Drug-induced liver injury (DILI) remains a major challenge in drug development, highlighting the need for reliable in vitro tools to assess hepatotoxicity and improve translation from animal to human studies. Models using hepatocytes from preclinical species are also needed to evaluate species-specific toxicity. In this study, we evaluated the basal function and DILI sensitivity of primary hepatocytes from human, monkey, rat, and dog cultured for up to 14 days in static monolayers, static spheroids, and spheroids cultured in a microfluidics-based microphysiological system (MPS). Hepatocyte function and injury responses were assessed using albumin, urea, and liver enzymes. Cells were exposed to species-specific DILI compounds chlorpromazine (CPZ), bosentan (BOS), and fialuridine (FIAU). Across platforms, human and monkey hepatocytes exhibited greater functional stability and sensitivity to DILI compounds than rat and dog hepatocytes. CPZ and BOS induced cytotoxicity primarily in human and monkey hepatocytes, while FIAU produced species-dependent effects consistent with known in vivo outcomes. The microfluidics-based MPS exhibited modestly improved hepatocyte spheroid function relative to static models, although limited MPS throughput constrained our ability for testing drugs beyond FIAU. Overall, these results demonstrate that integrating multi-species hepatocyte spheroids across static and microfluidic platforms enables comparative DILI assessment and supports improved preclinical-to-clinical translation.
The DrugMatrix database contains systematically generated toxicogenomics data from short-term in vivo studies for over 600 chemicals. However, most potential endpoints are missing due to a lack of experimental measurements. Therefore, we leveraged matrix factorization and machine learning methods to predict the missing values, which includes gene expression across eight tissues on two expression platforms along with paired clinical chemistry, hematology, and histopathology. We propose a method, ToxCompl, that applies systematic hybrid sampling guided by Bayesian optimization in conjunction with low-rank matrix factorization to predict the missing values. In-depth validation of the ToxCompl predicted data from machine learning, biological, and toxicological perspectives shows that the predicted differential gene expression aligns well with what would be anticipated. This includes examining the connectivity pattern of predicted gene expression responses, characterizing molecular pathway-level responses from sets of differentially expressed genes, evaluating known transcriptional biomarkers of tissue toxicity, and characterizing predicted apical endpoints. For example, we identified kidney toxicants using the transcriptional biomarker Havcr1. All measured and predicted DrugMatrix data (i.e., gene expression, clinical chemistry, hematology, and histopathology) are available to the public (https://rstudio.niehs.nih.gov/toxcompl/). Notably, predicted clinical chemistry of subtle effects and histopathological prediction are two areas we will continue to improve. The main advantage of the ToxCompl approach is that it drastically extends the toxicogenomic landscape into many data-poor tissues in the absence of acquiring additional experimental data, thereby allowing researchers to formulate mechanistic hypotheses about effects in tissues that have been underrepresented in the literature.
The development of New Approach Methodologies (NAMs) for genotoxicity testing is needed to support the transition towards animal-free human-relevant chemical genotoxicity assessment. In this context, we investigated the predictive power of two transcriptomic biomarkers developed in 2D cell systems (GENOMARK and TGx-DDI) in metabolically active 3D HepaRG spheroids to classify chemicals as genotoxic (GTx) or non-genotoxic (NGTx). Human-hepatic HepaRG spheroids were treated for 72h with three different groups of chemicals: A) five genotoxic chemicals: lasiocarpine, cyclophosphamide, ethyl methanesulfonate, aflatoxin B1, benzo[a]pyrene; B) five non-genotoxic non-carcinogens: aflatoxin B2, 2-deoxy-D-glucose, D-mannitol, caffeine, ampicillin trihydrate; and C) five non-genotoxic carcinogens: perfluorooctanoic acid, diethanolamine, thioacetamide, phorbol-12-myristate-13-acetate, di(2-ethylhexyl) phthalate. Gene expression data from the treated HepaRG spheroids were collected using the high-throughput TempO-SeqTM technology and analysed with the transcriptomic biomarkers GENOMARK and TGx-DDI. Results with both biomarkers generally aligned with existing knowledge for all three groups of tested chemicals, effectively distinguishing genotoxic from non-genotoxic chemicals. The results of this study indicate that combining the 3D HepaRG spheroid model with transcriptomic biomarkers offers a promising, human-relevant approach to complement traditional in vitro genotoxicity tests.
