Antibody-mediated broad sarbecovirus neutralization through ACE2 molecular mimicry

Understanding broadly neutralizing sarbecovirus antibody responses is key to developing countermeasures against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants and future zoonotic sarbecoviruses. We describe the isolation and characterization of a human monoclonal antibody, designated S2K146, that broadly neutralizes viruses belonging to SARS-CoV– and SARS-CoV-2–related sarbecovirus clades, which use angiotensin-converting enzyme 2 (ACE2) as an entry receptor. Structural and functional studies show that most of the virus residues that directly bind S2K146 are also involved in binding to ACE2. This allows the antibody to potently inhibit receptor attachment. S2K146 protects against SARS-CoV-2 Beta variant challenge in hamsters, and viral passaging experiments reveal a high barrier for emergence of escape mutants, making it a good candidate for clinical development. The conserved ACE2-binding residues present a site of vulnerability that might be leveraged for developing vaccines eliciting broad sarbecovirus immunity.

T he zoonotic spillover of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in a global pandemic causing more than 266 million infections and more than 5.2 million fatalities as of December 2021. Continued SARS-CoV-2 evolution leads to the emergence of variants of concern (VOCs) that are characterized by higher transmissibility, immune evasion, and/or disease severity. For pandemic preparedness, we need pan-sarbecovirus countermeasures, such as vaccines and therapeutics that are effective against all SARS-CoV-2 variants and divergent zoonotic sarbecoviruses (1).
To assess the role of somatic mutations for S2K146 binding and neutralization, we generated its inferred unmutated common ancestor (S2K146 UCA). Alignment with the UCA amino acid sequence reveals that S2K146 harbors seven and two somatic hypermutations in the heavy-and light-chain complementarity determining regions (CDR), respectively [V H (variable region of immunoglobulin heavy chain) identity: 94.4%; V L (variable region of immunoglobulin light chain) identity: 98.9%] ( fig. S2D). Except for WIV-1, S2K146 and S2K146 UCA showed no major cross-reactivity differences with a panel of RBDs representative of clade 1 sarbecoviruses, as determined by ELISA ( fig. S2E). Nevertheless, biolayer interferometry revealed that S2K146 bound to prefusion-stabilized SARS-CoV and SARS-CoV-2 S ectodomain trimers with enhanced avidities compared with S2K146 UCA (fig. S2F). Accordingly, S2K146 UCA showed a marked loss in neutralizing activity against both SARS-CoV S and SARS-CoV-2 S VSV pseudotypes (Fig.  1F). Our results suggest that somatic hypermutations associated with S2K146 affinity ma-turation are especially important for enhancing mAb avidity and potency.
To understand the sarbecovirus cross-reactivity of the RBM-specific S2K146 mAb, we determined a cryo-electron microscopy (cryo-EM) structure of the S2K146 Fab fragment in complex with the SARS-CoV-2 S ectodomain trimer at 3.2-Å resolution ( Fig. 2A, fig. S4, and table

S1
). Three-dimensional (3D) classification of the data led to the determination of a structure with three open RBDs, each bound to a S2K146 Fab, as well as a structure with two open RBDs and one closed RBD, with a Fab bound to each of them ( fig. S4). Our cryo-EM data show that the opening of two RBDs is enough to allow three Fabs to bind to an S trimer, as the remaining closed RBD can engage an S2K146 Fab owing to its angle of approach.
To overcome the conformational heterogeneity of the S2K146-bound RBDs relative to the rest of the S trimer, we used focused 3D classification and local refinement of the S2K146 variable domains and RBD to obtain a reconstruction at 3.2-Å resolution enabling unambiguous model building and providing a detailed view of the binding interface (Fig. 2B, fig. S4, and table S1). S2K146 recognizes an epitope in antigenic site I (17), which overlaps with the RBM and is partially masked when the three RBDs adopt a closed state leading to clashes between the mAb and a neighboring RBD (Fig. 2, A  and B, and fig. S1). The S2K146 paratope includes the heavy chain N terminus and CDRH1, CDRH2, and CDRH3, accounting for threequarters of the surface buried upon binding, with light chain CDRL1, CDRL2, and CDRL3 making up the rest of the interface. A total of 1000 Å 2 of the paratope surface is buried at the interface with the RBM through electrostatic interactions and shape complementarity.
