Transmission of HIV-1 Gag immune escape mutations is associated with reduced viral load in linked recipients

Abstract

In a study of 114 epidemiologically linked Zambian transmission pairs, we evaluated the impact of human leukocyte antigen class I (HLA-I)-associated amino acid polymorphisms, presumed to reflect cytotoxic T lymphocyte (CTL) escape in Gag and Nef of the virus transmitted from the chronically infected donor, on the plasma viral load (VL) in matched recipients 6 mo after infection. CTL escape mutations in Gag and Nef were seen in the donors, which were subsequently transmitted to recipients, largely unchanged soon after infection. We observed a significant correlation between the number of Gag escape mutations targeted by specific HLA-B allele-restricted CTLs and reduced VLs in the recipients. This negative correlation was most evident in newly infected individuals, whose HLA alleles were unable to effectively target Gag and select for CTL escape mutations in this gene. Nef mutations in the donor had no impact on VL in the recipient. Thus, broad Gag-specific CTL responses capable of driving virus escape in the donor may be of clinical benefit to both the donor and recipient. In addition to their direct implications for HIV-1 vaccine design, these data suggest that CTL-induced viral polymorphisms and their associated in vivo viral fitness costs could have a significant impact on HIV-1 pathogenesis.

Figure 1.

Southern African clade C Gag sequences tend to cluster together. Neighbor-joining phylogenetic tree of Gag (p17 and p24) clade C taxa is shown. These taxa (20 from each cohort) were randomly selected from the total pool of sequencing data available from the South African (ZA, yellow) and Zambian (ZM, red) cohorts. We used an HIV database (Table I) to obtain clade C consensus and sequences from other countries affected by the clade C epidemic. Other African countries are shown (blue), with 10 sequences each from Botswana (BW), Kenya (KE), and Malawi (MW), and 6 sequences from Zimbabwe (ZW). Non-African taxa (green) are represented by 10 sequences each from India (IN) and Brazil (BW). Bootstrap values >70% are shown.

Figure 2.

HLA-B–associated mutations in donor Gag affect HIV-1 plasma VL in linked recipients. (A and B) In linked recipients in the Zambian cohort (n = 114), the association between the numbers of mutations within or flanking known CTL epitopes (Table I) in HIV-1 Nef (A) and Gag (B) with VL is shown. (C and D) All HLA-I–restricted polymorphisms associated with Nef and Gag mutations, respectively, were included (P < 0.001 and q < 0.2), regardless of whether they were within or flanking known CTL epitopes (Table I and Table S1). (E and F) all HLA-B–restricted mutations (P < 0.001 and q < 0.2) for Nef and Gag, respectively. The horizontal bars represent median VL values. The Mann-Whitney U test was used to compare median VL between groups, and the nonparametric Spearman rank correlation test was used to determine whether a correlation exists between the number of mutations and VL.

Figure 3.

Recipients lacking HLA-B alleles associated with Gag mutations benefit most from these mutations. (A and B) The relationship between VL and HLA-B–associated polymorphisms in Gag is shown for recipients who lack (A) or carry (B) the HLA-B alleles associated with these mutations (Table I). (C) The relationship between VL and the number of Gag epitopes that are potentially targeted in the same recipients as in B. Potential epitopes are defined as the number of Gag polymorphisms that are associated with the HLA-B alleles of each recipient. Panel A represents 35 recipients, whereas B and C represent 50 recipients each. The horizontal bars represent median VL values. Statistics were performed as described in Fig. 2.

Figure 4.

Lower VL in recipients is not restricted to a select class of donor HLA-B alleles. VLs based on transmission of viruses with (closed squares) or without (open squares) Gag mutations restricted by individual HLA-I alleles are shown. The data represent only recipients (n = 36) who do not express HLA-B molecules associated with Gag polymorphisms, as shown in Table I. Only mutations present in at least two of the recipients are shown. The Gag mutation restricted by either HLA-B*13 or -Cw*02 could not be resolved to either individual allele. Squares represent median values, and error bars represent the interquartile range of the data.