Deep Learning-Based Virtual Screening Identifies Potential Small-Molecule Inhibitors of ASK1: Natural Product Lead...

Area of Science:

• Pharmacology
• Computational Chemistry
• Drug Discovery

Background:

• Apoptosis signal-regulating kinase 1 (ASK1) is a key therapeutic target for metabolic dysfunction-associated steatohepatitis (MASH).
• Natural products offer diverse chemical structures for identifying novel ASK1 inhibitors.
• Artificial intelligence-assisted drug discovery (AIDD) accelerates the exploration of small-molecule inhibitors.

Purpose of the Study:

• To identify novel natural product-derived ASK1 inhibitors for MASH treatment using AIDD.
• To systematically evaluate the binding mode, stability, and key interactions of a natural product candidate (CMNPD10921) with ASK1.

Main Methods:

• Virtual screening and molecular docking were employed to identify potential drug candidates from a natural product library.
• Molecular dynamics simulations and MM-GBSA calculations assessed the binding stability and free energy of the CMNPD10921-ASK1 complex.
• Residue energy decomposition analysis pinpointed key amino acid residues involved in ligand binding.

Main Results:

• CMNPD10921 was identified as a putative lead compound targeting the ASK1 active pocket.
• The CMNPD10921-ASK1 complex exhibited stable binding through hydrophobic interactions and hydrogen bonds with key residues.
• A computed binding free energy of -31.15 kcal/mol, driven by van der Waals forces, indicated favorable binding.

Conclusions:

• CMNPD10921, identified through AIDD, is a promising candidate for developing novel anti-MASH therapeutics.
• The study highlights the successful integration of deep learning, molecular simulation, and quantum chemical calculations in drug discovery.
• This research provides a novel molecular candidate targeting the ASK1 regulatory region for MASH treatment.