Screening and risk management solutions for steroidal estrogens in surface and wastewater

Background: The European Commission Implementing Decision EU 2015/495 included three steroidal estrogens, namely 17α-ethinyl estradiol, 17β-estradiol, and estrone, in the so-called “watch list” of the EU Water Framework Directive (WFD). The monitoring of these compounds is difficult because the detection limits of the majority of the available analytical methods cannot achieve the very low target concentrations required to meet proposed environmental quality criteria. In 2014, a combined Science-Policy Interface/Chemical Monitoring of Emerging Pollutants project was launched to meet this monitoring challenge. The project involved 24 research organizations and environmental agencies from 12 different countries. Methods: Sixteen surface water (SW) and 17 wastewater (WW) samples were collected across Europe and analysed using five in vitro effect-based and three chemical analytical methods. A general description of the project and data evaluation is provided by Könemann and colleagues in the companion publication 2018. In our study, we compared bioanalytical and chemical analytical results with regard to their application for aquatic status assessment. Therefore we considered the potential to predict population-relevant risks for aquatic organisms and the specificity and sensitivity of the various methods used in both approaches. Finally, we tested and discussed the applicability and relevance of previously suggested effect-based trigger values (EBT). Results and discussion: Results of the risk assessment based on chemical analytical data correlated highly with estrogenic activities (expressed as 17β-estradiol equivalents (EEQ) determined using effect-based methods), demonstrating the ability of the bioassays to predict the mixture risk caused by steroidal estrogens. For about 15% of SW and 40% of WW samples detection limits of chemical-analytical methods were too high to allow a status assessment, while detection limits of all effect-based methods were below proposed EBT. This demonstrates that effect-based methods are suitable for status assessment of surface waters. The in vitro effect-based methods were quite specific for steroidal estrogens and highly sensitive, meaning they have a low probability to detect false positive or negative results. After testing of three alternative EBT proposals, we concluded to use preliminary 400 pg/L EEQ as screening EBT corresponding to the AA-EQS of E2. Further test specific refinements in the application of this value are not excluded. Conclusions: We conclude that water quality assessment can progress from a purely analytical approach to effect-based monitoring, from single substance to known and unknown mixture assessment and from in vitro screening to population-relevant risk assessment. Despite a few limitations, effect-based in vitro methods are recommendable for monitoring steroidal estrogens under the WFD because they, a) are able to sensitively quantify the activity of steroidal estrogens in all surface and wastewater samples, b) are able to detect the combined effect of estrogen mixtures including unknown chemicals with estrogen receptor activating properties, c) allow an ecotoxicological status assessment using EBT to calculate risk quotients. This approach is similar to the risk ratio used in regulatory environmental risk assessments, but allows for an integrated mixture assessment. Outlook: The results of this study support the introduction of a holistic approach for the regulation of chemicals in the aquatic environment under the EU WFD, as proposed recently by EU Water Directors. An ecotoxicological status assessment for one of the most relevant modes of action of endocrine disruption will allow authorities responsible for water quality assessment to focus available monitoring resources and to improve the ecological status of EU waterbodies.