach by knocking down DDX24 in MEF cells and observed a comparative effect ( Figure 3C–3F ) . These data again indicate that DDX24 can exert a negative-regulatory effect on cytosolic RNA signaling events in the cell . Microarray analysis of polyI:C transfected MEF cells treated with or without anti-DDX24 RNAi confirmed that DDX24 suppressed IFNβ production as well as IFN-inducible genes such as CXCL10 ( Figure 3G ) . Thus , DDX24 may be a negative regulator of RNA-mediated type I IFN transcriptional activation ( Figure 3G ) . Since we observed that loss of DDX24 may augment RNA-mediated type I IFN production , we surmised that DDX24 suppression may in turn repress RNA virus replication because of prevalently high type I IFN levels . Two types of recombinant VSV were used for these studies to help facilitate this analysis . First we suppressed DDX24 in MEFs cells and infected with VSV expressing a luciferase gene ( VSV-Luc ) . We noted that suppression of DDX24 robustly prevented the replication of this virus as determined by luciferase levels and viral titers 8 or 24 hours post infection ( Figure 4A and 4B ) . As a control , we knocked down RIG-I , which is a key sensor of VSV-mediated type I IFN production [5] . Suppression of RIG-I unsurprisingly lead to an increase in VSV-Luc replication opposite to DDX24 suppression ( Figure 4A and 4B ) . Similar results were obtained using VSV expressing GFP ( VSV-GFP ) in MEFs treated with DDX24 siRNA ( Figure 4C–4E ) . HUVEC cells treated with RNAi targeting DDX24 likewise lead to a decrease in VSV replication ( Figure 4F ) . Collectively , our data indicate that DDX24 exerted suppressive effects on RNA-mediated innate immune signaling events in the cell controlled by RLRs . RIG-I and MDA5 have been reported to sense viral negative and positive stranded RNA species , respectively , to initiate the activation of NF-κB and IRF3 signaling pathways that results in the production of type I IFN and other primary innate immune genes [9] . Briefly , viral RNA interacts with the helicase domains of RIG-I/MDA5 , inducing conformational changes that enable the CARD domains to interact and activate IPS1/MAVS [19]–[22] ( Figure 5A ) . It was thus plausible that DDX24 could suppress RIG-I/MDA5 activity by competing with RNA activators . To evaluate this , we first examined the ability of DDX24 to directly associate with RNA species . Our analysis indicated that c-Myc-tagged DDX24 could be precipitated from 293T cells using biotinylated polyI:C or ssRNA compared to DNA species ( Figure 5B–5D and Figure S2 ) . Furthermore , in vitro binding experiments indicate that DDX24 could bind to the VSV-G gene transcript through its helicase C domain ( Figure 5F and G ) . We also observed that DDX24 was able to compete with RIG-I for RNA representing the VSV-G gene transcript ( Figure 5E ) . Thus , it is plausible that DDX24 may exert some negative-regulatory effect by directly competing for RIG-I/MDA5's activators . However , to evaluate whether DDX24 could directly inhibit RIG-I and/or MDA5 by means other than by plausible sequestration of RNA ligands , we transfected dRIG-I or dMDA5 constructs into 293T cells co-transfected with increasing amounts of DDX24 . The dRIG-I/dMDA5 constructs we used represented the amino terminal CARD domains ( dRIG-I aa 1–284; dMDA5 aa 1–349 ) that following expression , oligomerize without the requirement of RNA to activate IPS-1/MAVS and type I IFN signaling [5] , [30] . Thus , we were also able to determine whether DDX24 impeded RIG-I function independent of RNA . This experiment indicated that DDX24 could indeed inhibit dRIG-I and dMDA5-mediated signaling which is independent of RNA sequestration ( Figure 5H ) . Overexpressed IPS1/MAVS is also known to oligomerize to activate type I IFN activity without the requirement of RIG-I/MDA5 . Since similar experimentation indicated that DDX24 could also inhibit IPS1/MAVS signaling , we conclude that DDX24's regulatory influence was also exhibited downstream of RIG-I/MDA5 and IPS1/MAVS interaction ( Figure 5H ) . Indeed , we found that DDX24 could affect TBK1 activity suggesting that this helicase exerts its influence at the levels of RIG-I/MDA5's ability to regulate IRF3/7 dependent type I interferon production . We had also previously reported that FADD and RIP1 regulate RNA-mediated innate immune signaling [28] . We noted that expression of FADD or RIP1 could greatly facilitate dRIG-I's ability to augment type I IFN production ( Figure 5I and 5J ) . However , DDX24 was again able to inhibit this signaling process ( Figure 5I and 5J ) . This would suggest that DDX24 competes in innate immune complexes comprising FADD/RIP1 that are required for efficient type I IFN signaling . It should be noted that we observed that DDX24 predominantly inhibited type I IFN signaling compared to other signaling pathways . For example , DDX24 did not affect p53 signaling , general luc gene transcription or induce cell death ( Figure 5K and Figure S4 ) . Thus DDX24 is able to sequester RNA activators of type I IFN activation as well as additionally interfere with downstream signaling events that control RIG-I/MDA5 function . The regulation of IRF3 and NF-κB pathways are complex though are known to involve RIG-I/MDA5 invoking TBK1 to principally phosphorylate IRF3 which then along with activation of the NF-κB pathway activates IFNβ transcription [3] . These events produce type I IFN-inducible IRF7 that binds to and activates several IFNα genes to augment type I IFN production in a positive feedback manner [35] . To additionally evaluate the mechanisms of DDX24 activity , we inquired whether DDX24 affected IRF3 or IRF7 activity . However , we observed that DDX24 did not affect poly I:C or VSVdM dependent IRF3 phosphorylation or dimerization ( Figure 6A ) . Furthermore , we did not observe an inhibition of constitutively activated IRF3 ( SA ) –mediated induction of IFNβ-luc during overexpression of DDX24 ( Figure S5B and Figure S5C ) . To next evaluate the influence of DDX24 on IRF7 function , we examined the effects of DDX24 on an IFNα4 promoter driving luciferase which is strongly activated by IRF7 rather than IRF3 [36] . This experiment indicated that overexpression of DDX24 in 293T cells inhibited the ability of RIG-I and IRF7 to fully activate the IFNα4 promoter ( Figure 6D ) . Conversely , loss of DDX24 in 293T cells by RNAi treatment enhanced RIG-I and IRF7's ability to activate the IFNα4 promoter ( Figure 6E ) . Overexpression of FADD was observed to facilitate IRF7 signaling , which was significantly increased in the absence of DDX24 ( Figure 6F ) . To extend this analysis , we evaluated whether DDX24 could affect TBK1/IKKi's ability to phosphorylate IRF7 . Co-expression analysis confirmed that DDX24 could affect IRF7 phosphorylation ( Figure 6B ) . Thus , DDX24 could influence IRF7 function , which is a pivotal positive regulator of type I IFN production . Given this we next examined whether DDX24 disrupted FADD or RIP interactions with downstream components of the RLR pathway , which would eventually lead to attenuated IRF7 activation . We first tested the interaction of FADD and IRF7 , but no association was observed by co-immunoprecipitation after overexpressing both proteins in 293T cells ( data not shown ) . However , it has been reported that RIP1 could recruit IRF7 to the signaling complexes to positively regulate IRF7 activation [31] . We thus explored the association of DDX24 , RIP1 and IRF7 through co-immunoprecipitation experiments . Indeed , c-Myc-tagged DDX24 was observed to interact with RIP1 ( Figure 6C ) . However , no interactions between DDX24 and IRF7 were detected ( data not shown ) . Since IRF7 and DDX24 have been reported to bind to RIP1 , it is possible that DDX24 could disrupt the interactions between IRF7 and the RIP1 kinase . To examine this , we co-expressed RIP1 and IRF7 with or without DDX24 . This experiment confirmed that RIP1 could associate with IRF7 and that this event could be disrupted in the presence of DDX24 ( Figure 6G ) . It is therefore plausible that DDX24 may compete with RIP1 to bind IRF7 . Reciprocal co-IP experiments confirmed that RIP1 precipitated with IRF7 , an event that was impeded in the presence of DDX24 ( Figure 6H ) . Consistent with our findings , we observed increased RIP1 interaction with IRF7 when we knocked down DDX24 by siRNA in 293T cells ( Figure 6I ) . Collectively our data indicates that DDX24 negatively regulates RLR signaling at least in part by affecting RIP1/IRF7 interactions . To study the function of DDX24 in vivo , DDX24 knockout mice were generated using a gene trapping strategy . Mouse embryonic stem ( ES ) cells with the DDX24 genomic locus disrupted by a β-galactosidase/neomycin cassette placed between exons 6 and 7 were microinjected into C57BL/6 blastocysts to create chimeric mice ( Figure S4A ) . Although DDX24+/− mice appeared normal and were fertile , no viable homozygous mutant mice were observed in the first 300 pups derived from DDX24+/− intercrosses , suggesting that DDX24 deficiency results in embryonic lethality ( Table 1 and Figure S4B ) . In fact , we were not able to retrieve DDX24−/− MEFs from over 200 embryos isolated between E12 . 5 to E9 . 5 ( Table 1 and Figure S4C ) . We were only able to identify DDX24−/− embryos between E3 . 5 and E8 . 5 suggesting an embryonic lethal occurrence between these times ( Figure S4D ) . By histological analysis , we were able to identify embryos with severe embryogenesis defect at E7 . 5 ( Figure 4E ) . These data suggest that loss of DDX24 is lethal to embryos before E8 . 5 , suggesting an important role for this helicase , perhaps in addition to facilitating RLR signaling .
It has been previously reported that RIG-I/MDA5 associates with FADD , RIP1 and TRADD in a signaling complex that also involves IPS-1 and which is required for the stimulation of host defense genes [1] , [29] . Our data here indicate that an uncharacterized helicase DDX24 may negatively regulate these processes . DDX24 belongs to the DExH/D family , which contains at least 59 proteins conserved from bacteria to humans . DExH/D helicases are broadly involved in many RNA related processes such as transcription , translation , ribosome biogenesis and RNA transportation [37] . In addition , RIG-I , MDA5 and LGP2 have been reported to be key sensors in RNA-virus mediated innate immune signaling processes . However , several other RNA helicases have also been implicated in the regulation of host defense processes such as DHX9 , DDX60 and DDX3x which reportedly act as alternate RNA sensors in myeloid dendritic ( mDCs ) among other cells [38]–[40] . DDX1/DDX21/DHX36 complex may similarly be involved with sensing both short and long poly I:C via the adaptor TRIF [41] . Further , DHX9 and DHX36 have been reported to sense cytosolic CpG-DNA via MyD88 in pDCs [42] . Finally , DDX41 has been reported to sense intracellular DNA [43] . Although a majority of these helicases are able to bind RNA or DNA and may act as sensors , it is also possible that certain members such as DDX3x may also function downstream of nucleic acid recognition to affect multi-protein signaling complexes required for efficient primary innate immune gene transcription [44] , [45] . Although we observed that DDX24 could specifically bind to dsRNA or ssRNA , DDX24 did not show a binding affinity to DNA species , which suggests a preferable role in RNA dependent signaling . Furthermore , although DDX24 exerts RNA binding activity that is similar to the RLR's , DDX24 was able to inhibit the function of constitutively active RLR's that lacked RNA binding ability , such as dRIG-I and dMDA5 . This data would suggest that aside from being able to sequester RNA activators , DDX24 may additionally act as a competitive protein to disrupt the appropriate formation of downstream signaling complexes . For example , it has been reported that RIP1 is able to associate with IRF7 and is important for IRF7 activation . Possibly , RIP1 helps to recruit IRF7 to signaling complexes comprising TBK-1/IKKi . In this study , we have confirmed an interaction between RIP1 and IRF7 and demonstrated specific inhibition of this interaction by DDX24 . Cells overexpressing DDX24 exhibited attenuated TBK-1/IKKi dependent IRF7 phosphorylation , which would be in agreement with this model . Additionally , we investigated the potential interacting partner of DDX24 , but we did not observe direct evidence of DDX24 binding to IPS-1 or TRADD ( data not shown , Figure S2F ) , which are reported as FADD binding molecules [19] , [29] . Although no direct interactions between FADD and IRF7 were observed , FADD was able to greatly facilitate IRF7-dependent IFNα4 promoter activation . Moreover , this regulatory function is significantly enhanced in the absent of DDX24 , indicating the requirement for FADD on IRF7 activity . Given the fact that RIG-I associates with both FADD and RIP1 without disrupting the interaction , it is plausible to propose that FADD-RIP1 association acts as scaffold that facilitates the recruitment of IRF7 to the IPS-1 based signaling complex . Losing either FADD or RIP1 may lead to a disruption of the signaling the complex , which could potentially lead to insufficient recruitment of IRF7 to this complex . DDX24 , anchored into this complex by interacting with FADD and RIP1 likely disrupts the binding of IRF7 , and thus negatively regulates this pathway . DDX24 appears to be relatively ubiquitously expressed , not just prevalent in hematopoietic lineages , and it can be induced by IFNβ and poly I:C treatment . DDX24 was strongly associated with the nucleolus . However , our results from immunofluorescent and fractionation experiments suggest that FADD and DDX24 converge in the cytoplasm in quantities sufficient enough to regulate innate immune signaling events . Interestingly , this DDX24/FADD association is strengthened after the cells were treated with poly I:C for 6 hours as detected by co-immunoprecipitation in MEF cells . Consistently , we have also observed increased DDX24 in the cytoplasm in our immunofluorescent experiments . This suggests the translocation of DDX24 following RLR activation or that the cytoplasmic DDX24 is induced by poly I:C treatment . Similar to the immunofluorescent experiments , we observed elevated levels of cytoplasmic DDX24 after 24 hours of poly I:C stimulation . However , we did not observe significant changes in levels of nuclear DDX24 , which suggests a function unrelated to IFN signaling for nuclear DDX24 that remains unclear . Moreover , the interferon inducible characteristic of DDX24 provides extra clues on the mechanism of DDX24's regulation of RLR signaling . Indeed , we observed induction of poly I:C-inducible genes following 9 hours poly I:C treatment when DDX24 was knocked down , supporting our hypothesis that DDX24 negatively regulates IRF7 dependent signaling which happens at a late stage of RLR signaling . Interestingly , we also observed an inhibitory role of DDX24 at the early time point , which is possibly caused by DDX24-mediated inhibition of NF-κB . Furthermore , we observed a decrease in DDX24 knockdown efficiency , especially in the 9 hour-treated group , possibly due to the induction of DDX24 by the treatment , which could have attenuated the effects of DDX24 at later times following poly I:C treatment . Therefore , the generation of DDX24 deficient cells would be crucial for clarification of this observation and further analysis . Previously , FADD has been shown to play key roles in multiple cytoplasmic signaling processes such as apoptosis and innate immunity . The possible role of DDX24 in alternate mechanisms is underscored by demonstrating that DDX24−/− mice die early in embryonic development ( <e7 . 5 ) . Indeed , we were unable to obtain DDX24−/− MEFs for analysis . The severe embryonic phenotype clearly indicates that DDX24 is crucial to early embryogenesis . However , RLR deficient animals remain viable again suggesting that DDX24 has alternate functions in the cell [9] , [46] . It is noteworthy that FADD deficient mice also exhibit early embryonic lethality , but at later stage from E9 . 5 to E11 . 5 possibly due to a severe failure of cardiac development [47] . It remains to be seen whether DDX24 plays a role in these processes . Finally , it is worth noting that the negative regulation of innate immune gene transcription may play an important role in inflammatory disease . For example , defects in cytosolic DNA signaling that facilitate enhanced STING activity can lead to lethal inflammatory disease [48] . Mutations in the genes that negatively regulate STING have been found in patients suffering from severe inflammatory disease [49] . It remains to be seen whether defects in the RLR pathway may exert similar effects . In summary , we have further characterized a new IFN-inducible DExD/H helicase DDX24 that is involved in a negative-feedback role to regulate the RLR pathway and type I IFN production . Further understating these processes may shed light in causes of infectious disease and plausibly inflammatory disorders involving enhanced innate immune gene activity .
The protocol was reviewed and approved by the University of Miami Institutional Animal Care and Use Committee ( IACUC ) ( Protocol number: 11-043 RENEWAL 03 ) . This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . Yeast two-hybrid assay was performed using the Matchmaker Gal4 two-hybrid system ( Clontech ) according to the manufacturer protocols . Briefly , full length hFADD was cloned into the yeast bait vector , pGBT9 . pGBT9-FADD transformed yeasts were transfected with another plasmid pGAD-hDDX24 or pGADT7 vector . Yeasts with both plasmids transfected were selected on SD-Leu/-Trp plates . Single colonies on the plates were picked up and seeded on SD-His/-Leu/-Trp plates for verification of interactions between molecules . Gene trap mutated DDX24 embryonic stem cells ( RRK059 ) were purchased from BayGenomics . Chimera mice were produced by microinjection of heterozygous ES cells into E3 . 5 C57BL/6 blastocysts that were subsequently transferred to pseudo-pregnant foster mothers . Chimera male mice were bredwith control female C57BL/6 mice to transmit the mutated ddx24 alleleto the germline . Heterozygous mice were interbred to obtain DDX24−/− mice . Genotyping was performed by genomic DNA based PCR . The primer 5′-GCTAATTCCTGCCTGTATGACCTT-3′ was used in combination with either 5′-ATTCAGAGCAGGTTAACCCAGGAC-3′ for the wild type ddx24 allele , or the primer 5′-GACTGGTGAGTACTCAACCAAGTC-3′ for the mutant allele . Animals were generated at the University Of Miami School Of Medicine Transgenic Core Facility ( Miami , FL ) . Mice were allowed to access food and water freely and were housed at an ambient temperature of 23°C and at a 12 hour light/dark cycle . Animal care and handling was performed as per Institutional Animal Care and Use Committee guidelines . Timed matings were performed between mature DDX24 heterozygote mice and MEFs were obtained using a standard procedure . Briefly , embryos from E8 . 5 to E12 . 5 days were dissected free of surrounding tissues , washed in PBS and the heads and livers were removed . Each individual embryo was completely trypsinized for 15 minutes and cultured separately . MEFs were genotyped using same gDNA based method as described above . Primary DDX24+/+ and DDX24+/− MEFs before passage 6 were used for all experiments . Uteri from timed mating were formalin fixed and embedded in paraffin . Sections were cut at 6 µm . After dewaxing in xylene and rehydration in a series of graded ethanol , intermittent sections were stained with hematoxylin and eosin ( H&E ) in order to identify mutant embryos . Immuno-staining was performed after a heat and citrate based antigen retrieval of sections . The primary antibody used to detect Disabled-2 ( Dab2 ) was a mouse monoclonal antibody from BD Transduction Laboratories . The secondary antibody applied was a peroxidase conjugate ( Vector Labs , CA ) and sections were counterstained with hematoxylin . After timed mating of DDX24 heterozygotes , preimplantation embryos were flushed at E3 . 5 . The embryos were cultured upon an irradiated fibroblast feeder layer in ES cell media ( DMEM with 15% FBS , 1000units/ml ESGRO , 1× non-essential amino acids , 2 mM L-glutamine , 50 IU/ml penicillin and 50 mg/ml streptomycin ) until they had attached . They were then trypsinized , routinely fed and assayed for the appearance of ES cell clones . 293T cells ( ATCC ) , WT MEFs , DDX24+/+ and DDX24+/− MEFs were grown in DMEM supplemented with fetal bovine serum ( 10% ) and penicillin/streptomycin ( 1% ) . HUVEC cells were purchased from ATCC and cultured in EGM-2 media supplied with growth factors obtained from EGM Bullet kit . All cells were maintained at 37°C in a 5% CO2 laboratory incubator subject to routine cleaning and decontamination . Poly I:C ( Amersham ) was reconstituted in PBS at 2 mg/ml , denatured at 55°C for 30 min , and allowed to anneal at room temperature before use . Antibodies were obtained from following sources: rabbit anti-DDX24 ( A300-697A Bethyl ) ; mouse anti-FLAG M2 antibody , rabbit anti-c-Myc , mouse anti-HA ( Sigma ) ;rabbit anti-IRF3 , rabbit anti-GFP , mouse anti-GAPDH ( Santa Cruz Biotechnology , Inc . ) ; rabbit anti-phospho-IRF3 ( Upstate ) ; rabbit anti-phospho-IRF7 ( Cell signaling ) ; rabbit anti-Fibrillarin ( ab5821 ) , mouse anti β-actin ( Abcam ) ; mouse anti-RIP1 ( BD Science ) ; mouse anti-FADD ( cell signaling ) . Control scrambled ( D-001206–01-80 ) , mRIG-I ( L-065328-01 ) , mDDX24 ( L-042299-01 ) and hDDX24 ( L-010397-01 ) smart pool siRNAs were purchased from Dharmacon/Thermo Scientific . Plasmid or poly I:C transfection in 293T cells or MEFs were conducted using Lipofectamine 2000 ( Invitrogen ) transfection reagents in Opti-MEM ( Invitrogen ) according to the manufacturer's manual . HUVEC and MEFs siRNA transfections were performed using AMAXA HUVEC and MEF Nucleofectin Kit 1 according to the manufacturer's recommendations ( AMAXA Biosystems ) . Indiana strain of VSV was used in all experiments . Constructed VSVs ( VSV-GFP , VSV-luc and VSVdM ) were constructed in our lab . Briefly , keep the medium serum free during first two hours post infection and change the medium back to full medium until harvest . Expression vectors ( pcDNA3 . 1 , Invitrogen ) FLAG or GFP tagged RIG-I , MDA-5 , dRIG-I , dMDA5 , IPS1 , TBK1 , RIP1 , FADD , IRF3 were generated in our lab by polymerase chain reaction . N-terminal c-Myc-tagged or FLAG-tagged plasmids DDX24 , DDX24-N ( AA 1–577 ) and DDX24-C ( 578–859 ) were generated using pCMV-tag system from stratagene . Same fragment of DDX24 were also ligated into the EcoRI and SalI site of pGADT7 vector to generate pGADT7-DDX24 for yeast two hybrid experiments . Other plasmids used in this study , c-Myc-RIP1 , c-Myc-IRF7 , FLAG-IRF7 , and IFNα4-luc ( S Ning ) ; IFN-β Luc ( J Hiscott ) . IRF3 ( SA ) was purchased from invivogen . Total RNA was isolated by using RNeasy RNA extraction kit ( Qiagen ) and cDNA synthesis was performed with random hexamer primers using 5 µg of total RNA ( Invitorgen ) Real-time PCR was performed using a LightCycler 2 . 0 instrument and the TaqMan Gene Expression Assays ( Applied Biosystems ) : mIFNβ ( Mm00439546 ) , mIFNα2 ( Mm00833961 ) , mIRF7 ( Mm00516788 ) , mRIG-I ( Mm00554529 ) , mDDX24 ( Mm00517454 ) , hIFNhβ ( Hs00185375 ) , hIRF7 ( Hs00185375 ) , Luciferase ( 4331348 customized ) . Relative amount of mRNA was normalized to the 18S ribosomal RNA level in each sample . Alternatively , SYBR green systems from New England Biolabs ( DyNAmo SYBR Green qPCR Kit ) were used for human ddx24 detection . The human RNA samples used for ddx24 profiling were purchased from Ambion ( AM6000FirstChoice Human Total RNA Survey Panel ) . Primers used for human ddx24 were: Forward 5′-GCCGAATTTACAGGAATTAAAACTG-3′; Reverse 5′-GTCATCCACTACCAGGGCACCTGAGC-3′ . Primers used for human gapdhwere Forward 5′-atgacatcaagaaggtggtg-3′; Reverse 5′-cataccaggaaatgagcttg-3′ . Relative amount of mRNA was normalized to the gapdh RNA level in each sample . Briefly , 293T cells or MEF cells were seeded on 24-well plates and were transiently transfected with 50 ng/100 ng of the luciferase reporter plasmids together with a total of 600 ng of various expression plasmids or empty control plasmids . As an internal control , 10 ng/20 ng pRL-TK plasmids expressing Renilla protein was transfected simultaneously . Twenty four or 36 hours later , cells were lysed by adding 100 µl/well of Cell culture lysis buffer ( CCLR ) , and luciferase activity in the total cell lysate was measured by illuminometer . Immunofluorescence experiments were performed as follows . Briefly , cell monolayers were fixed in 4% paraformaldehyde for 10 min , washed with PBS and permeabilized with 0 . 2% Triton-X 100 in PBS for 5 min . After blocking in PBS containing 10% FBS for 20 min , samples were incubated 1 hour at 37°C or overnight at 4°C with appropriate primary antibody . After PBS washing for three times , samples were incubated 1 hour with secondary antibodies conjugated with Cy3 , Cy5 or FITC at 1: 200 dilution . Cells were washed again and incubated with 0 . 5 µg/ml DAPI solution for 5 min . Samples were then washed with PBS and mounted using prolong gold antifade reagent from invitrogen . Pictures were taken using an Olympus fluorescent microscope equipped with a digital camera and a Zeiss LSM-510 Confocal Laser Scanning Microscope . Cells were lysed in RIPA buffer on ice , followed by centrifugation . Cell lysates were separated by SDS-PAGE , transferred to PVDF membranes , and subjected to immunoblotting . For co-immunoprecipitation , expression vectors were transfected into 293T cells for 36 to 48 hours , cells were lysed in ice-cold NP40 IP buffer ( 50 mM Tris , pH 8 . 0 , 150 mM NaCl , 0 . 5% NP-40 , 50 mM NaF , 0 . 1 mM Na3VO4 , 1 mM DTT ) or RIPA buffer with protease inhibitors ( 100 mM PMSF , Leupeptin , Aprotinine , Pepstatin ) , and cell lysates were precipitated with appropriate amount of FLAG-M2 antibody ( SIGMA ) or endogenous antibodies overnight at 4°C . Following day , 30 µl of Protein G was added and incubated for 3 hours . RNA pulldown assay was performed using the same lysis buffer and method . Cell lysates were incubated precipitated with biotin labeled RNA/DNA for 3 hours before precipitated with streptavidin beads . All precipitates were washed with lysis buffer 3 times and proteins were released by 2× Sample Buffer after boiling and analyzed by SDS-PAGE . 293T cells recovered from 6-well dishes were lysed in 100 µl of native lysis buffer ( 50 mM Tris-Cl , pH 8 . 0 , 1% NP40 , 150 mM NaCl , 100 mg/ml leupeptin , 1 mM PMSF , 5 mM orthovanadate ) . Ten µg of protein was mixed with 2× native PAGE sample buffer ( 125 mM Tris-Cl , pH 6 . 8 , 30% glycerol , bromphenol blue ) and subjected to electrophoresis on non-denaturing 7 . 5% polyacrylamide gels . ELISAs for mouse IFNβ were performed using supernatants from cells where values are expressed as pg/ml ±S . E . as calculated from a standard curve derived from recombinant IFNβ provided in the ELISA kit ( PBL Interferon Source ) . Total RNA were purified and transcripts analyzed by Illumina Sentrix Chip Array ( Mouse WG6 version2 ) . Promoter analysis is performed by Genomatix MutInspector software . Statistical significance of differences in cytokine levels , mRNA levels , viral titers , and luciferase intensity in reporter assay and VSV-Luc-infected cells were determined using Student's t-test . The following information was arranged in the format of “Symbol” , “Accession numbers ( Human , Mouse ) ” . RIG-I ( O95786 , Q6Q899 ) ; MDA5 ( Q9BYX4 , Q8R5F7 ) ; IPS-1/MAVS/CARDIF/VISA ( Q7Z434 , Q8VCF0 ) ; TBK-1 ( Q9UHD2 , Q9WUN2 ) ; IKKi ( Q14164 , Q9R0T8 ) ; RIP1 ( Q13546 , Q60855 ) ; FADD ( Q13158 , Q61160 ) ; DDX24 ( Q9GZR7 , Q9ESV0 ) ; TRADD ( Q15628 , Q3U0V2 ) ; IRF3 ( Q14653 , P70671 ) ; IRF7 ( Q92985 , P70434 ) ; TLR3 ( O15455 , Q99MB1 ) ; TLR7 ( Q9NYK1 , P58681 ) ; TLR8 ( Q9NR97 , P58682 ) ; LGP2 ( Q96C10 , Q99J87 ) ; TLR9 ( Q9NR96 , Q9EQU3 ) ; STING/TMEM173/MITA/MPYS ( Q86WV6 , Q3TBT3 ) ; CXCL10 ( P02778 , P17515 ) ; IFNβ ( P01574 , P01575 ) ; DHX9 ( Q08211 , O70133 ) ; DDX60 ( Q8IY21 , E9PZQ1 ) ; DDX3X ( O00571 , Q62167 ) ; TRIF ( Q8IUC6 , Q80UF7 ) ; DHX36 ( Q9H2U1 , Q8VHK9 ) ; MyD88 ( Q99836 , P22366 ) ; DDX41 ( Q9UJV9 , Q91VN6 ) . | Innate immunity is the first and most rapid host defense against virus infection . Viral RNAs , which are generated during RNA virus replication in host cells , can be recognized through RIG-I-Like Receptors ( RLRs ) to transiently produce type I interferon , which further induce abundant interferon stimulated genes ( ISGs ) to clear viral infection . However , uncontrolled innate immune responses cause inflammatory diseases that are detrimental to the host . Therefore , a balanced innate immune response is critical to maintain homeostasis of the host . Thus , RLR signaling is tightly regulated by both positive and negative regulators . DDX24 , a helicase reported in this study , is an ISG that exerts an inhibitory effect on RLR dependent signaling . DDX24 hijacked adaptor proteins FADD and RIP1 in host cells to suppress viral RNA dependent interferon production and facilitated RNA virus replication in certain cells . Moreover , DDX24 deficient mouse embryos exerted early embryonic lethality , suggesting an important role for this helicase , perhaps in addition to regulating RLR signaling . In all , our results elucidate the role of DDX24 in RLR dependent signaling , and may shed light on innate immune gene regulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | DDX24 Negatively Regulates Cytosolic RNA-Mediated Innate Immune Signaling |
While some evidence supports the beneficial effects of integrating neglected tropical disease ( NTD ) programs to optimize coverage and reduce costs , there is minimal information regarding when or how to effectively operationalize program integration . The lack of systematic analyses of integration experiences and of integration processes may act as an impediment to achieving more effective NTD programming . We aimed to learn about the experiences of NTD stakeholders and their perceptions of integration . We evaluated differences in the definitions , roles , perceived effectiveness , and implementation experiences of integrated NTD programs among a variety of NTD stakeholder groups , including multilateral organizations , funding partners , implementation partners , national Ministry of Health ( MOH ) teams , district MOH teams , volunteer rural health workers , and community members participating in NTD campaigns . Semi-structured key informant interviews were conducted . Coding of themes involved a mix of applying in-vivo open coding and a priori thematic coding from a start list . In total , 41 interviews were conducted . Salient themes varied by stakeholder , however dominant themes on integration included: significant variations in definitions , differential effectiveness of specific integrated NTD activities , community member perceptions of NTD programs , the influence of funders , perceived facilitators , perceived barriers , and the effects of integration on health system strength . In general , stakeholder groups provided unique perspectives , rather than contrarian points of view , on the same topics . The stakeholders identified more advantages to integration than disadvantages , however there are a number of both unique facilitators and challenges to integration from the perspective of each stakeholder group . Qualitative data suggest several structural , process , and technical opportunities that could be addressed to promote more effective and efficient integrated NTD elimination programs . We highlight a set of ten recommendations that may address stakeholder concerns and perceptions regarding these key opportunities . For example , public health stakeholders should embrace a broader perspective of community-based health needs , including and beyond NTDs , and available platforms for addressing those needs .
A major challenge facing health systems in low-resource settings is the often ad-hoc and vertically silo-ed approach to organizational structure . Many global health leaders are promoting the integration of vertical programs into a shared delivery infrastructure to strengthen the efficiency , effectiveness , and sustainably of health systems and to optimize resources with the goal of promoting equitable and synergistic health improvements [1–3] . However , the term integration is not consistently defined and often incorporates a variety of ideas , including continuity of care , inter-organizational relationships , disease management , and others [4] . In addition , a variety of external and internal factors , including local and national politics and complex institutional pressures are also important drivers of , and barriers to , successful integration [5] . Understanding facilitators and barriers of integration within existing integrated programs offers the opportunity to improve public health integration efforts globally . The neglected tropical diseases ( NTDs ) are an example of a group of 17 diseases in which synergistic integration may be ideal given their high degree of geographic and population overlap . Furthermore , five of the world’s most prevalent NTDs are generally considered “tool ready” and are primarily controlled through mass drug administration ( MDA ) , including: lymphatic filariasis ( LF ) , onchocerciasis , schistosomiasis , soil transmitted helminths ( STH ) , and trachoma [6] . Schistosomiasis and STH MDA programs are typically delivered to pre-school and school-age children via school-based delivery platforms . LF , onchocerciasis , and trachoma programs , on the other hand , are typically delivered via community-based delivery platforms . LF and onchocerciasis MDA programs are uniquely delivered using community directed intervention ( CDI ) strategies . Triple drug administration ( TDA ) is one option for integrated treatment through the simultaneous provision of albendazole , ivermectin , and praziquantel or azithromycin . Studies suggest that TDA is clinically effective and cost-efficient , and the WHO recommends TDA of albendazole , ivermectin and praziquantel in areas that have had one to two previous rounds of treatment [8 , 11 , 18] . However , TDA is not widely utilized in co-endemic countries , even those with established integrated NTD programs . In addition to preventative treatment strategies such as MDA there are a number of other activities that can potentially be integrated between NTD programs . NTDs that are co-endemic in the same geographic area can integrate activities such as mapping , vector control , community sensitization and education , health worker training , surveillance , monitoring and evaluation , and disability management . In January 2012 an array of partners gathered to endorse the World Health Organization ( WHO ) NTD Roadmap and launch the London Declaration , a commitment to pursuing control or focal elimination of select NTDs by 2020 [7] . Integration of NTD programs is considered to be one of the most promising approaches for achieving the London Declaration goals , and integrated NTD programming has been endorsed and recommended by the WHO for NTD endemic countries to optimize program implementation [8 , 9] . However , while there is evidence suggesting beneficial effects of integration on NTD program coverage [10] and costs [11] , there is negligible information regarding best practices in operationalizing integration . Similarly there has been minimal examination of the potential detrimental impacts of integration . ” The lack of a systematic analysis of integration experiences and complexity limits the ability of programs to optimize NTD program implementation in co-endemic areas [12] . Given that integrated NTD programs are now encouraged in endemic countries , there is currently a unique opportunity to learn from stakeholders as they engage in integrated programming . In this study we aimed to identify how perceptions regarding the role , effectiveness , and implementation of integrated NTD programs differ among various NTD stakeholders . We also aimed to describe program integration and best practices for implementing integrated programs from each stakeholder’s perspective . We focus specifically on the NTDs for which MDA is the standard of care due to geographic and interventional congruencies , as well as to limit the scope of this analysis .
We conducted a qualitative cross-sectional study to identify and harmonize NTD stakeholder approaches to integrated program delivery . We identified seven primary stakeholder groups involved in and affected by integrated programming of NTDs , including partners at multilateral organizations , funding partners , implementation partners , national Ministry of Health ( MOH ) teams , districts MOH teams , volunteer rural health workers ( known as community drug distributors , CDDs ) , and community members participating in MDA campaigns . The conceptual model in Fig 1 outlines the stakeholders involved in integrated NTD programs as well as simplified descriptions of their driving interests and influences on NTD integration . The sampling methodology utilized the NTD stakeholder group ( N = 7 ) as the unit of analysis , as determined through a maximum variation approach . An overarching tenet of maximum variation is the understanding that each stakeholder group must be considered separately , and thus a distinct sampling frame and sampling strategy was identified for each group . The sampling frame for stakeholders working in multilateral organizations and funder organizations was individuals deemed as subject matter experts ( i . e . key informants ) . In this study , subject matter experts are defined as influential individuals recognized in the NTD domain as thought-leaders or policy-influencers , who often present and publish in the field . Subject matter experts must have been working in the field of NTDs for ten or more years . The sampling frame for the implementation partners and national MOH workers was limited to subject matter experts working in five highly co-endemic countries ( i . e . endemic to all five MDA-indicated NTDs ) and thus these stakeholders work in a variety of NTD-endemic countries throughout sub-Saharan Africa and Asia . The sampling frame for district and rural MOH workers was even further limited to one country of in-depth focus . The country of in-depth focus was selected because it is endemic for all five NTDs for which MDA is the standard of care , noted above . It is located in sub-Saharan Africa , and the population is largely rural . The name of the country cannot be provided , in accordance with Institutional Review Board ( IRB ) stipulations . The recruitment strategy used to identify all stakeholders—with the exception of community members—was purposive quota sampling of mutually exclusive key informant groups [13] . The strategy was deemed appropriate as this research aims to capture and equally value the range of NTD stakeholder perspectives , from influential key informants at the global level to program implementers at the local level . Only individuals working on two or more of the five NTDs for which MDA is the standard of care were recruited . For community members , the sampling strategy was random purposive sampling . During four community-based MDA campaigns in the country of focus , community members gathered to receive NTD health education and treatment at schools and health posts . Education was provided prior to treatment , and then treatment was delivered to community members throughout the day . During the campaigns , a translator or CDD made an announcement that community members may be approached and invited to participate in interviews regarding their experiences with NTD programs . The translator or CDD was instructed to approach every 5th woman who participated in MDA treatment , starting with the first woman treated . This probabilistic sampling strategy for community members was used as this group is much larger than the other stakeholder groups and we aimed to capture a more representative sample of these NTD stakeholders . Due to the large number of individual stakeholders in each stakeholder group , it was not possible to ensure complete data saturation . Semi-structured key informant interviews with a mix of respondent and informant style questions were used . Participants were asked both to describe the process through which they engage in integrated programs as well as their recommendations to others regarding how integration should be pursued . They were asked to explain their rationale for these recommendations , describe factors that facilitate success , and describe barriers that challenge successful integration . Many of the questions were similar across stakeholders in order to compare and contrast answers . However , questions were also specific to the experiences and roles of the particular stakeholder . All interviews were audio recorded and participants were required to provide verbal consent prior to the commencement of the interview . This methodological approach was granted exemption status from the University of Washington IRB committee under a minimal risk determination status . Exemption was granted with the understanding that no country names or personal identifiers of interviewees would be available . Transcripts were uploaded to the software program Atlas ti . ( V . 7 2012 ) . Coding of key themes involved a mix of applying in-vivo open coding and a priori thematic coding from a start list [14] . After the first round of coding was completed , a second round was conducted using an adaptation of the constant comparative coding method . Constant comparative methods help conceptualize and describe the variety of responses in the data [15] . This set of codes and memos were used to identify themes that highlight common trends across responses [14] . In qualitative research , themes are recurrent unifying concepts or statements about the topic of inquiry [16] . Within each theme group , we identified responses that were similar , opposing , or unique to each stakeholder group . By undertaking this two stage analysis process we utilized a mix of case-oriented and variable-oriented analytical strategies [17] .
A total of 41 interviews were conducted with stakeholders: 2 from multilateral organizations , 2 from funding partner agencies , 4 implementation partners , 5 national MOH health workers , 6 MOH district health workers , 8 CDDs , and 14 community members . Salient themes varied by stakeholder however dominant themes arising during analysis were relevant to ( 1 ) variations in definitions of “integration” , ( 2 ) differential effectiveness of integration according to the specific NTD activities integrated ( 3 ) community perceptions of integrated NTD programs , ( 4 ) the influence of funders on NTD integration , ( 5 ) perceived facilitators of integration , ( 6 ) perceived barriers to integration , and ( 7 ) the effect of integration on health system strength . Within each of these themes , we identified where feedback was similar , differed , or was unique to each stakeholder group . In the section that follows these themes are discussed in greater detail , with a selection of supportive quotes provided in context . All participants , with the exception of community members , were asked to provide a definition of the term “integration” within the context of NTD programs . In general , definitions were broad , even within stakeholder groups , and participants often provided rationales for integration rather than a working definition . Health workers , particularly CDDs , described integration of specific activities while individuals at multilateral , funder , and implementing organizations focused on definitions of integration relevant to upstream planning and measurement . A number of participants noted the lack of an existing definition but cautioned against being too definitive , citing the need to maintain flexibility in applications of the term . Conversely , several stakeholders noted the danger of having such ubiquitous , vague terminology . The lack of consensus regarding the definition of integration was evident in the many ways stakeholders used the term to describe various clinical , political , and organizational processes . However , the most common definition of integration across stakeholders was the act of coordinating specific activities and interventions when mutually relevant to two or more health programs . Some stakeholders also identified integration as a way to transfer knowledge between programs , particularly for diseases at different stages of elimination . Multiple stakeholders argued that there was a common misconception that integrating NTD programs for the five MDA-indicated diseases simply requires co-delivering all designated drugs . Rather , there are a number of activities involved in NTD control in addition to drug delivery ( Fig 2 ) . These activities can be integrated with one another or with complementary co-interventions such as safe water , sanitation , or nutritional interventions . But stakeholders unanimously noted that not all of these activities are contextually or scientifically appropriate for integration across NTDs . All of the specific program activities mentioned by stakeholders during interviews are discussed below . Community members had unique feedback regarding the acceptability and feasibility of integrated NTD programs . Amongst those interviewed there was unanimous fear of NTD-associated morbidities , particularly elephantitis and visible worms in stool . This fear translated to strong community demand for MDA programs . There were three points of constructive feedback from community members regarding MDA in general . First , during pre-treatment community sensitization and education it is often not clear which diseases are being treated simultaneously or during different MDA treatment rounds . Second , MDA programs can demand considerable time from community members to participate in all education , registration , and treatment activities . Depending upon the treatment schedule , these NTD activities can occur several times per year and are likely one of many community-based health campaigns in which community members are asked to participate . Lastly , some community members desired more treatments , particularly expanding praziquantel treatment to adults in schistosomiasis endemic areas . These points of feedback were echoed by volunteer CDDs . Women noted that men in particular are likely to be absent from MDA events when they perceive that the campaigns require a large amount of time throughout the year for each separate program . In general it was difficult for community members to discuss their preferences regarding integrated activities relative to vertical activities . When asked , one community member shrugged and said , “This is government decision , when they say to receive drugs , we just accept it , and we don’t have an option . ” ( CM9 ) . When asked about their potential participation in integrated treatments in the future , some community members feared taking too many tablets at once during integrated campaigns . Additionally , two community members mentioned concerns that further integration would actually make treatment days longer , especially if there are only a few health workers present at each campaign . However , the majority of community members independently noted that they are accustomed to taking several medications at once , repeatedly providing the example of malaria treatment regimens and pain killers . Most of those in favor of integrated MDA drug delivery focused on the time savings that would be accrued . Several district and rural health workers noted that community members are unknowingly already accustomed to integrated NTD programs through co-delivery of schistosomiasis and STH services in schools and LF and onchocerciasis programs in communities . One CDD was surprised that some might consider programs such as LF and onchocerciasis separately in the first place , “From our side , we only call it the oncho program…though I suppose we know the filariasis is there” ( CDD 2 ) . Similarly , although several programs have successfully integrated activities in some co-endemic countries , it is clear that integrated programming has led to some confusion on the part of CDDs and , as a result , community members . One CDD erroneously explained , “We combine treatments because Mectizan is there to prevent disease while albendazole is there to cure worms that are already too much so it’s good to combine them . One is for prevention and the other is for cure for which is already there” ( CDD3 ) . Many community members cited the expanded program on immunizations ( EPI ) as an example for why integrated community-based programs are acceptable and advisable . Community members noted that “first clinic” ( EPI and growth monitoring during child health weeks ) was appropriately integrated because all activities shared the common goal of promoting child growth . These community perceptions were echoed by district workers in the same communities . Interview participants identified four main facilitating factors of integration including the need for efficiency with limited resources , strong central leadership , conception and launching of new integrated programs , and continued relevancy post elimination . The need for program efficiency was the most frequently cited driver for and benefit of integration during stakeholder interviews . Specifically , MOH stakeholders at every level described financial and time efficiencies as the primary factors encouraging program integration in low income settings . For example , several national MOH health workers noted that when they don’t have adequate funding to hold a CDD training on one disease , they will utilize resources earmarked for another disease to provide an integrated training . Almost all of the stakeholders interviewed , with the exception of community members , stated that integrated strategies will be most effective if they are first institutionalized at the national level of a health system before being launched at district or local levels . Most multilateral , implementation , and MOH stakeholders said integration at the national level is most efficient if there is a single NTD coordinator overseeing disease-specific NTD focal persons . Such a leader can promote cross-disease coordination and improved communication at the national level which , according to MOH stakeholders , will trickle down to the peripheries . Additionally , several multilateral and donor partners remarked that it is easier to launch new integrated programs in a country than it is to integrate extant disease-specific programs due to entrenched institutional identities . One multilateral stakeholder noted that , in this capacity , NTD Steering Committees endowed with decision making power ( not simply endorsing entities ) are critical in the integration process to ensure that Integrated NTD Master Plans are available with actionable recommendations and integrated activities . Additionally , as countries work towards NTD elimination , some MOH stakeholders noted there are political incentives for national and sub-national health workers to integrate their workloads and programs . Specifically , it allows programs and personnel to have continued relevancy and expertise after single diseases have been locally eliminated . Stakeholders identified four primary barriers to effective NTD program integration: ( 1 ) a fear amongst MOH workers that they will lose their jobs or recognition of their work , ( 2 ) external timelines or funder pressures that do not allow for a lengthy integration processes , ( 3 ) tensions between school and community-based delivery proponents , and ( 4 ) the fact that some strong or well-funded programs do not see integration as a “win-win” . These barriers are driven by political and administrative factors . Stakeholders at all levels reported that integration is often not pursued or is inefficiently pursued when stakeholders perceive that they won’t be able to maintain some degree of disease-specific autonomy and recognition of program-specific markers of success . While leveraging strong programs to increase the coverage or efficiency of weaker programs is a primary rationale for integration , several stakeholders expressed concern that integrating strong and weak programs could be to the detriment of strong programs . However , when asked , stakeholders could not provide examples of a situation where this phenomena had occurred . Several district level MOH workers and implementing partners also expressed concerns that successful integrated programming might be used as a rationale for minimizing the financial support provided to sub-national disease control teams . MOH stakeholders unanimously noted that vertical funding mechanisms limit the ability for NTD programs to integrate at national and sub-national levels . Some stakeholders recommended that funders and implementation partners take more responsibility for ensuring greater collaboration between disease-focused groups while other stakeholders recommended that governments take more responsibility for ensuring coordination amongst NTD partners working in a country . Funder influence is discussed in greater detail in Theme 6 . Another identified barrier to achieving more extensive NTD program integration is the operational and political divides between community-based MDA programs ( LF , onchocerciasis , and trachoma programs ) and school-based MDA programs ( schistosomiasis and STH programs ) . For example , according to several implementation and multilateral stakeholders TDA has not been introduced or scaled-up in any country largely due to the political paradigm shift that would need to take place prior to providing treatments to children in community-based settings . Many of the community members recommended that all treatment take place during community campaigns so that parents can supervise their children . Without advocating for one delivery system over another , the majority of implementation partners and MOH stakeholders commented on the political complexity of choosing either school or community based delivery over the other . Some national stakeholders argued that more school-age children will miss integrated treatments if targeted in the community , while a number of CDDs argued that integrated community-based treatments could expand treatment to children who do not attend school . Several stakeholders noted that shared responsibility between Ministries of Education and Health means that there is an authority vacuum where both entities have a stake but neither claims ownership for the disease program . Lastly , many stakeholders highlighted that while integration is necessary and desirable in some locations , stakeholders should still track disease specific outcomes . A lack of disease-specific outcomes limits funder or partner ability to track impact and may diminish their desire to pursue integrated programs . Specifically implementation partners and multilateral stakeholders recommended disease-specific coverage metrics as indicators of integrated program success . Similarly , although stakeholders noted opportunities for NTDs to integrate with other community-based health programs within a comprehensive primary healthcare system , this broad integration is often not undertaken because stronger , well-funded programs often do not see integration with NTDs as a “win-win” . Many MOH stakeholders cited that integrated programs would be more effective if funder and partner resources were more easily integrated . A common theme arising in this regard was related to incentives for volunteer CDDs; given that programs rely on a largely volunteer workforce , the sustainability of the programs is compromised by changes in volunteer incentivizing . MOH stakeholders recommended that community volunteer incentive schedules align across NTDs . They also recommended that NTD incentives align with other disease programs , noting that well-funded programs such as HIV may compromise less-funded programs such as NTDs by offering larger incentives to health workers for similar work . According to the stakeholders interviewed , funders may unintentionally thwart integration due to their desire for limited resources to be spent efficiently , causing different funders to cover different geographies of a country in a piece-meal manner . Additionally , because some funders only provide resources for specific diseases their resources may not be available for integrated programs . As one stakeholder points out , there is a delicate balance that a funding organization must maintain to ensure that it has fidelity to its mission statement while ensuring maximum health impact . However most of the funder , multilateral , and implementation partner stakeholders remarked that funder and implementation partner culture appears to be changing , with increased emphasis on country needs and integrated programming . Newly established NTD coordinating committees might be a facilitating factor . Stakeholders reported that NTD integration can affect NTD delivery systems as well as the health-system more generally . NTD-specific effects include transferring knowledge from successful programs to scaling programs and strengthening the presence of NTDs on the global health agenda . Broader health system effects include general improvements in community based healthcare programs , encouraging a culture of learning from other disease delivery platforms , and progress towards universal primary healthcare coverage . A number of stakeholders across levels recommended that in order to maintain NTD scale-up success , integration be pursued in a “fluid manner” ( IMP3 ) based on the maturation of different disease programs and disease dynamics at a local level . In this regard , stakeholders repeatedly provided LF and STH as examples of programs for which integration might facilitate a transfer of knowledge within the NTD system; geographies that have successfully controlled LF and are transitioning to post-MDA surveillance while attempting to scale-up nascent STH programs could integrate activities to avoid losing progress made by LF programs . These disease programs have particular co-dependencies and a natural opportunity for synergy due to the fact that they utilize MDA of the drug albendazole . Several stakeholders also discussed how integration helped NTDs develop a larger presence on global health agendas and include a larger number of stakeholders . Various stakeholders noted that the benefit of integration to NTDs in terms of advocacy and fundraising has been indispensable . According to a number of MOH stakeholders , NTD program integration could also result in broad health system improvements if it encourages health worker efficiencies . These efficiencies could improve community participation in community-based healthcare activities generally . Multilateral and implementation partners noted that the act of implementing integrated health programs forces public health workers to learn from what works , to look to programs and platforms that have had success , and to think about what can be learned from them . Several stakeholders argued that a focus on learning from integrated programs will benefit both NTDs and country health systems more broadly . A subset of stakeholders across levels also mentioned that establishing an integrated community-based NTD platform affords the opportunity to provide preventative healthcare for NTDs and other conditions outside of formal treatment settings . This is important for achieving universal health coverage of preventative healthcare services and for disease elimination efforts in particular .
Recommendation 1: Countries may wish to establish a single NTD Coordinator for all NTDs for which MDA is the standard of care . Stakeholders recommended that this NTD coordinator could oversee disease-specific program managers and resource allocation , with frequent communication with the NTD Steering Committee . A leader with an integrated perspective and relevant competencies appears critical in contexts with changing vocabulary and foci , and can help facilitate alignment amongst stakeholders within a given health system [21] . Recommendation 2: Country-level NTD Steering Committees should be established or strengthened where already present . Weak committees should be strengthened by increasing decision making capacity and requiring implementation partners to present to the Committee prior to program launchings . In this way Steering Committees could also help coordinate donor funding or support MOH program managers working with multiple donors . Stakeholder theory suggests that meeting the multiple needs of stakeholders on Steering Committees would maximize overall systems effectiveness [22] . Steering Committees should help develop or review long-term integrated Master Plans that must include detailed planning regarding specific activities that will be integrated , how they will be integrated , and how integrated activities may be uniquely assessed for impact . According to study stakeholders , most existing integrated Master Plans do not meet these criteria . Recommendation 3: Healthcare integration in resource-limited settings is facing a similar definitional challenge that integrated care implementers have faced in higher income countries for decades [4] . Thus the NTD Steering Committee in each country should establish contextual definitions and rationales for integration . Rationales for integration should include evidence or hypotheses relevant to ( 1 ) scientific rationales for integration , ( 2 ) administrative rationales for integration , and ( 3 ) health system rationales for integration . According to strategic change management theory , articulating a shared need for change with consideration of diverse stakeholder roles positively influences implementation effectiveness and structural change efforts [20] . This process will provide a standardized manner for the Steering Committee to consider why they are pursuing integration and any potential unintended consequences . These rationales should be shared with all other levels of the MOH and Ministries of Education , where relevant . Recommendation 4: Funders and implementation partners should empower and work with NTD Steering Committees by coordinating closely with the Committee and ensuring that MOH priorities are paramount [23] . Evidence suggests that funder conditions can hamper resource allocation decisions following strategic planning or produce additional workload for health workers in the field [24] . Thus partners must also ensure that they are working closely with other institutions and organizations to ensure funds and activities are complementary . Contributing to a single integrated Master Plan may be a way to promote inter-organizational coordination from the onset . Recommendation 5: NTD integration is complicated and , given that many of the challenges to effective integration are procedural and behavioral rather than scientifically based , change management activities should be undertaken at the national level and used to clearly communicate integration definitions , rationales , and Master Plan strategies to peripheral district and local MOH offices . While evidence from other studies found topdown management of NTD programs to result in negative feedback from peripheral levels [23] , according to district-level stakeholders in this study , a strong core management structure is necessary to promote sustainable consensus on integrated NTD programming . There is a general understanding amongst stakeholders that effective integration is facilitated by having a collective vision , shared strategies , and common culture [25] . A number of change management strategies such as promoting a shared mental model of integrated care , for example , would create an inter-organizational and inter-professional environment necessary for delivery of integrated care [20] . Recommendation 6: Many community members cited EPI as an appropriately integrated program because all distinct health interventions clearly share the common goal of promoting child health . This suggests that , in order to promote program acceptability , it is up to NTD public health practitioners to ensure that integrated programs have a clearly expressed unified goal and that this goal is being communicated to the public . In communicating this unified message , community members should be made fully aware of what diseases they are receiving treatment for and why . This may involve changing the structure of current CDD training curriculum and prioritizing community sensitization efforts . Recommendation 7: The “justification for integrated delivery systems is to meet patients’ needs rather than providers” [26] . Public health stakeholders should embrace a broader perspective of community-based health needs and available platforms for addressing those needs . For example , most district health workers recommended that child health weeks , which are well attended and accepted , be expanded to include other community-based healthcare delivery activities , including MDA delivery . This will necessitate some technical adjustments to standard clinical procedures . Other evaluations of community-based programs such as EPI have also concluded that shared platforms may have broad health system benefits [27] . Ultimately multisectoral integration between activities such as nutritional intervention , water , sanitation , and MDA will result in more effective programs and thus shorter necessary durations of treatment [28] . However , broader approaches to community-based delivery must be designed carefully in fragile health systems so as not to induce operational problems affecting program quality [23] . Recommendation 8: Where appropriate , MOHs should incorporate TDA into drug delivery schedules . Most community members described the implementation method as hypothetically acceptable and other health programs , such as polio immunization initiatives , have similarly identified community dissatisfaction when the number of intervention rounds are high [29] . Yet many health workers do not know that TDA of albendazole , ivermectin , and praziquantel is approved by the WHO for the simultaneous treatment of onchocerciasis , LF , STH , and schistosomiasis [8] . According to stakeholders , promoting TDA will require more specific guidelines from the WHO as well as more deliberate attempts to bridge the political divide between school and community-based treatment approaches . These efforts will require multi-year advanced planning to sync previously vertical NTD programs within an integrated platform . Recommendation 9: Incentives and support systems for community volunteers and health workers should be aligned across NTDs and other community-based disease programs . This will require cooperation by multiple funding partners . By integrating approaches to volunteer recruitment and maintenance there may be greater sustained engagement overall . Such necessary activities would also align with the WHO-endorsed Joint Commitment to Harmonized Partner Action for Community Health Workers and Frontline Health Workers [30] . This recommendation could also be considered a part of process integration . Recommendation 10: Stakeholders should standardize and redesign subnational reporting systems to capture information regarding which NTD program activities are integrated with other activities . Stakeholders reported that current data collection methods are confusing for health workers and supervisors working on integrated programs , and aggregated field data do not provide information regarding the effectiveness of specific integrated activities . Such data are necessary for linking particular integrated interventions to programmatic outcomes or health impacts in non-experimental settings . These data must be shared promptly and transparently with the WHO to ensure global disease control benchmarks are accurately monitored . One limitation of this analysis is that on one-on-one interviews of community members and health workers took place in private rooms within clinics or in places of work . Responses may have been biased if interviewed individuals felt that their feedback may reach employers or community leaders . A second limitation is that the data analysis did not involve multiple coders and thus intercoder reliability was not possible to establish . Lastly , the stakeholders’ views and opinions reflect a subset of the NTD community and are not representative of all stakeholders . While patterns emerged , complete data saturation by stakeholder group was not sought , nor achieved . Further geographically-specific research should be conducted prior to the introduction of any relevant policy changes . Application of a social science approach allowed us to provide a theoretical understanding of a number of similarities and differences between different stakeholder perceptions of the complex process of NTD integration . In general , there was greater variation between groups than within and stakeholder groups provided unique perspectives , rather than contrarian points of view , on the same topics . The stakeholders identified more advantages to integration than disadvantages , however there are a number of both unique facilitators and challenges to integration from the perspective of each stakeholder group . These findings provide both explanatory as well as meditative information to NTD integration stakeholders . The ten recommendations provided draw from the qualitative data to highlight structural , process , and technical opportunities to maximize stakeholder interests while promoting more effective and efficient integrated NTD elimination programs . | Neglected tropical diseases are a group of parasitic , viral , and bacterial diseases that are often co-endemic in low resource settings . Five of these diseases ( lymphatic filariasis , onchocerciasis , schistosomiasis , soil transmitted helminths , and trachoma ) are addressed specifically through a method called mass drug administration , where entire at risk populations are targeted with preventative drug treatments . Because of the geographical and interventional overlap between these diseases , many experts recommend program integration as a method for accelerating their control or elimination . However the optimal approaches for operationalizing integrated programing has not been systematically assessed . We undertook a cross sectional qualitative research study with neglected tropical disease stakeholders to understand different stakeholder approaches to and perspectives on program integration . The stakeholders highlighted different definitions of the term “integration” , the differential effectiveness of specific activities when integrated , the influence of integration on community member engagement , the influence of funders on integrated programming , facilitators and barriers to effective integration , and the effects of integration on health system strength . Our analysis suggests that there are three types of integration to consider: structural , process , and technical . We use these categories to make ten recommendations to stakeholders that might be used to improve integrated neglected tropical disease programming moving forward . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"research",
"funding",
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"parasitic",
"diseases",
"health",
"care",
"bacterial",
"diseases",
"pharmaceutics",
"eye",
"diseases",
"neglected",
"tropical",
"diseases",
"onchocerciasis",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"health",
"systems",
"strengthening",
"government",
"funding",
"of",
"science",
"drug",
"delivery",
"helminth",
"infections",
"schistosomiasis",
"science",
"policy",
"ophthalmology",
"trachoma",
"health",
"care",
"policy"
]
| 2016 | Integrated Healthcare Delivery: A Qualitative Research Approach to Identifying and Harmonizing Perspectives of Integrated Neglected Tropical Disease Programs |
Various studies showed that chemotherapy can control schistosomiasis morbidity , but association of measures ( water supply , sewage disposal and increase of socioeconomic conditions ) is necessary for transmission control . A survey dealing with socioeconomic conditions , snail survey , contact with natural waters , and clinical and stool examinations was undertaken at an endemic area in the State of Minas Gerais , Brazil . The methodology used was the same for both evaluations ( 1981 and 2005 ) . Four hundred and seventy-five out of 1 , 474 individuals studied in 1981 could be contacted . From these , 358 were submitted to stool examination , and 231 of them were clinically examined . Patients eliminating S . mansoni eggs in their stools were treated . The results showed that the prevalence rate in Comercinho , a municipality of the State of Minas Gerais , Brazil , was substantially reduced to 70 . 4% and 1 . 7% in 1981 and 2005 , respectively , as well as the frequency of the hepatosplenic form ( 7% to 1 . 3% ) after five treatments effectuated between 1981 and 1992 . No other new case of this form was detected from 1981 onwards . Another important aspect to be considered was the improvement of people's living standard that occurred in the region after more than two decades' efforts ( better housing , professional skill and adequate basic sanitation ) . The control of morbidity and very significant decrease of schistosomiasis transmission in an area until then considered as hyperendemic was possible by means of association of successive specific treatments of the local population , together with the construction of privies , water supply in the houses and improvement of socioeconomic conditions .
Schistosomiasis is a social disease , found at poor rural regions and periphery of cities , with a precarious socioeconomic development , where the inhabitants have frequent contact with contaminated waters , as well no available of adequate sewerage system . WHO considers schistosomiasis as the second only to malaria in socioeconomic importance worldwide , and the third more frequent parasitic disease Public Health importance . [1] The main necessary step recommended for reduction of schistosomiasis morbidity is the treatment of individuals living in endemic areas [2] . The program for national control of schistosomiasis was launched in Brazil in 1975 , by SUCAM ( “Superintendência de Campanhas de Saúde Pública” ) – Ministry of Public Health , by means of the “Programa Especial de Controle da Esquistossomose ( PECE ) ” ( Special Program for Schistosomiasis Control ) , which directed its activities for the chemotherapeutic treatment with oxamniquine in large scale ( more than 13 million people were treated ) . As molluscicide , niclosamide was used , but in lower scale and irregular manner . Sanitation , safe water supply and health education were also measures adopted , but with less frequency [3] . Various early studies demonstrated that improvement of the sanitary conditions and treatment of positive patients contribute to reduce morbidity and prevalence of the disease [4]–[11] . A study carried out in Comercinho/MG , Brazil , in 1974 , clearly shows these facts . In that year , a staff from the Laboratory of Schistosomiasis , Research Center René Rachou/FIOCRUZ , under the leadership of one of the authors ( NK ) , performed the first survey on schistosomiasis mansoni ( prevalence of 70 . 4% ) . Census of the population , mapping of the town , clinical and stool examinations of the patients infected with S . mansoni were performed . However , the treatment of the local population could not be administered as it was intended , due to the appearance of lethal cases in Brazil with the use of hycanthone , the antischistosomal drug of choice at that occasion . After the discovery of a novel drug – oxamniquine – the researchers came back to Comercinho in 1981 . On that year , besides the above mentioned measures , other ones were taken , such as: identification of the intermediate host , socioeconomic survey , research on contact with natural waters , clinical examinations of the population , and intradermal reaction for this group , besides specific treatment of infected patients [12] . In 1984 and 1986 , the individuals that presented S . mansoni eggs in the feces , detected by means of examination of the local residents performed in the preceding year , were once more treated . In 1988 a new re-evaluation was undertaken according to the same methodology [13] . In 1992 , Rocha and Katz [14] re-examined the conditions of the area after five treatments with oxamniquine ( from 1981 to 1991 ) . From that date onwards the Prefecture of the town was in charge of the program for the control of schistosomiasis , and the treatments continued to be administered at the local Public Health Center by local physicians and technicians ( horizontalization of the Program for Schistosomiasis Control ) . In 2005 , a new clinical-epidemiological survey of the population living in the area in 1981 was carried out , focused on the following priorities: identification of the intermediate host; parasitological , clinical and socioeconomic evaluations of the population and evolution of contact with natural waters . In the present paper we compare the results of the last evaluation ( 2005 ) with data related to the inhabitants of the region in 1981 , when people were treated with antischistosomal drug for the first time .
Comercinho is a little town located at the Northeast of the State of Minas Gerais , macro-region of Jequitinhonha , Brazil , at a distance of 714 Km from the capital of the state . In 2005 , the population was estimated in 10 . 181 inhabitants and 3 . 340 of them were living in the urban area , where this study was done . The urban center has three public buildings pertaining to the Prefecture ( 1 for odontological attendance , the other ones for the “Programa de Saúde da Família ( PSF ) ” – Program for Family Health – and for the “Programa de Controle da Esquistossomose - PCE” – Program for the Control of Schistosomiasis . COPASA ( “Companhia de Saneamento de Minas Gerais” ) is responsible for the water supply and sewerage system . The household waste is daily collected in all the urban area , and the solid residues are deposited in a landfill situated 2 Km far from the urban center . The patients were informed about this new study , and a signed written informed consent was obtained from all patients ( including from parents/guardians for all the 7–14-year old children ) before admission to the study . This study was approved by the Ethical Committee for Human Research of the Research Center René Rachou/FIOCRUZ ( 02/2006-CEPSH/CPqRR ) , and by the Ethical Committee for Human Research of the Santa Casa Hospital , in Belo Horizonte/MG ( Statement n° 016/2006 ) . The inhabitants that participated in the study performed in 1981 , and were still living in the area , were interviewed by the technicians of the “Programa de Saúde da Família ( PSF ) ” . They answered a socioeconomic questionnaire , and they were also invited to be submitted to clinical and stool examinations . Collection of snails was performed within the urban area ( Sapê and Areia brooks ) . The snails were sent to the Mollusc Room at the Research Center René Rachou/FIOCRUZ in order to be identified and evaluated in relation to S . mansoni infection , by means of light exposure and by crushing between two glass plates . The staff of the “Programa de Saúde da Família ( PSF ) ” visited the housings of the participants in the study performed in 1981 , and the interview was held with the owner or user of the housing . The socioeconomic survey considered the following items: a ) insertion of the family head into the productive system; b ) individual occupation; c ) working place; d ) place of birth; e ) type of housing; f ) source of water supply . These topics were assessed according to the same parameters described by Costa in 1983 [12] . The patients received a recipient to collect the feces , identified with the same plot number attributed to them in 1981 . A stool sample was collected for preparation of two slides by the Kato-Katz method [15] . The eggs were counted , and an arithmetic average of the number of eggs per gram of feces was considered as an individual result . In order to know the number of eggs per gram of feces ( epg ) in the community the geometric average was used . The positive patients ( eliminating S . mansoni eggs in the feces ) received chemotherapeutic treatment for schistosomiasis and/or other helminthiases . Seven hundred fifty-nine school children ( 7 to 14-year-old ) were examined for evaluation of the current situation related to schistosomiasis . Clinical examination was performed by means of anamnesis and abdominal palpation . The clinical classification adopted was: type I ( intestinal ) , type II ( hepatointestinal ) or type III ( hepatosplenic ) [16] . The patients clinically examined answered a questionnaire on contact with natural waters , for evaluation of the frequency and reason for contact . The data were grouped as follows: washing clothes , fetching water , bathing , leisure ( swimming and/or fishing ) , professional activities ( watering garden , removing sand and crossing the stream ) . The chi-square test was used for comparison of frequency distribution ( 1981–2005 ) . The test was performed with 95% confidence , using the program SPSS version 11 . 5 .
One hundred and sixteen out of the 181 captured snails were found to be alive , and were identified as Biomphalaria glabrata . None of the alive snails examined were infected with Schistosoma mansoni . In 2005 , 475 out of 1 . 474 individuals that have participated in the study carried out in 1981 could be contacted . Stool and clinical examinations were performed in 1 . 329 and 836 , and in 358 and 231 individuals , in 1981 and 2005 , respectively . Table 1 shows comparison of the socioeconomic survey obtained in 1981 and 2005 . As can be seen in Table 1 , significant improvements were attained , such as substantial increase in the number of housings with safe water supply provided by the Public Service ( from 33 . 7% in 1981 to 96% in 2005 ) . Waste disposal using cess-pits or flush toilets was increased from 71 . 7% to 97 . 6% , provided by the population . In 1981 , only 34 . 2% of the housings were classified as type A ( considered of better quality ) , and 97 . 6% in 2005 . The proportion related to the heads of the households considered as skilled workers showed also a significant increase ( 6 . 6% to 22 . 8% ) ( Figure 1 ) . The school-children were used as indicators of the current situation of schistosomiasis mansoni in 2005 . Thus , the prevalence rate estimated in 759 school-children was 1% for S . mansoni , 1 . 7% for hookworms , and 0 . 4% for Ascaris lumbricoides . Table 2 shows the data related to the distribution of indicators for schistosomiasis mansoni in the re-evaluated group . The infection rate decreased dramatically ( from 70 . 4% in 1981 to 1 . 7% in 2005 ) . Infection rate in the age of group ( 30–40 years ) in 1981 was 69 . 2% in 130 persons; in 2005 it was 4 . 1% in 98 persons . The geometric average of eggs per gram of feces was 334 epg and 172 epg , in 1981 and 2005 , respectively . The patients presenting more than 500 epg in 1981 were 36 . 6% , whereas only one case could be detected in 2005 . Table 3 shows the results obtained with clinical evaluation in the studied years . The clinical form Type I was detected in 67 . 9% and 95 . 2% of the patients in 1981 and 2005 , respectively . The clinical forms Type II and III were observed in 25% and 3 . 5% , and 6 . 8% and 1 . 3% , in 1981–2005 , respectively . Among the signs and symptoms evaluated , abdominal pain and diarrhoea were the most reported at both study periods . Blood in the stools was present in 50% of the studied population in 1981 , but no similar case was reported in 2005 . Table 4 shows the results on contact with natural waters in Comercinho . In 1981 , the daily contact with natural waters was reported by 62% of the population , whereas in 2005 by just 25% . Also , a marked decrease related to the use of natural waters was detected , when the bi-weekly or less frequency was considered ( 51 . 9% in 2005 , and 13 . 7% in 1981 ) . As far as the reasons for contact with natural waters were concerned , in 1981 , 21% and 3 . 7% were due to leisure and professional activities , and in 2005 these activities were reported by 27 . 2% and 44 . 4% of the population .
In the last two decades , studies performed in Brazil [5] , [6] , [9] , [10] , [12] , [13] , [17]–[19] , [32] , and in other countries [20]–[27] , have described the distribution of infection , reasons and frequency of contact with natural waters , as well as other parameters related to schistosomiasis mansoni . In the present study , we studied a population from about 1 , 400 individuals that participated in a survey carried out in 1981 , comparing the results obtained with the current ones . Nevertheless , only 358 patients were re-examined , the other ones could not be observed , since some of them moved from the town or refused to participate . Comercinho/MG , Brazil , was considered as a hyperendemic area in 1981 ( 70 . 4% ) , but turned into a low endemic area ( 1 . 7% ) in 2005 . Administration of various treatments and quality of intervention measures produced an appreciable decrease in the prevalence of the disease ( 97 . 6% ) . The geometric average of the number of eggs per gram of feces obtained was 172 epg , lower than that reported by Costa in 1983 [12] ( 334 epg ) , or a little higher than the average found by Cury in 1991 [13] ( 105 epg ) , both of them also in Comercinho . It is note worthy that this actual average was obtained taking into account only 6 cases related to S . mansoni eggs discharged in the feces . Various studies in different regions demonstrated that the intensity of infection varies very much , and that reinfection after treatments in endemic areas show a lower number of eggs in the feces , when compared to the number detected pre-treatment [28] , [29] . The rate of splenomegaly was 1 . 3% in adults in 2005 , lower than that detected at the beginning of the project ( 6 . 8% ) . This fact may be connected with various treatments administered to the population along of 25 years . In fact , according to Kloetzel ( 1967 ) and Bina ( 1977 ) [28] , [30] , after specific treatment for schistosomiasis , even when reinfection occurs , it can be observed that the splenomegaly rate decreases significantly , and no new cases among the treated patients could be found . In Comercinho , no new case of hepatosplenomegaly could be detected after more than two decades of surveillance . The reasons and frequency for natural water contact frequently occurs in association with the socioeconomic standard of the population living in endemic areas , and depend on their needs and cultural habits . In Comercinho in 2005 , the main reasons for contact with natural waters pointed to professional activities , such as: watering vegetable-garden or farming , removing sand , crossing a brook , etc . ( 44 . 4% ) . The decrease of the related daily contact may be directly connected with the increase in the number of households with water supply . However , it was not possible to correlate directly the contact with natural waters to infection with S . mansoni , in the last survey , since only three positive patients mentioned contact with natural waters , with a biweekly frequency . Costa et al . ( 1987 ) [33] reported that the main risk factors responsible for splenomegaly in Comercinho were: absence of piped water , daily contact with natural waters and unskilled workers . Scott et al . ( 2003 ) [24] showed that many aspects , such as frequency , duration or time of contact , have influence on the infection rate . Certainly , the supply of safe water at town level diminished the incidence of schistosomiasis , since the existence of piped water in the housings reduced considerably the frequency and duration of contact with natural waters . In 231 patients clinically examined in 2005 , 95 . 2% presented intestinal clinical form of the disease , 3 . 5% showed hepatointestinal and 1 . 3% hepato-splenic forms , whereas at the beginning of the study , the percentages were 67 . 9% , 25 . 3% and 6 . 8% , respectively . The reversal of hepatomegaly and splenomegaly was deemed as significant . The importance of treatment and provision of sanitation for decrease of prevalence and morbidity control was previously emphasized . In Capitão Andrade , a small town in the State of Minas Gerais , Brazil , Conceição & Pereira ( 2002 ) [9] noticed that over a 21-year-period , from 1973 to 1994 , the prevalence decreased ( 60 . 8% in 1973; 32 . 2% in 1984 , and 27 . 3% in 1994 ) , whereas the evolution profile of the clinical forms was found to be satisfactory ( unaltered in 76 . 7% , clinical progression in 8 . 4% and regression in 14 . 9% ) . The reduction of both prevalence and severity of S . mansoni infection were ascribed to the treatment with oxamniquine administered in all infected individuals in 1984 , as well as to provision of piped water in the housings . In 2003 , those authors re-evaluated the area , and observed that the prevalence has also decreased ( 19 . 4% ) in relation to the preceding years , as well as the hepato-splenomegaly ( 5 . 8% in 1973 , 2 . 8% in 1984 , 2 . 3% in 1994 and 1 . 3% in 2003 ) . They observed that in spite of the significant reduction in the prevalence of infection without treatment at the initial phase ( 1973–1974 ) , followed by a specific treatment with oxamniquine in 1984–1994 , the rate of the severe forms and prevalence remained very high throughout the period 1994–2003 . During this time , people continued to receive treatment , but there were no improvements related to either basic sanitation or potable water supply , only sanitary education was strengthen . Thus , these facts led to the supposition that the high prevalence and severity of the clinical forms may have occurred due to reinfection [10] . In our laboratory , a study was devised to be carried out in Ravena , a district of Sabará , State of Minas Gerais , Brazil , in 1980 . Initially , the prevalence of schistosomiasis in Ravena was 36 . 7% , with an infection intensity of 229 epg ( geometric average related to positive individuals ) . No cases of hepato-splenic form could be detected . A specific treatment with oxamniquine in large scale was provided ( every four years , three treatments ) to patients discharging eggs in the feces . In 1992 , the local population was re-examined . When the study was initiated , 90% of the housings received safe water supply . The number of housings with an appropriate waste disposal also increased ( from 17% to 36% ) . In 1992 , the prevalence in the population decreased to 11 . 5% , and the average of eggs was 60 . 3 [34] , [35] . Recently , the same area was re-examined , i . e . , 27 years after the first clinical-epidemiological survey [11] . In this last survey , the prevalence was 2 . 5% , with an average of 21 epg . In the age group of 0–14 years old the positivity rate was 0 . 75 , whereas in 1980 this rate was 11 . 6% . Besides , 95% of the housings disposed of safe water supply , and more than 80% had appropriate waste disposal either by means of sewerage system , flush toilets or cess-pits . From 1990 onwards , the population was treated by a physician at the local health center , based on the results of stool examinations and by spontaneous plea . The living standard related to water contact in Ravena was modified throughout the years , since the majority of the population is no more in the habit of using natural contaminated water . In the last survey in Comercinho , 96% of the houses visited disposed of safe water supply by means of the public system , 97 . 6% had flush toilets or cess-pits for waste disposal , and 97 . 6% of the housings were classified as being of better quality ( Type A ) . The composition of a schistosomiasis control program varies according to two approaches: 1 . Control of morbidity , aiming at reducing the number of severe form of the disease; 2 . Control of transmission , by interrupting the evolutive cycle of the parasite . In the first case , the control of morbidity is specially undertaken by using chemotherapy , whereas the control of transmission requires treatment , safe water supply and appropriate waste disposal , environmental sanitation , and health education [2] , [31] . Currently , in Comercinho , low prevalence rate regarding the population in general and in previously treated individuals , low frequency of cases with hepatosplenic form , have clearly proved that control measures in association can led to interruption or significant decrease of transmission . At least , this clearly happens in Ravena and Comercinho . Finally , due to the effectiveness of the measures above mentioned , it is quite clear that the Brazilian Government should adopted the association of control measures mentioned in this study in order to attain schistosomiasis transmission control in the country . | A clinical-epidemiological reevaluation on schistosomiasis mansoni was performed in 2005 , in the urban area of a little town , Comercinho , MG , specifically focusing on the inhabitants of the same area in 1981 , when a first survey and treatment with oxamniquine were carried out . The surveys included: identification of the intermediary host , census , mapping of the city , socioeconomic survey , stool examination , clinical examination , research dealing with contact with natural waters , and treatment of the positive cases . From a population of 1 , 474 people studied in 1981 , 358 were submitted to stool examination , and 231 were clinically examined . From 1981 to 1992 five specific treatments were performed with oxamniquine and the last one with praziquantel . The results obtained were compared and demonstrated that the prevalence in Comercinho decreased significantly ( 70 . 4% to 1 . 7% ) , as well as the hepatosplenic form ( 7% to 1 . 3% ) in 1981 and 2005 , respectively . Significant improvement in the life quality ( improvement in the housing , professional qualification and basic sanitation ) were observed and must be considered important for the schistosomiasis control . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"public",
"health",
"and",
"epidemiology/epidemiology",
"infectious",
"diseases/helminth",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases"
]
| 2011 | Evaluation of a 25-Year-Program for the Control of Schistosomiasis Mansoni in an Endemic Area in Brazil |
Sequencing of whole tumor genomes holds the promise of revealing functional somatic regulatory mutations , such as those described in the TERT promoter . Recurrent promoter mutations have been identified in many additional genes and appear to be particularly common in melanoma , but convincing functional data such as influence on gene expression has been more elusive . Here , we show that frequently recurring promoter mutations in melanoma occur almost exclusively at cytosines flanked by a distinct sequence signature , TTCCG , with TERT as a notable exception . In active , but not inactive , promoters , mutation frequencies for cytosines at the 5’ end of this ETS-like motif were considerably higher than expected based on a UV trinucleotide mutational signature . Additional analyses solidify this pattern as an extended context-specific mutational signature that mediates an exceptional position-specific vulnerability to UV mutagenesis , arguing against positive selection . We further use ultra-sensitive amplicon sequencing to demonstrate that cell cultures exposed to UV light quickly develop subclonal mutations specifically in affected positions . Our findings have implications for the interpretation of somatic mutations in regulatory regions , and underscore the importance of genomic context and extended sequence patterns to accurately describe mutational signatures in cancer .
A major challenge in cancer genomics is the separation of functional somatic driver mutations from non-functional passengers . This problem is relevant not only in coding regions , but also in the context of non-coding regulatory regions such as promoters , where putative driver mutations are now mappable with relative ease using whole genome sequencing[1 , 2] . One important indicator of driver function is recurrence across independent tumors , which can be suggestive of positive selection . However , proper interpretation of recurrent mutations requires a detailed understanding of how somatic mutations occur in the absence of selection pressures . Somatic mutations are not uniformly distributed across tumor genomes , and regional variations in mutation rates have been associated with differences in transcriptional activity , replication timing as well as chromatin accessibility and modification[3–5] . Impaired nucleotide excision repair ( NER ) has been shown to contribute to increased local mutation density in promoter regions and protein binding sites[6 , 7] . Additionally , analyses of mutational processes and their sequence signatures have shown the importance of the immediate sequence context for local mutation rates[8] . Still , our understanding of mutational heterogeneity is incomplete , and it is not clear to what extent such effects can explain recurrent somatic mutations in promoter regions , which are suggested by some studies to be particularly frequent in melanoma despite several other cancer types approaching melanoma in terms of total mutation load[9 , 10] .
To characterize somatic promoter mutations in melanoma , we analyzed the sequence context of recurrently mutated individual genomic positions occurring within +/- 500 bp of annotated transcription start sites ( TSSs ) , based on 38 melanomas subjected to whole genome sequencing by the Cancer Genome Atlas[10 , 11] . Strikingly , of 17 highly recurrent promoter mutations ( recurring in at least 5/38 of tumors , 13% ) , 14 conformed to an identical 6 bp sequence signature ( Fig 1a and 1b ) . Importantly , the only exceptions were the previously described TERT promoter mutations at chr5:1 , 295 , 228 , 1 , 295 , 242 and 1 , 295 , 250[12 , 13] ( Fig 1c ) . The recurrent mutations occurred at cytosines positioned at the 5’ end or one base upstream of the motif CTTCCG ( Fig 1d ) , and were normally C>T or CC>TT transitions ( Fig 1a ) . Similar to most mutations in melanoma they thus occurred in a dipyrimidine context and were compatible with UV-induced damage through cyclobutane pyrimidine dimer ( CPD ) or 6–4 photoproduct formation[8 , 14] . Out of 15 additional positions recurrently mutated in 4/38 tumors ( 11% ) , 13 conformed to the same pattern , while the remaining two showed related sequence contexts ( Fig 1a ) . Many less recurrent sites also showed the same pattern ( S1 Table ) . The signature described here matches the consensus binding sequence of ETS family transcription factors ( TFs ) [15] , and the results are consistent with recent reports showing that ETS promoter sites are often recurrently mutated in melanoma[9] and that such mutations preferably occur at cytosines upstream of the core TTCC sequence[16] . Thus , while recurrent promoter mutations are common in melanoma , they consistently adhere to a distinct sequence signature , which may argue against positive selection as a major causative factor . The recurrently mutated positions were next investigated in additional cancer cohorts , first by confirming them in an independent melanoma dataset[17] ( S2 Table ) . We found that the identified hotspot positions were often mutated also in cutaneous squamous cell carcinoma ( cSCC ) [18] ( S3 Table ) as well as in sun-exposed skin[18 , 19] , albeit at lower variant frequencies ( S1 Fig , S4 Table ) . Additionally , one of the mutations , upstream of DPH3 , was recently described as highly recurrent in basal cell skin carcinoma[20] . However , we did not detect mutations in these positions in 13 non-UV-exposed cancer types ( S5 Table ) . The hotspots are thus present in UV-exposed samples of diverse cellular origins , but in contrast to the TERT promoter mutations they are completely absent in non-UV-exposed cancers . This further supports that recurrent mutations at the 5’ end of CTTCCG elements are due to elevated susceptibility to UV-induced mutagenesis in these positions . Next , we considered additional properties that could support or argue against a functional role for the recurrent mutations . We first noted a general lack of known cancer-related genes among the affected promoters , with TERT as one of few exceptions ( Fig 1a and S1 Table , indicated in blue ) . Secondly , the recurrent promoter mutations were not associated with differential expression of the nearby genes ( Fig 1a and S1 Table ) . This is in agreement with earlier investigations of some of these mutations , which gave no conclusive evidence regarding influence on gene expression[9 , 16 , 20] , although it should be noted that significant association was lacking also for TERT in this relatively small cohort . Lastly , we found that when comparing different tumors there was a strong positive correlation between the total number of the established hotspot positions that were mutated and the genome-wide mutation load , both in melanoma ( Fig 2a; Spearman’s r = 0 . 88 , P = 2 . 8e-13 ) and in cSCC ( S3 Table; r = 0 . 78 , P = 0 . 026 ) . This is again compatible with a passive model involving elevated mutation probability in the affected positions . Importantly , this contrasted sharply with most of the major driver mutations in melanoma , which were detected also in tumors with lower mutation load ( Fig 2b , S3 Table ) . These different findings further reinforce the CTTCCG motif as a strong mutational signature in melanoma . We next investigated whether the observed signature would be relevant also outside of promoter regions . As expected , numerous mutations occurred in CTTCCG sequences across the genome , but notably we found that recurrent mutations involving this motif were always located close to actively transcribed TSSs ( Fig 3a , 3b and 3c ) . We further compared the frequencies of mutations occurring at cytosines in the context of the motif to all possible trinucleotide contexts , an established way of describing mutational signatures in cancer[8] . As expected , on a genome-wide scale , the mutation probability for cytosines in CTTCCG-related contexts was only marginally higher compared to corresponding trinucleotide contexts ( Fig 4a ) . However , close to TSSs , the signature conferred a striking elevation in mutation probability compared to related trinucleotides , in particular for cytosines at the 5’ end of the motif and most notably near highly expressed genes ( Fig 4b–4d ) . Recurrent promoter mutations in melanoma thus conform to a distinct sequence signature manifested only in the context of active promoters , suggesting that a specific binding partner is required for the element to confer elevated mutation probability . CTTCCG elements have in various individual promoters been shown to be bound by ETS factors such as ETS1 , GABPA and ELF1[21] , ELK4[22] , and E4TF1[23] . This suggests that the recurrently mutated CTTCCG elements could be substrates for ETS TFs . As expected , matches to CTTCCG in the JASPAR database of TF binding motifs were mainly ETS-related ( S6 Table ) . Notably , recurrently mutated CTTCCG sites were evolutionarily conserved to a larger degree than non-recurrently mutated but otherwise similar control sites , further supporting that they constitute functional ETS binding sites ( S2 Fig ) . This was corroborated by analysis of top recurrent CTTCCG sites in relation to ENCODE ChIP-seq data for 161 TFs , which showed that the strongest and most consistent signals were for ETS factors ( GABPA and ELF1 ) ( S3 Fig ) . The distribution of mutations across tumor genomes is shaped both by mutagenic and DNA repair processes . Binding of TFs to DNA can increase local mutation rates by impairing NER , and strong increases have been observed in predicted binding sites for several ETS factors[6 , 7] . It is also established that contacts between DNA and proteins can modulate DNA damage patterns by altering conditions for UV photoproduct formation[24–27] . In upstream regions of XPC -/- cSCC tumors lacking global NER , we found that several of the established hotspot sites were mutated ( S7 Table ) and that the CTTCCG signature still conferred elevated mutation probabilities compared to relevant trinucleotide contexts ( Fig 5 ) , although to a lesser extent than in melanomas with functional NER ( Fig 4 ) . Transcription-coupled NER ( TC-NER ) may still be active in XPC -/- tumors , and the signature could thus theoretically arise due to blocking of TC-NER at CTTCCG elements . However , only upstream regions , which should not be subjected to this process , were considered in this analysis . Additionally , TC-NER is strand-specific[14] , but the signature was present independently of strand orientation relative to the downstream gene in XPC -/- tumors ( Fig 5a and 5b ) . The signature described here is thus unlikely explained by impaired NER alone , and other mechanisms , such as inhibition of other repair-related processes or favorable conditions for UV lesion formation at the 5’ end of ETS-bound CTTCCG elements , may contribute . Finally , we sought to experimentally test our proposed model that the observed promoter hotspots are due to localized vulnerability to mutagenesis by UV light . We subjected human melanoma cells and keratinocytes to daily UV doses for a period of 5 or 10 weeks and used an ultrasensitive error-correcting amplicon sequencing protocol , SiMSen-Seq[28] , to assay two of the observed promoter hotpots for mutations: RPL13A , the most frequently mutated site in the tumor data , and DPH3[10 , 20] ( Fig 6a ) . Between 36k and 82k error-corrected reads ( >20x oversampling ) were obtained for each of 16 different conditions ( Fig 6b and 6c ) . Strikingly , subclonal mutations appeared specifically in expected positions at both time points and in both cell lines at a frequency reaching up to 2 . 9% of fragments ( RPL13A , 10 weeks of exposure ) , while being absent in non-exposed control cells ( Fig 6d and 6e ) . As predicted by the tumor data , mutations occurred primarily at cytosines upstream of the TTCCG motif , with lower-frequency mutations occurring also in the central cytosines . Few mutations were observed outside of the TTCCG context despite presence of many cytosines in theoretically vulnerable configurations in the two amplicons ( Fig 6d and 6e , underscored ) . Interestingly , an atypical substitution pattern displayed by the DPH3 hotspot in the tumors , involving C>A and C>G in addition to the expected C>T transitions ( Fig 1a ) , was mirrored also in the UV exposure data ( Fig 6d ) . Our results from UV exposure of cultured cells further reinforce that recurrent mutation hotspots in promoters in melanoma arise due to an exceptional vulnerability to UV mutagenesis in these positions . In summary , we demonstrate that recurrent promoter mutations are common in melanoma , but also that they adhere to a distinct sequence signature in a strikingly consistent manner , arguing against positive selection as a major driving force . This model is supported by several additional observations , including lack of cancer-relevant genes , lack of obvious effects on gene expression , presence of the signature exclusively in UV-exposed samples of diverse cellular origins , and strong positive correlation between genome-wide mutation load and mutations in the affected positions . Crucially , exposing cells to UV light under controlled conditions efficiently induces mutations specifically in affected sites . These results point to limitations in conventional genome-wide derived trinucleotide models of mutational signatures , and imply that extended sequence patterns as well as genomic context should be taken into account to improve interpretation of somatic mutations in regulatory DNA .
Whole-genome sequencing data for 38 skin cutaneous melanoma ( SKCM ) metastases were obtained from the Cancer Genome Atlas ( TCGA ) together with matching RNA-seq data ( dbGap accession phs000178 . v9 . p8 ) . Mutations were called using SAMtools[29] ( command mpileup with default settings and additional options -q1 and–B ) and VarScan[30] ( command somatic using the default minimum variant frequency of 0 . 20 , minimum normal coverage of 8 reads , minimum tumor coverage of 6 reads and the additional option –strand-filter 1 ) . Mutations where the variant base was detected in the matching normal were not considered for analysis . Mutations overlapping germline variants included in the NCBI dbSNP database , Build 146 , were removed . The genomic annotation used was GENCODE[31] release 17 , mapped to GRCh37 . The TSS of a gene was defined as the 5’most annotated transcription start . Somatic mutation status for known driver genes was obtained from the cBioPortal[32 , 33] . RNA-seq data was analyzed with respect to the GENCODE[31] ( v17 ) annotation using HTSeq-count ( http://www-huber . embl . de/users/anders/HTSeq ) as previously described[34] . Differential gene expression between tumors with and without mutations in promoter regions was evaluated using the two-sided Wilcoxon rank sum test . The SKCM tumors were analyzed across the whole genome or in regions close to TSS , in which case only mutations less than 500 bp upstream or downstream of TSS were included . For the analysis of regions close to TSS the genes were divided in three tiers of equal size based on the mean gene expression level across the 38 SKCM tumors . The February 2009 assembly of the human genome ( hg19/GRCh37 ) was downloaded from the UCSC Genome Bioinformatics site . Sequence motif and trinucleotide frequencies were obtained using the tool fuzznuc included in the software suite EMBOSS[35] . The mutation probability was calculated as the total number of observed mutations in a given sequence context across all tumors divided by the number of instances of this sequence and by the number of tumors . The evolutionary conservation of genome regions was evaluated using phastCons scores[36] from multiple alignments of 100 vertebrate species retrieved from the UCSC genome browser . The analyzed regions were 30 bases upstream and downstream of the motif CTTCCG located less than 500 bp from TSS . Binding of transcription factors at NCTTCCGN sites was evaluated using normalized scores for ChIP-seq peaks from 161 transcription factors in 91 cell types ( ENCODE track wgEncodeRegTfbsClusteredV3 ) obtained from the UCSC genome browser . Whole genome sequencing data from sun-exposed skin , eye-lid epidermis , was obtained from Martincorena et al . , 2015[19] . SAMtools[29] ( command mpileup with a minimum mapping quality of 60 , a minimum base quality of 30 and additional option –B ) was used to process the data and VarScan[30] ( command mpileup2snp counting all variants present in at least one read , with minimum coverage of one read and the additional strand filter option disabled ) was used for mutation calling . Whole genome sequencing data from 8 cSCC tumors and matching peritumoral skin samples was obtained from Durinck et al . , 2011[37] . Whole genome sequencing data from cSCC tumors and matching peritumoral skin from 5 patients with germline DNA repair deficiency due to homozygous frameshift mutations ( C940del-1 ) in the XPC gene was obtained from Zheng et al . , 2014[18] . SAMtools[29] ( command mpileup with a minimum mapping quality of 30 , a minimum base quality of 30 and additional option –B ) was used to process the data and VarScan[30] ( command mpileup2snp counting all variants present in at least one read , with minimum coverage of two reads and the additional strand filter option disabled ) was used for mutation calling . For the mutation probability analysis of cSCC tumors with NER deficiency , an additional filter was applied to only consider mutations with a total coverage of at least 10 reads and a variant frequency of at least 0 . 2 . The functional impact of mutations in driver genes was evaluated using PROVEAN[38] and SIFT[39] . Non-synonymous mutations that were considered deleterious by PROVEAN or damaging by SIFT were counted as driver mutations . A375 melanoma cells were a gift from Joydeep Bradbury and HaCaT keratinocyte cells were a gift from Marica Ericson . Cells were grown in DMEM + 10% FCS + gentamycin ( A375 ) or pen/strep ( HaCaT ) ( Thermo Scientific ) . Cells were treated in DMEM in 10 cm plates without lids with 36 J/m2 UVC 254 nm ( equivalent to 6 hour daily dose at 0 . 1J/m2/min[40] , CL-1000 UV crosslinker , UVP ) , 5 days a week for 10 weeks . Cells were split when confluent and reseeded at 1:5 . Cells were frozen at -20°C . DNA was extracted based on Tornaletti and Pfeifer [41] . Briefly , cell pellets were lysed in 0 . 5 ml of 20 mM Tris-HCl ( pH 8 . 0 ) , 20 mM NaCl , 20mM EDTA , 1% ( w/v ) sodium dodecyl sulfate , 600 mg/ml of proteinase K , and 0 . 5 ml of 150 mM NaCl , 10 mM EDTA . The solution was incubated for two hours at 37°C . DNA was extracted twice with phenol-chloroform and once with chloroform and precipitated by adding 0 . 1 vol . 3 M sodium acetate ( pH 5 . 2 ) , and 2 . 5 volumes of ethanol . The pellets were washed with 75% ethanol and briefly air-dried . DNA was dissolved in 10 mM Tris-HCl ( pH 7 . 6 ) , 1 mM EDTA ( TE buffer ) ( all from Sigma Aldrich ) . DNA was treated with RNAse for 1 hr at 37°C and phenol-chloroform extracted and ethanol precipitated before dissolving in TE buffer . To detect and quantify mutations we applied SiMSen-Seq ( Simple , Multiplexed , PCR-based barcoding of DNA for Sensitive mutation detection using Sequencing ) as described[28] . Briefly , barcoding of 150 ng DNA was performed in 10 μL using 1x Phusion HF Buffer , 0 . 1U Phusion II High-Fidelity polymerase , 200 μM dNTPs ( all Thermo Fisher Scientific ) , 40 nM of each primer ( PAGE-purified , Integrated DNA Technologies ) and 0 . 5M L-Carnitine inner salt ( Sigma Aldrich ) . Barcode primer sequences are shown in S8 Table . The temperature profile was 98°C for 3 min followed by three cycles of amplification ( 98°C for 10 sec , 62°C for 6 min and 72°C for 30 sec ) , 65°C for 15 min and 95°C for 15 min . The reaction was terminated by adding 20 μL TE buffer , pH 8 . 0 ( Invitrogen , Thermo Fisher Scientific ) containing 30 ng/μL protease from Streptomyces griseus ( Sigma Aldrich ) at the beginning of the 65°C incubation step . Next , 10 μL of the diluted barcoded PCR products were amplified in a 40 μL using 1x Q5 Hot Start High-Fidelity Master Mix ( New England BioLabs ) and 400 nM of each sequencing adapter primer . Adapter primers are shown in S8 Table . The temperature profile was 95°C for 3 min followed by 40 cycles of amplification ( 98°C for 10 sec , 80°C for 1 sec , 72°C for 30 sec and 76°C for 30 sec , with a ramp rate of 0 . 2°C/sec ) . The 40 μL PCR products were then purified using Agencourt AMPure XP beads ( Beckman-Coulter ) according to the manufacturers’ instructions using a bead to sample ratio of 1 . The purified product was eluted in 20 μL TE buffer , pH 8 . 0 . Library concentration and quality was assessed using a Fragment Analyzer ( Advanced Analytical ) . Final libraries were pooled to equal molarity in Buffer EB ( 10 mM Tris-HCl , pH 8 . 5 , Qiagen ) containing 0 . 1% TWEEN 20 ( Sigma Aldrich ) . Sequencing was performed on an Illumina NextSeq 500 instrument at TATAA Biocenter ( Gothenburg , Sweden ) using 150 bp single-end reads . Raw FastQ files were subsequently processed as described[28] using Debarcer Version 0 . 3 . 0 ( https://github . com/oicr-gsi/debarcer ) . Sequence reads with the same barcode were grouped into families for each amplicon . Barcode families with at least 20 reads , where ≥ 90% of the reads were identical , were required to compute consensus reads . FastQ files were deposited in the Sequence Read Archive under BioProject ID PRJNA375726 . | Cancer is caused by somatic mutations that alter cell behavior . While such mutations typically occur in protein-coding genes , recent studies describe individual positions in gene regulatory regions ( promoters ) that are recurrently mutated in many independent tumors . This suggests that positive selection could be acting on these non-coding mutations , and that they may contribute to carcinogenesis . However , proper interpretation of recurrent mutations requires a detailed understanding of how such mutations arise in the absence of selection pressures , referred to as mutational heterogeneity . In this paper , we describe a distinct sequence signature that characterizes nearly all highly recurrent promoter mutations in melanoma . Additional analyses support that this sequence mediates an exceptional local vulnerability to UV-induced mutagenesis , explaining why mutations are frequently observed in these positions . Importantly , cultured cells exposed to UV light quickly developed mutations specifically in the expected sites . Our results have important implications for the interpretation of recurrent somatic mutation patterns in non-coding DNA . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"&",
"methods"
]
| [
"cancer",
"genomics",
"medicine",
"and",
"health",
"sciences",
"computational",
"biology",
"cancers",
"and",
"neoplasms",
"basic",
"cancer",
"research",
"oncology",
"mutation",
"genome",
"analysis",
"sequence",
"motif",
"analysis",
"mutation",
"databases",
"frameshift",
"mutation",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"bioinformatics",
"melanomas",
"biological",
"databases",
"somatic",
"mutation",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"genomic",
"databases",
"genomic",
"medicine"
]
| 2017 | Recurrent promoter mutations in melanoma are defined by an extended context-specific mutational signature |
The contemporary proteinogenic repertoire contains 20 amino acids with diverse functional groups and side chain geometries . Primordial proteins , in contrast , were presumably constructed from a subset of these building blocks . Subsequent expansion of the proteinogenic alphabet would have enhanced their capabilities , fostering the metabolic prowess and organismal fitness of early living systems . While the addition of amino acids bearing innovative functional groups directly enhances the chemical repertoire of proteomes , the inclusion of chemically redundant monomers is difficult to rationalize . Here , we studied how a simplified chorismate mutase evolves upon expanding its amino acid alphabet from nine to potentially 20 letters . Continuous evolution provided an enhanced enzyme variant that has only two point mutations , both of which extend the alphabet and jointly improve protein stability by >4 kcal/mol and catalytic activity tenfold . The same , seemingly innocuous substitutions ( Ile→Thr , Leu→Val ) occurred in several independent evolutionary trajectories . The increase in fitness they confer indicates that building blocks with very similar side chain structures are highly beneficial for fine-tuning protein structure and function .
Nature uses monodisperse , sequence-specific polymers to perform a plethora of functions essential to survival . While ancient life forms are believed to have harnessed RNA molecules to accelerate vital metabolic reactions [1]–[3] , proteins assumed most catalytic functions in the course of natural evolution . Proteins are generally superior to RNA as catalysts , principally due to their flexible backbone and chemically diverse side chains which enable construction of finely tuned active sites [4] , [5] . However , at the dawn of the protein world , it is unlikely that all of the current proteinogenic α-amino acids were employed . The genetic code more plausibly started from a subset of contemporary building blocks – extant in the primordial soup or produced by early life forms – and subsequently expanded to meet increasingly challenging metabolic needs [6] . Such a scenario has three corollaries: Here , we experimentally evaluate this conundrum by monitoring the evolution of a simplified enzyme — the CM constructed from a 9-amino acid alphabet [14] — while allowing it to freely acquire any of the remaining proteinogenic amino acids . The active site of this model primordial enzyme contains essentially the same functional groups as wild-type CM enzymes , but it exhibits lower stability and probably suffers from suboptimal alignment of these residues . This system is therefore ideally suited to investigate how expansion of the amino acid repertoire might enhance catalytic sophistication of enzymes beyond the contribution of novel reactivities , thus providing insight into the development of the genetic code .
MjCM is an all-helical , domain-swapped dimer that catalyzes the Claisen rearrangement of chorismate to prephenate , an essential transformation in the biosynthesis of aromatic amino acids [13] . The amino acid composition of a stable 93-residue variant [13] was previously simplified by two-stage combinatorial mutagenesis and in vivo selection . In the first step , the three helices in the enzyme were replaced with binary-patterned modules [16] , [17] of randomized sequence , whereby only hydrophobic side chains were allowed for residues facing the interior of the protein and surface residues were constrained to polar building blocks [18] . Active enzymes in the degenerate libraries were then identified by their ability to complement the CM deficiency of a genetically engineered bacterial host [19] . One of the resulting 14-letter CMs was further simplified in a second step by systematically replacing loop residues and an active site glutamine with amino acids from the restricted alphabet [14] . In vivo selection finally yielded a variant , referred to here as 9-CM ( Figure 1 ) , composed of only nine different amino acids , namely the four polar ( D , E , N , K ) and four non-polar ( F , I , L , M ) residues from binary patterning , plus Arg , which is essential for binding the anionic substrate . 9-CM mimics many of the properties expected for a primitive enzyme . Despite dramatic simplification , it adopts a specific tertiary and quaternary fold as judged by CD spectroscopy and size-exclusion chromatography and catalyzes a metabolically important chemical transformation . The active site has all the functional groups needed for catalysis , including three arginines to bind the substrate carboxylates and other polar residues that contribute hydrogen-bonding interactions in the transition state . Nevertheless , 9-CM displays molten globular characteristics [14] and is 11 kcal/mol less stable and >1000-fold less active than the wild-type MjCM dimer . Because it was impossible to sample all sequences accessible in a nine amino acid alphabet when 9-CM was evolved , and due to the combinatorial fashion in which it was generated , the enzyme presumably resides at a local fitness maximum . These features make 9-CM an ideal starting point to explore how evolutionary expansion of the amino acid alphabet — with concomitant introduction of novel side chains — impacts protein structure and function . Given the architecture of the genetic code , the evolvability of an enzyme is not solely based on its amino acid sequence . It also depends on codon usage , since different amino acid substitutions are accessible from different codons by point mutation [20] . Two different starting genes coding for 9-CM were therefore used for our evolution experiments: the gene obtained from the selection of 9-CM , enriched in NWV codons ( where N = A , G , C , T; W = A , T; V = A , G , C ) due to the mode of its construction [14] , [18] , and a synthetic version with a highly skewed nucleotide composition ( 89% A+T content ) . These are referred to as n9-cm and AT9-cm , respectively ( Table S1 ) . Besides offering alternative evolutionary trajectories , such distinctive starting points can potentially provide insight into the adaptation of genes . To monitor the evolution of 9-CM , we first introduced the two genes encoding the simplified enzyme into the genome of a CM-deficient E . coli strain [19] under control of the highly regulable tetracycline promoter ( Ptet ) [21] ( see Materials and Methods for details ) . Not surprisingly , protein production is strongly reduced when genes for 9-CM are expressed chromosomally or from A+T-rich constructs , as compared to plasmid-based templates or more balanced nucleotide compositions ( Figure S1 , lanes 1 , 4 , and 5 ) , causing a reduction in fitness . As a consequence , only the genomically reengineered strain harboring n9-cm grew under selective conditions in the presence of tetracycline , the activator of transcription . Chromosomal AT9-cm failed to complement the CM deficiency under all conditions tested . To evolve the weakly complementing n9-cm strain , we resorted to automated cultivation in the GM3 device , an apparatus that enhances the growth rate of bacterial cells in suspension and counterselects the formation of biofilms [22] . The turbidostat regime , which maintains the optical density of a microbial population at a fixed setpoint by diluting the culture with pulses of fresh nutrient medium , is well suited for improving enzyme activity; cumulative fixation of adaptive mutations in the gene ( s ) encoding the enzyme ( s ) catalyzing the limiting metabolic reaction ( s ) is expected to occur over successive generations of the cultivated cell population . In addition to this growth control by infusion of fresh nutrient medium , transfer of the bacterial population is conducted cyclically between two cultivation chambers which alternatively undergo a transient period of sterilization with concentrated sodium hydroxide . In this way , any mutant cell that could adapt by attaching to any inner surface of the device and thus escape selection for faster growth in suspension is actively destroyed during every purging cycle of the device . Figure 2 delineates , schematically , the history of the CM evolutionary process . In the early phase of the experiment , the growth medium was supplemented with high concentrations of tetracycline and phenylalanine but lacked tyrosine . These conditions permit growth of strains harboring weakly active CM variants . For later generations , L-Phe was omitted to increase selection stringency [19] , while tetracycline was included at all stages of the experiment , as its omission would favor escape mutations in regulatory elements ( i . e . the tet-regulon ) over CM adaptation . At arbitrary intervals , the CM genes of several clones from the evolving population were sequenced . In the course of optimization , 9-CM first acquired the I30T mutation , yielding 10-CM , which subsequently acquired the L61V mutation to give 11-CM ( Figure 1A , 1B , amino acid numbering according to MjCM ) . Each mutation occurred via a single base substitution at the DNA level , and extended the amino acid alphabet of the mutase by a single letter . The fact that these mutants arose spontaneously and outcompeted their respective parent suggests that both substitutions have beneficial effects on enzyme stability , in vivo concentration , activity , or some combination of these factors . To enable analogous selection experiments with the AT-rich gene , AT9-cm was subcloned into a high-copy plasmid under the control of the strong trc promoter ( pKECMT [23] ) . Increasing gene dosage was expected to compensate for possible gene expression problems , such as low rates of transcription/translation or low transcript stability . This construct indeed complemented the CM deficiency , and , after continuous evolution for 45 days in the turbidostat , faster growing variants were obtained . Isolation and sequencing of the responsible AT9-cm gene revealed a single AUU to ACU mutation , resulting in the same I30T substitution as observed for n9-cm . In addition , the helper plasmid pKIMP-UAUC had recombined with the plasmid coding for the CM variant at homologous promoter regions , as indicated by restriction digestions and DNA sequencing ( Figure S7 ) . Thus , two drastically different CM genes evolved by targeting the same locus , underscoring the importance of the I30T substitution . The observation that mutations occurred only at two very specific sites within the proteins suggests that there are surprisingly few hot-spot positions where substitutions can improve CM structure and function . Furthermore , the mutations of Ile30 to Thr and Leu61 to Val are surprisingly unassuming considering the other alphabet-expanding mutations that were accessible . To test whether this finding is an artifact of the genetic code's architecture , random cassette mutagenesis was performed at positions 30 and 61 simultaneously . A gene library constructed by PCR using degenerate oligonucleotides and n9-cm as a template was ligated into a high-copy plasmid ( pKT [21] ) , and subjected to selection in CM-deficient cells . A comparison of the sequences of complementing clones under different selection stringencies with library members grown under non-selective conditions revealed pronounced residue preferences at the hotspot positions ( Figure 3 and Figure S2 , Tables S2 and S3 ) . Under low selection stringency ( 1000 ng/mL tetracycline , Figure S2B ) , a range of small residues including Ser , Val , and Thr was selected . Upon increasing the selection stringency ( 100 ng/mL tetracycline , Figure S2C ) , a dramatic enrichment of Thr at each hot-spot position and , to a lesser extent , Ser ( pos . 30 ) and Val ( pos . 61 ) occurred in functional mutases . In contrast to the mutations Ile30Thr and Leu61Val , which are accessible via single-base substitution and which accumulated in the turbidostat selections , Leu61Thr requires two base substitutions , reducing its likelihood of spontaneous occurrence . In summary , regardless of starting gene and mode of mutagenesis , the simplified CM evolves in a defined way by conservative expansion of its building block repertoire dominated by subtle changes of residue size and polarity . The evolutionary trajectory of n9-cm was scrutinized by characterizing 9-CM , 10-CM , and 11-CM in vivo and in vitro . To separate the effects of CM evolution from alternative mechanisms of strain adaptation to the selection pressure , we cloned the CM genes into high-copy plasmids ( pKT ) under the control of Ptet . Four arbitrarily chosen CM variants that were subcloned after selection from the random cassette mutagenesis library were included for comparison . E . coli strains harboring these constructs were evaluated by in vivo complementation . The selection stringency of the assays was adjusted by varying the tetracycline concentration ( Figure S3 ) . While 9-CM only grows at concentrations of tetracycline >1000 ng/mL , the evolved variants still thrive when transcription is more strongly repressed ( Figure S3A ) . Interestingly , none of the four CM variants selected from the random cassette mutagenesis library outperforms 11-CM in these complementation assays ( Figure S3B ) . However , a positive control carrying a wild-type CM domain from E . coli ( EcCM ) [24] grows even in the absence of tetracycline . Western blot analysis of E . coli cells producing 9-CM and its evolved variants revealed that expansion of the amino acid alphabet leads to increased CM production ( Figure S1 , lanes 1–3 ) . Since the mutated codons found for 11-CM do not substantially differ from those of 9-CM with respect to codon usage , this result suggests that the evolved CMs fold into compact dimers more readily and are thus more resilient towards degradation in vivo . Biophysical characterization of the original clone ( 9-CM ) and the evolved variants ( 10-CM and 11-CM ) yielded insight into the benefit provided by the two mutations . The proteins , equipped with a C-terminal His6 tag , were produced in a CM-deficient production strain [25] and purified by Ni-NTA affinity and size-exclusion chromatography . All variants eluted as dimers from the size-exclusion column , with a small fraction of aggregates in the more readily produced mutant proteins ( Figure S4A ) . Circular dichroism ( CD ) spectroscopy confirmed that all variants are highly helical ( Figure S4B ) . CD was also used to assess stability ( Figure S4C ) . While the parent 9-CM unfolds non-cooperatively upon heating , the evolved 10-CM and 11-CM show modest cooperativity , with midpoints of transition at 60 and 65°C , respectively . The free energies of unfolding ΔGU0 ( H2O ) , estimated from titrations with GdmCl , show the same trend ( Figure 4A , Table 1 ) . While the double mutant 11-CM is stabilized compared to the simpler variant 9-CM by more than 4 kcal/mol , it is still far less stable than natural CMs [13] . Limited proteolysis with trypsin was used to assess the compactness of the CM structures ( Figure S5 ) . 9-CM is cleaved readily over a timescale of minutes , whereas the evolved mutases show greater resilience towards proteolysis . For 11-CM , an intermediate corresponding to CM ( 1–92 ) , i . e . a loss of the C-terminal residue plus the His6-tag ( Figure S5B ) , is not degraded further by trypsin over the course of approximately 1 hour . These results parallel the equilibrium unfolding experiments , and highlight the fact that the I30T and the L61V mutations incrementally increase protein compactness . Expansion of the 9-amino acid alphabet positively affected catalytic properties as well ( Figure 4B , Table 1 ) . The steady-state parameters show that 10-CM and 11-CM are roughly 3-fold and 10-fold more active than 9-CM , respectively , mostly due to an increase in kcat . Nevertheless , wild-type mutases are again more sophisticated; their activities are several orders of magnitude higher . Although the two mutations improve both stability and activity of the protein , additional refinement will clearly be required to reach wild-type fitness levels . To rationalize the improvement in stability and activity achieved during the evolutionary trajectory , dimeric models of each variant were constructed and compared to EcCM ( Figure 1C ) . A prototype 9-CM structure was created by homology modeling using I-TASSER [26] with the EcCM structure [27] as a reference . The structures for 10-CM and 11-CM were derived from that of 9-CM by in silico mutagenesis . One transition state analog ( TSA ) was placed into each active site . The structures were first energy-minimized and then subjected to a 10 ns molecular dynamics ( MD ) simulation at 293 K . Despite having only ca . 25% sequence identity , the energy-minimized homology-modeled structures of the simplified CMs are all essentially identical to that of EcCM ( Figure 1C ) . The structural stability , assessed in terms of secondary structure content and atom-positional root-mean-square deviation ( rmsd ) from the energy-minimized starting structure , is similar for EcCM and all of the evolved variants within the time frame of the simulations ( Figure S6 ) . The secondary structure is particularly well maintained , and comprises predominantly α-helical structure , in keeping with the results of the CD spectroscopy . Overall , these results show that the assumption of a similar structure for the reduced alphabet variants as for the wild-type EcCM is reasonable . While the global structure seems largely unperturbed in all of the simplified CMs , the MD simulations revealed how local interactions that are perturbed in the simplified CMs were restored during directed evolution . The environments of the residues that changed during the evolutionary trajectory were probed by calculating the fraction of the simulation frames in which at least one atom from a given residue falls within 0 . 6 nm of the Cα atom of residue 30 or 61 ( Figure 5A ) . The primary interaction partners for residue 30 are residues 82–89 of the same subunit , and residue 12 of the alternate subunit ( Figure 5B ) . The latter interaction occurs in nearly every structure saved during all simulations , whereas the interactions with residues 82–89 persist only for EcCM . Residue 61 of the wild-type EcCM interacts with residues 57 ( helix 2 ) , 65–68 ( the loop connecting helices 2 and 3 ) and 71 ( helix 3 ) of the same subunit , as well as Leu20 situated in helix 1 of the alternate subunit ( Figure 1C and Figure 5C ) . In 9-CM , the local environments of residues 30 and 61 are partially rearranged . The probabilities of contact of Ile30 with Asn82 and Glu86 are diminished , in part at the expense of aberrant proximity to residues 89 and 90 , or almost completely absent . Such distortions of the active site region may explain the reduced activity of the simplified mutases . In addition , the conformation of the loop between helices 2 and 3 appears to differ between EcCM and 9-CM , with helices 2 and 1′ being pushed apart in the latter ( Figure 1C ) . Upon mutation of Ile30 to Thr , helix 3 is repositioned , evident in the movement of residues 82 and 85 with respect to residue 30 and the loss of the interaction between Leu61 and Phe71 in 10-CM ( Figure 5A ) . Although the introduction of a threonine potentially creates new possibilities for hydrogen bonding , the overall size of this residue is likely to have been more decisive given that natural AroQ class CMs typically have a Gly , Ala , or Val residue at this position [13] . Moreover , hydrogen-bonding interactions between Thr30 and the TSA were detected to a significant extent only in one subunit of 10-CM , but not in the other subunit of 10-CM or either subunit of 11-CM . The subsequent shortening of the side-chain of Leu61 to Val in 11-CM shifts helix 2 , promoting contacts of residue 61 with helices 3 ( Phe71 ) and 1′ ( Leu20′ ) to better resemble the helix packing in EcCM . Interestingly , subunit symmetry appears to be correlated to CM activity , with the two subunits of more proficient CM variants being more similar than the subunits of their less evolved counterparts . In summary , the mutations I30T and L61V iteratively restore some of the structural features of highly active CMs , exemplified here by EcCM , lost upon simplification to 9-CM , bringing with them improvements in both the stability and the activity of the simplified CM .
In 1939 , Ernest Vincent Wright wrote the novel “Gadsby” , a story of 50 , 000 words , without a single use of the letter “e” . In this so-called lipogram , Wright was able to tell a complete tale fluently , an achievement previously claimed to be impossible because of the abundance of “e”s in the English language . Still , without the past tense , definite articles , many numbers and pronouns , the text lacks some of the finesse that can be achieved with the entire alphabet . The same principle holds true for biopolymers . Most simply , increasing the number of different building blocks in lattice models not only promotes uniqueness of folded structures , but also ameliorates folding kinetics [28] , [29] . Regardless , functional polymers can be constructed with extremely small monomer diversity . For example , Joyce and co-workers evolved a ribozyme that lacks cytidine [30] , and later extended their simplification to a two-nucleotide alphabet , the simplest possible [31] . To investigate the effect of added complexity , they evolved the cytidine-free ligase , allowing incorporation of the missing nucleotide [32] . With twelve changes ( seven of which were mutations to C ) , the catalytic activity was improved 20-fold , predominantly by stabilizing the active conformation and fine-tuning the structure . Here , we report how simplified proteins evolve when we let them access new building blocks . Upon prolonged in vivo selection for improved CM activity , a simplified CM constructed from a nine amino acid alphabet [14] acquired two mutations , I30T and L61V . These changes extend the alphabet to 11 letters , and improve the enzyme's stability by ca . 4 kcal/mol and its activity by a factor of 10 . According to molecular dynamics simulations , the overall protein structure and the architecture of the active site are similar in all CM variants , but helix packing is distorted in the simplified enzyme 9-CM . The I30T and L61V mutations restore proper helix packing , which simultaneously increases protein stability and fine-tunes active site interactions and thus improves catalytic activity . The evolutionary pathway for 9-CM is similar in three separate experiments starting from two structurally very different but synonymous genes , and two complementary strategies to introduce diversity , namely in vivo mutagenesis and random cassette mutagenesis . A transition resulting in the I30T mutation ( ATC to ACC in n9-cm , ATT to ACT in AT9-cm ) was rapidly selected in all evolution experiments conducted under competitive conditions . Subsequently , the mutation L61V was fixed in the long-term evolution of n9-CM . This outcome is paralleled when both positions are simultaneously varied using random cassette mutagenesis , whereby an enrichment of Thr and Ser at position 30 , and Thr and Val at position 61 was observed in more active CMs . The rapid fixation of I30T from two distinct genetic templates indicates that the evolutionary pathway of the simplified enzyme is governed by a specific hotspot ( residue 30 ) proximal to the active site of the protein . In addition , the mutations that arose are strikingly modest given that more drastic changes in physico-chemical properties were accessible at the hotspot positions , as well as throughout the entire protein . Random cassette mutagenesis confirmed , however , that these unspectacular mutations represented strongly competitive changes at the mutagenesis hotspots even in the absence of constraints due to the genetic code's architecture . While the amino acid set of 9-CM differs from likely primordial alphabets [15] , the conclusions drawn from this study are nevertheless relevant to the evolution of primitive functional proteins . The fact that all the mutations found in the CM selections extend the amino acid alphabet corroborates that 9-CM is situated at a local fitness maximum , thereby meeting the expectations for a primordial protein adapted to its producer's genetic code . Nevertheless , the mutations found in 11-CM do not contribute novel functionality to the active site . Instead , relatively modest changes in the building blocks , in this case shortening of the side chain by one methylene group ( Leu to Val , Ile to Thr ) and increased polarity as well as the potential to engage in H-bonding ( Ile to Thr ) , fine-tune the enzyme's structure , and consequently its activity . Similar improvements presumably occurred simultaneously for many different primordial enzymes , greatly enhancing the fitness of early life forms . Evolutionary experiments with proteins and organisms containing alternative or even expanded amino acid repertoires [33]–[36] suggest that the current genetic code is not an evolutionary dead end , but amenable to further natural and synthetic innovations . Our results show that incremental addition of even seemingly redundant building blocks can profoundly affect protein structure and activity , significantly enlarging the functional space of these biomolecules .
In vivo complementation assays for CM activity in KA12/pKIMP-UAUC [19] , [21] as well as protein production from the CM-deficient E . coli strain KA13 and in vitro characterization were carried out as previously described [14] . Growth of strains harboring CM variants was assessed from single colony streak-outs using a phenotypic growth scale according to colony size ( Table S5 ) . Genes encoding 9-CM under control of Ptet were introduced into the genome of the CM-deficient E . coli strain KA12 at the kdgK locus ( Figure S8 ) in a RecA-dependent manner according to published procedures [37] . E . coli cells harboring the chromosomally encoded n9-cm and the helper plasmid pKIMP-UAUC were grown in rich medium to an OD600 of 0 . 5 , washed and resuspended in minimal medium and directly used to inoculate a 19 . 5-mL pulse-feed alternating turbidostat [22] . At regular intervals , a conditional pulse of fresh medium was delivered ( if OD600>0 . 3 ) , effecting a 10% dilution . Steady-state bacterial populations are thus expected to contain approximately 4–5×109 cells . Once every 24 h , the growing culture was transferred to a backup chamber while the growth chamber was sterilized with NaOH , rinsed , and emptied . Selections with the AT-rich gene were conducted in pKECMT [23] under control of the strong trc promoter . KA12/pKIMP-UAUC cells were transformed with pKECMT-AT-9 cm , adapted to selective CM-medium ( i . e . reduced CM medium lacking Tyr ) , and used to inoculate two separate pulse-feed alternating turbidostats at an initial OD600 of 1 . 46 . After 45 days of continuous growth under selective conditions ( in the absence of Tyr , Phe and tetracycline ) , four and five clones from the individual cultures were sequenced and analyzed by restriction digestion of their CM-encoding plasmids ( Figure S7 ) . Dimeric models of the reduced-alphabet CMs were created by homology-modeling , based on the EcCM X-ray structure [27] , followed by in silico mutagenesis to generate 10-CM and 11-CM from 9-CM . One TSA molecule was added to each active site . All structures were first energy-minimized using the GROMOS 53A6 force field [38] , [39] prior to further analysis . Molecular dynamics ( MD ) simulations of each protein/ligand system in SPC water were run using the GROMOS05 software [40] under NpT conditions with a pressure of 1 atm and a temperature of 293 K . All analyses were carried out using the GROMOS++ suite of programs [41] . | Proteins are linear polymers of a set of typically 20 different amino acid building blocks . The amino acid sequence—encoded by a genetic template—directs the folding of newly synthesized proteins into compact 3D structures and dictates the function of the protein product . Monomers containing distinct physico-chemical properties and geometries allow the formation of highly sophisticated architectures , and diverse functional groups enable enzymes to catalyze a plethora of chemical transformations . Nevertheless , the biochemical rationale for the exact composition ( and particularly the redundancy ) of the proteinogenic amino acid alphabet , which contains multiple building blocks that are chemically similar , remains enigmatic . By subjecting a simplified enzyme—constructed from only nine different amino acids—to directed evolution , we were able to investigate the impact of amino acid diversity on protein function . The most prolific variant selected in the course of the experiments expanded its amino acid alphabet , albeit through two surprisingly subtle mutations ( isoleucine to threonine and leucine to valine ) . The mutations improve both stability and catalytic activity of the enzyme , thereby demonstrating that the presence of structurally similar amino acids specified by the genetic code is highly beneficial for protein fitness . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"biochemistry",
"forms",
"of",
"evolution",
"directed",
"evolution",
"evolutionary",
"biology",
"chemical",
"biology",
"biology"
]
| 2013 | Directed Evolution of a Model Primordial Enzyme Provides Insights into the Development of the Genetic Code |
Regulatory networks have evolved to allow gene expression to rapidly track changes in the environment as well as to buffer perturbations and maintain cellular homeostasis in the absence of change . Theoretical work and empirical investigation in Escherichia coli have shown that negative autoregulation confers both rapid response times and reduced intrinsic noise , which is reflected in the fact that almost half of Escherichia coli transcription factors are negatively autoregulated . However , negative autoregulation is rare amongst the transcription factors of Saccharomyces cerevisiae . This difference is surprising because E . coli and S . cerevisiae otherwise have similar profiles of network motifs . In this study we investigate regulatory interactions amongst the transcription factors of Drosophila melanogaster and humans , and show that they have a similar dearth of negative autoregulation to that seen in S . cerevisiae . We then present a model demonstrating that this stiking difference in the noise reduction strategies used amongst species can be explained by constraints on the evolution of negative autoregulation in diploids . We show that regulatory interactions between pairs of homologous genes within the same cell can lead to under-dominance — mutations which result in stronger autoregulation , and decrease noise in homozygotes , paradoxically can cause increased noise in heterozygotes . This severely limits a diploid's ability to evolve negative autoregulation as a noise reduction mechanism . Our work offers a simple and general explanation for a previously unexplained difference between the regulatory architectures of E . coli and yeast , Drosophila and humans . It also demonstrates that the effects of diploidy in gene networks can have counter-intuitive consequences that may profoundly influence the course of evolution .
Negative autoregulation is a network motif in which a transcription factor inhibits its own expression . Theoretical work has shown that this type of regulation reduces intrinsic noise and quickens the response time to environmental perturbations [1]–[3] and experiments using artificial gene regulatory circuits in E . coli have confirmed these predictions [2] . Negative autoregulation therefore represents a simple yet powerful mechanism to maintain cellular homeostasis in the face of environmental and metabolic perturbations and reduce the often substantial fitness costs that noise can incur [4] . Different organisms , however , vary a great deal in their use of the motif . In E . coli , close to 50% of transcription factors ( 82 out of 182 ) [5]–[8] have been shown to negatively autoregulate . In contrast , negative autoregulation is almost entirely absent amongst the transcription factors that have been studied in S . cerevisiae ( 3 out of 169 ) [6] , [8]–[11] . How can we account for this discrepancy ? In order to answer this , we looked at the extent to which negative autoregulation is used in other species . We interrogated systematic datasets on the regulatory interactions amongst the known transcription factors of D . melanogaster and humans and found a similar pattern to that observed in yeast: in D . melanogaster 3 out of 87 [12]–[14] and in humans 5 out of 301 [13]–[15] transcription factors negatively autoregulate ( see SI , Table S1 , S2 , S3 ) . Currently , there is no obvious way to account for this striking discrepancy between these organisms , despite widespread interest in the strategies they employ to tackle noise [1]–[4] , [16]–[18] . Here we develop a model , founded in biophysics , for the evolution of negative autoregulation in diploid species . We use it to support the hypothesis that a dearth of negatively autoregulating genes in yeast , flies and humans can be explained by constraints on the evolution of negative autoregulation that arise due to diploidy .
Previous theoretical work on the dynamics of gene expression under negative autoregulation has considered single genes and so is implicitly haploid [1]–[3] , [18] . Such models exclude the more complex interactions that occur due to cross-regulation between homologous gene copies within a diploid cell ( Fig . 1 ) . Here we characterise the expression dynamics and regulatory evolution of homologous pairs of negatively autoregulating genes , taking into account the cross-talk between alleles . We model negative autoregulation in a diploid using a set of ordinary differential equations that track changes in the mRNA and protein concentrations for each of a pair of alleles ( labelled with subscripts and ) , , , and . The total concentration of mRNA and protein in the diploid cell are given by the summed output of the two alleles and . Changes in mRNA and protein concentrations for the pair of alleles over time are given by ( 1 ) According to these equations , mRNA is transcribed at a ( usually low ) constant background rate , plus a rate due to negative autoregulation , that decreases as the total cellular protein level increases . Protein is produced from mRNA at the rate of translation , whilst protein and mRNA degrade with rates and , respectively . As in previous work [1] , [2] , we model the repression function in Eqs . 1 as a Hill functionwhere is the dissociation constant associated with the autoregulating transcription factor binding site . Smaller values of ( lower rates of dissociation ) indicate stronger regulation . The Hill coefficient governs the steepness of the function at the inflection point and hence determines how step-like regulation will be . In systems where transcription is regulated by a single binding site , has a Michaelis-Menten-like form , corresponding to a Hill coefficient of [2] , [19] , [20] . A single binding site is the simplest , and perhaps the most relevant case for evolving negative autoregulation , and it is the one we focus on here . We analyse the more general case of arbitrary Hill coefficient in the Methods and in the SI we show that our results also hold for different values of . In the absence of negative autoregulation ( i . e . , ) , mRNA is produced at the maximum rate of transcription . In this case , concentrations of mRNA and protein reach equilibrium values of and . Starting from these values , equilibrium mRNA and protein levels decrease with increasing autoregulatory binding strength ( decreasing ) . The minimum mRNA and protein levels are reached when negative autoregulation is strongest ( i . e . as ) . The resulting minimum equilibrium concentrations are and . In order to analyse the evolution of autoregulatory binding sites we consider two separate but related functions of negative autoregulation: faster response times and maintaining mRNA and protein homeostasis . First , to study the evolution of negative autoregulation for faster response times , we simply equate the fitness of a system with its response time ( i . e the time taken to return to equilibrium following a perturbation ) . We use Eqs . 1 to infer selection pressures on the strength of autoregulation , i . e . , the dissociation constant , by analysing how quickly genotypes with different autoregulatory binding strength return to equilibrium following a perturbation in protein level . To do this we calculate a genotype's “response time”: the time taken for cellular protein concentration to return to equilibrium following a perturbation . We model perturbations as a reduction of the protein level to a fraction of the equilibrium level . The value of varies continuously between and to encompass both small perturbations , for example those resulting from intrinsic noise in transcription and translation ( ) , and larger perturbations , for example those resulting from resource deprivation in the environment or following cell division [17] . We present results derived from numerical analysis of Eqs . 1 that are applicable to perturbations of any size . These are complemented with an analytical treatment of the response time of the system to small perturbations , based on its maximal eigenvalue ( see Methods ) , which allows us to develop an intuition for how autoregulating genes in diploids respond to perturbations . To study the evolution of negative autoregulation for homeostasis , we turn to stochastic simulations of negatively autoregulating genes , which allow us to assess the amount of intrinsic noise associated with gene expression . Previous work has shown that negative autoregulation can help maintain homeostasis in gene expression by reducing the amount of intrinsic noise in negatively autoregulating genes , compared to other genes [3] . In fact , reducing the response time of a gene to very small perturbations away from equilibrium , also decreases the intrinsic noise in gene expression . Therefore , the two functions of negative autoregulation we consider ( producing faster response times and reduced intrinsic noise ) are highly inter-related . To study the evolution of negative autoregulation for lower intrinsic noise , we equate the fitness of the system with the amount of intrinsic noise it displays ( i . e the ratio of the variance in gene expression to the mean gene expression level ) . We infer selection pressures on the strength of autoregulation , i . e . , the dissociation constant , by the intrinsic noise of genotypes with different autoregulatory binding strengths . These are determined by performing Monte Carlo simulations for a full , molecular model of transcription , translation and autoregulation ( see Methods ) . We first compare the response times of two homogozyotes whose alleles are identical in every respect except for the dissociation constant . One homozygote carries two copies of a resident allele with dissociation constant , the other carries two mutant alleles that have a decreased dissociation constant ( with ) and hence stronger autoregulatory binding . Numerical analysis of the system shows that homozygotes for the more strongly autoregulating allele ( with ) respond more quickly than homozygotes for the more weakly autoregulating allele ( with , Fig . 2a ) . This is true up to a value of , which provides the fastest response time attainable by the system and hence provides the optimal binding strength . Further increases in regulation beyond this value are not favoured and lead to overshooting the optimal binding strength . These results for diploid homozygotes mirror those obtained for haploids [2] ( see Methods ) and show that regulatory interactions between pairs of identical alleles do not , in themselves , diminish the beneficial effects of negative autoregulation . Negative autoregulation can therefore , in principle , function as a mechanism to produce faster response times in diploids just as it does in haploids . The results above depend on comparing homozygotes for alleles with different dissociation constants , and . The evolution of negative autoregulation , however , must occur through the stepwise accumulation of new mutations that are initially rare and found only in heterozygotes . In order to assess whether autoregulation can evolve in diploids , we therefore need to determine whether a mutant allele with a stronger binding site ( ) will confer a selective advantage to a heterozygote that also carries a resident allele with a weaker binding site ( ) . A mutation will be favoured and increase in frequency if a heterozygote is able to respond more quickly to perturbations than a homozygote carrying two copies of the more weakly binding resident allele . Numerical analysis of Eqs . 1 reveals that heterozygotes often have greater response times than homozygotes with the more weakly binding resident allele . Fig . 2b shows that heterozygotes only have improved response times when the resident allele binding strength is weak ( [1] , [2] ) or if the effect of a mutation that increases binding strength is small ( is small ) . As the resident allele binding strength increases ( i . e . increases ) an ever larger range of mutation sizes result in increased heterozygote response times ( Fig . 2b ) , resulting in under-dominance ( i . e . heterozygote disadvantage ) . Typicaly mutation sizes for transcription factor binding sites are in the range , [20]–[22] . In this range regulatory mutations are subject to under-dominance even when the resident allele has relatively weak binding strength , and increasingly so as the binding strength of the resident allele increases . As a consequence , the maximum binding strength that can evolve is likely to be significantly lower than in haploids ( Fig . 2 ) . Based on these results , we expect under-dominance to pose a significant barrier to the evolution of negative autoregulation in diploids . To better understand why under-dominance arises in this system , we calculated the eigenvalues associated with Eqs . 1 . These provide a measure of the rate at which the system returns to equilibrium following a small perturbation , and allow us to elucidate the relative contributions of the different alleles to the response dynamics of the gene pair . The maximal eigenvalue of Eqs . 1 for a heterozygote , , can be expressed as ( 2 ) ( see Methods ) where is the squared difference of the mean steady state expression levels of the two alleles in the heterozygote and is the maximal eigenvalue of a homozygote with protein concentration equal to that of the heterozygote at equilibrium , ( see Methods ) . Eq . 2 says that , even if increasing autoregulatory binding strength leads to a faster response time in a homozygote , this advantage is offset in the heterozygote by an amount , which measures how different the expression levels of the two alleles are ( it is analogous to the Fano factor , a measure of the spread in a probability distribution [3] ) . As the difference in the expression of the alleles increases , increases from to a maximum . We can understand why increasing the difference in allelic expression results in increased response time by considering the contribution of the individual alleles to the response time of the gene pair ( Fig . 3 ) . The level of negative autoregulation at each allele depends on the strength of its binding site and the amount of protein product present in the cell . In a heterozygote , the allele with the stronger binding site is more strongly suppressed ( compared to the same allele in a homozygote ) , since there is more protein available to bind to it . At the same time , the allele with the weaker binding site is less strongly suppressed compared to the same allele in a homozygote . As a result , the allele with the stronger binding site has a faster response time than in a homozygote , whilst the allele with the weaker binding site has a slower response time than in a homozygote . However , the overall effect tends to be to increase the response time of the heterozygote , because the dynamics of protein expression in the heterozygote are dominated by the allele with the weaker binding site ( Fig . 3 ) . Under-dominance for response time occurs across a wide range of parameter values , but can be avoided if mutations have small effects on binding site strength ( Fig . 2b ) . To determine whether a series of mutations with small effect could offer a feasible way for genes to evolve strong negative autoregulation in diploids , we carried out simulations of binding site evolution that incorporated established properties of real binding sites . Transcription factor binding sites in eukaryotes vary between and nucleotides in length , with an average of 10 nucleotides [23] . They have a small number of optimal sequences that bind the transcription factor with maximum affinity [20]–[22] , [24] , [25] . The binding strength of a site can be expressed as a function of the total binding energy of its sequence , so . This total binding energy is generated by the additive contributions of individual nucleotides to overall binding , . Individual contributions are set to for nucleotides that do not match the optimal sequence and for matched nucleotides [20]–[22] . Based on these properties , we performed simulations of the evolution of an autoregulatory binding site under selection for decreased response time . These took into account the empirical distribution of binding site length in model eukaryotes and the variation in contributions to binding strength across the binding site sequence ( see Methods ) . The values of were drawn from a uniform distribution in the interval . This sampling covers the empirically estimated range [20]–[22] . It also ensures that mutations of small effect ( ) occur frequently and so allows for the possibility that autoregulation could evolve via the accumulation of mutations with small effect . Evolution was started from a state of minimum affinity ( all nucleotides non-optimal ) and proceeded through a series of single nucleotide substitutions . A mutant was assumed to go to fixation if it resulted in a response time less than or equal to that of the resident . Simulations were carried out for both haploids and diploids ( for which the response time of mutants was evaluated in the heterozygote state ) . The results ( Fig . 4 ) confirm that under-dominance strongly constrains the evolution of negative autoregulation in diploids . Haploids readily evolved binding sites with dissociation constants close to . In contrast , the average binding strength in diploids was around 100 times weaker than and only a small proportion of sites reached binding strengths comparable to those of haploids . This shows that under realistic conditions , diploids will rarely be able to evolve the level of autoregulation observed in haploids . In order to investigate the evolution of negative autoregulation as a mecahnism to reduce intrinsic noise in diploids , we turned to stochastic simulations . Intrinsic noise in gene expression occurs because transcription and translation are inherently noisy processes: all genes experience constant fluctuations in their mRNA and protein levels . The greater intrinsic noise associated with a particular gene , the higher the variance in its expression level relative to the mean . Therefore , a natural way to characterise the amount of intrinsic noise associated with a gene is to measure the ratio of the variance to the mean expression level at equilibrium ( known as the Fano factor ) [3] . We performed molecular simulations that capture transcription , translation and degredation in the presence of negative autoregulation ( see Materials and Methods ) . Just as in our analysis of response times , we compared a resident allele with dissociation constant , to a mutant allele with dissociation constant . We compared the intrinsic noise ( as measured by the Fano factor ) in the resident homozygote to that of the heterozygte and the mutant homozygote , and thus determined whether under-dominance occurs in the evolution of negative autoregulation as a mechanism to reduce intrinsic noise . The results are shown in Fig . 5 . We find once again that under-dominance occurs . Whereas the optimal binding strength for a single negatively autoregulating binding site is found to be , the maximum evolvable binding strength ( i . e that which can evolve without encountering under-dominance ) is found to be , an order of magnitude weaker . A similar pattern occurs when steeper Hill coefficients are considered ( Fig . 5 ) . Therefore we conclude that under-dominance poses a barrier to the evolution of strong negative autoregulation both as a mechanism to speed response times and to reduce intrinsic noise . To test the generality of our findings , we also considered variation in other parameters ( see SI Fig . S1 , S2 , S3 , S4 , S5 and Text S1 ) . We first relaxed our assumption of a single binding site and explored the case of Hill coefficients , implying regulation through multiple , cooperatively acting binding sites . In line with the effect of increasing binding strength through changes in , we find that mutations increasing the Hill coefficient are subject to under-dominance ( see SI Fig . S1 , S2 and Text S1 ) . Therefore , a mutation that increases the strength of negative autoregulation is subject to the same evolutionary constraints , independent of whether they increase regulation by changing the dissociation constant or the Hill coefficient . We also considered variation in the rates of mRNA and protein degradation ( and ) to see whether they provide conditions in which the effects of under-dominance on autoregulatory binding strength can be avoided ( see SI Fig . S4 and Text S1 ) . Variation in the rate of mRNA or protein degradation did not remove the tendency for mutations that increase autoregulatory binding strength to be subject to under-dominance . However , as has been pointed out elsewhere [17] , [26] , faster rates of protein degradation result in faster response times , and regulation of protein degradation can reduce noise . As might be expected , the constraints we describe on the evolution of response times through stronger negative autoregulation do not preclude the evolution of response times through other mechanisms , such as changes in protein degradation rates .
We have put forward the hypothesis that regulatory interactions between homologous genes can generate deleterious effects that constrain the evolution of negative autoregulation . The predictions of our model show that the high incidence of autoregulation in E . coli and the dearth of negatively autoregulating genes in yeast , flies and humans can be reconciled by taking into account a simple biological attribute—ploidy . Importantly , the difference between haploid and diploid regulation dos not appear to be a mere correlate of the prokaryote-eukaryote divide . This was already suggested by the finding that the genetic networks of E . coli and yeast are—with the exception of their use of autoregulation— very similar [11] . More generally , our work demonstrates that regulatory evolution can be considerably complicated by the presence of multiple copies of a gene in a cell , as is typically the case for eukaryotes . By explicitly considering the evolution of regulatory interactions , we have highlighted constraints that would not be evident from an analysis of the functional properties of an existing regulatory interaction in isolation—strong negative autoregulation quickens the response of genes to perturbation , but it is hard to evolve for this purpose due to under-dominance . This evolutionary perspective needs to be absorbed into attempts at unravelling the function of regulatory networks in higher organisms , a key problem for systems biology .
We used simulations of the molecular dynamics within a cell to determine the amout of intrinsic noise of autoregulating genes in diploids . A model that tracks the number of mRNA and protein molecules for a negatively autoregulating gene within a haploid cell is described in [3] . We generalised this to account for diploidy . The state of the system is described by the number of mRNA molecules , and the number of protein molecules produced from the two alleles . The probability of a state is specified by the joint probability distribution . The transition probabilities for the system to move between states due to changes in and ( and , analogously , due to changes in and ) are given bywhere , is the rate at which mRNA molecules are transcribed from allele 1 , is the rate of mRNA degradation , is the rate at which mRNA is translated into protein and is the rate of protein degradation . As in the ODE model , is a function of the number of proteins present in the cell , such thatwhere is the maximum rate of mRNA transcription , and is the dissociation constant of the binding site of allele 1 . To calculate response times we first determined the equilibrium expression level of the system from the average of replicate Monte-Carlo simulations . We then reduced mRNA and protein levels to a fraction of the equilibrium level . The time for each replicate to return to equilibrium was measured and the average across the ensemble used as an estimate of the response time of the system . In order to determine how response times vary with the level of perturbation , simulations were run for values of between 0 and 1 in steps of 0 . 01 . Binding site evolution was modelled by generating a transcription factor binding motif with a length nucleotides and an optimal base associated with each nucleotide . As in other models of TF-DNA binding , when a given nucleotide was matched for for the optimal base it contributed an amount to binding energy , otherwise it contributed 0 [20]–[22] . Binding site lengths were drawn from an empirical distribution generated from the binding motifs of 454 eukaryotic transcription factors contained in the JASPAR CORE database [23] . The value of for each nucleotide was drawn from a uniform distribution in the interval . The optimal binding strength was determined numerically ( see Methods ) , using the values for the system parameters that are given in the legend of Fig . 4 4 . We excluded from our analysis any binding sites for which the total binding strength of the optimal sequence was too low to achieve the fastest response time ( i . e . , those sequences for which ) . Evolution started from a state of minimum affinity ( all nucleotides non-optimal ) and proceeded through a series of single nucleotide substitutions . At each time step , a random mutation was introduced into the binding site sequence , switching one nucleotide from the non-optimal to the optimal state . If the mutation resulted in a response time less than or equal response time of the resident , the mutant sequence was assumed to go to fixation in the population . Deleterious mutations that increased response times were assumed to be lost . The simulation was ended when no further advantageous mutations were available . Simulations were carried out for both haploids and diploids ( for which response time of mutants was evaluated in the heterozygote state ) . Here we derive results for the response time of a haploid autoregulating gene . We derive results for the general case in which autoregulation is described by a Hill function with arbitrary coefficient ( the analyses in the main text assumes ) . The set of ODEs describing transcription and translation of mRNA and protein at a single autoregulating gene are analogous to those given for one allele in Eqs . 1 for a pair of autoregulating genes in a diploid . In order to simplify the analysis of the system we make the change of variableswith and . The dynamics of the system can then be rewritten as ( 3 ) where . In general since and ( 4 ) is the rescaled form of the repression function described in the main text . Assuming that mRNA decays much faster than protein [2] , [3] , then , it follows that is small relative to and we can assume that transcription output goes to equilibrium rapidly . That is , we can take and hence that the quasi equilibrium condition holds . Substituting into Eqs . 3 , generates a 2-dimensional system that is well approximated by the 1-dimensional system ( 5 ) The Lyapunov exponent associated with Eq . 5 at equilibrium gives the rate at which the system returns to equilibrium following a small perturbation . It is given by ( 6 ) Eq . 6 is always negative . In what follows we will discuss only the magnitude of the Lyapunov exponent with the understanding that this quantity is always negative and therefore describes the rate at which the system returns to equilibrium . From Eq . 6 it is clear that a mutation which increases will always serve to decrease the Lyapunov exponent and thus increase the rate at which the system converges to equilibrium . We compare a wild-type binding site , with dissociation constant , to a mutant binding site with dissociation constant such that —meaning that the mutant has a stronger binding site than the wild-type . At equilibrium , the protein concentrations satisfy ( 7 ) It is simple to show that by differentiating , with respect to . Thus , strengthening the autoregulatory binding site ( i . e . , decreasing ) will lead to a decrease in the equilibrium protein concentration , and so with we always have . To calculate the value of for which is maximum , we note thatAt equilibrium , and the Lyapunov exponent can be written asand we can find the value of that results in the largest Lyapunov exponent . This is given by Translating this back into units of protein concentration , this means that the fastest response to small perturbations about equilibrium occurs when ( 8 ) Thus , mutations which increase the strength of negative autoreguation , ( and therefore decrease ) , will decrease response time provided the equilibrium protein concentration is , as discussed in the main text . The optimal binding site strength can be determined by calculating the value of which gives the optimal equilibrium protein concentration of Eq . 8 . In the general case of arbitrary , cannot be found analytically , but it can always be found numerically . The derivation of presented here is based on the assumption that perturbations of the system are small , in which case the dynamics of the system are well captured by its Lyapunov exponent . The optimal binding strength under perturbations of arbitrary size can be obtained by numerical integration of the system . As might be expected , the values obtained in this way are similar to those calculated for small perturbations above . For diploids we proceed in the same way as for a single gene , and obtain a 1-D system for expression of a pair of alleles ( with dissociation coefficients and ) ( 9 ) where We now consider the response time of a pair of autoregulating alleles in a diploid . When an organism is homozygous , both binding sites have the same dissociation constant , and Eq . 9 is of the same form as Eq . 5 for a haploid , and the results for response time in haploids can be applied . When an organism is heterozygous however , the results for haploids do not hold . We compare the Lyapunov exponents of a heterozygote with dissociation constants and , where , to a resident homozygote in which both binding sites have strength . At equilibrium the total protein concentrations satisfy ( 10 ) where is the equilibrium protein concentration of the ( resident ) homozygote and is the equilibrium expression of the ( mutant ) heterozygote . It is simple to show that . by differentiating Eq . 10 with respect to . Following a small displacement from equilibrium , under-dominance will occur if the heterozygote has a smaller Lyapunov exponent than the homozygote . The maximal Lyapunov exponent of the system is given by ( 11 ) for the homozygote , and ( 12 ) for the heterozygote , where referes to allele in a diploid carrying alleles and . We can observe that the squared difference in the mean allele expression , , is given by , which can be expanded to give Substituting this expression for in Eq . 12 we find ( 13 ) ( 14 ) Note that Eq . 14 is of the same form as Eq . 13 , with an additional term that depends on the ratio of the squared difference in allele expression , to the total expression . We can define to be the Lyapunov exponent associated with a homozygote of a given equilibrium expression and to be the Lyapunov exponent associated with a heterozygote of the same equilibrium expression and obtain Eq . 2 of the main text ( with ) . | All genes have to deal with intrinsic noise , and a variety of mechanisms have evolved to reduce it . One important mechanism of noise reduction for transcription factors is negative autoregulation , in which a gene product represses its own rate of transcription . Negative auotregulation occurs frequently in E . coli but , we find , occurs much more rarely in S . cerevisiae , D . melanogaster and humans . Whilst there are a great many important differences in the genetic architectures of these organisms , they tend to share , with the exception of negative autoregulation , similar profiles of network motifs . This makes the discrepancy in the degree of negative autoregulation all the more striking , as it lacks any obvious explanation . Our study presents a potential explanation , by comparing the evolvability of negative autoregulation as a noise reduction mechanism in haploids and diploids . We show that , in diploids , mutations that increase the strength of negative autoregulation at one gene copy often increase overall noise in gene expression . This results in under-dominance , in which heterozygotes are less fit than homozygotes . The result is that the evolution of negative autoregulation in diploids is significantly constrained . We verify our results using a combination of detailed molecular simulations and evolutionary simulations | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"biology",
"evolutionary",
"biology"
]
| 2013 | Under-Dominance Constrains the Evolution of Negative Autoregulation in Diploids |
Herpesvirus latency is generally thought to be governed by epigenetic modifications , but the dynamics of viral chromatin at early timepoints of latent infection are poorly understood . Here , we report a comprehensive spatial and temporal analysis of DNA methylation and histone modifications during latent infection with Kaposi Sarcoma-associated herpesvirus ( KSHV ) , the etiologic agent of Kaposi Sarcoma and primary effusion lymphoma ( PEL ) . By use of high resolution tiling microarrays in conjunction with immunoprecipitation of methylated DNA ( MeDIP ) or modified histones ( chromatin IP , ChIP ) , our study revealed highly distinct landscapes of epigenetic modifications associated with latent KSHV infection in several tumor-derived cell lines as well as de novo infected endothelial cells . We find that KSHV genomes are subject to profound methylation at CpG dinucleotides , leading to the establishment of characteristic global DNA methylation patterns . However , such patterns evolve slowly and thus are unlikely to control early latency . In contrast , we observed that latency-specific histone modification patterns were rapidly established upon a de novo infection . Our analysis furthermore demonstrates that such patterns are not characterized by the absence of activating histone modifications , as H3K9/K14-ac and H3K4-me3 marks were prominently detected at several loci , including the promoter of the lytic cycle transactivator Rta . While these regions were furthermore largely devoid of the constitutive heterochromatin marker H3K9-me3 , we observed rapid and widespread deposition of H3K27-me3 across latent KSHV genomes , a bivalent modification which is able to repress transcription in spite of the simultaneous presence of activating marks . Our findings suggest that the modification patterns identified here induce a poised state of repression during viral latency , which can be rapidly reversed once the lytic cycle is induced .
Herpesviruses are able to establish latent infections , enabling them to persist for the lifetime of their host [1] . During latency , no viral progeny is produced; instead , the largely quiescent genome persists as an extrachromosomal episome in the nucleus of the infected cell . Unfavorable conditions ( e . g . cell stress ) may trigger reactivation of such cells , leading to induction of the lytic cycle and completion of the viral lifecycle . In a healthy host , latently infected cells form a reservoir of chronic viral infection which is tightly controlled by the immune system . However , latently infected cells may also give rise to disease if the immunological control is lost . This is especially true for the members of the gammaherpesvirus subfamily , which are frequently associated with tumors in their natural host , in particular in immunosuppressed individuals . Kaposi Sarcoma-associated herpesvirus ( KSHV ) is etiologically linked to Kaposi Sarcoma ( KS ) , a tumor of endothelial origin , as well as at least two lymphoproliferative disorders , primary effusion lymphoma ( PEL ) and multicentric Castleman's disease ( MCD ) [2] , [3] , [4] . The majority of tumor cells in these malignancies exhibit a latent gene expression profile which has been extensively studied in cell lines established from PEL tumors [5] , [6] , [7] . These cells express a very limited contingent of viral genes , including the latency-associated nuclear antigen LANA ( encoded by ORF73 ) which permits replication of latent episomes , a viral cyclin D homologue ( v-Cyc/ORF72 ) , a viral homologue of a FLICE-inhibitory protein ( v-Flip ) encoded by ORF71 ( also termed K13 ) and Kaposin ( ORF K12 ) , a protein that can stabilize cytokine transcripts [7] , [8] , [9] , [10] , [11] , [12] . All of the above proteins are translated from alternatively spliced mRNAs transcribed from a single multicistronic locus; primary transcripts from the locus furthermore can give rise to 12 virally encoded microRNAs ( miRNAs ) [7] , [13] , [14] , [15] , [16] , [17] , [18] . It is thought that , together , these genes serve to ensure persistence of the latent infection and survival of the host cell . However , several of the latency genes have also been shown to exhibit tumorigenic properties in various experimental systems , supporting the idea that the viral latency program plays a causative role during onset and/or progression of KSHV associated tumors . The viral genes which encode components of the lytic or productive cycle are transcriptionally silent during latency . This quiescent state of infection can be overturned by forced expression of Rta ( the product of the ORF50 gene , also termed Lyta ) , a homologue of the Epstein-Barr virus ( EBV ) transactivator Rta [19] , [20] , [21] . Upon expression , Rta acts as a master-switch regulator which orchestrates the expression of downstream lytic genes , leading to massive amplification of viral genomes , followed by assembly of virions and , ultimately , death of the host cells and release of viral progeny [20] , [21] , [22] , [23] . How Rta and other lytic genes are kept silenced during latency is not understood , but it is very likely that epigenetic modifications play an important role during this process . This notion is supported by the fact that treatment of latently infected PEL cells with inhibitors of DNA methyltransferases as well as histone deacetylases induces lytic cycle replication , and that lytic cycle induction leads to profound chromatin rearrangements at several loci [24] , [25] , [26] , [27] . Furthermore , the ORF50 promoter was reported to be subject to DNA methylation in latently infected PEL cells whereas the latent ORF73 promoter remained unmethylated , and it has therefore been suggested that CpG methylation may actively repress Rta expression during latency [26] . The DNA methylation status of other regions of the KSHV genome , however , has so far not been analyzed . Likewise , the current knowledge about global histone modification patterns during viral latency is very limited . All studies of latent modification patterns have furthermore been performed in PEL cells and thus describe the epigenetic status during fully established latency . However , since the packaged virion DNA is devoid of DNA methylation as well as histones [28] and thus epigenetically naïve , such epigenetic modification patterns need to be re-established during each round of latent infection . Especially the early phase of a de novo infection thus represents a critical phase of the viral lifecycle . We have performed a comprehensive study of DNA methylation as well as histone modification patterns across the complete KSHV genome , in both PEL cells as well as a de novo infected endothelial cell line . We have observed highly distinct global patterns on the level of both DNA as well as histone modifications . Such patterns were furthermore highly similar in PEL cells and stably infected endothelial cells , suggesting a highly regulated modification program during latency establishment . However , whereas modified histones could be readily detected at early timepoints of a de novo infection , DNA methylation patterns evolved over significantly longer periods of time , suggesting they do not govern early latency expression patterns . Our analysis rather suggests that DNA methylation patterns evolve as a secondary result of the histone modification patterns , which are established early in the infection . Surprisingly , in spite of their quiescent state , latent KSHV episomes were also not devoid of activating histone marks: In fact , such marks occupied several lytic promoters soon after the de novo infection and were not stripped from the genomes in long term infected cells . However , concomitant with the appearance of these modifications , latent genomes were also subject to profound tri-methylation of lysine 27 of histone H3 ( H3K27 ) , a modification which can suppress transcription even in the presence of activating marks [29] , [30] , [31] . Thus , latent episomes bear the hallmarks of poised chromatin , an observation which is in line with the hypothesis that viral latency represents a meta-stable state of transcriptional repression which can be quickly reversed once the lytic cycle is induced .
In mammals , DNA methylation occurs almost exclusively by methylation of cytidine residues at CpG dinucleotides and is generally associated with transcriptional repression ( reviewed in [32] ) . As methylcytidine is prone to spontaneous deamination , an evolutionary consequence of DNA methylation is the relative scarcity of CpG dinucleotides in methylated genomes . In contrast to most members of the alpha- and betaherpesvirus subfamily , the majority of gammaherpesviruses show evidence of such CpG suppression , suggesting that these viruses are subject to DNA methylation [26] . Furthermore , the genomes of EBV as well as the rhadinovirus Herpesvirus saimiri ( HVS ) have been found to carry methylated CpG motifs at multiple loci in latently infected cells , suggesting that DNA methylation plays a role in the control of latent gammaherpesvirus gene expression patterns ( reviewed in [33] , [34] ) . So far , analysis of DNA methylation within latent KSHV genomes has been limited to the promoters of the gene encoding Rta ( i . e . ORF50 ) and the promoter upstream of ORF73/LANA which drives expression of the latency gene cluster . While no CpG methylation was detected in the region of the ORF73 promoter , the ORF50 promoter was found to be heavily methylated in the PEL derived cell line BCBL-1 [26] . As promoter activity was furthermore repressed by DNA methylation in an in vitro assay , it was suggested that CpG methylation actively suppresses expression of the lytic switch gene Rta during KSHV latency [26] . However , this hypothesis is complicated by the fact that the same study also found that the majority of samples from different KSHV-positive tumor samples did not harbor these methylation patterns . The authors suggest that their observations may have been due to the presence of lytic cells , which represent a sub-population among the mostly latently infected cells in some tumor types . Given the absence of comprehensive DNA methylation data for KSHV , we first sought to determine the global methylation status of viral episomes in PEL derived cell lines . For this purpose , we employed the MeDIP ( methylated DNA immunoprecipitation ) technique , which is based on the pulldown of methylated DNA using methylcytidine-specific antibodies [35] . The MeDIP samples were analyzed on a custom-designed , high-resolution microarray which covers both strands of the KSHV genome in non-overlapping , hybridization temperature-optimized 60mers . To obtain a quantitative measure of the extent of DNA methylation , we additionally devised positive and negative controls according to the scheme depicted in Figure 1A . As a negative control , we employed a bacterially amplified ( and hence CpG methylation free ) bacmid clone which carries the complete KSHV genome [36] . For a given DNA fragment , the amount of DNA which can be maximally recovered by MeDIP is dependent on the number of CpG motifs in that sequence . Consequently , the array hybridization patterns are a function of the relative degree of methylation as well as local CpG frequency . To control for such differences we generated a positive control by in vitro methylation of KSHV bacmids using the methylase M . SssI , which is specific for CpG dinucleotides . Restriction analysis confirmed complete methylation of the bacmid DNA ( Fig . 1B ) . Prior to immunoprecipitation , the untreated or in vitro methylated bacmid DNA was mixed with DNA from human cell lines to also control for any signals which may arise due to cross-hybridization of cellular DNA in the infected samples . The ratio of viral and cellular DNA was selected such that it is equal to that typically seen in KSHV-infected PEL cell lines and corresponds to a viral copy number of approximately 30 genomes per cell . Furthermore , all samples were spiked with a constant amount of in vitro methylated heterologous DNA , so that accurate normalization across individual array hybridizations could be performed . After normalization , the MeDIP values obtained from the samples or the positive control were corrected by subtracting the background values from the negative control ( see Material & Methods for details ) . To minimize the risk of investigating cell-line specific ( and hence potentially random ) modifications , we analyzed the global DNA methylation pattern of KSHV genomes in three different PEL-derived cell lines: BCBL1 , AP3 and HBL6 . The HBL6 line was originally established from a PEL tumor co-infected with KSHV and EBV and carries both viruses in a latent state [37] . BCBL1 and AP3 cells are KSHV positive , but negative for EBV . In Figure 2 , we present the results of our analysis of PEL cells , along with the data obtained from the positive control ( upper four solid graphs in each panel; see also Figure S1 for a more detailed view with a differentially scaled x-axis ) . The distribution of local CpG frequencies across the KSHV genome is also shown ( black line graph ) . As expected , the positive control yielded a signal distribution which showed a high degree of correlation with local CpG content ( Pearson correlation coefficient = 0 . 513 , see Table 1 ) . The results from the PEL-derived cell lines revealed that , indeed , KSHV genomes are subject to profound DNA methylation during latency . For all three lines , we observed global methylation profiles which were strikingly similar , with overall correlation coefficients ranging from 0 . 593 to 0 . 724 ( Table 1 ) . Furthermore , the profiles were clearly not a mere function of CpG content , as several regions showed low levels of DNA methylation in all three PEL lines , but not the positive control . One such region , extending approximately from nucleotides ( nts . ) 127301 to 128901 , harbors the major latency promoter upstream of ORF73 . The absence of methylation in this area is to be expected ( and has been noted before [26] ) , given the constitutive activity of the promoter in latently infected cells . However , our analysis revealed several additional loci which were not ( or only poorly ) methylated , despite their ( presumable ) transcriptional inactivity in latently infected cells . For example , the region between nts . 9701 and 12601 showed very little methylation in PEL lines compared to the positive control; this area is centered on the start position of the gene encoding the DNA polymerase ( ORF9 ) , which is exclusively expressed during the lytic cycle . While the methylation profiles of the three PEL lines were highly similar , the absolute degree of methylation was different: Across the complete KSHV genome , HBL6 reached approximately 88% of the MeDIP signal obtained for the positive control , followed by AP3 ( 54% ) and BCBL1 ( 51% ) . To investigate whether the observed methylation profiles were specific to PEL cell lines or a general feature of latent genomes , we next sought to analyze genomes from cells which had established stable latency after KSHV infection in vitro . While the infection of non-adherent cells ( including B cells ) with KSHV in vitro is very inefficient , a wide variety of adherent cells can be readily infected by incubating the cultures with supernatants from lytically induced PEL lines [38] . However , although KSHV rapidly adopts a latent expression profile in these cultures , most infected cells tend to loose the viral episomes over the following cell divisions [39] . Only a small percentage of cells ultimately succeeds in establishing stable latent episomes , which are then propagated with the same efficiency as the genomes in PEL cells . Previously , we have established the SLKp sub-line from in vitro infected SLK cells , a cell line of endothelial origin [39] . The SLKp line was generated by pooling seven KSHV-positive single cell clones which had been isolated from an infected bulk cultures at approximately 65 days post infection . SLKp cells are stably infected , carry approximately the same episome copy number as BCBL1 cells ( 30–40 copies/cell ) , and have a strictly latent expression profile [39] . We analyzed SLKp cells which had been in continuous culture for 6 months , corresponding to a total time span of approximately 8 months after the original infection . As shown in Figure 2 ( red graph ) , although the overall methylation levels in SLKp cells were substantially lower ( reaching approximately 9 . 6% of positive control levels; note the differentially scaled y-axis in Figure 2 ) , the observed profile was indeed highly similar to that seen in PEL cells , with the highest degree of similarity to BCBL1 cells ( correlation coefficient 0 . 712 , see Table 1 ) . Taken together , these results suggest that the distinct MeDIP profiles revealed during our analysis are non-random and represent a characteristic of latent KSHV episomes . In order to confirm that the relative MeDIP values identified during our microarray-based analysis are indeed an accurate measure of CpG methylation levels , we investigated a number of loci using independent methods . First , based on our analysis of BCBL1 cells , we chose 3 loci which had registered as being strongly methylated , and another 3 for which our initial analysis had suggested the absence of DNA methylation . As shown in Figure 3A , bulk bisulfite sequencing established near-complete methylation at the former and absence of methylation at the latter loci . We selected one of the loci ( labeled 1 in Figure 3A ) which had shown differential methylation in PEL and SLKp cells for further analysis . Figure 3B shows an enlarged representation of the corresponding section of the KSHV genome , along with the original MeDIP array data . As shown in Figure 3C , quantitative real-time PCR amplification of an ∼100 bp segment at the center of the region ( indicated by the black bar labeled “qPCR” in Figure 3B ) confirmed the overall lower degree of methylation in SLKp cells and suggested an intermediate degree of methylation in AP3 cells , which is in accordance with the array data for this position . As discussed later , we also investigated de novo infected SLK cells at 5 days post infection ( SLK-5dpi ) , which showed very little evidence of methylation . Next , we performed a PCR amplification of bisulfite converted total DNA from all samples and subjected the amplified region to digestion with the restriction enzyme TaqI ( combined bisulfite restriction analysis assay , COBRA ) . The recognition sequence of TaqI contains a CpG motif , and as only methylated cytosine residues are preserved during bisulfite conversion , absence of DNA methylation at the restriction site leads to TaqI resistance . As shown in Figure 3D , unmethylated bacmid DNA as well as DNA isolated from KSHV virions were completely unmethylated and hence resistant to TaqI restriction . Likewise , DNA from freshly infected SLK cells remained intact . In contrast , the amplification products from the in vitro methylated bacmid as well as BCBL1 and HBL6 were completely cleaved , in agreement with methylation of all 4 TaqI sites , and the products from AP3 cells and SLKp cells were incompletely digested , indicating an intermediate level of methylation . The latter contained a significant amount of undigested product , suggesting that the material represents a mixture of methylated and unmethylated DNA , presumably due to clonal differences in the original single cell clones . We hence determined the specific sequence of bisulfite converted DNA from two individual clones , and subjected the remaining samples to bulk sequencing . The results of this analysis are shown in Figure 3E and are in perfect accord with our COBRA analysis and MeDIP array results . Indeed , the two investigated SLKp clones showed differential methylation patterns at this particular locus , thus explaining the observed restriction patterns . We hence conclude that our array analysis accurately reflects CpG DNA methylation within the KSHV genome . As noted before , the global methylation patterns in PEL and SLKp cells were highly similar: If a locus was found to be methylated in one of the samples it tended to be methylated also in the others . Only very few loci were methylated in only one sample . Interestingly , the most prominent locus which showed differential methylation was a region encompassing nts . 70500 to 71700 , which includes the promoter governing expression of ORF50/Rta . Our analysis suggested profound methylation in the HBL6 line , but very little or no methylation in AP3 , BCBL1 and SLKp cells ( see Figure 2 ( second panel ) and enlarged depiction of the ORF50 locus in Figure 4A ) . This was surprising , as the ORF50 promoter has been previously reported to be abundantly methylated in BCBL1 cells [26] . To confirm our results , we performed bulk bisulfite sequencing of the region extending from the transcriptional start of ORF50 to a position approximately 1100 bp upstream ( nts . 70597 to 71681 ) with DNA isolated from AP3 , BCBL1 , HBL6 , SLK-5dpi and SLKp and cells ( Figure 4B ) . Additionally , we subjected the two overlapping amplification products from BCBL1 , SLKp and HBL6 cultures to COBRA analysis ( Figure 4C ) . The results clearly confirmed our MeDIP results and revealed near complete CpG methylation of the ORF50 promoter region in the HBL6 line , but no or only sporadic methylation in all other cells . We currently do not know the reason for the different BCBL1 methylation patterns detected in our study and that performed by Chen et al . [26] . It is possible that Chen and colleagues employed a different sub-clone of the BCBL1 line , or that the lines may have diverged while being cultured in the two labs . The methylation patterns we detected in HBL6 cells are very similar ( but not identical ) to those described by Chen et al . , and thus an agreement between both studies is that such patterns can principally evolve in PEL cells . However , regardless of the reasons for the different findings , our results clearly show that methylation of the ORF50 promoter is not a principal requirement for the maintenance of latency in PEL cells or in vitro infected endothelial cells lines . As the majority of cells in PEL and KS tumors are latently infected with KSHV , our findings may also provide an alternative explanation for the observation that the ORF50 promoter was found to be not or only poorly methylated in the majority of clones derived from such tissues [26] . SLKp cells exhibited global methylation patterns which were near-identical to those seen in the BCBL1 line , but were characterized by a significantly lower absolute level of DNA methylation ( approximately 1/5th of that seen in BCBL1 cells ) . This observation suggested to us that DNA methylation of KSHV episomes may progress slowly over time; hence the lower overall extend of methylation would be a result of the comparatively short period of time ( approx . 8 months , see above ) that has elapsed since the SLKp cells were originally infected . We therefore analyzed SLK cultures which had been freshly infected with KSHV . We choose a time point of 5 days post-infection for our analysis; at this time point , the cultures have adopted latent expression patterns and sporadic lytic cells are found only at very low frequency ( ∼0 . 01% ) [38] , [39] . Quantitative RT-PCR ( Figure 5A ) and immunofluorescence analysis ( Figure 5B ) confirmed efficient infection and absence of lytic gene expression ( note that the relatively high basal levels of lytic ORF50 and ORF59 transcripts in latent BCBL1 cultures ( Figure 5A , top panel ) can be significantly upregulated by lytic cycle induction ( bottom panel ) ; they stem from the small number ( approximately 0 . 3% , see Figure S4 ) of spontaneously reactivating cells present in uninduced BCBL1 cultures ) . Indeed , our array-based MeDIP analysis of global DNA methylation patterns of SLK-5dpi cultures revealed very little DNA methylation at this early time point of infection ( see graphs labeled SLK-5dpi in Figures 2 , 3B and 4A ) , reaching , on average , less than 1% of the levels observed for the positive control . DNA methylation was also virtually absent from the ORF50 promoter , in spite of the fact that the infected cultures had established a strictly latent infection . These findings support our hypothesis that , although DNA methylation may reinforce latent gene expression patterns at late timepoints of infection , ORF50 promoter methylation is principally not required to abolish or prevent Rta expression during KSHV latency . Given the absence of DNA methylation in our SLK-5dpi samples , we deemed it unlikely that this modification governs early KSHV latency expression patterns , and hypothesized that such patterns might rather be governed by histone modifications . Herpesvirus genomes are known to become rapidly chromatinized [40] after host cell entry , and the deposition of histone modifications is therefore expected to represent a much more dynamic process than DNA methylation . To investigate this hypothesis , we performed chromatin immunoprecipitation ( ChIP ) experiments from BCBL1 , SLKp cells or de novo infected SLK cultures and analyzed the precipitated DNA on our tiled microarrays using standard ChIP-on-chip protocols ( see Material & Methods for details ) . First , we investigated the distribution of two modifications which are commonly associated with active chromatin , using antibodies which are specific for Histone H3 acetylated at lysine 9 and/or 14 ( H3K9/K14-ac ) , or H3 molecules which are tri-methylated at lysine K4 ( H3K4-me3 ) . As shown in Figure 6 ( see also Figure S2 for a more detailed view ) , in BCBL1 as well as SLKp cells we observed global modification patterns which were highly similar when comparing any pairwise combination of either histone modification or cell line , with correlation coefficients ranging from 0 . 709 to 0 . 894 ( Table 2 ) . Furthermore , investigation of H3K4-me3 patterns in SLK-5dpi cultures showed that these patterns were indeed already fully established 5 days after de novo infection . Comparison with the previously observed CpG methylation patterns revealed a marked negative correlation between these histone modifications and DNA methylation , as most of the regions which had been found to be poorly methylated in BCBL1 or SLKp cells compared to the positive control showed abundant deposition of active histone marks . In accordance with the overall higher degree of DNA methylation , this negative correlation was most obvious in BCBL1 cells ( Pearson correlation coefficient = −0 . 530 , see Table 2 ) , but could also be clearly observed in SLKp cells ( correlation coefficient = −0 . 263 ) . Interestingly , while the highest density of CpG motifs within the KSHV genome is found at the terminal repeats ( TR , see rightmost region of the KSHV map ) , this is also the region which showed the highest levels of H3K9/K14-ac and H3K4-me3 enrichment . The latter is in agreement with the observation that , in spite of the high number of potential methylation sites , MeDIP signals were absent from this region ( Figure 2 ) . In fact , in bulk bisulfite sequencing reactions , we were unable to identify any DNA methylation within the terminal repeats in SLKp or BCBL1 cells ( data not shown ) . Hence , our data indicate that local deposition of active histone marks early during KSHV infection prevents the acquisition of DNA methylation over the ensuing cell divisions , ultimately leading to the establishment of the global methylation patterns as shown in Figure 2 . While it may account for the evolution of the observed DNA methylation patterns , the distribution of H3K9/K14-ac and H3K4-me3 modifications provides no immediate explanation for the establishment of latent expression profiles , as these marks were present on many loci which are transcriptionally inactive during latency . Notably , this also includes the ORF50 promoter , a finding which is in accordance with the absence of DNA methylation at this location in BCBL1 as well as SLKp cells . We therefore reasoned that latency may be determined by the presence of repressive marks rather than the absence of activating ones . Therefore , we analyzed two modifications commonly associated with silent chromatin: Tri-methylation of lysine 9 of histone H3 ( H3K9-me3 ) , which is a hallmark of constitutive heterochromatin , and tri-methylation of lysine 27 ( H3K27-me3 ) , a modification which is typically seen in facultative heterochromatin . As shown in Figure 7 ( lower graphs in each panel; see also Figure S3 for a more detailed view ) , in both BCBL1 and SLKp cells the H3K9-me3 modification was mainly restricted to two consecutive regions of the viral genome , spanning approximately nts . 33000 to 46000 ( ORF19-ORF25 ) and 100400 to 114400 ( ORF64- ORF67 ) . Both of these regions had shown relative poor occupancy with acetylated histones in our previous assay ( see Figure 6 ) , which is in agreement with the fact that these modifications in general are mutually exclusive . While the H3K9-me3 modification was most prominently detected in BCBL1 cells , SLKp cells did display a markedly less distinct pattern , and the modification was barely detectable in SLK-5dpi cultures . The ORF50 promoter was devoid of trimethylated H3K9 in all samples . Hence , H3K9-me3 is unlikely to be a major regulator of latent gene expression , at least not in the early phase of infection when latency is first established . Our findings are in agreement with a previous study that had investigated a number of select loci in ChIP experiments , and had found little to no H3K9 methylation at any of them [41] . Next , we analyzed the distribution of the H3K27-me3 modification across the viral genome . H3K27 tri-methylation is carried out by EZH2 , the enzymatic subunit of the polycomb PRC2 complex , leading to the recruitment of polycomb PRC1 complexes and thus gene silencing [29] , [30] , [31] , [42] , [43] , [44] . Tri-methylation of H3K27 has been shown to play important roles in developmental and differentiation processes , cell cycle regulation , mammalian X chromosome inactivation , stem cell identity and cancer [31] . One characteristic of H3K27 methylation is that , in contrast to H3K9-me3 , it can occupy promoters concurrently with activating modifications , specifically H3K4-me3 or H3K9-ac . Such regions are termed “bivalent” domains and have been found to be specifically enriched in embryonic stem cells , where they often occupy promoters which encode key factors involved in developmental regulation [42] . The presence of H3K27 methylation keeps these promoters silent in undifferentiated cells , but the chromatin remains in a “poised” state due to the simultaneous presence of activating marks . Decreasing levels of H3K27-me3 during the onset of differentiation allows such promoters to rapidly revert to an active state , hence further commiting the cell to terminal differentiation [42] . As shown at the top of each panel in Figure 7 , our analysis revealed that latent KSHV episomes in the BCBL1 and SLKp lines as well as SLK-5dpi cells were subject to abundant H3K27 tri-methylation . The modification was detected virtually across the complete genome , although most regions which had been found enriched in H3K9/K14-ac and H3K4-me3 modifications tended to be tri-methylated at H3K27 to a lesser extend ( compare with Figure 6 , see also correlation coefficients in Table 1 ) . A number of loci , however , displayed the hallmarks of bivalent chromatin , i . e . simultaneous presence H3K27-me3 and activating marks . Interestingly , the ORF50 promoter featured prominently among these regions , whereas the major latency promoter upstream of ORF73 showed very little or no H3K27 tri-methylation in all three samples . Importantly , in contrast to H3K9-me3 , the H3K27-me3 patterns were already present 5 days after de novo infection of SLK cultures . This also includes the ORF50 promoter , and our data thus suggest that a poised state of repression is imposed upon the ORF50 promoter early during the establishment of latency . Interestingly , two studies have recently found this modification to be present on herpes simplex virus genomes during latent infection in dorsal root or trigeminal ganglia [45] , [46] . Although only a small number of select promoters were investigated , this may indicate a general role for this modification during herpesvirus latency . As all of the KSHV-infected lines investigated here contain multiple copies of the viral episome , it may appear possible that the simultaneous detection of activating marks and the H3K27-me3 modification could be due to the existence of distinct , but separate episome populations . We think this is very unlikely , as we have observed the same patterns in BCBL1 cells and in vitro infected SLK cells . Thus , if distinct populations exist , they would have to be re-established in the exact same stoichiometry upon a de novo infection . However , to also directly investigate the presence of bivalent histone modifications , we have performed a sequential ChIP from BCBL1 cells , followed by qPCR amplification of two regions within the ORF50 promoter . As controls , we investigated the latent promoter upstream of ORF73 ( which exclusively carries activating marks ) and a region within the coding region of ORF21 ( which is subject to the H3K27-me3 modification , but is devoid of H3K9/K14-ac and H3K4-me3 marks ) . The location of the amplified regions and their histone modification profiles are depicted in Figure 8A . As shown in Figure 8B , when the first round of immunoprecipitation was carried out with an antibody specific for H3K9/K14-ac , the sequential ChIP using a H3K27-me3-specific antibody recovered material only from the ORF50 promoter , but neither of the two control regions . To confirm these results in the reverse direction , we also performed the first round of immunoprecipitation using the H3K27-me3-specific antibody and used a H3K4-me3 antibody for the second immunoprecipitation to probe for the presence of activating marks ( Figure 8C ) . Again , while the ORF21 region was recovered in the first round of ChIP , only the ORF50-specific sequences registered in both immunoprecipitation experiments , thus demonstrating bivalent modification of this promoter . If H3K27-me3 contributes to the silencing of the ORF50 promoter , these marks should also diminish upon lytic cycle induction . We therefore monitored the levels of H3K27-me3 during reactivation from latency . Indeed , sodium butyrate treatment of BCBL1 cells resulted in a progressive loss of H3K27-me3 at the ORF50 promoter , with a reduction to approximately 50% , 20% and 5% of the original levels after 24 , 48 and 72h of treatment , respectively ( Figure 8D ) . While this observation suggested efficient removal of H3K27-me3 , the magnitude of the effect at 48h and 72h post induction was surprising , given that the treatment only reactivates about 20% of all cells in the cultures ( as judged by staining for the late gene product ORF59 ) . A possible explanation for this observation is that the rapidly increasing numbers of replication products ( which are epigenetically naive ) exaggerate the effect at late time points , as they will lead to a relative increase in the percentage of unmodified episomes within the cultures . However , the 24h time point precedes the replication phase and accumulation of newly synthesized/packaged genomes thus cannot be responsible for the H3K27-me3 decrease observed early after induction . To further substantiate this assumption , we also monitored H3K9/K14-ac levels at the ORF50 promoter . We reasoned that , if the above is correct , prior to the onset of DNA replication we should first see an increase of the histone acetylation levels , followed by a decline as more and more replicated genomes accumulate . As shown in Figure 8D , this is precisely what we observed . Thus , upon lytic cycle induction , a reduction of H3K27-me3 and an increase of H3K9/K14-ac levels occur simultaneously at the ORF50 promoter and precede the DNA replication phase . In order to investigate whether a reduction of H3K27-me3 also results in an increase of the number of lytically reactivated cells in the absence of chemical inducers , we next generated BCBL1 and SLK cells which were stably transduced with a retrovirus that expresses the H3K27-me3-specific demethylase JMJD3 [47] . After antibiotic selection of the cultures for 12 days , the SLK cells were additionally infected with KSHV and analyzed 5 days later . In both lines , while the ectopically expressed JMJD3 protein was barely detectable on western blots ( data not shown ) , we nevertheless observed a reduction of total cellular H3K27-me3 levels of at least 50% ( Figure 9A ) . The reduction was less pronounced on the ORF50 promoter , which still exhibited about 70% and 80% of the H3K27-me3 levels seen in the vector controls of BCBL1 and SLK-5dpi cells , respectively ( Figure 9B ) . However , as shown in Figure 9C , despite of the moderate degree of this reduction , both cultures showed a marked increase in the overall levels of ORF50 transcription , which reached approximately twice the values as in the control cultures . When we stained the JMJD3-transduced BCBL1 cultures for the late gene product ORF59 , ( Figure 9D ) we furthermore observed a twofold increase in the percentage of spontaneously reactivated cells in the JMJD3-transduced BCBL1 cells ( ∼0 . 6% , compared to 0 . 3% in the control cultures ) . In addition , the JMJD3-transduced cells were also more responsive to lytic cycle induction by sodium butyrate treatment , resulting in the reactivation of 30% of the cultures ( compared to approximately 20% in the control cultures; Figure 9E ) . In comparison to PEL lines , SLK cells as well as most other de novo infected adherent cell lines exhibit an extremely low percentage of spontaneously reactivated cells [38] , [39] . Such cells do also not respond to chemical inducers which reactivate PEL cells ( e . g . phorbol esters ) , although the lytic cycle can be induced by ectopic Rta/ORF50 overexpression [38] . Their low frequency notwithstanding , in addition to the elevated ORF50 transcript levels we also observed an approximately threefold increase of the number of spontaneously reactivated cells in JMJD3-transduced SLK-5dpi cultures ( Figure 9E ) . Taken together , the above data suggest an important role for the H3K27-me3 histone modification during latent infection with KSHV . While our analysis has revealed highly distinct patterns of DNA and histone modifications , the question remains which factors determine such patterns . The molecular requirements for the recruitment of PRC2/EZH2 methyltransferase complexes in mammals are poorly understood , and so far no simple sequence motifs recognized by these complexes have been described . The results from our early infected SLK cultures would seem to indicate that PRC2/EZH2 complexes are recruited to KSHV genomes in a more global fashion . However , it is also possible that the modification is initially established at a small number of loci and rapidly spreads to neighboring regions during the earliest stages of a latent infection . Whatever the mode of deposition , loci with lower levels of H3K27-me3 are ultimately confined to those regions which carry activating H3K9/K14-ac and H3K4-me3 marks which , clearly , are present not only on latency promoters . Two independent studies have noted the presence of such marks in latent cultures before , using ChIP in conjunction with either PCR for a number of select loci , or a global promoter microarray [24] , [41] . Overall , the data from our high-resolution tiling arrays are in very good agreement with the findings reported in both studies . However , while Ellison and colleagues hypothesized that the detection of these marks may have been due to a low percentage of cells which undergo spontaneous reactivation , our study clearly shows that they are a hallmark of latent episomes: First and foremost , the patterns were not only detected in BCBL1 cells , but also in SLKp cultures , which are strictly latent and do not harbor any spontaneously reactivated cells . Second , for a low percentage of reactivated cells to leave a prominent footprint in the histone modification profile of the total population , one has to assume that they contain a disproportional high number of episomes which carry the lytic marks . Given the high copy number of de novo replicated genomes in reactivated cells , this seems a reasonable assumption ( provided that the lytically replicated genomes inherit the parental modification patterns , and that histones are removed only immediately before packaging ) . However , this does not apply to DNA methylation ( which is absent from replicated virion DNA ) . The fact that the global CpG methylation patterns observed in our study show a marked negative correlation with the activating histone marks thus strongly argues for their presence on latent episomes . So what signals trigger the initial recruitment of activating marks at the earliest timepoints of infection ? While this is , ultimately , an issue which will have to be resolved in future studies , when comparing our data with those of two recent studies which have performed genome-wide screens for Rta binding sites [24] , [48] , we noticed that a surprising number of sites mapped within or very close to the loci which were found to be enriched for activating histone marks during our investigation ( see Figure S5 ) . Interestingly , Rta is known to be expressed for a brief period of time within the first few hours of a de novo infection , before latency ensues [49] . Therefore , one attractive hypothesis is that binding of Rta may trigger the initial modification of H3K9/K14 and H3K4 at these sites . The onset of widespread H3K27 methylation then may lead to the silencing of the ORF50 promoter , establishing a poised state of repression which can be easily reverted upon reactivation . However , while this model may explain the initial deposition of activating marks , it does not provide a satisfactory explanation for their maintenance during the later stages of viral latency . So far , there is very little evidence for the propagation of activating histone marks through cell divisions; rather , it is thought that their preservation requires continuous transcriptional initiation . In contrast to DNA methylation or polycomb-associated H3K27-me3 marks , H3K9/K14-ac and H3K4-me3 modifications are therefore not considered inheritable ( and , therefore in a strict sense also do not represent epigenetic modifications ) . Thus , even if Rta is indeed responsible for the initial establishment of H3K27-me3 and H3K9/K14-ac marks , due to its rapid eradication upon establishment of latent expression patterns it cannot be responsible for their long-term maintenance . One possible explanation is that these loci represent preferred binding sites not only for Rta , but also for constitutively expressed host transcription factors . In this scenario , host factors could sustain the poised state of repression at H3K27-me3 enriched promoters , but additional stimuli would be required to return them to an active state . There are , however , also a few loci which are rich in H3K9/K14-ac and H3K4-me3 marks , but display very little H3K27 tri-methylation or DNA methylation . These include not only the constitutively active latency promoter upstream of ORF73 as well as the locus encoding the KSHV miRNA-cluster , but also ORFs K5/K6/K7/nut-1 , the region upstream of three of the four vIRFs ( vIRF1/K9 , vIRF-3 and vIRF-4 ) and the complete K15 gene region at the right end of the viral genome . vIRF-3 ( also termed latency-associated antigen 2 , LANA2 ) is known to be expressed in latent PEL cells [50] and K15 ( which encodes the latency associated membrane protein/LAMP ) has been originally identified as a latently expressed gene , although its expression is significantly upregulated during the lytic cycle [51] . Since the region immediately upstream of the K1 gene at the left side of the KSHV genome as well as the terminal repeats ( shown only to the right of the map in Figures 2 , 5 and 6; as the episome is circular they however also flank the left terminus of the genome ) are also highly enriched in activating marks but display very little H3K27-me3 , our data also support a previous report of K1 expression in latently infected cells [52] . However , as K1 transcription is strongly upregulated in lytic cells and LANA has been found to repress K1 gene expression [53] , whether K1 is indeed expressed at significant levels during latency is currently unclear . Interestingly , the region upstream of the K2 gene also displayed a high ratio of activating vs . repressive marks in BCBL1 cells ( which is less pronounced in SLKp or SLK-5dpi cells ) . K2 encodes v-IL6 , a viral homologue of IL6 which supports B cell growth and blocks interferon responses [54] , [55] . While v-IL6 has been reported to be expressed in latently infected PEL cells [54] , [56] , similar to K1 its expression is strongly upregulated by lytic cycle induction and it thus remains controversial whether it represents a latent gene . Although latent transcripts of unknown promoter origin have also been identified in the broader region encompassing ORFs K4 to K7 [57] , most of the remaining genes so far have not emerged as being latently expressed in experimental systems [58] . If continuous transcriptional initiation is required to maintain H3K9/K14-ac and H3K4-me3 marks , then additional factors may exist which stall RNA polymerase II at these loci . Alternatively , it is possible that above genes are transcribed during latency only at low level , or that their mRNAs are rapidly turned over such that they do not accumulate . Transactivation by Rta as well as transcript stabilization by other lytic gene products ( e . g . the product of ORF57 ) may then allow efficient expression of these genes once the viral genome is committed to productive replication . Taken together , our data thus provide a rationale for the observation that some of the above genes have been found to be expressed at low level in latent cultures . Importantly , they may also help to understand how the host environment may modulate the latent gene expression program . For example , IFN-α treatment of PEL cells has been shown to result in the transactivation of the K2 promoter via IFN-stimulated response element ( ISRE ) sequences [55] . It has been suggested that this enables the virus to sense innate immune responses and modify its gene expression in order to block them , a model which is strongly supported by the observation that the K2 promoter appears to be already primed for expression in latently infected PEL cells . Finally , there is also the question of the role of the profound DNA methylation which occurs at later stages of viral latency . It appears likely that these patterns are established as a consequence of the continuous presence of EZH2/PRC2 repressor complexes on viral DNA , as such complexes have been shown to directly recruit DNA methyltransferases ( DNMTs ) [59] . The absence of DNA methylation at loci which are devoid of H3K27-me3 would support this conclusion . Another contributing factor could be the delayed appearance of constitutive heterochromatin marks: While H3K9-me3 is restricted to a few regions , it may nevertheless support the recruitment of DNMTs to the viral genome ( this may be especially the case at the ORF64 locus , which displays the highest levels of H3K9-me3 but only moderate levels of H3K27-me3 ) . What are the functional consequences of DNA methylation ? Currently , this is a question that is difficult to answer . Based on the fact that SLK cells establish latent expression patterns in the absence of DNA methylation , and that BCBL1 and AP3 cells maintain latency in spite of the lack of DNA methylation at the ORF50 promoter , one may be tempted to think that this epigenetic mark is of no fundamental importance during KSHV latency . However , this is a conclusion which cannot be drawn . Compared with SLKp cells , de novo infected SLK-5dpi cells indeed display elevated levels of lytic gene expression and a higher number of spontaneously reactivated cells ( Figs . 5 and 9 ) . However , the generally low transcript levels together with the scarcity of lytic cells even in SLK-5dpi cultures complicate any general conclusion . Likewise , as the cells do not respond to chemical treatment with reagents that induce the lytic cycle in PEL cells , comparative studies of reactivation are difficult to perform . Studies employing Rta overexpression ( which can reactivate such cells [38] ) may be feasible , but they would be of limited value as the ectopic expression would artificially override one of the most critical steps of lytic reactivation . With regard to PEL cells , more lines will have to be studied to conclude whether absence or presence of methylation marks at the ORF50 promoter has an impact on the percentage of spontaneously reactivated cells and/or the response of such lines to chemical agents which induce the lytic cycle . At present , although such differences certainly exist between many PEL lines , this could in large part or entirely be a consequence of host cell differences . Lastly , even if DNA methylation should turn out to have no significant additive effect over the presence of repressive histone marks in vitro , this may be fundamentally different in vivo . Although the physiological triggers which reactivate in vivo latency reservoirs ( e . g . memory B cells ) are poorly understood , they are very likely to be much more specific than the broad pleiotropic effects induced by chemical agents such as phorbol esters or sodium butyrate . It is thus very conceivable that DNA methylation may represent an additional , functionally important block which augments repressive histone marks and re-inforces latent expression patterns during long-term latency in vivo , but which may be of lesser consequence in in vitro models of viral infection . Considering all of the above , many questions remain to be answered before the molecular mechanisms which govern establishment and maintenance of KSHV latency are fully understood . However , especially given the unexpected spatial and temporal patterns of histone modifications and DNA methylation revealed by our study , the data presented here provide important clues as to the host and viral factors which might be at work , and should greatly help to design further studies aimed at elucidating the role of epigenetic modifications during this crucial phase of the viral lifecycle .
The establishment of SLKp cells has been described before [39] . Briefly , endothelial SLK cells [60] were infected with KSHV in vitro and passaged for several weeks . Seven KSHV-positive single cell clones were selected from the long-term infected cultures and pooled to form the SLKp line . SLKp cells and the parental SLK line were cultured in DMEM supplemented with 10% fetal calf serum and penicillin-streptomycin ( 5 µg/ml ) . The KSHV-positive PEL cell lines BCBL1 [6] , HBL6 [61] and AP3 [62] were cultured in RPMI 1640 medium ( Invitrogen ) supplemented with 10% fetal calf serum and penicillin-streptomycin at a final concentration of 5 µg/ml . Concentrated supernatants of infectious KSHV virions were harvested from lytically induced BCBL1 cells as described [39] . De novo infection of SLK cells was performed by incubating 2×105 cells at 70% confluency for 2 hrs with 500 µl virus supernatant at a concentration of 1×108 KSHV genome equivalents per ml ( as determined by quantitative PCR ) in the presence of 8 µg/ml polybrene in EGM-2 medium ( Lonza ) . Generally , more than 95% of cells were infected , as judged by immunofluorescence analysis for LANA 48h after infection . For lytic reactivation of BCBL1 cells , sodium butyrate was added to the culture medium at a final concentration of 0 . 3 mM . Cells were fixated with 4% paraformaldehyde in PBS for 15 min , permeabilized with 2% Triton X-100 in PBS for 10 min , blocked with 3% BSA in PBS and incubated with primary antibodies specific for LANA or ORF59 ( Advanced Biotechnologies: #13-211-100 ) in blocking solution for 2 hrs . Cells were washed three times with PBS and incubated with secondary antibodies ( Alexa Fluor-555 goat anti mouse and −488 goat anti rabbit ) for another 2 hrs and analyzed by fluorescence microscopy . Western blot analysis of total cell lysates was carried out by standard SDS-PAGE and immunoblot protocols , using antibodies directed against histone tri-methylated at lysine 27 ( Upstate: #07-449 ) or , as a loading control , actin ( Santa Cruz: #SC-8432 ) . A retroviral JMJD3 expression construct was kindly provided by Paul Khavari [47] . The retroviral backbone MSCV ( Clontech ) was used as a negative control . Supernatants containing infectious viral particles were harvested 48 hrs post transfection of PhoenixGP cells ( Nolan Laboratory , http://www . stanford . edu/group/nolan/ ) . BCBL1 and SLK cells were transduced with recombinant retroviruses by spin inoculation at 300×g for 1 h , using undiluted supernatants in the presence of 8 µg/ml polybrene . After inoculation , cultures were maintained in medium containing 2 µg/ml puromycin for 12 days to select for transduced cells . Bisulfite sequencing was performed using the EpiTect Bisulfite Kit ( Qiagen ) , following the manufacturer's instructions . The method relies on a chemical reaction that leads to the conversion of all unmethylated cytosine residues to thymidines , allowing the identification of originally methylated cytosines after PCR amplification and sequencing of the locus of interest . The sequences of all bisulfite sequencing primers employed in this study are given in Table S1 . PCR products were sequenced directly ( bulk sequencing ) using either the forward or reverse primer from the original amplification . CpG methylation patterns were extracted from the bulk sequencing data using the BiQ Analyzer v2 . 0 software [63] . A combined bisulfite restriction analysis , short COBRA assay , has been described before [64] . Briefly , PCR products from bisulfite treated samples were digested with the restriction enzyme TaqI ( Fermentas ) and resolved on an agarose gel ( 3% ) . TaqI recognizes the nucleotide sequence TCGA , which contains a CpG dinucleotide . After bisulfite conversion , the site is only preserved if the original CpG motif was methylated ( note that the bisulfite conversion creates additional TaqI sites at methylated CpG motifs which are flanked by C and A residues , as the C in position −1 is converted to a T by the bisulfite reaction ) . RNA was isolated using the RNA-Bee ( Tel-Test , Inc . ) reagent . Contaminating DNA was removed by incubation with amplification grade DNase I ( Invitrogen ) and cDNA was prepared from random-primed RNA using Superscript III ( Invitrogen ) as per the manufacturer's instructions . Real-time quantitative PCR ( qPCR ) of cDNA or genomic DNA samples was performed using SensiMix SYBR Kit ( Quantace ) on a Rotorgene 6000 light cycler ( Corbett Life Science ) . For quantitation , standard curves were created using dilutions of genomic BCBL1 DNA over a range of at least 10000× . The sequences of all primer pairs used in this study are given in Table S1 . ChIP analysis was performed as described by Si et al . [65] and recommended by the array manufacturer ( Agilent Mammalian ChIP-on-chip protocol V10 . 0 , May 2008 ) , with some modifications . A detailed protocol of the procedure is given in Protocol S1 . Briefly , chromatin from 5×106 to 2×107 cells was cross-linked with 1% formaldehyde . After quenching of the reaction by the addition of glycine , cells were lysed and nuclei were isolated by centrifugation . Chromatin was extracted from the isolated nuclei and fragmented by sonication using a Bioruptor ( Diagenode ) to an average length of 100–500 bp . A portion of the total chromatin sample was set aside for the later preparation of input controls . Material from 1×106 cells was pre-cleared with salmon-sperm DNA protein-A agarose beads ( Upstate ) to reduce non-specific background and subjected to immunoprecipitation using 2 to 10 µg of antibodies specific for the histone modifications H3K9/K14-Ac ( Upstate: #06-599 ) , H3K4-me3 ( Upstate: #04-745 ) , H3K9-me3 ( Upstate: #17-625 ) or H3K27-me3 ( Upstate: #07-449 ) . After incubation for 16 hrs , chromatin-immunocomplexes were precipitated by the addition of protein-A agarose , washed , eluted and de-crosslinked overnight at 65°C . DNA was purified by phenol-chloroform extraction and ethanol precipitation . For preparation of input controls , 1/4th of the amount of chromatin used in the immunoprecipitation reactions was employed . Input samples were treated in an identical manner as the immunoprecipitated samples , starting with the de-crosslinking step . Both samples were subsequently subjected to whole genome amplification and labeling using a linker mediated PCR protocol ( Agilent Mammalian ChIP-on-chip protocol V10 . 0 , May 2008 ) , followed by microarray hybridization . Colocalization of bivalent histone marks was measured by use of a sequential ChIP assay . Prior to the first IP antibodies were incubated with protein-A agarose beads ( Upstate ) for 2 hrs at 4°C . Antibody bead complexes were washed twice with 0 . 2 M triethanolamine buffer ( Sigma ) . Beads and antibodies were coupled covalently by incubation with 20 mM dimethyl pimelimidate dihydrochloride ( DMP , Sigma ) in 0 . 2 M triethanolamine buffer on a rotating wheel for 30 min at RT , and the reaction was stopped by washing with 50 mM Tris-HCl ( pH 7 . 5 ) . Uncoupled antibodies were removed by pre-elution with 0 . 1 M acetic acid ( pH 3 . 0 ) for 5 min at RT . Beads were incubated with diluted chromatin samples for 16 hrs at 4°C . Washing and elution was performed in a identical manner as described above for the standard ChIP assay . 1/16th of the precipitated chromatin was de-crosslinked and was used to determine the efficiency of the first IP by real-time qPCR . The remainder was employed as the input for the second IP , which again was performed according to the standard ChIP protocol . Results were calculated as percent of the original input , i . e . the total amount of DNA which was subjected to the first round of immunoprecipitation . MeDIP analysis was essentially performed as described before [35] , [66] , [67] , [68] , with some modifications . A detailed protocol is given in Protocol S2 . Briefly , highly pure genomic DNA served as an input in the MeDIP procedure . Negative and positive controls were prepared by mixing genomic DNA from KSHV-negative SLK cells with unmethylated or in vitro methylated KSHV bacmid DNA [36] , respectively . The ratio of viral vs . cellular DNA was selected such that it mimics the episome content typically seen in KSHV-infected PEL cell lines and the SLKp line ( approx . 30–40 copies per cell ) . All DNA samples were sonicated to an average fragment size of 100–500 bp using a Bioruptor ( Diagenode ) . In order to allow quantification and normalization of the data , a constant amount ( 0 . 2 ng ) of in vitro methylated pCR2 . 1 plasmid was added per 5 µg of the sheared DNA . 1 µg of the sample was set aside as an input control , and the remainder was subjected to immunoprecipitation using 2 . 5 µg of a 5′-methylcytidine specific antibody ( MAb-5MECYT-100 , Diagenode ) . The precipitated immunocomplexes were harvested using Dynabeads M-280 Sheep anti-Mouse IgG ( Invitrogen ) . After washing , DNA was eluted and purified by phenol-chloroform extraction and ethanol precipitation . Input control samples were treated identical to the IP samples , starting with the ethanol precipitation step . The samples were subsequently analyzed by qPCR and/or microarray hybridization . For control and normalization purposes , we prepared in vitro methylated DNA from a bacmid containing the complete KSHV genome [36] or the pCR2 . 1 vector ( Invitrogen ) . DNA was methylated by incubating 15 µg of DNA with 40 units of the CpG methyltransferase M . SssI ( NEB ) for 2 hrs in 1× NEBuffer 2 containing 160 µM S-adenosylmethionine ( SAM ) . Fresh SAM was added and reactions were incubated for another 2 hrs . DNA was purified and the reaction was repeated once to ensure complete methylation . Complete methylation was confirmed by restriction analysis using methylation sensitive enzymes ( HpaII and MspI , Fermentas ) , and/or bisulfite sequencing of specific loci . Custom high-resolution KSHV microarrays were designed by shifting a sequence window of 60 nt . across both strands of the prototypic KSHV sequence ( type P , accession number NC_009333 ) as well as the terminal repeat unit ( KSU86666 ) . Probes with a length between 45 and 60 nucleotides were selected from these windows such that their melting temperature was close to the optimal Tm of 80°C . To also ensure complete coverage of type M KSHV strains , the resulting probe sets were aligned to the type M reference sequence ( NC_003409 ) and additional probes were designed in a identical manner for all regions with a length of 80 or more nucleotides which were not already covered by the original probe set . The length of all probes was subsequently adjusted to 60 nucleotides using sequences from a common linker ( ATAACCGACGCCTAA ) , and each probe was synthesized in duplicate on Agilent 8×15k custom microarrays . For normalization purposes , the array also contains probe sets which were generated in an identical manner to cover the adenovirus type 5 genome ( AY339865 ) as well as the pCR2 . 1 plasmid ( Invitrogen ) . 500 ng of MeDIP-input or ChIP-input controls , 500 ng of immunoprecipitated ChIP material and all of the MeDIP material were labeled with Cy3 and Cy5 using Agilent Genomic DNA Labeling Kit PLUS according to Agilent's recommendations . For normalization purposes , 0 . 1 ng of Adenovirus Type 5 DNA were added to each samples prior to the labeling procedure . After labeling , samples were purified using Microcon YM-30 filter columns ( Milipore ) , blocked using Agilent blocking solution and human cot-1 DNA ( Invitrogen ) , and hybridized using the Agilent Oligo aCGH/ChIP-on-chip Hybridization Kit at 65°C for 24 hrs in a rotating oven . Arrays were washed once with Oligo aCGH/ChIP-on-chip Wash Buffer 1 ( Agilent 5188–5221 ) at RT for 5 min and in Oligo aCGH/ChIP-on-chip Wash Buffer 2 ( Agilent 5188–5222 ) at 37°C for 1 min and scanned using a GenePix Personal 4100A scanner ( Axon Instruments ) . Primary array analysis and data normalization was carried out using GenePix Pro 6 . 0 software ( Axon Instruments ) . All MeDIP datasets were normalized using the methylated pCR2 . 1 plasmid which had been added to the samples prior to the immunoprecipitation , thus controlling for differences in MeDIP efficiency as well as labeling and array hybridization . Both channels were adjusted such that the average ratio of input vs . MeDIP signals across all pCR2 . 1-specific spots was 1 . Similarly , ChIP datasets were normalized using the adenovirus type 5 DNA that was added as a spike-in prior to labeling , hence correcting for errors during labeling , hybridization or scanning of the samples . To eliminate false positive spots , we hybridized DNA from KSHV-negative SLK cells and identified all probes which exhibited high levels of background hybridization ( i . e . , fluorescence levels that exceeded the mean value plus 1× the standard deviation of all KSHV-specific spots on the negative control array ) . These probes ( which mapped almost exclusively to repeat regions ) were permanently flagged in all datasets and not used for further analysis . While our arrays carry probes specific for the M and P types of the KSHV genome , the KSHV genomes from the BCBL1 and AP3 lines have not been fully sequenced and thus may deviate from the reference genomes at a few locations . To control for such sequence differences , we flagged all spots which exhibited fluorescence levels which did not exceed a background fluorescence threshold in the input channel , which was set to the mean fluorescence plus twice the standard deviation of all negative control features ( i . e . empty array features as well as spots containing irrelevant sequences , corresponding to all Agilent probes in the datasets which are labeled with “NC2_” and “ ( - ) 3xSLv1” ) . Note that , if sequence diversification leads to only a reduction of hybridization efficiency ( e . g . due to single nucleotide polymorphism , which will not abolish hybridization ) , this will not falsify our results as the hybridization efficiency will be reduced in input as well as the immunoprecipitated sample; the ratio will thus be unaffected . In addition to above quality controls , in each dataset we flagged all probes which exhibited more than 30% variance between duplicate spots . The 30% threshold corresponds to the mean variance plus twice the standard deviation exhibited by all KSHV-specific probes in all MeDIP experiments , thus removing all probes which show a significantly increased variance between individual spot repeats . MeDIP data were furthermore corrected by subtracting from each probe-specific signal the value observed in the negative control , i . e . the MeDIP sample representing the unmethylated KSHV bacmid in the background of cellular DNA . After normalization , an enrichment score was calculated for each of the probes , represented by the ratio of fluorescence signal intensities in the immunoprecipitated samples relative to the input control . As the average length of the immunoprecipitated MeDIP and ChIP fragments ( 100 to 500 bp ) is greater than that of the tiled probes ( 45 to 60 nucleotides ) , the resolution of our analysis was limited by the fragment length rather than the array design . To account for this fact , the data presented in Figures 2 to 6 , 7 and 8 were calculated by tiling overlapping sequence windows of 250 nucleotides across the KSHV genome , using a step size of 100 nucleotides to advance each window . The type M reference sequence ( NC_003409 ) was used for the HBL6 line , whereas the type P genome ( NC_009333 ) was used for all other cells . The KSHV specific probes were subsequently blasted against the window sequences , and each window was awarded an enrichment score represented by the average score of all probes which showed more than 90% identity with either strand of its sequence . All scores ( which were also used to calculate the Pearson correlation coefficients presented in Tables 1 and 2 ) are given in the Dataset S1 . All raw data , including the original GPR files as well as sequence and match location ( s ) of individual probes are available from the Gene Expression Omnibus ( GEO ) Database at http://www . ncbi . nlm . nih . gov/geo , under accession number GSE19907 . | A characteristic feature of herpesviruses is their ability to establish a latent infection during which most of the viral genes are silenced . As a consequence , no viral progeny is produced and the host cell remains viable . While the viral genome may persist in the nucleus of such cells indefinitely , it retains the ability to re-enter the lytic cycle and produce new virions if conditions in the cell become unfavorable . The molecular requirements for the establishment of latency are poorly understood , but are thought to depend on epigenetic modifications of the viral episome . Here , we report a genome-wide screen to investigate DNA methylation and histone modification patterns associated with latent infection by Kaposi Sarcoma-associated herpesvirus ( KSHV ) , a tumor virus linked to the development of several cancers . We find that latency is likely to be determined by modifications commonly associated with genes that are transcriptionally “poised” . The promoters of such genes harbor activating as well as repressive histone marks such that they are silenced , but they can be rapidly activated upon removal of the repressive marks . Our findings thus may explain how KSHV achieves efficient quiescence during latency , yet retains the potential to quickly revert to a fully active state upon induction of the lytic cycle . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"virology/persistence",
"and",
"latency",
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"genetics",
"and",
"genomics/gene",
"expression",
"virology/viruses",
"and",
"cancer",
"genetics",
"and",
"genomics/epigenetics"
]
| 2010 | The Epigenetic Landscape of Latent Kaposi Sarcoma-Associated Herpesvirus Genomes |
Thiamine deficiency is thought to be an issue in Cambodia and throughout Southeast Asia due to frequent clinical reports of infantile beriberi . However the extent of this public health issue is currently unknown due to a lack of population-representative data . Therefore we assessed the thiamine status ( measured as erythrocyte thiamine diphosphate concentrations; eThDP ) among a representative sample of Cambodian women of childbearing age ( 15–49 y ) and their young children ( 6–69 mo ) . Samples for this cross-sectional analysis were collected as part of a national micronutrient survey linked to the Cambodian Demographic and Health Survey ( CDHS ) 2014 . One-sixth of households taking part in the CDHS were randomly selected and re-visited for additional blood sampling for eThDP analysis ( 719 women and 761 children ) . Thiamine status was assessed using different cut-offs from literature . Women were mean ( SD ) 30 ( 6 ) y , and children ( 46% girls ) were 41 ( 17 ) mo . Women had lower mean ( 95% CI ) eThDP of 150 nmol/L ( 146–153 ) compared to children , 174 nmol/L ( 171–179; P < 0 . 001 ) . Using the most conservative cut-off of eThDP < 120 nmol/L , 27% of mothers and 15% of children were thiamine deficient , however prevalence rates of deficiency were as high as 78% for mothers and 58% for children using a cut-off of < 180 nmol/L . Thiamine deficiency was especially prevalent among infants aged 6–12 mo: 38% were deficient using the most conservative cut-off ( < 120 nmol/L ) . There is a lack of consensus on thiamine status cut-offs; more research is required to set clinically meaningful cut-offs . Despite this , there is strong evidence of suboptimal thiamine status among Cambodian mothers and their children , with infants <12 mo at the highest risk . Based on eThDP from this nationally-representative sample , immediate action is required to address thiamine deficiency in Cambodia , and likely throughout Southeast Asia .
Beriberi is a ‘forgotten disease’ [1–4] that remains a public health issue in Southeast Asia despite near eradication elsewhere [5–7] . Beriberi is caused by thiamine ( vitamin B1 ) deficiency , and is most serious in infants due to the rapid growth and development that occurs during this time , and the relatively high thiamine needs compared to body size [4 , 8] . Breast milk thiamine concentrations reflect maternal dietary thiamine intake [9] . As such , poor maternal thiamine status during pregnancy and lactation puts infants at risk of developing beriberi [9–11] , which can lead to death in hours of clinical presentation if not recognized or left untreated [6] . While infantile beriberi is the most serious outcome of thiamine deficiency , marginal thiamine status in the wider population causes fatigue , apathy , anorexia , and dizziness [12] , and with these the potential for decreased school performance and/or economic output . In addition , Israeli children who consumed thiamine-deficient infant formula in infancy , but did not develop beriberi , exhibited retarded neurological , cognitive , and cardiological development at age 5–7 y [13] , highlighting the importance of thiamine sufficiency in early life . Beriberi remains a problem in Southeast Asia , in part , because non-parboiled [14] , unfortified white rice is the dietary staple [8 , 12] . In Cambodia , white rice makes up an estimated 60% of daily dietary energy [15] . Although rice contains thiamine [8] , it is found only in the outer husk and bran , the vast majority of which is removed during the milling process [14] . In most rice-consuming cultures , polished white rice is preferred [14 , 16] for several reasons: organoleptic qualities , white rice as a status symbol [12] , and because removal of the lipid-rich outer bran increases shelf-life [14] . In Cambodia brown rice is a cultural dietary taboo; people were forced to eat brown rice during the Khmer Rouge regime [17] . Although there are several reports of infantile beriberi in Cambodia [5 , 18 , 19] , there is a lack of accurate prevalence data . The World Health Organization has suggested that in the absence of reliable information on the prevalence of beriberi or on biochemical markers of thiamine status , infant mortality curves could be indicative of thiamine deficiency being a health concern , with a peak in infant mortality around 3–4 months of age being suggestive of a high prevalence of beriberi [20] . We analyzed infant mortality data from the Cambodian Demographic Health Surveys ( CDHS ) 2000 , 2005 , 2010 and 2014 and found indeed a peak in mortality around 3 months of age ( Fig 1 ) . Thiamine status has traditionally been assessed using a functional indicator , erythrocyte transketolase activity coefficient [21 , 22] , however , this method has several shortcomings including the inactivation of transketolase during sample processing and storage , poor inter-assay precision [23] , and a tendency of this assay to underreport deficiency among chronically deficient individuals [24] . More recently , the biochemical assessment of the biologically active form of vitamin B1 , thiamine diphosphate , in whole blood or in erythrocytes has been advocated [2] . Erythrocyte thiamine diphosphate concentrations ( eThDP ) measured by high performance liquid chromatography ( HPLC ) overcomes several downfalls of the functional assay , and correlates well with erythrocyte transketolase activity coefficient [23] . Coats and colleagues reported that Cambodian mother-infant dyads in Prey Veng province had significantly lower whole blood thiamine diphosphate concentrations , regardless of infant clinical beriberi diagnosis , compared to American controls [5] . Lower eThDP were reported in a representative survey of Cambodian women of childbearing age residing in urban Phnom Penh and rural Prey Veng provinces compared to a small convenience sample of purportedly thiamine-replete Canadian women from Vancouver [25 , 26] . Unfortunately there is currently a lack of consensus on the most appropriate cut-offs to define suboptimal status or deficiency using whole blood ThDP or eThDP , and there are no nationally representative data available on biochemical thiamine status of any population group in Southeast Asia . Therefore , the objective of this study was to determine eThDP among women of childbearing age and their children aged 6–69 mo who participated in the most recent 2014 Cambodian Demographic and Health Survey ( CDHS ) [27] and the linked National Micronutrient Survey [28] to determine the prevalence of thiamine deficiency in this population using various cut-offs .
This biochemical thiamine analysis was part of the 2014 Cambodian National Micronutrient Survey [28] ( conducted June 2 –December 12 , 2014 ) , which was linked to the CDHS , a nationally-representative survey of adults aged 15–49 y and children from 24 Cambodian provinces [27] . Population proportionate to size sampling was used to select 611 villages from which 16 , 356 individual households were selected . Trained , Khmer-speaking enumerators visited all selected households and collected information on health outcomes including nutrition , fertility and family planning , morbidity and mortality , housing , and assets and wealth using a validated , standardized questionnaire [27] . One week to 2 months after the CDHS had visited the household , one sixth of the households were re-visited and biological samples were collected from mothers and their children . Full sampling and survey details can be found elsewhere [27 , 28] . With one sixth of the households re-visited , it was estimated that 935 mothers and children could be included in the micronutrient survey , but due to absence of mother or care-takers and refusal to participate , blood samples were collected from only 726 women and 781 children , and eThDP was measured in 726 and 761 samples , respectively . Pregnant women ( n = 7 ) were excluded , leaving 719 maternal samples for analysis . Hemoglobin concentrations were measured by the CDHS survey , and not repeated during the micronutrient survey . As hemoglobin was measured in only half of the children and women , data on hemoglobin and anemia prevalence is only available for 441 women and 476 children . The National Ethics Committee for Health Research , Phnom Penh , Cambodia granted ethical approval ( 057 NECHR 2014 ) for the 2014 Cambodian National Micronutrient Survey . Eligibility for participation included: having participated in the CDHS and have given permission to the CDHS team to be re-visited for the micronutrient survey , a child in the household aged 6–59 mo , neither mother nor child having evidence of severe or chronic illness , and mothers or care-taker providing written , informed consent . All data was anonymized . Nurses collected non-fasting , morning-time blood samples into trace element-free , heparin-coated tubes ( Vacuette , Greiner Bio One , Austria ) at a central village location . Samples were stored in a dark cooler box and transported to the nearest Provincial Health Centre within 6 h of collection . Samples were centrifuged ( 3000 g for 10 min ) , plasma and buffy coat were thoroughly removed , and erythrocytes were separated into 500 μL aliquots and frozen at -20°C . Frozen samples were then transported to the Department of Fisheries Post-Harvest Technologies and Quality Control Laboratory , Fisheries Administration in Phnom Penh for storage at -20°C . Once the survey was completed the samples were batch shipped on dry ice to Abbot Laboratories in Singapore for eThDP analysis . After arrival of the blood samples in Abbott Nutrition R&D Singapore Center , the samples were stored at -78°C . eThDP was measured using a modified method of Lu & Frank [29] , as reported elsewhere [30] . Briefly , samples were thawed on ice in a dark room with amber light . Trichloroacetic acid solution was added to precipitate protein out . After centrifugation , the supernatant was collected , washed with methyl tert-butyl ether , and subject to ultra-high performance liquid chromatography ( Agilent model 1290 system , Singapore ) with a fluorescence detector ( Agilent model G1321A , Agilent Technologies , Singapore ) and an autosampler ( Agilent model G4226A , Agilent Technologies , Singapore ) that allowed online pre-column derivatization with potassium ferricyanide . At least ten different thiamine status cut-offs are reported in literature for use in women and children [5 , 21 , 22 , 31–40] . Noteworthy are the Institute of Medicine ( IOM ) definitions of thiamine deficiency ( eThDP <70 nmol/L ) and marginal deficiency ( 70–90 nmol/L ) [22] , which are based on values from 68 healthy Dutch blood donors and laboratory staff aged 20–50 y [31] . In the original citation these are described as cut-offs for whole blood or red cells [41] , causing confusion over the correct biological sample for their use . If the IOM cut-offs represent whole blood ThDP , then values should be corrected for hematocrit to obtain eThDP values [34] . For example , the Coats et al . noted that their lab uses a reference range of 80–150 nmol/L for whole blood ThDP , or , if divided by hematocrit , 150–290 nmol/L for eThDP [5] . The Institute of Medicine cut-offs have not been employed here due to confusion surrounding use for whole blood versus erythrocyte ThDP . We have made a distinction between cut-offs being reported as falling below a ‘reference range’ , or ‘deficient’ or ‘marginally deficient’ , as it is unclear whether a value outside a reference range represents real deficiency . For example , a thiamine deficiency cut-off of eThDP <180 nmol/L was proposed by Mancinelli and colleagues to align with the 25th percentile eThDP of 103 healthy controls ( 45 men and 58 women , employees of University “La Sapienza” Hospital ) in Rome , however none of these subjects were thiamine deficient [38] . In best practice , an average of 120 subjects are needed to generate accurate reference limits for a given biomarker [42]; this has not been the case for the majority of thiamine cut-offs . We present four cut-off values describing suboptimal status below a reference range: the abovementioned eThDP <180 nmol/L [38]; eThDP <165 nmol/L , the lower bound of a 95% reference range ( 165–286 nmol/L ) of 48 ( 25 men and 23 women ) healthy hospital staff at Broadgreen Hospital , Liverpool , UK [32]; <150 nmol/L , corresponding to the eThDP reference range of the Mayo Medical Laboratories [5]; and <140 nmol/L , the lower limit of normal eThDP ( cut-off of lowest 2 . 5% ) of healthy blood donors in Christchurch , New Zealand; n unknown [33] . eThDP <148 nmol/L was used to categorize low thiamine status in two studies [35 , 36] , and originated as the lower bound of normal range ( 50–150 ng/mL packed cells ) from 21 healthy adults in Nashville , Tennessee [37] . Since the values of 148 and 150 nmol/L are close , we have used only <150 nmol/L as a cut-off in the current paper . Marginal thiamine deficiency has been described using one cut-off , eThDP between 120–150 nmol/L , a cut-off reported in [21 , 39] , but no details of these values are known . Two cut-offs for thiamine deficiency have been reported: eThDP <120 nmol/L , again that was used in [21 , 39] , but the origins of this cut-off are unknown; and <118 . 5 nmol/L , which was used to categorize thiamine deficiency in [35] , and is described as below the 95 percentile reference range ( 40–85 μg/L ) among healthy black South African adults in [40] . As these values are close again , we have used <120 nmol/L as cut-off for thiamine deficiency . Demographic characteristics were computed as mean ( SD ) or n ( % ) , and eThDP as mean ( 95% CI ) . Children’s eThDP and thiamine status are categorized by age category , 6–12 mo , 13–24 mo , 25–36 mo , 37–59 mo , and ≥ 60 mo . A t-test was performed to compare women and children’s eThDP , and eThDP among residents in rural and urban areas; a one-way ANOVA was employed to compare eThDP among different wealth quintiles , and children’s age categories ( with least significant difference post-hoc correction for multiple comparisons ) . Linear regression models were built to measure the association between eThDP and various independent variables . Variables were included in the linear regression model if P < 0 . 20 in bivariate correlation , and were entered stepwise into separate models for women and children . The following variables were evaluated for inclusion in the model for both women and children: province , wealth quintile , population density ( urban/rural ) , mother’s education , cigarette smoke in home , subsidized healthcare available for household , weight , height/length , age , and hemoglobin concentration . The model for women also included BMI , and the children’s model also included sex and birth order . Results were considered significant at P < 0 . 05 . Weight-for-age , height-for-age , BMI-for-age , and weight-for-height z-scores were calculated using WHO Anthro and WHO Anthro Plus software programs , otherwise all analyses were performed using SPSS for Macintosh version 23 . 0 ( IBM , Armonk , NY , USA ) .
Demographic characteristics are shown in Table 1 . Women were 30 ( 6 ) y , and the majority had a normal BMI ( 70%; 18 . 5–24 . 99 kg/m2 ) , were married ( 93% ) , and had attended some formal schooling ( 93% ) . Children were 41 ( 17 ) mo and 46% were girls; 28% ( n = 130 ) of children were wasted ( weight-for-age z score < -2 SD ) , and 39% ( n = 182 ) were stunted ( height-for-age z score < -2 SD ) . Children had a higher mean ( 95% CI ) eThDP of 174 nmol/L ( 171–179 nmol/L ) compared to women , 150 nmol/L ( 146–153 nmol/L; P < 0 . 001 ) ; Table 2 . eThDP did not differ between children living in rural ( 173 nmol/L , 169–177 nmol/L ) versus urban areas ( 180 nmol/L , 172–189 nmol/L; P = 0 . 14 ) , however , rural women had lower eThDP ( 146 nmol/L , 143–150 nmol/L ) compared to their urban peers ( 164 nmol/L , 157–171 nmol/L; P <0 . 001 ) . Compared to higher wealth quintiles , eThDP was lower among both children ( P = 0 . 04 ) and women ( P < 0 . 001 ) in lower wealth quintiles . Young children aged 6–12 mo had significantly lower eThDP ( 144 nmol/L , 130–159 nmol/L ) compared to older children aged 13–36 mo ( 176 nmol/L , 170–173 nmol/L; P < 0 . 001 ) and > 36 mo ( 177 nmol/L , 172–182 nmol/L; P < 0 . 001 ) ; eThDP in the latter two age groups did not differ ( P = 0 . 90 ) . Table 3 shows the percentage of women and children below selected cut-offs . Using the most conservative cut-off for thiamine deficiency ( eThDP < 120 nmol/L ) , 27% of mothers and 15% of children were thiamine deficient . Worrisome , 38% of infants were thiamine deficient . The following variables were included in the eThDP prediction linear regression models: for children , age ( mo ) , hemoglobin concentration ( g/dL ) , and household wealth quintile; for women , household wealth quintile , household qualification for subsidized health , province of residence , population density ( rural/urban ) , education level attended , and hemoglobin concentration ( g/dL ) were included . Hemoglobin concentration was the only predictor of children’s eThDP ( adjusted R2 = 0 . 024 , standardized β [95% CI] , 0 . 161 [3 . 2–11 . 3 g/dL] , P < 0 . 001 ) . The model for women’s eThDP ( adjusted R2 = 0 . 044 ) included wealth quintile ( standardized β [95% CI] , 0 . 209 [3 . 6–10 . 0] , P < 0 . 001 ) and hemoglobin concentration ( standardized β [95% CI] , 0 . 114 [0 . 6–7 . 9 g/dL] , P = 0 . 02 ) .
Here we present the first nationally representative biochemical thiamine status data from a country in Southeast Asia , a region where beriberi still exists [5–7] . Despite variation in the prevalence of thiamine deficiency by cut-off , there is clear evidence of suboptimal thiamine status among women of childbearing age and their children in Cambodia . Of highest concern are infants aged 6–12 mo ( n = 50 ) , of whom 38% was classified as thiamine deficient by the most conservative cut-off , and up to 70% using the most liberal one ( eThDP < 180 nmol/L ) . Although not included in this study , Cambodian infants aged 0–6 mo , who are at the highest risk for developing infantile beriberi [6 , 8] due to the relatively high thiamine needs compared to their body size [8] , are likely to have a poor thiamine status too . Given the peak in infant mortality around 3 months of age in Cambodia , combined with the biochemical evidence of a high prevalence of thiamine deficiency in the population , we are convinced that infant beriberi is a highly under recognized cause of death in Cambodia , and that infants <12 mo of age are at the highest risk for thiamine deficiency . Indeed , Kauffman et al . estimated that infantile beriberi might be responsible for 6% of overall infant mortality in Cambodia [44] . Whereas other causes of infant mortality have been addressed , leading to a considerable decrease in infant mortality over the last 2 decades [27] , the peak at 3 months of age has remained ( see Fig 1 ) . Children had significantly higher eThDP than their mothers ( 174 versus 150 nmol/L; P < 0 . 001 ) . This is consistent with a recent study in Cambodia in which we measured eThDP among women ( 18–45 y ) and their children ( 6–59 mo ) in Prey Veng province as part of a randomized controlled trial investigating thiamine-fortified fish sauce [45] . Women and children in the control group ( who received only nutrition education; 92 mothers and 87 children ) had mean ( 95% CI ) eThDP of 184 nmol/L ( 169–198 nmol/L ) and 213 nmol/L ( 202–224 nmol/L ) , respectively . Coats and colleagues reported similar values among Cambodian women of childbearing age: 141 and 150 nmol/L ( thiamine diphosphate in whole blood , corrected for hematocrit ) among mothers of infants with and without the clinical symptoms of infantile beriberi , respectively [5] . The fact that Coats et al . reported ThDP concentrations in infants with beriberi and without beriberi that are close or within to the lower reference ranges for ThDP ( <140 and <150 nmol/L [5 , 33] ) suggests that these cut-offs might be too conservative . Alternatively , perhaps eThDP ( or whole blood ThDP ) is not a good predicator of beriberi . However , reports of beriberi are not uncommon in Cambodia and throughout Southeast Asia [5–7 , 18 , 19 , 46 , 47] , suggesting that thiamine deficiency is an issue . Therefore our biochemical data suggesting a high prevalence of thiamine deficiency suggest that current cut-offs can indicate populations at risk for thiamine deficiency even though the cut-offs might not be clinically meaningful . These Cambodian values do differ greatly from older , previously reported values among older children and adults in Europe . The mean eThDP of British adolescent girls and boys ( n = 54 , 13–14 y ) was 226 . 8 nmol/L and 206 . 1 nmol/L , respectively [35] , which are similar to those reported for free-living British elderly women ( 247 nmol/L , n = 80 ) and men ( 218 nmol/L , n = 57 ) [36] . However , due to advances in HPLC equipment and sensitivity since these latter studies were published two decades ago [35 , 36 , 39] , there is merit in investigating whether different thiamine deficiency cut-offs need to be developed for adults and children . As shown in Table 3 , there is wide variation in the prevalence of thiamine deficiency and/or suboptimal thiamine status depending on the cut-off employed , therefore it is difficult to determine the severity of low thiamine status as a public health concern in Cambodia and the wider region . However , even with the most conservative cut-offs , >25% of the mothers , and 38% of the infants were classified as deficient in our study , making thiamine deficiency a serious public health concern , which is also reflected in the peak in infant mortality around 3 months of age . Wilkinson et al . used the lowest 2 . 5% of healthy blood donors in Christchurch , New Zealand to set their deficiency cut-off of eThDP < 140 nmol/L , but astutely noted that “the lower limit of normal , seen in a healthy population , cannot be assumed to be the upper limit of abnormal” [33] . While the cut-offs shown in Table 3 may be helpful in categorizing potentially at-risk individuals , it is clear that more research is required to develop clinically meaningful thiamine deficiency cut-offs . Although a recent review has not found compelling human nutrition trials to prioritize an update of recommended dietary thiamine intakes [48] , this may be due to a heavy focus on beriberi as an outcome . Thiamine is involved in important cell functions including glucose conversion and energy metabolism in the Kreb’s cycle and pentose phosphate pathway [49] . Thiamine also performs critical enzymatic functions in processes related to brain development and function , neuronal communication , as well as immune system activation , signaling and maintenance [50] . There is evidence that obesity may impair thiamine utilization and alter requirements: a recent American study reported thiamine deficiency in 15 . 5–29% of obese patients seeking bariatric surgery [51] . In addition , considerable evidence over the past century has linked thiamine deficiency to neurological problems including cognitive deficits and encephalopathy [52] . Even a short-term exposure to poor thiamine intake in early life may have long-term impacts on cognition [53 , 54] . It is difficult to establish new thiamine cut-offs without new human studies on these manifestations of thiamine deficiency that are distinctly different from beriberi . Perhaps a wider description of clinical syndromes of thiamine deficiency is needed , compiling all under one term , for example , Thiamine Deficiency Disorders , just as advancing insights into iodine deficiency in the early 1980s led to the use of the term Iodine Deficiency Disorders . And perhaps circulating thiamine concentrations might not sufficiently reflect thiamine status . We further urge future researchers to collect beriberi prevalence data from clinical settings with matched biochemical samples to better guide development of cut-offs that have clinical and/or physiological meaning . This study has several strengths , but most notably it is the first nationally representative evaluation of eThDP in a country in Southeast Asia ( while a national study in the Philippines using erythrocyte transkelotase activity is the most recent in the region in two decades [55] ) . Thiamine status was measured among women of childbearing age and children because , due to unequal household food distribution and higher needs relative to body size , these groups are at highest risk of nutritional deficiencies in low-income countries . These are also the most pertinent population groups for biochemical thiamine assessment because thiamine-deficient mothers confer a higher risk of infantile beriberi [9–11] , and in turn mortality [6] , to their children . There is evidence that increased thiamine intake improves biochemical thiamine status among mothers and their infants in Cambodia [30] . During a recent randomized controlled trial in Prey Veng we found that maternal consumption of thiamine-fortified fish sauce over 6 mo throughout pregnancy and lactation resulted in higher eThDP among mothers and infants , as well as breast milk thiamine , compared to a control sauce containing no thiamine [30] , indicating that this group shows potential for improvement in eThDP . However , thiamine deficiency may be common among the elderly [56] , and beriberi outbreaks among adult men have also been reported [47 , 57] , likely due to higher thiamine needs with increased physical activity and a high-carbohydrate diet [22 , 49 , 58] . Therefore future studies should include the full range of population groups . Dietary intake data was not collected in this study , therefore while it is well established that low dietary thiamine intake causes low biochemical thiamine status [22] , and that there is little thiamine available in the Cambodian diet [15 , 59] , we cannot provide direct causation for low eThDP in this population . Consistent with recent studies of thiamine status in Cambodia , we report low eThDP among a nationally representative sample of Cambodian women and their young children . Thiamine status classification varies dramatically depending on the cut-off employed , from 27% to 78% , and 15% to 58% among mothers and children , respectively . More research is required to develop more useful , clinically meaningful thiamine status cut-offs . However , in view of the peak in infant mortality around 3 months of age suggestive of infantile beriberi , immediate action is required to develop interventions to increase thiamine intake in Cambodia and the wider Southeast Asia region where thiamine deficiency and beriberi remain a public health concern . | Thiamine is an often-overlooked micronutrient of concern in Cambodia and throughout Southeast Asia , where reports of beriberi are not uncommon due to a diet of thiamine-poor , white , polished rice . Thiamine plays a critical role in cellular energy generation , and also modulates neuronal and neuromuscular transmissions . Thiamine deficiency can progress to beriberi , which can be fatal . Although several recent studies have investigated thiamine status and/or beriberi in the region , this is the first nationally-representative biochemical thiamine data from any country in Southeast Asia . Unfortunately , there is a lack of clinically meaningful cut-offs to interpret these data . Upwards of 10 different cut-offs exist , but many are simply the lower bounds of a reference range , and therefore do not align with clinical symptoms of beriberi . Using the most conservative cut-off from the literature , 27% of mothers and 15% of children were thiamine deficient . More research is required to develop more useful , clinically meaningful thiamine status cut-offs . In addition , given the distinctive peak in Cambodian infant mortality data suggestive of infantile beriberi , immediate action is required to improve the thiamine status in Cambodia , and likely in other countries in Southeast Asia . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"b",
"vitamins",
"children",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"chemical",
"compounds",
"geographical",
"locations",
"beriberi",
"organic",
"compounds",
"age",
"groups",
"nutrition",
"infants",
"families",
"chemistry",
"vitamins",
"nutritional",
"deficiencies",
"people",
"and",
"places",
"blood",
"organic",
"chemistry",
"asia",
"anatomy",
"neurology",
"physiology",
"population",
"groupings",
"cambodia",
"physical",
"sciences",
"biology",
"and",
"life",
"sciences",
"thiamine",
"micronutrient",
"deficiencies"
]
| 2017 | High prevalence of thiamine (vitamin B1) deficiency in early childhood among a nationally representative sample of Cambodian women of childbearing age and their children |
There is growing evidence that population-level brain activity is often organized into propagating waves that are structured in both space and time . Such spatiotemporal patterns have been linked to brain function and observed across multiple recording methodologies and scales . The ability to detect and analyze these patterns is thus essential for understanding the working mechanisms of neural circuits . Here we present a mathematical and computational framework for the identification and analysis of multiple classes of wave patterns in neural population-level recordings . By drawing a conceptual link between spatiotemporal patterns found in the brain and coherent structures such as vortices found in turbulent flows , we introduce velocity vector fields to characterize neural population activity . These vector fields are calculated for both phase and amplitude of oscillatory neural signals by adapting optical flow estimation methods from the field of computer vision . Based on these velocity vector fields , we then introduce order parameters and critical point analysis to detect and characterize a diverse range of propagating wave patterns , including planar waves , sources , sinks , spiral waves , and saddle patterns . We also introduce a novel vector field decomposition method that extracts the dominant spatiotemporal structures in a recording . This enables neural data to be represented by the activity of a small number of independent spatiotemporal modes , providing an alternative to existing dimensionality reduction techniques which separate space and time components . We demonstrate the capabilities of the framework and toolbox with simulated data , local field potentials from marmoset visual cortex and optical voltage recordings from whole mouse cortex , and we show that pattern dynamics are non-random and are modulated by the presence of visual stimuli . These methods are implemented in a MATLAB toolbox , which is freely available under an open-source licensing agreement .
Recent advances in brain recording techniques have led to a rapid influx of high spatial- and temporal-resolution datasets of large neural populations [1–4] . One of the major challenges in modern neuroscience is to identify and extract important population-level structures and dynamics from these datasets [5 , 6] . Traditionally , neural population activity has been mainly studied from the perspective of temporal synchrony or correlation , and relating correlated neural activity to brain functions has been the major focus of many studies in neuroscience during the past two decades [7 , 8] . However , growing evidence indicates that population-level brain activity is often organized into patterns that are structured in both space and time . Such spatiotemporal patterns , including planar traveling waves [9–11] , spiral waves which rotate around a central point [12–14] , source and sink patterns which expand or contract from a point [13 , 15] , and saddle patterns which are formed by the interaction of multiple waves [13] , have been observed at different neural levels within multiple recording techniques , including multi-electrode arrays [13 , 16–18] , voltage sensitive dye ( VSD ) imaging [9 , 12 , 19] , and electroencephalography ( EEG ) , electrocorticography ( ECoG ) , magnetoencephalography ( MEG ) and functional magnetic resonance imaging ( fMRI ) [20–24] . The functional role of these spatiotemporal patterns is a subject of active research: In spontaneous activity , propagating patterns have been shown to follow repeated temporal motifs instead of occurring randomly [13 , 15] , and are postulated to facilitate information transfer across brain regions [10 , 17] and carry out distributed dynamical computation [25] . In sensory cortices , stimuli can elicit repeatable propagating patterns [9 , 10 , 19 , 26 , 27] , and the properties of these waves can be linked to stimulus features . For instance , the phase and amplitude of traveling waves in the motor cortex and visual cortex correlate with reach target location [17] and with saccade size [18] , respectively , and the propagation direction of moving patterns in the visual cortex is sensitive to visual movement orientation [28] . These studies thus indicate that the ability to detect and analyze these patterns is essential for uncovering the principled dynamics of neural population activity and for understanding the working mechanisms of neural circuits [15 , 26 , 29 , 30] . In this study , to detect changes of neural signals happening across both space and time , we introduce velocity vector fields which represent the speed and direction of local spatiotemporal propagations . These vector fields allow us to make a novel conceptual link between spatiotemporal patterns in neural activity and complex patterns such as vortices or eddies found in the field of fluid turbulence [31–33] , in which these patterns are similarly characterized by using velocity fields of the underlying moving molecules . Velocity vector fields in our methods are computed by adapting optical flow estimation methods originally developed in the field of computer vision [34] . Optical flow techniques have previously been implemented to analyze brain activity [13 , 26–28] , but here we extend these methods to consider the amplitude and phase of oscillatory neural signals , allowing for a comprehensive analysis of neural spatiotemporal patterns . When constructed from oscillation phase , velocity vector fields are conceptually similar to phase gradient vector fields as often used in previous studies [15 , 18] . However , velocity vector fields provide a conceptual basis for us to adapt methods from turbulence to develop a unified methodological framework for analyzing neural spatiotemporal patterns . We show that by examining the critical points in a velocity vector field ( also called “stationary points” or “singularity points” ) , where the local velocity is zero [35] , different types of spatiotemporal patterns including spiral waves ( “foci” ) , source/sink patterns ( “nodes” ) and saddles can be detected . In addition to these complex wave patterns , neural systems can exhibit widespread synchrony and planar travelling waves . These types of activity are common to many physical and biological systems , and can be detected by introducing global order parameters calculated from velocity vector fields [36] . These methods thus enable the automatic detection of a diverse range of spatiotemporal patterns after user-defined parameters have been chosen; these parameters are discussed in detail in Methods and Materials . Aside from detecting these patterns , our methods can provide systematic analysis of pattern dynamics including their evolution pathways and their underlying spatiotemporal modes that exhibit intrinsic and inseparable spatial and temporal features , thus providing a novel alternative to existing dimensionality reduction techniques which instead separate space and time components [6] . We validate the effectiveness of all methods and their implementation in the toolbox through multiple approaches . Using synthetic data with known pattern activity , we show that spatiotemporal pattern detection is accurate and reliable even in noisy conditions . We then analyze local field potentials from multi-electrode arrays in marmoset visual cortex and whole-brain optical imaging data from mouse cortex to test our methodological framework across different recording modalities , species , and neural scales . We find that pattern properties including location and propagation direction are modulated by visual stimulus , and that patterns evolve along structured pathways following preferred transitions .
Neural activity , although appearing highly disordered at the single-neuron level , can form dynamical coherent structures such as propagating waves at the population level [37] . There are many other complex systems that display similar emergent pattern dynamics , including fluid turbulence , in which coherent flows and vortices emerge from interacting molecules that behave irregularly [31 , 32] . Velocity vector fields , which represent the direction and speed of fluid motion , are essential mathematical tools for detecting and analyzing coherent activity patterns embedded in turbulent flow [32]; studies using this approach typically separate activity patterns at different scales , independently detecting both large-scale flows and small-scale eddies [33] . In turbulent flow , velocity vector fields are typically measured by following the movement of tracer particles within the fluid [38] . Here , we introduce a method for calculating analogous velocity vector fields in neural signals , representing the local direction and speed of propagating activity at each recording site . As for in studies of coherent structures in turbulence , these velocity fields obtained in neural data provide a powerful conceptual framework for analyzing a diverse range of propagating wave patterns in the brain . For a data sequence D ( x , y , t ) , which may represent the raw recorded signal z or the amplitude A or phase θ of an oscillatory neural signal , extracted by using either the Hilbert transform [39] or complex Morlet wavelets [40] ( see Oscillatory data filtering in Methods and Materials ) , the velocity vector field w ( x , y , t ) = ( u ( x , y , t ) , v ( x , y , t ) ) represents the velocity in x- and y-directions at each location between time t and t+δt , where δt is the time step specified by the sampling frequency . If data contain multiple trials , the velocity vector field is computed iteratively for each trial . To calculate the velocity field w ( x , y , t ) , we adapt optical flow estimation techniques from the field of computer science , which were first developed by Horn and Shunck to track the motion between successive frames of a sequence of images [34] . In their original formulation , the optical flow is calculated by solving two constraints . The first is the data constancy constraint , D ( x+u , y+v , t+δt ) −D ( x , y , t ) =0 , ( 1 ) which specifies that the same data is present at time t and time t+δt , only shifted in space . To first order , this can be expressed as Ed=Dxu+Dyv+Dt≈0 , ( 2 ) where Ed is the error in the data constancy , Dx and Dy denote spatial derivatives , and Dt denotes the temporal derivative at the point x , y , t . The second is the spatial smoothness constraint , which specifies that the computed velocity fields contain smooth and continuous motion where possible . This constraint can be expressed as Es2=|∇u|2+|∇v|2 , ( 3 ) where Es quantifies the overall departure from smoothness , and ∇≐∂x , ∂y denotes the gradient operator . The velocity vector field can uniquely be defined by minimizing both these error terms: minu , v{∬[ρ ( Ed2 ) +αρ ( Es2 ) ]dxdy} , ( 4 ) for some regularization parameter α and penalty function ρ . Horn and Shunck used a quadratic penalty , ρx2=x2 , but this can lead to inaccuracies if the underlying data contains hard edges and adjacent regions moving in different directions [41] . More accurate velocity vector fields can be obtained by using the Charbonnier penalty , ρ ( x2 ) =x2+β2 , for a small positive constant β [42] . Eq 4 can be solved by linearizing its corresponding Euler-Lagrange equations , creating a unique velocity vector field w ( see Solving optical flow equations in Methods and Materials ) . Turbulence studies typically separate activity at different scales based on velocity fields [33] . We similarly implement independent procedures to detect global patterns ( plane waves and synchronous activity ) , which are active across the whole recording area , and complex wave patterns , including source , sink , spiral and saddle patterns , which are characterized by local activity around their central points . Complex spatiotemporal wave patterns , which are analogous to eddies , are organized around critical points where the velocity field has zero magnitude [35] . These complex wave patterns generate distinctive dynamics around their central critical points; in our methodology , we exploit this dynamical property to automatically detect and classify such patterns . In velocity fields , we identify critical points as locations where both x- and y-components of the velocity are zero by finding intersections of the bilinearly interpolated zero-level contours lines of u and v [43] . Each critical point is then categorized by the Jacobian matrix , J= ( ∂u∂x∂u∂y∂v∂x∂v∂y ) , ( 5 ) which is estimated at the critical point using bilinear interpolation from the corners of the surrounding 4-element cell of recording sites . Depending on the trace ( τ ) and determinant ( Δ ) of the Jacobian , critical points are classified as a node ( ∆>0 and τ2>4∆ ) , focus ( ∆>0 and τ2<4∆ ) , or saddle ( ∆<0 ) , and node and focus points are further classified as stable ( τ>0 ) or unstable ( τ<0 ) . These classes of critical points correspond to different types of wave patterns ( Fig 1 ) : Nodes expand or contract from a critical point , forming sources or sinks , respectively; saddles have one stable axis and one unstable axis , and are typically formed through interactions between different waves; and foci rotate around the critical point , thereby corresponding to spiral waves . In addition to their rotating motion , foci can also involve expansion or contraction from the critical point , forming spiral-out or spiral-in wave patterns , respectively . However , previous studies of spiral waves have not distinguished between spirals-out and spirals-in [12 , 14 , 44] . In our methods and toolbox , these patterns can therefore optionally be combined to facilitate direct comparison with other published results . Although complex wave patterns are classified only by the local properties of their central critical point , they can spread over larger regions of space . We thus develop a method for characterizing the spatial extent of complex wave patterns by using the winding number ( Poincaré index ) , which has a value of +1 for node and focus patterns and -1 for saddle patterns for all closed paths within the pattern’s extent around the location of the critical point [43] . We create approximately circular paths around the location of the critical point , and the winding number is estimated in each of these paths as windingnumber=12π∑k=1n ( θk+1−θk ) , ( 6 ) where θk is the angle of the k-th vector around a closed counter-clockwise path with n points , angles are subtracted circularly , and where θn+1=θ1 . We compute the winding number in paths of expanding size around a pattern’s center , and its spatial extent is defined by the largest area within which all computed winding numbers are consistent with the critical point type . This procedure therefore provides an efficient estimate how far wave patterns spread across the cortex , an important property of neural oscillatory activity . We next introduce methods for detecting and analyzing simple , large-scale patterns such as synchrony and planar waves by defining order parameters based on the velocity vector fields . We detect planar waves using an order parameter defined as the average normalized velocity [36]: φ ( t ) =||∑x , yw ( x , y , t ) ||∑x , y||w ( x , y , t ) || ( 7 ) This statistic is equivalent to phase gradient directionality [17] except it uses velocity vector fields instead of the phase gradient . Normalized velocity ranges from zero to one , with φ→1 as velocity vectors align to one direction , reflecting coherent motion across the recording area . Plane wave activity is therefore detected at times when φ is greater than some threshold value Tpw , which should be close to one ( Tpw=0 . 85 by default in the toolbox ) . If data has been band-pass filtered to extract the oscillation phase θ , we also detect large-spread synchronous activity using another order parameter , which is defined as the resultant vector length of phase across the recording area [45 , 46] , R ( t ) =1N|∑x , yeiθ ( x , y , t ) | , ( 8 ) where N is the number of spatial recording sites in phase maps . The resultant vector also ranges from zero to one , with R→1 as the phase of oscillations at all recording sites align to the same value , reflecting wide-spread synchrony . We note that the order parameter as defined in Eq 8 is similar to that used to characterize global synchrony in coupled phase oscillators [47] , and 1-R is commonly defined as the circular variance [45] . Synchronous activity is therefore detected at times when R is higher than another threshold value Tsyn , the default of which is Tsyn=0 . 8 in the toolbox . To test the performance of our pattern detection methods , we generated artificial data sets with simultaneous source and sink patterns active at the same frequency , located at random positions and propagating in random directions within a 12×12 spatial grid ( see Simulated data in Methods and Materials ) . We then added Gaussian white noise , band-pass filtered the signal , calculated velocity vector fields , searched for complex wave patterns in the velocity fields , and compared the detected pattern centers with their true locations . An example of this procedure is shown in Fig 2 , which shows calculated velocity fields and pattern centers between two frames of a simulated data set ( Fig 2A ) . The velocity fields depend on two parameters in the optical flow estimation procedure ( Eq 4 ) : The smoothness regularization parameter α , and non-linear penalty constant β . The smoothness regularization parameter α determines the weighting of the smoothness constraint compared to the data constraint , and thereby the overall smoothness of the velocity fields . Small values of α generate velocity fields that primarily capture local changes and are therefore sensitive to added noise , potentially leading to the detection of spurious , noise-driven patterns ( Fig 2B , left column , α = 0 . 1 ) . Large values of α are more robust to noise , but can over-smooth the data , creating mostly uniform flow fields that do not capture the underlying dynamics ( Fig 2B , right column , α = 1 . 5 ) . Reasonable values for α can range from ~0 . 1 to ~20 , depending on the size of the data , the dynamics of the propagations , and the level of noise; for example , reducing the spatial sampling frequency of a dataset reduces the number of grid spaces between complex patterns , typically requiring a lower value of α to effectively resolve individual patterns . The non-linear penalty constant β determines the degree of non-linearity of the penalty functions , with large values β≫1 resulting in a quadratic penalty and small values β≪1 in a more robust non-linear penalty . Small values of β give more accurate velocity vector fields for any regions with discontinuous motion in the underlying data [41 , 48] , but we find that such discontinuities are rare in neural recordings , so using large values of β generally gives similar results ( Fig 2B ) . In addition , when β is large and the equations are effectively quadratic , the optical flow procedure can typically converge much faster . The choice of appropriate values for α and β cannot be fully automated for a real dataset without making assumptions about the dynamics of the data . However , pattern detection accuracy can be evaluated in simulated datasets with specified properties and pattern dynamics , which can then be used to guide parameter choices in real data . Fig 3 illustrates the effectiveness of the pattern detection algorithm for one such set of properties and dynamics ( see Simulated data in Methods and Materials ) . The detected spatial position of patterns is most accurate when using small values of α ( α≤1 , Fig 3A ) . Using a quadratic penalty function ( β=10 ) generally gives more consistent results across a range of α values than a non-linear penalty function ( β=0 . 01 ) and results in fewer missing patterns ( Fig 3B ) , but using the non-linear penalty can give a lower false positive rate ( Fig 3C ) . Plotting the true positive rate against the false positive rate provides a clear way to examine the effectiveness of pattern detection across a range of parameters ( Fig 3D ) . We generally recommend using large values of β when examining new data sets , as this provides more reliable performance and faster processing overall . Additionally , pattern detection is largely unaffected by noise if the variance of the noise is equal or less than the variance of the pattern oscillations and remains fairly accurate for significantly greater noise levels ( Fig 3E–3G ) . To validate our methods and test for wave pattern activity in real neural data from different scales and imaging techniques , we examined previously published LFP recordings from the MT area of anaesthetized marmosets [49] and optogenetic voltage imaging recordings from a complete cortical hemisphere in awake mice [50] ( see Experimental recordings in Methods and Materials ) . Using our methodology , we searched for wave patterns within the phase and amplitude of oscillations across a range of frequency bands . Both datasets exhibited a rich repertoire of wave patterns which were successfully detected . Some examples of common pattern activity for each modality are shown in Fig 4 . In delta-band phase of the marmoset data , complex waves were commonly present across the whole 16 mm2 recording area , including saddle ( Fig 4A ) and spiral-out ( Fig 4B ) patterns . We also observed multiple complex wave patterns active simultaneously in different areas of the cortex , as shown for sink and saddle patterns in Fig 4C . In the mouse data , complex waves were present in the phase of slow ( 4 Hz ) oscillations , and these waves sometimes spread across the whole cortical hemisphere , including sink ( Fig 4D ) and spiral-in ( Fig 4E ) patterns . Large-scale propagating patterns were also present in the amplitude of 10 Hz oscillations , where multiple spreading patterns often interacted to form saddles ( Fig 4F ) . These examples demonstrate that complex wave patterns are present at multiple scales of brain activity , and that these patterns can be detected and quantified through our methodology . Having presented our pattern detection methods , we now demonstrate how these techniques can be used to examine the properties and dynamics of waves patterns in more detail , and how these properties can be further related to brain function . Directly tracking simple and complex wave patterns allows their location , movement direction , prevalence , duration and other properties to be collated across many occurrences . To validate the results of the pattern detection procedure , the properties of patterns detected in a real dataset can be compared to those of patterns detected in a surrogate dataset comprised of noise with similar characteristics to the original data ( see Pattern detection parameters and result validation in Methods and Materials ) . The processes of band-pass filtering and velocity vector field estimation can smooth data and may therefore generate spurious wave patterns in noise-driven surrogates . However , these patterns in surrogate data are typically more localized and transient than real neural wave patterns and can therefore be mostly removed if the minimum pattern spatial extent and duration parameters are sufficiently large . In neural recordings with genuine wave pattern activity , all pattern types will typically occur more frequently ( Fig 5A ) , be present for a larger proportion of recording time ( Fig 5B ) and last longer per occurrence ( Fig 5C ) than equivalent patterns in noise-driven surrogates . The properties of wave patterns can vary depending on brain state , recording location , or cognitive task , revealing relevant dynamical changes in the recorded neural system . An example of this is shown in Fig 6 , which compares properties of patterns in spontaneous and stimulus-evoked phase velocity fields from the same animal . During ongoing activity ( sustained blank screen stimulus ) , plane waves were active for much of the recording time and propagated in a range of directions ( Fig 6A , mean resultant vector length 0 . 28 ) . Complex wave patterns were also common and did not form randomly in space . Instead their central critical points were clustered around preferred locations ( Fig 6B ) , which were situated at different points in the recording array for node and saddle patterns . When relevant stimulus was presented ( coherently propagating dot fields turned on and off every two seconds ) , relatively fewer plane waves were active overall , but their propagations were more tightly distributed around one preferred direction ( Fig 6C , mean resultant vector length 0 . 42 ) . The presence of stimulus also affected the overall prevalence of critical point patterns , increasing the number of stable and unstable nodes and decreasing the number of saddles , and changed their patterns of distribution across space ( Fig 6D ) . Our methods can therefore be used to quantify changes in spatiotemporal pattern dynamics driven by different stimuli , cognitive tasks or behavioral states . Detected wave patterns can also be processed to reveal their temporal evolution dynamics . Brain activity evolves between different activity patterns in a complex and non-random way , but the mechanisms of these transitions are not well-understood [13 , 15] . Our methodology provides an ideal framework for exploring such dynamics: Once all patterns in a recording have been identified , common pattern transitions and motifs can easily be identified . We demonstrate some of these evolution dynamics in stimulus-evoked LFP recordings ( see Pattern evolution dynamics in Methods and Materials ) . Patterns were typically active for tens to hundreds of milliseconds , often then transitioning into a different pattern ( Fig 7A ) . The total number of transitions between all pairs of pattern types were counted across a recording , and the significance of these observed counts was established by comparison to the expected number of counts if all patterns began and ended at random times ( Fig 7B ) . Using this simple analysis , we observed that periods of plane wave and synchronous activity were usually interspersed by other pattern types , synchronous activity was highly likely to transition to or from all other pattern types , and patterns commonly evolved from sources to sinks and vice versa . This analysis illuminates the temporal dynamics of the spatiotemporal activity patterns present in the recording and provides quantitative measurements which can be linked to cognitive tasks or used to constrain models of cortical dynamics . Similar analyses can also facilitate tracking the movement of neural structures of interest across brain regions [15] , detecting repeated motifs in pattern dynamics [13] , or examining gradual changes in pattern dynamics corresponding to changes in brain states [50] . In neural recordings , amplitude and phase data at the same frequency reflect different properties of brain activity , with amplitude representing a combination of the coherence and overall activity of a local ensemble and phase representing the timing of its oscillations . Accordingly , these signals typically contain different spatiotemporal patterns , and both phase and amplitude patterns can be relevant and informative . Fig 8 illustrates the spatiotemporal profile of raw LFP data , filtered LFP data , and the amplitude and phase of filtered LFP data , again taken from marmoset visual area MT . The spatiotemporal dynamics in the raw data ( Fig 8A ) primarily reflect those in the oscillations with greatest power , but also contain a large amount of noise from other frequencies . Filtering the data to a narrow-band signal ( Fig 8B ) reduces the noise by extracting the patterns present in the chosen frequency band alone , but these wave patterns are typically complicated as they are influenced by two different types of propagating activity: amplitude patterns ( Fig 8C ) , which capture the movement of the overall shape of the wave and travel at the group velocity [51] , and phase patterns ( Fig 8D ) , which capture the progression of timing differences between electrodes and travel at the phase velocity . In a general oscillating system , the phase and amplitude are independent properties which have no a priory reason to affect each other . However , there is some evidence that phase and amplitude patterns can be related in some neural systems: Phase patterns in rabbit sensory cortices are more commonly observed around the formation of new amplitude patterns [52] , and spiral waves in mammalian neocortex exhibit consistently reduced amplitudes at their centers [12 , 44] . Examining both phase and amplitude separately may therefore uncover similar relationships in other experimental protocols and can reveal a more comprehensive understanding of the underlying dynamics of cortical circuits . For example , two simple patterns can be resolved from the complicated activity in Fig 8B: A gradually expanding activation from a point near the center of the recording array , as revealed by the amplitude velocity field in Fig 8C , and a plane wave propagating across the recording area , as revealed by the phase velocity field in Fig 8D . Whilst direct identification of wave patterns in velocity fields as described in the previous sections allows for patterns’ individual dynamics to be fully characterized , the procedure does not specify the extent to which these patterns contribute to the overall spatiotemporal dynamics of a recording . To address this , we introduce a complementary method for studying wave activity in neural recordings by using velocity field decomposition , which finds low-dimensional spatiotemporal modes that capture the majority of variance in the system . Dimensionality reduction techniques are commonly used for uncovering underlying neural mechanisms of brain function [6] . However , the majority of existing techniques use principal component analysis ( PCA ) or similar procedures that decompose data into independent spatial and temporal modes , obscuring activity that is not time-space separable such as propagating waves and patterns [53] . Some studies have used decomposition techniques to specifically detect waves by examining phase gradients in complex decompositions of data [26 , 29] . To identify dominant spatiotemporal patterns in our framework , we again obtain inspiration from the field of turbulence , in which dimensionality reduction is often directly applied to velocity fields , capturing low-dimensional spatiotemporal dynamics [33] . In turbulence , dimensionality reduction can be performed through a variety of different decomposition methods , including Reynolds decomposition , principal component analysis ( or proper orthogonal decomposition ) , and dynamic mode decomposition [54] . These techniques find modes capturing the majority of the energy in the system , which is not well-defined for velocity fields of neural data as it is in fluid flows , but some of these methods nonetheless can be adapted to find low-dimensional representations of the primary spatiotemporal dynamics of a neural recording . We implement a simple singular value decomposition ( SVD ) to extract the dominant spatiotemporal patterns from a time series of velocity fields in an efficient and parameter-free way . To reorganize the velocity fields ux , y , t , vx , y , t into standard form for decomposition with variables in columns and observations in rows , we combine spatial dimensions and rearrange indices to define u~t , r , v~t , r , for time t and recording site r . We then use two alternate approaches to combine ũ and ṽ . In the first approach , we concatenate the two matrices across recording sites to define the real matrix w˜re ( t , r' ) =[u˜|v˜] . In the second , we represent the velocity field as a complex number to form the complex matrix w˜co ( t , r ) =u˜ ( t , r ) +iv˜ ( t , r ) . In either case , the singular vector decomposition ( SVD ) is defined as w˜=TΣR* , ( 9 ) where w~ denotes w~re or w~co , T and R are unitary matrices , * denotes the conjugate transpose , and Σ is a rectangular diagonal matrix of positive numbers σi , called the singular values [55] . This operation finds orthogonal linear combinations of recording sites that explain the greatest variance in the velocity fields , and is closely related to PCA: if w~ has been shifted so that each recording site has zero sample mean , then R comprises exactly the principal component loadings , and σi2 are the principal component scores [56] . However , normalizing the velocities at each recording site is counterproductive in this application , as biases in propagation direction are an important component of wave dynamics . The k-th spatial mode , defined by the velocity field in the k-th column of R , explains a proportion of the overall variance given by σk2/∑iσi2 , and has a time course given by the k-th row of T . The vector SVD procedure is closely related to traditional PCA methods , as both techniques reduce the dimensionality of a dataset by extracting patterns that comprise the bulk of the variance and their evolution over time . The differences between these approaches are illustrated in Fig 9 , which again shows marmoset LFP data during moving dot-field stimulation . PCA typically decomposes data into orthogonal spatial modes ( Fig 9A ) , which comprise linear combinations of recording sites [6] . Vector SVD instead processes the velocity vector fields to extract spatial modes which are vector fields themselves ( Fig 9B ) , and therefore represent distinct propagation patterns in the underlying data . In both cases , each spatial mode has a corresponding time-course ( or temporal mode ) , describing its evolution across the duration of a recording ( Fig 9C and 9D ) . Although the dominant PCA modes explain more variance than their SVD counterparts , their temporal components reveal structured interactions between the dominant spatial modes ( Fig 9C ) , generating spatiotemporal activity patterns which are difficult or impossible to determine directly from the PCA modes . In contrast , SVD spatial modes directly reflect these spatiotemporal patterns , and their evolution over time represents the strength of different pattern types in response to stimulus . In Fig 9 , stimulus onset generates large , clear changes in spatiotemporal pattern dynamics revealed by SVD ( Fig 9D ) : Sink pattern activity increases dramatically but transiently ( shown by the large deflection in mode 3 ) ; plane waves ( modes 1 and 2 ) increase in activity more modestly , but change direction soon after stimulus onset ( as indicated by the sign change of mode one at 300 ms ) and are sustained for a longer period . These results suggest that stimuli can directly affect the dynamics of propagating wave patterns , but that these changes are obscured when using PCA or other decomposition methods that separate space and time . We find that in both spontaneous and stimulus-evoked LFP recordings , velocity vector fields in phase and amplitude at all frequencies display consistent dynamics: the most dominant modes typically reflect orthogonal directions of plane wave motion , and the next most dominant modes contain complex patterns including sources , sinks , spirals and saddles ( Fig 10A ) . Despite these similarities , the disparities between spatial modes in different recordings or conditions can reveal major differences in the underlying pattern dynamics , including the primary directions of plane wave motion , the center location of complex patterns , and the relative prevalence of different pattern types . As an example , we compare the dominant SVD modes during stimulus-evoked activity in Fig 9B to those during ongoing activity in the same recording ( Fig 10A ) . The four most dominant modes represent the same activity patterns , but they display slightly different dynamics: The primary propagation direction of plane waves changes ( shown by the direction of mode 1 ) ; the central locus of source , sink and saddle activity changes location ( shown by the critical point in nodes 3 and 4 ) ; and ongoing activity overall contains more plane wave and less source , sink and saddle activity ( as revealed by the percentage of variance explained ) . In the SVD method discussed thus far , each class of spatiotemporal pattern may be represented across multiple modes ( e . g . , modes one and two both reflect plane wave activity ) , making their overall prevalence more difficult to calculate . To address this issue , we also implement a modified SVD procedure that we call complex singular value decomposition ( cSVD ) , which treats velocity vectors as complex numbers . In this approach , temporal modes have both a real and imaginary component , allowing spatial modes to both scale and rotate over time: The amplitude of the temporal mode gives the relative strength of the pattern , and the argument of the real and imaginary components gives the angle by which all vectors are rotated . This approach effectively combines real SVD modes together ( Fig 10 ) : Modes with plane waves travelling in any direction are combined , as are modes containing source , sink and spiral patterns with the same center location , or saddles with the same center location . This allows the overall relative contribution of each type of activity pattern ( plane waves , expanding or contracting waves , saddle patterns ) to be accurately evaluated , but information about the direction of patterns is removed to the complex time evolution .
In this paper , we have introduced a methodological framework and associated MATLAB toolbox for the classification and analysis of propagating wave patterns in neural recordings , and illustrated these methods using simulated data , LFP recordings from marmoset visual area MT and whole-brain optical imaging data from mouse cortex . The toolbox is freely available under an open source agreement from [https://github . com/BrainDynamicsUSYD/NeuroPattToolbox] . As we have demonstrated , our methods provide a framework for uncovering the spatiotemporal organization principles of these patterns and for examining how they are related to brain function . We have introduced velocity vector fields to characterize how neural oscillatory signals change across space and time . Based on these vector fields a range of mathematical techniques including order parameters , critical point analysis and winding number calculation are uniquely combined to detect a diverse range of wave patterns and to characterize their key spatiotemporal organization properties . Our methods thus build upon the application of optical flow analysis for detecting wave patterns developed in previous studies [15 , 57 , 30 , 58] . As we have demonstrated , order parameters can be used to detect the presence of large-scale plane waves or synchronous activity , and critical point analysis can detect complex wave patterns , comprising sources , sinks , spirals-in , spirals-out , and saddles . Calculation of the winding number around critical points can then be used to measure the precise size of wave patterns , which may be useful in future studies to examine the spatial scale of neural features and effects across different frequencies [59] . These approaches allow multiple classes of waves to be tracked simultaneously and systematically . Applying these methods to experimental data , we successfully identified both small-scale wave patterns in LFP recordings from anaesthetized marmoset visual cortex and large-scale patterns in whole-brain optical recordings from awake mice . In both datasets , multiple coexisting patterns were commonly active and all patterns were significantly more prevalent than in noise-driven surrogate data ( Fig 5 ) . Furthermore , we showed that visual stimulation can change the direction of plane wave activity and the position of complex waves in marmoset area MT ( Fig 6 ) , and that these waves evolve between different types in a structured way beyond what is expected by chance ( Fig 7 ) . These results are consistent with previous studies associating visual stimuli and traveling waves [9 , 10 , 19] , and showing that neural dynamics evolve following preferred pathways [13 , 15 , 60] . However , unlike previous work , our methodology allows these patterns and their dynamics to be simultaneously detected and quantified , and places them into a framework of explicit pattern behavior to more precisely study underlying neural dynamics . In our methods , dominant spatiotemporal activity patterns can be extracted from a recording using novel vector field decomposition methods . These present a promising approach to the task of dimensionality reduction in large-scale neural recordings . Current dimensionality reduction methods typically process data into separable temporal and spatial modes which reproduce the dynamics of a recording [6] . However , these approaches find population structures that are often dominated by single-cell response properties and correlated activity [61] , and do not adequately capture activity patterns that are not time-space separable , such as propagating waves [53] . In contrast , vector field decomposition specifically targets propagating waves by directly extracting spatiotemporal pattern modes from data . We showed that stimulus onset in marmoset LFP recordings generated complicated responses in spatial PCA modes that are difficult to interpret , but clear effects on the activity of spatiotemporal pattern modes . We also showed that the dominant spatiotemporal modes are consistent across recordings but change in dynamics depending on cognitive function . In the future , this approach could be useful for examining how sensory stimuli and cognitive tasks affect wave dynamics of population-level responses , and for visualization and exploration of the underlying spatiotemporal activity in large neural data sets . Together , the detailed wave pattern tracking approach , and broad , parameter-free velocity decomposition approach provide a comprehensive analysis of spatiotemporal activity patterns in neural recordings . There are many ways that our methodology can be extended to explore spatiotemporal neural pattern dynamics beyond what has been presented in this paper . For instance , plane waves and synchronous activity are currently detected as global patterns active across the whole recording area , but it would be advantageous ( particularly in large-scale recordings ) to identify discrete local regions exhibiting these patterns . This would support accurate simultaneous analysis of brain areas displaying different dynamics , be useful for studying the spread of synchrony or plane wave propagations , and provide consistency with the localized nature of complex patterns as defined by the winding number . However , implementing localized order parameters across regions of different sizes would significantly slow the pattern detection procedure and introduce additional free parameters to the framework . Future work may develop more efficient methods to characterize localized regions displaying synchrony or planar propagations , potentially allowing entire cortical sheets to be fully and dynamically segmented into multiple interacting patterns . Additionally , our methods can be further extended to explore currently unknown mechanisms of wave pattern interactions in the brain . Firstly , localized patterns with complex dynamics that are active simultaneously may directly interact . Such interactions are prevalent in modelling studies including spiking neural networks [37] and neural firing rate models [62] , and they are theorized to be directly involved in distributed dynamic computation [25] . In experimental studies , interactions between sharp-wave ripple patterns in rat hippocampus can result in their reflection or annihilation [63] , but more complex interactions have not been examined . Secondly , patterns may interact across oscillations at different frequencies . Currently , the phase of low-frequency oscillations is known to influence the amplitude of high frequency oscillations in the brain [64 , 65] , but it is not clear how this cross-frequency coupling actually influences or is influenced by the underlying patterns in these systems . Finally , wave pattern dynamics may interact across multiple spatial and temporal scales in more complex ways , creating cascades of pattern dynamics comparable to energy cascades in turbulence studies [66] . By effectively detecting and analyzing neural spatiotemporal activity patterns simultaneously across multiple scales , our methods provide a framework for further exploring these key questions in future studies .
NeuroPatt includes two complementary methods to band-pass filter oscillatory neural data to extract amplitude or phase at a chosen frequency prior to detection of spatiotemporal patterns . The first method uses an eighth-order Butterworth filter as implemented in MATLAB’s Signal Processing Toolbox to filter data to a specified frequency range . This filter is applied in both forward and reverse directions to minimize phase distortion . The oscillation amplitude , A , and phase , θ , are then extracted from the analytic signal , zf+iz^f=Aeiθ , where z^f is the Hilbert transform [39] of the filtered data zf . The second method estimates the analytic signal at a specified center frequency using the complex Morlet wavelet transform to filter data and extract phase and amplitude with an optimal trade-off between time and frequency resolution [40] , as implemented by MATLAB’s Wavelet Toolbox . These two procedures give comparable outputs [67] , but each has advantages in different situations: The Hilbert transform allows the properties of the filtering to be precisely specified but can be invalid if the underlying frequency is not sufficiently narrow-band; the wavelet transform is usually faster to compute and always results in a valid analytic signal , but gives a less concretely defined frequency range . Both procedures are included in the toolbox , with the wavelet transform as the default option . Users without access to either the Signal Processing Toolbox or the Wavelet Toolbox can detect spatiotemporal patterns in unfiltered data , which are valid but can be contaminated by noise from multiple frequencies as illustrated in Fig 8 , or can calculate the analytic signal through other implementations of the Hilbert or wavelet transform for use in the later steps of the toolbox . Any band-pass filtering procedure necessarily involves some degree of temporal smoothing [68] , which can inhibit the extraction of precise timing information in later analysis steps . We note that this effect will not change the timing of maxima or minima in time series , as both wavelet and Hilbert filtering techniques do not distort signal phase , but they will smear out activity between these points . Velocity vector fields are calculated by solving the Euler-Lagrange equations corresponding to the minimization problem given by Eq 4: ρ′ ( Ed2 ) Dx[Dxu+Dyv+Dt]−α∇⋅[ρ′ ( Es2 ) ∇u]=0 , ρ′ ( Ed2 ) Dy[Dxu+Dyv+Dt]−α∇⋅[ρ′ ( Es2 ) ∇v]=0 , ( 10 ) Where ρ′ ( x2 ) = ( 2x2+β2 ) −1 . Note that for large values of β , ρ'x2 is approximately constant , and the optical flow for the Charbonnier penalty approaches that of the quadratic penalty . For clarity , we let ρd=1αρ′ ( Ed2 ) and ρs=ρ′ ( Es2 ) , and rewrite these equations as ρdDx[Dxu+Dyv+Dt]−ρs∇2u−∇ρs⋅∇u=0 , ρdDy[Dxu+Dyv+Dt]−ρs∇2v−∇ρs⋅∇v=0 , ( 11 ) where ∇2≐ ( ∂x2 , ∂y2 ) denotes the Laplace operator . These equations can be solved through fixed point iteration for the functions ρd and ρs after linearizing all other terms . In the toolbox , we approximate partial first derivatives with a five-point stencil 112-1 , 8 , 0 , -8 , 1 where possible ( or with centered or forward differences when close to edges ) , and approximate the Laplacian with a 2D five-point stencil [69] . If D represents phase data , it will contain temporal and spatial discontinuities where phase wraps from -π to +π , which invalidate linear difference stencils taken near these points . Instead , we approximate partial derivatives with centered or forward differences calculated using circular subtraction , θ1-θ2=modθ1-θ2+π , 2π-π . All figures in this report use parameters α=0 . 1 and β=10 unless otherwise specified . To create valid velocity vector fields , NeuroPatt has some restrictions on the format and content of input data sets . Firstly , data must be spatially arranged in a 2D square lattice of recording sites . This restriction exists primarily because of the optical flow estimation methods implemented in the toolbox , which assume spatial uniformity to maximize efficiency and accuracy . Alternate optical flow estimation methods exist for 3D data sets sampled volumetrically [70] or from non-uniform surfaces [71] , but these implementations require significant modifications to the methodology described here and are much more computationally intensive . Secondly , data must be consistent across multiple recording sites: Because velocity vector fields are computed based on local dynamics , any recording sites with erroneous activity can significantly influence the surrounding velocity vectors . As in previous studies using optical flow for brain recordings [58 , 72] , we recommend that highly noisy data are spatially filtered prior to the application of optical flow methods , and that any invalid or discontinuous recording channels are interpolated over . When amplitude spatiotemporal patterns are being examined , data should also be normalized across recording sites by subtracting the baseline or z-scoring to remove factors that could cause any regional bias , such as uneven electrode impedance or dye intensity . These processes are included as optional pre-processing steps in the toolbox . Finally , the changes between consecutive time steps in recorded signals must also be sufficiently small , as optical flow cannot be estimated if there are large discontinuities between frames . Such discontinuities can occur if signals are changing on a shorter time scale than the sampling frequency , which may be an issue for fast neural signals such as action potentials and high-frequency oscillations , or for recording techniques with low temporal resolution such as fMRI . There is no strict rule to determine if the sampling frequency is sufficiently high , but as a general guideline we suggest that signals at a single recording site should typically change by less than 10% of their maximum range between consecutive time steps . The toolbox will warn if this condition is not satisfied , potentially leading to non-convergent optical flow estimation or invalid velocity vector fields , or if the change in data is significantly below this threshold , indicating that it can be safely down-sampled for computational efficiency . The toolbox includes multiple parameters for the identification and tracking of spatiotemporal patterns . Firstly , the user can specify a minimum distance from the edge of the recorded area Ledge ( default 2 grid spaces ) for critical points to occupy , as velocity fields can be inaccurate and contain spurious critical points close to the boundary ( mainly due to the use of forward differences to approximate derivatives at these points ) . The user can also specify a minimum radius Lradius ( default 2 grid spaces ) for critical point patterns to occupy , to exclude small-scale and potentially noise-driven local patterns from analysis . Once both simple and complex patterns are detected in individual velocity fields , these individual observations are combined across time to identify persistent spatiotemporal patterns . NeuroPatt allows the user to specify a minimum duration tdur ( default 5 time steps ) for all patterns ( including global synchrony and plane waves ) : patterns which persist for less than this amount of time are discarded . To add some error tolerance to the process of linking patterns together over time , a maximum time gap tgap ( default 1 time step ) can be specified between successive instances of a pattern for it to still be counted as one spatiotemporal structure . Finally , complex patterns can move over time , so critical points of the same type in successive time steps are considered part of the same spatiotemporal pattern only if they are separated by less than a maximum displacement Ldisp ( default 0 . 5 grid spaces ) . The pattern detection methods and all relevant parameters can therefore be summarized as follows: Synchrony occurs when Rt>Tsyn at least every tgap+1 time steps for a period of at least tdur . Plane waves occur when φt>Tpw at least every tgap+1 time steps for a period of at least tdur . Nodes , foci and saddles occur when a critical point , at least Ledge away from the edge of the grid with a minimum spatial radius of Lradius , can be linked to another critical point of the same type within tgap+1 time steps and distance Ldisp , and this chain of critical points persists for at least tdur time steps . Nodes and foci with the same stability properties are typically treated as separate patterns . However , they represent the same type of motion ( expansion from or contraction to the critical point ) , so the toolbox can optionally group these critical points together for a more robust characterization of these pattern types . NeuroPatt contains a few key parameters that must be carefully selected by the user to ensure valid and accurate results . As illustrated in Fig 2 , the optical flow smoothness parameter α controls many properties of the computed velocity vector fields and can over-smooth the data and create spurious plane wave activity if too large or generate flow fields dominated by local noise if too small . To assist with the selection of appropriate values of these parameters , the NeuroPatt toolbox can automatically generate simulated datasets with recording size , sampling frequency and oscillation frequency identical to input data , allowing α and β to be optimized based on the user’s data , as shown in Fig 3 . Because the pattern dynamics are typically unknown prior to processing they must be guessed for the simulated data , but we find that in most cases the optimal parameter choices do not change significantly with the pattern types or sizes present . Even within valid velocity fields , the detection of plane waves and synchronized activity is largely dependent upon the thresholds Tpw and Tsyn , which are typically arbitrary parameters set by the user . This is a persistent problem in the detection of such activity patterns: Previous studies have identified plane waves using template matching [24 , 30 , 73] or though alignment statistics [17]; and synchronous activity through correlation or coherence measures [74 , 75] . All these methods rely on largely arbitrary thresholds to explicitly detect patterns . To assist with the choice of these thresholds in our methodology , we implement an optional visual inspection step in the toolbox , allowing users to view sample periods of plane wave and synchronous activity for various thresholds and display distributions of the underlying velocity field statistics across a recording . If such distributions are multi-modal , they can suggest meaningful boundary points for threshold values . To help to validate results , the NeuroPatt toolbox implements surrogate data generation to test results obtained from real data against results obtained from noise with the same basic dynamics as the input data . To achieve this , we construct time series comprising white noise for each recording site with the same mean and standard deviation of the corresponding site in the original data . We then repeat all processing steps including pre-processing , filtering , optical flow estimation and pattern detection with multiple random surrogate datasets and compare identified pattern statistics and dynamics with those obtained from the true data . If results are comparable between the real and surrogate datasets , it suggests that they may have been introduced through smoothing or other processing steps rather than being real effects in the data . This surrogate data testing process will therefore flag most false positive detections made by the toolbox . To ensure that all findings are robust to changes in parameters , we recommend that users verify that their results are consistent across a range of values for key parameters in the pattern detection process . Once simple and complex spatiotemporal patterns have been detected , the evolution dynamics between different pattern types can be quantified . For each pair of pattern types ( pA , pB ) , the observed number of transitions from pA to pB , nobspA→pB , can be counted in each trial by searching for instances where pB starts within a short time gap tgap of pA ending . This can be compared to the expected number of transitions if the patterns in the trial occurred at random times , nexp ( pA→pB ) =nAnBtgapttrial , ( 12 ) where nA and nB are the observed number of patterns pA and pB within the trial and ttrial is the total length of the trial in seconds . The fractional change between observed and expected transition counts is then defined as nobs/nexp-1 . We used paired t-tests with the Bonferroni correction for multiple comparisons to evaluate whether nobs and nexp were significantly different for each pattern transition across multiple trials of a recording . We test the pattern detection procedures in NeuroPatt by generating data sets with known pattern properties . To create a simulated wave pattern zsim with wavenumber k and angular frequency w , we use the formula zsim ( x , y , t ) =A ( x , y , t ) ei ( ks ( x , y , t ) −wt ) , ( 13 ) where A ( x , y , t ) is a function giving the spatial amplitude profile of the wave , and s ( x , y , t ) is a function giving the spatial phase profile . Using different functions for s allows different types of critical point patterns to be generated . All patterns are specified with an initial location ( x0 , y0 ) , and a constant velocity ( vx , vy ) , given in grid spaces per time step , so the location of a pattern at time t is xt , yt= ( x0+vxt , y0+vyt ) . For source or sink patterns , we use ssourcesink ( x , y , t ) = ( x−xt ) 2+ ( y−yt ) 2 , ( 14 ) for spiral patterns we use sspiral ( x , y , t ) = ( x−xt ) 2+ ( y−yt ) 2+1katan2 ( y−yt , x−xt ) , ( 15 ) where atan2 is the multi-valued inverse tangent , and for saddle patterns we use ssaddle ( x , y , t ) =|x−xt|−|y−yt| . ( 16 ) To ensure that all patterns are localized , we define A ( x , y , t ) as a symmetric two-dimensional Gaussian centered on the critical point location: A ( x , y , t ) =A0exp ( − ( ( x−xt ) 2+ ( y−yt ) 2 ) 2c2 ) , ( 17 ) where A0 is the maximum amplitude and c is the Gaussian width parameter . To generate complex datasets , we add multiple wave patterns and then add normally distributed white noise with mean 0 and standard deviation proportional to the amplitude of each grid point . For Fig 3 , we used a 12×12×10 spatiotemporal grid to generate datasets comprising two random critical point patterns , both with parameters w = 2π×0 . 01 and k = 2π/5 . Start positions , velocities , maximum amplitudes and Gaussian width parameters were randomized in each dataset: x0 and y0 were picked uniformly randomly but rejected if patterns were within 2 grid spaces of each other or an edge , vx and vy were picked randomly between −0 . 1 and +0 . 1 , A0 was picked between 1 and 2 and c was picked between 3 and 5 . Pattern detection in simulated data was performed with default parameters . To demonstrate methods in NeuroPatt , we analyze recordings from the middle temporal area of adult male marmosets ( Callithrix jacchus ) . Details of preparation are given previously [49 , 76] . Anesthesia and analgesia were maintained by intravenous infusion of sufentanil citrate ( 6–30 μg kg−1 h−1 ) and inspired 70:30 mix of N2O and carbogen ( 5% CO2 , 95% O2 ) . Dominance of low frequencies ( 1–5 Hz ) in the EEG recording and absence of EEG or electrocardiogram changes under noxious stimulus ( tail pinch ) were taken as the chief signs of an adequate level of anesthesia . Drifts towards higher frequencies ( 5–10 Hz ) in the EEG record were counteracted by increasing the rate of venous infusion or the concentration of anesthetic . The typical duration of a recording session was 48–72 h . Stimuli were presented on a cathode-ray-tube monitor ( Sony G500 , refreshed at 100 Hz , viewing distance 45 cm , mean luminance 45–55 cd m−2 ) , and comprised either a grey screen held at constant luminance for the duration of the recording ( 5–25 minutes , ongoing activity ) , or a pattern alternating every two seconds between a grey screen and a field of drifting circular white dots ( Weber contrast 1 . 0; dot diameter 0 . 4°; drift velocity 20 deg/s ) presented in a large , stationary circular window ( 30° ) . For dot fields , different motion directions ( 90° steps ) were presented pseudo-randomly , and the procedure was repeated until 100 repetitions were made for each of the four directions . Data were recorded using multielectrode arrays ( 10×10 electrodes , 1 . 5 mm length , electrode spacing 400 μm , Blackrock Microsystems ) . Recording surface insertion depth was targeted to 1 mm . To demonstrate pattern dynamics at a larger spatial scale and during waking activity , we also examine optical voltage recordings from awake mice , obtained with permission from Thomas Knöpfel . Details of recording have been described previously [50 , 77 , 78] . Briefly , excitatory neurons in mouse layer 2/3 were targeted with the gene encoding VSFS Butterfly 1 . 2 [79] , and mice were implanted with a head post and thinned skull cranial window . Image acquisition was performed with a dual emission wide-field epifluorescence macroscope during anesthesia induced by pentobarbital sodium ( 40 mg/kg i . p . ) . The data presented here were taken as anesthesia was wearing off , when mice were responsive to touch and exhibited spontaneous whisker and limb movement between recordings . Image sequences of 60 s duration were acquired at 50 Hz temporal resolution and 320 × 240 pixel spatial resolution , with each pixel corresponding to 33 × 33 μm of a projected cortical area . The voltage imaging signal was calculated as the ratio of mKate2 to mCitrine fluorescence after equalization of heartbeat-related fluorescence modulation . The resulting ratiometric sequences of voltage maps were then spatially down-sampled by a factor of 5 using the MATLAB function imfilter , to smooth over noise and reduce the density of calculated velocity fields . | Structured activity such as propagating wave patterns at the level of neural circuits can arise from highly variable firing activity of individual neurons . This property makes the brain , a quintessential example of a complex system , analogous to other complex physical systems such as turbulent fluids , in which structured patterns like vortices similarly emerge from molecules that behave irregularly . In this study , by uniquely adapting techniques for the identification of coherent structures in fluid turbulence , we develop new analytical and computational methods for the reliable detection of a diverse range of propagating wave patterns in large-scale neural recordings , for comprehensive analysis and visualization of these patterns , and for analysis of their dominant spatiotemporal modes . We demonstrate that these methods can be used to uncover the essential spatiotemporal properties of neural population activity recorded by different modalities , thus offering new insights into understanding the working mechanisms of neural systems . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"and",
"materials"
]
| [
"velocity",
"classical",
"mechanics",
"engineering",
"and",
"technology",
"statistics",
"signal",
"processing",
"marmosets",
"vertebrates",
"social",
"sciences",
"neuroscience",
"animals",
"mammals",
"signal",
"filtering",
"animal",
"models",
"primates",
"multivariate",
"analysis",
"singular",
"value",
"decomposition",
"mathematics",
"algebra",
"experimental",
"organism",
"systems",
"vision",
"waves",
"research",
"and",
"analysis",
"methods",
"monkeys",
"animal",
"studies",
"new",
"world",
"monkeys",
"mathematical",
"and",
"statistical",
"techniques",
"principal",
"component",
"analysis",
"physics",
"psychology",
"wave",
"propagation",
"eukaryota",
"linear",
"algebra",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"sensory",
"perception",
"amniotes",
"statistical",
"methods",
"organisms",
"motion"
]
| 2018 | Detection and analysis of spatiotemporal patterns in brain activity |
Nitric oxide ( NO• ) is generated by the innate immune response to neutralize pathogens . NO• and its autoxidation products have an extensive biochemical reaction network that includes reactions with iron-sulfur clusters , DNA , and thiols . The fate of NO• inside a pathogen depends on a kinetic competition among its many targets , and is of critical importance to infection outcomes . Due to the complexity of the NO• biochemical network , where many intermediates are short-lived and at extremely low concentrations , several species can be measured , but stable products are non-unique , and damaged biomolecules are continually repaired or regenerated , kinetic models are required to understand and predict the outcome of NO• treatment . Here , we have constructed a comprehensive kinetic model that encompasses the broad reactivity of NO• in Escherichia coli . The incorporation of spontaneous and enzymatic reactions , as well as damage and repair of biomolecules , allowed for a detailed analysis of how NO• distributes in E . coli cultures . The model was informed with experimental measurements of NO• dynamics , and used to identify control parameters of the NO• distribution . Simulations predicted that NO• dioxygenase ( Hmp ) functions as a dominant NO• consumption pathway at O2 concentrations as low as 35 µM ( microaerobic ) , and interestingly , loses utility as the NO• delivery rate increases . We confirmed these predictions experimentally by measuring NO• dynamics in wild-type and mutant cultures at different NO• delivery rates and O2 concentrations . These data suggest that the kinetics of NO• metabolism must be considered when assessing the importance of cellular components to NO• tolerance , and that models such as the one described here are necessary to rigorously investigate NO• stress in microbes . This model provides a platform to identify novel strategies to potentiate the effects of NO• , and will serve as a template from which analogous models can be generated for other organisms .
NO• is an uncharged , highly diffusible , membrane-permeable metabolite , generated by mammalian NO• synthases ( NOS ) for use in signaling and defense [1] , [2] . The diversity of functions performed by NO• , from pathogen detoxification to vasodilation , reflect its broad reactivity . NO• directly reacts with iron-sulfur ( [Fe-S] ) clusters , superoxide ( O2•− ) , and O2 , whereas its oxidized forms ( NO2• , N2O3 , and N2O4 ) damage thiols , tyrosine residues , and DNA bases [2]–[5] . Such widespread activity has made the biological effects of NO• difficult to predict [2] . For instance , if 1 , 000 NO• molecules entered a cell , what would become of them ? How many would disrupt an [Fe-S] cluster to form a protein-bound dinitrosyl-iron complex ( DNIC ) ? How many would autoxidize to form nitrogen dioxide ( NO2• ) and then react with another NO• to form nitrous anhydride ( N2O3 ) ? How many N2O3 would deaminate DNA bases ? These questions are representative of one unifying , fundamental question of NO• metabolism: how does NO• distribute within a cell ? The answer to this question lies in understanding the kinetic competition of NO• with its many intracellular targets . However , the NO• biochemical network is complex ( Figure 1 ) , contains numerous short-lived intermediates at low concentrations [6] , converges to only a few stable end-products [4] , and involves various damaged biomolecules that are continually digested or repaired [7] . Such complexity has necessitated the use of computational models to both interpret and predict the outcome of NO• treatment [4] . A number of kinetic models have been developed to simulate NO• chemistry in biological contexts [3] , [4] , [6]–[17] . Many of these models have focused on mammalian systems due to the importance of NO• in human physiology . Nalwaya and Deen [9] calculated steady-state concentration profiles of NO• , CO2 , O2•− , and peroxynitrite ( ONOO− ) in idealized mammalian cell cultures using a reaction–diffusion model , and explored the effect of varying the rates and locations ( extracellular , mitochondrial , or cytosolic ) of their generation . Their results suggested negligible spatial variation in species concentrations , and identified conditions under which the different cellular compartments serve as dominant sources or sinks . However , their model did not include the reactions of numerous intracellular metabolites that either directly react with NO• , or its autoxidation products ( NO2• , N2O3 , and N2O4 ) . Lancaster [3] constructed a non-diffusive , but more extensive kinetic model to encompass the complex reaction network of NO• and its autoxidation products with glutathione ( GSH ) and tyrosine in mammalian systems . This model allowed for predictions regarding the relative importance of the various NO•-consuming pathways under inflammatory and non-inflammatory regimes , and highlighted the dominance of oxidative reactions . Lim et al . [4] built upon the work of Lancaster [3] by incorporating additional antioxidants , as well as a separate membrane compartment to account for partitioning of certain species in the lipid-phase . Their model was developed to be representative of inflamed tissue in vivo and used to estimate steady-state intracellular concentrations of different reactive nitrogen species ( RNS ) , in addition to identifying their major sources and sinks in the cytosol and membrane compartments of mammalian cells . Interestingly , none of these models considered the interaction of NO• and its autoxidation products with [Fe-S] clusters , cytochromes , or DNA , and their treatment of the relevant enzymatic processes was limited to NO• dioxygenase and superoxide dismutase . Recently , Tórtora et al . [7] measured rates of ROS- and RNS-induced damage to the mitochondrial aconitase [4Fe-4S] cluster , and incorporated the reactions into a kinetic model of aconitase inactivation in the presence of O2•− and NO• . Since their focus was specifically on the inactivation of aconitase , they did not consider much of the extensive reaction network of NO• , O2•− , and their products . Bagci et al . [14] merged a mitochondrial apoptotic network [18] with a kinetic model of NO• chemistry [10] and extended treatment to include formation of N2O3 , NO2• and ONOO− , as well as their interactions with GSH , non-heme iron , and mitochondrial cytochrome c . However , their attention was primarily on the dynamics of the apoptotic response , and many RNS-related reactions and biological species that were not directly involved in apoptosis were omitted . Though previous models provide a firm foundation for modeling NO• in biological systems , none are sufficiently comprehensive to quantify the distribution of NO• among its many intracellular consumption pathways . Here , we describe the construction , experimental validation , and utility of a comprehensive model of NO• metabolism in Escherichia coli . This model includes NO• autoxidation , enzymatic detoxification , [Fe-S] damage , thiol and tyrosine nitrosation , DNA base deamination , tyrosine nitration , and the repair steps responsible for regeneration of RNS targets . A model of NO• stress with this level of detail has not been previously recognized for any organism . Using this model , we quantitatively explored the distribution of NO• consumption in E . coli , and predicted that the utility of the major aerobic NO• detoxification system ( Hmp ) depends on the NO• delivery rate and extends to environments with O2 concentrations as low as 35 µM ( microaerobic ) . We went on to experimentally confirm these predictions , thereby demonstrating the utility of this model to the study of NO• metabolism . This computational model will serve as a platform to quantitatively interrogate the kinetic competition of NO• with its many targets in E . coli , and assess the influence of various parameters on its distribution .
Upon diffusing into E . coli , NO• may be consumed directly through enzymatic detoxification ( Hmp , NorV , NrfA ) , or reactions with [Fe-S] clusters , O2•− , or O2 ( Figure 1 , Figures S1 , S2 ) . Several resulting nitrosative species , including NO2• and N2O3 , can further react to deaminate DNA bases , nitrosate protein and low molecular weight thiols , and nitrate tyrosine residues . To quantify how NO• distributes within a cell , we have constructed a comprehensive kinetic model of the NO• biochemical reaction network in E . coli , where autoxidation , detoxification ( Hmp , NorV , NrfA ) , [2Fe-2S] and [4Fe-4S] damage and repair , thiol nitrosation and denitrosation , DNA base deamination and repair , enzyme expression and degradation , tyrosine nitration , and reversible cytochrome inhibition are included ( Figure 1 ) . The model consists of 179 reactions , 132 chemical and biochemical species , and 163 kinetic parameters ( Tables S1 , S2 , S3 , Text S1 ) . Of the kinetic parameters , 24 have values that are uncertain , either due to variability or unavailability in literature ( Table S4 ) . An overview describing the construction of the model is presented in the Materials and methods section , whereas a more detailed description has been presented in Text S1 . Due to its scope and completeness , the model is suited to predict the distribution of NO• consumption among the available pathways in E . coli . For example , the fraction of NO• detoxified by Hmp , the amount of NO2• , N2O3 , and ONOO− formed , the quantity of [Fe-S] clusters and DNA bases damaged and repaired , the extent and duration of cytochrome inhibition , and amount of thiols nitrosated can all be calculated from model simulations . Further , the model allows parameter variation ( for example , enzyme mutation/deletion ) and quantification of the impact these alterations have on NO• metabolism . To substantiate the utility of the model , we first validated that the model could reproduce experimentally-measured NO• dynamics and make accurate predictions of experimental outcomes . We sought to validate that the model could capture NO• dynamics in E . coli cultures . Since extracellular NO• loss , including autoxidation and gas phase transport , was non-negligible , we bridged the intracellular model to the experimental system by adding an extracellular ( growth media ) compartment that accounted for autoxidation and gas-phase transport ( Materials and methods ) . Kinetic parameters specific to the extracellular compartment ( NO• delivery rate , NO• and O2 gas phase mass transfer coefficients , and NO• autoxidation rate ) were determined from experimental NO• and O2 measurements in the absence of cells ( Materials and methods , Figure S3 , Text S1 ) . In the experimental system , exponential-phase wild-type E . coli were treated with 0 . 5 mM dipropylenetriamine ( DPTA ) NONOate , and the concentration of NO• in the culture was monitored over time ( Materials and methods ) . The NO• concentration peaked rapidly to 9 . 7 µM following delivery of DPTA , and decreased at a steady rate for ∼0 . 6 hours , after which the concentration dropped quickly to submicromolar levels ( Figure 2A ) . Using a nonlinear least squares optimization algorithm , 39 uncertain parameters ( 24 kinetic constants and 15 species concentrations ) from the model were optimized to capture the experimentally-measured NO• concentration profile ( Materials and methods , Table S4 ) . Uncertain parameters were defined as those that were absent from literature , or those whose literature values had a high degree of variability . All other parameters were either set to their literature values , or measured independently in our experimental apparatus ( Tables S1 , S2 , S3 , Text S1 ) . Given that the optimization algorithm does not guarantee identification of the globally optimal solution , 100 independent sets of random initial parameter values were used ( Materials and methods ) . The optimized parameter set yielding the lowest sum of squared residuals ( SSR ) between the simulated and experimental [NO•] curve is presented in Figure 2A , and demonstrates the model's ability to capture NO• dynamics in a wild-type E . coli culture . For comparison , we took the three most comparable NO• models [3] , [4] , [9] , adapted them to our experimental conditions , and performed an analogous nonlinear least squares optimization in an attempt to capture the NO• dynamics of wild-type E . coli cultures ( Materials and methods ) . As depicted in Figure S4 , none of the alternative models could accurately simulate E . coli NO• dynamics . Quantitatively , the SSRs between the experimental data and the [NO•] curves predicted by the adapted , alternative models of Lim et al . [4] , Lancaster [3] , and Nalwaya and Deen [9] were , respectively , 200- , 200- , and 70-fold greater than that of the model presented here . These data convincingly demonstrate that the model presented here far exceeds current state-of-the-art kinetic models for simulation of microbial NO• metabolism . With an ability to simulate NO• dynamics confirmed , we sought to identify which of the 39 parameters adjusted by the nonlinear optimization procedure were informed by the process , and which had a negligible influence under these conditions . We varied each parameter individually and calculated the corresponding increase in SSR , keeping all other parameters at their optimized values ( Figure 2B ) . The analysis revealed that the Hmp NO• binding ( kHmp , NO•-on ) , and Hmp expression ( kHmp-exp , max and KHmp-exp , NO• ) parameters were the most influential , whereas the oxidation of NorV ( kNorV-O2 ) was of minor significance , but exhibited a greater effect than the remaining parameters , which were all negligible ( less than a 5% increase in SSR ) ( Figure S5 ) . This prompted us to identify the minimum biochemical reaction network necessary to simulate NO• dynamics in aerobic , wild-type E . coli cultures ( Materials and methods ) . As depicted in Table S5 , the model presented here can be simplified to include 17 reactions , 18 chemical and biochemical species , and 14 kinetic parameters without exceeding an overall 5% increase in SSR . While this simplified model can capture the NO• dynamics presented in Figure 2A , we note that it is not suitable for the calculation of additional NO• outcomes , such as the degree of [Fe-S] cluster damage or cytochrome inhibition , and it is not generally translatable to other experimental conditions , such as anaerobic environments . The comprehensive model , on the other hand , can perform such calculations and be applied under many more experimental conditions . The importance of parameters governing Hmp detoxification activity suggested a dominant role for this enzyme in the consumption of NO• under aerobic conditions , a result that is consistent with previous studies of NO• sensitivity in E . coli [19]–[22] . To quantitatively investigate the contribution of Hmp to NO• consumption , we calculated the cumulative , time-dependent distribution ( overall and intracellular ) of NO• for wild-type E . coli treated with DPTA using the optimized parameter values ( Figures 2C and 2D ) . The simulated distributions predicted that autoxidation of NO• in the media accounts for the majority of NO• removal shortly after DPTA addition , with loss to the gas phase comprising most of the remaining flux . By 45 min after delivery , the model predicted that cellular consumption of NO• had accumulated to match that of gaseous loss , and after 1 h became the primary sink . The predicted concentration of NO• dropped rapidly to submicromolar levels at 43 minutes post-dose , where it remained for the duration of the simulation , as Hmp continued to remove NO• as it was released by DPTA . The majority ( 78 . 1% ) of the total NO• released by DPTA was predicted to be consumed by the cells , while autoxidation in the media and loss to the gas phase accounted for 13 . 6% and 8 . 3% of the total NO• consumption , respectively . Virtually all ( 99 . 85% ) of the NO• consumed by the cells was predicted to be through Hmp detoxification , with most of the remaining 0 . 15% through oxidation by O2 and O2•− . Reduction by anaerobic detoxification enzymes ( NorV and NrfA ) and nitrosylation of [Fe-S] clusters was predicted to account for less than 0 . 03% and 0 . 04% of the cellular NO• consumption , respectively . To provide additional experimental evidence in support of these intracellular distributions , we experimentally validated that a mutant lacking the NorV enzyme ( ΔnorV ) consumed NO• at the same rate as wild-type under the experimental conditions tested ( Figure S6 ) . Given the importance of Hmp to the removal of NO• , we assessed the predictive power of the model by determining whether it could accurately predict NO• dynamics in a Δhmp mutant culture . We simulated a Δhmp mutant by fixing the Hmp expression rate to zero . All other model parameter values were left unchanged . As expected , the removal of Hmp was predicted to have a considerable effect on the cells' ability to remove NO• from the environment ( Figure 3A ) . Although the [NO•] curve simulated for the Δhmp culture closely matched that predicted for wild-type at early times ( 0 to ∼10 min ) after DPTA delivery , it started to diverge rapidly as Hmp began to dominate the consumption of NO• in the wild-type culture ( Figure 2C ) . The model predicted that the wild-type culture would quickly consume NO• to reach a submicromolar NO• concentration by 43 min , while the concentration of NO• in the Δhmp culture would gradually decline , requiring over 6 . 4 hours to achieve submicromolar levels . In contrast to wild-type cultures where it was predicted that most NO• would be converted to NO3− by Hmp , the model predicted that the majority of NO• in Δhmp cultures would be converted to NO2− through autoxidation ( Figure 3C ) . To experimentally confirm the Δhmp model predictions , we measured the concentration of NO• in a Δhmp culture after treatment with 0 . 5 mM DPTA under identical conditions as wild-type ( Figure 3B ) . In addition , we measured NO2− and NO3− in the Δhmp culture at 10 h post-dose , when it was predicted that over 99% of the donor had dissociated . The model-predicted NO• concentration curve and final NO2− and NO3− concentrations were in excellent agreement with the experimental data ( without further optimization of any parameters ) ( Figure 3B ) , validating the ability of the model to make accurate predictions regarding major perturbations to the system . To further investigate NO• clearance from the Δhmp culture , we simulated the corresponding intracellular distribution of NO• ( Figure 3D ) . In the Δhmp culture , consumption of NO• by cells was predicted to account for less than 1% of the total NO• delivered , compared to the 78 . 1% for wild-type cells . Over 73% of the NO• that was consumed through intracellular pathways was predicted to be by reaction with O2 or O2•− , while anaerobic enzymatic reduction ( NorV and NrfA ) and [Fe-S] nitrosylation accounted for the remaining 14 . 1% and12 . 6% , respectively ( Figure 3D ) . After validating the model , we sought to identify parameters that control the NO• distribution in E . coli cultures . We focused on experimentally-accessible model parameters to enable experimental validation of predictions . To identify control parameters , we performed a parametric analysis ( Materials and methods ) to assess the effect of varying each parameter on the distribution of NO• . Varied parameters included enzyme concentrations or maximum expression rates , initial concentration and release rate of the NO• donor , O2 concentration in the environment , and intracellular concentrations of GSH , amino acids , and energy metabolites ( Figure 4A , Table S6 ) . In addition to Hmp expression , the parametric analysis revealed NO• donor concentration and release rate , as well as O2 concentration , as important parameters governing the distribution of NO• consumption . Anaerobic NO• detoxification enzymes became the dominant mode of NO• removal within E . coli at lower O2 concentrations due to the loss of Hmp NO• dioxygenase activity , a decrease in the O2-mediated deactivation of NorV , and reduced repression of NrfA expression . The lower O2 concentration also decreased the rate of NO• autoxidation in the media , leaving intracellular reactions and escape to the gas phase as the two primary modes of NO• removal . Although it had little impact on the total NO• distribution , removing superoxide dismutase activity resulted in a small , but noticeable increase in the fraction of intracellular NO• consumed through reaction with O2•− to form ONOO− . Interestingly , the model predicted that higher donor release rates decrease the utility of Hmp in detoxifying NO• ( Figure 4B ) . This decrease can be attributed to the higher NO• concentrations achieved with faster release rates , which in turn enhance substrate inhibition due to the binding of NO• to the Hmp active site before O2 [23] . To further examine the effect of donor release rate on model dynamics , we simulated delivery of NO• to cultures at an increased rate , where Hmp contribution to NO• consumption was predicted to be largely reduced . The initial concentration of donor was maintained at 0 . 5 mM , but the release rate was increased from 1 . 34×10−4 s−1 ( 1 . 4 h half-life , DPTA ) to 1 . 35×10−3 s−1 ( 8 . 6 min half-life ) , the measured rate for the NO• donor propylamine propylamine ( PAPA ) NONOate ( Figure S7 , Text S1 ) . We performed simulations for wild-type and Δhmp cultures , and generated the corresponding NO• concentration profiles ( Figure 4C ) . The strong influence of NO• delivery kinetics on model dynamics are readily apparent when comparing the NO• concentration profiles simulated for PAPA ( Figure 4C ) with those for DPTA ( Figure 3A ) . The faster release rate of PAPA predicted a peak NO• concentration nearly four times that of DPTA ( 34 µM compared to 9 µM , respectively ) , and a large increase in similarity between the simulated wild-type and Δhmp [NO•] curves was observed . Although the predicted NO• concentration in the PAPA-treated wild-type culture dropped rapidly to submicromolar levels at a time similar to that of DPTA ( 37 min and 43 min , respectively ) , Δhmp entered this regime after 1 . 2 h when treated with PAPA , compared to the 6 . 4 h predicted for DPTA . We simulated the corresponding NO• distributions for PAPA-treated cultures to examine the participation of the different pathways in NO• removal . The elevated NO• concentrations simulated for the faster-releasing PAPA greatly increased flux through various consumption pathways , where over 99% of the total NO• consumption was predicted to occur within the first hour after dose for both wild-type and Δhmp ( Figures 5A and 5C , respectively ) . The activity of Hmp , however , was attenuated by the higher NO• concentration due to substrate inhibition ( see Text S1 ) , reducing its ability to participate in detoxification . When Hmp activity was restored and became the most rapid NO• removal pathway after ∼30 minutes , simulation results showed that over 90% of the total NO• had already been consumed through autoxidation and gas transfer pathways ( Figure 5A ) . As a result , the fraction of total NO• consumed by cellular pathways in the wild-type culture was predicted to decrease by nearly 10-fold ( 78 . 1% to 8 . 4% ) due to the increased NO• delivery rate ( compare Figures 2C and 5A ) . When treating with DPTA , the NO• concentration profile and distribution simulated for the Δhmp mutant ( Figure 3B ) were observed to differ greatly from those of wild-type ( Figure 2C ) , but were significantly more similar to wild-type when using PAPA as the donor ( Figures 5A and 5C ) due to the large reduction in Hmp-mediated NO• consumption predicted for the wild-type culture . The intracellular distribution simulated for wild-type treated with PAPA ( Figure 5B ) was still dominated by Hmp , despite its large reduction in activity . However , the proportion of intracellular NO• consumed through pathways other than Hmp was predicted to increase by over 15-fold ( 0 . 15% to 2 . 6% ) upon increasing the NO• delivery rate , suggesting that these other pathways maintain activity while Hmp is inhibited . Thus , the reduction in Hmp activity predicts a 15-fold increase in contribution by other intracellular pathways to the removal of NO• within the cell , including damage to biomolecules such as [Fe-S] clusters . To experimentally validate the prediction that the utility of Hmp decreases as the delivery rate of NO• increases , we measured and compared the ability of wild-type and Δhmp to remove NO• from the culture when dosed with PAPA . We observed excellent agreement between model-predicted and experimentally-measured NO• concentration profiles for the addition of PAPA to wild-type and Δhmp cultures , with no further optimization of model parameters ( Figure 6A ) . The peak concentration of NO• was underestimated by approximately 10% , which was also observed when measuring NO• release from PAPA in media without cells ( Figure S7 ) , suggesting that the disagreement was not associated with cellular parameters . In addition , the rate of NO• clearance by the wild-type cells was slightly overestimated . This could originate from the treatment of Hmp expression in the model , where a more extensive implementation of its governing regulatory network may improve the accuracy of the simulated transcriptional response of hmp expression to elevated levels of NO• . As predicted , the measured difference in time required to remove NO• from the culture between wild-type and Δhmp was small for PAPA ( 0 . 6 h difference in time to reach submicromolar levels ) , highlighting the decreased utility of Hmp under conditions of more rapid NO• release . These results demonstrate that the model can accurately identify parameters that control the distribution of NO• in bacterial cultures , and quantify the impact of their manipulation . In addition to NO• removal from the cell interior and surrounding environment , the model can be used to calculate the extent to which NO• affects various cellular targets , including [Fe-S] nitrosylation [24]–[26] and cytochrome inhibition [27] , [28] . Therefore , we utilized the model to evaluate the protective effect of Hmp with respect to [Fe-S] cluster damage ( Figure 6B ) and cytochrome bd inhibition ( Figure 6C ) . Due to the wide range of reaction rates reported for the nitrosylation of [Fe-S] clusters by NO• ( kNO•-[Fe-S] ) , the parameter value was varied across this range when predicting the extent of [Fe-S] damage . Simulated exposure of wild-type and Δhmp E . coli to 0 . 5 mM DPTA predicted a 2- to 4-fold reduction in the total concentration of [Fe-S] clusters damaged as a result of Hmp activity . When simulations were repeated for PAPA , however , the total [Fe-S] damage predicted for wild-type and Δhmp cultures differed by a maximum of 5% , in agreement with the predicted dependence of Hmp utility on NO• release rate . Furthermore , the duration of NO•-mediated cytochrome bd inhibition following DPTA treatment was predicted to greatly increase for the Δhmp culture relative to wild-type , requiring over 9 h ( compared to 0 . 7 h for wild-type ) for the concentration of NO•-bound cytochromes to drop below 50% of the total . Treatment with PAPA resulted in more similar cytochrome inhibition between the strains , with durations of 0 . 6 h and 1 . 5 h predicted for wild-type and Δhmp , respectively . Collectively , the results from these damage descriptors , in addition to the rate and distribution of NO• consumption , predicted a greater similarity in recovery from bacteriostasis between wild-type and Δhmp when treated with PAPA than with DPTA . To test the prediction , we monitored the optical density ( OD600 ) of each strain following treatment with 0 . 5 mM DPTA or PAPA ( Figure 6D ) . In agreement with the prediction , the duration of NO•-induced stasis was more similar between wild-type and Δhmp strains when using a faster NO• donor . Growth inhibition of Δhmp following PAPA treatment was less severe than that observed for DPTA , where cells exited stasis less than 2 h after wild-type , compared to over 10 h for DPTA . Hmp is considered the major aerobic enzyme responsible for NO• detoxification [19] , [21] , [29] , whereas NorV is considered the major anaerobic detoxification system [22] , [30] , [31] . Surprisingly , the parametric analysis suggested that Hmp remains dominant at O2 concentrations as low as 25 µM ( ∼14% air saturation [32] ) ( Figure 4A ) . To experimentally confirm that this was the case , we adjusted the experimental setup by adding N2-bubbling at a rate of 1 ml/s . In the presence of wild-type E . coli at an OD600 of 0 . 05 , an O2 concentration of 35 µM was achieved and maintained constant throughout the time course of a DPTA experiment ( Figure S8 ) . This concentration was over 5-fold less than air-saturated media ( 185 µM ) , but also above the 25 µM used in the parametric analysis . Due to the adjustment in experimental conditions , the model was similarly optimized for NO• dynamics from microaerobic wild-type E . coli cultures ( Material and methods ) , and found to capture the data very well ( Figure 7A ) . We note that N2-bubbling increased fluctuations in the NO• measurements , but the increased error was minor compared to the range of NO• concentrations investigated . Using the optimized model , we predicted the effect of genetic deletions of norV and hmp on the NO• dynamics . Consistent with the previous parametric analysis , NorV was identified as a negligible consumption pathway under microaerobic conditions ( 35 µM O2 ) , whereas Hmp was identified as the major NO• sink . These predictions were experimentally validated , and the results are presented in Figures 7B and 7C . These data demonstrate that the model is useful for studying sub-aerobic environmental conditions , and that the switch between Hmp-dominated and NorV-dominated NO• consumption regimes occurs at very low O2 concentrations . The corresponding extracellular and intracellular NO• distributions for wild-type , Δhmp , and ΔnorV under microaerobic conditions were simulated , and are presented in Figure 7 . Loss of NO• to the gas phase was predicted to largely increase for all strains ( 26% and 27% of the total NO• consumption for wild-type and ΔnorV , respectively , and 95% for Δhmp ) , due to the increased air-liquid surface area caused by the bubbling of N2 through the culture , as well as the reduced rate of autoxidation . Autoxidation was predicted to have negligible NO• consumption activity compared to the cellular and gas transport pathways ( 0 . 5% of the total for wild-type and ΔnorV , and 1 . 2% for Δhmp ) , owing to the reduced O2 concentration , as well as the lower peak NO• concentration ( ∼4 . 5 µM for all strains ) than was achieved under aerobic , non-bubbling conditions using DPTA ( ∼8–10 µM ) . Cellular consumption of NO• was still predicted to be the greatest sink of NO• for the wild-type and ΔnorV strains ( accounting for 74% and 73% of the total consumption , respectively ) , but only a minor pathway in the Δhmp culture ( 3 . 8% ) . The intracellular distributions ( Figure 7 ) for wild-type and ΔnorV cultures were still predicted to be dominated by Hmp detoxification ( both exceeding 98% of intracellular NO• consumed by Hmp ) , as was seen under aerobic conditions . The NO• consumed by Δhmp cells , however , was now predicted to occur primarily through NorV reduction ( 93% of the intracellular NO• ) , compared to the 14% contribution predicted for Δhmp in aerobic conditions . Overall , the simulation results predicted Hmp to be the primary mode of NO• consumption under O2 concentrations as low as 35 µM , but suggested an increased role of NorV reduction in the event that Hmp detoxification becomes unavailable .
NO• is a critical antimicrobial of the innate immune response whose utility originates from its ability to diffuse through cellular membranes [33] , deactivate bacterial enzymes [26] , inhibit respiration [28] , and react with O2 and O2•− to yield the reactive nitrogen species , NO2• , N2O3 , N2O4 , and ONOO− [24] . The biochemical reaction network of NO• includes both spontaneous and enzymatic reactions involving many short-lived species that decompose to several common end-products [4] . Increasing the complexity of this system is the continuous degradation and repair of damaged biomolecules , which regenerates targets for NO• and its reactive intermediates [34] . A quantitative description of how NO• distributes among these many pathways is critical to understanding immune function and pathogenesis , as well as to designing NO•-based and NO•-synergizing therapeutics [35]–[37] . However , the complexity of the NO• reaction network renders exhaustive experimental monitoring infeasible , and interpretation of measurements difficult [4] , [6] . To address these challenges , experimentally-informed computational models are required to explore the NO• reaction network . Though several kinetic models have been developed to study the chemistry of NO• in biological systems , of which the majority are mammalian , none have had sufficient breadth and depth to address the full range of effects of NO• exposure [1] , [38] . The model presented here is far more comprehensive than those constructed previously , incorporating the damage , modification , and repair of biomolecules , as well as enzymatic detoxification and transcriptional control . These functionalities allow focused investigation of intracellular components of the NO• network , such as [Fe-S] cluster and DNA damage , but also culture-wide prediction of the NO• distribution . We validated the utility of the model by demonstrating that it can reproduce NO• dynamics in a bacterial culture , make accurate predictions regarding large perturbations to the system , and identify parameters that control the distribution of NO• in bacterial cultures . Specifically , model simulations predicted that NO• autoxidation and Hmp-catalyzed detoxification were the primary sinks for NO• consumption in aerobic wild-type E . coli cultures . Oxidation of NO• has been shown in the past to be a major contributor to the consumption of NO• under certain conditions [3] , and the dominant role of Hmp in aerobic detoxification is in agreement with previous studies that have demonstrated its importance in tolerating NO• stress [21]–[23] . In addition , we used the model to ( 1 ) uncover a novel dependency of Hmp utility on the NO• delivery rate , and ( 2 ) discover that Hmp is the dominant cellular NO• detoxification system at dissolved O2 concentrations as low as 35 µM ( microaerobic ) . Both of these predictions were validated experimentally , thereby demonstrating the utility of the model for the study of NO• metabolism . Specifically , when treated with a fast-releasing NO• donor ( PAPA ) , the consumption of NO• and recovery from bacteriostasis was far more similar between wild-type and Δhmp E . coli than with a slower NO• donor ( DPTA ) . This effect arises from substrate inhibition of the Hmp active site caused by high NO•/O2 concentration ratios and the time required to synthesize Hmp [23] . An effect of NO• delivery on its toxicity has been observed previously in a mammalian system [39] , [40] , where it was shown that killing of human lymphoblastoid cells ( TK6 and NH32 ) was a function of both NO• concentration and cumulative dose . Here , we have demonstrated an influence of NO• delivery rate on the dynamics of NO• consumption and recovery in bacterial cultures , and also offered a detailed , mechanistic description of the observed dependence . In addition , we discovered that Hmp remains the major cellular detoxification system at dissolved O2 concentrations as low as 35 µM . This effect originates from the strong induction of Hmp expression upon NO• exposure even under anaerobic conditions [41] , [42] , and the rapid O2-mediated deactivation of NorV , the alternative NO• detoxification system that has been previously identified as critical for resisting NO• stress under anaerobic conditions [22] , [31] . These data demonstrate the flexibility of this method to different environmental conditions ( microaerobic ) , and provide support for the role of Hmp as a virulence factor [43] , [44] , since O2 concentrations at infections sites/in macrophages and neutrophils are typically hypoxic ( less than 50 µM O2 [4] , [45] , [46] ) . Interestingly , both NorV- and Hmp-type enzymes have been found to be virulence factors for numerous organisms [36] , [47]–[50] , and thus a quantitative understanding of the conditions under which each contributes to NO• clearance would be valuable for the study of their importance to virulence . The work presented here demonstrates the predictive accuracy and utility of a comprehensive model of NO• metabolism in E . coli . The scope of this model allows for detailed , quantitative exploration of numerous NO• network features and environmental conditions , including future investigations of the roles of O2 concentration and indirect NO• delivery , such as that observed for S-nitrosothiols [41] . Further , this model will prove useful for the optimization of NO•-synergizing and NO•-based therapeutics , which are being investigated as antibiotic alternatives for the treatment of both gram-positive and gram-negative infections , including those caused by Mycobacterium tuberculosis , Staphylococcus aureus , Pseudomonas aeruginosa , E . coli , and Acinetobacter baumannii [35]–[37] . Such therapies include NO•-releasing nanoparticles [36] , NO•-releasing dressings [37] , and rhodanines , which kill non-replicating mycobacteria through the potentiation of host-derived NO• [35] . Interestingly , the study by Sulemankhil and colleagues identified NO• release rate and dosage as important parameters governing the effectiveness of the examined dressings . The modeling approach presented here could provide a more quantitative understanding of how these potential therapeutics neutralize pathogens , and would prove useful for identifying methods to increase their potency through the quantitative identification of the NO• distribution pathways used by specific organisms . To achieve this potential utility , the modeling method described here must be adapted for use in organisms other than E . coli . To do this , the enzymatic reactions within the model would need to be removed , replaced , or augmented based on the systems harbored by the pathogen of interest , and uncertain parameters would need to be identified by training the model on experimental data , as performed here . In the event that an important reaction is missing from a model , stable NO• end products ( such as NO2− and NO3− ) would be measured and both metabolic databases and the organism's genome would be mined for model additions capable of capturing the experimental data . Potential additions would then be experimentally validated by measuring in vitro kinetics of samples purified from cultures of interest . Execution of these steps will produce models of NO• metabolism in pathogens , that will mirror utility and capabilities achieved by the kinetic platform described here .
All strains used in this study were E . coli K-12 MG1655 . The Δhmp and ΔnorV mutants were obtained from the Keio collection [51] , and transferred into the MG1655 background using the P1 phage method . Proper chromosomal integration and absence of gene duplication were checked by PCR . The hmp primers used were 5′-CCGAATCATTGTGCGATAACA-3′ ( forward ) and 5′-ATGATGGATACTTTCTCGGCAGGAG-3′ ( reverse ) for accurate integration , and 5′- TCCCTTTACTGGTGGAAACG-3′ ( forward ) and 5′-CACGCCCAGATCCACTAACT-3′ ( reverse ) for gene duplication . The norV primers used were 5′-CCAGCACATCAACGGAAAAA-3′ ( forward ) and 5′-ATGATGGATACTTTCTCGGCAGGAG-3′ ( reverse ) for accurate integration , and 5′-GACTGGGAAGTGCGTGATTT-3′ ( forward ) and 5′-CGGAAGCGTAAACCAGTCAT-3′ ( reverse ) for gene duplication . NO• donors ( Z ) -1-[N- ( 3-aminopropyl ) -N- ( 3-ammoniopropyl ) amino]diazen-1-ium-1 , 2-diolate ( DPTA NONOate ) and ( Z ) -1-[N- ( 3-aminopropyl ) -N- ( n-propyl ) amino]diazen-1-ium-1 , 2-diolate ( PAPA NONOate ) were purchased from Cayman Chemical Company . All other chemicals and reagents were purchased from Sigma Aldrich or Fisher Scientific , unless otherwise noted . E . coli from a frozen −80°C stock were inoculated into 1 ml of fresh LB broth and grown for 4 hours at 37°C and 250 r . p . m . 10 µl of the LB culture were used to inoculate 1 ml of MOPS minimal media ( Teknova ) containing 10 mM glucose . The minimal glucose culture was grown at 37°C and 250 r . p . m . overnight ( 16 h ) and used to inoculate 20 ml fresh MOPS glucose ( 10 mM ) in a 250 ml baffled shake flask to a final OD600 of 0 . 01 . The flask culture was grown at 37°C and 250 r . p . m . to exponential phase ( OD600 = 0 . 2 ) , at which point 4 ml was transferred to separate microcentrifuge tubes in 1 ml aliquots and centrifuged at 15 , 000 r . p . m . for 3 min at 37°C . To remove the culture media , 980 µl of the supernatant was removed and cell pellets were resuspended in 1 ml of pre-warmed ( 37°C ) 10 mM MOPS glucose media . Samples were combined in a 15 ml Falcon tube , and returned to the shaker ( 37°C , 250 r . p . m . ) . After 5 minutes , the resuspended culture was diluted to an OD600 of 0 . 03 in fresh , pre-warmed ( 37°C ) MOPS glucose ( 10 mM ) in a 50 ml Falcon tube with a final culture volume of 10 ml . The culture was stirred with a sterilized magnetic stirring bar , and immersed in a stirred water bath to maintain the temperature at 37°C . Growth was monitored until the OD600 reached a value of 0 . 05 ( approximately 45 minutes after diluting to OD600 of 0 . 03 ) , at which time the NO• donor solution ( DPTA or PAPA ) was added . On the day of use , the NONOate powder was dissolved in a chilled ( 4°C ) , sterile solution of 10 mM NaOH in deionized H2O , and stored on ice prior to delivery . After NONOate delivery , every half-hour ( DPTA ) or twenty minutes ( PAPA ) , 75 µl aliquots were removed to measure the OD600 ( Synergy H1 Microplate Reader , BioTek Instruments , Inc . ) . The concentration of NO• in the culture was monitored continuously over the course of the experiment using an ISO-NOP NO• sensor ( World Precision Instruments , Inc . ) . The electrode was calibrated daily , prior to use , according to the manufacturer's specifications . The microaerobic NO• consumption assay was performed using the same procedure , except N2 bubbling was included to reduce the dissolved O2 concentration . Immediately following the dilution of cells to an OD600 of 0 . 03 in the 50 ml Falcon tube , N2 gas ( 99 . 998% pure ) was bubbled into the culture through a sterile pipet tip at a constant flow rate of 1 ml/s . The O2 concentration was observed to drop quickly and stabilize at approximately 19% air saturation ( 35 µM ) within 15 minutes of initiating the N2 bubbling , where it remained for the duration of the experiment . The concentration of O2 was monitored continuously to ensure stable conditions throughout the assay ( Figure S8 ) . The concentration of dissolved O2 was measured using the FireStingO2 fiber-optic O2 meter with the OXF1100 fixed needle-type minisensor ( PyroScience GmbH ) . The sensor was calibrated according to the manufacturer's specifications , and the signal was automatically compensated for temperature fluctuations using the TDIP15 temperature sensor ( PyroScience GmbH ) during all O2 measurements . The concentration of NO2− and NO3− were measured using the Nitrate/Nitrite Colorimetric Assay Kit from Cayman Chemical Company , following the manufacturer's instructions . Briefly , Griess reagents were added to diluted samples to convert the NO2− to a purple azo compound , and quantified by measuring the absorbance at 540 nm using a microplate reader [52] . A calibration curve was generated using varying dilutions of a standard NO2− solution . The total NO2−+NO3− concentration of the samples were obtained by first converting the NO3− to NO2− using nitrate reductase , and then treating with Griess reagents . The NO3− concentration in the samples was calculated as the difference between the total NO2−+NO3− concentration and the NO2− concentration . All samples were measured in triplicate . All simulation calculations were performed using Matlab ( R2012a ) . The governing set of differential mass balances was integrated using the stiff numerical ODE integrator ( ode15s function ) . Optimization of model parameters was performed in Matlab using the lsqcurvefit function , which solves nonlinear least-squares minimization problems . Through an iterative process , the function identified parameter values yielding the lowest sum of squared residuals ( SSR ) between the experimentally-measured and model-simulated NO• concentration profiles . Since the nonlinearity of the minimization problem gives rise to local minima , we performed 100 independent optimizations , each initialized with a random set of parameter values ( within their allowed range ) . The parameter optimization procedure was used to determine the values of extracellular parameters specific to our experimental system: NO• donor dissociation ( kNONOate ) , transfer of NO• to the gas phase ( kLaNO• ) , and the rate of NO• autoxidation ( kNO•-O2 ) . Cell-free growth media was treated with 0 . 5 mM DPTA under conditions identical to the aerobic NO• consumption assay , and the resulting NO• concentration profile and final ( 10 h ) NO2− and NO3− concentrations were measured . The optimization yielded values of 1 . 34×10−4 s−1 ( 1 . 4 h half-life ) , 4 . 74×10−3 s−1 , and 1 . 80×106 M−2s−1 for kNONOate , kLaNO• , and kNO•-O2 , respectively ( see Text S1 for further detail ) . Figure S3 demonstrates excellent agreement between the predicted and measured [NO•] curve and final NO2− and NO3− concentrations when using the optimized parameter values . Of the cellular-related model parameters , 39 were classified as uncertain due to variability or unavailability in literature ( Table S4 ) . A parameter optimization was conducted to identify the set of parameter values yielding the lowest SSR between the simulated and experimentally-measured NO• concentration profile resulting from the addition of 0 . 5 mM DPTA to an aerobic , exponential-phase culture of wild-type E . coli . The predicted [NO•] curve using the optimal parameter set was in excellent agreement with the experimental data ( Figure 2A ) . For the microaerobic ( 35 µM O2 ) NO• consumption assay , uncertain parameters were re-optimized for the low-O2 environment due to expected changes in cellular properties and the effect of N2 bubbling on gas transfer rates . We note that differences in N2 bubble properties ( such as bubble size and lifetime ) caused by the presence of cells prevented the use of cell-free NO• measurements in determining extracellular parameters for this experimental setup . Instead , the simulated O2 concentration was fixed to 35 µM based on experimental observations ( Figure S8 ) , and the remaining extracellular and uncertain parameters ( total of 42 parameters ) were simultaneously optimized to best capture the NO• concentration curve measured for wild-type cells treated with DPTA under microaerobic conditions ( Table S7 ) . The optimal set of parameter values was able to accurately capture the experimentally-measured NO• dynamics in the microaerobic environment ( Figure 7 ) . An individual parametric analysis of the 42 optimized parameters was performed to determine those that had a significant impact ( greater than 5% increase in SSR ) on the predicted [NO•] curve in the microaerobic environment ( Figure S9 ) . Similar to the aerobic parametric analysis , Hmp-associated parameters ( kHmp , NO•-on , kHmp-exp , max , and KHmp-exp , NO• ) were found to strongly influence the predicted NO• dynamics . Parameters governing the rate of NONOate dissociation ( kNONOate ) and NO• transfer to the gas phase ( kLaNO• ) also demonstrated substantial control of the [NO•] curve upon variation . Finally , NorV expression ( kNorV-exp , max and KNorV-exp , NO• ) and inactivation ( kNorV-O2 ) parameters were found to have a significant impact on the SSR . Parametric analyses were used to evaluate the influence of model parameters on either the simulated [NO•] curve or the predicted distribution of NO• consumption in the culture . The effect of parameter variation on the [NO•] curve was quantified by the resulting change in SSR between the model-simulated and experimentally-measured NO• concentration profiles . Specifically , parameters were individually varied among 100 evenly-spaced points spanning their allowed range , and the resulting SSR at each parameter value was calculated . The effect of parameter variation on the SSR for aerobic ( Figure 2B , Table S4 ) and microaerobic ( Figure S9 , Table S7 ) wild-type E . coli cultures was evaluated . To quantify the effect of varying experimentally-accessible parameters on the predicted distribution of NO• consumption , parameters were individually varied among five logarithmically-spaced values spanning their permitted range ( Table S6 ) . Simulations were run for each different parameter set , and the final distribution of NO• consumption among the available pathways ( such as autoxidation , transport to the gas phase , Hmp-mediated detoxification , and [Fe-S] damage ) was calculated ( Figure 4A ) . Three existing models of NO• chemistry , developed by Lim et al . [4] , Lancaster [3] , and Nalwaya and Deen [9] , were individually assessed for their ability to simulate NO• dynamics in a culture of wild-type E . coli . The alternative models were constructed and adapted to our experimental system using the following procedure . Starting with the model presented in this study , all reactions absent in the alternative model were eliminated , except the release of NO• from a NONOate , and the NO• and O2 liquid-gas transport reactions . Reactions present in the alternative model that were not included in the present model ( due to the consumption or production of an unknown or nonspecific species , or the simplification of a more complex process in the present model ) were added to the adapted model . For Lancaster's model , the NO• formation and disappearance reactions , as well as the disappearance of NO2• and •OH , were not included because the rates of the disappearance reactions are user-defined , and the formation of NO• is accounted for by the NONOate dissociation reaction . The model described by Nalwaya and Deen contains a simplified reaction representing the consumption of NO• by a heme- and flavin-dependent dioxygenase , analogous to Hmp detoxification in E . coli . The reaction was included in the adapted model , and the associated bimolecular rate constant was allowed to vary during parameter optimization . Additionally , the rate parameters governing ONOO− and ONOOH reactions used by Nalwaya and Deen were adjusted for a pH of 7 . 6 , where the fraction of ONOO− in protonated form was calculated to be 12% [9] . Although Nalwaya and Deen do not include NO• autoxidation in their model , it was incorporated into the adapted version , as autoxidation is an important effect under the aerobic experimental conditions used in this study . Species concentrations in the alternative models were set to the same values or ranges used in the present model , except for a few minor differences . The concentrations of proteins and transition metal centers ( Mn+ ) in the model of Lim et al . were allowed a range of 5–8 mM and 1–500 µM , respectively . The protein concentration range was selected based on typical protein content reported for E . coli [119] , while Mn+ was allowed the same concentration range as [Fe-S] clusters in the present model , which assumes ∼5% of proteins contain [Fe-S] clusters [120] . The three adapted models were subjected to a parameter optimization procedure analogous to that used for the model presented here ( see “Parameter optimization” section above ) , where parameters classified as uncertain were varied to minimize the SSR between the predicted and experimental [NO•] curves . Ultimately , none of the three adapted models were able to capture the dynamics of NO• measured in wild-type E . coli cultures , yielding [NO•] curves with SSR values that were 200-fold ( Lim et al . and Lancaster ) and 70-fold ( Nalwaya and Deen ) greater than the SSR achieved by the present model ( Figure S4 ) . In order to identify the core set of reactions required to accurately simulate NO• dynamics in aerobic wild-type E . coli cultures ( Figure 2A ) , a systematic reduction of the model reaction network was performed using a two-tier process . In the first tier , reactions were sequentially deleted from the original network in a random order . After each reaction deletion , the SSR between the simulated and experimentally-measured [NO•] curve for DPTA-treated wild-type E . coli was calculated . If the SSR exceeded a 5% increase over the original SSR , the reaction deletion was undone . This process was repeated until no remaining reactions could be removed without exceeding the 5% increase in SSR . The entire model reduction process was repeated for a total of 100 iterations , each following a random sequence of reaction deletions . The reduced reaction network was selected as the set containing the least number of reactions . In the event of two or more minimum sets , the network yielding the lowest SSR was chosen . In the second tier , the minimal reaction network was further reduced through a similar reaction deletion process , except with the inclusion of a parameter optimization step . After deleting a reaction , any remaining parameters in the reduced model classified as uncertain ( Table S4 ) were re-optimized , following the nonlinear least-squares optimization procedure described above . If the optimization succeeded in decreasing the SSR to within 5% of the original SSR value , the reaction was removed from the final network . The final , minimum biochemical reaction network determined through this process is presented in Table S5 . | Nitric oxide ( NO• ) is a highly reactive metabolite used by immune cells to combat pathogens . Since the biological effects of NO• are governed by its broad reactivity , it is desirable to determine how NO• distributes among its many targets inside a cell . A quantitative understanding of this distribution and how it is controlled will facilitate the development of novel NO•-potentiating therapeutics . Here , we have constructed and experimentally validated a comprehensive kinetic model of NO• biochemistry within Escherichia coli that includes NO• autoxidation , respiratory inhibition , enzymatic detoxification , and damage and repair of biomolecules . Using this model , we investigated the control of NO• dynamics in E . coli cultures , and found that the primary aerobic detoxification system , NO• dioxygenase ( Hmp ) , functions as a dominant NO• consumption pathway under microaerobic conditions ( 35 µM O2 ) , and loses utility as the NO• delivery rate increases . We confirmed these predictions experimentally , thereby demonstrating the predictive power of the model . This model will serve as a quantitative platform to study nitrosative stress , provide a template from which models for other organisms can be generated , and facilitate the development of antimicrobials that synergize with host-derived NO• . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"bacteriology",
"microbial",
"metabolism",
"neurochemistry",
"small",
"molecules",
"enzymes",
"metabolic",
"networks",
"microbiology",
"host-pathogen",
"interaction",
"bacterial",
"biochemistry",
"escherichia",
"coli",
"biological",
"systems",
"engineering",
"prokaryotic",
"models",
"model",
"organisms",
"enzyme",
"kinetics",
"bacterial",
"pathogens",
"bioengineering",
"biochemistry",
"simulations",
"neurochemicals",
"nitric",
"oxide",
"metabolic",
"pathways",
"biology",
"systems",
"biology",
"biochemistry",
"biochemical",
"simulations",
"immunity",
"innate",
"immunity",
"computational",
"biology",
"metabolism",
"engineering"
]
| 2013 | A Kinetic Platform to Determine the Fate of Nitric Oxide in Escherichia coli |
Japanese encephalitis virus ( JEV ) is the causative agent of Japanese encephalitis , the leading cause of viral encephalitis in Asia . JEV transmission cycle involves mosquitoes and vertebrate hosts . The detection of JEV RNA in a pool of Culex pipiens caught in 2010 in Italy raised the concern of a putative emergence of the virus in Europe . We aimed to study the vector competence of European mosquito populations , such as Cx . pipiens and Aedes albopictus for JEV genotypes 3 and 5 . After oral feeding on an infectious blood meal , mosquitoes were dissected at various times post-virus exposure . We found that the peak for JEV infection and transmission was between 11 and 13 days post-virus exposure . We observed a faster dissemination of both JEV genotypes in Ae . albopictus mosquitoes , when compared with Cx . pipiens mosquitoes . We also dissected salivary glands and collected saliva from infected mosquitoes and showed that Ae . albopictus mosquitoes transmitted JEV earlier than Cx . pipiens . The virus collected from Ae . albopictus and Cx . pipiens saliva was competent at causing pathogenesis in a mouse model for JEV infection . Using this model , we found that mosquito saliva or salivary glands did not enhance the severity of the disease . In this study , we demonstrated that European populations of Ae . albopictus and Cx . pipiens were efficient vectors for JEV transmission . Susceptible vertebrate species that develop high viremia are an obligatory part of the JEV transmission cycle . This study highlights the need to investigate the susceptibility of potential JEV reservoir hosts in Europe , notably amongst swine populations and local water birds .
