Scientific Publications

With global restrictions on Bisphenol A (BPA), various BPA alternatives are increasingly found in ecosystems, raising concerns. This study focuses on the genotoxic potential of three emerging BPA alternatives — Bisphenol AF (BPAF), Bisphenol AP (BPAP), and Bisphenol PH (BPPH) — using an advanced in vitro 3D model system, spheroids, prepared from a Zebrafish (Danio rerio) liver cell line (ZFL). Their cytotoxicity was evaluated using the CellTiter-Glo® 2.0 assay, while their genotoxic potential was assessed using the comet assay, γH2AX assay, and toxicogenomic analysis. The BPA alternatives were more cytotoxic to ZFL spheroids than BPA. Non-cytotoxic concentrations caused transient DNA damage without a significant increase in DNA double-strand breaks (DSBs). The toxicogenomic analysis confirmed these findings, indicating activation of the TP53 DNA damage response pathway, the nucleotide excision repair (NER) and base excision repair (BER) mechanisms, likely in response to bulky DNA lesions and oxidative DNA damage. In addition, the gene expression analysis indicated the influence of the tested BPs on the endocrine system. Our results indicate that BPAF, BPAP and BPPH have considerable genotoxic potential and pose a significant ecotoxicological risk, underscoring the need for further investigation and careful consideration of these chemicals as BPA replacements. 

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.

This article reviews the application of multicriteria decision analysis (MCDA) in chemical alternatives assessment (CAA) and presents an overview of how the methodology has been applied within CAA. The study aimed to identify research that uses MCDA to identify the most harmful or least problematic chemicals and evaluate its current use in CAA. The study supports the Partnership for the Assessment of Risks from Chemicals (PARC) in developing a toolbox for safe and sustainable by design (SSbD). 520 studies were analysed, and 21 studies were included. Although MCDA in CAA is still emerging, it shows growth potential in decision analysis and chemical alternatives assessment. The reviewed studies cover various CAA applications and methodological approaches. Multiattribute utility theory (MAUT) is the most often used, followed by Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), ÉLimination Et Choix Traduisant la REalité (ELECTRE), and analytic hierarchy process (AHP). Experimental data and in silico data have been used with roughly equal frequency as input data. Group decision-making involving stakeholders with conflicting interests is rarely addressed, with parameter weighting and problem structuring usually handled by authors, sometimes with expert input. Another little discussed topic is the use of external normalisation of input data. In silico generated predictions on chemical alternatives’ properties come with varying degrees of uncertainty, remaining an issue in CAA with MCDA.

Chemical risk assessment can benefit from integrating informative biomarkers in human biomonitoring (HBM). Beyond exposure biomarkers, effect biomarkers inform on biological reactions in the body, potentially leading to adverse effects, while susceptibility biomarkers address inter-individual variability in exposure. DNA methylation of key genes shows promise as an effect biomarker but this epigenetic mark remains underexplored in the context of chemicals. Similarly, although some genetic polymorphisms are linked to increased chemical susceptibility, genetic biomarkers are rarely included in HBM. This mini-review highlights recent literature supporting the inclusion of genetic and epigenetic biomarkers in HBM. Subsequently, we elaborate on how Oxford Nanopore Technologies as sequencing method can efficiently measure these biomarkers simultaneously, even in non-invasive samples like saliva. Widely used in other fields, this experimental set-up could facilitate the design of large-population studies paving the way for a next generation risk assessment (NGRA) of chemicals.

Shinyscreen is an R package and Shiny-based web application designed for the exploration, visualization, and quality assessment of raw data from high resolution mass spectrometry instruments. Its versatile list-based approach supports the curation of data starting from either known or “suspected” compounds (compound list-based screening) or detected masses (mass list-based screening), making it adaptable to diverse analytical needs (target, suspect or non-target screening). Shinyscreen can be operated in multiple modes, including as an R package, an interactive command-line tool, a self-documented web GUI, or a network-deployable service. Shinyscreen has been applied in environmental research, database enrichment, and educational initiatives, showcasing its broad utility. Shinyscreen is available in GitLab (https://gitlab.com/uniluxembourg/lcsb/eci/shinyscreen ↗) under the Apache License 2.0. The repository contains detailed instructions for deployment and use. Additionally, a pre-configured Docker image, designed for seamless installation and operation is available, with instructions also provided in the main repository.

