A Novel Monoclonal Antibody Targeting a Large Surface of the Receptor Binding Motif Shows Pan-neutralizing SARS-CoV-2 Activity Including BQ.1.1 Variant

1 A Novel Monoclonal Antibody Targeting a Large Surface of the Receptor Binding Motif Shows Pan-neutralizing SARS-CoV-2 Activity Including BQ.1.1 Variant Leire de Campos-Mata1,#, Benjamin Trinité2,#, Andrea Modrego3,#, Sonia Tejedor Vaquero1, Edwards Pradenas2, Natalia Rodrigo Melero4, Diego Carlero3, Silvia Marfil2, Anna Pons-Grífols2, María Teresa Bueno-Carrasco3, César Santiago3, Ferran TarrésFreixas2, Victor Urrea2, Nuria Izquierdo2,5, Eva Riveira-Muñoz2, Ester Ballana2,5,6, Mónica Pérez7, Júlia Vergara-Alert7,8, Joaquim Segalés8,9, Carlo Carolis4,§*, Rocío Arranz3,§*, Julià Blanco 2,5,6,10,§*, Giuliana Magri1,11,§* 1Translational Clinical Research Program, Institut Hospital del Mar d’Investigacions Mèdiques (IMIM), Barcelona, Spain 2IrsiCaixa AIDS Research Institute, Hospital Germans Trias I Pujol, Campus Can Ruti, Badalona, Spain 3Centro Nacional de Biotecnología (CNB-CSIC), Madrid, Spain 4Centre for Genomic Regulation (CRG), The Barcelona Institute of Science and Technology, Barcelona, Spain 5CIBERINFEC, ISCIII, Madrid, Spain 6Germans Trias i Pujol Research Institute (IGTP), Can Ruti Campus, Badalona, Spain 7Unitat mixta d'Investigació IRTA-UAB en Sanitat Animal. Centre de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), Bellaterra, Spain 8IRTA. Programa de Sanitat Animal. Centre de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), Bellaterra, Spain 9Departament de Sanitat i Anatomia Animals, Facultat de Veterinària, Universitat Autònoma de Barcelona (UAB), Bellaterra, Spain 10Infectious Diseases and Immunity, Faculty of Medicine, University of Vic-Central University of Catalonia (UVic-UCC), Barcelona, Spain 11Present address: Immunology Unit, Department of Biomedical Sciences, Faculty of Medicine and Health Sciences, University of Barcelona, Barcelona, Spain #These authors contributed equally §These authors contributed equally *Correspondence: Carlo Carolis, [email protected]; Julià Blanco, [email protected]; Rocío Arranz, [email protected] and Giuliana Magri, [email protected] .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 2 Summary In the present study we report the functional and structural characterization of 17T2, a new highly potent pan-neutralizing SARS-CoV-2 human monoclonal antibody (mAb) isolated from a convalescent COVID-19 individual infected during the first wave of the COVID-19 pandemic. 17T2 is a class 1 VH1-58/κ3-20 antibody, derived from a receptor binding domain (RBD)-specific IgA memory B cell and developed as a human recombinant IgG1. Functional characterization revealed that 17T2 mAb has a high and exceptionally broad neutralizing activity against all SARS-CoV-2 spike variants tested, including BQ.1.1. Moreover, 17T2 mAb has in vivo prophylactic activity against Omicron BA.1.1 infection in K18-hACE2 transgenic mice. 3D reconstruction from cryogenic-electron microscopy (cryo-EM) showed that 17T2 binds the Omicron BA.1 spike protein with the RBD domains in “up” position and recognizes an epitope overlapping with the receptor binding motif, as it is the case for other structurally similar neutralizing mAbs, including S2E12. Yet, unlike S2E12, 17T2 retained its high neutralizing activity against all Omicron sublineages tested, probably due to a larger contact area with the RBD, which could confer a higher resilience to spike mutations. These results highlight the impact of small structural antibody changes on neutralizing performance and identify 17T2 mAb as a potential candidate for future therapeutic and prophylactic interventions. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 3 Introduction SARS-CoV-2, the etiological agent of COVID-19, has provoked one of the worst pandemics in human history, causing more than 6.7 million deaths registered worldwide (https://covid19.who.int/). The high level of virus circulation amongst humans and other species has led to the emergence of several variants of concern (VOCs) with progressively increased transmissibility and immune evasion capacity 1–3. In December 2021, the Omicron VOC became the globally dominant circulating strain, after replacing previous variants. The initial Omicron wave was caused by the BA.1 variant, followed by several Omicron sublineages (BA.2, BA.5 and BQ.1.1 among others). Compared with the ancestral variant identified in Wuhan (WH1), the spike protein of the BA.1 one contains at least 30 amino acid (aa) substitutions including 6 aa deletions and 3 aa insertions which are largely confined to the receptor binding domain (RBD) and the N-terminal domain (NTD), the two major antigenic sites targeted by neutralizing antibody responses 4. BA.2 shares 21 mutations with BA.1 and harbors an additional 8 specific mutations. BA.5 appears to have evolved from BA.2 with