QSAR, molecular docking and ADMET properties in silico studies of novel 4,5,6,7-tetrahydrobenzo[D]-thiazol-2-Yl derivatives derived from dimedone as potent anti-tumor agents through inhibition of C-Met receptor tyrosine kinase

Ossama Daoui; Souad Elkhattabi; Samir Chtita; Rachida Elkhalabi; Hsaine Zgou; Adil Touimi Benjelloun

doi:10.1016/j.heliyon.2021.e07463

QSAR, molecular docking and ADMET properties in silico studies of novel 4,5,6,7-tetrahydrobenzo[D]-thiazol-2-Yl derivatives derived from dimedone as potent anti-tumor agents through inhibition of C-Met receptor tyrosine kinase

Heliyon. 2021 Jul 3;7(7):e07463. doi: 10.1016/j.heliyon.2021.e07463. eCollection 2021 Jul.

Authors

Ossama Daoui¹, Souad Elkhattabi¹, Samir Chtita², Rachida Elkhalabi³, Hsaine Zgou⁴, Adil Touimi Benjelloun⁵

Affiliations

¹ Laboratory of Engineering, Systems and Applications, National School of Applied Sciences, Sidi Mohamed Ben Abdellah-Fez University, Fez, Morocco.
² Laboratory of Physical Chemistry of Materials, Faculty of Sciences Ben M'Sik, Hassan II University of Casablanca, Casablanca P.O. Box 7955, Morocco.
³ Laboratory of Applied Organic Chemistry, Faculty of Sciences and Technologies, Sidi Mohamed Ben Abdellah-Fez University, Fez, Morocco.
⁴ Polydisciplinary Faculty of Ouarzazate, Ibn Zohr University, Agadir, Morocco.
⁵ LIMAS, Faculty of Sciences Dhar El Mahraz, Sidi Mohamed Ben Abdallah University, Fez, Morocco.

Abstract

A quantitative structure-activity relationship (QSAR) study is performed on 48 novel 4,5,6,7-tetrahydrobenzo[D]-thiazol-2 derivatives as anticancer agents capable of inhibiting c-Met receptor tyrosine kinase. The present study is conducted using multiple linear regression, multiple nonlinear regression and artificial neural networks. Three QSAR models are developed after partitioning the database into two sets (training and test) via the k-means method. The obtained values of the correlation coefficients by the three developed QSAR models are 0.90, 0.91 and 0.92, respectively. The resulting models are validated by using the external validation, leave-one-out cross-validation, Y-randomization test, and applicability domain methods. Moreover, we evaluated the drug-likeness properties of seven selected molecules based on their observed high activity to inhibit the c-Met receptor. The results of the evaluation showed that three of the seven compounds present drug-like characteristics. In order to identify the important active sites for the inhibition of the c-Met receptor responsible for the development of cancer cell lines, the crystallized form of the Crizotinib-c-Met complex (PDB code: 2WGJ) is used. These sites are used as references in the molecular docking test of the three selected molecules to identify the most suitable molecule for use as a new c-Met inhibitor. A comparative study is conducted based on the evaluation of the predicted properties of ADMET in silico between the candidate molecule and the Crizotinib inhibitor. The comparison results show that the selected molecule can be used as new anticancer drug candidates.

Keywords: ADMET; C-Met inhibitors; Crizotinib; Molecular docking; QSAR.