Rapid elicitation of broadly neutralizing antibodies to HIV by immunization in cows

Abstract

No immunogen to date has reliably elicited broadly neutralizing antibodies to HIV in humans or animal models. Advances in the design of immunogens that antigenically mimic the HIV envelope glycoprotein (Env), such as the soluble cleaved trimer BG505 SOSIP, have improved the elicitation of potent isolate-specific antibody responses in rabbits and macaques, but so far failed to induce broadly neutralizing antibodies. One possible reason for this failure is that the relevant antibody repertoires are poorly suited to target the conserved epitope regions on Env, which are somewhat occluded relative to the exposed variable epitopes. Here, to test this hypothesis, we immunized four cows with BG505 SOSIP. The antibody repertoire of cows contains long third heavy chain complementary determining regions (HCDR3) with an ultralong subset that can reach more than 70 amino acids in length. Remarkably, BG505 SOSIP immunization resulted in rapid elicitation of broad and potent serum antibody responses in all four cows. Longitudinal serum analysis for one cow showed the development of neutralization breadth (20%, n = 117 cross-clade isolates) in 42 days and 96% breadth (n = 117) at 381 days. A monoclonal antibody isolated from this cow harboured an ultralong HCDR3 of 60 amino acids and neutralized 72% of cross-clade isolates (n = 117) with a potent median IC₅₀ of 0.028 μg ml^-1. Breadth was elicited with a single trimer immunogen and did not require additional envelope diversity. Immunization of cows may provide an avenue to rapidly generate antibody prophylactics and therapeutics to address disease agents that have evolved to avoid human antibody responses.

Extended Data Fig. 1. Serum collected at different timepoints over the course of immunization for cows 26 and 27 were evaluated for neutralization breath and potency on the 12-virus global indicator panel

Values represent serum ID₅₀ against the indicated virus and colored according to the legend.

Extended Data Fig. 2. Autologous serum timepoints tested for neutralization on BG505 pseudovirus

Neutralization against autologous virus emerged at the same time as breadth (d42) and increased in potency over time. Values represent serum ID₅₀.

Extended Data Fig. 3. Development of neutralization breadth at 35 days post immunization from cows in experiment 1

Serum collected at day 35 from experiment 1 were tested for neutralization breadth and potency on the 12-virus global indicator panel. Values represent serum ID₅₀.

Extended Data Fig. 4. Sorting strategy for isolating antigen-specific IgG+ B cells

(A) Cow PBMCs were sorted for IgG+ cells that bound to biotinylated BG505 SOSIP-AviTag conjugated on PE and APC fluorophores. (B) To isolate epitope-specific antibodies, unliganded BG505 SOSIP (blue) and BG505 SOSIP liganded with NC-Cow1 was used to antigen-sort memory B cells. Epitope-specific B cells are defined as binding unliganded SOSIP and not binding to liganded SOSIP. This sort strategy yielded the broadly neutralizing antibodies NC-Cow7 to NC-Cow10.

Extended Data Fig. 5. Functional screening and sequencing information for isolated antibodies

(A) Amplified heavy chains were paired with universal cow light chain or with NC-Cow1 light chain and tested for expression (anti-Fc ELISA), Ag binding (BG505 SOSIP), autologous neutralization (BG505 pseudovirus), and heterologous neutralization (Q23 pseudovirus). Sequence alignment of recovered heavy chains are listed underneath. As ultralong HCDR3 antibodies have been reported to pair with a single germline light chain (V30), amplified heavy chain genes were first paired with the universal cow light chain and screened for expression, binding to BG505 SOSIP, neutralization of BG505 virus, and neutralization of a clade A heterologous virus, Q23. From this dataset, 3 antibodies (named NC-Cow1, NC-Cow2, and NC-Cow3) were selected that showed autologous neutralization (all from d238) and corresponding native light chains were amplified, with success for only NC-Cow1. These three antibodies were expressed and purified at larger scale for additional characterization by maintaining native pairing for NC-Cow1 and pairing with germline V30 for NC-Cow2 and NC-Cow3. For the d70 timepoint, three heavy chains were selected that showed binding to BG505 SOSIP, but no neutralization, and these antibodies were produced with their native light chain pairs (NC-Cow4 to NC-Cow6). Finally, as NC-Cow1 could neutralize isolate Q23 in the neutralization screen, an additional sort with PBMCs from d381 was performed, but used BG505 SOSIP liganded with and without NC-Cow1 to enrich for epitope-specific antibodies. The enrichment yielded an additional 5 hits by neutralization screen, and 4 out of these 5 antibodies were produced at larger scale with their native light chain pairs (NC-Cow7 to NC-Cow10). Small-scale screening was also performed with all heavy chains paired with NC-Cow1 light chain and no significant increase in neutralization breadth was observed, although there were slight improvements in BG505 SOSIP affinity or autologous potency. (B) Nucleotide alignment of heavy chain sequences of isolated monoclonal antibodies.

