Clustering of hepatotoxins based on mechanism of toxicity using gene expression profiles

Toxicogenomics in drug discovery: from preclinical studies to clinical trials

Gene expression analysis applied to toxicology studies, also referred to as toxicogenomics, is rapidly being embraced by the pharmaceutical industry as a useful tool to identify safer drugs in a quicker, more cost-effective manner. Studies have already demonstrated the benefits of applying gene expression profiling towards drug safety evaluation, both for identifying mechanisms underlying toxicity, as well as for providing a means to identify safety liabilities early in the drug discovery process. Furthermore, toxicogenomics has the potential to better identify and assess adverse drug reactions of new drug candidates or marketed products in humans. While much still remains to be learned about the relevance and the application of gene expression changes in human toxicology, the next few years should see gene expression technologies applied to more stages and more programs of the drug discovery and development process. This review will focus on how toxicogenomics can or has been applied in drug discovery and development, and will discuss some of the challenges that still remain.

On the utility of pooling biological samples in microarray experiments

Over 15% of the data sets catalogued in the Gene Expression Omnibus Database involve RNA samples that have been pooled before hybridization. Pooling affects data quality and inference, but the exact effects are not yet known because pooling has not been systematically studied in the context of microarray experiments. Here we report on the results of an experiment designed to evaluate the utility of pooling and the impact on identifying differentially expressed genes. We find that inference for most genes is not adversely affected by pooling, and we recommend that pooling be done when fewer than three arrays are used in each condition. For larger designs, pooling does not significantly improve inferences if few subjects are pooled. The realized benefits in this case do not outweigh the price paid for loss of individual specific information. Pooling is beneficial when many subjects are pooled, provided that independent samples contribute to multiple pools.

Using in vitro models for expression profiling studies on ethanol and drugs of abuse

The use of expression profiling with microarrays offers great potential for studying the mechanisms of action of drugs of abuse. Studies with the intact nervous system seem likely to be most relevant to understanding the mechanisms of drug abuse-related behaviours. However, the use of expression profiling with in vitro culture models offers significant advantages for identifying details of cellular signalling actions and toxicity for drugs of abuse. This study discusses general issues of the use of microarrays and cell culture models for studies on drugs of abuse. Specific results from existing studies are also discussed, providing clear examples of relevance for in vitro studies on ethanol, nicotine, opiates, cannabinoids and hallucinogens such as LSD. In addition to providing details on signalling mechanisms relevant to the neurobiology of drugs of abuse, microarray studies on a variety of cell culture systems have also provided important information on mechanisms of cellular/organ toxicity with drugs of abuse. Efforts to integrate genomic studies on drugs of abuse with both in vivo and in vitro models offer the potential for novel mechanistic rigor and physiological relevance.

Microarray analysis in human hepatocytes suggests a mechanism for hepatotoxicity induced by trovafloxacin

Idiosyncratic drug toxicity, defined as toxicity that is dose independent, host dependent, and usually cannot be predicted during preclinical or early phases of clinical trials, is a particularly confounding complication of drug development. An understanding of the mechanisms that lead to idiosyncratic liver toxicity would be extremely beneficial for the development of new compounds. We used microarray analysis on isolated human hepatocytes to understand the mechanisms underlying the idiosyncratic toxicity induced by trovafloxacin. Our results clearly distinguish trovafloxacin from other marketed quinolone agents and identify unique gene changes induced by trovafloxacin that are involved in mitochondrial damage, RNA processing, transcription, and inflammation that may suggest a mechanism for the hepatotoxicity induced by this agent. In conclusion, this work establishes the basis for future microarray analysis of new compounds to determine the presence of these expression changes and their usefulness in predicting idiosyncratic hepatotoxicity. Supplementary material for this article can be found on the HEPATOLOGY website (http://interscience. Wiley.com/jpages/0270-9139/suppmat/index.htnd).

Comprehensive analysis of differential gene expression profiles on carbon tetrachloride-induced rat liver injury and regeneration

Microarray analysis of RNA from carbon tetrachloride (CCl4)-administered rat livers was performed at various time points to establish a global gene expression profile during injury and regeneration stages. A single dose of 1 ml/kg of CCl4 was given by ip injection, and the liver samples were obtained after 6, 24, 48 h, and 2 weeks. Histopathologic, biochemical, and immunohistochemical studies enabled the classification of the CCl4 effect into injury (6 and 24 h) and regeneration (48 h and 2 weeks) stages. The expression levels of 5180 clones on a custom rat gene microarray were analyzed and 587 clones yielded changeable gene expression on at least single time point. One hundred seventy-nine clones were classified as injury-specific clones, while 38 clones as regeneration-specific clones. Characteristic gene expression profiles could be associated with CCl4-induced gene expression with the disruption of lipid metabolism, which is known to cause the fatty liver induced by CCl4 treatment. In addition, induction of the transcripts for many ribosomal proteins was detected during the injury stage, particularly at the 24-h time point, despite the previous report of decreased protein synthesis rate upon CCl4 treatment. Several genes with known functions were also identified as CCl4-regulated genes. In conclusion, we established a global gene expression profile utilizing microarray analysis in rat liver upon acute CCl4 administration with a full chronological profile that not only covers injury stage but also later points of regeneration stage.

Toxicogenomics and systems toxicology: aims and prospects

Toxicogenomics combines transcript, protein and metabolite profiling with conventional toxicology to investigate the interaction between genes and environmental stress in disease causation. The patterns of altered molecular expression that are caused by specific exposures or disease outcomes have revealed how several toxicants act and cause disease. Despite these success stories, the field faces noteworthy challenges in discriminating the molecular basis of toxicity. We argue that toxicology is gradually evolving into a systems toxicology that will eventually allow us to describe all the toxicological interactions that occur within a living system under stress and use our knowledge of toxicogenomic responses in one species to predict the modes-of-action of similar agents in other species.

