化学学报 ›› 2011, Vol. 69 ›› Issue (21): 2618-2626.DOI: 10.6023/A1101082B 上一篇    下一篇

研究论文

硝基芳烃对梨形四膜虫毒性的定量构效关系解析

陈学勇1,2,韦朝海*,1,邓秀琼1,夏芳1,于旭彪1   

  1. (1华南理工大学环境科学与工程学院 工业聚集区污染控制与生态修复教育部重点实验室 广州 510006)
    (2广东省韶关市环境保护科学技术研究所 韶关 512026)
  • 收稿日期:2011-01-08 修回日期:2011-06-12 出版日期:2011-11-14 发布日期:2011-07-12
  • 通讯作者: 韦朝海 E-mail:cechwei@scut.edu.cn
  • 基金资助:

    国家自然科学基金;“十一五”国家科技支撑计划重点项目

An Analysis of the Toxicity of Nitroaromatic Compounds to Tetrahymena pyriformis Using the Quantitative Structure Activity Relationship Method

Chen Xueyong1,2 Wei Chaohai*,1 Deng Xiuqiong1 Xia Fang1 Yu Xubiao1   

  1. (1 Colleage of Environmental Science and Engineering, South China University of Technology, The Key Lab of Pollution Control and Ecosystem Restoration in Industry Clusters of Ministry of Education, Guangzhou 510006)
    (2 Institute of Environmental Science and Technology of Shaoguan City, Shaoguan 512026)
  • Received:2011-01-08 Revised:2011-06-12 Online:2011-11-14 Published:2011-07-12

应用半经验方法AM1计算了42种硝基芳烃的19个量子化学和理化参数, 结合部分取自文献的分子结构符和梨形四膜虫的急性毒性数据, 采用偏最小二乘法回归分析建立了硝基芳烃对梨形四膜虫的毒性定量预测模型, 对该类化合物的急性毒性机理进行了讨论并对所建立模型与文献中相关报道的模型的评价指标进行了比较. 模型的内部交叉检验和外部测试检验结果均显示所获得模型具有较高的拟合精度和较好的预测效果(R2=0.946, =0.921, P=5.5×10-18, F=453.53, SD=0.171), 其评价指标优于文献中报道的其它模型结果. 模型变量重要性分析表明, ELUMO, a, , ET, MW, CSVMR对硝基芳烃的毒性大小有较大的制约作用, 硝基芳烃对梨形四膜虫所表现出来的毒性随着ELUMO, ET, △E的增大而减少, 随a, , , △Hf,, MW, CSVMR的增大而增大. 导致ET和△Hf对梨形四膜虫的毒性大小贡献的不同步, 被认为是由于硝基芳烃对梨形四膜虫多种致毒机制共存所造成.

关键词: 定量构效关系, 偏最小二乘法, 硝基芳烃, 梨形四膜虫, 毒性机理

The toxicity of nitroaromatic compounds to Tetrahymena pyriformis was studied by quantitative structure activity relationship (QSAR). The partial least squares approach was empolyed to establish a quantitative model which included 19 quantum-chemical and physical-chemical parameters calculated from 42 nitroaromatic compounds based on the semi-empirical AM1 method, and included the toxicity data of Tetrahymena pyriformis and part of molecular structures reported in literatures. The reliability of the model was evaluated and compared with other reported models. The toxicity of nitroaromatic compounds was also analyzed. Results showed that the model had a good predictive capability (R2=0.946, =0.921, P=5.5×10-18, F=453.53, SD=0.171), indicating higher accuracy achieved as compared to other models. The parameters of ELUMO, a, , ET, MW, CSV and MR played important roles in the toxicity inhibition of nitroaromatic compounds, among which ELUMO, ET and △E had negative correlations with the toxicity, while a, , , △Hf, MW, CSV and MR had positive correlations with the toxicity. The difference in the contribution to the Tetrahymena pyriformis toxicity by ET and △Hf is possibly caused by different types of mechanisms responsible for their toxicity to the nitroaromatic compounds.

Key words: quantitative structure activity relationship (QSAR), partial least squares (PLS), nitroaromatic compounds, Tetrahymena pyriformis, toxicity mechanism