Structure-Based Design and In silico Evaluation of Novel Donepezil Analogues as Potent Inhibitors of Human Acetylcholinesterase for Alzheimer's Disease Therapy

Abstract

Alzheimer's disease (AD) accounts for 60 to 70% of dementias worldwide. This condition is primarily associated with the decrease in acetylcholine (ACh) levels, due to the action of human acetylcholinesterase (AChE). The lack of a curative treatment for Alzheimer's disease underscores the urgency to develop options that are more effective. This study aims to design new analogues of donepezil optimized for the effective inhibition of human AChE. Using a physics-based computer-aided molecular design (CAMD) approach, we developed a reliable quantitative structure-activity relationship (QSAR) model utilizing the crystallographic structure of AChE (PDB code: 7E3H) along with a series of 45 molecules derived from donepezil, of which 32 were used for the training set and 13 for external validation. The QSAR model was rigorously validated with this external validation set, demonstrating a high predictive power. The established QSAR model ( ) explains 95% of the variation in experimental biological activity ( ) based on the variation in binding free energy (∆∆Gcom). The variation in binding free energy in the biological environment is given by the equation: . A detailed investigation into the binding mode of inhibitors at the human AChE active site, combined with the energetic assessment of individual residue interactions, enabled the rational selection of molecular substituents necessary for the design of new, optimized analogues. Using an intuitive substitution method, a series of 59 new, more effective analogues of donepezil (DPZA) were designed, notably: DPZA46 ( ), DPZA36 ( ), DPZA20 ( ). The best-designed DPZA analogues exhibit a favorable theoretical pharmacokinetic profile. The stability of the best-designed DPZA inhibitors and their complexes formed with AChE has been confirmed through molecular dynamics simulations, thereby validating the obtained active conformations. These results suggest that the designed analogues, particularly DPZA46, are promising candidates worthy of further experimental investigation as potential anti-Alzheimer agents.