Objective: Sodium-glucose cotransporter 2 (SGLT2) inhibitor had been evaluated in clinical trials as the basic strategy of hyperglycemia handling in diabetes. However, because of SGLT2 inhibitors is the new class of oral antidiabetic, it is rare to be found in Indonesia, and it is costly.This study was intended to find compounds from Indonesian herbal database that show capability to be used as SGLT2 inhibitors through a pharmacophore-based virtual screening approach. Methods: The SGLT2 inhibitor pharmacophore models were made from 10 training sets of SGLT2 ligand inhibitors using the Ligand Scout 4.1.5. Ten pharmacophore models which had been made were validated using test set and decoy set methods to know how the performance of pharmacophore model worked. Virtual screening were then applied to the best pharmacophore model. Results: The model-1 pharmacophore was the best model, with values of 0.9080, EF1% = 56.5, EF5% = 56.5 and AUC100% = 0.87 which served as model for virtual screening. Model-1 consisted of one hydrophobic interaction, one aromatic ring, four hydrogen bond donors and five hydrogen bond acceptors. Virtual screening showed three compounds (Hits) with best pharmacophore fit scores according to model-1 among 1377 compounds, they were vitexin = 113.62; cucumerin A = 112.62; and cucumerin B = 113.51.
- SGLT2 Inhibitor
- Virtual Screening