UBC Faculty Research and Publications

Lack of a significant impact of Gag-Protease-mediated HIV-1 replication capacity on clinical parameters… Sakai, Keiko; Chikata, Takayuki; Brumme, Zabrina L; Brumme, Chanson J; Gatanaga, Hiroyuki; Oka, Shinichi; Takiguchi, Masafumi Nov 19, 2015

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

Item Metadata


52383-12977_2015_Article_223.pdf [ 1.76MB ]
JSON: 52383-1.0347312.json
JSON-LD: 52383-1.0347312-ld.json
RDF/XML (Pretty): 52383-1.0347312-rdf.xml
RDF/JSON: 52383-1.0347312-rdf.json
Turtle: 52383-1.0347312-turtle.txt
N-Triples: 52383-1.0347312-rdf-ntriples.txt
Original Record: 52383-1.0347312-source.json
Full Text

Full Text

Sakai et al. Retrovirology  (2015) 12:98 DOI 10.1186/s12977-015-0223-zRESEARCHLack of a significant impact of Gag-Protease-mediated HIV-1 replication capacity on clinical parameters in treatment-naive Japanese individualsKeiko Sakai1*, Takayuki Chikata1, Zabrina L. Brumme3,4, Chanson J. Brumme3, Hiroyuki Gatanaga1,5, Shinichi Oka1,5 and Masafumi Takiguchi1,2,6Abstract Background: HLA class I-associated escape mutations in HIV-1 Gag can reduce viral replication, suggesting that associated fitness costs could impact HIV-1 disease progression. Previous studies in North American and African cohorts have reported reduced Gag-Protease mediated viral replication capacity (Gag-Pro RC) in individuals express-ing protective HLA class I alleles including HLA-B*57:01, B*27:05, and B*81:01. These studies also reported significant positive associations between Gag-Pro RCs and plasma viral load (pVL). However, these HLA alleles are virtually absent in Japan, and the importance of Gag as an immune target is not clearly defined in this population.Results: We generated chimeric NL4-3 viruses carrying patient-derived Gag-Protease from 306 treatment-naive Japa-nese individuals chronically infected with HIV-1 subtype B. We analyzed associations between Gag-Pro RC and clinical markers of HIV-1 infection and host HLA expression. We observed no significant correlation between Gag-Pro RC and pVL in Japan in the overall cohort. However, upon exclusion of individuals expressing Japanese protective alleles HLA-B*52:01 and B*67:01, Gag-Pro RC correlated positively with pVL and negatively with CD4 T-cell count. Our results thus contrast with studies from other global cohorts reporting significantly lower Gag-Pro RC among persons expressing protective HLA alleles, and positive relationships between Gag-Pro RC and pVL in the overall study populations. We also identified five amino acids in Gag-Protease significantly associated with Gag-Pro RC, whose effects on RC were confirmed by site-directed mutagenesis. However, of the four mutations that decreased Gag-Pro RC, none were asso-ciated with reductions in pVL in Japan though two were associated with lower pVL in North America.Conclusions: These data indicate that Gag fitness does not affect clinical outcomes in subjects with protective HLA class I alleles as well as the whole Japanese population. Moreover, the impact of Gag fitness costs on HIV-1 clinical parameters in chronic infection is likely low in Japan compared to other global populations.Keywords: HIV-1, Replication capacity, Gag, Fitness© 2015 Sakai et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.BackgroundWithin an infected individual, immune pressure by cyto-toxic T lymphocytes (CTLs) induces the selection of HLA-associated escape variants in the HIV-1 genome. The persistence of HLA-associated escape variants upon transmission in turn shapes viral diversity at the popula-tion level [1]. CTL escape occurs throughout infection. Moreover, certain escape mutations compromise viral fitness. In some of these cases, their effects on immune evasion can be offset in part by their fitness costs though the effects of such mutations on plasma viral load (pVL) are complex and remain incompletely understood [2–9]. HIV-1 Gag is rich in immunodominant CTL epitopes, and a vital role of anti-Gag CTL responses on HIV-1 Open Access*Correspondence:  sakai001@kumamoto-u.ac.jp 1 Center for AIDS Research, Kumamoto University, Kumamoto 860-0811, JapanFull list of author information is available at the end of the articlePage 2 of 14Sakai et al. Retrovirology  (2015) 12:98 immune control has been demonstrated [10, 11; also reviewed in 12, 13]. In particular, associations between fitness-costly Gag escape mutations and decreases in pVL have been described [2, 5–8], notably for the immuno-dominant HLA-B*57:01-restricted TW10 and B*27:05-restricted KK10 epitopes, both of which are well studied because of their protective roles in HIV-1 disease out-come. Specifically, CTL-driven escape mutations in TW10 impose a mild fitness cost that, along with robust HLA-B*57:01-driven CTL responses, likely contribute to the significantly lower pVL observed in these subjects [5, 7, 14–17]. Notably however, fitness costs of escape in HLA-B*57:01 epitopes are generally rescued by the subsequent appearance of compensatory mutations [18, 19]. In contrast, a single amino acid change at Gag 264 (R264K) in the B*27:05-restricted KK10 epitope substan-tially compromises in vitro viral replication [20–22], but this mutation generally occurs late in infection due to the requirement for the upstream compensatory mutation S173A [7, 21]. As such, the appearance of R264K in vivo generally heralds HIV-1 progression [23, 24]. These stud-ies support a relationship between Gag escape mutations and pVL and underscore a critical role of Gag in HIV-1 disease outcome.In vitro studies also highlight the importance of Gag as an immune target. Certain HLA-associated mutations in Gag have been shown to impair Gag-Protease-mediated viral replication capacity in vitro (Gag-Pro RC). Gag-Pro RC is in turn positively associated with pVL in numer-ous global cohorts [18, 25–30]. In particular, reduced Gag-Pro RC has been observed in predominantly North American cohorts of elite controllers as well as individu-als in acute/early infection who express protective alleles B*13, B*27, B*57, and/or B*58:01 [18, 26], though the latter associations are less apparent during the chronic phase [18]. In addition, the impact of HLA-associated changes in Gag-Pro RC on pVL have also been reported in African cohorts chronically infected with HIV-1 subtype C [27–29]. A more recent study compared Gag-Pro RCs in treatment-naive Mexican and Barbadian HIV-1 sub-type B cohorts which differ in their frequency of protec-tive HLA alleles (8 % versus 34 % of B*27/57/58:01/81:01 in Mexico versus Barbados). The study revealed a higher burden of HLA-mediated Gag escape mutations and sig-nificantly lower Gag-Pro RC in chronically infected Bar-badian compared to Mexican individuals [30]. Moreover, Gag-Pro RC correlated positively with pVL, supporting the importance of Gag in HIV-1 disease outcome [30]. The significantly lower Gag-Pro RC and pVL in Barbados versus Mexico was attributed to the higher prevalence of protective HLA alleles in the former cohort, implying that protective alleles lower Gag-Pro RC, viral set points, and therefore, disease progression rates at the population level. These results are consistent with a previous study of nine distinct cohorts worldwide that demonstrates HIV-1 adaptation to HLA class I alleles at a population level, presumably as a result of transmission and persistence of certain escape mutations in the population [1]. Thus, HLA allele frequency shapes viral diversity, which could also affect CTL responses and HIV-1 disease outcome in the population.HLA class I distributions differ markedly in Asian com-pared with North American and African cohorts [31]. The “classical” protective alleles B*27:05 and B*57:01 are essentially absent in Japan: a published cohort of 504 treatment-naive HIV+ Japanese featured only one B*27:05+ and no B*57:01 patients [31–37]. Instead, B*52:01 and B*67:01 are associated with HIV-1 control in Japan [32]. Therefore, the role of Gag fitness in HIV-1 