Complex mixtures of organic chemicals extracted from representative but not directly related environmental samples (wastewater, surface water, fish), food items (drinking water, fish, milk) and human blood were tested in 22 in vitro bioassays targeting pathways associated with neurodevelopmental and reproductive health. Extraction methods were optimized to extract common chemicals across matrices capturing both persistent and nonpersistent, neutral and charged organic chemicals─albeit with some bias toward more hydrophilic chemicals over highly hydrophobic chemicals. Most bioassay end points─except genotoxicity─were responsive, with strongest effects observed higher up the food chain in fish and humans. Experimental mixture effects of 24 chemicals quantified in these extracts conformed to the mixture prediction model of concentration addition in the six most responsive bioassays, namely neurite outgrowth inhibition, mitochondrial membrane potential inhibition, transthyretin protein binding, sodium-iodide symporter inhibition and androgen receptor antagonism. Designed mixtures explained little of total bioactivity, indicating that many of the thousands of unannotated molecular features detected by nontarget analysis contribute to mixture effects. Preliminary effect-based trigger (EBT) values defined for water and food by extrapolation from safe levels of individual chemicals indicate no immediate health risks at these average contamination levels. The high complexity and multivalent bioactivity of these mixtures on neurodevelopmental and reproductive pathways necessitate further toxicological scrutiny.

Pharmaceuticals are increasingly recognized as contaminants of concern in aquatic environments. Sitagliptin, an antidiabetic drug that carries a C-CF3 group, which is a precursor of the persistent trifluoroacetic acid, is excreted largely unmetabolized and inefficiently removed in wastewater treatment plants, leading to its widespread detection in surface waters. The hyporheic zone, a region between surface water and groundwater, serves as a natural bioreactor with high microbial activity and diverse redox conditions, offering the potential for sitagliptin attenuation. This study explored the biotransformation of sitagliptin in hyporheic sediments under varying redox conditions through batch experiments and field observations. We showed that batch experiments can complement field observations to capture both mechanistic insights and their environmental relevance. Batch experiments revealed amide hydrolysis and N-acetylation of sitagliptin under anoxic conditions, with subsequent deamination and oxidation of transformation products under oxic conditions. Metagenome-resolved metaproteomics suggested Pseudomonas asiatica as a key player in the oxic transformation. Field analysis of pore water samples identified up to 6.47 µg L⁻¹ sitagliptin and ten transformation products with concentrations of up to 4.82 µg L⁻¹ . Amide hydrolysis products were the most abundant transformation products and preferentially formed under anoxic conditions. All investigated transformation products exhibited lower cytotoxicity and oxidative stress responses than sitagliptin in in vitro bioassays, highlighting the toxicity reducing potential of the hyporheic zone. By identifying conditions that promote sitagliptin transformation and characterizing its transformation products toxicologically, our work provides parameters for enhanced sitagliptin removal in aquatic environments and improved risk assessment of fluorinated trace organic contaminants.

Chemical contaminants are widely dispersed in the environment in all its dimensions, posing significant public health problems. Comprehensive knowledge of these stressors is a prerequisite for assessing the associated risk and implementing public policy measures to reduce the level of population exposure. Nontargeted suspect screening approaches broaden the knowledge of the chemical human exposome. We developed and used a suspect screening method based on large spectral libraries. Chemical profiling was based on a combined LC- and GC-HRMS approach. The methodology was applied to 16 samples spanning the environment, food, and health continuum. Using a combination of matching and scoring data, a total of 547 compounds were likely identified, from which the chemical structure of 63 molecules was confirmed to the highest level of certainty. Wastewater, and more generally environmental samples, had the highest number of chemicals detected, while fish samples had a lower number. Pharmaceuticals, pesticides, and personal care product-related compounds were found to be the most common compounds in and between the extracts, particularly in water and serum samples. Many natural and endogenous compounds were consistently annotated in the samples submitted for analysis, regardless of the compartment investigated.

