Targeting cathepsin B activity by natural inhibitors: a structural dynamics and network pharmacology approach for amyloid-beta dysregulation in Alzheimer's disease.

Alzheimer's disease (AD) is characterized by progressive cognitive decline and pathological accumulation of amyloid-beta (Aβ). Cathepsin B (CatB) has been implicated in both Aβ degradation and amyloidogenic APP processing; this compartment-dependent dualism makes CatB a complex but intriguing target, where selective inhibition of pathogenic activities (rather than global inhibition) is likely required for therapeutic benefit. In this research, we used an integrated computational pipeline to discover effective CatB inhibitors from a natural product-like compound library. A tiered virtual screening methodology (HTVS, SP, XP) was complemented by MMGBSA rescoring to yield F3382-3724, F6617-5583, and F6617-3074 as strong candidates. DFT B3LYP-D3/6-31G optimization confirmed F6617-5583 to possess the most stable electronic structure, followed by F3382-3724. Molecular dynamics simulations of 500 ns indicated that all complexes exhibited protein RMSD values below 2.0 Å, with F6617-5583 achieving conformational stability earlier, showing ligand RMSD values between 3.0-3.5 Å similar to the reference inhibitor. MMGBSA calculations identified F6617-5583 as the most potent binder (ΔG_bind = - 74.92 ± 3.10 kcal/mol), primarily stabilized by van der Waals and lipophilic interactions. Consistent interactions involving catalytic residues such as Trp30 and Trp221 facilitated stable ligand retention. PCA and Free Energy Landscape analyses further supported the localization of F6617-5583 in a low-energy conformational pocket indicative of strong and specific binding. A compound-gene interaction network constructed using NetworkX and Matplotlib revealed distinct connectivity patterns, providing insights into potential polypharmacological effects. Collectively, these results establish F6617-5583 as the lead CatB inhibitor, while F3382-3724 remains a promising secondary candidate for further optimization in Alzheimer's disease therapeutics.