Testing paradigm for prediction of development-limiting barriers and human drug toxicity

The financial investment grows exponentially as a new chemical entity advances through each stage of discovery and development. The opportunity exists for the modern toxicologist to significantly impact expenditures by the early prediction of potential toxicity/side effect barriers to development by aggressive evaluation of development-limiting liabilities early in drug discovery. Improved efficiency in pharmaceutical research and development lies both in leveraging "best in class" technology and integration with pharmacologic activities during hit-to-lead and early lead optimization stages. To meet this challenge, a discovery assay by stage (DABS) paradigm should be adopted. The DABS clearly delineates to discovery project teams the timing and type of assay required for advancement of compounds to each subsequent level of discovery and development. An integrative core pathology function unifying Drug Safety Evaluation, Molecular Technologies and Clinical Research groups that effectively spans all phases of drug discovery and development is encouraged to drive the DABS. The ultimate goal of such improved efficiency being the accurate prediction of toxicity and side effects that would occur in development before commitment of the large prerequisite resource. Good justification of this approach is that every reduction of development attrition by 10% results in an estimated increase in net present value by $100 million.

Chemical mixture toxicology: from descriptive to mechanistic, and going on to in silico toxicology

Because of the pioneering vision of certain leaders in the biomedical field, the last two decades witnessed rapid advances in the area of chemical mixture toxicology. Earlier studies utilized conventional toxicology protocol and methods, and they were mainly descriptive in nature. Two good examples might be the parallel series of studies conducted by the U.S. National Toxicology Program and TNO in The Netherlands, respectively. As a natural course of progression, more and more sophistication was incorporated into the toxicology studies of chemical mixtures. Thus, at least the following seven areas of scientific achievements in chemical mixture toxicology are evident in the literature: (a) the application of better and more robust statistical methods; (b) the exploration and incorporation of mechanistic bases for toxicological interactions; (c) the application of physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) modeling; (d) the studies on more complex chemical mixtures; (e) the use of science-based risk assessment approaches; (f) the utilization of functional genomics; and (g) the application of technology. Examples are given for the discussion of each of these areas. Two important concepts emerged from these studies and they are: (1) dose-dependent toxicologic interactions; and (2) "interaction thresholds". Looking into the future, one of the most challenging areas in chemical mixture research is finding the answer to the question "when one tries to characterize the health effects of chemical mixtures, how does one deal with the infinite number of combination of chemicals, and other possible stressors?" Undoubtedly, there will be many answers from different groups of researchers. Our answer, however, is first to focus on the finite (biological processes) rather than the infinite (combinations of chemical mixtures and multiple stressors). The idea is that once we know a normal biological process(es), all stimuli and insults from external stressors are merely perturbations of the normal biological process(es). The next step is to "capture" the biological process(es) by integrating the recent advances in computational technology and modern biology. Here, the computer-assisted Reaction Network Modeling, linked with PBPK modeling, offers a ray of hope to dealing with the complex biological systems.

Morphological and microarray analysis of T-2 toxin-induced rat fetal brain lesion

To examine morphological and gene expression changes induced by T-2 toxin in the fetal brain in detail, pregnant rats on day 13 of gestation were treated orally with a single dose of T-2 toxin (2 mg/kg) and sacrificed at 1, 3, 6, 9, 12 and 24 h after treatment (HAT). Histopathologically, the number of apoptotic neuroepithelial cells in the telencephalon increased from 1 HAT and peaked at 12 HAT. Based on the histopathological examinations, microarray analysis was performed at 6, 12 and 24 HAT. Microarray analysis showed that the expression of oxidative stress-related genes (heat shock protein 70 (HSP70) and heme oxygenase (HO)) was strongly induced by T-2 toxin at 12 HAT, the peak time point of apoptosis induction. The expression of mitogen-activated protein kinase (MAPK)-related genes (MEKK1 and c-jun) and other apoptosis-related genes (caspase-2 and insulin-like growth factor-binding protein-3 (IGF-BP3)) was also induced by the T-2 toxin treatment. The changes observed by microarray analysis were confirmed for four up-regulated genes (HSP70, HO, IGF-BP3 and VEGF-A) using real-time RT-PCR. Our results suggest that the T-2 toxin-induced apoptosis in the fetal brain is due to oxidative stress, and that the MAPK pathway may be involved in T-2 toxin-induced toxicity.

Gene expression profiles in pregnant rats treated with T-2 toxin

Pregnant rats on day 13 of gestation were treated orally with T-2 toxin at a single dose of 2 mg/kg and sacrificed at 24 hours after treatment. Histopathologically, apoptosis was increased in the liver, placenta and fetal liver. Microarray analysis was performed to examine the gene expression in the liver, placenta, and fetal liver. The results of microarray analysis showed that the changes in the expression of apoptosis genes, metabolic enzymes and oxidative stress-related genes were detected in these tissues. Suppression of phase I and II enzymes-related genes expression in the liver, and suppression of phase II enzymes-related genes expression in the placenta and fetal liver were observed. Semiquantitive RT-PCR analysis also showed the same results as those of microarray analysis. From the results of microarray analysis and histopathological examination, T-2 toxin seems to induce oxidative stress in these tissues, following the changes in metabolism-related genes expression. These changes may alter the intracellular environments resulting in the induction of apoptosis. Further studies on the gene expression profiles at the earlier time point are necessary to clarify the detailed mechanisms of T-2 toxin-induced toxicity in pregnant rats.

Multiplex mRNA assay using electrophoretic tags for high-throughput gene expression analysis

We describe a novel multiplexing technology using a library of small fluorescent molecules, termed eTag molecules, to code and quantify mRNA targets. eTag molecules, which have the same fluorometric property, but distinct charge-to-mass ratios possess pre-defined electrophoretic characteristics and can be resolved using capillary electrophoresis. Coupled with primary Invader mRNA assay, eTag molecules were applied to simultaneously quantify up to 44 mRNA targets. This multiplexing approach was validated by examining a panel of inflammation responsive genes in human umbilical vein endothelial cells stimulated with inflammatory cytokine interleukin 1beta. The laser-induced fluorescence detection and electrokinetic sample injection process in capillary electrophoresis allows sensitive quantification of thousands of copies of mRNA molecules in a reaction. The assay is precise, as evaluated by measuring qualified Z' factor, a dimensionless and simple characteristic for applications in high-throughput screening using mRNA assays. Our data demonstrate the synergy between the multiplexing capability of eTag molecules by sensitive capillary electrophoresis detection and the isothermal linear amplification characteristics of the Invader assay. eTag multiplex mRNA assay presents a unique platform for sensitive, high sample throughput and multiplex gene expression analysis.

