Toxicology in the 21st Century Program (Tox21) Genomic Toxicology

Tox21计划是来自NIH的研究人员之间的联邦合作，包括NCATS和国家环境健康科学研究所(NIEHS)、环境保护局(EPA)和食品和药物管理局(FDA)的国家毒理学计划(NTP)。这些机构共同努力，推进体外毒理学测试。Tox21计划由以下专业组成：基因组毒理学、系统毒理学和计算毒理学。基因组毒理学团队已经开发出高通量的基因表达技术，目标是每年从涉及大量人类基因的数十万个样本中产生数据。RNA-SEQ技术在独特的固态384孔板格式中进行了优化和强大的自动化。实验成功地在384个样本中的每个样本中对代表347个细胞应激基因的1041个分析进行了多路复用，这些分析又被汇集/多路用于单个NextGen测序反应，该反应包括大约2亿个序列。对NCATS RASL-Seq平台的全面评估显示，该平台具有良好的重复性(平均变异系数为10-11%)、高通量(样本：一周内6×384孔板)、与定量聚合酶链式反应(QPCR)90%的符合率、中等成本(&lt；12美元/样本)。利用RASL-seq平台分析了61种线粒体膜调节剂、19种烟草成分和19种抗病毒核苷类似物的剂量效应。数据分析被编写成一系列自动化步骤或管道，用于建立RASL-SEQ数据分析和RNA-SEQ全基因组基因表达数据分析。 Tox21基因组毒理学团队还改进了肝细胞、神经元和血管内皮细胞的细胞模型。Tox21第二阶段包括对在初级定量高通量筛选中发现的活性化合物的详细分析。二次筛选将需要改进的细胞模型，通常是永生化细胞或干细胞，在体外分化为有功能的肝细胞、神经元和心肌细胞。在神经元毒性方面，该团队比较了SH-SY5Y神经母细胞瘤细胞、LUHMES条件永生化人脑多巴胺能神经元和神经干细胞作为神经元毒性的模型。他们发现，LUHMES细胞对大多数化合物最敏感，最令人惊讶的是，从LUHMES细胞分化而来的神经元对11种毒物比亲本未分化的细胞系更敏感。 最后，Tox21基因组毒理学团队使用诱导多能干细胞(IPSC)技术来产生盘中病模型。例如，在烟草项目中，RASL-seq和RNA-seq都被用来帮助团队了解烟草成分对血管内皮细胞的影响。通过使用IPSC从患有和不患有血管疾病的吸烟者身上培养这些细胞，该团队能够确定一些患者是否在遗传上对这种疾病敏感或耐药。Tox21基因组毒理学团队发现，IPSC来源的内皮细胞(IECS)实际上对烟雾的反应与来自三个血管床的原代内皮细胞非常相似，验证了IEC模型。他们重复了一项RNA-SEQ实验，该实验确立了四个个体之间的遗传/表观遗传差异控制着对烟草烟雾的不同反应。这项实验利用了来自四个个体的12个IPSC来源的内皮细胞系，发现每个个体的三个细胞系具有非常高的重复性。他们还评估了19种烟草成分对人脐静脉内皮细胞和正常内皮细胞的剂量依赖性细胞毒性和基因表达反应。

The Tox21 program is a federal collaboration among researchers from NIH, including NCATS and the National Toxicology Program (NTP) at the National Institute of Environmental Health Sciences (NIEHS), the Environmental Protection Agency (EPA), and the Food and Drug Administration (FDA). These agencies work together to advance in vitro toxicological testing. The Tox21 program is comprised of the following specialties: Genomic Toxicology, Systems Toxicology, and Computational Toxicology. The Genomic Toxicology team has developed high-throughput gene expression technologies with the goal of annually generating data from hundreds of thousands of samples across a vast number of human genes. RNA-seq technology was optimized and robustly automated in the unique solid-state 384-well plate format. Experiments successfully multiplexed 1,041 assays representing 347 cellular stress genes in each of 384 samples, which were in turn pooled/multiplexed for a single NextGen sequencing reaction comprising approximately 200 million sequences. A thorough evaluation of the NCATS RASL-Seq platform showed excellent reproducibility (average 10-11% CV), high-throughput (samples: 6 x 384-well plates in one week), 90% concordance with qPCR, and moderate cost (<$12/sample). The RASL-seq platform was used to analyze dose-responses for 61 mitochondrial membrane modulators, 19 tobacco components and 19 antiviral nucleoside analog drugs. Data analyses were scripted into a series of automated steps, or pipeline for both RASL-seq data analyses and RNA-seq whole-genome gene expression data analyses established. The Tox21 Genomic Toxicology team also has improved cellular models for hepatocytes, neurons and vascular endothelial cells. Tox21 Phase II includes detailed analysis of compounds that are found to be active in primary quantitative high-throughput screening. This secondary screening will require improved cellular models, typically immortalized cells, or stem cells, which are differentiated in vitro into functional hepatocytes, neurons and cardiomyocytes. For neuronal toxicity, the team compared SH-SY5Y neuroblastoma cells, LUHMES conditionally immortalized dopaminergic human neurons and neuronal stem cells as models of neuronal toxicity. They found that LUHMES cells are most sensitive to most of these compounds, and most surprising, that neurons differentiated from LUHMES cells were more sensitive to 11 toxicants that the parent undifferentiated cell line. Lastly, the Tox21 Genomic Toxicology team has used induced pluripotent stem cell (iPSC) technology to generate disease-in-a-dish models. For example, in the tobacco project, both RASL-seq and RNA-seq were used to help the team understand the effects of tobacco components on vascular endothelial cells. By producing these cells using iPSC from smokers with and without vascular disease, the team was able to determine whether some patients are genetically susceptible or resistant to the disease. The Tox21 Genomic Toxicology team discovered that iPSC-derived endothelial cells (iECs) in fact responded to smoke very similarly to primary endothelial cells from three vascular beds, validating the iEC model. They repeated an RNA-seq experiment that establishes that genetic/epigenetic differences between four individuals govern distinct responses to tobacco smoke. This experiment utilized 12 iPSC-derived endothelial cell lines from the four individuals and found that the three cell lines from each individual were very reproducible. They also assessed the dose-dependent cytotoxicity and gene expression responses to 19 selected tobacco components on human umbilical vein endothelial cells as well as normal iECs.