A Novel Covalent Inhibitor Fragment for the SARS-CoV-2 Main Protease Identified by Target-Specific Deep Learning.

Area of Science:

• Virology
• Drug Discovery
• Computational Chemistry

Background:

• The SARS-CoV-2 main protease (M^pro) is crucial for viral replication and a key target for antiviral drug development.
• Developing novel M^pro inhibitors is essential due to the continuous evolution of SARS-CoV-2.
• Existing experimental data on M^pro inhibitors provides a foundation for computational drug discovery.

Purpose of the Study:

• To accelerate the discovery of new SARS-CoV-2 M^pro inhibitor compounds and fragment-like starting points.
• To develop and apply a target-specific deep learning workflow for identifying potential drug candidates.
• To prioritize novel, drug-like compounds from a large chemical library for experimental validation.

Main Methods:

• Developed a deep learning workflow integrating inhibitor prediction, solubility (logS), and lipophilicity (logP) models.
• Utilized molecular similarity analysis and literature mining to refine candidate prioritization.
• Screened a purchasable library of over 500,000 compounds.

Main Results:

• Identified 24 potential M^pro inhibitor candidates for experimental testing.
• Discovered a novel covalent inhibitor fragment (A02) with an IC₅₀ of 1.5 μM.
• Determined the crystal structure of M^pro bound to A02, revealing covalent modification of C145 and unique S3' pocket engagement.

Conclusions:

• The developed deep learning workflow effectively accelerates the screening and discovery of novel inhibitor leads.
• The identified fragment A02 represents a promising scaffold for developing new SARS-CoV-2 M^pro inhibitors.
• Target-specific deep learning approaches show significant potential for guiding future antiviral drug discovery efforts.