[Retracted] Two Myricetin‐Derived Flavonols from Morella rubra Leaves as Potent α‐Glucosidase Inhibitors and Structure‐Activity Relationship Study by Computational Chemistry

Diabetes mellitus (DM) is a chronic disease characterized by hyperglycemia, and oxidative stress is an important cause and therapeutic target of DM. Phytochemicals such as flavonols are important natural antioxidants that can be used for prevention and treatment of DM. In the present study, six flavonols were precisely prepared and structurally elucidated from Morella rubra leaves, which were screened based on antioxidant assays and α‐glucosidase inhibitory activities of different plant tissues. Myricetin‐3‐O‐(2″‐O‐galloyl)‐α‐L‐rhamnoside (2) and myricetin‐3‐O‐(4″‐O‐galloyl)‐α‐L‐rhamnoside (3) showed excellent α‐glucosidase inhibitory effects with IC50 values of 1.32 and 1.77 μM, respectively, which were hundredfold higher than those of positive control acarbose. Molecular docking simulation illustrated that the presence of galloyl group altered the binding orientation of flavonols, where it occupied the opening of the cavity pocket of α‐glucosidase along with Pi‐anion interaction with Glu304 and Pi‐Pi stacked with His279. Pi‐conjugations generated between galloyl moiety and key residues at the active site of α‐glucosidase reinforced the flavonol‐enzyme binding, which might explain the greatly increased activity of compounds 2 and 3. In addition, 26 flavonols were evaluated for systematic analysis of structure‐activity relationship (SAR) between flavonols and α‐glucosidase inhibitory activity. By using their pIC50 (‐log IC50) values, three‐dimensional quantitative SAR (3D‐QSAR) models were developed via comparative molecular field analysis (CoMFA) and comparative similarity index analysis (CoMSIA), both of which were validated to possess high accuracy and predictive power as indicated by the reasonable cross‐validated coefficient (q2) and non‐cross‐validated coefficient (r2) values. Through analyzing 3D contour maps of both CoMFA and CoMSIA models, QSAR results were in agreement with in vitro experimental data. Therefore, such results showed that the galloyl group in compounds 2 and 3 is crucial for interacting with key residues of α‐glucosidase and the established 3D‐QSAR models could provide valuable information for the prediction of flavonols with great antidiabetic potential.