Mechanism-Based Evaluation System for Hepato- and Nephrotoxicity or Carcinogenicity Using Omics Technology
We have been developing a carcinogenicity prediction system based on gene expression profiles focusing on omics technology to enable mechanism-based evaluations of toxicity to reduce the numbers of animals and toxicological endpoints required by animal studies. Here, we report the development of a mechanism-based evaluation system focused on chemically induced hepato- and nephrotoxicity or hepatic and renal carcinogenicity using a gene expression analysis with a DNA microarray. As a case study, the mode-of-action (MoA)/adverse outcome pathway (AOP) was constructed from the gene expression profiles and histopathological findings of carbon tetrachloride and cisplatin for hepatotoxicity and nephrotoxicity, respectively. Consequently, we developed an advanced toxicity evaluation system for hepato- and nephrotoxicity or hepatic and renal carcinogenicity based on the toxicity mechanisms. We also developed a new prediction system named “CARCINOscreen®” for evaluating the carcinogenic potentials of chemicals using the gene expression profiles of liver and kidney tissues from rats after a 28-day repeated administration. The prediction system could predict the carcinogenicity potential of a training chemical set including carcinogens and non-carcinogens with an accuracy of more than 90%. The marker genes established in this study are promising for the development of new effective in vitro testing methods in the future.
KeywordsAdverse outcome pathway (AOP) Gene expression profiles Hepatotoxicity Nephrotoxicity Carcinogenicity CARCINOscreen®
Of the more than 80,000 chemicals in commerce, rigorous safety testing and risk assessment has been carried on relatively few. As an example, rodent carcinogenicity test data available for less than 1,000 compounds in the US National Toxicology Program database. The carcinogenicity of chemicals in our environment is an important health hazard to humans. Carcinogenicity studies using rodents have long been the standard for evaluating the carcinogenic potential of chemicals ; however, such studies are time-consuming, expensive, and require large numbers of experimental animals. Therefore, the carcinogenic potential of many important chemicals remains untested. In addition to the carcinogenic potential, the hepatotoxicity and nephrotoxicity of xenobiotics, which include classical drugs, herbal medicines, and chemical products, represents a significant cause of liver and kidney diseases [2, 3]. To evaluate hazards of a compound, various toxicity studies are needed, leading to problems such as a high cost and long test period in regulatory sciences. The test guideline known as the “repeated dose 28-day oral toxicity study in rodents” (TG 407) adopted by the Organization for Economic Co-operation and Development (OECD) is used mainly in Japan and Europe as a screening toxicity test. If an initial response, such as a change in a gene expression level associated with toxic effects, could be detected, a single animal study might be capable of predicting various toxicity endpoints, including long-term toxicity. Under these circumstances, the development of an efficient hazard assessment system for chemicals is needed. Moreover, the promotion of a “3Rs” policy and the development of promising in vitro alternative test methods, are both progressing in toxicological studies.
Omics technology, such as gene expression analyses, can be used effectively for the identification and prediction of hazards. Toxicogenomics has been established as a powerful tool for elucidating the mechanisms of chemical toxicity, such as carcinogenicity [4, 5, 6], hepatotoxicity [7, 8] and nephrotoxicity [9, 10]. However, numerous unknown pathways or gene networks that lead to toxicity exist. For a better understanding of adverse outcome pathways (AOPs) and the expansion of mode of action (MoA) applications, the elucidation of pathways/networks or biomarkers to detect or predict in vivo toxicity is needed.
We participated in a 5-year ARCH-Tox project conducted by the Ministry of Economy, Trade and Industry (METI) in Japan with the aim of developing a new testing approach that would enable the evaluation of multiple endpoints (hepatotoxicity/nephrotoxicity, carcinogenicity and neurotoxicity) in a single 28-day repeated dose toxicity study using sets of marker genes selected based on toxicity mechanism such as MoAs or AOPs. Mechanism-based analysis using omics technology is expected to reveal new MoAs or AOPs, leading to the development of new in vitro assays.
