In silico Analysis of AMP-activated Protein Kinase and Ligand-based Virtual Screening for Identification of Novel AMPK Activators

Background: Adenosine-Monophosphate-Activated protein kinase (AMPK) is a conserved kinase that plays an important role in maintaining the homeostasis of cells. AMPK activation has a positive impact on treatment of diseases such as diabetes, obesity and cancer as well. This observation led to the development of AMPK activators. Certain naturally occurring compounds have also been known to activate AMPK. Methods: In this study, we retrieved the AMPK activators that include chemical drugs, xenobiotics and natural compounds and analyzed their interactions with AMPK via docking studies. Using this ligand dataset, a pharmacophore model was generated based upon ligand-based pharmacophore modeling strategy. The generated pharmacophore model was used to screen a library of ZINC database. The new hits which share the properties of our pharmacophore model were further analyzed via docking studies. Results: This study led to the identification of new chemical compounds which has the potential to activate AMPK. Even some of the screened hits showed better binding energies as compared to that of the ligand dataset used thus having the potential to activate AMPK more efficiently. The promising hits obtained after virtual screening of ZINC database were also checked against the Lipinski's rule of five. Conclusion: Compound 7 out of the 10 compounds showed best binding energies even more efficient than the ligand dataset itself.