Volatile organic compounds (VOCs) are ubiquitous inhaled pollutants. This review showcases current literature utilizing in vitro models of the human respiratory system to characterize the toxicity of VOCs. To map the existing evidence base, we conducted a scoping review following systematic search and screening procedures. Comprehensive searches of MEDLINE, Embase, Web of Science, CINAHL, PubMed, and CENTRAL identified 3,052 records. After screening, 144 original studies evaluating VOC exposures in human lung epithelial models met inclusion criteria. Overall, the current literature reflects substantial heterogeneity in cell models, exposure systems, and endpoints. Among 105 unique VOCs evaluated, acrolein, formaldehyde, toluene diisocyanate, and benzene were most frequently studied. Most investigations used submerged culture systems with liquid-phase VOC application, while fewer employed air-liquid interface (ALI) exposures that better mimic inhalation. Cytotoxicity, oxidative stress, pro-inflammatory signaling, and apoptosis were the most commonly measured endpoints, with oxidative stress frequently identified as an upstream driver of inflammatory and cytotoxic responses. However, mechanistic depth varied, and studies examining metabolism, barrier function, morphology, or transcriptomic regulation were relatively uncommon. Notably, chronic or repeated exposures were rarely conducted. Overall, this review highlights the variety of VOC exposure methods and commonly assessed biological endpoints, as well as the critical lack of more detailed mechanistic studies and somewhat limited VOC/mixture evaluation. The field is expansive but methodologically fragmented, underscoring the need for broader use of human respiratory cell models, more physiologically relevant exposure systems, improved dose characterization, and greater mechanistic resolution to advance understanding and throughput for VOC-induced pulmonary toxicity studies.
Adipose tissue plays a critical role in metabolic and endocrine function because it is essential for maintaining systemic energy homeostasis and other related physiological functions. Mammals have four types of adipose tissue: white adipose tissue (WAT), brown adipose tissue (BAT), beige or brite adipose tissue (BeAT), and pink adipose tissue (PAT). These adipose tissues release endocrine factors that modulate diverse processes such as energy storage and expenditure, appetite control, glucose homeostasis, insulin sensitivity, inflammation, lipid metabolism, tissue repair, thermogenesis, and milk production. Proper adipose tissue function relies on hormone receptors and signaling pathways that make the adipose tissues susceptible to disruption by endocrine-disrupting chemicals such as phthalates. Here, we review relevant research on the associations between phthalate exposures and abnormalities in WAT and BAT functions, including phthalate-induced changes in morphology, physiology, and gene expression effects. This review covers in vitro studies, in vivo studies in mammals, and studies in humans. We also discuss important gaps in the literature. Overall, the evidence indicates that phthalates adversely affect WAT and BAT functions. Further studies are needed to better elucidate the mechanisms through which phthalates act in the adipose tissues and to determine the effects of phthalates on human adipose tissues.
Electronic cigarettes (e-cigarettes), particularly nicotine-free (0%) formulations, are often perceived as safer alternatives to nicotine-containing products. However, the progressive and mechanistic basis of blood-brain barrier (BBB) injury following acute and sub-chronic 0% exposure remains poorly researched. In this study, we investigated whether short-term and prolonged 0% e-cigarette exposure alters BBB integrity using complementary in-vitro and in-vivo models. Using a well-established astrocyte-bEnd.3 co-culture BBB model, 24-hour exposure significantly reduced transendothelial electrical resistance (TEER) without increasing sodium fluorescein (NaF) permeability. Following prolonged 5-day exposure, TEER remained significantly reduced and was accompanied by increased NaF permeability. In contrast, permeability to 4 kDa and 70 kDa FITC-dextran tracers remained unchanged. Immunocytochemistry (ICC) demonstrated reduced claudin-5 mean fluorescence intensity (MFI) after 24 hours, along with reduced ZO-1 and claudin-5 MFI and disrupted junctional morphology following 5-day exposure. Western blot (WB) analysis additionally demonstrated a significant reduction in claudin-5 expression after 5 days. To assess translational relevance, male and female CD-1 mice were exposed to 0% aerosols for 7 or 14 days, with 7-day exposures producing no significant changes in BBB tight junction (TJ) proteins, inflammatory cytokines, or plasma cardiac troponin-l levels. 14-day exposures reduced ZO-1 expression in both sexes, occludin expression in males, and elevated pro-inflammatory cytokines IL-17A, GM-CSF, MCP-1, TNF-α, and IL-1α without affecting plasma troponin-1. Collectively, these findings demonstrate that repeated 0% e-cigarette exposure promotes progressive BBB dysfunction and a neuroinflammatory environment, highlighting the need to consider BBB vulnerability as a critical endpoint in 0% e-cigarette exposure toxicology.