Japanese encephalitis is one of the major viral encephalitides in Asia , with an estimated 68 , 000 human cases per year [1] . Up to 30% of the symptomatic cases are fatal , and long-term neurologic sequelae can occur in 30 to 50% of survivors [2] . Japanese encephalitis virus ( JEV ) is the causative agent of Japanese encephalitis , and is transmitted through the bite of an infected mosquito . JEV is a member of the Flavivirus genus in the Flaviviridae family and has a positive-sense RNA genome . The viral polyprotein is processed into 10 proteins: three structural proteins and seven nonstructural proteins . JEV strains can be differentiated into five genotypes ( 1 to 5 ) based on phylogenetic studies of the viral envelope protein sequences . Until recently , most of the strains of JEV at the origin of major epidemics in the South , East and Southeast Asia regions belonged to genotype 3 [3 , 4] . Recently a shift in prevalence from JEV genotype 3 to 1 has been observed in several Asian countries [5–7] . JEV genotype 5 was first isolated in Malaysia in 1952 , and is genetically and serologically distinct from other genotypes [8–10] . No other JEV genotype 5 strain had been identified until its recent isolation from Culex spp . mosquito pools in China in 2009 [11] and in South Korea in 2010 and 2012 [12 , 13] . Most of the vectors for JEV belong to the Culicinae subfamily in the Culicidae family . In most Asian countries , the main vector is Culex tritaeniorhynchus [7 , 14–18] , while Cx . annulirostris was identified as the main vector for JEV transmission in Australia [19 , 20] . Several secondary vectors are known to efficiently transmit JEV: Cx . annulirostris , Cx . annulus , Cx . fuscocephala , Cx . gelidus , Cx . sitiens or Cx . vishnui . The fact that JEV can be detected in field-caught mosquitoes belonging to numerous species , such as Cx . pipiens [12 , 17 , 21] , Aedes albopictus [7 , 22] , or Anopheles species [7 , 23] , poses the question if those mosquito species could also act as secondary vectors for JEV . The JEV enzootic cycle involves mosquitoes and amplifying vertebrate hosts , such as water birds and domestic swine [24] . Humans are considered as dead-end hosts , while they can be infected by JEV , they do not develop high levels of blood viremia , and thus cannot infect mosquitoes [25] . A fragment of JEV genome was detected in a pool of Cx . pipiens and in birds caught in 2000 and 2010 in Northern Italy [21 , 26] raising the threat of a putative emergence of the virus in Europe [27] . Recent studies have shown that Ae . detritus from England and Ae . japonicus japonicus from Germany were competent to transmit JEV [28 , 29] . These observations emphasize on the need to study the vector competence of European mosquito populations for JEV . Ae . albopictus is currently expanding its range , predominantly in temperate areas in North America and Europe , and this invasion raises a public health threat for pathogens transmitted by this vector , such as Zika and dengue viruses . Cx . pipiens is the most widely distributed species of mosquito in the world , and is typically found in temperate regions . Cx . pipiens complex mosquitoes play important roles in the transmission of several medically relevant pathogens such as West Nile virus ( WNV ) , Saint Louis encephalitis virus , and filarial worms [30–32] . In the present study , we evaluated the competence of Ae . albopictus and Cx . pipiens populations collected in the South of France for two representative strains of JEV , belonging to distinct genotypes . We found that both viruses could infect and disseminate to high efficiency in either vector and could be readily transmitted . We additionally evaluated the influence of mosquito salivary factors on viral pathogenesis and showed that they had no impact on the development of Japanese encephalitis in a mouse model for the disease . Overall , these findings highlight the need for investigation of the other factors that could contribute to JEV emergence in Europe .
The protocols and subsequent experiments were ethically approved by the Ethic Committee for Control of Experiments on Animals ( CETEA ) at the Institut Pasteur and declared to the French Ministère de l’Enseignement Supérieur et de la Recherche ( n° 000762 . 1 ) in accordance with European regulations . Experiments were conducted following the guidelines of the Office Laboratory of Animal Care at the Institut Pasteur . Euthanasia was performed by CO2 asphyxiation , followed by cervical dislocation . Anesthesia was performed by intraperitoneal injection of a mixture of Xylazine ( Rompun , 5 to 10 mg/kg ) and Kétamine ( Imalgène , 80 to 100 mg/kg ) . Cx . pipiens form pipiens and Ae . albopictus mosquito colonies were established in the laboratory using mosquitoes collected in Montpellier and Nice , in 2010 and 2011 , respectively . Eggs of each mosquito colony were hatched in tap water . Larvae were reared in plastic trays containing tap water supplemented with brewer’s yeast tablets and cat food . Adults were maintained at 27°C , 80% relative humidity with a 12 h:12 h light: dark cycle and were given continuous access to 10% sucrose solution . Mosquito Ae . albopictus C6/36 cells were maintained at 28°C in Leibovitz medium ( L15 ) supplemented with 10% heat-inactivated fetal bovine serum ( FBS ) . Baby hamster kidney-derived BHK-21 ( purchased from ATCC ) , chicken fibroblast-derived DF-1 ( obtained from Nadia Naffakh ) , and human kidney-derived HEK293T cells ( purchased from ATCC ) were maintained at 37°C in DMEM supplemented with 10% FBS . Mouse hybridomas producing the monoclonal antibody 4G2 anti-Flavivirus E were purchased from ATCC and a highly purified antibody preparation was produced by RD Biotech . The anti-mosquito saliva antibody was produced in house in rabbits exposed to mosquito bites . Horseradish peroxidase ( HRP ) -conjugated goat anti-mouse and anti-rabbit IgG antibodies were obtained from Bio-Rad Laboratories . Alexa Fluor 488-conjugated goat anti-mouse IgG antibody was obtained from Jackson ImmunoResearch . A molecular cDNA clone of JEV genotype 3 strain RP-9 was kindly provided by Yi-Lin Ling and was modified as described previously [33] . A molecular cDNA clone of JEV genotype 5 strain XZ0934 was described previously [33] . To produce infectious virus , the molecular clones were transfected into HEK293T cells using Lipofectamine 2000 ( ThermoFischer Scientific ) . At 3 days post-transfection , viral supernatants were collected and used to infect DF-1 cells in order to grow final virus stocks for experiments . For infections , C6/36 cells were seeded in 24-well tissue culture plates in L15 , supplemented with 2% FBS . Aliquots of virus were diluted in 200 μl of medium and added to the cells . Plates were incubated for 1 h at 28°C . Unadsorbed virus was removed by two washes with Dulbecco's phosphate-buffered saline ( DPBS ) and then 1 ml of L15 supplemented with 2% FBS was added to the cells , followed by incubation at 28°C until collection . BHK-21 cells were seeded in 24-well plates . Tenfold dilutions of virus samples were prepared in duplicate in DMEM and 200 μl of each dilution was added to the cells . The plates were incubated for 1 h at 37°C . Unadsorbed virus was removed , after which 1 ml of DMEM supplemented with antibiotics and antifungals , 1 . 6% carboxymethyl cellulose ( CMC ) , 10 mM HEPES buffer , 72 mM sodium bicarbonate , and 2% FBS was added to each well , followed by incubation at 37°C for 32 h . The CMC overlay was aspirated , and the cells were washed with PBS and fixed with 4% paraformaldehyde for 15 min , followed by permeabilization with 0 . 1% Triton-X100 for 5 min . After fixation , the cells were washed with PBS and incubated for 1 h at room temperature with anti-E antibody ( 4G2 ) , followed by incubation with HRP-conjugated anti-mouse IgG antibody . The assays were developed with the Vector VIP peroxidase substrate kit ( Vector Laboratories ) according to the manufacturer’s instructions . The viral titers were expressed as focus forming units ( FFU ) /ml . Seven day-old female mosquitoes were deprived of sucrose 24 h prior to the infectious blood meal . They were then allowed to feed for 2 h on blood-soaked cotton pledgets in the dark at 28°C . The infectious blood meal was comprised of washed rabbit erythrocytes ( obtained from animals housed at the Institut Pasteur animal facility ) , viral suspension , and ATP ( as a phagostimulant ) at a final concentration of 5 μM . The virus titer in the blood meal was adjusted to 8 x 106 FFU/ml . Blood-fed females were sorted and transferred into cardboard containers covered with mosquito nets . After exposure , engorged mosquitoes were maintained at 26°C , 80% relative humidity , with a 10 h: 10 h light: dark cycle with simulation of dawn and sunset during 2 h . Mosquitoes were dissected at various time points after oral exposure . For titrations , the mosquitoes or individual organs were collected in a tube containing 0 . 5 mm glass beads and 300 μl of DMEM supplemented with 2% FBS . The organs were ground for 30 sec at maximum speed , using a Minilystissue homogeneizer ( Bertin ) and stored at -80°C until analysis . Experiments were reproduced twice with 5 to 10 mosquitoes collected at each time point for dissection . JEV exposed mosquitoes were anesthetized at 4°C , legs and wings were removed and the bodies were attached to a glass slide using double-sided tape . The proboscis was manually inserted into a 10 μl low binding pipette tip filled with 10 μl DMEM containing 2% FBS . The tip contents were collected 30 min later in a tube . Two μl were transferred to a tube containing 2 μl SDS sample buffer and analyzed by dot-blot to verify the presence of saliva . Four μl were analyzed by FFA to determine virus titer . Ten to 20 mosquitoes were analyzed for each time point ( days 11 , 12 and 13 post-virus exposure ) and experiments were reproduced twice . Five days after emerging , mosquito females were blood-fed on mice previously anesthetized by intraperitoneal injection of a mixture of Xylazine ( 5 to 10 mg/kg ) and Ketamine ( 80 to 100 mg/kg ) . Three weeks later , 100 salivary glands ( SG ) were dissected and placed in 100 μl 1X PBS . SG extracts were prepared by sonicating the SG ( five times at 4 min each with a pulse ratio of 2 sec on / 2 sec off ) and centrifuging the crude extract at 10 , 000 g for 15 min at 4°C . The supernatant was transferred to clean tubes and stored at −80°C . The inocula used in our experiments contained the equivalent to a pair of SG . Protein lysates were prepared by cell lysis in radio-immunoprecipitation assay ( RIPA ) buffer ( Bio Basic ) containing protease inhibitors ( Roche ) . Equal amounts of proteins were loaded on a NuPAGE Novex 4–12% Bis-Tris protein gel ( ThermoFisher Scientific ) and transferred to a polyvinylidene difluoride membrane ( Bio-Rad ) using the Trans-Blot Turbo Transfer System ( Bio-Rad ) . After blocking the membrane for 1 h at room temperature in PBS-Tween ( PBS-T ) plus 5% milk , the blot was incubated overnight at 4°C with appropriate dilutions of the primary antibodies . The membrane was then washed in PBS-T and then incubated for 1 h at room temperature in the presence of HRP-conjugated secondary antibodies . After washes in PBS-T , the membrane was developed using Pierce ECL Western Blotting Substrate ( ThermoFisher Scientific ) and exposed to film . The saliva collected from each mosquito was blotted onto a nitrocellulose membrane . Two μl of DMEM containing 2% FBS and 1 μg of mosquito salivary gland extract were deposited on the membrane as negative and positive controls , respectively . The membranes were blocked for 1 h in PBS-T plus 5% milk and incubated overnight at 4°C with an anti-mosquito saliva antibody . The blots were then processed as indicated above for Western blotting . After dissection , midguts ( MG ) and salivary glands ( SG ) were placed on slides and the PBS removed . MG were fixed in acetone for 15 min . SG were fixed in 4% paraformaldehyde for 15 min . Both slides were dried and stored at 4°C until use . The MG and SG were then rehydrated in PBS for 15 min . The MG and the SG were incubated in Triton X100 ( 0 . 2% ) for 2 h and 15 min , respectively . They were then washed with PBS and incubated for 30 min with PBS + 0 . 1% Tween 20 containing 1% BSA . The slides were drained and incubated overnight at 4°C with anti-flavivirus protein E 4G2 antibody diluted in PBS , then washed with PBS . The slides were next incubated for 1 h with a fluorophore conjugated antibody , and washed with PBS . After washing , a drop of ProLong Gold Antifade reagent with DAPI ( ThermoFisher Scientific ) was placed on each slide and a cover slide was added . All preparations were examined using a fluorescence microscope ( Axioplan 2 Imaging , Zeiss ) . Three-week-old female BALB/c mice were housed under pathogen-free conditions at the Institut Pasteur animal facility . Groups of mice were anesthetized as described above , and were next intradermally inoculated with 50 FFU of JEV genotype 5 in absence or in presence of salivary gland extract or with JEV-infected saliva diluted in 100 μl of DPBS supplemented with 0 . 2% endotoxin-free serum albumin . An unpaired t test was used to compare quantitative data , and a Log-rank ( Mantel-Cox ) test was used to compare survival data . GraphPad Prism was used for all statistical analysis . Rabbit and mice were housed in Institut Pasteur animal facilities .
To assess the vector competence of European mosquitoes for JEV , we decided to use two molecular clones of viruses ( RP-9 and XZ0934 ) , which are representative of two currently circulating genotypes . The well-characterized genotype 3 strain , JEV RP-9 , was isolated from Cx . tritaeniorhynchus mosquitoes in Taiwan in 1985 [34 , 35] , while the genotype 5 strain , JEV-XZ0934 , was recently isolated from Cx . tritaeniorhynchus mosquitoes in China in 2009 [11] . For simplification , JEV-RP-9 and JEV-XZ0934 will be hereafter referred to as JEV g3 and JEV g5 , respectively . Both viruses were produced by transfection of cDNA into mammalian cells , as previously described [33] , followed by amplification of viral stocks in chicken fibroblasts DF-1 cells . Those viruses displayed comparable growth after infection of Ae . albopictus derived C6/36 cells ( Fig 1 , [33] ) . To evaluate the vector competence of European mosquito species for JEV , we exposed mosquitoes to either JEV g3 or JEV g5 by feeding on blood meals containing approximately 8 x 106 FFU of virus per ml . We note that the viremia in infected pigs or in birds can reach up to 107 PFU/ml , but is on average 104 PFU/ml [36–41] . While we offered blood meals that contained relatively high levels of virus , it is generally accepted that a greater quantity of virus is needed to infect mosquitoes orally with artificial mixtures than with viremic hosts [42] . For each experiment , 3 blood-fed mosquitoes were harvested immediately post-virus exposure , and the ingested virus titers were evaluated by FFA . The amount of ingested infectious virus was comprised between 400 and 9 , 000 FFU per mosquito , with an average titer of 4 , 000 FFU . Previous studies on vector competence of various species of mosquitoes for JEV have shown the peak for JEV infection and transmission occurs between 5 and 23 days after peroral infection [23 , 29 , 43–48] . Our preliminary studies showed that , under our experimental conditions , the majority of Cx . pipiens and Ae . albopictus were infected from 10 to 15 days post-virus exposure . We chose to focus collection times around the peak of viral transmission and harvested samples at 7 , 11 , 12 and 13 days post-virus exposure . We note that the survival rate of exposed mosquitoes dropped considerably after 2 weeks of infection , and consequently did not analyze the levels of mosquito infection beyond this point . First , we determined the infection rates in Ae . albopictus ( Fig 2A ) and Cx . pipiens ( Fig 2B ) mosquitoes by titrating the midguts harvested from mosquitoes . Next , we measured the levels of JEV infection in the heads of infected mosquitoes , and calculated infected dissemination rates ( Fig 2C and 2D ) . We did not observe any statistically significant differences in infection rates amongst genotypes for each mosquito species or by time after the infectious blood meal . We did note that dissemination of JEV was faster in Ae . albopictus mosquitoes , when compared with Cx . pipiens mosquitoes . Notably , at 7 days post-virus exposure , we found that 57 to 90% of Ae . albopictus mosquitoes were systemically infected , whereas only 26 to 36% of Cx . pipiens were ( Fig 2C and 2D ) . Last , we determined transmission rates through titration of saliva collected from blood-fed mosquitoes ( Fig 2E and 2F ) . While this is a method widely used to determined transmission rates , we observed that salivation assays are highly dependent on salivation efficiency , and that the levels of virus in saliva can sometimes be below the detection limit of our titration assay . Keeping in mind that this determination of the virus transmission rates has limitations , we observed that both mosquito species transmitted JEV at rates ranging from 20 to 63% for Ae . albopictus , and from 12 to 41% for Cx . pipiens ( Fig 2E and 2F ) . Next , we analyzed the levels of JEV g3 and g5 accumulation in the different mosquito organs that had been harvested ( Fig 3 ) . We noted that JEV levels in the midguts slowly decreased between 7 and 13 days post-virus exposure , while viral levels in heads and salivary glands increased over time , which is consistent with patterns of viral dissemination in mosquitoes . We noted that at 7 days post-virus exposure , the rates of salivary glands infection ranged from 40 to 80% for Ae . albopictus , and from 5 to 9% for Cx . pipiens . Viral infection of salivary glands has been shown to correlate well with infection of saliva [43] , and thus we hypothesize that Ae . albopictus mosquitoes were likely to transmit JEV at earlier times than Cx pipiens . Interestingly , in Ae . albopictus mosquitoes midguts , we observed a significant difference in the titers of JEV g5 when compared to JEV g3 titers ( Fig 3A ) . Notably , it appeared that JEV g5 accumulated to higher levels than JEV g3 at 7 days post-virus exposure , and to lesser levels at later infection times ( 11 to 13 days post-virus exposure ) . Additionally , we analyzed the distribution of JEV envelope protein in the organs of infected mosquitoes ( Fig 4 ) . First , we performed immuno-localization within organs harvested from Ae . albopictus mosquitoes at 14 days post-virus exposure , which corresponds to a peak in viral transmission ( Fig 4A ) . While envelope protein staining within the midgut was relatively weak , there was a strong staining of numerous cells within both lobes of salivary glands . Samples collected at 11 days post-virus exposure were also analyzed by western blotting and showed good detection of the envelope protein in midguts and salivary glands ( Fig 4B ) . To evaluate the levels of virus secreted in mosquito saliva , we collected mosquitoes at 11 , 12 and 13 days post-feeding on an infectious blood meal , and performed forced salivation . Since not all of the mosquitoes salivate when subjected to this assay , we also performed a survey of successful salivation . A fraction of the collected saliva was dotted on a membrane , and was next incubated with an antibody specific for mosquito saliva ( Fig 5A ) . We noted that both mosquito species efficiently salivated under our experimental conditions , and that the levels of actual salivation were above 40% for either mosquitoes ( Fig 5A ) . The collected saliva was then subjected to a standard infectivity assay to determine the levels of JEV transmitted in JEV-positive saliva at each time point ( Fig 5B and 5C ) . We noted that for both mosquito species , higher levels of virus were secreted in saliva at later times post-virus exposure , which mirrored the increase in viral load in salivary glands ( Fig 3C and 3F ) . The levels of infectious virus in saliva ranged between 2 and 200 FFU for JEV g3 ( 45 and 55 FFU in average for Ae . albopictus and Cx . pipiens , respectively ) , and between 2 and 196 FFU for JEV g5 ( 38 and 35 FFU in average for Ae . albopictus and Cx . pipiens , respectively ) . Next we assessed whether the virus transmitted by European mosquitoes was capable of developing a productive infection in mammalian hosts . To evaluate this , we used a previously characterized murine model for Japanese encephalitis , based on JEV g5 infection of 3-week-old BALB/c mice [33] . Three-week-old BALB/c mice were injected via intradermal route with JEV , as this mode of injection most resembles a mosquito bite . First , JEV-positive saliva samples collected from Ae . albopictus and Cx . pipiens mosquitoes ( Fig 5 ) were used as an inocula ( Fig 6A ) . Saliva containing various loads of virus was used , with a titer comprised between 7 and 98 FFU . A control group of mice were similarly injected with JEV grown from C6/36 cells , using a single dose of 50 FFU ( Fig 6A ) . As expected , the animals rapidly exhibited limb paralysis and encephalitis . We did not observe any significant differences in survival rates amongst the different inocula ( Fig 6A ) . The survival rate was between 33 and 40% , with a mean survival time of 11 to 12 . 5 days . The detection of JEV-specific antibodies showed that all surviving mice had been exposed to the virus ( S1 Fig ) . Since it was shown that mosquitoes can inject salivary components that influence the outcome of viral infection [49–51] , we next evaluated the impact of European mosquito salivary glands on JEV pathogenesis in a murine model . As described above , we used the intradermal route to inject 50 FFU of JEV g5 to 3-week-old BALB/c mice ( Fig 6B ) . For two groups of mice , the inoculum was mixed with salivary glands extracts obtained from Ae . albopictus or Cx . pipiens mosquitoes . In accordance with what was previously observed after injection of saliva collected from infected mosquitoes , we did not observe any significant difference in the development of JEV pathogenesis in presence of mosquito salivary glands ( Fig 6B ) . Overall these experiments show that European mosquitoes are fully competent at transmitting infectious JEV , but that saliva does not facilitate the development of viral pathogenesis in a susceptible murine model .
In recent years , the increase in locally acquired exotic arbovirus diseases in Europe can be linked to the presence of appropriate combinations of vectors and vertebrate hosts , which could ultimately lead to the establishment of these diseases in Europe [52–54] . Since 2010 , sporadic cases of locally acquired chikungunya and dengue fevers have been noted in Europe [55 , 56] . The driving forces behind these events are viraemic travelers and the increasing presence of competent vector species , such as Ae . aegypti and Ae . albopictus , in temperate regions . Likewise , the circulation of WNV and Usutu virus–two Flaviviruses—was reported in 10 European countries [57] . Two studies in Italy reported the infection of local Cx . pipiens populations with both WNV and Usutu virus [58 , 59] and there is an increase in WNV disease incidence in Europe [60] . While JEV RNA was recently detected in mosquitoes and birds in Northern Italy [21 , 26] , to date , human infections with JEV were only reported in travelers returning from endemic countries [61–63] . Our study shows for the first time that European strains of Cx . pipiens and Ae . albopictus are both competent vectors to transmit two genotypes ( 3 and 5 ) of JEV . High levels of infection , dissemination and transmission rates were observed in both vectors for either genotypes after oral exposure of mosquitoes to a blood meal containing virus at 8 x 106 FFU/ml . In the present study , we did not evaluate vector competence for viral strains belonging to the genotype 1 . Strains belonging to this genotype have displaced the prevalent genotype 3 in several countries in recent years [5–7] . Genotype 1 and 3 strains are genetically close when compared to the more distant genotype 5 strains [8–10] . Since we observed equivalent vector competence of European mosquitoes for genotypes 3 and 5 , we hypothesize that those mosquitoes will also be competent at transmitting genotype 1 viruses . Several mosquitoes from the Culex genus are established vectors for JEV . Cx . tritaeniorhynchus is the main vector in the enzootic cycle of JEV in tropical and subtropical regions of Asia . Interestingly , the vector competence of a Cx . pipiens molestus population from Taiwan was found to be similar to that of Cx . tritaeniorhynchus , when tested in laboratory conditions [46] , which is in line with our observations . Other investigators reported that Cx . pipiens populations from other world regions ( Cx . pipiens molestus from Uzbekistan , Cx . pipiens pallens from Korea , Cx . pipiens from the United States of America ) were less susceptible to JEV and were not always capable of transmitting the virus [50 , 64] . Of note , there has been some reports of isolation of JEV from field-caught mosquitoes along the years [12 , 17 , 21 , 65] , and all strains of JEV isolated from Cx . pipiens mosquitoes in Korea in 2012 belonged to the genotype 5 [12] . As strains belonging to the genotype 5 were only rarely isolated , one can wonder if the transmission cycles that are involved in the maintenance of those viruses involve mosquito and amplifying host species different from the established Cx . tritaeniorhynchus / swine model . Since the currently available vaccines do not confer full protection against JEV genotype 5 strains [66 , 67] , the risks of JEV g5 transmission to human populations must be carefully examined . Similarly to Cx . pipiens , field-collected Ae . albopictus mosquitoes were occasionally found positive for JEV [22 , 68] . Various transmission rates were observed in laboratory settings , from less than 17% for Australian populations [69] to 45% for Taiwanese populations [22] . In our experiments , European Ae . albopictus was also able to transmit different strains of JEV to high efficiency , which supports the hypothesis that European mosquito populations belonging to these two species have a better vector competence for JEV than populations isolated in other parts of the world . Our results also showed that the extrinsic incubation period ( i . e . the time between ingestion of the virus and the ability of the mosquito to become infectious ) for JEV is shorter in Ae . albopictus than in Cx . pipiens . We have not performed a formal analysis of the relative life span of our mosquito populations after ingestion of an infectious blood meal . If we assume that both mosquito species have similar life spans , then this would imply that Ae . albopictus mosquitoes can transmit JEV for a longer period than the French population of Cx . pipiens and therefore might be a more efficient vector . Specifically , host biting preferences may have consequences on the emergence of the disease in Europe and its transmission dynamics . Arbovirus circulation is defined by many aspects including the population dynamics of the mosquito vector , the extrinsic incubation period , and the population densities of the vertebrate amplifying hosts , all of which are influenced by environmental factors . In the case of JEV , the classic Cx . tritaeniorhynchus–pig transmission cycle was observed in Japan , a region of high pig farming density , but other species and scenarios could be invoked in regions where pig farming is less abundant , or where Cx . tritaeniorhynchus is not found [70] . An example of this is the 1995 outbreak of Japanese encephalitis in Australia that involved the presence of domestic pigs and high populations of Cx . annulirostris [19 , 20] . Ae . albopictus is considered to be an opportunistic feeder: it primarily feeds on mammalian hosts ( humans , wild and domestic animals ) but can also acquire blood from avian sources [71 , 72] . Analysis of feeding patterns in temperate regions showed that populations of Ae . albopictus in the United States of America mainly fed on mammals and rarely on birds [73] . Cx . pipiens mosquitoes feed mostly on birds ( 83% ) but also on mammals [74] . Interestingly , it was shown that 20% of Cx . pipiens emerging from diapause in temperate habitats fed on mammals [73] . Considering the natural cycle of JEV implying birds as reservoir and pigs as amplifying hosts , specificity in host preferences may have consequences on the possible emergence of the disease in Europe and its transmission dynamics . Favorable conditions for JEV emergence may be gathered in several places in Europe where pig breeding sites , bird sanctuaries and Ae . albopictus and/or Cx . pipiens mosquitoes coexist ( Marquenterre parc , Baie de Somme , France; Camargue , Rhone delta , France; Danube delta , Roumania ) . The last part of our study was to investigate the role of mosquito saliva in the transmission of JEV to mice . When insects take a blood meal , they trigger defensive responses from the vertebrate , such as hemostasis and various immune responses . The saliva proteins injected by the mosquito can counteract these defenses , through their angiogenic , anti-hemostatic , anti-inflammatory and immunomodulatory properties [75] . This complex interaction may significantly affect the evolution of the disease; notably co-injection of virus and saliva was shown to potentiate infection of the vertebrate host by arboviruses belonging to various families [49–51 , 76 , 77] . While we did not observe any enhancement of Japanese encephalitis disease in mice , in presence of salivary gland extract or of saliva collected from either European vectors , further studies are needed to evaluate the impact of saliva on viral burden in different organs . Additionally , we cannot exclude the possibility that JEV pathogenesis might be enhanced by salivary factors of other mosquito species , or that saliva from the two species tested in the present study might enhance pathogenicity in other mammalian species .
In this study , we have clearly demonstrated that European populations of Ae . albopictus and Cx . pipiens were efficient vectors for JEV transmission . Conditions for a putative emergence of JEV in Europe are linked to the possibility for an enzootic cycle to take place in temperate areas . In order to complete the infection cycle , JEV must be transmitted to a susceptible vertebrate host , capable of producing sufficient viral titers for subsequent acquisition by the insect vector . It is therefore important to further investigate whether any European swine or water birds populations can be infected with JEV and produce sufficiently high viremias to infect mosquitoes that feed on them . Such knowledge is critical to assess the potential for JEV to establish local transmission cycles similar to the closely related WNV in Northern Italy . | Japanese encephalitis virus ( JEV ) is the leading cause of viral encephalitis in Asia . JEV is maintained in a cycle involving mosquitoes and vertebrate hosts , mainly pigs and wading birds . Humans can be infected when bitten by an infected mosquito . Culex tritaeniorhynchus is the main vector of the disease in tropical and subtropical areas . The recent detection of JEV in birds and mosquitoes collected in Northern Italy has led us to evaluate the putative emergence of this arboviral disease in Europe . For this purpose , we have tested the competence of European populations of Cx . pipiens and Aedes albopictus to transmit this virus in a laboratory setting . We showed that these local mosquitoes could be infected and were capable of transmitting a pathogenic virus to mice . It is thus urgent to evaluate the risks of JEV emergence in European regions displaying a favorable environment for mosquito vectors , susceptible pigs and wading birds . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"animal",
"models",
"of",
"disease",
"geographical",
"locations",
"microbiology",
"saliva",
"animals",
"animal",
"models",
"physiological",
"processes",
"model",
"organisms",
"salivation",
"experimental",
"organism",
"systems",
"insect",
"vectors",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"animal",
"models",
"of",
"infection",
"animal",
"studies",
"epidemiology",
"exocrine",
"glands",
"mouse",
"models",
"disease",
"vectors",
"insects",
"hematology",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"blood",
"anatomy",
"physiology",
"salivary",
"glands",
"biology",
"and",
"life",
"sciences",
"europe",
"organisms"
]
| 2017 | European Aedes albopictus and Culex pipiens Are Competent Vectors for Japanese Encephalitis Virus |
The circadian clock integrates temporal information with environmental cues in regulating plant development and physiology . Recently , the circadian clock has been shown to affect plant responses to biotic cues . To further examine this role of the circadian clock , we tested disease resistance in mutants disrupted in CCA1 and LHY , which act synergistically to regulate clock activity . We found that cca1 and lhy mutants also synergistically affect basal and resistance gene-mediated defense against Pseudomonas syringae and Hyaloperonospora arabidopsidis . Disrupting the circadian clock caused by overexpression of CCA1 or LHY also resulted in severe susceptibility to P . syringae . We identified a downstream target of CCA1 and LHY , GRP7 , a key constituent of a slave oscillator regulated by the circadian clock and previously shown to influence plant defense and stomatal activity . We show that the defense role of CCA1 and LHY against P . syringae is at least partially through circadian control of stomatal aperture but is independent of defense mediated by salicylic acid . Furthermore , we found defense activation by P . syringae infection and treatment with the elicitor flg22 can feedback-regulate clock activity . Together this data strongly supports a direct role of the circadian clock in defense control and reveal for the first time crosstalk between the circadian clock and plant innate immunity .
Plants are challenged by various pathogens on a daily basis . Accumulating evidence implicates a role of the circadian clock in regulating plant innate immunity . The circadian clock is the internal time measuring machinery important for plant growth and development . However , our understanding of the molecular basis of how the circadian clock controls plant innate immunity is still in its infancy . Plants have evolved various mechanisms , some pre-formed and others induced , to ward off pathogen invasion . An example of pre-formed surface structures is the stomate , the natural opening important for photosynthetic gas exchange . This opening can provide a portal for pathogens to enter leaves; however , plants can also control the aperture of stomata to physically limit pathogens [1] , [2] . One type of induced defense is activated when plants recognize pathogen-associated molecular patterns ( PAMPs ) , which are conserved molecules or structures present in groups of related microbes . This defense , also termed PAMP-triggered immunity ( PTI ) , can be highly effective against non-adapted pathogens and provides a basal level of defense even against adapted pathogens [3] , [4] . Another type of induced defense is activated by plant resistance ( R ) proteins , which specifically recognize secreted pathogen effectors and subsequently activate effector-triggered immunity ( ETI ) . ETI , also termed R gene-mediated resistance , is a stronger and faster elaboration of PTI , and frequently results in hypersensitive cell death at the infection site [5] , [6] , [7] . The small molecule salicylic acid ( SA ) has been linked to signal transduction in PTI and ETI [8] , [9] , [10] . The circadian clock has profound influence on the fitness of organisms [11] , [12] , [13] , [14] , [15] , [16] . The core of the circadian clock is the central oscillator , which in Arabidopsis , is composed of multiple interconnected negative feedback loops that orchestrate biological adjustments independently of external stimuli [17] , [18] . Of these clock components , CIRCADIAN CLOCK ASSOCIATED1 ( CCA1 ) and its close homolog LATE ELONGATED HYPOCOTYL ( LHY ) are transcription factors that are involved in multiple feedback loops and function synergistically to regulate clock activity [19] , [20] , [21] . The role of the circadian clock in controlling plant innate immunity has long been proposed based on circadian-regulation of defense gene expression [22] , [23] , [24] , [25] , [26] . Direct evidence from several research groups has recently emerged to support such a role of the circadian clock . Under free running conditions , wild type Arabidopsis exhibits temporal oscillations in susceptibility to Pseudomonas syringae infection , which are disrupted by overexpression of CCA1 [27] . Misexpression of several clock genes , including CCA1 , compromises resistance to the bacterial pathogen Pseudomonas syringae and/or to the oomycete pathogen Hyaloperonospora arabidopsidis ( Hpa ) [27] , [28] , [29] . Interestingly , although lhy mutants exhibit similarly shortened circadian period as cca1 mutants , LHY was not shown to play a defense role against Hpa [28] . This raises the question of whether CCA1 is a dual function protein , affecting both the circadian clock and other non-clock related processes , as shown in the case of another central oscillator component GIGANTEA [30] . cca1-conferred disease susceptibility might be attributed to a role of CCA1 in regulating non-clock related processes rather than to its direct involvement in the circadian clock [31] . To better understand the role of CCA1 and LHY-mediated circadian clock in defense control , we tested plants misexpressing CCA1 and/or LHY for disease resistance to P . syringae and Hpa . We show that CCA1 and LHY loss-of-function mutants synergistically affect basal resistance and R gene-mediated defense against both pathogens . Disrupting the circadian clock caused by overexpression of CCA1 or LHY also results in severe disease susceptibility to P . syringae . The defense role of CCA1 and LHY against P . syringae is at least partially through circadian control of stomatal aperture but is SA-independent . Furthermore , we found that clock activity is modulated by P . syringae infection or treatment with the elicitor flg22 . These data further establish the role of the circadian clock in defense control and for the first time reveal crosstalk between the circadian clock and plant innate immunity .
To evaluate defense roles of CCA1 and LHY , we constructed the cca1-1lhy-20 mutant via a genetic cross in a Col-0 background that also contains the LUCIFERASE reporter gene driven by the CCA1 promoter ( ProCCA1:LUC ) . The single loss of function mutants , cca1-1 and lhy-20 , have shortened circadian periods of ProCCA1:LUC expression in constant light ( LL ) [11] . In LL , we confirmed that cca1-1lhy-20 had a much-shortened period ( 19 . 9±0 . 11 hr ) , compared with wild type ( wt ) Col-0 ( 24 . 4±0 . 09 hr ) ( Figure S1A and [19] ) . Although experiments in LL are important for establishing the involvement of the circadian clock in specific phenotypes , such experimental conditions can also be limiting . In entraining conditions ( e . g . , a 12 hr L/12 hr D cycle; LD ) , the altered period of clock mutants like cca1-1 and lhy-20 is not seen due to the entraining cycle , which imposes a 24 hr period ( Figure 1 ) . The clock remains important in such LD conditions , though , because the clock determines the phase of specific events with respect to as dawn and dusk . Mutants with altered period in LL typically exhibit altered phase in LD , with short period mutants exhibiting a leading ( early ) phase and long period mutants exhibiting a lagging ( late ) phase [32] . Moreover , interactions between the endogenous circadian clock and external LD cycles can results in phase differences , sometimes dramatic , when measured in LD versus LL . For example , the phase of maximal hypocotyl elongation during early seedling growth was shifted 8–12 hours between LD and LL conditions [33] , [34] . In their natural environment , plants do not usually encounter LL . Therefore in evaluating the role of the circadian clock on plant defense against pathogens , it is critically important to study plant-pathogen interactions in LD and to consider the potential influence of the circadian clock on the phases of rhythmic events that might influence the plant response to pathogen challenge . We show here that in LD the phases of cca1-1 and lhy-20 single mutants were leading with respect to that of wild type Col-0 , and that the cca1-1 lhy-20 double mutant exhibited a much earlier phase than either single mutant , consistent with the synergistic contribution of CCA1 and LHY in regulating clock activity ( Figure 1 and Figure S1B ) . Early phase was also reported with other cca1lhy mutants [20] , [21] . In addition , we found that plants overexpressing CCA1 ( CCA1ox ) , which display arrhythmic clock activity in LL [35] , also showed arrhythmic expression of ProCCA1:LUC in LD with an acute peak in response to lights on ( Figure 1 and S1B ) . Low ProCCA1:LUC activity in CCA1ox is consistent with CCA1 being a negative regulator of its own expression [35] . These results emphasize that altered function of the circadian clock can manifest in both LL and LD conditions . To test disease resistance of cca1-1 and lhy-20 plants , we performed infection experiments at Zeitgeber Time 1 ( Zeitgeber Time is the time relative to dawn; ZT1 is 1 hr after lights on ) or ZT13 ( 1 hr after lights off ) , two times of day associated with drastic changes of light regime . Plant leaves were pressure-infiltrated with virulent P . syringae pv . maculicola ES4326 strain DG3 ( PmaDG3 ) [36] . The infected plants were placed in either LD or LL . Bacterial growth assays at 3 days post infection ( 3 dpi ) revealed no significant difference among Col-0 , cca1-1 , lhy-20 , and cca1-1lhy-20 in either LD or LL ( Figure 2 and Figure S2 ) . Under natural conditions , P . syringae enters the apoplast of leaves through openings such as stomata and wounds . It is known that stomatal aperture is regulated by the circadian clock [37] , [38] . Therefore , infiltration of bacteria directly into plant tissue might bypass the influence of the circadian clock on stomatal defense . To test this possibility , we spray-infected with PmaDG3 Col-0 , cca1-1 , lhy-20 , and cca1-1lhy-20 at ZT1 and ZT13 in LD . We found that Col-0 supported over 10-fold more bacterial growth with ZT1 infection than with ZT13 infection ( Figure 3A and 3B ) , suggesting that Col-0 is more resistant at night than at dawn when spray-infected . Although we did not observe significant difference in bacterial growth between Col-0 and cca1-1 and lhy-20 single mutants , the double mutant cca1-1lhy-20 showed enhanced susceptibility to PmaDG3 when sprayed at ZT13 ( Figure 3A to 3C ) . Consistent with this result , we found that PmaDG6 ( an avirulent strain recognized by the resistance protein RPS2 in Col-0 ) [36] ) grew significantly more in cca1-1lhy-20 than in Col-0 and the single mutants with ZT13 infection ( Figure 3D and 3E ) . Together these data suggest that CCA1 and LHY share redundant functions to regulate both basal and RPS2-mediated defense against P . syringae . To further substantiate the role of CCA1 and LHY in defense regulation , we tested disease resistance of plants overexpressing CCA1 ( CCA1ox ) or LHY ( LHYox ) , which were shown to have arrhythmic clock activity in LL [35] , [39] . CCA1ox plants also exhibited clock arrhythmicity in LD ( Figure 1 and S1B ) . Disease resistance assays indicate that CCA1ox plants were more susceptible to PmaDG3 than Col-0 with infiltration infection in LD or LL ( Figure 2 and S2 ) . CCA1ox plants were also more susceptible than Col-0 to PmaDG3 and to PmaDG6 when spray-infected at ZT1 or ZT13 in LD ( Figure 3 ) . LHYox plants are in the Landsberg erecta ( Ler ) background , with which we used P . syringae pv . tomato DC3000 ( DC3000 ) to test disease resistance because this strain induces stronger disease symptoms in our hands than does PmaDG3 . Similar to CCA1ox plants , LHYox plants had more bacterial growth than Ler when infiltrated with DC3000 at ZT1 or ZT13 in LD ( Figure 4A ) . In addition , spray-infection at ZT1 or ZT13 in LD also gave similar results ( Figure 4B ) . Together , disruption of the circadian clock by misexpressing CCA1 and/or LHY compromises disease resistance to P . syringae , supporting a direct role of the circadian clock in defense regulation . Our data show that cca1-1lhy-20 was more susceptible with spray-infection and CCA1ox and LHYox plants displayed enhanced susceptibility with both spray and infiltration infections . These suggest that both stomata-dependent and -independent defense can be affected by misexpression of either of these two core oscillator genes . Consistent with this notion , a previous study showed that CCA1ox plants had increased CO2 assimilation and stomatal conductance [13] . To further test whether the defense role of CCA1 and LHY is linked to the control of stomatal pore size , we measured plant stomatal aperture at ZT1 and ZT13 in LD . Consistent with Col-0 being more resistant with spray-infection at ZT13 than at ZT1 , we found that stomatal aperture of Col-0 was much smaller at ZT13 than at ZT1 ( Figure 5A ) . Compared with Col-0 , the cca1-1 and lhy-20 mutants and CCA1ox plants showed similar stomatal aperture at ZT1 but had greater stomatal aperture at ZT13 ( Figure 5A ) . These data suggest that disrupting clock activity mediated by CCA1 and LHY could make plants less responsive to dark-induced stomatal closure at night , thereby enhancing access of P . syringae to the leaf interior . To further determine how these mutants respond to P . syringae infection , we measured stomatal aperture in the presence of PmaDG3 . PmaDG3 treatment was performed at ZT4 after plants had been exposed to light for four hours to ensure the opening of the stomata ( Figure S3 ) . At 1 hr post infection ( 1 hpi ) , we observed a 48 . 1% suppression of stomatal aperture in Col-0 , compared with mock treatment ( Figure 5C top and Table S1 ) . However , this suppression was much reduced in cca1-1 and lhy-20 and largely blocked in cca1-1lhy-20 and CCA1ox . P . syringae-induced stomatal closure was transient since both mock and PmaDG3-treated leaves showed similar stomatal aperture at 3 hpi ( Figure 5C bottom ) . Although exhibiting similar stomatal aperture at ZT1 and ZT13 ( Figure 5B ) , the LHYox plants also showed reduced suppression of DC3000-induced stomatal closure at 1 hpi ( 16 . 9% ) , compared with Ler control ( 51 . 6% ) ( Figure 5D and Table S1 ) . Hence , these results indicate that disrupting the circadian clock by CCA1 and LHY misexpression impairs plants' capacity of inducing stomatal closure in response to P . syringae . CCA1 but not LHY was previously shown to regulate resistance to the oomycete pathogen Hpa [28] . To test whether a contribution of LHY to Hpa resistance could be discerned in the double mutant cca1-1lhy-20 , we sprayed seven-day-old seedlings at ZT7 in LD with the virulent strain Hpa Emco5 or the avirulent strain Hpa Emoy2 ( recognized by the R protein RPP4 in Col-0 ) . We observed significantly more susceptibility to both Hpa strains in the cca1-1lhy-20 double mutant , compared to Col-0 and the single mutants ( Figure 6A and 6B ) while the CCA1ox plants were substantially more resistant to Hpa Emco5 ( Figure 6A ) . Our data are broadly in agreement with those previously reported [28] . The reason that we did not observe a significant difference between Col-0 and cca1-1 could be due to the difference in the infection time and/or Hpa strains used - Wang et al inoculated plants with the avirulent strain Hpa Emwa1 at dawn [28] while we used Hpa Emco5 ( virulent ) and Emoy2 ( avirulent ) in the afternoon in our experiments . Nevertheless , these data , together with the P . syringae data described earlier , demonstrate that CCA1 and LHY contribute synergistically to basal resistance and R-gene mediated defense against both bacterial and oomycete pathogens . What surprises us is the difference in response to P . syringae ( decreased resistance ) and Hpa ( enhanced resistance ) strains observed in CCA1ox plants . We speculate that there are distinct mechanisms that these plants use to defend against the two pathogens . Identification of defense-related genes controlled by CCA1 and LHY is critical to gain better understanding of the mechanism of action of CCA1 and LHY in defense regulation . To this end , we analyzed promoters of 571 genes for CCA1-binding site ( CBS ) and evening element ( EE ) , two cis elements known for CCA1 and LHY binding [40] , [41] , [42] . These 571 genes had been previously selected to construct mini-microarrays , consisting of three groups , selected ( 337 defense-related genes based on microarray experiments ) , empirical ( 127 empirical marker genes for various pathogen responses ) , and normalization ( 107 non-defense related genes whose expression levels were relatively stable among experiments with pathogen infection ) [43] . The online tool POBO [44] was used to analyze up to 3000 bp from the promoter regions of these genes , which do not include the coding sequences of neighboring genes , for an enrichment of CBS or EE motifs . The background for this analysis was generated using pseudo-clusters of 100 promoters of up to 3000 bp in length of randomly sampled Arabidopsis genes ( 1000 bootstrap replications were used in the sampling ) . Compared with the background , the CBS motif was found as often as expected by chance in the selected and empirical gene promoters ( Figure 7A and 7B ) but the motifs were found less frequently in the normalization gene promoters ( Figure 7C and Table S2 ) . When compared to the normalization genes , there was a greater than 40% increase of the cluster mean for the CBS motif in both selected and empirical genes . These observations suggest that although defense-related genes ( selected and empirical genes ) are not particularly enriched with the CBS motif , the non-defense related genes ( the normalization genes ) are slightly depleted of the motif . The enrichment of the EE motif was more pronounced in both selected and empirical genes , with about 200% increase of the cluster means when compared to the normalization genes ( Figure 7D–7F and Table S2 ) . Thus , these results suggest that defense-related genes are preferentially regulated by CCA1 and LHY . However , since the sample size in each group is small , caution should be taken when extrapolating this interpretation to the whole genome level . The frequency of CBS or EE motif per promoter region was quantified from the above three sets of genes ( Figure S4 ) . Among the genes analyzed , we found that GRP7 ( At2g21660; also known as COLD AND CIRCADIAN REGULATED 2 [CCR2] ) [45] , [46] had the most overrepresentation of the EE motif , with four EE within a 300 bp promoter region . One CBS motif was also found at 1294 bp of the GRP7 promoter . GRP7 is a key constituent of a slave oscillator regulated by the circadian clock [45] , [47] and also has been demonstrated to have roles in regulating floral transition and plant defense [48] , [49] . Expression of GRP7 was previously shown to be circadian regulated with a shortened circadian period in a cca1lhy double mutant and a disrupted pattern in CCA1ox plants [12] , [20] , [50] . However , GRP7 had never been explicitly established as a target gene of CCA1 and LHY . Our northern analysis confirmed circadian expression of GRP7 and showed that such expression was slightly affected by the cca1-1 mutation and became arrhythmic in CCA1ox in LL ( Figure S5 ) . We also observed disrupted expression of GRP7 in CCA1ox plants in LD ( Figure 8A ) . Thus , these data further confirm that GRP7 is regulated by CCA1 . GRP7 was previously demonstrated to regulate stomatal activity [51] . We found that similar to cca1-1lhy-20 and CCA1ox plants , stomatal aperture of grp7-1 was greater than that of Col-0 at ZT13 ( Figure 8B ) . In response to PmaDG3 infection , grp7-1 displayed 14 . 2% suppression of stomatal aperture whereas Col-0 showed 48 . 1% suppression at 1 hpi ( Figure 8C , S3 , and Table S1 ) , suggesting that grp7-1 has reduced responsiveness to PmaDG3 in stomatal closure . We further found that grp7-1 was significantly more susceptible to PmaDG3 than Col-0 when spray-infected at ZT13 in LD ( Figure 8D ) . Together our bioinformatic analysis and experimental evidence indicate that GRP7 is a target of CCA1 and/or LHY that regulates stomatal activity and modulates plant defense . SA is a key signaling molecule involved in both basal resistance and R gene-mediated defense . The accelerated cell death 6-1 ( acd6-1 ) mutant shows constitutive defense , high levels of SA , and extremely small size that is sensitized to the change of SA defense [52] , [53] . Thus , acd6-1 has been used as a convenient readout to gauge the effect of some known defense genes in regulating SA-mediated defense [54] , [55] , [56] . To determine whether CCA1 and LHY act through SA , we crossed cca1-1lhy-20 to acd6-1 and obtained homozygous double ( acd6-1cca1-1 and acd6-1lhy-20 ) and triple ( acd6-1cca1-1lhy-20 ) mutants . We found that both double and triple mutants resembled acd6-1 , displaying dwarfism and accumulating similar SA levels ( Figure 9A and B ) . However , when spray-infected with PmaDG3 at ZT13 , the double mutants were slightly more susceptible while the triple mutant was much more susceptible than acd6-1 ( Figure 9C ) . These results corroborate a synergistic interaction between CCA1 and LHY in clock and defense regulation . They also suggest that the defense role of CCA1 and LHY is largely SA-independent . Consistent with this notion , we found that in the absence of acd6-1 , the SA levels are comparable among Col-0 , cca1-1 , lhy-20 , cca1-1lhy-20 , and CCA1ox in LD ( Figure S6A ) . In addition , although more susceptible to P . syringae infection , LHYox plants were dwarf , showed spontaneous cell death , and accumulated high levels of SA ( Figure 4C , 4D , and S6B ) . Together , these results indicate that CCA1 and LHY act independently of SA to regulate resistance to P . syringae . Our data and those from other groups clearly indicate that plant innate immunity is an output event regulated by the circadian clock . However , it is not known whether this regulatory relationship is reciprocal with defense activation feeding back to affect clock activity . To test this , we infected Col-0 expressing the ProCCA1:LUC reporter with both virulent and avirulent P . syringae strains . Bioluminescence analysis indicated that the period of ProCCA1:LUC was significantly shortened in the presence of the virulent strain PmaDG3 or the avirulent strain PmaDG6 at a high dose ( OD = 0 . 1 ) ( Figure 10 and Table S3 ) . Similarly , infection of Col-0 seedlings expressing ProGRP7:LUC also resulted in period shortening of ProGRP7-controlled luciferase activity ( Figure S7 and Table S3 ) . These results suggest that clock activity is modulated by both basal and RPS2-mediated defenses . To further investigate which defense signaling pathway ( s ) are involved in the feedback-regulation of clock activity , we treated Col-0/ProCCA1:LUC seedlings with flg22 or benzo ( 1 , 2 , 3 ) thiadiazole-7-carbothioic acid ( BTH ) . Flg22 is a 22-aa synthetic peptide from the conserved region of flagellin proteins of P . syringae and elicits plant basal defense in a wide variety of plant species [4] , [57] . BTH is an agonist of SA that efficiently activates SA signaling [58] . We found that flg22 at both doses ( 1 µM and 10 µM ) significantly shortened the period of CCA1 expression . However , BTH treatment ( 10 µM and 300 µM ) did not change CCA1 promoter activity ( Figure 11A and Table S3 ) . To further test if SA could affect clock activity , we used a cotyledon movement assay [59] to gauge clock activity in the acd6-1 mutant , which constitutively accumulates high levels of SA [52] , [53] . We found that acd6-1 showed similar period , phase , and amplitude of the rhythm for cotyledon movement to Col-0 ( Figure 11B and S8 ) . Taken together , these data indicate that activation of flg22-triggered basal defense but not SA signaling can feedback to regulate clock activity .
Typical studies of the circadian clock have been performed under constant light ( LL ) conditions to emphasize the endogenous nature of the clock . In LL , perturbations of the circadian clock typically result in altered period length; for instance , loss of CCA1 or LHY function shortens circadian period . However , plants typically grow in LD cycles in which the environmental cycle entrains even a mutant clock to a 24-hour period . Under such LD conditions , perturbations in the circadian clock can manifest as alterations in phase for reporter gene expression ( Figure 1 and S1B and [20] , [21] ) as well as changes in a variety of other traits , including flowering time , metabolism , stomatal activity , gene expression patterns , and defense responses [12] , [13] , [20] , [29] , [50] , [60] , [61] . Thus , the effects of disrupted circadian clock could become apparent under LL and LD conditions . Several studies indicate that the circadian clock mediated by CCA1 and LHY regulates plant defense in both LL and LD ( [27] , [28] and this study ) . For instance , Bhardwaj et . al . showed that CCA1ox plants were more susceptible to P . syringe infection than wt in LL [27] . Here we extend this observation by showing that CCA1ox plants had enhanced susceptibility to P . syringae in both LL and LD ( Figure 2 , 3 , and S2 ) . In LD , enhanced susceptibility was also observed in cca1-1lhy-20 and LHYox plants to P . syringae strains ( Figure 3 and 4 ) and in cca1-1lhy-20 to Hpa strains ( Figure 6 ) . Consistent with our data , a single cca1 mutant showed compromised resistance to a different Hpa strain and affected expression of some defense-related genes in LD [28] . Together these studies firmly establish that plant innate immunity is an output regulated by the circadian clock under LL and LD conditions . While we mainly focus our analyses in this report on defense phenotypes regulated by CCA1 and LHY in LD , we also agree that we should use caution when interpreting our results since the effect of the circadian clock can manifest differently under different light conditions , including both differing daylengths and light intensities . For instance , it is possible that the degree of susceptibility to pathogen infection and the severity of stomatal change in response to dark and P . syringae infection could be different in LD from those in LL in cca1-1lhy-20 ( compared with wt ) . Alternatively , the amplitude , period , and/or phase of defense gene expression could be different in cca1-1lhy-20 ( compared with wt ) in LD from those in LL . Even different LD conditions could have different effects on clock activity . For example , Michael et al . [62] showed that the set of cycling transcripts increased with the number of different cycling conditions examined . We found that in 12 hr L/12 hr D , expression of GRP7 retained rhythmicity in CCA1ox , compared with that in wt , although the waveform was altered with baseline expression increased ( Figure 8A ) . However , Green et al observed more pronounced alterations in phase of GRP7 expression in CCA1ox ( compared with that in wt ) in seedlings growing in long or short daylengths ( 16 hr L/8 hr D or 8 hr L/16 hr D ) , with maximal transcript accumulation in the dark [12] . Such differences in the patterns of GRP7 transcript abundance could also be due to other reasons besides light conditions . Nonetheless , these observations together with those of Michael et al . [62] emphasize that to better understand the role of the circadian clock in defense control , analyses of defense phenotypes with plants misexpressing CCA1 , LHY , and/or other clock genes should be carried out in LL , DD , and different LD conditions for a comprehensive comparison . Although encountering pathogens at different times in a day , Arabidopsis plants were suggested to be more resistant in the morning than at night . To support this conclusion , wt plants demonstrated higher resistance and/or defense responses when infiltrated during the day than at night [27] , [63] . We also observed similar results in plants infiltrated with P . syringae in LL or LD ( Figure 2 , 4A , and S2 ) , thus supporting this conclusion . However , with spray-infection in LD , we observed the opposite phenotype; wt plants were more resistant at night than in the morning ( Figure 3 , 4B and 8D ) . During spray-infection , P . syringae initially lands on the leaf surface . Further invasion depends on the success of the bacteria in gaining entry into the host tissue via natural openings , such as stomata [1] , [2] . Consistent with enhanced disease resistance to sprayed P . syringae , plants in the evening have much smaller stomatal pore sizes than in the morning ( Figure 5A , 5B , and 8B ) . These two seemly contradicting results actually coalesce to suggest different mechanisms that plants use to defend against pathogens at different times of day , depending on the mode of pathogen invasion . As summarized in Figure 12A , at night plants might rely more on closed stomata to physically restrict pathogen invasion but stomata-independent defense is relatively low . If a pathogen can breach stomatal restriction ( i . e . being pressured into host tissue via infiltration in the laboratory ) at night , it can be more virulent to the host . However , with stomata widely open during the day , plants apparently compensate for enhanced pathogen access to the leaf interior with enhanced stomata-independent defense that is stronger during the day than at night . This cycling in host resistance means that plants can be more resistant to epiphytic pathogens at night than during the day . But in the presence of apoplastic pathogens , plants can activate stronger defense during the day than at night . Taken together , we conclude that plants rely on distinct mechanisms , involving stomata-dependent and stomata-independent defenses , to respond to pathogen attacks at different times of day . Our data suggest that both stomata-dependent and -independent defense can be affected by CCA1 , LHY , and its downstream target GRP7 . Consistent with such a role of CCA1 and GRP7 , these proteins are expressed in guard cells [51] . It is conceivable that CCA1 and/or LHY proteins directly affect the abundance of GRP7 via binding to its promoter at different times of day , which in turn regulates stomatal aperture and thereby stomatal defense ( Figure 11B ) . Since both CCA1 and GRP7 proteins are also found in other cell types besides the guard cells [51] , [64] , it is possible that CCA1/LHY/GRP7 also contribute to stomata-independent defense . GRP7 is unlikely to be the only target of CCA1 and LHY to regulate pathogen defense . First , our bioinformatic analysis suggests that a number of defense genes besides GRP7 might be preferentially regulated by CCA1 and LHY ( Figure 7 ) . And second , plants overexpressing GRP7 are not more susceptible to P . syringae ( J . Alfano and H . Kang , personal communications ) while CCA1ox and LHYox plants are more susceptible to P . syringae ( this study ) . Thus , CCA1 and LHY presumably act through multiple downstream target genes to regulate plant defense . Identification of these additional defense genes controlled by CCA1 and LHY should advance our understanding of the mechanisms by which the circadian clock regulates plant defense . Rhythmic variation in stomatal aperture is known to be regulated by the circadian clock [13] , [37] , [65] . Besides CCA1 and LHY , other genes encoding components of the central oscillator may also affect stomatal defense . For instance , a mutation in EARLY FLOWERING 3 ( ELF3 ) was recently shown to suppress stomatal closure and disease resistance [27] , [66] . ELF3 might act through the FLOWERING LOCUS T gene , which is highly expressed in stomata of the elf3 mutant and has been shown to affect stomatal activity [66] . In addition , the timing of cab expression1-1 ( toc1-1 ) mutant also shows defects in stomatal aperture [59] , [67] . It is tempting to speculate that ELF3-mediated defense is related to its role in stomatal control and TOC1 could also contribute to plant defense . However , further experiments are necessary to validate these speculations . Nevertheless , these observations suggest that the circadian clock can influence stomatal activity and possibly also stomatal defense via different pathways ( Figure 12B ) . Stomata have been proposed as a critical battleground during plant-bacterium interactions [1] , [2] . However , it is not known whether stomatal defense can also restrict the invasion of pathogens with different life styles from those of bacteria . The oomycete pathogen Hpa does not enter host organs through stomata; rather , germinating spores produce hyphae that penetrate between host epidermal cells and extend through the intracellular space in the mesophyll layer . However , near the end of the infection cycle , hyphal tips emerge through the stomata to the exterior of the leaf and then differentiate into spore-bearing structures [68] , [69] . Thus , it is possible that host control of stomatal aperture could influence this stage of the life cycle . Although the role of stomata in defense against Hpa has not been well established , the fact that cca1-1lhy-20 showed enhanced susceptibility to Hpa infection relative to the single mutants and wt suggests such a role of the circadian clock . Interestingly , while conferring enhanced disease susceptibility to P . syringae , CCA1ox heightened resistance to Hpa ( Figure 6A and [28] ) , suggesting that CCA1ox plants employ different mechanisms to defend against these two pathogens . However , it is not clear whether the enhanced Hpa resistance conferred by CCA1ox is related to the circadian clock or to another function resulting from CCA1 overexpression . While regulating multifaceted physiological activities of plants , the circadian clock can also be influenced by external signals , such as changes of light , temperature , hormones , and nutrients [32] , [70] , [71] , [72] , [73] , [74] . Here we show that infection with both virulent and avirulent P . syringae strains shortens circadian period in Arabidopsis ( Figure 10 and S7 ) . We further found that such feedback regulation can be recapitulated with flg22 treatment ( Figure 11A ) . Thus , defense activation can also serve as an input signal to regulate clock activity besides being an output of the circadian clock . Since flg22-triggered callose deposition and expression of genes involved in flg22 sensing and signal transduction were previously shown to be under circadian clock control [27] , we conclude that the clock-defense crosstalk involves flg22 signaling ( Figure 12B ) . Production of SA is circadian regulated [75] , however , activation of SA defense does not affect clock activity ( Figure 11 and S8 and [74] ) . Therefore , SA is an output of the circadian clock but does not serve as an input factor . Since our data showed that CCA1 and LHY act largely independently of SA , we speculate that other circadian clock components may act through SA as an output in defense control . What would be the advantages for plants to have clock-defense crosstalk ? A properly tuned circadian clock enhances growth vigor and confers better survival rate and competitive advantage [11] , [12] , [13] , [14] , [15] , [16] . Regulation of defense by the circadian clock suggests that timing of effective defense against pathogens is crucial for host fitness in the presence of pathogens . However , defense is an energy-costly process intricately connected with plant growth and development . A feedback regulation of the circadian clock by defense activation could be important for the host to balance growth , development , and defense responses , for instance , to redirect the energy from costly disease resistance to primary metabolism . Consistent with this idea , several phytohormones are potential components of the clock-defense circuitry . For instance , auxin regulates clock activity as an input [74] while auxin production and signaling are affected by the circadian clock and thus are clock output events [73] , [76] , [77] . Other hormones , such as abscisic acid , brassinosteroids , cytokinins , and gibberellic acid , have been shown to serve as clock inputs [74] , [78] . Interestingly , cytokinin affects the phase but not the period of the clock [74] , [79] , [80] . However whether these hormones are also on the output pathways of the circadian clock remains to be investigated . On the other hand , ethylene and jasmonic acid production and/or signaling are on the output pathways of the circadian clock [29] , [75] , [81] , [82] , [83] although ethylene does not serve as a clock input in Arabidopsis [82] . The role of jasmonic acid as a clock input is currently unknown . All these phytohormones have been implicated in defense control besides their critical roles in regulating plant growth and development [84] , [85] , [86] . Therefore further investigating the roles of these phytohormones in clock-defense crosstalk should shed light on the molecular mechanisms by which plants employ to regulate growth , development , and responses to pathogen invasion . Such information could potentially lead to a better control of plant growth and resistance to devastating pathogens , ultimately enhancing productivity of plants .
Unless otherwise indicated , all plants used on this paper are in the Columbia-0 ( Col-0 ) background and were grown in growth chambers with a 12 hr light/12 hr dark cycle , light intensity at 200 µmol m−2 s−1 , 60% humidity and 22°C . Single mutants ( acd6-1 , lhy-20 , and grp7-1 ) and plants overexpressing CCA1 ( CCA1ox ) or LHY ( LHYox ) were described previously [11] , [35] , [39] , [48] , [52] . cca1-1 was originally a Wassilewskija allele but was introgressed into Col-0 via five sequential backcrosses . The mutants cca1-1lhy-20 , acd6-1cca1-1 , acd6-1lhy-20 , and acd6-1cca1-1lhy-20 were made by genetic crosses and confirmed with PCR markers corresponding to individual mutations ( Table S4 and [54] ) . CCA1ox ( line #34 ) and grp7-1 seeds were from Elaine Tobin and James Alfano , respectively . P . syringae strains were grown at 28°C with King's B medium ( 10 g proteose peptone , 1 . 5 g K2HPO4 , 3 . 2 ml 1 M MgSO4 , and 5 g glycerol per liter ) containing the appropriate antibiotics for selection . Freshly cultured bacteria were collected , washed once , and resuspended to desired final concentrations in 10 mM MgSO4 for infiltration and spray infections or in sterile water for stomatal aperture measurement and bioluminescence analysis . For infiltration infection , the bacterial solution was pressured into the abaxial side of the fifth to seventh leaves of a plant with a 1 ml needleless syringe . For spray infection , the bacterial solution was mixed with Silwet L-77 ( Lehle Seeds ) at a final concentration of 0 . 04% and sprayed onto plants until the leaf surface was evenly wet . Bacterial growth and disease symptoms were analyzed as described previously [53] . Log transformed bacterial growth was used in statistical analysis . Hyaloperonospora arabidopsidis ( Hpa ) strains were propagated and prepared as previously described [56] , [87] . Seven day-old soil-grown seedlings were sprayed with a spore suspension ( 5×104 spores/ml in water ) containing the virulent strain Hpa Emco5 or the avirulent strain Hpa Emoy2 . Seven days post inoculation , sporangiophores on both sides of cotyledons were counted to determine the level of resistance . Hpa infections were conducted as blind experiments where plant genotypes were unknown to the experimenters until the completion of the experiments . All bacterial and Hpa infection experiments were repeated at least three times unless otherwise indicated . RNA extraction and northern blotting were performed as described [54] . Radioactive probes were made by polymerase chain reaction ( PCR ) with a specific antisense primer for a gene fragment in the presence of [32P] dCTP . Primers used for making probes were listed in Table S4 . Stomatal aperture was measured with 25-day-old plants as previously described [1] . Briefly , the fifth to seventh leaves were taken at the indicated times and mounted onto a glass slide at the abaxial side using Telesis 5 silicone adhesive ( Premiere Products , Inc . , CA ) . The top layer of the leaf was scratched off with a razor blade . Images of at least three random regions of the bottom layer of the leaf were taken immediately with a camera ( Canon Digital Rebel xsi , Japan ) connected to an inverted microscope ( Olympus Model IMT-2 ) . P . syringae treatment was performed at ZT4 when plants had been exposed to light for 4 hr to ensure that most stomata opened . The fifth to seventh leaves of plants were collected and immersed in PmaDG3 or DC3000 resuspensed in sterile water ( 108 cfu/ml ) or in water as mock treatment . At 1 hpi or 3 hpi , treated leaves were harvested and processed for stomata imaging . At least three leaves per genotype and per time point were taken for stomatal images . The stomatal aperture was determined by the ratio between the width and the length of a stoma , which was measured with the assistance of ImageJ ( version 1 . 45 ) . Seedlings expressing the reporter gene LUCIFERASE ( LUC ) under the control of promoters from CCA1 or GRP7 ( At2g21660; also called CCR2 ) [45] , [46] were grown on MS media with 2% sucrose in a 12 hr light/12 hr dark cycle at 22°C for 7–10 days . Seedlings were soaked in P . syringae resuspended in sterile water in the presence of 0 . 04% Silwet L-77 , flg22 ( 1 µM or 10 µM ) , or benzo ( 1 , 2 , 3 ) thiadiazole-7-carbothioic acid ( BTH; a SA agonist ) ( 10 µM or 300 µM ) , and transferred to 96-well plates containing 200 µl of MS media and 30 µl of a 2 . 5 mM D-luciferin solution . Mock treatments were conducted with sterile water with or without 0 . 04% Silwet L-77 . The seedlings were subsequently transferred to LL at 22°C . LUC activity was detected at 1 hr intervals for 7 days with a TopCount luminometer ( Perkin Elmer Life Sciences ) and analyzed with MetaMorph image software [88] . Flg22 was purchased from GenScript USA Inc . and BTH was a kind gift from Robert Dietrich ( Syngenta ) . For cotyledon movement , surface sterilized Arabidopsis seeds were grown on MS media with 2% sucrose for 6 days in a 12 hr light/12 hr dark cycle at 22°C and were transferred to 24-well cloning plates , one seedling per well . The seedlings were entrained for two more days in the 12 hr light/12 hr cycle at 22°C and were subsequently released into LL at 22°C . Cotyledon movement was recorded with multiple surveillance cameras every 20 min for 7 days and post-run image analysis was performed as described [88] . Up to 3000 bp promoter sequences of 571 genes [43] were downloaded from Athena ( http://www . bioinformatics2 . wsu . edu/cgi-bin/Athena/cgi/analysis_select . pl ) [89] . These genes were grouped into three sets , selected ( 337 defense-related gene based on microarray experiments ) , empirical ( 127 empirical marker genes for various pathogen responses ) , and normalization ( 107 non-defense related genes whose expression levels were relatively stable among experiments with pathogen infection ) [43] . Promoters of these genes were analyzed for the enrichment of CBS ( AA[AC]AATCT ) or EE ( AAAATATCT ) motifs , using the online tool POBO ( http://ekhidna . biocenter . helsinki . fi/poxo/pobo/ ) [44] . Pseudo-clusters of 100 promoters of up to 3000 bp in length of Arabidopsis genes , which do not contain the coding sequences of the neighboring genes and were sampled randomly from the entire Arabidopsis genome with 1000 bootstrap replications , were analyzed to generate the background as a control for each motif . The number of the CBS or EE motifs in gene clusters was quantified , using a Perl program . | Plants are frequently challenged by various pathogens . The circadian clock , which is the internal time measuring machinery , has been implicated in regulating plant responses to biotic cues . To better understand the role of the circadian clock in defense control , we tested disease resistance with Arabidopsis mutants disrupted in CCA1 and LHY , two key components of the circadian clock . We found that consistent with their contributions to the circadian clock , cca1 and lhy mutants synergistically affect resistance to both bacterial and oomycete pathogens . Disrupting the circadian clock caused by overexpression of CCA1 or LHY also results in severe disease susceptibility . Thus , our data further demonstrate a direct role of the circadian clock mediated by CCA1 and LHY in defense regulation . We also found that CCA1 and LHY act independently of salicylic acid mediated defense but at least through the downstream target gene GRP7 to regulate both stomata-dependent and -independent pathways . We further show that defense activation by bacterial infection and the treatment with the elicitor flg22 can also feed back to regulate clock activity . Together our study reveals for the first time reciprocal regulation of the circadian clock and plant innate immunity , significantly expanding our view of complex gene networks regulating plant defense responses and development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biology"
]
| 2013 | Crosstalk between the Circadian Clock and Innate Immunity in Arabidopsis |
The nosocomial pathogen Acinetobacter baumannii is a significant threat due to its ability to cause infections refractory to a broad range of antibiotic treatments . We show here that a highly conserved sensory-transduction system , BfmRS , mediates the coordinate development of both enhanced virulence and resistance in this microorganism . Hyperactive alleles of BfmRS conferred increased protection from serum complement killing and allowed lethal systemic disease in mice . BfmRS also augmented resistance and tolerance against an expansive set of antibiotics , including dramatic protection from β-lactam toxicity . Through transcriptome profiling , we showed that BfmRS governs these phenotypes through global transcriptional regulation of a post-exponential-phase-like program of gene expression , a key feature of which is modulation of envelope biogenesis and defense pathways . BfmRS activity defended against cell-wall lesions through both β-lactamase-dependent and -independent mechanisms , with the latter being connected to control of lytic transglycosylase production and proper coordination of morphogenesis and division . In addition , hypersensitivity of bfmRS knockouts could be suppressed by unlinked mutations restoring a short , rod cell morphology , indicating that regulation of drug resistance , pathogenicity , and envelope morphogenesis are intimately linked by this central regulatory system in A . baumannii . This work demonstrates that BfmRS controls a global regulatory network coupling cellular physiology to the ability to cause invasive , drug-resistant infections .
Infections with multidrug-resistant ( MDR ) bacteria pose a serious threat to public health . The ability to withstand multiple antibiotic treatments derives from a diverse array of genetic resistance determinants , many of which are deeply intertwined with aspects of microbial physiology including virulence [1 , 2] . Understanding how intrinsic resistance systems are controlled and how they modulate pathogen-host interactions promises to reveal ways to target MDR bacteria . An MDR pathogen that has emerged as a major cause of recalcitrant opportunistic infections in hospitals worldwide is Acinetobacter baumannii . A . baumannii is a Gram-negative rod that has evolved high rates of resistance to a broad array of antibiotics . Nosocomial diseases caused by A . baumannii are notoriously difficult to treat and manifest commonly as bacteremia , pneumonia , wound infections , and sepsis [3] . Modeling pathogenesis with A . baumannii has presented challenges . Immunocompetent mouse model hosts typically show resistance to the development of lethal disease upon infection with A . baumannii [3] . A number of bacterial products , many of which involve the multilayered cell envelope [4] , have been shown to promote fitness in hosts and also contribute to intrinsic drug resistance [3 , 5–7] . How the production of these factors is controlled by the pathogen remains poorly characterized . Understanding this control circuitry is complicated by the fact that Acinetobacter lacks several canonical systems that regulate antimicrobial stress responses in other organisms [8] . Our previous work has uncovered the existence of a regulatory network that responds to antibiotic stress and controls virulence in A . baumannii [5] . We found that production of an important protective component of the A . baumannii envelope , capsular exopolysaccharide , is transcriptionally upregulated by the antibiotic chloramphenicol at sub-minimal inhibitory concentration ( MIC ) [5] . Such treatment causes a dramatic phenotypic conversion of A . baumannii from an avirulent state to one that can cause lethal disease in mice [5] . Signaling from a highly conserved two-component system ( TCS ) , BfmRS , is required for transcriptional upregulation of capsule in response to antibiotics [5] . The phenotypes of bfmS and bfmR null mutants fit a model in which the BfmS receptor histidine kinase negatively regulates its cognate response regulator BfmR by altering its phosphorylation state [5] . BfmRS is critical for various functions relevant to nosocomial disease . Bacteria lacking BfmR show reductions in biofilm production [9] , resistance to certain antibiotics [10] , survival in human ascites fluid and serum [10] , and fitness in animal hosts [7 , 11] . In a mouse A . baumannii pneumonia model , bfmR mutants are the most severely attenuated within a saturating genomic transposon mutant pool , consistent with the BfmR-controlled regulatory network playing a central role in the disease process [11] . Moreover , bfmR mutants exhibit elongated morphologies [5 , 9] , and bfmS knockout allows bypass of toxicity due to a late-stage capsule biosynthesis defect [5] , consistent with roles for this system in envelope biogenesis . Despite its importance , we understand little about how BfmRS regulates gene expression , and how such control influences antibiotic resistance and the ability of the pathogen to cause opportunistic disease . Here , we have addressed these questions by analyzing the role of BfmRS in jointly promoting broad-range drug resistance , defense from host attack , and bacterial physiology . This analysis reveals that transcriptional control of cell envelope morphogenesis is fundamentally connected to both intrinsic antibiotic resistance and pathogenic potential in A . baumannii .
We previously found that antibiotics that stimulate capsule hyper-production augment the ability of A . baumannii to cause lethal systemic disease [5] . As BfmRS activity is necessary and sufficient for capsule hyper-production [5] , we hypothesized that activation of BfmRS signaling would similarly result in augmented virulence . To test this model , we measured serum killing and virulence of strains carrying bfmRS mutant alleles that are predicted to activate or abolish the TCS control circuit . Exposure of 60% rabbit serum for 1 hour reduced viable counts of the wild-type ( WT ) strain approximately 50-fold ( Fig 1A ) . Deletion of bfmS , predicted to activate the control circuit , resulted in complete resistance to killing by serum ( Fig 1A ) , phenocopying the high-level serum resistance produced by antibiotic pretreatment [5] . In contrast , loss of the TCS due to deletion of bfmRS resulted in efficient killing , consistent with a prior study [10] , with survival levels approximately 10−5 relative to the bfmS deletion ( Fig 1A ) . Killing was dependent on complement activity because no reductions in viability were observed with heat-inactivated serum ( Fig 1A ) . The altered complement susceptibility phenotypes of bfmRS mutants were specific to this locus; reintroduction of WT bfmRS into a ΔbfmRS strain produced levels of serum susceptibility that were similar to WT , while reintroduction of bfmR in the absence of bfmS or in the presence of a catalytically inactive bfmS allele [bfmS ( H234Q ) ] recapitulated the high-level resistance to serum seen in the hyperactive ΔbfmS strain ( Fig 1B ) . Enhanced serum complement resistance is tightly connected to increased virulence potential of A . baumannii in mammalian infection models [12–16] . To determine if loss of BfmS function similarly enhances virulence relative to WT , the mutant was tested in a systemic infection model employing immunocompetent mice which normally survive WT infection and restrict the organism [5 , 13] . Intraperitoneal infection with the ΔbfmS strain caused lethal , septic disease in approximately 60% of the inoculated mice , at a dose that was completely nonlethal with WT bacteria ( Fig 1C ) . These results show that BfmRS controls the ability to develop virulence in A . baumannii in a manner that mirrors the effects of sub-MIC antibiotic treatment . Interestingly , by contrast with sub-MIC chloramphenicol treatment , which partially inhibits bacterial growth [5] , bfmS mutation had no discernable effect on culture growth rates or yield , at least by using simple short term broth growth ( S1 Fig ) . The ΔbfmRS mutant showed logarithmic phase growth that was similar to WT but produced lower post-exponential phase growth yield ( S1 Fig ) . Given the ability of BfmRS to modulate bacterial defense against host antimicrobial killing and its described effects on resistance to some antibiotics [6 , 10] , we determined the ability of this system to control the outcome of a wide range of antibiotic interactions . We first compared bfmRS null and hyperactive mutants for their resistance to a set of 19 antibiotics by profiling populations of each strain for their colony formation efficiency ( CFE ) [5] on medium containing increasing concentrations of antibiotics . From the resulting CFE profiles ( shown in Fig 2B and S2A Fig ) , the MIC of each antibiotic against the different strains was determined ( see Materials and Methods and S3 Table ) . The effects of bfmRS mutation on each drug’s MIC are summarized in Fig 2A . We observed that knockout of bfmRS caused hypersensitivity to the vast majority of antibiotics tested , and the constitutively active ΔbfmS strain caused hyperresistance to a substantial subset ( Fig 2A ) . Although multiple diverse classes of antibiotics were affected , resistance to the β-lactam class of antibiotics , which targets assembly of the peptidoglycan component of the cell envelope , demonstrated the most pronounced dependency on bfmRS activity . Furthermore , in the absence of BfmS , resistance to β-lactams was enhanced for most of the drugs ( Fig 2A ) . In the case of mecillinam and imipenem , MICs of the ΔbfmRS strain were 32- and 16-fold lower , respectively , compared to WT ( Fig 2A and 2B ) . As a point of reference , loss of the key AdeIJK efflux system , a major intrinsic resistance determinant , caused a much lower degree of hypersensitivity ( 4- and 2-fold reductions in MIC with mecillinam and imipenem , respectively , compared to WT; Fig 2B ) , highlighting the profound defect caused by loss of BfmRS . Surprisingly , barely visible pinpoint colonies of the ΔbfmRS strain appeared at high frequency at 0 . 1 μg/ml of imipenem ( Fig 2B , arrowhead , and Fig 2C , circles ) , even though these were not apparent at lower or higher concentrations . One possible explanation is that a BfmRS-independent stress response triggered by imipenem , but normally masked in the WT strain , became apparent at this drug concentration in the bfmRS-null mutant . In the presence of the β-lactams ceftazidime , aztreonam , or sulbactam , the ΔbfmRS strain had modestly decreased CFE vs WT while the constitutively active ΔbfmS strain had much greater growth ( Fig 2A and 2B , S2A Fig ) . Hypersensitivity and hyperresistance with ΔbfmRS and ΔbfmS , respectively , were also seen with additional β-lactam antibiotics including ampicillin , carbenicillin , and cephalexin ( Fig 2A and 2B , S2A Fig ) . These resistance phenotypes were not apparent , however , with some envelope-targeting antibiotics , such as the non-β-lactam antibiotic phosphomycin , which inhibits peptidoglycan precursor synthesis [17] , or the MreB antagonist A22 , which inhibits cell wall synthesis during bacterial elongation [18] ( Fig 2A and S2A Fig ) . This indicates that BfmRS may be particularly protective against the bactericidal properties of β-lactams , which cause cell-wall assembly enzymes to malfunction [19] . The dependence of the observed resistance phenotypes on the function of BfmRS was assessed by recombining a variety of alleles into the ΔbfmRS strain at the parental chromosomal site and assaying for CFE with a subset of antibiotics . We observed that reintroduction of WT bfmRS rescued the WT phenotype ( Fig 2B , S2 Fig , S3 Table ) . Furthermore , introduction of bfmR alone into the ΔbfmRS strain increased CFE to levels identical to the ΔbfmS strain ( S2B Fig ) . Replacement of ΔbfmRS by bfmR together with the bfmS kinase-defective allele ( bfmS ( H234Q ) ) also resulted in hyper-resistance ( S2B Fig ) , consistent with the model that BfmS negatively regulates the resistance-promoting activities of BfmR . Given the enhanced antibiotic resistance and protection from serum killing that resulted from BfmRS activity , we hypothesized that this system would also modulate antibiotic tolerance during transient exposures to high levels of drug . To examine antibiotic tolerance , we measured bacterial viability after challenge with carbenicillin or the fluoroquinolone ciprofloxacin at concentrations well above the MIC ( Materials and Methods ) . With carbenicillin at 320 μg/ml ( 20X the WT MIC ) , viability of the WT strain slowly declined over time ( Fig 3A ) . Loss of culture optical density ( OD ) followed similar kinetics ( Fig 3B ) , consistent with lysis of the killed population . Strikingly , the ΔbfmS mutant showed no loss in viability or culture OD at this same concentration or at a higher concentration ( 640 μg/ml , 20X the MIC of ΔbfmS ) , used to account for the higher MIC compared to WT ( Fig 3A and 3B ) . By contrast , the ΔbfmRS mutant showed rapid killing at both 80 μg/ml ( 20X MIC ) or 320 μg/ml equivalently ( Fig 3A and 3B ) . Reintroduction of bfmRS or bfmR into the ΔbfmRS mutant rescued tolerance to the levels of the WT or hyper-resistant strains respectively ( Fig 3C and 3D ) , confirming that these phenotypes were dependent specifically on BfmRS . High-level survival was also observed with expression of bfmR+bfmS ( H234Q ) in the ΔbfmRS background ( Fig 3C and 3D ) , consistent with loss of BfmS activity causing increased tolerance . As a consequence of BfmRS function , there was a population-wide decrease in killing by carbenicillin accompanied by an increased duration of drug treatment necessary to drop viability . These results are consistent with of BfmRS function supporting increased antibiotic tolerance [20] . In killing assays employing high concentrations of ciprofloxacin ( 10μg/ml , or 20X MIC ) , WT bacteria showed a biphasic time-kill curve , with an initial rapid killing phase followed by a plateau phase between 3 and 6h ( Fig 3E ) consistent with a sub-population of tolerant cells , or persisters [20] . With this same concentration of drug ( which was also 20X MIC of ΔbfmS ) , ΔbfmS bacteria showed a rate of killing during the first hour of exposure that was similar to WT followed by a plateau resulting in a persistent population that had at least 10X CFU higher than WT ( Fig 3E ) . After repassaging several surviving ΔbfmS clones on medium lacking antibiotics , 100% ( 16/16 ) showed levels of ciprofloxacin sensitivity that were unchanged from non-treated parental bacteria , indicating that the enhanced survival rates were not due to mutation . In contrast to ΔbfmS , the ΔbfmRS mutant had an initial rate of killing that was more rapid than WT and had fewer survivors in the persister phase after treatment with either 10μg/ml or 2 . 5μg/ml ciprofloxacin , concentrations that are 80X and 20X the MIC of ΔbfmRS respectively ( Fig 3E ) . These results demonstrate that BfmRS , in addition to modulating virulence and drug resistance , mediates enhanced drug tolerance in A . baumannii . To identify transcriptional activity dependent on BfmRS function , RNA-seq analysis was performed on bfmRS mutants and the isogenic WT strain . Comparing ΔbfmRS with WT , 1827 chromosomal genes ( out of 3638 ) were differentially expressed based on a false-discovery rate ( q value ) < 0 . 05 ( Fig 4A and S1 Dataset ) . Deletion of bfmR alone resulted in 1774 genes showing altered expression , the majority of which overlapped with those altered in ΔbfmRS ( 1531 genes overlapping out of a total set of 2070 , or 74% ) ( Fig 4A and S2 Dataset ) . This is consistent with the model that BfmS signaling is dependent on its cognate response regulator BfmR , and the fact that the two bfmR mutants are phenotypically similar ( S3 Table ) [5] . Comparing ΔbfmS with WT , 600 chromosomal genes were differentially expressed ( Fig 4A and S3 Dataset ) , and these mostly ( ~74%; 443 genes ) overlapped with those whose expression was modified in the ΔbfmRS and ΔbfmR strains . Our A . baumannii reference strain ( ATCC 17978 ) contains 3 plasmids , and transcript levels of many plasmid genes were also affected ( S3A Fig ) , with ΔbfmS by-and-large decreasing expression of plasmid genes relative to WT ( S3B Fig ) . Together these results demonstrate that largescale changes in gene expression occur with change in BfmRS alleles . To link transcriptional reprogramming by BfmRS to function , we performed pathway enrichment analysis ( Materials and Methods ) , identifying the enriched Gene Ontology ( GO ) terms ( Fig 4B ) . Given the opposing phenotypes of ΔbfmS ( UP ) and ΔbfmRS ( DOWN ) mutants , we focused our analysis on the gene subsets whose transcription was up in ΔbfmS or down in ΔbfmRS ( Fig 4B , subsets indicated by boxes ) . In these gene subsets we observed enrichment of several processes related to the cell envelope ( Fig 4B , italicized terms ) . Based on the enrichment of cell wall growth , division , and morphogenesis terms , we examined the full set of proteins in A . baumannii involved in these pathways that had orthologs in E . coli [21] . We found that many of the genes in these pathways were under the coordinated control of BfmRS ( Fig 4C and 4D ) , as were the two chromosomally-encoded β-lactamases ( the AmpC-type cephalosporinase ADC , and the class D β-lactamase OXA-51 ) linked to the “response to antibiotic” term ( UP in ΔbfmS and DOWN in ΔbfmRS; Fig 4F ) . Our analysis found that outer membrane protein assembly pathways were also influenced by the state of BfmRS activity ( Fig 4G ) . Consistent with previous work [5] , expression of K locus genes responsible for capsule and other envelope glycans [14] was also altered as a result of bfmRS mutation ( Fig 4E ) . These findings indicate that global transcriptome modulation by BfmRS involves control of processes central to envelope homeostasis , consistent with a role in resistance to envelope-targeting antibiotics . Further analysis indicated that BfmRS modulates A . baumannii stress responses , particularly at the level of the cell envelope . These include responses to oxidative and osmotic stresses ( Fig 4H and 4I ) , as well as several drug efflux transporters that were modestly up-regulated as a result of the ΔbfmS allele ( Fig 4F ) . Multiple metabolic pathways also had altered regulation as a result of bfmRS mutation; this was the case in gene subsets with lowered transcription in ΔbfmRS , increased in ΔbfmS , or both ( Fig 4B ) as well as in other gene subsets ( S4 Fig ) . To validate our RNA-seq analysis , we measured the effect of BfmRS on transcription of genes associated with various envelope processes by using qPCR and transcriptional reporter fusions . We focused our qPCR analysis on genes encoding β-lactamases , synthases of the lipid carrier undecaprenyl-phosphate ( Und-P ) , proteins involved in cell wall degradation and cell septation , and a key lipoprotein biogenesis protein . Transcription levels of each of these genes were consistent with the RNAseq results , with significant up-regulation in the ΔbfmS strain in all cases , and down-regulation in the ΔbfmRS strain with most of the genes ( Fig 5A ) . To determine whether BfmRS regulates transcription initiation , we constructed transcriptional fusions of a GFP reporter ( Fig 5B ) to the promoter regions of several genes predicted by transcriptome analysis to be BfmRS-regulated and whose products function in the cell envelope . Reporter fusions to promoter regions upstream of oxa51 , slt ( periplasmic lytic transglycosylase responsible for cell wall turnover [18] ) , ygeR ( LytM domain , predicted to control daughter cell separation [22] ) , tolB , ompW ( outer membrane protein ) , ACX60_RS18040 ( predicted lipoprotein ) , and ACX60_RS13710 ( membrane glycoprotein [23] ) each demonstrated higher activity in ΔbfmS and lower activity in ΔbfmRS A . baumannii cells compared to WT ( Fig 5D ) . By contrast , GFP fusion to the promoter of a gene not displaying BfmRS-dependent enhancement of transcription in our RNA-seq analysis ( ACX60_RS09685 , predicted iron acquisition protein ) resulted in highest activity in the ΔbfmRS mutant ( Fig 5D ) . GFP fusion to the region upstream of adc did not result in significant reporter levels in the presence of any bfmRS allele ( Fig 5D ) , suggesting that the construct tested did not contain sufficient information for transcript initiation or stability . We next tested whether BfmRS is capable of regulating a subset of these promoter-reporter fusions in E . coli ( Fig 5C ) , which lacks highly similar orthologs of BfmRS . WT BfmRS stimulated the activity of the slt , tolB , ompW and ACX60_RS18040 promoters when introduced heterologously in E . coli ( Fig 5E ) . Moreover , co-introduction of BfmR with an inactive BfmS variant in which a critical catalytic residue has been substituted ( G494V [5] ) resulted in robust activation of all four reporters ( Fig 5E ) . These results are consistent with direct regulation of these target gene promoters by BfmRS . Reporter fusion to the oxa51 promoter region showed little activity when co-expressed with WT or variant BfmRS in E . coli ( Fig 5E ) , suggesting that transcription initiation from this region requires other elements present in A . baumannii . The ACX60_RS09685 promoter did not show robust stimulation by either WT or hyperactive BfmRS when heterologously expressed in E . coli ( Fig 5E ) , reproducing its behavior in A . baumannii . Multiple pathways related to envelope biogenesis and division showed coordinated transcriptional regulation by BfmRS . Therefore , the following analysis pursued the hypothesis that control of the cell envelope underlies the dramatic resistance phenotypes associated with altered BfmRS signaling . To solidify the connection between BfmRS control of cell wall homeostasis and antibiotic resistance , we next analyzed β-lactamase modulation by BfmRS . Transcription of the two chromosomal genes encoding β-lactamases ( adc and oxa51 ) was greatly increased in ΔbfmS and decreased in the ΔbfmRS mutant ( Fig 5A ) . By using mutants with deletions of these genes , we found that the chromogenic substrate nitrocefin enabled detection of activity from the ADC β-lactamase specifically ( S6A Fig ) , as noted previously [24] . WT A . baumannii bacteria produced a level of β-lactamase activity in cell sonicates that was decreased about 4-fold in ΔbfmRS ( Fig 6A ) . The ΔbfmS strain had WT levels of cell-associated β-lactamase activity , but had greatly increased levels relative to WT secreted into culture supernatants ( Fig 6A ) . Total cell protein in these supernatants was at or near the limit of detection , and at least 40-fold lower than that measured in cell sonicates ( S6B Fig and Materials and Methods ) , arguing against cell lysis playing a major role in enhanced β-lactamase release by the ΔbfmS strain . Increased extracellular secretion of β-lactamase due to bfmS inactivation was observed with additional A . baumannii clinical isolates ( ATCC 19606 and AB5075; Fig 6B ) . Deletion of the ADC β-lactamase resulted in increased sensitivity to mecillinam , ampicillin , and cephalexin , indicating the protein contributes to resistance to these antibiotics ( Fig 6C , S6C Fig , and S4 Table ) . We found that the filtered supernatant fraction from ΔbfmS cultures allowed WT bacteria to form colonies at higher efficiency on plates containing an otherwise inhibitory concentration of cephalexin ( Fig 6D ) . Enhanced resistance was due to ADC , because no growth enhancement was observed with culture supernatants from ΔbfmS Δadc double mutants ( Fig 6D ) . These results demonstrate that β-lactamase production is modulated by BfmRS and enhances resistance to the appropriate antibiotics . To understand further how BfmRS enhances β-lactamase production , we explored its relationship to the well-described β-lactamase control circuit , which depends on cell wall recycling by the AmpG permease and is activated by β-lactam-mediated cell wall damage [25–27] . Stimulation of β-lactamase production by hyperactive BfmR still occurred in the absence of the AmpG permease , consistent with regulation occurring independently of this cell wall recycling component ( Fig 6E; ΔampG ΔbfmS ) . This is in contrast to the prototypical muropeptide-sensing AmpR system , which requires AmpG permease for AmpC β-lactamase induction [28] . In fact , loss of AmpG increased the extracellular levels of β-lactamase in all strains , regardless of the BfmRS allele ( Fig 6E; compare ΔampG strains to isogenic controls ) , consistent with a lack of genetic interdependence between ampG and bfmRS . β-lactamase release in the absence of AmpG was not associated with wholesale release of cellular protein content , consistent with selective secretion rather than cell lysis ( S6D Fig ) . These results demonstrate that β-lactamase control by BfmRS occurs independently of AmpG-promoted recycling , and reveal two parallel pathways for control of β-lactamase production . To gain evidence that BfmRS participates in the response to β-lactam-induced cell wall damage , strains were exposed to graded increases of the β-lactam imipenem . At sub-MIC levels , there were corresponding increases in production of β-lactamase , including substantial release extracellularly ( Fig 6F ) without extensive release of cellular protein content ( S6E Fig ) . Parallel exposure of ΔbfmRS cells to imipenem at doses causing a similar degree of growth inhibition resulted in a muted response ( Fig 6F ) , although there was evidence for some induction of β-lactamase , consistent with a parallel BfmRS-independent response . These results support the model that A . baumannii β-lactamase production has diverged from canonical Gram-negative induction systems , requiring activation of multiple control pathways to maximize protection from β-lactams . Based on several results , we argue that in addition to control of the chromosomal β-lactamases , BfmRS confers β-lactamase-independent resistance to cell wall damaging antibiotics . Lack of BfmRS resulted in hypersensitivity to many β-lactams ( imipenem , carbenicillin , aztreonam , and ceftazidime ) whose potency was not affected by WT levels of the chromosomal β-lactamases ( compare Fig 2A with Fig 6C and S6C Fig ) . Furthermore , the degree of resistance conferred to antibiotics that were clear substrates of the β-lactamases ( mecillinam , ampicillin , and cephalexin ) could only partially account for the decreased overall resistance in the bfmRS null strain , in which β-lactamase production is decreased but not abolished ( compare Figs 2A and 6C ) . Finally , deletion of adc had only minimal effect on the resistance of ΔbfmS to ceftazidime and aztreonam ( S6C Fig; compare with Fig 2B ) , although a role for augmented OXA-51 levels in this enhanced resistance has not been examined . We considered the possibility that BfmRS augmentation of capsular polysaccharide [5] contributes to β-lactam resistance . A multi-gene deletion involving the K locus ( ΔKL3 ) that eliminates capsule production and causes a truncated lipoolosaccharide [5] did not substantially affect resistance levels against carbenicillin or aztreonam in a bfmRS+ strain or in a strain with a truncated ( null ) bfmS allele [5] ( S7A Fig ) . The ΔKL3 deletion showed increased CFE with ceftazidime , particularly in the bfmS-null strain ( S7A Fig ) . These findings argue against the model that BfmRS-mediated resistance to β-lactams derives from increased production of K-locus polysaccharides . We hypothesized that BfmRS control of peptidoglycan growth pathways that enhance cell division contributes to β-lactamase-independent resistance based on gene expression analysis ( Figs 4 and 5 ) . Consistent with this analysis and with a previous report [5 , 9] , we found that mutations in bfmRS had major effects on cell morphogenesis . Under conditions of exponential growth in LB medium in which WT bacteria grow as short rods ( median length 2 . 5 μm ) , length was greatly increased with ΔbfmRS cells ( 3 . 2 μm ) , while ΔbfmS cells assumed a smaller , more spherical morphology having a diameter of 2 . 0 μm ( Fig 7A and 7B; Materials and Methods ) . In addition , alterations in the degree of bacterial chaining were observed . Chains of two or more connected bacteria occurred at higher frequency in the ΔbfmRS strain compared to the WT and ΔbfmS strains ( Fig 7C and S5 Fig ) . These changes indicate that BfmRS increases the frequency of division and cell separation . To test the model that control of cell division and morphogenesis contributes to β-lactamase-independent resistance , we analyzed the effects of bfmRS mutation on A . baumannii morphology after challenge with imipenem . Sub-MIC levels of imipenem ( 0 . 03 μg/ml ) caused bacteria with WT BfmRS to grow as single or diploid spheres ( Fig 7D ) , a phenotype consistent with the known targeting of the cell wall elongation machinery by this antibiotic and related carbapenems [29 , 30] . We reasoned that BfmRS facilitates the intrinsic ability to bypass imipenem-mediated elongation block by promoting growth and division as spheres . Consistent with this notion , at 0 . 03 μg/ml imipenem , a concentration at which growth is dependent on BfmRS ( Fig 2B and S7B Fig ) , ΔbfmRS bacteria showed aberrant morphologies including larger , uneven , less spherical shapes that appeared to reflect division plane defects , including blebbing and incomplete pinching between connected spheroids ( Fig 7D ) . Aberrant non-spherical morphology was also evident with a lower dose ( 0 . 01 μg/ml , Fig 7D ) at which ΔbfmRS CFE is predicted to be 100% ( Fig 2B ) . Furthermore , the loss of the regulatory system appeared associated with osmotic fragility , as the CFE of bfmRS-null cells on graded levels of imipenem approached WT levels in the presence of 7 . 5% sucrose ( Fig 7E , left ) , albeit with smaller colonies than observed for the bfmRS+ control ( S7C Fig ) . Consistent with an osmoprotectant function , sucrose did not significantly modify CFE of WT or ΔbfmS with imipenem , nor did it boost ΔbfmRS CFE on medium containing ciprofloxacin , which does not cause cell lysis ( Fig 7E , right ) . Absence of BfmRS did not cause bacteria to become osmotically fragile when grown in medium lacking antibiotics ( S7D Fig ) , indicating that in the absence of drug-induced stress , the sacculus of this strain provides sufficient protection from irreversible osmotic damage . As an additional test of the model that BfmRS facilitates growth and division pathways bypassing elongation block , we determined the role of BfmRS in protecting from mutational disruption of cell elongation . We isolated mutants lacking the elongation-specific peptidoglycan transpeptidase , PBP2 , which is the target of mecillinam and imipenem [29 , 30] , and determined the genetic interactions of pbp2 mutations with bfmRS alleles . Single deletion of pbp2 produced viable bacteria that grew as spheres , reflective of a block in elongation system function ( Fig 7F ) , and caused slightly reduced growth in the post-exponential phase ( Fig 7G , WT vs Δpbp2 ) . These phenotypes resembled those caused by pbp2 knockout in Pseudomonas aeruginosa [31] . Notably , the ΔbfmRS deletion caused a marked exacerbation of the Δpbp2-associated growth defect ( Fig 7G , ΔbfmRS vs ΔbfmRS Δpbp2 ) , resulting in much lower yield during growth in liquid medium and small colonies on solid medium ( S7E Fig ) . An aggravating genetic interaction was not observed with the constitutively active ΔbfmS allele ( Fig 7G , ΔbfmS vs ΔbfmS Δpbp2 ) . Lack of BfmRS activity therefore strongly modulated the phenotype of the elongation-defective Δpbp2 mutant . The resilience of BfmRS-active cells challenged with varied elongation system lesions is consistent with the ability of BfmRS to permit cell propagation through enhanced division complex levels or activity . In E . coli , cells become hypersensitive to β-lactam inhibition of elongation when insufficient division complex levels are combined with knockout of the key lytic transglycosylase enzyme Slt [19] . Slt activity protects cells by preventing toxic peptidoglycan misincorporation resulting in shape defects [19] . Because absence of BfmRS results in extremely low expression of slt ( Fig 5 ) in the setting of reduced division complex function ( Figs 5 , 7A–7D , 7F and 7G ) , we hypothesized that restoring slt expression in ΔbfmRS bacteria would relieve toxicity due to elongation-targeting β-lactams . When slt expression is increased via induction from a multicopy plasmid , ΔbfmRS cells indeed show elevated resistance to low levels of imipenem ( 0 . 02μg/ml ) ( Fig 7H ) . Restoration of slt expression did not augment resistance at a higher concentration of imipenem ( 0 . 175μg/ml ) , or with ciprofloxacin ( Fig 7H ) . With these conditions , killing is not predicted to be effectively suppressed by osmoprotection ( Fig 7E ) . Together , these results are consistent with a model in which BfmRS counters β-lactam morphogenic toxicity through coordinate regulation of cell division and lytic transglycosylase activity . The above findings indicate that BfmRS protection from cell wall stress is closely linked to control of division and morphogenesis . To determine processes in which intrinsic resistance and division/morphogenesis are co-regulated , we isolated suppressor mutations that bypass ΔbfmRS antibiotic hypersensitivity . Several spontaneous mutants of ΔbfmRS were isolated that could grow on plates containing otherwise inhibitory concentrations of mecillinam or imipenem [29 , 30] . The resulting suppressor mutations were identified by whole-genome sequencing and were found to affect proteins having a range of functions ( S5 Table ) . Several of the affected gene products participate in translation , while a particularly strong suppressor was identified in a gene encoding a predicted Nudix hydrolase ( S5 Table; Fig 8 ) . Mutations altering tRNA synthesis and Nudix hydrolases cause increased cellular levels of ppGpp in other bacterial species [32–35] , with consequent increased resistance to mecillinam [34 , 35] . Mutations in genes without an obvious connection to ppGpp levels or mecillinam resistance were also isolated . These included mutation in the putative genes cspC ( cold-shock family RNA chaperone [36] ) , pbp1a ( bifunctional peptidoglycan synthase [21] ) , ctpA ( periplasmic protease [37] ) and ispB ( isoprenoid quinone synthesis [38] ) . We analyzed the intrinsic resistance and cell morphology of a subset of these suppressors , which were determined by whole-genome sequencing to have a single mutation in cspC ( frameshift ) , mnmA ( tRNA modification enzyme , G362S substitution predicted to be deleterious ) , or ACX60_RS05385 ( NUDIX hydrolase , gene disrupted by insertion sequence ) . These strains each showed increased CFE with mecillinam compared to the parental ΔbfmRS strain , with particularly robust suppression resulting from disruption of the NUDIX hydrolase ( Fig 8A , top ) . Resistance to sulbactam , which targets cell division , was also augmented ( Fig 8A , bottom ) , while in the case of mnmA and NUDIX mutations , survival in serum was enhanced ( Fig 8B ) . The increased resistance phenotype in these strains was linked to restoration of a shorter rod shape comparable to WT ( Fig 8C and 8D ) . With other suppressors , cell morphology was either not dramatically different from the parent ( pbp1a transpeptidase domain mutant , and ispB catalytic domain mutant ) or resulted in irregularly shaped , thin rods ( ctpA transposon insertional mutant ) ( S8A Fig ) . In suppressors showing restoration of a shorter , WT-like rod shape , transcription of multiple cell division and peptidoglycan degradation genes was also increased relative to the ΔbfmRS parent , indicating that they have effects on levels of bacterial transcription of a large swath of genes associated with the cell envelope ( Fig 8E ) . The phenotypes resulting from mutations in these strains link increased intrinsic resistance with proper control of cell division , phenocopying major functions of BfmRS .
Virulence and antibiotic resistance are intertwined in A . baumannii [2 , 6 , 39 , 40] . In this work we demonstrated that the pathogen employs BfmRS to coordinate both its ability to cause disease and to resist a broad range of antibiotics . We revealed that BfmRS functions to globally influence the transcriptome . Expression of genes involved in biogenesis and growth of multiple envelope components ( in addition to capsule [5] ) was highly dependent on the state of BfmRS signaling , pointing to transcriptional control of envelope homoeostasis as a major mechanism of resistance to various forms of antimicrobial attack . With the cell-wall targeting β-lactam antibiotics , BfmRS controls resistance through two strategies: β-lactamase production and physiological resistance to cell wall lesions . Our results indicate that BfmRS co-regulates the envelope biogenesis and division machineries , connecting its control of pathogenicity and drug resistance . This study demonstrates that a single genetic locus can control conversion of A . baumannii to a state of increased pathogenicity . A transition to increased virulence is driven by growth on sub-MIC levels of the translation inhibitor chloramphenicol [5] , and translation inhibition appears connected to BfmRS signaling . Full induction of capsular polysaccharides by chloramphenicol treatment requires BfmRS [5] , and several mutations suppressing bfmRS-null phenotypes isolated in this work are predicted to affect translation . Although our suppressor analysis has pointed to candidates that may act as transmitters of translation-level stress , such as mutations that affect the levels of ppGpp and cold shock proteins ( discussed further below ) , the precise signaling network connecting BfmRS with ribosomal stress is unclear . In addition , the end-targets under transcriptional regulation that mediate control of virulence remain to be elucidated . The breadth of the BfmRS modulon suggests that many factors are likely involved . In dissecting how this regulatory network functions in disease processes , an important approach would be to expand the use of animal models including those that directly examine bacteremia , pneumonia , and other forms of A . baumannii disease . The BfmRS system controls broad-range intrinsic antibiotic resistance in addition to determining pathogenicity . BfmRS-deficient mutants showed hypersensitivity to many antibiotics with a diverse range of targets , indicating that the system maintains the function of broadly protective intrinsic resistance factors . Transcriptome profiling pointed to multifaceted coordination of cell envelope processes which potentially bolster the envelope permeability barrier . This is supported by the low-level hypersensitivity of ΔbfmRS to hydrophobic antibiotics such as rifampicin and erythromycin , although the presumed envelope defects were not sufficient to cause altered susceptibility to the large peptide antibiotic vancomycin ( Fig 2A ) . BfmRS control of interactions with antibiotics appears to have multiple layers , as different bfmRS alleles caused impressive changes in resistance against β-lactams antibiotics , as well as altered persistence in assays after transient exposure to high concentrations of diverse antibiotics . We uncovered that BfmRS coordinates multiple defenses against cell wall lesions that can be roughly grouped into β-lactamase-dependent and -independent pathways . With the β-lactamase-dependent pathway , resistance is conferred against a subset of β-lactams through a non-canonical regulatory strategy that does not depend on traditional AmpG-mediated cell wall recycling . We also provide evidence that BfmRS controls a β-lactamase-independent resistance strategy , linked to the ability to counter toxicity induced by β-lactams . Our data support a model in which BfmRS maintains the integrity of the peptidoglycan sacculus in the face of β-lactam stress by directly controlling cell division and cell wall degradation pathways ( in which Slt plays a major role ) , preventing accumulation of toxic peptidoglycan structures . In addition , BfmRS may allow cells to counteract depletion of cell wall precursors , which is another aspect of β-lactam intoxication [19] . BfmRS increases transcription of many genes involved in production of precursors of the cell wall and other envelope structures . One key precursor is Und-P , a lipid carrier for intermediates in the biosynthesis of envelope carbohydrate polymers including peptidoglycan and capsule [41] . Increased Und-P levels may limit β-lactam toxicity and may also account for the ability of bfmS mutation to suppress lethality due to capsule assembly block [5] , a type of lesion that depletes Und-P due to sequestration into dead-end intermediates [42] . We isolated suppressor mutants that bypassed the β-lactam hypersensitivity caused by bfmRS deletion , to identify processes that may couple resistance to control of cell morphogenesis and possibly collaborate with BfmRS signaling . One set of suppressors included a cold shock family protein ( Csp ) . Csps form a large family of RNA-binding proteins that regulate gene expression [36] which can either be induced in response to low temperature or produced constitutively at 37°C [43] . Interestingly , some Csps are induced by treatment with chloramphenicol , a stress that also stimulates capsule overproduction and transition to a hypervirulent state in A . baumannii [5] . The A . baumannii csp gene that was affected in our suppressor analysis was very highly expressed during growth at 37°C ( S1–S3 Datasets ) , indicating that it probably has important functions at a variety of temperatures . An additional large group of suppressors of ΔbfmRS hypersensitivity had mutations predicted to slow translation . Many of these mutations are predicted to result in lower cellular aminoacyl tRNA levels , which is known to elevate ppGpp levels and activate the stringent response [35] . This response increases mecillinam resistance in E . coli for unknown reasons , although enhancements to peptidoglycan precursor pools or cell division capacity , and reductions in overall peptidoglycan synthesis and turnover , have been proposed as possible mechanisms [44–46] . Interestingly , BfmRS modulates the expression of several of the tRNA synthesis or modification genes that had suppressor mutations ( S8B Fig ) , consistent with feedback interactions in which control of translation rates interplays with envelope morphogenesis and division . We propose that the Acinetobacter BfmRS signaling cascade enhances physiological properties that are known in other Gram-negatives to be activated upon entry into the post-exponential phase of growth [47] . This model is supported by multiple findings: BfmRS increased tolerance to bactericidal antibiotic exposures , drove the construction of a cellular morphology that had reduced size relative to strains lacking the circuit , enhanced growth yield in post-exponential phase , and controlled the expression of post-exponential phase-associated genes known to protect against oxidative and osmotic stresses and redirect metabolism [47] . BfmRS is also known to mediate biofilm formation [9] , a lifestyle associated with post-exponential phase transition or starvation in Acinetobacter [48] . In E . coli and related Gram-negative bacteria , transcriptional reprogramming during transition to post-exponential phase is governed by the alternative sigma factor σS ( RpoS ) , whose levels and/or activity are augmented by a histidine-kinase signaling system ( Rcs ) [49] . Acinetobacter does not encode an ortholog of RpoS , or of the several components of the Rcs phosphorelay system , indicating that responses to starvation and other stresses are occurring via noncanonical pathways . Given the similarities between phenotypes of null and hyperactive alleles of BfmRS and Rcs , including antibiotic resistance [50] , capsule biosynthesis [5] , and biofilm formation [9 , 51] , BfmRS may represent a functional counterpart to the Rcs phosphorelay system . It is almost certain that BfmRS , like Rcs , acts both directly on target gene promoters as well as indirectly through additional regulators to exert control on its expansive modulon . Direct control of targets modulating the envelope ( including those identified in Fig 5E ) could have pleiotropic effects accounting for many indirect changes in gene expression . Based on the multiple cell envelope processes under BfmRS control , and the contribution of BfmRS to imipenem-induced β-lactamase production , envelope-level stress represents a potential input that is sensed by this system . Understanding the global control network in which BfmRS plays a key role will illuminate how A . baumannii optimizes its ability to grow under stress conditions and transitions to states of augmented virulence and resistance . Targeting this system therapeutically represents an attractive strategy to enhance the effectiveness of antibiotic treatments while inhibiting pathogenesis .
The bacterial strains used in this study are listed in S1 Table . A . baumannii strains were derivatives of ATCC 17978 unless otherwise noted . Bacteria were grown in Lysogeny Broth ( LB ) ( 10 g/L tryptone , 5 g/L yeast extract , 10 g/L NaCl ) . Gentamicin ( Gm , 15 μg/ml ) , kanamycin ( Km , 10 μg/ml ) , carbenicillin ( Cb , 100 μg/ml ) , and/or tetracycline ( Tc , 10 μg/ml ) ( Sigma ) were used at noted concentration in solid medium . Broth cultures were grown at 37°C in flasks with orbital shaking at 250 rpm or in tubes with rotation on a roller drum at 56 rpm . Growth was monitored by measuring absorbance spectrophotometrically at 600nm ( A600 ) . Plasmids and oligonucleotides ( Integrated DNA Technologies ) used in this study are listed in S1 and S2 Tables , respectively . All constructs were sequenced ( Genewiz ) before introduction into A . baumannii via electroporation . ΔbfmRS::aacC1 and ΔbfmR::aacC1 mutants were isolated via homologous recombination with previously described pSR47S ( KmR ) -based constructs [5] , using methods that minimized exposure to antibiotics during strain construction . KmS recombinant deletion mutants were isolated from GmR KmR merodiploids by sucrose counterselection on medium lacking Gm . Markerless , in-frame deletions of additional genes were isolated via homologous recombination as described previously [5] . To engineer altered bfmRS alleles for complementation tests , inverse PCR was performed on fragments cloned in pUC18 . The bfmR+ΔbfmS allele was generated with primers that encoded a 3X-FLAG tag fusion to the C-terminus of BfmR . The bfmR+bfmS ( H234Q ) allele was generated by using primers containing the engineered point mutation and a template encoding a 3X-FLAG tag fusion to the C-terminus of BfmS . Constructs were subcloned to pEGE148 and introduced in single copy at the ΔbfmRS::aacC1 locus via homologous recombination . Chromosomal alleles of bfmRS were also cloned in pJB1801 for expression in E . coli . A vector for transcriptional fusions to GFP , pEGE245 , was constructed by replacing the EcoRI-PstI fragment of pWH1266 with a synthetic oligonucleotide containing tandem rrnBT1 & T7Te terminators and a SacI restriction site , followed by incorporation of the SacI-PstI fragment of pFPV25 ( gift from Raphael Valdivia , Addgene plasmid # 20667 ) containing gfpmut3 . Promoter fragments upstream of the indicated genes were cloned into pEGE245 directly upstream of the promoterless gfpmut3 gene . A vector for inducible expression , pEGE305 , was constructed by replacing the EcoRI-PstI fragment of pWH1266 with a polylinker , followed by the lacIq T5-lacp fragment cloned from pCA24N-dinB ( gift of Veronica Godoy ) . The slt gene was cloned directly downstream of the T5-lacp promoter in pEGE305 . 8–10 week old female C57BL/6 mice were used for intraperitoneal infections . WT and ΔbfmS A . baumannii were grown in LB to early post-exponential phase . Systemic infections were initiated by intraperitoneal injection of approximately 108 bacteria suspended in 100ul PBS as previously described [5] . Animals showing signs of severe morbidity were euthanized by CO2 asphyxiation followed by cervical dislocation . Experiments were carried out in accordance with protocols approved by Tufts University Institutional Animal Care and Use Committee ( B2014-91 ) and in adherence with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . Approximately 105 bacteria grown to early post-exponential phase were diluted in PBS and incubated with 60% baby rabbit complement serum ( AbD Serotec ) for one hour at 37°C . Serum inactivated by heating ( 56°C for 30 minutes ) was used as a control . Reactions were stopped by placing on ice , and viable bacterial counts were determined by plating serial dilutions on LB agar followed by overnight incubation at 37°C . Antibiotic resistance was measured by CFE assays utilizing solid medium containing 2-fold serial dilutions of antibiotic as previously described [5] . The lowest antibiotic concentration at which CFE is less than 10−3 defines the MIC . MIC values were used to calculate relative resistance levels of mutant strains compared to WT . For assays measuring bacterial viability after transient treatment with antibiotics , bacteria were grown to early post-exponential phase ( A600 ~1 . 5–2 ) , antibiotic was added , and cultures continued growth for 6 hours , with samples taken at the indicated time points . Optical density of samples was measured as A600 , and the density of viable bacteria ( colony forming units ( CFU ) /ml ) was determined by washing samples with PBS , serially diluting in PBS , and plating onto solid medium lacking antibiotics . Bacteria surviving ciprofloxacin tolerance assays were tested for mutational resistance by passaging on LB agar without antibiotics , then assessing growth relative to naïve parental bacteria on solid medium containing 0 . 5 and 1 μg/ml ciprofloxacin . Mid-log phase bacteria were immobilized in agarose pads ( 1% in PBS ) , and images were acquired via phase-contrast on a Zeiss Axiovert 200m microscope with 100x/1 . 3 lens . Length of individual cells and chain number was measured by the Oufti software package [52] . Bacteria were cultured and harvested as for RNA-seq . RNA was purified and cDNA synthesized as described previously [5] . cDNA was amplified with Power SYBR Green Master Mix ( Applied Biosystems ) via a StepOnePlus system according to the manufacturer’s instructions for two-step RT-PCR . Target amplification efficiency was assessed by generating a standard curve with a dilution series of cDNA and was determined to be >95% for each primer pair . Controls lacking reverse-transcriptase were performed to confirm lack of signal from residual genomic DNA . Gene expression levels were quantified by using the 2-ΔΔCt method with two endogenous controls ( 16S and rpoC ) . Bacteria containing derivatives of pEGE245 were grown to early post-exponential phase in LB without antibiotics ( A . baumannii ) or LB with carbenicillin and tetracycline at 20 and 3 . 5 μg/ml , respectively ( E . coli with pJB1801::bfmRS ) . Cultures were transferred to 96-well plates ( Costar 3631 ) and diluted with PBS . Fluorescence ( excitation at 480 nm/emission at 520 nm ) and A600 were measured with a Biotek SynergyMx plate reader . Fluorescence units ( RFU ) were normalized by dividing by A600 , and RFU/A600 of parallel-cultured control bacteria containing the vector ( pEGE245 ) alone was subtracted to calculate background-corrected reporter fluorescence values . Bacteria were grown to early post-exponential phase , centrifuged , and cell pellets and supernatants collected . Cells were resuspended in ice-cold 0 . 1 M phosphate buffer ( pH 7 ) and periplasmic contents liberated by sonication for 2 min ( 10 s on , 5 s off duty cycle ) with a high-intensity cuphorn sonifier ( Branson ) chilled to 4°C . Samples were clarified by centrifugation , and extracts were diluted 5–10 fold in 0 . 1M phosphate buffer ( pH 7 ) . β-lactamase content was assayed spectrophotometrically with the colorimetric substrate nitrocefin ( 20μg/ml in 0 . 1M phosphate buffer , pH 7 ) by measuring absorbance at 486 nm every minute for 15 minutes at room temperature . β-lactamase activity was calculated as initial reaction rate ( Vmax ) /culture density ( A600 ) *dilution factor . Supernatant samples were assayed as above either directly or after dilution in phosphate buffer . In antibiotic treatment experiments , mid-log phase cultures ( A600 ~0 . 2 ) were treated with imipenem for 2 hours and processed as above . Total protein concentration was measured via Bradford microtiter assay ( Bio-Rad ) using BSA as standard . Bacteria were grown to mid-logarithmic or early post-exponential phase and then diluted 100-fold into ultrapure water . After incubation for 20-30minutes , 1/10th volume of 10X PBS was added , and bacteria were serially diluted with 1X PBS and spotted onto LB agar plates for enumeration of surviving CFU . Genomic library preparation for Illumina sequencing was performed using the small-volume Nextera tagmentation method [58] , with the following modifications: NEBNext High-Fidelity 2X PCR Master Mix ( NEB ) was used for amplification; adapter addition/library amplification was followed by 5 cycles of reconditioning PCR with adapter-specific primers; library QC/quantitation was performed directly after PCR by visualizing a sample of each library separated on a 2% agarose/TAE gel; multiplexed PCRs were purified by using a single Qiagen PCR purification column and size-selected via a PippinHT . Libraries were sequenced ( single-end 100bp or 150bp ) on a HiSeq2500 at TUCF-Genomics . Reads were aligned to the A . baumannii 17978-mff genome and variants identified by using Geneious software [59] or BRESEQ [60] . The predicted functional impact of substitution variants was determined by using PROVEAN [61] . | Infections with the hospital-acquired bacterium Acinetobacter baumannii are highly difficult to treat . The pathogen has evolved multiple lines of defense against antimicrobial stress , including a barrier-forming cell envelope as well as control systems that respond to antimicrobial stresses by enhancing antibiotic resistance and virulence . Here , we uncovered the role of a key stress-response system , BfmRS , in controlling the transition of A . baumannii to a state of heightened resistance and virulence . We show that BfmRS enhances pathogenicity in mammalian hosts , and augments the ability to grow in the presence of diverse antibiotics and tolerate transient , high-level antibiotic exposures . Connected to these effects is the ability of BfmRS to globally reprogram gene expression and control multiple pathways that build , protect , and shape the cell envelope . Moreover , we determined that resistance-enhancing mutations bypassing the need for BfmRS also modulate envelope- and morphology-associated pathways , further linking control of physiology with resistance in A . baumannii . This work uncovers a global control circuit that shifts cellular physiology in ways that promote hospital-associated disease , and points to inhibition of this circuit as a potential strategy for disarming the pathogen . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"antimicrobials",
"cell",
"walls",
"medicine",
"and",
"health",
"sciences",
"cell",
"cycle",
"and",
"cell",
"division",
"drugs",
"cell",
"processes",
"immunology",
"microbiology",
"dna",
"transcription",
"antibiotic",
"resistance",
"clinical",
"medicine",
"antibiotics",
"materials",
"science",
"pharmacology",
"hypersensitivity",
"cellular",
"structures",
"and",
"organelles",
"macromolecules",
"materials",
"by",
"structure",
"polymers",
"polymer",
"chemistry",
"antimicrobial",
"resistance",
"gene",
"expression",
"chemistry",
"cell",
"biology",
"clinical",
"immunology",
"peptidoglycans",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences"
]
| 2018 | A global regulatory system links virulence and antibiotic resistance to envelope homeostasis in Acinetobacter baumannii |
Genetics is believed to have an important role in intellectual disability ( ID ) . Recent studies have emphasized the involvement of de novo mutations ( DNMs ) in ID but the extent to which they contribute to its pathogenesis and the identity of the corresponding genes remain largely unknown . Here , we report a screen for DNMs in subjects with moderate or severe ID . We sequenced the exomes of 41 probands and their parents , and confirmed 81 DNMs affecting the coding sequence or consensus splice sites ( 1 . 98 DNMs/proband ) . We observed a significant excess of de novo single nucleotide substitutions and loss-of-function mutations in these cases compared to control subjects , suggesting that at least a subset of these variations are pathogenic . A total of 12 likely pathogenic DNMs were identified in genes previously associated with ID ( ARID1B , CHD2 , FOXG1 , GABRB3 , GATAD2B , GRIN2B , MBD5 , MED13L , SETBP1 , TBR1 , TCF4 , WDR45 ) , resulting in a diagnostic yield of ∼29% . We also identified 12 possibly pathogenic DNMs in genes ( HNRNPU , WAC , RYR2 , SET , EGR1 , MYH10 , EIF2C1 , COL4A3BP , CHMP2A , PPP1CB , VPS4A , PPP2R2B ) that have not previously been causally linked to ID . Interestingly , no case was explained by inherited mutations . Protein network analysis indicated that the products of many of these known and candidate genes interact with each other or with products of other ID-associated genes further supporting their involvement in ID . We conclude that DNMs represent a major cause of moderate or severe ID .
Intellectual disability ( ID ) is defined by significant impairment of cognitive and adaptive functions with onset before 18 years of age . It has an estimated worldwide prevalence of 1–3% , with moderate or severe forms of ID ( IQ<50 ) affecting up to 0 . 5 % of the population in Western countries [1] . We and others have reported that de novo point mutations ( including single nucleotide substitutions ( SNVs ) and small insertions/deletions , referred herein collectively as DNMs ) play a significant role in the genetics of ID [2]–[5] . Similarly , DNMs were found to be implicated in the etiology of other neurodevelopmental disorders overlapping with ID , such as autism spectrum disorders ( ASD ) , epileptic encephalopathy and schizophrenia [6]–[10] . DNMs represent the most extreme form of rare genetic variations; they are more deleterious , on average , than inherited variations because they have been subjected to less stringent evolutionary selection . Importantly , they provide a mechanism by which early-onset reproductively lethal diseases remain frequent in the population . This makes these mutations prime candidates for causing diseases that occur sporadically , and that decrease the reproductive fitness and incur a large degree of selection against phenotypes such as ID . Based on these considerations , we hypothesized that the contribution of DNMs is greater in more severe forms of ID . In order to explore this hypothesis , we performed high-depth exome sequencing in 41 trios consisting of individuals with moderate or severe ID and their healthy parents and assessed the contribution of DNMs to this condition .
We performed exome sequencing in 41 individuals with ID and their unaffected parents . We identified a total of 83 putative DNMs in as many genes within both coding and consensus splice site sequences . Sanger sequencing confirmed 81 of these as de novo and 2 as inherited from one of the parents ( Table S1 ) . All of these DNMs were represented by ≥25% of reads , suggesting that they are unlikely to be associated with somatic mosaicism . The fact that the mutant and wild-type peaks on Sanger chromatograms were comparable in size is consistent with this conclusion . The average DNM rate per trio was 1 . 98 , with only 3 trios containing no detectable DNMs ( Figure 1 ) . The observed de novo SNV rate in the consensus coding sequences ( CCDS ) was 1 . 56 events per trio or 2 . 58×10−8 per base per generation ( 64 SNVs in 2 , 477 , 702 , 175 CCDS bases sequenced at ≥10× in the 41 affected individuals ) , which is significantly higher than the expected population rate of 1 . 65×10−8 ( R binomial test , p = 0 . 0007 ) , or than the ones experimentally determined from exome sequencing studies in control trios ( 1 . 28×10−8 and 1 . 51×10−8 ) [2] , [4] . Considering only de novo SNVs affecting the coding and the canonical splice sites ( AG , GT at intronic positions −1/−2 and +1/+2 of the acceptor and donor splice sites , respectively ) , 73% were missense and 11% were nonsense and canonical splice site mutations . We found a significant excess of these de novo nonsense and splice site mutations in the probands of our cohort when compared to data from exome sequencing of 54 control trios with no family history of ID [4] , [11] or of 593 quartets , including unaffected siblings of individuals with ASD ( R binomial test , p = 0 . 0015 and p = 0 . 02 , respectively ) ( Table 1 ) [7] , [9] , [10] . Such an excess of deleterious DNMs suggest that at least a subset of them are pathogenic . Twelve DNMs were found in as many probands in genes previously associated with ID based on the documentation of deleterious DNMs in at least 4 unrelated individuals with similar phenotypes . Nine of these DNMs are Loss-of-Function ( LoF ) variants ( nonsense , frameshift and canonical splice variants ) and affect the following genes: ARID1B [OMIM 614556] [12] , CHD2 [OMIM 602119] [4] , [13] , FOXG1 [OMIM 164874] [14] , GATAD2B [OMIM 614998] [2] , [15] , MBD5 [OMIM 611472] [9] , [13] , [16]–[18] , MED13L [OMIM 608771] [7] , [19]–[21] , SETBP1 [OMIM 611060] [4] , [22] , TCF4 [OMIM 602272] [4] , [23]–[25] , and WDR45 [OMIM 300526] [26] , [27] ( Tables 2 and 3 ) . None of these 9 DNMs were found in public SNP databases . The phenotype of each of the probands is consistent with that of subjects previously described with mutations in these respective genes , with two exceptions ( Text S1 ) . Although truncating mutations in CHD2 have been reported in individuals with epileptic encephalopathy [4] , [6] , [13] , the individual described herein with a CHD2 frameshift mutation has no history of epilepsy , suggesting that LoF mutations in CHD2 are associated with greater clinical heterogeneity than initially expected . Another example of a gene associated with clinical heterogeneity in our dataset is SETBP1 . Missense mutations clustering in a conserved 11-bp coding region of SETBP1 have been reported to cause Schinzel-Giedon syndrome ( OMIM 269150 ) , a condition characterized by severe ID and specific craniofacial features [22] . In contrast , our case carried a de novo truncating mutation in SETBP1 and showed moderate non-syndromic ID without the typical craniofacial features of Schinzel-Giedon syndrome . Recent studies reported a similar phenotype in individuals with a truncating mutation in SETBP1 or microdeletions encompassing SETBP1 [4] , [28] . Collectively , these observations suggest that SETBP1 haploinsufficiency results in a different phenotype than that induced by the missense mutations reported in Schinzel-Giedon syndrome , which presumably lead to a gain-of-function or a dominant negative effect [22] . We conclude that all of these 9 DNMs are likely to be pathogenic . The three other DNMs in genes previously associated with ID include an in-frame insertion in GABRB3 [OMIM 137192] , a missense in TBR1 [OMIM 604616] and a missense in GRIN2B [OMIM 138252] ( Tables 2 and 3 ) . All of these DNMs affect conserved residues and are predicted to be damaging . Moreover , none of them were found in public SNP databases . Damaging missense mutations in GABRB3 have been previously documented in cases with ID and intractable epilepsy with various types of seizures [6] . Individual 1843 . 647 also showed ID and intractable epilepsy with a similar pattern of seizures as these cases ( Text S1 ) . DNMs in TBR1 have been found in patients with ID and the variable presence of ASD or growth retardation [8] , [9] , [21] , [29] , [30] . Individual 121 . 83 displayed a phenotype similar to previously described cases , including ID , ASD and growth retardation ( Text S1 ) . Finally , DNMs in GRIN2B have been associated with ID of variable severity with or without ASD and epilepsy [2] , [6] , [31] , [32] . Individual 838 . 321 showed severe ID , not walking and saying only one word at 16 years of age ( Text S1 ) . He has never had any seizures though his EEG revealed multifocal epileptic activity . Similar patterns of cognitive impairment were also reported in other patients with DNMs in GRIN2B [2] , [32] . Interestingly , the mutation identified in our case affects a residue located in the ligand-binding domain of the protein , like previously reported de novo missenses in GRIN2B [32] . We conclude that these three DNMs are also likely to be pathogenic . Among the remaining cases , 22 have predicted-damaging DNMs , including 7 LoF mutations ( 5 frameshifts , 1 nonsense , 1 consensus splice site ) , 13 missenses , 1 deletion , and 1 synonymous mutation whose predicted effect on splicing was confirmed by RT-PCR ( Figure S1 ) . Interestingly , deleterious DNMs in 6 of these genes ( HNRNPU [OMIM 602869] , WAC [OMIM 615049] , RYR2 [OMIM 180902] , MYH10 [OMIM 160776] , EIF2C1 [OMIM 606228] , COL4A3BP [OMIM 604677] ) have previously been reported in at least one individual with ID . We discuss hereafter the DNMs that we identified in these genes ( Tables 2 and 3 ) . HNRNPU [OMIM 602869] codes for a highly conserved protein that binds RNAs and mediates different aspects of their metabolism and transport . Chromosome 1q44 microdeletions have defined a critical region associated with ID and seizures that encompasses HNRNPU as well as two other genes [33] , [34] . Two truncating and one splice mutations in HNRNPU were subsequently identified in individuals with ID and seizures [6] , [13] , [24] . Two of these mutations occurred de novo whereas the origin of the other one was not elucidated . One of these individuals also showed ASD whereas the case with the splice mutation displayed syndromic features , including panhypopituitarism , bifid great toe and vertebral segmental defects . We identified an individual ( 1464 . 524 ) who carries a de novo truncating mutation ( c . 511C>T , p . Gln171* ) in HNRNPU . This mutation is located in an upstream coding exon present in all isoforms , thus having the potential to induce nonsense mRNA mediated decay [35] . Moreover , inspection of the Exome Variant Server ( EVS ) database ( 6500 exomes ) revealed no LoF variants in HNRNPU , indicating that haploinsufficiency of this gene is not tolerated . Our case displayed ID , epilepsy and ASD ( Text S1 ) , a phenotype that is similar to that of the other non-syndromic cases with DNMs in this gene , further supporting its involvement in ID . WAC encodes a nuclear protein that interacts with RNF20/40 to regulate histone H2B ubiquinitation , chromatin organization , and gene transcription [36] . De novo microdeletions encompassing WAC and a nonsense DNM in WAC in individuals with severe ID were recently reported [2] , [37] . Our subject ( 762 . 297 ) carries a truncating mutation in WAC ( c . 263_266del , p . Glu88Glyfs*103 ) . This mutation is located in an upstream coding exon present in all isoforms . Inspection of the EVS database revealed no LoF variants in WAC . Individual 762 . 297 showed moderate ID without any distinguishing features on clinical examination and brain imaging , a phenotype that is consistent with that observed in the previously reported patient with a truncating mutation in this gene ( Text S1 ) [2] . Our finding , thus , further supports the involvement of WAC in ID . RYR2 encodes the cardiac and brain-expressed calcium release channel ryanodine receptor 2 . Mutations in RYR2 are typically associated with exercise-induced ventricular and atrial arrhythmias . Virtually all reported mutations in RYR2 are missenses or in-frame deletions that are believed to confer a gain of function , resulting in an increase of Ca+ release [38] , [39] . We identified an individual ( 341 . 162 ) with ID , seizures , short stature and severe atrial arrhythmias ( Text S1 ) who carries a predicted-damaging de novo missense mutation in RYR2 ( c . 14864G>A , p . Gly4955Glu ) . Interestingly , 3 patients with seizures have previously been reported with DNMs in RYR2: 1 ) an individual with epileptic encephalopathy but presumably without a history of arrhythmia was recently found to carry a nonsense mutation ( c . 9568C>T , p . Arg3190* ) in RYR2 [6]; this DNM might not be disease-causing considering that the pathogenic impact of truncating mutations in RYR2 remains unclear and that inspection of EVS revealed 5 different heterozygous LoF mutations in RYR2; 2 ) an individual with cognitive impairment , intractable seizures , short stature and subclinical ventricular tachycardia was found to carry a missense mutation ( c . 12563T>C , p . Leu4188Pro ) [40]; and 3 ) an individual with intractable seizures but without cognitive impairment and arrhythmia was described with a missense mutation ( c . 14803G>A , p . Gly4935Arg ) [41] . It is noteworthy that the mutation found in this latter individual is in close proximity to that of our subject , affecting a highly conserved C-terminal region of the protein . Interestingly , mice heterozygous for the missense mutation p . R2474S in Ryr2 display generalized seizures and arrhythmias [42] . More recently , two brothers with ID , seizures and atrial arrhythmias were found to carry a missense mutation in CLIC2 ( OMIM 300138 ) , which maps to the X chromosome [43] . CLIC2 is a negative regulator of RYR2 . The mutation was shown to stimulate the release of Ca2+ by keeping the RYR2 channel in an open state , possibly due to a higher binding affinity for the RYR2 protein . The specificity of the phenotype observed in our subject and its similarity with that of other individuals with DNMs in RYR2 or with the mutation in CLIC2 suggest that the mutation identified herein may be causal . MYH10 encodes the non-muscle myosin heavy chain IIB that is critical for heart and brain development [44] , [45] . Loss of Myh10 function in mice results in embryonic lethality , hydrocephalus and neuronal migration defects but the cognitive and behavioural phenotype of heterozygous mice has not yet been reported . We identified a predicted-damaging de novo missense mutation ( c . 838C>T , p . Arg280Cys; individual 1871 . 656 ) in MYH10 , affecting its conserved motor domain , whereas another group recently reported a de novo truncating mutation ( c . 2722G>T , p . Glu908* ) in the same gene [46] . Both individuals displayed severe ID , microcephaly , and feeding difficulties as well as cerebral atrophy with increased intensities in bilateral basal ganglia and thalami on brain MRI ( Text S1 ) . The similarities between the phenotypes of these individuals raise the possibility that these mutations in MYH10 are pathogenic . O'Roak et al . ( 2012 ) also reported a predicted-damaging de novo missense mutation ( c . 794A>G , p . Y265C; NM_001256012 . 1 ) in the motor domain of MYH10 , in close proximity to the mutation identified herein , in a patient with ASD and moderate to severe ID . However no additional phenotypic data was available . Interestingly our patient with the MYH10 mutation also displayed autistic features . Inspection of EVS for potential LoF mutations in MYH10 showed the presence of a heterozygous frameshift deletion and a heterozygous splice site mutation . It is important to note , however , that these EVS variants were seen in single individuals and were not validated . DNMs in EIF2C1 and COL4A3BP have also been previously reported in single individuals with severe ID [2] , [4] . For each of these genes , the phenotype of the affected individuals appears similar to that of our subjects ( Text S1 ) . However , because of the lack of specific clinical features in these individuals , the occurrence of DNMs in unrelated subjects does not readily indicate pathogenicity , especially in the case of missense mutations whose functional consequences are not validated . Among the remaining cases , we also identified 6 predicted-damaging DNMs in genes ( SET [OMIM 600960] , EGR1 [OMIM 128990] , PPP1CB [OMIM 600590] , CHMP2A [OMIM 610893] , PPP2R2B [OMIM 604325] , and VPS4A [OMIM 609982] ) that play biological functions relevant to ID ( Table 2 ) . Inspection of the EVS database revealed no LoF variants in these genes , with the exception of a single heterozygous variant in PPP1C1B ( MAF = 1/12518 ) with a potential effect on splicing . In addition , some of these genes were found in proteomic studies to physically interact with the product of at least one ID-associated gene , further increasing the probability of their involvement in this disorder ( see below and Figure 2 ) . Each of these DNMs is discussed hereafter . SET encodes a widely expressed multifunctional nuclear protein that affects pathways involved in ID , such as chromatin remodelling and gene transcription [47] . SET physically binds SETBP1 [48] , whose disruption is known to cause severe ID ( see above ) . In addition , recent studies indicate that SET directly interacts with MCPH1 ( OMIM 607117 ) to ensure the proper temporal activation of chromosome condensation during mitosis [49] . Cells with SET knockdown exhibited abnormal condensed chromosomes similar to those observed in MCPH1-deficient fibroblasts . In addition , mutations that impair binding of MCPH1 to SET affect the ability of the former to rescue the abnormal chromosome condensation phenotype in fibroblasts from Mcph1 mutant mice . Recessive mutations in MCPH1 cause primary microcephaly , which is characterized by reduced brain size , without major structural abnormalities , and mild-to-moderate ID [50] . We identified a de novo deletion resulting in the creation of a premature stop codon in SET ( c . 699_701del , pTyr233* ) in an individual ( 115 . 81 ) with congenital microcephaly , normal brain MRI , and moderate ID without any other distinguishing feature ( Text S1 ) . The functional relationship between MCPH1 and SET and the phenotypical similarities between cases with mutations in MCPH1 and our subject suggest that the truncating DNM in SET may be pathogenic . EGR1 encodes a transcription factor that plays a key role in learning and memory [51] . We identified a de novo truncating mutation ( c . 1347_1348insA , p . Tyr450Ilefs*92 ) in EGR1 in an individual ( 670 . 267 ) with severe non-syndromic ID and acquired microcephaly ( Text S1 ) . Mice harbouring a heterozygous deletion of Egr1 showed synaptic plasticity , learning and memory impairments [52] , [53] . Due to the prominent role of EGR1 in learning and memory and the impact of its haploinsufficiency on cognition in mice , we postulate that the truncating DNM identified herein in EGR1 may be pathogenic . PPP1CB , which encodes a brain-enriched beta catalytic subunit of protein phosphatase 1 ( PP1 ) , and PPP2R2B , which encodes a neuron-specific B regulatory subunit of protein phosphatase 2 ( PP2A ) , have been shown to regulate synaptic plasticity pathways [54] , [55] . Individual 1439 . 518 carries a truncating mutation ( c . 909dupA , p . Tyr304Ilefs*19 ) in PPP1CB . This individual displayed severe ID , growth retardation and some dysmorphic features ( Text S1 ) . Individual 1841 . 646 carries a predicted-damaging missense mutation ( c . 413G>C , p . Arg138Pro ) in PPP2R2B . This individual showed ID , intractable seizures and autistic features ( Text S1 ) . The pathogenic impact of these mutations remains uncertain at this point . Among these candidate genes , CHMP2A and VPS4A are of special interest , as the proteins encoded by each are interacting partners . VPS4 ATPases play a critical role in the ESCRT pathway by recognizing membrane-associated ESCRT-III complexes and catalyzing their disassembly , a process that involves a direct interaction between CHMP2A and VPS4A [56] . The ESCRT-III pathway is involved in key cellular processes , including formation of endocytic multivesicular bodies , the abscission stage of cytokinesis , as well as centrosome and spindle maintenance [57] . Specific depletion of either CHMP2A or VPS4A proteins in cultured cells disrupts mitosis by inhibiting abscission and altering centrosome and spindle pole numbers [58] . We identified an individual ( 580 . 240 ) with a de novo frameshift insertion ( c . 286_287insC , p . Asn96Thrfs*35 ) in CHMP2A and another individual ( 985 . 382 ) with a predicted-damaging in-frame deletion ( c . 577_579delTCC , p . Ser193del ) in VPS4A ( Table 2 ) . Both subjects showed severe ID as they were non-ambulatory and non-verbal at 4 years of age ( Text S1 ) . Our findings , thus , raise the possibility that components of the ESCRT-III complex maybe involved in ID . To determine whether the genes identified here with predicted-damaging DNMs ( likely/possibly pathogenic or of yet unknown significance to ID ) ( Table 3 ) encode proteins that are physically interconnected , we performed protein-protein interaction network analysis using GeneMANIA ( http://www . GeneMANIA . org/ ) [59] . We also included in this analysis the known and candidate ID genes identified with predicted-damaging DNMs in other ID trio studies ( Table S2 ) [2]–[6] , [21] . This analysis showed that 11 out of the 24 proteins encoded by genes found herein with likely/possibly pathogenic DNMs interacted with either known or candidate ID genes , or with each other , further supporting their link to ID . Interestingly , we observed an enrichment for proteins implicated in glutamate receptor signaling pathways ( FDR q-value = 7 . 04e-6 ) in the generated network ( 38 interconnected proteins ) ( Figure 2 ) . Previous studies have shown an excess of functional DNMs over neutral ones in genes associated with glutamatergic systems in cases with non-syndromic ID , further supporting the critical involvement of this pathway in ID [3] . We also searched for the presence of rare inherited deleterious mutations ( truncating , splicing , predicted-damaging missense and insertions or deletions ) in genes associated with autosomal recessive or X-linked forms of ID , epilepsy or ASD ( see Table S3 for the complete list of inherited rare variants in each proband ) . We identified only one case ( 692 . 274 ) that could potentially be explained by such mutations . This individual is hemizygous for a predicted-damaging missense ( c . 7949G>A [p . Arg2650His]; NM_031407 . 6 ) in the E3 ubiquitin ligase gene HUWE1 , which is inherited from his healthy mother . Missense mutations in HUWE1 have been associated with moderate to severe X-linked ID with normocephaly or macrocephaly [60] . Our case showed severe ID ( non-verbal , non-ambulatory at 5 years of age ) with congenital microcephaly . Because of these phenotypical differences , it is thus unclear whether this variation in HUWE1 is pathogenic . In summary , our trio exome sequencing study identified deleterious DNMs in genes previously causally linked to ID in 12 cases out of the 41 studied herein , resulting in a molecular diagnostic yield of 29% . Recently , de Ligt et al . ( 2012 ) and Rauch et al . ( 2012 ) performed trio exome sequencing in individuals with severe ID and obtained a diagnostic yield , based on the presence of predicted-damaging point mutations in currently known ID genes , of 20% and 35% , respectively [2] , [4] , [21] . Overall , the contribution of inherited autosomal or X-linked recessive mutations appears limited in the three cohorts . The study of Rauch et al ( 2012 ) and ours were intentionally centered on sporadic cases , which might have created a bias against inherited mutations . However , it is important to emphasize that most cases with moderate or severe ID are sporadic , at least in Western societies . de Ligt et al . ( 2012 ) observed a proportionally smaller number of DNMs in their cohort when compared to that of Rauch et al . ( 2012 ) and ours . This difference may be related to the use of a different sequencing technology , which is associated with a lower depth , possibly accounting for the lower diagnostic yield observed in this study . Indeed , exploration of a subset of unexplained cases from this cohort using whole-genome sequencing revealed additional pathogenic DNMs in known ID genes , bringing the point mutation molecular diagnostic yield in this cohort to 34% [21] . Our study also provides evidence for the potential pathogenicity of 12 additional DNMs in as many genes . Some of these genes represent strong candidates . For instance , both HNRNPU and WAC map to small critical regions associated with ID , which were defined by a series of microdeletions . De novo truncating mutations in each of these genes were previously described in cases with severe ID . We now report additional truncating DNMs in these genes in cases with similar phenotypes as those already published , further supporting their involvement in ID . Similarly , we and others have identified damaging DNMs in RYR2 and MYH10 in patients with similar features . Finally , we discovered a truncating DNM in EGR1 , the haploinsufficiency of which affects learning and memory in mice . Although the characterization of additional cases will be needed to confirm the involvement of these candidate genes in ID , these results indicate that the contribution of DNMs to the pathogenesis of moderate or severe ID could be even greater than that suggested by the diagnostic rate observed in this study . In conclusion , our study suggests that DNMs represent a predominant cause of moderate or severe ID . High-depth trio-based exome sequencing is an effective method to establish molecular diagnosis in such cases .
The cases reported here ( 18 males , 23 females ) with moderate ( n = 12 ) or severe ( n = 29 ) ID were recruited at the Sainte-Justine Hospital ( Montreal , Canada ) , after the approval of the ethics committee , and informed consent was obtained from each participant or legal guardian . Inclusion criteria for the probands were: 1 ) absence of a history of ID , epilepsy or ASD in first or second-degree relatives; 2 ) moderate or severe ID with or without epilepsy or autistic features; 3 ) absence of pathogenic copy number variants as revealed by array comparative genome hybridization performed on a clinical basis ( using a 135k-feature whole-genome microarray ( SignatureChip OS2 . 0 manufactured for Signature Genomic Laboratories ( Spokane , WA , USA ) by Roche NimbleGen , Madison , WI , USA ) ; 4 ) absence of specific changes on brain imaging . The clinical description of the 41 affected individuals is summarized in Table S4 . For cases with likely or possibly pathogenic variants , a more detailed clinical description can be found in Text S1 . Genomic DNA ( 3 µg ) extracted from blood samples were used for exome capture and sequencing at the McGill University and Genome Quebec Innovation Center ( Montreal , Quebec , Canada ) using the Agilent SureSelect v4 exome capture kit , according to the manufacturer's recommendations , followed by 100 bp paired-end sequencing of each trio exomes on a single lane of the Illumina HiSeq2000 . Exome sequence data processing , alignment ( using a Burrows-Wheeler algorithm , BWA-mem ) , and variant calling were done according to the Broad Institute Genome Analysis Tool Kit ( GATK v4 ) best practices ( http://www . broadinstitute . org/gatk/guide/topic ? name=best-practices ) , and variant annotation was done using Annovar [61] . The median coverage of the target bases was 135× with 95% of the target bases being covered ≥10× . We focused on variants affecting the exonic regions and consensus splice site sequences ( defined herein as intronic bases up to positions −3 and +6 from the exon boundaries ) . Only variants whose positions were covered at ≥10× and supported by at least 4 variant reads constituting ≥20% of the total reads for each called position were retained . This typically yielded an average of ∼22 , 000 variants . This variant list was subsequently reduced to an average of ∼500 rare variants by filtering out those that are present in ≥0 . 5% of in-house exome data sets ( n = 600 ) from unrelated projects , as well as variants present in the 1000 Genome or in the Exome Variant Server ( EVS; http://evs . gs . washington . edu/EVS/ ) with minor allele frequencies ( MAF ) ≥0 . 5% . Putative DNMs ( typically <10/exome ) were then extracted from the rare variant list by further excluding those that were present in the exomes of the parents . The sequencing reads carrying putative DNMs were inspected visually in each trio , using the Integrative Genomics Viewer ( IGV ) [62] , to exclude obvious false positives . All putative DNMs were validated by bidirectional Sanger sequencing in the corresponding trio . | Intellectual disability ( ID ) is the most frequent severe handicap of childhood . Several observations indicate that genetic factors explain a large fraction of cases with ID . We and others have recently found that de novo mutations ( DNMs; genetic changes not transmitted from the parents ) represent a common cause of ID . To further assess the contribution of DNMs to the development of ID , we interrogated virtually all the genes of the genome in 41 affected children with moderate or severe ID and in their healthy parents . In 12 of the cases , we identified disease-causing DNMs in genes known to be associated with ID , resulting in a molecular diagnostic yield of 29% . We also found 12 possibly disease-causing DNMs in genes that were not previously causally linked to ID . Interestingly , many of the genes with deleterious DNMs uncovered by this study encode proteins that interact with each other and affect specific processes in brain cells . In contrast , we did not identify any inherited mutations that could explain our cases . We conclude that DNMs play a predominant role in moderate or severe ID . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"biology",
"and",
"life",
"sciences",
"medicine",
"and",
"health",
"sciences"
]
| 2014 | De Novo Mutations in Moderate or Severe Intellectual Disability |
The virulence regulator ToxR initiates and coordinates gene expression needed by Vibrio cholerae to colonize the small intestine and cause disease . Despite its prominence in V . cholerae virulence , our understanding of the direct ToxR regulon is limited to four genes: toxT , ompT , ompU and ctxA . Here , we determine ToxR’s genome-wide DNA-binding profile and demonstrate that ToxR is a global regulator of both progenitor genome-encoded genes and horizontally acquired islands that encode V . cholerae’s major virulence factors and define pandemic lineages . We show that ToxR shares more than a third of its regulon with the histone-like nucleoid structuring protein H-NS , and antagonizes H-NS binding at shared binding locations . Importantly , we demonstrate that this regulatory interaction is the critical function of ToxR in V . cholerae colonization and biofilm formation . In the absence of H-NS , ToxR is no longer required for V . cholerae to colonize the infant mouse intestine or for robust biofilm formation . We further illustrate a dramatic difference in regulatory scope between ToxR and other prominent virulence regulators , despite similar predicted requirements for DNA binding . Our results suggest that factors in addition to primary DNA structure influence the ability of ToxR to recognize its target promoters .
Bacteria emerge as pathogens by horizontally acquiring new genetic functions from their environment and neighboring organisms [1 , 2] . Vibrio cholerae , the etiological agent of cholera , is a paradigm of this process . Benign environmental V . cholerae isolates emerge as pandemic pathogens through the horizontal acquisition and incorporation of genetic elements encoding virulence factors into their progenitor genomes [3–5] . The factors gained by the benign progenitor genome include cholera toxin , encoded on the CTX prophage , and the colonization pilus TCP , along with regulators TcpP and ToxT , encoded on the Vibrio Pathogenicity Island 1 ( VPI-1 ) [6–9] . Moreover , current 7th pandemic V . cholerae strains are genetically distinguished from the previous 6th pandemic strains by the acquisition of two new horizontally acquired elements , Vibrio Seventh Pandemic islands 1 and 2 ( VSP-1 , 2 ) [5 , 10] . The acquisition of VSP-1 and 2 are thought to have promoted the emergence and dominance of 7th pandemic strains . The progenitor genome-encoded transcription factor ToxR plays a critical role in V . cholerae virulence and stress response . ToxR is a membrane-bound transcriptional regulator with a partner protein , ToxS , that enhances ToxR activity [4 , 11 , 12] . The major role of ToxR in pathogenesis is to act with TcpP and induce expression of toxT . ToxT then triggers expression of genes encoding colonization factors and cholera toxin , resulting in disease [13–17] . When overexpressed , or in the presence of bile , ToxR can also directly activate the genes encoding cholera toxin , ctxAB [18 , 19] . On the progenitor genome , ToxR directly regulates expression of V . cholerae’s major outer membrane proteins: OmpU and OmpT [20 , 21] . Expression of OmpU and OmpT is important for V . cholerae to survive host-relevant stresses including bile , antimicrobial peptides , and pH changes [22–25] . ToxR’s ability to regulate both progenitor-encoded and recently acquired DNA allows for new and existing gene functions to be coordinated , which has supported V . cholerae’s emergence as a successful pathogen . ToxR expression and activity are responsive to stimuli , including pH , oxygen , temperature , and metabolites [24 , 26–28] . Other transcription factors likely compete with ToxR for binding sites to control gene expression under different conditions [29–31] . The complexity of ToxR regulation may be necessitated by the many processes ToxR impacts [32 , 33] . Despite its critical role in virulence , ToxR has only been shown to directly regulate four target genes [15 , 20 , 34 , 35] . Here , we integrate chromatin-immunoprecipitation sequencing ( ChIP-seq ) data with gene expression data and phenotype studies to map the regulon directly controlled by ToxR . We identify ToxR regulation in several new roles affecting V . cholerae virulence and biofilm formation , which correlate with the emergence of 7th pandemic strains . Analysis of our ChIP data was unable to identify a motif that could explain how ToxR identified its target binding location in vivo . However , it did describe an affinity of ToxR for low GC-content locations that were frequently shared with the histone-like nucleoid structuring protein H-NS ( VC1130;VicH ) . Our results show ToxR antagonizes H-NS transcriptional regulation , and that this interplay controls V . cholerae host colonization and impacts biofilm formation . A comparison between ToxR and additional prominent virulence regulators TcpP and ToxT shows a unique global role for ToxR gene regulation .
ToxR is a major virulence regulator in V . cholerae , yet we only know of four genes that it can directly regulate: toxT , ompU , ompT and ctxA [13 , 15 , 19–21 , 36] . Microarray experiments performed under conditions that induce virulence factor expression have implicated ToxR in the regulation of more than 100 genes [33] , suggesting a much larger regulon . However , it is unclear how much of this regulation is direct . To determine the direct regulon of ToxR , we used chromatin-immunoprecipitation-sequencing ( ChIP-seq ) to identify ToxR binding sites across the genome . We ectopically expressed ToxR with a C-terminal V5 tag under control of an arabinose inducible promoter in 7th pandemic V . cholerae strain C6706 . This approach allows reproducible induction and immunoprecipitation of ToxR without prior knowledge of all the environmental factors that may control its expression . Expression levels of ToxR are shown S1A Fig . This method has proven effective for ChIP-seq in V . cholerae and other bacteria [37–40] . To confirm the DNA binding activity of the tagged ToxR , we induced its expression and performed ChIP as previously described [37 , 38] . Quantitative PCR ( qPCR ) analysis of ToxR ChIP DNA samples demonstrates that V5-tagged ToxR strongly binds known target sites in the toxT , ompT and ompU promoters , but not to a negative control site at the icd promoter ( S2 Fig ) . We performed ChIP-seq and identified genome-wide ToxR binding locations as previously described [37 , 38] . Alignment of sequencing reads from each sample gave average genome coverage of 41-fold . This depth of coverage allowed us to use a stringent false-discovery rate ( FDR ) cutoff of 0 . 001% to identify ToxR ChIP-enriched genomic regions , which are referred to as peaks . ChIP peaks are identified when the sequence coverage of a given genomic region in the experimental sample exceeds the non-immunoprecipitated input control sample at a rate specified by the FDR . ChIP peak enrichment ranged from 5—to 19-fold over the input . qPCR analysis of ChIP DNA generally showed a much higher fold enrichment ( S2 Fig ) . This is likely because computational ChIP-seq enrichment is a measurement of the average enrichment across the whole peak , while our qPCR analysis generally measures enrichment at specific locations within the peak . We compared the ToxR ChIP peak lists generated from two biological replicates and set a limit that a peak must be identified in both replicates to be included as a potential ToxR binding location for our analysis . Peaks meeting this standard were then manually curated for accuracy [41] . We associated a ToxR peak with a gene based on its proximity to promoters and translation start sites . With these criteria , a ToxR peak can associate with more than one gene if 1 ) the translational start sites of two or more genes are close together , or 2 ) if ToxR binds multiple sites that are too close together to be accurately separated by peak-calling algorithms [42] . In these cases we used published gene expression data and data generated in this study to interpret which gene ( s ) ToxR is likely to directly regulate . For example , there is a ToxR peak overlapping the 172 bases between divergently transcribed genes VC0844 and VC0845 . Previous studies have described ToxR affecting regulation of both genes [33 , 43] . Our analysis identified 35 ToxR peaks associated with 39 genes by our criteria ( Table 1 ) . Three ToxR peaks remained associated with more than one gene . The coordinates encompassing the raw ToxR ChIP-seq peak locations and their associated genes are given in S1 Table . Schematics of ToxR ChIP enrichment at select loci are shown in S3 Fig . One peak was identified covering each of the promoters for toxT , ompU , and ompT , validating our procedure for identifying ToxR binding locations . Table 1 shows several genes in horizontally acquired elements and genes that have previously been connected with ToxR regulation through microarray and additional studies . Analysis of the locations and functions of genes associated with ToxR peaks identified two overrepresented groups: 18% of the genes identified in this study are known or predicted to function in biofilm formation , and 40% are located on horizontally acquired elements . We identified ToxR peaks in the promoter regions of six genes and one small RNA ( sRNA ) all known or suspected to play a role in biofilm formation: ryhB , vpsL ( VC0934 ) , VC1145 , VC1330 , VC1599 , leuO ( VC2485 ) , and VC2697 [44–49] . These genes are all encoded on the progenitor genome [50] . ToxR was previously shown to induce leuO expression [51] . Our ChIP-seq analysis identified ToxR binding covering the leuO promoter region ( Table 1 ) , which shows that the observed positive regulation is likely direct . To further understand how ToxR regulates expression of genes involved in biofilm formation , we determined the impact of ToxR on the expression of ryhB , vpsL , and VC1599 . These genes were chosen because they have not been previously associated with ToxR regulation and encode diverse biological functions . RyhB is a small regulatory RNA involved in regulation of iron metabolism [46 , 52] . VC1599 is a diguanylate cyclase that produces the signaling molecule cyclic-di-GMP ( cdiGMP ) [45 , 53] . vpsL encodes a glycosyltransferase for Vibrio polysaccharide production and is the first gene of the Vibrio polysaccharide vps-II operon [44 , 54 , 55] . qPCR analysis of ToxR ChIP DNA confirmed our sequencing data and showed ToxR enrichment of ryhB , vpsL , and VC1599 promoter regions , but not of a negative control site ( Fig 1A ) . We used northern blots and quantitative reverse-transcription PCR ( qRT-PCR ) to determine ToxR regulation of ryhB , vpsL and VC1599 . Northern blot analysis showed that deletion of toxRS led to an increase in ryhB abundance , consistent with direct ToxR repression of ryhB expression ( Fig 1B ) . Deletion of toxRS alone did not affect vpsL or VC1599 expression ( S4 Fig ) . The free-living planktonic cells used for our gene expression assays might not recapitulate the environmental signals needed for ToxR regulation of vpsL and VC1599 utilized for biofilm formation [56] . In an attempt to bypass this potential signaling hurdle , we compared expression of vpsL and VC1599 in a toxRS deletion strain carrying either an empty vector or a vector with an arabinose inducible toxRS operon to specifically increase ToxRS levels . In this comparison , induction of toxRS led to an increase in vpsL expression and a decrease in VC1599 expression , supporting direct ToxR regulation of these genes ( Fig 1C ) . Our results establish both positive and negative control of biofilm-associated genes by ToxR . It also ties ToxR regulation to small regulatory RNAs and cdiGMP , both of which influence a wide spectrum of genes and biological processes [46 , 57] that may be responsible for indirect effects associated with ToxR regulation . We assessed the ability of wild type , toxRS , ryhB , VC1599 , and vpsL mutant strains to form biofilm in a static microtiter assay in rich broth at 30°C ( Fig 1D and 1E ) . The toxRS deletion strain showed reduced biofilm formation , supporting its regulatory role in this process . This phenotype was complemented by ectopic expression of toxRS ( S5 Fig ) . The requirement of ryhB , vpsL , and genes downstream of vpsL in the vps-II operon for biofilm formation was previously established [44 , 46 , 55 , 58] . Supporting those results , a vpsL in-frame deletion mutant and a ΔryhB::kanR mutant both showed a defect in biofilm formation ( Fig 1D and 1E ) . These phenotypes were complemented by ectopic expression of the respective gene ( S5 Fig ) . Overexpression of VC1599 had been shown to increase biofilm formation [45] . Supporting this observation our VC1599 deletion strain showed decreased biofilm production ( Fig 1D and 1E ) . This phenotype was complemented by ectopic expression of VC1599 from a plasmid , which led to biofilm overproduction ( S5 Fig ) . Loss of toxRS or vpsL decreased biofilm formation under the experimental conditions used for our assay . Positive regulation of vpsL by ToxR could explain the biofilm defect of our ΔtoxRS mutant . We tested the ability of a ΔtoxRSΔvpsL double mutant to form biofilm , as well as ΔtoxRSΔryhB::kanR and ΔtoxRSΔVC1599 double mutants . We did not observe a significant difference in biofilm formation for any double mutant relative to the ΔtoxRS mutant ( S6 Fig ) . The resolution of our assay may not be sufficient to identify synergies or additive effects of these mutants . Our ChIP-seq results showed that ToxR binds locations on all four of V . cholerae’s major acquired pathogenicity islands: VPI-1 , VPI-2 , VSP-1 , and VSP-2 ( Table 1 ) . In addition to the toxT promoter , our analysis shows ToxR binds the promoter regions of VPI-1 genes VC0824 ( tagD ) , VC0825 ( tcpI ) , VC0844 ( acfA ) , and VC0845 ( acfD ) ( Table 1 ) . qPCR analysis of ToxR ChIP DNA validated our sequencing results that ToxR binds the promoter regions of VC0824 ( tagD ) , VC0825 ( tcpI ) , and the promoter region shared by VC0844 ( acfA ) and VC0845 ( acfD ) ( Fig 2A ) . Combined with gene expression studies describing positive regulation of tagD , acfA , and acfD genes by ToxR , independent of ToxT [33 , 43 , 59] , our results support a direct role for ToxR in the positive regulation of these genes , expanding ToxR’s known targets on VPI-1 . While the function of these genes is under investigation , tcpI , acfA , and acfD are known to be required for V . cholerae colonization of a model host [16 , 60] . When overexpressed or activated by specific compounds , ToxR can activate ctxA expression [18–20] . However , the physiological relevance of this interaction is unclear . Our ChIP-seq analysis did not identify a ToxR binding site in the ctxA promoter , suggesting the event was either below our level of detection or does not occur to an appreciable extent in V . cholerae under our experimental conditions . Seventh pandemic V . cholerae is genetically distinguished from previous 6th pandemic strains by the presence of acquired islands VSP-1 and 2 . Little is known about the origin , content , and regulation of these islands , though VSP-1 carries at least one gene that influences the ability of V . cholerae to colonize the infant mouse model [38 , 61] . Our results show ToxR binding across the promoter regions of genes located on both VSP-1 and VSP-2 ( Table 1 ) . qPCR analysis of ToxR ChIP DNA validated that ToxR binds the promoter regions of VC0176 , VC0178 , VC0182 , and VC0183 on VSP-1 , and VC0490 and VC0493 on VSP-2 ( Fig 2B and 2C ) . Microarray analysis suggested that ToxR can repress of VC0176 , VC0490 , and VC0493 expression under virulence-gene inducing conditions [33] , supporting a direct role for ToxR in their regulation . To corroborate and expand ToxR regulation of VSP-1 and 2 , we used qRT-PCR to determine if ToxR regulated expression of selected VSP-1 and 2 genes . Deletion of toxRS alone did not affect expression of VSP-1 or 2 genes when V . cholerae was grown exponentially in rich broth ( S4 Fig ) . We again considered that conditions for ToxR regulation of VSP-1 and 2 genes were not recapitulated by exponentially growing cells in rich broth . We compared expression of VSP-1 and 2 genes in a toxRS deletion strain carrying an empty vector or a vector with an arabinose inducible toxRS operon . In this comparison , induction of toxRS led to repression of VC0176 , VC0178 , and VC0493 , supporting a direct role for ToxR regulation of VSP-1 and VSP-2 genes [33] ( Fig 3A ) . Considering the central role of ToxR in virulence regulation , we questioned whether the ToxR-regulated genes on VSP-1 also affected V . cholerae colonization . VC0178 was previously shown not to influence V . cholerae host colonization; however , VC0176 was not tested [38] . We constructed an unmarked VC0176 deletion mutant and tested its ability to colonize the infant mouse . We found that the ΔVC0176 mutant showed approximately a 5-fold defect in colonizing the infant mouse intestine in competition with the parental strain ( Fig 3B ) . No defect was observed when the strains were competed in liquid culture ( Fig 3B ) . The phenotype was complemented by ectopic expression of VC0176 ( S7 Fig ) . These results expand the regulatory role of ToxR on virulence islands VPI-1 and 2 , which are found in all pandemic V . cholerae strains . They further show that ToxR has gained control over expression of recently acquired genetic elements that define the current 7th pandemic strains , including a new VSP-1 colonization factor VC0176 . These results implicate ToxR as a regulatory hub for integrating expression of progenitor genome-encoded functions with newly acquired genes to promote V . cholerae fitness . Our results demonstrate that ToxR binds all four Vibrio pathogenicity islands , and implicates ToxR as a global regulator of horizontally acquired genetic elements . Horizontally acquired DNA generally has a lower GC-content than the progenitor genome [62] . For example , the average GC-content of the N16961 V . cholerae genome is 47% , while the average GC-content of VPI-2 and VSP-1 is 42% and 40% respectively [63 , 64] . Analysis of the DNA sequences comprising the ToxR ChIP-seq peak locations showed they contain an average GC-content of just 40% . This suggests that ToxR preferentially binds DNA with base composition more similar to acquired elements than to that of the progenitor genome average . This result agrees with the low GC-content of the predicted ToxR consensus binding motif ( TNAAA-N5-TNAAA ) , which was based on ToxR binding and/or activation of toxT , ompT , ompU , and ctxA promoters [15 , 20] . The preference for binding low GC-content DNA is shared with the histone-like nucleoid structuring protein ( H-NS ) that binds and silences horizontally acquired DNA [65] . V . cholerae H-NS binds and silences genes identified in our ToxR regulon study , including toxT and vpsL [31 , 66–68] . These observations prompted us to question if ToxR and H-NS may share additional genomic binding locations . We added a V5-tag to the C-terminus of the chromosomally encoded H-NS in V . cholerae C6706 to facilitate immunoprecipitation . We performed ChIP-seq for H-NS-V5 and determined its genome-wide binding profile under the same conditions as we used for ToxR ChIP-seq ( S2 Table ) . We compared the genome binding profiles and found that 39% of regions bound by ToxR were also identified in our H-NS ChIP-seq analysis ( Table 1 ) . Previous studies have shown genetic interactions between toxR , tcpP , and hns influence expression of the toxT promoter [31] , and that H-NS can directly regulate vpsL [54 , 66] . Our results suggest that ToxR might antagonize H-NS regulation at multiple locations to gain access to gene targets . Rather than a defined consensus motif , topology has been implicated as a critical factor controlling H-NS binding to DNA . Low GC-content DNA forms structures that are preferentially bound by H-NS [69–71] . Since DNA topology and H-NS binding changes with environmental conditions [65 , 69 , 72 , 73] we wanted to test if ToxR could antagonize H-NS binding in vivo , in the context of the bacterial cell . To do this , we introduced an empty or arabinose-inducible , toxRS-encoding plasmid into our V . cholerae strain containing V5-tagged H-NS . We induced toxRS expression with arabinose and performed ChIP against H-NS-V5 . We next used qPCR to determine H-NS enrichment at shared ToxR binding locations . We chose to examine the vpsL promoter on the progenitor genome , and toxT and VC0844-5 promoter regions on VPI-1 . At each location we found that H-NS occupancy decreased following induction of toxRS , indicating that ToxR can antagonize H-NS binding at these locations ( Fig 4A ) . These experiments were performed in the presence of the chromosomally-encoded toxRS . Thus , the impact of ToxR on H-NS binding may be even greater than observed here . As H-NS is a global silencer of horizontally acquired genetic material , our results indicate that ToxR has the ability to antagonize H-NS binding and bring the regulation of new genetic material under virulence gene control . ToxR is essential for V . cholerae virulence through its regulation of many genes important for host colonization and pathogenesis [13–17] . Supporting this role , our ΔtoxRS deletion strain was strongly outcompeted by the wild type strain in infant mouse intestinal colonization assays ( Fig 4B ) , which agreed with previous reports [74] . This defect was complemented by ectopic expression of toxRS ( S8 Fig ) . H-NS represses many virulence genes , and deletion of hns results in their induction [31] , suggesting deletion of hns should not impair V . cholerae intestinal colonization . Supporting our hypothesis the Δhns mutant did not show a significant defect in colonizing the infant mouse intestine in competition with the wild type strain ( Fig 4B ) . Our data showed that ToxR and H-NS both bind the promoter regions of many of the same genes that are important for V . cholerae virulence ( Table 1 ) . It also showed that ToxR could antagonize H-NS binding at shared binding locations ( Fig 4A ) . If ToxR antagonizes H-NS repression of important colonization factors , then deletion of H-NS should alleviate the need for ToxR regulation in intestinal colonization . To genetically test our hypothesis , we constructed a double ΔtoxRSΔhns mutant and assayed its ability to colonize infant mice ( Fig 4B ) . Agreeing with our hypothesis for the importance of ToxR’s genetic interaction with H-NS for colonization , the double ΔtoxRSΔhns mutant showed no competitive defect compared to wild type or the Δhns mutant alone . Ectopic expression of hns in the ΔtoxRSΔhns mutant produces a competition defect similar to that of ΔtoxRS mutant alone ( S8 Fig ) . Removing H-NS activity genetically eliminates the need for ToxR regulation in V . cholerae host colonization . We observed a similar genetic effect for biofilm formation , where deletion of hns compensated for the biofilm defect of the toxRS mutant ( Fig 4C ) . This effect was complemented by ectopic expression of hns , though not to wild type levels ( S9 Fig ) . This may be because expression of hns from a plasmid does not recapitulate H-NS levels necessary for normal biofilm regulation in our strain . Our results indicate that for both host colonization and biofilm formation , the major purpose of the ToxR regulation is to antagonize H-NS activity . ToxR co-operates with transcription factor TcpP to activate toxT gene expression [13–17] . Like ToxR , TcpP is a membrane-bound transcription factor with an enhancer partner protein , TcpH , and is responsive to environmental conditions and upstream regulation [7 , 26 , 31 , 75 , 76] . TcpP is only known to regulate toxT . The region of the toxT promoter that affects TcpP binding also shows low GC-content and low sequence complexity ( TGTAA-N6-TGTAA ) [77] . Given the similarity of TcpP’s and ToxR’s binding motifs , we hypothesized that TcpP may also directly regulate more genes , alone or in association with ToxR . Previous microarray studies found that deletion of tcpP changed the expression of 58 genes under conditions that activate colonization factor expression [33] , supporting a possible broader role for TcpP regulation . To define the regulon directly controlled by TcpP , we performed ChIP-seq in a similar manner as for ToxR . tcpP expression levels are shown in S1B Fig . qPCR analysis of TcpP ChIP DNA showed that the V5-tagged TcpP bound the toxT promoter , but not to a negative control locus ( Fig 5A ) . In stark contrast to ToxR ( and despite its relatively weak predicted binding motif constraints ) , our ChIP-seq analysis identified only three TcpP peaks in the entire V . cholerae genome ( Table 2 ) . We identified a strong TcpP peak upstream of toxT , agreeing with our initial validation of our TcpP construct ( Fig 5A ) . A schematic of ChIP-seq DNA enrichment at this site is shown in S3 Fig . In addition , we identified TcpP peaks upstream of VC1854 ( ompT ) and hypothetical gene VCA0536 . qPCR of TcpP ChIP DNA validated our sequencing data , showing TcpP binding of ompT and VCA0536 promoter regions , but not a negative control locus ( Fig 5A ) . Enrichment of TcpP at ompT and VCA0536 promoter regions was similar to enrichment at the toxT promoter . Microarray analysis previously suggested TcpP can repress ompT expression [33] . Supporting this observation , we found that ectopic expression of tcpPH in a ΔtcpPH mutant repressed ompT expression compared with the empty plasmid control ( Fig 5B ) . Along with toxT , ompT is now the second gene recognized as co-regulated by ToxR and TcpP . Moreover , TcpP repression of ompT shows that like ToxR , TcpP can act as either a transcriptional activator or repressor . VCA0536 has not previously been associated with TcpP regulation . VCA0536 encodes a putative cyclic di-GMP phospodiesterase that was found to be expressed in vivo by IVIAT [78] , and is affected by the biofilm regulator VpsT [57] . Induction of tcpPH activated VCA0536 expression compared to the empty plasmid control ( Fig 5B ) , supporting direct positive regulation by TcpP . Our results show that TcpP does regulate genes in addition to toxT , but does not share global regulation with ToxR despite similar predicted binding requirements . We computationally scanned seven V . cholerae genomes , including both El Tor and Classical strains , for previously determined ToxR ( TNAAA-N5-TNAAA ) and TcpP ( TGTAA-N6-TGTAA ) binding motifs [15 , 20 , 77] using FIMO motif search software [79] . We used a cut-off p-value of < 0 . 0001 to identify significant sequence matches . For each motif , we identified many more matching sites in the genomes than were identified in their respective ChIP-seq analysis ( S3 Table ) . This suggests that while primary DNA structure is undoubtedly important for ToxR and TcpP binding , the motif sequences alone are not sufficient to explain the selectivity of ToxR and TcpP binding in vivo These motifs were constructed based on a small set of binding locations; four for ToxR and only 1 for TcpP . To attempt to improve the specificity of these motifs , we analyzed our ChIP-seq data sets for ToxR and TcpP binding site motif sequences using GLAM2 motif predication software [80 , 81] . We screened motifs generated through our analysis by determining if they overlapped with experimentally proven binding sites for TcpP in the toxT promoter , and for ToxR in the toxT , ompU , and ompT promoters . For ToxR and TcpP , we analyzed their respective ChIP-seq data sets as a whole and as peaks found on genomic islands compared to peaks found on the progenitor genome . The V . cholerae N16961 genome has an average GC-content of 47% [50] . ToxR ChIP peak sequences found in genomic islands and on the progenitor genome had lower average GC-contents of 38% and 42% respectively . Using all ToxR ChIP peak sequences , we were able to generate a motif that overlapped the previously published sequence important for ToxR binding and regulation of the toxT , ompU , and ompT promoters ( Fig 6 ) . This motif resembles the previously published motif and , like it , showed low sequence complexity and low GC-content . We computationally scanned seven V . cholerae genomes for this new motif using FIMO and again found it present more times throughout the genome than were identified by our ToxR ChIP-seq analysis ( S3 Table ) . Use of this new motif alone also appears insufficient to predict locations bound by ToxR in vivo . We were unable to identify a TcpP binding motif from our ChIP peak dataset that also overlapped TcpP’s known binding site in the toxT promoter .
Our results indicate that ToxR directly controls a much larger gene set than previously recognized . This expands our understanding of virulence control and biofilm formation , and implicates ToxR as a broad regulator of acquired genetic information ( Fig 7 ) . ToxR expression level and activity are regulated by many environmental signals [26–28 , 82 , 83] . ToxR also competes and interacts with other proteins to control transcription of target genes [29–31] . These factors likely allow V . cholerae to differentially control subsets of the ToxR regulon depending on the environmental conditions . The exact protein levels and activity of ToxR during each stage of infection or in biofilm development are unclear . In an attempt to overcome unknown environmental signals and broadly identify genomic sites for ToxR binding , we chose to use ectopic ToxR expression . This approach allows reproducible induction and immunoprecipitation of ToxR without prior knowledge of all the factors that may control its expression , and has proven effective for elucidating transcription factor regulons in V . cholerae and other bacteria [37–39] . A concern of this approach is that ectopic expression of ToxR or TcpP may cause aberrant binding or transcriptional regulation . While this remains a possibility , theoretical [84] and experimental studies [37 , 38 , 40] indicate that transcription factor overexpression does not lead to significant off target binding in vivo . Supporting our approach , the 35 ChIP loci we identified for ToxR is relatively small compared to many other prokaryotic ChIP-seq studies , which identified anywhere from several dozen to several hundred binding sites for other transcription factors [40 , 85–87] . Also , the ctxA promoter has been shown to bind ToxR in vitro , but the in vivo relevance of this is uncertain [18–20] . We did not identify this interaction with ChIP-seq , supporting that the expression level of ToxR used in our study did not promote ToxR binding to all available sites in vivo . Our results indicate that ToxR regulation extends to all four V . cholerae pathogenicity islands , including VSP-1 and VSP-2 , which genetically define seventh pandemic strains . The ability of ToxR to regulate new VSP-1 and VSP-2 functions along with existing cellular processes may have helped promote the emergence of 7th pandemic strains . We identified a potential role for ToxR-regulated VSP-1 gene VC0176 in host colonization . VC0176 expression was found to be upregulated during intestinal colonization of the infant mouse model [88] . However , ToxR represses VC0176 expression and deletion of VC0176 results in a colonization defect . This suggests that ToxR may act on VC0176 to limit V . cholerae colonization at some point during the infection cycle , possibly in preparation for exiting the host . This is similar to the recent observation that ToxR can downregulate virulence gene expression through its regulation of leuO [51] . The ability of ToxR to gain direct control over VSP-1 and integrate it with existing virulence networks may have potentiated exploitation of VSP-1 gene functions and promoted the emergence of 7th pandemic strains . Our analysis identified additional ToxR regulated genes encoded on the progenitor genome , including those that function in biofilm formation . The positive regulation of vpsL by ToxR most adequately explains the defect in biofilm formation of the ΔtoxRS mutant under our conditions . vpsL is the first gene in the vps-II operon [44 , 55 , 58] . Thus , ToxR activity likely influences additional genes downstream of vpsL that are also important for biofilm formation . This model would also help explain how the deletion of hns elevates the ΔtoxRS biofilm defect . The regulatory relationship between toxR , ryhB , and VC1599 is less straightforward , but may be relevant for biofilm formation under different environmental conditions . ToxR regulation of ryhB and VC1599 could also be important for other aspects of V . cholerae biology , such as iron regulation , in which ryhB figures prominently . Deletion of toxR was recently shown to enhance biofilm formation of V . cholerae strain A1552 through an unknown mechanism in a standing culture in a silica tube [89] . The differences between those results and ours may be due to differences in assay conditions or , more likely , strain differences . Our studies used strain C6706 , while Valeru et al . used strain A1552 . Phenotypic differences between these strains have previously been observed with competency and Vibrio polysaccharide regulation , and may be attributed to strain variation in cAMP-CRP or quorum-sensing regulation [90 , 91] . It is worth noting that biofilms can enhance gene transfer [92–97] and ToxR is involved in both biofilm formation and broad regulation of acquired genes . It will be interesting to test if ToxR also enhances gene transfer or stability of acquired elements . Our results provide genetic evidence that the master regulator ToxR antagonizes H-NS activity at sites across the genome to affect important phenotypes . This result is consistent with previous studies describing interactions between H-NS , ToxT , and ToxR in regulating expression of toxT , tcpA and ctx [31 , 67] . Importantly , we demonstrate that deleting hns eliminates the requirement of ToxR for host colonization in modern 7th pandemic V . cholerae . This result suggests that the major role of ToxR in virulence is to antagonize H-NS repression of colonization factors . The mechanism of ToxR antagonism is unclear . Rather than one mechanism , the way in which ToxR and H-NS interact may vary with genomic location . Moreover , since H-NS gene silencing is regulated by environmental factors [69 , 72] , the interaction between H-NS and ToxR may change as V . cholerae cycles between host and environmental reservoirs . Our ChIP-seq analysis shows that ToxR and H-NS share certain binding locations across genome ( such as the toxT promoter ) , and induction of toxRS results in decreased DNA binding of H-NS . ToxR may directly compete with and displace H-NS at shared binding sites , as has been suggested for other H-NS/transcription factor interactions [67 , 98] . Rather than sequence alone , H-NS has an affinity for DNA structure , favoring the binding of curved DNA [99–101] , and is known to form nucleoprotein filaments that promote DNA silencing [102] . ToxR may bind and alter DNA topology near H-NS , which could destabilize its interactions with DNA . Alternatively , ToxR may directly interact with H-NS and destabilize its DNA association , as has been shown for phage protein Arn [103] . Understanding how ToxR recognizes its target DNA sequences will be important in deciphering its antagonism of H-NS . Our analysis of ChIP-seq peaks identified an expanded ToxR consensus DNA motif that may facilitate its DNA binding . However , the large number of locations of this motif in the genome compared to the number of ToxR binding sites we identified suggests that our motif is still inadequate to predict ToxR binding specificity in vivo alone . It is possible that a primary structure of A , T , G , and C that does dictate ToxR binding was left undiscovered by our analysis . Differences between predicted and actual in vivo binding sites were also observed for ToxT , which also has a low GC-content and low-complexity consensus motif . Computationally , the ToxT consensus motif ( toxbox ) maps to a large number of locations across the V . cholerae genome [104] . However , in vitro biochemical interactions between purified ToxT and fragmented V . cholerae genome identified just 199 ToxT binding sites [105] . Subsequently , in vivo ChIP-seq identified and validated only seven of these ToxT binding sites , which is in line with transcriptome studies of ToxT regulated genes [33 , 38] . In eukaryotic gene regulation , factors in addition to linear DNA sequence , including topology , partner proteins , and DNA localization , all contribute to in vivo selectivity of transcription factor DNA binding [106 , 107] . Analysis of 119 transcription factors from the ENCODE project database has shown up to 99% of motif locations in a genome are not bound by their respective transcription factor [108] . Like H-NS , ToxR has a propensity to bind low GC-content DNA . Thus , ToxR binding may also use DNA topology in addition to sequence . ToxR is also unique in that it is a membrane-bound transcription factor . ToxR’s localization may limit its access to genome locations in a packed nucleoid . Super-resolution microscopy has suggested that H-NS sequesters bound DNA into two compact clusters per chromosome in E . coli [109] . Similar nucleoid structuring in V . cholerae could also act to limit ToxR access to all genomic locations . Future localization and chromosome conformation capture studies may yield important information on factors in addition to primary DNA structure that dictate how ToxR reaches its target sequences . Continued research to understand how ToxR finds its regulatory targets may provide insight into the evolutionary trajectory of V . cholerae and its potential for future acquisition of foreign genes .
The animal experiments were performed with protocols approved by the University of Texas at Austin , Institutional Animal Care and Use Committee . Protocol number AUP-2013-00052 . The University of Texas at Austin animal management program is accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care , International ( AAALAC ) , and meets National Institutes of Health standards as set forth in the Guide for the Care and Use of Laboratory Animals ( DHHS Publication No . ( NIH ) 85–23 Revised 1996 ) . Strains and plasmids are listed in S4 Table . Strains were grown in Luria Broth ( LB; rich medium ) . The following antibiotic concentrations were used: carbenicillin 75 μg/mL , kanamycin 25 μg/mL , streptomycin 100 μg/mL and chloramphenicol 2 . 5 μg/mL for V . cholerae and 10 μg/mL for E . coli . Arabinose was used at 0 . 2% for induction . X-gal was used at 40 μg/mL . All cloning products were sequence-verified , and the nucleotide sequences of all primers used for cloning are listed in S5 Table . For in-frame gene deletions of toxRS , tcpPH , VC1599 , VCA0536 , vpsL and H-NS , genomic DNA surrounding the respective gene was amplified by crossover PCR and cloned into pWM91 or pSSK10 for subsequent sacB mediated allelic exchange as described [110 , 111] . For complementation constructs , the respective gene was amplified from chromosomal DNA and cloned into plasmid pBAD18 or pWKS30 [112 , 113] . For genes cloned into pWKS30 , the respective native promoter was also included . Full length ToxR and TcpP were cloned into pBAD18 with C-terminal 3XV5 tags as previously described [37–39] . Genes cloned into pBAD18 were induced by adding arabinose to the growth medium . Biofilm assays were performed essentially as described [114] . V . cholerae C6706 wild-type and mutants strains where grown overnight on LB agar plates . Each strain was back-diluted in a 5 mL culture of LB and grown to mid-log phase . The culture was then diluted 1:100 in fresh LB , and 100μL of the diluted culture was added to a round-bottom PVC microtiter plate in replicates of three . Strains were allowed to grow for 22 hours at 30°C . Planktonic cells were removed , and bound cells were washed twice with 200 μL sterile water and then stained with 0 . 1% crystal violet for 15 mins . Stain was removed and cells were washed three times with 200 μL PBS and allowed to air dry for 15–30 mins . Stain was then solubilized with 200 μL 95% ethanol for 15 mins . Finally , 125 μL of solubilized stain was transferred to a new 96-well , flat bottom polystyrene plate . The optical density was measured at 595 nm using a SpectraMax Plus384 absorbance microplate reader with SOFTmax Pro v6 . 2 . 2 software . ChIP was performed as previously described [37 , 38] . 50 mL of exponentially growing culture in LB was induced with 0 . 1% arabinose for 30 min at 37°C . No induction was required for H-NS ChIP . Formaldehyde was added to 1% final concentration and incubated at 25°C for 20 min with occasional swirling . Crosslinking was quenched by adding glycine to 0 . 5 M . Cell pellets were washed in 1X TBS and resuspended in lysis buffer ( 10 mM Tris pH 8 . 0 , 100 mM NaCl , 1 mM EDTA , 0 . 5 mM EGTA , 0 . 1% DOC , 0 . 5% N-lauroylsarcosine ) + protease inhibitor cocktail ( Sigma ) and 1 mg/mL lysozyme and incubated at 37°C for 30 min . The cells were sonicated 1X 30sec with a needle sonicator , and unlysed debris was pelleted by centrifugation . The lysate was sonicated for 20 min with a 10 s on/ 10 s off cycle ( QSonica; www . sonicator . com ) . Sheared samples had an average DNA fragment size of ~300bp with a spread of 50-800bp . A sample was taken as a non-immunoprecipitated input control for sequencing . Following clarification by centrifugation , 1/10 volume of 10% Triton X-100 in lysis buffer was added to each sample followed by 100 μl of Dynal-Protein G beads coated with anti-V5 monoclonal antibody ( Sigma ) and incubated overnight with rotation . The beads were washed 5X with ChIP RIPA buffer [50 mM HEPES pH 7 . 5 , 500 mM LiCl , 1 mM EDTA , 1% NP40 , 0 . 7% DOC] , then 1X in TE + 50 mM NaCl and resuspended in 100 μL elution buffer [50 mM Tris-HCl , pH 7 . 5 , 10 mM EDTA , 1% SDS] . Samples were incubated at 65°C for 30 min and the beads pelleted by centrifugation . Supernatants were incubated at 65°C overnight to reverse crosslinks . Samples were incubated with 8 μL of 10 mg/mL RNase A for 2 hr at 37°C , then 4 μL of 20 mg/ml proteinase K at 55°C for 2 hr , then purified . Experiments were repeated in at least biological duplicate . Sequencing sample preparation was performed as previously described [37] . Samples were sequenced using Illumina HiSeq . Data processing for ChIP-seq was performed as previously described [37–39] . Sequence reads were aligned to the V . cholerae N16961 genome using CLC genomic workbench software . CLC genomic workbench ChIP-seq software was used to compare control input and experiment alignments to identify peak enrichment . Our DNA sonication method results in an average DNA fragment size of ~300bp with a spread of 50-800bp . A transcription factor can occupy the extreme ends of up to an 800bp fragment allowing a raw peak to be called that spans up to ~1600bp . We have reported these maximum raw coordinates in S1 Table ( ToxR ) and S2 Table ( H-NS ) , without computational refinement . Peaks that were identified in both replicates were scored as real peaks . All motif studies were performed using the MEME Suite of motif-based sequence analysis tools [79–81] . Genome scanning for motifs was performed with FIMO version 82 with a stringent p-value cut-off of <0 . 0001 . FIMO returns sequences that match the input motif with a probability specified by the p-value . Identification of ToxR and TcpP binding motifs from ChIP-seq data was performed with both MEME and GLAM2 . We analyzed the respective ChIP-seq data as a whole , and separated into peaks found on genomic islands compared to peaks found on the progenitor genome . We screened motifs generated through our analysis to determining if they overlapped with the biochemically proven binding sites for TcpP in the toxT promoter , and for ToxR in the toxT , ompU , and ompT promoters . We focused on identification of ungapped motifs . We did not identify a TcpP motif that meets our criteria . We identified a ToxR motif using GLAM2 present in all ToxR ChIP-seq peak sequences that met our criteria . For ChIP-seq peak validation , relative abundance quantitative PCR ( qPCR ) was performed with Kapa Biosystems Sybr Fast One-Step qRT-PCR kit using 16S rDNA as the internal reference . Relative target levels were calculated using the ΔΔCt method , with normalization of ChIP targets to 16S rDNA signal [37] . For gene expression analysis , relative expression reverse-transcription quantitative PCR was performed with Applied Systems RNA-Ct one-step system . Relative expression levels were calculated using the ΔΔCt method , with normalization of gene targets to16S rRNA signals [37] . RNA was prepared from logarithmic cultures in triplicate under the same growth conditions used for ChIP-seq . Equal amounts of total RNA were separated on a 6% TBE-urea gel and transferred to Hybond N membrane . After crosslinking and prehybridization , membranes were incubated with 100 pmol of 32P labeled probe . Washed membranes were exposed to film overnight . Bands were quantified by densitometry . RyhB and 5S probes are listed in S5 Table . A modified version of the protocol of Baselski and Parker [115] was performed for infection and recovery of all strains . Strains were grown on selective medium overnight at 37°C . Wild-type and mutant strains were mixed together in LB . 50 μL of this competition mixture ( ∼50 , 000 bacteria ) was inoculated into a 5-day-old CD1 mouse pup ( Charles River Company ) . One strain carried an active lacZ allele . Serial dilutions of the competition mixture were plated on selective medium and enumerated to determine the input ratio of wild type and mutant strain . After incubation at 30°C for 18 hr the mouse pups were sacrificed and small intestines were removed and homogenized in 10 mL of LB . Serial dilutions were plated in LB + Sm100 + Xgal and enumerated to determine the output ratio of wild-type and mutant strain . The competitive index for each mutant is defined as the output ratio of mutant/wild-type strain divided by the input ratio of mutant/wild-type strain . Statistical significance was determined by comparing the resulting ratio to the ratio of WT versus WT lacZ− . At least five mice were tested for each mutant . Data were analyzed using GraphPad Prism 5 Software . Statistical significance between two groups was assessed using an unpaired two-tailed Student’s t test . Statistical significance when comparing more than two groups was assessed using a One-Way ANOVA analysis followed by a Tukey’s multiple comparison post-test . Standard error of the mean ( SEM ) is shown . The sequence data have been deposited with the NCBI’s Gene Expression Omnibus under Accession Number GSE72474 . | The transcription factor ToxR initiates a virulence regulatory cascade required for V . cholerae to express essential host colonization factors and cause disease . Genome-wide expression studies suggest that ToxR regulates many genes important for V . cholerae pathogenesis , yet our knowledge of the direct regulon controlled by ToxR is limited to just four genes . Here , we determine ToxR’s genome-wide DNA-binding profile and show that ToxR is a global regulator of both progenitor genome-encoded genes and horizontally acquired islands that encode V . cholerae’s major virulence factors . Our results suggest that ToxR has gained regulatory control over important acquired elements that not only drive V . cholerae pathogenesis , but also define the major transitions of V . cholerae pandemic lineages . We demonstrate that ToxR shares more than a third of its regulon with the histone-like nucleoid structuring protein H-NS , and antagonizes H-NS for control of critical colonization functions . This regulatory interaction is the major role of ToxR in V . cholerae colonization , since deletion of hns abrogates the need for ToxR in V . cholerae host colonization . By comparing the genome-wide binding profiles of ToxR and other critical virulence regulators , we show that , despite similar predicted DNA binding requirements , ToxR is unique in its global control of progenitor-encoded and acquired genes . Our results suggest that factors in addition to primary DNA structure determine selection of ToxR binding sites . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"biofilms",
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"pathogens",
"vibrio",
"regulatory",
"proteins",
"microbiology",
"dna-binding",
"proteins",
"regulator",
"genes",
"vibrio",
"cholerae",
"transcription",
"factors",
"gene",
"types",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"dna",
"bacteria",
"bacterial",
"pathogens",
"promoter",
"regions",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"molecular",
"biology",
"biochemistry",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
]
| 2016 | ToxR Antagonizes H-NS Regulation of Horizontally Acquired Genes to Drive Host Colonization |
Metagenome analysis of the gut symbionts of three different insects was conducted as a means of comparing taxonomic and metabolic diversity of gut microbiomes to diet and life history of the insect hosts . A second goal was the discovery of novel biocatalysts for biorefinery applications . Grasshopper and cutworm gut symbionts were sequenced and compared with the previously identified metagenome of termite gut microbiota . These insect hosts represent three different insect orders and specialize on different food types . The comparative analysis revealed dramatic differences among the three insect species in the abundance and taxonomic composition of the symbiont populations present in the gut . The composition and abundance of symbionts was correlated with their previously identified capacity to degrade and utilize the different types of food consumed by their hosts . The metabolic reconstruction revealed that the gut metabolome of cutworms and grasshoppers was more enriched for genes involved in carbohydrate metabolism and transport than wood-feeding termite , whereas the termite gut metabolome was enriched for glycosyl hydrolase ( GH ) enzymes relevant to lignocellulosic biomass degradation . Moreover , termite gut metabolome was more enriched with nitrogen fixation genes than those of grasshopper and cutworm gut , presumably due to the termite's adaptation to the high fiber and less nutritious food types . In order to evaluate and exploit the insect symbionts for biotechnology applications , we cloned and further characterized four biomass-degrading enzymes including one endoglucanase and one xylanase from both the grasshopper and cutworm gut symbionts . The results indicated that the grasshopper symbiont enzymes were generally more efficient in biomass degradation than the homologous enzymes from cutworm symbionts . Together , these results demonstrated a correlation between the composition and putative metabolic functionality of the gut microbiome and host diet , and suggested that this relationship could be exploited for the discovery of symbionts and biocatalysts useful for biorefinery applications .
Insects represent one of the most diverse groups of organisms on the planet that can adapt to the extremely diverse eco-environments . In particular , herbivorous insects can exploit a wide range of the plant species as food sources [1] . Insect gut symbionts play an essential role in the insect adaptation to various food types and they have been shown to be important for lignocellulosic biomass degradation , nutrient production , compound detoxification , and environmental adaptation [2]–[7] . Disrupting insect gut symbionts can significantly reduce the fitness of insects and can even cause serious diseases such as CCD ( Colony Collapse Disease ) [8] . Moreover , insect gut symbionts also were shown to be maternally inheritable from generation to generation , which suggests the symbiotic microbiota is a dynamic component of the competitive evolution between plants and herbivorous insects as well as a driving force for insect speciation [9] , [10] . For these reasons , insect gut symbionts have been the subject of extensive studies in recent years [10] . Previous studies highlighted several important features of some insect gut symbionts including their reduced genome size , convergent evolution , co-speciation , and complementary function with the host genome [11]–[15] . Recent studies also expanded our understanding of the roles of insect gut symbionts in non-conventional functions like nitrogen recycling , reproductive manipulation , pigment production and many other aspects related to insect fitness [16] , [17] . Despite the progress toward understanding insect-symbiont relationships , there is still much to be learned especially with regard to facultative symbionts . Moreover , limited research has focused on comparing the gut symboints from insect species that specialize on different food sources . For this reason , we systemically compared the gut enzyme activities and microbial diversity in several insect species relevant to biotechnology applications [2] , [3] , [18] . Previous studies comparing gut symbionts from woodbore ( Cerambycidae sp . , ( Coleoptera ) ) , silkworm ( Bombyx mori ( Lepidoptera: Bombycidae ) ) , and grasshopper ( Acrida cinerea ( Orthoptera: Acrididae ) ) suggested that the insect gut cellulytic enzyme activities were generally correlated with the lignocellulosic biomass composition in the food consumed [2] . Furthermore , the comparison of the microbial community structure of gut symbionts from woodbore , silkworm , grasshopper , and cutworm ( Agrotis sp . ( Lepidoptera:Noctuidae ) ) using DGGE ( Denaturing Gradient Gel Electrophoresis ) revealed significant differences in symbiotic community correlating with food adaptation [3] . Despite the progress , an in-depth understanding of the eco-evolutionary adaptation to food types requires metabolic and phylogenic analysis that cannot be offered by traditional approaches like DGGE [18] . Most of the previous comparative studies of symbionts from different insect species were either carried out with DGGE or focused on one or few symbiotic species [19] , [20] . Compared to those conventional techniques , new platforms like metagenomics could help define the function of symbionts in the food adaptation of insects and promote discovery of biocatalysts for biotechnology applications [18] . From the deep sea to the human intestine system , metagenome analysis has emerged as a major approach to study the composition , function , and evolution of various microbiota [21] . Metagenome analysis and metabolic reconstruction of the termite gut symbiotic microbiota revealed potential functionality in these microbiomes that might be required for biomass degradation , nutrient synthesis and other functions essential to the insect [22]–[24] . Moreover , those studies also highlighted the potential for biotechnology application of insect gut symbionts , since many potential glycosyl hydrolases ( GH ) family enzymes have been identified from the termite gut [24] . Further studies revealed the potential complementary function between the host and symobionts enzymes for highly efficient biomass degradation [23] . Despite the progress , previous research mainly focused on the metagenome sequencing of symbionts in single insect species or the same symbioint in different insect species [17] , [25]–[27] . Few studies have systematically compared the metagenomes of symbiotic microbiota from insect species with distinctly different diets , environmental adaptations , or life histories . This type of comparative metagenomics approach has the potential to substantially improve our understanding of the adaptive significance of insect gut symbionts for insect diet specialization as well as facilitates the discovery of novel biocatalysts for biorefinery applications . In this study , we selected three insect species that are from different insect orders and have different diets and life histories characteristics: grasshopper ( Acrida cinerea ( Orthoptera ) , cutworm ( Agrotis ipsilon ) ( Lepidoptera ) and termite , Nasutitermes sp . ( Isoptera: Termitidae ) . The grasshopper is a polyphagous insect specializing on different plant leaves , mainly from the monocot grass species . Previous studies revealed that the grasshopper diet contains about 37 . 2% of forbs , 58% of grasses and sedges and 4 . 8% of others [28] . The cutworm is also a polyphagous , generalist that can adapt to a broad range of food sources including cabbage , asparagus , bean , and other crucifers [29] . In contrast , the termite is monophagous insect that specializes on lignocellulosic biomass as a food source . The three insects also differ in life cycle . The cutworm is a holometabolous insect that undergoes complete metamorphosis with a pupal stage [30] , whereas the grasshopper and termite are hemimetabolous , having incomplete metamorphosis and juveniles with morphologies similar to adults [31] . Metagenome data from the gut symbiotic microbiota of grasshopper and cutworm were generated using Illumina Genome Analyzer , and these metagenome data were compared with the updated sequencing data from gut symbionts of the wood-feeding higher termite [24] . As one of the first comprehensive comparisons of insect gut symbiotic metagenome , the goal was to examine the relationships between the taxonomic and potential metabolic diversity of the insect gut microbiomes and the diets and life histories of their insect hosts at the community , metabolic pathway , and molecular levels . The analysis indicated that the composition of gut symbionts was correlated with their function in biomass degradation and nutrient biosynthesis . The metabolic reconstruction revealed the presence of specific pathways relevant to the utilization and transport of diverse carbohydrate sources in cutworm and grasshopper . The diversity , phylogenetic , metabolic , and functional analyses all supported the hypothesis that insects and their gut symbionts co-evolved with the food preferences of the insect toward optimal capacities in biomass degradation , macromolecule intake and utilization , complementary nutrient synthesis , and other aspects related to insect life style . In addition , we cloned 24 biomass degrading enzymes based on the predicted gene models and characterized four of them . Enzyme assays revealed that grasshopper cellulytic enzymes were generally more active than the cutworm cellulytic enzymes , which confirmed the presence of functional diversity at the protein . The enzyme characterization indicated that insect guts were useful resources for discovering novel biocatalysts for biorefinery applications .
Relative abundance of symbiotic microbial species in each insect gut was estimated based on the species distribution of the gene-coding sequences as annotated by the BLAST search . The cluster analysis of bacterial species distribution for the gut symbionts was shown in Figure 1 . It should be pointed out that Figure 1 only represented a rough estimation of the microbial species distribution because of the genome size variations in different symbionts , which complicated the data interpretation . Nevertheless , the comparison of the relative abundance of the bacteria phyla in the microbiota from the three different insect species revealed that the microbiota composition was rather different from each other and these differences might be relevant to the functions they provided for their insect hosts . The dominant groups differed among the three insect species . For the cutworm , the phylum Bacilli was the dominant group ( 24 . 14% ) , followed by Clostridia ( 4% ) , Erysipelotrichi ( 3 . 64% ) and γ-proteobacteria ( 1 . 43% ) ( Figure 1 ) . For the grasshopper , the most common bacterial genes were from γ-proteobacteria ( 25 . 16% ) , followed by Erysipelotrichi ( 3 . 51% ) , Clostridia ( 1 . 27% ) , and Bacilli , ( 0 . 84% ) , respectively ( Figure 1 ) . For both species , the most abundant groups comprised about 25% of the diversity , whereas the second most abundant groups comprised less than 5% . Even though the insects differed in microbial composition , there were some similarities that likely were related to function . Both Clostridia and Bacilli species have been shown to be the major groups of microbes responsible for biogas production and biomass conversion in microbial communities [32] . Many Clostridia species such as C . thermocellum and C . ljungdahlii are anaerobic Firmicutes known to have a robust capacity to use cellulose , hemicellulose , and other carbohydrate [33]–[35] . The presence of a large proportion of Clostridia was likely to be important for lignocellulosic biomass degradation [34] , [36] . However , the predominance of the γ-proteobacteria in grasshopper was unexpected , because γ-proteobacteria has not been shown previously to be involved in biomass utilization . However , recent work revealed that γ-proteobacteria might be important nutrient providers for host insects . For example , γ-proteobacteria as facultative or obligate endosymbionts were shown to play essential roles for insects like tsetse fly in the utilization of low nutrient food sources [37] . Similarly , the predominance of γ-proteobacteria in grasshoppers might be important for the utilization of the grasses , which characteristically have high fiber content . Compared to the grasshopper and cutworm microbiomes , the microbial composition of the termite microbiome reflected its unique adaptation to utilization of woody species , where both the Clostridia and the Spirochaetes species were predominant ( Figure 1 ) [24] . Additionally , the termite microbiome was composed of several major groups with more than 5% abundance . Morphologically diverse spirochaetes were consistently present in the hindgut of all termites [38] , and was found as ectosymbionts attached to the surface of cellulose-digesting protists [39] . Overall , the microbial populations of the cutworm , grasshopper and wood-feeding termite gut systems appeared to consist of taxa with known capacities for degrading and utilizing the different types of foods on which their insect hosts specialize . In addition to gene-coding sequence-based analyses , we also implemented two types of phylogenetic analyses . First , two partial 16S rRNA clone libraries were established from the PCR amplified 16S rRNA sequences using 515F/1492R primers . Sanger sequencing was used to sequence individual 16S rRNA clones as summarized in Table S1 . The phylogenetic analysis was presented in Figure 2 . The second phylogenetic analysis was based on the annotation of the contigs derived from the metagenome sequence assembly . The assembled contigs were first aligned to the 16S rRNA genes from the recent release of RDP database using blastn . The analysis resulted in 188 and 102 contigs assigned to be 16S rRNA for cutworm and grasshopper , respectively ( Table S1 ) . The most similar partial or complete 16S rRNA sequences from the database were used for the multiple sequence alignment and phylogenetic analysis using Maximum likelihood method ( RAxML ) . The analysis results were presented in Figure S1 . The results from the two types of analysis generally were consistent; although the phlygenetic analysis based on the annotated contigs ( Figure S1 ) provided a deeper coverage of microbial species and a better representation of uncultured species . The phylogenetic analyses ( Table S1 , Figure 2 , Figure S1 ) revealed three features . First , proteobacteria represented the most diverse group of the microbes in the microbiomes of both grasshopper and cutworm . Among the proteobacteria , γ-proteobacteria was the predominant taxa and the 16S rRNA sequences from cutworm and grasshopper formed two distinct clades , indicating the relatively independent evolution of the gut microbiome in the two species . The 16S rRNA-based phylogenetic analysis correlated well with the microbial abundance analysis using gene models ( Figure 1 ) . The studies confirmed the differences in abundance , phylogeny , and evolution of gut symbionts between cutworm and grasshopper . A second feature of the analyses was that the cutworm had more species of gut symbionts than grasshopper ( 188 vs . 102 , Figure S1 ) . We speculated that the greater diversity of symbionts in the cutworm gut as compared to that of the grasshopper might be relevant to its being both more of a dietary generalist . A third feature was the discovery of large number of uncultured species or unknown species . Uncultured species referred to the species that cannot be cultured in standard medium , whereas unknown species referred to those lacking taxonic information . Due to the deeper coverage of metagenomic sequencing compared to the PCR cloning library , Figure S1 showed almost 60% sequences were from uncultured or unknown species . The results highlighted our limited knowledge of the diversity of insect gut symbionts . It was proposed that the existence of many unculturable species might be related to the significant reduced genome and limited metabolic capacity of some symbiotic microbes [40]–[43] . The phenomena indicated that the metabolic capacity of insect gut microbiota should be considered as a whole instead of based on individual species . Another observation was that 14 and 10 16S rRNA sequences were assigned to Acetobacter pasteurianus ( AP011163 ) for cutworm and grasshopper , respectively ( Figure S1 ) . Acetobacter strains belong to acetic acid bacteria ( AAB ) , which are often found in various categories of fruits , flowers , and fermented foods [44] and some insect guts [45] . Acetobacter might have originally been acquired from the food sources of cutworm and grasshopper and subsequently become a more permanent symbiont for the two species or might occur as a transient resident . Acebacter can produce alcohol dehydrogenase ( ADH ) , which could potentially contribute to lignin oxidation for lignin degradation/modification in termite guts [46] , [47] . Overall , the phylogenetic analysis indicated correlations between microbial composition and function and insect diet preference . Metagenome sequencing provided more detailed functional comparisons of different gut symbionts using pathway analysis based on COGs ( Clusters of Orthologous Groups ) and KEGG ( Kyoto Encyclopedia of Genes and Genomes ) [48] , [49] . KEGG maps the genes within the biological pathways to derive potential functions [50] , whereas COG analysis uses evolutionary relationships to group functionally relevant genes [51] . The annotation of the cutworm and grasshopper gut microbiomes yielded 11 , 317 and 8954 hits for the COG database as well as 900 and 1105 hits for the KEGG pathways , respectively . D-ranks analysis was used to evaluate the relative enrichment of COG and KEGG gene categories in the cutworm and grasshopper gut symbiotic metagenomes compared to the termite metagenome . The enrichment or under-representation of COG categories were as shown in Figure 3 . Both cutworm and grasshopper gut symbionts were enriched in several metabolic pathways compared to termite gut symbionts . Cutworm gut symbionts were enriched with genes for carbohydrate transport and metabolism , and defense mechanisms ( P<0 . 05 ) relative to grasshopper symbionts . The diversity in carbohydrate metabolism genes correlated well with the taxonomic diversity of the gut microbiomes ( Figure S1 ) and were consistent with the hypothesis that the greater diversity in species composition and carbohydrate metabolism observed in the cutworm may be related to the broader diet preference and more complicated life histories of the cutworm compared to those of the grasshopper . The ontology analysis based on KEGG revealed similar patterns as shown in Table S2 , where flagella assembly in cell motility and type III secretion system ( P<0 . 05 ) are more enriched in termite gut symbionts than those of cutworm and grasshopper , although it is unclear why this would be so . Overall , the metagenomic composition of genes in all categories reflected their potential function in adaptation to insect diet and life history . A more detailed functional relevance can be derived from examination of specific pathways . Metabolic reconstruction provided comparison of potential biocatalyst functionality in four general COG categories and thus a means of relating the metabolic diversity and capability of the microbiome to the insect diet and life style . The ultimate goal of this research was to discover novel biocatalysts for biorefinery applications . We therefore cloned and characterized several enzymes for functional validation . A total of 24 ORFs of predicated plant polysaccharides degradation enzymes were PCR amplified using primers based on the assembled sequences ( Figure S2 ) . A total of 22 out of 24 ORFs amplified and the sequences of all of the amplicons were consistent with the assembled sequences ( Figure S2 ) . The results highlighted the reliability of the Illumina metagenomic sequencing and assembly to identify degredation enzymes . Our research represents one of the few metagenome sequencing efforts to rely mainly on the Illumina Genome Analyzer [69] . We further characterized an endoglucanase ( CW-EG1 and GH-EG1 ) and a xylanase ( CW-Xyn1 and GH-Xyn1 ) from both the grasshopper and cutworm guts , respectively . The selected enzymes were expressed and purified by a His-trap nickel column , as indicated by SDS-PAGE ( Figure S3 ) . The enzyme performance under different temperature and pH conditions was as shown in Figure S4 . All four of the enzymes exhibited activity , and the activities were significantly influenced by temperature and pH . Most enzymes had temperature optima at 60∼70°C and pH optima at 7 . 0–9 . 0 ( Figure S4 ) . This pH range correlates with the fact that many insect gut systems have a slightly basic environment [70] Considering that many traditional filamentous fungi enzymes had optimal activity in the weakly acidic pH range , the insect gut enzymes provided complementary capacity for biomass degradation . We further compared the specific activity of the same category of enzymes from cutworm and grasshopper gut microbiome . Interestingly , for both cellulase and xylanase , the grasshopper gut enzymes were significantly higher than those of cutworm ( P<0 . 05 , Figure 5 ) . The result correlated with our previous analyses of gut content activities , even though the differences could also result from the choice of enzymes and other factors [2] . The adaptation to relatively higher temperature made the enzymes good candidates for some biomass conversion applications . Together with many recent studies , our research indicated that insect gut symbionts are substantial resources for enzyme discovery for biorefinery applications . The relationship between the diversity and potential functional capabilities of the gut microbiomes and insect food preference is particularly relevant improvements in biomass degradation , and thus should be explored for biotechnology applications [71]–[75] . Due to the technical limitations , we particularly focused on the bacterial symbionts in this study . Nevertheless , the fungal and protozoal symbionts in insect guts were also widely studied for their biomass degradation capacity . These eukaryote symbionts should be investigated for their roles in biomass deconstruction , food and life history adaptation in the follow-up studies .
Metagenome analysis requires comprehensive coverage of most multiple species in the sample [76] . To obtain sufficient high-quality DNA for sequencing with Illumina Genome Analyzer , approximately 2000 third to fifth instar grasshoppers and 50 fourth to fifth instar cutworms were dissected to extract genomic DNA from gut symbionts . A recently developed indirect DNA extraction method was modified for the insect gut metgenomic DNA extraction [77] . The extracted metagenomic DNA were quantified by a Nano Drop ND-1000 spectrophotometer and characterized by electrophoresis . Moreover , the quality of the DNA was verified by PCR amplification of conserved 16S rRNA for bacteria and conserved 18S rRNA for insect host contamination [29] . The results confirmed that the metagenomic DNA is free from host DNA contaminations , because the 18S rRNA did not amplified . Metagenome sequencing of cutworm and grasshopper gut symbiotic microbioata was carried out using Illumina Genome Analyzer II ( Illumina , Inc . CA , USA ) with paired-end 76 base sequencing . Library construction was carried out following the manufacture's recommendation using Illumina Paired-End Sequencing Kit ( Cat . No . PE-102-1001 ) . Briefly , 2 to 5 µg metagenomic DNA was sheared by nebulization to generate DNA fragments and the ends were repaired with Klenow , followed by several steps to add the adapters . Adapter-ligated DNA fragments of length 300–350 bp were isolated from a 2% agarose gel using QIAquick Gel Extraction Kit . The fragments were then amplified by 11 cycles of PCR reaction to generate the DNA library at a concentration of 20–35 ng/µl . The median size of the library was evaluated using 2% agarose gel . The PHIX Control V2 Library was prepared by Illumina ( Cat . No CT-901-2001 ) and used for sequencing . Approximately 5 pmol DNA libraries were subjected to cluster generation and sequenced by DNA core of Institute of Plant Genomics and Biotechnology . The images were processed using version 0 . 3 of the GAPipeline software supplied by Illumina . After base-calling with GAPipeline software , the remaining 44 , 155 , 246 ( cutworm ) and 58 , 033 , 340 ( grasshopper ) reads ( each is about 76 bases ) were trimmed and assembled using Velvet version 0 . 7 . 55 ( http://www . ebi . ac . uk/~zerbino/velvet/ , European Bioinformatics Institute , EMBL-EBI ) . The resulted assembly consisted of 64 , 065 and 78 , 991 contigs for cutworm and grasshopper , respectively . The draft assembled contigs ( ≥100 bp ) were loaded into IMG/M ( http://www . jgi . doe . gov/m ) [78] . Before further analysis , the IMG/M system first carried out a gene model validation process , including editing overlapping CDSs , correcting start codons , and identifying missed genes and pseudogenes [78] . The predicted coding sequences ( CDSs ) and some functional RNAs were recorded with start/end coordinates in the contigs . The predicted genes were assigned to COGs ( clusters of orthologous groups ) based on RPS-BLAST ( reverse position specific BLAST ) and NCBI's Conserved Domain Database ( CDD ) , using an e-value threshold of 10−2 without low-complexity masking [79] . Genes were also probed against Pfam database using HMMER search ( http://hmmer . janelia . org/ ) [80] , [81] . Protein-coding sequences were further annotated for molecular function and pathways using KEGG pathways . In addition , the metagenome sequences and gene models were binned to rank domain , phylum , and class using PhyloPythia [82] . The phylogenetic analysis of 16S rRNA was carried out with two types of analyses . First , two clone libraries were prepared using PCR products amplified from cutworm and grasshopper gut metagenome DNA with one pair of primers broadly targeting the V3–V9 region of 16S rRNA . The primer sequences were 515F ( 5′-GTGCCAGCAGCCGCGGTAATACCTTGTTACGACTT-3′ ) and 1492R ( 5′-GGTTACCTTGTTACGACTT-3′ ) [83] . 87 and 97 near complete 16S rRNA V3–V9 region sequences were obtained for cutworm and grasshopper gut microbiome , respectively . The 16S rRNAs was then used for phylogenetic analysis . In addition to sequencing of the V3–V9 region , we also sought to reach a deep coverage of symbiotic species by analyzing the assembled metagenome sequences . 16S rRNA sequences were identified using BLASTN ( E<1×10−5 and a sequence length hit >50 nt ) search against the SSU rRNA genes from release 16 . 3 . 3 of the RDP database ( http://rdp . cme . msu . edu/ ) [84] , and the European Ribosomal RNA database ( http://www . psb . ugent . be/rRNA/index . html ) . Due to the high similarity , it is usually difficult to isolate the 16S rRNA genes from de novo assembly of metagenome data . A total of 96 and 53 partial and near complete 16S sequences were extracted from 188 and 102 assembled contigs for cutworm and grasshopper gut microbiomes , respectively . The sequences were then aligned with the NAST aligner [85] , and imported into an ARB database ( http://greengenes . lbl . gov ) [86] . The nearest aligned full length sequences were used for classification and phylogenetic tree construction using RAxML [87] . Phylogenetic analysis was carried out using the Minimum Evolution method with the sum of branch length = 5 . 0 [88] . The evolutionary distances were computed using the Maximum Composite Likelihood method with 1000 replicates of bootstrap tests [89] . In order to compare the metabolic pathways for different microbiota , the coding sequences were analyzed with KEGG and COG ( Clusters of orthologous groups ) . Both grasshopper and cutworm symbiotic metagenome and updated termite metageome data ( JGI IMG Database GOLD ID: GM00013 and Sample ID: GS0000048 ) [24] were compared . For KEGG analysis , all coding sequences were converted into KEGG orthologous ( KO ) groups , and the KEGG pathway annotation was extracted based on the latest release of KEGG version ( Release 55 . 1 , September 1 , 2010 ) . The COG assignment was based on RPS-BLAST and NCBI's Conserved Domain Database ( CDD ) . Only 4 . 95% , 3 . 48% , and 6 . 41% of predicted genes were assigned to KEGG pathway for grasshopper , cutworm , and termite gut microbiome , respectively . 39 . 4% , 44 . 41% , and 53 . 56% of coding sequences were assigned to COG terms for grasshopper , cutworm and termite gut microbiome , respectively . In order to further define the enrichment or under-representation of a KEGG pathway or a COG term in a certain microbiome , two metrics were used in this study . For the comparison of a protein family between a query metagenome and a reference metagenome , the D-scores were calculated using a binomial distribution . We calculated the D-score using ( f1–f2 ) /sqrt ( p*q * ( 1/n1+1/n2 ) ) , where f1 = x1/n1 = frequency of functional occurrence in query group , f2 = x2/n2 = frequency of functional occurrence in reference group , p = ( x1+x2 ) / ( n1+n2 ) = probability of occurrence , q = 1−p = probability of non-occurrence . Specifically , x1 was the number of a given function in query group , x2 was the number of a given function in reference group , n1 was total counts of all function occurrences in query group , and n2 was total counts of all function occurrences in reference group . Further analysis involved D-rank , a normalization ranking for each pair wise comparison . D-rank was calculated by adding the D-scores of all protein families assigned to a certain functional category and then normalized by the square root of the number of total categories [90] , [91] . In order to verify the quality of sequence assembly and discover novel biocatalysts , 24 predicted coding genes for carbohydrate degrading enzymes were amplified , among which 22 showed positive results . Among the 22 , four were expressed and analyzed . The same batch of sequenced metagenomic DNAs were used as template for PCR amplification . The PCR mixture ( 50 µl ) contained 5 µl of 10× PCR buffer , 4 µl of MgCl2 ( 25 mM ) , 1 µl of dNTP , 1 µl of each primer ( 10 mM ) , 37 µl of sterile Milli-Q water , 0 . 5 µl of Taqpolymerase ( AmpliTaq Gold DNA Polymerase , Applied Biosystems , CA , USA ) , and 0 . 5 µl of DNA templates . PCR were carried out under the following conditions: an initial denaturation at 94°C for 5 min; 35 cycles of denaturation at 94°C 30 s , annealing at 55°C 1 min , and extension at 72°C for 1 . 5 min . The final step of the PCR was an extension step at72°C for 7 min , followed by cooling at 4°C . The PCR products were analyzed by gel electrophoresis . Two predicted endoglucanase genes and two xylanase genes were cloned and expressed as described by Shi et al ( 2011 ) [29] . Briefly , the endoglucanase and xylanase genes were cloned into pET161 vector ( Cat No . K160-01 , Invitrogen , USA ) with a 6×His-tags . The enzyme expressions were induced in BL21 ( DE3 ) cells with 0 . 5 mM IPTG at 25°C for 5 hours . The expressed enzymes were purified through a 5-ml nickel affinity column in AKTA FPLC system ( GE healthcare , USA ) . Cellulase and xylanase activities were measured by the amount of reducing sugars released using dinitrosalicylic acid [92] . One unit was calculated as 1 µmol reducing sugar released per minute using glucose as standard . This Whole Genome Shotgun project was deposited at DDBJ/EMBL/GenBank under the accession AKYZ00000000 and AKZA00000000 for grasshopper and cutworm , respectively . The version described in this paper is the first version , AKYZ01000000 and AKZA01000000 . The Genbank ID for the four enzymes was as follows; cutworm EG1 is JX434086; grasshopper EG1 is JX434088; cutworm XYN1 is JX434089; and grasshopper XYN1 is KC155983 . | The symbiotic gut microbiome of herbivorous insects is vital for their ability to utilize and specialize on plants with very different nutrient qualities . Moreover , the gut microbiome is a significant resource for the discovery of biocatalysts and microbes with applications to various biotechnologies . We compared the gut symbionts from three different insect species to examine whether there was a relationship between the diversity and metabolic capability of the symbionts and the diet of their hosts , with the goal of using such a relationship for the discovery of biocatalysts for biofuel applications . The study revealed that the metabolic capabilities of the insect gut symbionts correlated with insect adaptation to different food types and life histories at the levels of species , metabolic pathway , and individual gene . Moreover , we showed that the grasshopper cellulase and xylanase enzymes generally exhibited higher activities than those of cutworm , demonstrating differences in capabilities even at the protein level . Together , our findings confirmed our previous research and suggested that the grasshopper might be a good target for biocatalyst discovery due to their high gut cellulytic enzyme activities . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"biotechnology",
"biofuels",
"biology",
"genomics",
"microbiology",
"agricultural",
"biotechnology",
"genetics",
"and",
"genomics",
"agriculture"
]
| 2013 | Comparative Genomic Analysis of the Endosymbionts of Herbivorous Insects Reveals Eco-Environmental Adaptations: Biotechnology Applications |
The rod-shaped bacterium Escherichia coli selects the cell center as site of division with the help of the proteins MinC , MinD , and MinE . This protein system collectively oscillates between the two cell poles by alternately binding to the membrane in one of the two cell halves . This dynamic behavior , which emerges from the interaction of the ATPase MinD and its activator MinE on the cell membrane , has become a paradigm for protein self-organization . Recently , it has been found that not only the binding of MinD to the membrane , but also interactions of MinE with the membrane contribute to Min-protein self-organization . Here , we show that by accounting for this finding in a computational model , we can comprehensively describe all observed Min-protein patterns in vivo and in vitro . Furthermore , by varying the system's geometry , our computations predict patterns that have not yet been reported . We confirm these predictions experimentally .
Nature presents an overwhelming variety of forms and patterns . While system specific conditions can play an important role for their formation , also a few general principles underlying biological pattern formation have been proposed in the past . A particularly attractive concept is the spontaneous formation of patterns in reaction diffusion systems as proposed by Alan Turing [1] . In this case , a ( small ) number of different constituents collectively form large-scale patterns . So far , however , only a few biological examples of bona fide Turing patterns are known [2] . An example of subcellular pattern formation due to reactions and diffusion of just two different constituents is provided by the Min system in the rod-shaped bacterium Escherichia coli [3] . This protein system forms a spatiotemporal oscillation in the cell , that is , a standing wave with a node in the cell center [4] , [5] , see Figure 1A , which plays an essential role in division site selection in E . coli . Whereas the oscillations emerge solely from the interactions between MinD , MinE and the membrane , the inhibitor of cell division MinC binds to MinD and is thus distributed similarly: it appears periodically at the cell poles , but is practically absent from the cell center . In this way , division occurs in the cell center leading to two daughter cells of the same size . While some models suggest that particular properties of the cell poles might play an essential role for Min-protein pattern formation [6] , a number of observations support the notion that the Min system can self-organize without any additional spatial cues . First of all , depending on the cell geometry and the Min-protein expression level , the protein pattern can change: In longer cells , standing waves with several nodes form [4] , see Figure 1B , whereas in shorter cells and for slightly over-expressed Min proteins , oscillations are replaced by stochastic switching of the proteins between the two cell halves [7] , [8] . In Y-shaped cells , the proteins visit the different arms in a way that depends on the arms' lengths [9] . Furthermore , in vitro studies of purified proteins found MinD and MinE to spontaneously organize into collective traveling waves [10] . Together , these observations suggest that the Min-protein patterns emerge from the intrinsic dynamics of these proteins , in particular , the exchange of proteins between the membrane , driven by the high affinity of MinD for the membrane when ATP is bound and a low affinity with ADP bound [11] . In addition , membrane-bound MinD recruits MinE , which in turn induces hydrolysis of the bound nucleotide by MinD and consequently MinD detachment from the membrane . These well-established processes are at the core of a number of computational models reproducing the Min-protein oscillations observed in E . coli [12] . The most popular mechanism studied through such models assumes that cooperative membrane-attachment of MinD is at the origin of pattern formation . In the simplest version , the rate of MinD attachment to the membrane increases in presence of membrane-bound MinD [13] . Several works on models implementing cooperative membrane attachment in various ways and complementing it by different side processes have shown that it can robustly generate the pole-to-pole oscillations observed in E . coli [14]–[16] even during septum closure [17] . Other works rather emphasized cooperative effects between already membrane-bound MinD [18] , [19] . However , in spite of more than a decade of theoretical analysis , there exists to date no comprehensive description of all Min-protein patterns observed in vivo and in vitro . Some evidence suggested that an N-terminal helix allows MinE to also interact with the membrane [20] , [21] , however , it remained unclear if this property was important for the self-organization of the Min system . Single molecule data obtained in vitro [22] as well as genetic , physiological , and structural analysis [23] finally provided evidence that the ability of MinE to interact with phospholipids allows it to remain bound to the membrane after MinD has detached , which could lead to the subsequent removal of several MinD dimers by one MinE dimer . In analogy to molecular motors that can perform several subsequent steps on a cytoskeletal filament , we call this property “MinE processivity” . This possibility had been proposed earlier on theoretical grounds as it offers a mechanism for the formation of MinE-rings [19] , [24] , [25] and was crucial for describing the guidance of Min-protein waves on patterned substrates [26] . In the present work , we perform a computational study to explore the consequences of this molecular property for large-scale pattern formation . To this end , we use deterministic and stochastic calculations in three dimensions . We show that MinE processivity provides a key to obtain a unified description of all previously described Min-protein patterns in vivo and in vitro . In addition , our analysis predicts hitherto unknown patterns , namely traveling waves in long and moving patches in aberrantly large cells . We confirm the existence of these states by fluorescence microscopy of living E . coli cells . Beyond the Min system , our findings highlight the importance of membrane-binding for subcellular pattern formation .
We first studied the behavior of Min protein patterns in cellular geometries . To this end , we solved the stochastic and deterministic dynamic equations in a cylindrical domain with hemispherical caps . The parameters used in this section are given in Table 1 . The values of the cytosolic diffusion constants have been measured in Ref . [45] . While there is no direct measurement of the diffusion constants for membrane-bound MinD , MinE , and MinDE , diffusion on membranes is usually two to three orders of magnitude smaller than in the bulk [46] . For larger values of these constants , the resulting patterns are broader and less well defined . Decreasing their values does not affect the patterns significantly . To determine the value of the maximal density of membrane-bound proteins , we use that close packing of MinD on the membrane would yield a density of about 1/ ( lateral extension of a MinD dimer ) , with the latter being approximately . To account for crowding of the membrane by other molecules we use a value roughly 10 times smaller , . The values of the various attachment and detachment rates have been chosen to match the experimentally observed patterns . Note , that for the parameter values given in Table 1 , the dominant path for MinE-induced MinD detachment involves MinE staying on the membrane . This corresponds to a high MinE processivity . Finally , we mostly considered the Min patterns in geometries of fixed size . Even under optimal growth conditions , E . coli gains only about 100 nm per oscillation period . As we show below , the patterns are robust against such changes . A major breakthrough in the understanding of Min-protein pattern formation has been achieved by studying the Min-dynamics in open geometries [10] , [22] , [26] , [54] . Experimentally , in vitro studies using supported lipid bilayers have allowed us to clearly establish the propensity of the Min proteins to self-organize [10] . Structural analysis suggested that binding to the membrane can also occur for MinE not associated with MinD [23] , providing a natural explanation for guiding Min-protein waves on structured surfaces [26] . In Figure 8 we present the result of a numeric solution of the dynamic equations ( 8 ) – ( 12 ) , where we have employed periodic boundary conditions in the x- and y-directions . Parameter values are given in Table 1 . The differences between these values and those used for the in vivo geometries reflect differences in the environmental conditions , notably the presence respectively absence of other macromolecules . Similar to the experimental observations , the Min proteins self-organize into traveling waves . The calculated wave profile presents the same features as in the experiment: the MinD profile increases at the wave front and then saturates until it sharply drops . The density of MinE increases more slowly than that of MinD . Towards the wave's trailing edge it exhibits a sharp increase and then drops rapidly . The parameter is increased in comparison to the value determined in the section ‘Min-protein patterns in cellular geometries’ . The presentation of the distribution's z-dependence in Figure 8C shows that the pattern is confined to a layer of about above the membrane . This result justifies a posteriori the use of effective 2d descriptions for the Min-protein dynamics [10] , [26] even though it is not obvious how to formally obtain the 2d equations from the 3d system . The propagation of the wave fronts can be understood by interpreting the space coordinate in Figure 8B as time: First cytosolic MinD binds to the empty membrane . The nonlinearity in the MinD binding term then leads to an increased binding rate and thus to an accelerated increase of the MinD density on the membrane . As soon as membrane-bound MinD is present , MinE starts to attach . As the MinE binding sites are abundant , the increase of the MinE density is roughly linear . With increasing MinE density , the net rate of MinD attachment decreases . Eventually , the MinE-induced detachment rate exceeds the attachment rate and the density of membrane-bound MinD decreases . This decrease is sharp at the waves trailing edge , because MinE processivity leads to an accumulation of MinE in this region . The sequence of Min protein patterns in vivo upon changing the system length can be intuitively understood from the mechanism underlying traveling waves in vitro . To this end , we introduce the diffusion length , which is the length a molecule typically diffuses before attaching to the membrane . For a diffusion constant D and an attachment rate ω it is given by . Now , consider a wave in a cell propagating in the direction of the long axis . The wave is sustained by molecules binding to the wave's leading edge after they have been released from the trailing edge . When the wave reaches a pole , the MinD dimers released from the membrane at the trailing edge can no longer bind at its leading edge . Instead , they diffuse away form the cell pole . If the cell length is on the order of , the proteins will preferentially bind at the opposite pole [55] , see Figure 9A . Similarly , with some delay , MinE released from the original wave , will bind at this pole , too , and a new wave traveling in the same direction as the original one is generated . If the system size is shorter , MinD binding will occur in a zone extending further from the new pole to the cell center because the ratio of diffusion length to the cell length has increased . As the affinity for MinE binding to MinD on the membrane is large , MinE will preferentially bind to the part of the MinD zone proximal to the cell center and the wave will move into the opposite direction compared to the original wave , see Figure 9B , thus giving rise to pole-to-pole oscillations . For even shorter cells , the distribution of cytosolic MinD and MinE is essentially homogenous as the diffusion lengths significantly exceeds the cell length . MinD and also MinE thus bind preferentially to zones of the highest MinD concentrations on the membrane and a stationary profile emerges , see Figure 9C . The picture presented here is thus somewhat different from the mechanism underlying the pole-to-pole oscillations proposed in Ref . [15] as we discuss below . Let us finally note , that it is harder to get an intuitive picture of the dependence of the Min-protein patterns on the total protein concentration and we refrain here from discussing this topic further .
In this work , we presented a computational study of self-organized pattern formation by MinD and MinE from E . coli . The equations , which notably account for membrane-binding of MinE , generate the patterns previously observed in living cells as well as Min protein waves on flat surfaces observed in reconstitution experiments . In addition , our analysis yielded two patterns that had not been reported before: In sufficiently long cells and for elevated protein levels , traveling waves emanating from one cell pole and propagating to the opposite pole should emerge . Secondly , in aberrantly large cells , the rotational symmetry of the pattern should be lost and a moving spot should form instead . Both predictions were confirmed experimentally . We conclude that the membrane-binding of MinE is an essential molecular feature to comprehensively describe large-scale pattern formation of the Min proteins . In vitro experiments on micropatterned membranes suggest an important role of MinE processivity for Min-protein pattern formation [26] , but it remains to be seen whether this is the case in vivo . In fact , comparing our system to the one proposed by Huang et al . [15] shows that MinE processivity can at least in part be replaced by a high rate of MinE binding to membrane-bound MinD ( they chose a rate orders of magnitude higher than we did ) . This leads to a different mechanism underlying the pole-to-pole oscillations and requires a finite MinD-ADP to MinD-ATP exchange rate for stabilizing standing waves with several nodes . It will be interesting to test experimentally which of the two possibilities is realized in living E . coli . Our description neglects many molecular details . For example , we did not consider explicitly a MinD dimerization step or the finite exchange rate of ADP for ATP for cytosolic MinD . Also , different expressions accounting for the binding of cytosolic MinE to membrane-bound MinD might be used . We analyzed several different expressions describing the effect that a single MinE dimer can induce detachment of several MinD dimers from the membrane . While these modifications led to quantitative differences , their analysis also revealed that details of the corresponding expressions are rather unimportant for the overall behavior of the system . As a consequence of the relatively simple reaction terms employed in our description , our model reveals some quantitative discrepancies compared to experimental observations . For example , the fluctuations present in the kymographs in Figure 3A and B are apparently larger than in the experimental kymographs in Figure 1 . In addition , the wave profile shown in Figure 8 differs from the experimentally determined [22] . However , complete quantitative agreement likely requires knowledge of more molecular details of the reactions involved . Note , however , that a quantitative comparison on the single cell level also requires precise measurements of the corresponding amount of MinD and MinE , which are currently not available . On a coarser level , though , our description seems to match the topology of the phase space . That is , we present one set of parameters , that correctly reproduces the sequence of patterns as cells grow and also correctly describes the appearance of stochastic switching and traveling waves in living cells with increasing protein levels . In contrast , the exact transition points differ in general from those observed in experiment and any coincidence would be fortuitous . Let us also emphasize that , experiments are now very much needed to constrain possible parameter values . Only with such data we can expect to make further significant progress in understanding Min protein patterns . In agreement with previous work , our analysis also showed that molecular noise has only a minor effect on the Min-protein patterns . Macroscopic signatures of molecular noise were only found under special conditions , namely , in short cells presenting stochastic switching and in large cells , where the Min proteins formed a rotating patch with a stochastically switching sense of rotation . Our description of the Min-protein dynamics can now be used to to design new experiments , for example , to test the interplay between the Min oscillations and Z-ring assembly in vivo or to determine conditions to generate Min-protein patterns inside vesicles in vitro . Such experiments could present important steps on the way to synthesize a system that is able to divide autonomously , that is , a minimal synthetic cell .
We used cells of the E . coli strain JS964 containing the plasmid pAM238 encoding for MinE and GFP-MinD under the control of the lac-Promoter [5] . Bacteria were grown overnight in a 3ml LB medium at 37°C . Cells were induced with Isopropyl – β – D-thiogalactopyranosid ( IPTG ) at a concentration of and incubated for 3–4 hours prior to measurements . During 1–2 hours prior to measurement , cells were kept at 30°C for better fluorescence . The optical density was less than 0 . 6 . During measurements , cells were in the exponential growth phase . The samples were kept at a temperature of 30°C using a Bachhoffer chamber . To keep bacteria from moving under the cover slip , we put them on an agar pad ( 1% agar solution in LB medium with a reduced yeast extract fraction , 10% , in order to lower background fluorescence ) . The fluorescence recordings were taken with an Olympus FV 1000 confocal microscope , at an excitation wavelength of 488 nm from a helium laser at low power . We used an Olympus UPLSAPO 60× , NA 1 . 35 oil immersion objective and recorded a frame every 3s . A measurement lasted 40min . During this period , the focus was manually readjusted at irregular intervals . A22 ( S- ( 3 , 4-Dichlorobenzyl ) isothiourea , HCl ) was purchased from Merck Millipore . Cells were imaged 2–3 hours after adding of A22 . We solve the dynamic equations ( 8 ) – ( 14 ) in the in vitro as well as in the in vivo geometry by using Comsol Multiphysics 4 . 1 which is a solver for partial differential equations based on the finite element method ( FEM ) . All computations with exception of those for Figure 4C were performed in 3d and no assumption was made about the symmetries of the solutions . For the calculations for the patterns in a bacterial geometry the maximal grid size was . For the calculations in the in vitro geometry , we used a maximal grid size of in the surface domain and of in the buffer domain . As initial condition we used homogenous distributions of cytosolic proteins with a random perturbation of 5–10% . The initial surface densities were chosen to be zero for the in vivo geometries . For the in vitro geometry the surface densities were different from zero in a semi-annulus to rapidly induce a spiral . The calculations for the growing cell presented in Figure 4C , the system length was increased by adding discrete pieces at one end of the interval . For the pattern shown in Figure 4 , the rate of growth was . On the added pieces , the protein densities of cytosolic MinD and MinE were initialized with the values and , respectively , whereas the densities of membrane-bound proteins were initially set to zero . To simulate the stochastic reaction diffusion kinetics ( 1 ) – ( 7 ) in three dimensions , we used MesoRD [56] , a tool to solve the stochastic Master Equation using a reaction diffusion Master Equation . It is based on the Next Subvolume Method [57] . | Cellular protein structures have long been suggested to form by protein self-organization . A particularly clear example is provided by the proteins MinC , MinD , and MinE selecting the center as site of cell division in the rod-shaped bacterium Escherichia coli . Based on binding of MinD to the cytoplasmic membrane and an antagonistic action of MinE , which induces the release of MinD into the cytoplasm , these proteins oscillate from pole to pole , where they inhibit cell division . Supporting the idea of self-organization being the cause of the Min oscillations , purified Min proteins were found to spontaneously form traveling waves on supported lipid bilayers . A comprehensive understanding of the Min patterns formed under various conditions remains elusive . We have performed a computational analysis of Min-protein dynamics taking into account the recently discovered persistent action of MinE . We show that this property allows to reproduce all observed Min-protein patterns in a unified framework . Furthermore , our analysis predicts new structures , which we observed experimentally . Our study highlights that mechanisms underlying the spontaneous formation of protein patterns under purified in vitro conditions can also generate patterns inside complex intracellular environments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | Membrane Binding of MinE Allows for a Comprehensive Description of Min-Protein Pattern Formation |
In bacterial genomes , gene order is not random . This is most evident when looking at operons , these often encoding enzymes involved in the same metabolic pathway or proteins from the same complex . Is gene order within operons nonrandom , however , and if so why ? We examine this issue using metabolic operons as a case study . Using the metabolic network of Escherichia coli , we define the temporal order of reactions . We find a pronounced trend for genes to appear in operons in the same order as they are needed in metabolism ( colinearity ) . This is paradoxical as , at steady state , enzymes abundance should be independent of order within the operon . We consider three extensions of the steady-state model that could potentially account for colinearity: ( 1 ) increased productivity associated with higher expression levels of the most 5′ genes , ( 2 ) a faster metabolic processing immediately after up-regulation , and ( 3 ) metabolic stalling owing to stochastic protein loss . We establish the validity of these hypotheses by employing deterministic and stochastic models of enzyme kinetics . The stochastic stalling hypothesis correctly and uniquely predicts that colinearity is more pronounced both for lowly expressed operons and for genes that are not physically adjacent . The alternative models fail to find any support . These results support the view that stochasticity is a pervasive problem to a cell and that gene order evolution can be driven by the selective consequences of fluctuations in protein levels .
It is well established that the chromosomal distribution of genes is not random , and in many genomes , genes that need to be coexpressed tend to cluster [1]–[4] . The coexpression of adjacent genes can be enforced by the action of bidirectional promoters [5] , simultaneous opening and closing of chromatin [6] , [7] , transcriptional spill-over [8] , or by inclusion within the same operon [9] , [10] . Given that prokaryotic operons generally contain functionally related genes that need to be expressed together [3] , gene order evolution in bacteria is often considered to be driven by coexpression ( although other scenarios have also been proposed to explain the origin of operons [11] , [12] ) . Such coexpression models do not obviously predict that within an operon there need be selection on gene order . However , a relationship between bacterial morphology and the relative order of genes in a cluster involved in cell division [13] provides some evidence for adaptive gene organization within an operon . Furthermore , a prior comparative genomics study found that horizontal transfer of operonic genes often involves in situ gene displacement by an ortholog from a distant organism without change of the local gene organization [14] , hinting at the presence of selection on intraoperonic gene order per se . Nevertheless , it remains unclear whether these phenomena are restricted to certain gene clusters only or whether they could be a more general property of bacterial operons , and most importantly , what selective forces might be responsible for these genomic patterns . Here , then , we ask whether gene order within operons is under selection and if so why ? In particular , we investigate whether gene order within metabolic operons of E . coli reflect the functional order of the encoded enzymes , i . e . , colinearity ( Figure 1 ) .
To examine whether gene order within metabolic operons reflect the functional order of the encoded enzymes , we focused on E . coli , where high-quality and high-coverage data are available on both biochemical pathways and operon structures . We compiled data on operons encoding at least two enzymes in the same biochemical pathway according to EcoCyc [15] , resulting in a list of 70 operons and 321 intraoperonic gene pairs ( Methods ) . For each intraoperonic enzymatic gene pair , we recorded whether their relative position in the operon corresponds to their functional order and hence displays colinearity . Approximately 60% of the 321 gene pairs showed colinearity , compared to 50% expected if intraoperonic gene order was random ( p = 0 . 0011 , from randomisation , see Methods ) . At first sight , the above result is unexpected as gene order should not affect the steady-state pathway productivity under the most simplistic scenario ( hypothesis 0 , see below ) . To confirm this , we built general mathematical models of operon expression and a linear metabolic pathway with four enzymes ( E1…E4 ) based on previous studies [16] , [17] . We used realistic model parameters ( see Methods and Table S1 ) and assigned identical enzyme kinetic parameters and a standard Michaelis-Menten rate law to all four enzymes . Thus , the metabolite concentrations are expressed as follows ( Equation 1 ) for the first three products ( i = 1 . . . 3 ) : ( 1 ) Where kcat is the enzyme turnover number , D is the dilution rate ( i . e . , growth rate of the cell ) , and Km is the Michaelis constant ( see Table S1 for values ) . Concentration of the first substrate ( S0 ) was fixed at 1 mM , and the initial concentrations of the other metabolites were set to zero . Metabolic pathway productivity was defined as the amount of end product synthesized during a given time period after operon induction . End product ( S4 ) formation is given by Equation 2 ( dilution of the end product is not considered as we are interested only in the total amount of S4 synthesized ) : ( 2 ) Operon expression was modelled following the “read-through” operon model of Swain [17] , in which ribosomes move directly from one gene to the next , hence translation events are completely correlated across intraoperonic genes . The formalization includes transcription initiation ( RNA polymerase binding and isomerisation ) , transcription elongation , mRNA degradation and dilution , ribosome binding ( ribosomes are bound to the first cistron , hence translation is read-through ) , translation , and protein degradation and dilution ( see Figure 2 ) . The rate of translation was fine tuned to achieve a delay between the appearances of consecutive gene products ( Ei ) that reflects empirically observed values , i . e . , 60 s ( see [18] , [19] ) . See Table S1 for model parameters . Our simulations confirm that at steady state , flux through the pathway is independent of gene order ( Table S2 ) . However , the excess of colinear metabolic operons indicates that the above model is missing something . We now consider three hypotheses , one old and two new , that have the potential to explain colinearity . In brief , the hypotheses suppose that colinearity is favoured because ( 1 ) it increases productivity associated with higher expression levels of the genes 5′ in operons [20] , ( 2 ) it provides a transient advantage immediately after up-regulation , and ( 3 ) it minimizes metabolic stalling owing to stochastic protein loss . The first hypothesis [20] proposes that increased productivity of a colinear arrangement could be attributed to a monotonically decreasing mRNA abundance profile along the operon , so-called polarity [21] . That a higher expression level of the first enzyme in the pathway might increase product yield has been experimentally verified in an engineered operon [20] . It has also been shown theoretically that , when the total amount of enzymes within the pathway is fixed ( i . e . , there is an upper limit of enzyme concentrations ) , maximal steady-state flux through unbranched pathways can be obtained by a monotonic decrease of enzyme concentrations along the path under some circumstances [22] . In a linear pathway where equilibrium constants of reactions are larger than unity and all enzymes have the same catalytic efficiency , mathematical models predict a decrease of flux control coefficients from the upper end to the lower end of the chain , and therefore , an accumulation of enzyme concentrations at the upper end of the pathway when flux is maximized [22] . If so , and if gene expression levels are not uniformly distributed within operons , then colinearity might confer an advantage by increasing steady-state flux . Indeed , our mathematical model of operon expression coupled with a chain of irreversible enzymatic steps confirms this expectation ( Table S3 ) when the effect of polarity is included in the model ( i . e . , by introducing degradation of ribosome-bound mRNA intermediates; see Table S3 legend for details ) . We should then observe higher colinearity in operons with an mRNA abundance profile decreasing from 5′ to 3′ . We extend our formalization to identify two further conditions under which colinearity will be under selection ( hypotheses 2 and 3 ) . One possibility is that colinearity presents a transient advantage after up-regulation prior to steady state . Consider a simple pathway with two enzymes , A and B . If the order in the operon is AB , then enzyme A can start processing its substrate while B is being synthesized . Conversely , if the order is BA , metabolism cannot start until translation of the second gene is finished , thus slowing the processing of a new metabolite . Our simulations support this possibility ( Figure 3 and Table S4 ) . Indeed , a comparison of different simulated intraoperonic gene orders showed that colinearity can increase pathway productivity by up to 8 . 49% within one cell-generation time following operon induction ( comparing the most colinear , ABCD , and least colinear , DCBA , arrangements ) owing to the temporal delay between the appearances of consecutive gene products [18] , [19] , which modulates substrate turnover when total enzyme amount is limited [16] , [23] . Moreover , we found that , on average , the effect of swapping the position of two intraoperonic genes depends on their chromosomal distance: swapping adjacent genes had the smallest effect on pathway productivity ( see Figure S1 ) . The above deterministic simulations fail to capture the fact that small numbers of molecules are frequently involved in the process of gene expression and could lead to significant stochasticity in protein abundance [24] . Whereas enzymes encoded in a highly expressed operon are likely to be always present in the cell whenever the operon is induced , stochasticity might play an important role in weakly expressed operons as enzymes could either decay or be diluted by cell division between two expression episodes [25] , hence recurrently stalling metabolism . Colinearity could minimize the effect of such stochastic enzyme losses by speeding up the reinitiation of stalled metabolic transformations , in a similar manner as it provides a transient advantage after up-regulation of an inactive pathway ( see hypothesis 2 ) . To formally examine this verbal argument , we also simulated stochastically our model to explore how gene order affects pathway productivity as a function of expression level . Different expression levels were simulated by varying the rate of RNA polymerase dissociation from DNA ( see Table S1 ) . First , we observe that whereas enzyme molecule numbers fluctuate at both low and high expression levels , enzyme levels frequently drop to zero only when the rate of expression is low ( Figure 4A ) . Importantly , typically all four enzymes are lost between expression bursts at very low transcription rates , which results in the complete stalling of the pathway . Second , we simulated the behaviour of two linear pathways encoded by the same operon , one of which was colinear with the operonic gene order and the other was anti-colinear ( thus the metabolic productivity of the two arrangements were directly comparable despite the stochastic nature of the simulations ) . Pathway performance was assessed after 50 cell-generation times following operon induction . Our analysis showed ( Figure 4B ) that whereas colinearity in a very lowly expressed operon ( mean±standard deviation [SD] protein copy number per cell = 2 . 4±6 . 1 ) can increase pathway productivity by 4 . 65% , this figure drops to 0 . 1% for a highly expressed operon ( 3 , 959 . 4±232 protein copies per cell ) . The effect can be attributed to stochasticity as the advantage of colinearity diminishes when low-expression simulations are run deterministically ( i . e . , average protein levels are controlled for ) , and closely resembles the high-expression stochastic simulation scenario with a minute 0 . 07% increase in pathway performance . Moreover , at low expression levels , even gene orders with an intermediate level of colinearity provide a clear advantage when compared to an anti-colinear arrangement ( 2 . 42% , on average ) . Thus , based on the above simulations , which appear robust against parameter variations ( see Table S5 ) , we expect that colinearity should primarily be a property of lowly expressed operons and of nonadjacent intraoperonic gene pairs . This is an unusual prediction , as more classically , the strength of selection is considered to be greater on more highly expressed genes because these present more opportunity for selection . Such a logic explains , for example , why highly expressed genes evolve slowly [26] , [27] . We can test the three theoretically viable hypotheses by reference to the data on which operons show colinearity . To test hypothesis 1 , we gathered a set of microarray expression data on wild-type E . coli grown on glucose minimal medium under aerobic and anaerobic conditions ( Methods ) . First , we asked whether operons in general display a decreasing level of mRNA abundances from the 5′ to 3′ end , as observed in the engineered zeaxanthin biosynthesis operon [20] and in the native lactose operon [21] . Indeed , we found an excess of intraoperonic gene pairs in which the gene located closer to the transcription start site has a higher mRNA abundance compared to those located more downstream ( p<10−6; Methods ) , although we still see many individual operons in which such a trend cannot be detected . This pattern also holds when only metabolic operons are investigated ( p<0 . 03 ) . From the above hypothesis , we would expect colinear arrangement only in those operons in which the 5′ genes have higher transcript levels than those located downstream . We hence asked whether the degree of colinearity differs between operons with and without a significantly decreasing mRNA abundance profiles . Using linear trend analysis [28] , we identified 26 and 23 operons showing a significantly decreasing mRNA abundance profile under aerobic and anaerobic conditions , respectively ( see Methods ) . Contradicting the prediction of the hypothesis , however , we failed to find an increased colinearity in these operons , neither using the aerobic ( p = 0 . 97 ) nor the anaerobic ( p = 0 . 36 ) expression dataset . To test the prediction of the stochastic stalling hypothesis that colinearity should be found predominantly in lowly expressed operons , we split the set of metabolic operons into two groups based on their mRNA expression levels: operons with average or higher log expression levels and operons with lower than average log expression levels . As predicted , we found ( Figure 5 ) that highly expressed operons have significantly lower degrees of colinearity than those expressed at lower levels ( mRNA measured under aerobic condition: p = 0 . 0011 , degrees of colinearity: 44 . 8% vs . 71 . 6% , anaerobic condition: p = 0 . 0006 , degrees of colinearity: 45% vs . 70 . 8% , ; see Methods ) . Indeed , for the highly expressed operons , there is no significant deviation from null ( Figure 5A ) . It should be noted that in the absence of genome-wide data on protein copy number fluctuations in E . coli , we used a population-averaged mRNA level as an inverse proxy for stochasticity in protein concentrations , an assumption that holds in yeast [29] and mammals ( L . D . Hurst , unpublished data ) . The fact that colinearity is most profoundly seen in low-abundance operons might also be compatible with hypothesis 1 and 2 , hence , independent of gene expression noise . First , the advantage of colinearity in the presence of polarity ( hypothesis 1 ) might be more profound in lowly expressed operons . However , simulating the model without stochastic effects predicts the opposite: polarity provides more advantage to colinearity when the expression level is high ( Table S3 ) . Furthermore , as explained above , we failed to find any empirical evidence for higher colinearity in operons with significant polarity effects . Second , in hypothesis 2 , the variation in expression intensity across conditions is what drives colinearity ( i . e . , it provides a transient advantage after up-regulation following an environmental shift ) . Operons that are highly expressed on average might also likely be constitutive in the expression , hence , potentially explaining a connection with abundance . If what matters is the rapid processing of metabolism on up-regulation of the operon , then variation per se may be what matters rather than mean dose . To examine this issue , we used an index of relative variability of mRNA levels measured under 213 conditions as a proxy for gene expression variability across different environments ( see Methods ) . In contrast to expectations , we find that the gene expression level of operons correlates positively , albeit weakly , with their expression variability ( r = 0 . 347 , p = 0 . 004 and r = 0 . 267 , p = 0 . 0268 under aerobic and anaerobic conditions , respectively ) . Moreover , no significantly increased colinearity can be detected in operons with high expression variability when controlling for expression level , and if there is any trend , it is in the opposite direction ( p = 0 . 057 and p = 0 . 128; see Methods ) . In contrast , the effect of mRNA abundance on colinearity remains significant when expression variability is controlled for ( p = 0 . 01 and p = 0 . 016 under aerobic and anaerobic conditions , respectively ) . We conclude that variation in the environmental specificity of operon expression cannot explain the higher incidence of colinearity in lowly expressed operons . The above mathematical analyses also predict that the impact of gene order rearrangement on metabolic pathway productivity should be most pronounced when the position of genes located distantly within the operon is interchanged ( Figure S1 ) . Therefore , one would expect to see more colinearity for distantly located gene pairs compared to those located adjacent in lowly expressed operons . We observe that in support of this expectation , in lowly expressed operons , gene pairs separated by a physical distance of at least one gene length show higher colinearity than those located closer ( Fisher exact test , p<0 . 005; the median gene length is 1 , 070 bp in our dataset ) . Thus , colinearity is more pronounced for distant enzymatic genes in the operon . All of the above tests presume that if selection favours a given gene order , that order should match the metabolic order . But is it necessarily the case that colinearity is always optimal for metabolic operons ? The presence of within-pathway regulatory interactions ( i . e . , when one enzyme is regulated allosterically or competitively by a product of another enzyme in the same pathway ) might impose additional requirements on gene order . If such regulation was more common for abundantly expressed operons , this could explain why colinearity is more common in lowly expressed operons . More specifically , it has been proposed that spatial colocalization of enzymes interacting via small molecule metabolites might enable faster feedback regulation and could be achieved by a closer physical proximity of the enzyme-coding genes within the operon [30] . Thus , there might be selection to place genes of interacting enzymes close to each other in the operon even if they are not colinear with the metabolic pathway . To test this possibility , we collected data on metabolite-level within-pathway enzymatic interactions from EcoCyc [15] and from a published dataset [31] based on the BRENDA database [32] ( Methods ) . In contrast to the above prediction , we found that the observed average gene distance between interacting enzyme pairs was not significantly different from that expected by chance ( p = 0 . 234 , n = 19 gene pairs ) , suggesting an absence of clustering of metabolically interacting genes in operons . To further investigate whether the presence of intraoperonic regulatory interactions has an effect on the extent of colinearity , we compared the degree of colinearity in operons with known regulatory interactions ( 11 operons ) to the rest of the dataset ( 59 operons ) . A randomisation test showed that the degree of colinearity is not lower in the set in which intraoperon regulations have been reported ( p = 0 . 089; see Protocol S1 ) . A similar result was obtained when we controlled for expression-level differences between the two groups ( p = 0 . 35 , for both aerobic and anaerobic conditions; see Protocol S1 ) . Thus , we failed to find evidence in support of the idea that within-pathway intraoperon metabolic regulation has an influence on gene order and might interfere with colinearity .
To our knowledge , the present study on E . coli metabolic operons provides the first systematic evidence that intraoperonic gene order is not random , but rather correlates with the functional order of the encoded enzymes . This is true , however , exclusively for lowly expressed operons , an otherwise curious result given that we usually expect selection to be strongest on highly expressed genes . Our analyses did not find support to the ideas ( 1 ) that colinear gene arrangement might be an adaptation to enable high steady-state pathway flux as a result of a decreasing mRNA abundance along the operon [20] , or ( 2 ) that colinearity presents a transient advantage following up-regulation of the operon in a changed environment . In contrast , the evidence supports the hypothesis that colinearity minimizes stochastic stalling of metabolism at low expression levels: constitutively , but lowly expressed operons are under stronger selection for optimal order as gene expression occurs in random episodes [25] , and enzymes from prior expression events might have been lost by decay or by cell division , potentially stalling metabolism . Colinear organization of operonic gene order could minimize any such stalling . This result underscores both the importance of stochastic events to cellular functioning and provides a further case history in which gene order appears to be an adaptation to ensure resilience to stochasticity [33] . An issue that we have not addressed is the potential role of horizontal transfer in establishing colinearity . According to the selfish operon hypothesis [11] , novel metabolic functions can be gained by horizontal transfer that moves sets of genes in unison . This hypothesis , the validity of which is questionable as an explanation for which genes reside in operons [4] , [10] , does not , as far as we can tell , have any specific prediction as to whether an operon should be colinear or not . Consider a pathway with enzymatics steps A → B → C . If all three genes are needed for successful horizontal transfer , then it should not matter whether the order is ABC , CBA , or any other variant . If transfer of two successive enzyme is selectively favourable ( e . g . , A and B , but not C ) , then transfer of AB from the operon ABC should be as viable as transfer of BA from operon BAC . Thus , there is no obvious reason why horizontal transfer of operons should impose any filter on colinearity . This accords with observation . Operons containing at least one gene gained by horizontal gene transfer ( 69% colinearity ) are not significantly more ordered than all other operons ( 53% colinearity ) , p = 0 . 13 . Conversely , operons with essential genes are unlikely to be gained by horizontal transfer according to the selfish operon hypothesis . As then expected , operons containing at least one essential gene ( 61 . 8% colinearity ) are not less ordered than the rest of operons ( 60% colinearity ) , p = 0 . 9 ( see Methods ) . The finding that operonic gene order could be an adaptation against noise in protein levels leaves at least one paradoxical problem: although we see evidence for selection on gene order in operons , gene order within operons shows especially high evolutionary conservation , indicating strong purifying selection on local gene organization [34] . Indeed , a comparison of E . coli metabolic operons with information on operon structure and orthology in Bacillus subtilis ( see Methods ) revealed that 70% of E . coli operons could not be matched to conserved B . subtilis operons , either due to the absence of orthologs , or due to the fact that the orthologs are no longer located in the same transcription unit . In another 22% of the operons , the relative order of the orthologs was completely conserved , and we detected only five operons ( 8% ) in which the relative position of orthologs had been rearranged . This freezing of gene order within operons may reflect the fact that intraoperonic inversions will place genes on the wrong strand . How can the two apparently contradictory findings be resolved ? We speculate that the selection is not on order within the operon per se , but rather selection on successful establishment of operons . Imagine that A and B reside next to each other and a mutation occurs that permits them to be coded in a polycistronic transcript , i . e . , operonization . If the operon is lowly expressed , our results suggest that selection for operonization will be stronger if the order is AB than if it is BA . Hence , a selective filter on operonization can explain the findings and be consistent with frozen operons after their establishment .
We considered a four-enzyme irreversible linear metabolic pathway coupled with operonic gene expression . The enzymes were assumed to operate according to standard Michaelis-Menten equations . All enzymes had the same velocities , turnover numbers , and Km values based on experimentally measured values of aspartate kinase I ( Table S1 ) . Cell generation time was 60 min , and all metabolites and enzymes were diluted accordingly ( D ) . Initially , all metabolites had zero concentrations except for the substrate of the first enzyme , which was fixed at 1 mM . Operonic gene expression was modelled following the read-through operon model by Swain [17]; see Figure 2 . Copasi version 4 . 4 . 28 was used to perform all simulations [35] . Stochastic simulations were carried out using a hybrid deterministic–stochastic simulation algorithm built into Copasi ( “Hybrid Runge-Kutta” ) to simulate gene expression and enzymatic reactions within one model ( default parameter values were used with the exception of Runge-Kutta step size , which was set to 0 . 1 ) . Operonic gene order and metabolic pathways were extracted from EcoCyc [15] v10 . 5 . We compiled a list of metabolic gene sets , in which each set consisted of genes belonging to the same pathway and encoded in the same operon . To enable the quantification of the extent of colinearity , we generated a list of nonredundant intraoperonic gene pairs with unambiguous metabolic pathway order ( cyclic pathways were excluded ) , resulting in 321 gene pairs from 70 operons and 73 pathways ( see Protocol S2 ) . Data on B . subtilis operon structures were obtained from BioCyc [36] and DBTBS [37] ( where information on a large number of experimentally characterized B . subtilis operons is available ) . Cross-species comparison of orthologs and chromosomal positions was performed using EcoCyc [15] . Colinearity of a set of operons was measured as the ratio of the number of colinear pairs to the total number of gene pairs across all investigated operons . A gene pair was considered colinear if the gene located closer to the 5′ end of the operon encodes an enzyme operating earlier in the same pathway compared to the downstream gene . To assess the statistical significance of colinearity in our dataset , we compared the observed level of colinearity to a distribution of colinearity values generated by randomizing gene order within each operon 100 , 000 times ( p-values were calculated by p = ( R+1 ) / ( N+1 ) , where R is the number of cases when the randomized level of colinearity is equal or greater than the observed level , and N is the number of randomizations ) . Affymetrix microarray gene expression data were obtained from ref [38] . We used log2-transformed normalized expression profiles of wild-type K-12 MG1655 strain grown on M9 glucose medium under aerobic and anaerobic conditions [38] . For each gene , we calculated the average expression value based on three ( aerobic ) and four ( anaerobic ) data points . Transcript abundance level of operons was defined as the mean of expression values of the constituent genes . To examine whether operons display a decreasing level of mRNA abundances from 5′ to 3′ end , we compiled a nonoverlapping set of gene strings from the whole E . coli genome , in which genes associated with a given string are always transcribed together . This resulted in 386 gene strings ( 2 , 199 within-string pairs ) containing at least two genes with expression data . We counted the number of cases where the 5′ member of a gene pair has a higher expression level than the downstream gene ( aerobic dataset: 1 , 274 pairs , anaerobic: 1 , 293 pairs ) and compared those cases to the values obtained by randomizing the positions of within-string genes ( p<10−6 for both conditions ) . A similar analysis was performed on our filtered set of metabolic operons ( 270 gene pairs in 65 operons; p = 0 . 0295 and p = 0 . 0224 for aerobic and anaerobic conditions , respectively ) . We also determined for each operon whether its mRNA abundance profile shows a significant monotonic decrease from 5′ to 3′ end . The presence of a monotonically changing abundance was tested by linear trend analysis [28] ( using gmodels R package ) . The direction of change was inferred from the Spearman rank correlation . After Bonferroni correction , we found 26 and 23 operons showing a significantly decreasing mRNA expression profile under aerobic and anaerobic conditions , respectively . To quantify gene expression variability across environmental conditions , we used compiled expression data for 213 conditions [39] . We calculated the SD of the log2-transformed expression values for each gene , which is invariant under a multiplicative change [40] . Expression variability of operons was defined as the mean of SD values of the constituent genes . To examine whether expression variability of operons correlates with the degree of colinearity when controlling for mRNA abundance levels ( as measured under aerobic and anaerobic glucose conditions ) , we used residuals from the linear regression of SD on expression level to classify operons into groups with higher or lower than average gene expression variability . Randomisation was employed to test whether the colinearity of these two groups were different . A similar procedure was followed to examine whether mRNA abundance of operons is associated with colinearity when expression variability is controlled for . Genes that have undergone horizontal transfer into the E . coli lineage since its split from the Vibrio lineage were previously identified using parsimony analysis of gene presence and absence data [41] . Data on gene essentiality were obtained from a recent functional genomic study in E . coli K12 , in which a systematic collection of in-frame , single-gene deletion mutants was constructed [42] . | In bacteria , different enzymes from the same metabolic pathway are often encoded within one transcriptional unit , an operon . There is also , we show , a tendency for the enzymes that are needed earlier in the pathway to feature earlier in the operon , so-called colinearity . Why might this be ? We test three ideas , one old and two new . The prior suggestion supposes that proteins of genes early in operons will be at a higher dose . Although some operons are like this , in general , we see no relationship of protein dose with colinearity . We also find no evidence that operons that frequently need up-regulation are any more likely to be colinear . A third model is , however , supported . If an operon is rarely expressed , then all the proteins for this part of metabolism can be lost by chance . Rebooting such metabolism is fastest if the operon is colinear . This model predicts , correctly , that colinearity should be more frequent in operons that are expressed at a low level . This result is important for at least two reasons . First , it supports the view that chance events ( such as protein loss ) within cells are important on a day-to-day basis . Second , it challenges the supposition that natural selection will be weakest on lowly expressed genes . Where chance events are concerned , natural selection can be strong on genes expressed at a low level . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics",
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/chromosome",
"biology",
"evolutionary",
"biology/genomics",
"evolutionary",
"biology/bioinformatics"
]
| 2009 | Stochasticity in Protein Levels Drives Colinearity of Gene Order in Metabolic Operons of Escherichia coli |
Voltage-gated ion channels are essential for electrical signaling in neurons and other excitable cells . Among them , voltage-gated sodium and calcium channels are four-domain proteins , and ion selectivity is strongly influenced by a ring of amino acids in the pore regions of these channels . Sodium channels contain a DEKA motif ( i . e . , amino acids D , E , K , and A at the pore positions of domains I , II , III , and IV , respectively ) , whereas voltage-gated calcium channels contain an EEEE motif ( i . e . , acidic residues , E , at all four positions ) . Recently , a novel family of ion channel proteins that contain an intermediate DEEA motif has been found in a variety of invertebrate species . However , the physiological role of this new family of ion channels in animal biology remains elusive . DSC1 in Drosophila melanogaster is a prototype of this new family of ion channels . In this study , we generated two DSC1 knockout lines using ends-out gene targeting via homologous recombination . DSC1 mutant flies exhibited impaired olfaction and a distinct jumpy phenotype that is intensified by heat shock and starvation . Electrophysiological analysis of the giant fiber system ( GFS ) , a well-defined central neural circuit , revealed that DSC1 mutants are altered in the activities of the GFS , including the ability of the GFS to follow repetitive stimulation ( i . e . , following ability ) and response to heat shock , starvation , and pyrethroid insecticides . These results reveal an important role of the DSC1 channel in modulating the stability of neural circuits , particularly under environmental stresses , likely by maintaining the sustainability of synaptic transmission .
Voltage-gated sodium channels are primarily responsible for the initiation and propagation of action potentials , playing a critical role in regulating neuronal excitability . They are members of a superfamily that also includes voltage-gated potassium channels and voltage gated calcium channel [1] . The sodium and calcium channels contain four homologous domains , whereas the potassium channels consist of tetramers of single-domain subunits . It is generally believed that sodium channels evolved from an ancient calcium channel [2] , [3] . Selectivity in sodium and calcium channels is strongly influenced by a ring of amino acids in the pore regions of the channels [4] . Sodium channels contain a DEKA motif ( i . e . , amino acids D , E , K , and A at the pore positions of domains I , II , III , and IV , respectively ) , whereas voltage-gated calcium channels contain an EEEE motif ( i . e . , acidic residues , E , at all four positions ) . Recently , a novel family of ion channel proteins that contains an intermediate DEEA motif in the pore regions has been found in a variety of invertebrate species [5] . DSC1 in Drosophila melanogaster and BSC1 in Blattella germanica , are prototypes of this new family of ion channels . We have previously shown that DSC1 and BSC1 are functionally and evolutionally intermediate between voltage-gated sodium and calcium channels and more permeable to Ca2+ even though the amino acid sequences are more closely related to sodium channels [6] , [7] . Interestingly , a recent study showed that a sodium channel-like channel containing a DEEA motif , NvNav2 . 1 , from the starlet sea anemone also conducts Ca2+ [5] . Therefore , it has been speculated that the DEEA motif could be a pore sequence from which sodium channels have evolved from calcium channels [2] , [5] . More strikingly , a recent phylogenetic analysis revealed the presence of the DEEA motif in a gene homologous to voltage-gated sodium channels in a single-celled choanoflagellate , indicating that evolution of sodium channels may have predated the origin of the nervous systems [2] . Intriguingly , this DEEA motif was retained in many other animal groups , such as ascidians , insects and cnidarians , but was lost in vertebrates [2] , [5] . The physiological role of this new family of sodium channel-like , but Ca2+-selective , ion channels in animal biology , however , remains mysterious . In this study , we investigate the role of the DSC1 channel in vivo . The DSC1 transcript and the DSC1 protein are found in a variety of tissues , such as brain , antennae , thorax , legs , and ovary [8] , [9] , suggesting a potentially broad role of DSC1 in insect biology . Previously , Anholt and colleagues conducted genetic and molecular analysis of a smell-impaired ( smi ) mutant , smi60E , which carried a P-element insertion in an intron of the DSC1 gene [10] . This insertion resulted in a 2-fold reduction in the steady-state level of the DSC1 transcript and the mutant had a slight reduction in olfactory response to benzaldehyde , implicating a role of the DSC1 channel in olfaction [10] . However , a complete knockout mutant of DSC1 has not been characterized , and it remains to be determined whether DSC1 has a broader role in modulating insect neurophysiology . We generated two DSC1 knockout lines and conducted a battery of behavioral analyses , followed by electrophysiological characterization of the giant fiber system ( GFS ) , a well-defined central neural circuit [11] , [12] , [13] , [14] . Our results not only confirm a role of the DSC1 channel in insect olfaction , but , unexpectedly , also show that the DSC1 channel has a unique role in regulating neuronal excitability , especially in extending the stability of neural circuits and behaviors under environmental stresses and insecticide exposure .
Although a P-element insertion has been found in an intron of the DSC1 gene in the smi60E mutant , which has partially reduced expression of the DSC1 transcript level and decreased olfactory response to benzaldehyde [10] , complete knockout mutants of the DSC1 gene would be ideal for assessing the full extent by which the DSC1 gene contributes to insect biology . We successfully used the method of ends-out targeted gene knockout via homologous recombination [15] to isolate two independent DSC1 knockout lines . The procedures for generating DSC1 knockout lines are summarized in Figure 1A . Twelve independent transgenic lines were mapped to the chromosome 2 ( where the DSC1 gene resides ) . To detect the targeting events in DSC1 knockout flies by PCR , two pairs of primers were used . Primer a corresponds to the inserted white gene , and primers b and c to the DSC1 genomic DNA sequences inside and outside of the donor construct ( Figure 1B ) . As expected , there was no detectable PCR product amplified from control flies using primer pairs a/b or a/c . A PCR product with the predicted length ( 3 . 5 kb ) was amplified from DSC1 transgenic donor flies using primer pair a/b but not a/c . On the other hand , predicted PCR products of 3 . 5 kb and 3 . 6 kb were amplified from two targeted transgenic lines using primers a/b and a/c , respectively ( Figure 1B ) . These two lines are named DSC1a and DSC1b . Southern blot analysis was also performed to confirm DSC1 knock-out . DNA from homozygous flies was digested with restriction enzyme KpnI and hybridized with a probe made from the DSC1 fragments ( Figure 1C ) . As predicted , a 6 . 9 kb band in the wild-type and two bands of 3 . 3 kb and 8 . 1 kb in DSC1a and DSC1b flies were detected ( Figure 1C ) . The successful construction of two independent DSC1 knockout lines provided us with a critical foundation for characterizing the role of the DSC1 channel in insect biology . Previous research has shown a broad distribution of the DSC1 transcript and the DSC1 protein in different tissues , such as brain , antennae , thorax , legs , and ovary [8] , [9] , yet , the DSC1 knockout flies are viable under the standard laboratory rearing condition and exhibit no morphological and developmental abnormalities and have a normal lifespan and weight , which enabled us to perform a variety of behavioral tests to assess in vivo function of the DSC1 channel in D . melanogaster . Because the DSC1 channel was implicated in olfaction in a previous study [10] , we first conducted an olfaction behavioral assay . As expected , the defect in olfaction was more severe in the DSC1a and DSC1b flies than that previously observed in the smi60E mutant ( Figure S1A and S1B ) . Next , we performed experiments to determine possible locomotion defects in the DSC1 knockout mutants . Climbing assays were carried out as described in Protocol S1 . After 30 sec , more than 80% of DSC1a and DSC1b flies were able to reach or pass the 10 cm bar and climbed into the second vial , with no significant difference from that of the w1118 control flies ( Figure S1C ) . However , DSC1a and DSC1b flies had a stronger tendency to jump or fly when disturbed by a gentle tap on the vial ( see Video S1 ) . The jumpy phenotype suggests a defect in the nervous system in the DSC1 knockout mutants since the DSC1 protein is not detected in muscles [8] . We noticed that this defect was especially evident during experiments to assess the response of the DSC1 knockout mutants to heat shock and starvation . Unlike other Drosophila ion channel mutants , DSC1 KO flies did not exhibit leg-shaking under ether anesthesia or temperature-sensitive paralysis . Both w1118 and mutant flies began to exhibit knockdown ( unable to walk or on their back ) 5 min into 40°C heat shock and all flies were knocked down at the end of 15 min heat shock ( Figure 2A ) . However , DSC1 knockout flies were significantly more jumpy ( frequently flying off from the wall or bottom of the vial ) than w1118 flies , particularly from the fifth to tenth minutes after heat shock ( Figure 2B ) . Furthermore , the recovery from heat shock was significantly delayed in DSC1 knockout flies when the vials were returned to room temperature ( Figure 2C ) . At the 10-min time point , 80% of w1118 flies and 40% of DSC1 knockout flies had resumed climbing . However , it took an additional 30 min for another 40% of DSC1 knockout flies to climb up again ( Figure 2C ) . Similarly , in experiments to determine the response of DSC1 knockout flies to starvation , DSC1a flies were significantly more jumpy than w1118 flies after starvation ( Figure 2D; see Video S2 ) . The DSC1 channel shares the highest sequence similarity with the Para sodium channel which is the primary target of pyrethroid insecticides and sodium channel blocker insecticides ( SCBI ) . We first attempted to determine whether the DSC1 channel is directly affected by these sodium channel-targeting insecticides by pharmacologically characterizing DSC1 channels expressed in Xenopus oocytes . However , DSC1 currents in oocytes exhibit significant rundown , which makes pharmacological analysis unfeasible . We therefore took an alternative approach to assess the involvement of DSC1 in possibly modulating pyrethroid sensitivity . Pyrethroids prolong the opening of sodium channels , causing increased neuronal excitability including repetitive firing and/or membrane depolarization ( depending on doses of pyrethroid , types of pyrethroids , and also nerve preparations ) [16] . Sustained action of pyrethroids eventually results in the complete blockage of signal transmission and lethality of poisoned insects [16] . We examined the susceptibility of DSC1 knockout flies to the lethal effect of pyrethroids . Flies from w1118 and both DSC1 knockout lines , DSC1a and DSC1b , were tested in a contact bioassay . We found that DSC1 knockout mutants were more susceptible than w1118 flies to all four pyrethroids tested , including two type I ( permethrin and bioresmethrin ) and two type II ( deltamethrin and fenvalerate ) pyrethroids ( Table 1 , Table S2 ) . Next , we examined whether the hyper-susceptibility of the DSC1 knockout mutant is specific to pyrethroids . We tested DCJW , an active metabolite of indoxacarb , which is a sodium channel blocker insecticide ( SCBI ) and has a mode of action opposite to pyrethroids ( i . e . , inhibiting neuronal excitability ) [17] . As shown in Table 1 , the susceptibilities of DSC1 knockout flies to DCJW were similar to that of w1118 flies . To determine whether the DSC1 channel may have effects on susceptibility beyond sodium channel-targeting insecticides , we tested the susceptibility of DSC1 flies to the insecticide fipronil , which causes neuronal hyperexcitability by blocking the GABA-gated Cl− channel [18] . As shown in Table 1 , w1118 and DSC1 knockout flies exhibited similar susceptibility to fipronil . These results suggest that DSC1 knockout flies are affected in response to neuronal stimulation by sodium channel activators . To identify potential defects in the nervous system of DSC1 knockout flies , we examined a well-defined adult neural circuit , the giant fiber system ( GFS; Figure 3A ) . The GFS mediates the jump-and-flight escape reflex in response to visual stimuli in Drosophila . The components of the GFS are depicted in Figure 3 , including the tergotrochanteral muscle ( TTM ) for jump and the dorsal longitudinal muscle ( DLM ) for flight . Briefly , the somas of giant fiber neurons are located in the brain with their large axons extending to the thorax , where the terminals of the giant fiber axon form synaptic connections with two different neurons: a large motorneuron that innervates the TTM and a peripherally synapsing interneuron ( PSI ) . The PSI axon crosses the midline and synapses with motor neurons , which innervate the DLM . The giant fiber pathway can be triggered from different sites ( the brain or thorax ) , recruited at different stimulation intensities . The responses can be recorded from DLM or TTM which represent two distinct branches: the GF-PSI-DLMn-DLM and GF-TTMn-TTM ( see Figure 3 legend for details ) . The time interval between the stimulus and the first muscle potential is termed the response latency . It reflects the chain of events from stimulation , initiation and conduction of action potentials , and synaptic transmission along the neuronal elements of the neural circuit to the innervated muscle . At higher stimulating voltages , which are sufficient to directly excite the giant fiber , muscle spikes can be recorded after a shorter time , defining a short-latency ( SL ) response ( Figure 3A ) . When lower stimulating voltages are used , the same set of muscle potentials appears but only after a longer delay because lower-intensity stimuli across the brain recruit the upstream synaptic activity afferent to the GF and thus trigger the muscle response with a longer delay ( Figure 3A ) . Thus , high and low voltage stimuli result in SL or long-latency ( LL ) responses , respectively . The refractory period ( RP ) is the characteristic time interval following the first stimulus during which a second stimulus fails to evoke another response . We found that the LL and SL responses from the DLM branch and their corresponding refractory periods were similar between undisturbed w1118 and DSC1 knockout flies ( Figure 3 and Figure S2 , Table S3 ) . However , the LL and SL responses and their refractory periods of flies under heat shock were significantly different . The LL response and the long-latency refractory period ( LLRP ) were decreased upon heat shock in both w1118 and DSC1 knockout flies ( Figure 3B and 3C , Table S3 ) . Notably , the reduction in LLRP was significantly more drastic in DSC1 knockout flies at 5 and 10 minutes after heat shock , coinciding with the time when the jumpy phenotype in DSC1 knockout mutants was most pronounced ( Figure 2B ) . At the end of the 15 min heat shock , the LLRP of w1118 flies also was drastically reduced , but no difference was observed between w1118 and DSC1 mutant flies , consistent with the eventual total paralysis of both w1118 and DSC1 knockout flies . In contrast to the LLRP results , w1118 and DSC1 knockout flies showed similar SL refractory periods ( SLRP ) ( Figure S2 ) . No difference in the latency of the LL response was observed between w1118 and DSC1 mutant flies in response to starvation ( Figure 3D , Table S4 ) . However , an apparently LLRP-specific defect was detected in DSC1 mutant flies when starved flies were assayed ( Figure 3E , Table S4 ) , consistent with the exaggerated jumpy phenotype of DSC1 knockout flies under starvation ( Figure 2D ) . Because the recovery from heat shock was delayed for the DSC1 knockout flies compared to w1118 flies ( Figure 2C ) , we examined the changes in GFS parameters during the recovery at 10 , 20 , 30 and 40 min . The reduced LL was recovered in both w1118 and DSC1 knockout flies within 10 minutes ( Figure 3F , Table S5 ) . Again , we detected a significant difference in the LLRP between w1118 and the DSC1 knockout flies ( Figure 3G ) . Whereas the LLRP was fully recovered in w1118 flies at 10 min , it was still significantly shorter at the end of 40 min for DSC1 knockout flies ( Figure 3G , Table S5 ) . The GFS is extremely robust to high frequency stimulation [11] , [19] . The greater reduction in the LLRP of DSC1 knockout flies in response to heat shock and starvation suggests these treatments destabilize the function of the GFS . Thus , we applied different stimulus protocols to localize the weaker links along the GF pathway as revealed by altered responses to repetitive stimulation . We challenged the GF-PSI-DLM and GF-TTM branches of the GFS with 5 trains of 30 stimuli of different frequencies , up to 200 Hz ( Figure 4A ) and recorded the SL responses from the DLM and TTM ( Figure 4B ) . Notably , unlike DLM , it is evident that the response from the TTM was significantly impaired in DSC1 knockout flies as compared to w1118 flies ( Figure 4B ) . Besides the drop in the overall mean frequency response , the frequency response of the GF-TTM branch in DSC1 knockout flies is also highly variable , which could undermine the stability or reliability of the motor output . The response in the DLM branch upon brain stimulation , however , was not different between w1118 and DSC1 knockout flies ( Figure 4B ) , which could be due to the fact that the peripherally synapsing interneuron ( PSI ) , a cholinergic neuron interposing between the GF and DLMn , is an intrinsically weaker point in high frequency response [20] , which masks any defect in the GF itself in DSC1 flies occurring at higher frequencies . Since the TTM branch does not involve the PSI , we could observe a defect in the following ability of the TTM branch . These findings demonstrate that the GF itself in mutant flies harbors electric signaling defects , suggesting an important role of the DSC1 channel in maintaining stable high-frequency signal transmission along the GF circuit . To determine whether the GF-TTM impairment exists beyond the GF at the motorneuron level ( see Figure 3A for circuit ) , we applied thorax stimuli to bypass the GF and directly recruit DLMn and TTMn . We found that the responses of DLM and TTM were similar between w1118 and DSC1 knockout flies ( Figure 4B ) , indicating that the following abilities of both the DLMn and TTMn were similar and well above those of the GF-TTM branch . Therefore , the primary defect can be located to the GF itself . We then examined the effects of heat shock on following abilities of the GF-TTM and GF-PSI-DLM branches . A significant decrease in frequency response was observed in the GF-PSI-DLMn branch in both w1118 and DSC1 knockout flies , presumably due to further weakening of PSI frequency response by high temperature ( Figure 4C ) . In contrast , the GF-TTM response was not obviously changed in w1118 flies but a tendency of enhanced frequency response was observed in DSC1 knockout flies , a potential indication of heightened excitability related to the heat shock-induced “jumpy” phenotype of the mutant flies . Furthermore , heat shock did not significantly alter the frequency response of the downstream motor neurons , TTMn and DLMn , in both w1118 and DSC1 knockout flies ( Figure 4C ) . Thus , the results demonstrate that different neuronal elements in the GF circuit are differentially affected in DSC1 knockout flies and that high temperature treatment could lead to responses very different from that in control flies . In particular , the central GF-TTMn transmission in the DSC1 knockout flies displayed severe defect and instability , as well as drastically different heat shock response , whereas the ability of TTMn or DLMn to drive the TTM and DLM via peripheral synapses remains largely intact in DSC1 knockout flies even after heat shock ( Figure 4C ) . To determine whether the enhanced sensitivity to pyrethroids may also be linked to an altered sensitivity of the GFS in DSC1 knockout flies , we examined the effect of pyrethroids on the activities of the DLM branch since the ability of the DLM branch to follow high frequency stimulation was not altered in DSC1 knockout flies ( Figure 4B ) . The flies were examined fifteen minutes after exposure to pyrethroids by topical application to the dorsal thorax ( See Protocol S1 ) . In response to a train of 50 pulses delivered at 100 Hz and 30 V , each stimulus evoked a muscle potential from the DLM and a total of 50 muscle potentials were observed in both w1118 and DSC1a knockout flies ( Figure 5A ) , which is consistent with the results in Figure 4B . The responses , however , were altered after pyrethroid exposure . In w1118 flies , the number of muscle potentials during the third 10 stimuli was reduced by bioresmethrin; and no muscle potentials were elicited in the remaining two 10 stimuli ( Figure 5B ) . DSC1a flies were more sensitive to bioresmethrin; a reduction in the number of muscle potential was already evident in the second 10 stimuli ( Figure 5B ) . Similarly , upon exposure to deltamethrin , the number of muscle potentials was reduced in the fifth 10 stimuli in w1118 flies . In DSC1a flies , the reduction in muscle potentials was detected in the second 10 stimuli and this reduction progressed in the remaining three groups of stimuli ( Figure 5C ) . Thus , both type I and type II pyrethroids affect the GFS , and the DSC1a flies are more sensitive to pyrethroid-induced inhibition , as compared to w1118 flies .
Four-domain , Ca2+-selective cation channels with a DEEA motif at the selectivity filter are interesting because their amino acid sequences and gating properties are intermediate between sodium and calcium channels . As such , these channels could potentially be an important evolutionary link between sodium and calcium channels . Although this family of cation channels has apparently been lost in vertebrates , they appear to be widespread in invertebrates . Despite the intriguing nature of the DEEA motif-containing cation channels , their role in animal physiology is not well understood . In this study , by taking advantage of the genetic tractability of Drosophila melanogaster , we assessed the role of DSC1 , a prototype of DEEA motif-containing cation channels , in insect biology by generating null DSC1 knockout mutants using gene-targeted knockout via homologous recombination . Our behavioral , pharmacological and electrophysiological analyses of these mutants uncover an important role of the DSC1 channel in maintaining the overall stability of neural circuits , particularly under stressful environmental conditions . Prior to our study , Anholt and colleagues [10] reported an olfactory phenotype in the Drosophila smi60 line , which has reduced ( by two-fold ) DSC1 transcript level associated with a P-element insertion in an intron . Our behavioral analysis of the DSC1 knockout flies confirmed this pioneering observation and further shows that the knockout lines have a more severe olfaction defect compared to the smi60 line . Besides the olfaction phenotype , however , we discovered that the DSC1 knockout flies also exhibit a prominent jumpy phenotype when disturbed . In particular , this defect was intensified under heat shock and starvation conditions . Intriguingly , DSC1 knockout flies are also more sensitive to pyrethroid insecticides . What could be the common neural/physiological processes that directly or indirectly affect olfaction , jumpiness , and response to heat shock , starvation and pyrethroids ? Our electrophysiological analyses using a well-defined neural circuit , the GFS , reveal that the DSC1 channel indeed contributes to nerve membrane excitability and may play an important role in balancing neuronal excitability and stability of synaptic transmission , thus providing an extended safety margin when the nervous system is required to operate beyond the normal functional range under challenging , extreme conditions . First , we detected a defect in the refractory period of LL responses in the DSC1 knockout flies upon heat shock and starvation . Consistent with earlier findings [21] , heat shock reduced LLRP . We found that the reduction in the LLRP was more drastic in DSC1 knockout flies than in w1118 flies . The difference was most pronounced at 10 min of heat shock , at which the jumpy phenotype was also most distinct . Intriguingly , starvation also reduced LLRP in DSC1 knockout flies ( but not for w1118 ) and triggered the jumpy phenotype . Thus , reduced LLRP correlates with the jumpy phenotype , heat shock , and starvation . The LL response of the GF pathway is driven by sensory input from vision , olfaction and other sensory systems . In contrast , the SL response reflects direct activation of the GF and bypasses input from all the sensory circuits [22] . We found that deletion of the DSC1 channel did not cause severe alteration in the SL response of the DLMs ( Figure S2 ) or its high-frequency responses ( Figure 4B ) , suggesting that one of the key functions of the DSC1 channel is in modulating the activities of the central neurons presynaptic to the giant fiber . Deletion of the DSC1 channel enhanced the electrical activities of these neurons . This conclusion is consistent with the high level of DSC1 expression in the sensory system , such as the optic lobes , which send sensory input directly or indirectly to the GF [4] . Second , examination of the GF-TTM branch allows us to assess the synapses that connect the giant fiber to the motor neurons , and to the NMJ [22] . The synaptic defect appears to be in the GF terminal itself , not at the TTMn output since the response of TTMn was not altered when the TTM response was recruited by direct thorax stimulation . Furthermore , heat shock enhanced the following ability of the GF-TTM branch , again not the responses of DLMn and TTMn . These findings show that the DSC1 channel is not only involved in the frequency responses of GF at room temperature ( Figure 4B ) , but also important for the regulation of GF function , possibly other neuronal elements in the nervous system when facing environmental challenges , such as heat shock ( Figure 4C ) . Third , destabilization of the GFS caused by the deletion of the DSC1 channel is also evident when depolarization at presynaptic neurons was intensified by the action of pyrethroids on sodium channels . Strikingly , both bioresmethrin ( type I pyrethroid ) and deltamethrin ( type II pyrethroid ) impaired the following ability of the SL response of the GFS during repetitive high voltage stimulation and this impairment was more severe in DSC1 knockout flies , compared with w1118 flies ( Figure 5 ) . The increased toxicity of pyrethroids on DSC1 knockout flies is therefore likely because DSC1 knockout flies have a hypersensitive nervous system . Based on these findings , we propose that the DSC1 channel function as a stability guarding system to keep insect nerve firing properties in check , as evident by the greatly enhanced variability and deteriorated frequency response in the GF-TTM SL response ( Figure 4B ) . This physiological function becomes even more apparent when the neural circuit is challenged under stressful conditions , such as heat shock , starvation or pyrethroid exposure ( Figure 3 , Figure 4 , and Figure 5 ) , which tend to hyper-stimulate the nervous system . How the DSC1/BSC1 cation channel exerts stability control over stress-enhanced neuronal hyperexcitability at the molecular level remains to be worked out . We have initiated such experiments in the larval neuromuscular preparation in which axonal action potentials and synaptic potentials can be directly recorded with microelectrodes . At present time , the preliminary results indicate defects in both axonal action potentials and neurotransmitter release . It is well known that synaptic transmission is regulated by calcium influxes . Calcium influx at the nerve terminals is mediated by voltage-gated calcium channels . Because the DSC1/BSC1 channel is a voltage-gated cation channel with relatively high permeability to Ca2+ [6] , [7] , it may have a fundamental role , together with classical voltage-gated calcium channels , in regulating Ca2+ fluxes at the nerve terminals . Intriguingly , unlike classical calcium channels , the DSC1 channel does not seem to play a critical role in regulating neurotransmitter release under normal laboratory condition since deletion of the DSC1 channel did not severely impair the functioning of the GFS under the normal laboratory condition . However , high-frequency stimulation , or heat shock , starvation or pyrethroid exposure , all of which induce excess neurotransmitter release [23] , intensify the neural defects in DSC1 knockout flies ( Figure 3 , Figure 4 , Figure 5 ) . The DSC1 channel could take part in the regulation of synaptic vesicle cycling , which is critical for synaptic transmission [24] . Both exocytotic release and re-uptake by endocytosis are controlled by Ca2+ influx [24] , [25] . Interestingly , the identity of the calcium channel that controls endocytosis at the presynaptic terminals has not been identified in insects , raising the intriguing possibility that the DSC1/BSC1 channel could be a candidate . Conceivably , retarded cycling of synaptic vesicles at the presynaptic terminals , upon hyper-stimulation of nerve terminals in the absence of DSC1 channels , would almost certainly contribute to the defects in locomotion ( jumpiness ) , and exasperated response to heat shock , starvation and pyrethroid exposure . It is also possible that deletion of the DSC1 channel somehow affect the expression or function of another ion channel , such as the sodium channel , that is important for maintaining the proper electric signaling in the nervous system . We examined the transcript level of the para sodium channel by quantitative-PCR , but did not detect a significant difference between w1118 and DSC1 knockout flies ( Figure S3 ) . Whether other ion channels in the nervous system are altered in the DSC1 knockout flies to compensate for the loss of the DSC1 channel remains to be determined . In summary , using D . melanogaster DSC1 knockout mutants we have provided strong genetic evidence for an important role of the DSC1 channel in insect neurobiology and neurotoxicology . The results described in this study not only provide fundamental insight into the in vivo function of a prototypic member of an historically elusive family of ion channel , but also may have significant implications for the development of new and safer insecticides . In particular , the DSC1/BSC1-family cation channel , which are absent in vertebrates , may be excellent targets for the development of a new generation of pesticides . In addition , because the DSC1 mutants are more susceptible to pyrethroids , DSC1 channel blockers may be useful to enhance the efficacy of pyrethroids , which are currently a key weapon against numerous agriculturally and medically important arthropod pests , including malaria-transmitting mosquitoes .
For constructing the donor transgene , a 6 . 6-kb DSC1 genomic DNA region was amplified in two 3 . 3-kb fragments ( Figure 1 ) by PCR ( primer sequences are listed in Table S1 ) using genomic DNA isolated from w1118 flies and platinum Taq DNA polymerase High Fidelity ( Invitrogen ) . The upstream and downstream 3 . 3 kb fragments encode IS1–5 , and IS6 and part of the linker connecting domain I and domain II , respectively . The DNA fragments were first cloned into PCR2 . 1 vector ( Invitrogen ) . A stop codon along with a MluI ( within ST1 ) or an EcoRI ( within ST2 ) cleavage site was then introduced in the middle of the each fragment , ( i . e . , ST1 in the upstream fragment and ST2 in the downstream fragment , Figure 1 ) , by PCR-mediated mutagenesis ( primer sequences are listed in Table S1 ) using Pfu DNA polymerase ( Stratagene ) . The PCR was carried out in a 50 µl reaction mixture for 18 cycles with 50 ng of DNA template and 2 units of Pfu DNA polymerase . The upstream fragment and the downstream fragments were then cloned into the pW25 vector ( kindly provided by Dr . Kent G Golic , University of Utah ) at the NotI and Acc65I sites , and the BsiWI and AscI sites , respectively . The gene knock-out strategy followed that reported by Rong and Golic [26] . Transgenic flies carry heat-inducible FLP recombinase ( 70FLP ) and I-SceI endonuclease ( 70I-SceI ) on chromosome 2 were kindly provided by Steve Crews ( University of North Carolina , Chapel Hill ) . The donor constructs were transformed into w1118 flies by standard P-element mediated transformation . The obtained transgenic flies that carry a donor construct on chromosome X or 3 were crossed with transgenic flies that carry both the heat-inducible 70FLP and 70I-SceI transgenes . The 3–4 day old progeny were heat-shocked at 38°C for 1 h and crossed to w1118 flies . When offspring with pigmented eyes were observed , the w+ gene was mapped to detect its mobilization to chromosome 2 ( to which DSC1 maps ) . Genomic DNA was isolated from 20–30 adult flies by standard methods . Southern blot analysis was performed using a DIG DNA Labeling and Detection Kit ( Roche ) according to the instruction manual . Briefly , DNA ( ∼10 µg ) was cleaved with KpnI , fractioned in a 0 . 8% agarose gel and transferred to nylon membrane ( Amersham Pharmacia Biotech ) . Blots were probed with DIG-labeled DNA containing DSC1 genomic DNA fragments ( see Figure 1C ) . Hybridization was carried out for 16 h at 42°C in DIG Easy Hyb ( Roche ) . Filters were washed twice for 15 min each at room temperature with 2× SSC containing 0 . 1% SDS , and twice for 30 min each at 68°C with 0 . 2× SSC containing 0 . 1% SDS . To avoid using damaged flies , a climbing assay ( See Protocol S1 ) was performed and only flies that could climb into the top vial were collected for the heat shock assay . Flies ( 10 flies in each vial ) were incubated in a 40°C humidified hybridization oven ( Hybaid , Thermo Scientific , Inc . ) with a glass front door which allowed direct observation of fly behavior during heat shock . The number of paralyzed flies was counted every minute during the 15-min heat shock period . Humidity in the chamber was maintained by including a plastic tray ( 8×12×3 cm ) containing water . Paralysis is defined as loss of an ability to walk . In addition , the number of jumps in each vial was recorded from the 5th to 10th minutes during heat shock . A jump was defined as flying off from the wall or bottom of the vial . Flies were returned to room temperature after the 15-min heat shock . A modified climbing assay was performed at various time points to determine recovery from the heat shock . The flies were tapped down to the bottom and given 30 seconds to stand up and climb . The number of flies that could climb on the wall of the vial was recorded . The assay was performed every 2 minutes during the first 20 minutes and every 5 minutes during the second 20 minutes of the recovery period . The method was modified from a previous study [27] . Thirty two- to three-day-old male flies were evenly divided into six groups ( i . e . , five flies/vial ) and raised on regular fly food for another two days before being transferred to vials containing 0 . 5% agar ( five ml ) instead of food . Flies were then transferred to fresh agar-containing vials daily and a jump phenotype similar to that induced by heat shock was examined daily for three days . The number of jumps per vial ( five flies ) was recorded during the first second to tenth seconds after a gentle tap on the vial . Deltamethrin , DCJW and fipronil were kindly provided by Bhupinder Khambay ( Rothamsted Research , Ltd . ) , Keith Wing and Daniel Cordova ( Dupont ) and Vince Salgado ( BASF ) , respectively . The method for contact bioassay was similar to that described in Hardstone et al . , [28] ( See Protocol S1 for details ) . The median lethal concentration ( LC50 ) and 95% confidential interval were calculated using the POLOplus software ( LeOra Software Company ) . LC50 values were considered as significantly different if the 95% confidence intervals did not overlap . Methods for recording GF-driven muscle potentials with electrical stimulation similar to those used in Engel and Wu [11] , [29] , [30] . Briefly , the fly was tethered to a tungsten hook and electric stimuli ( 0 . 1 msec , Grass S88 ) were delivered across the brain through two tungsten electrodes inserted in the compound eyes just beneath the cornea ( for long- and short-latency responses , anode in the right eye ) or on the junction of prothorax and metathorax ( for thorax stimulation of motor neurons , anode on the right side ) . The action potentials in the left tergotrochanteral muscle ( TTM , leg extensor ) and the right dorsal longitudinal muscle a ( DLMa , indirect flight muscle ) were recorded to indicate the giant fiber pathway output . Long and short latency response thresholds were assessed using single-pulse test stimuli with duration of 0 . 1 ms . Muscle responses were amplified by a DAM50 DC amplifier ( World Precision Instrument , Inc . ) and converted to digital signal by a Digidata 1440A ( Axon Instrument , Inc . ) coupled with Clampex 10 . 2 and Clampfit 10 . 2 software ( Axon Instrument , Inc . ) . Student's t-test was used to analyze the jump phenotype during heat shock and the climbing assay results . Two-way analysis of variance ( ANOVA ) with Tukey's test employed as the post hoc test were used to analyze the data of GFS recording . F-test was used to compare sample variances to detect differences in sample distribution . P<0 . 05 was set as the criterion for statistical significance . | Voltage-gated sodium and calcium channels are four-domain proteins that are essential for electrical signaling in neurons and other excitable cells . Recent genomic and functional analyses reveal a novel family of four-domain , Ca2+-selective cation channels in a variety of invertebrates , from sea anemones to insects . The amino acid sequences and gating properties of these channels appear to be intermediate between sodium and calcium channels; as such , these channels could potentially be an important evolutionary link between sodium and calcium channels . Despite the intriguing nature of this family of cation channels , their role in animal physiology remains mysterious . In this study , taking advantage of the genetic tractability of the fruit fly , Drosophila melanogaster , we examined the physiological role of such a channel , called DSC1 , in this model insect . We generated two DSC1 knockout lines and conducted behavioral and electrophysiological analyses . Our results show that the DSC1 channel contributes to neuronal excitability regulation and plays a unique role in retaining stability of the nervous system function in response to environmental stresses , including heat shock and starvation . Interestingly , the DSC1 knockout flies were also more susceptible to pyrethroid insecticides , which are used globally as a major weapon against the malaria-carrying mosquitoes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"agriculture",
"biology"
]
| 2013 | Role of the DSC1 Channel in Regulating Neuronal Excitability in Drosophila melanogaster: Extending Nervous System Stability under Stress |
The Eukaryotic RecA-like proteins Rad51 and Dmc1 cooperate during meiosis to promote recombination between homologous chromosomes by repairing programmed DNA double strand breaks ( DSBs ) . Previous studies showed that Rad51 and Dmc1 form partially overlapping co-foci . Here we show these Rad51-Dmc1 co-foci are often arranged in pairs separated by distances of up to 400 nm . Paired co-foci remain prevalent when DSBs are dramatically reduced or when strand exchange or synapsis is blocked . Super-resolution dSTORM microscopy reveals that individual foci observed by conventional light microscopy are often composed of two or more substructures . The data support a model in which the two tracts of ssDNA formed by a single DSB separate from one another by distances of up to 400 nm , with both tracts often bound by one or more short ( about 100 nt ) Rad51 filaments and also by one or more short Dmc1 filaments .
Meiotic recombination is a highly regulated process that faithfully repairs programed DSBs , ensuring accurate reductional chromosome segregation at meiosis I[1] . Following pre-meiotic DNA replication , Spo11 introduces DSBs across the genome . These DSBs are nucleolytically resected , revealing 3’ single strand DNA ( ssDNA ) tracts that are subsequently used to locate an intact , homologous double strand DNA ( dsDNA ) repair template . Upon completing the homology search , the ssDNA invades the intact dsDNA duplex creating a displacement-loop structure . The invading 3’ end serves as a primer for restorative DNA synthesis , facilitating the completion of the DNA repair process . During meiosis , the eukaryotic RecA homologs Rad51 and Dmc1 cooperate to promote homology search and strand exchange , the central step in homologous recombination[2] . Like RecA , Rad51 and Dmc1 form nucleoprotein filaments on ssDNA and catalyze strand exchange in vitro[3–6] . Rad51 is responsible for catalyzing strand exchange in vivo in mitotically cycling cells[7] . However , the meiosis-specific protein Dmc1 is the predominant meiotic strand exchange enzyme[8–10] . Rad51’s activity is inhibited during meiosis by the Hed1 protein[11 , 12] . Nonetheless , Rad51 plays an important non-enzymatic role , promoting Dmc1 assembly and directing it to invade a homolog chromatid rather than a sister chromatid[9 , 13–15] . Rad51 and Dmc1 form spatially associated repair complexes in accord with their genetic interaction . In spread meiotic S . cerevisiae nuclei , Rad51 and Dmc1 form DSB-dependent foci[16] . Rad51 and Dmc1 approximately co-localize as side-by-side , partially offset “co-foci” when viewed by widefield epifluoresence microscopy [16–18] . Combined with the propensity of Rad51 and Dmc1 to interact homotypically but not heterotypically[19–21] , this staining pattern led to speculation that Rad51 and Dmc1 homofilaments might occupy opposite ends of each DSB[17 , 18] . This speculation , along with a number of other observations , influenced the development of models of meiotic recombination involving asymmetric loading of Rad51 and Dmc1 on opposite DSB ends[1 , 15] . The asymmetric loading model in which Rad51 and Dmc1 homofilaments occupy opposite ends of a single DSB has awkward implications . First , the model implies that Rad51 is selectively loaded onto one member of a pair of ends , somehow avoiding the other end of each DSB . Second , Rad51 is required for normal assembly of Dmc1[16 , 22] , thus the asymmetric loading model requires Rad51 to accomplish this assembly function in trans on the opposite DSB ssDNA tract . Third , the asymmetric loading model implies that only the Dmc1-decorated end is capable of strand exchange , given the disposability of Rad51’s catalytic activity[9–11] . While the asymmetric loading model has been argued to account for the apparent differentiation of the two ends of a DSB[1 , 15] , both ends of a DSB can catalyze strand exchange as evidenced by multichromatid joint molecules that are normally disassembled by the Sgs1 helicase[23] . Here we present evidence contradicting the asymmetric loading model and supporting a model in which both Rad51 and Dmc1 can and often do load on both DSB ends . Additionally , we demonstrate that Rad51 and Dmc1 filaments are very short in vivo coating only about 100 nt of ssDNA .
When viewed by widefield microscopy , a substantial fraction of Rad51-Dmc1 co-foci appear to be arranged in pairs separated by up to 400 nm in sparsely populated regions of meiotic S . cerevisiae spread diploid ( 2N ) nuclei ( Fig 1A , arrowhead ) . It was unclear if this apparent pairing might simply be due to the fortuitous arrangement of unrelated foci in a crowded nucleus . To reduce the potential bias imposed by crowding , we constructed a tetraploid ( 4N ) strain hypomorphic for SPO11 . This tetraploid carries one wild type ( SPO11+ ) allele and three alleles that code catalytically inactive protein ( spo11-Y135F ) [24] . Spo11 is the transesterase that forms DSBs[25 , 26] . As expected for a strain with reduced DSB levels and larger nuclei , the average density of Rad51 ( and Dmc1 ) foci per unit area was lower in the spo11 hypomorphic tetraploid than in wild type diploids . However , individual nuclei with focus densities ranging from very low to the same as that in wild type diploid nuclei were observed , likely reflecting the homeostatic control of DSB levels that eventually compensate for reduced Spo11 activity[14 , 27–29] . When the analysis was restricted to tetraploid nuclei early in prophase ( 2 . 5 hr ) with low densities of foci , pairs of foci separated by ≤ 400 nm could clearly be seen ( Fig 1B–1D ) . Thus , Rad51-Dmc1 co-foci tend to be arranged in pairs . To quantitatively assess the prevalence of the paired co-focus architecture , the distance between each focus and its nearest neighboring focus was measured in the spo11 hypomorphic tetraploid strains . For example , the shortest distance between the centroid of a Rad51 focus and the centroid of the nearest neighboring Rad51 focus was determined ( solid lined arrow , Fig 1E , left ) . This process was repeated for each Rad51 focus in the nucleus , and such measurements from many nuclei were pooled to form a Rad51-to-Rad51 ( R→R ) nearest neighbor distribution ( Fig 1F , red ) . The limited resolution of light microscopy makes it difficult to distinguish foci separated by distances below 200 nm , resulting in estimates of inter-focus distances that decay below 200 nm with a minimum inter-focus distance of about 150 nm in this assay ( see Analysis section of Materials and Methods ) . To evaluate the likelihood that this distribution could be generated by the independent , random assortment of individual foci , a simulated Rad51-to-Rad51 nearest neighbor distribution was generated by randomly positioning foci within a nuclear area ( Fig 1F , blue , see Methods ) . For a given distance bin , a higher frequency of experimental pairs than of simulated pairs indicates that such pairs are enriched relative to the expectation based on the random arrangement of foci . Quantitative analysis of nearest neighbor distributions confirmed that both Rad51 and Dmc1 foci are non-randomly arranged in pairs separated by up to 400 nm ( Fig 1F and 1G ) . This enrichment of focus pairs separated by ≤400 nm was detected in both unselected nuclei and nuclei selected for low focus density ( S1A–S1D Fig ) . Additionally , paired foci are not simply due to local crowding , since ≤400 nm pairs are enriched in locally sparse regions ( S1E and S1F Fig ) . This analysis cannot make strong conclusions regarding the prevalence of pairing below 200 nm ( due to the diffraction limit of light ) . The method also has limited power to identify non-random spatial patterns at distances greater than 400 nm , because it reports only on the closest distance between structures . However , if a precise longer pairing distance were present , it could not account for more than 15% of Rad51 or Dmc1 foci because a subpopulation larger than that size would have been detected ( Fig 1F and 1G ) . We also used the nearest neighbor methodology to assess the spatial relationship of Rad51 foci to Dmc1 foci and vice versa ( Fig 1E , right ) . 84% of Rad51 foci are less than 200 nm from the nearest neighboring Dmc1 focus ( Fig 1H ) . Similarly , 85% of Dmc1 foci are less than 200 nm from a Rad51 focus ( Fig 1I ) . It should be noted that this analysis differs from that of measuring distances between resolved pairs of Rad51 , or resolved pairs of Dmc1 foci , because the use of different fluorophores to detect Rad51 and Dmc1 allows measurement of nearest neighbor distances below the resolution limit . The distributions of Rad51-Dmc1 distances support the longstanding observation that Rad51 and Dmc1 foci colocalize imperfectly in a side-by-side configuration[16–18] . In the context of a ≤400 nm pair , each focus in a pair could be a Rad51-Dmc1 co-focus , a Rad51-only focus , or a Dmc1-only focus . All permutations of pair composition were observed with no apparent difference between Rad51 and Dmc1 ( e . g . a Rad51 focus was paired with a co-focus equally often as a Dmc1 focus was paired with a co-focus ) . A conservative measurement indicates that at least 50% of all focus pairs contain two Rad51-Dmc1 co-foci . Furthermore , the staining intensity of Rad51 or Dmc1 within one half of a co-focus pair does not predict the staining intensity of Rad51 or Dmc1 within the other half of the co-focus pair ( S1G–S1J Fig ) . These results suggest each Rad51-Dmc1 co-focus assembles independently of its partner co-focus . The enrichment of Rad51 ( and Dmc1 ) pairing at distances ≤400 nm is not simply explained by the restriction of foci to positions along an underlying , but invisible , linear chromosome structure . In leptotene nuclei , Zip1 foci form along the otherwise invisible , linear chromosome axis . Yet , the Zip1 nearest neighbor distribution peaks at longer distances than the Rad51 distribution measured in the same nuclei ( Fig 2A–2E ) . Furthermore , the Zip1 nearest neighbor distribution resembles the distribution expected from a random arrangement of Zip1 foci ( Fig 2E ) . These observations are consistent with the possibility that individual pairs of Rad51-Dmc1 co-foci represent meiotic recombination complexes associated with a single DSB . Nuclei with a single DSB site provided evidence that each DSB often gives rise to a pair of foci . To further rule out the possibility that paired Rad51-Dmc1 co-foci represent adjacent DSBs along a single chromosome , recombination complexes were characterized in a spo11 strain heterozygous for the VDE cut site . This background is heterozygous for a cleavage site for the meiosis-specific endonuclease VDE on chromosome IV . During meiotic prophase , both sister chromatids are eventually cleaved and repaired via recombination with a homolog chromatid , resulting in about 86% 4:0 gene conversions[30 , 31] . The kinetics of cleavage and repair at the VDE cut site have not ( to our knowledge ) been assayed in single cells , but at any given time 0 , 1 , or 2 DSBs are expected to be present . In chromosome spread preparations , VDE-dependent Rad51-Dmc1 foci were readily observed . While only 4% ( 2/47 ) of spo11 nuclei were focus-positive , 47% ( 54/116 ) of wild type and 56% ( 49/88 ) of mnd1 nuclei were focus-positive in the spo11 VDE cut site heterozygote background . Nuclei with a single focus or a single pair of foci are the predominant classes observed in wild type cells ( Fig 2F and 2G and 2L; categories I and II ) representing 79% of total focus-positive nuclei . However , a more complex class of structures containing more than two Rad51 foci and/or more than two Dmc1 foci in close proximity to each other was observed in about 6% of focus-positive wild type nuclei ( Fig 2H and 2I and 2L; category III ) . Additionally , nuclei with distant cytological complexes ( greater than 1 μm apart ) were observed in about 15% of wild type nuclei: 9 . 6% contained no more than two Rad51 or two Dmc1 foci ( Fig 2J and 2L; category IV ) and 5 . 8% contained greater than two Rad51 or two Dmc1 foci ( Fig 2K and 2L; category V ) . Importantly , 11 . 5% of wild type nuclei with greater than two Rad51 foci or two Dmc1 foci ( categories III and V ) cannot be explained by the asymmetric loading model in which two Rad51-Dmc1 co-foci—one co-focus at each of two DSBs—is the most complicated predicted structure . Moreover , a model in which both Rad51 and Dmc1 can co-occupy each of the four ssDNA tracts associated with two DSBs readily explains these data . This interpretation is further supported by the accumulation of category III and V nuclei to levels up to 39% of focus-positive cells in mnd1 mutants defective for strand exchange ( Fig 2I and 2K and 2L ) [32–34] . We interpret distant cytological complexes as likely reflecting situations in which the VDE sites on both sister chromatids are simultaneously cleaved , breaking the sister chromatid pair entirely , such that the two halves of the broken chromosome separate in vivo and/or during the spreading procedure . Supporting this interpretation , the incidence of distantly separated foci was greater among nuclei with more than 2 Rad51 and/or Dmc1 foci ( 50% ) than among nuclei with only two foci ( 19% ) . Furthermore , the class of nuclei with distantly separated foci ( categories IV and V ) was more predominant in an mnd1 mutant , primarily due to an increase in category V nuclei ( Fig 2J and 2K and 2L ) . This observation suggests that blockage of Mnd1-mediated strand exchange results in the accumulation of DSBs leading to the complete breakage of chromosome IV , and loading of Rad51 and/or Dmc1 on all four associated ssDNA tracts . In conclusion , the range of structures observed in spo11 strains heterozygous for the VDE cut site are best explained if both Rad51 and Dmc1 can load onto both ends of a single meiotic DSB . Co-focus pairs can be seen at a strong meiotic recombination hotspot . To obtain further evidence that paired Rad51-Dmc1 co-foci form at a single DSB , we simultaneously visualized co-foci and fluorescent landmarks flanking the HIS4::LEU2 DSB hotspot on chromosome III . Specifically , tandemly repeated arrays of the bacterial transcription factor binding sites lacO and tetO were integrated about 60 kb away proximal to the centromere and 37 kb away distal to the centromere , respectively , relative to the HIS4::LEU2 DSB hotspot on chromosome III , in the spo11 hypomorphic tetraploid strain with only a single chromatid bearing the HIS4::LEU2 DSB hotspot . Expression of 3xHA-LacI and YFP-TetR allows the arrays to serve as chromosomal landmarks that can be visualized along with Rad51 or Dmc1 foci via immunostaining . Although the position of the lacO array was not always detectable ( due to partial proteolysis of the 3xHA-LacI fusion protein as demonstrated cytologically and by western blot ) , a significant subset of nuclei displayed both landmarks . Considering only those nuclei that display both landmarks and a low density of Dmc1 foci in the vicinity of the landmarks , 41% ( 16/39 ) had one resolvable pair of Dmc1 foci at the hotspot ( Fig 2M–2O ) ; the remainder had a single focus . Similarly , 39% ( 19/49 ) of selected nuclei had paired Rad51 . Given previous results showing that breakage of both sister chromatids is very rare at the HIS4::LEU2 hotspot ( only 11% of tetrads show evidence of two different recombination events in a less severe spo11 hypomorph than the one used in this study ) [35] , the data provide additional evidence that a single DSB produces a co-focus pair . Interestingly , the tetO landmark signal was split in 50% or more of nuclei , indicating separation of sister chromatids 37 kb away from the HIS4::LEU2 DSB hotspot ( for example , Fig 2O ) . Although the frequency of tetO splitting increases after meiotic induction , the vast majority of the observed splitting is SPO11-independent ( 76% vs . 67% of total nuclei in wild type and spo11 , respectively; n = 45 nuclei for each ) . When there was an optically resolvable pair of split tetO spots , they were separated by around 400 nm ( 381±144; range 240–1008; 94% are between 240 and 600 nm ) , similar to the distance between paired Rad51 ( and Dmc1 ) foci . To probe the relationship between the paired co-focus architecture and both the progression of recombination reactions and transitions in global chromosome structure , nearest neighbor distributions were determined in strand exchange-defective ( mnd1 ) and synapsis-defective ( zip1 ) mutants . When strand exchange was blocked in an mnd1 mutant[32–34] , the paired character of Rad51 ( and Dmc1 ) foci was maintained ( Fig 3A and 3C and 3D ) . The pairing of Rad51 ( and Dmc1 ) foci was also unaltered in the zip1 mutant which blocks the progress of recombination reactions after strand exchange ( Fig 3B–3D ) [36–38] . In zip1 mutants , nascent post-strand exchange intermediates promote homolog co-alignment at a distance of 400 nm or less , but the closer 100 nm alignment of lateral elements resulting from elongation of the synaptonemal complex does not occur[38] . Together , these mutants suggest that the ≤400 nm paired Rad51-Dmc1 co-focus architecture does not require strand exchange and is retained until synapsis . Further analysis demonstrated that the nearest neighbor distributions of Rad51-to-Dmc1 and Dmc1-to-Rad51 distances are also unaltered in strand exchange-defective and synapsis-defective mutants ( Fig 3E and 3F ) , suggesting that the side-by-side Rad51-Dmc1 configuration is also established prior to , and persists after , strand exchange . To characterize recombination complexes in more detail we used direct stochastic optical reconstruction microscopy ( dSTORM ) , a method with higher resolution than standard widefield microscopy[39 , 40] . We validated our experimental system by imaging long Rad51 filaments prepared by the same method as that used previously for analysis by electron microscopy[41–43] . The Rad51 filaments were assembled on dsDNA in vitro , deposited on a coverslip , immunostained , and imaged . As expected , reconstructed dSTORM micrographs clearly show long Rad51 filaments often exceeding one micron in length ( Fig 4A–4D ) . Following indirect immunostaining , the structures observed are 70 nm wide , 60 nm wider than the underlying protein filament as a consequence of both antibody decoration and the finite resolution of the imaging method . Given this apparent filament width , only filaments longer than about 70 nm have readily identifiable long axes in dSTORM reconstructions . Such elongated filaments were readily identified . After validating our dSTORM imaging procedure , we utilized the methodology to interrogate the molecular structures underlying widefield Rad51 and Dmc1 foci in vivo . Super-resolution light microscopy revealed that individual Rad51 ( and Dmc1 ) foci observed by widefield microscopy are often composed of multiple distinct substructures . These Rad51 and Dmc1 “super-resolution foci” , hereafter referred to as sr foci , appeared to be paired or to be members of small clusters separated by less than 200 nm ( Fig 4E and 4F ) . In our initial experiments , images generated by super-resolution analysis software displayed fine threads connecting close pairs of sr foci , but further analysis showed this feature of the images was artifactual ( for details see Methods , S1 Text , and S2 Fig ) . Elimination of this artifact revealed pairs or clusters of distinct staining sr foci . In accord with the visual prevalence of pairing/clustering , the 100 nm peak in the Rad51-to-Rad51 ( and Dmc1-to-Dmc1 ) nearest neighbor distribution was enriched relative to a randomly simulated distribution ( Fig 4I and S3A Fig ) . Furthermore , 28% of Rad51 sr foci that were within 200 nm of at least one sr focus were within 200 nm of more than one sr focus . The nearest neighbor distributions of Rad51 sr foci and Dmc1 sr foci were similar in wild type diploids and spo11 hypormorphic tetraploids ( Fig 4I and S3A–S3C Fig ) . Importantly , like standard resolution foci , Rad51 and Dmc1 sr foci are predominantly SPO11-dependent ( S4 Fig ) ; the very faint SPO11-independent staining observed is likely to reflect binding of Rad51 and Dmc1 at non-DSB sites[44] . Artificially blurring dSTORM micrographs generates images that display the same nearest neighbor distributions as those obtained by the standard widefield method , indicating that the images produced by the two modalities are congruent ( S5 Fig ) . Comparison of widefield and dSTORM images demonstrates that standard resolution Rad51 ( and Dmc1 ) foci are often composed of two or more constituent sr foci . In the context of paired standard resolution Rad51 foci ( see Fig 1 ) , each standard resolution focus contains an average of 1 . 53 ± 0 . 78 Rad51 sr foci ( Fig 4L–4O ) . Additionally , nuclei with greater than four Rad51 sr foci were observed in spo11 mutants heterozygous for the VDE cut site ( S6 Fig ) . These observations suggest that more than one Rad51 ( and more than one Dmc1 ) filament can occupy a single tract of ssDNA . It should also be noted that the dSTORM results indicate that many of the “single” foci seen by widefield , i . e . foci well-separated from their nearest neighbors , represent pairs or clusters of sr foci . Like the ≤400 nm pairs observed with widefield microscopy , the Rad51-to-Rad51 ( and Dmc1-to-Dmc1 ) nearest neighbor distributions obtained by dSTORM imaging were unaffected in strand exchange and synapsis mutants ( Figs 4G and 4H and 4J and 4K and S3C and S3D ) . These results suggest that sr focus clusters form independently of strand exchange and persist until synapsis . The Rad51 and Dmc1 structures observed with super-resolution microscopy are only slightly elongated ( Fig 4E–4H and 4P ) . The longest dimension of the image of Rad51 and Dmc1 sr foci in wild type nuclei is 114 ± 29 . 9 and 119 ± 37 . 7 nm , respectively . Although these distances are greater than the resolution of the technique , the images of sr foci are only slightly elongated with aspect ratios of 1 . 27 ± 0 . 25 and 1 . 31 ± 0 . 23 ( Fig 4P ) . The small size of sr foci suggests only a portion of the ssDNA formed by resection of a DSB is bound by Rad51 or Dmc1 ( see Discussion ) . We next asked if Rad51 and Dmc1 sr foci were altered when strand exchange is blocked . In wild type cells , Dmc1 foci are present and strand exchange occurs between 3 and 6 hours[8 , 13 , 16]; foci disappear and strand exchange is complete before 8 hours . If strand exchange is blocked in an mnd1 or a dmc1 mutant , DSB-associated ssDNA tracts become much longer than normal by 8 hours[8 , 13 , 34] . dSTORM microscopy of 3 . 5 hour mnd1 nuclei revealed a punctate Dmc1staining pattern very similar to that seen in wild type ( Fig 4H and 4P ) . However , the Dmc1 staining patterns seen in 8-hour mnd1 nuclei were dramatically different; they contained numerous elongated structures with contour lengths often reaching 250 nm ( Fig 5A–5D ) . This result indicates that elongation of Dmc1-containing structures is limited by Mnd1 function , likely because completion of Hop2-Mnd1-dependent strand exchange is associated with Dmc1 disassembly , as has been argued for Rad51 based on biochemical observations[45] . Importantly , no corresponding elongated Dmc1 structures have been observed in strand exchange-proficient cells ( for example , Fig 4F and 4P ) . Like Dmc1 staining , little or no difference in the Rad51 staining was seen in either mnd1 or dmc1 mutants , as compared to wild type , at 3 . 5 hours ( Figs 4G and 4P and 5E–5H ) . However , at 8 hours , mnd1 nuclei frequently displayed clusters of about 3–7 poorly resolved Rad51 sr foci ( Fig 5I–5L ) . These clusters lacked obvious elongated structure , in dramatic contrast to the fibrous staining patterns seen for Dmc1 using duplicate slides from the same cultures and time points ( compare Fig 5A to 5I ) . Similar clustering of Rad51 sr foci was seen at 8 hours in dmc1 cells ( Fig 5M–5P ) , suggesting Dmc1 does not influence formation of Rad51 clusters when strand exchange is blocked . The Rad51 staining results are in agreement with previous low resolution studies showing Rad51 focus staining intensity increases with time in dmc1 and mnd1 mutants[16 , 33] . The observation that elongated Dmc1 structures are seen at 8 hours in mnd1 mutants indicates that the small dimensions of sr foci in strand exchange-proficient cells truly reflect the dimensions of underlying structures in living cells , rather than a shortcoming of the chromosome spreading or staining methods . To further address this issue , we examined mitotic srs2 mutant cells overexpressing Rad51 . These mutant cells had previously been shown to form elongated structures by widefield microscopy[46] . In these cells , spread nuclei stained for Rad51 and imaged by dSTORM revealed filamentous structures with contour lengths of up to 1 . 5 μm ( Fig 5Q–5T ) . Thus , we have observed highly elongated Dmc1 structures ( Fig 5A–5D ) and Rad51 structures ( Fig 5Q–5T ) in vivo with dSTORM imaging . Importantly , the widths of these elongated Rad51 and Dmc1 structures match the 70 nm width of the images of Rad51 filaments assembled in vitro and imaged under identical conditions ( Fig 4A–4D ) . These results strongly suggest that the small Rad51 and Dmc1 sr foci observed in meiotic nuclei are indeed very short filaments . Furthermore , we can rule out several potential explanations for the size and arrangement of in vivo sr foci including disruption of filaments by the spreading procedure and incomplete antibody labeling . Finally , the clear contrast between elongated structures/filaments and clustered sr foci strengthens the conclusion that clustered sr foci associated with hyper-resected tracts of ssDNA represent distinct filamentous entities rather than an artifact of dSTORM imaging . In summary , dSTORM easily detects elongated Rad51 and Dmc1 filaments under our experimental conditions . Given this , we conclude that the Rad51 and Dmc1 sr foci observed in wild type meiosis represent underlying structures that are shorter than 40 protomers on average .
Assuming that Rad51 and Dmc1 sr foci represent the DNA bound helical filaments that promote strand exchange in vitro , the sr foci we observed by dSTORM suggest that the Rad51 and Dmc1 filaments that promote recombination in vivo are extremely short and that more than one Rad51 and/or Dmc1 filament can form on the same ssDNA tract . Images of Rad51 and Dmc1 sr foci are only about 115 nm long . Since the diameter of Rad51 and Dmc1 filaments are known to be 10 nm [43] , but sr foci are 70 nm wide , the images of sr foci observed with super-resolution microscopy following indirect immunostaining are likely about 60 nm larger in each dimension than the underlying protein complex . This difference can be accounted for by considering the size of the primary and secondary antibodies decorating the structure . Thus , assuming that RecA homolog structures represent the canonical nucleoprotein filaments[43] , they are about 55 nm long . Given that Rad51 and Dmc1 filaments contain about 2 nt per nm[41 , 43 , 47] , this corresponds to roughly 100 nt , 33 protomers of Rad51 ( or Dmc1 ) , and 5 turns of the helical nucleoprotein filament . This length estimate suggests that an individual Rad51 or Dmc1 filament typically occupies less than 15% of a typical 800 nt ssDNA tract [8 , 48] . An alternative , but in our view less likely , interpretation , is that RecA homologs coat a more substantial fraction of each ssDNA tract in a previously unknown compact configuration . Additionally , a significant fraction of the closely spaced ( <200 nm apart ) Rad51 sr foci likely represent loading of distinct filaments on the same tract of ssDNA as evidenced by detection of nuclei with more than 4 Rad51 sr foci in VDE cut site heterozygote nuclei and the accumulation of clustered Rad51 sr foci at late time points in strand exchange mutants . Although 100 nt filaments are quite small relative to those that have been studied in many biochemical experiments[49] , both ensemble and single molecule experiments have shown that only 8 nt is sufficient for recognition of homology by RecA-like strand exchange proteins [50 , 51] . Thus , the size of the structures we observe is more than sufficient to promote efficient recombination . The finding that Rad51 and Dmc1 structures are small relative to the average length of ssDNA tracts is in agreement with previous observations indicating that Rad51 foci display offset colocalization with foci formed by the recombination proteins RPA and Rad52 . These observations suggest that RPA and Rad52 can simultaneously occupy ssDNA segments adjacent to regions bound by Rad51 and Dmc1[52] . Furthermore , the lack of an inverse relationship between Rad51 and Dmc1 staining intensity in a single co-focus supports the hypothesis that ssDNA tracts are not completely bound by RecA homologs . All of these observations suggest that the protein composition and organization of a tract of ssDNA associated with a meiotic DSB is highly heterogeneous ( Fig 6 , top ) . The evidence presented here for loading of both Rad51 and Dmc1 on both DSB ends suggests that the two ends may be functionally identical with respect to homology search and strand exchange activity . The observed paired co-focus structure also suggests a simple molecular mechanism through which Rad51 could mediate Dmc1 assembly[2] . Binding of Rad51 to a ssDNA tract could enhance the efficiency of nearby Dmc1 filament initiation on that same tract [16 , 22 , 53 , 54] . Furthermore , although the two DSB ends engage the homolog in temporal succession[55] , the first invading end need not be predetermined . Rather , the two ends of a DSB might both be released from the tethered loop complex in which they are formed ( Fig 6 , left ) [56] . It is possible that both ends then compete to locate and invade a homolog chromatid with the “winner” maturing into the stable single end invasion intermediate[55] . Indeed , the presence of Dmc1—and thus its homology search and strand exchange activity—on both ends of a single DSB accounts for the existence of joint molecules connecting more than two chromatids[23] . These multichromatid joint molecules are not readily accounted for by the model positing loading of Rad51 and Dmc1 on opposite DSB ends , because the ability of Rad51 to form homology-dependent joint molecules is inhibited by Hed1 protein during meiotic prophase[11] . While our data argue that a large fraction of DSBs load both Rad51 and Dmc1 on both ends , a subset of structures in our micrographs do not fit into this class . At standard resolution , about 15% of Rad51 foci do not colocalize with Dmc1 foci ( and vice versa ) , representing at least one ssDNA tract lacking Dmc1 ( or Rad51 ) . At super-resolution , about 21% of Rad51 sr foci lack a neighboring Rad51 sr focus within 400 nm , indicating a Rad51-coated ssDNA end unaccompanied by another Rad51-coated ssDNA tract within 400 nm . These results could be explained by a number of non-mutually exclusive possibilities . First , some focus pairs may be too close together to be resolved , even by dSTORM . Second , it is likely that a substantial fraction of structures formed by Rad51 and Dmc1 are too small to be detected by our methods . Consistent with this , focus staining intensities vary dramatically with some foci being only barely detectable above background . Third , the spread nuclei analyzed represent static snapshots of a possibly dynamic and undefined recombinosome assembly and disassembly process that ultimately passes through a stage where both Rad51 and Dmc1 occupy both DSB ends . Fourth , some DSBs may be processed by alternate pathways that do not require Rad51 and Dmc1 loading on both ends , as observed in a rad51 mutant for example[13 , 22] . Despite the proposed molecular symmetry , our results suggest that DSB ends can separate to distances of up to , and only rarely longer than , 400 nm . This distance is reminiscent of the conserved distance of 400 nm or less at which homologs are initially aligned[57] . Homolog alignment occurs at sites of nascent strand exchange intermediates—called axial association sites—prior to engagement of the second end of the DSB[36 , 38 , 58] . Thus , we interpret Rad51-Dmc1 co-foci paired at distances up to 400 nm as representing the structures responsible for the strand exchange-dependent component of homolog alignment ( Fig 6 , right ) . This cytological pattern of spatial separation between pairs of foci composed of recombination proteins Rad51 and Mer3 , has previously been observed during leptotene/zygotene in Zea mays and Sordaria macrospora , respectively[58 , 59] . Yet , surprisingly , the paired Rad51-Dmc1 co-focus architecture depends on neither strand exchange nor synapsis in S . cerevisiae . Thus , we conclude that the two ends of a DSB are separated by up to 400 nm prior to and after strand exchange ( Fig 6 ) . In other words , we propose that the structure of a pre-strand exchange intermediate determines the distance at which homolog axes will be aligned . What determines the distance at which the two ends of a DSB are separated ? The variability up to 400 nm suggests that this length is not strictly determined by a fixed proteinacious scaffold , such as the Zip1 protein , which is responsible for the defined 100 nm spacing of lateral elements in the synaptonemal complex[60] . Instead , we propose that DSB formation within a chromatin loop[56] , followed by release of the two flexible chromatin arms anchored at sites of sister chromatid cohesion along developing axial elements , allows the two DSB ends to separate as observed . The distribution of sister chromatid splitting distances rarely exceeds 600 nm , suggesting that chromatin loops are about 600 nm in length , similar to estimates of loop size obtained by electron microscopy[61] . If this loop size is correct , the lengths of the two released chromatin arms sum to about 600 nm , a distance in reasonable agreement with the separation distances detected for pairs of Rad51 or Dmc1 foci . We note that the DSB-independence of the sister chromatid splitting favors our hypothesis that the ends passively separate and does not provide support for the idea that meiotic recombination events involve local loss of sister chromatid cohesion or disassembly of chromatin to form a long homology-searching tentacle capable of searching the nuclear volume without associated movement of chromosome axes [62] . It is also in agreement with the alternative possibility that the searching entity is a chromatin “arm , ” rarely longer than 400 nm , that extends from the axial element [62] . It is also important to note that although the ≤400 nm axis alignment distance is conserved in diverse organisms , meiotic chromatin loop size is not [57] . Thus , the similarity between focus separation distances and chromatin loop size could be unique to S . cerevisiae .
Tetraploid spo11 hypomorphic strains were constructed by mating a SPO11/spo11-Y135F a/Δ diploid with a spo11-Y135F/spo11-Y135F α/Δ diploid[63] . The a/Δ and α/Δ diploids were obtained by gene targeting with DNA constructs designed to delete the MATα or the MATa locus , respectively . The strain with cytological landmarks in Fig 2 was constructed by combining the HIS4::LEU2 DSB hotspot[55] , a tetO array , a lacO array , YFP-TetR , and 3xHA-LacI via genetic crosses[64] . lacO and tetO arrays were inserted into Chr III , centromere proximal and distal to the HIS4::LEU2 DSB hotspot , respectively , using the cloning-free method[65] . For the spo11 VDE cut site heterozygote experiment in Fig 2 , DKB 4571 was constructed by mating of YOC 3524 and YOC 3525[30] , provided by the Ohya Lab . DKB 5369 was constructed by transformation of YOC 3524 and YOC 3525 with a PCR product designed to introduce an mnd1 mutation followed by mating . See S1 Text for further details and S1 Table for genotypes . Transfer to sporulation medium was used to induce synchronous meiotic cultures and preparation of spread chromosomes was previously described[16] . For experiments requiring visualization of LacI-3xHA , 4 mM PMSF was added to spheroplast suspensions and the solutions used for spreading . Rabbit anti-Rad51 ( #159 ) and goat or guinea pig anti-Dmc1 ( #189 or #174 ) antibodies were utilized at 1:1000 dilutions . Chicken anti-GFP ( Invitrogen ) was used at 1:1000 , and mouse anti-HA ( Santa Cruz ) was used at 1:100 . Alexa fluor 488 and 594 labeled secondary antibodies ( Invitrogen ) were used to stain Rad51 and Dmc1 . Alexa fluor 647 and 750 labeled secondary antibodies ( Invitrogen ) were used to stain for GFP and HA . All secondary antibodies were used at 1:1000 , except Alexa fluor 750 which was used at 1:100 . For Fig 2A–2E , the goat anti-Zip1 antibody ( Santa Cruz , sc-15632 ) was used at 1:1000 and stained with 1:1000 Alexa 594 secondary antibody . Images were acquired on a Zeiss Axiovision 4 . 6 microscope at 100X magnification and adjusted for brightness and contrast on ImageJ/FIJI software . For all two-color experiments , proper registration of image pairs obtained with different filter sets was confirmed using fluorescent beads ( Molecular Probes , L-5241 ) . For dSTORM microscopy , spreads were stained with 1:1000 rabbit anti-Rad51 ( or 1:1000 goat anti-Dmc1 ) and then with 1:1000 Alexa fluor 647 secondary . 0 . 05% Triton X-100 was included in the TBS washes to reduce background . Coverslips were mounted on a depression well slide filled with 10 mM MEA ( prepared in PBS ) and sealed with a two-part curable rubber product called “Body Double” ( Smooth-On , Inc ) . Image sequences were acquired on a Leica SR GSD 3D microscope in 2D epifluorescence mode . Depletion with the 642 laser at 100% power was performed until the frame correlation dropped below 0 . 05 , then acquisition commenced at 60% laser power . At least 25 , 000 frames were acquired . Images were reconstructed with the QuickPALM plugin[66] on FIJI using: an input pixel size of 100 nm , a reconstruction pixel size of 20 nm , a minimum SNR of 5 . 00 , minimum symmetry of 0% , local threshold of 25% , maximum iterations per frame of 1000 , and 50 threads . Also , a FWHM of 2 pixels was used to eliminate an artifact in which a large fraction of adjacent structures appear to be connected by a thin , sparsely populated thread ( see S1 Text ) . This artifact is due to the almost simultaneous blinking of two adjacent fluorophores resulting in a “mis-called” event half way between two diffraction-limited blinks ( S2 Fig ) . A 0 . 75 pixel Gaussian blur was applied to each micrograph before analysis . For Fig 5Q–5T , Rad51 was over-expressed[46 , 67] . For Fig 4A–4D , in vitro Rad51 assembly reactions were performed as previously described[9] with the following exceptions . The assembly reaction included 0 . 5 nM 2 . 7 kb linear dsDNA ( 1 . 35 μM bp ) generated by asymmetric PCR of pRS306 with one biotinylated primer , 1 . 6 μM Rad51 , and 1 μM Hed1 to stabilize the filament[68] . The reaction was fixed with 3% PFA prepared in reaction buffer and then added to previously prepared coverslips coated with streptavidin by a modified version of a previously established protocol[69] . Slides were stained and imaged using dSTORM as described above . Custom written ImageJ macros designated 4-spot macro were used to generate the nearest neighbor distributions in Figs 1 and 2 and 3 and S1 . Specifically , the ( x , y ) coordinates of focal centroids were determined manually with the ImageJ multipoint selection tool within the context of the macro . Note that Rad51 and Dmc1 foci were assumed to be diffraction limited spots ( a valid assumption based on dSTORM micrographs ) . Thus , focus centroids as close as about 150 nm were often recognized as being distinct based on the fine appearance of staining structures ( elongation vs perfectly circular focus , appearance of two maxima , etc ) , despite the fact that they were closer together than the resolution limit ( around 250 nm ) . The nearest neighbor distributions were generated by coalescing output from the ImageJ macro using Excel . Simulated distributions were generated with custom written ImageJ macros ( see S1 Text for details ) . dSTORM reconstructions were scored in FIJI . The elliptical selection tool was fit to each observed sr focus . The ( x , y ) coordinates and various other descriptors of the ellipses were measured . Nearest neighbor distributions were determined in Excel workbooks , and simulations were performed in ImageJ as above . For the Zip1 experiment in Fig 2 , the ( x , y ) coordinates of both Rad51 and Zip1 foci were determined in early prophase nuclei . Nuclei were selected for having both Rad51 and Zip1 staining , but only nuclei with a completely punctate Zip1 staining pattern were chosen for analysis . Scoring and simulation of nearest neighbor positions were performed as above . For the spo11 VDE cut site heterozygote experiment in Fig 2 , unselected nuclei were scored , but only focus-positive nuclei are included in the analysis . Single foci ( category I ) include Rad51-only , Dmc1-only , and Rad51-Dmc1 co-foci . Similarly , paired foci include all varieties of single foci , located within 1 μm of each other ( category II ) . The “>2 Rad51 and/or Dmc1 foci” class ( category III ) includes structures where all of those foci are within 1 μm one another . The distant class ( category IV ) represents nuclei in which foci are separated by distances greater than 1 μm , but there are no more than 2 Rad51 or Dmc1 foci . Finally , the “>2 Rad51 and/or Dmc1 foci and distant” class ( category V ) consists of nuclei with two distinct cytological complexes separated by greater than 1 μm where the sum or Rad51 or Dmc1 foci is greater than two . Scoring the cytological landmark experiment in Fig 2 required multiple levels of filtering . First , only nuclei displaying lacO and tetO arrays were scored for experiments involving both landmarks . Meiotic proteolysis resulted in a large fraction of nuclei without lacO spots . Also , rare nuclei with >2 lacO or tetO spots were excluded from analysis . Furthermore , only nuclei with one Dmc1 focus within 300 nm of the point between the closest tetO and lacO foci and 0–2 Dmc1 foci within 1 μm of that Dmc1 focus were analyzed . All analysis was performed with ImageJ software . | During meiosis , a specialized form of chromosome segregation ensures that gametes contain only one copy of the parental chromosome complement . Accurate segregation of maternal and paternal chromosomes requires them to first become connected in pairs . Homologous recombination forms these needed connections . Connections between homolog chromosomes are made by forming and then repairing DNA double strand breaks . Rad51 and Dmc1 are structurally related enzymes that form complexes by binding DNA at sites of breaks , where they then function to promote break repair by searching for and invading corresponding unbroken DNA sequences on a homologous chromosome . In this paper , we describe several features of the recombination complex structure . We provide evidence that: 1 . both Rad51 and Dmc1 load onto both ends of a single DSB , accounting for the known activity of both ends of a DSB; 2 . the two ends of a DSB can separate by distances of up to 0 . 4 microns; 3 . Rad51 and Dmc1 complexes only occupy small segments of DNA ( about 100 bases ) ; 4 . multiple short Rad51 and Dmc1 complexes can occupy a single DSB end . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Small Rad51 and Dmc1 Complexes Often Co-occupy Both Ends of a Meiotic DNA Double Strand Break |
Schizosaccharomyces pombe Rad3 checkpoint kinase and its human ortholog ATR are essential for maintaining genome integrity in cells treated with genotoxins that damage DNA or arrest replication forks . Rad3 and ATR also function during unperturbed growth , although the events triggering their activation and their critical functions are largely unknown . Here , we use ChIP-on-chip analysis to map genomic loci decorated by phosphorylated histone H2A ( γH2A ) , a Rad3 substrate that establishes a chromatin-based recruitment platform for Crb2 and Brc1 DNA repair/checkpoint proteins . Unexpectedly , γH2A marks a diverse array of genomic features during S-phase , including natural replication fork barriers and a fork breakage site , retrotransposons , heterochromatin in the centromeres and telomeres , and ribosomal RNA ( rDNA ) repeats . γH2A formation at the centromeres and telomeres is associated with heterochromatin establishment by Clr4 histone methyltransferase . We show that γH2A domains recruit Brc1 , a factor involved in repair of damaged replication forks . Brc1 C-terminal BRCT domain binding to γH2A is crucial in the absence of Rqh1Sgs1 , a RecQ DNA helicase required for rDNA maintenance whose human homologs are mutated in patients with Werner , Bloom , and Rothmund–Thomson syndromes that are characterized by cancer-predisposition or accelerated aging . We conclude that Rad3 phosphorylates histone H2A to mobilize Brc1 to critical genomic domains during S-phase , and this pathway functions in parallel with Rqh1 DNA helicase in maintaining genome integrity .
