Combining molecular docking and QSAR studies for modelling the antigyrase activity of cyclothialidine derivatives

DNA gyrase is a well-established antibacterial target consisting of two subunits, GyrA and GyrB, in a heterodimer A2B2, where GyrB catalyzes the hydrolysis of ATP. Cyclothialidine (Ro 09-1437) has been considered as a promising inhibitor whose modifications might lead to more potent compounds against the enzyme. We report here for the first time, QSAR studies regarding to ATPase inhibitors of DNA Gyrase. 1D, 2D and 3D descriptors from DRAGON software were used on a set of 42 cyclothialidine derivatives. Based on the core of the cyclothialidine GR122222X, different conformations were created by using OMEGA. FRED was used to dock these conformers in the cavity of the GyrB subunit to select the best conformations, paying special attention to the 12-membered ring. Three QSAR models were developed considering the dimension of the descriptors. The models were robust, predictive and good in statistical significance, over 70% of the experimental variance was explained. Interpretability of the models was possible by extracting the SAR(s) encoded by these predictive models. Analyzing the compound–enzyme interactions of the complexes obtained by docking allowed us to increase the reliability of the information obtained for the QSAR models.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► Three QSAR models are made with cyclothialidine-like ATPase DNA GyrB inhibitors. ► 3D structures are created using GR1222222X core as reference. ► Molecular docking is applied to the dataset in the cavity of the enzyme. ► The best conformations achieved are used to calculate 3D descriptors for QSAR model. ► New inhibitors are designed with information from the QSAR and docking models.