Direct comparison of the hepatoma cell lines HepG2 and HepaRG has previously been performed by only evaluating a limited set of genes or proteins. In this study, we examined the whole-genome gene expression of both cell lines before and after exposure to the genotoxic (GTX) carcinogens aflatoxin B1 and benzo[a]pyrene and the nongenotoxic (NGTX) carcinogens cyclosporin A, 17beta-estradiol, and 2,3,7,8-tetrachlorodibenzo-para-dioxin for 12 and 48 h. Before exposure, this analysis revealed an extensive network of genes and pathways, which were regulated differentially for each cell line. The comparison of the basal gene expression between HepG2, HepaRG, primary human hepatocytes (PHH), and liver clearly showed that HepaRG resembles PHH and liver the most. After exposure to the GTX and NGTX carcinogens, for both cell lines, common pathways were found that are important in carcinogenesis, for example, cell cycle regulation and apoptosis. However, also clear differences between exposed HepG2 and HepaRG were observed, and these are related to common metabolic processes, immune response, and transcription processes. Furthermore, HepG2 performs better in discriminating between GTX and NGTX carcinogens. In conclusion, these results have shown that HepaRG is a more suited in vitro liver model for biological interpretations of the effects of exposure to chemicals, whereas HepG2 is a more promising in vitro liver model for classification studies using the toxicogenomics approach. Although, it should be noted that only five carcinogens were used in this study.

Unexpected hepatotoxicity is one of the major reasons of drugs failing in clinical trials. This emphasizes the need for new screening methods that address toxicological hazards early in the drug discovery process. Here, proteomics techniques were used to gain further insight into the mechanistic processes of the hepatotoxic compounds. Drug-induced hepatotoxicity is mainly divided in hepatic steatosis, cholestasis, or necrosis. For each class, a compound was selected, respectively amiodarone, cyclosporin A, and acetaminophen. The changes in protein expressions in HepG2, after exposure to these test compounds, were studied using quantitative two-dimensional differential gel electrophoresis. Identification of differentially expressed proteins was performed by Maldi-TOF/TOF MS and liquid chromatography-tandem mass spectrometry. In this study, 254 differentially expressed protein spots were detected in a two-dimensional proteome map from which 86 were identified, showing that the proteome of HepG2 cells is responsive to hepatotoxic compounds. cyclosporin A treatment was responsible for most differentially expressed proteins and could be discriminated in the hierarchical clustering analysis. The identified differential proteins show that cyclosporin A may induce endoplasmic reticulum (ER) stress and disturbs the ER-Golgi transport, with an altered vesicle-mediated transport and protein secretion as result. Moreover, the differential protein pattern seen after cyclosporin A treatment can be related to cholestatic mechanisms. Therefore, our findings indicate that the HepG2 in vitro cell system has distinctive characteristics enabling the assessment of cholestatic properties of novel compounds at an early stage of drug discovery.

The lack of accurate in vitro assays for predicting in vivo toxicity of chemicals together with new legislations demanding replacement and reduction of animal testing has triggered the development of alternative methods. This study aimed at developing a transcriptomics-based in vitro prediction assay for in vivo genotoxicity. Transcriptomics changes induced in the human liver cell line HepG2 by 34 compounds after treatment for 12, 24, and 48 h were used for the selection of gene-sets that are capable of discriminating between in vivo genotoxins (GTX) and in vivo nongenotoxins (NGTX). By combining transcriptomics with publicly available results for these chemicals from standard in vitro genotoxicity studies, we developed several prediction models. These models were validated by using an additional set of 28 chemicals. The best prediction was achieved after stratification of chemicals according to results from the Ames bacterial gene mutation assay prior to transcriptomics evaluation after 24h of treatment. A total of 33 genes were selected for discriminating GTX from NGTX for Ames-positive chemicals and 22 for Ames-negative chemicals. Overall, this method resulted in 89% accuracy and 91% specificity, thereby clearly outperforming the standard in vitro test battery. Transcription factor network analysis revealed HNF3a, HNF4a, HNF6, androgen receptor, and SP1 as main factors regulating the expression of classifiers for Ames-positive chemicals. Thus, the classical bacterial gene mutation assay in combination with in vitro transcriptomics in HepG2 is proposed as an upgraded in vitro approach for predicting in vivo genotoxicity of chemicals holding a great promise for reducing animal experimentations on genotoxicity.

Chemical carcinogens may cause a multitude of effects inside cells, thereby affecting transcript levels of genes by direct activation of transcription factors (TF) or indirectly through the formation of DNA damage. As the temporal profiles of these responses may be profoundly different, examining time-dependent changes may provide new insights in TF networks related to cellular responses to chemical carcinogens. Therefore, we investigated in human hepatoma cells gene expression changes caused by benzo[a]pyrene at 12 time points after exposure, in relation to DNA adduct and cell cycle. Temporal profiles for functional gene sets demonstrate both early and late effects in up- and downregulation of relevant gene sets involved in cell cycle, apoptosis, DNA repair, and metabolism of amino acids and lipids. Many significant transcription regulation networks appeared to be around TF that are proto-oncogenes or tumor suppressor genes. The time series analysis tool Short Time-series Expression Miner (STEM) was used to identify time-dependent correlation of pathways, gene sets, TF networks, and biological parameters. Most correlations are with DNA adduct levels, which is an early response, and less with the later responses on G1 and S phase cells. The majority of the modulated genes in the Reactome pathways can be regulated by several of these TF, e.g., 73% by nuclear factor-kappa B and 34-42% by c-MYC, SRF, AP1, and E2F1. All these TF can also regulate one or more of the others. Our data indicate that a complex network of a few TF is responsible for the majority of the transcriptional changes induced by BaP. This network hardly changes over time, despite that the transcriptional profiles clearly alter, suggesting that also other regulatory mechanisms are involved.