Immune-driven recombination and loss of control after HIV superinfection

Abstract

After acute HIV infection, CD8(+) T cells are able to control viral replication to a set point. This control is often lost after superinfection, although the mechanism behind this remains unclear. In this study, we illustrate in an HLA-B27(+) subject that loss of viral control after HIV superinfection coincides with rapid recombination events within two narrow regions of Gag and Env. Screening for CD8(+) T cell responses revealed that each of these recombination sites (approximately 50 aa) encompassed distinct regions containing two immunodominant CD8 epitopes (B27-KK10 in Gag and Cw1-CL9 in Env). Viral escape and the subsequent development of variant-specific de novo CD8(+) T cell responses against both epitopes were illustrative of the significant immune selection pressures exerted by both responses. Comprehensive analysis of the kinetics of CD8 responses and viral evolution indicated that the recombination events quickly facilitated viral escape from both dominant WT- and variant-specific responses. These data suggest that the ability of a superinfecting strain of HIV to overcome preexisting immune control may be related to its ability to rapidly recombine in critical regions under immune selection pressure. These data also support a role for cellular immune pressures in driving the selection of new recombinant forms of HIV.

Figure 1.

Loss of viral control in subject AC160. Longitudinal assessment of plasma viral loads (red line) and CD4⁺ T cell counts (dashed blue line) after acute HIV-1 infection. At day 503 (T4), viral loads dramatically increased to 380,000 copies/ml, followed by a steady decline in CD4⁺ T cell counts to <300 copies/μl.

Figure 2.

Loss of the immunodominant HLA-B27–restricted KK10 CD8 response after viral escape, and the development of a variant-specific response. (A) Longitudinal assessment of the WT (black) and L₂₆₈M variant-specific (gray) KK10 CD8 responses by IFN-γ ELISpot. The WT response declined at day 419, coincident with development of the L₂₆₈M mutation, after which the L₂₆₈M-specific response developed. Decline of both responses occurred at day 1,034 after development of the R₂₆₄K anchor mutation. (B) Assessment of CD8⁺ T cell polyfunctionality at four different time points. All 32 possible combinations of the 5 antigen-specific functions studied for each epitope are shown on the x axis, and the contributions of each epitope-specific CD8⁺ T cell population are indicated as bars (WT, red; L₂₆₈M, blue). Responses are grouped according to the number of functions (1, yellow; 2, cyan; 3, green; 4, blue; 5, red) and summarized by pie charts. (C) Dual tetramer staining for the KK10 WT and L₂₆₈M-specific response on day 664, illustrating two distinct non–cross-reacting populations of antigen-specific cells.

Figure 3.

Superinfection and recombination. (A) HIV gag sequences from subject AC160 and 19 other HIV chronic-infected subjects were compared using a neighbor-joining phylogenetic tree. Sequences from subject AC160 derived from the first year of infection (days 22, 83, and 419) cluster independently from sequences derived later in infection (days 503 and 545). Scale bar indicates the genetic distance along the branches, and bootstrap values >60 are shown. (B) SimPlot recombination analysis of a full-length HIV sequence derived from day 545 compared with viruses derived from pre-superinfection (day 83, green; day 419, red) and the superinfecting strain (day 503, blue) using a window of 100 bp and a step size of 10 bp. Two regions with double recombination breakpoints were observed in Gag and Env and Yates-corrected χ² values, and P values were calculated for each putative breakpoint. Breakpoints in gag were detected around positions 690 (χ = 5.8; P = 0.0165) and 822 (χ = 50.1; P < 0.0001), and breakpoints in env were detected around positions 6,047 (χ = 50.1; P < 0.0001) and 6,164 (χ = 30.1; P < 0.0001). By day 860 (T8), the breakpoints in gag were no longer detectable, supportive of a possible second recombination event in the recombinant strain that restored this region of the original superinfecting strain. (C) Longitudinal amino acid alignment of Gag sequences in AC160. Sequences derived from the primary infecting strain (black), the superinfecting strain (red), and the new recombinant form (blue) are aligned to clade B consensus. Early viral escape was observed in the B27-KK10 epitope boxed in red. Flanking sequences are deleted (//) to illustrate sequence diversity between the two strains. Sequences suggested to be involved in the first recombination event are shaded gray, whereas those involved in the second recombination event are shaded yellow. (D) Longitudinal amino acid alignment of Env sequences containing the Cw1-CL9 epitope are boxed in red. Sequences suggested to be involved in the third recombination event in Env are shaded gray.

Figure 4.

Abrogation of the Cw1-CL9 response after recombination in Env. Longitudinal development of the WT (black) and A₂₁₇T variant-specific (gray) CL9 responses as measured by IFN-γ ELISpot. The A₂₁₇T variant-specific response was not detected before recombination at day 419 (“not present”). At the time of superinfection at day 503, the WT response exhibited the highest magnitude, suggestive of having driven the recombination event to the A₂₁₇T form of CL9. Decline of both the WT and the A₂₁₇T -specific CL9 response was observed at day 1,034.