The objective of this investigation was to determine whether direct addition of BPA to cultured primary mouse or human naïve CD4+ T cells promotes IL-17A production and Th17 cell differentiation and to assess the role of aryl hydrocarbon receptor (AHR) in promoting IL-17A regulation. Kinetic studies showed IL-17 concentrations peaked by day 4 post-activation in the mouse and on day 6 in human CD4+ T cells under Th17 polarizing and nonpolarizing conditions. AHR activation in murine CD4+ T cells by TCDD or FICZ promoted IL-17A and IL-22 secretion compared with vehicle controls under non-Th17 polarizing conditions. In contrast, AHR antagonism by CH223191 promoted IL-17 production in human CD4+ T cells while IL-22 secretion was reduced compared with controls. AHR activation by FICZ in human CD4+ T cells decreased IL-17 while increasing IL-22 secretion. Conversely, AHR antagonism in murine CD4+ T cells suppressed IL-17A and IL-22 responses. BPA treatment of CD4+ T cells from either species did not produce a significant increase in IL-17A-positive cells or increased IL-17 secretion, whereas FICZ and CH223191 treatment induced IL-17A increases under nonpolarizing and/or polarizing conditions in CD4+ T cells from mouse and human, respectively. These studies suggest that BPA treatment of mature, naïve CD4+ T cells from either species does not induce IL-17A or IL-22 secretion, whereas AHR activation inversely regulates IL-17A responses between species. These results confirm published results showing CD4+ T cells are refractory to BPA-mediated effects on IL-17, suggesting reported BPA effects in murine CD4+ T cells may be strain-dependent.
The widespread presence of chemicals in the environment poses significant health risks to humans and wildlife. Existing traditional in-vivo approaches, while informative, are limited by considerable ethical, economic, and logistical challenges that prevent timely and effective chemical testing and management strategies. To overcome these challenges, new approach methodologies (NAMs) offer promising alternatives for high-throughput, mechanistic, and ethically conscious chemical toxicity testing. Key to testing chemicals for potential adverse effects is selection of appropriate bioassays that cover the range of biological or physiological pathways likely targeted by the contaminant of interest. In other words, chemical testing strategies need to cover a broad range of physiological systems to account for mechanistic specificity of contaminants. Researchers and regulators thus face the challenge of navigating this interdisciplinary and rapidly growing space. This in-depth review attempts to address this important challenge by identifying and describing existing in vitro bioassays that can be used to assess functional impacts of chemical contaminants across vital physiological systems: endocrine, immune, cardiovascular, neurological, and hepatic. By summarizing the current tools and methodologies, this paper provides a comprehensive guide for researchers and regulators to help identify relevant NAMs for use in modern chemical toxicity testing strategies.
Minimizing the use of animals as concurrent controls in in vivo studies directly supports the 3Rs principles of Replacement, Reduction, and Refinement. Current virtual control group (VCG) approaches primarily rely on historical control data, but their utility may be limited by cross-study variability from differences in study design, laboratory practices, and data heterogeneity. We propose GanCtrl, a generative AI approach to infer study-specific control data directly from time-matched treatment data. By generating synthetic controls analogous to concurrent controls, GanCtrl aims to mitigate biological, temporal, and technical biases inherent in VCG approaches based on historical control data. GanCtrl was applied to rat repeat-dose toxicity studies to simulate 38 clinical pathology endpoints under control conditions using corresponding treatment data. Synthetic controls closely approximated real concurrent controls, with differences smaller than intra- and inter-laboratory baseline variation and comparable to biological replicate variability, while also preserving the typical magnitude and distribution of control responses across studies. Importantly, synthetic controls enabled detection of drug-induced clinical pathology signals and maintained biologically relevant endpoint relationships, such as ALT-AST. For practical utility, toxicity assessments using GanCtrl-derived synthetic controls were compared with those using real concurrent controls and benchmarked against VCGs constructed from single-laboratory or combined multi-laboratory historical data. Although both approaches performed comparably in the single-laboratory setting, GanCtrl outperformed VCGs when data from multiple laboratories were combined. These findings suggest that GanCtrl offers a potential approach for generating VCGs that may reduce the use of concurrent control animals and advance the 3Rs.