two additional mutations and one reversion in the RBD and one deletion in the NTD 1. Since November 2022, BQ.1 and BQ.1.1 variants, which harbor 3 new mutations in key antigenic sites of the RBD, have overtaken BA.5 as the dominant variant across the USA and Europe 5. The presence of these mutations has not only increased the transmissibility of Omicron variants but has also caused a strong resistance to vaccine-induced antibody responses, leading to unprecedented levels of vaccine breakthrough infections worldwide1,6,7. Together with vaccines and antiviral drugs, neutralizing monoclonal antibodies (mAbs) targeting the spike protein of SARS-CoV-2 have been extensively used to treat individuals at the highest risk of severe COVID-19 8,9. During previous COVID-19 waves, therapeutic administration of mAbs was reported to be highly effective in preventing COVID-19-related hospitalization and death 10,11. However, most of the mAbs currently in use were developed against ancestral SARS-CoV-2 and they all lost or significantly reduced their activity against highly mutated Omicron sublineages 1,2,12. Therefore, there is an urgent need to develop new pan-SARS-CoV-2 neutralizing antibodies that are effective against current and future VOCS to provide additional therapeutic options to patients. Here we report the functional and structural characterization of a new human monoclonal antibody developed from a patient infected with the ancestral SARS-CoV-2 that shows pan-neutralizing activity against both pre-Omicron and Omicron SARS-CoV-2 variants, including BQ.1.1. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 4 Results Generation of human recombinant SARS-CoV-2-specific monoclonal antibodies To generate human mAbs capable of neutralizing SARS-CoV-2, we sorted 380 circulating RBD+ B cells from a convalescent COVID-19 individual who was infected during the first wave of the pandemic in Spain 13, using a biotinylated RBD protein from the ancestral SARS-CoV-2 strain as bait (Supplementary Figure 1A). RBD-specific B cells were further characterized according to the expression of CD27, CD21, CD11c, HLA-DR, IgM, IgD, IgA and the Ig light chain λ (Supplementary Figure 1B). mRNA from sorted cells were then reverse-transcribed and the immunoglobulin heavy chain (IGHV) and light chain variable (IGLV) regions were amplified by PCR following an established protocol 14. After Sanger sequencing, 5 RBD-specific IGHV and IGLV paired regions were cloned into expression vectors and generated as recombinant human monoclonal IgG1. Three of these mAbs were generated starting from RBD-specific CD21+CD27+ canonical IgG memory B cells whereas two of them, 17T2 and 54T1, were generated starting from RBD-specific IgA canonical memory B cells (Supplementary Table 1). As expected, all 5 mAbs showed relatively low levels of somatic mutations, which is consistent with the low level of hypermutation reported in RBD-specific antibodies following infection with the ancestral SARS-CoV-2 15 (Supplementary Table 1). These mAbs were first screened to confirm their reactivity profile by enzyme-linked immunosorbent Assay (ELISA) using recombinant RBD from the WH1 and the subsequent Beta (B.1.351), Gamma (P.1), Delta (B.1.617.2), Omicron BA.1 and Omicron BA.2 variants as immobilized proteins. As expected, all mAbs bound to the RBD from the ancestral variant, yet only 17T2 was able to efficiently recognize all variants tested, including RBD from highly mutated Omicron variants (Supplementary Figure 2). The binding affinity of 17T2 for the RBD from different SARS-CoV-2 variants was then assessed by surface plasmon resonance (SPR). This analysis confirmed the high-affinity binding with equilibrium dissociation constants (KD) in the subnanomolar range for all variants tested (Supplementary Table 2). Interestingly, 17T2 belongs to the IGHV1-58/κ3-20 clonotype, as many other potent neutralizing mAbs isolated from SARS-CoV-2-infected and/or vaccinated individuals 16–21 (Supplementary Table 1). 17T2 mAb shows high neutralization activity against all SARS-CoV-2 variants including Omicron sublineages To evaluate the functional activity of the selected antibodies, we tested their neutralization capacity using HIV reporter pseudoviruses expressing different SARS-CoV-2 spike proteins from variants ranging from WH1 to Omicron BA.1 (Supplementary Table 3) 22. Consistent with the reactivity data, .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 5 all antibodies showed neutralization of WH1 and D614G pseudoviruses. However, only 17T2 mAb maintained high neutralizing capacity against all variants tested, while the other mAbs markedly lost potency against Beta, Gamma, Mu or Omicron BA.1 variants (Figures 1A and 1B). 