Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, United States
Jeffrey Chang
Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, United States; Department of Physics, Harvard University, Cambridge, United States
Biological and Biomedical Sciences, Harvard Medical School, Boston, United States
Allison J Greaney
Basic Sciences Division and Computational Biology Program, Fred Hutchinson Cancer Research Center, Seattle, United States; Department of Genome Sciences, University of Washington, Seattle, United States; Medical Scientist Training Program, University of Washington, Seattle, United States
Basic Sciences Division and Computational Biology Program, Fred Hutchinson Cancer Research Center, Seattle, United States; Department of Genome Sciences, University of Washington, Seattle, United States; Howard Hughes Medical Institute, Seattle, United States
Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, United States; Department of Physics, Harvard University, Cambridge, United States; NSF-Simons Center for Mathematical and Statistical Analysis of Biology, Harvard University, Cambridge, United States; Quantitative Biology Initiative, Harvard University, Cambridge, United States
The Omicron BA.1 variant of SARS-CoV-2 escapes convalescent sera and monoclonal antibodies that are effective against earlier strains of the virus. This immune evasion is largely a consequence of mutations in the BA.1 receptor binding domain (RBD), the major antigenic target of SARS-CoV-2. Previous studies have identified several key RBD mutations leading to escape from most antibodies. However, little is known about how these escape mutations interact with each other and with other mutations in the RBD. Here, we systematically map these interactions by measuring the binding affinity of all possible combinations of these 15 RBD mutations (215=32,768 genotypes) to 4 monoclonal antibodies (LY-CoV016, LY-CoV555, REGN10987, and S309) with distinct epitopes. We find that BA.1 can lose affinity to diverse antibodies by acquiring a few large-effect mutations and can reduce affinity to others through several small-effect mutations. However, our results also reveal alternative pathways to antibody escape that does not include every large-effect mutation. Moreover, epistatic interactions are shown to constrain affinity decline in S309 but only modestly shape the affinity landscapes of other antibodies. Together with previous work on the ACE2 affinity landscape, our results suggest that the escape of each antibody is mediated by distinct groups of mutations, whose deleterious effects on ACE2 affinity are compensated by another distinct group of mutations (most notably Q498R and N501Y).