Interaction networks: coordinating responses to xenobiotic exposure

In the last decade the increased usage of '-omic' technologies, plus the sequencing of over 800 complete genomes has led to a vast increase in the amount of information available to the researcher for examining cellular responses to xenobiotics. Much effort has been put into the identification and analysis of expression profiles associated with pathobiological conditions and/or xenobiotic exposure. These profiles are commonly used in two applications. Firstly, comparative profile experiments are used to classify pathobiological states and for the screening of novel chemical entities to predict their action(s) on the body. Secondly, mechanistic investigations will gain information on the molecular mechanisms underlying toxic responses/pathobiological states. During the course of such analysis it has become increasingly clear that a series of highly refined interaction networks exist within the body, regulating both the sensitivity and selectivity of the body's response to pathobiological states/xenobiotic exposure. These interaction networks exist at several levels: Firstly, within individual cells, the interaction between factors that transmit xenobiotics signals will determine the overall cellular response. Secondly, intraorgan communication occurs between the different cell types/sub-types which makes up an organ, coordinating the overall organ response. Finally, interorgan interactions provide axes of response through the body.

Accelerating drug discovery

Although the evolution of '-omics' methodologies is still in its infancy, both the pharmaceutical industry and patients could benefit from their implementation in the drug development process

Discriminating different classes of toxicants by transcript profiling

Male rats were treated with various model compounds or the appropriate vehicle controls. Most substances were either well-known hepatotoxicants or showed hepatotoxicity during preclinical testing. The aim of the present study was to determine if biological samples from rats treated with various compounds can be classified based on gene expression profiles. In addition to gene expression analysis using microarrays, a complete serum chemistry profile and liver and kidney histopathology were performed. We analyzed hepatic gene expression profiles using a supervised learning method (support vector machines; SVMs) to generate classification rules and combined this with recursive feature elimination to improve classification performance and to identify a compact subset of probe sets with potential use as biomarkers. Two different SVM algorithms were tested, and the models obtained were validated with a compound-based external cross-validation approach. Our predictive models were able to discriminate between hepatotoxic and nonhepatotoxic compounds. Furthermore, they predicted the correct class of hepatotoxicant in most cases. We provide an example showing that a predictive model built on transcript profiles from one rat strain can successfully classify profiles from another rat strain. In addition, we demonstrate that the predictive models identify nonresponders and are able to discriminate between gene changes related to pharmacology and toxicity. This work confirms the hypothesis that compound classification based on gene expression data is feasible.

Inverse gene expression patterns for macrophage activating hepatotoxicants and peroxisome proliferators in rat liver

Macrophage activation contributes to adverse effects produced by a number of hepatotoxic compounds. Transcriptional profiles elicited by two macrophage activators, LPS and zymosan A, were compared to those produced by 100 paradigm compounds (mostly hepatotoxicants) using cDNA microarrays. Several hepatotoxicants previously reported to activate liver macrophages produced transcriptional responses similar to LPS and zymosan, and these were used to construct a gene signature profile for macrophage activators in the liver. Measurement of cytokine mRNAs in the same liver samples by RT-PCR independently confirmed that these compounds are associated with macrophage activation. In addition to expected effects on acute phase proteins and metabolic pathways that are regulated by LPS and inflammation, a strong induction was observed for many endoplasmic reticulum-associated stress/chaperone proteins. Additionally, many genes in our macrophage activator signature profile were well-characterized PPARalpha-induced genes which were repressed by macrophage activators. A shared gene signature profile for peroxisome proliferators was determined using a training set of clofibrate, WY 14643, diethylhexylphthalate, diisononylphthalate, perfluorodecanoic acid, perfluoroheptanoic acid, and perfluorooctanoic acid. The signature profile included macrophage activator-induced genes that were repressed by peroxisome proliferators. NSAIDs comprised an interesting pharmacological class in that some compounds, notably diflunisal, co-clustered with peroxisome proliferators whereas several others co-clustered with macrophage activators, possibly due to endotoxin exposure secondary to their adverse effects on the gastrointestinal system. While much of these data confirmed findings from the literature, the transcriptional patterns detected using this toxicogenomics approach showed relationships between genes and biological pathways requiring complex analysis to be discerned.

Environmental risk assessment of medicinal products for human use according to European Commission recommendations

Presented here, based on new recommendations of the European Commission, is an environmental risk assessment (ERA) of a selected group of pharmaceuticals for Phase I, environmental exposure assessment, and Phase II Tier A, initial environmental fate and effect analysis. This pharmaceutical group is composed of the 111 highest-selling human drug substances that have annual sales in Germany of more than 5,000 kg. The data required for this ERA came from analyzing: (1) sales annually (in kg or IU) of the 2671 active pharmaceutical drug substances (2001) on the German market in all medicinal products sold by pharmacies (with and without prescriptions) and used in hospitals in 1996-2001; (2) the use pattern of drug substances as categorized according to Anatomical Therapeutic Chemical (ATC) classification indexes ATC3 and ATC7; (3) data for excretion, toxicity, and metabolites of the 111 selected human drug substances; (4) the physicochemical properties of these substances; and (5) the degradability of selected drug substances in sewage treatment plants (STPs) by using a validated and accredited liquid chromatography-electrospray ionization tandem mass spectrometry method. A correction factor for the pharmaceutical therapeutic (PT) activity of metabolites, the PT(Index) (excretion rate/100) for drug substances and PT active metabolites was established to refine the predicted environmental concentration (PEC(SURFACEWATER)). A refinement of the PEC(SURFACEWATER) was carried out with the market penetration factor of the human drug substances in Germany. In addition, for effect analysis the predicted no-effects concentration (PNEC) was calculated using assessment factors. The estimated PEC results were validated with the exposure results of effluents of the STPs. All results on ERA of drug substances have been documented in a Microsoft Access 2000 database.