131T2 showed the lowest potency against pre-Omicron variants and no activity against Omicron BA.1 (Figure 1A). To further characterize 17T2 mAb we analyzed neutralization of pseudoviruses exposing the spike of newer Omicron subvariants (Supplementary Table 3) and SARS-CoV-1. Moreover, for comparative purposes, we assayed in parallel two well-characterized broad-spectrum neutralizing RBD-targeting mAbs: S2E12 and S309 21,23. S2E12 is a VH1-58/κ3-20-encoded class 1 antibody, like 17T2, whose antibody-binding epitope overlaps with the receptor binding motif (RBM) in the RBD 20, whereas S309 is a class 3 antibody isolated from a patient recovering from SARS, which binds to a conserved epitope outside the RBM 24. 17T2 mAb neutralized all SARS-CoV-2 variants tested (pre-Omicron variants WH1, D614G, Alpha, Beta, Gamma, Delta, and Omicron subvariants BA.1, BA.2, BA.4/5 and BQ.1.1) with IC50 values ranging from 38 to 2 ng/ml for WH1 and BA4/5, respectively. In contrast, no activity against SARS-CoV-1 was observed (Figures 1C and 1D). The wide SARS-CoV-2 neutralization spectrum of 17T2 was remarkably different from the other two broadly neutralizing mAbs. S2E12 showed higher potency against all pre-Omicron variants but had significant lower neutralizing activity against BA.1, BA.2 and no activity against BA.4/5 and BQ.1.1 (Figures 1C and 1D), while the panneutralizing antibody S309 maintained some activity against Omicron subvariants, but with a significantly lower potency compared to 17T2 (Figures 1C-E). 17T2 mAb has prophylactic activity against Omicron BA.1.1 in vivo in K18-hACE2 transgenic mice Next, we tested 17T2 in vivo prophylactic efficacy in K18-hACE2 transgenic mice. 17T2 mAb or an isotype control antibody (IgGb12) were administered intraperitoneally (10 mg/kg) 24 hours before challenge with 103 TCID50 of a BA1.1 SARS-CoV-2 isolate (Figure 2A). As previously reported 25,26, Omicron infection of K18-hACE2 transgenic mice resulted in a mild disease without significant changes in weight in infected animals as compared to uninfected ones (Figure 2B). However, isotype control-treated animals showed high viral loads in oropharyngeal swabs, lungs and nasal turbinates, both 3and 7-days post infection (dpi) (Figure 2C). In comparison, mice treated with 17T2 mAb showed significantly lower viral loads in lungs 3 dpi and in all tissues assayed (oropharyngeal swabs, lungs, and nasal turbinate) at 7 dpi (Figure 2C). The protective effect of 17T2 was confirmed in lung tissue by analyzing histopathological lesions and the SARS-CoV-2 nucleocapsid focal expression at 3 and 7 dpi by histology and immunohistochemistry, respectively (Figures 2D and 2E). .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 6 17T2 binds the Omicron BA.1 spike protein with the RBD domains in the up position and recognizes a large surface overlapping with the receptor binding motif We carried out cryo-EM analysis to understand the potency and breadth of 17T2-mediated neutralization of SARS-CoV-2 variants. For this reason, we solved the structure of the complex between the highly mutated Omicron BA.1 trimeric spike and the 17T2 Fab fragment, reaching a resolution of 3.46 Å (Figures 3A and 3B). Our analysis showed that 17T2 Fab binds to RBD in the “up” conformation with all particles containing 3 17T2 Fabs, each one bound to adjacent RBDs within a single spike trimer (Figures 3A and 3B). Due to the higher conformational dynamics in the 17T2 variable domains and the RBD regions, resolution in the contact area was lower than in the rest of the trimer. Local refinement was performed in this region, significantly improving local resolution to 4.41 Å (Figures 3C and Supplementary Figure 3). After refinement, we determined that 17T2 Fab binds to the left shoulder-neck region of the RBD that is only accessible in “up” conformation (Figures 3BD). The interaction site overlaps with the RBD in a similar manner to other class 1 VH1-58/κ3-20-- derived neutralizing mAbs 16–21. Fab 17T2/RBD interactions involve both the heavy chain (HC) and light chain (LC) of the antibody, covering 563Å2 for the HC and 295 Å2 for the LC of the total interaction surface (Figure 3D). 17T2 Fab uses complementary determinant regions (CDR) 1 to 3 of the HC and CDR1 and CDR3 of the LC to recognize residues 420 to 421; 455; 473 to 478; 480; 484 to 487, 489 and 493 of the SARS-CoV-2 RBD (Figures 3D and 3E). The contact area mostly involved Van der Waals interactions with a minor contribution from hydrogen bridges (Figure 3E). In addition, we identified a salt bridge formed between the D420 residue of the RBD and the R103 residue of H3 that stabilizes the binding of 17T2 to the RBD (Figure 3E). As previously described for structurally similar neutralizing antibodies 16,27, 17T2 is glycosylated at the N102 residue of H3 located near the left shoulder of the RBD (Figure 3C). Using PNGase-mediated de-glycosylation (Supplementary Figure 4A) we confirmed that this glycosylation had no effect on the affinity nor the neutralizing activity of 17T2 mAb (Supplementary Figure 