In vitro hepatic metabolic clearance data were generated for a diverse set of 207 chemicals to advance the collaborative initiative to establish an internal Threshold of Toxicological Concern (iTTC). The data reported herein are being used for chemical-specific physiologically based pharmacokinetic (PBPK) modeling to convert oral No Observable Adverse Effect Levels (NOAELs) into estimates of internal exposure. Hepatocyte assays were conducted at 2 concentrations (0.1 and 1 µM) using cryopreserved cells from multiple species, ensuring applicability to existing mammalian toxicity studies. The metabolic clearance measurements across chemicals varied significantly, ranging from 0 to 4,294 µl/min/106 cells at 0.1 µM and from 0 to 2,351 µl/min/106 cells at 1 µM. A substantial proportion of the chemicals (68% at 0.1 µM and 62% at 1 µM) exhibited clearance values below 30 µl/min/106 cells. Additionally, we observed a strong correlation (R = 0.8) between intrinsic clearance (CLint) values determined at the 2 concentrations. These data contribute to establishing robust iTTC values that can be utilized for: (i) extrapolating from an oral in vivo study to dermal and inhalation exposures, (ii) risk-based screening of aggregate exposures of a given substance from multiple routes of exposures, and (iii) risk-based screening of human biomonitoring results.
Ambient particulate matter (PM2.5) exposure is a major environmental risk factor for cardiopulmonary disease, but its effects on erythroid homeostasis remain incompletely understood. Although prior in vitro work indicates that PM2.5 can damage circulating red blood cells (RBCs), whether exposure alters erythropoiesis and the coordinated clearance of senescent RBCs has not been fully explored. Using a mouse model of whole-body exposure to concentrated ambient PM2.5 (CAP), we investigated associations between inhaled PM on erythroid output, splenic macrophage function, and lipid mediator signaling. CAP exposure was associated with suppressed erythropoietin levels, reduced circulating reticulocytes, and decreased erythroid precursor populations in the bone marrow, consistent with impaired erythropoiesis. Despite preserved splenic architecture, CAP-exposed mice exhibited reduced splenic iron and heme content, consistent with diminished erythrocyte turnover and processing. Targeted lipidomic profiling revealed broad suppression of proresolving lipid mediators in the spleen, with lipoxin A4 (LXA4) among the most consistently reduced species. Expression of the LXA4 receptor, ALX/FPR2, was also downregulated with prolonged exposure. Importantly, removal of CAP and return to filtered air resulted in normalization of splenic lipid mediator profiles, restoration of LXA4 levels, and recovery of erythroid parameters, including reticulocyte abundance and RBC stress markers. Together, these findings suggest that altered resolution signaling contributes to PM2.5-induced disruption of erythroid homeostasis and implicate macrophage-lipid mediator pathways in the hematologic response to environmental stress.
