Abstract: Aldose reductase is related to the occurrence of diabetic complication.The most commonly used anti-diabetes drugs often use this enzyme as the target,however,these drugs have a large number of toxic and side effects in their application.Therefore,search of safe aldose reductase inhibitors is one of the most important research fields in functional food and medical research.Previous reports showed that plant-derived flavonoids have strong aldose reductase inhibitory activity,however,the corresponding mechanism is still unclear.In this study,three-dimensional quantitative structure-activity relationship (3D-QSAR) of flavones inhibiting aldose reductase and their interaction mechanism were investigated by using molecular simulation methods.The 3D-QSAR models of 39 flavones as aldose reductase inhibitors were obtained by using the Topomer CoMFA method based on R-group search technology and the external prediction ability of this model was validated by a test set consisting of 12 samples.The multiple correlation coefficients of fitting,cross validation and external validation were 0.831,0.564 and 0.794,respectively.The molecular docking modes of flavones and aldose reductase were analyzed by the Surflex-dock method.The results showed that the different structures of flavones could lead to their different orientations in the hydrophobic cavity of aldose reductase,which resulted in their different performances in the inhibitory activity.When the distribution of substituent groups on the flavones could cater to the requirements of the stereo and electrostatic fields,the corresponding flavones could better interact with the hydrophobic cavity to improve its aldose reductase inhibitory activity.This study is helpful to develop the novel aldose reductase inhibitors and prompt the application of flavones in functional foods.