Zhang Huabei, Li Hua, Ma Qinqin
Department of Chemistry, Beijing Normal University, 19# Street Xinjiekou, Beijing 100875, China.
J Mol Graph Model. 2007 Jul;26(1):226-35. doi: 10.1016/j.jmgm.2006.11.005. Epub 2006 Dec 8.
Extensive 3D-QSAR studies were performed on 158 diverse analogues of 3-pyridyl ethers, which are excellent ligands of alpha4beta2 neuronal nicotinic acetylcholine receptor (NnAChR). Comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) techniques were used to relate the binding affinities with the ligand structures. Two QSAR models were obtained using CoMFA and CoMSIA techniques. The two QSAR models were proved to be statistically significant and have high predictive power. The best CoMFA model yielded the cross-validated q(2)=0.605 and the non-cross-validated r(2)=0.862. The derived model indicated the importance of steric (85.9%) as well as electrostatic (14.1%) contributions. The CoMFA model demonstrated the steric field as the major descriptor of the ligand binding. The best CoMSIA model gave q(2)=0.723 and r(2)=0.685. This model showed that steric (30.3%) and H-bond interaction (61.8%) properties played major roles in ligand binding process. The squares of correlation coefficient for external test set of 28 molecules were 0.723 and 0.685 for the CoMFA model and the CoMSIA model, respectively. The two models were further graphically interpreted in terms of field contribution maps. SAR studies were also performed on different series of compounds in order to get a more reasonable understanding of the interactions between the ligands and the receptor. With the results, we have also presumed some assistant elements as supplements to the traditional pharmacophoric elements. A crude vision of ligand localization in the ligand-binding pocket of the receptor was also obtained, which would favor for the docking study of this kind of ligands.
对158种不同的3-吡啶基醚类似物进行了广泛的3D-QSAR研究，这些类似物是α4β2神经元烟碱型乙酰胆碱受体（NnAChR）的优秀配体。采用比较分子场分析（CoMFA）和比较分子相似性指数分析（CoMSIA）技术将结合亲和力与配体结构相关联。使用CoMFA和CoMSIA技术获得了两个QSAR模型。这两个QSAR模型经证明具有统计学意义且具有较高的预测能力。最佳的CoMFA模型得到交叉验证的q(2)=0.605，非交叉验证的r(2)=0.862。推导的模型表明空间（85.9%）以及静电（14.1%）贡献的重要性。CoMFA模型表明空间场是配体结合的主要描述符。最佳的CoMSIA模型给出q(2)=0.723和r(2)=0.685。该模型表明空间（30.3%）和氢键相互作用（61.8%）性质在配体结合过程中起主要作用。对于28个分子的外部测试集，CoMFA模型和CoMSIA模型的相关系数平方分别为0.723和0.685。根据场贡献图对这两个模型进行了进一步的图形解释。还对不同系列的化合物进行了SAR研究，以便更合理地理解配体与受体之间的相互作用。根据这些结果，我们还推测了一些辅助元素作为对传统药效基团元素的补充。还获得了受体配体结合口袋中配体定位的粗略视图，这将有利于此类配体的对接研究。