During DNA replication cells are particularly vulnerable to loss of genetic information and mutation [1] . The DNA replication checkpoint pathway monitors the genome to detect and stabilize stalled forks , initiate repair , and delay mitotic entry until DNA damage is repaired [2] . Checkpoint activation during replication is triggered by the kinase ATR , which is crucial for maintenance of genome stability during S-phase [3] . Chromosomal instability , neuronal defects , and premature aging characterize human Seckel syndrome caused by ATR deficiency , whilst corresponding mutants of Saccharomyces cerevisiae Mec1 or Schizosaccharomyces pombe Rad3 display chromosomal instability and are hypersensitive to a wide spectrum of genotoxins [3] . ATR orthologs are important for maintaining genome integrity in the absence of genotoxins , yet little is known about the endogenous sources of ATR activation . Candidates include highly repetitive DNA , natural replication fork barriers ( RFBs ) , and chromosomal “fragile sites” , which may present obstacles to replication [1] , [4] , [5] . However , it is unknown whether specific chromosomal domains are responsible for Rad3 activation during an unperturbed cell cycle . One of the most rapid and highly conserved responses to DNA damage is phosphorylation of histone H2AX or histone H2A in yeast [6]–[8] . ATRMec1/Rad3 and the related kinase ATMTel1 catalyze this phosphorylation [6]–[8] . Phospho-H2AX ( γH2AX ) spreads over extensive chromatin domains flanking DNA double-strand breaks ( DSBs ) [9] , [10] . Mice deficient for γH2AX are immunocompromised , display sensitivity to ionizing radiation ( IR ) , and moderate genomic instability , while yeast with H2A mutations that abolish phosphorylation are moderately sensitive to a variety of DNA damaging agents [7] , [8] , [11] , [12] . Although typically associated with DSBs , γH2AX also forms in response to genotoxic stresses that stall or collapse replication forks [13]–[18] . The role of γH2AX in repair of replication-associated damage is largely unexplored . γH2AX may have a direct function in stabilization of stalled forks [15] , [18] and has been linked to the Fanconi Anemia/BRCA pathway for resolving stalled forks [17] . The best-characterized γH2AX-binding proteins are mammalian Mdc1 and fission yeast Crb2 , which associate with γH2AX through their C-terminal tandem BRCT domains [19] , [20] . We recently discovered that Brc1 is a second γH2A-binding protein in fission yeast [21] . Like Mdc1 and Crb2 , Brc1 uses a pair of C-terminal BRCT domains to bind γH2A . However , unlike Mdc1 or Crb2 , which play critical roles in the checkpoint responses to DSBs , Brc1 is specifically required for repair of replication-associated DNA damage [22] , [23] . Indeed , Brc1 mutants that cannot bind γH2A are sensitive to replication-specific genotoxins , as are fission yeast “htaAQ” mutants lacking the phosphorylation site in histone H2A [8] , [21] . Brc1 is structurally related to budding yeast Rtt107/Esc4 and mammalian PTIP [23]–[25]; however , the mechanisms by which these proteins protect genome integrity remains a mystery . Rad3 , γH2A and Brc1-defective mutants all have increased frequencies of spontaneous Rad22Rad52 foci during S-phase , indicating homologous recombination ( HR ) repair of stalled or collapsed replication forks [8] , [26] . This phenotype is shared with mutants lacking Rqh1 , a RecQ-like DNA helicase that is required for genome stability during S-phase and which has critical functions in maintaining the copy number of the repeated ribosomal RNA gene ( rDNA ) loci [27]–[30] . These phenotypes may be partially explained by endogenous DNA damage arising from normal metabolic processes , but other factors could be specific fragile sites , where DNA-bound protein complexes , transcriptional machinery , or other uncharacterized chromatin structures interfere with DNA replication . To address whether Rad3 is activated at specific chromosomal domains during replication , we mapped γH2A formation during an unperturbed S-phase using whole-genome microarrays . These studies , which by proxy reveal the site of action of ATRRad3 or ATMTel1 , show that γH2A decorates a surprisingly diverse array of chromosomal structures , including all heterochromatin domains . There are interesting similarities and differences with the genome-wide distribution of γH2A in budding yeast [31] . We also demonstrate that Brc1 binding to γH2A is critical in the absence of Rqh1 , indicating a crucial role for the Rad3-γH2A-Brc1 pathway in maintaining genome integrity during S-phase .
To address whether specific chromosomal domains trigger DNA damage responses in the absence of exogenous genotoxins , we wanted to map the distribution of γH2A in the S . pombe genome using chromatin immunoprecipitation and tiled microarray ( ChIP-on-chip ) analysis . Since γH2A ChIP-on-chip analysis has not been performed in S . pombe , we first tested if γH2A could be detected at a site-specific DNA DSB made by activating expression of the HO-endonuclease [32] . γH2A ChIP was performed with a phospho-H2A specific antibody [33] . As a control we used a strain in which both histone H2A genes ( hta1 and hta2 ) encode proteins that cannot be phosphorylated , hereafter referred to as the htaAQ mutant [8] . Upon HO-endonuclease induction , γH2A was detected in a broad domain spanning more than 40 kb surrounding the DSB , but was reduced within a 2 kb region around the break site , which is similar to previous observations in S . cerevisiae [34] ( Figure S1 ) . Next , to obtain a positive control site for the ChIP-on-chip experiments , we examined if γH2A can be detected at the only known fragile site in S . pombe , located in the mating-type ( MT ) locus . Mating-type switching involves DSB formation and recombination , but it does not elicit a checkpoint arrest , as evidenced by normal cell-cycle progression in switching strains . The configuration of the MT locus varies with strain mating type , so a typical MT locus of a standard h90 homothallic strain is shown in Figure 1A . The mat1 gene is actively expressed and determines the mating-type , which is either h- or h+ . Mat1 is replicated from the right side , which induces a DSB at a fragile site next to mat1 [35]–[37] . A polar replication fork barrier ( RFB ) , called RTS1 , blocks replication forks from the centromere . The induced DSB initiates recombination with inactive donor alleles , mat2 ( h+ ) or mat3 ( h− ) , located in a heterochromatic domain , to switch the mating-type at mat1 [37] . We checked γH2A formation at the mat1 DSB in a “Donorless” strain , which lacks the donor alleles , and uses sister-chromatid recombination to repair the break [36] , [38] . Since DSB formation at mat1 is transient and occurs during replication , ChIP analysis was performed on cells that were enriched in S-phase by using the “cdc25-22 block and release” protocol to synchronize cell cycle progression , as described in Materials and Methods [39] , [40] . We observed that γH2A was highly enriched near the DSB in the MT locus in S-phase , but not in G2 ( Figure 1B ) . Typical h− and h+ laboratory strains also contained γH2A in the MT locus ( data not shown ) . This demonstrated that γH2A is triggered by a transient DSB during mating-type switching and therefore the MT locus should be a reliable positive control in γH2A ChIP-on-chip analysis . The “Donorless” strain was used for genome-wide localization analysis of γH2A . This MT locus configuration has no adverse effects on viability [38] . Samples were collected from cultures synchronized in S-phase using cdc25-22 block and release . DNA was hybridized to whole-genome tiling microarrays with 20 bp resolution and the data was analyzed using the Model-based Analysis of Tiling Array ( MAT ) algorithm [41] . ChIP input DNA was used as a control for the microarray data analysis . Predicted γH2A enrichment at each site is displayed as “MAT score” as explained in the supporting information ( Text S1 ) . Note that MAT score is a measure of probability and is not a quantitative measure of protein amount . Approximately 400 statistically validated peaks were identified , all of which had MAT scores of 5 or higher . The resulting genome-wide landscape of γH2A encompassed a diverse array of genomic features , including the mating-type locus , the rDNA loci , and surprisingly all heterochromatin regions , including the centromeres and telomeres ( Figure 2A ) . γH2A formation was also observed at mobile genetic elements ( Tf2-type retrotransposons and wtf elements ) and in a subset of gene coding sequences ( Table S1 ) . To confirm γH2A enrichment at the identified loci and determine if the phosphorylation was S-phase specific , we monitored γH2A levels by conventional ChIP during the cell cycle . Cells were synchronized in G2 phase using cdc25-22 temperature arrest , and ChIP samples were collected every 30 min after return to permissive temperature . Cell cycle progression was assessed by monitoring the septation index . Quantitative PCR ( qPCR ) primers were designed in regions predicted to have high γH2A levels by ChIP-on-chip analysis . Our results showed that γH2A enrichment in all regions was low in G2 , increased and peaked in S-phase ( coinciding with the rise in septation index ) , and decreased as the cells reentered G2 ( Figure 2B ) . The highest level of enrichment was detected in the MT locus ( MT ) . The telomeres ( tel ) had intermediate γH2A levels , and lower signals were detected at the outer centromere repeats ( cen-dh ) . As predicted by ChIP-on-chip analysis , no enrichment was detected in the centromere core ( cnt1 ) . Transient formation of γH2A during replication suggests that these genomic regions are especially susceptible to replication fork arrest or collapse during S-phase . These data indicated that generation of γH2A is a normal event in S-phase . To gauge the level of γH2A relative to that caused by replication fork arrest or collapse , we released cds1+ or cds1Δ cells from the cdc25-22 arrest into media containing 12 mM hydroxyurea ( HU ) . The amount of γH2A in wild type cells treated with HU was comparable to untreated cells harvested in S-phase , whereas replication fork collapse caused by HU treatment of cds1Δ cells led to substantially higher γH2A levels ( Figure 2C ) . Next , we investigated the relative contributions of Rad3 and Tel1 checkpoint kinases to γH2A formation during unperturbed replication . Using the cdc25-22 arrest and release protocol to enrich cells in S-phase , we observed a large decrease in γH2A at every site in the absence of Rad3 , while tel1Δ mutants were similar to wild type ( Figure 2C ) . The effect of the rad3Δ mutation was weakest at the telomere , indicating a major role for Tel1 at telomeres , at least in the absence of Rad3 . Indeed , Rad3-defective cells have short telomeres and rad3Δ tel1Δ double mutants completely lose telomeres [42] , [43] . These facts are consistent with our data showing a high γH2A signal at the telomeres in G2-phase rad3Δ cells ( Figure 2C ) . As expected , deletion of both kinases completely abolished γH2A formation ( Figure S2 ) . The MT locus showed the highest γH2A levels on the array ( Figure 2A ) . Detailed analysis revealed a broad , non-uniform distribution of γH2A spanning approximately 50 kb around the mat1 DSB and the RTS1 barrier , with a trough immediately adjacent to the DSB and enrichment peaking within 10 kb of the DSB ( Figure 3A ) . The diagrams above the plot correlate locations of MT locus features to the microarray data and compare the microarray MT locus configuration to the Donorless strain . The main difference is at mat1 , due to different mating types , and in the silent region , where a LEU2 marker replaced the mat3 allele in the Donorless strain . The silent region in both strains is flanked by inverted repeats ( IR ) . Closer inspection revealed additional interesting features in the γH2A pattern . First , H2A phosphorylation was preferentially distributed in gene coding regions , an example of which is shown in Figure 3B . This trend was observed genome-wide and could be due to lower nucleosome occupancy in highly AT-rich S . pombe intergenic regions [44] . Second , we identified two γH2A peaks at the IR elements flanking the missing mat3 site , which has a gap in γH2A signal ( Figure 3C ) . These IRs act as boundary elements that prevent the spread of heterochromatin out of the MT silent region [45] . A comparison to histone modifications in this region revealed that γH2A colocalized with H3 K9 methylation , a marker of heterochromatin , which peaks at the IRs [45] . The specific boundary elements are B-box sequences bound by TFIIIC , a factor associated with RNA polymerase III [45] . We found that γH2A peaks are bounded by B-box sequences , similar to heterochromatin ( Figure 3C ) . There were no additional boundaries to γH2A spreading in the MT locus , and the overall signal attenuated as distance from the DSB increased . We next tested whether the RTS1 RFB alone was sufficient for triggering γH2A formation by using a strain called “smt0” , which contains the RTS1 barrier but lacks the DSB imprinting site [46] . We detected γH2A on the left side of RTS1 where the replication fork stalls at the barrier , but the signal greatly decreased on the right side of the DSB ( Figure 3D ) . Thus , although the majority of γH2A at the MT locus was due to the DSB , replication fork pausing at RTS1 also triggers γH2A formation . Both RTS1 barrier activity and DSB formation depend on the Swi1-Swi3 complex , which travels with the replisome and mediates replication fork pausing at natural barriers [47] , [48] . To confirm that H2A phosphorylation was dependent on RFB activity , γH2A ChIP was performed in swi1Δ or swi3Δ strains that contain RTS1 and the DSB site . The formation of γH2A at the MT locus was eliminated in both swi1Δ and swi3Δ cells ( Figure 3E ) . These data demonstrate that both a transient DSB and a natural replication fork barrier trigger γH2A , which depends on replication fork pausing mediated by the Swi1-Swi3 complex . We next examined whether γH2A can form at an ectopic RTS1 fork barrier . We used a strain where RTS1 was inserted in the Ade6 locus , between two direct repeats ( Figure 4A ) [49] . This is a strong polar replication fork barrier and outside the MT locus fork stalling at RTS1 promotes recombination , which occurs without fork breakage , DSB formation , or checkpoint activation [49] . As a control we used an “inactive” strain in which the barrier is oriented in the opposite direction , thereby avoiding fork arrest ( Figure 4A ) . ChIP analysis revealed a strong asymmetric enrichment of γH2A up to 5 kb away from the active RTS1 barrier ( Figure 4B ) . The majority of γH2A was located on the right side of the barrier , where fork stalling occurs and recombination is initiated [49] . There was comparatively little γH2A on the left side of the barrier and no change in γH2A levels was detected when RTS1 orientation was reversed . A primer near RTS1 in the MT locus was used as a positive control for the experiment and showed similar levels of γH2A in both strains . These data show that fork stalling at a polar fork barrier triggers formation of an asymmetric γH2A domain and suggests that γH2A may mark recombination hotspots in the genome . Multiple RFBs are present in the rDNA loci , which are located on the subtelomeric arms of chromosome 3 [50] , [51] . A large γH2A domain was detected in the rDNA , which decreased to background levels within 10 kb outside this region ( Figure 5A ) . The rDNA is organized into ∼150 tandem repeats , but the microarray probes represented only a few repeat units , as shown below the γH2A plot . The gaps between γH2A peaks reflect the absence of microarray probes at repetitive DNA sequences . Because rDNA sequences were poorly represented on the microarray we further examined the γH2A distribution using conventional ChIP analysis . A diagram of a single rDNA repeat is shown in Figure 5B . Each repeat consists of the 35S rDNA genes , a replication origin ( ars3001 ) , and four distinct replication fork barriers ( RFB1–3 and RFP4 ) . The rDNA is highly transcribed , and the RFBs between the repeats facilitate unidirectional replication to prevent head-on collisions between the replisome and transcription complexes [52] . Replication fork pausing at rDNA barriers also regulates recombination , which is necessary for dynamic maintenance of rDNA copy number [52] , [53] . Using the indicated primers ( Figure 5B ) , we detected γH2A enrichment throughout the rDNA locus , including the gene coding sequences and surrounding all four RFBs ( Figure 5C ) , as predicted by the ChIP-on-chip analysis . Enrichment was detected only in S-phase and not in G2-arrested cells , which suggests that RFB activity triggers γH2A formation . Interestingly , γH2A levels at the barriers correlated with barrier strength , with RFB1 being the strongest barrier , and RFB2 the weakest [51] . The Swi1-Swi3 complex is required for replication fork stalling at the three intergenic RFBs ( RFB1-3 ) [51] , [54] . The RFP4 barrier is atypical , and is thought to be caused by collisions between transcription and replication [51] . Deletion of Swi1 exacerbates fork stalling at RFP4 [51] . ChIP analysis in swi1Δ cells revealed high levels of γH2A in the rDNA in both G2 and S-phase , which contrasts the S-phase-specific γH2A formation in wild type cells ( Figure 5D ) . The high levels of γH2A could be caused by fork collapse and/or recombination due to replisome collisions with transcription at RFP4 . Increased recombination-associated Rad22 foci were observed in swi1Δ mutants , which supports these conclusions [48] , [55] . Overall , these results suggest that the Rad3-γH2A pathway has a role in the maintenance of rDNA during normal replication , and that perturbation of RFB activity increases genomic instability in the rDNA locus . Replication fork pausing can also occur at the highly transcribed tDNA genes [56] , [57] . However , we did not detect γH2A near tDNAs in our genome-wide analysis , with the exception of the centromeres , where clusters of tDNAs surround a γH2A domain located within the heterochromatic centromeric repeats ( Figure 6A ) . These tDNA clusters serve as boundary elements that prevent heterochromatin spreading from the centromere repeats [58] . In a similar manner , the tDNA clusters seem to prevent γH2A spreading from the repeats , because beyond the tDNAs the γH2A MAT scores drop sharply ( Figure 6A ) . The centromeric γH2A signal could be caused by replication through heterochromatin , which is discussed in the next section . There are some differences between the structure of RTS1 and rDNA RFBs and the tDNAs barriers . The tDNA barriers are nonpolar and weaker than RTS1 , and fork stalling at tDNAs does not depend on Swi1 [56] . However , Swi1 has a general role in facilitating replisome progression through tDNAs and the absence of Swi1 increases recombination at tDNAs , likely from increased collisions between the replisome and transcription [56] . We analyzed the effect of Swi1 deletion on γH2A levels near the leftmost tDNA cluster bordering centromere 2 ( Figure 6A ) . Primers were designed around the tDNATyr gene . ChIP in wild type cells showed no appreciable γH2A enrichment upstream of the tDNATyr ( Figure 6B ) . However , in swi1Δ mutants we detected increased γH2A levels around the tDNA cluster , as far as 11 kb away . Our data suggest that Swi1 is needed to promote replication fork progression through the tDNAs in a manner that avoids triggering γH2A formation . Interestingly , our genome-wide analysis revealed prominent γH2A peaks in all heterochromatic loci , including centromeres , telomeres , silent MT locus , and rDNA ( Figure 2A ) . Heterochromatin is established during S-phase , and is initiated by methylation of histone H3 on lysine 9 ( H3K9me ) by the Clr4 methyltransferase , followed by binding of Swi6 , a homolog of HP-1 , which regulates heterochromatin spreading and function [59] . Additionally , tandem repeat sequences called “dg/dh repeats” , located in the outer centromeres , subtelomeres , and the silent MT region , are transcribed into small interfering RNA ( siRNA ) to promote heterochromatin formation during S-phase [60] . Fission yeast centromeres consist of a central core ( cen ) , surrounded by inner ( imr ) and outer ( otr ) regions [59] . The otr contains dg/dh repeats and is the site of Clr4-dependent heterochromatin assembly . Detailed analysis revealed that γH2A was enriched in the otr and a part of the imr at all three centromeres ( Figure 7A ) . Remarkably , this pattern almost perfectly matches the distribution of heterochromatin markers Swi6 and H3K9me [61] . The spreading of γH2A beyond heterochromatin was blocked by tDNA clusters and inverted repeat ( IR ) boundary elements , which prevent heterochromatin spreading [58] , [61] . As discussed earlier , γH2A in the MT locus silent region was similarly restricted by IR boundary elements . The paucity of γH2A in the centromere core can also be attributed to the reduced density of H2A/H2B dimers from this part of the centromere [62] , [63] . It is intriguing that Rad3-dependent phosphorylation of H2A may be inhibited by the same boundary elements that prevent the spread of heterochromatic epigenetic modifications . The telomere sequence coverage on the S . pombe microarray is incomplete and the best coverage is of the subtelomere located on the left arm of chromosome 1 [61] . Detailed analysis showed that a large non-uniform γH2A domain extended up to 50kb away from the subtelomere ( Figure 7B ) . Like the centromere otr repeats , the subtelomere is modified by H3K9me and Swi6 , and contains dg/dh-like elements that are transcribed into siRNAs [59] . Comparison of the subtelomeric γH2A pattern to H3K9me revealed that γH2A colocalized with heterochromatin , but unlike at the centromeres , the signal spread more than 50 kb beyond the end of H3K9me chromatin [61] . This region of the subtelomere is also transcriptionally repressed , but via Clr6/Clr3 histone deacetylases , independently of Clr4 methylation [64] , [65] . It contains a significant number of meiotic genes , which are expressed during nitrogen starvation [66] . Thus , γH2A formation in subtelomeric regions is associated with two different types of repressed chromatin . The subtelomeres of chromosome 3 contain rDNA repeats , which also have H3K9 methylated heterochromatin [61] . Heterochromatin in the rDNA is limited from spreading by long terminal repeats ( LTRs ) [61] , however , γH2A enrichment continued past these boundaries ( Figure 5A ) . γH2A was also present at Tf2-type retrotransposons and S . pombe-specific wtf elements , which are transcriptionally repressed by Clr3/Clr6-mediated histone deacetylation [64] . The distribution of γH2A at Tf2s and wtfs was confined by LTRs ( Figure S3 ) . Due to high sequence similarity among the members of transposon families [67] it was not possible to distinguish whether all Tf2s and wtf elements , or only a sub-population , were enriched in γH2A . Functionally , the role of wtf elements is unknown , but they are very highly expressed during meiosis [66] , [67] . Overall , the association of γH2A with repressed chromatin during S-phase suggests that heterochromatin may impede replication fork progression . Gene coding regions enriched with γH2A were distributed on all three chromosomes ( Table S1 ) As mentioned earlier , the majority of these genes are located in subtelomeric regions , in silenced chromatin mediated by Clr6/Clr3 histone deacetylases ( Figure 7B ) . Analysis of gene ontology ( GO ) terms and expression data in the S . pombe GeneDB database ( www . genedb . org/genedb/pombe ) revealed that many of the γH2A enriched genes are involved in mating and meiosis or are upregulated in response to environmental stress ( Table S1 ) . The largest group of genes represented was cell adhesion proteins many of which are needed during mating and share a common structure in the form of internal tandem repeats [68] . An example of this is the SPBPJ4664 . 02 gene locus , which is predicted to code for a protein with ∼250 copies of 12-amino acid repeats according to the S . pombe GeneDB database ( www . genedb . org/genedb/pombe ) ( Figure S3 ) . Thus , our data suggests that induction of γH2A in genes could be triggered by either repressed chromatin or repetitive DNA sequences . The prominent enrichment of γH2A at heterochromatic loci suggested that a relationship exists between heterochromatin formation and γH2A during DNA replication . On one hand , heterochromatin contains many features that could impede replisome progression , including repetitive sequences , compacted higher-order DNA structures , and multiple DNA-bound proteins [4] , [69] . On the other hand , the striking similarity between heterochromatin distribution and γH2A prompted us to examine if cells lacking γH2A ( htaAQ mutant ) had defects in heterochromatin function . We tested chromosome segregation defects , loss of gene silencing , sensitivity to microtubule inhibitors , and telomere length alterations , but did not detect any defects ( data not shown ) . A crucial function of heterochromatin is recruitment of cohesin , and high levels of the cohesin subunit Rad21 are present at the centromere dg/dh repeats and in the subtelomeres [70] , [71] . γH2A has been shown to recruit cohesin to sites of induced DSBs [72] , [73] . However , ChIP analysis of Rad21 binding at the centromeres and subtelomeres did not show significant differences between wild type cells and htaAQ mutants ( Figure S4 ) . These results led us to conclude that γH2A either does not participate in heterochromatin function , or it is part of a redundant pathway , which masks any defects . We next tested whether γH2A formation was dependent on the presence of heterochromatin by examining γH2A levels in the absence of factors that mediate heterochromatin formation . ChIP of γH2A in clr4Δ and swi6Δ mutants in S-phase showed that deletion of Clr4 but not Swi6 led to decreased γH2A signals in the centromere dg/dh repeats ( Figure 7C ) . This suggests that Clr4-dependent heterochromatin formation triggers γH2A formation in the centromeres during replication . The lack of an effect in swi6Δ mutants indicates that a barrier to replication is still present in the centromeres . This idea is supported by studies of heterochromatin replication timing , which showed that centromere replication is partially impeded by Clr4-dependent heterochromatin in swi6Δ mutants [74] . Next , we analyzed γH2A at the telomeres in clr4Δ , swi6Δ , and taz1Δ mutants . Taz1 is a telomere end-capping protein that facilitates replication through the telomeres and regulates telomere homeostasis [75] , [76] . Since the microarray probe coverage was limited to the subtelomere , we extended our analysis of γH2A further into the telomeres . The subtelomere region is followed by telomere-associated sequence ( TAS ) elements , and then tandem telomeric repeats ( bottom diagram , Figure 7D ) . We observed high γH2A enrichment in the subtelomere , as predicted by the ChIP-chip analysis , but the signal rapidly decreased in the TAS region ( Figure 7D ) . The subtelomere contains Clr4-dependent heterochromatin , whereas the telomeric repeats bind Taz1 [61] , [70] , [77] . The distribution of γH2A was limited to the subtelomere region similar to H3K9me and Swi6 binding [70] , [77] . As a positive control for the ChIP experiments we confirmed that Taz1 binds in the TAS and the telomere repeats and not in the subtelomere ( Figure S5 ) . The subtelomeric γH2A signal decreased in both clr4Δ and swi6Δ mutants , but not in taz1Δ mutants ( Figure 7D ) . These data demonstrate that γH2A formation in the subtelomeres is linked to the presence of heterochromatin established by H3K9 methylation and Swi6 binding and is independent of Taz1 . Overall , our results indicate that transient γH2A formation in the centromeres and subtelomeres during DNA replication is associated with the presence of Clr4-dependent heterochromatin and partially depends on Swi6 . We recently discovered that γH2A is required for formation of both spontaneous and genotoxin-induced nuclear foci of Brc1 , a genome maintenance protein that has a role in replication fork stability and chromatin organization [21] , [23] . Spontaneous Brc1-GFP foci form in approximately 25% of wild type cells and about 60% of these foci are perinucleolar , indicating colocalization with rDNA [21] . Consistent with these observations , we detected γH2A at the rDNA in our Chip-chip analysis . Since γH2A levels were reduced in the absence of heterochromatin , we tested how deletion of Clr4 affected levels of spontaneous Brc1-GFP foci . We detected a reduction in Brc1-GFP foci in clr4Δ cells , from 30% to 12% ( Figure 8A ) . To test if Brc1 binds at γH2A sites identified in this study , we performed Brc1-GFP ChIP in wild type , htaAQ , and clr4Δ mutant cells ( Figure 8B ) . In addition to the rDNA , γH2A-dependent Brc1 binding was detected at the subtelomeres and in the outer centromere dh repeats ( Figure 8B ) . In clr4Δ cells Brc1 binding was reduced at the centromeres and telomeres , but not at the rDNA ( Figure 8B ) . These data show that activation of γH2A during replication recruits Brc1 to specific genomic regions , and that binding of Brc1 at centromeres and subtelomeres is associated with heterochromatin . Our discovery that γH2A recruits Brc1 to critical genomic features during S-phase suggests that Rad3 and γH2A protect genome integrity in response to endogenous replication-associated DNA damage . Indeed , a function for γH2A during unperturbed growth is indicated by the increased incidence of Rad22Rad52 HR repair foci in htaAQ cells [8] . A similar increase is seen in brc1Δ cells [21] . However , htaAQ cells do not display overt growth defects that would be suggestive of severe genomic instability . It is therefore likely that γH2A works redundantly with other genome maintenance factors to protect genome integrity during S-phase . One of these factors could be Rqh1 , a RecQ helicase that is required for rDNA locus stability [27] , [30] . As γH2A prominently decorates the rDNA loci during S-phase , we explored the genetic interactions between rqh1Δ and htaAQ mutations . As shown in dilution assays ( Figure 9A ) , we found that the growth of the rqh1Δ htaAQ cells is substantially compromised relative to either rqh1Δ or htaAQ cells [8] . These synergistic genetic interactions were maintained but apparently not enhanced in media containing hydroxyurea ( HU ) or camptothecin ( CPT ) , which stall or collapse replication forks . ( Figure 9A ) . We next attempted to define the function of γH2A that becomes critical in the absence of Rqh1 . Consistent with earlier studies [23] , we found that rqh1Δ and brc1Δ had strong negative genetic interactions , similar to those between rqh1Δ and htaAQ ( Figure 9A ) . Taken together , these data suggested that Brc1 binding to γH2A might be critical in the absence of Rqh1 activity . To address this possibility , we employed the brc1-T672A mutation of the BRCT5 domain of Brc1 , which through structural , biochemical and genetic studies was shown abolish Brc1 binding to γH2A and partially compromise Brc1 function [21] . Tetrad analysis revealed strong negative genetic interactions between rqh1Δ and brc1-T672A ( Figure 9B ) that were confirmed in dilutions assays ( Figure 9A ) . Microscopic examination of rqh1Δ htaAQ , rqh1Δ brc1-T672A , and rqh1Δ brc1Δ mutant cells revealed increased aberrant mitoses and chromosome segregation defects compared to parental strains ( Figure 9C ) , which suggests that Rqh1 is required for maintaining chromosome in the htaAQ and brc1-T672A mutant backgrounds . We also confirmed by ChIP analysis that γH2A levels are elevated at the rDNA in rqh1Δ mutants ( Figure 9E ) . From these data we conclude that the Rad3-γH2A-Brc1 pathway and Rqh1 DNA helicase work independently to maintain genome integrity .
We demonstrated that γH2A is formed at natural fork barriers such as RTS1 in the mating type locus and RFBs in the rDNA . We also showed that γH2A is linked to the regulation of barrier activity by the Swi1-Swi3 fork protection complex ( FPC ) . Deletion of Swi1 abolished γH2A formation at RTS1 in the MT locus , which indicates that the FPC triggers γH2A . Analysis of γH2A formation in the rDNA was more complicated than in the MT locus , since only three of the four rDNA barriers are dependent on the fork protection complex ( RFB1 , RFB2 and RFB3 ) but not RFP4 [51] , [54] . The first three are also dependent on specific DNA-binding proteins that constitute a barrier to the advancing replication fork , Reb1 at RFB2 and RFB3 , and Sap1 at RFB1 [50] , [54] . However , there is no known RFP4-binding protein and this barrier is likely caused by the collision of transcription and replication machineries at the highly transcribed rDNA locus [51] . Deletion of Swi1 increases fork stalling at RFP4 and leads to an increase in Rad22 foci , which suggests elevated levels of recombination [48] , [51] . Therefore , the high levels of γH2A observed at the rDNA in the absence of Swi1 may be associated with elevated recombination , suggesting that γH2A is a marker of recombination . This could also explain why we detected γH2A near the tDNA barriers only when Swi1 was deleted . Similar to RFP4 , the tDNAs are nonpolar fork barriers triggered by collisions between the replisome and transcription machinery [56] . The FPC suppresses recombination at the tDNAs and facilitates replisome progression through these sites [56] . The increase in γH2A at the tDNAs in swi1Δ cells again suggests that γH2A is associated with “hot-spots” of recombination . This observation is supported by our studies of γH2A formation at an ectopic RTS1 replication fork barrier inserted between two direct repeats of ade6 alleles [49] . We observed high levels of γH2A on the side where fork stalling occurs and recombination is initiated , which occurs in the absence of DNA DSBs[49] . Additionally , we observed an asymmetric distribution of γH2A at RTS1 , which contrasts the typical bimodal γH2A distribution found around DSBs [10] . Altogether , our data show that γH2A is triggered at natural replication fork barriers and suggests that phosphorylation of H2A is associated with recombination hotspots at stalled replication forks . The enrichment of γH2A in heterochromatin regions during DNA replication is the most intriguing outcome of our ChIP-on-chip analysis . We demonstrated that γH2A precisely colocalizes with Clr4-dependent H3K9me in the pericentromeric dg/dh repeats , the subtelomere , and IR elements flanking the MT silent region . We also detected γH2A in transcriptionally repressed subtelomeric regions and in retrotransposons , which are silenced by Clr3/Clr6-dependent histone deacetylation . Importantly , the formation of γH2A at the centromeres and in subtelomeric regions was dependent on the Clr4 methyltransferase , which mediates the crucial steps in heterochromatin establishment [80] . Our data suggests that either the active process of heterochromatin establishment or the resulting heterochromatin structure lead to replication fork pausing , stalling or collapse , which triggers recruitment of Rad3 and γH2A formation . Heterochromatin establishment occurs during DNA replication and involves assembly of multiple protein-DNA complexes , which could impede replisome progression [69] , [81] , [82] . Additionally , RNAi transcription occurs at the same time at the dg/dh repeats in the centromere and subtelomeres , which could induce collisions between the replisome and transcription machinery [81]–[83] . How replication fork progression is coordinated with heterochromatin formation is poorly understood . Since Swi6 facilitates early origin firing at the centromere dg/dh repeats [74] , heterochromatin is at least partially present when replication begins . Replication fork pausing has been detected in budding yeast centromeres [84] and structures suggestive of replication intermediates have been reported in fission yeast centromeres although no specific replication fork pause sites have been identified [85] . Replication fork pausing also occurs at the telomeres and is increased in the absence of Taz1 [76] . Thus , we were surprised that γH2A levels in the subtelomeres of taz1Δ mutants remained similar to wild type cells . We showed that γH2A forms in subtelomeric heterochromatin , and is rapidly depleted below background levels in the TAS region and telomere repeats , where Taz1 binds . This suggests that the transition from the subtelomere into the TAS is accompanied by a change in chromatin structure , which is consistent with reports that S . pombe telomeres may be non-nucleosomal or contain alternative nucleosomal configurations [75] , [77] . The DNA fragments containing stalled forks that were analyzed in taz1Δ mutants were located outside the subtelomeric γH2A domain [76] . Our data strongly suggests that γH2A formation in the subtelomeres is linked specifically to the presence of heterochromatin , independently of Taz1 . This is supported by observations that deletion of Taz1 does not change existing heterochromatin levels at the subtelomeres [77] and our data that γH2A formation is associated with Clr4 activity . There is accumulating evidence that replication of heterochromatin requires extra vigilance by genome maintenance factors [86] , [87] . Deletion of Rad3 destabilizes the centromeres and telomeres , resulting in increased gross chromosomal rearrangements [42] , [43] , [86] . We detected formation of γH2A at the subtelomeres in G2 phase in rad3Δ , which indicates unrepaired DNA damage , likely a result of telomere shortening through recombination [42] . At the centromeres γH2A formation temporally and physically coincides with recruitment of the Rad51 recombinase in early S-phase , where it is required for suppression of gross chromosomal rearrangements between centromere repeats [86] . The Smc5/6 complex , an essential factor for genome stability , also binds to heterochromatin regions during S-phase , where it is thought to suppress recombination [87] . Put together , formation of γH2A in heterochromatin coincides with DNA replication , and participates in either maintenance of replication fork stability , recombination , or repair of collapsed replication forks . This role may be partially conserved in mammals during replication of specialized chromatin domains such as the silent X chromosome , where γH2AX and BRCA1 foci form in S-phase [88] . A number of genetic studies support the idea that γH2A has a role in repair of replication-associated DNA damage , including our investigations of γH2A binding to Brc1 [21] . Brc1 binds to γH2A in response to endogenous replication stress and DNA damaging agents via its C-terminal BRCT domain . In this study we demonstrated that Brc1 is recruited by γH2A to the rDNA , centromeres , and telomeres . Since γH2A forms transiently during S phase , Brc1 recruitment suggests that these regions undergo DNA damage during replication . A majority of spontaneous γH2A-dependent Brc1 foci colocalize with the rDNA [21] and Brc1 is needed for viability in the absence of the Rqh1 helicase , which functions to maintain rDNA stability [23] . We showed that the rqh1Δ brc1-T672A mutant , in which Brc1 binding to γH2A is abolished , and the htaAQ rqh1Δ mutants have poor viability and severe chromosome segregation defects . Along with these genetic interactions , the increased γH2A levels in the rDNA of rqh1Δ mutants support the idea that γH2A may function in the maintenance of rDNA in a pathway mediated by Brc1 . Brc1 association with heterochromatin in the centromeres and telomeres indicates that the Rad3-γH2A-Brc1 pathway also functions to prevent DNA damage in heterochromatin . Interestingly , deletion of Brc1 is synthetic lethal with mutants of the Smc5/6 complex , which suppresses recombination in rDNA and heterochromatin [22] , [89]–[91] . It is noteworthy that the binding of Smc5/6 to centromeric heterochromatin is decreased in Clr4Δ cells [87] , similar to our observations of Brc1 . Our ideas are supported by studies of the putative Brc1 homolog in S . cerevisiae , Rtt107/Esc4 , which is involved in DNA repair at stalled replication forks , maintenance of rDNA stability , and chromatin silencing [23] , [92]–[96] . PTIP , a mammalian protein that is structurally related to Brc1 , has roles in both transcriptional regulation and maintenance of genomic stability [97] . The role of PTIP in the response to DNA damage is poorly understood , although there are indications that it participates in homologous recombination [98] . The recruitment of Brc1 to γH2A-sites during replication suggests that it and possibly Rtt107 or PTIP may function in maintaining genomic integrity in regions that are susceptible to replication stress , particularly the rDNA and heterochromatic loci . The recruitment of Brc1 to these sites is crucial in the absence of the Rqh1 RecQ helicase . It will be interesting to determine if γH2AX similarly recruits PTIP to fragile genomic sites in mammalian cells . The genome-wide distribution of γH2A in budding yeast was recently described [31] and allows us to compare genome protection mechanisms between the two yeast species . The patterns of γH2A formation are remarkably similar between these highly divergent organisms . Both studies identified γH2A domains at natural replication fork barriers , centromeres and telomeres , and repressed chromatin . This indicates that the functions of γH2A in genome stability in these regions are likely to be conserved in higher eukaryotes . Several γH2A sites found in budding yeast were absent in fission yeast , including tDNAs , LTRs , and replication origins . Fission yeast tDNAs are weak , nonpolar fork barriers [56] , and our data indicates that the Swi1-Swi3 complex mediates replication through these barriers in a manner that avoid triggering Rad3 activation , as evidenced by lack of γH2A at these sites . We identified γH2A near tDNAs only at the centromeres , where these elements are boundaries to heterochromatin spreading and regulate centromere functional organization in fission yeast [45] , [58] . Similarly , we observed association of γH2A with LTRs only where they flank Tf2 retrotransposons . These LTRs recruit CenpB-family proteins to assemble repressive structures called ‘Tf-bodies’[99] , which may impede replication fork progression and lead to γH2A formation . We did not observe γH2A loading at replication origins , where peaks of γH2A were detected in budding yeast . Fission yeast origins are more similar to mammalian cells as they are inefficient and lack strong regulation , whereas in budding yeast there are specific origins programmed to fire efficiently in S phase [100] . Due to the stochastic origin activity in fission yeast , the likelihood of fork stalling at a specific origin is low , and the accumulation of γH2A near origins may not be detectable in an average cell population . The most fascinating discovery of both studies is that γH2A associates with heterochromatic regions and repressed genes . This is interesting considering that the pathways of heterochromatin establishment in these yeasts are highly divergent [101] . In budding yeast gene silencing occurs primarily through histone deacetylation , but in S . pombe there are additional pathways that involve Clr4-dependent histone methylation , recruitment of HP-1 homologs , and RNAi transcription [101] , [102] . Heterochromatin structure of S . pombe is more similar to mammalian heterochromatin , particularly at the centromeres [101] . In S . pombe subtelomeres and centromeres γH2A forms transiently during DNA replication , but in S . cerevisiae γH2A is constitutively present in the subtelomeres , and transiently forms in the centromeres . Although there are indications that γH2AX has a role in chromatin silencing and chromosome segregation in other organisms [103] , [104] , neither yeast study was able to link γH2A to a heterochromatin-specific function . Put together , these data strongly suggest that repressed chromatin impedes replication fork progression , and that Rad3-γH2A activation may have a critical role in replication of heterochromatin regions in higher eukaryotes . During DNA replication the genome is very vulnerable to mutations and gross chromosomal rearrangements [1] . We demonstrated that genome-wide mapping of γH2A is a sensitive method for identifying regions of DNA replication stress . We also showed that γH2A may play a role in stabilizing replication forks stalled by natural impediments , such as RFBs , and that γH2A is linked to heterochromatin establishment during replication . γH2A-dependent Brc1 binding is a link between γH2A formation and mechanisms that ensure genome stability during DNA replication . Since γH2A modification in heterochromatin regions is conserved among several organisms , it is plausible that this modification has a yet-undiscovered role in heterochromatin function , which is difficult to detect due to redundant genetic pathways that compensate in its absence . ChIP-on-chip analysis of γH2A in a genetically tractable organism such as fission yeast will be a useful assay for studying genome-wide effects of mutations that impair DNA replication or checkpoint pathways .
Fission yeast strain genotypes are listed in Table S2 . Standard fission yeast methods were used as described previously [39] . Gene deletions and epitope-tagging were performed as described [105] . For synchronization of cells using cdc25-22 block and release , cells containing the temperature sensitive cdc25-22 allele were incubated at restrictive temperature ( 36°C ) for 4 hours to arrest the cell cycle in G2 [39] . Upon release to permissive temperature ( 25°C ) , the cells synchronously enter the cell cycle . Progression into S-phase was monitored microscopically by counting cells that contained septa , the appearance of which correlates with S-phase [39] . Cells were fixed for ChIP experiments when the septation index was between 60–80% . Ectopic expression of pRep41-N-GFP-brc1+ for microscopy and ChIP analysis was under the control of the thiamine-repressible nmt41 promoter . Induction of plasmid expression was performed in selective medium in the absence of thiamine for 18–20 hours . For γH2A and H2A Western blotting , histone-enriched protein extracts were prepared as previously described in [33] . Polyclonal anti-γH2A antibody ( courtesy of C . Redon ) was used for γH2A detection . Polyclonal anti-H2A antibody ( 07–146 , Millipore ) was used for H2A detection . Images were acquired on the Odyssey Infrared Imaging System ( LI-COR Biosciences ) . Cells were grown in YES liquid medium at 30°C , washed once in 1x PBS and imaged live for Brc1-GFP foci , or fixed overnight in cold 70% ethanol , washed once in 1x PBS , and mixed with 500 ug/ml DAPI for analysis of chromosome segregation defects . Cells were photographed using a Nikon Eclipse E800 microscope equipped with a Photometrix Quantix charge-coupled device camera . Error bars represent the standard deviation between 3 independent experiments , or range for 2 independent experiments . ChIP experiments were performed as described [20] with the following modifications . Cleared lysates were incubated with protein G Dynabeads ( Invitrogen ) pre-bound to anti-γH2A antibody ( Courtesy of C . Redon ) or anti-GFP antibody ( Roche ) . DNA was purified using the QIAquick PCR Purification Kit ( Quiagen ) . Quantitative PCR was performed on input and ChIP samples using the Chromo4 Real-Time PCR Detection System ( Bio-Rad ) using the iQ SYBR Green Supermix ( Bio-Rad ) . QPCR primers are listed in Table S3 with the exception of the subtelomere primers , which are from [70] and tDNA primers , which are from [87] . Percent of immunoprecipitated DNA ( %IP ) in the ChIP samples was calculated relative to the amount of DNA in the input samples . ChIP fold enrichment was calculated relative to actin . All error bars represent the standard error between experimental triplicates . Input and ChIP DNA samples were prepared using standard Affymetrix protocols and hybridized on Affymetrix S . pombe tiling arrays FR1 . 0 . See Supporting Information ( Text S1 ) for description of MAT score and other methods . | Eukaryotic genomes , which range in size from ∼107 to ∼1011 base pairs , are replicated with nearly absolute fidelity every cell cycle . This amazing feat happens despite the frequent stalling or collapse of replication forks . The checkpoint kinase ATR is activated by replication fork stalling and phosphorylates histone H2A in nucleosomes surrounding damaged DNA . As the genomic regions triggering ATR activation are largely unknown , we used a whole-genome microarray to map chromosomal domains enriched with phospho-H2A during DNA replication in fission yeast . This analysis identified specific sites , including natural replication fork barriers in ribosomal DNA repeats , retrotransposon elements , and most surprisingly , all heterochromatin regions . Phospho-H2A binds the genome maintenance protein Brc1 , and our genetic studies reveal that this molecular pathway becomes crucial in the absence of Rqh1 , a conserved DNA helicase that is linked to cancer predisposition . As the fission yeast and human genomes share many similarities , our study reveals genomic landmarks that could similarly trigger ATR activation in human cells and shows that phospho-H2A and Brc1 are a critical part of the network that maintains genome integrity during DNA replication . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"molecular",
"biology/dna",
"replication",
"molecular",
"biology/histone",
"modification",
"molecular",
"biology/recombination",
"molecular",
"biology/centromeres",
"molecular",
"biology/chromosome",
"structure",
"molecular",
"biology/dna",
"repair"
]
| 2010 | Rad3ATR Decorates Critical Chromosomal Domains with γH2A to Protect Genome Integrity during S-Phase in Fission Yeast |
Russell’s viper envenoming is a major problem in South Asia and causes venom induced consumption coagulopathy . This study aimed to investigate the kinetics and dynamics of venom and clotting function in Russell’s viper envenoming . In a prospective cohort of 146 patients with Russell’s viper envenoming , we measured venom concentrations , international normalised ratio [INR] , prothrombin time ( PT ) , activated partial thromboplastin time ( aPTT ) , coagulation factors I , II , V , VII , VIII , IX and X , and von Willebrand factor antigen . The median age was 39y ( 16–82y ) and 111 were male . The median peak INR was 6 . 8 ( interquartile range[IQR]:3 . 7 to >13 ) , associated with low fibrinogen [median , <0 . 01g/L;IQR:<0 . 01–0 . 9g/L ) , low factor V levels [median , <5%;IQR:<5–4%] , low factor VIII levels [median , 40%;IQR:12–79%] and low factor X levels [median , 48%;IQR:29–67%] . There were smaller reductions in factors II , IX and VII over time . All factors recovered over 48h post-antivenom . The median INR remained >3 at 6h post-antivenom but had reduced to <2 , by 24h . The aPTT had also returned to close to normal ( <50sec ) at 24h . Factor VII , VIII and IX levels were unusually high pre-antivenom , median peak concentrations of 393% , 307% and 468% respectively . Pre-antivenom venom concentrations and the INR ( r = 0 . 20 , p = 0 . 02 ) and aPTT ( r = 0 . 19 , p = 0 . 03 ) were correlated ( non-parametric Spearman analysis ) . Russell’s viper coagulopathy results in prolonged aPTT , INR , low fibrinogen , factors V , VIII and X which recover over 48h . Severity of clotting abnormalities was associated with venom concentrations .
Snake envenoming is a major health issue in the Asia-Pacific region with between 250 , 000 and 1 million cases occurring annually . [1] Russell’s viper ( Daboia russelii ) is one of the most medically important snakes in the region , [1 , 2] with bites causing death in 2 to 5% of cases , accounting for the majority of fatal snakebites in Sri Lanka[3] . Russell’s viper envenoming results in local effects , venom induced consumption coagulopathy ( VICC ) , mild neurotoxicity and renal injury . [3 , 4] VICC is the commonest systemic manifestation and in some cases results in mucosal bleedings and less commonly major bleeding including intracranial haemorrhage . [3 , 5–7] The in vitro procoagulant effects of Russell’s viper venom have been well characterised and a number of procoagulant toxins have been isolated and used in laboratory assays for decades . [8–10] Russell’s viper venom contains both factor X and factor V activators which trigger the clotting pathway early on , resulting in consumption of multiple clotting factors . [11–13] Previous studies have shown that VICC resulting from Russell’s viper envenoming causes overall haemostatic disturbances which manifest in prolonged prothrombin time ( PT ) /international normalised ratio ( INR ) and activated partial thromboplastin time ( aPTT ) , as well as decreased levels of fibrinogen , factor V and factor X and elevated D-Dimer concentrations . [14–17] However , there is limited information on the dynamics of clotting factor levels in Russell’s viper envenoming and their response to antivenom treatment . A number of studies have measured venom concentrations in patients with Russell’s viper envenoming , and some have suggested that there is recurrence of venom post-antivenom . [5 , 18] However , there is limited information on the dynamic relationship between venom concentrations and clotting factor levels , and whether the detection of venom post-antivenom is associated with further clotting factor consumption . This study aimed to explore the dynamic changes in venom and clotting factor levels in VICC , including the effect of antivenom treatment following Russell’s viper envenoming .
All patients ( >13 years old ) admitted with a suspected or definite snake bite between January 2007 and July 2009 were identified on admission to hospital . From these only definite cases were included if Russell’s viper venom was detected with a venom-specific enzyme immunoassay ( EIA ) . All admission samples were tested for Russell’s viper venom . VICC was defined as coagulopathy ( abnormal PT/INR ) with evidence of consumption ( low/undetectable fibrinogen or elevated D-Dimer more than 10 times the upper limit of normal ) and an INR > 1 . 5 . [20] Severe or complete VICC was defined as an INR > 13 ( unrecordable ) . Baseline data , including demographic data ( age and sex ) , information on the snake bite ( snake type , time of bite ) , clinical effects ( local effects: local pain , swelling , bruising , blistering and necrosis; systemic effects: features of coagulopathy including bleeding and neurotoxicity ) and antivenom treatment ( dose and time of administration ) were recorded prospectively for all patients . Research blood samples were collected from all patients on admission and then at regular time intervals during their admission . Blood was collected in citrated tubes for clotting times and coagulation studies and in serum tubes for venom-specific EIA . All samples were immediately centrifuged , aliquoted and frozen initially at -20°C , and then transferred to a -80°C freezer within 2 weeks of collection . All patients received Indian polyvalent snake antivenom manufactured by VINS Bioproducts Limited ( batch number: ASV 42C/06 , 1030 ) or BHARAT Serum and Vaccines Limited , India ( batch number: 5346KD4 , LY 55/05 , LY 32/04 , A5307035 ) . A sandwich EIA was used to measure Russell’s viper venom in serum samples and has previously been described . [7 , 21 , 22] In brief , polyclonal IgG antibodies were raised against Russell’s viper ( D . russelii ) venom in rabbits[23] . Antibodies were bound to microplates as well as being conjugated to biotin for a sandwich EIA with the detecting agent being streptavidin-horseradish peroxidase . All samples were measured in triplicate , and the averaged absorbance converted to a concentration by comparison with a standard curve based on serial dilutions of venom using a sigmoidal curve . The assay does not cross-react with Hypnale venom , the only other medically important snake in Sri Lanka that cause coagulopathy ( excepting Echis carinatus—saw-scaled viper—in the north ) . [22] Frozen citrated plasma samples were used for all clotting times and clotting factor studies including prothrombin time ( PT ) /international normalised ratio [INR] and activated partial thromboplastin time ( aPTT ) . Levels of factors I ( fibrinogen ) , II ( prothrombin ) , V , VII , VIII , IX and X , von Willebrand factor antigen ( VWF:Ag ) and D-Dimer were all measured . All assays were done using either standard coagulometric or immunoturbidimetric methods as provided by the manufacturer and were performed on a Behring Coagulation System ( BCS ) or Sysmex CA-1500 analyzer ( Dade Behring , Marburg , Germany ) [24] . Individual clotting factor levels were determined by mixing patient plasma with plasma deficient in the factor being measured and the time for clot formation measured in seconds . The amount of factor present in the sample was quantified by comparing with a standard or reference curve produced using serial dilutions of plasma deficient in the factor mixed with normal plasma , against the clotting time . The quantification of von Willebrand factor antigen ( VWF:Ag ) and D-Dimer was done using immunoturbidometric methods . The average number of samples collected from the patients was four ( range 1 to 10 ) . To describe the peak effect of the venom on the clotting pathway , the maximal ( longest ) clotting time or minimum ( lowest ) factor level was determined for each patient . Factor levels and clotting times were then reported as medians and interquartile ranges of the maximum ( PT/INR , aPTT , D-Dimer ) or minimum ( Fibrinogen , Factors II , V , VIII , IX , X ) for each level over the time course of the admission . For visual analysis of concentration time data , median factor concentrations were plotted versus time to provide empirical estimates of the average/median changes over time in the coagulation studies and the factor levels . Time zero was defined as the time of antivenom administration . This was done by binning the data based on the time post-snake antivenom and then calculating the median factor level/clotting time for each bin and the median time for each bin ( ie . for the specified time period ) . The median factor level/clotting time was then plotted versus the median time . For factors VII , VIII and IX , the median and interquartile range of the levels of these factors was taken from the bin where the peak occurred because of the unusually high pre-antivenom levels of these three factors . Separate plots were also made for complete VICC where the INR was unrecordable ( INR > 13 ) and partial VICC where the INR was abnormal but still recordable ( 1 . 5 < INR < 13 ) . To investigate whether there was an association between the venom load ( i . e . amount of venom delivered by the snake ) and the severity of VICC , correlations between pre-venom concentrations and the PT/INR , aPTT and all clotting factor levels were tested with non-parametric Spearman correlation analysis . In addition clotting tests and factor studies versus pre-antivenom venom concentrations were plotted ( with the line of best fit and 95% confidence intervals ) . All analyses and graphics were done in GraphPad Prism version 6 . 01 for Windows , GraphPad Software , San Diego California USA , www . graphpad . com .
There were 146 patients with Russell’s viper bites . The median age was 39 years ( Range: 16 to 82 years ) and 111 ( 76% ) were male . All patients had VICC ( low fibrinogen and elevated PT ) and 70 ( 48% ) had neurotoxicity . Local effects were reported in 134 ( 91% ) patients and systemic bleeding developed in 14 ( 10% ) patients . The median pre-antivenom venom concentration was 201ng/ml ( IQR: 74 to 435ng/ml; Range: 1 to 1521ng/ml ) which dropped to a median concentration of 2ng/ml ( IQR: 0 to 9ng/ml ) after the administration of antivenom . The median peak INR in the patients was elevated at 6 . 8 ( IQR: 3 . 7 to >13 ) as was the median peak aPTT of >180sec ( IQR: 91 . 3 to > 180sec ) . The abnormal clotting times were associated with a low fibrinogen [median , <0 . 01g/L; IQR: <0 . 01 to 0 . 9g/L] , low factor V levels [median <5%; IQR: <5 to 4%] , low factor VIII levels [median; 24%; IQR: 10 to 41%] and mildly decreased factor X levels [median 48%; IQR: 29 to 67%] over the course of the patient admissions . The median of the highest or median of the lowest factor concentrations/clotting times are given in Table 1 . There were smaller reductions in factors II , VII and IX over the course of the patient admission . The INR , fibrinogen , factors V and X recovered over 48 hours post-antivenom ( Figs 1 and 2 ) . The median INR remained greater than 3 at 6 hours post-antivenom , but had reduced to less than 2 , by 24 hours ( Fig 1 ) . The aPTT had also returned to close to normal ( < 50sec ) at 24 hours ( Fig 1 ) . There were smaller reductions in most factor levels in partial versus complete VICC ( S1 Fig ) . Factors VII , VIII and IX levels were very high prior to antivenom and then dropped dramatically into the normal range or to low levels ( FVIII ) post-antivenom ( Fig 2 ) . The median peak factor VII levels were 393% ( IQR: 85 to 698% ) , factor VIII levels were 307% ( IQR: 160 to 400% ) and factor IX levels were 468% ( IQR: 331 to 704% ) respectively ( Fig 2 ) . There was a statistical association between pre-antivenom venom concentrations and the INR ( r = 0 . 20 , p = 0 . 02 ) , aPTT ( r = 0 . 19 , p = 0 . 03 ) and factor IX ( r = -0 . 36 , p<0 . 001 ) , and there were trends for factor V ( r = -0 . 17 , p = 0 . 058 ) , factor X ( r = -0 . 17 , p = 0 . 05 ) and VWF:Ag ( r = -0 . 18 , p = 0 . 053 ) ( S2 Fig ) . There were no statistical associations for pre-antivenom venom concentrations and fibrinogen , factors II , VII , VIII and D-Dimer ( S2 Fig ) .
The study shows that VICC resulting from Russell’s viper envenoming is characterised by an elevated INR and aPTT associated with low fibrinogen , factor V , VIII and X levels . There was an association between the pre-antivenom venom concentrations and the severity of the coagulopathy , mainly with the INR and aPTT . The coagulopathy resolved over a period of 48 hours after the administration of antivenom consistent with VICC from other snakes . [24 , 25] An unusual finding was the very high levels ( above the normal range ) of factor VII , VIII and IX prior to antivenom treatment which then returned to normal ranges soon after antivenom treatment . It is unclear the exact reason for these high factor levels but may be related to venom activity in the sample in vitro . Russell’s viper venom contains factor V and factor X activators which convert these factors to their activated forms ( i . e . Va and Xa ) , explaining the low factor V and factor X levels in human envenoming ( Fig 3 ) . [11 , 12 , 26 , 27] Activation of factor V and X results in the formation of the prothrombinase complex ( XaVa ) which activates the whole clotting cascade by converting prothrombin to thrombin . This then leads to the consumption of fibrinogen , factor VIII and further consumption of factor V . The multiple factor deficiencies result in the prolonged INR and aPTT , and activation of the clotting cascade leads to increased fibrinolysis and therefore elevated D-Dimer levels . A similar pattern of factor deficiencies has been described in previous studies of patients envenomed by Russell’s vipers . [14–17] Four studies showed the initial drop in fibrinogen to low levels followed by a recovery over 24 to 48 hours . [14–17] Two studies also reported serial measurements of factor V and factor X , with initial low levels that recovered over 24 hours consistent with our study . [15 , 16] The recovery of the coagulopathy occurred over a period of 48 hours , although the rate of recovery differed for each of fibrinogen , factor V and factor VIII . The median INR was still greater than 3 at 6 hours , suggesting that 6 hours if too early to determine the effect of antivenom ( Fig 1 ) . The median INR was 2 . 3 at 12 hours and then less than 2 at 24 hours suggesting that patients had only a mild coagulopathy at this time ( Fig 1 ) . The presence of very higher factor VII , VIII and IX levels ( 392% , 313% and 463% respectively; Fig 2 ) prior to antivenom is an unexpected finding , although earlier studies have reported high values in a small number of patients . [15 , 16] A possible explanation for this finding is that the presence of Russell’s viper venom ( RVV ) factor X activator toxin in the sample results in falsely high factor levels . The in vitro activity of the toxin would appear similar to the activity of factor VII , VIII and IX . For example , in the case of factor VII , the presence of active RVV in the sample will result in a shorter clotting time because the RVV factor X activator has the same activity as factor VII ( i . e . both VII:TF and RVV factor X activator convert factor X to Xa ) . Interpolation of this shorter clotting time on the standard curve for factor VII results in a factor VII level greater than 100% ( S3 Fig ) . A similar phenomena occurs with both factor VIII and factor IX because the VIIIa:IXa complex also activates factor X to Xa ( intrinsic pathway ) . Previous studies that have measured factor VII , VIII and IX levels have also found normal to high values , but because they have only measured factors at one time point , they are difficult to interpret . [14 , 16] The time course of these three factors , with high levels only occurring in the pre-antivenom samples , and then normal factor VII levels or low factor VIII and IX levels post-antivenom , also suggests that these high levels prior to antivenom are due to the assay and not VICC ( Fig 2 ) . There was only moderate reduction in factor II , and moderate reductions in factors VII and IX after the initial high pre-antivenom values of the latter two . This is most likely because none of these factors are directly activated by the venom ( as are factors V and X ) , or are factors that are completely consumed when the clotting pathway is completely activated ( factor VIII ) . There is a large excess of factor II ( prothrombin ) , so even if all the fibrinogen is converted to fibrin ( consumed ) , prothrombin levels will only partially decrease . Factor VII and IX are also not activated and consumed in VICC . VWF:Ag levels were mildly elevated which is most likely an indirect consequence of the clotting pathway being activated . There was a rebound in venom concentrations approximately 40 hours after antivenom ( Figs 1 and 2 ) which we have previously shown to not be associated with recurrent coagulopathy . [28] A recent study has shown that this re-appearance of venom is due to bound venom being detected by the venom specific EIA . [29] This is entirely consistent with the results in this study . There was recovery of all clotting factors and clotting times despite this rebound in measured venom This further supports that venom-specific EIA is measuring bound venom post-antivenom . The study found a correlation between venom concentrations and the severity of the coagulopathy measured by the INR and aPTT . An earlier study by Than et al found a similar association between venom concentrations and the severity of the coagulopathy . [15] There was an association between venom concentrations and factor V and Factor X , consistent with the Factor V activator and Factor X activator toxins in the venom . There are a number of limitations of the study . Coagulation studies are best done on samples collected fresh from the patient . This was not possible in this study because of the resource limitations in the hospital where the patients were treated . However , all samples were immediately centrifuged , aliquoted and frozen to preserve the functional activity of the clotting factors . The results were in keeping with another study of VICC in Australian elapids which used a similar approach successfully . [24]Platelet counts , platelet function testing and tissue factor assays were not done in this study . These may play a role in haemostatic disturbances , particularly platelets , and should be investigated in future studies . Another limitation was that the decision to give antivenom was made by the treating clinician and not the investigators . However , the majority of these patients were recruited to a randomised controlled trial comparing two different infusion rates of antivenom . This meant that all patients had a standardised antivenom administration . | Snake envenoming is an important health issue in many parts of the world and coagulopathy is one of the commonest manifestations . Russell’s viper envenoming occurs throughout south-east and south Asia and most commonly causes a venom induced consumption coagulopathy which may be complicated by bleeding . There is a limited understanding of the time course of the coagulopathy , including the specific clotting factor deficiencies and their relationship to the time course of venom concentrations . In this study , we measure the venom concentrations and clotting factor concentrations throughout the course of the patient admission . The study confirms that Russell’s viper envenoming causes prolonged clotting times and deficiencies of factors V , X , VIII and fibrinogen which recover over 2 days post-antivenom . In addition , the prothrombin time and activated partial thromboplastin times correlated with venom concentrations . Factor VII , VIII and IX levels were very high and may be related to venom activity , although further research is required . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | Venom Concentrations and Clotting Factor Levels in a Prospective Cohort of Russell’s Viper Bites with Coagulopathy |
IFI16 ( gamma-interferon-inducible protein 16 ) , a predominantly nuclear protein involved in transcriptional regulation , also functions as an innate immune response DNA sensor and induces the IL-1β and antiviral type-1 interferon-β ( IFN-β ) cytokines . We have shown that IFI16 , in association with BRCA1 , functions as a sequence independent nuclear sensor of episomal dsDNA genomes of KSHV , EBV and HSV-1 . Recognition of these herpesvirus genomes resulted in IFI16 acetylation , BRCA1-IFI16-ASC-procaspase-1 inflammasome formation , cytoplasmic translocation , and IL-1β generation . Acetylated IFI16 also interacted with cytoplasmic STING and induced IFN-β . However , the identity of IFI16 associated nuclear proteins involved in STING activation and the mechanism is not known . Mass spectrometry of proteins precipitated by anti-IFI16 antibodies from uninfected endothelial cell nuclear lysate revealed that histone H2B interacts with IFI16 . Single and double proximity ligation microscopy , immunoprecipitation , EdU-genome labeled virus infection , and chromatin immunoprecipitation studies demonstrated that H2B is associated with IFI16 and BRCA1 in the nucleus in physiological conditions . De novo KSHV and HSV-1 infection as well as latent KSHV and EBV infection induces the cytoplasmic distribution of H2B-IFI16 , H2B-BRCA1 and IFI16-ASC complexes . Vaccinia virus ( dsDNA ) cytoplasmic replication didn’t induce the redistribution of nuclear H2B-IFI16 or H2B into the cytoplasm . H2B is critical in KSHV and HSV-1 genome recognition by IFI16 during de novo infection . Viral genome sensing by IFI16-H2B-BRCA1 leads to BRCA1 dependent recruitment of p300 , and acetylation of H2B and IFI16 . BRCA1 knockdown or inhibition of p300 abrogated the acetylation of H2B-IFI16 or H2B . Ran-GTP protein mediated the translocation of acetylated H2B and IFI16 to the cytoplasm along with BRCA1 that is independent of IFI16-ASC inflammasome . ASC knockdown didn’t affect the acetylation of H2B , its cytoplasmic transportation , and the association of STING with IFI16 and H2B during KSHV infection . Absence of H2B didn’t affect IFI16-ASC association and cytoplasmic distribution and thus demonstrating that IFI16-H2B complex is independent of IFI16-ASC-procaspase-1-inflammasome complex formed during infection . The H2B-IFI16-BRCA1 complex interacted with cGAS and STING in the cytoplasm leading to TBK1 and IRF3 phosphorylation , nuclear translocation of pIRF3 and IFN-β production . Silencing of H2B , cGAS and STING inhibited IFN-β induction but not IL-1β secretion , and cGAMP activity is significantly reduced by H2B and IFI16 knockdown during infection . Silencing of ASC inhibited IL-1β secretion but not IFN-β secretion during de novo KSHV and HSV-1 infection . These studies identify H2B as an innate nuclear sensor mediating a novel extra chromosomal function , and reveal that two IFI16 complexes mediate KSHV and HSV-1 genome recognition responses , with recognition by the IFI16-BRCA1-H2B complex resulting in IFN-β responses and recognition by IFI16-BRCA1 resulting in inflammasome responses .
RNA and DNA genomes of viruses are recognized by several host innate immune response sensors in different subcellular locations , resulting in antiviral responses of type 1 interferon ( IFN ) and inflammasome activation [1] . We have shown that IFI16 ( interferon inducible protein 16 ) , a resident nuclear protein involved in transcriptional regulation by an unknown mechanism , also functions as a nuclear sensor of innate immune inflammasome and IFN-β responses [2–5] . IFI16 detects the nuclear replicating episomal herpesvirus genomes of Kaposi's sarcoma-associated herpesvirus ( KSHV ) , Epstein-Barr virus ( EBV ) , and herpes simplex virus type-1 ( HSV-1 ) . This leads to IFI16-ASC-procaspase-1 inflammasome formation in the nucleus , which is transported to the cytoplasm leading into caspase-1 activation and pro-IL-1β/IL-18 cleavages [2–6] . We and others have also shown that independent of ASC , KSHV and HSV-1 genome recognition results in IFI16 interaction with STING in the cytoplasm , phosphorylation and nuclear translocation of IRF3 , IFN gene expression and IFN-β production [1 , 4 , 6–8] . Our recent studies have shown that BRCA1 , a DNA damage response ( DDR ) sensor and transcription regulator , is in complex with IFI16 in the uninfected cell nucleus . This BRCA1-IFI16 interaction increased during de novo KSHV , EBV and HSV-1 infection and in cells latently infected with KSHV and EBV , but not by bleomycin induced DDR or by cytoplasmic dsDNA vaccinia virus replication [1] . BRCA1 is a constituent of the genome recognition triggered IFI16-inflammasome that translocates to the cytoplasm . IFI16’s recognition of KSHV and HSV-1 genomes was inhibited without BRCA1 , demonstrating that sensing of viral DNA by IFI16 depends on its pre-existing complex with BRCA1 . In the absence of BRCA1 , the consequences of viral genome sensing , such as IFI16-inflammasome assembly , cytoplasmic localization , IL-1β production , and cytoplasmic IFI16-STING interaction , pIRF3 and IFN-β induction , were inhibited [1] . Our studies have also revealed that IFI16 undergoes acetylation by p300 ( histone acetyl transferase ) facilitating IFI16-ASC-procaspase-1 association , cytoplasmic translocation via Ran-GTP resulting in IL-1β production , and interaction with STING and IFN-β induction [9] . Leptomycin B treatment abrogated acetylated IFI16 translocation to the cytoplasm . ASC and STING knockdowns did not affect IFI16 acetylation demonstrating that this modification is upstream of inflammasome assembly and STING activation [9] . p300 inhibitor C646 or knockdown of p300 did not inhibit the association of IFI16 with KSHV and HSV-1 genomes signifying that increased nuclear acetylation of IFI16 is a post-nuclear genome recognition event that is common to IFI16-mediated inflammasome and IFN-β induction during KSHV , EBV , and HSV-1 infections [9] . Although IFI16 is the primary nuclear DNA sensor during HSV-1 infection , IFI16 interactions with cGAS ( cGAMP-Synthase ) and stabilization of IFI16 by cGAS have been reported [10] . cGAS senses cytosolic DNA leading to the production of second messenger cGAMP which activates STING to stimulate IFN-β production [11–14] . DNA damage induced leakage of self DNA into the cytoplasm has been shown to activate the IFN-β pathway [15] , while extra-chromosomal cytoplasmic histone H2B is suggested to be involved in aberrant self-or non-self-dsDNA recognition and induction of IFN-β [16] . We observed earlier that IFN-β was induced in the absence of ASC [6] and acetylated IFI16 was still detected in the cytoplasm of KSHV infected cells although the total and acetylated IFI16 levels were reduced by >3-fold compared to the levels in the presence of ASC [9] . We hypothesized that the cytoplasmic redistribution of IFI16 in ASC knockdown cells must be an inflammasome independent event which might be attributed to cytoplasmic export of acetylated IFI16 either alone or in complex with other nuclear proteins resulting in the activation of STING . To identify the IFI16 associated nuclear proteins involved in STING activation , uninfected endothelial cell nuclear lysate was precipitated by anti-IFI16 antibodies . Mass spectrometry ( MS ) of specific protein bands revealed that histone H2B was interacting with IFI16 . Here , we demonstrate that H2B is an essential component of the post-KSHV and HSV-1 genome recognition induced IFI16-mediated IFN-β production . Viral genome recognition by IFI16 led to the BRCA1 dependent p300-IFI16 interaction , acetylation of H2B and IFI16 in the nucleus , and their export via Ran-GTP . H2B-IFI16 along with BRCA1 interacted with cGAS and STING in the cytoplasm , resulting in pIRF3 induction and IFN-β production . Knockdown of H2B impaired the IFI16-mediated IFN-β response during KSHV and HSV-1 de novo infection and did not affect IFI16-inflammasome induction . Cytoplasmic distribution of H2B-IFI16 is also observed in cells latently infected with EBV . Collectively , these studies demonstrate that H2B is a crucial component in herpesvirus nuclear genome sensing by IFI16 and in the consequent innate IFN-β response .
To determine the identity of IFI16 associated nuclear proteins potentially involved in cytoplasmic activation of STING in the absence of ASC , nuclear lysates from uninfected primary human microvascular dermal endothelial ( HMVEC-d ) cells were IP-ed with anti-IFI16 antibodies and specific bands were analyzed by MS . Among the identified IFI16 interacting proteins , histone H2B ( ~13 kDa ) had the highest PEAKS score and coverage ( S1 Table ) . This interested us since apart from its epigenetic roles in the nucleus , extrachromosomal cytoplasmic histone H2B has been shown to be involved in the induction of IFN-β against small DNA fragments [17] . However , how H2B translocated into the cytoplasm and whether H2B plays a role in innate responses against nuclear genomes of herpes viruses were not known . To validate the MS data , cytoplasmic and nuclear fractions from uninfected human B ( BJAB-lymphoma ) , endothelial ( HMVEC-d ) and fibroblast ( HFF ) cells were IP-ed with anti-IFI16 and H2B antibodies and western blotted for various proteins . TATA-binding protein ( TBP ) and tubulin showed the purity of nuclear and cytoplasmic fractions , respectively , and the expression of IFI16 , H2B , BRCA1 , H2A and ASC proteins are shown by the input controls in Fig 1E and 1F . Results revealed the interaction of IFI16 and H2B only in the nuclear fractions in all three cell types examined ( Fig 1A–1D ) . We have previously demonstrated that IFI16 interacts with BRCA1 in the uninfected cell nucleus [1] . Interestingly , we also observed the interaction of H2B with BRCA1 only in the nuclear fractions of HMVEC-d , HFF and BJAB cells ( Fig 1B and 1D ) . Interactions between IFI16 and BRCA1 and between H2B and H2A were observed only in the nuclear fractions which served as positive controls ( Fig 1A–1D ) . In contrast , no apparent interactions of IFI16 with ASC and H2A or H2B with ASC were observed ( Fig 1A–1D ) . We performed PLA in uninfected BJAB , HMVEC-d and HFF cells as PLA detects an endogenous individual protein or interaction of two proteins based on the principle that if two epitopes or proteins are within the proximity of 40 nm or below , the PLA oligo probes linked to two secondary antibodies bound to primary antibody-antigen complexes can be amplified to give a PLA signal visualized as a fluorescent dot . PLA results demonstrated the close association ( interaction ) of IFI16 with H2B and BRCA1 and between H2B and BRCA1 only in the nucleus of uninfected cells ( Fig 1G–1J , red arrows ) . H2B-H2A association and IFI16-ASC association were used as PLA controls , and quantitation of the average dots per cell are shown in S1A–S1D Fig . We did not observe any association between IFI16 and H2A and between H2B and ASC ( S1E and S1F Fig ) . Primary , secondary or IgG control antibodies used to ascertain the specificity of PLA reactions did not show any amplified dots ( S2A Fig ) . In addition , IFA results also supported the association of IFI16 and H2B only in the nucleus of uninfected BJAB and HMVEC-d cells ( S2B Fig ) . These results validated our MS data and demonstrated the association of IFI16 with H2B in the nucleus of uninfected cells . Interaction of IFI16 with ASC and procaspase-1 along with BRCA1 results in inflammasome responses during KSHV , EBV and HSV-1 de novo infection and in cells carrying latent KSHV and EBV genomes [1–6] . Since we observed IFI16-H2B interaction in the nucleus of uninfected cells , we determined whether this interaction has any role in the inflammasome and IFN-β responses . Nuclear and cytoplasmic fractions from uninfected and KSHV infected HMVEC-d cells were IP-ed with anti-IFI16 or anti-H2B antibodies . Immunoblot results revealed an IFI16 and H2B interaction in the nuclear extracts of both uninfected and infected cells ( Fig 2A and 2B , lanes 1–5 ) . Interestingly , the IFI16-H2B interaction was also observed in the cytoplasmic extracts of cells infected with KSHV for 2 , 4 , and 12 h , which was reduced at 24 h p . i . ( Fig 2C and 2D , lanes 2–5 ) . In contrast , we observed little or no interaction in the cytoplasm of uninfected cells ( Fig 2C and 2D , lane 1 ) . IFI16 interacted with BRCA1 and ASC but not with H2A , and similarly , H2B interacted with BRCA1 and H2A but not with ASC in the nuclear extracts ( Fig 2A and 2B ) . Moreover , in the cytoplasmic extracts of infected cells and not in uninfected cells , we observed the interactions of IFI16 with ASC and BRCA1 but not with H2A . Similarly , H2B was IP-ed with BRCA1 but not with H2A and ASC in the cytoplasm ( Fig 2C and 2D ) . The expression levels of the proteins in nuclear and cytoplasmic extracts demonstrated that KSHV infection did not alter their levels ( Fig 2E and 2F , input controls ) . To confirm IFI16-H2B complex association and redistribution , uninfected and KSHV ( 4 h ) infected cells were subjected to PLA using anti-IFI16 and anti-H2B antibodies . We observed substantial IFI16-H2B PLA spots in the nucleus of uninfected and KSHV infected cells ( Fig 2G , red arrows; S2C Fig and Fig 2H ) . In contrast , significant levels of IFI16-H2B complex PLA spots were observed only in the cytoplasm of infected cells ( Fig 2G , yellow arrows and S2C Fig ) . When PLA and IFA were performed to assess the IFI16-H2B complex at various times of infection , we observed increased association of IFI16 and H2B in the cytoplasm of infected cells from 2 to 12 h p . i . which was reduced at 24 h p . i . ( Fig 2H , yellow arrows and S2D and S2E Fig ) . In cells infected with vaccinia virus replicating its dsDNA in the cytoplasm of infected cells , we have demonstrated the activation of the cytoplasmic AIM2-ASC inflammasome and not the IFI16-ASC inflammasome , as well as no significant increase in the IFI16-BRCA1 interactions and the absence of IFI16-BRCA1 in the cytoplasm [1 , 2] . When we examined the H2B-IFI16 in vaccinia virus infected cells ( 5 pfu/cell; 4h ) , we did not observe any significant redistribution of nuclear H2B-IFI16 or H2B into the cytoplasm ( S2G and S2H Fig ) . In addition , we did not observe any significant association of IFI16 and H2A by PLA ( S2F Fig ) . As demonstrated before [1 , 3 , 5] , PLA and IFA results showed increased IFI16-ASC association PLA spots in the cytoplasm of infected cells at 2 , 12 and 24 h p . i . which served as positive controls ( S3A , S3B and S3C Fig ) . These results demonstrated that IFI16 and H2B , and H2B and BRCA1 associate in the nucleus of uninfected and KSHV infected cells which redistribute to the cytoplasm only after infection and suggest that the presence of nuclear KSHV DNA is necessary for the H2B-IFI16 and H2B-BRCA1 cytoplasmic localization . To determine whether H2B-IFI16 and H2B-BRCA1 interactions and their cytoplasmic distributions observed during de novo KSHV infected cells ( Fig 2A–2H ) also occur during other herpesvirus infections , we examined these interactions during HSV-1 de novo infection . Uninfected and HSV-1 ( KOS ) infected HFF cells ( 1 pfu/cell; 4 h ) were subjected to a PLA reaction using anti-H2B , IFI16 and BRCA1 antibodies . We selected the 4 h time point as we have shown that beyond 4 h , IFI16 is not detected in the infected HFF cells as it is targeted and degraded by HSV-1 immediate early E3 ligase ICP 0 protein [1 , 4 , 6] . We observed the associations of H2B with IFI16 , H2B with BRCA1 and the associations of IFI16 with BRCA1 in the nucleus of uninfected and HSV-1 infected cells ( Fig 2I , 2J and 2K; red arrows ) . In contrast , significant associations of H2B with IFI16 , H2B with BRCA1 and IFI16 with BRCA1 were observed only in the cytoplasm of infected cells ( Fig 2I , 2J and 2K; yellow arrows ) . As demonstrated by us before [1] , PLA spots indicating the IFI16-BRCA1 association and their cytoplasmic distribution served as a positive control ( Fig 2K ) . These observations demonstrated that similar to KSHV , HSV-1 infection and the presence of nuclear viral genome also induces the redistribution of H2B-IFI16 and H2B- BRCA1 complexes to the cytoplasm of infected cells . We have shown constitutive IFI16-ASC-procaspase-1 inflammasome activation in association with BRCA1 in cells carrying multiple copies of latent KSHV episomal DNA and colocalization of IFI16 with the nuclear viral genomes [1 , 3] . Since we observed IFI16 and H2B interaction in the nucleus of uninfected B ( BJAB ) cells ( Fig 1A , 1B and 1G ) , we next determined their association and distribution during KSHV latent infection in B cells . Cytoplasmic and nuclear extracts from uninfected BJAB and KSHV latently infected BCBL-1 cells were IP-ed using anti-IFI16 and anti-H2B antibodies . Western blot analysis revealed the IFI16 and H2B interaction in the nucleus of BJAB and BCBL-1 cells , and interaction in the cytoplasm of BCBL-1 cells but not in BJAB cells ( Fig 3A and 3B ) . As demonstrated before , IFI16 interacted with BRCA1 both in the nucleus and cytoplasm of BCBL-1 cells but only in the nucleus of BJAB cells . These , along with H2B interactions with H2A in the nucleus of BJAB and BCBL-1 cells , served as positive controls , while little or no observed association between IFI16 and H2A and no interaction between H2B and ASC served as negative controls ( Fig 3A and 3B ) . Interestingly , we also observed the H2B interaction with BRCA1 in the cytoplasm of BCBL-1 cells but not in BJAB cells ( Fig 3B ) . Expression levels of these proteins in these cells are shown in Fig 3C , input controls . To further confirm the association and redistribution of IFI16-H2B , BJAB and BCBL-1 cells were tested by PLA using anti-IFI16 and anti-H2B antibodies . We observed considerable IFI16-H2B association PLA spots in the nucleus of BJAB and BCBL-1 cells ( Fig 3D , white arrows ) . In contrast , significant redistribution of IFI16-H2B PLA spots was observed only in the cytoplasm of BCBL-1 cells ( Fig 3D , red arrows and Fig 3E ) . Unlike the IP reactions in Fig 1 , we also observed a moderate increase in IFI16 and H2A association in the nucleus of infected BCBL-1 cells compared to BJAB cells ( Fig 3F and 3G ) . This could be due to the sensitivity of the PLA reaction detecting interactions that were either very weak or probably lost during IP-reactions . Nevertheless , these findings demonstrated that similar to de novo infection , latent KSHV infection induces the redistribution of the IFI16-H2B complex to the cytoplasm . Our previous studies have demonstrated the constitutive activation of IFI16 inflammasomes in association with BRCA1 in cells carrying multiple copies of latent EBV episomal DNA [1 , 5] , as well as the colocalization of nuclear IFI16 with the EBV genomes [5] . To determine whether H2B-IFI16 associate during EBV latent infection , EBV ( - ) BJAB , and EBV ( + ) LCL ( latency III ) and EBV ( + ) Akata ( latency I ) cells were subjected to IFA using anti-IFI16 , H2B and BRCA1 antibodies . H2B-IFI16 colocalization was observed in the nucleus of uninfected BJAB cells ( S4A Fig , white arrow ) . In contrast , significant colocalization of H2B-IFI16 was observed only in the cytoplasm of EBV+ LCL and Akata cells ( S4A Fig; red arrows ) . Similar results were also observed for H2B-BRCA1 colocalization ( S4B Fig ) as well as for IFI16-BRCA1 colocalization used as a positive control [1] ( S4C Fig ) . Collectively , these observations suggest that latent EBV infection induces H2B-IFI16 and H2B-BRCA1 redistribution to the cytoplasm . During de novo KSHV infection , viral genome recognition by nuclear IFI16 led into its acetylation by p300 and transport of acetylated IFI16 to the cytoplasm via Ran-GTPase [9] . Acetylation of H2B in the nucleus is one of the post-translational modifications essential for its function , such as interaction with DNA and proteins , transcription and chromatin remodeling [18 , 19] . We next determined whether IFI16 associated H2B is also acetylated during infection to aid in cytoplasmic translocation . PLA reactions were performed using combinations of anti-acetyl lysine , H2B and IFI16 ( rabbit or mouse ) antibodies , and a non-toxic concentration of p300 competitive inhibitor C646 ( 1 μM ) not affecting the viability of cells , viral entry or nuclear delivery of viral genome [9] . As expected , acetylated H2B PLA spots were detected in the nucleus and not in the cytoplasm of uninfected cells ( Fig 4A , top panel , white arrows ) . In contrast , we observed increased acetylated H2B PLA spots in the nucleus as well as in the cytoplasm of infected cells ( Fig 4A , middle panel , white and red arrows ) , which were significantly reduced by C646 ( Fig 4A , lower panel ) . As shown by us [9] , PLA analysis for acetylated IFI16 revealed increased IFI16 acetylation and its localization in the cytoplasm of KSHV infected cells ( positive control ) which was abolished in the presence of C646 ( S5A Fig ) . In addition , as before [9] , distribution of IFI16 to the cytoplasm during KSHV infection was restricted to the nucleus in the presence of C646 , demonstrating that only acetylated IFI16 redistributed to the cytoplasm ( S5B Fig ) . To verify the PLA results ( Fig 4A ) , cytoplasmic fractions from uninfected and KSHV infected ( 4 h ) cells untreated ( UT ) or treated with C646 were tested with anti-H2B and anti-IFI16 antibodies ( Fig 4B ) . Cytoplasmic H2B and IFI16 were detected only in the infected cells which was abolished by C646 ( Fig 4B , lanes 2 and 4 ) . Together , these findings demonstrated that similar to IFI16 , KSHV infection induced the acetylation of H2B and its distribution to the cytoplasm . Since we have shown that Ran-GTP assists the transport of acetylated IFI16 from the nucleus to the cytoplasm [9] , we examined whether Ran-GTP is also involved in the transport of acetylated H2B . Whole cell lysate ( WCL ) from uninfected and KSHV infected cells ( 4 h ) treated with or without C646 was IP-ed using anti-Ran-GTPase antibody . Immunoblot analysis of Ran-IFI16 association demonstrated that Ran was not associated with IFI16 in uninfected cells while a prominent association of Ran with IFI16 was observed in infected cells which was reduced by C646 ( Fig 4C , second panel , lanes 1–4 ) . Similarly , Ran was not associated with H2B in uninfected cells , and in contrast , a substantial association of Ran with H2B was detected in infected cells which was abolished by C646 ( Fig 4C , top panel , lanes 1–4 ) . Furthermore , PLA analysis demonstrated increased Ran-H2B association PLA spots only in the infected cell nucleus and cytoplasm ( Fig 4D , white and red arrows , respectively ) which was blocked by C646 ( Fig 4D ) . These results demonstrated that KSHV infection induces the increased H2B acetylation which is crucial for H2B-Ran association followed by transportation to the cytoplasm of infected cells . After their translation in the cytoplasm , IFI16 and H2B translocate to the nucleus via their NLS domains [20 , 21] . IFI16 redistribution to the cytoplasm during KSHV infection was inhibited by LPT [9] . To determine whether the H2B protein detected in the cytoplasm during KSHV infection represents newly synthesized molecules or redistributed from the nucleus , HMVEC-d cells pre-incubated with or without LPT ( 50 nM ) were infected with KSHV for 4 h or uninfected in the presence or absence of LPT . The concentration of LPT used did not show any toxic effect on HMVEC-d cells nor on KSHV entry into the cells and gene expression [9] . PLA analysis revealed the redistribution of H2B ( Fig 4E , red arrow ) and H2B-IFI16 complex ( Fig 4F , red arrow ) to the cytoplasm only in the infected cells which was significantly blocked by LPT and restricted to the nucleus ( Fig 4E and 4F , white arrows ) . In addition , western blot analysis using anti-H2B and anti-IFI16 antibodies with cytoplasmic fractions corroborated the finding that LPT blocked the redistribution of H2B and IFI16 to the cytoplasm ( Fig 4G ) . These results demonstrated that the H2B-IFI16 complex detected in the cytoplasm originated from the nucleus of infected cells . Although ASC knockdown abolished the IFI16-inflammasome , we could still detect reduced levels of IFI16 in the cytoplasm of KSHV infected HMVEC-d cells [9] . Hence , we sought to determine whether this IFI16 represented IFI16-H2B complex . Cytoplasmic extracts from HMVEC-d cells electroporated with siC ( control siRNA ) and siASC and infected with KSHV for 4 h or left uninfected were IP-ed with anti-acetyl lysine antibody . The knockdown efficiency of ASC is shown in Fig 4H , third panel . As seen in the PLA results ( Fig 4A , top panel ) , a basal level of H2B acetylation was observed in uninfected cells ( Fig 4H , panel 2 , lane 1 ) . An increase in acetylated cytoplasmic H2B levels in infected cells ( Fig 4H , panel 2 , lane 2 ) was observed , which was not affected by the absence of ASC ( Fig 4H , panel 2 , lanes 2 and 4 ) . Only a moderate decrease in the level of cytoplasmic acetylated IFI16 was observed by ASC knockdown in infected cells ( Fig 4H , panel 1 , lanes 2 and 4 ) . Taken together , these observations suggest that: a ) KSHV de novo infection increases nuclear H2B acetylation which is subsequently transported to the cytoplasm via Ran-GTP; b ) ASC does not play any role in H2B acetylation and its cytoplasmic transportation; and c ) IFI16-H2B is an independent complex distinct from the IFI16-ASC-procaspase-1-inflammasome complex formed during KSHV infection . After stimulation , STING , an ER membrane protein activates TBK1 , which in turn phosphorylates and activates IRF3 , and pIRF3 translocates into the nucleus to initiate IFN-β gene transcription [12] . With our observations of increased acetylation of H2B and IFI16 , H2B-IFI16 translocation to the cytoplasm independent of the IFI16-ASC inflammasome complex , and the induction of IFN-β in the absence of ASC in herpesvirus infected cells [6] together with the reported role of extra-chromosomal cytoplasmic H2B in interferon induction [17] , we next determined whether H2B along with IFI16 plays any role in STING activation to induce IFN-β . Cytoplasmic fractions from uninfected and KSHV infected HMVEC-d cells ( 2 , 4 , 12 , 24 h p . i . ) described in Fig 2F experiments were IP-ed with anti-IFI16 , H2B or STING antibodies , and the results demonstrated the interactions of IFI16 , H2B and STING ( Fig 5A , 5B and 5C ) . Interaction of IFI16 with STING was observed from 2 to 24 h p . i . , while the interaction of H2B with STING was reduced at 24 h p . i . ( Fig 5A , 5B and 5C ) . In contrast , we did not detect any such interactions in uninfected cells ( Fig 5A , 5B and 5C , lane 1 ) . STING expression was not affected during KSHV infection ( Fig 5D , input controls ) . To further confirm the interaction of IFI16 , H2B and STING , uninfected and KSHV ( 4 h ) infected cells were tested by PLA using anti-IFI16 , H2B and STING antibodies and quantitated ( S5C and S5D Fig ) . We observed substantial association between IFI16 and STING as well as between H2B and STING only in the cytoplasm of infected cells ( Fig 5E and 5F , red arrows ) , demonstrating that these associations are induced by KSHV infection . UV light treatment of KSHV abolishes its ability to express its genome . This process , however , does not affect the envelope and capsid of the virion , creating a virus that is still capable of entry into the virus and delivering the viral genome into the nucleus [2] . Previously , we have shown that the presence of nuclear viral genome but not viral gene expression is enough to induce the IFI16-ASC-procaspase-1 inflammasome activation in the nucleus and translocation into the cytoplasm [2] . Hence , we determined whether KSHV-induced H2B-STING and IFI16-STING is dependent on the presence of viral genome and/or viral gene expression . HMVEC-d cells were infected with 30 DNA copies/cell of UV-KSHV or live-KSHV and PLA reactions were performed for 2 , 4 and 24 h p . i . using anti-H2B , IFI16 and STING antibodies . No H2B-STING or IFI16-STING interactions were observed in the uninfected cells ( S5E–S5H Fig , top panel ) . In contrast , we observed significant H2B-STING and IFI16-STING interactions in UV-KSHV infected cells which were similar to live-KSHV infected cells , and we did not observe any significant change in the above associations between UV-KSHV and live KSHV infected cells ( S5E , S5F , S5G and S5H Fig ) . Taken together , these observations suggest that viral DNA sensing in the nucleus induce the translocations of H2B and IFI16 to the cytoplasm and their interactions with STING and viral gene expression is not required . To determine whether H2B-STING and IFI16-STING interactions in the cytoplasm observed in KSHV infected cells also occur during HSV-1 infection , uninfected or HSV-1 ( KOS ) ( 1 pfu/cell ) infected HFF cells were subjected to PLA reactions using anti-STING , H2B and IFI16 antibodies . We observed substantial levels of H2B-STING and IFI16-STING PLA interaction spots in the cytoplasm of infected cells ( Fig 5G and 5H , lower panels red arrow ) , which is in contrast to uninfected cells showing no or only a few spots of such interactions ( Fig 5G and 5H , top panels ) . These results demonstrate that HSV-1 infection induces the cytoplasmic associations of H2B-STING and IFI16-STING early during infection . Cytoplasmic and nuclear fractions from KSHV ( - ) BJAB and KSHV ( + ) BCBL-1 cells described in Fig 3C experiments were IP-ed using anti-IFI16 , H2B or STING antibodies . Tubulin and TBP western blots demonstrated the purity of the cytoplasmic and nuclear fractions , respectively ( Fig 5L ) . Western blot analysis revealed the interactions between IFI16 and STING as well as between H2B and STING only in the BCBL-1 cytoplasmic fractions ( Fig 5I , 5J and 5K , lane 3 ) . The expression of STING in BJAB and BCBL-1 cells is shown in the input controls ( Fig 5L ) . PLA results demonstrated the IFI16-STING and H2B-STING interactions only in the cytoplasm and not in the nucleus of BCBL-1 cells ( Fig 5M and 5N , red arrows , and S5I Fig ) . Specificities of these reactions are shown by the absence of PLA dots in single species primary anti-H2B or anti-IFI16 antibody reactions ( S6A and S6B Fig ) . To rule out the role of ASC and H2A in the H2B-IFI16-STING interaction , WCL from uninfected and KSHV-infected ( 4 h ) HMVEC-d cells were IP-ed with anti-STING or anti-ASC antibodies . Western blots showed no association of STING with H2A and between ASC and STING ( S6C Fig ) which was further confirmed by PLA analysis ( S6D and S6E Fig ) . Collectively , these results demonstrated the association of STING with IFI16 and H2B but not with ASC in the cytoplasm during KSHV de novo and latent infection . cGAS ( cGAMP-Synthase ) is a cytosolic DNA sensor [12 , 22] , and studies suggest that IFI16 , BRCA1 and cGAS are essential for IFN-β induction during HSV-1 infection of HFF cells [1 , 10] . Hence , we evaluated whether cGAS is part of the IFI16-H2B complex . Cytoplasmic fractions from uninfected and infected HMVEC-d cells were IP-ed with anti-cGAS antibodies . We observed the interactions of cGAS with IFI16 , BRCA1 and STING in the cytoplasmic extracts of KSHV infected cells at 2 , 4 , 12 and 24 h p . i . Interestingly , cGAS interacted with H2B at 2 , 4 , and 12 h p . i . with KSHV which was reduced at 24 h p . i . ( Fig 5O , lanes 2–5 ) . In contrast , very little or no association of the above proteins was observed in uninfected cell cytoplasm ( Fig 5O , lane 1 ) . Expression levels of cGAS remained unchanged in the cytoplasm of infected cells ( Fig 5P ) . Input controls for BRCA1 , IFI16 , H2B and STING were similar as in Figs 2F and 5D . These results suggested that cGAS interacts with IFI16 , H2B , BRCA1 and STING in the cytoplasm of KSHV infected HMVEC-d cells . Next , we determined the interactions of cGAS with IFI16 , H2B , BRCA1 and STING in cells latently infected with KSHV . Cytoplasmic and nuclear fractions from BJAB and BCBL-1 cells were IP-ed with anti-cGAS antibodies . Western blot analysis revealed that cGAS interacted with IFI16 , H2B , and BRCA1 in the nucleus of BJAB and BCBL-1 cells ( Fig 5Q , lane 2 and 4 ) . cGAS also interacted with IFI16 , H2B , BRCA1 and STING but only in the infected BCBL-1 cytoplasm , and in contrast , even though cGAS was detected in the cytoplasm of control BJAB cells ( Fig 5R , lane 1 ) , we did not observe any interaction with IFI16 , H2B , BRCA1 and STING ( Fig 5Q , lanes 1 ) . The expression levels of cGAS is shown in the input controls ( Fig 5R ) , and the levels of IFI16 , H2B , BRCA1 and STING is shown in Figs 3C and 5L as input controls . These observations indicated that KSHV latent infection also induces the interactions of cGAS , IFI16 , H2B , BRCA1 and STING in the cytoplasm of infected cells . Interactions of IFI16 and H2B with STING in the cytoplasm during KSHV infection prompted us to determine whether these proteins also interact with BRCA1 and cGAS in the cytoplasm of infected cells , and whether they form macromolecular complexes . For this , uninfected and infected HMVEC-d cells ( 4 h ) were subjected to double sequential PLA reactions with initial reactions for a ) IFI16 and H2B ( Fig 6A and 6C , green spots ) , BRCA1+H2B ( Fig 6B and 6D , green spots ) , and H2B+STING ( Fig 6E , green spot ) , respectively , followed by b ) second reaction indicated by red spots for H2B+STING ( Fig 6A and 6B ) , H2B+cGAS ( Fig 6C ) , BRCA1+cGAS ( Fig 6D ) and STING+cGAS ( Fig 6E ) , respectively . We observed the following interesting results: Taken together , these results demonstrated that during KSHV infection , nuclear H2B associated with IFI16 and BRCA1 , translocates to the cytoplasm and associates with cGAS-STING . These results further validated the IP-reactions in Fig 5 . Our recent studies demonstrated that BRCA1 knockdown impaired not only genome recognition by IFI16 but also the cytoplasmic IFI16-STING mediated IFN-β response during de novo KSHV and HSV-1 infection [1] . Previous studies also suggested that cGAS induced STING-dependent activation of IRF-3 signaling cascades [12 , 22] . Since BRCA1 and cGAS have been shown to be involved in KSHV and HSV-1 IFN-β responses , we investigated whether H2B has any role in IFI16-STING mediated signaling . HMVEC-d cells electroporated with siC , siBRCA1 , sicGAS and siH2B were uninfected or infected with KSHV for 4 h , and whole cell lysates were IP-ed with anti-IFI16 antibodies . We observed efficient knockdown of these proteins ( Fig 6F ) . In siC KSHV infected cells , STING was IP-ed with IFI16 which was abolished by H2B knockdown ( Fig 6G , compare lanes 1 and 2 with 7 and 8 ) . Similarly , little or no STING was IP-ed with IFI16 in BRCA1 knockdown infected cells ( Fig 6G , compare lanes 1 and 2 with lanes 3 and 4 ) , while cGAS knockdown reduced ( ~50% ) the levels of STING associated with IFI16 ( Fig 6G , compare lanes 1 and 2 with lanes 5 and 6 ) . In addition , BRCA1 and cGAS knockdown also hampered the interaction of IFI16-H2B by ~40% and ~25% , respectively ( Fig 6G ) . The bottom input panels of Fig 6G show the efficiency of H2B knockdown , the absence of off-target effects as well as STING levels . These results suggested a critical participation of H2B and BRCA1 in the cytoplasmic IFI16-STING interactions during de novo KSHV infection . Our studies show that in the absence of ASC , acetylated H2B and IFI16 are detected in the cytoplasm of infected cells ( Fig 4H ) . To rule out the effect of ASC on the H2B-STING interaction , HMVEC-d cells electroporated with siC and siASC were infected with KSHV for 4 and 24 h , and uninfected and infected cell WCL were IP-ed using anti-STING antibodies . ASC knockdown efficiency and expression levels of IFI16 , STING and H2B are shown in Fig 6H and 6I , bottom panels . The absence of ASC did not impact the interaction of H2B with STING or IFI16 with STING in cells infected with KSHV for 4 h ( Fig 6H , top three panels ) . In contrast , although IFI16 IP-ed with STING , the interaction between H2B and STING was not detected at 24 h p . i . ( Fig 6I , top three panels ) which is similar to the observations in Fig 5B and 5C demonstrating the reduced H2B levels and H2B-IFI16 interaction in the cytoplasm at 24 h p . i . Targeting of H2B by factors ( host and/or viral ) could be a potential reason for such a reduced interaction with STING during de novo KSHV infection . Nevertheless , these findings suggest that IFI16 , H2B , BRCA1 , cGAS and STING associate in the cytoplasm during KSHV de novo and latent infection that is independent of ASC . KSHV infection induces IFI16-mediated IL-1β and IFN-β secretion during de novo infection [1 , 2 , 9] . To determine the functional role of H2B , HMVEC-d cells electroporated with siC , siIFI16 , siBRCA1 , sicGAS , siSTING , siH2B or siASC and WCL from uninfected and KSHV ( 4 h ) infected cells were western blotted . We observed efficient knockdown of these proteins ( Figs 6F and 7A ) . To determine the levels of pTBK-1 and pIRF3 signaling molecules involved in IFN-β induction , WCL of these cells were immunoblotted with anti-pIRF3 , tIRF3 , pTBK-1 or tTBK-1 antibodies . We observed increased levels of pIRF3 and pTBK-1 in control siRNA ( siC ) KSHV infected cells ( Fig 7B , lanes 1 , 2 , 13 and 14 ) which were significantly decreased in siIFI16 , siBRCA1 , sicGAS , siSTING , and siH2B infected cells ( Fig 7B , lanes 3–12 ) . In contrast , KSHV infection induced pIRF3 and pTBK-1 levels that were not affected in siASC cells ( Fig 7B , lanes 15 and 16 ) . When supernatants from these cells were tested for IFN-β by ELISA , we observed increased IFN-β secretion ( ~80 pg/ml ) in siC-KSHV infected cells ( 4 h ) which was not affected in siASC knockdown cells ( Fig 7C ) . In contrast , significant reduction in IFN-β secretion was observed in siH2B , siIFI16 , siBRCA1 , siSTING or sicGAS cells infected with KSHV ( Fig 7C ) . Taken together , these results demonstrated that H2B , IFI16 , BRCA1 , STING , and cGAS play roles in the activation of TBK , IRF3 and induction of IFN-β . When the same supernatants from Figs 6F and 7A experiments were tested for the secreted IL-1β levels , significant reduction was observed only in siIFI16 and siASC KSHV infected cells in comparison to siC infected cells ( Fig 7D ) . In contrast , we did not observe any reduction in secreted IL-1β levels in sicGAS , siSTING and siH2B infected cells ( Fig 7D ) which clearly demonstrated that H2B , cGAS , and STING do not play any role in inflammasome activation and IL-1β secretion . HSV-1 infection induced IFN-β secretion in primary HFF cells [6 , 7] . To analyze the importance of H2B in IFN-β induction by HSV-1 , HFF cells electroporated with siC , siIFI16 , siBRCA1 , sicGAS , siSTING , siH2B or siASC were infected with HSV-1 ( 4 h ) , and supernatants were tested by IFN-β ELISA . WB analysis of WCL from the above cells showed significant knockdown efficiency of these proteins ( S6F Fig ) . We detected a significant level of IFN-β secretion ( ~280 pg/ml ) in siC-HSV-1 infected cells which was unaffected in siASC-infected cells ( Fig 7E ) . In contrast , a significant reduction in IFN-β secretion was observed in siH2B , siIFI16 , siBRCA1 , siSTING and sicGAS infected cells ( Fig 7E ) . Collectively , these findings demonstrated that similar to the role played by IFI16 , H2B , BRCA1 , STING and cGAS play a role in HSV-1 infection induced IFN-β production . cGAS induction results in the production of cGAMP ( cyclic GMP-AMP ) which in turn activates STING resulting in pTBK-1 , pIRF3 and interferon induction . To ascertain our findings that H2B and IFI16 are essential in IFN-β production through STING-mediated pathway during KSHV infection , we determined the level of cGAMP production . For this we used the THP1-Lucia cells expressing the secreted luciferase Lucia reporter gene under the control of an IRF-inducible promoter consisting of five IFN-stimulated response elements ( ISRE ) . The activation of STING by cGAMP induces the IRF3 phosphorylation , pIRF3 translocation into the nucleus and activation of ISRE resulting in the secreted luciferase . Uninfected HMVEC-d cells electroporated with siC , siH2B and siIFI16 were infected with KSHV for 4 h , lysed , treated with benzonase , and heat inactivated at 95°C for 5 min . 10 μl of these lysates or varying amounts of purified cGAMP were added to 1X105 THP-1-Lucia ISG cells . After overnight incubation , 10 μl culture supernatant was used in a luminescence assay as a measure of cGAMP produced during KSHV infection . A control luminescence assay measuring the cGAMP activity of pure cGAMP is shown in Fig 8A . We observed that KSHV infection induced a considerable level of cGAMP in siC HMVEC-d cells which was reduced significantly in the absence of IFI16 ( ~75% ) and H2B ( ~60% ) ( Fig 8B1 ) . As a specificity control that the observed results shown in Fig 8B1 are via STING , we used the same lysates from KSHV infected cells with STING knockout ( KO ) THP-1-Dual KO-STING cells expressing secreted Lucia luciferase gene under the control of ISG54 ( interferon-stimulated gene ) ISRE which can be induced by STING-dependent IRF3 as well as STING-independent IRF9 that is inducible by IFN-α/β . When ISRE-induced Lucia luciferase activity was measured , the recombinant human IFN-β used as control increased the luciferase activity in THP-1-Dual KO-STING cells which demonstrated the activation by a STING-independent pathway ( Fig 8B2 ) . In contrast , we observed only negligible levels of luciferase activity in the supernatant of cells incubated with the lysates from siC KSHV and IFI16 and H2B knocked down infected cells ( Fig 8B2 ) . These results demonstrated that the luciferase activity seen in Fig 8B1 was through the cGAS-STING dependent pathway . When these electroporated plus KSHV infected/uninfected cells ( from Fig 8B1 ) were examined by PLA , compared to uninfected cells , a considerable number of IFI16-cGAS PLA spots were observed in the cytoplasm of siC KSHV infected cells ( Fig 8C and 8D ) . In contrast , a significant reduction in the cytoplasmic IFI16-cGAS association was observed in H2B knockdown cells ( Fig 8C and 8D ) . Together with the results shown in Figs 5 , 6 and 7 , these results demonstrate that H2B is essential for IFI16-cGAS association , cGAMP induction and IFN-β production during de novo KSHV infection . To ascertain the results shown in Fig 7 that H2B doesn’t play a role in inflammasome activation and IL-1β secretion , uninfected HMVEC-d cells were electroporated with siC and siH2B and knockdown efficiencies were verified . These cells were infected with KSHV ( 30 DNA copies/cell ) for 4 h and tested by PLA reactions using anti-IFI16 and anti-ASC antibodies . As reported by us before [1 , 2] , compared to uninfected cells , we observed a significant number of IFI16-ASC interacting PLA spots in the cytoplasm of siC-infected cells ( Fig 8E and 8G ) . In addition , we did not observe any significant change in the association of IFI16 with ASC association in H2B knockdown cells ( Fig 8E and 8G ) . These results , together with the negligible effect on the secreted IL-1β levels in siH2B infected cells ( Fig 7D ) , demonstrated that IFI16-ASC inflammasome induction and their cytoplasmic translocation is independent of H2B . When these cells were examined by PLA with anti-IFI16 and anti-acetylated antibodies , compared to uninfected cells , a substantial level of acetylated IFI16 PLA spots were observed in the cytoplasm of KSHV infected cells ( Fig 8F and 8H ) . Knockdown of H2B resulted in ~30% reduction in the cytoplasmic acetylated IFI16 PLA spots ( Fig 8F and 8H ) . These results suggested that the absence of H2B doesn’t affect IFI16-ASC association and cytoplasmic distribution but reduces the level of cytoplasmic acetylated IFI16 during KSHV de novo infection . This reduction suggests that during KSHV infection , two distinct acetylated IFI16-H2B and IFI16-ASC complexes are formed in the nucleus and redistributed to the cytoplasm , and H2B knockdown results in the absence of acetylated IFI16-H2B in the cytoplasm . We have reported that IFI16 recognizes the episomal KSHV , EBV and HSV-1 genomes in the nucleus of infected cells resulting in IFI16 mediated innate inflammasome and IFN-β responses [1 , 2 , 4–6] . Herpesvirus genomes delivered in the nucleus as linear , naked dsDNA with nicks and breaks undergo rapid circularization and chromatinization [23] . Since H2B is an essential component of chromatin structure , we first determined whether H2B associates with the KSHV genome in the nucleus . HMVEC-d cells were infected with unlabeled or EdU-genome labeled KSHV ( 200 copies/cell ) for 2 h . A DNA mediated pull down assay was performed as described earlier [1] by first cross-linking protein-DNA and linking biotin-TEG azide to EdU-labeled viral DNA via a Click reaction . Following pull down by streptavidin , captured proteins were immunoblotted with anti-H2B and anti-H3 antibodies . Similarly , HSV-1 genome association with H2B during de novo infection in HFF cells was also performed . We observed the association of H2B with EdU-labeled KSHV and HSV-1 genomes early during infection ( Fig 9A and 9B , lane 2 ) . DNA purified from unlabeled or EdU-labeled KSHV or HSV-1 infected cells showed similar levels and served as input controls ( S7A and S7B Fig , lanes 1 and 2 ) . Upon streptavidin capture , DNA recovery was observed only from cells infected with EdU-labeled virus ( S7A and S7B Fig , lane 4 ) but not from cells infected with unlabeled virus ( S7A and S7B Fig , lane 3 ) . This demonstrated the specificity of the EdU genome pull down assay . Our recent studies showed that BRCA1 knockdown reduced the association of IFI16 with KSHV and HSV-1 genomes [1] . To determine the role of H2B in KSHV genome association by IFI16 , HMVEC-d cells transfected by siC , siH2B or siBRCA1 were infected with EdU-genome labeled KSHV and tested by PLA with anti-IFI16 ( mouse and goat ) antibodies and EdU-labeled genome was detected by an EdU reagent kit [1] . PLA results showed a considerable amount of EdU KSHV ( red dots ) and IFI16-IFI16 ( green dots ) association in the nucleus ( Fig 9C , top right panel , white arrows and Fig 9D ) which were reduced significantly in siH2B infected cells ( Fig 9C , middle right panel , white arrows and Fig 9D ) . This suggested that H2B has a significant role in KSHV genome recognition by IFI16 . Similarly , in PLA reactions to detect the EdU-labeled HSV-1 genome association with IFI16 during de novo HFF cell infection we observed substantial levels of EdU-HSV-1 and IFI16-IFI16 association in the nucleus ( Fig 9E , top right panel , white arrows and Fig 9F ) which was reduced significantly in siH2B infected cells ( Fig 9E , middle right panel , white arrows and Fig 9F ) . In addition , we observed significantly less IFI16 distribution in the cytoplasm of HMVEC-d and HFF cells infected by KSHV and HSV-1 in siH2B compared to siC infected cells ( Fig 9C and 9E , red arrows left panels ) . siBRCA1 was used as a positive control [1] and we observed substantial reduction in IFI16 association with KSHV and HSV-1 genomes ( Fig 9C–9F ) . Collectively , these observations revealed the H2B association with the KSHV and HSV-1 genomes and highlighted the essential role of H2B in the regulation of KSHV and HSV-1 genome recognition by IFI16 and correlated to the subsequent IFI16 mediated host innate IFN-β responses . To define the association of H2B with viral genomes further , HFF cells electroporated with siC , siH2B or siBRCA1 were infected with unlabeled or EdU-labeled HSV-1 ( 10 pfu/cell ) for 2 h and then protein-DNA cross-linking was performed as described in Fig 9A . The purified DNA from unlabeled or EdU-labeled HSV-1 infected cells showed similar levels in siC , siBRCA1 and siH2B electroporated cells ( S7C Fig , lanes 8–13 ) . However , DNA was recovered by streptavidin capture materials only from EdU-labeled HSV-1 infected cells ( S7C Fig , lanes 5–7 ) but not from unlabeled virus infected cells ( S7C Fig , lanes 2–4 ) . DNA shearing and short chromatin fragments were captured on streptavidin beads and pull down proteins were analyzed by immunoblotting using anti-IFI16 and H2B antibodies . The absence of any protein in unlabeled HSV-1 infected cells validated the specificity of these reactions ( Fig 9G , top two panels , lanes 1–3 ) . We observed substantial levels of IFI16 association with HSV-1 genome ( siC ) which was significantly reduced by siBRCA1 ( 80% ) as well as in siH2B ( 59% ) infected cells ( Fig 9G , top panel , lanes 4–6 ) . H2B association with viral genome was significantly reduced ( 79% ) in the siH2B cells and in contrast , we observed a substantial level of H2B association with HSV-1 genome in siBRCA1 with only ~8% less compared to siC cells ( Fig 9G , second panel , lanes 4–6 ) . This suggested that H2B associates with HSV-1 genome independent of its innate response functions mediated through its association with IFI16 and BRCA1 . Nevertheless , collectively our results suggested that H2B is critical in KSHV and HSV-1 genome recognition by IFI16 during de novo infection . Since we observed that H2B associates with IFI16 in the nucleus of uninfected and KSHV infected cells ( Figs 1 and 2 ) and H2B was also pulled down by KSHV and HSV-1 genome during de novo infection ( Fig 9A and 9B ) , we next determined whether H2B participates in viral genome recognition by IFI16 . HMVEC-d cells transfected with siC and siBRCA1 followed by infection with EdU-labeled KSHV were subjected to PLA reaction using anti-IFI16 ( mouse ) and anti-H2B ( rabbit ) antibodies . PLA revealed association of IFI16-H2B ( green dots ) and EdU-KSHV ( red dots ) in the nucleus of cells which were significantly reduced in BRCA1 knockdown cells ( Fig 9H ) . The average colocalized PLA spots ( yellow color ) per cell are presented in the bar graphs ( Fig 9I ) . These results suggested that IFI16 in complex with H2B associates with the KSHV genome and BRCA1 participates in this association . Our studies show that KSHV induces IFI16 and p300 interaction and p300 is required for KSHV induced acetylation of IFI16 [9] , and that BRCA1 knockdown results in inhibition of IFI16 translocation to the cytoplasm , and subsequent IL-1β and IFN-β induction [1] . Since KSHV or HSV-1 genome recognition by IFI16 alone or in complex with H2B ( Fig 9 ) was significantly reduced in the absence of BRCA1 and since BRCA1 is known to interact with p300 [24] , we hypothesized that besides involvement in IFI16’s ability to recognize viral episomal genomes , BRCA1 may also be involved in the post-genome recognition event of acetylation of IFI16 and H2B . To determine the role of BRCA1 in the acetylation of IFI16 , HMVEC-d cells electroporated with siC and siBRCA1 were left uninfected or infected with KSHV for 4 h and tested by PLA reactions using anti-IFI16 and anti-acetyl lysine antibodies . PLA analysis revealed that BRCA1 knockdown abolished the acetylation of IFI16 during KSHV de novo infection ( Fig 10A ) . To further verify the effect of BRCA1 on H2B and IFI16 acetylation , HMVEC-d cells electroporated with siC and siBRCA1 were left uninfected or infected by KSHV for different time points and WCL was IP-ed using anti-IFI16 and , anti-acetyl lysine antibodies and western blotted for IFI16 , BRCA1 , p300 and H2B . The IP results demonstrated that IFI16 and p300 interaction as well as IFI16 and H2B acetylation were inhibited in BRCA1 knockdown cells during KSHV de novo infection ( Fig 10B ) . Bottom panels showed input controls for BRCA1 , p300 , IFI16 and total protein acetylation levels ( Fig 10B , bottom panels ) . Collectively , these results demonstrated that BRCA1 is not only essential for genome recognition by IFI16 alone or in complex with H2B but also plays important roles in the recruitment of p300 for the acetylation of IFI16 and H2B , which leads into IFI16 interaction with ASC , inflammasome formation , transport into the cytoplasm , and IL-1β induction as well as inflammasome independent IFI16-H2B translocation into the cytoplasm , interaction with cGAS-STING and IFN-β induction .
The innate immune response is a very effective front line host defense against microbial pathogens including viruses . Eukaryotic nuclear proteins , besides being involved in localized functions , also mediate functions in the cytoplasm . For example , high mobility box protein 1 ( HMGB1 ) involved in transcriptional regulation and DNA organization also acts as an “alarmin” , exits the nucleus during necrosis and activates innate immune signaling by binding to viral RNAs and DNAs [25] . Extrachromosomal functions of histones have also been observed . An apoptotic stimulus , such as DNA damage induces the translocation of nuclear histone H1 . 2 to mitochondria by an unknown mechanism to promote the mitochondrial apoptotic pathway [26] . Extrachromosmal H2B was shown to be one of the potential mediators of the IFN-β response to cytoplasmic host DNA [17] . Our studies not only identify H2B as an innate immune sensor of nuclear herpesviral genomes but also define the potential mechanisms by which H2B mediates its extrachromosomal IFN-β response function during herpes viral infection . Our comprehensive studies ( Fig 10C ) demonstrate for the first time that: a ) histone H2B is in a complex with innate immune DNA sensor IFI16 and BRCA1 proteins in the nucleus . This is independent of its interactions with histone H2A . b ) H2B is a component of IFI16-BRCA1 , and sensing of the nuclear herpesvirus episomal genome results in BRCA1 dependent recruitment of p300 to IFI16 and subsequent acetylation of H2B and IFI16 . c ) Acetylated IFI16-H2B in association with BRCA1 is exported to the cytoplasm by Ran-GTP where they interact with cGAS and STING , resulting in pIRF3 induction and IFN-β production . d ) Independent of H2B-IFI16-BRCA1 , genome recognition by IFI16-BRCA1 leads to the formation of a BRCA1-acetylated IFI16-ASC-procaspase-1 inflammasome complex , which also translocates to the cytoplasm resulting in caspase-1 formation and subsequent cleavage of pro-IL-1β , and e ) Independent of its interactions with IFI16-BRCA1 , H2B associates with viral genome . We have shown that BRCA1 is associated with IFI16 in uninfected cells which increased in KSHV , HSV-1 and EBV infected cells . Also , BRCA1 is part of the IFI16-ASC-procaspase-1 inflammasome complex [1] . BRCA1 is essential for KSHV and HSV-1 genome recognition by IFI16 since in the absence of BRCA1 , IFI16’s association with the viral genome is significantly reduced ( Fig 9 ) [1] , resulting in decreased IFI16 cytoplasmic translocation , and inflammasome as well as IFN-β responses [1] . Significant reduction in IFI16’s association with viral genome ( Fig 9 ) in H2B knockdown cells and the near absence of an IFN-β response suggests that H2B , in association with IFI16-BRCA1 , is involved in viral genome sensing . However , results such as ~40% association of IFI16 with viral genome ( Fig 9G ) and induction of IL-1β secretion in the infected cells in the absence of H2B ( Fig 7D ) demonstrate that the function of viral genome recognition resulting in BRCA1-IFI16-ASC-procaspase-1 inflammasome complex formation is not affected . The IFI16-H2B complex is mainly involved in inflammasome independent STING-mediated IFN-β production as shown by the association of cytoplasmic IFI16-H2B-BRCA1 complex with cGAS and STING to form a signal hub and IFN-β production ( Fig 10C ) , absence of IFI16-STING association in H2B knockdown cells during KSHV de novo infection ( Fig 6G ) , absence of H2B interaction with ASC ( Fig 2B and 2D ) and absence of STING interaction with ASC ( S6C–S6E Fig ) . These observations are consistent with earlier studies demonstrating that ASC is not essential for IFN-β production [6 , 8] . Therefore , we surmise that nuclear viral genome sensing is mediated by at least two IFI16 complexes in which 1 ) recognition by the IFI16-BRCA1-H2B complex results in the IFN-β response , and 2 ) recognition by the IFI16-BRCA1 complex results in interactions with ASC and inflammasome formation . The BRCA1-IFI16 complex that we observed is not related to the host DDR responses induced by the addition of bleomycin in the uninfected HMVEC-d cells [1] . Though IFI16 association with viral genome was significantly reduced in the absence of BRCA1 , H2B association with viral genome was not affected ( Fig 9G ) . This suggests that independent of BRCA1 and IFI16 , H2B associates with viral genome probably to mediate its nucleosome associated functions , which are distinct from the IFI16-BRCA1-H2B mediated innate interferon response . This is supported by our earlier observations that there was no significant IFN-β production in the BCBL-1 cells [1 , 3] which could be due to a virus strategy to avoid antiviral effects as several viral latent proteins such as LANA-1 and vIRF-1 have been shown to block the pIRF nuclear functions of IFN response gene activation [27–29] . Nuclear histone H2A , H2B , H3 and H4 proteins form octamers , bind and package DNA into ordered nucleosome units consisting of two H2A-H2B dimers and an H3-H4 tetramer [30–32] . H2B undergoes several modifications such as acetylation , phosphorylation and ubiquitination that are essential for its various roles , and p300 is essential for the acetylation of H2B-H2A [16 , 18 , 19 , 33–37] . Our studies demonstrate that basal levels of H2B acetylation in the nucleus increased in KSHV infected cells and acetylation of IFI16 associated H2B is crucial for H2B’s cytoplasmic translocation along with IFI16 and BRCA1 and the consequent innate IFN-β response . Nuclear herpes viral genome recognition by IFI16 is dependent upon its association with BRCA1 but independent of its acetylation [9] . Increased IFI16-p300 interaction was observed only after nuclear herpes viral genome entry resulting in increased p300 activity only in the nucleus . This increased acetylation was not due to decreased activity of HDACs [9] . We speculated a role in IFI16-p300 interaction as BRCA1 is known to interact with p300 [24] . Absence of IFI16 interaction with p300 together with the absence of increased IFI16 and H2B acetylation in BRCA1 knockdown cells ( Fig 10B ) clearly demonstrate that besides its role in IFI16’s ability to recognize the episomal viral genomes [1] , BRCA1 is essential for the recruitment of p300 to the IFI16-H2B-BRCA1 complex leading to H2B and IFI16 acetylation . p300-mediated acetylation has been shown to modify the nucleosome structure to facilitate the disassociation and transfer of H2B-H2A from the nucleosome to histone chaperon NAP-1 [37] . Even though H2B interactions with H2A were observed in the nucleus of uninfected and infected cells , only H2B in association with IFI16 and BRCA1 was detected in the cytoplasm of virus infected cells ( Figs 1 and 2 ) . It is possible that acetylation of IFI16 and H2B probably leads to changes in their affinity and structure to facilitate their dissociations from the viral DNA leading to their association with RAN-GTP , transport to the cytoplasm leading into interaction with cGAS and STING then IFN-β production ( Fig 10C ) . The detection of IFI16 association with the viral genomes in KSHV and EBV in latently infected cells suggest that recognition , acetylation and relocalization of H2B-IFI16 is a dynamic continuous event with IFI16 always occupying the viral genomes [3 , 9] . KSHV doesn’t infect laboratory animals including primates . We have examined human tissue sections from normal skin , KSHV+ Kaposi’s sarcoma ( KS ) lesions , control lung and KSHV+ solid lung primary effusion B-cell lymphoma ( PEL ) lesions . IFI16 and ASC colocalization was not observed in the control skin and normal lung sections . In contrast , perinuclear cytoplasmic colocalization of ASC and IFI16 was observed in KSHV+ KS and PEL lesions [3] . These studies demonstrated the potential in vivo involvement of the IFI16-inflammasome in KSHV biology . cGAS , identified as a cytoplasmic DNA sensor [11 , 22 , 38–41] , was also detected in the nucleus and cytoplasm of HFF cells and immortalized oral keratinocyte cells [10] , and in the nucleus and cytoplasm of BCBL-1 cells . However , IFI16 is the primary sensor of HSV-1 DNA in the nucleus and cGAS is not involved in genome recognition [10] . cGAS is believed to stabilize IFI16 in HSV-1 infected cells [10] . However , it is not clear whether this interaction or stabilization occurs in the nucleus or in the cytoplasm . Nevertheless , cGAS and IFI16 knockdown impaired the IFN-β responses in HSV-1 infected HFF cells [10] . We also observed similar findings ( Fig 7 ) and in addition , demonstrate the interactions of cGAS with H2B , IFI16 , BRCA1 , and STING in the cytoplasm of cells during de novo infection as well as in latent infection ( Figs 5O , 5Q and 6 ) . A recent study also demonstrated the Interaction of cGAS with STING in the cytoplasm during Chlamydia tracomatis infection resulting in IFN-β production [42] . Our earlier studies have shown that the IFI16-inflammsome is not induced by the infection of HMVEC-d cells with lentivirus expressing KSHV proteins [2] which suggested that IFI16 perhaps doesn’t recognize the integrated lentvirus genome as foreign . A recent study with HCMV ( Ad169 ) infection for 6 h in IFI16 knockout ( KO ) human fibroblast cells suggests that IFI16 is not necessary for the IFN-β response [43] . However , as responses against HSV-1 or KSHV de novo infections are not examined in these studies under similar IFI16 KO conditions , it is premature to conclude the role of IFI16 in the IFN-β response during the complex biology of various herpesvirus infections . Moreover , our ongoing studies with HSV-2 and HSV-1 with human osteosarcoma cells in which IFI16 is knocked out by CRISPR [6] demonstrate that in the absence of IFI16 , the IFN-β response is significantly abrogated . IFI16 , H2B , BRCA1 and cGAS knockdown clearly demonstrate that a macromolecular complex of these molecules is necessary for STING activation and innate IFN-β response during KSHV and HSV-1 de novo infection ( Fig 7 ) . Moreover , H2B knockdown significantly reduced the IFI16-STING and IFI16-cGAS association and cGAMP production ( Figs 6 , 7 and 8 ) . Furthermore , THP-1-Dual KO-STING cells results ( Fig 8B2 ) confirmed that IFN-β induction is mainly mediated through the cGAS-STING pathway which demonstrates that H2B is essential for the IFI16-H2B-cGAS-STING-mediated IFN-β response . Whether cGAS stabilizes IFI16 in complex with STING and whether cGAS forms single or separate complex with H2B and IFI16 needs to be studied thoroughly which are beyond the scope of our present study . Similarly , the role of nuclear cGAS in the IFI16-STING-mediated innate IFN-β response requires additional studies .
Leptomycin B ( LPT ) , EdU ( 5-ethynyl-2’-deoxyuridine ) , and C646 ( Sigma-Aldrich ) , SlowFade Gold Antifade reagent with DAPI ( Life Technologies ) , Verikine human IFN-β ELISA kit ( PBL Assay Science ) and IL-1β ELISA kit were from RayBiotech , Inc . Human dermal microvascular endothelial ( HMVEC-d ) cells and human foreskin fibroblast ( HFF ) cells ( Clonetics ) , BJAB and BCBL-1 cells ( ATCC CRL 8799 and 2336 ) were grown as described earlier [2 , 4 , 5] . KSHV was purified from the supernatant of induced BCBL-1 cells using phorbol ester and virus DNA copy number was analyzed by real-time DNA-PCR [2] . In most of the experiments KSHV infection was done with 30 DNA copies/cell for 2 h in serum free medium , washed , and then replaced with complete medium for different time points of infection [2] . HSV-1 ( KOS ) production and viral titer using a plaque assay on Vero cells were performed as described earlier [4] . In most of the experiments , HSV-1 infection was done with 1 pfu/cell ( ~25 DNA copies/cell ) in serum free medium for 2 h , washed and replaced with complete medium and incubated for other time points . Antibodies are listed in Table 1 . The secondary antibodies conjugated to HRP against anti-rabbit , anti-goat and anti-mouse IgG and Alexa Fluor-488 , and -594 ( Molecular Probes ) and VeriBlot for IP secondary antibody ( HRP ) were purchased from Abcam . The BCBL-1 cells were induced using phorbol ester and KSHV DNA was labeled by adding EdU ( 5-ethynyl-2’-deoxyuridine ) ( 10 μM ) in the culture medium during lytic replication on the first and third day of induction [1] . The labeled viruses from the cell culture supernatant ( day 5 ) were purified and their genome copy number was analyzed by real-time DNA-PCR [2] . HSV-1 ( KOS ) genome was labeled by adding EdU to the Vero cell medium at 8 , 24 and 48 h post-infection [1] . On day 4 , the culture supernatant was collected and labeled virus was purified and titrated [4] . Cells were harvested and used for preparation of nuclear and cytoplasmic extracts using a nuclear complex Co-IP kit ( Active Motif Corp . ) . Nuclear and cytoplasmic proteins were estimated using BCA protein assay reagent ( Pierce ) , and purity of the fractions was determined by western blotting using anti-TBP and anti-β-tubulin antibodies , respectively . Cells were lysed in RIPA ( radioimmunoprecipitation assay ) lysis buffer ( 15 mM NaCl , 1 mM MnCl2 , 1 mM MgCl2 , 2 mM phenylmethylsulfonyl fluoride plus protease inhibitor cocktail ) , sonicated , and centrifuged at 10 , 000 rpm at 4°C for 10 min for western blot analysis . An equal amount of proteins were separated by SDS-PAGE , transferred to nitrocellulose and incubated with primary antibodies followed by HRP-conjugated secondary antibodies . Immunoreactive protein bands were detected by chemiluminescence ( Pierce ) as per manufacturer’s instructions . For IP , the harvested cells were lysed using IP lysis buffer ( 25 mM Tris-HCl , pH7 . 5 , 150 mM NaCl , 1% NP40 , 2 mM EDTA , 10% Glycerol , and protease inhibitor mixture ) and 150 to 200 μg of precleared whole cell lysates or extracted nuclear/cytoplasmic fractions were incubated overnight with primary antibodies at 4°C . The immune complexes were captured using protein A- or G-sepharose ( GE Healthcare , PA ) , washed thrice and examined by immunoblotting . Blots were scanned by an AlphaImager system ( Alpha Innotech Corp . ) and quantitated by ImageJ software . Primary HMVEC-d and HFF cells were electroporated with different siRNAs using a Neon Transfection System ( Invitrogen ) as per manufacturer’s instructions [1] . Briefly , subconfluent monolayer cells were harvested and washed with 1X PBS ( phosphate-buffered saline ) and resuspended ( 1x107 cells/ml ) in resuspension buffer R ( Invitrogen ) . Ten microliters of cell suspension plus 100 pmol of siRNA were mixed and then used for microporation at room temperature using a single pulse of 1350 V for 30 ms for HMVEC-d and 1700 V for 20 ms for HFF cells . Soon after microporation , cells were dispersed into complete medium and incubated at 37°C in a humidified 5% CO2 incubator . After 48 h electroporation , cells were used either for nuclear or cytoplasmic fractions or lysed in IP or RIPA buffer and knockdown efficiency was determined by immunoblots . siRNA oligonucleotides for BRCA1 and IFI16 ( siGenome SMART pool ) , STING ( smart pool: siGenome TMEM173 ) , ASC and C6orf150 ( cGAS ) ( Santa Cruz Biotechnology , Inc ) and a non-targeting siRNA pool were purchased from Thermo Scientific . For H2B siRNA , dsRNA was synthesized by Invitrogen ( stealth RNAi ) : histone H2B sense , 5′-UCC AAG GCC AUG GGC AUC AUG AAC U-3′; histone H2B antisense , 5′-AGU UCA UGA UGC CCA UGG CCU UGG A-3′ . The non-coding stealth siRNA was purchased from Invitrogen . Primary HMVEC-d and HFF cells seeded on glass chamber slides ( Nalgene Nunc International ) were uninfected , KSHV infected ( 30 DNA copies/cell ) , or HSV-1 infected ( 1pfu/cell ) , fixed for 15 min with 4% paraformaldehyde , and permeabilized using 0 . 2% Triton X-100 for 5 min . Cells were then washed and blocked using Image-iT signal enhancer ( Life Technologies ) for ~20 min followed by incubation with primary antibodies and then incubated with secondary antibodies conjugated with fluorescent dye . To detect EdU labeled viral genome , cells were fixed , permeabilized and blocked with Image-iT signal enhancer for 20 min . A CLICK reaction was performed for 30 min at RT using Click-iT EdU reaction additive ( Life Technologies ) , copper sulphate , EdU reaction buffer and Alexa Fluor 594 azide . Cells were observed by Nikon Eclipse 80i microscope , and analyzed with Metamorph digital imaging software . All images were acquired at 40X magnification . Cell culture supernatants from uninfected or virus infected cells ( ~3X105 ) were collected and levels of IFN-β and IL-1β secretion were measured [2 , 6] . The absorbance was read at 450 nm using a Synergy2 Biotek Plate Reader ( Biotek ) . Protein—protein interactions were studied using a DUOLink PLA kit ( Sigma ) as described earlier [1] . Briefly , uninfected and KSHV ( 30 DNA copies/cell ) infected HMVEC-d cells or HSV-1 ( 1 pfu/cell ) infected HFF cells were seeded in chamber microscope slides , fixed with 4% PFA for 15 min , permeabilized using 0 . 2% Triton X-100 for 5 min and then blocked with blocking buffer for 30 min at 37°C . For BJAB and BCBL-1 cells , equal numbers of cells were washed with PBS by centrifugation at 200xg at 4°C and spotted on glass slides , then fixed/permeabilized with pre-chilled acetone and blocked with blocking buffer . Cells were incubated with primary antibodies , washed and further incubated with species specific PLA probes ( PLUS and MINUS probes ) under hybridization conditions in the presence of two additional oligonucleotides to enable hybridization of PLA probes if they were in proximity of <40 nm . A ligation mixture was added to form a closed circle while amplification mixtures result in the formation of a concatemeric product extending from the oligonucleotide arm of the PLA probe . Finally , a detection mixture consisting of fluorescently labeled oligonucleotides was added , and the labeled oligonucleotides were hybridized to the concatemeric products . The signal was detected as a distinct fluorescent dot in the FITC green or Texas red channel and analyzed by fluorescence microscopy . Negative controls consisted of samples treated as described above but with only primary , secondary or control IgG antibodies . The average number of PLA dots per cell was quantified using DUOLink software . For double PLA , two independent PLA reactions were performed sequentially [1] . Briefly , the PLA reaction for IFI16 , H2B and STING was performed first using mouse anti-IFI16 and goat anti-H2B antibodies and detected by DUOLink green detection agent . Cells were then washed , blocked and subjected to a second PLA reaction with goat anti-H2B and rabbit anti-STING antibodies and detected with DUOLink red detection agent . HMVEC-d cells electroporated for 48 h with siC , siH2B and siIFI16 were left uninfected or infected with KSHV ( 30 DNA copies/cell ) for 4 h . Cells were lysed , treated with benzonase for 30 min at 37°C , and heat inactivated at 95°C for 5 min . 10 μl of heat inactivated lysates or varying amounts of purified cGAMP were added to the 1X105 THP-1-Lucia ISG ( InVivoGen ) and THP-1-Dual KO-STING cells ( InVivoGen ) pretreated with digitonin for 30 minutes . Cells were then incubated overnight , and 10 μl culture supernatant mixed with 50 μl of QUANTI-Luc luminescence assay solution ( InVivogen , San Diego , CA ) and cGAMP level assayed by the luminescence read on an ELISA plate reader . The EdU-labeled viral genome ( chromatin ) pull down method has been described earlier [1] . Briefly , HMVEC-d and HFF cells ( ~8x106 cells/ml ) with or without control H2B or BRCA1 siRNA for 48 h were infected with unlabeled or EdU labeled KSHV ( 200 DNA copies/cell ) and HSV-1 ( 10 pfu/cell ) for 2 h and then cross-linked with 1% formaldehyde for 10 min at 4°C . Unreacted formaldehyde was quenched using 0 . 125 M glycine for 10 min at 4°C and cells were harvested , permeabilized ( 0 . 1% Triton X-100 ) for 10 min and washed with PBS . Biotin was linked to EdU genome by a Click reaction using sequential addition of ( + ) -sodium-L-ascorbate ( 10 mM ) , biotin-TEZ azide ( 0 . 1 mM ) and copper ( II ) sulfate ( 2 mM ) for 30 min in the dark followed by adding 1% BSA and 0 . 5% Tween 20 for 10 min . The soluble proteins were isolated in 500 μl CL lysis buffer ( 50 mM HEPES , pH 7 . 8 , 0 . 25% Triton X-100 , 0 . 5% NP-40 , 150 mM NaCl , 10% glycerol plus protease inhibitors ) for 10 min at 4°C and centrifuged at 300xg . The pellet containing chromatin-protein complexes was washed with wash buffer ( 10 mM Tris-HCL , pH 8 . 0 , 0 . 5 mM DTT , 200 mM NaCl ) at 4°C for 10 min and then resuspended in 500 μl RIPA buffer ( 10 mM Tris-HCl , pH 8 . 0 , 0 . 1% Na-Deoxycholate , 0 . 1% SDS , 1% Triton X-100 and 140 mM NaCl plus protease inhibitor cocktail ) and chromatin was sheared by sonication . The sonicated extract was clarified by centrifugation ( 15 , 000xg ) for 10 min at 4°C and 1 mg of the extract was used for pull down using 50 μl of streptavidin magnetic beads . Beads with bound complexes were subjected to reverse protein-DNA cross-linking and proteins were eluted in 1X Laemmli sample buffer ( 95°C for 10 min ) for immunoblotting . To purify DNA , the complexes were eluted from beads in elution buffer ( 0 . 1 M NaHCO3 and 1% SDS ) and cross-linking was reversed by treating with 0 . 1 mg/ml RNase A and 0 . 3M NaCl for 30 minutes at 37°C and then incubated at 65°C for 2 h with 0 . 1 mg/ml Proteinase K . Eventually , DNA was column purified using a Qiagen DNA extraction kit as per manufacturer’s instructions . Data are expressed with means ± SD of at least three independent experiments using a Student’s T-test . p<0 . 05 was considered statistically significant . | Eukaryotic cells elicit innate immune responses against invading microbes including viruses . IFI16 , a predominantly nuclear protein , has emerged as an innate response nuclear DNA sensor . Recognition of nuclear KSHV , HSV-1 and EBV dsDNA genomes by IFI16-BRCA1 leads to IFI16 acetylation , cytoplasmic translocation of the BRCA1-IFI16-ASC-procaspase-1 inflammasome complex and IL-1β generation . Here , we demonstrate that histone H2B is associated with IFI16-BRCA1 in the nucleus under physiological conditions . Recognition of nuclear viral genomes by IFI16-H2B-BRCA1 leads to BRCA1-p300 mediated acetylation of H2B and IFI16 , and cytoplasmic transport of H2B-IFI16-BRCA1 via Ran GTP protein . The inflammasome independent cytoplasmic IFI16-H2B-BRCA1 complex interacts with cGAS and STING resulting in TBK1 and IRF3 phosphorylation , and nuclear pIRF3-mediated IFN-β induction . H2B knockdown inhibits IFN-β production while ASC silencing doesn’t affect IFN-β induction . Our studies identify H2B as an innate nuclear sensor and reveal that two IFI16 complexes mediate nuclear herpesviral genome recognition responses , IFI16-BRCA1-H2B-IFN-β responses and IFI16-BRCA1-inflammasome responses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"viruses",
"immunoprecipitation",
"dna",
"viruses",
"viral",
"genome",
"inflammasomes",
"mammalian",
"genomics",
"antibodies",
"cellular",
"structures",
"and",
"organelles",
"herpesviruses",
"microbial",
"genomics",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"viral",
"genomics",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"chemistry",
"kaposi's",
"sarcoma-associated",
"herpesvirus",
"animal",
"genomics",
"precipitation",
"techniques",
"cytoplasm",
"biochemistry",
"cell",
"biology",
"post-translational",
"modification",
"acetylation",
"virology",
"physiology",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"genomics",
"organisms"
]
| 2016 | Histone H2B-IFI16 Recognition of Nuclear Herpesviral Genome Induces Cytoplasmic Interferon-β Responses |
Chromosomal inversions have been an enduring interest of population geneticists since their discovery in Drosophila melanogaster . Numerous lines of evidence suggest powerful selective pressures govern the distributions of polymorphic inversions , and these observations have spurred the development of many explanatory models . However , due to a paucity of nucleotide data , little progress has been made towards investigating selective hypotheses or towards inferring the genealogical histories of inversions , which can inform models of inversion evolution and suggest selective mechanisms . Here , we utilize population genomic data to address persisting gaps in our knowledge of D . melanogaster's inversions . We develop a method , termed Reference-Assisted Reassembly , to assemble unbiased , highly accurate sequences near inversion breakpoints , which we use to estimate the age and the geographic origins of polymorphic inversions . We find that inversions are young , and most are African in origin , which is consistent with the demography of the species . The data suggest that inversions interact with polymorphism not only in breakpoint regions but also chromosome-wide . Inversions remain differentiated at low levels from standard haplotypes even in regions that are distant from breakpoints . Although genetic exchange appears fairly extensive , we identify numerous regions that are qualitatively consistent with selective hypotheses . Finally , we show that In ( 1 ) Be , which we estimate to be ∼60 years old ( 95% CI 5 . 9 to 372 . 8 years ) , has likely achieved high frequency via sex-ratio segregation distortion in males . With deeper sampling , it will be possible to build on our inferences of inversion histories to rigorously test selective models—particularly those that postulate that inversions achieve a selective advantage through the maintenance of co-adapted allele complexes .
Since their initial discovery in Drosophila melanogaster [1] , chromosomal inversions have been the topic of many analyses and much speculation . A growing body of literature suggests that inversions may play a role in speciation [2] , [3] , local adaptation [4] , and the maintenance of segregation distortion complexes [5]–[7] , among other potential selective mechanisms ( reviewed in [4] , [8] ) . Empirical surveys indicate that inversions are pervasive , and polymorphic inversions have been indentified in virtually all species that have been carefully scrutinized [8] , [9] . In many species , including plants , fungi , insects and humans , there is evidence that inversions respond to natural selection; however few genes or other chromosomal features that are the targets of selection have been unambiguously identified . Thus , the mechanisms of selection that affect most inversions remain unknown [8] . Owing to its position as a premier model and the facility with which inversions can be assayed cytologically , the Drosophila genus has been a favored system for studying polymorphic inversions in natural populations [10] . Nearly a century of work has yielded numerous lines of evidence that suggest strong selection governs the distributions of inversions in these species . Much of the earliest data consistent with selection on inversions was obtained from D . pseudoobscura ( reviewed in [10] , [11] ) . Although our analysis and discussion will focus on data from D . melanogaster , the patterns we observe may represent general phenomena and are consistent with evidence that has accumulated in a variety of other species [8] . Frequency clines of the most common D . melanogaster inversions are independently replicated on many continents , and quickly reestablish following colonization events [12]–[15] . Recurrent seasonal frequency shifts have been observed in numerous geographically diverse populations [14] , [16] . Finally , heterozygote superiority has been reported in both laboratory and natural populations [12] , [17] , [18] . Collectively , these findings suggest powerful selective mechanisms affect the distributions of polymorphic inversions in D . melanogaster . Despite continuing efforts , many unaddressed gaps remain in our understanding of the inversion polymorphisms of D . melanogaster . First , the breakpoints of only three inversions have been examined at the nucleotide level [19]–[21] . Second , largely due to a paucity of nucleotide data , few attempts have been made towards estimating the genealogical histories of polymorphic inversions , which may suggest selective mechanisms and inform tests of selective hypotheses ( although see [19]–[22] ) . Third , we have little data on the degree to which inversions affect polymorphism throughout chromosome arms . Finally , the selective pressures that affect the distributions of inversions in D . melanogaster have rarely been identified conclusively , with notable exceptions being inversions associated with the Segregation distortion complex [5] , [7] . Recently , Corbett-Detig et al . [23] developed a method of inversion breakpoint detection based on next-generation sequence data , and they applied this method to a large sample of African D . melanogaster genomes . In total , they identified eight polymorphic inversions in African and Cosmopolitan populations of D . melanogaster . Four inversions , termed “common cosmopolitan” , have been recovered in almost all populations worldwide [10] . These inversions have been the subjects of most population frequency assays and fitness assays in this species . Corbett-Detig et al . [23] also recovered two “rare cosmopolitan” inversions , In ( 3R ) Mo and In ( 3R ) K , and two “recurrent endemic” inversions that are only known from African populations , In ( 1 ) A and In ( 1 ) Be ( Table 1 ) [24] . Little work has focused on these rare cosmopolitan and endemic inversions , which are expected to be relatively young and therefore may provide information about the selective pressures that affects an inversion's initial rise in frequency . Here , we use these new tools in combination with data from two publically available D . melanogaster sequencing datasets [25] , [26] to investigate the genealogical histories of polymorphic inversions in these populations . Consistent with the demographic history of this species [25] , [27] , [28] , and previous work on inversions in this species [19] , [20] , [22] , our data support a recent African origin for most inversions . We examine the effects of these inversions on polymorphism throughout the genome as well as the selective models proposed to explain the initial rise in frequency and maintenance of inversions in natural populations; we find numerous examples that are qualitatively consistent with selection . Finally , conspicuous population genetic signatures suggest , and we confirm experimentally , that one X-chromosome inversion achieves a transmission advantage via sex-ratio distortion . In combination with deeper sampling , especially in ancestral African populations , it will be possible to build on our genealogical inferences to test a range of selective models in this species .
Because they remain in strong linkage disequilibrium with inversions , sequences surrounding breakpoints are often used as a means of investigating inversion genealogical histories [e . g . 19] , [20] , [22] . However , the utility with which standard genome assemblies can be used to estimate true levels of polymorphism is questionable . Pool et al . [25] note a strong correlation between sequencing depth and divergence from the reference sequence in the dataset produced by the second sequencing phase of the Drosophila population genomics project ( DPGP2 ) . They attribute this to reference bias , or the inherent ascertainment bias against non-reference alleles . It is straightforward to imagine that systematically underestimating polymorphism may downwardly bias estimates of the time since recent common ancestry . To mitigate this potential bias , we developed RAR . Briefly , this method works by aligning all reads to a reference sequence , and subsequently parsing reads and their pairs from particular genomic regions and de novo reassembling this set . This enables RAR to recruit reads that are not initially mapped to the reference , provided their paired-end does , and to resolve highly polymorphic regions including insertion and deletion polymorphisms . The interpretation of consensus quality is an unresolved issue in population genomics . Due to the additional complication of a de novo reassembly step and the difficulty of simulating data that accurately reflect true patterns of polymorphism , we favor an empirical confirmation of RAR consensus quality . Three of the strains sequenced as a part of DPGP2 have been studied extensively for PCR-based demographic analyses in this species [27] , [29] . To estimate the error rate of RAR , we downloaded more than 50 kb of PCR sequence data for each strain ( See Supplemental Table 1 for EMBL ascension numbers ) , and used RAR to rebuild the corresponding regions from next-generation short-read data . The majority of these sequences are derived from intergenic regions in one of the most diverse populations of D . melanogaster [25] . These sequences are therefore a conservatively challenging test of RAR's performance . In total , we identified 23 single-nucleotide mismatches between Sanger-PCR fragments and corresponding RAR sequences . After resequencing via PCR all fragments that contained a mismatch ( Supplemental Dataset S1 ) , we found that all discrepant sites matched the RAR consensuses . Thus , the point estimate for RAR's error rate is 0 . Assuming errors in the RAR assemblies are Poisson distributed , the upper 95% confidence interval of RAR's error rate corresponds to three errors , or approximately Q47 . While error rates are of interest , the most important and direct consequence of reference bias for population genetic inference is decreased polymorphism in resequenced individuals relative to the reference genome . In particular , reference bias is exacerbated by shallow sequencing depth [25] . We found that the RAR consensus sequences yielded nearly identical estimates of divergence to the reference genome as the Sanger-PCR sequences . The corresponding sequences produced by Pool et al . [25] using BWA and samtools [30] , [31] underestimate divergence from the reference by approximately 25% and align fewer bases ( Table 2 ) . To investigate the effect of sequencing depth on RAR's performance , we performed bootstrap replicates by discarding read pairs at random . RAR is robust to decreased sequencing depth , and can produce accurate , unbiased assemblies even with only 10% of reads retained ( ∼2× depth; Figure 1 ) . It is important to note that reference bias may persist in RAR assemblies . To whatever degree sequencing biases , such as biases in GC composition , are correlated with true patterns of genomic variation is an important potential confounding factor . That there are few models of these and related biases precludes an in-depth examination of this problem . Nonetheless , we do not observe any indirect effects of these or related biases in our validation , suggesting that RAR will be sufficient for our analyses and may be widely serviceable for a variety of other applications . We used RAR to rebuild intergenic regions immediately inside inversion breakpoints , which are expected to reflect the genealogical histories of inversions due to strongly suppressed recombination with the standard arrangement . Prior to all subsequent analyses , we removed sequences that appeared to result from exchange between arrangements ( see methods ) . If an inversion has fixed nucleotide substitutions since its formation , the time to the most recent common ancestor ( TMRCA ) is an underestimate of the true time since formation . This may be especially likely if inversions are prone to hitchhiking effects [32] due to reduced recombination . A divergence-based metric ( e . g . [22] ) , may be an appealing alternative for estimating the age of inversions . We calculated pairwise divergence ( π ) at breakpoint regions between inverted and standard haplotypes and between all standard haplotypes . Subtracting π among standard haplotypes from π between inverted and standard haplotypes and subsequently normalizing by the local mutation rate yields an estimate of the time since formation of an inversion [22] . At most breakpoints studied , this quantity is very small ( or negative; Table S2 ) , which suggests that inversions are very recently derived from the standard arrangement . Importantly , this is unlikely to stem from a bioinformatic artifact as we have taken strong precautions against underestimating divergence to a standard-arrangement reference haplotype . Hasson and Eanes [22] found that In ( 3L ) P is considerably more divergent from standard haplotypes than we estimated ( Table S1 ) . The contrast with our results likely stems from the differences in the selection of standard strains for comparison . As a result of a recent bottleneck , cosmopolitan populations have significantly decreased polymorphism relative to ancestral populations [25] , [33] . In the analysis of [22] , all but one of the standard strains are derived from cosmopolitan populations . This could cause the sampled standard sequences to be more closely related , to the exclusion of cosmopolitan inverted haplotypes , with which exchange is suppressed , than if standard strains had been selected from a diverse African population . When we recalculated the same divergence-based estimate using the standard French haplotypes in the DPGP2 dataset and all inversion-bearing haplotypes , the inverted haplotypes appear much more differentiated from the standard sequences , and our estimate ( 340 , 000 years ) is consistent with that of ref . [22] . The lack of genetic divergence between arrangements does not necessarily indicate a recent origin of inversions , as this pattern may also result from a combination of genetic exchange and occasional selective sweeps , which periodically eliminate variation , causing inversions to appear more recently derived from a standard haplotype than they are in actuality . We cannot formally exclude this explanation; however recent data from 3rd chromsome inversions of D . pseudoobscura [34] demonstrate that sequences near inversion breakpoint harbor more polymorphism than segments more distant from breakpoints and collinear regions of the genome . While it is possible that the two species differ in some other fundamental biological feature ( e . g . the mechanism or rates of recombination in heterokaryotypes , the frequency of selective sweeps , etc . ) , a simpler explanation is that the inversions of D . melanogaster have a more recent origin than those of D . psuedoobscura . The data from D . pseudoobscura therefore support the use of age estimates based on polymorphism among inverted haplotypes . Both inverted and standard allele frequency spectra , summarized as Tajima's D [35] and D′ [36] , are skewed towards rare alleles ( Table S3 ) . This skew is consistent with demographic models for the species that suggest a recent population expansion in African populations of D . melanogaster [27] , a recent range expansion , or pervasive selection [25] . However , inversion D′ values tend to be more negative than corresponding standard arrangements . In one case , In ( 1 ) Be , there are no segregating sites present on inverted haplotypes ( Table S3 ) . Given this excess of rare alleles , and paucity of polymorphisms , it is reasonable to suppose that most inversions have only recently achieved their present frequencies . Because sequences tightly linked to an inversion breakpoint effectively create a single rarely-recombining “locus” , it is not feasible to fit complex models to these data . Instead , we assume a simple model of exponential growth from the time of inversion formation through the present . This approach may be preferable to the minimum age estimate described in [19] , as it does not assume neutrality and demographic equilibrium of the population or an explicit effective population size . In addition , it is possible using our approach to quantify the variance of our estimate via an ABC method [28] , [37] , [38] . Although there are many advantages to our approach , we stress that these results should be interpreted cautiously because any simple model is unlikely to reflect an inversion's true history . In the case of In ( 2R ) NS's proximal breakpoint , we obtained a very low acceptance rate during the ABC run ( 1 . 52*10−5 ) , suggesting that the model may be a poor fit for this sequence . Other breakpoints' acceptance rates are at least one order of magnitude greater . While undoubtedly the age estimates are approximations , they should be sufficient for comparative purposes , and in the case of younger inversions , this model is likely a reasonable facsimile of the inversion's history . In two instances ( In ( 3L ) P and In ( 1 ) A ) , the posterior distributions obtained from opposite breakpoints of a single inversion were discordant ( Figure 2 ) . One plausible explanation is that by reducing local recombination rates , inversions increase the “range” of genetic hitchhiking [22] . Indeed , the large segment ( ∼2 MB ) of depressed polymorphism surrounding the distal breakpoint of In ( 1 ) A appears consistent with a hitchhiking explanation ( Figure 3 ) . In scans for selection within the Rwandan population , Pool et al . [25] found that the sequence corresponding to In ( 3L ) P's centromere-proximal breakpoint is the eighth-ranked genome-wide outlier region for Sweepfinder's [39] Λmax statistic , which is consistent with recent positive selection associated with this region of the genome . We therefore treat the oldest posterior distribution for each inversion as the better estimate , but we still provide the posterior distribution obtained for the other breakpoint in Figure 2 . Although none of the breakpoints , besides In ( 3L ) P's proximal breakpoint , are contained within the regions present in the 5% tail of Λmax distribution , polymorphism at other breakpoints may also be affected by selection on linked sites ( which may be very distant in an inversion ) . As such , a polymorphism-based approach may underestimate inversion ages , and it may be preferable to interpret these results as lower bounds of inversion ages . Under the assumptions of the exponential-growth model , we find that most inversions are quite young . Median age estimates range from 60 to 239 , 102 years ( 95% CI's 5 . 9–373 and 172 , 236–336 , 440 respectively; Figure 2; Table S3 ) , and these estimates are largely consistent with previously work on common cosmopolitan inversions [19] , [20] , [22] . As might reasonably be expected , all four endemic and rare cosmopolitan inversions appear to be younger than the four common cosmopolitan inversions . Although the majority of research has focused on the common cosmopolitan inversions ( In ( 2L ) t , In ( 2R ) NS , In ( 3R ) P , and In ( 3L ) P ) , this suggests that less-studied rarer inversions may prove to be instructive in addressing fundamental questions concerning a novel arrangement's initial increase in frequency . Based in large part on evidence from D . melanogaster , Andolfatto et al . [40] observed that polymorphic inversions in Drosophila species tend to be young relative to the TMRCA arrangement from which they are derived . Although our estimates are qualitatively consistent with this observation , particularly among the endemic and rare cosmopolitan inversions , our data suggest that many inversions segregating at moderate frequencies within D . melanogaster are significantly younger than previous findings . Although they estimated inversion ages based on a different method , it is noteworthy that Wallace et al . [34] found that many of the 3rd chromosome inversions of D . psuedobscura are an order of magnitude older than we find for D . melanogaster . As noted above , their data also indicate that nucleotide diversity is significantly higher in breakpoint-proximal regions . In short , it is unclear at present if the very young ages of inversions in D . melanogaster are a general feature of segregating inversions , or specific to this species . Undoubtedly , additional examples are needed to definitively address the apparent differences between species . However there is some evidence that the inversions of D . melanogaster are unusual with respect to its close relatives . D . melanogaster has more than 500 segregating arrangements , some of which are present throughout the species' range , but its sister species , D . simulans , harbors no inversions at polymorphic frequencies [10] , [41] , [42] . A single large paracentric inversion has fixed on the D . melanogaster lineage since its last common ancestor with D . simulans , and no inversions fixed since the common ancestor with D . yakuba approximately 13 million years ago , while the D . yakuba lineage acquired 28 inversions during this same timeframe [10] , [42] . One plausible explanation for these puzzling observations and the young ages of segregating inversions in this species , is that D . melanogaster's ancestors did not harbor polymorphic inversions , and this species' genome has only recently become tolerant of inversions . This hypothesis was originally proposed in Langley et al . [33] . It is unclear why D . melanogaster would shift from the ancestral state , which is retained in D . simulans and related island endemic species , but the evolution of inversion-tolerance could be expected to leave a detectable signal in the genome . Specifically , genes functionally important for achiasmatic segregation may be expected to show evidence of positive selection specific to D . melanogaster or to inversion-tolerant lineages , and we may observe geographically-structured selection associated with populations that contain different frequencies of segregating inversions . While previous studies [19] , [20] , [22] have attempted to estimate inversion ages , none has directly considered geographic origins . An analysis such as this may inform our understanding of inversion genealogical histories , and suggest potential selective mechanisms . One feature of D . melanogaster's demographic history is useful in this regard: this species emerged from ancestral African populations and colonized the rest of the world approximately 10 , 000–15 , 000 years ago [28] , [29] . During this expansion , cosmopolitan populations experienced a sharp bottleneck , which reshaped patterns of nucleotide variation genome-wide . This bottleneck event left a detectable signature in nucleotide data , and it can be used to estimate sub-Saharan vs . cosmopolitan origin for specific haplotypes [25] . Using a simple divergence metric , we judge six of the eight inversions studied to be African in origin ( In ( 2L ) t , In ( 2R ) NS , In ( 3L ) P , In ( 3R ) K , In ( 3R ) P , and In ( 1 ) A ) , as they are all approximately equally divergent from the African and French sequences and their nearest neighbor is invariably African—this is expected for African haplotypes given the demographic model proposed by ref . [25] ( Table S5 ) . In ( 1 ) Be and In ( 3R ) Mo , which are also the two youngest inversions studied , appear to be two exceptions to the predominantly African origins of inversions . Including the breakpoints , there are four large haplotypes in strong linkage disequilibrium with In ( 3R ) Mo ( Figure 3 ) [36] . FR310 , the only DPGP2 line that contains In ( 3R ) Mo , also contains these haplotypes , and the sequence at each is on average more divergent from African than French lines . Additionally , the least divergent individual haplotype is invariably French . Collectively , these considerations provide strong evidence that In ( 3R ) Mo is cosmopolitan in origin; however , note that In ( 3R ) Mo is not as closely related to other French genomes as they are to each other ( Table S5 ) , suggesting that this inversion may have originated in a different cosmopolitan population . In ( 1 ) Be has almost no segregating sites across the entire 1 . 7 Mb length of the eight samples of this inversion ( Figure 3 ) . We find that this haplotype is less divergent on average from French than from African genomes , and that its closest relative at each breakpoint is French , not African ( Table S5 ) . This suggests the inversion captured a cosmopolitan haplotype , a conspicuous finding in light of its young age and the fact that this inversion has never been reported outside of Africa even though many cosmopolitan populations have been extensively surveyed [10] . Despite their similar geographic origins , In ( 3R ) Mo displays the opposite distribution of In ( 1 ) Be . In ( 3R ) Mo has been identified almost exclusively in cosmopolitan populations , having only been reported from a single South African population [24] , which is likely highly admixed [43] . It is interesting to note that introgression patterns of cosmopolitan inversions appear to be the opposite of collinear regions of the genome , in which autosomal chromosomes exhibit significantly more cosmopolitan admixture in Africa than the X chromosome [25] . Predicted geographic origins of inversions agree with and lend further support to the polymorphism-based estimates . That is , both cosmopolitan inversions appear to be younger than the predicted time of the out-of-Africa migration of D . melanogaster ( Figure 2 ) , which suggests that we have not overestimated inversion origins based on polymorphism . Somewhat more compelling is the observation that at least one of the breakpoints of all African-originated inversions appears to be older than the predicted timing of the out-of-Africa demographic event ( Figure 2 , Table S4 ) . Although this is not a necessary condition for consistency between these analyses ( an inversion could originate on an African haplotype after the cosmopolitan expansions ) , the fact that we do not observe this pattern suggests that hitchhiking effects may not have drastically affected age estimates at both breakpoints of the same inversion . The haplotype admixture analysis of [25] sought to resolve sub-Saharan versus cosmopolitan ( admixed ) ancestry in a panel of African genomes that included chromosome arms bearing inversions . Although the question of recent population ancestry is distinct from that of inversion origin ( e . g . a haplotype carrying an arrangement that originated in Africa could be considered “cosmopolitan” if it went through the out-of-Africa bottleneck ) , it may be worthwhile to examine the potential influence of inversions on demographic inferences of this type . Whereas we focus on the comparison of chromosomes with known inversion genotypes , the analysis of Pool et al . [25] considers all Rwandan and French genomes as representative of cosmopolitan and African haplotypes regardless of arrangement , and regions around inversion breakpoints are often flagged as admixed by their analysis . This result may be due to powerful effects of inversions on haplotype structure , which is most pronounced when inversions are at different frequencies between populations . Consistent with this explanation , African strains that bear In ( 3L ) P , In ( 3R ) P , In ( 3R ) K and In ( 2L ) t ( all of which are present in the French population ) , are often identified as admixed near breakpoint regions . In particular , African In ( 3L ) P haplotypes are masked up to approximately 1 . 5 Mb from the breakpoints by the admixture-detection method of ref . [25] . This could result from this inversion's absence from the Rwandan “African reference” population . Especially in light of their strong population frequency differences and their extended chromosomal influence on diversity ( see below ) , it appears that inversions may strongly impact demographic analyses , potentially resulting in spurious inferences if they are not considered individually . Given the powerful effect of inversions on nucleotide sequences near to their breakpoints , it is natural to ask whether inversions also affect polymorphism in more distant regions of the chromosome . Neutral models of exchange predict that inversions should be genetically indistinguishable from standard haplotypes towards the middle of inverted regions shortly after achieving equilibrium frequencies [44] . Nonetheless , there are numerous instances in which different chromosome arms within the DPGP2 dataset produce substantially different estimates of nucleotide diversity [25] . In the French population , levels of polymorphism on the autosomal arms correlate with the number of inversions present . Removing lines bearing common inversions sharply reduces this effect ( Figure 4 , see also [25] ) . To investigate the effects of inversions on estimates of polymorphism in additional populations , we compared pairwise nucleotide diversity ( π ) on chromosome arm 3R in the France , Rwanda , and Gabon populations both including and excluding inversion-bearing haplotypes ( Figure 4 ) . In African populations , we observe only a modest effect of inversions on nucleotide diversity . Diversity in Rwanda is slightly increased , likely owing to the low frequency of In ( 3R ) P in this population ( Table S6 ) ; in Gabon nucleotide diversity is slightly reduced , perhaps because of the high frequency of In ( 3R ) P in this population and the low diversity within this arrangement . In the French sample , we observe a sizable ( ∼30% across all of 3R ) increase in nucleotide diversity when inversions are included ( Figure 4 , see also [25] ) . Inversion-mediated effects on nucleotide diversity may be pertinent on a genome-wide scale as well . Principle component analysis , as described in [25] and following the method of [45] , within the Rwandan samples suggests that inversions are responsible for the majority of genetic structure in this population ( Figure 5 ) . Importantly , this does not appear to be limited to breakpoint regions , instead inversions affect polymorphism through the majority of chromosome arms ( Figure 4 ) . Consistent with previous work in D . pseudoobscura and related species [46] , we observed increased differentiation between arrangements in regions up to 4 Mb outside inversion breakpoints ( Figure 4 ) . Note also that inversion-bearing chromosome arms are more closely related to individuals that have the same arrangement in other populations than to standard haplotypes within their own population ( Figure 6 ) . It is known that recombination in heterokarytopyes within inverted regions is infrequent [4] , [8] , [10] , [47] . Existing estimates in Drosophila suggest a neutral recombination rate of approximately 10−4 for double recombination events towards the center of inversions [47] , and theoretical predictions suggest exchange rates may be as large as 10−2 in the center of large inversions [44] , [48] . As the inversions studied here are young , some level of differentiation may be attributable to their unique origins and suppressed recombination . Still , even low rates of exchange are expected to rapidly eliminate genetic differentiation between arrangements [44] , [48] . A likely explanation of differential diversity associated with inversion-bearing haplotypes is that inversions migrate at different rates between populations than standard haplotypes . In particular , arm 3R inversions , especially In ( 3R ) P and In ( 3R ) K , increase diversity by ∼30% in the French population ( Figure 4 ) . Because different cosmopolitan populations typically contain similar sets of genetic variants [49] , it seems likely that most inversion bearing haplotypes present in the French sample are recent migrants from African or African-admixed populations . As described above , chromosome arms with fewer or no inverted haplotypes in this sample show concordant decreases in polymorphism , which supports a differential-migration interpretation . However the evolutionary drivers of inversion introgression remain largely unknown . Although inversions retain some genetic differentiation from standard haplotypes , in African samples FST decays quickly with increasing distance from breakpoints in all inversions except In ( 1 ) Be ( Figure 3 ) ; thus we expect it will soon be possible to test hypotheses regarding the selective maintenance of co-adapted alleles ( reviewed in [4] , [8] ) . Because we lack specific knowledge of neutral recombination rates in heterkaryotypes and population demographic models , and because the African sampling is distributed among many geographically diverse populations , we do not feel these data are suitable for a rigorous quantitative test of these selective hypotheses . Nonetheless , we do note numerous regions of decreased polymorphism and strong genetic differentiation between arrangements that are qualitatively suggestive of selective mechanisms in many inversions ( Figure 3 , Figure 7 ) . As noted in [33] , In ( 3R ) Mo , which is present in ∼12% of the strains in the DGRP sample , is in strong linkage disequilibrium with two large haplotypes that are not immediately adjacent to inversion breakpoints . One of these haplotypes lies outside the inversion , between the distal breakpoint and telomere ( Figure 3 ) . These haplotypes are shared with the single In ( 3R ) Mo bearing line in the DPGP2 dataset . Thus this pattern of long-range linkage disequilibrium is not limited to the Raleigh population and may instead be a geographically widespread phenomenon . In ( 1 ) A displays a similar pattern in the regions surrounding the distal breakpoint , with numerous haplotypes in strong linkage disequilibrium with the inversion ( Figure 3 ) . It is not clear whether these and other conspicuous patterns of variation are consistent with the maintenance of co-adapted alleles [e . g . 4] , [50] or selective sweeps [51] specific to one arrangement [33] . This important distinction is in some ways an extension of the ongoing debated between the relative prevalence of background [52] and positive selection [51] and demands further analysis updated with information of the rates of exchange of specific inversion heterokaryotypes . Likely resulting from the out-of-African bottleneck , the data show that inversions tend to be more genetically differentiated from the standard arrangement in cosmopolitan populations ( Figure 7 ) . It therefore appears that the ancestral African populations would be the best suited for fine-mapping alleles that are associated with alternative arrangements via differential selection , and future research in this field should concentrate on samples derived from this region . Although the hypothesis has received less attention in the literature , inversion breakpoint mutations may also be the target of selection that affects their evolutionary outcomes [21] . We find that few inversion breakpoints disrupt genic sequences and associated regulatory regions ( Table S7 ) . Of the three inversions with simple cut-and-paste breakpoints ( see [42] for a description of inversion breakpoint structures ) , In ( 3R ) Mo has one breakpoint situated in an exon , and In ( 2L ) t's distal breakpoint truncates the 3′ untranslated region of CG15387 . Many of the inversions with inverted duplications at each breakpoint also interrupt transcribed sequences . In most cases , the duplicated portion may retain an intact copy ( e . g . In ( 3R ) P [20] ) . However , two inverted-duplication bearing inversions , In ( 1 ) A and In ( 1 ) Be , have both breakpoints of the duplicated regions situated in single genes ( Table S7 ) . Thus , four inversions breakpoints may have produced structural or regulatory changes in genic sequences . In ( 3L ) P's distal breakpoint also interrupts a transcribed cDNA [21] , but presently there are no annotated transcripts that correspond to this region . While it is possible that breakpoint mutations are the selective mechanisms by which inversions achieve polymorphic frequencies [21] , we favor a model in which these effects are deleterious byproducts of the inversion formation . The proportion of breakpoints that interrupt genic sequences is significantly fewer than expected if we assume breakpoints form randomly with respect to genic sequences and uniformly across chromosome arms ( P = 0 . 000122; Permutation Test ) . Furthermore , two studies that focused on inversion fixations between Drosophila lineages do not report interrupted genic sequences associated with inversion breakpoints [42] , [53] . Thus , breakpoint mutations may often oppose inversions' fixation in natural populations . Deleterious consequences of interrupted genic sequences cannot be too severe , since all of the inversions that we studied are regularly homozygous in phenotypically normal isofemale lines , and those situated on the X chromosome must often exist in hemizygous states in natural populations . To this point , we have characterized inversion genealogical histories , and presented evidence that selective mechanisms affect their distributions . Of course , it would be of interest to identify specific sources of natural selection that affect the distributions of inversions . That In ( 1 ) Be arose on a cosmopolitan haplotype and is currently invading African populations suggests that this inversion may harbor a sex-ratio distortion complex . Though there are no known cases from D . melanogaster , X chromosome inversions in other Drosophila species are commonly associated with sex-ratio distortion [6] . There is prior evidence that cosmopolitan X chromosomes from this species can drive against African Y chromosomes [54] . Hence , a wealth of background information , in combination with suggestive population genetic signatures , prompted us to investigate sex-ratio distortion as one potential mechanism influencing In ( 1 ) Be . In seven of the eleven experimental crosses , we find significant evidence for distortion at the α = 0 . 05 level and in all crosses the average sex ratio trended towards females . Three of the experimental crosses remain significant after applying a Bonferoni correction for multiple testing . None of the control crosses , which are derived from many of the same populations as In ( 1 ) Be bearing strains , show significant evidence for a transmission bias ( Table 3 ) . It is formally possible that this inversion does not drive , but that the test lines we selected also happen to contain a segregation distortion complex outside of the inversion . Because it is polymorphic for this inversion , strain RG11N is a more definite control . We find that that male progeny from RG11N mothers that inherit In ( 1 ) Be transmit their X chromosome at higher than Mendelian expectations , while those that inherit the standard arrangement do not ( Table 3 ) . To exclude differential viability as an explanation , we counted all eggs laid by females mated to RG11N ( In ( 1 ) Be ) /ZS30 ( Y ) F1 males . Even when we conservatively assume that all preadult mortality is suffered by males , we find a significant excess of female offspring ( P = 0 . 0248; Table 3 ) . It is plausible that In ( 1 ) Be's recent rise in frequency is due to selection favoring this transmission bias . The strength of drive ( k∼0 . 541 ) , though weak by comparison to other sex-ratio distortion systems in many other Drosophila species ( reviewed in [6] ) , is substantial when compared to most selection coefficients in the genome and is similar to an existing estimate of drive strength of cosmopolitan X chromosomes against African backgrounds ( k∼0 . 61 ) [54] . In ( 1 ) Be is within the 95% confidence interval of map location for this complex , though this interval is very wide [54] . The Recovery Disrupter sex-ratio distortion complex , which may be the same as [54] studied , has been mapped more precisely to cytological band 1–62 . 9 , and has been suggested to act in natural populations [55] , [56] . This cytological band also contains the proximal breakpoints of both In ( 1 ) Be and In ( 1 ) A . It is possible that a locus near band 1–62 . 9 may have recurrently evolved drive during the recent evolutionary history of this species , and has acquired at least one recombination-suppressing inversion . This may explain the extreme proximity ( 816 bp ) of In ( 1 ) Be and In ( 1 ) A's proximal breakpoints ( but we note that breakpoint reuse may result from neutral mechanisms [42] ) . In pilot crosses , we did not observe sex-ratio distortion associated with In ( 1 ) A ( not shown ) . Since this inversion is considerably older than In ( 1 ) Be , suppressors specific to the driving allele captured by In ( 1 ) A may have achieved high frequencies , effectively masking distortion associated with In ( 1 ) A . Similar results have been inferred in other sex-ratio distortion systems in Drosophila ( reviewed in [6] ) . Testing In ( 1 ) A for distortion on a wider range of African and cosmopolitan lines is a target of future research .
The majority of existing work on inversions has focused on well-established polymorphisms . Young inversions provide a valuable counterpoint because they yield a glimpse of the mechanisms that lead to their initial rise in frequency . The forces involved in the initial rise need not be the same as the ones involved in long-term maintenance of inversions; examples from young inversions are essential to addressing this potential difference . In the case of In ( 1 ) Be , we have identified a likely mechanism for this inversion's rapid increase in frequency , namely , sex-ratio distortion . Young , rare inversions are also commonly associated with Segregation distortion haplotypes in this species [5] including one that is currently sweeping African populations [7] , suggesting that distortion may be a common means by which inversions initially achieve high frequencies . Therefore , testing inversions for segregation distortion may be a fruitful approach . Finally , for In ( 3R ) Mo , the linked-haplotypes found outside of the breakpoints also make attractive targets for genetic dissection . Numerous models of inversion evolution posit that selection favors different alleles in alternative arrangements . Because they will be more amenable to surveys via population genetic modeling , older inversions of D . melanogaster provide an ideal system to test these hypotheses . That is , because recombination has had more time to decouple selective and neutral processes , older inversions should afford better resolution of specific alleles in linkage disequilibrium with inversions . Though importantly , our data suggest that sampling should be focused on ancestral African populations where selective and demographic/neutral patterns of variation may be more easily distinguished . Though low levels of genetic differentiation remain between arrangements , we observe rapid decay of genetic differentiation with increasing distance from breakpoints . Hence , in combination with neutral estimates of recombination in heterokayotypes , it will soon be possible to test widely popular selective hypotheses that suggest inversions achieve high frequencies via maintaining linkage disequilibrium between co-adapted alleles . One important message of this work is that inversion data should be interpreted with caution . Inversions interact powerfully with diversity in the sequences immediately proximal to their breakpoints and chromosome wide . Failure to account for inversions may lead to spurious results ( nonetheless , it is probably wise to exclude these regions from population genetic analyses that assume normal recombination ) . As we have shown , inversions also have diffuse effects on polymorphism , which may further complicate demographic modeling by producing substantial arm-specific effects . Thus , even population genomic studies that are not focused specifically on inversions cannot ignore their presence in the data . The reverse is also true; population genetic analyses focused on inversions may be affected by sampling if the recent demographic history of the species is not explicitly considered . Far from being a nuisance in the analysis of population genomic data , inversions may prove fertile ground for the study of genome evolution and mechanisms of selection . We hope that our analysis will help reignite interest in naturally occurring inversions . Although studied extensively for almost a century , little progress has been made towards conclusively understanding the selection that affects inversion polymorphisms . With the increasing availability of genomic techniques , it is now possible to reopen many longstanding questions . In combination with an exceedingly well-curated reference genomes and an enormous body of literature , these bioinformatic and computational tools make the inversion polymorphisms of D . melanogaster an appealing model system once more .
The majority of analyses in this study are focused on sequence data from second sequencing phase of the Drosophila Population Genomics Project ( hereafter DPGP2; www . dpgp . org ) . See [25] for a description of this primary data set , which was the source of all African and European samples . We limited most our analyses to the target , ‘core’ genomes described in [25]; although we included all strains that contained inverted haplotypes regardless of core or addendum status in breakpoint analyses . To study inversions from one cosmopolitan population , we downloaded assemblies [26] and short read data ( NCBI SRTA ) for the Drosophila genetic resource panel ( DGRP ) , which is derived from a population from Raleigh , NC . For analyses of breakpoint regions , we apply RAR to generate unbiased sequence data . Because the dataset produced by Mackay et al . [26] is comprised of numerous sequencing technologies , we restricted breakpoint analyses to lines that had been sequenced with illumina paired-end reads . Analyses focused on comparative polymorphism across chromosome arms , such as windowed π and FST analyses , rely on the assemblies produced by [25] and [26] . Because Mackay et al . [26] did not mask regions that fail to inbreed , their assemblies contain substantial residual heterozygosity . These regions are obvious under cursory inspection [23] , [33] , so we masked all chromosome arms that demonstrated long tracks ( identified using 1/2 mb windows ) of residual heterozygosity from all analyses . We generated inversion genotypes for each line by including the breakpoint-spanning contigs produced by Corbett-Detig et al . [23] with the standard D . melanogaster reference sequence during initial mapping . Reads overlapping an inversion breakpoint by more than 20 bp were considered evidence that the stock bears the inversion . In all cases , there is a perfect correspondence between breakpoint genotypes , mate-pairs that span an inversion breakpoint ( where applicable ) , the inversion validations performed by [33] , and with our unpublished data . See Table S6 for a list of inversion genotypes in each line in DPGP2 . See supplemental tables S8 , S9 , and S10 for inversion genotypes for In ( 2L ) t , In ( 2R ) NS and In ( 3R ) Mo respectively in the DGRP lines . Results are provided in this format due to the presence of residual heterozygosity in DGRP sequence data . That is , we only report individuals that appear to be fixed for the inversion of interest , but do not want to give the impression that other lines may not also contain these inversions . We do not report other inversions , which are surely present in DGRP , because we did not attempt to validate any of these genotypes due to small sample sizes in that panel . See [33] for many PCR-validated inversion genotypes in lines derived from this panel . We aligned all short-read data to the D . melanogaster reference genome v5 . 31 [57] using BWA v0 . 5 . 9 [30] . We extracted all read-pairs if either read aligned within 500 bp of a region of interest , and de novo assembled all reads for each region using PHRAP v1 . 090518 ( http://www . phrap . org/phredphrap/phrap . html ) . PHRAP command line parameters used were ‘-forcelevel 10’ , ‘-minmatch 15’ , ‘-vectorbound 0’ , and ‘-ace’ . We converted these ‘ . ace’ assemblies to SAM format using custom perl scripts , and generated a consensus from the resulting alignment using samtools [31] , where we required a minimum depth of 3 and a minimum nominal quality of 50 . Finally , we extracted sequences of interest by using cross_match v1 . 090518 ( http://www . phrap . org/phredphrap/general . html ) to align the corresponding region from the reference genome to the consensus . We required a minimum alignment length of 100 bp ( Figure S1 ) . Three of the strains ( ZK84 , ZK131 , and ZK186 ) that were resequenced as a part of DPGP2 have been studied extensively via Sanger-PCR sequencing [27] , [29] ( EMBL ascension numbers are provided in Table S1 ) . For each , there is more than 50 kb of high quality X-chromosome sequence data available . This population is among the most diverse analyzed for this project or ever identified in the species . In addition , these sequences are primarily derived from intergenic regions , and are therefore a conservatively challenging test for RAR's performance . Using RAR , we assembled sequences corresponding to each available PCR fragment in each line . We resequenced via PCR those fragments where we identified mismatches between the RAR consensus and PCR-derived sequences . All PCR was performed on the original DNA extraction used for library preparation . The generated PCR traces were aligned to the original EMBL sequence and to the RAR assembly using clustalW version 2 [58] . We also experimented with additional iterations ( i . e . replacing the reference with corresponding RAR contigs and realigning all reads to this augmented reference sequence ) , but observed little improvement relative to a single reassembly ( not shown ) . To investigate the effect of decreasing read depth , we reran RAR after randomly discarding 10 to 90 percent of the reads ( in 10% intervals ) , on these same regions . We performed 100 bootstrap replicates at each proportion of reads discarded for each line . The resulting contigs were then aligned to both the Sanger-PCR sequences as well as the reference , and their divergence from each recorded . We assembled sequences for each line immediately inside each inversion's breakpoints using RAR . Alignments were performed using clustalW version 2 [58] . All alignments were inspected with assistance from PERL scripts , which we designed to flag problematic regions surrounding indels and SNPs shared between inverted and standard arrangements . Multiple alignments for all SNPs within 10 bases of an indel and all shared polymorphism was inspected manually . In all but two cases , shared polymorphisms were present on an inverted haplotype flanked with other shared polymorphisms . This is an expected signature of genetic exchange between arrangements , and we masked all sequences that we inferred resulted from recombination between inverted and standard arrangements . Finally , we estimated local rates of mutation by aligning the reference sequence from regions we used for demographic analyses with a recently improved D . simulans reference genome [59] . At In ( 3R ) P's distal breakpoint , genetic exchange with the standard arrangement has been extensive , and we could not confidently determine which samples retained the original haplotype that the inversion arose on . We therefore excluded this breakpoint from all subsequent analyses . For all other breakpoints , we were able to infer the original haplotype captured by the inversion event , and we discarded all recombinant haplotypes from downstream analyses . Empirical estimates of summary statistics , specifically π , θW , Tajima's D , and D′ , were obtained by treating all missing data and indels as complete deletions . Simulations were performed in ms [60] following a rejection-sampling approximate Bayesian computation approach . Briefly , we modeled each inversion as a population that has grown exponentially at the same rate since its formation through the present . We calculated the expected theta of the current inverted population as:Where L is the total length of the aligned sequences . d is the nucleotide divergence per site between D . simulans and D . melanogaster per site . t is an approximation of the time of divergence between these two species in generations ( 30 million ) . Ne is a widely used estimate of the D . melanogaster effective population size ( 106 ) , and f is the frequency of the inversion in the DPGP2 or DGRP datasets as appropriate . To accommodate uncertainty in our estimate of , we selected values for simulations from uniform ( 0 , *10 ) . We defined the tolerance for the number of segregating sites and π as no more than 5% different from empirically obtained data from the sequence alignments . We stored alpha , theta , and TMRCA of the sample for each accepted simulation , and ran each until at least 10 , 000 simulations were accepted for each inversion breakpoint . We obtained a posterior distribution of estimates for the age of each inversion as:Where is in units of substitutions per region per year , and TMRCA is in units of 4N0 . Thus this equation yields a distribution of the estimates of the age of the inversion in years . To estimate geographic origins , we compared breakpoints regions of each inversion with both the cosmopolitan ( FR ) and African ( RG ) populations . We noted both the average divergence to lines that bear the standard haplotypes in the French population and African populations , as well as the nearest neighbor . We judged each inversion as cosmopolitan if both the nearest neighbor was a standard French sequence , and the average divergence between the French population and this inversion was less than the average divergence from African sequences . We initially tested the eight DPGP2 strains identified as carrying In ( 1 ) Be for fixation of the inversion using PCR primers we have developed ( unpublished work ) . We identified three strains , GA191N , KR39 , and RG10 that have fixed In ( 1 ) Be . RG11N is segregating for this arrangement . For control crosses , we selected three strains from the same populations , RG22 , RG35 , and KR42 , and we confirmed that each is fixed for the standard X chromosome arrangement . Virgin females from these stocks were crossed to five strains , which are known from previous work to be susceptible to cosmopolitan sex-ratio distortion , K12 , C5 , C17 , ZS30 , and ZS53 [54] . The resulting male progeny were then crossed individually to two virgin Oregon-R females , aged 3 to 8 days . All crosses were performed in vials on standard corn-meal medium supplemented with yeast and maintained at 25°C . After three days we discarded the parents . We counted male and female offspring each day after the first flies emerged until 15 days after removing the parents . Crosses to test for differential viability as an explanation of the observed sex-ratios were performed identically , except that parents were flipped to a new vial every eight hours during the laying period , and we counted eggs immediately afterwards . Windowed summary statistics for inverted and standard populations were calculated based on the assemblies produced by Pool et al . [25] and Mackay et al . [26] . We masked all putatively heterozygous sites prior to this analysis . In both datasets , approximately 1% of non-reference alleles are heterozygous outside of residually heterozygous regions . Although this is a relatively small proportion , this practice of excluding heterozygous sites may be dangerous in serious quantitative analyses; however , for our purposes , which are largely oriented towards qualitative , broadscale observations , this is unlikely to present a major issue . We calculated π without applying any sampling thresholds . We calculated FST as described in [61] , excepting that we did not weight polymorphism estimates by the sample size . | Chromosomal inversions are known to respond to powerful natural selection in many species . Despite this evidence , little progress has been made towards understanding the nature of selection that affects inversions . Here , we utilize two recently released population-resequencing projects from D . melanogaster to address many of the unknown features of polymorphic inversions . We find evidence that inversions in this species are generally very young , with ages on the order of hundreds to tens of thousands of years , and that the majority of inversions originated in ancestral African populations . Inversions are also the source of the majority of genetic structure within populations and affect polymorphism chromosome-wide . We are able to confirm experimentally that one X-chromosome inversion achieves an advantage by selfishly increasing its transmission through males . Future work will build on our basic inferences to identify potential selective mechanisms and candidate genes in the other inversions studied . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Conclusion",
"and",
"Prospects",
"Methods"
]
| [
"population",
"genetics",
"biology",
"population",
"biology",
"genetics",
"and",
"genomics"
]
| 2012 | Population Genomics of Inversion Polymorphisms in Drosophila melanogaster |
Glucocorticoids are essential for life , but are also implicated in disease pathogenesis and may produce unwanted effects when given in high doses . Glucocorticoid receptor ( GR ) transcriptional activity and clinical outcome have been linked to its oligomerization state . Although a point mutation within the GR DNA-binding domain ( GRdim mutant ) has been reported as crucial for receptor dimerization and DNA binding , this assumption has recently been challenged . Here we have analyzed the GR oligomerization state in vivo using the number and brightness assay . Our results suggest a complete , reversible , and DNA-independent ligand-induced model for GR dimerization . We demonstrate that the GRdim forms dimers in vivo whereas adding another mutation in the ligand-binding domain ( I634A ) severely compromises homodimer formation . Contrary to dogma , no correlation between the GR monomeric/dimeric state and transcriptional activity was observed . Finally , the state of dimerization affected DNA binding only to a subset of GR binding sites . These results have major implications on future searches for therapeutic glucocorticoids with reduced side effects .
Glucocorticoids influence the activity of almost every cell in mammalian organisms , mainly through binding to the glucocorticoid receptor ( GR ) . In the absence of ligand GR primarily localizes in the cytoplasm while the activated GR-ligand complex is mainly nuclear . Once in the nucleus , the GR regulates gene expression by directly binding to specific DNA sequences or by the interaction with , and modulation of other transcription factors [1] . These two main mechanisms of action were historically named GR transactivation and GR transrepression , respectively [2] . Even though GR homodimerization is considered an essential step in the GR-transactivation pathway , it is still not clear whether GR dimerizes before [3]–[6] or after [7]–[9] DNA binding; or which regions of the protein are functionally involved in the homodimerization process [10] . Nevertheless , as GR transactivation was originally correlated with side effects of long-term clinical use of glucocorticoids , intense efforts have been made to design GR ligands with “dissociated” glucocorticoid properties that exclusively activate the transrepression pathway [11] . Since the current model of the GR mechanism of action states that the monomeric/dimeric status of the receptor defines its transcriptional activity , most of the rational drug design strategies have been focused on the search for ligands that promote the monomeric ( i . e . , transrepression ) form of GR [12] . GR is a modular protein organized into three major domains: the N-terminal ligand-independent activation function-1 domain; the central DNA-binding domain ( DBD ) ; and the C-terminal ligand-binding domain ( LBD ) [13] . Crystal structures of both DBD [14] and LBD [15] have been obtained separately but no reports have described a structure of the entire protein . The first crystal structure of the GR DBD revealed a dimerization region , and subsequent mutational studies partially defined a five amino acids sequence , named the D-loop , that could potentially be involved in GR dimer formation [8] . However , these earlier studies were performed with a GR fragment and entirely in vitro . Following this work , a point mutation within the human GR DBD ( A458T ) in the context of the entire protein was reported to be able to separate transactivation from transrepression and unable to dimerize [16] , although no direct evidence supported the latter conclusion . The human GRA458T , mouse GRA465T , and rat GRA477T have been commonly referred to as the “GRdim” mutants [17] . From a transcriptional standpoint , early studies characterized the GRdim mutant as unable to transactivate genes but able to transrepress both in vitro [16] and in vivo [18] . However , GRdim's inability to transactivate has been challenged after results that showed this mutant can induce gene expression in a sequence and context-dependent manner [19]–[21] . From a biophysical standpoint , the early GRdim studies established that dimerization was entirely dependent on the DBD region . However , a recent study confronted this idea by showing protein-protein interactions between GRdim molecules [22] . Here we performed in vivo mapping of the GR oligomerization state by using the number and brightness ( N&B ) method [23] . We present conclusive evidence showing dimerization of the GRdim mutant while an additional mutation in the LBD ( I634A ) severely compromises homodimer formation . Importantly , no correlation between oligomerization state , DNA binding , and transcriptional activity could be established . These results question a key paradigm in the quest for glucocorticoid “dissociated” ligands .
To determine the state of GR dimerization in living cells , we performed the N&B method [23] . This novel technique , based on moment-analysis , provides the average number of moving fluorescent molecules and their brightness at every pixel in the images ( Figure 1A and 1B ) . In the simplest case the brightness of an oligomer consisting of n monomers is n-times the brightness of the n-monomers . Therefore , N&B is a useful method to obtain the oligomerization state of proteins in living cells with high spatial resolution . Figure 1C shows the nuclear brightness ε ( i . e . , measure of fluorophore oligomerization ) corresponding to the wild-type enhanced green fluorescent protein ( eGFP ) -GR expressed in baby hamster kidney ( BHK ) cells . As we previously demonstrated [24] , ε values significantly increased ( approximately 2-fold ) in the nucleus of cells treated with dexamethasone ( Dex ) , consistent with a virtually complete population of GR dimers upon ligand addition . eGFP brightness is statistically indistinguishable from unstimulated eGFP-GR , indicating that nuclear eGFP-GR is mostly monomeric in the absence of ligand . Similar results were observed in the presence of the natural ligand corticosterone ( Cort ) ( Figure 1C ) . As a negative control , no GR dimer formation was observed in cells treated with the non-steroidal ligand compound A ( CpdA ) , in agreement with previous studies [25] . Recently , an in vitro study reported that the GR exists mostly as a monomer [26] . If that were the case in vivo , we would have detected an average brightness below two-fold in our system because of a linear-weighted-average combination from the contribution of the monomer/dimer population . Nevertheless , we cannot rule out the existence of a small population of monomeric molecules or even a small proportion of other oligomers . Previous reports have suggested that GR dimer formation is an irreversible process in vitro [26] . Thus , we evaluated the stability of GR dimers in vivo by performing washout experiments . Interestingly , Cort withdrawal significantly reduced the population of GR dimers ( Figure 1C ) , even though GR remained in the nucleus ( Figure 1D ) , demonstrating that dimerization is a reversible process in vivo . Importantly , Dex washouts did not affect dimerization most likely due to the high affinity of this ligand for the receptor [27] . As we previously described [24] , in our N&B assay there is an excess eGFP-GR molecules due to over-expression in comparison to accessible glucocorticoid response elements ( GREs ) at a given time . Moreover , any given GRE is only transiently bound by GR during physiological transcriptional activation [28] , [29] . Hence , the virtually complete population of dimers observed is more compatible with a DNA-independent model for GR dimerization . Next , we decided to test the oligomerization status of the GRdim mutant using N&B . Interestingly , treatment of eGFP-GRA465T expressing cells with Dex showed an increase in nuclear brightness virtually identical to that observed with the wild-type receptor ( Figure 2A ) , clearly demonstrating that this mutant is able to form dimers in vivo . Interestingly , the weaker natural steroid Cort also induces significant GRA465T dimerization although with less efficiency than Dex ( Figure 2A ) . Recently , it has been suggested that GR expression levels affects the dimerization status of the receptor [30] . Since we were working in an over-expression system due to the transient transfection of eGFP-GR , we decided to study GR oligomerization status in a model expressing physiological levels of the receptor . We generated mouse embryonic fibroblast ( MEF ) cell lines from a GR null mouse stably expressing a mouse eGFP-GR protein at endogenous levels ( Figure S1 ) . N&B analysis of the MEF cell lines showed that the wild-type GR fully dimerizes in the presence of Dex and that the GRdim also forms dimers , although with a slightly less efficiency than the wild-type receptor ( Figure 2B ) . Nuclear translocation was similar for both the GRwt and the GRA465T mutant ( Figure 2C ) . In conclusion , both ligand affinity and GR expression levels have no apparent effect on the dimerization status of the wild-type receptor . On the other hand , the GRdim oligomerization status is mildly sensitive to both ligand and receptor expression levels . Another alleged property of the GRdim is its inability to bind DNA [16] , [18] , although it has also been questioned [19] , [20] , [22] , [31] . To address this , we evaluated in vivo recruitment to DNA in the 3617 mouse cell line , which contains an amplified array of a GR responsive promoter structure ( the mouse mammary tumor virus [MMTV] array ) . Thus , eGFP-GR interactions with MMTV GREs can be directly visualized in living cells as a bright spot [32] . Figure 2D clearly shows array formation on both GRwt and GRA465T receptors . In summary , the GRdim seems to be able to dimerize and to bind DNA in vivo . The transcription factor NF-κB mediates key inflammatory pathways and its interaction with GR has been widely documented [1] . NF-κB is mainly composed of the heterodimer p50/p65 , although p65 homodimers have also been described [33] . The transrepression hypothesis sustains that the GR interacts with p65 exclusively as a monomer; however , this idea relies almost entirely on the GRdim paradigm . The fact that monomeric GR molecules like CpdA-GR complexes are able to transrepress [34] does not rule out the possibility that GR dimers would also be capable of transrepression . To test this , we assessed the “dimeric transrepression” hypothesis by analyzing the oligomerization state of mCherry-GR in the presence of GFP-p65 . Figure 3A shows N&B analysis of cells expressing mCherry or mCherryGR in the presence or absence of GPFp65 , and Figure 3B contains representative images of these cells . As expected , in the presence of ligand mCherryGR showed full GR dimerization ( Figure 3A ) . If GR interacts with p65 as a monomer , then GFP-p65 presence should decrease the population of mCherryGR dimers . However , no effect on mCherryGR oligomerization state is observed when GFP-p65 is present ( Figure 3A ) . Brightness analysis also confirms that GFP-p65 dimerizes upon TNF addition . Moreover , this dimerization state was maintained in the nucleus containing Cort-activated GR molecules . To evaluate mCherryGR and GFP-p65 interactions , cross correlation analysis of the intensity fluctuations [35] was performed on the same data set . When untagged GFP and mCherry particles were analyzed , a symmetric cross correlation ( brightness cross correlation [Bcc] ) centered on zero was observed ( Figure 3C ) , indicating an absence of interaction between the GFP-mCherry pair . On the contrary , mCherryGR and GFP-p65 showed an asymmetric , positive Bcc value ( Figure 3C ) , indicating an interaction between GR and p65 molecules under our experimental conditions . Overall , although we cannot directly measure the stoichiometry of the GR-p65 complex , the most parsimonious model that fits our data is the one where Cort-stimulated GR is interacting with p65 as a dimer . As demonstrated above , GRdim is able to form dimers in vivo ( Figure 2 ) . If the DBD dimerization surface is indeed compromised in the GRdim mutant , then another region of the protein must participate in GR-GR interactions . An interesting candidate is the LBD region , where a second dimerization surface has been described [15] but whose functional relevance has been questioned on the basis of studies performed with DBD mutants like GRdim [10] , [36] . According to the GR LBD/Dex crystal structure , the dimerization interface includes a central hydrophobic region made up of reciprocal interactions between residues in the βA strand and a network of hydrogen bonds involving residues of the H1–H3 loop [37] . In a previous report , we characterized a rigid steroid ligand , 21-hydroxy-6 , 19-epoxyprogesterone ( 21OH-6 , 19OP ) , which behaves as a GR agonist in transrepression assays but as an antagonist in transactivation ones [24] . According to molecular dynamics ( MD ) simulations , 21OH-6 , 19OP induces a dramatic change in the average position of the H1–H3 loop within GR's LBD [38] . N&B studies showed that 21OH-6 , 19OP is still able to induce GR dimerization ( Figure 4A and [24] ) , which suggests that GR-21OH-6 , 19OP complexes dimerize through the DBD dimerization surface since the H1–H3 loop is compromised . Consistent with this hypothesis , GRA465T dimerization was abrogated in cells treated with 21OH-6 , 19OP ( Figure 4B ) , even though this compound induced GR nuclear translocation ( Figure 4B , right panel ) . Together , these results suggest that GR form dimers in vivo through the combined action of the LBD and the DBD regions . This model explains how the MD predictions performed on the GR-21OH-6 , 19OP complex are only detected in vivo when the DBD is compromised ( i . e . , in cells expressing GRA465T ) . To further evaluate the functional contribution of the DBD and LBD regions on GR dimerization , we constructed the mutant eGFP-GRI634A , on the basis of the orthologous human mutation I628A ( residue localized at the βA strand ) previously reported to decrease by 10-fold the dimerization of LBD-LBD fragments in vitro [15] . Figure 4C shows that eGFP-GRI634A has a diminished ability to form dimers in the presence of 0 . 1 µM Dex , although its subcelluar localization remains nuclear ( Figure 4C , right panel ) . Activation of the receptor with 1 µM Dex slightly increases dimerization , supporting a previous report suggesting that the human I628A may have reduced affinity for Dex [15] . Interestingly , when we combined the mutations in the DBD and the LBD dimerization surface ( eGFP-GRA465T/I634A ) dimer formation was completely abolished with 0 . 1 µM Dex and severely compromised in the presence of 1 µM Dex ( Figure 4D ) , suggesting a combinatorial contribution of both domains on GR dimerization . Similar behavior of these mutants was observed in both 3617 cells and in the MEF cell line with low-expression levels of GR ( Figure S2 ) . Accordingly , we named the GRA465T/I634A mutant GRmon as it is defective in dimerization in vivo . Consistent with the N&B data , the Förster resonance energy transfer ( FRET ) assay also indicated that the GRmon is impaired in dimerization ( Figures 4E , 4F , and S3 ) . We next characterized the transcriptional activities of all GR mutants . In agreement with previous data [16] luciferase reporter assays showed that GRA465T has little transactivation activity ( Figure 5A ) but similar transrepression efficiency ( Figure 5B ) compared to the wild-type receptor . As originally reported [15] , GRI634A also promotes poor transactivation activity ( Figure 5A ) ; however , transrepression of an NF-κB reporter was not affected ( Figure 5B ) . The GRmon behaves similarly to the single DBD and LBD point mutants ( Figure 5 ) . Consistently , transcriptional activation of endogenous genes in the eGFPGR-MEFs cell lines showed a similar trend ( Figure 5C ) . Taken together , our results show no correlation between the dimeric/monomeric state of the receptor and its ability to transactivate or transrepress gene expression , at least in the context of reporter gene assays . We next analyzed the ability of the LBD mutants to bind to the MMTV array in 3617 cells . Similar to the GRwt and GRA465T ( Figure 2D ) , array formation was successfully observed upon Dex addition with GRI634A ( Figure 6A , white arrows ) . On the contrary , although a few cells were positively visualized ( unpublished data ) , we failed to observe a considerable number of cells with arrays in the presence of GRmon ( Figure 6A ) . To confirm these results with an average-population , quantitative approach , we performed chromatin immunoprecipitation ( ChIP ) assays using a GFP antibody on the MMTV array . In agreement with the imaging data , all single GR mutants can occupy the MMTV region but GRmon is poorly recruited ( Figure 6B ) . Interestingly , both GRA465T and GRI634A are able to bind DNA with less efficiency than their wild-type counterpart ( Figure 6B ) . The eGFP-GR mutants in ChIP experiments were expressed at a similar level ( Figure S4 ) . The recruitment of transcription factors to chromatin depends on a variety of complex events . An emerging paradigm suggests that the local chromatin structure of response elements contributes strongly to the tissue-specific action of many transcription factors [39] . In particular , in vivo GR recruitment to DNA is strongly dependent on the chromatin landscape , with most of the GR binding events occurring at pre-programmed chromatin ( i . e . , DNaseI hypersensitive sites prior to ligand treatment ) and only a small fraction of binding at de novo sites ( i . e . , DNaseI sites actively induced by the receptor ) [40] . To further characterize the GR mutants , we performed ChIP assays on a few pre-programmed or de novo sites . Irrespective of the sites analyzed , both the GRA465T and the GRI634A were able to bind chromatin ( Figure 6C and 6D ) , although their relative occupancy compared to the wild-type was site-specific . Interestingly , GRmon was recruited to most of the pre-programmed sites evaluated ( Figure 6D ) while no significant binding was observed to de novo sites ( Figure 6C ) . Finally , we evaluated GR recruitment to recently reported negative GREs ( nGREs ) [41] , which were suggested to be preferential binding sites of the monomeric GR [9] . To identify nGREs in 3134 cells , we overlapped all 1 , 147 putative nGREs conserved between human and mouse [41] with GR ChIP-seq data from 3134 cells [40] . Surprisingly , only five were found at GR binding sites in 3134 cells ( Figure S5 ) . From these , three were located near Dex-repressed genes as shown by microarray analysis [42] . ChIP results show no clear link between the mutants and the receptor's ability to bind nGREs ( Figure 6E ) . In summary , our data suggest that the dimeric status of the receptor neither defines its transactivation activity nor predicts its ability to bind chromatin in vivo . On the other hand , the monomeric form of GR seems to be less efficient in its ability to bind chromatin than the dimeric form of the receptor .
Studies mainly using the GRdim mutant suggested the dissociated model of GR action and led to the transrepression hypothesis [2] . This hypothesis states that suppression of inflammation by GR is mainly mediated by the transrepression mechanism , and is independent of GR transcriptional regulation through its direct binding to DNA . Accordingly , side effects of glucocorticoids were suggested to be dependent on GR dimerization , GR-GRE interaction , and the downstream consequence on gene regulation . This model has been the guiding principle in the search of new compounds with dissociated glucocorticoid properties [11] . Today this strategy is deeply criticized , not only because it is known that some glucocorticoid anti-inflammatory effects depend on gene activation [2] , [43] , but also because evidence against GRdim's alleged monomeric status and inability to bind DNA is accumulating [19] , [20] , [22] , [31] . Here , we demonstrate that the so-called GRdim is able to dimerize in vivo while the new mutant GRmon ( A465T/I634A ) is severely impaired in dimer formation . We have studied the oligomerization state of GR by the novel N&B technique , under both physiological and over-expressed GR levels . Independent confirmation that the GRmon is impaired in dimerization has been obtained by fluorescence lifetime imaging microscopy ( FLIM ) -FRET analysis . If the GRdim is still able to bind DNA and form dimers as demonstrated here and elsewhere [19] , [20] , [22] , [31] , why is this mutant unable to transactivate genes ? Recent studies have shown that the GRdim's residence time on DNA is ten times less than the one observed for wild-type GR [29] , in strict agreement with its diminished transcriptional activity according to the “hit and run” model of transcriptional activation [44] , [45] . Also , it has been shown in vitro that the dim mutation alters the allosteric effect that DNA exerts on GR , therefore varying the receptor's conformational states and perhaps changing the ability to interact with co-regulators [20] . Even though it has not been directly tested , GRdim's altered ability to interact with specific cofactors could explain why this mutant is able to induce the expression of genes whose promoters contain certain GREs and not others [19]–[21] . In other words , the dim mutation does not actually appear to abolish GR transactivation altogether but instead their effect depends on both gene and cellular context , producing an overall change in the whole transcriptional outcome . As an example , a microarray analysis performed in U-2 OS cells showed a very different pattern of gene regulation comparing wild-type and GRdim expressing cells [22] . Moreover , expression analysis performed in livers from wild-type and dim mice revealed that GRdim could induce gene expression when compared with wild-type GR [46] . Overall , there is compelling evidence that suggests that the transactivation versus transrepression model that arouse from the GRdim mouse phenotype was oversimplified and needs re-examination [2] , [43] . More genome-wide studies on the dim model will provide much needed insights in the mechanisms underlying the GRdim mice phenotype . The establishment in the community that transactivation is mediated by GR dimers and transrepression occurs exclusively through GR monomers has been built almost entirely under the GRdim paradigm [16] , [18] , [47] . However , here we find no correlation between the dimeric/monomeric state of the receptor and its ability to transactivate or transrepress reporter genes . For example , even though GRwt and GRdim are mainly dimeric the latter is severely impaired in transactivation compared to the wild-type GR . On the other hand , the GRmon is mainly monomeric but its transrepression efficiency is indistinguishable from the fully dimeric wild-type receptor . Hence , changing the relative population between dimers and monomers does not necessarily change the transcriptional outcome . In conclusion , GR dimerization appears necessary but not sufficient for transactivation and it is not required for transrepression . Nonetheless , given the fact that GR transcriptional activity is highly gene- and cell type- specific more studies are needed to properly evaluate the scope of this conclusion . Interestingly , our data suggest that Cort-GR molecules remain dimeric in the presence of GR/NF-κB interactions . Thus , the idea that transactivation could be dissociated from transrepression through manipulation of the oligomerization state of the receptor should be critically revised , if not entirely discarded . Overall , our results indicate that GR dimerization involves a more complex mechanism than previously anticipated . Moreover , we also challenge the view that transrepression is exclusively performed by the monomeric GR . This implies that the simplified monomer/dimer model equilibrium does not explain GR transactivation versus transrepression activity . It seems that the prevailing view was established without rigorous verification and new approaches for mitigating the side effects of chronic glucocorticoid treatment should be explored .
Dex and Cort were purchased from Sigma-Aldrich . CpdA [25] was purchased from Enzo Life Sciences . 21OH-6 , 19OP was prepared as previously described [48] . pEGFP-GR expresses the eGFP protein fused to the N-terminal end of the mouse GR [24] . pEGFP-GRA465T was generated by site-directed mutagenesis by TOP Gene Technologies . pEGFP-GRI634A and pEGFP-GRA465T/I634A were generated by site-directed mutagenesis by Stony Brook cloning facility ( Stony Brook University , New York , USA ) . mCherry-GR was previously described [49] . For FLIM-FRET experiments , the coding region of the super ( s ) REACh fluorophore [50] was subcloned into the N-terminal of the mouse GR sequence ( psREACh-GR ) . Briefly , the AgeI-BglII eGFP containing sequence of pEGFP-GR and pEGFP-GRA465T/I634A was replaced with sREACh cDNA PCR amplified from mGFP-10-sREACh-N3 ( Addgene , plasmid 21947 ) using the Herculase II fusion DNA polymerase system ( Agilent Technologies ) . The reverse primer contained an additional five bases , introducing 5′-TACTC-3′ into the plasmid prior to the BglII restriction site and so preserving the same linker as the eGFP variants . pMMTV-luciferase; pkB-luciferase , pRelA and pCMV-LacZ were previously described [24] . GFP-p65 was a kind gift from Alessandra Agresti [51] . The SV40T-expressing retroviral pBabe-largeT cDNA and the retroviral pWZL-neo plasmids were a gift from Kai Ge [52] . The coding region of the eGFP-GR mutants was cloned into the pWZL-neo vectors for retroviral transduction . Briefly , each eGFP-GR coding sequence was independently isolated by PCR ( using the high fidelity Herculase II polymerase ) with primers carrying BamHI and MfeI restriction sites: forward ( For ) atatggatccGTGAACCGTCAGATCCGCTAG and reverse ( Rev ) atcgCAATTGGGCAGCCTTTCTTAGTAAGGCAG . The purified fragment was subcloned into BamHI/MfeI sites of the pWZL-neo vector . BHK21 and Cos-7 cells were cultured in DMEM ( Invitrogen ) supplemented with 10% FBS ( Internegocios S . A . ) . 3134 and 3617 cells were cultured in DMEM and supplemented with 10% FBS ( Hyclone ) . The 3134 cell line is a mouse mammary adenocarcinoma cell line . It contains a large tandem array ( ∼200 copies ) of a mouse mammary tumor virus , Harvey viral ras ( MMTV-v-Ha-ras ) reporter . The 3617 cell line is a derivative of 3134 cell line expressing a GFP-tagged version of GR ( GFP-GR ) from a chromosomal locus under control of the tetracycline repressible promoter . Both cell lines were described previously [53] . In all cases , prior to glucocorticoid treatment cells were incubated at least 18 h in DMEM medium containing 10% charcoal-stripped FBS ( Hyclone ) . Heterozygous GR-deficient ( GR het ) mice were generated by crossing mice with one allele of GR exon 3 flanked by loxp sites [54] with mice expressing Cre driven by the b-actin promoter . Day 13 . 5 embryo bodies from a timed GR het×GR het mating were minced with scissors and forceps , digested with trypsin , and cultured in DMEM supplemented with FCS and glutamine at 37°C in 5% CO2 . GR-deficient MEFs were identified by PCR as being positive for the deleted allele and negative for the germline allele . Primary fibroblasts were immortalized via retroviral transduction with SV40 large T antigen . Briefly , 5 million Phoenix A cells were plated in a 10-cm dish 24 hours prior to transfection with 10 µg pBabe-SV40 ( Puro ) plasmid using JetPRIME transfection reagent ( Polyplus transfection ) according to the manufacturer's recommended protocol . Virus containing supernatant was collected 48 hours post-transfection and filtered through a 0 . 45 µM filter . Filtered virus-containing Phoenix cell supernatant was diluted with an equal volume of fresh media and polybrene was added to a final concentration of 5 µg/ml . 2 ml of this virus solution was used to infect 200 , 000 MEFs . 48 hours post-transduction the cells were challenged with 2 µg/ml puromycin ( SIGMA-Aldrich ) . Puromycin selection was complete in 3–4 days , however these large T antigen immortalized MEFs were maintained in media containing 2 µg/ml puro . The immortalized MEF cell lines ( wt and GR−/− ) were transduced with pWZL-GFPGR ( Neo ) as described above . These cells were selected with 500 µg/ml G418 ( Cellgro ) . After 15 days of Neomycin selection , cells were sorted by FACS according to their GFP expression into three categories ( low , medium , high ) . eGFP-GR levels were monitored by Western blot ( Figure S1 ) and medium expression cells were chosen for further studies . BHK21 and Cos-7 cells were transiently transfected with Lipofectin 2000 ( Invitrogen ) according to manufacturer's instructions . 3134 and 3617 cells were transfected with jetPRIME reagent ( VWR ) according to manufacturer's instructions . 3×105 BHK cells were transfected with 1 . 5 µg of pEGFP-GR or the mutant variants and incubated with vehicle , 100 nM Dex , 1 µM Dex , 100 nM Cort , 10 µM 21OH-6 , 19OP , or 10 µM CpdA for at least 1 h . Washout procedures consisted in washing the cells three times with pre-warmed ( 37°C ) PBS and then adding hormone-free media for 20–40 minutes before analysis . Measurements were done in a FV1000 confocal laser scanning microscope ( Olympus ) , with an Olympus UPlanSApo 60× oil immersion objective ( NA = 1 . 35 ) . The excitation source was a multi-line Ar laser tuned at 488 nm ( average power at the sample , 700 nW ) . Fluorescence was detected with a photomultiplier set in the pseudo photon-counting detection mode . 3617 cells were grown in the presence of 5 µg/ml tetracycline ( Sigma-Aldrich ) to inhibit the stable GFP-GR gene expression [27] , [53] , and transiently transfected with 1 . 5 µg of pEGFP-GR or the mutant variants , or a combination of mCherryGR and GFP-p65 as indicated . Cells were incubated for at least 30 min with 100 nM Dex , 100 nM Cort , or 300 nM Cort in the presence or absence of 10 ng/ml TNFα ( Sigma-Aldrich ) . Measurements were done in a LSM 780 laser scanning microscope ( Carl Zeiss , Inc . ) at the CCR Confocal Microscopy Core Facility ( NIH , Bethesda , Maryland , USA ) . We used a 63× oil immersion objective ( NA = 1 . 4 ) . The excitation source was a multi-line Ar laser tuned at 488 nm and or a 594 nm laser . Fluorescence was detected with a GaAsP detector in photon-counting mode . N&B measurements were done as previously described [23] with some modifications [24] . Briefly , for each studied cell a stack of 150–200 images ( 256×256 pixels ) were taken in the conditions mentioned above , setting the pixel size to 80–82 nm and the pixel dwell time to 6 . 3 or 10 µs . Each stack was further analyzed using the N&B routine of the “GLOBALS for Images” program developed at the Laboratory for Fluorescence Dynamics ( UCI , Irvine , California , USA ) . In this routine , the average fluorescence intensity ( <I> ) and its variance ( σ2 ) at each pixel of an image are determined from the intensity values obtained at the given pixel along the images stack . The apparent brightness ( B ) is then calculated as the ratio of σ2 to <I> while the apparent number of moving particles ( N ) corresponds to the ratio of <I> to B . In a previous work it has been demonstrated that B is equal to the real brightness ε of the particles plus one [23] . Therefore , ε at every pixel of images can be easily extracted from B measurements . Importantly , this analysis only provides information regarding the moving or fluctuating fluorescent molecules since fixed molecules will give B values equal to 1 . For the transactivation assay , 3×105 Cos-7 cells were co-transfected with 1 . 5 µg pMMTV-luciferase vector and 0 . 5 µg of pEGFP-GR vectors . For the NF-κB transrepression assay , 1 . 5 µg pkB-luciferase and 1 . 5 µg pRelA were used . In all cases , 0 . 5 µg pCMV-LacZ was added as transfection control . After transfection , cells were incubated in DMEM containing 5% charcoal-stripped FBS and incubated with 100 nM Dex for at least 18 h . Luciferase activity and β-galactosidase activity was measured as previously described [24] . 3617 cells were seeded to 22×22 mm glass coverslips in six-well tissue culture plates . Media contained 10% charcoal-stripped serum and tetracycline ( 5 µg/ml ) to prevent GFP-GR expression . Next day cells were transiently transfected with 2 µg total plasmid using JetPRIME ( VWR ) and the manufacturer's protocol . EGFP ( donor ) and sREACh ( acceptor ) plasmids were transfected at 1∶2 ratio to maximize the chances of seeing an interaction by FRET . 24 h after transfection cells were treated for 30 min with 100 nM Cort and fixed in paraformaldehyde added to media ( 4% final concentration ) for 15 min . Coverslips were washed 3× in PBS and mounted to microscope slides with Mowiol 4–88 containing 1 mg/ml p-phenylenediamine as anti-fade ( both Sigma-Aldrich ) . Images were acquired on a Leica DMI 6000 SP5 inverted confocal microscope with a 63× oil immersion objective of NA 1 . 4 ( Leica Microsystems ) . EGFP excitation at 850 nm was achieved with a femtosecond mode-locked ( 80 MHz repetition rate ) Mai-Tai HP pulsed , multi photon laser ( Spectra Physics ) . Fluorescence was collected using a HPM100 Hybrid Detector R3809U-50 ( Becker & Hickl; Hamamatsu Photonics ) through a band-pass GFP filter at ET 525/50 ( Chroma Technology Corp ) . Fluorescence decays were resolved by time-correlated single-photon counting ( TCSPC ) using a SPC830 acquisition board ( Becker & Hickl ) . Images were acquired in 256×256 pixel format collecting at least 1 , 000 photons per pixel over 2–5 min . Fluorescence transients were acquired with SPCImage software ( Becker & Hickl ) , analyzed according to single-life time decay , then exported to Image J ( NIH ) . An in-house Image J protocol permitted selection of the relevant pixels ( nucleus ) and derivation of histograms for the weighted mean average of the fluorescent lifetimes . These were plotted as frequency distributions normalized and integrated for area under the curve using Igor Pro ( WaveMetrics Inc ) . The weighted mean lifetime ( T ) was extracted from histograms of individual cells in Image J and converted to FRET efficiency relative to the GFP-GR control according to: FRET Efficiency ( % ) = 1− ( Tdonor/Tdonor+acceptor ) −100 to allow statistical analysis . 3134 cells were seeded in 150 mm tissue culture plates and the next day transiently transfected with 10 µg of pEGFP-GR or the mutant variants . Cells were collected the next day after 1 h of 100 nM dex treatment . ChIP was performed according to the standard protocol ( Upstate Biotechnology ) with a crosslinking step ( 1% formaldehyde at RT ) , followed by a quenching step with 125 mM glycine . Chromatin was sonicated by using the Bioruptor sonicator ( Diagenode ) with 15 s “on” and 15 s “off” for 30 cycles . Sonication efficiency was monitored by 2% agarose gel electrophoresis . Sonicated chromatin ( 400 µg ) was immunoprecipitated with an antibody against GFP ( Abcam ab290 ) . DNA isolated from immunoprecipitates , as well as input DNA , was used as a template for real-time PCR ( qPCR ) . Primers used for qPCR are ( 5′→3′ ) : MMTV , For TGGTTACAAACTGTTCTTAAAACGAGGATG and Rev CTCAGATCAGAACCTTTGATACCAAACC; LCN2 , For TCACCCTGTGCCAGGACCAA and Rev TGGGGAAGGGTGAGCAAGCT; GluL , For CACTTGGGCAAACATGGACGGT and Rev CACAAGAGGAAATGCCCCCCT; Mt2 , For CATAGCCAGGGCAGCCACAGAA and Rev GGCAATGCCTTCTTGACTCATTCC; SGK , For CACTTGGGCAAACATGGACGGT and Rev CACAAGAGGAAATGCCCCCCT; Mt1 , For TAGGGACATGATGTTCCACACGTC and Rev TTTTCGGGCGGAGTGCAGAG; Tgm2 , For CCACACATTGGTTTTGCTATGCTTG and Rev AATCATTTTCTCATTCCACACAGCC; Ampd3 , For GCCAGGACGTGGTGTTCAGGAT and Rev GGGCTGGAAATTCTCCTGCG; Sarc , For CCTCAGTCAGTGCTCAGTGG and Rev GGGACCAGATGGGATATCAG; Aebp1 , For CTCTTATGCAATCGTTGTCAGTAAATCT and Rev ATGATGAATGGTGCCTTACAGTCTC; Mocs1 , For ATTTGGCAGAGACTAGCCTGGAAATGAT and Rev CATCTTATGACCTACTTCCACCCCA; S100a4 , For ATGGGGTAAGGAGCGGAAGG and Rev CTGGACCCAGCCATGCCCTC . Standard curves were created by 4-fold serial dilution of an input template . The data presented are from four independent experiments . The MEFs cell lines were plated for 48 h in DMEM medium containing 10% charcoal-stripped FBS and then treated for 1 h with 100 nM Dex . RNA extraction was performed with the Nucleospin RNA-kit ( Clontech ) according to manufacturer's instructions . cDNA was made with the iScript cDNA Synthesis Kit ( Bio-Rad Laboratories , Inc . ) from 1 µg RNA . Upon dilution , cDNA was subjected to qPCR using the iQ SYBR green supermix ( Bio-Rad ) with the indicated primers . Primer sequences were designed to amplify only nascent RNA , using PCR amplicons that cross an exon/intron or UTR/intron boundary . Primer sequences are as follows: Mt1 , For CCTCACTTACTCCGTAGCTCCAGC and Rev TCCCGCCAAGCCTCTACAACTC; Mt2 , For GAACTCTTCAAACCGATCTCTCGTC and Rev TCCCAGAAATCCCGTCAGCA; SGK , For GGGAATGGTAGCGATTCTCATCG and Rev CGACGCCACACGCTAATCTG; Actin , For AGTGTGACGTTGACATCCGTA and Rev GCCAGAGCAGTAATCTCCTTCT . Chromatin samples from ChIPs experiments ( i . e . , inputs ) were separated by SDS–PAGE and transferred to PVDF membranes . Blots were probed with primary antibodies anti-GR ( sc-1004; 1∶1 , 000 ) , anti-actin ( sc-1615; 1∶1 , 000 ) ( Santa Cruz Biotechnology ) , or anti-GAPDH ( Abcam , ab-8245 , 1∶1 , 000 ) in Tris-buffered saline ( TBS ) containing 5% nonfat dry milk , followed by incubation with horseradish peroxidase ( HRP ) -conjugated anti-goat , anti-mouse , or anti-rabbit antibody ( Santa Cruz Biotechnology ) . All blots were visualized with the ECL kit ( Supersignal ) . Results were expressed as means ± SEM . Statistical analyses were performed with STATISTICA 7 . 0 ( StatSoft , Inc . ) and consisted of one-way ANOVA followed by Tukey's multiple comparisons tests . Differences were regarded as significant at p<0 . 05 ( bars with different superscript letters are significantly different from each other ) . Before statistical analysis , data were tested for homoscedasticity using Bartlett's test . In some cases , transformation of the variable ( x′ = √x ) were necessary . | The powerful anti-inflammatory and immunosuppressive action of glucocorticoids have made them one of the most prescribed drugs worldwide . Unfortunately , acute or chronic treatment may have severe side-effects . Glucocorticoids bind to the glucocorticoid receptor ( GR ) , a ligand-dependent transcription factor . GR regulates gene expression directly by binding to DNA or indirectly by modulating the activity of other transcription factors . It is currently accepted that the direct pathway is mostly responsible for glucocorticoids side-effects and that the oligomerization state of the GR ( whether it is a dimer or a monomer ) determines which pathway ( direct or indirect ) will prevail . Hence , scientists have tried to develop “dissociated ligands” able to specifically activate the GR indirect pathway . In the present work , we employed a novel microscopy method named the number and brightness assay , which measures GR oligomerization state inside the living cell . Our results suggest that—contrary to the established view—there is no clear correlation between the oligomerization state of GR and the mechanistic pathway the receptor will follow upon ligand binding . This discovery presents supporting evidence towards the increasing view of the inherent complexity of glucocorticoid action and might impact future approaches towards the design of safer synthetic glucocorticoids . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"nucleic",
"acids",
"biology",
"and",
"life",
"sciences",
"dna",
"nuclear",
"receptor",
"signaling",
"molecular",
"cell",
"biology",
"cell",
"signaling"
]
| 2014 | Live Cell Imaging Unveils Multiple Domain Requirements for In Vivo Dimerization of the Glucocorticoid Receptor |
The oral polio vaccine ( OPV ) contains live-attenuated polioviruses that induce immunity by causing low virulence infections in vaccine recipients and their close contacts . Widespread immunization with OPV has reduced the annual global burden of paralytic poliomyelitis by a factor of 10 , 000 or more and has driven wild poliovirus ( WPV ) to the brink of eradication . However , in instances that have so far been rare , OPV can paralyze vaccine recipients and generate vaccine-derived polio outbreaks . To complete polio eradication , OPV use should eventually cease , but doing so will leave a growing population fully susceptible to infection . If poliovirus is reintroduced after OPV cessation , under what conditions will OPV vaccination be required to interrupt transmission ? Can conditions exist in which OPV and WPV reintroduction present similar risks of transmission ? To answer these questions , we built a multi-scale mathematical model of infection and transmission calibrated to data from clinical trials and field epidemiology studies . At the within-host level , the model describes the effects of vaccination and waning immunity on shedding and oral susceptibility to infection . At the between-host level , the model emulates the interaction of shedding and oral susceptibility with sanitation and person-to-person contact patterns to determine the transmission rate in communities . Our results show that inactivated polio vaccine ( IPV ) is sufficient to prevent outbreaks in low transmission rate settings and that OPV can be reintroduced and withdrawn as needed in moderate transmission rate settings . However , in high transmission rate settings , the conditions that support vaccine-derived outbreaks have only been rare because population immunity has been high . Absent population immunity , the Sabin strains from OPV will be nearly as capable of causing outbreaks as WPV . If post-cessation outbreak responses are followed by new vaccine-derived outbreaks , strategies to restore population immunity will be required to ensure the stability of polio eradication .
Wild polioviruses ( WPVs ) have been eliminated from all but three countries [1 , 2] by mass vaccination with the oral polio vaccine ( OPV ) . The annual burden of paralytic polio infections has been reduced 10 , 000-fold since the start of vaccination efforts [1] . OPV has been the preferred vaccine for polio eradication because it costs less , can be reliably delivered by volunteers without medical training , and is more effective against poliovirus infection , relative to the inactivated polio vaccine ( IPV ) [3 , 4] . Unique among current human vaccines , the live-attenuated Sabin poliovirus strains in OPV are transmissible . This transmissibility provides additional passive immunization that enhances the effectiveness of OPV for generating herd immunity . However , the attenuation of Sabin OPV is unstable and so it can , in rare instances , cause paralytic poliomyelitis [5] and lead to outbreaks of circulating vaccine-derived poliovirus ( cVDPV ) with virulence and transmissibility comparable to that of WPV strains [6] . Thus , to complete the task of poliovirus eradication , vaccination with Sabin OPV must eventually cease [7] . The dual role of Sabin OPV as both a vaccine and a source of poliovirus is responsible for key uncertainties surrounding the ability of the Global Polio Eradication Initiative to achieve and sustain poliovirus eradication . Since the widespread introduction of polio vaccination , polio outbreaks have taken place in regions of low immunity against infection surrounded by regions of high immunity [8] , OPV campaigns implemented in outbreak response have been effective for interrupting transmission [3] , and cVDPV outbreaks have been rare consequences of the hundreds of millions of OPV doses administered every year [9] . However , after vaccination with OPV is stopped , population immunity against infection will progressively decline . If polioviruses are reintroduced , whether because of accidental or deliberate release [10–12] or because of sustained yet undetected transmission [2 , 13–15] , then large outbreaks may again occur . Outbreak control would require vaccination campaigns in affected countries and perhaps much more broadly , as has been done following recent type 2 cVDPV detections in Nigeria , Pakistan , the Democratic Republic of Congo , and Syria [16–19] . In this paper , we explore the implications of the accumulated evidence about polio infection and transmission for the long-term stability of polio eradication . Will it be possible to interrupt all polio outbreaks without restarting widespread OPV vaccination , now and at all times in the future ? Fundamental to this question is the ability of the Sabin polioviruses to circulate in low immunity populations . After OPV cessation , under what conditions will Sabin OPV remain the most effective tool for eliminating outbreaks without significant risks of causing more ? To address these questions , and building from primary literature and previous reviews and models [4 , 20–25] , we developed a comprehensive synthesis of the evidence for how within-host immunity , viral infectivity , and transmission dynamics fit together to explain the epidemiology of poliovirus transmission . We built a within-host model that summarizes the effects of immunization on poliovirus shedding and susceptibility . We then incorporated the within-host dynamics into a poliovirus transmission model using a household–community network framework , and we calibrated the model to field transmission studies . With the model , we explored how the average transmission rate in exposed communities varies with immunity , sanitation , number of social contacts , and poliovirus type . We identified conditions required for the Sabin strains to remain indefinitely as highly effective vaccines with low risks of causing outbreaks , and conditions in which they can be expected to transmit nearly as efficiently as WPV . Our results are discussed in the context of the established stability of OPV cessation in the developed world and the ongoing global Sabin 2 cessation . Previous models have also explored the effects of OPV cessation on Sabin and WPV transmission [20 , 26–33] . Our work shares a similar emphasis on individual immunity and infection dynamics [20 , 24 , 34] , but it differs in model structure , use of transmission data sources , and approach to epidemiological inference . The earlier work used compartmental models that assume individuals in large populations interact randomly [20 , 26 , 27 , 33] . Models of specific settings—places and times—were based on national polio surveillance data [27 , 33] , and poliovirus evolution was modeled by extrapolation between Sabin and WPV end points based on assumed intermediate transitions [26 , 27 , 33] . In contrast , our model is based on person-to-person transmission among family members and extrafamilial contacts . We calibrate specific settings to contact transmission data from field studies designed for that purpose [35–37] . We explore the effects of evolution by focusing on differences between the Sabin and WPV end points , and not the largely unstudied population genetics in between . In short , our model takes a bottom-up approach to modeling poliovirus transmission that complements existing work . Instead of drawing inferences about the unobserved conditions that affect transmission from observed outbreaks [27 , 33] , we draw inferences about unobserved properties of possible outbreaks from observed conditions that directly affect transmission .
To integrate knowledge of within-host immunity , shedding , and acquisition with between-host transmission , we built a multi-scale mathematical model . We first performed a quantitative literature review of clinical trial data to determine the impact of polio vaccination schedules containing OPV and/or IPV on poliovirus shedding after challenge with OPV . This resulted in an indirect measure of immunity—the OPV-equivalent antibody titer—which was used to model the associations between polio vaccination and shedding duration , concentration of virus in stool , and oral susceptibility to infection . Second , we reviewed in detail three historical transmission studies to parameterize a model of poliovirus transmission within households and between close extrafamilial contacts . We then extended the person-to-person model by defining the local reproduction number—a threshold parameter that summarizes the potential for epidemic transmission within a community that has homogeneous demographics , immune histories , and sanitation practices . The model was implemented in Matlab 2015b ( The Mathworks , Natick , MA ) and is available in S1 Code and at famulare . github . io/cessationStability . For all model parameters , see S1 Text Table A . We made simplifying assumptions while developing the model . First , we did not include the oral–oral transmission route . The studies known to us show that oral shedding occurs for shorter durations than fecal shedding and most often in individuals with low immunity [38–41] , and this route is likely more important in high sanitation settings [34 , 39 , 42] . Second , we ignored fascinating questions about the effects of genetic evolution on Sabin strain transmission , and so our Sabin transmission parameters are most applicable in the first few weeks after OPV vaccination [43–45] . Third , our model focuses on transmission from children to family members and extrafamilial contacts , and it ignores other person-to-person interactions and possible environmental transmission routes , all of which influence the absolute probability and severity of outbreaks [46–50] . In the Results , we explore how the limitations of a model with these assumptions are informative about the roles of transmission route , viral evolution , and contact structure in various settings . Fourth , because paralysis has no direct influence on transmission , we did not model the impact of vaccination on paralysis ( see Vidor and Plotkin [51] for a review ) . All model features that describe the fraction of subjects shedding after live poliovirus exposure were fit by maximum likelihood assuming binomial sampling . Models for positive-definite quantities ( concentration of poliovirus in stool , antibody titer ) were estimated by ordinary least squares on log ( quantity ) , and 95% confidence intervals ( CIs ) assume log-normality . Ninety-five percent CIs were estimated by parametric bootstrap with 1 , 000 replicates . To estimate bootstrap CIs of parameters that are conditionally dependent on previously estimated parameters , we propagated uncertainty by resampling known parameters from the 95% CIs prior to resampling the data and re-estimating the parameters under investigation . Model equations and more detailed discussions of design decisions and calibration results can be found in S1 Text . Differences in comparable quantities are considered statistically significant at α = 0 . 05 . Our within-host model describes shedding from and susceptibility to poliovirus infection . In the model , the ability of an infected individual to transmit polio depends on the duration of shedding and the concentration of poliovirus in their stool . Oral susceptibility to infection depends on the dose–response relationship for the probability that poliovirus ingested orally results in an infection , as detected by subsequent fecal shedding . Shedding duration , concentration , and oral susceptibility all depend on pre-exposure immunity and the poliovirus source , vaccine or wild . Immunity in our model is represented by the OPV-equivalent antibody titer ( denoted NAb ) —an indirect measure of immunity that is inferred from measurements of shedding duration and/or dose response ( first introduced by Behrend and colleagues [25] and called "mucosal immunity" therein ) . Previous reviews have demonstrated that when immunity is due to prior OPV immunization or natural WPV infection , homotypic ( of the same serotype ) serum neutralizing antibody titers ( measured as the geometric mean reciprocal dilution of serum that is able to neutralize 100 CID50 [the culture infectious dose that induces a cytopathic effect in 50% of infected cell or tissue cultures] of poliovirus ) are predictive of fecal shedding and susceptibility [25 , 52] . However , serum antibody titers induced by IPV alone , and heterotypic titers against type 2 from bivalent type 1 and 3 OPV ( bOPV ) are not predictive of shedding and susceptibility [25 , 53 , 54] . The OPV-equivalent antibody titer describes the impacts of vaccination histories containing IPV or bOPV on shedding and susceptibility in terms of equivalent serum antibody titers from homotypic OPV vaccination in children . This model is agnostic about the biophysical mechanisms of immunity that prevent fecal shedding and is not intended to represent immunoglobulin A concentration or other direct correlates of mucosal immunity [55] . Following the results of Behrend and colleagues [25] , we assumed that the typical immunologically naive individual with no history of poliovirus exposure ( "unvaccinated" ) and no measurable humoral immunity ( "seronegative" ) has an OPV-equivalent antibody titer of NAb = 1 by definition , that the typical OPV-equivalent titer at maximum achievable individual immunity is NAb = 2 , 048 ( = 211 ) , and that homotypic antibody titers for each poliovirus serotype are independent . The studies used to calibrate the within-host model span many countries , years , and types of immunization history . All included studies describe the fraction of subjects positive for poliovirus in stool after OPV challenge or WPV exposure as equal to the number of subjects shedding divided by the number tested at each time point . In many cases , the data were digitized from published figures that do not report variation in the number of samples for each time point , and so our sample sizes at each time point are often approximate . A summary of all included studies , with details about which studies contributed to which components of the model and reasons for study exclusion , appears in S1 Text Part B [36 , 41 , 53 , 54 , 56–72] , and the data are in S1 Code . Our model describes the effects of within-host dynamics on transmission among people who share a household and close social contacts outside the household . We assumed that transmission from infected person to recipient occurs by oral exposure to infected feces , for which the amount of poliovirus transmitted per exposure is determined by the shedding duration and concentration models , and recipient susceptibility is determined by the dose–response model .
Fig 7 summarizes our within-host model for the effects of immunity on shedding and susceptibility and how typical immunity levels relate to specific vaccination schedules . The shedding index ( Fig 7A ) is the expected total amount of virus shed per gram of stool after mOPV challenge . For a typical healthy child under 5 years of age—averaged over vaccination timing and waning—each of the first three doses of OPV increases the OPV-equivalent antibody titer by roughly a factor of 8 and decreases the expected amount of virus shed by a factor of 10 . To characterize settings with low OPV effectiveness [25 , 74–77] , we found that children who received at least six doses of tOPV in UP and Bihar [37] had similar OPV-equivalent antibody titers to healthy clinical trial subjects who received three tOPV doses . In our model , IPV boosting and OPV doses after the first three maintain maximum immunity . The heterotypic protection against type 2 from bOPV immunization is comparable to that of a single homotypic dose but does not accumulate with multiple doses . We inferred from the trial arms reviewed that the OPV-equivalent immunity of IPV-only is at most comparable to heterotypic immunity from bOPV , but we expect that the true impact is closer to none—the trial arms that showed the highest immunity ( Fig 1 ) likely included some incidental IPV boosting , with larger effects in older [41 , 61 , 65 , 69] versus younger [53 , 57 , 61] subjects in OPV-using countries , and negligible effects in older subjects in countries where OPV is not ubiquitous [67 , 93] . Susceptibility is also strongly impacted by immunity , with the expected fraction shedding after Sabin 2 challenge dropping below half at all relevant doses for NAb ≥ 64 ( Fig 7B ) . Our waning model ( S1 Text Eq F , Fig 4 ) predicts that without reinfection , typical peak OPV-equivalent antibody titers ( NAb = 2 , 048 ) decline to typical three-dose healthy child immunity ( NAb = 512 ) in 5 ( 4–7 ) months and to typical two-dose immunity ( NAb = 64 ) in an additional 4 ( 2–10 ) years . However , the model also predicts that it takes an additional 45 ( 15–160 ) years to fall to the equivalent of one-dose childhood immunity ( NAb = 8 ) and that residual immunity persists for life , as has been suspected previously [24 , 94] . This result is in disagreement with the conclusions of Abbink and colleagues [68] . They argued from the lack of correlation between serological boosting responses and shedding duration after OPV challenge that memory immunity in seronegative elderly does not protect against poliovirus shedding , but the study lacked a control group of never-exposed subjects to contrast deeply waned and truly naive immunity . As seen through metastudy , the OPV-equivalent immunity of the Abbink and colleagues seronegative elderly cohorts is similar to that of children who have received one dose of OPV . For heterotypic immunity against type 2 from bOPV , we predict that protection from shedding will be lost 13 ( 9–22 ) months after bOPV vaccination is stopped [95] . Fig 8 shows MLEs from our transmission model for the local reproduction number of WPV , Rloc ( Eq 3 ) , as functions of immunity and daily fecal–oral dose ( Fig 8A ) , and fecal–oral dose and the number of close social contacts outside the household ( Fig 8B ) . The value of Rloc , a measure of the average transmission rate in a community , depends linearly on the number of social contacts but varies across four orders of magnitude because of the strong effects of immunity and dose . Assuming one fecal–oral exposure per day ( see Methods: Transmission model: Calibration ) , the physiological range for the average fecal–oral dose maxes out at two milligrams of stool , corresponding to the upper bound of our estimate from UP and Bihar in 2003–2008 . When all children have typical three-dose childhood immunity or more ( NAb ≥ 512 ) , we estimated Rloc < 1 over the entire physiological range and thus that WPV persistence is impossible under universal tOPV immunization . In the absence of immunity , WPV epidemics are possible in all settings where sanitation practices permit the ingestion of roughly one microgram of stool per day or more . We identified three categories describing the transmission rate in different settings: low , where the fecal–oral route alone cannot sustain WPV transmission ( Rloc < 1 for all NAb ≥ 1 ) ; moderate , where WPV epidemics can occur in immunologically naive communities but not where at least one-dose OPV-equivalent immunity is common ( Rloc ≥ 1 only for NAb < 8 ) ; and high , where WPV can persist despite at least one-dose OPV-equivalent immunity in everyone ( Rloc ≥ 1 when NAb ≥ 8 but less than a protective threshold ) . Fig 9 shows the dependence of the local reproduction number on poliovirus strain and immunity for example low , moderate , and high transmission rate settings . In low transmission rate settings , epidemic transmission of any strain cannot occur without contributions from the unmodeled oral–oral transmission route . This result supports the long-held hypothesis that oral–oral transmission is critical in settings with good sanitation , supported by many observations that IPV alone—an effective intervention against oral shedding [38–42]—can block transmission and prevent outbreaks from importation in communities with high socioeconomic status [8 , 42 , 96] . In moderate transmission rate settings ( such as Houston 1960 [36] , Louisiana 1953–1955 [35] , or Matlab , Bangladesh 2015 [86] ) , immunologically naive populations can support WPV epidemics , but Rloc≲1 for the Sabin strains and one-dose OPV-equivalent immunity ( NAb = 8 ) are sufficient to block epidemic transmission of all strains . This result is consistent with the historical experience in middle- and high-development countries that WPV elimination rapidly follows the introduction of OPV vaccination [22 , 97–99] and that cVDPV outbreaks are unknown [9 , 43] outside of isolated communities with atypical immunological and social conditions [100–102] . In high transmission rate settings ( such as UP and Bihar 2003–2008 [37] ) , reinfection of previously immunized people can permit community-wide epidemics if typical immunity is below a threshold level . In the example shown , one-dose OPV-equivalent immunity ( NAb = 8 ) has little or no impact on Rloc for any poliovirus strain , and WPV elimination requires NAb > 64 for all . This result , that WPV could persist despite NAb > 8 for most children in UP and Bihar 2003–2008 , is supported by serosurveillance [103] . Prior to WPV elimination , the endemic dynamics of natural infection and vaccination conspire to maintain typical immunity levels near Rloc ( WPV ) ≈1 [84] , and thus the Sabin strains must have Rloc ( Sabin ) <1 , with Sabin 2 highest and Sabin 3 lowest . This result is consistent with the historical experience that vaccine-derived outbreaks have only been observed after genetic reversion has restored WPV-like properties in places where the WPV serotype has been eliminated [43 , 44] , and that type 2 cVDPVs are most common [9] . However , if poliovirus is reintroduced after elimination into a high transmission rate setting with insufficient immunity , our model predicts that epidemic dynamics will be similar for all strains: Rloc ( Sabin ) ≈Rloc ( WPV ) >1 is determined by the number of social contacts and is insensitive to differences in infectiousness of the Sabin strains . Our results above , combined with the observation that cVDPV outbreaks have only been observed at rates of roughly one per year per 250 million children at risk under 15 years of age [9] , indicate that settings where the transmission rate for the Sabin strains is high have been rare . To evaluate how community susceptibility to Sabin 2 transmission will change due to anticipated vaccination policy changes after WPV eradication [95] , we considered four scenarios for childhood immunity against type 2 poliovirus in Fig 10A . The tOPV×3 scenario describes pre-cessation populations in which all index persons , household members , and close social contacts had achieved maximum immunity prior to waning . The bOPV and tOPV×3 scenario applies in the first 2–3 years after type 2 cessation , when birth spacing [104] is such that the likely index child in a family has only received bOPV ( and possibly IPV ) , but older household members and their contacts have had tOPV . The bOPV scenario applies when two or more children in a typical household are born after type 2 cessation , and the naive scenario applies in settings where all OPV immunization has stopped . Prior and up to a few years after type 2 cessation , Rloc ( Sabin2 ) <1 almost everywhere . However , our model predicts that Rloc ( Sabin2 ) >1 will be common when typical households have more than one child born after type 2 cessation and where hygenic practices are comparable to those of UP and Bihar in the years preceeding WPV elimination . Some moderate transmission settings may also become susceptible to Sabin 2 outbreaks once all OPV vaccination is stopped . To relate local reproduction number to data that can be collected in the field , Fig 10B shows our MLEs for the fraction of index children , household members , and close social contacts who shed after mOPV2 challenge of the index child . In well-protected communities ( Rloc≪1 ) , the model predicts little to no measurable transmission from index children infected with Sabin 2 , but when Rloc≫1 , Sabin 2 transmission from index children to unvaccinated contacts will be nearly indistinguishable from WPV [24 , 35 , 37] . Fig 11 shows the sensitivity of the local reproduction number in immunologically naive settings to social distance , measured in terms of the fecal–oral dose ( Ths ) and the household member to social contact interaction rate ( Dhs ) . In moderate transmission rate settings , such as Houston 1960 , Rloc declines rapidly with increasing social distance , even in the absence of immunity . Relative to the calibrated parameters that describe transmission among close contacts , a 10-fold reduction in either fecal–oral dose or interaction rate reduces Rloc from near 1 to less than 0 . 1 . In moderate transmission rate settings , significant transmission requires regular , undiluted contact , and so Sabin 2 is unlikely to spread outside of the communities it is delivered to . However , in high transmission rate settings such as UP and Bihar 2003–2008 , Rloc can remain above 1 across two orders of magnitude in fecal–oral dose or interaction rate—and above 0 . 1 across three . Under these conditions , transmission does not require undiluted fecal–oral contact , and Sabin 2 can escape local communities via social interactions that take place only a few times per year .
We have shown how the effects of immunity on poliovirus shedding and susceptibility to infection interact with sanitation and local interfamilial relationships to determine community susceptibility to poliovirus transmission . We found that the local reproduction number is a useful threshold statistic for characterizing the transmission rate . The highest typical levels of OPV-equivalent immunity in our model predict Rloc < 1 for all strains in all settings . In low and moderate transmission rate settings , we inferred that the Sabin strains have Rloc ( Sabin ) <1 because of attenuated infectiousness relative to WPV ( 9 ) , and thus significant person-to-person Sabin transmission is unlikely regardless of population immunity . Moderate transmission rate settings are at risk of outbreaks from WPV or imported ( wild-like ) cVDPV but are unlikely to generate indigenous Sabin-derived outbreaks . However , in high transmission rate settings with low population immunity—a situation than can only exist in the absence of endemic transmission and OPV vaccination—our model predicts that the transmission rate of the Sabin strains , if reintroduced , will exceed all levels experienced prior to OPV cessation , approaching that of WPV and with highest risk for Sabin 2 . Other published mathematical models known to us have explored the effects of immunity on Sabin transmission [20 , 26 , 31 , 32] . Despite substantial methodological differences , all are in agreement that the Sabin strains will have reproduction numbers above one in high transmission rate settings with low population immunity . In addition to novel results for dose response and waning , the key innovation of our work is its direct connection from individual-level measures of shedding and susceptibility obtained by stool surveys to assessment of community susceptibility ( Fig 10 ) . A recent application of this model to a field transmission study in Matlab , Bangladesh [86] , found that moderate transmission rate conditions exist in a low-income , high-density community in the developing world , where comprehensive maternal and child healthcare and improved sanitation systems are in place [105] . The key limitation of our model is that , while it can predict when the outbreak risk from OPV vaccination is negligible , it cannot address the absolute probability , severity , or geographic scope of outbreaks when they are possible without incorporating additional structural assumptions and calibration data about socially distant transmission . We discuss the relevance of our results for interpreting the history of and implications for vaccination policy in the polio eradication endgame [95] below . Before polio vaccination , most people were immunized against subsequent polio infection by natural exposure to WPV at young ages . The Sabin strains dramatically lowered the burden of paralytic disease by producing unprecedentedly high levels of immunity and displacing WPV . OPV cessation is intended to eliminate the residual disease burden caused by the Sabin strains [5 , 18] , but stopping OPV vaccination will reduce global immunity against poliovirus transmission to unprecedentedly low levels . Many high-income countries with good sanitation and smaller family sizes have maintained polio elimination solely through the routine use of IPV [7 , 106 , 107] . Although IPV alone has little to no impact on susceptibility or shedding in stool ( Fig 7 ) , our results show that the fecal–oral route alone is incapable of supporting epidemic transmission in low transmission rate settings ( Fig 8 ) . When the oral–oral route is required to permit significant transmission , our results indicate that it is possible for IPV alone to prevent outbreaks by reducing oral shedding [34 , 38–41] . The Netherlands is an example of a country where IPV alone has been sufficient . In 1978 and 1992 , there were outbreaks of WPV , but virus was found almost exclusively within high-risk groups who refused vaccination , and no evidence of circulation in the well-vaccinated population was found [96 , 108–110] . Furthermore , many countries that could not have eliminated WPV with IPV alone a few decades ago appear now to be adequately protected . The US is an example . While there is some evidence that IPV alone could reduce WPV transmission among middle- and upper-class families in 1960 [39] , IPV vaccination of subjects with no prior exposure to live poliovirus had no impact on transmission for both Sabin and WPV strains in communities with low socioeconomic status [36 , 111] . However , since 2000 , the US has only used IPV [106] and yet has remained polio-free in all vaccinated populations [101] despite extensive international connections and cross-border mixing with OPV-using countries [112] . The 2013 WPV outbreak in Israel shows the limits of IPV to prevent transmission . Eight years after Israel switched from using both OPV and IPV to using IPV only , a type 1 WPV outbreak was tracked via sewage surveillance from February 2013 until April 2014 [113 , 114] . Most infections were found in children born after the switch despite 93+% coverage with two or more doses of IPV and waning immunity in older people [15 , 115] . A recent model estimated that the effective reproduction number of WPV among children in the Bedouin community in which transmission was most common was 1 . 8 [15]; the corresponding reproduction numbers for the Sabin strains , assuming our model of infectivity , are 0 . 4 and below . Our interpretation is that Israel in 2013 was an example of a moderate transmission rate setting where WPV can persist despite comprehensive IPV vaccination but the Sabin strains cannot [116–118] . In the above scenarios , our model predicts that OPV cessation is stable . OPV can be used to interrupt outbreaks of WPV or imported ( WPV-like ) cVDPV , and the persistence of vaccine-derived strains is unlikely within ( Fig 10A ) or outside ( Fig 11A ) the outbreak response zone . However , in high transmission rate settings with low immunity , we expect that Sabin transmission to unvaccinated contacts within outbreak response regions will be common ( Fig 10 ) and significant transmission to socially distant contacts will occur ( Fig 11B ) . In these settings , OPV cessation is inherently unstable—if poliovirus is reintroduced , there is no guarantee that transmission can be stopped and new cVDPV prevented without restarting OPV vaccination in all high transmission rate settings . Our conclusion that global OPV cessation is unstable follows from the inference that doses acquired via fecal–oral exposure can be much higher in the developing world than they were in the countries where Sabin OPV was first studied and where OPV cessation has already been successful ( Fig 8 ) . The time when instability will reveal itself is uncertain . Our model predicts that two or more children per family born after cessation are required to support Sabin 2 outbreaks in most high transmission rate settings ( Fig 10 ) . The median birth spacing in most bOPV-using countries is 24–36 months [104] . Thus , we predict that between early 2018 and mid-2019 , the risk of establishing type 2 cVDPV will increase substantially in many regions of the developing world that have not received post-cessation mOPV2 campaigns . The cross-immunity from bOPV against type 2 ( with or without IPV ) does not alter this conclusion . Our estimate of 2 to 3 years to increased cVDPV2 risk upon Sabin 2 reintroduction is consistent with predictions from other models [20 , 26 , 32 , 33] and is compatible with the known epidemiology of cVDPV2 outbreaks . The first known example of widespread circulation following a small release of Sabin 2 took place in Belarus in 1965 but was only confirmed as such in 2003 [119] . Two years after a local experiment in type 2 OPV cessation , tOPV given to 40 children likely spread Sabin-derived poliovirus throughout a city of 160 , 000 people for at least 10 months . In northern Nigeria , after widespread vaccine refusal in 2003 and 2004 [120] , restoration of tOPV vaccination seeded 12 independent type 2 Sabin-derived outbreaks , including the largest known outbreak of cVDPV2 in history [44] . The introduction of IPV in routine immunization globally between 2014 and 2016 aimed to provide protection against poliomyelitis to children born after OPV cessation [121] . But without substantial improvements in sanitation , IPV supply [7] , and routine immunization coverage , we expect that IPV alone is insufficient to protect against poliovirus circulation in all settings . In pursuit of high vaccine efficacy with low virulence [4 , 122] , Sabin selected strains that are 1 , 000–10 , 000 times less likely to cause paralysis than WPV [5] but only 4–10 times less infectious ( Table 1 ) . In the absence of population immunity , the differences in infectiousness are insufficient to limit transmission and prevent the evolutionary restoration of virulence [43] . As a consequence , Sabin OPV will be insufficient to guarantee protection from circulation in high transmission settings [20 , 123 , 124] . To secure polio eradication for all times and in all conditions , we believe improved vaccines that produce infection-blocking immunity without the risks of Sabin OPV are required . Genetically stabilized , engineered live vaccines are in development and promise the benefits of Sabin OPV without the risks [125–127] , and adjuvanted IPV may provide a complementary route to a new effective vaccine [128] . Regardless of the challenges detailed above , Sabin OPV vaccination is always preferable to natural infection by WPV or cVDPV . Thus , mass vaccination with OPV remains the most effective intervention to eliminate poliovirus transmission [3] , and the continued use of mOPV2 in regions experiencing type 2 outbreaks is warranted [18] despite concerns about poliovirus containment [19] . For risk mitigation , our model shows the value of healthy contact stool surveillance . The fraction of vaccine recipients and unvaccinated contacts shedding is a direct probe of population immunity and the local transmission rate , and our results provide a rubric to categorize the risk of subsequent outbreaks . Furthermore , with data about fecal–oral contamination ( whether from studies of other enteric diseases or sanitation ) , our model can be calibrated to predict transmission rates in the absence of poliovirus and may thus have predictive value far into the post-cessation future . To go from outbreak risk categorization to risk quantification , continuing work to better understand the relationships between local and nonlocal transmission is needed [129–131] . | Oral polio vaccine ( OPV ) has played an essential role in the elimination of wild poliovirus ( WPV ) . OPV contains attenuated ( weakened ) yet transmissible viruses that can spread from person to person . In its attenuated form , this spread is beneficial as it generates population immunity . However , the attenuation of OPV is unstable and it can , in rare instances , revert to a virulent form and cause vaccine-derived outbreaks of paralytic poliomyelitis . Thus , OPV is both a vaccine and a source of poliovirus , and for complete eradication , its use in vaccination must be ended . After OPV is no longer used in routine immunization , as with the cessation of type 2 OPV in 2016 , population immunity to polioviruses will decline . A key question is how this loss of population immunity will affect the potential of OPV viruses to spread within and across communities . To address this , we examined the roles of immunity , sanitation , and social contact in limiting OPV transmission . Our results derive from an extensive review and synthesis of vaccine trial data and community epidemiological studies . Shedding , oral susceptibility to infection , and transmission data are analyzed to systematically explain and model observations of WPV and OPV circulation . We show that in high transmission rate settings , falling population immunity after OPV cessation will lead to conditions in which OPV and WPV are similarly capable of causing outbreaks , and that this conclusion is compatible with the known safety of OPV prior to global cessation . Novel strategies will be required to ensure the stability of polio eradication for all time . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"infectious",
"disease",
"epidemiology",
"pathogens",
"immunology",
"microbiology",
"viruses",
"preventive",
"medicine",
"physiological",
"processes",
"vaccines",
"rna",
"viruses",
"infectious",
"disease",
"control",
"vaccination",
"and",
"immunization",
"antibodies",
"public",
"and",
"occupational",
"health",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"molting",
"proteins",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"poliomyelitis",
"enteroviruses",
"biochemistry",
"immunity",
"poliovirus",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"organisms"
]
| 2018 | Assessing the stability of polio eradication after the withdrawal of oral polio vaccine |
Obtaining informed consent for clinical trials is especially challenging when working in rural , resource-limited areas , where there are often high levels of illiteracy and lack of experience with clinical research . Such an area , a remote field site in the northeastern part of the state of Minas Gerais , Brazil , is currently being prepared for clinical trials of experimental hookworm vaccines . This study was conducted to assess whether special educational tools can be developed to increase the knowledge and comprehension of potential clinical trial participants and thereby enable them to make truly informed decisions to participate in such research . An informational video was produced to explain the work of the research team and the first planned hookworm vaccine trial , using a pedagogical method based on analogies . Seventy-two adults living in a rural community of Minas Gerais were administered a structured questionnaire that assessed their knowledge of hookworm , of research and of the planned hookworm vaccine trial , as well as their attitudes and perceptions about the researchers and participation in future vaccine trials . The questionnaire was administered before being shown the educational video and two months after and the results compared . After viewing the video , significant improvements in knowledge related to hookworm infection and its health impact were observed: using a composite score combining related questions for which correct answers were assigned a value of 1 and incorrect answers a value of 0 , participants had a mean score of 0 . 76 post-video compared to 0 . 68 pre-video ( p = 0 . 0001 ) . Similar improvements were seen in understanding the purpose of vaccination and the possible adverse effects of an experimental vaccine . Although 100% of participants expressed a positive opinion of the researchers even before viewing the film and over 90% said that they would participate in a hookworm vaccine trial , an increase in the number who expressed fear of being vaccinated with a novel vaccine was seen after viewing the video ( 51 . 4% post-video versus 29 . 2% pre-video ) . Increases were also seen in the proportion who thought that participation in a vaccine trial would be inconvenient or disrupt their daily activities . Even in rural , resource-limited populations , educational tools can be specially designed that significantly improve understanding and therefore the likelihood of obtaining truly informed consent for participation in clinical research . The observed changes in the knowledge and perceptions of the research participants about hookworm infection and the experimental hookworm vaccine demonstrate that the video intervention was successful in increasing understanding and that the subjects acquired knowledge pertinent to the planned research .
In research involving human subjects , the ethical relationship that must be established and maintained between investigators and research subjects is essential to successfully conduct investigational clinical trials of experimental drugs or vaccines , especially ones in which the participants are drawn from vulnerable populations [1] , [2] . In such studies , investigators must attempt to mitigate any possible manipulation of the research population during the recruitment process , and especially to ensure that the risks and benefits to which volunteers are going to be exposed are well understood [3] , [4] . The informed consent document is the traditional instrument utilized for this aim; by signing it , it is assumed that the volunteer has freely exercised their will , has formed their own evaluation and critique of the proposed research and has arrived at a truly informed decision about participation [5]–[8] . The process of obtaining informed consent becomes especially challenging when working in rural , resource-limited settings . Although there is no consensual definition of vulnerability , age , socioeconomic status , access to basic services such as health and sanitation , ethnic group , religion , cultural affiliation , and educational level are all characteristics that have been cited as indicating vulnerability , and which may therefore influence an individual's ability to consent to participation in clinical trials both in terms of the ability to exercise autonomy , but also to comprehend the proposed research [3] , [9] , [10] . With respect to the latter challenge , the informed consent form must frequently convey complex technical information and scientific concepts that are often not easily understandable , especially in populations with low levels of literacy [11]–[13] . To improve the quality of the information transmitted to potential study participants , researchers have , in general , increased the amount of information in informed consent documents [11] . However , a document that contains extensive and complex information may not convey a satisfactory understanding of the study procedures , or of the potential risks and benefits of participation in the research [14] . As an example , after evaluating the understanding of the information in an informed consent document for a research project in San Francisco , California , researchers found that despite using a form that had been simplified using language appropriate for a primary school level , the majority of individuals required more than one explanation of the study before satisfactorily comprehending it [12] . In that study , low literacy level and socioeconomic status were associated with an increased need for interventions that gave an improved comprehension of the information contained in the document . In addition to the issue of comprehension , recruiting volunteers into clinical trials being conducted in resource-limited settings is further complicated by the limited access to medical care that is frequently found in such areas . Frequently , potential research subjects may feel an obligation to participate in order to receive medical attention for themselves or their family members [15] . Such motivation could influence individuals to participate and undertake risks that they otherwise would not accept . Despite these very real issues related to obtaining informed consent from vulnerable populations living in resource-limited settings , it is often necessary to conduct clinical trials in such areas , particularly when the product being developed is meant to treat or prevent a disease that affects such a population . As one example , hookworm infection is one of the most prevalent chronic infections of humans , with an estimated 740 million cases worldwide , mostly in rural poor rural areas of the tropics and subtropics [16] . The two hookworms that infect humans are Necator americanus and Ancylostoma duodenale , with infection being transmitted through skin contact with soil contaminated with infective larvae . The major clinical manifestations result primarily from the loss of blood caused by adult worms that attach onto the intestinal wall , resulting in anemia which subsequently can lead to delays in cognitive development in children and reduction of well-being and productivity in adults [17] . Although effective chemotherapy exists to treat hookworm , current anthelminthics have important limitations , not least of which is that re-infection often occurs within a short time after treatment due to ongoing exposure [18] , [19] . To develop an alternative control tool , the Human Hookworm Vaccine Initiative ( HHVI ) is developing a vaccine to prevent the morbidity due to this parasitic infection . Since hookworm does not occur in the developed world , testing the safety and efficacy of vaccines targeting this parasite must be conducted in the rural , resource-limited areas where the disease is endemic , among populations which are frequently referred to as being “vulnerable . ” The HHVI has been preparing a trial site for studies of its investigational hookworm vaccines that is based in the town of Americaninhas , in the northeast part of the Brazilian state of Minas Gerais [20] , [21] . The first experimental vaccine being developed by the HHVI – the Na-ASP-2 Hookworm Vaccine – was tested there in a phase 1 clinical trial in 2007 . For this study , healthy adult volunteers were recruited from communities surrounding Americaninhas . In advance of this trial , studies were performed to assess the baseline knowledge of potential study participants in order to design appropriate educational interventions that could be used in the consenting process . Unfortunately , little is known about which are the more appropriate pedagogic models for informing populations involved in clinical trials , especially vulnerable populations that are economically or educationally disadvantaged . In the field of health education , purely informative pedagogic models do not usually result in modification of positions or attitudes , since behaviors are manifestations of firmly-held values and beliefs [22] . The ideal nature of an educational intervention that leads to acquisition of the knowledge necessary to make a conscious decision about participating in a research study is a matter of debate . Several authors have proposed a pedagogic model that makes use of the analogy [11] , [23] . Analogies are useful tools for forming mental constructs that simplify or render familiar what an individual is attempting to understand [24] . The use of analogies can introduce new scientific concepts or alter previously held ideas [25] , and can overcome barriers to learning by allowing an individual to make creative connections between pre-existing concepts and those related to the new knowledge being presented [23] . With this in mind , the current study aimed to develop an educational intervention based on analogies for a population resident in a hookworm endemic area of Brazil , and to evaluate its effectiveness in disseminating knowledge about the disease caused by hookworm , the experimental Na-ASP-2 vaccine that was about to be tested in their community , and about attitudes related to their willingness to participate in clinical trials of the vaccine .
The study was conducted in 2007 in the communities of Furtado , Beija-Flor and Jamir , all of which are rural areas endemic for hookworm located in the region surrounding the town of Americaninhas in the municipality of Novo Oriente de Minas , 500 kilometers northeast of Belo Horizonte , the capital of the Brazilian state of Minas Gerais . Americaninhas is located in a mountainous region with a tropical climate [20] , [26] . The population largely exists on subsistence farming of cassava , sugar cane , coffee and beans . They typically live in simple , hand-made dwellings of packed earth or adobe with roofs of corrugated iron . The population of Americaninhas consists of around 1000 people living in the urban center , with another 4000 living in the surrounding rural areas in smaller hamlets . Approximately 500 people live in the communities of Furtado , Beija-Flor and Jamir . This region was chosen for conducting clinical trials of experimental hookworm vaccines in view of the elevated prevalences of helminth infections that have been found during previous studies performed by the research team in the area: 68% for the hookworm N . americanus , 45% for Schistosoma mansoni , and 49% for Ascaris lumbricoides [20] . Such high prevalences are the result of socioeconomic and environmental conditions that favor the transmission and development of hookworm , such as a warm and humid climate , a lack of basic sanitation , and the low socioeconomic status of the population . Individuals were included in the study if they were between the ages of 16 and 50 years and had been resident in the study area for the previous 24 months; were infected with hookworm as determined during the course of the larger epidemiological study; and , had completed a course of treatment for hookworm with albendazole . All volunteers consented to participate in the research , as evidenced by their signature on the informed consent form approved by the ethical review committees of the Centro de Pesquisas Rene Rachou ( part of the Fundação Oswaldo Cruz – FIOCRUZ ) and the George Washington University Medical Center , and completion of a true/false questionnaire that assessed their understanding of the informed consent document . Volunteers had to respond correctly to all questionnaire questions prior to being considered for inclusion into the study . In the cases of volunteers unable to read , the consent form was read aloud to them in the presence of an independent witness who also signed the form after the volunteer affixed their thumbprint . The intervention was an educational video containing details about the proposed Na-ASP-2 vaccine trial , as well as explanations regarding the nature of research , the work of researchers , the reasons the area was chosen to test a vaccine against hookworm , and concepts of hookworm disease , vaccination , and the use of placebos in vaccine trials ( Videos S1 , S2 , S3 , S4 , S5 , S6 , S7 , S8 , and S9 ) . The video was filmed in the communities of Jamir and Beija Flor , and was produced using a pedagogical approach based on the use of analogies . In it , the daily experiences of local inhabitants , such as the cultivation of cassava , and the making of flour , sweets and cheese , are compared to the manufacturing of vaccines and to the experiments of researchers working in the laboratory . The characters featured in the film are actual inhabitants of the community who are representative of those from the rural interior of the state of Minas Gerais , thus facilitating identification of the viewer with the individuals on screen and enabling the learning process . The film opens with scenes showing typical day in the town of Americaninhas , with people enjoying themselves in the town square or observing the main street from their windows ( Video S1 ) . After these initial images , the film transitions to describing the production of a traditional regional sweet . Each step in its production is shown , starting with cultivation of the sugar cane from which the basic ingredient is derived , followed by extraction of cane juice using a machine , preparation of the other ingredients , and combining these in specific quantities to create a final , high-quality product . Interspersed with these images are those of FIOCRUZ researchers working in the laboratory , using machines and instruments to assist them in discovering ideal components that , when combined in the correct amounts , may produce an effective vaccine . An analogy is thereby constructed between typical experiences of the region such as the production of sweets and the manufacturing of a vaccine against hookworm . The video then shows people being interviewed about their knowledge of hookworm using language that is unique to the local population ( Videos S2 , S3 , and S4 ) . The individuals discuss the illness known locally as “amarelão” or the “illness of Jeca-tatu” ( after a popular cartoon character from the early 20th century ) , its mode of transmission , and the associated symptoms . Their perceptions of the researchers working in the area and what they're doing in the region , as well as the hookworm vaccine program and the possible adverse effects of such a vaccine , are also presented . In the final part of the video , members of the HHVI team speak on camera about the hookworm vaccine project , to clarify details of the planned clinical trial ( Videos S5 , S6 , S7 , S8 , and S9 ) . The presenters explain why hookworm is endemic in their community , the criteria for inclusion in the forthcoming vaccine trial , the adverse effects that the vaccine might cause , as well as how the vaccine is made . The video was shown to prospective vaccine trial participants in group sessions in their own communities . After presentation of the film , conversation was encouraged to discuss and debate what was viewed so that the presented knowledge could be consolidated and learned . Data were collected by means of a structured survey consisting of 45 questions , which was designed to assess knowledge about hookworm , vaccines ( in general and the Na-ASP-2 vaccine in particular ) , and the researchers , as well as attitudes related to their willingness to participate in clinical trials of hookworm vaccines ( Text S1 ) . The questionnaire consisted of a combination of true or false questions , multiple choice questions , and subjective questions answered according to the 5-point Likert scale ( ranging from “strongly disagree” to “strongly agree” ) . Survey questions were divided into three categories: a ) those assessing knowledge about hookworm ( Group 1 ) ; b ) those assessing knowledge about the hookworm vaccine and upcoming clinical trial ( Group 2 ) ; and , c ) those assessing the attitudes and feelings of individuals about illness due to hookworm , vaccines , and participation in vaccine trials ( Group 3 ) . The survey was first pilot tested on a group of 20 adults . After conducting this pilot test , modifications were made to improve the understanding of the tool by the general public . The survey was administered by specially trained interviewers at two distinct times: once immediately before viewing the educational video described above and then approximately two months later . Data were tabulated and analyzed using the SPSS software program ( version 15 ) . Of the 45 questions on the questionnaire , only questions 2 through 40 were included in the analysis ( the first question was for informational purposes whereas the final three concerned only those participants who had children ) . For questions from Groups 1 and 2 , the frequencies of each answer were summarized . Responses to these questions were dichotomized such that each correct answer was assigned a value of “1” and each incorrect answer a value of “0”; a response of “don't know” was also assigned a value of “0” as it was considered a lack of knowledge . Subsequently , the mean of the responses was calculated for both the pre- and post-film administrations of the questionnaire . Additionally , composite scores for knowledge about hookworm and knowledge about the experimental hookworm vaccine were obtained by calculating the mean of responses to questions that fell into these broad categories . For the knowledge about hookworm composite score , responses to 14 similar questions were combined ( #2–14 , #26 ) whereas for knowledge about the hookworm vaccine 6 questions ( #15 , #16 , #18–20 , #30 ) were combined . Questions in Group 3 were divided into two subgroups which were analyzed using different methods: for the first subgroup , answers were dichotomized such that an affirmative answer was assigned a value of “1” and a negative answer a value of “0” whereas for the second subgroup , the Likert scale consisting of five categories was used: −2 ( strongly disagree ) , −1 ( disagree ) , 0 ( neither agree nor disagree ) , 1 ( agree ) , and 2 ( strongly agree ) . As for the dichotomous responses , means were calculated for each Likert scale question and compared pre- and post-film; to do this , the “strongly disagree” and “disagree” categories were combined and assigned a value of “0” , and the “strongly agree” and “agree” categories were assigned a value of “1” . As for the first two groups of questions , a composite score for the attitudes and feelings of study participants was created by combining individual responses to 7 different questions from the questionnaire ( #21–23 , #27–29 , #31 ) . Student's paired t-test was used to compare pre-film and post-film means for both individual questions and the composite scores for Groups 1 and 2 . The chi-square test was used to compare proportions for questions with several possible responses . For all tests , a p value less than 0 . 05 was considered significant .
When assessed as a composite score , knowledge about hookworm improved significantly after viewing the informational video from a mean score of 0 . 68 before the video to 0 . 76 after the video ( p<0 . 0001 ) , demonstrating that significant understanding was acquired by the participants through the targeted educational intervention . When assessing knowledge about vaccines and the proposed clinical trials of the experimental Na-ASP-2 Hookworm Vaccine , a small improvement was also seen after viewing the video , with the mean of correct answers on the post-video questionnaire being significantly higher than the mean on the pre-video test ( 0 . 58 vs . 0 . 65 , p = 0 . 03 ) ( Table 1 ) . Table 2 describes in more detail the specific knowledge acquired about hookworm after viewing the film . The illness , recognized by the popular name “amarelão” ( “the big yellow” ) by 88 . 9% of participants prior to viewing the film , was identified by the more scientific name “ancilostomídeo” by 91 . 7% of the study subjects after seeing the film ( compared to only 79 . 2% before , p = 0 . 01 ) . The mode of transmission of hookworm , which in Brazil is sometimes confused with other worm infections such as A . lumbricoides and S . mansoni , was already correctly identified as being through contact of skin with contaminated soil by 95 . 8% of subjects before the film , which increased to 100% after viewing the video ( p = 0 . 08 ) . However , the common misconception that hookworm infections are acquired through contact with contaminated water and unwashed fruit or vegetables , was abandoned by a quarter of the subjects participating in the research after viewing the educational film ( 88 . 9% pre-film vs . 63 . 9% post-film , p<0 . 001 ) . After the educational intervention , 79 . 2% of participants recognized that individuals can be infected with hookworm burt be asymptomatic , compared to 66 . 7% who held this view prior to seeing the film ( p = 0 . 1 ) . The acquisition of this knowledge may be crucial because health surveillance is intimately associated with the recognition of the gravity of the disease and the acknowledgement that it can be a “silent” illness . When asked about the negative health consequences associated with hookworm infection , even before viewing the educational film , 90 . 3% of participants correctly identified anemia as an important result of hookworm infection , compared to 93 . 1% after the film ( p = 0 . 5 ) . Differences were found before and after the educational intervention when assessing the level of subjects' comprehension about the impact of illness due to hookworm in their community . Although even before viewing the video 84 . 7% of the population believed that hookworm was an important illness in their community , which increased slightly to 91 . 7% of participants afterward ( p = 0 . 2 ) , a significant change was seen in the more subtle question that asked whether participants thought that hookworm is not a serious illness because it can be easily treated: initially , 72 . 2% responded in the affirmative to this question whereas after viewing the educational film and hearing more about the long-term consequences of asymptomatic infection , only 45 . 8% held this viewpoint ( p<0 . 001 ) . Furthermore , after viewing the educational video more people understood that despite the existence of effective drug therapy for hookworm infection , there are major limitations with this treatment due to the probability of becoming re-infected , which in many cases can occur rapidly following treatment , although this increase in knowledge wasn't statistically significant ( 30 . 6% pre-video vs . 38 . 9% post-video , p = 0 . 2 ) . When evaluating the participants' knowledge about vaccines in general and the upcoming hookworm vaccine trial in particular , improvements in understanding were acquired after viewing the educational film ( Table 3 ) . Regarding the purpose of vaccination , the participants' knowledge before and after the educational intervention is shown in Figure 1 . Results from the pre-film survey demonstrate that a majority of participants ( 56 . 9% ) believed that the purpose of a vaccine is to treat an established illness . Although following the film , this association was still made by almost half of responders ( 45 . 8% ) , a significant increase did occur in those associating a vaccine with illness prevention ( 41 . 7% post-video vs . 20 . 8% pre-video , p = 0 . 005 ) . To assess knowledge about the possible adverse effects caused by vaccination , participants were asked what they thought could happen if an experimental vaccine were administered to them . The potential side effects of vaccination that they chose are presented in Figure 2 . Among the responses , the possibility of experiencing an allergic reaction upon being vaccinated , which was recognized by none of the interviewed participants prior to viewing the educational video , was cited frequently during the post-film test ( 43 . 1% ) , as was the possibility of experiencing arm redness ( at the site of injection ) following vaccination ( 15 . 3% post-video compared to 1 . 4% before , p = 0 . 003 ) . Of note , the educational intervention was associated with a significant reduction in the proportion of individuals who answered “other” ( 18 . 1% post-film vs . 54 . 2% pre-film , p<0 . 001 ) and a non-significant reduction in those who chose “don't know” as their response to this question ( 9 . 7% post-film vs . 18 . 1% pre-film , p = 0 . 15 ) . The participants' perceptions about the nature of the work being done by the FIOCRUZ research team in the study area are shown in Figure 3 . The level of comprehension regarding the work of the researchers underwent significant change following the presentation of the interventional video . Although the concept that the work of a researcher is to take care of people's health and treat disease remained the perception of 30 . 6% and 33 . 3% of the interviewed subjects , respectively , compared to 36 . 1% and 31 . 9% who listed these roles prior to viewing the film ( p = 0 . 5 and 0 . 9 , respectively ) , there was a significant increase in the proportion of participants who listed the function of the researcher as consisting of conducting studies on a new vaccine ( 27 . 8% post-film vs . 4 . 2% pre-film , p<0 . 001 ) . Regarding survey questions pertaining to the attitudes and feelings of the participants towards the researchers and participation in a hookworm vaccine trial , there was no significant change in the composite mean of responses before compared to after viewing the film ( 0 . 86 vs . 0 . 83 , p = 0 . 07 ) ( Table 4 ) . Similarly , no significant difference was found between the attitudes and feelings expressed pre-film and post-film when those questions evaluated using the 5-point Likert scale were combined as shown in Figure 4 . However , for both of these measures , the attitudes of the study participants were somewhat less favorable towards participating in future vaccine trials after viewing the informational video , even if these differences weren't statistically significant . Even before the educational intervention took place , more than 95% of the participants interviewed already expressed a favorable attitude toward the planned vaccine study , displayed confidence in the work of the researchers and expressed interest in learning more about hookworm and the experimental hookworm vaccine ( Table 5 ) . In fact , 100% of respondents agreed that the researchers are doing good work in their community , a percentage that remained unchanged after viewing the film . After the educational intervention , although no significant changes were recorded in the attitudes and feelings regarding aspects of the hookworm vaccine project , there were reductions in the number of individuals who were interested in participating in a future vaccine trial ( 95 . 8% before compared to 88 . 9% after ) and in those who said that their family would approve of their participation in a vaccine trial ( 86 . 1% before compared to 80 . 1% after ) . Closely related to these attitudes was the initial perception that participating in a vaccine study would result in focused attention on and treatment of health problems of study participants ( 95 . 8% ) , an aspect that certainly implies improvement of the health and the quality of life of each participating individual and indeed , even of others in the community ( 91 . 7% ) ( Table 5 ) . Regarding the attitudes and feelings expressed by those surveyed toward the researchers and the hookworm vaccine project , it is important to note that the proportion of people expressing the opinion that being a volunteer in a research study may complicate daily activities or prove inconvenient increased slightly after viewing the video: 15 . 3% and 23 . 6% held these views prior to watching the film compared to 23 . 6% and 27 . 8% after ( p = 0 . 2 and 0 . 07 , respectively ) . Furthermore , the percentage of participants who expressed fear of becoming ill or experiencing an adverse event also increased significantly , from 29 . 2% to 51 . 4% , respectively ( p<0 . 01 ) .
Our study has demonstrated that despite conducting research in a rural , resource-limited population that has limited access to routine health care and education , specially-designed educational activities can significantly improve understanding and therefore the likelihood of obtaining truly informed consent for clinical research . The observed changes in the knowledge and perceptions of the research participants about hookworm infection and the experimental hookworm vaccine clearly demonstrate that the targeted educational intervention was successful in increasing understanding and that the subjects acquired knowledge pertinent to the planned research . Importantly , the increase in knowledge appeared to be sustained , as the post-video questionnaire was conducted two months following the viewing of the film and not immediately afterward . The conceptual evolution following an educational intervention can be attributed to the non-cognitive educational methodology that was utilized . The video that was developed and evaluated as part of this study relied on the use of analogies – for example , comparing the production of a new vaccine to the making of a local sweet – to convey new scientific concepts related to hookworm and research . The analogy is a comparison based on similarities between the structures of two different fields of knowledge [27] . Reasoning through analogy is , therefore , a subjective internal process that is effectuated by the interaction between two mental fields . The results of the questionnaire demonstrate that this pedagogical methodology is effective in the population that was studied . The educational approach chosen for this study differs from conventional educational methods , in that it considers the lifestyle of people , their ideas , beliefs and values , and the specific cultural context . This leads to enhanced self-esteem , increased community participation , thus promoting the values of citizenship [28] . As outlined by Rice and Candeias , the traditional educational model in which information is simply provided to and individual or community has only temporary effects in terms of behavior change [29] . When the educational stimulus ceases , so too are its effects . The main criticism of the traditional cognitive approach is that it does not take into consideration the psychosocial and cultural determinants of health behaviors [30] . In a previous study that we conducted in a rural area of Minas Gerais where schistosomiasis is endemic , different health education approaches were assessed regarding their effectiveness in increasing the knowledge of schoolchildren with respect to the transmission and prevention of this parasitic disease [31] . This study demonstrated that in a group of schoolchildren whose education was based upon a model using analogies or social representations , levels of knowledge about schistosomiasis increased significantly , compared to those in which a cognitive model based on the simple presentation of information , or a control group that received no specific information about the disease . Regarding the specific changes in knowledge that were observed in the current study after the educational intervention , an increased appreciation of the illness caused by hookworm was observed , such that after watching the film , it was identified as an important affliction in the study area due to its endemnicity and its effect on people's lives . The understanding that hookworm infection is not a major health problem because it can be easily treated with anthelminthic medications was reduced following the film , serving as evidence of an improved understanding of the re-infection process wherein individuals are continuously at risk of infection despite anthelminthic treatment due to the environment in which they live . In analyzing the attitudes and perceptions of the participants , it appears that there were no significant changes in the subjects' notions in relation to the researcher and the benefits of participating in clinical trials . Capturing the participants' perspective before and after the educational intervention , no change was observed . Initially , an overall favorable opinion toward the researchers and the project was observed , with a full 100% agreeing that the investigators are doing good work in their community . This good-will and trust towards the researchers remained unchanged after viewing the educational film , and it translated into a high level of interest in participating in future vaccine trials , although this willingness decreased slightly after viewing the film , perhaps as a result of an increased understanding of the risks involved , as discussed below . According to the vast majority of the individuals who were studied , participating in a vaccine trial might not only benefit themselves but also bring benefit to others who are not participants , resulting in betterment of the community as a whole . Surprisingly , although some participants acknowledged that participating in a vaccine trial might interfere with their daily activities , the majority of those interviewed said that it would not be inconvenient to them . The time commitment involved , which for these individuals whose livelihoods depend on long hours of hard manual labor , could be significant but would apparently be outweighed by the perceived potential benefits of participation such as improvement of their own health , greater attention to treatment of health problems , and potential identification of health problems that are otherwise unrelated to the study or vaccination with an experimental product . A broad increase in understanding of the work being conducted by the research team after viewing the educational film was observed among those surveyed , although the idea that the project's purpose is to treat illness and take care of people's health – instead of to conduct research – remained prevalent despite having viewed the video . Regarding this specific point , it is important to highlight that understanding the research process is a challenge for investigators who are involved in the preparation of communities in advance of conducting clinical trials and beginning the individual informed consent process , especially in resource-limited communities with low levels of literacy and limited access to routine health care [32] . Since the Na-ASP-2 Hookworm Vaccine is an experimental vaccine , individuals participating in the planned study of this investigational product could potentially experience adverse reactions of variable degrees of severity , although in the first clinical trial of this vaccine administered to healthy , hookworm-naïve volunteers living in the United States , observed reactions consisted mostly of mild to moderate intensity injection site reactions such as pain , swelling and erythema [33] . After viewing the film in which the potential risks of participating in the proposed vaccine trial were described , an increase in apprehension related to participation was observed . This should not necessarily be seen as a negative outcome of the educational video , as it may in fact reflect a superior understanding of the risks involved when participating as a volunteer in a clinical vaccine trial – something that should be welcomed . Even though the educational intervention may have resulted in fewer people who would be willing to participate ( 88 . 9% after watching the video compared to 95 . 8% before ) , if those individuals who remained interested in volunteering were better informed , the intervention was successful . In contrast to an increased appreciation of the risks of participation in a vaccine trial , no significant changes were observed after the educational intervention in relation to the perceived benefits that might come from participating in the research . Among the perceived benefits of participating in a vaccine trial were improvements in health , treatment of illness , and a better quality of life . As demonstrated by our findings , the educational intervention utilized in this study was not uniformly successful . In several instances , erroneous perceptions of the study participants – such as the belief that the purpose of a vaccine is to treat disease , or that the role of the research is to take care of the health problems of the population – persisted in a significant proportion of those interviewed even after viewing the educational video . Obviously , although these misconceptions were reduced by the use of the educational video , further work remains to better inform potential research participants of the nature of research and the purpose of interventions that might be tested in future research trials . As with any ongoing research project , obtaining informed consent from volunteers is an on-going process in which individuals are continuously engaged in educational activities . In summary , not only is it important to assess the current level of understanding of potential vaccine trial participants prior to conducting clinical studies , it is also useful to design specially-tailored educational interventions to develop a more informed community that is able and willing to participate in such research . This process should be continuous , with frequent re-assessment of the understanding of individuals that results in revision of the educational materials . It is the ethical imperative of the investigators and research team to ensure that potential study participants have an adequate understanding of the research to be undertaken . In resource-limited areas such as our study site , this often requires more than simply reading the informed consent document , and may include targeted educational activities similar to the one tested at our study site . | Conducting clinical trials of new vaccines in rural , resource-limited areas can be challenging since the people living in these areas often have high levels of illiteracy , little experience with clinical research , and limited access to routine health care . Especially difficult is obtaining informed consent for participation in this type of research and ensuring that potential participants adequately understand the potential risks and benefits of participation . The researchers have been preparing a remote field site in the northeastern part of the state of Minas Gerais , Brazil , for clinical trials of experimental hookworm vaccines . A special educational video was designed based on the method of analogies to introduce new scientific concepts related to the researchers' work and to improve knowledge of hookworm , a disease that is highly prevalent in their community . A questionnaire was administered both before and after the video was shown to a group of adults at the field site , which demonstrated the effectiveness of the video in disseminating knowledge about hookworm infection and about the vaccine being developed . Therefore , even in a rural , resource-limited area , educational tools can be specially designed that significantly improve understanding and therefore the likelihood of obtaining truly informed consent for participation in clinical research . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"non-clinical",
"medicine/sociology",
"infectious",
"diseases/helminth",
"infections",
"non-clinical",
"medicine/bioethics",
"infectious",
"diseases/neglected",
"tropical",
"diseases"
]
| 2010 | Health Education through Analogies: Preparation of a Community for Clinical Trials of a Vaccine against Hookworm in an Endemic Area of Brazil |
Pathogenic microbes employ a variety of methods to overcome host defenses , including the production and dispersal of molecules that are toxic to their hosts . Pseudomonas aeruginosa , a Gram-negative bacterium , is a pathogen of a diverse variety of hosts including mammals and the nematode Caenorhabditis elegans . In this study , we identify three small molecules in the phenazine class that are produced by P . aeruginosa strain PA14 that are toxic to C . elegans . We demonstrate that 1-hydroxyphenazine , phenazine-1-carboxylic acid , and pyocyanin are capable of killing nematodes in a matter of hours . 1-hydroxyphenazine is toxic over a wide pH range , whereas the toxicities of phenazine-1-carboxylic acid and pyocyanin are pH-dependent at non-overlapping pH ranges . We found that acidification of the growth medium by PA14 activates the toxicity of phenazine-1-carboxylic acid , which is the primary toxic agent towards C . elegans in our assay . Pyocyanin is not toxic under acidic conditions and 1-hydroxyphenazine is produced at concentrations too low to kill C . elegans . These results suggest a role for phenazine-1-carboxylic acid in mammalian pathogenesis because PA14 mutants deficient in phenazine production have been shown to be defective in pathogenesis in mice . More generally , these data demonstrate how diversity within a class of metabolites could affect bacterial toxicity in different environmental niches .
The Gram-negative bacterium Pseudomonas aeruginosa , a pathogen of both plants and metazoans , is a prevalent and pernicious pathogen in persons who are immunocompromised or suffer from cystic fibrosis ( CF ) [1] , [2] . P . aeruginosa employs many mechanisms to antagonize its hosts , including the production of low molecular weight toxins [3] , [4] , [5] . Identifying toxins , the conditions under which they are produced , and the mechanisms by which they act , are of fundamental importance in understanding and combating the virulence of this clinically-important pathogen . The nematode Caenorhabditis elegans , which is found in decaying plants where many pathogenic microbes reside , is a useful model host for a variety of pathogens including P . aeruginosa [6] , [7] . PA14 is a virulent clinical isolate of P . aeruginosa that is capable of killing C . elegans [8] , [9] . Previous studies using C . elegans as a model system for PA14 pathogenesis determined that PA14 can kill C . elegans either as a consequence of intestinal infection or intoxication , depending on the media on which the bacteria are grown [8] , [9] . Nematodes die in hours when PA14 is grown on a nutrient rich agar containing peptone , glucose , and sorbitol ( PGS agar ) [8] . This type of killing is referred to as ‘fast killing’ . In contrast , when the bacteria are grown on a less rich medium , it takes several days for C . elegans to die [9] . This type of killing is referred to as ‘slow killing’ . Fast killing is thought to be mediated by diffusible toxins because exudates of PA14 grown on PGS agar are sufficient to kill C . elegans; worms need not be in the presence of live PA14 in order to be killed . In contrast , slow killing requires bacterial growth in the worm gut to effect pathogenesis . Experiments in mice have demonstrated that the fast killing toxin-based model is relevant in plant and mammalian pathogenesis because mutants that are defective in fast killing are also defective in Arabidopsis thaliana and murine PA14 infection models [8] . In this study we focus on identification of the toxins responsible for fast killing . In a previously published screen for PA14 mutants that exhibited reduced levels of fast killing , several isolates displayed reduced levels of the blue-green phenazine pigment pyocyanin [8] . Phenazines are a class of tricyclic aromatic molecules produced by P . aeruginosa and several other Gram-negative and Gram-positive bacteria [10] , [11] , [12] . Some phenazines , especially pyocyanin , have been shown to act as toxins against bacteria , fungi , or mammals as a consequence of their redox activities [12] , [13] , [14] , [15] . Only a subset of the previously identified C . elegans fast killing-deficient PA14 mutants were found to produce less pyocyanin than wild-type [8] , suggesting that other phenazines or a different class of molecules are involved . Moreover , phenazine toxicity has not been demonstrated directly in C . elegans [8] . In order to better understand the mechanisms of P . aeruginosa PA14 toxicity , we sought to identify the toxin molecules produced by PA14 that kill C . elegans . We demonstrate that three of the phenazines produced by PA14 can rapidly kill C . elegans: phenazine-1-carboxylic acid kills C . elegans at acidic pH; pyocyanin , a product of phenazine-1-carboxylic acid , kills C . elegans at neutral or basic pH; 1-hydroxyphenazine , a second product of phenazine-1-carboxylic acid , kills C . elegans in a pH-independent manner . We also show that under the conditions of the fast killing assay phenazine-1-carboxylic acid , not pyocyanin , is the primary toxin responsible for the rapid death of C . elegans in the presence of PA14 .
As described above , among the previously isolated PA14 mutants that are deficient in toxin-mediated killing of C . elegans on PGS agar , those with the largest reduction in toxicity were found to produce less of the phenazine pyocyanin than wild-type PA14 [8] . Pyocyanin is one of at least four phenazines that are produced by wild-type PA14 [10] , [16] , [17] ( Figure 1A ) . Phenazine-1-carboxylic acid , the precursor of all other phenazines produced by P . aeruginosa , is synthesized from chorismate by genes constituting the redundant phzA1-G1 and phzA2-G2 operons , each of which encodes a full set of functional phenazine-1-carboxylic acid biosynthetic enzymes [17] . Phenazine-1-carboxylic acid can be modified by other enzymes to make 1-hydroxyphenazine , phenazine-1-carboxamide , or pyocyanin [17] . To determine the importance of phenazines in the pathogenesis of an established C . elegans model host system , we tested the killing ability of a PA14 mutant that does not produce any phenazines ( Δphz ) . The Δphz mutant is missing both the phzA1-G1 operon and the phzA2-G2 operon , precluding the production of phenazine-1-carboxylic acid as well as the other phenazines [16] . Lack of phenazine production by the Δphz mutant was confirmed by metabolite profiling ( Table 1 ) . Wild-type or Δphz mutant PA14 bacteria were spread on PGS agar plates and allowed to grow for 24 hours at 37°C followed by 24 hours at 23°C . L4 stage worms were then placed on the agar . Worms were scored as live or dead based on whether or not movement could be elicited by tapping their heads gently with a thin wire . We found that the Δphz mutant is severely compromised in its ability to kill C . elegans compared to wild-type PA14 ( Figure 1B ) . Transformation of Δphz with a multicopy plasmid containing either the phzA1-G1 operon or the phzA2-G2 operon partially complemented the killing-deficient phenotype ( Figure 1B ) . Chemical complementation of the Δphz phenotype by the addition of 100 µg/mL of synthetic phenazine-1-carboxylic acid to the agar prior to bacterial growth restored the nematode-killing phenotype ( Figure 1C ) , suggesting that phenazine-1-carboxylic acid or another phenazine derived from phenazine-1-carboxylic acid could be a toxin involved in PA14-mediated killing of C . elegans . Alternatively , phenazine-1-carboxylic acid could function indirectly to regulate the production of a toxin that kills C . elegans . To determine if phenazines are sufficient to kill worms , we tested the killing activities of synthetic phenazines directly by adding them to PGS agar in the absence of bacteria . Under these conditions , 1-hydroxyphenazine killed worms at concentrations above 16 µg/mL with kinetics similar to a typical fast killing assay with PA14 , whereas phenazine-1-carboxylic acid , pyocyanin , and phenazine-1-carboxamide did not kill worms on a relevant time scale , even at much higher concentrations ( Figures 2A , S1A ) . To account for potential synergistic effects of phenazines with other metabolites produced by PA14 , we tested the killing activities of synthetic phenazines added to PGS agar after growth of a lawn of Δphz bacteria . After growth of the lawn , the bacteria were scraped off the agar , the agar was melted in a microwave , phenazines were mixed in at several concentrations , and the agar was allowed to cool prior to introduction of the worms . We refer to the melted and cooled agar upon which the Δphz bacteria had been grown as “Δphz agar” . Similarly to the data shown in Figure 2A , when 1-hydroxyphenazine was added to Δphz agar it killed worms rapidly at concentrations above 16 µg/mL , whereas pyocyanin and phenazine-1-carboxamide killed worms poorly , even at much higher concentrations ( Figures 2B , S2B ) . Interestingly , contrary to its activity on naive PGS agar , on Δphz agar phenazine-1-carboxylic acid also killed worms at concentrations above 16 µg/mL . These data , together with the observation that 1-hydroxyphenazine appeared to be somewhat more toxic to worms when added to Δphz agar than when added to naive PGS agar ( compare Figures 2A and 2B ) , suggested that the Δphz strain produces a factor or factors that enhance the toxicities of 1-hydroxyphenazine and phenazine-1-carboxylic acid . The data in Figure 2B showing that both 1-hydroxyphenazine and phenazine-1-carboxylic acid kill C . elegans when added to Δphz agar indicated that either or both of these phenazines could be responsible for PA14-mediated intoxication of C . elegans in the fast killing assay . In considering the relative contributions to nematode death of phenazine-1-carboxylic acid and 1-hydroxyphenazine produced by PA14 , we used metabolite profiling to determine the levels of phenazine-1-carboxylic acid and 1-hydroxyphenazine in PGS agar following growth of wild-type PA14 . We found that the amount of phenazine-1-carboxylic acid ( 53 µg/mL ) was greater than the level required for killing worms , whereas the amount of 1-hydroxyphenazine ( 1 . 4 µg/mL ) was insufficient to kill worms to a significant degree ( Table 1 ) . These observations suggested that phenazine-1-carboxylic acid is at least partially responsible for nematode killing under the conditions of our intoxication assay . To further investigate which phenazines are responsible for worm killing , we tested the killing abilities of PA14 mutants that are unable to synthesize pyocyanin , 1-hydroxyphenazine , or phenazine-1-carboxamide . A phzM mutant , which does not produce pyocyanin , killed worms more rapidly than did wild-type PA14 ( Figure 2C ) . A phzS mutant , which does not produce pyocyanin or 1-hydroxyphenazine [17] , [18] , [19] , did not show a significant difference in killing from wild-type . A phzH mutant , which is deficient in production of phenazine-1-carboxamide , also did not show a significant difference in killing from wild-type . The lack of production of phenazines in these mutants according to the previously described biosynthetic pathways shown in Figure 1A was confirmed by metabolite profiling ( Table 1 ) . These data , in combination with the fact that synthetic pyocyanin and phenazine-1-carboxamide do not kill worms at the concentrations they are produced under the conditions of our assay ( Figure 2B , Table 1 ) , are consistent with the conclusion that pyocyanin and phenazine-1-carboxamide do not play a significant role in fast killing . Moreover , the facts that the phzS mutant retains the ability to kill and that the level of 1-hydroxyphenazine produced by wild-type PA14 is insufficient to kill to a significant degree , are consistent with the conclusion that 1-hydroxyphenazine is either not necessary for killing or that it is one of multiple factors that cooperate to kill worms . We also observed that the phzM mutant , which kills more rapidly than wild-type PA14 , produced 34% more phenazine-1-carboxylic acid than wild-type ( 70 µg/mL vs . 53 µg/mL; Table 1 ) , which was the greatest amount of phenazine-1-carboxylic acid produced among the strains tested . Levels of phenazine-1-carboxylic acid were also slightly elevated with the phzH mutant ( 59 µg/mL ) . In contrast , the phzS mutant showed slightly depressed levels of phenazine-1-carboxylic acid ( 46 µg/mL ) . These data are consistent with the hypothesis that worm death requires phenazine-1-carboxylic acid . Furthermore , we observed no killing of nematodes on naive PGS agar that was supplemented with all four synthetic phenazines at concentrations comparable to those produced by wild-type PA14 ( data not shown ) , indicating that toxicity is not induced by the combination of phenazines in the absence of other factors . These data further support the conclusion that pyocyanin and phenazine-1-carboxamide are unlikely to play major roles in worm killing under the conditions tested . The data in Figure 2 suggested that phenazine-1-carboxylic acid is most likely the primary phenazine toxin responsible for nematode killing . Because phenazine-1-carboxylic acid killed worms when mixed with Δphz agar but not when mixed with naive agar , we reasoned that at least one additional factor provided by the bacteria is necessary for the toxicity of phenazine-1-carboxylic acid . In our attempts to identify this factor , we took into consideration precedents in the literature for the pH-dependence of the activity of bacterial toxins [20] , [21] . We found that the pH of PGS agar drops from approximately 6 prior to bacterial growth to between 4 and 4 . 5 after growth of PA14 . We tested the killing activities of phenazine-1-carboxylic acid and 1-hydroxyphenazine under different buffer conditions and found that killing by phenazine-1-carboxylic acid ( at 100 µg/mL ) was strongly pH dependent , with low pH supporting killing and neutral or higher pH preventing killing ( Figure 3 ) . In contrast , 1-hydroxyphenazine did not show pH-dependent toxicity at the same concentration . DMSO , the solvent for the phenazine solutions , did not kill worms under any of the buffer conditions tested , demonstrating that the worms are not dying only as a consequence of exposure to the low pH buffer . These observations explained why phenazine-1-carboxylic acid failed to kill worms when added to naive agar ( pH 6 ) , and considered together with the genetic and metabolite profiling data , suggest that phenazine-1-carboxylic acid is the primary toxin responsible for PA14-mediated killing of C . elegans under fast killing assay conditions . If phenazine-1-carboxylic acid is the primary toxic agent and if phenazine-1-carboxylic acid is only active at low pH , we reasoned that buffering the agar media to pH≥6 after growth of PA14 would block toxin-mediated killing activity of the PA14 agar . Although we observed a delay in killing when we raised the pH of the media to 7 with potassium phosphate , a substantial fraction of the worms died within seven hours ( Figure 4A ) . When we performed the same experiment with a phzM or phzS strain , we found that killing on the relevant time scale was abrogated ( Figure 4B & C ) . The phenotype of the phzH mutant was indistinguishable from wild-type under these conditions . Similar results were obtained when the pH was raised to 8 with Tris HCl , demonstrating that the phenomenon is not specific to potassium phosphate ( Figure S2 ) . Because both phzM and phzS are required for the synthesis of pyocyanin and phzH is not , these data suggested that pyocyanin might be toxic to worms at pH 7 or pH 8 . We tested the toxicity of pyocyanin by observing worm survival on PGS agar with synthetic pyocyanin ( 10 µg/mL ) and buffer added after growth of Δphz ( Figure 4D ) . Addition of pyocyanin caused worm death at pH 7 and pH 8 , but not at pH 6 . Moreover , the kinetics of worm death at pH 7 and 8 were similar to those observed with media on which wild-type PA14 were grown and subsequently buffered at pH 7 , with the toxic effects not evident until the 7-hour time point . Surprisingly , when we exposed worms to pyocyanin ( 10 µg/mL ) at pH 8 ( 50 mM potassium phosphate ) on PGS agar plates in the absence of bacteria , we observed no worm death within 7 hours ( data not shown ) , suggesting that there is an unknown non-phenazine product of P . aeruginosa that sensitizes C . elegans to pyocyanin . Together , these data indicate that three of four known phenazines produced by P . aeruginosa strain PA14 are toxic to C . elegans . Toxicity of the phenazines , however , varies depending on the pH of the media as well as the presence of other factors .
Pathogenesis of P . aeruginosa can occur through a variety of mechanisms including the production and excretion of toxins . In this study we identified three phenazine toxins produced by P . aeruginosa PA14 that can kill C . elegans under different environmental conditions . We showed that under the conditions of our assay phenazine-1-carboxylic acid is the predominant toxic phenazine produced by P . aeruginosa PA14 that kills C . elegans . Our work suggests that phenazine-1-carboxylic acid should be further studied as a potentially important contributor to toxin-mediated pathogenesis in other metazoan hosts besides C . elegans , including mammals . We found that the toxicity of phenazine-1-carboxylic acid to C . elegans requires an acidic environment . Acidic conditions are not uncommon , for example in wounds [22] , [23] , in the gut [24] , [25] , and intracellularly within lysosomes [23] , [26] and secretory granules [27] . Phenazine-1-carboxylic acid may act as a toxin in a variety of circumstances in those locations . The pH dependence of phenazine-1-carboxylic acid toxicity may be due to the protonation state of its carboxyl group , for which the pKa is 4 . 2 [28] . This hypothesis suggests that the uncharged acid species may be toxic and the negatively charged carboxylate benign , as has been observed for the antimicrobial activity of phenazine-1-carboxylic acid against Bacillus cereus and the fungus Gaeumannomyces graminis var . tritici [28] . Because the cytosol is buffered near neutral pH , we suspect that either the toxic effects of phenazine-1-carboxylic acid are occurring extracellularly or that the charge state affects the permeability of phenazine-1-carboxylic acid through the membrane . Neutral molecules typically traverse membranes more easily than charged molecules . After the uncharged phenazine-1-carboxylic acid species has diffused through the membrane , the neutral environment of the cytoplasm would result in its deprotonation . In its negatively charged carboxylate form it might be unable to diffuse out of the cell , resulting in its toxic accumulation within the cell . Phenazine-1-carboxamide , which differs from phenazine-1-carboxylic acid by only an amide group in place of the carboxyl group , is not toxic to C . elegans in our assay . The amide and the carboxyl groups should both be uncharged at acidic pH where phenazine-1-carboxylic acid is toxic and phenazine-1-carboxamide is not . Although it is possible that the chemical difference between these phenazines could affect their behavior due to other properties than their difference in pKa , the difference in toxicity of these two species suggests that charge state is not the sole determinant of phenazine toxicity . The zwitterionic species of pyocyanin is the predominant form at pH 7 and 8 , and the net neutral charge may facilitate its traversal through the membrane as well [15] . The pKa of pyocyanin is 4 . 9 [29] , which does not explain why pyocyanin is ineffective at killing nematodes at pH 6 , where the zwitterionic form should still predominate . However , as shown in Figure 4 , unlike the killing activities of phenazine-1-carboxylic acid and 1-hydroxyphenazine , the killing activity of pyocyanin appears to be dependent on a non-phenazine factor produced by P . aeruginosa . It is possible that the oxidation state of pyocyanin is altered by components of P . aeruginosa exudates on PGS agar . Phenazines , including pyocyanin , exert toxic effects in a variety of mammalian tissues through the generation of reactive oxygen species [8] , [12] , [13] , [15] , [30] , [31] . Oxidized and reduced phenazines also have different hydrophobicities [12] , which can affect their ability to permeate membranes or remain in aqueous solution . The oxidation state of pyocyanin has been shown to change as a function of cell density in PA14 liquid culture [32] , and changes in pH also affect its redox potential [33] . The difference in pyocyanin toxicity to nematodes in the presence and absence of PA14 exudates could be due to differences in the oxidation states of pyocyanin . It is also possible that the pH dependence of pyocyanin-mediated killing is due to pH-dependent activation of the accessory factor ( s ) and not of pyocyanin itself . Given the importance of pyocyanin as a toxin in a variety of systems , it would be valuable to identify these accessory factors and determine if they play a role in pyocyanin toxicity in other hosts . Studies have demonstrated that 5-methyl-phenazine-1-carboxylic acid ( 5MPCA ) , another phenazine produced by P . aeruginosa , is toxic to the fungus C . albicans [18] , [19] . 5MPCA is an intermediate in pyocyanin synthesis that is produced by PhzM acting on phenazine-1-carboxylic acid . Given that the phzM strain is the most toxic in our assay , and that the phzS strain , which results in the accumulation of 5MPCA , is no more toxic than wild-type , we think it is unlikely that 5MPCA produced by PA14 is a toxic species to C . elegans under the conditions of our assay . CF patients are highly susceptible to lung infections by P . aeruginosa [1] , [2] . CF results in abnormal hyperacidification of organelles of epithelial cells in the respiratory pathway as well as increased acidity of airway surface liquid as compared with healthy individuals [34] . Hyperacidification of the lung airway surface liquid reduces its antimicrobial effects in a porcine CF model [35] and has been speculated to have other effects on the CF lung , including tissue damage , inflammation , and thickening of the mucus [34] . Increased acidity could also lead to enhanced toxicity of phenazine-1-carboxylic acid in the P . aeruginosa infected CF lung . The concentrations of pyocyanin we determined that are active against nematodes are within the concentrations observed in human CF sputum and that have been shown to have physiological activity [36] , [37] , [38] , [39] , [40] , [41] . In wild-type PA14 grown on PGS agar , we detected 2 . 38 µg/mL ( 11 . 3 µM ) pyocyanin . Pyocyanin has been detected at concentrations of up to 27 . 3 µg/mL ( 130 µM ) in the sol phase of CF sputum [37] . Interestingly , a recent study showed that pyocyanin concentrations in sputum from CF patients correlated with the degree of severity of the disease , ranging from 7 . 7 µM in unobstructed airways to 46 . 8 µM in severely obstructed airways [36] . Phenazine-1-carboxylic acid concentrations also correlated with impairment of lung function in CF , with concentrations ranging from 5 . 2 µM in unobstructed airways to 39 . 9 µM in severely obstructed airways [36] . The concentrations we detected in PGS agar were considerably higher , at 52 . 7 µg/mL ( 235 µM ) , but we demonstrated that phenazine-1-carboxylic acid is toxic to nematodes at concentrations as low as 16 µg/mL ( 71 . 4 µM ) . Although this is still higher than concentrations observed in CF lungs , it is possible that worms are intrinsically more resistant to phenazine-1-carboxylic acid than mammalian cells . Pseudomonas infections are also common in wounds , which exhibit a variety of different pH conditions [22] , [23] . During inflammation , the pH of acute wounds falls below 6 , a level at which phenazine-1-carboxylic acid could exert its toxic effects . During early stages of healing , the pH increases to between 7 and 8 , where phenazine-1-carboxylic acid might be ineffective as a toxin . In that pH range pyocyanin could act as a toxin . Indeed , pyocyanin has been shown to inhibit wound repair and induce cellular senescence in an in vitro model of wound healing [41] . In agreement with previous observations [8] , [9] , we observed that L4 worms were more susceptible to killing by phenazines than adult animals . We also noticed that older L4 worms were killed more quickly than younger L4 worms ( B . Cezairliyan and R . Feinbaum , unpublished observations ) . We suspect that proximity to the L4 to adult molt plays a role in the susceptibility of L4 worms . Perhaps worms are most susceptible when they are first exposed to phenazines during the molt , when phenazines might be able to most easily due to the shedding of the cuticle . Earlier exposure to phenazines might allow worms time to detoxify them before reaching their most susceptible stage . It is also possible that specific gene expression patterns unrelated to toxin permeability during the L4 to adult molt are the cause of increased susceptibility . Identification of genes that reduce susceptibility to phenazines at this transition or that alter the developmental stage at which worms are susceptible to phenazines should help to elucidate the cause of the stage-dependent susceptibility of C . elegans to phenazines . An understanding of the regulation of toxin production in P . aeruginosa is critical for understanding its interactions with its hosts . Phenazine biosynthesis in P . aeruginosa is dependent on a variety of environmental factors in the media and is also largely dependent on cell density via quorum sensing signals [12] , [13] , [42] , [43] , [44] , [45] , [46] . In P . aeruginosa , the genes that are responsible for producing phenazine-1-carboxylic acid are present in duplicate in two differently-regulated operons in the genome . The genes that regulate production of other phenazines , however , are transcribed independently of one another and of the phzA1-G1 and phzA2-G2 operons , thus allowing for transcriptional regulation of the production of particular phenazines depending on the environmental stimuli [17] , [46] , [47] . Many phenazine-producing bacteria make more than one type of phenazine [11] . pH has been shown to influence the production of phenazine-1-carboxylic acid in Pseudomonas fluorescens and phenazine-1-carboxamide in Pseudomonas chlororaphis [48] , [49] . Moreover , P . aeruginosa grown under different conditions can produce other low molecular weight toxins that are lethal to C . elegans , including cyanide [5] , [50] and quinolone [4] . Our work suggests that one of the purposes that could be served by the diversity of phenazines is to have an armamentarium of molecules available that are active under different environmental conditions . Knowledge of the toxins that are effective in our nematode killing assay will allow for a better understanding of phenazine toxicity in C . elegans and in other hosts . Although the mutants identified in the screen for toxin-killing deficiency in PA14 were also reduced in pathogenesis in a mouse thermal injury model and an Arabidopsis leaf infiltration model [8] , it is still unclear how the nematode intoxication assay serves as a proxy for the mammalian and plant models , which are based on infection and colonization by live bacteria . It will be important to determine if the same phenazine toxins that we have identified in the killing of C . elegans play a role in the murine or plant models . If the toxicity in mammals is similarly based on phenazines , C . elegans will continue to serve as a useful tool in the identification of pathogenesis mechanisms of P . aeruginosa in mammals and plants . Host pathways in C . elegans can be probed genetically to identify mechanisms by which phenazine-1-carboxylic acid acts to kill the worm and whether the three toxic phenazines exert their lethal effects in similar ways in the host . A combination of proteomics , gene expression , and mutant analysis should help to shed light on toxicity mechanisms of phenazines in the host and the ways by which their production is regulated in the pathogen .
Worms were maintained on lawns of E . coli OP50 on nematode growth medium agar ( NGM ) plates prior to killing assays . Killing assays were performed on PGS agar ( 1% bacto-peptone , 1% glucose , 1% NaCl , 150 mM sorbitol , 1 . 7% bacto-agar ) as described [8] . 1-hydroxyphenazine was purchased from TCI America ( H0289 ) , pyocyanin from Cayman Chemical Company ( 10009594 ) , phenazine-1-carboxylic acid from Apollo Scientific ( OR01490 ) , and phenazine-1-carboxamide from Princeton Biomolecular Research ( PBMR030086 ) . All phenazine stocks were dissolved in DMSO . The PA14 Δphz mutant strain lacking both the phzA1-G1 and phzA2-G2 operons was kindly provided by Lars Dietrich and Dianne Newman [16] . Other PA14 mutant strains were obtained from a non-redundant transposon insertion library [51]: phzM ( mutant ID 40343 ) , phzH ( 39981 ) , phzS ( 44099 ) . All transposon insertion strains were sequenced to confirm the location of the transposon . The phzM and phzS genes are on opposite sides of the phzA1-G1 operon . phzM and phzS are coregulated to some degree with phzA1-G1 , but they are part of separate transcriptional units not containing other genes [17] , [46] . Thus , the transposons in phzM and phzS are unlikely to exert polar effects . The phzH gene is elsewhere in the genome and is not known to be cotranscribed with other genes [17] , [31] . C . elegans strain Bristol N2 [52] was used for killing assays . Killing assays were performed as previously described [8] . In order to have fourth larval stage ( L4 ) worms for killing assays , 10 gravid worms were picked to a 60 mm plate with NGM agar and a lawn of E . coli OP50 as a food source . Gravid worms were kept on plates for 16 hours at 15°C , after which they were removed . Plates were returned to 15°C for 8 hours , after which they were transferred to 20°C . Eggs hatched and grew to L4 stage approximately 36 hours after transfer . Proper staging of L4 worms was critical to the reproducibility of the assay , as worms that were younger or older were killed less quickly than L4 stage worms . Killing agar plates were prepared by spreading 5 µL of overnight culture of PA14 in LB on a 35 mm petri plate containing 4 mL of PGS agar ( 1% bacto-peptone , 1% glucose , 1% NaCl , 150 mM sorbitol , 1 . 7% bacto-agar ) . Plates were incubated for 24 hours at 37°C and then transferred to 23°C for 24 hours . L4 stage worms were put on the plates , which remained at room temperature until the completion of the assay . Worms were scored as live or dead based on movement elicited by tapping their heads gently with a thin wire . To mix the agars of plates seeded with different PA14 strains , bacteria were scraped off the surface of the agar with a cell scraper after which the agar was melted by heating in a microwave . The hot agars were mixed , repoured into plates , and allowed to cool . In experiments where phenazines or buffer were added , plate agar was melted , concentrated buffer and/or phenazine stock solution ( in DMSO ) was added and mixed , after which plates were repoured and allowed to cool . In order to determine absolute amounts of phenazines produced by different strains , a standard curve was constructed for each of the phenazines at concentrations of 1 , 4 , 16 , 32 , and 64 µg/mL . Killing agar plates seeded with Δphz were melted and doped with a range of concentrations of synthetic phenazines . Cool , dry plates were extracted with chloroform/methanol ( CHCl3∶MeOH∶agar = 2∶1∶1 ) . The organic layer was transferred to another vial and concentrated under a stream of nitrogen . Samples were dissolved in 200 µL of CHCl3 and stored at −80°C . Prior to LC-MS analysis , reconstituted samples were diluted 200 fold in CHCl3 and spiked with cytosporone B ( csn-B ) to the final concentration of 5 µM as a standard for normalization across samples . To obtain the most accurate concentrations of phenazines , samples from different strains were prepared in the same manner and at the same time as the standards . Three experimental replicates were used for each standard and bacterial sample . LC-MS analysis was performed using an Agilent 6410 LC-ESI-QQQ instrument in multiple reaction monitoring ( MRM ) mode . Samples were analyzed both in the negative and positive ion modes as previously described [53] . For the LC analysis in the negative ion mode , a Gemini ( Phenomenex ) C18 column ( 5 µm , 4 . 6 mm×50 mm ) was used together with a precolumn ( C18 , 3 . 5 µm , 2 mm×20 mm ) . Mobile phase A consisted of a 95/5 water/methanol mixture , and mobile phase B was made up of 60/35/5 isopropanol/methanol/water . Both A and B were supplemented with 0 . 1% ammonium hydroxide as solvent modifiers . For the LC analysis in the positive ion mode , a Luna ( Phenomenex ) C5 column ( 5 µm , 4 . 6 mm×50 mm ) was used together with a precolumn ( C4 , 3 . 5 µm , 2 mm×20 mm ) . Mobile phases A and B , as well as the gradient , were the same as that used for the negative ion mode analysis , except that , in this case , both A and B were supplemented with 0 . 1% formic acid and 5 mM ammonium formate as solvent modifiers . The gradient started at 0% B for 3 min at 0 . 2 mL/min and then linearly increased to 100% B over the course of 8 min at 0 . 5 mL/min , followed by an isocratic gradient of 100% B for 4 min at 0 . 5 mL/min before equilibration for 5 min at 0 . 5 mL/min . The total analysis time , including 3 min at 0 . 2 mL/min , was 20 min . MS analysis was performed with an electrospray ionization ( ESI ) source . The capillary voltage was set at 4000 V . The drying gas temperature was 350°C , the drying gas flow rate was 10 L/min , and the nebulizer pressure was 45 psi . For both ionization conditions , data was collected in the profile mode . Conditions for MRM experiment were optimized using the program called Optimizer ( Agilent ) . The resulting optimized conditions for each transition used to quantitate each phenazines are listed in Supplementary Table S1 . Each run was performed using a 1 µL injection of extract . For data analysis , ions corresponding to each phenazine and csn-B were extracted from the total ion chromatograms to obtain integrated mass ion intensities ( peak area; MSII ) for each ion . The levels of phenazines were normalized by dividing MSII of phenazines by MSII of csn-B detected in each chromatogram to give normalized MSII ( nMSII ) . nMSII of the standard samples was applied to construct standard curves . Finally , phenazine concentrations in the experimental samples were calculated according to the standard curves . | The bacterium Pseudomonas aeruginosa is a pathogen of a wide variety of organisms . It has been shown that P . aeruginosa factors that are critical for its toxicity to the nematode worm Caenorhabditis elegans are also important for its pathogenicity in mammals . In this report we show that phenazines , a class of small molecules produced by P . aeruginosa , act as lethal toxins against the worm . Under conditions relevant to mammalian pathogenesis , we identified one phenazine , phenazine-1-carboxylic acid , that is primarily responsible for killing worms . We found that the toxicity of this phenazine and one other phenazine , pyocyanin , are dependent on the pH of the media . We also identified a third toxic phenazine , 1-hydroxyphenazine , whose toxicity is not dependent on pH . These results show that the diversity of toxic molecules produced and released by P . aeruginosa may serve the bacterium to facilitate pathogenicity in a variety of environments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"bacteriology",
"biochemistry",
"gram",
"negative",
"microbial",
"pathogens",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"toxic",
"agents",
"bacterial",
"biochemistry",
"bacterial",
"pathogens",
"toxicology",
"pathogenesis"
]
| 2013 | Identification of Pseudomonas aeruginosa Phenazines that Kill Caenorhabditis elegans |
By applying REMD simulations we have performed comparative analysis of the conformational ensembles of amino-truncated Aβ10-40 peptide produced with five force fields , which combine four protein parameterizations ( CHARMM36 , CHARMM22* , CHARMM22/cmap , and OPLS-AA ) and two water models ( standard and modified TIP3P ) . Aβ10-40 conformations were analyzed by computing secondary structure , backbone fluctuations , tertiary interactions , and radius of gyration . We have also calculated Aβ10-40 3JHNHα-coupling and RDC constants and compared them with their experimental counterparts obtained for the full-length Aβ1-40 peptide . Our study led us to several conclusions . First , all force fields predict that Aβ adopts unfolded structure dominated by turn and random coil conformations . Second , specific TIP3P water model does not dramatically affect secondary or tertiary Aβ10-40 structure , albeit standard TIP3P model favors slightly more compact states . Third , although the secondary structures observed in CHARMM36 and CHARMM22/cmap simulations are qualitatively similar , their tertiary interactions show little consistency . Fourth , two force fields , OPLS-AA and CHARMM22* have unique features setting them apart from CHARMM36 or CHARMM22/cmap . OPLS-AA reveals moderate β-structure propensity coupled with extensive , but weak long-range tertiary interactions leading to Aβ collapsed conformations . CHARMM22* exhibits moderate helix propensity and generates multiple exceptionally stable long- and short-range interactions . Our investigation suggests that among all force fields CHARMM22* differs the most from CHARMM36 . Fifth , the analysis of 3JHNHα-coupling and RDC constants based on CHARMM36 force field with standard TIP3P model led us to an unexpected finding that in silico Aβ10-40 and experimental Aβ1-40 constants are generally in better agreement than these quantities computed and measured for identical peptides , such as Aβ1-40 or Aβ1-42 . This observation suggests that the differences in the conformational ensembles of Aβ10-40 and Aβ1-40 are small and the former can be used as proxy of the full-length peptide . Based on this argument , we concluded that CHARMM36 force field with standard TIP3P model produces the most accurate representation of Aβ10-40 conformational ensemble .
Aβ peptides linked to the development of Alzheimer’s disease ( AD ) are produced , through a normal cellular proteolysis , in a variety of alloforms , which differ with respect to sequence length and the extent of amino- or C-terminal truncation [1–3] . The most abundant form is a 40-residue version Aβ1-40 , which constitutes about 90% of all Aβ species in cerebrospinal fluid [4] . Virtually all Aβ peptides are highly amyloidogenic [5 , 6] and play a central role in amyloid cascade hypothesis , which explains AD pathogenesis on the basis of multi-stage aggregation of Aβ species . In this process , Aβ monomers represent initial species involved in spontaneous aggregation . Moreover , according to experimental studies fibril elongation is also largely driven by deposition of Aβ monomers to the fibril edges [7] . Generally , Aβ peptides display a high level of cytotoxicity [2 , 8–10] , which is related to their ability to readily bind to cellular lipid bilayers and disrupt their structure [11] . Although the mechanism of binding to lipid bilayers is likely to be concentration dependent , Aβ peptides predominantly bind as monomers rather than oligomers at nanomolar concentrations [12 , 13] . Aβ peptides belong to the class of intrinsically disordered proteins implying that they lack well defined native structure in aqueous environment . Indeed , experimental investigations , including solution NMR studies , have shown that the conformational ensemble of Aβ monomer in water is populated by heterogeneous coil-like conformations [14–17] . More recently , several NMR measurements , including chemical shifts , nuclear Overhauser effects , and J-couplings , have been used to confirm that Aβ peptides adopt generally random coil structures at neutral pH [18] . At the same time , careful analysis of electron paramagnetic resonance studies revealed that Aβ monomers still contain short structured regions ( His14-Val18 , Gly29-Ala30 , and Gly38-Val40 ) under normal physiological conditions [19] . Similarly , several NMR studies have pointed to a formation of a turn or bend structures in the sequence region ( Phe20-Ser26 ) between the central hydrophobic cluster ( Leu17-Ala21 ) and the C-terminal ( Ala30-Val40 ) [15 , 17] . Due to generally disordered state of Aβ in water , it is not surprising that Aβ conformations are highly dependent on solvent properties . For example , in the membrane-like environments Aβ conformational ensemble undergoes considerable reorganization manifested in the formation of helical structure in the sequence regions Glu15-Val24 and Gly29-Met35 [20–22] . Moreover , Aβ helix propensity is pH dependent illustrated by the observation that the Glu15-Val24 helix , but not the C-terminal helix , becomes destabilized at normal pH . Conformational plasticity of Aβ peptides is also consistent with mutagenesis studies . For example , Iowa ( D23N ) or Osaka deletion ( E22Δ ) mutants aggregate significantly faster than the wild-type [23 , 24] . Similarly , according to in vitro studies many single-point mutations mainly affecting hydrophobic or charged residues can either accelerate or reduce Aβ aggregation propensity [25 , 26] . One may expect that Aβ monomeric conformations , being the initial species involved in aggregation , are impacted by these single-point mutations . A survey of experimental findings presented above suggests that computational characterization of the conformational ensemble formed by Aβ monomers is important for understanding its aggregation and cytotoxicity . Several previous molecular dynamics studies have probed Aβ monomers in water . Garcia and coworkers have studied the conformations of Aβ1-40 and Aβ1-42 peptides using replica exchange molecular dynamics ( REMD ) simulations extended to microsecond timescales [16 , 17 , 27] . Consistent with the experiments , their analysis revealed generally disordered Aβ conformational ensemble augmented by several structured regions , especially in the C-terminal of Aβ1-42 , where a β-hairpin has been detected . Qualitatively similar conclusions have been reached in the recent REMD study conducted by Head-Gordon and coworkers [28] . In our previous studies , we have utilized REMD and all-atom CHARMM22 force field with CMAP corrections to investigate the conformations of amino-truncated Aβ10-40 peptide in water [29] . We found that , similar to the full-length peptide , Aβ10-40 samples predominantly turn and random coil structures , whereas helical and especially β-sheet propensities are low . Furthermore , the peptide almost completely lacks tertiary structure with most stable intrapeptide interactions forming between the amino acids adjacent along the sequence . In light of disordered state of Aβ monomer it is important to test the dependence of its conformational ensemble on the force fields employed in the simulations . Recently , such investigation has been carried out for Aβ1-40 monomer , which was probed using OPLS-AA/L , AMBERff99sb-ILDN , and CHARMM22* protein force fields and several water models [30] . All the three force fields predicted the formation of β-structure in the Leu17-Ala21 and Ala30-Leu34 regions , but with markedly different β propensities . In particular , OPLS-AA/L and AMBERff99sb-ILDN simulations revealed stable β-structures , whereas suppressed β fraction has been observed in the CHARMM22* force field . It is conceivable that the three force fields overestimate the β propensity in Aβ monomer , which is expected to be low according to the NMR studies [18] . Indeed , REMD study of two natively unfolded peptides , NTL9 ( 1–22 ) and NTL9 ( 6–17 ) , using AMBERff99sb-ILDN , CHARMM22/CMAP , and CHARMM36 force fields [31] has shown that both CHARMM force fields predict much smaller β fraction than AMBERff99sb-ILDN . Additionally , the parameterization of water may affect peptide conformational ensembles . This point has been recently demonstrated using REMD and CHARMM36 for two Ala-rich peptides and GB1 peptide , which form considerably more solvated and extended structures with modified TIP3P water model compared to its standard version [32] . Because previous investigations have emphasized the importance of force field parameterization , a natural question arises about how a force field affects sampling of amino-truncated Aβ10-40 peptide , which was extensively studied by us in the context of peptide-lipid bilayer interactions [33–36] . To address this question , we used all-atom REMD simulations and performed a systematic comparison of Aβ10-40 conformational ensembles in four protein force fields ( CHARMM36 , CHARMM22* , CHARMM22/CMAP , OPLS-AA ) and two water models ( standard and modified TIP3P ) . We show that although all force fields are consistent in predicting the Aβ propensity to form turn and random coil structures , they strongly disagree on the extent and distribution of tertiary interactions or helix and β propensities in Aβ monomer . We have also compared the J-coupling and residual dipolar coupling ( RDC ) constants computed from Aβ10-40 simulations with Aβ1-40 experimental data . Surprisingly , we found that in silico Aβ10-40 conformational ensemble produced by CHARMM36 force field agrees better with the experimental measurements than in silico Aβ1-40 ensemble simulated earlier [17 , 27] . Based on these comparisons we suggest that Aβ10-40 can serve as a proxy of the full-length Aβ1-40 peptide and the CHARMM36 force field with standard TIP3P water model provides most accurate reproduction of Aβ conformational ensemble .
To explore the impact of force field parameterizations , we have performed molecular dynamics ( MD ) simulations of Aβ10-40 peptide , which is an amino-truncated fragment of the full-length peptide Aβ1-40 ( Fig 1a ) . In all , we have investigated four all-atom protein force fields and two explicit water models [37 , 38] resulting in five simulation systems , which utilized CHARMM36 [39] with modified TIP3P water model ( denoted as C36 ) , CHARMM36 with standard TIP3P water model ( C36s ) , CHARMM22* with modified TIP3P water model ( C22* ) [40] , CHARMM22 with CMAP corrections and modified TIP3P water model ( C22cmap ) [41] , and OPLS-AA with modified TIP3P water model ( OPLS-AA ) [42] . The C22cmap system was already studied by us previously [29 , 43 , 44] and is used here to expand the force fields comparison . All simulation systems contained a single Aβ10-40 peptide , 4959 water molecules , and one sodium ion to set the net system charge to zero . In all , the simulation systems contained 15 , 354 atoms . Aβ termini were capped with acetylated and aminated groups . The charged states of amino acids corresponded to neutral pH ( in particular , histidines were deprotonated ) . For all force fields we used periodic boundary conditions with the cubic unit cell having the edge dimension of about 53 . 8 Å and resulting in the mass density of 0 . 9848 g/cm3 at 330 K . Non-bonded interactions were computed using a smooth switching functions acting within the interval of 8 and 12 Å . Electrostatic interactions were computed using particle mesh Ewald summation with the grid size of ≈ 1Å . All hydrogen associated covalent bonds except in water molecules were treated as rigid by applying ShakeH algorithm . Water molecules were treated as rigid using SETTLE algorithm . All simulations used the integration step of 1 fs . Full electrostatic evaluation frequency was set to 4 integration steps . To produce exhaustive sampling of Aβ10-40 conformational ensembles we have utilized canonical ( NVT ) replica exchange molecular dynamics ( REMD ) [45] , which is implemented in NAMD MD program [46] . For all systems we used R = 40 replicas distributed exponentially in the temperature range from 300 to 440 K . Canonical ensembles in the replicas were generated by applying underdamped Langevin dynamics with the damping coefficient γ = 5 ps−1 . Replica exchanges were attempted every 2 ps between all neighboring replicas along a temperature scale resulting in the average acceptance rates ranging between 27 and 29% . For each simulation system we have produced four REMD trajectories collecting in total 3 . 2 μs of sampling ( or 80 ns per replica ) . Each REMD trajectory has been initiated with random initial conformations equilibrated at 330 K with preliminary isothermal-isobaric simulations to set correct mass densities . In addition , each replica was equilibrated at its own temperature for an additional 1 ns . With this approach no sampling data from REMD trajectories needed to be discarded as non-equilibrated . Although probing REMD convergence is generally a difficult task ( see , e . g . , [47] ) , our analysis in S1 Text suggests that the resulting simulation times appear sufficient for sampling convergence . Secondary structures in Aβ were assigned using the STRIDE program [48] . A helical state includes α- , 310- , or π-helix conformations , whereas a β-strand state includes extended conformations or isolated bridges . Tertiary interactions were probed by side chain contacts . A contact occurs if the distance between the geometric centers of heavy atoms in two side chains is less than 6 . 5 Å . This cutoff approximately corresponds to the onset of hydration of side chains as their separation increases . We classified the contacts between residues i and j as long-range if |j − i| ≥ 5 or short-range otherwise . To explore the fluctuations in peptide backbone , we computed the standard deviations δϕ ( i ) and δψ ( i ) for backbone dihedral angles ϕ and ψ of a residue i . These standard deviations are referred to as root-mean-square fluctuations ( RMSF ) . To evaluate peptide dimensions we computed the radius of gyration , Rg , using the positions of side chain centers of mass and Cα atoms . To quantitatively evaluate the consistency between the experimental and computational conformational ensembles we have utilized two quantities . The first is the 3JHNHα coupling constants associated with three-bond coupling interaction between HN and Hα protons . These constants are sensitive to peptide’s secondary structure [49] . In this study we used three sets of experimentally measured J-coupling constants , Jexp , for Aβ1-40 peptide [16–18] . In silico J-coupling constants , Jcomp , were determined from the backbone dihedral angles using Karplus equation [50] J c o m p = A · c o s 2 ( ϕ - 60 ° ) + B · c o s ( ϕ - 60 ° ) + C , ( 1 ) where ϕ is a backbone dihedral angle and A , B , and C are the coefficients determined by fitting with the experimental data . We used three sets of coefficients reported by Pardi et al ( A = 6 . 4 , B = −1 . 4 , and C = 1 . 9 ) [51] , Brueschweiler et al ( A = 9 . 5 , B = −1 . 4 , and C = 0 . 3 ) [52] , and Vuister et al ( A = 6 . 51 , B = −1 . 76 , and C = 1 . 60 ) [53] . The N-terminal amino acid was excluded from our computation of J-couplings as its ϕ angle may be distorted by the capping group . As a second quantity we chose residual dipolar coupling ( RDC ) constants experimentally measured for Aβ1-40 peptide [54] . RDC measurements probe orientation of amide NH bonds in protein backbones with respect to external magnetic field , and can identify long-range structural correlations across biomolecular structure . To compute RDC constants from in silico structures we used the Prediction of ALignmEnt from Structure ( PALES ) program [55] . We processed our simulation structures by using default PALES settings , including the application of steric interaction model , which determines alignment orientation based on steric properties of a molecule . As described in S1 Text we applied global alignment of Aβ10-40 structures for RDC computations . Once the RDC constants were produced , they were multiplied by the scaling factors determined from the least-squares fitting that minimizes the deviation between the experimental and in silico data . To assess the similarity between experimental and computational quantities , we used the Pearson’s correlation coefficient ( PCC ) , root mean square deviations ( RMSD ) , and quality factor Q defined as Q = ∑ k ( D c o m p , k - D e x p , k ) 2 / ∑ k D e x p , k 2 , where Dexp , k and Dcomp , k are the experimental measurements and their computed counterparts available for amino acids k . The sum is taken over all amino acids for which experimental and computed values are simultaneously available . We applied Q to evaluate agreement for J-coupling or RDC constants . To preserve consistency with our previous studies all ensemble averages were computed using weighted histogram analysis method ( WHAM ) [56] of REMD data at 330K . J-coupling and RDC constants were computed at 300K , which is the closest simulation temperature to experimental conditions ( ≈ 280K ) [16–18] .