A critical understanding of how pests interact with active ingredients is essential for the development of new insect control solutions to maintain crop quality and quantity by reducing insect damage. Absorption of insecticides into insect bodies of targeted pest species is the first critical step that confounds the efficacy of insecticides. This study investigated how different feeding behaviour of two pests, Myzus persicae and Spodoptera littoralis, affects the absorption, metabolism, and excretion (AME) of seven insecticidally inactive test compounds. A feeding contact assay for the chewing pest (Lepidopteran larvae) and an oral ingestion assay for the sucking pest (aphids) was used to investigate the AME of test compounds with agrochemical-like structural motifs. The standardized assays comprised of an exposure period with treated diet and a subsequent depuration period with untreated diet. The results showed that S. littoralis larvae differed from M. persicae in their compound quantities absorbed into the insect body and in their excretion products at the end of the exposure or depuration periods. We suggest that this is caused by their different ingestion types and rates resulting in different absorption and excretion quantities. Further, we found differences in the metabolism (timing and biotransformation pathways) of compounds between both species. Notably, certain compounds remained detectable in both pests after the depuration period, suggesting compound and species-specific metabolism and excretion. Our results highlight the complex interplay between feeding biology of insects, in particular the critical role of excretion products, and the exposure to different compounds that lead to species-specific AME.

Human-impacted rivers often contain a complex mixture of organic micropollutants, including pesticides, pharmaceuticals and industrial compounds, along with their transformation products. Combining chemical target analysis for exposure with in vitro bioassays for effect assessment offers a holistic view of water quality. This study targeted the River Elbe in Central Europe, known for its anthropogenic pollution exposure, to obtain an inventory of micropollutant contamination during base flow and to identify hotspots of contamination. We identified tributaries as sources of chemicals activating the aryl hydrocarbon receptor quantified with the AhR-CALUX assay, including historically contaminated tributaries and a newly identified Czech tributary. Increased neurotoxicity, detected by differentiated SH-SY5Y neurons' cytotoxicity and shortened neurite length, was noted in some Czech tributaries. A hotspot for chemicals activating the oxidative stress response in the AREc32 assay was found in the middle Elbe in Germany. An increase in oxidative stress inducing chemicals was observed in the lower Elbe. While effect-based trigger values (EBT) for oxidative stress response, xenobiotic metabolism and neurotoxicity were not exceeded, estrogenicity levels surpassed the EBT in 14% of surface water samples, posing a potential threat to fish reproduction. Target analysis of 713 chemicals resulted in the quantification of 487 micropollutants, of which 133 were active in at least one bioassay. Despite this large number of bioactive quantified chemicals, the mixture effects predicted by the concentrations of the quantified bioactive chemicals and their relative effect potency explained only 0.002-1.2% of the effects observed in the surface water extracts, highlighting a significant unknown fraction in the chemical mixtures. This case study established a baseline for understanding pollution dynamics and spatial variations in the Elbe River, offering a comprehensive view of potential chemical effects in the water and guiding further water quality monitoring in European rivers.

Revision of the REACH chemical regulation should enable more realistic understanding and management.