During lead optimization, off-target activation of the aryl hydrocarbon receptor (AhR) was identified for Compound 25. AhR is a xenobiotic-responsive transcription factor associated with adverse outcomes in rodents (and potentially humans) and modulation of drug metabolism. In rats, marked hepatic induction of AhR target genes, Cyp1a1 and Cyp1a2, was observed following repeat dosing and increased with study duration. Compound 25 was inactive in conventional AhR luciferase reporter assays but induced robust Cyp1a1 expression in metabolically competent rat and human HEPATOPAC cultures, suggesting a metabolism-dependent mechanism. Extensive metabolite profiling demonstrated that Compound 25 undergoes complex oxidative and conjugative metabolism; however, specific AhR-active metabolite(s) could not be identified. AhR involvement was confirmed in vivo using AhR knockout and wild-type rats, in which Cyp1a1/2 induction occurred exclusively in wild-type animals. Chromatin immunoprecipitation further demonstrated increased AhR binding at Cyp1a1/2 promoters following Compound 25 treatment. In rat HEPATOPAC cultures, siRNA-mediated knockdown of Por and Cyb5a attenuated Cyp1a1 induction, supporting a requirement for metabolic competence. Consistent with AhR pathway engagement, induction of CYP1A1 in human HEPATOPAC was reduced by the AhR antagonist CH223191. Collectively, these findings demonstrate that AhR activation associated with Compound 25 is dependent on metabolic transformation and may not be detected in conventional reporter assays lacking metabolic capacity. This work highlights the value of metabolically competent hepatocyte models, combined with short-duration in vivo studies, for identifying and mitigating metabolism-dependent AhR liabilities during drug discovery.
Immune systems have evolved under constant pressure from pathogens and environmental challenges, leading to the emergence of conserved defense mechanisms across diverse organisms. Evidence indicates that environmental exposures perturb immune regulatory networks, particularly during development, when transcriptional programs governing hematopoiesis, immune cell differentiation, and inflammatory signaling are highly dynamic and sensitive to external stressors. Volatile organic compounds represent an important but incompletely understood source of immunological perturbation. Among these, benzene is a ubiquitous environmental contaminant associated with hematotoxicity and immune dysregulation; however, transcriptional responses to environmentally relevant low-level exposures during development remain poorly characterized. To determine whether benzene exposure engages conserved cross-species immune regulatory pathways, we performed a comparative transcriptomic analysis integrating developmental tissues from 3 vertebrate systems: human placenta, murine placenta, and zebrafish larvae. Bulk RNA sequencing datasets were analyzed to identify transcriptional responses associated with benzene exposure in experimental models (≤5 ppm) and with benzene adduct levels in maternal plasma for human samples. Because placental gene expression exhibits strong sexual dimorphism, murine datasets were stratified by fetal sex. Pathway- and network-level analyses were used to identify conserved biological responses. We observed a striking convergence on activation of innate immune pathways associated with neutrophil degranulation, IL-8 signaling, and Rho GTPase-mediated inflammatory responses. Further, network analyses identified CXCL8 and ERK1/2 as shared regulatory hubs linking transcriptional responses across datasets. Together, these findings uncover an evolutionarily conserved innate immune signature associated with benzene exposure during vertebrate development, suggesting that environmental chemical perturbations may disrupt fundamental immune regulatory programs across species.
Ethylene oxide (EtO) is a highly reactive industrial chemical and known human carcinogen with a mutagenic mode of action (MOA). Its genotoxicity is primarily mediated through alkylation of DNA, forming the mutagenic adduct O6-(2-hydroxyethyl)-2'-deoxyguanosine (O6-HE-dG), albeit in small quantities, and the more abundant but less- or nonmutagenic N7-(2-hydroxyethyl)guanine (N7-HE-G) adduct. However, dose-response relationships of these DNA adducts, particularly at low inhalation exposure levels (< 3 ppm), remain unknown. These data are necessary to inform the biological plausibility of different statistical dose-response models that have been applied to human or animal data used for cancer risk assessment. In this study, B6C3F1 mice were exposed to EtO (0 to 200 ppm) for 6 h/d over 28 consecutive days. DNA adducts in lung, liver, bone marrow, and mammary gland were quantified using highly sensitive mass spectrometry platforms. N7-HE-G was detected in all tissues and exposure groups, showing linear dose-response relationships in the low-dose range (≤ 1 ppm) and increased sharply and exposure-disproportionately in the high-dose range (≥ 50 ppm). Despite high sensitivity, O6-HE-dG was undetectable in any tissue at exposure < 50 ppm, reflecting adduct levels that are below the current quantifiable limit. At higher exposures (≥ 50 ppm), O6-HE-dG exhibited a dose-response pattern of N7-HE-G. Notably, the mammary gland, despite being anatomically distant from the site of inhalation, exhibited the second-highest levels of both adducts at higher doses. This study provides the first reliable quantitative dose-response evidence of DNA adducts in tumor target and nontarget (liver) tissues across a wide range of EtO exposures. The two DNA adducts differ markedly in their abundance, repairability and mutagenic potential and together provide a molecular MOA dose-response framework to provide the biological foundation for informing quantitative cancer risk assessment and genotoxic hazard characterization.