pathogenesis, or more broadly the mechanism of pro-tection, is likely to be different in Asian populations. As such, the impact of HLA-mediated immune pressure on viral fitness and its influence on HIV-1 pathogenesis need to be specifically defined for Asian populations. To address these issues, we investigated the effect of HLA-associated changes in Gag-Pro RC in 306 treatment-naive HIV+ Japanese patients chronically infected with HIV-1 subtype B in the presence or absence of protective HLA alleles. We also identified amino acid changes in Gag-Protease that altered Gag-Pro RCs and examined their associations with pVL. In contrast to studies of non-Asian cohorts that supported a role of Gag fitness in HIV-1 disease outcome, our results suggested that Japa-nese protective alleles did not lower Gag-Pro RC and that Gag fitness did not correlate with clinical markers in the whole Japanese population.ResultsRelationship between Gag‑Protease‑dependent replication capacity and clinical markers of disease progressionWe generated recombinant HIV-1NL4-3 containing Gag-Protease from 306 treatment-naïve Japanese subjects chronically infected with HIV-1 subtype B. Plasma viral sequences were determined for all patients in a previ-ous study [31], and the patient origin of all recombinant viruses was confirmed by maximum likelihood phyloge-netic analysis (Additional file  1). Recombinant viruses were used to infect a reporter CD4+ T-cell line, CEM-GXR, which naturally expresses CXCR4 and has been further engineered to express CCR5 and Tat-inducible GFP [38]. Gag-Protease-dependent replication capacities (Gag-Pro RCs) were determined based on the exponen-tial increase in the  % GFP+ cells during days 2–8 follow-ing infection and normalized against that of the parental HIV-1NL4-3 as described for North American and African studies [18, 25–30]. Similar to previous reports, Gag-Pro Page 3 of 14Sakai et al. Retrovirology  (2015) 12:98 RCs were normally distributed (Fig. 1a), with a mean of 1.12, a standard deviation (SD) of 0.13, and a range of 0.67–1.56. The mean Gag-Pro RC in the Japanese cohort was slightly higher than that observed in the North American cohort [18] but comparable to those reported more recently for an ART-naïve Mexican cohort [30].To examine the impact of Gag-Pro RC on HIV-1 patho-genesis, we investigated the relationship between Gag-Protease RC and clinical markers of HIV-1 infection (pVL and CD4 T-cell count). Overall, Japanese Gag-Pro RCs did not correlate with pVL (R = 0.092, P = 0.11; Fig. 1b) nor CD4 T-cell count (R = −0.078, P = 0.17; Fig. 1c). Not only did these results not achieve statistical significance, but the correlation coefficients for pVL and Gag-Pro RC in Japan were weaker than those reported in African (R  =  0.24) [28, 29] and Mexican/Barbadian (R  =  0.25) [30] chronic cohorts. The Japanese correlation coefficient R =  0.092 was also weaker to that reported in a North American chronic cohort (R  =  0.27 in the presence of protective HLA alleles; R = 0.12 for the total population) [18]. Together, this result suggested a minimal role for Gag fitness as a major determinant of virologic control in Japan compared to other global cohorts.Impact of HLA class I alleles on Gag‑Pro RCs and relationship with clinical parametersTo investigate the role of Gag as an HLA-mediated immune target, we stratified HLA class I alleles based on their median RCs (Additional file  2A-C). For all alleles with frequencies >3 % (n > 8) in the cohort, we did not Fig. 1 Gag-mediated replication capacities in a treatment naive Japanese cohort chronically infected with HIV-1 subtype B. a Distribution of Gag-Protease replication capacities. Recombinant HIV-1NL4-3 was constructed, using plasma-derived gag-protease from treatment-naïve HIV+ Japanese patients, and replication capacity for each virus was determined by in vitro infection. The values were normalized against mean growth rate of wild-type NL4-3. The samples were measured in triplicate. b, c Replication capacities and patients’ clinical parameters. Replication capacities of chimeric viruses were plotted against plasma viral load (b) or CD4 count (c) at the time of sample collection. Correlations were determined by Spearman’s rank correlation testPage 4 of 14Sakai et al. Retrovirology  (2015) 12:98 find any significant associations between HLA alleles and Gag-Pro RC. In addition, for each HLA allele, we com-pared the relative rankings of median RCs amongst the HLA alleles with those previously determined for pVL [32]. If Gag-Pro RC affects pVL, HLA alleles with lower Gag-Pro RCs would show a tendency toward lower pVL and vice versa. However, the relative RC ranking (in Additional file  2A-C) and pVL for each HLA allele did not match in most cases. We previously reported that A*33:03, B*44:03, B*40:06, C*14:03, and C*12:02 were associated with lower pVL in Japan [32], but none of these alleles were associated with lower Gag-Pro RC (Additional file  2A-C). Notably, the median Gag-Pro RCs in individuals expressing Japanese protective alleles HLA-B*52:01 and B*67:01 [32] were higher than the pop-ulation median (Additional file 2B), indicating that pro-tective-allele expression is not associated with lower Gag fitness in Japan. Indeed, no significant correlation was observed between median Gag-Pro RCs and median pVL for each HLA allele (Additional file 2D).To further explore the relationship between HLA, Gag-Pro RC, and HIV-1 clinical parameters, we grouped our cohort by HLA-allele expression and analyzed associa-tions between Gag-Pro RCs and HIV-1 clinical param-eters in each HLA group (Table 1). In two HLA-A alleles (A*02:01 and A*31:01), one B allele (B*07:02), and two C alleles (C*08:01 and C*08:03), Gag-Pro RC exhibited a positive correlation with pVL (p < 0.05), whereas Gag-Pro RC exhibited negative correlations with CD4 count at this threshold in seven alleles (HLA-A*02:06, HLA-B*46:01, B*48:01, B*59:01, C*01:02, C*08:01, and C*08:03; Table 1). However, only in HLA-C*08:01+ (n = 34) and C*08:03+ (n = 10) subjects, Gag-Pro RC correlated with both clini-cal parameters: a positive correlation with pVL (R = 0.36 and R  =  0.70, respectively) and an inverse correlation with CD4 count (R = −0.46 and R = −0.69, respectively; Fig. 2). We also examined groups of patients based on non-expression of particular HLA alleles. We observed significant correlations between clinical parameters and Gag-Pro RCs in the absence of eight HLA alleles (Table  1), including HLA-B*52:01. In individuals lack-ing the Japanese protective B*52:01 allele, Gag-Pro RCs correlated positively with pVL (R  =  0.15, P  =  0.021, q = 0.24, n = 234) and inversely with CD4 T-cell count (R = −0.19, P = 0.005, q = 0.16; Fig. 2). Similarly, Gag-Pro RC also showed significant associations with both pVL and CD4 count in individuals lacking HLA-C*12:02 (Fig.  2). HLA-B*52:01 and C*12:02 form a haplotype in Japan [31–36]. Indeed, all B*52:01+ subjects had C*12:02, and of 85 C*12:02+ subjects in our cohort, only three did not carry B*52:01. Positive correlations between Gag-Pro RC and pVL were additionally observed in the absence of A*11:01, A*33:03, B*39:01, B*44:03, C*07:02, and C*14:03 (Table 1), but none of these alleles associated with CD4 count. Taken together, our results stand in contrast to observations of lower Gag-Pro RC among individuals expressing classical protective HLA alleles in other global cohorts [18, 25–30]. HIV-1 control by protective HLA alleles is not associated with low Gag fitness in Japan.Table 1 Associations of Gag-Pro RCs with clinical parameters in each HLA class I alleleItalic HLA alleles have associations with both pVL and CD4 countn number of subjects, R Spearman’s rank correlationa HLA+ and HLA− indicate subjects with or without the particular HLA allele, respectivelyAssociations with pVL Associations with CD4 countHLA allele HLA + or −a n R P value q value HLA allele HLA + or −a n R P value q valueA*02:01 + 63 0.26 0.0428 0.243 A*02:06 + 61 −0.29 0.0231 0.279A*31:01 + 48 0.31 0.0357 0.238 B*46:01 + 31 −0.41 0.0264 0.279B*07:02 + 24 0.46 0.0288 0.238 B*48:01 + 14 −0.65 0.0194 0.279C*08:01 + 34 0.36 0.0377 0.238 B*59:01 + 14 −0.59 0.0344 0.318C*08:03 + 10 0.70 0.0365 0.238 C*01:02 + 92 −0.28 0.0082 0.163A*11:01 − 252 0.14 0.0277 0.238 C*08:01 + 34 −0.46 0.0088 0.163A*33:03 − 265 0.14 0.0288 0.238 C*08:03 + 10 −0.69 0.0399 0.328B*39:01 − 285 0.12 0.0385 0.238 B*52:01 − 234 −0.19 0.0045 0.163B*44:03 − 266 0.14 0.0244 0.238 C*12:02 − 231 −0.20 0.0024 0.163B*52:01 − 234 0.15 0.0209 0.238C*07:02 − 233 0.14 0.0325 0.238C*12:02 − 231 0.16 0.0135 0.238C*14:03 − 267 0.14 0.0251 0.238Page 5 of 14Sakai et al. Retrovirology  (2015) 12:98 Impact of Gag on clinical parameters in the absence of protective allelesIn our cohort, Gag-Pro RCs did not correlate with pVL in subjects carrying either of the Japanese protec-tive alleles, B*52:01 or B*67:01. However, exclusion of B*52:01 (n  =  234) resulted in associations of Gag-Pro RCs with clinical markers of infection (R  =  0.15, P = 0.021, q = 0.24 for pVL; Table 1). Therefore, we next examined correlations between Gag-Pro RCs and clini-cal parameters in the absence of these protective alleles (i.e. among B*52:01 and B*67:01 double-negative sub-jects; n = 224). In persons lacking both of these protec-tive alleles, Gag-Pro RCs exhibited a positive association with pVL (R = 0.16, P = 0.017) and a negative association with CD4 count (R = −0.21, P = 0.0019) (Fig. 3a). These associations are weaker than those observed in Mexi-can/Barbadian, African, and North American individu-als expressing protective HLA alleles (R  >  0.2 for these cohorts) but comparable in magnitude to the relation-ship between Gag-Pro RC and pVL in the North Ameri-can chronic cohort (R =  0.12, P =  0.0007) [18, 29, 30]. Indeed, after excluding B*52/B*67/C*12 alleles, the asso-ciations of Gag-Pro RC and clinical parameters became slightly stronger in each HLA allele- group analyzed in Table  1 (Additional file  3). These observations further support our result that Gag fitness does not associate with clinical parameters in subjects with protective HLA alleles in Japan.When we further explored the relationship between Gag-Pro RC and HIV clinical parameters in the B*52:01−B*67:01− subpopulation by multivariable analyses, a linear regression model adjusting for pVL and CD4 count indicated that Gag-Pro RC in this sub-set was driven largely by CD4 count (P = 0.043; β Esti-mate  =  −0.012  per  100  cell/mm3 increment) and not pVL (P = 0.47; β Estimate = 0.0091 per log10 increment). In other words, the model indicated that, after control-ling for pVL, Gag-Pro RC decreased by 0.012 units for each increment of 100  CD4  cells/mm3. Although the model identified CD4 count as an independent predictor of Gag-Pro RC, CD4 count explains only 3.7 % of popula-tion variation in Gag-Pro RC (Table 2). In another linear regression model additionally incorporating three HLA alleles—C*08:01, C*08:03, and C*12:02—that exhibited significant associations of Gag-Pro RC with both pVL and CD4 count in the B*52:01−B*67:01− subpopulation, the addition of HLA information to the model did not improve the model fit (P = 0.12). Taken together, results of the multivariable analyses suggest that apparent HLA associations with Gag-Pro RC, HIV clinical parameters, Fig. 2 Replication capacities for HLA alleles and their associations with clinical parameters. The presence (HLA-C*08:01 and C*08:03) or absence (HLA-B*52:01 and C*12:01) of four HLA alleles showed significant associations with pVL (top panels) as well as CD4 count (bottom panels). The asso-ciations were examined by Sperman’s rank correlation testPage 6 of 14Sakai et al. Retrovirology  (2015) 12:98 and HLA alleles are likely to be driven primarily by CD4 count rather than pVL in subjects lacking the protective B*52 and B*67 alleles. Moreover, these observations raise the possibility that Gag-Pro RC may not have a major impact on pVL even in the B*52−B*67− subpopulation.We next analyzed the impact of HLA associated poly-morphisms on pVL in 218 B*52:01− B*67:01− subjects based on the HLA associated polymorphisms deter-mined for the Japanese population [31]. Here, HLA-associated polymorphisms were defined based on a published list derived from plasma virus sequences from the present cohort (see methods and [31]; these Fig. 3 The relationship between HLA-associated-polymorphisms and Gag-Protease-mediated replication capacities in the HLA-B*52:01−B67:01− population. a Associations of replication capacities with plasma viral load and CD4 count in the B*52:01−B*67:01− population were tested by Spear-man’s rank correlation test. b–g An overall number of HLA-induced mutations in Gag (b), Protease (c), and both regions (d) were plotted against Gag-Protease-mediated replication capacities to assess their impact on viral fitness. Their effects on patients’ plasma viral load were also tested statistically (e–g)Table 2 Linear regression models investigating the rela-tionship between Gag-Pro RC and clinical factors (analysis limited to individuals lacking B*52/B*67)β Estimates for log10viral load are expressed per log10 increment. Estimates for CD4 count are expressed per 1 cell/mm3 incrementa Model: multiple r2 = 0.037, P = 0.015Variable Multivariable linear regression model parameters (Gag‑Pro RC)aEstimate P valueLog10pVL 9.13 × 10−3 0.47CD4 −1.16 × 10−4 0.043Page 7 of 14Sakai et al. Retrovirology  (2015) 12:98 included 94 HLA-associated polymorphisms in Gag and 16 in Protease). For each individual in this analysis, we counted the number of HLA-associated polymor-phisms present in their viral sequence (regardless of the host HLAs expressed). In doing so we observed no cor-relation between the number of polymorphisms in Gag, Protease, or Gag-Protease and Gag-Pro RCs (Fig. 3b–d). On the other hand, statistically significant negative cor-relations were observed between the number of HLA-associated polymorphisms in Gag (R = −0.15, P = 0.024) as well as Gag-Protease (R = −0.16, P = 0.018) and pVL (Fig. 3e–g). This suggests that HLA-associated polymor-phism burden in Gag impacts pVL without significantly affecting Gag-Pro RCs in the B*52−B*67− subpopulation, indicating the possible absence of correlation between Gag-Pro RC and pVL in this subpopulation. A simple explanation for the discrepancy would be that a combi-natorial effect by mutations in other parts of the virus may have resulted in the significant change in pVL even though the number of polymorphisms in Gag only was not enough to affect Gag replication per se.Impact of amino acid changes on Gag RCs in the absence of the protective allelesWe expanded our search for mutations in Gag that would affect Gag-Pro RCs in the B*52:01− B*67:01− subjects. In an exploratory analysis, we statistically identified amino acids in Gag that were associated with Gag-Pro RCs in persons lacking the protective B*52:01 and B*67:01 alleles. We identified amino acid variants at four posi-tions, Gag 79, 228, 286, and 357 that were associated with lower Gag-Pro RC and at one position, Gag 218, which was associated with a higher Gag-Pro RC (Table  3). To evaluate the potential combinatorial effects of these mutations on Gag-Pro RC, we assigned a score to each recombinant Gag-Protease viral sequence by assigning +1 for the presence of the RC-increasing residue and −1 for each of the four RC-decreasing residues and com-puting this sum. Overall, these scores exhibited a sig-nificant positive correlation with Gag-Pro RCs (R = 0.35, P = 0.03 × 10−6; Fig. 4a). This suggests a dose-depend-ent effect of these substitutions on replication capac-ity, which we sought to confirm in vitro. To do this, we focused on the amino acid changes that reduced Gag-Pro RC and introduced Gag Y79F and R286K substitu-tions (the most frequent variations at these positions; Additional file 4), along with Gag M228L into pNL4-3 by site-directed mutagenesis. NL4-3 naturally carries gly-cine