Broadly neutralizing antibodies targeting the HIV-1 envelope V2 apex confer protection against a clade C SHIV challenge

Abstract

Neutralizing antibodies to the V2 apex antigenic region of the HIV-1 envelope (Env) trimer are among the most prevalent cross-reactive antibodies elicited by natural infection. Two recently described V2-specific antibodies, PGDM1400 and CAP256-VRC26.25, have demonstrated exquisite potency and neutralization breadth against HIV-1. However, little data exist on the protective efficacy of V2-specific neutralizing antibodies. We created a novel SHIV-325c viral stock that included a clade C HIV-1 envelope and was susceptible to neutralization by both of these antibodies. Rhesus macaques received a single infusion of either antibody at three different concentrations (2, 0.4, and 0.08 mg/kg) before challenge with SHIV-325c. PGDM1400 was fully protective at the 0.4 mg/kg dose, whereas CAP256-VRC26.25-LS was fully protective even at the 0.08 mg/kg dose, which correlated with its greater in vitro neutralization potency against the challenge virus. Serum antibody concentrations required for protection were <0.75 μg/ml for CAP256-VRC26.25-LS. These data demonstrate unprecedented potency and protective efficacy of V2-specific neutralizing antibodies in nonhuman primates and validate V2 as a potential target for the prevention of HIV-1 infection in passive immunization strategies in humans.

Figure 1. Highlighter amino acid sequence alignment of env derived from the SHIV-325c stock and the parental HIV-1 env

Amino acid substitutions that differ from the parental HIV-1 CA325 env sequence are indicated in color. Dashes indicate amino acid sequences identical to the parental sequence. A total of 25 sequences from SHIV-325c were generated by SGA.

Figure 2. Plasma viral loads and CD4+ T-cell counts in rhesus macaques challenged with SHIV-325c

(A) All animals were infected after challenge with SHIV-325c and 6 out of 8 animals established chronic infection with mean viral loads set-point of 3.9 log RNA copies/mL. The red line represents the mean log RNA copies/mL. (B) During acute infection mean CD4+ T cell numbers declined by 27%.

Figure 3. Neutralization profiles of PGDM1400 and CAP256-VRC26.25 against a multi-clade panels of 208 pseudoviruses (A) and highlighting specifically the activity of both bNAbs against clade C pseudoviruses (B)

(A) Each bar represents the IC₈₀ (μg/ml) of PGDM1400 (upper graph) or CAP256-VRC26.25 (lower graph) against a single virus. Viruses are ranked according to increasing IC₈₀ values. Bars reaching the dotted line represent IC₈₀ values >50 μg/mL. The red bars highlight the IC₈₀ values for PGDM1400 (upper graph) or CAP256-VRC26.25 (lower graph) against SHIV-325c. (B) IC₈₀ values for PGDM1400 and CAP256-VRC26.25 against SHIV-325c (red dot) and against the clade C pseudoviruses included in panel (A). The horizontal bars represent the mean of IC₈₀ values of neutralizable pseudoviruses.

Figure 4. Neutralization of SHIV-325c by CAP256-VRC26.25, PGDM1400 and PG9

Neutralization was measured using replication competent challenge stock SHIV-325c infection of TZM-bl cells. All 3 antibodies showed 100% neutralization of SHIV-325c with no evidence of a plateau effect below 100% neutralization.

Figure 5. Protective efficacy of PGDM1400 and CAP256-VRC26.25-LS against SHIV-325c in rhesus macaques

Plasma viral RNA (log RNA copies/mL) are shown for animals that received placebo control (A), 2 mg/kg PGDM1400 (B), 0.4 mg/kg PGDM1400 (C), 0.08 mg/kg PGDM1400 (D), 2 mg/kg CAP256-VRC26.25-LS (E), 0.4 mg/kg CAP256-VRC26.25-LS (F), or 0.08 mg/kg CAP256-VRC26.25-LS (G). All placebo controls became infected. Protection was observed in all animals that received CAP256-VRC26.25-LS at all doses. 1/5 animals in the PGDM1400 2 mg/kg dose group and 3/4 animals in the 0.08 mg/kg PGDM1400 dose group were infected. The assay sensitivity limit was >50 RNA copies/mL.

Figure 6. Serum concentration of CAP256-VRC26.25-LS and PGDM1400 in antibody treated animals

bNAb concentrations were determined by ELISA. The results show average serum concentrations of 19.8 μg/ml, 3.3 μg/ml and 0.75 μg/mL on the day of challenge in the CAP256-VRC26.25-LS 2 mg/kg, 0.4 mg/kg and 0.08 mg/kg groups, respectively, and 6.9 μg/mL, 2.5 μg/mL and 0.22 μg/mL on the day of challenge in the PGDM1400 2 mg/kg, 0.4 mg/kg and 0.08 mg/kg groups, respectively.

Figure 7. Analysis of SHIV-325c V1V2 envelope sequences in breakthrough infections

Single genome envelope amplification was performed with plasma viral RNA on day 28. For one placebo control animal, H515, viral amplification failed and therefore is not shown here. The env sequence at the top presents the SHIV-325c molecular clone inoculum. The column on the right reports the frequency of distinct amplicons (based on unique sequence variations in V1/V2) per total env amplicons obtained from a given animal. Most sequence variation is observed in residues 165–175 in the V2 region. Mutations at position 167 are associated with escape to CAP256 V2-lineage antibodies and but are present in the challenge stock, PGDM1400 and placebo treated animals. Interestingly, animal M97, which developed infection in the highest PGDM1400 dose group had a fixed F210L mutation in all 27 amplicons (marked with ), that was not present in the challenge molecular clone or developed in any other animal.

Figure 8. Potency and breadth profiles of single and combination bNAbs against 200 clade C HIV-1 Env pseudoviruses

(A) Heatmaps of IC₈₀ values for single bNAbs and bNAb combinations respectively. Rows represent Env pseudoviruses, and columns represent single and combination bNAbs. Darker hues of red indicate more potent neutralization and grey cells indicate IC₈₀ above threshold. (B) Potency-breadth curves for single bNAbs and bNAb combinations are shown. IC₈₀ scores for combinations and single bNAbs were compared using Wilcoxon rank sum test. Antibody a, PGDM1400; Antibody b, CAP256-VRC26.25; Antibody c, PGT121.