Cross-reactivity of eight SARS-CoV-2 variants rationally predicts immunogenicity clustering in sarbecoviruses

A steep rise in Omicron reinfection cases suggests that this variant has increased immune evasion ability. To evaluate its antigenicity relationship with other variants, antisera from guinea pigs immunized with spike protein of SARS-CoV-2 variants of concern (VOCs) and variants of interest (VOIs) were cross-tested against pseudotyped variants. The neutralization activity against Omicron was markedly reduced when other VOCs or VOIs were used as immunogens, and Omicron (BA.1)-elicited sera did not efficiently neutralize the other variants. However, a Beta or Omicron booster, when administered as the 4th dose 3-months after the 3rd dose of any of the variants, could elicit broad neutralizing antibodies against all of the current variants including Omicron BA.1. Further analysis with 280 available antigen-antibody structures and quantification of immune escape from 715 reported neutralizing antibodies provide explanations for the observed differential immunogenicity. Three distinct clades predicted using an in silico algorithm for clustering of sarbecoviruses based on immune escape provide key information for rational design of vaccines.

Automated summary

A steep rise in Omicron reinfection cases indicates that this variant has enhanced immune evasion ability. Antisera from guinea pigs immunized with spike protein of SARS-CoV-2 variants of concern (VOCs) and variants of interest (VOIs) were cross-tested against pseudotyped variants to evaluate their antigenicity relationship. The study found that the neutralization activity against Omicron was significantly reduced when other VOCs or VOIs were used as immunogens. However, administering a Beta or Omicron booster as the 4th dose 3 months after the 3rd dose of any variant could elicit broad neutralizing antibodies against all current variants, including Omicron BA.1. Further analysis and quantification of immune escape from reported neutralizing antibodies provided explanations for the observed differential immunogenicity. An in silico algorithm predicted three distinct clades based on immune escape, which can be crucial for designing vaccines.