Open Collections

UBC Faculty Research and Publications

Early immune adaptation in HIV-1 revealed by population-level approaches Martin, Eric; Carlson, Jonathan M; Le, Anh Q; Chopera, Denis R; McGovern, Rachel; Rahman, Manal A; Ng, Carmond; Jessen, Heiko; Kelleher, Anthony D; Markowitz, Martin; Allen, Todd M; Milloy, M-J; Carrington, Mary; Wainberg, Mark A; Brumme, Zabrina L Aug 29, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12977_2014_Article_64.pdf [ 1.58MB ]
Metadata
JSON: 52383-1.0224006.json
JSON-LD: 52383-1.0224006-ld.json
RDF/XML (Pretty): 52383-1.0224006-rdf.xml
RDF/JSON: 52383-1.0224006-rdf.json
Turtle: 52383-1.0224006-turtle.txt
N-Triples: 52383-1.0224006-rdf-ntriples.txt
Original Record: 52383-1.0224006-source.json
Full Text
52383-1.0224006-fulltext.txt
Citation
52383-1.0224006.ris

Full Text

RESEARCH Open AccessEarly immune adaptation in HIV-1 revealed byConclusion: Cross-sectional host/viral genotype datasets represent an underutilized resource to identifyMartin et al. Retrovirology 2014, 11:64http://www.retrovirology.com/content/11/1/642British Columbia Centre for Excellence in HIV/AIDS, Vancouver, BC, CanadaFull list of author information is available at the end of the articlereproducible early pathways of HIV-1 adaptation and identify correlates of protective immunity.Keywords: Human immunodeficiency virus type-1 (HIV-1), Human leukocyte antigen (HLA) class I, CD8+ cytotoxicT-lymphocytes (CTL), Immune escape, HLA-associated polymorphism, Adaptation, Evolution, Acute/early infection,Population-level analysis, Statistical association with phylogenetic correction* Correspondence: zbrumme@sfu.ca†Equal contributors1Faculty of Health Sciences, Simon Fraser University, Burnaby, BC, Canadapopulation-level approachesEric Martin1,2†, Jonathan M Carlson3†, Anh Q Le1, Denis R Chopera1,4, Rachel McGovern2, Manal A Rahman1,Carmond Ng2, Heiko Jessen5, Anthony D Kelleher6, Martin Markowitz7, Todd M Allen8, M-J Milloy2,9,Mary Carrington8,10, Mark A Wainberg11 and Zabrina L Brumme1,2*AbstractBackground: The reproducible nature of HIV-1 escape from HLA-restricted CD8+ T-cell responses allows theidentification of HLA-associated viral polymorphisms “at the population level” – that is, via analysis of cross-sectional,linked HLA/HIV-1 genotypes by statistical association. However, elucidating their timing of selection traditionally requiresdetailed longitudinal studies, which are challenging to undertake on a large scale. We investigate whether the extentand relative timecourse of immune-driven HIV adaptation can be inferred via comparative cross-sectional analysisof independent early and chronic infection cohorts.Results: Similarly-powered datasets of linked HLA/HIV-1 genotypes from individuals with early (median < 3 months)and chronic untreated HIV-1 subtype B infection, matched for size (N > 200/dataset), HLA class I and HIV-1 Gag/Pol/Nefdiversity, were established. These datasets were first used to define a list of 162 known HLA-associated polymorphismsdetectable at the population level in cohorts of the present size and host/viral genetic composition. Of these 162known HLA-associated polymorphisms, 15% (occurring at 14 Gag, Pol and Nef codons) were already detectable viastatistical association in the early infection dataset at p ≤ 0.01 (q < 0.2) – identifying them as the most consistentlyrapidly escaping sites in HIV-1. Among these were known rapidly-escaping sites (e.g. B*57-Gag-T242N) and othersnot previously appreciated to be reproducibly rapidly selected (e.g. A*31:01-associated adaptations at Gag codons397, 401 and 403). Escape prevalence in early infection correlated strongly with first-year escape rates (Pearson’sR = 0.68, p = 0.0001), supporting cross-sectional parameters as reliable indicators of longitudinally-derivedmeasures. Comparative analysis of early and chronic datasets revealed that, on average, the prevalence ofHLA-associated polymorphisms more than doubles between these two infection stages in persons harboring therelevant HLA (p < 0.0001, consistent with frequent and reproducible escape), but remains relatively stable inpersons lacking the HLA (p = 0.15, consistent with slow reversion). Published HLA-specific Hazard Ratios forprogression to AIDS correlated positively with average escape prevalence in early infection (Pearson’s R = 0.53,p = 0.028), consistent with high early within-host HIV-1 adaptation (via rapid escape and/or frequent polymorphismtransmission) as a correlate of progression.© 2014 Martin et al., licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.distribution and HIV-1 diversity. We did so by drawingMartin et al. Retrovirology 2014, 11:64 Page 2 of 16http://www.retrovirology.com/content/11/1/64BackgroundHIV-1 escape from Human Leukocyte-Antigen (HLA)class I-restricted CD8+ T-lymphocytes (CTL) occurs ina broadly predictable manner based on the HLA allelesexpressed by the host [1]. Reversion of escape mutations,usually to consensus, upon HIV-1 transmission to anindividual lacking the restricting HLA also occurs re-producibly in many [2-5], though not all [6-8], cases.The reproducible nature of viral adaptation allows usto identify HLA-associated polymorphisms in HIV-1(that is, viral polymorphisms that are significantly over- orunder- represented among persons expressing a givenHLA allele) “at the population level” (that is, via the ana-lysis of cross-sectional, linked HLA/HIV-1 genotypes viastatistical association approaches that additionally correctfor various potential confounders [9-12]). Such studies arenormally undertaken in chronic infection, as the virushas undergone a majority of its within-host adaptationby this stage. For example, a recent population-levelstudy of >1800 chronically HIV-1 subtype B-infectedpersons identified over >2000 HLA-associated polymor-phisms across HIV-1, with a majority occurring in Gag,Pol and Nef [11].Though HLA-associated polymorphisms in HIV-1 canbe identified using cross-sectional approaches, their tim-ing of selection cannot be directly determined by thesemethods. Rather, temporal information is ideally estab-lished via detailed longitudinal study of untreated indi-viduals recently infected with HIV-1 (e.g.: [2,3,13-19]).However, identifying large numbers of recently-infectedpersons is challenging. Another consideration is that,given the current evidence and clinical recommenda-tions supporting HIV-1 treatment initiation in early in-fection [20], prospective longitudinal observational studyof untreated HIV-1 infection may no longer be feasiblenor ethical moving forward.As such, cross-sectional pretreatment host/viral geno-type datasets from individuals at different HIV-1 infec-tion stages enrolled in established (or future) cohortscould potentially provide alternate data sources to inferthe extent and time course of immune-driven HIV-1adaptation, including the earliest events post-infection,using population-level approaches. Though such ap-proaches have been investigated [21,22], they remainunderutilized in this context. Notably, population-levelapproaches offer one key advantage in that - by defin-ition - they specifically identify HIV-1 adaptations thatoccur reproducibly in persons expressing the restrictingHLA [21] (as opposed to longitudinal studies thatcharacterize immune escape dynamics in individualpersons, but cannot elucidate the extent to which suchpathways are shared between persons, e.g. [3,17-19]).As such, population-level studies may be particularlyuseful in identifying the HLA-restricted CTL escapeupon host and viral genotype data from early andchronic infection cohorts in North America, Europeand Australia (methods and [13,23-25]). Our final earlyand chronic datasets comprised 221 Gag, 203 Pol and219 Nef HIV-1 subtype B sequences per cohort, forwhich linked HLA class I types were available. Early cohortpatients were recruited a median of 88 [IQR 63–120] daysfollowing infection. All early and >75% of chronic patientswere antiretroviral naïve; the remainder were untreated attime of sampling.A total of 59 HLA class I alleles, classified at subtype-level (4-digit) resolution, were observed at a frequency >1%in the early and/or chronic cohorts; these comprised17 HLA-A, 23 HLA-B and 19 HLA-C alleles (Figure 1).Of these 59 alleles, the frequencies of 56 (94.9%) werecomparable between cohorts; only three alleles (HLA-A*02:06, A*30:02 and B*39:01) exhibited significantlymutations that are most rapidly and reproduciblyselected following HIV-1 infection.In an attempt to achieve these goals, we undertook aproof-of-concept study that compared the prevalence ofknown HLA-associated polymorphisms in HIV-1 Gag,Pol and Nef [11] in identically-sized cross-sectionalearly and chronic infection cohorts that were matchedas closely as possible for their HLA allele distributionsand their total HIV-1 diversity. Our main goals were:1) to assess the utility of population-level approaches toidentify the most reproducibly rapid escape mutations inHIV-1; 2) to estimate the extent of escape and reversionbetween early and chronic infection; and 3) to investigatewhether features related to population-level early im-mune escape signal can discriminate protective fromnon-protective HLA alleles.Results and discussionAssembling early and chronic infection cohorts matchedfor size, HLA and HIV-1 diversityOur study sought to demonstrate that the extent, re-producibility and relative timing (early versus later) ofHLA-driven escape in HIV-1 can be inferred via com-parative analysis of independent cross-sectional host/virus genotype datasets from different infection stages.This strategy ideally requires cross-sectional datasetsthat are identically powered with respect to host andviral genetic diversity (i.e. datasets that mimic longitu-dinal data as closely as possible, in that they differ onlywith respect to infection stage of the participants). Assuch, our first step was to assemble early and chronicHIV-1 subtype B cohorts of identical size that werematched as closely as possible for HLA class I alleledifferent frequencies