A transcriptome-based classifier to identify developmental toxicants by stem cell testing: design, validation, and optimization for histone deacetylase inhibitors

Test systems to identify developmental toxicants are urgently needed. A combination of human stem cell technology and transcriptome analysis was used here to provide proof-of-concept that toxicants with a related mode of action can be identified, and grouped for read-across. We chose a test system of developmental toxicity, related to the generation of neuroectoderm from pluripotent stem cells (UKN1), and exposed cells for six days to benchmark concentration (BMC) of histone deacetylase inhibitors (HDACi) valproic acid, trichostatin-A, vorinostat, belinostat, panobinostat and entinostat. To provide insight into their toxic action, we identified HDACi consensus genes, assigned them to superordinate biological processes, and mapped them to a human transcription factor network constructed from hundreds of transcriptome data sets. We also tested a heterogeneous group of ‘mercurials’ (methylmercury, thimerosal, mercury(II)chloride, mercury(II)bromide, 4-chloromercuribenzoic acid, phenylmercuric acid) (BMCs). Microarray data were compared at the highest non-cytotoxic concentration for all 12 toxicants. A support vector machine (SVM)-based classifier predicted all HDACi correctly. For validation, the classifier was applied to legacy data sets of HDACi, and for each exposure situation, the SVM predictions correlated with the developmental toxicity. Finally, optimization of the classifier based on 100 probe sets showed that eight genes (F2RL2, TFAP2B, EDNRA, FOXD3, SIX3, MT1E, ETS1, LHX2) are sufficient to separate HDACi from mercurials. Our data demonstrate, how human stem cells and transcriptome analysis can be combined for mechanistic grouping and prediction of toxicants. Extension of this concept to mechanisms beyond HDACi would allow prediction of human developmental toxicity hazard of unknown compounds with the UKN1 test system. We applied a test system (next referred as the UKN1 test system) of developmental toxicity, related to the generation of neuroectoderm from human embryonic stem cells (hESCs), and exposed cells for six days to the histone deacetylase inhibitors (HDACi) valproic acid, trichostatin-A, vorinostat, belinostat, panobinostat and entinostat, as well as to the mercurial compounds methylmercury, thimerosal, mercury(II)chloride, mercury(II)bromide, 4-chloromercuribenzoic acid and phenylmercuric acid. After 6 days, total RNA has been isolated and gene expression studies were performed applying the human Genome U133 plus 2.0 arrays. Untreated and vehicle controls (Ethanol and DMSO) have been also investigated. At least four independent experiments have been performed.

当前亟需可用于识别发育毒性物质的测试系统。本研究结合人类干细胞技术与转录组分析（transcriptome analysis），验证了可通过作用机制相似性识别有毒物质并进行交叉参照（read-across）分组的可行性。我们选取了与多能干细胞（pluripotent stem cells）分化生成神经外胚层相关的发育毒性测试系统UKN1，将细胞暴露于组蛋白去乙酰化酶抑制剂（HDACi）的基准浓度（BMC）下培养6天，受试抑制剂包括丙戊酸、曲古抑菌素A、伏立诺他、贝利司他、帕比司他与恩替诺特。为解析这些物质的毒性作用机制，我们鉴定了HDACi的共识基因，将其归类至上级生物学过程，并映射至由数百个转录组数据集构建的人类转录因子网络中。此外，我们还测试了一组异质性汞类化合物（甲基汞、硫柳汞、氯化汞(II)、溴化汞(II)、4-氯汞苯甲酸、苯基汞酸）的基准浓度（BMC）。我们在所有12种有毒物质的最高非细胞毒性浓度下，对其微阵列（microarray）数据进行了比对分析。基于支持向量机（SVM）的分类器可准确识别所有HDACi。为验证模型性能，我们将该分类器应用于HDACi的既往数据集，结果显示所有暴露场景下的SVM预测结果均与发育毒性程度相关。最终，基于100个探针组的分类器优化结果显示，仅需8个基因（F2RL2、TFAP2B、EDNRA、FOXD3、SIX3、MT1E、ETS1、LHX2）即可有效区分HDACi与汞类化合物。本研究数据证实了人类干细胞技术与转录组分析相结合，可用于有毒物质的机制分组与毒性预测。将该研究思路拓展至HDACi以外的作用机制，即可通过UKN1测试系统预测未知化合物对人类的发育毒性危害。本研究采用了基于人类胚胎干细胞（human embryonic stem cells, hESCs）分化生成神经外胚层的发育毒性测试系统（后文简称UKN1测试系统），将细胞分别暴露于HDACi类化合物（丙戊酸、曲古抑菌素A、伏立诺他、贝利司他、帕比司他、恩替诺特）以及汞类化合物（甲基汞、硫柳汞、氯化汞(II)、溴化汞(II)、4-氯汞苯甲酸、苯基汞酸）中培养6天。培养6天后，我们提取总RNA，并使用人类基因组U133 Plus 2.0芯片（human Genome U133 plus 2.0 arrays）开展基因表达分析。同时设置了未处理组与溶剂对照组（乙醇与二甲基亚砜，DMSO），所有实验至少重复4次独立实验。