Evaluating the antibody response elicited by diverse HIV envelope immunogens in the African green monkey (Vervet) model

Abstract

African Green (Vervet) monkeys have been extensively studied to understand the pathogenesis of infectious diseases. Using vervet monkeys as pre-clinical models may be an attractive option for low-resourced areas as they are found abundantly and their maintenance is more cost-effective than bigger primates such as rhesus macaques. We assessed the feasibility of using vervet monkeys as animal models to examine the immunogenicity of HIV envelope trimer immunogens in pre-clinical testing. Three groups of vervet monkeys were subcutaneously immunized with either the BG505 SOSIP.664 trimer, a novel subtype C SOSIP.664 trimer, CAP255, or a combination of BG505, CAP255 and CAP256.SU SOSIP.664 trimers. All groups of vervet monkeys developed robust binding antibodies by the second immunization with the peak antibody response occurring after the third immunization. Similar to binding, antibody dependent cellular phagocytosis was also observed in all the monkeys. While all animals developed potent, heterologous Tier 1 neutralizing antibody responses, autologous neutralization was limited with only half of the animals in each group developing responses to their vaccine-matched pseudovirus. These data suggest that the vervet monkey model may yield distinct antibody responses compared to other models. Further study is required to further determine the utility of this model in HIV immunization studies.

Figure 1

Design, expression and characterization of the CAP255 trimer. (A) The The CAP255 SOSIP.664 trimer was expressed in HEK 293F cells and purified by nickel affinity chromatography followed by size exclusion chromatography. The purity of the trimer was assessed by SDS-PAGE. (B) An HIV pseudovirus-based neutralization assay was used to determine the neutralization titer of monoclonal antibodies against the CAP255 pseudovirus. IC₅₀ values are shown in μg/ml with resistance being > 50 μg/ml (blue) and increasing sensitivity from 0.1–0.9 μg/ml (orange) to < 0.09 μg/ml (red). Trimer-specific antibodies are labelled as Y (yes) and monomer-specific antibodies are labelled as N (No). (C) Monoclonal antibody binding responses towards the trimer were assessed by ELISA. An antibody bead depletion assay was used to assess the binding of conformational antibody, PGT151 (D) A Spearman correlation was conducted in Graphpad Prism v9.5.1 to assess the relationship between binding and neutralizing titers. (E) Differential scanning calorimetry was used to assess the melting temperature of the CAP255 trimer.

Figure 2

Immunization and sampling schedule for the vervet monkey model. (A) Vervet monkeys were assigned into groups with Group 1A receiving BG505 SOSIP.664 only, Group 1B receiving CAP255 SOSIP.664 only and Group 1C receiving a combination between BG505, CAP255 and CAP256.SU SOSIP.664 trimers. Each group received their immunogen(s) through 2 subcutaneous injections of 50 μg (total = 100 μg) of trimer during each vaccination and 10 ml of blood was collected at each blood draw. (B) Group 1A (n = 4), Group 1B (n = 5) and Group 1C (n = 4) each had a distribution of male and female vervet monkeys within an age range of 5–12 years of age.

Figure 3

Immunization with SOSIP.664 trimers elicits cross-reactive binding antibodies. The presence of HIV Env-specific binding antibodies in plasma from BG505 (Group 1A), CAP255 (Group 1B) and or BG505, CAP255 and CAP456.SU combination (Group 1C) immunized vervet monkeys against homologous (red boxes) and heterologous antigens was assessed using an in-house ELISA. Binding is represented by an OD450nm value.

Figure 4

Comparison of binding antibody responses between immunization groups. Binding antibody responses measured through ELISA for each timepoint tested. The area under the curve was calculated for each animal across all timepoints and this was compared between Groups 1A, 1B and 1C for each antigen tested: (A) BG505, (B) CAP255, (C) CAP256.SU and (D) 1086C. Statistical significance was calculated in Graphpad Prism v9.5.1 using the Kruskal–Wallis test with Dunn’s multiple comparison test where *p < 0.05, **< 0.01, ***< 0.001, ****< 0.0001.

Figure 5

Immunization with SOSIP.664 trimers triggers antibodies with preserved antibody dependent cellular phagocytosis over time. ADCP activity was assessed in plasma from BG505 (Group 1A), CAP255 (Group 1B) and or BG505, CAP255 and CAP456.SU combination (Group 1C) immunized vervet monkeys against homologous (red boxes) and heterologous antigens. ADCP was measured longitudinally before and after the various immunizations. An ADCP score was calculated as the percentage of THP-1 cells that engulf trimer-coated beads multiplied by the geometric mean fluorescence intensity (MFI).

Figure 6

Comparison of antibody dependent cellular phagocytosis responses between immunization groups. ADCP activity was measured and the area under the curve was calculated for each animal across all timepoints and this was compared between Groups 1A, 1B and 1C for each antigen tested: (A) BG505, (B) CAP255 and (C) CAP256.SU. Statistical significance was calculated Graphpad Prism v9.5.1 using the Kruskal–Wallis test with Dunn’s multiple comparison test where *p < 0.05, **< 0.01, ***< 0.001, ****< 0.0001.

Figure 7

Neutralizing antibody titers following vervet monkey trimer immunizations. Plasma from the pre-immunization bleed and from after the 3^rd immunization were tested in our HIV pseudovirus assay for neutralization activity against 6 pseudoviruses; three tier 1 viruses: 6644.v2.c33, MW965.26 and SF162 and three Tier 2 viruses: BG505 + N332 (BG505), CAP255.2000.4D (CAP255) and CAP256.SU. The names in red indicate the vaccine-matched pseudovirus for each vervet monkey group. ID₅₀ values are shown with resistance being > 20 (blue) and increasing sensitivity from 20–100 (yellow); 101–300 (orange); 301–1000 (red) and > 1000 (maroon).