at Gag357 and therefore naturally contains a residue other than S at this site (357G). Note that the combina-tion of these four substitutions 79F+228L+286K+357G was observed in naturally occurring sequences from our cohort. The mutant NL4-3 carrying these four Gag-Pro RC-reducing changes exhibited 20  % decreased replica-tion capacity compared to wild-type (P < 0.0001; Fig. 4b). Thus, the in vitro result confirmed the effect of the amino acid changes on the viral replication capacity. Even though the presence of these amino acid changes in the HLA-B*52:01−B*67:01− subpopulation reduced viral replication capacity in  vitro, they did not signifi-cantly correlate with pVL in this subpopulation (Fig. 5a). To investigate the potential in  vivo importance of these amino acid changes in a non-Japanese cohort, we ana-lyzed associations between these variants and pVL in a North American cohort that previously reported stronger positive correlations between Gag-Pro RC and pVL [18]. In the North American cohort, B*52:01/C*12:02 preva-lence is <2  % and B*67:01 prevalence is 0  %, and thus these alleles are unlikely to influence the present analy-sis [18]. In contrast to the Japanese cohort, the North American cohort showed a weak association between pVL and mutation scores (R = 0.078, P = 0.049; Fig. 5b), suggesting a minor but significant impact of these poly-morphisms in  vivo. Individual analysis of mutations at each position also revealed significantly lower pVL among individuals harboring amino acid variants at Gag 79 (Y79 versus non-Y) and 228 (M228L) in the North American cohort (Fig.  5d). In contrast, no such rela-tionships were observed in the HLA-B*52:01−B*67:01− Japanese subpopulation (Fig.  5c). The observation that Table 3 Amino acid polymorphisms affecting viral replication capacities in the B*5201−/B*6701− populationaa+ sequences containing the amino acid variant in question, aa− sequences lacking the amino acid variant in questiona Consensus amino acids indicate those determined for the Japanese subjects. The amino acids in parentheses indicate clade B consensusB52−B67−n = 218Amino acid  positionAmino acid RC Sample number Median RC Mann–Whitney U testConsensus Variant aa+ aa− aa+ aa− P value q valueGag 79 Matrix F (Y)a Y Lower 147 71 1.10 1.13 0.0081 0.18218 Capsid V V Higher 58 160 1.13 1.10 0.028 0.26228 Capsid M L Lower 12 206 1.03 1.11 0.015 0.22286 Capsid K (R)a R Lower 115 103 1.09 1.13 0.029 0.26357 Capsid G S Lower 159 59 1.10 1.16 0.0074 0.18Page 8 of 14Sakai et al. Retrovirology  (2015) 12:98 Gag amino acid variants, at least at position 79 and 228, are associated with pVL in North America but not Japan further supports a lesser role of Gag fitness in modulating clinical outcomes in Japan, even in the HLA-B*52:01−B*67:01− subpopulation.DiscussionMounting evidence suggests the importance of immune responses against Gag in the control of HIV-1 infec-tion [reviewed in 12, 13]. In persons expressing protec-tive HLA-B*57:01 and B*27:05 alleles, targeting critical immunodominant epitopes in Gag slows disease pro-gression. In vitro viral replication studies, notably those using recombinant viruses carrying patient-derived Gag-Protease sequences, have contributed to this understand-ing by demonstrating significantly lower Gag-dependent viral replication in HIV-1 controllers from North Ameri-can cohorts [25, 26]. Positive correlations between Gag-mediated replication capacity and pVL have also been reported in HIV-1 subtype C in an African cohort [28, 29] and subtype B in Mexican and Barbadian cohorts [30] which indicated an association between Gag-Pro RC and the frequency of protective HLA alleles in the population.In contrast, our results indicate no significant asso-ciation of Gag-Pro RC with pVL and CD4 in the over-all Japanese cohort even though identical experimental methods were employed. This lack of significant correla-tion between Gag-Pro RC and pVL, possibly even in the B*52-B*67- subpopulation according to the multivari-ate analysis, suggests that Gag fitness does not substan-tially affect pVL in Japan and raises the hypothesis that Gag-mediated HIV-1 control is less critical in Japan than other populations. The difference could be due to the unique HLA distribution in Japan [31], which shapes viral diversity at a population level [1]. Particularly, the protective HLA-B*57:01 and B*27:05 alleles are virtually absent in Japan; instead, HLA-B*52:01 (allele frequency 10.4–11.1 %) and Asian-specific B*67:01 (allele frequency 1.1–1.7 %) are associated with HIV-1 control [Allele Fre-quencies database (http://www.allelefrequencies.net/default.asp); 32]. Intriguingly, neither B*52:01 nor B*67:01 was associated with low Gag-Pro RCs, and Gag-Pro RC did not correlate with clinical parameters in subjects with these protective alleles. As such, we hypothesize that the replicative consequences of B*52/B*67-driven immune escape mutations in Gag are minimal compared to the replicative effects of escape from protective alleles such as B*27 and B*57 in other populations. In support of this hypothesis, our previous results indicate that the number of HLA-associated amino acid changes in Pol, but not in Gag, negatively correlate with pVL [31]. The correlation was observed in the presence of HLA-B*52:01, suggest-ing that escape from HLA-B*52:01-restricted responses in Pol may confer fitness costs.The absence of HLA-C*12:02, an allele in tight link-age disequilibrium with HLA-B*52:01, also showed significant associations between RC and both clinical parameters. Our previous study demonstrated a negative correlation between pVL and B*52:01-associated amino acid changes in Pol whereas C*12:02-associated changes did not show a significant association [31]. However, the effect of B*52:01 cannot be separated in the present study due to the extremely low number of B*52:01 or C*12:02 single positive subjects. The presence of C*08:01 (n = 34) and C*08:03 (n =  10) also had strong associations with both clinical parameters, but these two alleles are not sig-nificantly associated with lower or higher pVL in chronic phase in our previous study [32]. Therefore, the clini-cal importance and a putative role of these associations require further investigation.We identified amino acid polymorphisms in Gag that were associated with alterations of Gag-Pro RC in the absence of protective HLA-B*52:01 and B*67:01 alleles. Of the four polymorphisms associated with lower RC, one was located in p17 (Y79) and the other three in p24 (M228, R286, G357). Our previous analysis of HLA-associated viral polymorphisms in this cohort indicated that changes at Y79 were associated with HLA-B*51:01, C*14:02, and C*14:03, changes at M228 were associated with C*0304, and changes at G357 were associated with B*07:02 [31]. Of note, R286 had been previously identi-fied as B*52:01-associated even though B*52-expressing subjects were excluded from the present analysis of RC-associated substitutions [31]. It is therefore possible that R286K imposes a fitness cost in B*52:01+ subjects, but Fig. 4 The impact of amino acid polymorphisms in Gag on viral fitness. a The x- and y-axes indicate mutation scores and replication capacities, respectively. Mutation scores were calculated by giving −1 for each polymorphism associated with decreased replication capacities in the HLA-B*52:01−B*67:01− subpopulation, i.e. at Gag 79, 228, 286, and 357, and +1 for a polymorphism at Gag 218, which was associated with an increased replication capacity. b Combinations of the Gag-Pro RC-decreasing mutations were generated on pNL4-3, and their effects on viral replication were investigated in vitroPage 9 of 14Sakai et al. Retrovirology  (2015) 12:98 its effect on Gag-Pro RC may have been indiscernible due to the presence of other RC-modulating mutations in these individuals. Moreover, the impact of R286K on pVL among B*52:01 individuals may be difficult to detect due to two conflicting forces at play: on one hand, R286K may slightly