Chemicals, Animal Test and Microarray Analysis
Histopathological findings of liver (CCl4)
Histopathological findings of kidney (cisplatin)
Total RNA was extracted from the liver samples using QIAzol (Qiagen, Hilden, Germany) and the RNeasy Mini Kit or miRNeasy Mini Kit (Qiagen), in accordance with the manufacturer’s protocol. The quality of the RNA samples was examined using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), and undegraded RNA samples were used for the experiments; for this study, we used RNA samples with RIN values of > 7.0 as an index of the high purity and integrity of the RNA samples.
Microarray analysis was performed as described previously . Briefly, three types of custom arrays, Toxarray III ver.2 and Agilent Whole Rat Genome Microarrays 8 × 60 K Toxplus ver.1 and ver.2, and the gene-expression-based carcinogenicity prediction system CARCINOscreen® were used for the microarray analysis. Global normalization was applied to one-color microarray data using GeneSpring GX 10 (Agilent Technologies). Lowess normalization was applied to two-color microarray data using Feature Extraction Software 184.108.40.206 (Agilent Technologies). The signal log2 ratio of the administration group vs. the vehicle control group was calculated using the mean normalized signal intensity in each group. The pathway or functional analysis for the DNA microarray data was performed using Ingenuity Pathways Analysis (IPA) software (Qiagen).
AOP-Based Mechanism of Hepatotoxicity Suggested by Case Study with Carbon Tetrachloride
AOP-Based Mechanism of Nephrotoxicity Suggested by Case Study with Cisplatin
Detection System for Hepato- and Nephrotoxicity
Toxicological findings and number of detection genes for hepato- and nephrotoxicity
Prediction System for Hepatic and Renal Carcinogenicity: CARCINOscreen®
Carcinogenicity is one of the most serious toxic effects of chemicals, and highly accurate methods for predicting carcinogens are strongly desired for the assessment on human health. We previously developed a prediction system named “CARCINOscreen®” for evaluating the carcinogenic potentials of chemicals using the gene expression profiles of liver tissues from rats after a 28-day repeated dose toxicity study . The prediction formula was generated using a support vector machine with predictive genes selected from 68 training chemical datasets; a predictive score was then calculated to predict the carcinogenic potentials of the tested chemicals. To ensure the accuracy of the prediction system, the chemicals were divided into three groups (Groups 1 to 3) according to the resulting hepatic gene expression profiles, and a prediction formula was generated for each group. The prediction system was capable of predicting the carcinogenicity of the training carcinogens and the non-carcinogens with an accuracy of 92.9%–100%. The final prediction result was determined based on the maximum prediction value obtained with three independent prediction formulas to establish the CARCINOscreen®. The system was able to accurately predict carcinogenicity in rats in 94.1% of the 68 training chemicals . Furthermore, we attempted to develop a quantitative PCR (qPCR)-based system as an alternative to the microarray-based CARCINOscreen® . The prediction accuracies of the qPCR-based alternative for training- and validation-phase trials were 82.8% and 86.4%, respectively .
Recently, we reported a renal carcinogenicity prediction system to predict chemical carcinogenicity in rats; a 28-day repeated-dose test was performed using male Crl:CD (SD) rats with 12 carcinogens and 10 non-carcinogens as the training dataset and five carcinogens and five non-carcinogens as the validation dataset . In this prediction system, the prediction accuracies for the training and the validation datasets were calculated to be 100% and 90%, respectively, while 4-hydroxy-m-phenylenediammonium dichloride (AMIDOL), a known non-renal carcinogen, was judged as being positive. Among the predictive genes, Hamp and Ranbp1 are known to be important for cell growth and cell cycle regulation, which are important events in carcinogenesis. Given our current limited knowledge of the genes responsible for renal carcinogenesis, the identification of candidate genes for chemical-induced renal carcinogenicity using this gene expression-based prediction method represents a promising advance in renal carcinogen identification .
