Germline-targeting HIV-1 Env vaccination induces VRC01-class antibodies with rare insertions

Abstract

Targeting germline (gl-) precursors of broadly neutralizing antibodies (bNAbs) is acknowledged as an important strategy for HIV-1 vaccines. The VRC01-class of bNAbs is attractive because of its distinct genetic signature. However, VRC01-class bNAbs often require extensive somatic hypermutation, including rare insertions and deletions. We describe a BG505 SOSIP trimer, termed GT1.2, to optimize binding to gl-CH31, the unmutated common precursor of the CH30-34 bNAb lineage that acquired a large CDRH1 insertion. The GT1.2 trimer activates gl-CH31 naive B cells in knock-in mice, and B cell responses could be matured by selected boosting immunogens to generate cross-reactive Ab responses. Next-generation B cell sequencing reveals selection for VRC01-class mutations, including insertions in CDRH1 and FWR3 at positions identical to VRC01-class bNAbs, as well as CDRL1 deletions and/or glycine substitutions to accommodate the N276 glycan. These results provide proof of concept for vaccine-induced affinity maturation of B cell lineages that require rare insertions and deletions.

Keywords: HIV-1 Env; NGS; bNAbs; deletions; germline-targeting; insertions; mouse model; neutralizing antibodies; vaccines.

Graphical abstract

Figure 1

Design and antigenicity of BG505 SOSIP germline trimer 1.2 (GT1.2) (A) Schematic linear representation of BG505 SOSIP.v4.1 GT1.2 with glycan occupancy data. All amino acid mutations compared with BG505 SOSIP.664 also present in BG505 SOSIPv4.1 GT1 are indicated in black, whereas the N279D substitution defines GT1.2. The glycan icons on top of the linear GT1.2 sequence represent the predominant type of glycan observed at that specific potential N-glycosylation site. ND, not determined. (B) Negative-stain electron micrograph of soluble GT1.2 trimers. (C) Surface plasmon resonance sensorgrams showing the specific binding signal in response units (Rus) on the y axes as a function of time during association and dissociation on the x axes for each of the antibody concentrations used (0.98 nM–1,000 nM). (D) Crystal structure of BG505 SOSIP.v4.1-GT1.2 (blue) trimer in complex with gl-PGV20 (orange) and PGT124 (cyan) Fabs at 3.8 Å resolution. (E) Side view of the crystal structure of the gl-PGV20 Fab bound to of GT1.2 (blue). (Right) Close-up view of the W100b hydrogen bond interactions of gl-PGV20 with N279 and N280 on GT1.2. (F) Superimposition of GT1.2 (blue), BG505 SOSIP.664 (orange) (PDB: 5CEZ), and GT1 (gray) (PDB: 5W6D).

Figure 2

GT1.2 primes CD4bs-directed VRC01-class serum responses in a gl-CH31 KI mouse model (A) Schematic of the simple prime-polish immunization regimen. The immunizations are indicated by syringes. (B) Normalized area under the curve (AUC) values of serum antibody binding to the indicated Env as measured by ELISA, for time point T1 (left) and T4 (right). (C) Midpoint titers (ID₅₀) of serum against VRC01-class signature viruses as described previously (LaBranche et al.⁵⁴) for each time point (T1, T4, and T5). (D) Schematic of the complex sequential immunization regimen. (E–G) Normalized AUC values of serum antibody binding to the indicated Env as measured by ELISA, for time point T2 (E), T3 (F), or T4 (G). The background color represents the priming group (GT1.2 trimer/NP vs. BG505). (H) CD4bs specificity of mice primed with GT1.2 (blue, n = 16) or BG505 (gray, n = 8) for each time point, represented by the BG505/BG505 D368R AUC ratio as measured by ELISA. (I) CD4bs specificity of mice primed with GT1.2 (blue, n = 16) or BG505 (gray, n = 8) at T7 for each of BG505 (clade A), AMC008 (clade B), or ZM197M (clade C) Envs as in (H). (J) Midpoint neutralization titers (ID₅₀) of serum from mice primed with GT1.2 (blue, n = 16) or BG505 (gray, n = 8) against VRC01-class signature viruses as in (C) for each time point (T2/T6/T7). Each dot represents an individual mouse.