Activators of the rat pregnane X receptor differentially modulate hepatic and intestinal gene expression

Ligand-mediated activation of the pregnane X receptor (PXR, NR1I2) is postulated to affect both hepatic and intestinal gene expression, because of the presence of this nuclear receptor in these important drug metabolizing organs; as such, activation of this receptor may elicit the coordinated regulation of PXR target genes in both tissues. Induction of hepatic and intestinal drug metabolism can contribute to the increased metabolism of drugs, and can result in adverse or undesirable drug-drug interactions. 2(S)-((3,5-bis(Trifluoromethyl)benzyl)-oxy)-3(S)phenyl-4-((3-oxo-1,2,4-triazol-5-yl)methyl)morpholine (L-742694) is a potent activator of the rat PXR and was characterized for its effects on hepatic and intestinal gene expression in female Sprague-Dawley rats by DNA microarray analysis. Transcriptional profiling in liver and small intestine revealed that L-742694 and dexamethasone (DEX) induced the prototypical battery of PXR target genes in liver, including CYP3A, Oatp2, and UGT1A1. In addition, both DEX and L-742694 induced common gene expression profiles that were specific to liver or small intestine, but there was a distinct lack of coordinated gene expression of genes common to both tissues. This pattern of gene regulation occurred in liver and small intestine independent of PXR, constitutive androstane receptor, or hepatic nuclear factor-4alpha expression, suggesting that other factors are involved in controlling the extent of coordinated gene expression in response to a PXR agonist. Overall, these results suggest that ligand-mediated activation of PXR and induction of hepatic, rather than small intestinal, drug metabolism genes would contribute to the increased metabolism of orally administered pharmaceuticals.

Effect of serum cholesterol on the mRNA content of amyloid precursor protein in rat livers

Genes that showed mRNA content profiles, which correlated with serum concentrations of total cholesterol (T.CHO), were screened from the microarray data of phenobarbital (PB)- or clofibrate (CLO)-treated rat livers, and the correlation was evaluated based on Spearman's correlation coefficient. Many genes involved in the cholesterol or bile acid metabolism were highly correlated such as UDP-glucuronosyltransferase-21, apolipoprotein A-I and cMOAT. The mRNA content of the amyloid precursor protein (APP) showed the 5th highest correlation among the 8799 probes in the Affymetrix Rat Genome U34 Array. In the livers of rats fed a high-cholesterol (1%) diet for 33 days, serum T.CHO levels increased by 4.6-fold, and the hepatic APP mRNA content also increased by 1.9-fold compared to the control group. These data suggest that the hepatic APP mRNA content was affected by serum T.CHO, and that hepatic APP was involved in cholesterol metabolism in rat livers.

Integration of clinical and gene expression endpoints to explore furan-mediated hepatotoxicity

Molecular techniques, such as cDNA microarrays, are being used to aid in the elucidation of the mechanisms of toxicity of a variety of compounds. In this study, we evaluate the molecular effects of furan in the rat liver. Sprague-Dawley rats were exposed to 4 or 40 mg/kg furan for up to 14 days. Furan induced an initial degenerative and necrotic phenotype that was followed by inflammation and fibrosis, consistent with previous observations for this compound. RNA was harvested from each lobe of the liver at several time points to observe whether lobe-specific gene expression effects occurred. Similar gene expression changes were observed in all lobes, however the magnitude of gene expression change was more pronounced in the right lobe. Finally, to help determine the correlation between gene expression changes and liver pathology, we applied traditional microarray visualization tools to the assessment of clinical chemistry and pathology parameters.

Gene expression profiling reveals multiple toxicity endpoints induced by hepatotoxicants

Microarray technology continues to gain increased acceptance in the drug development process, particularly at the stage of toxicology and safety assessment. In the current study, microarrays were used to investigate gene expression changes associated with hepatotoxicity, the most commonly reported clinical liability with pharmaceutical agents. Acetaminophen, methotrexate, methapyrilene, furan and phenytoin were used as benchmark compounds capable of inducing specific but different types of hepatotoxicity. The goal of the work was to define gene expression profiles capable of distinguishing the different subtypes of hepatotoxicity. Sprague-Dawley rats were orally dosed with acetaminophen (single dose, 4500 mg/kg for 6, 24 and 72 h), methotrexate (1mg/kg per day for 1, 7 and 14 days), methapyrilene (100mg/kg per day for 3 and 7 days), furan (40 mg/kg per day for 1, 3, 7 and 14 days) or phenytoin (300 mg/kg per day for 14 days). Hepatic gene expression was assessed using toxicology-specific gene arrays containing 684 target genes or expressed sequence tags (ESTs). Principal component analysis (PCA) of gene expression data was able to provide a clear distinction of each compound, suggesting that gene expression data can be used to discern different hepatotoxic agents and toxicity endpoints. Gene expression data were applied to the multiplicity-adjusted permutation test and significantly changed genes were categorized and correlated to hepatotoxic endpoints. Repression of enzymes involved in lipid oxidation (acyl-CoA dehydrogenase, medium chain, enoyl CoA hydratase, very long-chain acyl-CoA synthetase) were associated with microvesicular lipidosis. Likewise, subsets of genes associated with hepatotocellular necrosis, inflammation, hepatitis, bile duct hyperplasia and fibrosis have been identified. The current study illustrates that expression profiling can be used to: (1) distinguish different hepatotoxic endpoints; (2) predict the development of toxic endpoints; and (3) develop hypotheses regarding mechanisms of toxicity.

