Estrogenic activity of ternary UV filter mixtures in fish (Pimephales promelas) — An analysis with nonlinear isobolograms

Numerous estrogenic compounds are present in aquatic environments, but currently it is not well understood how compounds that differ in maxima and slope of their individual dose–response curves contribute to the overall mixture effect. In order to better understand such interactions we investigated 3 commonly used UV filters, for their estrogenic mixture activity and analysed their joint effects by using the concentration addition (CA) concept. Thereby, we extended the method of isoboles for analysis of 3 compounds that differ in maxima and slopes of their dose–response curves. 3-Benzylidene camphor (3BC), benzophenone-1 (BP1) and benzophenone-2 (BP2) are estrogenic in fish and act as pure- or partial estrogen receptor α agonists. First we exposed juvenile fathead minnows for 14 days to six concentrations of each UV filter alone to determine vitellogenin (VTG) induction curves, calculate equi-effective mixture concentrations and predict mixture effects. For 3BC, BP1 and BP2 significant VTG-induction occurred at 420, 2668, and 4715 μg/L, respectively. BP2 displayed a full dose–response curve, whereas 3BC and BP1 showed submaximal activity of 70 and 78%, respectively. Second, we exposed fish to 6 equi-effective mixtures (EC-NOEC, EC1, EC5, EC10, EC20, EC30) of these UV filters. Significant VTG-induction occurred at EC5 and higher. Submaximal activity of 67% as compared to the control EE2 (100 ng/L) was reached. The curves for the observed and predicted mixture activity agreed for mixture levels (EC10 to EC30), however, at EC-NOEC, EC1 and EC5, lower activity was observed than predicted by CA. Detailed isobolographic analysis indicate additivity at EC10 to EC30, and antagonism at low levels (EC-NOEC to EC5). Our data show for the first time, that for compounds with differences in maxima and slope, considerably more mixture combinations are additive than previously thought. This should be taken into account for hazard and risk assessment of UV filters and xenoestrogens.