The ENDOMIX perspective: how everyday chemical mixtures impact human health and reproduction by targeting the immune system† Authors: Sergio Gómez-Olarte, Verena Mailänder, Júlia Castro-Neves, Violeta Stojanovska, Anne Schumacher, Nicole Meyer, Ana C Zenclussen Published: 2024 ABSTRACT AbstractEndocrine-disrupting chemicals are natural and synthetic compounds found ubiquitously in the environment that interfere with the hormonal-immune axis, potentially impacting human health and reproduction. Exposure to endocrine-disrupting chemicals has been associated with numerous health risks, such as neurodevelopmental disorders, metabolic syndrome, thyroid dysfunction, infertility, and cancers. Nevertheless, the current approach to establishing causality between these substances and disease outcomes has limitations. Epidemiological and experimental research on endocrine-disrupting chemicals faces challenges in accurately assessing chemical exposure and interpreting non-monotonic dose response curves. In addition, most studies have focused on single chemicals or simple mixtures, overlooking complex real-life exposures and mechanistic insights, in particular regarding endocrine-disrupting chemicals' impact on the immune system. The ENDOMIX project, funded by the EU’s Horizon Health Program, addresses these challenges by integrating epidemiological, risk assessment, and immunotoxicology methodologies. This systemic approach comprises the triangulation of human cohort, in vitro, and in vivo data to determine the combined effects of chemical mixtures. The present review presents and discusses current literature regarding human reproduction in the context of immunotolerance and chemical disruption mode of action. It further underscores the ENDOMIX perspective to elucidate the impact of endocrine-disrupting chemicals on immune-reproductive health.Keywords: endocrine-disrupting chemicals, chemical mixtures, cohort studies, immunotoxicity, immune function, hormone receptors, organoid models, reproductive outcomes, mouse models Introduction Modern human populations are constantly co-exposed to several environmental chemicals that can induce transitory or permanent biological deviations from homeostasis, such as acute cellular responses, epigenetic reprogramming, or even mutagenesis [1, 2]. A group of heterogenous substances classified as endocrine-disrupting chemicals (EDCs) are of particular concern because they interfere with endocrine functions, which regulate development, growth, metabolism, and reproduction. These substances consist of naturally occurring molecules (e.g., phytoestrogens) and synthetic compounds utilized in several industrial products, including plastics (bisphenol A [BPA]), plasticizers (phthalates), pharmaceutical agents (diethylstilbestrol [DES]), surfactants (per- and polyfluoroalkyl substances [PFAS]), herbicides (atrazine), and metal elements (cadmium), among others [3]. EDCs modify not only the production, secretion, transport, and degradation of endogenous hormones but also their binding, signaling, and metabolism in hormone-producing and responsive cells, such as mature reproductive cells and immune cells [4]. Therefore, EDC exposure can impact human and environmental well-being, raising public health concerns that are being addressed with chemical regulatory frameworks in the European Union (EU). EDC exposure has been linked with an increased risk of neurodevelopmental disorders [5], metabolic syndrome triggering obesity and diabetes [6], thyroid dysfunction [7], male and female infertility [8], and mammal gland and prostate cancers [9]. Indeed, these chemicals are of special concern for reproduction and fertility since they mainly target steroid hormone receptors (SHRs) in hormone-responsive cells [10]. Furthermore, due to the role of immune-endocrine cross-talk during pregnancy and the expression of SHRs in immune cells [11], the immune system has been recognized as a key mediator of EDC effects [12]. The association between EDC exposure and disease outcome is frequently taken as demonstrated. However, most of the scientific evidence is limited to the scope of single chemicals or simple mixtures of two or three molecular-related substances but it rarely investigates EDC’s mechanistic mode of action [13], such as immunotoxicity. A systemic approach to studying EDC should integrate data from prospective human cohorts, complex in vitro bioassays, and adequate animal models while considering the exposure to real-life chemical mixtures during highly susceptible life stages. Most epidemiological and experimental studies have focused on correlating single exposures to EDCs with a specific outcome. However, a principle of “something from nothing” has been described, by which EDC mixtures can provoke biologically relevant alterations at levels that single chemicals do not induce measurable effects [14]. Thus, investigating the combined effect of EDC mixtures on health and reproduction is physiologically relevant and could provide a more accurate estimation of EDC exposure risk. This is in line with the exposome approach to EDC research, which comprises the assessment of life-course environmental exposures from fetal development onwards [15]. Here, we propose a more practical strategy involving the analysis of EDC-derived biological disturbances during critical stages of life, namely pregnancy, adolescence, and reproductive age [16]. This effort requires multicenter studies that harmonize epidemiological data and triangulate target organs and cellular mechanistic studies in