New limits proposed for the management of non-mutagenic impurities.

Multiple international guidelines exist that describe both quality and safety considerations for the control of the broad spectrum of impurities inherent to drug substance and product manufacturing processes. However, regarding non-mutagenic impurities (NMI) the most relevant ICH Q3A/B guidelines are not applicable during early phases of drug development leading to confusion about acceptable limits at this stage. Thus, there is need for more flexible approaches that ensure that patient safety remains paramount, while taking into consideration the limited duration of exposure. An EFPIA survey, which collected quantitative data from different types of studies applied to qualify impurities in accordance with ICH Q3A, shows that no toxicities could be attributed to any of the 467 impurities at any tested level in vivo. This data combined with earlier published toxicological datasets encompassing drug substances and intermediates, food related substances and chemicals provide convincing evidence that for NMIs, the application of a generic 5 mg/day limit for an exposure duration <6 months, and a 1 mg/day generic limit for life-long exposure, provides sufficient margins to ensure patient safety. Hence, application of these absolute limits to trigger qualification studies (instead of the relative limits described in Q3A/B), is considered warranted. This approach will prevent conduct of unnecessary dedicated impurity qualification studies and the resulting use of animals.

Acceptable intakes (AIs) for 11 small molecule N-nitrosamines (NAs).

Low levels of N-nitrosamines (NAs) were detected in pharmaceuticals and, as a result, health authorities (HAs) have published acceptable intakes (AIs) in pharmaceuticals to limit potential carcinogenic risk. The rationales behind the AIs have not been provided to understand the process for selecting a TD50 or read-across analog. In this manuscript we evaluated the toxicity data for eleven common NAs in a comprehensive and transparent process consistent with ICH M7. This evaluation included substances which had datasets that were robust, limited but sufficient, and substances with insufficient experimental animal carcinogenicity data. In the case of robust or limited but sufficient carcinogenicity information, AIs were calculated based on published or derived TD50s from the most sensitive organ site. In the case of insufficient carcinogenicity information, available carcinogenicity data and structure activity relationships (SARs) were applied to categorical-based AIs of 1500 ng/day, 150 ng/day or 18 ng/day; however additional data (such as biological or additional computational modelling) could inform an alternative AI. This approach advances the methodology used to derive AIs for NAs.

PPAR agonists reduce steatosis in oleic acid-overloaded HepaRG cells.

UNLABELLED: Although non-alcoholic fatty liver disease (NAFLD) is currently the most common form of chronic liver disease there is no pharmacological agent approved for its treatment. Since peroxisome proliferator-activated receptors (PPARs) are closely associated with hepatic lipid metabolism, they seem to play important roles in NAFLD. However, the effects of PPAR agonists on steatosis that is a common pathology associated with NAFLD, remain largely controversial. In this study, the effects of various PPAR agonists, i.e. fenofibrate, bezafibrate, troglitazone, rosiglitazone, muraglitazar and tesaglitazar on oleic acid-induced steatotic HepaRG cells were investigated after a single 24-hour or 2-week repeat treatment. Lipid vesicles stained by Oil-Red O and triglycerides accumulation caused by oleic acid overload, were decreased, by up to 50%, while fatty acid oxidation was induced after 2-week co-treatment with PPAR agonists. The greatest effects on reduction of steatosis were obtained with the dual PPARα/γ agonist muraglitazar. Such improvement of steatosis was associated with up-regulation of genes related to fatty acid oxidation activity and down-regulation of many genes involved in lipogenesis. Moreover, modulation of expression of some nuclear receptor genes, such as FXR, LXRα and CAR, which are potent actors in the control of lipogenesis, was observed and might explain repression of de novo lipogenesis. CONCLUSION: Altogether, our in vitro data on steatotic HepaRG cells treated with PPAR agonists correlated well with clinical investigations, bringing a proof of concept that drug-induced reversal of steatosis in human can be evaluated in in vitro before conducting long-term and costly in vivo studies in animals and patients.

