Insights into osteoporosis triggered by common antiepileptic drugs via network toxicology, molecular docking and molecular dynamics simulations

Background Antiepileptic drugs (AEDs), first-line treatments for epilepsy and related seizure disorders, are increasingly recognized as important risk factors for osteoporosis. However, the molecular mechanisms and specific targets underlying AED-induced bone loss remain largely unclear. Methods We explored these mechanisms using an integrated strategy combining network toxicology, molecular docking, molecular dynamics simulations, and in vitro experiments. Common targets of four AEDs and osteoporosis were identified through database screening, followed by protein-protein interaction network construction, hub gene identification, and functional enrichment analysis. Molecular docking was used to evaluate the predicted binding affinity between AEDs and primary hub targets, and 100-ns molecular dynamics simulations were applied to assess dynamic binding stability. Furthermore, colony formation assays and Western blots were performed in MC3T3-E1 pre-osteoblasts to support the computational predictions. Results We identified 47, 50, 46, and 48 potential osteoporosis-related targets for carbamazepine phenobarbital phenytoin (PHT), and valproic acid respectively. These targets were mainly enriched in key signaling pathways and cellular processes associated with bone metabolism. Topological analysis identified four primary hub targets: epidermal growth factor receptor, CASP3, peroxisome proliferator-activated receptor gamma, and PTGS2. AEDs and their targets exhibited favorable binding affinity in the molecular docking simulations (binding energy < −5 kcal/mol). Molecular dynamics simulations revealed heterogeneous dynamic stability among the complexes: PB-CASP3 and PHT-PPARG reached relatively stable conformational equilibrium, whereas CBZ-EGFR displayed notable protein residue flexibility in the equilibrated trajectory, and VPA-PTGS2 sustained persistent structural fluctuations during the simulation period. In vitro experiments demonstrated that all AEDs inhibited the clonogenic capacity of MC3T3-E1 cells, and Western blot results further indicated that AEDs differentially modulated the expression and activation levels of their corresponding primary hub targets, which was consistent with the computational predictions. Conclusion This study constructs an integrated research framework combining computational prediction and in vitro validation to explore potential molecular mechanisms underlying AED-induced osteoblast dysfunction and osteoporosis. The findings highlight candidate targets and provide a preliminary mechanistic basis for future rational clinical medication and skeletal protection in patients receiving long-term AED therapy.