We begin the comparison of different force fields with the detailed analysis of the conformational ensemble of Aβ10-40 monomer in CHARMM36 force field with modified TIP3P water model ( denoted as C36 ) . We first focus on the Aβ10-40 secondary structure . It follows from Fig 2 and Table 1 that its conformational ensemble is dominated by random coil ( 〈RC〉 = 0 . 49 ± 0 . 04 ) and turn ( 〈T〉 = 0 . 44 ± 0 . 03 ) , which together account for 93% of all amino acid states . Indeed , the turn structure is stable ( i . e . , its fraction 〈T ( i ) 〉 > 0 . 5 ) for few S1 residues ( His13 , His14 ) and within the long sequence span Phe19-Gly29 covering the part of hydrophobic region S2 , the entire hydrophilic S3 , and the beginning of C-terminal S4 . Random coil structure dominates the N- and C-terminals and the Gln15-Val18 region . The occurrence of helix and β-structure is negligible ( ≤0 . 04 ) . In fact , Fig 3 shows that the helical propensity 〈H ( i ) 〉 is weak throughout Aβ10-40 sequence being always ≲ 0 . 1 . To explore the structural fluctuations in Aβ10-40 backbone we have computed the root-mean-square fluctuations ( RMSF ) , δϕ ( i ) and δψ ( i ) , in backbone dihedral angles ϕ and ψ of amino acids i . Fig 4a shows that apart from enhanced fluctuations of Gly ψ angles the distributions δϕ ( i ) and δψ ( i ) are fairly uniform throughout Aβ sequence . The most rigid backbone conformation is observed for hydrophobic Phe20 . The values of δϕ and δψ averaged over the entire Aβ10-40 sequence are 55 . 2 ± 1 . 6° and 81 . 9 ± 3 . 8° , respectively . Relatively uniform distribution of the backbone fluctuations in Aβ sequence is consistent with the lack of well-defined secondary structure , such as helices or β-strands . We next investigate the Aβ10-40 tertiary structure by probing the formation of intrapeptide interactions . Fig 5a presents the peptide contact map 〈C ( i , j ) 〉 , which visualizes the probabilities of forming contacts between amino acid side chains . This figure suggests that there are very few stable interactions in Aβ peptide , i . e . , those occurring with the probability 〈C ( i , j ) 〉 > 0 . 35 . According to Table 2 there are no stable long-range ( |i − j| ≥ 5 ) contacts , and there are only two stable short-range ( |i − j| < 5 ) contacts , namely , Leu17-Phe19 and Asp23-Ser26 . The Leu17-Phe19 contact is likely to explain a rigid backbone conformation at Phe20 . The probability of forming a salt bridge Asp23-Lys28 , which is important for amyloid fibril assembly , is low being equal to 〈C ( 23 , 28 ) 〉 = 0 . 10 ± 0 . 03 . However , weak electrostatic interactions are formed between Glu22 and Lys28 ( 0 . 21 ± 0 . 13 ) and between Glu11 and Lys16 ( 0 . 20 ± 0 . 11 ) . Overall , the average number of all side chain contacts forming in Aβ monomer is 〈C〉 = 21 . 1 ± 1 . 4 , of which 10 . 7 ± 0 . 9 ( or 51% ) are long-range . Finally , we consider the probability distribution P ( Rg ) of Aβ10-40 radius of gyration Rg . Fig 6 shows that for C36 P ( Rg ) is broad with the maximum at Rg ≈ 15 Å . The equilibrium value of Rg , 〈Rg〉 , is 16 . 9 ± 0 . 5 Å , which is the largest among all force field considered ( see below ) . In addition , we have computed the average end-to-end distance 〈R1N〉 = 24 . 6 ± 1 . 2Å , which is also the largest among other force fields . Thus , Aβ10-40 peptide in C36 force field lacks stable secondary or tertiary structure and forms expanded conformations as illustrated in Fig 1b . To check the impact of water model , we repeated CHARMM36 REMD simulations of Aβ10-40 peptide using standard TIP3P water ( denoted as C36s ) . Following the analysis for C36 we have computed Aβ secondary and tertiary structure . Fig 2 and Table 1 demonstrate that in close agreement to C36 simulations Aβ10-40 primarily samples random coil ( 〈RC〉 = 0 . 47 ± 0 . 04 ) or turn ( 〈T〉 = 0 . 45 ± 0 . 03 ) conformations , which together represent 92% of all residue states . In contrast , helix or β-state occur rarely . Fig 3 further reveals that the residue-specific turn 〈T ( i ) 〉 and helix 〈H ( i ) 〉 propensities are almost identical to those observed in C36 . Similar to C36 , stable turn structure is present at His13-His14 and in the region Phe19-Gly29 . Consequently , the root-mean-square deviations ( RMSD ) between 〈T ( i ) 〉 and helix 〈H ( i ) 〉 distributions computed from C36s and C36 simulations are low ( 0 . 02 and 0 . 03 ) . Random coil distributions between C36s and C36 also nearly match . Furthermore , according to Fig 4a and 4b a very close agreement is observed in the RMSF distributions , δϕ ( i ) and δψ ( i ) , as confirmed by the low respective RMSD values ( 3 . 9 and 3 . 2° ) . As expected the average values 〈δϕ〉 and 〈δψ〉 computed using all amino acids are 54 . 6 ± 1 . 9° and 81 . 1 ± 5 . 0° , which are nearly identical to those of C36 . To analyze Aβ tertiary structure we have computed the contact map 〈C ( i , j ) 〉 , which is presented in Fig 5b . In general , C36s tertiary interactions are very similar to those seen for C36 . It follows from Table 3 that Aβ peptide has no stable long-range contacts and the same two stable short-range contacts as in C36 are formed . There are four common contacts among top five long-range interactions in C36s and C36 simulations ( Ala21-Ser26 , Glu22-Lys28 , Glu11-Lys16 , Val24-Gly29 ) , whereas three short-range contacts are shared between the two simulations ( Leu17-Phe19 , Asp23-Ser26 , Gly25-Asn27 ) . As in C36 the probability of forming the salt bridge Asp23-Lys28 is low ( 0 . 14 ± 0 . 07 ) . Overall , the RMSD computed between C36 and C36 contact maps is small ( 0 . 02 ) confirming their similarity . It is noteworthy , however , that in C36s simulations the peptide forms , on an average , more side chain contacts than in C36 , as their number reaches 〈C〉 = 23 . 5 ± 1 . 4 , of which 12 . 8 ± 0 . 8 ( or 54% ) are classified as long-range . Furthermore , comparison of the probability distributions P ( Rg ) for Aβ10-40 radius of gyration in Fig 6 reveals that the C36s distribution is systematically shifted to smaller Rg values suggesting that the C36s peptide is more compact . Indeed , from Fig 6 we determine that the equilibrium 〈Rg〉 is 15 . 9 ± 0 . 4Å , which is smaller than the C36 value . Thus , although the secondary and tertiary structure propensities in C36s and C36 simulations are in close agreement , the former generates more compact structures with larger number of intrapeptide interactions ( Fig 1c ) . CHARMM22* ( C22* ) is another version of CHARMM force field , which was developed to address conformational biases in CHARMM22cmap [40] . We performed REMD simulations of Aβ10-40 monomer using C22* and studied its conformational ensemble . As demonstrated by Fig 2 and Table 1 the peptide forms predominantly turn conformations ( 〈T〉 = 0 . 52 ± 0 . 01 ) , whereas the fraction of random coil ( 〈RC〉 = 0 . 30 ± 0 . 02 ) , while still significant , is reduced compared to C36 and C36s . A distinctive feature of C22* simulations is an elevated fraction of helical structure , which reaches 0 . 18 ± 0 . 03 ( a more than four-fold increase compared to C36 ) . Occurrence of β-states is negligible . Fig 3 further underscores differences in the secondary structure distributions at a residue level . The C22* turn fraction 〈T ( i ) 〉 is noticeably higher than the C36 turn propensities within hydrophobic S2 and hydrophilic S3 regions ( Phe19-Gly25 ) , where it exceeds 0 . 8 . Random coil occurs near the peptide terminals and within Lys16-Val18 . More importantly , C22* leads to formation of a marginally stable helix structure , particularly , in the hydrophobic C-terminal ( S4 region ) . Indeed , according to Fig 3a within the sequence interval Asn27-Val36 the helix fraction 〈H ( i ) 〉 approaches 0 . 4 in sharp contrast to C36 propensities . Consequently , the differences between turn and helix distributions observed in C22* and C36 simulations are reflected in high RMSD values , which are 0 . 16 and 0 . 19 , respectively . The RMSF distributions , δϕ ( i ) and δψ ( i ) , shown in Fig 4c display distinctive features characteristic of C22* force field . Specifically , the plot reveals two sequence regions with suppressed backbone fluctuations , Val17-Asp23 and Leu34-Val36 , which approximately coincide with the formation of stable turn and marginally stable helix structures . Furthermore , in C22* simulations for most sequence positions δϕ ( i ) is much smaller than in C36 . Overall , the average RMSF 〈δϕ〉 and 〈δψ〉 observed in C22* are 30 . 2 ± 1 . 5° and 66 . 1 ± 4 . 6° , i . e . , compared to C36 a particularly significant decrease by a factor of 1 . 8 is seen in 〈δϕ〉 . It is then not surprising that the RMSD for δϕ ( i ) and δψ ( i ) distributions computed between C22* and C36 simulations are 31 . 1° and 33 . 0° , respectively , which are about an order of magnitude larger than those comparing C36 force fields . The C22* equilibrium contact map 〈C ( i , j ) 〉 displayed in Fig 5c shows a formation of multiple side chain contacts , of which eight long-range and ten short-range contacts are stable . Moreover , Table 4 lists top five long-range contacts , one of which , the salt bridge Lys16-Asp23 , is formed with exceptionally high probability of 0 . 85 ± 0 . 02 . Two hydrophobic long-range contacts , Phe19-Val24 and Val24-Ala30 , also occur with high probabilities ( 0 . 52 ± 0 . 04 and 0 . 46 ± 0 . 08 ) . Interestingly , in C22* simulations Asp23-Lys28 salt bridge is effectively disrupted ( 0 . 06 ± 0 . 03 ) . Also , multiple very stable short-range contacts are observed , such as the helix-like contact Gly33-Gly37 ( 0 . 86 ± 0 . 03 ) . Overall , the equilibrium number of side chain contacts in Aβ10-40 is 〈C〉 = 29 . 0 ± 0 . 8 , of which 〈CLR〉 = 15 . 4 ± 0 . 3 ( or 53% ) are long-range . Compared to C36 , the values of 〈C〉 and 〈CLR〉 are larger by 37% and 44% , respectively . Finally , we note that none of the top five C22* long- or short contacts are observed in C36 or C36s simulations . Consequently , the RMSD value measuring the difference between C36 and C22* contact maps is much larger ( 0 . 12 ) than the RMSD comparing C36 and C36s force fields . As seen in Fig 6 the probability distribution P ( Rg ) for Aβ10-40 monomer computed from C22* simulations peaks at smaller values of Rg and is more narrow than for either of C36 force fields . From Fig 6 we find 〈Rg〉 = 14 . 5 ± 0 . 2Å , which is smaller than the respective values for C36 simulations ( 16 . 9 or 15 . 9 Å ) . The same conclusion applies to the end-to-end distance ( 〈R1N〉 = 20 . 0 ± 0 . 9 Å ) . In summary , compared to C36 simulations , C22* force field significantly enhances turn and , particularly , helical propensities , makes several sections of Aβ backbone rigid , and dramatically strengthens tertiary interactions as illustrated in Fig 1d resulting in peptide compaction . Therefore , there is little consistency between the conformational ensembles mapped using C22* and C36 force fields . In our previous work [29 , 43] , we have used REMD to study the conformational ensemble of the Aβ10-40 monomer using CHARMM22 force field with CMAP corrections ( denoted as C22cmap ) . Below we extend our previous C22cmap analysis to provide comparison with other force fields . We begin by focusing on Aβ secondary structure . Fig 2 and Table 1 demonstrate that in C22cmap Aβ mainly forms turn ( 〈T〉 = 0 . 49 ± 0 . 01 ) and random coil ( 〈RC〉 = 0 . 38 ± 0 . 01 ) structures . These observations are qualitatively consistent with other force fields . However , compared to both C36 the population of helical structure in C22cmap ( 〈H〉 = 0 . 12 ± 0 . 01 ) is elevated , whereas the formation of β structure is still rare . Fig 3 demonstrates residue-specific secondary structure propensities for C22cmap . Stable turn structure is observed in His13 , Phe19-Gly25 , Asn27-Gly29 , and Met35-Gly37 , which with the exception of the last region match well the C36 turn distribution . Random coil , 〈RC ( i ) 〉 , is observed in Tyr10-Glu11 , Gln15-Val18 , and Val39-Val40 , i . e . , it follows the respective C36 propensities . Fig 3a also reveals an appearance of marginally stable helix in the C-terminal region S4 ( Ile32-Val36 ) , which approximately coincides with the distribution of C22* helix . The RMSD values comparing C22cmap and C36 distributions of turn and helix structure are 0 . 11 and 0 . 12 , respectively , implicating moderate differences between the two systems . Fig 4d demonstrates that the fluctuations of backbone dihedral angles , δϕ ( i ) and δψ ( i ) , with the exception for Gly residues , agree generally well with those computed for C36 . However , in contrast to C36 distributions , the backbone fluctuations are suppressed in a wider interval of Val18-Val24 compared to a single position Phe20 in C36 . The averages 〈δϕ〉 and 〈δψ〉 are 49 . 5 ± 1 . 0° and 74 . 0 ± 1 . 2° , which are somewhat smaller than the respective averages computed from C36 simulations . The RMSD values comparing δϕ ( i ) and δψ ( i ) between C22cmap and C36 are 13 . 7° and 21 . 1° , respectively . These RMSD values are much larger than those comparing C36 and C36s force fields , but significantly smaller ( by 2 . 3 or 1 . 6 times , respectively ) than the RMSD probing the difference between C36 and C22* . In line with these observations , the C22cmap fluctuations in Fig 4d , particularly δϕ ( i ) , are generally higher than those observed for C22* . Fig 5d presents the equilibrium contact map 〈C ( i , j ) 〉 computed using C22cmap simulations . In all , we have detected the formation of nine short-range and one long-range ( Lys16-Asp23 ) stable contacts that stands in contrast to C36 results , which showed only two stable short-range contacts . The Asp23-Lys28 salt-bridge , which is disrupted in C22* or C36 , has also low probability of occurrence ( 0 . 15 ± 0 . 03 ) . Moreover , Table 5 demonstrates none of the top long- or short-range contacts formed in C22cmap force field are observed in any of C36 simulations . The overall number of side chain contacts formed in C22cmap is 〈C〉 = 26 . 2 ± 0 . 2 , of which 〈CLR〉 = 12 . 3 ± 0 . 2 ( or 47% ) are long-range . Compared to C36 simulations , 〈C〉 and 〈CLR〉 are larger by 24 and 15% , respectively . As a result , the RMSD comparing the C22 and C36 contact maps is moderately large ( 0 . 07 ) , exceeding the RMSD between C36 and C36s simulations ( 0 . 02 ) , but still being smaller than the RMSD comparing C36 and C22* ( 0 . 12 ) . Last , we examined the probability distribution P ( Rg ) plotted in Fig 6 . The distribution observed for C22cmap shows surprisingly good agreement with C36 ( particularly , C36s ) force fields . From P ( Rg ) we found 〈Rg〉 = 15 . 8 ± 0 . 2 Å , which is almost equal to that observed for C36s ( 〈Rg〉 = 15 . 9 Å ) , somewhat smaller than that of C36 ( 16 . 9Å ) , but larger than the C22* result ( 14 . 5Å ) . Taken together , C22cmap simulations exhibit a weak helix propensity in the C-terminal consistent with the formation of large number of stable short-range contacts that positions the C22cmap conformational ensemble in-between C36 and C22* ensembles ( see Discussion for further analysis ) . The representative structure of Aβ10-40 peptide in C22cmap force field is displayed in Fig 1e . To evaluate Aβ10-40 conformational ensemble in the force field unrelated to any CHARMM version , we have performed OPLS-AA REMD simulations . Similar to all previously considered force fields Fig 2 and Table 1 show that Aβ10-40 in OPLS-AA simulations adopt mainly turn ( 〈T〉 = 0 . 46 ± 0 . 02 ) or random coil ( 〈RC〉 = 0 . 40 ± 0 . 03 ) conformations , which together represent 86% of all amino acid states . However , in contrast to other force fields ( e . g . , C36 ) β-state fraction is elevated four-fold to 〈S〉 = 0 . 12 ± 0 . 02 , whereas α-helix occurrence is negligible . Fig 3 , which displays residue-specific secondary structure propensities , further underscores the characteristic OPLS-AA feature—an elevated sampling of β-structure in the S2 and S4 hydrophobic regions , where for some positions i 〈S ( i ) 〉 reaches ≈0 . 4 . The helix propensity across Aβ sequence is uniformly weak , whereas the turn and random coil fractions generally follow the other force field trends . Accordingly , the RMSD comparing the β distributions 〈S ( i ) 〉 in OPLS-AA and C36 simulations has a large value of 0 . 13 , whereas the RMSD for helix and turn propensities are much smaller ( 0 . 04 and 0 . 08 ) . Fig 4e displays the distributions of dihedral angle RMSF , δϕ ( i ) and δψ ( i ) , which are qualitatively similar to their C36 counterparts . In fact , the average 〈δϕ〉 = 54 . 4 ± 1 . 2° and 〈δψ〉 = 91 . 1 ± 1 . 8° are close to the C36 values . Consequently , the RMSD comparing δϕ ( i ) and δψ ( i ) distributions from OPLS-AA and C36 simulations are relatively small ( 20 . 2° and 12 . 8° ) . These findings indicate that the β-structure occurs in Aβ peptide transiently and does not strongly affect backbone fluctuations . The equilibrium contact map 〈C ( i , j ) 〉 displayed in Fig 5e clearly exhibits more extensive tertiary interactions forming in OPLS-AA comparing to C36 force field . Indeed , the numbers of all and long-range contacts are increased by about 50% and 100% , respectively , to 〈C〉 = 30 . 8 ± 0 . 5 and 〈CLR〉 = 21 . 1 ± 0 . 6 resulting in the largest fraction of long-range interactions of 69% among all force fields tested by us . Interestingly , even though OPLS-AA promotes tertiary interactions , very few of them qualify as stable . Specifically , in sharp contrast to C22* Table 6 lists only one stable long-range ( Val18-Leu34 ) and one stable short-range ( Tyr10-Val12 ) contacts . Among top five long-range contacts four are hydrophobic and three link the sequence regions S2 and S4 ( Val18-Leu34 , Val18-Val36 , Leu17-Met35 ) with the probabilities of occurrence close to the stability threshold . Also , unique to OPLS-AA simulations , the salt-bridge Asp23-Lys28 appears among the top five long-range contacts . Finally , no long- or short-range top five contacts are shared between OPLS-AA and C36 simulations . It is then not surprising that extensive but weak tertiary interactions formed in OPLS-AA simulations lead to a fairly large RMSD value measuring the difference between the contact maps obtained in OPLS-AA and C36 force fields ( 0 . 07 ) . Probability distribution P ( Rg ) for Aβ10-40 radius of gyration presented in Fig 6 is far more narrow and reaches maximum at the smallest Rg compared to any other force field . We determined from Fig 6 that the equilibrium value 〈Rg〉 is 13 . 5 ± 0 . 2Å , which is the smallest among the force fields tested . Similar observation holds for the end-to-end distance ( 〈R1N〉 = 15 . 6 ± 1 . 5 Å ) . Thus , OPLS-AA force field promotes extensive but flickering tertiary interactions resulting in Aβ10-40 collapse and moderate enhancement of β-structure as shown in Fig 1f . We have compared 3JHNHα-coupling and RDC constants computed from our simulations and measured experimentally ( see Materials and Methods and Fig 7a ) . ( It is important to note that experimental measurements refer to the full-length peptide Aβ1-40 , whereas our simulations have examined the amino-truncated fragment Aβ10-40 . This point is elaborated in Discussion . ) We first investigated the agreement between 3JHNHα-coupling constants . As stated in the Materials and Methods we used three sets of experimental 3JHNHα-coupling constants , Jexp , measured by Garcia and coworkers [16 , 17] and , more recently , by Bax and coworkers [18] . Furthermore , to compute 3JHNHα-coupling constants in silico , Jcomp , we applied Eq ( 1 ) with three different sets of Karplus equation coefficients [51–53] . Because a priori it is unclear which combination of experimental data and Karplus equation coefficients is more accurate , we considered all nine possible combinations and for each computed Jcomp ( i ) -coupling constants using Aβ10-40 sequence positions i with available experimental measurements . Consistency between Jcomp ( i ) and Jexp ( i ) was evaluated by calculating the root-mean-square deviation ( RMSD ) , quality function Q , and Pearson’s correlation coefficient ( PCC ) as described in the Materials and Methods . These quantities were then averaged over nine datasets and are presented in Table 7 . Judged by RMSD , PCC , and Q values as well as their errors the best agreement between experimental and computational J-coupling constants is observed for C36 and C36s force fields . The ranking of other force fields in the descending order of agreement with the experiment is OPLS-AA , C22cmap , and C22* . Notably , C22* shows very significant increase in RMSD ( by 41% ) compared to C36 and has effectively no correlation with Jexp ( i ) as measured by PCC ( 0 . 14 ) . To provide additional probe of Aβ10-40 conformational ensemble we have computed the RDC constants , RDCcomp ( i ) , as described in the Materials and Methods and S1 Text . To compare them with their experimental counterparts , RDCexp ( i ) , [54] shown in Fig 7b , we used the same metrics as for 3JHNHα-coupling constants . The results presented in Table 7 indicate that the best agreement between experimental and computational RDC constants is observed for C36s force field , which has the smallest RMSD or Q and the largest PCC of 0 . 65 . Compared to C36s force field , C36 exhibits larger RMSD and Q ( by ≈ 20% for both ) and significantly lower PCC ( a 25% decrease ) . With respect to RMSD and Q the worst agreement is seen for C22cmap and C22* , for which the RMSD or Q values are about two-fold larger than for C36s . Their PCC values are also lower than for C36s , but higher , especially of C22cmap , than for C36 . Measured by RMSD and Q the agreement with the experiment places OPLS-AA between C36 and C22cmap/C22* . However , OPLS-AA demonstrates the worst correlation between computational and experimental RDC constants as measured by PCC . Thus , taken together our data suggest that the force field providing the best consistency with the experimental data is C36s . If we disregard a negligible ( within the error ) difference in J-coupling RMSD for C36s and C36 ( Table 7 ) , this conclusion is supported by all comparison metrics . With the exception of RDC PCC , other metrics identify C36 as the next “best” force field . Again , with the exception of PCC for RDC all the metrics determine that C22* produces the worst agreement with the experimental data . The implications of our findings for the selection of force field and water model and for the differences between Aβ1-40 and Aβ10-40 conformational ensembles are presented in the Discussion .
Using REMD we have performed a comparative analysis of Aβ10-40 conformational ensembles generated by employing five force fields , which combine four protein parameterizations ( C36 , C22* , C22cmap , and OPLS-AA ) and two water models ( standard and modified TIP3P ) . As a reference in our analysis we took the recent modification of CHARMM force field , CHARMM36 , coupled with modified TIP3P water model . Its selection was motivated by recent tests showing that this force field provides the best agreement with experimental NMR data collected for six proteins [57] . Taken together , our results suggest several observations . First , all force fields produce fairly consistent distributions of secondary structure . According to Fig 2 and Table 1 all of them predict largely similar fractions of turn conformations ( varying between 0 . 44 and 0 . 52 ) and , to a lesser extent , random coil ( varying between 0 . 30 and 0 . 49 ) . In all force fields , the turn structure dominates within approximately the same sequence regions , His13 and His14 in S1 region and a sequence interval Phe19-Gly29 , whereas the random coil occurs at Aβ10-40 termini . Nevertheless , there are also significant variations among the force fields . For example , a unique feature of C22* force field is a considerable helix bias in the S3 and S4 regions resulting in 〈H ( i ) 〉 ≈ 0 . 4 for few positions . Similarly , OPLS-AA differs from other simulations by significant β-structure propensity in the S2 and S4 regions , where 〈S ( i ) 〉 peaks at ≈ 0 . 4 . Second , additional insight into peptide conformational ensemble is provided by the fluctuations in backbone dihedral angles , δϕ ( i ) and δψ ( i ) . Similar to secondary structure , backbone fluctuations in C36 and C36s are in excellent agreement as evidenced by their consistent average values and small RMSDs . C22cmap differs moderately from C36 by having slightly more rigid backbone ( i . e . , smaller average 〈δϕ〉 and 〈δψ〉 values ) , whereas OPLS-AA , in contrast , demonstrates enhanced backbone fluctuations . However , the force field clearly standing apart from others is C22* , which predicts suppressed fluctuations in two sequence regions ( Val17-Asp23 and Leu34-Val36 ) . This feature plus generally small fluctuations in ϕ angles result in the lowest averages 〈δϕ〉 and 〈δψ〉 as well as elevated , by an order of magnitude , RMSD value between C22* and C36 simulations as opposed to that between C36 and C36s . Therefore , when secondary structure and backbone fluctuations are considered together , the force fields can be ranked in the descending order of similarity to C36 as C36s , which is nearly identical to C36 , C22cmap and OPLS-AA , which moderately differ from C36 , and C22* , which reveals considerably differences due to helix formation and rigid backbone . It is important to consider our results in the context of other studies of force field propensities . C22cmap tendency to bias peptide conformational ensembles toward helical states has been documented for Ala pentamer [58] , to correct which C36 force field has been developed [39] . Our results show that C22cmap indeed produces a slightly elevated helix fraction in unstructured Aβ10-40 compared to C36 or C36s increasing it to 0 . 12 from 0 . 04 or 0 . 06 , respectively . However , this bias is weak compared to that of C22* , which demonstrates much stronger helix propensity in the C-terminal . Small differences in helix propensities between C22cmap and C36 have also been noted for the unstructured fragments of NTL9 peptides [31] . Our third observation is related to the distribution of tertiary interactions . In Aβ peptide C36 force field produces no stable long-range interactions and only two stable short-range contacts and , overall , it leads to the smallest numbers of tertiary contacts ( 21 . 1 ) and long-range contacts ( 10 . 7 ) . Aβ conformations in this force field are also the least compact being characterized by the largest radius of gyration ( 16 . 9 Å ) . Thus , we conclude that equilibrium C36 conformations are dominated by expanded structures lacking stable interactions . C36s force field , which differs from C36 solely by the water model , generates very similar conformational ensemble , which also lacks stable interactions and shares four or three common top long- and short-range contacts with C36 . Overall , the contact map RMSD for C36 and C36s is very low ( 0 . 02 ) . Nevertheless , the C36s numbers of contacts , including long-range , are slightly larger ( by 10–20% ) than in C36 . Additionally , compared to C36 C36s features slighly smaller Aβ radius of gyration . These results are consistent with recent comparison of standard and modified TIP3P water models , which showed that the latter enhances hydration and generates more open peptide conformations [32] . Although Aβ structures produced with C22cmap and C22* share some similarity ( three long- and short-range top contacts are common ) , C22* is by far unique in generating the largest number of stable long-range contacts ( eight as opposed to one in C22cmap ) . Some of these contacts , such as salt-bridge Lys16-Asp23 , are effectively always formed . A distinctive characteristic of C22cmap is the large number ( 9 ) of stable short-range contacts combined with very few ( 1 ) stable long-range interactions . According to contact map RMSD , C22cmap differs moderately from C36 ( 0 . 07 ) , whereas C22* deviates from C36 by far larger degree ( 0 . 12 ) . There are no common top long- or short-range contacts between C36 and C22cmap or C22* . The tertiary interactions also set OPLS-AA force field apart from all other simulations . OPLS-AA generates the largest number of all contacts and , more importantly , the largest number of long-range contacts , which is increased almost two-fold compared to other simulations ( except for C22* , with respect to which 〈 CLR 〉 increases one-third ) . As a result OPLS-AA conformations exhibit extremely large fraction of long-range interactions reaching almost 70% and , accordingly , Aβ adopts most compact structures ( 〈Rg〉 13 . 5 Å ) . Similar observation concerning Aβ1-40 peptide collapse has been made previously for OPLS-AA/L force field [30] . Interestingly , OPLS-AA tertiary contacts , although numerous , are weak suggesting that Aβ samples compact but still disordered states . OPLS-AA and C36 do not share common top interactions , whereas the contact map RMSD between the two is moderate ( 0 . 07 ) . Thus , using contact map RMSD and C36 as reference we rank the force fields in the descending order of similarity as C36s , C22cmap , OPLS-AA ( due to compact state ) , and C22* . It might be argued that the computed Aβ10-40 conformational ensembles are specific to 330K . To investigate this possibility we used our REMD sampling to recompute the secondary structure propensities for five force fields at 300K . Fig D and Table A in S1 Text represent the analogues of Fig 2 and Table 1 obtained at 300K . It is seen that the secondary structure propensities at 300K and 330K are qualitatively similar , differring by no more than 7% for the propensities > 0 . 1 . The same conclusion applies to the comparisons of the distributions of secondary structure along Aβ10-40 sequence , which demonstrate , for instance , that C22* helical propensity < H ( i ) > only slighly increases at 300K . Finally , decrease in temperature triggers no qualitative changes in tertiary interactions ( see S1 Text ) . Therefore , we surmize that Aβ10-40 conformational ensembles at 300K and 330K appear qualitatively similar . Combining the analysis of secondary and tertiary structures we make the following conclusions: Using REMD simulations we have computed the 3JHNHα-coupling and RDC constants for the amino-truncated peptide Aβ10-40 and compared them to the experimental measurements performed for the full-length peptide Aβ1-40 . Although Aβ1-40 and Aβ10-40 peptides are not entirely identical having 78% of sequence homology , their comparison can still be instructive as demonstrated below . We first evaluate the agreement between in silico Aβ10-40 and experimental Aβ1-40 data in light of similar assessments made in the literature for Aβ1-40 or Aβ1-42 peptides , in which identical peptide species were used both in the experiments and simulations [17 , 27] . According to our analysis ( Table 7 ) RMSD and PCC between the in silico and experimental J-coupling constants for the “best” C36s force field is 0 . 89 Hz and 0 . 44 , respectively . Similar comparison made for Aβ1-42 using AMBERff99SB force field and TIP4P-Ew water model [27] yielded the RMSD values of 0 . 96 Hz or 1 . 46 Hz depending on the specific set of Karplus equation coefficients [52 , 53] ( the average is 1 . 21 ) . The PCC values varied in the interval 0 . 4–0 . 5 . If we restrict our computations of the average RMSD and PCC to the two coefficient sets used by Sgourakis et al [27] , we obtain RMSD = 0 . 92 Hz and PCC = 0 . 44 . ( Note that previous studies have often performed fitting of Karplus equation coefficients to better represent experimental data [17 , 27] . However , we opted against this adjustment to provide more unbiased assessment of force fields . ) Our comparison of Aβ10-40 RDC constants produced with C36s to their Aβ1-40 experimental counterparts results in RMSD = 1 . 36 Hz and PCC = 0 . 65 . Analogous comparisons made for Aβ1-42 led to RMSD = 1 . 49 Hz and 0 . 35 ≲ PCC ≲ 0 . 45 [27] . A recent study has compared in silico and experimental J-coupling and RDC constants for three peptides , Aβ1-42 , Aβ1-40 , and Aβ1-42-M35ox using OPLS-AA/L force field and TIP3P water model [17] . Using the coefficients of Vuister and Bax [53] , the RMSD values for J-coupling distributions were 1 . 21 , 1 . 29 , and 1 . 06 Hz , respectively ( the average is 1 . 19 Hz ) , whereas the PCC values were 0 . 50 , 0 . 76 , 0 . 49 ( the average is 0 . 58 ) . If we again restrict our computations of the RMSD and PCC to the Vuister and Bax coefficients , we find RMSD = 0 . 85 Hz and the PCC = 0 . 48 . The comparison of the in silico and experimental RDC distributions for Aβ1-42 , Aβ1-40 , and Aβ1-42-M35ox [27] yielded the RMSD values of 1 . 66 , 1 . 69 , and 1 . 45 Hz ( the average is 1 . 6 Hz ) , and the PCC values of 0 . 39 , 0 . 50 , and 0 . 44 ( the average is 0 . 44 ) . Finally , the third study has performed similar comparison of in silico and experimental J-coupling and RDC data using AMBERff99SB force field [28] . The RMSD values for J-coupling distributions were 0 . 99Hz for both Aβ1-40 and Aβ1-42 . The RMSD values comparing RDC distributions were about 2 . 2Hz for both peptides . Thus , if we consider the RMSD values comparing Aβ1-40 and Aβ10-40 peptides against the RMSD comparisons made previously for identical peptides , it becomes clear that the difference in J-coupling constants between Aβ1-40 and Aβ10-40 is actually smaller than the reported values for Aβ1-42 , Aβ1-40 , or Aβ1-42-M35ox ( our RMSDs of 0 . 89 , 0 . 92 , or 0 . 85 Hz vs “their” RMSDs of 1 . 21 , 1 . 19 , or 0 . 99 Hz ) . The values of PCC calculated by us are about the same or slightly lower than those reported for the three peptides ( our PCCs of 0 . 44 , 0 . 44 or 0 . 48 vs “their” PCC in the range of 0 . 4–0 . 6 ) . As shown above these conclusions hold irrespective of computing the RMSD and PCC using all Karplus equation coefficient sets or only specific sets . Similarly , as measured by RMSD the agreement between in silico Aβ10-40 and experimental Aβ1-40 RDC data is much better than the previous comparisons , which involved identical in silico and experimental peptide species , such as Aβ1-42 , Aβ1-40 , or Aβ1-42-M35ox ( our RMSD of 1 . 36 vs “their” RMSD of 1 . 49 , 1 . 6 , or 2 . 2 Hz ) . The same conclusion is supported by PCC comparing RDC distributions , which is 0 . 65 in our study against the approximate range of 0 . 35 to 0 . 45 reported previously . Taken together the analysis above suggests two conclusions . First , if in silico Aβ10-40 and experimental Aβ1-40 J-coupling and RDC constants are generally in better agreement than these quantities computed and measured for strictly identical peptides , then the differences in the conformational ensembles of Aβ10-40 and Aβ1-40 are likely to be small or , at least , not exceeding the force field errors in reproducing the conformations of a specific peptide ( Aβ1-42 , Aβ1-40 , or Aβ1-42-M35ox ) . Therefore , guided by the previous validations of protein force fields against Aβ NMR data [16 , 17 , 27 , 28] we argue that our analysis supports using Aβ10-40 peptide as a proxy of the full-length Aβ1-40 . This conjecture has been made earlier by our [33 , 59] and other [60] groups . In this context , we note that the OPLS-AA/L simulations of the full-length Aβ1-40 have predicted stable β-structure in Leu17-Ala21 and Ile31-Val36 regions [30] . In line with our view of Aβ10-40 as a proxy of Aβ1-40 , the elevated β-structure propensity is observed in the same Aβ10-40 regions when sampled in our OPLS-AA simulations . Second , a good agreement between in silico Aβ10-40 and experimental Aβ1-40 J-coupling and RDC constants argues that CHARMM36 force field with standard TIP3P water model is possibly the best force field for reproducing Aβ conformational ensemble . Previous studies evaluating the force fields for their ability to reproduce Aβ experimental data have identified OPLS-AA with TIP3P water model as most accurate [16] . However , to our knowledge CHARMM force fields have never been directly evaluated against the distributions of Aβ J-coupling and RDC constants . In this study we addressed this issue . Recently , eight different force fields were evaluated using REMD simulations for their ability to reproduce small-angle X-ray scattering and NMR data for five natively unstructured peptides [61] . The authors have determined that CHARMM22* generates the conformational ensembles most consistent with the experiments . They also noted erroneous CHARMM36 propensity to sample left-handed α-helix conformations . However , our study did not reach the same conclusions for Aβ peptides suggesting that the selection of the “best” force field still depends on the peptide and details of simulations . Incidentally , an updated version of CHARMM36 force field has been recently released , which corrects left-handed α-helix bias [62] . However , in the case of Aβ10-40 peptides this modification appears as not critically necessary given the lack of Aβ left-handed α-helix in the original CHARMM36 force field . By applying REMD simulations we have performed comparative analysis of the conformational ensembles of amino-truncated Aβ10-40 peptide produced with five force fields , which combine four protein parameterizations ( CHARMM36 , CHARMM22* , CHARMM22/cmap , and OPLS-AA ) and two water models ( standard and modified TIP3P ) . Aβ10-40 conformations were characterized by the analysis of secondary structure , backbone fluctuations , tertiary interactions , and radius of gyration . In addition , using computed conformational ensembles we have calculated Aβ10-40 3JHNHα-coupling and RDC constants and compared them with their experimental counterparts obtained for the full-length Aβ1-40 peptide . Taken together , our study led us to several conclusions . First , all force fields predict that Aβ adopts unfolded structure dominated by turn and random coil conformations . Second , specific TIP3P water model does not dramatically affect secondary or tertiary Aβ10-40 peptide structure , albeit standard TIP3P model favors slightly more compact states . Third , although the secondary structures observed in CHARMM36 and CHARMM22/cmap simulations are qualitatively similar , their tertiary interactions show little consistency . Fourth , two force fields have unique features setting them apart from CHARMM36 or CHARMM22/cmap . Specifically , OPLS-AA reveals moderate β-structure propensity coupled with extensive , but weak long-range tertiary interactions leading to Aβ collapse . CHARMM22* exhibits moderate helix propensity and generates multiple , exceptionally stable long- and short-range interactions . There are no common frequent tertiary interactions between CHARMM36 and OPLS-AA or CHARMM22* force fields . Our investigation suggests that among all force fields CHARMM22* differs the most from CHARMM36 . Fifth , the analysis of 3JHNHα-coupling and RDC constants based on CHARMM36 force field with standard TIP3P model led us to an unexpected finding that in silico Aβ10-40 and experimental Aβ1-40 constants are generally in better agreement than these quantities computed and measured for identical ( 100% homologous ) peptides , such as Aβ1-40 or Aβ1-42 . On the basis of this observation we argued that the differences in the conformational ensembles of Aβ10-40 and Aβ1-40 are likely to be small and the former can be used as proxy of the full-length peptide . We also concluded that CHARMM36 force field with standard TIP3P model produces the most accurate representation of Aβ10-40 conformational ensemble . | Dependence of protein conformational ensembles on force field parameterizations limits the predictive power of molecular dynamics simulations . To address this problem , we evaluated five all-atom force fields for their consistency in reproducing the conformational ensemble of Alzheimer’s Aβ10-40 peptide . To generate conformational ensembles , we have used replica exchange molecular dynamics and computed Aβ10-40 secondary and tertiary structures . We found that , although all force fields predict Aβ10-40 unfolded structure , they strongly disagree on helix and β propensities and tertiary structure distributions . We have also calculated Aβ10-40 J-coupling and residual dipolar coupling constants and compared them with the experimental data for the full-length Aβ1-40 peptide . Unexpectedly , we determined that in silico Aβ10-40 and experimental Aβ1-40 constants are in better agreement than these quantities computed and measured previously for identical peptides , such as Aβ1-40 or Aβ1-42 . We then concluded that the conformational ensembles of Aβ10-40 and Aβ1-40 are similar and on this basis argue that CHARMM36 force field with standard TIP3P water model provides the most accurate description of Aβ10-40 . Although our objective was not to evaluate the biomolecular force fields in general , our study is expected to facilitate their proper selection for the simulations of Alzheimer’s peptides . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"geometry",
"simulation",
"and",
"modeling",
"probability",
"distribution",
"mathematics",
"protein",
"structure",
"prediction",
"protein",
"structure",
"research",
"and",
"analysis",
"methods",
"proteins",
"peptide",
"mapping",
"salt",
"bridges",
"chemistry",
"proteomics",
"molecular",
"biology",
"probability",
"theory",
"dihedral",
"angles",
"protein",
"structure",
"comparison",
"biochemistry",
"biochemical",
"simulations",
"electrochemistry",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"macromolecular",
"structure",
"analysis"
]
| 2017 | Is the Conformational Ensemble of Alzheimer’s Aβ10-40 Peptide Force Field Dependent? |
Genome-wide association studies ( GWAS ) have identified more than 90 susceptibility loci for breast cancer , but the underlying biology of those associations needs to be further elucidated . More genetic factors for breast cancer are yet to be identified but sample size constraints preclude the identification of individual genetic variants with weak effects using traditional GWAS methods . To address this challenge , we utilized a gene-level expression-based method , implemented in the MetaXcan software , to predict gene expression levels for 11 , 536 genes using expression quantitative trait loci and examine the genetically-predicted expression of specific genes for association with overall breast cancer risk and estrogen receptor ( ER ) -negative breast cancer risk . Using GWAS datasets from a Challenge launched by National Cancer Institute , we identified TP53INP2 ( tumor protein p53-inducible nuclear protein 2 ) at 20q11 . 22 to be significantly associated with ER-negative breast cancer ( Z = -5 . 013 , p = 5 . 35×10−7 , Bonferroni threshold = 4 . 33×10−6 ) . The association was consistent across four GWAS datasets , representing European , African and Asian ancestry populations . There are 6 single nucleotide polymorphisms ( SNPs ) included in the prediction of TP53INP2 expression and five of them were associated with estrogen-receptor negative breast cancer , although none of the SNP-level associations reached genome-wide significance . We conducted a replication study using a dataset outside of the Challenge , and found the association between TP53INP2 and ER-negative breast cancer was significant ( p = 5 . 07x10-3 ) . Expression of HP ( 16q22 . 2 ) showed a suggestive association with ER-negative breast cancer in the discovery phase ( Z = 4 . 30 , p = 1 . 70x10-5 ) although the association was not significant after Bonferroni adjustment . Of the 249 genes that are 250 kb within known breast cancer susceptibility loci identified from previous GWAS , 20 genes ( 8 . 0% ) were statistically significant associated with ER-negative breast cancer ( p<0 . 05 ) , compared to 582 ( 5 . 2% ) of 11 , 287 genes that are not close to previous GWAS loci . This study demonstrated that expression-based gene mapping is a promising approach for identifying cancer susceptibility genes .
Breast cancer is the most common cancer in women in the United States and in the world [1] . It is a heterogeneous disease and the two main subgroups of breast cancer are estrogen receptor ( ER ) -positive and ER-negative cancer . Genome-wide association studies ( GWAS ) have identified more than 90 susceptibility loci for breast cancer [2–20] , with only a few loci specific for ER-negative breast cancer [3 , 15 , 17] . Susceptibility loci for ER-positive loci are often the same as loci for overall breast cancer risk because most of breast cancers are ER-positive , especially in women of European or Asian ancestry [2 , 4 , 19] . Women of African ancestry are more likely to be diagnosed with ER-negative breast cancer compared to women of non-African ancestry [21–23] . To date , breast cancer GWAS have been conducted primarily in populations of European ancestry . The difference in linkage disequilibrium ( LD ) patterns and allele frequencies across ancestry groups may explain the apparent inconsistencies in GWAS findings from studies of women of European ancestry as compared to studies of women of African ancestry [24 , 25] . The strength and the direction of the association between causal variants and disease are expected to be consistent across populations , and thus cross-population validation provides further evidence of causation . In addition , trans-ancestry analysis could identify novel breast cancer susceptibility variants [26] . The variants discovered by previous GWAS along with previously known high-penetrance genes explain only a modest proportion of the heritability of breast cancer [2] . More genetic factors for breast cancer are yet to be identified , but power for discovery of new loci is limited by the sample size of existing GWASs . Moreover , the biologic significance of the variants identified by GWAS and the genes on which they act , are often unknown . Single nucleoid polymorphisms ( SNPs ) associated with disease traits are more likely to be expression quantitative trait loci ( eQTLs ) [27] , and regulatory variants can explain a large proportion of disease heritability [28] . Therefore , genes regulated by eQTLs can be used as an enrichment analysis unit to identify more genetic risk factors for breast cancer . Recently , gene-based approaches using eQTL information , such as PrediXcan , have been proposed , which can reduce the multiple testing burden in genome-wide analyses and have been used to identify novel genes for autoimmune diseases [29] . PrediXcan uses individual-level data to estimate the correlation between genetically predicted levels of gene expression and human traits to prioritize causal genes . MetaXcan computes the same correlation as PrediXcan , but does so using summary statistics from GWAS , which are much more readily accessible than individual level data [30] . To identify novel genes involved in breast cancer susceptibility , we utilized a gene-level expression-based association method , implemented in the MetaXcan software [30] , to infer gene expression levels using summary statistics from five GWASs . We used an additive prediction model of gene-expression levels trained in Depression Genes and Network ( DGN ) data [31] and examined the predicted expression of specific genes for association with overall breast cancer risk and estrogen receptor-negative breast cancer risk . The GWAS datasets were made available in dbGaP ( https://www . ncbi . nlm . nih . gov/gap ) through “Up For A Challenge ( U4C ) –Stimulating Innovation in Breast Cancer Genetic Epidemiology” launched by the National Cancer Institute . The DGN data included RNA sequencing data from whole blood of 922 genotyped individuals ( 463 cases of major depressive disorder and 459 controls ) , all of European ancestry . These individuals consisted of 274 males and 648 females with ages ranged from 21 to 60 .
Using logistic regression , we first conducted SNP-level GWAS analysis for overall breast cancer risk among 8605 breast cancer cases and 8095 controls , and for ER-negative breast cancer risk among 3879 cases and 10213 controls . The analyses were performed for each of the five GWAS datasets separately and summary statistics including log odds ratios and standard errors were generated . These summary statistics for each dataset were input to the software MetaXcan [30] to perform genome-wide gene-level expression association tests for 11 , 536 genes . Then , we performed meta-analysis of the results from individual MetaXcan analyses . Quantile-quantile plots of P-values from the meta-analysis showed little inflation ( Fig 1 ) . For overall breast cancer risk , there was no gene with a P-value that deviated from the null distribution ( Fig 1A ) , but for ER-negative breast cancer risk analysis , there were several genes with P-values smaller than expected , including TP53INP2 , HP , and DHODH ( Fig 1B ) . Table 1 lists the top genes with P-values less than 10−3 in the analyses of association between predicted gene expressions and overall breast cancer risk . The sign of Z score indicates the direction of association between genetically-predicted expression and breast cancer risk . None of the genes reached genome-wide significance when a Bonferroni threshold ( α = 4 . 33x10-6 ) was used . Of the 249 genes that are 250 kb within known susceptibility loci identified from previous breast cancer GWAS [2–4 , 17 , 32] , 12 genes ( 4 . 8% ) were statistically significant associated with overall breast cancer risk at nominal significance level of 0 . 05 , compared to 497 ( 4 . 4% ) of 11 , 287 genes that are not close to previous GWAS loci ( P for enrichment = 0 . 75 ) . Table 2 lists the genes with P-values less than 10−3 in the ER-negative breast cancer analysis . TP53INP2 was the top gene ( P = 5 . 35x10-7 ) , which surpassed the Bonferroni-corrected p-value threshold ( α = 4 . 33x10-6 ) . The false discovery rate for TP53INP2 was 0 . 0062 . Higher genetically-predicted TP53INP2 expression was associated with lower risk of ER-negative breast cancer . The gene with the second smallest P-value was HP , which had p-value of 1 . 70x10-5 , close to but not significant after Bonferroni correction . The false discovery rate for the HP gene was 0 . 098 . For the HP gene , higher expression was associated with higher risk of ER-negative breast cancer . Both genes are novel and no previous studies have found association between these two genes and breast cancer risk . Of the 249 genes that are 250 kb within known breast cancer susceptibility loci identified from previous GWAS , 20 genes ( 8 . 0% ) were statistically significant associated with ER-negative breast cancer ( p<0 . 05 ) , compared to 582 ( 5 . 2% ) of 11 , 287 genes that are not close to previous GWAS loci ( P for enrichment = 0 . 044 ) , suggesting a moderate enrichment for genes close to known susceptibility loci . There were six SNPs included in the prediction of the expression of the TP53INP2 gene , from 367 kb upstream to 159 kb downstream of the gene ( Table 3 ) . Five of the six SNPs ( except for rs8116198 ) were associated with overall breast cancer risk and ER-negative breast cancer risk ( at the nominal level of α = 0 . 05 ) , and the effects were consistently across studies ( none of the heterogeneity tests were significant ) . These associations were more significant for ER-negative breast cancer risk ( p values ranging from 5 . 0x10-4 to 1 . 8x10-6 ) than for overall breast cancer risk ( 7 . 0x10-4 to 1 . 4x10-4 ) . None of the SNP-level associations reached traditional genome-wide significance , thus they have not been reported in previous GWAS publications . However , our study showed the aggregate effects of these SNPs were significantly associated with ER-negative breast cancer after Bonferroni correction . We noticed that one of the six SNPs , rs8116198 , is monomorphic in the SBCGS data . Therefore , when MetaXcan was applied to the SBCGS data , the prediction of TP53INP2 expression was based on only five SNPs . To make our results more robust to missing and low quality genotypes , in the DGN prediction model , we used elastic net with 0 . 5 as the mixing parameter , which sets the degree of mixing between ridge regression and LASSO . In addition , the SNPs in the prediction were not necessarily causal but could be in LD with the causal SNPs . Fig 2 shows positions of the 6 eQTL SNPs for TP53INP2 in the cytoband 20q11 . 22 . Interestingly , there are several other genes in this region that were associated with ER-negative breast cancer , including MAP1LC3A , ITCH , and TRPC4AP ( Fig 2 and Table 2 ) . The 6 SNPs are located either in enhancer elements or in promotor regions ( Table 4 ) . The promotor/enhancer features of 4 SNPs were found in human mammary epithelial cells ( HMEC ) and breast variant human mammary epithelial cells ( HMEC . 35 ) , and the enrichment was statistically significant for both cell types ( both p<0 . 03 ) . There were 20 SNPs included in the prediction of the expression of the HP gene ( S1 Table ) . Thirteen of the 20 SNPs were associated with overall breast cancer risk and 17 were associated with the risk of ER-negative breast cancer ( at the nominal level of α = 0 . 05 ) , quite consistently across populations ( none of the heterogeneity tests were significant ) . The strengths of their associations were stronger for ER-negative breast cancer risk than for overall breast cancer risk . Interestingly , none of the associations for individual SNPs reached genome-wide significance , thus they have not been reported in previous GWAS publications . We used summary results from GAME-ON GWAS ( http://gameon . dfci . harvard . edu ) to replicate our study findings from the U4C . All the six eQTLs for the TP53INP2 gene were available in GAME-ON ( Table 5 ) . Five of the six SNPs that were associated with ER-negative breast cancer in the discovery phase ( using U4C datasets ) were all statistically significant in GAME-ON at the nominal 0 . 05 significance level . Gene-level test of TP53INP2 from MetaXcan gave a Z-score of -2 . 803 ( p = 5 . 1×10−3 ) for ER-negative breast cancer in GAME-ON . The gene-level test for overall breast cancer risk was not significant in GAME-ON ( Z-score = -1 . 627 , p = 0 . 10 ) . Because the GAME-ON ER-negative data included the BPC3 dataset , in order to show the independent replication , we tested association in the U4C ER-negative data excluding BPC3 , and found the Z-score for the TP53INP2 gene was -4 . 127 ( p = 3 . 67×10−5 ) . For the HP gene , the direction of association for 19 SNPs ( out of 20 ) were consistent between U4C and GAME-ON for ER-negative breast cancer risk , but only 2 SNPs were statistically significant at nominal 0 . 05 level in GAME-ON ( S2 Table ) . None of the SNPs were significantly associated with overall breast cancer risk in GAME-ON . In the gene-based analysis using GAME-ON data , the Z-score for overall breast cancer risk was 1 . 769 ( p = 0 . 077 ) and the Z-score for ER-negative breast cancer risk was 2 . 02 ( p = 0 . 043 ) . In addition , we tested this association in the U4C ER-negative data excluding BPC3 , and found the Z-score for the HP gene was 2 . 81 ( p = 5 . 1×10−3 ) .
In this gene-level expression-based genome-wide association analysis of five breast cancer GWAS datasets composed of individuals of diverse ancestry , we identified TP53INP2 ( 20q11 . 22 ) as gene with genetically-determined expression that is associated with ER-negative breast cancer . The gene-based analysis of aggregated eQTLs for a particular gene as an analysis unit can reduce the burden of multiple testing and provide a direction of association between expression of a specific gene and disease risk . We found that increased expression of TP53INP2 expression in whole blood was associated with a decrease in ER-negative breast cancer risk . In addition , we identified the HP gene in the 16q22 . 2 regions to have expression levels that are positively associated with ER- negative breast cancer . The TP53INP2 gene ( tumor protein p53-inducible nuclear protein 2 ) is 9150 base pairs long and codes for a 220 amino acid protein , which is a dual regulator of transcription and autophagy and is required for autophagosome formation and processing . One experimental study showed that overexpression of TP53INP2 severely attenuated proliferative and invasive capacity of melanoma cells , via p53 signaling and lysosomal pathways [34] . This inverse correlation between TP53INP2 expression and cancer proliferation is consistent with our finding that TP53INP2 expression inversely correlated with breast cancer risk . P53 is a transcription factor for TP53INP2 , and TP53 plays an important role in development of multiple cancers . Germline TP53 mutations cause Li-Fraumeni syndrome , characterized as a cluster of cancers including breast cancer [35] . Somatic TP53 mutation is a common event in ER-negative breast cancer [36] . As a downstream gene of p53 , TP53INP2 may affect breast cancer risk through p53 signaling pathway . Also , known as DOR ( diabetes- and obesity-regulated gene ) , TP53INP2 has been linked to obesity and diabetes [37] . TP53INP2 is also associated with triglycerides and cholesterol level . One experimental study found that dietary fat content influenced the expression of TP53INP2 expression in adipose and muscle tissues of mice [38] . This gene has been proposed to serve as a diagnostic biomarker for papillary thyroid carcinoma [39] but no study has linked its expression to cancer risk . Obesity has been convincingly correlated with breast cancer risk in numerous studies , although the relationship is complex and involves additional modifying factors [40 , 41] . Obesity has been associated with excess risk for breast cancer among postmenopausal women [42–46] , while in pre-menopausal women , obesity was associated with decreased breast cancer risk [40 , 43 , 47–49] . However , the underlying mechanisms for this association are still not fully understood . The identification of TP53INP2/DOR as breast cancer-related gene could provide novel insight on the mechanism for obesity-breast cancer relationship . In the 20q11 . 22 region , several other genes including MAP1LC3A , ITCH , and TRPC4AP were associated with ER-negative breast cancer risk . MAP1LC3A codes for a protein that is important in the autophagy process , and was found to be expressed at higher level in breast cancer tissues than in normal tissues [50] . E3 ubiquitin ligase ITCH plays a role in erythroid and lymphoid cell differentiation and immune response regulation , and ITCH was found to be important in the cross-talk between the Wnt and Hippo pathways in breast cancer development [51] . TRPC4AP is involved in Ca2+ signaling and is part of the ubiquitin ligase complex [52 , 53] . It is unclear which of these genes ( or their interactions ) play a role in breast cancer development , but the 20q11 . 22 locus is worthy of further investigation . Three of the six SNPs for TP53INP2 ( rs6060047 , rs11546155 , and rs1205339 ) are also shared by the genes MAP1LC3A and TRPC4AP . It is possible that the associations in these three genes are partly generated by the overlapped SNPs , which contribute to predicted expression levels of the three genes and , possibly , to the enrichment observed at this locus . The HP gene ( 16q22 . 2 ) is 6 , 491 base pairs long and codes for a 406 amino acid preprotein , which codes haptoglobin . Haptoglobin binds to hemoglobin to prevent iron loss during hemolysis . There are two allelic forms , Hp1 ( 83 residues ) and Hp2 ( 142 residues ) , which determine 3 major phenotypes [54] . Haptoglobin genotype has been linked to cardiocerebral outcomes among diabetic patients [55] . A small study found haptoglobin phenotypic polymorphism was associated with familial breast cancer [56] , but no studies have reported on the relationship between SNPs in this gene and breast cancer risk . Further larger studies could investigate the relationship between major HP genotype/phenotype ( HP1-1 , HP1-2 , and HP2-2 ) and breast cancer risk . The present study has several strengths , including its large sample size , diverse ancestry groups , a cross-replication approach , and a novel gene expression-based analysis method . The gene-level analysis method can combine eQTL SNPs in a biologically informative way to assess relationships between predicated gene expression and disease risk . Compared to SNP-based analysis , the gene-based analysis can gain power by reducing the multiple testing burden by about 100-fold and using external information on correlation between gene expression and SNPs from reference samples . In addition , this approach enables the detection of individual SNPs with weak effects on disease risk by leveraging combined effects of multiple SNPs on gene expression . For example , none of SNPs for TP53INP2 reached traditional genome-wide significance , but their aggregated effect via TP53INP2 expression was genome-wide significant . The gene-based method ( MetaXcan ) that we employed is an extension of the gene expression-based method ( PrediXcan ) [29] and allows the use of SNP-level summary statistics without the need to access individual-level genotype data [30] . The MetaXcan method has been shown to produce PrediXcan results accurately , and it is robust to ancestry mismatches between studies and reference/training populations [30] . With this property , we were able to use summary statistics from the GAME-ON consortium for external replication . Several limitations should be considered when interpreting the study findings . The gene expression-based association method relies on accurate prediction of gene transcript level from genotypes , i . e . identification of eQTLs , but eQTL identification depends on sample size of eQTL studies as well as tissue types . In the current study , we used the transcriptome prediction model that was developed using 922 RNA-seq samples from whole blood and genotype data [31] . Although it has been shown that models developed in whole blood were still useful for understanding diseases that affect other primary tissues [29] , we expect there to be a loss of power when studying non-blood diseases using whole blood eQTL data . As a sensitivity analysis , we performed the MetaXcan analysis using the prediction model from breast tissues of 183 donors of multiple ethnicities ( http://www . gtexportal . org ) . Only 4 , 308 genes had breast tissue specific eQTLs , and no eQTL was available for TP53INP2 , perhaps due to the small sample size . We found that DHODH ( P = 3 . 61×10−5 ) , ITCH ( P = 1 . 23×10−4 ) , and TRPC4AP ( P = 7 . 7x10-4 ) were among the top genes associated with ER-negative breast cancer risk , and TRPC4AP ( P = 1 . 68×10−5 ) and DHODH ( P = 1 . 12×10−4 ) among the top genes associated with the overall breast cancer risk using breast tissue eQTLs . In the enrichment analysis , we found that 7 ( 8 . 2% ) out of 85 genes that are close to known breast cancer susceptibility loci identified in previous GWAS were associated with ER-negative breast cancer and 6 ( 7 . 1% ) genes were associated with overall breast cancer risk; by contrast , of the 4223 genes away from previous GWAS loci , 199 ( 4 . 7% ) genes were associated with ER-negative breast cancer and 212 ( 5 . 0% ) genes were associated with overall breast cancer risk . Here , we have to consider the balance between tissue relevance and sample size in eQTL studies . Further investigations based on large , reliable eQTL datasets are desirable . In future studies , we will seek out larger samples of multi-ethnic breast tissue as training data to construct improved prediction models of gene expression and further investigate trans-ethnic associations for breast cancer . In conclusion , our study identified TP53INP2 and several other genes in the 20q11 . 22 region as potential susceptibility genes for ER-negative breast cancer using a novel gene-based analysis method that incorporates genetically determined gene expression . We demonstrated this gene-based method increases statistical power and may be helpful in searching for causal variants . Future studies need to determine whether the TP53INP2 gene is a true susceptibility gene for breast cancer and what are the underlying mechanisms for its association with ER-negative breast cancer .
The study was approved by the Institutional Review Board of the University of Chicago . The Epidemiology and Genomic Research Program within the National Cancer Institute launched a Challenge at the end of 2015 to inspire novel cross-disciplinary approaches to more fully decipher the genomic basis of breast cancer , called "Up For A Challenge ( U4C ) –Stimulating Innovation in Breast Cancer Genetic Epidemiology” . Several data sets were gathered and made available for use in dbGap ( https://www . ncbi . nlm . nih . gov/gap ) . Our study has two phases; the discovery phase included five U4C GWAS datasets ( Table 6 ) . Here , we refer them collectively as “U4C” data . These data were collected from three distinct ancestry groups . The BPC3 [16 , 18] and CGEMS study [15 , 20] were conducted in women of European ancestry . The ROOT [17] and AABC study [57] consisted of women of African ancestry . The SBCGS study was conducted in Chinese population [19] . For the analysis of overall breast cancer risk , we used four GWAS datasets: AABC , CGEMS , ROOT , and SBCGS . For the analysis of ER-negative breast cancer risk , we used datasets from AABC , BPC3 , ROOT , and SBCGS . All these dbGap datasets included imputed genotype data that were inferred based on reference haplotypes from the 1000 Genomes project . In the replication phase , we used the summary results from the meta-analysis of 11 breast cancer GWASs in the GAME-ON consortium ( http://gameon . dfci . harvard . edu ) . All participants were of European ancestry . The overall breast cancer analysis included 16 , 003 cases and 41 , 335 controls from 11 GWAS studies; The ER-negative breast cancer analysis included 4939 cases and 13128 controls from 7 GWAS studies . The dataset from one study ( BPC3; all ER-negative cases ) in GAME-ON consortium was also included the U4C datasets . Because only meta-analysis results were available from GAME-ON , we removed the BPC3 data from “U4C” dataset when we compared replication performance to avoid duplicate counting . Our gene level expression-based association analysis consists of three main steps . First , we conducted SNP-level genome-wide association tests and calculated summary statistics such log odds ratios and their standard errors . We used logistic regression model adjusting for eigenvectors from the principal component analysis and related covariates such as age . Genotypes were coded by an additive genetic model . Eigenvectors in principal component analysis were calculated using the method smartPCA , which is implemented in the software EIGENSOFT version 6 . 0 . 1 [58] . For the ROOT dataset , we adjusted for age , study sites , and the top 4 eigenvectors . For the AABC dataset , we adjusted for age , study sites , and top 10 eigenvectors . For CGEMS and SBCGS , we adjusted for age and the top three or two eigenvectors , respectively . The number of eigenvectors we adjusted for was chosen according to published papers from these GWASs [17 , 57] , as well as their association with case-control status . The logistic regression models were fit using software Mach2dat ( http://www . unc . edu/~yunmli/software . html ) or SNPtest [59] , depending on format of the datasets; the Mach2dat software was used for CGEMS and SBCGS and SNPtest was used for ROOT and AABC . For BPC3 , the GWAS summary statistics for ER-negative breast cancer have been pre-calculated in the dbGap release , so we used them directly . Second , we applied the gene level association method , MetaXcan [30] ( https://github . com/hakyimlab/MetaXcan ) , to each of the datasets listed in Table 6 . MetaXcan is an extension of the method PrediXcan [29] , which uses an additive genetic model to estimate the component of gene expression determined by an individual’s genetic profile and then identifies likely causal genes by computing the correlations between genetically predicted gene expression levels and disease phenotypes . MetaXcan infers the results of PrediXcan using summary statistics from GWAS , which are much more readily accessible than individual level data . In our study , as input files for MetaXcan , we used summary statistics from SNP-based analysis of each dataset obtained in step one . In addition , we used the whole blood genetic prediction model of transcriptome levels trained in the DGN data [31] , which can be downloaded from http://predictdb . hakyimlab . orghttps://s3 . amazonaws . com/predictdb/DGN-HapMap-2015/ . The DGN data provides a large reference sample of 922 individuals with both genome-wide genotype data and RNA sequencing data . The model trained in the DGN data can be useful in estimating gene expression levels and has been successfully applied to the Wellcome Trust Case Control Consortium ( WTCCC ) data in identifying genes associated with five complex diseases [29] . The DGN prediction model includes a ) weights for predicting gene expression using genotypes and b ) covariance of the SNPs that takes into account linkage disequilibrium . We tested the association between predicted expression levels of 11 , 536 genes for each of the two phenotypes , overall and ER-negative breast cancer risk , using the MetaXcan software . To construct the prediction model of expression levels using the DGN data , MetaXcan used SNPs with minor allele frequencies ( MAFs ) >0 . 05 . When MetaXcan was applied to the breast cancer GWAS data , only SNPs with MAFs >0 . 05 were used . We also looked up genes within 250 kb of the 93 breast cancer susceptibility loci identified in previous GWAS [2–4 , 17 , 32] . Third , we conducted meta-analysis to combine results from MetaXcan analyses for different datasets . The method described by Willer et al . with sample size as meta-analysis weight [60] was used . We also conducted SNP-level meta-analysis using a fixed effect model , as implemented in the software METAL ( http://genome . sph . umich . edu/wiki/METAL ) . False discovery rates were calculated using the Benjamini and Hochberg method [61] . For genes identified in the discovery phase using the U4C datasets , we conducted replication analysis using GAME-ON summary results using the same methods described above . For each top variant and gene identified in this study , we used HaploReg [33] and USCS Genome Browser to explore functional annotations of noncoding variants . Chromatin states ( promoters and enhancers ) , variant effect on regulatory motifs , and protein binding sites were assessed from available data from the ENCODE [62] and Roadmap Epigenomics Consortium [63] . Data from normal mammary epithelial cells ( HMEC , MYO , vMHEC ) were emphasized . | Although individual genetic variant-based genome-wide association studies have greatly increased our understanding of the genetic susceptibility to breast cancer , the genetic variants identified to date account for a relatively small proportion of the heritability . Shifting the focus of analysis from individual genetic variants to genes or gene sets could lead to the identification of novel genes involved in breast cancer risk . Here , we take advantage of a recently developed gene-level expression-based association method MetaXcan to examine the association of genetically-predicted expression levels for 11 , 536 genes across the human genome with breast cancer risk . The MetaXcan method uses external information on the effects of genetic variants on gene expression . We show that the TP53INP2 gene on human chromosome 20 is significantly associated with estrogen-receptor negative breast cancer ( P = 5 . 35×10−7 , Bonferroni threshold = 4 . 33×10−6 ) . The association is consistent across analyses of four datasets , representing European , African and Asian ancestry populations . As a downstream gene of p53 , TP53INP2 may affect breast cancer risk through p53 signaling pathway . Furthermore , TP53INP2 , also known as DOR ( Diabetes And Obesity-Regulated Gene ) , has been linked to obesity and diabetes , suggesting a novel biological pathway for the known association between obesity and breast cancer risk . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"genome-wide",
"association",
"studies",
"body",
"weight",
"medicine",
"and",
"health",
"sciences",
"breast",
"tumors",
"body",
"fluids",
"cancers",
"and",
"neoplasms",
"oncology",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"physiological",
"parameters",
"genome",
"analysis",
"obesity",
"research",
"and",
"analysis",
"methods",
"mathematical",
"and",
"statistical",
"techniques",
"gene",
"expression",
"breast",
"cancer",
"statistical",
"methods",
"genetic",
"loci",
"blood",
"anatomy",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"gene",
"prediction",
"computational",
"biology",
"human",
"genetics"
]
| 2017 | Trans-ethnic predicted expression genome-wide association analysis identifies a gene for estrogen receptor-negative breast cancer |
Third-generation cephalosporins are a class of β-lactam antibiotics that are often used for the treatment of human infections caused by Gram-negative bacteria , especially Escherichia coli . Worryingly , the incidence of human infections caused by third-generation cephalosporin-resistant E . coli is increasing worldwide . Recent studies have suggested that these E . coli strains , and their antibiotic resistance genes , can spread from food-producing animals , via the food-chain , to humans . However , these studies used traditional typing methods , which may not have provided sufficient resolution to reliably assess the relatedness of these strains . We therefore used whole-genome sequencing ( WGS ) to study the relatedness of cephalosporin-resistant E . coli from humans , chicken meat , poultry and pigs . One strain collection included pairs of human and poultry-associated strains that had previously been considered to be identical based on Multi-Locus Sequence Typing , plasmid typing and antibiotic resistance gene sequencing . The second collection included isolates from farmers and their pigs . WGS analysis revealed considerable heterogeneity between human and poultry-associated isolates . The most closely related pairs of strains from both sources carried 1263 Single-Nucleotide Polymorphisms ( SNPs ) per Mbp core genome . In contrast , epidemiologically linked strains from humans and pigs differed by only 1 . 8 SNPs per Mbp core genome . WGS-based plasmid reconstructions revealed three distinct plasmid lineages ( IncI1- and IncK-type ) that carried cephalosporin resistance genes of the Extended-Spectrum Beta-Lactamase ( ESBL ) - and AmpC-types . The plasmid backbones within each lineage were virtually identical and were shared by genetically unrelated human and animal isolates . Plasmid reconstructions from short-read sequencing data were validated by long-read DNA sequencing for two strains . Our findings failed to demonstrate evidence for recent clonal transmission of cephalosporin-resistant E . coli strains from poultry to humans , as has been suggested based on traditional , low-resolution typing methods . Instead , our data suggest that cephalosporin resistance genes are mainly disseminated in animals and humans via distinct plasmids .
Antibiotic resistance among opportunistic pathogens is rapidly rising globally , hampering treatment of infections and increasing morbidity , mortality and health care costs [1] , [2] . Of particular concern is the increased incidence of infections caused by Escherichia coli isolates producing extended-spectrum β-lactamases ( ESBLs ) , which has rendered the use of third generation cephalosporins increasingly ineffective against this pathogen [3] . During the 1990s , the most commonly encountered ESBL genes were blaTEM and blaSHV , and their spread occurred mainly through cross-transmission in hospitals . However , the epidemiology of ESBL-producing E . coli has changed . Nowadays , the most prevalent ESBL gene type is blaCTX-M [4] and infections with ESBL-producing E . coli also occur in the community [5] , [6] . The intestinal tracts of mammals and birds are important reservoirs for ESBL-producing E . coli [7] , but it is unclear to what extent these bacteria can spread to humans . Food may be an important source , since ESBL genes have been detected in food-producing animals , especially poultry [8] , [9] , and on retail meat [10] . The presence of ESBL-producing bacteria in food has been attributed to widespread use of antimicrobials , including third generation cephalosporins , in industrial farming practices [11] . In The Netherlands , antibiotic use and prevalence of antibiotic resistance in humans are among the lowest in Europe [12] , whereas antibiotic use in food-producing animals ranks among the highest in Europe [13] . These circumstances render The Netherlands particularly suitable to study the transfer of third-generation cephalosporin-resistant bacteria through the food-chain . Recent studies performed in The Netherlands suggested clonal transfer of ESBL-producing E . coli from poultry to humans [14]–[16] . However , these interpretations were based on typing methods that target a limited number of genes , and which may not have provided sufficient resolution to accurately monitor the epidemiology of pathogens [17] . In this study , we have therefore sequenced 28 ESBL-producing and four ESBL-negative E . coli strains that had previously been collected from humans , poultry , retail chicken meat and pigs and tested whether previous claims on the relationship between strains from different reservoirs could be confirmed at the whole-genome sequence level . Furthermore , we investigated the relatedness of cephalosporin resistance gene-carrying plasmids , which were derived from different backgrounds and reservoirs , at the genomic level .
We assessed the relatedness of ESBL-producing E . coli from humans , animals and food by using Whole-Genome Sequencing ( WGS ) . The genomes of 32 , mostly ESBL-producing , E . coli strains isolated in The Netherlands in the period 2006–2011 were sequenced ( Table 1 ) . One set of isolates ( n = 24 ) included five pairs of human and poultry-associated strains that had previously been found indistinguishable based on Multi Locus Sequence Typing ( MLST ) , plasmid typing ( pMLST ) and ESBL gene sequencing [15] , [18] . This set also included 11 human and poultry-associated isolates that carried an AmpC-type β-lactamase gene on an IncK plasmid [18] . The second set of isolates contained eight ESBL-producing strains that were isolated from pigs ( n = 4 ) and their farmers ( n = 4 ) ( Table 1 ) . Illumina sequencing yielded draft genomes with an average assembly size of 5 . 2 Mbp ( ±0 . 17 Mbp ) , consisting of an average number of 133 scaffolds ( ±41 ) of size ≥500 bp and a mean N50 of 153 kbp ( ±47 . 9 kbp ) ( S1 Table ) . WGS-based MLST and ESBL gene analysis provided good agreement with previous typing data . Previously obtained MLST profiles and WGS-based MLST profiles were in complete agreement with each other . Although ESBL genes had previously been detected by both microarray-based methods and Sanger sequencing [15] , the previously typed ESBL genes of four ( out of 28 ) strains were absent from their assembled genomes . In three of these cases ( strains 681 , 320 and 38 . 34 ) , we detected a blaTEM-1 or blaTEM-20 gene in the assembled genome , whereas a blaTEM-52 gene should have been found according to the typing data . Mapping the Illumina reads of these strains against their own assemblies showed that the assembled blaTEM genes contained several ambiguous positions pointing to the presence of more than one type of blaTEM gene ( most likely a combination of blaTEM-1 and blaTEM-52 ) in these strains ( S2 Table ) . In comparison , no ambiguous positions were found in the assembled blaTEM genes of other strains using the same mapping approach . In addition , the relative coverage of the assembled blaTEM genes of strains 681 , 320 and 38 . 34 was higher than that of the assembled blaTEM genes of other strains ( S2 Table ) . These findings suggested that strains 681 , 320 and 38 . 34 contain multiple nearly identical blaTEM genes ( i . e . blaTEM-1 and blaTEM-52 ) that hampered the correct assembly of these genes . The fourth inconsistency between WGS and typing data was the absence of blaCTX-M-1 from the assembly of strain 435 . Mapping the reads of strain 435 against the blaCTX-M-1 gene sequence did suggest the presence of this gene in the WGS data , but with a depth of around 1/10th the average genomic sequencing depth . Possible explanations include a relatively poor isolation efficiency of the blaCTX-M-1-carrying plasmid and/or the loss of this plasmid from the bacterial cells during culturing in the absence of antibiotics . The previous AmpC typing data [18] and our WGS data were in complete agreement . To assess the phylogenetic context of the sequenced strains within the genus Escherichia and Shigella , we used publicly available genome sequences of Escherichia ( n = 126 ) and Shigella ( n = 12 ) strains . Based on COG assignments , we identified 215 core proteins in the 170 analysed genomes , from which a concatenated core genome alignment of 170461 bp was built . A phylogenetic tree based on the 18169 variable positions in this alignment confirmed previous clustering based around phylogroups A , B1 , B2 , D , E and F ( Fig . 1 ) [19] . The sequenced strains clustered together in accordance with their ST . Strains did not cluster based on isolation source , year , plasmid or ESBL gene . The ESBL-producing strains were spread throughout the tree , indicating that acquisition of ESBLs arises in different E . coli genetic backgrounds and has occurred multiple times during evolution ( Fig . 1 ) . There were four clusters of ESBL-producing strains isolated from humans and animals/meat ( clusters I–IV , Fig . 1 ) . Cluster I contained human and pig isolates from two pig farms , with strains from farm A being particularly closely related . The other three clusters contained the five pairs of human and chicken isolates that had previously been considered indistinguishable based on traditional typing methods [15] . Among the five pairs of human and chicken isolates , the most closely related pairs were in cluster IV . The COG-based core genome alignment showed 171 SNPs between these strains , corresponding to 1003 SNPs/Mbp . To better elucidate the minimum number of SNPs between human and chicken isolates , we performed a core genome analysis using OrthoMCL [20] on the strains in cluster IV . For comparison , ten clonal O104:H4 strains from the 2011 German EHEC outbreak [21] and the four strains from pig farm A ( cluster I ) were included in this analysis ( Fig . 1 ) . We identified 3574 core proteins in this dataset translating to a concatenated nucleotide alignment of 3 . 34 Mbp . Within cluster IV there were 4216 SNPs between the most closely related isolates , corresponding to 1263 SNPs/Mbp . In contrast , only 0–6 SNPs ( 0–1 . 8 SNPs/Mbp ) were found between any two strains in the German EHEC outbreak and only 6 SNPs were found between farmer isolate FAH2 and any of its two related pig isolates , suggesting recent clonal transmission of E . coli between pig and human in farm A ( Fig . 2 ) . Given an estimated E . coli mutation rate of 2 . 3×10−7 to 3 . 0×10−6 substitutions per site per year [21] , [22] and an average E . coli genome size of 5 . 2 Mbp , the number of SNPs ( 1263/Mbp ) between the two most closely related human and chicken isolates largely exceeded the number of 3–41 SNPs that is expected to arise in 2 . 6 years ( the difference in isolation dates between both strains , Table 1 ) . Even if 10% of the detected SNPs were due to recombination , which is considerably more than the reported upper limit for recombinant DNA ( ∼3 . 5% ) in E . coli [19] , the number of SNPs due to mutation would exceed the expected maximum number of SNPs in case of recent clonal transmission . As the genetic distance between all other pairs of human and poultry isolates was even larger , our findings do not support a scenario of recent clonal transmission of ESBL-producing E . coli strains between humans and poultry . To investigate the possibility of horizontal spread of ESBLs via plasmids , we employed a Plasmid Constellation Networks ( PLACNET ) approach to reconstruct plasmids from WGS data [23] . Application of this approach resulted in the reconstruction of 147 plasmids ( average of 4 . 6±2 . 1 plasmids per strain ) , with plasmid sizes ranging from 1 . 1 kbp to 290 . 4 kbp ( Table 2 ) . The plasmid sizes showed a trimodal distribution ( Fig . 3 ) that was similar to the distribution previously reported for plasmids from a wide range of bacterial taxa [24] . The median size of large ( conjugative ) plasmids was 93 . 6 kbp ( n = 91 ) . Small plasmids could be further subdivided into two groups: one with a median size of 5 . 9 kbp ( n = 41 ) , predominated by mobilizable plasmids ( i . e . containing MOB genes ) and one with a median size of 1 . 7 kbp ( n = 15 ) , predominated by non-mobilizable plasmids . Based on the classification of their MOB genes [25] and using a hierarchical clustering analysis of gene content ( Fig . 4 ) , reconstructed plasmids belonged to a limited number of plasmid families , of which the most abundant ones were IncF-MOBF12 ( n = 38; average size of 107 . 4±57 . 7 kbp ) and IncI1-MOBP12 ( n = 26; average size of 95 . 7±20 . 0 kbp ) . Other abundant families included MOBP5 ( n = 25 ) , IncK ( n = 12 ) and MOBQ ( n = 11 ) . Finally , there were 18 , mostly small-sized , plasmids ( median size of 1 . 6 kbp; range of 1 . 1–106 . 3 kbp ) that were scattered throughout the dendrogram and could not be clearly subdivided into any family . A comparison between previous typing data and the PLACNET reconstructions showed that both data types were in excellent agreement with each other . First of all , the 11 strains that were previously found to contain an IncI1 plasmid were also found to contain such a plasmid using PLACNET . The sizes of these 11 reconstructed plasmids ( average size of 92 . 7 kbp±5 . 7 kbp ) were also in agreement with their previously estimated sizes on the basis of gel electrophoresis ( average size of 97 . 7 kbp±3 . 8 kbp ) [15] . Furthermore , the reconstructed plasmids for ten of these 11 strains had exactly the same ST as was previously found using pMLST . The only inconsistency was found for strain 38 . 34 , which should contain an IncI1/ST10 plasmid according to pMLST , whereas we reconstructed an IncI1/ST36 plasmid . However , IncI1/ST10 and IncI1/ST36 are single locus variants that differ by only one SNP ( http://pubmlst . org/plasmid/ ) , indicating that this inconsistency was not a result of PLACNET , but was likely due to typing errors . Of the 11 strains that had previously been found to contain an IncK plasmid , ten were also found to contain such a plasmid using PLACNET , the only exception being strain 1047 . We also examined to what extent we were able to correctly connect ESBL and AmpC genes to reconstructed plasmids . Of the 28 previously typed ESBL genes , 24 were correctly identified in their genomes ( Table 1 ) and among these , 15 were connected to a reconstructed plasmid . Four of the remaining nine unconnected ESBL genes ( blaCTX-M-1 in strains 1350 , 1365 , 1047 and 38 . 52 ) should have been connected to an IncI1 plasmid according to previous typing data ( Tables 1–2 ) . The reason that these ESBL genes remained unassigned was because they were located on relatively small scaffolds ( average size of 6 . 6 kbp ) that did not contain enough genetic information to unequivocally match them to a single plasmid using our reference database . For the 15 cases where we were able to connect an ESBL gene to a reconstructed plasmid , typing data indicating where the ESBL gene should be located was available for four cases ( strains 148 , 897 , 38 . 16 and 38 . 27 ) and for all these cases we had connected the ESBL gene ( blaCTX-M-1 ) to the correct plasmid ( IncI1/ST7 ) ( Tables 1–2 ) . Of the 11 AmpC ( blaCMY-2 ) genes , ten were connected to their correct plasmid ( IncK ) . The only exception was found again for strain 1047 for which we could not reconstruct an IncK plasmid ( Table 2 ) . The above findings show that PLACNET worked efficiently to assemble plasmids from WGS data , although the assignment of small scaffolds to plasmids can be problematic , as is illustrated above by the ESBL genes that were not linked to a specific reconstructed plasmid ( see also discussion below and in [23] ) . Fifteen ESBL genes were connected to a reconstructed plasmid , of which 13 were connected to an IncI1 plasmid . Frequently ( eight out of 13 ) , these IncI1 plasmids were also unequivocally linked to other antibiotic resistance genes , such as sul , dfrA , aadA or tet . We also found IncK plasmids that were commonly ( ten out of 12 plasmids ) associated with the AmpC β-lactamase-encoding gene blaCMY-2 ( Fig . 4 ) . As IncI1 and IncK were the only plasmid families that included reconstructed ESBL-/AmpC-containing plasmids in strains from both humans and animals/meat , we further investigated their potential role in the transfer of resistance genes through the food-chain . To this aim we built a gene content-based dendrogram that also included closely related and publicly available plasmid sequences . In the resulting dendrogram , all reconstructed ESBL-containing IncI1 plasmids , except the blaSHV-12-carrying plasmids p1A_2 and p9B_1 , clustered into one specific branch that did not contain any other previously sequenced plasmid ( Fig . 5 ) . This branch also contained 12 of the 13 reconstructed IncI1 plasmids that did not include an ESBL gene . Similarly , all of the reconstructed IncK plasmids , except p87A_5 , clustered into one specific branch that did not include any previously sequenced plasmid . These findings suggest the existence of IncI1 and IncK plasmids with a genetic profile distinct from previously characterised plasmids . We did not find any single gene that unequivocally explained the formation of the IncK branch , pointing to a delicate configuration of genes that gives these plasmids their unique genetic profile . However , for the IncI1 branch , we found a characteristic shufflon-related gene ( UniProt P10487 ) that was present in all 26 reconstructed IncI1 plasmids , but which was absent from related IncI1 plasmids ( Fig . 5 ) . To further characterise the IncI1 and IncK resistance plasmids , phylogenetic trees were built from the sequences of the reconstructed plasmids and their closest plasmid relatives . For the IncI1 phylogenetic reconstruction , the 23 plasmids belonging to the specific IncI1 branch as well as 27 related plasmids were included . An OrthoMCL analysis of these plasmids resulted in 8 core proteins ( S3 Table ) , corresponding to a concatenated nucleotide alignment of 8 . 6 kbp , including 763 variable positions . In the phylogenetic tree built from these variable positions the reconstructed IncI1 plasmids were assigned to four distinct branches ( Fig . 6A ) , each of which also contained previously characterised plasmids . However , the reconstructed plasmids within each branch were always more similar to each other than to any of these previously characterised plasmids . Two of the four branches , corresponding to IncI1/ST3 and IncI1/ST7 , contained reconstructed ESBL-harbouring plasmids from both humans and animals or meat . Further rounds of OrthoMCL analyses showed that the reconstructed plasmids within each of these two sets were highly similar to each other: a maximum of only four SNPs ( all attributable to p53C_2 ) was found in the 40 kbp plasmid core of the IncI1/ST3 subset , whereas no SNPs were found in the almost 50 kbp plasmid core of the IncI1/ST7 subset ( Fig . 6A ) . Similarly , a subset of the blaCMY-2-carrying IncK plasmids contained a plasmid core of almost 37 kbp with a maximum of 27 SNPs ( Fig . 6B ) , which were mostly attributable to p435_1 . Leaving out p435_1 from the comparisons revealed a maximum of only seven SNPs . These data strongly support the existence of ESBL-associated IncI1 and AmpC-associated IncK plasmids that have spread through phylogenetically distinct E . coli populations , possibly contributing to the dissemination of ESBLs and AmpC-type β-lactamases through the food-chain . To validate the conclusions drawn from the PLACNET reconstructions , we sequenced two strains ( 53C and FAP1 ) using long-read DNA sequencing technology ( Pacific Biosciences ) . Strain 53C was selected because it has both an IncI1 and an IncK plasmid , carrying blaCTX-M-1 and blaCMY-2 , respectively . Strain FAP1 was selected because it contained an IncI1 plasmid of the same lineage as the one in strain 53C ( Fig . 6A ) . The total amount of reconstructed plasmid sequence for strains 53C and FAP1 was 338 kbp and 319 kbp , respectively ( Table 2 ) . Genomes were assembled to an average depth of 66 . 7- and 77 . 0-fold , respectively , resulting in 11 contigs for strain 53C and five contigs for strain FAP1 ( S4 Table ) . Inspection of the contig sequences showed the presence of four large plasmids in both strains . These were assigned to Inc groups F , I1 ( carrying blaCTX-M-1 ) , I2 , and K ( carrying blaCMY-2 ) in strain 53C and F , I1 ( carrying blaCTX-M-1 ) , and I2 , in strain FAP1 . A single plasmid in strain FAP1 could not be assigned to an Inc group . Except for the IncI1 plasmid of FAP1 , all plasmid contigs could be circularized ( S4 Table ) . The plasmid content was in agreement with our reconstructions , except for two inconsistencies in strain FAP1: ( i ) PLACNET did not assign a blaCTX-M-1 gene to its IncI1 plasmid , and ( ii ) PLACNET reconstructed two IncF plasmids . Blast analysis of both reconstructed IncF plasmids against the FAP1 long-read assembly suggested that they should indeed have been merged into one single plasmid . The reason for this incorrect prediction by PLACNET is unclear , but in the constellation network the two plasmids were relatively far away from each other , suggesting that the IncF plasmid in FAP1 is a fusion between previously observed IncF plasmids present in the reference database . These data show that caution must be taken in case PLACNET predicts multiple plasmids of the same Inc group in one strain . For the remaining plasmids , blast analysis showed that the precision rate of PLACNET was high , ranging from 97–100% ( Table 3 ) . Also in terms of sensitivity , PLACNET performed well being able to recover 72 . 1–99 . 7% of the plasmids ( Table 3 ) . The plasmid regions that were not reconstructed by PLACNET mostly aligned with small scaffolds ( average size of 2 . 0±1 . 9 kbp , n = 33 ) in the assemblies built from Illumina short-read data , which indicates that these regions are difficult to assemble . Notably , these small scaffolds encoded many mobile element- , phage- , transposon- and integrase-associated proteins ( 29 . 7% of all predicted proteins in these scaffolds ) as compared to the correctly assigned scaffolds , where only 6 . 7% of the proteins had these predicted roles . These observations are in line with results obtained from the PLACNET validation analyses described in [23] and show that PLACNET efficiently reconstructs plasmids from WGS data . Finally , the PLACNET-based prediction that both IncI1 plasmids from strains 53C and FAP1 are highly similar ( Fig . 6A ) was confirmed by aligning the two complete IncI1 plasmid sequences assembled from the long-read sequencing data . Filtering out repetitively aligning regions resulted in a pairwise alignment of 94 . 8 kbp containing only 4 SNPs . These data further substantiate our conclusions regarding highly successful plasmid lineages disseminating cephalosporin resistance .
We assessed the epidemiology of ESBL-producing E . coli from humans , animals and food using WGS . Our findings strongly suggest the existence of highly successful ESBL-carrying plasmids facilitating transmission of ESBL genes between different reservoirs . This has important implications for our understanding of the dynamics of the spread of ESBL genes and for evaluating control measures . Several strains that were sequenced in this study and which originated from humans and poultry had previously been considered indistinguishable based on MLST , plasmid and ESBL gene typing , suggesting clonal transfer of these strains through the food-chain , to humans [15] . The claim that ESBL-producing E . coli strains from humans and poultry are frequently identical was also made in other studies that made use of traditional sequence-based typing methods [14] , [16] . However , as has been demonstrated for different bacterial pathogens and in varied contexts , especially bacterial outbreak investigations , WGS provides superior resolution over traditional typing methods in terms of ruling in and out epidemiological connections between strains [26]–[28] . Similarly , we demonstrate that conclusions on the clonal spread of ESBL-producing E . coli through the food-chain cannot realistically be drawn on the basis of traditional sequence-based typing methods , due to their insufficient discriminative power . More specifically , we found that none of the five pairs of human and poultry-associated isolates , previously typed as indistinguishable , were particularly closely related . The most similar pair of isolates differed by 1263 SNPs/Mbp compared to a difference of 1 . 8 SNPs/Mbp for known/expected clonally related isolates . Hence , inferences from classical typing-based studies regarding the extent of transfer of ESBL-producing E . coli strains from animals via food to humans and the burden of disease and mortality due to the use of third-generation cephalosporins in food production must be considered as highly speculative [11] . In fact , our findings strongly suggest that distinct plasmids disproportionately contribute to the spread of antibiotic resistance between different reservoirs . We have demonstrated the existence of highly similar cephalosporin resistance-encoding IncI1/ST3 ( 40 . 0 kbp core , 0–4 SNPs ) , IncI1/ST7 ( 49 . 7 kbp core , 0 SNPs ) , and IncK ( 36 . 9 kbp core , 0–27 SNPs ) plasmids in different reservoirs . Reconstructed blaCTX-M-1-carrying IncI1/ST3 plasmids were found in one human and two poultry isolates , blaCTX-M-1-carrying IncI1/ST7 plasmids were found in three human , two poultry , and one pig isolate; and blaCMY-2-carrying IncK plasmids were found in five human and four poultry isolates . The isolates carrying these plasmids belonged to evolutionarily distinct backgrounds ( IncI1 in phylogroups A , B1 and B2; IncK in phylogroups A , B1 , B2 , D and F ) , suggesting that these plasmids efficiently spread through E . coli populations and play an important role in the dissemination of ESBL and AmpC-type β-lactamases between different reservoirs . Based on their genetic content , the IncI1 and IncK plasmids in our dataset clustered into specific sub-branches that did not contain any previously characterised plasmid . However , phylogenetic analyses revealed that these sub-branches could be split into evolutionarily distinct plasmids , some of them being distantly related to previously sequenced plasmids . These findings suggest that evolutionarily distinct plasmids have been accumulating genes from the same genetic reservoir , resulting in plasmids with a similar genetic inventory . The reconstructed IncI1 plasmids all harboured a characteristic shufflon-related gene that was absent from previously characterised IncI1 plasmids . Shufflons are site-specific recombination systems that produce variable C-terminal extensions of the PilV adhesin , resulting in variations of recipient ability in IncI1 plasmid mating [29] . Whether this shufflon explains the promiscuous nature of ESBL-carrying IncI1 plasmids remains to be determined . One important question is to what extent the IncI1 and IncK resistance plasmids found in this study have spread beyond The Netherlands . Given the trees in Fig . 6 , it is clear that currently available plasmid sequences in public databases do not contain any plasmids that are particularly closely related to our reconstructed IncI1 and IncK plasmids . The pMLST repository ( http://pubmlst . org/plasmid/ ) shows that blaCTX-M-1-carrying IncI1/ST3 plasmids have been isolated from six different European countries , whereas blaCTX-M-1-carrying IncI1/ST7 plasmids have until now been isolated only from The Netherlands and Germany . The location of blaCMY-2 on an IncK plasmid , as found here , has only been occasionally reported before , in The Netherlands [30] , [31] , but also in Sweden [32] and Canada [33] . Future sequencing projects are needed to determine whether the previously identified plasmids isolated outside The Netherlands are closely related to those described here . We found that none of the human E . coli strains in our dataset were closely related to strains from poultry . In contrast , nine out of 17 human isolates ( 53% ) contained a blaCTX-M-1 or a blaCMY-2 gene located on plasmids that were highly similar to those found in poultry . These data cannot be interpreted to mean that clonal transfer of antibiotic resistant E . coli strains between poultry and humans does not occur , but rather that such transfer occurs less frequently than the transfer of resistance plasmids between both reservoirs . One drawback of our study is that we have used a relatively small sample size ( 32 strains ) . Future studies , using larger sample sizes , are needed in order to make more accurate estimates of the relative ( and absolute ) contributions of clonal versus plasmid transfer towards the spread of antibiotic resistance and the associated health-care burden . In addition , our study focuses on IncI1 and IncK plasmids . Future studies are needed that also focus on other plasmid families , such as IncF plasmids , which are commonly detected in E . coli from human infections and are associated with the dissemination of many virulence and antibiotic resistance determinants [34] , [35] . Conjugal transfer of plasmids carrying antibiotic resistance genes has been shown to frequently occur among Enterobacteriaceae in different environments , including milk , meat , and feces , even in the absence of antibiotic pressure [36] , [37] . Moreover , it has been shown that bla-carrying plasmids are readily transferred from invading Enterobacteriaceae to Enterobacteriaceae that are indigenous to the animal and human intestine and that the invading clone itself generally does not persist in the intestine [38] , [39] . Nonetheless , it is difficult to infer to what extent the reservoir of bla-type resistance genes in poultry contributes to the carriage of such genes by human E . coli strains . If successful plasmids are largely responsible for the rising prevalence of ESBL- and AmpC-producing E . coli in healthy humans , their emergence in poultry and humans may simply be a reflection of selection of strains carrying these plasmids due to antibiotic usage in human and veterinary medicine . A better understanding of the dynamics of ESBLs and other resistance genes in different hosts is needed to design effective control measures , both in the community and within health care settings . Our findings strongly suggest the occurrence of clonal transfer of ESBL-producing E . coli between pigs and pig farmers , which may well occur through direct contact or through aerosols . Whether such events represent a public health threat remains to be determined . The occurrence of transmission of ESBL-producing E . coli from poultry through the food-chain is less evident . The occurrence of highly-related plasmids that carry ESBL- and AmpC-type resistance genes among genotypically distinct E . coli strains from different sources is cause for concern because this suggests that plasmids can spread with relative ease between the different reservoirs and the spread of these plasmids may be exceedingly difficult to control . Clearly , there still remains an urgent need to minimize the use of third-generation cephalosporins in animal husbandry as this is an important selective pressure for the occurrence of ESBL- and AmpC-producing E . coli in animals raised for food production .
The genomes of 32 , mostly ESBL-producing , E . coli strains isolated from different reservoirs in The Netherlands in the period 2006–2011 , were sequenced . One set of isolates ( n = 24 ) has been studied previously using classical typing methods [15] , [18] . This set contained strains from human clinical infections ( n = 13 ) which had been obtained from geographically dispersed laboratories in The Netherlands , servicing secondary and tertiary care hospitals , general practitioners and long-term care facilities . Additional isolates were from chickens raised on production farms ( n = 4 ) and chicken retail meat ( n = 7 ) ( Table 1 ) . All 24 isolates were previously genotyped by MLST [40] ( http://mlst . warwick . ac . uk/mlst/dbs/Ecoli ) and plasmid characterization was previously performed using PCR-based replicon typing [31] , [41] and additional pMLST for IncI1 plasmids [42] , [43] ( http://pubmlst . org/plasmid/ ) . Detection of ESBL genes had been performed for all 24 strains using microarray analysis and gene sequencing [44] . In addition , detection of AmpC-type β-lactamase-encoding genes had been performed for 11 strains , using gene sequencing [18] . The association between ESBL/AmpC genes and plasmids was previously determined by both Southern blot hybridization and transformation [31] . Four non-ESBL-producing isolates were included as controls and were analysed for the carriage of plasmids that can incorporate ESBL genes via horizontal gene transfer . The second set of isolates contained eight ESBL-producing strains that had been isolated from three different pig farms in The Netherlands in 2011 ( Table 1 ) . These farm strains were part of a larger cohort that will be described in detail elsewhere ( Dohmen et al . , unpublished data ) . For one farm ( farm A ) , four strains were collected , two from different fecal pools of six unique pigs and two from the feces of different farmers . For each of the other two farms ( farms B and C ) , one strain was collected from a fecal pool of six pigs and one from the feces of a farmer . Detection of the ESBL ( blaCTX-M-1 ) gene was performed using a CTX-M-1 group-specific PCR and additional gene sequencing ( Dohmen et al . , unpublished data ) . Genomic DNA was isolated from cell pellets using the Ultraclean Microbial DNA isolation kit ( Mo Bio Laboratories , Inc . , Carlsbad , CA , USA ) according to the manufacturer's instructions . Strains were sequenced using Illumina HiSeq 2000 sequencing technology ( Illumina , Inc . , San Diego , CA , USA ) generating 90 bp paired-end reads from a library with an average insert size of 500 bp and a total amount of quality-filtered raw sequence of over 600 Mbp per strain . Quality filtering included the removal of duplicate reads and reads that contained ≥15 bp overlap with the adapter sequences . The corresponding paired-end reads were also removed in these cases . Reads were assembled de novo using SOAPdenovo v1 . 05 [45] . For each Illumina dataset , a range of different k-mer lengths ( 21–63 bp ) was empirically tested to obtain the assembly with the lowest number of scaffolds of size ≥500 bp . In cases where more than one assembly contained the lowest number of scaffolds , the parameters of choice to pick the best assembly were: the lowest number of contigs of size ≥200 bp , the highest N50 for the scaffolds , and the highest N50 for the contigs , in order of priority . Assembly statistics are reported in S1 Table . Two strains ( 53C and FAP1 , Table 1 ) were also sequenced on a Pacific Biosciences RS II instrument ( Pacific Biosciences , Inc . , Menlo Park , CA , USA ) . Libraries were prepared using the PacBio 20 kbp library preparation protocol . Size selection ( 5 kbp cut-off ) of the final libraries was performed using a BluePippin instrument ( Sage Science , Inc . , Beverly , MA , USA ) . Sequencing was performed using P4-C2 chemistry . Three and five SMRT cells were used for sequencing strains FAP1 and 53C , respectively , generating 159191 and 95263 reads and a total of 997 . 1 and 471 . 5 Mbp , respectively . Reads were assembled using HGAP v3 ( Pacific Biosciences , SMRT Analysis Software v2 . 2 . 0 ) . Minimus2 , part of the AMOS package [46] , was used to circularize contigs . The SMRT Analysis Software was used to map reads back to the contigs and correct sequences after circularization . Assembly statistics are reported in S4 Table . Publicly available sequence data were retrieved from GenBank ( ftp://ftp . ncbi . nih . gov/genomes/Bacteria and ftp://ftp . ncbi . nih . gov/genomes/Bacteria_DRAFT ) . Whole genome sequence data for 126 Escherichia and 12 Shigella species were downloaded in June 2012 , whereas sequence data for 4188 completely sequenced plasmids , 797 of them from Enterobacteriaceae , were downloaded in June 2013 . The strains that were sequenced in this study were annotated with RAST v4 . 0 [47] using default settings . Predicted proteins were assigned to Clusters of Orthologous Groups ( COG ) [48] as described previously [49] . On the basis of COG assignments , a core proteome was defined by ( i ) extracting , per analysed genome , all proteins with one or more COGs assignments and which represented the only protein in that given COG or combination of COGs and by ( ii ) selecting from those proteins the ones that occurred in all genomes analysed . Alternatively , in the cases where smaller genomic datasets were analysed ( see Results ) , core proteomes were determined by first subjecting all associated protein sequences to an all-vs-all blastp similarity search ( defaults settings , except for: -F ‘m S’; -e 1×10−5; -z [the total number of proteins used in the analysis] ) . Groups of orthologous proteins were determined from the blastp output using OrthoMCL v2 . 0 . 2 [20] . Orthologous groups with exactly one representative protein from each input genome were considered to be part of the core proteome . Core genome alignments were built as follows: for each group of orthologous proteins , the corresponding nucleotide sequences were extracted and aligned using Muscle v3 . 7 [50] , after which gaps were stripped from each alignment using trimAl v1 . 2 [51] . The resulting alignments were concatenated to yield a core genome alignment . Phylogenies were reconstructed by building maximum likelihood phylogenetic trees from the variable positions in core genome alignments using RAxML v7 . 2 . 8 [52] under the GTRCAT model . Confidence was inferred by running 100 or 1000 bootstrap replicates under the same model . Trees were mid-point rooted and visualised in MEGA v5 . 05 [53] . Bowtie2 [54] was used for mapping Illumina reads against scaffolded assemblies and gene sequences . MLST profiling of sequenced bacteria was performed using MLST v1 . 6 [55] . Pairwise large-scale nucleotide alignments were built using NUCmer v3 . 23 ( with –mum option ) , which is part of the MUMmer package [56] . Plasmid reconstructions were based on the Plasmid Constellation Networks ( PLACNET ) method of genome representation [23] . In short , for all genomes , a PLACNET representation that clusters all plasmid-associated contigs was built using ( i ) contig similarities with reference genomes , ( ii ) all possible contig linkages , and ( iii ) plasmid-specific relaxase and replication initiator genes . This information was implemented in a network , where genomic contigs , together with reference plasmid and genome sequences are shown as nodes . The nodes are linked by edges of homology and scaffolding information . As a result , contigs fall into clusters , the largest one being the chromosome and additional ones being plasmids . Manual curation of the resulting networks helped solving most of the remaining ambiguities . Reference data from GenBank contained 4188 plasmids and 2728 chromosomes . Contig similarity analysis was performed using megablast against these reference data . Contig homology edges were defined by the five best blast hits ( e-value <1×10−20 ) . Scaffolds were determined by mapping all reads against contigs using Bowtie2 [54] , and allocating as scaffold links all discordant paired-end reads that matched two different contigs . To provide additional evidence for the plasmid origin of a cluster , a blastp search against in-house databases containing plasmid-specific relaxases and replication initiator proteins was performed . Contigs encoding these proteins were tagged in the PLACNET . Plasmid Neighbour-Joining dendrograms were built based on previously described methodologies [57] using CD-HIT [58] to construct protein profiles and the Jaccard formula to calculate distance metrics between profiles . PLACNET results were validated as follows: for the two strains 53C and FAP1 , the scaffolds assigned to the reconstructed plasmids were queried using megablast against the assemblies resulting from Pacific Biosciences ( PacBio ) sequencing . The best blast hit ( e-value ≤1×10−10 ) was inspected to assess whether the scaffolds had been assigned to the correct plasmid . For each reconstructed plasmid the PLACNET precision rate was calculated by the formula [ ( assembly size of all correctly assigned scaffolds/assembly size of all correctly+incorrectly assigned scaffolds ) ×100%] . To assess PLACNET sensitivity ( the percentage of each plasmid sequence size that was recovered ) blast hits against the corresponding PacBio plasmids were collected ( e-value ≤1×10−10 , minimum of 250 identical residues ) . The sensitivity rate was calculated by the formula [ ( total nr . of non-overlapping aligning residues found by blast/size of the PacBio plasmid ) ×100%] . All sequence data have been deposited at DDBJ/EMBL/GenBank . Accession numbers for the Illumina sequence data are listed in Table 1 . Pacific Biosciences sequence data have been deposited with accession numbers PRJNA260957 for strain 53C and PRJNA260958 for strain FAP1 . | The rapid global rise of infections caused by Escherichia coli that are resistant to clinically relevant antimicrobials , including third-generation cephalosporins , is cause for concern . The intestinal tract of livestock , in particular poultry , is an important reservoir for drug resistant E . coli , but it is unknown to what extent these bacteria can spread to humans . Food is thought to be an important source because drug-resistant E . coli have been detected in animals raised for meat consumption and in meat products . Previous studies that used traditional , low-resolution , genetic typing methods found that drug resistant E . coli present in humans and poultry were indistinguishable from each other , suggesting dissemination of these bacteria through the food-chain to humans . However , by applying high-resolution , whole-genome sequencing methods , we did not find evidence for such transmission of bacteria through the food-chain . Instead , by employing a novel approach for the reconstruction of mobile genetic elements from whole-genome sequence data , we discovered that genetically unrelated E . coli isolates from both humans and animal sources carried nearly identical plasmids that encode third-generation cephalosporin resistance determinants . Our data suggest that cephalosporin resistance is mainly disseminated via the transfer of mobile genetic elements between animals and humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"bacteriology",
"antimicrobials",
"gram",
"negative",
"bacteria",
"medicine",
"and",
"health",
"sciences",
"genome",
"evolution",
"evolutionary",
"biology",
"food",
"chains",
"infectious",
"disease",
"epidemiology",
"plasmids",
"population",
"genetics",
"microbiology",
"gene",
"transfer",
"antibiotic",
"resistance",
"plant",
"science",
"antibiotics",
"genome",
"analysis",
"genetic",
"elements",
"forms",
"of",
"dna",
"plant",
"pathology",
"dna",
"microbial",
"genomics",
"veterinary",
"science",
"bacterial",
"genomics",
"antimicrobial",
"resistance",
"genetic",
"polymorphism",
"veterinary",
"microbiology",
"medical",
"microbiology",
"epidemiology",
"molecular",
"evolution",
"comparative",
"genomics",
"biochemistry",
"ecology",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"genomics",
"mobile",
"genetic",
"elements",
"computational",
"biology"
]
| 2014 | Dissemination of Cephalosporin Resistance Genes between Escherichia coli Strains from Farm Animals and Humans by Specific Plasmid Lineages |
During a blood meal , Lutzomyia intermedia sand flies transmit Leishmania braziliensis , a parasite causing tegumentary leishmaniasis . In experimental leishmaniasis , pre-exposure to saliva of most blood-feeding sand flies results in parasite establishment in absence of any skin damages in mice challenged with dermotropic Leishmania species together with saliva . In contrast , pre-immunization with Lu . intermedia salivary gland sonicate ( SGS ) results in enhanced skin inflammatory exacerbation upon co-inoculation of Lu . intermedia SGS and L . braziliensis . These data highlight potential unique features of both L . braziliensis and Lu . intermedia . In this study , we investigated the genes modulated by Lu . intermedia SGS immunization to understand their potential impact on the subsequent cutaneous immune response following inoculation of both SGS and L . braziliensis . The cellular recruitment and global gene expression profile was analyzed in mice repeatedly inoculated or not with Lu . intermedia . Microarray gene analysis revealed the upregulation of a distinct set of IFN-inducible genes , an immune signature not seen to the same extent in control animals . Of note this INF-inducible gene set was not induced in SGS pre-immunized mice subsequently co-inoculated with SGS and L . braziliensis . These data suggest the parasite prevented the upregulation of this Lu . intermedia saliva-related immune signature . The presence of these IFN-inducible genes was further analyzed in peripheral blood mononuclear cells ( PBMCs ) sampled from uninfected human individuals living in a L . braziliensis-endemic region of Brazil thus regularly exposed to Lu . intermedia bites . PBMCs were cultured in presence or absence of Lu . intermedia SGS . Using qRT-PCR we established that the IFN-inducible genes induced in the skin of SGS pre-immunized mice , were also upregulated by SGS in PBMCs from human individuals regularly exposed to Lu . intermedia bites , but not in PBMCs of control subjects . These data demonstrate that repeated exposure to Lu . intermedia SGS induces the expression of potentially host-protective IFN-inducible genes .
Leishmania protozoan parasites induce a broad spectrum of disease including cutaneous lesions and visceral leishmaniasis the latter being fatal if not treated . L . braziliensis parasites can be transmitted by Lu . intermedia sand flies in Central and South America where they are the leading cause of American cutaneous and mucocutaneous leishmaniasis . During a blood meal , the host is exposed to a variety of sand fly factors . Sand fly saliva contains many pharmacological agents aimed at obtaining the optimal amount of blood for nutrition , egg development and survival . In addition , the proteophosphoglycan gel which is synthesized by the parasites inside the fly midgut can exacerbate cutaneous leishmaniasis [1] , [2] . Individuals living in an endemic region are bitten by both uninfected and infected sand flies , and thus are repeatedly being exposed to sand fly saliva , leading progressively to the induction of an immune response to saliva . In Brazil , Lu . intermedia is the predominant sand fly species harboring L . braziliensis [3] , [4] and in Corte de Pedra , Bahia , the endemic area studied in this report , both Lu . intermedia and Lu . whitmani sand fly species exist sympatrically with fluctuations reported for these populations [5] . The role of sand fly salivary factors is also important in the establishment of infection and thus the outcome of disease . Salivary factors include mediators that circumvent the host's hemostatic responses by preventing blood clotting , vasoconstriction and platelet aggregation for optimal feeding [6] , [7] . Sand fly saliva is immunogenic and the immune response to salivary antigens modulates the microenvironment at the site of the bite with an impact on the development of disease . Co-inoculation of L . major parasites and sand fly salivary gland sonicate ( SGS ) from either Phlebotomus papatasi or Lutzomyia longipalpis leads to increased lesion sizes and parasite numbers [8] , [9] . In contrast , several studies demonstrated that pre-immunization with P . papatasi SGS , individual components of SGS , or even uninfected sand fly bites followed by infection with L . major resulted in protection characterized by decreased lesion sizes and parasite numbers compared to controls [8] , [10] , [11] . These studies suggest that the immune response associated with sand fly SGS may be detrimental to the establishment of Leishmania infection and salivary molecules may be included in the design of a vaccine against leishmaniasis . In contrast , pre-exposure to Lu . intermedia SGS surprisingly leads to enhanced disease development after infection with L . braziliensis . The exacerbated disease in these mice was associated with increased parasite burdens and low IFNγ/IL-4 ratios [12] . Pre-sensitization to Lu . intermedia SGS induced cell recruitment , an anti-SGS antibody response and a cell-mediated immune response [12] , [13] . To understand the parameters involved in the increased lesion development at the site of L . braziliensis inoculation in mice pre-exposed to Lu . intermedia SGS , we examined gene expression in the skin after repeated SGS inoculations . We wanted to understand the mechanisms by which Lu . intermedia SGS modulates the microenvironment and how it may enhance susceptibility to L . braziliensis infection . The genes that were most induced in mice were further analyzed in SGS-stimulated PBMCs of healthy individuals naturally pre-exposed to Lu . intermedia sand fly bites .
For animal studies , all animal protocols were approved by the Swiss Federal Veterinary Office and experiments were performed adhering to ethical guidelines established by this office . Recommendations in the guidelines for the care and use of laboratory animals were obtained from the Department of Security and Environment of the state of Vaud , Switzerland . The protocol has been approved by the Ethics and Veterinary Office of Regulations of the state of Vaud ( SAV ) , Switzerland under the administrative authorization number 1266-5 . For human studies , written informed consent was obtained from all enrolled subjects; all procedures were approved by the Ethical Committee of the Federal University of Bahia . Female BALB/c mice were purchased from Charles River ( Lyon , France ) , housed under pathogen-free conditions in the BIL Epalinges Center and used for experiments between 6–8 weeks old . Adult Lu . intermedia female sand flies were captured in Corte de Pedra , Bahia , Brazil . Entomological gathering was done on private land with permission from owners for the study to be conducted on their land and within their residences . Lu . intermedia sand flies were morphologically identified according to the identification key proposed by Young and Duncan [14] . Sand fly salivary glands were dissected and stored in groups of 20 pairs in 20 mL NaCl ( 150 mM ) , Hepes buffer ( 10 mM; pH 7 . 4 ) at −70°C . Immediately before use , salivary glands were disrupted by ultrasonication in 1 . 5 mL conical tubes . Tubes were centrifuged at 10 , 000×g for 2 min , and the resultant supernatant ( SGS ) was used for the studies . All SGS batches were below the limit of detection for endotoxin activity ( <0 . 01 EU/µg ) using the LAL QCL-1000 assay ( Lonza , Portsmouth , NH ) . L . braziliensis ( MHOM/BR/01/BA788 strain ) parasite which does not contain the Leishmania RNA virus [15] was used for experiments . The parasites were maintained in vivo in BALB/c mice and grown in vitro in M199 media ( GIBCO , Paisley , UK ) supplemented with 10% FCS ( PAA Laboratories , Pasching , Austria ) , 4% HEPES ( Amimed ) and 2% antibiotics ( penicillin , streptomycin , neomycin ) ( GIBCO ) . For infections , 1×106 stationary phase promastigotes with or without SGS ( equivalent of 1 pair of Lu . intermedia salivary glands ) in 10 µL PBS were injected intradermally into the ear . Mice were immunized with salivary gland sonicate supernatant ( SGS ) as previously described [13] . BALB/c mice ( at least 3–5 per group ) were immunized 3 times with SGS ( equivalent to 1 pair of Lu . intermedia salivary glands ) or PBS in 10 µL in the right ear at 2-week intervals . After 2 weeks , the opposing left ear was challenged with SGS ( equivalent to 1 pair of Lu . intermedia salivary glands ) in the presence or absence of 1×106 stationary phase L . braziliensis promastigotes . Ear lesion size was monitored weekly and measured using a caliper . To determine cellular content , ears were digested 2 weeks after challenge in the left ear using 0 . 2 mg/mL Liberase TL ( Roche , Rotkreuz , Switzerland ) for 2 h at 37°C followed by FACS analysis [16] . For cell surface molecules , mAb 24G2 was used to block FcRs and cells were stained using α-F4/80-biotin , α-Ly6C-FITC , α-Ly6G-APC/Cy7 ( clone 1A8 ) , α-MHCII-Alexa Fluor 700 from BioLegend ( San Diego , CA ) and α-CD11b-eFluor 450 , α-CD11c-PE/Cy5 , α-DEC205-APC , α-pan-NK CD49b-PE ( clone DX5 ) and streptavidin-PE/Cy7 from eBioscience ( San Diego , CA ) . All cell events were acquired on an LSRII flow cytometer ( BD Biosciences , San Jose , CA ) and analyzed using FlowJo ( Tree Star , Ashland , OR ) . Ears were harvested 2 weeks after challenge , homogenized using a tissue lyser ( Qiagen , Hilden , Germany ) and mRNA was extracted by the RNeasy Plus Mini kit ( Qiagen ) . For microarray analysis RNA was harvested from ears 2 weeks post inoculation and for each sample condition , three independent sets of 200 ng of total RNA were isolated and used as a template for probe generation using an Ambion WT expression kit ( Applied Biosystems , Foster City , CA ) and the cDNA was fragmented and labeled with WT DNA terminal labeling kit ( Affymetrix , Santa Clara , CA ) . Biotinylated sense strand fragments were hybridized to Affymetrix Mouse Gene 1 . 0 ST GeneChips using the Hybridization Control and Hybridization Wash and Stain kits at 45°C for 18 h . The stained array was scanned using an Affymetrix GeneChip Scanner 3000 7G to generate the CEL files . The chip data were imported with Partek Genomics Suite 6 . 5 ( Partek , Inc . , St . Louis , MO ) , normalized and summarized using the RMA ( Robust Multiarray Average ) algorithm . The relative log expression was examined to ensure that the data were properly corrected by normalization and that there were no outliers . Scatter plots were generated using Matlab 2012a ( MathWorks , Natick , MA ) and DataGraph 3 . 0 ( Visual Data Tools Inc . , Chapel Hill , NC ) . To identify expression changes between genotypes , a one-way ANOVA with contrast was performed by using the methods-of-moments . Quantitative real-time PCR was carried out using random 9-mers , M-MLV reverse transcriptase RNase H- ( Promega , Madison , WI ) and SYBR green on a LightCycler 480 system ( Roche ) . The primer sequences are listed in Table S1 . Thermal cycle conditions started with a 5 min denaturation at 95°C and 45 cycles at 95°C for 10 sec , 60°C for 10 sec and 72°C for 10 sec . The results were normalized to the housekeeping gene hypoxanthine phosphoribosyl transferase ( HPRT ) using the comparative threshold cycle method ( 2ΔΔCT ) for relative quantification [16] . Samples used in the present study were obtained from individuals enrolled in an epidemiological survey conducted in Corte de Pedra , Brazil , an endemic region for American cutaneous leishmaniasis , where Lu . intermedia sand flies transmit L . braziliensis [5] . Details of the area and patients are described elsewhere [17] . For the present study , individuals ( n = 7 ) were selected based on a positive ELISA for anti-Lu . intermedia salivary molecules; the cutoff OD values for a positive anti-Lu . intermedia SGS response were established using control individuals [9] . None of control individuals had history of Leishmania infection and all had a negative Leishmania skin test . For the control group , four individuals living in a non-endemic area of Salvador , Bahia were selected based on their lack of exposure to Lu . intermedia SGS as determined by serology using SGS-specific ELISAs . Following Ficoll-Hypaque gradient centrifugation , peripheral blood mononuclear cells ( PBMCs ) were resuspended in RPMI-1640 supplemented with 2 mM L-glutamine , penicillin ( 100 U/mL ) , streptomycin ( 100 µg/mL ) ( all from Invitrogen ) , and heat inactivated human serum AB Rh+ ( Sigma Chemical Co . , MO ) . PBMCs ( 3×106/mL ) were washed two times and resuspended in complete RPMI . Cells were plated in 24-well plates ( Corning Incorporated Life Sciences , Lowell , MA ) and incubated at 37°C , 5% CO2 in the presence or not of SGS ( equivalent to 1 . 5 pairs of salivary glands ) for 72 h . Following stimulation , cells were harvested and total RNA was extracted using Trizol ( Life Technologies , Rockville , MD ) , according to manufacturer's instructions . RNA was eluted in water and used for cDNA synthesis ( ImProm-II reverse transcription system-Promega ) . Real-time PCR was performed on the ABI Prism 7500 ( Applied Biosystems ) . The primer sequences are found in Table S1 . Thermal cycle conditions consisted of a two-min initial incubation at 50°C followed by a 10 min denaturation at 95°C and 50 cycles at 95°C for 15 sec and 60°C for one min each . Samples were analyzed in triplicate and the comparative method was used where gene expression cycle threshold ( Ct ) values were normalized to HPRT expression as determined by the equation ΔCt = Ct ( target gene ) −Ct ( hprt ) . Fold change was determined by 2−ΔΔCt , where ΔΔCt = ΔCt ( SGS ) −ΔCt ( medium ) [18] . Statistical analysis was performed using GraphPad Prism 5 software ( San Diego , CA ) . For murine experiments , a two-tailed Student's unpaired t-test was carried out . For human experiments , a nonparametric Mann-Whitney test was applied .
Repeated pre-exposure of BALB/c mice to Lu . intermedia SGS enhances susceptibility to L . braziliensis infection with a lesion beginning at 3 weeks post-infection ( Fig . 1A and 1B ) in line with previously published results [12] . Thus , we wanted to determine if differences in cellular recruitment due to pre-immunization with Lu . intermedia SGS prior to infection could explain the differences in disease status . Therefore , mice were repeatedly pre-immunized with SGS or inoculated with PBS and both groups were challenged with SGS in the contralateral ear . We examined the cellular infiltrate of the ear two weeks after SGS challenge , when the adaptive immune response is ongoing and the parasite has typically already established infection , despite a lack of detectable differences in lesion size . At this point , no significant differences were observed in the total number of cells , or the numbers of neutrophils , macrophages or DCs in the ears of mice pre-immunized with SGS compared to those inoculated with PBS ( Fig . S1 ) . As a result , we hypothesized that alterations in gene expression in response to repeated exposures to Lu . intermedia SGS may be modulating the local skin microenvironment , impacting the innate and adaptive immune responses and thus the outcome of disease . To examine the effect of SGS pre-immunization at the inoculation site , we carried out a microarray analysis in mice that were pre-immunized with SGS or inoculated with PBS and later challenged with SGS in the opposing ear dermis . The ear pinna was processed and analyzed two weeks after the last SGS challenge ( Fig . 1C ) . Overall , there were few differences in the global gene expression patterns between mice that were repeatedly pre-exposed to SGS and challenged with SGS compared to those inoculated with PBS and challenged with SGS . However , hierarchical clustering analysis revealed that 95 genes were increased and 60 genes were decreased in response to SGS pre-immunization compared to control mice ( Fig . 1D and 1E ) . Of the 155 transcripts modulated by SGS pre-immunization , only 49 transcripts have been annotated , or ascribed to a specific gene , and the rest are classified as hypothetical or unknown . Despite the majority of these genes being classified as hypothetical or unknown , many of the transcripts that were differentially expressed in response to SGS pre-sensitization are known to play a role in immune processes like antigen presentation and signaling as well as transcripts encoding for cytokines , chemokines and their receptors ( Fig . 1F ) . Of the 49 annotated genes differentially regulated with SGS pre-immunization , the microarray analysis revealed all but one of these annotated genes was increased upon SGS challenge in mice pre-immunized with SGS compared to those inoculated with PBS ( Table 1 ) . Of the 49 annotated genes , 4 transcripts had greater than 2-fold expression in SGS pre-exposed mice compared to controls; the gene most highly expressed in SGS pre-immunized mice compared to PBS-inoculated animals was CXCL9 . Mpeg1 , a transcript indicative of macrophage presence , as well as IL-1rl1 and TLR13 which are members of the toll-like superfamily of receptors , were also significantly elevated in response to SGS challenge in SGS pre-exposed mice compared to controls . The microarray results revealed an especially high frequency ( 14 . 3% of the annotated genes ) of genes induced in response to SGS pre-immunization to be IFN-inducible genes including immunity-related GTPases ( IRGs ) and guanylate-binding proteins ( GBPs ) [19]–[25] . Given the surprisingly large proportion of the modulation of IFN-inducible genes in mice pre-immunized with SGS compared to control mice , we carried out real-time qPCR for IFN-inducible genes as well as genes associated with IFN-induced responses on a biological replicate experiment to confirm the findings of the microarray analysis . Cells from mice pre-sensitized with SGS and challenged with SGS had a higher expression of Ifit1 , Irgm1 and Irgm2 compared to mice inoculated with PBS and challenged with SGS . Of note , despite these differences , challenge with SGS in mice pre-immunized or not with SGS had a higher expression of these genes compared to naïve mice , suggesting SGS inoculation alone can already induce this gene family ( Fig . 2A ) . In addition , pre-immunization with SGS also led to an increased expression of Stat1 , a signaling molecule responsible for the subsequent expression of IFN-inducible genes . CXCL9 , a chemokine involved in T cell migration induced by IFNγ , was also expressed at higher levels in SGS pre-exposed mice compared to control mice ( Fig . 2B ) . Despite a reduced number of IRG homologues in humans , some of the IFN-inducible genes modulated in the mouse upon SGS pre-immunization have homologues in humans [26]–[30] . In order to determine if the same genes were upregulated in humans naturally exposed to Lu . intermedia saliva , we isolated PBMCs from individuals living in Corte de Pedra , Brazil , an area endemic for L . braziliensis with active Lu . intermedia sand fly transmission [31] . Exposure to Lu . intermedia bites was determined based on a positive serology result for anti-Lu . intermedia SGS antibodies using a mean OD cutoff of 0 . 2711 ( +/− 0 . 1006 SD ) ( Carvalho et al . , unpublished data ) . Following PBMC isolation , cells were cultured in the presence or absence of Lu . intermedia SGS followed by mRNA isolation . PBMCs from these exposed individuals exhibited higher levels of Ifit1 , Irgm , Stat1 and CXCL9 mRNA in response to SGS compared to PBMCs isolated from people living in a non-endemic area ( Fig . 3 ) . Similarly , supernatants from PBMCs of individuals living in an endemic area stimulated with SGS also produced significantly more CXCL9 protein as measured by ELISA compared to controls ( data not shown ) . These data demonstrate that IFN-inducible genes are induced in response to SGS pre-sensitization in both the experimental model and in cells from human individuals pre-exposed to Lu . intermedia bites . Here , we show that SGS pre-immunization induces the expression of IFN-inducible genes , which are typically associated with a protective response as Irgm1−/− animals are highly susceptible to Leishmania infection ( mentioned as data not shown in [21] ) . However , pre-immunization with Lu . intermedia SGS has been reported to enhance L . braziliensis infection [12] . To evaluate the effect of the parasite on the local immune response induced by pre-immunization with SGS , mice were repeatedly pre-exposed to SGS or PBS and challenged with SGS in the presence or absence of L . braziliensis parasites . Gene expression profiling studies were carried out 2 weeks later . Mice that were pre-sensitized with SGS and challenged with SGS alone significantly upregulated the expression of Ifit1 , Irgm1 , Irgm2 , Stat1 and CXCL9 compared to PBS pre-inoculated mice in line with our microarray data ( Fig . 4 ) . However , the mice that were pre-exposed to SGS and challenged with SGS and L . braziliensis did not significantly upregulate the expression of these genes compared to controls ( Fig . 4 and Table S2 ) . Taken together , these data suggest that L . braziliensis parasites modulate host gene expression at the site of infection creating a more hospitable environment for parasite establishment which is associated with increased lesion development .
Many studies have suggested the anti-saliva response against sand fly species such as P . papatasi or L . longipalpis is detrimental for the establishment of Leishmania infection . In contrast , the Lu . intermedia anti-saliva response does not prevent the development of disease , but rather may modulate the outcome of infection . Studies in a mouse experimental model have demonstrated that L . braziliensis infection alone induces a strong Th1 cell immune response with high levels of IFNγ and elevated numbers of IFNγ-producing CD4+ and CD8+ T cells in the dLN [32]–[34] . The strong protective immune response characterized by the presence of IFNγ was thought to correlate with the strong resistance to L . braziliensis infection [34]–[39] . Here , we show that repeated pre-immunizations with Lu . intermedia SGS alters the skin microenvironment and induces the expression of a variety of genes involved in the immune response , especially from the family of IFN-inducible genes . Genomic analysis of the skin of mice pre-immunized with SGS reveals an inflammatory setting with an increase in genes involved in immune responses including antigen presentation and cell signaling . Genes associated with Th1 cell immune responses such a CXCL9 , a chemokine typically linked with the recruitment of Th1 cells , exhibited the greatest fold induction at >5 times over control mice . The IL-7R , also known to influence the Th1 cell immune response , was also elevated following SGS pre-exposure ( this study and [40]–[42] ) . Of note , cytokines typically associated with a Th2 cell immune response such as Chi3l1 ( Ym1 ) , or regulatory cytokines such as IL-10Rα were also detected at higher levels in SGS pre-immunized mice . In the periphery IFNγ binds to its receptor and initiates the JAK/STAT signaling pathway leading to the phosphorylation and translocation of STAT1 to the nucleus which induces the transcription of more than 2000 genes including effector molecules that suppress the growth and survival of intracellular pathogens ( Phox , iNOS , IDO , NRAMP1 , GTPases , Ifits and chemokines ) [21] . Remarkably , several IFN-inducible genes as well as the IFN signaling molecule , STAT1 , were upregulated at the site Lu . intermedia challenge in mice that were pre-exposed through immunization . For example , p47 GTPases such as Irgm1 ( formerly Lrg47 ) and Irgm2 ( Gtpi ) and p65 GTPases such as GBP6 and GBP8 were expressed at high levels in pre-immunized mice . Interestingly , IFN-inducible genes were similarly induced in SGS-stimulated PBMCs isolated from humans living in an area endemic for L . braziliensis with active Lu . intermedia sand fly transmission . It was not possible to perform skin biopsy in the human population studied due to ethical considerations; however , the expression of IFN-inducible genes in SGS-stimulated blood cells of individuals naturally exposed to sand fly bites was similar to that observed at the site of SGS challenge in mice . Collectively , these data demonstrate that the induction of IFN-inducible genes by SGS is also occurring in humans . To our knowledge this is the first report demonstrating an induction in the expression of IFN-inducible GTPases in response to vector saliva . These products have been well characterized for their role in host defense against viruses but they also contribute to resistance against protozoans . Mice deficient for either Irgm1 or many of the other GTPases are highly susceptible to infection with Toxoplasma gondii , Trypanosoma cruzi and Leishmania major , and many mimic the dramatic susceptibility phenotypes seen in IFNγR-deficient mice [21] . It should be noted that IFN-inducible genes are turned on in response to IFNγ but type I IFNs may also contribute , although to a lesser degree [21] . In our analysis neither IFNγ or type I IFNs were elevated in cells from mice pre-sensitized with SGS but this may be a reflection of the time point analyzed ( 14 days post inoculation ) . In this study mice were immunized with SGS to mimic one of the features of natural transmission of Leishmania where individuals are pre-exposed to several sand fly bites prior to deposition of parasites by the sand fly . A high dose of L . braziliensis promastigotes was co-inoculated in mice with SGS in an attempt to reproduce the cell recruitment rapidly observed at the site of infection after a sand fly bite . However , it is important to note that upon a blood meal , the sand fly is inoculating fewer parasites and also regurgitating many other factors including metacyclic promastigotes embedded in a proteophosphoglycan gel in a blood pool [2] . These factors are not all present during needle inoculation of the parasites and SGS . It is clear that further studies using natural sand fly infection will be required for a better understanding of the transmission dynamics during Leishmania infection . Repeated exposures to Lu . intermedia SGS followed by challenge with L . braziliensis parasites in the presence of SGS leads to an enhanced disease compared to control mice ( Fig . 1A and [12] ) . Interestingly , in this prior study the SGS pre-immunized mice challenged with L . braziliensis plus SGS , and analyzed two weeks later had a lower parasite load compared to mice not immunized with SGS , suggesting a transient protection conferred by SGS pre-immunization [12] . However , the trend was inversed from 3 weeks on and the SGS-immunized group showed increased lesion size and parasite load [12] . Additionally , the mice pre-immunized with SGS and subsequently infected with L . braziliensis plus SGS had lower levels of IFNγ to IL-4 ratios compared to mice inoculated with PBS and infected with L . braziliensis plus SGS at 2 weeks post infection . Thus , in that study , the highest levels of IFNγ were not associated with the decreased parasite numbers in vivo suggesting IFNγ is not the major factor contributing to parasite killing by macrophages at this time point . In our study , increased expression of IFN-inducible genes and of IFNγ was also not detected in the microarray performed in SGS pre-immunized mice 2 weeks post co-inoculation of L . braziliensis and SGS compared to PBS controls challenged with L . braziliensis plus SGS . In this and the previous study , the levels of IFNγ upon challenge with L . braziliensis and SGS were not elevated following SGS pre-immunization . Thus other factors may be involved in the transient control of parasite load observed by de Moura and colleagues [12] . Furthermore , higher concentrations of IFNγ are required for optimal parasite killing of L . braziliensis compared to L . major suggesting differences in the susceptibilities to IFNγ-mediated killing between different parasite strains [15] . Nevertheless , following L . braziliensis and SGS co-inoculation , both studies showed increased inflammatory lesions in the group pre-immunized with SGS compared to that injected with PBS . This increase in disease severity to L . braziliensis infection in SGS pre-immunized mice , corresponds to a silencing of many of the genes turned on by SGS pre-sensitization , including IFN-inducible genes . This suggests that the parasite is actively modulating the host's immune response to the SGS . Interestingly , this observation is consistent with previous findings showing a decreased ratio of IFNγ/IL-4 production in the dLNs of mice pre-exposed to SGS and challenged with parasites [12] . In the same line , the same group further reported that challenge with L . braziliensis plus SGS after SGS pre-immunization also silenced CXCL10 , another IFN-inducible gene [13] . Despite differences in the methodology used between these studies ( air pouch model in the former studies and needle inoculation in the ear pinna in the current study ) , the outcomes are going in the same direction . In addition , the impact of SGS on the skin microbiome which was shown to influence skin immunity may also contribute to the phenotype observed [43] . We show here that there is an obvious benefit for the parasite to down-modulate the IRG system expressed in response to SGS pre-exposure to allow for parasite establishment . However , modulation of IFN-inducible genes is most likely not the only mechanism for enhancing disease . It is unclear how the parasite is altering the host's response to the SGS in the present study and this will require further investigation . In conclusion , we have shown that in both humans and mice , an array of IFN-inducible genes were up-regulated in response to Lu . intermedia SGS pre-exposure . Interestingly , these genes were silenced when the parasite was present during the challenge . Given the marked changes in the skin microenvironment resulting from repeated exposures to Lu . intermedia SGS , and the different outcomes to Leishmania infection , understanding the relationship between pathogens and their homologous vectors is essential . Since SGS proteins from different sand fly species can either exacerbate or protect from disease , subsequent studies will aim to understand how the parasite is modulating SGS impact on the microenvironment [44] . This will help determine risk factors for disease development , markers of exposure and defining potential vaccine candidates . | Leishmaniasis is a vector-borne parasitic disease of serious public health importance . No efficient vaccine is currently available . Parasites are transmitted to mammalian hosts during sand fly bites . During this process , both parasites and sand fly salivary products are delivered into the skin . Immunization with salivary proteins from most sand fly species can protect mice against cutaneous leishmaniasis; however , immunization with sand fly saliva of Lutzomyia intermedia leads to aggravation of leishmaniasis . We investigated the impact of Lutzomyia intermedia saliva exposure on the development of immune response to Leishmania braziliensis , the major causative agent of tegumentary leishmaniasis in Brazil . To this end , we analyzed in mice the gene expression pattern induced by immunization with salivary gland extracts . Among the genes highly induced were the interferon-inducible genes known to contribute to resistance against parasite infections . These genes were also induced in blood cells of human individuals that were naturally pre-exposed to bites of Lutzomyia intermedia sand flies . Interestingly , subsequent infection with Leishmania braziliensis blocked the induction of these genes in mice . These data show that the induction of potentially protective genes by insect saliva can be altered by the infecting parasite . This should be considered when including salivary components in a vaccine . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"immunizations",
"immunity",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"parasitology"
]
| 2014 | Repeated Exposure to Lutzomyia intermedia Sand Fly Saliva Induces Local Expression of Interferon-Inducible Genes Both at the Site of Injection in Mice and in Human Blood |
During embryonic development , pattern formation must be tightly synchronized with tissue morphogenesis to coordinate the establishment of the spatial identities of cells with their movements . In the vertebrate retina , patterning along the dorsal-ventral and nasal-temporal ( anterior-posterior ) axes is required for correct spatial representation in the retinotectal map . However , it is unknown how specification of axial cell positions in the retina occurs during the complex process of early eye morphogenesis . Studying zebrafish embryos , we show that morphogenetic tissue rearrangements during eye evagination result in progenitor cells in the nasal half of the retina primordium being brought into proximity to the sources of three fibroblast growth factors , Fgf8/3/24 , outside the eye . Triple-mutant analysis shows that this combined Fgf signal fully controls nasal retina identity by regulating the nasal transcription factor Foxg1 . Surprisingly , nasal-temporal axis specification occurs very early along the dorsal-ventral axis of the evaginating eye . By in vivo imaging GFP-tagged retinal progenitor cells , we find that subsequent eye morphogenesis requires gradual tissue compaction in the nasal half and directed cell movements into the temporal half of the retina . Balancing these processes drives the progressive alignment of the nasal-temporal retina axis with the anterior-posterior body axis and is controlled by a feed-forward effect of Fgf signaling on Foxg1-mediated cell cohesion . Thus , the mechanistic coupling and dynamic synchronization of tissue patterning with morphogenetic cell behavior through Fgf signaling leads to the graded allocation of cell positional identity in the eye , underlying retinotectal map formation .
Map-like representation of sensory information is an evolutionary conserved principle of brain organization and function [1] . The point-to-point mapping of retinal ganglion cell ( RGC ) axons onto the midbrain tectum/superior colliculus of the vertebrate , is a hallmark example for the requirement of precise pattern formation during embryonic development , since mapping occurs according to the position of RGCs along the nasal-temporal ( anterior-posterior ) and dorsal-ventral axes of the retina . The topographic projections of RGC axons accurately preserve information on cell positions and neighborhood relationships in the retina as a continuous map of terminals in the tectum [2] . Cell-surface axon guidance molecules expressed in gradients across the retina and tectum control the formation of retinotopic connections [3]–[8] . Guidance molecule expression along the nasal-temporal retina axis is regulated by the nasal- and temporal-specific transcription factors Foxg1 , Foxd1 , SOHo , and GH6 [9]–[12] . However , expression of these factors in the retina is asymmetrical from the onset , indicating that they act downstream of nasal-temporal axis specification . Retinotopic mapping consequently occurs as a function of RGC position along molecular gradients within a coordinate system set by the major retinal axes . This suggests that axis formation and mapping are intimately connected developmental processes , but the nature and timing of the signals that establish cell positional identities in this coordinate system are largely unknown . Resolving the mechanisms underlying the allocation of positional identity to retinal cells is confounded by the complex morphogenetic rearrangements of forebrain tissues that occur during eye formation [13]–[16] . Morphogenesis of the retina begins with the lateral displacement of cells in the eye field to the site of future optic vesicle evagination [17] , [18] . Subsequently , cells continuously evaginate from the forebrain , steadily increasing the size of the optic vesicle . Next , the optic vesicle invaginates to form the two-layered optic cup , with the outer layer , which faces the surface ectoderm and lens , fated to become neural retina and the inner layer the retinal pigment epithelium . Cell movements from the presumptive pigmented epithelium into the neural retina may occur during this phase [19] . Later optic cup development depends on integrin-mediated focal adhesion at the basal side of the retinal epithelium [20] . However , although the anatomy of the optic cup is well described , dynamic in vivo analysis of its formation is lacking , and axial patterning of the prospective retina has not been studied prior to the completion of optic cup formation [21] , [22] . It thus remains to be determined how and when axial cell positions in the retina are specified and how the orientation of the retinal axes is related to the axes of the neural tube during eye morphogenesis . Fgfs have conserved functions in axial patterning of the neural tube . Originating from local organizers , they can control cell positional identities in adjacent regions , with several Fgfs often acting in a combinatorial manner [23] , [24] . We have previously shown that Fgf8 signaling contributes to nasal-temporal patterning of the retina , and proposed a combinatorial Fgf signal to exert full control over the specification of this retinal axis [25] , but the nature of the Fgfs involved , their dynamic requirement , and sites of action during eye morphogenesis are not resolved . We show here that a combined Fgf8/3/24 signal specifies the positional identities of cells along the nasal-temporal retinal axis in zebrafish embryos . Absence of all three factors leads to completely temporalized , and ectopic activation of Fgf signaling to completely nasalized retinae . Axis specification occurs very early , at the onset of optic vesicle evagination , when fgf8/3 are expressed in the dorsal forebrain and fgf24 in the dorsally located olfactory placode . During formation of the optic vesicle , Fgf signaling from these dorsal sources is required to confine expression of foxg1 ( and other future nasal genes ) to the dorsal half , and foxd1 ( and other temporal genes ) to the ventral half of the evaginating optic vesicle . Thus , at the moment of specification , nasal-temporal cell positional identities align parallel to the dorsal-ventral axis of the neural tube and asymmetrically relative to the dorsal sources of Fgf . By in vivo tracking of GFP-labeled nasal and temporal retina progenitor cells in transgenic lines , we further show how tightly synchronized morphogenetic cell movements and cell shape changes during optic cup formation lead to axis reorientation and the final nasal-temporal subdivision of the neural retina . This occurs as the result of two temporally concordant morphogenetic processes: ( 1 ) compaction of nasal retina progenitors already residing in the future neural retina domain—possibly by shortening along the lateral and elongation along the apical-basal cell axis—and ( 2 ) directed movement of temporal retina progenitors into that domain . In this process , Foxg1 promotes cohesion of nasal progenitor cells in an Fgf-dependent manner , thereby probably allowing the gradual addition of temporal progenitors to the growing neural retina . Thus , the dynamic coordination of pattern formation with neuroepithelial morphogenesis through Fgf8/3/24 signaling controls the final arrangement of axial cell positions in the retina .
Fgf8 is involved in nasal-temporal patterning of the retina , but loss of fgf8 results only in subtle patterning defects , demonstrating the presence of other unknown factors controlling this process [25] . To test whether Fgf8 acts in combination with other Fgfs , we studied the expression of nasal and temporal marker genes in embryos that lack two or more Fgfs . We find that fgf3 and fgf24 strongly interact with fgf8 in nasal-temporal patterning ( Figure 1A and 1B ) . In wild-type ( wt ) control embryos at 28 h , the Eph receptor epha4b is expressed in a temporal-to-nasal decreasing gradient , and the ephrin ligand efna5a in a complementary nasal-to-temporal decreasing gradient . In fgf8−/− mutants , epha4b expression expands into the dorsonasal retina , and efna5a expression is reduced in that region . This phenotype is enhanced in fgf8/3−/− double mutants; and in fgf8/24−/− double mutants , the changes in epha4b and efna5a expression are even stronger , now also affecting the ventronasal retina . Upon inactivation of all three fgfs , by fgf24-morpholino injection in fgf8/3−/− double mutants ( referred to as fgf8/3/24−/− ) , all retinal cells express epha4b and none efna5a . The same result is obtained upon blocking of all Fgf receptor signaling with a pharmacological inhibitor ( FgfR-inh . ) between the 1- and 5-somite stages ( ss ) . Fgf3−/− and fgf24−/− single-mutants have no detectable retinal patterning or eye defects ( unpublished data ) . Analysis of efna5a and epha4b expression levels in defined axial regions of the retina shows that stepwise elimination of fgf8/3 and -24 results in a graded phenotypic series: wt control ( normal nasal-temporal pattern ) <fgf8−/−<fgf8/3−/−<fgf8/24−/−<fgf8/3/24−/− or FgfR-inh . –treated embryo ( all temporal pattern ) ( Figure 1C ) . The requirement for fgf8/3/24 is more pronounced in the dorsonasal than in the ventronasal retina , as illustrated by the stepwise temporal-to-nasal expansion of epha4b ( Figure 1D ) . Thus , Fgf8/3/24 constitute a combined Fgf signal that fully controls nasal-temporal patterning of the neural retina . The eye undergoes highly complex morphogenetic movements during its evagination , and the locations of cells contributing to the future nasal and temporal retina have not been followed during this process . To investigate the initial orientation of the retinal axes , we studied the early expression of genes that are later restricted along the nasal-temporal retina axis . In wt 10ss embryos , the future nasal marker , foxg1 , is expressed in the dorsal leaflet of the evaginating optic vesicle ( Figure 1E , left ) , whereas the future temporal marker , foxd1 , ( see Materials and Methods and Figure S11 and S12 for foxd1 gene nomenclature ) is expressed in the ventral optic vesicle leaflet ( Figure 1F , left ) . Surprisingly , nasal-temporal markers , therefore , initially align with the dorsal-ventral axis of the neural tube . By 20ss , the future nasal markers , foxg1 and efna5a , are restricted to the anterior-medial optic cup ( Figure 1G and 1I , left ) , whereas the future temporal markers foxd1 and epha4b are expressed in a complementary posterior-lateral domain ( Figure 1H and 1J , left ) . From 24 h , these markers are expressed in the anterior nasal half and posterior temporal half of the retina ( Figure 1A and 1B and unpublished data ) . Time-lapse imaging of transgene-labeled retinal cells during optic cup morphogenesis supports the conclusion from these gene expression analyses that the dorsal-ventral axis of the optic vesicle corresponds to the later nasal-temporal axis of the retina ( see below ) . To determine when Fgfs impose nasal-temporal identity , we analyzed early nasal-temporal markers in embryos lacking Fgf8/3/24 . In 10ss fgf8/3/24−/− embryos , foxg1 expression is absent from the dorsal optic vesicle leaflet ( Figure 1E , right ) and foxd1 is expanded throughout the vesicle ( Figure 1F , right ) , indicating a complete nasal-to-temporal patterning shift . Similarly , in 20ss fgf8/3/24−/− embryos , no foxg1/efna5a expression is detectable in the optic cup . ( Figure 1G and 1I , right ) and foxd1/epha4b expression is expanded throughout the optic cup ( Figure 1H and 1J , right ) . This shows that Fgf8/3/24 control nasal-temporal patterning of the retina at the onset of optic vesicle evagination . We next addressed the source of the Fgfs that pattern the optic vesicle along its dorsal-ventral axis , the future nasal-temporal axis of the retina . Early , spatially restricted fgf8/3/24 expression relative to the domains of nasally/temporally expressed genes explains the patterning activity of Fgfs along the dorsal-ventral axis of the optic vesicle ( Figure 1K ) . At 5ss , towards the end of the requirement phase for Fgf signaling in nasal-temporal patterning , fgf8 and -3 are expressed in the dorsal forebrain , which is contiguous with , and close to , the foxg1-expressing , dorsal optic vesicle leaflet and distant to the foxd1-expressing , ventral leaflet . fgf24 is expressed in cells of the nascent olfactory placode [26] , at the hinge between the dorsal forebrain and the dorsal optic vesicle leaflet . Fgf24-expressing cells remain in close contact with the developing optic vesicle and cup during later morphogenesis ( Figure S1 ) . Thus , expression of future nasal markers occurs close to the dorsal source of Fgfs and expression of future temporal markers distant to it ( Figure 1L ) . Local and graded Fgf signaling in the dorsal optic vesicle leaflet is further supported by nested , Fgf-dependent expression of the Fgf pathway target genes erm , pea3 , spry2 , and spry4 ( Figure S2 ) . Since Fgfs are known mitogens and Foxg1 promotes neural progenitor proliferation [27]–[30] , we assessed whether the nasal-temporal asymmetry in Fgf signaling affects cell proliferation in the optic vesicle . Control embryos , stained for the mitosis marker Phospho-Histone H3 ( PH3 ) , show slightly more PH3-positive cells in the dorsal than the ventral optic vesicle leaflet at 5ss ( Figure 2A , top ) . This asymmetry becomes clearer at 10ss ( Figure 2B , top ) , when the apical side of the dorsal leaflet is often densely populated by PH3-positive cells , a pattern never observed in the ventral leaflet . In FgfR-inh . –treated embryos , this asymmetric proliferation pattern is lost ( Figure 2A and 2B , bottom ) . Counting and plotting of the mean ratios of dorsal/ventral leaflet PH3-positive cells shows the increasing asymmetry in proliferation in control embryos and its loss upon FgfR-inhibition: at 10ss , 2-fold more dividing cells are found in the dorsal than in the ventral leaflet in control embryos , whereas the ratio is near 1∶1 after FgfR inhibition ( Figure 2C ) . Analysis of the mean PH3-positive cell number per optic vesicle leaflet , shows that progenitor proliferation is selectively affected in the nasal retina primordium/dorsal optic vesicle leaflet after FgfR inhibition , whereas proliferation of temporal retina progenitors in the ventral optic vesicle leaflet is unchanged ( Figure 2D ) . Similarly , BrdU incorporation at 10ss is severely reduced in nasal progenitors of the dorsal optic vesicle leaflet upon FgfR inhibition . Treatment has no obvious effect on temporal progenitors in the ventral leaflet ( Figure 2E ) . Thus , Fgf signaling is selectively required for enhanced proliferation of nasal retinal progenitors during optic vesicle evagination , and this requirement coincides with the Fgf-dependent regulation of foxg1 in the dorsal optic vesicle leaflet . The unexpected initial alignment of future nasal-temporal markers along the dorsal-ventral axis raised the question how the nasal-temporal axis reaches its final anterior-posterior orientation . We thus analyzed the dynamic development of the nasal-temporal axis by in vivo imaging of eye formation in transgenic Tg ( -8 . 0cldnb:lynGFP ) zf106 embryos [31] , which we find express GFP in the nasal retina throughout development ( Figure 3A and S3 , see Materials and Methods ) . From 10- to 15ss , cldnb:GFP is expressed throughout the dorsal optic vesicle leaflet , but between 18- and 25ss , cldnb:GFP expression becomes progressively restricted to the dorsal half of the outer layer of the optic cup . The portion of the optic vesicle that contacts the lens ectoderm will form the neural retina , and henceforth , we use the term outer layer to describe this part of the forming optic cup . At 28 h , after completion of optic cup morphogenesis , cldnb:GFP is restricted to the nasal half of the neural retina . To investigate the mechanism that gradually restricts cldnb:GFP expression to the dorsal half of the outer layer of the optic cup , we performed time-lapse imaging between 18- and 24ss ( Figure 3B and S4 ) . cldnb:GFP expression initially reaches the distal limit of the optic cup , which we term the ridge , but within about 2 . 5 h , its distal limit is approximately nine cell diameters from the ridge . Time-lapse movies show pronounced outward cell movement in the inner layer of the optic cup towards the distal ridge ( Video S1 ) . This suggests a gradual displacement of cldnb:GFP-positive , nasal retina progenitor cells in the outer optic cup layer—the future neural retina domain—by a late movement of cldnb:GFP-negative cells , presumably from the inner layer , around the optic cup ridge . To determine whether outer layer cells are indeed displaced in this way , we followed the movement of DsRed2-expressing outer layer cell clones in transgenic membrane-GFP Tg ( Bactin:HRAS-EGFP ) vu119 embryos [32] . A representative cell-tracking experiment shows how a cell in the ridge region ( blue ) of the outer optic cup ( distance to ridge at 0 min: one cell diameter ) is gradually displaced dorsally and proximally ( distance to ridge at 1 h 25 min: six cell diameters ) , and as this happens , it elongates along the apical-basal axis ( apical-basal axis at 0 min: 22 µm , at 1 h 25 min: 41 µm ) ( Figure 3C and Video S2 ) . Notably , this lateral displacement within the optic epithelium occurs with the same kinetics as the displacement of the distal limit of cldnb:GFP expression ( compare Figure 3B to 3C ) . A representative cell positioned further dorsally ( white ) is barely displaced laterally and elongates only slightly ( apical-basal axis at 0 min: 34 µm , at 1 h 25 min: 45 µm ) ( Figure 3C ) . At 36 h , the clone in Figure 3C , which initially covered the complete extent of the outer optic cup layer ( see insets in Figure 3C ) , is restricted to the nasal hemiretina ( Figure S5; n = 5/5 analyzed outer layer clones ) confirming that outer layer cells are all initially destined for nasal retina . This suggested that cell movements from the inner optic cup layer around the distal ridge region gradually add nonnasal retina progenitors to the outer layer of the optic cup . This addition of cells occurs coincident with the elongation of the apical-basal axis of nasal progenitors already residing in the outer optic cup layer , suggesting a gradual compaction of the future neural retina epithelium . The gradual encroachment of GFP expression into the outer layer in the HGn42A enhancer trap line [33] is complementary to the restriction in cldnb:GFP expression ( Figure 3D ) . The insertion in HGn42A maps to a site 52-kbp downstream of the foxd1 locus , and GFP expression in this line recapitulates endogenous foxd1 expression in the prospective temporal retina ( Figure S6 ) . These results support the conclusion that HGn42A:GFP-positive , prospective temporal , retinal cells move around the distal optic cup ridge and displace the cldnb:GFP-positive nasal progenitors to the dorsal-proximal optic cup . Thus , the nasal-temporal axis of the retina is established by Fgf-dependent patterning of the optic vesicle along the dorsal-ventral axis of the neural tube . Only later , during optic cup formation , do temporal retina progenitors start to move into the definitive neural retina domain , while nasal retina progenitors already residing there regress and compact . Concomitant with anterior eye rotation , this leads to the final alignment of the nasal-temporal retina axis with the anterior-posterior body axis . Fgf8/3/24−/− embryos form smaller , but otherwise morphologically normal , retinae ( Figure S7 ) despite the complete loss of nasal-temporal polarity and reduced proliferation , suggesting global eye morphogenesis is not compromised in the absence of nasal-temporal patterning . However , we do find that Fgf signaling specifically and regionally affects epithelial cell morphology and behavior at the onset of optic cup formation . Increasingly reduced levels of cldnb:GFP expression in the dorsal optic vesicle leaflet after FgfR-inh . –treatment indicates that reporter expression depends on Fgf signaling ( Figure 4A and 4B; n = 7/7 ) . At 15ss , cells in the dorsal leaflet appear disorganized , whereas cells in the lower leaflet appear normal ( Figure 4B ) , the tight apical membrane apposition of the optic vesicle leaflets is lost , and the ventricle contains delaminated , weakly cldnb:GFP-positive ( unpublished data ) cells ( Figure 4C and 4D , n = 5/5 ) . The delaminated cells eventually undergo apoptosis , but viability of the disorganized cells in the optic vesicle neuroepithelium is not compromised ( Figure 4E ) . To test whether an early defect in apical-basal cell polarity causes later delamination , we studied the expression of the two apical markers aPKC and ZO1 . At 10ss , prior to cell delamination , expression of both markers is normal in FgfR-inh . –treated embryos compared to controls ( Figure S8A and S8B ) . At 15ss , after the onset of delamination , expression reflects the loss of apical membrane apposition and accumulation of cells in the optic vesicle ventricle of FgfR-inh . –treated embryos , but the general apical-basal polarity of the optic vesicle epithelium is not affected ( Figure S8C and S8D and unpublished data ) . To directly assess the requirement for FgfR-signaling in nasal retina progenitors , we transplanted Tg ( hsp70l:dnfgfr1-EGFP ) pd1–positive cells [34] that express a dominant-negative , GFP-tagged version of Fgfr1 under the control of a heat-shock promoter , into wt host embryos and heat-shocked the chimeras at the onset of optic vesicle evagination . At 12ss , embryos with dnfgfr1-EGFP clones in the dorsal optic vesicle leaflet show reduced foxg1 ( n = 10/12 ) and ectopic expression of foxd1 ( n = 8/12 ) ( Figure S9 ) , reminiscent of fgf8/3/24−/− embryos ( Figure 1E and 1F , right ) . Many abnormally cuboidal dnfgfr1-EGFP–expressing cells in the dorsal optic vesicle leaflet accumulate at the apical side of the neuroepithelium and often protrude and delaminate into the ventricle at 15ss ( Figure 4F , n = 7/8 , left and middle ) , similar to the effect of the FgfR-inh . Cell and epithelial morphology in control chimeras that express eGFP under the control of the heat-shock promoter is normal ( Figure 4F , n = 9/9 , right ) . Thus , misspecification of the dorsal optic vesicle leaflet in the absence of Fgfs—now foxg1-negative—leads to a defect in neuroepithelial integrity and subsequent loss of presumptive nasal retina progenitors . Consistent with a loss rather than a temporal misspecification of nasal progenitors , there is no perdurance of nasal cldnb:GFP expression in the retinae of FgfR-inh . –treated embryos at 36 h ( Figure 4G; n = 8/8 ) . The complementary expansion of temporal HGn42A:GFP expression in retinae of FgfR-inh . –treated embryos at 36 h indicates a complete loss of nasal cell fates ( Figure 4H ) . Although the behavior of prospective nasal cells is disrupted upon abrogation of Fgf signaling , an optic cup still forms . To explore how eye morphogenesis occurs in such circumstances , we followed the movements of HGn42A:GFP-labeled temporal progenitors by in vivo imaging in FgfR-inh . –treated embryos . GFP mRNA in HGn42A embryos is ectopically found in the dorsal optic vesicle leaflet after FgfR-inh . treatment , similar to the effect on foxd1 , but GFP protein maturation appears to lag behind , therefore allowing the tracking of ventral optic leaflet cells ( Figure 4J ) . At 20ss , more HGn42A:GFP-positive temporal progenitors have moved into the future neural retina domain of the outer optic cup layer in FgfR-inh . –treated embryos compared to controls ( Figure 4I; n = 7/7 ) . This suggests that the loss of misspecified foxg1-negative , nasal progenitors in the optic cup after FgfR inhibition causes an enhanced movement of temporal progenitors from the ventral leaflet into the future neural retina domain . To complement the analysis of morphogenetic movements in eyes lacking nasal identity , we created eyes that lack temporal identity by implanting Fgf8 beads adjacent to the nascent temporal retina , below the ventral optic vesicle leaflet ( Figure 5A and 5B ) . The movement of HGn42A:GFP-positive cells from the ventral optic vesicle leaflet into the outer optic cup layer is delayed after Fgf8 bead implantation compared to the control side of the same embryo ( Figure 5C; n = 8/9 ) . However , perdurance of HGn42A:GFP expression shows that these cells eventually reach their normal axial position in the retina at 36 h ( Figure 5D; n = 12/12 ) . Thus , the morphogenetic movement of temporal retina progenitor cells from the ventral optic vesicle leaflet around the distal ridge into the prospective neural retina can occur independent of correctly restricted foxg1 and foxd1 expression . The dependence of foxg1 expression upon Fgf signaling raised the possibility that Foxg1 may be a transcriptional mediator of some or all of the effects of Fgf signaling upon presumptive nasal progenitors . To test this idea , we studied foxg1 in loss- and gain-of-function assays . Live imaging of optic vesicles in memGFP-labeled embryos , shows that abrogation of Foxg1 using a translation-blocking morpholino ( foxg1MO; Figure S10 ) results in the delamination and accumulation of cells in the ventricle of the optic vesicle at 15ss ( n = 7/8 ) , when compared to controls injected with a 5-bp mismatch control morpholino ( foxg1-5MM-MO , n = 8/8 ) ( Figure 6H , left two panels ) . This phenotype is highly similar to the delamination observed in fgf8/3/24−/− embryos and after FgfR inhibition ( Figure 6H , right two panels ) . Except for this phenotype and the previously reported changes during telencephalic development [35] , this foxg1 morpholino does not create any morphological defects ( unpublished data ) , consistent with its specificity and the restriction of foxg1 expression to the developing forebrain at this stage . Live imaging reveals that only foxg1MO-injected DsRed2 donor cells delaminate when transplanted into noninjected HRAS-GFP–expressing hosts ( n = 9/10 ) ; neither host cells nor DsRed2-positive control transplanted cells ( n = 10/10 ) , nor foxg1-5MM-MO–injected cells show this phenotype ( Figure 6A–6C ) . Next , we performed clonal overexpression by transplanting foxg1cherry-injected DsRed2 donor cells into noninjected HRAS-EGFP hosts . Compared to clones of noninjected cells , which scatter by mixing with host cells , foxg1cherry-overexpressing cells form highly coherent clusters , with hardly any host cells intermingling at 10ss ( Figure 6D and 6E; n = 15/15 ) . Together , these results suggested that Foxg1 may act downstream of Fgf signaling to promote epithelial cohesion of cells in the nascent nasal retina . Supporting this hypothesis , we found that Foxg1 could rescue the delamination of nasal retinal cells that occurs upon abrogation of Fgf signaling . In chimeras in which Fgf signaling is blocked , foxg1cherry-overexpressing cells are rescued from delamination ( n = 12/12 ) , when compared to host cells and noninjected transplanted cells ( Figure 6F and 6G; n = 9/11 ) . This strongly suggests that Foxg1 promotes cell cohesion in the dorsal optic vesicle leaflet , and that the delamination and death of misspecified nasal progenitors observed in the absence of Fgf signaling is due to the lack of Foxg1 . In chimeras carrying foxg1-overexpressing cells , we observed that the large majority of coherent clones were found in the dorsal forebrain , dorsal optic vesicle leaflet , head mesenchyme surrounding the optic vesicle , and olfactory primordium ( unpublished data ) , all sites of high Fgf pathway activity . To explore this phenomenon , we tracked the lateral spreading/clustering of foxg1-overexpressing cell clones in response to exogenous Fgf provided from a bead . For this purpose , first , 20–25 cells from memGFP-labeled donors , either overexpressing foxg1cherry or cherry ( control ) protein , were transplanted into nonlabeled hosts at sphere stage . Directly afterwards , beads , either coated with recombinant Fgf8 protein or PBS ( control ) , were implanted next to the transplanted cell clone . Individual bead-implanted chimeras were then separately analyzed by live imaging at sphere ( directly after implantation ) , bud ( 6 h after implantation ) , and 15ss ( 12 h after implantation ) ( Figure 7A ) . Directly after implantation ( sphere ) , spreading of foxg1cherry-overexpressing clones in the presence of a PBS bead ( Figure 7B , top ) , foxg1cherry-overexpressing cells in the presence of an Fgf8 bead ( Figure 7C , top ) and cherry-overexpressing cells in the presence of an Fgf8 bead ( Figure 7D , top ) is very similar , when comparing the average of the measured maximal ( dmax ) , measured minimal ( dmin ) , and calculated median ( dmed ) distance of cells to the bead surface from four representative experiments for each condition ( Figure 7E , first radial plot ) . At bud , live imaging and measuring dmax ( Figure 7E , second radial plot ) of foxg1cherry-overexpressing cells in the presence of an Fgf8 bead ( Figure 7C , middle ) show slightly decreased spreading , when compared to foxg1 overexpression with a PBS bead ( Figure 7B , middle ) or cherry overexpression with an Fgf8 bead ( Figure 7D , middle ) . At 15ss , all analyzed foxg1cherry-overexpressing clones in the presence of an Fgf8 bead formed a tightly aggregated , single cluster at the site of bead implantation ( Figure 7C , bottom ) . Foxg1cherry-overexpressing clones in the presence of a PBS bead form several scattered clusters that do not coincide with the bead implantation site ( Figure 7B , bottom ) , and Fgf8 beads do not induce clustering of cherry-overexpressing clones ( Figure 7D , bottom ) . Analysis of dmax at 15ss revealed a significant reduction for foxg1cherry-overexpressing clones in the presence of an Fgf8 bead compared to foxg1cherry-overexpressing clones in the presence of a PBS bead ( p = 0 . 03305 ) or cherry-overexpressing clones in the presence of an Fgf8 bead ( p = 0 . 00005 ) ( Figure 7E , third radial plot ) . Plotting dmax over time shows that the spreading behavior of foxg1cherry-overexpressing clones in the presence of an Fgf8 bead diverges from bud stage onwards when compared to cherry-overexpressing cells in the presence of an Fgf8 bead . At 15ss , dmax of foxg1cherry-overexpressing cells in the presence of a PBS bead shows a high degree of variability between individual clones , probably depending on their relative location to endogenous Fgf sources ( Figure 7F ) . These results show that foxg1-expressing cells preferentially cluster around Fgf sources , suggesting a positive feedback between Fgf-dependent regulation of foxg1 gene expression and sustained cohesion of foxg1-expressing cells close to Fgf sources .
Experiments in chick embryos have long suggested specification of the nasal-temporal axis of the retina at early stages of eye morphogenesis [36] , and our previous work indicated that nasal-temporal axis formation requires early Fgf8 signaling [25] . We now show that a combined Fgf8/3/24 signal along the dorsal-ventral axis of the neural tube fully controls the nasal-temporal subdivision of the future neural retina already at the onset of optic vesicle evagination . This unexpected , early orientation of the future nasal-temporal axis raises the question of how its final alignment with the anterior-posterior body axis is achieved . After the onset of evagination , the optic vesicle enters a phase of growth and lateral extension [37] . We show that during this process , the dorsal leaflet of the optic vesicle , which is facing the ectoderm , consists only of nasal retinal progenitors . Subsequently , the optic vesicle is transformed into the two-layered optic cup . At this stage , nasal retina progenitors in the outer layer gradually regress into their final axial position by epithelial compaction , concordant with late movement of temporal retina progenitors from the inner layer , around the distal ridge of the optic cup . The synchronization of these morphogenetic processes is crucial for the final nasal-temporal subdivision of the neural retina and could explain why fate-mapping experiments previously revealed an alignment of nasal-temporal cell positions along the medial-lateral axis of the optic cup [19] . Thus , formation of the nasal-temporal axis occurs as the results of three processes: ( 1 ) axis specification in the optic vesicle along the dorsal-ventral axis of neural tube , ( 2 ) morphogenetic axis reorientation by cell morphology changes and directed cell movements , and ( 3 ) 90° anterior rotation of the optic cup [37] , leading to the alignment with the anterior-posterior body axis ( Figure 8A ) . All three Fgfs involved in nasal-temporal pattern formation originate asymmetrically relative to the future axis , consistent with their role in nasal fate specification in the dorsal optic vesicle leaflet . Fgf8 and -3 are both expressed in the dorsal forebrain [38] , [39] and fgf24 in cells of the olfactory placode [26] , [40] . The graded transformation of nasal into temporal retina fates upon stepwise elimination of fgf8/3/24 and the nested expression of Fgf target genes in the optic vesicle , suggest a morphogen-like mechanism with continuous determination of cell positional identities along an Fgf gradient . This would be similar to the mechanism of action of Fgf signaling during mesoderm development [41] . Such a gradient could arise by propagation of Fgf8 and Fgf3 within the neuroepithelium , since early , the optic vesicle is contiguous with the dorsal forebrain . In contrast , the source of Fgf24 lies outside the neural tube , and it thus must signal vertically through the basal side of the optic vesicle neuroepithelium . Bead implantations into the tissue surrounding the optic vesicle indicate that Fgf8 can also signal effectively via the basal side of the neuroepithelium . Interestingly , Fgf signaling occurs via the basal side of the epithelium during otic vesicle invagination [42] . We thus favor the model of an extra-neuroepithelial gradient of Fgf8/3/24 during optic vesicle patterning . Bead implants show that the ventral optic vesicle leaflet—the future temporal retina—is also competent to respond to Fgfs . The spatially restricted response to Fgfs in the dorsal optic vesicle leaflet could , therefore , occur by limiting signal spreading . One possibility is that the tight apposition between cells of the pre-lens ectoderm and the distal optic vesicle ridge prevents Fgf spreading to ventral cells , thereby leading to a sharp signaling threshold that underlies the precise dorsal-ventral restriction of foxg1 and foxd1 expression in the optic vesicle . Recently , Fgf19 has been suggested to act during lens and retina development , possibly downstream of Fgf8 and Fgf3 [43] . Fgf19 loss- and gain-of-function experiments can alter retinal gene expression levels , but fail to produce any consistent nasal-temporal patterning shift independent of severe abnormalities in optic cup morphology . Together with its unrestricted and late expression , this makes a specific contribution of Fgf19 to nasal-temporal patterning very unlikely . The previously reported cooperation of Fgf8 and Fgf3 during retinal neurogenesis [44] appears to be unrelated to the function of Fgf8/3/24 during early pattern formation and eye morphogenesis ( unpublished data ) . Experiments in mouse and chick embryos have shown a mutually repressive interaction between foxg1 and foxd1 [10] , [45] . This is supported by the sharp foxg1/-d1 expression boundary at the distal ridge of the optic vesicle . Additionally , we find that foxg1 and foxd1 expression always occurs in a mutually exclusive pattern in both Fgf loss- and gain-of-function experiments . Similar to the mouse telencephalon [46] , [47] , foxg1 in the optic vesicle may be directly activated by Fgf signaling . Since Fgf signaling only regulates the development of nasal fates , temporal fate specification and foxd1 expression in the ventral optic vesicle leaflet could either occur as default , in the absence of Fgf signals , and/or as response to a different signal , potentially of ventral origin . Foxg1 and -d1 are likely to act as direct regulators of efn/eph guidance cue expression [10] , [11] . Although Efna5a and Epha4b are guidance cues for retinal ganglion cell axons , we show that their spatially restricted expression is already established in the optic vesicle , shortly after that of foxg1 and foxd1 and long before neuronal differentiation and retinotectal map formation . Given that Efns/Ephs are known effectors of cell sorting at lineage-restricted compartment boundaries [48] , a potential , early role for Efn/Eph signaling could be maintenance of a nasal-temporal lineage boundary between cells in the dorsal and ventral leaflets of the optic vesicle during growth and morphogenesis . Indeed , the presence of such lineage-restricted compartments along the dorsal-ventral axis of the retina has previously been suggested by fate mapping experiments in chick embryos [49] . Foxg1 is best studied for its function in maintaining neural progenitor proliferation [27]–[29] . We find that the optic vesicle contains 2-fold more mitotic , foxg1-positive , nasal retina progenitors than mitotic , foxd1-positive , temporal retina progenitors . This asymmetry depends on Fgf signaling and coincides with the strict Fgf requirement for foxg1 expression . One role for foxg1 could thus be maintenance of a high rate of neural progenitor proliferation , close to the source of Fgfs , similar to the role proposed for Fgfs in the vertebrate spinal cord stem zone [50] . Changes in motility and/or shape of individual cells must be tightly balanced with cell–cell adhesion to assure tissue integrity during epithelial morphogenesis , but how positional identities defined during pattern formation contribute to regional differences in morphogenetic cell behavior is poorly understood [51] , [52] . We find that local cell behavior and epithelial integrity during optic cup morphogenesis directly depend on correct prior patterning of the optic vesicle by Fgfs . When foxd1-expressing , temporal progenitors move into the optic cup , foxg1-expressing , nasal progenitors increasingly regress dorsoproximally , leading to the definitive alignment of nasal-temporal fates in the primordium of the neural retina ( Figure 8B , left ) . In the course of this morphogenetic movement , nasal progenitors appear increasingly immotile and elongated , whereas temporal progenitors in the inner optic cup layer appear motile and more cuboidal . There are two plausible , but not exclusive , explanations for the driving force behind the morphogenetic cell movement and cell shape changes in the optic cup: on one hand , an active movement of temporal cells could exert a force within the plane of the optic cup neuroepithelium , leading to the gradual compaction of nasal progenitors already residing in the future neural retina domain ( pushing force ) . Fgf8 bead implants indicate that the movement of cells from the ventral optic leaflet into the neural retina can occur independent of their Fgf-dependent nasal-temporal fate . Thus , if this movement is active , it appears to be under the control of a yet unidentified factor . Alternatively , since the optic cup epithelium is overall highly coherent , the compaction of future nasal tissue could tow temporal progenitors into the neural retina , independent of active , lateral cell movements ( pulling force ) . The higher rate of temporal progenitor movement observed in the absence of Fgf signaling suggests that intact cohesion of nasal progenitors at least restricts morphogenetic movements of temporal cells . Evidently , compaction by cell elongation can only occur if the nasal progenitors , already residing in the future neural retina domain , have a high degree of lateral cohesion . In support of this , we find that in the absence of Fgf signaling and Foxg1 activity , nasal progenitors start to delaminate exactly at the onset of temporal progenitor movement into the optic cup ( Figure 8B , right ) . Together with the ability of Foxg1 to rescue delamination in the absence of Fgf signaling and the strong effect of Foxg1 overexpression on lateral cell spreading/clustering , this suggests a novel role for Foxg1 as a positive regulator of neuroepithelial cell cohesion . When Foxg1 is ectopically expressed , cell cohesion is more pronounced in regions of high Fgf signaling activity . There could be several explanations for this . First , Fgf signaling might enhance Foxg1-dependent foxg1 expression , translation or posttranslational efficacy , such that Foxg1 function is more effective in an Fgf-signaling environment . Indeed , it has been recently shown that Fgf signaling posttranslationally regulates the subcellular localization of Foxg1 [53] , indicating a role for persistent Fgf signaling on Foxg1 function . Furthermore , there could be a feed-forward mechanism whereby Foxg1 enhances an independent role for Fgf signaling in mediating cell cohesion ( Figure 8C ) . In support of this , recent results suggest a requirement for Fgf signaling in epithelial organization during collective cell migration in the lateral line primordium [54] , [55] and perhaps to a lesser extent in the parapineal nucleus [56] . In the lateral line primordium , Fgf signaling may promote epithelialization and the formation of apical junctional complexes between the polarized epithelial cells [54] . A failure in maintaining neuroepithelial integrity and junctional complexes could certainly contribute to the observed extrusion of prospective nasal cells in the absence of Fgf signaling . Additionally , chemotactic effects of Fgfs [57] could limit nasal progenitor spreading and thereby indirectly lead to epithelial compaction . The future identification of Fgf- and Foxg1-regulated effectors of cell adhesion will advance the understanding of this mechanism . The phenotype of fgf8/3/24−/− embryos is reminiscent of the general loss of neuroepithelial integrity in N-cadherin mutants [58] , [59] . It is thus possible that Fgf-dependent foxg1 expression is locally required for cadherin-mediated cell adhesion . However , a disruption of apical-basal cell polarity does not appear to be the cause of cell delamination in fgf8/3/24−/− . In summary , by in vivo tracking , the development of one retinal axis from its specification , through morphogenetic rearrangement until its final orientation , we show that retinal pattern formation and morphogenesis are tightly coordinated processes . Considering the differences of optic vesicle morphology between vertebrate species [60] , [61] , it will be important to assess the conservation of morphogenetic axis reorientation . In first support of this , a recent fate-mapping study suggests that nasal-temporal cell positions initially align along the dorsal-ventral neural tube axis also in chick embryos [62] . Interestingly , in frog embryos , late cell movement from the optic stalk contributes to formation of the ventral neural retina [63] , suggesting comparable cell movements may shape the retina along both of its major axes .
Fish were maintained and bred according to standard procedures [64] . AB or tupl wild-type and aceti282a ( acerebellar/fgf8 ) , ikat22030 ( ikarus/fgf24 ) , and fgf3t24149 mutant fish were used for intercrossing and to breed [38] , [65] , [66] . Adult carrier fish and mutant embryos were identified by direct sequencing after PCR on genomic DNA , using the following primers: Fgf3-forward: 5′-TCTTCAACCGAGAGTGTGAGTTTCTA-3′ , Fgf3-reverse: 5′-CGCTGACTCTCTCTAAGCTTGCGC-3′ , Fgf8-forward: 5′-AGACGGACACATTTGGGAGTCGAGT-3′ , Fgf8-reverse: 5′- AAGTCACAAAAGTGATGACTTTTTCAGATA-3′ , Fgf24-forward: 5′- TTGTATTTTGCAGCTCTGCTTGTGGTC-3′ , Fgf24-reverse: 5′- TGTGGCTGTGTCCAGATGTTGTACG-3′ . The following transgenic lines were used: Tg ( -8 . 0cldnb:lynGFP ) zf106 , HGn42A , Tg ( hsp70l:dnfgfr1-EGFP ) pd1 , Tg ( h2afv:GFP ) kca66 , Tg ( Bactin:HRAS-EGFP ) vu119 [31]–[34] , [67] , and a line expressing DsRed2 under the control of the Xenopus ef1a promoter [68] . The cldnb gene is not expressed in the retina , and nasal retina expression is only present in the single Tg ( -8 . 0cldnb:lynGFP ) zf106 line ( D . Gilmour , personal communication ) , thus representing a positional effect . The nature of the trapped enhancer is currently unknown . Whole-mount mRNA in situ hybridizations were done as described [38] . Triple inactivation of fgf8 , -3 , and -24 was achieved by injecting an fgf24 antisense morpholino [26] ( MO ) into fertilized eggs from fgf8/3 double-mutant carrier crosses . The fgf24-MO was titrated to 1 nl/embryo of a 0 . 2 mM MO solution in 1× Danieau's medium with 0 . 02 mg/ml Fast Green FCF ( Fluka ) by comparing fgf24MO-injected , fgf8 mutant embryos with the phenotype of genetic fgf8/24 double mutants . FgfR-inhibitor treatment with 5 µM SU5402 ( Calbiochem ) in E3 medium was done between the 1- and 5ss stages on dechorionated embryos in agarose-coated 24-well plates . Controls were treated with 0 . 05% DMSO in E3 . Polystyrene beads ( 40 µm; Polysciences ) were loaded with 250 µg/ml recombinant zebrafish Fgf8 ( A . Picker , unpublished data ) or recombinant mouse Fgf8b ( R&D Systems ) and implanted as previously described [25] . Two foxg1 antisense morpholinos generated the same cell delamination phenotype: foxg1-MO1 ( 5′-CTTTTCTTTCTCCCATATCCAACAT-3′ [35] ) and foxg1-MO2 ( 5′-CCCATATCCAACATCACAAGTAAG-3′ ) [69] As control for foxg1-MO1 , a 5-mismatch morpholino was used ( 5′- CTaTTgTTTCTCgCATATgCAAgAT-3′ ) . All foxg1 morpholinos were injected at: 0 . 5–1 nl of MO/embryo ( 1 mM ) . For foxg1 overexpression , embryos were injected with 0 . 5 nl of in vitro–transcribed RNA ( 100 ng/µl ) encoding Foxg1cherry . The specificity of translation blocking by foxg1-MO1 was tested by coinjection with foxg1cherry RNA into Tg ( h2afv:GFP ) kca66 embryos which ubiquitously express GFP-tagged histone 2a . All cell transplantations were carried out between the 40% and 50% epiboly stages at the animal pole . Dnfgfr1-expressing cell clones were created by transplantation of cells from Tg ( hsp70l:dnfgfr1-EGFP ) pd1 embryos into the animal pole of wt embryos at late blastula stages . Chimeras were heat-shock induced by transfer into 37°C E3 medium at the 3ss stage and subsequently incubated at 28 . 5°C . Controls chimeras carrying clones from a hsp70l:eGFP transgenic line ( S . Hans , unpublished data ) were treated identically . Sets of 8-bit grayscale images of dissected , flat-mounted eyes were captured after in situ hybridization and imported into ImageJ 1 . 32j ( http://rsb . info . nih . gov/ij/ ) . Using the “Analyze>Mean” option , mean intensities in eight axial , 30-µm2 regions were measured and further analyzed with Microsoft Excel . Analysis was done on identically processed embryos . The following primary antibodies were used: anti-caspase3 ( 1∶500 , Abcam ) , anti-Phospho-Histone H3 ( PH3 ) ( 1∶500 , Upstate Biotechnology ) , anti-aPKC ( 1∶500 , Santa Cruz Biotechnology ) , anti-BrdU ( 1∶200 , Roche ) , and anti-ZO1 ( 1∶500 , Zymed/Invitrogen ) . The secondary antibodies used are: Alexa Fluor 488 goat anti-rabbit , Alexa Fluor 488 goat anti-mouse , and Alexa Fluor 633 anti-mouse ( all 1∶1 , 000 , Invitrogen ) . Counterstaining was done using DAPI ( 1 µg/ml , Invitrogen ) and Alexa Fluor 564 phalloidin ( 1∶400 , Invitrogen ) . For counting PH3+ cells in optic vesicles , forebrains were dissected from embryos after staining and transversely split into an anterior and posterior half , which were then flat-mounted . PH3+ cells in stacks of consecutive , transverse , single optical sections , captured at 3-8-µm intervals , were counted according to their localization in the evaginating vesicle . Data and statistical analysis for quantification of PH3 was carried out using MS Excel and a two-tailed Student t-test . BrdU treatments were done by incubating the embryos in 5 mg/ml BrdU and 15% DMSO , for 20 min on ice followed by 10 min at 28 . 5°C just prior to fixation . BrdU detection was done as previously described [70] . Live embryos were imaged with an upright Leica TCS SP5 confocal microscope using a 63× dipping lens after immobilization in 1 . 2% LMP agarose in embryo medium . A 40× UV-corrected lens was used for imaging fluorescently stained embryos . Image analysis and assembly was done with ImageJ , Metamorph , Volocity , and Leica LAS software . Injection of 25–50 pg in vitro–transcribed palmitoylated mRFP [71] or lynGFP RNA/embryo was used as in vivo membrane counterstain . At the time of this study , the zebrafish genome contained one gene annotated as foxd1 ( GenBank accession number: NM_131271 . 1 ) and one gene annotated as foxd1 like ( GenBank accession number: NM_212913 . 1 ) . Phylogenetic sequence analysis revealed that zebrafish foxd1 like is orthologous to other vertebrate foxd1 genes , whereas zebrafish foxd1 is orthologous to foxd2 ( Figures S11 and S12 ) . Therefore , foxd1 like is referred to as foxd1 throughout this study . | The vertebrate brain contains a point-to-point representation of sensory input from the eye . This visual map forms during embryonic development , by neuronal cells of the retina sending targeted axon projections to the brain . Since the projection needs to wire up neighboring cell positions in the retina to neighboring target areas in the brain , all retinal cells must harbor a defined spatial coordinate as prerequisite for map formation . How such a retinal coordinate system is established and maintained in the dynamically evolving embryo is a fundamental , but unresolved , question . By combining genetic analysis and in vivo imaging in zebrafish embryos , we have tracked the developmental origin of cell coordinates in the retina . We find that three related Fgf signals emanating from outside the eye define relative cell positions in the retina very early , already at the onset of its formation . But the absolute position of retinal cells relative to the body axes is greatly rearranged during subsequent development . In this phase , surprisingly , the same Fgf signals that at first defined retinal cell positions now balance asymmetric cell movements and cell shape changes , which are required for harmonic retinal growth and the final alignment of cell coordinates in the eye . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"developmental",
"biology/embryology",
"genetics",
"and",
"genomics/animal",
"genetics",
"neuroscience/sensory",
"systems",
"developmental",
"biology/morphogenesis",
"and",
"cell",
"biology",
"cell",
"biology/cell",
"signaling",
"cell",
"biology/morphogenesis",
"and",
"cell",
"biology",
"cell",
"biology/developmental",
"molecular",
"mechanisms",
"developmental",
"biology/pattern",
"formation",
"neuroscience/neurodevelopment",
"developmental",
"biology/neurodevelopment",
"developmental",
"biology/molecular",
"development",
"cell",
"biology/cell",
"adhesion",
"developmental",
"biology/developmental",
"molecular",
"mechanisms",
"cell",
"biology/gene",
"expression"
]
| 2009 | Dynamic Coupling of Pattern Formation and Morphogenesis in the Developing Vertebrate Retina |
We describe here the Drosophila gene hydra that appears to have originated de novo in the melanogaster subgroup and subsequently evolved in both structure and expression level in Drosophila melanogaster and its sibling species . D . melanogaster hydra encodes a predicted protein of ~300 amino acids with no apparent similarity to any previously known proteins . The syntenic region flanking hydra on both sides is found in both D . ananassae and D . pseudoobscura , but hydra is found only in melanogaster subgroup species , suggesting that it originated less than ~13 million y ago . Exon 1 of hydra has undergone recurrent duplications , leading to the formation of nine tandem alternative exon 1s in D . melanogaster . Seven of these alternative exons are flanked on their 3′ side by the transposon DINE-1 ( Drosophila interspersed element-1 ) . We demonstrate that at least four of the nine duplicated exon 1s can function as alternative transcription start sites . The entire hydra locus has also duplicated in D . simulans and D . sechellia . D . melanogaster hydra is expressed most intensely in the proximal testis , suggesting a role in late-stage spermatogenesis . The coding region of hydra has a relatively high Ka/Ks ratio between species , but the ratio is less than 1 in all comparisons , suggesting that hydra is subject to functional constraint . Analysis of sequence polymorphism and divergence of hydra shows that it has evolved under positive selection in the lineage leading to D . melanogaster . The dramatic structural changes surrounding the first exons do not affect the tissue specificity of gene expression: hydra is expressed predominantly in the testes in D . melanogaster , D . simulans , and D . yakuba . However , we have found that expression level changed dramatically ( ~ >20-fold ) between D . melanogaster and D . simulans . While hydra initially evolved in the absence of nearby transposable element insertions , we suggest that the subsequent accumulation of repetitive sequences in the hydra region may have contributed to structural and expression-level evolution by inducing rearrangements and causing local heterochromatinization . Our analysis further shows that recurrent evolution of both gene structure and expression level may be characteristics of newly evolved genes . We also suggest that late-stage spermatogenesis is the functional target for newly evolved and rapidly evolving male-specific genes .
Much of the genetic novelty that accompanies speciation and organismal evolution is driven by reutilization of pre-existing genetic information . In an influential essay , Francois Jacob likened evolution to a process of tinkering [1] . After the primordial evolution of truly new macromolecules and mechanisms of replication , Jacob suggested that much of phenotypic novelty arises from reusing , recombining , and altering the function of available genes . This view of evolution is strongly supported by studies of new-gene evolution . The vast majority of newly evolved genes can be attributed to duplication of pre-existing genes . These duplication events can range from single-gene events to duplications of entire genomes [2 , 3] . Much of protein evolution also conforms to Jacob's view , as new proteins are often generated from shuffling pre-existing protein domains [4] . The tinkering view of evolution does not rule out the occasional generation of novel protein sequences . One question then is whether and by what mechanisms such novel proteins evolve . Presumably novel proteins derive from noncoding sequences that acquire appropriate transcriptional and translational regulatory sequences . Experimental evolution studies suggest that random sequences of proteins can acquire biological functions at frequencies that are , surprisingly , greater than miniscule [5] . A recent study reported the exciting finding of several such candidate de novo genes in Drosophila that may be functional , based on RNA expression analysis [6] . A large number of new exons have also been identified in rodents that appear to have derived from incorporation of intronic sequences into mRNAs [7] . A large fraction of eukaryotic genomes is composed of transposable elements ( TEs ) . Because the TE component of genomes can evolve rapidly , TEs have a major impact on genome evolution [8–10] . TEs are widely considered to be selfish parasites that are deleterious to their host's fitness [11] , and most are likely eliminated quickly by natural selection . However , accumulating evidence has also shown that TEs can serve as an important source of genetic variation and genomic novelty [8 , 12–14] . TEs can generate genetic novelty by at least five different mechanisms . First , the reverse transcriptase enzyme from retrotransposons can generate duplicated retroposed genes [15–17] or generate chimeric fusion genes [15 , 18–21] . Second , TEs can change expression patterns of adjacent genes by providing novel regulatory elements or by disrupting host gene regulatory functions [14 , 22–26] . Third , coding regions from TEs such as envelope proteins and transposase enzymes can be incorporated by the host species to modify or even create new host protein-coding genes [13 , 27] . Fourth , TEs can contribute to noncoding regions ( untranslated regions [UTRs] ) of genes [12 , 24 , 28–30] . Fifth , TE insertions create multiple highly homologous sites throughout the genome that can cause duplications by ectopic recombination [9 , 31] . Comparisons between genomes have proven to be a powerful way to identify new genes [32] and to reconstruct the early history of new-gene evolution [2] . This approach has provided new data for models and mechanisms leading to new-gene evolution [6 , 20 , 33–35] . Such a comparative approach is used here to investigate the evolutionary history of a recently evolved gene in Drosophila . We report the characterization of the gene hydra , located in the pericentric region of the X chromosome in D . melanogaster . hydra was originally reported as predicted gene CG1338 [36] . We show that hydra originated in the melanogaster subgroup of Drosophila and subsequently experienced dramatic changes in its gene structure , including the formation of multiple alternative first exons , most of which are associated with the transposon DINE-1 ( Drosophila interspersed element-1 ) . We also investigate the history and impact of these gene-structure changes at both the DNA sequence and RNA expression levels by comparison among species of the melanogaster subgroup . We suggest that new genes will tend to evolve rapidly in coding sequence ( CDS ) , gene structure , and gene-expression level . We also suggest that TE accumulation can cause local heterochromatinization that can in turn have a major impact on the evolution of gene structure and gene expression .
CG1338 is located near the pericentric region of the X chromosome in D . melanogaster ( cytological region 19E1 ) . We first became interested in CG1338 during a genome-wide analysis of the transposon DINE-1 in the D . melanogaster genome . We present evidence below that CG1338 contains nine duplicated first exons in D . melanogaster . Because of these nine duplicated exon 1s ( see Figure 1A ) , we propose to rename this gene hydra , after the nine-headed monster slain by Hercules . We found that the gene region surrounding hydra in D . melanogaster contains dense repeats of DINE-1 ( Figure 1A ) . DINE-1 , also named INE-1 or DNAREP1_DM , is a highly abundant transposon that is predominantly found in the heterochromatic regions of the D . melanogaster genome [37 , 38] and is believed to have invaded the Drosophila genome before the diversification of the melanogaster subgroup [37–40] . Further analysis showed that hydra is flanked by two blocks of tandem repeats , with duplicated fragments of DINE-1 in the 5′ end and an undefined 500-bp repeat in the 3′ end . Using available genomic contigs from various Drosophila species ( http://flybase . net/blast ) , we found that the ~7-kb region surrounding hydra , including the adjacent gene CG1835 , only exists in species of the melanogaster subgroup , but not in other species outside of the subgroup . BLAST searches using unique sequences from hydra identified no similar sequences in other regions of these genomes . We suggest that hydra evolved recently in the species of the melanogaster subgroup ( Figure 1B ) . This conclusion was confirmed by Southern hybridization under low stringency using hydra exons 2–4 from D . melanogaster as the probe with DNA extracted from D . melanogaster , D . simulans , D . yakuba , D . erecta , D . ananassae , D . pseudoobscura , and D . virilis . Hybridization signals were only detected in flies of the melanogaster subgroup ( i . e . , D . melanogaster , D . simulans , D . yakuba , D . erecta ) , but not in the other species ( unpublished data ) . TBLASTN searches to the nonredundant ( nr ) or whole-genome shotgun ( wgs ) databases found no hits to non-melanogaster subgroup Drosophila species with E values < 0 . 001 . According to the Flybase genome annotation ( http://www . flybase . org ) , hydra of D . melanogaster contains four exons , and two types of transcripts ( type A and type B ) , which are transcribed using alternative first exons . Our analysis of the hydra gene region suggested that there are seven additional putative first exons similar to the two annotated first exons located in a region of dense DINE-1s in D . melanogaster ( Figure 1A; Text S1 ) . These seven additional potential first exons share similarity with the two annotated exon 1s in their CDSs ( which are nine codons long ) and in their 5′ upstream regions . The first of these duplicated exons may be nonfunctional because it creates a stop codon when spliced to exon 2 , assuming that it has the same structure and splicing pattern as the other 8 exons . Alternatively , this first exon 1 may use a different 3′ splice site , although no such evidence was found in our 5′ rapid amplification of cDNA ends ( RACE ) analysis described below ( Table 1 ) . Our further analysis of multiple lines of D . melanogaster ( and D . simulans ) with Southern hybridization detected substantial length variation in the gene region , presumably due to length variation in the exon 1 region ( Figure 2 ) . Although we have not determined the sequence structure of these alleles , the length variation observed is consistent with the hypothesis that the number of alternative first exons varies in D . melanogaster . In contrast , there is only a single exon 1 in the orthologous region of hydra from the species D . yakuba and D . erecta ( Texts S5 and S6 ) , and there is no length polymorphism in this region in D . yakuba ( unpublished data ) . This suggests that the dramatic gene structure change by duplication of exon 1 in hydra occurred within ~5–13 million y , after the divergence of D . melanogaster and D . yakuba [41 , 42] . By comparing the orthologous sequences of hydra using the released assembled genomic sequences of D . simulans and D . sechellia , we found that this region has undergone local duplications in both species ( Figure 1B; Texts S2–S4 ) . Determining the precise structure of the hydra region in D . sechellia will require further sequence analysis due to low sequence coverage of this region . We have , however , confirmed by Southern analysis that D . simulans does have two copies of hydra ( Figure 2C ) . In D . simulans , we find that one of these copies has two potential exon 1s . In D . sechellia , we identified two unlinked contigs containing hydra-homologous sequences ( Figure 1B ) . Scaffold 8 contains the syntenic genes run and cyp6v1 flanking a single copy of hydra . This copy of hydra contains two putative exon 1s . Scaffold 600 contains two tandem copies of hydra , each of which contains two or three putative exon 1s . These data suggest the possibility that D . sechellia has three copies of hydra , although further analysis will be required to determine the relative locations of these apparent duplicated copies that are on different scaffolds . Despite this uncertainty , we were able to investigate the phylogenetic relationship of these copies of hydra . Using either the sequence between exons 2–4 , inclusive ( Figure 3A; Text S7 ) or of intron 1 ( unpublished data ) , we obtained similar phylogenetic trees that suggest that hydra duplicated independently in D . simulans and D . sechellia . Because both gene regions produce the same phylogeny and these sequences span over 1 , 300 bp in all species , we consider it unlikely that this phylogenetic signal of independent duplications could be due to gene conversion . Upon further investigation of the intronic sequences next to exon 1 , we found that seven out of the nine duplicated exon 1s in D . melanogaster and several of the duplicated exon 1s in both D . simulans and D . sechellia are immediately followed by DINE-1 sequences , while in other species of the melanogaster subgroup , including D . yakuba and D . erecta , there are no DINE-1s in this region . We thought it likely for two reasons that insertion of DINE-1 adjacent to hydra exon 1 occurred in the common ancestor of D . melanogaster , D . simulans , and D . sechellia . First , it seems highly unlikely that a TE would insert independently at the same location in different species . Second , DINE-1 is thought to have been active and then to have become transpositionally inactive before these species diverged [38 , 40] . To test this hypothesis and to understand the evolutionary history and mechanism that resulted in the exon 1 duplications of hydra , we performed phylogenetic analysis of all the available exon 1s ( Figure 3B; Text S8 ) . We found that D . simulans S2 and D . sechellia C3 are outgroups to all other exon 1s . Both of these exons are flanked by DINE-1 sequence , which is consistent with our above suggestion that DINE-1 insertion in this region is ancestral ( at time B in Figure 1B ) . If this inference of the ancestral state of exon 1 in these three species is correct , it is interesting to note that DINE-1 does not flank every exon 1 . Our data would then suggest that either DINE-1 was lost next to some duplicated exon 1s or that some exon 1s duplicated without duplicating the adjacent DINE-1 . The relationships among the remaining D . simulans and D . sechellia exon 1s suggest that exon 1 duplicated before the speciation of D . simulans and D . sechellia to generate the exon 1 groups S1/S3 and C1/C4/C6 , with another duplication event generating exon 1 group C2/C5/C7 ( Figure 3B ) . Both this tree and the one made using the rest of the locus ( Figure 3A ) suggest that after these exon 1 duplications , the entire locus duplicated independently in D . simulans and D . sechellia . Figure 3B also suggests that exon 1 duplicated independently in D . melanogaster . Based on the sequence similarity of predicted exon 1 sequences , the duplicated first exons of D . melanogaster can be divided into three groups: group A , containing the first , second , third , and fifth duplicated exon 1s; group B , containing the fourth , seventh , eighth , and ninth duplicated exon 1s; and group C , containing the sixth duplicated exon 1 . The first and the ninth exon 1s are the previously annotated exon 1s for transcript types RA and RB , respectively . Our phylogenetic analysis showed that exon 1s adjacent to each other are generally more closely related ( Figure 3B ) . These results suggest that both the exon 1s and their flanking DINE-1s were duplicated together , and that unequal crossing-over is the major mechanism generating the tandemly duplicated exons . RT-PCR analysis demonstrated that hydra is expressed in adult male testes in all three species tested: D . melanogaster , D . simulans , and D . yakuba ( Figure 4 ) . This result demonstrates that the expression pattern of hydra is conserved despite having a very different gene structure in these species . In one out of two stocks of D . melanogaster that we examined , we also detected a low level of expression in the male carcass from which testes were removed ( Figure 4B ) . This expression pattern is consistent with D . melanogaster EST data: among ten ESTs in GenBank that match hydra , nine are from testis cDNA libraries , and one is from a larval fat-body library . We further investigated the pattern of expression of hydra in D . melanogaster by in situ hybridization . We found that hydra is expressed most intensely in the proximal region of the testis ( Figure 5 ) , where mature sperm are formed , suggesting that hydra functions in late stages of sperm development . To study the impact of exon 1 duplication on gene expression , we asked if each of the duplicated exons could function as a transcriptional start site . To determine whether each putative exon 1 is functional in D . melanogaster , we needed to use a technique that could distinguish these very similar potential transcripts . We therefore performed 5′ RACE analysis using RNA extracted from adult male testes of D . melanogaster and analyzed the sequences of 58 clones . The previously described Ra and Rb transcripts differ by three nucleotides at positions −21 to −23 in their 5′ UTRs as well as by a nonsynonymous polymorphism in their second codon . We designate alternative first exons 1 , 2 , 3 , and 5 as class A ( similar to Ra ) , and alternative first exons 4 , 6 , 7 , 8 , and 9 as class B ( similar to Rb ) . Some of these sequences can be further distinguished by additional polymorphisms further 5′ in the 5′ UTRs . Six different groups of alternative transcripts could be potentially distinguished , with two of these groups containing two or three identical first exons ( Table 1 ) . Among our 58 5′ RACE clones , four out of six of these groups were represented at least twice , demonstrating that at least four different alternative first exons in hydra are functional . We also analyzed to see if each duplicated first exon contains core promoter sequences known to be important for the initiation of transcription of Drosophila genes [43] . We looked specifically for the TFIIB recognition element ( −37 to −32 bp , GGGCGCC or CCACGCC ) , TATA box ( −31to −26 bp; TATAAA ) ; initiator ( −1 to +4 bp; TCA ( G/T ) T ( T/C ) ) , and downstream promoter element ( +28 to +32 bp; ( A/G ) G ( A/T ) ( G/T ) ( G/A/C ) ) . We found that almost all duplicated exon 1s contain all four core promoter sequences , with two exceptions: the sixth and the ninth duplicated exon 1s do not contain the downstream promoter element and the TATA box , respectively . Since both of these exons are used ( Table 1 ) , our results suggest that these two elements are not essential for transcription of hydra . Using the sequences of the clones from 5′ RACE , we also mapped the sites for transcription initiation . We found that most transcripts of hydra do not use a fixed site for transcription initiation . Instead , “slippery promoters” were found to be used in different exon 1s , as described for other Drosophila genes by Yasuhara et al . [44] . We also tested whether hydra structural evolution might lead to expression-level differences between sibling species by quantifying hydra expression levels in five wild lines of D . melanogaster and in four wild lines of D . simulans ( Figure 6 ) . Strikingly , hydra was expressed at a substantially higher level in all D . simulans lines compared with all D . melanogaster lines . Among all pair-wise comparisons , the fold increase in D . simulans ranged from ~4 to ~150 , with the mean fold difference being ~22 . These data indicate that hydra has lower expression in D . melanogaster , despite having multiple alternative transcription start sites . hydra contains an intact reading frame in all melanogaster subgroup species , encoding a predicted protein of 302 , 280 , and 308 amino acids in D . melanogaster , D . simulans , and D . yakuba , respectively . We calculated Ka/Ks ratios of the coding region of hydra among species in the melanogaster subgroup ( Table 2 ) . The ratio is <1 in all comparisons , suggesting that hydra is evolving under some degree of functional constraint . To gain further insight into the evolution of hydra , we collected population samples from D . melanogaster and D . simulans , which have the highest interspecific Ka/Ks ratio of 0 . 75 ( Texts S9 and S10 ) . We found that the nucleotide diversity at both synonymous ( πs ) and nonsynonymous ( πa ) sites in D . melanogaster is similar to the average of other functional genes , with about ten times more polymorphism at synonymous sites relative to nonsynonymous sites ( Table 3 ) . In contrast , there is not much difference between πs and πa in D . simulans , where πs is much lower than the average of other genes for this species [45] . In order to further investigate whether this unusual pattern of nucleotide diversity ( similar levels of synonymous and nonsynonymous polymorphism ) in D . simulans is lineage specific , we studied polymorphism in D . yakuba . πs and πa of hydra in D . yakuba are similar to those of D . melanogaster , and πs is similar to the average value ( 0 . 0127 ) of six X-linked loci in D . yakuba [46] , suggesting that the unusual pattern of nucleotide diversity of hydra is restricted only to D . simulans . The approximately equal values of πs and πa in D . simulans may suggest that hydra is under reduced functional constraint in D . simulans . Two findings suggest , however , that hydra is not a pseudogene in D . simulans . First , it has an intact open reading frame , and second , hydra is expressed in D . simulans and has a similar expression pattern to other species ( see above ) . Results of the McDonald-Kreitman ( MK ) test [47] among these three species are shown in Table 4 . Neutral evolution of hydra can be rejected between D . melanogaster and D . yakuba , which we suggest is caused by the large number of nonsynonymous substitutions between these species . There is also a large amount of nonsynonymous divergence between D . melanogaster and D . simulans . We do not , however , reject neutrality in this case , likely due to the unusually low πs in D . simulans discussed above ( Table 3 ) . Thus , these results suggest that D . melanogaster is the most likely lineage in which hydra has experienced positive selection . To confirm this conclusion , we tested to see if the nonneutral evolution of hydra occurred in a lineage-specific pattern by performing polarized substitution analysis with the MK test ( Table 4 ) . Replacements were polarized to each species lineage using D . yakuba hydra as an outgroup sequence . The polarized MK test is significant only in the D . melanogaster lineage .
We identified the gene hydra in D . melanogaster , but could only identify orthologs within other melanogaster subgroup species . Using both BLAST searches of full-genome sequences with the 6-kb region flanking hydra and Southern hybridization , we were unable to identify hydra from D . ananassae , D . pseudoobscura , or D . virilis . We were , however , able to identify the syntenic region in these species ( Figure 1B ) . These data strongly suggest that hydra originated in the common ancestor of the melanogaster subgroup . All available evidence also suggests that hydra is a functional gene . First , all melanogaster subgroup species contain intact open reading frames with no stop codons , even though there are many insertions and deletions in the hydra CDS among species . Second , Ka/Ks values between all species are less than 1 , suggesting that hydra is under functional constraint ( Table 2 ) . Third , the level of polymorphism of hydra in D . melanogaster is similar to known functionally important genes ( Table 3 ) . Fourth , RT-PCR and in situ hybridization show that hydra is predominantly expressed in male testes ( Figures 4 and 5 ) . Several other cases of new-gene evolution in Drosophila have been studied [2 , 48] . Examples include ( 1 ) Jingwei , a chimeric gene that evolved as a result of the insertion of an Adh retrosequence into a duplicated locus in the D . teissieri/D . yakuba lineage [19]; ( 2 ) Sdic , which originated through a complex set of rearrangements , including a gene fusion between the gene encoding the cell adhesion protein annexin X and a cytoplasmic dynein intermediate chain ( Sdic encodes a novel sperm-specific dynein found only in D . melanogaster ) [49]; ( 3 ) Sphinx , a recently evolved gene apparently created by both retroposition and exon shuffling , which produces a novel noncoding RNA gene found only in D . melanogaster [33]; ( 4 ) mkg , a gene family generated originally as retroposed duplicates about 1–2 million y ago that further evolved by individual members undergoing gene fission and fusion events [35]; ( 5 ) K81 , a gene that arose through duplication by retroposition after the radiation of the melanogaster subgroup and acquired a new male germline–specific function [50]; ( 6 ) Dntf-2r , a rapidly evolving male-specific gene in D . melanogaster and sibling species , which originated from retroposition [34]; and ( 7 ) five protein-coding genes which appear to have originated in D . melanogaster from noncoding DNA and evolved male-biased expression [6] . Despite the diverse mechanisms involved in the creation of these young genes in Drosophila , there are two features common to most of them: ( 1 ) they evolve rapidly in sequence , including at nonsynonymous sites for protein-coding genes [19 , 20 , 34 , 49 , 51 , 52]; and ( 2 ) their expression is often male-specific , and in many cases testis-specific [6 , 52] . This expression pattern was also often found in a large survey of new retroposed genes [17] . hydra shares both of these characteristics . The Ka and Ks values between D . melanogaster and D . simulans were substantially higher than the average values of most other genes , and the Ka/Ks ratio ranged from 0 . 44 to 0 . 75 in all species compared ( Table 2 ) . These high values may be explained by accelerated functional divergence following the origin of new genes . However , the elevated Ka/Ks ratio could also result from an elevated mutation rate and reduced selective constraint . To further investigate the evolution of hydra , we analyzed the level of polymorphism and divergence among D . melanogaster , D . simulans , and D . yakuba . Our population genetics analyses rejected the null hypothesis of neutral evolution in D . melanogaster ( Table 4 ) . This rejection appears to be due to a significant excess of fixed differences at nonsynonymous sites in D . melanogaster , suggesting that positive selection has driven the evolution of hydra in D . melanogaster . The male-specific expression pattern is a second feature that hydra shares with other new genes . For many newly evolved genes , male-specific or testis-specific expression has been demonstrated by RT-PCR , but the spatial expression patterns are unknown , making it difficult to refine our knowledge of what aspects of testis development or spermatogenesis may be under selection . We determined that hydra is expressed most intensely in the basal ( proximal ) part of the testis in D . melanogaster by in situ hybridization ( Figure 5 ) . This localization pattern suggests that hydra may have a function in late-stage spermatogenesis or sperm differentiation . We note that hydra is X-linked , which suggests that it escapes from the X inactivation that occurs during spermatogenesis [53] . Our results also stand in contrast to results from genome-wide studies that suggest that genes with male-biased expression as well as newly evolved testis-specific genes tend to be autosomal [17 , 54 , 55] . Interestingly , the newly evolved testis-expressed genes Sdic and CG15323 are also located in polytene region 19 of the X chromosome [6 , 49] . Several studies have shown that genes with male-biased expression have significantly faster rates of evolution than genes with female-biased or unbiased expression [56 , 57] . This difference is caused primarily by a higher Ka in the male-biased genes . Sexual selection is likely to be driving this acceleration of male-biased genes , but functional tests will be required to understand fully the role of selection . Our study of hydra suggests that late-stage spermatogenesis may be the key developmental process to investigate further for functional differences between species . Our analysis of hydra reveals several aspects that are unusual or unique compared with other newly evolved genes . One novel feature of hydra is that it has experienced recurrent structural evolution in the lineages leading to both D . melanogaster and D . simulans ( Figure 1 ) . We can infer the likely occurrence of three independent structural changes to hydra . One is a duplication of exon 1 in the common ancestor of D . simulans and D . sechellia . Second is a duplication of the whole locus in D . simulans and D . sechellia . Although uncertainties remain about the structure of the hydra region in D . sechellia , our phylogenetic analyses suggest that hydra duplicated independently in these two species . Third is the further duplication of seven additional exon 1s in D . melanogaster . Our phylogenetic analysis suggests that these additional duplication events occurred only in D . melanogaster , rather than in the common ancestor of D . melanogaster , D . simulans , and D . sechellia , followed by loss in the latter two species . It is possible , however , that the phylogenetic signal could be erroneous if homogenization occurred by continual gene conversion and/or unequal crossing over among these short duplicated exons . The duplicated exon 1 in D . simulans and D . sechellia and most of the additional exon 1 duplications in D . melanogaster are flanked by partial DINE-1 elements . DINE-1 is not associated with hydra in D . yakuba or D . erecta . Genome-wide analyses of DINE-1 suggest that DINE-1 was transpositionally active and then silenced in the common ancestor of D . melanogaster and D . yakuba , followed by another round of activity and silencing in D . yakuba [39] . These findings would suggest then that DINE-1 inserted ancestrally in hydra , followed by loss in the D . yakuba lineage but not the D . melanogaster lineage . An alternative scenario is that DINE-1 may have inserted in hydra during a period of activity in the D . melanogaster/D . simulans ancestor after divergence from D . yakuba that was too brief to have a left a genome-wide footprint . Regardless of the timing of the initial DINE-1 insertion into hydra , the further duplications of exon 1 in D . melanogaster were unlikely to have occurred by DINE-1 transposition because alternative exon 1s two through eight contain UTR and CDSs that are adjacent to rather than within the DINE-1 sequences . As noted above , the phylogenetic relationships among these alternative exons 1s suggest that they originated by unequal crossing over . DINE-1 may have promoted these duplications by increasing the homology among exon 1s . While we lack evidence for a direct role of DINE-1 in driving hydra structural evolution , our results do suggest that a property of transposable elements that is usually considered to be deleterious—their ability to cause ectopic recombination and unequal crossing over—may also contribute to structural evolution of genes . We have observed other cases of structural variation associated with DINE-1 insertions in the melanogaster species complex ( unpublished data ) . For most genes , evolving under strong functional constraint , a DINE-1 insertion that causes structural changes in the locus is likely to be highly deleterious . However , for a newly evolved gene like hydra , there may be more flexibility in terms of tolerating gene structure changes . We also suggest that as a gene evolves in structure , it may accelerate the rate of CDS evolution . More generally , our evidence that hydra has undergone multiple and independent structural changes suggests that gene structure as well as CDSs may evolve rapidly in new genes , even without any association with transposable elements . One other example is the monkey king gene family in Drosophila , which has undergone rapid structural changes caused by gene fission [35] . A second novel aspect compared with other studies of newly evolved genes is that hydra expression level has dramatically changed between sibling species . Our quantitative RT-PCR results showed that the expression level of hydra between D . melanogaster and D . simulans is highly different ( Figure 6 ) . hydra in D . melanogaster is expressed at least 20-fold less on average than in D . simulans . Why is the expression level so different between these two closely related species ? Two hypotheses may explain the decreased hydra expression in D . melanogaster . First , it may be caused by competition or interference among the tandemly duplicated heads during the transcription initiation stage . In this model , incomplete transcription initiation complexes would form at multiple heads . Reduced transcription initiation could result from either sequestration of transcription factors or direct interference among promoters . Second , it may be the result of differences in the degree of heterochromatinization of the hydra region between D . melanogaster and D . simulans . hydra is located near the pericentromeric region of the X chromosome . This region may be under continuous pressure from encroachment of the nearby heterochromatic pericentromere , leading to species-specific differences in chromatin states . Increased numbers of TE insertions are one characteristic of heterochromatic regions , and the large number of DINE-1 insertions in D . melanogaster is consistent with the hypothesis that the hydra region is partially heterochromatic . Our views on the evolution of genome structure and function have changed dramatically in the past decade as more whole-genome sequences have become available in various species . Repetitive elements are a major cause of structural changes and instability in genomes . At the cytological level , chromosome rearrangements often occur at repetitive sites [31 , 58 , 59] . Our identification of the hydra gene suggests that repetitive elements also contribute to the formation and evolution of new genes by causing structural changes in gene organization , driving the evolution of new exons and transcription start sites , and promoting divergence in gene expression . One of our most striking findings is that hydra appears to have been created de novo rather than being a duplicate of a pre-existing gene . Two pieces of evidence support this hypothesis . First , hydra ( and CG1835 ) are found only in melanogaster subgroup species . The surrounding ~26 kb of DNA is found in other Drosophila species that lack hydra ( Figure 1B ) . Second , hydra shares no homology with any known or predicted proteins . Arguing against this hypothesis of de novo origin is the fact that all forms of hydra contain four predicted exons . One might expect that a recently evolved de novo gene would have a simpler one-exon structure , as has been observed for several candidate de novo genes in D . melanogaster [6] . Nevertheless , if hydra does have an ortholog in Drosophila outside of the melanogaster subgroup , it must have evolved so rapidly as to share no detectable similarity with any predicted proteins , including in multiple Drosophila genomes or other insect genomes that have been completely sequenced . If there is a highly diverged hydra parental gene or ortholog in other Drosophila , it also must have transposed to a novel genomic location in the melanogaster subgroup . Considering all the available data , we suggest instead that hydra evolved de novo from DNA sequence that inserted between run and cyp6v1 in the common ancestor of the melanogaster subgroup . The most common fate of newly formed genes is disappearance . For example , about 85% of the genes duplicated during the yeast whole-genome duplication have been lost in Saccharomyces cerevisiae [3 , 60] . New genes therefore can be viewed as analogous to mutations , with most being neutral or deleterious in effect and therefore subject to elimination . A common characteristic of successful new exons and new genes is a high rate of CDS evolution [7 , 20] , and hydra shows this evolutionary pattern . Our discovery of hydra suggests that gene expression level and gene structure may also evolve rapidly and recurrently in new genes . We suggest that the successful retention of new genes requires maximal variation .
Flies used in this study include: ( 1 ) isofemale lines of D . melanogaster ( WI ) from Sergey Nuzhdin ( University of California Davis , Winters , California , United States ) ; ( 2 ) strains of D . melanogaster from worldwide populations: California ( CA ) , Canton S ( CS ) , Taiwan ( TWN ) , Oregon R ( ORC ) , France 3V-1 ( Fr3V-1 ) ( gift of S . C . Tsaur , Academia Sinica , Taiwan ) ; ( 3 ) isofemale lines of D . simulans from African populations ( mad , sz , su , zk lines; gift of John Pool , Cornell University , Ithaca , New York , United States ) , and one line each from France ( W54 ) and Japan ( W60 ) ; ( 4 ) isofemale lines of D . yakuba ( cy lines ) from Africa ( gift of Peter Andolfatto , University of California San Diego , United States ) ; and ( 5 ) strains of Drosophila species , including D . yakuba ( Tai18E2 , and 286 . 82–90 ) , D . erecta , D . ananassae , D . virilis , and D . pseudoobscura ( gift of S . C . Tsaur ) . DNA was extracted from pools of ~50 male flies from each line by a standard phenol-chloroform extraction followed by ethanol precipitation . Primers to amplify and sequence the hydra gene regions from D . melanogaster , D . simulans , and D . yakuba were designed using the homologous sequences obtained from BLAST searches against Flybase genome databases ( http://www . flybase . org/blast ) . Primers for both D . melanogaster and D . simulans are IN1_F ( 5′-GCGTTTGTACACTTGGCAAC ) and 3UR ( 5′-TGATGTAGGAATATGCGTTGC ) , and for D . yakuba are yakE2_F ( 5′-CTTCAGAGAACCCCAACCAA ) and yakE3_R ( 5′-CTTGCAATTATCCGGATTGC ) . In D . simulans , the primer pair specifically amplifies the second duplicate of hydra , which is adjacent to the S3 alternative exon 1 ( Figure 1B ) . Other primer sequences and PCR amplification conditions are available upon request . PCR products were directly sequenced in both direction using ABI BigDye ( Applied Biosystems , http://www . appliedbiosystems . com ) technologies under slight modification of the manufacturer's suggested protocols . All alleles were sequenced on both strands . Sequences are deposited in GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) under accession numbers EF596837–EF596877 . Genomic DNA was extracted from 60–70 flies of each line with phenol-chloroform , was ethanol precipitated , and was diluted to a final concentration of about 5 μg/μl . A total of 15 μg DNA was then digested separately with SacII , separated on a 0 . 8% agarose gel , and transferred to a nylon membrane by Southern blotting . Probes were from PCR product using DNA template from D . melanogaster with primers hydra_rep500_R2 ( 5′-CAGT ( T/C ) AGACCTCTCTGAAATC ) and E2_146R ( 5′- GGTTGGAATCCTCTGGAGTGTTGG ) , and were prepared by DIG-labeling Nick translation kit ( Roche , http://www . roche . com ) . Hybridizations were done at 50 °C . Testes of adult males aged 2–5 d old were dissected in PBS buffer and immediately transferred to PBS buffer on ice . RNAs were extracted with Trizol ( Invitrogen , http://www . invitrogen . com ) . After a brief vortex mixing , the mixture was incubated at 4 °C for 15 min and then centrifuged at low speed to obtain phase separation . The aqueous phase was transferred to a fresh tube , to which an equal volume of isopropanol was added in the presence of 3 μl glycogen . The mixture was kept at −20 °C for 30 min , followed by centrifugation . The RNA pellet thus obtained was washed with 70% chilled ethanol , repelleted , and air dried . The final RNA pellet was dissolved in 20 μl DEPC-treated distilled water and was used in at least two RT reactions . RT was performed using the Superscript II RT kit obtained from Life Technologies ( http://www . invitrogen . com ) using oligo ( dT ) 12–18 as the primer . The RT products were dissolved in 20 μl sterile distilled water and 1 μl was used for each PCR . For 5′-RACE ) , first strand synthesis using the lock-docking oligo ( dT ) primer and 35 cycles of PCR amplification were performed using the SMART RACE cDNA Amplification Kit purchased from Clontech ( http://www . clontech . com ) . The hydra-specific primers used in this study are E3_110R ( 5′-TGGATTTTACGCCGCTCCT ) . 5′ RACE–derived PCR products were subcloned into the pGEM-T vector ( Promega , http://www . promega . com ) using a standard cloning procedure , and 61 clones were picked randomly for DNA sequencing . A total of 50 testes of each line were collected from 3-d-old males . Testes were homogenized in Trizol ( Invitrogen ) , and total RNA was phenol-chloroform extracted , ethanol precipitated , and cleaned up with RNeasy mini kit ( Qiagen , http://www . qiagen . com ) and RNase free DNase Set ( Qiagen ) . About 200 ng of total RNA was used for synthesizing cDNA with TaqMan Reverse Transcription Reagents ( Applied Biosystems ) using oligo-dT as the primer , following the manufacturer's instruction . Quantitative real-time PCR ( qPCR ) was performed using the Lightcycler-FastStart DNA Master SYBR Green I kit ( Roche ) , according to the manufacturer's instructions . Primers used were E2_222F ( 5′-ATGTGGCAAAGGTCCAGAAT ) and E3_110R ( 5′-TGGATTTTACGCCGCTCCT ) , which are complementary to exon 2 and 3 of hydra , respectively . Note that this primer pair cannot distinguish transcripts derived from the two duplicates of hydra in D . simulans . PCR quality and specificity was verified by melting curve dissociation analysis and gel electrophoresis of the amplified products . Relative transcript abundance of hydra was calculated using the second derivative maximum values from the linear regression of cycle number versus log concentration of the amplified gene . Amplification of the control gene GADPH was used for normalization . Sequences of the primers used are GADPH_F: 5′-AAGGGAATCCTGGGCTACAC and GADPH_R: 5′-CGGTTGGAGTAACCGAACTC . Digoxigenin-labeled antisense and sense control riboprobes were generated from a cDNA containing the sequence of exon 2 and 3 of hydra cloned into a pBluescript vector . Probes were synthesized using DIG RNA labeling mix ( Roche ) , followed by treatment with 2 U DNAse ( Promega ) and incubation in carbonate buffer for 40 min at 65 °C as described in [61] . Probes were then ethanol precipitated , air dried , and resuspended in 200 μl hybridization buffer . Probes were used at a dilution of 1:50 during hybridization . Testes were dissected in 1× PBS and kept on ice until fixation . Two fixations were done with 4% formaldehyde in 1× PBS for 20 min on ice . After washing twice in PBS/0 . 1% Triton-X , protease treatment and all following steps were done as described in [61] . Hybridizations were done at 60 °C overnight . Orthologs of hydra were from the following prepublication genome sequences: D . simulans and D . yakuba sequence from the Washington University Genome Sequencing Center ( http://genome . wustl . edu ) ; D . sechellia sequence from the Broad Institute ( http://www . broad . mit . edu ) ; and D . erecta sequence from Agencourt Bioscience ( http://www . agencourt . com ) . Sequences of hydra from these species were annotated by maximizing homology to the D . melanogaster sequence , and are shown in Texts S1–S6 . Sequences were aligned using ClustalW [62] , with some manual adjustment to keep gaps in-frame in coding regions . Phylogenetic analyses were performed using Mega 3 . 1 [63] . Sequence polymorphism and divergence analysis were done using DnaSP 4 . 10 [64] . The effective number of synonymous and nonsynonymous sites , Ka and Ks values , and pairwise divergence for synonymous sites based on Kimura's two-parameter model were estimated . | Similar groups of animals have similar numbers of genes , but not all of these genes are the same . While some genes are highly conserved and can be easily and uniquely identified in species ranging from yeast to plants to humans , other genes are sometimes found in only a small number or even in a single species . Such newly evolved genes may help produce traits that make species unique . We describe here a newly evolved gene called hydra that occurs only in a small subgroup of Drosophila species . hydra is expressed in the testes , suggesting that it may have a function in male fertility . hydra has evolved significantly in its structure and protein-coding sequence among species . The authors named the gene hydra after the nine-headed monster slain by Hercules because in one species , Drosophila melanogaster , hydra has nine potential alternative first exons . Perhaps because of this or other structural changes , the level of RNA made by hydra differs significantly between one pair of species . This analysis reveals that newly created genes may evolve rapidly in sequence , structure , and expression level . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"arthropods",
"eukaryotes",
"evolutionary",
"biology",
"drosophila",
"animals",
"insects"
]
| 2007 | Evolution of hydra, a Recently Evolved Testis-Expressed Gene with Nine Alternative First Exons in Drosophila melanogaster |
The concept of robustness of regulatory networks has received much attention in the last decade . One measure of robustness has been associated with the volume of the feasible region , namely , the region in the parameter space in which the system is functional . In this paper , we show that , in addition to volume , the geometry of this region has important consequences for the robustness and the fragility of a network . We develop an approximation within which we could algebraically specify the feasible region . We analyze the segment polarity gene network to illustrate our approach . The study of random walks in the parameter space and how they exit the feasible region provide us with a rich perspective on the different modes of failure of this network model . In particular , we found that , between two alternative ways of activating Wingless , one is more robust than the other . Our method provides a more complete measure of robustness to parameter variation . As a general modeling strategy , our approach is an interesting alternative to Boolean representation of biochemical networks .
Robustness , in the context of biological networks , broadly indicates that the system remains viable under different perturbations . Defining robustness in a precise form is a challenging task , given that robustness to different kinds of perturbations , e . g . , environmental variation , intrinsic fluctuations in chemical networks or changes due to mutations , might involve different features of an existing network [1] , [2] . In this paper , we are concerned with the robustness of functionality to changes in the kinetic parameters for a given network architecture . In an influential study of the Drosophila segment polarity network , robustness has been associated to the fractional volume of the region in parameter space associated with the wild type gene expression pattern [3] . In this paper we will see that the geometry of the space of feasible parameters contains additional information on essential aspects of robustness and fragility of the network . In the context of fitting biochemical kinetics models to time series data , investigators have looked at effects of small parametric perturbations on the quality of the fit . Sensitivity analysis [4] , [5] , namely considering the effect of changing parameters , one at a time , is a common practice by now . Brown and Sethna have looked at correlated changes of parameters and study how moving in different directions in parameter space affects the predictions [6] . Based on the eigenvalues and the eigenvectors of the Hessian of the cost function at the minimum , these authors and their collaborators find that , for many known biochemical networks , only a few directions in the parameter space have stiff constraints whereas the rest of the directions are “sloppy” [7] , [8] . In this work , we will consider the segment polarity network as an example and will explicitly characterize the region in parameter space where the network could be functional . The anisotropy in the shape of this feasible region will become apparent from our analysis . We should clarify that the robustness of a model to parameter variation , as measured by goodness of fit to data , is distinct from the robustness of the system functionality with respect to parameter variation from mutations . However , at a mathematical level , these two problems just give rise to different ways of scoring parameter choices for a model , and there is much that is parallel in the consideration of the shape of the regions that score well in each of these problems . The segment polarity network is part of a cascade of gene families responsible for generating the segmentation of the fruit fly embryo . Genes involved in initiating this pattern are transiently expressed , and interactions among the segment polarity genes should maintain and fine-tune this pattern as the embryo grows through cell division . Much of the information about this network comes from genetic analysis and are therefore of qualitative nature . In particular , we do not know many of the parameters necessary to describe this dynamical system . This is a common situation faced in modeling most biochemical networks . In their work on modeling the segment polarity network , von Dassow et al . [3] encountered the same problem . Their approach was to solve an ODE model of the network for random choices of parameters and then score the resultant expression patterns based on compatibility with the experimentally observed wild type pattern . If this score is found to be above a certain threshold , the given parameter combination is said to belong to the feasible region of the parameter space . Robustness of a particular architecture is then ascertained by the fractional volume of the feasible region , estimated from their simulation . Ingolia [9] looked at a set of criteria for bistability in particular submodules of the network and studied the extent to which these criteria describe this feasible region . In general , providing an approximate description of the structure of feasible region , even for a medium size biochemical network , remains an important challenge . One could also get some insight into the functioning of the network by constructing a model where each gene or gene product is mostly ON or OFF . For example , in the context of this particular network , Boolean models have been employed to study dependence upon initial state or the effect of deletion of particular components [10] . Unfortunately , addressing questions related to parameter dependence is not possible within the conventional Boolean framework . Therefore , we develop a new approximation , within which the treatment of our model shares the simplicity of Boolean analysis without sacrificing the possibility of exploring parameter dependence issues . This approximation enables us to explicitly characterize the feasible region in the parameter space of the model . If a point in the feasible region of parameters represents a functional biological system , then a mutation causes the system to jump to a new point . If this new point also belongs to the feasible region , the system is robust with respect to that mutation . Otherwise the mutation is deleterious . If the jump in the parameter space , caused by a mutation , is relatively large then the result of successive mutations is to quickly probe different regions of the parameter space . In this case , robustness essentially depends on the volume of the feasible region . On the other hand , if the jumps in the parameter space are relatively small , evolution of parameters due to successive mutations can be represented by a random walk in the parameter space . The idea of representing evolution as a continuous random process has already been used in the adaptive landscape approach [11] . In this case , the random walk exiting the feasible region in the parameter space corresponds to a deleterious mutation . The exit time distribution is very sensitive to the shape of the feasible region . Robustness to mutation is , now , related to the features of this distribution ( e . g . half-life , asymptotic decay rate , etc . ) [12] and therefore depends upon the shape and not just the volume of the feasible region . If we want to choose a single measure for robustness , the inverse of the asymptotic decay rate is a good candidate [12] . This measure is sensitive to the geometry ( both volume and shape ) of the feasible region . For example , even if the total volume of the feasible region is relatively large , existence of “narrow” directions will greatly affect the decay rate; or if the feasible region is constituted of several disconnected part , the decay rate will again be affected . In addition , it is independent of the initial condition . Also , in the theoretical case , where every mutation leads to a new , uncorrelated point in the parameter space , the inverse of the asymptotic decay rate is a simple function of the fractional volume of the feasible region . In our study , we will estimate half-life , a different but closely related measure of robustness . In case a single exponential in time gave the probability of remaining in the feasible region , these two measures of robustness would be proportional to each other . In practice , half-life depends partially on short time properties of the system and is initial condition dependent . On the other hand , measuring the asymptotic decay rate accurately for high dimensional stochastic system needs more computational effort than estimating half-life . Before we go on , let us explain what measure of distance we use when we talk about narrow or wide directions in the parameter space . If we consider the continuous random walk approximation to parameter evolution , then the short-time properties of diffusion set up a metric for the space of parameters . The metric tensor of this space is the inverse of the covariance matrix of infinitesimal displacements divided by the infinitesimal time interval . Once we have this metric , we could decide whether , from a generic point , the distance to reach the boundary in certain direction is relatively small or large . This definition of distance is closely tied to the time the system typically takes to diffuse over a certain separation . Once we characterize the feasible region in parameter space , we explore how the system fails as a result of such a random walk . For two alternative network models , we compare the exit time distributions . More importantly , we can see how , in a particular model , the feasible region is narrower in certain directions than in others . These narrow directions are associated with the predominant modes of failure of the system in the random walk process . We end by speculating how these methods could be extended to generic biochemical network models .
We propose an approach which retains the information about kinetic parameters , but , at the same time , keeps part of the simplicity of a Boolean model by having most genes either in the fully ON or the fully OFF state . We approach the problem by first solving the algebraic equations coming from the steady state conditions and writing the steady state solutions in terms of the parameters . Since one of the steady state solutions should match the wild type pattern , one can look for the constraints on parameters that yield this pattern . This procedure provides a family of conditions defining regions of feasible parameters for the wild type steady state . Although all of the parameters in the feasible region can maintain the desired pattern , one aspect we ignore is whether the system can reach the wild type pattern from particular initial conditions . In our analysis , we used the fact that many of the differential equations in the model involve terms of the Hill form:where X is the concentration of some species , κ is the dissociation constant and ν is the Hill coefficient . The steepness of the Hill function is characterized by the Hill coefficient ν . As X increases from zero and passes the threshold κ , the function φ has a transition from OFF to ON state . For moderately large Hill coefficient , this transition becomes quite steep , and φ is practically insensitive to the actual value of ν . In the model presented in reference [3] , ν is indeed found to be often quite large , between 5 to 10 [16] . Any such term may thus be replaced by a step function with two levels: Using this , the steady state gene expression is characterized by the following equations: ( 2 ) ( 3 ) ( 4 ) ( 5 ) ( 6 ) ( 7 ) ( 8 ) ( 9 ) ( 10 ) ( 11 ) ( 12 ) ( 13 ) ( 14 ) Here we use the same notation as in [3] . Xi , i = 1 , 2 , 3 , 4 , denotes the total concentration of the protein species X in cell i , with lower case xi referring to the concentration of the corresponding mRNA molecules . In addition , for three of the components involved in cell-to-cell communication , namely , external Wingless ( EWG ) , Patched ( PTC ) and HH , the concentration on each of the four cell faces could be different . For any of these components , the concentration in cell i at face j is denoted by Xi , j , i = 1 , 2 , 3 , 4 , j = 1 , 2 , 3 , 4 . For these three species , the sum of the concentration over all four faces of cell i is denoted by Xi , T . The adjacent cell face to face j of cell i is shown by Xi , lr . The opposite cell face to face j of cell i is shown by Xn , j+2 . Also , κXY denotes the dissociation constant for species Y corresponding to the binding that regulates the species X . The range for κXY is chosen to be between zero and one . The equations are in normalized form , meaning that the concentrations of the components have been scaled so that the maximal steady state level is one . The structure of this particular network allows one to draw several interesting conclusions immediately . For example , the steady state levels for HH and PTC are completely determined once one specifies the mRNA levels of en , hh and ptc ( this does not depend on the high Hill coefficient approximation ) . Assuming that en and hh are expressed only in the cell 3 , which is the case in the wild type pattern , it can be shown that ptc2 = ptc4 , and PTC2 , T = PTC4 , T . The reason is as follows . If ptc2>ptc4 , cell 2 ends up producing more PTC , part of which get bound to HH diffusing over from cell 3 . However , the symmetric nature of the diffusion leads to more PTC in cell 2 than in cell 4: PTC2 , T>PTC4 , T . Higher level of PTC results in higher rate of proteolysis of CI . Therefore , in the steady state , CIi is a decreasing function of PTCi and CNi is an increasing function of PTCi . This means that ( given en is not present in cells 2 and 4 , and therefore has no repressive effect on ci production ) ( 15 ) However CI is an activator and CN is a repressor of ptc , which together with Equation 15 implies ptc2<ptc4 , which contradicts the assumption we started with . Of course , we could have started with ptc2<ptc4 and again end up with contradiction ( for the formal proof , see , Chaves , Sengupta and Sontag , Geometry and topology of parameter space: investigating measures of robustness in regulatory networks , to appear in Journal Mathematical Biology ) . This argument shows that the concentration levels of ptc , PTC , CI , CN and PH is exactly the same in cells 2 and 4: ( 16 ) This observation will turn out to be quite significant for the following reason . The wg level in a cell is controlled by the CI-CN pathway and the postulated feedback [3] from internal WG ( IWG ) . Since cells 2 and 4 do not differ when it comes to CI and CN levels , any difference in the WG expression has to be attributed to the wg autoregulation . In order to analyze the wg sector , we note that , in this model , the EWG and IWG levels are uniquely determined by a set of linear equations once the wg levels are given . Solving these linear equations , using the periodic boundary conditions and the fact that wg is produced only in cell 2 , we find that: ( 17 ) This result is not surprising because the distribution of WG is determined by a symmetric diffusion process from the source in cell 2 , the only wg producing cell in each parasegment . Therefore , we expect cells 1 and 3 to have identical amounts of WG signaling . It turns out that EWG at the source , cell 2 , is higher than that of the flanking cells ( the formal proof is presented in the supplementary material ) . These observations have important consequences for the regulation of en , as explained below . Since en is expressed in cell 3 , we have: ( 18 ) This , together with Equation 17 , implies: ( 19 ) Had the en production been solely controlled by EWG , the model would have implied that if EWG3 is high enough to activate en in cell 3 , en will be also activated in cells 1 and 2 . This is why , in reference [3] , adding repression of en by CN was necessary to achieve the wild type expression pattern . The two new links introduced in reference [3] ( green lines in Figure 1B ) give rise to two positive feedback loops . The wg autoactivation gives rise to bistability , allowing cells 2 and 4 to have distinct levels of wg expression . The other loop ( En __| ci→CI→CN __| en→EN ) , generated by adding repression of en by CN , is required to prevent en from being expressed in cells 1 and 2 . This also requires CN to be expressed in those cells . The bistability of the EN-CI-CN system allows cells 1 and 3 to have different en level even when the external Wg signal is the same for both of them . We should note that autoactivation as a way for maintaining the WG expression is problematic in the following sense . In the model described above , wg is always activated via autoactivation and the preexisted CI-CN pathway never contributes to the pattern . This is in contrast with the experimental data , which suggests that HH signaling from the neighboring cell plays a crucial role in maintaining the wg expression . The fact that model [3] does not depend upon HH signaling for maintaining the expression of wg manifests itself when cell division is considered . In this model , both daughters of a cell in the wg-expressing stripe are able to retain the wg ON state through autoactivation . This causes the stripe to grow wider and wider over cell divisions . However , in wild type fly , the wg-expressing stripe should remain one cell wide . The daughter cell , which is further from the en-expressing stripe , and therefore not exposed to HH signaling , loses wg expression . This means that one stripe of WG is left after each division . Ingolia [9] has also noticed that in this model , IWG level must always be above KWGwg ( the autoactivation threshold ) in the cell that expresses wg . When we removed the CI-CN cycle for activation of wg from the simulation performed in reference [3] , the fraction of “good solutions” increased by a factor of 3 . This suggests that most of the time the CI-CN pathway is either not contributing to WG expression or it leads to misexpression of WG in cell 4 . The model is too dependent on the bistability of the two sub-networks with positive feedback for maintaining four cell expression patterns . One could avoid this problem by making some of the four cells special , either by inclusion of other genes in the network or by explicitly breaking the symmetry via introducing different gene expression rates from cell to cell for some of the genes already in the model . The major candidate for inclusion in the model is the Sloppy-paired protein ( SLP ) as has already been suggested by others [9] , [10] , [17] . SLP is only present in cells 1 and 2: . It is a necessary ( but not sufficient ) factor for activation of wg and it also represses en . In the presence of SLP , the reason en is not expressed in cell 1 despite WG signaling is that it is being repressed by SLP . Also , despite HH signaling , wg is not produced in cell 4 because SLP is not present there . With SLP added , the two new interactions introduced in [3] are not necessary anymore , and also WG expression will depend on the CI-CN pathway . In this paper , we will analyze the effect of including SLP . We keep SLP as an external factor meaning that the expression pattern of SLP is given . However , it can easily be incorporated into the network . If WG activates SLP , a positive feedback loop is formed which allows for bistability: both WG and SLP can be ON or both can be OFF . On the other hand , if EN represses SLP , another positive feedback loop is formed which again allows for bistability: SLP can be ON and en OFF or vice versa . We have also explored a model with explicitly different rates of production of ptc and ci from cell to cell which will be presented in a separate publication ( Chaves , Sengupta and Sontag , Geometry and topology of parameter space: investigating measures of robustness in regulatory networks , to appear in Journal Mathematical Biology ) . The work also presents a study complimentary to that presented in this paper . It provides an explicit geometric description of the feasible region by partitioning the region into components defined by algebraic inequalities , in other words , by constructing a cylindrical algebraic decomposition . Here , we consider two particular cases: We can explicitly write down the conditions characterizing the feasible region for these two models . The results are presented in Tables 1 and 2 ( see Materials and Methods for the derivation of these conditions ) . We could easily estimate the associated volume of feasible region by randomly choosing points in the parameter space and check whether they satisfy the appropriate conditions . As we discussed in the introduction , the fate of random walks , especially where they exit the feasible region , teaches us a lot about relative vulnerability of different constraints . We explore the feasible region by following random walks starting from random points . Whenever one of the random trajectories hits a boundary and exits the feasible region , we terminate the walk and keep track of the inequality that was violated . This process can be viewed as a simulation of parameter evolution due to mutations in a fitness landscape that looks like a plateau . The points in the feasible region have a constant high fitness , and the rest of the points have zero fitness . The result of the simulation is presented in Figure 3 . For the two models discussed , the graphs in Figure 3A show the probability of survival as a function of time . This is the probability that the random walk has not exited the feasible region in the first t steps . From the graph , we can easily measure T1/2 , defined as the time for which there is a 50% chance that the system has already suffered a deleterious mutation . As we discussed in the introduction , this number is a possible indicator of robustness . Figure 3B shows the histogram of violated conditions . The number below each bin indicates the corresponding condition in Tables 1 and 2 . The lead cause of failure in the von Dassow et al . model is the constraint on κWGwg whereas in the SLP model it is the constraints on κEWGen . Higher vulnerability of the SLP model with respect to the constraint on κEWGen can be understood by comparing condition 3 in Table 1 and the corresponding condition in Table 2 . In the SLP model there is a lower bound on κEWGen coming from the fact that κEWGen should be greater than EWG4 to prevent activation of en in cell 4 . However in the von Dassow et al . model , en is being repressed by CN and therefore there is no lower limit on κEWGen . One might raise the question of whether including repression of en by CN in the SLP model changes the constraints on κEWGen . In high Hill coefficient limit , adding this interaction does not change the conditions in Table 2 . To see this , notice that as was mentioned before , requiring CI and CN levels to be different in cells 1 and 2 forces us to have CN2 = CN4 = 0 . In cell 4 , CN is not expressed , and in cells 1 and 2 , en is already being repressed by SLP . Therefore , adding the possibility of en repression by CN does not change any of the constraints . If we consider the case where Hill coefficients in the CI-CN-PTC sector are small , the transition from high to low in concentration value for ptc-nullcline and CN-nullcline would not be sharp . Instead , the transition would happen over a wide range . This means that we would get a non zero value for CN4 . In that case , adding repression of en by CN can indeed help in maintaining the wild type pattern , thereby increasing the robustness of the model . The parameters κCIwg , κCNwg and κWGwg are related to alternative routes controlling wg expression . The first two parameters play an important role in deciding WG expression in the SLP model , while this role is played by κWGwg in the von Dassow et al . model . Comparison of the frequency of failure for conditions 2 and 5 in the histogram in Figure 3B suggests that controlling wg via the CI-CN pathway in the presence of SLP is the more robust way of achieving the target gene expression pattern for wg . What about adding the WG autoactivation to the SLP model ? If one just cares about producing the right four-cell pattern for en , hh and wg , then this addition could give rise to more solutions . However , as we discussed before , not having wg production to be sensitive to HH signaling from the neighboring cell is problematic and gives rise to wide stripes of wg expression under cell division . If we constrain the model so that wg is sensitive to HH signaling via CI-CN pathway , we find that adding wg autoactivation to a functional solution in the SLP model often leads to misexpression of wg in cell 1 or cell 3 , thereby shrinking the feasible region in parameter space .
Our results imply that the lack of robustness is not only dependent upon the size of the feasible region , but also upon the existence of critical directions along which this region is globally very narrow . We found relatively few constraints on the parameters given that we have specified the gene expression patterns for en , hh and wg in each of the four cells . Much has been said about the relation between the topology of the network and robustness . In practice , we found that it is not only the structure of the network but also the nature of the wild type expression pattern which plays an important role in the ultimate simplicity of the constraints that dictate robustness . For example , the fact that only one cell is expressing en and hh and that wg had no direct effect on the CI-CN-PTC sector allowed us to draw several conclusions about certain variables being the same in cell 2 and cell 4 . If one only pays attention to the network structure , wg indeed has an effect on the CI-CN-PTC sector via its effect on en . However , specifying the en expression pattern hides the influence of wg and helps us disentangle the constraints . The role of wg shows up only when one insists upon self-consistency , namely , the wg expression pattern is going to lead to the target en expression pattern . Simplicity of the final constraints is not a result of some obvious modularity in the network itself but some combination of the network structure as well as of the sparseness of the expression pattern . We cannot be sure that this is a general feature of robust genetic networks . A broader study , which takes into account the role of the wild type pattern on the robustness of a network , would be a welcome deviation from discussions centered purely on network architecture . We noted that capturing the CI-CN-PTC negative feedback in the Boolean model is difficult . For example , in the Boolean model constructed by Albert and Othmer [10] , they are forced into a situation where ptc mRNA is OFF but PTC protein is ON . This is achieved because of an exception made in PTC production rule , namely , PTC can continue to be in the ON state even if there is no ptc . Of course , this implausible rule results in a distribution of ptc and ci products which mimics the wild type pattern . For example cell 1 has less ptc but more CN compared to cell 2 . In our model , we partially capture the effect of the feedback . We can indeed get the ptc levels to vary between cell 1 and cell 2 . Unfortunately , we saw that in the high Hill coefficient model , producing different CN levels requires fine-tuning of the parameters . Therefore , we understand why von Dassow et al . find that setting the Hill coefficients in the CI-CN-PTC sector to be small enhances their chance of finding good solutions [16] . The present approach shows that , in addition to volume , the topology and geometry of the feasible region have important consequences for the robustness of a system . Of special interest is the structure of the boundary in the parameter space that separates between functional and non-functional systems . In the models studied here , it was possible to describe this boundary explicitly as a collection of constraints . For a generic biochemical network model with a scoring function it may not be feasible to explicitly write down the boundary surface corresponding to the threshold of functionality . However , one could generate a sampling of the boundary surface by following random walks in the parameter space until it hits the boundary of the functional region ( decided by a threshold score ) . Instead of what we did in this study , we could slightly alter our strategy and let the walk be reflected off the boundary . In that process the same walk would hit many neighboring points on the boundary surface . If one generates a large enough sample of boundary points , one could use methods like manifold learning [18] , [19] to approximately reconstruct the boundary . Contrast this method to boundary reconstruction from uncorrelated random sampling . One could generate many points some of which are inside the region and many others that are outside . Indeed , many machine learning techniques for classification involve learning decision boundaries from such data . However , when the good region has a very small fractional volume and many of the randomly sampled points outside this region are far from the decision boundary , most of the sampled points have very little impact on boundary reconstruction . The uncorrelated nature of the sampling is useful for getting a good estimate for the fractional volume , but makes the process of mapping the geometry inefficient . It would be better to take advantage of one good solution to generate other good ones for the purpose of exploring local geometry . Whether these approaches work for analyzing biologically motivated network models remains to be seen . For an arbitrary random network , with an equally arbitrary random choice of gene expression pattern , the feasible region could have a very complex structure and the methods outlined would not be particularly useful for characterizing it . The hope is that , for biologically relevant networks with wild type gene expression patterns , the feasible region may be quite simple , geometrically , and could be approximately described by the approaches suggested above . To summarize , our analysis of the segment polarity network provides us with insights regarding the constraints that are crucial for functioning of the system . We showed how the system is particularly vulnerable to parametric perturbations in certain directions in the parameter space . We believe that the ideas developed here could be applied to other regulatory networks , to explore how the shape of feasible region in the parameter space contributes to its robustness . Hill terms appear often in models of biochemical networks . A simpler model , obtained by replacing these terms with step function , could be useful , because such a model enjoys some of the simplicity of the Boolean networks , while retaining many of the parameters of the original model .
Here we analyze two particular cases: We first focus on case I . Equations 2–14 characterize this network . The wild type expression pattern for wg , en and hh is given in Equation 1 . Since en is only expressed in cell 3 , ci and ptc are expressed in all cells except cell 3: ( 20 ) This is because in the absence of EN , ci is basally expressed which also leads to production of ptc . We will allow Ti to take values between zero and one . The reason for the special , non-Boolean , treatment of ptc has to do with capturing the effect of the negative feedback loop in the CI-CN-PTC sector properly . This negative feedback loop leads to lower ptc level in cell 1 than in cells 2 and 4 , as we shall see . The ptc level in cells 2 and 4 turn out to be comparable ( T2 = T4 ) . This is also the experimentally observed expression pattern of ptc [20] . How could we ever get such an intermediate values in our approach ? First , from Equations 13 and 14 , in the cells where en is not expressed and therefore ci is not repressed , namely in cells 1 , 2 and 4 , we have CI+CN = 1⇒CI = 1−CN ( this does not depend on the high Hill coefficient approximation ) . Since ptc is regulated by CI-CN , we could draw one nullcline expressing ptc concentration as a function of CN . This curve is represented by the green graph in Figure 4 . We will call it the ptc-nullcline . Here it is assumed that the negative feedback on ptc coming from repression by CN is active . This means that CN and ptc are not expressed maximally . For ptc to be expressed , the activation by CI requires 1−CN>κCIptc⇒CN<1−κCIptc . In addition , we need CN to be smaller than κCNptc to avoid repression of ptc by CN . Thus , for values of CN smaller than the threshold of min ( 1−κCIptc , κCNptc ) , ptc is fully expressed . As CN passes this point , the value of ptc will drop sharply . In the high Hill coefficient limit , ptc will abruptly fall to zero . On the other hand , CN production itself is dependent upon PTC protein . PTC is a monotonically increasing function of ptc and a decreasing function of HH signaling . Therefore , for a fixed value of HH level , we can also look at the concentration of CN as a function of ptc . This provides us with the CN-nullcline which depends upon the HH signaling strength . If we think of CN as a function of ptc level , the transition in CN from low level to its highest value happens at a particular ptc threshold , where the PTC level is just enough to start producing CN . If the cell is exposed to more HH signaling , sequestering away a larger fraction of total Patched protein , one needs more ptc to reach this threshold . The blue and the red graphs in Figure 4 show the CN-nullclines for relatively higher and lower values of HH signaling levels , respectively . Because cell 1 receives less external HH signaling than cells 2 and 4 , generally the red curve could be associated to cell 1 and the blue one to cells 2 and 4 . The intersection points 1 and 2 determine CI , CN and ptc level in cell 1 and 2/4 , respectively . As we see , ptc value could indeed be higher in cell 2 than in cell 1 . However , CN concentration seems to be comparable in those cells . This is an artifact of our model where Hill coefficients are very large , which causes the transition from high to low in concentration value to happen in a very narrow range . The only way to have CN2 to be non-zero but different from CN1 is to be in the situation where the CN-nullcline for cell 2 is like the dashed blue line in Figure 4 . In this case , the ptc threshold for CN production in cell 2 is fine-tuned to be very close to maximal ptc level . In a model with small Hill coefficients in the CI-CN-PTC sector , we would get CN1>CN2 and ptc1<ptc2 without such fine-tuning . We will come back to this point later . We should point out that , in this study , we lay down the conditions only on the expression levels of key components en , wg and hh as specified in Equation 1 . The reason , other than the simplicity of analysis , is that we believe the requirement of proper segment formation lays much stronger constraints on these key components compared to the rest . It is not clear to us that the CI-CN-PTC negative feedback has an extremely important role in segment formation stage of development . The study of von Dassow et al . [3] also uses a scoring function which rewards wild type levels only for these key components . Having specified the requirements of functionality , let us now analyze what conditions are laid on the parameters of the model . Table 1 shows the set of inequalities characterizing the feasible region in the parameter space . Here we present the arguments leading to these conditions . The presence of EN in cell 3 requires the WG signaling for this cell to be above the activation threshold for en . This requirement is condition 3 in Table 1 ( recall that κXY can take value only between zero and one ) . Also , in this cell , EN will shut off the expression of ci ( Equation 12 ) which is necessary for the production of CI , ptc , PTC and PH . Therefore , none of those components are expressed in cell 3 . In cells 2 and 4 , the expression level of these components has been shown to be the same ( Equation 16 ) . Therefore , we only need to focus on the expression of these components in cells 1 and 2 . Let be the PTC level corresponding to the maximal ptc mRNA ( ptc = 1 ) in cell i . If the threshold to produce CN is above , then cell i would not produce CN . As we pointed out before , the presence of CN in cells 1 and 2 is essential to repress en in those cells . These facts together necessitate condition 1 in Table 1 . What would the CN level in cells 1 and 2 be when condition 1 is satisfied ? As one sees from Figure 5A , there are two possibilities depending upon whether min ( 1−κCIptc , κCNptc ) is smaller or larger than . The case corresponding to ptc-nullcline in solid green has been discussed before . This is the case where ptc levels are affected by the negative feedback , and CN level is equal to min ( 1−κCIptc , κCNptc ) , which is less than its maximal possible value of . When the ptc-nullcline is like the dashed green line in Figure 5 , CN levels in both cell 1 and cell 2 is equal to the maximal amount of , which is lower than min ( 1−κCIptc , κCNptc ) . In this case , the negative feedback is not active and ptc is maximally expressed ( ptc = 1 ) . We conclude that CN level is given by , which we call ZC . We will now discuss the conditions to be satisfied by ZC for proper expression pattern of en and wg . The en repression in cells 1 and 2 gives rise to condition 4 in Table 1 . The fact that CI-CN pathway should not activate wg in cell 4 is guaranteed by condition 2 in Table 1 . Consequently , WG in cell 2 has no contribution from CI-CN pathway ( remember that cells 2 and 4 have the same CI and CN levels ) and is solely produced by the autoactivation term . The autoactivation should only operate in cell 2 and nowhere else . This is condition 4 in Table 1 . von Dassow and Odell analyzed randomly generated solutions for the segment polarity model in reference [3] and plotted the marginal distribution of parameters ( see Figure 6 of [16] ) . We can relate their results to the constraints presented in Table 1 . From condition 1 , we expect κPTCCI to have tendency for lower values . From condition 2 , we expect κCNwg to have tendency for lower values and κCIwg for higher values . Also , in order to have higher values for ZC , we expect κCIptc to have tendency for lower values and κCNptc for higher values . From condition 3 and 4 , we expect κEWGen and κCNen to have tendency for lower values . From condition 5 , we expect κWGwg to have tendency for intermediate values . These expectations agree qualitatively with the results presented in Figure 6 of [16] . From Figure 6 of reference [16] , we see that many of the parameters are uniformly distributed . One should note that a uniform distribution for a certain parameter could arise from two different scenarios . It could be the case that changing the parameter over a wide range of values does not influence the final outcome of the network . The other possibility is that the effect of changing the particular parameter could be compensated by changes in other parameters in such a way that for each value of the parameter , there is roughly equal number of solutions . Now , let us contrast these set of conditions to the one obtained for the SLP model . Table 2 shows the conditions defining the feasible region for this case . For this regulatory network ( Figure 2 ) , instead of Equations 2 and 5 , we have: ( 21 ) ( 22 ) The rest of equations are the same as before ( Equations 3 , 4 and 6–14 ) . Since SLP is present only in cells 1 and 2 , wg has the possibility to be expressed only in those two cells . The decisive factor is CN levels in cells 1 and 2 ( remember that , in these cells , CI = 1-CN ) . In the wild type pattern , wg is expressed only in cell 2 and this means that CN levels cannot be the same in cells 1 and 2 . The only way to have less CN in cell 2 compared to cell 1 is to have . The condition corresponds to the plateau in the CN-nullcline for cell 2 being higher or equal to the maximal ptc level ( blue graph in Figure 5B ) . When it is higher , CN2 is zero and when it is fine-tuned to be equal , CN2 is between 0 and 1 . If we had , given that , we would have CN1 = CN2 = 0 . This is inconsistent with our requirement that CN1 and CN2 be different . Therefore , we have . For our discussion , we will ignore the fine-tuned cases , leaving us with condition 1 in Table 2 . This mean CN2 = 0 and which we again call ZC . The condition 2 in Table 2 guarantees the absence of wg in cell 1 . The fact that external WG signaling has to be strong enough in cell 3 to activate en but has to be weak enough in cell 4 not to produce en is coded in the condition 3 of Table 2 . To get an estimate for the fractional volume of feasible region in the parameter space , we randomly chose 106 parameter combinations and checked if they satisfy the conditions given in Tables 1 and 2 for the corresponding model . We perform the random walk by first selecting a random point , P0 , from the set of admissible parameters and follow successive random perturbations . Each component of is selected from an independent Gaussian distribution with a standard deviation of 2*10−3 . We follow this random walk until it hits a boundary and exits the space . This happens when one of the inequalities , which characterize the feasible region , is violated . Whenever the random walk exits the region , we record the time as well as the condition that was violated and therefore caused the exit . The parameter ranges were similar to those used in [3] , except that we facilitated the transport processes for hh and PTC . We simulated the random walk for 30 , 000 runs .
AMS thanks Pankaj Mehta for discussions that lead to the formulation of the high Hill coefficient version of the segment polarity network model . We also thank Viji Nagaraj for carefully reading the final manuscript . | Developing models with a large number of parameters for describing the dynamics of a biochemical network is a common exercise today . The dependence of predictions of such a network model on the choice of parameters is important to understand for two reasons . For the purpose of fitting biological data and making predictions , we need to know which combinations of parameters are strongly constrained by observations and also which combinations seriously affect a particular prediction . In addition , we expect naturally evolved networks to be somewhat robust to parameter changes . If the functioning of the network requires fine-tuning in many parameters , then mutations causing changes in regulatory interactions could quickly make the network dysfunctional . For predictions involving gene products being ON or OFF , we found a method that facilitates the study parameter dependence . As an example , we analyzed several competing models of the segment polarity network in Drosophila . We explicitly describe the region in the parameter space where the wild-type expression pattern of key genes becomes feasible for each model . We also study how random walks in the parameter space exit from the feasible region of a network model , allowing us to compare the relative robustness of the alternative models . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Acknowledgments"
]
| [
"computational",
"biology/systems",
"biology"
]
| 2009 | Shape, Size, and Robustness: Feasible Regions in the Parameter Space of Biochemical Networks |
When chromosomal DNA is damaged , progression through the cell cycle is halted to provide the cells with time to repair the genetic material before it is distributed between the mother and daughter cells . In Saccharomyces cerevisiae , this cell cycle arrest occurs at the G2/M transition . However , it is also necessary to restrain exit from mitosis by maintaining Bfa1-Bub2 , the inhibitor of the Mitotic Exit Network ( MEN ) , in an active state . While the role of Bfa1 and Bub2 in the inhibition of mitotic exit when the spindle is not properly aligned and the spindle position checkpoint is activated has been extensively studied , the mechanism by which these proteins prevent MEN function after DNA damage is still unclear . Here , we propose that the inhibition of the MEN is specifically required when telomeres are damaged but it is not necessary to face all types of chromosomal DNA damage , which is in agreement with previous data in mammals suggesting the existence of a putative telomere-specific DNA damage response that inhibits mitotic exit . Furthermore , we demonstrate that the mechanism of MEN inhibition when telomeres are damaged relies on the Rad53-dependent inhibition of Bfa1 phosphorylation by the Polo-like kinase Cdc5 , establishing a new key role of this kinase in regulating cell cycle progression .
During mitosis , different surveillance mechanisms ensure that the replicated genomic material is protected from damage and correctly distributed between the daughter and the mother cells . In this way , chromosomal DNA damage triggers a stress response pathway known as the DNA damage checkpoint ( DDC ) [1] , which arrests the cell cycle to provide the cells with time to repair the genomic material before further progressing into mitosis . The cells must also ensure that all the chromosomes are attached to the spindle , a bipolar array of microtubules that allows for the segregation and distribution of the chromosomes between the daughter and mother cells . The proper attachment of all kinetochores to the spindle is monitored by the spindle assembly checkpoint ( SAC ) , which otherwise delays the onset of anaphase [2] . Finally , in cells that display some type of asymmetry during mitosis , as it is the case of the budding yeast Saccharomyces cerevisiae , it is also essential that the spindle be correctly positioned with respect to the division site . In this organism , this is ensured by the spindle position checkpoint ( SPOC ) , which restrains mitotic exit when the spindle is not properly aligned along the mother-bud axis and perpendicular to the bud neck [3] . A similar mechanism , the centrosome orientation checkpoint ( COC ) , has also been described in Drosophila male germline stem cells [3] . The COC delays the commitment to mitosis in case of centrosome misorientation . Despite the diverse signals that trigger the DDC , the SAC and the SPOC , as well as the different cell cycle stages where these surveillance mechanisms are triggered , the three checkpoints have been shown to inhibit mitotic exit in S . cerevisiae by activating the two-component GTPase-activating protein ( GAP ) Bfa1-Bub2 [4] . This GAP inhibits Tem1 , a GTPase that initiates signaling by the Mitotic Exit Network ( MEN ) [5] . Activation of the MEN drives a sustained release of the Cdc14 phosphatase out of the nucleolus in anaphase , where it is sequestered from G1 to metaphase by its inhibitor Cfi1/Net1 [6] , [7] . Once released into the nucleus and the cytoplasm , Cdc14 reverts the phosphorylation events triggered by the mitotic cyclin-CDK complexes , leading to their inactivation and the completion of mitosis [5] . During a normal cell cycle , the activity of Bfa1-Bub2 is regulated through phosphorylation . At the onset of anaphase , Bfa1 is phosphorylated by the Polo-like kinase Cdc5 , which inactivates the GAP and promotes MEN signaling [4] , [8] , [9] . Cdc5 also plays an essential role during mitotic exit by recruiting the MEN kinase Cdc15 to the spindle pole bodies ( SPBs , the yeast equivalent of the centrosomes ) [10] . When the SAC or the SPOC are triggered , however , Bfa1 is maintained in a hypo-phosphorylated and therefore active state , which restrains mitotic exit [4] . The protein kinase Kin4 plays a key role in the SPOC by promoting the inhibitory action of Bfa1 on MEN signaling . When the spindle is not properly positioned , Kin4 phosphorylates Bfa1 impeding its inactivation by Cdc5 [11] , [12] . In addition , Kin4 actively excludes Bfa1 from the SPBs after SPOC activation [13] . Since Tem1 localization to the SPBs depends on Bfa1 [14] and it is essential for MEN signaling [15] , this also contributes to the inactivation of mitotic exit under these circumstances . While the role of Bfa1 and Bub2 in the inhibition of mitotic exit after the activation of the SAC and the SPOC has been extensively studied , the mechanism by which these proteins prevent MEN function after the generation of DNA damage is still unclear . In S . cerevisiae , a central regulator of the DDC is the Mec1 kinase , the yeast homolog to mammalian ATR , which activates Chk1 and Rad53 , two kinases that form parallel branches of the Mec1-dependent DDC pathway [1] . After DNA damage , Chk1 inhibits the metaphase to anaphase transition by stabilizing securin ( Pds1 ) and therefore maintaining separase ( Esp1 ) inactive [16] . Rad53 also inhibits the metaphase to anaphase transition by regulating Pds1 stability , but it has been additionally shown to prevent mitotic exit by regulating Bfa1-Bub2 [4] , [17] . The mechanism by which Rad53 inhibits Bfa1-Bub2 is , at present , not clear . The kinase activity of Cdc5 was initially found to be high in DNA-damaged cells [18] , while Bfa1 was described to be phosphorylated but active in a Rad53-dependent manner after DDC activation [4] . Thus , and since phosphorylation of Bfa1 by Cdc5 inhibits its activity , it was formally proposed that , in contrast to the SAC and the SPOC , the inhibition of mitotic exit after DNA damage is independent of the phosphorylation of Bfa1 by Cdc5 [4] , and that , under these circumstances , Rad53 would be then controlling the activity of Bfa1 by a yet-unknown mechanism [4] . More recently , however , it has been demonstrated that Cdc5 is in fact partially inhibited by Rad53 after the generation of DNA damage [19] . This inhibition restrains the elongation of the spindle by preventing the inactivation of the anaphase-promoting complex cofactor Cdh1 , thus limiting the accumulation of the bimC kinesin family proteins Cin8 and Kip1 [19] . Therefore , and based on these new observations , it is necessary to reevaluate whether Bfa1 phosphorylation by Cdc5 needs to be actively prevented also after DDC activation in order to maintain the GAP in an active state and prevent MEN signaling . Here , we have analyzed the regulation of mitotic exit in response to DNA damage . We have determined the consequences of the lack of Bfa1 activity on the functionality of the DDC in response to a wide variety of chromosomal DNA damage . Based on our results , we propose that the inhibition of MEN signaling is essential when telomeres are damaged , but it is not a general requirement for the functionality of the DDC . Additionally , we have analyzed the mechanism by which the DDC impedes MEN signaling , and the role of Cdc5 in this process . We have demonstrated that the Rad53-dependent inhibition of Cdc5 after DNA damage is essential to maintain Bfa1 in a hypo-phosphorylated form that inactivates Tem1 and thus inhibits mitotic exit .
The role of Bfa1-Bub2 in the response to DNA damage was originally determined by using the cdc13-1 allele [4] , [20] . Cdc13 is an essential protein that protects the telomere from degradation and regulates telomerase activity [21]–[24] . Cells carrying the cdc13-1 mutation cannot “cap” the telomeres and accumulate single-stranded DNA at the restrictive temperature , which triggers a DDC-dependent cell cycle arrest in G2/M [21] , [25] . Accordingly , cdc13-1 cells synchronized in G1 using pheromone accumulated as large budded cells with an undivided nucleus after their release into pheromone-free medium at 34°C , while DDC-deficient cdc13-1 rad53Δ sml1Δ cells could not hold this arrest ( Figure 1A , [26]; deletion of SML1 is necessary to bypass the lethality associated to the inhibition of the ribonucleotide reductase and the subsequent reduced dNTPs levels in rad53Δ cells [27] ) . Approximately 30% of cdc13-1 rad53Δ sml1Δ cells still remained arrested , but this percentage was reduced to only 15% when CHK1 was also deleted and the other branch of the DDC was thus additionally inactivated ( Figure S1A ) . The metaphase arrest observed for cdc13-1 cells at the restrictive temperature was also dependent on both Bfa1 and Bub2 ( Figures 1A and B , [20] ) , which indicates that inhibition of MEN by the two-component GAP must also be ensured after the telomeres are damaged . In order to analyze whether inactivation of the MEN is a general requirement for the functionality of the DDC , we tested the capacity of bfa1Δ cells to maintain a DDC-dependent cell cycle arrest in the presence of other types of chromosomal damage . Wild type , rad53Δ sml1Δ , and bfa1Δ cells were synchronized in G1 using pheromone . After pheromone washout , cells were released into medium containing zeocin , a chemical that generates DNA double strand breaks ( DSBs ) [28] . The DNA damage generated by zeocin led to an accumulation of large budded cells in the wild type strain ( Figures 1C and S1C ) , as previously observed for the cdc13-1 mutant at the restrictive temperature . The cell cycle arrest induced by zeocin treatment was dependent on the functionality of the DDC , since it could not be held in rad53Δ sml1Δ cells ( Figures 1C and S1C ) . However , and in contrast to what was observed for the cdc13-1 bfa1Δ mutant at 34°C , zeocin-treated bfa1Δ cells accumulated as large budded ( Figures 1C and S1C ) . The specific requirement for Bfa1 after telomere damage cannot be attributed to different levels of checkpoint activation , since it was still observed in cdc13-1 cells growing at 37°C and wild type cells treated with zeocin at the same temperature , for which the extent of DDC activation , as measured by the effect of RAD53 deletion on the percentage of large budded cells , was similar ( Figure S1D and E ) . Inhibition of MEN signaling was also not necessary to restrain cell cycle progression in response to the generation of a single unrepaired DSB induced by expression of the HO-endonuclease in cells in which an HO recognition site was introduced in chromosome II and that also carried a MATa allele that cannot be cleaved by HO ( MATa-inc ) . The induction of the DSB in these cells led to a Rad53-dependent cell cycle arrest ( Figures 1D and S1F ) . However , this DSB-induced arrest could still be maintained in bfa1Δ cells ( Figures 1D and S1F ) . Therefore , our results indicate that , while necessary to block the cell cycle in response to uncapped telomeres , inhibition of mitotic exit is not essential for the DDC to maintain a cell cycle arrest after generation of DSBs in the DNA . To further evaluate the role of Bfa1 in the response to DNA damage , we also assessed the survival of wild type , rad53Δ sml1Δ , and bfa1Δ cells after irradiation with gamma rays and UV , as well as their capacity to grow in the presence of different compounds that induce DNA damage ( camptothecin , methyl methanesulfonate ( MMS ) and hydroxyurea ( HU ) ) . While , as expected , rad53Δ sml1Δ cells showed poor survival after irradiation with both gamma rays and UV ( Figure 1E , [26] ) and reduced viability in the presence of DNA-damaging chemicals due to an impairment of the DDC ( Figures 1F and G , [29] ) , cells lacking Bfa1 behaved as wild type cells ( Figures 1E , F and G ) . Therefore , our results suggest that rather than a general role in the protection against DNA damage , Bfa1 plays a more specific role in the response of the cell to the damage induced by a failure in telomere capping . During a normal cell cycle , the activity of Bfa1-Bub2 is regulated through phosphorylation . At the onset of anaphase , Bfa1 is phosphorylated by Cdc5 , which inactivates the GAP and allows for MEN signaling [4] , [8] . When chromosomes are not correctly attached to the spindle and the SAC is activated , Bfa1 is maintained in a hypo-phosphorylated and therefore active state , which restrains mitotic exit [4] . The same happens when the spindle is not properly aligned and the spindle position checkpoint ( SPOC ) is activated [4] . In order to analyze the molecular mechanism by which damaged telomeres trigger a Bfa1-dependent inhibition of mitotic exit , we determined the phosphorylation status of a N-terminal 3HA-tagged version of Bfa1 in cells carrying the cdc13-1 allele . We also analyzed the phosphorylation of Bfa1 in cdc15-2 cells , which cannot exit mitosis at the restrictive temperature due to a block in MEN signaling downstream of Bfa1 [9] , [30] . The cells were synchronized in G1 at the permissive temperature using pheromone , and then released into pheromone-free medium at the restrictive temperature . As expected , the cdc15-2 mutant arrested in anaphase , while cdc13-1 cdc15-2 cells were blocked already in metaphase due to the activation of the DDC ( Figures 2A and B ) . This also indicates that the N-terminal 3HA-tag of Bfa1 does not affect its functionality . Phosphorylation of Bfa1 was analyzed in these cells after optimization of the experimental conditions to detect as many modified forms of the protein as possible . Bfa1 was heavily modified in the cdc15-2 arrest ( Figures 2C and D ) . These modifications were caused by phosphorylation of the protein , since they were no longer detected after phosphatase treatment ( Figure S2A ) . Interestingly , activation of the DDC in cdc13-1 cells at the restrictive temperature prevented hyper-phosphorylation of Bfa1 ( Figures 2C and D ) . Although Bfa1 was still phosphorylated to some extent , the two highest phosphorylated forms of the protein disappeared in the cdc13-1 arrest when compared to the cdc15-2 mutant . We next analyzed the role of Rad53 in the regulation of Bfa1 activity by determining the phosphorylation status of Bfa1 in cdc13-1 rad53-21 cells . The checkpoint-deficient rad53-21 allele does not need the additional deletion of SML1 [31] , which simplifies strain construction . As observed for cdc13-1 rad53Δ sml1Δ cells , the cdc13-1 rad53-21 mutant could not hold the metaphase arrest at the restrictive temperature ( Figure 2B ) . Accordingly , rad53-21 cells also displayed low survival after irradiation with gamma rays and UV ( Figure S2D ) and reduced viability in media containing camptothecin , zeocin , MMS or HU ( Figures S2E and F ) . In order to block anaphase progression , the cdc13-1 rad53-21 mutant also carried the cdc15-2 allele . Analysis of the spindle and nuclear morphologies indicated that the cdc13-1 rad53-21 cdc15-2 cells indeed elongated the spindle and arrested at this cell cycle stage ( Figures 2A and B ) . Bfa1 was found to be hyper-phosphorylated in cdc13-1 rad53-21 cdc15-2 cells at the restrictive temperature to the same extent than in a cdc15-2 mutant ( Figures 2C and D ) . The results were similar when the cdc14-3 allele was used instead of cdc15-2 in order to promote the final anaphase arrest ( unpublished observations ) . The inhibition of the hyper-phosphorylated forms of Bfa1 after telomere damage is dependent on Rad53 activity , and it is not exclusively associated to the rad53-21 allele , since the phosphorylation pattern of the protein was the same in cdc13-1 rad53Δ sml1Δ cdc15-2 cells at the restrictive temperature ( Figures S2B and C ) . Interestingly , and even though Bfa1 was not necessary to hold cell cycle progression after zeocin treatment , the Rad53-dependent inhibition of Bfa1 phosphorylation was also observed when cells were treated with this compound ( Figure S3A ) . We also analyzed the role of other components of the DDC in the phosphorylation of Bfa1 when telomeres are damaged . Mec1 is the main sensor of the DNA damage response in cdc13-1 cells , and triggers both the Rad53 and Chk1 branches of the DDC [1] . Accordingly , cdc13-1 mec1Δ sml1Δ cells did not arrest and Bfa1 was found to be hyper-phosphorylated at the non-permissive temperature ( Figures S3B and C ) . On the contrary , Tel1 was not required to maintain the cell cycle block when cdc13-1 cells were shifted at the restrictive temperature ( Figure S3C ) , and the hyper-phosphorylation of Bfa1 was effectively inhibited in the cdc13-1 tel1Δ mutant ( Figure S3B ) . This is in agreement with the fact that Rad53 is still phosphorylated and active in this mutant [32] . Finally , we checked whether the Chk1-dependent branch of the DDC plays any role in the regulation of Bfa1 phosphorylation . Although the cdc13-1 chk1Δ cdc15-2 mutant was also unable to hold the metaphase arrest induced by damaged telomeres at the restrictive temperature ( Figure 2E ) , Bfa1 was still hypo-phosphorylated in these cells in a Rad53-dependent manner ( Figure 2F ) . Therefore , our results demonstrate that Rad53 , but not Chk1 , inhibits the hyper-phosphorylation of Bfa1 after DDC activation . The Polo-like kinase Cdc5 phosphorylates Bfa1 in anaphase , which inhibits the GAP and activates MEN signaling [4] . It has been recently demonstrated that Rad53 partially inactivates Cdc5 after induction of DNA damage to protect Cdh1 from inhibition and therefore restrain spindle elongation [19] . Therefore , the partial inactivation of Cdc5 by Rad53 could also be determining the DDC-dependent hypo-phosphorylation of Bfa1 . To test this hypothesis , the cdc5-2 allele was introduced in cdc13-1 rad53-21 cells . At the restrictive temperature , cdc5-2 cells cannot exit mitosis . This phenotype can be recovered by deleting BFA1 , which indicates that Cdc5-2 is specifically impaired in its ability to inactivate Bfa1 [4] . After their synchronization in G1 using pheromone , cdc13-1 rad53-21 cdc5-2 cells were released at 34°C . Even though Cdc5-inactivation delayed cell cycle progression [4] , [17] , the cells finally reached anaphase ( Figure 3A ) , as previously shown [17] , [26] . However , and according to our hypothesis , the hyper-phosphorylation of Bfa1 observed in the cdc13-1 rad53-21 cdc15-2 mutant was prevented in cdc13-1 rad53-21 cdc5-2 cells ( Figures 3B and C ) . We obtained the same results using the cdc5-as1 allele to inactivate Polo-like kinase activity . Cdc5-as1 can be conditionally inhibited by adding the CMK-C1 ATP analog to the medium [33] . G1-synchronized cdc13-1 rad53-21 cdc5-as1 cdc15-2 cells were allowed to enter the cell cycle at 34°C in pheromone-free medium containing or not the CMK-C1 inhibitor . As previously observed with the cdc5-2 allele , inactivation of Cdc5-as1 prevented the hyper-phosphorylation of Bfa1 in the cells that escaped the DDC-dependent metaphase arrest ( Figures S4A and B ) . The hyper-phosphorylated forms of Bfa1 were also absent in anaphase-arrested cdc5-2 cells at the restrictive temperature ( our unpublished observations , [11] ) . This suggests that the two most heavily phosphorylated forms of Bfa1 are indicative of the Cdc5-dependent inhibition of the GAP in anaphase . Bfa1 localizes to the cytoplasmic side of the SPBs during mitosis [14] , and its localization was not significantly affected in cdc13-1 cells at the restrictive temperature ( Figure S5G ) or after treatment with zeocin ( our unpublished observations ) . Cdc5 also localizes to the SPBs during mitosis , where it phosphorylates Bfa1 ( Figure 3D , [5] ) . Interestingly , 3HA-Cdc5 strongly accumulated in the nucleus in cdc13-1 cells at the non-permissive temperature ( Figure 3E ) . This accumulation was also observed in cdc13-1 rad53-21 cells while in metaphase . However , the cells that managed to escape the DDC-induced arrest released 3HA-Cdc5 from the nucleus and the protein could be observed on the SPBs during anaphase , as in wild type cells ( Figure 3F ) . Although the strong accumulation of Cdc5 in the nucleus did not allow us to assess whether Polo kinase is loaded on the SPBs during the DDC-dependent arrest , our results suggest that Rad53 may additionally impair localization of Cdc5 to the SPBs , which would contribute to the inhibition of Bfa1 phosphorylation by Polo kinase . Interestingly , Bfa1 was still phosphorylated to some extent after Cdc5 inactivation in cdc13-1 rad53-21 cells , which indicates that this residual phosphorylation of the GAP is independent of the Polo-like kinase . Phosphorylation of a protein by Cdc5 is sometimes preceded by a priming phosphorylation of a Polo-binding site in the protein by cyclin-dependent kinases ( CDKs ) [34] , [35] . However , Bfa1 phosphorylation in cdc13-1 cells at the restrictive temperature is not dependent on CDK , since it was preserved after addition of the ATP analogue 1-NM-PP1 to cdc13-1 cells carrying the analogue-sensitive cdc28-as1 allele [36] ( Figure 4A ) . The protein kinase Kin4 plays a key role in promoting the inhibitory action of Bfa1 on MEN signaling after SPOC activation [11] , [12] . Kin4 impedes the inhibition of Bfa1 by Cdc5 [12] and actively excludes the GAP from the SPBs after the SPOC is triggered [13] . Since Tem1 localization to this structure depends on Bfa1 and it is essential for MEN signaling [15] , this exclusion also contributes to the inactivation of mitotic exit once the SPOC is activated . However , Kin4 does not determine Bfa1 phosphorylation in the cdc13-1-dependent arrest , and it is not necessary to maintain the functionality of the DDC ( Figures 4B and C ) . Therefore , the kinase that phosphorylates Bfa1 under these conditions remains to be identified , as it is also yet unclear whether this phosphorylation plays a role in the functionality of the DDC . Interestingly , deletion of BUB2 completely impairs the phosphorylation of Bfa1 , including the Cdc5-independent phosphorylation observed in cdc13-1 cells at the restrictive temperature ( Figures 4D and 4E ) . This result suggests that the Cdc5-independent phosphorylation could take place at the SPBs , since Bub2 is necessary for Bfa1 to localize on this structure . In any case , and independently of the nature of this remnant phosphorylation , our results demonstrate that the Rad53-inhibition of Polo kinase activity not only restrains spindle elongation [19] , but also promotes a Bfa1-dependent block of mitotic exit . According to our results , and even though mutants affected in either Rad53 or Bfa1 are deficient for the DDC in the presence of uncapped telomeres , they should display different phenotypes after induction of DNA damage . Rad53 should act upstream of Bfa1 , and in its absence neither spindle elongation ( Figure 2A ) nor mitotic exit ( Figure 1A ) could be halted by the DDC . On the contrary , in a bfa1Δ mutant mitotic exit should occur without spindle elongation taking place . To test this , we analyzed cell cycle progression and budding in cdc13-1 rad53-21 and cdc13-1 bfa1Δ cells . As previously indicated , and while the cdc13-1 mutant at the restrictive temperature arrested as large budded cells , cdc13-1 rad53-21 and cdc13-1 bfa1Δ cells could not hold the DDC-dependent arrest and entered a new cell cycle ( Figure 5A ) . However , and as previously shown for cdc13-1 rad53Δ sml1Δ cells ( Figure 1A ) , cdc13-1 rad53-21 cells exited mitosis faster than cdc13-1 bfa1Δ , as evidenced by the faster decrease in large budded cells ( Figure 5A ) . Furthermore , and as predicted , while most cdc13-1 rad53-21 cells that escaped the arrest elongated their spindles , carried out cytokinesis , and accumulated as unbudded cells , cdc13-1 bfa1Δ cells did not elongate their spindles and most of them entered a new cell cycle without cytokinesis , which led to an accumulation of rebudded cells with a single nucleus ( Figures 5A and 5B ) . A small percentage of cdc13-1 rad53-21 cells also entered a new cell cycle without cytokinesis , but in this case , and in contrast to the cdc13-1 bfa1Δ mutant , some rebudded cells had elongated their spindles before exiting mitosis and therefore showed two separated nuclei ( Figure 5B ) . The cdc13-1 rad53-21 bfa1Δ mutant behaved as the cdc13-1 rad53-21 ( Figure 5A ) , which further indicates that Bfa1 acts in the same pathway as Rad53 . According to this , and also in agreement with our analysis of Bfa1 phosphorylation ( Figure 2F ) , deletion of BFA1 accelerated the mitotic exit phenotype of cdc13-1 chk1Δ cells ( Figure S1B ) , which indicates that they are acting in different branches of the DDC . The previous genetic analysis was consistent with the viability observed for cdc13-1 , cdc13-1 rad53-21 , and cdc13-1 bfa1Δ cells when compared to wild type cells in a drop test . None of the strains carrying the cdc13-1 allele could grow already at 30°C ( Figure 5C ) . At 27°C , cdc13-1 showed very limited growth due to the cell cycle block induced by the presence of uncapped telomeres ( Figure 5C ) . Interestingly , and even though their viability was reduced when compared to wild type cells , inactivation of Rad53 by the introduction of the checkpoint-deficient allele allowed cdc13-1 rad53-21 cells to grow at this temperature to a higher extent than the cdc13-1 mutant ( Figure 5C ) . This indicates that the telomere damage induced by the cdc13-1 allele at 27°C , despite leading to a strong DDC-dependent cell cycle arrest , does not severely affect viability of the cells when the checkpoint is not functional . It is also worth noting that the cdc13-1 rad53-21 mutant could grow at 27°C because most of the cells that escaped the arrest elongated their spindles and carried out cytokinesis as they exited mitosis ( Figure 5A ) . In contrast , we have demonstrated that cdc13-1 bfa1Δ cells mainly exited mitosis as mono-nucleated and rebudded cells , which are not viable ( Figures 5A and B ) . Accordingly , the cdc13-1 bfa1Δ mutant at 27°C showed extremely limited viability at 27°C ( Figure 5C ) . Together , our results are consistent with a dual role of the Rad53-dependent branch of the DDC , which not only prevents spindle elongation [19] , but also impedes MEN signaling by maintaining Bfa1 in a hypo-phosphorylated and active form that inhibits Tem1 . To further demonstrate that inhibition of Bfa1 phosphorylation by Cdc5 is essential for the Rad53-dependent cell cycle arrest originated after DNA damage , we analyzed the effect of mutations that affect Bfa1 phosphorylation on the mitotic exit phenotype of rad53-21 cells . We first made use of the cdc5-2 mutant , which can elongate the spindle and reach anaphase , but it is specifically impaired in its ability to phosphorylate and inactivate Bfa1 ( Figures 3B and C , [4] ) . A cdc13-1 rad53-21 cdc5-2 mutant at the restrictive temperature accumulated large budded cells due to the anaphase block ( Figures 3A and 6A ) , although a small percentage of cells finally managed to escape the arrest ( Figure 6A ) . According to our results , and if Cdc5-dependent phosphorylation of Bfa1 plays a key role in restraining mitotic exit after DDC activation , the arrest observed for cdc13-1 rad53-21 cdc5-2 cells could be explained by a Bfa1-dependent inhibition of mitotic exit due to the inability of Cdc5 to phosphorylate the GAP , even though Rad53 is inactive . If so , we reasoned that deletion of BFA1 should accelerate the mitotic exit phenotype of the cdc13-1 rad53-21 cdc5-2 cells . Indeed , this was the case . Both the decrease in the percentage of large budded cells and the accumulation of rebudded cells were accelerated in the cdc13-1 rad53-21 cdc5-2 mutant in the absence of BFA1 ( Figure 6A ) . Additionally , the percentage of rebudded cells was increased and matched the level observed in cdc13-1 rad53-21 ( Figure 6A ) . Exit from mitosis was still delayed in cdc13-1 rad53-21 cdc5-2 bfa1Δ cells as compared to the cdc13-1 rad53-21 mutant probably due to problems in the progression through the cell cycle associated to the defect in Cdc5 [4] , [17] . To strengthen our conclusions , we followed a parallel approach . The Bfa1-4A mutant cannot be efficiently phosphorylated by Cdc5 ( Figures S5A and B , [37] ) . This mutant has been shown to symmetrically localize in anaphase [37] , and its localization was not significantly affected after telomere damage ( Figure S5G ) . Additionally , this mutant delayed mitotic exit in otherwise wild type cells [37] and it could hold the DDC-dependent cell cycle arrest induced by telomere damage ( Figure 6B ) , zeocin treatment ( Figure S5C ) or the treatment with other DNA damaging agents ( Figures S5D , E , and F ) , as expected by the fact that Cdc5-phosphorylation leads to inactivation of Bfa1 and promotes mitotic exit [37] . However , and according to our hypothesis , the Bfa1-4A mutant should delay the mitotic exit phenotype of cdc13-1 rad53-21 cells , since even though Cdc5 would be active , it could not promote MEN signaling by the inhibition of Tem1's GAP . Indeed , cdc13-1 rad53-21 BFA1-4A cells exited mitosis at the restrictive temperature later than a cdc13-1 rad53-21 mutant , as demonstrated by the decrease in the percentage of large-budded cells ( Figure 6B ) . Additionally , the rebudding percentage was lower than in the case of cdc13-1 rad53-21 cells ( Figure 6B ) . MEN signaling was only delayed and not completely avoided probably due to the fact that an active Cdc5 kinase promotes mitotic exit not only through Bfa1 inactivation but also downstream of the GAP 9 , 10 , 38–40 . The striking similarity between the cdc13-1 rad53-21 BFA1-4A and the cdc13-1 rad53-21 cdc5-2 mutants demonstrates that the inhibition of Bfa1 phosphorylation by Polo-like kinase plays an essential role in the Rad53-dependent arrest generated by telomere damage .
The inhibition of MEN signaling by the SPOC is essential for anaphase cells with mispositioned spindles in order to provide them with time to reposition their spindles before exiting mitosis [3] . Interestingly , the functionality of mitotic checkpoints that are triggered earlier in the cell cycle is also dependent on the active inhibition of mitotic exit . In all the previous cases , this inhibition of MEN signaling is achieved by means of the activation of Bfa1/Bub2 , a two-component GAP that negatively regulates Tem1 [4] , [41] . While the SAC and the SPOC maintain Bfa1 in an active state by preventing its phosphorylation by the Polo-like kinase Cdc5 , a different mechanism was originally proposed for the inhibition of MEN signaling by the DDC that would rely on a Rad53-dependent yet Cdc5-independent activation of Bfa1 [4] , [17] . Here , we have analyzed the inhibition of mitotic exit after generation of chromosomal DNA damage and the resulting activation of the DDC . This inhibition of MEN signaling seems to be specifically required to face certain types of damage . In this way , and while Bfa1-Bub2 is necessary to maintain a DDC-dependent cell cycle arrest due to the generation of uncapped telomeres , the viability of a bfa1Δ mutant is not affected as compared with that of wild type cells after the treatment with a wide variety of DNA damaging agents , including UV or gamma-rays and compounds that generate DSBs or replicative stress . In agreement with this , bub2Δ cells do not show increased sensitivity to MMS or UV irradiation [42] . Even though uncapped telomeres resemble one half of a DSB , the checkpoint response partially differs for both structures [43] . Our results are extremely interesting , and provide a new piece of evidence that demonstrates that the cells respond differently to DNA DSBs than to the presence of uncapped telomeres . A possible explanation for these observations is that telomere damage could be triggering a weaker G2/M arrest that would rely on the additional inhibition of mitotic exit by Rad53 . Our data , however , exclude this possibility , since Bfa1 is essential for the functionality of the DDC even when the DNA damage caused by telomere damage is increased to the same extent than in cells treated with zeocin . Instead , we favor a distinct alternative scenario , in which Bfa1-Bub2 could collaborate with the DDC to specifically protect the cells when telomere integrity is compromised . Interestingly , it has been recently shown that in cells from the marsupial Potorous tridactylis , laser-induced damage at the telomeres during anaphase causes cell cycle delay and cytokinesis failure [44] . Therefore , our results are in agreement with the idea of a specific requirement for mitotic exit inhibition when telomeres are damaged [44] . Our study also sheds light into the mechanism by which mitotic exit is inhibited in the presence of uncapped telomeres . To this end , we have analyzed the pattern of phosphorylation of Bfa1 both during a normal cell cycle and after checkpoint activation . Bfa1 is gradually phosphorylated throughout the cell cycle , reaching its highest phosphorylated status once the cells are in anaphase . The hyper-phosphorylation of Bfa1 during anaphase is dependent on the Polo-like kinase Cdc5 , and it is therefore likely to represent the phosphorylation events that inactivate the Bfa1-Bub2 GAP at this cell cycle stage [8] . We have demonstrated that in response to DNA damage , Bfa1 is maintained in a hypo-phosphorylated and active state that inhibits Tem1 activity and therefore impedes mitotic exit . Additionally , we have also shown that the maintenance of the active form of Bfa1 after DDC activation relies on the Rad53-dependent inhibition of the Polo-like kinase Cdc5 , but not on the Chk1-dependent branch of the checkpoint . Our results contrast with previous observations suggesting that Bfa1 was hyper-phosphorylated in a Rad53-dependent manner after DDC activation [4] . We do not know the basis for this discrepancy , which might be due to the use of a different genetic background or to the possibility that a 3HA-tag in the C-terminus [4] instead of the N-terminus of Bfa1 could be affecting its functionality after DDC activation . In any case , our results are fully consistent with our genetic analysis of the Rad53-Cdc5-Bfa1 pathway and are in agreement with a more recent report that demonstrates that Cdc5 activity is inhibited in a Rad53-dependent manner after DNA damage [19] . It is worth noting that the inhibition of the Cdc5-dependent hyper-phosphorylation of Bfa1 also takes place when cells are treated with zeocin , a compound that generates DSBs in the DNA . However , and as previously stated , our data clearly demonstrates that inhibition of Bfa1 is not required to maintain a DDC-dependent cell cycle arrest when DNA DSBs are generated . The specific requirement for mitotic exit inhibition after telomere damage suggests that additional mechanisms to inhibit cell cycle progression must be triggered when cells are exposed to other types of DNA damage that are not induced in the presence of uncapped telomeres , therefore relieving the requirement for the inactivation of the MEN by Bfa1-Bub2 . Ours and previous results demonstrate that Rad53 fulfills a dual role in preventing cell cycle progression in response to telomere damage: it restrains the metaphase-to-anaphase transition avoiding spindle elongation [19] , but it also prevents mitotic exit by maintaining Bfa1 in an active state that blocks MEN signaling ( Figure 6C ) . This dual action of Rad53 is in agreement with the different behavior in terms of mitotic exit observed at the restrictive temperature for cdc13-1 rad53-21 cells , which can promote both spindle elongation and mitotic exit , and mutants in which exclusively either spindle elongation ( e . g . , cdc13-1 bfa1Δ ) or mitotic exit ( e . g . , cdc13-1 rad53-21 BFA1-4A ) is blocked . The inactivation of only one of the two Rad53-dependent functions in a cdc13-1 background reduces the ability of the cells to escape from the G2/M arrest induced at the restrictive temperature as compared to cdc13-1 rad53-21 cells . Therefore , the Cdc5-dependent inhibitory phosphorylation of Bfa1 by Rad53 plays a key role in the maintenance of the DDC-dependent cell cycle arrest determined by telomere damage . Interestingly , and besides regulating the activity of Cdc5 , our results suggest that Rad53 may also regulate the localization of the Polo-like kinase to the outer plaque of the SPBs , where it phosphorylates and inhibits Bfa1 [4] . This could represent an additional mechanism by which Rad53 blocks mitotic exit in the presence of damaged telomeres . Even though Rad53 inhibits the hyper-phosphorylation of Bfa1 by Cdc5 , Bfa1 still displays a considerable degree of phosphorylation in the G2/M cell cycle arrest induced by the DDC in response to telomere damage . This phosphorylation of Bfa1 is Cdc5-independent , which is in agreement with previous observations [11] , [41] . We have demonstrated that this remnant phosphorylation is also not dependent on Clb-CDK activity or the Kin4 kinase . Therefore , our results suggest the existence of an additional kinase that phosphorylates Bfa1 during metaphase . Since deletion of BUB2 completely abrogates Bfa1 phosphorylation and the localization of Bfa1 and Bub2 on the spindle poles is interdependent [14] , it is likely that this Cdc5-independent phosphorylation of Bfa1 could also take place at the SPBs . At present , however , the identity of this kinase remains to be established , as it is also not known what is the actual role of this phosphorylation in the regulation of the activity of Bfa1 during a normal cell cycle or in the functionality of the different cell cycle checkpoints . One possibility is that this phosphorylation could be involved in the regulation of Bfa1 loading onto the SPBs , since Bfa1 and Nud1 ( its anchor on the SPB ) preferentially co-immunoprecipitate in their phosphorylated forms [45] . Based on our results , the inhibition of Bfa1 phosphorylation by Cdc5 is a common theme for all the mitotic checkpoints that rely on the inhibition of mitotic exit . In agreement with this observation , overexpression of Cdc5 not only inhibits Bfa1-Bub2 activity in anaphase , but it is also able to bypass the cell cycle arrest induced by the activation of the DDC and the SAC [4] , [40] , [46] . These surveillance mechanisms therefore mainly diverge in the strategies by which the inhibition of Polo-kinase is achieved in each case . In this way , while Kin4 plays a critical role in the functionality of the SPOC [11] , [12] , this kinase is dispensable for the DDC and the SAC . Even though DDC activation in mammalian cells mainly blocks mitotic entry , exit from mitosis and cytokinesis are also restrained in these cells in response to DNA damage [44] , [47] , [48] . Furthermore , Polo-like Kinase I ( Plk1 ) is inhibited after DNA damage in an ATM and ATR-dependent manner [49] , [50] , and it has been demonstrated to interact and co-localize in centrosomes and the midbody during mitosis with Chk2 , the mammalian homolog of Rad53 [51] . Therefore , our data could provide new insights into common mechanisms by which exit from mitosis is prevented when the DNA is damaged in higher eukaryotes .
All strains are derivatives of W303 and are described in Table S1 . Unless otherwise indicated , all the strains are RAD5 . Strain F1333 , which expresses Bfa1-GFP , was constructed by first linearizing the pRS304-BFA1-GFP with the NruI endonuclease and then integrating it within the BFA1 promoter in F533 , a strain in which the endogenous BFA1 gene is deleted . Strain F1367 , which expresses Bfa1-4A-GFP , was constructed as F1333 , but integrating the pRS304-BFA1-4A plasmid [37] . Immunofluorescence was performed as described in [15] . In brief , cells were fixed overnight at 4°C in 3 . 7% formaldehyde , washed twice with 0 . 1 M potassium phosphate buffer ( pH 6 . 4 ) , and resuspended in 1 . 2 M sorbitol/0 . 12 M KH2HPO4/0 . 033 M citric acid ( pH 5 . 9 ) . Fixed cells were digested for 15 min at 30°C with 0 . 1 mg/ml zymolyase-100T ( US Biological ) and 1/10 volume of glusulase ( Perkin Elmer ) . Anti-tubulin ( Abcam ) and anti–rat FITC ( Jackson ImmunoResearch ) antibodies were used at 1∶200 . 3HA-Cdc5 was detected using anti-HA antibody ( HA . 11; Covance ) at 1∶500 and anti–mouse Cy3 antibody ( Jackson ImmunoResearch Laboratories , Inc . ) at 1∶1000 . Samples for GFP and DAPI imaging were prepared as described in [15] . Microscope preparations were analyzed and imaged at 25°C using a DM6000 microscope ( Leica ) equipped with a 100×/1 . 40 NA oil immersion objective lens , A4 , L5 , and TX2 filters , and a DF350 digital charge-coupled device camera ( Leica ) . Pictures were processed with LAS AF ( Leica ) and ImageJ ( http://rsbweb . nih . gov/ij/ ) software . Cells were fixed in 70% ethanol , incubated for 12 h in phosphate-buffered saline with 1 mg/ml of RNase A , and stained for 1 h with 5 µg/ml propidium iodide . After sonication of the sample to separate single cells , DNA content was analyzed in a FACSCalibur flow cytometer ( Becton Dickinson ) . Protein extracts were prepared using the TCA precipitation method described in [11] and were loaded on 6% polyacrylamide gels . Electrophoresis was carried out using a SE600 Hoefer electrophoresis system . Samples subjected to phosphatase treatment were incubated with 150 U of bovine phosphatase alkaline ( Sigma ) for 12 h at 37°C in 50 mM Tris-HCl , 1 mM MgCl2 buffer ( pH = 9 ) . 3HA-Bfa1 was examined with monoclonal HA . 11 ( Covance ) at 1∶5000 and anti–mouse HRP-linked antibodies ( GE Healthcare ) at 1∶10000 . GFP-tagged proteins were analyzed using JL-8 Living colors® monoclonal antibody ( Clontech ) at 1∶1000 and anti-mouse HRP-linked antibody ( GE Healthcare ) at 1∶2000 . Pgk1 levels were measured using anti-Pgk1 antibody ( Invitrogen ) at 1∶10000 and anti-mouse HRP-linked antibody ( GE Healthcare ) at 1∶20000 . In all cases , the protein signal was detected using the Western Bright ECL system ( Advansta ) . | A key aspect of the division of cells is that the genomic material must be carefully duplicated , protected from damage , and correctly distributed during this process . When the cells detect problems that affect the integrity or the proper distribution of the genome , they trigger different surveillance mechanisms to stop the division process until the problem is fixed . The functionality of some of these surveillance mechanisms depends on the inhibition of the final stages of cell division , a process known as mitotic exit . This is the case for the DNA damage checkpoint ( DDC ) , which is triggered by chromosomal DNA damage . Here , we propose that the inhibition of mitotic exit is specifically required by the DDC only when the telomeres ( the chromosomal ends ) are damaged . Additionally , we have demonstrated that the DDC blocks mitotic exit by inhibiting the key cell cycle regulator Cdc5 , which triggers the inactivation of Bfa1 and Bub2 , two negative regulators of the mitotic exit process . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | Inhibition of the Mitotic Exit Network in Response to Damaged Telomeres |
Interferon regulatory factor ( IRF ) -1 is an immunomodulatory transcription factor that functions downstream of pathogen recognition receptor signaling and has been implicated as a regulator of type I interferon ( IFN ) -αβ expression and the immune response to virus infections . However , this role for IRF-1 remains controversial because altered type I IFN responses have not been systemically observed in IRF-1-/- mice . To evaluate the relationship of IRF-1 and immune regulation , we assessed West Nile virus ( WNV ) infectivity and the host response in IRF-1-/- cells and mice . IRF-1-/- mice were highly vulnerable to WNV infection with enhanced viral replication in peripheral tissues and rapid dissemination into the central nervous system . Ex vivo analysis revealed a cell-type specific antiviral role as IRF-1-/- macrophages supported enhanced WNV replication but infection was unaltered in IRF-1-/- fibroblasts . IRF-1 also had an independent and paradoxical effect on CD8+ T cell expansion . Although markedly fewer CD8+ T cells were observed in naïve animals as described previously , remarkably , IRF-1-/- mice rapidly expanded their pool of WNV-specific cytolytic CD8+ T cells . Adoptive transfer and in vitro proliferation experiments established both cell-intrinsic and cell-extrinsic effects of IRF-1 on the expansion of CD8+ T cells . Thus , IRF-1 restricts WNV infection by modulating the expression of innate antiviral effector molecules while shaping the antigen-specific CD8+ T cell response .
The rapid triggering of an IFN-αβ response results in the early control of virus infection in mammalian cells . Detection of RNA viruses occurs through the recognition of specific sequence motifs , secondary structure , or modification of viral nucleic acids by pattern recognition receptors ( PRR ) in the cytosol ( RIG-I and MDA5 ) and endosome ( TLR3 , TLR7 , and TLR8 ) [1] . PRR binding of viral RNA signals constituent adaptor molecules ( IPS-1 , TRIF , and MyD88 ) to activate transcription factors and induce expression of IFN-αβ genes . A current paradigm for IFN production after RNA virus infection describes a positive feedback model that is modulated by the transcription factors interferon regulatory factors ( IRF ) -3 and IRF-7 [2] , [3] . In the initial phase , viral nucleic acid sensing induces nuclear localization of IRF-3 , which stimulates gene transcription and production of IFN-β and IFN-α4 by infected cells . In the second phase , these IFNs bind to the common IFN-αβ receptor in a paracrine and autocrine manner and signal through the JAK-STAT pathway resulting in the induced expression of hundreds of interferon stimulated genes ( ISG ) ( e . g . , PKR , RNAse L , viperin , ISG15 , ISG54 , ISG56 , IFITM3 , and ISG20 ) , which limit viral replication through multiple mechanisms [4] , [5] , [6] , [7] , [8] , [9] . IRF-7 is both an ISG and a transcriptional activator and participates in an IFN amplification loop by inducing IFN-β and many subtypes of IFN-α [3] , [10] . West Nile virus ( WNV ) is a single-stranded positive polarity RNA virus in the Flaviviridae family that cycles in nature between birds and Culex mosquitoes . Humans , which are dead-end incidental hosts , can develop a febrile illness that progresses to flaccid paralysis , meningitis , or encephalitis [11] . WNV is emerging in the Western hemisphere as greater than 30 , 000 human cases of severe infection have been diagnosed in the United States since 1999 , and millions have been infected and remain undiagnosed [12] . Experiments in mice have identified immune mechanisms of control with significant contributions from inflammatory cytokines , chemokines , complement , B CD4+ , and CD8+ T cells ( reviewed in [13] , [14] ) . In particular , type I IFN ( IFN-αβ ) has an essential function in restricting cell and tissue tropism as IFN-αβR-/- or IFN-β-/- mice rapidly succumb to WNV infection [15] , [16] , [17] . Although IFN-β shapes a response that restricts WNV infection in many cells and tissues , its induction appears to differ in a cell type-dependent manner . In fibroblasts and neurons , IFN-β is triggered through an IPS-1-dependent and IRF-3 and IRF-7-dependent transcriptional signal [10] , [18] . However , additional transcription factors contribute to the regulation of IFN-β as gene expression and protein production were preserved in IRF-3-/- x IRF-7-/- myeloid cells after WNV infection [19] . Based on these studies , we hypothesized that additional transcriptional factors could regulate IFN-β and/or ISG expression . In particular , the function of IRF-1 in the context of infection by WNV or other flaviviruses remained unknown . We viewed IRF-1 as a candidate transcription factor because it had been reported to contribute to IFN-β induction in some experimental systems . IRF-1 was initially identified as a Newcastle disease virus ( NDV ) -induced transcription factor that activated IFN-β gene transcription in cell culture [20] , [21] , [22] . Other experiments suggested that IRF-1 activation also regulated genes that directly limit replication of several viruses ( encephalomyocarditis ( EMCV ) , NDV , hepatitis C ( HCV ) , yellow fever , Sindbis , and vesicular stomatitis ( VSV ) viruses ) independently of IFN production [23] , [24] , [25] . In addition to these functions , IRF-1 transduces part of the IFN-γ signal ( reviewed in [26] ) . Consistent with this , EMCV infection in IRF-1-/- fibroblasts was associated with a decrease in IFN-γ-stimulated genes [27] , and IFN-γ-induced upregulation of cytokines and chemokines in macrophages ( Mφ ) required the transcriptional activity of IRF-1 [28] , [29] , [30] , [31] . Studies with deficient mice have confirmed an important role for IRF-1 in controlling infection against EMCV or murine γ-herpesvirus 68 ( γHV68 ) [27] , [32] . However , IRF-1-/- mice did not show defects in the systemic type I IFN response in vivo after NDV infection [33] suggesting that it might regulate IFN genes in a pathogen and cell type-dependent manner . Independent of its possible effects on transducing IFN signals , IRF-1 has been defined as a tumor suppressor gene [34] , [35] , [36] that inhibits cell proliferation or enhances apoptosis in a manner dependent on its transcriptional activity ( reviewed in [37] , [38] ) . IRF-1 also regulates adaptive immune responses . Naïve IRF-1-/- mice exhibit blunted levels of CD8+ T cells in peripheral lymphoid organs in part , due to decreased class I MHC expression and impaired selection in the thymus [39] , [40] , [41] . Moreover , IRF-1-/- Mφ have decreased IL-12 production during bacterial and parasitic infection , whereas IRF-1-/- CD4+ T cells show reduced IL-12 receptor expression and are prone to TH2 skewing [42] , [43] , [44] . Nonetheless , because adoptive transfer of wild type peritoneal exudates cells restores induction of Th1 cells in IRF-1-/- mice , these defects are not entirely T cell intrinsic [45] . Finally , increased numbers of CD4+CD25+FoxP3+ regulatory T cells were reported in naïve IRF-1-/- mice [46] . Here , we evaluated the function of IRF-1 in the context of immunity to WNV infection . IRF-1-/- mice were vulnerable to WNV infection with enhanced viral replication in peripheral tissues and subsequent early dissemination to the brain and spinal cord . Despite a substantial baseline defect in the numbers of naïve CD8+ T cells after WNV infection , IRF-1-/- mice rapidly expanded antigen-specific IFN-γ-producing , granzyme B+ CD8+ T cells that were capable of killing targets and clearing virus from infected neurons . Although WNV-specific CD8+ T cells proliferated more rapidly in IRF-1-/- mice , they did not “catch up” quickly enough to mitigate the damage caused by early CNS dissemination that was due to impaired control of WNV in peripheral tissues .
To understand the contribution of IRF-1 in controlling WNV infection in vivo , wild type and IRF-1-/- C57BL/6 mice were infected subcutaneously with 102 PFU of a highly pathogenic WNV strain ( New York 2000 ) and monitored for survival . IRF-1-/- mice were more vulnerable to WNV infection , with a 0% survival rate and a mean time to death of 9 . 5±1 . 1 compared to age-matched wild type mice , which had a 65% survival rate and a mean time to death of 10 . 7±1 . 6 ( P<0 . 0001 , Fig . 1A ) . To begin to define how an IRF-1 deficiency increased the susceptibility of mice to WNV infection , we measured viral burden by fluorogenic quantitative RT-PCR or viral plaque assay at days 1 , 2 , 4 , 6 and 8 post infection in serum , peripheral organs ( draining lymph nodes , spleen and kidney ) and the CNS ( brain and spinal cord ) . Because of the early viral phenotype in peripheral tissues , and prior results with other viruses implicating IRF-1 as an activator of IFN-αβ gene transcription in cell culture [20] , [21] , [22] , we hypothesized that IRF-1 might contribute to induction of type I IFN responses locally in specific tissues or compartments . To test this , we measured levels of IFN-α and β mRNAs in the draining lymph nodes of WNV-infected mice at day 2 after infection . In wild type and IRF-1-/- mice , the induction of the IFN-α/β genes was comparable as both showed increased levels of IFN-α and β mRNA after WNV infection ( ∼10 to 20 fold increase for IFN-α and IFN-β ) ( Fig 2A ) . We also observed no statistically significant difference in induction of SOCS-1 and SOCS-3 in the draining lymph node , genes that are induced by IFN-γ [48] . Although IRF-3 and IRF-7 jointly regulate the induction of systemic IFN-αβ after WNV infection , their combined deficiency does not abrogate IFN accumulation in serum [19] . We therefore hypothesized that IRF-1 might contribute to the systemic levels of IFN-αβ in circulation after infection . Mice were infected with WNV and the presence of biologically active IFN in serum was monitored using a previously validated EMCV-L929 cell bioassay [47] , [49] . Type I IFN activity in the serum of infected wild type mice peaked at day 3 and then decreased by day 5 ( Fig 2B ) . IFN activity in the WNV-infected IRF-1-/- mice was equivalent or greater than that observed in wild type mice . For example , 5 to 10-fold higher levels of type I IFN activity were observed in IRF-1-/- mice at day 3 after infection ( P<0 . 001 ) . Thus , a deficiency of IRF-1 in vivo did not diminish early type I IFN levels in lymphoid tissues or circulation after WNV infection; these results are consistent with that observed previously after NDV infection [33] . To better understand the viral replication phenotype of IRF-1-/- mice , we investigated how an absence of IRF-1 affects WNV infection of primary mouse embryonic fibroblasts ( MEF ) and Mφ . While Mφ are physiologic targets of WNV infection in vivo [6] , [50] , [51] , we used MEFs as a comparison cell type because historically they have been used to study the significance of IRF in regulating cellular antiviral immune responses [3] , [19] , [52] . Multi-step viral growth curves performed in wild type and IRF-1-/- MEF showed no significant differences in WNV replication at any of the time points tested ( P > 0 . 2 ) ( Fig 3A ) . However , increased replication was observed at 48 and 72 hours after infection in IRF-1-/- Mφ when compared to wild type cells ( Fig 3B , 2 . 5-fold , P = 0 . 02 and 7 . 4-fold , P = 0 . 001 , respectively ) . Earlier studies showed that a defect in type I IFN signaling enhanced WNV infection of primary Mφ [15] while recent work indicated that IRF-3 and IRF-7 only partially regulated IFN-β induction after WNV infection in Mφ [19] . To define a possible role for IRF-1 in modulating IFN gene induction , cells were infected with WNV and levels of IFN-α and β mRNA were measured by qRT-PCR . Notably , levels of IFN-α and IFN-β were significantly higher at 48 hours in IRF-1-/- Mφ ( Fig 3D and F ) ; the increased levels were likely a consequence of greater viral replication , analogous to that seen in MyD88-/- Mφ [53] . Consistent with this , IFN-αβ gene induction was not altered in IRF-1-/- MEF ( Fig 3C and E ) . Thus , a deficiency in IRF-1 enhanced viral growth in Mφ yet this was not associated with a defect in the type I IFN response . Since previous studies have defined a role of IRF-1 in mediating the antiviral effects of IFN-γ signaling [54] , we evaluated the contribution of IRF-1 to the antiviral effects of either IFN-β or IFN-γ in Mφ after WNV infection . We pretreated wild type and IRF-1-/- Mφ with increasing concentrations of either IFN-β or IFN-γ , infected with WNV , and measured viral burden . Pretreatment with increasing amounts of IFN-β inhibited WNV replication in both wild type and IRF-1-/- Mφ ( 2- , 66- and 104-fold inhibition versus 1 . 6- , 20- and 193-fold inhibition , respectively , P>0 . 3 at 1 , 10 and 100 IU/ml ) ( Fig 4A ) . Whereas pretreatment with IFN-γ significantly decreased viral infection in wild type Mφ , it had virtually no inhibitory effect in IRF-1-/- cells ( 3 . 3- , 34- and 116-fold inhibition in wild type cells versus 1 . 3- , 1 . 6- and 2 . 9-fold inhibition in IRF-1-/- cells , P<0 . 05 at 1 , 10 and 100 IU/ml ) ( Fig 4B ) . Thus , in Mφ , IRF-1 is required to mediate the inhibitory effect of IFN-γ but dispensable for the antiviral activity of IFN-β against WNV infection . To determine whether the increased WNV replication in IRF-1-/- Mφ was associated with a paracrine antiviral effect of IFN-γ , we performed depletion experiments with a neutralizing anti-mouse IFN-γ monoclonal antibody ( MAb H22 ) [55] . To confirm the neutralizing capacity of the H22 MAb , wild type Mφ were pretreated with increasing amounts of IFN-γ that was pre-mixed with either H22 or an isotype control MAb , and then infected with WNV; at 48 hours , viral burden was quantified by plaque assay . Pre-incubation of IFN-γ with the isotype control MAb resulted in a dose-dependent inhibition in WNV infection ( Fig 4C ) . Since these controls confirmed that H22 MAb neutralized the IFN-γ-dependent antiviral activity against WNV , we performed multi-step viral growth in wild type and IRF-1-/- Mφ in the presence of H22 or the isotype control MAb , each in the absence of exogenous IFN-γ . Notably , WNV replication in the presence of the isotype control MAb gradually increased in both wild type and IRF-1-/- Mφ ( Fig 4D and 4E ) and exhibited growth kinetics virtually identical to that observed in the absence of MAb ( see Fig 3B ) . Similar results were obtained on WNV growth in the presence of H22 . These results imply that WNV-infected Mφ did not produce sufficient IFN-γ to explain the difference in viral growth between wild type and IRF-1-/- cells . Thus , although IRF-1 is essential for mediating the antiviral activity of IFN-γ against WNV in Mφ , the cell-intrinsic difference in replication in IRF-1-/- cells must occur independently of the IFN-γ response . Innate immune signaling can modulate the induction and quality of antigen-specific antibody responses after viral infection [18] , [56] . As a depressed antiviral antibody response can promote increased viremia and early dissemination of WNV [57] , we evaluated whether a deficiency of IRF-1 modulated humoral immune responses . Notably , similar or higher levels of WNV-specific IgM and IgG were detected in IRF-1-/- mice at days 5 and 8 after infection , and no difference in neutralization titer was observed ( Fig 5 ) . Thus , the virologic phenotype observed in IRF-1-/- mice likely was not due to a primary defect in B cell function . Naïve IRF-1-/- mice have quantitative defects in the number of CD8+ T cells in peripheral lymphoid organs because of impaired negative and positive selection in the thymus [39] , [40] , [41] . Moreover , IRF-1-/- mice also have defects in TH1 responses due to altered production of IL-12 by Mφ and hypo-responsiveness of T cells to the effects of IL-12 during bacterial and parasite infections [42] , [43] , [44] . Because WNV is controlled in part , by cytolytic and IFN-γ secreting CD8+ T cells that clear infection from peripheral and CNS tissues [58] , [59] , [60] , [61] , [62] , we evaluated whether the severe clinical phenotype after infection in IRF-1-/- mice was explained in part , by a failure of antigen-specific cells to expand and migrate to infected tissues . Initial studies confirmed a reduced percentage ( ∼5-fold , P<0 . 001 ) and number ( ∼25-fold , P<0 . 001 ) of CD8+ T cells in the spleens of naïve IRF-1-/- mice ( Fig 6A and B ) . In comparison , the percentage and absolute number of CD4+ T cells were largely intact in naive IRF-1-/- mice . However , within eight days of WNV infection , the total CD8+ T cell population expanded in IRF-1-/- mice , such that there was a smaller difference in percentage ( ∼3-fold ) and number ( ∼10-fold ) compared to wild type mice ( Fig 6C and D ) . Remarkably , ∼60% of splenic CD8+ T cells taken directly from WNV-infected IRF-1-/- mice expressed high levels of granzyme B ( GrB ) , establishing the presence of a large pool of potentially cytolytic CD8+ T cells ( Fig 6E , F , and K ) ; the majority of these GrB+ CD8+ T cells were specific for one Db-restricted immunodominant epitope on the WNV NS4b protein ( Fig 6I and J , and Fig S1 ) . In comparison , ∼0 . 2% of CD8+ T cells expressed GrB in naïve IRF-1-/- mice ( data not shown ) . Consistent with this , a significantly higher percentage ( ∼30% compared to ∼5% for wild type , P = 0 . 001 ) of IRF-1-/- CD8+ T cells produced IFN-γ in response to WNV NS4b peptide restimulation ex vivo ( Fig 6G , H , and K ) . This result was not specific to the NS4b epitope as similar results were observed after re-stimulation with an E protein-derived peptide ( E771: IALTFLAV ) that comprises a subdominant Kb-restricted T cell epitope ( data not shown ) . In comparison , neither wild type nor IRF-1-/- CD8+ T cells from WNV-infected mice produced IFN-γ in the absence NS4b or E peptide restimulation ( 0 . 1 to 0 . 2% IFN-γ+ CD8+ positive cells from wild type and IRF-1-/- mice , respectively ) . Overall , in the context of WNV infection , IRF-1-/- mice paradoxically had more robust antigen-specific T cell responses . The rapid and antigen-specific expansion of CD8+ T cells in IRF-1-/- mice after WNV infection was surprising , given the previously described defects in negative and positive T cell selection in the thymus and IL-12 responsiveness in the periphery . We hypothesized that one of several mechanisms could explain this phenomenon: ( a ) WNV-specific IRF-1-/- CD8+ T cells were inherently of higher affinity because they had been selected on IRF-1-/- stromal cells expressing lower levels of class I MHC molecules [39]; ( b ) IRF-1 directly or indirectly modulated the numbers of CD4+CD25+FoxP3+ regulatory T cells ( Treg ) after WNV infection . A decrease in Treg numbers was recently shown to cause an increase in numbers of WNV-specific CD8+ T cells [63]; or ( c ) IRF-1 regulated the threshold for activation-induced proliferation after antigen recognition , consistent with its tumor suppressor properties . We tested these three hypotheses in the following ways . To evaluate whether WNV-specific IRF-1-/- CD8+ T cells were of higher affinity than wild type CD8+ T cells , ex vivo restimulation was performed over a wide range of NS4b peptide concentrations . Notably , no difference was observed in the concentration of peptide or the time required for activation of CD8+ T cells and expression of IFN-γ ( Fig 7A and B ) . Moreover , we observed a slightly ( 2 . 4-fold , P = 0 . 004 ) decreased number of CD4+CD25+FoxP3+ Tregs in the spleens of WNV-infected IRF-1-/- mice , which could contribute to the skewing of the CD8+ T cell response ( Fig 7C ) . However , this reduction is an unlikely explanation because the CD8+ T cell response disparity in IRF-1-/- mice was far greater than that observed in FoxP3-/- mice , which entirely lack Tregs [63] . To assess how IRF-1 regulates the rapid expansion of CD8+ T cells after WNV infection , we performed competitive adoptive transfer experiments with wild type ( CD45 . 1 ) and IRF-1-/- ( CD45 . 2 ) CD8+ T cells into IRF-1+/+ RAG1-/- mice . Within 12 hours of transfer , slightly greater numbers ( ∼1 . 8-fold ) of wild type ( CD45 . 1 ) CD8+ T cells were observed in recipient RAG1-/- mice ( data not shown ) . Animals subsequently were infected with WNV and harvested 8 days later for profiling of CD8+ T cells . Notably , we observed a cell-intrinsic effect of IRF-1 on the proliferative potential of CD8+ T cells as the wild type ( CD45 . 1 ) cells rapidly expanded compared to IRF-1-/- ( CD45 . 2 ) counterparts ( 4 . 1 and 4 . 4-fold increase in percentage and number , respectively , P≤0 . 02; Fig 8A ) . Consistent with this , although the percentage of NS4b-specific IFN-γ+ or GrB+ CD8+ T cells was equivalent ( P > 0 . 6 ) in both the wild type ( CD45 . 1 ) and IRF-1-/- ( CD45 . 2 ) compartments ( Fig 8B ) , the total number of IRF-1-/- ( CD45 . 2 ) trended lower ( 3 . 5-fold lower for IFN-γ+ ( P = 0 . 09 ) and 4 . 7-fold lower for GrB+ cells ( P = 0 . 03 ) ) . Thus , CD8+ T cells lacking IRF-1 are at a competitive disadvantage compared to wild type cells within an IRF-1+/+ stromal environment . To determine whether an IRF-1 environment contributes to shaping the antigen-specific CD8+ T cell responses after WNV infection , we adoptively transferred wild type ( CD45 . 1 ) CD8+ T cells into IRF-1-/- ( CD45 . 2 ) or wild type ( CD45 . 2 ) mice , and compared antigen-specific responses of the donor and host CD8+ T cells ( Fig 8C ) . Remarkably , CD45 . 1 wild type CD8+ T cells in an IRF-1-/- mouse generated a larger WNV antigen-specific population compared to those observed in wild type mice ( Fig 8D and E ) . These values approached but did not attain those observed with unperturbed , WNV-infected IRF-1-/- mice . These observations suggest that IRF-1 expression outside of the T cell compartment also regulates the magnitude of the antigen-specific CD8+ T cell response during WNV infection . To further understand the rapid expansion of antigen-specific CD8+ T cells in IRF-1-/- mice , we evaluated the percentage and number of cells that were proliferating at the peak of the response , based upon the expression of Ki67 , a protein upregulated during the cell cycle . Ki67 expression was evaluated directly ex vivo , without peptide restimulation , in splenocytes of IRF-1-/- and wild type mice at 8 days after WNV infection . A significantly higher percentage ( 66 versus 10% P<0 . 008 ) and number ( 17×104 versus 1 . 8×104 cells , P<0 . 008 ) of WNV-specific CD8+ T cells in IRF-1-/- mice were proliferating at day 8 ( Fig 9A ) . To address why there was an increase in proliferation of antigen-specific CD8+ T cells in IRF-1-/- mice during WNV infection , CD8+ T cells were isolated ( ∼90% purity ) by negative selection , labeled with carboxyfluorescein diacetate , and stimulated with plate-bound anti-CD3ε and anti-CD28 in vitro . Notably , IRF-1-/- CD8+ T cells proliferated more rapidly than wild type cells at both 48 and 72 hours after stimulation ( Fig 9B ) . Collectively , these results confirm that IRF-1-/- CD8 T cells have an intrinsic capacity to proliferate more rapidly in the context of stimulation through the T cell receptor , suggesting that IRF-1 acts to regulate T cell proliferation . The accumulation of CD8+ T cells in the brain can be protective in the context of WNV infection depending on the numbers of migrating cells , the extent of neuronal infection , the timing of trafficking , and the inherent virulence of the viral strain [58] , [60] , [61] , [64] . To assess whether the increase in relative numbers of WNV-specific CD8+ T cells in the spleen of IRF-1-/- mice was also observed in the CNS , we evaluated leukocyte accumulation in the brain . Leukocytes were recovered from the brains of wild type and IRF-1-/- mice at day 8 post-infection after perfusion . Equivalent percentages and numbers of CD45high/CD11bhigh Mφ , CD45low/CD11bhigh activated microglia , and CD4+ T cells were observed in the brains of WNV-infected wild type and IRF-1-/- mice ( data not shown ) . In comparison , higher percentages ( >2-fold ) and total numbers ( ∼4-fold ) of Db-restricted NS4b tetramer+ CD3+ CD8+ T cells were detected in the brains of IRF-1-/- mice ( P<0 . 002 ) ( Fig 10A-C ) . Thus , a deficiency of IRF-1 enhanced accumulation of antigen-specific CD8+ T cells in the brains of WNV-infected mice . To determine the potential significance of the increased numbers of IRF-1-/- CD8+ T cells , we assessed their functional activity in vivo and ex vivo . When NS4b-peptide pulsed naïve splenocytes were adoptively transferred into WNV-infected wild type or IRF-1-/- mice , IRF-1-/- CD8+ T cells showed an equivalent if not slightly greater capacity to lyse target cells ( 98% versus 80% , P<0 . 03; Fig 10D ) . Importantly , NS4b-pulsed targets were not lysed when transferred into naïve wild type or IRF-1-/- mice . We next performed adoptive transfer studies with WNV-primed wild type or IRF-1-/- CD8+ T cells . To generate WNV-primed CD8+ T cells , wild type or IRF-1-/- mice were infected with WNV , and on day 7 , spleens were harvested and 2×105 NS4b tetramer+ CD8+ T cells were adoptively transferred into congenic five week-old wild type mice two days after WNV infection , and 4 days later ( day 6 after infection ) , brains were harvested to evaluate the effect on control of WNV infection . While addition of naïve CD8+ T cells did not affect viral burden in the brain [60] , wild type and IRF-1-/- CD8+ T cells both reduced viral titers ( Fig 10E , P<0 . 02 ) . To directly establish that IRF-1-/- CD8+ T cells could clear WNV from neurons , we used an ex vivo viral clearance assay with primary neurons derived from the cerebral cortex of wild type C57BL/6 mouse embryos [60] . One hour after infection , WNV-primed wild type or IRF-1-/- CD8+ T cells were added at an effector to target ( E∶T ) ratio of ten to one . At 48 hours after infection , the level of infectious WNV in the neuronal supernatants was measured by focus-forming assay . Addition of naïve CD8+ T cells does not reduce WNV infection in neurons [60] . In comparison WNV-primed CD8+ T cells significantly reduced ( ∼40 to 1400-fold , P<0 . 006 ) infectious virus production from infected cortical neurons , with an even greater ( 35-fold , P<0 . 02 ) inhibitory effect seen with IRF-1-/- compared to wild type CD8+ T cells ( Fig 10F ) . Collectively , these experiments suggest that WNV-primed IRF-1-/- CD8+ T cells are capable of controlling neuronal infection . However , this response is not sufficient to compensate for increased viral replication in peripheral tissues , which results in early and enhanced infection in the CNS of IRF-1-/- mice .
In this study , we identified IRF-1 as an essential regulator of the host immune response against WNV infection , and show that it governs processes of both the innate and adaptive immune responses that control outcome . IRF-1-/- mice were vulnerable to lethal infection with enhanced viremia , increased viral replication in peripheral tissues , altered tropism , and rapid dissemination into the CNS . Ex vivo analysis showed a cell-specific utilization of IRF-1 in controlling WNV replication . Although an absence of IRF-1 did not alter WNV infection in MEF , IRF-1-/- Mφ supported enhanced viral replication . Additionally , we identified a surprising phenotype with respect to the effects of IRF-1 on the induction of WNV-specific CD8+ T cells . Despite markedly fewer CD8+ T cells in naïve IRF-1-/- mice at baseline , we observed a rapid expansion in the periphery of antigen-specific CD8+ T cells that were IFN-γ+ and GrB+ , which was associated with increased accumulation in the brain . Although IRF-1-/- CD8+ T cells expanded rapidly and were capable of killing targets and efficiently clearing virus from infected neurons , they did not attain sufficient numbers quickly enough to mitigate the increased infection in the CNS caused by the compromised early innate control of WNV in peripheral tissues . Innate immunity and particularly type I ( IFN-αβ ) , II ( IFN-γ ) and III ( IFN-λ ) IFN responses orchestrate control of infection by many DNA and RNA viruses [65] , [66] , [67] . Analogously , IFN-αβR-/- [15] , IFN-γ-/- or IFN-γR-/- [68] , [69] , and IFN-λR-/- ( IL28A ) ( H . M . Lazear and M . Diamond , unpublished results ) mice all show enhanced susceptibility to WNV infection . The phenotype observed in the IRF-1-/- mice with WNV infection was in some ways similar to primary defects in IFN responses: increased replication in peripheral tissues at early time points that was associated with premature viral dissemination into the CNS . Because IRF-1 is important for efficient transduction of the antiviral signals downstream of IFN-γ [27] , the early virologic phenotype in WNV-infected IRF-1-/- mice might be expected to phenocopy IFN-γ-/- mice [68] , [69] . However , a comparison of the current data with published results revealed a more severe phenotype in IRF-1-/- mice , with higher and sustained infection in the spleen , altered tissue tropism with infection in the kidney , and higher titers in the brain and spinal cords between days 4 and 8 . Thus , IFN-γ-independent immune regulatory functions must explain the greater susceptibility of IRF-1-/- compared to IFN-γ-/- mice . IRF-1 is not required for induction of the IFN-αβ genes in MEF and Mφ in response to WNV infection , results that agree with experiments with NDV-infected fibroblasts [33] , [40] or reovirus-infected myocytes [70] . Instead , IRF-3 and IRF-7 appear as the primary IRF transcriptional activators downstream of RIG-I and MDA5 recognition and IPS-1 signaling for efficient IFN-αβ gene induction after WNV infection of MEF and Mφ [18] , [19] . The antiviral effect of IRF-1 was cell type-specific as enhanced WNV replication was observed preferentially in Mφ . An IRF-1-dependent antiviral signature is consistent with recent ectopic expression studies in human cells , which showed broad antiviral activity of IRF-1 against a range of viruses [25] . At present , we do not know the identity of the antiviral genes induced by IRF-1 that contribute directly to the virologic phenotype in IRF-1-/- Mφ . Comparative microarray analysis of WNV-infected wild type and IRF-1-/- MEF and Mφ is planned to define a set of uniquely induced genes that could function to limit WNV infection in a cell-type specific manner . Previous studies have reported deficits in T cell function in IRF-1-/- mice including low levels of CD8+ T cells in peripheral lymphoid tissue [39] , [40] , [41] . Based on this , we expected a blunted and dysfunctional CD8+ T cell response after WNV infection in IRF-1-/- mice; as antigen-specific CD8+ T cells contribute to clearance of WNV-infected neurons [58] , [59] , [60] , [61] , [62] , [71] , we anticipated uncontrolled viral replication in the brain prior to death of the animals . Although we confirmed lower levels of peripheral CD8+ T cells in naïve IRF-1-/- mice , surprisingly , we observed rapid proliferation of antigen-specific CD8+ T cells such that a high fraction ( ∼30% ) of CD8+ T cells in the spleen were directed against a single viral peptide epitope . Moreover , IRF-1-/- CD8+ T cells were fully capable of lysing target cells and clearing viral infection from neurons and the brain . This activity was consistent with a flattening of viral growth kinetics in the brain and spinal cord between day 6 and 8 in IRF-1-/- mice . In comparison , mice with targeted deletions in CD8+ T cells [58] or perforin [60] show logarithmic increases in CNS viral burden at this phase of infection . How does IRF-1 regulate CD8+ T cell responses after WNV infection ? Consistent with earlier studies , we observed a CD8+ T cell-intrinsic defect in proliferation and/or survival in vivo . This was most apparent in the increased ratio of wild type ( CD45 . 1 ) to IRF-1-/- ( CD45 . 2 ) CD8+ T cells within 12 hours of transfer to RAG1-/- mice . Additionally , by 8 days after infection , the ratio of total wild type ( CD45 . 1 ) to IRF-1-/- ( CD45 . 2 ) CD8+ T cells continued to increase . Given that the RAG1-/- mice are IRF-1 sufficient , this establishes a cell-intrinsic defect in CD8+ T cell expansion , and is consistent with bone marrow chimera reconstitution studies showing that IRF-1-/- thymocyte maturation was not restored in irradiated IRF-1+/+ mice [39] . This phenotype could be due to altered responsiveness to IL-12 , which has been observed in IRF-1-/- CD4+ T cells [43] , [44] , or possibly , other cytokines ( e . g . , IL-15 or IL-18 ) that regulate CD8+ T cell proliferation . Alternatively , IRF-1 signaling in antigen presenting or stromal cells may induce counter-regulatory networks or inhibit proliferation signals that result in tempered CD8+ T cell responses . Consistent with this possible function , we also observed blunted induction of CD4+CD25+FoxP3+ regulatory T cells in WNV-infected IRF-1-/- mice . A deficiency of regulatory T cells was previously shown to augment antigen-specific CD8+ T cell responses after WNV infection [63] , although not to the extent observed in IRF-1-/- mice . Within IRF-1-/- mice we observed an increase in the percentage and number of WNV-specific CD8+ T cells at the peak of infection due to increased proliferation . Our in vitro data supports this concept as IRF-1-/- CD8+ T cells proliferated to a greater extent than wild type cells after direct stimulation through the T cell receptor . Interestingly , adoptive transfer of wild type CD8+ T cells into an IRF-1-/- environment suggests that in vivo , a T cell extrinsic signal can drive the proliferation of antigen-specific cells . Historically , IRF-1 has been considered a negative regulator of cell proliferation , which in part explains its tumor suppressive activity ( reviewed in [37] ) . At present , it remains uncertain as to the identity of the T cell-extrinsic signal that enhances the proliferation of antigen-specific cells in IRF-1-/- mice , although differential cell-type specific induction of key cytokines ( e . g . , IL-10 or IL-12 ) was not observed ( Fig S2 ) . In summary , our experiments establish that IRF-1 has an essential function in the innate and adaptive immunity against WNV infection . In addition to its transduction of IFN-γ-dependent antiviral signals and functions in T cell lineage commitment in the thymus , IRF-1 appears to directly regulate antiviral genes in a cell-type specific and IFN-independent manner and be required for efficient expansion of regulatory T cells . Moreover , IRF-1 has both cell-intrinsic and -extrinsic functions in shaping CD8+ T cell responses , including stimulating both positive and negative regulatory networks depending on the cell type . Genetic profiling studies are planned with wild type and IRF-1-/- cells to identify novel gene signatures that can help define antiviral effector and immunomodulatory genes that inhibit viral infections and shape adaptive T cell responses .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee at the Washington University School of Medicine ( Assurance Number: A3381-01 ) . All inoculation and experimental manipulation was performed under anesthesia that was induced and maintained with ketamine hydrochloride and xylazine , and all efforts were made to minimize suffering . The WNV strain ( 3000 . 0259 ) was isolated in New York in 2000 [72] and passaged once in C6/36 Aedes albopictus cells to generate a stock virus that was used in all experiments . Wild type and congenic RAG1-/- C57BL/6 mice were obtained commercially ( Jackson Laboratories ) . C57BL/6 . SJL-Ptprca/BoyAiTac mice were purchased ( Taconic ) and are congenic with respect to C57BL/6 mice except at the Ly5 . 1 ( CD45 . 1 ) locus . IRF-1-/- mice were originally generated by T . Taniguchi [3] , [34] , [73] and obtained on a C57BL/6 background ( kindly provided by T . Taniguchi and K . Fitzgerald ) . All mice were genotyped and bred in the animal facilities of the Washington University School of Medicine under pathogen free conditions , and experiments were performed in strict compliance with Washington University Animal Studies guidelines . Eight to twelve week old mice were used for all in vivo studies . For peripheral infection , 102 PFU of WNV was diluted in Hanks balanced salt solution ( HBSS ) supplemented with 1% heat-inactivated fetal bovine serum ( FBS ) and inoculated by footpad injection in a volume of 50 µl . To monitor viral spread in vivo , mice were infected with 102 PFU of WNV by footpad inoculation and sacrificed at days 1 , 2 , 4 , 6 and 8 after inoculation . After cardiac perfusion with PBS , organs were harvested , weighed , homogenized and virus was titrated by standard plaque assay as described [57] . Viremia was measured by analyzing WNV RNA levels using fluorogenic quantitative RT-PCR ( qRT-PCR ) as described [15] . Total RNA was isolated from lymph nodes or primary cells by using the RNeasy kit according to the manufacturer's instructions ( Qiagen ) . During the isolation , to remove any contaminating DNA , samples were treated with RNAse-free DNAse ( Qiagen ) . IFN-α and β mRNA were amplified and quantified from total RNA by qRT-PCR as previously described [47] . The following primers and probes were used to amplify murine IFN-α and IFN-β mRNA: IFN-α , forward primer , 5′-CTTCCACAGGATCACTGTGTACCT-3′ , reverse primer , 5′TTCTGCTCTGACCACCTCCC3′ , probe , 5′-FAM-AGAGAGAAGAAACACAGCCCCTGTGCC-TAMRA-3′; IFN-β , forward primer , 5′-CTGGAGCAGCTGAATGGAAAG-3′ , reverse primer , 5′-CTTCTCCGTCATCTCCATAGGG-3′ , probe 5′-FAM-CAACCTCACCTACAGGGCGGACTTCAAG-TAMRA-3′ . To analyze the relative fold induction of IFN-α and IFN-β mRNA , 18S rRNA expression levels were also determined for normalization by using the Ct method as described [47] . Levels of biologically active IFN in serum were measured using an EMCV L929 cytopathic effect bioassay as described previously [6] . Results were compared with a standard curve using recombinant mouse IFN-α ( PBL Laboratories ) . For IFN inhibition assays , wild type or IRF-1-/- Mφ were pretreated with increasing doses of IFN-β or IFN-γ ( PBL Laboratories ) for 24 hours and then infected with WNV at an MOI of 0 . 1 . Supernatants were harvested 48 hours after infection and titered by plaque assay . For the IFN-γ neutralization assay , various concentrations of exogenous IFN-γ were mixed with 20 µg of a hamster anti-mouse IFN-γ neutralizing MAb ( H22 ) [55] or isotype control for 1 hour at 37°C . Alternatively , wild type Mφ were pretreated with the IFN-γ with or without MAbs mixture for 24 hours and then infected with WNV at an MOI of 0 . 1 . Supernatants were harvested 48 hours post infection and titered by plaque assay . To test whether H22 ( and neutralization of IFN-γ ) affected viral replication , multi-step growth curves with wild type or IRF-1-/- Mφ were performed in the presence of 20 µg of H22 antibody or isotype control after infection at an MOI of 0 . 01 . The levels of WNV-specific IgM and IgG were determined using an ELISA against purified WNV E protein [74] . The focus reduction neutralization assay was performed as described previously [75] . Intracellular IFN-γ or TNF-α staining was performed on splenocytes using a previously identified Db-restricted NS4B peptide or Kb-restricted E peptide in a re-stimulation assay with 1 µM of peptide and 5 µg/ml of brefeldin A ( Sigma ) as described [71] . The number and percentage of CD4+CD25+FoxP3+ regulatory T cells in the spleen of naïve or WNV-infected wild type or IRF-1-/- mice was measured using the regulatory T cell staining kit ( Ebioscience ) . Granzyme B ( Invitrogen , clone gb12 ) intracellular staining was performed without stimulation using the staining solutions for FoxP3 . Intranuclear staining for Ki67 antigen ( BD Pharmingen , clone B56 ) was performed on CD8+ T cells ex vivo without additional stimulation using the FoxP3 staining solutions . In some experiments WNV-specific CD8+ T cells were stained with a Db-restricted NS4B peptide tetramer ( NIH Tetramer Core Facility , Emory University , Atlanta , GA ) . Samples were processed by multi-color flow cytometry on an LSR or FACSCalibur flow cytometer ( Becton Dickinson ) and analyzed with FlowJo software ( Treestar ) . Equal numbers of naïve donor splenic CD8+ T cells from wild type ( CD45 . 1 ) or IRF-1-/- ( CD45 . 2 ) mice were harvested , purified by negative selection using antibody-coated magnetic beads ( anti-NK1 . 1 , anti-B220 , anti-CD4 , and anti-MHC class II; Miltenyi Biotec ) , and transferred ( 2×106 total cells ) into recipient RAG1-/- mice . One day later mice were bled to confirm the reconstitution , and then infected with WNV ( 102 PFU via a subcutaneous route ) . Eight days later , splenocytes were harvested , and analyzed for activation and cytokine expression as described above . A similar procedure was followed for the adoptive transfer of CD8+ T cells ( 2×106 total cells ) from wild type CD45 . 1 SJL donors into recipient wild type ( CD45 . 2 ) or IRF-1-/- ( CD45 . 2 ) hosts . Purity of transferred T cell populations , which ranged from 88 to 92% CD8b+ , was evaluated by flow cytometry . To evaluate the ability of activated CD8+ T cells to reduce viral burden , wild type and IRF-1-/- CD8+ T cells were isolated using antibody-coated magnetic beads at day 7 after infection and 2×105 NS4b tetramer+ cells were adoptively transferred by intravenous route 48 hours after subcutaneous WNV ( 102 PFU ) infection of 5 week-old wild type mice . Mice were euthanized at day 6 after infection for analysis of viral burden in the brain by focus forming assay . Naïve splenic CD8a+ T cells were purified from wild type and IRF-1-/- mice by positive selection using antibody-coated magnetic beads ( anti-CD8a , Miltenyi Biotec ) . After purification , CD8+ T cells were washed twice with PBS and incubated with 1 µM carboxyfluorescein diacetate succinimidyl ester in PBS for 10 minutes at 37°C . The reaction was quenched by adding an equal volume of RPMI 1640 supplemented with 10% FBS , 0 . 1 mM β-mercaptoethanol , penicillin , and streptomycin . After washing , purified labeled CD8+ T cells ( 106 ) were plated in individual flat bottom 96 well tissue culture treated plates in the RPMI medium described above , in the presence or absence of 1 µM anti-CD3ε ( Biolegend , clone 2c11 NA/LE ) and 10 µM anti-CD28 ( BD Biosciences , clone 37 . 51 ) . Cells were harvested at 24 , 48 and 72 hours , washed twice with PBS , and analyzed by flow cytometry . Quantification of infiltrating CNS lymphocytes was performed as previously described [76] . Briefly , wild type and IRF-1-/- mouse brains were harvested on day 8 after infection , dispersed into single cell suspension with a cell strainer and digested with 0 . 05% collagenase D , 0 . 1 µg/ml trypsin inhibitor TLCK , 10 µg/ml DNase I and 10 mM of HEPES ( Life Technologies ) in HBSS for 1 hour . Cells were separated by discontinuous Percoll-gradient ( 70/37/30% ) centrifugation for 30 min ( 850 x g at 4°C ) . Cells were then counted and stained for CD4 , CD8 , CD45 and CD11b with directly conjugated antibodies ( BD Pharmingen ) for 30 minutes at 4°C , and then fixed with 1% paraformaldehyde . WNV-specific CD8+ T cells were identified using a Db-restricted NS4B peptide tetramer . In vivo killing of target cells was performed as previously described [77] . Briefly , splenocytes from B6 . SJL ( CD45 . 1 ) mice were isolated . Half of the cells were labeled with carboxyfluorescein diacetate succinimidyl ester ( CFDA ) at 500 nM and the remainder was labeled with 5 nM CFDA . After labeling , cells labeled with 500 nM CFDA were pulsed for one hour at 37°C with 1 µM NS4B 2488-2496 peptide , whereas the 5 nM CFDA cells were not pulsed with peptide . Both sets of cells were counted and equal numbers were mixed and injected intravenously ( 107 cells total per mouse ) into recipient WNV-infected ( at day 8 after infection ) wild type or IRF-1-/- mice . After 8 hours , the mice were sacrificed and splenocytes were gated on CD45 . 1 cells ( donor cells ) . The percent killing of target cells was calculated: ( 1 – ( ratio immune/ratio naive ) ) ×100 . Ratio equals the number of NS4B peptide-coated targets/number of reference targets [77] . Purified CD8+ T cells were incubated with WNV-infected cortical neurons as described previously [6] . Briefly , cortical neurons were isolated and plated onto 24 well plates at 3×105 neurons per well . Four to five days after plating , neurons were infected with WNV at an MOI of 0 . 001 for one hour . Neurons were washed thrice with warm media and 2×106 ( effector∶target ratio of 10∶1 ) CD8+ T cells , isolated by positive selection from wild type or IRF-1-/- splenocytes at eight days post infection , were added . Individual wells were harvested at 48 hours after infection , and WNV titers were measured by focus-forming assay . For in vitro experiments , an unpaired T-test was used to determine statistically significant differences . For viral burden and T cell analysis , differences were analyzed by the Mann-Whitney test . Kaplan-Meier survival curves were analyzed by the log rank test . All data were analyzed using Prism software ( GraphPad Prism4 , San Diego , CA ) . Additional materials and methods are provided ( Text S1 ) . | Interferon regulatory factor ( IRF ) -1 is a transcription factor that has been implicated in immune regulation and induction of type I IFN gene expression . To better understand the contribution of IRF-1 to antiviral immunity , we infected cells and mice lacking IRF-1 with West Nile virus ( WNV ) , an encephalitic flavivirus . IRF-1-/- mice were uniformly vulnerable to WNV infection with enhanced viral replication and rapid dissemination into the brain and spinal cord . Studies in cell culture revealed a cell-type specific antiviral role as IRF-1-/- macrophages but not fibroblasts supported enhanced WNV replication . IRF-1 also had an independent effect on CD8+ T cell responses . Although fewer CD8+ T cells were observed in naïve animals , WNV-specific CD8+ T cells rapidly expanded in IRF-1-/- mice and retained the capacity to clear infection . Collectively , our studies define independent roles for IRF-1 in restricting WNV pathogenesis and modulating the protective CD8+ T cell response . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"viral",
"classification",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"adaptive",
"immunity",
"rna",
"viruses",
"emerging",
"infectious",
"diseases",
"animal",
"models",
"of",
"infection",
"biology",
"pathogenesis",
"viral",
"replication",
"immune",
"response",
"immunity",
"virology",
"innate",
"immunity"
]
| 2011 | Interferon Regulatory Factor-1 (IRF-1) Shapes Both Innate and CD8+ T Cell Immune Responses against West Nile Virus Infection |
An essential component of genome function is the syntax of genomic regulatory elements that determine how diverse transcription factors interact to orchestrate a program of regulatory control . A precise characterization of in vivo spacing constraints between key transcription factors would reveal key aspects of this genomic regulatory language . To discover novel transcription factor spatial binding constraints in vivo , we developed a new integrative computational method , genome wide event finding and motif discovery ( GEM ) . GEM resolves ChIP data into explanatory motifs and binding events at high spatial resolution by linking binding event discovery and motif discovery with positional priors in the context of a generative probabilistic model of ChIP data and genome sequence . GEM analysis of 63 transcription factors in 214 ENCODE human ChIP-Seq experiments recovers more known factor motifs than other contemporary methods , and discovers six new motifs for factors with unknown binding specificity . GEM's adaptive learning of binding-event read distributions allows it to further improve upon previous methods for processing ChIP-Seq and ChIP-exo data to yield unsurpassed spatial resolution and discovery of closely spaced binding events of the same factor . In a systematic analysis of in vivo sequence-specific transcription factor binding using GEM , we have found hundreds of spatial binding constraints between factors . GEM found 37 examples of factor binding constraints in mouse ES cells , including strong distance-specific constraints between Klf4 and other key regulatory factors . In human ENCODE data , GEM found 390 examples of spatially constrained pair-wise binding , including such novel pairs as c-Fos:c-Jun/USF1 , CTCF/Egr1 , and HNF4A/FOXA1 . The discovery of new factor-factor spatial constraints in ChIP data is significant because it proposes testable models for regulatory factor interactions that will help elucidate genome function and the implementation of combinatorial control .
