Transmission and accumulation of CTL escape variants drive negative associations between HIV polymorphisms and HLA

Abstract

Human immunodeficiency virus (HIV)-1 amino acid sequence polymorphisms associated with expression of specific human histocompatibility leukocyte antigen (HLA) class I alleles suggest sites of cytotoxic T lymphocyte (CTL)-mediated selection pressure and immune escape. The associations most frequently observed are between expression of an HLA class I molecule and variation from the consensus sequence. However, a substantial number of sites have been identified in which particular HLA class I allele expression is associated with preservation of the consensus sequence. The mechanism behind this is so far unexplained. The current studies, focusing on two examples of “negatively associated” or apparently preserved epitopes, suggest an explanation for this phenomenon: negative associations can arise as a result of positive selection of an escape mutation, which is stable on transmission and therefore accumulates in the population to the point at which it defines the consensus sequence. Such negative associations may only be in evidence transiently, because the statistical power to detect them diminishes as the mutations accumulate. If an escape variant reaches fixation in the population, the epitope will be lost as a potential target to the immune system. These data help to explain how HIV is evolving at a population level. Understanding the direction of HIV evolution has important implications for vaccine development.

Availability of a Diversely Avid CD8+ T Cell Repertoire Specific for the Subdominant HLA-A2-Restricted HIV-1 Gag p2419–27 Epitope

“…These characteristics provide a plausible explanation for the paucity of circulating SL9-specific CD8 ϩ T cell reactivity during the proinflammatory acute phase of infection, as well as its paradoxical dominance during the CD4-diminished chronic phase of infection (18,26,27). The observation that SL9 and its mutations are highly conserved across clades (28), often emerging sequentially or even re-emerging within the same infection (18,29), is consistent with the idea that this epitope may oscillate around an optimal viral "solution" to a host population rich in HLA-A2 alleles (22). Indeed, SL9 and its common variants may be "optimized" to elicit ineffectual CTL responses (30).…”

“…This has led to the suggestion that useful HLA-A2-restricted epitopes in circulating clade B viruses may have been lost because the clade B epidemic is historically older than the clade C epidemic and HLA-A2 is more prevalent in the West than in Africa (19,20). Loss of epitopes restricted by MHC alleles with high frequencies was also supported by studies in HIV-1 clade C-infected cohorts (21,22) and in SIV-infected macaques (23).…”

“…Because Gag-specific immunity may be particularly effective for the control of HIV-1 infection (8,22,36), the highly conserved TV9, which overlaps with the HLA-B57-restricted p24 peptide recognized by long-term nonprogressors (9, 40 -42), may be a potentially useful vaccine target for HLA-A2 carriers. The fact that it is subdominant in patients may be particularly advantageous in a therapeutic vaccine setting because the responding repertoire would not have been exhausted by the infection.…”

“…HLA typing was possible for 280 of these samples, of which 121 were HLA-A2 ϩ . OLP21 (PRTLNAWVKVVEEKAP, p24 [17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32] ) was recognized by 24 subjects, with most of this reactivity (14 of 24) likely directed against the HLA-B*1503-restricted VKV VEEKAF epitope frequently seen in HLA-B*1503-expressing subjects (21). Reactivity to the nonameric TV9 peptide was confirmed in one of two HLA-A2 ϩ HLA-B*1503 Ϫ samples tested that reacted to OLP21 (L8 47, 500 sfc/10 6 PBMCs; Fig.…”

Divergent adaptation of hepatitis C virus genotypes 1 and 3 to human leukocyte antigen-restricted immune pressure

“…Associations with an odds ratio Ͻ1 indicate overrepresentation of consensus in individuals that carry the respective HLA allele, indicating sites where the consensus viral sequence appears best adapted to T cell responses acting at these sites across the host population. 4,5,18,19 In order to minimize confounding through founder effects, 12 the analysis was adjusted for viral phylogenetic relatedness (see Statistical Methods).…”