A CD4-mimetic compound enhances vaccine efficacy against stringent immunodeficiency virus challenge

Abstract

The envelope glycoprotein (Env) trimer ((gp120/gp41)₃) mediates human immunodeficiency virus (HIV-1) entry into cells. The "closed," antibody-resistant Env trimer is driven to more open conformations by binding the host receptor, CD4. Broadly neutralizing antibodies that recognize conserved elements of the closed Env are potentially protective, but are elicited inefficiently. HIV-1 has evolved multiple mechanisms to evade readily elicited antibodies against more open Env conformations. Small-molecule CD4-mimetic compounds (CD4mc) bind the HIV-1 gp120 Env and promote conformational changes similar to those induced by CD4, exposing conserved Env elements to antibodies. Here, we show that a CD4mc synergizes with antibodies elicited by monomeric HIV-1 gp120 to protect monkeys from multiple high-dose intrarectal challenges with a heterologous simian-human immunodeficiency virus (SHIV). The protective immune response persists for at least six months after vaccination. CD4mc should increase the protective efficacy of any HIV-1 Env vaccine that elicits antibodies against CD4-induced conformations of Env.

Fig. 1

Structure of the CD4mc, BNM-III-170, used in this study. The trifluoroacetate (TFA) salt of BNM-III-170 was synthesized from 5-bromo-1-indanone following the procedures in ref.

Fig. 2

CD4mc-dependent HIV-1 neutralization by immunized monkey sera. Monkeys were immunized with either human serum albumin (HSA) (Group 1) or the Clade C HIV-1_CH505 gp120 (Groups 2 and 3). In the experiments shown in the green curves, recombinant HIV-1 encoding luciferase and pseudotyped with the heterologous Clade B HIV-1_JR-FL Env (left panels) or with the control amphotropic murine leukemia virus (A-MLV) Env (right panels) was incubated with 20 μM BNM-III-170 for 30 min at 37 °C; in the experiments shown in the magenta curves, the viruses were incubated with the equivalent volume of DMSO under the same conditions. The viruses were then incubated for 30 min at 37 °C with the indicated dilution of sera from the immunized monkeys (collected on the day of SHIV-C5 Challenge 1). The virus-serum mixtures were added to Cf2Th-CD4/CCR5 cells and, 48 h after continuous incubation at 37 °C, the cells were lysed and luciferase activity measured. The percentage of luciferase activity (which reflects the level of virus infection) relative to that seen for viruses without added serum is shown. The means and standard deviations of values obtained from triplicate assays are shown. Differences between the sets of curves obtained in the absence and presence of BNM-III-170 were evaluated for statistical significance using the statmod software package for R (http://bioinf.wehi.edu.au/software/compareCurves). (ns not significant)

Fig. 3

CD4mc-dependent ADCC killing by immunized monkey sera. Primary CD4⁺ T lymphocytes from three healthy HIV-1-negative individuals were activated, and then infected with SHIV-C5 virus for 48 h before performing the antibody binding and ADCC assays. a Surface staining of infected cells was done with 1:250 dilutions of sera collected from the monkeys on the day of SHIV-C5 Challenge 1. Serum was added in the presence of 50 μM BNM-III-170 or an equivalent volume of DMSO (negative control) at 37 °C for 1 h. Goat anti-human antibody (AF-647) was used as the secondary antibody. The Mean Fluorescence Intensity (MFI) of AF-647 staining on Aqua Vivid (Invitrogen)-negative cells (living cells) is shown, relative to that of mock-infected cells. See Supplementary Fig. 9 for an example of the FACS gating strategy. b To evaluate ADCC, SHIV-C5-infected primary CD4⁺ T lymphocytes were incubated with autologous peripheral blood mononuclear cells (Effector:Target ratio of 10:1) from the same donor for 4–6 h at 37 °C in the presence of a 1:250 dilution of monkey serum and 50 μM BNM-III-170 or equivalent volume of DMSO. The serum was collected from the monkeys on the day of SHIV-C5 Challenge 1. Infected cells were identified by staining with an Alexa Fluor (AF)-488-conjugated antibody against SIV p27 CA. The percentage of ADCC killing was determined with the following formula: [(% of p27 ⁺ cells in Targets plus Effectors)−(% of p27 ⁺ cells in Targets plus Effectors plus serum)]/(% of p27 ⁺ cells in Targets) by gating infected (p27⁺) live (Aqua Vivid-negative) target cells, as described. c To evaluate the effect of the A32 Fab on the susceptibility of SHIV-C5-infected cells to ADCC mediated by the serum samples, 5 μg/ml of A32 Fab was added to the ADCC assay described in b. The results shown represent the average of three experiments. Statistical significance was tested by One-way ANOVA. (****P < 0.0001; ns, not significant)

Fig. 4

SHIV-C5 challenges and duration of protective immune responses. a Monkeys were immunized with either human serum albumin (Group 1) or gp120 glycoproteins corresponding to the Envs of the transmitted/founder virus and sequential virus isolates (at weeks 53, 78 and 100) from an HIV-1_CH505-infected individual (Groups 2 and 3) (green). Immunized monkeys were challenged with SHIV-C5 mixed with BNM-III-170 (Groups 1 and 3) or DMSO (Group 2) (blue). To accommodate the complete study in this schematic diagram, the time intervals on the horizontal axis are approximated. b Kaplan–Meier curves show the percentage of the monkeys in each group that remained uninfected after three high-dose intrarectal challenges with the heterologous Tier-2/3 transmitted/founder SHIV-C5. The monkeys were boosted with human serum albumin (HSA) (Group 1) or HIV-1_CH505 gp120 (Groups 2 and 3) either two weeks (Challenges 1 and 2) or four weeks (Challenge 3) before the SHIV-C5 challenge. Infection of the Group 2 monkeys occurred at a rate expected for naive monkeys challenged with the same dose (3.5 animal infectious dose (AID₅₀) units) of SHIV-C5^, . The indicated P values were obtained using the log rank test (ns, not significant). The rate of infection of the Group 3 monkeys was significantly lower than that expected if gp120 immunization and BNM-III-170 treatment were simply additive (P = 0.0186, log rank test). c The results of the three SHIV-C5 challenges shown in (b) were used to calculate the infection rate, which is the number of infections/number of exposures. Compared with unvaccinated historical controls, the Jewell bias-corrected relative risk of SHIV-C5 acquisition in the Group 2 monkeys is 0.95 ± 0.24 (95% confidence interval). d The percentage of Group 3 monkeys that remained uninfected after SHIV-C5 challenges at different times following the last gp120 immunization is shown. The green bars indicate SHIV-C5 challenges that were conducted in the presence of BNM-III-170 (300 μM) and the magenta bar indicates a SHIV-C5 challenge performed in the absence of BNM-III-170. Each of the values obtained with BNM-III-170 significantly differs from that obtained in the absence of the CD4mc (P < 0.05, one-tailed Fisher exact probability test). The values obtained for the SHIV-C5 challenges performed with BNM-III-170 (green bars) do not significantly differ from each other

Fig. 5

HIV-1 Env sequences of circulating SHIV-C5 in the infected monkeys. Plasma viral RNA was isolated from SHIV-C5-infected monkeys at the time of peak viremia and used for cDNA synthesis, single-genome amplification and env sequencing. The phylogenetic relationship of the env sequences is shown in a circular cladogram