Hayashi T, Itagaki H, Fukuda T, Tamura U, Kato S
Shiseido Production Research Center, 1050 Nippa-cho, Kohoku-ku, Yokohama 223, Japan.
Toxicol In Vitro. 1994 Apr;8(2):215-20. doi: 10.1016/0887-2333(94)90185-6.
Seven in vitro test systems used to predict eye irritancy (EYTEX, SIRC cytotoxicity, HeLa cytotoxicity, chorioallantoic membrane (CAM), liposome, red blood cell and haemoglobin denaturation test system) were applied to 12 surfactants, and the results were subjected to multivariate analysis to evaluate the relative contributions of five factors. These factors were: (1) cellular plasma membrane destruction factor, (2) haemoglobin type protein denaturation factor, (3) EYTEX-type protein denaturation factor, (4) cytotoxicity factor and (5) an unknown (unidentified) factor. The results clarified the basis on which the findings of each test system were related to the Draize results. According to the analysis, the Draize eye irritation test could be explained by the contribution of the protein denaturation factor and cellular plasma membrane destruction factor. This provides support for a previous hypothesis that the major mechanisms of eye irritation are cellular plasma membrane destruction and protein denaturation. The CAM test value showed a higher correlation coefficient (0.906) with the Draize score than did the results of any of the other test systems; this was due to the fact that these two tests showed similar patterns of dependence on the five factors as indicated by the factorial analysis. The haemoglobin denaturation test system had the next highest correlation coefficient at about 0.75. Furthermore, by using further tests to make up the deficiency of the other necessary factors, a desirable battery system could be predicted; for example, the combination of the haemoglobin-denaturation and rat red blood cell tests would provide a result similar to that of the Draize test. This study should contribute to the development of a rational basis for prediction of eye irritancy of chemicals using in vitro test systems.
将七种用于预测眼刺激性的体外测试系统（EYTEX、SIRC细胞毒性、HeLa细胞毒性、绒毛尿囊膜（CAM）、脂质体、红细胞和血红蛋白变性测试系统）应用于12种表面活性剂，并对结果进行多变量分析，以评估五个因素的相对贡献。这些因素为：（1）细胞质膜破坏因子，（2）血红蛋白型蛋白质变性因子，（3）EYTEX型蛋白质变性因子，（4）细胞毒性因子和（5）一个未知（未识别）因子。结果阐明了每个测试系统的结果与Draize结果相关的依据。根据分析，Draize眼刺激试验可以通过蛋白质变性因子和细胞质膜破坏因子的贡献来解释。这为先前的一个假设提供了支持，即眼刺激的主要机制是细胞质膜破坏和蛋白质变性。CAM测试值与Draize评分的相关系数（0.906）高于其他任何测试系统的结果；这是因为这两种测试在因子分析所示的对五个因素的依赖模式上相似。血红蛋白变性测试系统的相关系数次之，约为0.75。此外，通过使用进一步的测试来弥补其他必要因素的不足，可以预测出理想的组合测试系统；例如，血红蛋白变性测试和大鼠红细胞测试的组合将提供与Draize测试相似的结果。这项研究应有助于为使用体外测试系统预测化学品的眼刺激性建立合理的基础。