the context of relevant EDC mixtures at real-life concentrations. EDC exposure is ubiquitous among global populations and poses a substantial health and economic burden, estimated at €163 billion (1.28% of EU Gross Domestic Product) per year in the EU, from which ~5% is attributed to reproductive disorders [17]. Even though the high costs of chemicals-induced disease and disability have been documented, there is a huge gap between science and policy that hampers the definition and enforcement of regulations focused on minimizing European citizens’ exposure to EDCs [18]. Within the funding frame of the EU’s Horizon Health Program, the multidisciplinary project “Understanding how endocrine disruptors and chemical mixtures of concern target the immune system to trigger or perpetuate disease” —ENDOMIX aims to determine immune-mediated health and reproductive impacts of EDC mixtures. This review outlines the conceptualization, methodology, and analytical approach behind ENDOMIX with a particular focus on fertility and reproductive outcomes. Firstly, we provide an overview of how EDCs can interfere with hormonal functions during pregnancy and reproductive development. Next, we summarize relevant literature about immune cell interactions susceptible to chemical stress and their potential involvement in the development of fertility and reproductive disorders. Lastly, ENDOMIX objectives and methodological approach are presented, underlining how novel findings regarding EDC toxicology are needed to bridge science-policy gaps. Challenges of identifying and testing real-life EDC mixtures The exposure to real-life chemicals and their mixtures is variable and dynamic among human populations, making it challenging to establish causality between EDCs and disease outset. Currently, both observational and experimental strategies to investigate EDC exposure effects present limitations. For instance, epidemiological studies have inherent pitfalls associated with reliable exposure assessment, such as difficulty in measuring EDC exposure during critical periods of life (e.g., fetal development and puberty), latency between the exposure to non-persistent EDCs and outcome onset, correlations with cofounding bias, and lack of unexposed control groups [19]. Experimental studies using in vitro and animal models assess variable chemical exposure and doses but sometimes their design does not account for non-monotonic dose responses (NMDR), low dose effects, or interactions (e.g., additive, synergetic, or antagonist) within chemical mixtures [20]. NMDR describes the biological effect of a substance that does not consistently increase or decrease with the dose [21], arguing against risk assessment strategies that predict chemical low dose safety solely based on extrapolations of high-dose adverse effects [22]. Consequently, the requirement for a modern strategy integrating epidemiological, risk assessment, and immunotoxicology methodologies has been proposed [23]. To our knowledge, such research initiatives have not been undertaken. In the past decades, component-based methodologies have been successful in addressing EDC mixture effects. In this approach, a restricted number of chemicals (usually <10) with defined concentrations and modes of action (e.g., antiandrogens and xenoestrogens) are mixed and tested [24]. However, these simplified EDC mixtures do not reflect the toxicological profiles of real-life co-exposures, involving at least thousands of chemicals. Several human biomonitoring (HBM) Europe-wide studies have measured EDCs in blood, plasma, and urine samples showing the vast heterogeneity of the internal exposome. Remarkably, a cross-sectional German study of children and adolescents (3–17 years old) detected 31 chemical substances above the quantification limit in urine; the components and their concentration (median in μg/g creatinine) in the mixture comprised 5 elements (0.05–6.89), 17 phthalates (0.93–21.36) and 3 substitutes (2.03–4.66), 6 polycyclic aromatic hydrocarbons (PAHs; 0.04–3.15), and BPA (1.6). Another HBM report of Spanish workers above age 16 found similar levels of phthalates (0.62–189.47) and substitutes (0.7–0.43) in urine, as well as perfluorononanoic acid and perfluorodecanoic acid (0.95 and 0.37 μg/L) in blood. Altogether, this underscores the complexity of the internal exposome and the challenges it poses for the risk assessment of EDC toxicity. The toxicological profile of a chemical mixture does not equal the sum of its single component’s effects. Substances within the mixture can display additive or synergistic interactions provoking adverse effects through various modes of action, even when each chemical component is below the no-observable-adverse-effect-level (NOAEL) [25]. Between 2000 and 2020, about half of in vivo studies on EDC exposure at low doses (300 biomarkers, including cytokines, chemokines, and chemokines receptors, among other functional proteins. The adverse effect of chemicals on epigenetic footprint, metabolome, and gut microbiome will also be considered. By this means, ENDOMIX will identify EDC mixtures of concern and potential intermediate biomarkers of exposure based on real-life chemicals already measured in the human cohorts: PCBs, OCPs, PFAS compounds, metals, parabens, phthalates, phenols, and organophosphate pesticides. Through this strategy, we will determine the most sensitive windows of susceptibility to specific profiles of EDC exposure and propose measures to prevent undesirable health effects. Objective IV: determine how relevant chemical