Anatomy of Omicron BA.1 and BA.2 neutralizing antibodies in COVID-19 mRNA vaccinees

SARS-CoV-2 vaccines, administered to billions of people worldwide, mitigate the effects of the COVID-19 pandemic, however little is known about the molecular basis of antibody cross-protection to emerging variants, such as Omicron BA.1, its sublineage BA.2, and other coronaviruses. To answer this question, 276 neutralizing monoclonal antibodies (nAbs), previously isolated from seronegative and seropositive donors vaccinated with BNT162b2 mRNA vaccine, were tested for neutralization against the Omicron BA.1 and BA.2 variants, and SARS-CoV-1 virus. Only 14.2, 19.9 and 4.0% of tested antibodies neutralize BA.1, BA.2, and SARS-CoV-1 respectively. These nAbs recognize mainly the SARS-CoV-2 receptor binding domain (RBD) and target Class 3 and Class 4 epitope regions on the SARS-CoV-2 spike protein. Interestingly, around 50% of BA.2 nAbs did not neutralize BA.1 and among these, several targeted the NTD. Cross-protective antibodies derive from a variety of germlines, the most frequents of which were the IGHV1-58;IGHJ3-1, IGHV2-5;IGHJ4-1 and IGHV1-69;IGHV4-1. Only 15.6, 20.3 and 7.8% of predominant gene-derived nAbs elicited against the original Wuhan virus cross-neutralize Omicron BA.1, BA.2 and SARS-CoV-1 respectively. Our data provide evidence, at molecular level, of the presence of cross-neutralizing antibodies induced by vaccination and map conserved epitopes on the S protein that can inform vaccine design.

S ince its first appearance in December 2019, more than 495 million cases of SARS-CoV-2 infections were reported worldwide, with over 6.1 million deaths. Effective vaccines against the virus that first appeared in Wuhan, China, have been developed with unprecedented speed. However, their ability to contain the global pandemic has been compromised by the inability to timely deliver vaccines to low-income countries and by the appearance of several antigenic variants which escaped the natural and vaccine-induced immunity [1][2][3] . The main variants that emerged so far, and are listed as variants of concern (VoCs), are named B.1.1.7 (Alpha), B.1.351 (Beta), P.1 (Gamma), B.1.617.2 (Delta), and B.1.1.529.1 (Omicron; BA.1) 4,5 . The latter one showed to be the most efficient in spreading into partially immune populations and in a few months from its appearance has conquered most regions of the world 6,7 . Shortly after the appearance of the Omicron variant BA.1, the sublineage BA.2 (B.1.1.529.2) was identified, and it is now replacing the initial BA.1 variant worldwide 8,9 . Previous reports have shown that the unprecedented number of mutations carried on the Omicron BA.1 and BA.2 S protein drastically reduce the neutralizing efficacy of sera from infected and vaccinated people and that this VoC can escape more than 85% of nAbs described in literature, including several antibodies approved for clinical use by regulatory agencies [10][11][12][13][14][15][16][17][18] . Despite these observations, recent reports have shown different profiles of immune evasion between omicron BA.1 and BA.2 [19][20][21] . While serum activity and neutralization efficacy of selected monoclonal antibodies against Omicron BA.1 and BA.2 have been reported, the functional and genetic anatomy of nAbs elicited in naïve (seronegative) and convalescent (seropositive) people immunized with two doses of the BNT162b2 mRNA vaccine remains to be explored. Taking advantage of our previous work 22 , we tested 276 human monoclonal antibodies able to neutralize the original SARS-CoV-2 virus isolated in Wuhan, for their ability to neutralize the Omicron BA.1 and BA.2 variants, and the distantly related SARS-CoV-1 virus. Our work unravels the genetic signature of cross-protective antibodies and mapped conserved sites of pathogen vulnerability on the S protein that can be used to design the next generation of sarbecovirus vaccines.

SARS-CoV
Functional repertoire analyses. nAbs VH and VL sequence reads were manually curated and retrieved using CLC sequence viewer (Qiagen). Aberrant sequences were removed from the data set. Analyzed reads were saved in FASTA format and the repertoire analyses were performed using Cloanalyst (http://www.bu.edu/ computationalimmunology/research/software/) 48,49 .
Statistical analysis. Statistical analysis was assessed with GraphPad Prism Version 8.0.2 (GraphPad Software, Inc., San Diego, CA). Nonparametric Mann-Whitney t test was used to evaluate statistical significance between the two groups analyzed in this study. Statistical significance was shown as * for values ≤0.05, ** for values ≤0.01, *** for values ≤0.001, and **** for values ≤0.0001.
Reporting summary. Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability
Source data are provided with this paper. All data supporting the findings in this study are available within the article or can be obtained from the corresponding author upon request. Source data are provided with this paper.