Discovery of PAR2 antagonists through fragment molecular orbital-based hotspot analysis and allosteric sodium site...

Protease-activated receptor 2 (PAR2) is a G protein-coupled receptor (GPCR) implicated in inflammatory diseases and cancer, representing a validated therapeutic target. In our previous study, we combined molecular dynamics (MD) and fragment molecular orbital (FMO) analyses to rationalize binding mode differences between PAR2 agonists and antagonists. Here, we advanced this framework into a practical ligand discovery platform that integrates FMO-based hotspot mapping, pharmacophore modeling, structure-based virtual screening, and functional validation. FMO analysis of PAR2 complexes with a peptide agonist, antibody, and antagonist identified residue-level binding hotspots that were translated into a pharmacophore model for screening ∼50,000 in-house compounds. Hit compounds were virtually screened by docking and MD simulations, with Na⁺ ion displacement at the conserved allosteric site serving as a functional filter. Application of a 1.4 Å cutoff excluded agonist-like perturbations while retaining antagonist-like Na⁺ displacement profiles. The initial hit, an indazole derivative S757-0124, engaged key hotspots and exhibited antagonist-like Na⁺ dynamics but suffered from poor solubility and moderate potency. Scaffold replacement with an sp³-rich tetrahydropyrazolopyridine core yielded P2L-003, which showed 35-fold improved potency, 11.6-fold higher solubility, and maintained reduced Na⁺ displacement characteristic of antagonist binding. Functional assays further confirmed that P2L-003 selectively antagonized PAR2 without affecting PAR1, PAR4, or ATP-mediated signaling, while Western blotting demonstrated dose-dependent inhibition of ERK1/2 and p38 phosphorylation. Together, these results highlight how integrating electronic precision from FMO, functional filtering from MD, and medicinal chemistry optimization provides a robust and generalizable workflow for GPCR antagonist discovery.