Children’s lead exposure in three mining-impacted residential areas (Črna, Mežica and Žerjav) was modelled using the Integrated Exposure Uptake Biokinetic Model for Lead in Children (IEUBK). Site-specific environmental and dietary Pb source values were determined for modelling. For the first time, dietary exposure from both market and local foods was studied in detail. Children (Group 1: 24–36 and Group 2: 36–48 months) geometric mean blood lead levels (BLLs) were predicted and lead uptake from multiple sources was quantified according to the different dietary exposure scenarios. Biomonitoring data were used for validation. Site-specific soil, house dust and local food Pb contents are higher than legislative and background levels, remaining a cause for concern. Drinking tap water concentrations and outdoor air contents were found in acceptable levels. The determined dietary exposures, ranging from 0.7 to 3.3 µg/kg bw/day, were above the benchmark dose level of 0.5 µg/kg bw/day for developmental neurotoxicity set for Pb in children, indicating a health concern. In general, the estimated BLLs matched reasonably well with the observed BLLs in the Črna and Mežica area for both age groups and in the Žerjav area for Group 2. For Group 1, in the Žerjav area, the output of the IEUBK model overestimated the actual BLLs of the children. For both groups, the primary exposure pathway in Žerjav is from soil/dust, ranging from 55.3 to 84.8%. In Črna and Mežica, soil/dust exposure ranged between 24.2 and 57.8% and between 26.4 and 61.7%, respectively, indicating that dietary exposure predominates when local foods are included. The results of our study also suggest that using the IEUBK default diet value would reduce the dietary exposure up to 25.2% in Črna, 24.2% in Mežica and 8.6% in Žerjav. One of the main findings is that a diet containing local foods can be an important source of lead in mining-contaminated areas.

This study investigates the immunomodulatory effects of BPA substitutes using in silico and in vitro approaches. A comprehensive in silico screening with Endocrine Disruptome of BPA and its 25 substitutes revealed that BPG, BPS-MAE, BPS-MPE, BPPH and PF201 have the highest potential to bind nuclear receptors. Based on the in silico ranking, 12 compounds were selected and tested in vitro to investigate their effects on THP-1 derived macrophages, Jurkat T cells and LCLs. The metabolic activity tests showed that BPA has IC₅₀ values of 99-182 µM, while the values for BPG, BPPH and BPP are 5-10 times lower. Comparison of the IC₅₀ values with the logP of the tested compounds showed that the BPA analogues affect cell viability in proportion to their lipophilicity (R² = 0.9185). The results of the cytokine release assays showed that the BPA substitutes generally stimulated the proinflammatory response in THP-1 macrophages in a statistically significant manner when exposed to the environmentally relevant concentrations and suppressed it in the micromolar range in all cell lines tested. Further analysis using RamosBlue reporter cells indicated that BPAP, BPG, BPP and BPPH alter the NF-κB/AP-1 signalling pathway. Moreover, BPAP, BPG, BPP, BPPH, BPS-MAE, BPS-MPE, BTUM and PF201 were predicted in silico to be TLR4-MyD88 inhibitors. The overall results of this study identify BPG, BPP, BPPH, BPZ and TCBPA as BPA analogues with the highest immunomodulatory potential. In addition, the immunomodulatory potential of the BPA alternatives Pergafast201 and BTUM was confirmed in vitro for the first time. This study emphasises the need for cautious evaluation of BPA substitutes due to their potential immunomodulatory effect, which could significantly affect public health.

