Computational Mutagenesis at the SARS-CoV-2 Spike Protein/Angiotensin-Converting Enzyme 2 Binding Interface: Comparison with Experimental Evidence.

Abstract

The coronavirus disease-2019 (COVID-19) pandemic, caused by the pathogen severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), started in China during late 2019 and swiftly spread worldwide. Since COVID-19 emergence, many therapeutic regimens have been relentlessly explored, and although two vaccines have just received emergency use authorization by different governmental agencies, antiviral therapeutics based neutralizing antibodies and small-drug inhibitors can still be vital viable options to prevent and treat SARS-CoV-2 infections. The viral spike glycoprotein (S-protein) is the key molecular player that promotes human host cellular invasion via recognition of and binding to the angiotensin-converting enzyme 2 gene (ACE2). In this work, we report the results obtained by mutating <i>in silico</i> the 18 ACE2 residues and the 14 S-protein receptor binding domain (S-RBD_CoV-2) residues that contribute to the receptor/viral protein binding interface. Specifically, each wild-type protein-protein interface residue was replaced by a hydrophobic (isoleucine), polar (serine and threonine), charged (aspartic acid/glutamic acid and lysine/arginine), and bulky (tryptophan) residue, respectively, in order to study the different effects exerted by nature, shape, and dimensions of the mutant amino acids on the structure and strength of the resulting binding interface. The computational results were next validated <i>a posteriori</i> against the corresponding experimental data, yielding an overall agreement of 92%. Interestingly, a non-negligible number of mis-sense variations were predicted to enhance ACE2/S-RBD_CoV-2 binding, including the variants Q24T, T27D/K/W, D30E, H34S7T/K, E35D, Q42K, L79I/W, R357K, and R393K on ACE2 and L455D/W, F456K/W, Q493K, N501T, and Y505W on S-RBD_CoV-2, respectively.