Integrated computational analysis of halogenated curcumin derivatives targeting GSK-3β: Pharmacokinetic, docking, MDS, and DFT approaches

Glycogen synthase kinase-3β (GSK-3β) is a multifunctional serine/threonine kinase involved in several diseases, such as cancer, diabetes, and neurodegenerative disorders. Its inhibition has gained significant attention as a potential therapeutic approach. This study aimed to determine the inhibitory potential of halogenated curcumin derivatives against GSK-3β using comprehensive computational approaches. ADME analysis, molecular docking, density functional theory calculations, and a 100 ns molecular dynamics simulation were performed. Pharmacokinetic parameters assessed drug-likeness properties , while electronic reactivity of the compounds was explored using DFT calculations. Molecular docking and MD simulation were conducted to examine binding energies and stability within the GSK-3β active site. Both compounds showed favorable physicochemical properties, demonstrating high gastrointestinal absorption, favorable molecular mass, and zero Lipinski violations. None of the compounds were P-gp substrates, and only HC1 was able to cross the blood-brain barrier. Both compounds showed strong docking scores (HC2 –13.367, HC1 –11.034) and interaction energies (HC2 –78.049, HC1 –66.696), which were higher than those of the control inhibitors Elraglusib and Tideglusib. HC2 formed hydrogen bonds with LYS:585 and VAL:635, while HC1 formed one hydrogen bond with LYS:585. DFT calculations showed that HC2 was slightly reactive (ΔE = 3.537 eV) than HC1 (5.879 eV). In a 100 ns MD simulation time, HC2 maintained RMSD between 2.0–2.8 Å and HC1 between 1.5–3.5 Å, and minimal RMSF fluctuations in binding site residues, confirming stable protein-ligand interactions. HC2 demonstrated better binding energy, electronic reactivity, and conformational stability compared to HC1, suggesting it as a promising GSK-3β inhibitor. However, further experimental validation is necessary to validate its therapeutic efficacy due to the inherent limitations of computational analyses.