To identify new potential xanthine oxidase (XO) inhibitors through hierarchical virtual screening.

The pharmacophore model was constructed based on receptor-ligand crystal complex. And a decoy set was used to verify the pharmacophore model. On this basis, the model was used to find potential XO inhibitors from ZINC natural-products database by combining with molecular docking, ADMET prediction and molecular dynamics simulation.

The optimized pharmacophore model contained two hydrogen bond acceptors, two hydrophobic centers, one negative ionizable center and one aromatic ring center. The docking results of the decoy set indicated that the constructed pharmacophore model had high sensitivity and specificity. Molecule ZINC0934445 which was obtained through hierarchical virtual screening was predicted to have good inhibitory activity and drug-like properties. And molecular dynamics simulation results also showed that its binding free energy was better than the positive drug febuxostat.

A variety of computer virtual screening technologies were comprehensively integrated to identify novel XO inhibitors. And one potential active molecule ZINC09344458 was discovered from natural products, which would relieve pain associated with gout. This study provided an alternative method for relevant experimental studies on the development of gout therapeutic drugs.