4B-D). The 17T2 Fab binds to a large and mostly conserved area of RBD, with mutated residues harbored by SARS-CoV-2 variants located at the edge of the contact surface (Figure 3F and Supplementary Table 4). Since 17T2 and S2E12 mAbs share high sequence identity but differ in their neutralizing activity against highly mutated Omicron subvariants, we compared the structures of their respective Fabs/spike complexes (Supplementary Figures 5A and 5B). This comparison revealed that S2E12 16,20,27 and 17T2 Fabs bind parallel to the longest axis of the hACE2 binding site. Nevertheless, the area of interaction between 17T2 and RBD is broader than the area between S2E12 and RBD, the latter being included in the 17T2 interaction area (Supplementary Figure 5C). Moreover, although the .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 7 structure of the two antibodies is highly similar, the amino acid side chains of 17T2 are located closer to the surface of the RBD (Supplementary Figure 5B), allowing for higher number of contacts with conserved residues. This fact probably contributes to the ability of this antibody to neutralize Omicron subvariants exposing S477N and F486V mutations, which strongly impact S2E12 binding. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 8 Discussion Here we describe the functional and structural characterization of 17T2, a new human mAb with broad neutralizing activity against all SARS-CoV-2 variants, including Omicron sublineages BA.1, BA.2, BA.4, BA.5 and BQ.1.1. Importantly, prophylactic administration of 17T2 mAb resulted in significant reduction of viral replication and microscopic lung lesions in a murine model of SARS-CoV-2 Omicron infection. 17T2 belongs to the class 1 VH1-58/κ3-20-derived antibodies that include several mAbs (i.e. S2E12, Cv2.1169, A23.58.1, AZD8895) with high neutralizing activity against pre-Omicron variants 16–18,20,21. Yet, compared to these structurally similar mAbs, 17T2 retains its potency against all Omicron sublineages tested, including the highly immune evasive variants BA.5 and BQ.1.1 1,28, with IC50 values ranging from 2 to 25.5ng/ml. The structural analysis of 17T2 Fab in complex with Omicron BA.1 spike trimer suggests complementary mechanisms to explain its broad neutralizing activity. On the one hand, the high antibody affinity allows for a complete blockade of all RBDs of the spike trimer stabilized in the “up” conformation (stoichiometry 3 Fab:1 spike trimer). On the other hand, when compared to S2E12, 17T2 shows a larger area of interaction with the RBM, which could confer higher tolerability to RBD mutations. Moreover, the presence of a salt bridge between the R103 in the CDR H3 and the D420 in a conserved region of the RBD participates in the stabilization of the complex, contributing to the extraordinarily high affinity of 17T2 to the RBD from multiple SARS-CoV-2 variants. Interestingly, D420 has been recently identified by mutagenesis as a potential site for escaping neutralization by some class 1 antibodies 29,30. The stability provided by D420 could explain why 17T2 mAb resists the F486V spike mutation present in the BA.4 and BA.5 variants which otherwise escape all other IGHV1-58-derived antibodies described thus far 1,17,31. IgG Fab glycosylation appears to be a key parameter in immunity with possible consequences on antigen binding and antibody activity 32. Our structural analysis revealed that 17T2 is glycosylated at residue N102 in the CDR H3 region, in close proximity to the area of contact with the RBD. However, PNGase mediated de-glycosylation of 17T2 mAb had no effect on the binding to RBD or on the neutralizing activity, excluding a potential role of Fab glycosylation in its functional activity. Yet, we cannot rule out its possible implication in the stability or immune modulatory properties of the antibody32. Our data indicates that 17T2 shows unique structural properties conferred by single point mutations in CDRs from both heavy and light chains (as compared to S2E12), enabling its potent and broad neutralizing activity. However, both the RBD exposure in the spike trimer and the presence of RBD .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 9 mutations in the outer 17T2 mAb binding area seem to induce minor variations in the neutralization potency against the tested variants (IC50 range 3-38 ng/mL). A four-fold change in IC50 was observed when comparing WH1 and D614G pseudoviruses, which share identical RBD sequences but differ in the RBD “up” and “down” conformational equilibrium. Similarly, mutations such as S477N, T478K, E484A (present in most Omicron subvariants) and F486V (present in BA.4/5 and BQ.1.1) are well tolerated, suggesting some level of plasticity in the mode of antibody binding. BQ.1.1 has been shown to be resistant to neutralization by all approved therapeutic mAbs, including bebtelovimab 1,12, and became