QSAR study for the soybean 15-lipoxygenase inhibitory activity of organosulfur compounds derived from the essential oil of garlic

Abstract

In this study, multiple linear regression (MLR) and partial least-squares (PLS) techniques were used for modeling the soybean 15-lipoxygenase inhibitory activity of a varied group of mono-, di-, and trisulfides derived from the essential oil of garlic. The structures of the compounds under study were characterized by means of calculated physicochemical parameters and several nonempirical descriptors, such as topological, geometrical, and quantum chemical indices. The results obtained indicate that the inhibitory activity is strongly dependent on the ability of the compounds to participate in dispersive interactions with the enzyme, as expressed by the solvent-accessible surface area (SASA) and the average distance/distance degree descriptor (ADDD) index. On the other hand, the high contribution of the lowest unoccupied molecular orbit term (LUMO) in the PLS models derived for the di- and trisulfides suggests that the solute's electron-acceptor capacity plays a fundamental role in the inhibitory activity exhibited for these compounds. Finally, the geometric features as expressed by the shape parameters included in the models indicate a low but not negligible positive contribution of molecular linearity in the enzyme-inhibitor binding. In summary, the developed quantitative structure-activity relationship approach successfully accounts for the potencies of organosulfur compounds acting on soybean 15-lipoxygenase and thereby offers both a guide for the synthesis of new compounds and a hypothesis for the molecular basis of their activity.