Genomic sequences facilitate both cooperative and competitive regulatory factor-factor interactions that implement cellular transcriptional regulatory logic . The functional syntax of DNA motifs in regulatory elements is thus an essential component of cellular regulatory control . Appropriately spaced motifs can facilitate cooperative homo-dimeric or hetero-dimeric factor binding , while overlapping motifs can implement competitive binding by steric hindrance . Cooperative and competitive binding are an integral part of complex cellular regulatory logic functions [1] , [2] . The binding of regulatory proteins to the genome cannot at present be predicted from primary DNA sequence alone as chromatin structure , co-factors , and other mechanisms make the prediction of in vivo binding from sequence empirically unreliable [3] . Thus it is not possible to use primary DNA sequence to determine the aspects of genome syntax that are employed in vivo . To discover novel pair-wise factor spatial binding constraints in vivo , we have developed a new method called GEM that simultaneously resolves the location of protein-DNA interactions and discovers explanatory DNA sequence motifs with an integrated model of ChIP-Seq or ChIP-exo reads and proximal DNA sequences . We define a binding event location as the single base position at the center of a protein-DNA interaction . GEM reciprocally improves motif detection using binding event locations , and binding event predictions using discovered motifs . In doing so , GEM offers a more principled approach than simply snapping binding event predictions to the closest instance of the motif , and indeed , GEM does not require that all binding events are associated with strong motifs . GEM offers both improved spatial accuracy of binding event predictions and improved motif discovery in ChIP-Seq and ChIP-exo datasets . GEM's unbiased computational approach has enabled us to discover novel binding constraints between transcription factors from sequenced ChIP experiments . These spatial constraints directly suggest biological regulatory mechanisms that will be useful in future studies . Other methods to resolve binding events in sequenced ChIP data identify statistically enriched regions of ChIP-Seq read density and the peak points of enrichment within those regions [4]–[9] , and binding calls can be offset from the bound site by dozens of bases [10] . Recent studies have integrated peak detection and motif discovery by including motif occurrences to score the significance of predicted binding events [11] , [12] , or by using ChIP-Seq read coverage as a positional prior to improve motif discovery [13] , [14] . However , no study has yet used the motif position information to reciprocally improve the spatial accuracy of binding event prediction . SpaMo studied the motif spacing using ChIP-Seq events to infer transcription factor complexes but the predicted motif spacing does not necessarily indicate in vivo binding in the specific cellular conditions [15] . Here we review our GEM derived results , discuss these results in the context of current data production projects , and detail our methods .
We compared GEM's spatial resolution to six well known ChIP-Seq analysis methods , including GPS [8] , SISSRs [6] , MACS [4] , cisGenome [7] , QuEST [5] and PeakRanger [9] . We used a human Growth Associated Binding Protein ( GABP ) ChIP-Seq dataset for our evaluation because GABP ChIP-Seq data were previously reported to contain homotypic events where the reads generated by multiple closely spaced binding events overlap [5] . Thus the GABP dataset offers the opportunity to test if integrating motif information and binding event prediction improves our ability to deconvolve closely spaced binding events with greater accuracy . We also evaluated the methods using ChIP-Seq data from the insulator binding factor CTCF ( CCCTC-binding factor ) [16] , as it binds to a stronger motif than GABP . These two factors are representative of relatively easy ( CTCF ) and difficult ( GABP ) cases for ChIP-Seq data analysis . They are also used by other studies as benchmarks allowing for the direct evaluation of our results . GEM performance on other factors may vary . We found that GEM has the best spatial resolution among tested methods . Spatial resolution is the average absolute value difference between the computationally predicted locations of binding events and the nearest match to a proximal consensus motif . From all observations , spatial resolution is corrected for a fixed offset by subtracting the mean difference before averaging the absolute value differences . To ensure a fair comparison , we used 428 shared GABP binding sites that are predicted by all seven tested methods and which contain an instance of the GABP motif within 100 bp . GEM exactly locates the events at the motif position in 56 . 5% of these events ( Figure 1A ) . For a dataset with a stronger consensus motif , ChIP-Seq data from CTCF , GEM exactly locates the events at the motif position in more than 90% of the shared events , significantly improving the spatial accuracy of predicted binding events over other methods ( Figure 1B ) . Alternative evaluations with all the binding sites that have a motif at a distance less than 100 bp are also performed for both GABP and CTCF data , and the results ( Figure S1 ) are similar to those above . Thus , GEM's joint model of ChIP-Seq read coverage and sequence is able to more accurately predict the location of binding sites than other approaches , which do not use motif information in their binding event predictions . GEM is also better at resolving closely spaced binding events [17] in the GABP data than the other methods we tested . For example , GEM uniquely detects two GABP events over proximal GABP motifs that are 32 bp apart on chromosome 2 ( Figure 1C ) . To evaluate binding deconvolution on a genome-wide scale , we identified 477 candidate clusters of closely spaced binding events . Each candidate cluster was detected as bound by all seven tested methods and contained two or more proximal GABP motifs separated by less than 500 bp . GEM identified two or more closely spaced events in 144 of the candidate clusters , significantly more than GPS ( 108 ) , SISSRs ( 77 ) , QuEST ( 77 ) , PeakRanger ( 36 ) , MACS ( 4 ) and cisGenome ( 5 ) ( Figure 1D ) . We tested GEM's ability to discover biologically relevant DNA-binding motifs in data from the ENCODE project [18] . We chose this large collection of experiments because we expected they would be representative of the typical range of ChIP-Seq data noise and sequencing depth . Noise can be caused by low antibody affinity and deviations from ideal experimental procedure . We used a set of 214 ChIP-Seq experiments and associated controls comprising 63 distinct transcription factors that were profiled in one or more cell lines by the ENCODE project and for which validated DNA-binding motifs exist in public databases ( Dataset S1 ) . GEM analyzed these ChIP-Seq data , and the most significant GEM-discovered motifs from each analysis ( Table S1 and Dataset S2 ) were compared to corresponding known binding preferences of the same transcription factors using STAMP [19] . A motif alignment with E-value less than 1e-5 was considered a match . For comparison , we also used four popular traditional motif discovery tools covering a range of computational techniques , including MEME [20] , Weeder [21] , MDScan [22] , and AlignACE [23] , and three ChIP-Seq oriented tools , POSMO [24] , HMS [13] and ChIPMunk [14] on the same data . A set of 100 bp sequences extracted from the 500 most highly ChIP-enriched GPS peaks calls are examined by the motif-finders MEME , Weeder , MDScan , AlignACE , or POSMO . For HMS and ChIPMunk , a set of 100 bp sequences and corresponding read coverage profiles are extracted from the 500 most highly ChIP-enriched GPS peaks calls . We found GEM outperforms all of the compared motif discovery approaches , even when allowing each method to make multiple motif predictions ( Figure 2 , Table S2 , S3 ) . Therefore , the GEM approach to integrating ChIP-Seq event detection with motif analysis not only improves the spatial resolution of binding events , but also more accurately finds the expected binding motifs present at those events . We note that GEM sometimes failed to find the known motif in datasets where one of the other algorithms succeeds . The complete evaluation is in Table S2 , S3 . We then tested GEM on ENCODE ChIP-Seq experiments for 9 distinct transcription factors with no publically described DNA binding motif . For 6 of these transcription factors , GEM discovers novel motifs that are consistent with expected binding sequences based on a small number of binding sites characterized in the literature , or similarity to the known binding preferences of related proteins ( Table S4 ) . For example , GEM confirms that BATF has a similar binding preference to other members of the AP1 family of transcription factors . The similar TGAC/G binding preference has previously been supported by EMSA assays on regions upstream potential BATF regulated genes [25] . ChIP-exo aims to improve transcription factor binding spatial resolution by extensively digesting ChIP fragments down to the DNA that is protected by the bound protein complex [26] . While ChIP-exo experiments provide high-resolution binding information , typical peak-finding methodologies may fail to achieve single-base resolution binding event predictions if they do not account for the properties of the ChIP-exo experiment . An example is provided by the published CTCF ChIP-exo experiment [26] , where ChIP-exo reads are bimodally distributed around binding sites on both strands because CTCF is cross-linked at two distinct sites of DNA . The published event predictions did not account for this characteristic distribution , and are thus often offset from CTCF binding motif instances . Since GPS and GEM automatically learn a model of sequence reads around binding events , GPS and GEM may be directly applied to ChIP-exo data without modification . We first verified that GEM's model of binding events is able to automatically adapt to the read distribution produced by the ChIP-exo protocol . We compared GEM's final computed read distribution to the expected empirical distribution of ChIP-exo and found that they were consistent ( Figure 3B and Figure S2 ) . GEM improves upon the spatial resolution of binding event detection over other methods for ChIP-exo data ( Figure 3A ) . To investigate the performance of GEM on ChIP-exo data , we compared the binding event predictions of GEM and GPS on ChIP-exo CTCF binding and the “middle of peak-pair” method from the original ChIP-exo study [26] . To ensure a fair comparison , we used 5074 shared binding sites that are predicted by all tested methods and that contain a strong CTCF motif match within 100 bp of the binding positions . The original ChIP-exo study [26] had 5 . 4% of the binding event calls centered on the motif match position , 40 . 3% of the calls within 10 bp , and an average spatial resolution of 15 . 85±15 . 29 bp . Applying GPS to the ChIP-exo data improved the spatial resolution , with 8 . 8% calls at 0 bp positions , 59 . 7% of calls within 10 bp , and average spatial resolution of 10 . 38±11 . 26 bp . Applying GEM to the ChIP-exo data located 76 . 5% calls exactly at the motif match positions , 89 . 7% of calls within 10 bp , and an average spatial resolution of 3 . 35±9 . 71 bp . These results demonstrate that GEM can significantly improve the spatial accuracy of ChIP-exo binding event predictions . We examined if GEM could detect pairs of transcription factors that bind to the genome with characteristic pair-wise spacing , beginning with the well-known hetero-dimeric pair Sox2-Oct4 [27] . In general , distance-constrained transcription factor binding cannot be predicted based solely on sequence motifs as motif presence does not guarantee binding . Such spatial binding constraints may be caused by combinatorial binding , alternative binding , binding that is orchestrated by multimeric protein complexes , or the spread of constrained enhancer syntax . We were able to discover Sox2-Oct4 transcription factor spatial binding constraints by combining GEM binding calls from Sox2 and Oct4 ChIP-Seq data . We applied GEM independently to mouse ES cell Sox2 and Oct4 ChIP-Seq data [15] to call the respective binding sites , and then computed the distance between Oct4 sites from Sox2 sites within a 201 bp window . The sequence strand of the GEM binding predictions is oriented using the Sox2 motif when a match to the motif is present . As expected , GEM predicted Oct4 binding sites are predominantly ( 630 sites out of 2525 in the 201 bp window ) located at −6 bp position relative to GEM predicted Sox2 sites ( Figure 4A and Figure S3 ) . However , this spacing cannot be observed from the binding calls of GPS or other event discovery methods alone because of their more limited spatial accuracy ( Figure 4B ) . An alternative approach is to snap binding calls to the nearest instance of the transcription factor's binding motif . We tested this approach using GPS binding calls as the starting points and found that the alternate approach captures fewer ( 277 sites out of 2753 ) instances of Oct4-Sox2 spatial binding constraints ( Figure 4C ) , presumably because some of the bound motifs do not pass the motif scoring threshold or because some unbound motif instances are located closer to the binding calls than the true motif instances . We next studied pair-wise binding relationships between 14 sequence-specific transcription factors ( Oct4 , Sox2 , Nanog , Klf4 , STAT3 , Smad1 , Zfx , c-Myc , n-Myc , Esrrb , Nr5a2 , Tcfcp2l1 , E2f1 and CTCF ) and two transcriptional regulators ( p300 and Suz12 ) in mouse ES cells by applying GEM to a large compendium of ChIP-Seq binding data [16] , [28] . Binding prediction by GEM enables the detection of 37 pairs of statistically significant spatial binding constraints , involving Oct4 , Sox2 , Nanog , Klf4 , Esrrb , Nr5a2 , Tcfcp2I1 , E2f1 , c-Myc , n-Myc and Zfx ( Figure S4 , the full list of TF pairs are in Table S6 , S7 , motifs are in Table S5 and Dataset S3 ) . Interestingly , we found that Klf4 , one of the ES cell reprogramming factors , exhibits strong distance-specific binding with many other factors , including Nanog , Sox2 , Zfx , c-Myc , n-Myc , E2f1 , Esrrb , Nr5a2 and Tcfcp2l1 ( Figure S5 ) . The discovered pair-wise spatial binding constraints reveal complex relationships among the factors . For example , Klf4 exhibits constrained binding with Sox2 but much less significantly with Oct4 ( Figure S5 ) . However , we did observe strong distance-specific binding between Oct4-Sox2 ( Figure 4A ) . This raises the question of whether the detected Klf4-Sox2 and Oct4-Sox2 spatial binding constraints are on the same genomic regions . We therefore studied all Sox2 bound regions that are co-bound with Klf4 . Out of a total of 5609 Sox2 bound regions with a Sox2 motif instance that can be oriented , 123 regions are co-bound by Klf4 at position +25 bp ( Figure 5A ) . However , only four regions show co-binding of Klf4 at position +25 bp and Oct4 at position −6 bp . More surprisingly , the distance-constrained Sox2/Klf4 regions are co-bound by 6 ES cell factors within a 70 bp window , including Sox2 ( at 0 bp ) , Nanog ( at 1 bp ) , Klf4 ( at 25 bp ) , Esrrb ( at 56 , 59 bp ) , Nr5a2 ( at 55 , 58 , 61 bp ) and Tcfcp2I1 ( at 66 , 69 bp ) . Inspecting the underlying sequences of these regions , we found that the binding motifs of these factors are embedded at the positions consistent with the binding positions ( Figure 5B ) . In addition to the consistent spatial arrangement of motifs , these sequences ( spanning from −70 bp to 100 bp ) exhibit a high degree of similarity . A subset of the sequences is shifted 3 bases by some insertion/deletions , consistent with the 3 bp shift of some of the factor binding positions . Comparing with p300 and H3K27ac ChIP-Seq datasets [29] , we found that almost all ( 119 out of 123 ) of these regions are bound by p300 , a histone acetyltransferase and transcriptional coactivator that predicts tissue-specific enhancers [30]; the majority of these regions are also marked by H3K27ac , a histone modification associated with active enhancers [29] , suggesting that they may be active enhancer regions ( Figure S6 ) . These results demonstrated that GEM analysis enables detection of coordinated binding of multiple factors that are driven at least partly by the underlying sequences . Of the 123 regions where Sox2 , Klf4 , and other sites display constrained spacing , 109 ( 89% ) are annotated instances of the RLTR9 ERVK family of long terminal repeat elements . It is interesting to note that while Bourque , et al . found an association between Oct4/Sox2 co-binding sites and other members of the ERVK repeat class [31] , we find a set of repetitive elements that encode the binding of Sox2 and other factors without Oct4 in ES cells . Kunarso , et al . suggested that transposable elements have rewired the core regulatory network of ES cells [32] . Our analysis found that the repetitive sequences constrain the in vivo binding of a number of key transcription factors in ES cells . We computed statistically significant pair-wise spatially constrained binding events between 46 transcription factors characterized in 184 ENCODE ChIP-Seq data sets in five different cell lines . Each transcription factor ChIP was processed independently by GEM so that we could assess any differences in observed binding between cell lines and biological replicates . We found that 390 pairs of transcription factors have significant binding distance constraints within 100 bp of each other ( Figure 6–7 , Figure S7 , S8 , S9 , S10 , the full list of TF pairs are in Table S8 , S9 ) . The number of pairs found in each cell line differed as did the number of transcription factors assayed: K562 ( 152 pairs/37 TFs ) , GM12878 ( 148 pairs/29 TFs ) , HepG2 ( 107 pairs/29 TFs ) , HeLa-S3 ( 48 pairs/15 TFs ) , and H1 ( 23 pairs/11 TFs ) . Certain factor-pairs exhibited a highly significant single binding spacing offset within 100 bp , such as the 4 bp distance between Egr1 and CTCF in K562 cells ( Figure 6 ) . Other factor pairs exhibited a large number of significant offsets , such as the 167 significant spacings between JunD and Max with the most significant being at 4 bp ( Figure 6–7 ) . Our analysis confirmed known interaction pairs MYC-MAX [33] , the FOS-JUN heterodimer [34] , and CTCF-YY1 [35] ( Table S8 , S9 ) . Observed novel genome wide spatial binding constraints include c-Fos:c-Jun/USF1 , CTCF/Egr1 , HNF4α/FOXA1 . We find that USF1 often binds 4 bp from c-Fos:c-Jun ( Figure 8A and Figure S11 ) . This binding is consistent with Fra1's facilitation of a complex between USF1 and c-Fos:c-Jun [36] . We find a significant number of cases where CTCF co-binds 4 bp from Egr1 ( Figure 8B and Figure S12 ) . Egr1 promotes terminal myeloid differentiation in the presence of deregulated c-Myc expression , and Egr1 has been implicated in down regulating c-Myc in conjunction with CTCF [37] . In addition , the co-binding of CTCF and Egr1 at the EPO regulatory region has been suggested [38] . FOXA1 binds at a large number of significant positions close to HNF4α ( total 4215 regions with a spacing within 30 bp , Figure 8C and Figure S13 ) , and there are also significant binding constraints between HNF4α and HNF4γ and FOXA1 , FOXA2 in HepG2 cells ( Table S8 , S9 ) . While co-binding of HNF4α/FOXA2 has been reported [39] , co-binding of HNF4α/FOXA1 , HNF4γ/FOXA1 and HNF4γ/FOXA2 are not known . We note that HNF4α and any one of FOXA1 , FOXA2 , or FOXA3 is sufficient to reprogram cells towards a hepatocytic fate [40] .
Collectively , our results demonstrate that it is now possible to reveal aspects of functional genome syntax by surveying in vivo binding relationships between transcription factors at high spatial resolution . Our analysis has been made possible by sequenced ChIP data and a new computational method , GEM , which provides exceptional spatial resolution . GEM makes binding predictions and observes spatial constraints by discovering significant events utilizing both motifs and observed read coverage information . Prior work has documented specific genomic regions extensively targeted by multiple transcription factors ( TFs ) [16] . However , we have shown that the functional syntax of DNA motifs in regulatory elements cannot be fully elaborated with the imprecise ChIP-Seq event calls provided by previous methods . Motif analysis approaches such as SpaMo discover enriched motif spacing by scanning a list of known motifs in sequences anchored by ChIP-Seq data of a single factor [15] . Since the existence of motif instances does not guarantee condition specific in vivo binding , SpaMo cannot confidently determine the spacing between binding events and the factors involved , especially for motifs that are shared by a family of TFs . Furthermore , SpaMo excludes repetitive sequences [15] . In contrast , GEM predicts binding based on uniquely-mapped reads and is able to detect spatial binding constraints in transposable elements . Such elements have been implicated in rewiring the core regulatory network of human and mouse ES cells [32] . We expect that the genome grammatical rules that are suggested here will be examined in further studies to elucidate mechanisms of transcriptional control , and potential protein-protein interactions that have regulatory consequences . Exploration of other genome grammatical constructs can be accomplished with the use of further ChIP experiments and GEM .
Initial protein-DNA binding event locations are predicted by GPS [8] , which employs a negative Dirichlet sparse prior . GEM discovers a set of enriched k-mers by comparing k-mer frequencies between positive sequences and negative control sequences . The positive set consists of 61 bp sequences centered on the predicted binding locations from Phase 1 , and a negative set consists of 61 bp sequences that are 300 bp away from binding locations and that don't overlap positive sequences . We count the number of positive and negative sequences that contain instances of each possible k-mer ( hit count ) , treating each k-mer and its reverse complement as the same sequence . A k-mer is considered enriched if the hypergeometric p-value [41] of its enrichment is less than 0 . 001 and it has at least 3-fold enrichment in terms of positive/negative hit count . In this study , values of k from 5 to 13 are used on each dataset , and the final k value is chosen as the one that gives the most significantly enriched primary PWM as described below . Each k-mer carries with it its expected offset from a binding event as averaged over the positive set . GEM next clusters the enriched k-mers into equivalence classes that describe similar DNA binding preferences ( Figure S14 ) . Each equivalence class is a collection of k-mers . A genomic sequence is said to match a k-mer equivalence class if the genomic sequence contains any of its component k-mers . GEM clusters enriched k-mers into k-mer equivalence classes by ( Figure S14 ) : After finding the primary k-mer equivalence class , GEM searches for other classes . To accomplish this , the previous seed k-mer is removed from the enriched k-mer pool and PWM motif occurrences are masked in the sequences . The process of building new k-mer equivalence classes is repeated until no more significantly enriched PWMs can be constructed . Rarely , a secondary motif PWM can become more significantly enriched than the primary motif . If this happens , the motif finding process is restarted using the seed k-mer of this secondary motif . Phase 4 of GEM uses the primary k-mer equivalence class to compute a Dirichlet prior that will be used for binding event discovery in Phase 5 . The genome is segmented into independent separable regions ( typically a few kb long ) by dividing at read gaps that are larger than 500 bp and further excluding regions that contain fewer than 6 reads [8] . At each evaluated genome region , we simultaneously search the occurrences of all the k-mers of the primary k-mer equivalence class using the Aho-Corasick algorithm [43] , and matches are marked at the expected binding event location for every matching k-mer . The position-specific prior for a sequence base is defined as the number of positive set sequences that contain one of the enriched k-mers whose binding offsets match that base . The concept of using informative positional priors for motif discovery has been explored previously [44] , [45] . GEM employs a generative mixture model that describes the likelihood of a set of ChIP-Seq reads being generated from a set of protein-DNA interaction events originating at specific DNA sequences . The model generates protein-DNA interaction events that are biased to occur at explanatory DNA sequences by a k-mer based positional prior . Each event then independently generates reads following an empirical read spatial distribution that describes the probability of reads given the distance from the event [8] ( see Figure 3B for an example ) . Formally , in an evaluated region of length M , we consider N ChIP-Seq reads that have been mapped to genome locations R = {r1 , … , rN} and M all possible protein-DNA interaction events at single base locations B = {b1 , … , bM} . We represent the latent assignments of reads to events that caused them as Z = {z1 , … , zN} , where indicator function 1 ( zn = m ) = 1 when read n is caused by the event m . The probability of a read n is based on a mixture of possible binding events: where M is the number of possible events; π denotes the parameter vector of mixing probabilities , and πm is the probability of event m; p ( rn | m ) is the probability of read n being generated from event m and can be determined from the empirical spatial distribution of reads given the event [8] . The overall likelihood of the observed set of reads is: We make two prior assumptions about the binding events: 1 ) binding events prefer to occur at the sequence specific DNA motif positions; 2 ) binding events are relatively sparse throughout the genome . To incorporate these assumptions , we place a negative Dirichlet prior [8] , [46] p ( π ) on binding event probabilities π: where αs is the uniform sparse prior parameter governing the degree of sparseness , αs>0; αm denotes the binding event specific prior parameter and its value is proportional to Cm , the positional prior count underlying event m ( as defined in Phase 4 ) : where μ is a parameter to tune the effect of motif based prior , 0≤μ<1 . In this study , we choose μ = 0 . 8 . The rationale is that if the k-mers mapped to position m have more occurrences at binding events genome wide , it is more likely to cause a binding event at that genome position . The parameter αm is scaled such that all the values of possible αm will be less than αs . Therefore the k-mer based prior will not force the model to predict a binding event at a motif position when the observed reads do not provide sufficient evidence of a protein-DNA interaction event . Since the k-mers underlying the possible binding event positions and their counts are known , the value of term −αs+αm remains constant when we estimate the parameters in the mixture model . Therefore , we can solve the mixture model using Expectation-Maximization ( EM ) algorithm [47] . The complete-data log penalized likelihood is: where 1 ( zn = m ) is the indicator function . In the E Step we have: where γ ( zn = m ) can be interpreted as the fraction of read n that is assigned to event m . In the M step , on iteration i we find parameter to maximize the expected complete-data log penalized likelihood: under the constraint . By simplifying we find the close-form solution of the maximization as: where Nm is the effective number of reads assigned to event m , or the binding strength of event m . Intuitively , the effective read count of an event is decreased by a pseudo-count αs for the sparseness penalty , and is increased by a pseudo-count αm for the k-mer motif at position m . If for event m , the value of πm becomes zero , the model is restructured to eliminate it [46] . The EM algorithm is deemed to have converged when the change in likelihood falls below a small value , for example 1e−5 . Since the value of term −αs+αm is negative , a binding event supported by enriched k-mers may still be eliminated if it is not sufficiently supported by read data . In addition , a binding event not supported by enriched k-mers may still survive if it is sufficiently supported by the read data . The predicted binding events are tested for significance as described previously [8] . Briefly , if a control dataset is available , we compare the number of reads in the ChIP event to the number of reads in the corresponding region in the control sample using a Binomial test . If control data is not available , we apply a statistical test that uses a dynamic Poisson distribution to account for local biases . To correct for multiple hypothesis testing , a Benjamini-Hochberg correction [48] is applied . It is worth mentioning that we only use read counts of events to test for significance . The read spatial distribution of binding events is updated after each round of binding event prediction . Phase 6 repeats Phase 2 and 3 motif discovery using the binding events predicted from Phase 5 . As described in the results section ( Figure 1 ) , the spatial accuracy of binding events discovered from Phase 5 ( GEM ) is significantly improved from Phase 1 ( GPS ) . Thus , these events will be more accurately centered on motifs and the performance of motif discovery is correspondingly improved . GEM is a stand-alone Java software that takes alignment files of ChIP-Seq reads and a genome sequence as input and reports a list of predicted binding events and the explanatory binding motifs . It can be downloaded from our web site ( http://cgs . csail . mit . edu/gem ) . For analysis with mammalian genomes , GEM requires about 5–15 G memory . 214 ENCODE ChIP-Seq datasets that have an embargo date before Oct 28 , 2011 and have known motifs in public databases were downloaded from the ENCODE project website [18] . 16 mouse ES cell factor ChIP-Seq datasets published in references [16] and [28] were downloaded from GEO . ChIP-exo data were provided by Ho Sung Rhee and B . Franklin Pugh . FastQ files of the ChIP-Seq/ChIP-exo data were then aligned with genome ( human hg19 , mouse mm9 ) using Bowtie [49] version 0 . 12 . 7 with options “-q --best --strata -m 1 -p 4 --chunkmbs 1024” . The GABP ChIP-Seq data was downloaded from QuEST website ( http://mendel . stanford . edu/SidowLab/downloads/quest/ ) and was pre-aligned to hg18 genome . GEM was applied to 214 ENCODE ChIP-Seq data . The motif PWMs output by GEM were collected . An alternate pipeline used the GPS peak-finder [8] to call binding events and used 7 different motif finding methods ( AlignACE v4 . 0 [23] , MDscan v2004 [22] , MEME v4 . 7 . 0 [20] , Weeder v1 . 4 . 2 [21] , POSMO v2 [24] , HMS v0 . 1 [13] and ChIPMunk v3 [14] ) to discover motifs independently . For AlignACE , MDscan , MEME and Weeder , 100 bp sequences were extracted from the top 500 peaks from each dataset , as suggested by the MEME Suite's documentation based on the typical resolution of ChIP-Seq peaks . For POSMO , we extracted a set of 100 bp sequences from the top 500 GPS peaks . This set of sequences provided superior results when compared with sequences taken from the top 5000 1000 bp sequences ( as suggested by the author of POSMO ) . For ChIP-Seq oriented methods , HMS and ChIPMunk , a set of 100 bp sequences and corresponding read coverage profiles were extracted from the top 500 GPS peaks . We found these conditions provided superior results than using sequences taken from the top 5000 200 bp sequences ( as suggested by the authors of these methods ) . MEME was run with “-nmotifs 6” and Weeder was run with option “large” . POSMO was run with options “5000 11111111 sequence_file 1 . 6 2 . 5 20 200” . ChIPMunk was run with options “6 15 yes 1 . 0 p:read_coverage_profile 100 10 1 4 random 0 . 41” . HMS was run with options “-w motif_width -dna 4 -iteration 100 -chain 50 -seqprop 0 . 1 -strand 2 -base read_coverage_profile -dep 2”; motif_width was determined by width of motif discovered by MEME for the same data . All other parameters were the defaults specified by the authors . We collected known binding preference motifs from the TRANSFAC [50] , JASPAR [51] , and Uniprobe [52] databases . We only include motifs of the factors of interest or motifs for the TF family but not motifs of factors in the same family because factors in the same family may have very different binding motifs . The list of database matrices is provided in Dataset S1 . Discovered motifs were compared to known motifs using STAMP [19] . A motif with E-value less than 1e-5 was considered a match . For each program , we counted the number of datasets that had a motif matching at least one known motif of that transcription factor . In some cases , the correct motifs are not matched by the first motif that a method outputs , but by the second or later motifs . Therefore we compare the motif-finding performance using the top 1 , top 2… or top 6 motifs . Little improvement is observed after the 6th motifs . The genome-wide performance of spatial resolution in ChIP-Seq event calls is evaluated as following . We define effective spatial resolution as the average absolute value of the distance between genome coordinates of predicted binding events and the middle of the corresponding high-scoring binding motif hit . Because the center of the motif hit may not represent the true center of a binding event , the offsets to the motif were centered by subtracting the mean offsets . We compare spatial resolution on the “matched” set of predictions that are called by all the methods and correspond to the same high-scoring binding motif . Only those events within 100 bp of a motif match are included in the calculation . An alternative evaluation with all the events that have a motif at a distance less than 100 bp is also performed . The genome-wide performance of proximal event discovery in ChIP-Seq data is evaluated as follows . For GABP dataset , we compared GEM against other 6 methods ( GPS , SISSRs , MACS , cisGenome , Quest and PeakRanger ) genome wide . We define a set of candidate sites that all have at least one event detected by all seven methods , and that contain two or more GABP motifs separated by less than 500 bp . We discovered 477 such sites . For each of these sites , we count the number of events discovered by different methods . GABP motif was retrieved from TRANSFAC database ( M00341 ) [50] . A motif score threshold of 9 . 9 , which is 60% of maximum PWM score , is used in this analysis . In this study , to test GEM's ability to automatically adapt to ChIP-exo data , we initialized GEM with a ChIP-Seq empirical read distribution , and ran GEM with one extra run ( phase 5 and 6 ) so that GEM could use more accurately positioned events to refine the read distribution and use it for final prediction . In practice , the user can directly initialize GEM with a ChIP-exo empirical read distribution ( provided with GEM software ) and apply GEM the same way as analyzing ChIP-Seq data . To study the in vivo binding spatial relationship between a pair of transcription factors A and B in the certain cell type and condition , we apply GEM independently to ChIP-Seq data from A and B to predict the respective binding sites . To compute the distribution of spacing between A relative to B , we compute the offsets of A binding sites from B binding sites within a 201 bp window . The sequence strand of the binding predictions is oriented using the B motif when a match to the motif is present , and B is placed in the middle of the window . The occurrences of A at each offset position are summed over all the B sites to produce the empirical spatial distribution . In this study , we evaluate three different methods to call binding sites: GEM binding calls , GPS binding calls , and GPS binding calls that are snapped to a motif within 50 bp if one is present . Another motif distance for snapping binding calls , 100 bp , was also tested and the result was very similar to the 50 bp distance . To determine if a specific spacing is significant , we compute the p-value of the number of occurrences of factor A at that offset position using a Poisson test . The parameter of Poisson distribution is set as the mean number of occurrences across all the positions in the [−400 bp −200 bp] and [200 bp 400 bp] windows , assuming there are no significant spatial binding constraints in these windows . The p-value is corrected for multiple hypotheses testing using Bonferroni correction by multiplying the p-value by the number of positions in the window and the total number of pair wise tests across all cell types . The significance threshold for corrected p-value is 1e−8 . Because the strand orientation of bound sequences cannot be oriented consistently when comparing multiple factor pairs , we report the absolute distance between the most significant interacting factor pairs in Figure 6 . | The letters in our genome spell words and phrases that control when each gene is activated . To understand how these words and phrases function in health and disease , we have developed a new computational method to determine what word positions in our genomic text are used by each genome regulatory protein , and how these active words are spaced relative to one another . Our method achieves exceptional spatial accuracy by integrating experimental data with the text of our genome to find the precise words that are regulated by each protein factor . Using this analysis we have discovered novel word spacings in the experimental data that suggest novel genome grammatical control constructs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"sequence",
"analysis",
"genomics",
"functional",
"genomics",
"regulatory",
"networks",
"biology",
"computational",
"biology",
"genetics",
"and",
"genomics"
]
| 2012 | High Resolution Genome Wide Binding Event Finding and Motif Discovery Reveals Transcription Factor Spatial Binding Constraints |
Persistent activity and match effects are widely regarded as neuronal correlates of short-term storage and manipulation of information , with the first serving active maintenance and the latter supporting the comparison between memory contents and incoming sensory information . The mechanistic and functional relationship between these two basic neurophysiological signatures of working memory remains elusive . We propose that match signals are generated as a result of transient changes in local network excitability brought about by persistent activity . Neurons more active will be more excitable , and thus more responsive to external inputs . Accordingly , network responses are jointly determined by the incoming stimulus and the ongoing pattern of persistent activity . Using a spiking model network , we show that this mechanism is able to reproduce most of the experimental phenomenology of match effects as exposed by single-cell recordings during delayed-response tasks . The model provides a unified , parsimonious mechanistic account of the main neuronal correlates of working memory , makes several experimentally testable predictions , and demonstrates a new functional role for persistent activity .
Memory allows an organism to store information about past events and then use it to modify its behavior in response to future ones . As a simple example , consider the classic delayed match-to-sample ( DMS ) task , where the appropriate behavior upon presentation of the test stimulus is conditional on the sample stimulus ( the past event ) . What are the mechanisms that support the storage of information about the sample stimulus and its retrieval upon the presentation of the test stimulus ? Electrophysiological studies have exposed two neuronal correlates of those mechanisms: ( i ) sample-selective persistent activity during the delay period ( see , e . g . , [1–3] ) ; ( ii ) differential test-period activity depending on whether the test stimulus matches or not the sample ( see , e . g . , [1 , 4–7] ) . Specifically , some neurons show enhanced responses in the match as compared to the non-match condition ( match enhancement ) , while others show the opposite pattern ( match suppression ) , despite the fact that stimuli are physically identical in the two conditions . Persistent delay activity , together with match effects , are basic neuronal hallmarks of temporary maintenance and manipulation of information . Persistent activity has been primarily associated with short-term storage , while match effects are thought to reflect the outcome of the comparison between stored and incoming information . Consistently with these putative roles , they have been observed across different short-term memory protocols besides the DMS task , such as DMS with intervening distractors [8 , 9] , delayed non-match-to-sample [10] , delayed paired-associate [11] , delayed categorization [12] and delayed cue-instructed go/nogo tasks [13] , and with different class of stimuli , such as natural images , fractal images , spatial locations , motion directions and pure tones . Persistent activity and match effects have also been observed across diverse cortical regions and , interestingly , the basic phenomenology exposed in regions traditionally associated to memory function , such as the pre-frontal , infero-temporal and parietal cortices [8 , 9 , 14] , is qualitatively very similar to the phenomenology observed in regions traditionally associated to sensory processing , such as auditory cortex [15] , area V4 [16] and area MT [17] . It is presently unclear how the information stored in the pattern of persistent activity is retrieved and compared with incoming stimuli . One theory holds that match enhancement and persistent activity on one hand , and match suppression on the other hand , are neuronal manifestations of two distinct , parallel mechanisms supporting memory storage and retrieval [1 , 18] . Enhancement reflects the operation of an active detector which signals the appearance of behaviorally relevant stimuli , stored by persistent activity during the delay period . Suppression , instead , reflects the operation of an automatic detector which signals stimulus repetitions regardless of their behavioral relevance , and its functioning is assumed to be independent of persistent activity . Current mechanistic modeling fully embraces this notion of parallel mechanisms ( see , e . g . , [19] ) . Match suppression , which is generally the most prominent effect , is understood as the result of activity-dependent fatigue , either at the single-cell or at the synaptic level , produced by the sample presentation . Match enhancement , on the other hand , is understood as the result of selective inputs from a different area , or from a functionally different neuronal sub-population within the same area , carrying information about the stimulus actively held in memory . In some cases , however , the reported experimental phenomenology does not appear to be fully consistent with the parallel mechanisms theory . In delayed paired-associate tasks , the response is conditional on whether the test stimulus matches the pair-associate of the sample stimulus . No repetition occurs . According to the theory , only match enhancement ( or no effect , depending on the identity of the test stimulus ) should be observed . Experiments report both enhancement and suppression [11] . In delayed match-to-category tasks where stimuli also do no repeat within a trial , similarly both enhancement and suppression are observed when the test matches the category of the sample [12] . In a recent study using a delayed cue-instructed go/nogo tasks with distractors , differential responses to neutral stimuli depending on the cue stimulus have been reported [13] . The theory would rather predict no response modulation , since neutral stimuli have no behavioral relevance nor they repeat during the trial . Finally and importantly , both match enhancement and selective delay-period activity ( together with match suppression ) have been observed in passive fixation tasks [14 , 20–24] , where the theory would predict only match suppression . In fact , in such a task none of the stimuli has behavioral relevance and , thus , there is no need for active repetition detection , let alone memory maintenance . Here , we propose an alternative to the parallel mechanisms theory , where match enhancement and suppression both result from a single mechanism: the modulation of network excitability brought about by the persistent activity . Information about the presented stimulus is stored across the delay period by a pattern of persistent activity in which , depending on the stimulus , some neurons increase their spiking rate while others decrease it . Changes in the spiking rate must be accompanied by changes in the neuron’s responsiveness to inputs , whereby , upon presentation of a subsequent stimulus , more active neurons will be more depolarized and thus more excitable , while less active neurons will be more hyperpolarized and thus less excitable [25] ( Fig . 1A ) . Changes in single neuron’s responsiveness entail modifications of the tuning properties of the network , which depend on the ongoing pattern of persistent activity . In our account the differential stimulus-evoked activity depending on previously presented stimuli—i . e . , the match effects—is a result of those modifications . We show that the above described mechanism , when implemented in a spiking model network , can reproduce consistently and quite naturally , most of the experimental phenomenology about match effects observed in short-term memory tasks , with no need for fatigue mechanisms nor functional differentiation among neurons .
Cortical neurons typically show visual responses to a large faction of stimuli , even after such stimuli have become fairly familiar [29 , 30] . To reproduce this feature , we consider that the presentation of a stimulus elicits additional inputs ( on top of the background input ) to all neurons in the memory representations . Such inputs , for each neuron and for each stimulus , are drawn independently at the beginning of the simulation—and kept fixed thereon—from two Gaussian distributions with same variance but different means . Inputs to neurons in the memory representation associated to the stimulus presented are drawn from the Gaussian with larger mean , while inputs to neurons in the remaining representations are drawn from the Gaussian with smaller mean ( inset in Fig . 1C ) . Means and variance of the two distributions are chosen to qualitatively reproduce the statistics of evoked cortical responses . The resulting firing rates distribution is shown in Fig . 1C . Evoked responses vary in a wide range , from quiescence to relatively high firing rates . The high-rates tail is mostly constituted by neurons in the memory representation of the stimulus being presented ( red ) . However , a significant fraction of neurons in the other memory representations ( orange ) also fires at high rates . For each neuron in the memory representations , we also measure the sparseness of its responses across the stimulus set via the sparseness index ( see Methods ) , which takes on values between 0 ( the neuron responds to all stimuli in the same way ) and 1 ( the neuron responds to a single stimulus ) . The resulting distribution is shown in Fig . 1D . The average sparseness index as well as its distribution across neurons are consistent with experimental estimates from visual responses evoked in ITC by familiar stimuli [30] . The pattern of stimulus-evoked activity generated by the model reproduces some of the basic features of cortical responses: sparseness at the network level , right-skewed long-tailed distribution of firing rates , and broad tuning curves at the single-cell level [30] . Next , we study how the features of stimulus-evoked activity are modified when the network is in a memory state . We consider the case in which the active memory representation is the one associated with the incoming stimulus . This corresponds to a match trial in the DMS task . Accordingly , we present the same stimulus twice , with the two presentations separated by a delay period . The first presentation strongly activates the corresponding memory representations ( black , Fig . 2A ) and , to a lesser extent , all the other representations ( red , Fig . 2A ) due to the broad selectivity of the evoked responses . At stimulus offset , the memory representation activated by the stimulus remains active at high rates , while the others become quiescent due to the increased level of inhibition ( blue , Fig . 2A ) . Such re-organization of the spiking activity following stimulus presentation is due to a ( self-sustained ) change in the distribution of the recurrent inputs in the network . Neurons in the active representation receive increased inputs as a result of the positive feedback via the strengthened recurrent synapses . The other neurons receive decreased inputs as a result of the negative feedback due to global inhibition . As a consequence , neurons in the active/inactive representation have a higher/smaller response to stimulus repetition ( Fig . 2A ) . Although the basic features of stimulus-evoked activity in the spontaneous state are qualitatively preserved in the memory state , significant quantitative changes occur . In Fig . 2B we plot the response to the first presentation ( sample ) vs . the second presentation ( match ) for the neurons in the active ( black ) and inactive ( red ) memory representations . All neurons in the active representation show enhanced responses to the second presentation , although to a different extent . Neurons that show the largest relative increase are those that had moderate responses upon sample presentation . Neurons in the inactive representations also show a wide range of suppressed responses to the second presentation . Neurons with weak responses upon sample presentation ( i . e . a significant fraction of the responsive neurons—see Fig . 1C ) show no response at all or very strong suppression upon match presentation . Neurons with moderate-to-strong responses show modest levels of suppression or no suppression at all . The memory-dependent modulation of single-neuron responsiveness entails a strong increase in the selectivity of the responses evoked by the second presentation ( compare the distributions of the sparseness index in Fig . 2C and Fig . 1D ) . In our model , the best stimulus for a neuron is typically the one for which the neuron exhibits mnemonic activity so that , upon repetition of that stimulus , its response significantly increases . For a different stimulus the neuron does not exhibit mnemonic activity and thus , upon repetition , its response is suppressed . This is clearly seen in Fig . 2D where we show four single-neuron tuning curves . Stimuli are ranked from best to worst according to the evoked response at sample presentation , i . e . when the network is in the spontaneous state ( in black ) . Upon repetition , i . e . when the network is in the corresponding memory state , the response to the best stimulus increases significantly , while those to the other stimuli decrease ( in red ) . Note the amount of sharpening as quantified by the sparseness indices . We consider the case in which the incoming stimulus is associated with a memory representation different from the one currently active . This corresponds to a non-match trial in the DMS task . Accordingly , we present two different stimuli separated by a delay period . This case requires the consideration of three different neuronal sub-populations: neurons belonging to the active representation , neurons belonging to the representation associated to the incoming stimulus , and neurons in the remaining representations . The time course of the average activity in these sub-populations ( together with the inhibitory population—blue line ) is shown in Fig . 3A . The presentation of the first stimulus ( sample ) activates the corresponding memory representation ( green ) , which then stays active during the delay period much as before ( Fig . 2A ) . The presentation of the second stimulus ( non-match ) evokes a strong response in the corresponding memory representation ( black ) , which eventually shuts down the mnemonic activity related to the first stimulus . The response to the second stimulus is , however , smaller than the response to the first stimulus due to the increased level of inhibition when the network is in a memory state . The neurons not belonging to the active representation nor to the one associated with the non-match exhibit suppressed responses to the second presentation ( red ) . This can be also seen in the distribution of red dots in the scatter plot in Fig . 3B . Interestingly , neurons in the representation active during the first delay ( green dots ) give a strong , transient response to the second stimulus , which is typically larger than the response evoked by the presentation of the associated stimulus ( i . e . , the sample ) . During the second presentation , these neurons receive additional inputs which , although smaller than the ones received by neurons in the representation of the stimulus currently presented ( black dots ) , elicit a strong response , due to the increased responsiveness brought about by persistent activity . These additional inputs are statistically the same as the inputs to all the remaining representations , nevertheless , they elicit dramatically different responses ( compare green and red dots in Fig . 3B ) . This most clearly illustrates the importance of the current network state in determining the responses to incoming stimulation . In Fig . 3C we report the distribution of the sparseness index measured by taking out the stimulus whose memory representation is active during the delay period . The distribution is bimodal as a consequence of the increased responsiveness of neurons in the active representation , which transiently respond to all stimuli , inducing a loss of selectivity . On the other hand , the selectivity of the neurons not belonging to the active representation is sharpened as a result of the suppression of weak/moderate responses . Such differential changes in selectivity are clearly shown in the two sets of tuning curves in Fig . 3D . In line with the analysis commonly performed on neurophysiological data , we average neuronal responses , across all neurons and all stimuli , conditional on whether responses are collected in match or non-match trials ( e . g . [8 , 9 , 24] ) . In Fig . 4A we plot the average activity across memory representations vs . time in match and non-match trials . Consistently with recordings in ITC and PFC [8 , 11] , we find an overall suppression effect: the network response is weaker in match than in non-match condition . However , the scatter plot in Fig . 4C shows that while the largest proportion of cells have a stronger response to non-match than to match ( red dots below the diagonal ) , there are fewer cells ( black dots ) which show the opposite effect [8 , 9 , 11 , 24 , 31] . Interestingly , in our model , cells which show match suppression with respect to sample also show a match response lower than the non-match , while cells which show match enhancement with respect to sample also show a match response larger than the non-match . The difference in the response to matching and non-matching stimuli is a result of transient changes in network excitability brought about by the persistent activity . The duration of the transient is mainly controlled by XE , the fraction of slow , NMDA-like recurrent excitatory currents ( see also Section Dependence on persistent activity , response selectivity and current dynamics ) . To illustrate this dependence we plot , in Fig . 4B , the average match and non-match responses for three different fractions of slow currents , for a presentation time of 2 seconds . The difference in the responses to match and non-match decays after a time that is proportional to the fraction of slow excitatory currents . Nevertheless , the two responses are still distinguishable after 1 second even for XE = 0 . 3 . This time interval is significantly longer than the typical presentation times used in experiments , which rarely exceed 500ms . It has been shown that , for a significant fraction of neurons , tuning curves differ in match and non-match conditions ( e . g . [24] ) . Consistently , we find that the response of most neurons when the stimulus is a match differs from their response when the same stimulus is a non-match . The vast majority of neurons in our simulation exhibits mixed effects , that is they show match larger than non-match response for some stimuli , and the opposite behavior for other stimuli ( Fig . 4D ) . In the experiments a significant fraction of cells shows non-mixed effects , i . e . match larger than non-match response or viceversa [8 , 11 , 24] . The dominance of mixed cells in our simulation is due to the fact that ( i ) all neurons in the sample we monitor have sharply-tuned persistent activity ( i . e . they exhibit persistent activity for a single stimulus ) ; and ( ii ) the set of stimuli we use to probe them is optimal , in the sense that for each neuron there is a stimulus that elicits persistent activity in that neuron . These conditions are likely not to be fulfilled in a real experiment . For instance , recordings from a cell showing persistent activity for all stimuli ( which are , typically , not a large number ) would result in tuning curves similar to the ones showed in the first two sub-plots in Fig . 3D . Similarly , recordings from a cell showing visual responses but never persistent activity would result in tuning curves similar to the ones showed in the second two sub-plots in Fig . 3D . In general , we would expect the number of mixed cells observed in experiments to increase with the number of stimuli used . We ask next whether the basic mechanism illustrated above is also able to account for the patterns of neural responses observed in DMS tasks in which other stimuli ( distractors ) are presented in between the sample and the test stimulus . In one of such protocols , the so-called ABBA protocol [9 , 18] , the behavioral response is conditional upon the repetition of the sample stimulus ( A ) while the repetition of the distractor ( B ) has to be ignored . In this case , cells were found that showed an enhanced response to the repetition of A—the behaviorally relevant match—while they showed suppressed responses ( as compared to the match response ) for the repeating distractor B . At the same time , cells exhibiting match suppression showed larger responses ( as compared to the match response ) for the repeating distractor [9 , 18] . It is worth pointing out that the parallel mechanisms hypothesis was formulated in order to account for these experimental observations , in particular for the observation that match enhancement was apparent only upon the repetition of the behaviorally-relevant stimulus [18] . Our model is able to reproduce the experimental findings described above in a regime where the mnemonic representation activated by the sample stimulus survives the presentation of a distractor with high probability , but is occasionnally disrupted by it . This regime can be obtained by reducing the amplitude of the external inputs during stimulus presentation [32] ( see Methods for details ) . Fig . 5A shows the neural activity in the mnemonic representation associated to the sample stimulus ( left panel ) and in the remaining representations ( right panel ) . The presentation of the sample activates the corresponding mnemonic representation but also , to a smaller extent , other representations ( see Fig . 2 ) . The subsequent presentation of a distractor fails to abolish the mnemonic activity related to the sample ( unlike in Fig . 3 ) , although it evokes responses in both the active and the inactive representations . We then proceed as in the experiment [9 , 18] . For each stimulus presented as a sample , we separate the neurons which show enhancement upon repetition ( i . e . , the ones in the corresponding mnemonic representation ) from the neurons which show suppression ( i . e . , the ones in the remaining representations ) . Next , we average the activity in these neuronal populations across all trials ( with distractors ) where the corresponding stimulus is presented as match , non-match and repeated non-match . The results of this analysis are shown in Fig . 5B . As can be seen , neurons eventually showing match enhancement exhibit suppressed responses , as compared to the match response , to repeating distractor presentations ( left panel ) . Similarly , neurons eventually showing match suppression exhibit enhanced responses , again as compared to the match response , to repeating distractor presentations . Another common observation in protocols with distractors is that suppression wanes with increasing number of distractors [8] . This finding can also be replicated in the regime where persistent activity survives distractor presentation most of the time , but is occasionally disrupted by it . The amplitude of the population-averaged response depends on whether the mnemonic representation active upon test presentation is the one corresponding to the sample . In particular , if the persistent activity survives the presentation of the distractors , significant suppression will be observed in the majority of cells upon test presentation ( see Match-Non Match effects ) . On the other hand , if the persistent activity is disrupted by the distractor presentation , the fraction of cells exhibiting match suppression will be smaller . The probability that the mnemonic representation associated to the sample is still active upon test presentation is a decreasing function of the number of distractors . In Fig . 6 we plot the average response of cells exhibiting match suppression ( i . e . belonging to the mnemonic representations inactive following sample presentation ) as a function of the number of intervening distractors . As can be seen , the response in match and non-match trials becomes more similar , i . e . the level of suppression decreases with increasing number of distractors . The same parameter regime can reproduce “standard” match , non-match effects , i . e . when no distractors intervene between sample and match or non-match ( “0 distractors” in Fig . 6 ) . Hence , importantly , in our model a unique set of parameters can account for standard match non match protocols , as well as for protocols with intervening stimuli . We analyze systematically how match effects depend on persistent activity ( PA ) , selectivity of the visual responses , and currents dynamics via both a simplified rate dynamics ( SRD -that converges to the stationary states described by mean field equations; see Methods ) and the spiking network dynamics ( SND ) . First , we study the dependence of match effects on the level of PA by manipulating the synaptic strength between neurons in the same memory representations ( J+ ) , keeping all the other parameters fixed . To quantify match effects we use the enhancement ( suppression ) index , defined as the ratio between match ( M ) and sample ( S ) response in the active ( inactive ) memory representation and the match-non match index , defined as the ratio between M and non-match ( NM ) response in a memory representation . We find that match effects increase with increasing levels of PA ( Figs . 7A–B ) . Neurons in the active memory representation show larger responses to M than to S , as well as larger responses to M than to NM ( black and blue line in Fig . 7B ) . The amplitude of both suppression and enhancement increases with increasing J+ due to the concomitant increase of the level of PA ( increasing responsiveness in the active representation ) and the level of global inhibition ( decreasing responsiveness in the inactive representations ) . In the absence of PA , all indices are equal to one , i . e . there is no modulation of response . The simplified rate dynamics ( SRD-full lines ) give results that are very close to the spiking neurons dynamics ( SND-dashed lines ) . Next , we study the dependence of match effects on the selectivity of the visual responses by manipulating the difference ( α ) between the average additional input to the memory representation associated to the stimulus presented ( α + β ) and the average additional input to the other representations ( β ) ; all other parameters are kept fixed . We find that match effects are quantitatively more important for low/moderate levels of selectivity , but are nevertheless preserved for strongly selective responses . For α close to zero , the response to stimulus presentation is not selective enough to allow the network to activate the corresponding memory representation . In this regime , all indices calculated via the SRD are equal to one ( full lines in Fig . 7C ) . For moderate α ( 0 . 1 mV < α < 0 . 8 mV ) , the network is able to activate the memory representation associated with the presented stimulus and , moreover , PA survives the presentation of a NM . As a result enhancement effects are quite strong ( compare the blue/black with the red/green curves ) . In the SND , instead , the stimulus presentation could stochastically activate any of the memory representations , thus yielding high fluctuations on the indices ( dashed lines in Fig . 7C ) and low agreement with SRD . For larger α , PA is disrupted by the presentation of a NM , and enhancement and suppression effects become more comparable . Note that the range of α for which PA survives the presentation of the NM increases with increasing J+ . In our model , match effects result from modifications of the transient network response following stimulus presentation . Accordingly , decreasing the fraction of recurrent inputs with slow dynamics , and thus speeding up transients , reduces both suppression and enhancement effects ( Fig . 7D ) . Nevertheless , even for relatively small fractions of slow recurrent inputs , match effects are still significant when the responses are averaged over the entire duration of stimulus presentation ( i . e . , 500ms—right panel in Fig . 7D ) . This is consistent with experiments showing that match effects are mostly evident during the early response phase [33] . We have focused so far on a regime in which the sensory and the mnemonic representations of a stimulus are highly correlated . The best stimulus for a neuron is typically the one for which it exhibits persistent activity during the delay period and , therefore , it is also the one which elicits the highest response upon match presentation . Thus , strong responses during the sample presentation will typically be enhanced in the match condition , while poor responses will be suppressed . In this regime , active mnemonic representations lead to a significant sharpening of the associated sensory representations upon stimulus repetition , as we have shown . This could be the relevant scenario for sensory-related areas . In higher order areas ( e . g . , the pre-frontal cortex ) , however , sensory and mnemonic representations appear to be less correlated . As frequently observed in the data , neurons which show strong selectivity during stimulus presentation do not necessarily show strong selectivity during the delay period . Vice-versa , some neurons have been shown to exhibit weakly selective response , or no response at all during stimulus presentation while developing strong selectivity during the delay period ( see , e . g . , [9] for a comparison between stimulus-evoked and delay activity in the inferotemporal and prefrontal cortices ) . To investigate the effects of reduced correlation between sensory and mnemonic representations , we manipulated neural response heterogeneity upon stimulus presentation by increasing the variance of the external inputs ( σs—see Methods for details ) . As σs increases , the distribution of inputs to the neurons in the mnemonic representation corresponding to the stimulus being presented becomes more and more similar to the distribution of inputs to the other representations and consequently , the neural responses in these two sets also become similar . As a result , selectivity of neural responses during sample presentation decreases . To quantify the effects of increased variance in the external inputs , we computed for each neuron in the mnemonic representations the correlation ( Pearson correlation coefficient ) between stimulus-evoked and delay activity across the set of stimuli . The correlation averaged across all neurons in the mnemonic representations is reported in left panel Fig . 8 ( black curve ) . Increasing the variance of the external inputs , steadily and significantly , reduces the correlation between the pattern of activity elicited by the sample and the subsequent pattern of persistent delay activity . This is due to the reduction of the proportion of neurons exhibiting both strong stimulus response and enhanced delay activity . For purpose of illustration , we also show in the right panel of Fig . 8 the tuning curves during sample presentation and delay period for four neurons . As can be seen , both neurons which share the same stimulus preference during sample and delay , and neurons which do not can be found . Note that increasing the variance of the external inputs has a significantly weaker effect on the correlation between the pattern of activity during the delay period and the pattern of activity elicited upon match presentation ( left panel of Fig . 8—red curve ) . Neurons exhibiting high ( low ) levels of delay activity will also tend to exhibit strong ( poor ) responses to match presentation , regardless of the extent of their response at sample . Thus , match effects are largely preserved in the presence of strong heterogeneity in neuronal responses .
By mechanistically linking persistent activity and match effects , our model explains naturally why the latter are observed in the same cortical regions where the former is present ( e . g . in ITC: [8 , 33 , 35–38]; in PPC: [39–41]; in PFC: [9 , 11 , 17 , 24] ) . Upon stimulus repetition , we find that suppression effects are dominant , involving 80% of cell/stimulus combinations . This is due to two factors: ( i ) cells have broad selectivity properties during visual response , i . e . they respond to most of the presented stimuli; ( ii ) delay period activity is sparse , i . e . it involves only a small fraction of the excitatory cells of the network . Both features are consistent with neurophysiological recordings in areas of the temporal and frontal lobe . It has been reported that some cells show complete adaptation , i . e . essentially no response to stimulus repetition , despite vigorous response to novel stimuli [42] . Accordingly , we find cells which , although active at sample , stop firing during match presentation ( Fig . 3B ) . In these cells , as in most of the selective cells in our network , the presentation of a new stimulus elicits a response ( Fig . 2B ) , consistent with the observation that cells that show a modulation of response following stimulus repetition are usually broadly tuned [8 , 42] . The model predicts that ( i ) for a given stimulus , ( early ) single-cell responses should be positively correlated with the level of delay period activity preceding the test presentation; ( ii ) the proportion of cells showing match enhancement increases with the proportion of cells showing persistent activity , and the higher the level of persistent activity , the larger the amplitude of the enhancement effect . This is consistent with comparisons between IT and PF cortex: enhancement effects are stronger , in proportion of cells involved and in amplitude , in PFC than in ITC , consistent with the fact that the proportion of cells showing persistent activity during the delay period is larger in PFC [9 , 11 , 12] . Similarly , both persistent activity and match effects increased significantly after training with a DMS task , although both were found also in naive animals [24]; ( iii ) repetition produces ( transiently ) a sparser representation of the stimuli . This is due to the fact that the pattern of persistent activity modifies the transient response to repeated stimuli by enhancing the response of the neurons most active during the delay period while suppressing the response of the other—less active—neurons . In a regime where sensory and mnemonic representations are highly correlated , our scenario therefore provides a mechanistic account for the sharpening model [1 , 5 , 6 , 42–45] , according to which some but not all neurons that initially respond to a stimulus , show suppression to the repetition of that stimulus , and , most importantly , suppression is stronger for the non preferred stimuli , i . e . neurons showing little or no suppression to a repeated stimulus are highly selective for it . Note that such a mechanism , in which persistent activity acts as a ‘matching filter’ that transiently sharpens the response to the test if it matches the sample , could also potentially account for priming phenomena [46] . When increasing the heterogeneity of the evoked responses , and thus decreasing the correlation between sensory and mnemonic representations , the model also accounts for cells showing suppression instead of enhancement upon repetition of their preferred stimulus ( see , e . g . , [37] ) , whose tuning curves would look like those shown in Fig . 8 . Miller et al . [8 , 9 , 18] used protocols with distractors , and found that ( i ) enhancement is observed for behaviorally relevant matches , but not for repeating distractors; ( ii ) both suppression and enhancement effects are still present after a few distractors are presented , but decay progressively with the number of distractors . Our model can reproduce these data in a stochastic scenario in which in most cases the distractors do not perturb the active mnemonic representation , but with some non-zero probability erase the memory from the system . This scenario is fully consistent with both the behavioral [8] and the electrophysiological data in PFC indicating resistance to distractors [9] . Electrophysiological evidence for resistance to distractors in ITC is more mixed ( e . g . , [9] ) , though we note some degree of resistance to distractors has been found both in ITC ( see , e . g . , [37 , 47] ) and in the entorhinal and perirhinal cortices [10 , 48] . Going one step forward , the model predicts that any manipulation ( e . g . , pharmacological ) affecting persistent activity should also have significant impact on match effects . Weakening or destroying persistent activity is expected to severely diminish the amplitude of match effects or abolishing them altogether . For example , it has been shown that the enhancement of the GABA-ergic neurotransmitter system , e . g . via benzodiazepines , slows down working memory processes [49 , 50] and impedes repetition suppression [51] . Alternatively , boosting persistent activity by reducing the inhibitory feedback [52] is expected to significantly reduce suppression effects , while increasing enhancement effects both in the proportion of cells exhibiting them and in their amplitude at the single-cell level . The simplest model for match suppression is some form of adaptation ( either firing-rate adaptation , or synaptic short-term depression ) on sufficiently long time scales [53] . Match enhancement could be accounted for by a separate population of pyramidal cells having predominantly facilitating synapses , as found in PFC [54] . These purely passive mechanisms are however hard to reconcile with evidence that inter-trial intervals erase match effects , as reported by [8] . Our model would be consistent with the fact that inter-trial intervals erase these effects , provided persistent activity is switched off during inter-trial intervals ( but see [55] ) . Note that introducing adaptation and/or short-term synaptic plasticity in our framework would not alter qualitatively the conclusions . Adaptation and short-term depression would tend to decrease quantitatively both enhancement and suppression effects , while short-term facilitation would tend to increase these effects . A combination of passive and active mechanisms—inspired by the parallel mechanisms theory—has been implemented in a recent study by [19] , through a network composed of two separate neural populations ( one exhibiting enhancement , the other suppression ) , receiving top-down inputs from a working memory area . The model was shown to be able to reproduce part of the experimental phenomenology reported in [8 , 9 , 18] , but leaves aside a rather critical issue: by assuming two separate sets of cells showing either match enhancement or suppression , the model cannot account for the experimental observation of mixed cells , which show both enhancement and suppression depending on the stimulus presented [24] . We note that , albeit the underlying mechanism is different , the patterns of neuronal activity produced by our model are the same as the ones produced by the Engel and Wang model [19] . Their learning circuit would hence be effective also in our network , providing a biologically realistic read-out mechanism . Another class of models for match effects relies on synaptic modifications induced by stimulus presentations [56 , 57] . These mechanisms are not mutually exclusive , and might cooperate to generate strong match signals . As our main goal in this work was to show that changes in excitability brought about by persistent activity were a viable mechanism for match effects , we have chosen the simplest possible spiking neuron model able to maintain patterns of persistent activity thanks to increased synaptic strength between populations of neurons [26] . In such a model , firing rates in persistent activity are too high and homogeneous , spiking irregularity too low , and the proportion of neurons showing match effects somewhat larger as compared with the experimental data , unless additional features such as short-term plasticity are added ( e . g . , [58 , 59] ) . We do not expect these additional features to change qualitatively the picture described in this paper , rather we expect them to bring model behavior quantitatively closer to experimental data ( see Effects of heterogeneity in neuronal responses ) . We note that the general scenario presented above would still hold if attractors are stabilized by other mechanisms ( e . g . , [60–62] ) provided enhanced activity entails increased single-neuron excitability . The theory we have presented demonstrates a new functional role for persistent activity beyond temporary memory storage . Our results show how persistent activity can be instrumental in the retrieval of the stored information and , potentially , in the context-dependent encoding of incoming information ( see , e . g . , [13] ) . In our implementation , persistent activity is the result of the network possessing multiple steady states of activity ( attractors ) each manifested by elevated firing rates in stimulus-selective sub-populations of neurons [26 , 27] . The fact that the network is in such an attractor modifies the transient response to incoming stimuli; such response could then be exploited by a readout network , which could easily solve any task involving comparisons of sample and match/non-match stimuli . This general scenario would still hold if attractors are stabilized by other mechanisms [60 , 61] . Our results also lay the groundwork for uncovering physiological/mechanistic substrates common to different types of mnemonic processing , by suggesting specific neuronal signatures of memory retrieval and possible underlying mechanisms .
In the following we describe the full spiking neurons network simulation used for the results reported in the main text . The behaviour of a single excitatory or inhibitory neuron in the network can be described ( below firing threshold ) by the dynamics of its membrane potential , which obeys τ E , I V i ( t ) ˙ = - V i ( t ) + I i ( t ) ( 1 ) with i = 1 … N , where N = NE + NI is the total number of neurons in the network , τE , I is the integration time constant of the membrane potential for excitatory and inhibitory neurons , and Ii ( in mV ) is the total current impinging on the ( post-synaptic ) neuron . When the membrane potential reaches the firing threshold θ , upon integration of the incoming current , then a spike is emitted , the membrane potential is reset to a value VR and the neuron remains refractory for a time τarp . The total current arriving to a postsynaptic neuron is due to the activity of its local ( pre-synaptic ) afferents and to the current elicited by external afferents , e . g . neurons in neighboring cortical areas , namely I i ( t ) = I i e x t ( t ) + I i r e c ( t ) ( 2 ) where I i e x t ( t ) is the external input current and I i r e c ( t ) is the recurrent current . The recurrent contribution to the post-synaptic current comes from local ( pre-synaptic ) afferents and is mediated by NMDA , AMPA and GABA receptors , so that I i r e c ( t ) = I i N ( t ) + I i A ( t ) - I i G ( t ) ( 3 ) where each of the currents follows its own temporal dynamics: τ N I ˙ i N ( t ) = − I i N + X E , I τ E , I ∑ j J i j ∑ k δ ( t − t k ) ( 4 ) τ A I ˙ i A ( t ) = − I i A + ( 1 − X E , I ) τ E , I ∑ j J i j ∑ k δ ( t − t k ) ( 5 ) τ G I ˙ i G ( t ) = − I i G + τ E , I ∑ j J i j ∑ k δ ( t − t k ) ( 6 ) where XE , I is the fraction of the charge coming from excitatory afferents which is mediated by NMDA receptors and elicits a slower current dynamics on the post synaptic neuron . The remaining fraction of the charge ( 1 − XE , I ) is mediated by AMPA receptors and elicits a faster current . The current coming from inhibitory afferents is mediated by GABA receptors . Upon arrival of a pre-synaptic spike at time tk , the postsynaptic current instantaneously receives a “kick” proportional to the synaptic efficiency J ( in mV ) , followed by an exponential decay with time constant τsyn , where τsyn = τA , N , G . In a regime of spontaneous activity , i . e . when no external stimulus is presented to the network , I i e x t ( t ) = μ i e x t + σ e x t τ η i ( t ) ( 7 ) where μ i e x t is the value of the external current extracted for each neuron from a Gaussian distribution with mean μ ¯ E e x t and μ ¯ I e x t , respectively for excitatory and inhibitory neurons , and variance σ B G 2 ( quenched noise ) while ηi ( t ) is a white noise process with < ηi ( t ) > = 0 and < ηi ( t ) ηj ( t′ ) > = δij δ ( t − t′ ) uncorrelated from neuron to neuron and σext is the amplitude of the temporal fluctuations around μ i e x t ( fast noise ) . Upon presentation of a stimulus each neuron belonging to the selective populations ( i . e . the memory representations ) receives an external current given by: I i e x t ( t ) = μ i s t i m + σ e x t τ η i ( t ) ( 8 ) where μ i s t i m = μ i e x t + μ i s e l ( 9 ) if i belongs to the selective foreground , the current elicited by the stimulus on each neuron , μ i s e l , is drawn from a Gaussian distribution with mean α + β and variance σ S 2; if i belongs to the selective background instead: μ i s t i m = μ i e x t + μ i s e l ( 10 ) where μ i s e l is drawn from a Gaussian distribution with mean β and variance σ S 2 . Those distributions are quenched , so that the presentation of a given stimulus elicits always the same current in a given selective neuron . The Equations for the membrane potential ( Equation 1 , together with the condition for spike emission and refractoriness ) and the Equations 4–6 and 7 for the currents are integrated using the Euler method with a time step dt = 0 . 1ms . The mean value of the external current μ ¯ E , I e x t is calculated using mean-field Equations ( see below ) such that the background activity is at a chosen value ν ‾ E s p , ν ‾ I s p . The value of the synaptic potentiation J+ between neurons belonging to the same selective population was chosen to ensure stable persistent activity . The network has been simulated using the standard protocols of match and non-match trials , as well as protocols in which other stimuli ( distractors ) , repeating or not , are presented in between the sample and the test . All protocols are described below . Figs . 2A . and 3A . show the time course of single match and non-match trials simulated with the spiking neurons network with α = 1 . 5mV and β = 1 . 8mV ( all the other parameters are listed in Table 1 ) . Protocols with intervening stimuli ( Figs . 5 and 6 ) are simulated with the spiking neurons network with α = 0 . 84mV and β = 1 . 7mV ( all other parameters in Table 1 ) . Average responses to stimuli -sample , match , non-match , repeated non-match- are calculated by counting the number of spikes discharged by selective foreground and selective background during the first 200ms of stimulus presentation . We solve mean field Equations to find the average firing rates of the network p+2 populations in stationary conditions [26] . In particular , we study the network state in the absence of stimulus presentation , i . e . during the spontaneous activity state and during the delay period of a DMS task . The set of parameters used to find the networks stationary states is given in Table 1 ( in bold ) ; we chose them to be compatible with realistic cortical anatomy . Via the mean field approach described above we are able to completely define the attractor landscape of the network , i . e . the available stationary states of its dynamics . As a next step , in order to study the network’s transient response during sample , match and non−match presentations , we use a simplified rate dynamics on the average population currents . We solve the dynamics for each epoch of a match and non—match trial , i . e spontaneous activity , sample presentation , delay period , match/non-match presentation ( see Protocols for the time duration of each epoch ) . Such model , without being as realistic and detailed as a network of spiking neurons , allows us to explore the relevant parameters’ space in a less time consuming way . The rate dynamics is not exact , but becomes a good approximation when the firing rates are low [64] and gives in general a good approximation of the spiking network dynamics ( see Fig . 5 in the main text ) . In accordance with the spiking neuron model ( see the above section ) , we consider the different kinetics of the AMPA , GABA , and NMDA receptors to describe the dynamics of the average current for each of the four functionally relevant populations in the network . Hence , the total average current afferent on each population is given by μ ¯ K = I K N + I K A - I K G + μ ¯ K e x t ( 23 ) where K = {sf , sb , 0 , I} , μ ‾ K e x t = μ ‾ E e x t ( μ ‾ I e x t ) for the excitatory ( inhibitory ) populations and IN ( IA ) is the contribution to the average current coming from excitatory afferents and mediated by NMDA ( AMPA ) , while IG is the contribution to the average current coming from inhibitory afferents and mediated by GABA receptors . Each of these components evolves in time according to its own time constant τ N I ˙ K N = - I K N + X K μ ¯ K r e c E ; τ A I ˙ K A = - I K A + ( 1 - X K ) μ ¯ K r e c E ; τ G I ˙ K G = - I K G + μ ¯ K r e c I where μ ¯ K r e c E + μ ¯ K r e c I = μ ¯ K r e c and μ ¯ K r e c are defined in Equations 19–22 . As for the spiking network , τN , τA and τG are , respectively , the decay time constants of NMDA , AMPA and GABA currents , while Xsf , sb , 0 ( XI ) is the fraction of NMDA current on the excitatory ( inhibitory ) population and γ = 1 . Note that , as before , the total average currents defined in 23 depend on the average firing rates of each population according to ν ¯ k = 1 σ B G 2 π ∫ - ∞ + ∞ exp - ( μ k e x t - μ ¯ k e x t ) 2 2 σ B G 2 Φ ( I k N + I k A - I k G + μ k e x t , σ ) d μ k e x t ; Having introduced slow and fast current dynamics in the simplified rate model allows to have a good agreement with the spiking neurons model also during stimulus presentation . Responses to sample , match and non-match are calculated either by averaging across the responses of all selective populations over the first T = 200ms of stimulus presentation ( unless otherwise stated ) or by averaging separately across selective foreground and selective background . We use p = 6 different stimuli , corresponding to the p = 6 memory representations in the network . All trials begin with a pre-stimulus interval ( 1s ) in which no external stimulus is presented and the network is in a regime of spontaneous activity . Subsequently , a first stimulus is presented for 0 . 5s ( sample ) . In match trials , after a delay period of 0 . 7s in which the external stimulus is removed , a stimulus identical to the sample is presented for 0 . 5s ( match ) . In non—match trials , after the delay period ( 0 . 7s ) , a stimulus different from the sample is presented for 0 . 5s ( non—match ) . In protocols with intervening non-repeated stimuli , sample presentation is followed , after a delay period , by either 1 or 2 -different- distractors and then by a test stimulus , which could be either a match or a non-match to the sample . In protocols with intervening repeated stimuli -i . e . the ‘ABBA’ type of trials- the sample is followed , after a delay period , by 2 repeating distractors and by a final match to the sample . Stimulus and delay period durations are kept the same as in “simple” match/non-match trials . As in [30] , we used the following index as a measure of selectivity S = 1 - A 1 - 1 n where A = ( ∑ i n ν i / n ) 2 ∑ i n ( ν i 2 / n ) and n is the number of stimuli , νi are the mean firing rates to a set of stimuli . S takes values between 0 and 1 , so that S = 0 when ν i = ν for all i ( A = 1 ) S = 1 when ν i = ν and ν j ≠ i = 0 ( A = 1 / n ) | Over short time periods , memories are stored by sustained patterns of spiking activity which , once initiated by the stimulus , persist over the entire retention interval . How the information stored by such persistent activity is later retrieved is presently unclear . Here we propose that , besides temporarily storing memories , persistent activity is also instrumental in their retrieval by transiently modifying the tuning properties of the underlying neuronal networks . We show that the mechanism proposed parsimoniously recapitulates the extensive experimental phenomenology on match effects observed in delayed-response tasks , where the information held in memory has to be compared with incoming , sensory-related information to act appropriately . The theory makes very specific , straightforwardly testable predictions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| []
| 2015 | Modulation of Network Excitability by Persistent Activity: How Working Memory Affects the Response to Incoming Stimuli |
Chikungunya virus ( CHIKV ) is a mosquito-borne virus currently transmitted in about 60 countries . CHIKV causes acute flu-like symptoms and in many cases prolonged musculoskeletal and joint pain . Detection of the infection is mostly done using RT-RCR or ELISA , which are not suitable for point-of-care diagnosis . In this study , a reverse transcription recombinase polymerase amplification ( RT-RPA ) assay for the detection of the CHIKV was developed . The assay sensitivity , specificity , and cross-reactivity were tested . CHIKV RT-RPA assay detected down to 80 genome copies/reaction in a maximum of 15 minutes . It successfully identified 18 isolates representing the three CHIKV genotypes . No cross-reactivity was detected to other alphaviruses and arboviruses except O'nyong'nyong virus , which could be differentiated by a modified RPA primer pair . Seventy-eight samples were screened both by RT-RPA and real-time RT-PCR . The diagnostic sensitivity and specificity of the CHIKV RT-RPA assay were determined at 100% . The developed RT-RPA assay represents a promising method for the molecular detection of CHIKV at point of need .
Chikungunya virus ( CHIKV ) is a mosquito-borne virus belonging to the genus alphavirus , family Togaviridae and spread by Aedes mosquitoes . CHIKV is the aetiological agent of chikungunya fever ( CF ) and was first isolated in 1952 from the serum of a febrile patient during an outbreak in Tanzania , Africa [1] . CHIKV infection is characterized by an abrupt onset of fever lasting two to five days , frequently accompanied by arthralgia . The disease is usually self-limiting , but joint pain symptoms can persists for weeks up to years [2–4] . Other common symptoms include muscle pain , headache , nausea , fatigue and rash similar to the dengue virus infection . Besides CHIKV , the Alphavirus genus includes viruses such as O'nyong'nyong virus ( ONNV ) , Ross River virus ( RRV ) , Barmah Forest virus ( BFV ) , Sindbis virus ( SNV ) and Mayaro virus ( MAYV ) [5] . Since its discovery in Africa , CHIKV has repeatedly caused outbreaks in Africa , India , Southeast Asia , the Middle East and Europe with irregular intervals [6–10] . Phylogenetic analysis of CHIKV showed that the virus has evolved into three distinct genotypes: Asian , West African and Eastern/Central /South African ( ECSA ) [11] . A single-base mutation E1-A226V in a strain of the ECSA genotype enhanced replication of the virus in Ae . albopictus , and led to a large-scale epidemic on La Reunion in 2005 [12] . This ECSA genotype was subsequently associated with epidemics in the Indian Ocean region , and the Asian genotype has been associated with recent outbreaks in the Pacific region [13] . The first local transmission of CF in the Americas was reported from the Caribbean islands in December 2013 . The local transmission of the disease has been reported in 45 countries or territories throughout world with more than 1 . 7 million suspected cases ( CCDR October 20 , 2015 , http://www . cdc . gov/chikungunya/geo/index . html ) . It appears to have been introduced twice to Brazil , once from the Pacific region and once from Africa [14] . Current diagnosis of CHIKV is based on three main laboratory methods: virus isolation , reverse transcription-polymerase chain reaction ( RT-PCR ) and serological tests such as plaque reduction neutralizing test ( PRNT ) , enzyme-linked immunosorbent assays ( ELISA ) or immunofluorescence test ( IFT ) . Commonly , blood and serum samples are used as specimens for CHIKV diagnosis . Depending on the type of sample and the time of sample collection relative to symptoms ( acute or convalescent phase of disease ) , an appropriate diagnostic method is applied to the samples [15] . A pronounced viraemia of up to 109 viral genomes can be observed mainly on days 1–6 after onset of disease and in some cases for longer [16] , therefore , virus isolation and RT-PCR are performed on acute phase specimens collected during the first week after onset of symptoms . Several RT-PCR assays have been published for the detection of CHIKV RNA in clinical specimens and mosquito samples [17–23] . Real-time RT-PCR based assays are suitable for clinical diagnosis due to the closed tube assay format , the option for quantification of viral load , high sensitivity and specificity . Serological tests are applied to either acute or convalescent phase samples for the detection of IgM and IgG anti-CHIKV antibodies . Serological diagnosis is confirmed by direct detection of IgM anti-CHIKV antibodies or by determining a four-fold increase in CHIKV-specific antibody titers in acute and convalescent samples by ELISA , IFT or PRNT tests [15] . Since the clinical picture of CHIKV is similar to that of DENV and Zika virus , a simple and rapid molecular assay would be needed to select the best treatment approach . The recombinase polymerase amplification ( RPA ) assay utilizes enzymes and proteins in order to allow the amplification of the DNA at a constant temperature ( 38–42°C ) [24] . The presence of the amplicon is detected via the exo-probe , which include an internal abasic site mimic ( tetrahydrofuran , THF ) flanking fluorophore and Quencher as well as pathogen-specific 30 and 15 bases at both the 5´ and 3´ prime ends , respectively . Upon the hybridization of the exo-probe to the complementary sequence the Exonuclease III cleaves at the THF site and the fluorescence signal is generated . RPA assay was successfully developed for the detection of DENV and YFV as well as other human and veterinary pathogens [25–32] . Moreover , a mobile suitcase laboratory was established for allowing the deployment of the RPA in the field for identifying Ebola virus infected case [33] . In this study , we developed and evaluated a reverse-transcription recombinase polymerase amplification ( RT-RPA ) assay as potential point-of-care ( POC ) diagnostic tool for rapid detection of CHIKV . The RT-RPA assay was designed by targeting the non-structure protein 1 ( nsP1 ) . The sensitivity and specificity of the method was evaluated using strains of the three genotypes of CHIKV and compared to reference RT-PCR methods . Finally , the performance of the RT-RPA assay was evaluated on acute-phase serum samples for clinical diagnosis of CHIK .
In total , the study included 78 patients with a history of sudden onset of fever , headache , fatigue , nausea , vomiting , rash , myalgia and severe and very painful polyarthralgia suggestive of CHIK infection . Fifty-eight plasma samples were provided by the Department of Microbiology , Faculty of Science , Mahidol University , Bangkok , Thailand ( Ethical approval ID: COA . No . MU-IRB 2010/251 . 3108 ) . Twenty CHIKV positive sera samples collected from patients suspected to be infected by CHIKV during routine medical examination , were provided by French National Reference Center for Arboviruses , Marseille , France . Patients had given oral consent according to the national ethical regulations . All used samples were handled anonymously . All the different CHIKV virus strains ( Table 1 ) used in this study were derived from cell culture and kindly provided by the European Network for Diagnostics of Imported Viral Diseases ( ENIVD ) ; the Bernard-Nocht Institute in Hamburg , Germany; French National Reference Center for Arboviruses , Marseille , France and the Pasteur Institute of Dakar in Senegal . The Robert-Koch Institute , Berlin in Germany and the Pasteur Institute de Dakar in Senegal provided different flavivviruses , alphaviruses and Rift Valley Fever Virus ( Table 2 ) for cross reactivity testing . Viral RNA was isolated from 140 μl aliquots of cell culture supernatants or serum sample , using the QIAamp Viral Mini Kit ( QIAGEN , Hilden , Germany ) according to the manufacturer's instructions . RNA was eluted in 60–100 μl of elution buffer and stored at -80°C until further use . To identify the most conserved regions in the CHIKV genome for primer and probe design of the RPA assay , we first analyzed an alignment of fifty CHIKV full genome sequences from the RNA virus database ( http://bioafrica . mrc . ac . za/rnavirusdb/ ) using GENEIOUS [34] . The nsP1 gene region from nucleotide position 131–306 nt was chosen for final RT-RPA assay design . In total , 196 Genbank CHIKV sequences containing the conserved nsP1 region were re-aligned using GENEIOUS . Based on the alignment , two forward primers , two reverse primers and a single probe ( Table 3 ) covering all known variants of CHIKV sequences were selected . In-vitro RNA was synthesized using the RiboMAX Large Scale RNA Production System-T7 ( Promega , Mannheim , Germany ) according to the manufacturer´s instructions and quantified using a Nanodrop ND-1000 spectrophotometer ( Thermo Scientific , Dreieich , Germany ) . CHIKV-genomic RNA was tested and quantified by an in-house real-time RT-PCR targeting the E gene using in vitro transcribed RNA standards as described previously [35] . The real-time RT-PCR assays is a CHIKV group-specific assay and was carried out in a one-step format using an ABI 7500 real-time PCR system and the AgPath-ID One-Step RT-PCR Kit . To quantify CHIKV genomic RNA copies , ten-fold serial dilutions of the standard in vitro transcribed RNA ranging from 5 to 5 x 105 RNA copies/reaction were tested within the same sample run . The sensitivity of this real-time RT-PCR was determined at nine RNA copies in 95% of the cases by probit regression . Additionally , a second real-time RT-PCR assay targeting the E gene based on 149 CHIKV sequences was used [36] . It was performed using the LightCycler 480 RNA Master Hydrolysis Probes ( Roche , Mannheim , Germany ) on the LightCycler 2 . 0 and the following temperature profile: RT at 63°C for 3 min , activation at 95°C for 30 sec and 45 cycles of 2-steps PCR at 95°C for 5 sec and 60°C for 15 sec . This real-time RT-PCR had an analytical sensitivity of 39 RNA molecules detected by performing the probit analysis . The real-time RT-RPA assay was performed in 50 μl reaction volume using the TwistAmp RT exo kit ( TwistDx , Cambridge , UK ) which provides all enzymes and reagents necessary for the reverse transcription step and DNA amplification in lyophilized pellets according to the manufacturer’s instruction . Briefly , 29 . 5 μl of rehydration solution were mixed with 7 . 2 μl of ddH2O , 420 nM of each primer and 210 nM of target specific RPA exo-probe . A total of 41 . 5 μl of this master mix was dispensed into each tube of the eight-tubes strip and 5 μl of RNA template was added to the master mix and mixed . Finally 46 . 5 μl of master mix/template solution was transferred to each lyophilized RPA pellet of the eight-tubes strip provided in the kit . For each sample , 3 . 5 μl magnesium acetate ( 280 mM ) was added into the lid of the reaction tubes and the tubes were closed carefully . Through centrifugation of tubes , the magnesium acetate was dropped into each reaction simultaneously to trigger the RT-RPA reaction . The reaction tubes were vortexed , centrifuged and then placed in the ESE Quant Tubescanner ( Qiagen Lake Constance GmbH , Stockach , Germany ) for real-time monitoring of fluorescence . The reaction was performed at 39°C for 15 minutes , with brief mixing and centrifugation of reaction tubes after four minutes . The resulting amplification curves were analyzed by ESEQuant Tube Scanner software Version 1 . 0 and threshold values were determined by slope validation i . e . slope ( mV/min ) values were compared in order to distinguish positive from negative results . A dilution range between 107 and 101 of the nsP1 RNA molecular standard was used to determine the CHIKV RPA assay sensitivity for two RPA primer pair combinations ( RF+RR3 or RF2+RR2 , Table 3 ) . A probit regression analysis was performed on eight RPA runs of each combination using STATISTICA ( StatSoft , Hamburg , Germany ) . Additionally , RNA extracts were prepared from 10-fold serial dilutions of two CHIKV strains culture supernatant . RNA extracts containing low viral RNA load were prepared in 5-fold serial dilution to determine the limit of detection ( LOD ) . Aliquots of these RNA extracts were stored at -80°C until further use . The sensitivity of the RF+RR3 combination was evaluated by testing RNA extracts simultaneously on CHIKV RT-RPA and real-time RT-PCR assays in order to compare results and determine the LOD of the RT-RPA assay . Additionally a panel of 18 different CHIKV strains including strains from the current CHIKV outbreaks representing all known three genotypes: Asian , West African and Eastern/Central /South African ( ECSA ) ( Table 1 ) were tested . In total , twenty-two viruses including eleven alphaviruses , ten flaviviruses and one phlebovirus were used to determine the assay cross reactivity ( Table 2 ) . Twenty CHIKV positive sera and 58 plasma samples with suspected CHIKV infection were tested by real-time RT-PCR and RT-RPA assays to evaluate diagnostic specificity and sensitivity of the assay . RNA was extracted from these clinical samples using the QIAamp Viral RNA Mini Kit and 5 μl of eluated RNA was tested in each assay .
For the development of the CHIKV-specific RT-RPA assay , two forward primers , two reverse primers and one RPA probe were designed in the conserved nsP1 target region . To identify the best primer combination , a total of four combinations of were tested on RNA sample of the CHIKV LR strain . Fig 1 illustrates the results of testing the primer combinations and indicates that all four primer combinations were able to amplify CHIKV RNA efficiently within a maximum of six minutes . However , the primer combination CHIKV RF/RR3 resulted in an earlier amplification starting point ( after 2 . 7 minutes ) with s higher fluorescence intensity than the other three combinations ( Fig 1 ) . The analytical sensitivity of both the RF+RR3 and RF2+RR2 combinations were determined using the in vitro transcribed RNA standard ( n = 8 ) ( S1 Fig ) . The calculated sensitivity at 95% probability was 23 and 4310 RNA molecules/reaction for RF+RR3 and RF2+RR2 , respectively ( Fig 2 ) . RNA extracts of a 10-fold serial dilution of two CHIKV strains culture supernatant ( LR and IN ) were tested by CHIKV RT-RPA using the RF+RR3 primers and real-time PCR to determine the analytical sensitivity . Comparative data are plotted in Fig 3 . The results indicated that both the RT-RPA and the real-time PCR detected RNA linearly over 5 log10 -steps of a serial dilution of viral RNA . Similar results were also obtained with the CHIKV IN strain ( S1 Table ) . To determine the LOD of the RT-RPA assay more precisely , six replicates of the RNA extracts containing low viral RNA loads as determined with real-time PCR , were diluted in a 5-fold dilution series in the range of the LOD calculated by probit analysis for the RT-RPA using in vitro transcribed RNA . Tested by RT-RPA , 6 of 6 ( 100% ) RT-RPA replicates were amplified at a dilution 10−5 containing 361 and 467 RNA genome equivalents ( GC ) /reaction ( rxn ) for the LR and the IN strain , respectively . At the dilution of 10−6 containing 80 and 92 GC/rxn for the LR and the IN strain , respectively , 5 of 6 replicates were amplified by RT-RPA assay . At the lowest dilution of 10−7 containing 10 GC/rxn only 2 of 6 and 3 of 6 replicates were detected for both the LR and the IN strain . Overall the probit detection limit in 95% of cases for real-time RT-PCR was 10 GC/rxn , whereas the LOD of RT-RPA was 80 GC/rxn . The sensitivity of the RT-RPA assay was assessed by testing 18 CHIKV strains representing all three genotype ( Table 1 ) , the specificity was tested with eleven different alphaviruses , ten flaviviruses and one phlebovirus ( Table 2 ) . The CHIKV RT-RPA assay utilizing RF+RR3 efficiently detected all 18 CHIKV strains and the O'nyong'nyong virus ( ONNV ) , while RT-RPA assay using RF2+RR2 amplified only the CHIKV strains . There was no cross reactivity found to other viruses of the cross reactivity panel ( Table 2 and S2 Fig ) . The diagnostic sensitivity and specificity of the CHIKV RT-RPA assay was assessed with plasma samples from 58 suspect CF cases and compared to two CHIKV specific real-time RT-PCR tests during a field trial in Bangkok , Thailand . Both CHIKV real-time RT-PCR detected 36 out of 58 sample ( 62% ) positive . The Ct values obtained by real-time RT-PCR for positive samples ranged from 20 . 19 to 36 . 02 ( 1 . 6x104 -1x108 GC/rxn ) . In comparison to real-time RT-PCR , RT-RPA correctly identified all 36 positive samples and did not detect the 22 negative samples with 100% sensitivity and specificity ( PPV: 1 , NPV: 1 , Fig 4 ) . Additionally , we tested 20 sera from acute CHIK patients from France . All three methods efficiently detected 20 out of 20 CHIKV positive samples .
CHIKV is treated symptomatically by use of painkillers and anti-inflammatory drugs . As there is no effective specific antiviral drug and currently only experimental CHIKV vaccine development . Laboratory diagnosis of CHIKV is very important for effective outbreak management including clinical management and vector control . Therefore , there is a need for a reliable , rapid and portable diagnostic method , which can be deployed in the field during outbreaks . In this study , we developed a RT-RPA assay for rapid detection of CHIKV in clinical samples , which can be easily deployed to rural health care centers or used in field investigations . We designed a highly sensitive set of RPA primers and exo-probe , which detect CHIKV down to 80 GC/rxn . This detection limit is slightly higher than the 10 GC/rxn detected by reference real-time RT-PCR methods . However , the sensitivity of RT-RPA was sufficient to detect CHIKV infection in clinical samples from Thailand and France with an overall sensitivity and specificity of 100% in comparison to real-time RT-PCR . Previous studies described high titer viraemia exceeding 109 GC/ml in the acute phase of CF [37 , 38] . In our study of 20 CF acute samples , viral load ranged from 1 . 6x106 -1x1010 GC/ml ( 1 . 6x104 -1x108 GC/rxn ) . The RT-RPA assay detected a panel of 18 different CHIKV strains of all known three genotypes . There was no cross-reactivity of the RT-RPA assay with tested common alpha- and arboviruses except for ONNV detected with the primer combination ( RF+RR3 ) , while the combination RF2+RR2 did not detect the ONNV ( S2 Fig ) . Cross reactivity of CHIKV RT-RPA assay to ONNV could be due to 85% similarity of the genomes of ONNV to CHIKV . RT-RPA primers RF/RR3 harbour four mismatches in RF and seven mismatches in RR3 to the ONNV sequences ( S3 Fig ) . This seems to reflect the fact that RPA assays have been shown to amplify target genes in the presence of up to nine mismatches [25 , 30] . Moreover , the position of the mismatch did not influence the amplification step as in the real-time PCR [29] . However , we chose primer pair RF/RR3 due to its faster amplification ( Tt: 3 . 3 ) and highly analytical sensitivity ( 23 RNA copies ) in comparison to the more specific primer pair RF2/RR2 ( Tt: 3 . 7 and analytical sensitivity: 4310 ) . If necessary , the latter can be used for the differentiation between CHIKV and ONNV e . g . when used in ONNV endemic regions of Africa whereas the former can be used in Asia , Europe and the Americas without any issue . The RT-RPA assay using RF/RR3 demonstrated a sensitivity of 100% on CHIKV samples of the recent external quality assessment ( EQA ) study for molecular diagnosis but also detected one sample containing ONNV . Both real-time RT-PCR tests used in this study showed no cross reactivity to ONNV which indicates superior specificity compared to commercial PCR methods tested in the EQA study , in which 46 . 2% showed cross reactivity to ONNV [39] . Two loop-mediated isothermal amplification ( LAMP ) assays for the detection of CHIKV have been published [40 , 41] . The LAMP assay require at least four primers and six binding sites and results can be usually observed visually after 30–60 minutes [42] . Recent LAMP assay developments begin to show shorter run times [43] . In contrast , the RPA assay is much faster ( 3–15 minutes ) and utilized two primers and one probe i . e . three binding sites . In addition , RPA reagents are provided in a lyophilized pellet stable at ambient temperature ( 25–38°C ) [26 , 28 , 33] , which allow independence from the cooling chain . The RPA is a promising technology to perform molecular assay at point of need . Nevertheless , the RPA primer and probe design is still challenging , as dozens of primers combinations must be tested in order to select a functional one . The current RPA assay protocol requires four pipetting steps , which is still less than the real-time PCR method but further development to miniaturize the assay is required . The cost of test per sample is around 5 USD . Lowering the cost to 1 USD will maximum its use in the affected countries . The CHIKV RPA assay presented here is a promising tool for CHIKV diagnostics at the point of need . Integration into a multimer or multiplex assay for simultaneous and differential detection of CHIKV , Dengue virus and Zika virus as well as an internal positive control would improve outbreak investigations , since the three viruses induce the same clinical picture upon infection and increasingly co-circulate in many parts of the world . Furthermore , combination with a simple extraction method for allowing isolation of virus or virus-infected cells from the whole blood and simple lysis protocol will maximize its employment at low resource settings . | CHIKV is transmitted to humans via mosquitos . CHIKV induces clinical signs similar to Influenza , Dengue , and Zika viruses . We have developed a molecular assay for the detection of CHIKV genome based on isothermal„recombinase polymerase amplification ( RPA ) assay”performed at 42°C . The result was obtained in maximum of 15 minutes , which is 4–6 times faster than the current molecular diagnostic techniques . Our CHIKV RPA assay is rapid and sensitive , as well as easy to use at the point of need . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"reverse",
"transcriptase-polymerase",
"chain",
"reaction",
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"chikungunya",
"infection",
"pathogens",
"immunology",
"tropical",
"diseases",
"microbiology",
"rna",
"extraction",
"alphaviruses",
"viruses",
"chikungunya",
"virus",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"molecular",
"biology",
"techniques",
"cross",
"reactivity",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"artificial",
"gene",
"amplification",
"and",
"extension",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"flaviviruses",
"polymerase",
"chain",
"reaction",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"o'nyong'nyong",
"virus",
"organisms"
]
| 2016 | A Field-Deployable Reverse Transcription Recombinase Polymerase Amplification Assay for Rapid Detection of the Chikungunya Virus |
Rickettsia bacteria are responsible for diseases in humans and animals around the world , however few details are available regarding its ecology and circulation among wild animals and human populations at high transmission risk in Brazil . The aim of this study was to investigate the occurrence of ticks and Rickettsia spp . in wild boars , corresponding hunting dogs and hunters . Serum samples and ticks were collected from 80 free-range wild boars , 170 hunting dogs and 34 hunters from southern and central-western Brazil , from the Atlantic Forest and Cerrado biomes , respectively , between 2016 and 2018 . Serum samples were tested by indirect immunofluorescent-antibody assay ( IFA ) to detect IgG antibodies against Rickettsia rickettsii , Rickettsia parkeri , Rickettsia bellii , Rickettsia rhipicephali and Rickettsia amblyommatis . Tick species were identified by morphological taxonomic keys , as previously described . A total of 164 ticks including A . sculptum , A . brasiliense and A . aureolatum were tested in PCR assays for Spotted Fever Group ( SFG ) Rickettsia spp . A total of 58/80 ( 72 . 5% ) wild boars , 24/170 ( 14 . 1% ) hunting dogs and 5/34 ( 14 . 7% ) hunters were positive ( titers ≥ 64 ) to at least one Rickettsia species . A total of 669/1 , 584 ( 42 . 2% ) ticks from wild boars were identified as Amblyomma sculptum , 910/1 , 584 ( 57 . 4% ) as Amblyomma brasiliense , 4/1 , 584 ( 0 . 24% ) larvae of Amblyomma spp . and 1/1 , 584 ( 0 . 06% ) nymph as Amblyolmma dubitatum . All 9 ticks found on hunting dogs were identified as Amblyomma aureolatum and all 22 ticks on hunters as A . sculptum . No tested tick was positive by standard PCR to SFG Rickettsia spp . The present study was the concomitant report of wild boar , hunting dog and hunter exposure to SFG rickettsiae agents , performed in two different Brazilian biomes . Wild boar hunting may increase the risk of human exposure and consequently tick-borne disease Wild boars may be carrying and spreading capybara ticks from their original habitats to other ecosystems . Further studies can be required to explore the ability of wild boars to infecting ticks and be part of transmission cycle of Rickettsia spp .
The genus Rickettsia ( family Rickettsiaceae; order Rickettsiales ) comprises gram-negative and obligate intracellular bacteria , which are phylogenetically classified into the spotted fever group ( SFG ) rickettsiae , the typhus group rickettsiae , the Rickettsia bellii group rickettsiae and the Rickettsia canadensis group rickettsiae [1] . Tick-borne rickettsioses have been placed into the SFG group , known of causing infection in animals and human beings [2 , 3] , and participating on enzootic or epizootic cycles among vertebrates and arthropod vectors [4] . Ixodid ticks have been described as the main natural reservoirs and vectors of rickettsiae , with transstadial and transovarial transmission in ticks [5] . Rickettsia rickettsii , the main etiological agent of spotted fever in Brazil , has been primarily transmitted to human beings by Amblyomma sculptum and Amblyomma aureolatum ticks [6 , 7 , 8] . Amblyomma sculptum , characterized by an aggressive behavior and multispecies parasitism , may be the most prevalent tick species in the Cerrado and degraded areas of the Atlantic Rainforest biomes [9 , 10] . On the other hand , A . aureolatum ticks have been mostly found in Atlantic Rainforest fragments , which may provide favorable abiotic conditions and native carnivores as primary hosts [8] . Wild boars ( Sus scrofa ) have been classified by Brazilian laws as exotic invasive species originated by Eurasian wild boars and their hybrids , with nationwide hunting officially permitted ( Normative Instruction 03/2013 ) as a strategy for population control and eradication [11] . Wild boars may invade natural and anthropic areas , not only competing for resources with native wildlife and livestock species , but also sustaining life cycle of ticks and tick-borne diseases [12] . As large-bodied , non-native and the most invasive mammal species , wild boars have been considered as potential hosts of A . sculptum ticks in Brazilian biomes , particularly the Pantanal floodplains [13 , 14] . Hunting dogs ( Canis familiaris ) have been the most popular method for wild boar tracking and hunting in Brazil [15] . Brazilian rural dogs accessing natural areas have been frequently found to show parasitism for A . aureolatum ticks along with antibodies for Rickettsia spp . , potentially increasing the risk of human infection when bringing infected ticks to household environment [16 , 17 , 18] . Density population of capybaras ( Hydrochoerus hydrochaeris ) in spotted fever-endemic areas of southeastern Brazil , mostly related to sugarcane crops production [19] , has been 40 times higher than those reported in natural environments [20] . Similarly , wild boar populations have also been associated to several cultivated areas of central-western , southwestern and southern Brazil [21] . Hence , it is reasonable to speculate that the overlapping of wild boar and capybara environments may have a synergic impact on occurrence of ticks and tick-borne diseases . Despite wild boars , hunting dogs and hunters in Brazil may be exposed to several tick-borne rickettsiae , no study to date has concurrently assessed this potential and alternative life cycle of spotted fever in wild boars , hunting dogs and hunters . Accordingly , the aim of the present work was to determine anti-Rickettsia antibodies and presence of ticks in wild boars , hunting dogs and hunters in two different Brazilian biomes ( Atlantic Forest and Cerrado ) .
This is a descriptive cross-sectional study of boars , hunting dogs , hunters and ticks parasitizing them . The study was conducted in preserved and degraded areas in the Atlantic Forest biome of southern Brazil , including the Vila Velha State Park ( belongs to Campos Gerais National Park ) and Palmeira , Curitiba , Castro , Ponta Grossa , Porto Amazonas and Teixeira Soares municipalities; and in degraded areas in the Cerrado biome of central-western Brazil , the Aporé ( Fig 1 ) . A total of 22 on-field expeditions were carried out from November 2016 to May 2018 , which included summer , autumn , winter and spring . Ticks were collected from wild boars , hunting dogs and hunters during all year seasons , which may have covered all possible species and stages . Wild boars blood samples were collected by intracardiac puncture immediately after death , by jugular puncture in dogs and by cephalic puncture in hunters . All samples were collected in tubes without anti-coagulant and kept at room temperature ( 25 °C ) until visible clot retraction , centrifuged at 1 , 500 revolutions per minute for five minutes , and serum separated and kept at -20 °C until processing . Wild boars were sampled in agricultural areas of Atlantic Forest and Cerrado biomes following legal hunting laws , along hunting dogs and hunters . Additionally , wild boars at the Vila Velha State Park ( belongs to Campos Gerais National Park ) were baited , photo-monitored , trapped and euthanized . Both hunting and trapping , along with handling of wild boar samples and ticks were authorized by the Brazilian Environmental Biodiversity System ( SISBIO license 61805–2 ) . Tick sampling of each wild boar was randomly obtained by time-independent collection , with ticks picked on all surfaces of the two body sides to ensure maximum yield . After such hunting activities , resting and blood samplings , dogs were carefully examined for ticks and hunters asked for self-examination for tick presence . All ticks obtained from wild boars , hunting dogs and hunters were collected , preserved in isopropyl alcohol and taken to the laboratory for taxonomic identification , which was performed following standard morphological keys [22 , 23 , 24] . Hunting dogs underwent annual deworming protocols , along flea and tick control according to visual infestation , done by their owners . Serum samples were individually tested by indirect immunofluorescent-antibody assay ( IFA ) for five Brazilian Rickettsia isolates: R . rickettsii strain Taiaçu , R . parkeri strain At24 , R . amblyommatis strain Ac37 , R . rhipicephali strain HJ5 and R . bellii strain CL as previously described [25 , 26] . Individual sera were initially screened at a 1:64 dilution against each of the rickettsial antigens . A fluorescein isothiocyanate-labeled rabbit anti-pig IgG dilution 1: 1 , 500 ( IgG , Sigma Diagnostics , St . Louis , MO , lot 048K4842 ) as conjugate was used for hunting wild boars samples , fluorescein isothiocyanate-labeled rabbit anti-dog IgG dilution 1:1 , 000 ( IgG , Sigma Diagnostics , St . Louis , MO , lot 102M4795V ) was used as conjugate for the hunting dogs samples , and fluorescein isothiocyanate-labeled rabbit anti-human IgG dilution 1:1 , 500 ( IgG , Sigma Diagnostics , St . Louis , MO , lot 038K4802 ) as conjugate was used for the hunter samples . In each slide , a serum previously shown to be non-reactive ( negative control ) and a known reactive serum ( positive control ) were tested up to the 1:64 dilution . In case of a positive reaction of testing serum , serial dilutions at two-fold increments were tested up to the endpoint titer . Serum showing for a Rickettsia species titer at least fourfold higher than those observed for the remaining Rickettsia species was considered possibly homologous to the first Rickettsia species , as previously determined [25 , 26] . A sample of 164 ticks was randomly selected , individually submitted to DNA extraction by the guanidine isothiocyanate technique [27] , and individually tested by standard PCR for tick mitochondrial 16S rRNA [28] and rickettsial gltA gene [29] . For each PCR run , a negative control ( water ) and positive control ( Rickettsia vini DNA ) were included [30] . This study has been approved by the Ethics Committee of Animal Use ( protocol number 059/2017 ) of the Federal University of Paraná , officially included as part of the annual activities of the City Secretary of Health at Ponta Grossa and approved by National Human Ethics Research Committee ( number 97639017 . 7 . 0000 . 0102 ) . In addition , the in-park trapping and tick collection have been authorized by the Environment Institute of Paraná ( authorization number 30/17 ) and by Chico Mendes Institute of Biology ( authorization number 61805–2 ) . The absolute and relative frequency of infection was calculated stratifying the observations according to the species and to the region in the country in which samples were collected . The frequency of Rickettsia spp . between different species was compared using chi-square test . Observed differences were considered to be significant when the resulting P-value was less than 0 . 05 . A map illustrating the sampling points was constructed using QGIS 2 . 18 . 18 .
Blood samples were collected , and ticks searched from 80 wild boars , 170 hunting dogs and 34 hunters . Samples from 60/80 ( 75 . 0% ) wild boars were obtained by legal hunting ( agricultural areas ) , while 20/80 ( 25 . 0% ) by trapping ( conservation unit area ) . Among hunting individuals , 24/60 ( 40 . 0% ) wild boars , 147/170 ( 86 . 5% ) hunting dogs and 27/34 ( 79 . 4% ) hunters were sampled at the Atlantic Forest biome , while 36/60 ( 60 . 0% ) wild boars , 23/170 ( 13 . 5% ) hunting dogs and 7/34 ( 20 . 6% ) hunters at the Cerrado biome . Through serologic analysis for Rickettsia spp . , 58/80 ( 72 . 5% ) wild boars , 24/170 ( 14 . 1% ) hunting dogs , and 5/34 ( 14 , 7% ) hunters were seropositive for Rickettsia spp . ( Table 1 ) . In addition , possible antigen involved in a homologous reaction ( PAIHR ) for R . rickettsii , R . bellii or R . rhipicephali were found in 4/80 ( 5 . 0% ) wild boars , R . bellii and R . amblyommatis in 2/170 ( 1 . 17% ) hunting dogs ( Table 1 ) . Among wild boars , IFA endpoint titers varied from 64 to 1 , 024 for R . rickettsii and R . bellii , 64 to 512 for R . parkeri and R . rhipicephali , and 64 to 256 for R . amblyommatis . IFA endpoint titers in hunting dog samples varied from 64 to 512 for R . rickettsii , R . bellii , R . rhipicephali , R . amblyommatis , 64 to 256 for R . rickettsii , and 128 to 1 , 024 for R . parkeri . Among hunters , IFA endpoint titers varied from 128 to 256 for R . rickettsii , 64 to 256 for R . parkeri , 64 to 128 for R . bellii , and 64 to 512 for R . rhipicephali and R . amblyommatis . Seropositivity for Rickettsia spp . was higher in wild boars when compared to dogs ( p-value = 0 . 001 ) and humans ( p-value = 0 . 001 ) but was similar between dogs and humans ( p-value = 1 . 000 ) . Despite Rickettsia spp . prevalence was statistically higher in southern than central-western Brazil for wild boars ( p-value = 0 . 002 ) , no significance was observed in hunting dogs ( p-value = 1 . 000 ) and hunters ( p-value = 1 . 000 ) . Ticks were collected from wild boars , hunting dogs and hunters during all year seasons , covering all possible species and stages . A total of 1 , 584 ticks were collected from wild boars , including 669 ( 42 . 2% ) adults of A . sculptum , 910 ( 57 . 4% ) Amblyomma brasiliense composed by 870 ( 54 . 9% ) adults and 40 ( 2 . 5% ) nymphs , 4 ( 0 . 24% ) larvae of Amblyomma spp . and one ( 0 . 06% ) nymph of Amblyomma dubitatum . All 9 ticks founded on hunting dogs were identified as A . aureolatum adults , and all 22 ticks obtained from the hunters as A . sculptum nymphs ( Table 2 ) . In addition , 24/44 ( 54 . 5% ) and 8/36 ( 22 . 2% ) wild boars had an average infestation of 32 . 7 and 81 . 5 ticks per animal in southern and central-western Brazil , respectively . Amblyomma sculptum was the dominant tick species infesting central-western wild boars , whereas A . brasiliense was so in southern wild boars . All A . aureolatum-infested dogs were from the southern region . A total of 164/1 , 584 ( 10 . 4% ) ticks , including 162 adults and 2 nymphal , were randomly selected for the detection of SFG rickettsial DNA by PCR . They belonged to one genus including 3 species: 4 A . sculptum from 2/44 ( 4 . 5% ) wild boars of southern Brazil , 53 A . sculptum from 8/36 ( 22 . 2% ) wild boars of central-western Brazil , 100 A . brasiliense from 24/44 ( 54 . 58% ) wild boars of southern Brazil and 7/147 ( 4 . 8% ) A . aureolatum from hunting dogs of southern Brazil ( Table 2 ) . No rickettsial DNA was detected in these ticks , despite of each of them yielded a visible amplicon in agarose gel through the PCR targeting the tick 16S rRNA gene .
The present study reports serological findings and molecular assays of Rickettsia spp and ticks of wild boars , simultaneous to their correspondent hunting dogs and hunters . Seropositivity for Rickettsia spp . was higher in wild boars when compared to dogs and humans but was similar between dogs and humans . Despite results have apparently shown a higher seropositivity of hunting dogs and hunters in southern than in central-western Brazil , differences were not statistically significant probably due to a reduced statistical power between the prevalence of groups formed by stratification according to region . Since this was not the aim of the present study , further studies should be conducted to fully establish differences on serological rickettsial titers of wild boars , hunting dogs and hunters among different Brazilian regions . The difference of seropositivity between southern and central-western wild boars could be related to dominant tick species , namely A . sculptum in central-western and A . brasiliense in southern Brazilian regions . Serological results herein may indicate that , if the A . sculptum populations infesting wild boars , dogs and hunters in central-western Brazil were infected by any SFG pathogenic rickettsiae , the infection rate would be very low or only few populations would be infected . In fact , the low rickettsial seropositivity in central-western wild boars , hunting dogs and hunters could be a result of the rare rickettsial infection in A . sculptum ticks [31 , 32] . The only exceptions may be the spotted fever endemic areas of southeastern Brazil , where some populations of this tick species may be infected by R . rickettsii [33 , 34] . Thus , all A . sculptum tested were negative for Rickettsia spp . in molecular analyses . On the other hand , the much higher seropositivity of wild boars , hunting dogs and even hunters in southern Brazil may suggest that the A . brasiliense populations from this region would be infected by one or more SFG rickettsiae , yet to be identified in further studies . To the best of our knowledge , no rickettsial agent has been identified in A . brasiliense yet . Wild boars have been suggested to play a potential role in the eco-epidemiology of rickettsioses . In Catalonia , Spain , 12/23 ( 52 . 2% ) and 19/23 ( 82 . 6% ) wild boars sampled were seropositive to Rickettsia slovaca , classified into the SFG and associated with Dermacentor marginatus ticks [35] . In Mississippi , USA 17/58 ( 29 , 3% ) feral swine were seropositive to the SFG pathogen R . parkeri [36] . Although capybaras have long been recognized as the major host of A . sculptum and amplifier species for R . rickettsii infection in Brazil [37] , future studies should be conducted to fully establish the role of wild boars as hosts , amplifiers and their association to human cases of R . rickettsii-caused spotted fever . Despite human beings have been considered less exposed to ticks ( and therefore rickettsiae ) than animals [38] , specific human activities such as hunting may increase the risk of exposure and consequently of disease . Not surprisingly , individuals from rural areas who visit forest areas , rivers and waterfalls have also shown higher incidence of spotted fever infection [39] . Unfortunately , no information was found about hunting habits of a non-fatal human case of spotted fever illness notified in a nearby area of southern Brazil and other two cases notified in nearby area of central-western Brazil [40 , 41] , which hinders the risk assessment for this activity in regard to Rickettsia spp . transmission . Important to remark that , as mentioned before , hunting is currently unlawful in Brazil . Actually , wild boar hunting has been officially considered as “controlling non-protected invasive exotic species” , therefore the only legal regulated form of hunting activity to date in Brazil ( Normative Instruction 03/2013 ) [11] . The tick species obtained herein on wild boars have been previously involved in Rickettsia spp . transmission to dogs and human beings [10 , 33] . Association of hunting practices with seroreactivity to Rickettsia spp . has been attributed to a higher exposure to Amblyomma spp . while hunting , since these ticks have been primarily associated with wildlife in Brazil [42] . Further studies that better estimate the prevalence of infection in these populations are required to better design control strategies . Although restricted to Brazilian Pantanal biome ( floodplains ) , feral pigs , Sus scrofa L . ( Artiodactyla , Suidae ) , have been previously suggested as hosts to A . sculptum [13 , 14] . For the first time , A . sculptum ticks were found in two wild boars of subtropical climate from southern Brazil . In a recent study about the distribution of A . sculptum in Brazil , it was shown that this tick is absent from most of the southern region , possibly due to more severe winter temperatures [10] . The repeatedly findings herein of both engorged adults ( successfully fed ) and engorged nymphs ( different stages ) of A . brasiliense and A . sculptum on wild boars in the Atlantic Forest and Cerrado biomes , respectively ( Table 2 ) , have shown host adaptation and spreading to two more Brazilian biomes , suggest that these tick species might be adapting and spreading to areas previously thought as unsuitable for their survival . All ticks collected from wild boars at the conservation unit area of Atlantic Forest were identified as A . brasiliense , probably due to predominant high humidity and lower temperatures , important for this tick species development [43] , naturally maintained in such areas by native peccaries ( Tayassu spp . ) as primary hosts . However , the relative higher presence ( tick average per animal ) of adult and nymph stages in wild boars may suggest overlapping of ecological niche , and higher traveling body area as competent A . brasiliense hosts . Although A . brasiliense has been considered aggressive to human beings [44 , 45] and such scenario may impact on higher tick and tick-borne disease spreading , R . rickettsii transmission by A . brasiliense ticks have been observed only under experimental conditions [46] , and absent in molecular surveys on natural environments [17 , 47 , 48 , 49] . Not surprisingly , no A . brasiliense tested herein by standard PCR was positive to Rickettsia spp . Hunting dogs in the present study were only found with A . aureolatum , corroborating to previous studies in dogs from rural areas nearby rainforest fragments and hunter activities [50 , 51 , 52] . In a previous study in southern Brazil , 19/133 ( 14 . 3% ) rural dogs were reported with ticks , including A . aureolatum [53] . These ticks were the second most prevalent among rural dogs of another study from southern Brazil , representing 52/153 ( 33 , 9% ) of the collected ticks [54] . Amblyomma aureolatum ticks have shown high susceptibility to R . rickettsii infection , and dogs as one of the most important hosts in spotted fever-endemic areas [51] . The infection by R . rickettsii may contribute to lower survival and reproduction in A . aureolatum females , resulting in low infection rates ( <10% ) under natural conditions [7] . While this assumption could be associated to the absence of rickettsial DNA in the A . aureolatum ticks of the present study , we are aware that we have tested only a small sample of ticks , precluding a more rational conclusion . Seven hunters in the present study became infested by A . sculptum ticks after hunting . Amblyomma sculptum is the most frequent human-biting tick in Brazil , and also one of the main vectors of R . rickettsii in the country [10 , 33] . These findings highlight hunters as a potential risk group for tick-borne spotted fever in Brazil . Since Rickettsia spp usually infect and remain inside host endothelial cells , molecular detection has usually failed when investigating blood samples [55] . Under experimental R . rickettsii-infected tick infestation , rickettsial DNA has been detected by PCR in only one of 32 ( 3 . 1% ) blood samples of infected capybaras , despite serological titers up to 16 , 384 [56] . In the same study , despite serological titers up to 32 , 768 , direct intraperitoneal inoculation has failed to provide rickettsial DNA detection in blood samples . Thus , in the present study , no molecular investigation was made on blood samples of wild boars , hunting dogs and hunters . Wild boars may be carrying and spreading capybara ticks from their original habitats to other ecosystems . In Florida , USA , wild boars have been found over long distances and different ecosystem , increasing contact to multiple tick species in their preferential microhabitat [12] . Besides higher-energy requirements obtained in long distance incursions , adult wild pigs have also larger body area [57] than capybaras , which might be an important characteristic of wild boars in spreading ticks in Brazil . Altogether , such overlapping distribution of wild boars and capybaras in Brazil may lead to synergistic spreading of vector ticks , particularly of R . rickettsii-caused spotted fever , locally called as Brazilian spotted fever . Wild boars may post an additional treat due to their highly adaptative capacity , spreading themselves to both intact and degraded areas of all six Brazilian biomes , including Atlantic Forest ( rainforest ) , Cerrado ( tropical savanna ) , Pampas ( open fields ) , Pantanal ( flood plains ) , Amazon ( rainforest ) and Caatinga ( semi-arid ) , as recently recognized by the Brazilian Ministry of Agriculture ( map in S1 Fig ) [58] . As already mentioned , Brazilian Spotted fever and other rickettsial agents have reportedly overlapped capybara occurrence , therefore wild boars may carry ticks and tick-borne diseases outside capybara original areas , currently restricted to gallery forests and seasonally flooded savannas such as the Atlantic Forest , Pantanal and Cerrado [59] . In such scenario , authors hypothesize that wild boars may overspread ticks and rickettsial diseases to Brazilian biomes lacking capybaras as the Caatinga biome , a dry area found on northern , northeastern and southeastern Brazil . In addition , the Brazilian Ministry of Environment has warned about the ineffectiveness of wild boar natural population control by Brazilian native predators , mostly due to low populations of already critically endangered species as pumas ( Puma concolor ) and jaguars ( Panthera onca ) , associated to wild boar groups weighting up to 220 kg , defending themselves by sticking together and returning the attacks with potential wounds by bites and tusks [60] . Although hunting increase may be necessary to successfully control wild boar populations , authors suggest a governmental nationwide establishment of sanitary hunting guidelines , conducted always with tick-bite prevention and early recognition of rickettsial disease symptoms . The present study has shown seropositivity for at least one Rickettsia species in wild boars , hunting dogs and hunters . Despite an expected lower exposure of humans to ticks ( and therefore rickettsiae ) than animals , specific human activities such as wild boar hunting may increase the risk of human exposure and consequently tick-borne disease . Wild boars may be carrying and spreading capybara ticks from their original habitats to other ecosystems lacking capybaras , with no effective natural predators . These results may provide important findings for public health action to prevent vector-borne diseases in overlapping areas of capybaras , wild boars , hunting dogs and hunters . Further studies should be conducted to fully establish the wild boar ability to infect ticks and its role on Rickettsia spp . transmission cycle . | The present study reported serological findings and molecular assays of Rickettsia spp and ticks of wild boars , simultaneous to their correspondent hunting dogs and hunters . Seropositivity for Rickettsia spp . was higher in wild boars when compared to dogs and humans but was similar between dogs and humans . Despite Rickettsia spp . prevalence was statistically higher in southern than central-western Brazil for wild boars , no significance was observed in hunting dogs and hunters . For the first time , A . sculptum ticks were founded in wild boars from the subtropical climate of southern Brazil . Despite human beings have been considered less exposed to ticks ( and therefore rickettsiae ) than animals , specific human activities such as wild boar hunting may increase the risk of exposure and consequently tick-borne disease . Wild boars may be carrying and spreading capybara ticks from their original habitats to other ecosystems . These results may provide important findings for public action planning to prevent neglected vector-borne diseases in overlapping areas of wild boars , hunting dogs and hunters . Further studies can be required to explore the ability of wild boars to infecting ticks and be part of transmission cycle of Rickettsia spp . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"ixodes",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"geographical",
"locations",
"microbiology",
"vertebrates",
"dogs",
"animals",
"mammals",
"rickettsia",
"developmental",
"biology",
"nymphs",
"signs",
"and",
"symptoms",
"ticks",
"bacteria",
"bacterial",
"pathogens",
"infectious",
"diseases",
"rickettsia",
"rickettsii",
"south",
"america",
"ecosystems",
"medical",
"microbiology",
"microbial",
"pathogens",
"life",
"cycles",
"disease",
"vectors",
"brazil",
"arthropoda",
"people",
"and",
"places",
"arachnida",
"eukaryota",
"diagnostic",
"medicine",
"ecology",
"fevers",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"amniotes",
"organisms"
]
| 2019 | Ticks and serosurvey of anti-Rickettsia spp. antibodies in wild boars (Sus scrofa), hunting dogs and hunters of Brazil |
For the last three decades , evolutionary biologists have sought to understand which factors modulate the evolution of parasite virulence . Although theory has identified several of these modulators , their effect has seldom been analysed experimentally . We investigated the role of two such major factors—the mode of transmission , and host adaptation in response to parasite evolution—in the evolution of virulence of the plant virus Cucumber mosaic virus ( CMV ) in its natural host Arabidopsis thaliana . To do so , we serially passaged three CMV strains under strict vertical and strict horizontal transmission , alternating both modes of transmission . We quantified seed ( vertical ) transmission rate , virus accumulation , effect on plant growth and virulence of evolved and non-evolved viruses in the original plants and in plants derived after five passages of vertical transmission . Our results indicated that vertical passaging led to adaptation of the virus to greater vertical transmission , which was associated with reductions of virus accumulation and virulence . On the other hand , horizontal serial passages did not significantly modify virus accumulation and virulence . The observed increases in CMV seed transmission , and reductions in virus accumulation and virulence in vertically passaged viruses were due also to reciprocal host adaptation during vertical passages , which additionally reduced virulence and multiplication of vertically passaged viruses . This result is consistent with plant-virus co-evolution . Host adaptation to vertically passaged viruses was traded-off against reduced resistance to the non-evolved viruses . Thus , we provide evidence of the key role that the interplay between mode of transmission and host-parasite co-evolution has in determining the evolution of virulence .
Understanding which factors determine the evolution of virulence–the negative effect of parasites on host fitness [1] , [2] –and how they act , is a long-standing goal in evolutionary biology , and central for the control of infectious diseases [3] , [4] . Indeed , changes in virulence have been associated with the effects of parasites on host population dynamics [1] , [5] , the reduction of ecosystem biodiversity [6] , [7] , and the emergence and re-emergence of infectious diseases [8] , [9] . In the last three decades considerable effort has been devoted to developing theoretical models that predict conditions favouring the increase or decrease of parasite virulence . Most models are based on the hypothesis that the level of virulence is determined by trade-offs between the within-host and between-host components of the parasite's fitness; this is known as the trade-off hypothesis [2] , [10]–[12] . According to the trade-off hypothesis the level of virulence is generally , but not always [13] , [14] , a consequence of optimizing the within-host multiplication and between-host transmission components of parasite fitness [10] , [13]–[16] . Two key assumptions underlie the trade-off hypothesis . First , virulence is positively correlated with parasite multiplication within the infected host; and second , greater parasite load in an infected host increases the probability of transmission to a susceptible , uninfected host . A trade-off occurs because higher virulence may also increase host mortality , reducing the infectious period and the probability of transmission . Experimental analyses have shown that multiplication rates and/or transmission rates are positively correlated with virulence for most parasites of humans , animals and plants when transmitted horizontally , i . e . , between host individuals that are not parent and offspring [2] , [17]–[19]; [ see 20] , [ 21 for exceptions] . However , a wide range of human , animal and plant parasites that cause severe diseases , are vertically transmitted , i . e . , from parent to offspring , or are transmitted both horizontally and vertically . Evolution of virulence in these parasites may challenge the trade-off hypothesis and derived models , since the presence of an alternative mode of transmission may reduce the relative importance of the trade-off between horizontal transmission and virulence . For instance , under strict vertical transmission , virulence should be negatively correlated with transmission rate [2] , [13] , [14] , [22]: The fitness of vertically transmitted parasites is highly dependent on host reproductive potential , as hosts need to reproduce for the parasite to infect new individuals . Since virulence , by definition , reduces host fitness , vertically transmitted parasites should evolve towards lower virulence to maximize their own fitness [10] , [13] , [14] , [23]–[25] . Accordingly , the ‘continuum hypothesis’ proposes that the optimum virulence in parasites transmitted both vertically and horizontally will vary along a continuum depending on the relative weight of each transmission mode on the parasite's fitness: parasites mostly vertically transmitted will tend towards lower virulence and parasites mostly horizontally transmitted will tend towards higher virulence [14] , [16] , [22] . Despite the abundance of theory , and the numerous examples of important parasites with vertical or mixed modes of transmission , the largest fraction of experimental analyses of the effect of transmission mode on the evolution of virulence has been done under strict horizontal transmission . Studies of parasite evolution under vertical transmission have mostly reported a negative correlation between virulence and rate of vertical transmission , supporting the ‘continuum hypothesis’ [26]–[33] . However , this might not be a universal trend . For instance , in some of the few plant-parasite interactions studied , virulence was not negatively correlated with vertical transmission [34] , [35] . Therefore , understanding the relationship between mode of transmission and virulence evolution requires further analysis and is in need of more experimental data from a larger variety of host-parasite systems [36] . Most theoretical and experimental analyses of virulence evolution overlook the important fact that virulence and transmissibility are not just parasite traits [8] . Rather , virulence and transmissibility are the result of the parasite's interaction with the host , and thus are potentially subjected to host-parasite co-evolution [8] , [37] , [38] . For instance , reduced virulence might result from selection on the parasite to increase the efficiency of its vertical transmission , or from reciprocal selection on the host to reduce the damage caused by the parasite during vertical transmission [3] , [39] . Selection on the host will thus result in higher tolerance , where tolerance is defined as the host's ability to reduce the effect of infection on its fitness [8] , [40] . For example , tolerance of Arabidopsis thaliana plants to virus infection is achieved through developmental reprogramming , so that resources are reallocated from the vegetative growth to reproduction [41] . Demonstrating co-evolution presents the daunting challenge of demonstrating reciprocal effects of both host and pathogen [37] . Thus , empirical evidence for the co-evolution of host- and parasite-related components of virulence is limited [8] , [37] . Here we analyse experimentally the role of the mode of transmission , vertical or horizontal , on the evolution of virulence , while also accounting for the effect of host-parasite co-evolution in virulence-related traits . For this , we used the plant-virus system Arabidopsis thaliana L . Heynh . ( Brassicaceae ) - Cucumber mosaic virus ( CMV , Bromoviridae ) . A . thaliana ( from here on , Arabidopsis ) has been developed as model organism for molecular and genetic analyses of a wide range of plant traits , and also is increasingly used in analyses of host-parasite co-evolution [e . g . 42–48] . The short life cycle [49] , [50] of Arabidopsis facilitates performing serial passage experiments of vertical transmission , and the large numbers of seeds produced by a plant [49] , [51] ensures the maintenance of parasite lineages between host generations even at low rates of vertical transmission . CMV is a generalist parasite with the broadest host range known for a plant virus , and its genomic structure , replication , gene expression and pathogenicity have been analysed extensively [reviewed in 52] , [ 53] . CMV isolates are highly diverse and they have been classified into two subgroups ( subgroup I and subgroup II ) based on the nucleotide sequence similarity of their genomic RNAs . CMV is horizontally transmitted by more than 70 species of aphids in a non-persistent manner , and vertically through seeds , with rates that vary depending on CMV and plant genotypes [53] . In Arabidopsis , the efficiency of CMV seed transmission ranges between 2 and 8% [unpublished data] . In this work , we serially passaged three CMV strains in one Arabidopsis genotype for five generations under three transmission modes: strict vertical transmission , strict horizontal transmission , and alternation of vertical and horizontal transmission . We monitored the vertical transmission rate through seeds , virus accumulation and virulence after each passage . After the last passage , we compared virus accumulation and virulence of the evolved and the non-evolved virus lineages , both in the original plant stock , and in the progeny of plants derived from the fifth passage of vertical transmission . Results indicate that vertical passaging led to adaptation of the virus to greater vertical transmission , which was associated with reductions of virus accumulation and virulence . Increases in seed transmission and reductions in virus accumulation and virulence were determined also by reciprocal host adaptation during vertical passages , which additionally reduced virulence and multiplication of vertically evolved viruses . Host adaptation to vertically passaged viruses was traded-off against reduced resistance to the non-evolved viruses .
We studied the evolution of three strains of CMV; Fny-CMV ( a well-characterized strain isolated in New York State , belonging to subgroup I of CMV strains ) , De72-CMV ( isolated from a species in the Brassicaceae in Central Spain and also belonging to subgroup I ) , and of LS-CMV ( a well-characterized strain isolated in New York State , belonging to subgroup II ) [53] in Arabidopsis under strict vertical , strict horizontal or alternated vertical and horizontal transmission ( Figure 1 , and see Material and Methods for experimental details ) . The vertical transmission rate , estimated as the percentage of CMV-infected seeds; virus accumulation , quantified as µg of viral RNA per gr of fresh plant tissue; and virulence , measured as the effect of infection on fecundity , estimated from total seed weight , 1− ( SWi/SWm ) ( i and m denote infected and mock plants , respectively ) , were determined in each of five passages of vertical transmission ( Figure 2 , Table 1 ) . CMV vertical transmission rate , virus accumulation and virulence in each passage were quantified in the same plants in which the virus was passaged . For each of the three virus strains , the five replicate lineages yielded similar values for seed transmission rate ( F4 , 25≤2 . 19; P≥0 . 120 ) , virus accumulation ( F4 , 25≤1 . 51; P≥0 . 248 ) , and virulence ( F2 , 72≤1 . 50; P≥0 . 251 ) . Thus , lineage was not considered as a factor for further analyses . Seed transmission rate increased as the number of vertical passages increased , either when the three CMV strains were analysed together ( F4 , 72 = 48 . 72; P<1×10−5 ) , or independently ( F4 , 25≥12 . 79; P≤1×10−5 ) . Indeed , a significant positive linear correlation between seed transmission rate and number of vertical passage was found either when the three strains were considered together ( r = 0 . 84; P<1×10−5 ) or individually ( r≥0 . 83; P<1×10−5 ) . Slopes and intercepts of these regression lines did not differ significantly between CMV strains ( F2 , 72≤1 . 03; P≥0 . 363 ) , indicating that seed transmission increased at the same rate for the three viral strains ( Figure 2a , Table 1 ) . Virus accumulation decreased as the number of vertical transmission passages increased , either considering all strains together ( F4 , 72 = 8 . 16; P<1×10−4 ) or separately ( F4 , 25≥6 . 17; P≤0 . 002 ) ( Figure 2b , Table 1 ) . A negative logarithmic correlation between virus accumulation and vertical passage was found considering all strains together ( r = −0 . 58; P<1×10−5 ) and independently ( r≥−0 . 69; P≤1×10−4 ) . The slope and intercept of the De72-CMV regression were different from those of the Fny-CMV and LS-CMV regressions , so that virus accumulation decreased more slowly in the former than in the latter two strains ( F2 , 72≥9 . 41; P≤0 . 010 ) ( Figure 2b ) . Virulence decreased as the number of passages increased for all viral strains together and individually ( F4 , 72≥6 . 53; P≤0 . 002 ) ( Figure 2c , Table 1 ) . A significant negative linear correlation between virulence and vertical passage was observed , either for all strains together ( r = −0 . 80; P<1×10−5 ) or separately ( r≤−0 . 78; P<1×10−5 ) . Slopes and intercepts of the regression lines for each viral strain were similar ( F2 , 72≤1 . 23; P≥0 . 298 ) . The analyses above strongly suggest that increases in seed transmission rate might be accompanied by reductions in virus accumulation and virulence . To explore this relationship , we analysed the association between these traits during the serial vertical transmission passages considering all strains together and independently ( Figure 3 ) . Regression analyses indicated that the seed transmission rate was negatively ( exponentially ) correlated with virus accumulation for all strains together ( r = −0 . 32; P = 0 . 005 ) and separately ( r≤−0 . 40; P≤0 . 049 ) ( Figure 3a ) . In addition , seed transmission rate was also negatively ( linearly ) correlated with virulence in all CMV strains ( r≤−0 . 56; P≤0 . 006 ) ( Figure 3b ) . Finally , virus accumulation and virulence were always positively ( exponentially ) correlated ( r≥0 . 45; P≤0 . 034 ) ( Figure 3c ) . Overall , these results indicate that the increase of seed transmission rate through vertical passages is associated with a reduction of virus accumulation and virulence in Arabidopsis . These changes might be due to virus evolution and/or host evolution . In the next sections we address these possibilities . We further analysed how the mode of transmission affected CMV accumulation , and the effect of infection on plant growth and fitness . To do so , virus lineages evolved under vertical , horizontal and alternated transmission ( from here on , referred to as vertically evolved , horizontally evolved and alternately evolved viruses ) , as well as the initial non-evolved strains , were inoculated in plants from the ‘original’ seed stock . In these plants , we determined the effect of virus infection on plant vegetative and reproductive growth , through the weight of the rosettes ( RW ) and inflorescences ( IW ) , respectively . We previously reported that tolerance of Arabidopsis to CMV is attained by reallocating resources from vegetative to reproductive structures [41] . For these analyses , we used the RW and IW ratios ( Traiti/Traitm , where i and m denote infected and mock-inoculated plants , respectively ) . We also quantified the effect on SW ( virulence ) and virus accumulation as described above . Vertical transmission rate in plants from the ‘original’ seed stock was quantified only for vertically evolved viruses . The five virus lineages for each combination of CMV strain and mode of transmission did not differ for any of the parameters estimated ( F≤2 . 27; P≥0 . 077 ) ( Table S1 ) . Thus , lineage was not considered as a factor for further analyses . Indeed , GLM analyses nesting lineage in mode of transmission did not change our results ( not shown ) . Virus accumulation differed between evolved and non-evolved viruses , and between evolved viruses under different transmission modes for the three CMV strains ( F3 , 135≥7 . 53;P≤1×10−5 ) . In general , evolved viruses of the three strains showed lower virus accumulation than non-evolved viruses ( P≤0 . 024 ) , with the exception of horizontally evolved Fny-CMV ( P = 0 . 378 ) ( Table 2 ) . In addition , the effect of the mode of transmission ( vertical , horizontal or alternated ) varied depending on the CMV strain . In Fny-CMV , virus accumulation was similar for vertically and alternately evolved viruses ( P = 0 . 154 ) , and in both cases was significantly lower than for horizontally evolved viruses ( P≤0 . 001 ) . In LS-CMV , the virus accumulation of vertically evolved viruses was lower than that of horizontally and alternately evolved viruses ( P<1×10−4 ) , which were not significantly different ( P = 0 . 898 ) ( Table 2 ) . The mode of transmission did not affect the accumulation of De72-CMV ( P≥0 . 200 ) ( Table 2 ) . The mode of transmission did not significantly affect the effect of infection on vegetative growth ( RW ratio ) of any CMV strain ( Table 2 ) ( F3 , 135≤0 . 73; P≥0 . 538 ) . The effect of virus infection on reproductive growth ( IW ratio ) ( Table 2 ) varied significantly among transmission modes ( F3 , 135≥3 . 29; P≤0 . 023 ) . In Fny-CMV and LS-CMV , the IW ratio was significantly higher in vertically and alternately evolved viruses than in horizontally evolved and non-evolved viruses ( P≤0 . 051 ) ( Table 1 ) . In De72-CMV , the effect of infection on the IW ratio was significantly lower in vertically evolved viruses than in the other three treatments ( P≤0 . 017 ) , which were not significantly different ( P>0 . 645 ) ( Table 2 ) . The effect of infection by Fny-CMV and LS-CMV on seed weight ( i . e . , virulence ) significantly differed among transmission modes ( F3 , 135≥5 . 15; P≤0 . 002 ) . In both strains , virulence was lower when viruses were vertically and alternately evolved than when they were horizontally evolved or non-evolved ( P≤0 . 030 ) . No differences in virulence were observed between evolved and non-evolved De72-CMV viruses regardless of transmission mode ( F3 , 135 = 1 . 36; P = 0 . 260 ) . Finally , Fny-CMV and LS-CMV viruses had significantly higher vertical transmission rates when vertically evolved than non-evolved viruses in plants from the ‘original’ stock ( F1 , 42≥4 . 42; P≤0 . 043 ) , while no differences were observed for De72-CMV ( F1 , 33 = 1 . 15; P = 0 . 289 ) ( Table 2 ) . Thus , CMV evolution under strict vertical transmission results in decreased virus accumulation and virulence . Vertical transmission also results in a reduced effect of infection on plant growth , particularly of the plant reproductive structures . These effects of passaging are higher in Fny-CMV and LS-CMV than in De72-CMV . Serial passages under strict vertical transmission may result not only in virus evolution , but also in host adaptation; that is , we might be selecting for plant individuals that transmit the virus to seed at higher rates . To analyse this possibility , seeds from non-infected plants of the fifth vertical passage representing plant lineages evolved with each of the three virus strains were grown and inoculated with the corresponding evolved lineages and with the non-evolved CMV isolates . In these plants , virus accumulation and effects of infection on plant growth ( RW and IW ) and on fecundity ( SW ) were analysed as in ‘original’ stock plants , but percentage of CMV seed transmission was not determined . For each combination of CMV strain and mode of transmission , the different virus lineages did not differ in any of the parameters estimated ( F≤2 . 45; P≥0 . 060 ) ( Table S2 ) , and therefore lineage was not considered as a factor . As above , GLM analyses using ‘lineage’ as a nested factor did not alter the results . Virus accumulation in evolved plants differed among treatments ( F3 , 200≥3 . 59; P≤0 . 015 ) . Fny-CMV and LS-CMV evolved viruses accumulated at lower levels than non-evolved viruses ( P<1×10−4 ) , while the opposite was observed in De72-CMV ( P≤0 . 055 ) ( Table 3 ) . The effect of the mode of transmission varied depending on the CMV strain . In Fny-CMV , virus accumulation of vertically evolved viruses was lower than that of alternately evolved viruses ( P = 0 . 001 ) , with intermediate accumulation values of horizontally evolved viruses ( P≤0 . 005 ) ( Table 3 ) . In LS-CMV , virus accumulation was not significantly different in vertically and alternately evolved viruses ( P = 0 . 446 ) , but was significantly less in horizontally evolved viruses ( P<1×10−4 ) . The mode of transmission did not affect the accumulation of De72-CMV ( P≥0 . 216 ) ( Table 3 ) . Differences in the effect of CMV infection in RW and IW were observed among the different modes of transmission ( F3 , 200≥6 . 27; P<1×10−4 ) , and followed similar patterns in the three virus strains ( Table 3 ) . In Fny-CMV and LS-CMV , vertically evolved viruses had significantly higher RW and IW ratios than the non-evolved viruses ( P<1×10−4 ) , with intermediate values for horizontally and alternately evolved viruses ( Table 3 ) . In De72-CMV the effect of infection on RW and IW ratios was less in vertically evolved viruses than in the other three treatments ( P≤0 . 041 ) , which were not significantly different among themselves ( P≥0 . 068 ) ( Table 2 ) . In all strains , virulence significantly differed among modes of transmission ( F3 , 200≥21 . 17; P<1×10−4 ) , with vertically evolved viruses always being less virulent than viruses in the other three treatments ( P<1×10−4 ) . In addition , virulence was significantly lower in alternately evolved viruses than in the other two treatments ( P≤0 . 052 ) ( Table 3 ) . Finally , we explored if there were differences in virus accumulation and virulence of vertically evolved viruses in plants derived from the fifth vertical passage with respect to whether they were infected by vertical transmission ( via seed ) or horizontal transmission ( mechanical inoculation ) . To do so , we compared the data above for plants mechanically inoculated ( Table 3 ) with results for these traits in vertically infected plants derived from the fifth vertical passage ( Table 1 ) . Virus accumulation was higher in the horizontally inoculated than in the vertically infected plants for lineages of the three strains combined ( F1 , 333≥450 . 11; P<1×10−5 ) , and for each strain individually ( F1≥720 . 45; P<1×10−5 ) . Accordingly , virulence was always lower in vertically infected plants ( F1≥12 . 58; P≤0 . 001 ) . In summary , CMV evolution under vertical transmission decreases virus accumulation , virulence and effect of infection on the growth of plants derived from the fifth vertical passage . Differences among modes of transmission were larger in these plants than in those from the ‘original’ stock for most traits , which suggests that plants changed during the vertical passages by increasing their resistance to virus infection . Whether these changes co-evolved with those observed in the viruses is analysed in the next section . If vertical transmission of CMV resulted in selection for Arabidopsis plants with better performance under virus infection , the effect of infection with evolved and non-evolved strains would differ between ‘original’ stock plants and plants derived from the fifth vertical passage . To test this hypothesis , virus accumulation , RW and IW ratios and virulence in ‘original’ stock plants were compared with plants derived from the fifth vertical passage ( Tables 2–3 ) . Vertical transmission rate was compared between plants of the ‘original’ seed stock ( Table 2 ) and plants of the fifth vertical passage ( fifth passage in Figure 2 ) . Fny-CMV and LS-CMV vertically evolved viruses accumulated to lower levels ( F1 , 100≥3 . 97;P≤0 . 054 ) , and infected plants had higher RW and IW ratios , and lower virulence ( F1 , 100≥3 . 61;P≤0 . 068 ) in plants derived from the fifth vertical passage than in ‘original’ stock plants ( Table 3 ) . The opposite was observed in the non-evolved viruses of these two strains ( F1 , 100≥7 . 23; P≤0 . 009 , for virus accumulation; and F1 , 100≥1 . 65; P≤0 . 085 , for RW and IW ratios , and virulence ) . The exception was that no differences in RW and IW ratios were observed between LS-CMV-infected plants of the ‘original’ stock and those derived from the fifth vertical passage ( F1 , 100≤1 . 68; P≥0 . 198 ) ( Table 3 ) . For De72-CMV , an increase of RW and IW ratios , and a reduction in virulence was found in plants derived from the fifth vertical passage relative to ‘original’ stock plants infected with vertically evolved viruses ( F1 , 80≥5 . 36; P≤0 . 023 ) , but virus accumulation was similar in both types of plants ( F1 , 80 = 0 . 34; P = 0 . 560 ) . Non-evolved viruses were marginally more virulent in plants derived from the fifth vertical passage than in ‘original’ stock plants ( F1 , 80≥2 . 97; P≤0 . 090 ) ( Table 3 ) . Vertical transmission was higher in plants of the fifth passage of vertical transmission than in plants of the ‘original’ stock ( compare Figure 2 and Table 2; F≥6 . 93; P≤0 . 013 ) . Fny-CMV and LS-CMV strains horizontally evolved ( Table 3 ) caused less reduction in IW and were more virulent in plants derived from the fifth vertical passage compared with ‘original’ stock plants ( F1 , 100≥4 . 69; P≤0 . 033 ) . In De72-CMV virus accumulation was higher in plants derived from the fifth vertical passage than in the ‘original’ stock plants ( F1 , 80 = 301 . 33; P≤1×10−4 ) . Finally , Fny-CMV and LS-CMV alternately evolved were more virulent in plants derived from the fifth vertical passage compared with ‘original’ stock plants ( F1 , 100≥4 . 40; P≤0 . 040 ) , but virus accumulation of De72-CMV was higher in ‘original’ stock plants than in plants derived from the fifth vertical passage ( F1 , 80 = 41 . 92; P≤1×10−4 ) . No significant differences were found in the rest of comparisons ( Tables 1–3 ) . Thus , differences in the traits analysed between ‘original’ stock plants and plants derived from the fifth vertical passage indicate that the latter group of plants have mechanisms to reduce virus accumulation , the effects of infection on plant growth and on fecundity ( virulence ) by vertically transmitted viruses , often at the cost of increased virus accumulation and/or virulence of non-evolved viruses . These results support the hypothesis of plant-virus co-evolution during vertical transmission passages .
Most experimental analyses of virulence evolution are based on the hypothesis that virulence , which is correlated with parasite multiplication , is determined by trade-offs with parasite transmission rate [2] , [10]–[12] . However , the factors that modulate this trade-off and how they act are only partially understood . Although theory has identified several of these potential modulators , their effect has seldom been analysed experimentally . Here we investigated the role of two such major factors in the evolution of virulence: the mode of transmission , and host adaptation in response to parasite evolution [8] , [14] , [16] , [22] . The paucity of information on this subject has been attributed , in part , to the lack of suitable experimental systems [2] . For our experiments , we used the plant virus CMV and its host plant Arabidopsis thaliana , a system with several traits suitable for our objectives: i ) CMV is a pathogen of Arabidopsis that is found at high incidence in Arabidopsis wild populations [45] , ii ) CMV is transmitted both horizontally and vertically in Arabidopsis [45; unpublished data] , iii ) Arabidopsis has a short generation time , which allows serial passages of vertical transmission to be performed in a reasonable time frame; and iv ) recovery of Arabidopsis plants to CMV infection has not been described . From the perspective of viral fitness , host recovery is equivalent to host death in which the virus can no longer replicate or be transmitted . Therefore , recovery would affect the transmission rate , potentially blurring the virulence-transmission trade-off [54] . We serially passaged three CMV strains in Arabidopsis under strict vertical or strict horizontal transmission , and quantified virulence and traits related to viral fitness in the evolved and non-evolved viruses . We used the effect of virus infection on plant fecundity as a measure of virulence . Virulence encompasses the negative effect of a parasite on host longevity and fecundity [8] , [55] , [56] . In vertically transmitted parasites , the effect of infection on host fecundity is the most relevant trait for transmission success [10] , [13] , [14] , [23]–[25] . Thus , host fecundity is the best proxy for virulence to analyse the vertical transmission-virulence trade-off . However , longevity of hosts infected with horizontally transmitted parasites , which is linked to the length of the infectious period , is the most obvious determinant of parasite transmission rate . Indeed , in such parasites longevity of infected hosts has been proposed to be a good proxy for virulence [57] . Although CMV infection significantly affects host fecundity , it has little effect on the lifespan of Arabidopsis [41] . This mimics the effect of sterilizing parasites on their hosts . For sterilizing parasites horizontal transmission may be correlated with host fecundity , in which case the effect on host fecundity is the best proxy for virulence [8] , [55] , [58] . As predicted by theory , our results show a key role of the mode of transmission in the evolution of virulence . The comparison of non-evolved viruses with those evolved under strict vertical transmission indicated that the latter increased their rate of vertical transmission , and that adaptation to this transmission mode was associated with the reduction of virus multiplication and virulence . In contrast , evolution under strict horizontal transmission did not result in changes of virus multiplication or virulence . We did not quantify the rate of vertical transmission of horizontally evolved lineages . However , the absence of evolution in virus multiplication and in virulence of horizontally passaged viruses − two traits that are correlated with vertical transmission rate in our system − may be suggestive of a lack of change in vertical transmission rate . Thus , our results support predictions of the models of virulence evolution based on the trade-off hypothesis [14] , [16] , [22] in that we found a negative correlation between rate of vertical transmission and virulence . This negative correlation was also found in previous studies with bacteria and insect parasites [26]–[28] , [30]–[33] , [59] , [60] , and in the only other study of a plant virus , Barley stripe mosaic virus ( BSMV ) in barley [29] . Predictions of the trade-off hypothesis were not supported by results of the only other reported analyses of a plant-parasite system , in which the rate of vertical transmission was positively correlated with virulence . However , this is an unusual system as it involves a sterilizing fungus . This fungus invades the plant reproductive structures , and more virulent fungal strains have better access to the seeds . Thus , more virulent strains compensate the higher reduction of plant fecundity by infecting more seeds than less virulent strains [34] , [35] . More plant-parasite systems need to be characterized to determine the generality of the trade-off predictions . Perhaps one of the most striking results of this work is the observed negative correlation between virus multiplication and transmission rate . In other plant-parasite systems higher parasite load is associated with higher percentage of infected seeds [29] , [34] , [35] . Limited knowledge on the mechanisms of CMV seed transmission hinders the interpretation of this result . CMV is present in the embryo , the endosperm , and the coat of infected seeds [61]; and the virus is thought to gain access to seed tissues either through the ovules or pollen [62] , or through the suspensor that connects the mother plant and the developing seed [63] . Hence , vertical transmission rate would be determined by the capacity of the virus to reach the seed during gametogenesis and/or while the suspensor is still functional , and by the ability of plant defense to block virus access to the seed . If this model holds for CMV and Arabidopsis , a negative correlation between virus multiplication and transmission rate would be explained: 1 ) if lower virus titer would result in a less efficient triggering of plant defenses that prevent seed infection; and/or 2 ) if serial passages of vertical transmission selects for virus variants with mutations that facilitate direct or indirect access for CMV to the seed even at low multiplication levels . Virulence and within-host multiplication were positively correlated in viruses evolved under strict vertical transmission in this study , consistent with the central assumption of the trade-off hypothesis [2] , [10]–[12] . This result is at odds with observations in BSMV and barley where neither vertical transmission nor virulence correlated with virus multiplication [29] , [64] . Virulence and within-host multiplication does not correlate across CMV and Arabidopsis genotypes , however , as genotype-specific host tolerance to virus infection uncoupled both traits [65] . Interestingly , the Arabidopsis genotype used in the present work ( Cen-1 ) was rated as a low-tolerance genotype [41] , so that the relationship between virulence and within-host multiplication will not be blurred by tolerance . Note also that the relationship between virulence and within-host multiplication is not linear ( Figures 2–3 ) . Thus , it is possible that this relationship could go unnoticed in some studies [66] if data are within the range in which this relationship is saturated . The reduction of CMV virulence associated with increased rates of vertical transmission is also associated with smaller effects of infection in both the vegetative and reproductive efforts of the host ( RW and IW ) . Vertically transmitted parasites may select for modifications of host life-history traits that enhance host reproductive success and parasite transmission . For instance , insects have increased survival when infected by bacteria such as Rickettsia or Wolbachia [67] , [68] , or increased reproductive time span when infected by microsporidian parasites [69] . The higher RW and IW ratios of Arabidopsis infected with vertically evolved viruses compared with non-evolved viruses might be interpreted similarly , as Arabidopsis fecundity is positively correlated with both the vegetative and the reproductive growth of the plant [41] . Thus , the observed evolution towards lower virulence may represent a selective advantage for the virus , as it increases the chances for its vertical transmission . In certain contexts , the parasite and its host may have sufficient alignment of interests to boost host-virus co-evolution: Parasite infection may be still detrimental for the host , but maximization of host fitness upon infection also maximizes the parasite fitness [70] . Such maximization might occur when parasites adapt to vertical transmission [3] , [71] , [72] . In this scenario , the host may also evolve to increase the parasite's rate of vertical transmission [39] , [73] , a possibility that has seldom been analysed [36] . Our data provide evidence that this may be the case for the Arabidopsis-CMV interaction , as vertical transmission rate was higher , and virulence was lower , in plants selected during serial passages of vertical transmission as compared with plants from the ‘original’ stock . Although these changes in the passaged plant were not enough to completely compensate the negative effect of CMV infection in plant fitness , they significantly increased plant fecundity as compared with ‘original’ stock plants . The observed changes in the host plant during vertical passages are related to increased resistance , i . e . , reduced parasite multiplication [74] . Resistance was particularly effective when infection was the result of vertical transmission compared with horizontal transmission ( Tables 1–3 ) , although we cannot rule out that this would reflect that plants infected horizontally may be weaker as they come from seeds already challenged ( unsuccessfully ) with CMV . In either case , a degree of plant resistance may paradoxically represent a benefit for CMV strains adapted to vertical transmission , as long as multiplication does not go below a threshold that importantly decreases seed transmission rate . These patterns of plant-virus co-evolution may be due to selection during passages . In wild Arabidopsis populations in Spain , the inbreeding coefficient , i . e . , probability of autozygosity [75] , ranges between 0 . 87 and 0 . 99 [76] . Hence , wild accessions are not homozygous at all loci , and genetic variation may be expected in an experimental plant population derived from a single individual . Therefore , selection may have acted upon genetic variation in the original stock plant . Evolution in the Arabidopsis-CMV interaction is compatible with conditions required for host-parasite co-evolution as defined by [37] in which hosts and pathogens exert reciprocal selection on each other . Additionally , we can speculate on genetic changes induced by virus infection , resulting either in genomic rearrangements or in epigenetic variation [77] , [78] that may explain the observed changes in the host plant . Since all passages of horizontal transmission were done in plants from the ‘original’ stock , we cannot address whether plant adaptation to horizontally passaged viruses occurred in our experiments . The hypothesis that adaptation to vertical transmission is advantageous for both the virus and the plant is in apparent contradiction with the observation that vertical transmission of CMV occurs only at low rates across Arabidopsis genotypes [45; this study] . However , adaptation in Arabidopsis came at the cost of increased virulence and increased multiplication of the non-evolved viruses , i . e . , a trade-off in host fitness when infected with non-evolved virus genotypes . In natural populations of Arabidopsis , CMV spreads both by horizontal and vertical transmission [45; unpublished data] , and it is reasonable to hypothesize that both the optimal level of vertical transmission for the host plant , and of virulence for the virus would be influenced by the observed adaptation trade-off . In nature , most virus genotypes are likely to be better adapted to horizontal than to vertical transmission , and plant genotypes adapted to vertical transmission would suffer from extra fitness penalties when infected by most virus genotypes . Thus , virulence evolution could not be explained only by trade-offs between the relative rates of vertical and horizontal transmission , as stated by the ‘continuum hypothesis’: our results illustrate the key role that the complex interplay between mode of transmission and host-parasite co-evolution has in determining virulence evolution . This type of interplay should be considered in theoretical and experimental analyses of virulence evolution , and in the design of better strategies for virulence management .
Three virus strains were used: Fny-CMV and De72-CMV belonging to subgroup I of CMV isolates , and LS-CMV , belonging to subgroup II . Fny-CMV and LS-CMV are well characterized and were derived from biologically active cDNA clones [79] , [80] by in vitro transcription with T7 RNA polymerase ( New England Biolabs , Ipswich , MA , USA ) . De72-CMV was obtained from a field-infected plant of Diplotaxis erucoides ( Brassicaceae ) , a host closely related to Arabidopsis [81] . Transcripts of Fny-CMV and LS-CMV , and purified viral RNA from De72-CMV were used to infect tobacco ( Nicotiana tabacum ) plants for virus multiplication . CMV virions from tobacco leaves were purified as described in [82] , and viral RNA was extracted by virion disruption with phenol and sodium dodecyl sulphate . Accession Cen-1 ( Centenera , Spain ) was selected from a panel of eighteen Arabidopsis accessions as it combined a higher efficiency of CMV vertical transmission − ranging between 2–8% depending on the virus isolate [unpublished data] − , with a relatively short life cycle [65] . For plant growth , seeds of Cen-1 were sown on filter paper soaked with water in single plastic Petri dishes , and stratified in darkness at 4°C for five days before transferring for germination to a growth chamber ( 22°C , 14 h light and 70% relative humidity ) . Five day-old seedlings were planted in soil in 10 . 5-cm-diameter pots ( 0 . 43 l volume ) , and grown in a greenhouse ( 25/20°C day/night , 16 h light ) . The three CMV isolates were serially passaged in Cen-1 plants five times by strict vertical transmission , strict horizontal transmission , or alternating both modes of transmission ( Figure 1 ) . To analyse virus evolution under strict vertical transmission , Cen-1 plants from a seed stock ( referred to as ‘original’ stock ) generously provided by Dr . Carlos Alonso-Blanco ( CNB-CSIC , Spain ) were mechanically inoculated with purified CMV RNA of the three CMV strains ( 100 ng/ml ) in 0 . 1 M Na2HPO4 when rosettes had 4–5 leaves ( stages 1 . 04–1 . 05 in [50] ) with five replicates per treatment . Seeds from each infected plant were harvested at complete senescence ( stage 9 . 0 as in [50] ) generating five independent vertically transmitted CMV lineages per virus strain . One hundred Cen-1 seeds per CMV-infected plant were grown as described above , and CMV infection in twenty-day-old plants was detected by dot-blot hybridization ( see below ) to estimate the percentage of seed transmission . Infected plants were allowed to complete their life cycle and seeds were harvested at senescence . One infected plant per CMV lineage was randomly selected to start the next plant generation , and the process was repeated for five generations; i . e . , five passages of vertical transmission ( Figure 1 ) . The percentage of seed transmission was determined in every passage as described above . Seeds from uninfected individuals of the fifth generation were also harvested . To analyse virus evolution under strict horizontal transmission , sap extracts from the fifteen Cen-1 plants inoculated to generate the vertically transmitted lineages were used to mechanically inoculate fifteen uninfected Cen-1 plants from the ‘original’ Cen-1 seed stock; these fifteen plants represented the first passage of horizontal transmission in five independent lineages for each virus strain . Sap extract from each of these fifteen infected plants was used to inoculate ten plants at stages 1 . 04–1 . 05 [50] again from the ‘original’ Cen-1 seed stock . CMV infection was detected fifteen days post-inoculation ( dpi ) by dot-blot hybridization ( see below ) . Sap extract from one randomly chosen infected plant per lineage was used to inoculate ten new plants of the ‘original’ stock of Cen-1 seeds . This procedure was repeated for five passages ( Figure 1 ) . In parallel , five viral lineages per CMV strain were evolved alternating vertical and horizontal transmission to generate five alternately transmitted lineages . To do so , 100 seeds from each of the fifteen plants used to generate the vertically transmitted CMV lineages were grown , and CMV infection was detected , as described above . Sap extracts from one infected plant per lineage were used to inoculate ten Cen-1 plants from the ‘original’ seed stock . The procedure for this treatment to this point is identical to that used for the strict horizontal treatment . However , seeds from these horizontally infected plants were harvested at complete senescence . One hundred seeds per plant were grown , virus infection was detected , and one plant per lineage was randomly chosen to generate the new vertically transmitted generation . This process was repeated until the third horizontal passage was completed , for a total of three vertical passages and three horizontal passages ( Figure 1 ) . Multiplication and virulence of non-evolved and evolved CMV strains after vertical , horizontal and alternated passages , referred to as vertically evolved , horizontally evolved and alternately evolved viruses , were analysed in plants from the ‘original’ seed stock and in plants derived from the fifth vertical transmission passage . Sap extracts from CMV-infected plants of the fifth vertical and horizontal passages were used to inoculate ten Cen-1 plants from the ‘original’ seed stock and ten plants from seeds from the fifth passage of vertical transmission per virus lineage . Only five plants were inoculated for each of the alternately evolved viruses . In the fourth vertical transmission passage , no infected seeds were detected for one of the De72-CMV lineages; therefore , only four replicates were analysed for this treatment . Plants derived from the fifth passage of vertical transmission were chosen randomly to represent plants passaged with each of the three CMV strains , and inoculated with evolved lineages of the corresponding strain . In addition , ten plants of the ‘original’ Cen-1 stock and ten plants from one randomly chosen uninfected replicate after five vertical passages per each of the three strains were inoculated with sap from Cen-1 plants infected with non-evolved Fny-CMV , De72-CMV and LS-CMV . Mock-inoculated plants , with ten replicates per seed stock , were included as a control . Virus multiplication , and effect of virus infection on rosette and inflorescence growth , and on seed production were determined in each plant as described below . CMV multiplication was quantified as virus RNA accumulation . Total nucleic acid extracts from four leaf discs ( 0 . 01 g fresh weight ) collected from four different rosette and inflorescence leaves were obtained using TRI-reagent ( Sigma-Aldrich , St . Louis , MO , USA ) . RNA quantification was done by dot-blot hybridization with 32P-labeled RNA probes obtained by transcription from cDNA clones representing the 3′ non-coding region of the three genomic RNAs , which is highly similar within a CMV isolate . For Fny-CMV and De72-CMV , a probe representing nucleotides 1933 to 2215 of Fny-CMV RNA3 ( GeneBank Acc . No . D10538 ) was used , and for LS-CMV the probe represented nucleotides 1861 to 2193 of LS-CMV RNA3 ( Acc . No AF127976 ) . Internal CMV standards for subgroup I ( Fny-CMV or De72-CMV ) , and subgroup II ( LS-CMV ) were included as a two-fold dilution series of purified RNA ( 0 . 5 to 0 . 001 µg ) in nucleic acid extracts from mock-inoculated Arabidopsis plants . RNA extracts from infected plants were blotted at different dilutions to ensure that hybridization signal was on the linear portion of the RNA concentration-hybridization signal curve . All hybridizations were done at 65°C overnight in 6× SSC , 5× Denhardt's mixture , 0 . 1% sodium dodecyl sulphate , and yeast tRNA at 50 mg/ml [83] . RNA hybridization signal was detected using a Typhoon 9400 scanner ( GE Healthcare , Chalfont St . Giles , UK ) after exposure of the Eu+2 store phosphor screens to the labelled samples , and CMV multiplication was quantified by using Image-Quant 5 . 2 ( Molecular Dynamics , GE Healthcare ) [84] . Virulence is defined as the negative effect of infection on host fitness [1] , [2] , a good proxy for host fitness being total fecundity . Our previous work showed that CMV infection does not affect the viability or weight of individual seeds in Cen-1 [56]; therefore we used total seed weight ( SW ) as a measure of host fitness . Thus , virulence was estimated as one minus the ratio of the total seed weight of infected ( SWi ) to total seed weight of mock-inoculated ( SWm ) plants , 1− ( SWi/SWm ) . We also measured plant dry weight at complete senescence after drying at 65°C until constant weight , as a measure of tolerance ( see Introduction ) . Rosette weight ( RW ) and inflorescence weight including seeds ( IW ) were measured separately . To quantify the effect of CMV infection on RW and IW , the value of each infected plant was divided by the mean value of the mock-inoculated plants ( Traiti/Traitm , i and m denote infected and mock-inoculated plants , respectively ) . In each vertical passage , the rate of seed transmission was determined as the percentage of infected individuals out of the 100 germinated plants per lineage . Two leaves of twenty-day-old plants were harvested and pooled in groups of ten individuals , and total nucleic acid extracts of these pools were obtained . We pooled leaves among plants because of the low percentage of plants infected [unpublished data] . The presence of CMV in each pool was detected by dot-blot hybridization with the same probes used for quantification of virus accumulation . As negative controls , total nucleic acid extracts from pools of ten twenty-day-old non-infected plants were used . Samples with hybridization signal more than two-fold higher the negative controls were considered as positive . Two leaves were harvested from each plant within each positive pool for detection of CMV individually by dot-blot hybridization as described above . In the ‘original’ stock plants infected with the vertically evolved CMV strains , five infected plants per lineage were randomly chosen and 100 seeds per plant were grown as described above . Five-day-old seedlings were harvested in pools of ten individuals and the presence of CMV in each pool was detected essentially as described above . Because we pooled seedlings for this analysis , seed transmission rate was estimated assuming that the number of seedlings infected follows a Poisson distribution [85] . Estimates obtained assuming a binomial distribution yielded the same results . Data on virus multiplication , seed transmission rate , and rosette , inflorescence and seed weights , and their various transformations , including virulence , were homoscedastic and were analysed using full factorial General Linear Models ( GLM ) . Virus strain/lineage , type of plant ( stock vs . passaged ) , and mode of transmission were considered as fixed effect factors . Significance of differences among classes within each factor was determined by Least Significant Difference ( LSD ) analyses . Linear associations between virus multiplication , seed transmission rate and virulence during serial passages was analysed by bivariate tests and using Pearson's correlation test [85] . To investigate whether non-linear models better explained these associations , we fitted them to logarithmic , exponential and quadratic models [86] . Regression lines were compared using ANOVA to test the equality of slopes and intercepts; non-linear curves were log transformed for this analysis . All statistical analyses were performed using the statistical software packages SPSS 21 . 0 ( SPSS Inc . , Chicago , IL , USA ) . | Virulence is a key property of parasites , and is linked to the emergence of new diseases and to the reduction of ecosystem biodiversity . Consequently , scientists have devoted a great effort to build theoretical models that predict which factors may modulate virulence evolution . However , whether ( and how ) these factors affect virulence evolution has been seldom analysed experimentally . Using the plant virus Cucumber mosaic virus ( CMV ) and its natural host Arabidopsis thaliana , we studied the role of two such factors: the mode of transmission , and host adaptation in response to parasite evolution . We serially passaged CMV under strict vertical and strict horizontal transmission , and a combination of both . Subsequently , we analysed differences in CMV seed ( vertical ) transmission rate , accumulation and virulence between evolved and non-evolved viruses . We also compared whether these differences varied in original plants and in plants evolved during vertical passaging . Vertical passaging increased CMV seed transmission , and reduced accumulation and virulence , while horizontal passaging had no effect . Changes during vertical passaging were determined also by reciprocal host adaptation , which additionally reduced virulence and accumulation of vertically transmitted viruses . Hence , we provide evidence that the interplay between the transmission mode and host-parasite co-evolution is central in determining virulence evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"organismal",
"evolution",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"evolutionary",
"biology",
"host-pathogen",
"interactions",
"coevolution",
"microbial",
"evolution",
"virology",
"viral",
"transmission",
"and",
"infection",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"adaptation",
"microbiology",
"pathogenesis",
"evolutionary",
"processes",
"viral",
"evolution"
]
| 2014 | Vertical Transmission Selects for Reduced Virulence in a Plant Virus and for Increased Resistance in the Host |
The major DNA repair pathways operate on damage in double-strand DNA because they use the intact strand as a template after damage removal . Therefore , lesions in transient single-strand stretches of chromosomal DNA are expected to be especially threatening to genome stability . To test this hypothesis , we designed systems in budding yeast that could generate many kilobases of persistent single-strand DNA next to double-strand breaks or uncapped telomeres . The systems allowed controlled restoration to the double-strand state after applying DNA damage . We found that lesions induced by UV-light and methyl methanesulfonate can be tolerated in long single-strand regions and are hypermutagenic . The hypermutability required PCNA monoubiquitination and was largely attributable to translesion synthesis by the error-prone DNA polymerase ζ . In support of multiple lesions in single-strand DNA being a source of hypermutability , analysis of the UV-induced mutants revealed strong strand-specific bias and unexpectedly high frequency of alleles with widely separated multiple mutations scattered over several kilobases . Hypermutability and multiple mutations associated with lesions in transient stretches of long single-strand DNA may be a source of carcinogenesis and provide selective advantage in adaptive evolution .
Accurate DNA synthesis and repair of DNA lesions assure low rates of mutation . Genetic defects in either lead to increased , genome-wide mutator phenotypes which can manifest in cancer predisposition [1] . Strong mutators can cause reduced fitness due to the accumulation of dysfunctional alleles which place them under continual negative selection pressure [2] , [3] . By contrast , there would be little selection against transient hypermutability within subpopulations of cells or within limited genomic regions [4] . High mutation frequencies during transient hypermutability might contribute to overall population mutability and could represent a prominent source of multiple mutations important for evolution and cancer . Transient increases in localized mutation can even be regulated and benefit an organism , as found for somatic hypermutability in the immunoglobulin genes where a coordinated system of proteins induce DNA damage and repair the lesions in an error prone manner [5] . Examples of unregulated transient hypermutability include multiple mutations [4] , [6] , adaptive mutations in non-dividing cells [7] , meiosis-associated hypermutability [8] , hypermutability associated with repair of double-strand breaks ( DSBs ) [9] and increased mutability associated with senescence in telomerase-deficient yeast [10] . While the mechanisms remain to be explained , transient stretches of single-strand ( ss ) DNA have often been proposed as intermediates in hypermutation . Extended regions of ssDNA would be expected in association with DSBs and with dysfunctional uncapped telomeres , where many kilobases of ssDNA can be formed by 5′→3′ resection [11] , [12] . The genome would be expected to be especially vulnerable to lesions in ssDNA since repair mechanisms such as base-excision repair ( BER ) , nucleotide excision repair ( NER ) and post-replication repair ( PRR ) operate only in double-strand ( ds ) DNA . Restoration of damaged ssDNA to a ds-state would be expected to require translesion DNA synthesis ( TLS ) polymerases that are tolerant of lesions in a template and are often error-prone [13] due to misincorporation during synthesis past lesions . While error-prone restoration of damaged long stretches of ssDNA to dsDNA might explain many examples of transient hypermutability , this hypothesis has yet to be substantiated . Long tracts of ssDNA formed at DSBs and uncapped telomeres ( and potentially in other cell contexts ) are known to cause checkpoint responses , where the fate of the cell is determined by a balance of arrest , adaptation , recovery and repair processes [11] , [14] , [15] . Direct estimates of extent of damage tolerance and amounts of damage-induced mutations in ssDNA formed in a chromosomal context are required to establish that damage in long ssDNA actually leads to hypermutability rather than to cell death . We , therefore , designed systems in the budding yeast Saccharomyces cerevisiae for the generation of defined , persistent long stretches of transient ssDNA in the vicinity of a unique DSB or an uncapped chromosomal telomere . Lesions could be induced in the ssDNA before restoration to dsDNA . The consequences of restoration could be assessed with genetic reporters located within the regions that give rise to ssDNA . We demonstrate here that damage tolerance in stretches of long ssDNA formed at DSBs and uncapped telomeres results in extremely high frequencies of single and widely-separated multiple mutations caused by two major types of DNA damaging agents .
Recently , we demonstrated that , following the induction of a DSB in a yeast chromosome by the I-SceI site-specific endonuclease , there is 5′→3′ resection extending up to 20 kb from the break , after which repair can be mediated by a short oligonucleotide spanning the region of repair around DSB [16] . In this system , considerable DNA synthesis is required to restore the region to the ds-state . We employed this break-resection-oligonucleotide repair system to investigate the mutability of long ssDNA ( Figure 1 ) . Two types of strains were created wherein a cassette containing both an inducible Gal-I-SceI and a DSB target was inserted into either LYS2 on the centromere ( DSB-cen ) side or URA3 on the telomere ( DSB-tel ) side of the forward mutation reporter CAN1 . Placing cells in galactose medium for 3–6 hr led to induction of the DSB and resection . DNA damage was then induced by ultraviolet light ( UV ) or methyl methane sulfonate ( MMS ) . The cells were subsequently transformed with a pair of 95 nt self-complementary oligonucleotides that also had sequence homology to both sides of the break to allow DSB repair . As described in [16] only a portion of the cells experience a DSB . The cells that repaired a DSB with oligonucleotides gave rise to colonies on selective medium lacking lysine ( –Lys ) or uracil ( –Ura ) . No Lys+ or Ura+ colonies ( <10−9 ) were observed in controls lacking oligonucleotides ( data not shown ) . In the absence of additional DNA damage the median frequencies of DSB-induced transformation in the wild-type varied from 48 to 205×10−5 for the cells with a break at the DSB-cen position and from 14 to 26×10−5 for a break at the DSB-tel position ( Tables S1 and S2 ) . While this was 10–100 fold lower than observed for an induced DSB at the TRP5 gene on chromosome VII ( [16] and Tables S1 and S2 ) , the frequencies were 120–1300 fold in excess over the corresponding “no-DSB” controls ( containing the Gal-I-SceI insert , but lacking the I-SceI cut site ) . DNA damage ( UV or MMS ) increased transformation frequencies in the no-DSB controls , possibly by introducing recombinagenic DNA lesions . However , cells experiencing a DSB followed by exposure to UV or MMS yielded transformation frequencies that were still considerably greater ( 40–160 times ) than for the “no-DSB+DNA damage” controls . Thus , selection of individual DSB repair events resulting from restoration of a disrupted gene by oligonucleotides ( Lys+ or Ura+ colonies ) provides the opportunity to specifically examine subpopulations that had undergone DSB repair . The frequencies of spontaneous can1 mutants in the total populations of the DSB strains as well as in “no-DSB” control strains that were carried through all stages of transformation were 0 . 1–1 . 0×10−4 ( Figures 2 , 3 and Tables S1 and S2 ) . However , the can1 mutation frequencies were 50–300 times greater among the Lys+ ( or Ura+ ) selected colonies that arose after DSB induction and oligonucleotide repair . This is in agreement with observations of increased spontaneous mutation rates in the vicinity of a DSB repaired by recombination with a homologous chromosome or with a repeat in the same chromosome [9] , [17] . Doses of UV and MMS which caused 2–8 fold increases over low mutation frequencies in “no DSB” controls also amplified 3–30 times the high mutation frequencies associated with DSB repair ( Figures 2 , 3 and Tables S1 and S2 ) . Importantly , the frequencies of damage-induced can1 mutations among the cells that repaired a DSB were 180- to 1800-fold greater than in the population of the control , no-DSB strains treated with the same agents ( Figures 2 , 3 ) . Strong increases in the frequencies of repair-associated can1 mutants were observed regardless of whether the DSB was centromere or telomere proximal . The DSB-repair-associated spontaneous and damage-induced mutagenesis was observed only if the DSB was in the vicinity of the mutation reporter CAN1 . A DSB in a different chromosome ( TRP5 in chromosome VII ) did not affect mutagenesis in CAN1 ( Tables S1 and S2 ) . We utilized the special features of the cdc13-1 mutation to explore damage-induced mutagenesis in ssDNA generated at the ends of chromosomes with uncapped telomeres . Cdc13 , in complex with other proteins , protects chromosome ends after DNA replication by initiating telomere capping [12] , [18] . The mutant cdc13-1 is severely defective in postreplicative telomere capping at the nonpermissive temperature ( 37°C ) resulting in G2/M arrest . This arrest is associated with up to 15 kb 5′→3′ end resection that can be detected by ssDNA formation in 3–10% of individual chromosome ends [18] , [19] . In order to assess mutations in the subtelomeric region , 34 kb of nonessential DNA was removed from the left end of chromosome V and the mutation reporter LYS2 was inserted near the de novo telomere ( Figure 4A ) . The chromosome end was truncated to eliminate subtelomeric Y′ and X repeats in order to facilitate targeted integration and prevent silencing of the LYS2 reporter [20] . After shifting to non-permissive temperature for 6 hr to cause G2/M arrest and 5′→3′ resection , the cdc13-1 cells were UV-irradiated and plated onto rich ( YPDA ) medium . Mutagenesis was assessed by replicating colonies to –Lys medium . The frequency of lys2 mutants among Lys− colonies or sectors provided an estimate of mutagenesis next to a telomere . The appearance of Lys− mutants in genes other than LYS2 provided a comparison of telomere-associated vs . genome-wide mutagenesis within the same cell population . In addition to LYS2 , there are at least 7 genes scattered across the genome mutations in which can lead to lysine auxotrophy ( [21] and http://db . yeastgenome . org/cgi-bin/locus . pl ? locuslys ) . Frequencies and allelism of Lys− mutants were also determined in UV-irradiated cells that were not subjected to the 37°C arrest ( Figure 5 and Table S3 ) . Almost no mutants ( <10−4 ) were observed in the no-UV controls . Exposure to 45 J/m2 had little impact on survival of the cdc13-1 mutant regardless of irradiation at 37° or 23°C . However , the frequency of UV-induced lys2 mutants in the 37°C-arrested cells was at least 60 times that in the non-arrested control cells kept at 23°C . The low frequency of genome-wide UV-mutagenesis , as indicated by Lys− auxotrophic mutations at other loci ( “Lys− ( LYS2 WT ) ” bars ) , was not affected by the cdc13-1 arrest at 37°C . Genome-wide location of mutations in the Lys- ( LYS2 WT ) -category was confirmed by allelism test which assigned more than a half of them to 6 out of 7 genes with expected Lys- mutant phenotype ( Table S3 ) . PCR and sequencing analyses of spontaneous and UV-induced can1 and lys2 mutants did not identify any gross chromosome rearrangements which might occur at a low frequency through elimination of up to 43 kb of the nonessential region at the left end of the chromosome V [22] . The low frequency of GCR is likely due the high probability of loss of a chromosome having either an unrepaired DSB or an uncapped telomere . The following types of mutations were identified: substitutions of a single base pair ( referred as simple base substitutions ) , small insertions or deletions of one to three adjacent nucleotide ( indels ) as well as complex mutations which included clusters of individual mutations separated by ≤10 nt ( Tables S4 , S5 , S6 , S7 , S8 , S9 , S10 , S11 , and S12 ) . With one exception all categories of mutations were found among the spontaneous and UV-induced mutants regardless of the possibility for ssDNA generation . These categories were also found in previous studies that examined spontaneous and UV-induced mutations in can1 [23] , [24] . The striking feature of UV-induced mutation spectra associated with DSB repair or with uncapped telomere arrest was the presence of many multiple mutant alleles with as many as 6 widely-spaced mutations ( Table 1 , Tables S4 , S5 , S6 , S7 , S8 , S9 , S10 , S11 , and S12; also see next section ) . The distances between adjacent mutations within multiple-mutant alleles were usually greater than 100 bp ( ∼90% ) , reaching more than a few thousand nucleotides in the longer LYS2 gene . There was no apparent clustering of mutations ( Tables S13 and S14 ) suggesting that they could have originated from independent UV-lesions within the same DNA molecule . The high incidence of multiple mutations indicates that the level of UV-induced mutagenesis associated with DSB-repair or uncapped telomere arrest is even greater than estimated based simply on frequencies of loss-of-function CAN1 or LYS2 mutations ( Figures 2 and 5 ) . Therefore , we sought to estimate the actual frequencies of mutation per kilobase based on truncated Poisson distribution as described in [25] ( Table 1 , Table S15 and Text S1 ) . To estimate the probability of mutation it is necessary to know what fraction of all ( single and multiple ) mutant alleles result in loss of function . The fraction is expected to be smallest for single mutations and increase with multiplicity , thus multiple mutations would provide more accurate estimates of mutation probability . Therefore , for each set of data two types of calculations were developed based only on alleles with multiple mutations as well as based on the entire experimental distributions as described in Table S15 . While both calculations gave comparable results , Table 1 presents calculations based on multiple mutations . Each truncated Poisson distribution was calculated with two assumptions: all alleles included in a calculation lead to loss of function and the probability of mutation is the same for all survivors . The range of calculated densities of UV-induced mutations was 0 . 8–2 . 2 per target ( Table S15 ) . In the case of hypermutability associated with uncapped telomere arrest we obtained a direct estimate of the mutation density . For that purpose we sequenced the1848 nt NPR2 ORF located in the immediate vicinity of a telomere ( Figure 4A ) . We looked for the presence npr2 mutations in the subsets of UV-induced lys2 mutants that originated from the G1 and G2-arrested cdc13-1 cells . There were no npr2 mutations associated with 9 lys2 mutants originating from UV-irradiated G1-cells ( Table S11 ) ; however , we found 8 mutations in the npr2 ORFs of the 12 UV-induced lys2 mutants associated with uncapped telomere arrest ( P<0 . 02 ) . The density of mutations in NPR2 derived from this direct measurement is 0 . 36 mutations per kb , which is close to the density of mutations in the more distant telomere LYS2 calculated on the basis of truncated Poisson distribution ( Table 1 ) . The fit between the experimental numbers of multiple mutants and the numbers expected from truncated Poisson distributions was high ( 0 . 4<P ( χ2 ) <0 . 9 ) indicating that the mutations were independent and that there was no preferential loss of ssDNA molecules with higher density of damage within the population of damaged molecules . However , the relative survival of damaged vs undamaged ssDNA remains to be established using systems that would provide synchronous generation of ssDNA by 5′→3′ resection in all cells ( such as site-specific DSB caused by inducible HO-endonuclease ) . Extrapolations of the calculated mutation densities yielded probabilities of 0 . 55–0 . 88 that cells contain at least one mutation ( Pmut = 1−P0 ) . This is 10 to 100 times greater than the observed frequencies of UV-induced mutants associated with DSB repair or uncapped telomere arrest ( Figures 2 , 5 and Tables S1 and S3 ) . This discrepancy could be explained if there is a high likelihood of UV-induced mutants not being detected and/or by a non-uniform distribution of UV-induced mutability among cells . The latter would be expected if UV-induced hypermutability in ssDNA is associated with variable amounts of 5′→3′ resection . In this case the extremely high likelihood of mutation would be a feature of only a small fraction of survivors where resected DNA extended through the reporter at the time of UV-irradiation . Regardless of the reason for the discrepancy , the high incidence of multiple mutant alleles provides evidence for exceptionally high levels of UV-induced hypermutability associated with DSB repair or uncapped telomere arrest . The primary products of UV-damage as well as sources of mutations are cyclobutane pyrimidine dimers and 6-4 photoproducts resulting from covalent linkages of adjacent pyrimidines [26] . Nearly all UV-induced simple base substitutions ( substitutions that were not the part of a complex mutation ) in single and multiple mutants associated with DSB-repair ( can1 ) or uncapped telomere arrest ( lys2 ) could be attributed to pyrimidine bases of the unresected strand ( Figure 6A and Table S16 ) . Importantly , the bias was always directed to changes at pyrimidines of the unresected strand regardless of the position of the DSB ( compare DSB-cen and DSB-tel ) . The large majority of pyrimidines mutated in the unresected strand were associated with di-pyrimidines ( 128 out of 147; derived from data in Table S16 ) . A strong bias towards pyrimidines in unresected strands was also observed for base substitutions identified within UV-induced complex mutations associated with DSB-repair and with uncapped telomere arrest ( 33 pyrimidines∶1 purine; derived from data in Table S18 ) . The strand bias associated with ssDNA-forming conditions contrasted with the nearly equal frequencies of UV-induced substitutions of purines and pyrimidines in either strand in the absence of a DSB or cdc13-1 arrest ( Figure 6A and Table S16 ) . Additional strand bias analysis for 39 multiple mutant alleles of CAN1 and LYS2 with more than one simple base substitution ( Figures 6B and 6C ) revealed that only 11 of 92 base substitutions in these alleles could be assigned to purines in the unresected strand ( P<0 . 001 for the hypothesis that mutations in purines and pyrimidines occur with probabilities proportional to their presence in the CAN1 and LYS2 ORFs ) . This supports the view that multiple mutations resulted mostly from independent UV-photoproducts rather than from extended , error-prone DNA synthesis triggered by a single UV lesion . In the DSB-cen strains the URA3 gene was located telomere proximal to CAN1 ( Figure 1A ) . In a small fraction ( 1–3% ) of UV-induced can1 mutants associated with DSB-repair the URA3 gene was also inactivated , mostly by single mutations ( Table S9 ) . Although the number was small , multiple mutations were also observed in URA3 . The lower incidence of multiple mutations could be due to the size of the ORF ( 804 nt , URA3; 1773 nt , CAN1 ) , a potentially lower probability of URA3 inactivation by mutations as well as to more distant location from a DSB than CAN1 . Importantly , all 22 base substitutions identified in URA3 resulted from mutations in pyrimidines in the strand that would remain after 5′→3′ resection into this gene . In total can1 ura3 double mutants ( Tables S8 and S9 ) contained from two to four mutations ( e . g . , isolates #15 and #28 ) most of which were strand-specific . Based on reported lengths for 5′→3′ resection at a site-specific DSB ( [11] and references therein ) , we propose that the multiple UV-induced mutations in both URA3 and CAN1 originated from damage in large continuous stretches of ssDNA . For the case of the induced DSB , the continuous ssDNA regions would have to be >6 . 3 kb ( distance between the DSB-cen and the 5′-end of URA3; see Figure 1A ) and >6 . 4 kb for the uncapped telomere ( distance between the Tel-de novo and the 5′-end of LYS2; see Figure 4A ) . The continuity of subtelomeric hypermutable regions can also be derived from the observed multiple , strand-biased UV-induced mutations in NPR2 and LYS2 ( previous section and Table S12 ) . These observations strongly support the view that the hypermutable region is created by 5′→3′ resection starting at the telomere and extending through both ORFs . Altogether , the strong strand bias of UV-induced single and multiple mutations demonstrates that long stretches of damaged ssDNA are the source of mutagenesis and that the ssDNA can be successfully restored to the ds-state even if it contains multiple lesions . To address the mechanism ( s ) of mutation during restoration of damaged ssDNA to complete chromosomes , we investigated the roles of genes involved in translesion DNA synthesis ( TLS ) and post-replication repair ( PRR ) on mutagenesis associated with DSB-repair ( Figures 2 , 3 and Tables S1 and S2 ) . Similar to the other forms of damage-induced mutagenesis [13] , hypermutability associated with DSB-repair required the translesion polymerase ( Pol ζ . Deletion of REV3 , encoding the catalytic subunit of Pol ζ eliminated most of the UV- and MMS-induced mutagenesis while removal of the translesion polymerase Pol ζ ( rad30Δ ) had no detectable effect ( Figures 2 and 3 ) . The remaining level of UV-mutagenesis associated with DSB-repair was generally decreased further in a rev3Δrad30Δ double mutant . We also examined the impact of a rev1Δ mutation [27] and a PCNA-mono-ubiquitination defect ( pol30-K164R ) [28] ) , which can lead to indirect disruption of Pol ζ TLS function . Both defects also prevented the DSB repair-associated mutagenesis providing additional support that mutagenesis relies on Pol ζ . Mutagenesis was not affected by deletions of the UBC13 or RAD5 genes , responsible for steps in PCNA poly-ubiquitination and error-free damage avoidance by template switching [28] . Similar to findings by Jeff Strathern and co-authors [17] , spontaneous hypermutability associated with DSB-repair was also reduced 2–30 fold by inactivation of Pol ζ mediated translesion synthesis ( Figures 2 and 3 ) . Possibly there is damage to ssDNA prior to DSB repair that results in spontaneous hypermutability . We considered the possible contribution of greater variability and apparent overall reduction in transformation frequencies , regardless of DNA damage , to these results . For example , if transformants arose through recombination between unbroken chromosomal DNA and oligonucleotides [29] , then there would be little contribution of ssDNA to mutagenesis . However , we found that most transformants were indeed associated with DSB repair . In experiments involving MMS treatment of the various mutants , the DSB-induced transformation frequencies were 30- to 1000-fold higher than found with the wild type no-DSB controls ( Table S2 ) . For the case of UV-irradiation , some TLS polymerase single and double mutants showed as little as a 2- to 10-fold increase in transformation frequencies over wild type no-DSB controls , while others had >10-fold increases ( Table S1 ) . However , since even for the 2-fold increase at least 50% of the transformants could be assigned to events associated with oligonucleotide mediated DSB repair , drastic effects of TLS mutations on reductions in mutagenesis could be detected . Similar to the observations with DSB-repair associated mutagenesis , the UV-induced hypermutability of LYS2 associated with uncapped telomere arrest depended on the translesion Pol ζ ( REV3 ) and was not affected by elimination of Pol ζ ( RAD30 ) ( Figure 5 ) . ( Note: the rad30 mutation caused a moderate increase in the frequency of Lys− ( LYS2+ ) cells independently of uncapped telomere arrest . This could be due to a stronger role for Pol η in error-free TLS among the unidentified LYS genes as compared to LYS2 . )
Utilizing systems that can create resected chromosomal DNA , we found that long stretches of ssDNA formed at DSBs and uncapped telomeres ( >6 kb based on the region of detectable mutations ) can be efficiently restored to the ds-state in spite of the presence of multiple UV-induced lesions . Lesions in ssDNA along with ssDNA itself were hypermutable and hypermutability relied on TLS Pol ζ . We established the involvement of a ssDNA intermediate in transient hypermutability based on the strand-biased spectrum of UV-induced single and multiple mutations . Importantly , strand biased multiple mutations provide a useful tool for detecting in vivo mutagenesis in single molecules containing long stretches of damaged ssDNA . UV or any other mutagen with a bias towards certain DNA bases would create a permanent signature of strand-biased multiple mutations in chromosomal regions containing long stretches of ssDNA at the time of acute induction of lesions . Importantly , the density of UV-induced mutations was estimated to be ∼1 per 1 . 5–3 kb , which is comparable to the density of pyrimidines dimers induced by UV as determined in yeast [30] , [31] . Thus , we conclude that many UV-lesions in ssDNA directly result in mutations during the DNA synthesis that leads to chromosome restoration . Based on the similarity in genetic controls and responses to UV and MMS , we propose that there is a general phenomenon whereby lesions in long ssDNA tracts lead to hypermutagenesis . Remarkably , the mutation frequencies can reach 0 . 4–0 . 8 per kb ( Table 1 ) . This is comparable to the very high frequencies found for somatic hypermutation of immunoglobulin genes [5] . Since survival in our experiments is high and the mutation load in the rest of the genome is low , we anticipate that damage-induced localized hypermutability associated with ssDNA can achieve extremely high levels . Transient long stretches of ssDNA can form in several ways . Tens of kilobases of ssDNA can be formed in yeast by 5′→3′ resection at site-specific DSBs [11] and uncapped telomeres [12] . Recently , we found that ssDNA is also formed by resection at random DSBs induced by ionizing radiation ( J . Westmoreland , W . Ma and M . Resnick , unpublished ) . Also , there are strong indications of ssDNA formation at DSBs in mammalian cells [32] , [33] . Since delays in DSB-repair or in telomere recapping would increase the time of exposure and the length of ssDNA tract , resulting in hypermutability , it will be important to identify factors responsible for coordination of steps in DSB-repair and/or telomere maintenance . These factors would suppress the potential for hypermutability associated with ssDNA . In addition to end-resection , there are several possible sources of ssDNA . Long ssDNA has been detected in cultured human lymphoma cells , although the mechanism of generation has not been ascertained [34] . Uncoupled leading and lagging strand synthesis resulting from occasional disruption of the replisome or DNA damage can lead to ssDNA regions [35] , [36] . The ssDNA created by replication fork uncoupling might be highly prone to damage-induced mutagenesis . The similarity in hypermutability associated with DSB-repair and uncapped telomere suggests that there is generally a strong potential for genome instability associated with transient stretches of long ssDNA . We observed a high incidence of widely-spaced multiple mutations caused by damage in ssDNA . Multiple mutations are an important genetic phenomenon that may affect species evolution and cancer incidence [4] , [6] , [37] . They have a greater likelihood of conferring a selective advantage as compared with each single mutation which might have just a modest , neutral or even negative effect on gene function [38] . Our work provides a simple molecular mechanism for the simultaneous occurrence of multiple mutations across regions that may be several kilobases long , recently described as “mutation showers” [6] .
All yeast strains were isogenic to CG379 [39] with the following common markers , MATα ade5-1 his7-2 leu2-3 , 112 trp1-289 ura3Δ All strain constructions were performed using methods of PCR-based gene disruption and direct genome modification by oligonucleotides as described in [16] , [40] , [41] and references therein . Deletion strains for genetic control studies were generated by inserting antibiotic resistance markers as described [42] . Single deletions were obtained by inserting the G418 resistance ( kanMX4 ) module in place of the chosen ORF . Double-deletion strains were obtained by switching the G418 resistance marker to nourseothricin resistance ( kanMX4 to natMX4 ) followed by kanMX4 insertion to replace the second ORF . All construction steps were verified by phenotype and by PCR . The site-specific pol30-K164R mutant was generated as described in [43] and verified by sequencing . The strain FRO-1 with a self-generating DSB cassette inside the TRP5 gene in the chromosome VII used in control experiments has been described [16] . At the first step of construction the LYS2 gene was inserted after position 34 , 193 nt in the chromosome V with the help of the tailed PCR primers: oYY_a - TTCTTACTCAGTGTGAACGTGTTCTAAATAAGTTCTTGTTCTAATTAATT-TAAGCTGCTGCGGAGCTTCC and oYY_b - AGATACGATTACTCCAGTTCCTCTTACAAGAAATGCATAAAAATAGTTAC-AATTACATAAAAAATTCCGGCGG ( targeting tails are underlined; LYS2-amplification tails are in shown in bold ) . Then , a CORE-I-SceI cassette containing the I-SceI gene under control of the GAL1 promoter , the hygromycin resistance gene ( Hyg ) and an I-SceI site was placed inside the LYS2 flanked by the following sequences: GCTTGCCGACGGCGGCTAAGCTCATAACATTGATAGTTGAAATAACATTTGGA - telomere proximal and CACAGTTGATATAATTATCCATAATGGTGCGTTAGTTCACTGGGTTTATCCATATGCCAAATTGAGGGA–centromere proximal ( insertion strategy and sequences of oligonucleotides are available upon request ) . At the second step the URA3 gene was inserted after position 29 , 682 in chromosome V with the help of the tailed PCR primers: oYY_5 - CAGTCTTTATATCAGTCTTTGCATGGCTTTGCATCTGATGCTGGCTCTACCGACTTCTCG-CAGAGCAGATTGTACTGAGAGTGCACC and oYY_6 - ACTCAAATATTTATCCACTTTGGATAGTATATACGTCAAATTCTTTTGGTATTTTATCGC-CGCATCTGTGCGGTATTTCACACCGC ( targeting tails are underlined; URA3-amplification tails are in shown in bold ) . Two independent isolates , YY-22 and YY-24 were used as wild type DSB-cen strains for this study . The control “no DSB” strains ( YY_122 and YY_124 ) differ from YY_22 and YY_24 in that they contain uncuttable , incomplete ( half ) I-SceI sites . In order to place the self-generating DSB cassette on the telomere side of CAN1 ( DSB-tel ) we first transformed YY_22 and YY_24 strains to Lys+ with “repairing” oligonucleotides oYY_15 and oYY_16 ( see below ) and selected the removal of the Gal-I-SceI-CORE insert from the LYS2 gene . Then , a CORE-I-SceI cassette containing the I-SceI gene under the GAL1 promoter , hygromycin resistance gene ( Hyg ) and an I-SceI site was placed inside the URA3 gene flanked by the following sequences: CAAGGAATTACTGGAGTTAGTTGAAGCATTAGGT ( telomere proximal ) and TATCCACATGTGTTTTTAGTAAACAAATTTTGGG ( centromere proximal ) to produce two identical DSB-tel strains YY-266 and YY-267 . In order to generate a chromosome end-truncation and also create a de novo telomere we inserted Gal-I-SceI CORE after 34 193 nt at the left end of chromosome V ( sequence context: TTCTTACTCAGTGTGAACGTGTTCTAAATAAGTTCTTGTTCTAATTAATT–on the telomere side and GTAACTATTTTTATGCATTTCTTGTAAGAGGAACTGGAGTAATCGTATCT–on the centromere side ) . After induction of the DSB , cells were transformed with the oligonucleotide TGTGTGTGGGTGTGGTGTGTGTGTGGGTGTGGTG-GTAACTATTTTTATGCATTTCTTGTAAGAGGAACTGGAGTAATCGTATCT consisting of a short stretch of T ( G ) 1–3 telomeric repeats ( shown in bold ) followed by sequence homologous to the DSB-end on the centromere side . Loss of CORE was selected and isolates with end-truncation and de novo telomere V were identified by both Southern hybridization of the PstI-digested genomic DNA with NPR2-probe and by PCR with the following pair of primers: CA-16–CACCACACCCACACAC , telomere repeat-specific and Tel5-FS-npr2–GAACATTTTGCCCAGCCTAGTA , NPR2-specific . Based on these tests the size of added telomeric repeats was 300–500 nt ( not shown ) . The LYS2 ORF with promoter was amplified with targeting tailed primers oYY_9 ( TAATATTACAACTTATTTCCGTAAATAAAGATAGTACACACGAATCCAAACGTTTATATAGTTAGCTCTG-TAAGCTGCTGCGGAGCTTCC ) and oYY_10 ( AGACAGAAGAGAAGGGTGTGAAACCACCTCTACCAAACACACCAAGAGATGAACCTAAATCAAATTTTCA-AATTACATAAAAAATTCCGGCGG ) and inserted between NPR2 and CIN8 close to the de novo telomere ( Figure 4A ) . Strain DAG_760 was generated by introducing the cdc13-1 temperature-sensitive mutation into this strain with the help of the replacement plasmid pVL2862 containing the mutant cdc13-1 allele and the URA3 marker ( gift from Vickie Lundblad ) . cdc13-1 strains were always grown at permissive temperature 23°C . Deletion mutants in the genes rev3 and rad30 controlling TLS polymerases were obtained in the strain DAG_760 . Two independent deletion isolates was studied for each genotype . Yeast strains were grown with agitation in liquid rich media ( YPDA ) for approximately 20 hr and then diluted 37 fold with fresh 2% galactose synthetic complete media to induce Gal-I-SceI expression for generation of a site-specific DSB . After 3 or 6 hr of growth in galactose cells were washed once with 50 ml dH2O and DNA damage was applied to the yeast suspension . The “No-DSB” control strains went through the same incubation and transformation steps as the strains that experienced an inducible DSB . In the case of UV , yeast were suspended in 25 ml of dH2O ( ∼106 cells/ml ) , placed into a 150 mm Petri dish and irradiated with UV-C ( 254nm , 1 J/m2×sec ) with continuous agitation for 20 or 45 seconds . In the case of MMS , cells were suspended in 5 ml of 50 mM sodium phosphate buffer ( pH 7 . 0 ) ( ∼107 cells/ml ) and then treated with 11 . 8 mM ( 0 . 1% ) MMS for 15 or 30 min . MMS was inactivated by adding Na2S2O3 to a final concentration of 5% ( w/v ) for 2 minutes . Treated and untreated yeast were washed once with 50 ml water followed by transformation to Lys+ or Ura+ in DSB-cen or DSB-tel strains , respectively , with a pair of complementary oligonucleotides ( Figure 1 ) introducing a silent change ( shown in small letters ) creating additional restriction sites ( underlined ) , AvaII ( DSB-cen ) or ApaI ( DSB-tel ) . DSB-cen strains were transformed with a pair of oligonucleotides oYY_15–AATGGTGCGTTAGTTCACTGGGTTTATCCATATGCCAAATTGAGGGAcCCAAATGTTATTTCAACTATCAATGTTATGAGCTTAGCCGCCGTCGG and oYY_16 - CCGACGGCGGCTAAGCTCATAACATTGATAGTTGAAATAACATTTGGgTCCCTCAATTTGGCATATGGATAAACCCAGTGAACTAACGCACCATT; DSB-tel were transformed with a pair of oligonucleotides oYY_47–TGTTCGTACCACCAAGGAATTACTGGAGTTAGTTGAAGCATTAGGgCCCAAAATTTGTTTACTAAAAACACATGTGGATATCTTGACTGATTTTT and oYY_48 - AAAAATCAGTCAAGATATCCACATGTGTTTTTAGTAAACAAATTTTGGGcCCTAATGCTTCAACTAACTCCAGTAATTCCTTGGTGGTACGAACA . can1 mutants were screened among Lys+ or Ura+ transformant colonies after replica plating onto media with 60 mg/ml of L-canavanine . The cdc13-1 strain DAG_760 or its derivatives with additional mutations in TLS polymerases , carrying LYS2 next to the de novo telomere in the truncated left arm of the chromosome V ( Figure 4A ) , were grown with agitation for 72 hr in YPDA at the permissive temperature 23°C . At this point the cell density reached 1 to 2×108 cells/ml and more than 95% cells were in G1 , based on cell morphology . Part of the culture was left at 23°C while the other part was diluted ten times with fresh YPDA and incubated at the non-permissive temperature 37°C for 6 hr . After additional incubation all cultures were diluted to 1–5×105 cells/ml . A portion of the suspensions were UV-irradiated ( 45 J/m2 ) . All suspensions were plated onto YPDA to yield around 100–500 colonies per plate . Colonies were replica plated onto synthetic complete and lysine deficient media . Single cell isolates from Lys− colonies or sectors were then verified for the phenotype and checked for allelism to the LYS2 gene . | A variety of error avoidance mechanisms assure low mutation rates across the genome . Genetic defects in DNA replication or repair can lead to genome-wide increase in mutation frequency that may result in cancer predisposition and genetic disease . Transient localized hypermutability drastically differs in its biological consequences from genome-wide mutators . Since genome-wide hypermutability can cause reduced fitness due to accumulation of dysfunctional alleles , mutators are under negative selection pressure . By contrast , there would be less selection against temporary hypermutability within limited genomic regions , suggesting a special role in adaptive evolution and carcinogenesis . Mechanisms of transient hypermutability are poorly understood . Long stretches of single-strand DNA have been implicated but not demonstrated as a source of localized transient hypermutability . Using sophisticated yeast genetic systems that we developed , we found that transient stretches of chromosomal single-strand DNA at double-strand breaks and that telomeres can tolerate multiple lesions and are highly prone to damage-induced mutations , including a very unusual class of widely spaced multiple mutations . The hypermutability relied on error prone translesion DNA synthesis . Our work demonstrates a simple in vivo mechanism for localized transient hypermutability extending over several kilobases that can result in widely spaced multiple mutations without severe mutation load in the rest of the genome . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
]
| [
"molecular",
"biology/dna",
"repair"
]
| 2008 | Hypermutability of Damaged Single-Strand DNA Formed at Double-Strand Breaks and Uncapped Telomeres in Yeast Saccharomyces cerevisiae |
Organogenesis is controlled by gene networks activated by upstream selector genes . During development the gene network is activated stepwise , with a sequential deployment of successive transcription factors and signalling molecules that modify the interaction of the elements of the network as the organ forms . Very little is known about the steps leading from the early specification of the cells that form the organ primordium to the moment when a robust gene network is in place . Here we study in detail how a Hox protein induces during early embryogenesis a simple organogenetic cascade that matures into a complex gene network through the activation of feedback and feed forward interaction loops . To address how the network organization changes during development and how the target genes integrate the genetic information it provides , we analyze in Drosophila the induction of posterior spiracle organogenesis by the Hox gene Abdominal-B ( Abd-B ) . Initially , Abd-B activates in the spiracle primordium a cascade of transcription factors and signalling molecules including the JAK/STAT signalling pathway . We find that at later stages STAT activity feeds back directly into Abd-B , initiating the transformation of the Hox cascade into a gene-network . Focusing on crumbs , a spiracle downstream target gene of Abd-B , we analyze how a modular cis regulatory element integrates the dynamic network information set by Abd-B and the JAK/STAT signalling pathway during development . We describe how a Hox induced genetic cascade transforms into a robust gene network during organogenesis due to the repeated interaction of Abd-B and one of its targets , the JAK/STAT signalling cascade . Our results show that in this network STAT functions not just as a direct transcription factor , but also acts as a "counter-repressor" , uncovering a novel mode for STAT directed transcriptional regulation .
Organogenesis is controlled by the activation of complex gene regulatory networks in precise positions of the organism [1] . Selector genes , or master regulator genes , encode transcription factors required for the expression of entire organogenetic gene regulatory networks and , when expressed ectopically , can induce the formation of additional organs at new locations [2 , 3] . Prominent examples are Eyeless ( Ey ) capable of inducing ectopic eyes; or the Hox proteins capable of inducing segment specific organs [1 , 4] . However , these genes by themselves are unable to provide all the information required to specify an organ . For example , the organ specified by a particular Hox protein varies at different positions of the segment depending on its interaction with tissue-specific transcription factors and signalling pathway effectors active in each region . Similarly , ectopic Ey expression can only induce the formation of additional eyes at certain locations of the imaginal discs showing that the functional outcome of these proteins is locally modulated . Once a regulatory gene network is selected , each network gene has to be precisely activated in time and space and this is usually controlled through the gene’s non-coding cis-regulatory modules ( CRMs ) . Thus , CRMs play a critical role since they act as integrators of specific combinations of transregulatory transcription factors and signalling pathway effectors , resulting in localized transcriptional activity . Examples of Hox-induced organogenetic networks include the formation of the corpora allata or the maxillary cirri by Deformed ( Dfd ) ; the development of the prothoracic or the salivary glands by Sex combs reduced ( Scr ) ; or the formation of the posterior spiracles by Abdominal-B ( Abd-B ) [5–8] . In all these cases the Hox protein is the most upstream activator of an organogenetic gene-cascade . This simple and linear view of the role of Hox genes in the establishment of gene networks , contrasts with what is known for other selector genes . Eye development shows a more complex situation where Ey is not simply upstream of a linear cascade but forms part of the gene network itself , collaborating with other transcription factors as Eyes absent , Sine oculis or Dachshund that are equally important to induce eye organogenesis [9] . The formation of the posterior spiracles of Drosophila is an excellent model to study how a Hox protein controls organogenesis [8 , 10] . The posterior spiracles constitute the only external opening of the trachea when the larva hatches so their development has to be completed during embryogenesis . Posterior spiracle organogenesis is induced in the eighth abdominal ( A8 ) segment when the Abdominal-B ( Abd-B ) Hox selector protein activates a series of primary targets that include the transcription factor genes cut ( ct ) , empty spiracles ( ems ) and spalt ( sal ) as well as the unpaired ( upd ) and upd2 genes encoding ligands for the JAK/STAT pathway receptor . Downstream of these primary targets , a number of secondary target genes are activated which require the activity of the primary targets to be expressed . Secondary targets include other transcription factors , but also genes encoding realizators of organogenesis , like non-classic cadherins involved in cell adhesion , RhoGAP and RhoGEF GTPase regulatory proteins influencing cytoskeletal organization , and cell polarity determinants [11] . Central among the cell polarity genes is crumbs ( crb ) , one of the major determinants of apico-basal polarity [12 , 13] . Crb is a transmembrane protein that localizes to the subapical region of all ectodermal cells where it is required to maintain the epithelial structure by interaction with cortical proteins also involved in apico-basal polarity regulation . Although Crb is ubiquitous in all ectodermal cells , its expression is increased in the cells that invaginate to form the posterior spiracles ( S1A and S1B Fig and [11] ) . The enhancement of crb transcription in the spiracle primordium is controlled by a posterior spiracle specific CRM regulated by JAK/STAT signalling ( S1D and S1E Fig ) . In turn , JAK/STAT signalling in the posterior spiracles depends on upd transcription that is controlled by Abd-B ( S1C and S1F Fig and [11] ) . Although this simple linear model , where the crb-spiracle enhancer expression is solely dependent on JAK/STAT activation , can explain why it is expressed in the spiracle primordia of A8 ( S1C and S1D Fig ) , it does not explain why this enhancer is not also activated in the tracheal pits , where upd is transcribed and the JAK/STAT signalling cascade is also active ( S1C and S1G–S1G” Fig ) . Thus , the regulation of the crb posterior spiracle CRM must be more complex to attain organ specificity . To understand how organ specific gene expression is regulated in a Hox-induced morphogenetic cascade during organogenesis , we have studied the transcriptional activation of crb in the posterior spiracles . Our results show that spiracle specific expression of this downstream secondary target requires the cooperation of the primary targets and the Abd-B Hox selector protein . Moreover , we show that the simple linear cascade initiated by Abd-B early in embryogenesis , soon becomes a complex network due to feedback loops set up by the primary targets . Furthermore , we find evidence that STAT can influence gene expression in a novel way , not just as a positive transcription factor , but also as a counter-repressor . The crb posterior spiracle expression is controlled by an unanticipated number of interactions between modular elements of the CRM that influence the final transcriptional outcome .
The reporter construct crb43 . 2-lacZ contains a 2kb crb intronic region driving posterior spiracle expression . It was shown that the activity of this reporter depends on activation of JAK/STAT signalling and that mutation of the STAT binding sites present in this element downregulate its activity [[11 , 14] and S1 Fig , compare E and F] . These initial studies suggested that the crb43 . 2 enhancer is a direct target of the JAK/STAT pathway . To confirm this observation , we expressed activated STAT-GFP in S2 cells and performed Chromatin Immunoprecipitation ( ChIP ) using an anti-GFP antibody ( S2A Fig ) . The enrichment of input recovery shows that STAT is able to bind directly to the crb enhancer , reinforcing the view that crb is a direct target of the JAK/STAT pathway . To identify the minimal sequence requirements for crb posterior spiracle specific expression we compared crb43 . 2 sequence among several Drosophila species ( S2B Fig ) . As the initial characterization of crb43 . 2 indicated the requirement of two low affinity STAT binding sites [14] , we tested a highly conserved 518 bp genomic fragment centred around these sites ( S2B Fig ) . The 518 bp element was cloned into a lacZ reporter transformation plasmid containing a minimal promoter and lacZ expression was analyzed in transgenic flies . Like the original crb43 . 2 enhancer , crb518 directs expression of the transgene specifically in the posterior spiracles ( S2C Fig ) . Moreover , the activity of crb518 is also regulated by JAK/STAT signalling as both inactivation of the pathway or mutation of the STAT binding sites result in the downregulation of the enhancer’s activity ( S2E and S2F Fig ) . These results show that the crb518 enhancer constitutes a minimal element able to direct specific expression of crumbs to the posterior spiracle in a JAK/STAT dependent manner . To understand the molecular mechanisms by which STAT activates crumbs in the posterior spiracles , we further dissected the crb518 minimal enhancer . Based on the position of the STAT binding sites , the crb518 enhancer can be subdivided into three regions ( Fig 1A ) : a region encompassing the first 204 nucleotides ( 1–204 , yellow ) , a 101 bp region containing the STAT binding sites ( 205–305 bp , red ) and a region including the last 213 nucleotides ( 306–518 bp , green ) . Transgenes were designed with different combinations of these regions and their expression patterns characterized . When the 1–204 bp region is deleted ( crb313 ) , spiracle expression is lost ( Fig 1C ) . This result suggests that the STAT binding sites although required , are not sufficient to direct expression to the posterior spiracles and that additional factors binding to the 1–204 fragment ( yellow ) are crucial . This conclusion is further supported by the inability of the crb101 construct , which contains the 205–305 fragment ( red ) including both STAT sites , to direct expression in the spiracles ( Fig 1D ) . These observations are in accordance with our finding that the crb305 enhancer , where the 306–518 fragment ( green ) is deleted , is able to direct expression to the posterior spiracle ( Fig 1E ) . Moreover , the activity of crb305 is still dependent on the JAK/STAT pathway as its expression is downregulated in the absence of all three unpaired ligands ( Fig 1F ) . Therefore the crb305 enhancer seemed to contain all the cis-regulatory elements required to direct expression to the posterior spiracle in a JAK/STAT dependent manner . Finally , we tested the expression of the crb204 construct , which contains the 1–204 fragment ( yellow ) but lacks the STAT binding sites . Surprisingly , the crb204 is able to direct expression to the posterior spiracles ( Fig 1G ) . Moreover , this expression is dependent on JAK/STAT pathway activity as in embryos mutant for the unpaired ligands the enhancer is downregulated ( Fig 1H ) . Taken together , these results show that crb204 contains the crb posterior spiracle specificity element . Moreover , the expression of this construct suggests that although the activity of the JAK/STAT pathway is required to activate the crb spiracle expression , the STAT binding sites are dispensable . This seems to contradict our previous observations where mutation of the STAT sites resulted in the downregulation of the larger crb43 . 2 and crb518 enhancers . The spiracle expression of the crb204 enhancer was unexpected not only because it did not contain the STAT binding sites known to be required for the larger enhancers’ activity but also because despite the absence of STAT sites the enhancer is still regulated by the JAK/STAT pathway . These results prompted us to consider the possibility of cryptic STAT binding sites in crb204 that could account for these observations . Therefore , we searched for putative additional STAT binding sites that differed in one or two nucleotides from the consensus TTCNNN ( N ) GAA sequence [14 , 15] . Using this relaxed criterion , we identified fifteen possible cryptic binding sites spread along crb305 ( S3A Fig , asterisks ) , twelve of them in crb204 . To determine if STAT is able to bind in vitro to any of these putative sites , EMSAs were performed with activated STAT92E protein incubated with oligonucleotides containing the different cryptic sites ( S3B Fig ) . As controls we used oligonucleotides containing the STAT consensus binding sites as well as mutated versions of these sites ( Materials and Methods ) . In these conditions , STAT92E is only able to bind the previously identified consensus sites , while it fails to bind the mutated STAT sites or any of the putative cryptic sites ( S3B Fig ) . Therefore , these experiments exclude the possibility of cryptic binding sites being responsible for the activity of crb204 . Taken together , these results suggest that the JAK/STAT pathway is required to specify expression in the posterior spiracles most likely by indirectly regulating factors binding to the 1–204 specificity element ( yellow ) . This detailed analysis reveals that STAT does not function as a crb spiracle enhancer coactivator , but has a different function in the context of the crb enhancer . Mutation of the STAT sites in both the original crb43 . 2 enhancer [11 , 14] and in the crb518 enhancer indicated that the STAT sites are required for full activation ( S1E and S1F and S2C–S2F Figs ) . However , the spiracle activity of crb204 excluded the possibility that the direct binding of STAT is necessary to direct expression to the posterior spiracles ( Fig 1G ) . These contradictory results can be accommodated in a model where the specificity element is silenced by a nearby repressor element; the function of STAT being to prevent the repression , thereby , allowing spiracle specific activation ( Fig 2A and 2A’ ) . In this hypothesis the JAK/STAT pathway would play a dual role: first , it would be required indirectly to activate the specificity element by regulating activator factors; and second , it would be required directly to overcome the repression of the enhancer by counteracting a repressor element present in the CRM . Mutation of the STAT sites would prevent the binding of STAT to the CRM disabling this second counter-repression role resulting in the downregulation of expression controlled by the specificity element . As our model predicts the existence of a repressor element , we set to identify its location by deletion of DNA in a crb518 enhancer with mutated STAT sites ( Fig 2B and 2C ) . The logic behind this experiment was that the deletion of the repressor element should allow the activity of the specificity element in the absence of STAT binding sites . We found that removal of the 306–518 fragment ( crb305-STAT-mt ) results in the expression of the enhancer in the posterior spiracles ( Fig 2B’ and 2C’ ) confirming the existence of a repressor element in the 306–518 distal region ( green region ) . To further support our findings we generated an additional construct fusing the putative repressor module to the crb spiracle specificity module ( Fig 2D ) . As expected from the presence of a repressor element , the expression of this transgene in the posterior spiracles is downregulated ( Fig 2D’ ) . To determine whether the STAT binding sites and the repressor element present in the 205–518 fragment are able to function as regulatory modules independent of the specific context of the crb spiracle specificity module , we fused the crb313 fragment ( red and green in Fig 1A ) , which does not drive expression in the posterior spiracles ( Fig 1C ) , to ems0 . 35 ( Fig 3A and 3B ) , an unrelated posterior spiracle enhancer of the empty spiracles gene [16] . Although the crb313-ems0 . 35 fusion construct generates a novel gut pattern of expression not present in any of the original constructs , ems0 . 35 spiracle activity is still present when fused to the crb313 element ( Fig 3A and 3B and 3D and 3E ) . In contrast , fusion of the crb313 element with the STAT sites mutated results in the loss of the posterior spiracle expression driven by ems0 . 35 ( Fig 3G and 3H ) . These results further support the presence of a repressor element in the 306–518 fragment that would be counteracted by the STAT sites , and reveal the modular nature of the crb518 CRM . Moreover , while the ems0 . 35 enhancer by itself is expressed independently of STAT function ( Fig 3B and 3C ) , it becomes STAT dependent when fused to the crb313 element ( Fig 3E and 3F ) . Taken together , these data show that crb518 is structured into three different regions , each with a specific function: an activator element driving spiracle specificity ( 1–204 , yellow ) , a STAT-binding element that contains the conserved STAT binding sites ( 205–305 , red ) and a repressor element ( 306–518 , green ) . Analysis of the 306–518 DNA sequence did not provide any strong indication regarding the nature of the repressor protein . To fine map the repressor region , we constructed transgenes where we added increasing size fragments distal to crb305-STAT-mt ( S4B Fig ) . In these constructs the position of the repressor element should be identifiable by its ability to inhibit the specificity spiracle element , which cannot be counteracted by STAT binding . Based on the sequence conservation between D . melanogaster and D . virilis ( S4A Fig ) the repressor containing region was divided into four elements: two highly conserved elements ( CE ) , CE1 ( 345–385 bp ) and CE2 ( 492–518 bp ) , intercalated by two non-conserved ( NC ) elements , NC1 ( 305–344 bp ) and NC2 ( 386–491 bp ) . Addition of NC1 and CE1 has no effect on the activity of the enhancer ( S4B–S4D Fig ) . However , when the NC2 element is included , the spiracle expression is downregulated ( S4E Fig ) . Taken together , these results map the position of the repressor element around the NC2 region distal to the STAT binding sites . The identification of a repressor element in the crb518 enhancer prompted us to determine its role in the regulation of crb in the posterior spiracle . The repressor domain does not seem to be involved in tissue-specific regulation as the specificity element ( crb204 , Fig 1G ) is sufficient to direct spiracle activation . In fact , we were unable to observe any ectopic activation in constructs where the repressor element was deleted . Therefore , it seemed plausible that the repressor element could be required for temporal regulation of crumbs expression in the posterior spiracles . To test this possibility , we analyzed the onset of expression directed by the different crb CRM variants ( Fig 4 ) . While crb43 . 2 and crb518 , induce reporter transcription from stage 11 ( Fig 4A'–4A” , 4B’–4B” ) , we observe that crb305 , which lacks the repressor element , induces earlier transcription starting at stage 10 ( Fig 4C–4C” ) . These results demonstrate that the repressor element does not have a major function on tissue-specificity as expression is still detected in the posterior spiracles , but it does affect the temporal regulation of the enhancer with the onset of transcription occurring at earlier stages of development . Our finding that STAT is not involved in the direct activation of the crb spiracle enhancer , made us look into the posterior spiracle gene cascade for alternative activators . To this end , using the 69B-Gal4 line , we tested whether any of the Abd-B primary targets expressed throughout the epidermis could activate ectopic crb305 expression . While ectopic expression of spalt ( sal ) , empty spiracles ( ems ) or cut ( ct ) have no effect , ectopic expression of unpaired resulted in ectopic activation of the spiracle enhancer ( Figs 5E and S5 ) , consistent with the requirement of JAK/STAT for crb expression in the posterior spiracles . Interestingly , despite 69B-Gal4 expressing unpaired throughout the embryonic epidermis , ectopic activation of the enhancer is mostly confined to the posterior segments where Abd-B is also expressed . This result raised the possibility of Abd-B having a direct input on the activation of the crb specificity element . To test this possibility , we determined the ability of Abd-B to bind to the crb CRM in S2 cells using ChIP analysis . We obtained a 9-fold increase in input recovery when performing ChIP with Abd-B confirming that Abd-B is able to bind directly to the crb enhancer ( Fig 6A ) . Analysis of the crb305 sequence shows the existence of eight putative Abd-B binding sites ( Fig 6B , asterisks ) . To determine if Abd-B could bind them directly , EMSAs were performed with oligonucleotides covering regions containing the predicted sites ( Fig 6B ) . As a control , we designed oligonucleotides containing point mutations abolishing Abd-B binding . Abd-B was expressed in S2 cells and cell extracts prepared and used in these assays . As seen in EMSA , all eight sites are bound by Abd-B with different affinities ( Fig 6C ) . Moreover , binding to oligonucleotides containing more than one Abd-B site , is only abolished when both sites are mutant . A similar additive binding has been described for other direct Abd-B targets [16–18] . Our results suggest that Abd-B may be required both directly and indirectly for the activation of crb in the posterior spiracles . Directly , through Abd-B’s binding to the crb specificity element and indirectly through its activation of upd transcription and thus JAK/STAT signalling . Similarly , STAT is also required directly and indirectly for crb’s spiracle expression: directly , to block the repressor’s element activity; and indirectly , because JAK/STAT signalling is also required for the activation of the crb specificity element ( Fig 1H ) . To study if Abd-B can induce the activation of the specificity domain in the absence of JAK/STAT signalling and vice versa , we analyzed the effect of ectopically expressing one in embryos mutant for the other . We performed these experiments using the crb305 enhancer that lacks the repressor domain , avoiding the interference of the repressor element . The ectopic crb305 expression in segments A8-A9 caused by ectopic upd is not present in Abd-B mutants ( compare Fig 5E and 5F ) showing that Abd-B provides an additional input in the activation of the crb305 enhancer . Similarly , ectopic Abd-B expression in Df ( 1 ) os1A embryos lacking all three upd ligands is not capable of activating the crb305 enhancer ( compare Fig 5C and 5D ) . Therefore , these results indicate that both STAT and Abd-B inputs are necessary in parallel for spiracle specificity rather than all the Abd-B activity being mediated by its activation of upd and JAK/STAT signalling . The above data show that both STAT and Abd-B are required for crb specificity element expression . As the function of STAT on the specificity element is indirect , it is likely to be mediated through the regulation of another transcription factor expressed in the posterior spiracles . This means that a still unidentified STAT target mediates STAT’s indirect function on the crb spiracle enhancer . Alternatively , Abd-B itself could be regulated by STAT via a feedback mechanism . If a feedback loop mechanism existed in the posterior spiracles , overexpressing Abd-B should result in the activation of endogenous Abd-B . To determine the existence of such feedback loop , the 69B-Gal4 line was used to ectopically activate UAS-Abd-Bm in the ectoderm of embryos containing the BAC-Abd-B-GFP ( Fig 7A ) . In this BAC element GFP-tagged Abd-B is regulated by the endogenous upstream Abd-B cis-regulatory sequences that are sufficient to drive normal expression in the A8 and A9 segments ( Figs 7B and 7C and S6A–S6A” ) . This set-up allows discriminating between the ectopically expressed Abd-Bm protein and any feedback induced Abd-B-GFP protein from the BAC element . Ectopic UAS-Abd-Bm induction results in ectopic expression of Abd-B-GFP , indicating that a feedback loop mechanism maintaining Abd-B expression does exist ( Fig 7D ) . We next tested whether this feedback loop could be mediated by the activation of the JAK/STAT pathway . To determine this , we repeated the experiment in a Df ( 1 ) os1A mutant background and found that most feedback-induced ectopic expression disappeared ( Fig 7E ) while feedback-independent expression in A8-A9 is maintained . We also ectopically expressed upd throughout the ectoderm using the 69B-Gal4 driver line and analyzed Abd-B-GFP expression . As with ectopic expression of Abd-Bm , Upd also results in the ectopic activation of Abd-B-GFP expression ( Fig 7F ) . As these results suggest the existence in Abd-B of an enhancer regulated by JAK/STAT signalling , we analyzed the Stark Lab genome collection of enhancers made from the Abd-B locus [19] . We found that a 2 . 2 kb fragment ( VT42855 ) drives expression in the posterior spiracles and contains a high density of STAT sites ( Fig 7A and 7G and 7I–7K ) . Chromatin immunoprecipitation with STAT-GFP shows an enrichment of input recovery of this fragment suggesting that STAT is able to bind directly to the Abd-B cis regulatory region ( Fig 7L ) . Moreover , VT42855 expression in the posterior spiracles is abolished in Df ( 1 ) os1A mutants ( Fig 7G and 7H ) confirming the existence of a JAK/STAT regulated enhancer in the Abd-B locus . Taken together , these results indicate the existence of a feedback loop mechanism used to maintain spiracle expression of Abd-B via activation of the JAK/STAT pathway .
Two main regulatory structures have been described during organogenesis: genetic cascades and genetic networks . Genetic cascades are characterized by the hierarchical interactions between its components , with one gene setting the stage and the others responding to it . In this system , the type of organ to develop is chosen by the upstream selector gene , which controls the different cell behaviours via intermediate transcription factors [8 , 20–22] . Gene networks differ by the absence of a single master gene . In a gene network , like that controlling eye development , not only several genes can trigger organogenesis but also all of these genes are necessary for organogenesis . In the absence of anyone of them , organogenesis cannot proceed , even if forced expression of the other master regulators is induced . [23–25] . This behaviour suggests that none of the regulators is exclusively upstream of the others and that activation of any of the master genes results in the activation of the other master genes through cross-regulatory loops [9] . Cross-regulatory feedback loops confer robustness to a regulatory system and their absence in early Hox induced organogenesis could be due to the fast development of the Drosophila embryo that precludes Hox genes to become integrated by feedback loops into the cascades they activate . The formation of the posterior spiracle is a typical example of organogenesis induced by a Hox gene cascade [11] . Here , Abd-B activates the primary targets that , in turn , activate secondary targets like Cad96C , Cad88C , Gef64C and Cad74A . The hierarchical organization of these targets was demonstrated by the observation that in Abd-B mutants the simultaneous activation of the four Abd-B primary targets can activate the expression of Cad96C , Cad88C , Gef64C and Cad74A [11] . Despite of this , here we have found evidence that during embryonic development , the Abd-B spiracle cascade is already transforming into a gene network . We have observed that Abd-B activates the JAK/STAT signalling pathway in the posterior spiracles through upd and that the JAK/STAT pathway feedbacks to enhance Abd-B expression . Two observations show that this is important to maintain robust crb expression: first , JAK/STAT pathway signalling only weakly activates the expression of the crb posterior spiracle enhancer in segments that do not express Abd-B; and second , JAK/STAT pathway activation is insufficient to activate the crb enhancer in the A8 segment of Abd-B mutant embryos ( Fig 5 ) . An indirect feedback loop affecting the Ubx Hox gene has been reported [26] . In the visceral mesoderm Ubx activates in PS7 the signalling molecule dpp . Dpp signals to the neighbouring PS8 cells to activate with Abd-A wingless expression . In turn , Wg signals from PS8 back to PS7 where it is required for the maintenance of Ubx expression . This case and the JAK/STAT—Abd-B interaction we describe here may represent the earliest examples of indirect autoregulatory loops transforming a Hox-cascade into a Hox-network . It would be interesting to analyze other organogenetic cascades to find out if the establishment of feedback loops is a common theme during development and at what stage they become active . During organogenesis the complex patterning information set by the transcription factors and signalling molecules upstream in the cascade has to be integrated to activate the realizator molecules modulating the organogenetic cell behaviours [27] . This is mediated through the cis regulatory modules of each realizator gene . In this study , we have focused on how the posterior spiracle gene cascade activates the crb polarity gene through a CRM that responds to JAK/STAT signalling . The observation that the spiracle enhancer requires the activity of upd , an Abd-B primary target , suggested that Abd-B activates crb indirectly , by setting STAT activity [11] . Our dissection of the crb-spiracle CRM has revealed that a spiracle specificity element can be separated from the STAT binding sites and that the “minimal” CRM is modular in nature . The crb-spiracle CRM studied here is composed by , at least , three independently acting elements: an activator element where the spiracle specificity resides; a STAT binding element that is only required to counteract a neighbouring repressor element; and a repressor element that interferes with the activity of the specificity module . We have found that the spiracle specificity element is bound by Abd-B suggesting that crb is a direct target of Abd-B . However , crb is different from other Abd-B primary targets , since the crb specificity element also requires for its expression STAT activation , itself a target of Abd-B . Because of its mixed character as a direct and an indirect target , we term crb a delayed-primary target: primary to reflect its direct regulation by Abd-B , and delayed because it can only respond to the presence of Abd-B after other Abd-B primary targets are active . The requirement of both Abd-B and JAK/STAT signalling for crb activation explains why the crb-spiracle enhancer is only active in A8 and not in more anterior segments where JAK/STAT signalling is also active , as well as why crb-spiracle enhancer expression in A8 is restricted dorsally despite Abd-B being expressed throughout this segment . The repressor element we have identified is “transplantable” and can function autonomously on an independent posterior spiracle enhancer . In fact , we have shown that the ems spiracle enhancer , that is STAT independent ( Fig 3C ) , can be engineered to become STAT dependent by the inclusion of the STAT binding and repressor modules found in crb ( Fig 3E and 3F ) . The observation that the specificity element in crb is sufficient to drive the activity of the enhancer , begs the question of why is there any need for the presence of further modules in the enhancer . As the specificity element cannot function until the repressor activity is counteracted by Abd-B induction of JAK/STAT signalling in the spiracle , the repressor module provides the CRM with a subtle system to delay the crb spiracle expression during development . The integration of the results in this work uncovers the existence of complex Abd-B and JAK/STAT dynamic interactions that are described by the model shown in Fig 7M . Early Abd-B binding to the crb specificity module cannot activate transcription due to the presence of the neighbouring repressor module . This repression is relieved when Abd-B activation of upd in the posterior spiracles activates STAT . Direct binding of STAT to the crb CRM relieves the repression . We hypothesize that STAT induces the function of another activator protein ( A ) that also binds to the specificity module ( Fig 7M ) . The existence of such activator driven downstream of STAT is necessary to explain why ubiquitous ectodermal Abd-B expression with the 69B-Gal4 line is unable to activate the crb305 enhancer in upd mutant embryos despite the absence of the repressor module ( Fig 5D ) . Besides these functions , our data show that STAT reinforces Abd-B expression in the spiracles through a feed back loop . The model in Fig 7M can explain why the inactivation of JAK/STAT signalling in Df ( 1 ) os1A embryos has a stronger effect on crb518 enhancer expression than the deletion of the STAT binding sites ( compare S2E and S2F Fig and similar observations using crb43 . 2 in Lovegrove et al . [11] ) . Mutation of the two STAT binding sites decreases crb expression because the repressor module is free to interfere with the specificity module . Mutation of Df ( 1 ) os1A has a larger influence on expression because it affects crb regulation at three levels: First , because , as in the previous case , the repressor module is free to interfere with the specificity module in the absence of active STAT; second , because the proposed Activator downstream of JAK/STAT will not be activated; and third , because the STAT feedback loop over Abd-B will not be activated . Although we still do not know all the players involved in the regulation of the crb posterior spiracle enhancer , this case provides an example of how gene networks and CRMs dynamically interact during development . The complex CRM controlling crb spatial and temporal regulation in the posterior spiracles provides a plastic regulatory platform that could be subtly modulated during evolution . Earlier crb activation could be obtained by loss of the repressor module and lower or delayed expression by removal of Abd-B or STAT binding sites . Such quantitative and temporal regulation changes in either direction could be rapidly established by selection of minor mutations in these elements . STAT proteins in vertebrates and invertebrates are considered to be a family of latent transcription factors activated by phosphorylation . After their activation , the cytoplasmic proteins dimerize and accumulate in the nucleus where they bind DNA through TTC ( 3-4N ) GAA sites to activate transcription [28 , 29] . STAT specific transcriptional activation has been suggested to require either cooperative interaction with other enhancer binding proteins , or interaction with co-activator proteins that stabilize or recruit RNA polymerase II complexes or destabilize chromatin [30 , 31] . In this view , transcription of specific downstream targets is due to the direct interaction of STAT with other activators expressed in the same cells . Initially , we expected the STAT dependent up-regulation of crb in the spiracles to be controlled using such a mechanism , however , we have found this not to be the case . Our experiments show that direct STAT binding to the posterior spiracle’s specificity element is not required for transcription . This result suggests that STAT does not function as a classical transcription factor stabilizing RNA polymerase II in the posterior spiracles , but rather acts as a counter-repressor protein . STAT function as a counter-repressor is to prevent that proteins binding to the repressor element interfere with the binding or function of transcriptional activators in the specificity module . Thus , we propose that besides its function as a transcriptional activator , STAT can control gene specific expression through a novel function: counter-repression . Although future work should identify the STAT92E domain controlling counter-repression , we can predict that if the counter-repression domain were different from the transactivation domain , deletion of the transactivation element would still result in a STAT92E protein capable of controlling transcription through counter-repression . The existence of an independent domain with such new function could explain published results showing that STAT proteins with a deleted transactivation domain are still active proteins and do not act as dominant negative transcription factors [32] . Counter-repression is not related to the non-canonical JAK/STAT control of heterochromatin stability described by Li and collaborators as this does not require direct STAT binding to its consensus DNA sites [33–35] , while we show that STAT counter-repression requires the presence of its binding sites . In summary , we have shown that the Hox protein Abd-B initiates a simple embryonic posterior spiracle cascade that soon evolves into a gene network . In this network , Abd-B and the JAK/STAT signalling pathway ( which is activated downstream of Abd-B in the spiracles ) interact to precisely control the temporal activation of the crb realizator gene . Importantly , our study uncovered a novel way for STAT proteins to control cell specific gene expression: permissive counter-repression .
The following Gal4 driver and UAS lines were used: 69B-Gal4 to drive expression in the whole ectoderm , UAS-Abd-Bm , UAS-upd , UAS-ct , UAS-sal , UAS-ems , UAS-grn . We used the Abd-BM1 null allele , the Df ( 1 ) os1A [deleting all three Upd ligands , [36]] . As reporter spiracle lines we used the crb43 . 2 . 1-lacZ [11] and ems0 . 35-lacZ [16] , while the 10xSTAT-dGFP reporter was used to detect global JAK/STAT activation [37] . The VT42855-Gal4 line was obtained from the Stark lab fly transcriptional enhancer collection http://enhancers . starklab . org/ . The [CH321-91P18] BAC-Abd-B-GFP stock is a gift from Rebecca Spokony , modENCODE consortium . To test if the BAC can rescue Abd-B function , we combined BAC-Abd-B-GFP with a UbxMX12 , abd-AM1 , Abd-BM8 triple mutant [38] ( S6 Fig ) . The embryos were grown at 25°C except the 69B-Gal4 UAS overexpression experiments that were grown at 29°C to increase Gal4 activity . The following primary antibodies were used: mouse anti-AbdB 1A2E , anti-Crb and anti-Ct 2B10 ( Developmental Studies Hybridoma Bank ) ; rabbit anti-GFP ( Invitrogen ) and mouse anti-βGal ( Promega ) . Secondary antibodies were conjugated to Alexa Fluor 488 , 555 ( Invitrogen ) . Confocal images were obtained with a Leica SP2-AOBS microscope and processed using ImageJ and Adobe Photoshop CS5 . We used antisense RNA probes for upd , Gal4 and lacZ in situ . The design of the constructs was based on the sequence conservation analysis of crb43 . 2 by comparison of the twelve Drosophila species and A . gambiae , A . mellifera and T . casteneum genomes using the UCSC Genome Bioinformatics ( http://genome . ucsc . edu/ ) . All crb fragments were subcloned into the pCasper-hs43-lacZ transformation plasmid . To this end , pCasper-hs43-lacZ was digested first with EcoRI , the 5’overhang filled with Klenow and finally digested with EclXI . Unless stated otherwise , all PCR reactions were performed using as DNA template the 43 . 2-pBluescript-SK plasmid containing the crb43 . 2 genomic region [11] . The primers used to generate the different constructs are listed in S1 Table . To generate the crb518-Z construct , the 43 . 2-pBluescript-SK plasmid was digested with DraII , the resulting 5’overhangs filled with Klenow and subsequently digested with EclXI . A 518 bp fragment was isolated , purified and subcloned into pCasper-hs43-lacZ . To construct crb518-STAT-mt-Z , mutation of the STAT binding sites was carried out in a two-step process . The first STAT site was mutated by site-directed mutagenesis as described in [39] using the stat4n1-mt primer . The resulting plasmid with the mutated first STAT binding site was then used as a template to perform the mutation of the second STAT site by PCR mutagenesis . A first PCR reaction was performed using the primer pair crb-stat2-mt and rev-crb . The product was purified and used as a reverse primer in a second PCR reaction using fw-crb as forward primer . The resulting PCR product containing both mutated STAT binding sites was digested with EclXI and subcloned into pCasper-hs43-lacZ . The crb305-Z , crb313-Z , crb101-Z , crb204-Z , crb305-STAT-mt-Z , crbΔ25-Z , crbΔ133-Z and crbΔ174-Z constructs were generated as follows: the region of interest was amplified by PCR using either 43 . 2-pBluescript-SK ( for crb305-Z , crb313-Z , crb101-Z and crb204-Z ) or crb518-STAT-mt ( for crb305-STAT-mt-Z , crbΔ25-Z , crbΔ133-Z and crbΔ174-Z ) as DNA template; the PCR products were digested wit EclXI and subcloned into pCasper-hs43-lacZ . The primer pairs used are as follow: fw-crb and rev-stat ( crb305-Z and crb305-STAT-mt ) , fw-stat and rev-crb ( crb313-Z ) , fw-stat and rev-stat ( crb101-Z ) , fw-crb and rev-crb204 ( crb204-Z ) , fw-crb and crbΔ25-rev ( crbΔ25-Z ) , fw-crb and crbΔ133-rev ( crbΔ133-Z ) and fw-crb and crbΔ174-rev ( crbΔ174-Z ) . To generate crb518Δ101-Z , a first PCR was performed using the primer pair fw-Δ101 and rev-crb . The resulting product was purified and used as a primer in a second PCR with the fw-crb primer . The resulting PCR product was digested with EclXI and subcloned into pCasper-hs43-lacZ . To generate the crb313-ems0 . 35-Z constructs PCR reactions were performed using the reverse primer rev-crb with fw-crb205-BHI . The PCR products were purified and digested with EclXI and cloned into pCasper-ems 0 . 35 [16] digested with EclXI and BamHI . All constructs were randomly inserted using P-element transformation . We analysed between three and ten independent lines to discard insertion-site position-effects . Injection was performed by the Drosophila Consolider-Ingenio ( CBM-SO , Madrid ) platform or by BestGene . This protocol , based on [40] , was done as previously described [14 , 16] . ChIP was performed using transiently transfected S2 cells . 1 x 107 cells were seeded in 10 cm cell culture dishes one day before transfection . For Abd-B ChIP , cells were transfected with either ( 1 ) 5 μg pUAST-Abd-Bm-HA and 5 μg pAC-GAL4 or ( 2 ) 5 μg of empty pUAST and 5 μg pAC-GAL4 plasmids . For STAT92E-GFP ChIP , cells were transfected with either ( 1 ) 3 . 5 μg pUAST-STAT92E-GFP , 3 . 5 μg pAC-HopTum-l and 3 . 5 μg pAC-GAL4 plasmids or ( 2 ) 3 . 5 μg empty pUAST , 3 . 5 μg pAC-HopTum-l and 3 . 5 μg pAC-GAL4 plasmids . 1/10 of cells were collected to monitor the protein expression by Western blot . Remaining cells were cross-linked , lysed and sheared to 350–1000 bp . Immunoprecipitations were performed using 6 μl of anti-HA antibody ( Abcam ) or 50 μl of anti-GFP conjugated magnetic beads ( MBL ) per 100 μg sheared chromatin . qRT-PCR was performed with primers crbQPCR1for ( 5’-TTCATTCATTTCCATGAACACA-3’ ) and crbQPCR1rev ( 5’-ATTCGTCGGTTTTCCTTGTC-3’ ) amplifying inside the crb43 . 2 enhancer sequence; AbdB Bac1for ( 5’-TTGGACAAATTCACATGCAA-3’ ) and AbdB Bac1rev ( 5’-GGCCAATGAACTTCCCTCTA-3’ ) amplifying inside the VT42855 enhancer sequence; and , as a control , with AbdB BacC-Afor ( 5’-TGAACTTAAATGCCGAATCAA-3’ ) and AbdB BacC-Arev ( 5’-CACAAGAAGTGCGTGACTGA-3’ ) amplifying in a sequence lacking STAT binding sites . The data are represented as recovered percentage from the input in HA-Abd-B-transfected cells or GFP-transfected cells against pUASt-empty-transfected cells . The complementary oligonucleotides ( Sigma Aldrich ) used to generate the radiolabelled probes in EMSAs to determine STAT92E and Abd-B binding are listed in S2 and S3 Tables respectively . Radioactively labelled probes were generated by annealing and subsequent end filling with [α-32P]dCTP . The conditions used were similar to those described previously [16] . Briefly , double-stranded , end-labelled DNA ( 50 , 000 cpm/binding reaction; 10 nM ) was incubated for 30 min at 4°C with 2 μl of cell extract lysate expressing each tested protein or 2 μl of the cell extract control and 50 mM NaCl , 5 mM EDTA , 0 , 5 mM DTT , 10 mM Tris-HCl ( pH 7 . 8 ) , 4% glycerol , 1 mM MgCl2 , and 1 mg of poly dI-dC as nonspecific competitors , in a final reaction volume of 20 μl . The reactions were run on a 5% polyacrylamide gel , in 0 . 5x Tris-borate-EDTA buffer to visualize complex formation by retardation of the 32P-labeled target DNA . In some experiments monoclonal anti-Abd-B or anti-GFP were incubated with aliquots of the reaction mixture for an additional 30 min . For each gel shift reaction , a control with cell extract of non-trans00FEcted cells was used to detect possible DNA binding by endogenous lysate factors . Gels were dried at 80°C in vacuum , exposed to a phosphorimager screen and detected by a typhoon scanner . | Organogenesis is controlled by gene networks activated by upstream selector genes . To address how the network organization changes during development and how the target genes integrate the genetic information it provides , we analyze in Drosophila the induction of posterior spiracle organogenesis by the Hox gene Abdominal-B ( Abd-B ) . Initially , Abd-B activates in the spiracle primordium a cascade of transcription factors and signalling molecules including the JAK/STAT pathway . We find that at later stages STAT activity feeds back into Abd-B , initiating the transformation of the Hox cascade into a gene-network . Focusing on a spiracle downstream target gene of Abd-B , we analyze how its cis regulatory elements integrate the dynamic network information set by Abd-B and the JAK/STAT signalling pathway during development . Our results also show that the well known transcription factor STAT can control gene expression as a “counter-repressor” , uncovering an alternative novel mode for STAT directed transcriptional regulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | JAK/STAT and Hox Dynamic Interactions in an Organogenetic Gene Cascade |
Dengue virus ( DENV ) and Zika virus ( ZIKV ) are members of the Flaviviridae and are predominantly transmitted via mosquito bites . Both viruses are responsible for a growing number of infections in tropical and subtropical regions . DENV infection can cause lethargy with severe morbidity and dengue shock syndrome leading to death in some cases . ZIKV is now linked with Guillain-Barré syndrome and fetal malformations including microcephaly and developmental disorders ( congenital Zika syndrome ) . The protective and pathogenic roles played by the immune response in these infections is unknown . Mucosal-associated invariant T ( MAIT ) cells are a population of innate T cells with potent anti-bacterial activity . MAIT cells have also been postulated to play a role in the immune response to viral infections . In this study , we evaluated MAIT cell frequency , phenotype , and function in samples from subjects with acute and convalescent DENV infection . We found that in acute DENV infection , MAIT cells had elevated co-expression of the activation markers CD38 and HLA-DR and had a poor IFNγ response following bacterial stimulation . Furthermore , we found that MAIT cells can produce IFNγ in response to in vitro infection with ZIKV . This MAIT cell response was independent of MR1 , but dependent on IL-12 and IL-18 . Our results suggest that MAIT cells may play an important role in the immune response to Flavivirus infections .
Dengue virus ( DENV ) and Zika virus ( ZIKV ) are members of Flaviviridae and both are transmitted mostly via mosquito bites . It is estimated that around 400 million people are infected with DENV annually[1] . DENV infection symptoms range from mild disease , to dengue fever , dengue hemorrhagic fever , and dengue shock syndromes , which can be fatal in some cases . The mechanisms by which DENV infection causes severe illness are not completely understood . An extensive immune activation , characterized by a cytokine storm , has been described in DENV infection , and host factors are also likely to be involved[2] . Conventional antiviral CD8+ T cells are activated and expanded following DENV infection[3] , and have been proposed to be protective by reducing the viral load[4] . Until recently , ZIKV had been understudied because the infection was thought to be associated only with a mild viral illness and of limited geographical distribution . In 2014 , the virus suddenly expanded its range dramatically and appeared in the Americas , leading to the most widespread ZIKV outbreak in history . It is now estimated that over 2 billion people are living in regions suitable for ZIKV transmission[5] . ZIKV infection is now linked with cases of Guillain-Barré syndrome[6] and with a plethora of fetal malformations including microcephaly , now called congenital Zika syndrome , following transmission from an infected pregnant woman to her developing fetus[7] . The protective or pathogenic roles of T cells in ZIKV infection remains to be investigated . Mucosal-associated invariant T ( MAIT ) cells are a population of innate T cells that represent 1–10% of T cells in the blood of healthy individuals[8] . They express a semi-invariant TCR using Vα7 . 2 coupled with Jα33 and a limited Vβ repertoire[9] . A small fraction of MAIT cells have been found to express Vα12 or Vα20[10] . Recent studies suggest that the TCR β-chain has some influence on TCR dependent activation of MAIT cells[11 , 12] . MAIT cells can be identified by the expression of Vα7 . 2 in combination with CD161 or the IL-18 receptor[13] . They have been shown to recognize microbial vitamin B2 ( riboflavin ) metabolites presented by the MHC class I-like protein MR1[14] . This allows MAIT cells to respond to a range of bacteria , mycobacteria , and yeasts[15] . MAIT cells can also be activated in a TCR independent way by IL-12 and IL18[16] , allowing them to respond to pathogens not producing riboflavin , such as viruses[17 , 18] . In chronic HIV-1 and HTLV-1 infections , MAIT cells are reduced in number and display impaired functionality in response to bacterial stimulation[19–21] . A similar MAIT cell impairement has been described in patients with chronic infections due to a primary immunodeficiency[22] . In this study , we investigated MAIT cells response in Flavivirus infection . We report that MAIT cells are activated in acute DENV infection and have a poor response to in vitro bacterial stimulation . We also report that MAIT cells can produce IFNγ in response to in vitro ZIKV infection . This response was dependent on IL-12 and IL-18 and was impaired in HIV-1-infected individuals .
25 DENV-infected individuals from Sao Paulo , Brazil , were enrolled in the study ( 9 males and 16 females , age 17 to 87 , Table 1 ) . Patients were diagnosed with DENV infection by detection of DENV NS1 antigen and/or IgM-specific antibodies using a commercially available rapid test ( Dengue Duo Test Bioeasy , Standard Diagnostic Inc . 575–34 , Korea ) or by detection of DENV RNA by real time PCR ( RT-PCR ) . Absolute cell counts were determined using an automated hematology analyzer ( Abbott Cell-Dyn 3700 Hematology Analyzer ) at the Hematology Laboratory , Hematology Service , at the Faculty of Medicine , University of Sao Paulo . The study was approved by the University of Sao Paulo institutional review board ( CAPPesq ) , and written informed consent was provided by all participants according to the Declaration of Helsinki . Buffy coats from healthy donors were obtained from the New York Blood Bank as approved by the George Washington University institutional review board . Samples from HIV-1-infected patients were obtained from the Jacobi Medical Center ( NY , USA ) and written informed consent was provided by all participants . This study was approved by Jacobi Medical Center and the George Washington University institutional review boards . All samples from all sites were anonymized . Minors were enrolled in the study , in which case legal guardians provided written informed consent according to the Declaration of Helsinki . Peripheral blood mononuclear cells ( PBMCs ) were isolated by density-gradient sedimentation using Ficoll-Paque ( Lymphoprep , Nycomed Pharma , Oslo , Norway ) . Isolated PBMCs were washed twice in Hank’s balanced salt solution ( Gibco , Grand Island , NY ) , and cryopreserved in RPMI 1640 ( Gibco ) , supplemented with 20% heat inactivated fetal bovine serum ( FBS; Hyclone Laboratories , Logan UT ) , 50 U/ml of penicillin ( Gibco ) , 50 μg/ml of streptomycin ( Gibco ) , 10 mM glutamine ( Gibco ) and 7 . 5% dimethylsulphoxide ( DMSO; Sigma , St Louis , MO ) . Cryopreserved cells from all subjects were stored in liquid nitrogen until used in the assays . For DENV-infected patients , samples were collected during the acute phase of infection ( before defervescence ) and 1 month after ( convalescent phase ) . Plasma was collected by centrifugation and stored at -80°C until used in the assays . Vero cells were obtained from the American Type Culture Collection ( ATCC , Manassas , VA , USA ) and maintained using Eagle’s Minimum Essential Medium ( ATCC ) supplemented with 10% fetal bovine serum ( FBS ) at 37°C with 5% CO2 . ZIKV MR766 ( ATCC ) was added to Vero cells at a MOI of 0 . 1 and incubated for 4–6 days . The supernatant was centrifuged at 12 000g for 5 min , filtered ( 0 . 44 μm ) , aliquoted and stored at –80°C . The viral titer was determined using plaque assays on Vero cells as previously described[23] . Briefly , virus stocks were serially diluted and adsorbed to confluent monolayers . After 1 h , the inoculum was removed and cells were overlaid with semisolid medium containing 1% carboxymethyl cellulose ( Sigma Aldrich , St-Louis , MO , USA ) . Cells were further incubated for 5 days , fixed in 4% formaldehyde ( Sigma Aldrich ) , and stained with 1% crystal violet in 20% ethanol ( Sigma Aldrich ) for plaque visualization . Titers were expressed as plaque forming units ( PFU ) per milliliter . In some experiments , ZIKV was heat inactivated by a 60 minutes incubation at 56°C . Cryopreserved specimens were thawed and washed , and counts and viability were assessed using the Countess Automated Cell Counter system ( Invitrogen , Carlsbad , CA ) . Cells were washed and stained in Brilliant Violet Stain Buffer ( BD Biosciences , San Jose , CA ) at room temperature for 15 min in 96-well V-bottom plates in the dark . Samples were then washed and fixed using Cytofix/Cytoperm ( BD Biosciences ) before flow cytometry data acquisition . Intracellular staining was performed in Perm/Wash ( BD Biosciences ) . mAbs used in flow cytometry: CD3 AF700 , CD3 PerCP-Cy5 . 5 ( both clone UCHT1 ) , CD8 BV711 ( clone RPA-T8 ) , CD38 APC-H7 ( clone HB7 ) , CD127 FITC ( clone HIL-7R-M2 ) , CD161 BV421 ( clone DX12 ) , CCR6 BV786 ( clone 11A9 ) , HLA-DR APC ( clone L243 ) , IFNγ APC ( clone B27 ) , and PD-1 PE-Cy7 ( clone EH12 . 1 ) were all from BD Biosciences , PLZF APC was from R&D Systems ( Minneapolis , MN ) , EOMES FITC ( clone WD1928 ) was from eBioscience and TCR Vα7 . 2 PE ( clone 3C10 ) was from Biolegend ( San Diego , CA , USA ) . Live/dead aqua fixable cell stain was from Life Technologies ( Eugene , OR , USA ) . Data were acquired on a BD LSRFortessa instrument ( BD Biosciences ) and analyzed using FlowJo Version 9 . 8 . 5 software ( TreeStar , Ashland , OR , USA ) . MAIT cell function was determined in vitro using paraformaldehyde-fixed E . coli stimulation ( one shot top10 , Life Technology , multiplicity of exposure 10 ) in the presence of 1 . 25 μg/ml anti-CD28 mAb ( clone L293 , BD Biosciences ) [24] or ZIKV at a MOI of 5 ( without anti CD28 mAb ) . E . coli was fixed for 5 minutes in 1% paraformaldehyde . PBMCs were further cultured for 24 hours at 37°C/5% CO2 in RPMI medium supplemented with 10% fetal bovin serum . Monensin ( Golgi Stop , BD Biosciences ) was added during the last 6 hours of the stimulation . In some experiments blocking antibodies against MR-1 ( 5μg/ml , clone 26 . 5 , Biolegend ) , IL-12p70 ( 10μg/ml , clone 24910 , R&D systems ) , and IL-18 ( 10μg/ml , clone 125-2H , MBL International , Woburn , MA , USA ) were added . IL-7 ( RayBiotech , Norcross , GA , USA ) and sCD14 ( R&D Systems ) were measured in plasma by ELISA following manufacturer’s instruction . All statistical analysis was performed using Graph Pad Prism version 6 . 0h for Mac OSX ( GraphPad Software , La Jolla , CA ) . The changes between acute and convalescent phases and before/after ZIKV stimulation with or without blocking antibodies were analyzed with Wilcoxon matched-pairs signed rank test . Associations between groups were determined by Spearman's rank correlation . P-values ≤ 0 . 05 were considered statistically significant .
We enrolled 25 individuals with acute DENV infection , and we followed them during the convalescent phase ( Table 1 ) . We evaluated MAIT cell ( defined as CD3+ CD161+ Vα7 . 2+ , Fig 1A ) frequency by flow cytometry and found no significant difference between acute and convalescent DENV infection ( Fig 1B ) . However , MAIT cell counts were decreased in the acute phase ( Fig 1C ) due to significant overall lymphopenia amongst infected patients ( S1 Fig ) . Next , we characterized the phenotype of MAIT cells in the acute and convalescent phases of DENV infection . MAIT cells had significantly increased co-expression of the activation markers CD38 and HLA-DR ( S1 Fig and Fig 1D ) , of the IL-7 receptor CD127 ( S1 Fig and Fig 1E ) , and of PD-1 ( S1 Fig and Fig 1F ) in the acute phase . We did not observe any difference in the expression of CCR6 by MAIT between the acute and convalescent phases ( Fig 1G ) . In chronic viral infections MAIT cell activation is associated with their reduced frequency[19 , 21] . Thus , we investigated if there was an association between the reduced MAIT cell count in the acute phase and their increased co-expression of CD38 and HLA-DR , and found a trend for an inverse association ( p = 0 . 0779 , S1 Fig ) . Next , we compared the results for MAIT cells during the convalescent phase to healthy controls from Brazil . We found that there was no difference in the co-expression of CD38 and HLA-DR between the convalescent and healthy controls individuals ( S2 Fig ) . PD-1 remained elevated during the convalescent phase of DENV infection ( S2 Fig ) and CD127 was decreased compared to healthy controls ( S2 Fig ) . Our results show that MAIT cells are activated and reduced in number in acute DENV infection . Because the majority of MAIT cells are CD8+ , we evaluated the response of conventional CD8 T cells in acute DENV infection . Conventional CD8 T cells had significantly elevated levels of co-expression of CD38 and HLA-DR in the acute phase and the levels of co-expression in the convalescent phase were similar to healthy controls ( S3 Fig ) . PD-1 was also elevated on CD8 T cell in the acute phase of infection . However , PD-1 levels in the convalescent phase trend to remain elevated compared to healthy controls ( S3 Fig ) . However , in contrast to MAIT cells , the levels of CD127 on conventional CD8 T cells were not different between the acute and the convalescent phase , or healthy controls ( S3 Fig ) . MAIT cells have been shown to have decreased expression of key transcription factors in chronic viral infections[21 , 25 , 26] . Therefore , we investigated if MAIT cells showed a similar decrease of Eomes and PLZF expression in acute DENV infection . We found that Eomes expression was reduced in convalescent DENV infection ( Fig 2A and 2B ) compared to the acute phase and healthy controls . However , we did not observe any difference in PLZF expression between acute and convalescent DENV infection or healthy controls ( Fig 2A and 2C ) . Our results suggest that different transcription factor expression profiles are associated with acute and chronic viral infections respectively . Increased pro-inflammatory cytokines levels in DENV infection have been associated with microbial translocation[27] . sCD14 is a marker of monocyte activation and is considered an indirect marker of microbial translocation[28] . Thus , we measured the levels of sCD14 in our cohort of DENV-infected subjects . Levels of sCD14 were significantly higher in the acute phase of infection than in the convalescent phase ( Fig 3A ) . Levels of sCD14 remained higher in convalescent DENV compared to healthy controls . However , we did not find any significant associations between the levels of sCD14 in acute DENV infection and co-expression of CD38 and HLA-DR by MAIT cells or with MAIT cell numbers ( S4 Fig ) . Because we found elevated expression of the IL-7 receptor by MAIT cells in acute DENV infection , we measured the levels of plasma IL-7 in acute and convalescent DENV infection but did not find any significant change ( Fig 3B ) . Next , to establish the functionality of MAIT cells , we investigated the in vitro response of MAIT cells from the acute and convalescent phases of DENV infection to in vitro stimulation with E . coli . There was no difference in IFNγ production by MAIT cells in the acute and convalescent phases of DENV infection in the absence of stimulation ( S5 Fig ) . MAIT cells in the acute phase produced significantly less IFNγ after E . coli stimulation compared to the convalescent phase ( Fig 4A and 4B ) . The MAIT cell IFNγ response in the convalescent phase was similar to the response of healthy controls ( S5 Fig ) . Interestingly , we found that the levels of sCD14 in acute DENV infection were inversely associated with the MAIT cell IFNγ response ( Fig 4C ) , possibly suggesting a role for monocyte activation in the poor MAIT cell response . Finally , we used in vitro infection with ZIKV to study the mechanism of MAIT cell activation in a different Flavivirus infection . MAIT cells from healthy individuals consistently produced IFNγ in response to in vitro ZIKV infection ( Fig 5A and 5B ) . In contrast to E . coli , the MAIT cell IFNγ response to ZIKV could not be blocked by a MR-1 blocking antibody ( Fig 5C ) . The IFNγ response from MAIT cells to ZIKV was partially reduced by blocking antibodies against IL-12 and IL-18 and was completely blocked when they were used in combination ( Fig 5C ) . We also investigated if viral replication was needed for the MAIT cell response to in vitro ZIKV infection . For this purpose , heat inactivated ZIKV was added to PBMCs and the MAIT cell IFNγ response was compared to the response obtained using replication competent ZIKV . We observed only a small reduction in IFNγ production by MAIT cells in response to ZIKV when using a heat inactivated virus ( Fig 5C ) , suggesting that viral replication is not needed for production of IL-12 and IL-18 and subsequent MAIT cell activation . Viral co-infections are common and understudied . In Brazil , many people living with HIV will be exposed to dengue or zika viruses . MAIT cells from HIV-1-infected individuals exhibit decreased functionality following stimulation with E . coli[19] . Thus , we evaluated the capacity of MAIT cells from HIV-1-infected subjects ( Table 2 ) to produce IFNγ in response to ZIKV infection and found that in 5 out of 6 individuals there was no increase in IFNγ production in response to ZIKV infection ( Fig 5D ) . We then directly stimulated PBMCs with IL-12 and IL-18 and found a similar IFNγ response by MAIT cells from HIV-1-infected and uninfected subjects ( Fig 5E ) , suggesting that MAIT cells from HIV-1-infected individuals have a normal capacity to respond to cytokine stimulation .
We found that MAIT cells are activated in acute human DENV infection as well as following in vitro ZIKV infection . However , in contrast to a previous study[17] , we did not find any significant changes in MAIT cell frequency between acute and convalescent DENV infection . Wilgenburg et al . focused on the study of CD8+ MAIT cells , while in this study , we also included CD8- MAIT cells . In addition , the difference in timing of sample collection might explain the differences between the two studies . We found that MAIT cell counts were decreased in parallel with the total lymphocyte count during the acute phase . We found that MAIT cells were activated during acute DENV infection , as had Wilgenburg and colleagues . IL-12 and IL-18 have been shown to trigger MAIT cell activation[16] , and monocyte production of IL-18 is required for MAIT cell in vitro response to influenza A virus ( IAV ) [18] . The levels of IL-12 and IL-18 are elevated in DENV infection[17 , 29–31] and could therefore be involved in MAIT cell activation , as shown here . We showed that MAIT cell IFNγ production following in vitro ZIKV infection also depended on IL-12 and IL-18 . Immune activation in acute DENV infection has been associated with elevated levels of LPS and of markers of microbial translocation[27 , 32] . This raises the possibility that MAIT cells could also be activated in a TCR-dependent way by microbial products during acute DENV . However , we did not find an association between the levels of sCD14 , an indirect marker of microbial translocation , and MAIT cell activation in acute DENV . Rather , we found that sCD14 was inversely associated with the in vitro IFNγ response of MAIT cells to E . coli . This suggests that monocyte activation could result in poor antigen presentation to MAIT cells . An alternative explanation could be of a temporary monocyte tolerance to stimulation induced by LPS . This could contribute to the reduced MAIT cell response in the acute phase . Finally , the elevated expression of PD-1 on MAIT cells during acute DENV infection could also contribute to the reduced IFNγ production . Further studies are needed to confirm that both MAIT cells and monocytes are involved in this defect . Chronic viral infections have been associated with a reduced expression of the transcription factors PLZF and Eomes by MAIT cells[21 , 25 , 26] . Interestingly , we found that DENV infection did not change the levels of PLZF expression in MAIT cells and their Eomes levels were reduced in convalescent compared to acute DENV and healthy controls . CD56+ MAIT cells have been shown to have a higher Eomes expression and a more robust response to IL-12 and IL-18 than CD56- MAIT cells [11] . Therefore , it is possible that the decrease in Eomes expression by MAIT cells in convalescent DENV infection is part of a feedback loop to decrease their response to cytokines . Another possibility is a decrease in the CD56+ subset of MAIT cells in blood following acute DENV . We have also observed a decreased expression of the IL-7 receptor ( CD127 ) by MAIT cells during the convalescent phase . IL-7 has been shown to increase MAIT cell response[26 , 33] . Thus , reduced levels of Eomes and CD127 could by a mechanism by which MAIT cells could prevent sustained activation following an acute infection . Patients that recovered from IAV infection had higher circulating MAIT cells than those that succumbed[18] and IFNγ production by MAIT cells has been shown to limit HCV replication in vitro[17] . Thus , there is increasing evidence that MAIT cells could play a protective role in viral infections . DENV and ZIKV infections are associated with a range of clinical symptoms . More studies are needed to investigate if MAIT cell frequency , functionality or activation status have an impact on the clinical outcome of DENV and ZIKV infections . In this regard , MAIT cell production of IFNγ could be part of an innate immune response to induce an anti-viral state and compromise Flavivirus replication . Levels of serum IFNγ have been reported to be inversely associated with DENV load and symptoms[34] . One limitation of our study is that we focused only on peripheral MAIT cells . MAIT cells are present in the skin[35 , 36] and skin resident MAIT cells may play a more important role in early innate defense following mosquito transmission of Flavivirus . MAIT cells from HIV-1-infected individuals have been shown to have a lower production of cytokines in response to E . coli stimulation[19] . In this study , we show that the cytokine mediated MAIT cell response to in vitro viral infection is also impaired . However , MAIT cells from HIV-1-infected subjects had a normal response to direct cytokine stimulation , suggesting that poor IL-12 and IL-18 production in response to ZIKV infection could be responsible for the impaired MAIT cell response in these individuals . This suggests that HIV-1-infected individuals could have a poor innate immune response to ZIKV and be at a higher risk to develop complications following Flavivirus infection . Case reports of HIV-1-infected individuals with ZIKV infection have been reported[37 , 38] , including one case with congenital Zika syndrome[39] . Defective MAIT cell activation could be one factor contributing to the increase incidence of severe dengue in HIV-1-infected subjects[40] . More studies are needed to determine if MAIT cells contribute to protection or to immunopathology during Flavivirus infections . Overall , our results show that MAIT cells are activated in response to DENV and ZIKV infections . This innate response was TCR-independent and defective in HIV-1-infected individuals . Further studies are necessary to determine the importance of MAIT cell responses in the clinical outcomes of Flavivirus infections . | Dengue virus ( DENV ) and Zika virus ( ZIKV ) are responsible for a growing number of infections in tropical and subtropical regions . DENV infection can cause dengue shock syndrome leading to death in some cases , while ZIKV is now linked with Guillain-Barré syndrome and congenital anomalies including microcephaly . The protective and pathogenic roles played by the immune response in these infection is unknown . Mucosal-associated invariant T ( MAIT ) cells are a population of innate T cells with potent anti-bacterial activity . MAIT cells have also been postulated to play a role in the immune response to viral infections . In this study , we found that MAIT cells are activated in acute DENV infection and in vitro following ZIKV infection . MAIT cell IFNγ response to ZIKV infection was TCR independent , but IL-12 and IL-18 dependent . IFNγ produced from MAIT cells could help limit viral replication . Further studies are needed to determine the protective or pathogenic role of MAIT cells in Flavivirus infections . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"blood",
"cells",
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"cloning",
"viruses",
"bacterial",
"diseases",
"rna",
"viruses",
"molecular",
"biology",
"techniques",
"cytotoxic",
"t",
"cells",
"cell",
"enumeration",
"techniques",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"escherichia",
"coli",
"infections",
"medical",
"microbiology",
"microbial",
"pathogens",
"t",
"cells",
"molecular",
"biology",
"cell",
"biology",
"flaviviruses",
"monocytes",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms",
"zika",
"virus"
]
| 2018 | MAIT cells are activated in acute Dengue virus infection and after in vitro Zika virus infection |
Systemic , life-threatening infections in humans are often caused by bacterial or fungal species that normally inhabit a different locale in our body , particularly mucosal surfaces . A hallmark of these opportunistic pathogens , therefore , is their ability to thrive in disparate niches within the host . In this work , we investigate the transcriptional circuitry and gene repertoire that enable the human opportunistic fungal pathogen Candida albicans to proliferate in two different niches . By screening a library of transcription regulator deletion strains in mouse models of intestinal colonization and systemic infection , we identified eight transcription regulators that play roles in at least one of these models . Using genome-wide chromatin immunoprecipitation , we uncovered a network comprising ∼800 target genes and a tightly knit transcriptional regulatory circuit at its core . The network is enriched with genes upregulated in C . albicans cells growing in the host . Our findings indicate that many aspects of commensalism and pathogenicity are intertwined and that the ability of this microorganism to colonize multiple niches relies on a large , integrated circuit .
Mammalian mucosal surfaces harbor trillions of microorganisms from all three domains of life [1]–[4] . While most of these microorganisms are harmless ( or beneficial ) to their host , a few of them are able to cross the host's protective barriers and colonize internal organs that offer little apparent resemblance to the microbe-laden mucosal surfaces . Indeed , many of the life-threatening infections in humans are caused by the very same bacterial or fungal species that typically compose our own microbiota . A hallmark of these opportunistic pathogens therefore is their ability to proliferate in disparate niches within the host . It remains an open question , however , whether the repertoire of genes that enables such pathogens to thrive in the host varies from one niche to the other . In this paper we investigate the case of C . albicans , the most prominent fungal species living on mucosal surfaces—particularly in the gastrointestinal ( GI ) tract—of warm-blooded animals [5]–[7] . While it is a member of the normal human microbiota , C . albicans can also cause mucosal disease in healthy hosts or produce systemic infections and colonize internal organs in people who have received surgical implants or whose immune systems have been compromised , such as AIDS patients or individuals receiving chemotherapy . Deep-seated infections often result in life-threatening conditions . In addition to the status of the host immune system [8] , [9] , the outcome of the C . albicans–host interaction depends on various products and functions encoded in the genome of the fungus , as multiple gene deletions render C . albicans avirulent in both mucosal and invasive animal models ( reviewed in [10] ) . For example , the production of extracellular hydrolases [11] , the ability to switch between yeast and filament forms [12]–[14] , and the production of small molecules [15] are all necessary for C . albicans to proliferate as either commensal or pathogen . The recent generation of relatively large collections of gene deletion mutants makes it now possible to carry out systematic and unbiased searches for genes and cellular functions employed by C . albicans to thrive in the host . To begin to dissect the repertoire of genes that enable C . albicans to colonize the mammalian GI tract and determine whether these genes also play a role during systemic infection , we screened a collection of 77 C . albicans transcription regulator ( TR ) mutants in mouse models that recapitulate these two niches . We focused on TRs because transcriptional circuits are central to the regulation of many biological processes . The subset of TRs we screened was chosen because their deletion in C . albicans produces neither significant growth defects nor anomalous colony morphologies under any of 55 different laboratory growth conditions that have been tested [16] . Thus , we expected to maximize the identification of regulators “dedicated” to biological processes directly connected to C . albicans proliferation in the host . This approach also minimized the retrieval of mutants with either large pleiotropic effects or with fitness defects not specific to life in the host . Here we report the identification of eight C . albicans TRs required for GI tract colonization , systemic infection or both . We elucidate the transcriptional circuitry controlled by these regulators using genome-wide experimental approaches , and find that the resulting network is enriched with genes that are upregulated when C . albicans grows in the host . Five of the identified TRs form a highly interconnected core network that regulates determinants of GI tract colonization as well as systemic infection indicating that both types of growth in the host require common circuitries . We find that cell surface remodeling and the acquisition of carbon and nitrogen are salient among the functions that C . albicans requires to proliferate in the host . Finally , we demonstrate that several of the gene products regulated by the identified TRs are in fact required for intestinal colonization or for systemic infection . Thus , the use of TRs as genetic entry points , combined with full-genome molecular biology methods , can identify regulators , circuits , and target genes needed explicitly for C . albicans to colonize different niches of mammalian hosts .
Sequence-specific DNA binding proteins that regulate transcription , or TRs , are major elements within the gene network of an organism . They are pivotal in orchestrating responses to external cues and in maintaining internal homeostasis in the face of fluctuations in the environment . TRs are thus likely to be critical components of the gene network that underlies the ability of C . albicans to inhabit the host . Our laboratory recently constructed a collection of 165 C . albicans TR homozygous deletion mutant strains consisting of two independently generated , fully vetted isolates of each deletion [16] . About 45% of the C . albicans TR deletion strains display no significant growth or colony morphology phenotype under any of 55 different laboratory growth conditions ( Figure 1A ) [16] , raising the possibility that their function may be revealed only in the context of the host . Hence , we focused on this subset of TRs ( n = 77 ) to carry out genetic screens in mouse models that recapitulate niches where C . albicans thrives ( Figure 1B ) . To minimize the number of animals required to screen the mutant strain library , we adopted the signature-tagged mutagenesis technique that our laboratory has successfully used to identify virulence factors in C . albicans [15] and has been employed in other fungi [17] and bacteria [18] as well . We used a mouse model of intestinal colonization in which immunocompetent antibiotic-treated Swiss Webster mice are orally inoculated with C . albicans by gavage [19] , [20] . While mice do not appear to be natural hosts for C . albicans but for the closely related yeast C . tropicalis [1] , the murine GI colonization model has been adopted as the standard in the field to evaluate C . albicans commensalism [21]–[23] . We assayed pools of 15–20 signature-tagged mutant strains; the relative abundance of the strains recovered from feces ( at 1 , 9 , and 21 d post-inoculation ) or intestinal contents ( at day 21 when mice were euthanized ) compared to the inoculum was determined by real time PCR ( using primers to the signature tags ) as described [15] . We were able to confidently monitor 72 C . albicans strains over the course of the experiment in three mice each . The level of depletion or accumulation of each mutant relative to the inoculum is shown in Figure 2A . We found that ∼1 , 000-fold reduction with respect to the inoculum is the limit of accurate detection for most strains in this assay ( that is , a log2 value of about −10 ) . While the actual values can vary from mouse to mouse , they do show a high degree of consistency across samples ( e . g . , intestinal contents versus fecal pellets at day 21 ) and across time points ( e . g . , mutants that became undetectable at day 9 remained so at day 21 ) . The weight and body condition of all inoculated mice were closely monitored throughout the experiment; no differences were observed between inoculated and control animals . For each mutant that showed a severe defect ( i . e . , those that fell below the level of detection of the assay at day 9 or 21 post-inoculation ) we tested an independently constructed deletion of the same gene . We also re-tested in this assay all the mutants that showed defects in our second screen , a systemic infection model ( see below ) . We focused on mutants with large effects ( >1 , 000-fold reduction relative to the inoculum ) and , to facilitate the statistical analysis , we converted the data to binary mode: presence or absence . The data displayed in this manner are shown in Figure 2B . Each mutant shown here was evaluated in at least six mice , with both independent isolates producing consistent results . Based on these criteria , six TR deletion mutants ( tye7Δ/Δ [orf19 . 4941] , rtg1Δ/Δ [orf19 . 4722] , rtg3Δ/Δ [orf19 . 2315] , lys144Δ/Δ [orf19 . 5380] , hms1Δ/Δ [orf19 . 921] , and orf19 . 3625Δ/Δ ) showed significant , large impairments in GI tract colonization ( p<0 . 02 ) while zcf21Δ/Δ ( orf19 . 4166 ) has a weaker defect ( p = 0 . 0503 ) . To verify that the phenotype is due to the deleted gene , we reintroduced an ectopic copy of the wild-type allele back into each mutant and found that it was able to restore colonization at least partially in all mutants ( Figure 3 ) . As noted previously , the mutants have no growth defect in standard laboratory conditions and none of the six regulators had been previously implicated in intestinal colonization . Of the six regulators , TYE7 is known to control carbohydrate metabolism in C . albicans [24] and contributes to the cohesiveness and correct hyphal formation of biofilms [25] , while HMS1 has recently been reported to be required for C . albicans morphogenesis at elevated temperatures ( 42°C ) [26] . Beyond the initial phenotypic screening describing the TR deletion collection [16] , no function has been ascribed to any of the other TRs in C . albicans , although in Saccharomyces cerevisiae RTG1 and RTG3 are key regulators of the mitochondrial retrograde response ( described below ) [27] , [28] . To determine whether a given C . albicans TR play a role in colonization of the GI tract as well as during systemic infection , we evaluated the fitness of the same set of 77 TR deletion mutant strains in a mouse model of disseminated candidiasis . We chose tail vein injection because this model has been adopted as the standard in the field to assess C . albicans virulence . Pools of 24 signature tagged mutant strains were assayed , and the relative abundance of the strains recovered from the kidneys of moribund BALB/c mice ( 2–4 d post-infection ) ( Figure 1B ) was compared to that in the infecting inoculum using real time PCR [15] . For about two-thirds of the mutant strains , two independent isolates were evaluated . Only one isolate was tested for the other third . Every strain was assayed in at least four mice . The results obtained for all the mutants are shown in Figure 4A and Table S1 . To consider a mutant strain as having fitness defect , both isolates had to exhibit consistent results ( none of the mutants for which only a single isolate was tested showed any defect ) . Based on this criterion , the screen revealed five TR deletion mutants ( rtg1Δ/Δ , rtg3Δ/Δ , zcf21Δ/Δ , lys14Δ/Δ [orf19 . 5548] , and hms1Δ/Δ ) with reduced fitness ( p<0 . 05 ) . These five mutants were retested individually and showed reduced virulence in single ( as opposed to pooled ) tail vein infections when time to illness was monitored ( Figure 4B–4D ) . Careful scrutiny of the other three TR deletion mutants with defects in GI tract colonization ( tye7Δ/Δ , lys144Δ/Δ , and orf19 . 3625Δ/Δ ) confirmed that they did not show abnormalities in our systemic infection model ( in agreement with this observation , [24] also found no defect for a tye7 mutant in related models of disseminated candidiasis ) . In sum , our two in vivo genetic screens uncovered eight TRs playing roles in the proliferation of C . albicans within the host . Three of these regulators ( RTG1 , RTG3 , and HMS1 ) exhibited significant impairment in both GI tract colonization and systemic infection ( Figure 4E ) . ZCF21 displays a significant fitness anomaly in systemic infection but only a weak and variable defect in GI tract colonization . LYS14 showed a defect in systemic infection but not GI tract colonization . ORF19 . 3625 , LYS144 , and TYE7 showed the opposite behavior being required for intestinal colonization but not systemic infection . We excluded ORF19 . 3625 from further study because it encodes a putative subunit of a histone remodeling complex and as such it is unlikely to be a specific regulator for a particular set of genes . To gain insights into the biological processes directed by the seven identified TRs , we determined the genes that they regulate . TYE7 is the only one of the regulators for which genome-wide data regarding its target genes in C . albicans are available ( see [24] ) . Thus , we carried out whole genome chromatin immunoprecipitation followed by array hybridization ( ChIP-chip ) for the remaining six TRs . As it might be predicted , the conditions typically used to grow C . albicans in the laboratory ( liquid culture in YPD medium at 30°C ) were not optimal to detect either binding of the TRs to their target promoters or changes in the expression of target genes ( i . e . , in expression arrays comparing wild-type versus TR deletion mutant strains ) . This was not unexpected because the mutants chosen for the screen have no significant phenotypes when tested under laboratory conditions; therefore , the identified regulators are likely to be active only under specific conditions within the host . To overcome this limitation we constructed fluorescent reporter strains ( yfp or gfp fused to each regulator's native promoter and YFP- or GFP-fused TR proteins ) and sought conditions that promoted either the expression or the nuclear localization ( in the case of fusion proteins ) of the fluorescent reporters . Among the conditions tested were ∼20 different cell culture media , 37°C ( the temperature in the host ) and the growth of cells on a semi-solid surface ( which may mimic growth on the surfaces within the host ) . Figure 5A summarizes the optimal growth conditions that were chosen to immunoprecipitate each regulator in vitro . In the case of ZCF21 , LYS14 , and LYS144 , we nonetheless had to increase their expression artificially using the TDH3 promoter to be able to immunoprecipitate them . Using stringent cutoffs to define statistically significant binding events ( see Materials and Methods ) , we established that the following number of intergenic regions are bound by each regulator: 79 for Hms1 , 51 for Lys14 , 47 for Lys144 , 237 for Zcf21 , and 215 for Rtg1 and Rtg3 ( Dataset S1 ) . The ChIP-chip profiles of Rtg1 and Rtg3 were identical to each other implying that these two proteins bind to DNA together . Indeed , the S . cerevisiae Rtg1 and Rtg3 orthologous proteins are known to form a heterodimer to bind to DNA ( reviewed in [29] ) . Using only the ChIP-chip data , we were able to derive DNA motifs ( i . e . , cis-regulatory sequences ) for each regulator ( Figure 5A ) . These sequences were significantly enriched in the bound regions compared to the remainder of intergenic regions ( Figure S1 ) . The motif that we derived for Rtg1/3 is similar to the reported binding sequence of their orthologs in S . cerevisiae ( GTCAC ) [29] . Likewise , the motif that we find for Lys144 resembles the reported binding sequence for its closest homolog in S . cerevisiae , Lys14 ( TCCRNYGGA ) [30] . The motif that we derived for Hms1 , a member of the basic helix-loop-helix family of TRs , matches the non-E-box consensus binding sequence ( ATCACCCCAC ) for SREBP1 , the prototypical member of the family [31] . Although the Lys14 motif that we generated differs from the sequence recognized by its homolog Lys14 in S . cerevisiae , we confirmed by gel mobility shift assays that the purified C . albicans Lys14 protein binds in vitro to the sequence that we identified ( Figure 5B ) . ( Phylogenetic reconstructions indicate that the closest homolog of S . cerevisiae LYS14 in C . albicans is LYS144 and not LYS14 , albeit the current nomenclature implies otherwise . In addition , as described below , C . albicans LYS144 and LYS14 have nothing to do with lysine biosynthesis regulation . ) We were unable to identify an ortholog in S . cerevisiae for C . albicans Zcf21 , so we could not perform an independent check of its motif . Taken together , the fact that we were able to derive motifs de novo from the ChIP data , and the similarity of these independently derived motifs to the sequences known to be bound by homologs in other species validate the dataset that we generated by genome-wide ChIP . All the binding events by the seven TRs ( including the Tye7 ChIP data from [24] ) translate into 808 putative target genes bound by at least one of the regulators ( binding events in intergenic regions between divergently transcribed genes were counted as two target genes ) ( Dataset S2 ) . The resulting network depicting the relationships among the regulators and all their target genes is shown in Figure 6A . It is apparent from the network's topology that many of the target genes are regulated by more than one TR . Moreover , the network displays no clear distinction between potential subsets of targets controlled specifically by regulators required for intestinal colonization and subsets controlled solely by regulators of systemic infection . In fact , there is no obvious partition among the sets of targets controlled by RTG1/3 , HMS1 , TYE7 , and ZCF21 even though the phenotypes ascribed to them are different: RTG1/3 and HMS1 were identified in both screens , TYE7 only in the GI tract model , and ZCF21 only in the systemic model . This study was designed to identify TRs that specifically control aspects of C . albicans that are needed in the host . A prediction of this idea is that the target genes identified in this study will be preferentially expressed when C . albicans is in the host . To test this prediction , we compared the list of ChIP targets that we identified to an independently generated gene expression dataset where C . albicans growing in the mouse intestine was compared to C . albicans growing under laboratory conditions . In this study [19] , Rosenbach et al . defined a collection of 408 genes that were upregulated during growth in the murine cecum relative to laboratory grown exponential and post-exponential phase cells ( in reference [19]'s table S3 ) . We found that C . albicans genes upregulated during growth in the mouse intestine are significantly overrepresented in the set of putative ChIP targets ( 153 out of 408 genes , p = 2 . 7×10−38 ) ( Figure 6B; Dataset S2 ) supporting a role for the identified regulators in controlling a gene expression program activated specifically in the host . The subset of 153 target genes upregulated when C . albicans is growing in the murine gut is not evenly distributed across the network ( Figure 6A ) . Rather they are predominantly located in the set controlled by Rtg1/3 ( 108 of 153 genes ) ( Dataset S2 ) suggesting that these two proteins are major regulators of GI tract colonization determinants . RTG1/3 controls mitochondrial retrograde signaling in S . cerevisiae ( reviewed in [29] ) . This pathway involves sensing and transmitting nutritional as well as mitochondrial signals to effect changes in nuclear gene expression; these changes lead to a reconfiguration of metabolism to accommodate cells to nutrient availability or to mitochondrial defects [29] . Based on our ChIP-chip results , Rtg1/3 appears to regulate similar functions in C . albicans and these functions seem to contribute to the ability of the fungus to proliferate in the GI tract . In support of this idea , the subset of Rtg1/3 targets upregulated in C . albicans cells growing in the intestine ( 108 genes ) is enriched with genes involved in metabolic functions ( e . g . , carbohydrate catabolic process [p = 3 . 28×10−5] ) . While there is a diverse set of biological functions and processes represented in the target genes in the identified network ( Figure 6A ) , two groups of membrane proteins are salient among the targets bound by the TRs required for intestinal colonization: First , about a third of the C . albicans genes annotated as encoding amino acid permeases ( GNP1 , HIP1 , CAN2 , AGP2 , GAP2 , and GAP6 ) are bound by Rtg1/3 . Moreover , Rtg1/3 and Hms1 bind upstream of STP2 , a gene encoding a major regulator of transcription of amino acid permeases in C . albicans [32] . And second , Lys144 binds upstream of each of four putative allantoate transporters ( DAL5 , DAL7 , DAL8 , and DAL9 ) and of ORF19 . 2065 , a gene whose ortholog in S . cerevisiae ( DAL2 ) encodes an enzyme involved in allantoate catabolism [33] . That these TRs may exert control on the acquisition of amino acids as well as of allantoate , a product of purine metabolism in some species , suggest that , in the gut , C . albicans adjusts its metabolic response to procure nitrogen from these molecules . We next wanted to test experimentally whether the target genes identified in this study were actually required for C . albicans to colonize the GI tract or for fitness during systemic infection . We reasoned that the most likely candidates to show strong effects would be those genes that are clearly bound by one or more of the TRs identified here and whose expression is upregulated when C . albicans is growing in the mouse compared to laboratory conditions . Of the 153 genes upregulated in the host ( Figure 6 ) , we focused on those bound by Hms1 and Rtg1/3 because these TRs showed phenotypes in both mouse models . We selected 18 genes that met these criteria ( Table S2 ) and successfully constructed signature-tagged homozygous deletion strains for 17 of these genes ( we were unable to make a homozygous deletion of orf19 . 1363 , raising the possibility that this gene may be essential ) and tested 15 of them in the mouse models of GI tract colonization and systemic infection ( orf19 . 1069 and orf19 . 4961 were excluded because their deletion results in severe growth defects in vitro ) . As described for the initial TR screen , we tested these mutants as a single pool in at least six mice . Three of the 15 homozygous deletion mutants ( gal10 [orf19 . 3672] , dfi1 [orf19 . 7084] , and hap41 [orf19 . 740] ) showed significantly reduced levels of GI tract colonization whereas one ( nce102 [orf19 . 5960] ) displayed reduced fungal burden in kidneys after tail vein infection ( Figure 7 ) . ( Although dfi1 did not meet statistical significance , it shows a trend towards reduced fungal burden , which is consistent with a previous report [34] ) . As predicted by the ChIP-chip–based network ( Figure 6A ) , Rtg1/3 and Hms1 regulate the expression levels of these targets ( Figure S2 ) . None of the identified genes had been previously implicated in intestinal colonization . While the C . albicans GAL10 gene encodes an enzyme of the galactose utilization pathway [35] , the ability to use galactose as a carbon source per se is unlikely to contribute to the mutant's inability to colonize the GI tract because we did not observe similar defects in the GAL1 mutant ( Figure 7 ) ( the GAL1 gene encodes another enzyme of the galactose utilization pathway ) . Rather , the colonization defect may be related to the anomalous cell wall ultrastructure found in the gal10 mutant , an observation supported by the increased sensitivity of the mutant to cell wall disturbing agents such as Congo Red [35] . DFI1 encodes a cell wall-linked protein that promotes invasive filamentation when C . albicans is grown in semi-solid medium [34] . Like the gal10 mutant , the C . albicans dfi1 mutant is hypersensitive to cell wall disturbing agents such as Congo Red and Caspofungin [34] . This observation implicates determinants of cell surface integrity in the ability of C . albicans to colonize the murine GI tract . DFI1 also appears to signal through the Cek1 kinase to promote adhesion in addition to filamentation [34]; both properties seem important for the fungus to endure in the intestine . Little is known about the function of HAP41 and NCE102 in C . albicans . HAP41 is a S . cerevisiae HAP4 homolog but lacks a DNA-binding domain . In S . cerevisiae , the heme-activated , glucose-repressed Hap2p/3p/4p/5p CCAAT-binding complex is a transcription activator and global regulator of respiratory gene expression . The C . albicans genome harbors multiple homologs of each of the subunits of the S . cerevisiae complex . Unlike other HAP gene transcripts , HAP41 does not respond to iron deprivation conditions in C . albicans [36] suggesting that its function may be different from its S . cerevisiae homologs . The S . cerevisiae NCE102 gene encodes a transmembrane protein localized to discrete membrane compartments [37] and has been implicated in protein export [38] and as a sensor of sphingolipids [37] . To our knowledge , this is the first report that C . albicans NCE102 plays a role in the host .
The collection of TR target genes in the network includes a large and diverse set of biological functions , but three broad functions/categories are most noticeable: ( 1 ) acquisition and metabolism of carbon; ( 2 ) acquisition and metabolism of nitrogen; and ( 3 ) transporters and cell surface proteins . The acquisition and metabolism of carbon and nitrogen are among the most prominent challenges faced by bacteria that live in the gut as well [46] . Moreover , bacterial pathogens that undergo mutations as well as gene gains/losses resulting in alterations of their metabolic capabilities often display a selective advantage [47] . Cell surface remodeling is a key strategy used by microorganisms to circumvent host defenses; in fact , the ability to do so has been demonstrated to contribute to the virulence of a broad range of pathogens including bacteria [48] , fungi [12] , [49] , and parasites [50] . In bacterial species that can turn from harmless commensals to life-threatening pathogens , surface proteins also appear to play major roles in the transition between commensalism and pathogenicity [51] . Carbohydrates consumed by the gut microbiota are typically oligo- or polysaccharides derived from diet , host mucosal secretion , or other resident ( or dietary ) microbes [46] . In the bloodstream , on the other hand , glucose is the only sugar available whereas in internal organs the carbohydrates available are probably those from the proteoglycans that form the extracellular matrix , an ubiquitous constituent of animal tissues [52] . This major difference in the potential source of carbon between the two locales suggest that the strategy that C . albicans employs to obtain carbohydrates in the GI tract should differ , at least in part , from the strategy used while in the bloodstream or internal organs . Consistent with this notion , we find that TYE7 , one of the major regulators of carbohydrate metabolism in C . albicans [24] , is needed to proliferate in the gut but not during systemic infection . RTG1/3 and HMS1 , both required not only for gut colonization but also for full fitness after bloodstream infection , bind upstream of a significant number of genes involved in hexose catabolism ( Dataset S2 ) . This function may be important during systemic infection because genes involved in the assimilation of alternative carbon sources have been found to be upregulated in C . albicans cells during infection of the mammalian kidney [53] and carbon metabolism has been implicated in the infections of other fungal pathogens as well [54] . Metabolic flexibility , in general , has been postulated to be a requisite for C . albicans infection due to the dynamic nature of host niches which contain complex arrays of nutrients [55] . How nitrogen is acquired by microorganisms living in the GI tract remains an open question . Several bacterial species that live in the gut , e . g . , Bacterioides , seem to rely on NH3 ( reviewed in [46] ) . Other sources could be amino sugars and proteins that are present in secreted mucus and epithelial cells , or amino acids derived from diet [46] . Consistent with the latter , Rtg1/3 , one of the C . albicans TRs controlling intestinal colonization , has a number of putative amino acid permeases among their target genes . In addition , we find that Lys144 binds upstream of each of four putative allantoate transporters raising the possibility that allantoate , a product of purine catabolism in some bacteria , is one of the sources of nitrogen for C . albicans in the gut . Contrary to what their nomenclature implies , neither Lys144 nor Lys14 seems to regulate lysine biosynthesis genes in C . albicans ( our ChIP data and phenotypic screen results in [16] ) . The majority of the pathogen-associated molecular patterns ( PAMPs ) that activate and modulate immune responses are cell wall components [56] , [57] . Indeed , C . albicans mutants that are unable to add particular carbohydrate moieties to their surface proteins are attenuated for virulence in mouse models of systemic infection . ZCF21 and LYS14 , the two TRs that influence the outcome of systemic infections but not colonization of the GI tract , have among their targets a significant number of genes encoding proteins predicted to be localized to the cell surface or enzymes that modify the cell wall structure such as the mannosyltransferase OCH1 and the glucosyltransferase ALG6 . Hence , our findings reveal two regulators that C . albicans employ to remodel its surface and indicate that these modifications are needed during systemic infection . In summary , our findings indicate that the ability of C . albicans to colonize multiple niches within a mammalian host relies on a large , integrated circuit that responds to different environmental conditions to effect major changes in metabolic functions , nutrient ( especially carbon and nitrogen ) acquisition , cell wall remodeling , and cell wall integrity . We propose that this “master circuit” allows C . albicans to adjust to disparate environments in the host and accounts for the close links between commensalism and pathogenicity .
The procedure used was essentially the one described [19] , [20] . Female Swiss Webster mice ( 18–20 g ) were treated with antibiotics ( tetracycline [1 mg/ml] ) , streptomycin [2 mg/ml] , and gentamycin [0 . 1 mg/ml] ) added to their drinking water throughout the experiment beginning 4 d before inoculation . Prior to inoculation , C . albicans strains were grown for ∼18 h at 30°C in YPD liquid medium , washed twice with PBS , and counted in a hemocytometer . Mice were orally inoculated with 5×107 C . albicans cells ( in a 0 . 1 or 0 . 2 ml volume ) by gavage using a feeding needle . Colonization was monitored by collecting fecal pellets ( produced within 10 min prior to collection ) at various days post-inoculation and cecum contents at the end of the experiment when the mice were killed . In the initial screening the fecal pellets and intestinal contents were used to prepare genomic DNA directly from the samples . In follow-up experiments , the mouse homogenates were plated onto Sabouraud medium containing ampicillin ( 50 µg/ml ) and gentamycin ( 15 µg/ml ) ( antibiotics were included to prevent the growth of contaminating bacteria ) . Genomic DNA was prepared from yeast scraped off the plates . The yields of DNA prepared directly from fecal pellets and intestinal contents were relatively low , hence one round of whole genome amplification ( using Sigma's GenomePlex Complete Whole Genome Amplification kit ) was used to generate adequate amounts of material before the qPCR analysis . Similar results were obtained with samples that were plated or that were processed directly . The 77 C . albicans deletion mutants screened are listed in Table S1 . We assayed pools of 15–20 signature-tagged mutant strains . The relative abundance of the strains recovered from feces ( at 1 , 9 , and 18 or 21 d post-inoculation ) or intestinal contents ( at day 18 or 21 ) compared with the inoculum was determined by real time PCR ( using primers to the signature tags ) as described [15] . Briefly , threshold cycle ( CT ) values were converted to a linear scale using the simple equation , linear value = 2−CT . Experiments comparing 15 strains resulted in 15 values for the inoculum ( I ) and another 15 for the recovered pool ( Rraw ) . Rraw values were multiplied by median ( I ) /median ( Rraw ) to generate normalized R values . Ultimately , R/I was calculated for each mutant strain . These ratios expressed as log2 values are shown in Figure 2A . Empirically we found that the limit of accurate qPCR detection for most strains was about 1 , 000-fold reduction in levels compared to the inoculum . This level of detection is consistent with the number of colonies , typically ∼10 , 000 , that we recovered after plating around 10 mg of the fecal pellet and intestinal content homogenates . The procedure used has been described by our laboratory [15] . We used the t-test to compare the log2 ( R/I ) of mutants to those of wild-type using a significance threshold of p<0 . 05 ( correcting for multiple comparisons ) . Ten female BALB/c mice ( 18–20 g ) were infected with wild-type C . albicans or one of the mutant strains by tail vein injection . Saturated C . albicans cultures were diluted 1∶25 in YPD and grown for ∼4 h at 30°C prior to infection . Cells were washed twice with sterile saline , counted in a hemocytometer and 5 . 2×105 cells ( in a 0 . 1 ml volume ) were injected in each mouse . Mice were monitored daily and sacrificed when moribund . The logrank test was used for statistical analysis Each TR was tagged with a 13-MYC or GFP tag at the C- or N-terminal end of the protein in a wild-type reference strain background . The tagged strains along with untagged controls were grown as indicated in Figure 5 and ChIP was carried out as described [60] with the following modifications: GFP-tagged regulators were immunoprecipitated with an anti-rGFP polyclonal antibody ( Clontech ) ; the DNA recovered after crosslink reversal was purified with QIAquick PCR purification columns ( Qiagen ) and amplified using the GenomePlex Complete Whole Genome Amplification kit ( Sigma ) . Input and immunoprecipitated DNA were fluorescently labeled and competitively hybridized to custom full-genome oligonucleotide tiling microarrays ( Agilent ) as described [44] . MochiView [61] was used for data visualization , identification of binding events , and DNA motif analysis . The microarray data were normalized using the global lowess method . The normalized log2 enrichment values ( IP/input ) for each probe were imported into MochiView and the software's default parameters were used to smooth the data and extract binding events ( peaks ) . The cutoff for the minimum value for peak inclusion was set at two or three standard deviations from the mean of the log2 enrichment values ( cutoffs were typically in the range of 0 . 6–0 . 8 ) . To ensure the generation of a high confidence dataset , in addition to the standard analysis performed by MochiView we manually curated all the extracted peaks using the following criteria: ( 1 ) ChIP data derived from untagged control strains immunoprecipitated with anti-MYC and anti-GFP antibodies were used to filter out non-specific peaks ( this function is incorporated in MochiView ) ; ( 2 ) peaks located within annotated ORFs were discarded; ( 3 ) peaks located around highly expressed genes ( particularly ribosomal genes ) were also discarded because based on our experience ( e . g . , [43] , [44] , [62] ) these places tend to bind to almost all DNA-binding proteins non-specifically; and ( 4 ) we only included peaks that were consistent in two independent biological replicates ( typically >80% of peaks were concordant in the replicates ) . Sequences of 500 nt centered on the midpoint of about 20–30 of the top-scoring peaks for each regulator were used to derive motifs in MochiView . The software's default parameters were employed . To assess the significance of the derived motifs , we compared their occurrence in the remaining peaks versus their occurrence in a set of random intergenic regions of the same length . This analysis was performed using MochiView's “enrichment” function . The logrank test was used to compare the persistence or depletion of the various C . albicans mutants in the murine GI tract ( Figure 2B ) and to compare the time-to-illness curves of monotypic infections ( Figure 4B–4D ) . The t-test ( two-tailed , comparison of unpaired samples ) was used to evaluate the significance of the log2 ( R/I ) values of the mutants versus the wild-type reference strain in the systemic infection screen ( correcting for multiple comparisons ) . The hypergeometric distribution was used to evaluate the significance of the overlap between sets of genes . The Gene Ontology Term Finder feature of the Candida Genome Database ( www . candidagenome . org ) was used to search biological processes or functions enriched in the various datasets . EMSAs were carried out as described previously [63] . Cells were grown to mid-late logarithmic phase in YPD or Todd Hewitt Broth at 30°C . Total RNA was prepared with the RiboPure-Yeast kit ( Ambion , Life Technologies ) following the manufacturer's instructions . Three micrograms of purified RNA per sample were used to synthesize cDNA with SuperScript II Reverse Transcriptase ( Invitrogen ) . Quantification of transcripts was performed by real-time PCR using SYBR green . Results were normalized to those of the actin gene ( ACT1 ) . The ChIP-chip data reported in this article have been deposited in the NCBI Gene Expression Omnibus ( GEO ) database under accession number GSE41237 . | Our skin and mouth , as well as our genital and gastrointestinal tracts , are laden with microorganisms belonging to all three domains of life ( bacteria , archaea , and eukaryotes ) . Much of the time these commensal microorganisms are not only harmless but provide advantages to us . However , when the host's defenses are compromised , some members of the normal flora , such as the fungus C . albicans , can cross the host's protective barriers and colonize virtually every internal organ causing life-threatening conditions . The environment found in the bloodstream and internal organs is presumably distinct from the mucosal surfaces where our flora typically resides . Whether opportunistic pathogens such as C . albicans rely on common or separate gene repertoires to thrive in each of these locales is largely unknown . To address this question we carried out genetic screens in mouse models that recapitulate niches where C . albicans thrives and used genome-wide experimental approaches to uncover the genes required to proliferate in each environment . In fact , the ability of C . albicans to colonize disparate niches within a mammalian host relies on a large , integrated circuit . Our observations suggest that at least some key gene circuits are not dedicated to one niche or another . Rather , thriving in various locales of the host seems to involve the complex regulation of multiple processes , which may allow C . albicans to adjust to different environments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"biology",
"microbiology",
"genetics",
"and",
"genomics"
]
| 2013 | Candida albicans Commensalism and Pathogenicity Are Intertwined Traits Directed by a Tightly Knit Transcriptional Regulatory Circuit |
Persistently low white blood cell count ( WBC ) and neutrophil count is a well-described phenomenon in persons of African ancestry , whose etiology remains unknown . We recently used admixture mapping to identify an approximately 1-megabase region on chromosome 1 , where ancestry status ( African or European ) almost entirely accounted for the difference in WBC between African Americans and European Americans . To identify the specific genetic change responsible for this association , we analyzed genotype and phenotype data from 6 , 005 African Americans from the Jackson Heart Study ( JHS ) , the Health , Aging and Body Composition ( Health ABC ) Study , and the Atherosclerosis Risk in Communities ( ARIC ) Study . We demonstrate that the causal variant must be at least 91% different in frequency between West Africans and European Americans . An excellent candidate is the Duffy Null polymorphism ( SNP rs2814778 at chromosome 1q23 . 2 ) , which is the only polymorphism in the region known to be so differentiated in frequency and is already known to protect against Plasmodium vivax malaria . We confirm that rs2814778 is predictive of WBC and neutrophil count in African Americans above beyond the previously described admixture association ( P = 3 . 8×10−5 ) , establishing a novel phenotype for this genetic variant .
A large proportion of healthy African Americans have been observed to have a white blood cell count ( WBC ) that is persistently lower than the normal range defined for individuals of European ancestry [1]–[5] . This condition , called “benign ethnic neutropenia” , can have important effects on medical decision-making , since WBC is a valuable indicator of immunocompetence , infection , and inflammation . To seek the genetic basis of benign ethnic neutropenia , we recently carried out an admixture mapping analysis in which we identified a locus on chromosome 1 where local inheritance of African or European ancestry is sufficient to account entirely for the epidemiological differences in WBC levels between African Americans and European Americans [6] . By genotyping samples from two epidemiological cohorts—the Health Aging and Body Composition Study ( Health ABC ) and the Jackson Heart Study ( JHS ) —at a panel of markers that were extremely differentiated in frequency between Africans and Europeans , we identified an approximately 900 kilobase locus on chromosome 1 ( 99% credible interval of 155 . 46–156 . 36 Mb ) where individuals with low WBC had increased African ancestry compared with the average in the genome . In the present study , we narrowed the region of association from 900 kb to a single base pair substitution that is likely to have a strong effect on variation in WBC . To achieve this , we increased our sample size from 1 , 550 in the initial study to 6 , 005 , by pooling samples from the Jackson Heart Study ( JHS ) , the Health ABC Study , and the Atherosclerosis Risk in Communities ( ARIC ) Study . We found that neutrophil count is responsible for the vast majority of the WBC association at the locus , and therefore focused on neutrophil count in the current analysis . We also showed that the genetic change that is probably responsible is the Duffy Null polymorphism ( rs2814778 , also called FY+/− ) , which is already known to protect individuals of African descent against Plasmodium vivax malaria infection [7] , [8] , and which has recently been associated with susceptibility to HIV infection and rate of progression to AIDS [9] . Our identification of this polymorphism as the probable cause of benign ethnic neutropenia should prompt further investigation of its effects on hematopoiesis and immunity .
We pooled 6 , 005 African American samples from three cohort studies: the Jackson Heart Study ( JHS ) , the Atherosclerotic Risk in Communities ( ARIC ) Study , and the Health , Aging and Body Composition ( Health ABC ) Study . For each sample , we required a high quality genome-wide admixture scan ( Materials and Methods ) , a genotype at SNP rs2814778 , body mass index ( BMI ) , age , gender , and a full differential white blood cell count ( with the exception that for Health ABC samples we did not require a measurement of bands ) . To explore correlations between the genetics and the phenotype , we first used the genotype at SNP rs2814778 , which occurs at position 155 , 987 , 755 in Build 35 of the human genome reference sequence , within the 99% credible interval defined by our previous admixture mapping study [6] . This SNP is also known as the “FY+/−” or “Duffy” variant , and the FY− allele is very highly correlated to West African ancestry . For example , it is completely fixed in frequency in West African and European American samples from the International Haplotype Map [10] ( although it is not completely fixed in larger sample sizes from these populations; see below ) . For Figure 1 and Tables 1 and 2 , we used the genotype at rs2814778 as a surrogate for ancestry because the genotype can be conveniently read out as a discrete value ( 0 , 1 or 2 copies ) rather than as a continuous value , and is extraordinarily correlated to ancestry ( r2>0 . 99 ) . Later , we demonstrate that there is in fact a slightly stronger association to neutrophil count for rs2814778 than for ancestry , which is important in showing that the FY− allele at this polymorphism may actually be responsible for low neutrophil counts , and is not just in admixture linkage disequilibrium with the causal allele . To test for heterogeneity in the strength of the genetic association to WBC among the different sample sets that comprised our study , we divided the samples into four groups . There were 658 samples from the Health Aging and Body Composition Study ( “Health ABC” ) , 1 , 969 samples from the JHS cohort only ( after randomly dropping samples until there was only one from each pedigree; “JHS only” ) , 2 , 476 samples from the ARIC cohort only ( “ARIC only” ) , and 902 samples that overlapped between JHS and ARIC ( “JHS-ARIC overlap” ) . For the JHS-ARIC overlap samples , we averaged all phenotype measurements , taken an average of 14 years apart at the time the participant entered each study ( and having a correlation coefficient of r2 = 0 . 37 ) , to provide a more precise estimate of the phenotype than would be available from either measurement alone . Table 1 presents the characteristics of each of the groups of samples . We found that all sets of samples showed quantitatively similar associations to the chromosome 1 locus . In particular , for neutrophil count , individuals carrying at least one European-type ( “FY+” ) allele of rs2814778 had 1 . 58–1 . 65 times higher values , depending on the study , than individuals homozygous for African ancestry ( FY−/− ) , a tight enough range that we decided to pool all four groups of samples for subsequent analyses . Despite the similar correlation of neutrophil count to local ancestry across studies , we observed that the correlation coefficient to “European carrier status” was significantly higher for JHS-ARIC overlap samples , ρ = 0 . 57 , than for the samples for which only one measurement was made: ρ = 0 . 52 for JHS-only ( P = 0 . 03 for a reduction ) and ρ = 0 . 50 for ARIC-only ( P = 0 . 002 for a reduction ) ( Table 1 ) . This is likely to reflect a more accurate assessment of basal neutrophil count when it was measured twice and averaged over different environmental conditions ( the baseline measurements in JHS and ARIC studies taken an average of 14 years apart ) than when it was measured only once . In support of this hypothesis , the JHS-ARIC overlap samples contributed more per sample to the statistical signal than those measured in only one cohort: 28% more per sample on average , which we calculated by dividing the LOD score they contributed by the total number of samples . Combining all samples ( n = 6 , 005 in total ) and working with normally transformed cell counts for each white blood cell lineage ( Materials and Methods ) we explored how counts of total WBC and each of the 6 differential counts were associated with ancestry at the locus , using the genotype at rs2814778 as a surrogate for ancestry ( Table 2 ) . There was strong evidence that the allele at the locus that contributed to high white blood cell count had an almost purely dominant effect . As shown in Table 2 , there was no significant difference in leukocyte counts between 247 African Americans with two copies of European ancestry at this locus ( FY+/+ ) and 1 , 647 African Americans who were FY+/− ( P>0 . 08 for WBC and all differential counts ) . By contrast , being FY+/+ or +/− ( 1 , 894 African Americans ) vs . FY−/− ( 4 , 111 African Americans ) was strongly associated to counts of all white blood cell types except bands ( P<<10−4; Table 2 ) . The dominant effect of European ancestry on white blood cell count is also visually apparent in Figure 1 , which shows the distribution of neutrophil count for individuals grouped according to genotype at rs2814778 . Persons carrying at least one FY+ allele had a distribution of neutrophil counts that was shifted by 1 . 3 standard deviations above that of persons who were FY−/− ( this was extraordinarily statistically significant: Z = 49 . 7 ) . By contrast , there was no significant difference between individuals who carried either one or two FY+ alleles ( Z = 0 . 6 ) . For further analysis , we pooled individuals who were carriers of the FY+ allele at this locus . The differential white blood cell count that was most significantly associated with ancestry was absolute neutrophil count ( calculated as total WBC multiplied by the percentage of neutrophils ) . The correlation ( ρ ) of normally transformed absolute neutrophil count to carrier status for the FY+ allele was 0 . 519 , which was higher than that of the general WBC phenotype originally mapped to the locus [6] ( ρ = 0 . 458 ) . In the 952 African Americans who had absolute neutrophil counts at least 1 s . d . below the mean ( roughly <1 , 800 /mm3 ) , the proportion of FY+ allele carriers was reduced by more than an order of magnitude compared with the genome wide average . Neutrophil count was responsible for the vast majority of WBC association at the locus . After controlling for neutrophil count in a regression analysis , only monocyte count ( ρ = 0 . 025 , P = 0 . 05 ) and basophil count ( ρ = −0 . 034 , P = 0 . 009 ) remained nominally associated , and these associations were not significant after correcting for the 6 hypotheses tested ( Table 2 ) . The weak evidence of association to monocyte and basophil counts may reflect a real effect , or may be a false-positive due to multiple hypothesis testing . It is also possible that the result may be an experimental artifact related to the Coulter Counter technology used to measure differential WBC . In these measurements , the positions of monocytes and basophils were near those of neutrophils in the plots used for cell classification . Even a small amount of misclassification among neutrophils , monocytes , and basophils ( a couple of percent ) could cause their counts to be artifactually correlated , contributing to the signals we observe in the context of measurements in large sample sizes . Since neutrophil count appears to drive at least the great majority of association , we focused on this WBC phenotype for all further analysis To assess whether the higher neutrophil count observed in European Americans compared with African Americans can be entirely accounted for by ancestry at the chromosome 1 locus , as is the case with total WBC [6] , we examined samples from the Health ABC study ( 1 , 331 European Americans and 658 African Americans ) . Among African Americans who could be classified with confidence as carrying at least one chromosome of European ancestry at the locus , the absolute neutrophil count did not differ from that of European Americans ( P = 0 . 99 ) . Thus , genetic variation at the chromosome 1 locus was sufficient to account for the entire epidemiological difference across these populations . The predictive effect of ancestry at the chromosome 1 locus was profound . Carrier status for the European-type ( FY+ ) allele at the rs2814778 variant predicted 26 . 95% of the variance in normally transformed neutrophil count , which was far more than the 3 . 37% predicted by genome-wide European ancestry proportion . After controlling for rs2814778 genotype , there was no longer any association to genome-wide European ancestry . Similarly , after controlling for rs2814778 genotype , BMI and gender only predicted 0 . 79% and 0 . 14% of variance in neutrophil count respectively , while smoking ( analyzed in JHS only ) only predicted 0 . 8% of the variance . Age was not significantly associated to neutrophil count in our data ( P = 0 . 25 ) . We did not analyze other phenotypes like hypertension and coronary artery disease status for their correlation to neutrophil count . Because of the relatively weak contributions of all the non-genetic predictors we analyzed , we focused subsequent analyses on genotype at the chromosome 1 locus uncorrected for covariates . We were able to place strong constraints on the frequency of the variant affecting neutrophil count by analyzing the distributions of neutrophil count for individuals with 0 , 1 and 2 copies of European ancestry at the chromosome 1 locus , which in practice we marked by the genotype at rs2814778 . The analysis in Figure 1A provides strong evidence of a dominant allele of European origin contributing to high neutrophil count . We modeled the frequency of the variant that causes high neutrophil count by defining 6 parameters . The frequency of this variant in Africans was specified as PA and its frequency in Europeans as PE . Individuals who were homozygous for the other allele were assumed to have a normal distribution of neutrophil count with mean μL and standard deviation σL , and carriers of the “high neutrophil” allele were assumed to have a normal distribution of neutrophil count with mean μH and standard deviation σH . Studying a grid of values of PA ( Figure 1B ) , and another grid of values of PE ( Figure 1C ) , we found the combination of the remaining variables that provided the best fit to the data , as assessed by a chi-square goodness-of-fit statistic . Given each set of 6 model parameters , we calculated a likelihood of the data for all 6 , 005 individuals . This resulted in a marginal likelihood surface for PA ( Figure 1B ) and PE ( Figure 1C ) , which we used to place constraints on these parameters . Fitting this 6-parameter model to the data , we inferred that the frequency of the allele contributing to high neutrophil counts was <4 . 9% in Africans and >95 . 2% in Europeans ( Figure 1B , C ) , and that the difference in frequency between populations was >91 . 9% . Compared to 3 . 54 million autosomal SNPs in the November 2006 Phase2 HapMap data set [10] , there were only 115 SNPs with a frequency differentiation at least this extreme , and only one in the region of admixture association: the SNP rs2814778 ( at position 155 . 99 Mb ) , the same SNP we used as a marker of ancestry . This variant already has a known phenotype—susceptibility to Plasmodium vivax malaria—but it had not been hypothesized to be associated with low white blood cell count until it was found to lie within this locus [6] . While rs2814778 is a plausible candidate , the locus we described previously [6] spans 900 kb , and there could in principle be other variants within this span—unreported in the literature or in genome variation databases—that have a high enough frequency differentiation to explain the signal . In what follows , we present additional lines of evidence to rule out the great majority of sites other than rs2814778 as consistent with explaining the signal . We used four strategies to increase the height of our admixture association peak and thereby to narrow the position of the allele affecting neutrophil count . First , we used the fact that with 6 , 005 samples in our admixture mapping analysis pooled across three studies ( Table 3 ) , we had a greater sample size than the 1 , 550 samples that were used initially [6] . Second , we designed an analysis ( Materials and Methods ) that used all of the samples instead of just the extremes of the distribution [6] . Third , we changed the phenotype from total WBC ( ρ = 0 . 458 correlated to ancestry at the locus ) to absolute neutrophil count ( ρ = 0 . 519 ) to obtain a sharper statistical signal . Finally , we genotyped the JHS samples ( including the JHS-ARIC overlap samples ) at additional ancestry informative markers , spaced at a density of about 1 every 400 kb across the peak , to increase spatial resolution . Using all 6 , 005 samples and the stronger phenotype of absolute neutrophil count , the LOD score ( log base 10 of the Bayes score ) rose to 363 . 1 ( Table 3; Figure 2 ) . The 99% credible interval was narrowed to ∼450 kb ( 155 . 957–156 . 407 kb ) , and still contained the Duffy null polymorphism at position 155 , 987 , 756 near the DARC gene ( Entrez GeneID: 2532 ) , as well as a handful of other genes listed in the lower panel of Figure 2 . We exploited the large sample size ( 6 , 005 individuals ) to test whether the rs2814778 variant predicted low neutrophil count more than would be expected from the association to ancestry [6] . This is a difficult problem since the genotype at this SNP is highly correlated to ancestry . By using the ANCESTRYMAP software and the data from all 6 , 005 African Americans , we estimated that the frequency of FY+ allele at rs2814778 is 0 . 2±0 . 1% in Africans and 99 . 3±0 . 4% in Europeans ( this frequency distribution is consistent with the allele frequencies inferred for the causal allele based on modeling of neutrophil counts in Figures 1B and 1C ) . Thus , if rs2814778 is the causal variant , there should be a small handful of individuals for whom the genotype at rs2814778 is discrepant with ancestry , who will be informative for our analyses . To estimate the number of individuals who we expect to be informative for testing association of rs2814778 above and beyond ancestry , we used the fact that the cohort has 18 . 2% European ancestry on average ( Table 2 ) . Thus , we expected there to be about 13 individuals who are homozygous for the Duffy null allele at rs2814778 but heterozygous for European ancestry: 13 = ( 6005 ) × ( 2×18 . 2%×81 . 8% ) × ( 0 . 7% ) . Similarly , we expected there to be about 8 individuals who are heterozygous at rs2814778 but homozygous for local African ancestry: 8 = ( 6005 ) × ( 81 . 8%×81 . 8% ) × ( 0 . 2% ) . To test for association to rs2814778 above and beyond ancestry , we first obtained estimates of European ancestry at the position of the SNP using the ANCESTRYMAP software [11] . We included rs2814778 in the ancestry estimation so that we could explicitly test whether the genotype at this SNP alone was more predictive of neutrophil count than this SNP plus flanking markers . This would be evidence that it was more associated than African ancestry itself . Our power to detect a signal was highest for JHS samples , which were genotyped at a high density at the chromosome 1 locus . Consistent with this observation , the 7 samples for which we could state with >50% confidence that the local ancestry was discrepant with the expectation from the rs2814778 genotype were all from JHS . We performed three regression analyses ( Table 4 ) to explore whether rs2814778 or ancestry status at the chromosome 1 locus was a better predictor of neutrophil count . ( a ) First , we obtained a χ2 statistic for association of carrier status for the rs2814778 FY+ allele to neutrophil count; ( b ) second , we obtained a χ2 statistic for association of carrier status for European ancestry to neutrophil count ( using the rs2814778 genotype in the estimate ) ; and ( c ) third , we obtained a χ2 statistic for association of both predictors together . We found that there was a significant difference between the strength of association of ancestry alone and ancestry and genotype together: ( c ) - ( b ) = 15 . 7 ( P = 3 . 8×10−5 ) . Testing for the reverse effect of ancestry above and beyond the genotype of rs2814778 produced no signal: ( c ) - ( a ) = 0 . 4 ( P = 0 . 74 ) . These results confirm that rs2814778 is predictive of neutrophil count , above and beyond the effect of ancestry . To search for additional alleles in the admixture peak that might be associated to neutrophil count beyond the main effect , we genotyped a dense panel of 193 SNPs across the region in 148 individuals with low neutrophil count ( <−0 . 7 standard deviations below the mean ) and 74 individuals with high neutrophil count ( 1 . 3–2 . 8 standard deviations above the mean ) . We chose only individuals for whom we were >99% confident of all African ancestry at the locus , based on genotyping information at flanking markers excluding rs2814778 , so that ancestry would not be a confounder of the analysis . We genotyped these individuals for a set of SNPs chosen using Tagger [12] to capture the great majority of common variation across the admixture peak in both West Africans and Europeans ( Materials and Methods ) . After the genotyping was complete , we had captured 94% of SNPs of >5% minor allele frequency in West Africans , and 96% of SNPs of >5% minor allele frequency in European Americans , both at a correlation of r2>0 . 8 ( Figure 3B ) . Case-control association analysis of these 193 SNPs identified only one , rs2814778 , that was significantly associated ( nominal P = 2 . 1×10−5; Figure 3A ) after a Bonferroni correction for 193 multiple hypothesis tests . Thus , there was no evidence of any allele in the region that is associated to neutrophil count beyond the effect that is already captured by rs2814778 . We genotyped 10 , 062 self-identified European Americans in the ARIC study for rs2814778 , searching for a decreased neutrophil count in association with the null allele . This analysis should have little power if the European American population is in Hardy-Weinberg equilibrium , since FY−/− homozygotes are expected to occur very rarely among Europeans: less than 1/10 , 000 based on the observed frequency of the null allele in this population ( 0 . 34 = 10 , 062×0 . 58%×0 . 58% ) . Interestingly , we observed 7 European Americans with FY−/− genotypes , a significant excess compared with expectation ( P<4×10−9 ) suggesting that European Americans harbor population substructure with variable levels of African ancestry . Among the FY−/− homozygotes we found a non-significant reduction in WBC associated with the null allele: WBC was observed to be 5 . 9±2 . 6 for the 7 FY−/− homozygotes , 5 . 9±1 . 8 for the 103 FY+/− heterozygotes , and 6 . 3±1 . 9 for the 9 , 952 FY+/+ homozygotes ( P = 0 . 06 with an additive model and P = 0 . 35 with a dominant model using 1-sided tests ) . Genotyping of rs2814778 in 1 , 339 self-identified European Americans from the Health ABC study identified 26 heterozygous individuals , and none homozygous for FY−/− . These analyses strongly increase the likelihood that a single nucleotide change at the site of the rs2814778 polymorphism is responsible for low neutrophil counts , and provide no evidence of any other allele contributing a signal . However , these findings do not rule out the existence of undiscovered variants in the admixture peak that are differentiated enough to explain the signal . If such variants existed , then rs2814778 could simply be a marker in linkage disequilibrium with the causative variant rather than being causal itself . We carried out an analysis in which we systematically ruled out the majority of other nucleotides in the region as potentially contributing to the signal . We examined genomic databases to identify DNA sequence fragments that are known to be of either African or European ancestry and that overlap the admixture peak , and considered nucleotides where all African chromosomes had one allele and all European chromosomes had the other . Based on our modeling in Figure 1 , it is likely that the causative variant is sufficiently differentiated that it would be found by this discovery strategy . We mined shotgun sequencing data from 6 individuals of West African ancestry and 5 individuals of European ancestry ( Materials and Methods ) . We restricted analysis to nucleotides for which we had a high sequence quality score ( Neighborhood Quality Score of ≥40 ) [13] , randomly sampling one sequence to represent each individual . We supplemented the shotgun data with the human genome reference sequence , which is comprised of a mosaic of sequence from 5 BAC clones across the region . We determined that 3 of the clones ( one from individual CIT978SK and two from individual RPCI-11 ) were of European ancestry , and 2 were of African ancestry ( both from RPCI-11 ) ( Figure 4 ) . Interestingly , since RPCI-11 was heterozygous for African and European ancestry at this locus , this individual is probably African American . Because RPCI-11 is the source for most ( ∼74% ) of the human genome reference sequence [14] , we conclude that much of the public human genome reference sequence is that of an African American , and includes a substantial amount of sequence of African ancestral origin . We found that 82 . 3% of the admixture peak was covered in at least one chromosome from each population ( an average of 2 . 2× European and 2 . 1× African coverage ) . Of the 817 SNPs we identified , 594 could be ruled out as not completely differentiated in frequency between the European and African chromosomes used in SNP discovery , an additional 79 could be ruled out as not sufficiently differentiated across populations based on data from the International Haplotype Map database [10] , and 49 could be ruled out by their allele frequencies in our own follow-up genotyping of HapMap samples . Thus , 88 . 5% ( = ( 594+79+49 ) /817 ) of SNPs discovered in the admixture peak could be ruled out . This allowed us to infer that 72 . 8% ( = 82 . 3%×88 . 5% ) of nucleotides can be excluded as causal for the observed major effect on variation in neutrophil count . These results provide yet another line of evidence that the rs2814778 single nucleotide change may itself be causing low neutrophil levels . There are now five reasons why we believe rs2814778 is likely to be the direct cause for low neutrophil count: ( 1 ) rs2814478 falls within the ∼450 kilobase admixture peak . ( 2 ) rs2814778 contributes a signal of association above and beyond the admixture signal , showing that the true underlying causal variant is in an even narrower region around rs2814478 . ( 3 ) rs2814778 is already known to have functional consequences , based for example on past molecular work showing that it affects expression of an antigen on red blood cells and thus modulates resistance to P . vivax malaria . ( 4 ) rs2814778 is known to have a frequency differentiation across populations that makes it consistent with the underlying causal variant; a degree of differentiation that is extremely unusual , with only 0 . 003% of known SNPs in the genome having a differentiation this extreme . ( 5 ) Genome sequencing data directly rule out about three quarters of other nucleotides in the admixture peak as containing the variant . ( We have not carried out a similar analysis of insertion/deletion polymorphisms , which are known to occur at about a tenth the rate of SNPs . ) We caution that an association study is always correlational , and can never by itself prove a functional effect of an allele . To prove causality , it is essential to follow up any association study with biological work . Nevertheless , the present study provides the best example of which we are aware of taking association analysis to its limit , and using association analysis to demonstrate a likely causal effect . Our study justifies further work to understand the biological mechanism by which a single nucleotide change at rs2814778 probably causes reduced neutrophil counts .
We have used admixture mapping to localize a variant affecting neutrophil levels to a region of about 450 kb centered on the Duffy null locus . We have further shown that the underlying variant must be >91% different in frequency between West Africans and European Americans , placing it among the top 115 HapMap SNPs in the genome in terms of allele frequency differentiation ( top 0 . 003% ) . Since only one SNP in the admixture peak , the Duffy null polymorphism rs2814778 , is known to be this differentiated in frequency across populations , we tested this SNP for evidence of association above and beyond the ancestry effect , and found a signal ( P = 3 . 8×10−5 ) . Finally , we ruled out the great majority of nucleotides across the region , apart from rs2814778 , as sufficient to cause the signal . Methodologically , these results provide a case-example of fine-mapping in a difficult context . We have moved from an initial association signal discovered via mapping by admixture linkage disequilibrium , to a more fine-grained association based on linkage disequilibrium inherited from the ancestral African and European populations , and finally to an analysis where we systematically excluded the great majority of nucleotides in the region as contributing to the association . The study is also novel in demonstrating the value of mapping in multi-ethnic and admixed populations . The variant could not have been mapped in non-African Americans ( either Africans or Europeans ) , since it is nearly fixed in both populations , showing how studying diverse populations is important in biology . For example , when we genotyped rs2814778 in more than 10 , 000 European Americans from the ARIC study , we could not obtain a replication despite the large sample size . The mechanism of low neutrophil count in persons homozygous for the FY− allele is unknown . Interestingly , Yemenite Jews also have a high frequency of the FY− allele [15] , which we hypothesize explains the occurrence of reduced neutrophil count in this group . The term “ethnic neutropenia” has been applied to persistently low neutrophil count in both Yemenite Jews and in populations of African ancestry , and the condition is clinically similar in these two populations . Persons with ethnic neutropenia have a reduced capacity to mobilize bone marrow neutrophil reserves in response to corticosteroids , despite normal cellularity and maturation of all cell lines in bone marrow aspirates [16]–[18] . Exercise-induced increments in neutrophil counts ( demargination ) are also smaller in persons with “ethnic neutropenia” than in healthy volunteers . Thus , the low neutrophil count is not the result of increased sequestration of neutrophils in the marginated granulocyte pools ( cells adherent to the endothelium of post-capillary venules ) [19] . The FY− allele of rs2814778 has a −46 T to C substitution ( non-coding strand ) in the Duffy Antigen Receptor for Chemokines ( DARC ) gene , which disrupts a binding site for the GATA1 erythroid transcription factor [20] . This substitution abolishes gene expression in erythrocytes but not other cell types , such as endothelial cells of the post-capillary venules [21] . The DARC gene product is a seven-transmembrane receptor that selectively binds “inflammatory” chemokines of both the CXC and CC families , including , for example , CXCL8 ( interleukin 8 ) and CCL5 ( RANTES ) , both of which are involved in neutrophil recruitment [22]–[25] . Unlike related chemokine receptors with signaling function , DARC lacks a G-protein binding motif . It is nevertheless capable of internalizing bound chemokine [21] , and it has been hypothesized to affect leukocyte recruitment to sites of inflammation through its role in trancytosis of chemokines through endothelial cells [24]–[26] . DARC on red blood cells might affect the number of circulating neutrophils through any of several mechanisms , for example , by modulating the concentrations of chemokines in vascular beds in the bone marrow [25] , or by acting as a chemokine “sink” to limit the stimulation and extravasation of circulating neutrophils , or through other mechanisms that are not yet understood . Interestingly , DARC-knockout mice , which lack DARC expression not only on red blood cells but also in other tissues , do not differ from wild-type mice in peripheral blood leukocyte levels [27] , but have a different phenotype , of increased bone mineral density [28] . When we tested whether an association with bone mineral density existed in African Americans in the Health ABC cohort , however , rs2814778 genotype was not significantly associated with either total ( n = 1 , 141 , P = 0 . 57 ) or femoral neck ( n = 1 , 141 , P = 0 . 43 ) bone mineral density . The expression of DARC on erythrocytes is known to modulate chemokine levels after endotoxin treatment [29]; thus the FY− allele could potentially have important effects in critically ill patients with sepsis . More subtle effects on innate immunity and inflammation could also exist through DARC modulation of chemokine concentrations in specific vascular and tissue microenviroments[25] . However , we were not able to directly demonstrate an effect of the FY− allele on health . When we carried out tests of association of FY− to a wide range of phenotypes in the Health ABC Study ( M . Nalls and T . Harris unpublished data ) , we did not find association to any phenotype . It is not surprising that health impacts of this variant are subtle , since the allele has risen to nearly 100% frequency in some African populations without being retarded in its rise by natural selection . This forms a sharp contrast with the HBB allele , which confers resistance to Plasmodium falciparum in heterozygous individuals ( analogous to FY− conferring resistance to Plasmodium vivax ) but causes sickle cell disease in homozygous form , and as a result has never risen to more than about twenty percent frequency in any population . Another recent study found that the FY− variant of DARC is associated to altered rates of HIV infection and disease progression , potentially suggesting a health effect of the low neutrophil count [9] . The authors found that there is a significantly slower rate of disease progression in FY−/− individuals infected by HIV-1 than in carriers of the FY+ allele . To explore this result , they performed functional studies showing that in vitro , HIV-1 attaches to erythrocytes via DARC and uses it as a means of transfer to target cells . They argued that such transfer might be impaired in FY−/− individuals , leading to slower disease progression . However , these authors also identified a second phenotype that is associated to the FY−/− genotype—a 40% higher rate of acquiring HIV-1 infection—that is difficult to attribute to the same mechanism , since the failure to express DARC on red blood cells might be expected to decrease access by HIV-1 to CD4-positive target cells . Potential explanations are that either low neutrophil counts or altered chemokine concentrations due the FY−/− genotype may have some role in modulating infection . These possibilities should be testable in the laboratory . It is also possible that differing frequencies of the FY− allele reflect stratification within the study population , and that differences in DARC expression are not actually involved in modulating susceptibility to infection or disease progression . An immediate consequence of our finding is that genotyping of rs2814778 ( or measuring Duffy antigen expression on red blood cells ) might be used as a diagnostic guide in clinical situations in African Americans , helping to set a baseline expected neutrophil count for patients , and to guide treatment . Reduced neutrophil count has been cited as a potential cause of treatment delay and of less intensive therapy for early-stage breast cancer in African American women , perhaps contributing to ethnic disparities in breast cancer survival [30] . It also may alter the course of cytotoxic therapy for inflammatory diseases such as rheumatoid arthritis and systemic lupus erythematosus , perhaps needlessly . Finally , “ethnic neutropenia” may contribute to a diminished leukocytic response to infection [31]–[34] , perhaps resulting in a lowered index of suspicion and delayed diagnosis of infection in selected patients . New studies are needed in each of these clinical settings to incorporate genotyping information from rs2814778 to help in the interpretation of neutrophil counts .
