Reproducible chemical-induced changes in gene expression profiles in human hepatoma HepaRG cells under various experimental conditions
Introduction
Toxicogenomics is becoming a valuable approach to predict toxicity of a new compound and/or its mechanism of toxicity. Microarray technologies are the most frequently used tools in such studies, as demonstrated by several recent publications. However, some problems have arisen regarding the reliability, reproducibility and cross-platform comparability of results from such studies, even when the same RNA samples and the same commercial microarrays are used (Irizarry et al., 2005a, Irizarry et al., 2005b, Larkin et al., 2005, Tan et al., 2003, Waring et al., 2004, Zhang et al., 2005). Indeed, when using different platforms, the number of up- and down-regulated genes has been shown to vary greatly between experiments. Nevertheless, continuous improvements are being made to the quality and conditions of use of microarrays and recent data have shown an increase in reproducibility of interlaboratory data, thus establishing the robustness of microarray technology (Dobbin et al., 2005, Guo et al., 2006, Kuo et al., 2006, Patterson et al., 2006, Petersen et al., 2005, Piper et al., 2002, Shi et al., 2006).
Even though most studies to date have been performed with reference RNA samples, a recent report has shown that a comparison of gene expression profiles from primary cultures of rat hepatocytes prepared in different laboratories is also very difficult (Beekman et al., 2006). These findings are of major concern since the liver is the main target organ for potentially toxic compounds. Moreover, data obtained in animals cannot be extrapolated with certainty to the human situation and therefore in vitro human cell models would appear to be the most appropriate alternatives for investigating xenobiotic hepatotoxicity in humans. Primary human hepatocytes and immortalized hepatocytes are widely used; however, both model systems have limitations. Primary human hepatocytes are considered to be the most pertinent model in spite of their scarcity and unpredictable availability, limited growth activity and life-span and early and variable phenotypic alterations in vitro (Guillouzo, 1998, Hewitt et al., 2007). Hepatocyte cell lines lack a variable and substantial set of liver-specific functions, especially the major cytochromes P450 (CYP) activities and consequently, are unsuitable for mimicking in vivo normal parenchymal cells.
A new human hepatoma cell line, named HepaRG, was recently shown to be capable of expressing most of the liver-specific functions, including the major CYPs involved in drug metabolism at levels comparable to those found in primary human hepatocytes (Aninat et al., 2006, Gripon et al., 2002). In this study, we attempted to determine whether HepaRG cells could represent a robust model for evaluating the toxic effects of chemicals by a transcriptomic approach. For this purpose, we used cells prepared and drug-exposed at different laboratory test sites and analyzed mRNA samples on two platforms using different generations of Agilent oligonucleotide microarrays.
As suggested by others (Beekman et al., 2006, McMillian et al., 2004, Reymann and Borlak, 2006, van Delft et al., 2005), a high chemical concentration causing limited cytotoxicity, if at all, has to be investigated in order to evaluate toxic effects together with an active transcriptional activity. We show here that an 8 mM phenobarbital (PB) concentration caused reproducible effects, characterized by more than one thousand constantly modulated genes, in spite of the use of widely different experimental conditions.
Section snippets
HepaRG cell cultures
The HepaRG cell line, derived from a liver tumor of a female patient suffering from hepatocarcinoma, was cultured as previously described (Gripon et al., 2002). These cells reach maximum differentiation when maintained for two weeks at confluence in the presence of 2% dimethyl sulfoxide (DMSO) and 5 × 10−5 M hydrocortisone hemisuccinate. They are composed of both hepatocyte-like and biliary-like cells (about 50% of each type) (Cerec et al., 2007). For these studies, HepaRG cells were initially
Cytotoxic-induced effects
Intracellular ATP content was comparable following a 20 h treatment with 8 mM PB, representing around 40–50% of control values, in all experiments with the exception of one in which the content dropped to around 22% (experiment B) (specific data not shown). Examination of PB-exposed HepaRG cell cultures under phase-contrast microscopy revealed some alterations in a fraction of hepatocyte-like and biliary epithelial-like cells; these morphological alterations were more pronounced in cultures from
Discussion
The emerging toxicogenomic technologies offer new opportunities in pharmaceutical drug discovery and development since they can represent a novel way to predict toxicity and to determine the mode of toxicity of new drugs. However, these technologies create many challenges, due in part to problems related to reproducibility, interpretation, understanding and sharing of data. Various parameters must be considered when comparing different sets of data, including the biological models, RNA
Acknowledgments
Carine Lambert was a recipient of a CIFRE contract. We thank the Ouest-Genopole transcriptomic platform (Rennes) for help with some experiments. This work was supported by Servier group and in part by an EEC contract (LIINTOP-STREP-037499). We are grateful to Drs Lydie Sparfel, Françoise Goldfain-Blanc, Nicolas Sajot and Anne Licznar for helpful comments; and Miss Anne Platel for technical help. We also wish to thank Dr. Wynne Ellis and Dr. Delphine Allorge for careful reading of the manuscript.
References (41)
- et al.
The human hepatoma HepaRG cells: a highly differentiated model for studies of liver metabolism and toxicity of xenobiotics
Chem. Biol. Interact.
(2007) - et al.
A new family of Cdc42 effector proteins, CEPs, function in fibroblast and epithelial cell shape changes
J. Biol. Chem.
(2001) - et al.
A gene expression signature for oxidant stress/reactive metabolites in rat liver
Biochem. Pharmacol.
(2004) - et al.
Gene expression profiling and differentiation assessment in primary human hepatocyte cultures, established hepatoma cell lines, and human liver tissues
Toxicol. Appl. Pharmacol.
(2007) - et al.
TRB3 protects cells against the growth inhibitory and cytotoxic effect of ATF4
Exp. Cell Res.
(2007) - et al.
Reproducibility of oligonucleotide microarray transcriptome analyses. An interlaboratory comparison using chemostat cultures of Saccharomyces cerevisiae
J. Biol. Chem.
(2002) - et al.
Effects of clofibric acid on mRNA expression profiles in primary cultures of rat, mouse and human hepatocytes
Toxicol. Appl. Pharmacol.
(2003) - et al.
Comparison of supervised clustering methods to discriminate genotoxic from non-genotoxic carcinogens by gene expression profiling
Mutat. Res.
(2005) - et al.
Microarray analysis of hepatotoxins in vitro reveals a correlation between gene expression profiles and mechanisms of toxicity
Toxicol. Lett.
(2001) - et al.
The role of the nuclear receptor CAR as a coordinate regulator of hepatic gene expression in defense against chemical toxicity
Arch. Biochem. Biophys.
(2003)