New Approach Methodologies (NAMs) in Next-Generation Risk Assessment (NGRA), integrating toxicokinetics (TK) with toxicodynamics (TD), provides an accurate evaluation of combined chemical exposures. This study assesses pyrethroids, which pose regulatory challenges due to their widespread use and cumulative exposure risks. A tiered NGRA framework was compared with conventional risk assessment (RA) to evaluate regulatory applicability. In Tier 1, ToxCast data established gene and tissue bioactivity indicators, facilitating hypothesis-driven hazard identification. Tier 2 examined combined risk assessments, rejecting the hypothesis of the same mode of action and highlighting inconsistencies in in vitro data and NOAEL/ADI correlations. Tier 3 applied Margin of Exposure (MoE) analysis and TK modeling to realistic exposure estimations for risk assessment screening based on internal doses, identifying tissue-specific pathways as critical risk drivers. Tier 4 refined bioactivity indicators using TK approaches to improve the NAM-based effect assessment, including an in vitro vs. in vivo comparison, with coherent results based on interstitial concentrations, though intracellular estimations remained uncertain. Tier 5 confirmed that dietary exposure in healthy adults is close to but below levels of concern, with bioactivity MoE values remaining below concern thresholds, and in vivo MoEs within the standard safety factors. Nevertheless, the MoEs are insufficient for addressing the additional non-dietary exposure expected from other pyrethroid uses such as biocides or pharmaceuticals. Results demonstrate that NGRA with TK-NAM-based TD offers a nuanced, regulatory-relevant framework for risk assessment. The proposed approach integrates the information on individual pyrethroids using bioactivity indicators; and the re-assessment of regulatory toxicity studies to select organ-relevant NOAELs allowed an improved in vitro-in vivo comparison, demonstrating the capacity of bioactivity-based MoEs for combined exposure assessments. This tiered approach provides key insights for regulatory decision-making, establishing a robust model for evaluating pyrethroids and similar chemical classes.

In vitro transcriptomics holds promise for high-throughput, human-relevant data but is not yet integrated into regulatory decision-making due to the lack of standardized approaches. For genotoxicity assessment, transcriptomic biomarkers such as GENOMARK and TGx-DDI facilitate qualitative and quantitative analysis of complex in vitro transcriptomic datasets. However, advancing their use in quantitative testing requires standardized methods for deriving transcriptomic Points of Departure (tPoDs) and linking them to in vivo responses. Herein, we investigated different approaches to calculate tPoDs and applied in vitro to in vivo extrapolation to obtain administered equivalent doses (AEDs). Human HepaRG cells were exposed for 72 h to 10 known in vivo genotoxicants (glycidol, methyl methanesulfonate, nitrosodimethylamine, 4-nitroquinoline-N-oxide, aflatoxin B1, colchicine, cyclophosphamide, mitomycin C, ethyl methanesulfonate, and N-Nitroso-N-ethylurea) from the highest concentration that induces up to 50 % cytotoxicity through a range of lower concentrations. Gene expression data was generated using a customized version of the TempO-Seq® human S1500 + gene panel. The GENOMARK and TGx-DDI biomarkers produced genotoxic calls for all of these reference genotoxicants. Next, we performed benchmark concentration (BMC) modeling to generate both genotoxicity-specific biomarker (tPoD_biomarkers) and generic tPoDs (tPoD _S1500+). High-throughput toxicokinetic models estimated the human AEDs for these tPoDs, which were compared with (a) previously reported genotoxicity-specific AEDs from other New Approach Methodologies, and (b) in vivo PoDs from animal studies. We found that the generic AEDs were more conservative than genotoxicity-specific biomarker AEDs. For six of the nine genotoxicants, transcriptomic AEDs were lower than the in vivo PoDs; refined kinetic models may improve predictions. Overall, in vitro transcriptomic data in HepaRG cells provide protective estimates of in vivo genotoxic concentrations, consistent with other in vitro genotoxicity testing systems.

Chemicals in the EU are mainly regulated based on their intended use. Each legal framework consists of requirements and guidance for hazard- and risk assessment, along with the associated decision processes e.g., registration or authorisation of chemicals for market access in the EU. As a single chemical may have multiple uses, it may be assessed under more than one framework, potentially leading to different assessment outcomes. To address this, the European Commission has introduced the ‘one substance, one assessment’ approach as part of the Chemicals Strategy for Sustainability. The aims of the approach include streamlining risk assessment processes and reducing duplication of work in assessing the same chemical. This study aimed to map the scope of chemicals subject to assessment in multiple legal frameworks and to illustrate the importance of coordination and communication in chemical assessment processes. This was achieved by identifying chemicals that are either registered or have received specific approval for the EU market, and analysing their presence in different legal frameworks. Our findings showed that almost one-tenth of the substances identified were listed under more than one framework. However, there was a notable lack of coherent chemical identifiers available to accurately identify chemicals across the frameworks. Additionally, we identified the presence of phthalates, bisphenols and PFAS in EU frameworks to illustrate how a group-based approach to chemical assessment could be applied across different legal frameworks.

