A multiscale screening strategy for the identification of novel xanthine oxidase inhibitors based on the pharmacological features of febuxostat analogues

Hyperuricemia (HUA), an abnormally high level of uric acid (UA) in the blood, is showing an increasing incidence and younger tendency worldwide. Xanthine oxidase (XO) inhibitors (XOIs) can reduce the UA production by inhibiting the XO activity, which has been considered as a promising strategy for HUA treatment. In this study, pharmacophore models were first established to extract the key features of potent XOIs, including febuxostat and topiroxostat. Multiple virtual screening was then carried out based on the best pharmacophore model, and 4 hits (N1-4) with satisfactory pharmacokinetic parameters by prediction and ideal docking modes were retrieved from the ZINC database. The 4 protein-hit complexes were then subjected to 50 ns molecular dynamics (MD), which verified the stability of these hits in the XO binding pocket. The binding free energies revealed that hits N3 and N4 might possess better stability and higher affinity with XO. The residue-based energy contribution indicated that some key residues, including Asn768, Lys771, Glu802, Arg880, Phe914, Phe1009, and Thr1010, could affect the binding of compounds with XO through hydrogen bonds, pi-pi stacking, and hydrophobic interactions. These results might provide helpful information for the design and development of novel XOIs.