Natural Product-derived Binders of Protein Arginine Deiminase 4 (PAD4) Predicted Using Molecular Modeling.

BACKGROUND: Natural products serve as vital scaffolds for novel drug discovery. PAD4 is an attractive target dysregulated in chronic diseases like Alzheimer's, cancer, and rheumatoid arthritis. Given the lack of clinically approved synthetic inhibitors, exploring structurally diverse natural products offers a promising strategy for novel drug development. METHODS: Molecular modeling using ligand-based virtual screening, consensus molecular docking using four scoring functions available on the Smina and Vina software, ADME-Tox consensus filtering, and molecular dynamics analysis with GROMACS were applied to virtually screen natural products from the COCONUT database as potential PAD4 binders. RESULTS: The docking validation parameters (redocking RMSD < 2 Å, AUC-ROC > 70 %) confirmed the effective discrimination of known PAD4 inhibitors from decoys. From the COCONUT database, 29,164 compounds were initially filtered by similarity to noncovalent PAD4 inhibitors. Among these, 10 compounds were selected as the top candidate inhibitors using a multi-faceted virtual screening protocol that combined a novel consensus pharmacological score with binding affinity, druglikeness, and ADMET evaluations. PASS predictions further corroborated their PAD4 inhibitory potential. Molecular dynamics of the top five compounds confirmed active-site stability and noncovalent interactions with catalytic residues Asp350, Arg374, His471, Asp473, and Cys645. Structurally, these candidates represent diverse natural chemical classes, including benzenoids, alkaloids, and amino acid derivatives. DISCUSSION: The multifaceted virtual screening conducted in this work resulted in confident prioritization of five potential PAD4 inhibitors. Further in vitro and in vivo studies are needed to confirm their functional inhibitory activity in diseases in which this target is dysregulated. CONCLUSION: Five natural chemistry-binders of PAD4 were identified by detailed molecular modeling studies. They could be used as chemical scaffolds for future medicinal chemistry drug design.