Kinetic rates of antibody binding correlate with neutralization sensitivity of variant simian immunodeficiency virus strains

Abstract

Increasing evidence suggests that an effective AIDS vaccine will need to elicit both broadly reactive humoral and cellular immune responses. Potent and cross-reactive neutralization of simian immunodeficiency virus (SIV) and human immunodeficiency virus type 1 (HIV-1) by polyclonal and monoclonal antibodies is well documented. However, the mechanisms of antibody-mediated neutralization have not been defined. The current study was designed to determine whether the specificity and quantitative properties of antibody binding to SIV envelope proteins correlate with neutralization. Using a panel of rhesus monoclonal antibodies previously characterized for their ability to bind and neutralize variant SIVs, we compared the kinetic rates and affinity of antibody binding to soluble envelope trimers by using surface plasmon resonance. We identified significant differences in the kinetic rates but not the affinity of monoclonal antibody binding to the neutralization-sensitive SIV/17E-CL and neutralization-resistant SIVmac239 envelope proteins that correlated with the neutralization sensitivities of the corresponding virus strains. These results suggest for the first time that neutralization resistance may be related to quantitative differences in the rates but not the affinity of the antibody-envelope interaction and may provide one mechanism for the inherent resistance of SIVmac239 to neutralization in vitro. Further, we provide evidence that factors in addition to antibody binding, such as epitope specificity, contribute to the mechanisms of neutralization of SIV/17E-CL in vitro. This study will impact the method by which HIV/SIV vaccines are evaluated and will influence the design of candidate AIDS vaccines capable of eliciting effective neutralizing antibody responses.

FIG. 1.

Quantitation of envelope in SIV/17E-CL and SIVmac239 virions. Virus particles were pelleted from cell culture supernatants and subjected to polyacrylamide gel electrophoresis under denaturing conditions prior to analysis by Western blotting. Reactivity with antibodies specific for Gag (p27) and envelope (gp120) demonstrated similar ratios of Gag to envelope for the two viruses.

FIG. 2.

Characterization of SIV rgp140 trimers. (A) Recombinant gp140 proteins from SIV/17E-CL and SIVmac239 were run under native conditions on polyacrylamide gel electrophoresis. Both rgp140 proteins migrated to approximately 420 kDa, while rgp120 migrated to the expected molecular size of 120 kDa. (B) SIV/17E-CL and SIVmac239 trimeric rgp140 proteins were characterized for binding to sCD4 by ELISA. Both trimeric proteins bound sCD4 at levels comparable to those of native viral (SIV/17E-CL) envelope proteins.

FIG. 3.

Relative affinities of rhesus MAb binding to SIV rgp140 in a ConA ELISA. Levels of binding to SIV/17E-CL rgp140 (solid bars) did not distinguish between neutralizing and nonneutralizing MAbs. In addition, no quantitative differences in rhesus MAb reactivity to SIV/17E-CL or SIVmac239 (shaded bars) rgp140 were observed.

FIG. 4.

MAb-rgp140 binding interactions do not discriminate between neutralizing and nonneutralizing MAbs. Kinetics of neutralizing (▪) and nonneutralizing (□) rhesus MAb binding to SIV/17E-CL identified no significant differences in the median k_a (A), k_d (B), or K_D (C).

FIG. 5.

Rates of MAb-rgp140 binding discriminate between neutralization-sensitive and neutralization-resistant SIV. (A and B) The kinetics of rhesus MAb binding to SIV/17E-CL (▪) and SIVmac239 (○) demonstrated higher k_a (A) and k_d (B) for SIVmac239 than for SIV/17E-CL; these differences were statistically significant (P = 0.0078 and 0.0156, respectively). (C) No difference in the median K_D was observed.

FIG. 6.

Lack of correlation between rhesus MAb affinities determined by SPR and ConA ELISA analyses. Rhesus MAb affinities for SIV/17E-CL rgp140 (▪) and SIVmac239 rgp140 (○) were determined by SPR (Biacore) and half-maximal binding in a ConA ELISA. As indicated by the regression analysis r² values (solid line, SIV/17E-CL; dashed line, SIVmac239), there was no correlation between affinities determined by SPR and those obtained in a solid-phase ELISA.