Dynamics of the SARS-CoV-2 main protease binding site

The COVID-19 pandemic has been controlled through quarantines, travel bans, mask mandates, and vaccines. However, due to viral mutations and the unlikelihood of reaching herd immunity, there remains a need for therapeutics. The SARS-CoV-2 main protease (MPro) is necessary for the virus’ replication, thus it is a promising drug target. Additionally, it can be crystallized consistently and its sequence has few changes across virus mutants. MPro crystal structures depict general residue-residue interactions and backbone conformations, but they do not account for the dynamics of flexible regions. Instead, molecular dynamics (MD) simulations are used to identify the flexibility of the binding site, data we hypothesize is essential for successfully designing drugs to bind to MPro using computational drug docking methods. 600 ns of all-atom, explicit solvent MD simulations of the MPro homodimer were conducted, resulting in 1.2 million frames of monomeric data. Principal component analysis (PCA) was conducted on pairwise Cα-Cα distances between selected binding residues within apo and holo PDBs and the MD frames. These distances provide an intuitive set of measurements to characterize the dynamics of the binding site. Contour mapping was used to visualize the presence of clusters in the space defined by a few of the most contributive PC's. Snapshots representative of the most common conformations within the distribution of data were identified through the use of clustering methods and will be used in future drug docking simulations.