Screening for the selective inhibitors of MMP-9 from natural products based on pharmacophore modeling and molecular docking in combination with bioassay experiment, hybrid QM/MM calculation, and MD simulation.

Matrix metalloproteinase-9 (MMP-9) has been considered as an attractive target involving cancer therapy. In this study, the 3D QSAR pharmacophore model of MMP-9 inhibitors is built, and its reliability is subsequently validated based on different methods. The built pharmacophore model consists of the four chemical features, including two hydrogen bond acceptors (HBA), one hydrophobic (HY), and one ring aromatic (RA). Among them, both HY and RA are found to be especially important features because they involve the interactions of inhibitors with the S1' pocket of MMP-9, which determines the selectivity of MMP-9 inhibitors. By combining pharmacophore model with molecular docking, the virtual screening is carried out to identify the selective MMP-9 inhibitors from natural products. The four potential selective MMP-9 inhibitors of natural products are found. One of them was used to carry out the bioassay experiment inhibiting MMP-9, and the estimated IC value of only 26.94 µM clearly shows its strongly inhibitory activity; besides, both the hybrid quantum mechanics/molecular mechanics (QM/MM) calculation and the molecular dynamics simulation are performed to examine the reliability regarding the binding mode of this inhibitor with MMP-9 active sites predicted by molecular docking. All the screened four natural products are found to well bind with the MMP-9 active sites by different kinds of interactions. Finally, the ADMET properties of screened four natural products are assessed. These screened MMP-9 inhibitors of natural products could be used as the lead compounds to perform structural modifications and optimizations in the future work. Communicated by Ramaswamy H. Sarma.