Mixture effects of organic micropollutants present in water: towards the development of effect-based water quality...

In this study we propose for the first time an approach for the tentative derivation of effect-based water quality trigger values for an apical endpoint, the cytotoxicity measured by the bioluminescence inhibition in Vibrio fischeri. The trigger values were derived for the Australian Drinking Water Guideline and the Australian Guideline for Water Recycling as examples, but the algorithm can be adapted to any other set of guideline values. In the first step, a Quantitative Structure-Activity Relationship (QSAR) describing the 50% effect concentrations, EC50, was established using chemicals known to act according to the nonspecific mode of action of baseline toxicity. This QSAR described the effect of most of the chemicals in these guidelines satisfactorily, with the exception of antibiotics, which were more potent than predicted by the baseline toxicity QSAR. The mixture effect of 10-56 guideline chemicals mixed at various fixed concentration ratios (equipotent mixture ratios and ratios of the guideline values) was adequately described by concentration addition model of mixture toxicity. Ten water samples were then analysed and 5-64 regulated chemicals were detected (from a target list of over 200 chemicals). These detected chemicals were mixed in the ratios of concentrations detected and their mixture effect was predicted by concentration addition. Comparing the effect of these designed mixtures with the effect of the water samples, it became evident that less than 1% of effect could be explained by known chemicals, making it imperative to derive effect-based trigger values. The effect-based water quality trigger value, EBT-EC50, was calculated from the mixture effect concentration predicted for concentration-additive mixture effects of all chemicals in a given guideline divided by the sum of the guideline concentrations for individual components, and dividing by an extrapolation factor that accounts for the number of chemicals contained in the guidelines and for model uncertainties. While this concept was established using the example of Australian recycled water, it can be easily adapted to any other set of water quality guidelines for organic micropollutants. The cytotoxicity based trigger value cannot be used in isolation, it must be applied in conjunction with effect-based trigger values targeting critical specific modes of action such as estrogenicity or photosynthesis inhibition.