Identification of novel discoidin domain receptor 1 (DDR1) inhibitors using E-pharmacophore modeling, structure-based virtual screening, molecular dynamics simulation and MM-GBSA approaches. (March 2022)

• Record Type: Journal Article
• Authors: Nada, Hossam; Lee, Kyeong; Gotina, Lizaveta; Pae, Ae Nim; Elkamhawy, Ahmed
• Is Part Of: Computers in biology and medicine. Volume 142(2022)
• Publication Date: 2022-03
• Subjects: Novel discoidin domain receptor 1 (DDR1) -- DDR1 inhibitors -- E-Pharmacophore modeling -- Structure-based virtual screening (SBVS) -- Molecular dynamics (MD) simulation -- MM-GBSA study; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2022.105217
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Dysregulation of the discoidin domain receptor (DDR1), a collagen-activated receptor tyrosine kinase, has been linked to several human cancer diseases including non-small cell lung carcinoma (NSCLC), ovarian cancer, glioblastoma, and breast cancer, in addition to several inflammatory and neurological conditions. Although there are some selective DDR1 inhibitors that have been discovered during the last two decades, a combination of elevated cytotoxicity, kinome selectivity and/or poor DMPK profile has prevented more in-depth studies from being performed. As such, no DDR1 inhibitor has reached clinical investigation to date, forming an urgent need to develop specific DDR1 inhibitor(s) using various drug discovery means. However, the recent discovery of VU6015929, a potent and selective DDR1 kinase inhibitor, with enhanced physiochemical and DMPK properties in addition to its clean kinome profile marked a milestone in the development of DDR1 inhibitors. Herein, VU6015929 was used to construct a 3D e-pharmacophore model which was validated via calculating the difference of score between the active compounds and decoys. The validated e-pharmacophore model was then utilized to screen 20 million drug-like compounds obtained from the freely accessible Zinc database. The generated hits were ranked using high throughput virtual screening technique (HTVS), and the top 8 small molecules were subjected to a molecular docking study and MM-GBSA calculations. Protein-ligand complexes of compounds 1, 2, 3 and the standard compound (VU6015929) were performed for 100 ns and compared with the DDR1 unbound protein state and the DDR1 bound to a co-crystallized ligand. The molecular docking, MD and MM-GBSA outputs revealed compounds 1 –3 as potential DDR1 inhibitors, with compound 2 displaying superior binding affinity, comparable binding stability and average binding free energy for the ligand-enzyme complex compared to VU6015929. Highlights: A 3D e-pharmacophore model was constructed based on VU6015929 and validated. The e-pharmacophore model was used to screen 20 million drug-like compounds. HTVS, molecular docking, and MM-GBSA calculations were carried out. MD studies revealed compounds 1 –3 as potential DDR1 inhibitors. Compared to VU6015929, compound 2 displayed superior binding affinity.