between cohorts (all p < 0.05;0.08 < q < 0.33) (Figure 1A-C). As such, our early andMartin et al. Retrovirology 2014, 11:64 Page 3 of 16http://www.retrovirology.com/content/11/1/64chronic cohorts were generally well-matched with re-spect to host HLA diversity.HIV-1 Gag, Pol and Nef diversity was also generallycomparable between the two cohorts. Mean patristic(pairwise) genetic distances between HIV-1 sequencesin early versus chronic datasets, measured in units ofsubstitutions per nucleotide site, were 0.076 (StandardDeviation [SD] ± 0.011) versus 0.071 (SD ± 0.010) re-spectively for Gag (Figure 1D left and middle panels),0.057 (SD ± 0.008) versus 0.053 (SD ± 0.008) for Pol,Figure 1 Early and chronic datasets are comparable with respect to h(total 56) observed at frequencies > 1% in the early and/or chronic datasetsdatasets were comparable with respect to all HLA allele frequencies exceptthe early cohort compared to the chronic cohort (denoted by “*” for p < 0.HIV-1 polymorphisms restricted by these three HLA class I alleles were asseUnrooted maximum-likelihood phylogenies of early (left), chronic (middle)0.01 substitutions per nucleotide site. Mean patristic (pairwise) genetic distacohorts; moreover, no gross cohort-specific clustering is observed in the coand 0.119 (SD ± 0.018) versus 0.120 (SD ± 0.021) for Nef(not shown). Moreover, no gross inter-cohort segrega-tion was observed in a combined HIV-1 Gag phylogeny(Figure 1D, right), indicating that neither cohort wasdominated by large epidemiologically linked clustersnor exhibited evidence of recent descent from distinctancestors. Together, these data suggest that our earlyand chronic datasets are similarly powered with respectto host and viral genetic diversity, and thus differ onlywith respect to infection stage.ost and viral diversity. The 17 HLA-A, 23 HLA-B and 19 HLA-C allelesare displayed in Panels A-C, respectively. The early and chronicHLA-A*02:06, A*30:02 and B*39:01 whose frequencies were higher in05 and “**” for p < 0.01, Fisher’s exact test). Note however that nossed in the present study (see Figure 2 and Additional file 1). Panel D:and combined cohort (right) Gag sequences, on a distance scale ofnces between Gag sequences were comparable for early and chronicmbined phylogeny.Martin et al. Retrovirology 2014, 11:64 Page 4 of 16http://www.retrovirology.com/content/11/1/64Defining the list of HLA-associated polymorphisms forinvestigation in cohorts of the present size andcompositionA total of 453 HLA-associated polymorphisms in Gag/Pol/Nef had previously been identified at q < 0.05 in anindependent cohort of N > 1800 individuals with chronicHIV-1 subtype B infection [11], which contained no over-lap with the cohorts studied here. These HLA-associatedpolymorphisms comprise “adapted” associations (HIV-1amino acids significantly over-represented in the presenceof the HLA allele in question) as well as “nonadapted”associations (HIV-1 amino acids significantly under-represented in the presence of the HLA allele). For ex-ample, at Gag codon 242 the nonadapted amino acidassociated with HLA-B*57:01 is the subtype B consensus“T” whereas the B*57:01 adapted form is “N”, denoted as“B*57:01-Gag-T242N”. The cohort wherein these HLA-associations were originally defined however [11] wasmore than seven times larger than the cohorts presentlystudied. Therefore, we do not have sufficient statisticalpower to interrogate all of them in the present study. Assuch, our next step was to define, from the published list[11], the subset of known HLA-associated polymorphismsthat is appropriate for study in cohorts of the present sizeand host/viral genetic composition.Theoretically, if all immune escape mutations, once se-lected, persisted for the remainder of the host’s lifetime,and if we had achieved perfect genetic matching betweenour early and chronic cohorts, then we could define anappropriate subset of HLA-associated polymorphisms byinterrogating our chronic cohort for the presence of theseN = 453 known HLA-associated polymorphisms. Thosedetectable at the population level in chronic infection,a stage when a majority of within-host adaptation hasalready occurred, would represent an appropriate sub-set for study in cohorts of the present size. Thus wefirst interrogated our chronic cohort for the presenceof these 453 published HLA-associated polymorphismsusing statistical association with phylogenetic correc-tion (see Methods and [12,26]), and in doing so identified157 (35%) “adapted” and “nonadapted” HLA associationsat p < 0.01 (corresponding to q < 0.01 in this analysis)(Figure 2 and Additional file 1). These comprised 54, 52and 51 HLA-associated polymorphisms in Gag, Pol andNef respectively.However, if genetic matching of early and chronic co-horts was imperfect, and/or if immune escape muta-tions occasionally arose transiently during infection,then defining HLA-associated polymorphisms based onthe chronic cohort only could potentially lead us tomiss some associations. Thus, we interrogated our earlycohort for this same list of published HLA-associatedpolymorphisms using identical methods. In doing so weidentified an additional 5 HLA-associated polymorphismsat p < 0.01, two occurring at Gag codon 401 and one eachoccurring at Nef codons 38, 102 and 133, that had notbeen identified in the chronic analysis (Figure 2 andAdditional file 1).We therefore took the union of these results, totalingN = 162 HLA-associated polymorphisms, as our defin-ition of known HLA-associated polymorphisms appro-priate for cohorts of the present size and composition(Figure 2 and Additional file 1). To further validate thissubset, we applied a published phylogenetically-correctedinteraction test (see Methods and [12,27]) to compare thestrengths of selection of these individual HLA-associatedpolymorphisms in early versus chronic cohorts. Given thatwithin-host HIV-1 adaptation increases over the infectioncourse via the selection of immune escape mutations, wewould expect higher strengths of association between therestricting HLA and the HIV-1 polymorphism at later ver-sus earlier stages: indeed, of the 162 HLA-associated poly-morphisms studied, the strengths of 101 (62%) of themdiffered significantly (p < 0.01, q ≤ 0.1) between early andchronic infection (Additional file 1). Note also that noneof the N = 162 polymorphisms were restricted by the threeHLA alleles (A*02:06, A*30:02 and B*39:01) whose fre-quencies differed significantly between cohorts. All down-stream analyses therefore focused on this list of N = 162HLA-associated polymorphisms.Fifteen percent of known HLA-associated polymorphismsare already detectable at the population level in earlyinfectionOur first objective was to assess the extent of population-level signal for HLA-driven escape in early HIV-1 infec-tion. Of the N = 162 HLA-associated polymorphismsidentified for study, 24 (15%), occurring at 14 uniquecodons in Gag, Pol and Nef, were detectable at the popu-lation level in the early cohort at a threshold of p ≤ 0.01(Table 1). In total, these 24 associations comprised 16%of those investigated in Gag (9 of 56), 6% of those inves-tigated in Pol (3 of 52) and 22% of those investigated inNef (12 of 54).As expected, among these were escape mutationsknown or previously observed to occur in the first yearof infection, including B*57:01 Gag-T242N, B*51:01RT-I135X and Int-L28I, C*03:04 Nef-V85L, A*11:01Nef-K92R and A*24:02 Nef-Y135F [2,13,16,18,19,21,28].Our findings therefore provide proof-of-concept that themost consistently rapid host adaptations in HIV-1 can beidentified using cross-sectional methods. It is notable that,by < 3 months post-infection, the magnitude of statisticalassociation between certain HLA alleles and their associ-ated viral polymorphisms is already very high, and in somecases not significantly different from their magnitudesof association in chronic infection. For example, theOdds Ratio [OR] of association between B*57 andMartin et al. Retrovirology 2014, 11:64 Page 5 of 16http://www.retrovirology.com/content/11/1/64Gag-T242N is 33 in early infection (p = 3 × 10−9) com-pared to 151 (p = 5 × 10−16) in chronic infection, which,though stronger during the latter stage, does not representa statistically significant difference (inter-cohort com-parison p = 0.3) (Table 1). This observation underscoresFigure 2 HLA-associated polymorphisms detectable at the populationGag, Pol and Nef “immune escape maps” indicate the codon location, specHLA-associated polymorphisms detectable at the population level in cohoramino acids (those under-represented in the presence of the HLA allele) areHLA allele in question (and usually represent the subtype consensus residuthe HLA allele) are red; these represent the HLA-associated “escape variant”most cases both forms are detectable at the population level at a given p-B*57:01 – associated nonadapted and adapted forms, respectively), whereathreshold (e.g. at Gag codon 12, “E” represents the B*49:01-associated nonadaAsterisks (*) and italicized text denote the five HLA-associations at Gag codonsthe early cohort at p < 0.01, but were p≥ 0.01 in the chronic cohort. Note thaonly (the remainder of RT is colored gray). Subsequent analyses focused on ththe rapid and highly reproducible nature of certainHLA-driven within-host adaptations in HIV-1, where,for certain mutations such as Gag-T242N, escape(and by extension our ability to detect this associ-ation via population-level methods) is already nearlevel in cohorts of the present size and genetic composition.ific amino acid residues and HLA restrictions of the N = 162 knownts of the present size and host/viral genetic composition. “Nonadapted”blue; these represent the “immunologically susceptible” form for thee). “Adapted” amino acids (those over-represented in the presence of. Adapted and Nonadapted associations are counted independently; invalue threshold (e.g. at Gag codon 242, “T” and “N” represent thes in other cases only one of the two forms is detectable at a givenpted form but no specific adapted form is detected at this