prevalent in the population worldwide, probably due to its advantage in evading humoral responses 33. Other new Omicron sublineages, such as XBB.1.5, a BA.2-derived recombinant, have been recently identified in the United States 34. This variant differs from BA.2 at ten residues of the RBD; however, none of the mutations affect the central 17T2 binding site, but rather they are located in the outer contact area and involve residues mutated in 17T2 sensitive variants (new mutation F486P and reversion R493Q)5. All things considered, novel broadly-active neutralizing mAbs with proven in vivo neutralizing activity, such as 17T2 mAb, are urgently needed, especially to treat immunocompromised patients and those individuals at high risk of developing severe COVID-19. While only a few human mAbs have shown resilience against Omicron sublineages, antibodies SA58 and SA55 have shown partial or full coverage in the Omicron landscape 5. Interestingly, these antibodies bind to the RBD in a region distant from the 17T2 epitope, and structural data35 suggests that the RBD could simultaneously accommodate all three broadly neutralizing antibodies. Therefore, the combined use of these antibodies could improve their clinical efficacy, as previously suggested for combinations of mAbs active against preOmicron variants 36. The identification of an anti-RBM broadly neutralizing antibody has several implications for future COVID-19 pandemic management. First, 17T2 has been cloned from a circulating IgA+ memory B cell isolated from a convalescent COVID-19 patient infected with the ancestral SARS-CoV-2. Therefore, our study provides evidence that infection with the ancestral virus could elicit broadly neutralizing antibodies, likely of mucosal origin, to SARS-CoV-2 variants not yet circulating. Additionally, and considering the benefits afforded by antibody-based therapies to treat COVID-19 patients and the excellent and broad neutralizing activity of 17T2 mAb in vitro and in vivo, we believe that 17T2 represents a promising candidate for future therapeutic and prophylactic interventions, either alone or in combination with other antibodies. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 16 refinements with CryoSPARC were performed, using a mask encompassing the RBD-Fab region (Supplementary Figure 3E). The final resolution was 4.41 FÅ map, which allowed for increased definition in this region (Supplementary Figure 3G). However, the side chains were not fully resolved. Iterative manual model building and real-space refinement were carried out in Coot 49 and in Phenix 50 and CCP-EM51. The validation of the model was done with Molprobity 52 (Supplementary Table 5), sofware integrated in Phenix suite. UCSF Chimera and ChimeraX were used for map fitting and manipulation 53. Statistical analysis All figures were generated in GraphPad Prism 9.0.0. Statistical analyses were performed using R v4.1.1. Unpaired datasets were analyzed using a Kruskal-Wallis with Dunn’s correction for multiple testing. Histopathological and IHC scores were compared using the Generalized Pearson Chi-Squared Test for ordinal data. Viral load comparisons were analyzed using a Petro-Prentice generalized Wilcoxon test. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 17 References 1. Wang, Q. et al. Antibody evasion by SARS-CoV-2 Omicron subvariants BA.2.12.1, BA.4 and BA.5. Nature 608, 603–608 (2022). 2. Cao, Y. et al. Omicron escapes the majority of existing SARS-CoV-2 neutralizing antibodies. Nature 602, 657–663 (2022). 3. Harvey, W. T. et al. SARS-CoV-2 variants, spike mutations and immune escape. Nat. Rev. Microbiol. 19, 409–424 (2021). 4. Dejnirattisai, W. et al. SARS-CoV-2 Omicron-B.1.1.529 leads to widespread escape from neutralizing antibody responses. Cell 185, 467-484.e15 (2022). 5. Cao, Y. et al. Imprinted SARS-CoV-2 humoral immunity induces convergent Omicron RBD evolution. Nature (2022) doi:10.1038/S41586-022-05644-7. 6. Kuhlmann, C. et al. Breakthrough infections with SARS-CoV-2 omicron despite mRNA vaccine booster dose. Lancet (London, England) 399, 625–626 (2022). 7. Jiang, X.-L. et al. Omicron BQ.1 and BQ.1.1 escape neutralisation by omicron subvariant breakthrough infection. Lancet. Infect. Dis. 23, 28–30 (2023). 8. Corti, D., Purcell, L. A., Snell, G. & Veesler, D. Tackling COVID-19 with neutralizing monoclonal antibodies. Cell 184, 3086–3108 (2021). 9. Taylor, P. C. et al. Neutralizing monoclonal antibodies for treatment of COVID-19. Nat. Rev. Immunol. 2021 216 21, 382–393 (2021). 10. Weinreich, D. M. et al. REGEN-COV Antibody Combination and Outcomes in Outpatients with Covid-19. N. Engl. J. Med. 385, e81 (2021). 11. Gupta, A. et al. Early Treatment for Covid-19 with SARS-CoV-2 Neutralizing Antibody Sotrovimab. N. Engl. J. Med. 385, 1941–1950 (2021). 12. Arora, P. et al. Omicron sublineage BQ.1.1 resistance to monoclonal antibodies. Lancet. Infect. Dis. 23, (2023). 