Minami Keiichi, Saito Toshiro, Narahara Masatoshi, Tomita Hiroyuki, Kato Hirokazu, Sugiyama Hisashi, Katoh Miki, Nakajima Miki, Yokoi Tsuyoshi
Drug Metabolism and Toxicology, Division of Pharmaceutical Sciences, Kanazawa University, Kanazawa, Japan.
Toxicol Sci. 2005 Sep;87(1):296-305. doi: 10.1093/toxsci/kfi235. Epub 2005 Jun 23.
In the field of gene expression analysis, DNA microarray technology has a major impact on many different areas including toxicogenomics, such as in predicting the adverse effects of new drug candidates and improving the process of risk assessment and safety evaluation. In this study, we investigated whether there is relationship between the hepatotoxic phenotypes and gene expression profiles of hepatotoxic chemicals measured by DNA microarray analyses. Sprague-Dawley rats (6 weeks old) were administered five hepatotoxicants: acetaminophen (APAP), bromobenzene, carbon tetrachloride, dimethylnitrosamine, and thioacetamide. Serum biochemical markers for liver toxicity were measured to estimate the maximal toxic time of each chemical. Hepatic mRNA was isolated, and the gene expression profiles were analyzed by DNA microarray containing 1,097 drug response genes, such as cytochrome P450s, other phase I and phase II enzymes, nuclear receptors, signal transducers, and transporters. All the chemicals tested generated specific gene expression patterns. APAP was sorted to a different cluster from the other four chemicals. From the gene expression profiles and maximal toxic time estimated by serum biochemical markers, we identified 10 up-regulated genes and 10 down-regulated genes as potential markers of hepatotoxicity. By Quality-Threshold (QT) clustering analysis, we identified major up- and down-regulated expression patterns in each group. Interestingly, the average gene expression patterns from the QT clustering were correlated with the mean value profiles from the biochemical markers. Furthermore, this correlation was observed at any extent of hepatotoxicity. In this study, we identified 17 potential toxicity markers, and those expression profiles could estimate the maximal toxic time independently of the hepatotoxicity levels. This expression profile analysis could be one of the useful tools for evaluating a potential hepatotoxicant in the drug development process.
在基因表达分析领域，DNA微阵列技术对包括毒理基因组学在内的许多不同领域产生了重大影响，例如在预测新药候选物的不良反应以及改进风险评估和安全性评价过程方面。在本研究中，我们调查了通过DNA微阵列分析测得的肝毒性化学物质的肝毒性表型与基因表达谱之间是否存在关联。给6周龄的Sprague-Dawley大鼠施用五种肝毒性物质：对乙酰氨基酚（APAP）、溴苯、四氯化碳、二甲基亚硝胺和硫代乙酰胺。测量肝脏毒性的血清生化标志物以估计每种化学物质的最大毒性时间。分离肝脏mRNA，并通过包含1097个药物反应基因（如细胞色素P450、其他I相和II相酶、核受体、信号转导子和转运蛋白）的DNA微阵列分析基因表达谱。所有测试的化学物质均产生了特定的基因表达模式。APAP与其他四种化学物质归为不同的簇。根据基因表达谱和血清生化标志物估计的最大毒性时间，我们鉴定出10个上调基因和10个下调基因作为肝毒性的潜在标志物。通过质量阈值（QT）聚类分析，我们确定了每组中的主要上调和下调表达模式。有趣的是，QT聚类的平均基因表达模式与生化标志物的平均值谱相关。此外，在任何肝毒性程度下均观察到这种相关性。在本研究中，我们鉴定出17个潜在的毒性标志物，并且那些表达谱可以独立于肝毒性水平来估计最大毒性时间。这种表达谱分析可能是在药物开发过程中评估潜在肝毒性物质的有用工具之一。