MLinvitroTox is an automated Python pipeline developed for high-throughput hazard-driven prioritization of toxicologically relevant signals detected in complex environmental samples through high-resolution tandem mass spectrometry (HRMS/MS). MLinvitroTox is a machine learning (ML) framework comprising 490 independent XGBoost classifiers trained on molecular fingerprints from chemical structures and target-specific endpoints from the ToxCast/Tox21 invitroDBv4.1 database. For each analyzed HRMS feature, MLinvitroTox generates a 490-bit bioactivity fingerprint used as a basis for prioritization, focusing the time-consuming molecular identification efforts on features most likely to cause adverse effects. The practical advantages of MLinvitroTox are demonstrated for groundwater HRMS data. Among the 874 features for which molecular fingerprints were derived from spectra, including 630 nontargets, 185 spectral matches, and 59 targets, around 4% of the feature/endpoint relationship pairs were predicted to be active. Cross-checking the predictions for targets and spectral matches with invitroDB data confirmed the bioactivity of 120 active and 6791 nonactive pairs while mislabeling 88 active and 56 non-active relationships. By filtering according to bioactivity probability, endpoint scores, and similarity to the training data, the number of potentially toxic features was reduced by at least one order of magnitude. This refinement makes the analytical confirmation of the toxicologically most relevant features feasible, offering significant benefits for cost-efficient chemical risk assessment.Scientific Contribution:In contrast to the classical ML-based approaches for toxicity prediction, MLinvitroTox predicts bioactivity for HRMS features (i.e., distinct m/z signals) based on MS2 fragmentation spectra rather than the chemical structures from the identified features. While the original proof of concept study was accompanied by the release of a MLinvitroTox v1 KNIME workflow, in this study, we release a Python MLinvitroTox v2 package, which, in addition to automation, expands functionality to include predicting toxicity from structures, cleaning up and generating chemical fingerprints, customizing models, and retraining on custom data. Furthermore, as a result of improvements in bioactivity data processing, realized in the concurrently released pytcpl Python package for the custom processing of invitroDBv4.1 input data used for training MLinvitroTox, the current release introduces enhancements in model accuracy, coverage of biological mechanistic targets, and overall interpretability.

Fish acute toxicity testing is used to inform environmental hazard assessment of chemicals. In silico and in vitro approaches have the potential to reduce the number of fish used in testing and increase the efficiency of generating data for assessing ecological hazards. Here, two in vitro bioactivity assays were adapted for use in high-throughput chemical screening. First, a miniaturized version of the Organisation for Economic Co-operation and Development (OECD) test guideline 249 plate reader-based acute toxicity assay in RTgill-W1 cells was developed. Second, the Cell Painting (CP) assay was adapted for use in RTgill-W1 cells along with an imaging-based cell viability assay. Then, 225 chemicals were tested in each assay. Potencies and bioactivity calls from the plate reader and imaging-based cell viability assays were comparable. The CP assay was more sensitive than either cell viability assay in that it detected a larger number of chemicals as bioactive, and phenotype altering concentrations (PACs) were lower than concentrations that decreased cell viability. An in vitro disposition (IVD) model that accounted for sorption of chemicals to plastic and cells over time was applied to predict freely dissolved PACs and compared with in vivo fish toxicity data. Adjustment of PACs using IVD modeling improved concordance of in vitro bioactivity and in vivo toxicity data. For the 65 chemicals where comparison of in vitro and in vivo values was possible, 59% of adjusted in vitro PACs were within one order of magnitude of in vivo toxicity lethal concentrations for 50% of test organisms. In vitro PACs were protective for 73% of chemicals. This combination of in vitro and in silico approaches has the potential to reduce or replace the use of fish for in vivo toxicity testing.

High-throughput cell-based bioassays can fulfill the growing need to assess the hazards and modes of toxic action (MOA) of ionic liquids (ILs). Although nominal concentrations (Cnom) are typically used in an in vitro bioassay, freely dissolved concentrations (Cfree) are considered a more accurate dose metric because they account for chemical partitioning processes and are informative about MOA. We determined the Cfree of IL cations in AREc32 and AhR-CALUX assays using both mass balance model (MBM) prediction and experimental quantification. Partition coefficients between membrane lipid-water (Kmw), serum albumin-water (Kalbumin/w), and cell-water (Kcell/w) as well as potential confounding factors (binding to a test plate and micelle formation) were determined to improve the MBM prediction. IL cations showed a higher affinity for both cell lines than that predicted by the MBM based on Kmw and Kalbumin/w. Their affinity for the AhR-CALUX cells was more than 1 order of magnitude higher than for the AREc32, signifying cell line-specific affinity. The MBM with an experimental Kcell/w accurately predicted Cfree. Evaluating cytotoxicity based on Cfree eliminated the leveling off of toxicity observed for hydrophobic IL cations (side chain cutoff), suggesting that Cnom underestimates the effects of compounds with high affinity for the assay medium. Cell membrane concentrations calculated from Cfree using Kmw were compared to the critical membrane burden to identify whether IL cations act as baseline toxicants. The IL cations carrying 16 carbons in the chain in the AREc32 assay and most of the IL cations in the AhR-CALUX assay were classified as excess toxicants. However, since the reasons for the deviation of experimental Kcell/w from MBM prediction remain unexplained, it is uncertain whether the cell membrane concentrations can be well predicted from Kmw used in this study. Therefore, future studies should aim to uncover the underlying causes of differing cell affinities observed across cell lines and model predictions.

