APPLICATION OF HUMAN LIVER STEM CELLS FOR RECEPTOR-MEDIATED TOXICOGENOMIC STUDY By Suntae Kim A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Biochemistry and Molecular Biology - Environmental Toxicology 2011 ABSTRACT APPLICATION OF HUMAN LIVER STEM CELLS FOR RECEPTOR-MEDIATED TOXICOGENOMIC STUDY By Suntae Kim The high drug development failure rates, the large response discrepancy among test animal species to evaluate drug efficacy, and the lack of mechanism-based high-throughput screening recently prompted US and EU legislation recommend the development of reliable human in vitro models for toxicity testing to replace, reduce and refine animal testing. In vitro models are widely used, but the abnormality of continuous cell lines and restricted acquisition, source inconsistency and instability of primary cells limit their use. Adult stem cells derived from intact human tissue provides an innovative alternative that may more accurately predict in vivo toxicity. The objective of this research was to evaluate HL1-1 human hepatic stem cell line as a viable model for receptor-mediated toxicogenomic studies using the aryl hydrocarbon receptor (AhR) and peroxisome proliferator-activated receptor  (PPAR) as prototypical ligand activated transcription factors for comparative toxicogenomic investigation. AhR and PPAR are targets for environmental contaminants and pharmaceutical reagents and complementary species-specific hepatic responses are currently being studied. Comprehensive time course and dose-response gene expression studies were conducted in HL1-1 cells to assess the differential gene expression elicited by AhR and PPAR ligands in comparison to other in vitro and in vivo models. Some conserved responses with overlapping biological functions were identified. Although subsets of conserved differentially expressed genes are consistent with the known in vivo responses, the results suggest that species- and model-specific gene expression profiles are linked to species-specific physiology. HL1-1 cell were also immortalized by hTERT stable transfection (HLhT1) to overcome cell senescence and life span limitations. Immortalized HLhT1 cells maintain pluripotency characteristics as defined by stem cell and oval cell marker protein expression. The expression of functional AhR and PPAR and their ligand responsiveness is also comparable to the parental cell line. Collectively, human liver stem cells are viable models and warrant further development for mechanistic investigation of receptor-mediated hepatic toxicity and high throughput screening. To my wife and family iv ACKNOWLEDGMENTS I received an invaluable life experience while obtaining by PhD degree at Michigan State University. Without the help and support from numerous outstanding individuals, I could not have accomplished it. I would like to thank all people who have offered encouragement and support and inspired me throughout my graduation program. First, I would like to express my sincere and most heartfelt thanks to my advisor and mentor, Dr. Tim Zacharewski. He has provided me extraordinary opportunity to study in genomics, stem cell research and bioinformatics which are very advanced and promising field in biomedical research. He has been understanding and encouraging and has helped me to develop scientifically, intellectually and professionally. I could not complete this research without his tremendous supports. I would also like to thank my Graduate Committee members, Drs. Kristina Chan, John LaPres, James Pestka and Kevin Walker for their availability, insight and unique perspectives in contributing to the success of my project during our meetings. I would like to express my special thanks to Drs. CC Chang and Trosko and their lab for all supports and guidance in stem cell research with resources, techniques and advices. I would like to thank all current and previous my lab members, Ania Kopec, Michelle Angrish, Agnes Forgacs, Rance Nault, Qi Ding, Lyle Burgoon, Ed Dere, Naoki Kiyosawa, Josh Kwekel, Bryan Mets, Cora Fong, Jeremy Burt, Darrell Boverhof and many others, including cooperative education students for their friendship, advice and support inside and outside of the lab. v I also owe my deepest gratitude to everyone who provides me tremendous efforts to review and help editing my humble English writing and presentations. Also I would like send my gratitude to all Korean friends who work in biology fields and were met in Christian community at Michigan State University and University of Michigan. They are great fellows in both academically and socially, and make me and my family feel like at home country. Finally, I wish to thank my family for their love and support. My parents, Jongkook Kim and Kyungbok Lee, my wife’s parents, Youngsoo Bae and Kiok Ryu, and my brothers and their family gave me endless supports and encouragement for my study and taking care of our children. Thank you for encouraging me in the right direction and always supporting my choices. And to my wife Heekyong Bae who is outstanding researcher and the most faithful supporter, and my children, Ayn and Younghoo, you are the most precious gift I got through my PhD years. vi TABLE OF CONTENTS LIST OF TABLES ...................................................................................................................... iix LIST OF FIGURES ...................................................................................................................... x LIST OF ABBREVIATIONS ................................................................................................... xiii CHAPTER 1 Review of the Literature: Application of Human Stem Cell in Receptor Mediated Toxicity....................................................................................................................................... 2 Current Needs of Alternative Methods for Toxicity Tests ..................................................... 2 Toxicogenomic Approaches ................................................................................................... 4 Nuclear Receptors and Aryl Hydrocarbon Receptor Mediated Toxicity................................ 5 Current Models for Receptor Mediated Hepatotoxicity ......................................................... 9 Application of Human Stem cells for Toxicology ................................................................ 11 References ............................................................................................................................. 14 CHAPTER 2 Rationale, Hypothesis and Specific Aims.............................................................................. 20 Rationale ............................................................................................................................... 20 Hypothesis............................................................................................................................. 20 Specific Aims ........................................................................................................................ 21 References ............................................................................................................................. 23 CHAPTER 3 Evaluation of Human Adult Hepatic Stem Cells as an in vitro Receptor-Mediated Toxicogenomic Model and Their Cell Engineering for Immortalization .......................... 25 Abstract ................................................................................................................................. 25 Introduction ........................................................................................................................... 26 Materials & Methods ............................................................................................................ 30 Results ................................................................................................................................... 33 Discussion ............................................................................................................................. 48 References ............................................................................................................................. 55 vii CHAPTER 4 Comparative Analysis of AhR-Mediated TCDD Elicited Gene Expression in Human Liver Adult Stem Cells ........................................................................................................... 60 Abstract ................................................................................................................................. 60 Introduction ........................................................................................................................... 61 Materials & Methods ............................................................................................................ 63 Results ................................................................................................................................... 71 Discussion ............................................................................................................................. 99 References ........................................................................................................................... 108 CHAPTER 5 PPAR-mediated responses in human adult liver stem cells: in vivo/in vitro and crossspecies comparisons .............................................................................................................. 113 Abstract ............................................................................................................................... 113 Introduction ......................................................................................................................... 114 Materials & Methods .......................................................................................................... 116 Results ................................................................................................................................. 126 Discussion ........................................................................................................................... 143 References ........................................................................................................................... 165 CHAPTER 6 Conclusions and Future Research ....................................................................................... 172 Comparative Studies for Further Evaluation ...................................................................... 172 Model System Improvements and Further Application of Human Liver Stem Cells......... 173 Alternative Technologies for Toxicogenomics................................................................... 174 References ........................................................................................................................... 177 viii LIST OF TABLES Table 1. Online resources for toxicogenomic studies ................................................................................... 6 Table 2. Gene names and primer sequences for QRT-PCR ........................................................................ 34 Table 3. Gene names and primer sequences for QRT-PCR ........................................................................ 72 Table 4. Functional categorization of putative primary response genes elicited by TCDD ....................... 82 Table 5. HL1-1 vs. HepG2 Overlapping Genes: Functional Categories .................................................... 89 Table 6. HL1-1 vs. Hepa1c1c7 Overlapping Genes: Functional Categories ............................................ 102 Table 7. HL1-1 vs. Mouse Hepatic Tissue Overlapping Genes: Functional Categories .......................... 103 Table 8. Gene names and primer sequences for QRT-PCR ...................................................................... 121 Table 9. Over-represented gene ontology groups induced following PP treatment. BP: biological process, CC: Cellular component ..................................................................................................... 140 Table 10. Over-represented gene ontology groups repressed following PP treatment. BP: biological process, CC: Cellular component ..................................................................................................... 141 Table 11. PP elicited differentially expressed genes used for pathway mapping analysis ......................... 144 Table 12. Cytotoxicity of PPs in human cell lines ...................................................................................... 155 ix LIST OF FIGURES Figure 1. TCDD time course study design for cell line comparison. ........................................................... 29 Figure 2. Immortalization of HL1-1 cells. ................................................................................................... 36 Figure 3. Basal mRNA and protein expression levels of AhR, PPAR and , ER. .................................. 38 Figure 4. Proliferation potential of HL1-1 cell............................................................................................. 39 Figure 5. Morphology of HL1-1 and HLhT1 cells....................................................................................... 41 Figure 6. Basal stem cell marker and hepatocyte differentiation marker gene expression level comparison. ............................................................................................................................................... 43 Figure 7. Basal AhR and NRs expression levels comparison in parental and immortalized cell lines and receptors response in HLhT1 cells........................................................................................ 46 Figure 8. Temporal TCDD-mediated CYP1A1 induction profiles comparison. ......................................... 50 Figure 9. Temporal profiles of prototypical TCDD-responsive genes in human liver stem cell (HL1-1) and immortalized liver cells (HLhT1). ........................................................................................ 51 Figure 10. The proliferation potential and morphology of HL1-1 cell. ......................................................... 65 Figure 11. HL1-1 TCDD time course, dose response and cycloheximide (CHX) study designs. ................. 67 Figure 12. Microarray experimental designs. ................................................................................................. 68 Figure 13. Basal AhR mRNA and protein expression in HL1-1 cells. .......................................................... 73 x Figure 14. CYP1A1 expression induction by TCDD in HL1-1 cells. ............................................................ 74 Figure 15. Number of genes exhibiting differential expression changes in the TCDD (A) dose-response (12 hr) and (B) time course study (10 nM TCDD). .................................................................... 76 Figure 16. HL1-1 TCDD time course hierarchical clustering. ....................................................................... 77 Figure 17. QRT-PCR verification of CYP1B1, ALDH1A3 and SLC7A5 microarray results in the time course (10 nM TCDD) and dose response (12 hr) studies. ................................................... 78 Figure 18. Putative primary TCDD responsive genes from CHX study. ....................................................... 81 Figure 19. Cluster analysis of HL1-1 and HepG2 differentially expressed genes. ........................................ 96 Figure 20. Comparative analysis of HL1-1, HepG2, Hepa1c1c7 and C57BL/6 hepatic tissue gene expression profiles. ............................................................................................................. 100 Figure 21. Time course study designs. ......................................................................................................... 124 Figure 22. Clofibrate and Wy-14,643 effects on (A) body weight and (B) relative liver weight (RLW). .. 127 Figure 23. Representative histopathology results from vehicle-, clofibrate-, and Wy-14,643-treated mice at 4 and 14 days. ..................................................................................................................... 128 Figure 24. Mouse hepatic temporal response comparison between clofibrate and Wy-14643 treatment.... 130 Figure 25. Hierarchical clustering of 410 differentially expressed genes by gene and time. ...................... 133 Figure 26. QRT-PCR verification of selected microarray time course results. ........................................... 135 Figure 27. Functional categorization of the top 50 genes differentially expressed by Wy-14,643 and clofibrate treatment. ............................................................................................................ 138 xi Figure 28. Pathway mapping analysis. ......................................................................................................... 149 Figure 29. Basal PPAR mRNA and protein levels. ................................................................................... 153 Figure 30. Cytotoxicity testing. .................................................................................................................... 154 Figure 31. Comparative analysis of in vivo mouse liver and in vitro human HL1-1 and HepG2 temporal gene expression elicited by Wy-14,643. ............................................................................. 156 Figure 32. Comparative analysis of in vivo mouse liver and in vitro human liver stem cell (HL1-1) time course studies with Wy-14,643 treatment. .......................................................................... 158 xii LIST OF ABBREVIATIONS ADMET Absorption, distribution, metabolism, elimination and toxicity AhR Aryl hydrocarbon receptor ANOVA Analysis of variance ARNT Aryl hydrocarbon receptor nuclear translocator Asc-2P L-ascorbic acid 2-phosphate bHLH/PAS Basic helix-loop-helix/PER-ARNT-SIM BP Biological process BW Body weight CC Cellular component cDNA Complementary deoxyribonucleic acid ChIP Chromatin immunoprecipitation CHX Cycloheximide CLO Clofibrate cpdl Cumulative population doubling level CV Central vein DAVID Database for Annotation, Visualization and Integrated Discovery dNTP Deoxynucleotide triphosphate DRE Dioxin response element DTT Dithiothreitol ED50 Effective dose 50 EDC Endocrine disrupting chemical xiii EDSP Endocrine disruptor screening program EPA Environmental Protection Agency ES cell Embryonic stem cell EU European Union FBS Fetal bovine serum FN Fibronectin FNF Fenofibrate GO Gene ontology HGF Hepatocyte growth factor HSP Heat shock protein hTERT Human telomertase reverse transciptase IARC International Agency for Research on Cancer iPS cell Induced pluripotent stem cell KEGG Kyoto Encyclopedia of Genes and Genomes LDB Ligand-binding domain MCD Methionine and choline deficient MEM Modified Eagle's minimum essential medium MeV Multiexperiment Viewer MSC Mesenchymal stem cells MSS Matrix similarity score MTT 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H tetrazorium bromide NAC N-acetyl-L-cysteine NAFLD Non-alcoholic fatty liver disease xiv NGS Next generation sequencing NR Nuclear receptor NRC National Research Council PCA Principal component analysis PND Postnatal day PP peroxisome proliferator PPAR Peroxisome proliferator-activated receptor  PPRE Peroxisome proliferator response elements PV Portal vein PWM Position weight matrix PXR Pregnane X receptor QRT-PCR Quantitative real-time PCR REACH the Registration, Evaluation, Authorization, and Restriction of Chemicals RLW Relative liver weight RXR Retinoid X receptor SD Standard deviation SDS Sodium dodecyl sulfate SE Standard error SNP Single-nucleotide polymorphism T 2,3,7,8-Tetrachlorodibenzo-p-dioxin TCDD 2,3,7,8-Tetrachlorodibenzo-p-dioxin TIMS Toxicogenomic information management system TSS Transcription start sites xv UCSC University of California Santa Cruz UTR Untranslated region V Vehicle VEH Vehicle WHO World Health Organization WY Wy-14,643 xvi CHAPTER 1 1 CHAPTER 1 REVIEW OF THE LITERATURE: APPLICATION OF HUMAN STEM CELL IN RECEPTOR MEDIATED TOXICITY CURRENT NEEDS OF ALTERNATIVE METHODS FOR TOXICITY TESTS Development of mechanism based alternative methods and in vitro models to replace/reduce/refine animal testing are a current issue to pharmaceutical and chemical industries and government. Pharmaceutical industries invest more than 20 billion US dollars annually for the development of novel drugs [1, 2]. New drug candidates must pass a series of preclinical tests and clinical trials in order to launch but less than 10% survive to reach the market [3]. Failures typically include safety problems and lack of effectiveness which are not predicted by tests in animal models before entering clinical trials. Toxicity of drug candidates is one of the most significant causes of attrition [4]. In many cases, the toxicity of candidate drugs is not detected until clinical trials and late stage failures contribute substantial costs of drug development. In order to improve success and reduce late stage attrition, earlier more accurate high throughput toxicity prediction is needed [5, 6]. Recent legislations in the European Union (EU) and US have bolstered alternative animal testing efforts. In 2006, the EU developed the Registration, Evaluation, Authorization, and Restriction of Chemicals (REACH) legislation which established provisions for improved identification of intrinsic properties of chemical substances. REACH places greater responsibility on industry to manage chemical risks and to provide additional safety information. 2 Estimated costs for registration and testing range from 2.3-100 billion Euro during the first 11 years [7]. Moreover, testing on animals for cosmetic purposes has been banned effective 2009, and requires the use of alternative methods [8]. In the US, the National Research Council (NRC) report “Toxicity Testing in the 21st Century: A Vision and a Strategy” recommends the development of novel tools and approaches including high-throughput in vitro methods, systems biology, and computer-based modeling for toxicity testing [9-11]. Endocrine disrupting chemicals (EDCs) are exogenous substances that alter functions of the endocrine system and consequently cause adverse health effects. EDCs include a wide variety of chemicals, such as natural and synthetic hormones, pesticides, industrial by-products, consumer products, and pollutants [12]. EDCs were declared a high research priority by the World Health Organization (WHO) in 2010 [13]. The US EPA has developed the Endocrine Disruptor Screening Program (EDSP), consisting of Tier 1 screening and Tier 2 testing, in order to be compliant with the Food Quality Protection Act and Safe Drinking Water Act amendments [14, 15]. Screening deficits for selected toxicities has prompted the need to develop new toxicity testing and assay methods approaches. Traditional toxicological tests typically involve animal experiments including multigeneration studies which require a large number of animals, are time-consuming and result in high costs. As an alternative, in vitro systems are used to gain mechanistic understanding of the health effects for preclinical safety testing and in vivo studies are then used to establish doselimiting toxicity and identify target organs [16]. Alternative in vitro methods have many advantages for toxicity testing including small set-ups minimal test substance requirements, lower costs, high numbers of replicates, miniaturization, and automation. In addition, emerging 3 novel technologies including imaging and comprehensive ‘omic’ technologies also stimulate developments of alternative methods. TOXICOGENOMIC APPROACHES Toxicogenomics is the study of the response of a genome to environmental stressors and toxicants. The ability to probe thousands of genes simultaneously has made genomics a prime technology for toxicology. Microarrays and other methods of gene expression profiling have served as useful tools for identifying the responses of toxicologically relevant genes and discerning the mechanisms invoking toxic effects. Moreover, toxicogenomics can identify biomarkers of disease and exposure to toxic substances. Toxicogenomic approaches are also highly sensitive and detect toxic responses at earlier time points and at lower doses than conventional approaches such as histopathology and clinical chemistry [17, 18]. Toxicogenomic studies are usually designed to correlate expression profiles with phenotype defined by conventional measures of toxicity. Phenotypic anchoring of molecular expression data involves determining the relationship between a particular expression profile and the pharmacological or toxicological phenotype of the organism. A key goal in toxicogenomics is to integrate data from different studies and analytical platforms to produce a richer and biologically more refined understanding of the toxicological response [19]. It integrates genetics, genomic-scale mRNA expression (transcriptomics), protein expression (proteomics), metabolite profiling (metabonomics) and bioinformatics with conventional toxicology to identify molecular changes and elucidate molecular mechanisms of toxicity and disease causation [20-24]. Integration of data can provide a more complete picture of the expression profiles that are associated with a particular treatment. There are a number of 4 public repositories available that provide combined profile data with associated biological, chemical and toxicological endpoints (Table 1). The rapid accumulation of genomic-sequence data and associated gene and protein annotation has catalyzed the application of gene-expression analysis to understanding the modes of action of chemicals and other environmental stressors on biological systems. Improvements of bioinformatics related on statistical analysis and presentation of microarray data make it easier to interpret ‘omic’ data. Prior to the current advances in bioinformatics, the most common way of reporting results of microarray studies involved listing differentially expressed genes, with little information about the statistical significance or biological pathways. New mathematical and graphical approaches have been developed to improve data presentation and interpretation. Furthermore, curated web-based tools and software applications have been developed to provide information on cellular location, physiological function, or disease association of a given gene (Table 1). These approaches for analysis of microarray data, initiate “pathway mapping” provide more biological relevance to the analyses. Moreover, predictive systems toxicology will gradually evolve, aided by knowledge that is systematically generated through literature mining, comparative analysis and iterative modeling of expression datasets. Nuclear receptors (NRs) regulate numerous interacting biological processes, which confound the identification of key events and molecular targets related to toxicity [17]. Global genomic profiling approaches are required to comprehensively decipher intricate NRs regulation mechanisms. NUCLEAR RECEPTORS AND ARYL HYDROCARBON RECEPTOR MEDIATED TOXICITY Nuclear receptors are ligand-activated transcription factors involved in the regulation of specific target genes associated with metabolism, development, and cell differentiation [17]. 