Interindividual variability in gene expression profiles in human hepatocytes and comparison with HepaRG cells.

Interindividual variations in functions other than drug metabolism activity, remain poorly elucidated in human liver. In the present study, the whole transcriptome of several human hepatocyte populations and the differentiated human HepaRG cell line have been analyzed and compared, using oligonucleotide pangenomic microarrays. We show that, although the variation in the percentages of expressed genes did not exceed 14% among the primary human hepatocyte populations, huge interindividual differences in the transcript levels of many genes were observed. These genes were related to various functions; in addition to drug metabolism, they mainly concerned carbohydrate, amino acid, and lipid metabolism. HepaRG cells expressed from 81 to 92% of the genes active in human hepatocytes and, in addition, a specific gene subset mainly related to their transformed status, some chromosomal abnormalities, and the presence of primitive biliary epithelial cells. Of interest, a relationship was evidenced between abnormal basal expression levels of some target genes and their corresponding previously reported fold changes in one of four human hepatocyte populations treated with the hepatotoxic drug troglitazone and not with other nonhepatotoxic peroxisome proliferator-activated receptor agonists (PLoS One 6:e18816, 2011). Taken together, our results support the view that HepaRG cells express most of the genes active in primary human hepatocytes and show that expression of most human hepatic genes can quantitatively greatly vary among individuals, thereby contributing to explain the huge interindividual variability in susceptibility to drugs and other environmental factors.

Comparative gene expression profiles induced by PPARγ and PPARα/γ agonists in rat hepatocytes.

Species-differential toxic effects have been described with PPARα and PPARγ agonists between rodent and human liver. PPARα agonists (fibrates) are potent hypocholesterolemic agents in humans while they induce peroxisome proliferation and tumors in rodent liver. By contrast, PPARγ agonists (glitazones) and even dual PPARα/γ agonists (glitazars) have caused idiosyncratic hepatic and nonhepatic toxicities in human without evidence of any damage in rodent during preclinical studies. The mechanisms involved in such differences remain largely unknown. Several studies have identified the major target genes of PPARα agonists in rodent liver while no comprehensive analysis has been performed on gene expression changes induced by PPARγ and dual PPARα/γ agonists. Here, we investigated transcriptomes of rat hepatocytes after 24h treatment with two PPARγ (troglitazone and rosiglitazone) and two PPARα/γ (muraglitazar and tesaglitazar) agonists. Although, hierarchical clustering revealed a gene expression profile characteristic of each PPAR agonist class, only a limited number of genes was specifically deregulated by glitazars. Functional analyses showed that many genes known as PPARα targets were also modulated by both PPARγ and PPARα/γ agonists and quantitative differences in gene expression profiles were observed between these two classes. Moreover, most major genes modulated in rat hepatocytes were also found to be deregulated in rat liver after tesaglitazar treatment. Taken altogether, these results support the conclusion that differential toxic effects of PPARα and PPARγ agonists in rodent liver do not result from transcriptional deregulation of major PPAR target genes but rather from qualitative and/or quantitative differential responses of a small subset of genes.

Study of oxidative DNA damage in TK6 human lymphoblastoid cells by use of the thymidine kinase gene-mutation assay and the in vitro modified comet assay: determination of No-Observed-Genotoxic-Effect-Levels.