The 6 , 005 human subjects for this study were drawn from three observational cohorts in which large numbers of phenotypic measurements had been made: the Jackson Heart Study ( JHS ) [35] , the Health , Aging and Body Composition ( Health ABC ) Study [36] , and the Atherosclerosis Risk in Communities ( ARIC ) Study [37] . From each of these studies , we only included African Americans for whom a complete differential white blood cell count was available , including measurement of neutrophils , bands , lymphocytes , monocytes , eosinophils and basophils . An exception was the Health ABC study , which did not provide a measurement of bands . In addition , we restricted our analysis to individuals for whom we had an admixture scan that passed all our quality control filters; for whom we had a genotype at the Duffy polymorphism , rs2814778; and for whom we had information on gender , age , and body mass index ( BMI ) . All WBC and phenotypic measures were from baseline data collected at the time of enrollment in each study . The JHS cohort [35] consists of 5 , 302 self-identified African American men and women recruited between September 2000 and March 2004 from the three counties surrounding Jackson , MS . Unrelated persons aged 35–84 were enrolled from three sources: previous ARIC participants ( 31% of the total ) , random selectees from a commercial listing ( 17% ) , and members of an age- and sex-constrained volunteer sample ( 30% ) . The remaining participants , at least 21 years old , were members of a nested family cohort . A total of 3 , 945 JHS participants had the required phenotypic data and were successfully genotyped for a panel of admixture mapping markers . A subset of 2 , 871 was included in our analysis after randomly dropping samples of related individuals until there was only one individual included per family . Of these , 1 , 969 were “JHS unique” samples that were present only in JHS , and 902 were “JHS-ARIC overlap” samples , representing persons who had participated in both the JHS and ARIC studies . For the “JHS-ARIC overlap” samples , we averaged the baseline measurements for each individual at the time of their entry into each cohort , an average of 14 years apart . The Health ABC cohort [36] consists of 3 , 075 men and women aged 70–79 who were enrolled between April 1997 and June 1998 . All were Medicare beneficiaries living near Pittsburgh or Memphis and all reported having no difficulty performing basic physical activities . Of the 1 , 281 participants who identified themselves as African American , 658 had complete genotype and phenotype data and were included in the current study . Of the participants who identified themselves as European American , 1 , 331 were analyzed for the purpose of comparison with African Americans . The ARIC cohort [37] consists of 15 , 792 randomly-selected participants aged 45–64 who were recruited between November 1986 and December 1989 , in roughly equal numbers , from field centers in Jackson , MS , Minneapolis , MN , Forsyth County , NC , and Washington County , MD . The cohorts of the latter three field centers represent the ethnic mix of their communities . The Jackson-based cohort ( n = 3 , 728 ) was limited to self-identified African Americans , and comprised 87 . 4% of all African Americans in ARIC ( 1 , 626 Jackson-based participants were later enrolled in JHS ) . A total of 3 , 378 African American participants were included in the present analysis after applying all data filters . Of these , there were 2 , 476 “ARIC unique” participants who were present only in ARIC . There were 902 “JHS-ARIC overlap” samples as described above . A total of 10 , 062 European American ARIC participants were also genotyped and analyzed with respect to a single variant , rs2814778 . For all three studies , cells in EDTA-anticoagulated venous blood were counted using a Beckman-Coulter Counter ( Beckman Coulter , Inc . , Fullerton , CA ) , which combines measures of electrical conductivity and light scatter to distinguish cell lineages in suspensions of unstained leukocytes , yielding an overall WBC and relative proportions for each of six leukocyte subgroups: neutrophils , monocytes , lymphocytes , basophils , eosinophils , and “band forms” , expressed as a percentage of total WBC . Absolute counts were obtained by multiplying the differential count ( a percentage ) by total WBC . To create a phenotype for analysis , all counts were rank-ordered within one of the four groups of samples ( Health ABC , JHS only , ARIC only and JHS-ARIC overlap ) , and then assigned a percentile . An inverse normal transformation was used to translate this percentile into a normally distributed phenotype . Genotyping of African American samples on the admixture mapping panels was performed using the Illumina BeadLab platform [38] , which can analyze a custom panel of 1 , 536 SNPs . We have developed three consecutive versions of a custom admixture mapping panel , each providing incrementally better coverage of the genome than the previous version , and all yielding excellent coverage . A total of 1 , 119 samples were genotyped in the “Phase 2” panel [39] ( Health ABC samples and 16% of JHS samples ) and 4 , 886 samples were genotyped in the “Phase 3” panel [6] ( ARIC samples and 84% of the JHS samples ) . All panels include the rs2814778 polymorphism in the DARC gene . The genotyping of the Health ABC and JHS samples was carried out at the Broad Institute of Harvard and MIT in Cambridge as previously described [6] . Genotyping of the ARIC samples was carried out at the Center for Inherited Disease Research ( CIDR ) in Baltimore , MD . Genotyping of the rs2814778 polymorphism in 10 , 062 ARIC European American samples and 1 , 339 Health ABC European American samples was done using the ABI TaqMan technology [40] . We used built-in data quality checks in the ANCESTRYMAP software [11] , [39] , [41] , [42] to remove SNPs that were not appropriately intermediate in frequency in African Americans compared to the West African or European American ancestral populations , or that had evidence of being in linkage disequilibrium ( LD ) with each other in these ancestral populations [11] . After this filtering , the JHS samples had 1 , 265–1 , 532 SNPs available for analysis , the Health ABC study samples had 1 , 128–1 , 385 SNPs , and the ARIC study samples had 1 , 277–1 , 529 SNPs . To refine the peak of admixture association , we used the Sequenom iPLEX platform [43] to genotype all JHS samples more densely in the region of highest interest on chromosome 1 ( 153 . 5–157 Mb in Build 35 of the reference sequence ) . After filtering to remove SNPs in LD in the ancestral populations or with poor genotyping performance , we had data from 9 markers across this region in JHS ( rs2768744 , rs2309879 , rs7528684 , rs1587043 , rs857859 , rs2814778 , rs11265198 , rs2494493 and rs11265352 ) , compared with 2 in the other studies ( rs2768744 and rs2814778 ) . The “C” allele of rs2814778 ( also “FY−” ) is known to be almost completely correlated to West African ancestry at the chromosome 1q23 . 2 locus . We therefore used this allele as a marker to study the epidemiological association of African ancestry to various WBC phenotypes ( Figure 1 and Tables 1 and 2 ) . In addition to using FY− as a surrogate for ancestry , in the final analyses of this study ( Figure 3 and Table 4 ) , we took advantage of the fact that the correlation between this allele and African ancestry , though >99% , is not perfect . Thus , we could test whether the allele is more predictive of neutrophil count than is African ancestry itself . To test for association of the chromosome 1 locus to counts of leukocytes other than neutrophils , we carried out a regression analysis between absolute counts of neutrophils and absolute counts of each of the other white blood cell types . This generated a residual value for each cell type after correcting for the effect of neutrophil count . We then carried out 2-sided tests for association of each of these residuals to carrier status for European ancestry at the chromosome 1 locus ( defined as having at least one FY+ allele at rs2814778 ) , using Z-scores to indicate the difference , in standard deviations , in the population means between groups of samples . These Z scores can be approximately translated into a P-value by referring to the corresponding percentile in the cumulative normal distribution function . The ANCESTRYMAP software [11] was used to better define the region of chromosome 1 associated to neutrophil count . Since the software is optimized for dichotomous traits , we divided the 6 , 005 samples into 12 strata ( with 399–593 samples each ) based on their normally transformed neutrophil counts ( Table 3 ) . For each stratum , we identified a risk model ( increased probability of observing a sample with 1 or 2 copies of European ancestry compared with the expectation from the genome-wide average ) that optimized the peak LOD score at the locus . This ranged from a 25-fold decrease in the relative probability of European ancestry for individuals with neutrophil counts <−1 . 5 standard deviations below the mean , to a 6 . 5-fold increase for those with neutrophil counts >1 . 5 standard deviations above the mean . For each stratum , the risk model and LOD score at the chromosome 1 locus are given in Table 3 . To use these strata to define an admixture peak , we carried out an admixture scan for each group separately , and then summed the LOD scores at loci interpolated every tenth of a centimorgan . The peak LOD score was 363 . 1 as shown in Table 3 . The 99% credible interval of 155 . 957–156 . 408 Mb in Build 35 of the human genome reference sequence was determined by the region where the score was within log10 ( e6 . 63/2 ) = 1 . 44 of its maximum ( Figure 2 ) . This is calculated from a likelihood ratio test , using the fact that a χ2 statistic with 1 degree of freedom of 6 . 63 corresponds to P = 0 . 01 . To assess whether genotype at rs2814778 is more predictive of neutrophil count than ancestry , we calculated two numbers for each DNA sample . First , we recorded whether the individual was a carrier of the FY+ “functional” allele ( = 1 ) at rs2814778 , or was homozygous for the FY− “null” allele ( = 0 ) ( homozygosity for the “null” allele abolishes expression of the Duffy antigen on red blood cells but not on other cell types ) . Second , we used the ANCESTRYMAP software [11] to estimate the probability that in the region spanning this SNP , at least one of the individual's chromosomes was of European ancestral origin . Importantly , we included the genotype of rs2814778 in the ancestry estimation . Thus , if neutrophil count were better correlated to rs2814778 genotype than to an ancestry estimate that included information from both rs2814778 and closely neighboring SNPs , it would indicate that the neighboring SNPs did not add relevant information , and that rs2814778 is either the causal variant or is in strong LD with the causal variant . To determine whether genotype at rs2814778 or ancestry at the chromosome 1 locus was the better predictor of neutrophil status , we carried out three regressions: To test for evidence of an association of the genotype at rs2814778 above and beyond ancestry , we subtracted the χ2 statistics of ( c ) - ( b ) . To test for evidence of association to ancestry above and beyond SNP genotype , we subtracted the χ2 statistics of ( c ) - ( a ) . To test whether there were SNPs apart from rs2814778 that contributed evidence for association at the chromosome 1 locus , we densely genotyped a subset of especially informative samples . To select cases and controls for fine-mapping , we identified individuals from JHS for whom we were >99% confident of African ancestry on both chromosomes at the admixture peak . By limiting cases and controls to persons with more confident estimates of entirely African local ancestry , it was easier to detect whether any signal of association was significant above and beyond the admixture association . For this analysis , the estimate of ancestry at the admixture peak using ANCESTRYMAP [11] excluded SNPs within the peak . Among individuals who had >99% confidence of entirely African ancestry at the locus , we identified 696 individuals who had a “low” neutrophil count , defined based on visual inspection of the distribution as an absolute count of <2 , 100/mm3 ( and corresponding to <0 . 7 s . d . below the population mean for the entire JHS sample ) . We also identified 77 individuals who had a “high” neutrophil count , defined as an absolute count of 5 , 100/mm3–9 , 100/mm3 ( 1 . 3–2 . 8 s . d . above the population mean ) . For genotyping , we selected a random subset of individuals with “low” neutrophil count , and all of the individuals with “high” neutrophil count . We successfully genotyped 148 subjects with low and 74 subjects with high neutrophil count that we could use in this analysis . To identify additional SNPs across the admixture peak that might be associated with neutrophil levels , we used the Tagger software [12] to choose a panel of SNPs from the International Haplotype Map database [10] that captured all SNPs of >5% minor allele frequency in West African samples with a correlation of r2>0 . 8 . Forcing these SNPs into the analysis , we chose additional SNPs across the region until we had similarly captured all SNPs in European Americans with >5% minor allele frequency . All SNPs identified in this way were selected for genotyping on the Sequenom iPLEX platform [43] . Cases and controls were successfully genotyped at a densely spaced panel of 193 tag SNPs across the ∼450 kilobase admixture peak . Each SNP was tested using a χ2 statistic assuming an additive effect on neutrophil count for each additional copy of the allele . We did not test a dominant or recessive model because none of the genotyped SNPs ( apart from rs2814778 ) had a frequency differentiation across populations consistent with that SNP explaining the admixture signal ( and thus being the main effect SNP; see above ) . We chose an additive model to search for variants that might modulate the neutrophil count beyond the main effect because it is known that for substantial minor allele frequencies this provides reasonable power to detect an association , whether the true underlying effect is dominant , additive , or recessive [44] . While we found that rs2814778 was more predictive of neutrophil count than ancestry , we were concerned that the variant might not itself be causal , but instead only in LD with the causal variant . To search for additional candidate SNPs in the region that are highly different in frequency between Africans and Europeans , we examined shotgun genome sequence data derived from public databases , from 6 individuals who were known to have West African ancestry across the region ( NA18507 , NA18517 , NA19129 , NA19240 , NA17109 and NA17119 ) , and 5 individuals who were known to have European ancestry across the region ( NA12156 , NA12878 , NA07340 , HuAA and HuBB ) . These sequences include 4 West Africans and 2 Europeans examined as part of a fosmid end-sequencing project [45] , 2 European Americans sequenced as part of the Celera Genomics human genome sequencing project [46] , [47] , 1 European American sequenced for the purpose of SNP discovery [48] , [49] , and 2 African Americans also sequenced for SNP discovery [48] , [49] , who we determined had entirely African ancestry at the locus by using the ANCESTRYMAP software [11] and by unpublished methods ( Simon Myers , Alkes Price and Alon Keinan , personal communication ) . For each individual , we only analyzed nucleotides for which we had a high quality base call at the locus ( Neighborhood Quality Score of ≥40 ) [13] . At sites where we had more than 1× coverage , we randomly sampled one sequence to represent the individual . To determine the ancestry of the human genome reference sequence across the admixture peak , we first observed that it was spanned by a mosaic of 5 fully sequenced Bacterial Artificial Chromosomes ( BACs ) , each representing a clone of 86 , 000–196 , 000 base pairs ( http://genome . ucsc . edu ) . The problem of determining ancestry of the human reference sequence across the region thus amounts to determining the ancestry of each of the clones separately . To do this , we obtained the allele of the human genome reference sequence at each of 284 HapMap SNPs across the region that had been genotyped in both West Africans and European Americans . For each window of 8 consecutive SNPs in HapMap , we calculated the likelihood that the human reference sequence was of African or European ancestry . To determine this likelihood empirically , we compared the findings to 120 phased European chromosomes and 120 phased African chromosomes from the HapMap database , counting the number of matches to the human reference sequence in each population over that window . We conservatively added 1 to the counts of the tested haplotype for each population to preclude an estimate of zero probability for either population . Under the assumption that only European and African ancestry were possible , the results showed with confidence that 3 of the clones were of European ancestry and 2 were of African ancestry ( Figure 4 ) . | Many African Americans have white blood cell counts ( WBC ) that are persistently below the normal range for people of European descent , a condition called “benign ethnic neutropenia . ” Because most African Americans have both African and European ancestors , selected genetic variants can be analyzed to assign probable African or European origin to each region of each such person's chromosomes . Previously , we found a region on chromosome 1 where increased local African ancestry completely accounted for differences in WBC between African and European Americans , suggesting the presence of an African-derived variant causing low WBC . Here , we show that low neutrophil count is predominantly responsible for low WBC; that a dominant , European-derived allele contributes to high neutrophil count; and that the frequency of this allele differs in Africans and Europeans by >91% . Across the chromosome 1 locus , only the well-characterized “Duffy” polymorphism was this differentiated . Neutrophil count was more strongly associated to the Duffy variant than to ancestry , suggesting that the variant itself causes benign ethnic neutropenia . The African , or “null , ” form of this variant abolishes expression of the “Duffy Antigen Receptor for Chemokines” on red blood cells , perhaps altering the concentrations and distribution of chemokines that regulate neutrophil production or migration . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/gene",
"discovery",
"genetics",
"and",
"genomics",
"genetics",
"and",
"genomics/population",
"genetics"
]
| 2009 | Reduced Neutrophil Count in People of African Descent Is Due To a Regulatory Variant in the Duffy Antigen Receptor for Chemokines Gene |
Axonal growth and guidance rely on correct growth cone responses to guidance cues . Unlike the signaling cascades that link axonal growth to cytoskeletal dynamics , little is known about the crosstalk mechanisms between guidance and membrane dynamics and turnover . Recent studies indicate that whereas axonal attraction requires exocytosis , chemorepulsion relies on endocytosis . Indeed , our own studies have shown that Netrin-1/Deleted in Colorectal Cancer ( DCC ) signaling triggers exocytosis through the SNARE Syntaxin-1 ( STX1 ) . However , limited in vivo evidence is available about the role of SNARE proteins in axonal guidance . To address this issue , here we systematically deleted SNARE genes in three species . We show that loss-of-function of STX1 results in pre- and post-commissural axonal guidance defects in the midline of fly , chick , and mouse embryos . Inactivation of VAMP2 , Ti-VAMP , and SNAP25 led to additional abnormalities in axonal guidance . We also confirmed that STX1 loss-of-function results in reduced sensitivity of commissural axons to Slit-2 and Netrin-1 . Finally , genetic interaction studies in Drosophila show that STX1 interacts with both the Netrin-1/DCC and Robo/Slit pathways . Our data provide evidence of an evolutionarily conserved role of STX1 and SNARE proteins in midline axonal guidance in vivo , by regulating both pre- and post-commissural guidance mechanisms .
Axonal growth and guidance are responsible for the correct formation of neural circuits . These processes rely on the tightly regulated response of the growth cone to both diffusible and membrane-bound guidance cues . In response to such cues , several intracellular signaling cascades are activated within the growth cone , leading to directional growth . For instance , the chemoattractant Netrin-1 binds to the receptor Deleted in Colorectal Cancer ( DCC ) at growth cones , activating several kinases and small GTPases , cyclic nucleotides , and calcium cascades , as well as cytoskeletal rearrangements [1–8] . In contrast , few reports have addressed the cross-talk mechanisms between axonal guidance and membrane dynamics and turnover , other than the fact that growth cones are filled by vesicles and express most SNARE ( Soluble NSF Attachment Protein REceptor ) and exocyst proteins [9–13] . A growing number of reports using in vitro approaches indicate that axon guidance mechanisms require the participation of SNARE-mediated exocytosis for chemoattraction and endocytosis for repulsion [14–18] . Thus , it has been demonstrated that the vSNARE ( vesicular SNARE ) VAMP2 is required for L1-mediated chemoattraction [19] and for Sema3A-induced chemorepulsion [17] , that the vSNARE Ti-VAMP and the tSNARE ( target SNARE ) SNAP25 are necessary for neurite outgrowth [20–22] , and that Syntaxin-1 ( STX1 ) is required for Netrin-1-mediated attraction of axons and migrating neurons [15 , 16] . However , the participation of these proteins in neural circuit formation in vivo is still controversial . For instance , mice deficient for the SNAP25 and VAMP2 proteins show virtually no neural circuitry defects but do present a severe alteration of evoked synaptic activity [23–25] . Ti-VAMP-deficient mice display behavioral defects but no alterations in gross brain morphology [26] . STX1A knock-out ( KO ) mice show only mild cognitive defects and a normal brain structure [27] and axonal defects have not been described in STX1B KO [28] . In a previous study we showed that STX1A is required for the navigation of dorsal commissural in the chick spinal cord [16] . Syntaxin-1 loss-of-function in Drosophila and chick embryos results in motor axonal defects [29] . Drosophila melanogaster displays neural expression of a synaptobrevin ( VAMP ) gene , namely n-synaptobrevin ( n-syb ) [30] and a SNAP25 homolog , Snap25 [31] . Mutations in these components of the core SNARE complex give rise to neurotransmitter release phenotypes [32] . A single D . melanogaster STX1 homolog , Syntaxin1A ( Syx1A ) , has been identified that shows strong homology to rodent STX1A [33–35] . Mutations in this gene are homozygous lethal , with severe alleles resulting in early embryonic death . In D . melanogaster , loss of Syx1A abolishes synaptic transmission [33 , 36] , and other secretion phenotypes , such as soft cuticle and undigested yolk , have also been reported [33] . In addition , Syx1A is involved in cell membrane formation during cellularization [37] and in the condensation of the embryonic CNS [33] . In addition , Syx1A has been reported to affect the properties of neuronal membranes and influence membrane dynamics throughout development [38] . However , in D . melanogaster , in contrast to vertebrates , Syx1A has yet to be directly implicated in axonal guidance . Here we systematically inactivated SNARE genes in three species , Drosophila melanogaster , chick , and mouse , to determine the role of SNARE proteins in CNS midline axonal guidance in vivo . We show the involvement of this protein complex , in particular STX1 , in D . melanogaster and chicken axonal guidance , reporting aberrant phenotypes . Furthermore , we provide the first description of abnormal midline axonal defects in mouse embryos double mutant for STX1A and STX1B . Our results point to an evolutionarily conserved mechanism of SNARE complex proteins in midline axonal guidance .
To determine whether Syx1A affects axon guidance at the midline of D . melanogaster embryos , we started by analyzing the CNS phenotypes of Syx1A mutant embryos . For this purpose , we used zygotic mutants , since Syx1A has a maternal contribution and its function is required for proper cellularization [37] . In D . melanogaster , CNS axons of wild-type ( wt ) embryos are found in a stereotypic ladder-like arrangement . Within each neuromere , two commissures link the two halves of the nervous system . Individual neuromeres are connected by axons running in discrete fascicles in the lateral connectives . An antiserum that specifically recognizes D . melanogaster Syx1A revealed that , like Frazzled ( Fra , the D . melanogaster DCC homolog ) [39] , Syx1A is expressed in developing axons in the embryo from stage 13 . At later stages , this isoform is expressed at high levels on commissural and longitudinal axons in the developing CNS ( Fig 1A ) . Embryos of the null Syx1AΔ229 genotype [33] displayed no detectable Syx1A protein in the ventral nerve cord ( VNC ) when stained with the same antibody ( Fig 1B ) . To analyze axonal midline phenotypes , we stained embryos with HRP , which marks all axons in the CNS , and anti-Fasciclin II antibodies ( anti-FasII ) , which label lateral fascicles ( Fig 1C ) . At stage 15 or later , FasII identifies three major axonal tracts , which are visible as parallel straight lines: the medial ( FasII-m or 1st ) , intermediate ( FasII-i or 2nd ) and lateral ( FasII-l or 3rd ) fascicle . In wt conditions , these FasII-positive axons do not cross the midline . Examination of the commissural and longitudinal pathways labelled by HRP and anti-FasII in the VNC of Syx1A mutants revealed specific defects in diverse axonal pathways . From stage 14 onwards , by analyzing FasII-positive axons , we detected guidance defects in the VNC of 50% of the embryos ( n = 38 ) . These defects included aberrant midline crossing , as well as abnormal arrangement of ipsilateral fascicles ( Fig 1D , 1E and 1F ) . Of these embryos , 18% displayed strong defects ( Fig 1D and 1F ) and 32% weak defects ( Fig 1E and 1F , see Materials and Methods for quantification methods and definition of weak and strong phenotypes ) . FasII-positive fibers never crossed the midline in wt conditions . In contrast , Syx1A embryos displayed clear abnormal crossing of FasII-positive axons , thereby suggesting that longitudinal fibers aberrantly cross the midline in Syx1A mutants ( Fig 1G ) . When HRP staining ( a pan-axonal marker ) was used , control embryos presented regularly spaced commissures ( Fig 1C and 1H ) . However , embryos with the strong Syx1A phenotype showed collapsed commissures with no clear separation between anterior and posterior ones ( Fig 1D and 1H ) . We analyzed and quantified commissural phenotypes in the VNC of Syx1A embryos displaying phenotypes ( weak and strong phenotypes , 50% of embryos ) . Collapsed commissures ( 98% Fig 1H , arrows , and 1I ) and “fuzzy” commissures were detected in many segments ( 36% , Fig 1H , asterisk , and 1I ) , as well as thinning of longitudinal fascicles between segments ( 57% , Fig 1H , arrowhead , and 1I ) . On average , Syx1A mutant embryos displayed more than one defect in VNC axonal guidance per embryo ( Fig 1G ) . Among these defects , the strongest were axonal midline crosses ( Fig 1D , short arrow , 13% ) , commissural thinning ( Fig 1H and 1I , 55% ) , and thinner longitudinal fascicles ( Fig 1H and 1I , 32% ) , which were not detected in controls . The most penetrant phenotypes were fasciculation defects ( Fig 1E , long arrow , 58% ) , but Syx1A mutant embryos also showed fascicle collapse ( Fig 1D and 1E , arrowhead , 42% ) . Variability among these axonal phenotypes between individuals possibly reflects the differential contribution of the maternally deposited gene product . These results show that loss of Syx1A function induces axonal guidance defects in both commissural and longitudinal axons at embryonic stages of fly CNS development . We next used the same approach to study whether genetic loss of additional components of the SNARE core complex ( SNAP25 , VAMP2 and Ti-VAMP/VAMP7 ) also alters the commissural and longitudinal pathways in D . melanogaster . To do so , we examined mutants for nSyb ( human VAMP2 ortholog ) [40] and Snap25 [41] , and analyzed the D . melanogaster ortholog of Ti-VAMP/VAMP7 ( Vamp7 source:Flybase ) . Overall , the nSyb , Snap-25 , and Vamp7 phenotypes were weaker than the Syx1A one ( Fig 2A–2D ) . When analyzing FasII positive axons for fasciculation defects , the strongest axonal guidance phenotypes were found in Snap25 mutants , in which 74% embryos displayed fascicle collapse , defasciculation , or both ( Fig 2B and 2F , n = 35 ) . 57% of nSyb mutants showed fascicle collapse , defasciculation , or both ( Fig 2C and 2F , n = 41 ) . In addition , a small percentage of nSyb embryos ( 4% , n = 41 ) showed axonal midline crosses of FasII positive axons ( Fig 2C , arrow ) . Vamp7 mutants exhibited the weakest phenotypes ( Fig 2D and 2E n = 31 ) . We could detect stronger phenotypes when all axons were visualized with BP102 antibody . Again , the strongest phenotypes were detected in SNAP25 mutant embryos ( Fig 2H and 2K ) where 6 . 9% of embryonic segments showed fuzzy commissures and 27 . 5% showed thinning of longitudinal fibers ( n = 80 segments ) . nSyb mutants displayed intermediate phenotypes , fuzzy commissures in only 3 . 1% of segments and thinning of longitudinal fibers in 10 . 6% ( n = 80 ) ( Fig 2I and 2K ) . In Vamp7 embryos we could only detect thinning of commissures ( in 5% of embryonic segments , n = 80 ) ( Fig 2J and 2K ) . These results indicate that Vamp7 , nSyb , and Snap25 can also influence D . melanogaster axonal guidance at the midline , but to a lesser extent than Syx1A . Next , we used in ovo electroporation of double-stranded RNAs derived from STX1A , SNAP25 , VAMP2 , and Ti-VAMP to study the role of these genes in dI1 commissural axon guidance in the chicken spinal cord . In untreated controls and in EGFP-expressing control embryos , most commissural axons followed a stereotypic trajectory ( Fig 3A and 3B ) . The vast majority of the dI1 axons crossed the floor plate and turned rostrally along the contralateral floor-plate border . See Methods for details on the quantification method . In contrast , we found that the down-regulation of all SNARE-complex proteins ( STX1A , SNAP25 , VAMP2 , and Ti-VAMP ) generated defects in commissural axon guidance , as axons either failed to enter or to cross the floor plate , or failed to turn into the longitudinal axis along the contralateral floor-plate border ( Fig 3C–3G ) . After silencing STX1A aberrant axon pathfinding was found at 36% of the DiI injection sites per embryo ( Fig 3C and 3G ) . We found overall similar percentages of injection sites with aberrant phenotypes when SNAP25 , VAMP2 or Ti-VAMP were down-regulated ( Fig 3G ) . When electroporated at embryonic day 3 ( E3/HH17-18 ) most axons reached the floor plate and entered the midline area in all groups ( Fig 3H ) . A detailed analysis of the aberrant phenotypes indicated a significant increase in floor-plate stalling in embryos where STX1A , VAMP2 , or Ti-VAMP was silenced ( Fig 3I ) . Similarly , turning of post-crossing axons was affected in all experimental groups ( Fig 3J ) . In a separate series of experiments , we compared electroporation of dsSTX1A at HH13/14 ( E2 ) with electroporation at HH17/18 ( E3 ) ( Fig 3K ) . These experiments revealed an effect of timing of gene silencing on the severity of the phenotype . Electroporation at E3 , when axons are starting to grow in the dorsal spinal cord , resulted in normal axon guidance at 56 . 4% of all DiI injection sites per embryo . In contrast , electroporation of neural precursors at E2 resulted in normal axon navigation only at 23 . 8% of all DiI injection sites . Importantly , failure of some axons to enter the floor plate was observed in 3/8 embryos electroporated early but was never observed in the embryos electroporated late . These data support the notion that the silencing of SNARE proteins in the chicken spinal cord leads to various commissural axon guidance defects . To study the involvement of STX1 in commissural axon guidance in mammals , we generated double KO mice for the two STX1 isoforms , STX1A and STX1B . Double KO embryos died just after birth and displayed strong motor abnormalities , thereby suggesting severe alterations in nervous system organization . To address axonal phenotypes in the midline , we examined the commissural pathway in the spinal cord in E12 embryos [42 , 43] . In wt embryos , commissural axons stained with TAG-1 antibodies were organized as a narrow axonal bundle extending from the dorsal spinal cord towards the floor plate without invading the motorneuron area ( Fig 4A and 4B ) . In contrast , in STX1A/B ( -/- ) mice , TAG-1-positive commissural axons were still directed towards the ventrally located floor plate , although their organization differed from that of controls . Instead of forming a narrow bundle , axons in STX1A/B ( -/- ) mice were clearly defasciculated , invading the entire mantle zone . Individual fibers or bundles invading lateral motorneuron areas were frequently observed ( Fig 4C–4F ) . However , axons appeared to reach the floor plate . To support these observations , we quantified the width of the commissural pathway at three dorso-ventral levels ( Fig 4C–4G ) . At each one , we found a significant difference in the width of commissural axon bundles between wt and double STX1 KO mice . To confirm these findings , we stained spinal cord sections with Robo3 antibodies , a marker of pre-crossing and post-crossing commissural fibers . In wt embryos , Robo3-stained fibers formed a tight fascicle directed towards the floor plate . Moreover , post-commissural fibers in the ventral spinal cord were heavily stained ( S2A and S2B Fig ) . In contrast , commissural fibers in STX1A/B ( -/- ) embryos exhibited wider ipsilateral fascicles , often invading the lateral ( motorneuron ) domains in the ventral spinal cord . Many ipsilateral commissural axons in mutant embryos were tipped with growth cones indicating that they failed to reach the floor plate . Consistent with this observation , the bundles of post-crossing commissural axons ( located in the ventral spinal cord ) were markedly decreased in STX1A/B null-mutant embryos ( S2C and S2D Fig ) . Taken together , our observations indicate that the lack of STX1 results in aberrant commissural axon growth on both ipsilateral and contralateral sides . To confirm the above observations , we also examined commissural axon navigation in open-book preparations of E12 spinal cords ( Fig 5 ) . Comparable to our observation in the chicken spinal cord , we found aberrant navigation of dI1 axons at the floor plate . In the double STX1 KO embryos , almost no axons were found to cross the midline and to turn properly along the contralateral floor-plate border ( Fig 5D and 5E ) . Axons mainly failed to reach the contralateral border of the floor plate . In embryos lacking STX1A but expressing STX1B from one or two wt alleles , axon guidance was still compromised , exhibiting intermediate phenotypes ( Fig 5C , 5F and 5G ) . A quantitative analysis showed a pronounced decrease in normal axonal trajectories in mutant compared to wt embryos ( Fig 5F ) . In wild-type mice we found normal axon trajectories at 75 . 6±6 . 3% of the DiI injection sites per embryo . In contrast , in mice lacking STX1A but expressing one or two wild-type allele ( s ) of STX1B normal trajectories were only seen at 10 . 2±5 . 4% or 21 . 7±12 . 3% of the DiI injection sites , respectively . In double KO mutants axon navigation was affected even more strongly , as normal trajectories were only seen at 1% of the injection sites ( Fig 5F and 5G ) . A detailed analysis of the different guidance defects revealed a problem in floor-plate stalling in all mutants in addition to a failure to turn rostrally along the contralateral floor-plate border . Failure to enter the floor plate was only found in mutants but never in wild-type mice ( Fig 5G ) . Embryos having at least one wt allele of either STX1A or STX1B exhibited weaker axonal defects than the double KO mice ( Fig 5G ) . To confirm the expression of SNARE proteins in spinal cord commissural axons we performed immunocytochemical analyses in neuronal cultures . Dissociated mouse commissural neurons were identified with two antibody markers: DCC and Robo3 antibodies . Cultures were co-immunostained for different SNARE proteins including STX1A , VAMP2 , SNAP25 and Ti-VAMP . Confocal images revealed that commissural growth cones , identified by the expression of DCC and Robo3 , co-expressed all the studied SNARE proteins ( Fig 6 ) . We also observed variable degrees of colocalization between DCC/Robo3 receptors and SNARE proteins . These results indicate that embryonic commissural axons express the SNARE proteins analyzed in the present study . We next addressed whether STX1A is required for Slit-2 and Netrin-1 actions . Tissue explants of chick dorsal commissural neurons dissected from control embryos and grown on laminin are known to extend neurites readily in the absence of Slit-2 ( Fig 7A ) . In the presence of Slit-2 , neurite length was strongly reduced ( Fig 7B ) . In contrast , neurite growth from explants taken from embryos electroporated with dsRNA derived from STX1A did not differ in the absence or presence of Slit-2 ( Fig 7C , 7D and 7I ) , indicating that the absence of STX1A results in a markedly reduced Slit-2 response of chick commissural axons . We have previously shown that the blockade of STX1 with Botulinum Toxin C1 reduces Netrin-1 induced chemoattraction in mouse spinal cord explants [16] . To confirm this finding , we co-cultured in collagen gels dorsal spinal cord explants from STX1 KO embryos with Netrin-1 expressing cell aggregates . In comparison with explants cultured with control cells ( exhibiting radial growth , Fig 7E ) , wild-type explants confronted to aggregates of Netrin-1 expressing cells showed strong chemoattraction ( Fig 7F and 7J ) . In contrast , Netrin-1 induced chemoattraction was absent in spinal cord explants derived from double STX1A/B KO embryos ( Fig 7G , 7H and 7J ) . Our previous studies have shown that exposition to Netrin-1 increases DCC surface expression in growth cones and that such increase is not diminished after blockade of STX1 [15 , 16] . Here we performed experiments to determine whether STX1 regulates Robo3 surface expression in the presence of Netrin-1 . Mouse embryonic commissural neurons were cultured and stained for the differential immunolabeling of surface and intracellular Robo3 protein pools ( S3 Fig ) . The data show that in control commissural growth cones , Robo3 trafficking and distribution is not substantially altered by Netrin-1 incubation . In contrast , STX1A/B knock-out growth cones incubated with Netrin-1 exhibited a decrease in surface Robo3 signals ( S3D and S3E Fig ) . These findings suggest that the inactivation of STX1 alters Robo3 trafficking and/or surface expression , probably by increasing Robo3 internalization . During our analysis of midline guidance defects in Syx1A mutant embryos , we detected that the arrangement of the longitudinal fascicles was shifted towards the CNS midline compared to controls ( Fig 8A and 8B; n = 20 ) . In order to clarify whether the FasII–positive axonal fascicles were closer to the midline in Syx1A mutants , we quantified these distances at embryonic stage 16 and compared them to wt , frazzled ( fra ) and robo2 mutants ( Fig 8C–8E ) . Overall , Syx1A and robo2 mutations induce a shift of FasII-positive fibres ( both FasII-m and FasII-i ) towards the midline . Regarding both FasII-m and FasII-I , the opposite occurs in fra mutant embryos , thereby suggesting that Syx1A interferes with a midline repression pathway ( Fig 8A–8E ) [44] . In order to address whether Syx1A is involved in the Slit/Robo pathway , we generated double mutants for Syx1A and robo2 and examined the positioning of the longitudinal tracts in robo2;Syx1A embryos . As shown in Fig 8F–8J , the longitudinal fascicle distance to the midline decreased in the double mutant embryos ( Fig 8I and 8P ) , where the strongest phenotypes showed a complete collapse of the tracts in the midline ( Fig 8I , asterisk ) , as reported for slit ( sli ) mutants [45] . This result indicates that Syx1A and robo2 interact genetically . In molecular terms , our above results suggest that Syx1A is involved in axonal repulsion in the midline . In vertebrates , it has been reported that Netrin1/DCC axonal guidance is coupled to exocytosis through STX1 [15] . Therefore , we examined whether differences in Syx1A levels could affect the Drosophila fra axonal guidance phenotypes . We found that the fra mutant phenotype was significantly aggravated by removing one copy of Syx1A ( Fig 8K–8O ) , thereby suggesting that Syx1A levels severely interfere with the Netrin/Fra guidance pathway .
Here we used loss-of-function models in three species ( fly , chick and mouse embryos ) to examine the role of SNARE proteins in midline crossing of commissural axons . We report aberrant axonal phenotypes in the D . melanogaster midline and in the chicken spinal cord . Furthermore , we generated double mutant STX1A/B mice that show abnormal commissural phenotypes in the murine spinal cord . These findings indicate an evolutionarily conserved role of SNARE complex proteins in midline axonal guidance . The guidance of commissural axons at the midline of the spinal cord is a complex process regulated by many molecules . For instance , the interaction between Netrin-1 and DCC attracts pre-crossing axons toward the floor plate [42 , 43] . Axon growth towards the floor plate is also altered in mice lacking VEGF/Flk1 signaling [46] . Midline crossing was shown to depend on attractive ( Axonin-1/TAG1 with NrCAM; [47] ) and repulsive interactions ( Slit/Robo; [48]; Sema3B/NrCAM/PlexinA1/Neuropilin; [49] [50] [51 , 52] ) . The rostral turn of post-crossing commissural axons was shown to depend on morphogen gradients formed by Wnts [53 , 54] and Shh [55 , 56] , although the latter is also involved in attraction of commissural axons toward the ventral midline in parallel to Netrin-1 [57 , 58] . In addition , an interaction between axonal Semaphorin6B and its ligand PlexinA2 [59] and interactions between the SynCAMs [60 , 61] are also involved in the turning of post-crossing commissural axons into the longitudinal axis . Our results in D . melanogaster indicate that the SNARE complex is involved in midline axonal guidance . We show that Ti-Vamp/Vamp7 , nSyb , and Snap25 influence axonal guidance at the midline , but to a lesser extent than Syx1A . In general , these phenotypes may be incomplete Drosophila KOs , because in this fly there is a maternal contribution for all these components . Additionally , it is known that there is a high level of redundancy in D . melanogaster . In this regard , it has been reported that Syb can replace nSyb and Snap24 can replace Snap25 [40 , 41] . Regarding the Syx1A phenotypes , we can conclude that Syx1A loss of function induces commissural and ipsilateral axonal phenotypes . Interestingly , some of the phenotypes observed in Syx1A mutants resemble the loss of function phenotypes of the Robo/Slit pathway [62 , 63] . We detected midline crosses of FasII-positive axons and a strong genetic interaction with robo2 . These findings suggest that Syx1A is involved in repulsive midline guidance in Drosophila . In addition , most of the ipsilateral phenotypes detected are also common in frazzled ( fra , the D . melanogaster homolog of DCC ) , and fra phenotypes are aggravated by decreased Syx1A levels . Furthermore , similar ipsilateral phenotypes to Syx1A are observed in netrin mutant ventral nerve cords [44] , as well as in embryos mutant for Heparan Sulfate Proteoglycans ( HSPGs ) [64] and Hh [65] , thereby suggesting that Syx1A may participate in various guidance pathways in D . melanogaster not just in axonal guidance but also in fasciculation/defasciculation events . Recent studies suggest that whereas axonal attraction requires exocytosis , chemorepulsion relies on endocytosis [14 , 18 , 66 , 67] . In addition , SNARE proteins have been reported to be involved in both exo- and endocytosis [68] . Our genetic interaction experiments support the notion that Syx1A participates in both Netrin/Fra attraction and Slit/Robo repulsion . The data obtained in chicken reinforce the idea that the silencing of SNARE proteins induces an overall increase in various commissural guidance defects , with the silencing of STX1A leading to defects in all the guidance steps analyzed . Silencing STX1A at E2 , on day before neurons start to extend axons rather than E3 , just before they start to extend axons , also resulted in axons failing to enter the midline area . This phenotype resembled previously reported findings on the role of STX1A in DCC-mediated guidance of pre-crossing commissural axons towards the midline [16] . Based on these results , we conclude that SNARE proteins make a crucial contribution to the navigation of chick commissural axons at the floor plate . SNARE proteins are involved in both attractive [16] and repulsive ( this study ) decision-making steps in axon guidance , as STX1A , VAMP2 and Ti-VAMP silencing leads to a substantial increase in pre-crossing and post-crossing phenotypes . These phenotypes are in agreement with those reported in previous studies [15 , 16] , confirming the involvement of STX1 in the regulation of chemoattractive guidance pathways for commissural axons . The phenotypes seen in the chicken spinal cord resemble those observed after silencing Calsyntenin-1 and RabGDI in dI1 neurons [69] . In the absence of Calsyntenin-1 and RabGDI , Robo1 is not transported to the growth cone surface , resulting in axonal stalling in the floor plate . Furthermore , silencing Calsyntenin-1 , but not RabGDI , prevents the expression of the Wnt receptor Frizzled-3 on the growth cone surface , leading to failure of post-crossing axons to turn rostrally in response to the Wnt gradient [54] . These findings are consistent with the analyses of commissural axon navigation at the midline in STX1 KO mice . In double mutants , the absence of STX1A and STX1B proteins prevented midline crossing . In Netrin-1 [70 , 71] and DCC KO [44 , 70] mice , pre-commissural fibers display strong phenotypes , including defasciculation , aberrant trajectories , and a complete failure to cross the floor plate . Weaker phenotypes have been observed in mouse mutants for other genes involved in commissural axonal guidance , including Sonic Hedgehog and the VEGF/FlK1 pathways [46 , 57] . The phenotype described here in STX1A/B mutant mice is reminiscent of , that in Netrin-1 and DCC KO mice , with fewer axons reaching the floor plate , lateral invasion of motorneuron territories , and fewer post-crossing commissural axons , although the phenotype is much weaker and many fibers still reach the ipsilateral floor-plate border . A likely explanation could be that although STX1A/B might be required for correct sensing of Netrin1/DCC guidance [15 , 16] , compensatory mechanisms may derive from the possible lack of STX1A/B requirement in other commissural attractive pathways ( e . g . , SHH or VEGF/Flk1 ) . Our findings indicate that STX1A/B not only affects pre-crossing axons but strongly affects midline crossing and post-commissural axonal guidance . This conclusion is supported by our findings in Drosophila , which revealed a genetic interaction between Syx1A and the Robo pathway , as well as by the present in vitro experiments showing that STX1 is required for both Netrin-1 attraction and Slit-2 repulsion ( Fig 7 ) . Our previous studies showed that the lack of STX1 did not affect DCC surface expression induced by Netrin-1 [15] . In the present study we show that the lack of STX1 results in a decreased surface expression of the Robo3 receptor in response to Netrin-1 . It is therefore possible that the lack of STX1 may alter the surface expression of other Robo family members ( eg Robo1 ) and Frizzled-3 , the receptors used by post-crossing axons . This idea is supported by our STX1 loss-of-function data , showing that the lack of this SNARE protein abolished the responsiveness of post-crossing commissural axons to Slit , both in vitro ( explant experiments ) and in vivo ( chick and mouse analyses ) . We thus propose that , in contrast to DCC expression , the effect of STX1 loss-of-function on post-crossing axons may be explained in part by preventing the surface expression of axon guidance receptors required for midline crossing and post-crossing navigation . Robo3 is an atypical Robo receptor that has been proposed recently to potentiate DCC-mediated attraction to Netrin-1 , but without binding Slits [72 , 73] . Thus , the observation that Netrin-1 in STX1-deficient growth cones results in decreased Robo3 surface expression , supports the notion of a possible contribution of Robo3 membrane downregulation to explain the reduced Netrin-1 chemoattraction found in STX1 loss-of-function models . Many guidance molecules and their receptors involved in axonal guidance are conserved between vertebrates and invertebrates ( e . g . Robo , Hh , and Netrins ) [57 , 65 , 74] . Our systematic analysis of the phenotypes at the CNS midline of fly , chick , and mouse embryos mutant for STX1 unveils an evolutionarily conserved role for STX1 in midline axonal guidance . Overall , the ipsilateral phenotypes reported are consistent with the participation of STX1 in Netrin-1-dependent axonal guidance , as proposed previously [15 , 16] . In addition , here we describe post-commissural phenotypes that are reminiscent of those found in Robo , NrCAM and VEGF loss-of-function models [46 , 47 , 75 , 76] , thereby suggesting that STX1 underlies various signaling pathways . Furthermore , the phenotypes described herein for other SNARE proteins point to the participation of SNAP25 , VAMP2 and Ti-VAMP in midline axonal guidance , although the precise implication and relevance of individual SNARE proteins to specific axonal guidance signaling complexes remains to be determined . We propose that the coupling of the guidance receptor cell machinery to proteins that regulate exocytosis is a general and conserved mechanism linking chemotropic signaling to membrane trafficking [29] .
The following stocks are described in FlyBase ( http://flybase . org ) : Syx1AΔ229 , SNAP-25 ( Df ( 3L ) 1-16 ) , nSybd02894 , Ti-VAMP ( P[CG1599G7738] ) , fra3 and robo2 . All alleles used are genetic nulls . Wild-type control is yw . D . melanogaster stocks and crosses were kept under standard conditions at 25ºC . D . melanogaster embryos were staged as described by Campos-Ortega and Hartenstein [77] and stained following standard protocols . For immunostaining , embryos were fixed in 4% formaldehyde for 20 min . We used antibodies that recognize FasII ( mAb1D4 , DSHB ) , βGal ( Promega ) , Affinity-Purified Anti-HRP TRITC ( Jackson immunoResearch ) , and mAbBP102 ( DSHB ) . We used biotinylated HRP ( Amersham ) or non-biotinylated HRP ( GE Healthcare ) , Alexa488 , Alexa-555 and Alexa-647 , Cy2 , Cy3 and Cy5 conjugated secondary antibodies ( Jackson ImmunoResearch ) . For HRP histochemistry , the signal was amplified using the Vectastain-ABC kit ( Vector Laboratories ) when required . In addition , the signal for the DAB reaction was intensified with NiCl2 , except for double stainings , where it was omitted from one of the reactions . DIC photographs were taken using a Nikon Eclipse 80i microscope . Fluorescent images were obtained with a confocal microscope ( Leica TCS-SPE-AOBS and TCS-SP2-AOBS systems ) and processed using Fiji [78] and Adobe Photoshop . Images are maximum projections of confocal Z-sections . Fertilized chicken eggs were obtained from a local supplier , and embryos were staged following Hamburger and Hamilton [79] . Electroporations were performed either at HH13-14 ( E2 ) or at HH17-18 ( E3 ) . Images were acquired using a confocal spinning disk microscope ( Olympus BX61 ) . Mouse embryos aged 11 or 12 days ( E11 and E12 respectively ) were used for the experiments . To obtain the tissue samples , pregnant female mice were sacrificed by cervical dislocation . A small portion of the tail was cut for further genotyping . Embryos were then fixed with 4% paraformaldehyde , and spinal cord sections were immunostained with mouse anti—TAG1 ( mouse , Hybridoma Bank ) , followed by α-mouse IgM biotinylated ( goat , Chemicon ) and by Streptavidine Fluorescent FITC ( 490 nm , GE Healthcare ) . Sections were routinely stained with bisbenzimide . Confocal images were acquired using a Leica TCS SP5 microscope with 20x and 40x oil-immersion objectives . An average of 4–6 embryos per genotype and 10–15 slices per embryo were analyzed . For the quantification of TAG1 staining we used 4 KO and 3 WT embryos ( 32 sections from KO embryos and 8 from WT embryos ) . Data are presented as the means ± SEM . Statistical significance was determined using two-tailed Student’s t-test . Differences were considered significant at p<0 . 05 . Embryonic mouse spinal cord sections were also stained with goat α-Robo3 antibodies ( 1:100 , RD Systems ) followed by incubation with α-goat Alexa-488 secondary antibodies . Experiments with chicken embryos were approved by the Cantonal Veterinary Office Zurich . All the experiments using animals were performed in accordance with the European Community Council directive and the National Institute of Health guidelines for the care and use of laboratory animals . Experiments were also approved by the local ethical committees . The percentage of axons displaying abnormal FasII phenotypes was quantified via anti-FasII stainings by an observer with no knowledge of the genotype . Defects were categorized as “midline crossing” , “defasciculation” or “fiber collapse” . Due to the variability of fasciculations and occasional fiber collapses in wt embryos , we allowed for up to 2 of these defects to be considered “normal” or “background” and therefore set our “zero” defects at this control level . From 2 to 5 defects , we considered that embryos had a weak phenotype , while with more than 6 defects they were considered to have a strong phenotype . The percentage of axons displaying abnormal midline crossing phenotypes was quantified via anti-HRP or BP102 stainings by an observer with no knowledge of the genotype . Defects were categorized as “fuzzy commissures” , “collapsed commissures” or “thinning of longitudinals” . Fascicle distances to the midline were quantified via anti-FasII stainings , where the distance between fascicles was measures . Distances were measured in arbitrary units , between media , intermediate and lateral fascicles and these values were normalized to the length of the axonal tracks in the VNC . All data were analyzed statistically , SEMs were calculated and statistic significance assessed by Student’s t-test . The analysis of commissural axon trajectories was performed as described previously [47 , 55] . In brief , fertilized eggs were windowed on the second or third day of incubation . Extra-embryonic membranes were removed to access the spinal cord in ovo . A plasmid encoding EGFP ( 20 ng/μl ) and the dsRNA ( 250 ng/μl ) derived from STX-1A , SNAP-25 , VAMP2 , or Ti-VAMP in PBS were injected into the central canal using glass capillaries . dsRNA was produced by in vitro transcription as described previously [80] . For the production of dsRNA derived from STX1A , SNAP25 , VAMP2 , or Ti-VAMP , we used ESTs obtained from Geneservices ( now Source BioScience ) [55] . For visualization and control of injection quality and quantity , 0 . 04% Trypan blue was added . For electroporation , we used platinum electrodes connected to a BTX square wave electroporator . Electrodes were positioned parallel to the longitudinal axis of the lumbosacral spinal cord of the chicken embryo , as detailed previously [55 , 81] . Five pulses of 26 V and 50 ms duration with a 1-s interpulse interval were applied . After electroporation , eggs were sealed with Scotch tape and incubated for another 2 or 3 days . The gene silencing specificity was verified by using two independent and non-overlapping sequences for the generation of the dsRNAs , when available . dsRNA sequences used in the present study were shown to downregulate the targeted proteins by 25%-66% [29] . Because in these conditions only about 60% of the cells are efficiently transfected , small decreases in the total amount of protein can still be indicative of efficient knock-down by dsRNA transfection . For the analysis of commissural axon guidance , embryos were sacrificed between stages 25 and 26 [79] . The spinal cord was removed from the embryo , opened at the roof plate ( “open-book” preparation ) , and fixed in 4% paraformaldehyde for 30–60 min . The trajectories of dI1 commissural axons at the lumbosacral level of the spinal cord were visualized by the application of the lipophilic dye Fast DiI ( dissolved at 5 mg/ml in ethanol; Invitrogen ) to the cell bodies . Care was taken to exclusively label the dorsal-most population of commissural neurons ( dI1 neurons ) to avoid confusion with more ventral populations that have distinct pathfinding behavior . Only DiI injections sites that were in the appropriate location in the dorsal-most part of the spinal cord and within the region expressing fluorescent protein were included in the analysis . Quantification of the percentage of injections sites with axons displaying abnormal phenotypes was done by a person blind to the experimental condition . The injection sites were classified as ‘normal’ when the axons crossed the floor plate and turned rostrally along the contralateral floor-plate border . When at least 50% of the labeled fibers failed to reach the contralateral border of the floor plate , the DiI injection site was judged as ‘floor-plate stalling’ , when at least 50% of the fibers reaching the floor-plate exit site failed to turn into the longitudinal axis , the DiI injection site was considered to exhibit ‘no turning’ . Because stalling of all or most axons in the floor plate prevented the analysis of the turning phenotype , the quantification did not include a separate analysis of this phenotype between the different groups . In Fig 3G–3J , the average percentages of DiI injection sites per embryo with the respective aberrant phenotypes are shown . Mutant STX1B mice were generated using a gene-trapping technique [82] . Mice ( strain C57BL/6 from Charles River Laboratories ) were cloned from an ES cell line ( clone OST68841; Texas Institute for Genomic Medicine , TIGM ) . The ES cell clone contained an insertion of the Omnibank Vector VICTR24 in the first exon of the STX1B gene identified from the TIGM gene trap database and was microinjected into C57BL/6 host blastocysts to generate germline chimeras using standard procedures . The retroviral OmniBank Vector VICTR24 ( S1A Fig ) contained a splice acceptor sequence ( SA ) followed by a 5’ selectable marker β-GEO , a functional fusion between the β-galactosidase and neomycin resistance genes , for identification of successful gene trap events followed by a polyadenylation signal ( pA ) . Insertion of the retroviral vector into STX1B led to the splicing of the endogenous upstream exons into this cassette to produce a fusion transcript that was used to generate a sequence tag ( OST ) of the trapped gene by 3′ RACE [82] . More information on the gene trap strategies can be obtained from the TIGM website ( http://www . tigm . org/ ) . Chimeric mice were born three weeks later . Male chimeras where then mated with wt C57BL/6 to obtain germline transmission . We obtained four founders and used them to establish the colony . The derived F1 mice were screened by PCR . Genotyping of tail DNA was accomplished using PCR with forward and backward primers for the wt locus ( 5’-AAT CCG AAC AGA CTG AGA TAC ATT -3’; 5’-aGA GTT GGG CGG AAG GTA CAA GAG -3’ ) and two primers for the LTR mutant locus ( 5’-ATA AAC CCT CTT GCA GTT GCA TC-3’; 5’-AAA TGG CGT TAC TTA AGC TAG CTT GC-3’ ) . A 330-bp band was amplified for homozygous wt mice , a 270-bp and 200-bp band for homozygous mutant mice , and the three bands for heterozygous mice ( S1A and S1B Fig ) . Western blot analyses demonstrated absence of STX1B protein in STX1B mutants ( S1C Fig ) . STX1A mutant mice were a kind gift from Thomas Sudhof [83] . STX1A and STX1B mutant mice were mated to produce double heterozygous mutant mice . The double heterozygous mutant mice were then mated with each other , and the genotypes of their offspring were determined by PCR . Open-book preparations of mouse spinal cords were essentially done as described above for chicken embryos . Embryos were collected and dissected at E12 . Spinal cords were removed , opened at the roof plate and pinned down in a Sylgard dish in 4% PFA for 20–50 minutes . Commissural axons were traced with Fast DiI ( 5 mg/ml in ethanol ) by incubating the open-book preparations in PBS for at least 3 days before mounting in PBS between two 24x24 mm coverslips sealed with vacuum grease . Animal experimentation was conducted according to the European and National ( Spanish ) guidelines . The experimental protocol was approved by the local University Committee ( CEEA-UB , Comitè Ètic d´Experimentació Animal de la Universitat de Barcelona ) and by the Catalan Government ( Generalitat de Catalunya , Departament de Territori I Sostenibilitat ) with the approval number #9431 . Embryonic brains were lysed in hypotonic buffer ( 150 mM NaCl , 50 mM Tris pH 7 . 2 , 2 mM EDTA , 1% Triton X-100 , and protease inhibitors ( Complete , Mini Protease Inhibitor Cocktail Tablets , Roche ) ) . SDS-sample buffer was added to the lysates , and the proteins were analyzed by SDS-PAGE and Western blot . Proteins were transferred onto nitrocellulose membranes , which were blocked with 5% non-fat dry milk in Tris-HCl buffered saline ( TBS ) containing 0 . 1% Tween 20 , and incubated overnight at 4°C with mouse anti-STX1 ( HPC-1 clone 1:500–1000 , Sigma ) and anti-actin ( 1:10000 , Millipore ) antibodies . After incubation with secondary antibodies , blots were developed following the ECL method ( Amersham Pharmacia Biotech ) . Spinal cord dorsal neurons were isolated from E11 mouse embryos . After dissection of the alar plate , tissue was treated with Trypsin 1x for 3 minutes at 37°C and with DNase and Fetal Bovine Serum at 37°C for 10 minutes . 50000 cells were seeded on 200 mm2 wells , previously treated with 0 . 5 mg/mL poly-D-lysin coverslips . Cells were maintained in vitro for 16–24 hours in Neurobasal medium with B27 1x , Glutamax 1x , Penicillin/Streptomycin 1x and Fetal Bovine Serum 10% . Neurons were fixed in 4% paraformaldehyde , and incubated with mouse anti-DCC ( BD Pharmigen ) and goat anti-Robo3 ( RD Systems ) antibodies ( 1:100 ) . Mouse anti-STX1 ( Sigma ) , mouse anti-VAMP2 ( Synaptic Systems ) , mouse anti-SNAP25 ( Covance ) ; mouse anti-Ti-VAMP ( Abcam ) were incubated in blocking buffer to a final concentration of 1:100 . Double immunodetection for DCC and SNARE proteins required incubation in blocking buffer containing anti-mouse IgG Fab specific antibodies . Alexa 488 anti-mouse and anti-goat secondary antibodies , and Alexa 568 anti-goat antibodies were incubated in blocking buffer ( 1:100 ) . Neurons were imaged in a Leica TCS SP5 confocal microscope using a 63x oil-immersion objective . To analyse Robo3 receptor surface expression , dorsal spinal cord neurons from wild-type and STX1A/B KO embryos were prepared as above . We followed the protocol described in [15] . Briefly , cultures were incubated with either 300 ng/ml of Netrin-1 ( +Netrin-1 ) or with BSA ( -Netrin-1 ) for 30 min at 37°C . After fixation , cultures were blocked with 10% of Normal Horse Serum ( NHS ) in PBS for 2 hours and incubated without detergent with primary goat anti-Robo3 ( RD Systems , Robo3 1:100 ) . Afterwards , cells were washed with PBS and incubated with secondary antibodies ( Alexa Fluor 488 anti-goat , 1:50 ) in excess to block all the primary antibody epitopes . After several washes in PBS , cells were blocked again with 10% NGS and incubated with goat anti-Robo3 primary antibody , in the presence of 0 , 3% Triton X-100 . Cells were then washed and incubated with Alexa Fluor 568 anti-goat ( both at 1:50 , Jackson ) . Cells were washed , stained with DAPI and mounted in Mowiol . Cells were imaged in a Leica TCS SP5 microscope using a 63x oil-immersion objective . Z stacks of 8–12 confocal planes were acquired and images were processed with Fiji software . 20 growth cones per condition were analyzed with GraphPad software to quantify surface and inside receptor expression . Statistics were calculated with a two-way ANOVA and a Tukey’s multiple comparison post-test , p<0 . 05 . Dorsal ( alar plate ) spinal cord explants were dissected from E11 mouse embryos . Tissue explants were co-cultured as described [16] , along with cell aggregates of HEK293T cells stably transfected with a pCEP4-rNetrin-1c-myc construct or with control HEK293T cells . Explants and cell aggregates were embedded in a collagen matrix and maintained in vitro for 48h in Neurobasal supplemented media ( Penecillin/Streptomycin 1x , Glutamax 1x and B27 1x ) . Cultures were fixed with 4% paraformaldehyde and immunolabeled with mouse anti-βIII-tubulin ( 1:1000 , Covance ) in blocking buffer . Images were acquired with an Eclipse E1000 microscope using a 10x objective . Axon elongation was quantified by calculating the area occupied by the axons in both the proximal and distal quadrants . Data was statistically analysed as above . Explants of commissural neurons were dissected from untreated and experimental chicken embryos at HH25 . Explants were cultured in serum-free DMEM medium with GlutaMax , 5 mg/ml Albumax , N3 and 1 mM sodium pyruvate . Eight-well Lab-Tek dishes ( Nunc ) were coated with polylysine ( 20 μg/ml ) and laminin ( 10 μg/ml ) . After 24 h control medium or medium containing Slit-2 ( 100 ng/ml; R&D Systems ) was added to the explants . After an additional 18 h , cultures were fixed in 4% PFA and the average neurite length of each explant was measured using the cellSens program ( Olympus ) . | Syntaxin-1 is a core factor in tethering synaptic vesicles and mediating their fusion to the cell membrane at the synapse . Thus , Syntaxin-1 mediates neurotransmission in the adult nervous system . Here we show that this protein is also involved in axonal guidance in the CNS of vertebrates and invertebrates during the development of the nervous system: our systematic analysis of the phenotypes in the nervous system midline of fly , chick , and mouse embryos mutant for Syntaxin-1 unveils an evolutionarily conserved role for this protein in midline axonal guidance . Further , we also dissect the contribution of other proteins regulating neuronal exocytosis in axonal development . We propose that the coupling of the guidance molecule machinery to proteins that regulate exocytosis is a general mechanism linking chemotropism to axonal growth . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"nervous",
"system",
"neuroscience",
"animals",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"nerve",
"fibers",
"embryos",
"drosophila",
"research",
"and",
"analysis",
"methods",
"chemoattractant",
"axon",
"guidance",
"embryology",
"developmental",
"neuroscience",
"spinal",
"cord",
"animal",
"cells",
"axons",
"insects",
"arthropoda",
"cellular",
"neuroscience",
"eukaryota",
"neuroanatomy",
"cell",
"biology",
"axon",
"guidance",
"anatomy",
"phenotypes",
"neurons",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
]
| 2018 | A conserved role for Syntaxin-1 in pre- and post-commissural midline axonal guidance in fly, chick, and mouse |
Chemotherapy continues to have a major impact on reducing the burden of disease caused by trypanosomatids . Unfortunately though , the mode-of-action ( MoA ) of antitrypanosomal drugs typically remains unclear or only partially characterised . This is the case for four of five current drugs used to treat Human African Trypanosomiasis ( HAT ) ; eflornithine is a specific inhibitor of ornithine decarboxylase . Here , we used a panel of T . brucei cellular assays to probe the MoA of the current HAT drugs . The assays included DNA-staining followed by microscopy and quantitative image analysis , or flow cytometry; terminal dUTP nick end labelling to monitor mitochondrial ( kinetoplast ) DNA replication; antibody-based detection of sites of nuclear DNA damage; and fluorescent dye-staining of mitochondria or lysosomes . We found that melarsoprol inhibited mitosis; nifurtimox reduced mitochondrial protein abundance; pentamidine triggered progressive loss of kinetoplast DNA and disruption of mitochondrial membrane potential; and suramin inhibited cytokinesis . Thus , current antitrypanosomal drugs perturb distinct and specific cellular compartments , structures or cell cycle phases . Further exploiting the findings , we show that putative mitogen-activated protein-kinases contribute to the melarsoprol-induced mitotic defect , reminiscent of the mitotic arrest associated signalling cascade triggered by arsenicals in mammalian cells , used to treat leukaemia . Thus , cytology-based profiling can rapidly yield novel insight into antitrypanosomal drug MoA .
Chemotherapy is central to the control of the neglected tropical diseases caused by African trypanosomes ( Trypanosoma brucei spp ) , South American trypanosomes ( Trypanosoma cruzi ) and Leishmania spp; the related kinetoplastid parasites [1] . The current drugs suffer problems with complex administration , efficacy , toxicity and resistance [2] . There are a small number of drugs in clinical trials for these diseases but there remains a desperate need for new and improved drugs . An understanding of drug mode-of-action ( MoA ) would aid the development of these new drugs , but our knowledge of how the current antitrypanosomals work is lacking [1] . This has also been the case for drugs currently in clinical trials and for other promising compounds that emerged from phenotypic cell-based screening campaigns . These gaps in knowledge will only become more acute given the current trend of phenotypic screening and the typically high attrition rate during the development of compounds that emerge from target-based screening [1] . In the case of Human African Trypanosomiasis ( HAT ) , there are five current drugs , and two in clinical trials [1] . The disease progresses from stage 1 to stage 2 , when parasites invade the central nervous system , which is typically lethal without treatment . The current drugs are eflornithine , melarsoprol , nifurtimox , pentamidine and suramin . Nifurtimox and eflornithine were introduced most recently and are used in combination to treat stage 2 disease in West Africa [3]; eflornithine , however , is particularly challenging to administer . Melarsoprol use has been largely phased out , as it is highly toxic [4] . Resistance to melarsoprol , due to the disruption of a parasite aquaglyceroporin [5 , 6 , 7] , is also now widespread , but this drug remains the only available treatment for stage 2 disease in East Africa; the parasite sub-species found in this region is not susceptible to eflornithine [8] . Pentamidine and suramin are used to treat stage 1 disease in West and East Africa , respectively . The new orally active drugs in clinical trials are acoziborole / SCYX-7158 [9] and fexinidazole [10] . With the exception of eflornithine and acoziborole , the mechanisms of action for the antitrypanosomal drugs above are not understood in any detail . Eflornithine enters trypanosomes via the amino-acid transporter AAT6 [11] and inhibits ornithine decarboxylase [12] and , only recently , acoziborole was shown to target an mRNA maturation factor known as CPSF3 [13] . Almost all antitrypanosomal drugs emerged based on an ability to selectively target trypanosomes or to reduce parasite viability in cell culture or in animal models . Rather than differences between host and parasite intracellular targets , the selective efficacy of suramin , melarsoprol and pentamidine is at least partly due to trypanosome-specific uptake mechanisms [14] . Suramin enters trypanosomes via an endocytic pathway involving a bloodstream stage-specific invariant surface glycoprotein ISG75 [14] , and variant surface glycoprotein [15] . The action of suramin is enhanced by the import of ornithine , via an amino acid transporter [16] , and its metabolism by ornithine decarboxylase , such that eflornithine is antagonistic [14] . Suramin inhibits pyruvate kinase in vitro but the drug may also occupy ADP/ATP binding sites in other enzymes [17] , none of which have been validated as targets in vivo . Melarsoprol is an arsenical drug that enters trypanosomes via an adenosine transporter [18] and an aquaglyceroporin , AQP2 [5] , acting primarily by forming a stable adduct , known as Mel T , with the antioxidant , trypanothione [19] . Pentamidine , like melarsoprol , enters trypanosomes via AQP2 [5] . Indeed , pentamidine inhibits the glycerol permeability of AQP2 [20] but this particular activity has little impact on parasite viability . Rather , this DNA-binding drug [21] becomes highly concentrated in the cell and collapses trypanosome mitochondrial membrane potential [22] . Pentamidine remains a low nanomolar antitrypanosomal agent against parasites lacking mitochondrial ( kinetoplast ) DNA , however , which display only 2 . 5-fold resistance [23 , 24] . Nifurtimox and fexinidazole are both nitro pro-drugs that are activated by a putative ubiquinone nitroreductase ( NTR ) located in parasite mitochondria [14 , 25 , 26] , but it is unknown whether these drugs kill parasites primarily by disrupting mitochondrial functions or whether the toxic metabolites access targets outside the mitochondria . Cytology-based profiling can facilitate antibiotic discovery efforts [27] and a selection of cellular assays have been previously applied to antitrypanosomal compounds [28]; but this previous study examined only one of the current HAT compounds ( pentamidine ) and employed only two of the assays described below . We now report cytology-based profiling for T . brucei to probe the MoA of all five antitrypanosomals used in patients . We describe a panel of assays that assess cell cycle progression , nuclear and mitochondrial DNA content , mitochondrial DNA replication , nuclear DNA damage , mitochondrial membrane potential , and lysosome structure and function . Using these assays , we show that each drug tested induces specific and distinct cellular perturbations , yielding novel insight into the MoA of the antitrypanosomal agents . Follow-up studies revealed a melarsoprol-induced mitotic defect that is dependent upon a specific set of kinases .
The potency of the antitrypanosomal drugs used in the clinic varies widely . The 50% effective growth inhibitory concentration ( EC50 ) determined against bloodstream-form T . brucei in culture is in the low nM range for pentamidine ( 2 . 5 nM ) , suramin ( 27 nM ) and melarsoprol ( 7 nM ) but is in the low μM range for eflornithine ( 15 μM ) and nifurtimox ( 2 . 6 μM ) ; a 6 , 000-fold differential between the most potent ( pentamidine ) and least potent ( eflornithine ) . It is important to note that , since EC50 values are typically determined over three to four days , they may fail to reflect the rate at which growth is inhibited or whether the compound is cytocidal or cytostatic at a particular concentration . We examined the growth profiles of bloodstream-form trypanosomes treated with each of the five clinical antitrypanosomal drugs at 1 x EC50 and 5 x EC50; see Materials and methods . All drugs had a relatively moderate impact on growth at 1 x EC50 , as expected ( Fig 1 ) . In contrast , growth at 5 x EC50 revealed a clear difference between eflornithine , which is known to be cytostatic [29] , and the other drugs , which were all demonstrably cytocidal over four days ( Fig 1 ) . We selected 5 x EC50 exposure for 24 hours for subsequent assays . These concentrations and this time-point were selected to achieve a balance between allowing robust primary phenotypes to develop but to minimise the emergence of secondary effects associated with loss of viability . Our first cellular assay was a simple examination of each of the five populations of drug-treated cells for defects in gross cellular morphology by phase-contrast microscopy . We found that the majority of suramin-treated cells became enlarged and distorted ( see below ) . We also looked for the ‘BigEye’ phenotype , which is associated with an endocytosis-defect and observed following treatment with the N-myristoyltransferase inhibitor DDD85646 [30] , however this phenotype was not seen . Arguably the simplest and most widely used fluorescent-staining method for T . brucei is DAPI-staining of DNA followed by microscopy . This is particularly informative for T . brucei because cells in which the single mitochondrial genome , or kinetoplast , has segregated are easily visualised and scored , alongside the unsegregated or segregated nuclear genome [31] . We , therefore , examined DAPI-stained cells by microscopy following drug-exposure . Cells were scored according to the number of nuclei ( n ) and kinetoplasts ( k ) . In non-treated cells ( Fig 2A , NT ) , we found that ~80% of cells displayed a 1n1k pattern ( primarily G1 + S-phase ) , ~15% of cells were 1n2k ( primarily G2 ) and ~5% of cells were 2n2k ( post-mitosis ) . We exercised some caution when analysing these data for drug-treated cells , and focused on only robust phenotypes that preceded or were coincident with loss-of-viability . The analysis revealed a major increase in the proportion of cells with more than two nuclei following suramin treatment ( 79% ) , while eflornithine treatment also yielded an appreciable increase , with 11% of cells having more than two nuclei ( Fig 2A , yellow bars ) . In addition , we noted ~5% of cells lacking a kinetoplast ( 0k ) following pentamidine treatment ( Fig 2A , orange bar , see below ) . This loss of kinetoplast DNA is consistent with previous observations [28] . DNA staining followed by flow cytometry allows the rapid analysis of large numbers of cells and can reveal relative cellular ( primarily nuclear ) DNA-content . Accordingly , we next examined propidium iodide stained cells by flow cytometry following drug-exposure . Untreated cells showed the characteristic profile , with a G2/M population that was approximately half the magnitude of the G1 population ( Fig 2B , top-left and red profiles ) . In this assay , both nifurtimox-treated and pentamidine-treated cells conformed to the profile observed for untreated cells . In contrast , and consistent with the microscopy analysis above , we observed >70% of cells with >2n DNA content following suramin treatment; eflornithine treatment also yielded an appreciable increase in this category ( Fig 2B , blue profiles ) . Only melarsoprol treatment yielded a second , distinct and notable perturbation using this assay; the G2/M population increased from 51% to 83% , relative to the G1-population ( Fig 2B , blue profile ) . Since we had not scored an increase in the proportion of 2n cells by microscopy following melarsoprol-treatment , this particular flow-cytometry profile indicated an increase in cells with a replicated but unsegregated nuclear genome . Thus , DNA-staining with microscopy and flow cytometry suggested that suramin inhibited cytokinesis and that melarsoprol inhibited mitosis ( see below ) . The TUNEL ( Terminal deoxynucleotidyl transferase dUTP Nick End Labelling ) assay allows the fluorescent labelling and detection of blunt DNA ends , following programmed cell-death ( PCD ) for example; notably , conventional PCD is not thought to operate in trypanosomatids [32] . Using this assay , TUNEL-signals were not readily detected in trypanosome nuclei . In contrast , and as previously reported [33] , we observed robust signals associated with kinetoplasts ( Fig 3A ) . We found approximately 25% of control cells to be TUNEL-positive ( Fig 3A ) . Eflornithine , nifurtimox and melarsoprol-treatment did not significantly alter the proportion of cells with detectable TUNEL-signals when compared to cells that had not been exposed to drug ( Fig 3A ) , but the proportions of TUNEL-positive cells were significantly reduced following pentamidine or suramin-treatment ( Fig 3A ) . This was likely due to loss of kinetoplast DNA following pentamidine-treatment ( see above ) and may have been due to limited repeated rounds of kinetoplast DNA replication in multi-nucleated cells following suramin-treatment ( see below ) . TUNEL signals were found to be cell cycle dependent ( Fig 3B ) , consistent with the high concentration of transient nicked DNA ends expected to be present during the replication of thousands of minicircles [34] . Indeed , the majority ( 91% ) of elongated kinetoplasts ( in S-phase cells ) and a substantial proportion ( 22% ) of segregated kinetoplasts ( in G2 cells ) were TUNEL-positive , while we observed very few TUNEL-positive kinetoplasts in G1 or post-mitotic cells ( Fig 3C ) . The TUNEL-signals were consistently observed at opposite poles of extended kinetoplasts and , when present , on both segregated kinetoplasts . Notably , the appearance of kinetoplast-associated TUNEL-signals remained synchronised even in those suramin-treated cells with four kinetoplasts; all four were either negative or positive in each cell ( Fig 3D ) . Thus , TUNEL revealed those cells that are progressing through kinetoplast S-phase and indicated continued synchronisation , even in cells with four kinetoplasts . We previously identified trypanosome γH2A , a modified histone H2A that is phosphorylated at the C-terminus and that accumulates at sites of nuclear DNA double-strand breaks [35] . To assess nuclear DNA damage in drug-treated cells , we used a γH2A antibody in an immunofluorescence assay; methyl methanesulfonate ( MMS ) is a radiomimetic DNA-damaging agent and was included as a positive control ( Fig 3E ) . We found no significant differences in the proportion of cells with nuclear γH2A foci compared to untreated cells ( Fig 3E ) , suggesting that none of the current drugs kill trypanosomes by forming double-strand breaks in nuclear DNA . Notably , the nitro pro-drugs , nifurtimox and fexinidazole , although found to be mutagenic by Ames test ( Salmonella typhimurium based assay ) , are not thought to be genotoxic to mammalian cells [36 , 37] . MitoTracker fluorescence staining is dependent upon mitochondrial membrane potential . We treated trypanosomes with antitrypanosomal drugs and scored cells by microscopy for an extended MitoTracker signal , as observed in non-treated cells ( Fig 4A , NT panels ) . Significantly fewer cells scored positive for extended signals following either nifurtimox or pentamidine-treatment , compared to non-treated cells ( Fig 4A ) . There were also qualitative differences in signals resulting from drug treatments; whereas pentamidine eliminated the detectable signal , nifurtimox-treatment produced an intense and discreet signal adjacent to the kinetoplast ( Fig 4A ) . Notably , mitochondrial membrane potential was maintained following suramin-treatment ( Fig 4A ) , despite the gross morphological perturbations observed in these cells . LysoTracker is highly selective for acidic organelles . We treated trypanosomes with antitrypanosomal drugs and scored cells for LysoTracker signals by microscopy . Non-treated cells displayed a single signal between the nucleus and kinetoplast ( ~60% ) , as expected for the trypanosome lysosome ( Fig 4B , NT panels ) . Melarsoprol substantially reduced the proportion of LysoTracker-positive cells ( Fig 4B ) , but this failed to achieve statistical significance . In the case of suramin , which is known to accumulate in trypanosomes through receptor-mediated endocytosis [14] , acidification of the lysosomal compartment does not appear to be perturbed ( Fig 4B ) . The variability we observe in this assay suggests either that additional replicates will be desirable in future , or that this particular assay will be of value only when major lysosomal perturbation occurs . Using the DAPI , TUNEL and MitoTracker assays following pentamidine-treatment , we observed kinetoplast-loss , a reduced proportion of positive cells and a loss of mitochondrial membrane potential , respectively . To quantify these effects , we scored each phenotype following 12 , 24 , and 48 h of drug-exposure ( Fig 5A–5C ) . These analyses revealed a sharp increase in kinetoplast-negative cells in the 48-h DAPI-stained population ( Fig 5A ) and a progressive decline in TUNEL-positive ( Fig 5B ) and MitoTracker-positive cells ( Fig 5C ) at each time-point . DAPI-stained cells appeared to reveal progressively diminished kinetoplast DNA rather than loss in one step . For a quantitative and objective assessment of kinetoplast DNA-staining intensity , we adapted an ImageJ-based approach [31]; see Materials and methods . ImageJ efficiently identified nuclei and kinetoplasts in untreated cells ( Fig 5D , red and green , respectively in the upper panel ) , but failed to identify very low-intensity kinetoplasts in pentamidine-treated cells ( Fig 5D , open green circle in the lower panel ) . A quantitative analysis indicated that the kinetoplasts that are still detected following pentamidine-treatment display reduced surface area and signal intensity relative to the control population; values are expressed relative to nuclei in the same cell ( Fig 5D ) . These results confirm progressive loss of kinetoplast DNA induced by pentamidine . Nifurtimox-treatment produced an intense and discreet MitoTracker signal adjacent to the kinetoplast ( Fig 4A ) . To explore this effect further , we stained the nuclear-encoded F1β-subunit of the mitochondrial ATP-synthase ( Tb927 . 3 . 1380 ) in drug-treated and MitoTracker-stained cells and examined these cells by microscopy . The ATP-synthase signal was largely coincident with the extended MitoTracker signal in control cells ( Fig 6A , NT panels ) and was also coincident with the focal MitoTracker signal in nifurtimox-treated cells ( Fig 6A , Nif panels ) , possibly indicating a structural defect in mitochondria in the latter cells . We next assessed F1β-subunit expression by protein blotting of whole-cell extracts , which revealed a substantial and specific reduction in abundance following nifurtimox-treatment ( Fig 6B ) . Depletion of the ATP-synthase component may reflect a more widespread depletion of mitochondrial proteins , consistent with activation of this pro-drug by a mitochondrial NTR [14 , 25 , 26] . Using the same assays , pentamidine-treated cells displayed a diffuse and cytosolic , rather than mitochondrial , ATP-synthase signal ( Fig 6A , Pen panels ) . In this case , protein import into mitochondria [38] may be defective as a result of loss of mitochondrial membrane potential , as indicated by the diminished MitoTracker signal ( Fig 6A , Pen panels ) . Consistent with this hypothesis , the total cellular F1β signal detected by protein blotting remained constant following pentamidine treatment ( Fig 6B ) . DNA staining followed by microscopy or flow cytometry suggested that melarsoprol inhibited mitosis . Indeed , a closer inspection of DAPI-stained cells revealed a sub-set ( >10% ) with “conjoined” nuclei ( Fig 7; top panels , double-arrowheads ) . For an objective assessment of nuclear DNA-staining intensity in these cells , we used the quantitative ImageJ-based approach; see Materials and methods . In this case , nuclei were compared to kinetoplasts in ‘1n2k’ cells , revealing both increased nuclear surface area and signal intensity following melarsoprol-treatment ( Fig 7 ) . These quantitative results confirmed the mitotic defect resulting from melarsoprol treatment . Cytology-based profiling may be combined with orthogonal approaches to probe antitrypanosomal compound MoA . We have found high-throughput genetic screening to be particularly informative [13 , 14] . A prior RNA interference ( RNAi ) Target Sequencing ( RIT-seq ) screen revealed a highly significant ( P = 2 . 3 × 10−9 ) over-representation of kinases , comprising three hits among the top seven in this screen; these were a putative mitogen-activated protein kinase ( MAPK11 , Tb927 . 10 . 12040 ) , a putative MAPK kinase ( MKK4 , Tb927 . 8 . 5950 ) and a putative MAPK kinase kinase ( STE11 , Tb927 . 10 . 1910 ) [14] . It is also notable that trypanosome CDC14 ( Tb927 . 11 . 12430 ) was a hit in this screen , since human Cdc14 stabilises Wee1 , a key kinase that inhibits mitosis [39] . Thus , knockdowns that promote mitosis may partially counteract the inhibitory effects of melarsoprol , but the RIT-seq screening hits noted above have not previously been independently validated nor further characterised . To explore our hypothesis , we further characterised these kinases . All three were expressed with a C-terminal 12-myc epitope tag and the tagged proteins were all found to localise to the trypanosome cytosol ( Fig 8A ) . Since RIT-seq screening fails to yield clonal knockdown strains [14] , we assembled pairs of independent knockdown strains for two of the kinases ( Tb927 . 10 . 12040 and Tb927 . 8 . 5950 ) . Both knockdowns , confirmed by monitoring the epitope-tagged proteins ( Fig 8B ) , reproducibly and significantly ( P < 0 . 0001 ) increased melarsoprol EC50 by 1 . 7 +/- 0 . 07 fold and 1 . 5 +/- 0 . 01 fold , respectively ( Fig 8C ) , as predicted by the RIT-seq screen [14] . Finally , we determined whether knockdown alleviated the melarsoprol-induced , conjoined-nuclei phenotype and found that this was indeed the case for both kinases ( Fig 8D ) . Thus , the melarsoprol-induced mitotic defect is kinase-dependent .
Cytocidal or cytostatic compounds identified using phenotypic approaches may target proteins , nucleic acids , membranes or other metabolites . They may also exhibit polypharmacology , killing cells by perturbing multiple pathways . Even compounds developed as target-based therapies may kill trypanosomes by ‘off-target’ mechanisms . Determining mechanism of action remains a major challenge for these drugs and compounds and an improved understanding of how they kill parasites could present new opportunities in terms of developing more potent compounds , delivering compounds to their targets more effectively or devising combination therapies that minimise the likelihood of resistance . Eflornithine kills trypanosomes by inhibiting ornithine decarboxylase but the mode-of-action of the other four current antitrypanosomal drugs is not known . Our studies indicate that cytology-based profiling can provide a rapid and effective means to yield insight into drug mode-of-action and we now present additional insights into the mode-of-action for all current drugs used to treat human African trypanosomiasis . Suramin was found to produce cells with more than two nuclei , indicating a defect in cytokinesis with continued mitosis . Multi-nucleated cells were still stained by MitoTracker , indicating that this organelle retained membrane potential . DAPI-staining and TUNEL ( terminal dUTP nick end labelling ) assays indicated that nuclear:nuclear and nuclear:kinetoplast replication remained synchronised in these cells . Indeed , although the TUNEL assay failed to reveal any nuclear DNA damage induced by the drugs used here , it did provide an excellent marker for the kinetoplast replication cycle , producing robust signals consistent with the presence of DNA-ends at antipodal sites on the kinetoplast during minicircle DNA replication [34] . In this regard , TUNEL provides a useful marker for kinetoplast S-phase . Pentamidine is a DNA-binding drug [21] that collapses trypanosome mitochondrial membrane potential [22] and induces loss of kinetoplast DNA [28] . Data from Saccharomyces cerevisiae indicated pentamidine accumulation in the mitochondrion but also inhibition of translation in these cells [40] . In addition , metabolomic studies indicated that pentamidine is unlikely to act through the inhibition of any specific metabolic pathways [41] . We find that pentamidine-induced loss of kinetoplast DNA is progressive , suggesting progressive loss of maxi- and minicircles; the latter are present in thousands of copies per kinetoplast [34] . Kinetoplast DNA loss is revealed by both DAPI-staining and TUNEL-assay . Loss of kinetoplast DNA is expected to disrupt membrane-potential since the A6-subunit of the ATP-synthase , required to maintain this potential , is encoded by kinetoplast DNA [42] . Thus , we suggest that kinetoplast DNA is indeed a primary target for pentamidine , involving inhibition of mitochondrial type II topoisomerase action , as previously suggested [43] . What remains unclear is whether loss of mitochondrial membrane-potential is solely a consequence of the kinetoplast defect or whether this reflects an independent response to pentamidine . Nifurtimox is a pro-drug that is activated by a mitochondrial NTR [26] , but it is unknown whether the toxic metabolite ( s ) access ( es ) targets outside the mitochondrion , or whether parasite killing is primarily due to disruption of mitochondrial functions . We found that nifurtimox , like pentamidine , also disrupted the MitoTracker signal . Kinetoplast defects were not observed , however , and the MitoTracker staining pattern was distinct to that seen with pentamidine . Nifurtimox also induced loss of a nuclear-encoded ATP-synthase subunit , while data from pentamidine-treated cells indicated that ATP-synthase was still expressed , but not imported , when membrane potential was perturbed . Thus , we suggest severe disruption of mitochondrial structure and function by nifurtimox , consistent with damage to targets in the organelle where the drug is activated . Fexinidazole is another nitro pro-drug , currently in clinical trials , that is also activated by mitochondrial NTR [25] . These drugs may act through similar antitrypanosomal mechanisms . Melarsoprol forms potentially toxic adducts with trypanothione [19] , and we now show that this drug increases the proportion of cells with a replicated but unsegregated nuclear genome . This indicates a defect in mitosis . The identification of multiple putative kinases in a loss-of-function screen for melarsoprol-resistance suggested a role for a signalling cascade in melarsoprol susceptibility [14] . The current findings led us to consider a more specific role for these putative kinases in the control of mitosis . Indeed , these proteins are putative MAPK , MAPKK and MAPKKKs and our results now indicate that the mitotic defect is less pronounced when the putative MAPK ( Tb927 . 10 . 12040 ) and MAPKK ( Tb927 . 8 . 5950 ) are depleted . We suggest that these kinases negatively control mitosis , as part of normal quality control following DNA replication . Bypass of this surveillance could allow cells to continue to grow , possibly accumulating , but tolerating , melarsoprol-induced ( oxidative ) damage . Thus , knockdowns that promote mitosis may partially counteract the inhibitory effects of melarsoprol . Arsenicals are used to treat leukemia but are also themselves mutagens . There is indeed evidence that arsenic induces DNA damage in yeast [44] . Notably , the MoA in yeast involves activation of a MAP kinase pathway [45] and this also appears to be the case in mammalian cells [46] , where mitotic arrest occurs as a result of induction of a mitotic spindle checkpoint [47] . We did not detect any evidence for nuclear DNA damage in T . brucei following melarsoprol treatment but , as detailed above , did find a melarsoprol-dependent and kinase-dependent mitotic defect . Thus , arsenical MoA may operate through a common kinase signalling cascade leading to mitotic arrest in both trypanosomes and mammalian cells . Notably , like the Myt1 kinase , which phosphorylates Cdc2 and controls mitosis in Xenopus [48] , Tb927 . 10 . 1910 has a putative transmembrane domain; Tb927 . 10 . 1910 also has a putative guanylate cyclase domain . These findings illustrate how cytology and genetic approaches can converge to yield insights into drug MoA in trypanosomes . In this case , we propose a common MoA for arsenicals in human cells and in trypanosomes . Cytology-based approaches can provide rapid and cost-effective methods for quantitative profiling of cellular responses to drugs . The drugs can also be considered as chemical probes for exploring parasite biology . Taking a systematic cytology-based approach with antitrypanosomal drugs of uncertain MoA , our findings indicate target organelles and structures for pentamidine , nifurtimox and melarsoprol; the kinetoplast , the mitochondrion and the nucleus , respectively . Further analysis , guided by the primary assays , indicated destruction of mitochondrial ATP-synthase by nifurtimox and a mitotic defect induced by melarsoprol . These assays should also provide novel insights into MoA when applied to cells treated with other antitrypanosomal compounds , including those with known primary targets . Common profiles will allow compounds to be clustered based on their primary MoA . New assays and bespoke assays can be incorporated as appropriate and these may be guided by outputs from orthogonal genetic , proteomic , metabolomic or computational approaches . The approach may also be usefully applied to the other parasitic trypanosomatids , T . cruzi and Leishmania . These and other approaches should reveal those particularly susceptible pathways that can be prioritised and targeted by antitrypanosomal therapies .
Lister 427 derived T . brucei ( clone 221a ) bloodstream form cells were grown in HMI-11 in the presence of antitrypanosomal drugs . Cultures were initiated at 2 x 105 cells . ml-1 and incubated at 37°C in a humidified 5% CO2 atmosphere . The drugs were applied at five times the EC50 , as determined using a standard AlamarBlue assay [49]; for EC50 determination , drug exposure was for 66–67 h and AlamarBlue incubation was for 5–6 h . Plates were read on an Infinite 200 Pro plate-reader ( Tecan ) . The EC50 values of the antitrypanosomal drugs used were: Eflornithine 15 μM; Melarsoprol 7 nM; Nifurtimox 2 . 6 μM; Pentamidine 2 . 5 nM; Suramin 27 nM . Methyl methanesulfonate ( MMS , Sigma ) was used at 0 . 0003% . To monitor cumulative trypanosome growth , cultures were grown in the presence of drug , with four technical replicates of cell density counted at 6 , 9 , 24 , 48 , 72 and 96 h . Cultures were diluted with fresh media containing the appropriate drug , such that cell density never exceeded 2 x 106 cells . ml-1 . All tagging and RNAi constructs were transfected into 2T1 T . brucei cells [50] . RNAi was induced with tetracycline ( 1 μg/ml ) for three days prior to initiating EC50 determination . Cells were grown in the presence of drug at 5 x EC50 for 24 h , unless stated otherwise . Cells were then fixed in 2% v/v paraformaldehyde for 30 min at 4°C before being washed three times in PBS . Fixed cells were dried onto slides before staining with antibodies ( outlined below ) . Slides were washed in PBS and mounted in Vectashield containing DAPI ( Vector Laboratories ) ; 4 , 6-diamidino-2-phenylindole . Scoring of phenotypes was carried out by counting 100 cells per condition , and by two of us , with the data combined . For immunofluorescence analysis ( IFA ) , cells dried onto slides were permeablised in 0 . 5% Triton X100 / PBS for 20 min and washed three times in PBS before blocking in 50% foetal bovine serum FBS / PBS . Primary antibody ( α-γH2A ) [35] was diluted 1:250 and applied for 1 h , slides were washed three times in PBS , and secondary antibody ( fluorescein-conjugated goat anti-rabbit ) was diluted 1:100 and then applied for 1 h . Antibodies were applied in 3% FBS / PBS . The F1β-subunit of the mitochondrial ATP-synthase was similarly detected using a polyclonal rabbit antiserum directed against the Crithidia fasciculata ATP synthase ( 1:500 ) , which cross-reacts with the T . brucei orthologue [51 , 52] . All phase and epifluorescence images were captured on an Eclipse E600 microscope ( Nikon ) using a Coolsnap FX ( Photometrics ) charged coupled device camera and processed in Metamorph 5 . 4 ( Photometrics ) . MitoTrackerRed ( Invitrogen ) was added to cultures at 100 nM . Cultures were incubated for 5 min under standard conditions before parasites were harvested by centrifugation at 1000 x g for 10 min before fixing as outlined above . LysoTracker ( Invitrogen ) was added to cultures at 50 nM . Cultures were incubated under standard conditions for 1 h before parasites were harvested by centrifugation at 1000 x g for 10 min before fixing as outlined above . Cells on slides were fixed , dried and permeablised as described above for IFA . Reaction mix from the ‘In Situ Cell Death Detection Kit ( fluorescein ) ’ ( Roche ) was applied to cells for 1 h as per the manufacturer’s instructions . Cells were harvested and washed in ice cold PBS and resuspended in 300 μl PBS before the addition of 700 μl of methanol and stored at 4°C for 12 h . Cells were then harvested at 400 x g for 10 min at 4°C and resuspended in 1 ml PBS . 10 μg/ml RNAse A and 10 μg/ml propidium iodide was added to each sample and incubated at 37°C for 45 min in the dark . Analysis was performed on an LSRII flow cytometer ( BD Biosciences ) , and data analysis was conducted in FlowJo ( Tree Star ) . All images of DAPI fluorescence were captured at 40 ms exposure time for consistency and to avoid overexposure . They were analysed using an ImageJ plug-in [31] modified to enable cell cycle analysis from DAPI and phase contrast images . DAPI alone was suitable for the identification of nuclei and kinetoplasts , since kinetoplasts do not generally overlap with nuclear DNA signals in bloodstream form T . brucei . As in the original ImageJ plug-in , two kinetoplasts are counted only when no longer linked by pixels with signal . A subset of the original macros were retained and modified , as necessary: Measure K/N Signal , Cell Analysis , K/N Count Summary , and Save Analysis . First , the Measure K/N Signal tool thresholds images using built-in ImageJ functions , and extracts object area values from DAPI images . Next , the script applies the K-means clustering algorithm [31] to separate kinetoplasts and nuclei into respective clusters , returning values for the maximum kinetoplast area and minimum nucleus area . These values are then passed to the Cell Analysis tool which finds objects in phase images , creates binary copies of the phase and DAPI image and identifies kinetoplasts and nuclei , counting how many lie within each cell . K/N Count Summary and Save Analysis tools function identically to the original macros described in [31] . Macro scripts are available on request . For tagging native Tb927 . 8 . 5950 , Tb927 . 10 . 12040 and Tb927 . 10 . 1910 alleles with 12-myc epitope tags at the C-terminus , we used the following primer-pairs , to amplify 893 , 1038 and 769 bp fragments , respectively: 5950MF ( GATCAAGCTTGATCCATGTGTAGTTGAC ) and 5950MR ( GATCTCTAGAGGATACTGGTGAACCATC ) ; 12040MF ( GATCAAGCTTGGCACACTTTCACCACGAT ) and 12040MR ( GATCTCTAGACTCAACGGAACCCACATATT ) ; 1910MF ( GATCAAGCTTGCGTGTATCTAGGCATGGA ) and 1910MR ( GATCTCTAGAAAGGGAAAAAAGTG ) . These primer-pairs incorporate HindIII and XbaI sites , respectively ( italics ) . The resulting fragments were cloned in the pNATx12MYC construct [53] . The resulting pNAT5950-12myc , pNAT12040-12myc and pNAT1910-12myc constructs were linearized by digestion with Bstz171 , EcoRV and EcoRV prior to transfection , respectively . For knockdown of Tb927 . 8 . 5950 or Tb927 . 10 . 12040 using RNA interference , we used the following primer-pairs , respectively: 5950RF2 ( GATCTCTAGAGGATCCAAACGACCCAAGTTGGAGAG ) and 5950RR2 ( GATCGGGCCCGGTACCGCTTCCAGCGTCCATGTATT ) ; 12040RF2 ( GATCTCTAGAGGATCCATTCTTGGTGAGTTGCTGGG ) and 12040RR2 ( GATCGGGCCCGGTACCACTCTCATCATACACCGCCC ) . These primer-pairs incorporate XbaI / BamHI and ApaI / KpnI sites , respectively ( italics ) . The resulting fragments were cloned in the pRPaiSL construct [53] . The resulting pRPa5950-RNAi and pRPa12040-RNAi constructs were linearized by digestion with AscI prior to transfection . Total cell extracts were separated on SDS-polyacrylamide gels and subjected to standard western blotting analysis . Duplicate gels were generated and one was stained with Coomassie and the other was used to produce the nitrocellulose blot . Blots were blocked in 5% milk in TBST and washes were performed in TBST ( 0 . 05% Tween ) . Blots were then probed with mouse α-c-Myc primary antibody ( 1:5000; 9E10 , Source Biosciences ) or the ATP synthase primary antibody ( 1:500 ) . Following incubation with the appropriate secondary antibodies ( 1:2000; Pierce ) , membrane was washed and visualised using the Amersham enchanced chemiluminescence ( ECL ) detection system ( GE Healthcare Life Sciences ) according to the manufacturer instructions . | African trypanosomes cause devastating and lethal diseases in humans and livestock . These parasites are transmitted among mammals by tsetse flies and circulate and grow in blood and tissue fluids . There are several drugs available to treat patients but , despite their use for many decades , we know relatively little about how they work . We reasoned that exposure of trypanosomes to each drug , followed by microscopic examination of cellular structures , would reveal the major cellular compartments , structures or growth phases affected . For example , we examined two major DNA structures , and cellular compartments known as mitochondria . We found that two drugs thought to act in mitochondria did indeed disrupt this compartment , but in completely different ways . Another drug stopped cell growth at a specific point in the cycle . An arsenic-based drug , related to anti-leukaemia drugs , perturbed the nuclear DNA division cycle , indicating that arsenicals may kill parasites and cancer cells by similar mechanisms . Thus , the ‘chemical-biology’ profiles we observe illuminate distinct killing mechanisms . A similar approach can now be used to assess new drugs , and the insights may help to develop improved anti-parasite therapies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"membrane",
"potential",
"electrophysiology",
"parasitic",
"protozoans",
"mitosis",
"trypanosoma",
"brucei",
"trypanosoma",
"brucei",
"gambiense",
"protozoans",
"dna",
"replication",
"pharmaceutics",
"mitochondria",
"dna",
"bioenergetics",
"cellular",
"structures",
"and",
"organelles",
"kinetoplasts",
"chromosome",
"biology",
"drug",
"therapy",
"biochemistry",
"trypanosoma",
"eukaryota",
"cell",
"biology",
"nucleic",
"acids",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles",
"organisms"
]
| 2018 | Insights into antitrypanosomal drug mode-of-action from cytology-based profiling |
Paragonimiasis is a food-borne trematodiasis leading to lung disease . Worldwide , an estimated 21 million people are infected . Foci of ongoing transmission remain often unnoticed . We evaluated a simple questionnaire approach using lay-informants at the village level to identify paragonimiasis foci and suspected paragonimiasis cases . The study was carried out in an endemic area of Lao People's Democratic Republic . Leaders of 49 remote villages in northern Vientiane Province were asked to notify suspected paragonimiasis patients using a four-item questionnaire sent through administrative channels: persons responding positively for having chronic cough ( more than 3 weeks ) and/or blood in sputum with or without fever . We validated the village leaders' reports in ten representative villages with a door-to-door survey . We examined three sputa of suspected patients for the presence of Paragonimus eggs and acid fast bacilli . 91 . 8% of village leaders participated and notified a total of 220 suspected patients; 76 . 2% were eventually confirmed; an additional 138 suspected cases were found in the survey . Sensitivity of village leaders' notice for “chronic cough” and “blood in sputum” was 100%; “blood in sputum” alone reached a sensitivity of 85 . 7% . Our approach led to the identification of three previously unknown foci of transmission . A rapid and simple lay-informant questionnaire approach is a promising low-cost community diagnostic tool of paragonimiasis control programs .
Paragonimiasis is a food-borne trematodiasis and belongs to the so-called neglected tropical diseases although today an estimated 21 million people are infected [1] and 293 millions are at risk for infection [2] . Six of the 40 known species of Paragonimus genus [3] may lead to human infection , and may provoke severe and prolonged lung disease: chronic cough , haemoptysis and protracted pleural effusions . Paragonimiasis is endemic mainly in Asia but transmission foci are also known in African and South-American countries . In endemic areas numbers of cases increased recently such as in India [4] , and extra-pulmonary ectopic localizations in skin , liver and brain have been described [5] , [6] , which may lead to severe disease . Recent studies indicate that full recovery of pulmonary disease can not always be achieved with current treatment options [7] . Paragonimus spp . displays a typical trematode life-cycle with two intermediate hosts . Freshwater snails and crabs act as first and second intermediate hosts , respectively . Humans acquire infection by ingesting metacercariae present in flesh of crabs or crayfish . Paratenic hosts such as omnivorous mammals , i . e . wild boars , may transmit the parasite [8] . Dogs and cats are domestic reservoirs . Due to its symptoms of chronic cough and haemoptysis Paragonimus spp . infection is frequently confused with pulmonary tuberculosis and leads to inappropriate treatment [9] . Praziquantel and triclabendazole are efficacious medicines [10] . Lao people possess a deeply culturally rooted habit of raw food consumption including meat , fish , and crabs and crayfish . Transmission of food-borne helminthiasis such as opisthorchiasis [11] or trichinellosis [12] , [13] are widespread . Associated with the widely distributed intermediate crab host it can be assumed that paragonimiasis is much more prevalent than currently acknowledged . Diagnosis is rarely made at village and district level where village health workers and health personnel are largely unaware of this lung disease . We evaluated the performance of a rapid lay-informant questionnaire approach to identify paragonimiasis patients and foci of transmission . A 4-item questionnaire concerning typical paragonimiasis symptoms ( chronic cough , haemoptysis , absence of fever ) was sent through the administrative system to the village leaders . They were asked to notify patients with corresponding symptoms . In a subsequent step the village leaders' information was validated , and the suspected paragonimiasis patients' sputa examined .
Oral informed consent was obtained from all study participants enrolled . Ethical clearance for the study and all procedures was obtained from the Council of Medical Sciences , Ministry of Health , Vientiane , Lao PDR , including the use of oral consent which was documented on a spreadsheet . Oral consent was used as the illiteracy rate is high in rural Lao PDR . Approval for the study was obtained from provincial and district health authorities and village leaders . The study was carried out between February and April 2005 in Hinheub district ( Vientiane province , Lao PDR ) with 23 , 788 inhabitants in 49 villages ( National Census Data 2003 ) . It is a rural , mountainous , multiethnic district approximately 120 km north of Vientiane . The district has one district hospital located at the district capital Hinheub Tay . In 2003 in Naphong village ( approximately 20 km southeast Hinheub Tay ) 12 paragonimiasis cases were diagnosed in 33 patients suffering from chronic cough [14] . Follow-up investigation showed that four Paragonimus species are transmitted in Naphong and two neighbouring villages ( P . heterotremus , P . bangkokensis , P . harinasutai , P . westermani ) by Potamon lipkei and Chulathelphusa brandti crabs [15] P . heterotremus was confirmed in a human infection [16] . A suspected case of paragonimiasis was defined as a patient with a chronic cough of lasting longer than 3 weeks and/or with a sputum with blood ( red brownish colour ) without nocturnal fever . The questions of a 4 item questionnaire were: ( i ) Do you have a cough for more than three weeks ? ( ii ) Do you have bloody sputum ? ( iii ) Is your sputum coloured red or brown ? ( iv ) Do you suffer from nocturnal fever ? The questionnaire was translated to Lao . Pre-tests in Vientiane confirmed the clarity of the questions . The questionnaire was sent from the district health office to the leaders of all villages through the routine administrative system . A cover letter explained the purpose of the questionnaire . Village leaders were requested to ask the 4 questions ( with the help of the village health workers ) in each household , note any patient answering positively to one of the questions i ) to iii ) and to send the filled form back to the district health office . The questionnaire was picked up at the office by the research team . Villages were ranked according to the number of notified chronic cough patients and grouped in quintiles . Ten villages were identified for the validation: from each quintile 2 randomly selected villages . In each village a medical team ( 2 medical doctors ) carried out a door-to-door survey to identify suspected paragonimiasis patients using the same lay-informant questionnaire . In addition they examined all patients notified by the village leader . Suspected paragonimiasis cases of the four largest villages underwent a sputum examination . On two consecutive days three sputum samples were collected per patient: first sputum taken on first contact; a sputum container was given to the patient to collect a second sputum sample early mornings; on contact on the consecutive day the third sputum container was collected . Direct examination of ( unstained ) sputum using a light microscope was performed on site to detect Paragonimus ova . Routine examination ( Ziehl-Neelsen stain ) was eventually performed in the district hospital for each sputum sample to diagnose infection with acid fast bacilli ( AFB , Mycobacterium tuberculosis ) . Paragonimiasis was confirmed if at least one Paragonimus spp . egg was detected in at least one sputum examination . Pulmonary tuberculosis ( TB ) was confirmed if at least one sputum sample was positive for AFB . Paragonimiasis cases were treated with praziquantel ( 3×25 mg/kg/day for 3 days ) according to the Lao National Treatment guideline for district hospitals [17] . Patients diagnosed with TB were referred to the TB national program in the nearby provincial hospital of Phonhong . All data was entered in EpiInfo ( version 6 . 04 , CDC Atlanta ) . All records were cross-checked against original data sheets . Analyses were performed with STATA , version 8 ( Stata Corp . , College Station , TX , USA ) . The questionnaire return rate ( # returned/# distributed questionnaires ) , and the questionnaires' sensitivity , specificity , and positive and negative predictive values were calculated . The survey data performed by medical team and results of sputum examination served as gold standard for village leaders' notification and confirmation of suspected paragonimiasis cases , respectively .
Figure 1 summarizes the study procedures . 91 . 8% of village leaders ( 45 of 49 ) returned the questionnaire . They notified 220 suspected paragonimiasis cases corresponding to 3 . 2% of the population of the ten validation villages ( range per village: 0 . 8%–13 . 8% ) ; 76 . 2% of whom were confirmed by the research medical team visiting the villages . The number of notified suspected cases of paragonimiasis per population unit decreased with increasing village size ( r = −0 . 329 , Figure 2 ) . Additional 138 suspected paragonimiasis cases were identified during the door-to-door validation survey by the medical team , which were not previously notified by the village leaders . The number of newly identified suspected cases varied between the villages ( range: 0–42 ) and increased with increasing population size ( r = 0 . 72 ) . The number of newly identified suspected cases per population was also positively correlated with village population size ( r = 0 . 21 , Figure 2 ) . In total 276 suspected paragonimiasis cases were found of whom 129 patients sputum ( 46 . 7% ) examination could be performed ( Table 1 ) . Seven patients were Paragonimus-egg positive ( 5 . 4% ) and came from three different villages; 5 of whom were diagnosed in Namthome village ( infection rate 10 . 0% ) . AFB was diagnosed in one patient from Namthome village . Table 2 depicts the sensitivity , specificity and predictive values for ( confirmed ) paragonimiasis of the medical assessment of haemoptysis and the symptoms reported in the village leaders' questionnaire . The analysis was done on all subjects ( n = 102 ) with a complete data set . All Paragonimus –egg positive patients suffered from haemoptysis resulting in a 100% sensitivity of haemoptysis assessed by a medical doctor . 34 of 41 suspected paragonimiasis cases with medically assessed haemoptysis were Paragonimus – egg negative ( positive predictive value 17 . 1% ) . Village leaders' diagnostic criteria of “chronic cough” with “blood in sputum” had a 100% sensitivity . Notified cases with “blood in sputum” alone had a lower sensitivity of 85 . 7% . Village leaders missed to report one paragonimiasis patient suffering from haemoptysis . The utilisation of reports with regard to “fever” did not improve the diagnostic performance of the village leaders . When “no fever” was included as a criteria the sensitivity was lower compared to the analysis where the criteria “with or without fever” was used . All positive predictive values were between 9 . 7% and 44 . 4% .
In a paragonimiasis endemic district in Lao PDR we evaluated the performance of a rapid questionnaire approach with the objective to identify new paragonimiasis patients and foci of transmission . We used simple questions on “chronic cough” , “blood in sputum” and “absence of fever” used by village leader to screen the village households . Village leaders were highly responsive to our request . A large number of suspected paragonimiasis patients were notified of whom a considerable three-quarter matched our case definition . However , the village leaders missed half of the suspected cases as the door-to-door survey showed . In the sputum examination we confirmed seven paragonimiasis patients in villages where no cases had previously been diagnosed . In one village 5 patients were found . The village leaders' notification of patients with chronic cough and/or blood in sputum had a high sensitivity . The assessment of “blood in sputum” alone resulted in a 100% sensitivity . Our lay-informant questionnaire approach allowed identifying three new , previously unknown foci of paragonimiasis . In comparison to the high costs of other active case detection methods such as cross-sectional parasitological and/or serological surveys , our approach was performed at very low expenses which were limited to the direct costs of the sputum examination . We conducted the sputum examination in the villages . In a routine implementation of this approach , the diagnosis could be performed at the primary health care level which would further reduce costs . The transport of the questionnaires and screening at village level was performed voluntarily . Our study benefited from a high response rate of the village leaders . Their notification based on inquiries in the villages with the help of the village health workers was reasonably accurate . Three quarters of the suspected paragonimiasis patients could be confirmed . There was a considerable variation in the number of notified suspected cases of paragonimiasis per population unit . Village leaders from smaller villages reported higher rates of suspected cases indicating that in smaller villagers more efforts could be employed in each household to identify suspected persons . This interpretation is in line with the finding that the medical team found more additional suspected cases of paragonimiasis per population in larger villages . Thus , this questionnaire approach has an optimal functioning in areas with small to medium size villages . Additional activities could be proposed to the village leaders such as village meetings with household representatives to overcome this limitation . In our approach no incentives was given to the village leaders or any activity performed to identify the suspected cases . It was surprising that village leaders notify only half of the suspected paragonimiasis patients . “Chronic cough of lasting longer than three weeks” and “blood in sputum” seem to be easily recognisable and detectable symptoms . They should be detected also by lay-person . The reasons for this underreporting are not clear but we observed it also in a larger study in six provinces on tuberculosis cases detection using the same question of chronic cough [18] . We suspect that the highly prevalent symptom of chronic cough is very unspecific and often regarded as banal or even “normal” , and therefore not perceived as worthy to notify . We found that the village leaders report on “blood in sputum” had a higher and in combination of with the presence of “chronic cough” a maximal sensitivity . The inclusion of the presence or absence of reported nocturnal fever did not improve the diagnostic performance . Typically patients suffering from pulmonary paragonimiasis do not have regular fevers [9] , in contrast to tuberculosis patients . For this lay-informant case notification process this symptom was not useful . With this lay-informant questionnaire approach we did not intent to assess the prevalence of paragonimiasis in the community but rather to identify foci of ongoing transmission where later control activities can be instituted , such as information education and communication and/or diagnostic and treatment activities . Therefore , it is a community diagnostic tool applicable for paragonimiasis control . Similar questionnaire approach has been developed for schistosomiasis haematobium [19] . It is based on the macro haematuria ( blood in urine ) reported by school-teachers upon interrogation of their pupils . The approach proved to be highly cost-effective . Today , it is applied in the Schistosomiasis Control Initiative in eight African Countries ( www . schisto . org ) . Our approach on paragonimiasis has the additional advantage that it can be combined with tuberculosis . An earlier study showed that the approach increased the detection rate of tuberculosis [18] . Alternatively , an approach using Geographic Information Systems ( GIS ) in combination with Remote Sensing Techniques ( RST ) could be developed to delineate areas of transmission of paragonimiasis as it has been done for other trematodiasis . Example given , studies on Schistosomiasis japonicum in China showed that the presence of intermediate snail hosts of Onchomelania spp . can be predicted by environmental factors such as water availability , vegetation , altitudes , and temperature [20] , [21] . A similar mapping tool could be developed for predicting the presence of intermediate snail and crab of Paragonimus spp . In combination with information on risky nutritional practices of habit to eat raw foodstuff , accurate risk maps might be developed . The inconveniences of this strategy are the high costs and specialist health professionals involved , both most likely not available locally in Lao PDR and other paragonimiasis endemic areas . In addition , it is not certain that sufficient ecological information on the Paragonimus intermediate hosts are available to guarantee the development of such an approach . Direct sputum examination is known to have a low sensitivity . In particular in children and elderly sputum examinations are often negative or irrelevant due to the fact that good quality sputum are almost impossible to obtain [22] . This fact may explain the relatively large number of Paragonimus –egg negative sputum specimen in patient with haemoptysis identified in the village . Thus , additional , more sensitive sputum examination techniques ( e . g . coloration , centrifugation , or sputum collection over 24 hours ) or serological methods must be carried out [22] . Skin and immunodiagnostic tests based on parasite antigens are highly sensitive [23] . Lately , a molecular diagnostic procedure ( PCR ) has become available [24] . In the wake of increased awareness for neglected tropical diseases in general and the food-borne trematodiasis in particular , paragonimiasis deserves more attention . The current report shows that the development of a simple community diagnostic tool for improved control of paragonimiasis is feasible . | Paragonimiasis is a neglected pulmonary disease provoked by a food-borne trematode parasite . The infection may develop into severe pulmonary disease , often diagnosed with delay and confused with tuberculosis . Globally an estimated 21 millions people are infected . Human infection is acquired through consumption of raw crab , crayfish or wild boar . Typically infections occur clustered in foci of few to several villages where nutritional habits allow transmission . A major challenge for control is to identify the transmission foci . We evaluated a questionnaire approach using lay-informants at the village level to identify paragonimiasis foci and suspected cases . We sent a 4-item questionnaire to 49 village-leaders of a district in rural Lao PDR asking them to report patients with key symptoms of paragonimiasis , i . e . “chronic cough” and “blood in sputum” . The evaluation showed that lay-informants' report had a high sensitivity to identify suspected cases of paragonimiasis using “blood in sputum” as indicator . The approach allowed identifying 3 new , previously unknown foci of transmission in the district . We conclude that lay-informant questionnaires using easily identifiable key symptoms are simple to carry out and are promising low-cost tools for paragonimiasis control . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"public",
"health",
"and",
"epidemiology/epidemiology",
"infectious",
"diseases/helminth",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"infectious",
"diseases/respiratory",
"infections",
"public",
"health",
"and",
"epidemiology/environmental",
"health",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
]
| 2009 | Rapid Identification of Paragonimiasis Foci by Lay Informants in Lao People's Democratic Republic |
Zika virus ( ZIKV ) , an emerging flavivirus , has recently spread explosively through the Western hemisphere . In addition to symptoms including fever , rash , arthralgia , and conjunctivitis , ZIKV infection of pregnant women can cause microcephaly and other developmental abnormalities in the fetus . We report herein the results of ZIKV infection of adult rhesus macaques . Following subcutaneous infection , animals developed transient plasma viremia and viruria from 1–7 days post infection ( dpi ) that was accompanied by the development of a rash , fever and conjunctivitis . Animals produced a robust adaptive immune response to ZIKV , although systemic cytokine response was minimal . At 7 dpi , virus was detected in peripheral nervous tissue , multiple lymphoid tissues , joints , and the uterus of the necropsied animals . Notably , viral RNA persisted in neuronal , lymphoid and joint/muscle tissues and the male and female reproductive tissues through 28 to 35 dpi . The tropism and persistence of ZIKV in the peripheral nerves and reproductive tract may provide a mechanism of subsequent neuropathogenesis and sexual transmission .
Zika virus ( ZIKV ) , once a little-studied member of the family Flaviviridae , forcefully emerged across the Western Hemisphere in 2015–16 . As of October , 2016 , the Centers for Disease Control and Prevention ( CDC ) lists 60 countries worldwide , including the continental U . S . , that have reported autochthonous transmission of the virus , primarily through an Aedes spp mosquito vector ( https://www . cdc . gov/zika/geo/active-countries . html ) . The virus is also believed to be endemic in multiple countries in Africa and Southeast Asia . The World Health Organization ( WHO ) has estimated that 3–4 million individuals will be infected with ZIKV in the next year [1] . While an estimated 80% of all infections are asymptomatic or subclinical , the remaining 20% of ZIKV infections often resembles infection by co-circulating dengue ( DENV ) or chikungunya ( CHIKV ) viruses , with typical symptoms including fever , rash , headache and arthralgia , although ZIKV does appear to have a fairly distinctive association with conjunctivitis [2 , 3] . ZIKV was first isolated in 1947 , from serum taken from a febrile sentinel rhesus macaque ( RM ) in the Zika Forest region of Uganda [4 , 5] . Although reported instances of human ZIKV-associated disease during the 20th century had been sporadic with generally mild disease , large outbreaks were reported in Yap State , Micronesia in 2007 and in French Polynesia in 2013 , resulting in ~900 and 30 , 000 symptomatic cases , respectively [6 , 7] . Illness during these outbreaks was initially characterized as self-limiting and did not require hospitalization . However , 74 patients in French Polynesia who experienced confirmed or probable ZIKV infection later presented with neurological complications . Over half of these were characterized as Guillain-Barré syndrome ( GBS ) ; the remainder included various encephalitides , paraesthesia , facial paralysis and myelitis . Similarly , increases in GBS have been reported in 12 countries worldwide , together with laboratory confirmation of ZIKV infection associated with these cases [8] . Of particular concern is the presumed causal relationship between ZIKV and microcephaly in developing fetuses . In several cases , ZIKV infection of the mother during pregnancy , as well as the presence of ZIKV in the amniotic fluid or tissue of fetuses showing evidence of microcephaly was reported [9–12] . However , the specific mechanisms by which ZIKV causes fetal neurological defects in humans remain unknown . Mouse models , both with and without intact innate immune signaling , of ZIKV infection during pregnancy have proven susceptible to infection of placental and fetal tissue , resulting in intrauterine growth restriction and fetal death [13 , 14] . However , differences in placental architecture and fetal development in the mouse vis-à-vis humans suggest that certain aspects of ZIKV pathogenesis during pregnancy may not be reflected in the murine infection model [15] . Non-human primates , by virtue of their relatedness to humans , are valuable models for the study of human disease . For example , experimental infection of RM with yellow fever virus ( YFV ) results in viscerotropic disease that closely parallels the course observed in humans [16 , 17] . Additionally , we have recently developed a model of CHIKV infection in the RM that recapitulates several aspects of human disease , including joint tropism and inflammation , viremia , and robust innate and adaptive immune responses [18 , 19] . Both DENV [20] and WNV [21] infection of RMs results in detectable viremia and immune response , although infection is not associated with overt pathology . Prior to the current epidemic , the outcome of ZIKV infection in RM model had not been well characterized . However , the fact that the virus was originally isolated from a febrile RM suggests that viral replication , immune response , and aspects of pathogenesis may be modeled in RMs . Here , we report outcomes of infection of adult RMs with a ZIKV strain currently circulating in the Western hemisphere ( a 2015 isolate from Puerto Rico ) . Sub-cutaneous inoculation of animals produced transient , detectable viremia and viruria , as well as clinical symptoms described in human ZIKV infections ( e . g . fever , rash , conjunctivitis ) . These data are comparable to those additional recently published studies of ZIKV infection in RM [22–24] . Herein we also extend these data by examination of tissue tropism during infection . Cohorts of animals necropsied at day 7 , 28 or 35 pi indicated that viral RNA was present within secondary lymphoid tissues , joints , peripheral nervous tissue and organs of the female reproductive tract . Further , a robust immune response , including production of neutralizing antibodies , was observed in all infected animals . Our data suggest that RM may provide a useful model for the study of ZIKV pathogenesis , as well as a platform for the testing of vaccines or anti-viral therapeutics .
All Zika virus infection experiments utilizing animals were performed in compliance with guidelines established by the Animal Welfare Act for housing and care of laboratory animals and conducted in accordance with Oregon National Primate Research Center ( ONPRC ) Institutional Animal Care and Use Committee approved protocol ( IACUC #0993 ) . RM studies were performed in ABSL-3 or ABSL-2 containment facilities at the Oregon National Primate Research Center ( ONPRC ) , which are accredited by the Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) International . Appropriate procedures were utilized in order to reduce potential distress , pain and discomfort . Ketamine ( 10 mg/kg ) was used to sedate the animals during all procedures including routine blood draws performed by trained veterinary staff . Rhesus monkeys were fed standard monkey chow twice daily and the amount was matched to each animal according to body weight , age and sex and intake as monitored . Animals also received daily food supplements and other enrichment devices . The infected animals were caged with partners or caged separately but within visual and auditory contact of other animals in order to promote social behavior . At the designated time points , the animals were euthanized according to the recommendations of the American Veterinary Medical Association 2013 panel on Euthanasia . Zika virus train PRVABC59 was isolated by the Centers for Disease Control ( CDC ) from an individual in Puerto Rico in December 2015 [25] . PRVABC59 was obtained from the CDC , and passaged twice in C6/36 cells ( American Type Culture Collection , ATCC ) . To prepare virus stock , infected C6/36 tissue culture supernatant was concentrated through a 20% sorbitol cushion and titered in Vero cells ( ATCC ) using a focus-formation assay . The virus stock was sequenced and found to conform to the previously described sequence ( Genbank accession #KU501215 . 1 ) with the following four single base pair substitutions: G-1964-T ( Envelope protein V to L; frequency 82 . 7% ) ; T-3147-C ( NS2A protein M to T; frequency 14 . 5% ) ; C-5676-T ( NS3 protein S to F; frequency 36 . 8% ) ; and C-7915-T ( NS5 protein silent mutation; frequency 13 . 5% ) . All cells were cultured in Dulbecco’s Modified Eagle Medium ( DMEM; Corning ) containing penicillin-streptomycin-glutamine ( PSG; Corning ) and 5–10% fetal calf serum ( FCS; HyClone ) . Vero cells were grown at 37°C and C6/36 cells were grown at 28°C . Serial dilutions of virus were plated in 96-well plates seeded with Vero cells , allowed to adsorb for 1 h , followed by overlay with 0 . 5% carboxymethyl-cellulose ( CMC; Sigma ) . At 30 h pi , cells were fixed with 4% paraformaldehyde , washed twice with PBS and blocked/ permeabilized for 1 h in PBS supplemented with 2% normal goat serum ( NGS; Sigma ) and 0 . 4% triton X-100 . Cells were then washed twice with PBS followed by incubation with 0 . 3 μg/ml anti-flavivirus monoclonal antibody 4G2 [26] in PBS supplemented with 2% NGS for 1 h , washed twice more with PBS , incubated with anti-mouse IgG-horseradish peroxidase ( Santa Cruz Biotech ) for 1 h , and washed twice with PBS . Foci were visualized by incubation with the Vector VIP peroxidase substrate kit ( Vector Labs ) according to manufacturer’s specifications and counted using an ELIspot reader ( AID ) . Seven Indian-origin RMs ( 3 females and 4 males ) were divided into three cohorts ( Fig 1 contains a description of the animals within each cohort ) . Cohort 1 was infected subcutaneously with a total of 1x104 , 1x105 , or 1x106 focus forming units ( ffu ) of ZIKV diluted in 1ml of PBS and delivered by ten 100μl injections into the hands and arms bilaterally . These doses are comparable to the 1x105 pfu median dose of the flavivirus West Nile virus ( WNV ) previously determined to be delivered by the bite of infected Culex spp mosquitoes [27] . Cohorts 2 and 3 were similarly infected with 1x105 ffu . Peripheral blood and urine ( collected in the cage pan ) samples were obtained at 0 , 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 14 , 21 , 28 and 35 dpi . Peripheral blood mononuclear cells ( PBMCs ) and plasma samples were separated by centrifugation over lymphocyte separation medium . PBMCs were analyzed for immune cell phenotype and frequency by flow cytometry . Plasma was assessed for viral loads by qRT-PCR and the levels of cytokines by Luminex multiplex-bead based assay , as described below . Urine was assessed for viral RNA by qRT-PCR and for infectious virus by co-culture on C6/36 cells and focus-forming assays using Vero cells . Cohort 1 was euthanized at 28 dpi , Cohort 2 animals at 7 dpi , and Cohort 3 at 35 dpi . Samples of tissues ( joints , muscles , organs , brain , spinal cord , peripheral nerves , glands , and lymph nodes ) and biological fluids ( cerebral spinal fluid , blood , and urine ) were collected and stored in RNAlater , Trizol ( RNA isolation ) , medium ( virus isolation ) as well as fixed and embedded in paraffin . Plasma anti-ZIKV antibody concentrations were measured by end point dilution ELISA . For this assay , high-binding polystyrene 96-well plates ( Corning ) were coated with PBS containing a 1:1000 dilution of 4x108 ffu/ml stock of purified ZIKV particle preparations The plates were incubated overnight at 4°C and then blocked with PBS containing 2% milk and 0 . 05% Tween ( ELISA-Block ) for 1 hr at room temperature . Plates were washed with 0 . 05% Tween-PBS ( ELISA-Wash ) and incubated with two-fold dilutions of RM plasma in ELISA-Block starting at a dilution of 1:50 . The plate was incubated at room temperature for 2 hrs . Plates were washed several times with ELISA-Wash and then incubated with secondary anti-monkey IgM or IgG ( Rockland , Inc . ) conjugated with horseradish peroxidase for 30 mins . Plates were washed with ELISA-Wash and bound secondary antibody was detected using the OPD substrate ( Life Technologies ) followed by HCl stop the assay . The plates were read within 10 minutes using a Synergy HTX Microplate Reader ( BioTek ) at 490nm . Endpoint titers of ZIKV binding antibodies were determined using a Log/Log transformation method and the results were analyzed and graphed using GraphPad Prism v6 software . Neutralization assays were used to measure the concentration of serum that can neutralize 50% of a fixed number of ZIKV ( 50% Plaque Reduction Neutralization Test ( PRNT50 ) ) . Sera from infected RM were serially diluted 4-fold from starting dilutions of 1:10 and following dilution were mixed with an equal volume of ZIKV ( 30–50 PFU ) for final serum dilutions ranging from 1:20 to 1:5120 . Sera and virus were incubated for 1hr at 37°C . The mixtures were added to individual wells of 24-well plates seeded with Vero cells at 90% confluence for 1hr at 37°C on a rocker and then overlaid with 1% methylcellulose in OPTI-MEM ( Gibco ) . Plates were incubated for 3 days at which time the cells were fixed and counterstained with methyl blue . Plaques were visualized and counted using a light box . Raw counts were entered into GraphPad Prism v . 6 . 0 , converted to a percent of mock neutralized input virus and PRNT50 values were calculated utilizing the sigmoid dose-response curve fitting function with upper and lower limits of 100 and 0 , respectively . RNA from tissue samples , blood , urine , and cerebrospinal fluid ( CSF ) was isolated using TRIzol ( Invitrogen ) according to the manufacturer’s protocol . ZIKV RNA levels were measured by a one-step quantitative real time reverse transcription polymerase chain reaction assay ( qRT-PCR ) using TaqMan One-Step RT-PCR Master Mix ( Applied Biosystems ) . 250ng total RNA from tissue samples or 1/10th volume of RNA isolated from 100 μl of liquid samples was used in each reaction . Primers and probes were as follows: Forward: 5’-TGCTCCCACCACTTCAACAA ( ZIKV PRVABC59 genome sequence nucleotides 9797–9816 ) ; Reverse: 5’-GGCAGGGAACCACAATGG; ( complement of nucleotides 9840–9857 ) ; and TaqMan probe: 5’ Fam-TCCATCTCAAGGACGG-MGB ( nucleotides 9819–9834 ) . Forward and reverse primers were used at 250 nM in the reaction , and the probe at 200 nM . Validation of the qRT-PCR assay is shown in S1 Fig . For RNA standards , RNA was isolated from purified , titered stock of ZIKV ( PRVABC59 ) . RNA yield was quantified by spectrometry and the data was used to calculate genomes/μl . Focus-forming units ( ffu ) / μl was calculated based on titer of stock . ZIKV RNA was serially diluted 1:10 into Vero cell RNA ( 25 ng/μl ) and amplified in triplicate using primers and conditions described above . Tissues were homogenized in 1ml of DMEM cell culture medium containing 5% FBS and PSG plus approximately 250μl of SiLiBeads using a bead beater ( Precellys 24 homogenizer , Bertin Technologies ) , and cellular debris were pelleted by centrifugation ( 5 , 000 × g for 2 min ) . A 50μl or 500μl sample of the clarified lysate was applied to one well of a 6-well plate of C6/36 cells for seven days . Supernatant titers from these cultures were transferred to Vero cells , incubated at 37°C for an additional 5 d and assayed for the presence of infectious virus by indirect immunofluorescent staining using mAb 4G2 and an anti-mouse IgG Alexa-488 conjugated secondary antibody . This method proved must sensitive in isolation of infectious virus , as compared to initially culturing tissue samples with Vero cells . Flow cytometry was used to quantify the immune cell phenotype as well as the level of cellular proliferation and activation for peripheral blood mononuclear cells ( PBMCs ) isolated at the time points defined above . The panel of antibodies used for the analysis of innate immune cells consisted of HLA-DR , CD14 , CD11c , CD123 , CD20 , CD3 , CD8 , CD16 , and CD169 . To differentiate between monocyte/macrophages , DCs , and NK cells the following gating strategy was utilized: monocyte/macrophages ( CD3-CD20-CD14+HLA-DR+ ) , myeloid DCs ( CD3-CD20-CD14-HLA-DR+CD11c+ ) , plasmacytoid DCs ( CD3-CD20-CD14-HLA-DR+CD123+ ) , other DCs ( CD3-CD20-CD14-HLA-DR+CD123-CD11c- ) , and natural killer ( NK ) cells ( CD3-CD20-CD8+CD16+ ) . The percentage of activated cells ( CD169+ ) within each subset was calculated as a representation of the cellular activation profile [28] . T cells were analyzed with the following panel of antibodies directed against CD4 , CD8β , CD95 , CD28 , CD127 and for intracellular levels of Ki67 ( proliferation marker ) . The T cell subset was identified as CD4+ or CD8+ and within the CD4+ and CD8+ T cell subsets , the naïve ( CD28+CD95- ) , central memory ( CD28+CD95+ ) , and effector memory ( CD28-CD95+ ) subsets are displayed . B cells were analyzed using the following antibodies: CD3 , CD20 , CD27 , and IgD to delineate naïve ( CD3-CD20+CD27-IgD+ ) , memory ( CD3-CD20+CD27-IgD- ) and marginal-zone like B cells ( CD3-CD20+CD27+IgD+ ) as well as Ki67 to identify proliferating cells . The percentage of proliferating ( Ki67+ ) B and T cells within each subset was calculated as well as for granzyme B ( activation marker ) . The gating strategies and definition of the different cellular subsets were performed as previously described [18] . Phenotyping was performed using an LSRII instrument ( BD bioscience ) and the data was analyzed with FlowJo Software ( TreeStar ) . Monkey Cytokine Magnetic 29-plex Panel ( Luminex Platform Kit from Invitrogen ) was used to quantify cytokine and chemokine expression in blood plasma and CSF samples . According to the manufacturer’s instructions , antibody-conjugated polystyrene magnetic beads were plated onto a 96-well plate and washed with buffer . Beads were incubated with a 7-point standard curve along with 25μl of rhesus monkey plasma or CSF plus 25μl of blocking buffer for 2h . Beads were washed with wash buffer and labeled with the biotinylated detector antibody for 1hr . Beads were washed and then incubated with Streptavidin conjugated to R-Phycoerythrin for 30 minutes and washed . After final wash , cytokines were identified and quantified using a Luminex 200 Detection system ( Luminex ) . Statistical analysis was performed using Sidak’s multiple comparison tests and data was graphed using GraphPad Prism v6 software . Whole blood chemistry analysis was performed on a VetScan VS2 system ( Abaxis , Union City CA ) using the 14-analyte Mammalian Comprehensive Diagnostic Profile Panel ( #500–0038 ) according to the manufacturer instructions . Complete necropsies were performed and tissues were collected for microscopic examination . Tissues were fixed in 10% buffered formalin , embedded in paraffin , sectioned at 5μm and stained with hematoxylin and eosin . In situ hybridization studies were performed on formalin fixed paraffin-embedded ( FFPE ) tissue sections of 5μm using two different Zika-specific commercial RNAscope Target Probes ( Advanced Cell Diagnostics , Hayward , CA; catalog #464531 and #463781 ) complementary to sequences 866–1763 and 1550–2456 , respectively . Pretreatment , hybridization and detection techniques were performed according to manufacturer’s instructions . In the absence of control specimens of Zika virus infected cells/tissues , FFPE brain tissue from mice infected with West Nile Virus [29] were used as positive controls . The target probe #463781 detected WNV infected brain tissue while the probe #464531 did not . Both probes detected ZIKV in the test specimens . Tissue sections were counterstained with hematoxylin . As a negative control , a probe specific for Influenza A virus ( ACD catalog #313241 was used to probe contiguous sections of Zika positive tissues . Spleen and lymph node tissues were dissected at necropsy to produce single cell suspensions by first pushing the tissues through a wire mesh filter followed by extensive washing , and lysis of red blood cells . These cell preparations were put through a 70μm filter and counted prior to being frozen in RPMI containing 50% fetal calf serum plus 10% DMSO in liquid nitrogen . For MACS the cells were thawed , pelleted by low speed centrifugation ( 1 , 500 rpm for 10 minutes ) and resuspended in MACS buffer at a concentration of approximately 2x107 cells/mL . Prior to magnetic separation , cells were passed through a 70μM filter , to remove cell clumps . The cells were then divided into two aliquots of 1x107 cells/mL per tissue type . One aliquot was used to obtain pure populations of CD14+ macrophages and CD3+ T cells and the second to obtain pure populations of CD20+ B cells and CD1c+ dendritic cells . All cell types were isolated using a two-step magnetic isolation method with RM-specific reagents ( MACS , Miltenyi Biotec , Germany ) . For isolation of T cells and macrophages , approximately 1x107 cells were incubated with magnetic beads coated with anti-CD14 ( Miltenyi Biotech ) for 15 min at 4°C . The cells were then washed and resuspended in 500μL MACS buffer before being loaded onto a LD magnetic separation column . After sample loading the column was washed with 2mL of MACS buffer . The magnetically labeled CD14+ macrophages retained on the column were eluted by removing the column from the magnetic field and flushing with MACS buffer using the provided plunger . A portion of the eluted cells were saved for purity analysis via flow cytometry , the remaining eluted cells were pelleted and resuspended in Trizol reagent . The CD14-depleted cells in the flow-through were pelleted , resuspended in MACS buffer and incubated with magnetic beads coated with anti-CD3-biotin antibody ( Miltenyi Biotech ) for 10 min at 4°C , followed by subsequent incubation with anti-biotin MicroBeads ( Miltenyi Biotech ) for 15 min . The cells were then washed and resuspended in 500μL MACS buffer before being loaded onto a MS magnetic separation column for positive selection . Labeled CD3+ T cells , retained on the column , were flushed out of the column with 1mL MACS buffer with the provided plunger in the absence of a magnetic field . Eluted cells were analyzed for purity and vRNA as above . For isolation of CD20+ B cells and CD1c+ DCs , the total cell mixtures were first incubated with a FcR blocking reagent ( MACS ) and magnetic beads coated with anti-CD1c-PE ( MACS ) for 5 mins at 4°C followed by a 15 mins incubation with magnetic bead coated with anti-CD20 ( MACS ) . Cells were then washed , resuspended in 500μL MACS buffer , and loaded onto a LD magnetic separation column . The CD20+ B cells retained on the column were eluted with MACS buffer as described above , and a portion was analyzed for purity via flow cytometry . The remaining eluted cells were pelleted and resuspended in Trizol reagent . To positively select for CD1c+ DCs the B cell-depleted flow-through fraction was pelleted , resuspended in MACS buffer , and then incubated with anti-PE Microbeads ( MACS ) for 15 mins at 4°C . The cells were then washed and resuspended in 500μL MACS buffer before being loaded onto a MS magnetic separation column . The CD1c+ DCs retained on the column were eluted and analyzed as above . Luciferase assays were performed as previously described [30] . Briefly , the firefly luciferase ( LUC ) open reading frame downstream of an NF-κB promoter element was transduced via lentivector ( Qiagen ) into RM fibroblasts whose functional lifespan was extended through the stable introduction of human telomerase . Cells were grown , infected , and treated in 96 well plates as indicated . After adding Steady Glo lysis and luciferin reagent ( Promega ) luminescence was read on a BioTek Synergy plate reader .
Three cohorts of RMs were used for this study ( Fig 1 ) . The first cohort consisted of two adult females and one adult male . Animals were infected with 1x104 , 1x105 , or 1x106 focus forming units ( ffu ) of ZIKV ( PRVABC59 ) . The infectious dose was divided over 10 subcutaneous injections over bilateral hands and arms . Blood and urine were sampled daily through 10 dpi , as well as on 14 , 21 , and 28 dpi . Euthanasia was performed at 28 dpi and tissues collected at necropsy for analysis of viral loads . A second cohort , consisting of two adult RM ( one male , one female ) was infected with 1x105 ffu , followed by daily sampling of blood and urine through 7 dpi , at which time animals were euthanized as above . The third cohort , consisting of two adult male RM , was infected with 1x105 ffu followed by daily sampling of blood and urine through 35 dpi . All animals developed a transient fever , rash on the arms and upper torso , as well as lymphadenopathy of their axillary lymph nodes . Additionally , 3 of 7 animals developed conjunctivitis lasting 3–5 days . None of the infected animals experienced weight loss or signs of clinical disease other than those described above . Analysis of blood chemistry revealed no significant changes following ZIKV infection ( S2 Fig ) . All infected animals developed plasma viremia as detected by RT-qPCR of viral genomes that typically peaked at 2 dpi and was detectable out to 5–7 dpi ( Fig 2A and 2B ) . We were unable to titer virus directly from plasma samples . However , infectious virus , detected by co-culture of plasma with C6/36 cells , was observed in indicated cases between 2 to 4 dpi ( Fig 2A , stars ) . Viral RNA in the urine was detected from 3–10 dpi with peak levels at 5 dpi ( Fig 2A and 2B ) . We detected viral RNA positive urine samples outside of the initial 3–10 dpi window , which is consistent with other reports of ZIKV infections of NHP [22] . This finding indicates that ZIKV infection in NHP is dynamic and remains persistent . Following euthanasia and necropsy , RNA was isolated from individual tissues and the viral genomes were quantified by qRT-PCR ( viral loads of positive tissues in Fig 3A and 3B; complete list of tissues samples in S1 Table ) . At 7 dpi ( cohort 2 , 1x105 ffu ) , viral RNA was detected in multiple tissues: lymphoid tissue—including lymph nodes distributed throughout the body as well as the spleen; joints—most prominently joints near the site of inoculation but in some instances more distal joint tissue as well; peripheral nervous tissue , specifically the sciatic nerve , brachial plexus and trigeminal ganglion . Additionally , viral RNA was found associated with the spinal cord ( cervical , lumbar and thoracic ) , but not in CSF or in the brain , at this time point . This may indicate neurological tropism , but an inability to effect retrograde transport of infectious virus into the CNS or that infection of the CNS requires additional time . Viral RNA was detected in the kidney and bladder of the male animal ( 27679 ) , although not in the testes or prostate . Viral RNA was found in the uterus of the female ( 24504 ) . We were able to co-culture infectious ZIKV in C6/36 insect cells from homogenates of the axillary and inguinal lymph nodes , finger joints , kidney and bladder derived from the male monkey ( 27679 ) ( Fig 3A , black arrows ) . Together these data indicate that ZIKV quickly disseminates to many tissues throughout the body including lymph nodes/spleen , peripheral nerves , and skin as well as the genital/urinary tract . At 28 dpi , ( cohort 1 , inoculated with 1x104 1x105 and 1x106 ffu ) viral RNA was still detected in both lymphoid and joint tissues in all animals ( Fig 3B ) . In general , vRNA tissue distribution was greatest for the 1x105 and 1x106 ffu infected animals . The axillary ( draining ) lymph nodes and spleen showed the highest level of viral RNA in all three animals , while other lymph nodes were positive for at least 2 of 3 animals . Joint tissues close to the site of inoculation ( wrist and finger ) were also positive in all three animals 28 dpi . Additional joints , muscles of the arms and legs , and heart were positive for ZIKV RNA in subsets of animals . In animal 25421 ( female , 1x106 ffu ) viral RNA was detected in the reproductive tissues ( uterus and vagina ) suggesting that the virus can infect these tissues and persist there for at least 4 weeks post infection . This finding may have important implications for viral transmission and fetal infections during pregnancy . Viral RNA was also detected in the sciatic nerve and eyes from this subject . Interestingly , ZIKV RNA was detected in the cerebellum of animal 24561 ( female , 1x104 ffu ) , indicating penetration to the CNS . Co-culture of homogenates from tissues collected at 28 dpi with C6/36 cells did not amplify infectious virus . At 35 dpi , ( cohort 3 , inoculated with 1x105 ffu ) positive viral RNA detection occurred in neuronal tissues , lymph nodes , and joint/muscle tissues ( Fig 3C ) . Animal 26023 displayed extensive neuronal tissue involvement with viral RNA detected in the occipital and parietal lobes of the brain , lumbar region of the spinal cord , dorsal root ganglia , brachial plexus , and eye . In Animal 26021 , ZIKV RNA was not detected in the brain but was present in the trigeminal ganglia , as well as cervical , lumbar and thoracic regions of the spinal cord and peripheral nerves ( brachial plexus and sciatic nerve ) . Interestingly , in situ hybridization on cross sections of sciatic nerve using ZIKV-specific chromogenic probes detected robust virus RNA levels in the perineurial adventitial space from animal 26023 ( Fig 3D ) . Virus was not detected in the nerve fibers . Both animals had viral RNA in their axillary lymph nodes . These results combined with the viral detection data from the day 28 animals confirm the long-term persistence of ZIKV RNA in neuronal , lymph node and joint/muscle tissues . Histologic examination of sections taken from tissues of infected RM found few specific abnormalities , although several areas of inflammation were observed ( S3 Fig ) . An uncharacteristic prostatitis characterized by interstitial neutrophilic and lymphoplasmacytic cellular infiltrates and glandular microabscesses were noted 7dpi in animal 27679 infected with 1x105 ffu ( S3A Fig ) . Minimal perivascular lymphocytic or lymphoplasmacytic inflammatory cell infiltrates were present in sections of skin from the upper torso affected with a rash for both animals examined 7dpi ( S3B Fig ) . Viral RNA was also detected in this area of skin in this animal . Similarly , variable perivascular inflammatory infiltrates composed of lymphocytes , eosinophils and plasma cells were observed in the joints and muscles of animal 24504 ( S3C & S3D Fig ) . Focal lymphohistiocytic inflammation was associated with a meningeal vessel in the cerebrum of the high dose ( 1x106 ffu ) animal 25421 at 28 dpi suggestive of an ongoing infection of the brain vasculature ( S3E Fig ) . Animal 25147 ( 28 d pi , 1x105 ffu ) had focal lymphocytic infiltration of the dorsal root ganglion of the cervical spinal cord ( S3F Fig ) . The lack of correlation of detection of viral RNA with sites of inflammation in the prostate , brain , and DRG may indicate highly focal areas of infection , or clearance of virus prior to resolution of inflammation . In order to determine which cell types within lymphoid tissues were positive for viral RNA , we sorted macrophage , dendritic cell , B-cells and T-cells from the splenocytes and axillary lymphocytes by positive magnetic bead selection ( S4 Fig ) . RNA was isolated from each cell population and ZIKV RNA quantified by qRT-PCR . As shown , at 28 dpi RNA is primarily found in the macrophage and B cell subsets with reduced levels in DC subsets but rarely present in the T cell fractions ( Fig 4A ) . In situ hybridization with ZIKV- and Influenza-specific probes detected ZIKV but not Flu RNA in multiple axillary lymph node follicles from animal #25421 ( Fig 4B ) , confirming the presence of the ZIKV RNA in the macrophage , B cell and DC rich regions of the germinal center . Overall , this data indicates that ZIKV spreads to multiple tissue types and the infection of many of these tissues persists in macrophages , as well as other cell types for at least 4–5 weeks post infection . We performed a detailed phenotypic analysis of immune cell subsets by flow cytometry to characterize activation of innate immune cells ( monocyte/macrophage/DC/NK cells ) as well as adaptive immune cell proliferative responses ( T and B cells ) . We also characterized cytokines and antibodies present in the sera of infected RM . Within 1–2 days pi , all of the animals showed innate immune cell activation , as demonstrated by the presence of CD169+ staining ( Fig 5 ) . RM 24961 ( 1x104 ffu ) displayed a more protracted innate immune response , compared to 25421 ( 1x106 ffu ) , 25147 , 26021 and 26023 ( 1x105 ffu ) . While all animals showed an increase in CD169+ monocytes and DCs at 2–4 dpi , the number of activated cells waned between day 8–10 pi in animals infected with 1x105 or 1x106 ffu , while the number of activated cells in the animal infected with 1x104 ffu did not return to baseline levels until 14–21 dpi . Cytokine expression in the plasma largely did not change following infection . However , expression of 4 cytokines ( IL-1RA , MCP-1-CCL2 , IP-10-CXCL10 , and I-TAC-CXCL11 ) was induced over background levels in the plasma ( Fig 6A–6D ) . Expression of these cytokines was elevated within the first several days post infection but returned to baseline levels by 10 dpi . Low levels of cytokine activation in vivo may be an indirect effect of routine ketamine treatment [31] or a result of direct inhibition by ZIKV of innate immune pathways that direct synthesis and secretion of pro-inflammatory cytokines . To examine the latter possibility we employed a reporter assay for which the readout is luciferase expression that responds to NF-κB or JAK/STAT pathway ( type I IFN ) activation [32 , 33] . As shown in Fig 6E and 6F , rhesus fibroblasts infected with ZIKV for 48h at 5 FFU/cell showed significantly diminished LUC signal relative to uninfected cells following treatment with either poly ( I:C ) or human IL-1β . These represent distinct NF-κB-terminal signaling pathways with poly ( I:C ) induction resulting from activation of the TLR3 pattern recognition receptor and TRIF adaptor protein as well as IL-1β triggering the IL1 receptor and associated MyD88 adaptor protein . Similarly , activation of JAK/STAT signaling by IFNβ1 treatment was also repressed by ZIKV infection ( Fig 6G ) . These results agree with previous observations that ZIKV infection promotes degradation of STAT2 and subsequent inhibition of type I IFN signaling [34] . As such we hypothesize that ZIKV exhibits an inhibitory phenotype that operates downstream of the convergence of these pathways , likely targeting activation of NF-κB itself but delimiting the mechanisms associated with this point will require further experimentation . Proliferating CD4+ and CD8+ T-cells were present in all infected animals by 6–8 dpi . CD8+ T cell proliferative responses ( Ki67+ cells ) were evident at 6 dpi , maximal at 8–9 dpi and returned to background levels by 14 dpi ( Fig 7B and 7D ) . Both central memory and effector memory CD4+ T cell proliferative responses were maximal at 7 dpi ( Fig 7A and 7C ) but took longer to return to baseline levels compared to the CD8+ T cells . Consistent with these findings , Granzyme B expression in naïve and central memory CD4+ and CD8+ T cells peaked between 7 and 10 days post infection ( S5 Fig ) . B cell proliferative burst responses were maximal at 14 dpi ( Fig 7E–7G ) . Interestingly , when comparing the cohort 1 animals , the B-cell proliferative responses in RM 24961 ( 1x104 ffu ) were observed slightly earlier than in RMs 25147 ( 1x105 ffu ) and 25421 ( 1x106 ffu ) and represented a greater percentage of cells within each subset . Proliferating T-cells also appeared for a longer time post infection in this animal ( Fig 7A–7D ) . ZIKV virion-reactive IgM and IgG antibodies in sera were quantified by ELISA . Levels of anti-ZIKV IgM became detectable between 7–10 dpi and were maintained through 28 dpi in 2 of 3 animals , while one animal ( #25421 ) showed reduced titers after 10 dpi . ( Fig 8A ) . IgG levels increased beginning between d8 to d14 pi and plateaued around 21 dpi ( animals #25147 and #25421 ) or continued to increase in the low dose animal ( #24961 ) . Western blotting of ZIKV infected cell lysates revealed that antibody responses targeted at least two proteins that were 38 and 55 kDa , respectively , consistent with viral proteins NS1 and E ( S6 Fig ) , which elicit antibody responses during ZIKV infection of humans and are also major antibody targets during other flavivirus infections [35 , 36] . The neutralizing capacity of the ZIKV-directed antibodies was quantitated , and robust neutralizing antibody responses were detected at 28 or 35 dpi ( Fig 8C and 8D ) in all animals , regardless of the infectious dose .
In 1947 , the Zika virus was originally isolated in Uganda from a febrile RM used as a sentinel in a study of yellow fever virus transmission . Given this initial finding , we hypothesized that ZIKV infection of RM could be used as a model of viral replication , pathogenesis , and immune response . Our results demonstrate that the 2015 Puerto Rico ZIKV strain productively infects the RM , characterized by clinical symptoms comparable to that described in human infection , such as fever , rash , and conjunctivitis . The RM further develops viremia , viruria , widespread tissue infection and a robust adaptive immune response . These data are generally in agreement with recently published studies of RM infected with a 2013 isolate from French Polynesia [22] . Because ZIKV has shown a breadth of tissue tropism not seen in other human flavivirus infections , we sought to characterize distribution of ZIKV in RMs as broadly as possible . As such , our study significantly advances what is known about the tissue tropism of ZIKV on 7 , 28 and 35 dpi , a key feature of infection not examined by previous RM studies at these time points . Analysis of viral genome load in tissues revealed a tropism for lymphoid , joint and peripheral nerve tissue . The apparent persistence of viral RNA in various tissues after resolution of primary viremia is not unique to ZIKV . Persistence of flaviviruses in the infected host after cessation of viremia and recovery from clinical symptoms ( if any ) has been observed in several instances . In humans who have been infected with WNV , prolonged viruria ( in several cases >6 years ) is sometimes observed [37] . This viral persistence may correlate with long-term neurological and renal sequelae [38 , 39] . Models of WNV persistence in RM , mice , and hamsters indicate that virus can persist for months in tissues of the CNS , kidney , and lymphoid organs , and the presence of virus also correlates with long term neurological sequelae and motor neuron loss in hamsters [21 , 40–43] . Herein , we demonstrate that ZIKV RNA was detected in peripheral nerves ( 5 out of 7 RM ) and spinal cord ( 4 out of 7 RM ) and viral RNA can persist in these tissues for up to 5 weeks post infection , which could lead to long-term neuropathology . Indeed , post-recovery complications from ZIKV infection , primarily GBS or other neurologic manifestations , have been documented in the French Polynesian and subsequent South American outbreaks . GBS is characterized by the presence of certain autoreactive antibodies and immune cells and is often associated with previous infection , including viral infection . It is unknown how either acute or chronic ZIKV infection might contribute to GBS or other symptoms , and warrants further study . Additionally , viral RNA was detected in tissues of the male and female reproductive tracts , which may have implications for the association of ZIKV infection of pregnant women with aberrant fetal development and sexual transmission . These findings also correspond to a similar observation of ZIKV RNA in the genital tract of a human female [44] , and may also suggest a mechanism of a recently described instance of female to male sexual transmission [45] . In male RMs , we were unable to detect viral RNA in the testes , which is , perhaps , surprising given the reports of male to female ZIKV sexual transmission . However , we were able to detect viral RNA in the prostate and seminal vesicles , which may represent a potential reservoir and mode of sexual transmission . In addition , the presence of virus in the bladder and urine suggests virus seeding into the semen in the urethra may also be a possible route of transmission . Further intensive study with regard to sexual transmission is clearly warranted . We also note that the challenge route during sexual transmission may affect the biological outcomes to ZIKV infection including pathogenesis . While our studies were designed to mimic mosquito transmission , further studies to elucidate the effect of route of infection on disease are possible in RM . Viral RNA detected in the vagina and uterus of infected females may also be relevant to the association between ZIKV infection during pregnancy and microcephaly/fetal abnormalities . Interestingly , early results from one such study observed prolonged viremia in females infected during the first trimester of pregnancy , leading the authors to speculate that the fetus may be the source of this virus , or that the placenta may serve as a primary reservoir of ongoing ZIKV infection [22] . Our results suggest that other tissues within the infected dam , such as the female reproductive tract , lymphoid or joint tissue may also be considered as a potential source of persistent virus , given that pregnancy is associated with partial immune suppression , the pregnant animal maybe unable to completely clear viral infection . Again , much more intensive analysis of how ZIKV affects pregnancy in RM will be informative . Several recent studies have examined ZIKV infection in vivo using various mouse models . These include ZIKV challenge of mice deficient in type I or type I and II interferon ( IFN ) receptors , in which infection is lethal . Additionally , some strains of immunocompetent mice also appear susceptible to ZIKV infection with regard to displaying detectable viremia and susceptibility of fetuses to developmental defects , although the mechanisms underlying the outcomes in this model are unclear . In type I IFN receptor deficient mice ( IFNAR-/- or A129 ) viral tropism appears broader than we observe in the infected RM . Notably , high viral loads were observed in the brain and testes of these mice , tissues that in the RM , had undetectable or minimal levels of viral RNA . These results suggest that restriction of viral tropism in vivo may be due to the action innate immune factors . Therefore , robust and highly relevant animal models of disease are critical to advancing our understanding of ZIKV disease pathogenesis , immune responses , and potential vaccination and antiviral strategies and these will include both mouse and nonhuman primate models . Although it is a flavivirus , ZIKV clinical and virologic features are distinct and strikingly different from other flavivirus infections , challenging pre-existing assumptions about how this virus behaves in an intact host . Results presented here establish the outstanding potential of the RM ZIKV model for dissecting many of these unique features—specifically effective infectious dose , tissue tropism , fluid compartment infection , and chronicity of infection—in a host that both naturally develops disease and has a fully intact immune system . The magnitude and ongoing expansion of the current ZIKV outbreak calls for rapid initiation of more comprehensive studies to further validate and expand this RM model as well as begin to explore specific in vivo questions that are critical to controlling the ZIKV epidemic . | Although it was first identified almost 70 years ago , Zika virus had rarely been associated with pathology in humans until the 21st century . Recent outbreaks in the South Pacific and the Americas have been characterized by numerous confirmed cases , some involving neurologic sequelae and , of most concern , birth defects following infection of pregnant women . Here , we present the results of experimental infection of adult rhesus macaques with a strain of Zika virus isolated during the recent epidemic in the Western hemisphere . Following infection , Zika virus was detected in the sera and urine of all infected animals . Further , we detected virus in multiple tissues of infected animals as late as 35 days post infection , indicating viral persistence . The apparent tropism of the virus for tissues of the peripheral nervous system as well as the reproductive tracts of males and females has implications for the further characterization of the mechanism ( s ) of Zika virus pathogenesis . Additionally , this model provides a platform for development and testing of preventative or therapeutic interventions to combat the emergence of this virus . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"body",
"fluids",
"immune",
"physiology",
"pathogens",
"immunology",
"microbiology",
"viruses",
"urine",
"lymph",
"nodes",
"rna",
"viruses",
"lymphatic",
"system",
"antibodies",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"t",
"cells",
"immunoassays",
"blood",
"plasma",
"antibody-producing",
"cells",
"biochemistry",
"blood",
"cell",
"biology",
"flaviviruses",
"anatomy",
"b",
"cells",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms",
"zika",
"virus"
]
| 2017 | Zika Virus infection of rhesus macaques leads to viral persistence in multiple tissues |
Coordinated cell migration during development is crucial for morphogenesis and largely relies on cells of the neural crest lineage that migrate over long distances to give rise to organs and tissues throughout the body . Recent studies of protein arginylation implicated this poorly understood posttranslational modification in the functioning of actin cytoskeleton and in cell migration in culture . Knockout of arginyltransferase ( Ate1 ) in mice leads to embryonic lethality and severe heart defects that are reminiscent of cell migration–dependent phenotypes seen in other mouse models . To test the hypothesis that arginylation regulates cell migration during morphogenesis , we produced Wnt1-Cre Ate1 conditional knockout mice ( Wnt1-Ate1 ) , with Ate1 deletion in the neural crest cells driven by Wnt1 promoter . Wnt1-Ate1 mice die at birth and in the first 2–3 weeks after birth with severe breathing problems and with growth and behavioral retardation . Wnt1-Ate1 pups have prominent defects , including short palate and altered opening to the nasopharynx , and cranial defects that likely contribute to the abnormal breathing and early death . Analysis of neural crest cell movement patterns in situ and cell motility in culture shows an overall delay in the migration of Ate1 knockout cells that is likely regulated by intracellular mechanisms rather than extracellular signaling events . Taken together , our data suggest that arginylation plays a general role in the migration of the neural crest cells in development by regulating the molecular machinery that underlies cell migration through tissues and organs during morphogenesis .
Coordinated cell migration during development is crucial for tissue and organ morphogenesis from early gastrulation to adulthood . The largest cell populations that are capable of long-range migration at different developmental stages originate from the neural crest lineage . Neural crest cells are of mesenchymal morphology and migrate from the trunk into different areas of the developing embryo . These cells express a distinct subset of markers , including Wnt1 and others [1]–[8] , at or before the onset of migration . Recent studies of protein arginylation demonstrated an essential role of this poorly understood posttranslational modification in mammalian embryogenesis and suggested that arginylation is a previously unknown major signaling mechanism that regulates multiple physiological pathways . Knockout of arginyltransferase ( Ate1 ) in mice leads to embryonic lethality and severe cardiovascular defects , including abnormal heart septation , underdeveloped myocardium , and impaired angiogenesis [9] . Remarkably , all these phenotypes resemble the phenotypes seen in the mouse models with knockouts of various genes implicated in cell migration , leading to the hypothesis that the mechanisms underlying cell migration may be the primary targets for regulation by arginylation ( see [10] for review ) . It has been found that a large number of proteins in vivo are arginylated [11]–[23] , including a prominent subset of cytoskeletal targets that play direct mechanistic roles in cell migration . Arginylation of beta actin in cultured fibroblasts regulates lamella formation and the structure of the cell leading edge [24] . Other proteins involved in cell adhesion and migration , such as talin , spectrin , filamin , myosin , etc . are also arginylated in different mouse tissues [23] . All these data suggest that arginylation may be a general mechanism of the regulation of cell movement in different physiological events , however the role of arginylation in directional cell migration in culture and during embryonic development has never been studied before . To test the hypothesis that arginylation regulates cell migration during morphogenesis , we produced and analyzed a conditional knockout mouse model with Ate1 deletion driven by neural crest-marking Wnt1 promoter ( Wnt1-Ate1 mouse line ) . These mice exhibit perinatal lethality and severe morphogenesis defects resulting from poorly developed neural crest-derived structures , suggesting that Ate1 indeed regulates the migration of neural crest cells that give rise to these structures in embryogenesis . Studies of cell migration patterns in embryos and in culture show that Ate1 knockout results in an overall delay in the migration , likely regulated at the intracellular level , and that Ate1 knockout cells co-cultured with wild-type tend to ‘ride’ on the migrating cells rather than move on their own . Taken together , our data indicate that arginylation regulates tissue and organ morphogenesis by affecting the intracellular mechanisms that drive the migration of the mesenchymal cells of the neural crest lineage .
To produce an Ate1 conditional knockout we first generated an ‘Ate1-floxed’ mouse line with the first three exons of the Ate1 gene flanked by LoxP sites ( Figure S1 ) . We have previously shown that exons 1 and 2 are essential for the formation of all four Ate1 isoforms [25] and that deletion of the region encoded by these exons from the Ate1 sequence leads to the abolishment of Ate1 activity in yeast complementation assays [26] . Control experiments ( data not shown ) confirmed that deletion of the exons 1–3 using the Cre-driven recombination in the Ate1-floxed line resulted in embryonic lethality similarly to the previously described Ate1 knockout [9] , [27] . To produce a neural crest-specific Ate1 knockout , we crossed the Ate1-floxed line with a commercial mouse strain ( Wnt1-Cre ) where Cre recombinase is expressed under neural crest-inducing Wnt1 promoter , resulting in Ate1 deletion in subsequently derived neural crest and some other cell types . These mice , termed Wnt1-Ate1 mice , were used in the present study . To confirm the efficiency of the Ate1 knockout in these mice , we analyzed the Ate1 protein expression by immunohistochemistry of the sagittal sections of E12 . 5 and E16 . 5 Wnt1-Ate1 embryos probed with rat monoclonal antibody that recognizes all four Ate1 isoforms ( see Figure S2 for antibody characterization and Figure S3 and Figure S4 for embryo staining ) . In control embryos Ate1 expression was observed in most tissues and organs ( data not shown ) , consistent with our previous data that Ate1 is expressed throughout the embryo [9] , [25] . In contrast , Ate1 expression in the conditional knockout mice was prominently excluded from the midbrain region and the enteric nerves , as well as parts of the peripheral nervous system ( Figure S3 and Figure S4 ) , the structures that are derived from Wnt1 expressing cells [6] , [28] . No prominent areas with missing Ate1 expression were observed anywhere else in the embryo , however it is possible that Wnt1-expressing Ate1 knockout cells in other areas ( such as , the somite regions along the back ) mixed with other cell populations , making them difficult to detect . It is also possible that in such areas the normal levels of Ate1 are reduced , making the knockout cells poorly distinguishable from the background . To address this possibility and further confirm the efficiency of the Cre transgene expression , we crossed Wnt1-Cre mice to the R26R Rosa reporter mouse strain , in which prominent LacZ expression occurs in all Cre-expressing tissues [29]–[32] . X-gal staining of Wnt1-Cre embryos at E9 . 5 showed that Cre expression occurred as expected , with the majority of staining observed in the head and the somite region ( Figure S14 ) . Unlike the complete Ate1 knockout mice that die just past mid-gestation ( E12 . 5–E14 . 5 ) [9] , [27] , Wnt1-Ate1 mice survived until birth and exhibited perinatal lethality ( Table 1 ) . Over 60% of the pups died on the day of birth ( P0 ) , and another 24% died during the following 3 weeks , with less than 13% of mice surviving to adulthood . Such a variability in the mortality rate could be explained by the variability in the migratory patterns of neural crest cells between individual embryos ( see below ) . The conditional knockout mice that died at P0 had severe breathing problems . Unlike their littermates they breathed frequently and irregularly and appeared to be gasping as if from the lack of air ( Video S1 ) . Within a few hours , these mice became bloated , with abnormally enlarged stomachs , and died ( Figure 1A , left ) . Post-mortem dissection showed large amounts of air in their stomach ( Figure 1A , right ) , suggesting that the air was misdirected there from the respiratory tract , possibly through the digestive system , and accumulated due to extensive breathing . Wnt1-Ate1 mice that died at later stages after birth ( mostly within the first week , Table 1 ) showed no visibly abnormal breathing , however these mice appeared severely retarded , with delayed growth and significantly smaller body size compared to their littermates ( Figure 1B ) and apparent inability to actively move and explore the surrounding space ( Video S2 ) . Their appearance and behavior could be explained by malnutrition caused by non-lethal alterations in their ability to breathe and feed compared to their wild-type littermates . It is also possible , however , that these defects were caused by additional physiological and/or neurological abnormalities , caused by Ate1 deletion in other neural crest cell populations . The remaining 12 . 8% of Wnt1-Ate1 mice were able to grow to adulthood . These mice were fertile and able to produce healthy , surviving pups , however they were consistently smaller than their littermates and had visible facial abnormalities ( short snouts and abnormally shaped skulls , Figure 1C ) . Thus , despite the variations in the severity of the Wnt1-Ate1 phenotype , over 87% of the conditional knockout mice died at or soon after birth , and all of them exhibited varying degrees of defects . It has been previously found that mouse knockouts of the genes implicated in neural crest cell migration and neural crest-dependent morphogenesis are often accompanied by the defects in palate and other craniofacial structures that originate from the neural crest cell lineage [33]–[36] . Palate defects in particular are known to correlate with breathing problems . It is hypothesized that shortened or cleft palates affect the separation between the digestive and respiratory tract , causing the inhaled air to be directed to the stomach rather than the lungs and resulting in air accumulation in the stomach and lethality similar to that seen in Wnt1-Ate1 mice at P0 [35] , [37] , [38] . To test whether Wnt1-Ate1 mice have defects in the palate and/or surrounding structures , we performed postmortem analysis of the pups that died at P0 with breathing abnormalities , by removing the lower jaw and visually analyzing the throat and the roof of the oral cavity . Several defects were observed during this analysis . First , the majority of the analyzed mutant pups had an abnormally large entrance to the nasopharynx and a reduced area normally covered by the soft palate ( denoted by two shorter perpendicular arrows and one longer arrow in Figure 2A , respectively ) . In control animals ( Figure 2A , left ) the entrance to the nasopharynx was tightly shut and resembled a vertical slit surrounded by a small area of soft tissue . In the mutant animals ( Figure 2A , right ) , the nasopharynx entrance appeared partially opened , assuming a triangular shape with a larger surrounding area , suggesting that this structure was prevented from closing either because of being structurally defective , or because the musculature that controls it did not operate properly . Closer observations ( Figure 2B ) showed that while in control mice ( left ) the entrance to the nasopharynx was partially closed by the soft palate , in the mutants ( right ) the soft palate in this area was either short ( not shown ) or missing ( Figure 2B ) , leaving a gap that would be expected to permit the air to travel unrestrictedly throughout the passages connected to the oral cavity in this area . To confirm that the defects we see are indeed palate defects , we fixed the newborn Wnt1-Ate1 pups and their control littermates and sectioned them sagittally through the middle , in the area where the palate appears as the horizontal line separating the oral and the nasal cavity . Consistent with our morphological observations , palates in all the conditional knockout pups analyzed at P0 appeared short , unable to reach the area of the throat where the separation between the trachea and the esophagus occurs ( Figure 2C ) . The average distance between the end of the palate and the epiglottis in the mutant mice was approximately 10 times larger than that in control ( Figure 2D ) . Thus , Wnt1-Ate1 knockout mice , similar to other mouse models with neural crest migration defects , have severe malformations of the soft palate that likely cause the breathing problems and misdirected air flow resulting in early postnatal lethality ( Figure 1A and Video S1 ) . Since Wnt family genes , in addition to the neural crest , have been implicated in the functioning of other organs , including lungs [39] , [40] , we tested whether , in addition to the palate , other defects in Wnt1-Ate1 mice may contribute to the breathing problems and/or early lethality . Histological examination of the sagittal sections of the Wnt1-Ate1 embryos at P0 revealed no prominent abnormalities in the major organs or structures throughout the body ( data not shown ) . Lungs in the Wnt1-Ate1 mice appeared collapsed compared to the wild-type , with little or no air in the alveolae and hemorrhaging in the air passages , however no visible lung defects were observed in the E16 . 5 , E18 . 5 and E19 . 5 embryos recovered before birth that did not have a chance to breathe ( data not shown ) . Therefore , given that Wnt1-Ate1 pups have breathing abnormalities that may indirectly affect the lungs , it appears unlikely that these mice have an independent structural lung defect . Since neural crest cells contribute to the formation of bone and cartilage in the head , we next examined the skeletons of newborn Wnt1-Ate1 and control mice at P0 by staining with alizarin red S and alcian blue 8GS that interact with bones and cartilage to color them red and blue , respectively [41] ( Figure 3 ) . While the overall bone structure and skeletal architecture appeared normal in Wnt1-Ate1 mice ( data not shown ) , prominent abnormalities were seen in the development of the frontal bones , the neural-crest-derived parts of craniofacial skeleton that contribute to the top of the skull [32] , [42] . In control mice , frontal bones came close together , leaving only a narrow slit along the top of the skull ( Figure 3A , top images ) . In contrast , in the mutant mice , frontal bones appeared smaller and narrower and were unable to meet on top of the skull , leaving a wide gap that exposed the cranial cavity beneath ( Figure 3A , bottom images ) . Measurements of the ratios between the width of the gap and the width of the skull showed that in the mutant the gap occupied on average almost 1/3 of the skull width , an area almost 4 times larger than in wild-type ( Figure 3B ) . This defect was observed in all the analyzed mutants , even those that did not exhibit breathing defects at birth . To test whether frontal bone defects are also seen in the Wnt1-Ate1 mice that survive to adulthood , we euthanized several adult Wnt1-Ate1 animals and matching wild-type controls and stained their skeletons similarly to the way described above for the newborn mice . Consistent with the defects seen in the newborns , adult surviving Wnt1-Ate1 mice had short , deformed frontal bones that had altered shape and rougher outline and often appeared incompletely closed ( arrows in Figure 3C ) and unable to meet in the middle ( arrowheads in Figure 3C ) . Nasal bones also appeared shortened , leading to the overall shortening of the snouts as seen in the intact animals ( Figure 1C ) . Therefore , some neural crest-derived parts of the craniofacial skeletons are affected in Wnt1-Ate1 mice regardless of the severity of other phenotypic changes seen in these mice . Since , in addition to craniofacial structures , Wnt1-Ate1 mice also show prominent Ate1 deletion in the enteric neurons and some deletion in the peripheral nervous system ( Figure S3 and Figure S4 ) , we tested whether the mutants have abnormal distribution of peripheral nervous system structures and gut neurons that might indicate defects in gut innervation often associated with neural crest-related developmental abnormalities ( reviewed in [43]–[45] ) by staining embryo sections and whole mount guts excised from E16 . 5 Wnt1-Ate1 embryos with antibody to the neuron projection marker beta-III tubulin . This staining revealed no abnormalities in the peripheral nervous system ( Figure S4 ) or gut neuronal network ( Figure S5 ) , suggesting that the peripheral nervous system and enteric neurons , despite being Ate1-deficient , were able to position normally during embryogenesis . To address the question whether Wnt1-Ate1 mice exhibit any defects in neural crest cell migration , we used the Wnt1-Ate1-R26R reporter conditional knockout line and analyzed the distribution of LacZ -expressing cells in wild-type and Wnt1-Ate1-R26R embryos at E9 . 5 . Several litters , whose embryonic stage was determined by counting the somites , were analyzed , and the comparisons were made between embryos with comparable somite numbers . In the analyzed embryos , the somite numbers ranged between 22 and 26 in wild-type and between 21 to 28 in Wnt1-Ate1 , consistent with the expected somite numbers at this stage . While the X-gal staining of the migrating neural crest cells in these embryos , especially with the larger somite numbers , was heavily masked by staining of other organs originating from Wnt1-expressing cells , such as midbrain ( Figure S14 ) , cell migration patterns could be clearly observed in the somite regions and near the pharyngeal arches , where the migrating cells appeared as streams of LacZ-expressing ‘dots’ arranged in different patterns from the back to the front of the embryo ( Figure 4 ) . In wild-type ( Figure 4A and 4B ) , prominent populations of cells migrated from back to front as continuous lines ( from the third pharyngeal arch down to the somites ) and as triangular ‘streams’ directed toward the third and fourth pharyngeal arches and along each of the somites . In Wnt1-Ate1 , several of these migration patterns were affected with different degrees of severity . Embryos with 24+ somites had altered cell distribution in the streams migrating toward the third and fourth pharyngeal arches and in the line migrating over the upper area of the trunk , which appeared diffuse , with fewer cells present in those areas ( Figure 4C and 4D ) . In some embryos ( Figure 4E ) , migration toward the pharyngeal arches appeared normal , but X-gal staining in the upper trunk area appeared weaker than control , indicating a reduced number of migrating cells in that area . In the embryos at a slightly earlier developmental stage ( 21–23 somites , Figure 4F–4H ) these differences were more obvious , resulting in much lower overall levels of the X-gal-stained cells visible in these areas . Such extremely affected embryos also appeared smaller than wild-type embryos or knockout embryos with less severe phenotypes ( see Figure S14 ) . To further confirm this result , we performed in situ hybridization of whole mount embryos for the neural crest marker Sox10 . Analysis of embryos at E9 . 5 and E10 . 5 showed similar changes in the migration patterns in Wnt1-Ate1 mice compared to control ( Figure 4I and 4J and Figure S15 ) . Finally , to obtain an independent confirmation of altered migration in Ate1 knockout neural crest cells , we isolated neural crest explants from E8 . 5 Wnt1-Ate1-R26R mice , and observed the migratory patterns of LacZ-expressing cells after incubation for 48 hourrs in culture ( Figure S6 ) . While in control explants ( Figure S6 , left panels ) , many of the LacZ-expressing cells during this time emigrated from the original cell mass and traveled to the periphery of the expanding explant as groups or individual cells , in the knockout LacZ-expressing cells traveled to a shorter distance as streams of cells mostly connected to the original explant , without venturing out on their own ( Figure S6 , right panels ) . All of the examined explants behaved consistently with each other , suggesting that this altered migration pattern occurs universally in response to Ate1 knockout . Therefore , consistent with the situation in situ , Ate1 knockout in the neural crest cells results in impairment of their migration in culture . Since most of the developmental defects observed in Wnt1-Ate1 knockout mice are related to the size reduction of the structures that are normally derived from the neural crest cells ( such as palate or frontal bones ) , it is possible that in addition to decreased cell migration some of these defects are due to other reasons , such as decreased proliferation or increased apoptosis in the Ate1 knockout neural crest cells . To test for possible increase in apoptosis , we used TUNEL assay to stain E9 . 5 Wnt1-Ate1 embryos ( Figure S7 ) or E12 . 5 Ate1 knockout embryos ( not shown ) and found no differences in the amount or distribution of the stain in the wild-type and knockout littermates . We also stained sections of Wnt1-Ate1 embryos with antibodies to the apoptotic cell marker cleaved caspase 3 and found no differences in the staining intensity or patterns between wild-type and Wnt1-Ate1 embryos ( see Figure S8 for representative images ) . To test for possible changes in cell proliferation , we stained embryo sections for the cell proliferation marker phospho-histone H3 and found no differences in the staining intensity or distribution between wild-type and Wnt1-Ate1 embryos ( Figure S9 ) . To further test this possibility , we isolated neural crest explants from E8 . 5 and E9 . 5 Ate1 knockout littermate embryos and , after two days in culture , labeled the actively proliferating cells with 5-bromo-2-deoxyuridine ( BrdU ) . No difference was found between the numbers or distribution of actively proliferating cells in wild-type and Ate1 knockout explants ( Figure S10 ) . These results suggest that Ate1 knockout does not cause changes in the rates of cell proliferation or apoptosis during embryogenesis , pointing to the fact that the defects observed in Wnt1-Ate1 knockout mice are due primarily to the impairment in neural crest cell migration . To further characterize impairments in cell migration induced by Ate1 knockout , we analyzed the motility of cultured Ate1 knockout fibroblasts , derived from the back portion of the E12 . 5 Ate1 knockout embryos and immortalized by continuous passaging in culture as described in [9] , [24] . This model was chosen as one of the closest cell culture models of neural crest cell migration . Indeed , these fibroblasts are mesenchymal cells derived from the back portion of the embryos where the majority of the migrating neural crest cells are also found , and they are morphologically indistinguishable from the cells composing the neural crest explants in culture . Therefore , such fibroblasts can be expected to behave in the same manner as the neural crest cells in terms of their motile properties , and using these cells as a model in culture makes it possible to perform much more detailed tests than possible with neural crest explants . Ate1 knockout fibroblasts have been previously shown to have a defective lamella [24] , however no changes in the speed or directionality of their migration have been reported . To test whether these cells move slower than wild-type , we performed scratch wound assays by removing an area of a dense cell monolayer and taking time lapse images of cells moving from the dense area to the scarce . The observations were performed over long periods of time similar to the estimated duration of many neural crest-dependent migratory events in development ( 10–15 hours ) . While individual Ate1 knockout cells during shorter stretches of time were capable of moving at speeds similar to wild-type ( not shown ) , the migration speed of the Ate1 knockout cell monolayers over continuous periods of time was nearly 4 times slower than that of wild-type cells ( average speed of 11 . 5 µm/h compared to 41 . 3 µm/h in wild-type ) ( Figure 5 , Video S3 and Video S4 ) . This difference constitutes a significant delay in the overall migration , and , when transferred to an in vivo environment of a developing embryo would be likely to create significant morphogenic defects . Cell migration in culture and in situ is mediated by attachment to the substrate and forming a connection between the intracellular actin cytoskeleton and the extracellular matrix via focal adhesions . To test whether the slow migration speeds in Ate1 knockout cells were due to their impaired adhesion on the intra- or extracellular side , we first tested whether these cells are capable of creating a local extracellular environment that favors migration . To do this , we stained wild-type and Ate1 knockout non-permeabilized cell monolayers for fibronectin , a major extracellular matrix component that directs the migration and adhesion of the mesenchymal cells and is secreted by these cells in culture and in situ ( reviewed in [46] ) . No differences were found in the amount or distribution of fibronectin per area in each culture ( Figure 6A ) , suggesting that Ate1 knockout cells are capable of creating and utilizing the same local extracellular environment as wild-type . Thus , the migration defects in Ate1 knockout cells do not originate at the local extracellular level . We next tested whether the intracellular defects in Ate1 knockout cells could be responsible for slower migration speeds . To do this , we compared the number of focal adhesions ( the structures that connect cells to the extracellular matrix and thus are responsible for cell movement ) at the edge of the cell monolayer migrating into the wound in wild-type and Ate1 knockout cells by staining these cells with three different focal adhesion markers , paxillin , talin , and focal adhesion kinase ( FAK ) ( Figure 6B and Figure S11 ) . While in wild-type cultures cells moving into the wound developed prominent focal adhesions that appeared firmly anchored to the substrate ( Figure 6B , top panel , and Figure S11 , left panels WT ) , in Ate1 knockout cells focal adhesions appeared smaller and scarcer , sometimes difficult to detect ( Figure 6B , bottom panel , and Figure S11 , right panels ) . This decrease in focal adhesion size and number was accompanied by a reduced level of paxillin but did not correlate with the changes in talin and FAK protein level ( Figure S11 , middle panels KO ) , suggesting that the decrease in focal adhesion constituted a genuine structural defect rather than a secondary effect of down-regulation of any of these markers . Manual counting of the number of prominent elongated paxillin-containing structures per µm of the wound edge , or the area of talin-containing structures per leading edge in each cell , revealed an almost 10-fold reduction in the focal adhesion area and number in Ate1 knockout cells compared to wild-type ( Figure 6B and Figure S11 , right panels ) . Such a prominent difference is highly likely to severely affect both the cell's ability to attach to the substrate and the resulting cell migration . Thus , Ate1 knockout-dependent impairment in cell migration likely originates at the intracellular level . To further study the focal adhesion defects in Ate1 knockout cells we quantified the total area of talin and paxillin-containing focal adhesions in single wild-type and Ate1 knockout cells , and found that , consistent with the results obtained in the moving cell monolayer , the focal adhesion area in single cells is significantly reduced ( Figure S12 ) . This result suggests that the focal adhesion defects in Ate1 knockout cells are the intrinsic property of these cells and not the secondary result of the impaired movement of these cells along the substrate . Finally , to corroborate this observation with the situation in situ , we stained embryo sections for talin and quantified the staining intensity in the migrating subpopulations of neural crest cells at the sides of the neural tube ( Figure S13 ) . While this assay measures not only talin localized to the adhesion sites but the total talin level in the tissue ( which is similar in wild-type and knockout , as seen in Figure S11 ) , the un-localized talin is expected to show a reduced , more diffuse signal compared to talin bound at the focal adhesion sites . Consistent with this hypothesis and the results of focal adhesion quantification in culture , the talin staining in situ was reduced in Wnt1-Ate1 embryos by a small but statistically significant number ( Figure S13 ) , suggesting that migrating neural crest cells in these embryos have focal adhesion defects that may prevent them from proper attachment and movement through the tissues during embryogenesis . Unlike in the complete knockout , when cells migrate in situ in a conditional mouse model , in many cases knockout cells are found in the immediate vicinity of the wild-type and are expected to co-migrate during different developmental events . Therefore , the straight comparison between wild-type and Ate1 knockout cell movements does not necessarily reflect the complex situation in which these cells co-exist in situ , potentially affecting each others' migration . To test the motility of Ate1 knockout cells in the environment of the surrounding wild-type cells , we performed wound healing migration assays of wild-type cells co-cultured with Ate1 knockout cells labeled with stably transfected GFP . Control experiments showed that GFP-expressing cells moved at rates similar to the regular Ate1 knockout cells when cultured separately from wild-type ( data not shown ) . After the scratching of the wound the wild-type cells recovered first; even though areas with GFP-expressing cells at the edge were chosen for observation , shortly after the start of the time lapse , wild-type cells moved to the front and continued leading the way for the entire observation period ( Figure 7A and Video S5 ) . Ate1 knockout cells moved prominently slower than wild-type , however , unlike in individual cultures , co-cultured Ate1 knockout cells showed a less significant difference in migration speed from the wild-type cells ( around 2-fold , Figure 7B ) and were able to cover larger distances . This observation suggests that wild-type cells are capable of aiding the knockout cells along to cover larger distances over a period of time . Observations of the time-lapse images ( Video S5 ) suggested that rather than moving , Ate1 knockout cells , when possible , ‘rode’ on the expanding wild-type monolayer , possibly due to their poorer substrate attachment , and that their ability to cover greater distances was likely due to passive , rather than active movement . Reciprocal assay , with GFP-labeled wild-type cells co-cultured with unlabeled knockout cells , confirmed that , independently of the GFP transfection , wild-type cells indeed migrated faster and filled the wound sooner than the knockout ( Figure 7C ) . It should be noted that in the co-culture assay the wild-type cells on average moved somewhat slower than when cultured individually . A possible explanation for this effect could be that in this assay we did not wait for the edge of the monolayer to recover after the scratch wound and the delay may be simply due to the additional time the dislodged cells at the edge needed to re-attach and polarize . Thus , Ate1 knockout in the mesenchymal cells results in greatly delayed migration speeds that originate at the intracellular level and are likely to result in severe morphogenic defects in vivo .
Our data show that knockout of the arginyltransferase Ate1 in the cells of the neural crest lineage results in multiple morphogenic defects and perinatal lethality in mice . It has been previously shown that complete Ate1 knockout in mice leads to embryonic lethality and defects in cardiovascular development and angiogenesis [9] that are reminiscent of the defects seen in mouse models with knockout of genes implicated in cell adhesion and migration during embryogenesis [10] . Here we show for the first time that Ate1 deletion in the migratory subpopulations of the neural crest cells leads to delayed development and reduced size of the neural crest-derived organs and tissues , suggesting that Ate1-dependent migration of the neural crest cells is essential for normal embryogenesis . We have previously shown that Ate1 knockout embryonic fibroblasts have leading edge defects that arise from abnormalities in the non-arginylated actin cytoskeleton [24] . Here we found that in addition to the abnormal leading edge Ate1 knockout cells also have impaired adhesion to the extracellular matrix and that the rate of their motility is significantly delayed due to the impaired intracellular machinery rather than to the extracellular environment . It is possible that the defects seen in Wnt1-Ate1 embryos may have additional underlying mechanisms , including impaired epithelio-mesenchimal transformation that could hinder the size of the migratory cell population , or abnormal responses to the signals that coordinate the migratory events on the organismal level . However , the intracellular changes observed in response to Ate1 knockout are likely sufficient to induce major morphogenic defects even without perturbations in other important developmental processes . Wnt1-Ate1 mice show defects of varying severity , with a small fraction of mice surviving to adulthood . Our data show that the impairments of neural crest cell migration patterns observed in Wnt1-Ate1 knockout are variable from embryo to embryo , suggesting that local variations in the migration speeds of the Ate1 knockout cells can contribute to the phenotype variability . Our data on co-culture of wild-type and Ate1 knockout cells further suggest that Ate1 knockout cells during migration and tissue expansion can ‘ride’ on the neighboring expanding wild-type tissues , reaching further destinations in the embryo than they could reach on their own . This seems to be especially likely in the case of the enteric neurons that migrate as individual cells on the walls of the expanding gut . Indeed , the migrating speeds of these neurons have been calculated to exceed the average speed of migrating mesenchymal cells in culture and in situ by at least a factor of 2 [10] , suggesting that the expanding tissues greatly aid them along in reaching their destination and covering the entire gut wall . Consistent with this fact , Wnt1-Ate1 mice have no defects in gut innervation , while having significant defects in craniofacial migratory neural crest tissues , suggesting that gut neuron precursors in Wnt1-Ate1 knockout reach their destinations mostly by ‘riding’ , which is not affected by Ate1 knockout–related impairments in cell migration speeds . In addition to the neural crest-derived craniofacial structures , Wnt1-expressing cells in the neural tube give rise to the peripheral nervous system and large parts of the brain , including the entire midbrain . In Wnt1-Ate1 mice parts of the peripheral nervous system and the midbrain region are prominently devoid of Ate1 expression ( Figure S3 and Figure S4 ) . While we observed no morphological defects in the midbrain or the layout of the peripheral nervous system in the mutant mice , the Wnt1-Ate1 pups that survived for 1–3 weeks past birth exhibited severe behavioral retardation , delayed growth , and apparent difficulty with feeding . While some of these defects , especially the feeding abnormalities , could result from the shortened palates that prevent normal food intake , others could result from the neurological problems , such as reduced innervation leading to the impaired neuromuscular functions and reduced expansion in the thoracic cage ( which may result in breathing abnormalities and lung collapse ) , or from the defects in the sensory organs that are controlled by the midbrain , and possibly from other midbrain-related physiological abnormalities . It has been previously suggested that Ate1-dependent arginylation plays an important role in nerve growth and regeneration [17] , [47] , leading to an interesting possibility that some of the Wnt1-Ate1 mice may have neurological abnormalities of varying severity originated from the developmental defects not directly related to the neural crest cell migration . This possibility requires further study . Despite prominent Wnt1 expression in the head and chest area and somites , we saw no visible reduction in Ate1 staining in such neural crest-derived structures as teeth , endocrine glands , and heart , and no abnormalities were detected in these structures in Wnt1-Ate1 knockout mice . It is possible that in these areas Wnt1-expressing cells exist in a mixture with other cell populations , making them difficult to distinguish on tissue sections . In such case , their migration through the embryo could be heavily aided by surrounding tissues and impairment in their migratory machinery would be less visible . It is also possible that these cells normally express lower levels of Ate1 , making the Ate1 knockout and the corresponding migratory defects difficult to detect . Further investigation of these differences may lead to important discoveries of the role of arginylation in the development of these organs . Our data show that uniform impairment of the intracellular arginylation-dependent mechanisms of cell migration produces different effects in different Wnt1-expressing migratory cell subpopulations during development . These differences may be explained by the diverse mechanisms by which these subpopulations migrate in situ ( Figure 8 ) . Cells of the cranial neural crest migrate in large groups from the upper trunk region into the craniofacial areas and contribute to the soft palate , frontal bones , and other craniofacial structures . Vagal and sacral neural crest cells migrate individually along the developing gut from the back and the front , to give rise to the enteric nervous system . Trunk neural crest cells migrate over relatively short distances forming pigment cells , ganglia , and some other structures . Based on the timing of the corresponding developmental events and the distances these cells cover overall , it has been calculated that migration in these subpopulations occurs at different speeds [10] , and therefore such migration on the organismal level would be differentially affected by impairment of the same intracellular mechanisms . In the trunk , where cells migrate slowly over short distances , 2–4-fold reduction in the motility rate would be practically unnoticeable on the developmental scale . Consistent with this , we did not observe significant defects in the structures that develop from these cells due to Ate1 knockout . In the gut , cells migrate individually and while the migration speed of these cells is believed to be the fastest , up until now it has been unknown how much of this speed is contributed by the movement of the expanding tissues rather than the migrating cells themselves . Our data that enteric neurons are positioned normally despite the Ate1 knockout in these cells points to a possibility that their migration is heavily aided by tissue growth and gut elongation and that the cells themselves do not need to move fast to reach their final position . Cells contributing to craniofacial skeleton migrate as large subpopulations , with predicted speeds close to those experimentally observed for neural crest and other mesenchymal cells in culture and in situ ( ∼40 µm/hour , see [10] for review ) . Our data suggest that the migration of these cells depends heavily on the intracellular mechanical components that drive cell adhesion and movement , and that this cell population is affected the most by the impairment in these mechanisms . This is consistent with the results of other studies , where mouse knockouts of the cell migration-related genes ( such as fibronectin , cell adhesion molecules , etc . , see [10] for review ) leads to similar defects . Recent data suggest that many proteins involved in different physiological events are arginylated [23] and the list of identified arginylation targets is by far not complete . A prominent subset of the identified arginylated proteins ( such as actin , etc . ) are directly implicated in cell migration and it is very easy to suggest how lack of arginylation could lead to the defects in these proteins and affect cell migration by impairing the intracellular cytoskeleton-related machinery . For example , talin , a major focal adhesion protein , has been found to be arginylated on Ala1903 , and the present study suggests that talin-containing focal adhesions are significantly reduced in Ate1 knockout , pointing to a possibility that talin arginylation is one of the key players in the Ate1 regulation of cell adhesion . Studies of the regulation of these , and other proteins by arginylation and its contribution to Ate1-dependent cell migration constitute exciting directions of further research .
Mice with the exons 1–3 of the Ate1 gene flanked by LoxP sites ( Ate1-floxed mice ) were generated by introducing a targeting construct into the corresponding genome region in a cassette containing the floxed allele of the Ate1 genomic region fused with an frt-flanked Neo gene . The targeting vector was constructed using recombineering technique as described in [48] . A 12 , 375 bp genomic DNA fragment ( position Chr 7: 130 , 302 , 694–130 , 315 , 068 Mouse Feb 2006 Assembly ) containing exon 1–4 of the gene was retrieved from BAC clone RP23-92D13 . A loxP sequence was inserted 657 bp upstream of exon 1 and A frt-neo-frt-loxP cassette was inserted into the intron 3 , 505 bp downstream of exon 3 . Thus a fragment of 5 , 019 bp genomic DNA containing exons 1–3 of the Ate1 gene was floxed ( see Figure S1A ) . For ES cell targeting , the targeting vector was linearized with Not1 and electroporated into D1 ES cells derived from F1 hybrid blastocysts of 129S6 x C57BL/6J by Gene Targeting & Transgenic Facility at University of Connecticut Health Center . 192 G418 resistant ES colonies were isolated , and 32 clones were screened by nested PCR using primers outside the construct paired with primers inside the neo cassette . The sequences for primers used for ES cell screening were as follows: 5′ arm forward primers: ATE Scr F1 ( 5′- GTCTCACTTCCCTTCCTTAG -3′ ) and ATE Scr F2 ( 5′- ATTACCAGTGCTCGGTGCTT -3′ ) . Reverse primers: Loxp scrR1 ( 5′–GAGGGACCTAATAACTTCGT-3′ ) and loxp scrR2 ( 5′-GGAATTGGGCTGCAGGAATT-3′ ) . 3′ arm forward primers: frt scr F1 ( 5′-TTCTGAGGCGGAAAGAACCA-3′ ) and frt scr F2 , ( 5′-CGAAGTTATTAGGTGGATCC-3′ ) ; Reverse primers: ATE Scr R1 ( 5′- tcagtggttctcaacctgtg -3′ ) and ATE Scr R2 ( 5′- caggggttacctaagaccat -3′ ) ; 7 out of 32 clones were PCR positive for both arms and were expanded . The genotypes were confirmed after ES cell expansion . For chimera generation and F1 mice genotype analysis three clones ( 1B3 , 1A4 and 1H2 ) were aggregated with 8-cell embryos of CD-1 strain . The aggregated embryos were transferred to pseudopregnant recipients and allowed to develop to term . 25 chimeric mice were identified by coat color . Five chimeras ( 2 each for1B3 and 1A4 , 1 for 1H2 ) were mated with CD-1 females to test for germline transmission . All of them were demonstrated as germline chimeras . The neo cassette was removed by mating the chimeras with ROSA26FLP1 ( Jax stock#: 003946 ) homozygous females . To obtain Wnt1-Ate1 mice , Ate1-floxed mice were crossed with the mouse strain Tg ( Wnt1-GAL4 ) 11Rth Tg ( Wnt1-cre ) 11Rth/J ( The Jackson Laboratory ) expressing Cre recombinase under the neural crest-inducing Wnt1 promoter . The following primers were used for genotyping of the final mouse strain ( see Figure S1B for a typical genotyping gel ) : for Ate1-floxed allele , Ate1gtLoxF ( 5′-TGCCTCCAGCATTGGATGAA-3′ ) and Ate1gtLoxR ( 5′-CCATGGGTCTCCAATTTGCA-3′ ) ; For ROSA locus and Wnt transgene , primers recommended by the Jackson Laboratory were used as described on their web site for each corresponding strain . Rat monoclonal antibodies against a mixture of full-length bacterially expressed Ate1-1 and Ate1-2 were custom produced by Absea ( http://www . absea-antibody . com ) . Reactivity with individual Ate1 isoforms was verified by Western blots against purified bacterially expressed Ate1 , isoforms 1–4 . Clone 6F11 that interacts equally with all four Ate1 isoforms was used for section staining shown in Figure S3 and Figure S4 . For gross anatomical examination , mice at different postnatal stages were euthanized , dissected and observed for possible organ and tissue abnormalities , and individual organs were collected and compared between the mutants and the matching controls . For obtaining the images of the roof of the mouth and nasopharynx entrance shown in Figure 2 newborn mice were fixed in 4% paraformaldehyde in PBS and stored at 4°C . Fixed samples were washed in PBS and lower jaws were removed , and then the palate and the entrance of nasopharynx were observed under a dissection microscope . For histological analysis , mice and isolated mouse organs including X-gal stained embryos were fixed in 4% paraformaldehyde in PBS , paraffin embedded , and sectioned . For observation and analysis of general organ morphology sections were stained with hematoxylin and eosin . For immunohistochemistry shown in Figure S3 and Figure S4 , paraffin-embedded sections were deparaffinized with xylene , re-hydrated with sequential methanol∶PBS series ( 95∶5 , 70∶30 , 50∶50 and 30∶70 ) , washed with PBS , blocked with PBS supplemented with 0 . 1% Triton X-100 and BSA , and treated with anti-Ate1 ( described above , 1∶50 ) , anti-beta-III tubulin ( R&D Systems , MAB1195 , 1∶100 ) , anti-phospho histone H3 ( Santa Cruz Biotechnology , sc-8656 , 1∶500 ) , anti-cleaved caspase 3 ( Cell Signaling Technology , #9661 , 1∶100 ) or anti-talin ( Santa Cruz Biotechnology , sc-7534 , 1∶500 ) antibodies in PBS supplemented with 0 . 1% Triton X-100 . After washing with 0 . 1% Triton X-100 in PBS , samples were treated with fluorescent dye-conjugated secondary antibodies , washed with PBS and mounted in Aqua Poly/Mount ( Polysciences , Inc . , 18606 ) . Samples were observed under a fluorescent microscope . Whole-mount skeletons of newborn and adult mice were stained with alizarin red S and alcian blue 8GS as described in [41] . Fixation and blocking in guts excised from embryos at E16 . 5 was performed as described in [49] . After blocking , guts were incubated with anti beta-III tubulin antibody for 2 days at 1∶100 dilution in 5% BSA in TST buffer ( 20 mM Tris-HCl ( pH 8 . 0 ) , 150 mM NaCl , 0 . 1% Triton X-100 ) , washed with TST , and treated with FITC-conjugated secondary antibody at 1∶100 dilution in TST . After washing in TS buffer ( 20 mM Tris-HCl ( pH 8 . 0 ) , 150 mM NaCl ) , samples were mounted in OxyFluor at 1∶100 dilution in TS , and observed under confocal microscope . Riboprobe generation and in situ hybridization were performed according to [50] with a modification in post-antibody wash as follows: after the treatment with the antibody , embryos were washed 6 times for 1 hour each and stored at 4°C overnight . This washing process was done 3 times over the course of 3 days . After color development , embryos were fixed , dehydrated , rehydrated and cleared according to the protocol available online ( http://www . med . upenn . edu/mcrc/histology_core/wholemount . shtml ) . Embryos at E8 . 5 or E9 . 5 were collected from pregnant female mice into PBS , and neural tube area was dissected from the back of the embryos using 27 gauge needles . Dissected tissues were cultured in 1∶1 of DMEM/F10 supplemented with 10% FCS and antibiotics on glass bottom dishes ( Matek ) pre-coated with 10 µg/ml fibronectin in PBS for 1 hour at room temperature . Explants were cultured for 2 days and used for further analyses . Immortalized wild-type and Ate1 KO embryonic fibroblasts were cultured as described in [24] . Wound healing assays to measure the migration speeds of these cells were performed by growing the cells to a confluent monolayer in tissue culture dishes , followed by scraping off a portion of the monolayer , and the migration of the cells into the resulting wound was recorded as time-lapse images at 2 min intervals over 10–15 hour observation period and analyzed using Metamorph imaging software ( Molecular Devices ) . To analyze the migration of Ate1 KO cells in co-culture with wild-type cells , wound healing experiment were performed with GFP-labeled Ate1 KO cells co-plated with wild-type cells as a mixture at the ratio of KO∶WT 1∶2 to 1∶6 . For reciprocal experiments , GFP-labeled wild-type and unlabeled Ate1 KO cells were used . For immunostaining , WT and Ate1 KO cell monolayers were scraped to produce the wound . After 4 hours , cells were fixed with 4% paraformaldehyde , permeabilized with 1% Triton X-100 , and stained with mouse monoclonal anti-paxillin antibody ( BD Biosciences ) , rhodamine-phalloidin ( Sigma ) , mouse anti-talin clone 8d4 ( Sigma ) , and mouse anti-FAK , clone 2A7 ( Upstate Biotechnology ) in different combinations . For immunostaining for fibronectin , cells attached on the cover glasses were fixed in 4% paraformaldehyde and stained with rabbit polyclonal anti-fibronectin antibody ( Sigma ) . To observe only the extracellular fibronectin , cells were not permeabilized or washed with Triton X-100 before or after fixation . Fetuses or neural crest explants were fixed with 4% paraformaldehyde at 4C for 1 hour , rinsed for 30 minutes with rinse buffer ( 0 . 2 M sodium phosphate pH 7 . 3 , 2 mM magnesium chloride , 0 . 02% IGEPAL and 0 . 01% sodium deoxycholate ) 3 times , and incubated overnight at 37°C in the staining solution ( 5 mM potassium ferricyanide , 5 mM potassium ferrocyanide and 1 mg/ml X-gal in rinse buffer ) . Samples were post-fixed with 4% paraformaldehyde and stored in 70% ethanol . TUNEL assay were performed in fetuses at E9 . 5 and E12 . 5 using In Situ Cell Death Detection Kit , Fluorescein ( Roche ) . Fetuses were fixed in 4% paraformaldehyde in PBS overnight at 4°C , washed 3 times in PBS and incubated with 18 . 7 µg/ml proteinase K in 10 mM Tris/HCl , pH 7 . 5 for 30 minutes at 37°C . After washing 3 times in PBS , samples were incubated in the Label Solution supplemented with Enzyme Solution for 1 hour at 37°C , then washed in PBS 3 times and photographed under a fluorescent microscope . For negative control , samples were incubated in Label Solution without Enzyme Solution . For positive control , samples were treated with 100 U/ml DNase I and 1 mg/ml BSA in 50 mM Tris/HCl , pH 7 . 5 for 20 minutes at room temperature , followed by incubation with Label Solution supplemented with Enzyme Solution . Cell proliferation in the neural crest explants from E8 . 5 and E9 . 5 fetuses was analyzed by detecting BrdU incorporation using BrdU Cell Proliferation Assay kit ( Calbiochem , QIA58 ) . After 2-day culture , explants were incubated with 1∶2000 BrdU Label in culture medium for 2 hours . Explants were washed twice in DMEM without serum , and fixed with Fixative/Denaturing solution at room temperature for 30minutes . After washing in 70% ethanol , samples were air-dried and incubated with 1∶100 Anti-BrdU antibody in Antibody Diluent at room temperature for 1 hour . Samples were washed twice in Wash Buffer ( 1∶20 Plate Wash Concentrate in water ) , twice in PBS , and treated with 1∶200 Cy3 conjugated secondary antibody ( Jackson ImmunoResearch Laboratories ) in PBS at room temperature for 1 hour . Samples were washed 3 times in PBS and photographed under a fluorescent microscope . | Formation of many organs during development depends on the coordinated migration of individual cells and cell layers throughout the embryo . The majority of migrating cells originate from the neural crest lineage that gives rise to peripheral neurons , ganglia , pigment cells , and craniofacial structures , as well as parts of other organs in the body . Recent studies have implicated arginylation—a poorly understood protein modification—in the regulation of basic mechanisms that underlie cell migration . Here we test the role of arginylation in neural crest cell migration during mouse development by constructing and examining a mouse model with arginylation-deficient neural crest cells . We find that these mice die at or soon after birth and exhibit severe defects in the development of distinct neural crest-derived structures . Our findings uncover a previously unknown mechanism of the regulation of neural crest cell migration during development , and shed light on general principles of neural crest migration in vivo . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"developmental",
"biology/pattern",
"formation",
"cell",
"biology/morphogenesis",
"and",
"cell",
"biology",
"developmental",
"biology/organogenesis",
"developmental",
"biology/morphogenesis",
"and",
"cell",
"biology"
]
| 2010 | Arginylation-Dependent Neural Crest Cell Migration Is Essential for Mouse Development |
Although many methods are available to test sequence variants for association with complex diseases and traits , methods that specifically seek to identify causal variants are less developed . Here we develop and evaluate a Bayesian hierarchical regression method that incorporates prior information on the likelihood of variant causality through weighting of variant effects . By simulation studies using both simulated and real sequence variants , we compared a standard single variant test for analyzing variant-disease association with the proposed method using different weighting schemes . We found that by leveraging linkage disequilibrium of variants with known GWAS signals and sequence conservation ( phastCons ) , the proposed method provides a powerful approach for detecting causal variants while controlling false positives .
Next-generation DNA sequencing technologies allow discovery of genetic variants across the full spectrum of allele frequencies , thereby enabling exhaustive screens for association between diseases and variants . Numerous statistical methods [1] have been developed for analyzing sequence variants . Recently , these methods have increasingly focused on rare variants because of their functional implications [2] , [3] , documented roles in disease etiology [4]–[7] , potential contributions to missing heritability [8] , and to the associations reported between common diseases and common variants [9] . Testing rare variants one at a time tends to have low power with a realistic sample size , especially in the presence of low penetrance and allelic heterogeneity ( multiple variants at the same locus conferring risk ) . In an attempt to overcome this problem , grouping-based strategies have been proposed . These approaches typically involve grouping qualifying variants based on their location within a gene or pathway and testing the aggregated effect of the resulting set of variants . Several strategies for aggregating variant effects have been proposed: from simply collapsing rare variants [10] to summing weighted counts of minor alleles [11]–[13] , and to more sophisticated approaches that attempt to incorporate the effect of both protective and neutral alleles [14]–[18] . Grouping-based testing remains an active area of methodological research and new methods continue to be developed . However , there are two significant limitations of any grouping-based test . First , the performance is critically dependent on the extent to which the grouping strategy reflects the genetic architecture of the disease being investigated . In order for grouping to be an effective strategy one must put ‘weight’ on those variants that are truly important . Putting ‘weight’ on unimportant or null variants will add noise to the statistic resulting in a loss of power . Since the genetic architecture of any disease is unknown , it is not always obvious which grouping strategy to prefer . Second , a significant grouping-based test implicates a genomic region or pathway and not specific genetic variants . Thus , the ultimate goal of identifying individual causative mutations remains elusive . For these reasons , some are reconsidering association testing of rare variation via grouping-based tests [19] . Here we introduce a Bayesian multi-variant liability regression model that does not involve grouping and tracks directly to individual variants . More generally , this model falls into a broader framework of Bayesian hierarchical modeling , which attempts to estimate all regression coefficients ( variant effects ) simultaneously by imposing variant-specific shrinkage on the estimated effects . This class of approaches has seen wide and successful applications in predicting quantitative traits , particularly in agricultural species [20]–[22] . However , its effectiveness in the analysis of sequence data in human disease studies is yet to be investigated . There are two attractive features of the proposed approach . The first is that all variants are jointly analyzed so as to reduce bias of estimated effects by borrowing information from and/or accounting for other variants . The second feature is its ability to quantitatively incorporate prior information to weight individual variants based on their prior likelihoods of causality . Both features are expected to enhance power and reduce false discoveries . The weighting scheme is entirely customizable and can represent multiple sources of information . Here we have explored weighting schemes that incorporate two sources of information: sequence conservation ( a natural measure of functionality of mutations [23] , [24] ) , and linkage disequilibrium ( LD ) with significant SNPs detected by a genome wide association study ( GWAS ) . The motivation to utilize LD between candidate variants and prior GWAS results is an attempt to mimic the real situation for most common diseases where large scale GWAS have been performed before any sequence data are analyzed and to effectively leverage that prior information . GWAS signals can guide the search for causal variants from sequence data both by defining candidate regions and by ‘tagging’ causal variants . The latter is the very assumption behind GWAS and has been recently shown to happen relatively often , even between rare causal variants and common tags [9] . We first tested the proposed Bayesian approach in fully simulated scenarios . We focused on causal variants at low and intermediate frequency ( five MAF intervals between 0 . 2%–5% ) and an intermediate genotypic relative risk ( GRR ) of 3 under a dominant model , which is when single variant tests would likely underperform given a realistic sequencing sample size . We evaluated performance by the power of the methods to correctly identify the precise causal variants ( as opposed to association signals ) and their false positive rates . We then performed the Bayesian analysis on a set of exome sequencing samples , for which we assigned case/control statuses according to well-documented disease causing variants . We considered Crohn's disease and hepatitis C virus ( HCV ) treatment induced anemia . For Crohn's disease , there are two SNPs ( rs2066844 , MAF = 5 . 29%; rs2066845 , MAF = 1 . 10% ) in the NOD2 gene that are known to be causal [25] , [26] . Furthermore , GWAS run by the Wellcome Trust Case Control Consortium [27] found a signal ( rs17221417 , MAF = 25 . 41% ) in the NOD2 locus that is significantly ( p = 9 . 4×10−12 ) associated with Crohn's disease . For HCV treatment induced anemia , rs6051702 was identified as the most significant signal ( MAF = 17 . 81% , p = 1 . 1×10−45 ) in a GWAS scan [28] , and in the same study , two functional variants ( rs1127354 , MAF = 7 . 55%; rs7270101 , MAF = 13 . 85% ) in the ITPA gene were also identified as causal . A subsequent biochemical study confirmed that these variants are indeed causal and described biochemically how they protect against treatment induced anemia [29] . Using these confirmed causal variants and GWAS signal for a given trait , we simulated case/control data from the available exomes , assuming heritability ( on liability scale ) = 10% , and assessed performance of the Bayesian approach .
Across the 200 replicates , the number of candidate variants ranged from 862 to 1150 , with a mean number 991 . Each data replicate was analyzed by the single variant test ( benchmark ) and the Bayesian multi-variant liability regression model with ‘no weight’ , ‘r weight’ , ‘phastCons weight’ , and ‘r×phastCons weight’ . As expected , increasing causal MAF led to an increase in the power of all methods ( Table 1 ) . Notably , even in the lowest MAF range ( 0 . 2–1% ) , the best method , Bayesian model with ‘r×phastCons’ weight , achieved an average ( of the three causal variants ) power of 0 . 34 and was able to detect at least one causal variant 67% of the time . When MAF was raised to 4–5% , the average power was 0 . 4 for single variant test , 0 . 7 for Bayesian models with r weight and no weight , and 0 . 9 for Bayesian models with phastCons and r×phastCons weight . FPR was overall controlled at a very low level ( Figure S4 ) . Across all scenarios , the maximum averaged FPR was about 7 out of 1000 , as produced by single variant test on non-causal variants on the causal chromosome . As expected , FPR on the causal chromosome was consistently higher than that on the null chromosome . Next we describe in more detail results from the comparison among different methods . In summary , by simulation we showed that Bayesian multi-variant liability regression model with informative weight assigned to variants substantially improved the power to detect causal variants , compared with single variant test and unweighted Bayesian model . In particular , we found that the product of r and phastCons constitutes a better weight than either alone in terms of power and FPR , especially at low causal MAF . We then applied our method to real exome sequence data with simulated phenotypes . Information about the two NOD2 causal variants ( rs2066844 and rs2066845 ) and the two ITPA variants ( rs1127354 and rs7270101 ) is given in Table 2 . In both data sets , while LD of causal variants with GWAS signal was higher than most of the non-causal variants , there existed non-causal variants with higher LD ( Figure S1 , S2 , lower panel ) . Such occasional high LD with GWAS signal for non-causal variants was also observed in the previous simulated genomes data . Similarly , for some non-causal variants their phastCons scores were higher than those of causal variants . This is not surprising because variants not causal for one disease may be causal for other diseases . For example , in the NOD2 data , 28% of non-causal variants had higher composite phastCons scores than the causal variant rs2066844 ( 0 . 32 ) , and in the ITPA data 34% of non-causal variants had higher scores than the causal variant rs7270101 ( 0 . 17 ) . However , it was rare for non-causal variants to have both high r and high phastCons . This made r×phastCons an attractive weighting scheme because it incorporates both measures to discriminate causal from non-causal variants . In Figure 3 ( A ) and ( B ) , we first show a Manhattan plot from single variant test and one from Bayesian liability model with r×phastCons weight , based on one representative example out of the 100 simulated data sets . We then summarize results from both methods by displaying for each candidate variant the proportion of the 100 simulations where it was detected ( i . e . , being declared as significant ) . As seen in Figure 3 ( A ) , r×phastCons weights had a clear pattern in the NOD2 sequence variants: the two causal variants ranked 4th and 5th among all 100 variants; with a few exceptions , almost all non-causal variants had very small weights . The last panel illustrates by replicating simulations how well each method identified causal vs . non-causal variants . For the relatively common causal variant rs2066844 ( MAF = 5% ) , it was almost always identified by both methods across the 100 simulations . For the low-frequency causal variant rs2066845 ( MAF = 1% ) , however , proportion of detection was about 40% and 20% by the Bayesian model and single variant test , respectively . Considering the low heritability ( 10% ) and modest sample size ( ∼700 ) , 40% is a substantial improvement over 20% for detecting causal variant at an allele frequency of 1% . In the meantime , false detection among non-causal variants produced by either method was negligible , likely a result of the absence of linkage disequilibrium in this region . Compared with variants in the NOD2 region , variants in the ITPA region had a less clear pattern in r×phastCons ( Figure 3 ( B ) ) : causal variant rs1127354 ranked at the 4th whereas causal variant rs7270101 ranked at the 21st . The lower weight of rs7270101 had to do with the fact that it was an intronic SNP and was less conserved . While the Bayesian model outperformed single variant test for rs1127354 ( MAF = 7 . 55% ) by a large margin ( 80% vs . 30% ) , for rs7270101 ( MAF = 13 . 85% ) the pattern was opposite: the proportion of single variant far exceeded that of the Bayesian model ( 80% vs . 40% ) . However , the higher power of single variant test was accompanied by a large number of false positives , as opposed to virtually no false positives by the Bayesian model . This is in fact a strength of our Bayesian model in that it is able to select the correct causal variants out of many SNPs in LD . In addition , the two causal variants were always the top ranked variants in the Bayesian results while they were lower ranked in the single variant test . This makes prioritization much more accurate and efficient by the Bayesian model .
Here we developed and evaluated a Bayesian multi-variant regression approach for detecting causal variants in sequence data . We first tested its performance in simulated data using an intermediate risk ( GRR = 3 ) , a moderate and realistic sample size ( 500 cases and 500 controls ) , and a range of minor allele frequencies for the causal variants . Compared with the standard single variant test , our method , when coupled with informative prior weights , showed a clear advantage in statistical power . Application of the method to real exome sequence data in order to identify known causal variants implicated through GWAS showed similar results . In particular , while the high LD among variants in the ITPA data created great ambiguity in interpreting the results from single variant test , the Bayesian model using r×phastCons to weigh variants was able to effectively discriminate causal from non-causal variants in terms of their effect size estimates . The Bayesian multi-variant regression model entails leveraging informative prior information on variants so as to better distinguish causal variants from non-causal variants . In this study , the uses of different kinds of weights give rise to different prior specifications ( shrinkage ) of variant effects , which in turn affect posterior estimation of variant effects and their statistical significance . As we have shown in the Results section , power and FPR of the Bayesian regression model were greatly dependent on the choice of prior weighting scheme . A good weighting scheme is data-dependent and often requires combining different pieces of information . In the full simulation study , which served as a proof of concept , we generated data assuming that variant causality is correlated with its conservation as encoded by phastCons . We found that the combination of significant associations from GWAS and variant phastCons rendered the Bayesian model a highly improved performance in power than one without weighting . However , while there is a tendency for causal variants to be evolutionarily more conserved for many diseases , this may not be always true , especially for late on-set diseases that are not subject to natural selection . In such scenarios , some form of prediction for variant effect , such as the PolyPhen scores [35] for coding variants would be a more appropriate functional measure . Using prior biological knowledge in genetic association studies has increasingly been adopted recently . Several studies have proposed using variable significance threshold for variants . For example , the Prioritized Subset Analysis ( PSA ) [36] partitions variants to two subsets , a ‘prioritized set’ and a ‘complementary set’ , where the prioritized set contains variants that are more likely to harbor causative loci . It then controls false discovery rate in each set separately , resulting in different thresholds on nominal p values for prioritized and complementary variants . The partitioning of variants into two different priority sets can also be iteratively optimized given the data [37] . Alternatively , a more sophisticated method is to have variant specific threshold or weight based on their functionality [38] . In addition , grouping-based association tests can also aggregate variants differentially according to some prior knowledge [12] , [39] . One essential difference between our method and previous methods is that we fit all variants simultaneously while the variable threshold methods often estimate significance for variants individually . As in the common theme of Bayesian shrinkage regression methods , our knowledge on variant effect appears only in its prior variance , since all effects are shrunk toward a single prior mean , which is zero . As such , the weight only indicates magnitude of variant effect but not direction . Indeed , for variants with high phastCons weights , it is unlikely for their minor alleles to have a protective role in disease liability . Thus , it is desirable to assign a positive prior mean for these variants in order to enhance the power of discovery . On the other hand , for variants with near zero weights , we still can use zero mean as prior . These considerations require a different prior specification than presented here as well as more complicated posterior computations [40] , which will be explored in further extension of the present work . An alternative Bayesian approach to identifying variants associated with disease is Bayesian model selection via Bayes factor . A representative example under this framework is the Bayesian risk index method [41] , which allows for uncertainty of inclusion of variants and the direction of the effect . As opposed to fixing the number of variants , this feature offers the method some advantage when the proportion of causal variants is low in a region under study . Specifically , it provides a global evidence of a set of variants for their association with the disease and , if there is a global association , it furthers asks which variants are driving the association . Our method differs from the Bayesian risk index in the following essential ways . First , in our multi-variant regression model , a fully Bayesian treatment is used for the estimation of variant effects , whereas the computation of Bayes factor ( as in Bayesian risk index ) relies on approximation of marginal likelihood using maximum likelihood estimates of variant effects . That is , our method captures the uncertainty of variant effects through prior distributions . Second , variant-specific weights can be readily incorporated in our regression model , which has not been made possible in the Bayesian risk index . Furthermore , Bayes factor is a quantity that evaluates evidence in favor of a specific model; typically a value greater than 3 . 2 is interpreted as positive support . However , deciding upon a significant threshold for Bayes factor and thereby making decision is non-trivial and requires assumptions such as the relative cost of type II error to type I error [42] . Hence , our choice of using a test statistic followed by permutation test to control family wise type I error appears to be more convenient for declaring significance of a variant . We tested our method on a relative small candidate region ( 1 Mb ) around GWAS signals for convenient demonstration of the method , mainly for computational feasibility as many replicates needed to be run in the simulation study . The method can be extended to a larger candidate region to account for the situation that some causal variants could be far away from GWAS signals . Also , while we focused on low frequency candidate variants in our analyses , the method can readily encompass both common and rare variations . In both cases where the number of variants to be fitted in the model would be increased substantially , sample size needs to be increased accordingly to ensure accurate estimation of variant effects . On the other hand , as computing time grows when a large number of variants are analyzed , it may be useful to prescreen variants based on one or more criteria ( e . g . , exonic versus intergenic ) . In addition , another incentive for prescreening is that , with a limited sample size , adding more non-causal variants to the model would lead to less accurate effect size estimation for causal variants . In fact , the same argument holds for single variant test where prescreening can alleviate the burden of multiple testing . A common problem with a multiple regression model is multicollinearity among variables . In our context , this is caused by high LD between variants . In particular , variances of effect estimates increase for variants that are in LD with at least one other variant in the same model . This often leads to false rejection of a true association . In our method , this problem is mitigated by differential shrinkage on variant effects through differential weighting . That is , despite being highly correlated in their genotypes , variants in high LD can still be distinguished by additional information such as the phastCons score . This may explain why the supplement of phastCons to ‘r weight’ improved power substantially . We also performed a small experiment ( Text S3 ) where highly correlated variants were clustered and tagged . We tested the ability of the Bayesian model with ‘no weight’ to detect variant clusters that contained the causal variant as opposed to its ability to detect individual causal variants without clustering and tagging . Interestingly , clustering improved the power from 56% to 96% for causal MAF between 4–5% . However , the resolution to individual variants was lost by clustering and tagging . Nonetheless , when an informative weight is not available , this provides an efficient way to narrow down to variant clusters that contain causal variants . Our method is implemented to be able to include covariates such as principal components of SNP genotypes . However , an open issue exists with the permutation procedure in its validity in the presence of confounding covariates . In this case , naive shuffling of disease statuses would break the confounding structure observed in the original data . There are existing methods that can effectively preserve relationships between confounding covariates and genotypes as well as between covariates and disease statuses [43] . | The decline in DNA sequencing cost permits the interrogation of potentially all variants across the entire allele frequency spectrum for their associations with complex human diseases and traits . However , the identification of causal variants remains challenging . Existing single variant tests do not distinguish between causal association and association induced by linkage disequilibrium and tend to be underpowered for rare or low-frequency variants , whereas variant grouping methods do not identify individual causal variants . We propose a novel Bayesian hierarchical regression approach that estimates effects of multiple variants on a disease trait simultaneously and incorporates prior information on the likelihood of causality . By simulation , we show that by combining linkage disequilibrium with known genome wide association signals and functional conservation , the proposed method , the first of its kind , is powerful to correctly detect causal variants . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
]
| [
"genetics",
"biology",
"computational",
"biology"
]
| 2013 | Leveraging Prior Information to Detect Causal Variants via Multi-Variant Regression |
Ebola virus ( EBOV ) causes acute hemorrhagic fever in humans and non-human primates with mortality rates up to 90% . So far there are no effective treatments available . This study evaluates the protective efficacy of 8 monoclonal antibodies ( MAbs ) against Ebola glycoprotein in mice and guinea pigs . Immunocompetent mice or guinea pigs were given MAbs i . p . in various doses individually or as pools of 3–4 MAbs to test their protection against a lethal challenge with mouse- or guinea pig-adapted EBOV . Each of the 8 MAbs ( 100 µg ) protected mice from a lethal EBOV challenge when administered 1 day before or after challenge . Seven MAbs were effective 2 days post-infection ( dpi ) , with 1 MAb demonstrating partial protection 3 dpi . In the guinea pigs each MAb showed partial protection at 1 dpi , however the mean time to death was significantly prolonged compared to the control group . Moreover , treatment with pools of 3–4 MAbs completely protected the majority of animals , while administration at 2–3 dpi achieved 50–100% protection . This data suggests that the MAbs generated are capable of protecting both animal species against lethal Ebola virus challenge . These results indicate that MAbs particularly when used as an oligoclonal set are a potential therapeutic for post-exposure treatment of EBOV infection .
Ebola virus ( EBOV ) is a filovirus causing severe viral haemorrhagic fever in humans and non-human primates ( NHPs ) [1] . There are five species of EBOV: Zaire ebolavirus ( ZEBOV ) , Sudan ebolavirus ( SEBOV ) , Cote d'Ivoire ebolavirus ( CIEBOV ) , Reston ebolavirus ( REBOV ) , and Bundibugyo ebolavirus ( BEBOV ) [2] . ZEBOV has the highest virulence with a case fatality rate of 60–90% [1] , [3] . Although several attempts have been made to treat EBOV infections [4]–[8] , there are currently no commercially approved vaccines or effective therapies , therefore new treatments are needed . Several studies have been conducted to determine the immune correlates of protection in EBOV infections either by following natural infections , or in in vivo animal models [9]–[15] . Both T and B cell immunity was analysed and it was believed that a strong early humoral immune response may have been a factor in survival [11] , [16] , [17] . Additionally , in fatally infected patients EBOV-specific IgG was absent , and IgM levels were low in comparison to the survivors [16] . The passive transfer of immune sera or whole blood was tested but its effectiveness is still controversial as it has not consistently provided protection [10] . However , in mice experiments EBOV-specific sera was sufficient for improving survival [10] , [18] , [19] . The key target for developing effective neutralizing antibodies ( NAb ) is suspected to be the surface glycoprotein ( GP ) [20] . EBOV GP is the only protein on the surface of the virus and is responsible for receptor binding , viral entry , and cellular tropism [20]–[24] . GP-specific NAb generated in several species were protective in some animal models , however , the NAb titres are low in natural infections and their effectiveness in humans remains to be confirmed [10] , [25]–[27] . Antibodies blocking viral entry , by binding the receptor or preventing viral fusion would be ideal candidates for improving survival . Additionally , the primary pathology of EBOV haemorrhagic fever is vascular injury and coagulation abnormalities , and GP has been shown to cause cytotoxicity and vascular permeability [28] , [29] . In fact GP-induced cytotoxicity has been correlated with mortality rates in the different EBOV viral species [28] , [30] . Taken together this suggests that prophylactic and post-exposure treatment strategies involving antibodies specific for the EBOV GP would be an effective intervention for an Ebola infection . Monoclonal antibodies ( MAbs ) against ZEBOV GP have been created previously and tested in several animal models as a post-exposure therapeutic [26] , [27] , [31]–[34] . However , the ability of each of the MAbs to improve survival in a lethally infected animal varied considerably . Some MAbs were able to protect mice completely yet guinea pigs partially [32] , [33] . One neutralizing MAb KZ52 was 100% efficacious in guinea pigs , but did not protect NHPs [26] , [35] . Overall , there are a variety of mechanisms employed by MAbs to improve survival , and the ability of the MAb to neutralize the virus is not essential . The MAbs tested so far are not 100% efficacious in all animal models therefore further research is needed for more effective antibodies . The goal of this study was to test a panel of MAbs specific for the ZEBOV GP for their efficacy in protecting mice and guinea pigs from a lethal ZEBOV infection . Previously , 8 ZEBOV GP-specific MAbs had been generated using the VSVΔG/ZEBOVGP vaccine as the immunogen [36] . A preliminary study characterizing the MAbs found they all improved survival in mice infected with a high dose of mouse adapted-ZEBOV ( MA-ZEBOV ) [36] . As the MAbs were effective in the mouse model it is possible that these MAbs could be used as a post-exposure therapeutic for a ZEBOV infection . In this study optimization of a post-exposure protocol is undertaken in both the mouse and guinea pig model in order to determine the various treatment parameters , including the dose , treatment time , and MAb combination , that are required to provide complete protection .
All infectious animal work was performed in the biosafety level 4 biocontainment laboratory at the Public Health Agency of Canada , and approved by the Canadian Science Centre for Human and Animal Health Animal Care Committee following the guidelines of the Canadian Council on Animal Care . Animals were acclimatized for 10 days prior to the start of the experiment , and fed and monitored daily pre- and post-infection . The recombinant virus VSVΔG/ZEBOVGP containing the Zaire ebolavirus , strain Mayinga , glycoprotein ( GP ) in place of the VSV glycoprotein ( G ) has been described previously [37] . The mouse-adapted ( MA-ZEBOV ) and guinea pig-adapted ( GA-ZEBOV ) ZEBOV strain Mayinga viruses were described previously [38] , [39] . The creation of 8 MAbs ( 1H3 , 2G4 , 4G7 , 5D2 , 5E6 , 7C9 , 7G4 , 10C8 ) has been described previously [36] . Briefly , 6–8 week old Balb/C mice were immunized with 107 pfu VSVΔG/ZEBOVGP intraperitoneally ( ip ) , at 0 , 4 , and 8 weeks . A final boost with Zaire ebolavirus like particles ( eVLPs ) was performed before harvesting spleen cells and fusing with SP2/0 myeloma cells according to Kohler and Milstein [40] . The generation of the ZEBOV GP/VP40 eVLPs have been described previously [41] . The hybridomas were grown in Hybridoma SFM ( Invitrogen ) , 1 mM L-Glutamine , 1× Antibiotic-Antimycotic ( Invitrogen ) , in roller bottles at 37°C , 5% CO2 . Supernatant was cleared by centrifugation and concentrated ten times using an Amicon Stirred Cell system with a 30 kDA MWCO filter ( Millipore ) . The antibodies were purified on a HiTrap Protein G HP column ( GE Healthcare ) using Protein A Binding Buffer and IgG Elution Buffer ( Thermo Scientific ) according to manufacturers' instructions . Positive fractions were pooled , concentrated , then buffer exchanged into PBS using a 10 kDa MWCO Centriprep unit ( Millipore ) . Antibody purity , assessed by gel electrophoresis and coomassie blue staining was >98% . The 5–6 week old female Balb/C mice from Charles River , ( Quebec , Canada ) , were injected ip with the indicated amount of ZEBOV GP-specific MAbs in 100 µl PBS at the times indicated either before or after i . p . infection with 1 , 000 LD50 of MA-ZEBOV . Female guinea pigs ( Hartley strain ) , approximately 250 g , from Charles River , were challenged with 1 , 000 LD50 of GA-EBOV i . p . . At the indicated times post-infection the guinea pigs were treated i . p . with 1 ml of the MAb diluted in PBS . Naive control animals received PBS only . Clinical signs of infection and body weight were monitored for two weeks after challenge and survivors were followed three times longer than the death of the last control animal . ZEBOVGP-specific MAbs were serially diluted from 1/100–1/12 , 800 in DMEM . Starting concentrations were 3 . 75 , 3 . 46 , 4 . 34 mg/ml for 1H3 , 2G4 , and 4G7 , respectively . The MAbs were added to an equal volume of 104 pfu/ml VSVΔG/ZEBOVGP , diluted with DMEM , in order to provide 200 pfu/well . The virus-antibody combination was incubated at 37°C for 1 hour before adding 150 ul/well to a confluent 12 well tissue culture plate seeded with Vero E6 cells . After a 1 hour incubation , 1 ml of MEM 2% FBS , 1% low melting point agarose was added per well . Plates were incubated at 37°C 5% CO2 for 48 hours before adding 1 ml of 0 . 2% w/v crystal violet , 3 . 7% Formaldehyde , 2% Ethanol to each well for visualization of the plaques . The assay was performed in triplicate , and a positive control ( virus with no antibody ) and a negative control ( no virus ) incorporated . The percent reduction was calculated by averaging the count of the triplicate wells and comparing the number of plaques in the test sample against the number of plaques in the positive control ( 1− ( Test Sample plaques/Positive control plaques ) ) ×100 = % reduction . The log rank statistical test was performed for the Survival curve using the GraphPad Prism 4 software program . The survival curve for the MAb treated animals were compared to the survival curve for the PBS control group .
Previously , 8 MAbs specific for the glycoprotein ( GP ) of ZEBOV had been generated [36] . An initial characterization demonstrated they bound to a variety of GP segments , and that all 8 MAbs were able to pull down ZEBOV GP1 , 2 in an immunoprecipitation assay . In the current study , we further characterized the MAbs using a plaque reduction neutralization assay ( PRNT50 ) . The PRNT50 demonstrated that MAbs 1H3 , 2G4 , and 4G7 were neutralizing with a PRNT50 at a 1/200 , 1/800 , and 1/6 , 400 dilution , respectively ( Figure 1 ) . All of the other MAbs were non-neutralizing with 5D2 showing the highest degree of neutralization at 38% ( data not shown ) . Preliminary experiments in mice suggested these MAbs would be effective as a therapeutic for a MA-ZEBOV infection [36] . Therefore , a variety of parameters were assessed in order to establish the most effective treatment protocol . An in vivo mouse model was utilized to determine the protective efficacy of the individual MAbs ( Table 1 ) . Each MAb was injected either 1 day before ( −1 ) or after ( +1 ) a MA-ZEBOV infection ( 1 , 000 LD50 ) in Balb/C mice . All control mice receiving PBS only had a mean time to death of 6 . 6 and 5 . 0 days for the −1 and +1 day treatment , respectively . In contrast , mice treated with MAbs ( 100 µg ) demonstrated either partial or complete protection . For the −1 day protocol , the MAbs 5D2 , 5E6 , 7C9 were most effective with a 73–87% survival rate , in comparison to the other 5 MAbs ( 1H3 , 2G4 , 4G7 , 7G4 , 10C8 ) where survival rates ranged from 0–7% . Alternatively , every MAb performed better when given at 1 day post-infection ( dpi ) , with survival rates ranging from 40–100% . Overall , the level of protection against lethality varied with each MAb , and it appears that , in general , the MAbs are more effective when given 1 day after a lethal MA-ZEBOV infection . Since the MAbs were most effective when given 1 day after the lethal MA-ZEBOV infection , this treatment protocol was used to determine the most effective dose for protection ( Table 2 ) . A dose response was observed , and some MAbs were more potent than others for a given dose . The lowest doses providing complete protection from lethality for 5D2 , 5E6 , 7C9 , and 7G4 were 12 . 5 , 25 , 50 , and 100 µg , respectively . MAbs 4G7 and 10C8 demonstrated an 83% survival rate at the highest dose of 100 µg . MAbs 1H3 and 2G4 were not included as they were not very effective at the highest dose in the first experiment ( Table 1 ) . In the partially protected groups of mice , the mean time to death ranged from 6 . 40 to 8 . 20 days in comparison to the control mice ( 5 . 80 days ) . Overall , the various MAbs varied in their potency in providing protection against a lethal MA-ZEBOV infection in mice . Using the most effective MAb dose of 100 µg , the treatment time was extended in both directions in order to determine the optimal time for treatment , and to see how late treatment can be given before the survival rate declines ( Table 3 ) . A single dose of 100 µg for each MAb was injected either 1 or 4 days before a lethal MA-ZEBOV infection , or at 1 , 2 , or 3 dpi , and survival followed . Pre-treatment of the mice 4 days before infection with 1H3 , 2G4 , or 7G4 did not result in survival , whereas the other MAbs provided 30–90% protection . In the majority of cases , treatment 1 day before infection resulted in lower survival rates than 4 days before infection . Of the 8 MAbs , the most effective MAbs for pre-treatment were 5D2 , 5E6 , and 7C9 . They had the highest survival rates ( 73–90% ) and worked almost equally well on both days −4 and +1 . Extending the start of treatment after infection also had noticeable effects upon survival . Treating the mice on 1 or 2 dpi was the most effective treatment time , with some MAbs ( 5D2; 100% , 5E6; 93% , 7G4; 100% , 10C8; 87% ) being more protective when given at 1 dpi , and the others ( 1H3; 50% , 2G4; 70% , 4G7; 100% , 7C9; 90% ) when given at 2 dpi . Delaying the treatment to 3 dpi resulted in no survival for all MAbs except for 4G7 ( 10% survival rate ) . In general post-exposure treatment worked better than prophylactic treatment for the majority of the MAbs , except 5D2 , 5E6 , 7C9 which were highly effective at improving survival in mice both before and after the infection . All MAbs were once again tested individually in the guinea pig model . The MAbs were given i . p . at 1 dpi with 1 , 000 LD50 of GA-ZEBOV , and survival followed ( Table 4 ) . The PBS controls all died with a mean time to death of 7 . 7 days . In those groups receiving treatment , with the exception of 2G4 and 4G7 , none of the guinea pigs survived , but the mean time to death was significantly extended ( range of 9 . 4–11 . 7 days , p<0 . 050 ) . For 2G4 or 4G7 , the survival rate was 60% , demonstrating that the MAbs can provide levels of protective efficacy individually in the more stringent guinea pig model . Since individual MAbs were partially protective in the guinea pig , a second injection of the 3 neutralizing MAbs ( 1H3 , 2G4 , and 4G7 ) was included on 2 dpi ( Figure 2 ) . The guinea pigs were divided into 6 groups ( n = 6 ) , with one control group receiving only PBS , and 5 groups each receiving one of the non-neutralizing MAbs at 1 dpi , followed by the neutralizing MAb combination at 2 dpi . The PBS control treated animals all died with a mean time to death of 7 days . In contrast , all of the MAb treated groups demonstrated complete survival , except for 10C8 ( 83% ) . The treatment also improved morbidity as the MAb-treated groups maintained their weight in contrast to the controls that lost 6–7% of their weight by 4 and 5 dpi . This demonstrates that a combination of MAbs is an effective post-exposure treatment in guinea pigs . As two of the neutralizing antibodies were shown to be more effective at improving survival in guinea pigs ( Table 4 ) , the 3 ZEBOV GP-specific neutralizing MAbs were delivered as a combination alone to see if they would be sufficient as a therapy for an EBOV infection ( Table 5 ) . The combination of neutralizing MAbs ( 1 . 5 mg 1H3+1 . 5 mg 2G4+2 mg 4G7 ) was given to guinea pigs either 1 day before , or 1 , 2 or 3 days after a 1000 LD50 GA-ZEBOV infection , and survival followed . The PBS control group all died , with a mean time to death of 6 . 58±0 . 59 days . In contrast all animals receiving the neutralizing MAb combination at 2 dpi . survived . When the treatment was given on 3 dpi the percent survival dropped to 66 . 7% with a mean time to death of 11 . 17±3 . 09 days . Receiving the combination either one day before or after resulted in a survival rate of 50% , with the meantime to death of 11 . 17±3 . 09 and 7 . 92±0 . 42 , respectively . Overall , the neutralizing MAb combination improved survival in all treatment protocols with the 2 dpi treatment protocol being the most effective .
In this study 8 ZEBOV GP-specific MAbs were tested for their efficacy in protecting against a ZEBOV infection in both a mouse and guinea pig model; and a post-exposure protocol for guinea pigs was optimized . Individually , each MAb extended survival partially , or completely after a lethal dose of MA-ZEBOV in mice , whereas only the 2G4 or 4G7 treated groups demonstrated a 60% survival rate against a GA-ZEBOV infection in guinea pigs . The dose response in the mouse experiment suggests some MAbs were more potent than others at improving survival , with 100% protection with 12 . 5 µg 5D2 in comparison to an 83% survival rate with 100 µg of 4G7 or 10C8 ( Table 2 ) . In general , the MAbs worked best for both animal models when given after the start of the infection , particularly at 1 and 2 dpi , before efficacy started to decrease at 3 dpi This suggests that there may be a limited time period in which to begin treatment after becoming infected . Within this 2 day time span , some MAbs were more effective when given at 1 dpi ( 5D2 , 5E6 , 7G4 and 10C8 ) , and others at 2 dpi ( 1H3 , 2G4 , 4G7 , and 7C9 ) in the mouse model . It is possible that since the MAbs and virus were both injected ip that the MAbs might inhibit ZEBOV infection of cells and extend life . As the MAbs were only partially effective when given individually in the guinea pig model , a combination of 3 neutralizing MAb ( 1H3 , 2G4 , 4G7 ) at 2 dpi was tested in guinea pigs and found to provide complete protection , and prevent morbidity . Each of these MAbs binds to different regions of GP1 , 2 . 1H3 and 4G7 bind to the N- and C-terminus of GP1 , respectively , whereas 2G4 binds to GP2 [36] . Targeting multiple regions of GP appears to be a successful strategy . It is possible that a variety of mechanisms for preventing infection are employed by the 8 MAbs that is reflected in the differing amounts of MAbs needed , and the time of treatment in which they are most effective . Some MAbs may have more affinity for their epitope , or the epitope may be more accessible in the natural conformation . There is precedence for this as two ZEBOV-specific neutralizing MAbs KZ52 and JP3K11 were found to neutralize ZEBOV by distinct mechanisms [42] . Based on the various studies using ZEBOV-specific MAbs as a therapy for EBOV infections , it appears that there is no way of predicting which MAb would provide complete protection . However , initial testing must begin in mice and guinea pigs in order to make the initial determination about what therapeutic approach might be best to test in NHPs and humans . There have been several attempts at producing MAbs against ZEBOV GP however no clear pattern has emerged suggesting which primary sequence domain of GP is most immunogenic or whether neutralizing antibodies are more successful [26] , [27] , [31]–[34] . To date only neutralizing MAbs 133/3 . 16 , 226/8 . 1 and KZ52 have shown the capacity to improve survival in guinea pigs [32] , [33] , [35] . MAbs 133/3 . 16 and 226/8 . 1 only provided partial protection , while 25 mg/kg of KZ52 was completely protective in guinea pigs when given at −1 or +1 hours , but had later failed to protect NHPs [26] . Individually , none of the MAbs in this study were as effective as KZ52 in the guinea pig model . However , a combination of MAbs was more effective , and the treatment could be delivered as late as 2 days after infection . This is a significant extension from the 1 hour post-infection required for KZ52 . This is an important consideration as it is often difficult to begin treatment as early as 1 hour after an infection . All 8 MAbs in this study were originally selected by screening for their SSS coating antigen [36] . Theoretically , the ability to bind to the natural conformation may be more advantageous as it is more likely that the epitopes would be available and not hidden , or that the MAbs would be able to interfere with events required for viral entry such as receptor binding , and membrane fusion . Preventing entry into the cell would decrease the overall infection levels thereby giving the immune system a better chance at controlling the infection . There are many characteristics that make an antibody effective and it may not be the same mechanism for any two MAbs . Overall , the MAbs generated in this study and the optimized protocols demonstrate their potential as a post-exposure therapeutic against a ZEBOV infection . Because previous KZ52 antibody treatments that proved effective in guinea pigs later failed to protect NHPs , it is vital that further evaluation of the neutralizing MAb combination protocol should be conducted in NHPs . | Ebola virus ( EBOV ) causes acute hemorrhagic fever in humans and non-human primates with mortality rates up to 90% . So far there are no effective treatments available . This study evaluates the protective efficacy of 8 monoclonal antibodies ( MAbs ) against the Ebola virus surface glycoprotein , in mice and guinea pigs . Various combinations and doses of the neutralizing and non-neutralizing MAbs were tested , and a post-exposure treatment protocol was determined . There was 100% survival when guinea pigs received a mix of 3 neutralizing MAbs two days after a challenge with 1 , 000 LD50 of guinea pig-adapted EBOV . This data suggests that the MAbs generated are effective as a post-exposure therapeutic for a lethal Ebola virus infection . Development of a post-exposure therapeutic for an Ebola virus infection is vital due to the high lethality of the disease , the relative speed in which it kills , and the fact that no vaccine has been approved for human use . Additionally , is it unlikely that preventative vaccines will be employed , because Ebola virus infections occur primarily in Africa , and to date have only killed approximately 2 , 300 people making it financially unfeasible for a mass vaccination . Therefore , having an effective therapy in the event of an outbreak would be extremely beneficial . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"infectious",
"diseases",
"immunology",
"biology"
]
| 2012 | Ebola GP-Specific Monoclonal Antibodies Protect Mice and Guinea Pigs from Lethal Ebola Virus Infection |
Long non-coding RNAs ( lncRNAs ) have been recognized as key players in transcriptional regulation . We show that the lncRNA steroid receptor RNA activator ( SRA ) participates in regulation through complex formation with trithorax group ( TrxG ) and polycomb repressive complex 2 ( PRC2 ) complexes . Binding of the SRA-associated RNA helicase p68 preferentially stabilizes complex formation between SRA and a TrxG complex but not PRC2 . In human pluripotent stem cells NTERA2 , SRA binding sites that are also occupied by p68 are significantly enriched for H3K4 trimethylation . Consistent with its ability to interact with TrxG and PRC2 complexes , some SRA binding sites in human pluripotent stem cells overlap with bivalent domains . CTCF sites associated with SRA appear also to be enriched for bivalent modifications . We identify NANOG as a transcription factor directly interacting with SRA and co-localizing with it genome-wide in NTERA2 . Further , we show that SRA is important for maintaining the stem cell state and for reprogramming of human fibroblasts to achieve the pluripotent state . Our results suggest a mechanism whereby the lncRNA SRA interacts with either TrxG or PRC2 . These complexes may then be recruited by various DNA binding factors to deliver either activating or silencing signals , or both , to establish bivalent domains .
Histone H3 modifications involving lysine 4 trimethylation ( H3K4me3 ) and lysine 27 trimethylation ( H3K27me3 ) represent activating and repressive histone marks , respectively . However , when present together , as they are in bivalent sites , they mark genes that are poised for induction . Genes carrying the bivalent modification include those involved in differentiation of pluripotent stem cells . Two distinct histone modification machineries , associated with the trithorax group ( TrxG ) complex and with polycomb repressive complex 2 ( PRC2 ) , are responsible for methylating H3K4 and H3K27 , respectively . TrxG complexes comprise at least four protein components , WDR5 , RBBP5 , ASH2L and an H3K4 methyltransferase such as MLL ( MLL1-4 ) , whereas EZH2 , EED and SUZ12 are core components of PRC2 . Establishment of bivalent domains involves delivery of these two complexes to their target regions . Both MLL1 and MLL2 containing complexes deliver trimethyl marks to H3K4 , and MLL2 is required for this modification at bivalent sites in mouse embryonic stem cells [1 , 2] . CpG islands ( CGIs ) have been reported to play an important role in recruitment of TrxG and PRC2 complexes via several CGI-binding proteins [3] . In addition , TrxG complex has been shown to be recruited directly by DNA sequence-specific transcription factors Oct4 [4] and estrogen receptor α ( ERα ) [5] . Similarly , at least one component of the PRC2 complex , SUZ12 , can be targeted directly by the transcription factor CTCF [6] . Moreover , PRC2 target genes can recruit the complex through interaction with short RNAs transcribed from the 5’ ends of those genes [7–9] . We note that although under some solvent conditions PRC2 may exhibit non-specific interaction with RNA [9 , 10] , the experiments reported here , carried out in nuclear extracts or in PBS buffer , clearly show specificity for SRA . A growing number of long non-coding RNAs ( lncRNAs ) have been implicated in recruitment of TrxG or PRC2 complexes to their target genes [11] . Two groups of lncRNAs may be categorized according to whether TrxG or PRC2 complexes bind to them , defining activating and repressive lncRNAs respectively . The first category of activating lncRNAs , which recruit TrxG complexes to their target genes via WDR5 , includes Hottip [12] , NeST [13] and NANCI [14] . In contrast , examples of lncRNAs belonging to the second category of repressive lncRNAs , which recruit PRC2 complex to its binding sites , are Xist [15] , Hotair [16] and Braveheart [17] . The complete PRC2 complex has been shown to bind highly selectively to Hotair and RepA/Xist , as compared with control RNA [18] . Of the three core components of PRC2 comprising EZH2 , SUZ12 and EED , it has been shown recently that EZH2 and SUZ12 possess a high affinity for RNA binding , whereas EED helps to increase RNA binding specificity to the complex [18] . Recently , a novel technique , Chromatin Isolation by RNA Purification ( ChIRP ) , has provided a powerful method to map the location of lncRNAs genome-wide [19] . Using this technique , the lncRNA HOTAIR was shown to co-localize with the PRC2 complex and H3K27me3 genome-wide , supporting its functional role in tethering PRC2 to target genes . Similar techniques have been utilized to map the distribution of the lncRNA Xist , which also has a domain that recruits the PRC2 complex , along the X chromosome [20 , 21] . Although these and other lncRNA species have been shown to deliver either “activating” or “silencing” histone modifications , it is not clear whether they can function coordinately to create bivalent domains . The lncRNA steroid receptor RNA activator ( SRA ) can be recruited to DNA through interactions with proteins that bind either directly or indirectly to DNA [22] . For example , SRA has been shown to interact directly with ERα [23] , which binds to specific DNA sequences , and to co-activate ERα target genes [24] . It also forms a complex with the DEAD box RNA helicase p68 , which in turn interacts with the DNA binding protein MyoD [25] . We have reported previously that SRA and p68 form a complex with CTCF and are crucial for insulator function of CTCF at the IGF2-H19 locus [26] . Furthermore , it has been shown that SRA can interact with EZH2 [27] , suggesting that it might be involved in silencing functions associated with the PRC2 complex . In addition , SRA also interacts with HP1 gamma and LSD1 to repress progesterone receptor target genes [28] . In possible contradiction of that repressive function is the observation that knockdown of SRA in HeLa cells results in decreased expression of the majority of significantly changed genes [29] . In this study , we show that the lncRNA SRA is capable of binding TrxG and PRC2 . Direct interaction with the complexes is specific for sense SRA as compared with the control , its anti-sense counterpart . SRA-p68 interaction strengthens recruitment of a TrxG complex but does not affect PRC2 . We find that CTCF binding sites that are also occupied by SRA , are more likely to have bivalent marks . We also find that SRA/p68 associates with NANOG , a master transcription factor in pluripotent stem cells . These results show that SRA can associate with TrxG and PRC2 complexes to deliver either activating or repressive histone modifications , and that the choice can be modulated by proteins with which it associates . They also suggest a mechanism in which the bivalent state may be controlled at certain sites , including those occupied by NANOG , through recruitment of SRA and its associated histone modifying enzymes in pluripotent stem cells .
To confirm that SRA interacts with the RNA helicase p68 and CTCF [26] , an RNA pull down assay was performed using nuclear extract from human pluripotent stem cells NTERA2 and in vitro transcribed biotinylated antisense SRA and sense SRA . Western blot analysis showed that sense SRA specifically recruits p68 and CTCF ( Fig 1A and S1 Fig ) supporting our previous report [26] . p72 , another RNA helicase known to interact with SRA , was also pulled down by SRA . Next , to detect a possible association between SRA and TrxG and/or PRC2 in nuclear extract , the RNA pull down assay was employed to probe for WDR5 and EZH2 proteins , respectively . Both WDR5 and EZH2 were pulled down selectively by sense SRA suggesting that SRA interacts with TrxG and PRC2 complexes ( Fig 1A ) . WDR5 is shared by several TrxG complexes: interaction with both MLL1 and MLL2 was detected in these pull down experiments ( Fig 1A ) , as were related complexes containing histone methyltransferases SETD1A and SETD1B ( S2 Fig ) . RNA immunoprecipitation experiment showed that SRA was retrieved by anti-WDR5 and anti-SUZ12 , indicating an association between SRA and TrxG/PRC2 complexes in vivo ( S3 Fig ) . An in vitro RNA pull down assay similarly revealed an interaction between SRA and either recombinant TrxG or PRC2 complexes indicating that the binding between SRA and the two epigenetic machineries is direct ( Fig 1B ) . The selective properties of the SRA sense strand , in contrast to the antisense strand , are consistent with a specific interaction between the RNA and the two histone modifying complexes . To determine which components of TrxG and PRC2 mediate the interaction with SRA , individual recombinant proteins were used in the RNA pull down . Among major TrxG components , sense SRA specifically retrieved WDR5 , whereas it pulled down both the EED and SUZ12 components of the PRC2 complex ( Fig 1C ) . This result indicates that SRA interacts with TrxG through WDR5 and with PRC2 via EED and SUZ12 . Purified EZH2 , when not part of the PRC2 complex , shows no selective affinity for sense as compared to anti-sense SRA ( Fig 1C ) . To a lesser extent this is true for RBBP5 , which as an isolated component shows some binding to anti-sense SRA , unlike other members of the TrxG complex ( Fig 1B ) . It is clear however that the full complexes , and most of their components , exhibit selective binding to sense SRA . Domain mapping analysis , in which the 5’ or 3’ halves of the SRA molecule are separately tested for their ability to interact with TrxG and PRC2 complexes , suggests that the TrxG and PRC2 complexes preferentially bind to the 5’ and 3’ regions of SRA , respectively ( Fig 1D and S4 Fig ) . We note that the secondary structure of SRA [30] harbors distinct domains that might be specialized to interact with the TrxG and PRC2 complexes . These observations raise the question whether SRA might simultaneously bind to both TrxG and PRC2 , thereby in principle allowing for delivery of both activating and silencing marks . Co-immunoprecipitation experiments were performed using recombinant TrxG and PRC2 complexes in the presence of either antisense or sense SRA . Immunoprecipitation of RBBP5 resulted in an enrichment of EED when sense SRA was present in the reaction ( Fig 1E ) . Similarly , immunoprecipitation of EZH2 led to an enrichment of WDR5 in the presence of sense , but not antisense SRA . These results indicate that TrxG , PRC2 and SRA are present in the same complex . However they do not distinguish between a complex in which a single SRA molecule binds both TrxG and PRC2 , and , for example , a complex containing two or more SRA molecules , each separately carrying either TrxG or PRC2 . Nonetheless , the experiment in Fig 1D suggests that the binding domains on SRA for each complex are largely independent of each other and should be capable of binding both complexes at once To determine whether SRA displays the same bi-faceted binding properties in vivo , shRNA silencing of SRA was employed to deplete SRA expression in NTERA2 ( S5 Fig ) . Immunoprecipitation of RBBP5 co-precipitated EZH2 in control knockdown cells ( Fig 1F ) . However , this interaction of EZH2 and RBBP5 was reduced in SRA knockdown cells . This result is consistent with the in vitro interaction assay and suggests that SRA may be capable of delivering both activating and silencing histone modifications to sites where it is bound . The lncRNA SRA and RNA DEAD box helicase p68 have been implicated as acting together in transcriptional regulation , yet their mechanism of action remains elusive . If SRA in the absence of other components can recruit both the TrxG and PRC2 complexes , what role does p68 play ? We therefore sought to establish whether p68 might modulate SRA/TrxG/PRC2 interactions , altering the affinity of SRA for these complexes . An SRA pull down assay shows that the amount of interacting TrxG complexes is increased when p68 is present in the reaction ( Fig 2A ) . This property of p68 to promote TrxG recruitment by SRA is not due to an interaction between p68 and TrxG , since p68 does not directly associate with TrxG complexes ( S6 Fig ) . In contrast , the ability of SRA to pull down PRC2 complex is not altered by p68 ( Fig 2A ) . We obtained similar results using the p68 homolog p72 . To confirm in vivo the function of p68 in promoting SRA and TrxG interaction , RNA immunoprecipitation was carried out with an antibody recognizing RBBP5 after using shRNA to knock down p68 ( S7 Fig ) . The result shows that enrichment of SRA bound to TrxG complex , but not to PRC2 , was reduced in p68 knockdown cells ( Figs 2B and S8 ) . These results thus reveal a role of p68 in facilitating interaction between the lncRNA SRA and the activating epigenetic machinery of the TrxG complex . SRA interacts directly with TrxG and PRC2 complexes . The function of PRC2 involves methylation of histone H3 lysine 27 . The TrxG complex carrying MLL2 is responsible for trimethylation of histone H3 lysine 4 in mouse embryonic stem cells [1 , 2] , particularly at bivalent sites . We therefore asked whether SRA might be present at bivalent domains . To this end , we utilized the ChIRP technique [19] to pull down the lncRNA SRA from chromatin of the human pluripotent stem cells NTERA2 . Using next generation sequencing , we identified 7 , 899 SRA-binding sites genome-wide ( see Methods ) . Comparing SRA with profiles of H3K4me3 and H3K27me3 in NTERA2 generated by the ENCODE project , we find that 1 , 570 and 735 sites representing 20% and 9 . 3% of total SRA binding sites possess respectively either the H3K4me3 or H3K27me3 modification exclusively ( Fig 3A ) . Among SRA binding sites , 894 regions representing 11% have the bivalent domain signature ( Fig 3A , 3D and S9 Fig ) . Taken together , about 40% of SRA sites carry at least one of these modifications . Of all bivalent domains we mapped , 8% are associated with SRA binding . Gene classification analysis reveals that SRA-bound regions are associated with differentiation and embryonic development genes ( Fig 3B ) . This result is consistent with the observed interaction in vitro and in vivo between SRA and TrxG/PRC2 complexes , and with a role for SRA in targeting histone modifications , including bivalent modifications , in pluripotent stem cells . Because p68 facilitates interaction between SRA and WDR5 containing complexes , we asked whether sites of H3K4me3 modification might be enriched at genomic regions occupied by both SRA and p68 relative to those occupied by SRA alone . Chromatin immunoprecipitation ( ChIP ) sequencing of p68 in NTERA2 identified 14 , 131 binding sites genome wide; functions of many associated genes are involved in embryonic development ( Fig 3C ) . It is obvious from our data that many sites of H3K4 or H3K27 methylation are associated neither with p68 nor SRA , consistent with the existence of multiple mechanisms for delivering those modifications . However if we focus on the role of SRA and its interaction with p68 , we find that 16% of SRA binding sites are also occupied by p68 ( Fig 3D–3F and S10 Fig ) . Furthermore 21% of SRA/p68 binding sites are located at bivalent sites that harbor both H3K4me3 and H3K27me3 marks ( S11 Fig ) . Interestingly , we observe a significant 19% ( 47% versus 28% ) increase ( p-value < 10−4 , Fisher’s exact test ) in sites carrying the H3K4me3 modification at genomic regions occupied by both SRA and p68 compared with those occupied by SRA but lacking p68 ( Fig 3E ) . To investigate whether p68 facilitates modification of H3K4me3 , we performed ChIP-PCR of the histone mark at selected p68-bound genes upon silencing of p68 . Depletion of p68 led to a decrease in H3K4me3 occupancy at half of the p68-bound genes we examined ( S12 Fig ) . On the other hand , the presence of p68 at SRA binding sites has an insignificant effect on the extent of H3K27me3 modification ( 23% versus 20% ) ( Fig 3F ) . The genome-wide accumulation of H3K4me3 at p68-associated SRA binding sites thus suggests a role in vivo for p68 in facilitating SRA mediated H3K4 methylation , consistent with our observations in vitro that p68 stabilizes SRA-TrxG interaction . We have previously shown that CTCF , a DNA binding protein , interacts with p68/SRA in nuclear extracts , and that p68 binding is essential to CTCF dependent insulator function at the human IGF2/H19 imprinted locus [26] . However unlike the interactions of SRA with TrxG or PRC2 , the interaction between CTCF and p68/SRA is indirect ( S13 Fig ) . Analysis of SRA ChIRP data from the pluripotent stem cells NTERA2 cells shows that not all CTCF sites are associated with SRA . Nonetheless , recruitment of SRA by CTCF increases the probability that the site will also be bivalent: 14 . 3% of sites occupied by both CTCF and SRA also carry bivalent marks , whereas only 7 . 3% of CTCF sites not associated with SRA are bivalent ( S14 Fig , p-value < 10−4 ) . The presence of SRA at CTCF binding sites thus correlates with the presence of bivalent domains . We next asked whether the core transcription factors NANOG , OCT4 and SOX2 , which have been shown to occupy sites at bivalent genes in human pluripotent stem cells [31 , 32] , might interact with SRA as a means to recruit the lncRNA to their target genes . RNA pull down experiments using either nuclear extract or recombinant proteins reveal a direct association between SRA and NANOG , but our data do provide evidence for such association of OCT4 or SOX2 ( Fig 4A and 4B ) . Further , co-immunoprecipitation of p68 and NANOG in the presence of sense or antisense SRA shows that SRA facilitates specific complex formation between p68 and NANOG ( Fig 4C ) . Using a publicly available ENCODE database of NANOG ChIP-seq in human embryonic stem cells , we find that 16% of SRA binding sites detected in NTERA2 cells overlap with NANOG ( S15 Fig ) . Unlike for CTCF , we do not find a correlation between SRA co-localization with NANOG and the abundance of bivalent domains . However , 16 . 5% of NANOG-SRA binding sites also show bivalent association ( S16 Fig ) . At NANOG-SRA binding sites , the H3K4me3 mark associates with 75% of regions when p68 is present ( NANOG/SRA/p68/K4Me3 vs all NANOG/SRA/p68 ) compared with 51% of this modification at these regions without p68 ( NANOG/SRA/K4Me3 no p68 vs all NANOG/SRA no p68 ) ( S16B Fig , p-value < 10−4 ) . In contrast , a 5% reduction of H3K27me3 co-occupancy is observed for NANOG-SRA binding sites when p68 is present ( S16C Fig ) . Thus , similar to the above observation for all SRA associated sites , the presence of p68 at NANOG-SRA binding sites appears to facilitate the establishment of H3K4 methylation . As the TrxG and PRC2 complexes are important for reprogramming of somatic cells toward induced pluripotent stem cells [4 , 33 , 34] , we tested whether SRA is also important for this process . Human fibroblasts were transfected with a plasmid encoding OCT4 , SOX2 , c-MYC and KLF4 , and were grown under feeder-free human pluripotent stem cell conditions for 30 days . We find that , when SRA expression is depleted , the numbers of alkaline phosphatase and SSEA3 positive colonies are reduced ( Fig 4D and 4E ) . This observation suggests that , similar to TrxG and PRC2 complexes , SRA is a crucial factor for the reprogramming of fibroblasts toward induced pluripotent stem cells . Additionally , we find that silencing of SRA leads to a decrease in number of cells expressing the pluripotent stem cell marker SSEA3 , while the number of cells expressing the differentiation marker A2B5 is increased ( S17 Fig ) . This result indicates that SRA is important for maintaining the stem cell state . However , because silencing of SRA results in a decrease in self-renewal , we are unable to carry out experiments to study the effects of SRA depletion on histone modifications while maintaining the stem cell identity of NTERA2 cells .
The enzymatic mechanisms and cofactors underlying H3K4 and K27 trimethylation have been well characterized . However , little is known about mechanisms that could selectively generate a bivalent domain , which carries both kinds of methylation marks . In the present study , we have identified SRA as a lncRNA interacting with both the TrxG and PRC2 complexes . As discussed in the Introduction , several lncRNAs have been shown to bind either to TrxG or PRC2 [12 , 13 , 15–17] . To date , the only lncRNA known to interact with both TrxG and PRC2 is Fendrr [35] . However , it is not known whether the interaction between Fendrr and the two histone modifying complexes is direct , or whether Fendrr can deliver those complexes simultaneously . In NTERA2 cells , 11% of SRA-binding sites genome-wide overlap with bivalent domains , and another 29% are associated with sites carrying either H3K4me3 or H3K27me3 . This suggests that , depending upon the site , SRA can deliver either or both of these modifications , in the latter case consistent with the presence of a bivalent mark . Although SRA possesses a potential to interact with both TrxG and PRC2 , 20% of SRA-binding sites are occupied by H3K4me3 but not H3K27me3 , whereas only 9% of SRA-binding sites are marked by H3K27me3 but not H3K4me3 . Our finding therefore supports a preferred role of SRA as a transcriptional co-activator [29] . SRA frequently functions with p68 as a complex that can in turn interact with a variety of DNA-binding transcription factors such as MyoD . But as shown here for SRA-NANOG , SRA in some cases does not require the assistance of p68 . Our data nonetheless show that the presence of p68 enhances interaction between SRA and the TrxG complex in experiments carried out either with purified components or with nuclear extracts ( Fig 2 ) . The role of p68 in increasing SRA-TrxG interaction is analagous to that of ATRX , which increases interaction between Xist and PRC2 [36] . Consistent with these observations , the presence of p68 at SRA sites in NTERA2 cells in vivo increases the co-occupancy between SRA and H3K4me3 from 29% to 52% ( Fig 3E ) . These findings reveal the mutual relationship between p68 and SRA in transcriptional activation . Many DNA-binding transcription factors have been reported to interact with SRA , either directly or indirectly [22] . Our study shows that SRA directly interacts with the homeodomain transcription factor NANOG , which occupies regulatory elements of many genes associated with bivalent domains in human pluripotent stem cells [31 , 32] . We find that SRA and NANOG share binding sites genome-wide . NANOG is a key transcription factor required for self-renewal of human and mouse embryonic stem cells [37–39] and for establishment of pluripotency [40] . Similar to the latter function of NANOG , TrxG and PRC2 complexes are also important for reprogramming of the pluripotent state [4 , 33 , 34] . Our results suggest that NANOG recruits SRA and its associated TrxG and PRC2 complexes as part of the mechanism for establishing the pluripotency of induced pluripotent stem cells , and at least in some cases plays a role in establishing and/or maintaining bivalent domains ( see model in S18 Fig ) . Our results also show that SRA localization sites are widespread in the genome , and that they are likely to be involved at those sites in delivery of both activating and silencing histone modifications . The SRA/TrxG/PRC2 complexes can be recruited directly or indirectly to binding sites on DNA through interaction with a variety of transcription factors , only some of which have so far been identified . CTCF is a ubiquitous factor that appears to contribute to establishment of bivalent states at sites where SRA is also present . In addition to recruiting both MLL1 and MLL2 , which trimethylate H3K4 , SRA recruits both SETD1A and SETD1B , raising the possibility that it may mediate histone H3 monomethylation as well as trimethylation . Many other factors ( such as MyoD and NANOG ) are lineage specific; it will be important to investigate in other cell types the interaction of the SRA/TrxG/PRC2 complexes with lineage specific transcription factors , and their role in establishing patterns of histone modification important for regulation of gene expression .
A plasmid containing SRA sequence ( BC067895 . 1 ) was purchased from Open Biosystems . The SRA coding sequence was subcloned into pLITMUS28i ( New England Biolabs ) for in vitro transcription ( see below ) . The following plasmids were used for in vitro transcription/translation: pSG5-MYC encoding p68 and p72 ( gift from Prof . Frances V . Fuller-Pace , University of Dundee , UK ) ; pcDNA3 . 1-NANOG ( Addgene ) . See S1 Table for the list of antibodies used in this study . Human pluripotent stem cell line NTERA2 was grown in DMEM supplemented with 10% FBS ( Gibco ) at 37°C under a humidified atmosphere of 5% CO2 in air . At confluent , cells were passaged every three days using 0 . 25% trypsin ( Gibco ) . For establishment of NTERA2 stable knockdown cell lines , the plasmids pMLP-shRNA targeting SRA , p68 or scramble control ( transOMIC ) were linearized by NdeI and transfected into 1x106 cells using nucleofector ( Amaxa ) according to manufacturer’s protocol . Cells were immediately grown in DMEM-F12 plus 10% FBS . On day 3 , stable cell lines were selected using puromycin at 3 μg/ml final concentration . RNA pull down experiments were performed as previously described [41] . First , DNA fragments encoding full length , 5’ and 3’ domains of lncRNA SRA were cloned into pLITMUS28i ( New England Biolabs ) , and the DNA sequence was confirmed by sequencing . To generate antisense or sense SRA transcripts , the plasmid containing full length SRA was linearized by StuI or BglI , respectively . Biotinylated SRA was in vitro transcribed using HiScribe T7 In Vitro transcription kit ( New England Biolabs ) in the presence of biotin-14-CTP ( Invitrogen ) according to the instruction manuals . Transcribed RNA products were DNase-treated ( Ambion ) , purified by ethanol precipitation and verified by northern blotting . For RNA pull downs using nuclear extract , 3 μg of in vitro transcribed RNA was prepared in RNA structure buffer ( Tris-Cl pH 7 . 5 , 0 . 1 M KCl , 10 mM MgCl2 ) and incubated at 78°C for 3 min . The RNA was then gradually cooled down to 37°C . Five hundred micrograms of NTERA2 nuclear extract , prepared using NE-PER Nuclear Protein Extraction Kit ( Pierce ) , was mixed with the RNA in immunoprecipitation buffer ( PBS plus 0 . 1% Triton X-100 , 1 mM DTT , protease inhibitor cocktail , PMSF , 80 U RNase inhibitor ) in a total volume of 500 μL . The reaction was incubated for 4 hr at 4°C with rotation . MyOne Streptavidin C1 beads were prepared according to manufacturer’s recommendation , and used at 50 μL per sample . The RNA-beads complex was further incubated overnight . Beads were washed five times with immunoprecipitation buffer and boiled with 50 μL of SDS loading buffer . Twenty microliters was loaded onto Novex precast gel ( Invitrogen ) . For RNA pull down using recombinant proteins , 0 . 3 μg of RNA was used per pull down reaction with 3 μg of protein complex or 1 μg of individual protein . TrxG and PRC2 complexes were purchased from BPS Bioscience and Cayman Chemical . NANOG , OCT4 and SOX2 were purchased from Fitzgerald Industries International . The RNA helicases p68 and p72 were in vitro translated using the TNT Coupled Reticulocyte Lysate System ( Promega ) . A plasmid encoding luminescence protein was used as negative control ( Promega ) . Recombinant NANOG was also produced by in vitro translation using a Wheat Germ System ( Promega ) . All in vitro translated proteins were verified by western blotting . Ten microliters of translated protein product was used per RNA pull down reaction . For in vitro co-immunoprecipitation in the presence of antisense or sense SRA , the RNAs were transcribed without Biotin-14-CTP . Three micrograms of TrxG and PRC2 complexes or 10 μL of in vitro translated p68 and NANOG were used for co-immunoprecipitation in 200 μL of immunoprecipitation buffer . Antibodies for immunoprecipitation were used at 3 μg including mouse anti-RBBP5 ( MABE220 , Upstate ) , mouse anti-EZH2 ( MA5-15101 , Thermo Scientific ) and rabbit anti-DDX5 ( A300-523A , Bethyl Laboratories ) . For co-immunoprecipitation using nuclear extract , 500 μg of NTERA2 nuclear extract was mixed with 3 μg of relevant antibodies in a total of 500 μL of immunoprecipitation buffer . The reaction was incubated for 4 hr at 4°C with rotation . Protein A and protein G conjugated magnetic beads were prepared according to manufacturer’s recommendation ( Invitrogen ) , and used at 50 μL per sample . The complex was then further incubated overnight . Beads were washed five times with immunoprecipitation buffer and boiled with 50 μL of SDS loading buffer . Twenty microliters was loaded onto Novex precast gel ( Invitrogen ) . RNA was extracted using TRIzol reagent ( Invitrogen ) and DNase-treated ( DNA-free kit , Ambion ) . Complementary DNA synthesis was performed with 1 μg RNA using a Maxima First Strand cDNA Synthesis Kit ( Thermo Scientific ) . qPCR was carried on by using Power SYBR Green PCR Master Mix ( Applied Biosystems ) in a total volume of 20 μl each well with 7900HT real-time PCR system ( Applied Biosystems ) . Gene expression was normalized by expression level of ACTB . Primer sequences are available upon request . Twenty million cells were fixed with 1% formaldehyde in PBS for 10 min at room temperature . The fixation was quenched by adding glycine at 125 mM final concentration and incubated further for 5 min . Cells were washed and collected by centrifugation at 1500 rpm for 5 min . Nuclear extract was prepared by using NE-PER Nuclear Protein Extraction Kit ( Pierce ) . Three micrograms of antibody was added to 500 μg of the nuclear extract in immunoprecipitation buffer ( PBS , 1 mM DTT , protease inhibitor cocktail , PMSF , 80 U RNase inhibitor ) in a total volume of 500 μL . The complex was incubated at 4°C for 4 hr . Protein A and protein G conjugated magnetic beads were used at 50 μL per sample . The complex was then further incubated overnight . Beads were washed five times and resuspended in 100 μL proteinase K buffer ( 10 mM Tris-Cl pH 7 . 5 , 100 mM NaCl , 1 mM EDTA , 0 . 5% SDS ) with 5 μL proteinase K ( New England Biolabs ) . Samples were incubated at 50°C for 45 min with shaking , and boiled at 95°C for 10 min . Samples were mixed with 500 μL Qiazol by vigorous vortexing , and were incubated at room temperature for 10 min . RNA extraction was then performed using miRNeasy mini kit ( Qiagen ) . qPCR was employed to detect RNA binding . ChIP was performed according to the manufacturer’s instruction ( Active Motif ) . Briefly , 2 x 107 cells were fixed with 1% formaldehyde in PBS for 10 min at room temperature . The fixation was then quenched by adding glycine . Cells were washed and collected by centrifugation at 1500 rpm for 5 min . Nuclei were sonicated twice using Bioruptor ( Diagenode ) at maximum power , 30 sec ON and 30 sec OFF for 7 . 5 min to obtain chromatin fragments ranging from 200–1000 bp . Fifty micrograms of sheared chromatin was used per IP with 3 μg antibody . Retrieved DNA fragments were purified by QIAquick PCR Purification Kit ( Qiagen ) or ethanol precipitation . Primer sequences for ChIP are listed in S2 Table . ChIRP analysis was performed according to published protocols with minor modifications based on ChIRP and Capture Hybridization Analysis of RNA Targets ( CHART ) techniques [19 , 42 , 43] . Briefly , 3x107 cells were fixed with 1% glutaraldehyde for 10 min at room temperature with shaking . The fixation was stopped by adding glycine . Crosslinked cells were washed with PBS , and resuspended in 1 ml swelling buffer ( 25 mM HEPES pH 7 . 3 , 10 mM KCl , 0 . 1% NP-40 , 1 mM DTT , PMSF ) . Samples were incubated at 4°C for 30 min with shaking , and were collected by centrifugation . The pellet was resuspended with 350 μL of ChIRP lysis buffer , and was sonicated using Bioruptor ( Diagenode ) at maximum power , 30 sec ON and 30 sec OFF for 7 . 5 min of 6 cycles to obtain chromatin fragments ranging from 100–1000 bp . Sheared chromatin was then collected by centrifugation . Two hundred micrograms of sheared chromatin sample was pre-cleared for 1 hour using 100 μL of Ultralink-streptavidin beads ( Pierce ) at room temperature with shaking . The sample was then centrifuged , and supernatant was collected . The pre-cleared chromatin was used per hybridization reaction with 10 μL of 100 μM pooled 3’ Biotin TEG oligonucleotide probes ( Integrated DNA Technologies ) . SRA probes were designed to cover SRA transcript at nucleotide position 124 to 1473 ( accession number NR_045587 . 1 ) ( See S3 Table for the probe sequences ) . LacZ probes were employed as negative control [19] . The sample and the probes were hybridized at 37°C for 4 hours with shaking . Once the hybridization was completed , 100 μL of C-1 magnetic beads ( Invitrogen ) was mixed with the sample to pull down the biotinylated probes . DNA was eluted in the presence of 12 . 5 mM D-Biotin ( Invitrogen ) . DNA was ethanol precipitated and subjected to library preparation . Library preparation was performed using TruSeq ChIP Sample Preparation Kit ( Illumina ) or MicroPlex Library Preparation Kit ( Diagenode ) according to manufacturer’s instruction . Three biological triplicates were used for ChIRP-seq and ChIP-seq . Briefly , 5–10 ng of DNA starting material , which was quantified by Qubit ( Invitrogen ) , was used for each biological sample . The DNA was end-repaired , 3’ adenylated , and ligated with adapters . Then the ligated DNA was size-selected to obtain DNA fragments at 250–300 bp by agarose gel electrophoresis . The purified DNA was amplified to enrich the library . The final PCR product was purified by Agencourt AMPure XP beads ( Beckman Coulter ) and was submitted to the NIDDK Genomic Core Facility for high-throughput sequencing using Illumina HiSeq2500 . The sequencing was performed with the run type of single-end , 50 bp read . Data were aligned against the human genome version human_hg19 , and were exported into BAM file format . Aligned reads of SRA ChIRP-seq and p68 ChIP-seq were filtered with SAMTools program to remove duplicates and tags with a map quality score less than 20 [44] . MACS version 1 . 4 . 2 was used for peak calling with a threshold of p-value less than 10−5 for p68 and 10−4 for SRA [45] . The resulting BED files from each of biological triplicate samples were intersected using the Bioconductor package ChIPpeakAnno [46] . For SRA ChIRP-seq analysis , the SRA probe dataset was further subtracted against a LacZ probe dataset using bedtools [47] . The dataset was further screened for possible homologies with the RNA probes used in the analysis . No significant number of sequences was found with zero , one or two mismatches . Intersection between any of SRA , p68 , NANOG and CTCF was performed with maximum distance of 500 bp between peaks . For the binding comparison between the above factors and H3K4me3 and H3K27me3 , a distance less than 2 kb was allowed . BED files were visualized and exported using IGV [48] . H3K4me3 and H3K27me3 ChIP-seq data were taken from the ENCODE project of NTERA2 cells . NANOG and CTCF ChIP-seq data were taken from the ENCODE project of human embryonic stem cells H1 . Gene classification analysis was performed using GREAT [49] . The Fishers Exact test to measure peak enrichment was taken from the Fisher’s exact function from the R package for statistical computing [50] . SRA ChIRP- and p68 ChIP-Sequencing data were submitted to GEO Datasets under accession number GSE58641 . Human fibroblast cell line WI-38 at 1x106 cells were transfected with a single plasmid encoding the four reprogramming factors OCT4 , SOX2 , c-MYC and KLF4 [51] using Nucleofector with scrambled siRNA or ON-TARGETplus siRNA targeting SRA ( Dharmacon ) . Transfected fibroblasts were plated in six-well plates under Essential 8 medium ( Invitrogen ) . The siRNA knockdown was also performed consecutively at first and second week post-transfection using Lipofectamine RNAiMAX Reagent ( Invitrogen ) . Plates were collected on day 30 and were analyzed for expression of surface markers of human pluripotent stem cells . Alkaline phosphatase staining was performed using Alkaline Phosphatase Detection Kit ( Millipore ) per manufacturer’s instruction . Cells were fixed with 4% PFA and were incubated with a monoclonal antibody against SSEA3 ( gift from Prof . Peter W . Andrews , University of Sheffield , UK ) . A goat anti-mouse secondary antibody ( IgG+IgM ) conjugated with FITC ( Cayman Chemical ) was then used for visualization under a fluorescence microscope ( EVOS ) . Single cells were collected by trypsinization , and resuspended with 10% FBS in PBS buffer . Primary antibodies SSEA3 and A2B5 ( 1 in 10 dilution ) were added into cell suspensions containing 1x105 cells in 100 μL volume . P3X was used as a negative control antibody . The reaction was incubated on-ice for 30 min , washed with PBS and resuspended with 10% FBS in PBS . One μL of FITC-conjugated Goat anti-Mouse antibody ( Cayman Chemical ) was added to the reaction . The reaction was incubated on-ice for another 30 min . Flow cytometric analysis was performed using Cytomic FC500 ( Beckman Coulter ) . | Long non-coding RNAs ( lncRNAs ) can play an important role in regulation of gene expression . In a number of cases , individual lncRNAs have been shown to interact with either the trithorax group ( TrxG ) or polycomb repressive complex 2 ( PRC2 ) protein complexes , which deliver histone modifications associated respectively with transcriptionally active or inactive chromatin . Here we show that the lncRNA , SRA , unusually forms complexes with both TrxG and PRC2 complexes . Consistent with this property , some SRA binding sites in human pluripotent stem cells overlap with bivalent domains , which carry both kinds of histone modifications . We find that SRA complexed with the helicase protein , p68 , shows enhanced binding of TrxG complex , but not of PRC2 . This is reflected in genome wide enriched ‘activating’ histone modifications at SRA sites also occupied by p68 . We show that in human pluripotent stem cells SRA also interacts with NANOG , a principal determinant of pluripotency , and is important for maintenance of the pluripotent state . SRA may be involved in the delivery of histone modifications associated with either activation or silencing of gene expression , and in some cases could deliver both . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Association of the Long Non-coding RNA Steroid Receptor RNA Activator (SRA) with TrxG and PRC2 Complexes |
The central terminals of primary afferent fibers experience depolarization upon activation of GABAA receptors ( GABAAR ) because their intracellular chloride concentration is maintained above electrochemical equilibrium . Primary afferent depolarization ( PAD ) normally mediates inhibition via sodium channel inactivation and shunting but can evoke spikes under certain conditions . Antidromic ( centrifugal ) conduction of these spikes may contribute to neurogenic inflammation while orthodromic ( centripetal ) conduction could contribute to pain in the case of nociceptive fibers . PAD-induced spiking is assumed to override presynaptic inhibition . Using computer simulations and dynamic clamp experiments , we sought to identify which biophysical changes are required to enable PAD-induced spiking and whether those changes necessarily compromise PAD-mediated inhibition . According to computational modeling , a depolarizing shift in GABA reversal potential ( EGABA ) and increased intrinsic excitability ( manifest as altered spike initiation properties ) were necessary for PAD-induced spiking , whereas increased GABAAR conductance density ( ḡGABA ) had mixed effects . We tested our predictions experimentally by using dynamic clamp to insert virtual GABAAR conductances with different EGABA and kinetics into acutely dissociated dorsal root ganglion ( DRG ) neuron somata . Comparable experiments in central axon terminals are prohibitively difficult but the biophysical requirements for PAD-induced spiking are arguably similar in soma and axon . Neurons from naïve ( i . e . uninjured ) rats were compared before and after pharmacological manipulation of intrinsic excitability , and against neurons from nerve-injured rats . Experimental data confirmed that , in most neurons , both predicted changes were necessary to yield PAD-induced spiking . Importantly , such changes did not prevent PAD from inhibiting other spiking or from blocking spike propagation . In fact , since the high value of ḡGABA required for PAD-induced spiking still mediates strong inhibition , we conclude that PAD-induced spiking does not represent failure of presynaptic inhibition . Instead , diminished PAD caused by reduction of ḡGABA poses a greater risk to presynaptic inhibition and the sensory processing that relies upon it .
Synaptic inhibition regulates transmission of sensory signals through the spinal cord . Importantly , numerous chronic pain conditions are associated with diminished inhibition [1–5] and pharmacological blockade of inhibition at the spinal level has been shown to reproduce many features of those chronic pain conditions [6–9] . Decreased transmitter release , reduced GABAA/glycine receptor function , and altered chloride regulation are all potential disinhibitory mechanisms , but pre- and postsynaptic inhibition are not equally susceptible to certain pathological changes; for instance , the potassium-chloride co-transporter KCC2 is not expressed in primary afferent neurons , meaning disinhibitory effects of KCC2 downregulation [10] are attributable entirely to reduced postsynaptic inhibition , in cells that express KCC2 . KCC3 is expressed in some primary afferents and can extrude chloride under isosmotic conditions [11 , 12] but it remains unknown whether KCC3 is altered under pathological conditions . Yet selective disruption of presynaptic inhibition can cause mechanical and thermal hypersensitivity [13] and presynaptic expression of the α2 GABA receptor subunit is necessary for the antihyperalgesic effect of diazepam [14] . These observations affirm that presynaptic GABAAR-mediated inhibition also plays a key role in nociception . Pre- and postsynaptic inhibition in spinal cord are mechanistically distinct . Postsynaptically , in mature spinal neurons , the reversal potential associated with GABAAR ( EGABA ) is normally around -70 mV [10] , meaning GABAAR activation reduces depolarization caused by concurrent excitatory input . Presynaptically , in the central terminals of primary afferents , EGABA is normally around -35 mV because chloride is actively loaded into primary afferents by the sodium-potassium-chloride co-transporter NKCC1 [13 , 15–17] , thus GABAAR activation causes depolarization . Contrary to the presumed excitatory effect of depolarization , primary afferent depolarization ( PAD ) mediates inhibitory effects via sodium channel inactivation and shunting [18–21] . However , PAD can sometimes trigger spikes that conduct antidromically , thus producing what is referred to as a dorsal root reflex ( DRR ) [22] . One theory holds that antidromically conducted spikes mediate an inhibitory effect by colliding with and blocking othrodromically conducted spikes originating in the periphery [23 , 24]; however , collisions are unlikely since the latency to travel the full length of the nerve is short relative to the interspike interval at realistic spiking rates . PAD-induced spikes are unlikely to trigger synaptic release from the PAD-affected branch because spike amplitude is attenuated , but PAD-induced spikes that manage to propagate to adjacent , PAD-free branches may trigger synaptic release [25] . The experiments required to test these model predictions are prohibitively difficult . The above theory was formulated for large myelinated proprioceptive afferents involved in locomotion; in contrast , within smaller afferents responsible for nociception , the prevailing view is that PAD-induced spikes occur only under pathological conditions and that DRRs contribute to neurogenic inflammation and hypersensitivity [22 , 26] . Within this context , PAD-induced spiking is thought to represent conversion of PAD from an inhibitory process to an excitatory one [22] . With respect to biophysical mechanisms , PAD-induced spiking requires GABAAR activation [27] and NKCC1-mediated chloride loading [28] . Enhanced chloride loading and the consequent depolarizing shift in EGABA has been hypothesized to facilitate PAD-induced spiking [29 , 30] . Nerve injury increases NKCC1 protein levels and PAD [13 , 31] , and although total NKCC1 expression is not altered by inflammation [32 , 33] , NKCC1 membrane trafficking and phosphorylation are affected by painful stimuli [34] . Notably , inflammation causes a depolarizing shift in EGABA [35] and promotes DRRs in C and Aδ fibers [36] . Increased GABAAR density and reduced low-threshold potassium channel density have also been hypothesized to promote DRRs [35 , 37] but the full set of requirements for PAD-induced spiking remains unclear . We sought to identify which biophysical changes , alone or together , enable PAD-induced spiking and how such changes impact PAD-mediated inhibition .
Starting with computer simulations , we co-varied EGABA and intrinsic excitability ( controlled by βw; see Methods ) while keeping ḡGABA fixed at 2 nS/pF . The light grey and dark grey regions of the resulting 2-D bifurcation diagram ( Fig 1A ) show the EGABA and βw combinations that produce transient and repetitive spiking , respectively . Spiking pattern was determined by the response to GABA conductance “steps” . To more accurately simulate different forms of synaptic transmission , other conductance waveforms were tested: phasic inhibition via intrasynaptic GABAAR was modeled by a “fast” synaptic waveform ( τrise = 2 ms; τdecay = 20 ms; see Eq 6 ) ; tonic inhibition corresponds to the sustained component of the conductance step , but we also tested a “slow” synaptic waveform with intermediate kinetics to simulate spilled-over GABA asynchronously activating extrasynaptic GABAAR ( τrise = 20 ms; τdecay = 200 ms ) . Fig 1B shows responses to each stimulus waveform for parameter combinations labeled a-f on Fig 1A . Under control conditions used in the experiments described in this study ( EGABA = -35 mV and βw = -20 mV; point b ) , GABA conductance caused depolarization but no spiking . PAD-induced repetitive spiking required a combined depolarizing shift in EGABA and βw ( point e ) whereas transient spiking required a smaller increase in βw ( point c ) and could result solely from a large change in EGABA . By comparison , an isolated change in βw could not enable PAD-induced spiking . As illustrated in panel c of Fig 1B , slow-onset GABAAR input required stronger input to elicit spiking because transient spiking involves a spike initiation mechanism that is sensitive to the rate of depolarization [40] . Next , we repeated the 2-D bifurcation analysis for different ḡGABA values to produce a family of curves ( Fig 1C ) . The dashed curve demarcating the minimum requirements for transient spiking shifted downward as ḡGABA was increased . The solid curve demarcating the minimum requirements for repetitive spiking also shifted downward for an initial increase in ḡGABA but shifted rightward as ḡGABA was increased further , indicating that GABAAR activation is maximally excitatory at intermediate densities . Somatic recordings have demonstrated somatic ḡGABA between 0 . 2 and 0 . 5 nS/pF [35] and the absolute ḡGABA values reported by Chen et al . [13] correspond to approximately 0 . 1 nS/pF after conversion to densities based on estimated surface areas . Axonal ḡGABA may differ from somatic ḡGABA ( given precedents for differential ion channel distribution [41] ) but measuring ḡGABA in central axon terminals is prohibitively difficult . Our experimental approach does not rely on measuring axonal ḡGABA but , instead , was designed to determine the minimum ḡGABA required ( for different EGABA and intrinsic neuronal excitability ) to enable PAD-induced spiking; comparing this value against measured ḡGABA ( in the soma ) reveals whether the density of native GABA receptors is sufficient to evoke spiking under different conditions . It remains unclear what ḡGABA would be necessary to evoke spiking in central axon terminals . To test the simulation predictions described above , we conducted experiments in acutely dissociated DRG somata using an approach distinct from previous studies . Rather than activating native GABAARs by puffing GABA ( which would produce a current whose conductance , reversal potential and kinetics are not easily measured or independently manipulated ) , we used dynamic clamp to apply a virtual conductance whose parameters are precisely and independently controllable . In this way , we quantified the minimum virtual ḡGABA required to elicit spiking under different conditions . Importantly , because virtual ḡGABA can be higher than native ḡGABA , the density of native GABAAR does not limit our studies; indeed , failure of GABA puffs to evoke spikes in previous studies [13 , 35 , 42] suggests that somatic ḡGABA is normally too low to produce spikes , but ḡGABA may be higher in central axon terminals . In dynamic clamp , the voltage recorded from a neuron is passed to a computer , which , in real time , uses voltage to calculate current that is injected back into the patched neuron , thereby introducing a virtual conductance [43] . This approach allows manipulations to be applied like in computer simulations but to real neurons , such that we can avoid modeling the neuron ( and making any assumptions about intrinsic excitability ) and test directly how virtual GABAAR input affects native voltage-gated channels controlling spike initiation . Notably , photostimulation-based testing of axonal excitability has revealed transient spiking comparable to that observed in somata [39] but the excitability of central axon terminals remains uncertain . If central axon terminal and somatic excitability are similar , then the requirements for PAD-induced spiking ascertained for the soma can be extrapolated to those terminals; on the other hand , if those terminals are more excitable , they would operate farther to the right on the excitability axis depicted in the inset of Fig 2A . To begin , we tested virtual GABA conductances in neurons from naïve animals before and after reproducing the hyperexcitability associated with nerve injury by blocking Kv1-type potassium channels with 4-AP [44 , 45]; this corresponds in the model to setting βw to less negative values . Testing with different EGABA and stimulus waveforms , we systematically increased ḡGABA to try to elicit spiking . As illustrated for a typical cell in Fig 2A , PAD was most likely to produce spiking after application of 4-AP and a depolarizing shift in EGABA to -20 mV . Fig 2B summarizes the proportion of cells in which PAD-induced spiking was observed under each test condition . For cells from naïve animals tested with EGABA = -35 mV , 4-AP increased the proportion exhibiting PAD-induced spiking but not significantly ( p = 0 . 079; Fisher’s exact test ) whereas the 4-AP effect was highly significant for EGABA = -20 mV ( p = 0 . 004 ) . Shifting EGABA from -35 mV to -20 mV significantly increased the proportion of cells exhibiting PAD-induced spiking both before and after 4-AP ( p < 10−3 and 10−4 , respectively ) , consistent with the NKCC1 hypothesis of DRR generation [29 , 30] . But as predicted by our simulations , the proportion of cells with PAD-induced spiking was most significantly increased by the combination of 4-AP and a depolarizing shift in EGABA ( p < 10−9 ) . Within this data set , two cells were subsequently identified as outliers based on analysis of the minimum ḡGABA needed for PAD-induced spiking ( see below ) ; removing those outliers did not substantively alter the statistical results reported above . Based on cells that exhibited PAD-induced spiking before and after 4-AP for EGABA = -20 mV , the minimum ḡGABA needed to elicit spiking was significantly reduced from 0 . 49 ± 0 . 07 nS/pF ( mean±SEM ) to 0 . 16 ± 0 . 03 nS/pF by 4-AP ( p = 0 . 005 , Tukey test following ANOVA described below ) ( Fig 2C left ) . Plotting the same data against soma diameter revealed a trend towards higher minimum ḡGABA for smaller cells , but soma diameter did not have a significant effect ( p = 0 . 61 ) and nor did it interact significantly with the 4-AP effect ( p = 0 . 29; two-way repeated measures ANOVA ) ( Fig 2C right ) . Notably , we report all conductances as densities to correct for the direct effect of membrane surface area on our measurements; however , soma diameter is known to correlate with fiber type [46] , and so the insignificant effect of cell size ( after normalization by surface area ) argues that minimum ḡGABA does not differ significantly between myelinated ( A ) and unmyelinated ( C ) neurons . Of the cells that exhibited PAD-induced spiking for both EGABA values after 4-AP , the minimum ḡGABA needed to elicit spiking was significantly reduced from 0 . 30 ± 0 . 07 nS/pF to 0 . 11 ± 0 . 02 nS/pF by shifting EGABA from -35 mV to -20 mV ( p = 0 . 022 , paired t-test ) ( Fig 2D ) . Of the 10 cells tested with both fast and slow gGABA waveforms at EGABA = -20 mV after 4-AP , 7 responded to both stimuli with transient spiking and 2 responded with repetitive spiking . Among transient spiking cells , the slow waveform required higher ḡGABA than the fast waveform to elicit transient spiking ( 0 . 46 ± 0 . 09 nS/pF vs 0 . 27 ± 0 . 09 nS/pF ) which , although not a statistically significant difference ( p = 0 . 25; paired t-test ) , is consistent with a spike initiation mechanism sensitive to the rate of depolarization . By comparison , the two repetitive spiking cells required exactly the same minimum ḡGABA for the fast and slow waveforms , consistent with a spike initiation mechanism sensitive only to the amplitude of depolarization [40] . Comparing the responses to gGABA steps and ramps illustrates that the latter are far less effective in eliciting transient spiking ( Fig 2E ) . All of these experimental data are consistent with simulation results in Fig 1 and S1 Fig . Like 4-AP , nerve injury increased the proportion of cells exhibiting PAD-induced spiking ( bars on right side of Fig 2B ) . Compared against naïve cells without 4-AP , nerve injury caused no change in the proportion of cells exhibiting PAD-induced spiking for EGABA = -35 mV ( p = 1 ) whereas it did significantly increase that proportion for EGABA = -20 mV ( p = 0 . 028; Fisher’s exact tests ) . Nerve injury and treatment of naïve cells with 4-AP resulted in a similar proportion of cells with PAD-induced spiking when tested with EGABA = -35 mV and -20 mV ( p = 1 and 0 . 40 , respectively ) . Among nerve-injured cells , shifting EGABA from -35 mV to -20 mV significantly increased the proportion with PAD-induced spiking ( p = 0 . 001 ) . Consistent with the combined effects of 4-AP and altered EGABA , the proportion of cells with PAD-induced spiking was most significantly increased by the combination of nerve injury and a depolarizing shift in EGABA ( p < 10−5 ) . Testing with current injection ( Istim ) confirmed that 4-AP had the intended effect of increasing excitability yet , despite responding to Istim steps with repetitive spiking , most neurons responded to gGABA steps with transient spiking , as illustrated in Fig 3A . Specifically , PAD-induced repetitive spiking was not observed in any nerve-injured neurons and was seen in only two neurons after 4-AP application . All neurons were tested with a broad range of ḡGABA to confirm that repetitive spiking could not eventually be achieved by applying a stronger conductance . Increasing ḡGABA above the minimum required to elicit transient spiking consistently caused attenuation of the spike height and clamped the subsequent voltage near EGABA ( Fig 3B ) . Based on our simulation results ( see Fig 1A ) , we reasoned that the lack of repetitive spiking was due to 4-AP or nerve injury not causing a sufficient increase in excitability . To test this hypothesis , we further increased excitability by using dynamic clamp to introduce a virtual sodium conductance like that upregulated after nerve injury [45] . As predicted , PAD-induced repetitive spiking was made possible by this additional manipulation ( Fig 3C ) . Although we managed to reproduce PAD-induced repetitive spiking , the extent of the required manipulations suggests that naturally occurring pathological changes cause few neurons to become sufficiently hyperexcitable that PAD will induce repetitive spiking . That said , if the central terminals of axons are more excitable ( i . e . more prone to repetitive spiking ) than the soma , PAD would be more likely to elicit repetitive spiking than suggested by our data . The above results demonstrate that depolarizing GABA current can induce transient spiking under conditions associated with nerve injury . This does not , however , exclude PAD from retaining its inhibitory effects , especially given that inhibition stems from sodium channel inactivation and shunting . In fact , although PAD may induce a single spike at its onset , shunting effects persist as long as GABAAR are activated . This raises the important question of whether more spikes ( arising in the periphery or ectopically in the soma or a neuroma ) are blocked by PAD than are induced by PAD in the central axon terminals . Our initial model did not include sodium channel inactivation for the sake of simplicity; therefore , our next step was to modify the model so that a certain proportion of sodium channels , controlled by parameter p , experience inactivation ( Eqn . 7 ) . Using this new model , we set βw to 0 mV to facilitate repetitive spiking and conducted 2-D bifurcation analysis to determine the p and EGABA combinations associated with different effects of PAD ( Fig 4A ) . The grey region shows parameter combinations for which a gGABA step ( 2 nS/pF ) applied alone elicits spiking ( sample traces a and d in Fig 4B ) . The green region shows parameter combinations for which the same gGABA step inhibits spiking induced by Istim steps ( sample traces c-e in Fig 4B ) . Importantly , the green and grey regions overlap , thus demonstrating that PAD can induce spikes yet nonetheless block spikes originating by other means . Fig 4C shows the 2-D bifurcation analysis repeated for different ḡGABA values . The region of PAD-induced spiking remained unchanged ( not illustrated ) but the region of PAD-mediated inhibition expanded as ḡGABA was increased , suggesting that stronger GABAAR activation manages to terminate spiking despite a smaller proportion of inactivatable sodium channels . To measure PAD-mediated inhibition in real DRG neurons , we combined gGABA and Istim steps as done for simulations in Fig 4B . Fig 5A shows a typical neuron in which Istim elicited repetitive spiking . Interposing a gGABA step during the Istim step caused reduction or complete cessation of repetitive spiking depending on ḡGABA and EGABA . Note that spikes occurring during the gGABA step were shorter than those occurring outside the gGABA step , consistent with the shunting effect of the virtual GABA conductance . Applying the gGABA step before the onset of Istim confirmed that the former could elicit transient spiking yet still inhibit the repetitive spiking otherwise driven by Istim ( Fig 5B ) . Using the same stimulus sequence , we measured rheobase ( i . e . the minimum Istim required to elicit spiking ) for each level of ḡGABA ( Fig 5C ) . Rheobase was significantly increased by increments in ḡGABA ( p = 0 . 013 , two-way repeated measures ANOVA ) but was not significantly affected by EGABA ( p = 0 . 52 ) ( Fig 5D ) . These data confirm that PAD elicited in the cell body of DRG neurons mediates shunting inhibition even under conditions in which it can induce spiking . Activation of the calcium-activated chloride channel ANO-1 in primary afferent neurons can evoke or exacerbate pain , especially under inflammatory or neuropathic conditions [47–50] . Notably , intracellular chloride tends to be elevated under those conditions ( see Introduction ) , which may explain why ANO-1 activation is excitatory rather than inhibitory . Consistent with this , ANO-1 modulation of spiking evoked by current injection is sensitive to intracellular chloride level [51] but demonstration that ANO-1 itself evokes spiking was based on a chloride reversal potential of -18 mV [49] . Given its activation requirements [52] , we predicted that ANO-1 channels would not be activated by the GABAergic input underlying PAD; recall that GABAAR activation is necessary for PAD [22] . Nonetheless , to rule out a contribution by ANO-1 , we repeated virtual PAD experiments ( like in Fig 2 ) before and after blockade of ANO-1 channels by bath-applied 10 μM T16Ainh-A01 ( A01 ) ( Fig 6 ) . Based on the pipette solution , the chloride reversal potential for ANO-1 was -20 mV but EGABA for virtual gGABA was set to -35 or -20 mV in dynamic clamp . As predicted , ANO-1 blockade had no significant effect on the minimum ḡGABA needed to evoke spiking for EGABA = -20 mV ( p = 1 . 0 , paired t-test; Fig 6A ) and nor did it significantly affect the depolarization evoked by different ḡGABA for EGABA = -35 mV ( p = 0 . 77 , two-way repeated measures ANOVA; Fig 6B ) or have any effect on rheobase , input resistance , or resting membrane potential ( p > 0 . 05 , paired t-tests ) . The data above are based exclusively on capsaicin-responsive cells ( see Fig 6C ) since ANO-1 channels are expressed primarily in cells that express TRPV1 [47] . Notably , the response to capsaicin was reduced by ANO-1 blockade ( Fig 6D ) , consistent with Takayama et al . [49] and thus verifying the efficacy of our A01 . Based on these results , we conclude that ANO-1 channels are not activated and , therefore , do not contribute to PAD under our experimental conditions . All simulations described thus far were conducted in a single compartment model . This adequately simulates spike initiation occurring in proximity to the recording electrode , as occurs when recording from an isolated DRG soma . Although spontaneous or PAD-induced spiking may arise at the site of PAD , an important inhibitory effect of PAD in the intact fiber is to block the orthodromic propagation of spikes originating in the periphery . To test for conduction block , we converted our single-compartment model into a 3-compartment model ( Fig 7A ) . Although still very simple compared with past models used to study this topic [e . g . 19 , 25 , 53] , this model suffices to qualitatively illustrate key points relevant for the present study . Each compartment was further subdivided into equipotential segments . Based on its small diameter and the absence of nodes , this model simulates continuous propagation in an unmyelinated fiber . By applying GABA conductance to the middle compartment , we tested if that conductance can induce spikes ( originating within that compartment ) and/or block the propagation of other spikes ( evoked at the far end of adjacent compartment ) . For an EGABA value of -35 mV , gGABA never evoked spiking ( consistent with the single-compartment model ) but it did block spike propagation ( Fig 7B , left column ) . Interestingly , blocked propagation could occur even in the absence of sodium channel inactivation , therein supporting claims that shunting mediated by gGABA mediates an inhibitory effect . When EGABA was shifted to -20 mV , gGABA evoked a single spike that propagated in both directions away from the center compartment ( Fig 7B , right column ) . Yet despite this excitatory effect , propagation of other spikes was blocked in two of the three conditions illustrated . Sample traces were chosen to illustrate that large gGABA could block propagation in the absence of sodium channel inactivation but a smaller gGABA could achieve the same effect when combined with sodium channel inactivation . Fig 7C demonstrates that sodium channel inactivation can accumulate over time , thus eventually blocking spikes traveling as part of a train . These results confirm that PAD does not abruptly lose its inhibitory effects once able to induce its own spikes .
Using computer simulations and an experimental approach distinct from previous studies , we have identified which pathological changes are necessary and sufficient to enable PAD-induced spiking . We determined that a depolarizing shift in EGABA is necessary yet insufficient to enable PAD-induced spiking in most DRG neurons . An increase in intrinsic excitability ( i . e . altered spike initiation properties ) is also necessary , especially to enable PAD-induced repetitive spiking . Neurons may experience both changes after nerve injury or inflammation , meaning PAD-induced spiking could occur in certain pathological conditions [22 , 26 , 29]; however , other factors such as the requirement for fast depolarization suggest that PAD-induced spiking is probably rare ( see below ) , but this depends on the excitability of central axon terminals , which still remains uncertain . Intriguingly , our data also suggest that PAD continues to mediate presynaptic inhibition under conditions in which it can induce transient spiking . Although seemingly paradoxical , the co-existence of excitatory and inhibitory effects has been observed previously in studies of presynaptic inhibition in crayfish [54] and is consistent with the biophysical mechanisms responsible for each effect . This is unlike what happens postsynaptically in spinal neurons , where paradoxical excitation develops only after inhibition fails [10 , 55] . Our data argue that increased PAD has a net inhibitory effect , meaning paradoxical excitation via enhanced PAD poses less risk to somatosensory processing than disinhibition caused by reduced PAD . The GABA conductance density required for PAD-induced spiking under normal conditions is evidently quite high , so much so that we were able to elicit spiking in only 2 of 29 neurons despite testing with virtual ḡGABA several times greater than the typical density measured in somata [13 , 35] . This is consistent with previous failures to elicit spikes by puffing GABA on the soma [13 , 35 , 42] . Puffed GABA also failed to elicit calcium signals when applied to the central terminals of GCaMP-expressing primary afferents [13] , and Verdier et al . [56] observed GABA-induced spiking in only 4 of 77 neurons tested in the trigeminal nucleus . The value of ḡGABA in central axon terminals remains an open question but evidence points to reduced expression of presynaptic GABAARs following nerve injury [13 , 57 , 58] , which suggests that presynaptic inhibition is weakened by reduction of ḡGABA rather than ḡGABA becoming strong enough that PAD induces spiking . That said , the minimum ḡGABA needed for PAD-induced spiking is reduced by increased neuronal excitability ( Fig 2C ) and by a depolarizing shift in EGABA ( Fig 2D ) . Unlike an increase in ḡGABA , which increases inhibitory effects due to shunting , increased neuronal excitability and depolarized EGABA can encourage PAD-induced spiking without enhancing PAD-mediated shunting . Studying transient spiking cells in the chick cochlear nucleus , Monsivais and Rubel [59] found that depolarizing GABAAR input could elicit spiking after blockade of the low-threshold potassium current known to be responsible for transient spiking [60] . The same GABAAR input normally inhibited stimulus-evoked spiking by activating the low-threshold potassium current and thereby elevating spike threshold [59] . Those data are entirely consistent with results presented here . Notably , PAD-induced spiking would be more likely in central axon terminals if those terminals are more excitable that we have assumed based on extrapolation from somatic data . Intracellular chloride could be depleted during PAD if chloride uptake via NKCC1 became saturated ( at least transiently ) and thus failed to keep pace with chloride efflux via activated GABAA channels . The potential for altered chloride concentration is exacerbated by the small caliber of central axon terminals , especially C fibers , since intracellular volume is small compared to surface area [61] . Chloride depletion , if it occurred , would cause an activity-dependent hyperpolarizing shift in EGABA , the implication being that EGABA may be near -20 mV only at the onset of GABAAR activation . Given that PAD-induced spiking depends on a depolarized EGABA value , a hyperpolarizing shift would discourage PAD-induced repetitive spiking . That said , the transient spiking observed in our dynamic clamp experiments was not due to chloride depletion since the virtual GABA current is mediated by current injection through the patch pipette rather than by chloride efflux across the cell membrane . In effect , PAD-induced repetitive spiking may be more difficult to evoke under natural conditions , and transient spiking may rely even more heavily on abrupt depolarization than our experiments suggest . Following on the above points , both simulations and experiments demonstrated that smaller pathological changes in EGABA and/or excitability are required to enable PAD-induced transient spiking than are required for PAD-induced repetitive spiking . This has important implications . Even if sustained , PAD is likely to produce only one spike at its onset ( if it produces any spikes at all ) and will likely not produce any spikes unless its onset is abrupt . This is because transient spiking involves a spike initiation mechanism that is sensitive to the rate of depolarization [40] . Sensitivity to gGABA onset kinetics would be inconsequential if presynaptic inhibition was phasic , which is to say that the GABAARs are clustered within the synaptic cleft and therefore receive an abrupt pulse of GABA upon its vesicular release [62] , but evidence points toward a more tonic mode of action ( unlike the phasic inhibition studied in the crayfish neuromuscular junction [63] ) as outlined below . Recording from mammalian primary afferent terminals to measure the activation kinetics ( and density ) of the GABAAR current is prohibitively difficult , but immunocytochemical evidence argues that C fiber terminals are devoid of gephyrin [64] . Since gephyrin is usually necessary for GABAAR clustering [65] , its absence suggests that GABAARs are distributed more diffusely . Electrophysiological evidence for high-affinity GABAARs in primary afferent neurons [37] supports this view since such receptors have a δ subunit [66] in place of the γ subunit that is necessary for clustering [62 , 67] . If primary afferent GABAARs are indeed distributed extrasynaptically , and are thus activated asynchronously as GABA diffuses beyond the synaptic cleft , then gGABA will have slow onset kinetics and is unlikely to elicit transient spiking . Only the most hyperexcitable fibers ( i . e . those capable of PAD-induced repetitive spiking ) are likely to exhibit any PAD-induced spiking . And whereas PAD-induced transient spiking relies on abrupt GABAAR activation , PAD-mediated inhibition does not; instead , PAD-mediated inhibitory effects will last throughout the duration of the PAD . In other words , slow activation of extrasynaptic GABAARs–arguably the most likely scenario at least for C fiber terminals ( see above ) –will not cause PAD-induced spiking but will cause PAD-mediated inhibition . Notably , dorsal root reflexes ( DRRs ) have typically been studied using electrical stimulation of a nerve or dorsal root to synchronously activate a large number of afferent fibers [e . g . 68] . Notwithstanding differential conduction latencies , such input will evoke a large burst of GABA release , causing GABAAR activation that is ideally suited for PAD-induced transient spiking . It is not obvious that those same fibers would exhibit PAD-induced spiking under more natural conditions ( i . e . less synchronous inputs ) . However , Dubuc et al . [69] observed antidromic spiking in 19% of cat dorsal root fibers during fictive locomotion . It has long been recognized that dorsal root reflexes are more common in certain afferents ( e . g . stretch receptors ) with direct evidence for DRRs being weakest in C fibres [22] . However , Lin et al . [36] reported spontaneous and von Frey-evoked antidromic spiking in all fiber types and , moreover , found that intradermal capsaicin selectively increased antidromic spiking in C and Aδ fibers . Based on more recent observations , including data presented here , one may suspect that chloride regulation , GABA receptor clustering and/or intrinsic excitability differ between afferent types . Somatic recordings suggest that important differences do indeed exist [70] but definitively resolving this requires comparison of axon terminals ( rather than somata ) and is therefore technically difficult . Notably , Dubuc et al . [69] observed repetitive antidromic spiking , as have others [e . g . 71] , which argues that the excitability of certain afferent terminals is quite high . The role of axonal excitability warrants closer attention in future studies . Observation that cooling increases DRRs [72] likely holds important clues . Please see [5] for a recent review of other factors . As already explained , PAD-induced spiking does not equate with failure of presynaptic inhibition . The resilience of presynaptic inhibition is best appreciated by comparing how pre- and postsynaptic inhibition fail . As KCC2 is downregulated postsynaptically , EGABA undergoes a depolarizing shift that directly compromises the inhibitory effect of GABAergic input [61] . The same shift in EGABA that reduces postsynaptic inhibition is what eventually results in paradoxical excitation . This shift from inhibition to paradoxical excitation is evidently not what happens presynaptically . In primary afferent terminals , the changes required for paradoxical excitation–a shift in EGABA and increased excitability–do not undermine the inhibitory effect; in fact , the relatively high ḡGABA required for PAD-induced spiking also encourages PAD-mediated inhibition . This conclusion contradicts past assumptions on this matter . Furthermore , whereas the risk of paradoxical excitation increases postsynaptically during sustained GABAergic input ( because of chloride accumulation ) , presynaptically , the balance shifts towards inhibitory effects over time as sodium channel inactivation accumulates and if intracellular chloride is depleted . The greatest risk to presynaptic inhibition is reduced PAD rather than enhanced PAD . To conclude , we have demonstrated that combined changes in EGABA and intrinsic excitability enable PAD-induced transient spiking . However , unless neurons become so hyperexcitable that PAD can induce repetitive spiking , slow ( asynchronous ) activation of extrasynaptic GABAARs is unlikely to elicit any spiking . On the other hand , PAD will continue to mediate presynaptic inhibition . In practical terms , our results suggest that presynaptic inhibition is a viable therapeutic target whose enhancement carries little risk of causing paradoxical excitation .
All experiments were approved by the University of Pittsburgh IACUC and by The Hospital for Sick Children Animal Care Committee . Starting from a previously published model [45 , 73] , our single compartment , conductance-based model is described as follows: CdVdt=Istim−g¯Nam∞ ( V ) ( V−ENa ) −g¯Kw ( V−EK ) −gleak ( V−Eleak ) −gGABA ( t ) ( V−EGABA ) ( 1 ) where activation variable m changes instantaneously with voltage V according to m∞ ( V ) =0 . 5[1+tanh ( V−βmγm ) ] , ( 2 ) whereas w changes more slowly according to dwdt=ϕww∞ ( V ) −wτw ( V ) , ( 3 ) w∞ ( V ) =0 . 5[1+tanh ( V−βwγw ) ] , ( 4 ) τw ( V ) =1cosh ( V−βw2γw ) . ( 5 ) Neuronal excitability was varied by changing parameter βw [38] . Injury-induced hyperexcitability can be reproduced by shifting βw from its normal value of around -20 mV to less negative values [73] . Setting βw to less negative values reflects a multitude of potential injury-induced molecular changes including reduced KV1-type potassium current , which we model experimentally using 4-AP application , and increased sodium current , which we model experimentally using dynamic clamp ( see below ) ; the effect of such changes , occurring alone or together , is to alter spike initiation [45] . All other neuronal parameters were fixed as reported previously [38] at the following values: C = 2 μF/cm2; sodium conductance ḡNa = 20 mS/cm2 , ENa = 50 mV , βm = -1 . 2 mV , γm = 18 mV; potassium conductance ḡK = 20 mS/cm2 , EK = -100 mV , ϕw = 0 . 15 , γw = 10 mV; leak conductance gleak = 2 mS/cm2 , Eleak = -70 mV . Stimulating current Istim was not applied unless indicated . Maximal GABA conductance density ḡGABA and reversal potential EGABA were varied . Units for ḡGABA were converted to nS/pF for comparison with experimental measurements . The normal EGABA value in primary afferent is around -35 mV based on measurements using different techniques [12 , 13 , 35 , 42] . GABA conductance was activated as a step or as a synaptic waveform described by gGABA ( t ) =g¯GABAx[−e−tτrise+e−tτdecay] , ( 6 ) which comprises an exponential rise to maximum ( with time constant τrise ) followed by an exponential decay back to baseline ( with τdecay ) . The peak is normalized to 1 by factor x before being scaled by ḡGABA . Kinetics are reported in the Results section . For simulations reported in Figs 4 and 7 , sodium channel inactivation h was applied to a proportion of sodium channels defined by p , thus giving the following current balance equation CdVdt=Istim−pg¯Nam∞ ( V ) h ( V−ENa ) − ( 1−p ) g¯Nam∞ ( V ) ( V−ENa ) −g¯Kw ( V−EK ) −gleak ( V−Eleak ) −gGABA ( t ) ( V−EGABA ) . ( 7 ) Changes in h are described by the same equations used to describe w ( Eq 3–5 ) where βh = -28 mV , γh = -14 mV , and ϕh = 0 . 005 . All simulations in single compartment models were conducted in XPP . Bifurcation analysis was conducted using AUTO via the XPP interface . The multicompartment model was built in NEURON . Ion channels were modeled as above except that both ḡNa and ḡK were increased to 30 mS/cm2 . Additional parameters were as follows: axial resistivity Ra = 150 Ω·cm , diameter = 1 μm , compartment length = 1 mm , d_lambda = 0 . 01 . GABA conductance ḡGABA was modeled as a uniform density throughout the middle compartment . All experiments were carried out on adult ( 200–340 g ) male Sprague-Dawley rats ( Harlan , Indianapolis , IN and Charles River , Montreal , Quebec ) . A subset of animals received spinal nerve ligation ( SNL ) 2–5 days before terminal experiments [74] . Under isoflurane anesthesia , the paraspinal muscles were separated to access the L6 process , which was carefully removed . The L5 spinal nerve was tightly ligated with 6–0 silk suture . All nerve-injured animals maintained motor function but developed neuropathic pain as inferred by guarding of the affected paw . To collect DRG neurons , rats were deeply anesthetized by subcutaneous injection of anesthetic cocktail ( 1 ml/kg of 55 mg/ml ketamine , 5 . 5 mg/ml xylazine , and 1 . 1 mg/ml acepromazine ) or by isoflurane ( 4% for induction; 2 . 5% for maintenance ) . DRG ( L4 and L5 in naïve animals; L5 in nerve-injured animals ) were surgically removed to chilled MEM-FBS culture media and desheathed . DRG were then enzymatically treated for 45 minutes in culture media composed of 89% MEM , 370 units/ml penicillin and 370 μg/ml streptomycin , 1% MEM vitamin solution ( all from Life Technologies ) , and 1 . 2 mg/ml collagenase Type 4 ( Worthington Biochemical Corp ) . DRG were mechanically dissociated by trituration with a fire-polished Pasteur pipette , and further enzymatically treated for 5 minutes in Ca2+- and Mg2+-free Hanks’ balanced salt solution ( HBSS; Life Technologies Inc ) , containing 2 . 5 mg/ml trypsin ( Worthington Biochemical Corp ) and 0 . 02% sterile ethylenediaminetetraacetic acid ( EDTA; Sigma-Aldrich Canada Ltd ) . Trypsin activity was subsequently inhibited by the addition of MEM-FBS supplemented with 0 . 625 mg/ml MgSO4 ( Caledon Labs ) . Dissociated cells in MEM-FBS were plated on glass coverslips previously coated by a solution of 0 . 1 mg/ml poly-D-lysine , and incubated in MEM-FBS at 37°C , 5% CO2 , and 90% humidity for 2 h . Coverslips were then transferred to a HEPES-buffered Leibovitz’s L-15 media containing glutamine ( Life Technologies Ltd ) , 10% FBS , 100 units/ml of penicillin and 100 μg/ml streptomycin , and 5 mM D-glucose ( Caledon Labs ) and stored at room temperature until used for experiments for 2–28 hours later . Spiking properties do not change appreciably over this period and nor do neurites develop based on storage at room temperature , omission of laminin from coverslips , and the growth factor-free culture medium used . Coverslips with cultured cells were transferred to a recording chamber constantly perfused with room temperature , oxygenated ( 95% O2 and 5% CO2 ) artificial cerebral spinal fluid containing ( in mM ) 126 NaCl , 2 . 5 KCl , 2 CaCl2 , 2 MgCl2 , 10 D-glucose , 26 NaHCO3 , 1 . 25 NaH2PO4 . Cells were recorded in the whole-cell configuration with >70% series resistance compensation using an Axopatch 200B amplifier ( Molecular Devices; Palo Alto , CA ) . Electrodes ( 2–5 MΩ ) were filled with a recording solution containing ( in mM ) 125 KMeSO4 , 5 KCl , 10 HEPES , 2 MgCl2 , 4 ATP , 0 . 4 GTP as well as 0 . 1% Lucifer Yellow; pH was adjusted to 7 . 2 with KOH and osmolality was between 270 and 290 mosmol/L . For experiments on the contribution of ANO-1 channels , KMeSO4 was reduced to 67 mM and KCl was increased to 63 mM to give ECl = -20 mV . Data were low-pass filtered at 2 KHz , digitized at 20 KHz using a CED 1401 computer interface ( Cambridge Electronic Design , Cambridge , UK ) , and analyzed offline . Virtual GABA conductance was applied via dynamic clamp using Signal 5 software ( CED ) . The virtual conductance was modeled as a step or as a synaptic waveform described by Eqn . 6 . To express the virtual conductance as a density and thus exclude direct effects of cell size , we normalized absolute conductance values by membrane capacitance C because C is proportional to the surface area of the cell . Capacitance was measured for each cell based on responses to small ( 50 pA ) hyperpolarizing current steps , where C = τmembrane / Rin . To increase cellular excitability in neurons from naïve animals , potassium channels were blocked with bath applied 4-aminopyridine ( 4-AP ) . In a subset of experiments with 4-AP , a virtual voltage-dependent sodium conductance was also inserted via dynamic clamp using the equations and parameters reported by Ratté et al . [45] . Neurons from nerve-injured animals are already hyperexcitable and were not , therefore , subject to manipulations ( i . e . 4-AP or virtual sodium conductance ) intended to increase excitability . All data and computer code are available from the corresponding author upon request . | Postsynaptic GABAAR mediate inhibition by causing hyperpolarization or by preventing ( shunting ) the depolarization caused by concurrent excitatory input . Presynaptic GABAAR work differently , in the spinal cord at least . Because of their higher-than-equilibrium intracellular chloride concentration , the central terminals of primary afferent fibers are depolarized by activation of GABAAR . This so-called primary afferent depolarization , or PAD , nonetheless reduces spike propagation and synaptic release from those fibers because of shunting effects and sodium channel inactivation . But those inhibitory effects can be diminished under certain pathological conditions; in fact , the emergence of dorsal root reflexes suggests that PAD can become paradoxically excitatory . The biophysical basis for this paradoxical excitation has been hinted at by experiments , but here , for the first time , we use computational modeling and dynamic clamp experiments to decipher how distinct contributing factors interact to enable PAD-induced spiking . Our results suggest that PAD-induced spiking requires a shift in GABA reversal potential plus changes in intrinsic excitability that allow for repetitive spiking during sustained depolarization . Inhibitory effects of PAD are retained under conditions in which GABAAR activation causes transient spiking and are only lost if GABAAR activation can evoke repetitive spiking . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"action",
"potentials",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"chemical",
"compounds",
"depolarization",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"ion",
"channels",
"nerve",
"fibers",
"neurotransmitters",
"sodium",
"channels",
"animal",
"cells",
"axons",
"proteins",
"chemistry",
"chlorides",
"biophysics",
"gamma-aminobutyric",
"acid",
"physics",
"biochemistry",
"cellular",
"neuroscience",
"cell",
"biology",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"neurophysiology"
]
| 2016 | Combined Changes in Chloride Regulation and Neuronal Excitability Enable Primary Afferent Depolarization to Elicit Spiking without Compromising its Inhibitory Effects |
Giardia trophozoites attach to the intestinal microvilli ( or inert surfaces ) using an undefined “suction-based” mechanism , and remain attached during cell division to avoid peristalsis . Flagellar motility is a key factor in Giardia's pathogenesis and colonization of the host small intestine . Specifically , the beating of the ventral flagella , one of four pairs of motile flagella , has been proposed to generate a hydrodynamic force that results in suction-based attachment via the adjacent ventral disc . We aimed to test this prevailing “hydrodynamic model” of attachment mediated by flagellar motility . We defined four distinct stages of attachment by assessing surface contacts of the trophozoite with the substrate during attachment using TIRF microscopy ( TIRFM ) . The lateral crest of the ventral disc forms a continuous perimeter seal with the substrate , a cytological indication that trophozoites are fully attached . Using trophozoites with two types of molecularly engineered defects in flagellar beating , we determined that neither ventral flagellar beating , nor any flagellar beating , is necessary for the maintenance of attachment . Following a morpholino-based knockdown of PF16 , a central pair protein , both the beating and morphology of flagella were defective , but trophozoites could still initiate proper surface contacts as seen using TIRFM and could maintain attachment in several biophysical assays . Trophozoites with impaired motility were able to attach as well as motile cells . We also generated a strain with defects in the ventral flagellar waveform by overexpressing a dominant negative form of alpha2-annexin::GFP ( D122A , D275A ) . This dominant negative alpha2-annexin strain could initiate attachment and had only a slight decrease in the ability to withstand normal and shear forces . The time needed for attachment did increase in trophozoites with overall defective flagellar beating , however . Thus while not directly required for attachment , flagellar motility is important for positioning and orienting trophozoites prior to attachment . Drugs affecting flagellar motility may result in lower levels of attachment by indirectly limiting the number of parasites that can position the ventral disc properly against a surface and against peristaltic flow .
Giardiasis is caused by acute or chronic infection with the single-celled , zoonotic parasite Giardia intestinalis [1] . Giardiasis is one of the most prevalent intestinal protozoal parasitic infections worldwide [2] , resulting in several hundred million acute cases of malabsorptive diarrhea each year . The parasite persists in the environment as a dormant , infectious cyst [3] , [4] . Infection of humans or other mammals is initiated by the ingestion of cysts from contaminated water or food [5] . Following ingestion , giardial cysts travel to the small intestine of the animal host , excyst and transform into the flagellated trophozoite . To avoid peristalsis and colonize the small intestine , trophozoites attach to the intestinal villi via a specialized microtubule structure , the ventral disc . The mechanism of attachment has been proposed to involve suction generated either by the ventral disc itself or by the regular beating of the ventral flagella [6] , [7] . Both the molecular mechanism of attachment and the precise role of flagellar motility in attachment remain controversial . Trophozoites are bilaterally symmetrical with a flattened teardrop shape ( ∼15 µm long by 5 µm wide and 5 µm thick ) and possess a complex microtubule cytoskeleton that includes eight flagella [8] . Giardia's flagella generate complex movements essential for motility , cell division , and access to suitable sites for attachment on the intestinal villi [9] , [10] . The eight flagella are organized as four pairs: the anterior , the caudal , the posteriolateral and the ventral flagella ( Figure 1A ) . Giardia axonemes possess long cytoplasmic regions that exit the cell body as membrane-bound flagella . All eight flagella have a canonical motile structure consisting of nine outer doublet microtubules surrounding the central microtubule pair , radial spokes and dynein arms [11] . While the role of flagellar motility in attachment remains speculative [9] , the coordinated and differential beating of Giardia's eight motile flagella are known to be critical to cellular motility and division , and are possibly involved in encystation/excystation or chemotactic sensing [12] . The most widely held model of giardial attachment , the “hydrodynamic model” [7] , [13] , contends that flagellar motility is necessary for the initiation and maintenance of giardial attachment to surfaces . Specifically , the ventral flagella were proposed to produce a hydrodynamic current generating a suction pressure under the adjacent ventral disc . The model postulates that surrounding fluid is drawn through presumptive channels that initiate at the ventrolateral flange , flows under the marginal groove and lateral crest at the perimeter of the disc , and eventually exits at a channel in the posterior lip of the disc into the ventral groove , where the ventral flagella were thought to exit from the cell body ( see Figure 1B ) . Cytological evidence has not corroborated the existence of these channels; thus , support for the “hydrodynamic model” has remained strictly observational or theoretical [7] , [13] . Prior investigations have not distinguished between ventral flagellar beating that causes attachment and flagellar beating that merely coincides with attachment . We were interested in attachment mechanics and the precise contribution of flagellar beating to attachment , either directly via hydrodynamic suction [7] , [13] , or indirectly via cellular positioning prior to attachment . We examined the role of ventral flagellar beating during the early stages ( positioning ) and later stages ( maintenance ) of attachment in live trophozoites . Using Total Internal Reflection Microscopy ( TIRFM ) of trophozoites labeled with a fluorescent plasma membrane dye , we defined distinct stages of attachment based on cellular and ventral disc contacts with the substrate surface ( Figure 1 ) . To test whether flagellar motility is required for giardial attachment , we used a morpholino-based knockdown [14] of the axonemal central pair protein PF16 to generate a strain with defects in flagellar beating . Knockdown of giardial PF16 resulted in various defects in all flagella , including defects in the rate of flagellar beat and/or flagellar length ( Figure 2 ) . Secondly , we constructed a strain with defects specific to the ventral flagellar waveform by overexpressing a dominant negative [15] , [16] ventral flagella-specific alpha2-annexin ( Figure 3 ) . By assessing attachment in both types of trophozoites with defective flagellar motility , we demonstrate that defects in flagellar beating and coordination do not significantly affect attachment , with respect to disc contacts with the substrate surface or the ability to withstand normal forces and shear forces ( Figure 4 ) . Deficiencies in flagellar motility do , however , result in slower attachment during earlier stages when motility is required for positioning the ventral disc against the substrate surface ( Figure 5 ) . Thus , we propose that flagella contribute indirectly to attachment by positioning the cell , but ventral flagellar beating , specifically , is not directly involved in generating suction forces underneath the ventral disc .
To investigate the direct or indirect contributions of flagellar beating to attachment , we defined the general stages of attachment and detachment based on cell body and ventral disc surface contacts using TIRFM of trophozoites stained with a plasma membrane dye ( Figure 1 , panels C-F ) . TIRFM uses an evanescent wave that penetrates only 100 nm into the sample medium , enabling selective visualization of surface regions of cells . Trophozoites first skim and contact the surface with the anterior section of the ventrolateral flange ( Figure 1 ) . Secondly , the perimeter of the ventral disc touches the surface , forming a continuous contact , or “seal” at the area of the lateral crest . The lateral shield then presses against the substrate , followed by the bare area region within the ventral disc . We noted bare area contact in 76% of attached cells ( n = 97 cells ) . During maintained attachment , we observed a continual surface contact via the lateral crest seal , around the entire ventral disc . Giardial attachment to biological or inert surfaces is reversible and occurs within seconds [17] , [18] , [19]; here attachment to a glass substrate occurred in less than one to several seconds . In prior work using transmitted light contrast techniques this seal was postulated to be a channel [6] , [7] . The lateral crest seal is the first indicator that the hydrodynamic model as described is invalid [6] , [7] . Detachment begins with release of the bare area from the surface , followed by release of the lateral shield ( Figure 1 ) . The disc seal becomes discontinuous , specifically at the posterior lip , and finally the cell detaches to swim in the medium . When the cell skims ( Video S1 ) the disc seal and ventrolateral flange remain in close contact with the substrate while the bare area region lifts . Bare area contact reappears when the cell pauses or reattaches . Ventral flagellar beating is close to the surface and thus readily imaged in TIRFM . The beating of other flagella can only be observed when those flagella are motile and come within 100 nm of the substrate . In contrast to previous reports [7] , we did not observe an arched groove between the disc perimeter and the ventrolateral flange ( Figure 1 ) . The ventrolateral flange does not have an arched profile and remains flat against the substrate , as does the lateral shield . Most notably , we did not observe a “Y”-shaped ventral channel present between the posterior lip of the ventral disc that continues into a ventral-caudal stem , postulated to conduct a hydrodynamic current [6] , [7] . In contrast , we observed a complete and continuous disc seal ( Figure 1 ) . The surface contacts observed using TIRFM are consistent with ultrastructural SEM and TEM images of attached cells as well as interference-reflection microscopy [20] . We measured a wild type ventral flagellar mean beat frequency of 9 Hz with a synchronous waveform beat from base to tip along the longitudinal axis of the cell , similar to previous reports [10] . In contrast to prior work that used trophozoites scraped from mouse intestine [6] , [7] , we determined the mean amplitude of the ventral flagellar waveform to be 2 . 04 µm . Synchronous ventral flagellar beating was observed once the trophozoite made a seal with the lateral crest ( Figure 1 and Video S1 ) . Notably , a change in frequency and amplitude during skimming and swimming correlated with changes in the directional motility of the cell , as previously reported [6] , [7] . PF16 was first characterized in the green alga Chlamydomonas reinhardtii as a highly conserved armadillo-repeat protein localizing to the C1 microtubule of motile flagella [21] , [22] , and required for proper flagellar waveforms and motility [23] , [24] , [25] . We expected the Giardia PF16 homolog to have a similar function in generating proper waveforms and motility in the eight giardial flagella . To test whether ventral flagellar beating is necessary to create hydrodynamic flow for attachment , we used a PF16-specific morpholino ( see Methods and [14] ) to block translation of the protein . The anti-PF16 morpholino knockdown was confirmed via Western blot and immunostaining ( see Figure S1 ) using an integrated PF16::HA strain [26] . As with other flagellates [21] , [22] , [27] , the knockdown of PF16 resulted in significant flagellar motility defects . Twenty-four hours after morpholino electroporation , the anti-PF16 treated trophozoites exhibited erratic behavior in all flagella . Both the wild type and the PF16 mispair control had ventral flagella of similar lengths ( 14 . 4 µm for the membrane-bound portion , measured from the flagellar exit point at cell body to the flagellar tip ) , a sigmoidal ventral flagellar beat pattern , and similar beat frequencies ( 9 Hz ) and amplitudes ( 2 . 04 µm ) ( Figure 2 and Video S2 ) . The anti-PF16 morpholino-treated cells sustained amplitudes similar to wild type but displayed an erratic flagellar beat . Twenty-four hours after anti-PF16 morpholino introduction , 71% of cells exhibited a significantly decreased ventral flagellar beat ( mean = 4 Hz ) with 200 millisecond pauses ( Figure 2 ) . Flagellar motility is also thought to be required for later stages of cytokinesis in Giardia [28] . Forty-eight hours after anti-PF16 knockdown , many cells in the population lagged in their ability to complete cytokinesis so that many daughter cells remained connected via their posterior cell bodies . Total paralysis was not observed , but ventral flagellar beating slowed to 2 . 3 Hz , a significant reduction from 9 Hz found in wild type ( n = 50 ) . We also observed that the membrane-bound regions of the ventral and caudal flagella were one quarter and one third , shorter , respectively ( Figure 2 , Video S2 ) . A significant number of cells also exhibited dorsal flexion paralysis which could result in detachment , thus this time point was not included in the cell attachment assays . Using TIRFM with live imaging , we investigated the ability of the anti-PF16 morpholino transformant with flagellar beating defects to form proper surface contacts on glass coverslips ( see Figure 2 ) . Despite the significant defects in flagellar beat rate ( Figure 2 and Video S2 ) , trophozoite surface contacts in anti-PF16 morpholino-treated cells were similar to wild type ( Figure 1 ) . We observed no defects in the lateral body contacts , bare area contacts or the continuity of the disc perimeter seal . To assay the ability of the PF16-knockdown trophozoites with defective flagellar beating to maintain attachment , we next challenged live morpholino-treated trophozoites with two biophysical assays . First , using a centrifuge assay of increasing normal forces , we noted that the anti-PF16 knockdown population maintains attachment against normal centrifugal forces up to 2 . 1 nN ( Figure 4 ) similar to wild type trophozoites [21] . To assay the ability to withstand shear forces , we used a flow cell assay [20] . The anti-PF16 morpholino knockdown trophozoites were able to maintain attachment when challenged with 1 . 5 nN of shear force , equal to the mispair morpholino control ( Figure 4 ) . The prediction of the hydrodynamic model is that trophozoites would detach once the flagellar beat decreased , and thus would be incapable of maintaining steady state attachment when challenged with force . Despite the fact that the ventral flagella beat erratically and were noticeably shortened in length ( see above ) , the PF16 knockdown trophozoites retained the ability to initiate and maintain attachment comparable to wild type trophozoites . Annexins are membrane-scaffold proteins that generally link the cytoskeleton to the periphery of negatively charged , acidic phospholipid membranes in a Ca+2-regulated manner [29] . Several annexin homologs have been shown to localize specifically to various pairs of flagella [30] . Alpha2-annexin was previously shown to localize to the ventral flagella [30]; thus , dominant negative annexins could specifically inactivate the waveform of the ventral flagella . We confirmed the localization of alpha2-annexin to the ventral flagella ( Figure 3 ) in both live and fixed cells using a GFP tag [30] . Alpha2-annexin::GFP localizes to 87% of the cell population , strongly to the ventral flagella ( signal intensity mean = 1650 ) , and to a lesser degree , the plasma membrane of the ventral disc ( signal intensity mean = 700 ) and the cell . We observed that the ventral flagellar waveform , synchrony , beat rate and frequency in the alpha2-annexin::GFP strain equaled that of the WBC6 wild type strain . We measured a negligible decrease in amplitude at 1 . 71 µm , as compared to 2 . 04 µm in wild type , yet we observed no growth or attachment defects in the alpha2-annexin::GFP strain . To test the particular role of ventral flagellar beating in attachment , we created a strain with defects in the ventral flagellar waveform caused by overexpression of a tetracycline-inducible dominant-negative alpha-2 annexin . Specifically , we modified amino acid residues in two of four high-affinity calcium-binding domains in the giardial alpha-2 annexin from asparagine to alanine ( D175A , D275A ) , which has previously been shown to generate dominant negative annexins [31] ( Figure S1 ) . We observed and quantified significant defects in 82% of the alpha2-annexin ( D175A , D275A ) ::GFP population , specifically in the amplitude of the ventral flagella waveform ( as compared to wild type ( reviewed recently in [9] ) at both 24 and 48 hours after induction of the alpha2-annexin dominant negative construct . Specifically , the amplitude of the ventral flagellar waveform was significantly decreased from 2 . 04 µm in wild type to 0 . 85 µm in the dominant negative strain . A C-terminal GFP tag allowed visualization of the dominant negative alpha2-annexin::GFP , which localized to the ventral flagella plasma membrane ( Figure 3 ) and somewhat to the ventral disc , albeit with a weaker signal than the alpha2-annexin::GFP strain . Two of the four active calcium-dependent binding sites were mutated , leaving only two with the ability to bind the membrane , likely resulting in lower signal intensity . Because overexpression of the dominant negative alpha2-annexin resulted in defects of the ventral flagellar waveform , we assessed the ability of this strain to attach using both TIRFM and live biophysical assays of normal [19] and shear forces . Because prolonged exposure to fluorescence microscopy can induce changes in flagellar beating , we limited our observations to less than 30 minutes in temperature-controlled , closed environments . Flagellar beat measurements were captured with TIRFM at very low ( 10 ms ) exposures for two seconds , and then confirmed with phase contrast microscopy . Trophozoites overexpressing the dominant negative alpha2-annexin could still form a seal at the ventral disc perimeter ( Figure 3 ) , and could resist increased normal and shear forces despite the observed defects in ventral flagellar waveform ( Figure 4 ) . We did observe a decrease in the ability of the dominant negative alpha2-annexin strain to withstand normal forces in the centrifuge assay as compared to wild type that could be attributed to an increased rigidity in the plasma membrane of the ventral disc as well as the ventral flagella . Rather than being directly involved in generating a hydrodynamic current , flagellar motility could be important for positioning trophozoites so that the ventral disc is oriented parallel to the substrate . Defects in cellular positioning would not necessarily affect the overall number of cells attached but could slow the initial rate at which trophozoites attach to surfaces . Using live imaging we observed that the anti-PF16 morpholino-treated trophozoites often settled near the substrate , yet were oriented incorrectly with the disc side facing away from the surface . We then used time-lapse imaging of live trophozoites attaching to the bottom of a 96-well plastic cell culture plate and quantified the number of cells able to attach at specific intervals over a thirty-minute period ( Figure 5 ) . As compared to wild type and mispair morpholino controls , the anti-PF16 morpholino cells took longer to attach to the substrate ( Figure 5 ) and had a decreased skimming motility as compared to wild type . Moreover , the rate of attachment was significantly decreased at each time point over a thirty-minute period in the trophozoites with defective flagellar beating .