The S2K146 footprint on the SARS-CoV-2 RBD highly resembles that of the ACE2 receptor, with 18 of 24 epitope residues shared with the ACE2 binding site, including key ACE2 contact positions L455, F486, Q493, Q498, and N501 (Fig. 2, C and D). Moreover, electrostatic interactions formed between S2K146 and the SARS-CoV-2 RBD recapitulate some of the contacts involved in ACE2 binding, such as with residues K417, Y449, Y489, Q493, and G502 (Fig. 2, E and F). Although some S2K146 contact residues are mutated in several variants, such as K417 (Beta and Gamma), L452 (Delta, Epsilon, and Kappa), E484 (Beta, Gamma, and Kappa), and N501 (Alpha, Beta, and Gamma), the retention of neutralization of these variants suggests that the binding interface is resilient to these residue substitutions (Fig. 1E and fig. S2B). The cross-reactivity with and broad neutralization of SARS-CoV by S2K146 may be partially explained by the strict conservation or conservative substitution of nine and four epitope residues relative to SARS-CoV-2, respectively (Fig. 2, G and H, and fig. S5A), consistent with the ability of both RBDs to bind human ACE2.
S2K146 therefore overcomes the mutational plasticity of the RBM, which is implicated in immune evasion, by targeting residues required for binding to the ACE2 receptor. This is supported by S2K146 recognition of the reconstructed RBD ancestor of SARS-CoV and SARS-CoV-2 ( fig. S5, A and B), where human ACE2 binding first arose during sarbecovirus evolution (34), in line with the hypothesis that human ACE2 binding participates in conferring S2K146 susceptibility. As S2K146 does not compete with broadly neutralizing sarbecovirus mAbs targeting other antigenic sites, such as S309 (21) and S2X259 (19) (figs. S1 and S6), they could be combined in a cocktail to enhance breadth further and set an even higher barrier for emergence of escape mutants.
To prospectively evaluate the impact of antigenic drift on S2K146 neutralization, we mapped RBD mutations that affect mAb binding using deep mutational scanning (DMS) of a yeast-displayed RBD mutant library covering all possible single residue substitutions in the Wuhan-Hu-1 RBD background (30). S2K146 binding was reduced by only a restricted number of amino acid substitutions compared to S2E12, which binds an overlapping but distinct epitope (Fig. 3, A to D, and  fig. S7, A and B). All these mutations correspond to RBD residues buried upon ACE2 recognition (F456, A475, E484, F486, N487, and Y489) (Fig. 3B). Only one of these residue substitutions (Y489H) is accessible through a single-nucleotide change and could escape S2K146 recognition with a penalty on ACE2 binding affinity smaller than an order of magnitude, as determined by DMS data (30). None of the individual mutations present in the recently identified SARS-CoV-2 Omicron VOC affected S2K146 binding (Fig. 3B), although the effect of the full constellation of mutations remains to be evaluated. Conversely, DMS profiling of the S2K146 UCA revealed a greater number of binding-escape mutations, including some residue substitutions present in Omicron (e.g., Q493R or Q498R) (Fig. 3B). Therefore, the hotspot targeting of S2K146 on residues that are constrained in SARS-CoV-2 evolution appears to be a direct consequence of mAb affinity maturation.
To explore whether our escape map was consistent with in vitro viral evolution under mAb pressure, a replication competent VSV-SARS-CoV-2 S Wuhan-Hu-1/D614G chimera (35) was passaged in cell culture in the presence of the S2K146 mAb. Consistent with the DMS data, Y489H was the sole mutation resulting from a single nucleotide substitution that was detected in all the 36 neutralizationresistant plaques sampled (Fig. 3E, fig. S8A, and table S2). SARS-CoV-2 residue Y489 forms multiple interactions with S2K146 CDRH3 and accounts for~10% of the total epitope buried surface area (Fig. 3E), in line with the major impact of the Y489H substitution on mAb neutralization. Of all the mutations at position 489 identified by DMS to reduce S2K146 binding (Fig. 3B), the Y-to-H substitution had the lowest impact on ACE2 binding, which might explain why it was the sole neutralization escape mutant selected upon passaging.