mixtures disrupt molecular mechanisms to perpetuate disease We will expand the in vitro analysis of the chemical mixtures to placenta co-culture models, organoids, and barrier models that have been developed in our laboratory. Remarkably, dose-relationship assays and single chemical controls will be considered in the experimental design when needed. ENDOMIX will employ an in vitro trophoblast-based barrier model to access EDC effects on placenta functionality. The placenta barrier model will be exposed to EDC mixtures and co-culture with immune cells that can eventually infiltrate the decidua (NK and T cells), in both steady-state and inflammatory conditions. A second phase will involve 3D placenta models, such as spheroids and organoids generated by culturing trophoblast cell lines and primary cells with adequate extracellular matrices as previously published [231, 232]. The organoids will be co-cultured with complementary cell types (e.g., fetal endothelial cells) and immune cells to address EDC mixtures’ effect on cell interactions. Afterward, the following endpoints will be analyzed: cellular viability, migration, invasion, gene expression of cell-specific markers, and hormone production. In the third phase, we will cross-scale chemical mixture doses to animal studies. Special emphasis will be placed on transgenerational 3R-conform mouse models to validate the observed toxic effects on human cohorts and placenta barrier and organoid models. In this manner, we can triangulate reproductive outcomes, alterations in the placenta functionality, and modulation of pregnancy-relevant immune cell populations associated with EDC exposure. Upon chemical exposure of female mice during pregnancy and lactational periods, we will determine changes in metabolic and reproductive health (e.g., vaginal opening, estrus cycle, and sperm) [233] parameters among the female and male offspring. The impact of EDC in utero exposure on the offspring will also be investigated in settings of induced asthma [234, 235] and rheumatoid arthritis [236]. Altogether, this robust design involving physiologically relevant in vitro bioassays will reduce the use of animals for experimentation and enable exploring the underlying mechanisms governing chemical disruption. Objective V: integrate human cohort, in vitro and in vivo data The last objective involves adopting the weight-of-evidence (WoE) approach to concatenate, weigh the lines of evidence, and draw conclusions about the immunotoxicity of the tested EDC mixtures. The WoE data input will comprise epidemiological, in vitro, target organ/barrier models, and in vivo results from ENDOMIX, as well as existing literature and database-derived information. Moreover, we will employ the automated strategy Network and Dynamical Reasoning Assembler (INDRA) to discover molecular pathways targeted by EDC mixtures of concern. INDRA classifies and sorts out redundant information from published papers and databases and applies predictive AI models (e.g., natural language processing) to generate a framework of mechanistic networks of disease outcomes [237]. Accordingly, the resulting INDRA risk statements (e.g., BPA alters androgenic activity) can be implemented in rule-based and network models, such as knowledge graphs, and compiled in a comprehensive guide for EDC risk assessment in terms of human health and reproduction. Possible policy implications of ENDOMIX Overall, ENDOMIX aims to provide novel scientific-based evidence concerning the relevance of updating the toxicological risk assessment paradigm, from the traditional chemical-by-chemical approach toward a more realistic mixture risk assessment (MRA) concept, as previously proposed [238]. We have established a rigorous system biology strategy that integrates epidemiological, in vitro, and in vivo data to validate the toxicological relevance of MRA. In this sense, we expect to uncover chemical interactions targeting the immune system and posing a risk to human health and reproduction, as a complement to new assessment methods developed by other EU-funded initiatives (EuroMix, EDC-MixRisk, EUToxRisk, HBM4EU, and JRC) [239]. In concrete, to influence policy-making and the development of EU regulations, we will provide research-based recommendations about hazard and risk assessment. These include a guide of co-exposure patterns for specific consumer groups, depending on lifestyle and diet, as well as hazardous and unintentional chemical mixtures derived from key industries, such as pharmaceuticals, medical implants, tobacco, clothing, and food. To encourage the acceptance of innovative base guidelines, the project involves scientific advisors and a dedicated Stakeholder Board that will advocate for future policies. Likewise, citizens will be invited to participate in dissemination and educational activities, promoting awareness of how to mitigate exposure to real-life chemical mixtures that threaten human health and our natural environment. Potential limitations of the ENDOMIX approach First, ENDOMIX anticipates to uncover the health impact of realistic EDC mixtures by identifying representative mixtures of concern. Nonetheless, the inferred mixtures may overrepresent EDC that are commonly reported in toxicological databases and HBM studies, neglecting chemicals with short half-lives that can often only be detected in repeated biospecimen measurements [240, 241]. This exposure assessment bias could affect the selection of individual EDCs and the preparation of the mixtures for the bioassays. Second, we can solely test the