13. de Campos-Mata, L. et al. SARS-CoV-2 sculpts the immune system to induce sustained virusspecific naïve-like and memory B-cell responses. Clin. Transl. Immunol. 10, (2021). 14. Wardemann, H. & Busse, C. E. Expression Cloning of Antibodies from Single Human B Cells. Methods Mol. Biol. 1956, 105–125 (2019). 15. Gaebler, C. et al. Evolution of antibody immunity to SARS-CoV-2. Nature 591, 639–644 (2021). 16. Dejnirattisai, W. et al. The antigenic anatomy of SARS-CoV-2 receptor binding domain. Cell 184, 2183-2200.e22 (2021). .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 18 17. Planchais, C. et al. Potent human broadly SARS-CoV-2-neutralizing IgA and IgG antibodies effective against Omicron BA.1 and BA.2. J. Exp. Med. 219, (2022). 18. Kreer, C. et al. Longitudinal Isolation of Potent Near-Germline SARS-CoV-2-Neutralizing Antibodies from COVID-19 Patients. Cell 182, 843-854.e12 (2020). 19. Robbiani, D. F. et al. Convergent antibody responses to SARS-CoV-2 in convalescent individuals. Nature 584, 437–442 (2020). 20. Tortorici, M. A. et al. Ultrapotent human antibodies protect against SARS-CoV-2 challenge via multiple mechanisms. Science 370, 950–957 (2020). 21. Starr, T. N. et al. SARS-CoV-2 RBD antibodies that maximize breadth and resistance to escape. Nat. 2021 5977874 597, 97–102 (2021). 22. Trinité, B. et al. Previous SARS-CoV-2 Infection Increases B.1.1.7 Cross-Neutralization by Vaccinated Individuals. Viruses 13, (2021). 23. Chen, Y. et al. Broadly neutralizing antibodies to SARS-CoV-2 and other human coronaviruses. Nat. Rev. Immunol. (2022) doi:10.1038/S41577-022-00784-3. 24. Pinto, D. et al. Cross-neutralization of SARS-CoV-2 by a human monoclonal SARS-CoV antibody. Nature 583, 290–295 (2020). 25. Halfmann, P. J. et al. SARS-CoV-2 Omicron virus causes attenuated disease in mice and hamsters. Nature 603, 687–692 (2022). 26. Tarrés-Freixas, F. et al. Heterogeneous Infectivity and Pathogenesis of SARS-CoV-2 Variants Beta, Delta and Omicron in Transgenic K18-hACE2 and Wildtype Mice. Front. Microbiol. 13, (2022). 27. Liu, C. et al. The antibody response to SARS-CoV-2 Beta underscores the antigenic distance to other variants. Cell Host Microbe 30, 53-68.e12 (2022). 28. Wang, Q. et al. Alarming antibody evasion properties of rising SARS-CoV-2 BQ and XBB subvariants. Cell (2022) doi:10.1016/J.CELL.2022.12.018. 29. Greaney, A. J. et al. Mapping mutations to the SARS-CoV-2 RBD that escape binding by different classes of antibodies. Nat. Commun. 12, (2021). 30. Francino-Urdaniz, I. M. et al. One-shot identification of SARS-CoV-2 S RBD escape mutants using yeast screening. Cell Rep. 36, (2021). 31. Dong, J. et al. Genetic and structural basis for SARS-CoV-2 variant neutralization by a twoantibody cocktail. Nat. Microbiol. 6, 1233–1244 (2021). 32. van de Bovenkamp, F. S., Hafkenscheid, L., Rispens, T. & Rombouts, Y. The Emerging Importance of IgG Fab Glycosylation in Immunity. J. Immunol. 196, 1435–1441 (2016). 33. Miller, J. et al. Substantial Neutralization Escape by the SARS-CoV-2 Omicron Variant BQ.1.1. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 19 bioRxiv 2022.11.01.514722 (2022) doi:10.1101/2022.11.01.514722. 34. Yue, C. et al. Enhanced transmissibility of XBB.1.5 is contributed by both strong ACE2 binding and antibody evasion. bioRxiv 2023.01.03.522427 (2023) doi:10.1101/2023.01.03.522427. 35. Cao, Y. et al. Rational identification of potent and broad sarbecovirus-neutralizing antibody cocktails from SARS convalescents. Cell Rep. 41, 111845 (2022). 36. Barnes, C. O. et al. SARS-CoV-2 neutralizing antibody structures inform therapeutic strategies. Nature 588, 682–687 (2020). 37. Pradenas, E. et al. Virological and Clinical Determinants of the Magnitude of Humoral Responses to SARS-CoV-2 in Mild-Symptomatic Individuals. Front. Immunol. 13, (2022). 38. Connor, R. I., Chen, B. K., Choe, S. & Landau, N. R. Vpr is required for efficient replication of human immunodeficiency virus type-1 in mononuclear phagocytes. Virology 206, 935–944 (1995). 39. Ou, X. et al. Characterization of spike glycoprotein of SARS-CoV-2 on virus entry and its immune cross-reactivity with SARS-CoV. Nat. Commun. 11, (2020). 40. Brustolin, M. et al. Protection against reinfection with D614or G614-SARS-CoV-2 isolates in golden Syrian hamster. Emerg. Microbes Infect. 10, 797–809 (2021). 41. Vidal, E. et al. Chronological brain lesions after SARS-CoV-2 infection in hACE2-transgenic mice. Vet. Pathol. 59, 613–626 (2022). 42. Katsamba, P. S. et al. Kinetic analysis of a high-affinity antibody/antigen interaction performed by multiple Biacore users. Anal. Biochem. 352, 208–221 (2006). 43. de la Rosa-Trevín, J. M. et al. Scipion: A software framework toward integration, reproducibility and validation in 3D electron microscopy. J. Struct. Biol. 195, 93–99 (2016). 44. Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. Nat. Methods 14, 331–332 (2017). 