To evaluate the fitness of the Y489H mutant, we carried out a competition assay in which replicating VSV chimeras harboring the Wuhan-Hu-1/D614G S with or without the Y489H substitution were mixed at equal titers and passaged together without mAb. Because of the fitness cost associated with the mutation, which dampens~4.5-fold the 1:1 ACE2-binding affinity to the SARS-CoV-2 RBD ( fig. S8B), the Y489H S chimera was outcompeted by the Wuhan-Hu-1/D614G S chimera after only four rounds of passaging (Fig.  3F). Accordingly, only 29 out 2.9 million genomes were found to harbor the S Y489H mutation, underscoring the rarity of and the fitness cost imposed by this residue substitution (Fig. 3G). Collectively, these data illustrate the high barrier for emergence of escape mutants imposed by the S2K146 mAb, making it a good candidate for clinical development.
S2K146 targets antigenic site Ia, which overlaps with the RBM, indicating that mAb binding would compete with ACE2 attachment to the RBD via steric hindrance (Fig. 4A). Indeed, S2K146 inhibited binding of the SARS-CoV-2 and SARS-CoV RBDs to human ACE2 in a concentration-dependent manner, as measured by competition ELISA (Fig. 4B). As S2K146 conformationally selects for open RBDs, we assessed whether the mAb could promote shedding of the S 1 subunit from cell surfaceexpressed full-length SARS-CoV-2 S, similar to some other RBD-specific mAbs (17, 19, 24, 33). as efficiently as the RBM-targeting S2E12 mAb, whereas the control mAb S2M11, which locks S in the prefusion closed state, did not (19) (Fig. 4C). Furthermore, S2K146 Fab triggered fusogenic rearrangement of a wild-type-like S ectodomain trimer, as previously described for several SARS-CoV and SARS-CoV-2 neutralizing mAbs (20,(36)(37)(38) (Fig. 4D). Thus S2K146-mediated sarbecovirus neutralization relies on competitively blocking viral attach-ment to the ACE2 receptor and putative inactivation of S trimers at the surface of virions before encountering host cells.
The efficient S2K146-induced S 1 shedding could explain the lack of FcgRIIa and FcgRIIIa activation, which we used as a proxy for Abdependent cellular phagocytosis and Abdependent cellular cytotoxicity, respectively ( fig. S9, A and B). However, S2K146 also did not activate FcgRIIa and triggered FcgRIIIa only weakly when target cells expressed an uncleavable prefusion-stabilized SARS-CoV-2 S protein ( fig. S9, C and D). The greater efficiency of S2E12 for activating FcgRIIIa, relative to S2K146, might be explained by the different angles of approach at which these two mAbs bind to the RBD ( fig. S9, E and F).
Next, we evaluated the therapeutic activity of S2K146 against challenge with the SARS-CoV-2 Beta VOC in a Syrian hamster model of infection (39,40). S2K146 was administered in doses of 1, 5, and 10 mg/kg of body weight via intraperitoneal injection 24 hours after intranasal challenge with SARS-CoV-2, and the lungs of the animals were collected 3 days later for the quantification of viral RNA and replicating virus. In parallel, six animals were administered 1 mg/kg of the ultrapotent S2E12 mAb for benchmarking (33). Viral RNA loads in the lungs were reduced by~1, 4, and 3 orders of magnitude after receiving 1, 5, and 10 mg/kg of S2K146, respectively (Fig. 4E). Viral replication in the lungs was completely abrogated for the 5 and 10 mg/kg groups and reduced by greater than 2.5 orders of magnitude for the 1 mg/kg group (Fig. 4F). Overall serum mAb concentrations measured at day 4 after infection inversely correlated with viral RNA loads and infectious virus in the lungs (fig. S10, A and B). S2K146 therefore effectively protects against SARS-CoV-2 challenge in vivo in a stringent therapeutic setting.
pan-sarbecovirus vaccines to protect from future zoonotic transmission events. The broadly neutralizing sarbecovirus mAb S309 was isolated from a survivor of a 2003 SARS-CoV infection, and its derivative (sotrovimab) has received emergency use authorizations in several countries around the world for the early treatment of mild-tomoderate COVID-19 in adults and some pediatric patients who test positive for SARS-CoV-2 by direct viral testing and who are at high risk for progression to severe COVID-19, including hospitalization or death (19,27,29,53). S309 has proven resilient to the emergence of SARS-CoV-2 variants in preclinical studies, possibly owing to targeting of a conserved RBD epitope with very limited mutational tolerance (20,53). The mechanism of S2K146mediated ACE2 molecular mimicry also provides a high barrier for emergence of escape mutants in spite of the known mutational plasticity of the SARS-CoV-2 RBM (30). Therefore, the discovery of the S2K146 mAb might be a milestone for future treatment of COVID-19 patients and for pandemic preparedness against divergent sarbecoviruses.