Gene expression profiling in Ishikawa cells: a fingerprint for estrogen active compounds.

Several anthropogenous and naturally occurring substances, referred to as estrogen active compounds (EACs), are able to interfere with hormone and in particular estrogen receptor signaling. EACs can either cause adverse health effects in humans and wildlife populations or have beneficial effects on estrogen-dependent diseases. The aim of this study was to examine global gene expression profiles in estrogen receptor (ER)-proficient Ishikawa plus and ER-deficient Ishikawa minus endometrial cancer cells treated with selected well-known EACs (diethylstilbestrol, genistein, zearalenone, resveratrol, bisphenol A and o,p'-DDT). We also investigated the effect of the pure antiestrogen ICI 182,780 (ICI) on the expression patterns caused by these compounds. Transcript levels were quantified 24 h after compound treatment using Illumina BeadChip Arrays. We identified 87 genes with similar expression changes in response to all EAC treatments in Ishikawa plus. ICI lowered the magnitude or reversed the expression of these genes, indicating ER dependent regulation. Apart from estrogenic gene regulation, bisphenol A, o,p'-DDT, zearalenone, genistein and resveratrol displayed similarities to ICI in their expression patterns, suggesting mixed estrogenic/antiestrogenic properties. In particular, the predominant antiestrogenic expression response of resveratrol could be clearly distinguished from the other test compounds, indicating a distinct mechanism of action. Divergent gene expression patterns of the phytoestrogens, as well as weaker estrogenic gene expression regulation determined for the anthropogenous chemicals bisphenol A and o,p'-DDT, warrants a careful assessment of potential detrimental and/or beneficial effects of EACs. The characteristic expression fingerprints and the identified subset of putative marker genes can be used for screening chemicals with an unknown mode of action and for predicting their potential to exert endocrine disrupting effects.

Genomic profiling uncovers a molecular pattern for toxicological characterization of mutagens and promutagens in vitro.

The aim of this study was to evaluate the suitability of global gene expression profiling for the characterization and identification of mutagens and promutagens in vitro. To enable detection of both mutagenic and promutagenic compounds, we cotreated HepG2 cells with a rat liver S9 fraction as metabolic activation system (MAS), supplementing the limited drug metabolic capability of HepG2 cells. Illumina BeadChip arrays were used to quantify gene expression changes after treatment with three well-known mutagenic, three promutagenic, as well as two non-genotoxic reference compounds for a period of 24 or 48 h. Statistical data analysis revealed 91 genes being most representative for the (pro-)genotoxic response. Several processes such as cellular differentiation and the complex interactive regulation of the stress and DNA damage response via the transcriptional modulators STAT1, SP1, and P53 were differentially regulated. The gene set evaluated was further used to predict the genotoxic characteristics of N-nitrosodiethylamine (DEN) after its metabolic activation. Although no clear response could be established in P53 activation experiments, DEN was classified correctly as nongenotoxic without S9 and genotoxic in the presence of the MAS by means of its transcriptomic pattern. Our data support that mechanistic profiling in vitro is a useful tool compared with single endpoint detections to predict genotoxicity.

Activation of P53 in HepG2 cells as surrogate to detect mutagens and promutagens in vitro.

The current genotoxicity tests of the standard in vitro battery, especially those using mammalian cells, are limited by their low specificity and highlight the importance of new in vitro tools. This study aimed to evaluate the suitability of HepG2 cells for assaying mutagens and promutagens. We determined P53 activity as surrogate genotoxicity endpoint in HepG2 cells. Our results revealed a significant P53-induction by actinomycin D, methyl methanesulfonate and etoposide. Prior to the investigation of promutagens we characterized HepG2 cells by analyzing the expression of 45 genes involved in xenobiotic metabolism and measuring the activity of selected Cytochrome-P450 (CYP) enzymes. We determined a limited metabolic capacity prompting us to employ a co-treatment with rat liver S9 as metabolic activation system (MAS) for promutagens. While cyclophosphamide showed an elevation of activated P53 in the presence of S9, 7,12-dimethylbenz[a]anthracene and aflatoxin B(1) responded without the MAS. Inhibition of cellular CYP3A4 or CYP1A/1B suppressed the aflatoxin B(1)- and dimethylbenz[a]anthracene-mediated P53 response, respectively, indicating that HepG2 cells are capable of metabolizing these compounds in a CYP1A/B/3A4-dependent manner. In summary, our results indicate that P53 activation in HepG2 cells combined with a MAS can be used for the identification of new (pro)genotoxicants.