Incorporation of metabolic enzymes to improve predictivity of reporter gene assay results for estrogenic and anti-androgenic activity.

Identification and monitoring of so-called endocrine-disrupting compounds has received ample attention; both the OECD and the United States Environmental Protection Agency (US EPA) have designed tiered testing approaches, involving in vitro bioassays to prioritize and partly replace traditional animal experiments. Since the estrogen (ER) and androgen (AR) receptor are frequent targets of endocrine disrupting chemicals, bioassays detecting interaction with these receptors have a high potential to be of use in risk assessment of endocrine active compounds. However, in many bioassays in vivo hepatic metabolism is not accounted for, which hampers extrapolation to the in vivo situation. In the present study, we have developed a metabolic module using rat liver S9 as an add-on to human cell-based reporter gene assays. The method was applied to reporter gene assays for detection of (anti-) estrogens and (anti-) androgens, but can be extended to cell-based reporter gene assays covering a variety of endpoints related to endocrine disruption.

Incorporation of a metabolizing system in biodetection assays for endocrine active substances.

The use of in vitro assays is important for the biodetection of endocrine active substances (EAS), reducing and replacing the in vivo studies required for regulatory assessment. However, this approach often fails to take into account the role of biotransformation on the activity of the test substances. A method incorporating an S9 metabolic system into the CALUX-reporter gene assays for estrogen receptor α- and anti-androgen receptor -mediated activities has been developed. Methoxychlor, which is known to exhibit increased estrogenic and anti-androgenic activities after biotransformation, was used to set up the method in ERa and anti-AR CALUX. For the anti-androgenic assay, stanozolol was used as a competing agonist not metabolized by S9. The method was first applied in both agonist and antagonist modes to methoxychlor and bisphenol A, as positive and negative controls, respectively. Then, benzo(a)pyrene and flutamide were also tested for their potential of bioactivation. Co-treatment with S9 successfully increased the ERα agonist and AR antagonist potency of methoxychlor; no change was observed for bisphenol A. Incubation with S9 also enhanced the anti-androgenic activity of flutamide. Interestingly, the metabolism of benzo(a)pyrene by the S9 resulted in an increased estrogen receptor-mediated transcriptional activation; any increase in the potency was only minor. It is likely that both enzyme kinetics and metabolite stability have influenced these effects, which would affect the composition of the final metabolite mixture. Together these results demonstrate the relevance of including biotransformation in in vitro bioassays for the detection of EAS.