Toxicogenomics in drug development

Toxicogenomics represents the merging of toxicology with technologies that have been developed, together with bioinformatics, to identify and quantify global gene expression changes. It represents a new paradigm in drug development and risk assessment, which promises to generate a wealth of information towards an increased understanding of the molecular mechanisms that lead to drug toxicity and efficacy, and of DNA polymorphisms responsible for individual susceptibility to toxicity. Gene expression profiling, through the use of DNA microarray and proteomic technologies will aid in establishing links between expression profiles, mode of action and traditional toxic endpoints. Such patterns of gene expression, or 'molecular fingerprints' could be used as diagnostic or predictive markers of exposure, that is characteristic of a specific mechanism of induction of that toxic or efficacious effect. It is anticipated that toxicogenomics will be increasingly integrated into all phases of the drug development process particularly in mechanistic and predictive toxicology, and biomarker discovery. This review provides an overview of the expression profiling technologies applied in toxicogenomics. and discusses the promises as well as the future challenges of applying this discipline to the drug development process.

Applications of eTag™ assay platform to systems biology approaches in molecular oncology and toxicology studies

We have developed a universal eTag trade mark multiplex assay platform that can be uniquely applied to survey the molecule profiles of biologic systems in sub-global large-scale analyses. The effectiveness of eTag trade mark assays when applied to focused system biology studies in molecular oncology and predictive toxicology is herein described while reviewing the current methods commonly used. The multi-analyte and multi-parameter assay approach for parallel analysis will form the basis of an emerging paradigm of multiplexed molecular profiling for signaling pathway networks and various aspects of drug development processes.

Differentiation of human hepatoma cells during confluence as revealed by gene expression profiling

Certain human hepatocarcinoma cells undergo differentiation when grown at confluence. In order to understand the basis for this differentiation, we investigated the phenotypic changes occurring during confluent growth of the human hepatoma B16A2 cell line. The global gene expression profile of B16A2 cells grown during confluence for 5 weeks was investigated using microarrays containing complementary sequences corresponding to approximately 10,000 genes, and compared with profiles of adult human liver and HepG2 cells. The major part of gene products detected were shared by all three systems and the hepatoma cell lines expressed surprisingly high levels of liver-enriched transcription factors. During confluence of B16A2 cells, the majority of transcriptional changes monitored were directed towards the phenotype of adult human liver in vivo, although the changes accounted for less than 10% of those necessary to acquire a native hepatic phenotype. Several markers of liver differentiation and regeneration were changed in similar manner as observed in developing liver and during liver regeneration. In conclusion, the data indicate that differentiation in vitro of the B16A2 cell line during confluence partially resembles that of hepatic differentiation and regeneration in vivo, implying a partial normalization of a low differentiated phenotype.

Acute Molecular Markers of Rodent Hepatic Carcinogenesis Identified by Transcription Profiling

Currently, the only way to identify nongenotoxic hepatocarcinogens is through long-term repeat dose studies such as the 2 year rodent carcinogenicity assay. Such assays are both time consuming and expensive and require large amounts of active pharmaceutical or chemical ingredients. Thus, the results of the 2 year assay are not known until very late in the discovery and development process for new pharmaceutical entities. Although in many cases nongenotoxic carcinogenicity in rodents is considered to be irrelevant for humans, a positive finding in a 2 year carcinogenicity assay may increase the number of studies to demonstrate the lack of relevance to humans, delay final submission and subsequent registration of a product, and may result in a "black box" carcinogenicity warning on the label. To develop early identifiers of carcinogenicity, we applied transcription profiling using several prototype rodent genotoxic and nongenotoxic carcinogens, as well as two noncarcinogenic hepatotoxicants, in a 5 day repeat dose in vivo toxicology study. Fluorescent-labeled probes generated from liver mRNA prepared from male Sprague-Dawley rats treated with one of three dose levels of bemitradine, clofibrate, doxylamine, methapyrilene, phenobarbital, tamoxifen, 2-acetylaminofluorene, 4-acetylaminofluorene, or isoniazid were hybridized against rat cDNA microarrays. Correlation of the resulting data with an estimated carcinogenic potential of each compound and dose level identified several candidate molecular markers of rodent nongenotoxic carcinogenicity, including transforming growth factor-beta stimulated clone 22 and NAD(P)H cytochrome P450 oxidoreductase.

Surrogate tissue analysis: monitoring toxicant exposure and health status of inaccessible tissues through the analysis of accessible tissues and cells

Genomics and proteomics have made it possible to define molecular physiology in exquisite detail, when tissues are accessible for sampling. However, many tissues are not accessible for human diagnostic evaluations or experimental studies, creating the need for surrogates that afford insight into exposures and effects in such tissues. Surrogate tissue analysis (STA) incorporating contemporary genomic and proteomic technologies may be useful in determining toxicant exposure and effect, or disease state, in target tissues at the pre- or early clinical stage. We present here a discussion of STA based on presentations given at the Society of Toxicology's 2003 annual meeting's "Innovations in Applied Toxicology" symposium. Speakers at the symposium (Box 1) discussed various potential applications of STA, including the use of peripheral blood lymphocytes (PBLs) as a source of genetic biomarkers to monitor radiation exposure; the use of gene expression analysis of PBLs and hair follicles as a means to monitor the impact of toxicants on inaccessible organs; the characterization of disease-associated gene signatures in peripheral blood mononuclear cells (PBMCs) of renal cell carcinoma (RCC) patients; the use of sperm RNA to determine genetic and environmental effects on sperm development in the testis; and the use of serum protein profiles to monitor the development and progression of various cancers. Also discussed are some of the challenges that must be overcome if the utility of STA is to be proven, and thus permit researchers to move this concept from the laboratory to the clinical environment.

Gene expression analysis in response to lung toxicants: I. Sequencing and microarray development

A key challenge in measuring gene expression changes in the lung in response to site-selective toxicants is differentiating between target and nontarget areas. The toxicity for the cytotoxicant 1-nitronaphthalene is highly localized in the airway epithelium. Target cells comprise but a fraction of the total lung cell mass; measurements from whole lung homogenates are not likely to reflect what occurs at the target site. Additionally, the use of generic microarrays to measure expression in airway epithelium may not provide a good representation of transcripts present at the site of toxic action. cDNA libraries from airway and alveolar subcompartments of rat lung were sequenced for the development of a custom microarray representative of these lung regions. We identified 7,460 nonredundant rat lung sequences. Nearly 30% of the sequences on this array are not present on the Affymetrix Rat GeneChip 230. A 20,000-element microarray was developed that delineates differences in gene expression between subcompartments. This is the first in a series of articles employing this microarray for detecting gene expression changes during acute injury produced by 1-nitronaphthalene and subsequent repair.