lower replication capacity, but on the other hand, it confers the ability to escape from CTL responses and thus persist in vivo. For B*52:01− subjects, the B*52-driven R286K immune-escape mutation may have been transmitted to B*52:01− recipients, which may have Fig. 5 The impact of the amino acid polymorphisms on pVL. The x-axis indicates mutation scores as in Fig. 4. The y-axis represents pVL. Correlations between pVL and mutation scores were tested by Spearman’s rank correlation test for the HLA-B*52:01−B*67:01− Japanese (a) and North American (b) cohorts described in [18]. c, d The effect of mutations at each position on pVL was examined in the HLA-B*52:01−B*67:01− Japanese (c) and the North American (d) populations by the Mann–Whitney U test. The underlined samples indicate the presence of amino acid changes that reduce Gag-Pro RCPage 10 of 14Sakai et al. Retrovirology  (2015) 12:98 resulted in the observed lower Gag-Pro RC (similar to the reduced RCs observed when the B*57:01-restricted T242N is transmitted to B*57:01− recipients [8, 9] ). Site-directed mutagenesis confirmed that the four amino acid changes significantly reduced Gag-Pro RC. However, the reduction in Gag-Pro RCs by these amino acid changes did not lead to lower pVL even in the B*52−B*67− sub-population, suggesting a possible lack of a significant impact of Gag fitness on pVL in Japan.Even though our data argue against a significant role of Gag fitness as a major determinant of HIV-1 disease outcome in Japan, they do not completely exclude the possibility that Gag fitness subtly influences pVL. The amino acid changes may have significantly decreased pVL, but other mutations in viral genome could have restored viral replication capacity in  vivo. These sce-narios still suggest that pVL does not reflect change in Gag fitness, and that Gag replication has only a mini-mal impact on HIV-1 pathogenesis in Japan. Further studies, including longitudinal analyses, will be neces-sary to identify such compensatory mutations. Alter-natively, HLA-associated polymorphisms in Gag could actually be the result of immune pressure. In this case, a high immune pressure in Gag would generate immune-escape mutations, but the reduction of pVL may have occurred through cross-reactive CTL function rather than mutation-induced Gag fitness cost. A recent func-tional study on CTL responses reported 13 epitopes (eight in Gag and five in Pol) involved in HIV-1 con-trol in Japan [39]. B52* and B*67 targeted eight out of the 13 epitopes, including seven of the eight epitopes located in Gag, suggesting the importance of the Gag as an immune target by the Japanese protective alleles [39]. Surprisingly, however, only one of the eight dis-ease-controlling CTL epitopes targeted by B*52 induced escape mutations [39]. These results suggest that CTL responses directly control pVL by killing infected cells via HLA-mediated recognition of Gag epitopes, rather than via escape-mediated fitness costs. In addition to sustained responses to Gag epitopes that are difficult for the virus to escape [39], HLA-mediated protective effects in Japan may also be mediated by responses to Pol epitopes that eventually induce fitness-costly escape mutations within that viral protein, that partially offset the advantages gained by escape from CTL responses [31].Gag fitness has an effect on CD4 count in the Japa-nese population who do not express protective alleles, in contrast to global cohorts. At least in this subpopula-tion, escape mutations in Gag may impose fitness costs that are clinically advantageous. However, viral polymor-phisms that weakly affected pVL did not have an impact on Gag fitness even in this B*52−B*67− subpopulation, implying a minor contribution of Gag fitness to HIV-1 clinical outcome. Fitness costs of immune-driven muta-tions in other viral proteins may have more significant impact on HIV-1 disease progression in Japan [31]. These observations are in contrast to the previous reports which indicated significant correlations between Gag-Pro RCs and clinical parameters in Mexican/Barbadian and African chronic cohorts as well as elite controllers and subjects carrying protective HLA alleles in predomi-nantly Caucasian chronic cohorts [18, 25, 27–30].Previously, one study analyzed Gag-Pro RC in 156 treatment-naive Japanese individuals sampled from 1994 to 2009 and reported reduced Gag-Pro RCs among more recent isolates [40]. In contrast to our results, they observed a significant association between Gag-Pro RC and pVL. The discrepancy between these two studies may be due to the characteristics of the cohorts and/or study time course. The present study analyzed samples collected from 2008 to 2011 whereas the previous one focused on older samples. Newer viruses may have more chance to accumulate mutations that could affect viral replication dependently or inde-pendently of those occurring in Gag-Protease. Muta-tions located outside of Gag-Protease could affect pVL but not Gag-Pro RCs measured in  vitro, which would reduce the correlation between Gag-Pro RC and pVL. Sequence comparisons identified significantly differ-ent amino acid frequencies at 35 out of 500 amino acid positions in Gag and eight out of 100 positions in Pol between the two cohorts, which may have affected dif-ferent outcomes in the two cohorts (data not shown). In addition, cohort-level differences in HLA prevalence (e.g. in the other study, the frequencies of A*24+ and C*03+ individuals were 52.6 and 34.6  %, respectively whereas the frequencies of these alleles in the present study were 62.4 and 42.2 %, respectively) may also have affected the results.Notably, an association between Pol replication capacity and HIV-1 disease progression has also been reported in North American cohorts [41–46]. One of these studies suggested decreased reverse transcriptase-integrase-dependent RC in HIV-1 elite controllers [41]. In another study B*52:01-associated change at amino acid I220 in integrase was shown to decrease viral replication capacity [42]. Consid-ering that Pol enzymatic functions directly affect viral replication, mutations in Pol may meaningfully impact viral fitness and pVL. Possibly, HLA-associ-ated HIV-1 polymorphisms in Gag-Protease may not affect overall HIV-1 disease outcome in Japan as only a few alleles, including HLA-C*08:01 and C*08:03, had relatively strong correlations between Gag-Pro RC and clinical markers of infection. Additionally, Page 11 of 14Sakai et al. Retrovirology  (2015) 12:98 the present study comprised a cross-sectional anal-ysis at the chronic phase of HIV infection as previ-ously done in the other global cohort studies except for the North American cohort. Thus, we cannot exclude a possibility of a significant impact of Gag fitness on clinical outcome at different time points: e.g. at the acute/early phase. Further study, including longitudinal analyses, will be necessary to determine a possible role of Gag fitness at the earlier phase as well as Pol fitness in HIV-1 disease progression in Japan.ConclusionsOur data support a lack of a significant impact of Gag fitness on pVL in treatment-naive Japanese individuals chronically infected with HIV-1. Moreover, in contrast to non-Asian chronic cohorts where Gag-Pro RC fit-ness may modulate HIV clinical parameters especially in subjects with protective HLA alleles, the protective effects of Japanese B*52 and B*67 alleles do not seem to be associated with Gag fitness. Thus, the impact of Gag fitness on HIV clinical parameters is population dependent.MethodsStudy subjectsA total of 430 treatment-naive Japanese subjects chroni-cally infected with HIV-1 subtype B were recruited for a previous study [31] and enrolled at the National Center for Global Health and Medicine (NCGHM) from 2008 to 2011. HLA typing and the preparation of cDNA sam-ples from patient plasma were performed as described in the previous study. For the present study, specimens from 306 of the original 430 participants were randomly