Structural survey of HIV-1-neutralizing antibodies targeting Env trimer delineates epitope categories and suggests vaccine templates

Abstract

HIV-1 broadly neutralizing antibodies are desired for their therapeutic potential and as templates for vaccine design. Such antibodies target the HIV-1-envelope (Env) trimer, which is shielded from immune recognition by extraordinary glycosylation and sequence variability. Recognition by broadly neutralizing antibodies thus provides insight into how antibody can bypass these immune-evasion mechanisms. Remarkably, antibodies neutralizing >25% of HIV-1 strains have now been identified that recognize all major exposed surfaces of the prefusion-closed Env trimer. Here we analyzed all 206 broadly neutralizing antibody-HIV-1 Env complexes in the PDB with resolution suitable to define their interaction chemistries. These segregated into 20 antibody classes based on ontogeny and recognition, and into 6 epitope categories (V1V2, glycan-V3, CD4-binding site, silent face center, fusion peptide, and subunit interface) based on recognized Env residues. We measured antibody neutralization on a 208-isolate panel and analyzed features of paratope and B cell ontogeny. The number of protruding loops, CDR H3 length, and level of somatic hypermutation for broadly HIV-1 neutralizing antibodies were significantly higher than for a comparison set of non-HIV-1 antibodies. For epitope, the number of independent sequence segments was higher (P < 0.0001), as well as the glycan component surface area (P = 0.0005). Based on B cell ontogeny, paratope, and breadth, the CD4-binding site antibody IOMA appeared to be a promising candidate for lineage-based vaccine design. In terms of epitope-based vaccine design, antibody VRC34.01 had few epitope segments, low epitopeglycan content, and high epitope-conformational variability, which may explain why VRC34.01-based design is yielding promising vaccine results.

Accurate Prediction of Antibody Resistance in Clinical HIV-1 Isolates

“…Structural studies revealed two of the three amino acid positions were located at the VRC01 epitope and thus can be critical to VRC01 binding and neutralization ( Figure 1c) (12). Additionally, the top three features of bNAb 8ANC195 classifier account for a total variable importance of 48.47% and include Env residues 234 and 276, which must be glycosylated in order for 8ANC195 to bind and neutralize Env (Figure 1d, Supplementary Table S2) (13). Not all of the top discriminative features, however, involved bNAb epitope residues, suggesting that distal regions outside the epitope also affect neutralization susceptibility of HIV-1 strains ( Supplementary Table S4).…”

“…The buried surface area between antibody and antigen was calculated using NACCESS software (12,13). The epitope and paratope residues for each antibody were defined as residues with non-zero buried surface area.…”

Beyond Hot Spots: Biases in Antibody Somatic Hypermutation and Implications for Vaccine Design

“…In particular, antibodies capable of broad neutralization of HIV have particularly high levels of SHM ( 128 , 131 , 132 ) and tend to be enriched for rare substitutions ( 104 , 133 , 134 ) (Figure 1 ). Extraordinary levels of SHM (15–35% nucleotide mutations) are characteristic of antibodies targeting HIV ( 135 ), and elevated levels of SHM have also been observed in other types of chronic infection and in systemic autoimmune disorders ( 136 ).…”

“…Indeed, while several recent studies in transgenic mice have elicited B cells enriched for substitutions present in the targeted mature antibody ( 144 – 146 ), none have yet specifically elicited critical rare substitutions or fully recapitulated the neturalization activity of the target antibodies. Notably, however, the most successful example focuses on PGT121 ( 146 ), which contains fewer rare substitutions than many other HIV antibodies ( 104 , 134 ). It may, therefore, be more prudent to choose lineage-based vaccine design targets by avoiding those with functionally important rare substitutions ( 33 , 134 ).…”

“…Notably, however, the most successful example focuses on PGT121 ( 146 ), which contains fewer rare substitutions than many other HIV antibodies ( 104 , 134 ). It may, therefore, be more prudent to choose lineage-based vaccine design targets by avoiding those with functionally important rare substitutions ( 33 , 134 ).…”