Overview of an interlaboratory collaboration on evaluating the effects of model hepatotoxicants on hepatic gene expression

DNA microarrays and related tools offer promise for identification of pathways involved in toxic responses to xenobiotics. To be useful for risk assessment, experimental data must be challenged for reliability and interlaboratory reproducibility. Toward this goal, the Hepatotoxicity Working Group of the International Life Sciences Institute (ILSI) Health and Environmental Sciences Institute (HESI) Technical Committee on Application of Genomics to Mechanism-Based Risk Assessment evaluated and compared biological and gene expression responses in rats exposed to two model hepatotoxins--clofibrate and methapyrilene. This collaborative effort provided an unprecedented opportunity for the working group to evaluate and compare multiple biological, genomic, and toxicological parameters across different laboratories and microarray platforms. Many of the results from this collaboration are presented in accompanying articles in this mini-monograph, whereas others have been published previously. (Italic)In vivo(/Italic) studies for both compounds were conducted in two laboratories using a standard experimental protocol, and RNA samples were distributed to 16 laboratories for analysis on six microarray platforms. Histopathology, clinical chemistry, and organ weight changes were consistent with reported effects. Gene expression results demonstrated reasonable agreement between laboratories and across platforms. Discrepancies in expression profiles of some individual genes were largely due to platform differences and approaches to data analysis rather than to biological or interlaboratory variability. Despite these discrepancies there was overall agreement in the biological pathways affected by these compounds, demonstrating that transcriptional profiling is reproducible between laboratories and can reliably identify affected pathways necessary to provide mechanistic insight. This effort represents an important first step toward the use of transcriptional profiling in risk assessment.

Identification of platform-independent gene expression markers of cisplatin nephrotoxicity

Within the International Life Sciences Institute Committee on Genomics, a working group was formed to focus on the application of microarray technology to preclinical assessments of drug-induced nephrotoxicity. As part of this effort, Sprague-Dawley rats were treated with the nephrotoxicant cisplatin at doses of 0.3-5 mg/kg over a 4- to 144-hr time course. RNA prepared from these animals was run on a variety of microarray formats at multiple sites. A set of 93 differentially expressed genes associated with cisplatin-induced renal injury was identified on the National Institute of Environmental Health Sciences (NIEHS) custom cDNA microarray platform using quadruplicate measurements of pooled animal RNA. The reproducibility of this profile of statistically significant gene changes on other platforms, in pooled and individual animal replicate samples, and in an independent study was investigated. A good correlation in response between platforms was found among the 48 genes in the NIEHS data set that could be matched to probes on the Affymetrix RGU34A array by UniGene identifier or sequence alignment. Similar results were obtained with genes that could be linked between the NIEHS and Incyte or PHASE-1 arrays. The degree of renal damage induced by cisplatin in individual animals was commensurate with the number of differentially expressed genes in this data set. These results suggest that gene profiles linked to specific types of tissue injury or mechanisms of toxicity and identified in well-performed replicated microarray experiments may be extrapolatable across platform technologies, laboratories, and in-life studies.

Clofibrate-induced gene expression changes in rat liver: a cross-laboratory analysis using membrane cDNA arrays

Microarrays have the potential to significantly impact our ability to identify toxic hazards by the identification of mechanistically relevant markers of toxicity. To be useful for risk assessment, however, microarray data must be challenged to determine reliability and interlaboratory reproducibility. As part of a series of studies conducted by the International Life Sciences Institute Health and Environmental Science Institute Technical Committee on the Application of Genomics to Mechanism-Based Risk Assessment, the biological response in rats to the hepatotoxin clofibrate was investigated. Animals were treated with high (250 mg/kg/day) or low (25 mg/kg/day) doses for 1, 3, or 7 days in two laboratories. Clinical chemistry parameters were measured, livers removed for histopathological assessment, and gene expression analysis was conducted using cDNA arrays. Expression changes in genes involved in fatty acid metabolism (e.g., acyl-CoA oxidase), cell proliferation (e.g., topoisomerase II-Alpha), and fatty acid oxidation (e.g., cytochrome P450 4A1), consistent with the mechanism of clofibrate hepatotoxicity, were detected. Observed differences in gene expression levels correlated with the level of biological response induced in the two in vivo studies. Generally, there was a high level of concordance between the gene expression profiles generated from pooled and individual RNA samples. Quantitative real-time polymerase chain reaction was used to confirm modulations for a number of peroxisome proliferator marker genes. Though the results indicate some variability in the quantitative nature of the microarray data, this appears due largely to differences in experimental and data analysis procedures used within each laboratory. In summary, this study demonstrates the potential for gene expression profiling to identify toxic hazards by the identification of mechanistically relevant markers of toxicity.

Identification of putative gene based markers of renal toxicity

This study, designed and conducted as part of the International Life Sciences Institute working group on the Application of Genomics and Proteomics, examined the changes in the expression profile of genes associated with the administration of three different nephrotoxicants--cisplatin, gentamicin, and puromycin--to assess the usefulness of microarrays in the understanding of mechanism(s) of nephrotoxicity. Male Sprague-Dawley rats were treated with daily doses of puromycin (5-20 mg/kg/day for 21 days), gentamicin (2-240 mg/kg/day for 7 days), or a single dose of cisplatin (0.1-5 mg/kg). Groups of rats were sacrificed at various times after administration of these compounds for standard clinical chemistry, urine analysis, and histological evaluation of the kidney. RNA was extracted from the kidney for microarray analysis. Principal component analysis and gene expression-based clustering of compound effects confirmed sample separation based on dose, time, and degree of renal toxicity. In addition, analysis of the profile components revealed some novel changes in the expression of genes that appeared to be associated with injury in specific portions of the nephron and reflected the mechanism of action of these various nephrotoxicants. For example, although puromycin is thought to specifically promote injury of the podocytes in the glomerulus, the changes in gene expression after chronic exposure of this compound suggested a pattern similar to the known proximal tubular nephrotoxicants cisplatin and gentamicin; this prediction was confirmed histologically. We conclude that renal gene expression profiling coupled with analysis of classical end points affords promising opportunities to reveal potential new mechanistic markers of renal toxicity.