threshold).401 and Nef codons 38, 102 and 133 that were defined via detection int HIV-1 RT genotyping was performed for codons 1–400 of this proteinis list of HLA-associated polymorphisms.oprods18142otaona1:0ectMartin et al. Retrovirology 2014, 11:64 Page 6 of 16http://www.retrovirology.com/content/11/1/64Table 1 HLA-associated polymorphisms detectable at the pProtein HLA HIVpolymorphismaEarly infection ChOdds Ratiob p-valuec OdGag B*57:01 T242N 33 3 × 10−9 15C*07:04 M378x 3.6 0.006 3.5A*31:01 K397R 6.2 1 × 10−7 83A*31:01 I401L 5.1 0.005 1.9A*31:01 R403K 1.5 0.007 ~4Pol (RT) B*51:01 I135x 2.1 0.004 25Pol (Int) B*51:01 L28I 16 0.009 5Nef B*37:01 E38D 13 0.006 ~2C*03:04 V85L 2.4 0.002 4A*11:01 K92R 5.3 0.009 8.4C*03:04 H102x 3.2 0.001 ~1A*23:01 F143Y 8.4 0.01 85B*57:01 x133I 6 0.004 1.4A*24:02 Y135F 2.3 0.004 15aA total of 24 associations, occurring at 14 unique HIV-1 codons, are listed. The tcounted individually (e.g. B*57:01-Gag-T242N comprises two associations– the nform was not detected in early infection are denoted by a lowercase “x” (e.g. B*5early and chronic infection at p ≤ 0.01; those italicized represent associations detbmaximal in B*57:01-expressing persons < 3 monthspost-infection.Of note, HLA-associated HIV-1 polymorphisms withstrong early population-level escape signal also includedunderstudied viral sites. Notable among these wereA*31:01-associated polymorphisms at Gag codons 397,401 and 403, the first of which represented the secondstrongest p-value detected in the early cohort (OddsRatio = 6.2, p = 1 × 10−7, Table 1). These associationsare located within the novel A*31:01-restricted CR9CD8+ epitope originally characterized via detailed lon-gitudinal analysis of a single HIV-1 subtype B-infectedindividual [18,19]. By definition, population-level stud-ies identify viral adaptations that occur reproducibly inpersons expressing the restricting HLA; as such, thepresent results extend those of the original individual-level study [18] by indicating that escape within CR9 isboth rapid and highly consistent in HLA-A*31:01-ex-pressing persons. By extension, a lack of population-level early escape signal does not necessarily mean thata given site never escapes early: rather, it indicates thata given site does not reproducibly escape early (or atleast does not do so to an extent that achieves statis-tical significance in a dataset of the present size). Forexample, very rapid (<30 days) escape was previouslyWhere both nonadapted and adapted forms for a given HLA are identified, the macWhere both nonadapted and adapted forms for a given HLA are identified, the lowits nonadapted (E38x) form; the p-value for the adapted (x38D) form at this stage is*Bioinformatically predicted CTL epitopes are denoted by asterisks (*); the remainder aBold letters indicate the position within the epitope where the HLA-associated polymoin its C*03-adapted form.dFor each HLA-associated polymorphism in the table, its strength association in earphylogenetically-corrected interaction test (see Methods and [12,27]). The p-valuesulation level in early HIV-1 infectionnic infection CD8+ epitope* p-value(early vs. chronic)dRatiob p-valuec (HIV codon coordinates)6 × 10−17 TSTLQEQIGW (240–248) 0.30.0002 IMMQRGNF (377–383)* 0.78 × 10−10 CGKEGHIAR (395–403) 0.50.01 CGKEGHIAR (395–403) 0.21 × 10−7 CGKEGHIAR (395–403) 0.0025 × 10−11 TAFTIPSI (128–135) 0.00017 × 10−5 LPPIVAKEI (28–36) 0.20.002 LEKHGAIT (37–45)* 0.20.0002 AALDLSHFL (83–91) 0.64 × 10−6 AVDLSHFLK (84–92) 0.20.25 none 0.15 × 10−8 RYPLTFGWCF (134–143) 0.10.29 YTPGPGIRY (127–135) 0.22 × 10−21 RYPLTFGW (134–141) 0.0005l 24 is reached because HLA-associated nonadapted and adapted forms aredapted T and the adapted N). Cases where a specific non-adapted or adapted1-RT-I135x). Polymorphisms in bold represent associations detectable in bothable in only the early cohort with p < 0.01.documented within the A*01-restricted GY9 epitope(Gag codons 71–79) in two HIV-1 subtype C-infectedpersons [19], but no evidence of reproducible early escapeGag codon 79 in A*01-expressing persons was observed inour early dataset, suggesting that rapid A*01-driven escapeat this position is atypical in HIV-1 subtype B.Taken together, population-level analyses extend thoseof individual-level studies by identifying escape mutationsthat are rapidly and reproducibly selected across patients.Our observation that 15% of known HLA-associatedpolymorphisms, notably those in Nef and Gag, are alreadydetectable < 3 months post-infection, underscores the pre-dictable and rapid nature of HIV-1 adaptation despite eachindividual’s unique combination of host HLA and trans-mitted virus genetics. Further, the detection of substantialpopulation-level escape signal within unknown or under-studied CD8+ epitopes in HIV-1 (Table 1) argues for con-tinued efforts to map novel epitopes commonly targetedduring this critical infection stage.Can population-level approaches identify transient earlyescape pathways?Recent longitudinal studies have revealed that immuneescape is often characterized by the initial appearance oftransient mutant forms that often retain some ability toximum absolute Odds Ratio is shown.est p-value is shown. Note the chronic p-value for B*37:01-Nef-E38D refers to0.07.re published (http://www.hiv.lanl.gov/content/immunology/tables/tables.html).rphism occurs. Note the C*03-restricted AALDLSHFL epitope has been publishedly versus chronic infection was compared using a previously-describedof these comparisons are listed in this column.be targeted by existing (or de novo) CTL [18,29], whichthen drive the selection of more effective escape variantsthat ultimately become fixed within the host [3,17,18]. Ifsuch “transient” escape pathways are reproducible acrosshosts, we wondered whether population-level approachescould theoretically be used as exploratory tools to identifythem. If so, we reasoned that such transient escapes woulddisplay stronger population-level escape signal in earlycompared to chronic infection (since, in some persons,the early variant would be subsequently replaced with an-other, thereby reducing population-level signal in laterstages). Although our inter-cohort comparative analysisrevealed no HLA-associated polymorphisms that dis-played significantly stronger signal in early versus chronicinfection (Additional file 1 and data not shown), we werenevertheless intrigued by the five HLA-associated poly-morphisms in Gag and Nef that exhibited population-levelescape signal of p < 0.01 in our early cohort but p ≥ 0.01 inthe chronic cohort (Table 1), suggesting these as possibletransient escape pathways.Indeed, analysis of available longitudinal bulk plasmavariant at the earliest timepoint post-infection, that wassubsequently replaced by a non-adapted form (and/or amixture of the two) within a year of infection. In both ofthese cases the association is located at position 7within the epitope, which is consistent with transientearly escape mutations representing incomplete TCRrepertoire escape variants [18,29]. The idea that escapemutations, once selected, may not always persist for thelifetime of the host is also supported by within-host re-version of certain escape mutations in very advanceddisease [22]. We thus cautiously interpret the data tosuggest that cross-sectional approaches could theoreticallybe used to identify reproducible HLA-driven adaptationsthat represent “transient” early escape variants in some in-dividuals, though such findings would require validationin independent cohorts, as well as experimentally.Escape prevalence in early infection correlates withlongitudinal first-year escape ratesAnother objective was to investigate to what extentcross-sectional data could be used to infer the extentca1/02/1ancalMartin et al. Retrovirology 2014, 11:64 Page 7 of 16http://www.retrovirology.com/content/11/1/64HIV-1 RNA Nef sequences from seven B*57:01 express-ing individuals identified one case where an individualharbored the Nef-133I adapted mutation at the earliestsampled timepoint 30 days post-infection, which was re-placed by V at 86 days post-infection and then by a mix-ture of I/V at 228 days-post infection (Table 2). Similarly,analysis of available bulk plasma Gag sequences fromseven A*30:01 expressing persons identified one casewhere an individual harbored the adapted Gag-401 LTable 2 Examples of possible transient early HLA-driven esearly versus chronic infectionHLA-associated HIV-1 polymorphism Patient HLAA*31:01-Gag-I401L A0301/3101 B4403/3503 C040B*57:01-Nef-x133I A2402/2902 B4403/5701 C060A small number of HLA-associated polymorphisms, including A*31:01-Gag-I401Lversus chronic infection (see Table 1). Though these differences were not statistithey could represent potential examples of transient escape. In support of this hypoharbored the HLA-associated adapted variant at a given HIV-1 codon at the earliest(and/or a mixture of the two), consistent with transient early escape at these positioand time course of immune-driven HIV-1 adaptation. Assuch, we first wished to demonstrate that early escape fre-quencies calculated cross-sectionally predict rates of im-mune escape calculated longitudinally. Published first-yearrates of escape were available for 27 optimally-describedCD8+ T-cell epitopes [13] which contained one or moreHLA-associated polymorphisms investigated in the presentstudy. For example, the estimated first year escape ratefor the Gag-TW10 epitope (Gag240–249) is 38.36% perpe at HIV codons with stronger population-level signal inDays post-infection Bulk plasma seq. Adapted to HLA?401 143 L yes226 L yes309 I no485 I/L partial563 I no683 I/L partial601 30 I yes31 I yes60 I yes86 V no123 V no228 [I/V] partial361 [I/V] partial396 [I/V] partiald B*57:01-Nef-x133I, showed stronger population-level escape signal in earlyly significant (Table 1, last two columns), we nevertheless hypothesized thatthesis, the above table provides examples of two cases where a patienttimepoint post-infection, that subsequently give way to a non-adapted formns in these patients.person-month [13], while the prevalence of theGag-T242N mutation among B*57-expressing personsin our early dataset is 67% (Figure 3). As expected, lon-gitudinal first-year CD8+ epitope escape rates correlatedsignificantly with HIV-1 polymorphism prevalence amongpersons expressing the relevant HLA in our early infectiondataset (Pearson’s R = 0.68, p = 0.0001; Figure 3). Because~40% of the patients in the present early infection cohortwere included in the published longitudinal study [13], were-analyzed our data with these overlapping patients re-moved, and observed that the correlation remained strong(Pearson’s R = 0.55, p = 0.0035, not shown). This supportsHLA-associated escape mutation prevalence calculatedcross-sectionally at < 3 months post-infection as a reliablesurrogate marker of first year escape rates calculatedlongitudinally.Martin et al. Retrovirology 2014, 11:64 Page 8 of 16http://www.retrovirology.com/content/11/1/64Figure 3 Escape prevalence in early infection correlates withlongitudinal first-year escape rates. A total of 27 HLA-associatedpolymorphisms in Gag (orange), Pol (green) and Nef (purple)occurring within optimally defined CTL epitopes, for which first-yearepitope-specific rates of escape were previously published [13], wereinvestigated. A significant positive correlation is observed betweenthe proportion of persons expressing the restricting HLA andharboring the relevant polymorphism (“proportion escaped”) in earlyinfection and the published first-year epitope escape rate, providingproof-of-concept that the relative timecourse of early escape in HIV-1can be inferred using cross-sectional methods. In the case where agiven epitope contained multiple HLA-restricted polymorphic sites,the site exhibiting the maximum “proportion escaped” was used.For figure clarity, only a subset of well-known epitopes are labeledfor interest.Inferring the extent of host adaptation via comparativeanalysis of cross-sectional data from early and chronicinfectionWe next wished to use our cross-sectional early andchronic cohorts to quantify the extent of HLA-drivenescape occurring between these two infection stages.For this analysis, we specifically defined “escape” as thespecific adapted viral form associated with a given HLAallele at a given HIV-1 codon - for example, Gag 242 Nis the B*57:01-associated adapted form at this position.This adapted list comprised N = 74 HLA-associatedpolymorphisms (25, 24, and 25 in Gag, Pol and Nef re-spectively) (Figure 2 and Additional file 1). We calcu-lated the prevalence of each of these polymorphisms inpersons expressing the relevant HLA allele in our earlyversus chronic cohorts, thus allowing us to estimate theextent of within-host HIV-1 adaptation between thesetwo stages. Overall, the median “percentage escaped”(defined as the % of individuals expressing the relevantHLA and harboring the HIV-1 polymorphism of inter-est) was 23.8% [Interquartile range (IQR) 5.3-44.4%] inearly infection versus 55.1% [IQR 28.4-73.0%] in chronicinfection (p < 0.0001; Figure 4A). This indicates that, onaverage, escape prevalence in persons expressing therestricting HLA allele more than doubles between theseinfection stages. Breaking the analysis down by HIV-1protein, the median early versus chronic escape preva-lence was 23.5% [IQR 15.1-49.7%] vs. 55.6% [IQR 29.7-85.7%] in Gag, 11.3% [IQR 1.2-33.3%] vs. 50.5% [IQR25.2-69.2%] in Pol, and 31.3% [7.5-63.0%] vs. 54.6%[21.4-73.3%] in Nef (all p ≤ 0.001, not shown). This isconsistent with early escape occurring predominantly inGag and Nef [13,14,19], while escape in Pol is generallyslower but nevertheless approaches comparable levelsby chronic infection.Though summary statistics are informative, individualpolymorphisms differ widely in their timing and extent ofselection over the infection course. For this reason, detailson polymorphism prevalence in persons expressing vs. notexpressing the relevant HLA in early and chronic infec-tion, along with their statistical measures of association,are provided in the Additional file 1. We highlight someexamples here. First, for a substantial minority of poly-morphisms (notably those in Table 1), escape is rapid,reproducible and largely complete within < 3 monthspost-infection. For example, 67% of B*57:01-expressingindividuals already harbored Gag-242 N in early infec-tion, a proportion that increased to 83% in the chronicphase (Figure 4A). Noting that Gag-242 N frequency wasonly 5.8% among persons lacking B*57:01 in early infec-tion (Additional file 1), these results are consistent withescape having already occurred in over two-thirds of B*57-expressing persons by < 3 months post-infection [2,13],with an additional minority escaping somewhat later.ndl),c inththes tachinMartin et al. Retrovirology 2014, 11:64 Page 9 of 16http://www.retrovirology.com/content/11/1/64Figure 4 Estimated extent of escape and reversion between early a“adapted” (escaped form) HIV-1 polymorphisms investigated (N = 74 totathe relevant polymorphism (“proportion escaped”) in early versus chroniescape prevalence in persons expressing the restricting HLA allele moreHLA-associated “adapted” polymorphisms are broken down in terms ofdefined as population-level signal of p < 0.05 in early infection) as well a(“lower” vs. “higher”, where the former is defined as <30%). The size of ethe “pie slices” denote the breakdown of polymorphisms by HIV-1 proteWhile the prevalence of Gag-242 N is low in the gen-eral population (5.8% among B*57:01-negative personsand ~1% among persons lacking an allele belonging tothe B58 supertype), other polymorphisms are quiteprevalent in circulation, but are nevertheless signifi-cantly enriched among HLA-expressing persons in earlyinfection (Figure 4A). In this case, their high earlyprevalence is attributable to both frequent transmissionand reproducible early escape. For example, both theA*31:01-associated Gag-403 K and C*03:04-associatedNef-85 L polymorphisms are observed at >40% preva-lence in HIV-1 subtype B sequences, but their preva-lence is ~55% and ~72% respectively among personsexpressing the relevant HLA < 3 months post-infection(Odds Ratios 1.5 and 2.4 respectively, p < 0.01, Table 1and Additional file 1). The observation that population-level approaches are capable of detecting strong escapesignals despite high polymorphism background frequen-cies has previously been demonstrated in high-poweredchronic infection cohorts [11]; the present study extendsthis to demonstrate such signals can also be detected veryearly in infection, in more modestly-powered datasets.Overall, if one uses the original criterion of earlypopulation-level statistical signal of p ≤ 0.01 to defineHIV-1 sites that predominantly escape early, 15% (11of 74) of adapted polymorphisms fall into this category;statistics and examples of HLA-associated polymorphisms in each categoryHLA-associated adapted HIV-1 polymorphism in the absence of the restrictigenerally consistent with slow reversion of many transmitted escape mutaprotein: Gag (orange), Pol (green) and Nef (purple); those mentioned in thechronic infection. Panel A: For each of the specific HLA-associatedthe proportion of persons expressing the restricting HLA and harboringfection are depicted as linked pairs. The data indicate that, on average,an doubles between early and chronic infection. Panel B: The N = 74ir relative timeline of escape (“earlier” vs. “later”, where the former isheir relative prevalence/transmission frequency in the populationpie reflects the proportion of polymorphisms in each category, while(orange, green and purple for Gag, Pol and Nef respectively). Summaryusing a more liberal threshold of p < 0.05, this increasesto 21.6% (16 of 74) (Figure 4B).The remaining 78.4% (58 of 74) polymorphisms gener-ally reproducibly escape later than 3 months followinginfection (Figure 4B). It is important to note that laterescape can occur because CTL responses against theseregions generally arise later during infection (i.e. there isno immune pressure on these epitopes in early infection),or because CTL responses arise relatively early, but escapedoes not reproducibly occur rapidly in a significant pro-portion of individuals expressing the relevant HLA allele.Among these later-escaping polymorphisms are thosewhose population background (transmission) frequenciesare generally low, and those whose background fre-quencies are generally high. The B*51:01-associatedIntegrase-32I polymorphism at position 5 of the B*51-restricted LI9 epitope (Integrase28–36) provides an ex-ample of the former. In early infection, its frequency inB*51:01-expressing persons is 5%, not significantlydifferent from background, but this rises to 64% bychronic infection (Figure 4A and Additional file 1).The LI9 epitope is known to be consistently targeted inB*51-expressing persons early after infection [30,31].The observation that this epitope ultimately escapes viaInt-32I in >60% of B*51:01-expressing persons suggeststhis epitope is under strong, sustained and reproducibleare also provided. Panel C: The proportion of persons harboring anng HLA allele in early versus chronic infection is shown. The data aretions [7,14,34]. In all panels, polymorphisms are colored by HIV-1text are labeled.Martin et al. Retrovirology 2014, 11:64 Page 10 of 16http://www.retrovirology.com/content/11/1/64CD8+ T-cell pressure by B*51 in vivo, where delayed es-cape is likely explained by a combination of mutational/fitness constraints and both intra-individual (“vertical”)and inter-individual (“horizontal”) CD8+ T-cell immu-nodominance hierarchies [19,32,33].An example of a later-escaping polymorphism with highpopulation background frequency is B*44:02-Gag-312E. Lo-cated at position 7 of the B*44:02-restricted AW11 epitope(Gag306–317), it represents the HIV-1 subtype B consensusresidue at this codon. Its >60% frequency in both B*44:02and non-B*44:02-expressing persons in early infectionreflects its high transmission frequency, rather thanearly selection by B*44:02. Nevertheless, by chronic in-fection, 83% of B*44:02-expressing persons harboredGag-312E, consistent with later escape (Figure 4A). Afull categorization of HLA-associated “adapted” polymor-phisms in terms of “earlier” vs. “later” escaping (defined asearly p < 0.05 vs. p ≥ 0.05 respectively) and “lower” vs.