Pre-clinical development of a recombinant, replication-competent adenovirus serotype 4 vector vaccine expressing HIV-1 envelope 1086 clade C

Abstract

Background: There is a well-acknowledged need for an effective AIDS vaccine that protects against HIV-1 infection or limits in vivo viral replication. The objective of these studies is to develop a replication-competent, vaccine vector based on the adenovirus serotype 4 (Ad4) virus expressing HIV-1 envelope (Env) 1086 clade C glycoprotein. Ad4 recombinant vectors expressing Env gp160 (Ad4Env160), Env gp140 (Ad4Env140), and Env gp120 (Ad4Env120) were evaluated.

Methods: The recombinant Ad4 vectors were generated with a full deletion of the E3 region of Ad4 to accommodate the env gene sequences. The vaccine candidates were assessed in vitro following infection of A549 cells for Env-specific protein expression and for posttranslational transport to the cell surface as monitored by the binding of broadly neutralizing antibodies (bNAbs). The capacity of the Ad4Env vaccines to induce humoral immunity was evaluated in rabbits for Env gp140 and V1V2-specific binding antibodies, and HIV-1 pseudovirus neutralization. Mice immunized with the Ad4Env160 vaccine were assessed for IFNγ T cell responses specific for overlapping Env peptide sets.

Results: Robust Env protein expression was confirmed by western blot analysis and recognition of cell surface Env gp160 by multiple bNAbs. Ad4Env vaccines induced humoral immune responses in rabbits that recognized Env 1086 gp140 and V1V2 polypeptide sequences derived from 1086 clade C, A244 clade AE, and gp70 V1V2 CASE A2 clade B fusion protein. The immune sera efficiently neutralized tier 1 clade C pseudovirus MW965.26 and neutralized the homologous and heterologous tier 2 pseudoviruses to a lesser extent. Env-specific T cell responses were also induced in mice following Ad4Env160 vector immunization.

Conclusions: The Ad4Env vaccine vectors express high levels of Env glycoprotein and induce both Env-specific humoral and cellular immunity thus supporting further development of this new Ad4 HIV-1 Env vaccine platform in Phase 1 clinical trials.

Figure 1. Ad4Env vector design.

The HIV-1 env gene sequence was derived from 1086 clade C and inserted into the Ad4 virus E3 region. The use of a shuttle plasmid encoding the Env sequence and the Ad4 plasmid to obtain the final vaccine product is described in Materials and Methods.

Figure 2. Western blot analysis of env transgene protein expression.

A549 cells were infected with Ad4Env120, Ad4Env140 or Ad4Env160 recombinant viruses and harvested two to three days later. Comparable amounts of each Env protein (approximately 2 µg) were run on a SDS-PAGE gel and transferred to nitrocellulose. Env proteins were labelled with a mouse anti-HIV Env 1086 clade C (VRC-C 3B3) antibody which was then detected with a goat anti-mouse HRP antibody.

Figure 3. Recognition of cell surface Env 1086 clade C glycoprotein by bNAbs.

A549 cells infected at a concentration 5 × 10⁸ vp/mL of Ad4Env160, (A) and Ad4Env160K→N (B). Cells were harvested 18 hours after infection and flow cytometry performed using 10, 1.0, 0.1, and 0.01 µg/mL concentrations of bNAb as primary and goat anti-human R-PE as secondary antibody. The reactivity of each bNAb at the highest concentration against the negative control Ad4wt-infected cells is shown in each panel (indicated by “wt 10 µg/mL”). Any non-specific background staining shown was not evident at lower concentrations of bNAbs. Binding of the secondary antibody alone is indicated by “2° Ab only”.

Figure 4. Immunogenicity of Ad4Env160, Ad4Env140, and Ad4Env120 recombinant viruses in rabbits.

Three rabbits per group were immunized on days 0 and 28 with each of the Ad4Env 1086 clade C recombinant vectors and booster immunizations were performed on day 84 with recombinant Env 1086 clade C 140 glycoprotein. Ad4Env vectors were administered by the i.m route with the exception of one group which also received the Ad4Env140 vector by the i.n. route. Positive control rabbits were immunized three times with the recombinant Env 1086 clade C gp140 in Rehydragel on days 0, 28 and 84. Negative control rabbits were immunized three times with Ad4wt virus, i.m., on days 0, 28 and 84; antibody titers were all <50 for this group. Serum samples were collected on days 0 (pre-vaccine), 28, 84 and 112. The ELISA coating antigens were: (A) Env 1086 clade C 140 glycoprotein; (B) Env140 1086 clade C V1V2 polypeptide; (C) Env A244 AE strain V1V2 polypeptide; and (D) clade B Env CASE A V1V2 scaffold fusion protein. For each of the ELISA antigens, statistical analysis using the t-test was performed comparing antibody titers induced by the Ad4Env160 vector versus titers induced by each of the other Ad4Env vectors or by the Env140 1086 clade C glycoprotein. Groups with a statistical difference (p ≤ 0.05) in geomean ED₅₀ antibody titers relative to Ad4Env160 vector immunogen are denoted by an asterisk.

Figure 5. Ad4Env160K→N vector-induced Env-specific cell-mediated immunity in mice.

C57BL/6 × BALB/c F1 (CB6F1) mice (6 animals/group) were immunized two times, days 0 and 28, with Ad4Env160K→N recombinant virus, Ad4wt virus, and recombinant Env gp140 (rgp140). On day 56 the mice were sacrificed and splenocytes pooled and tested in an IFNγ ELISPOT. Whole splenocytes (A), CD4⁺ (B), and CD8⁺ (C) T cells were evaluated. 15-mer peptides from the HIV-1 consensus clade C Env gp160 sequence, Env gp140 1086 clade C protein, and heat denatured Ad4wt virus were used as target antigens.