RV144 vaccine imprinting constrained HIV-1 evolution following breakthrough infection

Abstract

The scale of the HIV-1 epidemic underscores the need for a vaccine. The multitude of circulating HIV-1 strains together with HIV-1's high evolvability hints that HIV-1 could adapt to a future vaccine. Here, we wanted to investigate the effect of vaccination on the evolution of the virus post-breakthrough infection. We analyzed 2,635 HIV-1 env sequences sampled up to a year post-diagnosis from 110 vaccine and placebo participants who became infected in the RV144 vaccine efficacy trial. We showed that the Env signature sites that were previously identified to distinguish vaccine and placebo participants were maintained over time. In addition, fewer sites were under diversifying selection in the vaccine group than in the placebo group. These results indicate that HIV-1 would possibly adapt to a vaccine upon its roll-out.

Keywords: HIV-1; sieve analysis; vaccine; within-host evolution.

Figure 1.

Sieve signatures distinguishing vaccine and placebo groups persisted over time. At twelve Env signature sites, the proportion of participants having the consensus amino acid is shown for vaccine (red) and placebo (blue) groups at (A) diagnosis and (B) 6 months post-diagnosis. Error bars in (B) indicate 95 per cent confidence intervals across jackknifed samples. (C) Proportion of transitions (median values across jackknifed samples) between consensus and non-consensus amino acids across sampling times in vaccine and placebo recipients. The per-participant VR scores corresponding to the twelve sieve signature sites that were calculated at (D) diagnosis and (E) 6 months post-diagnosis were significantly higher in vaccine recipients (P-values are shown for Mann–Whitney U t tests).

Figure 2.

Mutations at sites associated with B or T cell pressure in Env did not differ across treatment groups. The number of mutations at known Ab contact sites (n = 164) per placebo (blue) and vaccine (red) participant at (A) diagnosis when comparing sequences with the CRF01_AE consensus and (B) 6 months post-diagnosis when comparing these sequences with those sampled at diagnosis. Swarm plots show the number of mutations after correcting for differences in the number of sequences per participant. (C) The proportion of sites with mutations considering sites known as CTL epitopes and sites without CTL epitopes (data from vaccine and placebo groups combined). (D) The number of CTL epitopes predicted on sequences sampled at diagnosis in vaccine and placebo recipients. (E) The number of CTL epitopes matching CRF01_AE gp120 vaccine insert 92TH023 in vaccine and placebo recipients at diagnosis. (F) Mean sequence identity between vaccine and placebo recipients and CRF01_AE gp120 vaccine insert 92TH023. P-values are shown for pairwise Student’s t tests (parenthetical P-values show results after correcting for the number of sequences in (a and b)).

Figure 3.

Minor variants increased at sieve signature sites in placebo but not in vaccine recipients. The percentage represented by non-consensus amino acids in vaccine (red) and placebo (blue) recipients at each signature site at (A) diagnosis and (B) 6 months post-diagnosis. This is calculated for infections where the consensus corresponding to the signature site is found in the majority of sequences from that individual; the number of participants with zero values is noted in parentheses. Median values across jackknifed samples are shown in (B). The ratio of the mean number of non-consensus (minor) amino acids per consensus-majority participant in vaccine and placebo recipients at (C) diagnosis and (D) 6 months post-diagnosis at each Env signature site. (C and D) Dashed lines indicate an equal number of non-consensus amino acids between vaccine and placebo recipients; values above the dashed line indicate a higher number in vaccinees. (E) The mean ratio of non-consensus amino acid frequencies at each signature site across sampling times in placebo and vaccine recipients. The dashed line indicates the value if there was no difference in frequencies across sampling times; values above the dashed line indicate an increase in non-consensus amino acids between diagnosis and 6 months post-diagnosis.

Figure 4.

Diversifying selection observed in placebo but not vaccine recipients. The number of sites under purifying and diversifying selection per participant is shown for vaccine (red) and placebo (blue) recipients at (A) diagnosis and (B) 6 months post-diagnosis. (C) Per-participant sites under selection at 6 months post-diagnosis in Env-gp120 and Env-gp41. Significant pairwise differences are indicated with the associated P-value (Student’s t test).

Figure 5.

The lack of neutralization breadth in vaccine recipients was not associated with the number of potential escape mutations at Ab contact sites. Scatterplots of the neutralization breadth (measured 3 years after diagnosis) against the number of mutations at Ab contact sites in sequences from vaccine (n = 18, red) and placebo (n = 39, blue) recipients. Dashed lines indicate the threshold for broad neutralization breadth (70 per cent of viruses). Mutations are scored based on (A) sequences sampled at diagnosis compared to the CRF01_AE consensus and (B) sequences sampled 6 months later and compared to those sampled at diagnosis. Participants with homogeneous founder populations are shown with closed circles, and those with heterogeneous founders with open circles. Best-fit linear regressions are shown (solid lines) along with R² and P-values. (C) Boxplot of neutralization breadth measured 1 year and 3 years post-diagnosis for participants infected with homogeneous or heterogeneous founders. A significant pairwise difference seen with neutralization data obtained 3 years post-diagnosis is indicated with the associated P-value (Student’s t test).