5 Table 1. Online resources for toxicogenomic studies Online Resource Name  Omic Data Resources  GEO  ArrayExpress  CEBS  TG‐GATES  SMD    Gene Signature Resources  DrugMatrix  NextBio  GeneSigDB  ArrayExpress  Connectivity Map    Toxicity Data Resources  TOXNET  NTP database  Comparative Toxicogenomics  Database (CTD)  ACToR  ToxRefDB  ExpoCast  ToxCast  DssTox  Carcinogen Potency Database  URLs    http://www.ncbi.nlm.nih.gov/geo/ http://www.ebi.ac.uk/arrayexpress/ http://www.niehs.nih.gov/research/resources/databases/cebs/ http://toxico.nibio.go.jp/ http://smd.stanford.edu/     https://ntp.niehs.nih.gov/drugmatrix/index.html http://www.nextbio.com/b/nextbio.nb http://compbio.dfci.harvard.edu/genesigdb/ http://www.ebi.ac.uk/arrayexpress/ http://www.broadinstitute.org/genome_bio/connectivitymap.html     http://toxnet.nlm.nih.gov/ http://ntp‐apps.niehs.nih.gov/ntp_tox/index.cfm http://ctd.mdibl.org/                         http://actor.epa.gov/actor/faces/ACToRHome.jsp http://www.epa.gov/ncct/toxrefdb/ http://www.epa.gov/ncct/expocast/ http://www.epa.gov/ncct/toxcast/ http://www.epa.gov/ncct/dsstox/index.html http://potency.berkeley.edu/             6   Table 1 (cont’d) Online Resource Name  Gene Annotation Information  UCSC Genome browser   NCBI Entrez Gene   OmicBrowser    Biological Ontology  DAVID Bioinformatics  Gene Ontology Database  Ingenuity  GSEA  GeneGo  Open Biomedical Ontologies  CoPub  PantherDB  ToppGene    Pathway Resources  KEGG Pathway   Biocarta  GenMapp  Pathway Interaction Database  Reactome  NetPath  WikiPathways  SMPDB  PantherDB  URLs    http://www.genome.ucsc.edu/ http://www.ncbi.nlm.nih.gov/gene http://omicspace.riken.jp/db/genome.html     http://david.abcc.ncifcrf.gov/ http://www.geneontology.org/ http://www.ingenuity.com/ http://www.broadinstitute.org/gsea/index.jsp http://www.genego.com/ http://www.obofoundry.org/ http://services.nbic.nl/copub/portal/ http://www.pantherdb.org/panther/ontologies.jsp http://toppgene.cchmc.org/     http://www.genome.jp/kegg/pathway.html http://www.biocarta.com/genes/allPathways.asp http://www.genmapp.org/ http://pid.nci.nih.gov/ http://www.reactome.org/ReactomeGWT/entrypoint.html http://www.netpath.org/ http://www.wikipathways.org/index.php/WikiPathways http://www.smpdb.ca/ http://www.pantherdb.org/pathway/index.jsp                                         7   Endogenous or xenobiotic ligands bind the ligand-binding domain (LDB) at the carboxy terminus resulting in receptor activation. The activated receptor binds to recognition elements and recruits accessory proteins such as co-activators, co-repressors, and basal transcriptional factors to regulate target gene expression. NRs are attractive drug targets due to their regulatory role of a wide range of physiologic responses. In addition, many pharmaceutical agents along with environmental chemicals are NR ligands. Most adverse effects of EDCs are mediated by nuclear receptors and a number of therapeutic compounds including antibiotics, anticonvulsants, hypolipidemics, and cancer therapies target these receptors [25]. The peroxisome proliferator-activated receptors (PPARs) and pregnane X receptor (PXR) are members of the nuclear receptor superfamily that dimerize with the retinoid X receptor (RXR) and bind to specific regions on the DNA of target genes to modulate their expression. PPARs are known for their role in fatty acid metabolism and glucose homeostasis [26] and are useful drug targets for hyperlipidemia, diabetes and obesity. Additionally, the PXR plays an important role in xenobiotic sensing and regulation of genes involved in expression of phase I and II metabolizing enzymes to protect the liver and other organs from potentially harmful compounds [27-29]. PXR has a large number of exogenous ligands, many of which have been identified as endocrine-disrupting chemicals [17]. NRs and the aryl hydrocarbon receptor (AhR), a basic helix-loop-helix/PER-ARNT-SIM (bHLH/PAS) transcription factor, share similar modes of action as ligand-activated transcription factors which regulates the expression of genes involved in pleiotropic responses [17]. 2,3,7,8Tetrachlorodibenzo-p-dioxin (TCDD) is the prototypical AhR ligand. It is a ubiquitous, bioaccumulative environmental contaminant that causes various effects including endocrine, immunotoxicity, hepatotoxicity, teratogenesis, and multi-site tumor promotion [30]. A variety of 8 studies have demonstrated the carcinogenicity of TCDD which acts as a multi-site carcinogen and liver tumor promoter in rodents [31]. TCDD was also listed as an established human carcinogen by the International Agency for Research on Cancer (IARC) in 1997 [32]. Most, if not all of these effects are mediated by the AhR [33]. Receptor-mediated effects are attractive targets for toxicity screening and drug development, due to their role as drug and environmental toxicant targets and their high association with the human diseases. Despite numerous studies assessing NR and AhR-mediated toxicity using animal models, extrapolation to humans has identified a significant number of discordances [34]. Although the AhR is evolutionary conserved, responses elicited by TCDD and related compounds vary widely across species [35-38]. Induction of PPAR by chronic exposure to the peroxisome proliferators (PPs) fibrate drugs resulted in peroxisome proliferation and the development of hepatocarcinomas in mice. However, primates and humans appear to be non-responsive to these adverse effects. Species differences in PPARα function, particularly between mice and humans, are attributed to the level of PPARα expression, ligand activation, and biological responses [39-42]. PPARα humanized mice results suggest that species differences in the receptor only partially explain the differential susceptibility [43]. CURRENT MODELS FOR RECEPTOR MEDIATED HEPATOTOXICITY The liver is a vital organ essential for metabolism. It receives venous blood from the small intestine, stomach, pancreas and spleen containing high levels of nutrients, xenobiotics and other compounds. Due to reactive metabolite formation resulting from metabolism, the liver is a frequent target of toxicity. Most toxicogenomics studies so far have examined hepatotoxicants as the liver is the primary source of xenobiotic metabolism and detoxification and because liver injury is the principal reason for withdrawal of new drugs from the market. Notably, liver 9 toxicity and alterations of hepatic function are the most frequent reasons for pre-clinical drug development failure. To assess the risk of specific compounds in humans, many toxicity studies are performed by animal tests. However, it is difficult to accurately predict the hepatotoxicity of new compounds in humans and is often observed only in the clinical trials. The underlying reason for this may be related to marked species differences and is due to extensive use of animal models. Due to species differences and to reduce animal use, there is a need for a reliable human in vitro hepatotoxicity test systems. Significant effort has been devoted to establish predictive human hepatic cell models that could be assayed in vitro. In vitro models such as continuous cell lines, primary cells and organ slices, have been used to investigate ADMET (absorption, distribution, metabolism, elimination and toxicity) properties[44]. The principle approach of in vitro models is to assess the mode of action, including toxicant-target interaction and dose/time dependent responses [45]. In general, the more similar the model system is to a human target tissue, the more likely the results will predict in vivo human responses. Therefore, mechanism-based approach using an in vitro model system derived from human target tissue will provide more accurate predictive models of in vivo, responses relevant to efficacy and safety [46]. Although in vitro models do not always replicated in vivo responses, human based in vitro models are more likely to predict in vivo human responses and are preferred to predict mechanism of action of the chemicals [6, 45, 47, 48]. Hepatocytes constitute 60% of the liver cell population (80% of the volume) and are the major functional cells performing metabolic, endocrine and secretory functions [49]. Typical available models are based on human cancer cell lines or primary cells isolated from biopsies, but these have significant drawbacks. For example, human hepatoma cell lines such as HepG2, 10 poorly replicate phenotypic and functional native hepatocytes [50-53]. Cancer cells may also have mutations that result in the loss of significant portions of a chromosome. The preferred option, with respect to functional differentiation, is human primary hepatocytes. However, availability, viability and variability limit their usefulness. Furthermore, the purity of the primary isolations is also a point of concern as related non-parenchymal cells contaminate the final cell preparations [45, 47, 54]. Primary hepatocytes also do not proliferate and gradually lose their metabolic activity after maintaining in the cell culture media several weeks [51]. Hence, there is a need for novel models for human hepatocytes. Since mature hepatocytes are difficult to maintain and grow in vitro, liver stem or precursor cells derived from adult liver are potential alternatives. Limitation of mature hepatocytes may be overcome by the generation of differentiated hepatocytes from adult or embryonic stem cells or immortalization of differentiated hepatocytes. APPLICATION OF HUMAN STEM CELLS FOR TOXICOLOGY Stem cells are undifferentiated cells with the capacity for unlimited or prolonged selfrenewal and the ability to give rise to differentiated cells. The regenerative capacity of the liver is well established in models of partial hepatectomy or hepatotoxic injury [55, 56]. Evidence from several studies indicates the presence of resident stem cells in the adult liver [49, 57-59] and the oval cells located in the canals of Herring or Cholangioles appear to represent the progeny of pluripotential liver stem cells which are capable of generating hepatic lineages. Stem cells have the capacity of self-renewal and differentiation into virtually any cell type making them an attractive in vitro model. Human stem cells provide an attractive in vitro alternative and a potentially unlimited source of human cells [60, 61]. Normal adult human stem cells isolated from liver tissue might provide an alternative system which can proliferate and 11 closely mimics human responses [60, 62, 63]. Another advantage is the use of cells derived from a single stem cell line which would minimize donor and preparation variability. Moreover, there is increasing evidence that the origin of cancer is the adult stem cell population and their derivatives, which are targets for transforming mutations/epigenetic changes [64, 65]. Genomic profiling of nuclear receptor-mediated responses following toxicant exposure has confirmed the role of nuclear receptors in maintaining homeostatic conditions and mediating toxicity, distinguishing chemicals with similar or diverse mechanisms of toxicity and identifying potential gene expression markers of toxicity [17, 18]. Genomics can effectively be used in vitro and across non-rodent animal models to confirm pathways of toxicity. Mechanism-based approaches for in vitro model system may also lead to predictive models for toxicity screening [66]. The application of a toxicogenomic approach to human liver stem cells will provide global gene expression profiles leading to species comparison and the identification of species specific molecular mechanisms of toxicity. Taken together, toxicogenomic approach to human liver stem cells is a novel toxicity screening assay to identify mechanistically-based mode of action for early nuclear receptor-mediated toxicity. The use of human stem cells technologies could aid in toxicity testing during drug development and reduce the need for in vivo testing. 12 REFERENCES 13 REFERENCES 1. Morgan, S., et al., The cost of drug development: a systematic review. Health Policy, 2011. 100(1): p. 4-17. 2. Dimasi, J.A., Risks in new drug development: approval success rates for investigational drugs. Clin Pharmacol Ther, 2001. 69(5): p. 297-307. 3. DiMasi, J.A., et al., Trends in risks associated with new drug development: success rates for investigational drugs. Clin Pharmacol Ther, 2010. 87(3): p. 272-7. 4. Watkins, P.B., Drug safety sciences and the bottleneck in drug development. Clin Pharmacol Ther, 2011. 89(6): p. 788-90. 5. Dalgetty, D.M., et al., Progress and future challenges in stem cell-derived liver technologies. Am J Physiol Gastrointest Liver Physiol, 2009. 297(2): p. G241-8. 6. Pritchard, J.F., et al., Making better drugs: Decision gates in non-clinical drug development. Nat Rev Drug Discov, 2003. 2(7): p. 542-53. 7. Williams, E.S., J. Panko, and D.J. Paustenbach, The European Union's REACH regulation: a review of its history and requirements. Crit Rev Toxicol, 2009. 39(7): p. 553-75. 8. Taylor, K., C. Casalegno, and W. Stengel, A critique of the EC's expert (draft) reports on the status of alternatives for cosmetics testing to meet the 2013 deadline. ALTEX, 2011. 28(2): p. 131-48. 9. Gibb, S., Toxicity testing in the 21st century: a vision and a strategy. Reprod Toxicol, 2008. 25(1): p. 136-8. 10. Krewski, D., et al., Toxicity testing in the 21st century: implications for human health risk assessment. Risk Anal, 2009. 29(4): p. 474-9. 11. Dix, D.J., et al., The ToxCast program for prioritizing toxicity testing of environmental chemicals. Toxicol Sci, 2007. 95(1): p. 5-12. 12. Soto, A.M. and C. Sonnenschein, Environmental causes of cancer: endocrine disruptors as carcinogens. Nat Rev Endocrinol, 2010. 6(7): p. 363-70. 13. Alonso-Magdalena, P., I. Quesada, and A. Nadal, Endocrine disruptors in the etiology of type 2 diabetes mellitus. Nat Rev Endocrinol, 2011. 14. Stokes, W.S., Selecting appropriate animal models and experimental designs for endocrine disruptor research and testing studies. ILAR J, 2004. 45(4): p. 387-93. 14 15. Willett, C.E., P.L. Bishop, and K.M. Sullivan, Application of an integrated testing strategy to the U.S. EPA endocrine disruptor screening program. Toxicol Sci, 2011. 123(1): p. 15-25. 16. Hartung, T., From alternative methods to a new toxicology. Eur J Pharm Biopharm, 2011. 77(3): p. 338-49. 17. Woods, C.G., J.P. Heuvel, and I. Rusyn, Genomic profiling in nuclear receptor-mediated toxicity. Toxicol Pathol, 2007. 35(4): p. 474-94. 18. Morgan, K.T., et al., The hepatic transcriptome as a window on whole-body physiology and pathophysiology. Toxicol Pathol, 2005. 33(1): p. 136-45. 19. Waters, M.D. and J.M. Fostel, Toxicogenomics and systems toxicology: aims and prospects. Nat Rev Genet, 2004. 5(12): p. 936-48. 20. Tennant, R.W., The National Center for Toxicogenomics: using new technologies to inform mechanistic toxicology. Environ Health Perspect, 2002. 110(1): p. A8-10. 21. Mattes, W.B., et al., Database development in toxicogenomics: issues and efforts. Environ Health Perspect, 2004. 112(4): p. 495-505. 22. Gant, T.W. and S.D. Zhang, In pursuit of effective toxicogenomics. Mutat Res, 2005. 575(1-2): p. 4-16. 23. Boverhof, D.R. and T.R. Zacharewski, Toxicogenomics in risk assessment: applications and needs. Toxicol Sci, 2006. 89(2): p. 352-60. 24. Carlson, E.A. and J.B. Silkworth, Toxicogenomics in human health risk assessment. Toxicol Appl Pharmacol, 2009. 236(2): p. 254-6; author reply 257-60. 25. Moore, J.T., J.L. Collins, and K.H. Pearce, The nuclear receptor superfamily and drug discovery. ChemMedChem, 2006. 1(5): p. 504-23. 26. Dreyer, C., et al., Control of the peroxisomal beta-oxidation pathway by a novel family of nuclear hormone receptors. Cell, 1992. 68(5): p. 879-87. 27. Kretschmer, X.C. and W.S. Baldwin, CAR and PXR: xenosensors of endocrine disrupters? Chem Biol Interact, 2005. 155(3): p. 111-28. 28. Timsit, Y.E. and M. Negishi, CAR and PXR: the xenobiotic-sensing receptors. Steroids, 2007. 72(3): p. 231-46. 29. Matic, M., et al., Pregnane X receptor: promiscuous regulator of detoxification pathways. Int J Biochem Cell Biol, 2007. 39(3): p. 478-83. 15 30. Denison, M.S. and S. Heath-Pagliuso, The Ah receptor: a regulator of the biochemical and toxicological actions of structurally diverse chemicals. Bull Environ Contam Toxicol, 1998. 61(5): p. 557-68. 31. Knerr, S. and D. Schrenk, Carcinogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in experimental models. Mol Nutr Food Res, 2006. 50(10): p. 897-907. 32. McHale, C.M., et al., Microarray analysis of gene expression in peripheral blood mononuclear cells from dioxin-exposed human subjects. Toxicology, 2007. 229(1-2): p. 101-13. 33. Okey, A.B., An aryl hydrocarbon receptor odyssey to the shores of toxicology: the Deichmann Lecture, International Congress of Toxicology-XI. Toxicol Sci, 2007. 98(1): p. 5-38. 34. Olson, H., et al., Concordance of the toxicity of pharmaceuticals in humans and in animals. Regul Toxicol Pharmacol, 2000. 32(1): p. 56-67. 35. Silkworth, J.B., et al., Comparison of TCDD and PCB CYP1A induction sensitivities in fresh hepatocytes from human donors, sprague-dawley rats, and rhesus monkeys and HepG2 cells. Toxicol Sci, 2005. 87(2): p. 508-19. 36. Boverhof, D.R., et al., Comparative toxicogenomic analysis of the hepatotoxic effects of TCDD in Sprague Dawley rats and C57BL/6 mice. Toxicol Sci, 2006. 94(2): p. 398-416. 37. Nohara, K., et al., Comparison of the 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)induced CYP1A1 gene expression profile in lymphocytes from mice, rats, and humans: most potent induction in humans. Toxicology, 2006. 225(2-3): p. 204-13. 38. Kransler, K.M., B.P. McGarrigle, and J.R. Olson, Comparative developmental toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the hamster, rat and guinea pig. Toxicology, 2007. 229(3): p. 214-25. 39. Lawrence, J.W., et al., Differential gene regulation in human versus rodent hepatocytes by peroxisome proliferator-activated receptor (PPAR) alpha. PPAR alpha fails to induce peroxisome proliferation-associated genes in human cells independently of the level of receptor expresson. J Biol Chem, 2001. 276(34): p. 31521-7. 40. Klaunig, J.E., et al., PPARalpha agonist-induced rodent tumors: modes of action and human relevance. Crit Rev Toxicol, 2003. 33(6): p. 655-780. 41. Ammerschlaeger, M., et al., Characterization of the species-specificity of peroxisome proliferators in rat and human hepatocytes. Toxicol Sci, 2004. 78(2): p. 229-40. 42. Gonzalez, F.J. and Y.M. Shah, PPARalpha: Mechanism of species differences and hepatocarcinogenesis of peroxisome proliferators. Toxicology, 2007. 16 43. Morimura, K., et al., Differential susceptibility of mice humanized for peroxisome proliferator-activated receptor alpha to Wy-14,643-induced liver tumorigenesis. Carcinogenesis, 2006. 27(5): p. 1074-80. 44. Li, A.P., Screening for human ADME/Tox drug properties in drug discovery. Drug Discov Today, 2001. 6(7): p. 357-366. 45. MacGregor, J.T., et al., In vitro human tissue models in risk assessment: report of a consensus-building workshop. Toxicol Sci, 2001. 59(1): p. 17-36. 46. Burczynski, M.E., et al., Toxicogenomics-based discrimination of toxic mechanism in HepG2 human hepatoma cells. Toxicol Sci, 2000. 58(2): p. 399-415. 47. Abbott, A., Cell culture: biology's new dimension. Nature, 2003. 424(6951): p. 870-2. 48. MacGregor, J.T., SNPs and chips: genomic data in safety evaluation and risk assessment. Toxicol Sci, 2003. 73(2): p. 207-8. 49. Tanaka, M., et al., Liver stem/progenitor cells: their characteristics and regulatory mechanisms. J Biochem, 2011. 149(3): p. 231-9. 50. Harris, A.J., S.L. Dial, and D.A. Casciano, Comparison of basal gene expression profiles and effects of hepatocarcinogens on gene expression in cultured primary human hepatocytes and HepG2 cells. Mutat Res, 2004. 549(1-2): p. 79-99. 51. Wilkening, S., F. Stahl, and A. Bader, Comparison of primary human hepatocytes and hepatoma cell line Hepg2 with regard to their biotransformation properties. Drug Metab Dispos, 2003. 31(8): p. 1035-42. 52. Zhang, Z.Y., et al., Preferential inducibility of CYP1A1 and CYP1A2 by TCDD: differential regulation in primary human hepatocytes versus transformed human cells. Biochem Biophys Res Commun, 2006. 341(2): p. 399-407. 53. Guillouzo, A., et al., The human hepatoma HepaRG cells: a highly differentiated model for studies of liver metabolism and toxicity of xenobiotics. Chem Biol Interact, 2007. 168(1): p. 66-73. 54. Boess, F., et al., Gene expression in two hepatic cell lines, cultured primary hepatocytes, and liver slices compared to the in vivo liver gene expression in rats: possible implications for toxicogenomics use of in vitro systems. Toxicol Sci, 2003. 73(2): p. 386402. 55. Alison, M.R., S. Islam, and S. Lim, Stem cells in liver regeneration, fibrosis and cancer: the good, the bad and the ugly. J Pathol, 2009. 217(2): p. 282-98. 56. Michalopoulos, G.K. and M.C. DeFrances, Liver regeneration. Science, 1997. 276(5309): p. 60-6. 17 57. Herrera, M.B., et al., Isolation and characterization of a stem cell population from adult human liver. Stem Cells, 2006. 24(12): p. 2840-50. 58. Gaudio, E., et al., New insights into liver stem cells. Dig Liver Dis, 2009. 41(7): p. 45562. 59. Zhang, L., et al., The stem cell niche of human livers: symmetry between development and regeneration. Hepatology, 2008. 48(5): p. 1598-607. 60. Rohwedel, J., et al., Embryonic stem cells as an in vitro model for mutagenicity, cytotoxicity and embryotoxicity studies: present state and future prospects. Toxicol In Vitro, 2001. 15(6): p. 741-53. 61. Davila, J.C., et al., Use and application of stem cells in toxicology. Toxicol Sci, 2004. 79(2): p. 214-23. 62. Rolletschek, A., P. Blyszczuk, and A.M. Wobus, Embryonic stem cell-derived cardiac, neuronal and pancreatic cells as model systems to study toxicological effects. Toxicol Lett, 2004. 149(1-3): p. 361-9. 63. Carere, A., A. Stammati, and F. Zucco, In vitro toxicology methods: impact on regulation from technical and scientific advancements. Toxicol Lett, 2002. 127(1-3): p. 153-60. 64. Ahuja, Y.R., V. Vijayalakshmi, and K. Polasa, Stem cell test: a practical tool in toxicogenomics. Toxicology, 2007. 231(1): p. 1-10. 65. Alison, M.R. and M.J. Lovell, Liver cancer: the role of stem cells. Cell Prolif, 2005. 38(6): p. 407-21. 66. Suk, W.A., K. Olden, and R.S. Yang, Chemical mixtures research: significance and future perspectives. Environ Health Perspect, 2002. 110 Suppl 6: p. 891-2. 18 CHAPTER 2 19 CHAPTER 2 RATIONALE, HYPOTHESIS AND SPECIFIC AIMS RATIONALE Receptor-mediated toxicity is an attractive target for toxicity test and drug development, since receptors are a major drug and toxicant target and due to their high association with human diseases. Numerous studies to assess nuclear receptor and AhR-mediated toxicity have used animal models. However, extrapolating animal model data to humans has limited success [1]. Alternatively, the use of in vitro human-based model systems has focused more attention in toxicity tests [2]. Even though human-derived continuous cells, primary cells and tissue slice have been used toxicity testing, they are limited by abnormalities introduced into continuous cell lines, and the cost, instability and availability of primary cells/tissue model systems [3-5]. Therefore, more efficient and reliable, high throughput human-derived in vitro model system to assess the human toxicity warrant developed. This study will focus on developing a novel in vitro model system for receptor-mediated toxicity screening employing normal human liver stem cells isolated from adult liver tissue, an attractive alternative that provides high proliferative capacity and an origin from normal liver tissue. HYPOTHESIS Normal adult human liver stem cells are a viable in vitro model for receptor-mediated human gene response study related with toxicity. 20 SPECIFIC AIMS To address this hypothesis, the following specific aims with human liver stem cell characterization and engineering, and cross-model & cross-species comparative toxicogenomic analysis will be used: 1. Evaluate HL1-1 cells as an in vitro receptor-mediated toxicity model system by determining the expression and functionality of AhR and PPAR and cell engineering for its practical application. 2. Establish baseline quantitative temporal and dose-dependent data on global gene expression in HL1-1 human cells elicited by AhR agonists and evaluate the modelspecific and model-conserved gene expression responses. 3. Establish baseline quantitative temporal data on the hepatic effects elicited by PPAR agonists in an in vivo mouse model and on gene expression in HL1-1 human cells and evaluate the model-specific and model-conserved gene expression responses. 21 REFERENCES 22 REFERENCES 1. Olson, H., et al., Concordance of the toxicity of pharmaceuticals in humans and in animals. Regul Toxicol Pharmacol, 2000. 32(1): p. 56-67. 2. Li, A.P., Screening for human ADME/Tox drug properties in drug discovery. Drug Discov Today, 2001. 6(7): p. 357-366. 3. MacGregor, J.T., et al., In vitro human tissue models in risk assessment: report of a consensus-building workshop. Toxicol Sci, 2001. 59(1): p. 17-36. 4. Abbott, A., Cell culture: biology's new dimension. Nature, 2003. 424(6951): p. 870-2. 5. Boess, F., et al., Gene expression in two hepatic cell lines, cultured primary hepatocytes, and liver slices compared to the in vivo liver gene expression in rats: possible implications for toxicogenomics use of in vitro systems. Toxicol Sci, 2003. 73(2): p. 386402. 23 CHAPTER 3 24 CHAPTER 3 EVALUATION OF HUMAN ADULT HEPATIC STEM CELLS AS AN IN VITRO RECEPTOR-MEDIATED TOXICOGENOMIC MODEL AND THEIR CELL ENGINEERING FOR IMMORTALIZATION ABSTRACT Primary and continuous cell lines are valuable in vitro models that have been widely used to provide insights into mechanisms of toxicity. However the abnormality of continuous cell lines and a high cost and instability of primary cells limit their use as ideal model systems. The availability of human adult stem cells provides an innovative in vitro model that may more accurately predict in vivo toxicity. In this study, the human hepatic adult stem cell line, HL1-1, was assessed as a viable model for receptor-mediated toxicogenomic studies. Furthermore, HL11 cell were immortalized by hTERT stable transfection (HLhT1) to overcome its limited application due to cell senescence and limited life span. The mRNA and protein expression levels of the Aryl hydrocarbon receptor (AhR) and nuclear receptors (PPAR and , ER and PXR) were measured using QRT-PCR and Western blots, respectively in HL-1 and HLhT1 cells. The functionality of each receptor was assessed by treatment with prototypical agonist for each receptor and monitoring gene expression changes with QRT-PCR. TCDD, clofibrate and rifampicin were used as prototypical agonists and changes in mRNA expression of CYP1A1 (AhR), CPT1A (PPAR), and CYP3A4 (PXR) were evaluated. Expression level and responsiveness of the AhR and PPAR was confirmed in HL1-1 cells, however no changes were observed for the PXR. The immortalized HLhT1 cell maintains comparable expression of stem 25 cell and oval cell marker proteins (OCT4, AFP, VIM and THY1) relative to the parental cells. In addition, HLhT1 cell express functional AhR and PPAR and their responsiveness is also comparable to the parental cell line. Collectively, human liver stem cells, HL1-1 and HLhT1, are promising toxicogenomic models for mechanistic investigation of receptor-mediated hepatic toxicity. INTRODUCTION The liver is a central organ essential for the metabolism. The blood entering the liver consists of 75% of venous blood from the portal vein and 25% of arterial blood from the hepatic artery. All of the venous blood returning from the small intestine, stomach, pancreas and spleen converge into the portal vein and the liver receive all absorbed nutrients and xenobiotics. Hepatic metabolism leads to formation of reactive metabolites and it make liver as a frequent toxicity target. Liver toxicity and alterations of liver function are the most frequently occurring reasons for toxicology among drug molecules. However, it is difficult to predict hepatotoxic actions of new compounds in humans, because it is mainly related to distinct species differences and toxicity tests are based on the extensive use of animal models. Consequently, there is a demand for a reliable human test system and much effort has been directed to establish predictive human hepatic cell populations. The preferred human in vitro models are primary hepatocytes and hepatoma cell lines, but these have significant demerits. Continuous cell lines, while convenient for assessing the mechanistic toxicity, are mostly derived from diseased tissue or are transformed, which may lead to abnormal responses [1, 2]. The best option today, regarding functional aspect, is isolated human primary hepatocytes, but issues related to their acquisition, instability and variability from various source cause practical constraints that limit their application. Hence, there is a great need for novel improved models for human hepatocytes. 26 Liver stem cells derived from the human liver tissue are a valuable for novel human in vitro hepatic models and a potentially unlimited source of human cells [3, 4]. Stem cells are generally defined as clonogenic cells capable of both self-renewal and multilineage differentiation. In the adult organism, stem cells mediate tissue homeostasis and repair, e.g. the regenerative capacity of the liver is well established in models of partial hepatectomy or hepatotoxic injury [5, 6]. The isolation and culture of liver stem cells from various sources have been reported [7-11]. Human liver stem cells have lot of advantages including the fact that they are non-transformed intact cells from human tissue with proliferating and differentiation ability. Moreover, there is an increasing evidence on the origin of cancer in adult stem cell populations and their derivatives, which are the target of transforming mutations and epigenetic changes [12, 13]. Thus, they could be a potential direct target for certain toxic end points, such as tumor development. Receptor-mediated toxicity is an attractive target for toxicity screening assays and drug development, due to the role of receptors as a major drug and environmental toxicant target and their high association with the human diseases. A number of therapeutic compounds including hypolipidemic agents, antibiotics, cancer therapies, and anticonvulsants target nuclear receptor [15]. Nuclear receptors are members of a superfamily of ligand-activated transcription factors, which are involved in the regulation of specific target genes associated with metabolism, development, and cell differentiation [14]. Endogenous or xenobiotic ligands bind the ligandbinding domain of the receptor resulting in activation of the receptor. Activated receptor binds to recognition elements, recruits accessory proteins such as co-activators, co-repressors, and basal transcriptional factors to regulate the expression of target genes. 