Impact of immune-driven sequence variation in HIV-1 subtype C Gagprotease on viral fitness and disease progression.

Abstract

Understanding of the viral and host factors that determine time for progression to acquired immunodeficiency syndrome (AIDS) in individuals infected with human immunodeficiency virus type 1 (HIV-1) could aid in the design of an effective HIV-1 vaccine. Human leukocyte antigen (HLA) class I profile is strongly and consistently associated with differential rates of HIV-1 disease progression, however the mechanisms explaining this are not well understood. It has been hypothesised that “protective” HLA alleles select escape mutations in functionally important epitopes in the conserved group specific antigen (Gag) protein resulting in HIV-1 attenuation, which may result in slower disease progression. Many of the studies investigating the fitness cost of Gag escape mutations have concentrated on a few pre-selected mutations and have not assessed fitness consequences in the natural sequence background. Furthermore, the majority of studies have focussed on HIV-1 subtype B, while HIV-1 subtype C is the most prevalent subtype worldwide. Therefore, in the present study, a large population-based approach and clinically-derived Gag-protease sequences were used to comprehensively investigate the relationship between immunedriven sequence variation in Gag, viral replication capacity and markers of disease progression in HIV-1 subtype C chronic infection. The influence of Gag function on HIV-1 disease progression was further investigated in early HIV-1 subtype C infection. It was also hypothesised that Gag may contribute significantly to overall HIV-1 fitness and towards fitness differences between HIV-1 subtypes. Materials and Methods Recombinant viruses encoding Gag-protease, derived from antiretroviral naïve HIV-1 subtype C chronically (n=406) and recently (n=60) infected patients as well as a small subset of HIV-1 subtype B chronically infected patients (n=25), were generated by electroporation of an HIV-inducible green fluorescent protein (GFP)-reporter T cell line with plasmaderived gag-protease PCR products and linearised gag-protease-deleted NL4-3 plasmid. The replication capacities of recombinant viruses, as well as intact HIV-1 isolates from peripheral blood mononuclear cells of patients chronically infected with HIV-1 subtype C (n=16), were assayed in the GFP-reporter T cell line by flow cytometry. Replication capacity was defined as the slope of increase in percentage infected cells from days 3-6 following infection, normalised to the growth of a wild-type NL4-3 control. Replication capacities were related to patient HLA alleles and markers of disease progression (viral load, CD4+ T cell count, and rate of CD4+ T cell decline in chronically infected patients, and viral set point and rate of CD4+ T cell decline in recently infected patients). Replication capacities were compared between isolates and recombinant viruses encoding Gag-protease from the same isolates, as well as between HIV-1 subtype B and C recombinant viruses matched for viral load and CD4+ T cell count. Bulk sequencing of patient -derived gagprotease amplicons was performed and mutations were identified that were significantly associated with altered viral replication capacity. The fitness effect of some of these mutations was directly tested by site-directed mutagenesis followed by assay of the mutant viruses. Results In HIV-1 subtype C chronic infection, protective HLA-B alleles, most notably HLA-B*81 (p<0.0001), were associated with lower replication capacities. HLA-associated mutations at low entropy sites (i.e. conserved sites) in or adjacent to Gag epitopes were associated with lower replication capacities (p=0.02), especially the HLA-B*81-associated 186S mutation in the TL9 epitope (p=0.0001). The fitness cost of this mutation was confirmed in site-directed mutagenesis experiments (p<0.001), and the co-varying mutations tested did not significantly compensate for this fitness cost. Replication capacity also correlated positively of an HIV-inducible green fluorescent protein (GFP)-reporter T cell line with plasmaderived gag-protease PCR products and linearised gag-protease-deleted NL4-3 plasmid. The replication capacities of recombinant viruses, as well as intact HIV-1 isolates from peripheral blood mononuclear cells of patients chronically infected with HIV-1 subtype C (n=16), were assayed in the GFP-reporter T cell line by flow cytometry. Replication capacity was defined as the slope of increase in percentage infected cells from days 3-6 following infection, normalised to the growth of a wild-type NL4-3 control. Replication capacities were related to patient HLA alleles and markers of disease progression (viral load, CD4+ T cell count, and rate of CD4+ T cell decline in chronically infected patients, and viral set point and rate of CD4+ T cell decline in recently infected patients). Replication capacities were compared between isolates and recombinant viruses encoding Gag-protease from the same isolates, as well as between HIV-1 subtype B and C recombinant viruses matched for viral load and CD4+ T cell count. Bulk sequencing of patient -derived gagprotease amplicons was performed and mutations were identified that were significantly associated with altered viral replication capacity. The fitness effect of some of these mutations was directly tested by site-directed mutagenesis followed by assay of the mutant