Deciphering the action of indole analogues on selected targets for neuroprotection in Alzheimer’s disease: network pharmacology, in silico, and in vitro approaches

Abstract

Alzheimer’s disease is the leading cause of cognitive decline worldwide, and the lack of effective disease-modifying treatments has increased interest in synthetic small-molecule scaffolds as alternative therapeutic options for AD. This study combined network pharmacology, molecular docking, molecular dynamics simulations, and in vitro experimental validation to evaluate a series of indole-based analogs as candidates for treating Alzheimer’s disease. We constructed protein–protein interaction networks centered on Alzheimer’s-related targets using STRING and Cytoscape. Hub-gene analysis prioritized GSK3B, STAT3, and MMP2 for downstream investigation. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analyses were performed with the ClueGO plug-in in Cytoscape to identify key pathways involved in disease progression. Binding affinities between the indole analogs and target proteins were assessed by molecular docking using Molsoft ICM-Pro, complemented by predictions of ADMET properties and blood–brain barrier permeability. The two selected compounds and the native ligand were subjected to single-replica molecular dynamics simulations using CHARMM-GUI and NAMD to evaluate the stability of each protein–ligand complex. For biological validation, we tested five indole-based analogs at multiple concentrations in N2a neuroblastoma cells using MTT-based cellular assays under normoxic, hypoxic, and glutamate-induced injury conditions, followed by western blot assessment of GSK-3β phosphorylation. Under normoxia, the MTT-derived IC₅₀ reflects intrinsic cytotoxicity and does not measure neuroprotection; in the injury models, the reported values are half-maximal protective concentrations. Among the tested compounds, A13 showed the most consistent profile across molecular docking, stable molecular dynamics behavior, favorable MM/GBSA binding free energy (−8.42 kcal/mol), acceptable ADMET properties, predicted blood–brain barrier permeability, and neuroprotective activity. In the glutamate-induced injury model, A13 and A33 showed half-maximal protective concentrations of 1.87 and 2.28 μM, respectively. Western blot analysis suggested treatment-associated modulation of GSK-3β Ser9 phosphorylation. By combining computational and experimental strategies, this study identified A13 as a candidate GSK-3β-modulating neuroprotective lead rather than a confirmed GSK-3β inhibitor, supported by convergent structural and cellular evidence that warrants further mechanistic validation and continued development as a therapeutic candidate for Alzheimer’s disease.

Keywords: Alzheimer’s disease, indole analogs, network pharmacology, molecular docking, molecular dynamics simulation