27 The peroxisome proliferator-activated receptors (PPARs) are known for their role in fatty acid metabolism and glucose homeostasis [16] and have been identified as useful drug targets for hyperlipidemia, diabetes and obesity. Long-term treatment of peroxisome proliferators (PPs) in rodents results in the formation of hepatocellular tumors by a non-genotoxic mechanism, while humans appear to be non-responsive to these adverse effects. The pregnane X receptor (PXR) plays an important role in xenobiotic sensing and regulation of genes involved in phase I and II metabolizing enzymes to protect the liver and other organs from potentially harmful compounds [17-19]. PXR has a large number of exogenous ligands, many of which have been identified as endocrine-disrupting chemicals [14]. NRs and the aryl hydrocarbon receptor (AhR) share similar mode of action as ligand-activated transcription factors. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is the prototypical AhR ligand [20]. TCDD is ubiquitous and bioaccumulates in the environment causing various adverse and biological effects in experimental animals and humans including endocrine, immuno- and hepato- toxicity, teratogenesis, and multi-site tumor promotion [21, 22]. In this study, HL1-1 human adult liver stem cells are evaluated as a novel model for receptor-mediated toxicogenomic studies. Also, a clonal population of immortalized cells, referred to as immortalized HL1-1 (HLhT1) cells, was obtained and characterized. These liver stem cells isolated from adult liver tissue could be served as novel human in vitro models to elucidate molecular mechanism of receptors-mediated conserved and divergent responses between species. 28 Figure 1. TCDD time course study design for cell line comparison. Cells were treated with either 10 nM TCDD or 0.1% DMSO and harvested at 1, 2, 4, 8, 12, 24, or 48 hrs post-treatment. N = 3. 29 MATERIALS & METHODS CULTURE AND TREATMENT OF CELL LINE HL1-1, L1SV1 and HLhT1 cells were maintained in a modified MCDB 153 media (Keratinocyte-SFM, Invitrogen Corporation, Carlsbad, CA) supplemented with N-acetyl-Lcysteine (NAC) (2 mM), L-ascorbic acid 2-phosphate (Asc-2P) (0.2 mM), nicotinamide (5 mM) (referred to as K-NAC medium) [23]. The calcium concentration of this medium is 0.09 mM. The growth factors/hormones in the culture medium include rEGF (5 ng/mL), bovine pituitary extract (50 mg/mL) and 10% fetal bovine serum (FBS) (Hyclone, Logan, UT). NAC, a potent antioxidant, found to promote the self-renewal of glial precursor cells [24]. Asc-2P is a stable precursor to provide a constant concentration of ascorbate in the culture medium and ascorbate promotes cell growth and cell viability maintenance [25]. Nicotinamide is the poly (ADP-ribose) polymerase inhibitor and is known to prolong survival of primary cultured hepatocyte [26, 27]. D medium is utilized for cell life span expanding study and is a modified Eagle’s minimum essential medium (MEM) medium supplemented with 5% FBS, NAC (2mM), Asc-2P (0.2mM). HepG2 cell lines were maintained in phenol-red free DMEM/F12 media (Invitrogen) supplemented with 5% FBS (Hyclone). Combined antibiotics (50 g/mL gentamycin (Invitrogen), 100 U/mL penicillin and 100 g/mL streptomycin (Invitrogen)) were applied for all 6 2 cell culture media. For the chemical treatment, 1  10 cells were seeded into a 25 cm cell culture flask (#430639, vent cap) (Corning Inc., Corning, NY) and incubated under standard conditions (5% CO2, 37C). For the receptor response studies, cells were treated with 10 nM TCDD, 50 M Wy-14,643 (Sigma-Aldrich), 30 M rifampicin (Sigma-Aldrich), or DMSO (Sigma-Aldrich) vehicle treatment and harvested at 24 or 48 h post-treatment. For the TCDD 30 time course studies, cells were treated with 10 nM TCDD or DMSO (Sigma-Aldrich) vehicle treatment and harvested at 1, 2, 4, 8, 12, 24 or 48 h post-treatment (Figure 1). PROLIFERATION POTENTIAL OF HL1-1 CELL The cumulative population doubling level (cpdl) determined to examine proliferation 5 potential of HL1-1 cell. From the earliest passage of HL1-1 (cpdl = 26.5) stock, 1X10 cells of 2 were plated in 75 cm flask (#431464, vent cap) (Corning Inc., Corning, NY) and grown in KNAC medium containing 10% FBS until near confluence. To quantify the final cell yield they are subcultured continuously until no change in cpdl. The population doubling (pd) at each subculture was calculated by using the following equation: pd = ln (Nf /Ni)/ln 2, where Ni and Nf are initial and final cell numbers, respectively, and ln is the natural log. The pds of continuous subculture were added to obtain cpdl. IMMORTALIZATION OF LIVER CELLS 5 Early passage HL1-1 cells were plated in three 60 mm plate (5x10 per plate). After overnight incubation, the cells were transfected with the human telomerase reverse transcriptase plasmid (pBabe-hygro-hTERT, a gift from Robert A. Weinberg, The Whitehead Institute for Biomedical Research, Department of Biology, Massachusetts Institute of Technology, Cambridge, MA) by Lipofectamine (Life Technologies, Inc., Gaithersburg, MD) (Figure 2). After 4 days incubation in the K-NAC medium, stably transfected clonal populations were selected in 100 mg/ml of hygromycin (Sigma-Aldrich). The surviving actively proliferating clonal cultures were established by the trypsin glass ring cloning of drug-resistant colonies and propagated to sufficient number of cells for storage in liquid nitrogen. After selection, these cells were used for determination of proliferation potential and further characterizations and were 31 compared with parental HL1-1 cells and the L1SV1 cells, another immortalized cell line transformed with simian virus 40 large T antigen (SV40T). PROTEIN PREPARATION AND WESTERN BLOT Cell and tissue lysates for Western blot analysis were prepared in RIPA buffer (1x PBS, 0.1% SDS, 1% IGEPAL CA-630 and 0.5% Na-Deoxycholate) with protease inhibitor cocktail tablet (Roche Diagnostics, Mannheim, Germany) and quantified using the modified Lowry assay (Bio-Rad, DC Protein Assay, Hercules, CA). Cell lysate proteins were electrophoretically separated on a denaturing 12% SDS-polyacrylamide gel and transferred onto nitrocellulose membrane (Amersham Biosciences Inc., Piscataway, NJ). The membranes were probed with anti-human AhR (N-19), PPAR- (H-98), PPAR- (H-100), ER- (H-184) and PXR (N-19) antibody followed by horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology Inc., Santa Cruz, CA). Immunochemical staining and fluorescence detection on X-ray film was performed using the SuperSignal West Dura substrate (Thermo Fisher Scientific, Inc., Waltham, MA). RNA ISOLATION AND QUANTITATIVE REAL-TIME PCR ANALYSIS Cells were harvested by scraping in the presence of TRIzol Reagent (Invitrogen) and o preserved at -80 C until RNA isolation. Total RNA was isolated according to the manufacturer’s protocol and resuspended in RNA Storage Solution (Ambion Inc., Austin, TX). The RNA samples were quantified spectrophotometrically (A260) and assessed for purity by A260/A280 ratio and by visual inspection on a denaturing agarose gel. Quantitative real-time PCR (QRT-PCR) was performed as verification of microarray data for selected genes. For each sample, 1.5 g of total RNA was reverse transcribed by SuperScript II reverse transcriptase 32 (Invitrogen) using an anchored oligo-dT primer as described by the manufacturer’s instructions. 1.5 L of cDNA template was used in a 30 L PCR reaction containing 0.1 M of forward and reverse gene-specific primers designed using Primer3 [28] and SYBR Green PCR reaction mixture (Applied Biosystems, Foster City, CA). PCR amplification was conducted in MicroAmp Optical 96-well reaction plates (Applied Biosystems) on an Applied Biosystems PRISM 7500 Sequence Detection System. A dissociation protocol was performed to assess the specificity of the primers and the uniformity of the PCR products. Target gene cDNAs were quantified using a standard curve of log copy number versus threshold cycle (Ct). The copy number of each sample was standardized to the geometric mean of -actin (ACTB) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), to control for differences in RNA loading, quality, and cDNA synthesis [29]. For graphing purposes, the relative gene expression levels were scaled such that the expression level of the time-matched vehicle treated control group was equal to 1. Official gene names and abbreviations, forward and reverse primer sequences, and product length for the genes verified by QRT-PCR are listed in Table 2. RESULTS RECEPTOR EXPRESSION AND FUNCTIONALITY OF IN HL1-1 CELLS In order to assess HL1-1 cells as a feasible model for receptor-mediated toxicogenomic studies, basal expression levels of nuclear receptors and AhR were evaluated. Both mRNA and protein levels were analyzed by QRT-PCR and Western blotting, respectively. The basal transcripts of the AhR, ER, PPAR and  and PXR were measured in normal proliferating HL1-1 cells by QRT-PCR (Figure 3A). The purity of PCR product was inspected by the dissociation curve analysis and agarose gel electrophoresis (Figure 3B). The AhR expression was shown to be highest and the expression levels of ER and PXR were the lowest based on 33 Table 2. Gene names and primer sequences for QRT-PCR RefSeq Gene name Gene symbol Entrez Gene ID NM_001101 actin, beta ACTB NM_002046 glyceraldehyde-3-phosphate dehydrogenase peroxisome proliferative activated receptor, alpha cytochrome P450, family 4, subfamily A, polypeptide 11 adipose differentiation-related protein aryl hydrocarbon receptor GAPDH 2597 PPARA 65260 CYP4A11 1579 ADFP 123 AHR 157873 peroxisome proliferative activated receptor, gamma estrogen receptor 1 (ER) PPARG 109732 ESR1 165444 CYP1A1 13076 NM_005036 NM_000778 NM_001122 NM_001621 NM_005037 NM_000125 NM_009992 cytochrome P450, family 1, subfamily A, polypeptide 1 60 34 Forward Primer Reverse Primer CATCCCCCAA AGTTCACAAT GGCCTCCAAG GAGTAAGACC GCAGAAACCC AGAACTCAGC GAGGAATGCC TTTCACCAGA ACACCCTCCTG TCCAACATC TGTTGGACGTC AGCAAGTTC TGCAGGTGATC AAGAAGACG TAAATGCTGCC ATGTTCCAA AAGTGCAGAT GCGGTCTTCT AGTGGGGTGG CTTTTAGGAT AGGGGTCTAC ATGGCAACTG ATGGCCCAGT GTAAGAAACG GTTGAGCCTTC CTCAGTTGG GCATTGCGGA ACACTGAGTA TGGTGCCCAG AATAATGTGA TGGAAGAAGG GAAATGTTGG AATGCAAAGG GGTCTGTGTC AAAGTAGGAG GCAGGCACAA Product Size (bp) 125 147 141 125 103 197 117 113 140 Table 2 (cont’d) RefSeq NM_000104 Gene name Entrez Gene ID CYP1B1 1545 ALDH1A3 220 SLC7A5 8140 CYP3A4 1576 NM_198253 cytochrome P450, family 3, subfamily A, polypeptide 4 telomerase reverse transcriptase TERT 7015 NM_002701 POU class 5 homeobox 1 5460 NM_001134 alpha-fetoprotein POU5F1 (OCT4) AFP NM_003380 vimentin VIM 7431 NM_006288 Thy-1 cell surface antigen THY1 7070 NM_000693 NM_003486 NM_017460 cytochrome P450, family 1, subfamily B, polypeptide 1 aldehyde dehydrogenase 1 family, member A3 solute carrier family member 5 Gene symbol 174 35 Forward Primer Reverse Primer CACCAAGGCT GAGACAGTGA GGTGGACCTG CTTACAGAGC AGGAGCCTTC CTTTCTCCTG CAAGACCCCTT TGTGGAAAA GCGTTTGGTGG ATGATTTCT GTACTCCTCGG TCCCTTTCC CTTGTGAAGCA AAAGCCACA CCCTCACCTGT GAAGTGGAT GGACTGAGAT CCCAGAACCA GATGACGACT GGGCCTACAT TCCACGGTATG CACTAACCA CTGCAAACCCT AAGGCAGAG CGAGGCGACT TTCTTTCATC CAGGGCCTCG TCTTCTACAG CAAAAACCCT GGCACAAACT CCCTCTTCAGC AAAGCAGAC GCTTCAACGG CAAAGTTCTC ACGAAGGCTC TGGTCCACTA Product Size (bp) 164 165 181 187 151 168 122 91 124 A B EcoRI MluI hTERT SnaBI BamHI SalI HL1‐1 SV40E GAG SV40 Large T‐antigen hTERT Auto induction L1SV1 hTERT HYGRO HLhT1 5’ LTR pBABE-HYGRO-hTERT (~8.7 kb) pUC ORI L1SV1: by SV40 large T-antigen transfection HLhT1: by hTERT (human telomerase reverse transcriptase) transfection 3’ LTR AMP Figure 2. Immortalization of HL1-1 cells. (A) Map of hTERT (human telomertase reverse transciptase) express plasmid (pBABE-HYGRO-hTERT) (B) HL1-1 cell immortalization strategy. L1SV1 cell: immortalized by SV40 large T-antigen transfection. HLhT1 cell: immortalized by hTERT transfection. Morphology of HL1-1 (C), HLhT1 (D), and L1SV1 cells (E). The cultures contain serpiginous shaped cells for HL1-1 and HLhT1 cells and cuboid shaped cells for SV1C1 cells. For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this dissertation. 36 Figure 2 (cont’d) C D E 37 A B PCR product NRs/HK Genes Ratio (Log) QRT-PCR quantification 1 0 2 3 4 5 6 -1 -2 -3 -4 -5 -6 AhR PPAR PPAR PXR ER C AhR (122 kDa) PPAR(55 kDa) PPAR(67 kDa) ER(66 kDa) Figure 3. Basal mRNA and protein expression levels of AhR, PPAR and , ER. Various nuclear receptors detected by QRT-PCR and in protein levels by western blot. (A) The mRNA levels of the receptors were measured by QRT-PCR and normalized to several housekeeping genes (HK) and (B) PCR products were verified by agarose gel. 1. 100bp ladder, 2. AhR, 3. PPAR, 4. PPAR, 5. PXR, 6. ER. (C) Protein levels were determined by Western blots. 38 Population Growth (cpdl) HL1-1 Growth Curve 55 K-NAC 1:1 1:1 + FN 50 45 40 35 30 25 0 20 40 60 80 100 Time of Subculture (days) Figure 4. Proliferation potential of HL1-1 cell. Cumulative pd of HL1-1 in K-NAC medium with 10% FBS (K-NAC), K-NAC:D (1:1) medium with 10% FBS (1:1) or 1:1 medium in fibronectin coated plate (1:1 + FN). 39 QRT-PCR results. The protein level of the AhR, ER, PPAR and  was confirmed by western blotting (Figure 3C). The PXR was not detected in western blotting because of the protein expression level was too low. For investigation of the AhR and PPAR-responsiveness of the HL1-1 cells, TCDD and Wy-14,643 treatments were performed and induction of CYP1A1 and ADFP, proto-typical regulated genes, were measured by QRT-PCR, respectively. The functional responsiveness results are reported in Chapter 4 and 5, respectively. PROLIFERATION POTENTIAL OF THE HL1-1 CELL AND THEIR IMMORTALIZATION Several studies report on expanding cell life span by changing media conditions [30, 31]. Three media conditions, K-NAC medium with 10% FBS (K-NAC), the 1:1 mixture of K-NAC and D medium with 10% FBS (1:1), and fibronectin coated flask and K-NAC:D (1:1) with 10% FBS (FN), were tested for culture condition optimization for HL1-1 cell life span expansion. In 112 days after five passages, the lifespan of HL1-1 cells reached 52.3, 52.0, and 53.4 cpdl from initial 26.5 cpdl in the K-NAC, 1:1, and FN medium conditions, respectively (Figure 4A). Application of different media conditions did not expand the life span of HL1-1 cells significantly. Previous reports have indicated that ectopic hTERT expression enables cells to circumvent senescence [32, 33]. For immortalization of HL1-1 cells, the cells were transfected with a plasmid carrying an hTERT cDNA (pBabe-hygro-hTERT) and selected with hygromycin (100 mg/ml). Ten days after the selection, surviving colonies were isolated and propagated. Among of them, the best proliferation ability and maintaining primitive cell morphology with high hTERT expression was selected (HLhT1). From continuous subculturing, HLhT1 cells maintain proliferation ability up to more than 6 months culturing with 18 passages without any 40 A B C D Figure 5. Morphology of HL1-1 and HLhT1 cells. (A) HL1-1 cells at low cell densities. (B) HLhT1 cells at low cell densities. (C) HL1-1 cells at high cell densities. (D) HLhT1 cells at high cell densities. 41 sign of senescence or morphological change (Figure 5). HLhT1 were compared with other three other cell lines, including parental HL1-1 cells, L1SV1 immortal liver cell line derived from HL1-1 transfected with SV40 large T-antigen and with human hepatoma HepG2 cells. Similar to the parental HL1-1 cells, immortalized HLhT1 and L1SV1 cells express comparable levels of stem cell marker (OCT4) and the oval cell markers (α-fetoprotein, AFP; vimentin, VIM; and Thy-1 cell surface antigen, THY1) (Figure 6). In contrast, compared to HepG2, HL1-1 cells and their immortalized derivatives (HLhT1 and L1SV1) showed lower expression of AFP, but higher expression of VIM and THY1. Basal expression levels of hTERT were the order of HLhT1, HepG2, L1SV1 and HL1-1 cells. Finally, expression of albumin (ALB), which is differentiated hepatocyte marker protein, was much higher in HepG2 cells compared to human liver stem cells. RECEPTOR GENE EXPRESSION AND RESPONSE COMPARISON IMMORTALIZED CELL For receptor-mediated response comparison between cell lines, basal expression levels of nuclear receptors and AhR were evaluated. The basal transcripts expression level of each receptor was measured in normal proliferating HL1-1, L1SV1 and HLhT1 cells by QRT-PCR (Figure 7A). The AhR expression level was comparable among all cells lines, the levels of PPAR were lower in L1SV1 and HLhT1 cells, and the expression of PPARwas higher in HLhT1 compared to parental HL1-1 cells. Basal expression of PXR was very low in all three cell lines. The AhR, PPAR and PXR receptor responsiveness was evaluated in HLhT1 cells using 10 nM TCDD, 50 µM Wy-14,643 and 30 µM rifampicin, respectively. After 12 h treatment, the induction of prototypical regulated genes, i.e. CYP1A1 (AhR) and ADFP (PPAR was ~50-, and 3.5-fold, respectively, as measured by QRT-PCR (Figure 7B). However, rifampicin did not 42 VIM AFP 6 log (Copy Number) 5 4 3 2 1 0 7 6 5 4 3 2 1 0 2 H ep G T1 Lh L1 H ep H H -1 2 G 1 H Lh T SV 1 L1 H L1 -1 -1 L1 SV 1 log (Copy Number) 7 Figure 6. Basal stem cell marker and hepatocyte differentiation marker gene expression level comparison. Stem cell marker (OCT4), liver oval cell marker (AFP, VIM and THY1), telomerase reverse transcriptase (hTERT) and hepatocyte differentiation marker (ALB) were examined by QRT-PCR and compared in HL1-1, L1SV1, HLhT1 and HepG2 cells. Bars are mean ± SE for the average log value of target gene copy number, N = 3. 43 ep G 2 0 -1 44 2 1 G 2 H ep 3 1 THY1 H Lh T 4 SV 1 5 log (Copy Number) 6 L1 H L1 -1 H H Lh T1 SV 1 L1 -1 L1 H log (Copy Number) Figure 6 (cont’d) OCT4 4 3 2 1 0 -1 -2 45 2 -1 G 0 ep 1 H 2 1 hTERT H Lh T 3 SV 1 4 log (Copy Number) 5 L1 -1 2 1 G L1 ep H H H Lh T SV 1 L1 L1 H log (Copy Number) Figure 6 (cont’d) ALB 7 6 5 4 3 2 1 0 -1 -2 Receptor/HK genes Ratio (Log) A 0 HL1-1 L1SV1 HLhT1 -1 * -2 * -3 -4 -5 -6 -7 AhR PPAR PPAR PXR Figure 7. Basal AhR and NRs expression levels comparison in parental and immortalized cell lines and receptors response in HLhT1 cells. (A) Comparison of basal mRNA expression levels of AhR, PPAR and , PXR in HL1-1, L1SV1 and HLhT1 cells. The gene expression ratio is represented by PPAR gene expression levels normalized to the housekeeping genes (HK) HPRT, GAPDH and ACTB. (B) Responsiveness of AhR, PPAR and PXR proteins in HLhT1 cells was determined by TCDD (CYP1A1), Wy-14,643 (ADFP) and rifampicin (CYP3A4) treatments, respectively. Bars are mean ± SE, N = 3. 46 Figure 7 (cont’d) Fold Change 60 CYP1A1 5 * 50 40 30 20 3 2 0 0 TCDD 6 Wy‐14,643 Rifampicin Treatment CYP3A4 5 Fold Change * 4 1 10 4 3 2 1 0 ADFP 6 70 Fold Change B TCDD Wy‐14643 Rifampicin Treatment 47 TCDD Wy‐14643 Rifampicin Treatment induce the PXR-regulated CYP3A4 expression. These results suggest that HLhT1 cells possess functional AhR and PPAR and are responsive to TCDD and Wy-14,643 treatment. TCDD-INDUCIBLE TEMPORAL GENE EXPRESSION PROFILE COMPARISON For AhR-mediated response comparison between cell lines, HL1-1, L1SV1 and HLhT1 cells were treated with 10 nM TCDD or DMSO vehicle and temporal changes in targeted gene expression were evaluated using QRT-PCR. Induction of CYP1A1, was confirmed all cell lines at all time points, however with different maximum responses between cells (L1SV1: ~27-fold, HL1-1: ~1,600-fold, HLhT1: ~5,000-fold) (Figure 8). In addition, the TCCD-regulated CYP1B1, SLC7A5 and ALDH1A3 were among the primary response genes with putative dioxin response elements (DREs) induced in the HL1-1 cells (discussed in Chapter 4). Temporal profile and induction level of these genes was comparable between immortalized HLhT and parental HL1-1 cells (Figure 9). DISCUSSION In the present study we characterized normal adult human liver stem cells (HL1-1) and their immortalized cell lines (L1SV1 and HLhT1) and evaluated their application for receptor mediated toxicogenomic studies. HL1-1 possesses stem cell characteristics such as high proliferation potential (cpdl = 49), multipotent differentiation and the ability of anchorage independent growth [8]. HL1-1 cells appear as serpiginous cells in morphology, especially when they are growing at low cell density or in growth factor-deprived medium, similar to cell morphologies observed in some stem cell from human bone marrow or precursor cells from rat and human pancreatic islets as well [34, 35]. Liver oval cells express vimentin, α-fetoprotein, thy-1 and the hematopoietic stem cell markers, CD34 and SCF/c-kit, which are not expressed in hepatocytes or bile duct cells [36]. HL1-1 express several mesenchymal and oval cell markers 48 (vimentin, α-fetoprotein and thy-1) and the embryonic stem cell marker (Oct-4) indicating a partial commitment to the hepatic lineage. The basal expression of AhR, ER, PPAR and  mRNA and protein levels was confirmed in HL1-1 cells. However, PXR protein was not detected and treatment of the prototypical human PXR agonist (rifampicin) [37], failed to induce CYP3A4 expression. PXR, generally regarded as a sensor activated by exogenous and endogenous chemicals, regulates a large number of enzymes involved in oxidation, conjugation and transport of drug, xenobiotic and endobiotic molecules [38]. Due to its physiological role, PXR activation can affect drugdrug interactions and result in altered clinical responses [39]. Deficiency of functional PXR protein in HL1-1 cells could limit their application in testing of hepatocyte related functions. Induction of PXR expression by differentiation has been previously reported in bipotential mouse embryonic liver cells [40]. Moreover, during mammalian hepatocyte differentiation, expression regulation of HNF1a and PXR by the transcription factor HNF4a has been reported [41, 42]. Accordingly, the PXR deficiency may be overcome by the generation of differentiated hepatocytes from liver stem cells. When growing in a modified Eagle's MEM (minimum essential medium) with high calcium (1.8 mM) and hepatocyte growth factor (HGF), most HL1-1 cells were morphologically changed into larger size with multiple nuclei and differentiated to albumin expressing hepatocytes [8]. The use of cells derived from monoclonal stem cell line would be beneficial compared to primary hepatocytes to minimize the donor and preparation variability. HL1-1 cells maintained their proliferative ability for more than 50 cumulative population doublings over 10 to 15 sub-passages. Such proliferative capacity indicates that adult stem cells isolated from a human liver tissue could potentially generate trillions of cells. 49 HL1-1 CYP1A1 900 600 300 * * * * * 30 * 20 * 10 * * * * Fold change 40 * Fold change Fold Change 1200 HLhT1 CYP1A1 L1SV1 CYP1A1 6000 5000 1500 * * 1000 * 500 * * 0 1T 2T 4T 8T 12T 24T 48T Time (hr) 0 1T 2T 4T 8T 12T 24T 48T Time (hr) 0 * 1T 2T 4T 8T 12T 24T 48T Time (hr) Figure 8. Temporal TCDD-mediated CYP1A1 induction profiles comparison. Temporal TCDD-mediated CYP1A1 induction profiles in human liver stem cells (HL1-1) and immortalized liver cells (L1SV1 and HLhT1) measured by QRT-PCR. All fold changes were calculated relative to time-matched vehicle controls. Bars are mean ± SE for the average fold change, *p < 0.05 vs. control, N = 3. 50 HL1-1 CYP1B1 HL1-1 SLC7A5 * 30 20 * * * 30 * * 20 * * * * 1T 2T 4T 8T 12T 24T 48T Time (hr) Fold change Fold change * 20 10 1T 2T 4T 8T 12T 24T 48T Time (hr) HLhT1 SLC7A5 14 12 10 8 6 4 2 0 20 40 0 * 30 0 1T 2T 4T 8T 12T 24T 48T Time (hr) HLhT1 CYP1B1 50 10 40 * 15 10 * * * 5 0 * * * * * * 1T 2T 4T 8T 12T 24T 48T Time (hr) HLhT1 ALDH1A3 6 * * * Fold change 0 * * Fold change 40 10 50 * Fold change Fold change 50 HL1-1 ALDH1A3 5 4 3 2 * * * * * 1 1T 2T 4T 8T 12T 24T 48T Time (hr) 0 1T 2T 4T 8T 12T 24T 48T Time (hr) Figure 9. Temporal profiles of prototypical TCDD-responsive genes in human liver stem cell (HL1-1) and immortalized liver cells (HLhT1). All fold changes were calculated relative to time-matched vehicle controls. Bars are mean ± SE for the average fold change, *p < 0.05 vs. control N = 3. 51 Although HL1-1 cells have high proliferation ability, they eventually become senescent. Normal somatic cells lose telomeric DNA at rates of 50 to 200 bp per cell division because of replication-associated attrition [43]. Critically shortened telomere which is incapable of looping leads to growth arrest signal activation and stops cell proliferation [44]. To overcome cellular senescence, genes related to cell cycle regulation and telomere maintenance were targeted for cell engineering. Human or primate hepatocytes have been previously immortalized using Simian virus 40 large T-antigen (SV40T) and the catalytic subunit of the telomerase (hTERT) and these immortalized cells are non-tumorigenic and express hepatocyte genes [45-47]. By means of transfection of a pBABE-hygro retroviral vector expressing hTERT, we isolated a transduced clone referred to as HLhT1 which grows as a continuous cell line in vitro and can be amplified and cryopreserved efficiently. QRT-PCR analysis of hTERT expression in HLhT1 cells demonstrated stable proviral integration. Prototypical gene responses in HLhT1 cells with transactivation of AhR and PPAR showed concordant responses with their parental cells. Hence, immortalized HLhT1 cells harbor a phenotype expected from the parental HL1-1 liver stem cells. Compared to HLhT1 cells, L1SV1 cells immortalized by SV40T transduction appear morphologically different and exhibit lower level of CYP1A1 induction by TCDD treatment. SV40T is viral oncoprotein which targets, binds and inactivates tumor suppressor p53 and Rb proteins and drive cells into the cell cycle [48, 49]. The stabilization of p53 and subsequent transcription of p53-dependent genes results in cell cycle arrest and/or apoptosis. T antigen blocks this response by binding to p53 and preventing p53-dependent transcription in downstream of p53 pathway [50]. As a result, although infinite cell replication has been induced by expressing the SV40T, genetic transformation will lead to greater susceptibility for cancer by 52 blocking tumor suppressor p53 activity [46, 51]. Compared to SV40T, hTERT transduced immortalization has the advantage of being untransformed and non-tumorgenic after transplantation [47, 52, 53]. Telomerase reconstitution does not induce a transformed phenotype and is likely to be genoprotective throughout the expanded proliferative lifespan of immortalized cells. There is increasing evidence that adult stem cells and their derivatives are targets for carcinogenesis [12, 13]. Oval cells are also involved in human liver disease condition caused by alcohol, hepatitis C virus and hemochromatosis which are also associated with increased incidence of hepatocellular carcinoma or cholangiocarcinoma [54, 55]. There is also evidence that oval cells give rise to hepatocellular carcinoma in mouse as well [56]. Since liver stem cells could be a potential direct target for carcinogenesis, these cells may be used to develop an in vitro model to study the mechanism of human liver carcinogenesis. In summary, these results provide a novel in vitro model system for receptor-mediated toxicogenomic study employing human liver stem cells isolated from adult liver tissue. Human liver stem cells are an attractive alternative cell model, due to the high proliferative capacity, their intactness from normal liver tissue origin and differentiation ability. Furthermore, this study provides insight into cellular engineering approaches for the immortalization of human liver stem cells using hTERT transduction and their comparable AhR-mediated responsiveness with parental cells for extension of utilization. Further comparative studies with novel human liver stem cell model and conventional models may help elucidate the differences in the mechanisms of action of species-specific responses. 53 REFERENCES 54 REFERENCES 1. Abbott, A., Cell culture: biology's new dimension. Nature, 2003. 424(6951): p. 870-2. 2. Wilkening, S., F. Stahl, and A. Bader, Comparison of primary human hepatocytes and hepatoma cell line Hepg2 with regard to their biotransformation properties. Drug Metab Dispos, 2003. 31(8): p. 1035-42. 3. Rohwedel, J., et al., Embryonic stem cells as an in vitro model for mutagenicity, cytotoxicity and embryotoxicity studies: present state and future prospects. Toxicol In Vitro, 2001. 15(6): p. 741-53. 4. Davila, J.C., et al., Use and application of stem cells in toxicology. Toxicol Sci, 2004. 79(2): p. 214-23. 5. Alison, M.R., S. Islam, and S. Lim, Stem cells in liver regeneration, fibrosis and cancer: the good, the bad and the ugly. J Pathol, 2009. 217(2): p. 282-98. 6. Michalopoulos, G.K. and M.C. DeFrances, Liver regeneration. Science, 1997. 276(5309): p. 60-6. 7. Conigliaro, A., et al., Isolation and characterization of a murine resident liver stem cell. Cell Death Differ, 2008. 15(1): p. 123-133. 8. Chang, C.C., Tsai,J.L., Kuo,K.K., Wang,K.H., Chiang,C.H., Kao,A.P., Tai,M.H. and Trosko,J.E., Expression of Oct-4, alpha fetoprotein and vimentin and lack of gapjunctional intercellular communication (GJIC) as common phenotypes for human adult liver stem cells and hepatoma cells. Proc. Am. Assoc. Cancer Res., 2004. 45: p. 642. 9. Masson, N.M., et al., Hepatic progenitor cells in human fetal liver express the oval cell marker Thy-1. Am J Physiol Gastrointest Liver Physiol, 2006. 291(1): p. G45-54. 10. Malhi, H., et al., Isolation of human progenitor liver epithelial cells with extensive replication capacity and differentiation into mature hepatocytes. J Cell Sci, 2002. 115(Pt 13): p. 2679-88. 11. Suzuki, A., et al., Clonal identification and characterization of self-renewing pluripotent stem cells in the developing liver. J Cell Biol, 2002. 