In hepatotoxicity and nephrotoxicity, marker genes can be selected based on toxicity mechanisms such as MoA or AOP, enabling a detection accuracy of more than 90% for five kinds of toxicity findings in both the liver and kidney. For carcinogenicity, the CARCINOscreen® system predicted the carcinogenic potential of a training compound set that included non-carcinogens with a more than 90% accuracy for the liver and kidney. Furthermore, we developed a qPCR-based prediction system as an alternative to the microarray-based CARCINOscreen® for rat liver carcinogenicity. The prediction performance of the qPCR-based CARCINOscreen®, as well as its user-friendliness and cost effectiveness, suggests that this method is promising for application in primary health hazard assessments. These results suggested that omics technology, such as gene expression analysis, can be used effectively for hazard identification and prediction. From now on, the application of urine and blood samples, which are non- or semi-invasive to animals, might be more important as a contribution to the 3Rs policy. Blood and urine samples are used in metabolomics and proteomics approaches with a high frequency, and these techniques may also be powerful tools for the identification of toxicity mechanisms and to resolve issues in which changes in gene expression levels are not always correlated with the phenotypes.
This study was supported by a grant from the Ministry of Economy, Trade and Industry, Japan (ARCH-Tox).
- 2.Robert CJ, Stephen MR (2015) Hepatotoxicity: toxic effects on the liver. In: Robert CJ, Stephen MR, Lawrence HL (eds) Principles of toxicology: environmental and industrial applications, 3rd edn. Wiley, New York, pp 125–138Google Scholar
- 3.Lawrence HL (2015) Nephrotoxicity: toxic responses of the kidney. In: Robert CJ, Stephen MR, Lawrence HL (eds) Principles of toxicology: environmental and industrial applications, 3rd edn. Wiley, New York, pp 139–155Google Scholar
- 5.Fielden MR, Adai A, Dunn RT 2nd, Olaharski A, Searfoss G, Sina J, Aubrecht J, Boitier E, Nioi P, Auerbach S, Jacobson-Kram D, Raghavan N, Yang Y, Kincaid A, Sherlock J, Chen SJ, Car B, Predictive Safety Testing Consortium, Carcinogenicity Working Group (2011) Development and evaluation of a genomic signature for the prediction and mechanistic assessment of nongenotoxic hepatocarcinogens in the rat. Toxicol Sci 124(1):54–74Google Scholar
- 6.Yudate HT, Kai T, Aoki M, Minowa Y, Yamada T, Kimura T, Ono A, Yamada H, Ohno Y, Urushidani T (2012) Identification of a novel set of biomarkers for evaluating phospholipidosis-inducing potential of compounds using rat liver microarray data measured 24-h after single dose administration. Toxicology 295(1–3):1–7CrossRefGoogle Scholar
- 9.Marin-Kuan M, Nestler S, Verguet C, Bezençon C, Piguet D, Mansourian R, Holzwarth J, Grigorov M, Delatour T, Mantle P, Cavin C, Schilter B (2006) A toxicogenomics approach to identify new plausible epigenetic mechanisms of ochratoxin a carcinogenicity in rat. Toxicol Sci 89:20–34CrossRefGoogle Scholar
- 11.Matsumoto H, Yakabe Y, Saito K, Sumida K, Sekijima M, Nakayama K, Miyaura H, Saito F, Otsuka M, Shirai T (2009) Discrimination of carcinogens by hepatic transcript profiling in rats following 28-day administration. Cancer Inform 13:253–269Google Scholar
- 22.Lewis BK, Brian GS (2004) Anatomy and physiological of the kidney. In: Jerry BH, Robin SG (eds) Toxicology of the kidney, 3rd edn. CRC Press, New York, pp 1–36Google Scholar
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.