Comparison of gene expression changes induced in mouse and human cells treated with direct-acting mutagens

Exposure to DNA-damaging agents can elicit a variety of stress-related responses that may alter the gene expression of numerous biological pathways. We used Affymetrix microarrays to detect gene expression changes in mouse lymphoma (L5178Y) and human lymphoblastoid (TK6) cells in response to methyl methanesulfonate (MMS; a prototypical alkylating agent) and bleomycin (a prototypical oxidative mutagen). Cells were treated for 4 hr, and RNA was isolated either at the end of the treatment or after a 20-hr recovery period. Two concentrations of each agent were used based on cytotoxicity levels and Tk mutant frequencies. Our microarray data analysis indicated that MMS and bleomycin gene expression responses were considerably different in mouse cells versus human cells. The results also suggested that more comprehensive cellular responses to MMS and bleomycin occurred in TK6 cells than in L5178Y cells. In contrast to L5178Y cells, the response of TK6 cells to MMS and bleomycin was characterized by the induction of p53-dependent genes that are involved in DNA repair, cell cycle regulation, and apoptosis. It appears that the induction of DNA damage by MMS in human TK6 cells mediated cytotoxicity and led to decreased cell survival. This may explain the greater sensitivity of TK6 cells to cytotoxic effects of MMS compared to L5178Y cells. Bleomycin exerted comparable cytotoxic effects in the two cell lines. Overall, these studies were unable to identify distinctive gene expression changes that differentiated bleomycin from MMS in either TK6 cells or mouse lymphoma cells.

Toxicogenomic approach for assessing toxicant-related disease

The problems of identifying environmental factors involved in the etiology of human disease and performing safety and risk assessments of drugs and chemicals have long been formidable issues. Three principal components for predicting potential human health risks are: (1) the diverse structure and properties of thousands of chemicals and other stressors in the environment; (2) the time and dose parameters that define the relationship between exposure and disease; and (3) the genetic diversity of organisms used as surrogates to determine adverse chemical effects. The global techniques evolving from successful genomics efforts are providing new exciting tools with which to address these intractable problems of environmental health and toxicology. In order to exploit the scientific opportunities, the National Institute of Environmental Health Sciences has created the National Center for Toxicogenomics (NCT). The primary mission of the NCT is to use gene expression technology, proteomics and metabolite profiling to create a reference knowledge base that will allow scientists to understand mechanisms of toxicity and to be able to predict the potential toxicity of new chemical entities and drugs. A principal scientific objective underpinning the use of microarray analysis of chemical exposures is to demonstrate the utility of signature profiling of the action of drugs or chemicals and to utilize microarray methodologies to determine biomarkers of exposure and potential adverse effects. The initial approach of the NCT is to utilize proof-of-principle experiments in an effort to "phenotypically anchor" the altered patterns of gene expression to conventional parameters of toxicity and to define dose and time relationships in which the expression of such signature genes may precede the development of overt toxicity. The microarray approach is used in conjunction with proteomic techniques to identify specific proteins that may serve as signature biomarkers. The longer-range goal of these efforts is to develop a reference relational database of chemical effects in biological systems (CEBS) that can be used to define common mechanisms of toxicity, chemical and drug actions, to define cellular pathways of response, injury and, ultimately, disease. In order to implement this strategy, the NCT has created a consortium of research organizations and private sector companies to actively collaborative in populating the database with high quality primary data. The evolution of discrete databases to a knowledge base of toxicogenomics will be accomplished through establishing relational interfaces with other sources of information on the structure and activity of chemicals such as that of the National Toxicology Program (NTP) and with databases annotating gene identity, sequence, and function.

Separation and detection methods for covalent drug–protein adducts

Covalent binding of reactive metabolites of drugs to proteins has been a predominant hypothesis for the mechanism of toxicity caused by numerous drugs. The development of efficient and sensitive analytical methods for the separation, identification, quantification of drug-protein adducts have important clinical and toxicological implications. In the last few decades, continuous progress in analytical methodology has been achieved with substantial increase in the number of new, more specific and more sensitive methods for drug-protein adducts. The methods used for drug-protein adduct studies include those for separation and for subsequent detection and identification. Various chromatographic (e.g., affinity chromatography, ion-exchange chromatography, and high-performance liquid chromatography) and electrophoretic techniques [e.g., sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), two-dimensional SDS-PAGE, and capillary electrophoresis], used alone or in combination, offer an opportunity to purify proteins adducted by reactive drug metabolites. Conventionally, mass spectrometric (MS), nuclear magnetic resonance, and immunological and radioisotope methods are used to detect and identify protein targets for reactive drug metabolites. However, these methods are labor-intensive, and have provided very limited sequence information on the target proteins adducted, and thus the identities of the protein targets are usually unknown. Moreover, the antibody-based methods are limited by the availability, quality, and specificity of antibodies to protein adducts, which greatly hindered the identification of specific protein targets of drugs and their clinical applications. Recently, the use of powerful MS technologies (e.g., matrix-assisted laser desorption/ionization time-of-flight) together with analytical proteomics have enabled one to separate, identify unknown protein adducts, and establish the sequence context of specific adducts by offering the opportunity to search for adducts in proteomes containing a large number of proteins with protein adducts and unmodified proteins. The present review highlights the separation and detection technologies for drug-protein adducts, with an emphasis on methodology, advantages and limitations to these techniques. Furthermore, a brief discussion of the application of these techniques to individual drugs and their target proteins will be outlined.