156(1): p. 173-84. 12. Ahuja, Y.R., V. Vijayalakshmi, and K. Polasa, Stem cell test: a practical tool in toxicogenomics. Toxicology, 2007. 231(1): p. 1-10. 13. Alison, M.R. and M.J. Lovell, Liver cancer: the role of stem cells. Cell Prolif, 2005. 38(6): p. 407-21. 55 14. Moore, J.T., J.L. Collins, and K.H. Pearce, The nuclear receptor superfamily and drug discovery. ChemMedChem, 2006. 1(5): p. 504-23. 15. Woods, C.G., J.P. Heuvel, and I. Rusyn, Genomic profiling in nuclear receptor-mediated toxicity. Toxicol Pathol, 2007. 35(4): p. 474-94. 16. Dreyer, C., et al., Control of the peroxisomal beta-oxidation pathway by a novel family of nuclear hormone receptors. Cell, 1992. 68(5): p. 879-87. 17. Kretschmer, X.C. and W.S. Baldwin, CAR and PXR: xenosensors of endocrine disrupters? Chem Biol Interact, 2005. 155(3): p. 111-28. 18. Timsit, Y.E. and M. Negishi, CAR and PXR: the xenobiotic-sensing receptors. Steroids, 2007. 72(3): p. 231-46. 19. Matic, M., et al., Pregnane X receptor: promiscuous regulator of detoxification pathways. Int J Biochem Cell Biol, 2007. 39(3): p. 478-83. 20. Okey, A.B., An aryl hydrocarbon receptor odyssey to the shores of toxicology: the Deichmann Lecture, International Congress of Toxicology-XI. Toxicol Sci, 2007. 98(1): p. 5-38. 21. Denison, M.S. and S. Heath-Pagliuso, The Ah receptor: a regulator of the biochemical and toxicological actions of structurally diverse chemicals. Bull Environ Contam Toxicol, 1998. 61(5): p. 557-68. 22. Knerr, S. and D. Schrenk, Carcinogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in experimental models. Mol Nutr Food Res, 2006. 50(10): p. 897-907. 23. Kim, S., et al., Comparative analysis of AhR-mediated TCDD-elicited gene expression in human liver adult stem cells. Toxicol Sci, 2009. 112(1): p. 229-44. 24. Smith, J., et al., Redox state is a central modulator of the balance between self-renewal and differentiation in a dividing glial precursor cell. Proc Natl Acad Sci U S A, 2000. 97(18): p. 10032-7. 25. Chepda, T., et al., Monitoring of ascorbate at a constant rate in cell culture: effect on cell growth. In Vitro Cell Dev Biol Anim, 2001. 37(1): p. 26-30. 26. Inoue, C., et al., Nicotinamide prolongs survival of primary cultured hepatocytes without involving loss of hepatocyte-specific functions. J Biol Chem, 1989. 264(9): p. 4747-50. 27. Mitaka, T., The current status of primary hepatocyte culture. Int J Exp Pathol, 1998. 79(6): p. 393-409. 28. Rozen, S. and H. Skaletsky, Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol, 2000. 132: p. 365-86. 56 29. Vandesompele, J., et al., Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol, 2002. 3(7): p. RESEARCH0034. 30. Maheshwari, G., et al., Cell adhesion and motility depend on nanoscale RGD clustering. J Cell Sci, 2000. 113 ( Pt 10): p. 1677-86. 31. Neupane, M., et al., Isolation and characterization of canine adipose-derived mesenchymal stem cells. Tissue Eng Part A, 2008. 14(6): p. 1007-15. 32. Bodnar, A.G., et al., Extension of life-span by introduction of telomerase into normal human cells. Science, 1998. 279(5349): p. 349-52. 33. Counter, C.M., et al., Dissociation among in vitro telomerase activity, telomere maintenance, and cellular immortalization. Proc Natl Acad Sci U S A, 1998. 95(25): p. 14723-8. 34. Zulewski, H., et al., Multipotential nestin-positive stem cells isolated from adult pancreatic islets differentiate ex vivo into pancreatic endocrine, exocrine, and hepatic phenotypes. Diabetes, 2001. 50(3): p. 521-33. 35. Lin, T.M., et al., Accelerated growth and prolonged lifespan of adipose tissue-derived human mesenchymal stem cells in a medium using reduced calcium and antioxidants. Stem Cells Dev, 2005. 14(1): p. 92-102. 36. Alison, M., Liver stem cells: a two compartment system. Curr Opin Cell Biol, 1998. 10(6): p. 710-5. 37. Bolt, H.M., Rifampicin, a keystone inducer of drug metabolism: from Herbert Remmer's pioneering ideas to modern concepts. Drug Metab Rev, 2004. 36(3-4): p. 497-509. 38. Jones, S.A., et al., The pregnane X receptor: a promiscuous xenobiotic receptor that has diverged during evolution. Mol Endocrinol, 2000. 14(1): p. 27-39. 39. Ma, X., J.R. Idle, and F.J. Gonzalez, The pregnane X receptor: from bench to bedside. Expert Opin Drug Metab Toxicol, 2008. 4(7): p. 895-908. 40. Ochsner, S.A., et al., Transcriptional profiling of bipotential embryonic liver cells to identify liver progenitor cell surface markers. Stem Cells, 2007. 25(10): p. 2476-87. 41. Li, J., G. Ning, and S.A. Duncan, Mammalian hepatocyte differentiation requires the transcription factor HNF-4alpha. Genes Dev, 2000. 14(4): p. 464-74. 42. Kamiya, A., F.J. Gonzalez, and H. Nakauchi, Identification and differentiation of hepatic stem cells during liver development. Front Biosci, 2006. 11: p. 1302-10. 43. Allsopp, R.C., et al., Telomere length predicts replicative capacity of human fibroblasts. Proc Natl Acad Sci U S A, 1992. 89(21): p. 10114-8. 57 44. Hoare, M., T. Das, and G. Alexander, Ageing, telomeres, senescence, and liver injury. J Hepatol, 2010. 53(5): p. 950-61. 45. Allain, J.E., et al., Immortalization of a primate bipotent epithelial liver stem cell. Proc Natl Acad Sci U S A, 2002. 99(6): p. 3639-44. 46. Pfeifer, A.M., et al., Simian virus 40 large tumor antigen-immortalized normal human liver epithelial cells express hepatocyte characteristics and metabolize chemical carcinogens. Proc Natl Acad Sci U S A, 1993. 90(11): p. 5123-7. 47. Wege, H., et al., Telomerase reconstitution immortalizes human fetal hepatocytes without disrupting their differentiation potential. Gastroenterology, 2003. 124(2): p. 432-44. 48. Manfredi, J.J. and C. Prives, The transforming activity of simian virus 40 large tumor antigen. Biochim Biophys Acta, 1994. 1198(1): p. 65-83. 49. Pipas, J.M., SV40: Cell transformation and tumorigenesis. Virology, 2009. 384(2): p. 294-303. 50. Giono, L.E. and J.J. Manfredi, The p53 tumor suppressor participates in multiple cell cycle checkpoints. J Cell Physiol, 2006. 209(1): p. 13-20. 51. Farwell, D.G., et al., Genetic and epigenetic changes in human epithelial cells immortalized by telomerase. Am J Pathol, 2000. 156(5): p. 1537-47. 52. Jiang, X.R., et al., Telomerase expression in human somatic cells does not induce changes associated with a transformed phenotype. Nat Genet, 1999. 21(1): p. 111-4. 53. Morales, C.P., et al., Absence of cancer-associated changes in human fibroblasts immortalized with telomerase. Nat Genet, 1999. 21(1): p. 115-8. 54. Alison, M.R., Liver stem cells: implications for hepatocarcinogenesis. Stem Cell Rev, 2005. 1(3): p. 253-60. 55. Lowes, K.N., et al., Oval cell numbers in human chronic liver diseases are directly related to disease severity. Am J Pathol, 1999. 154(2): p. 537-41. 56. Dumble, M.L., et al., Generation and characterization of p53 null transformed hepatic progenitor cells: oval cells give rise to hepatocellular carcinoma. Carcinogenesis, 2002. 23(3): p. 435-45. 58 CHAPTER 4 Kim S, Dere E, Burgoon LD, Chang CC, and Zacharewski TR: Comparative analysis of AhRmediated TCDD-elicited gene expression in human liver adult stem cells. Toxicol Sci 2009, 112: 229-44. 59 CHAPTER 4 COMPARATIVE ANALYSIS OF AHR-MEDIATED TCDD ELICITED GENE EXPRESSION IN HUMAN LIVER ADULT STEM CELLS ABSTRACT Time course and dose-response studies were conducted in HL1-1 cells, a human liver cell line with stem cell-like characteristics, to assess the differential gene expression elicited by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) compared to other established models. Cells were treated with 0.001, 0.01, 0.1, 1, 10 or 100 nM TCDD or DMSO vehicle control for 12 hrs for the dose response study, or with 10 nM TCDD or vehicle for 1, 2, 4, 8, 12, 24 or 48 hrs for the time course study. Elicited changes were monitored using a human cDNA microarray with 6,995 represented genes. Empirical Bayes analysis identified 144 genes differentially expressed at one or more time points following treatment. Most genes exhibited dose-dependent responses including CYP1A1, CYP1B1, ALDH1A3 and SLC7A5 genes. Comparative analysis of HL1-1 differential gene expression to human HepG2 data identified 74 genes with comparable temporal expression profiles including 12 putative primary responses. HL1-1 specific changes were related to lipid metabolism and immune responses, consistent with effects elicited in vivo. Furthermore, comparative analysis of HL1-1 cells with mouse Hepa1c1c7 hepatoma cell lines and C57BL/6 hepatic tissue identified 18 and 32 commonly regulated orthologous genes, respectively, with functions associated with signal transduction, transcriptional regulation, metabolism and transport. Although some common pathways are affected, the results suggest that TCDD elicits species- and model-specific gene expression profiles. 60 INTRODUCTION 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is a ubiquitous environmental contaminant that elicits a broad spectrum of responses including endocrine disruption, immunotoxicity, hepatotoxicity, teratogenesis, and tumor promotion [1]. It is a multi-site carcinogen and liver tumor promoter in rodents [2], and classified as a human carcinogen by the International Agency for Research on Cancer (IARC) [3]. The effects are mediated by the aryl hydrocarbon receptor (AhR) [4], a ligand activated basic helix-loop-helix/PER-ARNT-SIM (bHLH/PAS) transcription factor. Upon ligand binding, the cytosolic AhR undergoes a conformational change, which leads to the dissociation of chaperone proteins and translocation to the nucleus where it heterodimerizes with the aryl hydrocarbon receptor translocator (ARNT). The heterodimer complex binds to dioxin response elements (DREs) in the regulatory region of target genes to modulate gene expression. AhR-mediated changes in gene expression are believed to account for the toxicity of TCDD and related compounds. In vivo and in vitro models have been used to investigate the mechanisms of action of TCDD and to assess its potential toxicity to humans and other environmentally relevant species. However, the relevance of data extrapolation from experimental models to target species has been questioned [5]. In general, minimizing the extrapolation between experimental models and the relevant species is expected to more accurately assess the potential toxicity to the target of concern. Although cell lines model may reflect some in vivo responses, their utility for predicting in vivo toxicity is limited. Nevertheless, human based in vitro models are expected to more accurately indicate potential in vivo human toxicity. For example, human based models are preferred in drug development to investigate the mechanisms of action and assess potential toxicities of drug candidates [6-9]. Immortalized cell lines, primary cells and organ slices have 61 been used as drug screening tools [10] and to assess potential mechanisms, including toxicanttarget interaction and dose/time dependent responses [7]. However, while continuous cell lines are convenient, they are transformed or derived from diseased tissue, which may not accurately reflect normal tissue responses [6]. In contrast, the availability, use, cost, safety, and stability of primary cells and human tissue continues to be a factor. Human stem cells provide an attractive alternative and potentially unlimited source of normal cells [11, 12]. They are defined by their self-renewal and differentiation capabilities, and classified as either embryonic or adult stem cells based on their developmental status. Normal adult human stem cells isolated from liver tissue is an alternative model for toxicity studies, that may more closely mimic human tissue [11, 13, 14]. Stem cells are also amenable to high throughput screening to rank and prioritize chemicals and drug candidates that warrant further investigation or development [12, 15, 16]. HL1-1 cells were derived from adult normal liver tissue. They exhibit high proliferation potential, express stem cell (Oct-4) and liver oval cell markers (AFP, vimentin and Thy-1), and have the ability to differentiate into hepatocytes [17]. In this study, HL1-1 cells were used to investigate the time- and dose-dependent gene expression changes elicited by TCDD. HL1-1 gene expression data were also compared to other in vitro (HepG2 and Hepa1c1c7 cells) and in vivo (C57BL/5 hepatic tissue) data sets treated with TCDD using comparable study designs and data analysis approaches. Although some common pathways are affected, the results suggest that TCDD elicits species- and model- specific gene expression profiles. 62 MATERIALS & METHODS DERIVATION OF HL1-1 HUMAN LIVER CELL LINE HL1-1 human liver cell line was derived from a normal healthy liver section obtained during the surgical resection of a male (age 49) with hemangioma. This cell line has been previously shown to possess liver stem cell characteristics [17, 18] including, (1) high selfrenewal ability, cumulative population doubling level (cpdl) about 50, (2) deficiency in gap junctional intercellular communication, (3) expression of Oct 4, and liver stem cell markers, alpha-fetoprotein, vimentin and Thy-1, and (4) the ability to differentiate into albumin producing cells. The morphology and proliferation potential of HL1-1 cells are shown in Figure 10. The finite lifespan of HL1-1 cells and the non-tumorigenic nature of HL1-1 cells immortalized by 6 SV40 large T-antigen (no tumor developed in 6 mice injected with 4 x 10 cells at 2 sites per mouse, our unpublished results) indicate that HL1-1 cells are normal non-tumorigenic cells. The cells at approximately passage 8 with cumulative population doubling level of about 40 were used for this study. CULTURE AND TREATMENT OF CELL LINE HL1-1 cells [17] were maintained in a low calcium (0.09mM) modified MCDB 153 medium (Keratinocyte-SFM, Invitrogen Corporation, Carlsbad, CA) supplemented with 2 mM N-acetyl-L-cysteine, 0.2 mM L-ascorbic acid 2-phosphate and 5 mM nicotinamide (referred to as K-NAC medium). The medium contains rEGF (5 ng/mL), bovine pituitary extract (50 ug/mL) 6 and 10% fetal bovine serum (FBS) (Hyclone, Logan, UT). 1  10 cells were seeded into vented 2 25 cm cell culture flasks (Corning Inc., Corning, NY) and incubated under standard conditions (5% CO2, 37C). For time course studies, cells were treated with 10 nM TCDD (S. Safe, Texas 63 A&M University, College Station, TX) or DMSO (Sigma, St. Louis, MO) and harvested at 1, 2, 4, 8, 12, 24 or 48 hrs (Figure 11). For dose-response studies HL1-1 cells were treated with DMSO (vehicle control) or TCDD (0.001, 0.01, 0.1, 1, 10, 100 nM) for 12 hrs (Figure 11). For the cycloheximide (CHX) study, HL1-1 cells were divided into four treatment groups: (1) 10 nM TCDD, (2) DMSO as a vehicle control, (3) pre-treatment with 10 µg/mL CHX(Sigma-Aldrich, St. Louis, MO) for 1 hr prior to treatment with DMSO, or (4) pre-treated with 10 µg/mL CHX for 1 hr prior to treatment with 10 nM TCDD, and then harvested at 4 and 12 hrs (Figure 11). PROTEIN PREPARATION AND WESTERN BLOT Cell lysates were prepared in RIPA buffer (1x PBS, 0.1% SDS, 1% IGEPAL CA-630 and 0.5% Na-Deoxycholate) containing a protease inhibitor cocktail (Roche Diagnostics, Mannheim, Germany). Protein concentrations were measured with a modified Lowry assay (Bio-Rad, DC Protein Assay, Hercules, CA). Cell lysates (20 ug) were separated on a 10% SDS-PAGE, transferred onto nitrocellulose membranes (Amersham Biosciences Inc., Piscataway, NJ), and probed with anti-human AhR (N-19) antibody followed by horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology Inc., Santa Cruz, CA). The blot was imaged by immunochemical staining and fluorescence detection on X-ray film by the ECL method (Amersham Biosciences Inc.). RNA ISOLATION All treatment groups were harvested in 1.0 mL Trizol Reagent (Invitrogen) and stored at o 80 C. Total RNA was isolated according to the manufacturer’s protocol, resuspended in RNA Storage Solution (Ambion Inc., Austin, TX), and quantified spectrophotometrically (A260). The purity and integrity of each sample was assessed by the A260/A280 ratio and gel electrophoresis. 64 A B Cumulative Population Doublings Cpdl of HL1-1 60 50 40 30 20 10 0 0 50 100 150 200 Days in culture Figure 10. The proliferation potential and morphology of HL1-1 cell. (A) The proliferation potential of HL1-1 cell clone as determined by cumulative population doubling level (cpdl) in continual growth 5 and subculture from 1x10 cells. (B) HL1-1 cell morphology. 65 MICROARRAY EXPERIMENTAL DESIGN Gene expression changes were measured on custom human cDNA arrays containing 9,684 features representing 6,995 unique genes. For the time course study, an independent reference study design was used that included three replicates. Time-matched TCDD treated and vehicle samples were co-hybridized at each time point (Figure 12A). Dose-dependent changes in gene expression were analyzed using a spoke design with three replicates at 12 hrs (Figure 12B). Dye swaps were also performed for both time course and dose response studies to account for dye biases. For CHX studies, a 2X2 factorial design was used to facilitate appropriate statistical comparisons between all four treatment groups (Figure 12C) [19]. If TCDD-elicited gene response were sustained or enhanced by cycloheximide co-treatment, a direct effect independent of down stream translational activities was inferred and classified as a putative primary response. However, if the gene expression response was attenuated by CHX, it was assumed that the response was dependent on protein production(s), which were blocked by CHX, and therefore classified as putative secondary responses. Candidate genes that passed the first filter were further investigated by comparing CHX treatment alone to vehicle and co-treatment compared to CHX alone in order to exclude unclassified genes whose change in expression was affected by CHX. ANALYSIS OF DIFFERENTIAL GENE EXPRESSION PCR amplified cDNAs were robotically spotted onto epoxy-coated slides (SchottNexterion, Duryea, PA) by an Omnigrid arrayer (GeneMachines, San Carlos, CA) equipped with Chipmaker 3 pins in a CHP3 printhead (Telechem International Inc., Sunnyvale, CA) at the Research Technology Support (http://www.genomics.msu.edu). Facility, Michigan State University Selected clones were obtained from EPAMAC, Research 66 CHX* -1 Dosed with TCDD or DMSO vehicle 0 1 2 4* 8 12†* 48 (hr) 24 Cell Harvest † : Dose response study * : CHX co-treatment study Figure 11. HL1-1 TCDD time course, dose response and cycloheximide (CHX) study designs. For the time course study, HL1-1 cells were treated with either 10 nM TCDD or 0.1% DMSO and harvested at 1, 2, 4, 8, 12, 24, or 48 hrs post-treatment. For the dose response study, HL1-1 cells were treated with 0.001, 0.01, 0.1, 1, 10, 100 nM TCDD or 0.1 % DMSO vehicle and harvested 12 hrs post-treatment (as indicated †). For the CHX study, 10 g/mL CHX was treated 1 hr prior to 10 nM TCDD or 0.1% DMSO treatment. Treatment groups were harvested at 4 and 12 hrs post-treatment (as indicated *). Three replicates were conducted in all studies. 67 A 1T 2T 4T 8T 12T 24T 48T 1V 2V 4V 8V 12V 24V 48V B C V T CHX 100nM CHX +T 10nM 0.001nM V6 V1 V5 V2 V4 V3 0.01nM 1nM 0.1nM Figure 12. Microarray experimental designs. Microarray experimental designs for (A) HL1-1 TCDD time course, (B) dose response and (C) CHX study. (A) Temporal gene expression changes were analyzed using an independent reference design that results in two independent labeling of each sample. Numbers indicate time of cell harvesting (hours), T indicate TCDD treatment and V indicate DMSO vehicle treatment. (B) Dose response gene expression changes were analyzed using a spoke design. Each dose treatment sample was compared with independent vehicle control. (C) CHX gene expression changes were analyzed using a 2x2 factorial design. This design allows for multiple comparisons to identify significant changes in gene expression between treatments. Each arrow represents a single microarray where arrow heads represent Cy5-labeled samples and arrow tails represent Cy3-labeled samples and double headed arrows indicate dye swap labeled on different arrays. 68 Genetics and Van Andel Research Institute. Detailed protocols for processing of microarrays including the labeling of the cDNA probe are available at http://dbzach.fst.msu.edu/interfaces /microarray.html. Briefly, the Genisphere 900 3DNA Array Detection (Genisphere Inc., Hatfield, PA) indirect incorporation kit was used to generate cDNA samples for hybridization according to manufacturer’s protocol. After 20 hrs of cDNA hybridization, slides were washed and rehybridized with a Cy3:Cy5 (1:1) dendrimer mixture to indirectly incorporate dyes at the Cy3and Cy5-dendrimer-tagged cDNA for 16 hrs. Slides were then scanned at 635 nm (Cy3) and 532 nm (Cy5) using a GenePix 4000B Array Scanner (Molecular Devices, Union City, CA). Images were analyzed for feature and background intensities using GenePix Pro 6.1 (Molecular Devices). MICROARRAY DATA NORMALIZATION AND ANALYSIS Data were normalized using a semi-parametric approach [20]. Empirical Bayes analysis was used to calculate posterior probabilities (P1(t) value) of expression change on a per gene and time point or dose group basis using the model-based t-value [21]. The data were filtered using a P1(t) and fold change to obtain the most reproducible differentially expressed genes for initial analysis and interpretation. All raw and normalized data were stored in the toxicogenomic information management system (TIMS) dbZach which supports microarray data management, mining, visualization and knowledge management [22, 23]. Expression changes that passed the criteria were analyzed by hierarchical clustering (GeneSpring 6.0, Agilent Technologies Inc., Santa Clara, CA and Multiexperiment Viewer (MeV) in TM4 software [24]) using uncentered Pearson correlation with average linkage. Normalization and empirical Bayes analysis were performed using SAS 9.1 (SAS Institute, Cary, NC) and R 2.0.1 (http://www.r-project.org). 69 Dose response analysis was performed using Graph Pad Prism 4.0 (GraphPad Software, San Diego, CA). Functional categorization of genes was performed using Database for Annotation, Visualization and Integrated Discovery (DAVID) analysis as a gene ontology tool [25]. QUANTITATIVE REAL-TIME PCR (QRT-PCR) ANALYSIS QRT-PCR was performed for a selected number of genes to verify microarray data. Total RNA (1.5 g) was reverse transcribed by SuperScript II according to the manufacturer’s protocol (Invitrogen). cDNA products were then amplified with gene specific primers designed using Primer3 [26] and SYBR Green PCR reaction mixture (Applied Biosystems, Foster City, CA) on an Applied Biosystems PRISM 7500 Sequence Detection System. Input copy number was quantified using a standard curve of log copy number versus threshold cycle (Ct). The copy number of each sample was standardized to the geometric mean of -actin (ACTB) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) to control for differences in RNA loading, quality, and cDNA synthesis. For graphing purposes, the relative gene expression levels were scaled such that the expression level of the time-matched vehicle treated control group was equal to 1. Official names and abbreviations, forward and reverse primer sequences, and product length are listed in Table 3. IDENTIFICATION OF DRES Intergenic regions 10kb upstream of RefSeq annotated transcription start sites (TSS) were obtained from the University of California Santa Cruz (UCSC) Genome Browser and deposited into dbZach [23]. Core DRE sequences (5’-GCGTG-3’) were computationally identified, extended by the flanking 7 base pairs and scored using a position weight matrix (PWM) application [27]. The PWM was developed using experimentally verified dioxin response 70 elements (DREs); DRE sequences are from genome sequence. Putative DREs are those with a matrix similarity score greater than 0.80 [28-30]. RESULTS EXPRESSION AND FUNCTIONALITY OF AHR IN HL1-1 CELLS AhR mRNA was detected in HL1-1 cells, C57BL/6 hepatic tissue, human HepG2 and mouse Hepa1c1c7 cells (Figure 13A). AhR protein expression was confirmed by Western blot (Figure 13B). Treatment of HL1-1 cells with TCDD resulted in the dose-dependent induction (>600-fold) of CYP1A1 mRNA with an EC50 of 8.30 nM (Figure 14), which is less sensitive compared to other human in vitro models (HepG2: 0.68 nM [31], HepG2: 0.2 nM and fresh hepatocytes: 0.14 nM [32]). There was no indication of TCDD treatment mediated differences with respect to growth or physical appearance. Collectively, these results confirm the functionality and responsiveness of HL1-1 AhR to TCDD. MICROARRAY ANALYSIS OF TCDD INDUCIBLE DOSE AND TEMPORAL GENE EXPRESSION PROFILES Temporal and dose-dependent changes in gene expression were determined using custom human cDNA microarrays consisting of 9,684 features representing 6,995 unique genes. Differentially expressed genes were identified using P1(t) values greater than 0.9999 and |fold change| > 1.5 as the criteria. 113 unique genes (72 induced; 41 repressed) exhibited dose dependent regulation with a majority exhibiting differential expression between 1 - 100 nM TCDD (Figure 15A). EC50s ranged from 0.18 nM for ACSL3 to 37.3 nM for MOBP, while induction and repression ranged from 40- to -2-fold for genes CYP1B1 and GALNT1, respectively. 71 Table 3. Gene names and primer sequences for QRT-PCR RefSeq  Gene name  Human  NM_001101  actin, beta  Gene  symbol  Entrez  Gene ID  ACTB  60  AGTGGGGTGGCTTTTA GGAT  AGGGGTCTACATGGCA ACTG  TGGTGCCCAGAATAAT GTGA  AAAGTAGGAGGCAGG CACAA  GATGACGACTGGGCCT ACAT  TCCACGGTATGCACTA ACCA  CTGCAAACCCTAAGGC AGAG  Actb 11461 Gapdh  14433  Ahr  11622  TCTCCAGGGAGGAAG AGGAT  TGTGAGGGAGATGCTC AGTG  TCTGAGGTGCCTGAAC TCCT  GAPDH  NM_009992  cytochrome P450, family 1,  subfamily A, polypeptide 1  NM_000104  cytochrome P450, family 1,  subfamily B, polypeptide 1   NM_000693 aldehyde dehydrogenase 1  family, member A3   NM_003486  solute carrier family 7 (cationic  amino acid transporter, y+  system), member 5   CYP1A1  actin, beta, cytoplasmic NM_008084  glyceraldehyde‐3‐phosphate  dehydrogenase  NM_013464  aryl‐hydrocarbon receptor  Reverse Primer  CATCCCCCAAAGTTCA CAAT   2597  GGCCTCCAAGGAGTAA GACC  157873  TGTTGGACGTCAGCAA GTTC  13076  AAGTGCAGATGCGGTC TTCT  1545  CACCAAGGCTGAGACA GTGA  220 GGTGGACCTGCTTACA GAGC  8140  AGGAGCCTTCCTTTCTC CTG  NM_002046  glyceraldehyde‐3‐phosphate  dehydrogenase  NM_001621  aryl hydrocarbon receptor   Mouse  NM_007393 Forward Primer  AHR  CYP1B1   ALDH1A3  SLC7A5   72 GCTACAGCTTCACCAC CACA  GTGGACCTCATGGCCT ACAT  ACCAGAACTGTGAGG GTTGG  Product  Size (bp)  125  147  197  140  164  165 181    123 125  155  A B AhR/HK genes Ratio 0.10 HL1-1 0.08 HepG2 Hepa 1c1c7 Mouse liver 0.06 0.04 0.02 0.00 HL1-1 HepG2 Hepa1c1c7 Mm liver Figure 13. Basal AhR mRNA and protein expression in HL1-1 cells. (A) AhR mRNA levels were measured by QRT-PCR and normalized to several housekeeping (HK) genes. HL1-1: human liver stem cell, HepG2: human hepatoma cell, Hepa1c1c7: mouse hepatoma cell, and Mm liver: C57BL/6 mouse liver. Error bars represent the SEM for the average. (B) AhR protein expression in HL1-1 cells, HepG2 cells, Hepa1c1c7 cells and C57BL/6 mouse liver detected by Western analysis. The mw is estimated to be 112 and 95 kDa for human and rodent AhR, respectively. 73 A B 1000 800 600 Fold Induction Fold Induction 1250 EC50 = 8.30nM 400 200 0 10 -4 10 -3 10 -2 10 -1 10 0 10 1 * 1000 * 750 * * 500 * * 250 0 10 2 1 TCDD (nM) 2 4 8 12 24 48 Time (hr) Figure 14. CYP1A1 expression induction by TCDD in HL1-1 cells. QRT-PCR verification of CYP1A1 mRNA levels from the dose response (12 hr) (A) and time course (10 nM TCDD) (B) studies in HL1-1 cells treated with TCDD. The EC50 value for CYP1A1 expression was 8.3 nM. Error bars represent the SEM for the average fold change. The asterisk (*) indicates p < 0.05. 74 In the time course study, 144 unique genes were differentially regulated by TCDD at one or more time points (Figure 15B). Hierarchical clustering identified induced and repressed expression with three distinct, temporal clusters of early (2-4 hrs), mid (8-12 hrs), and late (24-48 hrs) responses (Figure 16). Early and mid time point groups showed vast differences in their expression profiles, illustrating a temporal cascade of responses. A subset of responsive genes including CYP1B1, ALDH1A3 and SLC7A5 was verified by QRT-PCR (Figure 17). There was good agreement between the microarray and QRT-PCR data for the temporal expression profiles with comparable EC50 values. IDENTIFICATION OF PUTATIVE TCDD PRIMARY RESPONSE GENES FROM CHX STUDIES HL1-1 cells were pretreated with CHX to inhibit de novo protein synthesis in order to identify putative primary gene expression responses mediated by the AhR. Following treatment 78 and 203 differentially expressed genes (P1(t) > 0.9999 and |fold change| > 1.5) were identified at 4 and 12 hrs, respectively (Figure 18). These genes were classified into putative primary, secondary or unclassified groups. 