Ensemble docking based virtual screening of SARS-CoV-2 main protease inhibitors

Area of Science:

• Structural biology
• Drug discovery
• Computational chemistry

Background:

• The COVID-19 pandemic spurred rapid acquisition of X-ray structures for viral drug targets, including SARS-CoV-2 main protease (Mpro).
• Mpro is a validated target for direct-acting antiviral therapies against SARS-CoV-2.
• Selecting an optimal structural ensemble for virtual screening of Mpro inhibitors is challenging.

Purpose of the Study:

• To develop and validate a semi-automated procedure for selecting representative SARS-CoV-2 Mpro structures.
• To perform a docking-driven virtual screening campaign to identify novel Mpro inhibitors.
• To discover new chemotypes with confirmed enzyme inhibition.

Main Methods:

• A semi-automated active site Root Mean Square Deviation (RMSD) based method was employed for ensemble selection from SARS-CoV-2 Mpro crystallographic data.
• The developed procedure was compared against alternative ensemble selection strategies.
• Validation was performed using curated, activity-annotated compound libraries.

Main Results:

• The RMSD-based procedure provided an effective means for selecting a diverse and relevant set of Mpro structures.
• Prospective virtual screening identified a novel chemotype of non-covalent Mpro inhibitors.
• Experimentally confirmed enzyme inhibition was achieved for the identified thienopyrimidinone derivatives.

Conclusions:

• The developed semi-automated ensemble selection method streamlines structure-based drug discovery for viral targets like SARS-CoV-2 Mpro.
• The identification of thienopyrimidinone derivatives represents a significant advancement in the search for novel COVID-19 antiviral agents.
• This approach facilitates the discovery of new chemical entities targeting essential viral enzymes.