羟基多溴二苯醚生物活性的QSAR研究

摘要

Quantitative structure-activity relationship (QSAR) models are developed to predict the bioactivities of hydroxylated polybrominated diphenyl ethers (OH-PBDEs). The bioactivities are inhibition of steroidogene-sis mediated by CYP19, hormone activity to thyroid receptor-β (TR-β) , and binding interaction with thyroxine - binding globulin (TBG) and transthyretin (TTR). The VSMP (variable selection and modeling based on the prediction) technique is used to select optimal subset from large-size molecular descriptors. Multiple linear regression ( MLR) method is employed to construct the QSAR models. The correlation coefficient of leave-one-out cross validation ( Q2) and fitting correlation coefficient of models ( R2) of four QSAR models are greater than 0. 95 and 0. 97 respectively. The statistics parameters show that overall models are robust and have good fitting and prediction for selecting bioactivities. The standard regression coefficients of models reveal that the 3D structure and aromaticity of OH-PBDEs are the most important factors for the bioactivities. In addition, the number and the position of bromine play an important role in the bioactivities of OH-PBDEs.%采用VSMP方法从大量的分子结构描述符中筛选最优子集,用多元线性回归方法分别构建了羟基多溴二苯醚(hydroxylated polybrominated diphenyl ethers,OH-PBDEs)对细胞色素CYP19介导类固醇生成的抑制活性,甲状腺受体-β(thyroid receptors,TR-β)的激素活性以及与甲状腺素结合球蛋白(thyroxine-binding globulin,TBG)和甲状腺转运蛋白(transthyretin,TTR)之间结合能力的定量构效关系(quantitative structure-activity relationship,QSAR)模型.模型的留一法交叉验证(leave-one-out cross validation,LOO-CV)相关系数Q2和拟合相关系数R2均高于0.95和0.97,表明模型具有良好的稳健性、拟合能力和预测能力.QSAR模型表明OH-PBDEs的分子三维结构和芳香性是其活性的重要影响因素.另外,分子中溴原子数量和取代位置对OH-PBDE生物活性具有重要影响.