47 genes at 4 hrs and 53 genes at 12 hrs were putatively determined to be primary responses. A total of 78 unique genes were identified as putative primary responses with 22 genes in common at both time points. Putative primary responses were associated with xenobiotic and lipid metabolism, cell cycle regulation, transcription regulation, transport and signal transduction (Table 4). This included the prototypical xenobiotic metabolism related genes such as cytochrome P450s and aldehyde dehydrogenases which were highly induced. Other primary responses included genes associated with lipid metabolism (APOM, PLD3, ST8SIA1, ACSL3), cell cycle regulation (CDCA5, KANK1, FHIT, CTGF), and transcriptional regulation (MXD3, DEAF1,SERTAD2). The regulation of transcription factors and subsequent changes in gene expression is a hallmark 75 B 100 Number of Differentially Expressed Genes Number of Differentially Expressed Genes A 80 60 40 20 0 120 100 80 60 40 20 0 0.001 0.01 0.1 1 10 100 1 Dose (nM) 2 4 8 12 24 48 Time (hr) Figure 15. Number of genes exhibiting differential expression changes in the TCDD (A) dose-response (12 hr) and (B) time course study (10 nM TCDD). The number of differentially expressed genes exhibited dose-dependent induction and steady increase between 1 and 8 hr, followed by a decrease at 12 hr but further increases at 24 and 48 hr. 76 5.0 Expression ratio 2.0 1.0 0.5 0.2 1hr 2hr 4hr 8hr 12hr 24hr 48hr Figure 16. HL1-1 TCDD time course hierarchical clustering. Hierarchical clustering illustrates the induction and repression of 273 differentially expressed features representing 155 unique genes. Differentially expressed genes clustered according to early (2-4 hr), mid (8-12 hr), and late (24-48 hr) time points. 77 Figure 17. QRT-PCR verification of CYP1B1, ALDH1A3 and SLC7A5 microarray results in the time course (10 nM TCDD) and dose response (12 h) studies. Fold changes were calculated relative to time-matched vehicle controls. Bar (left axis) and lines (right axis) represent QRT-PCR and cDNA microarray data, respectively. EC50 values were calculated from QRT-PCR and cDNA microarray (parenthesized value) data, respectively. Results are represented as the average of three biological replicates. QRT-PCR data error bars represent the SEM for the average fold change. The asterisk (*) indicates p < 0.05 for QRT-PCR. 78 Figure 17 (Cont’d) Dose response 60 60 16 ALDH1A3 * 45 SLC7A5 45 12 EC50 = 2.38 nM (1.29 nM) * EC50 = 3.04 nM (2.09 nM) 9 * 30 8 4 30 * 6 * 15 0 12 6 45 EC50 = 1.70 nM (5.59 nM) * 30 15 15 4 -4 -3 -2 -1 0 1 Log Dose (nM) 0 2 0 * 2 15 -4 -3 -2 -1 0 1 Log Dose (nM) 2 0 0 Fold change (Array) Fold change (QRT-PCR) CYP1B1 8 60 * 3 -4 -3 -2 -1 0 1 Dose (nM) 0 2 Time Course 60 60 16 ALDH1A3 * 45 SLC7A5 * 45 12 12 * 6 45 * 15 0 * * * 1 2 * 4 8 12 24 48 Time (h) * 30 8 * * 0 * 4 30 2 15 1 2 6 * 15 4 0 9 * * 30 * 4 8 12 24 48 Time (h) 79 0 0 3 1 2 4 8 12 24 48 Time (h) 0 Fold change (Array) Fold change (QRT-PCR) CYP1B1 8 60 of TCDD action [33]. Immune response genes (IL1A, IL1B, CD8A), signal transduction related genes (MAPK7, PRKCB) and transporters (SLC2A1, SLC7A5, MTCH2) were also identified as putative primary responses. Computational analysis of the regulatory region of these putative primary responses revealed that 71 of the 78 genes had one or more putative DREs with a matrix similarity score greater than > 0.8 (Table 4). COMPARISON OF HL1-1 GENE EXPRESSION WITH HUMAN HEPATOMA HEPG2 CELLS Temporal changes in gene expression elicited by TCDD in HL1-1 cells were compared to intra-laboratory time course studies conducted in other model systems. 251 HL1-1 genes were identified as differentially expressed in the time course study using relaxed filtering criteria (P1(t) > 0.999 and |fold change| > 1.3) to include responses on the margins. Using the same relaxed criteria and the same study design, cDNA microarray, and data analysis strategy, 1,057 HepG2 genes were identified as differentially expressed at one or more time points following treatment with 10 nM TCDD (Dere et al., manuscript in preparation). Only 74 common genes were differentially regulated in HL1-1 and HepG2 cells by TCDD (Table 5, Figure 19A). Of these, 55 exhibited similar temporal expression patterns (38 induced; 17 repressed (Table 5, Figure 19B)), and 12 were classified as putative primary responses. Of the 19 genes exhibiting divergent regulation (7 induced in HL1-1 but repressed in HepG2; 12 repressed in HL1-1 but induced in HepG2 (Table 5, Figure 19C)), none were putative primary responses in the HL1-1 based on the CHX study. HL1-1 specific responses were associated with immune and lipid/xenobiotic metabolic processes (Figure 19D), while HepG2 specific responses were involved in the cytoskeleton, calcium signaling and lipid metabolism. Other functional associations included transport, cell cycle, signal transduction and transcriptional regulation. 80 4 hr CHX Study (78 active genes) 12 hr CHX Study (203 active genes) Primary response Secondary response Unclassifi able Primary response Secondary response Unclassifi able 47 20 11 53 95 |Fold change| >1.5 P1(t) > 0.9999 55 4hr (47) 25 12hr (53) 22 78 Unique putative primary responsive genes 31 Figure 18. Putative primary TCDD responsive genes from CHX study. Microarray analysis identified 78 and 203 TCDD-responsive genes at 4 and 12 hr, respectively. CHX co-treatment identified 47 and 53 genes classified as putative primary responsive genes at 4 and 12 hr, respectively. 78 unique putative primary responses were identified at both time points. 81 Table 4. Functional categorization of putative primary response genes elicited by TCDD Functional a Category Drug metabolism Lipid metabolism Entrez GeneId Gene Symbol Gene Name 1543 CYP1A1d cytochrome P450, family 1, subfamily A, polypeptide 1 1545 CYP1B1 cytochrome P450, family 1, subfamily B, polypeptide 1 224 ALDH3A2 aldehyde dehydrogenase 3 family, member A2 220 ALDH1A3 aldehyde dehydrogenase 1 family, member A3 8733 GPAA1 glycosylphosphatidylinositol anchor attachment protein 1 homolog (yeast) 5919 RARRES2 retinoic acid receptor responder 2 55937 APOM apolipoprotein M 51084 CRYL1 crystallin, lambda 1 23646 PLD3 phospholipase D family, member 3 9108 MTMR7 myotubularin related protein 7 6489 ST8SIA1 2181 ACSL3 2896 GRN ST8 alpha-N-acetylneuraminide alpha-2,8sialyltransferase 1 acyl-CoA synthetase longchain family member 3 granulin 82 DRE b Count 4 hr FC 12 hr FC TC/V C/V 507 4136 28 9.0 9.0 1.3 11.1 6 10.0 8.3 -1.2 3 8.6 7.5 9 1.8 8 13 7 T/V c T/V c TC/V C/V 222 26794 513 14.9 2.9 5.5 15.1 -1.4 -1.1 5.3 12.0 -1.3 2.8 1.5 1.5 3.7 1.9 2.4 1.8 1.2 3.0 3.9 1.6 8 7 6 1.2 2.4 1.6 1.2 2.7 2.9 1.0 1.0 1.6 2.0 2.2 1.1 1.1 10.1 2.7 -1.2 1.0 -1.1 6 3.1 4.9 1.3 1.7 6.9 1.5 4 1.7 2.2 1.5 2.2 5.4 2.4 7 -2.0 -1.8 1.0 1.2 2.7 -1.2 6 -2.2 -3.1 -1.1 -1.2 -3.2 1.2 Table 4 (cont'd) Functional a Category Regulation of cell cycle Entrez GeneId Gene Symbol 23189 KANK1 5069 PAPPA Regulation of transcription 113130 CDCA5 2272 FHIT 5270 SERPINE 2 5764 PTN 1490 CTGF 83463 MXD3 10522 DEAF1 22938 SNW1 10062 NR1H3 2002 ELK1 84759 PCGF1 22936 ELL2 9792 SERTAD2 10370 CITED2 Gene Name KN motif and ankyrin repeat domains 1 pregnancy-associated plasma protein A, pappalysin 1 cell division cycle associated 5 fragile histidine triad gene serpin peptidase inhibitor, clade E, member 2 pleiotrophin connective tissue growth factor MAX dimerization protein 3 deformed epidermal autoregulatory factor 1 SNW domain containing 1 nuclear receptor subfamily 1, group H, member 3 ELK1, member of ETS oncogene family polycomb group ring finger 1 elongation factor, RNA polymerase II, 2 SERTA domain containing 2 Cbp/p300-interacting transactivator, with Glu/Asp-rich carboxy-terminal domain, 2 83 DRE Count b 6 c 4 hr FC TC/ T/V C/V V 2.2 2.1 1.0 12 hr FC c T/V TC/V C/V 1.6 1.7 -1.3 3 2.5 2.4 1.6 1.7 5.8 1.7 2 1 0 8.6 1.2 2.9 14.8 1.2 2.6 1.4 1.4 1.6 2.8 1.8 2.0 33.3 2.4 7.1 2.1 1.4 1.9 0 2 13 13 1.8 -2.3 2.4 1.2 2.6 -2.2 2.8 1.1 1.9 1.0 -1.1 1.0 1.6 -1.7 2.1 1.7 5.8 -1.9 9.2 1.9 1.8 1.7 1.0 1.3 8 8 2.0 2.4 1.6 1.8 1.1 1.2 2.6 3.3 2.8 3.7 1.4 1.8 7 2.5 4.9 2.0 2.4 9.5 2.8 6 4 1.6 1.4 1.7 -1.1 -1.1 -1.4 1.7 2.1 3.0 1.5 1.1 -1.3 1 3 1.6 -1.9 6.7 -1.7 2.8 -1.1 2.0 1.4 8.7 2.3 4.4 1.0 Table 4 (cont'd) Functional a Category Signal transduction Entrez GeneId Gene Symbol Gene Name 2889 RAPGEF1 11156 29990 51768 136 3553 3552 5598 5579 4322 56990 4986 925 4982 Rap guanine nucleotide exchange factor (GEF) 1 PTP4A3 protein tyrosine phosphatase type IVA, member 3 PILRB paired immunoglobin-like type 2 receptor beta TM7SF3 transmembrane 7 superfamily member 3 ADORA2B adenosine A2b receptor IL1B interleukin 1, beta IL1A interleukin 1, alpha MAPK7 mitogen-activated protein kinase 7 PRKCB protein kinase C, beta MMP13 matrix metallopeptidase 13 (collagenase 3) CDC42SE2 CDC42 small effector 2 OPRK1 opioid receptor, kappa 1 CD8A CD8a molecule TNFRSF11B tumor necrosis factor receptor superfamily, member 11b 84 4 hr FC c 12 hr FC c DRE b Count T/V TC/V C/V T/V TC/V 12 1.7 1.4 1.2 2.2 2.3 1.1 10 1.8 1.7 1.1 1.4 2.7 1.2 10 1.3 1.4 1.1 1.6 1.9 1.1 9 1.2 1.1 -1.1 1.6 3.0 1.2 9 6 2 2 1 0 1.9 7.6 3.6 1.6 2.2 2.5 2.5 19.3 14.2 6.4 3.1 3.6 1.2 6.7 9.2 2.7 1.0 1.5 1.9 4.1 2.6 1.7 1.3 1.4 6.4 76.1 51.1 5.7 3.2 4.6 1.5 17.2 13.6 2.4 -1.7 1.3 0 0 2 1 2.0 1.4 -1.1 -1.7 2.5 1.5 1.0 -1.8 1.0 1.3 1.1 1.0 -1.7 1.9 1.5 -1.1 -1.5 4.8 1.8 -3.0 1.5 2.1 1.3 1.0 C/V Table 4 (cont'd) Functional a Category Transport Entrez GeneId Gene Symbol 29066 CLCN3 6583 SLC22A4 6566 SLC16A1 6513 SLC2A1 8140 SLC7A5 23788 MTCH2 RNA processing Regulation of translation Protein biosynthesi s 56342 2091 2332 1965 PPAN FBL FMR1 EIF2S1 1983 EIF5 51665 ASB1 51065 RPS27L 51081 MRPS7 DRE b Count Gene Name zinc finger CCCH-type containing 7A solute carrier family 22 (organic cation transporter), member 4 solute carrier family 16 (monocarboxylic acid transporters), member 1 solute carrier family 2 (facilitated glucose transporter), member 1 solute carrier family 7 (cationic amino acid transporter, y+ system), member 5 mitochondrial carrier homolog 2 (C. elegans) peter pan homolog (Drosophila) fibrillarin fragile X mental retardation 1 eukaryotic translation initiation factor 2, subunit 1 alpha, 35kDa eukaryotic translation initiation factor 5 ankyrin repeat and SOCS boxcontaining 1 ribosomal protein S27-like mitochondrial ribosomal protein S7 85 4 hr FC c 12 hr FC c T/V TC/V C/V T/V TC/V C/V 13 1.6 1.6 1.0 1.3 1.7 -1.1 13 2.2 4.1 1.9 2.2 7.5 3.2 11 2.4 2.2 1.1 2.0 4.5 1.5 9 3.3 6.5 1.4 1.7 8.2 1.0 7 9.6 12.2 1.0 5.1 26.6 1.3 4 2.9 5.0 1.0 1.6 5.8 -1.1 10 7 6 5 2.4 1.7 1.8 1.6 3.8 2.8 2.7 3.3 1.5 1.3 1.1 2.6 1.5 1.6 1.6 1.8 4.9 4.4 4.4 9.4 1.5 1.3 1.2 5.3 9 -1.6 -1.3 1.0 -1.8 -2.0 -1.4 6 1.1 1.1 1.0 1.6 1.9 1.3 5 7 1.7 -2.2 1.6 -2.0 1.0 1.0 1.6 1.0 1.4 2.6 1.1 -1.6 Table 4 (cont'd) Functional a Category Apoptosis Entrez GeneId Gene Symbol 10059 DNM1L 79370 BCL2L14 317 APAF1 Protein modification 20018 5 10413 64710 KRTCAP 2 YAP1 NUCKS1 90701 SEC11L3 8507 ENC1 2589 GALNT1 DNA metabolic process 10951 CBX1 10606 PAICS 6742 SSBP1 Gene Name dynamin 1-like BCL2-like 14 (apoptosis facilitator) apoptotic peptidase activating factor keratinocyte associated protein 2 Yes-associated protein 1, 65kDa nuclear casein kinase and cyclindependent kinase substrate 1 SEC11-like 3 (S. cerevisiae) ectodermal-neural cortex (with BTB-like domain) UDP-N-acetyl-alpha-Dgalactosamine:polypeptide Nacetylgalactosaminyltransferase 1 (GalNAc-T1) chromobox homolog 1 (HP1 beta homolog Drosophila ) phosphoribosylaminoimidazole carboxylase, phosphoribosylaminoimidazole succinocarboxamide synthetase single-stranded DNA binding protein 1 86 DRE Count b 12 c c 4 hr FC TC/ T/V C/V V 1.6 1.7 1.1 T/V TC/V -1.4 1.0 2.1 12 hr FC C/V 9 2.1 2.5 1.2 1.8 5.0 1.3 1 -1.2 -1.2 -1.1 -1.7 -2.2 -1.7 15 2.5 4.0 1.3 1.4 3.9 1.3 4 1 1.9 1.9 1.8 1.9 -1.2 -1.1 2.1 1.4 3.4 1.8 1.3 -1.2 1 4 1.9 -2.1 2.5 -2.0 1.2 1.0 -1.6 -1.6 -2.2 -2.2 1.3 1.4 0 -5.9 -5.8 -1.3 -2.4 -7.1 1.1 15 1.1 1.3 1.3 1.6 1.9 1.5 8 1.2 1.2 1.0 1.6 1.3 -1.1 2 1.4 2.0 1.8 1.5 4.4 2.5 Table 4 (cont'd) Functional a Category Etc Entrez GeneId Gene Symbol Gene Name 90488 C12orf23 DRE Count b c 4 hr FC TC/ T/V C/V V 1.2 1.1 1.1 12 hr FC T/V TC/V chromosome 12 open reading 10 1.8 3.2 frame 23 84079 ANKRD2 ankyrin repeat domain 27 (VPS9 9 2.2 3.6 -1.2 7 domain) 79798 ARMC5 armadillo repeat containing 5 7 1.6 2.8 1.1 1.7 2.4 23313 C22orf9 chromosome 22 open reading 7 1.5 1.5 1.1 1.8 2.4 frame 9 11346 SYNPO synaptopodin 5 2.4 7.5 2.5 1.7 9.8 57223 SMEK2 SMEK homolog 2, suppressor of 3 2.5 2.0 1.2 2.0 2.5 mek1 (Dictyostelium) 4077 NBR1 neighbor of BRCA1 gene 1 0 1.4 3.1 1.5 1.9 4.6 22948 CCT5 chaperonin containing TCP1, 6 -2.2 -1.7 1.3 -1.5 -1.7 subunit 5 (epsilon) 3188 HNRPH2 heterogeneous nuclear 6 -1.7 -1.4 1.0 -1.3 1.3 ribonucleoprotein H2 (H') a. Funtional categories was performed using an in-house Gene Ontology tool b. DRE identified in -10kb to transcriptional start site (TSS) and 5' UTR c. Expression fold changes (FC) determined by microarray analysis and numbers in bold font indicate |FC| > 1.5 T/V: TCDD treatment vs vehicle, T+C/V: TCDD & CHX co-treatment vs vehicle, C/V: CHX treatment vs vehicle d. Gene expressiion data was measured by QRT-PCR 87 c C/V 1.8 1.1 1.2 3.2 -1.2 2.7 1.5 2.1 GENE EXPRESSION PROFILE COMPARISON TO OTHER MODEL SYSTEMS Comparison of HL1-1 differential gene expression to mouse Hepa1c1c7 cell [34] and C57BL/6 hepatic tissue [35] were also examined using relaxed filtering criteria (P1(t) > 0.999 and |fold change| > 1.3) to include those responses approaching the initial criteria (P1(t) > 0.9999, |fold change| > 1.5). All of these studies used comparable study designs, cDNA microarray platforms and data analysis strategies. 5505 orthologous genes, defined by HomoloGene (http://www.ncbi.nlm.nih.gov/ HomoloGene/), were represented across the human and mouse cDNA arrays (Figure 20A). Comparison of HL1-1 to C57BL/6 liver tissue, and Hepa1c1c7 cells identified only 32 and 18 common differentially expressed genes, respectively (Figure 20B, Table 6). However, not all common differentially expressed genes exhibited the same expression pattern. For example, of the 18 genes differentially expressed in both HL1-1 cells and Hepa1c1c7 cells, 10 exhibited the same pattern (6 induced and 4 repressed genes), while 8 were divergently regulated (6 genes induced in HL1-1 cells were repressed in Hepa1c1c7; 2 genes repressed in HL1-1 were induced in Hepa1c1c7). Similar analyses were conducted between HL1-1 and C57BL/6 hepatic tissue differential gene expression data sets (Figure 20B, Table 7). Across all four models, only three genes (IRF1, SLC12A7 and ID3) were differentially expressed with only one gene (IRF1) exhibiting the same expression pattern. Functional annotations of the common TCDD regulated genes were associated with cell cycle regulation, and development. Several collagenases were differentially expressed in HL1-1 and mouse Hepa1c1c7 hepatoma cells, although some responses (e.g., CDC25B) were divergently regulated suggesting species-specific effects (Table 7) [27, 30, Sun, 2004 #172]. Comparison of the functional annotation of differentially expressed genes in HL1-1 cell and 88 Table 5. HL1-1 vs. HepG2 Overlapping Genes: Functional Categories Entrez Gene Name GeneId Co-induced (38) 7545 Zic family member 1 (oddpaired homolog, Drosophila) 9792 SERTA domain containing 2 Gene Symbol HL1-1 Time Points FC FC HepG2 Time Points ZIC1 2.6 2,4,8,12,24,48 SERTAD2* 2.2 1,2,4,12,24,48 THRAP5 1.7 2,4,8 9.1 2,4,8,12,24,4 8 3.6 1,2,4,8,12,24 ,48 1.7 8,12,24,48 IRF1 1.6 4,8,12 1.3 12 10265 iroquois homeobox protein 5 IRX5 1.6 2,4 1.7 2,4,8,12,24 58508 myeloid/lymphoid or mixedlineage leukemia 3 7799 PR domain containing 2, with ZNF domain 6256 retinoid X receptor, alpha MLL3 1.6 1,2,12,24,48 1.3 24 PRDM2 1.5 2,4,12,24,48 1.4 8,12,24 RXRA 1.4 2,4,8,12 1.8 4,8,12,24,48 POU6F1 1.4 8,12,24,48 2.2 8,12,24,48 PPAN* 2.1 2,4,8,12,48 SLC7A5* 9.1 1,2,4,8,12,24,48 2.2 2,4,8,12,24,4 8 1.4 24 10025 thyroid hormone receptor associated protein 5 3659 interferon regulatory factor 1 5463 POU domain, class 6, transcription factor 1 56342 peter pan homolog (Drosophila) 8140 solute carrier family 7 (cationic amino acid transporter, y+ system), member 5 89 Functions transcription regulator activity transcription regulator activity transcription regulator activity transcription regulator activity transcription regulator activity transcription regulator activity transcription regulator activity transcription regulator activity transcription regulator activity RNA splicing transport Table 5 (cont'd) Entrez Gene Gene Name GeneId Symbol 7529 tyrosine 3YWHAB monooxygenase/tryptophan 5monooxygenase activation protein, beta polypeptide 6513 solute carrier family 2 (facilitated glucose transporter), member 1 10723 solute carrier family 12 (potassium/chloride transporters), member 7 9962 solute carrier family 23 (nucleobase transporters), member 2 5579 protein kinase C, beta 1 54978 chromosome 2 open reading frame 18 196883 adenylate cyclase 4 83604 transmembrane protein 47 118672 chromosome 10 open reading frame 89 22934 ribose 5-phosphate isomerase A (ribose 5-phosphate epimerase) 51102 mitochondrial trans-2-enoylCoA reductase HL1-1 FC Time Points 3.0 1,2,4,8,12 HepG2 FC Time Points 2.7 1,2,4,8,12,24 Functions transport SLC2A1* 2.2 1,2,4,8,12,24, 48 1.3 8,24 transport SLC12A7 1.5 2,4,8 2.0 8,12,24,48 transport SLC23A2 1.5 2,3,12,48 2.1 4,8,12,24,48 transport PRKCB1* 1.6 12,24 C2orf18 2.1 1,2,4,8,12,24, 48 1.4 8,12,24 3.4 ADCY4 1.4 8,12,24 1.3 TMEM47 C10orf89 1.4 8,12,24 1.4 8,12,24 2.1 1.8 RPIA 2.5 2,4,8,12,24,4 8 MECR 1.4 8,12,24 90 25.7 signal transduction 4,8,12,24,48 signal transduction 12 signal transduction 8,12,24,48 transmembrane 8,12,24,48 regulation of translation 1,2,4,8,12,24, carbohydrate 48 metabolism 1.7 12,24 lipid metabolic process Table 5 (cont'd) Entrez Gene Name GeneId 10 N-acetyltransferase 2 (arylamine Nacetyltransferase) 6303 spermidine/spermine N1acetyltransferase 5682 proteasome (prosome, macropain) subunit, alpha type, 1 7298 thymidylate synthetase 706 benzodiazapine receptor (peripheral) 8453 cullin 2 1848 dual specificity phosphatase 6 92335 protein kinase LYK5 2683 UDP-Gal:betaGlcNAc beta 1,4- galactosyltransferase, polypeptide 1 928 CD9 molecule 1080 cystic fibrosis transmembrane conductance regulator, ATP-binding cassette 2821 glucose phosphate isomerase Gene Symbol NAT2 HL1-1 FC Time Points 1.7 12,24,48 HepG2 FC Time Points 1.6 24,48 SAT 1.4 2,4,8,24 PSMA1 1.6 2,4,8,12,24, 48 TYMS 1.7 2,4,8,12,48 BZRP CUL2 DUSP6 2.3 2,4,8,12,24, 48 1.5 48 1.4 12,24,48 LYK5 B4GALT1 1.4 24,48 1.4 12,24 2.8 8,12,24,48 1.9 8,12,24,48 CD9 1.5 48 1.6 24,48 CFTR 1.3 24,48 3.9 4,8,12,24,48 GPI 2.4 2,4,8,12,24, 48 1.7 24,48 91 1.4 12,24 22.6 1,2,4,8,12,24, 48 1.7 8,12,24 22.4 1,2,4,8,12,24, 48 1.5 12 1.9 1,2,4,8,12,24 Functions Caffeine metabolism amino groups metabolism protein metabolism DNA metabolic process cell proliferation cell proliferation regulation of cell cycle cell cycle regulation of apoptosis developmental process developmental process immune system process Table 5 (cont'd) Entrez Gene Name GeneId 2152 coagulation factor III (thromboplastin, tissue factor) 3108 major histocompatibility complex, class II, DM alpha 83699 SH3 domain binding glutamic acid-rich protein like 2 220004 chromosome 11 open reading frame 66 Co-repressed (17) 2181 acyl-CoA synthetase long-chain family member 3 1490 connective tissue growth factor Gene Symbol F3 HL1-1 FC Time Points 1.7 2,4,8,12,24 HepG2 FC Time Points 1.5 8,12,24 HLA-DMA 1.5 1 1.5 1,4,8 SH3BGRL2 1.4 8,12,24 2.1 8,12,24,48 C11orf66 1.7 2,4,8,12,24 Functions immune system process immune system process SH3 domain binding 1.5 12 ACSL3* -2.2 2,4,8,12,24,48 -1.7 2,12,24,48 CTGF* -2.3 2,4,8,12,24,48 -2.0 2,12,24,48 22948 chaperonin containing TCP1, subunit 5 (epsilon) 80273 GrpE-like 1, mitochondrial (E. coli) 2589 UDP-N-acetyl-alpha-Dgalactosamine:polypeptide Nacetylgalactosaminyltransferase 1 (GalNAc-T1) CCT5* -2.1 2,4,8,12,24,48 -2.3 2,4,12,24,48 Lipid metabolism cell differentiation protein folding GRPEL1 -1.3 24 -1.3 24,48 protein folding GALNT1* -3.1 2,4,8,12,24,48 -1.6 4,8,12,24 protein modification process 23788 mitochondrial carrier homolog 2 (C. elegans) 7376 nuclear receptor subfamily 1, group H, member 2 MTCH2* -1.5 4 -1.5 24,48 transport NR1H2 -2.4 2,4,8,12 -1.7 24,48 transcription regulator activity 92 Table 5 (cont'd) Gene Symbol BUD31 HL1-1 FC Time Points -1.8 24,48 HepG2 FC Time Points -1.7 12,24,48 8507 ectodermal-neural cortex (with BTB-like domain) 317 apoptotic peptidase activating factor 1902 endothelial differentiation, lysophosphatidic acid G-proteincoupled receptor, 2 4074 mannose-6-phosphate receptor (cation dependent) 3925 stathmin 1/oncoprotein 18 ENC1* -1.6 4,8,12,24,48 -1.4 48 APAF1* -1.5 12,24,48 -1.7 24,48 EDG2 -1.6 8,12,24,48 -1.7 24,48 M6PR -1.6 24,48 -1.7 12,24 STMN1 -1.4 24,48 -1.5 24 231 aldo-keto reductase family 1, member B1 (aldose reductase) 4913 nth endonuclease III-like 1 (E. coli) 5420 podocalyxin-like AKR1B1 -2.3 24,48 -1.4 24 NTHL1 -2.3 24,48 -1.4 24 PODXL -1.4 48 -1.4 24 9768 KIAA0101 KIAA0101 -1.5 12,24,48 -1.7 24,48 Entrez Gene Name GeneId 8896 BUD31 homolog (yeast) HL1_Up/HepG2_Down (7) 23204 ADP-ribosylation factor-like 6 interacting protein 54606 DEAD (Asp-Glu-Ala-Asp) box polypeptide 56 ARL6IP 1.9 12,24,48 -1.3 48 DDX56 1.7 4,8,12,24,48 -1.4 4 93 Functions transcription regulator activity developmental process regulation of apoptosis signal transduction signal transduction signal transduction carbohydrate metabolism DNA repair immune system process intracellular endomembrane system rRNA processing Table 5 (cont'd) Entrez Gene Gene Name GeneId Symbol 5786 protein tyrosine phosphatase, PTPRA receptor type, A 4851 Notch homolog 1, translocation- NOTCH1 associated (Drosophila) 5578 protein kinase C, alpha PRKCA HL1-1 FC Time Points 1.6 2,4,8,12,24,48 HepG2 FC Time Points -1.5 24,48 1.6 8,12,24 -1.5 24,48 1.4 24,48 -1.5 12,24,48 11332 acyl-CoA thioesterase 7 ACOT7 1.3 48 -1.4 24 57608 KIAA1462 KIAA1462 1.6 8,12,24,48 Functions signal transduction transcription regulator activity regulation of apoptosis lipid metabolic process -1.3 48 HL1_Down/HepG2_Up (12) 199 allograft inflammatory factor 1 652 bone morphogenetic protein 4 10293 TRAF interacting protein 834 caspase 1, apoptosis-related cysteine peptidase (interleukin 1, beta, convertase) 3399 inhibitor of DNA binding 3, dominant negative helix-loophelix protein 7077 TIMP metallopeptidase inhibitor 2 AIF1 -3.2 2,4,24,48 1.8 4,8,12,24,48 BMP4 -1.9 1,2,4,8,24 1.9 8,12,24,48 TRAIP -1.8 2,4,8 1.6 8,12,24,48 CASP1 -1.3 2,48 1.4 24 ID3 -2.5 2,4,24,48 1.9 4,8,12,24,48 regulation of transcription TIMP2 -1.3 48 1.6 24,48 cell proliferation 94 regulation of cell cycle regulation of cell cycle regulation of apoptosis regulation of apoptosis Table 5 (cont'd) Entrez Gene Name GeneId 85440 dedicator of cytokinesis 7 9200 protein tyrosine phosphataselike (proline instead of catalytic arginine), member A 22937 sterol regulatory element binding proteins (SREBF) chaperone 214 activated leukocyte cell adhesion molecule 2745 glutaredoxin (thioltransferase) 2806 glutamic-oxaloacetic transaminase 2, mitochondrial (aspartate aminotransferase 2) Gene Symbol DOCK7 PTPLA HL1-1 FC Time Points -1.5 48 -1.4 8,24 SCAP -1.8 4,24,48 2.6 4,8,12,24,48 lipid metabolic process ALCAM -1.6 24,48 1.5 12,24,48 immune response GLRX -1.5 48 2.0 12,24,48 transport GOT2 -1.3 48 1.5 12,24 amino acid metabolism * Putative primary responses based on the CHX co-treatment study 95 HepG2 FC Time Points 1.4 24 1.3 12 Functions cell differentiation developmental process A C HL1-1 (251) HepG2 (1,057) HepG2 HepG2 HL1-1 HL1-1 1 177 74 983 2 4 8 12 24 48 1 2 4 8 12 24 48 (hours) III IV Figure 19. Cluster analysis of HL1-1 and HepG2 differentially expressed genes. (A) 74 differentially expressed genes were identified in both HL1-1 and HepG2 cells following TCDD treatment. (B) Cluster analysis of the 55 genes that exhibited the same regulation in both HL1-1 and HepG2 cells. I: Gene set induced in both cell lines. II: Gene set repressed in both cell lines. The asterisk (*) indicates putative primary responses based on the CHX study. (C) Cluster analysis was performed on the 19 divergently regulated genes. III: Gene set induced in HL1-1, but repressed in HepG2. IV: Gene set repressed in HL1-1, but induced in HepG2. (D) Examples of specific HL1-1 and HepG2 gene expression responses grouped by functional annotation. 96 Figure 19 (cont’d) C HepG2 HL1-1 HL1-1 HepG2 1 2 4 8 12 2448 1 2 4 8 12 24 48 1 2 4 8 12 24 48 1 2 4 8 12 24 48 (hours) * * * * I * * * * * * II * 97 Figure 19 (cont’d) D 1 2 HL1-1 4 8 12 HepG2 24 48 1 2 4 8 12 24 48 (hours) xenobiotic / lipid metabolic process transport immune response translation transport cytoskeleton calcium signaling pathway/ calcium ion binding lipid metabolic process Expression Ratio 0.3 1.0 3.0 98 hepatic mouse tissue identified transcriptional regulation, signal transduction, metabolism, apoptosis and transport as being commonly regulated by TCDD (Table 7). In vitro and in vivo comparisons have previously reported that some genes were divergently regulated [34]. For example, the transcription regulation-related genes CEBPZ and ID3, and the signal transductionrelated genes DUSP6, ERBB3 and MTMR7 were divergently regulated in HL1-1 cell compared to hepatic mouse tissue. DISCUSSION In this study, AhR-mediated gene expression in HL1-1 human liver stem cells was compared to other in vitro and in vivo TCDD data sets. The normalcy and self-renewal properties of HL1-1 cells provide a unique, in vitro system that may more accurately reflect in vivo human responses to TCDD. As with other models, HL1-1 cells express a functional AhR, as evident by western analysis and the induction of several AhR gene battery members, including CYP1A1 and aldehyde dehydrogenases. Although comparable functional pathways are affected in each model, further examination indicates that different genes within common functions were differentially regulated. Furthermore, there were several examples of orthologs exhibiting divergent regulation (e.g., induced in one model but repressed in another). Interestingly, functional categorization of TCDD elicited differential gene expression is consistent with reported species-specific responses [30, 36] suggesting that HL1-1 cells may more accurately reflect in vivo human responses to TCDD. Human HepG2 cells are a popular model for investigating hepatotoxicity. Comparison of HepG2 and HL1-1 gene expression profiles revealed the co-regulation of several putative primary responses including CYP1A1, PRKCB, PPAN, SERTAD2, SLC2A1 and SLC7A5. In contrast, several divergently regulated genes not classified as primary responses were expressed 99 Figure 20. Comparative analysis of HL1-1, HepG2, Hepa1c1c7 and C57BL/6 hepatic tissue gene expression profiles. (A) The number of genes represented on the human and mouse cDNA arrays. In total 5,505 orthologs were represented on both platforms. (B) Comparative analysis of the TCDD elicited temporal gene expression of HL1-1, HepG2, Hepa1c1c7 and C57BL/6 mice hepatic tissue (Mm liver) studies. Collectively, 251, 1057, 770 and 1465 differentially expressed genes were identified from HL1-1, HepG2, Hepa1c1c7 and mouse liver TCDD time course study, respectively, using relaxed filtering criteria (P1(t) > 0.999 and |fold change| > 1.3) to include responses on the margins. 74, 18 and 32 differentially expressed genes were identified when comparing HL1-1 cells to HepG2 cells, Hepa1c1c7 cells, and C57BL/5 hepatic tissue, respectively. Further analysis identified 13 conserved genes between HL1-1, HepG2 and mouse liver, 3 conserved genes between HL1-1, HepG2 and Hepa1c1c7, and 8 conserved genes between HL1-1, Hepa1c1c7 and mouse liver. 