Bisphenol A (BPA) is a well-known endocrine disruptor linked to numerous adverse health outcomes and was, therefore, banned in food-contact materials in the European Union. Numerous alternatives are now in commerce, but their health hazards are often inadequately addressed. This study compared BPA and 26 alternatives in six in vitro bioassays for cytotoxicity, endocrine disruption, xenobiotic metabolism, adaptive stress responses, mitochondrial toxicity, and neurotoxicity. We developed a cumulative specificity ratio score that integrates the degree of specific activation and overall toxicological activity across a test battery, enabling direct comparison of BPA with its alternatives. Several alternatives with close structural resemblance showed similar or stronger activation of the estrogen receptor α (ERα) than BPA. The lack of estrogenicity for several BPA alternatives, e.g., 4-(4-phenylmethoxyphenyl)sulfonylphenol (BPS-MPE), was accompanied by a shift toward peroxisome proliferator-activated receptor γ (PPARγ) activation, a receptor that is not relevant for BPA itself. Some alternatives additionally inhibited mitochondrial functions and caused neurotoxicity. Simulated phase I metabolism reduced the cytotoxicity of all alternatives except for methyl bis(4-hydroxyphenyl)acetate (Bz) and 4-[[4-(allyloxy)phenyl]sulfonyl]phenol (BPS-MAE), while estrogenic activity remained unchanged or decreased. This study demonstrates the utility of bioassays for rapid hazard assessment and comparative evaluation, suggesting that many BPA alternatives are regrettable substitutes, although 2,2,4,4-tetramethyl-1,3-cyclobutanediol (TMCD) is a potentially more benign alternative.

Many micropollutants occur in aquatic environments due to anthropogenic activities leading to environmental risks to ecosystems. However, their combined effects remain underexplored, particularly in developing countries. A panel of in vitro bioassays indicative of xenobiotic metabolism (AhR, PPARγ and PXR), estrogenic activity via estrogen receptor alpha (ERα) activation, glucocorticogenic activity (GR), developmental neurotoxicity (SH-SY5Y), oxidative stress response (AREc32), mitochondrial membrane potential inhibition and cytotoxicity was used to investigate water quality in the Guandu River basin, a major source water catchment area in Rio de Janeiro State, Southeastern Brazil. The contribution of dissolved and suspended particulate matter (SPM) phases was considered, and the environmental relevance of effects was assessed against effect-based trigger (EBT) values in dry and wet seasons. Increased ERα activity was the most concerning endpoint and primary risk driver. SPM contributed more to cytotoxicity than to specific effects. Seasonal influences such as changing climate and hydrological conditions only had a minor role in explaining variation in the main river. In total, 269 chemicals were detected with target analysis, mostly pharmaceuticals and pesticides, but they explained only up to 1% of measured effects in AhR, PPARγ, SH-SY5Y and AREc32 assays. 1,2-Benzisothiazolinone, Mebendazole, Diuron, Benzothiazole, Diclofenac and 2-Benzothiazolesulfonic acid were identified as relevant toxicity drivers for neurotoxicity, xenobiotic metabolism and oxidative stress response. Attention is drawn to the pollution level in the investigated waters from a tropical developing country, which has similar water pollution levels as observed worldwide.

