Department of Chemistry, Isfahan University of Technology, Isfahan 84156-83111, Iran.
Chem Biol Drug Des. 2011 Oct;78(4):612-21. doi: 10.1111/j.1747-0285.2011.01183.x. Epub 2011 Aug 25.
As three-dimensional (3D) structure of the GABA(A) α5 was not determined, the crystal structure of 2Vl0E at 3.3 Å resolution which is a ligand-gated K(+) channel was used as a template in homology modeling, and the result was used in molecular dynamic simulation for obtaining its conformation in a water sphere. The resulted conformation of the receptor was used for docking with the most potent of imidazo[1,5-a][1,2,4]-triazolo[1,5-d][1,4,] benzodiazepines drugs to find out binding sites and consequently the types of the interaction between the drugs and receptor. The results showed that π-π interaction of the drugs with three phenylalanine and tyrosine residues plays an important role in determining the potency of the inhibitors. The obtained information relating to the binding sites of the receptor was utilized for docking all the drugs into the receptor and find out optimized conformation for each drug, used in structure-based quantitative structure-activity relationship (QSAR) model for calculation of descriptors. Then, selected descriptors were related to the binding affinity and selectivity of the drugs using multiple linear regression and least squares-support vector regression. Finally, the results of target- and ligand-based QSAR models were compared, resulted the superiority of the structure-based QSAR to the ligand-based model.
由于 GABA(A)α5 的三维结构尚未确定，因此在同源建模中使用了配体门控 K(+)通道的 2Vl0E 的晶体结构（分辨率为 3.3Å）作为模板，并将结果用于分子动力学模拟，以获得其在水球中的构象。所得受体构象用于与最有效的咪唑并[1,5-a][1,2,4]-三唑并[1,5-d][1,4,]苯并二氮杂*类药物对接，以找出结合位点，从而确定药物与受体之间的相互作用类型。结果表明，药物与三个苯丙氨酸和酪氨酸残基的π-π相互作用在确定抑制剂的效力方面起着重要作用。与受体结合位点相关的信息用于将所有药物对接入受体，并找出每种药物的最佳构象，用于基于结构的定量构效关系（QSAR）模型计算描述符。然后，使用多元线性回归和最小二乘支持向量回归将选定的描述符与药物的结合亲和力和选择性相关联。最后，比较了基于靶标和基于配体的 QSAR 模型的结果，结果表明基于结构的 QSAR 优于基于配体的模型。
