Rational design of a covalent ACE2 decoy receptor that broadly neutralizes SARS-CoV-2 variants

The ongoing evolution of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and the emergence of variants that evade existing vaccines and antibody therapies necessitate novel, potent, and broad-spectrum antiviral strategies. Enhancing therapeutic proteins with additional covalent binding capabilities, such as minibinders and nanobodies, reportedly potentiates their antiviral efficacy by irreversibly capturing the viral cell-entry protein. However, viral mutations that interfere with covalent bonding or reduce viral affinity with therapeutic proteins might compromise the efficacy of this strategy. Therefore, in this study, we aimed to develop a broadly neutralizing covalent angiotensin-converting enzyme 2-Fc (ACE2-Fc) decoy using a rational design strategy that integrates functional genomics with structural information. Using this approach, we targeted a highly conserved and functionally constrained residue on the viral receptor-binding domain (RBD) and identified tyrosine 473 (Y473) as an optimal target. We engineered ACE2-Fc constructs by replacing glutamate 23 (E23) and threonine 27 (T27) with the non-canonical amino acid-fluorosulfate-L-tyrosine (FSY), generating E23FSY and T27FSY constructs. These constructs formed a specific and efficient covalent bond with Y473 of the RBD. Notably, this covalent capture was retained against the highly mutated Omicron BA.5 RBD. In pseudovirus neutralization assays, both E23FSY and T27FSY exhibited markedly enhanced potency against both wild-type-like (D614G) and Omicron variants compared to their non-covalent counterparts. These results demonstrate that using an inherently escape-resistant decoy receptor to covalently target evolutionarily constrained residues on the viral RBD is a highly efficient strategy for creating potent, broad-spectrum covalent inhibitors against rapidly evolving viruses such as SARS-CoV-2.