This work discusses reuse of chemical data across disciplines and the role of various data initiatives and projects including PARC, NORMAN-SLE, MassBank, WorldFAIR, PSDI and NFDI4Chem to facilitate increased data sharing. Improved machine-readable chemical data supports global research and interdisciplinary methodologies crucial for sustainable development and achievement of UNESCO’s Open Science priorities and the UN Sustainability Development Goals. Examples of success and ongoing approaches include integrating toxicology and chemical exposure data using ontologies, linking specialised chemical data collections with larger repositories such as PubChem, and developing IUPAC International Chemicals Identifier (InChI) extensions for nanomaterials and mixtures. National data infrastructure projects in the UK and Germany focus on digitising and standardising chemical research data management workflows, aiding scientists in data collection, storage, processing, analysis, disclosure, and reuse. These global initiatives aim to enhance chemical data interoperability to solve real-world problems, foster collaboration, and promote innovation while considering sustainable data resources beyond individual projects.

Nephrotoxicity caused by drug or chemical exposure involves complex mechanisms as well as a temporal integration of injury and repair responses in different nephron segments. Distinct cellular transcriptional programs regulate the time-dependent tissue injury and regeneration responses. Whole kidney transcriptome analysis cannot dissect the complex spatio-temporal injury and regeneration responses in the different nephron segments. Here, we used laser capture microdissection of formalin-fixed paraffin embedded sections followed by whole genome targeted RNA-sequencing-TempO-Seq and co-expression gene-network (module) analysis to determine the spatial–temporal responses in rat kidney glomeruli (GM), cortical proximal tubules (CPT) and outer-medulla proximal tubules (OMPT) comparison with whole kidney, after a single dose of the nephrotoxicant cisplatin. We demonstrate that cisplatin induced early onset of DNA damage in both CPT and OMPT, but not GM. Sustained DNA damage response was strongest in OMPT coinciding with OMPT specific inflammatory signaling, actin cytoskeletal remodeling and increased glycolytic metabolism with suppression of mitochondrial activity. Later responses reflected regeneration-related cell cycle pathway activation and ribosomal biogenesis in the injured OMPT regions. Activation of modules containing kidney injury biomarkers was strongest in OMPT, with OMPT Clu expression highly correlating with urinary clusterin biomarker measurements compared the correlation of Kim1. Our findings also showed that whole kidney responses were less sensitive than OMPT. In conclusion, our LCM-TempO-Seq method reveals a detailed spatial mechanistic understanding of renal injury/regeneration after nephrotoxicant exposure and identifies the most representative mechanism-based nephron segment specific renal injury biomarkers.

Exposure to respiratory sensitizers (RSs) is the leading cause of occupational asthma. Although the prospective identification of RSs is important, there currently exists no OECD-approved test guideline for this endpoint. The adverse outcome pathway for respiratory sensitization consists of key event (KE) 1: binding of the respiratory sensitizer to a protein, KE2: activation of lung epithelial cells, KE3: activation of dendritic cells, and KE4: T-cell response. Here, we focused on KE2 by investigating whether measuring this KE could contribute to prospectively identify respiratory sensitizers. To mimic real-life exposure, cells were exposed via the air. We used an air–liquid interface model comprising the human bronchial epithelial cell line Calu-3. Exposure to the RS piperazine (3, 18, and 100 mg/m3) dose-dependently increased IL-6 production at dose levels that did not affect the other parameters tested (barrier integrity, cell metabolism, cytotoxicity, and IL-8 production). IL-6 has been linked to asthma in humans. Exposure to the RS chloramine-T (30, 300, and 3000 ng/cm2) showed only minor effects on the parameters tested. Exposure to the RS 2,4-toluene diisocyanate (10, 33, and 100 mg/m3) at the highest dose level clearly affected all parameters tested. Disrupted barrier function has been linked to asthma. In conclusion, this study may possibly suggest that different respiratory sensitizers may differentially impact KE2. Further research is needed to elucidate this.

As part of its response to the European Citizens’ Initiative “Save Cruelty-Free Cosmetics – Commit to a Europe Without Animal Testing”, the European Commission committed to developing a roadmap to eliminate the use of animals for the safety testing of chemicals. Five animal protection non-govern­mental organizations arranged a roundtable in June 2024 to support this effort, bringing together experts from different fields. The roundtable aimed to identify important elements and organizational structures for shaping the roadmap and explored practical steps to move to a regulatory system that does not rely on animal testing. Discussions covered revising current chemicals legislation, accel­erating the acceptance of non-animal methods, creating EU databases to improve data sharing, strengthening cooperation and communication, building skills and expertise in non-animal methods, monitoring progress, and increasing funding. This report summarizes the main ideas and recommendations from the roundtable to help guide the Commission’s work on the roadmap.