Figure 3

GT1.2 priming but not BG505 priming selects for rare VRC01-class sequence features, including multi-residue insertions and deletions (A) Simplified representation of the immunization regimen used in Figures 2A–2D. (B) Violin plot showing the number of total amino acid substitutions in the IGHV region for each group (left) and dot plot showing the median IGHV mutation frequency (%) for each mouse in separate groups (right). (C) Dot plot showing the mean number of improbable mutations per IGHV region per mouse defined as a <2% probability in the absence of selection. (D) Total and VRC01-class amino acid mutations in the IGHV1-2 region for recovered IgG sequences for each of the groups as outlined in (A). The staggered black line shows the expected level of VRC01-class mutations as expected to be introduced by random SHM in IGHV1-2 (Briney et al.³⁷). (E) Stacked bar graph showing the frequency of sequences per group selecting for VRC01-class mutations. N/A, not applicable; defined by the limits of the model in Briney et al. (F) Dot plots showing the frequency of highly infrequent mutational events in recovered sequences. Each dot represents an individual mouse. (G) Dot plots showing the frequency of substitutions at known CH31 contact residues. Each dot represents an individual mouse. ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗∗, p < 0.001.

Figure 4

GT1.2-primed VRC01-class mAbs display diverse binding and neutralization capacities (A) Total and VRC01-class amino acid mutations in the IGHV1-2 region as in Figure 3D for the 52 selected mAbs, including minVRC01 (purple), min12A21 (green), and BG24 (orange). (B) Stacked bar graph (left) showing whether the residue at the positions indicated on the x axis are germline (green) has mutated into a key VRC01-class residue (purple) or another residue (gray). (Right) Distribution of the number of key mutations in each individual mAb. (C) Amino acid alignment showing multi-residue insertions (red) in the CDRH1 (top) or in the FWR3 (bottom) of selected mAbs. (D) Midpoint binding titers (EC₅₀) of each of the selected mAbs. The color corresponds to the starting concentration used (orange, 1 μg/mL; green, 10 μg/mL; purple, 50 μg/mL). (E) Midpoint neutralization titers (IC₅₀) of a subset of selected mAbs. Each dot corresponds with an individual mAb. (F) Midpoint neutralization titers (IC₅₀) of a subset of selected mAbs against VRC01-class signature viruses. (G) Midpoint neutralization titers (IC₅₀) of a subset of selected mAbs against an N276Q global panel as indicated on the x axis.

Figure 5

Rare insertions and deletions are important for mAb-Env interactions and might establish quaternary contacts (A) ELISA binding to Envs expressed as the area under the curve (AUC) for each of the original mAbs and mAbs that had their specific indel removed. Each dot represents an individual experiment. (B) Bio-layer interferometry (BLI) sensorgrams showing binding of A23 (blue) and A27 (purple) with their indel events removed. (C) Midpoint neutralization titers (IC₅₀) of A27 and A27ΔFWR3 against the viruses indicated. (D) Amino acid sequence alignment with ARMADiLLO mutation probabilities (left) and phylogenetic tree (right) of an A27 lineage reconstructed by identifying the shortest path between gl-CH31 and A27 using NGS repertoire reads from the mouse from which A27 was isolated. (E) Structural representation (top view) of a BG505 SOSIP Env (PDB: 6NNJ) in complex with bNAbs CH31 (PDB: 6NNJ), VRC01 (PDB: 3NGB), 3BNC60 (PDB: 5VBL), and an AlphaFold2-Multimer-modeled structure of A27. (F) Side view of residue D72_A27 in the FWR3 insertion extending toward K207_gp120.