Analysis of gene expression in carbon tetrachloride-treated rat livers using a novel bioarray technology

The present study successfully utilizes a new ADME Rat Expression Bioarray, containing 1040 metabolism- and toxicology-linked genes, to monitor gene expression from the livers of rats treated with carbon tetrachloride (CCl(4)). Histopathological analysis, hierarchical clustering methods, and gene expression profiling are compared between the control and CCl(4)-treated animals. A total of 44 transcripts were found to be altered in response to the hepatotoxin, 19 of which were upregulated and 25 were downregulated. Some of these gene expression changes were expected and concurred with previously published data while others were novel findings.

An evaluation of a low-density DNA microarray using cytochrome P450 inducers

The aim of this study was to validate a low-density DNA microarray "Rat HepatoChip", which contains 59 genes from a range of potential toxic markers and drug metabolism-related genes. Liver mRNA was isolated from rats dosed with six different chemicals, dexamethasone, troleandomycin, miconazole, clotrimazole, and methylclofanapate, which are all known to induce different cytochrome P450 genes, and isoniazid, which does not cause histopathological changes. Replicate microarrays were used to measure the variability in the chips and in the process. The average variability in signal between different chips observed in triplicate experiments was 33% ranging from 21 to 39% depending on genes. We also demonstrated a strong correlation between the liver histopathology and the gene expression profiles indicating that the gene expression profile reflects histopathological changes. These results suggest that the Rat HepatoChip microarray may provide a fast and effective tool for assessing the toxicity profile of developmental drug candidates during the drug discovery process.

Gene expression dose-response changes in microarrays after exposure of human peripheral lung epithelial cells to nickel(II)

Occupational exposure to nickel compounds is associated with lung cancer risk; both genotoxic and epigenetic mechanisms have been proposed. For comprehensive examination of the acute effects of nickel(II) acetate on gene expression in cultured human peripheral lung epithelial HPL1D cells, microarray analyses were carried out with cDNA chips (approximately 8000 cDNAs). Cells were exposed for 24 h to nontoxic (50, 100, and 200 microM) or toxic (400, 800, and 1600 microM) nickel(II) concentrations. Cluster analysis was applied to the 868 genes with > or = 2-fold change at any concentration. Two main clusters showed marked up- or down-regulation at the highest, toxic concentrations. The data further subdivided into 10 highly cohesive clusters with high probability, and of these only 2 had the same response trend at low nontoxic as at high concentrations, an observation of clear relevance to the process of high- to low-dose extrapolation in risk assessment. There were 113 genes showing > or = 2-fold change at the three lower nontoxic concentrations, those most relevant to in vivo carcinogenesis. In addition to expected responses of metallothionein, ferritin, and heat-shock proteins, the results revealed for the first time changed expression of some potential cancer-related genes in response to low-dose Ni(II): RhoA, dyskerin, interferon regulatory factor 1, RAD21 homologue, and tumor protein, translationally controlled. Overall, most of the genes impacted by nontoxic concentrations of nickel(II) acetate related to gene transcription, protein synthesis and stability, cytoskeleton, signaling, metabolism, cell membrane, and extracellular matrix.

The toxicogenomics of nuclear receptor agonists

Toxicogenomics is the study of the structure and output of the genome as it responds to adverse xenobiotic exposure. Large-scale transcriptional analysis, made possible through microarray technologies, enables us to study and understand the complexity of the biological effects of drugs and chemicals, with the ultimate goal of separating wanted effects from adverse effects. Nuclear receptors are attractive targets for drug discovery because, as ligand-activated transcription factors, they coordinately regulate the expression of at least hundreds of genes that, in turn, control much of cellular metabolism. Through toxicogenomics, it is becoming possible to understand the therapeutic effects of agonists within the context of toxic effects, classify new chemicals as to their complete effects on biological systems, and identify environmental factors that may influence safety or efficacy of new and existing drugs.

Model systems in drug discovery: chemical genetics meets genomics

Animal model systems are an intricate part of the discovery and development of new medicines. The sequencing of not only the human genome but also those of the various pathogenic bacteria, the nematode Caenorhabditis elegans, the fruitfly Drosophila, and the mouse has enabled the discovery of new drug targets to push forward at an unprecedented pace. The knowledge and tools in these "model" systems are allowing researchers to carry out experiments more efficiently and are uncovering previously hidden biological connections. While the history of bacteria, yeast, and mice in drug discovery are long, their roles are ever evolving. In contrast, the history of Drosophila and C. elegans at pharmaceutical companies is short. We will briefly review the historic role of each model organism in drug discovery and then update the readers as to the abilities and liabilities of each model within the context of drug development.

Combined histomorphometric and gene-expression profiling applied to toxicology

We have developed a unique methodology for the combined analysis of histomorphometric and gene-expression profiles amenable to intensive data mining and multisample comparison for a comprehensive approach to toxicology. This hybrid technology, termed extensible morphometric relational gene-expression analysis (EMeRGE), is applied in a toxicological study of time-varied vehicle- and carbon-tetrachloride (CCl4)-treated rats, and demonstrates correlations between specific genes and tissue structures that can augment interpretation of biological observations and diagnosis.

OligoDesign: Optimal design of LNA (locked nucleic acid) oligonucleotide capture probes for gene expression profiling

We report the development of new software, OligoDesign, which provides optimal design of LNA (locked nucleic acid) substituted oligonucleotides for functional genomics applications. LNAs constitute a novel class of bicyclic RNA analogs having an exceptionally high affinity and specificity toward their complementary DNA and RNA target molecules. The OligoDesign software features recognition and filtering of the target sequence by genome-wide BLAST analysis in order to minimize cross-hybridization with non-target sequences. Furthermore it includes routines for prediction of melting temperature, self-annealing and secondary structure for LNA substituted oligonucleotides, as well as secondary structure prediction of the target nucleotide sequence. Individual scores for all these properties are calculated for each possible LNA oligonucleotide in the query gene and the OligoDesign program ranks the LNA capture probes according to a combined fuzzy logic score and finally returns the top scoring probes to the user in the output. We have successfully used the OligoDesign tool to design a Caenorhabditis elegans LNA oligonucleotide microarray, which allows monitoring of the expression of a set of 120 potential marker genes for a variety of stress and toxicological processes and toxicologically relevant pathways. The OligoDesign program is freely accessible at http://lnatools.com/.