Molecular modeling studies on phosphonic acid-containing thiazole derivatives: design for fructose-1,6-bisphosphatase inhibitors.

Presently, an in silico modeling was carried out on a series of 63 phosphonic acid-containing thiazole derivatives as fructose-1,6-bisphosphatase (FBPase) inhibitors using CoMFA/CoMSIA and molecular docking methods. The CoMFA and CoMSIA models using 51 molecules in the training set gave r (cv) (2) values of 0.675 and 0.619, r ( 2 ) values of 0.985 and 0.979, respectively. The systemic external validation indicated that our CoMFA and CoMSIA models possessed high predictive powers with r (0) (2) values of 0.995 and 0.994, r (m(test)) (2) values of 0.887 and 0.860, respectively. The 3D contour maps of the CoMFA and CoMSIA provided smooth and interpretable explanation of the structure-activity relationship for the inhibitors. Molecular docking studies revealed that a phosphonic group was essential for binding to the AMP binding site, and some key features were also identified. The analyses of the 3D contour plots and molecular docking results permitted interesting conclusions about the effects of different substituent groups at different positions of the common scaffold, which might guide the design of novel FBPase inhibitors with higher activity and bioavailability. A set of 60 new analogues were designed by utilizing the results revealed in the present study, and were predicted with significantly improved potencies in the developed models. The findings can be quite useful to aid the designing of new fructose-1,6-biphophatase inhibitors with improved biological response.