Extended Data Fig. 6. NC-Cow1 is not polyreactive to human antigens

(A) NC-Cow1 was tested for antigen reactivity in a HEp2 assay compared to the known polyreactive antibody 4E10, and negative and positive control sera supplied by the manufacturer. (B) NC-Cow1 was also tested for reactivity with a range of typical human autoantigens by ELISA as well as for binding to solubilized membrane (SMP) and cytosolic preparations (SCP) from CHO cells. Values are optical density values (OD405) at a 100 μg/ml dose. Black line indicates cut-off values as indicated by the manufacturer.

Extended Data Fig. 7. Epitope mapping of NC-Cow1

(A) Representative 2D class averages of cow Fabs NC-Cow1 and NC-Cow2 bound to BG505 SOSIP trimers to demonstrate CD4bs site specificity. The Fabs appear at slightly different angles relative to the trimer, perhaps due to some flexibility between the Fab and HCDR3. In the 2D class averages for NC-Cow1, the HCDR3 appears as a faint density bridging between the Fab and CD4bs. (B) Top and side views of 3D reconstruction of NC-Cow1 bound to BG505 SOSIP trimer with previously published Env trimer structure (green, PDB 5CEZ) and cow Fab (orange, PDB 5IJV) docked in. The body of NC-Cow1 Fab is approximately 20 Å away from the CD4 binding site, which is the estimated length of the ultralong HCDR3. (C) NC-Cow 1 was tested for binding by ELISA to BG505 or JRCSF gp120 captured from lysed virions with PGT121 (V3-glycan epitope) and VRC01 (CD4bs epitope) included for comparison. Values are EC₅₀ in mg/ml. NC-Cow1 was also tested on BG505 or JR-CSF pseudoviruses and corresponding alanine mutants with PGT121 (V3-glycan epitope) and 12A12 (CD4bs epitope) included for comparison. Values are IC₅₀ in mg/ml. (D) NC-Cow1 was tested for binding to WT and D368R protein. Antibodies VRC01 (CD4bs) and 14e (V3) were included for comparison.

Extended Data Fig. 8. Effects of pH on binding of NC-Cow1 and CD4bs antibodies to gp120

(A) CD4bs antibodies PGV04, 12A12, 3BNC60, and b12 were tested for binding to gp120 (isolate 92BR020) by ELISA in buffers at different pHs (3.5, 4.0, 4.5, 5.0, 5.5, and 6.0). (B) NC-Cow1 and VRC01 were tested for binding to BG505 gp120 in PBS buffer (pH 7.4) compared to simulated vaginal fluid (SVF), pH 4.5.

Figure 1. Cows develop broadly neutralizing serum antibodies when immunized with the HIV envelope trimer mimic BG505 SOSIP

(a) Schematic of SOSIP immunization experiments in cows. (b) Sera from all four cows were tested for neutralization on a 12-virus “global isolate” panel that serves as an indicator for breadth and potency. Presented values are neutralization ID₅₀ titers. Murine Leukemia Virus (MLV) was included as a negative control.

Figure 2. Rapid development of neutralization breadth and potency in cows

(a) Sera collected longitudinally for cow #26 and cow #27 were tested for neutralization on the 12-virus indicator panel. Neutralization ID₅₀ titers are presented for each virus by color at each timepoint. Percent neutralization breadth is indicated by the black line. (b) Sera at select timepoints were tested on a larger cross-clade 117-virus panel. Presented values are neutralization breadth and median neutralization ID₅₀ titers grouped by virus clade.

Figure 3. Isolation and characterization of a broadly neutralizing cow antibody

(a) HCDR3s are shown for antibodies NC-Cow1–10. The sequence for germline IGHV1-7/VHBUL, IGHD8-2, and IGHJ10 regions are shown at the top. HCDR3 length (L), number of cysteines within HCDR3 (#Cys), and HCDR3 numbering are listed. Cysteines within HCDR3 are highlighted yellow, with those conserved with the germline underlined, and the cysteine and tryptophan defining HCDR3 are highlighted cyan. (b) NC-Cow1 was analyzed for neutralization breadth and potency on a 117-virus panel. (c) NC-Cow1–10 antibodies map to the VRC01-class epitope by competition ELISA. (d) To determine if the HCDR3 of NC-Cow1 is effective at neutralization on its own, four chimeras were tested on the 12-virus global isolates indicator panel. An IC₅₀ value of 50μg/ml was used a cut-off for neutralization.