Nowadays, there is clear progress in using the threshold concept in genetic toxicology, but its demonstration and acceptance in risk assessment is still under debate. Although it has been accepted for some non-DNA-reactive agents for which mechanisms of action were demonstrated, there is a growing weight of evidence to also support the existence of thresholded dose-responses for DNA-reactive agents. In this context, we have recently shown in human TK6 lymphoblastoid cells, that DNA-oxidizing agents [potassium bromate, bleomycin and hydrogen peroxide (via glucose oxidase)] produced non-linear dose-responses in the in vitro micronucleus test, thus allowing the determination of No-Observed-Genotoxic-Effect-Levels (NOGELs). Therefore, the aim of the present study was to focus on the analysis of thresholded dose-response curves in order to further investigate the existence of NOGELs for these same directly DNA-damaging agents, by use of other genotoxicity endpoints. Mutation frequency was determined after a 1-h treatment in the thymidine kinase (TK) gene-mutation assay. Primary DNA damage, especially oxidative DNA damage, was also assessed after 1h of treatment, followed - or not - by a 23-h recovery period, with the modified version of the comet assay (i.e. with the glycosylases Fpg and hOgg1). Overall, our analysis demonstrates that there is convincing evidence to support the existence of thresholded dose-responses for DNA-oxidizing agents. The determination of NOGELs depends on the genotoxic endpoint studied and consequently requires different genotoxicity assays performed concurrently. NOGELs could only be defined for the induction of chromosomal aberrations and gene mutations, i.e. for an effect-endpoint but not for primary DNA damage, i.e. for an exposure-endpoint. Further statistical analyses of these data are now required in order to draw conclusions on the exact level of the thresholds.

Study of gene expression profiles in TK6 human cells exposed to DNA-oxidizing agents.

During the last decade, there has been clear progress in using threshold in risk assessment but its acceptance by scientists is still under debate. Contrary to indirect DNA-damaging agents, DNA-reactive agents have been assumed to have a non-threshold mode of action, as they directly induce DNA lesions that potentially can be converted into mutations. However, in recent years there is a growing number of data establishing threshold doses even for these DNA-reactive compounds. Indeed, there are several defence and repair mechanisms that provide protection and that may be responsible for genotoxic thresholds. In this context, we recently showed that DNA-oxidizing agents exhibit a thresholded dose-response in vitro with respect to chromosomal alterations. We have hypothesized the involvement of different cellular responses whose nature and efficiency depend on the stress level. The aim of this study was to develop a more complete understanding of these underlying mechanisms. We investigated global gene expression profiles of human lymphoblastoid TK6 cells after exposure to potassium bromate and hydrogen peroxide (via glucose oxidase). Cells were treated for 1h and mRNAs were isolated either immediately at the end of the treatment or after a 23-h recovery period. Our results showed that cells have developed elaborate cellular responses to oxidative stress in order to maintain genomic integrity. Many of altered genes were redox-sensitive transcription factors such as p53, NF-kappaB, AP-1 and Nrf2. Their downstream target genes and signalling pathways were subsequently activated leading mainly to the induction of antioxidant defenses, inflammation, cell cycle arrest, DNA repair and cell death. Overall, our study allowed the identification of key events involved in the thresholded response observed after DNA-oxidizing agents exposure and shows the usefulness of the combination of standard in vitro genotoxicity assays with gene expression profiling technology to determine modes of action, particularly for critical risk assessment.

Dose- and time-dependent effects of phenobarbital on gene expression profiling in human hepatoma HepaRG cells.

Phenobarbital (PB) induces or represses a wide spectrum of genes in rodent liver. Much less is known about its effects in human liver. We used pangenomic cDNA microarrays to analyze concentration- and time-dependent gene expression profile changes induced by PB in the well-differentiated human HepaRG cell line. Changes in gene expression profiles clustered at specific concentration ranges and treatment times. The number of correctly annotated genes significantly modulated by at least three different PB concentration ranges (spanning 0.5 to 3.2 mM) at 20 h exposure amounted to 77 and 128 genes (p< or =0.01) at 2- and 1.8-fold filter changes, respectively. At low concentrations (0.5 and 1 mM), PB-responsive genes included the well-recognized CAR- and PXR-dependent responsive cytochromes P450 (CYP2B6, CYP3A4), sulfotransferase 2A1 and plasma transporters (ABCB1, ABCC2), as well as a number of genes critically involved in various metabolic pathways, including lipid (CYP4A11, CYP4F3), vitamin D (CYP24A1) and bile (CYP7A1 and CYP8B1) metabolism. At concentrations of 3.2 mM or higher after 20 h, and especially 48 h, increased cytotoxic effects were associated with disregulation of numerous genes related to oxidative stress, DNA repair and apoptosis. Primary human hepatocyte cultures were also exposed to 1 and 3.2 mM PB for 20 h and the changes were comparable to those found in HepaRG cells treated under the same conditions. Taken altogether, our data provide further evidence that HepaRG cells closely resemble primary human hepatocytes and provide new information on the effects of PB in human liver. These data also emphasize the importance of investigating dose- and time-dependent effects of chemicals when using toxicogenomic approaches.

