Homology modeling and 3D–QSAR study of benzhydrylpiperazine δ opioid receptor agonists. (December 2019)

• Record Type: Journal Article
• Authors: Pan, Chenling; Meng, Hao; Zhang, Shuqun; Zuo, Zhili; Shen, Yuehai; Wang, Liangliang; Chang, Kwen-Jen
• Is Part Of: Computational biology and chemistry. Volume 83(2019)
• Publication Date: 2019-12
• Subjects: δ opioid receptor agonists -- 3D-QSAR -- Homology modeling -- Molecular dynamics simulation -- Molecular docking; Chemistry -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; Biochemistry -- Data processing; Biology -- Data processing; Molecular biology -- Data processing; Periodicals; Electronic journals; 542.85
• DOI: 10.1016/j.compbiolchem.2019.107109
• Languages: English
• ISSNs: 1476-9271

Abstract

Graphical abstract: Highlights: A 3D-QSAR study of benzhydrylpiperazine δ opioid receptor agonists is reported. CoMFA and CoMSIA models with good statistical values were obtained. 3D structure of the active form of human δ opioid receptor is developed by homology modelling. Conformational changes of active human δ opioid receptor is analyzed by molecular dynamics simulations. The interaction between benzhydrylpiperazine agonists and human δ opioid receptor is evaluated by molecular docking. Abstract: The binding affinity of a series of benzhydrylpiperazine δ opioid receptor agonists were pooled and evaluated by using 3D-QSAR and homology modeling/molecular docking methods. Ligand-based CoMFA and CoMSIA 3D-QSAR analyses with 46 compounds were performed on benzhydrylpiperazine analogues by taking the most active compound BW373U86 as the template. The models were generated successfully with q 2 value of 0.508 and r 2 value of 0.964 for CoMFA, and q 2 value of 0.530 and r 2 value of 0.927 for CoMSIA. The predictive capabilities of the two models were validated on the test set with R 2 pred value of 0.720 and 0.814, respectively. The CoMSIA model appeared to work better in this case. A homology model of active form of δ opioid receptor was established by Swiss-Model using a reported crystal structure of active μ opioid receptor as a template, and was further optimized using nanosecond scale molecular dynamics simulation. The most active compound BW373U86 was docked to the active site of δ opioid receptor and the lowest energy binding pose was then used to identify binding residues such as s Gln105, Lys108, Leu125, Asp128, Tyr129, Leu200, Met132, Met199, Lys214, Trp274, Ile277, Ile304 and Tyr308. The docking and 3D-QSAR results showed that hydrogen bond and hydrophobic interactions played major roles in ligand-receptor interactions. Our results highlight that an approach combining structure-based homology modeling/molecular docking and ligand-based 3D-QSAR methods could be useful in designing of new opioid receptor agonists.