Characterizing the Relationship Between Neutralization Sensitivity and env Gene Diversity During ART Suppression

Abstract

Although antiretroviral therapy (ART) successfully suppresses HIV-1 replication, ART-treated individuals must maintain therapy to avoid rebound from an integrated viral reservoir. Strategies to limit or clear this reservoir are urgently needed. Individuals infected for longer periods prior to ART appear to harbor more genetically diverse virus, but the roles of duration of infection and viral diversity in the humoral immune response remain to be studied. We aim to clarify a role, if any, for autologous and heterologous antibodies in multi-pronged approaches to clearing infection. To that end, we have characterized the breadths and potencies of antibody responses in individuals with varying durations of infection and HIV-1 envelope (env) gene diversity as well as the sensitivity of their inducible virus reservoir to broadly neutralizing antibodies (bNAbs). Plasma was collected from 8 well-characterized HIV-1⁺ males on ART with varied durations of active infection. HIV envs from reservoir-derived outgrowth viruses were amplified and single genome sequenced in order to measure genetic diversity in each participant. IgG from plasma was analyzed for binding titers against gp41 and gp120 proteins, and for neutralizing titers against a global HIV-1 reference panel as well as autologous outgrowth viruses. The sensitivity to bNAbs of these same autologous viruses was measured. Overall, we observed that greater env diversity was associated with higher neutralizing titers against the global panel and also increased resistance to certain bNAbs. Despite the presence of robust anti-HIV-1 antibody titers, we did not observe potent neutralization against autologous viruses. In fact, 3 of 8 participants harbored viruses that were completely resistant to the highest tested concentration of autologous IgG. That this lack of neutralization was observed regardless of ART duration or viral diversity suggests that the inducible reservoir harbors 'escaped' viruses (that co-evolved with autologous antibody responses), rather than proviruses archived from earlier in infection. Finally, we observed that viruses resistant to autologous neutralization remained sensitive to bNAbs, especially CD4bs and MPER bNAbs. Overall, our data suggest that the inducible reservoir is relatively resistant to autologous antibodies and that individuals with limited virus variation in the env gene, such as those who start ART early in infection, are more likely to be sensitive to bNAb treatment.

Keywords: ART suppression; HIV-1; autologous antibodies; broadly neutralizing antibodies; diversity; env gene.

Figure 1

Genetic diversity of participant reservoir virus increases with active infection length. (A) A minimum of 3 SGS sequences were used to generate consensus sequences for each QVOA well. The protein sequence of each well was aligned to consensus subtype B and AG and a maximum likelihood tree was generated. Wells containing multiple viruses were excluded. Each dot represents one QVOA outgrowth virus well and is color-coded by participant. The tree is midpoint rooted for visualization and bootstraps over 90% are shown. (B) For all participants a minimum duration of infection was calculated using the date of diagnosis and date of ART initiation. For individuals for whom an HIV^- date was confirmed, an estimated length of infection was calculated using an estimated date of infection as the midpoint between confirmed HIV^- and HIV⁺ time points. For each person, these estimated times of active infection are listed next to the average pairwise distance (APD) of Env for their inducible reservoir viruses.

Figure 2

Association between reservoir diversity and bNAb resistance. Each virus was tested for neutralization sensitivity against a panel of 10 HIV-1 bNAbs. The geometric IC₅₀ of each individuals’ virus-bNAb pairing was calculated by bNAb epitope: (A) CD4bs antibodies include VRC01, VRC07-523, 3BNC117, and N6, (B) MPER antibody 10E8v4-V5F-100cF (because there is only one bNAb, IC₅₀ is plotted), (C) V2-apex antibodies PGDM1400, CAP256.VRC26.25, and PG09, and (D) V3-glycan targeting antibodies PGT121 and 10-1074. Geometric mean IC₅₀s for each virus are color-coded by individual and grouped by high (>6%APD) or low (<2% APD) Env diversity. Comparisons were performed by Mann-Whitney. * signifies p < 0.05. *** signifies p < 0.001. ns signifies p > 0.05.

Figure 3

All participants exhibit detectable anti-HIV antibodies. IgG binding for each person was assessed by indirect ELISA against (A) YU2 gp120 and (B) DIII gp41. EC₅₀s for each participant against gp120 (C) and gp41 (D) were graphed by high (>6%APD) and low (<2% APD) Env diversity. EC₅₀s for gp120 (E) and gp41 (F) were graphed against duration of ART. Comparisons were performed by Mann-Whitney and correlations by nonparametric Spearman.

Figure 4

Duration of infection and reservoir diversity are associated with neutralization of global HIV-1 viruses. Geometric mean IC₅₀ of IgGs against a global HIV-1 reference panel were calculated for each individual and graphed against (A) Env diversity and (B) duration of ART. Correlations were calculated by nonparametric Spearman.

Figure 5

Sensitivity of the inducible reservoir to autologous IgG and to bNAbs targeting multiple epitopes. (A) IC₅₀ values for each individual’s IgG against autologous QVOA-derived viruses. Viruses resistant at 50µg/mL of IgG were denoted with “>50”. (B–E) Geometric mean IC₅₀s of bNAbs were calculated for viruses resistant to autologous neutralization (IC₅₀ >35µg/mL) and those that were sensitive. IC₅₀s were graphed by epitope targeted: (B) CD4bs antibodies include VRC01, VRC07-523, 3BNC117, and N6, (C) V2-apex antibodies include PGDM1400, CAP256.VRC26.25, and PG09, (D) MPER antibody 10E8v4-V5F-100cF, and (E) V3-glycan antibodies include PGT121 and 10-1074. Geometric mean IC₅₀s are color-coded by individual, and the limit of detection (50μg/ml) is indicated with a dotted line. * signifies p < 0.05. *** signifies p < 0.001. ns signifies p > 0.05.