viruses. Results In HIV-1 subtype C chronic infection, protective HLA-B alleles, most notably HLA-B*81 (p<0.0001), were associated with lower replication capacities. HLA-associated mutations at low entropy sites (i.e. conserved sites) in or adjacent to Gag epitopes were associated with lower replication capacities (p=0.02), especially the HLA-B*81-associated 186S mutation in the TL9 epitope (p=0.0001). The fitness cost of this mutation was confirmed in site-directed mutagenesis experiments (p<0.001), and the co-varying mutations tested did not significantly compensate for this fitness cost. Replication capacity also correlated positivelyof an HIV-inducible green fluorescent protein (GFP)-reporter T cell line with plasmaderived gag-protease PCR products and linearised gag-protease-deleted NL4-3 plasmid. The replication capacities of recombinant viruses, as well as intact HIV-1 isolates from peripheral blood mononuclear cells of patients chronically infected with HIV-1 subtype C (n=16), were assayed in the GFP-reporter T cell line by flow cytometry. Replication capacity was defined as the slope of increase in percentage infected cells from days 3-6 following infection, normalised to the growth of a wild-type NL4-3 control. Replication capacities were related to patient HLA alleles and markers of disease progression (viral load, CD4+ T cell count, and rate of CD4+ T cell decline in chronically infected patients, and viral set point and rate of CD4+ T cell decline in recently infected patients). Replication capacities were compared between isolates and recombinant viruses encoding Gag-protease from the same isolates, as well as between HIV-1 subtype B and C recombinant viruses matched for viral load and CD4+ T cell count. Bulk sequencing of patient -derived gagprotease amplicons was performed and mutations were identified that were significantly associated with altered viral replication capacity. The fitness effect of some of these mutations was directly tested by site-directed mutagenesis followed by assay of the mutant viruses. Results In HIV-1 subtype C chronic infection, protective HLA-B alleles, most notably HLA-B*81 (p<0.0001), were associated with lower replication capacities. HLA-associated mutations at low entropy sites (i.e. conserved sites) in or adjacent to Gag epitopes were associated with lower replication capacities (p=0.02), especially the HLA-B*81-associated 186S mutation in the TL9 epitope (p=0.0001). The fitness cost of this mutation was confirmed in site-directed mutagenesis experiments (p<0.001), and the co-varying mutations tested did not significantly compensate for this fitness cost. Replication capacity also correlated positively with baseline viral load (p<0.0001) and negatively with baseline CD4+ T cell count (p=0.0004), but not with subsequent rate of CD4+ T cell decline (p=0.73). In HIV-1 subtype C recent infection, replication capacities of the early viruses did not correlate with subsequent viral set points (p=0.37) but were significantly lower in individuals with below median viral set points (p=0.03), and there was a trend of correlation between lower replication capacities and slower rates of CD4+ T cell decline (p=0.09). Overall, the proportion of host HLA-specific Gag polymorphisms in or adjacent to epitopes was negatively associated with replication capacities (p=0.04) but host HLA-B-specific polymorphisms were associated with higher viral set points (p=0.01), suggesting a balance between effective Gag CD8+ T cell responses and viral replication capacity in influencing viral set point. A moderate statistically significant correlation was found between the replication capacities of whole isolates and their corresponding Gag-protease recombinant viruses (p=0.04) and the replication capacities of the subtype C recombinant viruses were significantly lower than that of the subtype B recombinant viruses (p<0.0001). The subtype-specific difference in the consensus amino acids at Gag codons 483 and 484 was found in site-directed mutagenesis experiments to largely contribute to the fitness difference between subtypes, possibly by influencing budding efficiency. Discussion The data support that protective HLA alleles, in particular HLA-B*81, attenuate HIV-1 through HLA-restricted CD8+ T cell-mediated selection pressure on Gag. Results suggest that viral replication capacity determined by sequence variability in Gag-protease has an impact on HIV-1 disease progression, but also indicate that a balance between HLA-driven fitness costs and maintenance of effective CD8+ T cell responses is important in determining clinical outcome. Gag-protease was observed to significantly contribute to overall HIV-1 replication capacity and variability in this region between HIV-1 subtypes B and C is suggested to partly explain the difference in viral fitness between these subtypes. Specific mutations in Gag-protease associated with viral attenuation were identified and it was also observed that mutations in conserved Gag regions carried the greatest cost to HIV-1 replication capacity. Overall, the data support the concept of, and may assist in the rational design of, an HIV-1 vaccine in which immune responses are directed towards several conserved epitopes, particularly in Gag, with the aim to constrain immune escape (thereby maintaining effective CD8+ T cell responses) and attenuate HIV-1 (in the event of partial escape), resulting in slower disease course and reduced HIV-1 transmission at the population level.