Reproducible chemical-induced changes in gene expression profiles in human hepatoma HepaRG cells under various experimental conditions.

The use of in vitro human liver cell models is an attractive approach in toxicogenomic studies designed to analyze gene expression changes induced by a toxic chemical. However, in such studies, reliability, reproducibility and interlaboratory concordance of microarrays, as well as the choice of the most suitable cell model, remain a matter of debate. This work was aimed at evaluating the robustness of microarray technologies and the suitability of the highly differentiated human HepaRG cell line in the investigation of gene expression changes induced by a toxic compound in human liver. The influence of various experimental conditions including cell cultures grown at different test sites, different generations of microarrays, RNA analysis platforms and softwares, was tested on gene expression profiles induced by a 20h treatment with an 8mM concentration of phenobarbital as the toxic compound. As many as 1099 genes (p-value<0.01 and 1.5-fold-change), representing 74% and 30% of the signature genes detected with Agilent 22 and 44K pangenomic microarrays, respectively, were shown to be modulated in common in six independently performed experiments. The most modulated genes included both those known to be regulated by phenobarbital, such as cytochromes P450 and membrane transporters, and those involved in oxidative stress, inflammation and apoptosis, typifying a toxic insult. These data provide strong support for the use of a toxicogenomic approach for the in vitro prediction of chemical toxicity, and for the choice of human HepaRG cells as a promising model system for human hepatotoxicity testing.

[In silico, in vitro, in omic experimental models and drug safety evaluation]. / La place des méthodes in silico, in vitro, in omic dans l'évaluation de la sécurité des médicaments.

Over the last few decades, toxicology has benefited from scientific, technical, and bioinformatic developments relating to patient safety assessment during clinical and drug marketing studies. Based on this knowledge, new in silico, in vitro, and "omic" experimental models are emerging. Although these models cannot currently replace classic safety evaluations performed on laboratory animals, they allow compounds with unacceptable toxicity to be rejected in the early stages of drug development, thereby reducing the number of laboratory animals needed. In addition, because these models are particularly adapted to mechanistic studies, they can help to improve the relevance of the data obtained, thus enabling better prevention and screening of the adverse effects that may occur in humans. Much progress remains to be done, especially in the field of validation. Nevertheless, current efforts by industrial, academic laboratories, and regulatory agencies should, in coming years, significantly improve preclinical drug safety evaluation thanks to the integration of these new methods into the drug research and development process.

[Reducing the use of laboratory animals]. / La réduction du recours a l'animal de laboratoire.

Since 1959, when Russel and Burch formulated the 3Rs principle (Reduce, Replace, Refine), the scientific community has been attempting to reduce the use of laboratory animals for research purposes. Current regulatory guidelines take this principle into account. Thanks to scientific and technical progress, and advances in bioinformatics, new tools are now available that reduce the need for laboratory animals, albeit without totally replacing them. Implementation of the International Conference on Harmonization recommendations in 1990 represented a major step forward, notably by helping to avoid duplication of studies using laboratory animals. The use of animals for cosmetics testing is now forbidden in the European Union. Although new in vitro and in silico models remain to be validated, they are proving particularly useful during the early stages of product development, by avoiding experimental studies of chemicals that are ineffective or excessively toxic. The success of these measures is reflected in the results of a European study showing a fall, between 1996 and 2005, in the number of laboratory animals used for research and development, despite a large increase in overall research activities. The challenge for the next decade is to amplify this trend.