Exposure to fine particulate matter (PM2.5) air pollution is associated with an increased cardiometabolic disease risk. However, although exposure timing is recognized as an important toxicity determinant, the chronotoxicity of PM2.5 exposure is less explored. To evaluate whether PM2.5 sensitivity depends on exposure timing, adult male C57/BL6 mice were exposed to concentrated PM2.5 (CAP, 6 h/day, 30 days) either during the inactive (light) phase (Zeitgeber time, ZT1 to 7) or the active (dark) phase (ZT17 to 23). Metabolic health was assessed by measuring body weight, fasting blood glucose, and plasma insulin levels. Oxidative stress and inflammatory responses in aortas and lungs were examined by quantitative real-time (qRT)-PCR, and pulmonary and circulating redox changes were measured calorimetrically. Although CAP exposure during the active phase increased blood glucose levels, inactive-phase exposure more profoundly increased antioxidant enzyme and inflammatory mRNA abundance in lungs and aortas. Inactive-phase exposure also intensified pulmonary and systemic lipid peroxidation and increased the depletion of circulating nitric oxide and lung glutathione. Examining the ability to protect against CAP-induced pulmonary oxidative stress, we found that the pulmonary antioxidant enzyme mRNA expression showed circadian rhythmicity that peaked during the active phase. These data suggest that a superior pulmonary antioxidant defense potential during the active phase could contribute to the protection against the PM2.5-induced vascular and pulmonary effects but not its impact on glucose homeostasis. Our study identified that exposure sensitivity depends on the exposure timing, which is of significance as it informs on timely susceptibility windows that could help to mitigate PM2.5 toxicity.
The potential carcinogenic and genotoxic effects of radiofrequency electromagnetic fields, particularly those emitted by mobile communication systems, have raised public health concerns. A previous study by the U.S. National Toxicology Program suggested increased incidences of gliomas and cardiac schwannomas in rats exposed to high levels of RF radiation. To evaluate these findings, an international collaborative study was initiated between Japan and Korea. Male Hsd: Sprague Dawley® SD® rats were exposed to 900 MHz CDMA-modulated RF-EMFs at a whole-body specific absorption rate of 4 W/kg for 18 hours and 20 minutes daily over two years. The study included a 28-day preliminary toxicity study, genotoxicity assays (alkaline comet and micronucleus tests), and a two-year carcinogenicity assessment. All procedures followed OECD guidelines and Good Laboratory Practice. No statistically significant increases in the incidences of neoplastic or non-neoplastic lesions were found in any major organ, including the brain, heart, and adrenal glands. Genotoxicity assays revealed no evidence of DNA damage or chromosomal aberrations in RF-exposed rats. A higher survival rate in the RF-exposed group, likely due to lower body weight and food consumption, was observed. This study performed in Japan, jointly planned and executed by Japan and Korea, provides strong evidence that long-term exposure to 900 MHz RF-EMFs did not produce reproducible carcinogenic or genotoxic effects in male rats. Combined with data from the Korean counterpart study, these results are expected to contribute to future international assessments of the carcinogenic potential of electromagnetic radiation.
Aryl hydrocarbon receptor nuclear translocator 2 is a member of the basic helix-loop-helix/Per-ARNT-Sim family of transcription factors involved in responding to various environmental, metabolic, and chemical signals. Initially known for its involvement in neurodevelopment, ARNT2 is now recognized as an essential binding partner for other transcription factors, including single-minded homologs 1 and 2, neuronal PAS domain protein 4, hypoxia-inducible factor 1, and plausibly Aryl hydrocarbon receptor to regulate stress adaptation, synaptic plasticity, immune signaling, and energy balance pathways. The role of ARNT2 has also been implicated in a plethora of pathologies, including inflammation, cardiovascular diseases, neurological disorders, metabolic diseases, and cancers. This review summarizes the current knowledge of ARNT2's structure, regulation, interacting partners, and its toxicopathological significance. A better understanding of ARNT2 biology may open new avenues for its characterization as a molecular target and designing novel therapeutic strategies across multiple diseases.