selected for analysis. The median pVL and CD4 count for this N = 306 cohort were 4.5 log10 RNA copies/ml (IQR: 4.1–5.1) and 299 cells/mm3 (IQR: 151–407), respectively. The study was approved by the Ethical Committees in the Faculty of Life Science, Kumamoto University and the NCGHM and conducted in accordance with the Declara-tion of Helsinki.For comparison with the Japanese cohort, a data set derived from a published North American (Canada and USA) cohort comprising 803 untreated chronic subjects infected with HIV-1 subtype B was used for the analysis [18]. Median pVL and CD4 count for the cohort were 5.1 log10 RNA copies/ml (IQR: 4.7–5.5) and 273  cells/mm3 (IQR: 130–420), respectively [18].Generation of chimeric viruses and in vitro replication assaysThe production of chimeric viruses and Gag-Pro RC determination was performed as described previously [18, 25–30]. Briefly, cDNAs were generated from plasma viral RNA for a separate study [31]. Plasma-derived gag-protease DNA fragments were generated from the cDNAs by nested PCR, using TaKaRa ExTaq Hot Start Version (Takara Bio Inc., Shiga, Japan). Chimeric HIV-1NL4-3 encoding plasma-derived gag-protease was gener-ated by electroporating gag-protease PCR fragments with linearized gag-protease-deficient pNL4-3 (delta gag-pro NL4-3) into a CEM-derived GFP-reporter T-cell line engineered to carry LTR-GFP, CXCR4, and CCR5 (CEM-GXR, a generous gift from T. Miura and M.A. Brock-man) [38] . Stock titers were determined by infecting CEM-GXR and measuring GFP+ cells, using flow cytom-etry. To determine Gag-Pro RCs, CEM-GXR cells were infected with chimeric viruses at a MOI of 0.03 at day 2, and the percentage of HIV-infected (GFP-expressing) cells was monitored daily for 2–8  days post-infection. Gag-Pro RC was calculated as the natural-log slope of the percentage of GFP-expressing cells during the expo-nential phase of viral spread and normalized against the mean value obtained with wild-type NL4-3 (i.e. NL4-3 RC  =  1.00). Virus titration and RC experiments were done in triplicate.For the validation of virus stocks, we isolated RNA from all virus stocks, using EZ1 Virus Mini Kit v2.0 (Qia-gen, Valencia, CA). Gag-protease DNA fragments were generated by RT-PCR, using SuperScript III First-Strand Synthesis System for RT-PCR (Life Technologies, Carls-bad, CA) for cDNA synthesis and TaKaRa ExTaq Hot Start Version for subsequent nested PCR. Viral sequences were analyzed on the ABI3500 genetic analyzer (Applied Biosystems, Carlsbad, CA) and compared with previ-ously determined gag-protease sequences from patients’ plasma viruses [31] by constructing a maximum-likeli-hood phylogenetic tree (PhyML) [47]. Phylogenies were viewed using Figtree (http://tree.bio.ed.ac.uk/software/figtree).Definition and analysis of HLA‑associated polymorphisms in Gag‑ProteaseA total of 110 HLA-associated polymorphisms, occur-ring at 61 unique codons in Gag-Protease, were previ-ously defined in a cohort of 430 treatment-naive Japanese subjects using phylogenetically-corrected methods [31]. In the present study we wished to identify HLA-asso-ciated polymorphisms that modulate Gag-Pro RC. For this purpose, we compared Gag-Pro RCs in the presence or absence of each of these HLA-associated amino acid substitutions and assessed significance using the Mann–Whitney U test. Figures depicting amino acid variation at Gag-Pro-RC-modulating codons were created using WebLogo 3 (http://weblogo.threeplusone.com) [48, 49] to visualize sequence conservation at each position.Page 12 of 14Sakai et al. Retrovirology  (2015) 12:98 Site‑directed mutagenesis and mutant virus stocksY79F, M228, and R286K substitutions were engineered into HIV-1 pNL4-3 by PCR overlap extension [50, 51], using PCR primers containing specific nucleotide changes: tat to ttc for Y79F, atg to ctg for M228L and aga to aag for R286K. These specific triplet codons were selected for each substitution based on higher frequencies in human codon usage, to exclude possi-ble negative effects of rare codon usage in  vitro. PCR fragments containing specific nucleotide changes were digested with BssHII (at nucleotide position 711 in pNL4-3) and SbfI (at 2838 in pNL4-3) followed by sub-cloning into pNL4-3 digested with the same restriction enzymes. The subcloned fragments were sequenced to confirm specific nucleotide changes and the absence of other mutations.pNL4-3 carrying specific mutations were transfected into HEK293T cells by lipofectamine 2000 (Life Tech-nologies, Carlsbad, CA) according to the manufac-turer’s instructions. Virus stocks were harvested 48  h post-transfection, and titers were determined by infect-ing CEM-GXR cells and subsequently measuring GFP+ cells. Replication assays were performed in two sets of triplicate for each sample as described above.Statistical analysisAssociations between Gag-Pro RC, clinical parameters and mutation scores were determined by Spearman’s rank correlation. Differences in pVL in context of the presence or absence of amino acid substitutions at four locations in Gag were determined by the Mann–Whit-ney U test. Multiple comparisons were addressed using q-values.The relationship between Gag-Pro RC and pVL/CD4 count in the population subset that did not express protective HLA class I alleles (HLA-B*52:01−B*67:01− subset) was also investigated using multivariable lin-ear regression. The primary model included only log10 plasma viral load (per log10 increment) and CD4 count (per 1 cell/mm3 increment) as variables. A second model additionally incorporated specific HLA alleles (C*08:01, C*08:02, C*12:02) that exhibited significant associations between Gag-Pro RC and both pVL and CD4 count as variables; here, model estimates are expressed in terms of individuals harboring (versus the reference group of indi-viduals not harboring) the HLA allele in question. Multi-variable analyses were performed using R (version 3.1.2).Ethics statementThe study was approved by the Ethical Committees in the Faculty of Life Science Kumamoto University (540) and the National Center for Global Health and Medicine. Informed consent was obtained from all patients, and the study was conducted in accordance with the Declaration of Helsinki.AbbreviationsHLA: human leukocyte antigen; pVL: plasma viral load; RC: replication capacity; CTL: cytotoxic T lymphocytes.Authors’ contributionsKS designed, acquired, analyzed data and drafted the manuscript. TM made substantial contributions to conception, design, and analysis of data and drafted the manuscript. TC and CJB performed the statistical analysis and contributed to drafting the manuscript. ZL generated and analyzed data and revised the manuscript critically for important intellectual content. HG and SO generated clinical data and samples. All authors read and approved the final manuscript.Author details1 Center for AIDS Research, Kumamoto University, Kumamoto 860-0811, Japan. 2 International Research Center for Medical Sciences, Kumamoto University, Kumamoto 860-0811, Japan. 3 British Columbia Centre for Excellence in HIV/AIDS, Vancouver, BC, Canada. 4 Faculty of Health Sciences, Simon Fraser University, Burnaby, BC V5A 1S6, Canada. 5 National Center for Global Health and Medicine, Tokyo 162-8655, Japan. 