Genetic toxicity assessment: employing the best science for human safety evaluation. Part II: Performances of the in vitro micronucleus test compared to the mouse lymphoma assay and the in vitro chromosome aberration assay.

The in vitro micronucleus test is commonly used in the early stages of pharmaceutical development as a predictive tool for the regulatory mouse lymphoma assay or in vitro chromosome aberration test. The accumulated data from this assay leads to the suggestion that it could be used as an alternative to the chromosome aberration test or the mouse lymphoma assay in the regulatory genotoxicity battery. In this paper, we present the results of the in vitro micronucleus test on L5178Y mouse lymphoma cells with 25 compounds from Servier research and have compared these results to those obtained in the genotoxicity regulatory battery. All the negative compounds were also negative in the in vitro micronucleus assay. Among the 14 positive compounds, two of them, positive in the mouse lymphoma assay, were found negative in the in vitro micronucleus test. However, this apparent discordance was likely to be due to cytotoxicity- or high concentration-related false positive responses in the mouse lymphoma assay. In addition, we confirmed that the in vitro micronucleus assay is useful for detecting aneugens, especially, when cells in metaphasis and multinucleated cells are also scored and when cells are allowed to recover after the long treatment. On this series of compounds, the in vitro micronucleus assay showed high sensitivity and possibly a better specificity than the mouse lymphoma assay. Thus, the in vitro micronucleus assay was shown to be at least as adequate as the mouse lymphoma assay or the in vitro chromosome aberration test to be used in the standard genotoxicity battery.

Genetic toxicity assessment: employing the best science for human safety evaluation part IV: a strategy in genotoxicity testing in drug development: some examples.

The minimal three-test battery of the International Conference on Harmonization guideline has been in use since 1997 for the development of new pharmaceuticals (ICH, 1997). After a 10-year experience of this core battery in regulatory genotoxicity testing, everywhere the time has come for reflection about what was learned from this experience. Different aspects of genotoxicity testing are currently being debated under different organizations (HESI, 2006; IWGT, 2007; Kirkland et al., 2007). The main concerns are to develop relevant strategies and adequate complementary tests to the minimal battery, appropriate for each specific case to assess risk for humans when in vitro positive results or findings in rodent bioassays for carcinogenicity are found. In this article, an example of an in-house decision tree is shown, with some options which can contribute to the current reflections. Additionally, tools built for early genotoxicity are presented.

The European Registered Toxicologist (ERT): Current status and prospects for advancement.

Following its inception in 1994, the certification of European Registered Toxicologists (ERT) by EUROTOX has been recognized as ensuring professional competence as well as scientific integrity and credibility. Criteria and procedures for registration are contained in the ERT "Guidelines for Registration 2012". The register of ERT currently has over 1900 members. In order to continue the harmonisation of requirements and processes between national registering bodies as a prerequisite for official recognition of the ERT title as a standard, and to take account of recent developments in toxicology, an update of the ERT Guidelines has been prepared in a series of workshops by the EUROTOX subcommittees for education and registration, in consultation with representatives of national toxicology societies and registers. The update includes details of topics and learning outcomes for theoretical training, and how these can be assessed. The importance of continuing professional development as the cornerstone of re-registration is emphasised. To help with the process of harmonisation, it is necessary to collate and share best practices of registration conditions and procedures across Europe. Importantly, this information can also be used to audit compliance with the EUROTOX standards. As recognition of professionals in toxicology, including specialist qualifications, is becoming more important than ever, we believe that this can best be achieved based on the steps for harmonisation outlined here together with the proposed new Guidelines.