Trophozoites attach to both biological substrates such as the intestinal microvilli ( in vivo attachment ) and inert substrates such as plastic or glass ( in vitro attachment ) ; however , the precise contacts of the ventral disc or the trophozoite cell body with the surface have not been defined . Giardial attachment has been broadly defined as the number of cells that remain adhered to a given surface after an experimental treatment [33] , [36] , [41] , [46] , [47] , [48] . Based on these definitions , attachment has been quantified using three types of experimental approaches: 1 ) direct/indirect counts of attached and unattached cells [33] , [36] , [41] , [46] , [47] , [48]; 2 ) live imaging [7] , [17] , [18] , [38] , [49] , [50]; and more recently 3 ) a novel centrifuge assay of normal attachment force [19] . Most attachment assays have counted the number of adherent trophozoites at the population level after long incubation periods ( ∼2-24 hours ) , as opposed to quantifying the attachment dynamics of individual trophozoites under physiological conditions comparable to the host . Attachment generally has not been correlated with cell viability despite the common understanding that Giardia detaches when dividing [28] , [45] , when non-viable , or when exposed to oxygen or low temperature [46] . Using TIRFM to capture attaching trophozoites ( Figure 1 ) , we demonstrate that a binary ( on/off ) conception of attachment is misleading and overly simplistic . Giardial attachment occurs as a stepwise process proceeding in degrees of cellular contact with the surface ( Figure 1 ) . Four stages of attachment include skimming , disc seal formation ( via the lateral crest ) , lateral shield contact and bare area contact ( Figure 1 ) . In each stage the disc remains concave , with only the disc edges and later the bare area contacting the surface . While the timing of these stages can vary from less than one second to several seconds , the stages of surface contacts during attachment and detachment ( Figure 1 ) always occur in this stepwise fashion . Quantifying cell surface contacts also permits the assessment of attachment defects resulting from drugs or potential molecular genetic disruptions of the attachment mechanism . During detachment , the disruption of surface contacts of the cell body and ventral disc occur in reverse order to the stages of attachment ( Figure 1 ) . Movements of the caudal pair of flagella are thought to generate the flexing of the posterior trophozoite “tail” region , indirectly resulting in detachment [10] . Our TIRFM analysis indicates that the tail region does not flex toward the surface prior to detachment and thus do not support this notion . Nonetheless , whether lateral tail flexion or dorsal tail abduction causes detachment still needs to be directly tested . Our analysis of surface contacts also has specific implications for giardial attachment models ( summarized recently in [8] ) . Soloviev and Holberton [7] proposed that a hydrodynamic force generated by ventral flagellar beating created a negative pressure differential under the adjacent disc to cause suction ( also see Figure 1 ) . The ventral flagella would theoretically create a fluid flow transmitted through a ventral disc channel toward a posterior disc cavity ( Figure 1B ) . Thus , hydrodynamic-based suction would be contingent upon the presence of an “arched profile” of the ventrolateral flange , a ventrolateral channel around the perimeter of the disc and a hypothetical “disc portal at the posterior rim of the disc” [51] . Trophozoite surface contacts using TIRFM ( Figure 1C-F ) demonstrate that the disc perimeter forms a continuous seal with the surface . Further , we do not observe either an anterior or a posterior channel when cells are attached . What was previously considered to be a putative channel is , in fact , the lateral crest of the disc pressed against the substrate to form the seal . The anterior portion of the cell , including the ventrolateral flange , may be a flexible region . We used two molecular genetic approaches to generate trophozoites with flagellar motility defects to test further whether the ventral ( or any ) flagellar beating is necessary for giardial attachment . First , we generated general flagellar beating defects by knocking down the giardial homolog of PF16 ( Figure 2 ) , a component of the central pair apparatus of axonemes [27] that localizes to the C1 microtubule of motile ( “9+2” ) flagella . In Chlamydomonas , mutations in pf16 result in paralyzed flagella [22] , and in trypanosomes RNAi of pf16 results in erratic flagellar twitching [21] , [25] . PF16 knockdown can result in axonemal ultrastructural defects , paralyzed flagella , or poorly beating flagella and can ultimately result in axonemes lacking the C1 microtubule [21] . Knockdown of PF16 in Giardia caused a significant decrease in flagellar beat frequency ( Figure 2 ) yet did not cause complete paralysis of flagellar motility . Transient paralysis or pausing did occur every six to eight beat cycles . We also observed shortened ventral and caudal flagella , with one caudal flagellum consistently shorter than the other ( Figure 2 ) . This may be a preliminary indication that one caudal flagellum is older than the other caudal flagellum . Alternatively this observation supports findings in Chlamydomonas that indicate that when a new flagellar axoneme is under construction , length regulation is not limited to the new flagellum , but affects the pair as a whole [52] . To generate ventral flagellar beating defects specifically , we created and overexpressed a dominant negative version of the alpha2-annexin in trophozoites ( Figure 3 and Figure S1 ) . Based on the observed ventral flagellar defects , alpha2-annexin is a presumptive component of the ventral flagellar membrane scaffold . Annexins mediate interactions between the cytoskeleton and the plasma membrane [31] , [53] , and in Giardia , the flagella-specific annexins may regulate the stabilization of flagellar membranes by linking axonemal microtubules to the plasma membrane [54] . We show that parasites are still able to maintain proper surface contacts , despite a ventral flagella waveform of less than half that of wild type due to the overexpression of the dominant negative alpha2-annexin . The fraction of cells able to maintain attachment under normal and shear forces was slightly reduced but because the alpha2-annexin protein localizes to the plasma membrane of the ventral disc as well as the flagella , we surmise that increased ventral disc membrane rigidity may affect the disc attachment dynamics . Eighty-two percent of the trophozoites exhibit the defective flagellar motility phenotype . If ventral flagellar motility were essential for attachment as predicted by the hydrodynamic model [7] , only 18% of the cells would be expected to remain attached due to the lack of penetrance of the phenotype . We observed that over two-thirds of parasites remained attached when compared to the uninduced construct ( Figure 4 ) ; there is not a statistical difference between these percentages . Despite this slight reduction , the formation of proper surface contacts ( as visualized by TIRFM of individual cells with impaired motility ( Figure 3 ) ) supports the argument that ventral flagellar beating is not directly responsible for attachment . Thus , both the PF16 knockdown and the overexpressed dominant negative , ventral flagella-specific annexin strain could initiate and maintain attachment , as measured by the degree of surface contact ( Figures 2 and 3 ) or in live imaging-based biophysical assays ( Figure 4 ) . The ability of attached cells to resist shear and normal forces , despite a decreased waveform amplitude or flagellar beat rate indicates that maintenance of attachment is independent of fluctuations in flagellar motility [7] , and does not support the hydrodynamic model . While ventral flagellar beating is coincident with attachment , ventral flagellar beating neither directly causes nor directly results from the process of attachment . Trophozoites colonize the small intestine after excystation , and flagellar motility is likely required for orientation and positioning the trophozoite against the intestinal villi and for resisting peristaltic currents . Therefore , independent of creating a hydrodynamic current , flagellar motility could have an indirect role in positioning and orienting trophozoites with the ventral disc parallel to in vivo or in vitro surfaces prior to attachment . Overall deficits in flagellar motility should affect both rotational motility ( via the anterior flagella ) and skimming motility ( via the ventral flagella ) . Over time , trophozoites with aberrant motility may settle and attach to surfaces , but the time required for orientation and positioning prior to attachment could be significantly longer . In support of this idea , the anti-PF16 morpholino knockdown , with universal defects in flagellar motility and/or length ( Figure 2 ) , did take significantly longer to attach at each time point over thirty minutes during time-lapse imaging ( Figure 5 ) , and often settled to the substrate with the ventral disc up or remained swimming near the substrate ( Video S4 ) . Anterior flagellar motility is proposed to be responsible for rotational movements [10] , thus anterior flagellar defects resulting from the PF16 knockdown would affect trophozoite orientation . Beating of the ventral flagella has been proposed to generate forward movement [10] , and thus the disruption of ventral flagellar function ( Figure 3 ) contributed to the inability of trophozoites to efficiently skim . Skimming motility allows the trophozoite to remain close to the substrate while searching for a desirable attachment location . Temporal lags in attachment due to flagellar motility defects might even result in more significant decreases in attachment in vivo due to consistent peristaltic flow . Models of giardial attachment are not mutually exclusive , and it is clear that site recognition , flagellar motility , and disc-mediated suction each contribute to in vivo attachment . With respect to site recognition , ligand-specific interactions could be involved in the parasite's selective colonization of the small intestine [32] , [33] , [34] , [35] , [36] in conjunction with flagellar motility . Despite a suggested role for sugars or lectins in mediating specific interactions of Giardia with host cells in vivo ( reviewed in [55] ) , lectin-mediated site recognition is not necessary for attachment in vitro . Once a site is recognized in the host , and flagellar motility positions the trophozoite , attachment may occur directly via a suction-based mechanism; suction is reported to be sufficient for in vitro attachment [19] . In the absence of a hydrodynamic current created by ventral flagellar beating to generate a negative pressure differential or suction underneath the disc , we propose that suction could be generated directly via a conformational change of the ventral disc . In this model , the lateral crest would first initiate the disc “seal” as observed in TIRFM ( Figure 1 ) . Next , a negative pressure differential would occur under the ventral disc via conformational changes of principal disc structures ( MTs , microribbons , crossbridges and/or motor proteins ) . These structures would then relax back to their original conformation , producing a pressure differential between the arched disc and the substrate , consistent with TEM studies [11] . Alternatively , the ventral disc of the trophozoite could undergo a conformation change via protrusion of the bare area ( as seen in the TIRFM , Figure 1 ) . This change in the ventral disc volume would result in decreased fluid pressure due to the displaced fluid . The low pressure under the disc , compared to the surrounding high pressure of the environment would result in a pressure differential that may explain Giardia's mode of suction-based attachment . Flagellar motility prior to attachment is a key factor in Giardia's pathogenesis and colonization of the host small intestine . The work presented here underscores that flagellar motility is important for positioning and orienting trophozoites prior to attachment . The consequence of inhibition of flagellar motility is a decrease in number of attached cells in vivo as is apparent in vitro ( Figure 5 ) . Thus , drugs affecting flagellar motility could indirectly result in lower levels of attachment by limiting the number of cells that can position the ventral disc properly against a surface and against peristaltic flow .
G . intestinalis strain WBC6 ( ATCC 50803 ) trophozoites were maintained in culture at 37°C in modified TYI-S-33 medium with bovine bile [56] in sterile 13 ml screw-capped disposable tubes ( BD Falcon ) and incubated upright without shaking . For imaging , trophozoites were also grown on coverslips placed in 8-well dishes in a sealed chamber ( PlasLabs ) and gassed with 100% N2 to maintain a low oxygen atmosphere . The chamber was incubated at 37°C prior to live cell imaging . Integration of an HA-tagged version of the pf16 gene permitted the assessment of morpholino knockdown in the absence of a specific anti-PF16 antibody . The C-terminal portion of the PF16 gene was cloned in frame to a 3HA-tag and then into a pJET vector containing a neomycin selectable marker as previously described [26] . NruI was used to linearize the vector . Trophozoites were transformed with linearized vector by electroporation as previously described [26] , resulting in C-terminal fusion of a 3HA tag to at least one endogenous copy of the PF16 gene . Transformants were selected with 200 µg/ml neomycin G418 ( Sigma ) . Endogenous integration of the construct was confirmed using PCR primers specific to the N terminus of the pf16 gene and the 3XHA epitope tag . The localization of PF16 to all eight axonemes was verified with immunostaining using a monoclonal anti-HA antibody ( Sigma H9658 ) at a 1∶100 dilution and an Alexa 594 goat anti-mouse IgG secondary antibody ( Invitrogen ) at a 1∶200 dilution . To knock down the giardial axonemal central pair PF16 homolog ( GiardiaDB GL50803_16202 ) , anti-sense morpholino oligonucleotides ( GeneTools ) were designed to the 5′ flanking region and first codons [14] with the following sequence: 5′ TACGACGAAGCGATTAGTTGCCATG 3′ . Anti-PF16 morpholino oligonucleotides ( 100 µM final concentration ) were electroporated into log phase trophozoites as previously described [14] . Morpholino-transformed cells were then incubated for 24 and 48 hours before the phenotype was assessed . To control for off-target effects of the electroporation or of the electroporated morpholino oligonucleotides , both sterile water or a morpholino with five mismatches ( 5′ TATGACAAAGCGGTTAGTAGCCATA 3′ ) were also electroporated . Following electroporation of PF16 specific morpholino oligonucleotides or controls , cell morphology , flagellar beating , and attachment were assessed in both live and fixed trophozoites ( see below ) . To determine the extent of knockdown , Western blotting was used to assay the HA-tagged PF16 levels in crude preparations of wild type G . intestinalis WBC6 or in extract from the integrated HA-tagged PF16 strain . Giardia PF16-3HA protein was detected using a 1∶2000 dilution of anti-HA antibody ( mouse monoclonal , Sigma H9658 ) and an HRP-conjugated secondary antibody ( Bio-Rad ) at 1∶4000 dilution . The blot was also probed with anti-actin antibody at 1∶1500 to verify equal loading . The degree of PF16 knockdown was quantified using the Alpha Innotech Gel imaging and documentation system ( Cell Biosciences ) . By adapting a methodology that has been used to create dominant negative mutant forms of kinesin or dynamin in Giardia [15] , [16] , we constructed dominant negative mutations in the alpha2-annexin gene ( D122A , D275A ) . We expected ventral flagellar defects with overexpression of alpha-2 annexin as it was previously shown to localize to the membrane-bound portions of the ventral flagella [30] . The conserved asparagines ( D ) were changed to alanines ( A ) , as has been done previously for human annexins [29] . Alpha2-annexin :GFP fusions containing dominant negative mutations were placed under the control of a tetracycline-inducible promoter and the mutant protein was overexpressed in Giardia [57] . GFP tagging permitted the identification and characterization of trophozoites with significant levels of overexpression . Using the plasmid pTetGFPC . pac [15] , we first created the tetracycline-inducible C-terminal alpha2-annexin::GFP fusion vector , pTetA2::GFPC . pac . We PCR amplified the alpha2-annexin gene ( GiardiaDB GL50803_7796 ) using Giardia genomic DNA as a template with the following oligonucleotide primers: TA2giaF: 5′ GATCAGGCGCGCCATGCCGAAGCTATCCCAGATCGTCGC 3′ TA2giaR: 5′ ACCGGTAGAGCGCCGGCTCCGGCTCCGGCCGCTGCGCCCTCCCTTAGGCGCCAGAGGGTACAGAG 3′ The PCR amplicon yielded alpha2-annexin flanked by 5′ AscI and 3′ AgeI restriction sites , permitting subcloning into pTetGFPC . pac . G . intestinalis strain WBC6 was transformed by electroporation with roughly 50 µg of pTetA2::GFPC . pac DNA using the GenePulserXL ( BioRad ) as previously described [58] with the following modifications: 375V , 1000 µF , 25 ohms . Episomes in transformants were maintained by antibiotic selection using 50 µg/ml puromycin ( Sigma ) [58] . . To create the pTetA2_D122A_D275A::GFP . pac dominant negative we used site-directed mutagenesis ( Stratagene Quik-Change Site-Directed Mutagenesis Kit ) with pTetA2::GFPC . pac as a template and the following PCR primers: For D122A: A2g122F: 5′ TTCATGAAGGCTGTCGGCCG 3′; A2g122R: CGGCCGACAGCCTTCATGAA 3′; and for D275A: A2g275F: 5′ GGTGCTTTGCTAAGCGCA 3′; A2g275R: TGCGCTTAGCAAAGCACC . The two point mutations ( D122A and D275A ) were created within the alpha2-annexin gene contained in the pTetA2::GFPC . pac construct; point mutations were confirmed by DNA sequencing . Constructs were electroporated into Giardia as described above . Induction of expression of alpha2-annexin in inducible strains was achieved by using 15 µg/ml of doxycycline per 12 ml culture for 24-48 hours . The maximal induction of transgenes occurred at 6-8 hours , and continued for over 48 hours after removal of doxycycline ( see Video S1 ) . Induction and overexpression of alpha2-annexin ( D122A , D275A ) was confirmed using RT-PCR . Total cellular RNA was isolated from uninduced cells and from induced alpha2-annexin ( D122A , D275A ) cells at 24 and 48 hours after induction using the Cells-to-cDNA kit ( Ambion ) . GFP overexpression was compared using the relative method of quantification [59] , and GFP expression levels were normalized to the single copy giardial actin gene . Overexpression was determined from comparisons of normalized GFP expression in induced time points to uninduced controls . Thus , for quantitative analysis of expression , 1 µl aliquots of the cDNA synthesis reactions were used in subsequent actin ( actF 5′ CCTGAGGCCCCCGTGAATGTGGTGG 3′ and actR 5′ GCCTCTGCGGCTCCTCCGGAGG 3′ ) and GFP-specific ( GFPF 5′ GAGCTGTTCACCGGGGTGGTGCCC 3′ and GFPR 5′ CGGGCATGGCGGACTTGAAGAAGTCGTGC 3′ ) PCR amplifications with DyNamo HS SYBR Green qPCR Master Mix ( Finnzymes ) . QPCR was performed with the Opticon 2 system ( Bio-Rad ) . To demonstrate that RNA samples were not contaminated with DNA , control cDNA synthesis reactions were performed in the absence of reverse transcriptase . Immunostaining and paraformaldehyde fixation of the alpha2-annexin::GFP and PF16::HA strains was performed as previously described [28] with anti-TAT1 tubulin ( a kind gift from Keith Gull's laboratory ) or anti-HA ( Sigma ) antibodies at 1∶100 with Alexa 594 secondary antibody at 1∶400 ( Invitrogen ) . Images were collected with Metamorph image acquisition software ( MDS Technologies ) using a Leica DMI 6000 wide-field inverted fluorescence microscope with a PlanApo 100X , NA 1 . 40 oil immersion objective and captured with a Q imaging Rolera-MGi EMCCD . Serial sections were acquired at 0 . 2 µm intervals , and deconvolved using Huygens Professional deconvolution software ( SVI ) . For presentation purposes , 2D maximum intensity projections were created from the 3D data sets . Simple histogram adjustments were made to increase visualization of the dominant negative alpha2-annexin::GFP ( D122A , D275A ) strain . TIRFM uses evanescent waves that selectively illuminate and excite fluorophores in restricted regions of the specimen adjacent to the glass-water interface . This evanescent field decays exponentially away from the source , penetrating only about 100 nm into the sample [60] . For TIRFM , trophozoites were resuspended in 1X HEPES Buffered Saline ( HBS ) and incubated on ice for 10 minutes . To stain cell membranes , trophozoites were incubated for an additional 5 minutes on ice with CellMask Orange ( final concentration of 2 µg/ml; Invitrogen ) . Stained cells were concentrated by centrifugation ( 900 x g for 5 minutes ) and resuspended in 500 µl of warmed 37°C 1X HBS prior to imaging . A simple imaging chamber was created by mounting a coverslip to a standard slide with parallel lines of double-sided adhesive tape to define an imaging chamber . Cells were loaded into the chamber using a wide-bore pipette , and the edges were sealed with melted VALAP ( equal parts Vaseline , lanolin , and paraffin ) . This chamber provided a microoxic environment sufficient for short-term imaging experiments up to one hour . Live cell imaging was performed in a microscope stage incubator ( OkoLab ) at temperature of 35-37°C using a 515 nm laser with 10 ms exposures at 30-60 ms intervals for less than two seconds; no CCD gain was used . Images were collected with a QuantEM 512 SC EMCCD camera ( Photometrics ) on a 3i Marianas inverted spinning disk confocal microscope system . The TIRF angle was achieved with a 100X 1 . 46 NA oil immersion objective . Controls for the axial plane showed loss of signal/resolution past 200 nm , which was confirmed with axial confocal controls . Slidebook software ( Intelligent Imaging Innovations ) was used for minor image processing such as cropping and 2D intensity plots . To assess flagellar beating and motility in live trophozoites , we used live imaging with DIC microscopy . Dead and/or unattached cells were decanted from culture tubes 1 to 4 hours prior to imaging , and fresh medium was added . The culture was then incubated on ice for 15 minutes , and pelleted by centrifugation at 900 X g at 4°C . Cell pellets were resuspended in 500 µl 37°C medium and transferred into a 35 mm glass bottom Petri dish ( MatTek ) . The Petri dish was placed in a closed chamber and gassed with N2 . The cells were allowed to attach to the dish for 1 hour before the dish was removed and the lid was sealed with Parafilm . This chamber provided a microoxic environment sufficient for short-term imaging of 1 to 4 hours . Flagellar length and potential motility defects in the ventral flagellar pair ( synchrony , waveform , and frequency ) were assessed using live cell imaging . Flagellar length was measured for the membrane-bound portions from the flagellar pocket exit point to the distal flagellar tip . The synchrony of the ventral flagellar pair was observed by visually confirming whether ventral flagella beat in unison [6] . Ventral flagellar waveform was classified as sigmoidal ( wild type ) or abnormal and scored by the measurement of ventral flagellar amplitude and wavelength . Amplitude was measured by drawing a line from the basal bodies to the tip of the cell posterior and distance is measured from the line to the peak of the first wave [6] using Metamorph image acquisition software ( MDS Technologies ) . Flagellar wavelength was measured by drawing a line from the first wave peak ( proximal to the disc ) to the next peak toward the tip . Ventral flagellar beat frequency has been previously reported to be 18 Hz [7] , [10] , [13] . Based on these measurements we satisfied proper Nyquist sampling by imaging at least twice the published frequency , thus capturing 36-45 images/second with 22-30 ms exposures . The effect of shear forces on live trophozoites attached to a glass substrate was imaged using a syringe pump to create laminar flow with a temperature controlled Harvard Apparatus RC-31 parallel plate flow cell chamber ( Warner Instruments ) mounted onto an inverted Nikon Eclipse TS100 microscope with a 10X/0 . 25NA ADL objective and a Retiga 2000R CCD ( Qimaging ) as previously described [18] . The RC-31 chamber was fitted with a 100 µm narrow slot chamber gasket . The syringe pump was attached to the flow chamber via PE-10 and PE-90 tubing and a three-way stopcock for introduction of cells . For each time point , we chilled one 13 ml tube of giardial culture on ice and pelleted cells by centrifugation at 900 x g for 5 minutes . Cells were resuspended in 1 ml chilled medium and transferred into a 1 ml syringe . Cells were kept on ice no longer than 15 minutes . Shear force experiments were performed on trophozoites 24 hours after the introduction of the PF16 morpholino or 24 hours after induction of the alpha2-annexin dominant negative construct . Specifically , the flow chamber was pre-warmed with 37°C 1X HBS for 5 minutes followed by warmed medium for 1 minute . A volume of 250 µl of trophozoites was introduced via a 1 ml syringe with an 18 gauge blunt needle into the chamber via a port on a three-way stopcock added to the media input line . Next , trophozoites were flowed into the flow cell chamber and allowed to attach for 5 minutes . The line was rinsed of floating cells until none were seen , at a rate of 0 . 5 ml/min . A pre-assay image was taken ( Time = 0 ) . Cells were then challenged with a 3 ml/min or greater laminar flow rate [18] for 1 to 3 minutes as images were captured via phase contrast at 10 second intervals . As a control for detachment , 5% bleach at 37°C was introduced into the chamber , whereby all cells detached within 5 seconds . The rate ( greater than 3 ml/minute ) and chamber area ( 208 mm ) were then converted to a “shear” force of 1 . 5 nN . To quantify the fraction of cells that maintained attachment over a range of shear forces , cell counting was performed manually or using Metamorph image acquisition software ( MDS Technologies ) . The proportion of attached morpholino-treated cells and/or alpha2-annexin dominant negative trophozoites were normalized to the wild type attached trophozoites . Defects in the normal forces of attachment in trophozoites were assayed at the population level using a physical attachment assay [19] . Briefly , PF16 morpholino or alpha2-annexin dominant negative trophozoites were cultured and pelleted as above . Cells were resuspended in 10 ml of chilled media , and then 3 ml were transferred to custom sample holders capped with thick , circular glass slides . The cells were incubated at 37°C for 1 hour in a microoxic chamber ( see above ) to allow attachment to the glass slides . Sample holders were centrifuged at 37°C in a hanging bucket centrifuge ( Sorvall RC5C HB4 rotor 07 ) at 5 , 000 and 10 , 000 rpm ( 518 pN and 2 . 1 nN normal force ) . Non-centrifuged controls were prepared in the same manner and incubated at 37°C . Immediately after centrifugation , the glass slides were removed from the chamber and five fields ( ∼5000 cells ) were imaged and counted in phase contrast ( see above ) . The proportion of trophozoites maintaining attachment was normalized to the non-centrifuged control within that run ( the number of attached cells following centrifugation was divided by the number of cells attached in the non-centrifuged sample to determine the fraction of cells detached ) . The ability of cells to initiate attachment was measured by assaying the number of cells attached in one microscopic field over a 30 minute time period . First , the medium in a 6 ml culture was exchanged with 1X HBS at pH 7 . 0 and incubated on ice for 10 minutes . To stain the cell membranes for cell counting , CellMask Orange ( final concentration of 2 µg/ml; Invitrogen ) was added , and the cells were incubated on ice for an additional 5 minutes . Stained cells were pelleted by centrifugation and resuspended in 250 µl of chilled 1X HBS ( 1 . 1×106 cells per ml suspension concentration ) . For time-lapse imaging of attachment , 50 µl of stained cells were transferred to a microwell ( Corning ) overlaid with mineral oil and placed in a heated , closed stage chamber . The total number of attached trophozoites was imaged using a 10x/0 . 25 NA objective and quantified using time-lapse epifluorescence microscopy over a range of time intervals from 1 to 30 minutes . Total cell attached were counted using Metamorph image acquisition software . Alpha2-annexin ( XP_001706958; GiardiaDB: GL50803_7796 ) , PF16 ( XP_001705527: GiardiaDB: GF50805_16202 . | Giardia is a widespread , single-celled , intestinal parasite that infects millions of people and animals each year . Colonization of the small intestine is a critical part of Giardia's life cycle in any host . This colonization is initiated when cells attach to the intestinal wall via a specialized suction cup-like structure , the ventral disc . In the host , Giardia moves by beating four pairs of flagella; movement of the ventral pair has been implicated in attachment . This study shows that the beating of the flagella is not important for attachment , but rather for positioning Giardia close to the intestinal wall prior to attachment , and thus disproves the commonly held model of giardial attachment . This work implies that drugs targeting Giardia motility could prevent or slow attachment , leading to lower rates of infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"molecular",
"cell",
"biology",
"cell",
"biology",
"cell",
"adhesion",
"microbial",
"pathogens",
"biology",
"microbiology",
"quantitative",
"parasitology",
"parasitology"
]
| 2011 | Giardia Flagellar Motility Is Not Directly Required to Maintain Attachment to Surfaces |
The co-evolution of myxoma virus ( MYXV ) and the European rabbit occurred independently in Australia and Europe from different progenitor viruses . Although this is the canonical study of the evolution of virulence , whether the genomic and phenotypic outcomes of MYXV evolution in Europe mirror those observed in Australia is unknown . We addressed this question using viruses isolated in the United Kingdom early in the MYXV epizootic ( 1954–1955 ) and between 2008–2013 . The later UK viruses fell into three distinct lineages indicative of a long period of separation and independent evolution . Although rates of evolutionary change were almost identical to those previously described for MYXV in Australia and strongly clock-like , genome evolution in the UK and Australia showed little convergence . The phenotypes of eight UK viruses from three lineages were characterized in laboratory rabbits and compared to the progenitor ( release ) Lausanne strain . Inferred virulence ranged from highly virulent ( grade 1 ) to highly attenuated ( grade 5 ) . Two broad disease types were seen: cutaneous nodular myxomatosis characterized by multiple raised secondary cutaneous lesions , or an amyxomatous phenotype with few or no secondary lesions . A novel clinical outcome was acute death with pulmonary oedema and haemorrhage , often associated with bacteria in many tissues but an absence of inflammatory cells . Notably , reading frame disruptions in genes defined as essential for virulence in the progenitor Lausanne strain were compatible with the acquisition of high virulence . Combined , these data support a model of ongoing host-pathogen co-evolution in which multiple genetic pathways can produce successful outcomes in the field that involve both different virulence grades and disease phenotypes , with alterations in tissue tropism and disease mechanisms .
The establishment and spread of Myxoma virus ( MYXV; genus Leporipoxvirus; family Poxviridae ) in the wild European rabbit ( Oryctolagus cuniculus ) population of Australia in 1950 initiated the textbook case study of host-pathogen co-evolution on a continental scale [1 , 2] . The virus was novel to the European rabbit having evolved in the Brazilian tapeti ( Sylvilagus brasiliensis ) . In the tapeti MYXV induces an innocuous , localized cutaneous fibroma from which the virus is mechanically transmitted by mosquitoes or fleas . However , MYXV proteins that had evolved to suppress immune clearance and facilitate virus persistence in the natural host overwhelmed the immune system of the European rabbit causing the disseminated , lethal disease myxomatosis [2 , 3] . In Australia MYXV was released into naïve rabbit populations as a biocontrol agent . The initial virus , a strain known as SLS with a case fatality rate ( CFR ) estimated at 99 . 8% [4] , was rapidly replaced by moderately attenuated viruses , which by permitting longer survival of the infected rabbit were more likely to be transmitted by mosquitoes . The majority of these attenuated viruses still maintained relatively high CFRs of 70–95% [5 , 6] . Simultaneously , there was very strong selection pressure for the evolution of genetically resistant rabbits [7 , 8] . It is likely that the increased resistance in the rabbit population also drove selection for increased virulence in the virus to maintain transmissibility , as highly attenuated viruses transmitted poorly [9 , 10 , 11] . This large-scale evolutionary “experiment” is especially informative because it was repeated on a continental scale as MYXV was subsequently released in Europe . In June 1952 , a landholder in France inoculated two wild rabbits with a strain of MYXV ( Brazil Campinas/1949 ) , now termed the Lausanne ( Lu ) strain . From this starting point , MYXV spread through the wild and domestic rabbit populations of Europe [12] . Myxomatosis was detected in wild rabbits in Britain in October 1953 , probably due to the illegal release of an infected rabbit from France [13] . Despite attempts at control , the virus became established and spread throughout the wild rabbit population [14] , which was eventually reduced to perhaps 1% of the pre-myxomatosis level . Strikingly , although the European release involved a different starting strain , with different insect vectors and ecological conditions , it resulted in essentially the same outcome in terms of virulence evolution [1 , 12] . To facilitate evolutionary studies , field isolates of MYXV were classified into virulence grades from 1 to 5 based on average survival times ( AST ) in small groups of laboratory rabbits . The progenitor type viruses , killing 100% of infected rabbits , were of grade 1 virulence , while grade 5 viruses were highly attenuated with CFRs <50% . Most field isolates collected following the initial radiation in Australia were of grade 3 virulence with CFRs of 70–95% [5 , 6] . The grade 3 classification was later split into grade 3A and 3B to provide greater resolution [15] . Although the initial virus isolates in Britain were of grade 1 virulence [5] , attenuated viruses were detected within 12 months [16 , 5] . A large scale study of the virulence of UK MYXV isolates from 1962 revealed a similar evolutionary pattern to Australia , with the majority of isolates being of grade 3 virulence [15] . Studies of UK MYXV isolates from 1975 and 1981 confirmed the predominance of grade 3 viruses , but also showed that grade 2 viruses ( with CFRs of >95% ) had become much more common than in Australia; over 90% of viruses tested in 1981 were grade 3A or grade 2 , implying CFRs of >90% [17] . Genetic resistance to MYXV was documented much later in Britain than in Australia , but then rapidly increased in the wild rabbit population [18 , 19] and may again have driven selection for higher virulence . Although there have been detailed studies of the ecology , transmission , virulence and resistance of MYXV in Britain , little is known about the genetic and phenotypic basis of MYXV evolution and whether and how it parallels the evolutionary process seen in Australia . Indeed , previous studies have largely focused on early virus isolates sampled between 1954 and 1955 [20 , 21] . To address this central question in viral evolution we determined the genome sequences of 21 MYXV isolates sampled between 2008 and 2014 in Scotland and England . Importantly , we characterise the phenotype of a number of these viruses in laboratory rabbits compared to the progenitor Lu strain and reveal major changes in disease pathogenesis .
The prototype Lu sequence [22 , 23] consists of 161 , 777 nucleotides of double-stranded DNA with closed single stranded hairpin loops at the termini and duplicated terminal inverted repeats ( TIRs ) of 11 , 577 bp . The virus encodes 158 unique open reading frames ( ORFs ) , 12 of which are duplicated in the TIRs . The UK viruses descend from the Lu strain that was released into Europe as a biological control ( Fig 1 ) . The earliest sequences are from the grade 1 virulence Cornwall strain ( England/Cornwall/4-54/1 ) isolated in April 1954 and the grade 3 Sussex strain ( England/Sussex/9-54/1 ) from September 1954 and which quickly diverged from the introduced virus [20 , 21] . This divergence is captured in a phylogenetic analysis of these viruses along with an additional early isolate ( Belfast/1955 ) sequenced here , 21 viruses from 2008–2013 ( Table 1 ) , and a number of other European viruses ( Fig 1 ) . Notably , the viruses from Perthshire , Scotland can be divided into two lineages , with those sampled in 2008 ( lineage 1 ) phylogenetically distinct from those present in 2010–2013 ( lineage 2 ) . In 2009 , both lineages were present in the Perthshire population and it is possible that our limited sampling has not detected other examples of co-circulation . Within lineage 1 , the viruses sampled in 2008 are also distinct from those sampled in 2009 , while there is no obvious distinction within the sequences of lineage 2 from 2009–2013 . The three viruses sequenced from Yorkshire , sampled between 31/12/2008 and 8/3/2011 , represent a third distinct UK lineage . Despite the difference in progenitor viruses in Australia and Europe the subsequent evolution of these viruses is strongly clock-like . Using a Bayesian approach and a strict molecular clock the mean evolutionary rate for the 32 European viruses was estimated to be 0 . 99 x 10−5 nucleotide substitutions per site , per year ( subs/site/year ) ( 95% HPD values of 0 . 90–1 . 09 x 10−5 subs/site/year ) , while the equivalent value for the 25 Australian viruses was 1 . 03 x 10−5 subs/site/year ( 95% HPD values = 0 . 86–1 . 21 x 10−5 subs/site/year ) . Very similar rates were obtained using a variety of data sets and nucleotide substitution , molecular clock and demographic models ( Fig 1 ) . In addition , a regression of root-to-tip genetic distance against year of sampling for the combined Australian and European data set revealed strong temporal structure ( R2 = 0 . 93 ) , with a mean evolutionary rate of 1 . 04 x 10−5 subs/site/year that was very close to that estimated using the Bayesian approach for the entire data set at 1 . 02 x 10−5 subs/site/year ( 95% HPD values = 0 . 94–1 . 10 x 10−5 subs/site/year ) ( Fig 1 ) . The similarly of rates among viruses sampled on different continents suggests that their high evolutionary rate is largely a reflection of rapid background mutation as suggested for other pox viruses [25] . Under these evolutionary rates it is estimated that the two MYXV lineages from Perthshire shared a common ancestor between 1956 and 1963 , while the lineage leading to the Yorkshire viruses originated between 1953 and 1955 ( Fig 1 ) . Across all the UK viruses there were 162 non-synonymous mutations , 137 synonymous mutations and 26 insertion/deletion events within ORFs compared to Lu; 51 genes had no mutations and a further 23 only possessed synonymous changes ( Fig 2A ) . A comparison with the mutations observed in the Australian isolates ( Fig 2B ) revealed that different genes tended to show the highest numbers of mutation in each case . Indeed , only the M017L gene exhibited frequent mutation in both data sets ( Fig 2C ) . Overall , 23 genes contained no mutations among both the UK and Australian sequences and a further 23 had only synonymous changes ( S1 Table ) . As previously reported for MYXV in Australia [20 , 21] , single or multiple nucleotide insertions/deletions ( indels ) leading to the predicted disruption of ORFs were relatively common ( Table 2 ) . Disruptions of genes previously identified as having major virulence functions and leading to likely loss of function of the encoded protein occurred in M002L/R [26]; M004L/R [27 , 28]; M005L/R [29 , 30]; M148R [31] and M153R [32 , 33] . In addition , there was loss of the M009L ORF in Perthshire lineage 1 and by two independent mutations in the Yorkshire lineage , and of the M036L ORF in Perthshire lineages 1 and 2 . There was also an adjacent mutation in M036L in the early Sussex and Nottingham strains , with a possible reversal of this disruptive mutation in the Yorkshire lineage ( S1 Fig ) . Single viruses with gene disruptions were found in all three lineages: M135R ( Perthshire 1527 ) and M008 . 1L/R ( Perthshire 2409 ) have been shown to have virulence functions [34 , 35] . M009L has also been lost in most modern Australian viruses , as well as in some European isolates and in the Californian MSW strain of MYXV [20 , 21 , 36 , 37 , 24] , suggesting that this gene is not essential . In addition to indels that disrupted ORFs , there were a number of large and small indels within genes that were not disruptive ( S2 Table ) . Moreover , there were single nucleotide indels in multiple intergenic homopolymer regions and larger deletions in some blocks of intergenic repeat sequence elements . These will not be considered further . Temporal regulation of most MYXV genes has been predicted on the basis of conserved early , late or intermediate promoter motifs [22 , 38] . However , the transcription start sites of most MYXV mRNAs have not been mapped and hence actual expression may differ from that assigned [39 , 31 , 40] . In the UK sequences , mutations upstream of the M000 . 5L/R , M001L/R , M008 . 1L/R , M019L , M033L , and M153R genes were located close to potential promoter sequences and could conceivably alter transcription [41 , 42] . However , any effect was likely to be limited , with the possible exception of a mutation in the M153R putative promoter sequence in the Perthshire lineage 2 viruses which could conceivably decrease promoter activity . This mutation was also present in the Australian WS6 1071 virus . To evaluate how the genetic divergence from the Lu progenitor has affected disease phenotypes in the UK viruses , groups of six laboratory rabbits were infected with representative viruses from Perthshire lineages 1 and 2 , and all three Yorkshire lineage viruses , and their virulence and disease phenotypes compared to rabbits infected with the Lu progenitor virus . The virulence grade of each isolate was estimated using the method of Fenner and Marshall ( 1957 ) [5] . These virulence assignments were necessarily inferred since rabbits were euthanized and survival times ( ST ) estimated rather than using death as an endpoint ( Table 3 ) . Kaplan-Meier plots show the actual ST estimates rather than the normalized values ( Fig 3 ) . The Lu strain was tested as a control and had a similar AST to previous reports [5] . Notably , the grade 1 Yorkshire 135 isolate had a significantly lower ST than all other viruses tested including Lu . In our animal experiments the disease caused by Lu was indistinguishable from previous descriptions of Lu as the prototype European virus [5] , with the exception that we did not see the copious nasal discharge , likely because of the absence of Pasteurella multocida in the upper respiratory tract of the specific-pathogen-free rabbits . Notable features of Lu compared to the infections with the recent virus isolates were extreme swelling of the eyelids and lips , large size of the primary lesion , large numbers of secondary cutaneous lesions and a precipitous clinical decline between days 10 and 12 ( S3 Table; S4 Table ) . A striking feature of infection with some viruses from all three recent UK lineages was acute collapse resembling septic shock with relatively mild signs of myxomatosis . This was distinct from the disease caused by Lu . Hemorrhages in multiple tissues , massive pulmonary oedema and swollen , pale or granular livers were also frequently but not universally present , although the degree of pathology may have depended on timing of euthanasia or death . Aggregates of coccoid bacteria were often present in multiple tissues but with no apparent cellular inflammatory response ( Fig 4; S5 Table ) . These rabbits often had higher virus titres in liver and lung compared to rabbits infected with Lu ( S6 Table ) . Overall , disease phenotypes could be divided into: ( i ) a nodular cutaneous or “myxomatous” disease with prominent primary lesions at the inoculation site and secondary cutaneous lesions on ears , head , body and legs as seen with Lu , Perthshire 1527 and Yorkshire 127 viruses , or ( ii ) a disease that resembled the “amyxomatous” phenotype described in Europe [44 , 45 , 46] and characterized by a poorly defined primary lesion and no or very few secondary cutaneous lesions . This second phenotype was seen with Perthshire 1792 , 2082 , 2282 , Yorkshire col and Yorkshire 135 , while Perthshire 1537 had an intermediate phenotype ( Fig 5; S4 Table; S5 Table ) . Acute collapse was only seen with the amyxomatous infections . Other features of myxomatosis such as swollen heads , ears , eyelids and perineum were , to some degree , common to all infections . Prolonged incubation periods described for some amyxomatous viruses [44] were not seen . Distinctive differences were also present in the pathology of the acute collapse amyxomatous infections compared with Lu and the myxomatous phenotype ( S5 Table ) . Bacteraemia was not a feature of the Lu infections . Although bacteria were observed in a necrotic focus in the liver of one rabbit infected with Lu , these were associated with an acute inflammatory response . The large numbers of neutrophils seen deep in cutaneous tissues and within lymph nodes in the Lu infections ( Fig 5H ) were absent in rabbits with the acute collapse syndrome and lymph nodes and spleens tended to be more depleted of lymphocytes in these rabbits . Late clinical signs in longer surviving or recovering rabbits were fairly typical of those described for myxomatosis caused by moderately attenuated viruses [5] , with the exception that the amyxomatous viruses did not induce secondary lesions ( S4 Table; S5 Table ) . The prolonged duration of high virus titres in the epidermis of primary or secondary lesions or in sites such as eyelids or ears is critical for transmission by arthropod vectors [9] . In general , longitudinal biopsy samples showed that levels of virus in the primary lesions , measured by qPCR , increased over the first 10 days to > 108 copies/mg and were then reasonably stable , albeit with reduced numbers of rabbits available for biopsy at later time points ( S2 Fig ) . However , two virus infections had consistently lower virus loads: the grade 5 Perthshire 1527 and the grade 2/3 Yorkshire 127 strain . Both viruses had the nodular myxomatous phenotype and the lower loads were probably due to cell destruction in the epidermis . Despite the limited nature of the primary lesion in the amyxomatous phenotypes ( Fig 5I ) they had very high levels of virus . Similar results were obtained with titres measured by plaque assay on autopsy samples ( S6 Table ) . Titres in the Lu infected rabbits were also relatively low , likely because of the highly scabbed and degenerate nature of the lesion ( S6 Table; Fig 5 ) . Biopsies were not collected from rabbits infected with Yorkshire 135 or Lu . Taken together with the histological and gross appearance of the primary lesions , these results indicate that the tissue response to the amyxomatous viruses is entirely different to that induced by Lu , but that this is not due to reduced virus replication . Despite the observed differences in disease phenotype and virulence , viruses within each lineage exhibit limited sequence divergence . For example , Yorkshire 127 caused the nodular cutaneous phenotype while the closely related Yorkshire 135 and Yorkshire col caused the amyxomatous phenotype ( Fig 3 ) . All three Yorkshire viruses have lost the functional domain of the M005L/R gene and have disrupted M009L and M153R genes ( Table 2 ) . There are six amino acid differences between Yorkshire 135 and Yorkshire Col and seven between Yorkshire 135 and Yorkshire 127 ( S7 Table ) . The Perthshire lineage 1 viruses are more complicated , as the 2008 viruses ( 1527 , grade 5 and 1537 , grade 3/4 ) have a disrupted M002L/R gene and Perthshire 1527 has a disrupted M135R gene; both are virulence determinants in Lu [34] . These genes are intact in the amyxomatous 2009 Perthshire 1792 virus ( grade 2 ) . As with the Yorkshire lineage , these viruses only differ at a small number of amino acid sites ( S8 Table ) . Both Perthshire lineage 2 viruses tested had the amyxomatous phenotype and were of grade 3 virulence . Apart from the premature termination of M008L/R in 2082 , there are only four amino acid differences between these viruses ( S9 Table ) . Phenotypically , it was difficult to differentiate these grade 3 viruses from the grade 2 Perthshire 1792 and Yorkshire Col . Overall , these results suggest that single amino acid changes can have a major impact on disease phenotype and virulence gene disruption may be compensated by epistatic mutations or other mechanisms .
Our genome-scale evolutionary analysis reveals that multiple lineages of MYXV have circulated in UK rabbits . In particular , the single lineage of viruses from Yorkshire and the two lineages present in Perthshire clearly diverged relatively early in the epizootic and have evolved independently ever since . This separation of the English and Scottish viruses could reflect a simple biogeographic division and a lack of virus gene flow , particularly since the European rabbit flea ( Spilopsyllus cuniculi ) is the main arthropod vector in the UK so that virus spread depends on movement of rabbits carrying fleas [47 , 48] . However , the phylogenetic separation between the two Scottish lineages is harder to explain as they were sampled within three kilometres of each other . Because these two lineages differ in the range of temporal sampling ( 2008–2009 ) and ( 2009–2013 ) it is possible that the later sampled lineage is a more recent invader into the study area and has outcompeted the previously existing lineage . Anecdotally , in 2009 this study site experienced a high mortality of rabbits due to myxomatosis , compatible with the possible invasion of a new strain into the area . Importantly , our comparison of MYXV genome sequences from the UK and Australia confirms previous conclusions that there is no single pathway to attenuation from the progenitor viruses or from attenuation back to virulence [20] . Indeed , it is striking that there are almost no shared mutations between the viruses from the two radiations despite the large number of complete genomes now sequenced . Hence , evolutionary success in these large genome DNA viruses has clearly resulted from the exploration of multiple evolutionary pathways along which different disease phenotypes appear . Indeed , our animal trials reveal that the clinical phenotype of a number of the UK viruses showed dramatic changes compared to the progenitor Lu virus , as well as within and between the modern viral lineages . Generalized disease seems critical for efficient virus transmission in European rabbits , with rabbits that survive infection ( and therefore control virus replication ) being poor transmitters [10] . In addition , resistance is manifest as control of virus replication rather than prevention of infection [49 , 50 , 51] , so is likely to select for virus mutations that can overcome this control . The emergence of genetic resistance in the wild rabbit population likely shifted selection towards more virulent viruses ( when tested in non-resistant rabbits ) to maintain this nexus between virulence and transmission , in turn setting up an arms race between host and virus . As we describe here , this can lead to dramatic changes in the disease phenotype in non-resistant rabbits . There is an implicit idea that changes in virulence will be due to mutations in genes involved in immunomodulation or host-range functions [40] . The role of many MYXV genes in virulence has been defined by single gene knock-out studies using the Lu strain or an early French derivative , the T1 strain [52] . In particular , the M005L/R and M153R genes have each been shown to have major virulence functions . Rabbits infected with knock-outs of either gene had a much lower CFR: 30% for ΔM153R and 0% for ΔM005L/R compared to 100% for Lu [32 , 29] . However , all three Yorkshire viruses have mutations that are predicted to disrupt both these genes causing loss of key functional domains [33 , 30] but have CFRs of nearly 100% . This suggests three possible explanations for retained virulence: ( i ) epistatic mutations compensating for the loss of these genes; ( ii ) a mechanism for suppressing reading frame disruptions; or ( iii ) functional activity retained by the truncated protein ( potentially in a new role ) [53] . Although it seems likely that unique amino acid substitutions are often responsible for alterations in virulence , the number of such amino acid changes evidently makes specific virulence determinants difficult to identify . Similarly , the Californian MSW strain of MYXV , which is found in S . bachmani in North America and is the most virulent strain of MYXV described for European rabbits [5 , 54] , has disrupted multiple virulence genes , suggesting that multiple epistatic mutations play a role in virulence determination [36] . As well as broad trends in virulence during the early radiation , changes were also observed in the clinical appearance of infected rabbits , with a relatively rapid evolution of a flat lesion morphology in both Australia and Europe rather than the domed SLS and Lu lesions [5 , 15] . More recently , the amyxomatous phenotype in European isolates has been distinguished from the nodular type of disease by having few or no cutaneous lesions and , in some cases , apparently prolonged incubation periods [44 , 55 , 56] . For some Australian isolates the amyxomatous phenotype is seen in laboratory rabbits , although the same virus causes a nodular phenotype when tested in resistant wild rabbits suggesting that changes in the pathogenesis of the disease have occurred due to selection in resistant wild rabbits [57] . Combined , these data strongly suggest that the accumulation of mutations in field strains of MYXV has caused changes in the pathogenesis of myxomatosis , such that we now see a spectrum of disease types that depend on the interactions between the virus genome and the genetics of the rabbit and non-genetic ( rabbit ) factors such as microbial flora , parasites , and abiotic environmental factors including temperature [58] . As an example , field isolates of European amyxomatous viruses tested in specific pathogen-free laboratory rabbits caused relatively minor disease with few fatalities . However , the same viruses tested in rabbits from commercial rabbitries caused significant disease with severe bacterial bronchopneumonia as the most common cause of death [46 , 59] . Different environmental conditions and vectors may therefore facilitate selection of virus strains that are more successful in particular niches . For example , in the farmed domestic rabbit populations in Europe where there has been no selection for resistance , we may expect low virulence strains predominantly transmitted by contact , strains with prolonged incubation periods [60 , 61] , or high virulence strains that can overcome imperfect vaccination [60 , 56 , 37] . With the exception of Yorkshire 127 , rabbits that died or required euthanasia early in the course of the disease had very different clinical signs from those infected with Lu . Hemorrhage and acute pulmonary oedema were common together with high titres of virus in lungs and liver . In some cases , large numbers of coccoid bacteria were present in multiple tissues , but did not elicit a visible cellular inflammatory response . Lymphocyte depletion from lymph nodes and spleens was relatively common . Despite extremely high virus titres , there was very limited pathology in the epidermis and dermis of the primary inoculation site . This suggests an acute overwhelming of the rabbit immune response triggered by high viral titres in critical tissues . This outcome is also clearly distinct from the secondary gram negative bacterial infections ( Pasteurella multocida , Bordetella bronchiseptica ) described in the upper respiratory tract for rabbits infected with the progenitor viruses or the bacterial bronchopneumonia described with isolates from rabbit farms [59] . In our study , rabbits that did not die of acute disease developed more typical signs of myxomatosis , although upper respiratory tract occlusion and discharge was relatively mild , possibly reflecting the specific-pathogen free status of the rabbits . Whether the difference in survival time and clinical disease between the acutely affected animals and the more chronically affected longer term survivors is related to genetic factors in the outbred rabbits or some stochastic factor early in the course of disease is not clear , but these animals clearly have a different form of the disease . Virulence , using the definitions of Fenner and Marshall ( 1957 ) , essentially meant the AST . However , this raises the question of what virulence means in terms of how a strain of MYXV causes disease ? Does a more virulent virus cause a different disease , or are there many pathways to death in an infected rabbit such that the phenotype seen may be due to which particular mechanism occurred in an individual rabbit . Thus , in one animal we see hemorrhage and pulmonary oedema , yet in another we see acute death without pulmonary oedema and hemorrhage , which might have developed if the animal had survived a few hours longer . It is possible that some of the longer-term survivors have a milder form of the disease at this stage and will go on to develop the more typical form of myxomatosis , and this pathway seems to predominate in attenuated viruses such as Perthshire 1527 . Clearly , virulence in this case is a more nuanced concept than generally depicted in studies of its evolution . The parallel evolution of virulence in MYXV in the Australian and British epizootics was evidently not accompanied by the acquisition of similar mutational changes . Our detailed examination of genomics and disease phenotypes of recent isolates of MYXV from the UK radiation reveals that highly virulent and highly attenuated viruses were present in the field , but that disruptions to major virulence genes were not necessarily associated with attenuation . More striking was that the disease caused by many of these viruses was clinically distinct from that caused by the progenitor Lu strain , with alterations in tissue tropism and pathogenesis in acutely affected rabbits , again demonstrating that the virus is able to explore many pathways to evolutionary success .
Sampling was performed according to field procedures approved by the Institutional Animal Care and Use Committee of The Pennsylvania State University ( IACUC # 26383 and 34489 ) . Animal experiments were conducted under protocols approved by the Institutional Animal Care and Use Committee , Pennsylvania State University ( IACUC # 33615 and 42748 ) . All animal work adhered to the guidelines laid out in the Guide for the Care and Use of Laboratory Animals . 8th ed . National Research Council of the National Academies . National Academies Press Washington DC . The virus isolates sequenced in this study are listed in Table 1 . Samples were taken from rabbits with clinical myxomatosis gathered at multiple locations on two sites , the first located in Perthshire in central-eastern Scotland , and the second in North Yorkshire , England , collected as part of other field studies [62 , 63 , 64 , 65] . An early isolate sampled in Belfast , Northern Ireland in 1955 , was also sequenced . All viruses were isolated in RK-13 cells and passaged between 1 and 3 times to prepare seed and working stocks , from which virus DNA was prepared [66] . An aliquot of virus from the DNA preparations was used for rabbit infections . Virus genomes were sequenced on three different platforms: the Illumina HiSeq 2000 and MiSEq , and the Ion Torrent . For the HiSeq200 , template viral DNA was processed using a TruSeq DNA sample preparation kit ( Illumina ) to produce a multiplex library for sequencing . Briefly , extracted viral genomic DNA ( gDNA ) was sheared with a Covaris AFA system , creating fragments of 50 to 7 , 000 bp . After end-repair , purification , and 3′ adenylation , bar-coded sequencing adapters were ligated , and 400- to 500-bp fragments were purified . Fragment enrichment and clean-up were performed with AMPure XP beads . Individual library components were quantitated by quantitative PCR ( qPCR ) , normalized , and pooled into a final sequencing library consisting of eight different viral genomes ( this included seven MYXV strains that were analyzed in a separate study ) , which was run on an Illumina HiSeq2000 to generate 100-bp paired-end reads . For the MiSeq , libraries were produced using the Nextera XT DNA kit ( Illumina ) . Extracted DNA samples were quantified using a Qubit fluorometer and 1ng of each sample was used as input DNA . The standard workflow was followed: duel index barcoding of the tagmented DNA was done according to the low plexity requirements and 1 . 8x AMPure XP beads were used to purify the library DNA . Library normalization was performed using Illumina beads . Multiplexing of the final library occurred according to Illumina recommendations . Briefly , 5 μl of each of the 14 finished , bead-normalized libraries were combined into a library pool . Next , 24 μl of this mix was transferred to a new tube containing 576 μl HT1 buffer , mixed well , and placed at 96°C for 2 minutes to denature , followed by cooling on ice for at least 5 minutes . Denatured 8pM PhiX was then combined with the denatured library pool in a total volume of 600 μl and a final concentration of 5% to produce the final sequencing pool . Sequencing was performed on an Illumina MiSeq using either 2x75bp V3 or 2X250 V2 paired-end kits , yielding approximately 14 . 5M paired-end reads for each run . Isolates 1527 and 2282 were sequenced on the Ion Torrent . Genomic DNA was sheared and converted into libraries with the Ion Xpress Plus fragment kit ( Ion Torrent ) by following the manufacturer’s instructions . Briefly , 200ng of gDNA was sheared for 20 minutes followed by purification , nick repair and adapter/barcode ligation . The DNA libraries were then size selected on the E-Gel SizeSelect ( Invitrogen ) platform to yield insert sizes of ~200 bp . Libraries were quantitated on the Bioanalyzer ( Agilent ) and combined in equimolar amounts to make the final sequencing pool . This pool was sequenced on the Ion Torrent with a 316 chip and a 200 base read length target , yielding 2 . 6M useable reads . Demultiplexed reads were quality trimmed using the trim . pl perl script ( http://wiki . bioinformatics . ucdavis . edu/index . php/Trim . pl ) and assembled with the Velvet de novo assembler iterated across a range of k-mers from 45 to 65 for each assembly [67] . Contigs were ordered into a single scaffold for each genome using the Abacas . pl script [68] and the Lu genome as reference ( GenBank accession AF170726 ) , and for each assembly the k-mer that generated the most complete coverage of the reference genome was selected for finishing and downstream analysis . The quality of each scaffold was verified by remapping the untrimmed reads to the assembly using Smalt ( http://www . sanger . ac . uk/science/tools/smalt-0 ) . One region of ambiguous assembly was amplified by PCR and sequenced using Sanger methodology to confirm the assembly . A nucleotide deletion within a homopolymer run in the M153R gene was also confirmed by Sanger sequencing . In every case , only one complete or near complete copy of the terminal inverted repeat ( TIR ) was assembled at either the 5’ or the 3’ end . The Belfast 1955 isolate was assembled de novo on a 100K sub-sample of the cleaned , paired-end reads using CLC Genomics ( version 8 ) with a word and bubble size of 30 nt and 150 nt , respectively . This yielded two contigs corresponding to the core genome ( ~138K ) and TIR ( ~11K ) . The TIR contig was duplicated and reverse complemented before manually assembling onto the core genome , and then all the cleaned , paired-end data was re-mapped back to confirm final assembly . Genome annotation was transferred from the Lu strain to the newly sequenced MYXV genomes using the Rapid Annotation Transfer Tool [69] . EMBL flatfiles of transferred gene models were then inspected and compared to the Lu reference using the Artemis Comparison Tool [70]; incorrect models were corrected , and new gene models added where transfer had not occurred . Nucleotide sequence accession numbers: all genome sequences generated here have been deposited in GenBank ( https://www . ncbi . nlm . nih . gov/ ) under accession numbers KY548792-KY548813 ( S10 Table ) . The 22 MYXV genome sequences determined here were combined with 35 complete genomes available on GenBank , representing 25 from the Australian outbreak ( including the SLS release strain ) and 10 from Europe ( including the Lu release strain ) ( S10 Table ) . These sequences were initially aligned in MUSCLE [71] and adjusted manually , resulting in a final sequence alignment data set of 57 sequences 163 , 645 bp in length . Because the sequences are highly conserved , the locations of synonymous and non-synonymous mutations in these sequences were determined manually . An initial phylogenetic tree of these sequences was inferred using the maximum likelihood procedure available in the PhyML package [72] . This analysis utilized the HKY+Γ4 model of nucleotide substitution and NNI+SPR branch-swapping . To test for the presence of recombination we utilized the RDP , Genecov and Bootscan methods ( with default settings ) available within the RDP4 package [73] . No significant evidence for recombination was found . To determine the rate of MYXV evolution we first assessed the degree of clock-like structure in the data using a regression of root-to-tip genetic distances on the ML tree inferred above against the year of virus sampling using TempEst [74] . As this analysis revealed strong temporal structure ( see Results ) , we next inferred the rates and dates of viral evolution using the Bayesian Markov chain Monte Carlo ( MCMC ) approach available in the BEAST package [75] . For this analysis we used a range of nucleotide substitution ( HKY+Γ4 , GTR+Γ4 ) , molecular clock ( strict , relaxed uncorrelated lognormal ) and demographic ( constant , Bayesian skyride ) models . As these gave strongly overlapping results we based our analysis on the simplest model: HKY+Γ4 , strict clock , constant population size ( Fig 1 ) . All analyses were run twice and for sufficient time ( 100 million generations ) to ensure that convergence was achieved , with statistical uncertainly manifest in values of the 95% highest posterior distribution ( HPD ) . The posterior distribution of trees from the HKY+Γ4 , strict clock , constant population size run was also used to infer a maximum clade credibility ( MCC ) tree ( Fig 1 ) . The degree of support of individual nodes is depicted as posterior probability values . New Zealand White male laboratory rabbits ( Oryctolagus cuniculus ) of four months of age were purchased from Harlan Laboratories ( Oakwood facility ) . Rabbits were specific-pathogen-free for Pasteurella multocida and Bordetella bronchiseptica . Animals were housed in individual cages on a 12h light regime , fed 125 g of standard pellets per day and allowed 10 days to acclimate in the facility prior to infection . Groups of six rabbits were inoculated with 100 pfu of virus intradermally in the rump and monitored closely over the course of the infection . Daily clinical examination included: rectal temperature , body weight , primary lesion size and shape at the inoculation site , secondary lesion size and distribution , plus semi-quantitative scoring on a 0 to 3 scale for demeanour , eyelid swelling , ear swelling , anogenital swelling , scrotal oedema , blepharoconjunctivitis , nasal discharge and respiratory difficulty . Food and water intake were recorded and fecal and urinary output monitored by inspection of collecting trays under the cages . Rabbits were euthanized based on the degree of clinical severity using respiratory difficulty , depression , inanition , reluctance to move , weakness on handling , weight loss and failure to eat or drink as indicators; any rabbit exhibiting pain or with a subnormal temperature was immediately euthanized . To monitor virus replication at the primary inoculation site , 1 mm diameter dermal punch biopsies were collected: in each group , three rabbits were sampled at day 5 post-infection and three at day 7; thereafter , each surviving rabbit was sampled at 5 day intervals . DNA was prepared using the DNeasy kit ( Qiagen ) . Rabbits were autopsied as soon as possible after death and bodies refrigerated if autopsy was delayed . Samples of the primary lesion and other tissues were collected for virus titration and histology but only from euthanized rabbits or rabbits that died within 1–2 hours prior to autopsy . Blood samples were collected from the marginal ear vein at days 0 and 10 , or following euthanasia , by cardiac puncture . Hematology was performed by the Centralised Biological Laboratory Facilities , the Pennsylvania State University . Because of the unusual virulence of the Yorkshire 135 virus , we tested whether there was any adventitious agent in the virus preparation by challenging immune rabbits with Yorkshire 135 . The only reaction was a swelling at the inoculation site , which resolved by day 6 . This is typical of what is seen when immune rabbits are challenged . While this does not completely exclude an adventitious agent that was only pathologic in the context of a highly immunosuppressive MYXV infection , it strongly supports the hypothesis that the peracute disease seen with Yorkshire 135 was indeed due to MYXV . To enable comparison with previous studies of MYXV , survival times ( ST ) from inoculation to death were estimated for rabbits that were euthanized as follows: ( i ) moribund rabbits were assigned the time of euthanasia as the ST; ( ii ) rabbits that were not expected to survive the next 24 hours were assigned an additional ST of +12 hours; and ( iii ) rabbits euthanized for humanitarian reasons were assigned a ST of +48 hours . Animals found dead were assigned a ST half-way between the time of last observation and finding the body . Average survival times ( AST ) for each group were calculated from individual ST normalized using the procedure of Fenner and Marshall ( 1957 ) [5] as log10 ( ST-8 ) and then back-transformed; a survival time of 60 days was assigned to rabbits that recovered or were alive at the end of the trials and considered likely to recover . If more than two rabbits survived , the virulence grade was assigned based on the CFR and clinical severity . Virulence grades were based on Fenner and Marshall ( 1957 ) [5] as modified by Fenner and Woodroofe ( 1965 ) [43] ( Table 4 ) . Data were also analysed using Kaplan-Meier survival plots ( using actual inferred survival times rather than the normalized survival times ) and tested for statistical significance by log rank test implemented in SigmaPlot . Quantitative PCR ( qPCR ) was performed on an ABI 7500-fast machine , using the Quantifast Sybr green kit ( Qiagen ) , by amplification of a 126 bp fragment ( nt 584–710 ) from the M080R gene from DNA extracted from primary lesion biopsies . This was quantified on a standard curve using a linearized control plasmid containing a 642 bp region of the M080R gene ( nt 241–883 ) . None of the UK viruses have mutations in this sequence . Virus titres were expressed as genome copy number/mg tissue . The qPCR primers used were: M080 qPCR Forward: 5' TATCAAACAACCTCCGCATACC 3' ( M080R 584–605 ) and M080 qPCR Reverse: 5' CTCCCATAACGCTTCCGAC 3' ( M080R 710–692 ) Samples of the primary lesion , lung , liver , spleen , and right popliteal lymph node were collected at autopsy from euthanized rabbits . Tissues were homogenized by Tissuelyser ( Qiagen ) . Virus was titrated on RK-13 cell monolayers as previously described [49] with titres expressed as pfu/g of tissue . | Species jumps and subsequent pathogen evolution are of increasing importance in a globally connected world . The co-evolution of myxoma virus and the European rabbit following the introduction of the virus into Australia in 1950 is the canonical case of host jumping and host-pathogen co-evolution on a continental scale . This natural experiment was repeated with the release of a separate strain of myxoma virus in Europe . On both continents moderately attenuated strains of virus became dominant while rabbits were selected for resistance to myxomatosis . Here we examine the genotypic and phenotypic evolution of myxoma virus in Great Britain compared to Australia and show that despite ecological convergence and equivalent evolutionary rates , the virus has followed distinct evolutionary pathways on both continents with few shared mutations . Furthermore , we reveal novel mechanisms of pathogenesis and tissue tropism compared to the progenitor virus , and that the disruption of virulence genes is compatible with high virulence . This suggests that mutations have occurred that can compensate for the loss of virulence genes driven by the nexus between virulence and transmission in an ongoing host-pathogen arms race . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"organismal",
"evolution",
"medicine",
"and",
"health",
"sciences",
"microbial",
"mutation",
"poxviruses",
"myxoma",
"virus",
"pathology",
"and",
"laboratory",
"medicine",
"genome",
"evolution",
"insertion",
"mutation",
"pathogens",
"microbiology",
"vertebrates",
"rabbits",
"animals",
"mammals",
"animal",
"models",
"viruses",
"mutation",
"signs",
"and",
"symptoms",
"dna",
"viruses",
"experimental",
"organism",
"systems",
"microbial",
"evolution",
"research",
"and",
"analysis",
"methods",
"genomics",
"medical",
"microbiology",
"microbial",
"pathogens",
"lesions",
"molecular",
"evolution",
"gene",
"disruption",
"viral",
"evolution",
"diagnostic",
"medicine",
"virology",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"amniotes",
"organisms"
]
| 2017 | Genomic and phenotypic characterization of myxoma virus from Great Britain reveals multiple evolutionary pathways distinct from those in Australia |
Cysteine-rich receptor-like kinases ( CRKs ) are transmembrane proteins characterized by the presence of two domains of unknown function 26 ( DUF26 ) in their ectodomain . The CRKs form one of the largest groups of receptor-like protein kinases in plants , but their biological functions have so far remained largely uncharacterized . We conducted a large-scale phenotyping approach of a nearly complete crk T-DNA insertion line collection showing that CRKs control important aspects of plant development and stress adaptation in response to biotic and abiotic stimuli in a non-redundant fashion . In particular , the analysis of reactive oxygen species ( ROS ) -related stress responses , such as regulation of the stomatal aperture , suggests that CRKs participate in ROS/redox signalling and sensing . CRKs play general and fine-tuning roles in the regulation of stomatal closure induced by microbial and abiotic cues . Despite their great number and high similarity , large-scale phenotyping identified specific functions in diverse processes for many CRKs and indicated that CRK2 and CRK5 play predominant roles in growth regulation and stress adaptation , respectively . As a whole , the CRKs contribute to specificity in ROS signalling . Individual CRKs control distinct responses in an antagonistic fashion suggesting future potential for using CRKs in genetic approaches to improve plant performance and stress tolerance .
Receptor protein kinases play key roles in mediating perception of extracellular signals . These signals trigger intracellular signalling cascades allowing cells to respond and adapt to internal and external stimuli . Receptor kinases contain an extracellular signal-sensing domain connected by a single transmembrane domain to an intracellular protein kinase domain [1] . During evolution , different systems for the same function have emerged in animals and plants: animals deploy receptor-tyrosine kinases whereas plants utilize receptor-like kinases ( RLKs ) , which are dual specificity serine/threonine and tyrosine kinases [2–4] . In contrast to mammals , all sequenced plant genomes contain a large number of RLKs , as illustrated by Arabidopsis and rice which encode more than 600 and 1100 RLKs in their genomes [1] , respectively . The highly diverse extracellular regions of RLKs typically contain one or more protein domains or combinations of different domains . These domains have been used to divide RLKs into different sub-groups [1] . To date only a few RLKs have been functionally characterised but expression analyses have linked a large number of RLKs to many different physiological processes and signalling networks in plant development , pathogen defence , and abiotic stress response [5–8] . The cysteine-rich receptor-like kinases ( CRKs , originally referred to as domain of unknown function 26 [DUF26] RLKs [9] ) represent one of the largest groups of RLKs with 44 members in Arabidopsis thaliana [7] . Most CRKs have a typical RLK domain architecture , but three CRKs ( CRK43 , CRK44 and CRK45 ) consist only of the cytoplasmic domain reminiscent of receptor-like cytoplasmic kinases ( RLCKs ) [1] . The extracellular domain of CRKs encompasses two copies of the DUF26 domain ( PF01657; http://pfam . sanger . ac . uk/family/PF01657; stress-antifung domain ) , which contains three cysteine residues in a conserved configuration ( C-X8-C-X2-C ) and is a predicted target for redox modifications . The DUF26 domain is also present in eight Arabidopsis PLASMODESMATA-LOCATED PROTEINs ( PDLPs ) [10] . The domain structure of these PDLPs is similar to CRKs but lacks the intracellular protein kinase , analogous to the leucine-rich repeat ( LRR ) receptor-like proteins ( RLPs ) . Experimental evidence suggests that PDLPs are involved in the regulation of cell-to-cell communication and are important for pathogen defence [11 , 12] . Furthermore , more than 50 secreted proteins in Arabidopsis contain DUF26 domains but their roles have so far not been elucidated . Several CRKs show elevated transcript levels in response to salicylic acid ( SA ) and pathogens [13–17] as well as ozone ( O3 ) and drought [7 , 18] . Altered transcript abundance due to conditions affecting cellular redox and reactive oxygen species ( ROS ) balance [7 , 19] , and the presence and spacing of the conserved cysteines in the DUF26 domain suggest that CRKs might be connected to ROS and redox signalling [7 , 14 , 20] . However , the functional role of the DUF26 domain is still unclear . Previous studies have suggested the involvement of some CRK family members in pathogen defence and osmotic stress . Overexpression of CRK4 , CRK5 , CRK19 , and CRK20 induced hypersensitive response-like ( HR-like ) cell death [13 , 14] and overexpression of CRK4 , CRK6 , CRK13 , and CRK36 resulted in enhanced tolerance to the bacterial pathogen Pseudomonas syringae pv . tomato DC3000 ( Pto DC3000 ) [21 , 22] . Also a loss-of-function mutant crk20 showed a slight reduction in Pto DC3000 growth [23] . A Medicago truncatula CRK , SymCRK , was found to be involved in preventing early senescence and defence responses during symbiotic interactions [24] . Knock-down of CRK36 resulted in increased sensitivity to abscisic acid ( ABA ) and osmotic stress [25] , and altered seed germination 6 ( asg6; crk2 ) , a mutant deficient in CRK2 function , has been associated with changes in seed germination in response to ABA [26] . A mutation in crk7 led to slightly increased sensitivity to extracellular ROS [27] , and a mutation in crk5 resulted in impaired stomatal conductance , accelerated senescence as well as enhanced cell death in response to ultraviolet radiation [28] . Given the large number of CRKs and several transcript profiling experiments which suggest that CRKs are involved in a variety of environmental responses [6–8 , 18 , 19] , it is surprising how little is known about the physiological and biochemical functions of this RLK family . However , based on transcriptional analysis CRKs might have far more complex functions , for example also in signalling in response to N-acetylglucosamine ( GlcNAc ) oligomers in the plant cell wall [29] . In this study we describe a comprehensive phenotypic analysis of a T-DNA insertion collection for the entire CRK gene family . This phenomics approach revealed novel roles for CRKs in control of plant development and biotic and abiotic stress adaptation . In spite of high amino acid sequence similarity , we observed that many CRKs mediate specific functions , with CRK2 and CRK5 playing predominant roles in growth regulation and stress adaptation , respectively . Our results imply a model for CRK function , placing CRKs as putative elements between ROS production and downstream signalling leading to pathogen- and abiotic stress-induced stomatal closure . This provides a framework for future detailed analysis of the molecular mechanisms underlying CRK signalling .
Transcriptional analyses have shown that CRK genes are responsive to several external stimuli . However , only limited information is available on their physiological roles . Based on the amino acid sequences of the coding region , the 44 Arabidopsis thaliana CRKs ( plus the putative pseudogene CRK35 At4g11500 and the truncated CRK9 At4g23170 ) form five distinct groups ( Fig 1A ) . Similar groups can be identified in phylogenetic trees based on the intracellular kinase domain ( S1A Fig ) as well as on the extracellular region ( S1B Fig ) . This suggests co-evolution of the extra- and intracellular domains of Arabidopsis CRKs . A group of six CRKs , group I , constitutes a basal clade which forms a sister group distinct from groups II-V . The position of CRK45 , At4g11890 , which lacks the extracellular and transmembrane region , is ambiguous as it clusters with low bootstrap support with the basal group in the phylogenetic tree based on the entire coding region ( Fig 1A ) but as a sister to groups II and III in the tree based on the kinase domain ( S1A Fig ) . Thus , CRK45 is not assigned to any CRK subgroup . Group I CRKs are distributed across chromosomes 1 , 4 and 5 and only CRK2 and CRK3 are located next to each other , whereas genes encoding CRKs in groups II-V are , with the exception of CRK4 on chromosome 3 , located on chromosome 4 and organized in repeats forming clusters of CRK genes ( S2 Fig ) . In order to investigate the function of individual CRK family members , a collection of 82 T-DNA insertion mutants was compiled from the Nottingham Arabidopsis Stock Centre ( NASC; Figs 1B and S3 and S4 and S1 Table ) . CRK35 ( putative pseudogene ) and CRK9 ( truncated ) were excluded and , under our conditions , no homozygous T-DNA insertion lines could be obtained for crk27 , crk34 , and crk44 in the Col-0 background ( Figs 1B and S3 and S1 Table ) . A total of 50 homozygous T-DNA insertion lines representing 41 crk mutants were isolated . In 21 crk lines the corresponding wild-type CRK transcript was not detected while levels were reduced in thirteen lines ( Figs 1B and S4 and S1 Table ) . Of those thirteen lines seven have an insertion in an exon suggesting that they would produce a truncated protein while five lines carried insertions in introns . Only one line , crk7-1 , with reduced transcript level of a CRK carried an insertion in the 5’ untranslated region ( UTR ) . The status of the crk1-1 , crk1-2 , crk32 , crk33 , and crk46 mutants is unclear ( Figs 1B and S3 and S1 Table ) since transcripts for those four CRKs ( CRK1 , CRK32 , CRK33 , CRK46 ) were not detectable in the Col-0 wild type . In three crk lines where insertions were located in the 5’ UTR or the upstream promoter region , the corresponding CRK transcript abundance was increased compared to wild type . In eight lines no differences in corresponding CRK transcript levels were detected between the T-DNA insertion line and wild type ( Figs 1B and S4 and S1 Table ) . Of those lines , two carried the insertions in the exonic region , thus possibly resulting in a truncated protein; one line carried the insertion in an intron , while the remaining five lines carried insertions in the regions upstream of the start codon ( S3 Fig ) . The eleven lines in which corresponding CRK transcript levels were increased or not altered compared to Col-0 wild type were excluded from subsequent analyses ( Fig 2 ) . Thus a total of 39 lines were used ( Fig 2 ) . Based on quantitative PCR ( qPCR ) analysis 27 of the T-DNA insertion lines contained a single insertion , twelve contained two insertions while the rest contained more than two insertions in their genomes ( Fig 1B and S1 Table ) . An age-matched seed collection was generated and used for all subsequent phenotypic analyses where we investigated the role of CRKs in aspects of growth/development , abiotic stress responses , biotic stress responses , photosynthesis and stomatal regulation ( Fig 2 and S2 Table ) . Most crk mutants displayed normal morphology similar to the Col-0 wild type ( S5 Fig ) . By contrast , crk2 displayed a clear dwarf phenotype ( Fig 3A ) . Complementation of the crk2 mutation restored morphology similar to the Col-0 wild type ( Fig 3A ) . The crk5 mutant was slightly smaller compared to Col-0 wild type , in particular after five weeks of growth [28] , and overexpression of CRK5 in the mutant background led to slightly larger rosettes . Under long day conditions ( 16h light/8h darkness ) 34 crk mutants displayed early senescence ( Figs 3B and S6A ) . Under the same conditions , four crk mutants flowered earlier than Col-0 wild type ( Figs 3C and S6B ) . The only mutant that flowered later compared to Col-0 wild type was the dwarf crk2 ( Fig 3C ) . Germination was delayed in 32 crk mutants as shown by analysis of endosperm rupture ( Fig 3D and S3 Table ) . Pavement cell density was reduced in eleven crks compared to Col-0 wild type ( Figs 3E and S6C ) . Germination and pavement cell density were not altered in the dwarf crk2 compared to Col-0 wild type ( Fig 3D and 3E ) . In addition , crk28 , crk29 , and crk42 showed slightly longer roots compared to Col-0 wild type seedlings ( Fig 3F ) . In summary , our results suggest that the CRKs are not exclusively involved in stress and pathogen responses , as previously suggested , but in addition contribute to the regulation of specific developmental processes . The only mutant displaying a clear dwarf phenotype was crk2 which supports the earlier predictions that CRK2 might be involved in growth regulation in response to ABA [26] . Elevated production of ROS is part of the response to many abiotic stresses [20] . CRK transcript levels were strongly regulated in response to abiotic stresses ( S7 Fig ) [7 , 19] including ozone ( O3 ) and ultraviolet radiation ( UV ) . O3-induced ROS formation in the extracellular space rapidly induced transcript accumulation for several CRKs , while high light-induced ROS formation in the chloroplasts showed no effect [7] . In line with this , high light stress did not induce extensive damage in the crk mutants ( S8 Fig ) . Only crk2 and crk45 showed significant light stress-induced electrolyte leakage that differed from the response in wild type plants after nine or 24 hours , respectively . Elevated ROS production in chloroplasts can also be induced by methyl viologen ( MV , also known as Paraquat ) , which leads to increased superoxide production in the reducing end of the PSII . Moreover , chloroplastic ROS formation can also be induced by 3- ( 3 , 4-dichlorophenyl ) -1 , 1-dimethylurea ( DCMU ) , which causes increased production of singlet oxygen by affecting the redox status of the plastoquinone pool . The use of MV or DCMU allowed the assessment of the crk responses to increased chloroplastic ROS production by measurement of photosynthetic energy transfer , which is a more sensitive measurement compared to electrolyte leakage . Treatment of plants with MV or DCMU resulted in stronger photoinhibition in several lines ( e . g . crk2 , crk5 , crk8 , crk17 , crk20 , crk40 , crk45 ) compared to wild type , depending on the ROS inducer that was used ( Figs 4A–4C and S9 ) . Photosynthetic impairment of the crk5 mutant was rescued by overexpression of CRK5 [28] . In summary , these results suggest that some CRKs are involved in sensing or adaptation to changes in ROS or redox balance in the chloroplast and could be involved in signal transduction processes that culminate in regulation of photosynthetic electron transport . Like most RLKs , CRKs are predicted to localize to the plasma membrane and no CRKs have been identified in studies exploring the chloroplast proteome [30 , 31] . Thus , at the moment it is not clear how the CRKs communicate with the chloroplast but they might participate in communication between apoplast and chloroplast similarly to what has been described during PAMP-triggered immunity [32] . Even though chloroplasts and peroxisomes are the main sources of intracellular ROS in plants , extracellular ROS production is also important in the response abiotic stimuli [20] . Extracellular ROS production can be specifically induced by exposure to the air pollutant O3 or infiltration with the enzymatic system Xanthine-Xanthine Oxidase ( X+XO ) [33] . Several crk mutants displayed differential responses to extracellular ROS in comparison to Col-0 wild type plants . Responses to O3 were generally subtle ( Figs 4D and S10 ) . At 9 . 5 hours after the onset of O3 treatment crk2 and crk31 showed increased electrolyte leakage while at later timepoints crk3 , crk13 , crk20 , crk25 , crk28 , crk41 , and crk42 , showed elevated electrolyte leakage ( Fig 4D ) . Enhanced O3-induced cell death , visible as lesion formation , was observed for crk2 , crk3 , crk5 , crk7-1 , crk7-2 , crk11 , crk13 , crk19-2 , crk20 , crk22 , crk23-2 , crk24 , crk28 , crk31 , crk37 , crk38 , and crk46 ( S11 Fig ) . In response to treatment with X+XO , crk19-2 showed increased electrolyte leakage while crk4 , crk11 , crk21-1 , crk23-1 , crk23-2 , crk29 , crk37 , crk38 and crk46 showed reduced electrolyte leakage compared to Col-0 wild type ( Figs 4E and S12 ) . In response to UV-A and-B , crk2 , crk5 , crk40 , and crk42 displayed significantly elevated electrolyte leakage compared to Col-0 wild type indicating more damage similar to lesion simulating disease 1 ( lsd1 ) , which was used as positive control ( Figs 4F and S13A ) . Complementation of crk5 rescued the hypersensitivity to UV-A and–B radiation [26] . Exposure to salt ( NaCl ) is another environmentally relevant abiotic stress . Overall , twenty one crk mutants , including crk2 , crk5 , crk8 , crk11 , crk28 crk29 crk37 , and crk45 , showed delayed germination compared to the Col-0 wild type on medium containing 120 mM NaCl ( Figs 4G and S13B and S1C ) . After six days of growth on medium supplemented with 120 mM NaCl 13 crk mutants still maintained the delayed germination ( S13C Fig ) . While previous studies have emphasized the roles of CRKs in the response to pathogens and cell death regulation [13 , 14 , 21–23] , our observations suggest that CRKs are also important regulators of the response to abiotic stresses , such as UV-A and–B , salt , and O3 , possibly through extracellular ROS . In addition , our results suggest that the CRKs could be involved in controlling processes that indirectly regulate photosynthetic electron transport in the chloroplast . Stomata are key structures in the control of plant responses to drought stress , pathogen infection , and other stimuli . Control of the stomatal aperture is a complex process involving plant hormones , most prominently ABA , ROS , and calcium ( Ca2+ ) signalling to mediate and integrate plant-derived and environmental signals [34] . Transcriptional analysis suggested that several CRKs are involved in the control of drought responses ( S7 Fig ) [18] . Most CRKs displayed lower transcript abundance in guard cells compared to total leaf based on microarray meta-analysis ( S14A Fig ) while according to qPCR several CRKs displayed higher transcript levels in guard cells ( S14B Fig ) . The difference between different methods to analyze gene expression could be due to the different preparation methods for guard cells . For the qPCR experiments enzymatic digestion was used for the isolation of guard cell protoplasts which might cause an additional pathogen treatment . As several CRKs have been described to show elevated transcript levels in response to pathogen or elicitor treatments , this could likely be the source of this difference . Water loss from detached leaves or rosettes can be used as a measure of initial stomatal aperture and the rate of stomatal closure . Water loss was enhanced in crk2 , crk5 , and crk31 , as indicated by rapid decrease of rosette weight due to impaired stomatal regulation ( Fig 5A and S4 Table ) . Complementation of the crk2 and crk5 mutations rescued water loss phenotypes of these mutants ( Fig 5B and 5C ) . Several other crks , notably crk45 , lost less water after detachment compared to Col-0 wild type ( Fig 5A ) . Complementation of the crk45 mutation using an overexpression construct ( see materials and methods ) rescued the phenotype and led to increased water loss compared to the mutant and Col-0 wild type ( Fig 5D ) . Stomatal openness ( measured as the ratio of width to length ) in response to the plant hormone ABA was altered in several crk mutants ( Figs 5E and S15A ) , suggesting that CRKs may also participate in ABA-dependent control of the stomatal aperture . The crk22 , crk24 , crk37 , and crk46 mutants showed stronger ABA-induced stomatal closure compared to Col-0 wild type ( Figs 5E and S15A ) . Similar to other mutants [35] , we found a negative correlation between stomatal length and density for crks ( Figs 5F and S15B ) . Furthermore , the analysis showed a cluster of crks that had reduced stomatal density and increased stomatal length as compared to Col-0 ( S15B Fig ) . To compare microscopic measurements of the stomatal aperture with stomatal function we measured stomatal conductance in gas exchange experiments using intact , soil-grown rosettes [36] . Thirteen crks displayed slightly increased basal steady-state stomatal conductance under control conditions ( S16A Fig ) . Under conditions that induce rapid stomatal closure [36] ( elevated CO2 , darkness , pulse of O3 ) the rapid decrease in stomatal conductance was less pronounced in crk5 and crk31 compared to wild type plants ( Figs 5G–5I , S16B–S16D , S17 , S18 , S19 and S20 ) . These two mutants also exhibited increased water loss ( Fig 5A ) . Complementation of the crk5 mutant restored wild type-like stomatal responses in the mutant in response to O3 , darkness and CO2 ( Fig 5J and 5K ) . Some crk mutants showed slightly increased stomatal closure compared to Col-0 wild type in response to the stimuli tested ( S16 , S17 , S18 , S19 and S20 Figs ) . In response to CO2 , stomatal closure of crk31 was also somewhat ( but not statistically significantly ) reduced ( Figs 5I and S19 ) . In addition to the major regulators , many additional CRKs may be involved in stomatal closure but are compensated by the redundancy within the gene family . This is suggested by the fact that the overall responses to O3 , CO2 , and darkness correlate significantly even when including the non-significant responses ( Fig 6A–6C ) . This could result from compensation that is not perfect , as would be expected with genes which have slightly different structures ( Fig 6A–6C ) . CRKs have not been implicated in the regulation of stomatal openness and closure previously . Our findings suggest that specific CRKs are involved in controlling basal stomatal aperture and stomatal responses to environmental stimuli , which are critical to plant survival . In addition , CRKs also participate in the regulation of stomatal numbers . Apoplastic ROS production also plays an important role in pathogen defence [37] . It can be triggered by treatments with pathogen- , microbe- , or damage-associated molecular patterns ( PAMPs; MAMPs; DAMPs , respectively ) , for example flg22 , a peptide derived from flagellin , an integral component of the bacterial flagellum [38 , 39] . Basal ROS production was slightly reduced in thirteen and elevated in two crk lines ( S21 Fig ) . In response to elicitation with flg22 eleven crk lines displayed significantly increased ROS production while ROS production was decreased in crk2 , crk3 , crk13 , and crk31 ( Figs 7A and S22A ) . This suggests that CRKs might be involved in control of ROS production , though the mechanism is not clear . Several crk lines were more susceptible than the wild type to surface infection with the hemi-biotrophic bacterial pathogen Pseudomonas syringae pv . tomato DC3000 ( Pto DC3000; Figs 7B and S22B ) . Notably , elevated flg22-induced ROS production did not fully match the responses to Pto DC3000 infection ( Figs 7A and 7B and S22 ) . The crk5 and crk28 mutants exhibited normal ROS production ( Fig 7A ) but showed increased disease symptoms ( Fig 7B ) , while crk23 showed elevated flg22-induced ROS production but did not differ from Col-0 wild type in its susceptibility to Pto DC3000 . The crk20 and crk29 mutants were also more susceptible to infection by Pto DC3000 in spite of the elevated ROS production . The different level of ROS production in the crk mutants compared to Col-0 wild type after elicitation with flg22 suggests that at least some CRKs might act through ROS signalling pathways rather than direct pathogen perception . To test this hypothesis , we measured PAMP-induced stomatal closure in Col-0 wild type and crk plants . PAMP perception through RLKs leads to NADPH oxidase activation and extracellular ROS production and induces stomatal closure [40–42] . While ROS production was normal or even elevated in most crks , stomatal closure triggered by flg22 was impaired in several mutants including crk5 , crk17 , crk20 , and crk28 ( Figs 7C and S23A ) and corresponds to their increased susceptibility to Pto DC3000 ( Figs 7B and S22B ) . In response to the PAMP chitin , a fungal cell wall component , stomatal closure was reduced in several crk mutants including crk2 , crk6 , crk10-2 , crk10-4 , crk12 , and crk19-2 ( Figs 7D and S23B ) . Significant ABA- , flg22- , and chitin-induced stomatal responses were mostly mediated by different CRKs ( Fig 8A and 8B ) ; only a few CRKs participated in more than one process . Responses to the PAMPs flg22 and chitin might be mediated mostly by different CRKs ( Fig 8C ) . Again , in addition to the major regulators , the responses overall show a significant correlation even when including also the non-significant crk mutants ( Fig 8A–8C ) . This may be due to the redundancy within the gene family , similar to the result observed in stomatal conductance ( Fig 6A–6C ) . Most crk mutants that were affected in stomatal immunity displayed higher transcript abundance in guard cell protoplasts compared to whole , untreated leaves ( S14B Fig ) . While Pto DC3000 can infect plant leaves through stomata [43–45] , powdery mildews follow a different strategy by penetrating and colonizing epidermal cells [46–48] . Infection of wild type Col-0 with the biotrophic virulent powdery mildew fungus , Golovinomyces orontii ( Go ) , or the non-host powdery mildew fungus Blumeria graminis f . sp . hordei ( Bgh ) resulted in altered expression of several CRK genes ( S24A Fig ) and susceptibility towards the pathogens was also affected in crks . Several crks showed increased susceptibility to Go , whereas responses to Bgh were more subtle . Specifically , crk2 and crk5 displayed less visible mildew symptoms in response to Go infection ( Fig 9A and 9B ) , whereas crk17 , crk20 , crk23-1 , crk23-2 , crk25 , crk28 , crk32 , and crk38 were more susceptible ( Figs 9A and 9B and S24B and S24C ) . Furthermore , crk20 and both alleles of crk23 ( crk23-1 and crk23-2 ) showed consistently enhanced pigmentation of the leaves following Go infection ( Fig 9B ) . Additionally , crk1-1 , crk17 , crk25 , and crk32 showed this response also to the non-pathogenic Bgh ( Fig 9C ) . While previous reports have already suggested that CRKs participate in defence against bacterial pathogens , our results show that CRKs are also involved in the defence against fungal pathogens . CRKs participate in the control of pathogen-induced ROS signalling , stomatal responses and pre-invasive immunity .
This study addresses the physiological and cellular roles of the CRK protein family—with 44 members , one of the largest subgroups of RLKs in Arabidopsis ( Fig 2 ) . As a result of the large-scale phenotyping of a crk T-DNA insertion mutant collection we were able to identify clear and specific phenotypes for several crk mutants . Clustering of all phenotypic differences of crk mutants compared to Col-0 wild type allowed the generation of a genetic and phenotypic framework ( Figs 10 and S25 ) . Previous reports have linked ectopic overexpression of individual CRKs with pathogen defence and regulation of cell death [12 , 13 , 20] and CRK45 , one of the few CRKs lacking ecto- and transmembrane domains , was found to interact with pathogen effectors in a large-scale screen [49] . However , it is unclear whether other CRKs could be direct targets for pathogen effectors . Meta-analysis of microarray data ( S7 Fig ) suggested the involvement of CRKs in response to a variety of additional stimuli beyond pathogen defence , including O3 , UV , light , salt , and drought stress . Several crks displayed phenotypes in response to those stimuli ( Figs 10 and S25 ) . In addition , growth and development were altered in several crks ( Figs 10 and S25 ) . Roles for ROS/redox signalling in biotic or abiotic stress response have been shown repeatedly over the last decade [20 , 50] . Similarly , plant development [51] , root growth [52 , 53] , senescence [54] , germination [55] , cell expansion [56] , flowering [57–59] , and cell cycle control [60] are tightly integrated with ROS and redox-dependent processes . This might suggest that CRKs could be connected to ROS/redox signalling in both stress and developmental processes . Three CRKs ( CRK27 , CRK34 , and CRK44 ) might fulfil critical roles for plant survival . Their expression in different plant tissues and organs did not reveal any striking patterns ( S5 Table ) and no assumption towards their function . In the Col-0 genetic background no T-DNA insertion lines were obtained for CRK27 , CRK34 , and CRK44 . Their roles will however need to be verified in the future . Overall , some of the most striking phenotypes were found for crk2 and crk5 , members of the basal group I and group V , respectively ( Fig 10 ) . Some phenotypes of crk2 might be caused by its dwarf morphology . However , this , together with the underlying cause of the dwarfism in crk2 , will require more detailed analysis in the future . Earlier studies have suggested that the high degree of amino acid sequence similarity between CRK family members would be the main reason for redundancy and lack of loss-of-function mutant phenotypes [13 , 14 , 21 , 22] . Our findings suggest that CRKs do not function in an exclusively redundant fashion . Specific CRKs , for example CRK2 of the basal phylogenetic group I , could function as primary regulators while others might provide calibration for more fine-tuned responses . This would offer an explanation for the intriguingly large number of CRKs in Arabidopsis thaliana , and also in other species . Tissue and cell specificity of various CRKs might also be a reason for the large number of different genes , as they would be under different transcriptional regulation . However , according to eFP browser ( http://bar . utoronto . ca/efp/cgi-bin/efpWeb . cgi ) [61] most CRKs seem to be present in low levels in most tissues ( S5 Table ) . Only CRK2 and CRK3 show strikingly higher transcript abundance in guard cells , hypocotyl and in vascular tissues ( S5 Table ) . Expression of CRKs might however also be regulated in response to external stimuli . This is the case for CRK7 [27] and CRK5 [28] . Regulation of the stomatal aperture is an important factor in the response to a wide range of stimuli [45 , 62 , 63] and has been shown to involve ROS signalling [64] . Several crks showed differences in ABA- or stress-induced stomatal closure ( Fig 11A ) . Thus , it is tempting to link the crk phenotypes defective in ABA signalling with altered ROS signalling . Two crks ( crk5 and crk31 ) showed defective responses to stomatal closure induced by abiotic factors . Expression of CRK5 is not restricted to guard cells ( S14B Fig ) , underlining the importance of whole-leaf processes mediated by CRK5 in stomatal movements . CRKs might also be involved in determining basal openness of stomata . Interestingly , different CRKs control basal steady-state stomatal openness and stomatal responsiveness to stimuli . Stomatal closure can also be triggered by application of PAMPs , e . g . flg22 or chitin . Signal transduction from receptor-mediated PAMP perception by intracellular signalling and ROS production down to stomatal closure is becoming increasingly better understood [45 , 65] . There are convergence points in the signalling pathways of ABA- and PAMP-induced stomatal closure , for example ROS production by the NADPH oxidase RBOHD [34 , 41 , 66] . Consistent with enhanced disease symptoms upon Pto DC3000 infection of several crks , mutations in several CRKs impaired flg22-triggered stomatal immunity ( Fig 11A ) . The results suggest that PAMP perception and the earliest signalling events upstream of ROS production are likely not affected in crk mutants . Chitin-induced stomatal closure was also compromised in crks , but different crks were affected in chitin-induced stomatal closure compared to flg22 ( Fig 11A ) suggesting that CRKs could provide signalling specificity . Even though powdery mildew fungi do not use stomata as an infection route but directly penetrate epidermal cells , several crks impaired in chitin-induced stomatal immunity also displayed enhanced susceptibility to Go . This indicates that CRKs are also signalling components in guard cell function-independent plant defence responses . Interestingly , crk mutants that were impaired in stomatal immunity were not altered in ABA-induced stomatal closure ( Fig 11A ) . Together , this indicates that many CRKs might fulfil independent functions in PAMP- or ABA-triggered processes in guard cells ( Fig 11A ) , while a few CRKs might control common , and presumably basal , aspects of guard cell function ( Fig 11B ) . While CRKs in Arabidopsis thaliana are involved in immune signalling evidence from legumes suggests that CRKs might also participate in the control of symbiosis or even in distinguishing between pathogenic or beneficial microbes [22] . This suggests that there might be additional functions for CRKs which cannot be addressed in Arabidopsis . The most prominent feature of the CRK protein family is the presence of two cysteine-rich DUF26 domains ( with C-X8-C-X2-C-motifs ) in the extracellular region . Structural analysis of the DUF26 domain of ginkbilobin-2 ( Gnk2 ) , a Ginkgo biloba protein containing a single DUF26 domain with a proposed function as an antimicrobial protein , suggests that the cysteines form disulphide bonds [67 , 68] . The role of the CRK ectodomain is still unknown . It could either bind a ligand ( peptide or other ) or be crucial for the formation of complexes with other receptors . Recently , it has been suggested that the DUF26 domain in Gnk2 might be involved in mannose binding [69] . The residues required for mannose binding in Gnk2 are , however , not conserved in the DUF26 domains of the CRKs . It has also been suggested that the cysteines in the DUF26 domain could be a target for redox modification which might lead to a conformational change , for example through opening of disulphide bridges . However , redox regulation of CRK ectodomain structure and ligand binding might not be mutually exclusive . A connection between CRK function and redox or ROS-related processes is also suggested by the strict and specific regulation of genes encoding CRKs under ROS-producing conditions [7 , 19] . Through this CRKs might participate in feedback regulation of ROS production where they might sense extracellular ROS and be part of a “ROS amplification loop” . This would place the CRKs in the “ROS wave” [70] by perceiving ROS from neighbouring cells and transducing the signal into the cytosol , subsequently regulating NADPH oxidase activity and signal propagation ( Fig 11B ) . This is particularly interesting since the precise control and adjustment of ROS production in response to different stimuli is still unresolved even though activation and regulation of NADPH oxidases and other ROS-producing enzymes through protein phosphorylation and RLKs is becoming better understood . Proteins with DUF26 domains are restricted to plants but other cysteine-rich domains could fulfil analogous functions to CRKs with respect to sensing of extracellular ROS in other organisms . The data shown here suggests that one of the main functions of CRKs could be to provide signalling specificity downstream of extracellular ROS production . However , it is unclear how this regulation might work exactly . Recent evidence suggests that CRKs might be able to interact with pattern recognition receptors [22] but the specificity and the precise role of this interaction will require further investigation . It suggests however , that CRKs might act in concert with other receptors and RLKs , possibly also during the regulation of plant development and during abiotic stress responses . From the phenotypic framework ( Figs 2 and 10 ) it will now be possible to dissect the molecular mechanisms through which the CRKs function . Conceptually , perception of cell-to-cell or environmental signals through CRKs could follow different modes of action as positive or negative regulators . However , it is likely that CRKs have multiple rather than single downstream targets . How CRK signalling is integrated in synergistic or antagonistic fashion might be highly process specific . The genetic and phenotypic framework and the proposed models for modes of CRK action will allow targeted and detailed mechanistic analysis of CRK function in the future . Ultimately , this will allow improvement of plant growth and tolerance to complex environmental challenges . Our results demonstrate that discovery of subtle phenotypic responses and aspects , which might otherwise be missed , can be facilitated with thorough phenotypic analysis of comprehensive mutant collections for large gene families , instead of studying individual family members .
All T-DNA insertion crk lines were obtained from the Nottingham Arabidopsis Stock Centre ( NASC , http://nasc . life . nott . ac . uk/ ) and were confirmed by PCR ( primers are listed in S1 Table ) . An age-matched seed collection was generated and used for all experiments . The seed collection has been donated to the European Arabidopsis Stock Centre in Nottingham ( http://www . arabidopsis . info ) and can be obtained from there . Siliques were harvested at maturity and dried at room temperature for 10 days prior to the collection of seeds . Freshly harvested seeds were after-ripened for 3 months at 20°C ( approximately 30% relative humidity ) in darkness and further used in germination tests . After stratification for 3 days at 4°C , seeds were grown in a mixture of soil and perlite ( 3:1 ) or on Jiffy Peat Pellets in the growing room under the following conditions: 8/16 h photoperiod , temperature 22/18°C ( day/night , respectively ) , relative humidity of 70 ± 5% , and PAR ( 100–150 μmol m-2 s-1 ) . Experiments were performed on 4 week-old plants , unless otherwise stated . For pathogen assays and stomatal analysis Arabidopsis thaliana plants were grown on general soil ( Arabidopsis mix , John Innes Centre , Norwich ) , or for infection assays on Jiffy pellets ( Jiffy Products , Norway ) under 10 h or 16 h of light at 20–22°C and 65% humidity . Mutant fls2 lines have been described previously [71] . O3 exposure and high-light treatments started at 9 am and were continued for 6 h . 18-day-old plants were used for O3 experiments . Leaf rosettes were harvested 7 h after the start of the treatment then washed with ultra-pure water and transferred into 15 ml ultra-pure water for electrolyte leakage measurements ( n = 4 ) . Experiments were repeated twice . Ozone ( 350 ppb ) and high-light ( 1430 μmol m-2s-1 photosynthetically available radiation [PAR; 400–700 nm] , 11 mW m-2 UV-B radiation [280–315 nm] , 25 . 4 W m-2 UV-A radiation [315–400 nm] ) treatments were performed at the Research Unit Environmental Simulation of the Helmholtz Zentrum München ( Germany ) in the walk-in-size chambers and in a small sun simulator respectively . Spectral measurements were performed using a double monochromator system TDM300 ( Bentham , Reading , England ) . Arabidopsis plants were cultivated on multiplication substrate ( Floradur ) mixed with quartz sand ( Dorfner ) in the respective ratio 5:1 . After a 2-day pretreatment at 4°C , pots were cultivated under the following conditions: 250 μmol m-2 s-1 PAR under the exclusion of UV radiation ( <400 nm ) , under 12h-day length ( day: 23°C , 70% relative humidity; night: 18°C , 90% relative humidity ) . Extracellular superoxide was generated by vacuum infiltration of 1 mM xanthine ( X ) and 0 . 1 U ml-1 xanthine oxidase ( XO , Sigma-Aldrich ) into the leaf discs from 4-week-old plants as previously described [33 , 72] . Cell death was monitored by electrolyte leakage measurements with a conductivity meter ( Mettler Toledo ) at the indicated times after the end of the treatment . High-throughput confocal imaging was performed using the Opera microscope ( PerkinElmer , Germany ) as published [73] . For quantification of cell numbers , cotyledons were stained with propidium iodide according to Lucas et al . [74] and measured using PDQUANT as described previously [73] . Bacterial inoculation assays were performed as described previously [75] . Briefly , Pto DC3000 was sprayed onto leaf surface at 108 CFU ml-1 and disease symptoms were scored 3 days post inoculation . For mildew infection , plants were grown under 8/16 h photoperiod ( 200 μmol m-2 s-1 ) at 22°C ± 1°C . Three to four week old plants were inoculated in a settling tower with about 1 spore per mm2 of a virulent Golovinomyces orontii ( Go ) powdery mildew isolate or 10 spores per mm2 of a non-host Blumeria graminis f . sp . hordei ( Bgh ) powdery mildew isolate . Symptoms and mildew coverage was assessed after 6 days . Mildew coverage in percent per plant was scored from digital images using the image processing software ImageJ ( http://imagej . nih . gov/ij/ ) . The experiment was conducted three times with 5 replicate plants . Plants for qPCR were grown as above , samples were taken at 6 , 16 , and 24 h after inoculation with Go or Bgh . Five plants were pooled into one sample and the experiment was conducted three times . RNA extraction and cDNA synthesis was as described earlier [46] ( primer sequences for CRK transcripts [7] ) . Relative CRK gene expression was analyzed by the comparative CT method . CT values were normalized to 18S rRNA and expression between uninfected control and Go/Bgh powdery mildew infected plants were compared using the 2-ΔΔCt method . Significant differences were determined according to student’s t test . Guard cells were isolated as described [76] and cDNA was generated from RNA isolated from guard cell protoplasts . Expression of CRKs was compared with cDNA from total RNA isolated from untreated Arabidopsis leaves . Actin-2 ( At3g18780 ) , YLS8 ( At5g08290 ) , PP2A ( At1g13320 ) , TIP41 ( At4g34270 ) , and At4g35510 were used as normalization genes for the analysis using the 2-ΔΔCt method . ROS assays were performed as described previously [41] . Briefly , 16 leaf discs were excised per genotype of four weeks-old plants and treated with 1 μM flg22 . ROS was measured with a Varioskan multiplate reader ( Thermo Fisher Scientific , USA ) for 35 min . For ultraviolet light source , UVC 500 Crosslinker ( Hoefer Pharmacia Biotech , San Francisco , CA , USA ) equipped with three UV-B lamps ( type G8T5E , Sankyo Denki , peak wavelength 306 nm ) and two UV-A lamps ( type TL8WBLB , Philips , peak wavelength 365 nm ) were used . Plants were exposed to single radiation episode until a cumulative dose of 1500 mJ cm-2 was reached ( roughly 10 minutes ) . After 4 days leaves were excised , fresh weight measured ( g ) and transferred into 50 ml falcon tubes containing 35 ml MilliQ water . The relative electrolyte leakage was measured with a conductance meter ( WTW , INOLAB Cond Level 1 ) and calculated as a ratio between the value obtained after 1 h incubation and the total electrolyte leakage evaluated after autoclaving the samples . The O-J-I-P test was performed as described [77] using FluorCam and the associated software ( Photon System Instruments , Czech Republic ) . Plants were dark-adapted for 30 min prior to measurement . The maximum quantum yield of primary photochemistry ( φPo ) , the size of the plastoquinone pool ( qPQ ) and the total dissipation of untrapped excitation energy from photosystem II ( PSII ) reaction center ( DLo/RC ) were calculated . φPo represents the probability that an absorbed photon is trapped by the reaction center and used for primary photochemistry , reducing QA to QA- . Analysis of the fluorescence transients was made on whole rosettes and included two sets of plants: one set sprayed with 32 μM MV and kept in 8/16 photoperiod for two days and a second set sprayed with 20 μM DCMU ( dissolved in 75% ethanol ) and kept in darkness for two hours . To assess germination of crk mutants , endosperm rupture assays were performed by placing after-ripened seeds in 9 cm Petri dishes ( 30 seeds per dish , three independent biological replicates ) on 1% agar with addition of 0 . 01% PPM ( Plant Preservative Mixture , Plant Cell Technology , USA ) . All assays were performed at 20°C under PAR of 200 μmol m-2 s-1 . Testa and endosperm rupture were assessed every 5 hours up to 51 hours of imbibition . A seed was considered as germinated when the radicle protruded through both envelopes . Seeds were sterilized , germinated and grown as described [78] . Root length measurements were performed 8 days after stratification in two independent assays with 6 plants each . Interesting lines were additionally screened twice . To measure bolting , flowering and senescence germinated seeds were transferred to soil and grown with 16 h-light/8 h-dark photoperiods at 22°C . Plants were considered bolting at the first appearance of the inflorescence , flowering at the opening of the flower petals and senescing at the first yellowing of the rosette leaves . Salt treatment assays were performed by placing seeds in Petri dishes ( 15 seeds per lines , three independent biological replicates ) on MS medium containing 1% sucrose , 1% agar , buffered to pH 5 . 7 with 2 . 8 mM MES , with or without 120 mM NaCl; under 12h-day length ( 100–150 μmol m-2 s-1 , day: 22°C—night: 19°C , 70% relative humidity ) . A seed was considered as germinated when the 2 cotyledons were visible . For each line , germination rate was expressed as the ratio ( germination percentage on NaCl plates /germination percentage on control plates ) , at 5 and 6 days after stratification . Plants were grown as described in the plant materials section . Weight of detached whole rosettes was followed until 4 h at room temperature ( five 3-week old plants per lines , for three independent biological replicates ) . To analyse steady-state stomatal conductance and stomatal responses to darkness , elevated CO2 and O3 , 21–26 days old plants and a custom made gas exchange device were used . As crk2 had reduced growth rate , older plants ( 26 to 32 days old ) were analysed . The device and plant growth conditions have been described previously [36 , 79] . First , plants were inserted into the device and kept at 150 μmol m-2 s-1 light , 65% air humidity and ambient CO2 concentration ( 400 ppm ) until stomatal conductance had stabilized . To address stomatal response to darkness , light was switched off , to address elevated CO2-induced stomatal closure , CO2 concentration was increased to 800 ppm and in order to address stomatal closure induced by O3 , a three min pulse of 500–600 ppb of O3 was applied . In all experiments , stomatal conductance was followed for 32 min from application of closure-inducing stimuli . CRK2 and CRK45 coding regions were cloned into pDONRzeo ( Invitrogen ) via Gateway site-specific recombination . Coding regions were then assembled together with CaMV 35S-promoter and Venus YFP [80] C-terminal tag into the pBm43GW [81] expression vector , using MultiSite Gateway technology ( Invitrogen ) , creating a translational fusion protein . Primers are listed in S1 Table . Constructs were transformed into the corresponding T-DNA insertion plants by GV3101 Agrobacterium-mediated floral dipping [82] . For selection of successful transformants , seeds were plated on ½ MS media supplemented with 1% sucrose and 20 μg/mL Basta . Plants were grown for one week in vitro before being transferred to soil . T1 plants were used for experiments with each plant constituting an individual insertion event . CRK5 complementation/over-expression lines have been described in Burdiak et al . [28] . Two individual homozygous T3 lines with a single insert were used . All data analysis was carried out using R , version 2 . 15 . 1 . All experiments were made commeasurable by computing the Z score statistic of the comparison of each of the alleles versus the Col-0 wild type reference using appropriate statistical test for the given data set . Sample sizes across different experiments were normalised by bootstrap sampling such that in each bootstrap data set the number of samples for a given condition was set to n = 15 . The Z score for each experiment was computed as the mean of the Z score estimates from bootstrap data sets . In cases where the null distribution of the statistical test did not follow Z distribution ( e . g . with Mann-Whitney test ) , it was approximated with the Z-distribution . To reduce the number of false positives produced by the analysis , the Z-scores in each experiment were transformed to P-values and corrected using Benjamini-Hochberg false discovery rate adjustment . The adjusted P-values were then transformed back to obtain adjusted Z-scores . Adjusted Z-score values >2 or < -2 ( corresponding to a false discovery rate < 5% ) were considered to be statistically significant . In experiments involving a time course and many measurements , each time point was analysed separately to have statistical power comparable to non-time course experiments . The list of specific statistical models and comparisons that were used to get Z scores and subsequently used for construction of heat maps are displayed in S6 Table . Unless stated otherwise , the Z score statistic from the statistical model was obtained comparing each crk genotype to Col-0 wild type . Comparisons applying linear models were carried out with multcomp R package ( http://cran . r-project . org/web/packages/multcomp/index . html ) . Heatmap of adjusted Z scores was constructed using hierarchical clustering with Ward’s method and applying Euclidean distance metric . Phylogenetic trees were estimated using MEGA6 [83] . Alignments were carried out using Muscle [84 , 85] . Trees for the entire coding region were estimated using all positions while trees for the kinase domains and extracellular regions were estimated using complete deletion for gaps . The initial guide tree was estimated using maximum parsimony . 1000 bootstrap replicates were used for all trees . Statistical significances were estimated by constructing separate linear mixed model for each of the crk genotypes including Col-0 wild type as the reference data , modelling different time points as fixed effects . Comparisons were carried out using multcomp package for R ( http://cran . r-project . org/web/packages/multcomp/index . html ) and applying single step p-value adjustment for multiple comparisons . In light stress and fresh weight experiments , the model consisted of genotype , time and their interaction as fixed effects , and experiment replicate as random effect . Each crk genotype was compared to Col-0 in all time points . In O3 , X+XO , salt experiments , the linear mixed model consisted of genotype , time and their interaction as fixed effects , and experiment replicate as random effect . Each crk genotype was compared to Col-0 under treatment in all time points . In Go scoring , the leaf coverage percentage was modelled with genotype as fixed effect and experiment replicate as random effect . In germination assay analysis , the linear mixed model consisted of genotype , time and their interaction as fixed effects . In both experiments , each crk genotype was compared to Col-0 . Pre-processing of the gene expression data and the accession numbers in databases has been previously described [86] . In brief , the data was downloaded from NASCArrays ( http://affymetrix . arabidopsis . info/narrays/experimentbrowse . pl ) , ArrayExpress ( http://www . ebi . ac . uk/microarrayas/ae/ ) , Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/ ) , and The Integrated Microarray Database System ( http://ausubellab . mgh . harvard . edu/imds ) . Arrays were normalised with Robust Multi-array Average ( RMA ) [87] , and log2 ratio of the mean of treatment and control expressions across biological replicates was computed , resulting in 141 differential expression profiles . Bayesian Hierarchical Clustering of CRKs present on the arrays was carried out using R package BHC [88] using log2 fold change ±1 as discretization threshold . | Receptor-like kinases ( RLKs ) are important regulators in signal transduction in plants . However , the large number of RLKs and their high sequence similarity has hampered the analysis of RLKs . One of the largest subgroups of RLKs , the cysteine-rich receptor-like kinases ( CRKs ) , has been suggested to be involved in mediating the effects of reactive oxygen species ( ROS ) . While ROS are recognized as important signalling elements with a large variety of roles in plants , their ligands and achievement of signalling specificity remain unknown . Using insertion mutants we analysed the roles of CRKs in plant development and stress responses and show that CRKs have important roles as mediators of signalling specificity during regulation of stomatal aperture . Our study shows that , despite their large number and high sequence conservation , individual CRKs have intriguingly distinct functions in different aspects of plant life . This makes the CRKs promising candidates for future studies of their biochemical function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Large-Scale Phenomics Identifies Primary and Fine-Tuning Roles for CRKs in Responses Related to Oxidative Stress |
Although genome-wide association studies ( GWAS ) have identified hundreds of complex trait loci , the pathomechanisms of most remain elusive . Studying the genetics of risk factors predisposing to disease is an attractive approach to identify targets for functional studies . Intracranial aneurysms ( IA ) are rupture-prone pouches at cerebral artery branching sites . IA is a complex disease for which GWAS have identified five loci with strong association and a further 14 loci with suggestive association . To decipher potential underlying disease mechanisms , we tested whether there are IA loci that convey their effect through elevating blood pressure ( BP ) , a strong risk factor of IA . We performed a meta-analysis of four population-based Finnish cohorts ( nFIN = 11 266 ) not selected for IA , to assess the association of previously identified IA candidate loci ( n = 19 ) with BP . We defined systolic BP ( SBP ) , diastolic BP , mean arterial pressure , and pulse pressure as quantitative outcome variables . The most significant result was further tested for association in the ICBP-GWAS cohort of 200 000 individuals . We found that the suggestive IA locus at 5q23 . 2 in PRDM6 was significantly associated with SBP in individuals of European descent ( pFIN = 3 . 01E-05 , pICBP-GWAS = 0 . 0007 , pALL = 8 . 13E-07 ) . The risk allele of IA was associated with higher SBP . PRDM6 encodes a protein predominantly expressed in vascular smooth muscle cells . Our study connects a complex disease ( IA ) locus with a common risk factor for the disease ( SBP ) . We hypothesize that common variants in PRDM6 can contribute to altered vascular wall structure , hence increasing SBP and predisposing to IA . True positive associations often fail to reach genome-wide significance in GWAS . Our findings show that analysis of traditional risk factors as intermediate phenotypes is an effective tool for deciphering hidden heritability . Further , we demonstrate that common disease loci identified in a population isolate may bear wider significance .
Intracranial aneurysms ( IA ) are berry-shaped pouches at the branching sites of cerebral arteries . 2–5% of the world population is estimated to harbor IA [1] . Most IA go unnoticed during one's lifetime . However , when they become symptomatic , it is usually due to rupture , causing subarachnoid hemorrhage ( SAH ) . SAH is devastating intracranial bleeding , and half of those with SAH die within a year [2] , [3] . SAH affects the working age population , with a median age of 55 [4] . Its incidence in Finland is 19/100 000/year [5] , [6] , triple than that of the rest of the world . The reason for this higher than average incidence is unknown . Aneurysmal SAH places a heavy burden on society both emotionally and financially . The strongest known non-modifiable risk factor of SAH is family history of the disease , and the strongest modifiable risk factors are smoking , excessive alcohol intake , and hypertension [7] . An important step in tackling SAH is to understand why IAs develop . Our understanding of the environmental and genetic background of IA formation is limited . Positive family history of IA or SAH , older age and female sex increase the risk of developing IA [1] . Of the general cardiovascular risk factors , smoking has been shown to increase the risk of IA formation [8] , and high blood pressure has long been speculated to do so [9] . The high , often undocumented , prevalence of high blood pressure in the control populations is likely the reason why it frequently fails to reach statistical significance as an IA risk factor [1] . Chronic hypertension may contribute to IA formation by imposing constantly high shear stress on vascular walls [9] . Multiple factors , such as familial aggregation of the disease , make a genetic contribution likely to the risk of IA . A minority of IAs show familial aggregation ( under 10% ) [7] . Linkage studies in IA families have highlighted numerous genetic regions and a recent exome sequencing study identified coding mutations in familial thoracic aortic aneurysm with intracranial aneurysm [10] . However , the majority of IA is sporadic . Sporadic IA is a complex disease and no gene with a certain role has been identified yet . Recent genome-wide association studies ( GWAS ) [11] , [12] involving Finnish IA patients , have attempted to decipher the complex genetic background of IA . From these studies , five loci emerged with strong association to IA ( p<5E-07 , posterior probability of association –PPA>0 . 5 ) , with the highest statistical significance at 9p21 . 3 , a risk locus of multiple cardiovascular diseases . Further 14 loci exhibited suggestive association to IA ( 0 . 1≤PPA<0 . 5 ) . Despite the success of GWAS in identifying IA susceptibility loci , the pathomechanism by which they contribute to IA formation remains elusive . We hypothesize that hypertension , a strong modifiable risk factor of IA , may possess an overlapping genetic background with IA . To test this hypothesis , we analyzed the IA loci so far identified , in well-characterized population-based cohorts consisting of more than 210 000 individuals with blood pressure measurements .
41 SNPs from 19 independent IA loci [13] were first analyzed for association with blood pressure in the national Health 2000 survey ( H2000 ) [14] discovery cohort of 1581 individuals without blood pressure lowering medication ( Table S1 ) . We adjusted the analysis for age and gender ( ROBUST model ) . The most significant association ( p<0 . 1 ) were observed at 2q33 . 1 with diastolic blood pressure ( DBP ) and with mean arterial pressure ( MAP ) , at 4q31 . 23 and 19q13 . 12 with DBP , and at 5q23 . 2 with systolic blood pressure ( SBP ) , DBP , and MAP . We did not detect association with pulse pressure ( PP ) ( Table S1 ) . Next , we wanted to analyze the independence of the association signals observed . We tested all 19 loci SNPs adjusting for further factors known to affect blood pressure , namely smoking habits , alcohol consumption , and body mass index ( BMI ) ( ADVANCED model ) . There was no tendency of association with DBP at 4q31 . 23 and 19q13 . 12 with the ADVANCED model . The strength of the association decreased for the four SNPs at the 2q33 . 1 locus for SBP , but increased marginally for DBP , and MAP . At 5q23 . 2 the strength of association increased substantially for most blood pressure measurements , such as SBP , DBP , and MAP ( Table 1 and Table S2 ) for all three SNPs tested . The IA risk alleles at 5q23 . 2 were associated with elevated blood pressure . To confirm the initial association signals at the 2q33 . 1 and 5q23 . 2 loci observed in the H2000 discovery cohort , we tested them for association with blood pressure in three additional population-based cohorts from Finland . SNPs at 2q33 . 1 failed to show significant association with DBP and MAP in any of the replication cohorts ( the Cardiovascular Risk in Young Finns Study-YFS [15] , [16] , the Northern Finland Birth Cohort 1966-NFBC1966 [17] , and the Helsinki Birth Cohort Study-HBCS [18] ) . When the results were combined from all cohorts in a fixed effect meta-analysis , they remained non-significant ( Table 2 ) . At 5q23 . 2 SNPs showed significant association with SBP in YFS and NFBC1966 ( Table 1 ) . In HBCS , although consistent in the direction of the effect , the association remained suggestive . When the results were combined from all cohorts in a fixed effect meta-analysis , we detected significant association with SBP at 5q23 . 2 ( prs570682 = 4 . 80E-05 , prs2287696 = 6 . 81E-05 , prs335206 = 3 . 01E-05 ) ( Table 1 ) . Comparing the mean SBP of the study participants stratified for their 5q . 23 . 2 genotypes indicated a positive correlation between the number of risk alleles and higher SBP for all three SNPs tested ( Figure 1 ) . Study participants homozygous for the risk allele ( C , in the case of rs335206 ) , had on average 1 . 3 Hgmm higher SBP compared to those who were homozygous for the protective allele , and 0 . 9 Hgmm higher than those with the heterozygous genotype . This effect size is comparable to those of most blood pressure loci identified by The International Consortium for Blood Pressure Genome-wide Association Studies ( ICBP-GWAS ) consortium [19] . The observed linear effect of risk allele count is strongly suggestive of a true association . Association at 5q23 . 2 with DBP ( prs570682 = 0 . 02 , prs2287696 = 0 . 04 , prs335206 = 0 . 03 ) and MAP ( prs570682 = 0 . 0007 , prs2287696 = 0 . 0010 , prs335206 = 0 . 0004 ) showed a reduction of significance when results were combined from all cohorts ( Table S2 ) . To test whether the association at the 5q23 . 2 locus is unique to the Finnish cohorts , we attempted to replicate the association with the three SNPs in the multinational cohort ICBP-GWAS [19] . All three SNPs showed significant association with SBP ( prs570682 = 0 . 0065 , prs2287696 = 0 . 00079 , prs335206 = 0 . 0014 ) in the ICBP-GWAS cohort of 200 000 individuals of European descent . The risk allele for elevated SBP in the ICBP-GWAS cohort was the same as in our meta-analysis of four Finnish population-based cohorts . When the results from the four Finnish cohorts were combined with the ICBP-GWAS results in a fixed effect meta-analysis , the strength of the association increased with all three SNPs tested ( Table 3 ) . The strongest association was observed with rs2287696 ( pALL = 8 . 13E-07 ) . This suggests that the variant at 5q23 . 2 is a common risk factor present in multiple populations of European descent . Further loci or results for DBP or MAP were not tested for association in ICBP-GWAS , since they failed to show significant association in our replication cohorts . All three tested SNPs at 5q23 . 2 reside in intronic regions of the gene PR domain containing 6 ( short form: PRDM6 ) and showed comparable p-values . To further explore the associated region in an attempt to pinpoint the causative variant , we examined all 1000 Genomes variants around PRDM6 in the four Finnish cohorts ( Figure 2 ) . The strongest association was observed with rs163189 ( p = 6 . 12E-06 ) near rs570682 and rs2287696 , in the second intron , where the most significantly associated SNPs clustered . All five of the strongest associated SNPs are located within a 4 . 7 kb region at 122 . 4 MB ( Human genome build 36 ) , surrounding a Sterol regulatory element binding transcription factor 1 ( SREBP1 ) binding site ( Figure 2 ) [20] .
Hypertension , a leading cardiovascular risk factor , is a strong modifiable risk factor for IA and its deadly rupture . Our study establishes a genetic link between elevated SBP and IA formation . Further , we demonstrate the benefits of using population isolates for mapping complex disease loci valid in multiple populations . 5q23 . 2 was identified as a suggestive IA risk locus by Yasuno and colleagues [13] in a multinational GWAS including Finnish IA patients . The strength of the association at 5q23 . 2 in their study mainly came from the Finnish cohort ( Figure S1 ) . However , albeit weaker , association to IA at 5q23 . 2 was observable in all cohorts tested by Yasuno and colleagues . In the two tier approach we applied , the suggestive aneurysmal locus at 5q23 . 2 showed robust association to blood pressure traits in three cohorts ( namely the discovery cohort H2000 , and the replication cohorts NFBC1966 and YFS ) . The trend of the effect was the same while the association remained suggestive with blood pressure traits in the HBCS . HBCS participants' average age was higher ( 61 years ) than that of the rest ( 36 years ) ( Figure S2 ) . With age , the relative contribution of genetic predisposition and lifestyle may change , potentially accounting for the less significant association in HBCS . In our meta-analysis of candidate loci the most significant association was observed at 5q23 . 2 in PRDM6 . Although an association can be observed throughout the whole gene , fine-mapping of the region with 1000 Genomes variants revealed the focus of association to be within a 4 . 7 kb region in the second intron ( Figure 2 ) . PRDM6 encodes an epigenetic modulator of transcription with roles in endothelial [21] and vascular smooth muscle cells ( SMC ) [22] . PRDM6 has a critical role in arterial wall SMC , where it is predominantly expressed . PRDM6 participates in the phenotypic switch between proliferative and differentiating vascular SMC phenotypes [22]; when active , PRDM6 inhibits differentiation and promotes proliferation . Excess vascular SMC proliferation is an important pathomechanism in hypertension , and it exacerbates the vascular wall remodeling often seen in IA [23] , [24] . When vascular SMCs re-enter the cell cycle to proliferate , they lose their contractile qualities . Distinct from extracranial arteries , cerebral arteries lack an external elastic lamina and the adventitia is weakly developed , making them inflexible , and less resistant to stress [25] . It is possible that when SMC proliferation further stiffens cerebral arteries , they become incapable of adjusting to shear stress , and give way to IA formation . This is a plausible explanation to why the intracranial manifestation of a supposedly generalized vasculopathy can be so distinct . Intriguingly , excessive vascular SMC proliferation is part of the pathomechanism of the strongest common IA risk locus at 9p21 . 3 [26] . However , to test possible causality , examination of whether the risk variant at 5q23 . 2 is associated with higher PRDM6 activity is necessary . Although the causative variant remains elusive , we succeeded in narrowing down the associated region markedly . The 4 . 7 kb region showing the strongest association harbors a SREBP1 binding site . SREBP1 is a transcription factor governing cellular lipid biosynthesis . Highlighting its biological significance in vascular traits , non-synonymous mutations in SREBP1 cause spontaneous hypertension in rats [27] . It is possible that common variants facilitate SREBP1 binding , and thus , as shown by Zhou and colleagues [28] , cause vascular SMC proliferation . We propose that this effect is conveyed by PRDM6 activation . Although both the location and the function of the gene highlight PRDM6 as a likely candidate , it is not the only plausible gene near the association signal . Centrosomal protein of 120 KD ( short form: Cep120 ) is just downstream from the region of association ( Figure 2 ) . Cep120 is a centrosomal protein with preferentially high expression in neuronal progenitors during development [29] . Cep120 could contribute to IA risk by causing perturbation in the neurovascular niche . This is the first study establishing a shared genetic background at 5q23 . 2 for IA and its important risk factor , high blood pressure . However , both IA [30] , [31] and hypertension [32] have shown linkage to 5q23 . 2 in previous studies . Resequencing the genomic region in families that previously showed linkage to 5q23 . 2 might reveal penetrant variants causing familial IA or severe high blood pressure , or possibly both . Notably , Vasan and colleagues [33] found that rs17470137 , less than 8 kb downstream from PRDM6 , is associated with aortic root size , a feasible proxy of blood pressure [34] . GWAS are designed to identify associations , they do not prove causality . Deep resequencing of the associated region may improve the fine mapping and guide closer to the causative variant , or even uncover it , although resequencing efforts of GWAS regions have had limited success [35] . A further limitation of our study is that we were unable to address whether the identified risk variant at 5q23 . 2 increases the risk of developing IA as a consequence of elevated SBP ( causality between high SBP and IA ) or whether the variant modifies vessel wall structure in a way that elevates SBP and increases IA risk as a pleiotropic effect ( Figure S3 ) . A study conducted in a cohort characterized both for IA and blood pressure would likely be a more suitable way of addressing this question . Unfortunately , to the best of our knowledge , such a large-scale cohort does not currently exist . The identified risk variant , however , is unlikely to confer its effect solely by increasing blood pressure , as leading hypertension risk loci fail to show association with IA ( data not shown ) . Yet , the mechanical effect of elevated BP on the vessel wall , likely exacerbates IA formation . The significance of the association identified in our study awaits confirmation in other ethnicities . To further decipher the genetics of IA , it is important to test if genetic links can be established between IA and other strong risk factors , such as smoking and alcohol consumption . In conclusion , our results highlight the link between IA and blood pressure .
Four Finnish population-based cohorts were included in our study ( Table 4 ) . These cohorts were not characterized for IA . We utilized genome-wide genotyped participants with available blood pressure data , excluding those on blood pressure medication and those for whom blood pressure medication data was not available ( nexcluded = 1373 ) . In our two tier approach , the discovery cohort ( ndiscovery = 1581 ) was a subsample of the H2000 [14] . The H2000 study was carried out in several regions of Finland from fall 2000 to spring 2001 , and was designed to provide information on the health of the Finnish population . A subset of this cohort , consisting of metabolic syndrome cases and matched controls , was genotyped and utilized in this analysis . The replication cohort ( nreplication = 8312 ) consisted of the YFS ( n = 1874 ) [15] , [16] , the NFBC1966 ( n = 5361 ) [17] , and the HBCS ( n = 1077 ) [18] . YFS participants were recruited from all around Finland for a large follow-up study on cardiovascular risk factors in young individuals in 1980 . Clinical data are from the follow-up at age 27 performed in 2007 . NFBC1966 comprises individuals born in 1966 in the two northernmost provinces of Finland ( Oulu and Lapland ) . Clinical examinations took place at the follow-up at age 31 in 1997 . The HBCS participants were recruited from the Helsinki region . The study examines the impact of fetal environmental factors on childhood and adult life . Clinical examinations took place during 2001–2004 . HBCS participants had the highest average age ( Figure S2 ) . The ICBP-GWAS represents a union of numerous prior blood pressure GWAS consortia to create a discovery meta-analysis of over 200 000 individuals of European ancestry . NFBC1966 is part of the ICBP-GWAS; however , this overlap does not represent a significant risk for bias , due to the small relative contribution of NFBC1966 to the ICBP-GWAS results . All Finnish cohorts were genotyped using Illumina arrays ( Illumina Inc . San Diego , CA , USA ) : Illumina Infinium HD Human610-Quad BeadChip for H2000 , Illumina HumanCNV370-Duo BeadChip for NFBC1966 , and Illumina Human670K custom BeadChip for YFS and HBCS . For SNPs to be successfully genotyped , a per individual and per marker success rate minimum of 95% was defined as default . 36 out of 41 candidate SNPs were successfully genotyped in all cohorts . For SNPs with no directly genotyped data available , we imputed genotypes with MACH [36] using HapMap CEU from Phase II as the reference panel ( further referred to as HM2 imputed data ) . If a SNP was not present in HM2 imputed data , we used genotypes imputed with IMPUTEv2 using the 1000 Genomes pilot data CEU panel ( August 2009 haplotypes ) combined with HapMap Phase 3 ( Public Release #2 ) haplotypes as the reference panel [37] , extended with Finnish specific HapMap Phase 3 haplotypes [38] ( further referred to as 1000G+HM3 imputed data ) . All missing genotypes were imputed , so the number of individuals included in the analyses for each SNP is the same and equals the final number ( Table 4 ) . Candidate loci were selected based on IA GWAS results [13] . Loci associated with IA with PPA≥0 . 1 were included ( Table 2 ) . PPA was calculated as described by Yasuno et al [12] . Briefly , a uniform prior probability of association of 1/10 000 was assumed for all SNPs and used to provide a probabilistic measure of evidence . We tested 41 SNPs from 19 independent loci . We defined SBP , DBP , MAP , and PP as quantitative outcome variables . MAP was counted as the average of SBP and DBP ( ( SBP+DBP ) /2 ) and PP as the difference of the two ( SBP-DBP ) . We tested all 19 loci in the discovery cohort ( H2000 ) , and those showing suggestive association ( uncorrected p<0 . 1 ) with any outcome variable were tested in the replication cohorts ( YFS , NFBC1966 , and HBCS ) . Association analyses with an additive genetic model were performed with ProbABEL [39] for HM2 imputed data , and with SNPTESTv2 [40] , [41] for 1000G+HM3 imputed data . The analyses were adjusted for age and gender ( ROBUST model ) , or for age , gender , smoking habits , alcohol consumption , and BMI ( ADVANCED model ) . Additionally , in the metabolic syndrome case-control subset of the H2000 cohort we corrected for case-control status in both models . Population stratification was corrected for by calculating principal components from genome-wide SNP data and including significant principal components in the association models as covariates . Association results were combined in a fixed effect meta-analysis with MetABEL [39] for HM2 imputed data , and with METAv1 . 2 [42] for 1000G+HM3 imputed data . The best result at 5q23 . 2 in PRDM6 was tested for association in the ICBP-GWAS [19] cohort of 200 000 individuals of European descent . In the ICBP-GWAS association with SBP was tested by linear regression assuming an additive model and correcting for age , age-squared and BMI . To test the per-allele effect size of risk alleles on blood pressure , we calculated the mean blood pressure for the three genotypic states for the three 5q23 . 2 SNPs using Plink v1 . 07 [43] . Results were plotted using the Microsoft Excel charts function . To further investigate the strongest associated locus , we analyzed all 1000 Genomes variants , with minor allele frequency greater than 1% , in and around PRDM6 . We took uncertainty of imputation into account by using the maximum likelihood estimates of the reference allele counts as genotypes ( these estimates may be fractional and range from 0 to 2 ) . Fine mapping of the 5q23 . 2 region was performed with 1000G+HM3 imputed data . Results were plotted with LocusZoom [44] . | When multiple genes or genetic regions contribute to the inherited risk of a disease , it is referred to as a complex disease . Genome-wide association studies ( GWAS ) aim to detect common genetic variations that associate with complex traits or diseases . Although GWAS have been successful in identifying strongly associated genetic loci , they lack the means to point out true , but less strong , associations . Studying conditions that are related to the disease of interest can help sort out less strong associations . Intracranial aneurysms ( IA ) are berry-like dilations in cerebral arteries . Most IAs do not give symptoms until they bleed , causing a highly fatal form of stroke . Half of the people who suffer bleeding of an IA die . IA is a complex disease . Both inherited risk and environmental factors contribute to the risk of developing IA . Women , smokers , those with high alcohol intake or high blood pressure are more prone to develop IA and bleeding . GWAS found 19 genetic regions increasing the risk of IA . Here we show that one of these loci , on the long arm of chromosome 5 , in addition to raising IA risk also increases systolic blood pressure . We speculate that the cause is modified vascular wall structure . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"neurological",
"disorders",
"neurology",
"genetics",
"biology",
"genetics",
"and",
"genomics",
"cardiovascular"
]
| 2012 | Intracranial Aneurysm Risk Locus 5q23.2 Is Associated with Elevated Systolic Blood Pressure |
Gram-negative pathogens express fibrous adhesive organelles that mediate targeting to sites of infection . The major class of these organelles is assembled via the classical , alternative and archaic chaperone-usher pathways . Although non-classical systems share a wider phylogenetic distribution and are associated with a range of diseases , little is known about their assembly mechanisms . Here we report atomic-resolution insight into the structure and biogenesis of Acinetobacter baumannii Csu and Escherichia coli ECP biofilm-mediating pili . We show that the two non-classical systems are structurally related , but their assembly mechanism is strikingly different from the classical assembly pathway . Non-classical chaperones , unlike their classical counterparts , maintain subunits in a substantially disordered conformational state , akin to a molten globule . This is achieved by a unique binding mechanism involving the register-shifted donor strand complementation and a different subunit carboxylate anchor . The subunit lacks the classical pre-folded initiation site for donor strand exchange , suggesting that recognition of its exposed hydrophobic core starts the assembly process and provides fresh inspiration for the design of inhibitors targeting chaperone-usher systems .
All gram-negative bacteria express fibrous adhesive organelles that mediate targeting to sites of infection . The major class of these adhesive pili ( or fimbriae ) is assembled via the classical , alternative and archaic chaperone-usher ( CU ) pathways [1] . CU pili are linear polymers made of subunits capable of either self-polymerisation or assembly with other subunits [2 , 3] . The CU fibre can possess rich binding properties [3–5] , which facilitate binding to host cell receptors , as well as mediate biofilm formation through self-association [6] and interactions with abiotic surfaces [7] . The biogenesis of CU fibres requires a periplasmic chaperone and outer membrane assembly platform termed the usher [2] . Although these assembly proteins are conserved within the three CU pathway families , little sequence homology exists between the different CU pathways , which suggests distant phylogenetic relationships [1] . Among the three CU systems , the archaic ( also termed σ ) pathway assembles the largest class of pili [1] . Whereas the classical and alternate CU systems are restricted to β- and γ-proteobacteria , members of the archaic CU family are present in α- , β- , γ- , and δ-proteobacteria , whilst also in phyla Cyanobacteria and Deinococcus-Thermus . Furthermore , archaic systems are associated with bacteria that cause some of the most severe diseases in humans , animals , and plants [1] . Archaic Csu pili mediate the formation of Acinetobacter baumannii biofilms , which contribute to high rates of nosocomial infections [7] . This pilus is formed from four subunits , namely CsuA/B , CsuA , CsuB , and CsuE , and is assembled using the CsuC-CsuD chaperone-usher secretion machinery [7 , 8] . The alternative or α CU pathway is a highly divergent family with a wide phylogenetic distribution [1] . This pathway includes CFA/I-like fimbriae , which are the primary adhesins of human enterotoxigenic Escherichia coli , a major cause of mortality in young children from developing countries . The E . coli common pilus ( ECP ) also belongs to the alternative pathway and is associated with both disease-causing and commensal strains [9] . ECP is composed of the EcpA and EcpD subunits , which are assembled using two periplasmic chaperones , EcpB and EcpE , and the EcpC usher [10] . The classical CU pathways , namely β , γ , κ and π , are relatively conserved and they assemble a large variety of structures that are primary associated with the virulence of animal and human pathogens . The classical systems have been studied for several decades and their biogenesis is now understood in exquisite detail . The periplasmic chaperones form a binary chaperone-subunit complex by occupying a hydrophobic cleft created by the absence of a β-strand from the subunit immunoglobulin ( Ig ) like fold , in a process known as donor strand complementation ( DSC ) [11 , 12] . Fibre subunits are subsequently assembled by donor strand exchange ( DSE ) , in which the N-terminal extension from an incoming subunit displaces the chaperone via a "zip-in-zip-out" mechanism [13 , 14] and provides the necessary β-strand [14 , 15] . This process occurs at the entrance to the usher pore and is facilitated [16] by optimal positioning of the incoming chaperone-subunit complex by the usher [17] . Although archaic and alternative systems ( grouped under the term ‘non-classical’ ) have a far wider phylogenetic distribution and are associated with a broader range of diseases than their classical equivalents , little is known regarding their precise assembly mechanisms . Recent structural analysis of the subunits from two alternative systems confirm that their biogenesis is governed by the general principles of DSC and DSE [6 , 18] , although the lack of sequence similarity between chaperones suggests that the assembly process for the non-classical pathways could differ from the classical systems substantially . Here , we report atomic-resolution insight into the structure and biogenesis of Acinetobacter baumannii Csu and Escherichia coli ECP pili assembled via the archaic and alternative pathways , respectively , whilst also highlighting some important deviations from the classical assembly mechanism . The non-classical chaperones , unlike their classical counterparts , maintain subunits in a substantially unfolded state by utilising a register-shifted DSC and a distinct subunit C-terminal carboxylate anchor . The extreme dynamic nature of this chaperone-bound subunit arrangement allows for a more flexible mode of DSE initiation during polymerisation . Furthermore , this mechanistic distinction represents an attractive target for the rational design of new antimicrobials .
The csu gene cluster encodes four different pilus subunits ( CsuA/B , CsuA , CsuB , CsuE ) ( Fig 1A ) . Based on size , positioning within the operon and levels of expression [7 , 8] , we reasoned that CsuA/B is the major shaft-forming subunit and CsuE is a tip subunit . In this scenario , CsuA/B should be capable of polymerisation whilst CsuE , the proposed tip subunit , should not . To verify this , we purified the subunits after over-expression in E . coli and examined their ability to polymerize in vitro . In the absence of the chaperone , only low levels of expression were detected , however upon co-expression with CsuC , these levels dramatically increased . Moreover , the subunits were successfully co-purified via the His-tagged CsuC ( CsuC-His6 ) by Ni2+-affinity chromatography , suggesting that they form stable chaperone complexes ( Fig 1B ) . SDS-PAGE of purified CsuC-CsuA/B complexes after incubation at room temperature revealed a ladder of bands with sizes corresponding to a dimer , trimer , tetramer and larger multimers of the CsuA/B subunit . In contrast , electrophoresis of the CsuC-CsuE complex resulted in a single band of the CsuE monomer ( Fig 1B ) . Boiling the CsuC-CsuA/B sample disrupted the ordered CsuA/B aggregation , resulting in a single band for CsuA/B ( Fig 1B ) . This behaviour has been previously observed for major subunits from classical CU systems [19] and together this confirms that CsuA/B is the major subunit capable of spontaneous polymerisation in presence of the chaperone . Although subunits CsuA and CsuB are not expressed as efficiently as CsuA/B [8] , they have similar size and might also be capable of polymerisation , serving as either adaptors or forming finer shaft structures . Upon closer inspection of N-terminal sequences of CsuA/B , CsuA and CsuB , a clear pattern of alternating hydrophilic-hydrophobic residues is observed , characteristic for N-terminal donor strands of classical pilins ( Fig 1A ) . To test whether CsuA/B polymerizes via DSC , we substituted the first 12 residues with a His6-tag ( His6-CsuA/B ) and co-expressed it with CsuC . Analysis of the co-purified complex by SDS-PAGE prior to and after boiling show no ladder of CsuA/B polymers , indicating that this N-terminal segment is responsible for assembly ( Fig 1C ) . To confirm that the sequence forms a donor strand we prepared single alanine substitution mutants of the two largest hydrophobic residues , Leu10 and Ile12 , in a tagless CsuA/B construct and co-expressed these with CsuC-His6 . SDS PAGE analysis revealed a significant reduction of high molecular mass polymers , particularly for the Ile12Ala mutant ( Fig 1C ) . Moreover , simultaneous mutation of three hydrophobic residues ( Val8 , Leu10 and Ile12 ) abolished CsuA/B polymerisation completely . These results provide convincing evidence that the assembly of archaic pili is based on DSC . To gain insight into the structure and assembly of archaic CU pili , we determined the crystal structure of a CsuC-CsuA/B pre-assembly complex , composed of His6-CsuA/B and CsuC ( to avoid polymerisation ) . Crystals were readily obtained in spacegroup P6422 and the structure was solved using Se-SAD phasing to a resolution of 2 . 4 Å ( Fig 2 ) . The CsuC chaperone has a canonical Ig-like fold with two 7-stranded β sandwich domains ( D1 and D2 ) oriented at ∼90° angle ( Figs 2 and 3A ) . Despite the lack of sequence similarity , comparison of CsuC and classical chaperones with known structures revealed a significant similarity in D1 . For example , 100 equivalent D1 Cα atoms of CsuC and the Yesinia pestis Caf1M chaperone superimpose with RMSD of 1 . 9 Å ( Z-score = 12 . 8 , S2 Table ) ( Fig 3A ) . The largest structural differences occur at the edges of the β sandwich domain . CsuC has an additional β-strand D1 ( Fig 3A and 3B ) . The C1-D1 hairpin protrudes from domain D1 towards domain D2 , closing the entrance to the inter-domain cleft . This additional sequence is present in all archaic chaperones ( S1 Fig ) , suggesting that this blockade of inter-domain cleft plays an important functional role . Domain D2 is less similar to the equivalent domain in classical chaperones ( Z-score = 5 . 9 , S2 Table ) . The principal difference occurs in the position of β-sheet D2C2F2G2 ( Fig 3A ) ; in Caf1M this is rotated with respect to β-sheet A2B2E2 by 35–50° , where as in CsuC this is 60–85° . The nearly orthogonal packing of β-sheets renders the β-barrel in CsuC more open than for Caf1M , although the β-barrel is covered by an additional helix from the E2-F2 loop ( helix 2 ) . Whilst the refinement statistics for the structure of the CsuC-CsuA/B complex are good ( S1 Table ) , approximately 40% of the CsuA/B sequence was not evident in electron density maps , whilst another 7% has very poor electron density ( Cα atom B-factors higher than 80 A2 ) . Furthermore , this is also reflected in the gradual increase in B-factors for CsuA/B outside of the chaperone interface ( Fig 4A ) . To provide further insight into the structural heterogeneity in CsuA/B within the CsuC-bound complex , we prepared a 15N-labelled CsuC-CsuA/B sample and acquired a 1H-15N TROSY NMR spectrum ( Fig 4B ) . While the spectrum shows good chemical shift dispersion consistent with a significant ordered structure , a larger than expected distribution of amide line widths is observed and ~15% of the expected non-proline resonances are absent from the spectrum . This strongly implies that within the context of a pre-assembly complex a large portion , presumably within CsuA/B , displays dynamic conformational exchange on an intermediate timescale . Although our data does not rule out that this large region adopts an alternative conformational state , it more likely that it exists in many different states and exchanges between them . Additional ordered conformational states would manifest as multiple NMR resonances for the same residue . We observe no evidence for this within the NMR spectrum for the chaperone-subunit complex . In fact , the NMR spectral properties more akin to that for a molten globule and is consistent with our crystallographic data , in which no clear electron density was resolved for almost half the CsuA/B sequence . The remaining structure of the CsuA/B subunit reveals a double β-sheet sandwich comprising strands A , B , and E ( β-sheet 1 ) and C and F ( β-sheet 2 ) and β-strand D , which switches between the sheets ( Figs 2 and 4C ) . The three strands of β-sheet 1 ( A , B , and E ) are interrupted in the middle by aperiodic regions . Conserved cysteines 16 and 62 form a disulphide bridge linking the beginning of β-strand A with the end of β-strand B´ . The N-terminal sequence up to Ser13 is disordered and the large unstructured sequences are located between β-strands A and A´ , A´ and B , C and D , D´ and E , and E´ and F . The most striking structural difference between CsuA/B and classical subunits is in the degree to which subunits are folded when in complex with the chaperone . In the majority of available structures of preassembly complexes from classical systems , the entire sequence ( 99–100% ) of the chaperone-bound subunit ( except for the N-terminal extension ) is highly ordered ( S3 Table , e . g . Yersinia pestis capsular subunit Caf1 in complex with Caf1M shown in Fig 4D ) . In contrast to the classical systems , nearly half of CsuA/B is disordered or displays very poor electron density . Nevertheless , the ordered part of CsuA/B provides sufficient information to conclude that , as with classical systems this pilin has the incomplete Ig-like fold in a six-stranded β-sandwich , in which the absent 7th strand ( G ) leaves a large hydrophobic cleft . The ordered part of CsuA/B shows limited structural similarity to Caf1 with a Z-score of 3 . 7 ( Fig 4D ) . CsuA/B interacts predominantly with D1 of the chaperone CsuC via edge strands to form a closed “super-barrel” with a common core . The chaperone A1 and G1 strands are hydrogen bonded to the subunit A and F strands , respectively . Four large hydrophobic residues from the chaperone G1 strand ( Val110 , Phe112 , Met114 , Tyr116 ) are donated to the subunit to compensate for the missing G strand . In addition , strand A1 provides several hydrophobic residues stabilising the super-barrel ( S2 Fig ) . Superposition of CsuC-CsuA/B and Caf1M-Caf1 complexes revealed that CsuA/B is situated closer to the chaperone D2 than Caf1 ( Fig 5A ) . To explore this global difference , we compared the position of donor strand residues in CsuC and Caf1M . In the classical chaperones the donor strand motif can vary in length ( from 3 to 5 hydrophobic donor residues as in Caf1M ) , but it starts from the same position 1 in the classical donor strand register , corresponding to Ile134 in Caf1M . However , in CsuC , the donor strand motif is shifted towards the C-terminal end of β-strand G1 by two residues ( or one donor residue ) . It starts with the highly conserved Tyr116 and corresponds to position 0 using the classical donor strand register ( Figs 5B , 3B , and S1 ) . The subunit and donor residue motif in the chaperone are shifted in the same direction along strand G1 , arguing that the donor strand determines the position of the subunit . Strand F is one residue shorter in CsuA/B than in classical subunits and the C-terminal Phe152 is situated in the centre of the complex , within the interdomain cleft of the chaperone ( Fig 2A ) . The C-terminal carboxylate of Phe152 interacts with two highly conserved residues of CsuC ( Fig 6A and 6B ) . One oxygen atom of the carboxylate forms an ionic interaction with invariant Arg89 ( S1 Fig ) in β-strand F of D1 . The other oxygen atom of the carboxylate is hydrogen bonded to the hydroxyl group of the highly conserved Tyr196 from β-strand F of D2 . Interestingly , the hydroxyl group of Tyr196 is also hydrogen bonded to the conserved Arg174 . This basic residue serves as an acceptor of electrons , strengthening the hydrogen bond between the hydroxyl group of Tyr196 and C-terminal carboxylate ( Fig 6B ) ; therefore , Arg174 also contributes as a part of the carboxylate anchoring mechanism . To study the contribution of this network to the subunit binding , we constructed point mutants of CsuC , namely Arg89Ala , Arg174Ala , and Tyr196Phe . Wild type CsuC and the mutant versions were co-expressed with His6-CsuA/B and assessed with Ni2+-affinity pulldowns ( Fig 6C ) . Arg89Ala and Tyr196Phe mutations reduced the subunit recovery equally , suggesting that both ionic ( carboxylate-Arg89 ) and hydrogen ( carboxylate-hydroxyl group of Tyr196 ) bonds contribute to complex formation . Furthermore , combining these mutations led to near zero recovery of the subunit . The Arg174Ala mutation had a measureable effect on periplasmic levels of CsuA/B , but was less dramatic and supports a secondary role for Arg174 in positioning the carboxylate-Tyr196 . Subunit C-terminal carboxylate binding in archaic chaperones is notably different from that of the classical chaperones . Classical chaperones bind the carboxylate via two highly conserved basic residues ( Arg20 and Lys139 in Caf1M ) , which have no analogues in archaic chaperones . Furthermore , these residues are located in D1 , whereas in archaic systems only one anchoring residue is provided by D1 and two are located in D2 ( Fig 3B ) . Subunit polymerisation within the alternative assembly system , ECP , cannot be explained fully by the classical CU mechanism . This is because it is necessary to insert a large tryptophan side chain from the middle of the EcpA N-terminal extension deeply within the core of an adjacent subunit during assembly [6] . Our observations that archaic chaperones maintain their subunits in a partially folded preassembly state via a distinct binding mode , led us to suggest that alternative systems could employ a similar mechanism . Furthermore , archaic and alternative CU assembly systems pre-date the evolution of the classical pathways , suggesting that they share similar non-classical CU features . To test this we first demonstrated an interaction between the major pilin subunit EcpA and its cognate chaperone , presumed to be EcpB . We co-expressed EcpA ( with a Trp11Ala mutation to abrogate self-polymerisation: EcpAW11A ) in the E . coli periplasm with His-tagged EcpB ( EcpB-His6 ) , followed by purification via Ni2+-affinity chromatography . Subsequent gel filtration and SDS page analysis revealed that a tight complex is formed between EcpB and EcpA with a 1:1 stoichiometry ( S3 Fig ) and this was also confirmed by subsequent NMR analyses ( see later ) . We next crystallized free EcpB and determined its structure using I-SIRAS phasing to 2 . 4 Å resolution ( S1 Table ) . EcpB consists of the two characteristic Ig-like domains as seen in all other chaperones and has an overall boomerang-like shape ( Fig 7A ) . Surprisingly , it superimposes poorly with any subunit free and subunit bound chaperone structures solved thus far ( S2 Table ) . However , when we compared the individual domains , considerable structural similarity could be identified , particularly for D1 . As expected , the superposition revealed that EcpB is structurally more related to CsuC ( Z-score = 14 . 2 ) than classical chaperones ( e . g . Caf1M , Z-score = 12 . 0 ) ( S2 Table ) . No topological differences were found in D1 between EcpB and CsuC , but both possess an additional strand D1 that is absent in the classical chaperones . Less similarity is observed in D2 between EcpB , CsuC and the classical Caf1M ( Z-scores of 3 . 0–3 . 1 , S2 Table ) . Whereas the β-sheet packing in D2 in subunit-bound CsuC is almost orthogonal , in EcpB , the two β-sheets pack nearly parallel to each other , at an angle of just 15–30° . Domain 2 of EcpB does not possess helix 1 , common in classical chaperones , or helices 1 and 2 , characteristic for archaic chaperones ( Figs 3 , 7 , and S1 ) . Instead , it has a short helix located in a long loop between β-strands C2 and D2 . The C2-D2 loop is flanked with a pair of conserved cysteines ( 170 and 179 ) , which form a disulphide bond ( S1 Fig ) that is not seen in classical chaperones . The alternative chaperones can be grouped into two subfamilies ( S1 and S4 Figs ) . One group have significant similarity to EcpB , whereas the other group includes CfaA , which are involved in assembly of class 5 fimbriae in enterotoxigenic E . coli and Yersinia pestis . The crystal structure of chaperone CfaA has been recently determined [20] , providing an opportunity to compare chaperones between these subfamilies ( Fig 7B ) . Expectedly , superposition of N-terminal domains in EcpB and CfaA shows significant structural similarity ( Z-score of 12 . 7 , S2 Table ) . Surprisingly , structural similarity between C-terminal domains is low ( Z-score = 5 . 4 , S2 Table ) . The angle between the β-sheets in CfaA is more similar to classical chaperones and considerably smaller than in EcpB . EcpB and CfaA also display very large differences in the relative domain orientation; when superimposed over D1 ( Fig 7B ) the C-terminal ends are separated by over 30 Å . A detailed comparison of CsuC , EcpB and CfaA reveals that , as in CsuC , the donor strand motifs are shifted towards the C-terminus of β-strand G1 relative to the classical chaperones ( Fig 8A ) . Large residues , Ile117 in EcpB and Tyr120 in CfaA , occupy position P0 . The other positions ( P1-3 ) in EcpB are occupied by alanines of which only one is structured in the subunit-free chaperone structure ( S1 Fig ) . To confirm the role of the P0-3 residues in DSC , we replaced each with glycine and examined whether these mutants bind to the EcpA subunit . Mutant or wild type EcpB was co-expressed in E . coli with EcpAW11A , purified by Ni2+-affinity chromatography from the periplasm and analysed by SDS PAGE ( Figs 8D and S5 ) . Although Ala113Gly had little effect on complex formation , mutations Ala111Gly , Ala115Gly and Ile117Gly practically abolished capture of the subunit , suggesting that DSC is disrupted . The lack of an effect for Ala113Gly is not surprising as we predict that Ala113 occupies the P2 pocket of EcpA . This is where Trp11 is buried during pilus formation ( S6 Fig ) and we envisage that this pocket should be partially open within the EcpB-EcpA complex to enable the introduction of the large tryptophan side chain . Archaic and alternative chaperones share two highly conserved residues ( S1 Fig ) : proline ( Pro59 in CsuC and Pro60 in EcpB ) and arginine ( Arg89 in CsuC and EcpB ) . The proline is the only invariant residue for the entire CU super-family ( Fig 3B ) [21] . It introduces a kink in β-strand D1 switching it between the β-sheets of D1 ( S7 Fig ) . The highly conserved arginine is only present in archaic and alternative chaperones , representing the most characteristic sequence feature of these non-classical systems . Superposition of D1 from CsuC , EcpB , and CfaA reveals that the arginine is located in the same position in all three structures ( Fig 8B and 8C ) . Since Arg89 in CsuC is essential for the binding of the C-terminal carboxylate of CsuA/B , we assumed that the corresponding residue in the alternative chaperones is involved in binding the subunit C-terminus . Furthermore , two other residues of CsuC implicated in binding ( Tyr196 and Arg174 ) superimpose well with identical residues in EcpB ( Tyr169 and Arg148 , respectively ) , suggesting that these residues contribute to anchoring of the subunit C-terminal carboxylate in a similar manner . In CfaA , a similar pair of residues ( Tyr182 and Arg154 ) is located deeper within the inter-domain cleft , due to a two-residue shift of the tyrosine and arginine towards the termini ( Fig 8C ) . This shift is a characteristic feature of CfaA-like chaperones ( S1 Fig ) . To examine the role of Arg89 , Tyr169 , and Arg148 in EcpB in the interaction with EcpA , we created Arg148Ala and Tyr169Phe mutants , and also a double Arg89Ala , Tyr169Phe mutation . We next analysed their ability to recover EcpA in the E . coli periplasm ( Figs 8D and S5 ) . As for CsuC , mutation of Tyr169 in EcpB dramatically decreased its chaperone function and when both Arg89Ala and Tyr169Phe mutations are present , essentially no subunit can be recovered . Although mutation of Arg148 did not reduce the subunit-binding capability of EcpB significantly , our mutational data suggest that archaic and alternative chaperones share the same anchoring mode of subunit C-terminal carboxylate . To confirm whether EcpB maintains EcpA in a partially folded conformation , as observed for the CsuC-CsuA/B complex , we used NMR to compare the solution states of self-complemented EcpA ( EcpAsc ) representing the final fibre-inserted conformation [6] , free EcpB and the EcpB-EcpAW11A complex ( Fig 9 ) . Inspection of 2D 1H-15N HSQC spectra for free EcpA and EcpB ( Fig 9A and 9B ) reveals excellent chemical shift dispersion with resonances observable for all non-proline amides , consistent with the fully folded domains observed in the crystal structures . However , the 1H-15N TROSY NMR spectrum of the EcpB-EcpAW11A complex ( Fig 9C ) displays features characteristic of less ordered regions in the structure . This is highlighted by the high number of peaks at ~8 . 0 ppm when compared to the free components as well as significant variations in amide line-widths . To explore any conformational differences within the pre-assembly complex we transferred resonance assignments made on EcpAsc to the NMR spectrum of EcpB-EcpAW11A [6] . Taking a conservative nearest neighbour approach we were able to assign only ~54% of the EcpA sequence within the complex ( Fig 9D ) . We then measured chemical shift differences for EcpA resonances in free and bound spectra and mapped them onto a docked model based on the CsuC-CsuA/B structure ( Fig 9E ) . Minor shifts of up to one line width difference were grouped together ( blue ) and those experiencing greater than one line width shift or were either broadened beyond detection or shifted beyond this were categorized as major shifts ( red ) . Strikingly , major chemical shift perturbations are consistent with observations from the crystal structure of CsuC-CsuA/B , where they not only localise to regions in EcpA that interact directly with the chaperone , but more significantly many lie far from the EcpB interface and colocalise with same regions in CsuA/B , for which electron density was not observed . Furthermore , superimposing NMR spectra for the chaperone EcpB in its free form and the EcpB-EcpAW11A complex ( Fig 9D ) reveals equally dramatic and widespread chemical shift differences , which is in stark contrast to results obtained from a similar NMR study on FimC-FimH from the classical Fim system [22] . In this study , chemical shift differences occurred at the direct interface with FimH and there was also an absence of any changes in D2 of FimC chaperone suggesting that the domain orientation is preserved . Our data on EcpB-EcpAW11A suggest that the two domains in EcpB undergo a substantial reorientation upon formation of the complex . The subunit EcpA is trapped at an early folding intermediate in which a large portion of the structure remains conformationally heterogeneous and not ordered . Taken together our results suggest that alternative and archaic systems are closely related and define a new non-classical pathway , where their chaperones transport partially folded subunits to the usher . This is in contrast to the classical systems , where subunits are substantially folded , but maintained in an assembly competent conformation .
Our new structural and biochemical data on the non-classical CU pathway show that , although donor strand complementation governs specific chaperone-subunit and subunit-subunit interactions across all CU families , major differences in how the classical and non-classical pathways implement this mechanism exist . It has been suggested that the chaperone-subunit association begins with the binding of the C-terminal end of the subunit to the inter-domain cleft of the chaperone [23] . The vital role of this step in the biogenesis of CU organelles explains why the carboxylate anchoring residues are among the most highly conserved residues in chaperones . However , the distinct differences between anchoring mechanism of archaic and classical chaperones suggest a large evolutionary distance between these systems . The striking feature of non-classical chaperones is the direct involvement of D2 in the anchoring mechanism . To fulfil this role , the domain must be precisely positioned in the subunit bound conformation of the chaperone . Interestingly , the relative orientation of the two chaperone domains in the archaic CsuC-CsuA/B and classical complexes ( e . g . Caf1M-Caf1 ) is nearly identical ( Figs 3A and 7B ) . In contrast , the domain orientation in subunit-free conformations can be varied significantly , as seen in the structure of EcpB ( Fig 7B ) , which implies that a substantial rearrangement in the relative domain orientation occurs during the formation of the chaperone-subunit complex . Movement of the domains apart from one another would significantly decrease the affinity of the chaperone for the subunit in non-classical systems and hence may provide a mechanism for subunit release during the DSE assembly on the usher platform . CU fibre assembly does not require energy from external sources . It has been shown that classical CU chaperones preserve a proportion of the folding free energy of subunits , which is later used to facilitate fibre formation [14 , 24] . Our data suggest that this free energy is , at least in part , stored in a new relative domain orientation adopted by the chaperone when in complex with the subunit . The relative motion of the two chaperone domains during subunit release could be a source of the physical force necessary for translocation of the fibre through the usher pore . The nature and the extent of the changes in domain orientation could be tuned for the size of the subunit secreted and the architecture of the final pilus . This would also explain why the adhesive tip of the ECP system ( EcpD ) has a dedicated chaperone , as the tip subunit is the largest of all the CU pathways at ~60 kDa . Although the chaperone-bound subunits in classical systems are highly structured , they represent the high-energy intermediate and upon DSE the subunit undergoes a structural rearrangement [14] that releases this free energy [24] . Here , we have discovered that archaic and alternative chaperones , unlike their classical counterparts , maintain pilus subunits in a state that exhibits significant disorder . The structure of CsuA/B reveals a loosely packed hydrophobic core , resembling a folding intermediate such as a molten globule . As such , classical and non-classical chaperones appear to trap subunit-folding intermediates at very different stages . Most free energy of folding is released at the stage of hydrophobic core collapse , therefore the non-classical chaperones may preserve significantly more folding free energy of subunits . In classical systems , the subunit assembly proceeds via an usher-coordinated , stepwise zip-in-zip-out mechanism . This involves the gradual replacement of donor strand G1 of the chaperone by free donor strand Gd of the incoming subunit [13 , 14] , which is initiated by an insertion of a hydrophobic side chain at the C-terminal end of strand Gd into a vacant P5 pocket of the acceptor cleft [13 , 25 , 26] . Our study shows that subunits from non-classical systems are devoid of such a pre-folded DSE initiation site , as this region ( P4-5 ) is disordered , which raises the important question of how the zip-in-zip-out process could be initiated in these systems ? In classical chaperone-subunit complexes the donor strand G1 from the chaperone is buried within the compactly folded subunit , however in the non-classical CsuC-CsuA/B complex , it is significantly more accessible to the solvent and only transiently covered by the poorly structured mini-strand A´ . The attacking Gd strand could intercalate fully between strands B and G1 , which could then be followed by the displacement of chaperone strand G1 by the subunit donor strand Gd . In contrast to the classical zip-in-zip-out DSE , lateral replacement could start from any pocket of the acceptor cleft . Interestingly , such a mechanism would explain how the large side chain of Trp11 is introduced in the hydrophobic core of EcpA during DSE [6] . Classical systems possess a bulky hydrophobic residue at the C-terminal end of the donor strand , whereas the largest donor residue of EcpA , Trp11 , is located centrally . The side chain of Trp11 is deeply inserted into pocket P2 of the cleft ( S6A and S6B Fig ) and it would seem more likely that this residue initiates DSE by attacking the disordered region of the subunit laterally ( S8 Fig ) . In the following events , one half of the donor strand in EcpA ( residues 1–13 ) may participate in zip-in-zip-out DSE and the other half ( residues 14–17 ) could stabilise the folding of the remaining part of the subunit . The partially unfolded nature of the chaperone-bound subunit in non-classical systems offers a highly flexible mechanism for assembly . DSE initiation could occur at lower pockets of the acceptor cleft than P5 , which could be followed by a lateral replacement of the donor strand or by a combination of lateral replacement and the classical zip-in-zip-out mechanism . Adaptability of this system to accept larger side chains within the central region of the donor strand side chains could also increase the stability of the final assembled fibre . It is also conceivable that the classical mode of utilising folded subunits has evolved more recently and suggests that it may be a highly refined and more efficient assembly process . It should be noted , that although our study demonstrates a distinct “non-classical” mechanism for Csu and Ecp systems , other systems might show a more mixed type of assembly . For example , β-fimbriae , which are currently considered as classical systems , are in fact closer to the non-classical types described here . Classical subunit anchoring residues are absent in β-fimbriae chaperones , but analogous residues to the non-classical chaperones are present ( S9 Fig ) . Inhibiting the biogenesis of virulence pili at the level of periplasmic assembly is a highly promising strategy for the prevention of infections caused by antibiotic resistant Gram-negative pathogens . Here , we have shown that although the subunit C-terminal carboxylate-binding site is present in both classical and non-classical chaperones , its precise makeup is different between the two families . It may therefore require separate approaches to target this critical subunit-binding site with inhibitors of classical and non-classical CU pathways . At the same time , our discovery of the core-exposed conformation of the chaperone-bound subunit in non-classical systems suggests a novel inhibition strategy: potentially it should be possible to inhibit the DSE step of non-classical pathways by specific targeting of the accessible core . Another attractive target for inhibition is the usher-binding site on the chaperone [27] . Here , both classical and non-classical chaperones present a conserved hydrophobic surface for recruitment by the usher ( S10 Fig ) , therefore , a broad range inhibitor could potentially be developed against this interaction .
Expression plasmids were constructed using a procedure that we previously developed for the expression of fimbrial subunits [4] . Synthetic genes of CsuC fused to a 6His-tag ( 6H ) , CsuA/B and CsuE were ordered from GenScript . Each of the genes was delivered on plasmid pUC57 . The DNA fragment coding for CsuC-6H was inserted into the expression pET101 vector ( Invitrogen ) using restriction enzyme sites EcoRI and SacI to produce plasmid pET101-CsuC6H . 6H was removed by reverse PCR using primers CsuC-6Hdel-R and -F ( S4 Table ) to yield plasmid pET101-CsuC . The nucleotide sequence , encoding residues 29–37 of CsuA/B ( residues 4–12 in the mature protein sequence ) , was replaced by a 6H-coding fragment with a reverse PCR using primers 6H-CsuAB-R and -F ( S4 Table ) . The modified gene of CsuA/B ( 6HCsuA/B ) was cut out with restriction enzymes NheI and SacI and inserted into the same sites in pET101-CsuC to create the CsuC and 6HCsuA/B co-expression plasmid pET101-CsuC-6HCsuA/B . To produce plasmids pET101-Csu6H-CsuA/B and pET101-Csu6H-CsuE , co-expressing CsuC6H with wild type subunits CsuA/B and CsuE , respectively , the CsuA/B or CsuE genes were cloned into plasmid pET101-CsuC6H using restriction sites for NheI and SacI . Full-length ecpB including the N-terminal periplasmic signal sequence and incorporating a C-terminal His-tag was cloned into the pET28b vector ( pET28ecpB ) using In-Fusion ( Clontech ) . Full-length ecpA including the N-terminal periplasmic signal sequence was cloned into the pBAD vector using In-Fusion ( Clontech ) . Trp11 was then mutated to an alanine to prevent auto-aggregation using reverse PCR ( pBADecpAW11A ) . All oligonucleotides are listed in S4 Table . Mutagenesis of CsuC and CsuA/B genes in plasmids pET101-CsuC-6HCsuA/B and pET101-CsuC6H-CsuA/B , respectively , and EcpB in plasmid pET28 , were performed by reverse PCR using primers listed in S4 Table . CsuC-CsuA/B expression and purification is described in [28] . EcpB expression and purification is described in [29] . Wild type and mutant EcpB-EcpAW11A complexes were expressed by co-transformation of E . coli BL21 ( DE3 ) with pET28ecpB and pBADecpAW11A , followed by growth in LB media at 37°C , induction at OD600nm 0 . 6 with 0 . 5 mM IPTG and 0 . 05% L-arabinose , and incubation overnight at 18°C . For 15N-labelled samples , expression was undertaken in M9 media supplemented with 15NH4Cl . Cells were harvested by centrifugation and samples purified by Ni2+-affinity chromatography . NMR samples were further purified with an S200 gel filtration column ( GE healthcare ) . Determination of the crystal structure of CsuC:CsuA/B is described in [28] . Model building and refinements were performed by PHENIX refinement module . Manual corrections were done with molecular modelling program COOT ( Emsley P . , et al . , 2010 ) . Crystals of EcpB were obtained as described in ( Garnett et al . , 2015 ) . Derivative crystals were obtained by soaking native crystals for 30s in 0 . 5 M NaI , 15% ( v/v ) glycerol , 15% ( w/v ) PEG 5000 MME prior to freezing . I-SAD data were collected and initial phases were calculated by SIRAS using SHELXD and SHARP . Automated model building was carried out with ArpWarp , refinement with Refmac and manual model building in COOT . 1H -15N TROSY NMR spectra were collected on 15N-labelled samples ( 250 μM CsuC-CsuCA/B complex , 150 μM EcpB-EcpAW11A complex ) in 20 mM HEPES pH 7 . 0 , 100 mM NaCl , 10% D2O at 298K on a Bruker Avance II 800 spectrometer equipped with TCI cryoprobe . A 1H-15N HSQC NMR spectrum of free EcpB ( 250 μM ) in the same buffer was collected at 298K on a Bruker Avance III 600 spectrometer equipped with TCI cryoprobe . Data were analysed with in-house script using NMRview [30] . The coordinates and structure factors have been deposited in the Protein Data Bank under accession codes 5D6H and 5DFK for the CsuC-CsuA/B complex and EcpB chaperone , respectively . | Gram-negative pathogens depend on fibrous adhesive organelles to attach to target tissues and establish infection . The major class of these organelles is assembled via the classical , alternative and archaic chaperone-usher ( CU ) pathways . CU pathways are recognized as promising new targets for the next generation of antibacterial drugs . The recently discovered archaic and alternative systems are of particular interest , as they are implicated in biofilm formation of antibiotic resistant pathogens , have a wider phylogenetic distribution and are associated with a broader range of diseases than the classical systems . Here , we report an atomic-resolution insight into the structure and assembly mechanism of two such biofilm-forming organelles assembled via the archaic and alternative pathways . We show that the archaic and alternative systems are structurally related , but their assembly mechanism is strikingly different from the classical assembly pathway . Whereas the classical chaperones deliver folded subunits to the usher assembly platform , non-classical chaperones apply a unique binding mechanism to maintain subunits in substantially unfolded state . The open subunit core allows for a new mode of strand replacement during polymerisation , and also represents an attractive target for the rational design of antimicrobials . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| []
| 2015 | Structural Insight into Archaic and Alternative Chaperone-Usher Pathways Reveals a Novel Mechanism of Pilus Biogenesis |
γ-herpesviruses ( γHVs ) have developed an interaction with their hosts wherein they establish a life-long persistent infection and are associated with the onset of various malignancies . One critical virulence factor involved in the persistency of murine γ-herpesvirus 68 ( γHV68 ) is the viral homolog of the Bcl-2 protein ( vBcl-2 ) , which has been implicated to counteract both host apoptotic responses and autophagy pathway . However , the relative significance of the two activities of vBcl-2 in viral persistent infection has yet to be elucidated . Here , by characterizing a series of loss-of-function mutants of vBcl-2 , we have distinguished the vBcl-2-mediated antagonism of autophagy from the vBcl-2-mediated inhibition of apoptosis in vitro and in vivo . A mutant γHV68 virus lacking the anti-autophagic activity of vBcl-2 demonstrates an impaired ability to maintain chronic infections in mice , whereas a mutant virus lacking the anti-apoptotic activity of vBcl-2 establishes chronic infections as efficiently as the wild-type virus but displays a compromised ability for ex vivo reactivation . Thus , the vBcl-2-mediated antagonism of host autophagy constitutes a novel mechanism by which γHVs confer persistent infections , further underscoring the importance of autophagy as a critical host determinant in the in vivo latency of γ-herpesviruses .
Apoptosis and autophagy , characterized by distinctive morphological and biochemical changes , are tightly regulated processes essential for homeostasis , development , and human diseases [1] , [2] . Once triggered by internal inducers , such as DNA damage and viral replication , or by external stimuli , such as the engagement of the TNF receptor , apoptosis proceeds through a cascade of programmed internal proteolytic digestion , resulting in the collapse of cellular infrastructure , mitochondrial potential , genomic fidelity , and cell membrane integrity ( for review see [1] , [3] ) . Therefore , apoptosis represents an important effector of host immunity by eliminating virally-infected cells whose survival might otherwise prove harmful to the host [3] . In contrast to the self-destructing apoptotic program , cellular autophagy ( Greek for ‘self-eating’ ) allows cells to engulf cytoplasmic materials , including long-lived proteins or aberrant organelles , into specialized double membrane-bound vesicles and deliver them to lysosomes for degradation and turnover ( for review see [4] , [5] ) . Although originally characterized as a cellular response to nutrient deprivation , autophagy has been increasingly recognized essential for protecting cells against pathogens [6] . Neuronal overexpression of the autophagy protein Beclin1 confers resistance to Sindbis virus infections [7] . Similarly , depletion of beclin1 in plants aggravates the tobacco mosaic virus-induced hypersensitive response ( HR ) [8] . In addition to digesting cellular components , autophagy has been indicated to sequester virions and bacterial components for degradation [9] , [10] . Thus , autophagy constitutes , in addition to apoptosis , an important host antiviral response [10] . However , the relative contributions and coordination of these two important pathways during viral infection remain largely unknown . Yet as distinct as they are , the apoptotic and autophagic machinery converge at a number of points . One direct crosstalk between these two pathways is mediated in part by the functional and physical interaction of Beclin1 , an essential autophagy activator , with Bcl-2 , a prototype apoptosis inhibitor [11] , [12] . Cellular Bcl-2 was originally discovered as an oncogenic protein in B-cell lymphomas , since then a number of proteins belonging to the Bcl-2 family have been identified , each possessing the signature of Bcl-2 homology domain ( BH ) . The Bcl-2 family consists of both anti-apoptotic ( e . g . , Bcl-2 , Bcl-XL , and Bcl-w ) and pro-apoptotic ( e . g . , Bax , Bak , Bid , and Bad ) members , which cooperate by forming homo- or hetero-dimers to regulate the cell's commitment to apoptosis [13] . A major mechanism by which the anti-apoptotic Bcl-2 proteins block apoptosis involves an extended hydrophobic groove on the surface of the proteins that serves as a binding-pocket for the α-helical BH3-domain of the pro-apoptotic Bcl-2 family proteins [14] , [15] . Aside from its ability to interact with and inhibit pro-apoptotic family members like Bax and BH3-only proteins , the hydrophobic pocket of Bcl-2 also binds Beclin1 ( the mammalian ortholog of yeast Atg6 ) , which is part of a class III PI3 kinase complex required for the initiation of autophagosome membrane [16] , [17] . In fact , the anti-autophagic action of Bcl-2 closely mirrors its capacity to bind and inhibit Beclin1 [12] . Intriguingly , structural analysis of Beclin1 revealed that it possesses a putative α-helical BH3 domain , which allows Beclin1 to dock into the hydrophobic pocket of Bcl-2 . As such , Beclin1 is recently considered as a novel BH3-only protein [18] , [19] . Although it remains unknown how Bcl-2 discriminates among its targets , the dual roles of Bcl-2 in apoptosis and autophagy suggest that a coordinated regulation may exist for Bcl-2 to conduct these two activities [20] . Given the important role of Bcl-2 in cell survival , many viruses have evolved to encode structural and functional orthologs of Bcl-2 ( vBcl-2s ) to prevent the premature death of the infected cells from sustained viral replication and associated diseases [21] , [22] . All sequenced γ herpesviruses ( γHV ) encode a homolog of Bcl-2 , including Epstein-Barr virus ( EBV ) , Kaposi's sarcoma-associated herpesvirus ( KSHV ) , herpesvirus saimiri ( HVS ) , and the murine γ herpesvirus 68 ( γHV68 ) [22] , [23] , [24] . The vBcl-2 of γHV68 ( also referred to as M11 ) has been implicated in preventing Bax toxicity in yeast and blocking apoptosis in cultured cells when induced by diverse apoptotic stimuli [25] , [26] . However , γHV68 vBcl-2 appears to be dispensable for acute infection in vivo and lytic replication in vitro , instead it is proved to be essential for efficient viral persistent replication as well as reactivation from latency , characteristic of all γHVs [25] , [27] . Analysis of the three-dimensional structure reveals that vBcl-2 of γHV68 has limited sequence similarity to Bcl-2 but virtually adopts a fold similar to that of Bcl-2 [25] , [28] . The seven-helix bundle ( α1-7 ) of vBcl-2 forms a globular structure , where helices 2 , 3 , 4 , and 5 define an extended hydrophobic surface cleft allowing vBcl-2 to associate with BH3 domains , especially those of Bak and Bax [25] , [28] . Mutations within the BH3-binding groove abolished the ability of γHV68 vBcl-2 to interact with Bax and Bak , block apoptosis , and abrogate vBcl-2 function in persistent replication and reactivation from latency in vivo [25] . Thus , it is generally believed that vBcl-2 functions in vivo predominantly by binding and inhibiting pro-apoptotic Bcl-2 family proteins [25] . Yet , recent evidence favors a central role for vBcl-2s of γHV in blocking autophagy by directly interacting with Beclin1 via the BH3-binding groove of vBcl-2 [12] , [19] , [29] . Furthermore , the binding of purified vBcl-2 protein to the Beclin1-derived peptide appears to be the tightest when compared to peptides from the pro-apoptotic proteins , including BAX , BAK , BIM , PUMA , BID , and Noxa [28] . Unlike its cellular counterpart , this vBcl-2-Beclin1 complex can not be easily displaced by other BH3-only molecules , such as Bid or Bim [15] , [28] . Accordingly , vBcl-2 exhibits an enhanced capacity for autophagy inhibition than cellular Bcl-2 [28] , [30] . These findings raise the possibility that evasion of autophagy might also account for the biological effects of vBcl-2 in viral lifecycle and/or pathology . Nonetheless , due to the engagement of the hydrophobic surface groove of vBcl-2 by both the pro-autophagic BH3 domain of Beclin1 and the pro-apoptotic BH3 domain [28] , mutations of vBcl-2 identified so far that disrupt Beclin1 binding and inhibition of autophagy also abolish its capacity to interact with BH3 peptides and inhibit apoptosis , adding to the complexity of genetically dissecting the in vivo role of vBcl-2-mediated autophagy inhibition and the vBcl-2-mediated antagonism of apoptosis in γHV68 infection . In this study , we used loss of function mutagenesis to determine the role of vBcl-2-mediated anti-autophagy versus vBcl-2-mediated anti-apoptosis in the context of γHV68 infection . We found that a Beclin1-binding-deficient vBcl-2 mutant virus , which is impaired in autophagy inhibition but retains intact anti-apoptotic activity , was compromised in the maintenance of latency , though the initial viral establishment of latency was unaffected . In contrast , anti-apoptosis-defective vBcl-2 mutant virus infection was associated with a normal latent load but was largely impaired in efficient ex vivo reactivation from latency . Our findings thus demonstrate an as yet undefined function of autophagy in controlling viral infections . Unlike what was previously thought that anti-apoptosis features prominently the functions of vBcl-2 in vivo , our study reveals that an evasion of autophagy-mediated host innate immunity serves as a key aspect of γHV68 replication and pathogenesis .
Beclin1 was originally identified as an interactor of Bcl-2 by a yeast two-hybrid screen [7] . The α-helical structure of the N-terminal region of Beclin1 ( residues 88–150 ) mimics the BH3 domain of pro-apoptotic Bcl-2 family members , allowing it to associate with the hydrophobic BH3-binding groove on the surface of vBcl-2 [19] , [28] . However , the Beclin1 peptide ( KD 40 nM ) binds to vBcl-2 with a much higher affinity than is observed for the Bak ( KD 76 nM ) and Bax peptides ( KD 690 nM ) [28] , raising the possibility that Beclin1 may not necessarily share binding sites with the pro-apoptotic Bcl-2 family members for the hydrophobic groove of vBcl-2 . To dissect the specific interacting domain of vBcl-2 for Beclin1 binding , we created a series of N- and C-terminal deletion mutations of vBcl-2 ( Figure 1 ) , and tested for their abilities to associate with the BH3-like domain ( residue 88–150 ) of Beclin1 in the GAL4-based yeast two-hybrid assay [31] . Wild-type ( WT ) vBcl-2 readily interacted with Beclin1 BH3-like domain in the yeast two-hybrid assay . In contrast , a vBcl-2 mutant with a triple alanine substitution at the conserved residues of Ser85 , Gly86 , and Arg87 ( hereafter termed as vBcl-2 AAA ) within the BH3 binding groove that has been shown to abrogate BH3 peptide binding of vBcl-2 , lost the ability to interact with Beclin1 ( Figure 1 ) . This was consistent with previous observations demonstrating that the hydrophobic groove of vBcl-2 is important for this function [25] . When screening our truncation mutants , we found that the C-terminal truncations of vBcl-2 up to α7 helix showed little to no effect on Beclin1 binding; yet a further truncation up to helix α6 ( e . g . vBcl-2 Δα6/α7/TM mutant ) severely impaired Beclin1 interaction in yeast ( Figure 1 ) , suggesting that the C-terminal boundary of vBcl-2 for Beclin1-binding lies within the α6 helix because no deletions from this end was tolerated . Of particular interest , a deletion of the BH2 domain only , which by analogy to the equivalent domain in Bcl-2 and Bcl-xL has been shown to abolish binding and inhibition of Bax [32] , [33] , did not prevent vBcl-2 from binding Beclin1 in yeast ( Figure 1 ) . This data suggests that the BH2 region of vBcl-2 is dispensable for Beclin1 interaction . In contrast to the C-terminal moiety , removal of a small segment of the α1 helix at the N-terminus abrogated the ability of vBcl-2 to interact with Beclin1 ( Figure 1 ) . In fact , any segment truncation from the N-terminus of vBcl-2 resulted in the complete loss of Beclin1 binding ( Figure 1 ) . Thus , our results indicate that the minimal region required for Beclin1 interaction in yeast involves α helices 1-6 of vBcl-2 . Although the N-terminal α1 helix is not part of the core hydrophobic α helices within the BH3-binding groove , it does appear to be critical for mediating Beclin1 interaction in yeast . We next confirmed our yeast two-hybrid results in mammalian cells . 293T cells were transfected with the WT or mutant forms of vBcl-2 and/or V5-tagged Beclin1 , followed by co-immunoprecipitation ( co-IP ) assays . Consistent with the yeast two-hybrid binding data , deletion of the N-terminal α1 helix of vBcl-2 abolished Beclin1 binding , as also seen with the AAA mutant of vBcl-2 ( Figure 2A ) . The loss of binding activity was not due to defects in protein expressions , since both the Δα1 and AAA vBcl-2 mutants were expressed at equivalent levels to WT in transfected cells ( Figure 2A ) . Yet , deletion of the α7 or BH2 , one of the central components of the vBcl-2 hydrophobic cleft , had no significant effect on the interaction between vBcl-2 and Beclin1 , as was seen with the deletion mutation of the C-terminal hydrophobic ‘tail’ ( ΔTM ) ( Figure 2A ) . Similar results were also observed with endogenous Beclin1 in 293T cells , in that removal of the α1 helix but not the BH2 domain abolished endogenous Beclin1 binding ( Figure 2B ) . These data thus indicate that while the BH2 region is structurally important for assembling the hydrophobic core on the surface of vBcl-2 , it is dispensable for vBcl-2-Beclin1 interaction , whereas the N-terminal α1 helix of vBcl-2 serves as a Beclin1-interacting domain ( Figure 2B ) . These data are consistent with those collected from the yeast two-hybrid assay . Although the N-terminal α1 helix deletion has been previously shown not affecting the overall folding of Bcl-2 family proteins [28] , [34] , it remains possible that the inability of the vBcl-2 Δα1 constructs to bind Beclin1 could reflect the loss of proper folding of the protein . To clarify this , we tested whether the mutants of vBcl-2 retain the ability to associate with other BH3-domain-containing molecules . Bak has been previously shown to have the highest affinity to vBcl-2 in vitro among the pro-apoptotic Bcl-2 proteins [28] . We then performed in vitro GST pull-down assay using the bacteria purified GST-fused Bak protein that was incubated with the cell lysates of 293T cells transfected with the HA-tagged WT or mutant forms of vBcl-2 ( Figure 2C ) . The TM domain of Bak was removed ( referred to as GST-BakΔTM ) to increase its solubility in E . coli . In agreement with previous studies [25] , [28] , we observed that Bak was able to associate with the WT and ΔTM mutant of vBcl-2 , but not with the vBcl-2 AAA mutant ( Figure 2C ) . No interaction was detected between the vBcl-2 and purified GST alone , indicating that the vBcl-2-Bak interaction was specific ( Figure 2C ) . Notably , we found that the Δα1 mutant that lacks Beclin1-binding retained its ability to interact with Bak , whereas the ΔBH2 mutant that was able to bind to Beclin1 failed to interact with Bak ( Figure 2C ) . Thus , the loss-of-function phenotype of the vBcl-2 mutants for Beclin1 or Bak binding , particularly that of Δα1 and ΔBH2 , is less likely due to misfolding of the mutant vBcl-2 protein , rather , it implies that the mechanisms of vBcl-2 for binding with Beclin1 and Bak involve distinct contact sites within the hydrophobic groove of vBcl-2 . The interaction of Bcl-2 with Beclin1 largely correlates to its anti-autophagic activity [15] . We then assessed the effects of the vBcl-2 mutants binding to Beclin1 , in particular that of Δα1 and ΔBH2 mutants , on Beclin1-dependent autophagy . NIH3T3 cells stably expressing empty vector ( NIH3T3 . Vector ) , WT vBcl-2 ( NIH3T3 . WT ) , or the mutant forms of vBcl-2 , including the Δα1 , AAA , Δα7 , ΔBH2 and ΔTM mutants , were generated . To measure autophagy levels , we initially used the fluorescent autophagosome marker GFP-LC3 ( a mammalian homologue of the yeast Atg8 ) , which redistributes from a diffused cytosolic/nuclear staining to a punctate pattern in the cytoplasm upon autophagy stimulation [35] . We found that , consistent with their ability to co-IP Beclin1 , the vBcl-2 ΔBH2 , Δα7 , and ΔTM mutants suppressed autophagy in these cells as effectively as WT did under nutrient depletion and rapamycin treatment , the established inducers of autophagy ( Figure 3A and 3B ) . In contrast , the Δα1 mutant and AAA mutant of vBcl-2 , which were unable to interact with Beclin1 , failed to inhibit both starvation- and rapamycin-induced autophagy in the cells ( Figure 3A and 3B ) . In accord , a significantly reduced number of autophagosomes per cell profile was observed in cells expressing WT and the ΔBH2 mutants , but not the Δα1 and AAA vBcl-2 mutants ( Figure S1A ) . Immunoblotting was then performed with an antibody against LC3 to further measure autophagy in vBcl-2-expressing cells . During autophagosome formation , cytosolic LC3 ( LC3-I ) undergoes a covalent conjugation to phosphatidylethanolamine ( PE ) to yield a lipidated form of LC3 , LC3-II , which displays higher electrophoretic mobility [36] . Consistent with the results of the GFP-LC3 puncta assay ( Figure 3A and 3B ) , the conversion of LC3 from LC3-I to LC3-II was much reduced in WT and ΔBH2-expressing cells compared to that in Δα1- and AAA-expressing cells under normal and rapamycin treatment conditions ( Figure 3C ) . It should be noted that the divergent features of the vBcl-2 mutant proteins in autophagy inhibition were not due to their differing protein expression since all tested mutants were expressed at levels equivalent to WT vBcl-2 in stably transfected cells ( Figure S1B ) . Furthermore , all of the vBcl-2 mutants exhibited punctate cytoplasmic staining in the cells , similarly to the WT , except that the ΔTM mutant of vBcl-2 showed modest nuclear staining ( Figure S1C ) . By analogy to the role of the equivalent region in Bcl-2 relatives , this hydrophobic ‘tail’ probably serves as a membrane anchor sequence in vBcl-2 . While the C-terminal hydrophobic tail is not required for the function of vBcl-2 , it may contribute to vBcl-2 by ensuring the proper subcellular localization of the protein . These data collectively demonstrate that the BH2 domain of the hydrophobic groove of vBcl-2 is not essential for suppressing Beclin1-mediated autophagy and its elimination does not affect Beclin1 binding , whereas the elimination of α1 helix leads to the loss of both Beclin1 binding and autophagy suppressing activity , reflecting a striking correlation between the ability of vBcl-2 to bind Beclin1 and its protection from Beclin1-mediated autophagy . To further address whether vBcl-2 inhibits Beclin1-mediated autophagy in virally infected cells , we generated recombinant γHV68 viruses that express HA-tagged WT ( referred to as HA-WT ) or mutant forms of vBcl-2 including AAA , Δα1 and ΔBH2 mutants ( referred to as HA-AAA , HA-Δα1 and HA-ΔBH2 , respectively ) from its normal context in the viral genome , using the bacterial artificial chromosome ( BAC ) system ( for detail please see Material and Methods ) . The genomic integrities of all recombinants were confirmed by restriction enzyme mapping and Southern blot analyses ( Figure S1D ) . The vBcl-2 protein with the predicted molecular weight of 18 kDa was detected by immunoblotting using an anti-HA antibody with all of the recombinant γHV68 viruses in lytically infected 3T3 fibroblast cells ( Figure S2A ) . Furthermore , the genetic manipulation of vBcl-2 did not affect expression of the neighboring v-cyclin ( ORF72 ) in the recombinant viruses ( Figure S2B ) . NIH3T3 cells were then infected with WT γHV68 or the recombinant γHV68 virus expressing HA-tagged WT vBcl-2 or its mutant derivatives . We found that cells infected with the recombinant γHV68 expressing HA-tagged WT vBcl-2 exhibited comparable levels of autophagy to those of WT γHV68-infected cells , suggesting that HA tagging does not affect vBcl-2 function ( Figure 4 ) . Notably , WT γHV68-infected cells exhibited levels of autophagy indistinguishable from those of mock-infected cells ( Figure 4 ) . In contrast , 3T3 cells infected with the Beclin1-binding-deficient vBcl-2 mutant viruses ( γHV68 vBcl-2Δα1 and γHV68 vBcl-2AAA ) showed significantly higher levels of autophagosome accumulation than those infected with the WT and ΔBH2 mutant viruses ( Figure 4 ) . These data indicate that γHV68 infection can trigger cellular autophagy , which is antagonized by vBcl-2 through Beclin1 inhibition . Taken together , vBcl-2 efficiently inhibits Beclin1-mediated autophagy in transfected and virally infected cells , and that this activity requires the α1 helix of vBcl-2 , whereas the BH2 domain is dispensable for the anti-autophagic activity of vBcl-2 . Given the pivotal role of vBcl-2 in apoptosis inhibition , it is important to know if the regions of vBcl-2 , which is required for binding and inhibiting Beclin1 , are equally or differentially required for blocking apoptosis . To this end , we compared the abilities of WT or the vBcl-2 mutants to confer apoptosis resistance . Upon treatment with TNFα and cycloheximide ( CHX ) for 12 h , NIH3T3 cells stably expressing WT , ΔTM , or the Beclin1-binding deficient Δα1 vBcl-2 mutant survived significantly better than those expressing the empty vector , ΔBH2 , or AAA vBcl-2 mutant ( Figure 5A ) . Further quantification of the apoptotic cells via TUNEL staining revealed that the removal of the BH2 domain resulted in the failure of vBcl-2 in inhibiting apoptosis and the accumulation of apoptotic cells ( Figure 5B ) . In contrast , deleting the α1 helix or the TM region did not affect the ability of vBcl-2 in suppressing apoptosis ( Figure 5B ) . Equivalent results were obtained when we used propidium iodide ( PI ) staining to determine the accumulation of sub-G1 cells , which are considered to be apoptotic , via flow cytometry ( Figure S3 ) . Finally , the robust activation of caspase-3 , an early event in apoptosis , was detected in cells stably expressing the vBcl-2 ΔBH2 and AAA mutants , whereas the expression of WT or the vBcl-2 Δα1 mutant significantly blocked caspase-3 activation elicited by TNFα/CHX- ( Figure 5C ) . This further supports the notion that the deletion of the BH2 domain but not the α1 region seriously attenuates the ability of vBcl-2 to suppress caspase-dependent apoptosis . Taken alongside the autophagy analysis data , these results clearly demonstrate that the vBcl-2-mediated inhibition of apoptosis differs in important respects with its anti-autophagic activity . As summarized in Table 1 , the deletion of the α1 helix in vBcl-2 that prevents Beclin1 binding and autophagy inhibition generally has little effects on vBcl-2' anti-apoptotic activity . In contrast , the removal of the BH2 region of vBcl-2 that abolishes vBcl-2's ability to block host-cell apoptosis has minimal effect on vBcl-2-mediated anti-autophagy . Thus , vBcl-2-mediated antagonism of autophagy can be structurally and functionally distinguished from its previously defined apoptosis inhibition activity , which then provides a general basis for evaluating their functional contributions in vivo during viral infections . To determine the role of the vBcl-2-mediated inhibition of autophagy and apoptosis during viral infection , we first examined the in vitro growth properties of the recombinant γHV68 viruses expressing HA-tagged WT and the mutant forms of vBcl-2 in comparison to WT γHV68 in both BHK21 cells and NIH3T3 cells ( Figure 6A and S4 ) . The γHV68 vBcl-2Δα1 and ΔBH2 mutant viruses , as well as the vBcl-2AAA mutant , grew with the same kinetics as WT γHV68 in cultured cells ( Figure 6A and S4 ) , suggesting that the vBcl-2-mediated inhibition of autophagy and apoptosis are not required for lytic replication in vitro , which is consistent with the previous reports that γHV68 does not require vBcl-2 to replicate in vitro [25] , [27] , [37] . In accord with their growth in vitro in fibroblast cells , vBcl-2 mutant γHV68 viruses replicated at levels comparable to WT γHV68 in the lungs of intranasally infected BALB/c mice 5 or 7 days postinfection ( dpi ) , as measured by plaque assay ( Figure 6B , left panel ) and real-time PCR ( Figure 6B , right panel ) . No statistically significant differences were detected among γHV68 WT , the γHV68 mutant lacking vBcl-2 ( vBcl-2-null ) , and the γHV68 recombinants expressing HA-tagged WT or mutant derivatives of vBcl-2 ( Figure 6B ) . Independent isolates of γHV68 containing the Δα1 or the ΔBH2 mutations of vBcl-2 replicated normally in the lungs , arguing against the possibility of chance mutations having occurred elsewhere in the recombinant virus genomes ( Figure S5A ) . These data collectively support a dispensable role for vBcl-2-mediated anti-autophagy and anti-apoptosis in viral lytic replication both in vitro and in vivo . After immune clearance of acute replication , γHV68 establishes latency in splenocytes , macrophages , and dendritic cells [38] , [39] , [40] . Disruption of vBcl-2 has been indicated to abrogate γHV68 from establishment of latency and/or reactivation [25] , [27] . To determine which of the two activities of vBcl-2 , anti-autophagy versus anti-apoptosis , might be primarily responsible for vBcl-2 function in vivo , we next evaluated the capacities of the recombinant γHV68 vBcl-2 mutant to confer chronic infection in mice in comparison to that of WT γHV68 . BALB/c mice were intranasally infected with 5 , 000 PFU of γHV68 WT or mutants . By 12 and 14 days after infection , when WT γHV68 had reached its peak latent load in the spleen , the splenocytes were isolated and the viral latent loads were assessed by an infectious center assay as previously described [41] . Virus-driven splenomegalies were found in all infected mice , but no significant differences in spleen cell numbers were observable among the samples ( data not shown ) . The virus titer of the vBcl-2 mutant including vBcl-2Δα1 , vBcl-2ΔBH2 , vBcl-2AAA , and vBcl-2 null , was similar to the WT in BALB/c mice ( Figure 7A and 7B , left ) . Consistent with the infectious center data , the vBcl-2 mutant viruses showed peak levels of viral DNA loads comparable to the WT γHV68 at 12 and 14 dpi , suggesting the normal amplification of latent viruses in the spleen ( Figure 7A right , 7B right ) . Similar observations could be made with the independently derived Δα1 and ΔBH2 mutant viruses ( Figure S5B ) . Our results thus indicate that the loss-of-function mutations of vBcl-2 in autophagy or apoptosis inhibition , or both , have no appreciable impact on the establishment of viral latency in spleens , consistent with earlier finding that vBcl-2 is not required for the establishment of latency by γHV68 [37] . Given the lack of a role for vBcl-2 during early times of latent infection , we extended our analyses to determine whether vBcl-2-mediated autophagy and/or apoptosis affect viral maintenance of splenic latency at later time points . No significant difference in splenic latency between WT and the vBcl-2 mutant γHV68 was detected at day 21 ( Figure 7C ) . However , by 28 days postinfection , the titers of vBcl-2 mutant viruses , including the independently derived vBcl-2Δα1 and vBcl-2ΔBH2 mutants , were dropped 6- to 10- fold when compared to the WT ( P<0 . 001; Figure 7D up and S5C ) . Notably , this defect was not transient , but persisted 35 days and 42 days postinfection with a substantial , greater than 10-fold reduction in infectious center titers between WT and the vBcl-2 mutant γHV68 , which represents an approximate 90% decrease in the frequency of latent viruses able to reactivate ex vivo . ( Figure 7E and 7F up ) . These results indicate that there was a sustained deficiency of the vBcl-2 mutant viruses in the maintenance of latency after infection . A contraction of latently infected splenocytes was apparent at 42 dpi for both the WT and vBcl-2 mutant viruses ( Figure 7F up ) . In all of the analyzed mice , preformed infectious viruses were undetectable in equivalent , freeze-thawed spleen samples ( data not shown ) . Taken together , these data suggest that although vBcl-2 mutant viruses initially establish latency at levels equivalent to that of WT , there seems to be a steady decline of the latent virus reservoir at later time points in mice infected with the virus lacking a functional vBcl-2 . Since the infectious center assay does not distinguish between reductions in viral latent loads versus a failure of the latent virus itself to reactivate , we quantitated the viral genomes per splenocyte sample by real-time PCR to further measure the degree of viral latency . In agreement with the reduced infectious center titers at later time points , the viral genome load of the vBcl-2Δα1 mutant virus was severely reduced compared to that of WT viruses over the day 28 to day 42 time course in repeated experiments ( Figure 7D , 7E , and 7F down , and S5D ) . In fact , the γHV68 vBcl-2Δα1 mutant , which expresses an anti-autophagy defective vBcl-2 , was nearly as impaired as the AAA and vBcl-2-null mutant viruses ( Fig . 7D , 7E , 7F down ) . The close correlation between viral genome loads and the frequency of latent γHV68vBcl-2Δα1 reactivation ex vivo at later times postinfection substantiates a severe latency defect of the γHV68 vBcl-2Δα1 virus . Since deletion of the α1 helix abolished vBcl-2's anti-autophagic activity but retained its anti-apoptotic function , the impaired latency associated with the vBcl-2Δα1 mutant virus infection at later times thus suggests that autophagy evasion by vBcl-2 plays an important role in maintaining γHV68 latent infection in splenocytes , whereas vBcl-2-mediated anti-apoptosis may not be absolutely required or sufficient for maintaining γHV68 latency . Indeed , the vBcl-2ΔBH2 mutant defective for apoptosis inhibition yet retaining Beclin1-binding and autophagy inhibition intact maintained viral genome loads equivalent to WT virus ( Figure 7D , 7E , 7F down and S5D ) , arguing that the ΔBH2 mutation had no significant impact on viral latency . Yet , in marked contrast to the normal viral genome loads and splenocyte numbers ( Figure S5E ) in the ΔBH2 mutant virus-infected mice , the infectious center titer of the vBcl-2ΔBH2 mutant was significantly lower than WT γHV68 at later times , as previously described ( Figure 7D , 7E , and 7F up ) . The disparity between the viral genome load and the latent viral titer argues that although the γHV68 vBcl-2ΔBH2 mutant virus is capable of maintaining a WT-level viral DNA load , it is unable to efficiently reactivate from latency in the spleen between day 28 and day 42 after infection . Since the BH2 domain is involved in the ability of vBcl-2 to inhibit apoptosis but not autophagy , this result thus suggests that the inhibition of host apoptosis by vBcl-2 is required for efficient ex vivo reactivation from the latent state particularly at later time points after infection , which is consistent with the previous report that the γHV68 vBcl-2 is required for latency reactivation [37] . The viral capacity of latency maintenance is prerequisite for establishing lifelong persistent infection of γHV and is often associated with various malignancies . Our study of the vBcl-2 mutant γHV68 viruses thus indicates that vBcl-2-mediated anti-autophagy and anti-apoptosis may play distinct role in γHV68 persistent infection , in that autophagy evasion by vBcl-2 is particularly required for the maintenance of viral latency , while the vBcl-2-mediated inhibition of apoptosis may play a role during upon viral reactivation .
Here we provide evidence that the vBcl-2-mediated Beclin1 binding and autophagy inhibition is necessary for the maintenance of γHV68 latent infection , whereas the capability of vBcl-2 to antagonize the host apoptosis response is required for efficient viral reactivation from latency ex vivo . Our study thus for the first time indicates that the vBcl-2-elicited anti-autophagy and anti-apoptosis activities are functionally and genetically distinct , also suggesting that the evasion of autophagy represents a critical step in the lifecycle and/or pathogenesis of γHVs . The anti-apoptotic Bcl-2 proteins are structurally characterized by a hydrophobic surface groove that can accommodate the BH3 domain of the pro-apoptotic Bcl-2 family members as well as the BH3-like domain of Beclin1 . Structural alignments of the BH3-like domain of Beclin1 and the BH3 domain of the pro-apoptotic Bcl-2 proteins revealed highly conserved topology and groove contact sites despite overall sequence variability , leading to the conclusion that Beclin1 is a putative BH3-only protein [15] , [18] , [19] . However , no apparent apoptosis induction activity has been found with Beclin1 in an in vivo context [15] . Moreover , our study indicates that despite their structural overlap , Beclin1 and pro-apoptotic Bcl-2 proteins interact with vBcl-2 through two discrete modes of binding that are dependent on a distinct region of vBcl-2 . We show that removing the BH2 domain from vBcl-2 does not affect vBcl-2's capacity to bind and suppress Beclin1 , but it significantly dampens its anti-apoptotic activity . By contrast , deleting the α1 helix does not affect vBcl-2's capacity to suppress apoptosis , yet strikingly impairs its anti-autophagic activity . We thus propose that the anti-apoptotic function of vBcl-2 is not required for its effect on autophagy inhibition and vice versa . Compared to a previous study of the vBcl-2-Beclin1 interaction in vitro [19] , our in vivo data further demonstrates that mutations of vBcl-2 outside the vBcl-2-Beclin1 BH3 domain interface ( e . g . vBcl-2Δα1 ) also affect vBcl-2-Beclin1 binding affinity , probably by altering the conformation of the hydrophobic cleft composed of the BH1 , BH2 , and BH3 domains . It can be speculated that not only do different BH3 domains have distinct binding footprints on the vBcl-2 surface groove as previously described [19] , but that vBcl-2 undergoes different conformational changes when bound to distinct BH3 domain sequences . Thus , our data , in conjunction with recent findings [19] , [28] , provides a molecular explanation for the distinctly different roles of vBcl-2-mediated apoptosis and autophagy regulation in living cells . γHV68 vBcl-2 is required for persistent viral replication and reactivation of the virus from latency [25] , [27] , [37]; two types of biological activities have been described: anti-apoptosis and anti-autophagy . However , it has not been possible to experimentally delineate their relative contributions to overall vBcl-2 functions in vivo . Our studies have allowed us to genetically distinguish the role of vBcl-2-mediated blockage of autophagy in vivo from vBcl-2-mediated anti-apoptosis by constructing a recombinant mutant virus that has the ability to block apoptosis but is unable to inhibit autophagy in infected cells . This mutant virus is highly attenuated in maintaining viral latency , suggesting that the vBcl-2-mediated inhibition of host-cell apoptosis is not sufficient to confer persistent infection , but rather that the vBcl-2-mediated blockage of Beclin1-dependent autophagy is required for the efficient maintenance of viral latency , a prerequisite for subsequent reactivation and transmission . This finding seems to be without precedent because vBcl-2 homologs have not been recognized to directly contribute to viral latent infection by interfering with the host autophagy machinery . Our data , however , do not rule out an important role for apoptosis , likely provided by other viral factors , in maintaining viral latency in vivo but , instead , they identify a vBcl-2-associated autophagy defect in chronic infection of γHV68 . Analogous to γHV68 vBcl-2 , KSHV-encoded vBcl-2 has been found to target Beclin1-dependent autophagy more strongly than cellular Bcl-2 [12] , [30] . Given their poor overall amino acid homology to cellular Bcl-2 family members , the conservation of a mechanism of autophagy inhibition strongly supports the notion that interfering with the host autophagic machinery likely represents a common strategy for latent infection shared by these , and possibly other persistent γHVs . Nonetheless , it remains possible that the impaired latency that we observed with the vBcl-2 mutant γHV68 virus is not simply due to the disabled anti-autophagic activity of vBcl-2 but other as-of-yet undefined mechanisms . Future studies of viral replication and pathogenesis in mice lacking functional autophagy genes should help address this contingency . Notably , two previous studies have also demonstrated the functional role of vBcl-2 during γHV68 chronic infection but with slightly different results [37] . In contrast to our findings that vBcl-2 is required for efficient maintenance of γHV68 latency , Gangappa et al . did not observe evident defects in the splenic latency of the vBcl-2 mutant [37] . As previously noted [27] , [42] , this discrepancy is potentially due to the use of different viral administration routes: intraperitoneal inoculation in Gangappa et al . versus intranasal inoculation in our study . On the other hand , de Lima et al . observed a reduced efficiency in the initial establishment of γHV68 splenic latency in the absence of a functional vBcl-2 as early as day 14 postinfection , which was subsequently recovered 6 months postinfection [27] . However , in our study with the vBcl-2 mutant viruses , the latency defect was not detected until 4∼6 weeks postinfection , the period in which a contraction of latently infected splenocytes was apparently observed . The basis for the differences between these data and our findings is not yet clear . It is possible that the higher dose inoculation ( 2×104 PFU ) used by de Lima et al might provoke stronger proinflammatory responses in the lung , which may presumably affect the initial viral seeding and amplification in the spleen . Alternatively , the vBcl-2 may carry out additional activities other than anti-apoptosis and anti-autophagy that contribute to the chronic infection of γHV68 . Despite the discrepancies revealed in different experimental settings , the lack of absolute ablation of latency upon infection with the γHV68 vBcl-2 mutant suggests a complex nature of γHV68 persistence , involving multiple viral factors and cellular processes . Nevertheless , our studies on the role of vBcl-2 , particularly its anti-autophagy function , in the maintenance of splenic latency highlight the importance of autophagy during γHV68 infection . Despite the fact that vBcl-2 is primarily expressed during the γHV68 lytic cycle and that it efficiently blocks apoptosis in cell culture and in transgenic models [25] , [26] , [37] , γHV68 vBcl-2 is dispensable for the initial evasion of apoptosis in acute infections , as also shown by Gangappa et al . and de Lima et al . [27] , [37] . In this respect , one would expect that other anti-apoptotic lytic γHV68 genes be involved in the acute phase of γHV68 infections . Indeed , recent work by Feng et al . [43] indicates that γHV68 encodes a mitochondrion-associated anti-apoptotic protein ( vMAP ) that effectively antagonizes apoptosis and is required for the lytic replication of γHV68 in cell culture . It will then be of interest to test whether vMAP can also antagonize the host autophagy response and whether autophagy is involved in acute infections of γHV68 . On the other hand , the preferential roles of vBcl-2 at the late stages of latent infection and ex vivo reactivation from latency strongly imply that its expression likely continues through into latent infection [27] , [37] . This view is strengthened by the detection of the vBcl-2 transcript in latently infected cells and/or tissues , albeit in low abundance [41] , [44] , [45] , [46] . However , it should be noted that because the vBcl-2-encoding M11 gene is interposed in the opposite orientation between the v-cyclin gene and LANA/ORF73 , with its transcript overlapping with ORF73 , it remains possible that the RT-PCR signal for vBcl-2 may correspond to an ORF73 mRNA encoded on the opposite strand . Due to the complex nature of transcription across this region , a strand specific transcript mapping may be necessary for clarifying the expression profile of vBcl-2 during different phases of viral infection in mice . While dispensable for lytic replication , vBcl-2 has been previously indicated to be required for ex vivo reactivation [25] , [37] . We further extended this view by showing that the reactivation efficiency of vBcl-2 correlates with its anti-apoptotic ability , not anti-autophagic effect . We found that a mutant strain of γHV68 that is specifically impaired in the apoptosis-inhibitory activity of vBcl-2 , while remaining competent for autophagy inhibition , exhibited normal levels of splenic latency but inefficient ex vivo reactivation of the virus from the latently infected cells , suggesting that apoptosis evasion by vBcl-2 is particularly important during the reactivation process of the virus from latency . Notably , this viral phenotype , associated with the ΔBH2 mutant , was revealed at day 28 dpi but not at earlier time points of latency . This implies that the viral maintenance of latency represents a genetically distinct phase of γHVs infection and involves different viral and/or host factors . In this scenario , it is more likely that additional apoptosis inhibitors of γHV68 and/or the host proteins may be required for , or at least directly involved , for ex vivo reactivation in the early stage of latency , compensating for the anti-apoptotic effect of vBcl-2 . Therefore , our studies do not preclude the importance of apoptosis regulation for ex vivo reactivation at the early stage of latency , but rather have identified a vBcl-2-associated deficit in latency maintenance . Such a deficit would also presumably reduce the virulence of γHV68 . Furthermore , it has been set forth that the poor viability of explanted murine B cells may conceivably affect the efficiency of ex vivo reactivation . It is thus possible that the removal of the BH2 domain , which mitigates the anti-apoptotic properties of vBcl-2 , may affect the survival of explanted B cells , thereby indirectly impacting the efficiency of the ex vivo reactivation of the virus . However , this possibility did not manifest itself in the early time points of viral latency , suggesting that survival of latently infected B cells in culture does not play a critical role in dictating the phenotype of the γHV68 vBcl-2 ΔBH2 mutant and that the vBcl-2-mediated inhibition of host apoptosis may be more directly involved in the reactivation programming of γHV68 . Nevertheless , our study of the requirement of vBcl-2 for both latency maintenance and reactivation process is consistent with vBcl-2 being expressed in latently infected tissues [45] . It also points to the unique protective activities of vBcl-2 mutants in apoptosis and autophagy with respect to distinct phases of viral infection , and also their coordinated effects on γHV68 persistency and/or pathogenesis . γHVs have developed a unique mode of interaction with the host where they establish lifelong latency and may be reactivated throughout the life of the host , which has been associated with the onset of various malignancies . Although it remains to be fully understood what factors govern the establishment and maintenance of latency , our study clearly demonstrates that sustained γHV68 latency in splenocytes requires the vBcl-2-mediated inhibition of the host autophagy machinery . But , how autophagy functions and what accounts for its effect are not yet understood . Given the substantial contributions of autophagy to the quality control of cytoplasmic components in host cell , most simply , autophagy induction may promote the degradation of cytosolic viral protein ( s ) essential for the maintenance of latency . Alternatively , the ‘autophagic digestion’ of viral latent antigens may facilitate its MHC class II presentation and cytotoxic T cells ( CTL ) recognition , as recently exemplified by the nuclear antigen 1 of Epstein-Barr virus ( EBNA1 ) [47] . Additionally , autophagy may help to deliver a ‘viral signal’ to TLR-containing endosomes , thus stimulating the IFNα production that has been proven to be important to the control of acute γHV68 infection as well as latency [48] , [49] , [50] , [51] . It is also possible that the autophagy induced when Beclin1 is unchecked by vBcl-2 can trigger cell death of latently infected cells , such a scenario is supported by the fact that a Beclin1 mutant unable to bind to Bcl-2 induces caspase-independent autophagic cell death [12] . The observations that autophagy may promote the sequestration and digestion of replicating viruses inside the host cell as described in HSV-1 [52] , [53] could also provide an attractive explanation for the restriction of persistent infection of γHV68 , but direct evidence is missing . While it is not yet clear by which mechanism ( s ) autophagy restricts viral persistency , none of these mechanisms are mutually exclusive and there may be other consequences of autophagy function relating to the activation of the host immune responses against γHV68 and latency , as well . Further studies examining the molecular details involved in the vBcl-2-mediated inhibition of autophagy will expand our understanding of both autophagy and γHV-associated pathogenesis and reveal novel targets for antiviral therapy . In conclusion , we have described a crucial role for the viral evasion of autophagy in latent viral infections . Beyond its established anti-apoptotic functions , vBcl-2 targets the host autophagy effector protein Beclin1 and this activity of vBcl-2 is essential to the viral maintenance of latency . Our findings thus indicate that two host innate immune pathways , autophagy and apoptosis , both targeted by vBcl-2 , actually conduct hitherto unexpected and distinctive roles in protecting against viral infections . Future studies will aim to analyze in detail the molecular mechanisms of autophagy that contributes to controlling γHV infection .
All mice handling was performed in accordance with the Animal Research Committee guidelines of the University of Southern California ( USC ) and the University of California , Los Angeles . All methods used herein have also been approved by the USC Animal Research Committee . BALB/c mice were obtained from Charles River Laboratories , Inc . ( Wilmington , MA ) . All mice ( ∼6-week old , n = 5∼8 per pool ) were infected intranasally with 5 , 000 plaque-forming units ( PFU ) of γHV68 viruses under brief halothane anesthesia and the infected mice were sacrificed at 5 and 7 days post-infection ( dpi ) to measure acute infection in the lungs or at 12 dpi , 14 dpi , 21 dpi , 28 dpi , 35 dpi , and 42 dpi to measure viral latent load in the spleen . NIH3T3 , BHK21 and 293T cells were cultured in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal bovine serum , 2 mM L-glutamine , and 1% penicillin-streptomycin ( Gibco-BRL ) . Transient transfection was performed with Fugene 6 ( Roche ) , Lipofectanine 2000 ( Invitrogen ) , or Calcium phosphate ( Clontech ) . NIH3T3 stable cell lines were established using a standard protocol of selection with 2 µg/ml of puromycin ( Sigma-Aldrich ) . The wild-type ( WT ) γHV68 virus , pBAC/γHV68 virus [54] , and its mutant derivatives were all propagated in BHK21 cells for in vitro studies and in NIH3T12 cells for in vivo studies . A DNA fragment corresponding to the γHV68 vBcl-2 coding sequence was amplified from S11 genomic DNA . The PCR-amplified vBcl-2 DNA was then cloned into a modified pEF-IRES-puro vector ( Invitrogen ) encoding an N-terminal HA tag ( pEF-HA-vBcl-2 ) . Mutations in the vBcl-2 gene were generated by PCR ( Hi-Fidelity PCR kit , Roche ) with oligonucleotide-directed mutagenesis . Specifically , vBcl-2 Δα1 ( lacking the N-terminal 21 residues ) and ΔTM ( lacking the C-terminal 20 residues ) deletion constructs were amplified from the pEF-HA-vBcl-2 vector using specific primers; The vBcl-2 ΔBH2 ( lacking residues 129-144 of BH2 domain ) and Δα7 [lacking residues 130–134 ( NHFPL ) ] mutants were created via two-step PCR mutagenesis . The HA-vBcl-2 AAA mutant with alanine substitutions at the Ser85-Gly86-Arg87 residues was created using a Quickchange site-directed mutagenesis kit ( Stratagene ) . All of the PCR products with the indicated vBcl-2 mutations were then cloned in frame into the XhoI/MluI sites of the pEF-IRES-puro vector , for both transient and stable expression . All mutant constructs were completely sequenced to ensure the presence of the desired mutation and the absence of secondary mutations . Constructs expressing the HA-tagged Bcl-2 family proteins were kindly provided by J . Marie Hardwick ( John Hopkins university ) . The Beclin1-V5 plasmid has been described previously [29] . For yeast-two hybrid analyses , vBcl-2 and its mutant derivatives were cloned into the EcoRI/BamHI sites of the yeast plasmid pGBKT7 ( Clontech ) , which carries the S . cerevisiae TRP1 gene as a selectable marker . The BH3-like domain ( residues 88-150 ) of Beclin1 was subcloned into the EcoRI/XhoI sites of the pGADT7 vector ( Clontech ) , harboring the LEU selection marker . To produce a GST fusion protein of Bak ( GST-BakΔTM ) from E . coli , the PCR product of Bak cDNA with a deletion of the C-terminal TM region was subcloned into the EcoRI/XhoI sites of pGEX4T-1 . All constructs were sequenced using an ABI PRISM 377 automatic DNA sequencer . To make specific mutants of γHV68 ( i . e . HA-WT , HA-Δα1 , HA-ΔBH2 , and HA-AAA ) , the two-step bacteriophage lambda Red-mediated homologous recombination method was performed using the γHV68 bacterial artificial chromosome ( BAC ) clone in GS1783 ( an E . coli strain containing an arabinose-inducible I-SceI gene , provided by G . Smith , Northwestern University Medical School ) as previously described [55] . Briefly , PCR was used to generate constructs containing the kanamycin-resistance ( KanR ) gene with the mutated vBcl-2 gene . This kanamycin cassette was then inserted into the γHV68 BAC clone by homologous recombination and a markerless mutation was achieved through the deletion of the kanamycin resistance gene using I-SceI . Consequent mutations in the BAC DNAs were confirmed by DNA sequencing and the genomic integrity of the mutated BAC MHV-68 was investigated by restriction enzyme digestion and southern blot analysis as previously described [54] . vBcl-2-null BAC was generated by the in vitro MuA transposition of signature tagged transposon [54] . All BACs were reconstituted into infectious viruses by transfecting the BAC DNA along with the Cre recombinase-expressing plasmid , which removes BAC vector sequence , into NIH3T12 cells using Lipofectamine Plus reagent ( Invitrogen ) . The produced viruses were purified as single clone by limiting dilution and then amplified in NIH3T12 cells . The purified viral stock was tittered by plaque assays , using a monolayer of Vero cells overlaid with 1% methylcellulose . By 5 days post-infection , the cells were fixed and stained with 2% crystal violet in 20% ethanol . Plaques were then counted to determine infectious titer . To analyze the interactions between the Beclin1 and vBcl-2 mutants , yeast strain AH109 , expressing the BH3-like domain of Beclin1 fused to the Gal4 activation domain in the pGADT7 plasmid , was used to transform pGBKT7 plasmids containing the mutants of vBcl-2 , and the transformants then assayed for α-galactosidase activity , as previously described [56] . Quantitative GFP-LC3 light microscopy autophagy assays were performed in NIH3T3 stable cells expressing the WT or mutant forms of vBcl-2 , then transfected with a GFP-LC3-expressing plasmid [35] . Autophagy was then induced by starvation or rapamycin treatment . For starvation , the cells were washed three times with PBS and incubated in Hank's solution ( Invitrogen ) for 4 h at 37°C . Alternatively , the cells were cultured in DMEM containing 1% FBS and 2 µM rapamycin ( Sigma-Aldrich ) for 6 h . LC3 mobility shift was detected by immunoblotting as previously described [29] . For autophagy levels during viral infection , NIH3T3 cells were transfected with GFP-LC3 , then infected with recombinant γHV68 WT or mutant viruses at an MOI of 5 , and fixed 18 h after infection . NIH3T3 cells stably expressing the WT or mutant forms of vBcl-2 were seeded at 1×106 cells per well into 6-well plates for 24 h . The cells were then treated with fresh medium containing 2 ng/ml tumor necrosis factor alpha ( TNFα ) plus 1 µg/ml cycloheximide ( CHX ) for up to 12 h . For the cell viability assay , the cells were stained with trypan blue for dye exclusion . For the analysis of apoptotic cells , the samples were prepared using an DEADEND™ Fluorometric TUNEL system kit ( Promega ) according to the manufacturer's instructions . Nuclei were counterstained with 4 , 6-diamidino-2-phenylindole ( DAPI ) . Fluorescence microscopy analyses were performed with an Olympus IX-70 microscope . The percentage of TUNEL-positive cells was determined against the number of DAPI-stained nuclei . For the PI staining assay , the cells were collected with the cell dissociation buffer ( Sigma-Aldrich ) and then fixed with 70% of ethanol overnight at −20°C . Fixed cells were washed twice with PBS , and incubated in PBS containing propidium ( PI; 5 µg/ml ) , RNase A ( 1 mg/ml ) , and Triton X-100 ( 0 . 5% ) at room temperature for 30 min . Fluorescence emitted from the propidium–DNA complex was measured using FACScan flow cytometry . Cells containing hypodiploid DNA were considered apoptotic . The data was analyzed using Cell Quest ( BD Bioscience ) . For caspase-3 activity assay , the cells were harvested after treatment , washed three times with PBS and fixed with fixation medium ( Invitrogen , Catalog# GAS001S ) for 15 min , permeabilized with Permeabilization Medium ( Invitrogen , Catalog# GAS002S ) for another 15 min , and then stained with PE-conjugated anti-Caspase3 active form ( BD biosciences #550821 ) for flow cytometry analysis . Data was analyzed by FlowJo-6 . 4 . For apoptotic levels during viral infection , NIH3T3 cells were infected with recombinant γHV68 WT or mutant viruses at an MOI of 5 , and apoptosis was assessed by TUNEL staining and nuclei counterstaining as described above . BHK21 cells and NIH3T3 cells were seeded at 2×105 cells per well into 6-well plates for the single-step growth curve with a multiplicity of infection ( MOI ) of 5 . 0 , or at 1×105 cells per well for multi-step growth curves with an MOI of 0 . 1 . The samples were harvested at various time points post-infection , subjected to three freeze-thaw cycles , then titered by plaque assay in triplicate as previously described [37] . To determine the virus titer in the infected lungs , the lungs were homogenized in 1 ml of DMEM and the infectious viruses in the homogenate supernatants was measured by three independent plaque assays . For infectious center assay , which measures the amount of the latent virus that is able to reactivate from the latently infected B cells , single cell suspensions of splenocytes were prepared from the infected spleens and co-cultivated with a monolayer of Vero cells overlaid with 1% methylcellulose . The Vero cells were incubated further for 5 days , then fixed and stained with 2% crystal violet in 20% ethanol . Plaques were then counted to determine the infectious centers [57] . A majority of the samples in the assay for preformed viruses resulted in no plaque , with a minority of samples displaying 1 to 2 plaques per ∼107 splenocytes . For quantification of viral genome load from the infected cells/tissues , total genomic DNA from the infected organs was prepared and subjected to quantitative real-time PCR , as previously described [54] . Briefly , total genomic DNA from the infected lungs or the spleen tissues was extracted using a DNeasy Tissue Kit ( QIAGEN , Valenia , Calif . ) , according to manufacturer's instructions . γHV68 ORF56-specific primers ( forward primer: 5′-GTAACTCGAGACTGAAACCTCGCAGAGGTCC-3′; reverse primer: 5′-CCGAAGCTTGCACGGTGCAATGTGTCACAG-3′ ) were used in the assay . The DNA templates were mixed with 2× Master mix ( Biorad iQ™ SYBR® Green Supermix ) and PCR was performed at 95°C for 15′ and 45 cycles of 95°C for 30″ , 60°C for 30″ , and 72°C for 30″ , followed by melting curve analyses . 100–500 ng of DNA was analyzed in duplicate for each sample and compared with a standard curve of a BAC plasmid containing the γHV68 genome , serially diluted with uninfected cellular DNAs and amplified in parallel . Amplification and detection were performed using Opticon II ( MJ Research ) . The specificity of the amplified products was confirmed by agarose gel electrophoresis . Quantitative analyses of v-cyclin transcript were performed using SYBR GreenER™ qPCR Kit ( Qiagen ) on a DNA Engine Opticon® 2 continuous Fluorescence Detection System ( MJ Research , Incorporated , Waltham , MA ) . Total RNA was extracted from the infected cells using Trizol ( Invitrogen ) and 100 ng of purified total RNA was reverse transcribed to cDNA using a cDNA synthesis kit ( Invitrogen ) . The PCR reaction was set according to the manufacturer's recommendations . Briefly , after an initial 5 minutes of denaturation at 95°C , thermal cycling was performed at 94°C for 45″ , 57°C for 1′ , and 72°C for 1′ for a total of 40 cycles followed by a melting curve analyses . The amount of RNA was normalized with the quantified β-Actin in each sample . The primer sets for amplification of orf72 were: forward , 5′-GGAGCAACAACAGCTGACAA-3′; reverse , 5′-GTGATTAGCACTGGGCGTTT-3′ . The primer sets for β-Actin were: forward , 5′-CGAGGCCCAGAGCAAGAGAG-3′; reverse , 5′-CGGTTGGCCTTAGGGTTCAG-3′ . Quantitative experiments were performed at least three times , including a no-template control each time . The size of the amplified products was confirmed by agarose gel electrophoresis . For immunoblotting , the polypeptides were resolved by SDS-PAGE and transferred onto a PVDF membrane ( Bio-Rad ) . The membranes were blocked with 5% non-fat milk , and probed with the indicated antibodies . Goat antibodies coupled to horseradish peroxidase specific to mouse or rabbit immunoglobulins were used as secondary antibodies ( diluted 1∶10 , 000 , Sigma-Aldrich ) . Immunodetection was achieved with a chemiluminescence reagent ( Pierce ) and detected by a Fuji Phosphor Imager ( BAS-1500; Fuji Film Co . , Tokyo , Japan ) . For immunoprecipitation , cells were harvested and then lysed in a 1% NP40 lysis buffer supplemented with complete protease inhibitor cocktail ( Roche ) . After pre-clearing with protein A/G agarose beads for 1 h at 4°C , whole-cell lysates were used for immunoprecipitation with the indicated antibodies . Generally , 1–4 µg of the commercial antibodies was added to 1 ml of the cell lysate , which was then incubated at 4°C for 8–12 h . After addition of protein A/G agarose beads , incubation was continued for another 2 h . Immunoprecipitates were extensively washed with an NP40 lysis buffer and eluted with an SDS-PAGE loading buffer by boiling for 5 min . For in vitro GST pull-down assay , GST by itself or a GST-BakΔTM fusion protein was purified from E . coli strain BL21 ( DE3 ) ( Promega ) . 293T cell lysates were incubated with glutathione beads containing the GST fusion protein in a binding buffer ( 20 mM HEPES [pH 7 . 4] , 100 mM NaCl , 1% NP-40 , and protease inhibitors ) for 2 h at 4°C . The glutathione beads were then washed four times with the binding buffer , and the proteins associated with the beads were analyzed by SDS-PAGE and subjected to immunoblot assay with the phosphorimager . NIH3T3 stable cells grown on 8-well chamber slides were fixed with 2% ( w/v ) paraformaldehyde in PBS for 20 min , permeabilised with 0 . 2% ( v/v ) Triton X-100 for 15 min and blocked with 10% goat serum ( Gibco-BRL ) for 1 h . Primary antibody staining was performed using antiserum or purified antibodies in 1% goat serum for 1–2 h at room temperature . The cells were then extensively washed with PBS and incubated with diluted secondary antibodies in 1% goat serum for 1 h . The cells were mounted using Vectashield ( Vector Laboratories , Inc . ) . The confocal images were acquired using a Leica TCS SP laser-scanning microscope ( Leica Microsystems , PA ) fitted with a 100× Leica objective ( PL APO , 1 . 4NA ) and Leica image software . Statistical analyses were performed using unpaired t-tests . Values are expressed as mean±SEM of at least three independent experiments unless otherwise noted . A P value of ≤0 . 05 was considered statistically significant . | Autophagy ( ‘self-eating’ , lysosome-dependent degradation and recycling of the intracellular components in response to stress ) and apoptosis ( ‘self-killing’ , cells commit suicide in response to stress ) are important host defense mechanisms against viral infections . γ-herpesvirus 68 ( γHV68 ) encodes a Bcl-2 family protein , vBcl-2 , that effectively antagonizes both autophagy and apoptosis and is required for chronic viral infection and pathogenesis . However , the relative contributions of the vBcl-2-mediated evasion of autophagy and apoptosis to γHV68 persistent infection remain largely unknown . Here , we characterized a series of vBcl-2 mutants to genetically and functionally distinguish these closely related activities of vBcl-2 in vitro and in vivo . We have found that the inhibition of autophagy by vBcl-2 is important for maintaining latent infections , while the anti-apoptotic activity of vBcl-2 is largely involved in efficient viral reactivation from latency . Our findings thus reveal a novel paradigm for the vBcl-2-mediated evasion of autophagy and apoptosis during chronic viral infection , identifying a vital role for autophagy in controlling γHV68 latent infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"biology",
"virology",
"microbiology",
"molecular",
"biology"
]
| 2009 | Viral Bcl-2-Mediated Evasion of Autophagy Aids Chronic Infection of γHerpesvirus 68 |
Caspase-dependent cleavage of antigens associated with apoptotic cells plays a prominent role in the generation of CD8+ T cell responses in various infectious diseases . We found that the emergence of a large population of autoreactive CD8+ T effector cells specific for apoptotic T cell-associated self-epitopes exceeds the antiviral responses in patients with acute hepatitis C virus infection . Importantly , they endow mixed polyfunctional type-1 , type-2 and type-17 responses and correlate with the chronic progression of infection . This evolution is related to the selection of autoreactive CD8+ T cells with higher T cell receptor avidity , whereas those with lower avidity undergo prompt contraction in patients who clear infection . These findings demonstrate a previously undescribed strict link between the emergence of high frequencies of mixed autoreactive CD8+ T cells producing a broad array of cytokines ( IFN-γ , IL-17 , IL-4 , IL-2… ) and the progression toward chronic disease in a human model of acute infection .
The fate of the enormous number of apoptotic cells that derive from effector Tcells undergoing apoptosis after performing their functions during acute or chronic infections remain to be determined [1] , [2] . Phagocytosis of apoptotic cells by dendritic cells ( DCs ) leads to the processing of apoptotic cell-associated antigens and the cross-presentation of the resulting peptides on major histocompatibility complex ( MHC ) class I molecules [3]–[6] . This phenomenon seems crucial for inducing either cross-priming or cross-tolerance of CD8+T cells , based on the presence or absence of various infectious or danger signals influencing the switch from tolerogenic immature ( i ) DCs to mature ( m ) DCs with high stimulatory and migratory capacities [3]–[7] . In previous studies , we found that the proteome of apoptotic T cells includes prominent caspase-cleaved cellular proteins and that a high proportion of distinct epitopes in these fragments ( apoptotic epitopes ) can be cross-presented by DCs to a wide repertoire of autoreactive CD8+ T cells [8] . Recent reports have confirmed the role of caspase cleavage in the processing and presentation of epitopes that are derived from apoptotic cells in different models [9]–[11] . In chronic HIV infection , these autoreactive CD8+ T cells correlate with the proportion of apoptotic CD4+ T cells in vivo and are involved in establishing polyclonal T cell activation that in the long run results in generalized T cell dysfunction/depletion [8] . In addition , apoptotic cells derived from activated T cells ( in contrast to those derived from resting T cells or from non-lymphoid cells ) retain the expression of CD40 ligand ( L ) and can then condition CD40+ DCs to acquire high capacities to prime or cross-prime autoreactive T cells [12] , [13] . This mechanism is consistent with the evidence that the signals provided by CD40L+ apoptotic cells and not those provided by conventional apoptotic cells facilitate the emergence of autoreactive T cell responses to apoptotic self-antigens [12] , [13] . Successful priming of naïve CD4+ or CD8+ T cells results in the generation of both effector memory T ( TEM ) cells expressing various differentiation programs ( type-1 , -2 , -17 ) , according to the environment in which they are exposed [14]–[21] , and central memory T ( TCM ) cells that promptly proliferate and generate new waves of effector cells on demand [22]–[24] . The transcription factor T-box-containing protein expressed in T cells ( T-bet ) is the master regulator of the type-1 cell differentiation program that is associated with the production of IFN-γ , which is required for the development of protective immune responses against intracellular pathogens [15] . GATA-binding protein 3 ( GATA-3 ) controls the development of the type-2 cell lineage that is characterized by the production of IL-4 , -5 , and -13 , which is critical for immunity against helminths and other extracellular pathogens [15] . Retinoid acid-related orphan receptor ( ROR ) -γt in mice and the human ortholog RORC in humans represent the master regulators of type-17 cell differentiation that leads to the production of IL-17 , which is specifically required for protection against several types of extracellular and intracellular bacterial infections [14] , [16]–[18] . All these ( type-1 , -2 , -17 ) functions can elicit either protective or harmful effects , depending on whether they are executed by pathogen-specific or autoreactive T cells or whether the pathogen-specific are involved during an acute resolving infection or a chronic infection , respectively . Here we used the hepatitis C virus ( HCV ) infection as a human model of acute infection that generally undergoes chronic progression to verify whether CD8+ T cells that are specific for apoptotic self-epitopes have a distinct effector type-1 , -2 , or -17 phenotype , to distinguish which of them is associated with the fate of a viral infection ( recovery versus chronicity ) , and to ascertain the mechanisms whereby these responses are induced and maintained .
We analyzed longitudinally the responses of 18HLA-A2+ patients with acute HCV infection . The follow-up ranged from the onset of acute disease ( clinical onset ) to 15–24 months ( the sixth month being considered the time of conversion from an acute to a chronic infection ) . Of the 18 patients , 6 patients had a self-limited infection and 12 patients exhibited a chronic evolution of infection ( Table 1 ) . Initially , the effector responses were determined by the capacity of freshly isolated CD8+ T cells from either HLA-A2+ patients or healthy controls to form IFN-γ spots ( in an enzyme-linked immunospot [ELISPOT] assay ) within 4 to 6 hours ( h ) of contact with nine pools of synthetic apoptotic peptides ( Table S1A–C ) , eight pools of HCV genotype 1c , or genotype 2c peptides selected for their capacity to bind the HLA-A2 molecule [8] , [25] , or nine pools of overlapping peptides spanning the entire sequence of the HCV genotype 3a ( Table S2A–E ) . The different HCV genotype-related peptides were matched with the viral genotype infecting the single patients . Each peptide pool was tested in triplicate . The synthetic apoptotic peptides used were prepared according to the sequence of caspase-cleaved proteins that had been previously identified by the proteomic analyses of apoptotic T cells ( i . e . , fragments of actin cytoplasmic 1 [ACTB] , heterogeneous nuclear ribonucleo protein [ROK] , lamin B1 [LAM1] , non muscle myosin heavy chain 9 [MYH9] , vimentin [VIME] , or proteasome component C2 [PSA1] ) [8] . We found that the apoptotic ( but not the viral ) epitope repertoire recognized by IFN-γ+CD8+ TEM cells was significantly larger in patients undergoing chronic infection than in those undergoing recovery ( Fig . 1A , B and Fig . S1 ) . Interestingly , the mean number of IFN-γ spots promptly formed by CD8+ TEM cells in response to each pool of apoptotic epitopes ( but not viral epitopes ) was directly correlated with the viral ( plasma HCV-RNA ) load , thus supporting the relationship between these responses and chronic evolution ( Fig . 1C ) . None of the 21 HLA-A2+ healthy donors exhibited significant effector responses against any of the apoptotic or viral peptides ex vivo ( data not shown ) . The HLA-restriction of these responses was demonstrated both by blocking responses with an appropriate anti-class I mAb and by determining that no response was observed in HLA-A2− patients ( data not shown ) . We enumerated specific CD8+ T cells directly in the peripheral blood of HLA-A2+patients or healthy donors by using pentamers of HLA-A*0201 molecules complexed to either apoptotic ( MYH9478–486 , MYH9741–749 , VIME78–87 ) or viral epitopes ( NS31073–1081 , NS31406–1415 , Core132–140 ) that had been previously identified as the most immunogenic among all patients tested ( Fig . 1D ) . Control HLA-A*0201 pentamers complexed to a non-natural irrelevant peptide were unable to stain CD8+ T cells in all peripheral blood mononuclear cells ( PBMCs ) tested ( data not shown ) . The pentamer values were significantly higher in both in patients experiencing chronic infection and in patients with self-limited infection ( at all the time points tested ) than in 20 HLA-A2+ healthy donors ( Fig . 1E , F ) . However , in contrast to the ELISPOT assay showing frequencies of IFN-γ+CD8+Tcells specific to apoptotic peptides significantly higher in patients experiencing chronic infection than in patients with self-limited infection ( Fig . 1A ) , the total frequencies of either apoptotic or viral epitope-specific CD8+ T cells , as detected by pentamers , did not differ between patients undergoing chronic or recovery evolution at all the time points tested ( Fig . 1D–F ) . This difference may be explained by the finding that each single pentamer+ cell population can simultaneously contain ( rare ) naïve T cells , many TCM cells and several types of TEM cells with the same epitope specificity , as well as T cells with a “stunned phenotype” ( representing the reducing capacity of cells to perform effector functions ) [26] , whereas ELISPOT assay only identifies IFN-γ+ cells in our system . To detect different effector functions within the CD8+pentamer+ T cells , we analyzed the frequencies of freshly isolated CD8+pentamer+ T cells that produced a wide array of cytokines ( IL-17 , IFN-γ IL-4 , IL-2 within a few h of contact with the relevant peptides and optimal concentrations of anti-CD28 mAb , which served as a surrogate costimulatory signal . Irrelevant cytokine production was observed when either apoptotic epitope- or viral epitope-specific CD8+pentamer+ T cells of 20 HLA-A2+ healthy individuals were stimulated with this procedure ( data not shown ) . Importantly , apoptotic epitope-specific CD8+pentamer+ TEM cells promptly produced notable and sustained amounts of all the cytokines tested within a few h of contact with the relevant epitopes , much more in patients experiencing chronic infection than in those undergoing infection resolution ( Fig . 2A , B ) , in all time points tested ( Fig . S2 ) . By contrast , the virus-specific CD8+pentamer+ TEM cells produced lower amounts of the same cytokine in both categories of patients without any differences between them ( Fig . 2A , B and Fig . S3 ) . Peptide dose-response curves of cytokine-producing CD8+pentamer+ cells emphasized this difference ( Fig . 2C , D ) . Time course analyses , performed longitudinally throughout the follow-up in all patients , revealed that the frequencies of polyfunctional apoptotic epitope-specific CD8+ TEM cells were significantly higher in patients experiencing chronic infection ( Fig . 3A , B ) . These responses were sustained over time in relation to the sustained viral load ( HCV-RNA copies ) and alanineaminotransferase ( ALT ) levels only in patients who evolved into chronic infection ( Fig . 3A , B ) . Then , the majority of these cell frequencies , as well as the serum biomarkers of viral hepatitis , tended to decline considerably later in patients who evolved into chronic infection than in those resolving infection ( Fig . 3A , B ) . By contrast , no substantial difference was revealed in the time course of the virus-specific effector response between the two categories of patients ( Fig . 3C , Fig . S3 ) . Notably , the polyfunctional responses in the majority of patients were maintained by the parallel presence of different antigen-specific CD8+ T cell subsets , each of which produced a single cytokine in all time-points tested ( mixed polyfunctional populations ) ( Fig . S4A , B ) . Therefore , the minority of patients showed cells simultaneously producing significant amounts of IFN-γ and IL-17 ( type 1/17 cells ) , or cells simultaneously producing significant amounts of IL-17 and IL-4 ( type 2/17 cells ) ( Fig . 2A , B and Fig . S4A , B ) . Importantly , the frequencies of CD8+pentamer+ TEM cells promptly producing IFN-γ or IL-17 in response to the relevant apoptotic epitopes , but not to the viral epitopes ( data not shown ) , were directly correlated with the plasma viral load or the serum ALT levels ( Fig . 4A–D ) . Fresh apoptotic epitope-specific CD8+pentamer+ TEM cells promptly produced IFN-γ or IL-17 ex vivo within a few h of contact with DCs that had been pulsed with apoptotic T cells ( i . e . , through the cross-presentation mechanism ) ( Fig . 5A , B ) . The cross-presentation resulted in a marked decrease in IFN-γ or IL-17 production when apoptotic cells had been previously treated with a selective caspase-3 inhibitor ( C3I ) ( Fig . 5A , B ) . This phenomenon was confirmed in five independent patients ( Fig . 5B ) . DCs alone , despite known to endogenously express high levels of the ubiquitous ( long-lived ) cellular proteins ( vimentin , non-muscle myosin , actin , heterogeneous nuclear ribonucleoprotein , lamin B1… ) ( 14 ) , were unable to directly stimulate the related specific CD8+ T cells ( Fig . 5A , B ) . The frequencies of apoptotic epitope-specific CD8+ T cells ( but not those of viral epitope-specific CD8+ T cells [data not shown] ) correlated with the number of circulating apoptotic T cells ( Fig . 5C ) . The percentage of apoptotic T cells in PBMCs was significantly higher in patients than in the 20 healthy donors tested ( 11 . 0±7 . 7 versus 3 . 9±3 . 9; P< . 001 ) . To verify if type-17 CD8+ TEM cells specific to apoptotic self-antigens in the long run acquire functional plasticity in vivo [15] , [18] , [27]–[29] , we monitored ( from the clinical onset of infection up to 24 months ) selected patients showing a notable number of CD8+ TEM cells promptly producing IL-17 within few h of contact with the relevant apoptotic epitopes at the clinical onset . During the course of the follow-up , the frequency of type-17 CD8+ TEM cells exhibited a progressive increase , followed by the emergence of type-1/17 cells in response to apoptotic epitopes ( Fig . 6A ) . These responses were associated with both the maintenance of the type-17 transcription factor RORC and the appearance of the type-1 transcription factor T-bet ( Fig . 6B , C ) . This scenario was observed both in the 3 patients showing type-17 CD8+ TEM cells specific for the MYH9741–749 epitope ( Fig . 6 ) , and in additional 3 patients showing type-17 CD8+ TEM cells specific for different self-epitopes epitope ( data not shown ) . By contrast , representative fully polarized type-1 CD8+ TEM cells strictly maintained this phenotype throughout the follow-up period in all patients studied ( Fig . S5online ) . To determine whether antigen-specific type-17 CD8+ T cells can reprogram their phenotype and convert into type-1/17 CD8+ TEM cells in vitro ( situation which may mimic the type-17 conversion into type-1/17 phenotype in vivo ) , we used anti-CCR6 and anti-CCR4 mAbs [30] to sort IL-17–producing cells from antigen-stimulated CD8+ T cells ( purity >98% type-17 pentamer+CD8+ T cells ) ( Fig . 6D ) . These cells were then restimulated in vitro with irradiated autologous PBMCs ( acting as antigen-presenting cells [APCs] ) that had previously been pulsed with the relevant peptide in the presence of either a mixture of IFN-γ and IL-12 ( polarizing toward the type-1 phenotype ) or a mixture of TGF-β , IL-6 , IL-23 , and IL-1β ( polarizing toward the type-17 phenotype ) [14] , [31] . After 10–12 days of culture in IL-2 conditioned medium , the cells were tested for their capacity to produce both IL-17 and IFN-γ in response to the peptide plus APCs . Interestingly , CD8+ T cells that had been cultured in the presence of the type-17 polarizing cytokines maintained or increased the original type-17 phenotype , whereas CD8+ T cells that had been cultured in the presence of the type-1 polarizing cytokines switched ( in a notable proportion ) toward the type-1 phenotype ( Fig . 6D ) . To understand why the self-epitope-specific cells of patients undergoing resolution display significantly lower polyfunctional functions than patients experiencing chronic infection ( Fig . 2A–D and Fig . 3A , B ) , we performed a series of functional experiments . First , we ruled out the possibility that apoptotic epitope-specific CD8+ T cells from patients undergoing recovery expressed intrinsic defects of effector cell functions . Indeed , apoptotic epitope-specific CD8+ T cells from the two categories of patients yielded similar cytokine responses after stimulation by polyclonal mitogens ( i . e . , phorbol 12-myristate 13-acetate [PMA] and ionomycin [iono] ) ( Fig . S6 ) . Second , the majority of CD8+pentamer+ T cells ( both apoptotic epitope-specific and viral epitope-specific ) in both categories of patients were either CD45RO+CD127+ ( TCM cells ) or CD45RO+CD127− ( TEM cells ) ( Fig . S7A , B ) . This finding suggested that the CD127−CD8+ TEM cells , which promptly produced the vast array of cytokines tested within a few hours of contact with the relevant peptides ( Fig . S7C ) , were likely derived from the CD8+ TCM cells rather than naïve cells in both categories of patients in vivo . Then , we evaluated whether the apoptotic epitope-specific CD8+ TEM cells were less polyfunctional in patients undergoing infection resolution because they were conditioned by a more severe programmed death ( PD ) -1-dependent exhaustion [32] in comparison to those from patients experiencing chronic infection . We found a similar PD-1 expression in apoptotic epitope-specific CD8+ T cells between patients undergoing infection resolution and those experiencing chronic infection ( Fig . S8A ) . This result might argue against the possibility of a more severe PD-1-dependent exhaustion of apoptotic epitope-specific CD8+ TEM cells from patients resolving infection . However , we cannot exclude that the two groups of patients might express different levels of PD-1 ligands ( i . e . , in inflamed liver ) that might provide different threshold of PD-1 dependent exhaustion in vivo . PD-1 upregulation was also shown in HCV-specific CD8+ T cells without any significant difference between the two categories of patients ( Fig . S8B ) . To determine the functional capacity of PD-1 expression , we first selected PBMCs containing either viral epitope-specific PD-1+CD8+ T cells or apoptotic epitope-specific PD-1+CD8+ T cells from patients undergoing infection resolution or those experiencing chronic infection . In the presence or absence of a blocking anti-PD-L1 mAb or isotype control mAb in vitro , these cells were stimulated with the relevant peptide and anti-CD28 . After 10 days of culture in IL-2-conditioned medium , cells were double-stained with the appropriate pentamers and anti-CD8 mAb , stimulated or not with autologous APCs plus the peptide , processed for intracellular cytokine staining ( ICS ) with mAbs to IFN-γ , IL-17 , and IL-4 , and analyzed by flow cytometry . Apoptotic epitope-specific pentamer+CD8+ T cells produced notable amounts ofIFN-γ or IL-4 after 10 d of stimulation , and even more in the presence of a blocking anti-PD-L1 mAb ( Fig . 7A ) . By contrast , IL-17 production under the same conditions failed to increase but rather decreased during the 10 d of antigen stimulation in vitro , irrespective of the presence of anti-PD-L1 , emphasizing the possible functional instability of this cell population ( Fig . 7A ) . Cumulative experiments with PBMCs from a total of eight patients confirmed that the PD-1/PD-L1 blockade resulted in an increase of IFN-γ or IL-4 production by both apoptotic epitope-specific pentamer+CD8+ T cells ( Fig . 7B ) and viral epitope-specific pentamer+CD8+ T cells ( data not shown ) , whereas the production of IL-17 was not affected . Importantly , the degree of increase in the responses exhibited by both apoptotic epitope-specific pentamer+CD8+ T cells and viral epitope-specific pentamer+CD8+ T cells was virtually the same between patients undergoing infection resolution and those evolving into chronic infection ( data not shown ) . Taken together , these findings suggest the following . First , the apoptotic epitope-specific PD-1+CD8+ T cell responses are gently modulated by PD-1 because they are highly polyfunctional in patients experiencing chronic infection ex vivo . Second , the PD-1 blockade does not seem a principal cause of the decreased responsiveness exhibited by apoptotic epitope-specific CD8+ T cells from patients undergoing infection resolution in comparison to responses from patients who have developed a chronic infection , given that the degree of increase in the effector responses upon PD-1/PD-L1 blockade was very similar between the two categories of patients . However , we cannot exclude that other PD-1 ligands or the differential PD-L1 expression by inflamed liver can intervene in favoring T cell exhaustion or dysfunction in vivo , thus explaining the decreased responsiveness exhibited by apoptotic epitope-specific CD8+ T cells from patients undergoing infection resolution [33] . Finally , we hypothesized that differences in T cell receptor ( TCR ) avidity might account for the different apoptotic epitope-specific CD8+ T cell responsiveness between patients experiencing chronic infection and those undergoing infection resolution . To assess TCR avidity , we evaluated the dissociation kinetics of peptide/HLA-A*0201 pentamer binding to antigen-specific CD8+ T cells that were isolated from the two groups of patients [34] . Specifically , we stained fresh CD8+ T cells with saturating amounts of HLA-A*0201 pentamers that were complexed to either apoptotic or viral epitopes and an anti-CD8 mAb for 45 min at room temperature . Cells were then washed and incubated at 4°C with saturating amounts of an anti-HLA-A2 mAb to prevent rebinding of pentamers during the pentamer dissociation assay . The rate of decay was measured by flow cytometry at appropriate time points . We obtained linear decay plots of the natural logarithm of the normalized fluorescence versus time in all experiments performed , indicating that the pentamer decay was occurring stochastically and that the resulting pentamer staining half-lives ( t1/2 ) should be proportional to the t1/2 of respective pentamer/TCR complexes ( Fig . 8A ) . The t1/2 for apoptotic epitope-complexed pentamer staining to fresh CD8+ T cells from patients experiencing chronic infection was significantly longer than the decay of pentamer staining to CD8+ T cells from patients undergoing infection resolution ( Fig . 8A , B ) . By contrast , the t1/2 for viral epitope-complexed pentamers to CD8+ T cells did not differ between patients experiencing chronic infection and those undergoing infection resolution ( Fig . 8A , B ) . Control experiments in which HLA-A*0201 pentamers were complexed to a non-natural irrelevant peptide showed that the t1/2 for staining to CD8+ T cells was undetectable ( data not shown ) .
Here we demonstrate for the first time that the multispecificity , magnitude , and polyfunctional ( type-1 , -2 , -17 ) strength of CD8+ TEM cell responses directed to apoptotic self-epitopes were wide and robust during the acute phase of HCV infection , particularly in patients experiencing chronic progression compared with those undergoing infection resolution . The responses were directly correlated with the plasma viral load , the serum ALT levels or the number of circulating apoptotic T cells , and were then sustained over time in relation to the viral persistence . Our parallel study still in progress indicates that similar autoreactive CD8+ T cell responses in chronically infected patients are recruited in the inflamed livers ( Fig . S9 ) , are related with the signs of hepatic damage , and decrease in relation with the decline or the disappearance of the viral load upon antiviral therapy ( interferon plus ribavirin@ ) ( data not shown ) . Altogether these results suggest that strong CD8+ T cell responses against apoptotic self-epitopes arise and are maintained in HCV infection and may potentially contribute to the liver immunopathology through the production of high levels of inflammatory cytokines . Recently , several models of chronic viral infection demonstrated that virus-specific CD4+ or CD8+ T cells producing elevated levels of IL-17 correlated with either viral persistence or a wasting syndrome with a multiple organ neutrophil infiltration [20] , [35] , [36] . Currently , our data suggest that the emergence of high frequencies of mixed autoreactive CD8+ T cells producing a broad array of cytokines ( including IL-17 ) is prominent in patients undergoing chronic progression in the human model of acute HCV infection . By contrast , the frequencies of virus-specific effector cells ( producing the different cytokines analyzed ) were extremely low as compared with the apoptotic epitope-specific . Our data are coherent with the majority of studies revealing that the magnitude of HCV-specific CD8+ T cell effector responses does not correlate with the clinical or viral outcome in acute HCV infection [37] . In particular , HCV-specific CD8 T cells have been reported to express increased levels of PD-1 and an exhausted phenotype ( weak proliferation , IFN-γ production , and cytotoxicity ) [37]–[41] . Although depletion studies in the chimpanzee model are consistent with a role of CD8+T cells as primary effector of protective immunity [42] , studies in natural HCV infection were unable to find clear correlations between HCV-specific CD8+ T cell responsiveness and outcome of infection [37]–[41] , [43] , [44] . It is possible that the mechanisms that control HCV in the long term lie not exclusively on these conventional functions , but they are also displayed by some other subset of immune mediators , including HCV-specific antibodies [45]or CD4+ T cells [26] , [40] , [46] . Consistent with this finding , in vivo depletion of CD4+ T cells from HCV-recovered chimpanzees abrogates protective CD8+ T cell–mediated immunity upon rechallenge [47] , which suggests that CD4+ T cell help is required for the generation and maintenance of protective CD8+ T cells . Therefore , the viral immunological correlates of infection should be detected by multiparametric analyses ( antibody , CD4+ , CD8+ responses… ) rather than individual analysis that may underestimate the multiple immunological variables related to infection outcome . In this respect , our study suggests that the apoptotic epitope- more than the virus-specific CD8+ cell responses discriminate patients with different infection outcome . The observation that cross-presentation of apoptotic T cells by DCs requires caspase-dependent cleavage of apoptotic self-antigens to promptly activates specific CD8+ TEM cells ex vivo indicates that this mechanism might be operative in the induction of the related polyfunctional autoreactive responses in vivo . This possibility is emphasized by the finding that the frequencies of apoptotic epitope-specific CD8+ T cells correlated with the number of circulating apoptotic T cells . Consistently with our previous observations ( 14 ) , cross-presentation of apoptotic cells plays a key role in activating autoreactive CD8+ T cells , as caspase-dependent cleavage of cell-associated ( long-lived ) proteins ( such as vimentin , non-muscle myosin , actin , heterogeneous nuclear ribonucleoprotein , lamin B1… ) is required to efficiently target the related fragments to the processing machinery of DCs . By contrast , live DCs alone , despite known to express the whole form of the same ubiquitous ( long-lived ) cellular proteins ( 14 ) , are unable to stimulate the related specific CD8+ T cells by direct presentation mechanism , likely because they do not possess the caspase-cleavage program required for the presentation of these proteins ( 14 ) . Collectively , these data suggest that these autoreactive CD8+ T cells may perform their functions through the by-stander effect of the pro-inflammatory cytokines upon cross-presentation of apoptotic cells rather than by the direct killing of cells endogenously expressing the related self-antigens . The strong production of IFN-γ and IL-17 may favor the triggering of recruitment of inflammatory cells , which contribute to the immunopathology [48]–[50] . Our study provides a possible explanation for why the enormous expansion of activated T cells , during persisting viral infections , is only minimally attributable to virus-specific T cells [8] . Inflamed tissues ( including the HCV-infected liver ) are generally infiltrated by several billions of activated lymphocytes and the rate of apoptotic cells derived from them by far exceeds that originated by the turn-over of epithelial cells ( i . e . , hepatocytes ) [51] . The demonstration that apoptotic cells derived from activated T cells ( in contrast to those derived from epithelial cells ) are CD40L+ and then condition CD40+ DCs to prime T cells [12] , [13] , suggest that they are the most important source of apoptotic self-antigens capable to cross-prime CD8+ T cell responses in an inflamed microenvironment . However , we cannot exclude that also apoptotic hepatocytes may amplify this phenomenon in an inflammatory context , because they might potentially generate the same caspase-cleaved antigenic fragments described in apoptotic T cells [8] , and be cross-presented by DCs . Recent data have clearly stressed the importance of infections in inducing and maintaining autoimmunity [52] . In particular , the initial emergence of apoptotic antigen-specific T cells in acute HCV infection may be dependent on virus-specific T cells that can provide the first waves of apoptotic substrates , upon performing their effector function . This mechanism may be maintained in patients evolving towards the viral persistence , and be further amplified by the apoptotic antigen-specific T cells themselves providing further waves of apoptotic antigens . Additional studies , even in appropriate experimental models , are required to ascertain the role of these autoreactive responses in the chronic evolution of infection . Moreover , it could be of interest to investigate if an expansion of mucosal associated invariant type-17 CD8+ T cells may participate in the high IL-17 production [53] , as well as if other additional cytokines including IL-10 [54] may increase the polyfunctionality of apoptotic epitope-specific CD8+ T cells . IL-17 production can account for the transient intra-hepatic infiltration of neutrophils found only in the early phase of acute HCV infection [55] . Then , these responses timely decline likely through the simultaneous presence of autoreactive type-1 , -2 , and -17 CD8+ TEM cells , which may regulate each other , and even the capacity of type-17 CD8+ TEM cells to express a certain degree of plasticity [15] , [18] , [27]–[29] , and to convert to type 1/17 CD8+ TEM cells . Therefore , the environmental setting during an acute inflammatory disease seems to be addressed to guarantee the coexisting polarization of type-1 , -2 , -17 , -1/17 CD8+ TEM cells , and even type-2/17 CD8+ TEM cells , likely to limit excessive damage by fine-polarized type-1 or type-17 CD8+ TEM cells . In this context , it is intriguing the observation that the polyfuctional autoreactive CD8+ T cells express high PD-1 levels in vivo and are limited only partially by the inhibitory PD-1 capacity in vitro . A possible explanation of this is that these autoreactive CD8+ T cells are primed later and thus submitted to a less prolonged antigenic stimulation than the virus-specific , that in contrast show a profound exhaustion in patients with HCV infection [32] . Recently , the persistent antigenic stimulation has been demonstrated to cause down regulation of T-bet , which results in more severe exhaustion of virus-specific CD8+ T cells [56] . However , the tuning of autoreactive CD8+ TEM cell functions by PD-1 , as well as the high frequencies of apoptotic epitope-specific CD8+ TEM cells producing IL-4 , may contribute to limit excessive functional responses over time [19] , [54] , [57] , [58] . In support of this hypothesis , our study in patients with long-term chronic HCV infection demonstrates that liver-infiltrating CD8+ TEM cells specific to apoptotic self-epitopes produce levels of cytokines significantly lower than patients with acute hepatitis ( data not shown ) . An important facet of our findings is that they demonstrate a link between the TCR avidity of autoreactive CD8+ T cells and the difference in the responsiveness of apoptotic epitope-specific CD8+ T cells exhibited by patients experiencing chronic infection and those undergoing infection resolution . The dissociation kinetics of peptide/HLA-A*0201 pentamer binding to antigen-specific CD8+ T cells clearly demonstrated that the t1/2 for apoptotic epitope-complexed pentamer staining to CD8+ T cells from patients experiencing chronic infection was significantly longer than the decay of pentamer staining from patients undergoing infection resolution . The t1/2 for pentamer staining was detected on freshly isolated CD8+ T cells , suggesting that TCR avidity measured by this system likely reflects what occurs in vivo . In the original study , this methodological approach made it possible to postulate that T cells with TCRs that bind peptide/MHC complexes for a longer duration are selectively preserved in comparison to T cells that express TCRs with lower avidity [34] . Recent studies suggested that in response to different microbial infections , initially naïve T cells with a wide range of avidity are efficiently recruited and expanded [59] , [60]; subsequently , those with lower avidity undergo premature contraction , whereas those with higher avidity are selected because of a more prolonged expansion and correlate with protection [59] , [60] . Our results provide an additional challenge to this model , and demonstrate that the TCR avidity of autoreactive CD8+ T cells specific for apoptotic self-epitopes was significantly higher in patients undergoing chronic infection than in those resolving infection . The selection of the autoreactive CD8+ T cells with higher avidity likely occurs because of a sustained stimulation by apoptotic antigens [8] . By contrast , lower avidity CD8+ T cells in the presence of weaker stimuli would undergo rapid contraction , as seen in the peripheral blood of patients with self-limited HCV infection . The viral persistence may provide the conditions that influence the availability of sustained apoptotic antigenic stimuli . However , our model does not exclude the possibility that cross-reactive CD8+ T cells , even expressing dual TCR [61] , may intervene in this process . Finally , our results may provide an important platform for the design of innovative therapeutic strategies to re-engineer protective immune responses in persisting infections . In addition , further studies will ascertain whether polyfunctional CD8+ T cells that are specific to apoptotic epitopes could predict chronic infection in other acute ( i . e . , HBV or HIV ) infections that develop viral persistence . The detection of these autoreactive CD8+ T cells may also be relevant in determining whether the contraction or the quality variation of the polyfunctional responses can be used as biomarkers to verify the protective effects of conventional or innovative antiviral therapies [62] , [63] .
The study cohort included 18HLA-A2+patients with acute HCV infection ( 5 women , 13 men , median age 34 years , range 22–57 years ) , according to the ethical guidelines of the 1975 Declaration of Helsinki and priori approval by the Ethics Committee of the Italian National Institute of Health: written informed consent was obtained from all patients . Diagnosis of acute HCV infection was based on ( 1 ) high levels of serum ALT; ( 2 ) seroconversion assessed by third generation enzyme linked immunosorbent assay , or anti-HCV positivity at the time of the diagnosis with an anti-HCV negative test in the previous 12 months; ( 3 ) presence of HCV-RNA in at least the first serum sample , and ( 4 ) sudden onset of liver disease symptoms . Alternative causes of acute hepatitis , such as other viral infection , autoimmunity , alcohol , drugs , and toxins were excluded . Patients with concomitant immunological disorders or with HIV coinfection were also excluded from the study . PBMCs were isolated and T cell clones were generated as previously described [64] . CD8+ T cells were purified from PBMCs by positive selection coupled to magnetic beads ( MiltenyiBiotec ) as previously described [54] . Flow cytometry analysis demonstrated >99% CD8+ cells in the positively purified population and <5% in the CD8-depleted population . To purify antigen-specific type-17 CD8+ cells , PBMCs were stimulated with the relevant peptide plus anti-CD28 ( 4 µg/ml ) ( BD Pharmingen ) . Then , cells were stained with allophycocyanin ( APC ) -labeled anti-CCR6 ( R&D System ) and phycoerythin-cyanine 7 ( PE-Cy7 ) -labeled anti-CCR4 ( BD Pharmingen ) and processed with FACSAria ( Becton Dickinson ) to sort CCR6+CCR4+ cells: >98% of these cells both produced IL-17 in response to the relevant peptide and were susceptible of staining with the relevant pentamers , as detected below . Spontaneous apoptosis of PBMCs from patients was determined by staining with Annexin-V ( ApoAlert Apoptosis Kit , Clontech Laboratories Inc ) , propidium iodide ( PI ) ( Sigma-Aldrich ) and PE-Cy7-labeled anti-CD3 ( BD Pharmingen ) before and after 18 h incubation at 37°C . Immature ( i ) DCs were derived from peripheral monocytes that had been purified by positive selection with anti-CD14 mAb coupled to magnetic beads ( MiltenyiBiotec ) . CD14+ cells were incubated for 5 days in RPMI 1640 medium containing 5% FCS , 2 mM glutamine , 1% nonessential amino acids , 1% sodium pyruvate , 50 µg/ml kanamycin ( Gibco BRL ) , 50 ng/ml rGM-CFS ( Novartis Pharma ) , and 1000 U/ml rIL-4 ( gently provided by A . Lanzavecchia , Bellinzona , CH ) . Mature DCs were obtained by a 40-h stimulation of iDCs with soluble rCD40L molecules ( Alexis Biochemicals , Alexis Corporation ) . The definition of monocyte-derived DCs was based on their surface phenotype profile by staining with anti-CD14 , anti-CD86 ( Caltag Laboratories ) , anti-CD1a , anti-CD1c , anti-CD11c , anti-CD32 , anti-CD80 ( BD PharMingen ) mAbs , Annexin-V ( ApoAlert Apoptosis Kit , Clontech Laboratories Inc ) , PI ( Sigma-Aldrich ) , and the appropriate secondary labeled antibodies ( BD PharMingen ) . Highly purified CD8+ T cells ( 1×105 ) from PBMCs were stimulated for 4–6 h with nine independent pools of apoptotic peptides ( Table S1A–C ) , eight independent pools of viral-peptides ( genotype 1b ) , eight independent pools of viral-peptides ( genotype 2c ) [8] , [25] , or nine pools of overlapping peptides spanning the entire HCV genotype 3a ( Chiron Mimotopes ) ( Table S2A–E ) , and irradiated autologous CD8-depleted PBMCs , used as APCs , and tested by an ELISPOT assay , as described [8] . Each peptide pool contained 5 µg/ml of each single peptide . PBMCs were incubated with APC-labeled–HLA-A*0201 pentamers ( complexed to vimentin78–87 [LLQDSVDFSL] , non-muscle myosin478–486 [QLFNHTMFI] , or non-muscle myosin741–749 [VLMIKALEL] peptide ) for apoptotic epitopes and APC-labeled–HLA-A*0201 pentamers ( complexed to HCV-NS31073–1081 [CINGVCWTV] , HCV-NS31406–1415 [KLVALGINAV] or HCV-Core132–140 [DLMGYIPAV] peptide ) for viral epitopes ( ProImmune Limited , Oxford , United Kingdom ) , in FACS buffer ( PBS 1× , 2% human AB serum ) for 10 min at 37°C , followed by washing and further incubation with APC-Cy7-labeled mAb to CD8 ( BD Pharmingen , San Diego , CA ) , fluoresceinisothiocyanate ( FITC ) -labeled anti-PD-1 ( BD Pharmingen ) , PE-labeled anti-CD127 ( BD Pharmingen ) , FITC-labeled anti-CD45RO ( Caltag Laboratories , Burlingame , CA ) for 20 min at 4°C . Negative controls were obtained by staining cells with an irrelevant isotype-matched mAb . Cells were washed , acquired with a FACSCanto flow cytometer and FACSDiva analysis software ( Becton Dickinson ) or FlowJo software version 7 . 5 . 5 ( Tree star , Inc . San Carlos , CA ) . PBMCs were stained with pentamers and mAb to CD8 , and then stimulated with or without different concentrations of the corresponding uncomplexed peptide ( ProImmune Limited ) plus anti-CD28 mAb ( 4 µg/ml ) ( BD Pharmingen ) , or with PMA ( 50 ng/ml ) plus ionomycin ( 1 µg/ml ) ( Sigma Aldrich , Milan , Italy ) , for 6 h at 37°C . At the 2nd h , 10 µg/ml Brefeldin A ( Sigma-Aldrich ) was added . Cells were washed , fixed and permeabilized using Cytofix/Cytoperm solution ( BD Pharmingen ) at 4°C for 20 min , re-washed with Perm Wash Buffer ( BD Pharmingen ) , and intracellularly stained with different combinations of Alexa Fluor 488-labeled anti-IL17A ( eBioscience San Diego , CA ) , PE-labeled anti-IFN-γ , PE-labeled anti-IL-2 , FITC-labeled anti-IL-4 ( BD Pharmingen ) , PE-labeled anti-RORC ( eBioscience ) or purified anti-T-bet ( Santa Cruz Biotechnology Santa Cruz , California ) for 30 min at 4°C . When stained with unlabeled specific antibody to detect T-bet , cells were washed and stained with the appropriate secondary FITC-labeled antibody . Cells were washed , acquired with a FACSCanto flow cytometer and FACSDiva analysis software ( Becton Dickinson ) or FlowJo software ( Tree star ) . Cloned CD8+CD95+ T cells ( 10–100×106 ) were incubated in the presence or absence of 14 µg/ml C3I ( Z-DEVD-FMK ) , or a negative caspase control ( K , Z-FA-FMK ) ( BD Pharmingen ) for 1 h at 37°C in a 24-well plate . Then , cells were induced to apoptosis by incubation with 500 ng/ml anti-Fas ( anti-CD95 mAb [clone CH11] , Upstate Biotechnology ) for at least 6 h . Apoptotic cells were determined by staining with Annexin-V ( ApoAlert Apoptosis Kit , Clontech Laboratories Inc ) , PI ( Sigma-Aldrich ) and flow-cytometry analysis . PBMCs were double-stained with pentamers and mAb to CD8 and cultured with iDCs that had been pulsed or not with apoptotic cloned T cells . After 6–8 h , cells were tested for their capacity to produce IL-17 and IFN-γ by ICS as described above . Cells were washed , acquired with a FACSCanto flow cytometer and analyzed with FACSDiva analysis software ( Becton Dickinson ) or FlowJo software ( Treestar ) . PBMCs were stained with saturating amounts of APC-labeled-HLA-A*0201 pentamers and APC-Cy7-labeled-CD8 ( BD Pharmingen ) for 45 min at room temperature [34] . Then , cells were washed three times with buffer ( 2% FCS , 0 . 01 sodium azide in PBS ) and resuspended in 500 µl of buffer with saturanting amounts of mAb to HLA-A2 ( BB7 . 2 , ATCC ) . At various time points ( 0 , 30 min , 1 h , 2 h and 3 h ) , an aliquot cells was washed and the fluorescence intensity was determined by flow cytometry analysis . Double staining using an anti-human TCRα/β ( BD Pharmingen ) and pentamers was performed in parallel to normalize pentamer fluorescence against the expressed TCR . The values were then normalized to percent of the total fluorescence at the initial time point and plotted on a logarithmic scale . t1/2 are determined by calculating the ( ln2 ) /mean slope value of plots of the natural logarithm ( ln ) of the pentamer fluorescence normalized for the TCR fluorescence . The slope is equivalent to ln ( Fa/Fb ) /t , where Fa is the normalized fluorescence at the start of the interval , Fb is the normalized fluorescence at the end of the interval , and t is the length of the interval ( minutes ) . PBMCs were incubated for 10 days at 37°C with specific peptides ( apoptotic or viral peptides ) , human rIL-6 ( 50 ng/ml ) , rIL-1β ( 10 ng/ml ) , rIL-23 ( 50 ng/ml ) and rTGF-β ( 10 ng/ml ) ( R&D Systems ) for the Th17 cell polarization . For the Th1 cell polarizing condition , PBMCs were antigen-stimulated in the presence of recombinant human rIL-12 ( 10 ng/ml ) and rIFN-γ ( 100 U/ml ) ( R&D Systems ) . Recombinant IL-2 was added on day 4 of culture ( 50 U/ml ) . On day 10 , cells were stained with surface antibodies , pentamers , anti-IL-17A , anti-IFN-γ , anti-IL-2 and anti-IL-4 mAbs . Cells were washed , acquired with a FACSCanto flow cytometer and analyzed with FACSDiva analysis software ( Becton Dickinson ) or FlowJo software ( Tree star ) . All statistical analyses were performed with Prism 4 ( GraphPad ) software using nonparametric Spearman's correlation test , nonparametric Mann-Whitney U-test for unpaired data and Wilcoxon test for paired data . The differences were considered significant at P<0 . 05 . actin cytoplasmic 1 [ACTB] P60709 heterogeneous nuclear ribonucleoprotein [ROK] P61978 lamin B1 [LAM1] P20700 non muscle myosin heavy chain 9 [MYH9] P35579 vimentin [VIME] P08670 proteasome component C2 [PSA1] P25786 | The emergence of a large population of mixed polyfunctional ( type-1 , -2 , -17 ) CD8+ T cell effector responses specific for apoptotic T cell-associated self-epitopes rather than the dysfunction or altered quality of virus-specific CD8+ T cells is associated with the progression toward chronic disease in the human model of acute HCV infection . The chronic evolution is associated with the selection of autoreactive CD8+ T cells with higher T cell receptor avidity , whereas those with lower avidity undergo prompt contraction , as seen in patients undergoing infection resolution . We suggest that these autoreactive responses are secondary to the viral persistence and can participate to the HCV-related immunopathology . This data has implications for the prognosis and therapy of infections undergoing chronic evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"immune",
"cells",
"immune",
"activation",
"antigen-presenting",
"cells",
"immunity",
"to",
"infections",
"immunology",
"adaptive",
"immunity",
"hepatitis",
"immune",
"defense",
"hepatitis",
"c",
"infectious",
"diseases",
"hiv",
"t",
"cells",
"biology",
"immune",
"response",
"antigen",
"processing",
"and",
"recognition",
"clinical",
"immunology",
"immunity",
"viral",
"diseases",
"autoimmunity"
]
| 2012 | Polyfunctional Type-1, -2, and -17 CD8+ T Cell Responses to Apoptotic Self-Antigens Correlate with the Chronic Evolution of Hepatitis C Virus Infection |
Distribution networks—from vasculature to urban transportation pathways—are spatially embedded networks that must route resources efficiently in the face of pressures induced by the costs of building and maintaining network infrastructure . Such requirements are thought to constrain the topological and spatial organization of these systems , but at the same time , different kinds of distribution networks may exhibit variable architectural features within those general constraints . In this study , we use methods from network science to compare and contrast two classes of biological transport networks: mycelial fungi and vasculature from the surface of rodent brains . These systems differ in terms of their growth and transport mechanisms , as well as the environments in which they typically exist . Though both types of networks have been studied independently , the goal of this study is to quantify similarities and differences in their network designs . We begin by characterizing the structural backbone of these systems with a collection of measures that assess various kinds of network organization across topological and spatial scales , ranging from measures of loop density , to those that quantify connected pathways between different network regions , and hierarchical organization . Most importantly , we next carry out a network analysis that directly considers the spatial embedding and properties especially relevant to the function of distribution systems . We find that although both the vasculature and mycelia are highly constrained planar networks , there are clear distinctions in how they balance tradeoffs in network measures of wiring length , efficiency , and robustness . While the vasculature appears well organized for low cost , but relatively high efficiency , the mycelia tend to form more expensive but in turn more robust networks . As a whole , this work demonstrates the utility of network-based methods to identify both common features and variations in the network structure of different classes of biological transport systems .
Transport networks—which are a subset of complex networks commonly studied using methods from network science [1]—represent structures throughout which entities are transferred between different regions of the system . Such networks are prevalent in both the engineered and natural world . One classic example is an urban transit system , where stations correspond to network nodes and where physical routes , such as roads or railways , correspond to network edges along which traffic can flow [2–4] . Examples from biology include vasculature networks , which allow for the distribution of blood to various parts of an organism , or collections of neurons and larger-scale brain areas connected by physical pathways , which allow for the transmission of electrical signals throughout the network . Importantly , all of these networks are spatial in the sense that the nodes and edges are embedded into real space [5] . The physical nature and spatial embedding of these systems often imposes costs associated with building and maintaining network infrastructure , and these costs can in turn constrain the network’s topology [6] , for example , by making spatially long-distance connections improbable or by constraining the density of connections in the network . On the other hand , pressures that may compete against wiring minimization include those driving network efficiency and robustness . Tradeoffs between these desirable network features can vary across systems , and a quantitative understanding of such tradeoffs may directly inform the design of optimal spatial transport networks [7–14] . In this study , we focus on characterizing the network organization of biological distribution systems , which are indeed subject to the competing pressures of maintaining low material costs while achieving high efficiency and robustness . However , not all biological distribution systems are the same . Some can constitute an entire organism—such as mycelial fungi—in which the physical cords making up the organism can be represented as edges in a network , and in which branching , fusion , or end points among those cords can be represented as nodes in a network [15–19] . Past work has shown that these systems appear to strike an intermediate balance between cost and efficiency , enabling the organism to achieve competing goals . In addition , their network architecture can change and adapt over time in ways that can strengthen beneficial features , such as increased formation of cross-links and loops that aid in robustness to damage and allow for parallel flow pathways [15 , 16 , 18 , 20–24] . Alternatively , distribution systems can form only a small part of a larger organism—as is the case with cortical vasculature—in which a pial network on the surface of the brain routes blood to penetrating arterioles , that in turn supply the underlying tissue [25 , 26] . This vasculature system can also be modeled as a network , in which edges represent surface vessels , and nodes represent branching points among vessels or penetrating arterioles [27] that connect to and source an underlying three-dimensional system of microvessels . Previous investigations have found that the pial network of the middle cerebral artery in rodent brains forms a robust foundation of interconnected loops that can withstand damage and re-route flow in the presence of occlusions [27 , 28] . In both the mycelial and vasculature systems studied here , planar distribution networks—whose nodes are distributed in two-dimensional space—must transport fluid and nutrients efficiently in the face of constraints on the total amount of material that they can support . But in spite of these commonalities , the two networks exist and have evolved in inherently different environments , which may directly impact the sorts of evolutionary pressures that the different networks experience . For example , the main role of the surface vasculature network is to transport blood to tissue that is part of a larger organism . On the other hand , for mycelial fungi , the network is the organism itself and is not necessarily constrained to serve or occupy a set region of space . Moreover , brain vasculature resides in a controlled environment within the confines of the skull , whereas mycelial networks live in and must adapt to often unprotected and varied environmental conditions [17 , 29–31] . In addition , while directed flow and growth are known to be important in both vasculature and fungi , the mechanisms behind long-distance transport of nutrients and maturation are different in the two systems [17] . However , most past work has focused on characterizing these two classes of distribution networks separately from one another ( see , for example , [25–27] and [15 , 17 , 20] ) , with little attention paid to how the varying habitats , function , and development of vasculature vs . mycelial systems may directly affect their network architectures . Thus , in contrast to prior studies , here we carry out a comparative analysis to make progress in understanding the similarities and differences in network organization across these two classes of natural transport systems . To quantitatively compare and contrast the mycelial fungi and vasculature , we consider the binary network of nodes and edges that represents the structural backbone of connectivity underlying each system . We then utilize a set of methodologies rooted in network science to examine the organization of those networks . We first investigate the mycelial and rodent brain vasculature systems using a set of measures that probe different aspects of network topology , from those that evaluate local connectivity to those that assess hierarchical organization . We find that the two types of distribution networks exhibit some similarities , but also many differences in regard to certain network features across varying topological and spatial scales , despite the fact that both kinds of systems yield highly constrained , planar network layouts . In the second part of our study , we carry out a network-based analysis that purposefully takes into account the importance of the spatial embedding of the distribution systems , and we examine network correlates of properties that are more directly relevant to the function of these systems . In particular , by comparison to a set of spatially-informed null models , we examine relationships between network measures of wiring length , transport efficiency , and robustness , and compare and contrast the associated tradeoffs across the different network classes . This analysis uncovers some clear distinctions in how the two types of distribution networks balance competing goals , which we hypothesize may be markers for the systems’ functions or reflect the environment in which those functions must be performed . Taken together , our work demonstrates the utility of network science for characterizing and distinguishing different kinds of transport networks in biology , and underscores the fact that different systems can exhibit resemblances as well as important variations in their network structure . We hope that this study can inform future modeling and empirical work in this area , and that it provides a useful entry for more complex comparative investigations that consider both connectivity as well as edge weights ( vessel/cord radii ) , in systems where this information is known .
In order to study the vasculature and mycelia using methods from network science , we first construct an undirected and unweighted adjacency matrix A for each network under consideration , with elements defined as A i j = { 1 if there is an edge between nodes i and j , 0 otherwise . ( 1 ) This yields a binary connectivity matrix that captures the topological structure of the underlying system . For the mycelial systems , the physical cords making up the organism are assigned to edges of the network represented in the adjacency matrix A , and the branching , fusion , or end points among those cords ( as well as the inocula ) are represented as nodes in the network . The networks describing the mycelia are 2-dimensional and planar , and all nodes have corresponding spatial coordinates in 2D . An example of an unweighted network from P . I . is shown in Fig 1A . Red points correspond to network nodes and gray lines correspond to the location of edges . See Fig . A in the S1 Text for additional examples of the unweighted network representations from the other species of mycelial fungi . For the vasculature systems , edges of the network represented in the adjacency matrix A correspond to the location of vessel segments through which blood flows , and nodes are either junctions where surface vessels merge , or are penetrating arterioles that dive into and source the underlying neocortical microvasculature . In this study , we include both types of nodes in the network representation so that we can account for the vessels forming the surface backbone of loops as well as the many vessels that branch off the backbone and lead to penetrating arterioles , both of which are important functionally [27] . Furthermore , modeling the full network rather than just the subset of edges in loops allows for a fairer comparison with the mycelial systems , and a more complete quantification of network wiring lengths and other network measures involving pathways of edges . The vasculature networks are also 2-dimensional and planar , and all nodes have 2D spatial coordinates . We note that vessel radii information was not available for this data set , so we therefore focus on characterizing and comparing the unweighted network connectivity . An example of the entire vasculature network of the middle cerebral artery from a rat brain is shown in Fig 1B . Red nodes are branching points among the surface vessels and blue nodes are penetrating arterioles; gray lines represent the locations of edges . See Fig . B in the S1 Text for an example vasculature network from a mouse brain . In Table 1 , we give the number of nodes ( N ) and number of edges ( M ) in each network . It is important to note that because transport systems are physical in nature , there is often additional information that can be incorporated into analyses due to the embedding of their networks into real space . This is useful to consider , because in such spatial networks , the connectivity and the geometric layout of network elements are likely to be interwined and constrained by one another . The networks studied here are embedded in two dimensions , so a given node i has a spatial coordinate , {xi , yi} , which allows the Euclidean distance between nodes i and j , Dij , to be computed . In reality , the cords or vessels also have some radius rij , and one could thus construct a weighted network representation of the system , in which edges of the corresponding adjacency matrix are weighted by a function of both length and radius [15 , 17 , 19–21 , 25 , 26] . However , since cross-sectional area information was not available in the rodent vasculature experiments , we consider only the edge length in this study . Although the spatial coordinates of nodes and edge lengths do not encompass the full geometric structure of the underlying system , they still capture important information about how the system is laid out in space . Furthermore , in conjunction with the network connectivity , incorporating spatial information about inter-node distances allows one to estimate quantities such as wiring length and more relevant assessments of network efficiency ( see Network measures for details ) . These kinds of metrics complement and augment network-based analyses that consider only network topology with no regard to spatial embedding . However , we do acknowledge that radial information is crucial for understanding distribution systems , and this will be important to include in future work . Network science provides a powerful mathematical foundation for the principled study of diverse complex networks . Within this framework , there are many different methods and metrics that one can use to quantify network properties . Below ( and in more detail in the S1 Text ) , we describe the measures utilized in this study . In order to place various measures of network organization into context , or to compare and contrast empirical networks of varying size , it is useful to normalize network properties computed on a given network G by their corresponding values in null model networks that have the same number of nodes as G . Null models are often canonical networks that preserve certain features of the original network ( e . g . , density ) or are networks that have idealized or extreme topological and/or spatial organization ( e . g . , lattice-like , random , or fully connected ) [49] . In order to make statistical comparisons between the two types of distribution networks—mycelial fungi vs . vasculature—we first group all mycelial networks together and all vasculature networks together , and use two-sample t-tests to compare network measures between the two groups . Statistically significant differences between the two groups with respect to the mean value of a particular measure ( which we denote using an overbar symbol and subscripts “F” or “V” for fungi and vasculature , respectively ) are indicated by a p-value < 0 . 05 . Because there are only a small number of networks in the individual subgroups of fungi ( i . e . , P . I . , P . V . 1 , P . V . 2 , R . B . ) and vasculature ( i . e . , rat , mouse ) , rather than making comparisons at the level of each subgroup—which would not permit a robust statistical analysis—we use the method of first grouping the data and then examining overall differences or similarities between the broad classifications of network type .
While there are some general similarities between the vasculature and mycelial networks , differing habitats or environmental pressures could lead to distinct structural features . Indeed , we quantified some of these distinctions in the previous section , Characterization of network architecture with graph-theoretical measures . Here we conduct further network-based analyses that more explicitly take into account the spatial embedding of these systems , and we compare the vasculature and fungi using a set of network properties motivated specifically by the functional requirements of biological distribution networks .
There are a number of methodological considerations—especially concerning the construction of the network representations of the vasculature and fungi—that are useful to comment on . First , it is crucial to point out that we have used a simplified network representation of the distribution systems in that information about cord or vessel radii was neglected . The main reason for this was that these details were unavailable for the vasculature . However , it is known that tube area is important functionally , and can significantly alter transport in distribution networks since it affects flow resistance . Including this additional physical property would thus yield a more complete analysis , both in terms of quantifying network construction costs as well as measuring network properties such as physical efficiency . Indeed , past studies on mycelial networks have found that the presence of radial thicknesses confers improved transport characteristics to the system [16 , 18 , 20] . It would be interesting to take this into account in future work to investigate whether the inclusion of radial information in the weighting of the network reduces or enhances the differences found between the vasculature and mycelial systems . It is also important to point out that the definition of wiring length employed in this study assumes a linear relationship between edge length and edge “cost” . In reality this might not be the case . For example , longer connections might be disproportionately more expensive to build and maintain , or there might be an offset cost to create connections at all . Furthermore , we note that the network property of physical efficiency quantifies transport capabilities only in terms of shortest paths , and in the case studied here , assumes bi-directionality of flow; an understanding of transport along indirect walks is also relevant in distribution systems , as is the notion of directed transport that allows for the movement of nutrients through long distances . The extension of traditional network measures and null models to include this type of information may lead to considerable insight into structure-function relationships in biological transport networks more generally . For example , network communicability [63] naturally includes information on walks of all lengths between pairs of nodes , and could be one valuable candidate for further analyses . In addition , it is worth noting that the completeness of the network representations are also subject to the resolution at which fine-scale edges can be traced or digitized . We also note that in the network representation of the mycelial networks , the food source itself was considered as an imposed ‘super-node’ that connects to the cords incident on its boundary . Since the mycelial networks grow outward from the inocula , the density of edges around that node may be higher compared to more peripheral regions of the network , which could introduce some heterogeneity in the architecture . In terms of Rentian scaling in particular , we would then expect partitions that surround the ‘super-node’ to have a higher number of boundary-crossing edges . These partitions might be a cause for some spread or disruption in the spatial scaling relationship , or perhaps bias the exponent towards higher values . However , at least upon visual inspection of the examples of log m vs . log n in Fig 5 and Fig . H in the S1 Text , this does not appear to be a salient effect . Since we sample the network numerous times to compute the scaling , most of the partitions will not include the inocula , and this likely prevents the ‘super-node’ from dominating the scaling . However , it is important to be aware of the issues that the ‘core-like’ structure in the network can induce . In regard to the vasculature networks , it is important to remark on the fact that nodes correspond to either branching points among surface vessels or places where penetrating arterioles descend into the underlying tissue . Furthermore , many of the penetrating arterioles originate from offshoots that stem from the network of surface vessels and that terminate at the end of a surface edge in a single point [27] . Though this is a meaningful organizational feature of the vasculature that we chose to include in our primary examination , it is still interesting to consider whether results hold if we consider a “reduced” network that does not contain the stubs leading to single penetrating arterioles . To this end , we carried out an analysis comparing the complete vasculature networks to reduced versions in which isolated penetrating arterioles ( and the corresponding surface edges connecting to them ) were removed , and we also compared the reduced vasculature networks to the mycelial networks in order to test the robustness of the main results of this study ( full details are provided in the S1 Text ) . In summary , we found that though a number of network properties differ between the full and reduced vasculature networks , these differences are slight and do not change the conclusions or significance of the main results in terms of how the organization of the vasculature networks compare to that of the mycelial networks . Aside from those already mentioned , there are several other directions for future work . For example , one could try to further understand the observed differences between the structure of the vasculature and fungi , but perhaps more importantly , investigate the causes and actual functional consequences of those differences . We suggested that some of the observed variabilities in network organization may be due to the different environmental habitats or function of the two kinds of transport systems . However , to more solidly establish this will certainly require both further experimental work , as well as more extensive theoretical models . For example , it would be interesting to build off of prior work [9 , 14 , 16 , 19 , 20 , 25 , 27 , 29–31 , 55 , 61 , 64] and continue to design experiments and construct models that describe how a distribution network evolves in a physically unbounded and dynamically variable environment , vs . in a spatially constrained and more regulated environment , and investigate how these differences affect network adaptation and the resulting network architecture . Furthermore , while this work considers only the static network structure , an improved analysis should also consider the relation of that structure to experimental data on flow throughout the network , or should investigate the functional consequences of certain structural network properties using models of flow and transport . Understanding how transport in turn affects network development is also important [16 , 61] . Another intriguing direction would be to further examine the properties of distribution networks that are embedded not in 2-dimensions , but in 3-dimensions [26 , 65] , where relationships between topology and geometric structure or spatial layout may be more complex . In this study , we used network-based methods to quantitatively compare and contrast the structural skeleton of mycelial networks and vasculature networks from the surface of rodent brains . Both systems are two-dimensional , planar , biological transport networks , whose organization arises through natural processes in the absence of global rules for network construction . However , despite similarities , these networks have different developmental and transport mechanisms , and exist in distinct environments . While such differences are likely to differentially impact the network structure of vasculature and mycelia , little work has focused on comparative analyses of network traits across the two systems . Here we characterized the vasculature and mycelial networks using both classic network metrics , as well as a collection of measures inspired by the functional requirements and physical constraints of and on these systems . We quantified and described both conserved features of network organization across the two types of transport networks , as well as differences in topological structure and in tradeoffs between network correlates of wiring length , efficiency , and structural robustness . This work illustrates the utility of network science to uncover common organizational properties across different kinds of networks , as well as variabilities that may be important to or reflect differences in the evolution , functional requirements , or capabilities of different systems . | Distribution networks such as vasculature systems or urban transportation pathways are prevalent in our world . Understanding how different kinds of transport systems are organized to allow for efficient function in their environments and in the presence of constraints on material costs is currently an open area of investigation . In this study , we use methods from network science to compare and contrast the structure of two different classes of biological distribution networks: mycelial fungi and rodent brain vasculature . While each of these systems have been studied separately , less work has focused on understanding the diversity of their network organization . Here , we first examine several measures that characterize network connectivity on varying scales , finding that—although both systems have highly constrained network layouts—there are quantifiable differences in their architectures . Furthermore , using network analyses that specifically consider the embedding of these transport networks into real space , we observe that the two types of systems display distinct tradeoffs in network correlates of material cost , efficiency , and robustness . Together , our results provide evidence that while different distribution networks have general resemblances , they also exhibit variable design features that could reflect differences in their functions , environmental conditions , or development . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"neural",
"networks",
"mathematical",
"models",
"neuroscience",
"fungal",
"structure",
"fungi",
"mycelium",
"network",
"analysis",
"biological",
"transport",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"mycology",
"mathematical",
"and",
"statistical",
"techniques",
"centrality",
"biochemistry",
"eukaryota",
"biology",
"and",
"life",
"sciences",
"metabolism",
"organisms"
]
| 2018 | Comparing two classes of biological distribution systems using network analysis |
Arbovirus vector dynamics and spread are influenced by climatic , environmental and geographic factors . Major Chikungunya and Dengue fever outbreaks occurring the last 10 years have coincided with the expansion of the mosquito vector Aedes albopictus to nearly all the continents . We characterized the ecological ( larval development sites , population dynamics , insemination and daily survival rates ) and genetic ( diversity , gene flow , population structure ) features of two Aedes albopictus populations from distinct environments ( rural and urban ) on Réunion Island , in the South-West Indian Ocean . Microsatellite analysis suggests population sub-structuring Ae . albopictus populations . Two genetic clusters were identified that were significantly linked to natural versus urban habitats with a mixed population in both areas . Ae . albopictus individuals prefer urban areas for mating and immature development , where hosts and containers that serve as larval development sites are readily available and support high population densities , whereas natural environments appear to serve as reservoirs for the mosquito .
In recent years , the emergence of arboviruses and some of their vectors has caused major health and economic problems worldwide . Chikungunya ( CHIK ) , an arbovirus infection that was not considered to be a major health problem before 2005 , recently caused a major pandemic affecting Africa , Asia and to a lesser extent Europe . The pandemic began in Kenya and the South-West Indian Ocean in 2005 [1] , [2] , with a separate focus in Central Africa [3] , [4] that then spread to Europe [5] and Asia [6] . Thousands of people were affected with incidence rates up to 75% in Lamu , Kenya [7] . In addition to low levels of immunity against CHIV in the human population , emergence of epidemic transmission has been attributed to changes in vector competence [8] , ecology [9] , [10] and dynamics [11] . It is hypothesized that an amino acid replacement in the E1 envelope glycoprotein arose in response to selection for efficient transmission by Aedes albopictus especially in locations where Ae . aegypti was absent or less abundant [8] , [12] , [13] . Therefore , the vectors incriminated for this pandemic were primarily Ae . albopictus and to a lesser extent Ae . aegypti [14] , [15] . Aedes albopictus originated in Asia [16] and has extended its range in the last 20 years across many parts of the world . It is now recognized as a competent vector of numerous arboviruses [14] , [17] . Aedes vector dynamics and spread are affected by climatic [18] , [19] , environmental and geographic factors [20] , [21] . These vector species are known to be short-distance migrants and their dynamics are influenced by their environment [22] . The flight ranges of Aedes albopictus may increase when females fail to find a suitable site for oviposition or blood-meals . Its abundance varies from year to year and is affected by the inter-annual climate variability [23] . Indeed , understanding the factors that determine the vectors' habitat and population dynamics at a micro-scale is a major challenge but could help improve the efficiency of vector control . Réunion Island is situated in the South-West Indian Ocean , East of Madagascar . The habitat is predominantly composed of houses with gardens and more than 300 gullies spread throughout the island . The gullies cross urban environments and natural areas , providing potential mosquito production sites . Nevertheless , this habitat has never been evaluated for its impact on human health as a potential reservoir for mosquitoes , especially Ae . albopictus , the dominant species on the island [15] . Indeed , the population densities , dynamics or flow between the gullies and the urban environments have never been investigated . In this study , we seek to examine the population ecology ( larval development sites , longevity and insemination rates ) and genetic structure during two seasons in two locations , including gullies and urban areas .
All volunteers are co-authors and provided informed oral consent as the IRB approved the use of oral consent . Oral consent was obtained before starting the whole study , after clear explanation of what would stand in the study . All entomological surveys and gathering made on private lands or in private residences were made with the owners/residents permission and presence . We chose two gullies ( approximately 300 meters long ) close to an urbanised area bordering the gully ( 13 houses were surveyed ) : one situated on the eastern part ( Chemin Sévère , close to the main city Saint Benoît ) , the other on the western side of the island ( Bassin Plat , close to the main city Saint Pierre ) . The distance between the gullies and houses was less than 20 meters . Both sites are infested with Ae . albopictus; in 2007 the average Breteau Index was 38 in Saint-Benoît and 28 in Saint-Pierre ( Agence Régionale Santé [ARS] ) and had a high number of CHIKV cases reported in 2005–2006 ( Bulletin Épidémiologique Hebdomadaire , Institut de Veille Sanitaire , N° 38-39-40 , 21 October 2008 ) . Seroprevalences of antibodies against CHIKV estimated after the epidemic were 48% and 38% in the Saint Benoît and Saint Pierre , respectively [24] . Houses surveyed were built in concrete and/or wood and tin- roofed . Houses had , as in most urban areas on La Réunion , a large grass-garden combined with numerous varieties of fruit trees , flower beds , bushes and other flowering shrubs . Entomological surveys were conducted 1 gully and 13 nearby houses in both the Chemin Severe ( East ) and Bassin Plat ( West ) sites the austral winter in 2006 ( July/August ) and summer in 2007 ( February/March ) . Container surveys for immature mosquitoes and human landing collections for adult mosquitoes were carried out on the same day at each site . Adult samplings were performed under a shaded environment ( for example in the gully in the east that was under the shade of a bamboo grove , in the west under shade of Schinus therebenthifolius grove , in urban parts under the shade of fruit trees ) . At each season about 2 weeks were needed for each site to perform all entomological surveys , larvae and adult collections . Females Ae . albopictus were dissected to determine parity and the number of spermathecal capsules filled . Once the parity status [25] of each female had been determined , the three spermathecal capsules were placed in a drop of saline water on a glass slide covered with a glass cover slip and examined for sperm under a microscope . Each mosquito was ground in 200 µl of 2% CTAB with a glass bid using a Mix Miller MM 400 set at 30 Hz and left for 5 min at 65°C . Then , 200 µl of chloroform was added and mixed gently . After a centrifugation ( 12 000 rpm , 5 min ) , the upper phase was collected and 200 µl of isopropanol added . The mix was centrifuged for 15 min ( 12 000 rpm ) and the isopropanol was removed . Then , an extra step of 70% ethanol was carried out to purify the DNA . After the removal of the 70% ethanol , the DNA was dried using a speed-vac and eluted with 20 µl of water . Of the six primers available from the literature for Ae . albopictus [26] , one ( AealbB52 , Table 2 ) was not variable and the genetic resolution obtained using the remaining five markers was not considered adequate . We screened some Ae . aegypti microsatellite markers available from the literature and selected two new markers ( AEDC and 34–72 , Table 2 ) . In parallel we developed an enriched microsatellite bank , from which we identified two additional markers ( see Table 2 ) . The extracted DNA of each sample was used as a template for the amplification of a set of 10 microsatellite markers AealbB52 , AealbB51 , AealbA9 , AealbB6 , AealbD2 , AealbF3 , alb212 , 34–72 , AEDC , alb222 ( Table 2 ) . These markers were selected for polymorphism , size , and low numbers of null alleles . Two were from the newly developed set , eight were from Ae . albopictus ( 6 ) and Ae . aegypti ( 2 ) literature ( Table 2 ) . A total of 342 adults were genotyped with these markers ( Table 3 ) . Genomic ( 10 ng ) DNA was used for amplification with the QIAGEN multiplex PCR Master Mix kit ( ref . 206145 ) according to the manufacturer's instructions in a final volume of 15 µL . One of each pair of primers was fluorescently end-labelled with the fluorochromes NED , VIC , PET or FAM . Two primer mixes were used in 15 µL at a final concentration of 400 nM . The programme consisted of denaturation at 94°C for 5 min followed by 30 cycles at 94°C for 45 s , 56°C for 1 min 30 s , 72°C for 45 s , with a final elongation step for 30 min at 60°C . Then , 2 µL of the DNA was diluted from 1/100 to 1/60 according to PCR products . The diluted PCR product was mixed with 10 . 7 µL of ultra-pure Hi-Di-formamide TM and 0 . 3 µl of size marker ( GeneScan 500Liz ) , and loaded onto an ABI Prism 3130 Genetic Analyser automated sequencer . Allele sizes were determined using GeneMapper v4 . 0 . A mixed-data factor analysis was carried out on our datasets containing a combination of continuous and ordinal variables . The resulting components were used in regression models and tested with an ANOVA . A step-by-step analysis was done with all the different factors , removing the less significant factor at each step . At each step , we compared the tested model and the previous model until a significant difference appears between the two models . The final retained model was the model before this significant difference appears . We used a model with multivariate normal random effects , using Penalized Quasi-Likelihood . This general linear model is used to fit generalized linear models , specified by giving a symbolic description of the linear predictor and a description of the error distribution . Specifically , the GLMM is assumed to be of the form g ( μ ) = Xβ+Ze where g is the link function , μ is the vector of means and X , Z are design matrices for the fixed effects β and random effects e respectively . Furthermore the random effects are assumed to be i . i . d . N ( 0 , σ2 ) . The 14 studied factors are the width , depth , volume , class and type of the breeding site , type of habitat , the location , organic matter , effect of the sun , the season , the year and the presence of Anopheles and Culex and the type of water quality . The year was considered as a randomised factor . The models and the ANOVA were carried out on a dataset from 728 immature production sites . All statistical analyses of this part were performed with R software .
A total of 11 , 528 Ae . albopicutus larvae and pupae were collected from 728 potential larval development sites ( Table 1 ) . Of these , figure the 8 , 634 immature individuals collected from 630 containers or natural habitats located on the east side of the island compared to 2 , 890 collected in 99 containers on the west side of island . Abundance of larval habitats were comparable during the winter and summer collections ( Table 1 ) . Although , slightly more potential larval habitats were observed in gullies compared to household collections ( 396 versus 333 ) , significantly more larvae and pupae were collected in houses ( nearly 2 fold 7 , 565 versus 3 , 959 ) . The average number of immature Ae . albopictus ranged from: 8 to 13 in natural production sites and small containers; 37 to 46 in plates under flowerpots , tyres and big containers; and 82 in basins and tanks . Six of the 14 factors tested ( see material and method section ) , significantly influenced the number of immature mosquitoes present in the production sites ( Table 4 ) . The width , volume and nature of the breeding site were correlated with the number of immature Ae . albopictus , as well as the mosquito's habitat and the location . The number of Aedes individuals was significantly higher in the biggest and widest breeding sites . We observed significant differences between the average numbers of immature Ae . albopictus ( +/− SE ) from natural and plant immature production sites ( 13 . 10±2 . 58 ) compared to the artificial immature production sites ( 42 . 65±3 . 54 ) . The total number of immature mosquitoes was higher in the east than in the west . However , average productivity was significantly higher in the west ( 47 . 31±6 . 07 ) than in the east ( 19 . 96±2 . 29 ) . A significant difference was observed between the number of immature mosquitoes from gullies or urban areas . On average , Ae . albopictus immature production site productivity was 13 . 53±3 . 32 in gullies and 37 . 53±3 . 32 in urban areas . The number of immature Ae . albopictus was not correlated to the presence of sun ( yes/no ) , water quality ( clear/tinted/polluted ) , the presence of organic matter , the season ( winter/summer ) and the presence of Anopheles and Culex . A total of 851 Ae . albopictus adult females were dissected to determine parity and the number of spermathecae that were inseminated . Overall 70 . 2% were parous ( 598/851 ) . On both sides of the island , the proportion of parous females was higher in houses than in gullies ( Figure 1 ) . In contrast , the effect of season on parity differed between the two sides of the island; in the west parous rate in the summer was 56 . 1±1 . 9% compared to 81 . 0±2 . 1% during the winter ( Figure 1 ) whereas in the eastern sites there was no significant difference between the parous rates during the two seasons . The daily survival rate was higher in urban areas and during the winter . Thus , younger mosquito populations were found in gullies ( Table 5 ) . The dissection of the spermathecal capsules of the same 851 adult females showed that 193 had empty spermathecae , 335 had one filled spermathecae , 304 had two filled spermathecae and 19 had three filled spermathecae ( Figure 2 ) . Independent of location , the number of filled spermathecal capsules in Ae . albopictus was higher in urban than gully habitats ( ANOVA; Df = 1; F = 140 . 85; P<0 . 0001 ) . On average , the number of empty spermathecae was significantly higher in the gully areas than in urban areas ( Chi2 , P<0 . 0001 Figure 2 ) . There were significantly more females caught in urban areas with one full spermathecal capsule compared to females captured in the gully areas , independent of location ( Figure 2 ) . However , in the western location no significant differentiation was noticed between the different habitats for females with two filled spermathecae ( Chi2 , P = 0 . 25 ) unlike in the east ( Chi2 , P<0 . 0001 ) . From a total of 342 mosquitoes collected , 11 produced no PCR products for fewer than six of the loci and were discarded ( Table 3 ) . The loci AealbB52 were monomorphic for our populations . The nine remaining loci had 4 to 17 alleles each , with allelic richness ranging from 4 to 13 . On the 36 combinations of pairs of loci , only four combinations were in linkage disequilibrium . Population structure among samples was investigated using assignation probabilities provided by Structure . Two groups ( DK = 300 ) were identified of 139 and 114 mosquitoes , respectively ( Figure S1 ) . It was considered that an individual assignation probability in the [0 . 30; 0 . 70] interval belonged to a hybrid genotype and the others belonged to pure populations . A total of 78 hybrids were detected ( Figure S2 ) . Significant deviation from Hardy-Weinberg Equilibrium was detected in seven markers ( Table 6 ) . The recorded deviations are likely due to null alleles because in each case there was an excess of homozygotes . The AMOVA showed that most of the variation was distributed within individuals ( 81% ) , but also between clusters identified in Structure ( 8 . 35% ) ( Table 7 ) . The genetic differences between cluster 1 and cluster 2 account for more genetic variance ( 8 . 35%; Fct = 0 . 084 , P<0 . 001 ) than those among habitats within clusters ( 1 . 5%; Fsc = 0 . 017 , P<0 . 001 ) . No significant differentiation was found among clusters between types of immature production sites ( artificial/natural; data not shown ) or season ( summer/winter; data not shown ) . The two clusters ( plus a hybrid population ) were assigned using Structure software , following the microsatellite analysis ( see above ) . The 331 individuals analysed and assigned to one of the populations were studied in relation to their location ( east/west ) , the season when they were sampled ( summer/winter ) , their habitat ( gully/urban ) and the nature of the immature production sites ( natural/artificial ) ( Table 5 , 7 , Figure S2 ) . When the dataset was partitioned into the two locations sampled ( east/west ) , we observed significant differences in the genetic distribution of these clusters according to their habitat ( Figure 3 ) . In the west , no difference was observed between the three clusters according to habitat . However , in the east , individuals from cluster 1 were significantly more present in the urban habitat and cluster 2 significantly in the gully habitat . Interestingly , the hybrid cluster was found in almost equal proportion in both gully and house habitats ( Figure 3 ) . No significant differentiation was found in terms of the nature of the immature production sites ( artificial/natural ) .
We have shown that urban areas are preferred by Ae . albopictus for mating and oviposition . This is likely due to host availability and the existence stable and abundant artificial containers that serve as larval development sites facilitating large mosquito densities . Gullies and other natural environments however , are potential reservoirs for Ae . albopictus on Réunion Island , for re-colonising the urban areas after a population reduction ( for example , following vector control ) . Nevertheless , when available suitable larval development sites are abundant , low production of mosquitos and population structuring is observed . This suggests that females have a preference for certain habitats and reproductive isolation depending on the habitat . An important consequence of the existence of highly clustered , local spatial patterns is that if some houses are missed during vector control operations , it is possible that the remaining intact mosquito clusters could subsequently repopulate the area . These results underline the need to use new control methods as an alternative to chemical control , such as the sterile insect technique . | The objective of our research was to study the movements the mosquito Aedes albopictus . This mosquito transmits more than 20 viruses to humans throughout the world and is the vector of the recent major epidemics of Dengue and Chikungunya on Reunion Island and the Indian Ocean Region and is , therefore , of great interest for human health . We set out to determine whether reservoirs of populations could be found in natural environments and whether or not these populations are capable of re-colonising urban areas . Until now , only limited data has been available on the population dynamics of Aedes albopictus in this part of the world , information critical for guiding vector control strategies and predicting or preventing epidemics . We chose two areas where a serious CHIKV epidemic occurred . We then used genetic markers and ecological data to estimate patterns of gene flow and behaviour . We were able to demonstrate that populations were structured with limited gene flow despite observing migration . We found that Ae . albopictus preferred urban areas for mating and to lay their eggs because of the availability of hosts and permanent containers that favoured higher mosquito densities . We also show , however , that natural environments are reservoirs for re-colonisation of urban areas . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"infectious",
"diseases",
"ecology",
"genetics",
"neglected",
"tropical",
"diseases",
"arboviral",
"infections",
"biology",
"genetics",
"and",
"genomics"
]
| 2013 | Evidence of Habitat Structuring Aedes albopictus Populations in Réunion Island |
Entamoeba histolytica is a pathogen that during its infective process confronts the host defenses , which damages the amoebic plasma membrane ( PM ) , resulting in the loss of viability . However , it is unknown whether amoebic trophozoites are able to repair their PM when it is damaged . Acid sphingomyelinases ( aSMases ) have been reported in mammalian cells to promote endocytosis and removal of PM lesions . In this work , six predicted amoebic genes encoding for aSMases were found to be transcribed in the HM1:IMSS strain , finding that the EhaSM6 gene is the most transcribed in basal growth conditions and rendered a functional protein . The secreted aSMase activity detected was stimulated by Mg+2 and inhibited by Co+2 . Trophozoites that overexpress the EhaSM6 gene ( HM1-SM6HA ) exhibit an increase of 2-fold in the secreted aSMase activity . This transfectant trophozoites exposed to pore-forming molecules ( SLO , Magainin , β-Defensin 2 and human complement ) exhibited an increase from 6 to 25-fold in the secreted aSMase activity which correlated with higher amoebic viability in a Ca+2 dependent process . However , other agents that affect the PM such as hydrogen peroxide also induced an increase of secreted aSMase , but to a lesser extent . The aSMase6 enzyme is N- and C-terminal processed . Confocal and transmission electron microscopy showed that trophozoites treated with SLO presented a migration of lysosomes containing the aSMase towards the PM , inducing the formation of membrane patches and endosomes in the control strain . These cellular structures were increased in the overexpressing strain , indicating the involvement of the aSMase6 in the PM injury repair . The pore-forming molecules induced an increase in the expression of EhaSM1 , 2 , 5 and 6 genes , meanwhile , hydrogen peroxide induced an increase in all of them . In all the conditions evaluated , the EhaSM6 gene exhibited the highest levels of induction . Overall , these novel findings show that the aSMase6 enzyme from E . histolytica promotes the repair of the PM damaged with pore-forming molecules to prevent losing cell integrity . This novel system could act when encountered with the lytic defense systems of the host .
It is well established that the plasma membrane ( PM ) is the most important component of the cell that maintains its integrity and homeostasis during physiological changes . However , its integrity is regularly infringed by many mechanical or biochemical factors that endanger cell viability . The eukaryotic cells have developed mechanisms that help repair injuries and prevent the release of cytoplasmic components , thus restoring complete functionality or allowing their elimination through apoptosis [1 , 2] . The first sign of damage to the PM is the uncontrolled entry of Ca+2 , which activates different mechanisms in order to repair the damaged membrane [1 , 3] . The repair of injuries by a spontaneous reorganization of membrane phospholipids can only occur if the membrane damage is not recurrent and the lesion is under 0 . 2 μm in diameter [3 , 4] . More frequent or larger lesions need the replacement of the damaged area by the elimination of the lesion for which two main mechanisms have been described . One repair mechanism is mediated by the annexins which are Ca2+ sensors that merge at the site of damage in the PM and promote its elimination [4–7] . Another repair mechanism involves the exocytosis of lysosomal acid sphingomyelinase ( aSMase ) that can trigger the formation of endosomes that internalize the lesion regenerating the integrity of the PM [7–9] . After damage to the PM and the uncontrolled entry of Ca2+ , there is a recruitment of lysosomes to the site of the lesion and then they fuse to the PM , releasing the aSMase which hydrolyzes sphingomyelin into ceramide; this , in turn , favors the formation of endosomes that internalize the lesion and restores the integrity of the membrane [8–11] . Intestinal amoebiasis is a parasitic infection that affects humans and is caused by the protozoan Entamoeba histolytica , resulting in 40 , 000 to 100 , 000 deaths annually worldwide [12–14] . E . histolytica infection is a multifactorial process in which adaptation is the key for parasite survival , which is dictated by specific molecules and the mechanisms to confront the host immune system . The first line of innate immune defense which amoebae confront is the mucus layer that acts as a protective barrier that prevents damage to the intestinal epithelial cells . When the trophozoites overcome this first barrier , the epithelial cells secrete potent pro-inflammatory mediators and chemokines that recruit immune cells , such as neutrophils , which release reactive oxygen species [15 , 16] and activated macrophages that release nitric oxide [16 , 17] , both damaging the plasma membrane by lipid oxidation . Another protection mechanism of the intestinal epithelial cells is the production and secretion of antimicrobial peptides , such as LL-37 [18] and defensin 2 [19] , which has been showed to damage the trophozoites in vitro . These peptides destabilize and alter the PM of E . histolytica , causing an increase in its permeability , reducing the viability of the amoebae . The human complement system is one of the most effective strategies to prevent the dissemination of trophozoites , once the amoebae activate the complement system , the membrane attack complex is formed which lyses the trophozoites [20 , 21] . In response to the attack of the human immune system , the amoeba confronts it and neutralize the potential damage through molecules called virulence factors . The most relevant molecules are: i ) Gal/GalNAc lectin that binds to colonic mucin and promotes adhesion to the host cell [22 , 23] and linking to C8 and C9 subunits interrupting the assembly of the complement on the membrane of the trophozoites [24]; ii ) cysteine proteases that degrade antibodies such as IgA and IgG [25 , 26] and the C3 convertase involved in the activation of the classic complement pathway and the amplification of the inflammatory process [21]; also , these enzymes degrade IL-1β , reducing the production of reactive oxygen species and nitric oxide by neutrophils and activated macrophages [27–29]; iii ) amoebapores , which are small pore-forming proteins that have the ability to lyse eukaryotic cells in a cell contact-dependent manner [30–32] . Also , there is another set of molecules called virulence determinants that are indirectly involved in the pathogenic process by regulating the expression of virulence factors , with high gene expression plasticity and conferring adaptation and survival of the amoeba [33] . E . histolytica is an infectious agent that during the invasive process , either in the intestine or the liver , confronts the different defenses of the host , such as inflammatory response and the complement or the antimicrobial peptides that can directly damage its PM , requiring a mechanism for membrane repair and prevent its cell lysis . In the present work , we evaluated the participation of aSMases in the PM damage repair in E . histolytica . The analysis of the E . histolytica genome revealed the presence of six genes encoding for aSMases that are transcribed under basal growth conditions . The EhaSM6 coding sequence generates a functional recombinant protein with aSMase activity . The aSMase activity secreted by the virulent strain HM-1:IMSS is stimulated by Mg+2 and inhibited by Co+2 , similarly as for the recombinant enzyme . The EhaSM6 overexpression in E . histolytica trophozoites induces an increase in the secreted activity and tolerance to lysis with pore-forming molecules ( β-Defensin 2 , human complement , Streptolysin O and Magainin ) , in a Ca2+-dependent process . This PM damage induced an increase in the expression of EhaSM1 , 2 , 5 and 6 genes , meanwhile oxidative stress induced an increase in all of them . Furthermore , we found that the damage to the PM of E . histolytica induces the exocytosis of lysosomal aSMase , resulting in the formation of membrane patches and endosomes . We propose that membrane damage promotes the migration of lysosomes towards the exposed site of the lesion where they secrete aSMase that induce the endosome formation and thus internalize the lesion , regenerating the integrity of the PM , favoring the amoebic viability and survival . We report for the first time that E . histolytica possess a mechanism of PM damage repair mediated by aSMase .
A search for annotated sequences encoding aSMases was done in the E . histolytica genome database [34] . Six genes annotated as putative aSMases-like phosphodiesterases were found: EhaSM1 ( EHI_040600 ) , EhaSM2 ( EHI_172510 ) , EhaSM3 ( EHI_118110 ) , EhaSM4 ( EHI_100080 ) , EhaSM5 ( EHI_147020 ) and EhaSM6 ( EHI_125660 ) . The analysis of the six identified coding sequences for aSMases revealed that they present homology at the protein sequence level . The identified sequences showed conserved domains , but overall , the homology ranged between 11 to 50% , with identity ranging between 11 to 50% ( S1 Fig ) . The alignment of the predicted amoebic sequences with those corresponding to aSMases in other organisms ( Homo sapiens and Mus musculus ) ( S1 Fig ) showed that the all the amoebic sequences exhibited the amino acids involved in catalysis , that have been described for these enzymes , which includes conserved amino acids important for catalysis , sites for metal coordination , the hydrophilic-aromatic cluster , the substrate recognition site "NX3CX3N" [35] , the cysteines involved in the processing of the C-terminal region associated with the activation of the enzyme [36 , 37] , and N-terminal signal peptide that suggests these enzymes are secreted [38] ( Fig 1 ) . These conserved residues match with those reported for the aSMases of the nematode Caenorhabditis elegans , which has a 30% homology with the human aSMase [39] . The in silico analysis of amoebic aSMases revealed that coding sequence renders proteins ranging between 46–49 kDa with a predicted signal peptide of 14 to 20 residues . Also , no transmembrane regions or signals for endoplasmic reticulum retention were identified , which suggests that these proteins are secreted . In addition , four aSMases have a calcineurin-like region ( EhaSM1 , EhaSM2 , EhaSM5 , and EhaSM6 ) ( Fig 1 ) , which suggests their participation in response to stress conditions [40] . qRT-PCR analysis showed that all the putative aSMase genes are transcribed and the EhaSM6 showed the highest expression ( 60% ) under basal growth conditions ( S2 Table ) . We found that the six predicted amoebic aSMases genes are transcribed in the HM1 strain under basal growth conditions . Although these putative aSMase enzymes share sequence similarity with well-characterized mammalian aSMases , it is not known if these genes code functional aSMases proteins . The expression analysis showed that EhaSM6 exhibited the highest expression , for this reason , we cloned the complete ORF in a suitable vector and the recombinant protein was produced in E . coli . The purified rEhaSM6 exhibited aSMase activity stimulated by Mg2+ and inhibited by Co2+ ( S3 Table ) . In contrast with previous studies showing that the mammalian enzyme was stimulated by Zn2+ [41 , 42] , the amoebic rEhaSM6 was not significantly affected by Zn2+ ( S3 Table ) . The aSMases of E . histolytica have a predicted signal peptide , therefore , the secreted aSMase activity was determined from trophozoites in the exponential growth phase . After 10 min incubation , the secreted activity of HM-1:IMSS trophozoites was determined in the presence of different concentrations ( 0–20 mM ) of divalent cations ( Mg2+ , Mn2+ , Ca2+ , Co2+ or Zn2+ ) , likewise , the EGTA chelator was used to scavenge for trace amounts of metal ions . Mg2+ at a concentration of 20 mM stimulated the activity 2 . 1-fold , while Ca2+ and Zn2+ did not show a significant effect . Mn2+ and EGTA at a concentration of 20 mM inhibited the activity in approximately 40% , while Co2+ totally inhibited the activity at concentrations even lower than 1 mM ( Fig 2A ) , presenting the same cation requirements or inhibitory effect as for the rEhaSM6 . To evaluate the biological role of aSMases in the amoebic trophozoites , the aSMase6 was further studied by generating a construct to overexpress it . Transfectant trophozoites overexpressing the EhaSM6 gene fused to an HA tag in the C-terminus were obtained . The transcript level of EhaSM6 was evaluated in the overexpressing strain ( HM1-SM6HA ) by qRT-PCR . The transcript level was 7 . 2-fold higher in comparison with the control strain transfected with the empty plasmid ( HM1-HA ) ( Table 1 ) . We measured the secretion of aSMase activity in the control and overexpressing strains at different times of trophozoite incubation . The results indicate that the activity of secreted aSMase in the overexpressing strain increases 3-fold in the first 10 min of incubation in comparison with the control strain . After 30 min the secreted activity remains constant ( Fig 2B ) . Secreted aSMase activity was not detected in trophozoites incubated at 4 °C . SLO is an exotoxin from group A streptococci , which is toxic to eukaryotic cells due to its affinity for cholesterol , but at controlled doses promotes the formation of membrane pores that in consequence triggers the secretion of aSMase required for membrane repair in mammalian cells [43 , 44] . The effect of SLO on the secretion of aSMase activity in trophozoites of HM1-HA and HM1-SM6HA strains was evaluated . Different concentrations of toxin were tested ( 0 . 8–6 . 4 ng/μL ) , and the secreted activity of aSMase in trophozoites of amoebic strains was determined . Both strains showed an increase in the secreted activity , reaching a maximum of 8-fold with 1 . 6 ng/μL of SLO ( Fig 3A ) . The secreted aSMase activity in the HM1-SM6HA is at least 2-fold higher than for HM1-HA in all the SLO treatment conditions , except with 6 . 4 ng/μL where there is a decrease in activity for both strains with respect to toxin-free amoebas . Also , after SLO treatment the amoebic viability was analyzed . The control HM1-HA strain showed a gradual decrease in cell viability , reaching 15% of viability at 3 . 2 ng/μL of SLO , this is in contrast with the 65% of viability for the EhaSM6 overexpressing strain ( Fig 3B ) . Interestingly , 6 . 4 ng/μL of SLO , a lytic concentration of the toxin , is associated with the lowest levels of enzyme secretion , suggesting that the secreted activity undergoes an activation process . Therefore , a greater increase in the secreted activity of aSMase is related to enhanced amoebic viability after SLO treatment . The viability of trophozoites of both strains was also evaluated using Live/Dead staining . After treatment with 1 . 6 ng/μL of SLO for three min , the viability of HM1-HA and HM1-SM6HA strains were 25 . 45% and 77 . 84% , respectively ( Fig 3C ) . The viability values were similar to those obtained by the exclusion of trypan blue ( Fig 3B ) . To rule out that the rise in aSMase activity in the supernantat after SLO tratment , alcochol dehydrogenase ( ADH ) , a cytoplasmic enzyme , activity was measured in E . histolytica trophozoites . Activities of aSMase and ADH were monitored over time to determine a correlation to cellular integrity . By measuring the release of the cytoplasmic marker enzyme ADH in trophozoites of both strains , only 15% of total activity was released from amoebae without and with low SLO exposition ( 1 . 6 ng/μL ) . While the secreted aSMase activity was present in amoebic supernatants without SLO treatment and a significant increase of the secreted activity ( >8-fold ) was observed when amoebae were exposed to the same low concentrations of SLO . In contrast , lytic SLO concentration ( 6 . 4 ng/μL ) released a substantial ADH activity in comparison with the very low aSMase activity detected for both amoebic strains . It became apparent that the release of ADH was accompanied by cellular disintegration caused by high SLO concentration ( 6 . 4 ng/μL ) . Using the tool of ADH release as a criterion for cellular disintegration , the majority of amoebae appeared to be intact in DMEM medium ( S2 Fig ) . At least in this condition , the majority of aSMase activity has been released by viable amoebae . With regard to aSMases , the increased activity detected in the supernatants were found to depend on the SLO concentration and amoebic viability , indicating that the aSMase is secreted and activated only upon plasma membrane damage response in viable trophozoites with active secretory machinery . The Ca2+ gradient that infiltrates through cell membrane lesions activates cytoplasmic proteins and induce the exocytosis of lysosomes at the site of the lesion [45] . Lysosomes release their cargo , including the aSMase , followed by massive endocytosis that triggers the process of wound repair [8 , 44 , 46] . The secreted aSMase activity and viability of amoebic trophozoites were evaluated in the presence of 1 . 6 ng/μL of SLO but using a Ca2+-free DMEM medium . No increase of secreted aSMase activity was observed in trophozoites of HM1-HA and HM1-SM6HA strains treated with SLO in the absence of extracellular Ca2+; in contrast with the increase in enzymatic activity in the presence of extracellular Ca2+ . The secreted activity of aSMase in the absence of toxin was not affected by the presence or absence of Ca2+ in both strains ( Fig 4Aa ) . The results show that after inducing damage with SLO in the presence of 1 . 8 mM of Ca2+ for three min , the viability of the control strain is 30% , while that of the EhaSM6 over-expressing strain is 80% . In contrast , in the absence of Ca2+ and damage with SLO , the viability decrease was more evident , since the viability for both strains is less than 11% in comparison to their controls in the absence of damage under the same conditions ( Fig 4Ab ) . Therefore , Ca2+ is a necessary cation for exocytosis of lysosomes and the secretion of aSMase activity , which correlates with the preservation of amoebic viability as it has been demonstrated in other cell types that trigger the repair of damage to the PM [47] . As it has been shown above ( Fig 2 ) , the aSMase activity from E . histolytica was strongly inhibited by Co2+ . We perform an assay of plasma membrane repair after SLO exposition in the presence of Co2+ 0 . 5 mM . The Fig 4Bc shows that after the exposition to the SLO , the secreted aSMase activity was inhibited by Co2+ , which also inhibited the membrane repair capacity of the amoebae and consequently decreased their viability ( Fig 4Bd ) . This strongly suggests that aSMase6 is one of the enzymes that initiate the repair mechanism of damage to the PM of the amoeba . To evaluate if the aSMase6 is involved in the repair of membrane damage generated by another membrane damaging agent , we used the antimicrobial peptide β-Defensin 2 . During an intestinal human infection , several antimicrobial molecules are produced , where β-defensins are the most common [48] . Defensins are cysteine-rich cationic peptides of low molecular weight , 3 to 5 kDa , that bind to the membranes of microorganisms rich in anionic phospholipids [49] and integrate into the membranes inducing the formation of pores [50] . In this study , the HM1-HA and HM1-SM6HA strains were exposed to increasing concentrations of β-Defensin 2 . Concentrations of 20–30 ng/mL of β-Defensin 2 induced the highest secretion levels of aSMase activity reaching 6-fold and 7-fold for the control and overexpressing strains respectively , which is 2-fold higher in the overexpressant than the control strain ( Fig 5Aa ) . Higher levels of secreted activity correlate with higher capability of PM repair and consequently greater viability ( Fig 5Ab ) . At higher concentrations of β-Defensin 2 , the levels of secreted activity decreases , as does the viability of trophozoites , again suggesting that the secretion of aSMases is a process that requires active cellular metabolism . The human complement system is an effective defense mechanism in the elimination of pathogens , which culminates in the formation of large pores in the membrane , which alters the osmotic equilibrium of the pathogen [51 , 52] . The effect of the complement on the secreted aSMase activity and viability of the trophozoites of HM1-HA and HM1-SM6HA strains was evaluated . The results obtained indicate that after incubation with the complement , both strains significantly increase the activity of secreted aSMase , showing a 3-fold increase in the overexpressant with respect to the control strain ( Fig 5Bc ) . Furthermore , viability results ( Fig 5Bd ) show that amoebae interacting with complement diminish their viability , but this decrease was more evident for the control strain than for the overexpressing strain , which bears a more efficient repair mechanism . Complement exposition induced the highest levels of secreted aSMase activity in E . histolytica than all the pore-forming peptides evaluated in this work . Another plasma membrane disrupting agents were evaluated such as Magainin , an antimicrobial peptide and , Triton X-100 , an non-ionic detergent . Magainin is a cathelicidin that interacts with the plasma membrane forming permeable ion channels [47 , 48] . Trophozoites exposed to increasing concentrations of Magainin exhibit similar results to those obtained with the other pore-forming molecules evaluated in this work , where the higher secreted aSMase activity of the HM1-SM6HA correlated with the higher amoebic viability ( S3A Fig ) . When Triton X-100 was used at concentrations that acts as a permeabilizer [53] , a slight increase in the secreted activity of aSMase was observed , but at higher concentrations , it produces lysis of the amoebae and very low levels of activity was detected ( S5B Fig ) . It has been reported that the generation of ceramide , mainly a product of the hydrolysis of sphingomyelin by SMases , is associated with different stimuli such as oxidative stress generated by H2O2 , generating cellular responses such as apoptosis and cell proliferation [54] . To determine the possible role of aSMases in response to damage by oxidative stress in E . histolytica , the effect on the secreted activity of aSMases and cell viability in the HM1-HA and HM1-SM6HA strains was determined . Trophozoites of HM1-SM6HA strain showed a significant increase in the secreted activity of the aSMase in all peroxide concentrations evaluated with respect to that observed in the HM1-HA control strain ( Fig 6A ) ; these increase in the activity of aSMase is related to a maintenance of the viability ( Fig 6B ) in each of the peroxide concentrations evaluated in the HM1-SM6HA strain . However , the level of secreted aSMase activity is lower than those obtained with pore-forming molecules . The above evidence strongly indicates that the aSMase6 of E . histolytica is involved in membrane repair . To evaluate the events related to PM damage repair induced by SLO in HM1-HA and HM1-SM6HA strains , trophozoites labeled with Lysotracker from both strains were incubated at 37 °C with FITC labeled-Dextran and 1 . 6 ng/μL of SLO for short times . The HM1-HA control and HM1-SM6HA overexpression strains without exposing to SLO show the presence of lysosomes ( red ) and few endosomes ( green ) in the cytoplasm of the amoebic trophozoites ( Fig 7A and 7B ) . HM1-HA trophozoites treated with SLO ( 1 . 6 ng/μL ) change their classic amoebic shape to a rounded shape after 1 min of incubation , and the migration of lysosomes to the PM is observed , some of them seems fused to PM forming elongated structures ( patch-like ) , this resembles the structures formed in some damaged eukaryotic cells . In addition , the number of endosomes increases and some of them are also located in the plasma membrane , where few of them co-localize with lysosomes ( yellow ) forming structures with a patch-like shape . After three min of SLO exposure , the patches on the plasma membrane disappear and the number of endosomes increases in the subcortical region of the PM . After 10 min of SLO exposure , the polarization of the endosomes is lost , and vesicles exhibiting the label for both lysosomes and phagosomes are observed , suggesting the formation of endolysosomes ( yellow ) ( Fig 7A ) . The mobilization of vesicles resembles the mechanism of membrane damage repair in HeLa cells , fibroblasts and lymphoblasts , after exposure to SLO , where the lysosomes fuse to the membrane and release the aSMase which promotes the injury site endocytosis [8 , 44 , 46 , 47] . Transfectant trophozoites that overexpress aSMase6 ( HM1-SM6HA strain ) exposed to SLO under the same conditions described above , after one min of exposition , trophozoites showed a greater number of patches showing colocalization of the lysosome and endosome markers . After three min of SLO treatment , the patches are no longer present , and the endosomes are observed in the inner side of the plasma membrane without co-localization with lysosomes . After 10 min , a large number of endosomes fused with lysosomes ( endolysosomes ) were observed . We theorize that the formation of these structures is a consequence of efficient endocytosis of the lesions in the PM ( Fig 7B ) in the HM1-SM6HA strain . Endosomes per cell were quantified , showing an average of 2 . 8 , 8 . 2 , 5 . 2 and 29 . 8 endosomes per amoeba in strains HM1-HA , HM1-HA with SLO , HM1-SM6HA and HM1-SM6HA with SLO , respectively . Likewise , the amoebae of each of the strains from identical conditions as above were lysed and the FITC fluorescence emitted by the endosomes was quantified ( S5 Table ) , which corroborates the data above . These results suggest that the aSMase6 overexpression induce an efficient repair mechanism evidenced by a significant increase in the viability of the trophozoites ( Fig 5B ) , reinforcing the role of aSMase6 in PM repair . With the above observations that revealed an intense mobilization of vesicles towards the plasma membrane , we sought the ultrastructural analysis of the trophozoites exposed to SLO by transmission electron microscopy ( TEM ) . Trophozoites of HM1-HA and HM1-SM6HA strains , adhered to a plate , were treated with 1 . 6 ng/μL of SLO for one and three min , as described above . The trophozoites without SLO exposition , show a typical amoeboid morphology and cell structures , including several vesicles with little content inside ( v ) and glycogen ( g ) ( Fig 8Aa and 8Ba ) . After one min of toxin exposure , the trophozoites of HM1-HA strain showed a polarization of vesicles towards the PM ( Fig 8Ab ) , and after three min the vesicles presented residues of the PM ( Fig 8Ac ) . Transfectant trophozoites that overexpress aSMase6 exposed for one min to SLO showed a stronger polarization of vesicles towards the site of membrane damage , displacing glycogen to the opposite end of the cell ( Fig 8Bb ) . After three min , an increase of vesicles containing membrane residues was observed ( Fig 8Bc ) . The vesicles involved in membrane damage repair are mainly located polarized to one side of the PM because the SLO exposure was performed on adhered trophozoites , however when cells in suspension were used , the migration of vesicles was homogenously distributed around the PM and the glycogen was located on the center of the trophozoite ( S4 Fig ) . In summary , fluorescence and transmission electron microscopy results strongly suggest that trophozoites of E . histolytica present a PM damage repair mechanism that renders viable trophozoites after PM damage . The results of confocal microscopy revealed that amoebae treated with toxin had an active process of endocytosis . To demonstrate that the secreted aSMase6 of E . histolytica participates in the repair mechanism by inducing the formation of endosomes , the amoebae of the HM1-HA strain were incubated with the protein rEHaSM6 and FITC -Dextran for 1 minute at neutral and acid pH . The results show that the trophozoites without the addition of the recombinant protein both at neutral pH ( Fig 9Aa ) and acid ( Fig 9Ba ) , present an average of 10 . 2 and 7 . 5 endosomes per trophozoite , respectively . When the recombinant protein is added to the trophozoites for one minute at neutral pH in which the enzyme has no activity , the number of endosomes was similar to that of the control in this same condition ( Fig 9Ab ) ; meanwhile at acid pH an increase of 5-fold in the number of endosomes with respect to the untreated control was observed ( Fig 9Bb ) . When Co2+ 0 . 5 mM was added to the HM1-HA strain at acidic pH , there was a decrease of endosome formation with respect to the amoebae without exposition to Co2+ ( Fig 9Ca ) . In the presence of the recombinant protein and Co2+ 0 . 5 mM , the induction of endosome formation was not observed ( Fig 9Cb ) . These results corroborate the participation of the aSMase6 in the formation of endosomes , which , are important structures for the internalization of the lesion in damaged PM . Our results suggest that amoebic trophozoites , in response to SLO exposition , fuse the lysosomes to the PM and thus releasing the aSMase along with its content into the extracellular milieu , generating a transient acidic environment necessary for aSMase activity as occurs in mammalian cells [44–46] . Cysteine proteases ( CP ) are present inside the lysosomes and thus should be released along with the cargo when lysosomes fuse with the PM during damage [55 , 56] . The CP activity secreted by HM1-HA and HM1-SM6HA strains was determined in the absence and presence of SLO . The HM1-HA and HM1-SM6HA strains exhibited similar levels of CP activity in the absence of damage , however , when the damage with the toxin is induced , this activity increases by 53% and 130% , respectively ( S5 Fig ) , supporting the observation that the lysosomes indeed fuse with the PM releasing their cargo . EhaSM6 gene code for a protein of 48 . 8 kDa with a predicted signal peptide indicating that the protein is secreted as reported for mammalian aSMases [57 , 58] . The cysteines located at the C-terminal suggests that the aSMase6 could be processed , as occur for the mammalian aSMase where one of the cysteines of this region is susceptible to proteolytic processing [36 , 59] . The HM1-SM6HA strain was used to detect the aSMase6 protein by Western blot using anti-HA antibodies . Despite the high levels of transcript in the over-expressing strain , two weak bands were revealed , suggesting that most of the protein was processed in its C-terminal end . Proteins of 53 . 5 and 50 . 9 kDa were detected in total homogenate , which corresponds to the predicted molecular weight of the unprocessed protein at the C-terminal region without and with N-terminal processing , respectively ( Fig 10A , lane 2 ) . A protein of 53 . 5 kDa was detected in the supernatant , which could correspond to the secreted aSMase6 unprocessed in the C-terminal or with post-translational modifications ( Fig 10A , lane 3 ) . The small differences found in the molecular weight of the proteins in contrast to the predicted requires further investigation of post-translational modifications of this enzyme . To evaluate the C-terminal processing of the amoebic aSMase6 , the HM1-SM6HA-RFP strain that produces the HA-tagged enzyme fused to a Red Fluorescent Protein ( RFP ) [60] in the C-terminal region was used . The activity of the enzyme was not disturbed by the addition of this tag since the supernatant activity is comparable to the HA fused protein described above ( S6 Fig ) . Two proteins of 72 and 23 . 5 kDa were immunodetected with anti-HA antibodies in the total homogenate and one associated with the secreted protein with a molecular weight of 72 kDa ( Fig 10B ) . In the HM1-SM6HA-RFP strain , the 72 kDa protein corresponds to the unprocessed protein , while the 23 . 5 kDa protein corresponds to the HA-RFP tag released from the C-terminal region of the aSM6HA-RFP protein ( Fig 10B ) . These results indicate that there is a processing of the C-terminal end of the aSMase6 . Analysis of the supernatant in the HM1-SM6HA-RFP strain revealed that only one protein of 72 kDa is detected corresponding to unprocessed aSM6HA-RFP protein , in contrast with the HM1-SM6HA strain where two bands were detected . If the N- and C-terminal unprocessed protein is present in the HM1-SM6HA-RFP strain , it is not distinguishable from the processed because the difference between these proteins is only of two kDa . The processed aSMase6 ( N and C-terminal region ) that would have a molecular weight of 47 kDa , in theory , should be more abundant than the remaining unprocessed protein that we observed in both strains . With the overall results , we demonstrate that the processing of aSM6 occurs at the N-terminal region to remove the signal peptide and the C-terminal region as described for the mammalian enzyme . With the compiling results shown in this work , we suggest that the aSMase6 of E . histolytica , responds to the damage to the plasma membrane aiding to maintain cell viability . However , the amoeba has five additional aSMase genes ( Fig 1 ) , which until now their function remains unknown . The transcription levels of the EhaSM1 , EhaSM2 , EhaSM3 , EhaSM4 and EhaSM5 genes besides the EhaSM6 , were evaluated by qRT-PCR in the control ( HM1-HA ) and overexpressing ( HM1-SM6HA ) strains , before and after damage with SLO . Table 1 shows that the overexpressing strain has a 7 . 2-fold increase in the expression of the EhaSM6 gene with respect to the parental strain , while the other 5 genes did not show a significant change in their expression . In response to PM damage with the SLO toxin , both strains showed an increase of the EhaSM6 gene of 4 . 1 and 9 . 4-fold in the control and overexpressing strains , respectively . In response to SLO , there was an overexpression of 4 , 3 and 2 . 5-fold of EhaSM1 , EhaSM2 , and EhaSM5 , respectively . This suggests that , in addition to the aSMase6 , the aSMase1 , aSMase2 , and aSMase5 could also be involved in the repair of membrane damage . In contrast , the genes that code for the aSMase3 and aSMase4 proteins did not show changes in their expression , suggesting they are not participating in the PM repair process or they exhibit a different regulation because of the absence of calcineurin region . When evaluating the expression of the aSMase genes in E . histolytica trophozoites exposed to β-Defensin 2 30 ng/ml ( S4 Table ) or hydrogen peroxide 0 . 5 mM ( Table 2 ) , we found similar results with β-Defensin 2 to those obtained with amoebae exposed to SLO ( Table 1 ) where the genes encode for aSMase3 and aSMase4 did not present changes in their expression . However , with amoebae exposed to hydrogen peroxide , all six aSMase genes showed a significative increase in the HM1-HA and HM1-SM6HA strains ( Table 2 ) , where the EhaSM6 gene again presented the highest expression , even greater than that obtained with SLO and β-Defensin 2 . Overall , these results suggest that aSMases 1 , 2 , 5 , 6 could be involved in the repair of plasma membrane damage caused by pore-forming molecules such as SLO and β-Defensin 2 , while more widespread membrane damage , such as those caused by hydrogen peroxide , all aSMases could be involved .
