Neutralizing antibody responses in macaques induced by human immunodeficiency virus type 1 monovalent or trivalent envelope glycoproteins

Abstract

A major goal of efforts to develop a vaccine to prevent HIV-1 infection is induction of broadly cross-reactive neutralizing antibodies (bcnAb). In previous studies we have demonstrated induction of neutralizing antibodies that did cross-react among multiple primary and laboratory strains of HIV-1, but neutralized with limited potency. In the present study we tested the hypothesis that immunization with multiple HIV-1 envelope glycoproteins (Envs) would result in a more potent and cross-reactive neutralizing response. One Env, CM243(N610Q), was selected on the basis of studies of the effects of single and multiple mutations of the four gp41 glycosylation sites. The other two Envs included R2 (subtype B) and 14/00/4 (subtype F), both of which were obtained from donors with bcnAb. Rhesus monkeys were immunized using a prime boost regimen as in previous studies. Individual groups of monkeys were immunized with either one of the three Envs or all three. The single N610Q and N615Q mutations of CM243 Env did not disrupt protein secretion, processing into, or reactivity with mAbs, unlike other single or multiple deglycosylation mutations. In rabbit studies the N610Q mutation alone or in combination was associated with an enhanced neutralizing response against homologous and heterologous subtype E viruses. In the subsequent monkey study the response induced by the R2 Env regimen was equivalent to the trivalent regimen and superior to the other monovalent regimens against the virus panel used for testing. The 14/00/4 Env induced responses superior to CM243(N610Q). The results indicate that elimination of the glycosylation site near the gp41 loop results in enhanced immunogenicity, but that immunization of monkeys with these three distinct Envs was not more immunogenic than with one.

Figure 1. Deglycosylation mutants of HIV-1 strain CM243 gp140 Env.

Panel A: Locations of point mutations in the gp41 ectodomain coding sequence that eliminate potential N-linked glycosylation sites. Panel B: Reduced SDS-PAGEanalysis of selected mutants of immunoprecipitated metabolically-labeled CM243 gp140 envelope glycoproteins using polyclonal rabbit antisera. Glycoproteins were transiently expressed in cells by plasmid transfection and infection with recombinant vaccinia viruses. Panel C: Analysis of precursor processing of metabolically-labeled CM243 gp140 wild type and glycosylation mutants by immunoprecipitation and reduced SDS-PAGEanalysis using a gp41 specific mAb (D61). Arrows point to gp41 ectodomains. Wild type CM243, single mutant N610Q and N615Q were processed to gp120 and gp41 ectodomain subunits. Where processing was evident the single N610Q resulted in the largest shift in apparent molecular weight of the gp41 ectodomain, an approximately 5–6 kDa loss. Panel D: Cell-cell fusion activity of selected CM243 mutants and wild type full length Envs using the β-gal reporter gene assay in transient transfected expressed constructs.

Figure 2. Monoclonal antibody and CD4 binding of the deglycosylation mutants of HIV-1 strain CM243 gp140 Envs.

A. Reactivity of CM243 gp140 wild type and glycosylation mutants to a panel of conformation-dependent and -independent monoclonal antibodies by immunoprecipitation and reduced SDS-PAGE analysis of metabolically labeled gp140s and autoradiography. While equivalent reactivities with mAbs D19, T50, D38, and T3 were seen in both wild type and mutant gp140 s, differential reactivities were found with mAbs that map to the Cluster I region in gp41. These are conformation dependent, oligomeric specific anti-gp41 antibodies. B. Analysis of mAb binding to CD4i epitopes following sCD4 binding to purified gp140 proteins by immunoprecipitaion and SDS-PAGE and western blot detection using rabbit polyclonal antisera.

Figure 3. Induction of neutralizing antibody responses by mutant forms of CM243 Env containing deglycosylation mutations in NZW rabbits.

There were five groups of three animals each: the groups were wild type gp140 (CM243), N610/615/624/636Q, N615Q, N610/615Q and N610Q. Rabbits were immunized four times with the indicated mutant gp140s. and sera were obtained for testing 2 weeks after the fourth doses. Sera were tested in serial two-fold dilutions for neutralization of virus pseudotyped with CM243 Env. Results obtained from the gp140 immunized rabbits are shown in black circles and for control (adjuvant only) rabbits in open circles. Results shown are means and standard deviations of three separate assays, each in triplicate, with sera collected following the final immunization.

Figure 4. Immunoglobulin G responses of monkeys measured by enzyme linked immunosorbent assay using gp140 antigens of strains R2, 14/00/4, and CM243.

A. Geometric mean titers are shown for the 6 monkeys in each of the immunization groups against each of the gp140 antigens. B. Ratios of GMT of sera of monkeys in group receiving the same antigen as the test/GMT of sera from monkeys receiving the heterologous antigens. Sera were substantially more reactive with homologous than heterologous antigens, and trivalent immunization resulted in responses against each antigen similar to homologous monovalent immunization.

Figure 5. Neutralizing antibody responses of monkeys against viruses pseudotyped with subtype B, C, and A/E envelope glycoproteins.

Sera were tested in serial dilutions starting at 1∶5. Sera with less than 50% neutralizing activity at 1∶5 were assigned inhibitory dilution 50% (ID₅₀) of 2.5 (log dilution = 0.4). Dotted vertical lines indicate groupings of Clade B, C, and A/E pseudoviruses tested for neutralization.