45. Zivanov, J. et al. New tools for automated high-resolution cryo-EM structure determination in RELION-3. Elife 7, (2018). 46. Zhang, K. Gctf: Real-time CTF determination and correction. J. Struct. Biol. 193, 1–12 (2016). 47. Punjani, A., Rubinstein, J. L., Fleet, D. J. & Brubaker, M. A. cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nat. Methods 14, 290–296 (2017). 48. Sanchez-Garcia, R. et al. DeepEMhancer: a deep learning solution for cryo-EM volume postprocessing. Commun. Biol. 4, (2021). 49. Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D. Biol. Crystallogr. 66, 486–501 (2010). 50. Liebschner, D. et al. Macromolecular structure determination using X-rays, neutrons and .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 20 electrons: recent developments in Phenix. Acta Crystallogr. Sect. D, Struct. Biol. 75, 861–877 (2019). 51. Wood, C. et al. Collaborative computational project for electron cryo-microscopy. Acta Crystallogr. D. Biol. Crystallogr. 71, 123–126 (2015). 52. Davis, I. W., Murray, L. W., Richardson, J. S. & Richardson, D. C. MOLPROBITY: structure validation and all-atom contact analysis for nucleic acids and their complexes. Nucleic Acids Res. 32, (2004). 53. Pettersen, E. F. et al. UCSF Chimera--a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612 (2004). .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 21 Acknowledgements We acknowledge access to the cryo-EM CNB-CSIC facility in the context of the CRIOMECORR project (ESFRI-2019-01-CSIC-16) and we thank the staff of the Protein Technology Unity (CRG) for the help in protein production. This study was supported by the COVID-19 call grant from Generalitat de Catalunya, Department of Health (to GM), grant Miguel Servet research program (to GM), and partially funded by the crowdfunding initiative #joemcorono and the Fundació Glòria Soler (to JB). A.P-G. was supported by a predoctoral grant from Generalitat de Catalunya and Fons Social Europeu (2022 FI_B 00698). Author Contributions L.D.C.-M. produced the monoclonal antibodies, designed and performed experiments, analyzed and discussed data and reviewed and edited the manuscript. S.T.V. produced the monoclonal antibodies, designed and performed experiments and discussed data. N.R.M. performed SARS-CoV-2 antigens and mAbs expression and purification and carried out SPR experiments. B.T. carried out neutralization assays and in vivo experiments with K18-hACE-2 mice, analyzed and interpreted the data and contributed to writing of the manuscript. E.P. and S.M. carried out neutralization assays. A.P.-G., F.T.- F. and N.I. carried out in vivo experiments with K18-hACE-2 mice. E.R.-M. and E.B. performed VL quantification. J.V.-A. carried out immunohistochemistry and histology experiments and analyzed and interpreted the data. M.P. carried out immunohistochemistry and histology experiments. J.S. carried out immunohistochemistry and histology experiments and analyzed and interpreted the data. E.B. analyzed and interpreted the data. A.M. carried out computational aspects of image processing and structure determination, interpreted and analyzed the cryo-EM structures and binding analyses and carried out experimental aspects of cryo-EM. M.T.B.-C. collected cryo-EM data. D.C. carried out computational aspects of image processing and structure determination. C.S. interpreted and analyzed the cryo-EM structures and binding analyses. R.A. carried out experimental aspects of cryo-EM, interpreted and analyzed the cryo-EM structures and binding analyses and wrote the manuscript. C.C. designed and performed experiments, carried out SPR experiments, discussed data and reviewed and edited the manuscript. J.B analyzed and interpreted the data and wrote the manuscript. G.M. designed the study, produced the monoclonal antibodies, analyzed and interpreted data, and wrote the manuscript. Competing Interests Statement .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 22 Unrelated to the submitted work, J.B. is founder and shareholder of AlbaJuna Therapeutics, S.L; J.B. reports institutional grants from HIPRA, NESAPOR Europe, and MSD. The authors declare no other competing conflicts of interests. This work is protected by intellectual property rights through the patent EP22382940, published by G.M., S.T.V., L.D.C.-M., C.C., J.B.A., B.T., E.P. (co-authors of this work). Data and materials availability The data that support this study are available from the corresponding authors upon reasonable request. Cryo-EM data have been deposited in the Electron Microscopy Data Bank under accession codes EMD-16453 for SARS-CoV-2 Spike Trimer in complex three 17T2 Fabs and EMD-1643 for RBD/17T2 Fab and its associated atomic models have been deposited in the Protein Data Bank under accession code 8C89. Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead Contact, Giuliana Magri ([email protected]). .