Multiscale computational modeling integrated with in vitro evaluation of green-synthesized...

Glioblastoma multiforme (GBM) is a highly aggressive brain tumour with limited therapeutic options, largely owing to the poor blood-brain barrier (BBB) permeability of current drugs. Quinazolinone derivatives represent an important class of heterocycles with diverse pharmacological potential; however, their activity against U87 glioblastoma cells has not been previously reported. In this study, a series of 2,3-dihydroquinazolin-4(1 H)-one derivatives (DHQs) were synthesised through a sustainable PEG-400-mediated multicomponent protocol performed in a sealed tube, providing an efficient and environmentally benign route to access this pharmacologically important scaffold. To identify potential glioblastoma inhibitors, a multiscale computational pipeline integrating DFT descriptors, ADME screening, molecular docking, molecular dynamics (MD) simulations, MM/GBSA, principal component analysis (PCA), and free energy landscape (FEL) calculations was employed. Among the synthesised molecules, compound 1h emerged as the most promising candidate, exhibiting the highest binding affinity towards EGFR (-9.11 kcal mol⁻¹) and favourable CNS-relevant physicochemical properties. MD simulations confirmed the structural stability of the 1h-3POZ complex for over 100 ns, as supported by low RMSD values, restricted residue fluctuations, and a stable free-energy profile. Experimental validation using the MTT assay on U87 glioblastoma cells demonstrated that compound 1h exhibited potent cytotoxicity (IC₅₀ = 16.57 ± 0.90μM), significantly outperforming temozolomide. Overall, this study presents the first integrated green-synthetic and computational-experimental evaluation of DHQs against U87 cells, highlighting compound 1h as a promising lead for glioblastoma drug discovery.