Entamoeba histolytica is able to invade human tissues by means of several molecules and biological properties related to virulence . Pathogenic amoebae use three major virulence factors , Gal/GalNAc lectin , amoebapore and proteases , to lyse , phagocytose , kill and destroy a variety of cells and tissues in the host [61–64] , while its counterpart is the defensive response of the host that is characterized by humoral and cellular immune reactions [52 , 65 , 66] . The host-parasite relationship is based on a series of interplays between host defense mechanisms and parasite survival strategies . In the present work , we characterized the aSMase activity and its role in PM repair . The genome of E . histolytica has six genes annotated as aSMases , which are actively transcribed and the Ehasm6 gene has the most abundant transcript production rate . Amoebic aSMase sequences present low homology with other previously reported aSMases , such as the aSMases of C . elegans that shows 30% of homology with human and murine aSMases [39] . The predicted amino acid sequences of amoebic aSMases show a low homology with the eukaryotic aSMases reported , however , it presents the essential amino acid residues for the catalysis described for this type of enzymes [35] . There are no reports on aSMases sequences in other protozoa . The amoebic aSMases have a signal peptide in the N-terminal region required for secretion , similar to other aSMases associated to lysosomes such as the human aSMase where a mutant lacking the signal peptide has no enzymatic activity and is not secreted [42 , 67] . The presence of cysteines in the C-terminal region suggests posttranslational processing related to the enzyme activation as reported in other aSMases [36 , 37] . Likewise , the amoebic aSMases have predicted residues for the coordination with cations , which orientates both the enzyme and the substrate in a suitable manner for the reaction as reported for H . sapiens [68 , 69] , C . elegans [39] and M . tuberculosis [70] aSMases . In this work , we focused on the EhaSM6 gene , which exhibited the highest expression in E . histolytica trophozoites under basal conditions of growth . This gene codes for a functional protein as demonstrated in the protein expressed in E . coli , showing activity against sphingomyelin , being stimulated by Mg2+ , inhibited by Co2+ , and exhibited no effect by Zn2+ . This is in contrast with other eukaryotic aSMases that require Zn2+ for activity [41 , 68] . The amoebic nSMases were stimulated by Mn2+ and partially inhibited by Zn2+ [71] , while its counterpart in eukaryotes and prokaryotes responds to Mg2+ [72 , 73] . The aSMase activity was spontaneously secreted by E . histolytica trophozoites under standard culture conditions and exhibited the same effect of bivalent cation observed with the recombinant aSMase6 . The aSMase activity released into the supernatant of trophozoites may be the contribution of the aSMases encoded by the six genes of E . histolytica . The transfected cell line of E . histolytica overexpressing the EhaSM6-HA gene exhibited a 2-fold increase of secreted aSMase activity suggesting that it is mainly due to the overexpression of the EhaSM6 gene . Reports of aSMases from mammalian cells indicate that they are lysosomal enzymes involved in the hydrolysis of sphingomyelin to produce ceramide , an important second messenger lipid associated with several cellular responses to stress , cell growth , differentiation , and apoptosis in eukaryotes [54] . In mammalian cells , the S-SMase is secreted spontaneously and its deficiency has been implicated in pathologies such as atherosclerosis [74 , 75] , while the lysosomal aSMase has been detected after induction of stress and involvement in the repair of damage to the PM [44 , 76] . Therefore , it is of our interest to investigate if the E . histolytica trophozoites have a mechanism of plasma membrane repair mediated by aSMases which would allow it to survive the attack of lytic components of the host defense systems . E . histolytica trophozoites of HM1-HA strain treated with the SLO showed an increase in the secreted aSMase activity , which in turn is 10-fold higher in the aSMase6 over-expressing strain , being these amoebae more resistant to the damage caused by the toxin than the control strain , showing a direct relationship between the level of aSMase secreted and amoebae viability . The aSMase6 over-expressing trophozoites exposed to another pore-forming molecules such as Magainin , β-Defensin 2 and human complement exhibited an significant increase in the secreted aSMase activity which correlated with higher amoebic viability in a Ca+2 dependent process . SLO molecules bind to cholesterol-containing target membranes to assemble , form rings that penetrate into the apolar domain of the lipid bilayer , resulting in the formation of pores of up to 30 nm in diameter . Membrane damage by SLO is basically analogous to channel formers , namely , the C5b-9 complement complex and some human antimicrobial peptides such as β-Defensin 2 [77] . Previous reports have shown in mammalian cells that after the damage to the PM by SLO , there is an intracellular flow of Ca2+ through the lesion that triggers the repair mechanism [8 , 44 , 78] . In E . histolytica trophozoites , extracellular Ca2+ seems to be an indispensable requirement for the secretion of aSMase activity in response to pore forming molecules exposition and for maintenance of viability , since in the absence of Ca2+ , the activity of the secreted enzyme is null , and the viability is brought down completely . These results suggest that the aSMase6 of E . histolytica is secreted extracellularly during the induction of damage to the plasma membrane by SLO in a process Ca2+ dependent , and it is involved in the maintenance of amoebic viability . This process is related to the PM integrity restoration as shown for mammalian cells [44 , 47 , 79] , but thus far , it has not been reported for E . histolytica trophozoites . In mammalian cells , it has been described that after the increase of intracellular Ca2+ , multiple calcium sensors such as synaptotagmin ( Syt ) VII , dysferlin and SNARE that promote lysosomal exocytosis towards the site of the lesion become involved in the PM repair [80–82] . The genome of E . histolytica encodes a large number of calcium-binding proteins , many of these proteins are unique to the amoebae , indicating that it has extensive Ca+2 signaling pathways [83] , but only a few events mediated by this cation have been described . Even less known is the process of repair of damage to the membrane of E . histolytica . The SNARE complex and calcium-binding proteins , such as EhCaBP1 [84] , could be involved in the exocytosis of lysosomes and the formation of endosomes in response to the increase in intracellular calcium concentration . There is evidence that some components of the SNARE complex can function in the trafficking of vesicles to the PM [85] , as well as the EhRab7 and EhRab11 proteins that are involved in the biogenesis , acidification and trafficking of lysosomes , as well as in the trafficking of late endosomes and phagosomes [86–88] . It is likely that E . histolytica Rab proteins facilitate and regulate the kinetics of anchoring and pairing of the SNARE complex and promote exocytosis of lysosomes to the site of PM damage , similar to the already reported mechanism in mammalian cells [89 , 90] , however , this still remains to be determined in E . histolytica . Hydrogen peroxide , an important mediator of acute lipid oxidative injury , alters the fluidity and generates a leaky plasma membrane-associated with lipid peroxidation [91] , also induced an increase of secreted aSMase in E . histolytica , but to a lesser extent . Hydrogen peroxide , a primary form of ROS in mammalian cells has been proposed as second mesangers in mammalian cells to mediate cellular responses , activating the aSMase translocation and activation [92] . The mechanism by which hydrogen peroxide induces aSMase secretion in E . histolytica remains to be investigated . The C-terminal processing of human aSMases has been reported , which may exist associated with lysosomes or released extracellularly [93] , which arise from the post-translational modifications during its vesicular trafficking and maturation [37 , 41 , 94] . To address the C-terminal processing and to explain the weak signal with anti-HA antibodies , the HA labeled aSMase6 was fused to the optimized Red Fluorescent Protein ( RFP ) for E . histolytica expression [60] , and immunodetected with anti-HA antibodies . A protein corresponding to HA-RFP was detected corroborating the C-terminal processing . The C-terminal processing found in mammalian cells during the maturation of the enzyme not only modifies its molecular weight but also implying that this processing involves the elimination of the Cys 629 of the enzyme which increases its activity [36 , 59] . It is still necessary to deepen into the processing and activation of the aSMases of E . histolytica . Confocal microscopy analysis of trophozoites exposed to SLO at early times showed that there is a migration of lysosomes to the PM and the formation of endosome structures in a "patch-like" arrangement which are transient structures that prevent the exit of the cytoplasmic components . Although these structures have been observed in mammalian cells after the induction of PM damage , they have not been well characterized . It has been suggested that they arise from the rapid and massive formation of endocytic vesicles , which accumulate near the site of the lesion [80 , 95–97] . The secretion of the aSMase by exocytosis of lysosomes is accompanied by the release of its content to the extracellular medium , determined by an increase in the activity of secreted CPs and generating an acidification of the extracellular medium on the periphery of the trophozoites ( S7 Fig ) after the SLO treatment allowing the aSMase activity on the sphingolipid substrate on the PM . The lysosomes are secretory vesicles that can release their content which may include lysosomal proteases [46] and transient acidification can be generated extracellularly at sites of lysosomal exocytosis [98 , 99] . Also , the role of released proteases from the lysosomes and their involvement in the resealing process of the plasma membrane needs to be elucidated , as described for mammalian cells [46] . The lysosome biogenesis that controls transport , maturation , and secretion of CPs and probably aSMases that could have an important role in the pathogenesis as well as housekeeping functions unrelated to parasitism and virulence in E . histolytica [100] . The increase in endosome formation after treatment with SLO is higher in aSMase6-HA overexpressing trophozoites compared with the control strain and trophozoites without treatment . This process has been observed after a few minutes of SLO exposure in NRK , HeLa or Jurkat cell-lines [44 , 101] . After massive endosomes formation , they fuse to lysosomes because they share fluorescence signal between lysosomes and endosomes following the vesicular traffic route , which has been described to be fully functional in amoebas , even though it lacks morphologically defined organelles such as the endoplasmic reticulum , Golgi apparatus , instead the amoebae have a high content of vesicles , many of which are associated with the functions of these organelles [102 , 103] . Exocytosis of aSMase by wounded cells promotes endocytosis and plasma membrane repair by the generation of the secondary messenger ceramide [44] . Sphingomyelin enriched lipid domains or ‘‘rafts” may serve as substrate pools for SMase-induced formation of ceramide microdomains that act as platforms from which these signal transduction cascades originate [104] . By analogy , aSMases could be involved in the production of ceramide associated with the stress responses in E . histolytica during the invasive process of the host [105 , 106] , suggesting that conversion of plasma membrane sphingomyelin to ceramide by this lysosomal enzyme promotes lesion internalization . TEM analysis shows the polarization of vesicles after exposure to SLO toxin , suggesting the participation of these vesicles in maintaining the integrity of the membrane at early damage-associated events . As the damage progresses , it is possible to observe an increase in polarized endosomes which internalize fragments of the damaged membrane , and the amoebae that repaired successfully the damage remain viable . In summary , the results presented here , suggest three main events after the amoebic membrane injury: first , the internalization of calcium through the lesion is important in the repair process that activates the lysosome exocytosis and aSMase release to initiate the repair process mechanism; second , the generation of patches formed from the fusion of lysosomes and endosomes at the damaged site , which momentarily prevent cell lysis , and third , the endocytosis-dependent generation of ceramide by the aSMase activity . The data presented here are consistent with the repair mechanism mediated by aSMase in mammalian cells [44 , 107] . This is the first report showing that E . histolytica has the machinery to repair PM damage mediated by aSMase6 . Secretion of aSMase has been detected in mammals after exposing the cells to different types of stress , in particular , it has been shown to be activated in response to damage to the PM , such as the caused by bacterial toxins , which generate small ( 0 . 5–5 nm ) or large ( 20–100 nm ) pores dependent on the concentration [44 , 47 , 108]; or viral , bacterial and parasitic pathogens ( EBOV , Neisseria gonorrhoeae , Staphylococcus aureus , Pseudomonas aeruginosa , Trypanosoma cruzi ) , which are associated with regions rich in cholesterol destabilizing the PM , thus generating lesions [109–113] , leading to expect that amoebae could respond to PM damage caused by other molecules besides SLO toxin . During an intestinal human infection , several pore forming molecules are produced , where β-defensins are the most common [48] . The β-defensin 2 and the antimicrobial peptide magainin , isolated from Xenopus laevis , which is a cathelicidin , similar to the antimicrobial peptide LL-37 , have been reported to have a lytic effect in E . histolytica trophozoites , interacting directly with the anionic phospholipids of the plasma membrane through the amphiphilic α-helix and forming permeable ion channels , resulting in depolarization , irreversible cytolysis and finally amoebic cell death [19 , 114 , 115] . Our results indicate that the overexpressing strain HM1-SM6HA is more resistant to damage with this peptide in comparison with the parental strain , which also correlated with higher secreted aSMase activity . Interestingly , after SLO or β-Defensin 2 exposure there is an increase of gene expression of EhaSM1 , EhaSM2 , EhaSM5 y EhaSM6 , suggesting the potential participation of other members of aSMase gene family besides aSMase6 in the PM damage repair mechanism proposed in this work . Also , there is no change in gene expression for EhaSM3 and EhaSM4 genes after SLO treatment . The four EhaSM genes that respond to membrane damage , each one possesses a single calcineurin domain , while the two genes that do not respond to damage lack these domains . Calcineurin is a protein phosphatase regulated by Ca2+/calmodulin conserved in eukaryotes , and has been associated with stress response in yeast; this is activated when there is an increase in the concentration of cytosolic Ca2+ in response to internal or external signals , causing the activation of the Ca2+-calmodulin-binding domain and then subsequently binds to calcineurin , thus dephosphorylating the target proteins that modulate various biological processes that allow cell survival [116–118] . E . histolytica has genes that encode for calcineurin , but to date , there are no reports describing the processes in which they participate . Unlike what happens with pore-forming molecules , exposure to hydrogen peroxide increased the expression of the six genes that encode for aSMases in E . histolytica . In all the conditions evaluated , the EhaSM6 gene exhibited the highest levels of induction . The above suggests that the expression of aSMasas in E . histolytica could be selective to different types of cellular stress . The defensive response of the host is characterized by humoral and cellular immune reactions . The presence of amoebic trophozoites causes the infiltration of neutrophils , lymphocytes , and macrophage , and serum factors are amoebicidal through the activation of the alternative complement pathway [119] . While host cells elaborate diverse mechanisms for pathogen expulsion , amoebae have also developed complex strategies to modulate host immune response and facilitate their own survival [52 , 120] . In addition to virulence factors , there are other amoebic molecules , termed virulence determinants , that participate in the pathogenicity process by promoting the survival of parasites while confronting host defenses , allowing the parasites to harm the host [33] . It is not surprising that E . histolytica possess a mechanism of damage repair to the plasma membrane mediate by aSMase for maintaining trophozoites viability and to confront with various lytic agents such as antimicrobial peptides , bacterial toxins , and the complement system . Recently we reported that the nSMase3 of E . histolytica participate in hemolytic and cytotoxic activities , while , the nSMase1 and nSMase2 are involved in the cytopathic activity [121] . There are still new factors that remain to be elucidated and characterized to fully comprehend the virulence mechanism in this parasite . Further characterization of aSMases in E . histolytica is necessary to uncover their role in virulence as well as in cell signaling . The study of the relationship between aSMases and virulence is currently in process . Taking together all the results presented in this work , the damage of the trophozoites of E . histolytica with a sub-lethal concentration of SLO , induce the entry of Ca2+ , which favors the migration of the lysosomes to the periphery of the cell , fuses with the plasma membrane and pour their content of aSMases to the outside of the cell . The secreted aSMases produce ceramide favoring the internalization of the lesion for its degradation in phagolysosomes . The pores generated by the PM damage are rapidly blocked by patch-like structures of lysosomes and endosomes that prevent the lysis of the trophozoite and immediately begin the internalizing the lesion . The aSMase6 overexpression favors the repair of the lesion and the survival of the trophozoites of E . histolytica . The plasma membrane damage repair mediated by aSMase in E . histolytica is summarized in Fig 11 .
Trophozoites of E . histolytica HM-1:IMSS and transfected trophozoites of the same strain were cultured under axenic conditions in Diamond’s TYI-S-33 medium supplemented with 10% adult bovine serum ( Microlab Laboratories , Mexico ) at 36 °C [122] . The transfectant strains were grown in the presence of G418 ( Sigma-Aldrich , St . Louis , MO , USA ) , as a selective agent . The encoding sequence of the amoebic aSMase6 ( EHI_125660 ) was obtained by PCR from E . histolytica HM-1:IMSS cDNA using sense and antisense oligonucleotides containing the appropriate restriction sites ( S1 Table ) . PCR product was cloned into pGEM-T ( Promega ) and subcloned into pRSET ( Invitrogen ) . The resulting plasmids pRSET-aSM6 encoding the EhaSM6 fused to a 6X histidine tag was verified by sequencing and used to transform E . coli BL21 AI cells ( Invitrogen ) . For overexpressing the recombinant protein ( rEhaSM6 ) , BL21 AI culture was grown to a cell density of ∼0 . 4–0 . 6 OD 600 at 37 °C . To maximize the yield of recombinant soluble protein , E . coli cells were cultured for 4 h at 25 °C with 0 . 2% L-arabinose as an inductor . Soluble rEhaSM6 was purified under native conditions for the detection of aSMase activity . Briefly , harvested E . coli BL21 AI cells were resuspended with 50 mM NaH2PO4 , 300 mM NaCl , 10 mM imidazole ( pH 8 ) at 2 ml per gram of wet weight . Lysozyme was added to the cell suspension at a final concentration of 10 mg/ml , and the mixture was incubated 30 min , followed by three freeze-thaw cycles . The cell lysate was centrifuged at 10 , 000×g for 20 min , and the rEhaSM6 was purified from the supernatant fraction using Ni-NTA Agarose ( Qiagen ) as directed by the manufacturer . The purified proteins were applied to a Hi-Trap desalting column ( Amersham Pharmacia Biotech ) and eluted with 100 mM Tris–HCl , 20 mM NaCl pH 7 . 5 buffer . The activity of the purified recombinant aSMase6 was determined using the Amplex Red Sphingomyelinase Assay Kit ( Molecular Probes ) as directed by the manufacturer . Briefly , a two-step assay was performed in 50 mM sodium acetate pH 5 . 0 containing sphingomyelin ( 0 . 5 mM ) , 2% Triton X-100 , with 0 . 5 μg of the recombinant protein at 37 °C for 1 h . Subsequently , 100 mM Amplex Red reagent , 2 U mL−1 horseradish peroxidase , 0 . 2 U mL−1 choline oxidase and 8 U mL−1 alkaline phosphatase were added and mixed with 25 μL of 100 mM Tris–HCl ( pH 8 . 0 ) . After incubation for 20 min at 37°C; the fluorescence at 582 nm was measured , with excitation at 556 nm , using a Fluoroskan Ascent FL ( Thermo Scientific ) luminescence spectrometer . In all the assays , background from the enzyme-free controls were routinely subtracted from the activities of samples containing enzyme extracts . The EhaSM6 coding sequence was PCR amplified from cDNA using sense and antisense oligonucleotides containing the appropriate restriction sites ( S1 Table ) . A sequence tag consisting of three tandem repeats of hemagglutinin ( HA ) peptide was added at the C-terminus of aSMase by cloning the PCR products into pEhEx [123] . PCR fragments were digested with SmaI and XhoI and ligated into corresponding sites of the expression vector pEhEx . HM-1:IMSS trophozoites were transfected with the construct paSM6HA by liposome-mediated transfection as previously described [124 , 125] . EhaSM6 over-expressing transfectants were selected with 40 μg/mL of G418 and maintained as stable cell lines . Over-expression of EhaSM6 gene was determined by quantitative real-time PCR ( qPCR ) . Also , the paSMHA-RFP construction was designed to obtain the over-expressing strain of HA-tagged sphingomyelinase fused to a Red Fluorescent Protein ( RFP ) [60] in the C-terminal region ( HM1-SM6HA-RFP ) . Trophozoites in the exponential phase of growth were harvested . 3x105 amoebae were placed per well of a 24-well cell culture plate and incubated at 37 °C in TYI-S-33 for 2 h . Adhered trophozoites were washed two times with Ca2+ free DMEM medium ( Gibco , Life Technologies , Carlsbad , CA ) and further incubated in 500 μL of DMEM medium containing 1 . 8 mM Ca+2 ( Gibco , Life Technologies , Carlsbad , CA ) at 37°C for the indicated time . After each incubation time , the collected supernatant was centrifugated at 15 , 890 x g for 3 min to remove any detached cells and the secreted aSMase activity was determined using the Amplex Red Sphingomyelinase Assay Kit , as described above . Supernatants from the secreted activity of aSMase were collected and the activity of alcohol dehydrogenase was evaluated as described previously [126] . Briefly , the buffer contains a 50 mM glycine/ NaOH buffer , pH 9 . 5 , NADP+ 0–2 mM , 20 mM 2-propanol , and 150 μL of the sample is added to give a final volume of 200 μL . The reduction rate of NADP+ was evaluated at an absorbance of 340 nm at 25 °C for 60 min . Adhered amoebae in a 24 well cell culture plate were washed with DMEM medium without Ca2+ , and then replenish with complete DMEM medium ( 1 . 8 mM Ca2+ ) pre-warmed at 37 °C and PM damage was performed by adding streptolysin-O ( SLO ) ( Sigma-Aldrich ) , the antimicrobial peptides Defensin and Magainin II ( Sigma-Aldrich ) , and Triton X-100 ( Sigma-Aldrich ) . After exposing the trophozoites to various incubation times , the viability of trophozoites was determined ( see below ) , and cell-free supernatants were collected and assayed for the secreted aSMase activity as described above . Assays for susceptibility to human complement lysis were carried out with trophozoites during the logarithmic phase of growth . A previously published protocol was followed with some modifications [127 , 128] . Briefly , a total of 1×106 trophozoites were incubated in buffer ( PBS , 0 . 5 mM MgCl2 , 1 . 25 mM CaCl2 ) with 50% normal human serum for 20 min at 37 °C . As a control for amoebic viability , trophozoites were incubated with heat-inactivated normal human serum ( 30 min at 56 °C ) . Trophozoites were centrifuged at 804 xg for 5 min , resuspended in 100 μl of PBS and stained with 0 . 2% Trypan blue dye ( Microlab ) to assess cell viability . The viability of the amoebae was measured by the exclusion of trypan blue dye . The average number of dead trophozoites that resulted from the incubation with heat-inactivated serum was subtracted from the average number of dead parasites incubated with normal human serum . Adhered amoebae in a 24-well cell culture plate were washed with DMEM medium without Ca2+ , then replenish with complete DMEM medium ( 1 . 8 mM Ca2+ ) pre-warmed at 37 °C and , different concentrations of H2O2 ( 0 , 0 . 2 , 0 . 5 and 1 mM ) were added . The amoebas were incubated for 10 min to induce oxidative stress in the cells . After this time , the aSMase activity present in the supernatants and the viability of the trophozoites by trypan exclusion were evaluated . To evaluate the viability of trophozoites after SLO o Magainin II exposition , three methods were used: i ) The viability of amoebae was measured by the exclusion of trypan blue dye . After plasma membrane damage , trophozoites were centrifuged at 804 xg for 5 min , resuspended in 100 μL of PBS and stained with 0 . 2% Trypan blue dye ( Microlab ) . Viability was determined by counting the number of cells that did not incorporate the dye by light microscopy and counting 100 total trophozoites; ii ) Live/Dead kit ( Molecular Probes ) was used according to the manufacturer's instructions . In brief , calcein-AM , and ethidium homodimer were added to trophozoites previously washed in PBS . Cells were incubated in these reagents for 10 min at room temperature and fluorescence was examined by epifluorescence microscopy ( Zeiss Axioskop 40 ) . For the staining of lysosomal compartments , live amoebae were growth in TYI-S-33 medium containing 2 μM Lysotracker Red DND-99 ( Molecular Probes ) for 14 h at 36°C . To label early endosomes , the amoebae stained with Lysotracker in exponential growth were harvested and transferred to coverslips placed in a 24 well cell culture plate and incubated at 37 °C for 2 h . The trophozoites were washed twice with Ca2+ free DMEM and further incubated in 500 μL of DMEM medium supplemented or not with SLO , and simultaneously exposed to 0 . 5 μg μL-1 fluorescein isothiocyanate ( FITC ) -labeled dextran ( 10 , 000 MW , Molecular Probes ) followed by incubation at 37 °C for 1 , 3 or 10 min . The cells were fixed for 10 min at room temperature with 3 . 7% paraformaldehyde in PBS and then incubated for 10 min at room temperature with 2 μM Höescht 33342 ( Sigma-Aldrich , St . Louis , MO ) . Trophozoites were washed extensively with PBS , mounted using Vectashield ( Vector Laboratories , Inc , USA ) and recorded using a Zeiss LSM 700 confocal microscope . Proteinase activity was measured using the synthetic peptide ZArg-Arg-pNA ( Bachem ) as a substrate [129] . 20 μl of cell-free supernatants collected from the secreted aSMase activity assays were combined with 180 μl of PBS and 2 μl of the 10 mM stock substrate for 2 h at 37 °C , reading every 15 min . The release of p-nitroaniline was measured in a microplate reader ( Multiskan Go Thermo Scientific ) at 405 nm . One unit of activity is defined as the number of micromoles of substrate hydrolyzed per min . RNA was isolated from 5×106 log-phase E . histolytica trophozoites using the Trizol reagent ( Invitrogen ) following the manufacturer’s protocol including DNase I ( Qiagen ) treatment . RNA was quantified , purity checked by absorbance at 260 and the 260/280 nm ratio respectively using a GeneQuant spectrophotometer ( GE Healthcare ) , and integrity of isolated RNA was verified by gel electrophoresis . For first-strand cDNA synthesis , 3 μg of total RNA ( DNA-free ) isolated from amoebic trophozoites was reverse transcribed using Oligo ( dT ) and reverse transcriptase from the SuperScript II RT-system ( Invitrogen ) according to manufacturer’s instructions . For qPCR experiments , sense and antisense primers were designed ( S1 Table ) to amplify approximately 150 base pairs of the target gene sequences . qPCR was performed using the Step One Real-Time PCR System ( Applied Biosystems ) and Fast SYBR Green Master Mix ( Applied Biosystems ) following the manufacturer’s protocol . Relative quantification was carried out using the delta-delta Ct method [130] and E . histolytica gapdh gene transcript was used as house-keeping control . Two biological replicates were analyzed in triplicates . The trophozoites were harvested at the exponential phase of growth ( 5×106 cells/ml ) and washed twice with PBS pH 7 . 0 . Trophozoites were re-suspended in lysis buffer ( 100 mM Tris–HCl pH 7 . 4 supplemented with 0 . 05 mM E64 and 1% Triton X-100 ) and disrupted with a hand homogenizer . Amoebic cell extracts were separated by SDS-PAGE and transferred to a nitrocellulose membrane ( Hybond , Amersham Biosciences ) . Western blotting was performed using mouse anti-HA monoclonal antibody ( 1:500 ) ( Invitrogen ) as primary antibody and horseradish peroxidase-conjugated goat anti-mouse IgG ( Amersham Pharmacia Biotech ) as a secondary antibody and visualized using the alkaline phosphatase conjugate substrate kit ( Bio-Rad ) . For equal amounts of protein , the concentration was determined by the DC Protein Assay ( Bio-Rad ) . Trophozoites that had been exposed to SLO were washed once with PBS and twice with 0 . 1 M sodium cacodylate buffer at 37 °C and fixed for 3 h with 2 . 5% glutaraldehyde in 0 . 2 M sodium cacodylate buffer , pH 7 . 4 . Fixed trophozoites were washed twice with 0 . 1 M sodium cacodylate buffer , post-fixed with 1 . 0% osmium tetroxide in 0 . 1M sodium cacodylate at 4 °C , dehydrated with ethanol at increasing concentrations and treated with propylene oxide . Trophozoites were embedded in EmBed 812 epoxy resins , polymerized blocks were cut using an ultramicrotome , and thin sections were stained with 2% uranyl acetate and 2% lead citrate . Trophozoites morphology was analyzed by transmission electron microscopy ( TEM ) with a JEM-1010 JEOL at 80 keV . | The host-amoeba relationship is based on a series of interplays between host defense mechanisms and parasite survival strategies . While host cells elaborate diverse mechanisms for pathogen elimination , Entamoeba histolytica trophozoites have also developed complex strategies to counteract host immune response and facilitate its own survival while confronting host defenses . E . histolytica exposed to pore-forming proteins such as β-Defensin 2 , human complement and Streptolysin O ( SLO ) , increases the activity of secreted aSMase , which is related to greater amoebic viability . Other agents that affect plasma membrane ( PM ) may also increase secreted aSMase but to a lesser extent . SLO form pores in the PM of E . histolytica trophozoites that initiates the uncontrolled entry of Ca2+ , recognized as the primary trigger for cell responses which favors the migration of the lysosomes to the periphery of the cell , fuses with the PM and release their content , including aSMase to the external side of the cell . The secreted aSMase favoring the internalization of the lesion for its degradation in phagolysosomes . During the early stages of PM damage , the pores are rapidly blocked by patch-like structures that prevent the lysis of the trophozoite and immediately begin internalizing the lesion . The aSMase6 overexpression favors the repair of the lesion and the survival of E . histolytica trophozoites . Pore-forming proteins induced an increase in the expression of EhaSM1 , 2 , 5 and 6 genes , meanwhile oxidative stress induced an increase in all of them . Here we report , for the first time , that E . histolytica possess a mechanism for PM damage repair mediated by aSMase similar to the system described in mammalian cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"trophozoites",
"parasite",
"groups",
"lysosomes",
"medicine",
"and",
"health",
"sciences",
"vesicles",
"parasitic",
"protozoans",
"parasitology",
"physiological",
"processes",
"apicomplexa",
"protozoans",
"cellular",
"structures",
"and",
"organelles",
"endosomes",
"entamoeba",
"histolytica",
"amoebas",
"cell",
"membranes",
"eukaryota",
"cell",
"biology",
"physiology",
"secretion",
"biology",
"and",
"life",
"sciences",
"organisms"
]
| 2019 | Plasma membrane damage repair is mediated by an acid sphingomyelinase in Entamoeba histolytica |
Despite a growing appreciation of their vast diversity in nature , mechanisms of speciation are poorly understood in Bacteria and Archaea . Here we use high-throughput genome sequencing to identify ongoing speciation in the thermoacidophilic Archaeon Sulfolobus islandicus . Patterns of homologous gene flow among genomes of 12 strains from a single hot spring in Kamchatka , Russia , demonstrate higher levels of gene flow within than between two persistent , coexisting groups , demonstrating that these microorganisms fit the biological species concept . Furthermore , rates of gene flow between two species are decreasing over time in a manner consistent with incipient speciation . Unlike other microorganisms investigated , we do not observe a relationship between genetic divergence and frequency of recombination along a chromosome , or other physical mechanisms that would reduce gene flow between lineages . Each species has its own genetic island encoding unique physiological functions and a unique growth phenotype that may be indicative of ecological specialization . Genetic differentiation between these coexisting groups occurs in large genomic “continents , ” indicating the topology of genomic divergence during speciation is not uniform and is not associated with a single locus under strong diversifying selection . These data support a model where species do not require physical barriers to gene flow but are maintained by ecological differentiation .
Molecular sequence analyses of microbial populations commonly reveal discrete clusters of sequence diversity indicative of closely related , but distinct , coexisting , lineages [1]–[9] . Such clusters are sometimes given the status of species , especially when they are shown to be ecologically distinct [6] . Resolving the evolutionary mechanisms that cause the formation and maintenance of these clusters holds the key to understanding the process of speciation in clonally reproducing , asexual microorganisms [10] . In the absence of geographic barriers , speciation depends upon the balance between gene flow holding lineages together and selection pulling them apart [11] , [12] . For asexual microorganisms , two primary theoretical models explain the formation of sequence clusters , each tipping the scale in the opposite direction of the recombination-selection balance . The first emphasizes the importance of selection driving ecological specialization . This model predicts that the persistent coexistence of sequence clusters can result from sequential selective sweeps of adaptive mutations to different niches [13] . Persistent “ecotypes” are held together by the cohesive force of genetic drift or periodic selection and kept apart by selection for niche-specific adaptations or against niche-specific maladaptations [13] , [14] . This model can incorporate low levels of gene flow of universally adaptive genetic material without limiting the effects of periodic selection differentially occurring in the ecotypes [15] . Ecological differentiation , correlated with decreased recombination , has been observed using multilocus sequence markers [6] and recently through whole genome analysis in Escherichia coli strains that have evolved to inhabit different environments [16] . A second model relies on barriers to recombination in the absence of selection to explain persistent sequence clusters [17] . It demonstrates that clusters will form if the effects of recombination are lower than mutation . In this neutral model , when recombination is greater than mutation , recombination plays a cohesive role that is strong enough to prevent the formation of persistent independent clusters of sequences , unless there is a significant physical barrier to gene flow between them [17]–[19] . For microorganisms , many such barriers can be hypothesized [20] . The most often mentioned and broadly distributed among microbial taxa is caused by mismatch repair recognition [21] , which reduces the frequency of homologous recombination ( HR ) between divergent sequences . This type of barrier has been shown to lead to the formation of persistent diverging and independent clusters of sequences by allowing recombination within but not between groups that diverge through genetic drift [17] , [22] . Whether recombination barriers or selection play the primary role in driving divergence of sequence clusters and whether the balance between these processes results in the maintenance of independent species in microorganisms is a controversial topic [16] , [23] , [24] . Many Bacteria and Archaea exhibit significant rates of HR and other forms of horizontal gene flow [25]–[28] . However , for each of the known mechanisms of horizontal transfer ( transduction , transformation , and conjugation ) only a small region of the chromosome may be transferred with each event . Whether this level of recombination among coexisting strains is strong enough to overcome ecological specialization and periodic selection and how the balance between recombination and selection will affect the topology of speciation across the chromosome in microorganisms is only beginning to emerge with the advent of whole genome sequencing [16] , [29] . To examine the mechanisms of divergence and maintenance of independent species in Archaea , we sequenced the complete genomes of 12 strains of the thermoacidophilic Archaeon Sulfolobus islandicus from a single hot spring from the Mutnovsky Volcano region in Kamchatka , Russia . This location was selected because the S . islandicus population from the Mutnovsky volcano has been shown to be geographically isolated [30] , [31] , thus allowing us to investigate evolutionary processes occurring within well-defined geographic boundaries .
The 12 S . islandicus genomes had an average size of 2 . 64 Mb ( Table 1 ) , of which approximately 86% was shared by all strains . The genomic sequences of the strains were very similar , with pairwise genetic distances ranging from 0 . 01% to 0 . 35% ( Table 1 ) . Shared genomic regions were used to infer the clonal genealogy using ClonalFrame , which accounts for the possibility of HR disrupting the signal of vertical genetic inheritance [35] . The relationships reconstructed using the full genomes ( Figure 1B ) confirmed those resolved by MLSA ( Figure 1A ) , and further resolved bifurcations among strains that had unclear relationships using only seven marker loci . Based on this clonal genealogy , HR events were reconstructed using ClonalOrigin [36] . Overall , ClonalOrigin estimated that each nucleotide was substituted by recombination with a higher probability than mutation with a ratio estimated between 1 . 8 and 13 [37] . This value is above the threshold predicted to prevent divergence among sequence clusters in simulations of neutral populations with a small effective population size of 105 [17] . Using ClonalOrigin we were able to map recombinant fragments between donor and recipient genomes . As shown in Figure 2 , the pattern of recombination between branches of the tree differed from the values expected under the coalescent model with constant recombination . The observed number of HR events was higher than expected within two groups , but lower than expected between them . The first group contains seven strains ( M . 16 . 27 , M . 16 . 46 , M . 16 . 13 , M . 16 . 23 , M . 16 . 43 , M . 16 . 47 , and M . 16 . 30 ) , hereafter called the Blue group , and the second contains three strains ( M . 16 . 4 , M . 16 . 40 , and M . 16 . 02 ) , hereafter called the Red group . This pattern of higher gene flow within than between two groups ( Red and Blue ) coexisting in the same hot spring fits the biological species concept [10] , [38] , [39] . The absolute numbers of events ( Figure S1 ) show that there are rare transfers between divergent sets of strains in the Red and Blue groups , indicating that there is not a complete barrier to gene flow , but rather a relative decrease in the number of recombination events between groups . The two intermediate and nearly identical strains ( M . 16 . 12 and M . 16 . 22 ) receive recombinant fragments at a higher relative frequency from the Blue group than from the Red group ( Figure 2 ) but serve as a donor at a lower frequency than expected to both the Blue and Red groups . We excluded them from further analysis because it was unclear whether they should be included in the Blue group or kept separate . Microorganisms such as S . islandicus engage in promiscuous , non-homologous gene flow through horizontal gene transfer , which is often mediated by integration of a diversity of mobile elements such as viruses and plasmids [30] , [40] . To test whether there is evidence for non-homologous gene flow among populations , we mapped variation in genome content onto the core gene phylogeny . In total , we identified 48 non-core segments longer than 5 kb that are found in only some of the 12 strains ( Table S4 ) . Thirty-four of these can be explained by a single gain or loss event on a single branch along the phylogenetic tree ( Figure 1B ) , including 12 gains by a single genome ( Table S4 ) . The remaining 14 non-core segments have distributions that require multiple events to be explained ( e . g . , gene flow between strains or gene gain followed by differential gene loss ) . Only four of these ( 36 , 38–40 in Table S4 ) could result from exchange between strains in the Blue and Red groups . The distribution of non-core regions of the 12 S . islandicus genomes is therefore consistent with a low level of non-homologous gene flow between the two groups , analogous to the pattern of homologous gene flow observed above . The Red and Blue groups identified in Figure 2 could be either diverging over time or could have evolved independently and started to converge through a process of multiple migration with introgression [41] , [42] . To differentiate these two hypotheses , we studied the distribution of recombination events between the two species through coalescent time as inferred by ClonalOrigin . Figure 3 shows that the percentage of the recombinant events from the Blue to the Red groups is decreasing over coalescent time ( Figure 3A ) . This indicates that the groups are progressively diverging in a manner that is consistent with ongoing speciation . As shown by MLSA in Figure 1A and Table S3 , high levels of differentiation do not occur between hot springs . This indicates that the speciation we have observed in one spring has in fact occurred at a larger scale within the Mutnovsky population , which has been shown to be geographically isolated when compared to similar populations from North America [30] , [31] . We cannot exclude the possibility that one of these two groups initially diverged elsewhere and migrated to the Mutnovsky Volcano . The decrease in gene flow between groups suggests that if this were the case , ongoing migration is decreasing over time . Our observation of recent , low levels of gene flow between groups demonstrates the coexistence and interaction of these groups of strains . Therefore , we investigated mechanisms that either drive or maintain the independence of these groups as they coexist . Although recombination rates in this Archaeon , as in other microorganisms , are low relative to sexual eukaryotes , the two primary models for speciation in microorganisms predict that either barriers to recombination , or diversifying selection , between the two types is necessary to explain their maintenance and ongoing divergence . We first investigated possible physical barriers to recombination between the two species . Neutral divergence of lineages can occur when there is a decrease in recombination with genetic divergence resulting from mismatch recognition [21] . This relationship has been identified in many bacterial and eukaryotic species [43] , [44] . We tested this hypothesis by examining whether regions of the chromosome with higher divergence exhibited lower frequencies of recombination than those with lower divergence . This analysis therefore looks at the variation in rates of recombination along the chromosome rather than between particular partners as above . We found no correlation between genetic divergence and recombination frequency when all recombination events from the populations were pooled ( slope 0 . 016 , R2 = 6 . 3e-07 , Figure S2A ) . This contrasts with ( i ) the same analysis in the Bacillus cereus group [36] , where a negative correlation was found ( slope −3 . 28 , R2 = 0 . 29 , Figure S2B ) , ( ii ) experimental data for Bacteria tested over the same range of genetic distances ( Figure S2C ) [17] , [43]–[46] , and ( iii ) metagenomic analysis of the Archaeon Ferroplasma [47] . This lack of correlation between sequence divergence and rate of recombination is , however , consistent with the mechanisms of recombination reported for another Sulfolobus species , S . acidocaldarius , in which short tracts of 20–22 nt are incorporated during transformation with sequence identity of only 2–3 nt required at either end of the import [48] . In addition , the lack of an identified mutSL system in this species may result in the failure to prevent recombination between divergent sequences [49] . Homologous recombination and non-homologous gene flow through mobile elements has been observed in laboratory cultures between strains of S . acidocaldarius [50] , [51] . In this system , DNA transfer between cells is thought to occur through pilin-mediated aggregation and conjugation [51] , [52] . We considered physical barriers that could prevent aggregation and conjugation among sympatric species of S . islandicus . The most highly variable sequence in the genome is that of the large subunit of the S-layer protein that covers the cell surface of S . islandicus and other microorganisms [53] . With one exception , alleles of the S-layer cluster into three distinct groups: Blue , Red , and the strains M . 16 . 12 and M . 16 . 22 . The notable exception , M . 16 . 30 , is the most divergent Blue strain , which possesses an allele most similar to the Red group ( Figure S3 ) . The incongruence of the M . 16 . 30 strain possessing the Red allele but showing a history of recombination with the Blue strains suggests that S-layer divergence did not pose the barrier to recombination that resulted in the divergence of the Red and Blue groups . The Red and Blue S-layer alleles are highly divergent ( with 12% nucleotide substitutions ) compared to the rest of the genome , especially in the surface regions of the protein that are shown to be highly glycosylated in other species [54] . The allele in the Blue group appears to have been acquired by horizontal gene transfer , as it does not fit the core gene phylogeny of other sequenced Sulfolobus strains ( Figure S3 ) [30] . Two amino acid changes are identified between the Red and Blue alleles of the ups pili believed to be responsible for aggregation and possible DNA exchange in contact with the S-layer [52] , [55] . We cannot exclude the possibility that the history of recombination shown in M . 16 . 30 is not consistent with its current genotype because of the recent acquisition of the novel allele by the Blue group or the recent transfer of this allele between a strain belonging to the Red group and M . 16 . 30 . In this case , the S-layer may serve as a barrier to current gene transfer , analogous to the prezygotic barrier in macroorganisms , possibly reinforcing the isolation between lineages following their initial differentiation . The specificity of restriction enzymes could serve as barrier to recombination between species [20] . Because restriction enzymes are difficult to recognize bioinformatically , we investigated the distribution of methyltransferases that can confer specific protection to their linked restriction systems among the genomes of our 12 S . islandicus strains . While variation exists between strains , none of it was consistent with the distinction between the Blue and Red groups of S . islandicus strains , indicating that these systems are not providing a barrier to genetic exchange between the two groups . Although there may be physical barriers to the transfer and integration of DNA between the Red and Blue groups of strains that we have not yet identified , the set of possible mechanisms we have tested thus far do not appear to contribute to the decrease in gene flow between the Red and Blue groups that we observe . In the absence of physical barriers to gene flow , diversifying selection may be driving these two incipient species apart or maintaining their differentiation . We note that the relatively low frequency of recombination observed in microorganisms , as compared with sexual eukaryotes , facilitates selection-driven divergence . We examined the differences in genotypes and phenotypes between the two groups to identify possible loci under differential selection that could cause ecological differentiation . First , we estimated the ratio of non-synonymous to synonymous rates of substitutions ( dN/dS ) between the Blue and Red groups and performed the MacDonald-Kreitman test [56] for all core genes . No indication of diversifying selection ( values significantly greater than 1 . 0 ) was found . Overall , our estimate of the average dN/dS was 0 . 42 , which is consistent with recent divergence of these two groups in which purifying selection has not yet cleared mildly deleterious non-synonymous substitutions [57] , [58] . Genomic loci that are under differential selection between two diverging species have been identified as outlier loci in genomic islands associated with divergent alleles differentially fixed between populations [59] , [60] . In total , we identified 8 , 185 informative single nucleotide polymorphisms ( SNPs ) ( excluding indels ) within the Blue and Red groups of strains . Of these , 4 , 232 ( 52% ) were fixed differences between the two groups . This high number of fixed differences between the two groups of strains relative to recently diverged sexual species [61] , [62] may result from the relatively low rates of non-reciprocal gene flow occurring in Sulfolobus . We calculated FST values based on individual non-indel SNPs for each gene and in sliding windows of 10 kb across the genome ( Figure 4 ) . FST measures the level of differentiation between two groups [63] , [64] . Low values indicate more diversity within than between groups that can result from either constrained divergence between groups by purifying selection or the exchange of alleles through recombination . High values indicate differentiation with more variation between groups than within . This occurs in regions that are under diversifying selection or where there are low levels of recombination [63] . This analysis revealed that the majority of the core chromosome exhibits high FST values and appears to be differentiated . Of the 1 , 883 genes shared between the Blue and Red groups in which variation was detected , only 466 genes ( 25% ) exhibit FST values lower than 0 . 5 . As shown in Figure 4 , although the majority of the genome is highly differentiated ( FST>0 . 5 ) lower levels of differentiation ( FST<0 . 5 ) between the two species occur exclusively in three large regions of the genome ( ranging in size from approximately 265 Kb to 770 Kb ) , separated by differentiated genomic “continents” ( ranging in size from 290 Kb to 370 Kb ) [65] where no lower values are observed . Within these larger regions , smaller differentiated islands , between 5 Kb and 230 Kb , in length exist separated by contiguous regions of low differentiation between 10 Kb and 30 Kb long . The broad topology of differentiation across the chromosome is consistent with either differential selection among many loci within the chromosome ( located within the three regions of high differentiation ) , different levels of gene flow in different regions of the chromosome ( higher gene flow in regions of lower differentiation ) , or both [64] , [65] . Many loci under weak selection would explain our failure to identify loci under diversifying selection using the patterns of non-synonymous to synonymous substitutions or the MacDonald-Kreitman test . This pattern is also consistent with the prediction that , due to the mechanisms of gene flow in microorganisms , different regions of the chromosome will exhibit different patterns of speciation [29] . Chromosomal regions that are less susceptible to recombination between species will differentiate first even if they are not associated with ecological differentiation . Interestingly , most non-core genes occur in regions of low differentiation between the two groups ( Figure 4 ) . This is inconsistent with novel genes driving differentiation between species or with the import of novel gene islands decreasing recombination and promoting speciation [16] , [66] . This lack of differentiation in regions of the chromosome with variable gene content supports the earlier observation and experimental data that suggest long regions of high sequence homology is not required for recombination . While the basis for the existence of these genomic “continents” of differentiation is unknown , they highlight the importance of using whole genomes when investigating speciation . To determine whether differences in gene content between the Red and Blue groups could be responsible for their ecological differentiation , we identified four contiguous regions of more than 5 Kb in which more than half of the genes are shared exclusively by all members of a group ( Table S4 ) . Two of these contain genes of unknown function ( 1 and 3 , Table S4 ) , with one having signatures of being an integrated plasmid fragment ( 1 on Table S4 ) . Two additional islands , one in each species , with functional annotations , were identified . The first island ( 2 in Table S4 ) , present only in the Red strains , contains six subunits ( TmoA αβ , B , C , D , and E ) and three accessory proteins of a putative Toluene-4-monoxygenase system . This region is present in many previously sequenced strains of S . islandicus [30] and was probably lost by the Blue group after its divergence from the Red group . The Blue strains share an island of four subunits ( α , β , δ , and γ ) of a putative respiratory nitrate reductase system ( 4 in Table S4 ) . Active nitrate reductases have been observed in other Archaea [67] but not in Sulfolobales , while a putative monooxygenase operon , similar to that in the Red group , was reported to be active in S . solfataricus [68] . In both of these regions , dN/dS values and divergence were similar to the estimates for the rest of the genomes . Both of these islands occur in the large , second region of lower differentiation and high variation in gene content identified in Figure 4 . Finally , we observed a difference in growth characteristics that might result from ecological differentiation . Triplicate cultures were grown that showed that the two groups differ in heterotrophic growth characteristics in rich media developed for Sulfolobus in the laboratory [31] . The Red strains have a shorter lagging time , higher growth rate , and higher culture density than the Blue strains ( Figure 5 ) . The growth difference between the Red and Blue groups was statistically significant ( t test , p<0 . 001 ) . Although growth differences could result from many aspects of Sulfolobus physiology , these data provide a phenotypic basis for the definition of two species that is commonly suggested as necessary in microbial taxonomy [69] , [70] . The sequencing of multiple genomes from closely related strains of S . islandicus coexisting in the same hot spring has allowed us to identify evidence for sympatric speciation in this natural archaeal population . Without typical barriers to recombination associated with genetic distance , the mechanism of speciation is likely to be ecological differentiation among many loci throughout highly differentiated regions of the chromosome or by differential islands of gene content . Two incipient species are persistent in the Mutnovsky Volcano region , as MLSA of 42 strains collected from six springs in 2010 shows a very similar structure to that observed in 2000 ( Figure 1A ) . This supports ecological differentiation that prevents competitive exclusion resulting in extinction of one type on this time scale [71] , [72] . Speciation driven by ecological divergence rather than physical barriers to gene flow has been increasingly observed in sexual eukaryotes [11] , [73]–[76] . Furthermore , the genomic pattern of differentiation under these circumstances has been theoretically predicted [77] and empirically demonstrated to exist in “continents” of differentiation using genomic analyses [65] . These initial reports identifying the genomic pattern of differentiation among species in model organisms of the domains Archaea and Eukarya point towards a possible unified genomic process of speciation across these two domains .
Ninety-seven S . islandicus strains were isolated from eight hot springs located in the Mutnovsky Volcano region in the Kamchatka Peninsula ( Russia ) , with specifications listed in Table S1 . Cultures were colony isolated on four different media: DT ( dextrin and tryptone ) spread plate as described in [31] , [78] , DTO plate containing DT plate plus an overlay of 0 . 002% Gelrite ( Sigma ) , DTS spread plate containing DT plate plus an overlay of 0 . 002% colloidal sulfur , and a DTSO plate containing a DTS plate as described above plus an overlay of 0 . 002% Gelrite ( Sigma ) as shown in Table S1 . Each isolate was subjected to three additional rounds of colony purification , and then grown in liquid phase followed by DNA extractions as detailed elsewhere [30] , [31] . The set of seven MLSA loci including the primer sequence , PCR conditions , reactant concentrations , and sequencing conditions were selected as a subset of the 12 loci described in detail elsewhere [32] . Sequences of unique alleles are deposited in GenBank under the accession numbers HQ123504-HQ123512 , HQ123518-HQ123527 , HQ123532-HQ123534 , and HQ123541-HQ123543 , and JQ339286–JQ339304 . MLSA data were evaluated for all seven loci using the Clonal Frame V1 . 2 software [35] . Runs of 250 , 000 iterations , after 100 , 000 burn-in iterations , were found to have reached convergence based on between run comparisons . Arlequin 3 . 5 [34] was used for performing FST calculations to test for differentiation using MLSA data between springs or between two years of sampling . Significance of FST values was determined through comparison to a permutation test , with p<0 . 05 considered significant . Ten strains from a single hot spring designated M . 16 , which is approximately 25 cm in diameter , were selected for de novo sequencing . Genomic preparations were done using a scaled up version of the protocol used for previous DNA extractions [30] . Briefly , 1 , 000 ml of S . islandicus cultures were concentrated and the pellet treated with 10 ml of 1× GES and 7 . 5 ml of 7 . 5 M ammonium acetate . After complete cell lysis , 15 ml of phenol∶chlorophorm∶iso-amyl alcohol ( 25∶24∶1 ) was added to the mix , homogenized , and subsequently centrifuged . Aqueous layer containing DNA and RNA mix was recovered for subsequent isopropyl alcohol and ethyl alcohol DNA purification . Extracted DNA was treated with 1 unit of RNAse I ( New England Laboratories ) for 30 min at 37°C . DNA quality and concentration was evaluated by gel electrophoresis and spectrophotometry . Genome sequencing was done using a 454 Life Sciences GS-FLX sequencer ( Roche ) at the University of Illinois Core Sequencing Facility ( www . biotec . illinois . edu ) . All 10 strains were initially “shotgun sequenced” reaching 8–20× coverage . Six strains ( M . 16 . 02 , M . 16 . 22 , M . 16 . 23 , M . 16 . 40 , M . 16 . 43 , and M . 16 . 47 ) were further selected for a second round using “454 paired-end sequencing , ” increasing the genome coverage to 28–51× ( Table S5 ) . Genome assembly was done in three steps . First , reads were assembled using the GS De Novo Assembler V . 2 . 0 . 00 ( Roche ) . Second , the GS assemblies were evaluated using the Consed V . 19 software [79] , where contigs exceeding more than twice the expected average coverage were reassembled using the “miniassembly” tool with default settings . For the remaining contigs , a fragment of 700 bp was removed from both ends and the removed fragments reassembled individually with the miniassembly tool; if the miniassembly tool produced a single contig , then the fragment was joined back into the contig , but if more than one contig was formed from reassembly , then the fragments were left as new contigs to be resolved in the next assembly step . This second assembly step addresses two possible artifacts: ( a ) separating reads with sections of similar sequence , but belonging to different copies of repetitive or duplicated elements , forming short contigs with abnormally high coverage , and ( b ) resolving partially assembled reads with masked regions ( not contributing to contig consensus ) at both ends of a contig . The third assembly step used comparative genomics to organize and scaffold contigs of the draft genomes using the two completed M . 16 genomes [30] followed by PCR sequencing of small gaps . MUMmer 3 . 0 [80] , the “move contigs” tool from the Mauve V . 2 . 3 . 1 [81] and ABACAS [82] software , was used to generate the draft genomes of the 10 newly sequenced strains described in Table 1 . Comparative genomics and paired-end data predicted very few gaps in the draft genomes , none of which is likely longer than 4 or 7 Kb . These gaps were joined by short strings of “N” to artificially close draft genomes to facilitate gene prediction and genome analysis . Four of seven genomes from the Blue group and all three genomes from the Red group were closed to exclude possibility of missing genes in gaps of draft sequences that could contribute to ecological differentiation between them . The draft versions of all genomes were deposited as a Whole Genome Shotgun project at DDBJ/EMBL/GenBank under the accession AHJK00000000 , AHJL00000000 , AHJM00000000 , AHJN00000000 , AHJO00000000 , AHJP00000000 , AHJQ00000000 , AHJR00000000 , AHJS00000000 , AHJT00000000 . The version described in this paper is the first version , AHJK01000000 , AHJL01000000 , AHJM01000000 , AHJN01000000 , AHJO01000000 , AHJP01000000 , AHJQ01000000 , AHJR01000000 , AHJS01000000 , AHJT01000000 and are also available at http://www . life . illinois . edu/Sulfolobus_islandicus . ORFs in the newly sequenced genomes were predicted and automatically annotated using the Rapid Annotation with Subsystems Technology ( RAST ) V2 . 0 software and FIGfams set of protein families [83] . Identification of the core genomic regions was done using the ProgressiveMauve algorithm [81] on the set of 12 M . 16 genomes . The alignment contains regions shared among all genomes , shared by a subset , or unique to one genome . The blocks shared by all genomes were split where gaps of more than 20 alignment positions were found , resulting in 155 core region fragments . The cumulative length of the core region for each genome was recorded and the fraction that they represent of each strain genome is detailed in Table 1 . The core region was then analyzed using ClonalFrame with default parameters [35] . Runs of 10 , 000 iterations were found to have reached convergence based on between-runs comparisons . The clonal genealogy reconstructed by ClonalFrame is shown in Figure 1B . ClonalOrigin is a Bayesian method to perform approximate inference under the coalescent model with gene conversion [84] using whole microbial genomes [36] . Assuming the correctness of the clonal genealogy reconstructed by ClonalFrame ( Figure 1B ) , ClonalOrigin reconstructs recombination events that represent a deviation from such vertical inheritance . Each recombination event is characterized by the genomic segment it affects , as well as an origin and destination on the clonal genealogy . By summarizing these last two properties across all events , we reconstructed the overall pattern of genetic flux between branches of the tree ( Figures 2 and S1 ) . We also studied the distribution of recombination segments along the genome and found no correlation with the level of polymorphism in 1 , 000 bp windows of the alignment ( Figure S2 ) . The absolute recombination rate was estimated by ClonalOrigin to be 2 . 41×10−4 per site per coalescent unit of time . A coalescent unit of time is equal to the average length of a generation multiplied by the effective population size . ClonalOrigin also estimated the ratio of frequencies of recombination and mutation ( ρ/θ ) and the average tract length of recombination events ( δ ) . The relative effect of recombination and mutation ( r/m ) was estimated using the formula of Jolley et al . 2005 [37]: r/m = ( ρ/θ ) *δ*π , where π is the average pairwise distance between two genomes . This formula is correct assuming that any two genomes are equally likely to recombine , but here we found ( Figure 2 ) that recombination happens more often between genomes from the same lineage so that r/m might be overestimated by this formula . We also calculated r/m taking the formula above with π equal to the average pairwise distance of two genomes in the same lineage . This would be correct if recombination happened only between members of the same lineage , but here we know that recombination also happens across lineage boundaries ( Figure 2 ) . Thus , this second calculation is likely to underestimate r/m , and taken together , the two calculations above provide us with a lower and upper bound on r/m . Recombination events were also evaluated by manual pairwise scanning methods using the Recombination Detection Program 3 software [85] for all core genomic regions . A genomic section was identified as evidence of a HR event after finding significant results ( p<0 . 01 ) with at least three of the following four tests as implemented in the Recombination Detection Program 3 software: RDP , GENECONV , MaxChi , and Bootscan . Overall , these results show similar patterns to the ClonalOrigin analysis . Boundaries of variable segments from the Mauve alignment are defined where there are core regions greater than 5 kb . Variable gene segments that are smaller than 5 kb in length or contain less than half variable gene segments were excluded from analysis because the majority result from insertion elements . Each variable segment was investigated for its distribution among the 12 genomes . This allowed us to identify composite segments with different distributions among strains . There was one segment per genome that was excluded from analysis due to a complicated pattern of shared and unique content that made segment boundaries very difficult to assign . In M . 16 . 27 , this segment is 74 kb long and located at 997 , 394–1 , 071 , 175 in the genome . To identify integrated mobile elements , variable gene segments were compared to a database of Sulfolobus mobile elements [40] including elements that are integrated into S . islandicus genomes [30] , [86] using BLASTN ( e<0 . 001 , −F f ) and to the NCBI nr database using BLASTX ( e<1E-5 , −F f ) . BLASTN was used to compare variable segments to the rest of the S . islandicus genomes from the M . 16 hot spring in order to assess Mauve's assignment of variation . The genome in which each variable segment was longest was compared to each other genome in which the segment is present , and the level of nucleotide identity was calculated between each of them as the number of matching nucleotides divided by the length of the match as reported by BLASTN and averaged over all of the pairs . The coverage of the longest segment was also calculated pairwise as with percent identity , with the total length of matching nucleotides divided by the length of the longest segment and averaged over all pairs . Genes present in the variable segments that separate Blue from Red were compared to NCBI's nr database with BLASTP ( e<0 . 001 , −F f ) if they were complete in every genome from either Blue or Red and if they were not core genes . A table containing the position and nucleotide of every SNP in the core genome alignment was exported from the genome alignment using the “Export SNPs” tool from the Mauve software . SNPs found within and between the Blue and Red species were used for sliding window FST evaluations . FST values were calculated for sliding windows of 10 , 000 bps moving in 5 , 000 bp steps across the genome of M . 16 . 27 . Arlequin 3 . 5 [34] was used for calculating FST to test for differentiation in the M . 16 populations . Low regions were defined as beginning and ending where windows of FST values were less than 0 . 5 . FST values were not calculated for empty windows or where sequence found in M . 16 . 17 is not present in all strains . FST values for genes were calculated using the same methods but applying the coordinates of each gene to SNPs exported from the core alignment from Mauve rather than sliding windows . Pairwise dN/dS ratios were calculated using the ORF clusters identified by MCL analysis [87] . All clusters containing a single copy of an ORF per genome ( 2 , 187 ) were evaluated for all pairwise dN , dS , and dN/dS ratios using the SNAP program ( http://hiv-web . lanl . gov/ ) [88] with the Nei and Gojobori method as described elsewhere [89] . Sequence alignments for clusters that resulted in dN/dS values greater than 1 . 0 were manually checked to resolve homopolymer indels from the 454 sequence data . The 12 M . 16 strains were evaluated for their growth characteristics in rich liquid media . As initially described by Whitaker et al . ( 2003 ) [31] , culturing conditions were at pH 3 . 5 , 75–78°C and with media containing a basal mineral solution , 0 . 1% Dextrin ( Fluka ) and 0 . 1% Tryptone ( Difco ) . All cultures were initiated using exponentially growing cultures to inoculate 50 ml liquid cultures in 250 ml flasks at an estimated starting concentration of 2×103 cells/ml . Cultures were incubated under static conditions and growth was followed by OD600 reads up to the sixth day of incubation . | Microorganisms from the bacterial and archaeal domains of the tree of life comprise the greatest breadth of biodiversity on earth . Yet the essential evolutionary process of speciation ( through which biodiversity is generated ) is poorly understood in microbes . At issue is the fundamental question of whether gene flow among individuals of clonally reproducing microorganisms is rapid enough to provide coherence within—and prevent speciation between—coexisting lineages . We use complete sequencing of microbial genomes to observe speciation in action . We focus on Archaea called Sulfolobus islandicus gathered from a geothermal hot spring from the Mutnovsky Volcano in Kamchatka , Russia , whose physical isolation allows us to pinpoint evolutionary processes to one location . Contrary to the theoretical predictions for microbes , we provide evidence that two novel lineages are in the process of becoming ecologically distinct and evolutionarily independent despite the fact that they recombine . The divergence we observe is not happening uniformly across the genome because certain genomic regions are more prone to become differentiated between species than others . This genomic view of the process of speciation occurring within a single natural microbial population contributes to our understanding of the generation of biodiversity in Archaea and furthers our understanding of speciation across the tree of life . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Material",
"and",
"Methods"
]
| [
"biology",
"genomics",
"evolutionary",
"biology",
"microbiology",
"population",
"biology",
"genetics",
"and",
"genomics"
]
| 2012 | Patterns of Gene Flow Define Species of Thermophilic Archaea |
Arthropod-borne pathogens are transmitted into a unique intradermal microenvironment that includes the saliva of their vectors . Immunomodulatory factors in the saliva can enhance infectivity; however , in some cases the immune response that develops to saliva from prior uninfected bites can inhibit infectivity . Most rodent reservoirs of Yersinia pestis experience fleabites regularly , but the effect this has on the dynamics of flea-borne transmission of plague has never been investigated . We examined the innate and acquired immune response of mice to bites of Xenopsylla cheopis and its effects on Y . pestis transmission and disease progression in both naïve mice and mice chronically exposed to flea bites . The immune response of C57BL/6 mice to uninfected flea bites was characterized by flow cytometry , histology , and antibody detection methods . In naïve mice , flea bites induced mild inflammation with limited recruitment of neutrophils and macrophages to the bite site . Infectivity and host response in naïve mice exposed to flea bites followed immediately by intradermal injection of Y . pestis did not differ from that of mice infected with Y . pestis without prior flea feeding . With prolonged exposure , an IgG1 antibody response primarily directed to the predominant component of flea saliva , a family of 36–45 kDa phosphatase-like proteins , occurred in both laboratory mice and wild rats naturally exposed to X . cheopis , but a hypersensitivity response never developed . The incidence and progression of terminal plague following challenge by infective blocked fleas were equivalent in naïve mice and mice sensitized to flea saliva by repeated exposure to flea bites over a 10-week period . Unlike what is observed with many other blood-feeding arthropods , the murine immune response to X . cheopis saliva is mild and continued exposure to flea bites leads more to tolerance than to hypersensitivity . The immune response to flea saliva had no detectable effect on Y . pestis transmission or plague pathogenesis in mice .
Yersinia pestis , the etiologic agent of plague , is adapted to flea-borne transmission and is a highly invasive , virulent pathogen . Infected fleas typically transmit small numbers of Y . pestis into the dermis while attempting to feed on a mammalian host . The bacteria are able to rapidly disseminate from the flea bite site to the draining lymph node to cause bubonic plague . After extensive multiplication in the lymph node , the bacteria spread systemically . The high bacteremia level required to infect fleas is typically fatal to the vertebrate host [1] , [2] . Adaptation to bloodfeeding on vertebrate hosts has independently evolved many times in the arthropods [3] , [4] , and in each case the arthropod had to overcome the hemostatic and other defense efforts of its host . This is accomplished primarily by a multitude of pharmacologically active molecules present in the saliva that are injected into the bite site . Arthropod saliva contains a diversity of anti-hemostatic , anti-inflammatory , and immunomodulatory effectors [5]–[8] . Vector-borne pathogens are introduced into a unique microenvironment in the skin that includes this salivary cocktail . It is now well-established that the natural transmission route can influence infection dynamics and differs from needle-injection models . For example , injection of Leishmania spp . with sand fly salivary gland extract into naïve mice leads to increased infectivity , higher parasite burdens and increased pathology compared to needle inoculation of parasites alone [9]–[11] . Vector feeding or vector salivary gland extract is known to enhance infectivity of other arthropod-borne diseases , including bacteria [12] , viruses [13]–[15] , and parasites [16] . Furthermore , exposure to vector saliva in uninfected bites results in an immune response to salivary components , and this can affect transmission and pathogenesis when the animal is later fed upon by an infected vector . In some cases , a history of exposure to uninfected bites can lead to protection . For example , mice previously exposed to uninfected sand fly bites are more resistant to cutaneous leishmaniasis [17] . Immunity to salivary components from past exposures was also shown to inhibit transmission of Borrelia burgdorferi by Ixodes scapularis [18] and transmission of Plasmodium yoelii by Anopheles stephensi [19] . With mosquito transmission of West Nile virus ( WNV ) and tsetse transmission of Trypanosoma brucei , prior exposure to vector saliva exacerbated disease [20] , [21] . However , in other studies of mosquito transmission of WNV or Plasmodium , and tick transmission of Lyme disease spirochetes , prior vector exposure had no obvious effect [22]–[24] . The salivary proteins of the rat flea Xenopsylla cheopis and the cat flea Ctenocephalides felis have been characterized [25] , [26] . In addition to known anti-hemostatic and anti-inflammatory effectors such as apyrase and adenosine deaminase enzymes , esterase , Antigen-5 family proteins , and antimicrobial peptides commonly found in the saliva of other blood-feeding arthropods , flea saliva contains some unique proteins . Most prominently , a large family of closely related acid phosphatases , probably enzymatically inactive , comprises the major protein component of flea saliva; the FS family and several other peptides are also unique to flea saliva [25] , [26] . The function of these flea-specific protein families is unknown . Maintenance of Y . pestis depends primarily on flea-rodent transmission cycles . Most rodents harbor a permanent ectoparasitic flea fauna that feed on them daily . However , the immune response to flea saliva and how it might affect plague transmission dynamics has not been characterized . To date , laboratory studies of flea-borne transmission of Y . pestis have utilized naïve rodents with no previous exposure to fleas , which is not the natural situation . In this study , we examined the effect of flea saliva on early events in pathogenesis in an intradermal injection model of bubonic plague and characterized the murine immune response to flea saliva . We also evaluated whether pre-exposure to uninfected flea bites and pre-existing immunity to components of flea saliva influence the transmission and disease progression of flea-borne plague .
The fully virulent wild-type Y . pestis strain 195/P [27] was used in all experiments . Bacteria were grown in brain-heart infusion broth overnight at 28°C , transferred into LB broth and grown for 24 h at 28°C without aeration . The culture was brought to 20% glycerol and stored in aliquots at −80°C . The titer of aliquots used for injections was tested periodically by limiting dilution on tryptose blood agar plates in triplicate and there was no change in colony forming units ( CFU ) /ml of the stock over the course of the study . Specific-pathogen-free , 6–12 week old female C57BL/6 mice ( Harlan Laboratories ) were used for all experiments . All experiments involving animals were approved by the Rocky Mountain Laboratories , National Institute of Allergy and Infectious Diseases , National Institutes of Health Animal Care and Use Committee and were conducted in accordance with all National Institutes of Health guidelines ( Animal Protocol Approval #2010-48 ) . Salivary glands were dissected from adult X . cheopis fleas in sterile cell culture grade PBS and transferred to tubes containing PBS , two pairs of salivary glands/µl , and stored at −80°C . Salivary glands were subjected to four freeze/thaw cycles and sonicated on ice ( 50% power , 5 s on/off pulses for 2 min using a Vibracell VCX130 , Sonics and Materials , Newtown , CT ) . The SGE was cleared by centrifugation ( 8000× g for 5 min ) and total protein quantitated by Qubit Quant-iT Protein Assay Kit ( Invitrogen , Carlsbad , CA ) . Halt Protease Inhibitor Cocktail ( Thermo Fisher Scientific , Atlanta , GA ) was added according to manufacturer's recommendation and SGE was stored at −80°C . Xenopsylla cheopis fleas were from laboratory colonies and maintained as previously described [28] . Capsules used to contain fleas while feeding on mice were constructed by cutting the needle end from 10 ml plastic syringes and covering the open end with nylon mesh . The plunger was removed to put fleas into the syringe and replaced to contain them . Twenty-five fleas starved for 4–7 days were placed in a feeding capsule . Mice were anesthetized by subcutaneous injection of ∼70 µg/3 µg ketamine/xylazine per gram body mass . The mesh side of a feeding capsule was secured with tape on the ear pinna of anesthetized mice and fleas were allowed to feed for 30–40 min . Alternatively , a small patch of fur was shaved from the side of the mouse's abdomen and the flea capsule secured with tape over the skin for feeding . Aliquots of Y . pestis were thawed and diluted in PBS to the desired concentration . There were four treatment groups: 1 ) mice injected intradermally ( id ) in the ear pinna with ∼250 CFU Y . pestis in a total volume of 10 µl ( Y . pestis-only group ) ; 2 ) mice that received just the flea feeding procedure described above ( fleas-only group ) ; 3 ) mice that received flea feeding on the ear followed immediately by id injection in the same ear with ∼250 CFU Y . pestis in a total volume of 10 µl ( fleas + Y . pestis group ) ; and 4 ) mice injected id in the ear with 10 µl PBS ( control group ) . Samples of 5–10 mice from each group were euthanized at 3 , 6 , 12 , and 24 h post-infection ( pi ) . Ears were collected into tubes with 70% EtOH [29] . Superficial parotid lymph nodes ( using the nomenclature of Van den Broeck et al [30] ) were collected into tubes with 2 ml PBS without Ca++ or Mg++ . In a separate experiment , fleas were allowed to feed both on the ear and side of naïve mice as described above . At 3 , 6 , 12 , 24 and 48 h after feeding , three mice were euthanized . The ear and a skin biopsy from the feeding site on the side of each mouse was collected and fixed in 10% neutral buffered formalin ( NBF ) for histological staining . The contralateral ear and skin biopsy were taken as controls . Ears were removed from EtOH and blotted dry . Ears were carefully peeled apart , separating the two skin layers , and floated dermal side down in a 6-well non-tissue culture treated plate . Wells contained 3 ml RPMI medium ( Sigma , Atlanta , GA ) with 25 mM HEPES pH 7 . 5 , 1 . 5% NaHCO3 , 50 µg/ml DNAse I ( Worthington Biochemical Corporation , Newark , NJ ) and 26 U/ml Liberase TM ( Roche Diagnostics , Chicago , IL ) . Preparation of single cell suspensions from ear and superficial parotid lymph node samples and determination of bacterial load numbers were done as previously described [31] . Aliquots of 50 µl of single cell suspensions from each sample were dispensed into 96-well round bottom microtiter plates and stained with 1∶200 dilutions of antibodies ( BD Pharmingen or eBioscience ) : anti-Ly-6G ( clone 1A8 , FITC labeled ) , anti-CD11b ( clone M1/70 , labeled with phycoerythrin-Cy7 ) , and anti-F4/80 ( clone BM8 , allophycocyanin labeled ) . Rat IgG2a and IgG2b were used as isotype controls . Cells were stained for 30 min at 4°C , spun at 650× g for 1 min and fixed with IC Fixation Buffer ( eBioscience , San Diego , CA ) for 1 h at 4°C . Cells were spun at 650× g for 1 min and resuspended in PBS with 1% fetal bovine serum . Cell phenotype data were acquired on a Partec CyFlow ML flow cytometer and analyzed with FloMax ( Partec ) and FloJo ( Tree Star ) software . Gating strategies were as previously described [31] . Neutrophils were defined as Ly-6G+F4/80− . Neutrophils that expressed high levels of CD11b [29] ( CD11bhigh ) were defined as activated neutrophils [31]–[35] . Macrophages were defined as F4/80+Ly6G− cells . Three groups of five mice each received three different flea exposure regimens for ten weeks . Group A mice were fed on by 25 fleas once per week; Group B were exposed to 50 fleas once per week; and Group C 25 fleas twice per week . All flea feeds were done on a shaved area on the side of the mouse . After each exposure , fleas were individually examined under a dissecting microscope , and the number of fleas that had fed ( containing fresh blood in the midgut ) was recorded . Mice were tracked individually to determine the total number of flea bites for each mouse . After five weeks of exposure , blood samples were taken to collect serum for detection of antibody to salivary proteins ( 5-week sera ) . At the end of the 10-week exposure period , mice were exposed a final time on each ear , 25 fleas per ear; 12 h later the mice were euthanized . For each mouse , one ear was removed and processed for flow cytometry analysis as described; the other ear and a skin biopsy from the flea feeding site on the side of the mouse was fixed in NBF for histological staining ( Group A mice were not sampled for histology ) . Blood samples were taken to collect serum for detection of antibody to salivary proteins ( final sera ) . Tissue samples fixed in NBF were embedded in paraffin , sectioned , and stained with hematoxylin and eosin . For each tissue sample , 4–12 sections were examined and subjectively categorized by a board-certified veterinary pathologist ( D . Gardner ) and assigned a numerical inflammation severity score from 0 to 2: 0 = within normal limits ( i . e . , not different from unbitten controls ) ; 1 = minimal inflammation: very few to low numbers of inflammatory cells in the dermis and/or subcutis; 2 = mild inflammation: low to moderate numbers of inflammatory cells within the dermis and/or subcutis; inflammatory cells are detectable at 4–10X magnification and may aggregate together . Sera collected from flea-exposed mice were screened for IgG response to flea salivary proteins by Western blot . SGE ( 5 µg/lane ) was separated by SDS-PAGE on 4–20% polyacrylamide gradient gels and transferred to 0 . 2 µm nitrocellulose using a Criterion blotting apparatus ( BioRad , Richmond , CA ) . Blots were blocked in 5% dried skim milk in Tris-buffered saline ( TBS ) overnight at 4°C . Blots were cut into strips and incubated with serum samples diluted 1∶250 in 2% dried skim milk in TBS with 0 . 05% Tween 20 for 2 h at room temperature with gentle agitation . Blots were washed in TBS-Tween then incubated with goat anti-mouse IgG ( Invitrogen ) at 1∶10 , 000 for 1 h at room temperature . Blots were washed again in TBS-Tween and developed using the BCIP/NBT liquid substrate ( Sigma Life Science , Atlanta , GA ) . A polyclonal antibody raised to SGE in mice ( prepared by Lampire Biological Laboratories , Pipersville , PA ) was used at 1∶10 , 000 as a positive control; naive mouse serum at 1∶250 served as a negative control . To quantitate the IgG response to SGE , an ELISA was developed . Costar 96-well flat bottom high-binding EIA plates ( Fisher Scientific , Pittsburg , PA ) were coated with SGE in 0 . 05 M carbonate/bicarbonate buffer , pH9 . 6 ( 100 µl/well at 0 . 5 ng/µl ) overnight at 4°C . Plates were blocked with 5% dried skim milk in PBS-0 . 05% Tween-20 for at least 2 h at 28°C , then incubated with unknown sample sera or naïve mouse serum at 1∶250 in 2% dried milk in PBS-Tween for 2 h at 28°C . After washing with PBS , goat anti-mouse IgG horseradish peroxidase conjugate ( Thermo Scientific ) was added at 1∶20 , 000 in 2% dried milk in PBS-Tween and incubated for 1 h at 28°C . Plates were washed with PBS-Tween and developed using the Ultra TMB-ELISA substrate ( Thermo Scientific ) . Color development was stopped with 2 M H2SO4 and absorbance of wells read at 450 nm on a Synergy 2 microplate reader ( Bio Tek Instruments , Winooski , VT ) . Sera were tested in triplicate . For each ELISA run a standard curve was built using serial 2-fold dilutions ( 1∶1250–1∶1280K ) of the polyclonal anti-SGE serum . The 1∶2500 dilution was arbitrarily assigned a value of 10 , 000 antibody units ( U ) . A standard curve of log10 ( U ) plotted against A450 was fitted to the 4-parameter logistic regression model in GraphPad Prism v . 5 . 01 ( GraphPad Software , Inc . , La Jolla , CA ) , with the hill slope constrained to 1 . 0 and bottom parameter constrained to the average negative control value . The log10 ( U ) of unknown sera was interpolated from the standard curve . Sera with values > mean of the negative controls +2 SD were considered positive . Similar ELISAs were developed to quantitate the IgG1 , IgG2a , IgG2c , IgM , and IgE responses . Anti-mouse secondary antibodies to these antibody isotypes were obtained from Thermo Scientific ( α-IgM , α-IgE , and α-IgG2c ) or Jackson ImmunoResearch ( α-IgG1 and α-IgG2a ) . Sera from 20 wild Rattus norvegicus from Los Angeles , CA were obtained during surveys conducted in 2003 . Rats were combed for ectoparasites and the only species found was X . cheopis . Sera were tested for antibodies against SGE by Western blot as described above . Two groups of 20 mice each received contrasting exposure regimens to uninfected fleas . One group was fed on by 25 fleas once per week for 5 weeks ( low exposure ) ; the second group was fed on by 25 fleas twice per week for ten weeks ( high exposure ) . After five weeks of exposure a serum sample was taken from both groups . X . cheopis fleas were infected with Y . pestis by using an artificial feeding device [36] and monitored for proventricular blockage as previously described [37] . Three blocked fleas were used to challenge individual mice for 1 h on a shaved area on the side of the abdomen . After the 1 h feeding period the fleas were examined microscopically and the number that attempted to feed ( fresh blood in the esophagus ) was recorded . Naïve ( not pre-exposed to fleas ) mouse controls were similarly challenged by blocked fleas . Mice were monitored for the appearance of illness ( lethargy , ruffled fur , hunched posture , reluctance to respond to external stimuli ) and euthanized; time to terminal disease in hours was recorded . Triturated spleen and blood samples recovered from each mouse after euthanasia were cultured on blood agar plates to confirm Y . pestis infection . Final sera were taken from all survivors at the end of the experiment . Bacterial loads in infected mice and IgG responses in mice from the challenge experiments were compared using Student's t test . The association between bacterial loads and neutrophil recruitment in the ear and draining lymph node , and the association between number of flea bites and IgG response , was tested by Pearson correlation analysis . For flow cytometric data , groups were compared by Kruskal-Wallis nonparametric ANOVA , followed by Dunn's multiple comparison test to detect differences between treatments or timepoints . Survival curves in the challenge experiment were compared by the Mantel-Cox logrank test . Analyses were done using GraphPad Prism software ( version 5 . 01 ) .
Insect bites often cause local cutaneous inflammatory reactions , ranging from mild erythema to papule formation and edema , largely determined by salivary components [38]–[41] . The only obvious dermal sign after 22–28 X . cheopis fleas fed on the ear and abdominal skin of naïve mice were occasional small discrete erythematous spots , with no swelling or papule formation . Figure 1 shows representative histological examples of inflammation observed in mouse skin within 48 hrs of flea feeding . Three mice showed minimal inflammation ( severity score = 1 ) in the dermis of the ear ( Fig . 1C ) or abdomen ( Fig . 1D ) compared to controls ( Fig . 1A and B ) . One mouse had a focus of moderate inflammation ( severity score = 2 ) in the ear ( Fig . 1E ) . In ten of 15 mice examined , skin from flea-fed areas was indistinguishable from unbitten control skin . Arthropod saliva can modulate the migration and defense responses of innate immune cells . Consequently , in naïve animals ( with no prior exposure to the vector ) , injection of a pathogen into the skin where an uninfected vector has recently fed , or coinjection of the pathogen with vector SGE , often results in enhanced disease progression compared to injection of the pathogen alone [7] , [9] , [12]–[16] . We compared neutrophil and macrophage recruitment following id injection of fully virulent , wild-type Y . pestis 195/P into the ear of two groups of naïve mice , one of which had received 11–22 flea bites on the ear immediately before injection ( fleas + Y . pestis group ) and one which had not ( Y . pestis-only group ) . Two other groups of mice received flea bites only or id injection of PBS . The mean ± standard deviation of inocula for all mice infected with Y . pestis was 281±62 CFU . Bacterial loads were measured in the ear ( Fig . 2A ) and draining lymph node ( Fig . 2B ) of mice at different times after id infection with Y . pestis , with and without the presence of flea feeding . The Y . pestis-only and fleas + Y . pestis groups did not differ significantly from each other in the number of CFU recovered at any timepoint . At 3 h pi , both groups of infected mice had significantly greater neutrophil recruitment ( total Ly-6G+ cells ) in the ear than control mice receiving only flea feeding ( fleas-only ) or an id injection of PBS ( Fig . 3A and B ) . At 6 h pi the number of neutrophils in mice infected with Y . pestis ( with or without prior flea feeding ) decreased , but increased in the fleas-only treatment group . At 12 h pi , the fleas-only group had significantly higher % total neutrophils ( P<0 . 05 ) compared to the PBS controls ( Fig . 3A ) . By 24 h pi , neutrophils returned to PBS control level in the fleas-only group , but showed a variable response in the Y . pestis infected groups , with some mice similar to controls and some mice showing an influx of activated ( Ly-6G+CD11bhigh ) neutrophils . Overall , the Y . pestis-only and fleas + Y . pestis groups had the same kinetics: both showed early neutrophil recruitment at 3 h , which decreased significantly through 12 h pi and began to increase at 24 h as disease progressed . These treatments did not differ significantly from each other at any timepoint . In contrast , the fleas-only treatment showed a significant increase in neutrophils peaking at 6–12 h and declining to control levels by 24 h . In the draining lymph node , there were some differences among treatments in total neutrophils , but these mostly fell within the range of values seen in completely unmanipulated controls ( Fig . 3C and D ) . The exception to this was at 24 h , where some individuals in the Y . pestis-only and fleas + Y . pestis groups showed an increased influx of neutrophils , while other individuals were still in the range of control mice . This variation likely reflects differences in the progression of bubonic plague among individual mice [29] . In the fleas-only group , neutrophils in the lymph node remained in the range of control mice throughout the experiment . The reduction in neutrophils before 24 h pi in mice infected with Y . pestis is represented again in the graph of % total neutrophils vs . bacterial load ( Fig . 2 ) . In the ear , at 3 and 6 h , there was detectable neutrophil recruitment which was reduced to PBS control level by 12 h . This occurred whether or not mice were exposed to flea bites before infection ( Fig . 2C and D ) . Only after bacterial loads exceeded 4 log10 in ear or lymph node did we again observe some mice with neutrophil responses greater than PBS controls . There was no correlation between log10 CFU and % Ly-6G+ cells in the ear in either treatment ( Y . pestis-only: r = −0 . 21 , P = 0 . 21; fleas + Y . pestis: r = −0 . 17 , P = 0 . 28 ) . In the draining lymph node , neutrophil numbers did not begin to exceed those in PBS controls until 24 h pi , even in mice with 4–5 log10 bacteria ( Fig . 2E and F ) , with or without flea feeding . In the Y . pestis-only group , there was no correlation between log10 CFU and % Ly-6G+ cells ( r = 0 . 08 , P = 0 . 70 ) . In the fleas + Y . pestis group there was a significant positive correlation between log10 CFU and % Ly-6G+ cells ( P = 0 . 007 ) , but with a low r2 ( 0 . 354 ) . The correlation was dependent on the three mice at 24 h pi with an influx of neutrophils well above those seen in PBS controls ( Fig . 3D ) ; removing these outliers resulted in a non-significant correlation ( P = 0 . 061 ) . In both groups infected with Y . pestis macrophage recruitment in the ear was similar to that of neutrophils: at 3 h pi these mice had significantly greater macrophage recruitment compared to fleas-only mice ( P = 0 . 002 ) , which decreased to PBS control level by 12 h pi ( Fig . 3E ) . The Y . pestis-only and fleas + Y . pestis groups did not differ significantly from each other at any timepoint . In the fleas-only group , the presence of macrophages was not different from controls at 3 h pi but rose significantly at 6 h pi ( P<0 . 01 ) and remained elevated for the rest of the experiment , significantly greater than the Y . pestis-only group at 24 h pi ( P<0 . 001 ) . In the lymph node , there were some statistically significant differences between groups , but macrophages made up a very small percentage of cells , generally less than 1% in all treatments ( Fig . 3F ) . Also , median percentages of F4/80+ cells of all treatments were within the range of values seen in unmanipulated mouse controls . A second goal of this study was to determine if an anamnestic or hypersensitivity response develops to flea saliva and if it affects flea-borne plague transmission dynamics . Three groups of mice were fed upon by uninfected X . cheopis flea bites throughout a 10-week period according to the regimen shown in Table 1 . Histological examination of Groups B and C revealed no evidence of a hypersensitivity reaction as described in other arthropod-host systems [42] , [43] . After 10 weeks , four out of five Group C mice showed minimal inflammation ( severity score = 1 ) in the ear ( Fig . 4A ) and abdominal skin ( Fig . 4B ) , or were indistinguishable from control tissue ( severity score = 0 ) . One mouse in this group showed mild inflammation ( severity score = 2 ) in the skin 12 h after flea feeding ( Fig . 4D ) . Group B mice had a similar pattern: two mice had a score of 2 in the ear ( Fig . 4C ) and two mice had a score of 2 in the skin . In addition , one out of five mice had a score of 1 in the ear or skin . Throughout the course of flea exposure , mice were observed at 1 and 2 days after feeding , and none showed any swelling or delayed type hypersensitivity response at the feeding site . Some had transient reddening of the skin which did not last from one flea feeding episode to the next . Overall , mice with a history of flea exposure showed little inflammation or hypersensitivity in response to flea feeding . Mice with the lowest exposure to flea bites ( Group A ) had inconsistent seroconversion , with three out of five mice showing little or no IgG response to X . cheopis SGE in immunoblots ( Fig . 5 ) . The higher exposure regimens ( Groups B and C ) generated more consistent antibody responses to SGE ( Fig . 5 , Table 1 ) . There was an increase both in numbers of positive mice and in the magnitude of the IgG response between sera collected after 5 weeks of exposure to fleas and final sera taken after ten weeks of exposure . There was a significant correlation ( P = 0 . 001 , r2 = 0 . 22 ) between the total number of flea bites received by an individual mouse and its IgG log10 ( U ) . Most of the IgG response was directed to the prominent protein band of 40–43 kDa , which corresponds to the phosphatase family proteins [25] , the major component of X . cheopis SGE . ( Fig . 5 ) . Reactivity to a ∼100 kDa SGE antigen was also seen in mice with the highest exposure to flea bites ( Group C ) . Results of IgG1 , IgG2a and IgG2c-specific ELISA showed antibody against SGE was highly skewed toward the IgG1 subtype with very little production of IgG2a or IgG2c ( Table 2 ) , indicating flea feeding stimulated a Th2-biased response in mice . Minimal IgM responses were found in a few of the final sera from Groups B and C ( Table 1 ) , and none of the sera tested had detectable IgE by ELISA . The IgG response to flea saliva was also surveyed in the sera of wild brown Norway rats ( R . norvegicus ) trapped in Los Angeles , where the sole ectoparasitic flea species is X . cheopis . As with the C57BL/6 mice , rat serum IgG reactivity to SGE proteins was variable , with the flea phosphatases being immunodominant ( Fig . 5 ) . A band at about 55 kDa seen in three mice also appeared frequently in the rat samples; the 100 kDa band was clearly present in only one of the rats . Rats were trapped during August and November , representing seasons of high and low flea index , respectively . There was no discernable correlation between immunoblot results and date trapped or rat age ( adult , subadult , or juvenile ) or weight ( 55–370 g ) . In naive mice after a first exposure to fleas , recruitment of neutrophils and macrophages peaked at 6–12 hours in ear tissue ( Fig . 3 ) . At the end of their 10-week flea exposure regimen ( Table 1 ) , Group A , B and C mice were exposed to flea bites a final time on the ear . Twelve hours later the presence of neutrophils and macrophages was assessed by flow cytometry and compared to the naïve mouse 12 h timepoint ( Fig . 3 ) . Total neutrophils ( Ly-6G+ cells ) from Group C mice , a high exposure treatment , was significantly less ( P<0 . 05 ) than the response seen in the low exposure Group A ( Fig . 6A ) . In addition , fewer activated neutrophils ( Ly-6G+CD11bhigh cells ) were present in Groups B and C than in Group A , with a statistically significant difference between Group A and B ( P<0 . 05 , Fig . 6B ) . Figure 6C shows % neutrophils activated , the percentage of Ly-6G+ cells also CD11bhigh . Group B was significantly less than both Group A ( P<0 . 01 ) and the naïve mice group ( P<0 . 05 ) . Finally , mice with higher exposure to fleas ( Groups B and C ) had a much lower macrophage response than Group A ( vs . Group C: P<0 . 05 ) or naïve mice ( vs . Group B: P<0 . 05; vs . Group C: P<0 . 001 ) ( Fig . 6D ) . Overall this indicates a reduction in neutrophil and macrophage recruitment 12 h after flea feeding in experimentally exposed mice compared to naive mice receiving flea bites for the first time . To determine if a history of exposure to flea bites and the resulting immune response to flea saliva affects transmission dynamics , progression , or severity of disease , we challenged naïve and sensitized mice with Y . pestis by the natural , flea-borne infection route . Two such experiments were done ( Table 3 ) , comparing naïve mice to mice exposed to 25 fleas once per week for 5 weeks ( low exposure group ) or twice per week for 10 weeks ( high exposure group ) . By immunoblot , mice from the low and high exposure groups had qualitatively very different serum IgG levels to X . cheopis SGE ( Fig . 7A , B ) . These sera were also quantitatively analyzed by ELISA ( Table 1 ) . The low and high groups differed significantly in log10 ( U ) values ( t-test , P = 0 . 004 ) , and represent two contrasting treatment levels for challenge with Y . pestis by flea bite . The low and high exposure mice were challenged in tandem with naïve control mice by allowing three fleas that had been infected and become blocked with Y . pestis 195/P to feed on them . Table 3 shows data from individual mice in the challenge experiments . Survival curves ( Fig . 7C , D ) of the low and high group did not differ from their naïve control group by logrank test ( low: P = 0 . 50 , high: P = 0 . 92 ) . Exposed mice also did not differ in time to terminal disease compared to control mice ( t-test; low: P = 0 . 96 , high: P = 0 . 32 ) .
The immune sensitization of humans and other mammals to mosquito , sandfly , and flea bites often follows a characteristic five-stage sequence that evolves with repeated exposure [42] , [44] . The initial bites experienced by a naïve animal usually do not produce any observable skin reaction ( stage I ) . After a week or so of continued exposure , a delayed-type hypersensitivity response develops , typified by pruritic papules or vesicles that appear ∼24 hours after the bite ( stage II ) . As exposure continues , an immediate-type hypersensitivity response is seen within 30 minutes of the bite , which subsides but is followed by a delayed-type reaction ( stage III ) . With prolonged frequent exposure to bites , the delayed-type response no longer develops , leaving only the immediate-type response ( stage IV ) . Finally , desensitization or tolerance to the saliva develops , with no further skin reactivity ( stage V ) . Histologically , an influx of mononuclear cells is seen at stage II; in later stages neutrophils , eosinophils and basophils are also prominent [45] , [46] . Serum IgG and IgE antibodies specific to vector salivary proteins can be demonstrated [47]–[51] . This reactivity syndrome is indicative of an allergic or hypersensitivity response to insect bites . It has been described in guinea pigs [52] and dogs [46] following exposure to the cat flea Ctenocephalides felis , and allergic dermatitis in dogs and cats due to flea-bite hypersensitivity is an important veterinary concern [53] . In this study we found that C57/BL6 mice exposed to X . cheopis flea bites did not follow the stereotypical pattern described above . During the ten-week period in which a total of 35 mice received an average of 27–56 flea bites per week , no evidence of an immediate- or delayed-type hypersensitivity response was observed . The only dermal sign , regardless of the duration of exposure , was a transient non-papular , non-edematous erythematous area that was sometimes present on the skin immediately after feeding , probably the result of minor blood leakage due to anticoagulant effects of flea saliva . Flow cytometry data showed that flea bites elicited only a mild inflammatory response , evidenced by a small increase in neutrophil and macrophage recruitment that tended to subside after prolonged exposure ( Fig . 3 , 5 , 6 ) . No obvious mononuclear cell , eosinophil , basophil , or mast cell response was ever detected by histopathology ( Fig . 1 , 4 ) . Identical results were seen in BALB/c mice exposed to ∼20 fleas per week for 10 weeks ( Table S1 , Fig . S1 ) . Our results with mice are consistent with a previous study of the response of Sprague-Dawley rats ( R . norvegicus ) to X . cheopis flea bites over a four-week period , which also reported no obvious skin reaction and only a slight increase in neutrophils and mononuclear cells noted by histopathology [54] . In addition to monitoring the local immune response at the dermal bite site , we also characterized the murine adaptive immune response to flea saliva after prolonged exposure to flea bites . Mice produced antibodies to salivary proteins after different intensities of exposure to X . cheopis , with a general trend of increasing IgG with increasing total number of flea bites ( Fig . 5; Tables 1 , 2 ) . The IgG subtype production ( mostly IgG1 with very little IgG2a or IgG2c ) indicates a strong Th2 polarization in response to flea saliva . Most studies of exposure to salivary antigens from ticks [55] , sand flies [56] , tsetse [48] , and mosquitoes [57] also have shown a Th2 bias by isotype antibody production and cytokine profile . Although we were not able to simulate continuous infestation with fleas , as occurs in nature , serum from wild rats collected in Los Angeles , where X . cheopis is the only important flea ectoparasite , showed an antibody profile similar to our flea-exposed mouse sera . Thus , the detected murine response is not an artifact of exposure schedule , the use of laboratory-colonized X . cheopis or an inbred mouse strain . For both laboratory mice and wild rats , the immunodominant antigen was the family of 36–45 kDa acid phosphatase-like proteins , the predominant component of flea saliva . Antibodies to SGE proteins of ∼55 and ∼100 kDa were also detected . A similar immunoblot profile was observed in mice exposed to cat flea ( C . felis ) bites [58] . The function of the phosphatase-like family is unknown . There are ten identified transcripts , all with amino acid changes in their catalytic sites that presumably eliminate phosphatase activity [25] . However , they all have a basic pI >8 . 5 and still may be able to bind negatively charged substrates . One possibility is that they bind polyphosphate released by activated platelets [59] . If flea saliva is able to locally deplete polyphosphate this would inhibit platelet aggregation and blood coagulation [60] , [61] . An IgE response to salivary components is commonly associated with the allergic response described above to arthropod bites in lab animals as well as natural hosts [47]–[50] , [62] . In contrast , we did not detect serum IgE in any of the mice in our experiments , in keeping with the lack of any obvious allergic reaction at the flea bite sites . Similarly , a study of dogs found that animals allergic to flea bites produced high levels of flea antigen-specific IgE , but a group of dogs exposed to fleas constantly from a young age that showed no reaction to flea bites had IgE levels not significantly different from that of unexposed controls [63] . The authors concluded that chronic exposure to fleas resulted in tolerance in these animals to flea allergens . Overall the innate , adaptive and hypersensitivity immune responses of mice to X . cheopis saliva appear to be quite limited . As suggested by Vaughan et al . ( 1989 ) with rats [54] , mice may have a sort of adaptive tolerance to X . cheopis that is not seen in unnatural hosts such as guinea pigs [45] . In addition , because they live in close association with their hosts and require frequent blood meals , X . cheopis may have coevolved to not induce resistance in their usual hosts . We observed no reduction in feeding success of fleas used in any of the mouse sensitization trials , even in groups feeding on mice after twenty previous exposures that had strong antibody responses to SGE . This is consistent with the Vaughan et al . ( 1989 ) rat study in which no differences in the number of fleas that fed , blood meal size , or flea longevity was observed between fleas that fed on naïve rats compared to fleas fed on rats that had been sensitized to X . cheopis [54] . Thus , the immune response of mice and rats to X . cheopis appears to be one of tolerance rather than resistance . This can be contrasted to the strong hypersensitivity responses that develop to the bites of sandflies and hard ticks , which act to deter blood-feeding [64]–[67] . Unlike X . cheopis , sandflies and most hard ticks do not live in close contact with or feed repeatedly on an individual host . The dermal microenvironment into which arthropod-borne pathogens are transmitted is acutely influenced by the pharmacological and immunomodulatory effects of vector saliva . Added to this is the anti-saliva immune response of hosts with previous exposure to uninfected bites . There is now substantial evidence that this unique context , bypassed by needle-injection models , can significantly influence the infectivity of arthropod-transmitted pathogens . For example , in naïve animals the infectivity of both the sandfly-borne parasite Leishmania and the tick-borne bacterium Borrelia is enhanced by the immunomodulatory properties of the vector's saliva , but in sensitized animals the delayed-type hypersensitivity reaction at the bite site and acquired immune response to saliva are detrimental to pathogenesis [17] , [18] . In the case of West Nile virus infection , uninfected Culex tarsalis mosquitoes feeding on the footpads of mice followed immediately by needle injection of virus resulted in higher viremia at 24 and 48 h pi compared to needle infection alone [24] . This enhancement of infection was the same in mice presensitized to Culex tarsalis saliva . Unlike other vector-pathogen systems , repeated prior exposure of mice to uninfected flea bites had no significant effect on transmission , mortality , or time to disease after challenge with Y . pestis-infected fleas compared to naïve controls ( Fig . 7 , Table 3 ) . Injection of Y . pestis , either associated with flea bites or not , stimulated innate cell recruitment at 3 h which subsided at later time points despite concurrent bacterial replication . This is consistent with several studies showing that Y . pestis inhibits the inflammatory response until late in the disease progression [31] , [68]–[70] . In addition , injection of Y . pestis in association with flea bites did not enhance or inhibit bacterial replication or dissemination , in keeping with a previous study which reported no difference in bubonic plague pathogenesis following id injection of BALB/c mice with or without the presence of X . cheopis SGE in the inoculum [71] . In summary , by all parameters tested here , previous exposure to flea bites had no effect on Y . pestis infection in mice , and the inflammation observed in naïve mice exposed to fleas was inhibited in the presence of Y . pestis . The generally non-stimulatory nature of flea bites , the host tolerance to them , and the anti-inflammatory faculties of Y . pestis likely explain why exposure history to flea saliva did not affect plague transmission and pathogenesis . | The saliva of blood-feeding arthropods contains a variety of components that prevent blood clotting and interfere with the immune system of the vertebrate host . These properties have been shown to enhance or inhibit the transmission of different pathogens transmitted by arthropods . Yersinia pestis , the bacterial agent of plague , is maintained in nature by flea to rodent transmission cycles . Most rodents live in close association with fleas and are constantly being bitten by them , but the influence this has on plague transmission is unknown - previous studies used laboratory animals which have never experienced a flea bite . We found that flea bites caused a mild inflammatory response in mice , and eventually an antibody response to components of flea saliva , but did not significantly affect pathogenesis . The transmission of Y . pestis by infected fleas and the incidence rate of bubonic plague mortality were the same in mice that had been exposed to frequent uninfected flea bites and mice with no prior exposure to fleas . Therefore , in contrast to what has been shown for many other arthropod-borne disease systems , vector saliva did not enhance or inhibit Y . pestis infection in mice , regardless of the immune status of the host to flea saliva . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"blood",
"cells",
"immune",
"cells",
"immunology",
"microbiology",
"parasitology",
"disease",
"susceptibility",
"neutrophils",
"bacterial",
"pathogens",
"antibody",
"response",
"white",
"blood",
"cells",
"inflammation",
"animal",
"cells",
"medical",
"microbiology",
"microbial",
"pathogens",
"immune",
"response",
"cell",
"biology",
"clinical",
"immunology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"ectoparasites"
]
| 2014 | Evaluation of the Murine Immune Response to Xenopsylla cheopis Flea Saliva and Its Effect on Transmission of Yersinia pestis |
European liver fluke Opisthorchis felineus , causing opisthorchiasis disease , is widespread in Russia , Ukraine , Kazakhstan and sporadically detected in the EU countries . O . felineus infection leads to hepatobiliary pathological changes , cholangitis , fibrosis and , in severe cases , malignant transformation of bile ducts . Due to absence of specific symptoms , the infection is frequently neglected for a long period . The association of opisthorchiasis with almost incurable bile duct cancer and rising international migration of people that increases the risk of the parasitic etiology of liver fibrosis in non-endemic regions determine high demand for development of approaches to opisthorchiasis detection . In vivo magnetic resonance imaging and spectroscopy ( MRI and MRS ) were applied for differential assessment of hepatic abnormalities induced by O . felineus in an experimental animal model . Correlations of the MR-findings with the histological data as well as the data of the biochemical analysis of liver tissue were found . MRI provides valuable information about the severity of liver impairments induced by opisthorchiasis . An MR image of O . felineus infected liver has a characteristic pattern that differs from that of closely related liver fluke infections . 1H and 31P MRS in combination with biochemical analysis data showed that O . felineus infection disturbed hepatic metabolism of the host , which was accompanied by cholesterol accumulation in the liver . A non-invasive approach based on the magnetic resonance technique is very advantageous and may be successfully used not only for diagnosing and evaluating liver damage induced by O . felineus , but also for investigating metabolic changes arising in the infected organ . Since damages induced by the liver fluke take place in different liver lobes , MRI has the potential to overcome liver biopsy sampling variability that limits predictive validity of biopsy analysis for staging liver fluke-induced fibrosis .
Two liver fluke species of the genus Opisthorchis—O . felineus and O . viverrini—are known as human pathogenic agents . This fluke parasite inhabits the bile duct of the host liver , causing local mechanical damage and chronic inflammation , and having general toxic effects on the whole body [1] . Bile duct dilatation and mechanical obstruction accompanied by bile sludge formation are common complications of chronic opisthorchiasis [2] . Generally , liver fluke infection causes hepatobiliary pathological changes and leads to cholangitis , cholecystitis , and cholelithiasis . The most severe complication of opisthorchiasis is malignant transformation . There is a significant association between cholangiocarcinoma ( CAA ) and O . viverrini infection; that liver fluke is considered carcinogenic ( group I ) to humans [3–5] . O . felineus is classified by IARC as group 3 due to limited experimental data [6] . However , in sources published in the Russian language , the association between CAA and chronic long-term O . felineus infection has been emphasized [7 , 8] . In whole , periductal fibrosis is considered an important risk factor for bile duct cancer and is associated with CCA development [9 , 10] . In the meantime , there is no significant association between liver fluke infection and cirrhosis , at least for closely related species O . viverrini and Clonorchis sininsis [5] . Chronic liver diseases , mainly liver fibrosis , represent a public health problem worldwide [11] . The generally discussed disease entities associated with liver fibrosis are nonalcoholic fatty liver disease ( NAFLD ) , alcoholic liver disease , chronic viral hepatitis and chronic cholestasis , primary biliary cholangitis , and primary sclerosis cholangitis [12] . Fibrosis induced by liver flukes is commonly out of focus , though chronic infection caused by helminths of the genus Opisthorchis leads to periductal fibrosis and severe abnormalities in the hepatobiliary system [3] . It is clear that the pathogenesis of liver fibrosis depends on the underling etiology [13] . The main reason for complicated recognition of the cause of fibrotic or inflammatory process consists in the fact that liver responds to various injuries in a limited number of ways [14] . Parasitic liver flukes of the genus Opisthorchis are widespread in Eurasia—in Russia ( mainly West Siberia ) , Ukraine , Kazakhstan ( O . felineus ) , and South Asia ( O . viverrini ) . In the endemic regions more than 750 million people are at high risk of liver fluke infection [5] . There are some reports about O . felineus infected people in EU countries [15 , 16] . It is very important to note that most infected persons , especially those with chronic infection ( liver fluke lifespan may exceed 25 years [4] ) , show no symptoms at all or non-specific symptoms which are very difficult to recognize , so the infection is frequently neglected [17 , 18] . Taking into consideration increasing international migration of people , it is essential to bear in mind the risk of the parasitic etiology of liver fibrosis in the non-endemic regions [19] . In time , non-detected opisthorchiasis infection may be a cause of false diagnosis and complications in the treatment strategy of choice [17] . Furthermore , it is very important to exclude liver fluke infection , when assessing the transplantation potential of living liver donors . Search for non-invasive tests for liver fibrosis diagnosis is the main trend in hepatology , since the “gold-standard” method—liver biopsy—has some shortcomings , such as invasiveness , complications , and sampling variability [11 , 13 , 20] . MR imaging-based techniques for liver fibrosis assessment are very promising and are being actively developed currently [21] . Magnetic resonance imaging ( MRI ) and magnetic resonance spectroscopy ( MRS ) share common physical background and are based on a nuclear magnetic resonance ( NMR ) phenomenon [22] . In the presence of a static magnetic field , for a nucleus of spin 1/2 , absorption of energy emitted by a radiofrequency coil induces "flips" of a magnetic moment to another orientation . This "spin flip" places some of the spins in their high energy state . Relaxation processes which return nuclei back to the lower energy state after switching off the radiofrequency signal come with generation of a measurable amount of the radio frequency signal . The produced NMR signals received by the radiofrequency coil allow to generate MRS and MRI after Fourier transformation . MRI has become an increasingly important imaging technique for investigating patients with hepatic and biliary disorders [23 , 24] . Although MRI examination is more expensive in comparison with computed tomography and ultrasound , this modality is widely used due to higher spatial resolution [25] and absence of ionizing radiation . MRS methods offer an opportunity to assess relative tissue metabolite concentrations based on the chemical shift phenomenon [26] and are very useful in clinical and biomedical studies for examining metabolic changes in vivo non-invasively [25 , 27] . In particular , 1H MRS is successfully used to determine relative lipid concentration in hepatic tissue [28] , whereas 31P MRS allows to detect phosphorylated metabolites , including high-energy phosphates [25] . MRS methods are implemented to investigate the hepatic metabolic state and are applied as non-invasive diagnostic techniques for studying hepatobiliary pathology [29–31] . Despite the fact that MRI and MRS have become increasingly important imaging techniques for investigation of patients with liver and biliary disease [23 , 24] , there are few papers describing application of in vivo MR-techniques for studying liver fluke infection [32–35] and no papers depicting the use of these methods for investigating O . felineus . In this study we used magnetic resonance techniques ( MRI/MRS ) for differential assessment of liver abnormalities induced by O . felineus in an experimental animal model . We also found out correlations between the MR-findings and the histological data as well as the data of the biochemical analysis of liver tissue .
The experimental protocol was approved by the Bioethics Review Committee of the Institute of Cytology and Genetics , Siberian Branch , Russian Academy of Sciences , Novosibirsk ( No . 30 from 20 . 11 . 2015 ) . All animal experiments were conducted according to the principles of the Guide for the care and use of laboratory animals [36] . Metacercariae of O . felineus were obtained from naturally infected fish ( Leuciscus leuciscus ) caught in the river in the endemic areas of Western Siberia ( the Tom , Tomsk ) , Russia . A permit was not required to collect these fish according Federal law of the Russian Federation No . 166 “About fisheries and conservation of water biological resources” from 20 . 12 . 2004 . For the experiment , 5-week-old male hamsters ( n = 8 ) ( Mesocricetus auratus ) were infected intragastrically with 50 metacercariae per hamster , according to the previously described protocol [37] . Age-matched intact male hamsters ( n = 8 ) were used as controls . The hamsters were housed two in a cage ( OptiRAT ) under conventional conditions and were permitted ad libitum access to food and water . The animals were handled in pathogen-free environment . At 8 weeks post-infection , the hamsters were scanned with MRI and MRS in vivo . Following the MR examinations , the subsets of the infected and intact animals were deeply anesthetized with carbon dioxide and euthanized by decapitation . Blood and liver samples were collected for examination from each hamster from the control ( n = 8 ) and infected ( n = 8 ) groups . Serum and blood analyses were performed using routine procedures ( Methods in S1 Text ) . All 1H/31P MR experiments were performed on a horizontal tomographic scanner with magnetic field intensity of 11 . 7 T ( Bruker , Biospec 117/16 USR , Germany ) . Prior to MR examinations , the animals were fasted overnight . The animals were anaesthetized with gas anesthesia ( Isofluran; Baxter Healthcare Corp . , Deerfield , IL ) using a Univentor 400 Anesthesia Unit ( Univentor , Zejtun , Malta ) . The animal body temperature was maintained with a water circuit installed into the table bed of the tomographic scanner , which maintained the temperature of 30°C on its surface . A pneumatic respiration sensor ( SA Instruments , Stony Brook , NY ) was placed under the lower body part , which allowed to control the anesthesia depth . Liver samples ( from posterior segments of the right lobe ) were fixed in 10% buffered formalin , and embedded in paraffin . Tissue sections were cut into 4–5 μm-thick slices and stained with hematoxylin and eosin . Histological analysis was performed with the optical microscope Axiostar plus ( Carl Zeiss , Germany ) . For differential staining of collagen in the liver samples , Van Gieson’s method was used . Fibrosis was graded according to METAVIR score [39] . Hepatic steatosis was graded on a 0–3 scale through visual estimation of the percentage of hepatocytes containing intracellular vacuoles of fat [40] . Two samples of liver tissue were taken from posterior segments of the right lobe and medial segments of the left lobe for each animal . The samples were divided into five pieces in accordance with subsequent examinations and immediately placed in liquid nitrogen . Homogenization was performed using Tissue Grinders ( PELLET PESTLE® Cordless Motor , Kimble Chase , TN ) on ice . Statistical analyses were performed using IBM SPSS Statistics for Windows , Version 21 . 0 ( Armonk , NY: IBM Corp . ) . Data were presented as median ( range ) for data with non-normal distribution and mean ( SD ) for data with normal distribution . The data were tested by the Shapiro—Wilk test for normality . The differences among continuous variables with normal distribution were analyzed by the t-test and among continuous variables with non-normal distribution—by the Mann-Whitney test . For correlation analyses , the Person ( for continuous variables ) and Spearman ( for ordinal variables ) correlations were used . The P value below 0 . 05 was considered as significant .
Histological analysis data showed that O . felineus infection resulted in derangement of the liver architecture , whereas the trabecular pattern mostly remained intact . Hepatocytes had various sizes and patchy exhibited cloudy swelling . Cholangitis in the portal tract was accompanied by pronounced periductal and mild portal fibrosis as well as focal cystic dilatations of the intrahepatic bile ducts . Small bile ductular proliferation and irregular periportal ( mostly moderate ) infiltration of inflammatory cells , with predominant population of lymphocytes and histiocytes , occurred ( Fig 1 ) . Fibrosis was irregular; patchy , incomplete , thin fibrous septa ( portal to portal and portal to central bridge ) were observed ( Fig 1A , 1B and 1C ) . In whole , chronic cholangitis ( mild to moderate inflammatory activity , A1-A2 according to METAVIR score ) and mild chronic hepatitis were diagnosed , and the stage of fibrosis varied from 1 to 2 according to METAVIR score . There were no lipid droplets in the tested liver samples . No pathological changes took place in the liver in the reference group ( Fig 1D and 1E ) . The results of blood and serum analyses are given in Results in S1 Text and Table in S1 Table . In the infected group of hamsters the levels of alanine aminotransferase ( ALT ) and gamma-glutamyl transferase ( GGT ) increased markedly . Significant elevation of cholesterol , triglycerides , and low-density lipoproteins ( LDL ) in serum of the infected animals was detected . All the above listed serum parameters correlated with the fibrosis stage . The concentration of albumin in serum of the infected hamsters was lower than in the reference group and was accompanied by a statistically significant increase in the urea concentration . However , there was no correlation between these parameters ( r = -0 . 271 , p = 0 . 309 ) . The liver to body index increased in the infected animals , while the spleen to body index remained unchanged ( S1 Table ) . The results of liver tissue biochemical analysis are given in Table 1 . There was a statistically significant rise in the cholesterol level in opisthorchiasis , and the cholesterol to phospholipid ratio also elevated in the infected liver . Total lipid content , triglycerides and phospholipids as well as glycogen did not differ between the control and infected groups of hamsters . A significant decrease in the protein concentration occurred in the infected livers . However , the calculated total protein content in the entire liver did not differ between both control and experimental groups due to enlargement of the organ in the infected hamsters . It is important to note , that infection did not lead to lowering of the ATP concentration in liver tissue . A T2-weighted image of hamster liver infected with O . felineus for 8 weeks displayed pronounced bile duct abnormalities ( Fig 2A , 2B and 2C ) . Hyperintense areas extended around the bile duct are found in different lobes of the liver in all infected for 8 weeks hamsters . The most common MR findings included periductal enhancement , intrahepatic bile duct dilatation , and bile duct wall thickening . We detected a hyperintense zone extending from the dilated bile duct due to inflammation and fibrosis , without pronounced peripheral bile duct enhancement . In the control group the liver parenchyma was homogenous and showed no dilatation of the bile ducts ( Fig 2D , 2E and 2F ) . Analysis of the proton spectra ( Fig 3 ) showed no statistically significant differences in the content of lipid compounds in the livers of the control and infected hamster groups ( Table 2 ) . We found no correlation between the concentration of lipids in liver tissue determined according to biochemical analysis and certain characteristic peak areas in in vivo 1H MRS spectra . While the triglyceride and cholesterol concentrations , measured by biochemical analysis of liver tissue , correlated with 2 . 2 to 0 . 9 ppm peak ratio ( r = 0 . 561 , p = 0 . 041 for both ) , the triglyceride level correlated with 2 . 8 to 0 . 9 ppm peak ratio ( r = 0 . 537 , p = 0 . 032 ) . In fact , a peak at 0 . 9 ppm is attributed to protons in the methyl group of cholesterol , phospholipids and triglycerides , whereas peaks at 2 . 2 ppm and 2 . 8 ppm are assigned to protons bonded with carbons alpha and polyunsaturated carbons of triglycerides [45] [46] . More specifically 2 . 8 ppm/0 . 9 ppm peak area ratio is associated with a polyunsaturated bond [47 , 48] . Thus , according to the 1H MRS study , there were no significant changes in the lipid profile in the liver of the infected hamsters . Fig 4 presents typical 31P MR spectra of the hamster liver . The calculated fractions of various phosphorylated metabolites are given in Table 3 . Analysis of phosphorous spectra showed that the hepatic PME level increased in the infected group of hamsters , while the PDE levels remained unchanged . Significantly lower levels of NTP-α were detected in the infected group of hamsters , whereas the peak area attributed to other high-energy compounds , such as NTP-β and NTP-γ , in the livers of the control and infected hamsters had no statistically significant differences . PME and NTP-α peak areas correlated with the fibrosis stage ( r = 0 . 710 , p = 0 . 002 and 0 . 726 , p = 0 . 001 , respectively ) . Correlation analysis revealed that PME and NTP-α peak areas significantly correlated with serum ALT , AST , cholesterol , HDL , cholesterol to phospholipids ratio in liver tissue as well as with the fibrosis stage ( S2 Table ) . NTP-γ to NTP-α as well as NTP-γ to NTP-β ratios did not differ , whereas NTP-β to NTP-α ratio markedly increased in the infected group . Moreover , NTP-β/NTP-α ratio correlated with the fibrosis stage ( r = 0 . 622 , p = 0 . 010 ) , cholesterol content ( r = 0 . 622 , p = 0 . 005 ) and cholesterol to phospholipids ratio in the liver ( r = 0 . 664 , p = 0 . 005 ) . The calculated intracellular pH did not differ in the control and infected groups . 5’-AMP-activated protein kinase ( AMPK ) has been implicated in the control of hepatic glucose and lipid homeostasis , including fatty acid and sterol synthesis . It has been proposed to act as a ‘metabolic master switch’ mediating cellular adaptation to environmental or nutritional stress factors [49] . As illustrated in Fig 5 the amount of AMPK , phosphorylated by Thr172 residue of α-subunit ( phospho-AMPK ) , decreased in O . felineus infected liver . This means that the activity of kinase in the infected liver is lower than that in the healthy one . Statistical analysis showed that the relative levels of phospho-APMK in the hamster liver correlated with the NTP-α level , determined according to 31P MRS , ( r = 0 . 585 , p = 0 . 028 ) , cholesterol ( r = -0 . 726 , p = 0 . 003 ) and triglycerides ( r = -0 . 590 , p = 0 . 026 ) concentrations in serum as well as the fibrosis stage ( r = -0 . 607 , p = 0 . 021 ) .
O . felineus infection leads to significant pathological changes in the bile duct and liver of the hamsters , which results in chronic cholangitis and pronounced periductal/mild portal fibrosis . The key biochemical marker of hepatic parenchymal cell injury is the observed 3-fold increase in ALT in serum of the infected hamsters occurring due to the release of the enzymes from liver tissue into the circulation . The ALT level elevation in opisthorchiasis has been emphasized by several authors [50 , 51] . The observed hepatomegaly in the infected hamsters , a common sign of parasitic infection [52] , is caused rather by the inflammatory process in the liver [53 , 54] . The lowering of the protein concentration in liver tissue ( per gram ) of the infected animals may indicate an impaired protein synthesis function of hepatocytes . In the meantime , there is no difference in the total protein content calculated per whole liver between the infected and intact hamsters . Consequently , physiological adaptation to a metabolic demand as a driving mechanism of liver enlargement in the infected hamsters cannot be excluded [55] . O . felineus-induced liver lesions are clearly identified by MRI . The key MRI finding observed in our study is the pronounced intrahepatic biliary ductal dilatation with extension of T2-hyperintense damaged zone around the dilated bile duct . These spreading hyperintense zones correspond to fibrosis in adjacent to the bile duct hepatic tissue clearly identified by the histological analysis . The observed hyperintense signal from the bile duct wall is related to the inflammatory process diagnosed in the bile duct according to the histopathalogical study . It is important to emphasize the difference of the O . felineus MRI-pattern from that of closely related liver fluke infections . Choi et al . [33] described that the hallmark of radiologic findings in clonorchiasis is diffuse , mild dilatation of the intrahepatic biliary tree , especially the peripheral bile ducts , without any evidence of obstruction . In our study the dilatation was sufficiently pronounced . Yet , MRI in humans may slightly differ from MRI in animal models due to differences in size of the organ . At the same time , the T2-weighted image of O . felineus infected liver demonstrated wider and more pronounced hyperintense zones in liver tissue around the dilated bile duct , compared to the O . viverrini infected liver [34] . The origin of these differences is probably related to the fact that O . felineus causes more pronounced liver damage , as opposed to O . viverrini [56] . Therefore , the development of an approach to differential diagnosis of O . felineus infection from other helminth infections requires further investigation in human population . There is an additional reason for implementing of MRI for opisthorchiasis diagnosis . Since bile duct dilatation and T2-enchancement have taken place in different liver lobes of the infected individuals , liver biopsy sampling variability rises and limits predictive validity of biopsy analysis for staging liver fluke-induced fibrosis . In comparison with microscopic stool examination , MRI is a more expensive technique for diagnosing O . felineus infection , but it allows clinicians to evaluate opisthorchiasis-related liver damage [57] . Moreover , MRI can be very helpful for opisthorchiasis detection in infected people in non-endemic countries . Morphological and functional damage of the liver that resulted from mechanical , inflammatory , and toxic injury may lead to disturbance of the metabolic processes in the liver . Our 1H MRS data indicated no changes in the lipid profile of the liver and no fat droplets were observed in the liver of the infected hamsters according to histological examination . At the same time , biochemical analysis of hepatic tissue showed the elevated cholesterol concentration in liver tissue . Since MR techniques are sensitive only to molecules with a high degree of rotational molecular motion , the observed lipid resonances arise only from the intracellular fat droplets , whereas membrane lipids are not detectable using MRS [48] . Thus , the combination of MRS and biochemical data allows us to assume that cholesterol in the liver accumulates in the cellular membrane and the cholesterol to phospholipids ratio rises , respectively . This process may lead to loss of fluidity of liquid-crystalline domains in the cell membrane [58] . In turn , an increase in membrane rigidity may affect conformational freedom of integral membrane proteins and , consequently , their function . Therefore , the function of canalicular membrane transporters of cholesterol can be impaired even more , leading to enhancement of cholesterol excretion breakdown [59] . Observed in our study hyperlipidemia in combination with the increased cholesterol concentration and elevated cholesterol to phospholipid ratio in liver tissue indicate the disturbance of lipid metabolism in O . felineus infection . Alteration of lipid metabolism in parasite infection is emphasized by several authors [60–62] . For example , rise in the cholesterol/phospholipid ratio in liver plasma membrane and hyperlipidemia accompanied by strong accumulation of lipids in hepatic tissues during ancylostomiasis has been reported [60 , 63] . In case of O . felineus infection , cholesterol accumulates in the liver , however , no triglyceride and phospholipid accumulation occurs . It is little known about the main source and proportion of nutrients for O . felineus [37 , 64] , but liver flukes inhabiting the bile ducts are likely able to uptake nutrients from the bile . Young et al . reported [65] that closely related species O . viverrini and C . sinensis transcribe genes encoding enzymes and accessory proteins directly involved in processing of bile constituents , and can utilize free lipids . Notably , these genes are absent in blood flukes and tapeworms . Generally , cholestasis and biliary injury are associated with marked abnormalities of cholesterol metabolism and pronounced elevations in serum cholesterol and LDL [66 , 67] . Here mechanisms that can lead to cholesterol accumulation in liver tissue should be pointed out . At first , injury of epithelial bile duct cells can result in cholesterol excretion failure . Secondly , infection-associated cellular injury may need extra cholesterol for new membrane synthesis and be beneficial to the host for protection from harmful effects of the stimuli [68] . The last assumption has good concordance with our 31P MRS data . The increased PME level has been hypothesized to be associated with intensified cell membrane synthesis [27 , 69] and regeneration [70 , 71] . The observed in our study correlations of PME and NTP-α peaks with the ALT , AST , cholesterol , HDL levels in serum and the cholesterol/phospholipid ratio in liver tissue of the infected hamsters supported the relation of these spectral parameters to liver injury and fibrosis . However , there are some doubts concerning the power of the 31P MRS technique for assessing the degree of hepatic fibrosis [72] . Published literature has significant variations in the discovered correlation between 31P MRS peaks or its ratios and various liver abnormalities . Nonetheless a fibrosis marker , emphasized by the majority of authors , is the PME/PDE ratio; changes in the PME/PDE are thought to be associated with an increase in the regenerative efforts made by the damaged liver [73] . In our study we found neither changes in this ratio in O . felineus infection nor its correlation with fibrosis [72] . Thus , the correlation of NTP-α and PME peaks with fibrosis observed in our study can mainly reflect the metabolic changes in the liver , i . e . intensification of gluconeogenesis [74–76] . The assumption about intensification of gluconeogenesis in the liver in O . felineus infection is supported by the rise in PME as well as PME/NTP-α , PME/NTP-γ ratios in the infected group of hamsters . PME resonance , apart from PC and PE ( that are believed to represent phospholipid cell membrane precursors ) , also includes gluconeogenesis intermediates , such as glucose-6-phosphate ( G6P ) and 3-phosphoglycerate ( 3PG ) [27 , 69 , 74] . It is essential to describe the origin of the metabolic alteration . The critical role in local regulation of metabolism is played by the energetic status of the liver , especially the level and proportion of high-energy phosphates . In 31P MRS spectra , NTP-γ , -β and -α peaks are attributed to ATP , but also contain resonances from other triphosphates [77] . The NTP-β peak contains information almost exclusively from ATP [78] . In our experiment , the level of NTP-β as well as the Pi/NTP-β ratio in the liver of the infected hamsters did not change . Note that the ATP concentration in the liver of O . felineus infected animals , determined by the chemiluminescent assay , also maintained at a constant level . Thus , it may be concluded , that at this stage of infection the bioenergetics of the liver is not significantly disturbed . Nonetheless , NTP-α and–γ peaks , apart from ATP , contain a contribution from ADP [77] in turn NAD+/NADP+ and NADH/NADPH resonances are the components of the NTP-α peak . 31P MRS data showed that Pi/NTP-α and Pi/NTP-γ as well as NTP-β/NTP-α ratios increased markedly in the infected group . Although 31P MRS did not allow us to make an accurate conclusion about the level of AMP and ADP , the observed spectral alteration indicated the changes in the high-energy phosphates ratio in the liver of the infected hamsters . In turn , this proportion plays a crucial role in cell metabolism regulation [79] . A drop in the AMP/ATP ratio results in dephosphorylation and inactivation of AMPK [49 , 80 , 81] . The decrease in the phospho-AMPK level in O . felineus infected liver and the correlation of the relative phospho-AMPK level with NTP-α peak intensity observed according to western blot analysis support our above-mentioned suggestions . So , when the AMP level is low and the ATP level is normal , glycolysis is nearly switched off and gluconeogenesis is promoted [82] . In turn , acetyl-CoA in the presence of adequate stores of ATP and low AMP levels is diverted to lipid synthesis [83] . Thus , lowering of the AMPK activity leads to a decrease in HMG-CoA reductase phosphorylation and promotes cholesterol synthesis [49] . Since acetyl-CoA carboxylase ( ACC ) is a substrate of AMPK , depletion of AMP increases the catalytic activity of ACC and the level of malonyl-CoA , which is a critical precursor for biosynthesis of fatty acids [49 , 80] . The invert correlations of the phospho-AMPK level with triglycerides and cholesterol serum concentration can be considered as an indirect evidence of lipid synthesis intensification in the infected animal liver . The mechanism underlying the decrease in the AMPK activity has to be further investigated , but it is important to note that , for example , TNF-α and IL-6 cytokines suppress AMPK [84 , 85] . At the system level , chronic liver injury and inflammation per se provoke chronic stress followed by a rise in the cortisol/corticosterone level . There are some publications about elevation of plasma cortisol in animals infected by parasitic worms [86 , 87] . This glucocorticoid ( -s ) has several effects on metabolism , in particular , it provokes mobilization of non-hexose substrates from extrahepatic tissues and stimulates synthesis of glucose in the liver . Simultaneously , chronic glucocorticoid excess is not associated with increased lipolysis in liver tissue; moreover , it may cause activation of hepatic lipogenesis [88 , 89] . What is more , determined in our study glycogen concentration in liver tissue of the experimental hamsters did not reduce during infection . Additionally , glucocorticoids demonstrate anti-inflammatory and immunosuppressive properties that can contribute to success of longstanding invasion of liver flukes . The contribution of cortisol/corticosterone to the observed metabolic disorder should be investigated further . Apart from injury , inflammation and regeneration , the specific parasite-host interaction , i . e . molecules excreted by parasite adult worms during infection , may be the factor ( s ) leading to the observed metabolic changes [60 , 90] . In fact , the parasite’s goal is to take over the host , to provide comfortable conditions for its own habitat . Arguably , Opisthorchiidae flukes are unable to synthesize cholesterol de novo . As an alternative , for example , O . viverrini has molecular pathways to acquire , transport and process cholesterol from external lipids [65] . Thus , biosynthesis of cholesterol by the host is very important for the parasite . Here the question arises , whether the parasite may evolve some opportunity ( excreted molecular stimuli ) to regulate the cholesterol metabolism of the host . Indeed , Adam et al . observed marked proliferation of smooth endoplasmic reticulum ( ER ) in hepatocytes of animals infected with O . viverrini [91] . In turn , enzymes in smooth ER are involved in lipid synthesis , including phospholipids and steroids . For energy production the Opisthorchiidae fluke uses anaerobic and aerobic glycolysis [65]; during anaerobic glycolysis glucose is catabolized most likely to lactate . In spite of the fact that the parasite's genome encodes all of the tricarboxylic acid cycle enzymes living in microaerobic environment , it cannot effectively metabolize end-products of glycolysis along oxidative pathways and , therefore , they have to be excreted into the surrounding tissues of the host [92] . Further end-products of glycolysis may be utilized by the host liver in gluconeogenesis ( preferable in the liver ) or oxidative energy producing pathways . Recently in in vitro experiments it has been shown that excretory products of O . viverrini altered glycolysis/gluconeogenesis in normal human cholangiocytes ( H69 cell line ) [93] . The proposed changes of the high-energy phosphate ratios , as well as glucose metabolism , particularly , the gluconeogenesis rate , in O . felineus infected liver require additional investigation . However , the hypothesis about prevention of AMPK signaling activation by the parasite to retain the biosynthetic function of the liver , for the organ to produce cholesterol and glucose that are essential for liver fluke survival , seems not incurious . Having analyzed the whole set of the obtained data , we may conclude that O . felineus infection disturbs hepatic metabolism of the host . The liver is considered to be a metabolic power station of mammalians , where cholesterol homeostasis relies on an intricate network of cellular processes , which deregulations can lead to severe pathologies [94] . In turn , the decreased AMPK activation , indicated in the infected hamster liver , is known to be implicated in metabolic disorders and associated with cancer risk [95] . Future investigations may shed light on the mechanism that leads to the observed metabolic changes and evaluate the contribution of O . felineus infection to known metabolic disorders . The knowledge about the capabilities of the parasite to regulate metabolism of the host may be applied for seeking a target for anti-helminth therapy and serve as the basis for developing new approaches to human metabolic disease regulation .
The use of MR-techniques for investigating O . felineus infected liver has some advantages . MRI and MRS provide valuable information about the severity of liver impairments induced by opisthorchiasis . MR images of O . felineus infected liver have a characteristic pattern that differs from that of other liver fluke infections . The MRS findings point out the metabolic changes in the infected liver and identify the areas in which further scientific investigation is required . Further MR examination of infected humans allows clinicians to implement these methods in routine diagnosis of opisthorchiasis . | Opisthorchiasis caused by Opisthorchis felineus is a fish-borne parasitic worm infection spread in Russia and some European countries . The morbidities provoked by O . felineus infection are cholangitis and bile duct fibrosis . Long-term infection is associated with high risk of developing cholangiocarcinoma , a generally incurable bile duct cancer type . Due to lack of specific symptoms , O . felineus infection often escapes detection . Thus , opisthorchiasis diagnosis , especially in non-endemic regions , is a serious problem for physicians . In the paper , an animal model of O . felineus induced opisthorchiasis has been evaluated by in vivo magnetic resonance spectroscopy ( MRS ) and magnetic resonance imaging ( MRI ) . Application of MR-techniques allowed to detect a characteristic MRI pattern of O . felineus infected liver , get valuable information about the severity of organ impairments , and bring to light some metabolic changes provoked by the helminth in the liver . Since damages take place in different liver lobes , MRI has the potential to overcome sampling variability of liver biopsy that limits liver fluke-induced fibrosis staging . The use of MR-techniques is very advantageous for investigating parasitic infection . Collection of experimental MR-data gives a new impulse to examination of infected humans and encourages to implement these methods in routine diagnosis of infections , including but not limited to opisthorchiasis . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion",
"Conclusion"
]
| [
"biliary",
"system",
"medicine",
"and",
"health",
"sciences",
"liver",
"diagnostic",
"radiology",
"fibrosis",
"vertebrates",
"animals",
"mammals",
"liver",
"diseases",
"magnetic",
"resonance",
"imaging",
"developmental",
"biology",
"gastroenterology",
"and",
"hepatology",
"research",
"and",
"analysis",
"methods",
"hamsters",
"lipids",
"imaging",
"techniques",
"liver",
"fibrosis",
"cholesterol",
"bile",
"ducts",
"biochemistry",
"rodents",
"radiology",
"and",
"imaging",
"magnetic",
"resonance",
"spectroscopy",
"diagnostic",
"medicine",
"anatomy",
"biology",
"and",
"life",
"sciences",
"amniotes",
"organisms"
]
| 2017 | Magnetic resonance imaging and spectroscopy for differential assessment of liver abnormalities induced by Opisthorchis felineus in an animal model |
Adenosine is a constituent of many molecules of life; increased free extracellular adenosine indicates cell damage or metabolic stress . The importance of adenosine signaling in basal physiology , as opposed to adaptive responses to danger/damage situations , is unclear . We generated mice lacking all four adenosine receptors ( ARs ) , Adora1−/−;Adora2a−/−;Adora2b−/−;Adora3−/− ( quad knockout [QKO] ) , to enable investigation of the AR dependence of physiologic processes , focusing on body temperature . The QKO mice demonstrate that ARs are not required for growth , metabolism , breeding , and body temperature regulation ( diurnal variation , response to stress , and torpor ) . However , the mice showed decreased survival starting at about 15 weeks of age . While adenosine agonists cause profound hypothermia via each AR , adenosine did not cause hypothermia ( or bradycardia or hypotension ) in QKO mice , indicating that AR-independent signals do not contribute to adenosine-induced hypothermia . The hypothermia elicited by adenosine kinase inhibition ( with A134974 ) , inosine , or uridine also required ARs , as each was abolished in the QKO mice . The proposed mechanism for uridine-induced hypothermia is inhibition of adenosine transport by uridine , increasing local extracellular adenosine levels . In contrast , adenosine 5′-monophosphate ( AMP ) –induced hypothermia was attenuated in QKO mice , demonstrating roles for both AR-dependent and AR-independent mechanisms in this process . The physiology of the QKO mice appears to be the sum of the individual knockout mice , without clear evidence for synergy , indicating that the actions of the four ARs are generally complementary . The phenotype of the QKO mice suggests that , while extracellular adenosine is a signal of stress , damage , and/or danger , it is less important for baseline regulation of body temperature .
The nucleoside adenosine is incorporated covalently into many molecules of life , including those involved in information archiving and translation , energy storage and use , regulation of gene expression and function , and intermediary metabolism . Free , extracellular adenosine levels are normally low , but are increased by cell damage and metabolic stressors , such as hypoxia and seizures . Elevated extracellular adenosine thus indicates danger or damage to tissues and elicits adaptive responses . These responses are typically adaptations to the danger state , including selective regulation of perfusion and dampening of ( presumably excessive ) inflammatory/immune responses . A central question in adenosine physiology is the contribution of adenosine to basal homeostasis versus its clear role in stressed states [1 , 2] . Adenosine signals via four G protein–coupled adenosine receptors ( ARs ) , A1AR , A2AAR , A2BAR , and A3AR . A1AR and A3AR are generally coupled to Gi , reducing cAMP and considered inhibitory , while A2AAR and A2BAR are typically coupled to Gs/Gq and are stimulatory , increasing cAMP/Ca++ [3] . AR activation causes a plethora of physiology , with some actions unique to one AR and others with contributions from multiple ARs . Approved AR drugs include adenosine for paroxysmal supraventricular tachycardia ( A1AR agonism ) , adenosine and regadenoson for myocardial perfusion imaging ( A2AAR agonism , causing coronary artery vasodilation [4] ) , istradefylline for Parkinson’s disease ( A2AAR antagonism [5] ) , and the nonselective AR antagonist theophylline for asthma . Caffeine , another nonselective AR antagonist , is found in coffee and certain medications for treatment of drowsiness , headache , or migraines . Selective AR ligands are being investigated for additional indications [6 , 7] . Other drugs , such as dipyridamole and low-dose methotrexate , increase local adenosine levels , thus secondarily activating ARs . One physiologic effect of adenosine is hypothermia [8] . Adenosine-induced hypothermia was initially attributed to brain A1AR [9 , 10] , possibly through ARs in the nucleus of the solitary tract [11] or preoptic area [12 , 13] . A1AR may also regulate torpor ( a hibernation-like state with hypothermia and reduced metabolism , heart rate , and respiration ) [14] , but neither A1AR nor A3AR is required for fasting-induced torpor [15] . Additionally , activation of mouse mast cell A3ARs also causes hypothermia [16–18] , via histamine release and signaling through histamine H1 receptors [19 , 20] . Hypothermia is also caused by A2AAR agonists , which produce vasodilation , and by A2BAR agonists , which may be acting centrally [21] . Thus , agonism of any of the four ARs can produce hypothermia . It is unknown if adenosine also causes hypothermia by additional mechanisms . Could cellular uptake and phosphorylation of adenosine increase adenosine 5′-monophosphate ( AMP ) /ATP ratios , activating the energy sensor adenosine monophosphate–activated protein kinase ( AMPK ) [22] ? AMPK is proposed to have a role in signaling torpor/hypothermia [23] . We produced Adora1−/−;Adora2a−/−;Adora2b−/−;Adora3−/− mice ( quad knockout [QKO] mice ) to examine this idea and study the basal physiology and pathophysiology of mice lacking all four ARs .
QKO mice ( Adora1−/−;Adora2a−/−;Adora2b−/−;Adora3−/− ) were produced by crossing the previously characterized individual AR knockout mice [18 , 24–26] as detailed in Mice in Materials and methods . The Adora1− allele was slightly underrepresented in the progeny [27 , 28] , without pairwise interaction with genotype at the other three loci . QKO mice at 2 months of age appeared to breed normally , with litter sizes similar to controls ( control , 9 . 8 ± 0 . 6 per litter , n = 11 litters versus QKO 9 . 6 ± 0 . 9 per litter , n = 5 litters; P = 0 . 85 , unpaired t test ) . Breeding of 5-month-old QKO mice produced smaller litters ( QKO litters of 1 , 6 , 3 , and 7 pups versus control litters of 8 , 9 , 10 , and 9 pups ) . In addition , the QKO mice showed decreased survival starting at about 15 weeks of age ( Fig 1A ) . Two independent lines of Adora1−/− mice also show decreased survival ( https://www . jax . org/strain/014161 and George Yuning Huang , National Institute of Diabetes and Digestive and Kidney Diseases , personal communication ) ; to our knowledge , decreased survival has not been reported for Adora2a−/− , Adora2b−/− , or Adora3−/− mice . The QKO mice exhibited some eye inflammation and dermatitis , like Adora2a−/− mice ( Dorian McGavern , National Institute of Neurological Disorders and Stroke , personal communication ) . Growth , metabolism , and blood characteristics were measured in cohorts of male and female mice ( summarized in Table 1 ) . At 8–16 weeks of age , chow-fed mice had a slightly lower body weight due to reduced lean , but not fat , mass ( S1 Fig and S2 Fig ) . These mice also showed improved glucose tolerance and lower serum free fatty acid , cholesterol , and insulin-like growth factor-1 ( IGF-1 ) levels ( S1 Fig and S2 Fig ) . Corticosterone levels were similar to controls with intact diurnal rhythm ( control versus QKO: 8 AM 26 . 0 ± 5 . 2 versus 19 . 5 ± 4 . 3 ng/mL and 6 PM 149 . 6 ± 18 . 1 versus 103 . 9 ± 14 . 3 ng/mL; n = 10/group ) . Blood monocyte and polymorphonuclear leukocyte counts were reduced ( S1 Table ) . Serum chemistries revealed increased alkaline phosphatase and confirmed the low cholesterol levels ( S2 Table ) . On a high-fat diet , QKO mice were remarkably like control mice . An independent cohort of chow-fed older male mice was slightly heavier ( body weight , white adipose tissue [WAT] , brown adipose tissue [BAT] , and liver , which likely accounts for the higher leptin and insulin; S3 Table ) . In these mice , there was no difference in β-hydroxybutyrate or thyroxine ( T4 ) , while triiodothyronine ( T3 ) was increased and corticosterone was reduced . Overall , QKO and control mice showed some phenotypic differences , which were generally neither striking nor present in all cohorts . Kidney function ( as assessed by serum creatinine and urea nitrogen , S2 Table ) were similar in control and QKO mice . Water intake was similar in control ( 3 . 71 ± 0 . 11 g/day , n = 16 ) and QKO ( 4 . 00 ± 0 . 28 g/day , n = 17; P = 0 . 36 ) mice . Necropsies of male , 32-week-old QKO and control mice ( three of each ) did not reveal any gross differences in the heart , lungs , liver , kidneys , spleen , gastrointestinal tract , testes , epididymis , seminal vesicles , urinary bladder , preputial gland , or brain . Similarly , no histological abnormalities were apparent in kidney ( with tubular lipidosis in both control and QKO mice ) , heart , lung , spleen , WAT , or BAT . Control and QKO male mice tolerated surgical implantation of intraperitoneal ( i . p . ) sensors for core body temperature ( Tb ) similarly ( 100% survival in 13 QKO and 12 controls ) . The light phase Tb , dark phase Tb , and Tb span , a measure of Tb variation , were the same in QKO and control mice ( Table 2 ) . Tb and activity had a similar diurnal rhythm in QKO and control mice , and all mice entered torpor when challenged with a 24-hour fast ( Fig 1B ) . QKO and control mice increased Tb and activity similarly in response to handling for food removal ( Fig 1B ) or the stress of being placed in a cage previously occupied by an unfamiliar male . mRNA levels in WAT and BAT showed variable increases in Ucp1 , but not Cidea or Cox8b ( S4 Table ) . These data indicate that ARs are not required for baseline control of Tb , induction of torpor , or bidirectional Tb regulation . We investigated heart rate and blood pressure by telemetry in unanesthetized , freely moving female mice . Implantation of radio transmitters requires unilateral carotid artery occlusion; survival from this surgery was reduced in the QKO mice ( 5/11 [45%] QKO versus 34/42 [81%] non-QKO mice survived; P = 0 . 017 by χ2 ) . The telemetered QKO mice had a higher light phase heart rate and lower dark phase blood pressure , but the small sample size limits the conclusions ( Table 3 ) . The usual diurnal effects ( dark phase increases in systolic , diastolic , and mean arterial blood pressure ( MAP ) , and in heart rate , Tb , and physical activity ) were similar in QKO and control mice . Thus , ARs are not required for diurnal rhythmicity of blood pressure and heart rate . Adenosine causes hypothermia in wild-type mice , via each of the four ARs . The 100 mg/kg dose of adenosine equals about 1/10th of the molar ATP content of the body , so instantaneous conversion of the adenosine ( 1 adenosine + 2 ATP → 3 ADP ) would change a body’s ATP/ADP ratio from 10 to 2 and could signal an energy-depleted state , inducing hypothermia , independent of ARs . When male QKO mice were treated with adenosine , no hypothermia occurred . Similarly , treatment of female QKO mice with adenosine had no clear effect on Tb , heart rate , blood pressure , and activity , while in control mice , adenosine reduced Tb , heart rate , MAP , and activity and increased pulse pressure ( Fig 2A–2E ) . These results demonstrate that adenosine actions at A1AR , A2AAR , A2BAR , and/or A3AR mediate the full effects of adenosine-induced hypothermia , bradycardia , and hypotension . This result argues against the hypothesis that pharmacologic dosing of adenosine produces an AR-independent energy-depletion signal . Lipopolysaccharide ( LPS ) stimulates the innate immune system , increasing cytokine levels and causing Tb changes in a dose-dependent manner , with lower doses increasing Tb and higher doses causing hypothermia followed by fever [29 , 30] . Adora2b−/− mice are reported to have increased basal and LPS-stimulated cytokine levels [31] and , pharmacologically , A2AAR [32] and A3AR [18] agonists inhibited LPS-induced cytokine production . We examined the QKO mice to see if their response to LPS was exaggerated . At a low dose ( 10 μg/kg , i . p . ) , LPS increased Tb slightly in control ( Tb at 60–360 minutes versus vehicle: +0 . 52 ± 0 . 19°C , P = 0 . 018 ) and , similarly , in QKO ( +0 . 58 ± 0 . 18°C , P = 0 . 007 ) mice . At a dose of 50 μg/kg , the QKO mice showed less ( P = 0 . 015 ) increase in Tb in this interval ( +0 . 89 ± 0 . 19°C , P = 0 . 0009 in control versus +0 . 31 ± 0 . 12°C , P = 0 . 023 in QKO ) . At a higher LPS dose ( 250 μg/kg , i . p . ) the control mice demonstrated an increase in Tb , while the QKO mice had a mild early reduction in Tb , followed by a Tb increase ( Fig 3A–3E ) . These data are consistent with a subtly augmented Tb response to LPS in the QKO mice . We next studied the cytokine response to LPS ( 250 μg/kg , i . p . ) at 2 hours after dosing . Of the 31 analytes with detectable levels , 24 increased upon treatment with LPS . Interestingly , there was no significant effect of genotype or genotype × treatment interaction on plasma levels of any of these 31 proteins ( S5 Table ) . Thus , in the QKO mice we observed neither a change in basal cytokine levels , nor an exaggerated cytokine response to LPS , suggesting that the innate immune system was not activated because of the lack of ARs . AMP is a proposed natural regulator of torpor , and injection of large doses causes hypothermia [33] . The contribution of ARs to AMP-induced hypothermia is debated [15 , 34–37] . In the QKO mice , the hypothermia caused by AMP ( 500 mg/kg , i . p . ) was attenuated , but not abolished ( Fig 4A–4D ) . Thus , AMP-induced hypothermia has both AR-dependent and AR-independent components . Adenosine kinase phosphorylates adenosine to AMP , keeping intracellular adenosine levels low , thereby drawing extracellular adenosine into cells down its concentration gradient . It has been suggested that inhibition of adenosine kinase causes hypothermia via AR-independent mechanisms [38] . We confirmed that treatment of wild-type mice with the adenosine kinase inhibitor A134974 [39] caused hypothermia in a dose-dependent manner ( Fig 5A–5D ) . In QKO mice , A134974 ( 5 mg/kg , i . p . ) had no effect on heart rate , blood pressure , or Tb , while a 10-fold lower dose caused hypothermia , bradycardia , and hypotension in control mice ( Fig 5E–5I ) . These results suggest that the acute hypothermia and cardiovascular effects of adenosine kinase inhibition are due to increased extracellular adenosine acting on ARs , with no reason to invoke non-AR mechanisms . Caffeine is a brain-penetrant micromolar antagonist of all four human ARs [40] and three of the mouse ARs ( not A3AR ) [41 , 42] . Caffeine can also act at multiple other sites . In control mice , caffeine ( 30 mg/kg , i . p . ) caused a prolonged increase in Tb , activity , and blood pressure , all of which were attenuated or lost in the QKO mice ( Fig 6A–6J ) . These data suggest that these actions of this dose of caffeine are mediated by ARs . Inosine is formed by deamination of adenosine and has no known dedicated receptors , but has in vivo actions with multiple proposed mechanisms . Treatment of wild-type mice with inosine ( 200 mg/kg , i . p . ) produced hypothermia and hypoactivity . Both effects were lost in the QKO mice ( Fig 7A–7D ) . The hypothermia was also lost in KitW−sh/W−sh mice , which lack mast cells ( Fig 7E–7H ) , and in Adora3−/− but not Adora1−/− mice ( Fig 7I and 7J ) . The most parsimonious explanation of the data is that inosine is acting on mast cell A3AR , causing mast cell activation , including histamine release and hypothermia [19 , 43 , 44] . Inosine is probably not increasing extracellular adenosine by transporter inhibition [45 , 46] , as this would cause hypothermia via other ARs , in addition to A3AR ( see below ) . A recent study found that plasma uridine levels decreased with feeding and increased with fasting and suggested that uridine was linked to the reduction in Tb during fasting [47] . We replicated the effect of exogenous uridine ( 1 , 000 mg/kg , i . p . ) to cause hypothermia in control mice . However , the uridine did not cause hypothermia in QKO mice ( Fig 8A–8D ) . The uridine-induced hypothermia was distinct from inosine-induced hypothermia , as it was intact in Adora1−/− mice and attenuated but not eliminated in Adora3−/− , Adora1−/−;Adora3−/− , and KitW−sh/W−sh mice ( Fig 8E–8J ) , suggesting action via A3AR and also A2AAR , and/or A2BAR . How does uridine cause hypothermia via the ARs ? Uridine and adenosine are both substrates for the equilibrative nucleoside transporter 1 ( ENT1 ) , competing with each other for transport [45 , 46] . An ENT1 inhibitor ( 6-S-[ ( 4-Nitrophenyl ) methyl]-6-thioinosine [NBMPR] , 1 mg/kg , i . p . ) had slight effects on Tb by itself , but greatly amplified the hypothermia caused by adenosine ( 100 mg/kg , i . p . ) ( Fig 9A–9D ) . We identified a uridine dose ( 1 , 000 mg/kg , i . p . ) with a modest effect on Tb and tested its effect on adenosine-induced hypothermia , finding an increased effect ( Fig 9E–9H ) . We next measured plasma adenosine in uridine-treated ( 2 , 000 mg/kg , i . p . ) wild-type mice at 30 minutes after dosing , when uridine levels peak [48] . Uridine levels were low ( ≤2 μM ) in vehicle-treated controls and 3 , 360 ± 190 μM in uridine-treated mice ( Fig 9I ) ; uridine treatment tended to increase plasma adenosine levels ( 0 . 842 ± 0 . 086 μM in vehicle-treated versus 1 . 33 ± 0 . 06 μM in uridine-treated mice; P = 0 . 08 ) ( Fig 9J ) . Taken together , these results suggest that uridine inhibits adenosine transporters such as ENT1 , thereby increasing extracellular adenosine levels and causing hypothermia via mast cell A3AR and other ARs .
We were surprised by the baseline phenotype of mice lacking all ARs . We had hypothesized that QKO mice might have a basally activated immune system and not survive or be healthy . Young QKO mice turned out to be viable and breed readily . We did observe reduced blood monocyte and neutrophil levels , but we do not know the mechanism . Like Adora1−/− mice ( https://www . jax . org/strain/014161 ) , they showed decreased survival starting at about 4 months of age , with no identified cause of death . While the QKO mice showed no mortality from the surgery for i . p . implantation of Tb telemeters , they did exhibit increased mortality from surgery for cardiovascular telemeter implantation , which requires ligation of one carotid artery . A crucial difference is that carotid ligation is an ischemic event , increasing brain adenosine , which is neuroprotective via multiple ARs [52] . These protective effects cannot occur in the QKO mice , confirming adenosine’s beneficial effects in certain “danger” situations . While LPS treatment showed a subtly modified effect on Tb , the QKO mice had cytokine levels ( basal and LPS-stimulated ) that were the same as those in control mice . This differs from the reported elevated basal and LPS-stimulated TNF-α and IL-6 levels in Adora2b−/− mice [31] . Possible explanations for this difference include LPS dose , genetic background , and effect of the other AR genotypes . However , most literature on the role of ARs in modulating the LPS response examines the effect of exogenous drugs [18 , 32 , 44 , 53] , not of AR loss . Basal cytokine levels were reported unchanged with loss of A2AAR [32] or A3AR [18] . This evidence again suggests that adenosine is more important for “emergency physiology” in danger/damage situations , and not under basal conditions . Adenosine physiology can be difficult to study . Adenosine , the natural ligand , is locally cleared in a few seconds [54] , with a plasma half-life of about 1 minute in humans [55] . The rapid clearance means that experimentally , massive adenosine doses are used , with a complicated relationship to endogenous physiology . One solution is use of synthetic ligands , which increase the effect duration and likely provide supraphysiologic activation . It is not known if AR physiology from transient adenosine binding differs from the prolonged activation from agonist ligands . Some synthetic ligands are selective for one AR , but ligand selectivity can vary by >1 , 000-fold between species [41] . The use of knockout mice is a powerful tool for ensuring ligand selectivity , with loss of drug effect in the null being strong evidence for an “on-target” action , due to the identified AR . However , QKO mice develop from conception without any ARs , which may elicit compensatory mechanisms , underestimating the role of the missing AR . Therefore , a more prominent phenotype of the QKOs might appear in different experimental stress models . Finally , ARs can heterodimerize with other G protein-coupled receptors and use multiple signaling routes [56] , so it is possible that probing adenosine effect may require understanding of dimerization partners and differential effects on signaling pathways . The clear result that adenosine-induced hypothermia is absent in QKO mice makes these mice extremely valuable for investigating the possible role of adenosine in physiology thought to be caused by other agents . For example , the AR contribution to AMP-induced hypothermia is controversial [15 , 34–37] . The attenuation of AMP-induced hypothermia in the QKO mice demonstrates that both AR-independent and AR-dependent mechanisms contribute . Additionally , the QKO mice enable study of the non-AR mechanism . In contrast , the QKO mice demonstrate that the acute hypothermia caused by an adenosine kinase inhibitor is completely attributable to ARs . Some of the hypothermia is due to the increased extracellular adenosine activating A1AR [38] . While possible explanations for the remaining A134974-induced hypothermia have been proposed , the QKO results dictate that the non–A1AR-mediated hypothermia must occur via A2AAR , A2BAR , and/or A3AR . Inosine does not have identified dedicated receptors . Mechanisms that might explain inosine’s neuroprotective and immunomodulatory actions include all four ARs and non-AR routes ( reviewed in [57] ) . Inosine can activate A3AR on mouse mast cells [43 , 58] . Inosine has also been reported to be an agonist at A1AR [59] , A2AAR [60 , 61] , and A2BAR [62] . Our QKO and follow-up experiments demonstrate that inosine-induced hypothermia is due to A3AR agonism . Caffeine is an AR antagonist [40 , 41] . At higher concentrations , it inhibits phosphodiesterases [63] , γ-aminobutyric acid A ( GABAA ) receptors [63] , and glycine receptors [64] and activates ryanodine receptors [65] . We examined a 30 mg/kg dose , which approximates the human plasma exposure of 2–3 cups of coffee [66] . Our data confirm that the effects of caffeine ( 30 mg/kg ) to increase Tb and activity are mediated by ARs , likely A2AAR [42 , 67 , 68] . The protective effects of guanosine in the central nervous system ( CNS ) have been studied , revealing apparent adenosine-dependent and -independent effects , but a guanosine receptor has not been identified [69–71] . While more mechanistic studies of guanosine’s effects are warranted , we were unable to explore this with QKO mice , because a high , solubility-limited dose of guanosine ( 60 mg/kg , i . p . ) [70] failed to lower Tb in wild-type mice . We were surprised by the QKO results demonstrating that uridine-induced hypothermia requires ARs . Uridine ( at doses reaching ≥10 mM in plasma ) causes hypothermia in mice and rats [48] . A recent study found that physiological plasma uridine levels decreased with feeding and increased with fasting ( albeit in a much lower range , 4–10 μM ) and suggested that uridine was linked to the reduction in Tb that occurs during fasting [47] . The result that uridine-induced hypothermia is absent in QKO mice caused us to investigate further . Uridine itself does not bind to ARs [72] , is not a substrate for adenosine kinase [73] , and does not block adenosine deaminase [74] . However , uridine is a substrate for transport by ENT1 and a competitive inhibitor of adenosine transport by ENT1 [45 , 46] , and we showed that uridine mimicked the effects of the selective ENT1 inhibitor , NBMPR . Thus , we propose that the pharmacologic effects of uridine are due at least in part to inhibition of ENT1 , preventing adenosine uptake into cells . The increased extracellular adenosine activates multiple ARs ( mast cell A3AR and other ARs ) . This may explain the sleep-promoting and anticonvulsant effects of uridine ( reviewed in [75] ) . It seems likely that this AR-dependent mechanism for Tb regulation with large exogenous uridine doses does not explain Tb effects of changes of uridine levels in the physiologic range . The QKO mice described here are a valuable research tool to study AR-independent effects of adenosine . Adenosine is an obligatory product of S-adenosylmethionine ( SAM ) –dependent transmethylation reactions , which include DNA methylation and histone methylation . Through mass action , adenosine provides negative feedback control for transmethylation—low levels of adenosine drive DNA methylation , whereas high levels of adenosine block DNA methylation [76] . The QKO mice would be an ideal investigative tool to study the epigenetic mechanisms of adenosine in the absence of potentially confounding ARs . We monitored growth and metabolic parameters , Tb , activity , and cardiovascular end points . Other assays might expose additional differences in the QKO mice . While the mice have a mixed genetic background , both QKO and control mice were bred using a scheme to minimize founder/inbreeding effects . Lastly , mice with germline AR deletion may have undergone adaptive changes during development , masking deficits due to loss of ARs [77 , 78] . Use of conditional AR alleles with deletion in adulthood would avoid this potential confound . We have demonstrated that the physiology elicited by adenosine , uridine , inosine , caffeine , and ( partially ) AMP requires ARs . It is notable that the QKO mouse phenotype replicates that of individual AR knockout mice , without evidence for synergy , suggesting that the actions of the four ARs are generally complementary . The phenotype of the QKO mice supports the hypothesis that extracellular adenosine has less of a contribution to baseline physiology and is likely more important for its roles as a signal of stress , damage , and/or danger .
Studies were approved by the Animal Care and Use Committee of National Institute of Diabetes and Digestive and Kidney Diseases , animal protocol K016-DEOB-17 . Mice were anesthetized with isoflurane or ketamine and xylazine . Euthanasia of anesthetized mice was performed by exsanguination via retro-orbital bleeding followed by cervical dislocation . Male C57BL/6J and KitW−sh/W−sh ( JAX 012861 ) mice were obtained from the Jackson Laboratory . Adora1−/− [24] , Adora2b−/− [26] ( JAX 022499 ) , and Adora3−/− [18] mice were on a C57BL/6J genetic background . The Adora2a−/− mice [25] ( JAX 010685 ) were on a mixed background . Mice were genotyped as described ( S3 Fig ) [15 , 21] . Mice were singly housed at about 22°C with a 12:12-hour light–dark cycle . Chow ( NIH-07 , Envigo , Madison , WI ) or high-fat diet ( D12492 , 60% kcal fat , 5 . 24 metabolizable kcal/g; Research Diets ) and water were available ad libitum . Mice were studied ≥7 days after any operation or prior treatment . Reuse of mice tends to reduce physical activity levels , presumably due to acclimation . No specific effort was made to acclimate mice to handling in individual experiments . To generate QKO mice , mice heterozygous at all four loci ( Adora1+/−;Adora2a+/−;Adora2b+/−;Adora3+/− ) were crossed . Of 334 progeny ( 310 successfully genotyped at all four loci ) , the Adora2a , Adora2b , and Adora3 alleles were transmitted in the expected mendelian ratios . The Adora1− allele was slightly underrepresented ( +/+:+/−:−/− , observed 109:146:65 versus expected 80:160:80 , χ2 P = 0 . 0007 ) , as observed previously [27 , 28] . There was no pairwise interaction of the Adora1−/− genotype with genotype at the other three loci ( the mice with Adora1−/− genotype had the expected 1:2:1 +/+:+/−:−/− ratios for each of the Adora2a , Adora2b , and Adora3 loci ) . The QKO mice studied were the second- or third-generation progeny from the quad heterozygote intercross . Controls were +/+ or +/− at the four AR loci and are also second- or third-generation progeny of quad heterozygote intercross . Compounds ( source; vehicle ) were obtained as follows: 5′-AMP ( Sigma A1752; saline ) , adenosine ( Sigma; 10% DMSO in saline ) , uridine ( Sigma 1707114; saline ) , A134974 ( Sigma A2846; saline ) , caffeine ( Sigma; saline ) , dipyridamole ( Sigma; 10% DMSO in saline ) , EHNA ( erythro-9- ( 2-Hydroxy-3-nonyl ) adenine hydrochloride , Tocris 1261; 10% DMSO; 10 mg/kg dose chosen to avoid phosphodiesterase inhibition [79] ) , inosine ( Sigma I4125; saline; dose based on [80] ) , NBMPR ( 6-S-[ ( 4-Nitrophenyl ) methyl]-6-thioinosine , Tocris 2924; warm 10% DMSO in saline ) , and LPS ( Sigma L6511; saline ) . Tb and activity were continuously measured by telemetry , using i . p . G2 E-mitters , ER4000 energizer/receivers , and VitalView software ( Starr Life Sciences , Oakmont , PA ) , with data collected each minute [81] . Average Tb over 0–60 minutes and nadir Tb during 0–90 minutes after drug injection were used to quantitate hypothermic effects . Average activity over 0–60 minutes captures the reduced activity that accompanies hypothermia . Total minutes <34°C , a better measure of prolonged hypothermia , did not add useful information . Inhibitors were dosed 20–25 minutes before agonists . Continuous ambulatory intra-arterial blood pressure , heart rate , physical activity , and Tb were measured with radio transmitters ( HD-X11 , Data Sciences International , St Paul , MN ) implanted under ketamine and xylazine anesthesia in the carotid artery as described [82] . Data were sampled at 1 , 000 Hz using a PhysioTel RPC-1 receiver and Ponemah v6 . 30 ( Data Sciences International ) software , with 1-minute averages used for analysis . The pulse pressure is the difference between the systolic and diastolic blood pressure , expressed as a percentage of the systolic pressure . The following assays were used: β-hydroxybutyrate ( colorimetric , #K632 , BioVision ) , corticosterone ( RIA , #07120102 , MP Biomedicals ) , T3 ( ELISA , #T3043T-100 , Calbiotech ) and T4 ( ELISA , #T4044T-100 , Calbiotech ) . Cytokines were measured in plasma obtained 2 hours after dosing [83] LPS ( 250 μg/kg , i . p . ) or vehicle using the Mouse Magnetic Luminex Assay Kit ( LXSAMSM , R&D Systems ) , performed by the NHLBI Flow Cytometry Core . Plasma adenosine and uridine measurement: 30 minutes after dosing with uridine or vehicle , C57BL/6J mice were anesthetized with isoflurane , and blood was drawn retro-orbitally into an equal volume of ice-cold stop solution ( 5 μM EHNA , 200 μM dipyridamole , 4 mM EDTA in saline ) to prevent generation or loss of adenosine [84 , 85] . Samples were centrifuged ( 4°C , 13 , 000 rpm , 3 minutes ) , 50 μL of plasma was pipetted into 200 μL of methanol containing internal standards ( 13C9 , 15N2-uridine and 13C5-adenosine , Cambridge Isotopes ) , vortexed ( 5 minutes ) , centrifuged ( 4°C , 14 , 000 rpm , 10 minutes ) , and the supernatant was dried at room temperature under nitrogen and resuspended in 125 μL H2O . Samples were analyzed by UPLC-MS/MS using a Vanquish UPLC ( Thermo Scientific ) and Altis LC-Triple quadrupole mass spectrometer ( Thermo Scientific ) with heated electrospray ionization ( HESI-II , Thermo Scientific ) in positive ion mode ( 3 , 500 V ) . Injection volume was 1 μL ( autosampler at 5°C ) , using a Waters Cortecs C18 + 2 . 7 μm , 2 . 1 × 100 mm UPLC column ( 35°C ) at 350 μL/minute with solvent A ( 0 . 1% formic acid in H2O ) and solvent B ( 0 . 1% formic acid in methanol ) . Initially , 99% solvent A for 0 . 25 minutes , linearly changing to 10% A at 2 . 15 minutes , remaining at 10% A until 3 minutes , linearly changing to 99% A by 3 . 5 minutes , and stabilizing for 3 minutes . Transitions quantitated were ( m/z ) : adenosine , 268→119 and 268→136; 13C5-adenosine 273→119 and 273→136; uridine , 245→96 and 245→113; and 13C9 , 15N2-uridine , 256→101 and 256→119 . Adenosine was calibrated ( R2 ≥ 0 . 999 ) from 1 . 0 to 2 , 500 ng/mL and uridine from 0 . 25 to 750 μg/mL in the original samples . Literature plasma adenosine levels are 100 to 1 , 000 nM in human [54] and 145 to 6 , 100 nM in mouse [38 , 86 , 87] . Metabolic phenotyping , including glucose and insulin tolerance tests and hormone and metabolite profiles , were performed as previously described [88] . Complete blood count with differential and serum chemistry tests were performed by the Department of Laboratory Medicine , NIH Clinical Center . Statistical analyses were all two tailed , performed using Prism , and are detailed in S6 Table . | Elevated extracellular adenosine generally indicates metabolic stress or cell damage and regulates many aspects of physiology . We studied “QKO” mice lacking all four adenosine receptors . Young QKO mice do not appear obviously ill , but do show decreased survival later in life . QKO mice demonstrate that adenosine receptors are not required for growth , metabolism , breeding , and body temperature regulation . QKO mice are missing the pharmacologic effects of adenosine on body temperature , heart rate , and blood pressure . Therefore , all of these effects are mediated by the four adenosine receptors . We also determined that the hypothermic effects of a pharmacologic adenosine kinase inhibitor ( A134974 ) , uridine , or inosine each requires adenosine receptors . The uridine-induced hypothermia is likely due to its inhibition of adenosine uptake into cells . QKO mouse physiology appears to be the sum of the individual knockout mice , without evidence for synergy , indicating that the actions of the four adenosine receptors are generally complementary . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"caffeine",
"glycosylamines",
"medicine",
"and",
"health",
"sciences",
"alkaloids",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"physiological",
"parameters",
"signs",
"and",
"symptoms",
"body",
"temperature",
"inosine",
"adenosine",
"cardiology",
"chemistry",
"blood",
"pressure",
"uridine",
"heart",
"rate",
"biochemistry",
"rna",
"diagnostic",
"medicine",
"nucleic",
"acids",
"nucleosides",
"hypothermia",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"vascular",
"medicine",
"glycobiology"
]
| 2019 | Physiology and effects of nucleosides in mice lacking all four adenosine receptors |
Trypanosomes , protozoan parasites of medical importance , essentially rely on post-transcriptional mechanisms to regulate gene expression in insect vectors and vertebrate hosts . RNA binding proteins ( RBPs ) that associate to the 3’-UTR of mature mRNAs are thought to orchestrate master developmental programs for these processes to happen . Yet , the molecular mechanisms by which differentiation occurs remain largely unexplored in these human pathogens . Here , we show that ectopic inducible expression of the RBP TcUBP1 promotes the beginning of the differentiation process from non-infective epimastigotes to infective metacyclic trypomastigotes in Trypanosoma cruzi . In early-log epimastigotes TcUBP1 promoted a drop-like phenotype , which is characterized by the presence of metacyclogenesis hallmarks , namely repositioning of the kinetoplast , the expression of an infective-stage virulence factor such as trans-sialidase , increased resistance to lysis by human complement and growth arrest . Furthermore , TcUBP1-ectopic expression in non-infective late-log epimastigotes promoted full development into metacyclic trypomastigotes . TcUBP1-derived metacyclic trypomastigotes were infective in cultured cells , and developed normally into amastigotes in the cytoplasm . By artificial in vivo tethering of TcUBP1 to the 3’ untranslated region of a reporter mRNA we were able to determine that translation of the reporter was reduced by 8-fold , while its mRNA abundance was not significantly compromised . Inducible ectopic expression of TcUBP1 confirmed its role as a translational repressor , revealing significant reduction in the translation rate of multiple proteins , a reduction of polysomes , and promoting the formation of mRNA granules . Expression of TcUBP1 truncated forms revealed the requirement of both N and C-terminal glutamine-rich low complexity sequences for the development of the drop-like phenotype in early-log epimastigotes . We propose that a rise in TcUBP1 levels , in synchrony with nutritional deficiency , can promote the differentiation of T . cruzi epimastigotes into infective metacyclic trypomastigotes .
Gene expression regulation is required to balance the synthesis of the necessary protein components that a cell needs to survive , divide or differentiate . In eukaryotes , the first layer of regulation is at the level of transcription . Beyond this layer of regulation reside several mechanisms regulating gene expression at the post-transcriptional level . The complexity of these post-transcriptional mechanisms , operating over protein-coding transcripts , covers from mRNA processing in the nucleus , to silencing in cytoplasmic foci [1] . RNA-binding Proteins ( RBPs ) are crucial for these processes to be achieved in a controlled fashion , recognizing specific sequences or structural motifs mainly in the non-coding 3´untranslated regions ( 3´-UTR ) of mRNAs . There are many identifiable RNA-binding domains ( RBDs ) , of which the RNA-Recognition Motif ( RRM ) is the best-characterized [2] . However , the exact function of an RBP cannot be inferred by the presence of one or more RBDs . Recent findings suggest that intrinsically disordered sequences and low complexity ( LC ) domains accompanying RBDs in RBPs could play an important role in protein-protein interactions , as well as in the recruitment of other proteins for the formation of ribonucleoprotein ( RNP ) complexes [3] . Trypanosomes have proven to be very interesting models for the study of RBPs and post-transcriptional regulation of gene expression . These are single-celled parasites affecting humans and domestic animals , in which post-transcriptional mechanisms have an enormous influence on the final outcome of gene expression [4 , 5] . The genome organization of trypanosomes and related kinetoplastid organisms is highly unusual: intron-less protein-coding genes are organized into large polycistrons . A few transcription start sites with no canonical RNA polymerase II promoter sequences are in charge of transcription initiation . The synthesis of the polycistron is concomitant with its processing by trans-splicing and polyadenylation , giving rise to mature monocistrons . Thus , multiple open reading frames , unrelated in function , are synthesized . As such , post-transcriptional mechanisms such as degradation , silencing and translation efficiency of mRNA seem to be the central events in the regulation of gene expression in trypanosomes [5 , 6] . Trypanosoma cruzi is the parasite responsible for Chagas disease in the Americas . This protist presents a complex life cycle , alternating between a vertebrate host and an insect vector [7] . These dissimilar environments force T . cruzi to accomplish differentiation processes to cope with different nutrients , immune responses and temperatures . Metacyclogenesis is the process by which non-infective replicating epimastigotes develop into infective non-replicating metacyclic trypomastigotes in the hindgut of the insect . This process is characterized by the repositioning of the mitochondrial DNA ( Kinetoplast ) towards the posterior region of the cell body [8] . After a blood meal , metacyclic trypomastigotes are released together with the faeces , allowing them to infect the host , either through mucous membranes or through small abrasions produced from scratching . After cellular infection of the host , metacyclic trypomastigotes are released from the parasitophorous vacuole , allowing them to differentiate into amastigotes , which replicate in the cytoplasm . After a defined number of divisions , amastigotes differentiate into trypomastigotes , which are released from the infected cell , allowing them to infect new cells . Alternatively , trypomastigotes can be ingested by an insect vector in a blood meal , which then develop into non-infective replicating epimastigotes in the insect , closing the cycle [7] . How trypanosomes accomplish these transformation processes without transcription initiation regulation at every single gene has been a matter of study by many research groups . Several lines of evidence point towards the role of RBPs in regulating the beginning of developmental programs to which trypanosomes commit at the differentiation outset [9–12] . In general , most experimental approaches have been able to demonstrate the impact of RBPs on mRNA abundance , suggesting roles as modulators of mRNA stability [13 , 14] . However , only recently we are beginning to understand the effect of RBPs on the translational modulation of mRNA in trypanosomes [15 , 16] . In this line of evidence , trypanosomes have shown to respond to starvation stress by shutting down protein synthesis and protecting mRNAs from degradation through their recruitment to mRNA granules [17 , 18] . TcUBP1 is one of the best characterized proteins from starvation-induced mRNA granules in T . cruzi , and is one of the most characterized RBPs in trypanosomes [17 , 19–21] . In this work , we show the growth phase-dependent phenotypic result of the ectopic inducible expression of this single RRM-containing RBP . TcUBP1 expression in early-log epimastigotes promoted a drop-like phenotype with several characteristics of epimastigotes undergoing metacyclogenesis . However , when TcUBP1 was induced in late-log epimastigotes we obtained complete progression to infective metacyclic trypomastigotes . Functional characterization of TcUBP1 by tethering it to the 3’ UTR of a reporter mRNA , unveiled its role as a translational repressor . Furthermore , we demonstrate that TcUBP1-induced differentiation process is dependent on both the N and the C-terminal Q-rich LC sequences , highlighting the influence of these uncharacterized domains .
The transformation of non-infective epimastigotes into infective metacyclic trypomastigotes ( metacyclogenesis ) involves changes in the pattern of expressed genes , resulting in important morphological and functional differences between these developmental forms of T . cruzi [22] . This differentiation process can be stimulated in vitro by incubating epimastigotes from a late logarithmic phase of growth in a chemically defined medium with the given name of Triatomine Artificial Urine ( TAU ) , followed by an incubation in TAU supplemented with glucose and amino acids [8] . Fully developed metacyclic trypomastigotes emerge after 96 hs of incubation . While assessing TcUBP1 levels during TAU in vitro metacyclogenesis of wild type ( wt ) non-infective replicative epimastigotes to infective non-replicative metacyclic trypomastigotes , we found that TcUBP1 protein levels increased significantly after 72 h of differentiation onset ( Fig 1A ) . To reduce the burden of complexity of the differentiation process , and to better understand the effect of the rise in TcUBP1 levels , we made use of a Tetracycline ( Tet ) -based inducible system [23] to express TcUBP1-GFP ectopically in non-infective epimastigotes . We decided to use a GFP-fusion protein to allow easy tracking of recombinant protein expression , considering that TcUBP1-GFP fusion has the same dynamic localization patterns as endogenous TcUBP1 [17 , 19] . GFP by itself was expressed in the same way to serve as a negative control . A concentration of 0 . 05 μg/ml Tet was sufficient to generate a detectable increase in the levels of TcUBP1-GFP and GFP ( Fig 1B and 1C ) . At this level of expression , TcUBP1-GFP was distributed in the cytoplasm and nucleus ( S1 Fig , top panel ) , as expected [17 , 19] . However , higher Tet concentrations promoted preferential TcUBP1-GFP accumulations in the nucleus ( S1 Fig , bottom panel ) . In these parasites , cytoplasmic TcUBP1-GFP was lower than in parasites induced with a lower concentration of Tet . We have previously documented the nucleocytoplasmic nature of TcUBP1 by nuclear accumulation under arsenite stress [19 , 24] . We suggest that excessive protein concentration in the nucleus is likely to be the result of TcUBP1 that is not engaged in RNP complexes . This observation argues for a stringent regulation of TcUBP1 levels in the cytoplasm . Because of this , we continued our experiments with parasites expressing TcUBP1-GFP at low Tet concentrations ( 0 . 05 μg/ml ) , which would reflect a protein localization that is similar to the wt protein . Flow cytometric analysis showed low expression leakage in uninduced cells , and a high percentage of fluorescent cells after Tet addition ( Fig 1D ) . Expression of TcUBP1-GFP could be detected 24 hs after induction . At this level of expression , the level of total TcUBP1 ( endogenous plus ectopic ) was 80–90% higher than endogenous TcUBP1 level in uninduced parasites ( Fig 1E ) . TcUBP1-GFP levels remained constant for 5 days , after which Tet was re-added ( Fig 1E ) . Strikingly , induced ectopic expression of TcUBP1-GFP in parasites from an early logarithmic phase of growth promoted a shift in the position of the kinetoplast as compared to induced GFP parasites ( Fig 1F , compare upper and bottom panels ) . In TcUBP1-GFP parasites we observed an orthogonal kinetoplast related to the axis that is formed by the flagellum and nucleus ( Fig 1F , upper panel ) . In order to quantitate this effect , we measured the angle that is formed between the flagellum and the kinetoplast using the center of the nucleus as the vertex , hereafter FNK angle . In parasites ectopically expressing TcUBP1-GFP we obtained a mean FNK angle of 81° with a standard deviation ( SD ) of 40° , while parasites expressing GFP displayed a mean FNK angle of 21° with a SD of 15° ( Fig 1G , S2 and S3 Figs , S1 Table ) , showing a statistically significant difference ( p<0 , 0001 ) . This kinetoplast disposition is similar to that observed in intermediate forms of wt parasites under the first 24–48 h of in vitro metacyclogenesis [22 , 25] ( S4A Fig ) , or in the parasites developing in the rectum of the insect vector Triatoma infestans [25] . In some parasites , repositioning of the kinetoplast almost surpassed the nucleus , showing a similar position of the kinetoplast in parasites at 48 h of in vitro-induced metacyclogenesis ( magnification Fig 1F , S4A and S4B Fig ) . This particular phenotype was noticed in 68% of TcUBP1-GFP expressing parasites , being practically absent in GFP expressing cells ( 5% ) ( Fig 1H , S2 and S3 Figs ) . TcUBP1-GFP expressing parasites also showed a shape change , with a wider and rounded cell body as compared to parasites expressing GFP ( Fig 1F ) . We used the term drop-like phenotype for these parasites , as previously defined by Kollien and Schaub for epimastigotes with this morphology [26] . Drop-like forms have been considered as naturally-occurring insect-dwelling intermediate forms between epimastigotes and metacyclic trypomastigotes , having a round posterior end and a kinetoplast beside or at the posterior side of the nucleus [25] . Drop-like parasites expressing TcUBP1-GFP could easily be noticed two days after Tet addition , comprising the majority of the population from day four onwards ( Fig 1I ) . In our hands , parasites with a drop-like phenotype could be detected from the first 24h of in vitro metacyclogenesis ( S4C Fig ) , comprising the majority of the population at the end of the incubation , thus suggesting that these forms can be observed during the differentiation process to infective forms . We would like to highlight that the development of these phenotypic changes only takes place when TcUBP1-GFP is located in the cytoplasm , and not when TcUBP1-GFP accumulates in the nucleus at higher Tet levels ( S1 Fig ) . This observation argues for a cytoplasmic role of TcUBP1 in promoting this phenotype . In order to discard an unspecific differentiation stimulus due to ectopic expression , we analyzed if the expression of another RRM-type RBP could also induce these phenotypic changes . For this , we ectopically induced the expression of TcRBP35-GFP ( TcCLB . 510661 . 230 ) [27] , which was found to be transiently enriched during metacyclogenesis by a differential proteomic approach [28] . No phenotypic changes were found after 5 days of expression of this RBP ( S5 Fig ) , suggesting that induced ectopic expression of any RBP does not induce spontaneous unspecific differentiation . Overall , these results strongly suggest that ectopic expression of TcUBP1 in early-log epimastigotes leads to a phenotypic alteration with the hallmarks of T . cruzi metacyclogenesis . This putative differentiation event in parasites ectopically expressing TcUBP1-GFP made us wonder if it was concomitant with the expression of infective-stages biomarkers . To answer this , we assessed the expression of one of the most characterized virulence factors from T . cruzi , trans-sialidase containing the Shed Acute Phase Antigen repeats ( TS-SAPA ) [29] . TS-SAPA is a marker expressed by metacyclic trypomastigotes and blood trypomastigotes [30] , and not by epimastigotes [31] ( Fig 2A and 2B ) . TcUBP1-GFP expressing parasites , displaying orthogonal kinetoplast and drop-like phenotype , were positive for TS-SAPA staining ( Fig 2C ) , while control parasites were not stained ( Fig 2C ) . Maximal TS-SAPA staining was observed after 10 days of induction ( Fig 2C ) . Induced and uninduced control parasites expressing GFP showed no staining ( Fig 2C ) . Given the high proportion of parasites expressing TcUBP1-GFP , we could also detect the expression of TS-SAPA by Western blot ( Fig 2D ) , showing a similar band pattern as cell-derived trypomastigotes ( T ) , used as a positive control in this experiment . The obtained band pattern is the consequence of the expression of multiple genes [32] . It is noteworthy that uninduced TcUBP1-GFP transgenic parasites showed no expression of TS-SAPA , either by immunofluorescence or Western blot , reflecting that expression is promoted by the rise of TcUBP1-GFP levels . To further push the characterization of drop-like forms , we tested their resistance to lysis by human complement . This feature is only found in trypomastigotes ( metacyclic or cell-derived ) and amastigotes , and not in epimastigotes , through the expression of a variety of complement regulatory proteins [33] . We found a statistically significant increased resistance to lysis by complement in TcUBP1-GFP expressing drop-like parasites as compared to uninduced parasites ( Fig 2E ) . This result suggests that TcUBP1-GFP expression also promotes the readiness to encounter an innate immune response from a host . Trypomastigote Small Surface Antigen ( TSSA ) , a molecule only expressed by cell-derived trypomastigotes [34] ( S6A Fig ) , was not detected in TcUBP1-GFP or GFP-induced parasites ( S6B Fig ) , as expected . The differences between the expressed protein markers and the resistance to complement-mediated lysis suggests that drop-like forms could correspond to developmental intermediates of metacyclogenesis , and that TcUBP1-GFP expression is not triggering unregulated gene expression . Altogether , these results suggest that ectopic expression of TcUBP1 in parasites in early logarithmic phase of growth promotes a differentiation process with the hallmarks of metacyclogenesis . TcUBP1-GFP ectopic expression in early logarithmically growing parasites promotes many of the characteristics of the developmental differentiation process from non-infective epimastigotes to infective metacyclic trypomastigotes ( Figs 1 and 2 ) . However , these parasites did not progress completely to the metacyclic trypomastigote form . Given that metacyclogenesis in T . cruzi requires a nutritional stress response [35] , we induced the expression of TcUBP1-GFP in epimastigotes at late logarithmic phase of growth . This condition , which is more favorable for differentiation due to nutritional stress , resulted in the spontaneous development of metacyclic trypomastigotes at a level of 10-fold when compared to control cultures after 5 days of induction ( Fig 3A ) . Given that ectopic expression using pTcINDEX can already be detected after 24 h of induction , this period of Tet-induction is consistent with the time course of a TAU-induced metacyclogenesis ( S4A Fig ) . Metacyclic trypomastigotes generated in vitro after TcUBP1-GFP induction featured a kinetoplast at the posterior end of the cell , an elongated nucleus shape ( Fig 3B ) , metacyclic-like motility and they did not adhere to glass , all of which are characteristic features of metacyclic trypomastigotes . TcUBP1-GFP derived metacyclic trypomastigotes also expressed TS-SAPA ( Fig 3B ) , as expected . To further push parasites expressing TcUBP1-GFP into differentiation , we performed in vitro metacyclogenesis in TAU-3AAG medium . A significant difference in the number of metacyclic trypomastigotes was once again observed in TcUBP1-GFP induced cultures as compared to control uninduced or GFP expressing parasites ( Fig 3C ) . It is important to point out that the CL Brener stock of parasites used in this work is one of the strains with the lowest capacity to differentiate to metacyclic trypomastigotes [36] , thus explaining the low efficiency of this method . As an alternative way to analyze the spontaneous development of metacyclic trypomastigotes from epimastigotes , we performed infection of VERO cells using the same number of total parasites ( epimastigotes plus metacyclic trypomastigotes ) after induction of TcUBP1-GFP or GFP . We found 20% of infected cells using TcUBP1-GFP expressing parasites , while we only observed 2–3% of infected cells when using uninduced control parasites ( Fig 3D and 3E ) . Uninduced and induced GFP-expressing control parasites showed reduced ( 2–3% ) infection rates ( Fig 3E , S7 Fig ) , basically because in these cultures there are almost no metacyclic trypomastigotes , just epimastigotes . From this result we can make two conclusions; on one hand , that the spontaneous development of metacyclic trypomastigotes induced by TcUBP1-GFP expression can be detected by a method that is independent of the subjective discrimination between epimastigotes and metacyclic trypomastigotes by their morphology; on the other hand , that TcUBP1-GFP induced metacyclic trypomastigotes are infective in vitro , confirming their identity as a real developmental stage form . To confirm that the observed amastigotes were inside the cells , and not attached to the cell surface , we performed a double immunofluorescence assay with differential staining before and after permeabilization , which showed that amastigotes were indeed inside the cells ( Fig 3F ) , as expected . We can conclude that TcUBP1 ectopic expression in early-log epimastigotes promotes a drop-like phenotype , with similar characteristics to metacyclogenesis intermediate forms ( Fig 3G ) ; whereas late-log epimastigotes ( high density populations undergoing nutritional stress ) respond to ectopic expression of TcUBP1 by developing into infective metacyclic trypomastigotes ( Fig 3G ) . Previously , we have proposed that TcUBP1 can decrease the half-life of SMUG mRNAs in vivo [20] , suggesting that it plays a destabilizing effect on this mRNA . The effect of TcUBP1 on epimastigote differentiation shown in this work could be mediated by a decrease in mRNA abundance due to enhanced degradation , or by an effect on mRNA translation . To answer this , we developed a versatile reporter system to tether proteins to a reporter mRNA in T . cruzi epimastigotes using the λN-BoxB approach . For this system we made use of the trypanosome’s capacity of combined polycistronic transcription and post-transcriptional processing to obtain two mature RNA molecules from a single DNA construct ( Fig 4A ) . As a reporter mRNA we used the coding sequence for Firefly Luciferase . In the 3’ UTR of this cistron we introduced five BoxB tracts . In the second cistron we introduced the coding sequence of the 22 amino acid RNA-binding domain of λ bacteriophage antiterminator protein N ( λN ) , followed by a Flag tag and a multiple cloning site , that would allow us to clone the sequence coding for the RBP of interest ( Fig 4A ) . BoxB structures are recognized by λN , thus allowing the tethering of λN fusions to the 3’ UTR of the reporter mRNA [37 , 38] . Furthermore , this in vivo system also allows determining if the tethered RBP can modulate mRNA abundance and/or translation by measuring luciferase mRNA levels and luciferase activity , respectively . The normalization applied to the obtained values is shown in Fig 4B . For baseline measurements we used λN-GFP tethered to the 3’ UTR of luciferase reporter mRNA . This allowed us to compare the results obtained with different RBPs to those obtained with GFP . As a control of performance for our system we analyzed the effect of TcPABP1 , whose ortholog has been previously characterized in a similar system in T . brucei [39] . As expected , TcPABP1 promoted mRNA stability and translation efficiency ( Fig 4C ) , resembling the behavior observed in an analog system used in T . brucei [39] . TcUBP1 only showed a 1 . 9-fold decrease in reporter mRNA abundance ( Fig 4C ) . However , we found that it could induce an 8-fold reduction in translation efficiency of the reporter mRNA , suggesting that TcUBP1 could be a potent translational repressor . To further test the system , we also included other RBPs ( TcRBP4 from the RRM family group , and TcZFP2 and TcZFP3 from the CCCH Zinc Finger Family ) , for which a functional characterization is lacking in T . cruzi . TcRBP4 increased both mRNA abundance and translation almost four-fold when compared to GFP , suggesting a role as an mRNA stabilizer and a translational promoter ( Fig 4C ) . TcZFP2 showed highly variable results on mRNA abundance , while it did not show an effect on translation ( Fig 4C ) . TcZFP3 promoted an increase on mRNA abundance , which did not prove to be statistically significant ( Fig 4C ) . However , it did promote a decrease of 4-fold in reporter mRNA translation when compared to GFP , suggesting that TcZFP3 might be a translational repressor as TcUBP1 ( Fig 4C ) . In summary , these results are in concordance with the proposed role for TcPABP1 , thus validating this T . cruzi in vivo tethering system . It allowed us to determine the putative translational repression feature of TcUBP1 , making it a valuable diagnostic tool to functionally characterize RBPs in T . cruzi . In order to confirm the , up to now , unnoticed function of TcUBP1 as a translational repressor , we used three additional approaches . First , we made use of the SUnSET method [40] , in which puromycin is used in very low amounts for a short time period , and whose incorporation in neosynthesized proteins directly reflects the rate of mRNA translation in vivo . To validate this system , we verified the puromycin labeling of nascent polypeptides in wt parasites , and confirmed that puromycin is not incorporated when translation is completely inhibited by cycloheximide ( CHX ) ( Fig 5A ) . Early-log epimastigotes ectopically expressing TcUBP1-GFP or GFP for 5 or 10 days were incubated with puromycin , and its incorporation was detected by immunoblot with the anti-puromycin antibody . Puromycin incorporation was reduced in a time-dependent manner in parasites expressing TcUBP1-GFP ( Puromycin/Tubulin ratio = 5 . 6 ) , but not in uninduced or GFP-induced parasites ( Puromycin/Tubulin ratio = 9 . 7 and 9 . 9 , respectively ) ( Fig 5B ) . Puromycin incorporation was still observed in TcUBP1-GFP-expressing parasites five days after Tet addition , suggesting that the normal turnover of proteins required to survive was not affected in these cells ( Fig 5B ) . As a control of complete translation inhibition we used CHX , which dropped this ratio to 0 . 4 . Analysis of Puromycin/Tubulin ratio at 10 days after induction emphasized the effect of induced TcUBP1-GFP . Comparing CHX and induced TcUBP1-GFP induction we can conclude that TcUBP1-GFP expression did not completely block translation . We can suggest a role of TcUBP1 as a repressor of translation , operating selectively on certain mRNAs , or globally , by partially reducing translation rates of most mRNAs , but allowing the translation of mRNAs whose protein products will define the metacyclic trypomastigote or drop-like forms . These results , confirm our previous observation of the capability of TcUBP1 to repress translation in the in vivo tethering system , thus supporting a putative role for this RBP in the repression of protein translation in T . cruzi . Second , we determined the translational status of the parasites by analyzing their polysome profiles . Using 15–50% sucrose gradients we separated ribosome subunits ( 60S and 40S ) and monosomes ( 80S ) , which are not actively involved in protein synthesis , and polysomes , which are multiple ribosomes loaded onto single mRNA molecules and thus are an indication of active protein synthesis . When comparing the profile from induced TcUBP1-GFP cells to control profiles ( uninduced TcUBP1-GFP and induced GFP cells ) we found an overall reduction on the polysomal fraction as compared to control profiles ( Fig 5C ) . This was concomitant with an increase in monosomes , indicating translational repression [41] . The comparison of induced GFP and uninduced TcUBP1-GFP polysome profiles showed no differences , suggesting that the induced expression from the pTcINDEX vector is not affecting the translational status of the cells ( Fig 5C ) . A control profile using EDTA showed the disruption of 80S ribosomes into 60S and 40S ribosomal subunits , serving as a control of complete translational inhibition . These results suggest that the translational status in TcUBP1-GFP expressing parasites is reduced , although not inhibited , as compared to control parasites . Finally , we evaluated mRNA granule formation in transfected parasites by fluorescence in situ hybridization . It is known that mRNA granules develop in response to starvation stress in T . cruzi epimastigotes [17] . Transcripts stored in granules are translationally silenced and protected from degradation , awaiting favorable nutritional conditions to resume translation [42] . In T . cruzi , an optimal starvation stimulus typically promotes the formation of mRNA granules in 80–100% of an untransfected population ( Fig 5D ) [17] . Because of this , we analyzed formation of mRNA granules in epimastigotes expressing TcUBP1-GFP or GFP under suboptimal starvation stimulus ( see Materials and methods ) . In this experiment we applied the suboptimal starvation stress 36 h after addition of Tet , not allowing parasites to acquire the drop-like phenotype . The suboptimal starvation stimulus promoted the formation of mRNA granules in 20% of uninduced or GFP induced control parasites , while ectopic expression of TcUBP1-GFP promoted mRNA granules in 80% of induced parasites ( Fig 5E and 5F ) . Also , in TcUBP1-GFP induced cultures the biggest mRNA granule from each cell was almost twice as big as those from uninduced or GFP induced control parasites ( Fig 5G ) , while the number of granules per cell was unaltered in the four populations ( Fig 5H ) . These results support the role of TcUBP1 as a translational repressor by enhancing the condensation and silencing of mRNA in granules under suboptimal starvation stress . Metacyclogenesis is characterized by a massive reduction in translation rates that is concomitant with growth arrest , both in vivo and in vitro [44] . If TcUBP1 is a translational repressor that promotes metacyclogenesis , a growth arrest should be expected . Ectopic expression of TcUBP1-GFP showed a remarkable inhibition of parasite proliferation compared to non-induced cultures ( Fig 6A ) . Growth curves derived from control GFP-expressing parasites showed that inducible expression of the transgene did not affect growth ( Fig 6B ) . As an example of growth arrest induced by translation inhibition , we performed a growth curve of wt epimastigotes in the presence of CHX . The treatment with this translational inhibitor promoted growth arrest sooner than TcUBP1-GFP ( Fig 6C ) . This could be explained by the expected delay of 24 h to reach the desired TcUBP1-GFP levels induced from the pTcINDEX vector . To distinguish between parasite death and proliferation inhibition as a cause of TcUBP1-GFP reduced growth rates , we determined cell viability by incubating parasites with low concentrations of propidium iodide and analyzed them by flow cytometry at days 5 and 10 after Tet addition . We did not observe differences in the amount of dead cells in any of the time points between the studied cultures , indicating that TcUBP1-GFP did not induce cell death ( Fig 6D ) . To test if global reduction of translation promotes differentiation from epimastigotes to metacyclic trypomastigotes , we incubated wt late logarithmic parasites with CHX for 5 days . We observed 86% of dead cells by propidium iodide incorporation ( S8 Fig ) , suggesting that global translational inhibition leads to cell death and that it is not a driver for metacyclogenesis . To analyze if TcUBP1-GFP-induced growth arrest was reversible , we removed Tet from induced cultures at day 5 by washing the cells , and evaluated parasite number , kinetoplast position and TS-SAPA staining for 10 more days ( Fig 6E , 6F and 6G ) . In parasites where TcUBP1-GFP was induced , there was no recovery from the growth defect ( Fig 6E ) . Control populations of induced GFP or uninduced TcUBP1-GFP showed no signs of growth defect . After 5 days from Tet addition , TcUBP1-GFP induced parasites showed 64% of cells with an orthogonal kinetoplast ( Fig 6F ) , and internal staining with the anti-TS-SAPA antibody ( Fig 6G ) . This staining is compatible with previous results using this antibody in epimastigotes undergoing metacyclogenesis [45] . After 5 days from Tet removal ( day 10 of culture ) the proportion of parasites with an orthogonal kinetoplast was 61% ( Fig 6E ) . These parasites showed reduced TcUBP1-GFP fluorescence , while TS-SAPA staining was increased ( Fig 6G ) . At 10 days from Tet removal ( day 15 of culture ) , 53% of parasites still showed an orthogonal kinetoplast ( Fig 6E ) , while there was no sign of TcUBP1-GFP fluorescence ( Fig 6G ) . In contrast , staining with the anti-TS-SAPA antibody was still high . Overall , these results strongly suggest that the transient increase in TcUBP1 levels can promote the beginning of the metacyclogenesis process , which can not be reversed by returning TcUBP1 levels to the previous state , thus showing a commitment with differentiation . TcUBP1 is a protein with modular architecture . A single central RRM provides mRNA binding and target specificity , flanked by N and C-terminal LC sequences , composed by 38 and 43% glutamines , respectively ( Fig 7 ) [24] . N and C-terminal LC sequences seem to be intrinsically disordered structures , whose structure could not be determined earlier [46] . In order to determine the influence of the different TcUBP1 domains and regions on the parasite differentiation phenotype , we ectopically expressed different protein mutants and fragments fused to GFP ( Fig 7 ) . We analyzed TS-SAPA staining , the presence of an orthogonal kinetoplast and the drop-like phenotype in early logarithmic epimastigotes expressing these constructs . As expected , full length TcUBP1-GFP showed all the hallmarks previously analyzed . A triple point RNP1 RRM mutant ( mutTcUBP1 ) , which was previously shown to be unable to bind to RNA or to be recruited to mRNA granules [19] , did not show any of the novel phenotypes of unmodified TcUBP1 . This suggests that TcUBP1-induced differentiation is dependent on binding to RNA . Moreover , TcUBP1 fragments lacking the N ( ΔN ) or C-terminal ( ΔQ ) glutamine-rich LC sequences , did not resemble the effects exerted by TcUBP1 ( Fig 7 ) . These two constructs contain the functional RRM of TcUBP1 , can bind to RNA in vitro [19 , 47] , and can be recruited to mRNA granules [17] . Thus , a functional RRM together with a single LC sequence , either on its N or C-terminal end , is not enough to promote full commitment to the differentiating phenotype . The smallest portion of TcUBP1 with the capability to bind to RNA ( ΔNΔQG2 ) , corresponding to the RRM , was also unable to reproduce the effects of full length TcUBP1 . The complete C-terminal region , comprising the C-terminal LC sequence , or the C-terminal LC sequence alone , did not show any difference to control GFP-expressing parasites ( Fig 7 ) . These results show that only full length TcUBP1 is able to promote the differentiation phenotype . Therefore , removal of low complexity sequences , with previous unknown function , makes TcUBP1 dysfunctional to reproduce this novel phenotype . To sum up , we propose that TcUBP1 engages in multiple interactions involving N and C-terminal Q-rich LC domains while bound to RNA .
It is well established that post-transcriptional mechanisms have a huge impact on trypanosome regulation of gene expression [48] . These mechanisms also dictate the developmental transformation trypanosomes suffer during differentiation from non-infective to infective forms , as a response to environmental stimulus [49] . Current tools allow the analysis of mRNA and proteins that are up or downregulated as a response to external differentiation signals . This is the case of RBP6 in T . brucei [9] , whose coding mRNA was found to be increased by a 13-fold in trypanosomes from the proventriculus . By overexpressing TbRBP6 in cultured noninfectious procyclic trypanosomes , the developmental stages that are observed in tse tse fly were recapitulated , including the generation of infective metacyclic forms expressing the variant surface glycoprotein [9] . However , most mRNAs or proteins that are upregulated in this manner typically correspond to surface proteins or virulence factors associated to the final developmental form [50 , 51] . Also , In T . brucei , the CCCH Zinc Finger protein TbZFP1 is transiently elevated during differentiation from bloodstream to procyclic trypomastigotes forms , while its genetic ablation compromises the parasite differentiation program [10 , 11] . It was also demonstrated that the ectopic expression of TbZFP2 results in a dramatic procyclic stage-specific remodeling of the trypanosome , similar to the morphogenic events of differentiation [11] . Additionally , forced TbRBP10-expression in procyclics induced a switch towards bloodstream-form mRNA expression patterns , with concomitant growth inhibition . Tethering of TbRBP10 to a reporter mRNA inhibited translation , and halved the abundance of the bound mRNA [12] . More recently , it has been shown that TbRBP10 expression promotes the differentiation to the bloodstream state [16] . These lines of evidence support the role of RBPs as potential master regulators of differentiation in trypanosomes . However , data for RBPs promoting differentiation is completely lacking in T . cruzi . In this work , we show that TcUBP1 is enriched during metacyclogenesis in T . cruzi . Previously , De Godoy and coworkers also detected a rise in TcUBP1 levels during metacyclogenesis in a quantitative proteomic approach [28] . TcUBP1 associates to RNA secondary structures rather than to specific sequences [21] , being these structures highly represented in a multitude of mRNAs , including transcripts coding for surface proteins , metabolism enzymes and protein kinases [21 , 52] . This would imply that it can associate to many different transcripts , thus supporting a role for TcUBP1 as a putative regulator of post-transcriptional gene expression . Indeed , ectopic expression of TcUBP1-GFP in early-log epimastigotes cultures showed several hallmarks of parasites undergoing differentiation to metacyclic trypomastigotes . During developmental differentiation of epimastigotes to infective non-replicating metacyclic trypomastigotes the kinetoplast is repositioned to the posterior end of the parasite [25 , 53] . The kinetoplast transits beside the nucleus , during which it can be found in an orthogonal position respective to the axis formed by the flagellum and nucleus in intermediate developmental forms ( S4A Fig ) . Kollien and Schaub adopted the term drop-like morphology for forms found in the rectum of Triatoma infestans starved insects infected with T . cruzi [26] . Drop-like forms could be easily detected as soon as 48 hours after TcUBP1-GFP induction . It takes 24 hours for TcUBP1-GFP to be expressed , so we can conclude that the first emerging drop-like forms acquire this morphology 24 hours after TcUBP1 ( endogenous plus ectopic ) achieves a critical level to initiate translational repression . After the first 48 hours of TcUBP1-GFP induction drop-like forms accumulate , reaching more than half of the population by day 5 . Drop-like parasites derived from wt in vitro differentiated parasites in TAU medium ( S4C Fig ) emerge as soon as 24 hours after incubation . If we compare these two results we conclude that TAU induced metacyclogenesis is a much stronger inducer of metacyclogenesis , although both stimulus seem to operate at similar rate once a critical level of TcUBP1 is reached . In parasites showing the drop-like phenotype we were able to detect TS-SAPA , a protein involved in infection , which is exclusively expressed by infective trypomastigotes of T . cruzi , but not by amastigotes or epimastigotes [29] . We were also able to determine that these parasites do not express TSSA , a surface protein exclusively expressed in bloodstream trypomastigotes [34] . These observations are consistent with the development towards the metacyclic form by TcUBP1-expressing parasites . The expression of TS-SAPA , but not TSSA , clearly argues against unregulated gene expression havoc induced by ectopic expression of this RBP . As an alternative to the use of antibodies to detect stage-specific surface molecules we evaluated the resistance to lysis by complement in drop-like forms . During the differentiation of epimastigotes to metacyclic trypomastigotes , parasites also undergo a physiological change that confers them the capacity to evade complement-mediated lysis [54] . Thus , epimastigotes are not protected from the lytic effects of the innate immune response , while metacyclic trypomastigotes can circumvent it . The increased resistance to complement-mediated lysis by TcUBP1-GFP-induced drop-like forms suggests that complement regulatory proteins become expressed during this morphological metamorphosis [33] , preparing them for the encounter with an immune system . Ectopic expression of TcUBP1 in late logarithmic epimastigotes , which is a culture condition where differentiation is enhanced [36] , promoted spontaneous differentiation under LIT medium culture , and enhanced metacyclogenesis under induced differentiating conditions ( TAU medium ) . Metacyclic trypomastigotes promoted by ectopic expression of TcUBP1-GFP are functional infective forms as shown by infection of VERO cells in culture ( Fig 3C and 3D ) . This result provides an observer-independent way to determine the development of infective metacyclic trypomastigotes induced by TcUBP1 , as compared to control cultures . Furthermore , the intracellular development of TcUBP1-induced metacyclics into amastigotes implies that the progression of the life cycle is not compromised . Thus , it is important to highlight the full functionality of these parasites regarding infection , amastigogenesis and replication . Previously , we have used TcUBP1-GFP ectopic expression in T . cruzi epimastigotes to localize this RBP in mRNA granules [17] , and to show that TcUBP1 is a nucleocytoplasmic shuttling protein [19] . In fact , it seems that cytoplasmic TcUBP1 levels are strictly regulated , since overexpression at high Tet concentrations ( 0 . 5 μg/ml ) promotes nuclear accumulation of this RBP ( S1 Fig ) , probably as a mechanism to compensate for the presence of free TcUBP1 levels in the cytoplasm . In fact , parasites induced with the higher Tet concentrations did not show the drop-like phenotype , suggesting that TcUBP1-GFP can only induce differentiation in the cytoplasm , where translation occurs , and not in the nucleus . However , never before we were able to notice the kinetoplast relocalization and the drop-like phenotype described in this work . This can be explained because of the differences in the way TcUBP1-GFP has been expressed in our previous works . Before , we used a constitutively expressing vector such as pTEX [55] , which requires long antibiotic selection times . This selection process could have been selecting parasites that compensate TcUBP1-induced growth deficiencies , thus diluting and loosing parasites with the drop-like phenotype and with an orthogonal kinetoplast . In fact , pTEX-TcUBP1-GFP transgenic cultures usually contain a low number of parasites expressing TcUBP1-GFP , suggesting that these cells struggled to survive . Here , we have used the Tet-inducible pTcINDEX vector , with which we were able to induce the expression of TcUBP1-GFP in almost the whole population of cells at demand , at similar levels as with the pTEX vector . We can conclude that inducible ectopic expression in trypanosomes is a suitable experimental approach to raise the levels of an RBP and determine its effects on differentiation , phenotypic transformations and expression of differentiation-markers . However , attention should be paid to the levels of ectopic expression; if we had used the higher Tet concentration without previously knowing anything about TcUBP1 , we would have concluded that TcUBP1 is a predominantly nuclear protein that does not have any effect on metacyclogenesis , which is not . Recently , translational modulation has been proposed as a major influence on the resulting protein products in trypanosomes [56] . Proteomic and ribosome profiling approaches have shown numerous differences in protein abundance between developmental stages both in T . cruzi and T . brucei [57 , 58] . Yet , translational modulation has been more elusive to determine , probably because it is difficult to confirm that a given protein is responsible for repressing/enhancing translation . Here , we have developed our own in vivo T . cruzi system to examine the function of any protein when tethered to a reporter 3’ UTR . The virtue of this system relies on its ability to determine if a protein factor affects mRNA abundance and/or translation efficiency . This analysis is performed independently of sequence preference and without the influence of other potential factors binding to neighboring RNA structures . This is the first time that an RBP-tethering system is used in T . cruzi . In T . brucei it has been previously used to determine that , as expected , PABP1 can increase mRNA abundance and translation when tethered to a reporter CAT mRNA , while RBP10 can promote decreased mRNA abundance and translation [12] . Unlike the T . brucei system , which used Northern blot for quantitation of mRNA molecules [12 , 16] , we have implemented the use of Real-Time RT-PCR for the quantification of the reporter and Neomycin resistance mRNAs . In our tethering system , TcPABP1 behaved in the same way as in T . brucei , thus validating our experimental approach . A similar system has been used in T . brucei to show that SCD6 has a role in translational repression [15] . Erben and coworkers have used this kind of experimental approach to identify many potential mRNA regulators with no previously annotated function , or functions unrelated to mRNA metabolism [39] . In our in vivo RBP tethering system we found that TcZFP3 behaved like an mRNA stabilizer and a translational repressor . Unfortunately , contrastable information for the function of this protein is lacking in T . cruzi . In T . brucei , ZFP3 has been proposed to enhance the translation of procyclin EP1 mRNA directly [59] . More recently , ectopic expression of TbZFP3 showed to increase the abundance of two reporter mRNAs bearing the Rbp23 or SmB 3’ UTR [59] . However , this increase in mRNA abundance did not correlate with an increase in CAT activity , suggesting that ZFP3 could also stabilize and repress translation of mRNA [17] . This hypothesis is in clear agreement with the results we obtained for TcZFP3 in our tethering system . We would also like to highlight the potential function of TcRBP4 as an mRNA stabilizer , promoting the translation of the reporter mRNA . This function is the opposite of the one of TcUBP1 , showing an unbiased output for the system . Although mRNA granules from trypanosomes , which contain TcUBP1 , accomplish translational repression in response to a stressful environmental condition , TcUBP1 potential as a translational repressor was never examined before . Previous work from our group proposed an mRNA destabilizing function for TcUBP1 [20] , which was specific for SMUG mRNA and not for other mRNAs . Our new evidence does not rebut this previous hypothesis . The destabilizing effect of TcUBP1 could not be confirmed statistically in our tethering system , probably because TcUBP1 needs another context of protein factors in a RNP complex , which is probably found in SMUG 3’ UTR . Using our T . cruzi tethering system we were able to determine that TcUBP1 can significantly repress translation . Central to the interpretation of our observations was the notorious effect of TcUBP1-GFP ectopic expression in protein translation by using the SUnSET method [40] . This experimental approach was used and validated for the first time in trypanosomes in this work . It allowed us to confirm that TcUBP1-GFP expressing parasites are experiencing a reduced translation rate as compared to uninduced or GFP induced controls ( Fig 5B ) . Although we did not compare its sensitivity to classic [35S]-Met incorporation methods , it is clear that non-radioactive methods provide ease of use , especially when working with pathogens . The translational repression induced by the ectopic expression of TcUBP1-GFP was confirmed by qualitatively analyzing the polysomes profile , showing that translation is reduced although not inhibited . This metabolic condition is clearly different from that found in starved parasites , which lack polysomes , show granules of TcDhh1 ( mRNA granules ) and do not differentiate [60] . Another evidence arguing for a translational repression function in TcUBP1 is the enhancement in the formation of mRNA granules under suboptimal starvation stress ( Fig 5C and 5D ) . SCD6 was previously proposed as a translational repressor [15] , promoting the formation of mRNA granules in T . brucei [43] . Unlike TcUBP1 , TbSCD6 seems to promote the recruitment of silenced mRNAs into granules without affecting ongoing translation [43] . Although TcUBP1 did not enhance the number of granules per cell , it did promote an increase in the size of mRNA granules ( Fig 5E and 5F ) . We can hypothesize that ectopic expression of TcUBP1 could be increasing the speed at which mRNA granules are formed by removing certain mRNAs from the translation pool , leading to a faster response to starvation with increased granule size as compared to control parasites . Previously , TcUBP1-interacting mRNA targets were found to be enriched in genes coding for proteins involved in metabolism . From these transcripts , an RNA motif termed UBP1m was identified in the 3´ UTR of these mRNAs [21] . Accordingly , in a posterior in silico analysis UBP1m was shown to be enriched in transcripts coding for surface proteins , metabolism enzymes and protein kinases [52] . This could be interpreted as TcUBP1 being involved in a regulatory cascade , repressing the translation of several factors required to maintain the epimastigote life form , which in turn triggers metacyclogenesis . As such , the expression of TS-SAPA could be an indirect consequence of TcUBP1 ectopic expression . Unfortunately , these putative key differentiation inhibitors are unknown in T . cruzi . The association between translational repression and differentiation in T . cruzi has been demonstrated before for the kinase TcK2 . TcK2 might be acting as a sensor of cytoplasmic heme , being activated in its absence , thus phosphorylating eIF2α [61] . Phosphorylation of eIF2α is a known trigger to arrest growth and translation initiation in many cell types , including T . cruzi [35] . In consequence , this signal seems to potentiate the differentiation of epimastigotes to metacyclic trypomastigotes as a result of translation inhibition . It was likewise observed for TcUBP1-GFP expressing parasites that a severe defect in growth affected these cells ( Fig 6 ) . This growth defect could be recapitulated by inhibiting translation with CHX . In spite of this , translational inhibition with CHX did not promote differentiation of epimastigotes , suggesting that a developmental program which selectively represses the translation of specific factors should be triggered , either by eIF2α phosphorylation , by a rise in TcUBP1 levels , or by other unexplored mechanisms . TcUBP1-induced growth arrest , kinetoplast relocalization and expression of TS-SAPA proved to be irreversible after washing out Tet ( Fig 6E , 6F and 6G ) . We can suggest that these parasites are committed with a developmental program that cannot be turned off . Our results are in accordance with Domingo-Samanes et al . work , who established that the differentiation process in T . brucei is an irreversible and unidirectional bistable switch , involving irreversible commitment steps; and that differentiation is independent of the signal after commitment [62] . Full length TcUBP1 is required for the progression into infectivity , an affirmation that is supported by the mutational analysis we performed over its whole primary structure ( Fig 7 ) . This conclusion is based upon several results . On one hand , it is clear that the differentiation progression is dependent on the binding of TcUBP1 onto mRNA targets; the mutant of the RNP1 octapeptide of the RRM could not recapitulate the effect of unmodified TcUBP1 . We have previously shown that this mutant TcUBP1 RRM is unable to bind to RNA in vitro , and is also unable to associate to mRNA granules [19] . Furthermore , complete deletion of the RRM also rendered an effect-less protein , probably due to the incapacity of TcUBP1 LC sequences to bind to RNA [19] . On the other hand , and to our surprise , elimination of either the N or C-terminal LC sequences of TcUBP1 also prevented the differentiation effect observed for the full-length protein ( Fig 7 ) . As such , we adopted the provisional hypothesis that both LC sequences of TcUBP1 might be involved in the repression of translation , ultimately triggering the process of differentiation . LC sequences have been proposed to orchestrate the dynamic assembly of RNP granules in yeasts and mammalian cells [3] . One of the most emblematic cases of LC-containing RBPs is Fused in Sarcoma ( FUS ) . FUS has been a model protein for the demonstration of the liquid-droplet nature of RNA granules [63] . However , it is not completely clear how translational repression can be modulated by RBPs containing LC domains , as we suggest here for TcUBP1 . Current evidence indicates that LC sequences promote homo and heterotypic interactions for the development of liquid droplets and amyloid-like structures [64] . Interactions of this kind by TcUBP1 could be promoting translational repression by recruiting different mRNA-protein complexes , thus forming submicroscopic foci that promote silencing of the bound RNA . It is tempting to speculate that LC sequences might be the driving force for the biogenesis of the initial puncta that give birth to RNA granules [65] . In summary , we provide mechanistic evidence of the role of the RBP TcUBP1 on T . cruzi differentiation through translation repression . Together , our results deepen the knowledge of gene expression regulation that occurs during differentiation from non-infective to infective parasite forms .
Trypanosoma cruzi epimastigotes , strain CL Brener , were cultured in liver infusion tryptose ( LIT ) medium with 0 . 001% bovine hemin and 10% heat-inactivated fetal calf serum ( LIT complete medium ) at 28° C . Experiments were performed with cultures at early logarithmic phase of growth ( 1–2 x 107/ml ) or at late logarithmic phase of growth ( 5–6 x 107/ml ) . To determine parasite densities , we used a hemocytometer without fixation . Transgenic parasites containing the Luciferase reporter in pTEX vector were maintained in 200 μg/ml G418 . For the differentiation into metacyclic-trypomastigotes we essentially followed the protocol described by Bonaldo et al . [66] . Briefly , cultures of epimastigotes at 5 x 107 cells/ml were collected by centrifugation and resuspended at a density of 5 x 108 cells/ml in Triatomine Artificial Urine ( TAU ) medium [190 mM NaCl , 17 mM KCl , 2 mM MgCl2 , 2 mM CaCl2 , 8 mM phosphate buffer ( pH 6 . 0 ) ] . After 2 h of incubation at 28 °C the cells were diluted to 5 x 106 per ml in TAU supplemented with 10 mM L-proline , 50 mM L-glutamic acid , 2 mM L-aspartic acid and 10 mM glucose and incubated at 28 °C for 96 h . For tethering assays , transfected parasites were cultured in agitation at 28 °C at a density of 3 x 107 parasites/ml . For starvation studies , parasites were previously washed twice with PBS . Optimal starvation stimulus was defined as incubating late logarithmic epimastigotes cultures ( 5–6 x 107/ml ) in PBS ( plus Tet for induced cultures ) for 24 h . Suboptimal starvation stimulus was defined as incubating mid logarithmic epimastigotes cultures ( 3 x 107/ml ) in PBS ( plus Tet for induced cultures ) for 24 h . To analyze the percentage of parasites with mRNA granules a minimum of 200 cells were analyzed per culture through 5 different experiments . To define the size of granules , the area of the biggest granule from 20 different parasites for each culture was measured using ImageJ . To analyze the number of granules per cell , 40 parasites with mRNA granules were analyzed . For time-course growth curves , a culture of 3 x 107 per ml was used as a starter inoculum for the initial culture starting at 5 x 106 parasites/ml [67] . Cultures were incubated in agitation at 28 °C in the presence or absence of Tet or cycloheximide . For recovery studies , induced cultures were washed with fresh media 5 days before the addition of Tet , and allowed to resume growth for 5 more days in complete LIT medium without Tet . For the construction of the pTcINDEX-GFP vector we PCR-amplified the GFP open reading frame ( ORF ) using Primers GFP Fwd BamHI ( cgggatccCATGGTGAGCAAGGGCGAGGAGC ) and GFP Rev BglII ( cgagatctTTACTTGTACAGCTCGTCCATGCC ) from pTEX-GFP vector [17] . After digesting this insert with BamHI and BglII , it was cloned into the BamHI site of pTcINDEX [23] . After checking the orientation of different clones by PCR , a clone with the GFP ORF in the correct orientation was selected for sequencing . This clone contains the GFP ORF flanked by a BamHI site in its 5’ end , while the BamHI at the 3’ end is destroyed by ligation with the BglII end . After BamHI digestion and removal of 5’ phosphates in the resulting pTcINDEX vector , we subcloned the full length TcUBP1 ORF , or its RNP1 mutant and several fragments that were previously cloned into the BamHI site of the pTEX-GFP vector [17 , 19] . Transfection and selection of parasites was essentially performed as previously described [23] . Briefly , epimastigotes were first transfected by electroporation with circular pLEW13 DNA , after which they were selected in complete medium containing 200 μg/ml G418 . Stably transformed parasites , obtained after six weeks , were re-transfected with pTcINDEX-GFP constructs and selected with 100 μg/ml Hygromycin , together with G418 . Induction of recombinant proteins from the pTcINDEX-GFP vector was performed by the addition of Tet at 0 . 05 μg/ml final concentration , or otherwise stated . Parasites reached the desired induction levels after 24 h of Tet addition . For incubation times longer than five days , Tet was re-added at 0 . 05 μg/ml at day five of the time curve . For all protein extracts , parasites were first washed once with PBS . For Western blot , parasites were lysed for 15 min in TBS plus 0 . 5% NP-40 , with the addition of 100 μM trans-Epoxysuccinyl-L-leucylamido ( guanidino ) butane ( E64 ) and 1 mM Phenylmethylsulfonyl Fluoride ( PMSF ) ( Lysis Buffer ) . After lysis , samples were centrifuged at 10 , 000 g for 10 min at 4° C . Sample Buffer ( Final concentration: 50 mM Tris-HCl pH 6 . 8; 2% Sodium Dodecyl Sulfate , SDS; 10% Glycerol; 0 . 02% Bromophenol Blue; 100 mM Dithiothreitol ) , was added to the supernatant and immediately boiled for 3 min . Afterwards , samples were treated with DNase I ( Sigma ) for 15 min at room temperature to reduce DNA viscosity . For Western blot , samples were loaded onto 10% SDS-polyacrylamide gels . Gels were transferred to Immobilon-NC transfer membranes ( Millipore ) , probed with rabbit anti-TcUBP1 [21] , mouse anti-GFP ( Roche ) , mouse anti-Flag ( Sigma ) , rabbit anti-puromycin ( a kind gift from Dr . Walter from UCSF ) , and mouse anti-tubulin ( Sigma ) . Rabbit anti-TS-SAPA recognizes a 12 amino acid repeat denominated Shed Acute Phase Antigen ( SAPA ) , which is in the C-terminal part of the molecule , was a kind gift from Dr . Campetella from UNSAM [68] . We used IRDye secondary anti-rabbit antibody and anti-mouse antibodies ( LI-COR ) . Detection was performed using the Odyssey Imaging System ( LI-COR ) . Quantitation of protein bands was performed using Image Studio Lite software ( LI-COR ) . Parasites from uninduced or Tet induced cultures ( 48 hs ) were washed with PBS . Cells ( 1 x 105 ) were analyzed in a CyFlow space cytometer ( Partec , Germany ) . For determination of the percentage of non-viable cells in parasite cultures , cells were washed in cold PBS and then resuspended in PBS supplemented with 2 mM glucose . After that , 1 μg/ml propidium iodide ( PI ) was added 5 minutes before FACS analysis as previously described [69] . The percentage of PI stained cells was analyzed in a CyFlow space cytometer ( Partec , Germany ) . Parasites were washed with PBS and attached to glass slides pretreated with 0 . 01% Poly-L-lysine for 30 min . The excess of parasites was removed and the slides were incubated with 4% paraformaldehyde in PBS for 20 min , washed with PBS and permeabilized with 0 . 2% saponin . Fixed and permeabilized cells were washed and incubated with primary antibody diluted in PBS supplemented with 2% BSA for 1h at room temperature . The slides were washed with PBS , incubated with Alexa 568 goat anti-rabbit IgG ( Life technologies ) , washed with PBS and mounted in the presence of 10 μg/ml of 4-6-diamidino-2-phenylindole ( DAPI ) . Fluorescence In Situ Hybridization ( FISH ) of mRNA was performed essentially as described [17] . Microscopic analyses were performed in an Eclipse E600 Microscope ( Nikon , Tokyo , Japan ) coupled to a SPOT RT color camera ( Diagnostic Instruments ) . To determine the percentage of parasites with kinetoplast repositioning , we analyzed 239 cells for TcUBP1-GFP and 103 for GFP . For TS-SAPA expression determination , we used anti-TS-SAPA . To determine the percentage of spontaneous metacyclic trypomastigotes we analyzed 250 cells for TcUBP1-GFP and for GFP each time in four independent experiments . Transfected epimastigotes were grown to a concentration of 2 x 107/ml , induced or not with Tet for five days , washed once in PBS , and resuspended at 4 x 106/ml in PBS . To analyze complement-mediated lysis , 10% human serum was added . Cells were incubated at 28°C and the number of motile parasites was determined by cell count at 10 , 20 , 30 and 60 min . To monitor protein synthesis , we used the SUnSET method [40] . Briefly , puromycin is an aminonucleoside antibiotic produced by Streptomyces alboniger . It is a structural analog of aminoacyl tRNAs , which is incorporated into the nascent polypeptide chain and prevents elongation . When used in minimal amounts , puromycin incorporation in neosynthesized proteins reflects directly the rate of mRNA translation in vitro . Wt parasites were incubated with CHX ( 50 μg/ml ) or vehicle for 30 minutes at 28° C and then with puromycin ( 10 μg/ml ) or vehicle for 30 minutes at 28° C . After that , parasites were washed and lysed . Puromycin incorporation was determined by Western blot using an anti-puromycin antibody . Protein loading was monitored using an anti-tubulin antibody . Transfected parasites were cultured for 5 and 10 days . After that , 3 x 107 parasites were incubated with puromycin ( 10 μg/ml ) or vehicle for 30 minutes at 28° C . Parasites were washed and lysed . Puromycin incorporation was determined by Western blot using an anti-puromycin antibody . Protein loading was monitored using an anti-tubulin antibody . Western blots were performed using secondary antibodies coupled to Alexa 800 and Alexa 680 . To determine fluorescence intensity blots were analyzed using LI-COR software . For each gradient we used 5x108 parasites . When indicated , parasites were induced with Tet ( 0 . 05 μg/ml ) for 5 days . Before harvesting , cells were incubated with CHX ( 100 μg/ml ) to prevent ribosomal run-off , or EDTA ( 10 mM ) to dissociate ribosomal subunits , for 30 min . , after which were washed in PBS containing CHX or EDTA , respectively . Parasites were lysed at 2 , 5x106 cells/μl in Polysome buffer ( 20 mM Tris-HCl pH 7 . 6 , 120 mM KCl , 5 mM MgCl2 ) supplemented with 0 . 5% NP-40 , 1mM PMSF , 100 μM E64 , 80U/ml RNasin ( Promega ) , and 100 μg/ml CHX or 10 mM EDTA . After clearing by centrifugation at 20 , 000 g ( 10 min , 4° C ) , the soluble lysate was layered on a linear 15–50% sucrose gradient ( 12 ml ) , prepared in polysome buffer . Gradients were centrifuged for 2 , 5 hours at 35 . 000 rpm in a Beckman SW41Ti rotor at 4° C . Absorbance at 254 nm was monitored by using the UV detector of a ÄKTA HPLC system and pumping the gradient from bottom to top using a peristaltic pump . For infection assays we used late logarithmic parasites expressing GFP or TcUBP1-GFP , which were induced with Tet for 5 days . After 5 days , there was a mixture of epimastigotes and metacyclic trypomastigotes in the TcUBP1-GFP culture . Cell invasion assays were carried out by seeding these cultures of parasites ( 2 x 107 total cells ) onto each well of 24-well plates containing 13-mm diameter round glass coverslips coated with 1 x 104 VERO ( African green monkey kidney ) cells ( American Type Culture Collection , VA ) . The medium was removed every 24 h for 72 h . After that , wells were washed with PBS and incubated with 4% paraformaldehyde in PBS for 20 min . Following extensive washing in PBS , cells were incubated for 10 min with NH4Cl and blocked for 1 h in PBS containing 2% ( w/v ) BSA and 2% ( v/v ) normal goat serum . Extracellular ( attached ) parasites were labeled with the addition of a T . cruzi-infected mouse serum followed by Alexa Fluor 568 conjugated secondary antibody ( Invitrogen ) . Intracellular parasites were subsequently labeled with the addition of a T . cruzi-infected rabbit serum ( diluted in PBS with 0 . 5% saponin ) followed by Alexa Fluor 488 conjugated secondary antibody ( Invitrogen ) . The coverslips were washed and mounted in the presence of 10 μg/ml of 4-6-diamidino-2-phenylindole ( DAPI ) . Microscopic analyses were performed in an Eclipse E600 Microscope ( Nikon , Tokyo , Japan ) coupled to a SPOT RT color camera ( Diagnostic Instruments ) . For this construction , we used pTEX vector [55] and the Lambda ( λ ) bacteriophage antiterminator protein N ( λN ) peptide strategy to tether proteins to RNAs [37] . The RNA binding domain of the λN is used to tag the RBP . The reporter mRNA includes the sequence that codes for Firefly luciferase , which was obtained from the pGL3-Basic vector ( Promega ) . Specific sequences of 19 nt ( GGGCCCUGAAGAAGGGCCCUUUCCUUU ) that are binding sites for λN ( boxB ) were inserted into the target RNA . These were artificially synthesized ( Macrogen ) . To generate λN fusion proteins a synthetic sequence that codes for λN domain was designed ( MDAQTRRRERRAEKQAWKAANGGS ) ( Macrogen ) . The last three amino acids ( GGS ) serve as a flexible union between λN domain and the fusion partner . Also , a FLAG tag , a small multiple cloning site and 3 stop codons in each frame were added following the λN coding-sequence . This allowed us to assess the expression of the fusion protein by Western blot , and clone the RBPs as λN-fusion proteins . The intergenic region between the T . cruzi actin I and II genes was used between the boxB sites and the λN-coding cistron . This sequence provides a stable 3´UTR for the reporter mRNA and sites for post-transcriptional processing for the reporter mRNA and for the sequence that codes for the λN-fusion protein . In this manner , using pTEX vector we could express the reporter mRNA that codes for luciferase reporter mRNA and also the λN-fusion proteins ( Fig 4A ) . The major advantage of this system derives from the small size of the λN peptide and its target sequence , preventing interferences with the fused protein . Cloning of GFP , TcPABP and TcUBP1 into pTEX-luciferase vector was performed by subcloning from the pTEX-GFP vector [67] using the enzyme BamHI for TcPABP and TcUBP1 and the enzymes BamHI and HindIII for GFP . For the cloning of TcRBP4 we PCR-amplified the open reading frames ( ORF ) using Primers Fwd BamHI ( ggatccATGCGGAGTCGGAGCAGC ) and Rev BamHI ( ggatccCGTTCCTTCTCTTCTTCGTCCA ) . For TcZFP2 we PCR-amplified the open reading frames ( ORF ) using Primers Fwd BamHI ( ggatccATGTCCTACCCGAATCGTTATG ) and Rev HindIII ( aagcttTCACTGGGTCTGTGCGGGC ) . For TcZFP3 we PCR-amplified the open reading frames ( ORF ) using Primers Fwd BamHI ( ggatccATGCAGGGGTATTTTGCACTC ) and Rev HindIII ( aagcttTTATGACGCCGGCGTTTCTC ) . After checking the orientation of different clones by PCR , clones with the mentioned ORFs in the correct orientation were selected for sequencing . Luciferase activity was determined using the kit Bright-Glo Luciferase Assay System ( Promega ) . Briefly , 1 . 5 x 107 parasites were lysed and luciferase activity was measured following the kit instructions . Luminescence was normalized to the amount of protein present in the sample determined by the Bradford assay . Total RNA was isolated from uninduced and induced parasites with Trizol Reagent according to manufacturer’s instructions ( Life Technologies ) . Total RNA was purified using Direct-zol RNA MiniPrep columns ( Zymo Research ) . RNA integrity was evaluated by 1% agarose gel electrophoresis . Samples were incubated with RQ1 DNAse ( Promega ) followed by enzyme inactivation . First strand cDNA was synthesized from total RNA samples using Superscript II reverse transcriptase ( Life Technologies ) , following manufacturer´s instructions . Real time quantitative PCR ( qPCR ) was performed using Kapa Sybr Fast Universal Kit ( Biosystems ) with primers described below . Luciferase mRNA abundance was measured with the oligonucleotides F 5’-CAATGGGACGTATGGGACAT-3’ and R 5’-TCGTCTGTCACCACCCATAG-3’ . Neomycin resistance gene ( Neo ) mRNA was measured with oligonucleotides F 5’-TTTTTGACCGTGGTGGGAGG-3’ and R 5’-TTGATCCCTTCTCACAGCGG-3’ . The reactions were carried out with the 7500 Real Time PCR System from Applied Biosystems . After selecting stably expressing parasite populations , we used real time RT-PCR to determine luciferase reporter and Neo mRNA levels , while protein levels were determined by Western blot using anti-Flag and anti-tubulin antibodies . Luciferase activity was determined by luminescence and total protein was measured by Bradford assay . Luciferase activity was normalized to the amount of protein in the sample determined by Bradford ( L ) , RBP protein levels was normalized to tubulin protein levels ( P ) and luciferase mRNA levels was normalized to the Neomycin resistance gen mRNA ( Neo ) levels contained in the pTEX vector ( RNA ) . The abundance ( A ) of the reporter mRNA was established as the quotient between RNA and P . The translation efficiency ( T ) of the reporter mRNA was established as the quotient between luciferase activities ( L ) normalized to RNA abundance ( RNA ) and the amount of RBP ( P ) ( Fig 4B ) . Data analysis was performed with GraphPad Prism 6 . 0 ( GraphPad Software , La Jolla , CA , USA ) . Statistical differences were assessed by analysis of variance ( ANOVA ) with Tukey post-hoc analysis for multiple comparisons or with Dunnett post-hoc analysis for multiple comparisons to one control group . Differences with a p-value <0 . 05 were considered significant . For tethering assays data analysis , folds increase over 2 vs . control was considered significant . | Trypanosoma cruzi epimastigotes proliferate in the midgut of the hematophagous insect vector . Insect vectors can spend long periods of time without feeding , during which epimastigotes differentiate to infective metacyclic trypomastigotes in a process termed metacyclogenesis . This metamorphosis involves multiple phenotypic changes , involving the repositioning of the mitochondrial DNA ( kinetoplast ) and expression of virulence factors . Here , we show that the RBP TcUBP1 is transiently enriched during metacyclogenesis , and that ectopic expression of TcUBP1 promotes these phenotypic changes in epimastigotes , finally leading to infective metacyclic forms . Using four different approaches we found that TcUBP1 promotes translational repression together with growth arrest , both of which are characteristics of metacyclogenesis . Mechanistically , we show that low complexity regions in TcUBP1 could be involved in translational repression leading to phenotypic changes , suggesting their involvement in the formation of silenced ribonucleoprotein complexes . We conclude that TcUBP1 can act in a post-transcriptional regulatory cascade by repressing translation of multiple mRNA targets , thus promoting irreversible phenotypic changes leading to metacyclic infective forms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"messenger",
"rna",
"microbiology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"protozoan",
"life",
"cycles",
"developmental",
"biology",
"trypomastigotes",
"protozoans",
"epimastigotes",
"cellular",
"structures",
"and",
"organelles",
"kinetoplasts",
"gene",
"expression",
"life",
"cycles",
"biochemistry",
"rna",
"trypanosoma",
"cruzi",
"trypanosoma",
"eukaryota",
"cell",
"biology",
"nucleic",
"acids",
"protein",
"translation",
"genetics",
"biology",
"and",
"life",
"sciences",
"protozoology",
"organisms"
]
| 2018 | Translational repression by an RNA-binding protein promotes differentiation to infective forms in Trypanosoma cruzi |
Both linear mixed models ( LMMs ) and sparse regression models are widely used in genetics applications , including , recently , polygenic modeling in genome-wide association studies . These two approaches make very different assumptions , so are expected to perform well in different situations . However , in practice , for a given dataset one typically does not know which assumptions will be more accurate . Motivated by this , we consider a hybrid of the two , which we refer to as a “Bayesian sparse linear mixed model” ( BSLMM ) that includes both these models as special cases . We address several key computational and statistical issues that arise when applying BSLMM , including appropriate prior specification for the hyper-parameters and a novel Markov chain Monte Carlo algorithm for posterior inference . We apply BSLMM and compare it with other methods for two polygenic modeling applications: estimating the proportion of variance in phenotypes explained ( PVE ) by available genotypes , and phenotype ( or breeding value ) prediction . For PVE estimation , we demonstrate that BSLMM combines the advantages of both standard LMMs and sparse regression modeling . For phenotype prediction it considerably outperforms either of the other two methods , as well as several other large-scale regression methods previously suggested for this problem . Software implementing our method is freely available from http://stephenslab . uchicago . edu/software . html .
Both linear mixed models ( LMMs ) and sparse regression models are widely used in genetics applications . For example , LMMs are often used to control for population stratification , individual relatedness , or unmeasured confounding factors when performing association tests in genetic association studies [1]–[9] and gene expression studies [10]–[12] . They have also been used in genetic association studies to jointly analyze groups of SNPs [13] , [14] . Similarly , sparse regression models have been used in genome-wide association analyses [15]–[20] and in expression QTL analysis [21] . Further , both LMMs and sparse regression models have been applied to , and garnered renewed interest in , polygenic modeling in association studies . Here , by polygenic modeling we mean any attempt to relate phenotypic variation to many genetic variants simultaneously ( in contrast to single-SNP tests of association ) . The particular polygenic modeling problems that we focus on here are estimating “chip heritability” , being the proportion of variance in phenotypes explained ( PVE ) by available genotypes [19] , [22]–[24] , and predicting phenotypes based on genotypes [25]–[29] . Despite the considerable overlap in their applications , in the context of polygenic modeling , LMMs and sparse regression models are based on almost diametrically opposed assumptions . Precisely , applications of LMMs to polygenic modeling ( e . g . [22] ) effectively assume that every genetic variant affects the phenotype , with effect sizes normally distributed , whereas sparse regression models , such as Bayesian variable selection regression models ( BVSR ) [18] , [19] , assume that a relatively small proportion of all variants affect the phenotype . The relative performance of these two models for polygenic modeling applications would therefore be expected to vary depending on the true underlying genetic architecture of the phenotype . However , in practice , one does not know the true genetic architecture , so it is unclear which of the two models to prefer . Motivated by this observation , we consider a hybrid of these two models , which we refer to as the “Bayesian sparse linear mixed model” , or BSLMM . This hybrid includes both the LMM and a sparse regression model , BVSR , as special cases , and is to some extent capable of learning the genetic architecture from the data , yielding good performance across a wide range of scenarios . By being “adaptive” to the data in this way , our approach obviates the need to choose one model over the other , and attempts to combine the benefits of both . The idea of a hybrid between LMM and sparse regression models is , in itself , not new . Indeed , models like these have been used in breeding value prediction to assist genomic selection in animal and plant breeding programs [30]–[35] , gene selection in expression analysis while controlling for batch effects [36] , phenotype prediction of complex traits in model organisms and dairy cattle [37]–[40] , and more recently , mapping complex traits by jointly modeling all SNPs in structured populations [41] . Compared with these previous papers , our work makes two key contributions . First , we consider in detail the specification of appropriate prior distributions for the hyper-parameters of the model . We particularly emphasize the benefits of estimating the hyper-parameters from the data , rather than fixing them to pre-specified values to achieve the adaptive behavior mentioned above . Second , we provide a novel computational algorithm that exploits a recently described linear algebra trick for LMMs [8] , [9] . The resulting algorithm avoids ad hoc approximations that are sometimes made when fitting these types of model ( e . g . [37] , [41] ) , and yields reliable results for datasets containing thousands of individuals and hundreds of thousands of markers . ( Most previous applications of this kind of model involved much smaller datasets . ) Since BSLMM is a hybrid of two widely used models , it naturally has a wide range of potential uses . Here we focus on its application to polygenic modeling for genome-wide association studies , specifically two applications of particular interest and importance: PVE estimation ( or “chip heritability” estimation ) and phenotype prediction . Estimating the PVE from large-scale genotyped marker sets ( e . g . all the SNPs on a genome-wide genotyping chip ) has the potential to shed light on sources of “missing heritability” [42] and the underlying genetic architecture of common diseases [19] , [22]–[24] , [43] . And accurate prediction of complex phenotypes from genotypes could ultimately impact many areas of genetics , including applications in animal breeding , medicine and forensics [27]–[29] , [37]–[40] . Through simulations and applications to real data , we show that BSLMM successfully combines the advantages of both LMMs and sparse regression , is robust to a variety of settings in PVE estimation , and outperforms both models , and several related models , in phenotype prediction .
We begin by considering a simple linear model relating phenotypes to genotypes : ( 1 ) ( 2 ) Here is an -vector of phenotypes measured on individuals , is an matrix of genotypes measured on the same individuals at genetic markers , is a -vector of ( unknown ) genetic marker effects , is an -vector of 1 s , is a scalar representing the phenotype mean , and is an -vector of error terms that have variance . denotes the -dimensional multivariate normal distribution . Note that there are many ways in which measured genotypes can be encoded in the matrix . We assume throughout this paper that the genotypes are coded as 0 , 1 or 2 copies of a reference allele at each marker , and that the columns of are centered but not standardized; see Text S1 . A key issue is that , in typical current datasets ( e . g . GWAS ) , the number of markers is much larger than the number of individuals . As a result , parameters of interest ( e . g . or PVE ) cannot be estimated accurately without making some kind of modeling assumptions . Indeed , many existing approaches to polygenic modeling can be derived from ( 1 ) by making specific assumptions about the genetic effects . For example , the LMM approach from [22] , which has recently become commonly used for PVE estimation ( e . g . [24] , [44]–[46] ) , is equivalent to the assumption that effect sizes are normally distributed , such that ( 3 ) [Specifically , exact equivalence holds when the relatedness matrix in the LMM is computed from the genotypes as ( e . g . [47] ) . [22] use a matrix in this form , with centered and standardized , and with a slight modification of the diagonal elements . ] For brevity , in this paper we refer to the regression model that results from this assumption as the LMM ( note that this is relatively restrictive compared with the usual definition ) ; it is also commonly referred to as “ridge regression” in statistics [48] . The estimated combined effects ( ) , or equivalently , the estimated random effects , obtained from this model are commonly referred to as Best Linear Unbiased Predictors ( BLUP ) [49] . An alternative assumption , which has also been widely used in polygenic modeling applications [18] , [19] , [34] , and more generally in statistics for sparse high-dimensional regression with large numbers of covariates [50] , [51] , is that the effects come from a mixture of a normal distribution and a point mass at 0 , also known as a point-normal distribution: ( 4 ) where is the proportion of non-zero and denotes a point mass at zero . [This definition of follows the convention from statistics [19] , [50] , [51] , which is opposite to the convention in animal breeding [27] , [32]–[34] , [40] . ] We refer to the resulting regression model as Bayesian Variable Selection Regression ( BVSR ) , because it is commonly used to select the relevant variables ( i . e . those with non-zero effect ) for phenotype prediction . Although ( 4 ) formally includes ( 3 ) as a special case when , in practice ( 4 ) is often used together with an assumption that only a small proportion of the variables are likely to be relevant for phenotype prediction , say by specifying a prior distribution for that puts appreciable mass on small values ( e . g . [19] ) . In this case , BVSR and LMM can be viewed as making almost diametrically opposed assumptions: the LMM assumes every variant has an effect , whereas BVSR assumes that a very small proportion of variants have an effect . ( In practice , the estimate of under LMM is often smaller than the estimate of under BVSR , so we can interpret the LMM as assuming a large number of small effects , and BVSR as assuming a small number of larger effects . ) A more general assumption , which includes both the above as special cases , is that the effects come from a mixture of two normal distributions: ( 5 ) Setting yields the LMM ( 3 ) , and yields BVSR ( 4 ) . we can interpret this model as assuming that all variants have at least a small effect , which are normally distributed with variance , and some proportion ( ) of variants have an additional effect , normally distributed with variance . The earliest use of a mixture of two normal distributions for the regression coefficients that we are aware of is [52] , although in that paper various hyper-parameters were fixed , and so it did not include LMM and BVSR as special cases . Of the three assumptions on the effect size distributions above , the last ( 5 ) is clearly the most flexible since it includes the others as special cases . Obviously other assumptions are possible , some still more flexible than the mixture of two normals: for example , a mixture of three or more normals . Indeed , many other assumptions have been proposed , including variants in which a normal distribution is replaced by a distribution . These variants include the “Bayesian alphabet models” – so-called simply because they have been given names such as BayesA , BayesB , BayesC , etc . – that have been proposed for polygenic modeling , particularly breeding value prediction in genomic selection studies . Table 1 summarizes these models , and some other effect size distributions that have been proposed , together with relevant references ( see also [53] and the references there in ) . Among these , the models most closely related to ours are BayesC [34] and BayesR [35] . Specifically , BayesC without a random effect is BVSR , and with a random effect is BSLMM ( which we define below ) . BayesR models effect sizes using a mixture of three normal components plus a point mass at zero , although the relative variance for each normal distribution is fixed . Given the wide range of assumptions for effect size distributions that have been proposed , it is natural to wonder which are the most appropriate for general use . However , answering this question is complicated by the fact that even given the effect size distribution there are a number of different ways that these models can be implemented in practice , both in terms of statistical issues , such as treatment of the hyper-parameters , and in terms of computational and algorithmic issues . Both these types of issues can greatly affect practical performance . For example , many approaches fix the hyper-parameters to specific values [27] , [32] , [33] , [40] which makes them less flexible [34] , [54] . Thus , in this paper we focus on a particular effect size distribution ( 5 ) , which while not the most flexible among all that could be considered , is certainly more flexible than the one that has been most used in practice for estimating PVE ( LMM ) , and admits certain computational methods that could not be applied in all cases . We consider in detail how to apply this model in practice , and the resulting advantages over LMM and BVSR ( although we also compare with some other existing approaches ) . A key contribution of this paper is to provide new approaches to address two important practical issues: the statistical issue of how to deal with the unknown hyper-parameters , and the computational issue of how to fit the model . Notably , with the computational tools we use here , fitting the model ( 5 ) becomes , for a typical dataset , less computationally intensive than fitting BVSR , as well as providing gains in performance . With this background , we now turn to detailed description of the model , its prior specification and its computation algorithm . In this paper we focus on the simple linear model ( 1 ) with mixture prior ( 5 ) on the effects . However , the computational and statistical methods we use here also apply to a more general model , which we refer to as the Bayesian Sparse Linear Mixed Model ( BSLMM ) , and which includes the model ( 1 ) with ( 5 ) as a special case . The BSLMM consists of a standard linear mixed model , with one random effect term , and with sparsity inducing priors on the regression coefficients: ( 6 ) ( 7 ) ( 8 ) ( 9 ) where is an -vector of random effects with known covariance matrix . In referring to as the “random effects” we are following standard terminology from LMMs . Standard terminology also refers to the coefficients as “fixed effects” , but this phrase has a range of potential meanings [55] and so we avoid it here . Instead we use the term “sparse effects” for these parameters to emphasize the sparsity-inducing prior . It is straightforward to show that when , BSLMM is equivalent to the simple linear model ( 1 ) with mixture prior ( 5 ) on the effects . However , our discussion of prior specification , computational algorithms , and software , all apply for any . When we say that ( 6 ) is equivalent to ( 1 ) with ( 5 ) , this equivalence refers to the implied probability model for given and the hyper-parameters . However , and are not equivalent ( explaining our use of two different symbols ) : in ( 6 ) the random effect captures the combined small effects of all markers , whereas in ( 1 ) these small effects are included in . Since both our applications focus on the relationship between and , and not on interpreting estimates of or , we do not concern ourselves any further with this issue , although it may need consideration in applications where individual estimated genetic effects are of more direct interest ( e . g . genetic association mapping ) . A related issue is the interpretation of the random effect in BSLMM: from the way we have presented the material is most naturally interpreted as representing a polygenic component , specifically the combined effect of a large number of small effects across all measured markers . However , if there are environmental factors that influence the phenotype and are correlated with genotype ( e . g . due to population structure ) , then these would certainly affect estimates of , and consequently also affect estimates of other quantities , including the PVE . In addition , phenotype predictions from BSLMM will include a component due to unmeasured environmental factors that are correlated with measured genotypes . These issues are , of course , not unique to BSLMM – indeed , they apply equally to the LMM; see [56] and the response from [57] for relevant discussion . Finally , given the observation that a mixture of two normals is more flexible than a point-normal , it might seem natural to consider modifying ( 6 ) by making the assumption that comes from a mixture of two normal distributions rather than a point-normal . However , if then this modification is simply equivalent to changing the values of . The BSLMM involves ( hyper- ) parameters , , and . Before considering prior specification for these parameters , we summarize their interpretations as follows: Appropriate values for these parameters will clearly vary for different datasets , so it seems desirable to estimate them from the data . Here we accomplish this in a Bayesian framework by specifying prior distributions for the parameters , and using Markov chain Monte Carlo ( MCMC ) to obtain approximate samples from their posterior distribution given the observed data . Although one could imagine instead using maximum likelihood to estimate the parameters , the Bayesian framework has several advantages here: for example , it allows for incorporation of external information ( e . g . that most genetic markers will , individually , have small effects ) , and it takes into account of uncertainty in parameter estimates when making other inferences ( e . g . phenotype prediction ) . For the mean and the inverse of error variance , , we use the standard conjugate prior distributions: ( 10 ) ( 11 ) where and denote , respectively , shape and rate parameters of a Gamma distribution . Specifically we consider the posterior that arises in the limits , and . These limits correspond to improper priors , but the resulting posteriors are proper , and scale appropriately with shifting or scaling of the phenotype vector [58] . In particular , these priors have the property that conclusions will be unaffected by changing the units of measurement of the phenotype , which seems desirable for a method intended for general application . Prior specification for the remaining hyper-parameters is perhaps more important . Our approach is to extend the prior distributions for BVSR described in [19] . Following [19] we place a uniform prior on : ( 12 ) where is total number of markers being analyzed . The upper and lower limit of this prior were chosen so that ( the expected proportion of markers with non-zero ) ranges from to . A uniform prior on reflects the fact that uncertainty in in a typical GWAS will span orders of magnitude . A common alternative ( see e . g . [18] , [34] ) is a uniform distribution on , but as noted in [19] this puts large weight on large numbers of markers having non-zero ( e . g . it would correspond to placing 50% prior probability to the event that more than half of the markers have non-zero , and correspond to placing 90% prior probability to the event that more than 10% of the markers have non-zero ) . To specify priors for and , we exploit the following idea from [19]: prior specification is easier if we first re-parameterize the model in terms of more interpretable quantities . Specifically we extend ideas from [19] to re-parameterize the model in terms of the ( expected ) proportion of phenotypic variance explained by the sparse effects and by the random effects . To this end , we define PVE ( the total proportion of variance in phenotype explained by the sparse effects and random effects terms together ) and PGE ( the proportion of genetic variance explained by the sparse effects terms ) as functions of , and : ( 13 ) ( 14 ) where the function V ( x ) is defined as ( 15 ) These definitions ensure that both PVE and PGE must lie in the interval . PVE reflects how well one could predict phenotype from the available SNPs if one knew the optimal as well as the random effects ; together with PVE , PGE reflects how well one could predict phenotype using alone . Since PVE and PGE are functions of , whose distributions depend on hyper-parameters , the prior distribution for PVE and PGE depends on the priors assigned to these hyper-parameters . In brief , our aim is to choose priors for the two hyper-parameters and so that the induced priors on both PVE and PGE are roughly uniform on 0 and 1 . ( Other distributions could be chosen if desired , but we consider this uniform distribution one reasonable default . ) However , because the relationship between the distribution of PVE , PGE and the hyper-parameters is not simple , we have to make some approximations . Specifically , we introduce as approximations ( they are ratios of expectations rather than expectations of ratios ) to the expectations of PVE and PGE , respectively: ( 16 ) ( 17 ) where is the average variance of genotypes across markers , and is the mean of diagonal elements in . In other words , and , where and are the th elements of matrices and , respectively . See Text S1 for derivations . Intuitively , the term captures the expected genetic variance contributed by the sparse effects term ( relative to the error variance ) , because is the expected number of causal markers , is the expected effect size variance of each causal marker ( relative to the error variance ) , and is the average variance of marker genotypes . Similarly , the term captures the expected genetic variance contributed by the random effects term ( relative to the error variance ) , because is the expected variance of the random effects ( relative to the error variance ) when the relatedness matrix has unit diagonal elements , while properly scales it when not . The parameter provides a natural bridge between the LMM and BVSR: when BSLMM becomes the LMM , and when BSLMM becomes BVSR . In practice , when the data favors the LMM , the posterior distribution of would mass near 0 , and when the data favors BVSR , the posterior distribution of would mass near 1 . In summary , the above re-parameterizes the model in terms of instead of . Now , instead of specifying prior distributions for , we rather specify prior distributions for . Specifically we use uniform prior distributions for : ( 18 ) ( 19 ) independent of one another and of . Since and approximate PVE and PGE , these prior distributions should lead to reasonably uniform prior distributions for PVE and PGE , which we view as reasonable defaults for general use . ( If one had specific information about PVE and PGE in a given application then this could be incorporated here . ) In contrast it seems much harder to directly specify reasonable default priors for ( although these priors on do of course imply priors for ; see Text S1 ) . Note that we treat and as approximations to PVE and PGE only for prior specification; when estimating PVE and PGE from data we do so directly using their definitions ( 13 ) and ( 14 ) ( see below for details ) . To facilitate computation , we use the widely-used approach from [52] of introducing a vector of binary indicators that indicates whether the corresponding coefficients are non-zero . The point-normal priors for can then be written ( 20 ) ( 21 ) ( 22 ) where denotes the sub-vector of corresponding to the entries ; denotes the sub-vector of corresponding to the other entries , ; and denotes the number of non-zero entries in . We use MCMC to obtain posterior samples of ( ) on the product space , which is given by ( 23 ) The marginal likelihood can be computed analytically integrating out ; see below for further details . We use a Metropolis-Hastings algorithm to draw posterior samples from the above marginal distribution . In particular , we use a rank based proposal distribution for [19] , which focus more of the computational time on examining SNPs with stronger marginal associations . We use the resulting sample from the posterior distribution ( 23 ) to estimate PVE and PGE as follows . For each sampled value of , we sample a corresponding value for from the conditional distribution . We then use each sampled value of to compute a sampled value of PVE and PGE , using equations ( 13 ) and ( 14 ) . We estimate the posterior mean and standard deviation of PVE , PGE , from these samples . The novel part of our algorithm is a new efficient approach to evaluating the likelihood that considerably reduces the overall computational burden of the algorithm . The direct naive approach to evaluating this likelihood involves a matrix inversion and a matrix determinant calculation that scale cubically with the number of individuals , and this cost is incurred every iteration as hyper parameter values change . Consequently , this approach is impractical for typical association studies with large , and ad hoc approximations are commonly used to reduce the burden . For example , both [37] and [41] fix to some pre-estimated value . As we show later , this kind of approximation can reduce the accuracy of predicted phenotypes . Here , we avoid such approximations by exploiting recently developed computational tricks for LMMs [8] , [9] . The key idea is to perform a single eigen-decomposition and use the resulting unitary matrix ( consisting of all eigen vectors ) to transform both phenotypes and genotypes to make the transformed values follow independent normal distributions . By extending these tricks to BSLMM we evaluate the necessary likelihoods much more efficiently . Specifically , after a single operation at the start of the algorithm , the per iteration computational burden is linear in ( the same as BVSR ) , allowing large studies to be analyzed . Full details of the sampling algorithm appear in Text S2 . Software implementing our methods is included in the GEMMA software package , which is freely available at http://stephenslab . uchicago . edu/software . html .
Both the LMM and BVSR have been used to estimate the PVE [19] , [22] . Since the LMM assumes that all SNPs have an effect , while BVSR assumes that only a small proportion of SNPs have an effect , we hypothesize that BVSR will perform better when the true underlying genetic structure is sparse and LMM will perform better when the true genetic structure is highly polygenic . Further , because BSLMM includes both as special cases , we hypothesize that BSLMM will perform well in either scenario . To test these hypotheses , we perform a simulation using real genotypes at about 300 , 000 SNPs in 3 , 925 Australian individuals [22] , and simulate phenotypes under two different scenarios . In Scenario I we simulate a fixed number of causal SNPs ( with ) , with effect sizes coming from a standard normal distribution . These simulations span a range of genetic architectures , from very sparse to highly polygenic . In Scenario II we simulate two groups of causal SNPs , the first group containing a small number of SNPs of moderate effect ( or ) , plus a second larger group of SNPs of small effect representing a “polygenic component” . This scenario might be considered more realistic , containing a mix of small and larger effects . For both scenarios we added normally-distributed errors to phenotype to create datasets with true PVE = 0 . 6 and 0 . 2 ( equation 13 ) . We simulate 20 replicates in each case , and run the algorithms with all SNPs , including the simulated causal variants , so that the true PVE for typed markers is either 0 . 6 or 0 . 2 ( if we excluded the causal variants then the true PVE would be unknown ) . Figure 1A and 1C , show the root of mean square error ( RMSE ) of the PVE estimates obtained by each method , and Figure 1B and 1D summarize the corresponding distributions of PVE estimates . In agreement with our original hypotheses , BVSR performs best ( lowest RMSE ) when the true model is sparse ( e . g . Scenario I , or in Figure 1A , 1C ) . However , it performs very poorly under all the other , more polygenic , models . This is due to a strong downward bias in its PVE estimates ( Figure 1B , 1D ) . Conversely , under the same scenarios , LMM is the least accurate method . This is because the LMM estimates have much larger variance than the other methods under these scenarios ( Figure 1B , 1D ) , although , interestingly , LMM is approximately unbiased even in these settings where its modeling assumptions are badly wrong . As hypothesized , BSLMM is robust across a wider range of settings than the other methods: its performance lies between LMM and BVSR when the true model is sparse , and provides similar accuracy to LMM when not . Of course , in practice , one does not know in advance the correct genetic architecture . This makes the stable performance of BSLMM across a range of settings very appealing . Due to the poor performance of BVSR under highly polygenic models , we would not now recommend it for estimating PVE in general , despite its good performance when its assumptions are met . We also compare the three methods on their ability to predict phenotypes from genotypes , using the same simulations . To measure prediction performance , we use relative prediction gain ( RPG; see Text S1 ) . In brief , RPG is a standardized version of mean square error: RPG = 1 when accuracy is as good as possible given the simulation setup , and RPG = 0 when accuracy is the same as simply predicting everyone to have the mean phenotype value . RPG can be negative if accuracy is even worse than that . Figure 2 compares RPG of different methods for simulations with PVE = 0 . 6 ( results for PVE = 0 . 2 are qualitatively similar , not shown ) . Interestingly , for phenotype prediction , the relative performance of the methods differs from results for PVE estimation . In particular , LMM performs poorly compared with the other two methods in all situations , except for Scenario I with , the Scenario that comes closest to matching the underlying assumptions of LMM . As we expect , BSLMM performs similarly to BVSR in scenarios involving smaller numbers of causal SNPs ( up to in Scenario I ) , and outperforms it in more polygenic scenarios involving large numbers of SNPs of small effect ( e . g . Scenario II ) . This is presumably due to the random effect in BSLMM that captures the polygenic component , or , equivalently , due to the mixture of two normal distributions in BSLMM that better captures the actual distribution of effect sizes . The same qualitative patterns hold when we redo these simulation comparisons excluding the causal SNPs ( Figure S1 ) or use correlation instead of RPG to assess performance ( Figure S2 and Figure S3 ) . For a potential explanation why LMM performs much less well for phenotype prediction than for PVE estimation , we note the difference between these two problems: to accurately estimate PVE it suffices to estimate the “average” effect size reliably , whereas accurate phenotype prediction requires accurate estimates of individual effect sizes . In situations where the normal assumption on effect sizes is poor , LMM tends to considerably underestimate the number of large effects , and overestimate the number of small effects . These factors can cancel one another out in PVE estimation , but both tend to reduce accuracy of phenotype prediction . To obtain further insights into differences between LMM , BVSR and BSLMMM , we apply all three methods to estimate the PVE for five traits in two human GWAS datasets . The first dataset contains height measurements of 3 , 925 Australian individuals with about 300 , 000 typed SNPs . The second dataset contains four blood lipid measurements , including high-density lipoprotein ( HDL ) , low-density lipoprotein ( LDL ) , total cholesterol ( TC ) and triglycerides ( TG ) from 1 , 868 Caucasian individuals with about 550 , 000 SNPs . The narrow sense heritability for height is estimated to be 0 . 8 from twin-studies [22] , [59] . The narrow sense heritabilities for the lipid traits have been estimated , in isolated founder populations , to be 0 . 63 for HDL , 0 . 50 for LDL , 0 . 37 for TG in Hutterites [60] , and 0 . 49 for HDL , 0 . 42 for LDL , 0 . 42 for TC and 0 . 32 for TG in Sardinians [61] . Table 2 shows PVE estimates from the three methods for the five traits . PVE estimates from BVSR are consistently much smaller than those obtained by LMM and BSLMM , which are almost identical for two traits and similar for the others . Estimates of PVE from both LMM and BSLMM explain over 50% of the narrow sense heritability of the five traits , suggesting that a sizable proportion of heritability of these traits can be explained , either directly or indirectly , by available SNPs . These results , with LMM and BSLMM providing similar estimates of PVE , and estimates from BVSR being substantially lower , are consistent with simulation results for a trait with substantial polygenic component . One feature of BSLMM , not possessed by the other two methods , is that it can be used to attempt to quantify the relative contribution of a polygenic component , through estimation of PGE , which is the proportion of total genetic variance explained by “large” effect size SNPs ( or more precisely , by the additional effects of those SNPs above a polygenic background ) . Since the PGE is defined within an inevitably over-simplistic model , specifically that effect sizes come from a mixture of two normal distributions , and also because it will be influenced by unmeasured environmental factors that correlate with genetic factors , we caution against over-interpreting the estimated values . We also note that estimates of PGE for these data ( Table 2 ) are generally not very precise ( high posterior standard deviation ) . Nonetheless , it is interesting that the estimated PGE for height , at 0 . 12 , is lower than for any of the lipid traits ( ranging from 0 . 18 for TG to 0 . 46 for TC ) , and that all these estimates suggest a substantial contribution from small polygenic effects in all five traits . To assess predictive performance on real data , we turn to the Wellcome trust case control consortium ( WTCCC ) 1 study [62] , which have been previously used for assessing risk prediction [63]–[65] . These data include about 14 , 000 cases from seven common diseases and about 3 , 000 shared controls , typed at a total of about 450 , 000 SNPs . The seven common diseases are bipolar disorder ( BD ) , coronary artery disease ( CAD ) , Crohn's disease ( CD ) , hypertension ( HT ) , rheumatoid arthritis ( RA ) , type 1 diabetes ( T1D ) and type 2 diabetes ( T2D ) . We compared the prediction performance of LMM , BVSR and BSLMM for all seven diseases . Following [64] , we randomly split the data for each disease into a training set ( 80% of individuals ) and a test set ( remaining 20% ) , performing 20 such splits for each disease . We estimated parameters from the training set by treating the binary case control labels as quantitative traits , as in [5] , [9] . [This approach can be justified by recognizing the linear model as a first order Taylor approximation to a generalized linear model; we discuss directly modeling binary phenotypes in the Discussion section . ] We assess prediction performance in the test set by area under the curve ( AUC ) [66] . Figure 3 shows AUC for the three methods on all seven diseases . As in our simulations , we find BSLMM performs as well as or better than either of the other two methods for all seven diseases . Indeed , the performance of BSLMM appears to compare favorably with previous methods applied to the same dataset [63]–[65] ( a precise comparison with previous results is difficult , as some studies use slightly different splitting strategies [63] , [65] and some do not perform full cross validation [64] ) . As might be expected from the simulation results , BVSR performs better than LMM in diseases where a small number of relatively strong associations were identified in the original study [62] ( CD , RA and T1D ) and worse in the others . We obtained qualitatively similar results when we measured performance using the Brier score instead of AUC ( Text S3; Figure S4 ) . Finally , we caution that , although BSLMM performs well here relative to other methods , at the present time , for these diseases , its prediction accuracy is unlikely to be of practical use in human clinical settings . In particular , in these simulations the number of cases and controls in the test set is roughly equal , which represents a much easier problem than clinical settings where disease prevalence is generally low even for common diseases ( see [64] for a relevant discussion ) . In addition to the WTCCC dataset , we also assess perdition performance using a mouse dataset [67] , which has been widely used to compare various phenotype prediction methods [37]–[39] . The mouse dataset is substantially smaller than the human data ( , with exact numbers varying slightly depending on the phenotype and the split ) . This makes it computationally feasible to compare with a wider range of other methods . Therefore , we include in our comparisons here five other related approaches , some of which have been proposed previously for phenotype prediction . Specifically we compare with: See Text S1 for further details . Following previous studies that have used these data for prediction [37]–[39] we focused on three quantitative phenotypes: CD8 , MCH and BMI . These phenotypes have very different estimated narrow sense heritabilities: 0 . 89 , 0 . 48 , and 0 . 13 respectively [69] . Table S1 lists estimates of some key quantities for the three traits – including PVE , PGE and – obtained from LMM , BVSR and BSLMM . All three methods agree well on the PVE estimates , suggesting that the data is informative enough to overwhelm differences in prior specification for PVE estimation . Following [37] , [38] , we divide the mouse dataset roughly half and half into a training set and a test set . As the mice come from 85 families , and individuals within a family are more closely related than individuals from different families , we also follow previous studies and use two different splits of the data: the intra-family split mixes all individuals together and randomly divides them into two sets of roughly equal size; the inter-family split randomly divides the 85 families into two sets , where each set contains roughly half of the individuals . We perform 20 replicates for each split of each phenotype . It is important to note that the intra-family split represents an easier setting for phenotype prediction , not only because individuals in the test set are more related genetically to those in the training set , but also because the individuals in the test set often share a similar environment with those in the training set ( specifically , in the intra-family split , many individuals in the test set share a cage with individuals in the training set , but this is not the case in the inter-family split ) . We apply each method using genotypes only , without other covariates . We obtain effect size estimates in the training dataset , and assess prediction performance using these estimates in the test set by root of mean square error ( RMSE ) , where the mean is across individuals in the test set . We contrast the performance of other methods to BSLMM by calculating the RMSE difference , where a positive number indicates worse performance than BSLMM . We perform 20 inter-family splits and 20 intra-family splits for each phenotype . Figure 4 summarizes the prediction accuracy , measured by RMSE , of each method compared against BSLMM . Measuring prediction performance by correlation gives similar results ( Figure S5 ) . For the low-heritability trait BMI , where no measured SNP has a large effect , all methods perform equally poorly . For both the more heritable traits , CD8 and MCH , BSLMM consistently outperformed all other methods , which seem to split into two groups: LMM , LMM-Bayes and Bayesian Lasso perform least well and similarly to one another on average; BVSR , BayesA-Flex , BayesC and BSLMM-EB perform better , and similarly to one another on average . A general trend here is that accuracy tends to increase as model assumptions improve in their ability to capture both larger genetic effects , and the combined “polygenic” contribution of smaller genetic effects ( and possibly also confounding environmental effects correlating with genetic background ) . In particular , the distribution underlying BayesA-Flex , which has a long tail that could capture large effects , performs noticeably better than either the normal or double-exponential distributions for effect sizes underlying LMM and Bayesian Lasso . Comparisons of pairs of closely-related methods yield additional insights into factors that do and do not affect prediction accuracy . The fact that BSLMM performs better than BSLMM-EB illustrates how the approximation used in BSLMM-EB can degrade prediction accuracy , and thus demonstrates the practical benefits of our novel computational approach that avoids this approximation . Similarly , the superior performance of BayesA-Flex over BayesA ( which performed poorly; not shown ) also illustrates the benefits of estimating hyper parameters from the data , rather than fixing them to pre-specified values . The similar performance between BVSR and BayesC , which fit the same model but with different priors , suggests that , for these data , results are relatively robust to prior specification . Presumably , this is because the data are sufficiently informative to overwhelm the differences in prior . The average computational time taken for each method on the Mouse data is shown in Table 3 . Some differences in computational time among methods may reflect implementational issues , including the language environment in which the methods are implemented , rather than fundamental differences between algorithms . In addition , computing times for many methods will be affected by the number of iterations used , and we did not undertake a comprehensive evaluation of how many iterations suffice for each algorithm . Nonetheless , the results indicate that our implementation of BSLMM is competitive in computing speed with the other ( sampling-based ) implementations considered here . In particular , we note that BSLMM is computationally faster than BVSR . This is unexpected , since BSLMM is effectively BVSR plus a random effects term , and the addition of a random effects term usually complicates computation . The explanation for this is that the ( per-iteration ) computational complexity of both BSLMM and BVSR depends , quadratically , on the number of selected SNPs in the sparse effects term ( ) , and this number can be substantially larger with BVSR than with BSLMM , because with BVSR additional SNPs are included to mimic the effect of the random effects term in BSLMM . The size of this effect will vary among datasets , but it can be substantial , particularly in cases where there are a large number of causal SNPs with small effects . To illustrate this , Table 4 compares mean computation time for BSLMM vs BVSR for all datasets used here . For simulated data with a small number of causal SNPs , BSLMM and BVSR have similar computational times . However , in other cases ( e . g . PVE = 0 . 6 , S = 10 , 000 in Scenario I ) BSLMM can be over an order of magnitude faster than BVSR . In a sense , this speed improvement of BSLMM over BVSR is consistent with its hybrid nature: in highly polygenic traits , BSLMM tends to behave similarly to LMM , resulting in a considerable speed gain .
We have presented novel statistical and computational methods for BSLMM , a hybrid approach for polygenic modeling for GWAS data that simultaneously allows for both a small number of individually large genetic effects , and combined effects of many small genetic effects , with the balance between the two being inferred from the data in hand . This hybrid approach is both computationally tractable for moderately large datasets ( our implementation can handle at least 10 , 000 individuals with 500 , 000 SNPs on our moderately-equipped modern desktop computer ) , and is sufficiently flexible to perform well in a wide range of settings . In particular , depending on the genetic architecture , BSLMM is either as accurate , or more accurate , than the widely-used LMM for estimating PVE of quantitative traits . And for phenotype prediction BSLMM consistently outperformed a range of other approaches on the examples we considered here . By generalizing two widely-used models , and including both as special cases , BSLMM should have many applications beyond polygenic modeling . Indeed , despite its increased computational burden , we believe that BSLMM represents an attractive alternative to the widely-used LASSO [70] for general regression-based prediction problems . Although it was not our focus here , BSLMM can be easily modified to analyze binary phenotypes , including for example , a typical human case-control GWAS . For PVE estimation , one can directly apply BSLMM , treating the 1/0 case-control status as a quantitative outcome , and then apply a correction factor derived by [24] to transform this estimated PVE on the “observed scale” to an estimated PVE on a latent liability scale . This correction , for which we supply an alternative derivation in Text S3 , corrects for both ascertainment and the binary nature of case-control data . For phenotype prediction , one can again directly apply BSLMM , treating the 1/0 case-control status as a quantitative outcome , as we do here for the WTCCC dataset , and interpret the resulting phenotype predictions as the estimated probability of being a case . Although in principle one might hope to improve on this by modifying BSLMM to directly model the binary outcomes , using a probit link for example , we have implemented this probit approach and found that not only is it substantially more computationally expensive ( quadratic in instead of linear in ) , but it performed slightly worse than treating the binary outcomes as quantitative , at least in experiments based on the mouse phenotypes considered here ( Text S3 and Figure S6 ) , which is consistent with previous findings in quantitative trait loci mapping [71] . This may partly reflect inaccuracies introduced by the known greater computational burden and corresponding mixing issues with probit models ( e . g . [72] ) which are magnified here by the polygenic nature of the traits , and partly reflect robustness of linear models to model misspecification . The computational innovations we introduce here , building on work by [8] , [9] , make BSLMM considerably more tractable than it would otherwise be . Nonetheless , the computational burden , as with other posterior sampling based methods , remains heavy , both due to memory requirements ( e . g . to store all genotypes ) and CPU time ( e . g . for the large number of sampling iterations required for reasonable convergence ) . Although more substantial computational resources will somewhat increase the size of data that can be tackled , further methodological innovation will likely be required to apply BSLMM to the very large datasets that are currently being collected . In addition to providing a specific implementation that allows BSLMM to be fitted to moderately large datasets , we hope that our work also helps highlight some more general principles for improving polygenic modeling methodology . These include: One question to which we do not know the answer is how often the mixture of two normal distributions underlying BSLMM will be sufficiently flexible to capture the actual effect size distribution , and to what extent more flexible distributional assumptions ( e . g . a mixture of more than two normals , or a mixture of distributions with degrees of freedom estimated from the data ) will produce meaningful gains in performance . It seems likely that , at least in some cases , use of a more flexible distribution will improve performance , and would therefore be preferable if it could be accomplished with reasonable computational expense . Unfortunately some of the tricks we use to accomplish computational gains here may be less effective , or difficult to apply , for more flexible distributions . In particular , the tricks we use from [8] and [9] may be difficult to extend to allow for mixtures with more than two components . In addition , for some choices of effect size distribution , one might have to perform MCMC sampling on the effect sizes directly , rather than sampling , integrating out analytically , as we do here . It is unclear whether this will necessarily result in a loss of computational efficiency: sampling reduces computational expense per update at the cost of increasing the number of updates necessary ( sampling by integrating over analytically ensures faster mixing and convergence [74] , [75] ) . Because of these issues , it is difficult to predict which effect size distributions will ultimately provide the best balance between modeling accuracy and computational burden . Nonetheless , compared with currently available alternatives , we believe that BSLMM strikes an appealing balance between flexibility , performance , and computational tractability . | The goal of polygenic modeling is to better understand the relationship between genetic variation and variation in observed characteristics , including variation in quantitative traits ( e . g . cholesterol level in humans , milk production in cattle ) and disease susceptibility . Improvements in polygenic modeling will help improve our understanding of this relationship and could ultimately lead to , for example , changes in clinical practice in humans or better breeding/mating strategies in agricultural programs . Polygenic models present important challenges , both at the modeling/statistical level ( what modeling assumptions produce the best results ) and at the computational level ( how should these models be effectively fit to data ) . We develop novel approaches to help tackle both these challenges , and we demonstrate the gains in accuracy that result in both simulated and real data examples . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"genome-wide",
"association",
"studies",
"animal",
"genetics",
"statistics",
"quantitative",
"traits",
"mathematics",
"biostatistics",
"biology",
"heredity",
"genetic",
"association",
"studies",
"genetics",
"human",
"genetics",
"genetics",
"of",
"disease",
"statistical",
"methods",
"genetics",
"and",
"genomics",
"complex",
"traits"
]
| 2013 | Polygenic Modeling with Bayesian Sparse Linear Mixed Models |
In plants and fungi , small RNAs silence gene expression in the nucleus by establishing repressive chromatin states . The role of endogenous small RNAs in metazoan nuclei is largely unknown . Here we show that endogenous small interfering RNAs ( endo-siRNAs ) direct Histone H3 Lysine 9 methylation ( H3K9me ) in Caenorhabditis elegans . In addition , we report the identification and characterization of nuclear RNAi defective ( nrde ) -1 and nrde-4 . Endo-siRNA–driven H3K9me requires the nuclear RNAi pathway including the Argonaute ( Ago ) NRDE-3 , the conserved nuclear RNAi factor NRDE-2 , as well as NRDE-1 and NRDE-4 . Small RNAs direct NRDE-1 to associate with the pre-mRNA and chromatin of genes , which have been targeted by RNAi . NRDE-3 and NRDE-2 are required for the association of NRDE-1 with pre-mRNA and chromatin . NRDE-4 is required for NRDE-1/chromatin association , but not NRDE-1/pre-mRNA association . These data establish that NRDE-1 is a novel pre-mRNA and chromatin-associating factor that links small RNAs to H3K9 methylation . In addition , these results demonstrate that endo-siRNAs direct chromatin modifications via the Nrde pathway in C . elegans .
Small regulatory RNAs can silence gene expression in the nucleus by establishing repressive chromatin states . This process , termed Transcriptional Gene Silencing ( TGS ) , was first observed in plants , where small RNAs direct DNA methylation and histone modifications ( reviewed in [1] ) . In addition , the fission yeast , Schizosaccharomyces pombe has been an important model in defining the role of small RNAs in heterochromatin formation . In S . pombe , small RNAs direct the formation of heterochromatin primarily at repetitive DNA elements surrounding centromeres [2] , [3] . At these repetitive elements , nascent RNAs , transcribed by RNA Polymerase II ( RNAP II ) , serve as platforms for the assembly of RNAi machinery . For instance , the RNA Induced Transcriptional Silencing ( RITS ) complex , composed of the Argonaute Ago1 , the chromodomain protein Chp1 , and the glycine and tryptophan ( GW ) -motif-containing protein Tas3 , is guided to nascent transcripts by Argonaute and centromeric siRNAs [4] . The RITS complex recruits chromatin-modifying machinery , such as the histone methyltransferase Clr4 , to genomic sites of nuclear RNAi [5] , [6] . Clr4 catalyzes the methylation of Histone H3 on Lysine 9 ( H3K9me ) [7] . H3K9me is a conserved molecular mark of heterochromatin [8] . Thus , in plants and S . pombe , small RNAs play a central role in regulating chromatin dynamics . The role of TGS and heterochromatin formation in metazoan silencing processes is less clear [3] . Experimentally provided small RNAs can elicit transcriptional silencing and induce heterochromatic marks in metazoans . In mammalian cells , experimentally provided siRNAs directed against promoter regions can lead to transcriptional silencing and induce heterochromatic marks [9]–[12] . Paradoxically , experimentally provided small RNAs can also enhance transcription and decrease H3K9me marks [13] , [14] . In C . elegans , experimentally provided siRNAs are bound by the Ago NRDE-3 in the cytoplasm , and escorted into the nucleus [15] . NRDE-3/siRNA ribonucleoprotein complexes bind nascent transcripts and recruit the conserved nuclear RNAi factor NRDE-2 . The Nrde pathway inhibits RNA Polymerase ( RNAP ) II during the elongation phase of transcription , and directs the deposition of H3K9me marks at genomic sites that exhibit homology to experimentally introduced siRNAs [16] . How and if endogenously expressed small regulatory RNAs silence gene expression in metazoan nuclei is unclear . Dicer deficient mouse embryonic stem cells express high levels of centromeric repeat RNAs and exhibit altered heterochromatic marks at centromeres [17] . In Drosophila , heterochromatic marks , including H3K9me and HP1 , are mislocalized in flies lacking components of the RNAi machinery such as Piwi , Aubergine , and Homeless [18] . In addition , the Drosophila Ago-like protein PIWI binds small RNAs , termed piRNAs , and associates with chromatin [19] . Loss of piwi has variable effects on chromatin states at genomic sites homologous to piRNAs [20]–[24] . Finally , in C . elegans , animals lacking two RNAi-related factors: the RNA-dependent RNA Polymerase EGO-1 , or the Ago CSR-1 , exhibit large-scale changes in chromosomal H3K9me patterns during germline development [25] , [26] . Thus , endogenous small regulatory RNAs have been implicated in chromatin regulation in metazoans . However , a direct link has yet to be established , and the molecular mechanisms by which this might occur are unknown . Here we show that the endogenous small RNAs , termed endo-siRNAs , direct H3K9me marks at discrete genomic loci in C . elegans . Small RNA-directed H3K9 methylation requires the Nrde pathway and results in the inhibition of transcription from these loci . In addition , we identify two novel nuclear RNAi factors termed NRDE-1 and NRDE-4 , and show that these factors are required for small RNA-directed H3K9 methylation . Finally , we show that small RNAs direct NRDE-1 to associate with pre-mRNA and chromatin of genes , which have been targeted by RNAi . Thus , the Nrde pathway links endogenously expressed small regulatory RNAs to the regulation of transcription and chromatin dynamics in C . elegans .
We previously reported a forward genetic screen that identified two genes ( termed nrde-2 and nrde-3 ) required for nuclear RNAi [15] , [16] . The mechanism ( s ) by which NRDE-2/3 silence nascent transcripts and inhibit RNAP II transcription are unknown . To understand this mechanism we continued screening for nuclear RNAi factors . >80% of the nrde alleles identified in our original genetic screen were alleles of nrde-3 ( Table S1 , [15] ) . To maximize our chances of identifying novel nuclear RNAi factors , we performed our modified screen in animals harboring ectopic copies of nrde-3 ( nrde-3::gfp ) , which was integrated into the genome on chromosome V ( Figure S1 ) . eri-1 encodes an exonuclease that negatively regulates RNAi [27] . Our original screen was conducted in an eri-1 ( − ) genetic background . Our modified screen was conducted in eri-1 ( + ) animals ( Figure S1 ) . Our modified screen identified twenty-three alleles of nrde-2 , nineteen alleles of nrde-1 , nine alleles of the RNA-dependent RNA Polymerase ( RdRP ) rrf-1 , four alleles of nrde-4 , and one additional nrde allele , which complements the known nrde genes , but has not yet been assigned a nrde gene designation ( Figure 1a , and Table S1 ) . Here we report the identification and characterization of nrde-1 and nrde-4 . We first focused our attention on characterizing the role of nrde-1 in nuclear RNAi . Three lines of evidence indicate that nrde-1 functions with nrde-2 and nrde-3 to silence nuclear-localized RNAs during nuclear RNAi . First , NRDE-1 , like NRDE-2/3 , is required for RNAi-based silencing of nuclear-localized RNAs . For instance , the lir-1 and lin-26 genes are expressed in an operon; these genes are co-transcribed as a polycistronic pre-mRNA , which is spliced into distinct mRNAs in the nucleus before export to the cytoplasm [28] , [29] . lir-1 ( − ) mutant animals are viable , whereas lin-26 ( − ) mutant animals exhibit a lethal phenotype [30] . RNAi targeting lir-1 induces a lethal phenotype , indicating that lir-1 RNAi silences the nuclear-localized lir-1/lin-26 RNA [15] , [30] . nrde-2 ( − ) and nrde-3 ( − ) animals are viable following lir-1 RNAi , indicating that NRDE-2 and NRDE-3 are required for lir-1-mediated silencing of the lir-1/lin-26 RNA [15] , [16] . nrde-1 mutant animals were also viable when exposed to lir-1 RNAi , indicating that , like NRDE-2 and NRDE-3 , NRDE-1 is required for silencing of the lir-1/lin-26 RNA ( Table 1 ) . Similarly , NRDE-1/2/3 are required to silence the nuclear-localized lin-15b/lin-15a RNA . lin-15b and lin-15a genes are encoded in an operon . Mutations in lin-15b or lin-15a alone produce no obvious phenotype , but animals harboring mutations in both lin-15b and lin-15a exhibit a Multi-vulva ( Muv ) phenotype [31] . RNAi targeting lin-15b induces a Muv phenotype , indicating that lin-15b RNAi silences the nuclear-localized lin-15b/lin-15a RNA [15] , [16] . nrde-1/2/3 mutant animals do not exhibit a Muv phenotype in response to lin-15b RNAi , indicating that NRDE-1/2/3 are required for silencing the lin-15b/lin-15a RNA ( Table 1 ) . Second , nuclear-localized siRNAs direct a NRDE-2/3 dependent inhibition of RNAP II during the elongation phase of transcription [16] . For instance , lin-15b RNAi inhibits RNAP II transcription 3′ to the site of lin-15b RNAi ( Figure 1b ) . RNAi-mediated inhibition of RNAP II transcription is dependent upon NRDE-2 and NRDE-3 [15] , [16] . nrde-1 was also required to link small RNAs to RNAP II inhibition; in nrde-1 mutant animals , lin-15b RNAi did not result in transcription inhibition ( Figure 1b ) . Thus , like nrde-2/3 , a wild-type copy of the nrde-1 gene is required for RNAi to inhibit transcription elongation . Third , we conducted a genetic analysis using double mutant combinations of the Nrde factors . This analysis indicated that nrde-1 functions in a genetic pathway with nrde-2 and nrde-3 ( Figure S2 ) . Taken together , these data argue that nrde-1 is a component of the Nrde silencing pathway . To determine the molecular identity of nrde-1 , we used a single nucleotide polymorphism ( SNP ) -based mapping approach [32] . We mapped nrde-1 to a 0 . 86cM interval on Chromosome III that contained 42 genes . The open reading frame ( ORF ) c14b1 . 6 lies within this mapping interval . Sequencing of c14b1 . 6 from three independent nrde-1 alleles revealed three mutations in c14b1 . 6 ( Figure 1c ) . Two of these alleles encode premature stop codons , and therefore likely reveal the null phenotype of nrde-1 . Expression of a wild-type copy of c14b1 . 6 was sufficient to rescue the Nrde phenotype associated with nrde-1 ( see below ) . We conclude that c14b1 . 6 corresponds to nrde-1 . Analysis of nrde-1 expressed sequence tags ( ESTs ) indicated that nrde-1 encodes a protein containing 793 amino acids [33] . Database searches revealed that nrde-1 is conserved in other nematode species , but these searches failed to detect any obvious orthologues of nrde-1 outside nematodes . In addition , these database searches did not identify any obvious protein domains within NRDE-1 . We assessed the sub-cellular distribution of NRDE-1 . We constructed a NRDE-1 and Green Fluorescent Protein fusion protein ( NRDE-1::GFP ) , which encodes GFP 5′ to a full length copy of nrde-1 . We observed fluorescence in nuclei of NRDE-1::GFP expressing animals ( Figure 1d ) . NRDE-1::GFP rescued Nrde phenotypes associated with nrde-1 ( − ) animals ( Figure 1e ) , suggesting that the NRDE-1::GFP expression pattern reflects the expression pattern of endogenous NRDE-1 . We conclude that NRDE-1 is a nuclear localized protein . The Ago protein NRDE-3 binds siRNAs in the cytoplasm and transports these siRNAs to the nucleus to facilitate nuclear RNAi [15] . NRDE-3 can bind small RNAs generated from exogenously provided dsRNAs , which are termed exogenous ( exo ) siRNAs . NRDE-3 also associates with endogenously expressed small RNAs termed endo-siRNAs [15] . NRDE-3 shuttles siRNAs from the cytoplasm to the nucleus; NRDE-3 localizes to the nucleus when bound to either endo or exo-siRNAs , and localizes to the cytoplasm in the absence of these siRNAs [15] . We asked if NRDE-1 was required for NRDE-3/siRNA shuttling . NRDE-3 retained the ability to bind endo-siRNAs in nrde-1 ( − ) animals , indicating that nrde-1 is not required for loading NRDE-3 with siRNAs ( Figure 2a ) . In addition , NRDE-3 remained localized in the nucleus in nrde-1 ( − ) animals , indicating that NRDE-1 activity was not required for NRDE-3 shuttling ( Figure 2b ) . These data suggest that NRDE-1 functions downstream of NRDE-3 siRNAs transport . Following exposure to dsRNA , NRDE-3 associates with un-spliced RNAs ( pre-mRNA ) that exhibit sequence homology to the trigger dsRNA . The association of NRDE-3 with pre-mRNA is dependent upon the ability of NRDE-3 to localize to the nucleus , the ability of NRDE-3 to bind siRNAs , and is restricted to those pre-mRNAs that have been targeted by RNAi [15] . Thus , siRNAs direct NRDE-3 to associate with pre-mRNAs . To test the idea that NRDE-1 functions downstream of NRDE-3 shuttling , we asked if NRDE-1 was required for the association of NRDE-3 with pre-mRNA in response to RNAi . We performed NRDE-3 RNA Immuno-Precipitation ( RIP ) and found that , in response to lin-15b RNAi , NRDE-3 retained the ability to bind lin-15b pre-mRNA in nrde-1 ( − ) animals , indicating that NRDE-1 functions downstream of NRDE-3/pre-mRNA association during nuclear silencing events ( Figure 2c ) . Our genetic screen identified nine alleles of the gene rrf-1 ( Figure 1a , Table S1 ) . rrf-1 encodes one of four C . elegans RNA-dependent RNA Polymerases ( RdRPs ) [34] . We sought to position rrf-1 in the nuclear RNAi pathway . In animals lacking RRF-1 , NRDE-3 binds fewer small RNAs , suggesting that RRF-1 may generate the small RNAs bound by NRDE-3 [15] . Consistent with this idea , in rrf-1 ( − ) animals , NRDE-3/lin-15b pre-mRNA association was reduced relative to rrf-1 ( + ) animals , indicating that RRF-1 acts upstream of NRDE-3/pre-mRNA association during nuclear silencing ( Figure 2c ) . Taken together , these data indicate that our genetic screen is identifying components of the Nrde pathway that function both upstream and downstream of NRDE-3-mediated siRNA transport . We asked if NRDE-1 was recruited to pre-mRNA following RNAi . We performed NRDE-1 RNA Immuno-Precipitation ( RIP ) experiments in animals exposed to lin-15b dsRNA . lin-15b RNAi induced a ∼30–70× enrichment in un-spliced lin-15b RNA that co-precipitated with FLAG::NRDE-1 ( Figure 2d , 2e ) . The dpy-28 gene encodes a subunit of the C . elegans dosage compensation complex [35] . We tested if dpy-28 dsRNA would induce NRDE-1-dpy-28 pre-mRNA association . Following dpy-28 RNAi , NRDE-1 associated with dpy-28 pre-mRNA ( Figure 2d ) . Finally , dpy-28 or lin-15b RNAi did not result in enrichment of NRDE-1 with lin-15b or dpy-28 pre-mRNA , respectively , indicating that the association of NRDE-1 with pre-mRNA ( induced by RNAi ) is sequence specific ( Figure 2d ) . We were concerned that NRDE-1 might associate with pre-mRNA targets , in vitro , during sample preparation . To address this issue we pooled extracts from animals exposed to lin-15b dsRNA , and extracts from NRDE-1::GFP expressing animals not exposed to lin-15b , dsRNA and failed to detect an association of NRDE-1 with lin-15b pre-mRNA , indicating that NRDE-1/pre-mRNA interactions likely occurs in vivo ( Figure 2e ) . Taken together , these data show that NRDE-1 associates with pre-mRNAs that have been targeted by RNAi . NRDE-1 co-precipitating pre-mRNA was enriched for RNA sequences encoded at , or near , the site of RNAi- relative to sequences encoded 5′ or 3′ to the site of RNAi ( Figure 2e ) . We have previously shown that NRDE factors fail to associate with pre-mRNA sequences encoded 3′ to the site of RNAi due to RNAi-mediated inhibition of transcription elongation [16] . We investigated the apparent lack of pre-mRNA sequences encoded 5′ to the site of RNAi and found that , while the NRDE factors fail to associate with un-spliced RNA 5′ to the site of RNAi , the Nrdes do associate with spliced RNA 5′ to the site of RNAi ( Figure S3 ) . Splicing is thought to occur co-transcriptionally [29] . Therefore , the apparent lack of NRDE-1/pre-mRNA association 5′ to sites of RNAi may be due to co-transcriptional splicing of nascent transcripts . We investigated the genetic requirements of NRDE-1/pre-mRNA association . In nrde-2 ( − ) animals , RNAi failed to induce an association of NRDE-1 with pre-mRNA ( Figure 2e ) . In addition , ∼10× less lin-15b pre-mRNA co-precipitated with NRDE-1 in nrde-3 ( − ) animals than in nrde-3 ( + ) animals ( Figure 2e ) . We conclude that the recruitment of NRDE-1 to pre-mRNAs by small RNAs requires NRDE-2 and is largely dependent upon NRDE-3 ( see discussion ) . In plants and S . pombe small RNAs direct the methylation of Histone 3 Lysine 9 ( H3K9me ) . Histone methylation results from small RNA-mediated recruitment of histone methyltransferase enzymes to genomic sites exhibiting sequence homology to small RNAs [5] . RNAi also directs H3K9 methylation in C . elegans [16] . nrde-2 is required for RNAi-mediated H3K9 methylation in C . elegans [16] . The mechanism by which the C . elegans Nrde pathway mediates H3K9 methylation is unknown . We conducted H3K9me Chromatin Immuno Precipitation ( ChIP ) to determine if NRDE-1 was required to link small RNAs to H3K9 methylation . lin-15b RNAi induced a ∼30× increase in H3K9me marks at the lin-15 locus ( Figure 3 ) . In nrde-1 ( − ) animals , however , lin-15b RNAi had no effect on the methylation status of chromatin at the lin-15b gene ( Figure 3 ) . We conclude that NRDE-1 is required to link small RNAs to H3K9 methylation at a genomic site that has been targeted by RNAi . We asked if the NRDE factors themselves might become associated with chromatin in response to RNAi . In order to address this question , we performed NRDE-1/2/3 ChIP experiments before or after exposure of animals to dsRNA . In response to lin-15b RNAi , we did not detect any significant increase in the association of NRDE-2 or NRDE-3 with chromatin at the lin-15b gene ( Figure 4a ) . Interestingly , NRDE-1 precipitated ∼6× more lin-15b DNA following lin-15b RNAi ( Figure 4a ) . In nrde-2 ( − ) and nrde-3 ( − ) animals , lin-15b RNAi failed to trigger an increase in lin-15b DNA that co-precipitated with NRDE-1 ( Figure 4b ) . We conclude that NRDE-1 is able to IP chromatin of a gene that has been targeted by RNAi , and that the association of NRDE-1 with chromatin requires NRDE-2 and NRDE-3 . It is possible that the ability of NRDE-1 to co-precipitate with chromatin may occur as an indirect consequence of NRDE-1/pre-mRNA interactions . To address this issue , we turned our attention to nrde-4 . We mapped and cloned nrde-4 ( Figure S4 ) . nrde-4 is predicted to encode a protein containing 788 amino acids [33] . Database searches revealed that nrde-4 is conserved within other nematode species , but not in other species . nrde-4 encodes a predicted bipartite nuclear localization signal ( NLS ) and no other obvious protein domains ( Figure S4 ) . NRDE-4 is required for silencing nuclear localized RNAs ( Table 1 ) , for linking small RNAs to the inhibition of transcription ( Figure 1b ) , and for linking small RNAs to H3K9 methylation ( Figure 3 ) . Interestingly , the recruitment of NRDE-1 ( and NRDE-2/3 ) to pre-mRNA was largely unaffected in animals lacking NRDE-4 ( Figure 2c , 2e , Figure S5 ) . NRDE-4 was , however , required for recruitment of NRDE-1 to chromatin in response to RNAi ( Figure 4b ) . These data indicate that NRDE-4 functions downstream of NRDE-1/2/3/pre-mRNA interactions during nuclear RNAi . These data also demonstrate that the ability of NRDE-1 to associate with chromatin is dissociable from the ability of NRDE-1 to associate with pre-mRNA , supporting the idea that NRDE-1 associates with chromatin at genomic sites targeted by RNAi . C . elegans express at least three types of endogenous small RNAs; the microRNAs , the piRNAs , and the endo-siRNAs . A sub-set of the endo-siRNAs requires ERI-1 for their expression [36] , [37] . NRDE-3 associates with the ERI-1-dependent endo-siRNAs , but not the other classes of endogenous small RNAs [15] , [37] . Five lines of evidence cumulatively argue that ERI-1 dependent endo-siRNAs are able to direct the deposition of H3K9me marks in C . elegans . First , in animals that fail to express endo-siRNAs H3K9me marks are depleted at genomic regions exhibiting sequence complementarity to endo-siRNAs . For instance , e01g4 . 5 siRNAs are amongst the most abundant endo-siRNAs expressed in C . elegans [36] . eri-1 ( − ) animals do not express e01g4 . 5 endo-siRNAs ( [36] , [37] and Figure 5a ) . We conducted H3K9me ChIP and detected a ∼6× depletion of H3K9me marks at the e01g4 . 5 gene in eri-1 ( − ) animals ( Figure 5b ) . The changes in H3K9me marks were restricted to genomic regions exhibiting homology to endo-siRNAs; surrounding genomic regions , which are not homologous to known small regulatory RNAs , did not exhibited altered H3K9me marks ( Figure 5b ) . Second , in nrde-1/2/3/4 mutant animals we observed a similar localized depletion of H3K9me marks at e01g4 . 5 ( Figure 5c and Figure S6 ) . Third , the e01g4 . 5 pre-mRNA was over-expressed 2–5× in eri-1 and nrde-1/2/3/4 mutant animals ( [15] , Figure S7 , and data not shown ) . Fourth , we performed NRDE-1 RIP and quantified the amount of e01g4 . 5 pre-mRNA that co-precipitated with NRDE-1 . We conducted this experiment in both nrde-2 ( + ) and nrde-2 ( − ) animals as NRDE-2 is required for NRDE-1 recruitment to pre-mRNAs in response to feeding RNAi . We found that NRDE-1 associated with ∼5× more e01g4 . 5 pre-mRNA in nrde-2 ( + ) animals than in nrde-2 ( − ) animals ( Figure 5d ) . These data suggest that NRDE-1 can associate with pre-mRNAs that are homologous to endo-siRNAs , and that this process depends upon components of the Nrde pathway . Five , we detected a subtle and complex , yet reproducible , increase in transcription at the e01g4 . 5 gene in eri-1 and nrde-1/2/4 mutant animals ( Figure 5e and Figure S8 ) . Taken together , these data indicate that e01g4 . 5 endo-siRNAs are able to direct chromatin modification in C . elegans and that this process requires the Nrde pathway . Lastly , we investigated the generality of small RNA-mediated chromatin regulation in C . elegans . We queried seven additional genomic sites that exhibit sequence homology to eri-1-dependent endo-siRNAs . At four of these loci H3K9me marks were depleted in eri-1 and nrde-1/2/3/4 animals ( Figure 5f ) . At three of these loci , no significant differences in H3K9me marks were observed . We conclude that Nrde-dependent endogenous small RNA-mediated chromatin modification occurs at multiple loci in C . elegans .
In S . pombe silencing factors assemble upon nascent transcripts during nuclear RNAi [3] . Here , we present evidence that nascent transcripts serve a similar role in C . elegans . The Ago NRDE-3 is guided to nascent transcripts via base pairing between NRDE-3 bound siRNAs and nascent transcripts [15] . In nrde-1 mutant animals , NRDE-3 can still associate with the target pre-mRNA , but nuclear silencing does not occur ( Figure 2c ) . Thus , NRDE-3 bound siRNAs provide the information of where to silence , but additional downstream factors , such as NRDE-1 , are required for silencing to occur . NRDE-3 is required for recruitment of NRDE-2 to pre-mRNA in response to RNAi [16] . NRDE-3 and NRDE-2 are required for the recruitment of NRDE-1 to pre-mRNAs in response to RNAi ( Figure 2e ) . Thus , we propose that the NRDE factors assemble in a hierarchical manner on pre-mRNA; NRDE-3 identifies pre-mRNAs , and in association with NRDE-2 , recruits NRDE-1 to pre-mRNAs that have been targeted by RNAi ( Figure 6 ) . What is the role of NRDE-4 in nuclear RNAi ? Our preliminary investigation has shown that: NRDE-4 is required to link small RNAs to transcription and chromatin regulation . Interestingly , we find that NRDE-4 is not required for small RNA-directed NRDE-1/pre-mRNA association , but is , required for the recruitment of NRDE-1 to chromatin . Therefore , it seems reasonable to speculate that one role of NRDE-4 during nuclear RNAi may be to load/stabilize NRDE-1 on chromatin , following the recruitment of NRDE-1 to pre-mRNAs by NRDE-2/3 ( Figure 6 ) . What is the role of NRDE-1 in nuclear RNAi ? In response to RNAi , NRDE-1 co-precipitates with both pre-mRNAs and chromatin . We did not detect an association of NRDE-3 or NRDE-2 with chromatin despite the fact that NRDE-2/3 are able , like NRDE-1 , to associate with pre-mRNA in response to RNAi ( Figure 4a ) . These data hint that NRDE-1 may possess a chromatin associating property not exhibited by NRDE-2/3 . We considered the possibility that NRDE-1 might IP chromatin indirectly via pre-mRNA/RNAP II intermediates . However , we found that in nrde-4 ( − ) animals , NRDE-1 is recruited to pre-mRNAs by RNAi , but does not become associated with chromatin ( Figure 2e and Figure 4b ) . These data demonstrate that the RNA and chromatin associating properties of NRDE-1 can be separated . Additionally , we find that NRDE-1 association with RNA occurs predominantly 5′ to the site of RNAi , whereas NRDE-1 association with chromatin occurs predominantly 3′ to the site of RNAi ( Figure 2e and Figure 4a ) . Taken together , these data argue that NRDE-1 associates with chromatin in response to RNAi , and that NRDE-1 interacts with pre-mRNAs first and chromatin second during nuclear silencing processes ( Figure 6 ) . The question then becomes; what is the role of NRDE-1 at chromatin ? Here we show that small RNAs promote H3K9 methylation in C . elegans . We show that experimentally introduced small RNAs are sufficient to direct H3K9me marks at genomic sites targeted by RNAi . We also show that small RNAs are necessary to establish H3K9me marks; in animals that fail to express endogenous siRNAs , H3K9me marks are depleted at genomic sites homologous to endo-siRNAs . In S . Pombe , the RNAi machinery directs H3K9 methylation at pericentromeric repeats via recruitment of the H3K9 methyltransferase Clr4 to pre-mRNAs exhibiting homology to pericentromeric siRNAs [5] . Interestingly , fungi lacking H3K9me , due to loss of Clr4 , fail to express abundant pericentromeric siRNAs [38] . Thus , H3K9me and the RNAi machinery are thought to comprise a self-reinforcing loop that facilitates heterochromatin formation at pericentromeric regions in S . pombe [39] , [40] . We find that , in C . elegans , RNAi directs both H3K9 methylation and the association of NRDE-1 with chromatin . These data hint that C . elegans may employ a similar strategy as S . pombe for establishing heterochromatin; e . g . RNAi promotes H3K9 methylation and H3K9 methylation may help recruit components of the RNAi machinery , such as NRDE-1 , to chromatin . In order to test this model , the C . elegans methyltransferase ( s ) responsible for depositing H3K9me marks in response to RNAi will need to be identified . We find that H3K9me marks become distributed throughout a gene that has been targeted by RNAi ( Figure 3 ) . These data raise several interesting questions . First , how do H3K9me marks spread from the site of RNAi , and how are these marks prevented from spreading into adjacent genes ? A simple model posits that the deposition of H3K9me marks ( directed by small RNAs ) is coupled to transcription in C . elegans . In other words , the act of transcription may alter chromatin in such a way as to permit ( and limit ) H3K9me spreading . Another question that arises is; what is the connection between H3K9 methylation and RNAP II transcription in C . elegans ? We show that both endo-siRNAs and exo-siRNAs direct H3K9 methylation , which correlates with decreases in transcription . These data are consistent with the established repressive role of H3K9 methylation on transcription [8] . We find that RNAi-directed H3K9me marks peak 3′ to sites of RNAi ( Figure 3 ) . In addition , we find that NRDE-1 associates with chromatin predominantly 3′ to sites of RNAi ( Figure 4a ) , and RNAP II transcription is inhibited by RNAi predominantly 3′ to the site of RNAi ( Figure 1b ) . Therefore , H3K9me marks and NRDE-1 may contribute to the inhibition of RNAP II elongation by small RNAs ( Figure 6 ) . It should be noted , however , that while H3K9me marks peak 3′ to sites of RNAi , we observe H3K9 methylation throughout genes targeted by RNAi , hinting that H3K9me marks alone may not be sufficient to inhibit RNAP II transcription in C . elegans ( Figure 3 ) . In S . pombe small RNAs primarily target repetitive genomic elements . RNAi-directed heterochromatization at pericentromeric repeats permits efficient segregation of chromosomes during meiosis [41] . In plants , small RNAs silence genomic regions enriched in transposons , pericentromeric regions , and rRNA genes [1] . Here we show that ERI-1-dependent endo-siRNAs direct the establishment of heterochromatic marks on chromatin . The biological role ( s ) of this small RNA-mediated chromatin regulation in C . elegans is unknown . The ERI-1-dependent endo-siRNAs are anti-sense to several hundred cellular mRNAs [36] . In general , these mRNAs appear to be poorly conserved and repetitive , hinting that these mRNAs may represent the products of dead and dying genes [42] . The purpose of nuclear RNAi may be to prevent expression of these dysfunctional genes . Alternatively , these mRNAs may simply serve as templates for the creation of small RNAs , which , in turn , regulate chromatin dynamics . There are 26 Agos encoded in the worm genome , in addition to nrde-3 [43] . We have detected pleiotropic fertility defects exhibited by nrde-1/2/4 ( − ) , but not nrde-3 ( − ) , animals , hinting that other Ago proteins and , perhaps , other types of small RNAs , may engage NRDE-1/2/4 to promote H3K9 methylation during development ( Figure S9 ) . In support of this idea , we find that the recruitment of NRDE-1 to pre-mRNAs and chromatin , in response to RNAi , is not completely abolished in animals harboring null alleles of nrde-3 ( Figure 2e ) . These data support the idea that other Ago proteins may engage the Nrde pathway to elicit nuclear silencing and chromatin regulation in C . elegans ( Figure 2e , Figure 6 ) . The identification of these Ago factors and their small RNA partners will be important for unraveling the cellular connections that exist between endogenous small RNAs and chromatin dynamics in metazoans .
N2 , ( YY160 ) nrde-1 ( gg088 ) , ( YY186 ) nrde-2 ( gg091 ) , ( YY158 ) nrde-3 ( gg066 ) , ( YY453 ) nrde-4 ( gg129 ) , ( GR1373 ) eri-1 ( mg366 ) , ( YY191 ) eri-1 ( mg366 ) ; nrde-1 ( gg088 ) , ( YY468 ) eri-1 ( mg366 ) ; nrde-4 ( gg129 ) , ( YY268 ) nrde-1 ( gg088 ) ; ggIS12[nrde-3p::3xflag::gfp::nrde-1] , ( YY464 ) nrde-1 ( gg088 ) ; nrde-2 ( gg091 ) ; ggIS12 , ( YY459 ) nrde-1 ( gg088 ) ; nrde-3 ( gg066 ) ; ggIS12 , ( YY462 ) nrde-1 ( gg088 ) ; nrde-4 ( gg129 ) ; ggIS12 , ( YY174 ) ggIS1[nrde-3p::3xflag::gfp::nrde-3] , ( YY225 ) rde-1 ( ne219 ) ; ggIS1 , ( YY228 ) nrde-1 ( gg088 ) ; ggIS1 , ( YY454 ) nrde-4 ( gg129 ) ; ggIS1 , ( YY230 ) rrf-1 ( pk1417 ) ; ggIS1 , ( YY346 ) nrde-2 ( gg091 ) ; ggIS28[nrde-3p::3xflag::gfp::nrde-2] . For FLAG::GFP::NRDE-1 ( referred to as GFP::NRDE-1 when assaying NRDE-1 expression or FLAG::NRDE-1 when referring to NRDE-1 immunoprecipitation or western blotting ) the nrde-1 coding region and predicted 3′UTR were amplified by PCR from genomic N2 DNA and inserted into the pSG082 plasmid 3′ to the nrde-3p::3xFLAG::GFP . Low copy integrated transgenes were generated by biolistic transformation [44] . RNAi experiments were conducted as described previously [45] . The lir-1 and unc-15 bacterial clones were taken from the Ahringer library [46] . The lin-15b clone was described previously [15] . RIPs were performed as described previously [15] . Hypochlorite-isolated embryos were used for all RIPs . FLAG::NRDE-1 and FLAG::NRDE-3 proteins were immuno-precipitated with anti-FLAG M2 antibody ( Sigma , A2220 ) . ChIP experiments were performed as described previously [16] . Hypochlorite-isolated embryos were used for ChIP experiments . Isolated embryos were snap-frozen in liquid-Nitrogen before performing ChIP . FLAG::NRDE-1 , FLAG::NRDE-2 , and FLAG::NRDE-3 proteins were immuno-precipitated with anti-FLAG M2 antibody ( Sigma , A2220 ) . H3K9me3 antibody was from Upstate ( 07-523 ) . NRO was performed as described previously [16] . Hypochlorite-isolated embryos were used for NROs . RNAs were converted to cDNA by the iScript cDNA Synthesis Kit ( Bio-Rad , 170–8890 ) following the vendor's protocol . | Chromatin consists of DNA and proteins . Chromatin can exist in many different states . The state of chromatin in highly regulated in order to ensure that genes are expressed correctly . RNAs play an important role in the regulation of chromatin . For example , in plants and fungi small RNAs drive the formation of heterochromatin , a repressive chromatin state . Many types of small RNAs have been identified in animal cells , but the functions of these small RNAs are largely unknown . Using the nematode C . elegans as a model system , we identified a small RNA pathway that regulates the state of chromatin . We report the identification of two new factors , termed NRDE-1 and NRDE-4 , which act in this nuclear small RNA pathway . NRDE-1 and NRDE-4 link small RNAs to chromatin regulation . Additionally , we show that endogenously expressed small RNAs , termed the endo-siRNAs , direct the post-translational modification of histone proteins , which are core components of chromatin . These results establish a direct connection between small RNAs and chromatin regulation in animals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"model",
"organisms",
"genetics",
"biology",
"genetics",
"and",
"genomics"
]
| 2011 | A Pre-mRNA–Associating Factor Links Endogenous siRNAs to Chromatin Regulation |
A fundamental challenge in the post-genome era is to understand and annotate the consequences of genetic variation , particularly within the context of human tissues . We present a set of integrated experiments that investigate the effects of common genetic variability on DNA methylation and mRNA expression in four human brain regions each from 150 individuals ( 600 samples total ) . We find an abundance of genetic cis regulation of mRNA expression and show for the first time abundant quantitative trait loci for DNA CpG methylation across the genome . We show peak enrichment for cis expression QTLs to be approximately 68 , 000 bp away from individual transcription start sites; however , the peak enrichment for cis CpG methylation QTLs is located much closer , only 45 bp from the CpG site in question . We observe that the largest magnitude quantitative trait loci occur across distinct brain tissues . Our analyses reveal that CpG methylation quantitative trait loci are more likely to occur for CpG sites outside of islands . Lastly , we show that while we can observe individual QTLs that appear to affect both the level of a transcript and a physically close CpG methylation site , these are quite rare . We believe these data , which we have made publicly available , will provide a critical step toward understanding the biological effects of genetic variation .
With the widespread application of highly parallel SNP genotyping arrays much of the recent effort in human genetics has focused on defining the role of genetic variation in disease and physical traits . A small subset of this work , however , has attempted to examine the more proximal effects of genetic variation on mRNA and protein levels [1]–[5] . This has the potential to inform on several levels; first , it is a critical step toward understanding the pathobiological consequences of genetic variants linked to disease; second , it affords the opportunity to form inferences regarding relationships between genes based on patterns of co-regulation; and third , it provides a more complete view of multiple levels of regulation of gene expression than that provided by the traditional reductionist method [6] , [7] . Epigenetic alterations , including DNA methylation , histone modification and RNA mediated gene silencing , are defined as heritable changes in gene function that occur without an alteration of the underlying DNA sequence and which afford a level of transcriptional regulation above and beyond DNA sequence [8] . DNA methylation , which occurs at discrete CpG dinucleotide motifs , is believed to be an important mediator of gene expression; this observation has been most frequently linked to DNA methylation at CpG islands , regions of the genome that contain a high density of CpG sites , often proximal to gene promoter regions . A classical inverse relationship between the extent of DNA methylation at CpG islands and expression levels of the proximal gene product has been most often described [8] . To date the relationship between genetics , DNA methylation and gene expression is one that has been largely and necessarily confined to observations at single loci and transcripts in individual cell systems or tissues . The recent development of genome-scale technologies provides unprecedented opportunities to expand upon these experiments . The integration of genetic , epigenetic and expression data promises to provide general observations regarding the relationship between genetic variation and expression . Beyond these observations these data can be readily mined to unravel the network of effects associated with genomic variants . This may reveal some of the rather cryptic intermediate events that occur between DNA variant and phenotype . Because of our interest in genomic regulation of expression and neurological disorders we embarked upon a series of experiments to provide a brain region-specific contextual framework for genetic and epigenetic regulation of gene expression . We were particularly interested in mapping the effects of common genetic variation on gene expression and DNA methylation; the widespread adoption of genome wide association studies for disease and traits has generated a large number of associated loci , and such a map would allow these loci to be associated with a biological consequence . We obtained frozen brain tissue from the cerebellum ( CRBLM ) , frontal cortex ( FCTX ) , pons ( PONS ) and temporal cortex ( TCTX ) from 150 subjects ( total 600 tissue samples ) . We undertook three separate assays across this series; first , genome-wide SNP genotyping; second , assay of 27 , 578 CpG methylation sites in each of the four brain regions; third , mRNA expression profiling of 22 , 184 transcripts in all four brain regions . Here we discuss the results of these experiments , particularly in the context of integrated datasets to define expression quantitative trait loci ( eQTL ) and CpG methylation quantitative trait loci ( methQTL ) , where quantitative trait loci ( QTL ) are correlations between genetic variation within genomic region ( s ) ( in this case single nucleotide polymorphisms ) and a quantitative trait ( DNA methylation or expression ) , and detailing differences and similarities across brain regions .
To assess whether differences in CpG methylation and expression were consistently different between brain regions , our initial analyses focused on global comparison of these measures across tissues . Performing a hierarchical cluster analysis [9] of these data demonstrated that the four brain regions have different epigenetic and expression profiles ( Figure 1A and 1B ) . Expression pattern differences were distinct between cerebellum , pons and cortical tissue , with frontal and temporal cortices clearly separating within the mRNA dataset , but overlapping within the CpG methylation data . These data show that CpG methylation and mRNA levels vary measurably between brain regions . These data are in agreement with previous work that demonstrated distinct patterns of gene expression and DNA methylation in human cerebellum compared to cortical tissues [10] , [11] . The next analysis was limited to the CpG methylation and mRNA datasets of those probes where an Illumina detection p-value of less than or equal to 0 . 01 was observed in 95% of samples analyzed in each of the four tissue region . This provided data on a total of 27 , 277 unique CpG methylation sites and 8 , 076 probes against individual mRNA transcripts that were present in all four brain regions ( Figure S2 ) . The distribution of observed CpG methylation levels and transcript abundance was plotted as a histogram for each tissue ( Figure 1C and 1D ) . A high proportion of CpG sites interrogated within this assay were infrequently methylated , whereas a smaller proportion were highly methylated , a trend that was apparent across tissues . This distribution of DNA methylation ( Figure 1C ) closely matches distributions of DNA methylation reported previously by sequencing in multiple human cell types [12] , [13] . CpG sites that are within CpG Islands [14] , [15] were predominately unmethylated . Grouping of mRNAs into high through low expression groups revealed abundance profiles similar to those previously reported ( Figure 1D ) . We next compared CpG methylation and mRNA expression levels at individual loci directly between each possible pair of tissue regions ( Figure 1C and 1D , scatterplots ) . In general levels of DNA methylation and expression were quite similar between tissues . Measures within frontal and temporal cortices were consistently the most alike and cerebellar tissue provided the most distinct profile of the four regions . A primary aim underlying these experiments was to examine the extent of genetic control of DNA methylation and expression within brain tissues . To investigate this process , we undertook a series of QTL analyses . From the 537 , 411 genotyped SNPs that passed quality control filtering we then imputed genotypes for 2 , 545 , 178 SNPs using MACH [16] and HapMap CEU phase data . After additional quality and analysis specifications filtering of the imputed genotypes 1 , 629 , 853 SNPs ( average ) were used for analysis . The QTL analysis of each tissue region was performed separately so we expanded our trait selection from those that were detected in 95% of samples across all four brain regions to those that were detected in 95% of samples within a specific brain region . Changing this trait selection threshold based on what is present within a specific brain region allowed us to analyze additional mRNA transcripts that are well detected within one or more brain regions but not all four tissue regions . The number of SNPs , CpG sites and mRNA transcripts tested for each brain tissue can be found in Table 1 . With these data for each tissue we then performed linear regression of allele dosage against each measure using CpG methylation or expression of mRNA as the dependent variable and genotype as the independent variable . This regression analysis was performed with Plink [17] to correlate allele dosage with the quantitative trait . We corrected for number of tests per trait by permutation , computing a genome-wide empirical p-value for each of the ∼1 . 6 million SNPs tested against each individual trait . To correct for the number of traits tested within each brain region by assay type an FDR threshold for significance was determined based on the empirical p-values from the proceeding step . This yielded a necessarily conservative threshold for significance [18] . Prior to analysis each trait was adjusted using available biological and methodological covariates in an attempt to reduce the influence of systematic confounding effects ( Text S1 ) . Post hoc we annotated significant QTLs as cis if the SNP lay within 1MB of either the CpG methylation site in question or the transcript being tested; all other SNP-dependent variable tests were designated as trans . Notably , because designation of cis and trans QTL tests was performed post hoc , there was no distinction in terms of level of statistical correction between these groups . Our previous data from the human cerebral cortex [2] , as well as data from HapMap lymphoblastoid cell lines ( LCL ) [5] have suggested that SNPs proximal to genes including SNPs upstream of the transcriptional start site ( TSS ) within the gene and downstream of the transcript end site ( TES ) have a greater influence on gene expression than those further away . This is presumably because genetic variation around promoter elements , splice sites and 3′ UTR affects transcription , splicing and mRNA stability [19] that results in a relative enrichment of cis over trans eQTLs . To see if this observation generalized to multiple brain regions and also to QTLs linked to CpG methylation , we plotted all significant QTLs by genomic position of the SNP and transcript or SNP and CpG site ( Figure 2 ) . Previous work has shown that SNPs are weakly correlated with the methylation status of CpG islands as compared to other DNA characteristics such as sequence and structure [15]; however we detected a large number of significant correlations detected between genetic variation and the methylation status of CpG sites ( Table 2 ) . Genome-wide visualization of detected methQTLs illustrated a strong positional effect of this relationship with an excess of the number and magnitude of cis associations ( Figure 2A ) . The detected methQTLs accounted for between 18% and 88% of corrected methylation levels at individual loci . The number of CpG sites with methQTLs that were significant after correction for genome-wide multiple testing ranged from 1085 ( 4% ) in the cerebellum to 1417 ( 5 . 1% ) in the temporal cortex ( Table 2 ) . All significant methQTL results can be found in Table S3 . Detection of cis QTL for DNA methylation was more likely when the CpG site was outside of an island ( as defined in [14] ) ; while 42% of sites on the array are within a CpG island , only 18% of the CpG sites with a significant QTL were in islands . A similar number of mRNA transcripts with significant eQTLs were detected in each of the four brain regions , ranging from 280 ( 3 . 2% ) in the pons to 391 ( 4 . 2% ) in the temporal cortex ( Table 3 ) . These eQTLs accounted for between 18% and 77% of corrected expression levels of associated transcripts between individuals . All significant eQTL results can be found in Table S4 . Proportionally a similar number of QTLs were observed for CpG methylation and mRNA expression . In order to determine whether SNP variability was more influential in CpG methylation or expression , we performed a random sampling , with 10 , 000 iterations , of QTL calculations for each independent measure in a core set of 100 individuals ( Text S1 ) . The results of this analysis show on average that 2 . 34% of CpG sites and 1 . 99% of mRNAs significantly correlate with QTLs . Each mRNA correlated with twice as many SNPs , on average , than CpG sites did . However the average R2 of the SNPs correlated with mRNAs was equal to CpGs ( Figure S6 ) . The abundance of cis QTL for CpG methylation and mRNA expression prompted us to examine the distribution of cis methQTLs and eQTLs ( Figure 2C–2L ) . This revealed that both the number of significant methQTLs and the strength of association between SNP and DNA methylation level were inversely correlated with physical distance between the genetic and epigenetic variants in question ( Figure 2C–2F ) . Furthermore , this relationship was also evident for cis eQTLs ( Figure 2H–2K ) . The average distance between correlated cis SNP and trait was 81Kb for CpG sites and 121Kb for mRNA transcripts . The largest effect QTLs for both cis methQTLs and eQTLs tended to be present in all four tissues tested ( Figure 2G and 2L ) . Of the mRNA transcripts where a cis eQTL was significantly detected in at least one brain region 53% have been previously reported . This number increased to 70% when analysis is limited to those transcripts with a cis eQTL consistently detected in all four tissues [2]–[4] ( Table S2 ) . To assess the enrichment of detected cis QTLs relative to those in trans we calculated the number of observed and possible cis and trans QTLs for CpG methylation and mRNA expression levels . The calculation of possible cis QTLs was performed by counting the number of all testable , genotyped SNPs within 1Mb of the CpG probe or transcript in question , regardless of significant association; calculation of possible trans QTLs was performed by counting the number of all testable , genotyped SNPs throughout the genome , excluding those within 1Mb of the CpG probe or transcript in question , again , regardless of significant association . This analysis was repeated for every tested CpG probe and transcript . These data showed an extreme enrichment of cis methylation and expression QTLs relative to trans ( ∼4400-fold and ∼7300-fold respectively ) . Although it should be noted that this calculation of enrichment does not take into account the differences in power to detect cis versus trans affects , thus the extreme nature of these fold changes may be inflated . While the average distance between correlated cis SNP and trait was 81Kb for CpG sites and 121Kb for mRNA transcripts when cis is defined as 1Mb , the peak enrichment of the number of significant cis QTLs was observed when the threshold distance for what is considered cis was set at ∼45bp for CpG sites and ∼68kb for mRNA transcripts ( Figure 3 ) . Based on our results that reached statistical significance many QTLs appear to be tissue specific , where 49% of CpG sites and 54% of mRNA transcripts with a significant cis QTL were only detected within one tissue . Table 4 shows summary counts of CpG sites and mRNA transcripts with a QTL found in all four tissues . Because this analysis relies on a threshold for significance , it has the potential to be misleading , discounting QTLs that do not quite reach the threshold for significance . Thus , in order to compare detected methQTLs and eQTLs between tissues , we selected every SNP-CpG methylation pair and SNP-transcript pair that passed the defined threshold for significance in at least one tissue . We then compared R2 values for each of these SNP-CpG pairs or SNP-transcript pairs in all four tissues , including results from tissues where the SNP-CpG pair or SNP-transcript pair was non-significant , using ternary plots ( Figure 4 ) . The majority of large effect and many moderate effect QTLs were shared across the four brain regions ( Figure 4 ) when significant effects from a tissue are compared with corresponding ( possible non-significant ) effects from another tissue . Of interest , a subset of trans methQTLs ( defined here as any SNP CpG pair not within 1MB of each other ) was also shared between regions and had high R2 values . This suggests that , while methQTLs show cis enrichment , there are also a number of robust distal effects where SNP and CpG methylation show significant association . These plots illustrate that the majority of eQTLs or methQTLs with strong effect sizes were consistent across tissues . For example , a large effect eQTL was found for CHURC1 , which encodes a protein proposed to be involved in transcriptional regulation , in all tissues ( Figure S7 and as reported previously [3] ) . However , there were also rare , but observable , events where a large effect QTL was detected within a single tissue and was completely absent in the other three tissues . For example the cis eQTL for PPAPDC1A , encoding a phosphatidic acid phosphatase that displays hydrolase and phosphotase activity at the membrane , has a large effect that appears to be restricted to the cerebellum , despite reliable detection of the transcript in all four brain regions ( Figure 4Q and 4R–4T and Figure S8 ) . We next sought to investigate whether the observed methQTLs and eQTLs represent individual loci where the underlying variant influences both CpG methylation and gene expression . These data revealed that in general , there was little co-association between methQTLs and eQTLs . Of the SNP , CpG and mRNA combinations considered ( see Materials and Methods ) , 4 . 8% ( average across tissues ) were significant as both a methQTL and eQTL; this includes 2 . 6% ( average of 13 sites per tissue ) of CpG methylation sites where a significant methQTL is present and 8 . 2% ( average of 11 transcripts per tissue ) of mRNA transcripts that have a significant eQTL . For this 4 . 8% with a shared QTL both the mean distance between the CpG site and mRNA TSS and the strength of the correlation differed from that of the considered combinations . For the shared QTLs the distance between CpG site and mRNA TSS is 27 . 5Kb whereas in all considered the mean distance was 394 . 4Kb . The mean R2 between CpG methylation and mRNA expression within the shared QTL set was 0 . 255 as compared to 0 . 033 in the considered set . Within these possibly shared QTLs , almost half of the significant methQTLs and eQTLs were correlated in the same direction , i . e . the CpG methylation level was positively correlated with the mRNA expression level ( Figure S9 ) . However; when considering only those shared QTLs where the CpG site was within a CpG island ( as defined in [14] ) , 91% of these CpG methylation sites and mRNA expression pairs are inversely correlated , in line with the traditional view that increased CpG island methylation is associated with decreased expression at geographically local transcripts ( Figure 5 ) .
Elucidating the genetic control of biological processes is a critical issue in the post genome era . This will inform on the basic biological level and lead to clearer understanding of the pathophysiology underlying human disease . The work we have described here and the public release of the data resulting from this effort , aims to facilitate an understanding of the initial consequences of common genetic variation on DNA methylation and gene expression in brain ( GEO Accession Number: GSE15745; dbGAP Study Accession: phs000249 . v1 . p1 ) [20]–[22] . Our data show clearly that patterns of expression are measurably different across brain tissues . QTL analyses reveal an abundance of eQTLs and , as previously reported , significant eQTLs are predominantly cis in nature [2] , [5] . It is notable , particularly given the systematic differences in expression patterns among the four tissues , that the majority of large effect eQTLs detected were consistent across brain tissues . Large effect tissue-specific QTLs are also observable in the current data set , suggesting that there are some genetic effects on expression that are dependent on the tissue type used irrespective of expression levels of the mRNA . Our results that reached statistical significance are consistent with recent work showing that many eQTLs are likely tissue specific , but they typically have a smaller effect and are more distant [23]–[25] . However within our data when we observe a QTL that is only significant in a single tissue , examination of the correlation between the same SNP and trait in other tissues generally reveals a suggestive , although non-significant , correlation in the other tissues; thus both large and many moderate size effects appear to be shared . While this aspect of our data may seem at odds with the previous studies which in part used purified cells or cell lines , the current samples within our study were exclusively from heterogeneous tissues . Thus one might expect to see quantitative trait loci that are generalizeable across tissues because in essence each tissue is a heterogeneous mix of cells and we are more likely to detect QTLs that are in a majority of cell types . The detection of cell type specific QTLs is apparently still feasible , however , such correlations will have to be large enough to detect within the background of a heterogeneous tissue . We have shown for the first time that there are large numbers of methQTLs in human tissue . It is worth noting , however , that the CpG methylation sites assayed here have not been defined through experimental work; as such a proportion of these may be consistently non-methylated , and thus our estimate of the proportion of CpG sites with an associated methQTL may be an underestimate . As for mRNA expression , DNA methylation patterns are sufficiently different to predict the originating tissue; however , we also see that SNP variation strongly influences the level of DNA methylation at sites that show intermediate levels of methylation . The strength of effect for methQTLs is similar to those for eQTLs and as discussed above eQTLs and methQTLs tend to be found consistently across tissues , supporting the robustness of these loci . Additionally methQTLs also show cis-enrichment , suggesting that local genomic variants influence the propensity of a particular genomic site to be methylated . Combined analysis of methQTLs and eQTLs did not reveal a simple relationship between genetically influenced DNA methylation and expression . QTLs that influenced both DNA methylation and expression levels of a physically close transcript were identified , these were in the minority; for the most part strong methQTLs were not also QTLs for expression . It is plausible that this observation reflects a genuine biological disconnect between CpG methylation and gene expression . Alternatively , this may be a result of poor power within the series to detect such effects or , as discussed below , a consequence of differences in cellularity confounding our results . Notably , when analyzing joint methQTL-eQTLs where the methylation site in question resided in a CpG island , the classical inverse relationship between CpG methylation and expression was clearly observed . Increasing the sample size of the current study will likely yield a larger number of QTLs , particularly those with small to moderate effect size . It would also be of interest to extend this initial work to other tissues , which will not only be important in defining tissue specific QTLs , but also will afford us the opportunity to examine QTLs at transcripts not typically detected in brain . Additionally , we have limited our sampling to a population that represents only a small component of worldwide human genetic diversity [26] . Adding tissues from other populations would provide an understanding of genetic control of expression in different groups and a framework for interpretation of GWAS results in these groups . Furthermore , inter-population comparison would help refine the basis by which differences in genetic architecture affect expression . We also note that our coverage of DNA modifications is currently incomplete as there are other epigenetic modifications , such as acetylation , that are not currently accessible to high throughput techniques , and many more methylation sites exist than assayed here . Whether genotypic differences influence other components of the epigenome is unknown . Finally , in a complex organ such as brain cellularity will confound analyses correlating measures that vary between cell types . Thus , CpG methylation to mRNA correlations pose a considerable challenge in heterogeneous tissues . In the context of human tissues this problem may be solved by selection of cell type by laser dissection or by culture and terminal differentiation of cohorts of pluripotent stem cells . Both solutions have considerable technical limitations at this time . In summary , we show data that we argue convincingly demonstrates QTLs for DNA methylation and expression exist across human tissues . The data presented here provides an initial basis for understanding how genetic variance in humans influences epigenetic marks and expression; we argue that these observations may be useful in the understanding of gene contributions to human phenotypes . As well as providing a more complete model of the complex control of gene expression in the brain , our data may be useful in moving rapidly from locus to mechanism of disease .
Frozen tissue samples of the cerebellum ( CRBLM ) , frontal cortex ( FCTX ) , caudal pons ( PONS ) and temporal cortex ( TCTX ) were obtained from 150 neurologically normal Caucasian subjects ( Text S1 ) . 100–200mg aliquots of frozen tissue were sub-dissected from each tissue from all 150 subjects resulting in 600 tissue samples and used for methylation assay and expression assays . Genotyping was performed using Infinium HumanHap550 beadchips ( Illumina ) to assay genotypes for 561466 SNPs , from the cerebellum tissue samples . CpG methylation status was determined using HumanMethylation27 BeadChips ( Illumina ) , which measure methylation at 27 , 578 CpG dinucleotides at 14 , 495 genes . Profiling of 22 , 184 mRNA transcripts was performed using HumanRef-8 Expression BeadChips ( Illumina ) as previously described [27] . The threshold call rate for inclusion of the subject in analysis was 95% . All 150 subjects had a call rate greater than 95% , and were included in the subsequent analyses ( average call rate = 99 . 86%; range 97 . 72%–99 . 95% , based on the missing procedure within the PLINK v1 . 04 software toolset [17] . The gender of the subjects reported by the brain banks was compared against their genotypic gender using PLINK's check-sex algorithm , which determines a sample's genotypic gender based on heterozygosity across the X chromosome . A gender discrepancy was detected for one subject and excluded from further analysis . To confirm the ethnicity of the samples , Identity-By-State ( IBS ) clustering and multidimensional scaling analyses were performed within PLINK using the genotypes from the brain samples that had been merged with data from the four HapMap [28] populations ( n = 32 Caucasian ( CEU ) , 12 Han Chinese , 16 Japanese and 24 Yoruban non-trio samples previously genotyped by Illumina and assayed on the Infinium HumanHap500 version genotyping chips ) . Two samples were outliers based on population and excluded from further analysis ( Figure S1 ) . Genotype data of the samples were compared for cryptic relatedness using the Identity-By-Descent ( IBD ) procedure within PLINK . No samples were found to be from related individuals . Mach software version 1 . 0 . 16 [16] and HapMap CEU phase data ( release 22 ) were used to impute genotypes for ∼2 . 5 million SNPs . Imputed SNPs were excluded if the linkage disequilibrium r2 values between imputed and known genotypes was less than 0 . 3 , and if their posterior probability averages were less than 0 . 8 for the most likely imputed genotype . For each of the four tissue regions , SNPs were also excluded if: ( a ) call rate was less than 95% , ( b ) Hardy-Weinberg equilibrium ( HWE ) p-value was less than 0 . 001 , and ( c ) the SNP had less than 3 minor homozygotes present . Exact numbers of SNPs used per brain tissue and assay type are shown in Table 1 . Raw intensity values for each probe were transformed using the rank invariant normalization method [29]–[31] and then log2 transformed for mRNA analysis . Four samples were excluded from further analysis , as they were outliers based on overall probe detection rate and/or mean expression level; one CRBLM , one FCTX and two PONS . The threshold call rate for inclusion of samples in the analysis was 95% . Based on this metric , 9 CRBLM , 2 FCTX , 6 PONS and 8 TCTX samples were excluded from further analysis . The remaining brain samples had an average detection rate of 99 . 84% ( range 95 . 0% to 99 . 98% ) . The gender of the samples reported by the brain banks were compared against their assayed gender based on values of methylation from CpG sites on the X chromosome . Four samples with gender discrepancies were detected and were removed from subsequent analysis; two CRBLM , one PONS and one TCTX . The resulting HCL sample tree based on all Chromosome X probes , after removal of these four individuals is shown in Figure S3 . Performing a Hierarchical Clustering ( HCL ) [9] of the sample profiles using the TM4 MeV version 4 . 1 . 01 tool [32] , with Euclidian distances and ‘Average Linkage clustering’ resulted in the samples separating fairly well by brain tissue region . Where in mRNA separation of samples into 4 clusters matching brain tissue region was clear . The frontal and temporal cortices could not be separated in within CpG data . For clustering all detected data was used for mRNA; only detected data for autosomal probes was used in clustering of the CpG data , otherwise sub-clusters based on gender appeared . The HCL samples trees were saved a Newick tree files and plotted again using the HyperTree tool ( http://hypertree . sourceforge . net/ ) . Traits were excluded from analysis if they were detected in less than 95% of samples for each tissue region . For each tissue region and trait type ( CpG and mRNA ) the 95% threshold was determined using total number of analyzable samples for this pairing of region and trait . A probe is considered detected for a sample if the reported Illumina Detection p-value was less than or equal to 0 . 01 . Numbers of analyzable samples per trait ( or assay ) type and region are available in Table 1 overlap of detected probes among brain tissues is shown as Venn diagram in Figure S2 . Prior to quantitative trait loci analysis each trait was adjusted using the available biological and methodological covariates in an attempt to remove the influence of these potentially confounding affects . In R each trait was regressed using the following model:Where Y is the trait profile ( log2 normalized mRNA expression intensities and raw values of CpG DNA methylation ) and X1 … Xn represent the biological covariates Age and Gender and the methodological covariates post mortem interval ( PMI ) , which Brain Bank the samples was from and which preparation/hybridization batch the sample was processed in . Within this model gender , tissue bank and batch where treated as categorical covariates . After fitting each trait to the model the residuals from the model are kept and represent the trait in following analyses . Thus variance attributable to gender , age , post-mortem interval , tissue source and hybridization batch are removed prior to QTL analysis . These covariate data are available in Table S1 . Histograms showing the proportion of traits that are potentially impacted by these covariates are shown in Figure S4 ( CpG ) and Figure S5 ( mRNA ) . For each of the four brain regions , a regression analysis was performed on the residuals described in the preceding section for ( a ) mRNA expression and ( b ) the methylation values generated for CpG sites . The trait residuals were then used as the quantitative phenotype for that probe in genome-wide association analysis looking for quantitative trait loci . These analyses were performed using the assoc function within Plink , which correlates allele dosage with change in the trait . Each of the four tissue regions was analyzed separately , and independent genome-wide association analyses were performed looking for ( a ) expression quantitative trait loci ( eQTLs ) for mRNA and ( b ) methylation quantitative trait loci ( methQTLs ) for CpG sites . The Plink toolset quantitative trait association analysis fits data to the following model:Where Y is the quantitative trait and ADD represent genotypes encoded as allele dosage . See Plink Quantitative trait association and Linear and Logistic models documentation for more information . To correct for the large number of SNPs tested per trait , a genome-wide empirical p-value was computed for the asymptotic p-value for each SNP by using 1 , 000 permutations of swapping sample labels of the traits , using the maxT permutation functionality provided within Plink . A permutation based method using label swapping of the traits is an appropriate method of test correction [18] for these analysis as it is not dependent on these quantitative traits having a normal distribution and also allows the linkage disequilibrium of the genomic regions being tested against the traits to be maintained . To correct for the number of traits being tested per tissue region , a false discovery rate ( FDR ) threshold was determined based on the empirical p-values using the fwer2fdr function of the multtest package in R version 2 . 6 . 1 . Empirical p-values were allowed to exceed this threshold if their linkage disequilibrium r2 was greater than or equal to 0 . 7 with a SNP with empirical values within the FDR threshold . Sequence variants within the sequence of the probe used to assay individual traits may cause differential hybridization and inaccurate expression and possible methylation measurements . To exclude this confound , the sequences of probes with significant correlation to a trait were examined for the presence of polymorphisms using CEU HapMap data , and , if present , that QTL was removed from the result set . Within each tissue , we selected every pairing of CpG methylation sites and mRNA transcripts where the CpG methylation site was within 1Mb of the mRNA TSS and both the CpG methylation site or mRNA transcript had a significant cis QTL . These CpG-mRNA pairs were then expanded into triplets of SNP and CpG-mRNA pair , where the SNP was significantly correlated in cis with either the CpG methylation site or mRNA transcript of the CpG-mRNA pair . Data resulting from these experiments is available online ( GEO Accession Number: GSE15745; dbGAP Study Accession: phs000249 . v1 . p1 ) . | In this paper , we describe a comprehensive assessment of the correlation between common genetic variability across the human genome , gene expression , and DNA methylation , within human brain . We studied the cerebellum , frontal cortex , temporal cortex , and pons regions of 150 individuals ( 600 tissue samples ) . In each tissue , we assessed 27 , 578 DNA methylation sites and the expression level of 22 , 184 genes . Our research shows that DNA methylation and RNA expression patterns differ between brain regions . Further , we show that DNA genotype is correlated with gene expression and DNA methylation , particularly when the genetic variation is close to the DNA methylation site or gene . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics/bioinformatics",
"genetics",
"and",
"genomics/gene",
"expression"
]
| 2010 | Abundant Quantitative Trait Loci Exist for DNA Methylation and Gene Expression in Human Brain |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.