Genotoxicity of tungsten carbide-cobalt (WC-Co) nanoparticles in vitro: mechanisms-of-action studies.

We showed previously that tungsten carbide-cobalt (WC-Co) nanoparticles (NP) can be used as a nanoparticulate positive control in some in vitro mammalian genotoxicity assays. Here, we investigate the mechanisms of action involved in WC-Co NP genotoxicity in L5178Y mouse lymphoma cells and primary human lymphocytes, in vitro. Data from the micronucleus assay coupled with centromere staining and from the chromosome-aberration assay show the involvement of both clastogenic and aneugenic events. Experiments with the formamidopyrimidine DNA glycosylase (FPG)-modified comet assay showed a slight (non-significant) increase in FPG-sensitive sites in the L5178Y mouse lymphoma cells but not in the human lymphocytes. Electron paramagnetic resonance spin-trapping results showed the presence of hydroxyl radicals (•OH) in WC-Co NP suspensions, with or without cells, but with time-dependent production in the presence of cells. However, a significant difference in •OH production was observed between human lymphocytes from two different donors. Using H2O2, we showed that WC-Co NP can participate in Fenton-like reactions. Thus, •OH might be produced either via intrinsic generation by WC-Co NP or through a Fenton-like reaction in the presence of cells.

Safety pharmacology: an essential interface of pharmacology and toxicology in the non-clinical assessment of new pharmaceuticals.

Safety pharmacology studies are defined as the studies that investigate the potential undesirable pharmacodynamic effects of a substance on physiological functions in relation to exposure. In consequence, these studies are an integral part of the non-clinical safety assessment of new pharmaceuticals, in association with toxicological studies. A retrospective shows the evolution of the discipline in these last years. Safety pharmacology studies are of special interest, and some drawbacks and pitfalls must be considered (i.e. invasive methods, difficulties related to GLP (good laboratory practices) requirements, choice of a strategy). In the future, some priority should be given to education, promotion of scientific activities, reinforcement of the links between pharmacologists and toxicologists and implementation of relevant guidelines.

Value of in vitro models for the assessment of drug-induced haematotoxicity.

BACKGROUND: A new antipsychotic compound induced unexpected red cell hypoplasia (reticulocytopenia, red marrow hypoplasia) in rats dosed orally for 7 days. MATERIALS AND METHODS: Since an erythropoietin-mediated pathogenesis was excluded, in vitro tests on rat and human bone marrow cells were performed with measurement of formation of late erythroid (CFU-E) and granulocyte-macrophage (CFU-GM) colony-forming units after incubation with the drug. CFU-E together with growth factors were cultured for 2 days (rat) or 7 days (human) and CFU-GM was cultured for 7 days (rat) or 10 days (human). RESULTS: The drug induced inhibition of erythroid progenitors and myeloid progenitors for both species from 3 x 10(-5) mol/L, with the concentration inhibiting the growth of 50% (IC50) consistent with drug plasma levels measured in rats. CONCLUSION: These cloning assays on rat bone-marrow cells were shown to be adequate models for determining the haematotoxicity of the agent and to be predictive of human toxicity. With only a small amount of compound required, they can be used as screening tools to detect haematotoxic potential of candidate drugs.

Tungsten carbide-cobalt as a nanoparticulate reference positive control in in vitro genotoxicity assays.

