Screening and Experimental Validation of Modified Gandou Decoction-Targeted Inhibitors for Alleviating AD Components via Network Pharmacology, Machine Learning, and Molecular Dynamics Simulation

Shixin Ye¹Shun Zhang¹Liangdong Zhang¹Guorong Peng²Ming Xie^1,3Xiongfeng Huang^4,5*Yousheng Hu^1,6*

• ¹Fuzhou Medical College, Nanchang University, Fuzhou, China
• ²Fuzhou Medical College, Nanchang University, fuhzou, China
• ³Teaching and Research Section of Anatomy, Basic Medical College, Fuzhou Medical College, Nanchang University, fuzhou, China
• ⁴Fuzhou Medical College of Nanchang University, Fuzhou, China
• ⁵Department of Postgraduate, Jiangxi University of Chinese Medicine, Nanchang, China
• ⁶Fuzhou Key Laboratory of Chronic Disease Research, Fuzhou Medical College, Nanchang University, fuzhou, China

Background Alzheimer's disease (AD) is a neurodegenerative disease characterized by abnormal accumulation of β-amyloid (Aβ) and hyperphosphorylation of the Tau protein. Currently, there is a lack of effective and safe therapeutic approaches. In Traditional Chinese medicine (TCM), Gandou Decoction has shown significant efficacy in improving cognitive decline and dementia-related symptoms, but its specific mechanism remains unclear. Methods This study systematically analyzed the active components and anti-AD mechanism of Modified Gandou Decoction (MGD) by integrating network pharmacology, machine learning, molecular docking, molecular dynamics (MD) simulation, and in vitro experimental validation. Obtain the components of Chinese medicines in MGD from TCMSP and screen them via ADMET; obtain AD targets by combining databases and select core targets through machine learning; verify their effects through various analyses and experiments. Results A total of 21 potential active molecules of MGD and 68 intersection targets were screened out. Among them, 8 core targets (EIF2AK2, PPARG, BACE1, ESR1, GSK3B, ACE, CASP3, MAPK14) were confirmed to be significantly associated with AD pathology by gene expression difference analysis (P ≤ 0.05). KEGG enrichment analysis showed that MGD mainly intervenes in the amyloid production pathway, the MAPK pathway, and the IL-17 pathway. Molecular docking demonstrated that the majority of the 21 potential active compounds exhibited strong binding affinities to the 8 core targets. Moreover, some potential active molecules exhibited better binding energy and similar binding modes compared with known inhibitors when binding to the corresponding target proteins. Molecular dynamics simulation showed that Alisol B, a potential active component of MGD, could stably bind to BACE1, EIF2AK2, and CASP3. In vitro cell experiments confirmed that Alisol B could significantly reverse okadaic acid-induced damage in SH-SY5Y cells (p < 0.001). Conclusion MGD exerts its anti-AD effect through its potential active component Alisol B, which binds to target proteins BACE1, EIF2AK2, and CASP3, and synergistically inhibits Aβ production, Tau phosphorylation, and neuroinflammatory processes through multiple pathways. This study provides a foundation for developing MGD-derived natural products for AD treatment, although the precise mechanisms require further experimental validation.