“higher” background (estimated transmission) frequency(defined as <30% vs. p ≥ 30% respectively), is provided inthe Additional file 1. A graphic depicting the proportionof HLA-associated polymorphisms in each of these cat-egories, broken down by HIV protein, is provided inFigure 4B.The extent of reversion of HLA-associated polymor-phisms over time can be similarly estimated by calculat-ing their prevalence in persons lacking the relevant HLAin the early versus chronic cohorts. The overall medianpercentage of individuals harboring a given polymorph-ism in the absence of the restricting HLA allele wascomparable in early (13.7% [IQR 4.9-34.2%]) and chronic(14.9% [4.2-34.9%]) infection (p = 0.15; Figure 4C), consist-ent with slow reversion reported for many polymorphisms[7,14,21,34]. Note that inferred reversion frequencies meritcautious interpretation in cases where polymorphisms areselected by multiple alleles (e.g. the seemingly stable preva-lence of Gag-147 L in A*25:01-negative individuals is likelydue in part to its selection by B*13:02 and B*57:01 [11], as-sociations that were not investigated in the present study).Nevertheless, results confirmed that HLA-B*57:01-Gag-T242N reverts between early and chronic infection(though RT-245E, Int-122I, Int-124N, or Nef-116N revertslowly or not at all, as reported previously [2,21,35]).The reversion analysis additionally revealed novel sitesof potential interest. For example, the early escapingA*31:01-Gag 397R polymorphism (Table 1 and [18,19])displayed evidence of reversion, suggesting that thismutation may have a high fitness cost.Host adaptation-related features distinguish protectiveand non-protective HLA class I allelesLastly, we wished to identify adaptation-related featuresthat discriminate protective from non-protective HLAalleles, defined here as their published hazard ratiosfor progression to AIDS [HR-AIDS] in natural historystudies [36]. Although the timecourse of viral escape isinfluenced by complex factors including epitope immuno-dominance hierarchies, strength of selection, mutational/fitness constraints and transmitted virus characteristics[17,19,34,37-40], we reasoned that HLA alleles that re-strict polymorphisms that are already highly prevalentin early infection (due to rapid escape and/or frequenttransmission) would be generally unfavorable for HIV-1control. Thus, for all HLA alleles for which ≥2 adaptedpolymorphisms were investigated in the present study(N = 17 alleles total), we computed their mean “percent-age escaped” in early infection. That is, we took theprevalence of each of these adapted polymorphisms inpersons expressing the relevant HLA allele in our earlycohort (displayed in Figure 4A), and, for each HLA al-lele, computed the mean of these values.Consistent with our hypothesis, we observed a positivecorrelation between an HLA allele’s average extent ofadaptation in early infection, and its HR-AIDS (Pearson’sR = 0.53, p = 0.028; Figure 5A). Of note, HLA-B*57:01appears as somewhat of an outlier, exhibiting higherthan expected escape prevalence in early infection givenits protective nature. We hypothesize that the reasonB*57 can maintain sustained HIV-1 control despite rapidescape in some epitopes (Table 1 and [2,13,16,18,28]) isbecause the early B*57-restricted CD8+ response oftensimultaneously targets more than one epitope, notablyin p24Gag [31,41-43], where escape is accompanied byfitness costs [44-46].HLA-associated polymorphisms identified at the popu-lation level mark viral sites under strong, reproduciblein vivo immune pressure by individual HLA alleles [11].We thus further hypothesized that HLA alleles for whichescape was substantial (i.e. selected in a high proportionof persons expressing the relevant HLA) but generallydelayed (i.e. selected post early-phase) would tend to bemore protective. As such, for the same set of HLA-associated polymorphisms we computed their fold-change in escape between early and chronic infection,and calculated the mean of these values per HLA allele.By this measure, alleles for which the majority of escapehad already occurred in early infection would exhibitlow subsequent fold-changes, whereas alleles selectingescape mutations that generally occurred later in infec-tion would exhibit fold-changes reflecting the extent ofselection pressure on these sites in persons expressingthe relevant HLA. Consistent with our hypothesis, weobserved an inverse correlation between an HLA allele’sHR-AIDS and its average fold-increase in escape in chronicversus early infection (Pearson’s R = −0.54, p = 0.025;Figure 5B). Of note, B*27 appears as an additional out-lier in this analysis, possibly due to escape in the criticalGag-KK10 epitope requiring nearly a decade in somerostigrocorshipigfros (Ple,Martin et al. Retrovirology 2014, 11:64 Page 11 of 16http://www.retrovirology.com/content/11/1/64individuals [32,47,48] due to its high fitness/mutationalbarrier [33,49].Overall, our findings are consistent with a high earlyburden of adaptation to host HLA (either via rapid escapeor frequent polymorphism transmission) as a correlate ofHLA-associated progression risk. Conversely, HLA allelesfrom which HIV-1 escape is substantial and reproducibleyet occurs on a delayed timescale appears to be a correlateFigure 5 Adaptation characteristics as correlates of HLA-associated pand HLA-C (green) alleles for which ≥2 adapted polymorphisms were inveRatio of progression to AIDS (x-axis) was derived from historic published seexpressing that allele and harboring the specific viral HLA-associated polymadapted polymorphisms investigated (y-axis). A significant positive relationsuggesting that in general, high early escape prevalence is a correlate of hfold-increase in escape in chronic versus early infection was calculatedsignificant inverse relationship is observed between these two variablealleles are those from which HIV-1 escape is substantial and reproducibof protection. Taken together with observations that pro-tective HLA alleles contribute substantially to the totalHIV-specific CD8+ response in early infection [31,38], thatthey impose broad selection pressures on HIV-1 [11]; andthat some CD8+ epitopes escape slowly despite sustainedCD8+ targeting [19,48], results suggest that the capacityto exert consistent, substantial and sustained pressure,ideally on multiple epitopes, from which the virus can onlyescape on a relatively delayed timescale, is a correlate ofprotection.Some limitations of our study merit mention. Firstly,our early and chronic datasets are relatively modestlypowered by association testing standards, so it was notpossible to examine all published HLA-associated poly-morphisms in HIV-1 subtype B. Secondly, due to thelack of information on duration of infection for chronicpatients, it is likely that the chronic cohort comprisedpatients at a range of infection stages. Inclusion of somechronic patients with less advanced infection couldunderestimate the extent of escape at this stage. Finally,although care was taken to match our early and chronicdatasets as closely as possible for HIV and HLA geneticdiversity and distribution, it is essentially impossible toachieve perfectly matched datasets. As such, we cannotConclusionsIn conclusion, our results provide proof-of-concept thatstatistical association approaches can be applied to cross-rule out small differences in statistical power between co-horts, and therefore should interpret candidate “transient”early escape results with some caution.gression risk. Colored dots denote individual HLA-A (red), HLA-B (blue)ated in the present study (N = 17 alleles total). Each HLA allele’s Hazardonverter studies [36]. Panel A: For each HLA, the proportion of personsphism in early infection was calculated as the mean of all HLA-associatedis observed between these two variables (Pearson’s R = 0.53, p = 0.028),her HLA-associated progression risk. Panel B: For each HLA, the meanm all HLA-associated adapted polymorphisms investigated (y-axis). Aearson’s R = -0.54, p = 0.025), suggesting that in general, protectiveyet occurs on a delayed timescale.sectional host/viral genetic datasets to identify the mostrapidly selected HLA-associated polymorphisms in HIV-1that are also reproducibly selected across patients. As such,results from these types of population-level studies com-plement those of individual-level longitudinal analyses thatcannot assess inter-patient reproducibility. Furthermore,the extent and relative timing of individual escape events(in terms of early versus later in the infection course) canalso be inferred from cross-sectional data. In particular, wedemonstrate that high escape prevalence in early infection(either due to rapid selection and/or high transmission fre-quency) is a correlate of HLA-associated progression riskwhile reproducible later escape (a surrogate of consistentimmune selection on a given site in persons expressingthe relevant HLA) is a correlate of protection.Given that longitudinal observational studies of untreatedpersons are incompatible with current recommendationsfor early HIV-1 treatment initiation [20] and treatment asprevention [50], cross-sectional analyses of pretreatmenthost/viral genotypes could provide relevant alternative toolsto advance our knowledge of HIV-1 adaptation, includingthe earliest events post-infection. We suggest that studiessuch as the present one be undertaken with expandedcross-sectional cohorts, comprised of individuals at variousMartin et al. Retrovirology 2014, 11:64 Page 12 of 16http://www.retrovirology.com/content/11/1/64clinical stages of infection, including from different HIV-1subtypes.MethodsEarly and chronic infection cohortsThe early cohort was comprised of HIV-1 subtype Binfected patients recruited through various observa-tional seroconverter studies including