With the increasing human exposure to nanoparticles (NP), the evaluation of their genotoxic potential is of significant importance. However, relevance for NP of the routinely used in vitro genotoxicity assays is often questioned, and a nanoparticulate reference positive control would therefore constitute an important step to a better testing of NP, ensuring that test systems are really appropriate. In this study, we investigated the possibility of using tungsten carbide-cobalt (WC-Co) NP as reference positive control in in vitro genotoxicity assays, including 2 regulatory assays, the mouse lymphoma assay and the micronucleus assay, and in the Comet assay, recommended for the toxicological evaluation of nanomedicines by the French Agency of Human Health Products (Afssaps). Through these assays, we were able to study different genetic endpoints in 2 cell types commonly used in regulatory genotoxicity assays: the L5178Y mouse lymphoma cell line and primary cultures of human lymphocytes. Our results showed that the use of WC-Co NP as positive control in in vitro genotoxicity assays was conceivable, but that different parameters have to be considered, such as cell type and treatment schedule. L5178Y mouse lymphoma cells did not provide satisfactory results in the 3 performed tests. However, human lymphocytes were more sensitive to genotoxic effects induced by WC-Co NP, particularly after a 24-h treatment in the in vitro micronucleus assay and after a 4-h treatment in the in vitro Comet assay. Under such conditions, WC-Co could be used as a nanoparticulate reference positive control in these assays.

Comparative gene expression profiles induced by PPARγ and PPARα/γ agonists in human hepatocytes.

BACKGROUND: Several glitazones (PPARγ agonists) and glitazars (dual PPARα/γ agonists) have been developed to treat hyperglycemia and, simultaneously, hyperglycemia and dyslipidemia, respectively. However, most have caused idiosyncratic hepatic or extrahepatic toxicities through mechanisms that remain largely unknown. Since the liver plays a key role in lipid metabolism, we analyzed changes in gene expression profiles induced by these two types of PPAR agonists in human hepatocytes. METHODOLOGY/PRINCIPAL FINDINGS: Primary human hepatocytes and the well-differentiated human hepatoma HepaRG cells were exposed to different concentrations of two PPARγ (troglitazone and rosiglitazone) and two PPARα/γ (muraglitazar and tesaglitazar) agonists for 24 h and their transcriptomes were analyzed using human pangenomic Agilent microarrays. Principal Component Analysis, hierarchical clustering and Ingenuity Pathway Analysis® revealed large inter-individual variability in the response of the human hepatocyte populations to the different compounds. Many genes involved in lipid, carbohydrate, xenobiotic and cholesterol metabolism, as well as inflammation and immunity, were regulated by both PPARγ and PPARα/γ agonists in at least a number of human hepatocyte populations and/or HepaRG cells. Only a few genes were selectively deregulated by glitazars when compared to glitazones, indicating that PPARγ and PPARα/γ agonists share most of their target genes. Moreover, some target genes thought to be regulated only in mouse or to be expressed in Kupffer cells were also found to be responsive in human hepatocytes and HepaRG cells. CONCLUSIONS/SIGNIFICANCE: This first comprehensive analysis of gene regulation by PPARγ and PPARα/γ agonists favor the conclusion that glitazones and glitazars share most of their target genes and induce large differential changes in gene profiles in human hepatocytes depending on hepatocyte donor, the compound class and/or individual compound, thereby supporting the occurrence of idiosyncratic toxicity in some patients.

Gene Expression Changes Induced by PPAR Gamma Agonists in Animal and Human Liver.

Thiazolidinediones are a class of Peroxisome Proliferator Activated Receptor γ (PPARγ) agonists that reduce insulin resistance in type 2 diabetic patients. Although no detectable hepatic toxicity has been evidenced in animal studies during preclinical trials, these molecules have nevertheless induced hepatic adverse effects in some treated patients. The mechanism(s) of hepatotoxicity remains equivocal. Several studies have been conducted using PCR analysis and microarray technology to identify possible target genes and here we review the data obtained from various in vivo and in vitro experimental models. Although PPARγ is expressed at a much lower level in liver than in adipose tissue, PPARγ agonists exert various PPARγ-dependent effects in liver in addition to PPARγ-independent effects. Differences in effects are dependent on the choice of agonist and experimental conditions in rodent animal studies and in rodent and human liver cell cultures. These effects are much more pronounced in obese and diabetic liver. Moreover, our own recent studies have shown major interindividual variability in the response of primary human hepatocyte populations to troglitazone treatment, supporting the occurrence of hepatotoxicity in only some individuals.