Computational insights into taxifolin's therapeutic mechanisms for Alzheimer’s disease

Abstract

Background: Alzheimer's disease (AD) poses a significant challenge for research. Taxifolin is a natural flavonoid and has potential in protecting against Alzheimer’s by inhibiting oxidative stress. This study aims to elucidate the multi-target effects of Taxifolin on Alzheimer’s disease by computational techniques. Materials and Methods:  The target genes were mapped into a PPI network using the STRING database. Hub genes were identified using Cytoscape software. Gene Ontology and KEGG pathway analyses were performed to identify AD-related pathways. Finally, docking analysis was performed using CDOCKER and Autodock, and GROMACS molecular dynamics simulations were performed to study the ligand's behaviour. Results and Discussion: The results revealed promising drug-like properties, as assessed by Lipinski’s rule and ADMET predictions. A total of 47 human genes showed significant similarity (≥ 0.70), with 10,234 targets linked to Alzheimer’s disease. Of these, 673 genes were highly associated with the disease (GDA > 0.1). A Venn diagram identified 16 overlapping genes, including BACE1, DPP4, PIK3CA, MTOR, ESR2, and APP. Network analysis revealed interactions among Taxifolin, MTOR, SERPINE1, ESR2, PIK3CA, and NOS3. Taxifolin docking against Alzheimer’s targets identified 5T4B as the best hit, with the lowest binding energy in CDOCKER (−43.57 kcal/mol) and AutoDock Vina (−8.8 kcal/mol). A 100-ns MD simulation confirmed a stable 5T4B–taxifolin complex, showing structural stability (RMSD/RMSF), compactness (Rg), solvent exposure (SASA), and persistent hydrogen bonds. MM-PBSA analysis supported strong binding, primarily driven by van der Waals and electrostatic interactions. Conclusion: Hence, the study offers insights into the multi-target mechanisms of Taxifolin against AD.