100 Figure 20 (Cont’d) A B HUMAN (6,995) MOUSE (8,284) 177 1,490 5,505 HepG2 1057 HL1-1 251 74 983 2,779 vs. HepG2 74 61 10 0 3 HL1-1 251 219 Mm Liver 1465 32 vs. Mm 32 1433 14 5 10 vs. Hepa1 18 HL1-1 251 233 101 Hepa1c1c7 770 18 752 Table 6. HL1-1 vs. Hepa1c1c7 Overlapping Genes: Functional Categories Homolo  Entrez   Gene name  ‐gene ID  GeneId  Co‐induced (6)    1658  3659  interferon regulatory factor 1  50009  84658  integrin, beta 1  49860 113130  cell division cycle associated 5 20548  4322  matrix metallopeptidase 13 (collagenase 3)  36176  4826  neuronatin  9070 23299  bicaudal D homolog 2 (Drosophila)     MCM3 minichromosome maintenance deficient 3   (S. cerevisiae) associated protein  cyclin‐dependent kinase inhibitor 1C (p57, Kip2)  inhibitor of DNA binding 3, dominant negative   helix‐loop‐helix protein  hyaluronoglucosaminidase 2    regulation of transcription  cell adhesion / migration  cell cycle progression collagen catabolism  development  cytoskeleton   MCM3AP   cell cycle arrest CDKN1C  ID3  cell cycle arrest  negative regulation of  transcription  carbohydrate metabolism  Functions    Co‐repressed (4)  2902 8888  Gene   Symbol    IRF1  ITGB1  CDCA5 MMP13  NNAT  BICD2 58  1633  1028  3399  7776    8692      HYAL2      HL1‐1 up/ Hepa1c1c7 down (6)  21312  10723  solute carrier family 12 (potassium/chloride transporters),  member 7  2225  6279  S100 calcium binding protein A8 (calgranulin A)  48120  5329  plasminogen activator, urokinase receptor  41451  994  cell division cycle 25B  9866 54940  OCIA domain containing 1 2492  7288  tubby like protein 2    SLC12A7    ion transport  S100A8  PLAUR  CDC25B  OCIAD1 TULP2  inflammatory response  signal transduction  regulation of cell cycle  cell adhesion visual perception        SEMA3C  immune response  UBR2 ubiquitin cycle     HL1‐1 down/ Hepa1c1c7 up (2) 36201  10512  sema domain, immunoglobulin domain (Ig),   short basic domain, secreted, (semaphorin) 3C  26151 23304  ubiquitin protein ligase E3 component n‐recognin 2      102 Table 7. HL1-1 vs. Mouse Hepatic Tissue Overlapping Genes: Functional Categories Gene   Homolo Entrez   Gene name  Symbol  ‐gene ID  GeneId  Co‐up (14)  68035 1545  cytochrome P450, family 1, subfamily B, polypeptide 1  CYP1B1 1658 3659  interferon regulatory factor 1  IRF1  3832 2002  ELK1, member of ETS oncogene family  ELK1  32049 4851  Notch homolog 1, translocation‐associated (Drosophila)  NOTCH1  55621 20457 1848  dual specificity phosphatase 6 2065  v‐erb‐b2 erythroblastic leukemia viral oncogene  homolog 3 (avian)  99732 9108  myotubularin related protein 7  15780 11332  acyl‐CoA thioesterase 7  49860 113130  cell division cycle associated 5 50009 84658  integrin, beta 1  68520 6513  solute carrier family 2 (facilitated glucose transporter),  member 1  41088 51099  abhydrolase domain containing 5 2080 5682  proteasome (prosome, macropain) subunit, alpha type,  1  10683 57099  apoptosis, caspase activation inhibitor        Co‐down (5) 1577 2896  granulin  3278 2181  acyl‐CoA synthetase long‐chain family member 3  26151 23304  ubiquitin protein ligase E3 component n‐recognin 2  10859 57704  glucosidase, beta (bile acid) 2 10966 83698  calneuron 1  103 Functions  xenobiotic metabolizing enzyme regulation of transcription  regulation of transcription  regulation of transcription  DUSP6 ERBB3  signal transduction signal transduction  MTMR7  ACOT7  CDCA5 ITGB1  SLC2A1  signal transduction  lipid metabolism  cell cycle cell adhesion / migration  carbohydrate transport  ABHD5 PSMA1  AVEN    aromatic compound metabolism ubiquitin‐dependent protein  degradation  apoptosis    GRN  ACSL3  UBR2  GBA2 CALN1  cell‐cell signaling  lipid metabolism  ubiquitin cycle  bile acid metabolism calcium ion binding  Table 7 (cont'd) Homolo Entrez   Gene name  ‐gene ID  GeneId  HL1‐1 up/ Mouse liver down (9)  3045 2791  guanine nucleotide binding protein (G protein), gamma  11  20621 5786  protein tyrosine phosphatase, receptor type, A  199 3990  lipase, hepatic  13090 84759  polycomb group ring finger 1  20420 928  CD9 molecule 21312 10723  solute carrier family 12 (potassium/chloride  transporters), member 7  48120 5329  plasminogen activator, urokinase receptor  11174 53838  chromosome 11 open reading frame 24  12876 83699  SH3 domain binding glutamic acid‐rich protein like 2        HL1‐1 down/ Mouse liver up (4)  1229 214  activated leukocyte cell adhesion molecule  4210 10153  CCAAT/enhancer binding protein zeta 1633 3399  inhibitor of DNA binding 3, dominant negative helix‐ loop‐helix protein  2902 8888  MCM3 minichromosome maintenance deficient 3  associated protein  104 Gene   Symbol  Functions  GNG11 signal transduction PTPRA  LIPC  PCGF1  CD9 SLC12A7  signal transduction  lipid catabolism  regulation of transcription  cell adhesion ion transport  PLAUR  C11orf24  SH3BGRL2   chemotaxis  oxidoreductase  unknown   ALCAM  CEBPZ ID3  antimicrobial humoral response  regulation of transcription transcription corepressor activity  MCM3AP  DNA replication  at later time points suggesting that they may be secondary effects of TCDD. In addition, a number of cell specific responses were observed. For example, HL1-1 specific responses included xenobiotic and lipid metabolism and immune responses that correlated with in vivo mice effects [35], while HepG2 exhibited cytoskeleton and calcium signaling responses related to cell-adhesion, tumor cell motility and tumor promotion [37-39]. This may be due to differences in basal expression levels or other unique cell characteristics [40]. For example, primary rat hepatocytes cultured on standard collagen had basal gene expression levels more comparable to whole liver rather than rat hepatoma cells [41]. Comparisons of HL1-1 cells to other in vitro (i.e., Hepa1c1c7 cells) and in vivo (i.e., hepatic tissue from C57BL/6 mice) models identified several common TCDD responses, as well model specific differential gene expression. Although the structure, function, and mechanism of action of the AhR is highly conserved [42], comparative toxicogenomic and computational DRE search studies suggest that TCDD elicited gene expression profiles may be species-specific. Computational analysis of the regulatory regions of orthologs using a position weight matrix suggests that DREs are not conserved between humans, mice and rats [27]. Moreover, in vivo rat vs. mouse [30], and in vitro (HepG2 vs. Hepa1c1c7 vs. H4IIE) (Dere et al., manuscript in preparation) studies indicate that the hepatic gene expression profiles are species specific, despite the conserved induction of xenobiotic metabolizing genes. Moreover, identified putative primary responses differ between species (Dere et al., manuscript in preparation). Nevertheless, TCDD does affect common pathways across species and models, but appears to do so by regulating the expression of non-orthologous genes within those pathways. Collectively, these studies not only demonstrate the utility of HL1-1 cells but also the limitations of extrapolating from in vitro models to in vivo effects. Although in vitro models 105 have the advantage of reducing the complexity of a tissue response to allow a more focused analysis of the effects of TCDD on a specific cell type, it does not replicate other interactions that may be important in eliciting the toxic responses observed in vivo [34]. However, in addition to being a normal human cell which may more accurately reflect human responses relative to rodent models, HL1-1 cells are also stem-like which may be novel targets of toxicity [12]. 106 REFERENCES 107 REFERENCES 1. Denison, M.S. and S. Heath-Pagliuso, The Ah receptor: a regulator of the biochemical and toxicological actions of structurally diverse chemicals. Bull Environ Contam Toxicol, 1998. 61(5): p. 557-68. 2. Knerr, S. and D. Schrenk, Carcinogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in experimental models. Mol Nutr Food Res, 2006. 50(10): p. 897-907. 3. IARC, IARC Working Group on the Evaluation of Carcinogenic Risks to Humans: Polychlorinated Dibenzo-Para-Dioxins and Polychlorinated Dibenzofurans. Lyon, France, 4-11 February 1997. IARC Monogr Eval Carcinog Risks Hum, 1997. 69: p. 1631. 4. Okey, A.B., An aryl hydrocarbon receptor odyssey to the shores of toxicology: the Deichmann Lecture, International Congress of Toxicology-XI. Toxicol Sci, 2007. 98(1): p. 5-38. 5. Olson, H., et al., Concordance of the toxicity of pharmaceuticals in humans and in animals. Regul Toxicol Pharmacol, 2000. 32(1): p. 56-67. 6. Abbott, A., Cell culture: biology's new dimension. Nature, 2003. 424(6951): p. 870-2. 7. MacGregor, J.T., et al., In vitro human tissue models in risk assessment: report of a consensus-building workshop. Toxicol Sci, 2001. 59(1): p. 17-36. 8. Pritchard, J.F., et al., Making better drugs: Decision gates in non-clinical drug development. Nat Rev Drug Discov, 2003. 2(7): p. 542-53. 9. MacGregor, J.T., SNPs and chips: genomic data in safety evaluation and risk assessment. Toxicol Sci, 2003. 73(2): p. 207-8. 10. Li, A.P., Screening for human ADME/Tox drug properties in drug discovery. Drug Discov Today, 2001. 6(7): p. 357-366. 11. Rohwedel, J., et al., Embryonic stem cells as an in vitro model for mutagenicity, cytotoxicity and embryotoxicity studies: present state and future prospects. Toxicol In Vitro, 2001. 15(6): p. 741-53. 12. Davila, J.C., et al., Use and application of stem cells in toxicology. Toxicol Sci, 2004. 79(2): p. 214-23. 13. Rolletschek, A., P. Blyszczuk, and A.M. Wobus, Embryonic stem cell-derived cardiac, neuronal and pancreatic cells as model systems to study toxicological effects. Toxicol Lett, 2004. 149(1-3): p. 361-9. 108 14. Carere, A., A. Stammati, and F. Zucco, In vitro toxicology methods: impact on regulation from technical and scientific advancements. Toxicol Lett, 2002. 127(1-3): p. 153-60. 15. Ahuja, Y.R., V. Vijayalakshmi, and K. Polasa, Stem cell test: a practical tool in toxicogenomics. Toxicology, 2007. 231(1): p. 1-10. 16. Ek, M., et al., Expression of drug metabolizing enzymes in hepatocyte-like cells derived from human embryonic stem cells. Biochem Pharmacol, 2007. 74(3): p. 496-503. 17. Chang, C.C., Tsai,J.L., Kuo,K.K., Wang,K.H., Chiang,C.H., Kao,A.P., Tai,M.H. and Trosko,J.E., Expression of Oct-4, alpha fetoprotein and vimentin and lack of gapjunctional intercellular communication (GJIC) as common phenotypes for human adult liver stem cells and hepatoma cells. Proc. Am. Assoc. Cancer Res., 2004. 45: p. 642. 18. Tai, M.H., et al., Oct4 expression in adult human stem cells: evidence in support of the stem cell theory of carcinogenesis. Carcinogenesis, 2005. 26(2): p. 495-502. 19. Yang, Y.H. and T. Speed, Design issues for cDNA microarray experiments. Nat Rev Genet, 2002. 3(8): p. 579-88. 20. Eckel, J.E., et al., Normalization of two-channel microarray experiments: a semiparametric approach. Bioinformatics, 2005. 21(7): p. 1078-83. 21. Eckel, J.E., et al., Empirical bayes gene screening tool for time-course or dose-response microarray data. J Biopharm Stat, 2004. 14(3): p. 647-70. 22. Burgoon, L.D., et al., dbZach: A MIAME-compliant toxicogenomic supportive relational database. Toxicol Sci, 2006. 90(2): p. 558-68. 23. Burgoon, L.D. and T.R. Zacharewski, dbZach toxicogenomic information management system. Pharmacogenomics, 2007. 8(3): p. 287-91. 24. Saeed, A.I., et al., TM4: a free, open-source system for microarray data management and analysis. Biotechniques, 2003. 34(2): p. 374-8. 25. Dennis, G., Jr., et al., DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol, 2003. 4(5): p. P3. 26. Rozen, S. and H. Skaletsky, Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol, 2000. 132: p. 365-86. 27. Sun, Y.V., et al., Comparative analysis of dioxin response elements in human, mouse and rat genomic sequences. Nucleic Acids Res, 2004. 32(15): p. 4512-23. 28. N'Jai, A., et al., Comparative temporal toxicogenomic analysis of TCDD- and TCDFmediated hepatic effects in immature female C57BL/6 mice. Toxicol Sci, 2008. 103(2): p. 285-97. 109 29. Kopec, A.K., et al., Comparative toxicogenomic examination of the hepatic effects of PCB126 and TCDD in immature, ovariectomized C57BL/6 mice. Toxicol Sci, 2008. 102(1): p. 61-75. 30. Boverhof, D.R., et al., Comparative toxicogenomic analysis of the hepatotoxic effects of TCDD in Sprague Dawley rats and C57BL/6 mice. Toxicol Sci, 2006. 94(2): p. 398-416. 31. Zeiger, M., et al., Inducing effects of dioxin-like polychlorinated biphenyls on CYP1A in the human hepatoblastoma cell line HepG2, the rat hepatoma cell line H4IIE, and rat primary hepatocytes: comparison of relative potencies. Toxicol Sci, 2001. 63(1): p. 6573. 32. Silkworth, J.B., et al., Comparison of TCDD and PCB CYP1A induction sensitivities in fresh hepatocytes from human donors, sprague-dawley rats, and rhesus monkeys and HepG2 cells. Toxicol Sci, 2005. 87(2): p. 508-19. 33. Reymann, S. and J. Borlak, Transcriptome profiling of human hepatocytes treated with Aroclor 1254 reveals transcription factor regulatory networks and clusters of regulated genes. BMC Genomics, 2006. 7: p. 217. 34. Dere, E., et al., In vivo-in vitro toxicogenomic comparison of TCDD-elicited gene expression in Hepa1c1c7 mouse hepatoma cells and C57BL/6 hepatic tissue. BMC Genomics, 2006. 7: p. 80. 35. Boverhof, D.R., et al., Temporal and dose-dependent hepatic gene expression patterns in mice provide new insights into TCDD-Mediated hepatotoxicity. Toxicol Sci, 2005. 85(2): p. 1048-63. 36. Boutros, P.C., et al., Dioxin-responsive AHRE-II gene battery: identification by phylogenetic footprinting. Biochem Biophys Res Commun, 2004. 321(3): p. 707-15. 37. Puga, A., A. Maier, and M. Medvedovic, The transcriptional signature of dioxin in human hepatoma HepG2 cells. Biochem Pharmacol, 2000. 60(8): p. 1129-42. 38. Monteiro, P., et al., Dioxin-mediated up-regulation of aryl hydrocarbon receptor target genes is dependent on the calcium/calmodulin/CaMKIalpha pathway. Mol Pharmacol, 2008. 73(3): p. 769-77. 39. Tannheimer, S.L., et al., Carcinogenic polycyclic aromatic hydrocarbons increase intracellular Ca2+ and cell proliferation in primary human mammary epithelial cells. Carcinogenesis, 1997. 18(6): p. 1177-82. 40. Harris, A.J., S.L. Dial, and D.A. Casciano, Comparison of basal gene expression profiles and effects of hepatocarcinogens on gene expression in cultured primary human hepatocytes and HepG2 cells. Mutat Res, 2004. 549(1-2): p. 79-99. 41. Boess, F., et al., Gene expression in two hepatic cell lines, cultured primary hepatocytes, and liver slices compared to the in vivo liver gene expression in rats: possible 110 implications for toxicogenomics use of in vitro systems. Toxicol Sci, 2003. 73(2): p. 386402. 42. McGregor, D.B., et al., An IARC evaluation of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans as risk factors in human carcinogenesis. Environ Health Perspect, 1998. 106 Suppl 2: p. 755-60. 111 CHAPTER 5 Kim S, Kiyosawa N, Burgoon LD, Chang CC, and Zacharewski TR: PPAR-mediated responses in human adult liver stem cells: comparative in vivo/in vitro cross-species studies. Toxicol Sci (Submitted August 19, 2011) 112 CHAPTER 5 PPAR-MEDIATED RESPONSES IN HUMAN ADULT LIVER STEM CELLS: IN VIVO/IN VITRO AND CROSS-SPECIES COMPARISONS ABSTRACT The peroxisome proliferator-activated receptor α (PPARα) is a ligand activated transcription factor that regulates a variety of biological processes including lipid metabolism and energy homeostasis. Peroxisome proliferators (PPs) are non-genotoxic carcinogens in rodents, but humans are resistant to peroxisome proliferation and carcinogenesis. In this study, we examined differential gene expression elicited by clofibrate (CLO) and Wy-14,643 (WY) in C57BL/6 mouse liver compared to responses in human HepG2 hepatoma and HL1-1 adult stem cells. Mice were gavaged with sesame oil, 300 mg/kg CLO or WY for 2, 4, 8, 12, 18 or 24 h, or every 24 h and sacrificed after 1, 4 or 14 days. Although no significant changes in body weight gain were observed, WY induced relative liver weight at 4 and 14 days. Genome-wide hepatic gene expression analysis identified 719 and 1,443 unique genes that were differentially expressed by CLO and WY, respectively (|fold change|>1.5, P1(t)>0.99). Functional analysis associated the differentially expressed genes with lipid metabolism, transport, cell cycle and immune response. Using relaxed statistical criteria, ~90% of CLO and ~75% of the WY differentially expressed genes were in common between both treatments. Hierarchical clustering revealed only early time points (2-8h) clustered together. Complementary QRTPCR studies in human HL1-1 and HepG2 cells treated with 50 µM WY or DMSO for 1, 2, 4, 8, 12, 24 or 48 h identified a minimal number of conserved orthologous responses (e.g., Pdk4, Adfp and Angptl4) 113 while some genes (i.e., Bmf, a tumor suppressor) exhibited induction in human cells but repression in mice. These data suggest that PPs elicit species-specific PPAR-mediated gene expression. INTRODUCTION Peroxisome proliferators (PPs) are a diverse class of compounds that include industrial chemicals, drugs and endogenous steroids and lipids [1, 2]. While synthetic PPs, such as fibrates, are commonly used to treat dyslipidemia, other including pesticides, plasticizers and solvents are inadvertently introduced into the environment. They elicit a broad spectrum of biochemical and physiological responses such as liver hyperplasia and hypertrophy, peroxisome proliferation and increases in mitochondrial, peroxisomal, and microsomal fatty acid oxidation [3] mediated by peroxisome proliferator-activated receptors (PPAR), a family of ligand-activated transcription factors (TFs) [4]. PPARs exert their effects following ligand binding and dimerization with the retinoid X receptor (RXR). The heterodimer complex interacts with peroxisome proliferator response elements (PPRE) in the regulatory regions of target genes to modulate gene expression. Given their role in lipid metabolism and regulation by exogenous ligands, PPARs are active drug development targets for the treatment of hyperlipidemia, diabetes and obesity [5, 6]. PPARα and -γ are the most widely studied of the three isoforms (PPARα, -β/δ and -γ). PPARα and β/δ regulate catabolic energy metabolism, while PPARγ regulates anabolic lipid metabolism [2]. Studies with PPARα agonists, such as fibrates, suggest PPAR activation induces fatty acid oxidation in hepatic peroxisomes [7]. Fibrates were introduced as treatments for metabolic disorders in the early 1960s with the release of clofibrate with the subsequent development of other fibrates, including fenofibrate, gemfibrozil, bezafibrate and ciprofibrate [6, 8]. However, results from rodent carcinogenesis studies suggested fibrates and other 114 hypolipidemic drugs (i.e. Wy-14,643 and tibric acid) induce hepatic carcinogenesis curtailing their clinical use [9]. Gene knockout studies confirmed that PP-induced pleiotropic responses, including peroxisome proliferation and the development of non-genotoxic hepatocarcinomas, are PPARα-mediated [10, 11]. The human risk of PP-induced hepatocarcinomas is uncertain. Clinical studies suggest humans are resistant to PP-induced peroxisome proliferation and hepatic carcinogenesis [12-14]. PP-treated of PPARα humanized mice exhibit diminished peroxisome prolifereration and hepatocarcinoma incidence, suggesting that structural differences between human and mouse PPAR may lead to differential susceptibility [15, 16]. There are also significant differences between species in the levels of expression, structure and function of PPARα, as well as coregulator proteins and PPRE distribution [17-20]. In order to further elucidate the molecular basis for PPARα-mediated species-specific responses, comparative cross-species studies are required. There are many models, such as humanized mice [21] to human hepatoma cell line and primary hepatocyte [22] were implemented to study cross-species comparison to human. Human adult stem cells isolated from liver tissue are an attractive in vitro alternative with many advantages including the fact that they are a non-transformed intact cell from human tissue with proliferating and differentiation ability. They will provide an unlimited source of intact human cells and may be more predictive of human responses. It could also be a potential target for certain toxic end points, such as tumor development [23-25]. In this study, we examined CLO- and WY-elicited gene expression in C57BL/6 mouse liver and compared with WY-elicited gene expression in adult HL1-1 human stem cells and HepG2 human hepatoma cells to further investigate species-specific responses to PPs. 115 MATERIALS & METHODS HUSBANDRY Female C57BL/6 mice, ovariectomized by the supplier on postnatal day (PND) 20, were obtained from Charles River Laboratories (Raleigh, NC) on PND 25. The immature ovariectomized mouse was used to negate potential interactions with estrogens produced by the developing ovaries [26], and to facilitate comparisons with other data sets obtained using the same model, study design and analysis methods [27-30]. Mice were housed in polycarbonate cages containing cellulose fiber chip bedding (Aspen Chip Laboratory Bedding, Northeastern Products, Warrensberg, NY) maintained at 40-60% humidity and 23°C with a 12 h dark/light cycle (7 am - 7 pm). Animals were allowed free access to de-ionized water and Harlan Teklad 22/5 Rodent Diet 8640 (Madison, WI). Mice were acclimatized for 4 days prior to dosing. TREATMENTS AND NECROPSY Mice (n = 5) were orally gavaged once for the acute study or daily for sub-chronic studies with 300 mg/kg body weight CLO (Sigma-Aldrich, St Louis, MO) or WY (Sigma-Aldrich) in 0.1 ml of sesame oil (Sigma-Aldrich) vehicle. Time-matched vehicle controls (n = 5) were similarly gavaged with sesame seed oil vehicle. In acute studies, mice were sacrificed 2, 4, 8, 12, 18, and 24 h post-dose by cervical dislocation (Figure 21A) while sub-chronic animals were sacrificed 24 h after the last dose (1, 4 and 14 days) (Figure 21B). Body and liver weights were recorded and sections of the left lateral liver lobe (approximately 0.1 g) were snap-frozen in liquid nitrogen and stored at -80°C. The right lateral lobe was fixed in 10% neutral buffered formalin (NBF, VWR International, West Chester, PA) for histological analysis. All procedures were performed with the approval of the Michigan State University All-University Committee on Animal Use and Care. 116 CULTURE AND TREATMENT OF CELL LINES HL1-1 cells [31] were maintained in a modified MCDB 153 media (Keratinocyte-SFM, Invitrogen Corporation, Carlsbad, CA) supplemented with N-acetyl-L-cysteine (2 mM), Lascorbic acid 2-phosphate (0.2 mM) (referred to as K-NAC medium) and 50 g/mL gentamycin (Invitrogen). Added growth factors/hormones included rEGF (5 ng/mL), bovine pituitary extract (50 mg/mL) and 10% fetal bovine serum (FBS) (Hyclone, Logan, UT). HepG2 cells were maintained in phenol-red free DMEM/F12 media (Invitrogen) supplemented with 5% FBS (Hyclone), 50 g/mL gentamycin (Invitrogen), 100 U/mL penicillin and 100 g/mL 6 2 streptomycin (Invitrogen). 1  10 cells were seeded into a 25 cm cell culture flask (#430639, vent cap) (Corning Inc., Corning, NY) and incubated under standard conditions (5% CO2, 37C). For the time course studies, cells were treated with 50 uM WY (Sigma-Aldrich) or 0.1% (v/v) DMSO (Sigma-Aldrich) vehicle and harvested at 1, 2, 4, 8, 12, 24 or 48 h post-treatment (Figure 21C). MTT ASSAY Cell viability was assessed at 0, 3, 10, 30, 50, 100, 200, 300, 500, 1000, 1500 and 2000 M clofibrate, fenofibrate and Wy-14,643 at 24 h using the 3-(4,5-dimethyl-2-thiazolyl)-2,5diphenyl-2H tetrazorium bromide (MTT) assay (Sigma-Aldrich). Briefly, MTT (5mg/mL) was added following 24 h incubation and incubated for an additional 3 h with MTT. DMSO was then added to solubilize the MTT formazan, and absorbance was measured at 595 and 650 nm. Corrected absorbance was determined by subtracting the 650 nm value from the 595 nm background value. Relative cell viability was calculated as a percentage of control. Cell viability relative to dose was plotted to delineate PP concentrations that depressed MTT- 117 formazan production by 50% (LC50; Yoshii, 1997). LC50 values are expressed as the mean ± SEM from three separate experiments. PROTEIN PREPARATION AND WESTERN BLOT Cell and tissue lysates for Western blot analysis were prepared in RIPA buffer (1x PBS, 0.1% SDS, 1% IGEPAL CA-630 and 0.5% Na-Deoxycholate) with protease inhibitor cocktail tablet (Roche Diagnostics, Mannheim, Germany) and quantified using the modified Lowry assay (Bio-Rad, DC Protein Assay, Hercules, CA). Cell and tissue lysates were electrophoretically separated on a denaturing 12% SDS-polyacrylamide gel and transferred onto nitrocellulose membrane (Amersham Biosciences Inc., Piscataway, NJ). Membranes were probed with antihuman rabbit PPAR (H-98) polyclonal antibody followed by horseradish peroxidaseconjugated secondary antibodies (Santa Cruz Biotechnology Inc., Santa Cruz, CA). Immunochemical staining and fluorescence detection on X-ray film was performed using the SuperSignal West Dura substrate (Thermo Fisher Scientific, Inc., Waltham, MA). RNA ISOLATION Total RNA was isolated from left lateral liver sections using Trizol Reagent (Invitrogen, Carlsbad, CA). Samples were removed from -80°C storage and homogenized in 1 ml Trizol Reagent using a Mixer Mill 300 tissue homogenizer (Retsch, Germany). Cells were harvested in o TRIzol Reagent (Invitrogen) and stored at -80 C. Total RNA was isolated according to the manufacturer’s protocol and resuspended in RNA Storage Solution (Ambion Inc., Austin, TX). RNA samples were quantified spectrophotometrically (A260) and assessed for purity by A260/A280 ratio and by visual inspection on a denaturing agarose gel. 118 MICROARRAY ANALYSIS OF DIFFERENTIAL GENE EXPRESSION Treated and time-matched vehicle control samples were hybridized to independent arrays on the same array slide using one-color labeling (Cy3) of Whole Mouse Genome 4 × 44 K Oligo Microarray Kit (Agilent Technologies, Inc, Santa Clara, CA) with three biological replicates performed at each time point (Figure 21D). Microarray analysis was performed according to the manufacturer’s protocol (Agilent Manual: G4140-90040 v. 5.7). The microarrays were scanned at 532 nm (Cy3) using a GenePix 4000B microarray scanner (Molecular Devices, Union City, CA). Images were analyzed using GenePix Pro 6.0 (Molecular Devices). MICROARRAY DATA NORMALIZATION AND STATISTICAL ANALYSIS Data were normalized using a semi-parametric approach. Model-based t-values were calculated from normalized data, comparing treated and vehicle responses per time-point. Empirical Bayes analysis was used to calculate posterior probabilities (P1(t) value) of activity on a per gene and time point basis using the model-based t-value [32]. Normalization and empirical Bayes analysis were performed using SAS 9.2 (SAS Institute, Cary, NC) and R 2.11.0 (http://www.r-project.org). The data were filtered using a P1(t) and fold change cutoff to identify differentially regulated genes for subsequent analysis and interpretation. All raw and normalized data were stored in dbZach, which supports microarray data management, mining, visualization and knowledge management [33]. Differentially expressed genes were analyzed by hierarchical clustering (Multiexperiment Viewer (MeV) in TM4 software [34]) using Pearson correlation with average linkage. Dose response analysis of cytotoxicity was performed using Graph Pad Prism 5.0 (GraphPad Software, San Diego, CA). Annotation and functional categorization of differentially regulated genes was performed using Database for Annotation, Visualization and Integrated Discovery (DAVID) [35]. Unless stated otherwise, all data were 119 analyzed by analysis of variance (ANOVA) followed by Tukey’s post hoc tests. Differences between treatment groups were considered significant when p < 0.05. QUANTITATIVE REAL-TIME PCR (QRT-PCR) ANALYSIS QRT-PCR was performed as verification of microarray data for selected genes. For each sample, 1.5 g of total RNA was reverse transcribed by SuperScript II reverse transcriptase using an anchored oligo-dT primer as described by the manufacturer’s instructions (Invitrogen). 1.5 L of cDNA template was used in a 30 L PCR reaction containing 0.1 M of forward and reverse gene-specific primers designed using Primer3[36] and SYBR Green PCR reaction mixture (Applied Biosystems, Foster City, CA). PCR amplification was conducted in MicroAmp Optical 96-well reaction plates (Applied Biosystems) on an Applied Biosystems PRISM 7500 Sequence Detection System under 10 min initial denaturation and enzyme activation at 95C, followed by 40 cycles of 95C for 15 s and 60C for 1 min. A dissociation protocol was performed to assess the specificity of the primers and the uniformity of the PCR products. Target gene cDNAs were quantified using a standard curve of log copy number versus threshold cycle (Ct). The copy number of each sample was standardized to the geometric mean of two house-keeping genes, -actin (ACTB) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), to control for differences in RNA loading, quality, and cDNA synthesis [37]. For graphing purposes, the relative gene expression levels were scaled such that the expression level of the time-matched vehicle treated control group was equal to 1. Official gene names and abbreviations, forward and reverse primer sequences, and product length for the genes verified by QRT-PCR are listed in Table 8. 