Refining Molecular Docking Strategy for Aldose Reductase by Incorporating Specificity Pocket States and Methodological Evaluation

ABSTRACT Aldose reductase (AR) is a nicotinamide adenine dinucleotide phosphate (NADPH)‐dependent enzyme that catalyzes the reduction of aldehydes to alcohols and mediates the first step of the polyol pathway by converting glucose into sorbitol. The polyol pathway becomes excessively activated under hyperglycemic conditions contributing to diabetic long‐term complications. Although drug development targeting AR inhibition has been continuously pursued to prevent these complications, challenges remain due to the AR's induced‐fit mechanism and structural flexibility. In this study, we evaluated the binding affinity‐ and pose‐prediction accuracy of four docking methods—Vina, AutoDock4 (AD4), Protein‐Ligand ANTSystem (PLANTS), and CarsiDock—using Pearson correlation coefficient (r) and root‐mean‐square deviation (RMSD), respectively, and investigated how the specificity pocket state (open or closed) affects docking accuracy to identify optimal conditions for virtual screening targeting AR. The results demonstrated that AutoDock Vina outperformed other docking methods, and the open‐state structure 1PWL was identified as a suitable conformation capable of accommodating diverse binding behaviors of ligands effectively. These findings highlight the importance of selecting appropriate docking methods and structural conformations that account for the conformational diversity of binding pocket and suggest that the combination of Vina with the 1PWL structure provides a particularly effective strategy for AR‐targeted drug design.