Alzheimer’s disease (AD) is a complex neurodegenerative condition marked by a gradual decline in cognitive abilities, a reduction in acetylcholine (ACh) levels, and the accumulation of β-amyloid (Aβ) plaques. In a healthy brain, approximately 80% of ACh is broken down by acetylcholinesterase (AChE). Meanwhile, butyrylcholinesterase (BChE) serves a supportive function, gaining significance as AChE activity diminishes during the progression of Alzheimer’s disease. Modern therapeutic approaches focus on creating dual inhibitors of AChE and BChE that also aim to diminish Aβ-amyloidogenesis through interactions with the peripheral anionic site (PAS). This study combined fragment-based molecular design (CReM), AI-assisted retrosynthetic feasibility assessment, and in silico evaluation (docking, molecular dynamics, and ADMET profiling) to identify novel Donepezil derivatives as potential dual AChE/BChE inhibitors. A series of 10,000 derivatives were developed through computational methods and carefully assessed based on stringent drug-likeness, synthetic accessibility, and medicinal chemistry standards. This was succeeded by comprehensive ADMET profiling. Nine candidates were identified with predicted CNS pharmacokinetics, adequate toxicological profiles and reduced cytochrome P450 liabilities. Molecular docking yielded improved predicted binding affinities relative to Donepezil. Several derivatives, particularly D4 and D5, showed dual-site binding poses spanning both the catalytic gorge and the PAS. MD simulations indicated the stability of these poses over 100 ns. These computational results suggest that the proposed derivatives may preserve or improve upon Donepezil’s pharmacokinetic profile while offering potentially balanced AChE/BChE inhibition and anti-amyloidogenic activity, pending experimental validation.
