A new glimpse on the active site of SARS-CoV-2 3CLpro, coupled with drug repurposing study.

Coronavirus disease 2019 (COVID-19) is caused by novel severe acute respiratory syndrome coronavirus (SARS-CoV-2). Its main protease, 3C-like protease (3CLpro), is an attractive target for drug design, due to its importance in virus replication. The analysis of the radial distribution function of 159 3CLpro structures reveals a high similarity index. A study of the catalytic pocket of 3CLpro with bound inhibitors reveals that the influence of the inhibitors is local, perturbing dominantly only residues in the active pocket. A machine learning based model with high predictive ability against SARS-CoV-2 3CLpro is designed and validated. The model is used to perform a drug-repurposing study, with the main aim to identify existing drugs with the highest 3CLpro inhibition power. Among antiviral agents, lopinavir, idoxuridine, paritaprevir, and favipiravir showed the highest inhibition potential. Enzyme - ligand interactions as a key ingredient for successful drug design.

Evaluation of SARS-CoV-2 main protease and inhibitor interactions using dihedral angle distributions and radial distribution function.

In order to evaluate the interactions between a potential drug candidate like inhibitor N3 and the residues in substrate binding site of SARS-CoV-2 main protease ( M pro ), we used molecular docking and dynamics simulations. The structural features describing the degrees of folding states of M pro formed by beta-barrels and alpha-helices were analyzed by means of root mean square deviation, root mean square fluctuation, radius of gyration, residue velocity, H-bonding, dihedral angle distributions and radial distribution function. All of the residues forming ligand binding domain (LBD) of M pro lie within the allowed region of the dihedral angle distributions as observed from the equilibrating best pose of M pro -N3 system. Sharp peaks of radial distribution function (RDF) for H-bonding atom pairs (about 2 Å radial distance apart) describe the strong interactions between inhibitor and SARS-CoV-2 M pro . During MD simulations, HSE163 has the lowest residue speed offering a sharp RDF peak whereas GLN192 has the highest residue speed resulting a flat RDF peak for the H-bonding atom pairs of M pro -N3 system. Along with negative values of coulombic and Lenard-Jones energies, MM/PBSA free energy of binding contributed by the non-covalent interactions between M pro and N3 has been obtained to be -19.45 ± 3.6 kcal/mol. These physical parameters demonstrate the binding nature of an inhibitor in M pro -LBD. This study will be helpful in evaluating the drug candidates which are expected to inhibit the SARS-CoV-2 structural proteins.