Prioritizing sites and chemicals of concern in the large transboundary river using publically available high-throughput bioefects data

Recent environmental monitoring approaches aim to identify chemicals of concern based on the prediction of biological targets they are likely to impact. One rapid screening approach for prioritization of environmental chemistry datasets is to estimate relative risk of chemicals based on the exposure-activity ratios (EARs). EARs integrate chemical occurence and potency data using high-throughput screening (HTS) data from the Toxicity Forecaster (ToxCast) database. The aim of this study was to: i) prioritize the sites along the middle course of the large transboundary river in the South Eastern Europe (Sava River) using EARs, and ii) highlight chemicals likely contributing to the observed biological activity by combining EARs and empirical in-vitro tests of estrogenic (ER), androgenic (AR) and glucocorticoid (GR) activity. Water samples were collected monthly in winter 2018 at 13 sites along the whole Croatian section of the Sava River. A total of 34 water samples were analysed using liquid chromatography/time-of- flight mass spectrometry (UHPLC-QTOF-MS). Out of 549 analysed organic contaminants, 184 were detected in Sava samples. To prioritize sites we calculated the cumulative EAR for each site (EARcumulative - sum of all individual EARs for each chemical - molecular target combination), as well as total EAR for each chemical (sum of EARs for all molecular targets ; EARtotal), and the EAR for each chemical-molecular target pair (EARtarget). The five sites with the highest EARcumulative were either in close proximity to major cities (two sites) or downstream of large tributaries and/or in the areas with intensive agriculture (three sites). The highest EARtarget values were observed for the nuclear receptors (androgen, estrogen and glucocorticoid in particular). Consequently, chemicals that exert estrogenic, androgenic and/or glucocorticoid activity were identified as chemicals of concern (e.g., 17-beta estradiol, mometasone fuorate and flumethasone had the highest EARtotal). EAR predictions were supported by the results of the empirical in vitro testing, two sites with the highest in vitro biological activity were also among the highest ranked sites by EAR. Future work will use mechanistic HTS data and EARs to idenify adverse outcome pathways for aquatic keystone species such as decapod crustaceans.