120 Table 8. Gene names and primer sequences for QRT-PCR RefSeq Gene name Mouse NM_007393 actin, beta, cytoplasmic NM_008084 glyceraldehyde-3-phosphate dehydrogenase NM_011144 peroxisome proliferator activated receptor alpha NM_015729 acyl-Coenzyme A oxidase 1, palmitoyl NM_013743 pyruvate dehydrogenase kinase, isoenzyme 4 NM_010011 cytochrome P450, family 4, subfamily a, polypeptide 10 NM_177406 cytochrome P450, family 4, subfamily a, polypeptide 12a NM_013464 aryl-hydrocarbon receptor Gene symbol Entrez Gene ID Actb 11461 Gapdh 14433 Ppara 19013 Acox1 11430 Pdk4 27273 Cyp4a10 13117 Cyp4a12a 277753 Ahr 11622 NM_007408 adipose differentiation related protein NM_020581 angiopoietin-like 4 Adfp 11520 Angptl4 57875 NM_138313 BCL2 modifying factor Bmf 171543 121 Forward Primer Reverse Primer GCTACAGCTT CACCACCACA GTGGACCTCA TGGCCTACAT TCTGTGGGCT CACTGTTCTG CTCCCACTCT GGTCTTCCTG GGCCATCCAT GTAGGAGAGA ACCTACGTGC TGAGGTGGAC TTCCAGTCTC CTTGCCTGTC ACCAGAACTG TGAGGGTTGG ACTGGCTGGT AGGTCCCTTT GGAAAAGATG CACCCTTCAA CCTTTGCTGG AGCCAAGTAG TCTCCAGGGA GGAAGAGGAT TGTGAGGGAG ATGCTCAGTG AACTACCTGC TCAGGGCTCA AGCCACCATG ATTGAAGTCC GCCTGTGGGA AATAGGATGA CTGTTGGTGA TCAGGGTGTG GGTGCAGGTT AGGGAGATCA TCTGAGGTGC CTGAACTCCT CCTCAGACTG CTGGACCTTC TGCTGGATCT TGCTGTTTTG TGCAGACAGA TCCAGTCCAG Product Size (bp) 123 125 177 124 107 107 127 155 75 113 112 Table 8 (cont'd) RefSeq Human NM_001101 Gene name Gene symbol Entrez Gene ID actin, beta ACTB 60 glyceraldehyde-3-phosphate dehydrogenase peroxisome proliferative activated receptor, alpha acyl-Coenzyme A oxidase 1, palmitoyl pyruvate dehydrogenase kinase, isozyme 4 cytochrome P450, family 4, subfamily A, polypeptide 11 adipose differentiation-related protein angiopoietinlike protein 4 GAPDH 2597 PPARA 65260 ACOX1 51 PDK4 5166 CYP4A11 1579 ADFP 123 ANGPTL4 51129 NM_001003940 Bcl2 modifying factor BMF 90427 NM_001001547 CD36 molecule CD36 948 NM_002046 NM_005036 NM_004035 NM_002612 NM_000778 NM_001122 NM_139314 122 Forward Primer Reverse Primer CATCCCCCAA AGTTCACAAT GGCCTCCAAG GAGTAAGACC GCAGAAACCC AGAACTCAGC TTTCTTCACTG CAGGGCTTT TCTGAGGCTG ATGACTGGTG GAGGAATGCC TTTCACCAGA ACACCCTCCT GTCCAACATC TCCGTACCCT TCTCCACTTG CCTGAGAACT GAGCCCAGAC AGATGCAGCC TCATTTCCAC AGTGGGGTGG CTTTTAGGAT AGGGGTCTAC ATGGCAACTG ATGGCCCAGT GTAAGAAACG GGAAAGGAGG GATTTTGAGC CAAACATTCA GGAAGCAGCA GTTGAGCCTT CCTCAGTTGG GCATTGCGGA ACACTGAGTA AGTACTGGCC GTTGAGGTTG GAGTGAGTTC CTGGCTTTGC GCCTTGGATG GAAGAACAAA Product Size (bp) 125 147 141 115 137 125 103 124 112 150 Table 8 (cont'd) NM_002810 proteasome (prosome, macropain) 26S subunit, non-ATPase, 4 NM_004159 proteasome (prosome, macropain) subunit, beta type, 8 (large multifunctional peptidase 7) NM_001540 heat shock 27kDa protein 1 PSMD4 Entrez Gene ID 5710 PSMB8 5696 HSPB1 3315 NM_000128 coagulation factor XI F11 2160 NM_000063 complement component 2 C2 717 RefSeq Gene name Gene symbol 123 Product Size (bp) 105 Forward Primer Reverse Primer AGCCATTCGA AATGCTATGG CACGGGTAGT GGGAACACTT GCTACCCTTTC CCTCCAGTC GACAACGCCT CCAGAATAGC ACGAGATCAC CATCCCAGTC AAGCAGTGTG AATGGGTTCC GCGTTGCCAT TATCACCTTT CTTTACTTGGC 91 GGCAGTCTC TACAAACACCA 133 AGCCCCTTC AGGCTGCTGAT 94 CACCTCAGT 142 Figure 21. Time course study designs. The in vivo time course studies consisted of a single dose. Animals were sacrificed 2-24 hrs after exposure (A) or received a daily dose and sacrificed 1-14 days after exposure (B) with either 300 mg/kg body weight clofibrate, Wy-14,643 or sesame seed oil vehicle, n = 5. (C) For the in vitro time course study, HL1-1 or HepG2 cells were treated with either 50 uM Wy-14,643 or 0.1% DMSO and harvested 1-48 hrs post treatment, n = 3. (D) Microarray experimental design for the mouse in vivo time course study. Temporal gene expression changes were analyzed using an independent reference design. PP treated samples and time-matched vehicle controls were hybridized on independent arrays using one-color labeling (Cy3), n = 3. Mouse 4 × 44 K Oligo Microarrays (Agilent Technologies, Inc, Santa Clara, CA) were used for global gene expression analysis. T: PP treatment and V: sesame oil vehicle treatment. 124 Figure 21 (Cont’d) A Mouse in vivo single dosing study Clofibrate: 300 mg/kg 0 Wy-14,643: 300 mg/kg (hrs) 2 4 8 12 18 24 Sacrifice B Mouse in vivo multiple dosing study 1 2 0 3 4 5 6 7 8 9 10 11 12 13 Dosed once daily (Days) 1 Sacrifice 4 C Human in vitro TC study Wy-14,643: 50 M 0 (hrs) 1 2 4 8 12 48 Harvest 24 D 2T 4T 8T 12T 18T 24T 1dT 4dT 14dT 2V 4V 8V 12V 1dV 4dV 14dV 18V 24V 125 14 RESULTS BODY WEIGHT, RELATIVE LIVER WEIGHT (RLW) AND HISTOPATHOLOGY PPs did not elicit significant changes in body weight or RLWs following acute exposure in mice (Figure 22). However, RLWs were induced (p < 0.05) by 300 mg/kg WY at 4 and 14 days, and by 300 mg/kg CLO at 14 days compared to time- and vehicle-matched controls. Although there were no significant histological findings following acute exposure (>24 h), subchronic treatments (4-14 days) induced hyperplastic and hypertrophic hepatocyte growth and hepatomegaly, with no signs of irreversible hepatocellular injury (Figure 23). TEMPORAL GENE EXPRESSION CHANGES Gene expression was assessed using Agilent microarrays containing ~44,000 oligonucleotide probes representing ~34,200 annotated genes of which 21,000 are unique. Genes were identified as differentially regulated if the P1(t) > 0.99 and |fold change| > 1.5 compared to time-matched vehicle controls. In total, 719 and 1,443 genes were differentially expressed by clofibrate and Wy-14,643, respectively, at one or more time points. The number of differentially expressed genes steadily increased between 2 and 8 h following both treatments, with additional increases elicited by Wy-14,643 after 24 h and fewer changes elicited by clofibrate after 12 h (Figure 24A). Hierarchical clustering revealed early treatments (2-8 hrs) cluster by time, whereas later doses (12-24 hrs) cluster by treatment (Figure 25). In total, 321 genes were induced while 398 genes were repressed by clofibrate with changes ranging from 81.3-fold induction for Pdk4 (pyruvate dehydrogenase kinase, isoenzyme 4) to 7.9-fold repression for Bmf (Bcl2 modifying factor). Wy-14,643 induced 1,060 genes, while 383 genes were repressed, with fold changes ranging from +123-fold induction for Pdk4 to -177-fold repression for Fmo3 (flavin containing monooxygenase 3). The greatest overlap in the 126 A Body Weight 40 vehicle clofibrate Wy-14,643 Weight (g) 30 20 10 0 2 4 8 12 18 24 Time (h) 1d 4d 14d B Relative Liver Weight Relative liver weight (%) 20 vehicle clofibrate Wy-14,643 15 * 10 * * 5 0 2 4 8 12 18 24 Time (h) 1d 4d 14d Figure 22. Clofibrate and Wy-14,643 effects on (A) body weight and (B) relative liver weight (RLW). Immature ovariectomized C57BL/6 mice were orally gavaged with 300 mg/kg clofibrate, Wy14,643 or sesame oil vehicle at time 0 and every 24 hrs thereafter. Mice were sacrificed 2, 4, 8, 12, 18 or 24 hrs after the initial dose for the acute study and 1, 4 or 14 days for the sub-chronic study. Bars are mean ± standard error (SE). *p < 0.05 compared to vehicle controls within a time-point, N = 5. 127 A Vehicle B clofibrate C Wy-14,643 4 days D E F 14 days Figure 23. Representative histopathology results from vehicle-, clofibrate-, and Wy-14,643-treated mice at 4 and 14 days. Sub-chronic Wy-14,643 treatments (C, F) induced hyperplastic and hypertrophic hepatocyte growth and hepatomegaly without any signs of irreversible hepatocellular injury. Bars=50um. 128 number of differentially expressed genes between clofibrate and Wy-14,643 occurred 8 h postdose (Figure 24A). To include marginal responses, relaxed filtering criteria (P1(t) > 0.9 and |fold change| > 1.3) were applied. The number of differentially regulated genes increased to 3,600 and 6,639 for clofibrate and Wy-14,643, respectively (Figure 24B), with an overlap of 90% for clofibrate and 65% for Wy-14,643. Correlation analysis of the 1,221 overlapping differentially expressed genes revealed a high correlation between clofibrate and Wy-14,643 gene expression profiles indicating similarities in significance and expression profiles (Figure 24C). A subset of prototypical mouse PPAR responsive genes (Acox1, Pdk4, Cyp4a10 and Cyp4a12) was verified by QRT-PCR. Overall, there was good agreement between microarray and QRT-PCR data for temporal gene expression profiles and induction levels (Figure 26). There was some evidence of data compression for Pdk4 and Cyp4a12 when microarray data was compared to QRT-PCR due to the limited dynamic fluorescence intensity range (0 ~ 65,535) [27]. FUNCTIONAL ANNOTATION OF CONSERVED RESPONSIVE GENES Fifty common differentially expressed genes with the highest induction at 8 h which is the most overlapping time point between clofibrate and Wy-14,643 treatments were functionally annotated with the Database for Annotation, Visualization and Integrated Discovery (DAVID) (Figure 27). Over-represented functions were associated with lipid or carbohydrate metabolism, transport, cell cycle and immune response. For example, over-represented lipid or carbohydrate metabolism genes included prototypical PPAR responsive genes such as Acot2 and 4 (acylCoAs hydrolyze), Cd36 (fatty acid transport), Cpt1 (mitochondrial-oxidation), Pdk4 (glucose metabolism) and Cyp4s (medium- & long-chain fatty acid metabolism). However, there were 129 Figure 24. Mouse hepatic temporal response comparison between clofibrate and Wy-14643 treatment. (A) Number of genes exhibiting differential expression (P1(t) > 0.99 and |fold change| >1.5) in response to clofibrate (CLO) or Wy-14,643 (Wy). (B) Comparison of genes differentially expressed following clofibrate and Wy-14,643 treatments. Venn analysis using stringent filtering criteria (P1(t) > 0.99 and |fold change| > 1.5) and relaxed criteria (P1(t) > 0.9 and |fold change| > 1.3) are shown. Shaded areas represent number of marginal differentially expressed genes which are included by relaxed criteria. The numbers represent unique genes. (C) Correlation plot of genes differentially expressed by clofibrate and Wy-14,643 using the relaxed filtering criteria. Correlation analysis was used to compare significance and expression profile to identify similarities and differences between clofibrate and Wy-14,643 temporal data sets. A majority of genes (80.5%) were found within the upper right-hand quadrant indicating clofibrate and Wy14,643 elicit similar gene expression and significance profiles. 130 Figure 24 (Cont’d) Number of Differentially Expressed Genes A 1000 CLF Wy 800 600 400 200 0 2 4 8 12 18 24 1d 4d 14d Time (h) B |Fold change| >1.5 P1(t) > 0.99 |Fold change| >1.3 P1(t) > 0.9 CLF (719) CLF (719) 354 Wy (1,443) 365 74 280 365 1,078 Wy (1,443) 576 502 Comparison between CLF and WY active genes Overlap: 645/719 = 89.7% of CLF active genes 941/1,443 = 65.2% of WY active genes 131 Figure 24 (Cont’d) C 4.0% Correlation of fold change expression 80.5% 1.0 Correlation of significance 0.5 -1.0 -0.5 0.5 1.0 -0.5 -1.0 1.5% 14.0% 132 Wy-2h Wy-4h Wy-8h CL-12h Wy-14d Wy-18h Wy-1d CL-24h CL-4d CL-2h CL-4h CL-8h Wy-4d Wy-12h Wy-24h CL-18h CL-1d CL-14d 0.3 1.0 3.0 Figure 25. Hierarchical clustering of 410 differentially expressed genes by gene and time. Both Wy-14,643 and clofibrate treatment clustered together at early time points (2 ~ 8 h). At later time points (12 h+) the clustering is separated according to chemical treatment. Each bar represents fold change of gene expression compared with time-matched vehicle control. Purple bars indicate clustered groups. Wy: Wy-14,643, CL: clofibrate. 133 temporal differences between clofibrate and Wy-14,643 elicited differential gene expression. For example, Wy-14,643 induced gene expression at all time points, while induction by clofibrate diminished after 8 h. Among the 1,797 unique genes differentially expressed by clofibrate or Wy-14,643, 1,110 were induced and 653 were repressed, while 34 showed a mixed response. Overrepresented induced genes were involved in the proteasome complex, PPAR signaling, fatty acid metabolism, peroxisome organization, and mitochondrial transport (Table 9). All of these over-represented functions are consistent with the role of PPAR in lipid metabolism. Moreover, the differential expression of DNA repair and cell cycle related genes is consistent with reports of PPAR induced DNA damage and cell proliferation, related to tumor induction [38, 39]. For repressed genes, over-represented functions were associated with amino and nucleotide sugar metabolism, complement and coagulation cascades, and glycosphingolipid biosynthesis, oxidation reduction, and immune response (Table 10). The insulin signaling, lipid process and storage, and high-density lipoprotein particle related pathways were also enriched as well as apoptosis and ErbB signaling. PATHWAY MAPPING ANALYSIS Over-represented functions, such as lipid metabolism, were mapped onto KEGG pathways (Table 11, Figure 28A). Several fatty acid transport & oxidation, VLDL receptor and lipoprotein lipase-related genes were induced at all time points. However, other lipogenesis and HDL component-related genes were down regulated at late time points. Proteasome subunit components as well as heat shock proteins (HSPs) were also induced (Table 11, Figure 28B). The 26S proteasome and HSPs play key roles in protein maintenance and endoplasmic reticulum (ER) stress protection. In contrast, genes associated with complement and coagulation cascades 134 Figure 26. QRT-PCR verification of selected microarray time course results. Proto-typical PPAR responsive genes are indicated by the official gene symbol. The same RNA used for microarray analysis was examined by QRT-PCR. Bars (left axis) and lines (right axis) represent QRT-PCR and microarray data, respectively. Bars are mean ± SE for the average fold change, *p < 0.05 for QRT-PCR, N = 3. 135 Figure 26 (Cont’d) Wy-14,643 Clofibrate 12 Acox1 * 2 4 8 12 18 Time (hr) 24 800 3 2 2 2 * 4 4 * 2 1 1 0 0 * * 4 * 3 * 2 4 8 12 18 Time (hr) 24 200 300 Pdk4 150 Pdk4 * 600 0 150 240 * 120 * 180 400 90 100 * 120 60 * 200 0 50 * 2 * 4 * 8 12 18 Time (hr) 60 30 * 24 0 0 136 2 4 8 12 18 Time (hr) 24 0 Fold change (Array) Fold change (QRT-PCR) 5 * 6 * * * 4 0 Acox1 8 8 6 5 6 Fold change (Array) Fold change (QRT-PCR) 10 6 10 12 Figure 26 (Cont’d) 20 * 15 Cyp4a10 8 15 * 10 * 0 * 8 * 2 4 10 * 4 * * 5 50 8 12 18 Time (hr) Cyp4a12 24 * 0 0 4 * 50 2 2 4 60 40 6 8 12 18 Time (hr) 24 0 30 Cyp4a12 * 25 * 20 30 20 15 20 10 10 2 4 2.0 8 12 18 Time (hr) * 20 * 0 40 30 30 24 10 0 0 10 5 2 4 2.0 2.0 AhR 8 12 18 Time (hr) 24 0 2.0 AhR 1.5 1.5 1.5 1.0 1.0 1.0 1.0 0.5 0.5 0.5 0.5 2 4 8 12 18 24 Time (hr) 0.0 0.0 137 2 4 8 12 18 24 Time (hr) 0.0 Fold change (Array) 1.5 0.0 Fold change (Array) * * 2 50 * 40 Fold change (QRT-PCR) 10 6 5 Fold change (QRT-PCR) 10 Fold change (Array) Fold change (QRT-PCR) Cyp4a10 20 Figure 27. Functional categorization of the top 50 genes differentially expressed by Wy14,643 and clofibrate treatment. Genes were ranked by induction ratio and functionally annotated using the Database for Annotation, Visualization and Integrated Discovery (DAVID). Each bar represents the average fold change in gene expression compared with time-matched vehicle control. 138 Figure 27 (Cont’d) Wy-14,643 Clofibrate 2h 4h 8h 12h 18h 24h 1d 4d 14d 2h 4h 8h 12h 18h 24h 1d 4d 14d Lipid/ CHO Metabolism Transporter Cell cycle Regulation Immune Response Others 139 Table 9. Over-represented gene ontology groups induced following PP treatment. BP: biological process, CC: Cellular component Category  KEGG_PATHWAY  mmu03050  mmu03320  mmu00071  mmu01040  mmu00280  mmu04115  mmu00020  mmu02010  GOTERM_BP  GO:0006631  GO:0007031  GO:0006839  GO:0030163  GO:0070585  GO:0006412  GO:0006511  GO:0008610  GO:0006457  GO:0016042  GO:0007006  GO:0006637  GO:0006413  GO:0051004  GO:0043574  GO:0015908  GO:0006281  GO:0010883  GO:0007049  GOTERM_CC  GO:0005739  GO:0005777  GO:0000502  Term  Count   Proteasome  PPAR signaling pathway  Fatty acid metabolism  Biosynthesis of unsaturated fatty acids Valine, leucine and isoleucine  degradation  p53 signaling pathway  Citrate cycle (TCA cycle)  ABC transporters   fatty acid metabolic process  peroxisome organization  mitochondrial transport  protein catabolic process protein localization in mitochondrion  translation  ubiquitin‐dependent protein catabolic  process  lipid biosynthetic process  protein folding  lipid catabolic process  mitochondrial membrane organization  acyl‐CoA metabolic process translational initiation  regulation of lipoprotein lipase activity  peroxisomal transport  fatty acid transport DNA repair  regulation of lipid storage  cell cycle    mitochondrion  peroxisome  proteasome complex  140 22 25 17 11 11 11 6 7   38 12 14 60 10 40 21 32 19 19 8 7 9 3 3 4 20 3 42 P Value  Benjamini   5.04E‐14  8.87E‐12 1.68E‐11  1.48E‐09 3.33E‐09  1.95E‐07 2.08E‐06  9.15E‐05 3.54E‐04  1.03E‐02 8.46E‐03  1.17E‐01 3.76E‐02  2.86E‐01 5.38E‐02  3.71E‐01     2.18E‐12  4.72E‐09 2.15E‐09  1.55E‐06 1.67E‐07  5.17E‐05 3.80E‐07  8.23E‐05 9.38E‐07  1.85E‐04 1.29E‐06  2.33E‐04 6.39E‐05  5.31E‐03 1.47E‐04  1.02E‐02 1.48E‐04  9.96E‐03 2.93E‐04  1.70E‐02 3.36E‐04  1.90E‐02 4.77E‐04  2.61E‐02 7.76E‐04  4.11E‐02 8.34E‐03  2.54E‐01 2.59E‐02  5.35E‐01 2.95E‐02  5.78E‐01 3.04E‐02  5.85E‐01 5.06E‐02  7.01E‐01 8.40E‐02  8.11E‐01       199 1.89E‐41  8.03E‐39 35 8.86E‐18  4.17E‐16 25 1.42E‐15  5.56E‐14 Table 10. Over-represented gene ontology groups repressed following PP treatment. BP: biological process, CC: Cellular component Category  Term  Count KEGG_PATHWAY    mmu00520  Amino sugar and nucleotide sugar  metabolism  mmu04610  Complement and coagulation cascades  mmu00604  Glycosphingolipid biosynthesis  mmu00330  Arginine and proline metabolism mmu04910  Insulin signaling pathway  mmu00983  Drug metabolism  mmu04210  Apoptosis  mmu04142  Lysosome  mmu05214  Glioma  mmu02010  ABC transporters  mmu04012  ErbB signaling pathway  GOTERM_BP    GO:0009063  cellular amino acid catabolic process GO:0055114  oxidation reduction  GO:0007264  small GTPase mediated signal  transduction  GO:0006955  immune response  GO:0007010  cytoskeleton organization GO:0016042  lipid catabolic process  GO:0008202  steroid metabolic process  GO:0008652  cellular amino acid biosynthetic  process  acute inflammatory response  GO:0002526  GO:0010876  lipid localization  GO:0032869  cellular response to insulin stimulus  GO:0008610  lipid biosynthetic process  GO:0019915  lipid storage  GOTERM_CC    GO:0005764  lysosome  GO:0005783  endoplasmic reticulum  GO:0005739  mitochondrion  GO:0034364  high‐density lipoprotein particle  141 P Value  Benjamini   8 4.97E‐04  7.78E‐02 10 5 8 12 6 8 9 6 5 7 6.96E‐04  1.15E‐03  1.56E‐03  5.34E‐03  1.96E‐02  2.33E‐02  4.08E‐02  5.76E‐02  5.86E‐02  6.49E‐02  5.52E‐02 6.05E‐02 4.96E‐02 1.35E‐01 3.01E‐01 2.95E‐01 3.64E‐01 4.54E‐01 4.40E‐01 4.38E‐01       9 35 18 1.82E‐04  1.80E‐03  1.91E‐03  1.56E‐01 3.81E‐01 3.59E‐01 26 20 11 12 6 3.86E‐03  4.23E‐03  6.80E‐03  8.86E‐03  9.59E‐03  5.14E‐01 4.83E‐01 5.07E‐01 5.15E‐01 5.13E‐01 8 9 5 15 3 1.04E‐02  3.77E‐02  3.85E‐02  4.69E‐02  5.56E‐02  5.28E‐01 7.90E‐01 7.90E‐01 8.22E‐01 8.52E‐01     18 46 52 3   2.79E‐05  1.19E‐04  4.13E‐02  8.28E‐02  8.28E‐03 8.83E‐03 4.84E‐01 6.01E‐01 pathway, which are related to immune and inflammatory responses, were down regulated (Table 11, Figure 28C). IN VITRO PPAR EXPRESSION AND CELL CYTOTOXICITY Basal PPAR mRNA expression levels were comparable in HL1-1 human liver stem cells, HepG2 human hepatoma cells and Hepa1c1c7 mouse hepatoma cells and mouse liver based on RT-PCR (Figure 29A). Protein expression was also confirmed by western blot in each model (Figure 29B). The cytotoxic effects of Wy-14,643, clofibrate and fenofibrate were examined in HL1-1 and HepG2 cells. Overall, HL1-1 cells were more resistance than HepG2 to PP treatment (Table 12 and Figure 30). Wy-14,643 elicited the greatest cytotoxicity with an LC50 value of 740 and 459 uM in HL1-1 and HepG2 cells, respectively. In HL1-1 cells, fenofibrate elicited minimal cytotoxicity (LC50 2,600 uM) followed by clofibrate (1,300 uM). In HepG2 cells, clofibrate was the least toxic (1,470 uM) followed by fenofibrate (767 uM). Neither cell line exhibited cytotoxicity below 100 uM. COMPARISON OF MOUSE IN VIVO GENE EXPRESSION WITH HUMAN IN VITRO MODELS HL1-1 and HepG2 temporal gene expression responses elicited by 50 uM Wy-14,643 were compared for selected genes by QRT-PCR (Figure 21C). Orthologous lipid/cholesterol metabolism genes including PDK4, ADFP and ANGPTL4, showed comparable responses across species although there were temporal and magnitude differences (Figure 31). However, several species-specific responses were identified. For example, BMF (Bcl2 modifying factor), a gene associated with apoptosis signaling and tumor suppression was repressed in mouse liver but induced in HL1-1 and HepG2 cells. Other PPAR responsive genes, including ACOX1, CD36 and CPT1A, were induced in mouse liver but not differentially expressed in HL1-1 cells (Figure 142 32). Furthermore, proteasome component (PSMD4 and PSMB8), heat shock protein (HSPB1) and coagulation factor (F11) genes were differentially expressed in mice treated with PP, but were not changed in HL1-1 cells. DISCUSSION Temporal gene expression differences between Wy-14,643 and clofibrate that can be partially explained by their different plasma half-lives (clofibrate ~2 hrs [40]; Wy-14,643 ~1 h [41]) and pharmacodynamic properties. In addition, Wy-14,643 exhibits non-linear absorption and elimination behavior when compared to clofibrate [41]. Cell-based transactivation assays, with limited metabolic capabilities, also indicate Wy-14,643 (EC50 = 0.63uM compared to 50uM for clofibrate) is more potent and has greater (30-150-fold) relative binding affinity for PPARα [42, 43]. Although PPARα typically has a short half-life due to proteasome degradation, ligand binding protects it from ubiquitination, resulting in prolonged ligand-induced differential gene expression [44]. Collectively, these differences are consistent with the larger number of differentially expressed genes and prolonged expression elicited by Wy-14,643 when compared to clofibrate. Analysis of differentially expressed genes identified over-represented functions associated with fatty acid metabolism. Several fatty acid oxidation related genes, including mitochondrial -oxidation (Cpt1, Acad, Hadha, and Ucp2), peroxisomal -oxidation (Acox1) and microsomal -hydroxylation (Cyp4As), were strongly induced at all time points. Fatty acid transport (Cd36), lipoprotein lipase (Lpl), and hormone sensitive lipase (Hsl) were also induced, while lipid catabolism, HDL particle and insulin signaling related genes were repressed. These changes are consistent with increases in fatty acid-metabolism and transport, and the repression of lipogenesis resulting in the clearance of hepatic and plasma lipids. 143 Table 11. PP elicited differentially expressed genes used for pathway mapping analysis GeneID  Symbol  Wy  Time points  Clo  Time points  Functional  Description  Lipid metabolism related genes           De novo synthesis    20787  Srebf1      Down 8h  activate  lipogenesis  14104  Fasn  Down  12h  Down 12h  malonyl CoA ‐>  palmitate (C16:0)  FA transport    11520  Adfp  Up  2,8,12h  Up    FA transport  related  12491  Cd36  Up  4,8,12,18,24h,4, Up  4,8,12,18,24h, FA translocalase  14d  4,14d  (FAT)  19299  Abcd3  Up  4,8,12,18,24h,4, Up  4,8,12h  FA peroxisomal  14d  import        Fatty acid metabolism (oxidation)  12894  Cpt1a  Up  4,12,18,24h  Up  4,12h  mitochondrial ‐ oxidation   12895  Cpt1b  Up  2,4,8,12,18,24h, Up 2,4,8,12,18h rate limiting  4,14d  enzyme  11363  Acadl  Up  4,8,12,18,24h,4, Up 4,8,12,18h mitochondrial ‐ 14d  oxidation  11364  Acadm  Up  24h      11409  Acads  Up  24h,14d      97212  Hadha  Up  4,8,12,18,24h,4, Up  2,4,8,12,18,24 14d  h  11430  Acox1  Up  4,8,12,18,24h,4, Up  4,8,12,18h  peroxisomal ‐ 14d  oxidation  93732  Acox2  Up  24h      13117  Cyp4a10  Up  4,8,12,18,24h,4, Up 4,8,12,18,24h, microsomal ‐ 14d  14d  oxidation  277753  Cyp4a12a  Up  2,4,8,12,18,24h, Up  2,4,8,12,18,24 4,14d  h,4,14d  666168  Cyp4a31  Up  4,8,12,18,24h,4, Up  4,8,12,18,24h  14d  Esterification        13350  Dgat1  Up  24h      esterification  67800  Dgat2  Down  8,12,18h,14d  Down 8h    Lipid secretion        17777  Mttp  Up  4,8,12,18,24 h Up 4h make VLDL. Lipid  secretion  144 Table 11 (cont'd) GeneID  Symbol  Wy  Lipoprotein related genes  HDL related  11808  Apoa4  Down  Time points  Clo     Time  points     18h,4,14d  Down 18h,14d  Down 18h  20778  Scarb1  Down  8,18h,4d  11307  Abcg1  26357  Abcg2  Up  Up  Up  8,12h  18830  Pltp  Up  14d  8,12,18,24h,4,14 d  24h,4,14d        Up    4h  Up  4,8h              Up     14d Up  4h  VLDL related  17777  Mttp  Up  4,8,12,18,24 h  22359  Vldlr  Up  8,12,18,24  h,4,14d  Chylomicron related  228357  Lrp4  Up  Lipase related  11814  Apoc3  16956  Lpl  16890  Lipe  Down  Up  Up  24h  4,14d  8,12,18,24h,4,14 d  4,12,24h  145 Functional  Description     HDL component,  LCAT activator  receptor for HDL  (detect ApoA)  cholesterol transfer  to ApoA1 to make pre  b‐HDL  CE & TG transfer   between lipoproteins    make VLDL. Lipid  secretion  receptor for VLDL    receptor for  Chylomicron  remenant    inhibit LPL  lipoprotein lipase hormone sensitive  lipase  Table 11 (cont'd) GeneID  Symbol  Wy  Proteasome pathway  alpha subunit  26440  Psma1  Up  19166  Psma2  Up  19167  Psma3  Up  26441  Psma4  Up  26442  Psma5  Up  26443  Psma6  Up  26444  Psma7  Up  73677  Psma8  Up  beta subunit  19170  Psmb1  Up  26445  Psmb2  Up  26446  Psmb3  Up  19173  Psmb5  Up  19175  Psmb6  Up  19177  Psmb7  Up  16913  Psmb8  Down  16912  Psmb9  Down  19171  Psmb10  Down  26S subunit  Up  19182  Psmc3  23996  Psmc4  Up  67089  Psmc6  Up  70247  Psmd1  Up  21762  Psmd2  Up  19185  Psmd4  Up  66998  66413  17463  57296  67151  69077  66997  23997  59029  Psmd5  Psmd6  Psmd7  Psmd8  Psmd9  Psmd11  Psmd12  Psmd13  Psmd14  Up  Up  Up  Up  Up  Up  Up  Up  Up  Time points  Clo    Time  points    Functional  Description    18,24h,14d  18,24h  24h,14d  24h  4,8,12,18,24 h 18,24h,14d  24h  4,14d                  20S Core particles  alpha subunits              24h,14d  18,24h  24h  18,24 h,4,14d  24h  24h,14d  8,18,24 h,4,14d  8,12,18h  12,18h                     Down 8,12h,4d  Down 8,12h        Up    Up        Up   Up    Up  146   4,12h      24h  24h  18,24h  24h  12,24h  4,8,12,18,24  h,4,14d  24h  4,12,18,24 h  4,24h  24h  24h  12,24h  4,12,24h  24h  18,24h,14d        8h    4d   8h    8h  20S Core particles  beta subunits  Immunoproteasome  subunits    Regulatory particles  PA700 (Base)  Regulatory particles  PA700 (Lid)  Table 11 (cont'd) GeneID  Symbol  other subunit  107047  Psmg2  103554  Psme4  228769  Psmf1  Chaperone (Hsp)  193740  Hspa1a  15519  Hsp90aa1  Wy  Time points  Clo  Time points        Up    4h  Functional  Description    Assembly  chaperone  PA200  PI31     Hsp70  Hsp86/90  Up  24h  Up  Up       Up  Up  2,4,8,12,24h  Up  2,4,8,24h,4,14d Up  2,4,12h  2,4,8h (4d  down)  24h  4,12,24h  4,8,12,18,24  h,4,14d  4,8,12,18,24  h,4,14d  24 h,4,14d  2,4h  Up  Up    4h  Hsp110  8,12,18h,4,14d Hsp25/27  Up  8,18h  Hsp60    Up  Hsp10  Hsp105/110  Up      2,4,8h (4d  down)  8,12h        15516  Hsp90ab1  Up  15525  Hspa4  Up  15507  Hspb1  Up  15510  Hspd1  Up  15528  Hspe1  15505  Hsph1  Up  Up  18415  Hspa4l  Up  24h,4,14d Complement & coagulation cascades  pathway  complement factor  12261  C1qbp  Up  24h  12260  C1qb  Down  18h,4d  50909  C1r  Down  18h,4,14d 50908  C1s  Down  18h,4,14d  12263  C2  Down  8,12,18,24  h,4,14d  12268  C4b  Down  18h,4,14d  230558  C8a  Down  18,24h,4,14d 110382  C8b  Down  18,24h,4,14d  69379  C8g  Down  14d  12279  C9  Down  18,24h,4,14d  12628  Cfh  Down  18h,4,14d  50702  Cfhr1  Down  12,18h,14d  18636  Cfp  Down  18h  Down  8h  Down  8h  Down  8h  Down  8h               147                      Complement  cascade  Table 11 (cont'd) GeneID  Symbol  Wy  Time points  coagulation factor  14066  F3  Up  12,24h 14067  F5  Down  8,18h,4,14d  14068  F7  Down  8,12,18,24  h,4,14d  109821  F11  Down  8,12,18,24  h,4,14d  others  17174  Masp1  Down  8,12,18h,4,14d 18816  Serpinf2  Clo  Time points      Functional  Description    Coagulation cascade Down  8h  Down  8,12h  Down  8,12h      Down  18h  148       activation of the  lectin pathway  Progression of fibrosis A Acetyl‐CoA 2.3.1.16 CoA CoA 2.3.1.16 2.3.1.9 Acetoacetyl‐ CoA 1.1.1.35 (S)‐3‐Hydroxyl Butanoyl‐CoA 3‐Oxo‐ Hexanoyl‐CoA 1.1.1.211 3‐Oxo‐ Hexanoyl‐CoA 1.1.1.35 1.1.1.211 (S)‐3‐Hydroxyl Butanoyl‐CoA (S)‐3‐Hydroxyl Butanoyl‐CoA 4.2.1.17 Figure 28. Pathway mapping analysis. (A) Fatty acid metabolism pathway, (B) Proteasome pathway, (C) Complement & coagulation cascades pathway. Differentially expressed genes by Wy-14,643 were mapped to the targeted pathway using KEGG Mapper pathway mapping tools (http://www.genome.jp/kegg/ mapper.html). Colored genes represent differentially expressed genes. Red: induction, blue: repression. 149 Figure 28 (Cont’d) B IFNG PSMC PSMD PSMA2 PSMA6 PSMA4 PSMA3 PSMA1PSMA7 PSMB2 PSMA5 PSMB5 PSMB3 PSMB4 PSMB6 PSMB1PSMB7 PSMB8 PSMB9 PSMB10 PSMB5 PSMB6 PSMB7 PSME2 PSME1 Processed protein Polypeptide antigen 150 Figure 28 (Cont’d) C MASP2 MASP1 SERPIN G1 C4BP 151 PPs also induced stress modifier genes in mouse liver, which maintain the proteome. Proteome maintenance genes consist of heat shock proteins (Hsps, chaperones) and proteasome components (26S proteosome). Hsps stabilize unfolded proteins preventing aggregation while the ubiquitin-proteasome pathway removes aged, damaged, and misfolded proteins. Induction of proteome maintenance by PPARα suggests a protective role in response to oxidative stress [45], since liver and primary hepatocytes from null mice are more sensitive to chemical induced oxidative stress [46]. In the present study, except alternative β form subunits (Psmb8, 9 and 10) for the immuno-proteasome assembly, most of proteasome subunits were induced. Proteasome inhibitor Psmf1 (PI31) was also induced which interferes with the maturation of immuneproteasome precursor complexes [47]. Moreover, most Hsp genes were induced by PP in mouse liver. In fact, induction of stress responsive genes activation was shown at late time points (8 ~ 24 h), possible due to oxidative stress as a result of fatty acid oxidation from increased peroxisomal activation. The over-representation of genes associated with cell cycle and apoptosis is consistent with hepatomegaly as well as hyperplastic and hypertrophic hepatocyte growth. The induction of several cell cycle and DNA damage repair genes, such as Chek1, Prkdc, Mcm, and Rad51, by Wy-14,643 is abolished in PPARα null mice [48]. This suggests sustained PPARα activation leads to impaired cell cycle regulation and increased DNA damage by oxidative stress possibly contributing to hepatocarcinogenicity in rodents since Wy-14,643 is considered a non-genotoxic hepatocarcinogen in rodents [38, 39]. In mouse liver, repressed genes were associated with immune and acute inflammatory responses consistent with PP repression reported in cynomolgus monkey [49] and rat [50]. The complement cascade triggers inflammatory responses while the coagulation pathway contributes 152 B PPAR /HK genes Ratio (Log) A 0 -1 HL1-1 HepG2 Hepa 1c1c7 Mouse liver -2 -3 -4 -5 -6 HL1-1 HepG2 Hepa1c1c7 Mouse liver Figure 29. Basal PPAR mRNA and protein levels. (A) PPAR mRNA levels in HL1-1, HepG2, and Hepa1c1c7 cells as well as C57BL/6 mouse liver were examined by QRT-PCR using species-specific primers. PPAR mRNA levels are normalized to the geometric mean of housekeeping genes (HK) (Hprt, Gapdh and Actb). Bars are mean ± SE, N = 3. (B) Representative Western blot for PPAR. 153 A B HepG2 HL1‐1 RelativeAbs595 1.0 0.5 0.0 1.0 0.5 0.0 0 1 2 3 Log Dose (uM) 4 Wy14,643 FNF CLF 1.5 RelativeAbs595 Wy14,643 FNF CLF 1.5 0 1 2 3 Log Dose (uM) 4 Figure 30. Cytotoxicity testing. Cytotoxicity of Wy-14,643, fenofibrate (FNF) and clofibrate (CLF) were measured by MTT assay in (A) human liver stem cells HL11 and (B) human hepatoma cells HepG2. Bars are mean ± SE, N = 3. 