Computer-aided drug design of some KRAS G12C inhibitors: Targeting the covalent and allosteric binding site for cancer therapy

Abstract

KRAS is one of the most investigated anticancer targets today, owing to its common mutations in a variety of lethal malignancies. KRAS G12C has been the focus of cancer research since the identification of the druggable allosteric binding site located in close vicinity with G12C mutation. Herein, we present a computer-assisted drug discovery strategy for discovering new and selective KRAS G12C covalent inhibitors. With the screening of an online database that holds millions of assayed and curated compounds targeted against KRAS G12C, a total of 1001 compounds were generated. Furthermore, these ligands were subjected to a ligand-filtering process to identify those with an acrylamide warhead group that could bind covalently with the Cys12 nucleophile of KRAS G12C and like sotorasib, the only clinically approved drug against the therapeutic cancer target till date. The resulting 202 compounds were then subjected to a covalent-docking virtual screening experiment, leading to the identification of three putative binders (CID_13728717, CID_146235508, and CID_146235420) against the KRAS G12C in its inactive GDP-bound conformation. Subsequently, ADMET modeling and molecular dynamics investigations showed why the compounds may be regarded as drug-like and stable covalent binders of KRAS G12C, respectively. In particular, the molecular enumeration profile of CID_13728717 was the most satisfactory, excelling more than the standard sotorasib in this computationally driven drug discovery research. Hence, it is most likely to emerge as an effective KRAS G12C drug after extensive experimental efforts.