Computational study of the strong binding mechanism of SARS-CoV-2 spike and ACE2

The spike protein of SARS-CoV-2 (SARS-CoV-2-S) helps the virus attach to and infect human cells. With various computational methods applied in this work, the accessibility of its RBD to ACE2, its key residues for stronger binding to ACE2 than the SARS-CoV spike (SARS-CoV-S), the origin of the stronger binding, and its potential sites for drug and antibody design were explored. It was found that the SARS-CoV-2-S could bind ACE2 with an RBD-angle ranging from 52.2º to 84.8º, which demonstrated that the RBD does not need to fully open to bind ACE2. Free energy calculation by an MM/GBSA approach not only revealed much stronger binding of SARS-CoV-2-S to ACE2 (ΔG=-21.7~-29.9 kcal/mol) than SARS-CoV-S (ΔG=-10.2~-16.4 kcal/mol) at different RBD-angles but also demonstrated that the binding becomes increasingly stronger as the RBD-angle increases. In comparison with the experimental results, the free energy decomposition disclosed more key residues interacting strongly with ACE2 than with the SARS-CoV-S, among which the Q493 might be the decisive residue variation (-5.84 kcal/mol) to the strong binding. With the mutation of all 18 different residues of SARS-CoV-S on the spike-ACE2 interface to the corresponding residues of SARS-CoV-2-S, it was found that the mutated SARS-CoV-S has almost the same binding affinity as SARS-CoV-2-S to ACE2, demonstrating that the remaining mutations outside the spike-ACE2 interface have little effect on its binding affinity to ACE2. Simulation of the conformational change pathway from “down” to “up” states disclosed 5 potential ligand-binding pockets correlated to the conformational change. Taking together the key residues, accessible RBD-angle and pocket correlation, potential sites for drug and antibody design were proposed, which should be helpful for interpreting the high infectiousness of SARS-CoV-2 and for developing a cure.