effects of the prioritized single chemicals and their mixture in our complex barrier and organ models due to resources and cost restrictions. Although this can limit the validation of key toxic events for some EDC mixtures, we anticipate partially overcoming the gap via in silico models based on HT screening results and human cohort data. Finally, we will screen and prioritize thousands of potential EDCs focused on the mixture concept of concentration addition, which could overlook modulating or antagonistic chemical interactions. Yet, additivity and synergy are the most frequent mixture effects linked to adverse health outcomes [13]. Concluding remarks In the last decades, the risk assessment of EDC exposure in humans has been addressed using independent and non-standardized epidemiological and experimental designs. Thus far, most of the studies have focused on single chemicals and neglected their combined adverse effects on hormone-regulated organ systems, such as the immune system. In addition, EDC outcomes among vulnerable populations like pregnant women, newborns, and children are barely considered in risk and hazard assessment or studies addressing chemical impact on health. Therefore, the lack of robust scientific evidence regarding the health impact of real-life EDC mixtures has impaired the establishment of EU regulations to mitigate exposure in susceptible populations. In this sense, the ENDOMIX consortium tackles this knowledge gap with an interdisciplinary and systemic approach that brings together renowned toxicologists, epidemiologists, environmental immunologists, clinical and data scientists, and risk assessment experts. ENDOMIX’s main objective is to provide novel insights into the mechanistic relationship between chemical-driven immune modulation and adverse effects on health and male and female reproduction. Our approach involves characterizing EDC mixtures of concern in large human cohorts and assessing their direct toxic effects on immune cells, organoids, physiological barriers, and transgenerational animal models. The project envisions the triangulation of epidemiological, in vitro, and in vivo data by using statistical and computational strategies to identify intermediate biomarkers and toxicological endpoints of exposure during critical life stages. The end goal of ENDOMIX is to deliver pertinent evidence regarding EDC-induced immunotoxicity and science-to-policy translation guidelines to European society, comprising stakeholders, health professionals, the scientific community, and policymakers. Therefore, enabling the construction of a regulatory framework and empowered decision-making to minimize the EDC burdens on public health and the natural environment. Ethical considerations The ENDOMIX consortium conducts research with human data and animals under the appropriate legislation declared by the Declaration of Helsinki, ICH guideline for Good Clinical Practice, ISO guidelines on Good Clinical Practice ISO 14155:2020, the European Directive 2001/20/EC on Good Clinical Practice, and the EU General Data Protection Regulation (2016/679). Footnotes † Grant Support: This work was funded by the European Union’s European Health and Digital Executive Agency under grant agreement no. 101136566. Contributor Information Sergio Gómez-Olarte, Department of Environmental Immunology, Helmholtz Centre for Environmental Research GmbH—UFZ, Permoserstraße 15, 04318, Leipzig, Germany. Verena Mailänder, Department of Environmental Immunology, Helmholtz Centre for Environmental Research GmbH—UFZ, Permoserstraße 15, 04318, Leipzig, Germany. Júlia Castro-Neves, Department of Environmental Immunology, Helmholtz Centre for Environmental Research GmbH—UFZ, Permoserstraße 15, 04318, Leipzig, Germany. Violeta Stojanovska, Department of Environmental Immunology, Helmholtz Centre for Environmental Research GmbH—UFZ, Permoserstraße 15, 04318, Leipzig, Germany. Anne Schumacher, Department of Environmental Immunology, Helmholtz Centre for Environmental Research GmbH—UFZ, Permoserstraße 15, 04318, Leipzig, Germany; Perinatal Immunology, Saxonian Incubator for Clinical Translation (SIKT), Medical Faculty, Leipzig University, Philipp-Rosenthal-Straße 55, 04103, Leipzig, Germany. Nicole Meyer, Department of Environmental Immunology, Helmholtz Centre for Environmental Research GmbH—UFZ, Permoserstraße 15, 04318, Leipzig, Germany; Perinatal Immunology, Saxonian Incubator for Clinical Translation (SIKT), Medical Faculty, Leipzig University, Philipp-Rosenthal-Straße 55, 04103, Leipzig, Germany. Ana C Zenclussen, Department of Environmental Immunology, Helmholtz Centre for Environmental Research GmbH—UFZ, Permoserstraße 15, 04318, Leipzig, Germany; Perinatal Immunology, Saxonian Incubator for Clinical Translation (SIKT), Medical Faculty, Leipzig University, Philipp-Rosenthal-Straße 55, 04103, Leipzig, Germany. Funding The ENDOMIX project has received funding from the European Union’s European Health and Digital Executive Agency under grant agreement No. 101136566. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Health and Digital Executive Agency. Neither the European Union nor the granting authority can be held responsible for them. Conflict of interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Author contributions S.G.O.: conceptualization, investigation, writing, and editing; V.M. and J.C.N.: investigation and writing. V.S., A.S., and N.M.: review and editing; A.C.Z.: conceptualization, editing, and funding acquisition. All authors read and approved the final manuscript. References