154 Table 12. Cytotoxicity of PPs in human cell lines    IC50 (M)     Wy‐14,643  fenofibrate  clofibrate  HL1‐1  740 ± 141  2607 ± 712  1301 ± 223  HepG2  459 ± 108  767 ± 113  1473 ± 233  155 700 180 150 400 120 * 300 90 200 100 0 60 * * 2 4 10 Fold change (QRT-PCR) 500 6 4 * 0 8 12 18 24 Time (hr) 6 4 2 0 3 Fold change (QRT-PCR) Fold change (QRT-PCR) HL1‐1 8 12 18 24 Time (hr) 0 2 1 1 2 6 4 * * * 2 0 4 8 12 24 48 Time (hr) * * 1 2 4 8 12 24 48 Time (hr) 8 PDK4 ADFP * 30 Fold change (QRT-PCR) Fold change (QRT-PCR) 4 ADFP 40 HepG2 2 8 PDK4 20 * 10 0 8 2 4 0 10 * 8 30 * 12 Adfp Fold change (Array) Fold change (QRT-PCR) * Fold change (Array) Mouse liver 600 12 210 Pdk4 1 2 6 * * * * 2 0 4 8 12 24 48 Time (hr) * 4 1 2 4 8 12 24 48 Time (hr) Figure 31. Comparative analysis of in vivo mouse liver and in vitro human HL1-1 and HepG2 temporal gene expression elicited by Wy-14,643. Fold changes were calculated relative to time-matched vehicle controls. Bars (left axis) and lines (right axis) represent QRT-PCR and microarray data, respectively. Bars are mean ± SE for the average fold change, *p < 0.05 for QRT-PCR, N = 3. 156 Figure 31 (Cont’d) 8 8 2.0 Angptl4 4 2 2 0 Fold change (QRT-PCR) Fold change (QRT-PCR) 4 1.5 1.5 1.0 1.0 0.5 0.5 * 2 4 8 12 18 24 Time (hr) 0.0 ANGPTL4 8 12 18 24 Time (hr) BMF * Fold change (QRT-PCR) HL1‐1 Fold change (QRT-PCR) 4 4 12 10 2 8 6 4 3 * * * 2 1 2 0 1 2 0 4 8 12 24 48 Time (hr) 4 8 12 24 48 Time (hr) BMF 6 Fold change (QRT-PCR) Fold change (QRT-PCR) ANGPTL4 HepG2 2 8 8 4 2 0 1 1 2 4 157 * 2 0 4 8 12 24 48 Time (hr) * 6 1 2 4 8 12 24 48 Time (hr) 0.0 Fold change (Array) 6 Fold change (Array) 6 0 Mouse liver 2.0 Bmf 12 600 Fold change (QRT-PCR) Mouse liver 700 Pdk4 * 10 8 400 0 * * Fold change (QRT-PCR) HL1‐1 4 1.0 0.5 0 0.0 2 * * 2h 4h 8h 12h 18h 24h Time * 0 2h 4h 8h 12h 18h 24h Time * 4 ADFP 3 * 2 1 10 * 2 * * 0 2h 4h 8h 12h 18h 24h Time 4 ANGPTL4 BMF * 3 8 2 6 * * * * 4 1 1 1h 2h 4h 8h 12h24h48h Time 2h 4h 8h 12h 18h 24h Time 12 5 PDK4 0 1.5 4 4 3 6 * 6 * 200 100 Bmf Angptl4 2 500 300 2.0 8 Adfp 2 1h 2h 4h 8h 12h24h48h Time 0 1h 2h 4h 8h 12h24h48h Time 0 1h 2h 4h 8h 12h24h48h Time Figure 32. Comparative analysis of in vivo mouse liver and in vitro human liver stem cell (HL1-1) time course studies with Wy14,643 treatment. All fold changes were calculated relative to time-matched vehicle controls. Bars represent QRT-PCR (human in vitro) and microarray data (mice in vivo), respectively. Bars are mean ± SE for the average fold change, *p < 0.05 for QRT-PCR, N = 3. 158 Figure 32 (Cont’d) 5 Acox1 4 6 3 4 2 2 Cpt1a 1 10 5 0 Fold change (QRT-PCR) 8 15 5 HL1‐1 10 Cd36 Fold change (uArray) Mouse liver 20 4 2h 4h 8h 12h 18h 24h Time CD36 2h 4h 8h 12h 18h 24h Time ACOX1 0 5 4 3 1h 2h 4h 8h 12h24h48h Time 2 2h 4h 8h 12h 18h 24h Time CPT1A 3 2 1 * 1 0 5 4 3 2 0 1 0 1h 2h 4h 8h 12h24h48h Time 159 0 1h 2h 4h 8h 12h24h48h Time Figure 32 (Cont’d) Fold change (uArray) Mouse liver 5 4 2.0 Psmd4 4 3 2.0 Hspb1 3 1.0 2 2 0.5 2h 4h 8h 12h 18h 24h Time 5 4 F11 1.5 1.0 1 Fold change (QRT-PCR) Psmb8 1.5 0 HL1‐1 5 PSMD4 0.0 2h 4h 8h 12h 18h 24h Time 5 4 0.5 1 PSMB8 0 2h 4h 8h 12h 18h 24h Time 5 4 HSPB1 0.0 2h 4h 8h 12h 18h 24h Time 5 4 3 3 3 3 2 2 2 2 1 1 1 F11 1 0 1h 2h 4h 8h 12h24h48h Time 0 1h 2h 4h 8h 12h24h48h Time 160 0 1h 2h 4h 8h 12h24h48h Time 0 1h 2h 4h 8h 12h24h48h Time to the innate immune system. The PP-induced anti-inflammatory response inhibits nuclear translocation of nuclear factor-kB (NF-kB) p65 subunit and phosphorylation of the activator protein-1 (AP-1) c-jun subunit [51], which reduces pro-inflammatory signals such as interleukin1 (IL-1), tumor necrosis factor-α (TNF-α), and inducible nitric oxide synthase (iNOS) as well as cyclooxygenase-2 (COX-2) expression [51-53]. PPARα also induces anti-inflammatory genes such as interleukin-1 receptor antagonist (IL-1ra) [54]. Moreover, leukotriene B4 (LTB4), a powerful chemotactic inflammatory eicosanoid and endogenous PPARα ligand, triggers feedback inhibition through activation of the β- and ω-oxidation pathways that degrade LTB4. PPARα null mice also have a prolonged inflammatory response when compared to wild-type controls [55]. The anti-inflammatory response and improved lipid profiles elicited by PPs have also been investigated in methionine and choline deficient (MCD) diet-induced steatohepatitis in mice [56] and as a dyslipidemia treatment for non-alcoholic fatty liver disease (NAFLD) patients [57, 58]. Comparison of in vivo mouse liver and in vitro human liver cell (i.e., HL1-1 and HepG2) responses elicited by Wy-14,643 identified common as well as divergent responses. In HL1-1 cells, lipid trafficking and metabolism related genes including CD36, PDK4, ADFP and ANGPTL4 were induced, but ACOX1 and CPT1A, which regulate fatty acid oxidation, were not affected by PP. In addition, the differential expression of lipid metabolism related genes was generally lower in HL1-1 cells compared to mouse liver. In addition, proteasome maintenance and coagulation factor genes were also unresponsive to Wy-14,643 in HL1-1 cells. Some of these species-specific responses can be explained by differences in PPARα expression levels, ligand activation, and biological responses [20, 43, 59, 60]. For example, primates and humans appear to be resistant to PP-mediated peroxisome proliferation induction and hepatocarcinoma 161 development due to lower PPARα expression levels [18, 61] and the expression of splice variants [62]. Differences in the promoter regions of PPARα target genes may also account for some species-specific response [63]. Furthermore, the expression of different coactivator proteins (i.e. PBP/MED1) necessary for PPARα-mediated transactivation between species may be a contributing factor [19, 64, 65]. The lack of exogenous soluble factors, extracellular matrix components, and cell-cell interactions can also affect in vitro cellular functions [66]. For example, HL1-1 cells lack communication with other hepatic non-parenchymal sub-populations such as Kupffer, biliary, endothelial and stellate cells that likely affect responsiveness and function. The species–specific expression of the tumor suppressor Bmf is intriguing since PPAR agonists increase hepatocellular replication and inhibit apoptosis in rodents, but induce apoptosis in human models [67, 68]. The Bcl-2 family of proteins, can be either pro-apoptotic or antiapoptotic regulators in cell proliferation/differentiation and contribute to tumorigenesis when their balance is altered. Bmf, a Bcl-2 family member, is pro-apoptotic gene which is repressed by Wy-14,643 in mouse liver, but is induced in human cell models. It binds and neutralizes antiapoptotic Bcl-2 proteins triggering Bax/Bak activation and caspase-mediated cell apoptosis [69]. Moreover, rodent-specific peroxisomal -oxidation induction produces hydrogen peroxide, and the potential for oxidative damage. Overall, -oxidation related gene changes are greater in rodent than human models and consistent with rodent-specific tumorigenesis elicited by PPs. The species-specific divergent regulation of Bmf may be another factor contributing to PP carcinogenicity in rodents but not humans. In summary, Wy-14,643 elicited gene expression responses highly overlapped with clofibrate-mediated responses and reflected PPARα-mediated regulation of fatty acid metabolism, 162 proteome maintenance, cell cycle, immune suppression and anti-inflammatory responses. HL1-1 human liver stem cells expressed functional PPARα and some prototypical PPARα response genes were differentially expressed by Wy-14,643 treatment. However there are many mouse response genes which were not responding or differentially regulated in HL1-1. This study also provides insight into key events that may contribute to the species-specific tumorigenic effects of PPs that warrant further investigation. 163 REFERENCES 164 REFERENCES 1. Desvergne, B. and W. Wahli, Peroxisome proliferator-activated receptors: nuclear control of metabolism. Endocr Rev, 1999. 20(5): p. 649-88. 2. Pyper, S.R., et al., PPARalpha: energy combustion, hypolipidemia, inflammation and cancer. Nucl Recept Signal, 2010. 8: p. e002. 3. Ahmed, W., et al., PPARs and their metabolic modulation: new mechanisms for transcriptional regulation? J Intern Med, 2007. 262(2): p. 184-98. 4. Peraza, M.A., et al., The toxicology of ligands for peroxisome proliferator-activated receptors (PPAR). Toxicol Sci, 2006. 90(2): p. 269-95. 5. Berger, J.P., T.E. Akiyama, and P.T. Meinke, PPARs: therapeutic targets for metabolic disease. Trends Pharmacol Sci, 2005. 26(5): p. 244-51. 6. Lalloyer, F. and B. Staels, Fibrates, glitazones, and peroxisome proliferator-activated receptors. Arterioscler Thromb Vasc Biol, 2010. 30(5): p. 894-9. 7. Auboeuf, D., et al., Tissue distribution and quantification of the expression of mRNAs of peroxisome proliferator-activated receptors and liver X receptor-alpha in humans: no alteration in adipose tissue of obese and NIDDM patients. Diabetes, 1997. 46(8): p. 1319-27. 8. Thorp, J.M. and W.S. Waring, Modification of metabolism and distribution of lipids by ethyl chlorophenoxyisobutyrate. Nature, 1962. 194: p. 948-9. 9. Reddy, J.K., D.L. Azarnoff, and C.E. Hignite, Hypolipidaemic hepatic peroxisome proliferators form a novel class of chemical carcinogens. Nature, 1980. 283(5745): p. 397-8. 10. Lee, S.S., et al., Targeted disruption of the alpha isoform of the peroxisome proliferatoractivated receptor gene in mice results in abolishment of the pleiotropic effects of peroxisome proliferators. Mol Cell Biol, 1995. 15(6): p. 3012-22. 11. Peters, J.M., R.C. Cattley, and F.J. Gonzalez, Role of PPAR alpha in the mechanism of action of the nongenotoxic carcinogen and peroxisome proliferator Wy-14,643. Carcinogenesis, 1997. 18(11): p. 2029-33. 12. IARC, Peroxisome Proliferation and Its Role in Carcinogenesis. IARC Technical Report, 1995(No. 24.). 13. Ashby, J., et al., Mechanistically-based human hazard assessment of peroxisome proliferator-induced hepatocarcinogenesis. Hum Exp Toxicol, 1994. 13 Suppl 2: p. S1117. 165 14. Cattley, R.C., et al., Do peroxisome proliferating compounds pose a hepatocarcinogenic hazard to humans? Regul Toxicol Pharmacol, 1998. 27(1 Pt 2): p. 47-60. 15. Cheung, C., et al., Diminished hepatocellular proliferation in mice humanized for the nuclear receptor peroxisome proliferator-activated receptor alpha. Cancer Res, 2004. 64(11): p. 3849-54. 16. Morimura, K., et al., Differential susceptibility of mice humanized for peroxisome proliferator-activated receptor alpha to Wy-14,643-induced liver tumorigenesis. Carcinogenesis, 2006. 27(5): p. 1074-80. 17. Tugwood, J.D., et al., Peroxisome proliferator-activated receptors: structures and function. Ann N Y Acad Sci, 1996. 804: p. 252-65. 18. Palmer, C.N., et al., Peroxisome proliferator activated receptor-alpha expression in human liver. Mol Pharmacol, 1998. 53(1): p. 14-22. 19. Yu, S. and J.K. Reddy, Transcription coactivators for peroxisome proliferator-activated receptors. Biochim Biophys Acta, 2007. 1771(8): p. 936-51. 20. Ammerschlaeger, M., et al., Characterization of the species-specificity of peroxisome proliferators in rat and human hepatocytes. Toxicol Sci, 2004. 78(2): p. 229-40. 21. Yang, Q., et al., The PPAR alpha-humanized mouse: a model to investigate species differences in liver toxicity mediated by PPAR alpha. Toxicol Sci, 2008. 101(1): p. 132-9. 22. Rakhshandehroo, M., et al., Comparative analysis of gene regulation by the transcription factor PPARalpha between mouse and human. PLoS One, 2009. 4(8): p. e6796. 23. Patrawala, L., et al., Side population is enriched in tumorigenic, stem-like cancer cells, whereas ABCG2+ and ABCG2- cancer cells are similarly tumorigenic. Cancer Res, 2005. 65(14): p. 6207-19. 24. Visvader, J.E. and G.J. Lindeman, Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nat Rev Cancer, 2008. 8(10): p. 755-68. 25. Alison, M.R., S. Islam, and S. Lim, Stem cells in liver regeneration, fibrosis and cancer: the good, the bad and the ugly. J Pathol, 2009. 217(2): p. 282-98. 26. Crisafulli, C., et al., PPAR-alpha contributes to the anti-inflammatory activity of 17betaestradiol. J Pharmacol Exp Ther, 2009. 331(3): p. 796-807. 27. Boverhof, D.R., et al., Temporal and dose-dependent hepatic gene expression patterns in mice provide new insights into TCDD-Mediated hepatotoxicity. Toxicol Sci, 2005. 85(2): p. 1048-63. 166 28. Kopec, A.K., et al., Comparative toxicogenomic examination of the hepatic effects of PCB126 and TCDD in immature, ovariectomized C57BL/6 mice. Toxicol Sci, 2008. 102(1): p. 61-75. 29. Kopec, A.K., et al., PCB153-elicited hepatic responses in the immature, ovariectomized C57BL/6 mice: comparative toxicogenomic effects of dioxin and non-dioxin-like ligands. Toxicol Appl Pharmacol, 2010. 243(3): p. 359-71. 30. Kopec, A.K., et al., Automated dose-response analysis and comparative toxicogenomic evaluation of the hepatic effects elicited by TCDD, TCDF, and PCB126 in C57BL/6 mice. Toxicol Sci, 2010. 118(1): p. 286-97. 31. Kim, S., et al., Comparative analysis of AhR-mediated TCDD-elicited gene expression in human liver adult stem cells. Toxicol Sci, 2009. 112(1): p. 229-44. 32. Eckel, J.E., et al., Empirical bayes gene screening tool for time-course or dose-response microarray data. J Biopharm Stat, 2004. 14(3): p. 647-70. 33. Burgoon, L.D. and T.R. Zacharewski, dbZach toxicogenomic information management system. Pharmacogenomics, 2007. 8(3): p. 287-91. 34. Saeed, A.I., et al., TM4: a free, open-source system for microarray data management and analysis. Biotechniques, 2003. 34(2): p. 374-8. 35. Dennis, G., Jr., et al., DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol, 2003. 4(5): p. P3. 36. Rozen, S. and H. Skaletsky, Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol, 2000. 132: p. 365-86. 37. Vandesompele, J., et al., Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol, 2002. 3(7): p. RESEARCH0034. 38. Reddy, J.K., et al., Mitogenic and carcinogenic effects of a hypolipidemic peroxisome proliferator, [4-chloro-6-(2,3-xylidino)-2-pyrimidinylthio]acetic acid (Wy-14, 643), in rat and mouse liver. Cancer Res, 1979. 39(1): p. 152-61. 39. Suzuki, T., et al., Evaluation of in vivo liver genotoxic potential of Wy-14,643 and piperonyl butoxide in rats subjected to two-week repeated oral administration. Arch Toxicol, 2010. 84(6): p. 493-500. 40. Cayen, M.N., Disposition, metabolism and pharmacokinetics of antihyperlipidemic agents in laboratory animals and man. Pharmacol Ther, 1985. 29(2): p. 157-204. 41. NTP, NTP Technical Report on the Toxicity Studies of Wy-14,643 (CAS No. 50892-23-4) Administered in Feed to Sprague-Dawley Rats, B6C3F1 Mice, and Syrian Hamsters. Toxicity Report Series, 2007. no. 62(NIH Publication no. 08-4419). 167 42. Willson, T.M., et al., The PPARs: from orphan receptors to drug discovery. J Med Chem, 2000. 43(4): p. 527-50. 43. Klaunig, J.E., et al., PPARalpha agonist-induced rodent tumors: modes of action and human relevance. Crit Rev Toxicol, 2003. 33(6): p. 655-780. 44. Blanquart, C., et al., Peroxisome proliferator-activated receptor alpha (PPARalpha ) turnover by the ubiquitin-proteasome system controls the ligand-induced expression level of its target genes. J Biol Chem, 2002. 277(40): p. 37254-9. 45. Ren, H., et al., Regulation of Proteome Maintenance Gene Expression by Activators of Peroxisome Proliferator-Activated Receptor alpha. PPAR Res, 2010. 2010: p. 727194. 46. Anderson, S.P., et al., The transcriptional response to a peroxisome proliferatoractivated receptor alpha agonist includes increased expression of proteome maintenance genes. J Biol Chem, 2004. 279(50): p. 52390-8. 47. Zaiss, D.M., et al., PI31 is a modulator of proteasome formation and antigen processing. Proc Natl Acad Sci U S A, 2002. 99(22): p. 14344-9. 48. Qu, A., et al., PPARalpha-dependent activation of cell cycle control and DNA repair genes in hepatic nonparenchymal cells. Toxicol Sci, 2010. 118(2): p. 404-10. 49. Cariello, N.F., et al., Gene expression profiling of the PPAR-alpha agonist ciprofibrate in the cynomolgus monkey liver. Toxicol Sci, 2005. 88(1): p. 250-64. 50. Kramer, J.A., et al., Transcription profiling distinguishes dose-dependent effects in the livers of rats treated with clofibrate. Toxicol Pathol, 2003. 31(4): p. 417-31. 51. Ramanan, S., et al., PPARalpha ligands inhibit radiation-induced microglial inflammatory responses by negatively regulating NF-kappaB and AP-1 pathways. Free Radic Biol Med, 2008. 45(12): p. 1695-704. 52. Delerive, P., et al., Peroxisome proliferator-activated receptor alpha negatively regulates the vascular inflammatory gene response by negative cross-talk with transcription factors NF-kappaB and AP-1. J Biol Chem, 1999. 274(45): p. 32048-54. 53. Delerive, P., et al., Induction of IkappaBalpha expression as a mechanism contributing to the anti-inflammatory activities of peroxisome proliferator-activated receptor-alpha activators. J Biol Chem, 2000. 275(47): p. 36703-7. 54. Stienstra, R., et al., The Interleukin-1 receptor antagonist is a direct target gene of PPARalpha in liver. J Hepatol, 2007. 46(5): p. 869-77. 55. Devchand, P.R., et al., The PPARalpha-leukotriene B4 pathway to inflammation control. Nature, 1996. 384(6604): p. 39-43. 168 56. Ip, E., et al., Administration of the potent PPARalpha agonist, Wy-14,643, reverses nutritional fibrosis and steatohepatitis in mice. Hepatology, 2004. 39(5): p. 1286-96. 57. Madej, A., et al., Effects of fenofibrate on plasma cytokine concentrations in patients with atherosclerosis and hyperlipoproteinemia IIb. Int J Clin Pharmacol Ther, 1998. 36(6): p. 345-9. 58. Fernandez-Miranda, C., et al., A pilot trial of fenofibrate for the treatment of nonalcoholic fatty liver disease. Dig Liver Dis, 2008. 40(3): p. 200-5. 59. Lawrence, J.W., et al., Differential gene regulation in human versus rodent hepatocytes by peroxisome proliferator-activated receptor (PPAR) alpha. PPAR alpha fails to induce peroxisome proliferation-associated genes in human cells independently of the level of receptor expresson. J Biol Chem, 2001. 276(34): p. 31521-7. 60. Gonzalez, F.J. and Y.M. Shah, PPARalpha: Mechanism of species differences and hepatocarcinogenesis of peroxisome proliferators. Toxicology, 2007. 61. Walgren, J.E., D.T. Kurtz, and J.M. McMillan, Expression of PPAR(alpha) in human hepatocytes and activation by trichloroacetate and dichloroacetate. Res Commun Mol Pathol Pharmacol, 2000. 108(1-2): p. 116-32. 62. Gervois, P., et al., A truncated human peroxisome proliferator-activated receptor alpha splice variant with dominant negative activity. Mol Endocrinol, 1999. 13(9): p. 1535-49. 63. Varanasi, U., et al., Identification of a peroxisome proliferator-responsive element upstream of the human peroxisomal fatty acyl coenzyme A oxidase gene. J Biol Chem, 1996. 271(4): p. 2147-55. 64. Jia, Y., et al., Transcription coactivator PBP, the peroxisome proliferator-activated receptor (PPAR)-binding protein, is required for PPARalpha-regulated gene expression in liver. J Biol Chem, 2004. 279(23): p. 24427-34. 65. Matsumoto, K., et al., Critical role for transcription coactivator peroxisome proliferatoractivated receptor (PPAR)-binding protein/TRAP220 in liver regeneration and PPARalpha ligand-induced liver tumor development. J Biol Chem, 2007. 282(23): p. 17053-60. 66. Guillouzo, A., Liver cell models in in vitro toxicology. Environ Health Perspect, 1998. 106 Suppl 2: p. 511-32. 67. Kubota, T., et al., Fenofibrate induces apoptotic injury in cultured human hepatocytes by inhibiting phosphorylation of Akt. Apoptosis, 2005. 10(2): p. 349-58. 68. Martinasso, G., et al., Involvement of PPARs in Cell Proliferation and Apoptosis in Human Colon Cancer Specimens and in Normal and Cancer Cell Lines. PPAR Res, 2007. 2007: p. 93416. 169 69. Puthalakath, H., et al., Bmf: a proapoptotic BH3-only protein regulated by interaction with the myosin V actin motor complex, activated by anoikis. Science, 2001. 293(5536): p. 1829-32. 170 CHAPTER 6 171 CHAPTER 6 CONCLUSIONS AND FUTURE RESEARCH The preceding studies have established and evaluated human liver stem cells as a novel human in vitro model for toxicogenomic studies. Human liver HL1-1 stem cells were shown to express functional AhR and PPAR, and their gene expression profiles were identified using a toxicogenomic approach. Responses of HL1-1 cells were compared with human hepatoma cells HepG2 and mouse in vivo hepatic expression to elucidate molecular mechanisms of receptormediated toxicity. Receptor activation elicited complex temporal and species-conserved and specific gene expression responses, which could be related to physiological and toxicological outcomes of exposure. Many cross-species, cross-model conserved response genes were consistent with previously reported receptor-mediated physiological outcomes such as tumor development and alterations in lipid metabolism. Although some common pathways are affected, the data suggest that species-specific receptor-mediated gene expression profiles imply speciesspecific regulons and provide insight into key events that may contribute to the species-specific effects which warrant further investigation. However, new research opportunities have also been identified that warrant further exploration in order to further elucidate species-specific receptormediated mechanisms of toxicity as outlined below. COMPARATIVE STUDIES FOR FURTHER EVALUATION Data presented in Chapter 4 and 5 provide baseline quantitative response data on the human liver stem cells and mouse liver following treatment with TCDD (AhR ligand) and Wy14,643 (PPAR ligand), which can be used in future comparative studies. 172 Functional categorization of response genes a significant overlap between lipid metabolism related genes regulated by AhR and PPAR. These data suggest a possible AhR/PPAR cross-talk in lipid metabolism modulation which is in agreement with previously published interactions [1, 2]. Cotreatment studies using AhR and PPAR inducers could further elucidate potential AhR/PPAR cross-talk mechanisms involved in lipid metabolism modulation. Freshly isolated primary hepatocytes from intact liver tissues are often referred as the gold in vitro model standard for evaluating hepatic metabolism and toxicity testing [3]. TCDD toxicogenomic studies utilizing primary hepatocytes from human, mouse and rat were recently conducted in this lab (Forgacs et al., in preparation). To further assess human liver stem cells as an in vitro model, the results from human primary hepatocytes should be compared to human liver stem cells from Chapter 4, which will expand the understanding of AhR-mediated molecular level responses in human. Additionally, the same approach could be applied to PPAR-mediated gene expression profiles using comparative analysis with recently published human primary hepatocytes data [4]. MODEL SYSTEM IMPROVEMENTS AND FURTHER APPLICATION OF HUMAN LIVER STEM CELLS The development of immortalized HLhT1 cell line and their AhR- and PPAR-mediated gene expression responses comparison with parental HL1-1 cell was described in Chapter 3. Basal expression level of receptors and their responses are comparable to parental cell, as confirmed using a subset of responsive genes (CYP1A1, CYP1B1, SLC7A5 and ALDH1A3). Immortalized human liver stem cells are a very promising alternative model, which is amenable to high-throughput and mechanistic studies. However, further characterization including wholegenome gene expression profiling is required. 173 In Chapter 3, the development of hepatocyte differentiation methods was suggested which may re-establish the expression of PXR and other hepatic metabolizing enzymes. Liver stem cells give rise to both hepatocytes and bile duct epithelial cells also known as cholangiocytes. The growth and differentiation of hepatoblasts is regulated by various extrinsic signals and complex regulatory mechanisms [5, 6]. However, hepatocyte differentiation methods development from ES (embryonic stem) cells, MSC (mesenchymal stem cells) and even IPS (induced pluripotent stem) cells have been attempted by many groups [7, 8]. A major limitation of differentiated stem cell application in standardized high-throughput studies is the prominent heterogeneity of cell types. However, transgenic molecular enrichment and selection, and overexpression of transcription factors facilitate enhancement of the purity of the targeted differentiated cells [9]. Two dimensional in vitro cultures on plastic surfaces and growing in log phase in high oxygen tension, unlike in vivo situation, may cause non-relevant responses. To develop cell culture conditions closely mimicking the in vivo conditions, application of structural material resembling the complex extracellular environment such as Matrigel and three dimensional organoids co-cultured with their normal stromal cells providing specific organ-requiring microenvironmental factors has been proposed [10]. Additionally, clinical application of liver stem cells to regeneration of the liver parenchyma as a therapeutic option of end-stage liver disease (cirrhosis) is being investigated as an alternative approach to liver repair and regeneration [11]. ALTERNATIVE TECHNOLOGIES FOR TOXICOGENOMICS Microarrays are limited in terms of probe coverage due to integrity and completion of sequence information and gene annotation. Additionally, the data from microarrays could be affected by probe affinity to target genes and non-specific hybridization. Recently, genome-wide 174 sequencing has been enabled in biomedical research with the development of modern NGS (next generation sequencing). ChIP (Chromatin immunoprecipitation)-sequencing and RNA- sequencing analysis with NGS can be utilized for motif finding, assessment of differential gene expression, novel target gene discovery, splice isoform expression, and SNP (single-nucleotide polymorphism) annotations with their functional relation [12]. Application of NGS should be considered in future studies to provide more comprehensive analyses. Because of the limited intra-laboratory resources, -omics data from public repositories such as GEO (www.ncbi.nlm.nih.gov/geo), ArrayExpress (www.ebi.ac.uk/arrayexpress), CEBS (http://cebs.niehs.nih.gov), and DrugMatrix (https://ntp.niehs.nih.gov/drugmatrix) could also be used for comparative analyses. The growth of publicly available -omics data sets will increasingly drive integrated computational meta-analysis. Meta-analysis allows merging of several -omics data sets such as genomics, proteomics and metabolomics into knowledge through cross-examination of data and network reconstruction. Additional integrative analyses with RNA interference perturbations will provide more definitive evidence of functionally important connections and relationships within regulatory pathways. Systems integration of different experimental data into a consistent meta-analysis will guide more closely to the true biology involved in toxicity. 175 REFERENCES 176 REFERENCES 1. Villard, P.H., et al., PPARalpha transcriptionally induces AhR expression in Caco-2, but represses AhR pro-inflammatory effects. Biochem Biophys Res Commun, 2007. 364(4): p. 896-901. 2. Shaban, Z., et al., AhR and PPARalpha: antagonistic effects on CYP2B and CYP3A, and additive inhibitory effects on CYP2C11. Xenobiotica, 2005. 35(1): p. 51-68. 3. Madan, A., et al., Effects of prototypical microsomal enzyme inducers on cytochrome P450 expression in cultured human hepatocytes. Drug Metab Dispos, 2003. 31(4): p. 421-31. 4. Rakhshandehroo, M., et al., Comparative analysis of gene regulation by the transcription factor PPARalpha between mouse and human. PLoS One, 2009. 4(8): p. e6796. 5. Tanaka, M., et al., Liver stem/progenitor cells: their characteristics and regulatory mechanisms. J Biochem, 2011. 149(3): p. 231-9. 6. Gaudio, E., et al., New insights into liver stem cells. Dig Liver Dis, 2009. 41(7): p. 45562. 7. Ochiya, T., Y. Yamamoto, and A. Banas, Commitment of stem cells into functional hepatocytes. Differentiation, 2009. 8. Gallicano, G.I. and L. Mishra, Hepatocytes from induced pluripotent stem cells: a giant leap forward for hepatology. Hepatology, 2010. 51(1): p. 20-2. 9. Ahuja, Y.R., V. Vijayalakshmi, and K. Polasa, Stem cell test: a practical tool in toxicogenomics. Toxicology, 2007. 231(1): p. 1-10. 10. Trosko, J.E. and C.C. Chang, Factors to consider in the use of stem cells for pharmaceutic drug development and for chemical safety assessment. Toxicology, 2010. 270(1): p. 18-34. 11. Zhao, Q., et al., Stem/progenitor cells in liver injury repair and regeneration. Biol Cell, 2009. 101(10): p. 557-71. 12. Werner, T., Next generation sequencing in functional genomics. Brief Bioinform, 2010. 11(5): p. 499-511. 177