6 Nuffield Department of Medicine, University of Oxford, Oxford, UK. AcknowledgementsWe thank Drs. Richard Harrigan and Bruce Walker for their cohort data; Drs. Toshiyuki Miura and Mark Brockman for generously providing CEM-GXR cells and/or RC data; Mari Hasegawa, Yoshiki Eto, Yoshiko Tamura, and Lin Zhansong for technical assistance, and Sachiko Sakai for secretarial assistance.T.C. is a JSPS Research Fellow. Z.L.B. is the recipient of a New Investigator Award from the Canadian Institutes for Health Research and a Scholar Award from the Michael Smith Foundation for Health Research (MSFHR). This study was supported by the Global COE program Global Education and Research Additional filesAdditional file 1: Figure S1. Comparison of gag-protease nucleotide sequences from viral stocks and patients’ plasma viruses (total N = 306). (A) A maximum-likelihood phylogenetic tree was constructed for gag-protease sequences from all samples to verify their identities (n = 287). (B) The rest of 19 samples were analyzed separately due to the presence of gaps in the original sequences.Additional file 2: Figure S2. Replication capacities for HLA alleles preva-lent in Japan and their associations with clinical parameters. (A-C) All HLA alleles occurring at a phenotypic frequency of greater than approximately 3 % (n > 8) were examined for their associations with viral load. The median (white line), minimum and maximum values (whiskers), and inter-quartile range (box) were indicated. The black vertical lines indicate the median RC for the whole population (N = 298). HLA alleles were arranged from lowest (top) to highest (bottom) by their mean replication capacities. (D) The relationship between median Gag-Pro RCs and median pVL. Cor-relations were examined by Spearman’s rank correlation test. Each circle represents median Gag-Pro RC and pVL for a group of subjects expressing one of the HLA alleles.Additional file 3: Table S1. Associations of Gag-Pro RCs with clinical parameters in each HLA-class-I allele in the absence of B*52:01/B*67:01/C*12:02.Additional file 4: Figure S3. The positions and variations of amino acid sequences involved in the reduction of Gag-Pro RCs in the HLA-B*52:01−/B*67:01− population. The arrows indicate the positions of amino acid changes. Amino acid numbering below each sequence was based on the HIV-1 HXB2 strain. Consensus amino acids are indicated on the left and mutations on the right of each logo. Hydrophilic residues (RKDNQ) were shown in blue, neutral residues (SGHTAP) in green, and hydrophobic residues (YVMCLFIW) in black.Page 13 of 14Sakai et al. Retrovirology  (2015) 12:98 Center Aiming at the Control of AIDS, launched as a project commissioned by the Ministry of Education, Science, Sports, and Culture, Japan (K.S. and M.T.), Grant-in-Aid for Scientific Research (B) Grant Number 26293240 (M.T.), and Grants-in-Aid for AIDS Research from the Ministry of Health, Labor, and Welfare, Japan (M.T.).Competing interestsThe authors declare that they have no competing interests.Received: 30 April 2015   Accepted: 8 November 2015References 1. Kawashima Y, Pfafferott K, Frater J, Matthews P, Payne R, Addo M, Gatanaga H, Fujiwara M, Hachiya A, Koizumi H, Kuse N, Oka S, Duda A, Prendergast A, Crawford H, Leslie A, Brumme Z, Brumme C, Allen T, Brander C, Kaslow R, Tang J, Hunter E, Allen S, Mulenga J, Branch S, Roach T, John M, Mallal S, Ogwu A, Shapiro R, Prado JG, Fidler S, Weber J, Pybus OG, Klenerman P, Ndung’u T, Phillips R, Heckerman D, Harrigan PR, Walker BD, Takiguchi M, Goulder P. Adaptation of HIV-1 to human leukocyte antigen class I. Nature. 2009;2009(458):641–5. 2. 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, Allen SA, Hunter E. Role of transmitted Gag CTL polymorphisms in defining replicative capacity and early HIV-1 pathogenesis. PLoS Pathog. 2012;8:e1003041. 3. Frater AJ, Brown H, Oxenius A, Günthard HF, Hirschel B, Robinson N, Leslie AJ, Payne R, Crawford H, Prendergast A, Brander C, Kiepiela P, Walker BD, Goulder PJ, McLean A, Phillips RE. Effective T-cell responses select human immunodeficiency virus mutants and slow disease progression. J Virol. 2007;81:6742–51. 4. Matthews PC, Prendergast A, Leslie A, Crawford H, Payne R, Rous-seau C, Rolland M, Honeyborne I, Carlson J, Kadie C, Brander C, Bishop K, Mlotshwa N, Mullins JI, Coovadia H, Ndung’u T, Walker BD, Heckerman D, Goulder PJ. Central role of reverting mutations in HLA associations with human immunodeficiency virus set point. J Virol. 2008;82:8548–59. 5. Martinez-Picado J, Prado JG, Fry EE, Pfafferott K, Leslie A, Chetty S, Thobak-gale C, Honeyborne I, Crawford H, Matthews P, Pillay T, Rousseau C, Mul-lins JI, Brander C, Walker BD, Stuart DI, Kiepiela P, Goulder P. Fitness cost of escape mutations in p24 Gag in association with control of human immunodeficiency virus type 1. J Virol. 2006;80:3617–23. 6. Brumme ZL, Tao I, Szeto S, Brumme CJ, Carlson JM, Chan D, Kadie C, Frahm N, Brander C, Walker B, Heckerman D, Harrigan PR. Human leuko-cyte antigen-specific polymorphisms in HIV-1 Gag and their association with viral load in chronic untreated infection. AIDS. 2008;22:1277–86. 7. 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-lymphocyte response in Gag is associated with a dramatic reduction in human immunodeficiency virus type 1 replication. J Virol. 2007;81:12382–93. 8. Goepfert PA, Lumm W, Farmer P, Matthews P, Prendergast A, Carlson JM, Derdeyn CA, Tang J, Kaslow RA, Bansal A, Yusim K, Heckerman D, Mulenga J, Allen S, Goulder PJ, Hunter E. Transmission of HIV-1 Gag immune escape mutations is associated with reduced viral load in linked recipients. J Exp Med. 2008;205:1009–17. 9. Chopera DR, Woodman Z, Mlisana K, Mlotshwa M, Martin DP, Seoighe C, Treurnicht F, de Rosa DA, Hide W, Karim SA, Gray CM, Williamson C, CAPRISA 002 Study Team. Transmission of HIV-1 CTL escape variants provides HLA-mismatched recipients with a survival advantage. PLoS Pathog. 2008;4:e1000033. 10. Janes H, Friedrich DP, Krambrink A, Smith RJ, Kallas EG, Horton H, Casimiro DR, Carrington M, Geraghty DE, Gilbert PB, McElrath MJ, Frahm N. Vaccine-induced gag-specific T cells are associated with reduced viremia after HIV-1 infection. J Infect Dis. 2013;208:1231–9. 11. Ferrando-Martínez S, Casazza JP, Leal M, Machmach K, Muñoz-Fernández MÁ, Viciana P, Koup RA, Ruiz-Mateos E. Differential Gag-specific poly-functional T cell maturation patterns in HIV-1 elite controllers. J Virol. 2012;86:3667–74. 12. Ahlers JD, Belyakov IM. Lessons learned from natural infection: focusing on the design of protective T cell vaccines for HIV/AIDS. Trends Immunol. 2010;3:120–30. 13. Goulder PJ, Watkins DI. Impact of MHC class I diversity on immune control of immunodeficiency virus replication. Nat Rev Immunol. 2008;8:619–30. 14. 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, Walker B, Klenerman P, Steyn D, Goulder P, Phillips R, Bloemfontein-Oxford Collaborative G, vanVuuren C, Frater J. Progression to AIDS in South Africa is associated with both reverting and compensatory viral mutations. PloS One. 2011;6:e19018. 15. 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 cytotoxic T-lymphocyte pressure on human immunodeficiency virus type 1 Gag alter capsid interactions with cyclophilin A. J Virol. 2007;81:12608–18. 16. 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 reversion after transmission. Nat Med. 2004;10:282–9. 17. Boutwell CL, Rowley CF, Essex M. Reduced viral replication capacity of human immunodeficiency virus type 1 subtype C caused by cytotoxic-T-lymphocyte escape mutations in HLA-B57 epitopes of capsid protein. J Virol. 2009;83:2460–8. 18. Brockman MA, Brumme ZL, Brumme CJ, Miura T, Sela J, Rosato PC, Kadie CM, Carlson JM, Markle TJ, Streeck H, Kelleher AD, Markowitz M, Jessen H, Rosenberg E, Altfeld M, Harrigan PR, Heckerman D, Walker BD, Allen TM. Early selection in Gag by protective HLA alleles contributes to reduced HIV-1 replication capacity that may be largely compensated for in chronic infection. J Virol. 2010;84:11937–49. 19. Gijsbers EF, Feenstra KA, van Nuenen AC, Navis M, Heringa J, Schu-itemaker H, Kootstra NA. HIV-1 replication fitness of HLA-B*57/58:01 CTL escape variants is restored by the accumulation of compensatory muta-tions in Gag. PLoS One. 2013;8:e81235. 20. Ammaranond P, van Bockel DJ, Petoumenos K, McMurchie M, Finlayson R, Middleton MG, Davenport MP, Venturi V, Suzuki K, Gelgor L, Kaldor JM, Cooper DA, Kelleher AD. HIV immune escape at an immunodominant epitope in HLA-B*27-positive individuals predicts viral load outcome. J. Immunol. 2011;186:479–88. 