Direct oral anticoagulants are an important and relatively new class of synthetic anticoagulant drugs commonly used for the pharmacotherapy of thromboembolic disorders. However, they still have some limitations and serious side effects, which continuously encourage medicinal chemists to search for new active compounds acting as human-activated coagulation factor X (FXa) inhibitors. Isosteviol is a nontoxic hydrolysis product of naturally occurring stevioside and possesses a wide range of therapeutic properties, including anticoagulant activity. The present contribution describes the in silico design of novel oxime ether isosteviol derivatives as well as a molecular modeling approach based on QSAR analysis and a docking simulation for searching for novel isosteviol-based compounds as potential FXa inhibitors. The elaborated ANN model, encompassing topological and geometrical information, exhibited a significant correlation with FXa-inhibitory activity. Moreover, the docking simulation indicated six of the most promising isosteviol-like compounds for further investigation. Analysis showed that the most promising derivatives contain heterocyclic, aromatic, five-membered moieties, with substituents containing chlorine or fluorine atoms. It is anticipated that the findings reported in the present work may provide useful information for designing effective FXa inhibitors as anticoagulant agents.
Keywords: FXa inhibitor; QSAR; artificial neural networks; docking simulation; isosteviol-like FXa inhibitors; molecular modeling; thrombosis.