The partition dynamics of organic micropollutants between water and suspended particulate matter (SPM) in riverine ecosystems differs between dry and wet weather, as demonstrated at two sites at the Ammer River, Germany. One site was impacted by a wastewater treatment plant (WWTP) and the other by runoff of a mixed agricultural/urban area. Liquid and gas chromatography coupled to high-resolution mass spectrometry were used to quantify 415 organic chemicals, and their mixture effects were characterized with three in vitro bioassays indicative of the activation of the aryl hydrocarbon (AhR) and peroxisome proliferator-activated (PPARγ) receptors and the oxidative stress response. During wet weather, the total chemical concentrations and bioactivities in the water increased, but the concentrations in SPM did not change. As SPM levels increased, the SPM-bound chemicals contributed 6-16% to the overall concentrations in the water column during wet weather but only 0.1-0.9% during dry weather. The mixture effects were more strongly associated with SPM under wet conditions, particularly for AhR activity, where SPM accounted for over 90% of the observed effects. The AhR activity may therefore serve as an indicator for assessing the risks of SPM-related pollution in rivers. The high SPM-bound mixtures' activation of AhR and oxidative stress response during rain were primarily caused by polycyclic aromatic hydrocarbons, indicating a major contribution of road runoff.

The commitment to develop a roadmap for phasing out the use of animals for chemical safety assessments was part of the European Commission’s response to the European Citizens’ Initiative “Save Cruelty-Free Cosmetics – Commit to a Europe Without Animal Testing”. The roadmap aims to outline milestones and specific actions to be implemented in the short to long term to ultimately phase out animal testing for chemical safety assessments. To advance this goal and help define a structure of the roadmap, a multi-stakeholder roundtable workshop was organized by five animal protection non-governmental organizations in June 2024. The roundtable aimed to explore and define key elements and organizational structures for shaping the roadmap and identify pathways to facilitate the transition to a non-animal testing regulatory framework. Participants discussed a range of critical issues such as revising legislation and guidance, facilitating validation/qualification and regulatory acceptance, strengthening coordination, providing education and training in non-animal approaches, transparency and accessibility to data, establishing metrics to measure progress, and securing funding. The importance of a multi-faceted approach integrating scientific, regulatory, policy, ethical, societal, and practical dimensions was emphasized, along with the critical role of transdisciplinary collaboration and combining diverse knowledge, ideas, and technologies to achieve optimal outcomes. This report summarizes the main findings and discussion points and provides concrete recommendations. These are intended to facilitate the Commission’s work to develop the roadmap and may serve as a valuable resource for similar initiatives worldwide.

Pharmaceuticals are increasingly recognized as contaminants of concern in aquatic environments. Sitagliptin, an antidiabetic drug that carries a trifluoromethyl group, which is a precursor of the persistent trifluoroacetic acid, is excreted largely unmetabolized and inefficiently removed in wastewater treatment plants, leading to its widespread detection in surface waters. The hyporheic zone — a region between surface water and groundwater — serves as a natural bioreactor with high microbial activity and diverse redox conditions, offering the potential for sitagliptin attenuation. This study explored the biotransformation of sitagliptin in hyporheic sediments under varying redox conditions through batch experiments and field observations. Furthermore, we showed that batch experiments can complement field observations to capture both mechanistic insights and their environmental relevance. Batch experiments revealed amide hydrolysis and N-acetylation of sitagliptin under anoxic conditions, with subsequent deamination and oxidation of transformation products under oxic conditions. Metagenome-resolved metaproteomics suggested Pseudomonas asiatica as a key player in the oxic transformation. Field analysis of pore water samples identified up to 6.47 µg L⁻¹ sitagliptin and ten transformation products with concentrations of up to 4.82 µg L⁻¹. Amide hydrolysis products were the most abundant transformation products and preferentially formed under anoxic conditions. All investigated transformation products exhibited lower cytotoxicity and oxidative stress response than sitagliptin in in vitro bioassays, highlighting the detoxification potential of the hyporheic zone. By identifying conditions that promote sitagliptin transformation and characterizing its transformation products toxicologically, our work provides parameters for enhanced sitagliptin removal in aquatic environments and improved risk assessment of fluorinated trace organic contaminants. ![Figure][1] Graphical abstract ### Competing Interest Statement The authors have declared no competing interest. Deutsche Forschungsgemeinschaft, GRK 2032/2 [1]: pending:yes

