Cross-neutralizing antibodies bind a SARS-CoV-2 cryptic site and resist circulating variants

The study addresses how to develop antibodies and vaccine targets that maintain neutralization across SARS-CoV-2 variants and related sarbecoviruses. Despite effective vaccines and EUA antibody therapies, emerging variants reduce neutralization by many potent RBM-directed antibodies and can diminish vaccine efficacy. Prior cross-neutralizing antibodies (e.g., CR3022, S309) target conserved, non-RBM epitopes but can be impacted by certain variant mutations. The authors aim to elicit and characterize cross-neutralizing antibodies that recognize conserved regions of the RBD, withstand prevalent variant mutations, and elucidate their mechanisms to guide broad-spectrum antibody therapeutics and pan-sarbecovirus vaccine design.

Most potent SARS-CoV-2 neutralizing antibodies target the RBM to block ACE2, but many are strain-specific and vulnerable to variant escape. Two classes of cross-neutralizing antibodies have been described: CR3022-like antibodies bind a cryptic epitope distal to the RBM and may neutralize via spike disruption; S309-like antibodies target a conserved glycan-containing epitope distinct from RBM. Some cross-reactive antibodies lose potency against variants (e.g., S309 affected by B.1.1.7). Broadly neutralizing antibodies can map conserved antigenic sites to inform pan-sarbecovirus vaccine design. The authors build on this foundation by identifying antibodies that bind a different conserved, cryptic RBD site with resistance to circulating variant mutations.

- Immunization and antibody generation: Mice were immunized either with SARS-CoV-2 spike (S) trimer alone or with a combination of SARS-CoV-2 S, SARS-CoV S, and MERS-CoV RBD. Hybridomas were generated and screened for cross-reactivity to SARS-CoV-2 and SARS-CoV S proteins, with selection for RBD-binding clones. - Neutralization assays: Lead monoclonal antibodies (mAbs) were tested for neutralization using lentiviral pseudotyping (LVpp) bearing SARS-CoV-2 or SARS-CoV spikes, VSV-based SARS-CoV-2 pseudovirus, and authentic SARS-CoV-2 live virus focus reduction neutralization tests (FRNT) in BSL-3. Variant neutralization was assessed using LVpp for B.1.1.7, B.1.351, and P.1, and authentic B.1.351. - Binding kinetics and competition: Surface plasmon resonance (SPR) measured affinities (KD) to recombinant S-2P trimers and RBDs (SARS-CoV-2 and SARS-CoV). SPR-based blocking assays determined ACE2 competition and epitope competition among mAbs. - Structural studies: Fabs of 7D6, 6D6, and 16D8 were complexed with SARS-CoV-2 RBD and crystallized. X-ray crystallography determined structures at high resolution (7D6:RBD 1.40 Å; 6D6:RBD 1.92 Å). Epitope mapping identified contact residues and conservation across variants and sarbecoviruses. Structural docking placed Fab-RBD complexes onto spike trimers (open and closed states) to assess steric clashes with neighboring NTDs. - Biochemical and biophysical analyses of spike disruption: Size-exclusion HPLC and SDS-PAGE of mixtures of S-2P with 7D6 or 6D6 Fabs assessed depolymerization. Cryo-EM imaging of S-2P alone and in complex with Fabs visualized antibody-mediated trimer disruption. - Spike shedding assay: Full-length wild-type spike expressed on 293T cells was incubated with IgG and Fab forms of 7D6 and 6D6; flow cytometry quantified S1 shedding over time (5, 60, 120 min). - Additional experimental details: Ethics approvals; cell lines (H1299-ACE2hR, BHK21-hACE2, 293T, Vero E6, CHO, sf9, High Five); cloning and expression of S-2P and RBD constructs in insect cells; ELISAs for binding; determination of RBD mutant binding (e.g., N501Y, K417T/E484K/N501Y, L452R); data processing for crystallography and cryo-EM; conservation analyses using GISAID and other databases.

