A naturally arising broad and potent CD4-binding site antibody with low somatic mutation

Abstract

The induction of broadly neutralizing antibodies (bNAbs) is a potential strategy for a vaccine against HIV-1. However, most bNAbs exhibit features such as unusually high somatic hypermutation, including insertions and deletions, which make their induction challenging. VRC01-class bNAbs not only exhibit extraordinary breadth and potency but also rank among the most highly somatically mutated bNAbs. Here, we describe a VRC01-class antibody isolated from a viremic controller, BG24, that is much less mutated than most relatives of its class while achieving comparable breadth and potency. A 3.8-Å x-ray crystal structure of a BG24-BG505 Env trimer complex revealed conserved contacts at the gp120 interface characteristic of the VRC01-class Abs, despite lacking common CDR3 sequence motifs. The existence of moderately mutated CD4-binding site (CD4bs) bNAbs such as BG24 provides a simpler blueprint for CD4bs antibody induction by a vaccine, raising the prospect that such an induction might be feasible with a germline-targeting approach.

Fig. 1.. Isolation and characterization of antibody BG24 from donor 391370.

(A) Neutralization data of donor 391370’s and patient 3’s purified IgG against a 20-virus fingerprinting panel (f61). The average median inhibitory concentrations (IC₅₀s) in micrograms per milliliter are shown from duplicate neutralization measurements. NT, not tested. (B) Fingerprinting analysis of f61 neutralization for patient 3 and donor 391370. (C) Plasma viral load and peripheral blood CD4⁺ T cell counts of donor 391370 over time. The arrow indicates the time point for BG505 SOSIP.664 bait sorting. (D) Sorting of single BG505 SOSIP.664⁺ IgG⁺ memory B cells. (E) Neutralization breadth and potency of BG24 on a 126-virus cross clade panel. Neutralization testing performed in duplicates, average shown. (F) Somatic hypermutation (SHM) analysis of VRC01-like bNAbs, shown as % amino acid changes relative to putative germline variable gene sequence.

Fig. 2.. BG24 recognition of HIV-1 Env has features in common with VRC01-class bNAbs.

(A) Side and top views of the 3.9-Å x-ray structure of the BG24–BG505–10-1074 complex colored by components (dark gray, gp41; light gray, gp120; shades of blue, 10-1074 Fab; shades of brown, BG24 Fab). (B) Surface representation of gp120 (gray), with main loops at the CD4bs colored (yellow, CD4bs loop; blue, D loop; orange, V5 loop) and BG24 shown as cartoon representation. N-linked glycans modeled in the structure are shown as cyan sticks with electron density contoured at 1.5σ. (C) Stick representation of residue level contacts between VRC01-class signature residues in BG24_HC (brown) with gp120 (gray). Dashed black lines indicated potential for H-bond interactions.

Fig. 3.. BG24’s CDR3 sequence motifs are uncommon among VRC01-like bNAbs.

(A) Stick representation of residue level contacts between residues in BG24’s CDRL3 loop (yellow) with gp120 (gray). Potential H-bond interactions are shown as black dashed lines. VRC01 CDRH3 is also shown in (B) (cyan). (B) Stick representation of residue level contacts between residues in BG24’s CDRH3 loop (brown) with gp120 (gray). The CDRH3 loop of VRC01 (PDB 6VX8) is also shown (cyan). Potential H-bond interactions are shown as black dashed lines. (C) CDRH3 sequence alignment of VRC01-like antibodies that lack a Trp residue at the −5 position. (D) Neutralization data for in-common (n = 61) cross-clade viruses of BG24- and VRC01-like antibodies that lack a Trp residue at the −5 position. The geometric mean IC₅₀ value against antibody-sensitive strains is indicated by the horizontal black line. The percentage of non-neutralized strains is indicated on the top for each antibody. Analysis of neutralization and graphing was done using the Antibody Database (v 2.0) (73).

Fig. 4.. Improvements to BG24 neutralization potency and breadth.

(A) Neutralization data of engineered BG24 constructs against the global 12-virus panel. The average mean IC₅₀s in micrograms per milliliter are shown from duplicate neutralization measurements. (B) Side and top views of the 3.5-Å single-particle cryo-EM reconstruction of the BG24_PG20-CDR2-v2–DU422–10-1074 complex colored by components (dark gray, gp41; light gray, gp120; shades of blue, 10-1074 Fab; shades of brown, BG24 Fab). (C) Cartoon and stick representation of BG24_PG20-CDR2-v2 CDRH2 (wheat) at the gp120 (gray) interface. The CDRH2 loop from the BG24-BG505 complex (cyan) is overlaid with amino acid mutations between the two constructs labeled. Potential H-bond interactions are shown as black dashed lines. (D) Modeling of the Phe⁴³ gp120 pocket with BG24_G54W pocket-filling mutation highlighted (wheat). Electron density contoured at 7σ is shown for gp120 and BG24. (E) Surface representation of gp120 Phe⁴³ pocket (gray) and BG24_G54W pocket-filling mutation (wheat).

Fig. 5.. BG24 has comparable in vivo efficacy to VRC01 in HIV_YU2-infected humanized mice.

(A) Antibody monotherapy of humanized mice infected with HIV_YU2. Left graphs show absolute viremia (y axis) in mice treated with antibody monotherapy (BG24, n = 6, dark blue; VRC01, n = 6, dark green) or untreated control mice (n = 6, gray) over the course of the experiment (x axis, days). Right graphs show relative log drop after initiation of antibody therapy (∆log₁₀ copies/ml). Thick blue/green and thick dashed gray lines indicate the mean viral load of treated and untreated mice, respectively. Mice were infected 3 weeks before therapy initiation and received 1 mg of IgG as a loading dose followed by biweekly administration of 0.5 mg for 3 weeks. The dotted line at the bottom indicates the limit of accuracy of the quantitative polymerase chain reaction assay (384 copies/ml). Data from one experiment. (B) Plasma HIV-1 Env sequences obtained 4 weeks after initiation of therapy from mice treated with BG24 (top) and VRC01 (bottom), respectively. Letters show amino acid mutations relative to the HIV_YU2 molecular clone. Residues numbered according to HIV-1_HXB2. (C) HIV-1 Env sequences obtained from donor plasma RNA. Letters indicate amino acid mutations compared with consensus clade B (blue letters) shown on top. Black letters in the consensus sequence indicate amino acids also observed at each position with lower frequencies. Gray boxes indicate gaps. Yellow columns indicate PNGSs. Residues are numbered according to HIV-1_HXB2.

Fig. 6.. The role of SHM in BG24 recognition of gp120.

(A) Sequence alignment of BG24 and BG24 minimally mutated constructs with germline sequences. Somatic mutations compared to germ line are shown with BG24 paratope residues derived from SHM shown as red. Complete paratope is labeled below sequence alignment. (B) Paratope residues from germline V genes (green), somatic mutation (red), and CDR3 loops (orange) are shown as sticks on BG24. (C to F) Representative mutations that increase contacts of mature BG24 with CD4bs. Model of germline BG24 (pink) was superposed with the BG24–BG505–10-1074 structure. (G) Neutralization data of engineered minimally mutated BG24 constructs against the global 12-strain viral panel. The average mean IC₅₀s in micrograms per milliliter are shown from duplicate neutralization measurements.