The persistent nature of the environmental contaminants per- and polyfluoroalkyl substances (PFAS) has recently received considerable attention, particularly because of their adverse effects on immune system functionality in the context of vaccine responses to infectious diseases. Following COVID-19 vaccination, some studies have shown a significant negative correlation between serum PFAS concentrations and the humoral immune response to the SARS-CoV-2 spike protein vaccination. However, the influence of PFAS on the cell-mediated immune response to SARS-CoV-2 spike protein post-COVID-19 vaccination remains underexplored. In the present study, we investigated the impact of a human blood-relevant PFAS mixture, containing perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA), perfluorohexane sulfonate (PFHxS), perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), and perfluoroundecanoic acid (PFUnDA) on innate (monocytes and NK cells), cell-mediated (T cells) and B cells adaptive immune responses in COVID-19-vaccinated female and male healthy donors. Human peripheral blood mononuclear cells (PBMCs) were exposed to a mixture of the six PFAS at real life concentrations and subsequently stimulated with the SARS-CoV-2 spike peptide. We report a significant upregulation of IFNγ production in T and NK cells, particularly among male donors exposed to high concentrations of the PFAS mixture. Conversely, we observed a decrease in the total B-cell population, particularly among female donors. A significant reduction in the secretion of the pro-inflammatory chemokines MIP-1α (CCL3) and MIP-3α (CCL20) was observed at high PFAS mixture concentrations. Overall, these findings suggest that high PFAS exposure may differentially affect immune responses in a sex-specific manner, with a potential impact on vaccine efficacy.

Under evolutionary pressure, the kinematic and energetic characteristics of animal locomotion have been optimized for survival. We investigated the kinematics and energetic performance of zebrafish eleutheroembryo escape swims triggered by electrical stimuli in fluids of increasing viscosity. Eleutheroembryos exhibited a decrease in both tail movement frequency and swimming velocity in more viscous environments, while the amplitude of body curvature remains constant. We then combined experimental imaging of freely swimming eleutheroembryos with Navier-Stokes numerical simulations. The results showed that the mechanical power output was initially maximal and remained essentially stable with increasing viscosity, while the cost of transport was linearly correlated with viscosity. Eleutheroembryos maximize neuromuscular power output during the fast-start escape response, enabling them to potentially escape predators under all circumstances in a natural environment. This model may be used to identify genetic and toxicological factors that reduce the mechanical power developed by the neuromuscular system or induce a loss of efficiency in its use.

Parkinson’s disease (PD), the second most common neurodegenerative disorder, is characterized by the progressive loss of dopaminergic neurons in the substantia nigra pars compacta, leading to motor and non-motor symptoms. The neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) has been extensively used in different animal species to develop chemical models of PD. This study aimed to evaluate the effects of acute exposure to MPTP (3 × 150 mg/kg, intraperitoneally) on adult zebrafish by assessing the neurochemical, transcriptional, and motor changes associated with PD pathogenesis. MPTP treatment resulted in a significant decrease in brain catecholamines, including dopamine, norepinephrine, and normetanephrine. Additionally, a trend towards decreased levels of dopamine precursors (tyrosine and L-DOPA) and degradation products (3-MT and DOPAC) was also observed, although these changes were not statistically significant. Gene expression analysis showed the downregulation of dbh, while the expression of other genes involved in catecholamine metabolism (th1, th2, mao, comtb) and transport (slc6a3 and slc18a2) remained unaltered, suggesting a lack of dopaminergic neuron degeneration. Behavioral assessments revealed that MPTP-exposed zebrafish exhibited reduced motor activity, consistent with the observed decrease in dopamine levels. In contrast, the kinematic parameters of sharp turning were unaffected. A significant impairment in the sensorimotor gating of the ASR was detected in the MPTP-treated fish, consistent with psychosis. Despite dopamine depletion and behavioral impairments, the absence of neurodegeneration and some hallmark PD motor symptoms suggests limitations in the validity of this model for fully recapitulating PD pathology. Further studies are needed to refine the use of MPTP in zebrafish PD models.

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.