the Acute Infectionand Early Disease Research Program (AIEDRP) sites inBoston and New York (USA), Sydney (Australia), a privatemedical clinic in Berlin (Germany), and observationalcohort studies in Montreal and Vancouver (Canada)[13,23,24]. Infection dates for the patients in the earlycohort were estimated as described in [13,23]. Briefly,for patients with positive HIV RNA (>5,000 copies/ml)or detectable serum p24 antigen but a negative HIV-1enzyme immunoassay (EIA), 4 weeks were subtractedfrom the negative EIA date. For patients with positiveEIA but negative/indeterminate Western blot, 6 weekswere subtracted from the positive EIA date. For patientswith negative detuned EIA, 4 months were subtractedfrom this date. For the remainder, infection dates wereestimated as the midpoint between the last negative andthe first positive HIV test. Clinical histories were incor-porated into infection date estimates where available.To maximize our power to detect HLA-associatedpolymorphisms in the early infection stage, all availableearly infection patients were included in the presentstudy. For each of these patients, the sample closestto ~3-months following the estimated date of infectionwas selected, yielding a median sampling distribution of88 days [Interquartile Range 63–120 days] post-infectionfor early samples. In contrast, the chronic cohort was as-sembled from the baseline (pre-therapy) timepoint from atotal of more than 300 HIV-1 subtype B infected individ-uals initiating antiretroviral therapy in British Columbia,Canada, and untreated HIV-1 subtype B infected individ-uals in Boston, USA [13,25]. Time since infection is un-known for individuals in the chronic cohort, however themedian CD4 count at sampling was 250 [IQR 147–360]cells/mm3 for this group. To create HIV-1 gene-specificchronic infection datasets of equal size to the early cohort,that were also matched as closely as possible for HLAclass I and HIV-1 diversity of the early cohort, chronic pa-tients were selected from the total group using an iterativeprocess to achieve the closest matching of HIV-1 andHLA distributions (Figure 1). All early and >75% ofchronic patients were antiretroviral naïve; the remainderwere untreated at time of sampling.Ethics statementAll patients provided written informed consent. Ethicalapproval was obtained through the institutional reviewboards at the Massachusetts General Hospital, the BCCentre for Excellence in HIV/AIDS and Simon FraserUniversity.HIV-1 and host (HLA class I) genotypingHIV-1 RNA was extracted from plasma using standardmethods. Gag, Pol (including protease, codons 1–400of Reverse Transcriptase, and Integrase), and Nef wereamplified in separate nested RT-PCR reactions usingHIV-1 subtype B-specific primers. Amplicons were bulk-sequenced bidirectionally on a 3130xl and/or 3730xlautomated DNA sequencer (Applied Biosystems). Chro-matograms were analyzed using Sequencher v5.0 (Gene-codes) or RECall [51] with nucleotide mixtures called ifthe height of the secondary peak exceeded 25% of theheight of the dominant peak (Sequencher) or 20% of thedominant peak area (RECall). HIV-1 sequences were con-firmed as subtype B using the recombinant identificationprogram (RIP; http://www.hiv.lanl.gov/content/sequence/RIP/RIP.html) and aligned to the HIV-1 subtype B refer-ence strain HXB2. Phylogenetic trees were constructedusing PhyML [52] and visualized using FigTree (http://tree.bio.ed.ac.uk/software/figtree/). Pairwise genetic dis-tances were computed from newick treefiles usingPATRISTIC [53]. HLA class I typing was performed usingsequence-based methods [54] and imputed where neces-sary to high resolution using a machine learning algorithm([55]; http://research.microsoft.com/en-us/projects/bio/mbt.aspx#HLA-Completion). HIV-1 sequences from co-horts where REBs allow public sequence deposition havebeen deposited in GenBank: accession numbers are Gag(KJ869442 - KJ869609), Protease-RT (KJ869900 - KJ870015),Integrase (KJ869610 - KJ869735), Nef (KJ869736 - KJ869899).A full summary of polymorphism frequencies, brokendown by HLA allele carriage and infection stage isprovided as Additional file 1.Definition and identification of HLA-associatedpolymorphismsThe published reference list of N = 453 HLA-associatedpolymorphisms in HIV-1 subtype B Gag, Pol and Nefsequences was defined in an independent internationalcohort of >1800 individuals chronically infected withHIV-1 subtype-B using phylogenetically-informed methodsat q < 0.05 [11]. The cohort used to define these associa-tions had no overlap with the early and chronic cohortsstudied here [11]. Briefly, to identify HLA-associatedpolymorphisms in linked HIV/HLA datasets, maximumlikelihood phylogenetic trees (one per HIV-1 gene) areconstructed, and a model of conditional adaptation isinferred for each observed HIV-1 amino acid at eachcodon. The amino acid is assumed to evolve independ-ently along the tree until it reaches the tips, represent-ing the present host. Selection via host HLA-mediatedpressures and HIV-1 amino acid covariation is directlyMartin et al. Retrovirology 2014, 11:64 Page 13 of 16http://www.retrovirology.com/content/11/1/64modeled using a weighted logistic regression, in whichthe individual’s HLA repertoire and covarying HIV-1amino acids are used as predictors, and the bias is deter-mined by the inferred possible transmitted sequences(as inferred via reconstruction of the amino acid frequen-cies at the penultimate internal nodes in the phylogeny)[12]. Here, the null hypothesis is that the observed aminoacids at the tree tips are explained by the phylogeny only,whereas the alternative hypothesis is that they are betterexplained by the presence of a specific HLA (or covaryingHIV-1 amino acid) in the present host. To identify whichfactors (HLA and/or HIV-1 covariation) contribute tothe selection pressure, a forward selection procedure isemployed where the most significant association is addedto the model in an iterative fashion, with p-values com-puted using the likelihood ratio test. Statistical significanceis reported using q-values [56], the p-value analogue ofthe false discovery rate (FDR). Q-values denote the ex-pected proportion of false positives among resultsdeemed significant at a given p-value threshold; for ex-ample, at q ≤ 0.05, we expect 5% of identified associa-tions to be false positives.HLA-associated polymorphisms are classified into twocategories: (1) “Adapted forms”, amino acids significantlyoverrepresented in the presence of the HLA allele inquestion, which represent the putative escape forms asso-ciated with that HLA at that codon, and (2) “Nonadaptedforms”, amino acids significantly underrepresented in thepresence of the HLA allele in question, which representthe immunologically susceptible form associated with thatHLA at that codon. In most cases, HLA-associated nona-dapted forms represent the subtype consensus amino acidwhile adapted forms represent polymorphic variants – butexceptions exist.To identify an appropriate subset of known HLA-associated HIV-1 polymorphisms that are appropriatefor study in datasets of the present size (N ~ 200) andhost/viral genetic distribution, we interrogated our earlyand chronic infection cohorts for these N = 453 pub-lished polymorphisms [11] using the phylogenetically-corrected methods described above. As described in theresults, this yielded a subset of N = 162 HLA-associatedpolymorphisms detectable in our early and/or chronicdatasets (Figure 2 and Additional file 1).Our analyses also featured comparisons of thestrength of selection of HLA-associated polymorphismsbetween early and chronic cohorts, undertaken using apreviously-described phylogenetically-corrected inter-action test [12,22,27]. Briefly, we took the union of allHLA-associated polymorphisms detectable at the popu-lation level in either the early or chronic cohorts (N = 162).For each association on the list, we constructed aphylogenetically-corrected logistic regression model usingthe restricting HLA as a predictor. Using a likelihoodratio test, we compare this model to a more expressiveone that includes an additional interaction term that as-signs “1” if the individual expresses the restricting HLAallele and is in the chronic cohort, or “0” otherwise. Thisallows us to obtain a p-value testing the null hypothesisthat HLA-associated selection at that site is not signifi-cantly different in early versus chronic cohorts.Statistical analysesFisher’s exact test was used to compare HLA class I al-lele frequencies between cohorts. The Mann–Whitneypaired test was used to compare the prevalence of HLA-associated polymorphisms in the presence/absence oftheir restricting HLA, in early versus chronic cohorts, asthese data were non-normally distributed. Pearson’s cor-relation was used to investigate the relationship betweenearly escape prevalence and published first-year rates ofescape [13], as well as with published HLA allele-specificHazard Ratios for progression to AIDS [36], as thesedata did not significantly violate the assumption thatvalues were drawn from a normal distribution. In singleanalyses, significance is denoted by p < 0.05. In the caseof multiple tests, q-values are used [56]; thresholds aredefined throughout the paper. All tests of significancewere two-tailed.Additional fileAdditional file 1: A full summary of HLA-associated polymorphisms,their HIV genomic locations and directions of association (adaptedvs nonadapted), and their observed frequencies in early versuschronic infection in persons harboring vs. not harboring therestricting HLA allele are provided in columns A-L and O-T. P-valuesand q-values of association are also provided for early infection (ColumnsM-N) and chronic infection (Columns U-V). P- and q-values comparingthe strength of selection of individual HLA-associated polymorphismsbetween early and chronic stages are provided in Columns W-X. Finally,all HLA-associated “adapted” polymorphisms are categorized with respectto their relative timescale of escape (early vs. later; Column Y) as well astheir relative background frequency in the population (lower vs. higher;Column Z).Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsEM and JMC designed and executed data analyses. AQL, DRC, RM, MAR, andCN performed HLA class I and/or plasma HIV-1 genotyping. HJ, ADK, MM,TMA, MJM, MC, and MAW contributed patient specimens and/or data. ZLBconceived and designed the study, analyzed data, and wrote the manuscript.All authors read and approved the final manuscript.AcknowledgementsWe thank Colin Shen and Zhixing (Samuel) Tan for laboratory assistance. Wethank Chanson Brumme, Conan Woods and Daniel MacMillan for databaseassistance. We thank Richard Harrigan and Bruce D. Walker for data accessand mentorship. We thank Mark Brockman for helpful discussions.This work was supported by operating grants from the Canadian Institutesfor Health Research (CIHR) MOP-93536 and HOP-115700 to ZLB. The VIDUSand ACCESS projects are funded by the National Institute on Drug Abuse,NIH (RO1DA011591 and RO1DA021525). This project has been funded inMartin et al. Retrovirology 2014, 11:64 Page 14 of 16http://www.retrovirology.com/content/11/1/64whole or in part with federal funds from the Frederick National Laboratoryfor Cancer Research, under Contract No. HHSN261200800001E. The contentof this publication does not necessarily reflect the views or policies of theDepartment of Health and Human Services, nor does mention of tradenames, commercial products, or organizations imply endorsement by theU.S. Government. This Research was supported in part by the IntramuralResearch Program of the NIH, Frederick National Lab, Center for CancerResearch. EM was supported by a Master’s Scholarship from the CanadianAssociation of HIV Research and Abbott Virology. AQL is the recipient of aCIHR Frederick Banting and Charles Best Masters award. DRC was therecipient of a CIHR CANADA-HOPE fellowship. M-JM is supported bypost-doctoral fellowships from CIHR and the Michael Smith Foundation forHealth Research (MSFHR). ZLB is a recipient of a CIHR New InvestigatorAward and a MSFHR Scholar Award.The funding bodies played no role in the design, collection, analysis, orinterpretation of data, nor in the writing of the manuscript or the decision tosubmit it for publication.Author details1Faculty of Health Sciences, Simon Fraser University, Burnaby, BC, Canada.2British Columbia Centre for Excellence in HIV/AIDS, Vancouver, BC, Canada.3Microsoft Research, Los Angeles, CA, USA. 4KwaZulu-Natal Research Institutefor Tuberculosis and HIV, Nelson R. Mandela School of Medicine, Universityof KwaZulu-Natal, Durban, South Africa. 5Jessen-Praxis, Berlin, Germany. 6KirbyInstitute, University of New South Wales, Sydney, Australia. 7Aaron DiamondAIDS Research Center, The Rockefeller University, New York, NY, USA. 8RagonInstitute of MGH, MIT and Harvard University, Cambridge, MA, USA. 9Facultyof Medicine, University of British Columbia, Vancouver, BC, Canada. 10Cancerand Inflammation Program, Laboratory of Experimental Immunology, LeidosBiomedical Research Inc, Frederick National Laboratory for Cancer Research,Frederick, MD, USA. 11Lady Davis Institute, McGill University, Montreal,Canada.Received: 30 May 2014 Accepted: 24 July 2014Published: 29 August 2014References1. Goulder PJ, Watkins DI: HIV and SIV CTL escape: implications for vaccinedesign. Nat Rev Immunol 2004, 4:630–640.2. Leslie AJ, Pfafferott KJ, Chetty P, Draenert R, Addo MM, Feeney M, Tang Y,Holmes EC, Allen T, Prado JG, Altfeld M, Brander C, Dixon C, Ramduth D,Jeena P, Thomas SA, St John A, Roach TA, Kupfer B, Luzzi G, Edwards A,Taylor G, Lyall H, Tudor-Williams G, Novelli V, Martinez-Picado J, Kiepiela P,Walker BD, Goulder PJ: HIV evolution: CTL escape mutation and reversionafter transmission. Nat Med 2004, 10:282–289.3. Henn MR, Boutwell CL, Charlebois P, Lennon NJ, Power KA, Macalalad AR,Berlin AM, Malboeuf CM, Ryan EM, Gnerre S, Zody MC, Erlich RL, Green LM,Berical A, Wang Y, Casali M, Streeck H, Bloom AK, Dudek T, Tully D, NewmanR, Axten KL, Gladden AD, Battis L, Kemper M, Zeng Q, Shea TP, Gujja S,Zedlack C, Gasser O, et al: Whole genome deep sequencing of HIV-1reveals the impact of early minor variants upon immune recognitionduring acute infection. PLoS Pathog 2012, 8:e1002529.4. Crawford H, Prado JG, Leslie A, Hue S, Honeyborne I, Reddy S, van der StokM, Mncube Z, Brander C, Rousseau C, Mullins JI, Kaslow R, Goepfert P, AllenS, Hunter E, Mulenga J, Kiepiela P, Walker BD, Goulder PJ: Compensatorymutation partially restores fitness and delays reversion of escapemutation within the immunodominant HLA-B*5703-restricted Gagepitope in chronic human immunodeficiency virus type 1 infection.J Virol 2007, 81:8346–8351.5. Fryer HR, Frater J, Duda A, Roberts MG, Phillips RE, McLean AR: Modellingthe evolution and spread of HIV immune escape mutants. PLoS Pathog2010, 6:e1001196.6. Leslie A, Kavanagh D, Honeyborne I, Pfafferott K, Edwards C, Pillay T, HiltonL, Thobakgale C, Ramduth D, Draenert R, Le Gall S, Luzzi G, Edwards A,Brander C, Sewell AK, Moore S, Mullins J, Moore C, Mallal S, Bhardwaj N,Yusim K, Phillips R, Klenerman P, Korber B, Kiepiela P, Walker B, Goulder P:Transmission and accumulation of CTL escape variants drive negativeassociations between HIV polymorphisms and HLA. J Exp Med 2005,201:891–902.7. Schneidewind A, Brumme ZL, Brumme CJ, Power KA, Reyor LL, O'Sullivan K,Gladden A, Hempel U, Kuntzen T, Wang YE, Oniangue-Ndza C, Jessen H,Markowitz M, Rosenberg ES, Sekaly RP, Kelleher AD, Walker BD, Allen TM:Transmission and long-term stability of compensated CD8 escapemutations. J Virol 2009, 83:3993–3997.8. Cornelissen M, Hoogland FM, Back NK, Jurriaans S, Zorgdrager F, Bakker M,Brinkman K, Prins M, van der Kuyl AC: Multiple transmissions of a stablehuman leucocyte antigen-B27 cytotoxic T-cell-escape strain of HIV-1 inThe Netherlands. AIDS 2009, 23:1495–1500.9. Moore CB, John M, James IR, Christiansen FT, Witt CS, Mallal SA: Evidenceof HIV-1 adaptation to HLA-restricted immune responses at a populationlevel. Science 2002, 296:1439–1443.10. Brumme ZL, John M, Carlson JM, Brumme CJ, Chan D, Brockman MA,Swenson LC, Tao I, Szeto S, Rosato P, Sela J, Kadie CM, Frahm N, Brander C,Haas DW, Riddler SA, Haubrich R, Walker BD, Harrigan PR, Heckerman D,Mallal S: HLA-associated immune escape pathways in HIV-1 subtype BGag, Pol and Nef proteins. PLoS ONE 2009, 4:e6687.11. Carlson JM, Brumme CJ, Martin E, Listgarten J, Brockman MA, Le AQ, ChuiCK, Cotton LA, Knapp DJ, Riddler SA, Haubrich R, Nelson G, Pfeifer N, DezielCE, Heckerman D, Apps R, Carrington M, Mallal S, Harrigan PR, John M,Brumme ZL: Correlates of protective cellular immunity revealed byanalysis of population-level immune escape pathways in HIV-1. J Virol2012, 86:13202–13216.12. Carlson JM, Listgarten J, Pfeifer N, Tan V, Kadie C, Walker BD, Ndung'u T,Shapiro R, Frater J, Brumme ZL, Goulder PJ, Heckerman D: WidespreadImpact of HLA Restriction on Immune Control and Escape Pathways ofHIV-1. J Virol 2012, 86:5230–5243.13. Brumme ZL, Brumme CJ, Carlson J, Streeck H, John M, Eichbaum Q, BlockBL, Baker B, Kadie C, Markowitz M, Jessen H, Kelleher AD, Rosenberg E,Kaldor J, Yuki Y, Carrington M, Allen TM, Mallal S, Altfeld M, Heckerman D,Walker BD: Marked epitope- and allele-specific differences in rates ofmutation in human immunodeficiency type 1 (HIV-1) Gag, Pol, and Nefcytotoxic T-lymphocyte epitopes in acute/early HIV-1 infection. J Virol2008, 82:9216–9227.14. Duda A, Lee-Turner L, Fox J, Robinson N, Dustan S, Kaye S, Fryer H,Carrington M, McClure M, McLean AR, Fidler S, Weber J, Phillips RE, FraterAJ: HLA-associated clinical progression correlates with epitope reversionrates in early human immunodeficiency virus infection. J Virol 2009,83:1228–1239.15. Li B, Gladden AD, Altfeld M, Kaldor JM, Cooper DA, Kelleher AD, Allen TM:Rapid reversion of sequence polymorphisms dominates early humanimmunodeficiency virus type 1 evolution. J Virol 2007, 81:193–201.16. Yager N, Robinson N, Brown H, Flanagan P, Frater J, Fidler S, Weber J,Phillips R: Longitudinal analysis of an HLA-B*51-restricted epitope inintegrase reveals immune escape in early HIV-1 infection. AIDS 2013,27:313–323.17. Fischer W, Ganusov VV, Giorgi EE, Hraber PT, Keele BF, Leitner T, Han CS,Gleasner CD, Green L, Lo CC, Nag A, Wallstrom TC, Wang S, McMichael AJ,Haynes BF, Hahn BH, Perelson AS, Borrow P, Shaw GM, Bhattacharya T,Korber BT: Transmission of single HIV-1 genomes and dynamics of earlyimmune escape revealed by ultra-deep sequencing. PLoS ONE 2010,5:e12303.18. Goonetilleke N, Liu MK, Salazar-Gonzalez JF, Ferrari G, Giorgi E, Ganusov VV,Keele BF, Learn GH, Turnbull EL, Salazar MG, Weinhold KJ, Moore S, Letvin N,Haynes BF, Cohen MS, Hraber P, Bhattacharya T, Borrow P, Perelson AS,Hahn BH, Shaw GM, Korber BT, McMichael AJ: The first T cell response totransmitted/founder virus contributes to the control of acute viremia inHIV-1 infection. J Exp Med 2009, 206:1253–1272.19. Liu MK, Hawkins N, Ritchie AJ, Ganusov VV, Whale V, Brackenridge S, Li H,Pavlicek JW, Cai F, Rose-Abrahams M, Treurnicht F, Hraber P, Riou C, Gray C,Ferrari G, Tanner R, Ping LH, Anderson JA, Swanstrom R, Cohen M, Karim SS,Haynes B, Borrow P, Perelson AS, Shaw GM, Hahn BH, Williamson C, KorberBT, Gao F, Self S, et al: Vertical T cell immunodominance and epitopeentropy determine HIV-1 escape. J Clin Invest 2013, 123:380–393.20. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for theuse of antiretroviral agents in HIV-1-infected adults and adolescents. Bethesda(MD): Department of Health and Human Services (DHHS); 2013.21. Fryer HR, Frater J, Duda A, Palmer D, Phillips RE, McLean AR: CytotoxicT-lymphocyte escape mutations identified by HLA association favorthose which escape and revert rapidly. J Virol 2012, 86:8568–8580.22. Huang KH, Goedhals D, Carlson JM, Brockman MA, Mishra S, Brumme ZL,Hickling S, Tang CS, Miura T, Seebregts C, Heckerman D, Ndung'u T, WalkerB, Klenerman P, Steyn D, Goulder P, Phillips R, van Vuuren C, Frater J:Martin et al. Retrovirology 2014, 11:64 Page 15 of 16http://www.retrovirology.com/content/11/1/64Progression to AIDS in South Africa Is Associated with both Revertingand Compensatory Viral Mutations. PLoS One 2011, 6:e19018.23. Brockman MA, Chopera DR, Olvera A, Brumme CJ, Sela J, Markle TJ, MartinE, Carlson JM, Le AQ, McGovern R, Cheung PK, Kelleher AD, Jessen H,Markowitz M, Rosenberg E, Frahm N, Sanchez J, Mallal S, John M, HarriganPR, Heckerman D, Brander C, Walker BD, Brumme ZL: Uncommonpathways of immune escape attenuate HIV-1 integrase replicationcapacity. J Virol 2012, 86:6913–6923.24. Poon AF, McGovern RA, Mo T, Knapp DJ, Brenner B, Routy JP, Wainberg MA,Harrigan PR: Dates of HIV infection can be estimated for seroprevalentpatients by coalescent analysis of serial next-generation sequencingdata. AIDS 2011, 25:2019–2026.25. Cotton LA, Kuang XT, Le AQ, Carlson JM, Chan B, Chopera DR, Brumme CJ,Markle TJ, Martin E, Shahid A, Anmole G, Mwimanzi P, Nassab P, Penney KA,Rahman MA, Milloy MJ, Schechter MT, Markowitz M, Carrington M, WalkerBD, Wagner T, Buchbinder S, Fuchs J, Koblin B, Mayer KH, Harrigan PR,Brockman MA, Poon AF, Brumme ZL: Genotypic and Functional Impact ofHIV-1 Adaptation to Its Host Population during the North AmericanEpidemic. PLoS Genet 2014, 10:e1004295.26. Carlson JM, Brumme ZL, Rousseau CM, Brumme CJ, Matthews P, Kadie C,Mullins JI, Walker BD, Harrigan PR, Goulder PJ, Heckerman D: Phylogeneticdependency networks: inferring patterns of CTL escape and codoncovariation in HIV-1 Gag. PLoS Comput Biol 2008, 4:e1000225.27. Chikata T, Carlson JM, Tamura Y, Borghan MA, Naruto T, Hashimoto M,Murakoshi H, Le AQ, Mallal S, John M, Gatanaga H, Oka S, Brumme ZL,Takiguchi M: Host-specific adaptation of HIV-1 subtype B in the Japanesepopulation. J Virol 2014, 88:4764–4775.28. Turnbull EL, Baalwa J, Conrod KE, Wang S, Wei X, Wong M, Turner J,Pellegrino P, Williams I, Shaw GM, Borrow P: Escape is a more commonmechanism than avidity reduction for evasion of CD8+ T cell responsesin primary human immunodeficiency virus type 1 infection. Retrovirology2011, 8:41.29. Brackenridge S, Evans EJ, Toebes M, Goonetilleke N, Liu MK, di Gleria K,Schumacher TN, Davis SJ, McMichael AJ, Gillespie GM: An early HIVmutation within an HLA-B*57-restricted T cell epitope abrogates bindingto the killer inhibitory receptor 3DL1. J Virol 2011, 85:5415–5422.30. Tomiyama H, Sakaguchi T, Miwa K, Oka S, Iwamoto A, Kaneko Y, TakiguchiM: Identification of multiple HIV-1 CTL epitopes presented byHLA-B*5101 molecules. Hum Immunol 1999, 60:177–186.31. Altfeld M, Kalife ET, Qi Y, Streeck H, Lichterfeld M, Johnston MN, Burgett N,Swartz ME, Yang A, Alter G, Yu XG, Meier A, Rockstroh JK, Allen TM, JessenH, Rosenberg ES, Carrington M, Walker BD: HLA Alleles Associated withDelayed Progression to AIDS Contribute Strongly to the Initial CD8(+) TCell Response against HIV-1. PLoS Med 2006, 3:e403.32. Goulder PJ, Phillips RE, Colbert RA, McAdam S, Ogg G, Nowak MA,Giangrande P, Luzzi G, Morgan B, Edwards A, McMichael AJ, Rowland-JonesS: Late escape from an immunodominant cytotoxic T-lymphocyteresponse associated with progression to AIDS. Nat Med 1997, 3:212–217.33. Schneidewind A, Brockman MA, Yang R, Adam RI, Li B, Le Gall S, Rinaldo CR,Craggs SL, Allgaier RL, Power KA, Kuntzen T, Tung CS, LaBute MX, Mueller SM,Harrer T, McMichael AJ, Goulder PJ, Aiken C, Brander C, Kelleher AD, Allen TM:Escape from the dominant HLA-B27-restricted cytotoxic T-lymphocyteresponse in Gag is associated with a dramatic reduction in humanimmunodeficiency virus type 1 replication. J Virol 2007, 81:12382–12393.34. Herbeck JT, Rolland M, Liu Y, McLaughlin S, McNevin J, Zhao H, Wong K,Stoddard JN, Raugi D, Sorensen S, Genowati I, Birditt B, McKay A, Diem K,Maust BS, Deng W, Collier AC, Stekler JD, McElrath MJ, Mullins JI:Demographic processes affect HIV-1 evolution in primary infectionbefore the onset of selective processes. J Virol 2011, 85:7523–7534.35. Brumme ZL, Brumme CJ, Heckerman D, Korber BT, Daniels M, Carlson J,Kadie C, Bhattacharya T, Chui C, Szinger J, Mo T, Hogg RS, Montaner JS,Frahm N, Brander C, Walker BD, Harrigan PR: Evidence of Differential HLAClass I-Mediated Viral Evolution in Functional and Accessory/RegulatoryGenes of HIV-1. PLoS Pathog 2007, 3:e94.36. O'Brien SJ, Gao X, Carrington M: HLA and AIDS: a cautionary tale. TrendsMol Med 2001, 7:379–381.37. Liu Y, McNevin J, Cao J, Zhao H, Genowati I, Wong K, McLaughlin S,McSweyn MD, Diem K, Stevens CE, Maenza J, He H, Nickle DC, Shriner D,Holte SE, Collier AC, Corey L, McElrath MJ, Mullins JI: Selection on thehuman immunodeficiency virus type 1 proteome following primaryinfection. J Virol 2006, 80:9519–9529.38. Streeck H, Jolin JS, Qi Y, Yassine-Diab B, Johnson RC, Kwon DS, Addo MM,Brumme C, Routy JP, Little S, Jessen HK, Kelleher AD, Hecht FM, Sekaly RP,Rosenberg ES, Walker BD, Carrington M, Altfeld M: Human immunodeficiencyvirus type 1-specific CD8+ T-cell responses during primary infection aremajor determinants of the viral set point and loss of CD4+ T cells. J Virol2009, 83:7641–7648.39. Prince JL, Claiborne DT, Carlson JM, Schaefer M, Yu T, Lahki S, Prentice HA,Yue L, Vishwanathan SA, Kilembe W, Goepfert P, Price MA, Gilmour J,Mulenga J, Farmer P, Derdeyn CA, Tang J, Heckerman D, Kaslow RA, AllenSA, Hunter E: Role of transmitted Gag CTL polymorphisms in definingreplicative capacity and early HIV-1 pathogenesis. PLoS Pathog 2012,8:e1003041.40. Song H, Pavlicek JW, Cai F, Bhattacharya T, Li H, Iyer SS, Bar KJ, Decker JM,Goonetilleke N, Liu MK, Berg A, Hora B, Drinker MS, Eudailey J, Pickeral J,Moody MA, Ferrari G, McMichael A, Perelson AS, Shaw GM, Hahn BH,Haynes BF, Gao F: Impact of immune escape mutations on HIV-1 fitnessin the context of the cognate transmitted/founder genome. Retrovirology2012, 9:89.41. Borghans JA, Molgaard A, de Boer RJ, Kesmir C: HLA alleles associated withslow progression to AIDS truly prefer to present HIV-1 p24. PLoS ONE2007, 2:e920.42. Streeck H, Lichterfeld M, Alter G, Meier A, Teigen N, Yassine-Diab B, Sidhu HK,Little S, Kelleher A, Routy JP, Rosenberg ES, Sekaly RP, Walker BD, Altfeld M:Recognition of a defined region within p24 gag by CD8+ T cells duringprimary human immunodeficiency virus type 1 infection in individualsexpressing protective HLA class I alleles. J Virol 2007, 81:7725–7731.43. Brennan CA, Ibarrondo FJ, Sugar CA, Hausner MA, Shih R, Ng HL, Detels R,Margolick JB, Rinaldo CR, Phair J, Jacobson LP, Yang OO, Jamieson BD: EarlyHLA-B*57-restricted CD8+ T lymphocyte responses predict HIV-1 diseaseprogression. J Virol 2012, 86:10505–10516.44. Brockman MA, Schneidewind A, Lahaie M, Schmidt A, Miura T, Desouza I,Ryvkin F, Derdeyn CA, Allen S, Hunter E, Mulenga J, Goepfert PA, Walker BD,Allen TM: Escape and compensation from early HLA-B57-mediated cytotoxicT-lymphocyte pressure on human immunodeficiency virus type 1 Gag altercapsid interactions with cyclophilin A. J Virol 2007, 81:12608–12618.45. Martinez-Picado J, Prado JG, Fry EE, Pfafferott K, Leslie A, Chetty S,Thobakgale C, Honeyborne I, Crawford H, Matthews P, Pillay T, Rousseau C,Mullins JI, Brander C, Walker BD, Stuart DI, Kiepiela P, Goulder P: Fitness costof escape mutations in p24 Gag in association with control of humanimmunodeficiency virus type 1. J Virol 2006, 80:3617–3623.46. Crawford H, Lumm W, Leslie A, Schaefer M, Boeras D, Prado JG, Tang J,Farmer P, Ndung'u T, Lakhi S, Gilmour J, Goepfert P, Walker BD, Kaslow R,Mulenga J, Allen S, Goulder PJ, Hunter E: Evolution of HLA-B*5703 HIV-1escape mutations in HLA-B*5703-positive individuals and theirtransmission recipients. J Exp Med 2009, 206:909–921.47. Feeney ME, Tang Y, Roosevelt KA, Leslie AJ, McIntosh K, Karthas N, WalkerBD, Goulder PJ: Immune escape precedes breakthrough humanimmunodeficiency virus type 1 viremia and broadening of the cytotoxicT-lymphocyte response in an HLA-B27-positive long-term-nonprogressing child. J Virol 2004, 78:8927–8930.48. Gao X, Bashirova A, Iversen AK, Phair J, Goedert JJ, Buchbinder S, Hoots K,Vlahov D, Altfeld M, O'Brien SJ, Carrington M: AIDS restriction HLAallotypes target distinct intervals of HIV-1 pathogenesis. Nat Med 2005,11:1290–1292.49. Kelleher AD, Long C, Holmes EC, Allen RL, Wilson J, Conlon C, Workman C,Shaunak S, Olson K, Goulder P, Brander C, Ogg G, Sullivan JS, Dyer W, JonesI, McMichael AJ, Rowland-Jones S, Phillips RE: Clustered mutations in HIV-1gag are consistently required for escape from HLA-B27-restrictedcytotoxic T lymphocyte responses. J Exp Med 2001, 193:375–386.50. Montaner JS: Treatment as prevention: toward an AIDS-free generation.Top Antivir Med 2013, 21:110–114.51. Woods CK, Brumme CJ, Liu TF, Chui CK, Chu AL, Wynhoven B, Hall TA,Trevino C, Shafer RW, Harrigan PR: Automating HIV drug resistancegenotyping with RECall, a freely accessible sequence analysis tool. J ClinMicrobiol 2012, 50:1936–1942.52. Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, Gascuel O: Newalgorithms and methods to estimate maximum-likelihood phylogenies:assessing the performance of PhyML 3.0. Syst Biol 2010, 59:307–321.53. Fourment M, Gibbs MJ: PATRISTIC: a program for calculating patristicdistances and graphically comparing the components of geneticchange. BMC Evol Biol 2006, 6:1.54. Cotton LA, Rahman MA, Ng C, Le AQ, Milloy MJ, Mo T, Brumme ZL: HLAclass I sequence-based typing using DNA recovered from frozen plasma.J Immunol Methods 2012, 382:40–47.55. Listgarten J, Brumme Z, Kadie C, Xiaojiang G, Walker B, Carrington M,Goulder P, Heckerman D: Statistical resolution of ambiguous HLA typingdata. PLoS Comput Biol 2008, 4:e1000016.56. Storey JD, Tibshirani R: Statistical significance for genomewide studies.Proc Natl Acad Sci U S A 2003, 100:9440–9445.doi:10.1186/s12977-014-0064-1Cite this article as: Martin et al.: Early immune adaptation in HIV-1revealed by population-level approaches. Retrovirology 2014 11:64.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionMartin et al. Retrovirology 2014, 11:64 Page 16 of 16http://www.retrovirology.com/content/11/1/64Submit your manuscript at www.biomedcentral.com/submit

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0224006/manifest

Comment

Related Items