DESIGNING NOVEL QUINAZOLINE LIGANDS AS POTENTIAL EGFR-TARGETED ANTICANCER AGENTS: AN INTEGRATED MOLECULAR DOCKING, MOLECULAR DYNAMICS AND ADMET STUDY

Abstract

Background: Quinazoline is an established scaffold for a number of tyrosine kinase inhibitors that target the epidermal growth factor receptor (EGFR), which is a major therapeutic target in non-small cell lung cancer (NSCLC). The limitations of conventional drug development methods and the continuous push toward acquired resistance continue to the rational computational design of novel chemotypes and analogues. Objective: The goal of this study was to design and computationally evaluate twelve novel quinazoline derivatives (QZ-01 – QZ-12) against EGFR by an integrated structure-based workflow. Methods: The crystal structure of the EGFR (PDB 7AEI) was prepared for docking. The designed derivatives were screened with AutoDock Vina (PyRx) and were compared with erlotinib, gefitinib and afatinib. The validation of docking was undertaken by re-docking of the co-crystallized ligand and then by carrying out 100-ns molecular dynamics (MD) simulations of six selected derivatives and three reference inhibitors. The behaviour of RMSD, RMSF and hydrogen bond were evaluated and the physicochemical and Lipinski properties and predicted ADMET properties were used for the final prioritisation. The docking scores of the 12 designed derivatives varied from −9.80 to −8.60 kcal/mol. QZ-10 was the leading compound (−9. kcal/mol), followed by QZ-06 (−9.60 kcal/mol) and QZ-12 (−9.50 kcal/mol). All designed compounds had better docking scores than any of the three docked compounds in the reported set. When re-docked, the RMSD was 1.18 Å, which is lower than the 2.0 Å cutoff value. The designed compounds were compared based on the mean protein RMSD (QZ-10: 0.119 nm) and mean hydrogen-bond value (QZ-10: 4.5).The designed compounds were compared by the mean protein RMSD (QZ-10: 0.119 nm) and the mean hydrogen-bond value (QZ-10: 4.5) after the 100-ns MD simulation. None of the 12 derivatives were predicted to have any Lipinski violation, high gastrointestinal absorption, low blood–brain barrier penetration, non-major toxicity alert, and low CYP risk. The following ranks from the prioritization were: integrated 1, 2 and 3 were QZ-10, QZ-06 and QZ-12 respectively. Conclusion: It was concluded that QZ-10 is the best computational lead and QZ-06 and QZ-12 are strong contenders for experimental exploration. Since the study is computational, the results are not proof of anticancer activity, but hypotheses of priority