Design, 3D QSAR modeling and docking of TGF-β type I inhibitors to target cancer

Transforming growth factor-β (TGF-β) family members plays a vital role in regulating hormonal function, bone formation, tissue remodeling, and erythropoiesis, cell growth and apoptosis. TGF-β super-family members mediate signal transduction via serine/threonine kinase receptors located on the cell membrane. Variation in expression of the TGF-β type I and II receptors in the cancer cells compromise its tumor suppressor activities which direct to oncogenic functions. The present study was aimed to screen the potent TGF-β type I inhibitors through atom based 3D-QSAR and pharmacophore modelling. For this purpose, we have chosen known TGF-β type I inhibitors from the binding database. The PHASE module of Schrodinger identified the best Pharmacophore model which includes three hydrogen bond acceptors (A), one hydrophobic region (H), and one ring (R) as the structural features. The top pharmacophore model AAAHR.27 was chosen with the R2 value of 0.94 and validated externally using molecules of the test set. Moreover the AAAHR.27 model underwent virtual screening using the molecules from the NCI, ZINC and Maybridge database. The screened molecules were further filtered using molecular docking and ADME properties prediction. Additionally, the 7 lead molecules were compared with a newly identified compound "SB431542" (well known TGF-β type I receptor inhibitor) and Galunisertib, a small molecule inhibitor of TGF-β type I, under clinical development (phase II trials) using the docking score and other binding properties. Also a top scored screened molecule from our study has been compared and confirmed using molecular dynamic simulation studies. By this way, we have obtained 7 distinct drug-like TGF-β type I inhibitors which can be beneficial in suppressing cancers reported with up-regulation of TGF-β type I. This result highlights the guidelines for designing molecules with TGF-β Type I inhibitory properties.