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 23 Figure legends Figure 1. Pan-neutralizing activity of 17T2 mAb. (A). Neutralizing activities of the selected mAbs against the indicated SARS-CoV-2 variants. Values are in ng/mL and color coded as indicated in the bottom of the figure. (B) Neutralization curves of 17T2 mAb against the pseudoviruses tested in panel A. Values are mean ± SD of duplicate samples. (C) Neutralization curves of 17T2 (red), S309 (grey) and 2E12 (blue) mAbs against the indicated SARS-CoV-2 variants or SARS-CoV-1. Values are mean ± SD of duplicate samples (D) Summary of IC50 values from panel C in ng/mL and color coded as indicated in the bottom of the figure. (E) Impact of Omicron subvariants on neutralization capacity. IC50 values for pre-Omicron variants (WH1, D614G, Alpha, Beta Gamma and Delta) were grouped and compared to Omicron subvariants (BA.1, BA.2, BA.4/5 and BQ.1.1). Only comparisons of Omicron variants for the indicated mAbs are shown * (p<0.05, Kruskal-Wallis test with correction for multiple comparisons). Figure 2. Prophylactic 17T2 protection against SARS-CoV-2 BA1.1 infection in K18-hACE2 transgenic mice. (A) Schematic description of the experimental setting. Transgenic K18-hACE2 mice were administered intraperitoneally with either 17T2 (17T2 mAb, n=10) or an isotype control (IgGb12, n=10). After 24h, treated animals were intranasally challenged with an Omicron BA.1.1 SARS-CoV2 isolate (n=20), or PBS (Uninfected Control Group) (n=4). Mice were monitored for 7 days. Euthanasia was performed 3and 7-days post-infection (dpi) (n=5 for each treated group per timepoint, n=2 uninfected per timepoint) for sample and tissue collection. Created with Biorender.com. (B) Relative K18-hACE2 transgenic mice weight-loss follow-up. Groups include: uninfected (grey, n=4), 17T2 mAb-treated and infected (blue, n=10), and isotype control-treated and infected (black, n=10) animals. Solid lines and bars represent mean± SD. (C) SARS-CoV-2 viral RNA load quantification (copies/mL) on different tissues: oropharyngeal swab, lung and nasal turbinate at 3 and 7 dpi in 17T2 mAb-treated (blue dots, n=5 per timepoint), isotype control-treated (black squares, n=5 per time point) infected animals. Limit of detection is represented by a dashed line. Statistically differences were determined using a Peto-Prentice generalized Wilcoxon test. (D) Histopathological and immunohistochemical scores of lungs from infected K18-hACE2 mice. Lesion (broncho-interstitial pneumonia) scoring: 0 = no lesion, 1 = mild lesion, 2 = moderate lesion, 3 = severe lesion. IHC scoring: 0 = no antigen, 1 = low and multifocal antigen, 2 = moderate and multifocal antigen, 3 = high and diffuse antigen. # indicates an animal which presented focal expression only visible on 1 out of 5 lung sections: it was scored as 1 but with minimal detection of virus replication. All other positive scores showed multifocal distribution on multiple lung sections. Comparisons were performed using an Asymptotic Generalized Pearson Chi-Squared Test for ordinal data with pairwise comparisons. (E) .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint 24 Representative histopathological and immunohistochemical findings, at 3 and 7 dpi, in K18-hACE2 transgenic infected mice treated with either 17T2 mAb or an isotype control. Histological slides were stained with hematoxylin and eosin, and immunohistochemistry ones were counterstained with Hematoxylin. Scale = 80 µM. Figure 3. Structural and functional characterization of complex 17T2 Fab fragment with Omicron BA.1 spike using cryo-EM. (A and B) Top and side views of the cryo-EM map of Omicron BA.1 spike trimer with three 17T2 Fab fragments bound to three open RBDs. The core of the spike is shown in grey, the RBDs in pink, and the heavy chain (HC) and light chain (LC) 17T2 of the Fab in green and blue, respectively. (C) Structure of the RBD and 17T2 Fab after local refinement. The interaction zone between the RBD and the Fab is shown in cartoon representation where the three CDRs from each chain are distinctively colored and the N-glycosylation is indicated in orange. (D) CDRs that are involved in binding with RBD, specifically, interacting in the region of its left shoulder and neck (front and back view, respectively). (E) A detailed view of some interactions between 17T2 Fab and RBD. The main chains are colored in grey and the side chains of the residues involved in the interaction are shown in green for HC, in blue for LC, and in pink for RBD. (F) Locations of SARSCoV-2 variant mutations on RBD relative to 17T2 epitope site that is shown as a black line (front and back view, respectively). The information about the variants and mutated residues can be found in Supplementary Table 4. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 23, 2023. ; https://doi.org/10.1101/2023.01.20.524748doi: bioRxiv preprint