- Discovery of cross-neutralizing antibodies: From two immunization strategies, seven RBD-targeting cross-neutralizing mAbs were identified; 7D6 and 6D6 (and 16D8) were selected for detailed study. - Neutralization potency: • LV pseudovirus (SARS-CoV-2): 7D6 IC50 = 2.56 μg/mL; 6D6 = 8.91 μg/mL; 16D8 = 3.52 μg/mL; CR3022 > 284 μg/mL. • LV pseudovirus (SARS-CoV): 7D6 = 10.11 μg/mL; 6D6 = 1.67 μg/mL; 16D8 = 1.21 μg/mL; CR3022 = 7.11 μg/mL. • VSV pseudovirus (SARS-CoV-2): 7D6 = 0.04 μg/mL; 6D6 = 0.21 μg/mL; 16D8 = 0.26 μg/mL. • Authentic SARS-CoV-2 (WT): 7D6 = 2.23 μg/mL; 6D6 = 1.77 μg/mL; 16D8 = 5.30 μg/mL. • Authentic B.1.351: 7D6 = 13.73 μg/mL; 6D6 = 3.95 μg/mL. - Binding affinities (SPR, KD, nM): • SARS-CoV-2 RBD: 7D6 = 0.003; 6D6 = 0.01; 16D8 = 0.004. • SARS-CoV RBD: 7D6 = 0.03; 6D6 = 0.37; 16D8 = 0.08. • SARS-CoV-2 S-2P: 7D6 = 5.89; 6D6 = 0.03; 16D8 = 0.04. • SARS-CoV S-2P: 7D6 = 5.04; 6D6 = 2.14; 16D8 = 13.90. - Epitope and structure: 7D6 and 6D6 bind a nearly identical, cryptic RBD site distal to the RBM, involving residues primarily in loops η1, β1, β5, and β7–β8. The 7D6 interface buries ~796 Ų (HCDRs 73.3%); 6D6 buries ~1,019 Ų (HCDRs 54.7%). Key RBD residues include 346–355 and 466–471 (7D6) and 351–357 and 457–471 (6D6), with 18 shared residues. The shared 7D6/6D6 site shows high conservation (89% across sarbecoviruses) and is 100% conserved among tested SARS-CoV-2 variant contact residues. Crystal structures determined at 1.40 Å (7D6:RBD) and 1.92 Å (6D6:RBD). - Mechanism of neutralization: Structural modeling shows bound Fabs clash with adjacent NTDs in both RBD-up and RBD-down states (occluded volumes ~3,700 ų when RBD up; ~11,800–13,100 ų when RBD down), indicating a cryptic epitope. HPLC and cryo-EM demonstrate antibody-induced spike trimer disruption. Flow cytometry shows up to ~63% S1 shedding after 120 min with both IgG and Fab forms, supporting neutralization via spike destabilization/S1 shedding. - Variant resistance: The 7D6/6D6 epitope excludes common variant mutations (e.g., N501Y, K417N/T, E484K/Q, L452R/Q, T478K, F490S). 7D6 and 6D6 maintain binding to RBD mutants and show nearly unchanged pseudovirus neutralization of B.1.1.7, B.1.351, and P.1; authentic B.1.351 shows some reduction but retained activity. REGN10933 control showed decreased activity against B.1.351 and P.1. - Antibody classification: Based on epitope location and competition, 7D6/6D6 group with Class 1 (non-RBM, S309-like region but non-overlapping), distinct from RBM blockers (Class 2) and CR3022-like cryptic binders (Class 3). 7D6/6D6 likely neutralize by destabilizing the spike, differing functionally from S309. - Immunization strategy: Combined immunization with multiple sarbecovirus spikes yielded more cross-neutralizing RBD-directed mAbs than immunization with SARS-CoV-2 spike alone.

The study identifies two antibodies that address the need for variant-resilient neutralization by targeting a conserved, cryptic RBD site distinct from the RBM. Structural and biochemical data show that 7D6/6D6 binding induces steric clashes with neighboring NTDs on the spike trimer, destabilizing the prefusion spike and promoting S1 shedding, thereby preventing productive ACE2 engagement. Their epitopes avoid the prevalent variant mutations, explaining preserved binding and neutralization across major variants tested. The high conservation of the 7D6/6D6 site across sarbecoviruses highlights its value as a target for pan-sarbecovirus vaccine design. Furthermore, the combined immunization approach enriched for cross-neutralizing responses, suggesting vaccination regimens incorporating multiple sarbecovirus antigens could focus immunity on conserved epitopes. These antibodies could complement existing EUA therapies by adding a non-overlapping, mutation-resistant component to antibody cocktails.

Two cross-neutralizing mouse monoclonal antibodies, 7D6 and 6D6, were elicited that bind a conserved, cryptic RBD site with high affinity and neutralize SARS-CoV-2 and SARS-CoV. High-resolution structures define an epitope distal to the RBM that clashes with adjacent NTDs, with functional data confirming spike destabilization and S1 shedding as a neutralization mechanism. The antibodies retain activity against multiple circulating variants because their epitope excludes common RBD mutations. These findings expand the repertoire of mutation-resistant epitopes for therapeutic antibody development and provide a blueprint for rational pan-sarbecovirus vaccine design. Future work should include humanization and affinity maturation of these antibodies, in vivo efficacy studies, and immunogen designs that better expose the cryptic epitope to drive broadly neutralizing responses.

- Antibodies are murine and require humanization/chimeric grafting and potential affinity maturation before clinical use. - In vivo protective efficacy has not yet been demonstrated. - The cryptic epitope may be transiently accessible; immunogen designs to reliably expose this site need development and validation. - While major variants were tested (B.1.1.7, B.1.351, P.1), broader testing across additional and future variants is necessary. - Mechanistic conclusions are supported by structural modeling, HPLC, cryo-EM 2D analyses, and cell-based shedding assays; high-resolution cryo-EM of intact immune complexes on spike could further strengthen mechanistic insights.