The risk assessment of chemicals relies on multiple tools to quantify the ecological responses of ecosystems to existing chemical pollution. These tools are broadly categorized into three major groups: toxic pressure assessments, bioassays, and ecological monitoring. Here, we examine the strengths and limitations of these approaches, their current level of implementation for freshwater ecosystems across Europe, and their ability to evaluate the impacts of chemicals under field conditions. Additionally, we analyze the correspondence between results obtained from these tools when applied to a monitoring dataset from German streams. Our evaluation showed that no single tool can perfectly characterize the environmental impacts of chemical mixtures. However, each provides distinct lines of evidence, enabling the identification of chemicals driving ecological risks and the biological endpoints most likely to be affected, with ecological monitoring tools having the potential to show long-term ecosystem impairment. Finally, we propose recommendations to better understand the discrepancies between the outcomes of different methods and explore their potential integration into a unified water quality evaluation framework.

The eco-exposome represents the totality of chemicals present in an organism. To understand how internal exposure of fish relates to chemicals originating from external media (water, sediment), we conducted a 21-d caging study using fathead minnow (Pimephales promelas, FHM) as model species. Four sites in/at Lake Superior were chosen that reflect sources of two broad groups of environmental contaminants: Two pond sites with a legacy contamination by persistent organic pollutants (POPs), one site close to a wastewater treatment plant (WWTP) outlet showing more recent and regularly discharged compounds, and one creek expected to show a mixed contamination from both compound groups. We determined total water concentrations, freely dissolved concentrations in sediment pore water, and the FHM's body burden of organic micropollutants after 2 and 21 d of exposure. Of the 456 target compounds analyzed in FHM, 123 were quantified in water, 165 in sediment and 100 in FHM tissue samples. Chemical profiles and concentrations at the different study locations varied according to their classification as legacy or recently contaminated sites, with the site impacted by the WWTP showing the highest overall concentrations. Only 77 substances quantified in FHM were also detected in water and/or sediment and after applying additional quality and consistency measures, 37 of the 100 substances detected in FHM could not be directly linked to water and/or sediment. Therefore, chemical concentrations in water and sediment cannot simply predict the eco-exposome in fish, which underscores the need of body burden analysis to comprehensively understand an organism's exposure.

One of the primary criteria for a suitable drug biomarker for wastewater-based epidemiology (WBE) is having a unique source representing human metabolism. For WBE studies, this means it is important to identify and monitor metabolites rather than parent drugs, to capture consumption of drugs and not fractions that could be directly disposed. In this study, a high-throughput workflow based on a human liver S9 fraction in vitro metabolism assay was developed to identify human transformation products of new chemicals, using α-pyrrolidino-2-phenylacetophenone (α-D2PV) as a case study. Analysis by liquid chromatography coupled to high resolution mass spectrometry identified four metabolites. Subsequently, a targeted liquid chromatography – tandem mass spectrometry method was developed for their analysis in wastewater samples collected from a music festival in Australia. The successful application of this workflow opens the door for future work to better understand the metabolism of chemicals and their detection and application for wastewater-based epidemiology.

Human biomonitoring studies typically capture only a small and unknown fraction of the entire chemical universe. We combined chemical analysis with a high-throughput in vitro assay for neurotoxicity to capture complex mixtures of organic chemicals in blood. Plasma samples of 624 pregnant women from the German LiNA cohort were extracted with a nonselective extraction method for organic chemicals. 294 of >1000 target analytes were detected and quantified. Many of the detected chemicals as well as the whole extracts interfered with neurite development. Experimental testing of simulated complex mixtures of detected chemicals in the neurotoxicity assay confirmed additive mixture effects at concentrations less than individual chemicals’ effect thresholds. The use of high-throughput target screening combined with bioassays has the potential to improve human biomonitoring and provide a new approach to including mixture effects in epidemiological studies.