The role of receptor binding and immunity in SARS-CoV-2 fitness landscape: A modeling study

Despite extensive research on SARS-CoV-2 ACE2 binding and transmissibility, their relationship concerning varying immunity remains unclear. SARS-CoV-2 receptor-binding domain (RBD) sequences from Italy in GISAID, combined with multiple deep mutational scanning data, were used to calculate ACE2 binding score and immune escape during the pandemic. We developed a COVID-19 transmission model that decomposed the effective reproduction number into three time-varying components: viral infectiousness (representing fitness), host susceptibility, and contact rate. After model fitting, a rugged fitness landscape, spanned by ACE2 binding score and virus-perceived effective immunity (adjusted for viral immune escape), was observed with peaks corresponding to individual variants of concern (VOCs) (alpha, delta, and omicron (BA.1^∗ and BA.2^{∗ )}). Increasing effective immunity with reduced ACE2 binding corresponded to decreasing virus fitness peaks from alpha to delta. Among omicron sub-lineages, which exhibited enhanced immune escape, BA.2^∗ reached a fitness optimum, with ACE2 binding slightly lower relative to BA.1.1. The findings help explain the evolution of SARS-CoV-2.