Existing approaches for human health risk assessment of chemicals have overall provided a high level of protection for EU citizens. However, the established assessment schemes face numerous challenges. Ethical and scientific concerns about using animals for safety testing have triggered awareness of the need for a paradigm shift, requiring new concepts for chemical risk assessment complementing and, in the long run, replacing existing schemes. Next-Generation Risk Assessment (NGRA) using new approach methodologies (NAMs) is commonly regarded as the way forward. However, adequately meeting regulatory needs is challenging. The European Partnership for the Assessment of Risks from Chemicals (PARC) supports the development of NGRA frameworks and their implementation at all levels, from developing adverse outcome pathways (AOPs), NAMs and integrated approaches to testing and assessments (IATAs), to designing new conceptual approaches and formulating strategic roadmaps. A particular strength of PARC is its focus on interaction and collaboration with stakeholders from all sectors of the chemical risk assessment community to promote cooperation, advance research, increase knowledge of chemical risk assessment and train methodological skills. Its results will help launch European and national strategies to reduce risks posed by hazardous chemicals, to reduce animal testing and to implement NGRA strategies in regulatory practice.

Access to information about chemicals in products and articles is critical for supporting enforcement of chemical regulations, assessing risks from chemicals, allowing informed consumer choices, and enabling product circularity. In this work, we identified and evaluated available databases (DBs) on chemicals in products and articles from the literature using a defined protocol and from European national market surveillance authorities, nongovernmental agencies, and industrial sector groups using questionnaires. This is the first comprehensive review of DBs that provide information about chemicals in products and articles. A majority of these DBs are heterogeneous in terms of scope, ontologies, and data structures. Among the 57 identified DBs, 49 identified specific substances and only 30 reported their concentration in their products. In addition, 35 DBs included hazard information and 27 DBs provided safety information about products or chemicals. The analysis highlights the lack of comprehensive or accessible data on chemicals in products and articles for most categories of products/articles and jurisdictions. The limitations of existing DBs were attributed to scattered regulatory information requirements, a lack of data for unregulated substances, the complexity of supply chain communication, and confidentiality issues. In response to these challenges, we identified opportunities for improving existing information transfer structures and exploring alternative data sources to promote product and article safety and circularity.

Finding relevant chemicals in the vast (known) chemical space is a major challenge for environmental and exposomics studies leveraging nontarget high resolution mass spectrometry (NT-HRMS) methods. Chemical databases now contain hundreds of millions of chemicals, yet many are not relevant. This article details an extensive collaborative, open science effort to provide a dynamic collection of chemicals for environmental, metabolomics, and exposomics research, along with supporting information about their relevance to assist researchers in the interpretation of candidate hits. The PubChemLite for Exposomics collection is compiled from ten annotation categories within PubChem, enhanced with patent, literature and annotation counts, predicted partition coefficient (logP) values, as well as predicted collision cross section (CCS) values using CCSbase. Monthly versions are archived on Zenodo under a CC-BY license, supporting reproducible research, and a new interface has been developed, including historical trends of patent and literature data, for researchers to browse the collection. This article details how PubChemLite can support researchers in environmental and exposomics studies, describes efforts to increase the availability of experimental CCS values, and explores known limitations and potential for future developments. The data and code behind these efforts are openly available. PubChemLite can be browsed at https://pubchemlite.lcsb.uni.lu ↗.

The AOP-Wiki, a knowledge database for Adverse Outcome Pathways (AOPs), requires an efficient way to present an overview of its content for the reconstruction of networks by experts in a given domain. We have developed a user-friendly tool using artificial intelligence, the AOP-networkFinder, that retrieves AOPs of interest, modifies them in a desirable manner, and finally visualizes networks built around the retrieved AOPs. Our tool constructs AOP networks by connecting AOPs that use the same Key Events (KEs) in a versatile but controlled manner. Genes related to these KEs are also displayed. The constructed networks can then be exported as images or to Cytoscape for further fine-tuning and statistical analysis. The AOP-networkFinder allows users to comprehensively identify relationships between KEs and visualize the overall structure of an AOP both quickly and easily. This is immensely beneficial to researchers who need to understand the complex interplay between different KEs and the overall pathway they are studying and helps them to build further networks of interest while logging relevant information about changes within the network. These efforts are in line with the FAIR principles, which are crucial attributes for any developed databases and tools for optimizing use and reuse in a dynamically changing landscape of AOP-Wiki.

