Search for potentially biased epidermal growth factor receptor (EGFR) inhibitors through pharmacophore modelling, molecular docking, and molecular dynamics (MD) simulation approaches

Abstract

Epidermal growth factor receptor (EGFR), being one of the most crucial receptor in cancer therapy, has been selected as a potential target for the present study. Ligand-based pharmacophore model (n = 30, R²=0.93 with root mean square deviation = 1.14, ΔCost = 144.27 and configuration cost = 21) was developed and validated with Fischer’s randomisation (at 95% confidence), test set (n = 225, R² pred = 0.81), external data set (n = 13, R² pred = 0.95) and decoy set (n = 70), further the model has been used to search for novel EGFR inhibitors. The validated model was used for virtual screening of zinc database. A pool of 115,948 candidate molecules was screened through the model. Subsequently, molecules having predicted IC₅₀<0.2 µM were selected for screening through drug-like properties filter. Based on pharmacokinetic profile (ADMET study), Lipinski’s rule of five and Veber’s rule, 62 molecules were shortlisted for molecular docking. Using consensus docking, five hit molecules were selected, which were further considered for molecular dynamics simulation. Additionally MM-GBSA analysis was carried which showed that affinity of hits towards the receptor of three compound mainly ZINC305, ZINC131796 and ZINC131785 were similar to the standard vanedtinib. The simulation, performed for 100 ns, revealed that two hit molecules, namely ZINC305 and ZINC131785, showing potential interactions at the ligand-binding domain of EGFR protein with good ligand–protein stability.

Communicated by Ramaswamy H. Sarma

Keywords:

• EGFR inhibitor
• Pharmacophore modelling
• virtual screening
• molecular docking
• molecular dynamics simulation

Acknowledgments

Megha Jethwa wants to acknowledge University of Calcutta for granting the financial assistance for work. Aditi Gangopadhyay acknowledges the Council of Scientific and Industrial Research (CSIR), New Delhi, India, for providing financial support. Financial support from SERB MATRICS is also gratefully acknowledged.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the SERB MATRICS project under Grant no. MTR/2019/000343, dated 11th February 2020.