21. Schneidewind A, Brockman MA, Sidney J, Wang YE, Chen H, Suscovich TJ, Li B, Adam RI, Allgaier RL, Mothé BR, Kuntzen T, Oniangue-Ndza C, Trocha A, Yu XG, Brander C, Sette A, Walker BD, Allen TM. Structural and functional constraints limit options for cytotoxic T-lymphocyte escape in the immunodominant HLA-B27-restricted epitope in human immunode-ficiency virus type 1 capsid. J Virol. 2008;82:5594–605. 22. Kaslow RA, Carrington M, Apple R, Park L, Muñoz A, Saah AJ, Goedert JJ, Winkler C, O’Brien SJ, Rinaldo C, Detels R, Blattner W, Phair J, Erlich H, Mann DL. Influence of combinations of human major histocom-patibility complex genes on the course of HIV-1 infection. Nat Med. 1996;2:405–11. 23. Goulder PJ, Phillips RE, Colbert RA, McAdam S, Ogg G, Nowak MA, Gian-grande P, Luzzi G, Morgan B, Edwards A, McMichael AJ, Rowland-Jones S. Late escape from an immunodominant cytotoxic T-lymphocyte response associated with progression to AIDS. Nat Med. 1997;3:212–7. 24. 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, Jones I, McMichael AJ, Rowland-Jones S, Phillips RE. Clustered mutations in HIV-1 gag are consistently required for escape from HLA-B27-restricted cytotoxic T lymphocyte responses. J Exp Med. 2001;193:375–86. 25. Miura T, Brockman MA, Brumme ZL, Brumme CJ, Pereyra F, Trocha A, Block BL, Schneidewind A, Allen TM, Heckerman D, Walker BD. HLA-associated alterations in replication capacity of chimeric NL4-3 viruses carrying gag-protease from elite controllers of human immunodeficiency virus type 1. J Virol. 2009;83:140–9.Page 14 of 14Sakai et al. Retrovirology  (2015) 12:98  26. Miura T, Brumme ZL, Brockman MA, Rosato P, Sela J, Brumme CJ, Pereyra F, Kaufmann DE, Trocha A, Block BL, Daar ES, Connick E, Jessen H, Kelleher AD, Rosenberg E, Markowitz M, Schafer K, Vaida F, Iwamoto A, Little S, Walker BD. Impaired replication capacity of acute/early viruses in persons who become HIV controllers. J Virol. 2010;84:7581–91. 27. Wright JK, Naidoo VL, Brumme ZL, Prince JL, Claiborne DT, Goulder PJ, Brockman MA, Hunter E, Ndung’u T. Impact of HLA-B*81-asso-ciated mutations in HIV-1 Gag on viral replication capacity. J Virol. 2012;86:3193–9. 28. Wright JK, Novitsky V, Brockman MA, Brumme ZL, Brumme CJ, Carlson JM, Heckerman D, Wang B, Losina E, Leshwedi M, van der Stok M, Maphu-mulo L, Mkhwanazi N, Chonco F, Goulder PJ, Essex M, Walker BD, Ndung’u T. Influence of Gag-protease-mediated replication capacity on disease progression in individuals recently infected with HIV-1 subtype C. J Virol. 2011;85:3996–4006. 29. Wright JK, Brumme ZL, Carlson JM, Heckerman D, Kadie CM, Brumme CJ, Wang B, Losina E, Miura T, Chonco F, van der Stok M, Mncube Z, Bishop K, Goulder PJ, Walker BD, Brockman MA, Ndung’u T. Gag-protease-mediated replication capacity in HIV-1 subtype C chronic infection: associations with HLA type and clinical parameters. J Virol. 2010;84:10820–31. 30. Juarez-Molina CI, Payne R, Soto-Nava M, Avila-Rios S, Valenzuela-Ponce H, Adland E, Leitman E, Brener J, Muenchhoff M, Branch S, Landis C, Reyes-Teran G, Goulder P. Impact of HLA selection pressure on HIV fitness at a population level in Mexico and Barbados. J Virol. 2014;88:10392–8. 31. 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 Japanese population. J Virol. 2014;88:4764–75. 32. Naruto T, Gatanaga H, Nelson G, Sakai K, Carrington M, Oka S, Takiguchi M. HLA class I-mediated control of HIV-1 in the Japanese population, in which the Protective HLA-B*57 and HLA-B*27 alleles are absent. J Virol. 2012;86:10870–2. 33. Hashimoto M, Kinoshita T, Yamasaki M, Tanaka H, Imanishi T, Ihara H, Ichi-kawa Y, Fukunishi T. Gene frequencies and haplotypic associations within the HLA region in 916 unrelated Japanese individuals. Tissue Antigens. 1994;44:166–73. 34. Saito S, Ota S, Yamada E, Inoko H, Ota M. Allele frequencies and haplo-typic associations defined by allelic DNA typing at HLA class I and class II loci in the Japanese population. Tissue Antigens. 2000;56:522–9. 35. Tanaka H, Akaza T, Juji T. Report of the Japanese Central Bone Marrow Data Center. Clin Transpl. 1996;10:139–44. 36. Tokunaga K, Ishikawa Y, Ogawa A, Wang H, Mitsunaga S, Moriyama S, Lin L, Bannai M, Watanabe Y, Kashiwase K, Tanaka H, Akaza T, Tadokoro K, Juji T. Sequence-based association analysis of HLA class I and II alleles in Japanese supports conservation of common haplotypes. Immunogenet-ics. 1997;46:199–205. 37. Itoh Y, Mizuki N, Shimada T, Azuma F, Itakura M, Kashiwase K, Kikkawa E, Kulski JK, Satake M, Inoko H. High-throughput DNA typing of HLA-A, -B, -C, and -DRB1 loci by a PCR-SSOP-Luminex method in the Japanese population. Immunogenetics. 2005;57:717–29. 38. Brockman MA, Tanzi GO, Walker BD, Allen TM. Use of a novel GFP reporter cell line to examine replication capacity of CXCR4- and CCR5-tropic HIV-1 by flow cytometry. J Virol Methods. 2006;131:134–42. 39. Murakoshi H, Akahoshi T, Koyanagi M, Chikata T, Naruto T, Maruyama R, Tamura Y, Ishizuka N, Gataganaga H, Oka S, Takiguchi M. Clinical control of HIV-1 by cytotoxic T cells specific for multiple conserved epitopes. J Virol. 2015;89:5330–9. 40. Nomura S, Hosoya N, Brumme ZL, Brockman MA, Kikuchi T, Koga M, Naka-mura H, Koibuchi T, Fujii T, Carlson JM, Heckerman D, Kawana-Tachikawa A, Iwamoto A, Miura T. Significant reductions in Gag-protease-mediated HIV-1 replication capacity during the course of the epidemic in Japan. J Virol. 2013;87:1465–76. 41. Brumme ZL, Li C, Miura T, Sela J, Rosato PC, Brumme CJ, Markle TJ, Martin E, Block BL, Trocha A, Kadie CM, Allen TM, Pereyra F, Heckerman D, Walker BD, Brockman MA. Reduced replication capacity of NL4-3 recombinant viruses encoding reverse transcriptase-integrase sequences from HIV-1 elite controllers. J Acquir Immune Defic Syndr. 2011;56:100–8. 42. Brockman MA, Chopera DR, Olvera A, Brumme CJ, Sela J, Markle TJ, Martin E, Carlson JM, Le AQ, McGovern R, Cheung PK, Kelleher AD, Jessen H, Markowitz M, Rosenberg E, Frahm N, Sanchez J, Mallal S, John M, Harrigan PR, Heckerman D, Brander C, Walker BD, Brumme ZL. Uncommon path-ways of immune escape attenuate HIV-1 integrase replication capacity. J Virol. 2012;86:6913–23. 43. Barbour JD, Hecht FM, Wrin T, Segal MR, Ramstead CA, Liegler TJ, Busch MP, Petropoulos CJ, Hellmann NS, Kahn JO, Grant RM. Higher CD4+ T cell counts associated with low viral pol replication capacity among treat-ment-naive adults in early HIV-1 infection. J Infect Dis. 2004;190:251–6. 44. Daar ES, Kesler KL, Wrin T, Petropoulo CJ, Bates M, Lail A, Hellmann NS, Gomperts E, Donfield S, Hemophilia Growth and Development Study. HIV-1 pol replication capacity predicts disease progression. AIDS. 2005;19:871–7. 45. Goetz MB, Leduc R, Wyman N, Kostman JR, Labriola AM, Lie Y, Weidler J, Coakley E, Bates M, Luskin-Hawk R, Long Term Monitoring Study (CPCRA060). Terry Beirn Community Programs for Clinical Research on AIDS. HIV replication capacity is an independent predictor of disease progression in persons with untreated chronic HIV infection. J Acquir Immune Defic Syndr. 2010;53:472–9. 46. Ng OT, Laeyendecker O, Redd AD, Munshaw S, Grabowski MK, Paquet AC, Evans MC, Haddad M, Huang W, Robb ML, Reynolds SJ, Gray RH, Wawer MJ, Serwadda D, Eshleman SH, Quinn TC. HIV type 1 polymerase gene polymorphisms are associated with phenotypic differences in replication capacity and disease progression. J Infect Dis. 2014;209:66–73. 47. Guindon S, Gascuel O. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol. 2003;52:696–704. 48. Crooks GE, Hon G, Chandonia JM, Brenner SE. WebLogo: a sequence logo generator. Genome Res. 2004;14:1188–90. 49. Schneider TD, Stephens RM. Sequence logos: a new way to display con-sensus sequences. Nucleic Acids Res. 1990;18:6097–100. 50. Ho SN, Hunt HD, Horton RM, Pullen JK, Pease LR. Site-directed mutagen-esis by overlap extension using the polymerase chain reaction. Gene. 1989;77:51–9. 51. Heckman KL, Pease LR. Gene splicing and mutagenesis by PCR-driven overlap extension. Nat Protoc. 2007;2:924–32.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 redistributionSubmit your manuscript at www.biomedcentral.com/submit


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items