Human activities severely impact biodiversity, particularly in freshwater lakes. These habitats provide critical ecosystem services and, at the same time, suffer from river inflow, agricultural runoff, and urban discharge. DNA-based techniques are preferred for monitoring biodiversity due to their effectiveness. However, pinpointing the causes of biodiversity decline across landscapes poses challenges due to the complex interactions between biodiversity and environmental drivers. In this study, we used an explainable multimodal machine learning approach that can integrate different types of data, such as biological, chemical, and physical data, to discover potential causes of biodiversity dynamics. This is done by identifying relationships between environmental drivers—plant protection products, physico-chemical parameters and typology- and community biodiversity changes in 52 lake ecosystems. By analyzing benthic and pelagic lake communities, we found significant correlations between biodiversity and environmental drivers, such as plant protection products. Furthermore, our analysis allowed us to identify factors within these drivers responsible for biodiversity dynamics. Specifically, insecticides and fungicides were identified as the most important factors, followed by 43 physico-chemical factors, including many heavy metals. Our holistic, data-driven approach provides insights into large-scale biodiversity changes and could inform conservation efforts and regulatory interventions to protect biodiversity from pollution.

Assessment of potential impacts of chemicals on the environment traditionally involves regulatory standard data requirements for acute aquatic toxicity testing using algae, daphnids, and fish (e.g., Organisation for Economic Co-operation and Development [OECD] test guidelines 201, 202, and 203, respectively), representing different trophic levels. In line with the societal goal to replace or reduce vertebrate animal testing, alternative bioassays were developed to replace testing with fish: the fish cell line RTgill-W1 acute toxicity assay (OECD test guideline 249) and the zebrafish embryo acute toxicity test (zFET, OECD test guideline 236). However, previous studies revealed the lower sensitivity of the RTgill-W1 cell line assay and zFET for some neurotoxic chemicals and allyl alcohol, which is presumably biotransformed in fish to the more toxic acrolein (which is predicted well through the cell line assay). To provide an additional alternative to acute fish toxicity, in this study we analyzed historic ecotoxicity data for fish and daphnids from the EnviroTox Database. We found a considerable variability in acute fish median lethal concentration and acute daphnids median effect concentration values, particularly for neurotoxic chemicals. Comparing sensitivity of these taxonomic groups according to different neurotoxicity classification schemes indicates that fish rarely represent the most sensitive trophic level of the two. Exceptions here most prominently include a few cyclodiene compounds, which are no longer marketed, and a chemical group that could be identified through structural alerts. Moreover, daphnids are more sensitive than fish to acrolein. This analysis highlights the potential of the Daphnia acute toxicity test, which is usually a standard regulatory data requirement, in safeguarding the environmental protection level provided by the RTgill-W1 cell line assay and the zFET. This research, rooted in decades of efforts to replace the fish acute toxicity test, shifts the focus from predicting fish toxicity one-to-one to emphasizing the protectiveness of alternative methods, paving the way for further eliminating vertebrate tests in environmental toxicology.

Endocrine-disrupting chemicals (EDCs) are natural or synthetic substances that are able to interfere with hormonal systems and alter their physiological signaling. EDCs have been recognized as a public health issue due to their widespread use, environmental persistence and the potential levels of long-term exposure with implications in multiple pathological conditions. Their reported adverse effects pose critical concerns about their use, warranting their strict regulation. This is the case of bisphenol A (BPA), a well-known EDC whose tolerable daily intake (TDI) was re-evaluated in 2023 by the European Food Safety Authority (EFSA), and the immune system has been identified as the most sensitive to BPA exposure. Increasing scientific evidence indicates that EDCs can interfere with several hormone receptors, pathways and interacting proteins, resulting in a complex, cell context-dependent response that may differ among tissues. In this regard, the neuronal and immune systems are important targets of hormonal signaling and are now emerging as critical players in endocrine disruption. Here, we use BPA and its analogs as proof-of-concept EDCs to address their detrimental effects on the immune and nervous systems and to highlight complex interrelationships within the immune–neuroendocrine network (INEN). Finally, we propose that Receptor for Activated C Kinase 1 (RACK1), an important target for EDCs and a valuable screening tool, could serve as a central hub in our toxicology model to explain bisphenol-mediated adverse effects on the INEN.