Predicting the immunotoxic potential of chemical mixtures based on their endocrine activity profiles via computational modeling and in vitro bioassays

Exposure to endocrine disrupting chemicals (EDCs) is a known threat to the environment and human health, because EDCs can interfere with many developmental processes. As immune cells rely on signaling pathways than can be impacted by EDCs, the disruption of endocrine processes exhibits the potential to directly or indirectly adversely alter immune response. We employed a high-throughput in silico workflow to identify EDCs that are likely to alter endocrine and metabolic processes with direct or indirect immunotoxic potential. Initially included 83,693 chemicals were reduced to a priority list of 6,620 and matched with blood concentrations as reported in human biomonitoring studies. Mixtures in representative concentration ratios were prepared for five main groups of EDCs with five chemicals per mixture, covering per- and polyfluorinated compounds, halogenated pesticides, phthalates, phenols, and polycyclic aromatic hydrocarbons, and one mixture containing all 25 chemicals. Single compounds and mixtures were tested in high-throughput bioassays covering potential indirect modulators of immune response such as endocrine signaling through activation of estrogen receptor, androgen receptor, glucocorticoid receptor and aryl hydrocarbon receptor, thyroid hormone signaling through agonism and antagonism on the thyroid hormone receptor, thyroperoxidase inhibition, and transthyretin binding, or more direct endpoints such as pro-inflammatory signaling through the activation of nuclear factor κ-light-chain-enhancer of activated B cells, as well as redox and metabolic imbalance via oxidative stress response and mitochondrial disruption. 19 out of 33 bioactive mixtures confirmed concentration additivity across all tested endpoints within 3-fold deviation, indicating a clear relevance of commonly detected EDCs mixtures as potential immunomodulators.