High-Throughput Transcriptomics Platform for Screening Environmental Chemicals

In vitro to in vivo extrapolation modeling to facilitate the integration of transcriptomics data into genotoxicity assessment

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 (tPoDbiomarkers) 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.

TempO-seq and RNA-seq gene expression levels are highly correlated for most genes: A comparison using 39 human cell lines

Recent advances in transcriptomics technologies allow for whole transcriptome gene expression profiling using targeted sequencing techniques, which is becoming increasingly popular due to logistical ease of data acquisition and analysis. As data from these targeted sequencing platforms accumulates, it is important to evaluate their similarity to traditional whole transcriptome RNA-seq. Thus, we evaluated the comparability of TempO-seq data from cell lysates to traditional RNA-Seq from purified RNA using baseline gene expression profiles. First, two TempO-seq data sets that were generated several months apart at different read depths were compared for six human cell lines. The average Pearson correlation was 0.93 (95% CI: 0.90-0.96) and principal component analysis (PCA) showed that these two TempO-seq data sets were highly reproducible and could be combined. Next, TempO-seq data was compared to RNA-Seq data for 39 human cell lines. The log2 normalized expression data for 19,290 genes within both platforms were well correlated between TempO-seq and RNA-seq (Pearson correlation 0.77, 95% CI: 0.76-0.78), and the majority of genes (15,480 genes, 80%) had concordant gene expression levels. PCA showed a platform divergence, but this was readily resolved by calculating relative log2 expression (RLE) of genes compared to the average expression across cell lines in each platform. Application of gene ontology analysis revealed that ontologies associated with histone and ribosomal functions were enriched for the 20% of genes with non-concordant expression levels (3,810 genes). On the other hand, gene ontologies annotated to cellular structure functions were enriched for genes with concordant expression levels between the platforms. In conclusion, we found TempO-seq baseline expression data to be reproducible at different read depths and found TempO-seq RLE data from lysed cells to be comparable to RNA-seq RLE data from purified RNA across 39 cell lines, even though the datasets were generated by different laboratories using different cell stocks.

Deciphering per- and polyfluoroalkyl substances mode of action: comparative gene expression analysis in human liver spheroids

Understanding the mechanisms by which environmental chemicals cause toxicity is necessary for effective human health risk assessment. High-throughput transcriptomics (HTTr) can be used to inform risk assessment on toxicological mechanisms, hazards, and potencies. We applied HTTr to elucidate the molecular mechanisms by which per- and polyfluoroalkyl substances (PFAS) cause liver perturbations. We contrasted transcriptomic profiles of PFOA, PFBS, PFOS, and PFDS against transcriptomic profiles from established liver-toxic and non-toxic reference compounds, alongside peroxisome proliferator-activated receptors (PPARs) agonists. Our analysis was conducted on metabolically competent 3-D human liver spheroids produced from primary cells from 10 donors. Pathway analysis showed that PFOS and PFDS perturb many of the same pathways as the known liver-toxic compounds in the spheroids, and that the cholesterol biosynthesis pathways are significantly affected by exposure to these compounds. PFOA alters lipid metabolism-related pathways but its expression profile does not closely match reference compounds. PFBS upregulates many degradation-related pathways and targets many of the same pathways as the PPAR agonists and acetaminophen. Our transcriptional analysis does not support the claim that these PFAS are DNA-damaging in this model. A multidimensional scaling (MDS) analysis revealed that PFOS, PFOA, and PFDS cluster together in the same multidimensional space as liver-damaging compounds, whereas PFBS clusters more closely with the non-liver-damaging compounds. Benchmark concentration-response modeling predicts that all the PFAS are bioactive in the liver. Overall, our results show that these PFAS produce unique transcriptional changes but also alter pathways associated with established liver-toxic chemicals in this liver spheroid model.

Integration of new approach methods for the assessment of data-poor chemicals

The use of new approach methods (NAMs), including high-throughput, in vitro bioactivity data, in setting a point-of-departure (POD) will accelerate the pace of human health hazard assessments. Combining hazard and exposure predictions into a bioactivity:exposure ratio (BER) for use in risk-based prioritization and utilizing NAM-based bioactivity flags to indicate potential hazards of interest for further prediction or mechanism-based screening together comprise a prospective approach for management of substances with limited traditional toxicity testing data. In this work, we demonstrate a NAM-based assessment case study conducted via the Accelerating the Pace of Chemical Risk Assessment initiative, a consortium of international research and regulatory scientists. The primary objective was to develop a reusable and adaptable approach for addressing chemicals with limited traditional toxicity data using a NAM-based POD, BER, and bioactivity-based flags for indication of putative endocrine, developmental, neurological, and immunosuppressive effects via data generation and interpretation for 200 substances. Multiple data streams, including in silico and in vitro NAMs, were used. High-throughput transcriptomics and phenotypic profiling data, as well as targeted biochemical and cell-based assays, were combined with generic high-throughput toxicokinetic models parameterized with chemical-specific data to estimate dose for comparison to exposure predictions. This case study further enables regulatory scientists from different international purviews to utilize efficient approaches for prospective chemical management, addressing hazard and risk-based data needs, while reducing the need for animal studies. This work demonstrates the feasibility of using a battery of toxicodynamic and toxicokinetic NAMs to provide a NAM-based POD for screening-level assessment.

TPD-seq: A high throughput RNA-seq method to derive transcriptomic points of departure from cell lines

There is growing scientific and regulatory interest in transcriptomic points of departure (tPOD) values from high-throughput in vitro experiments. To further help democratize tPOD research, here we outline 'TPD-seq' which links microplate-based exposure methods involving cell lines for human (Caco-2, Hep G2) and environmental (rainbow trout RTgill-W1) health, with a commercially available RNA-seq kit, with a cloud-based bioinformatics tool (ExpressAnalyst.ca). We applied the TPD-seq workflow to derive tPODs for solvents (dimethyl sulfoxide, DMSO; methanol) and positive controls (3,4-dichloroaniline, DCA; hydrogen peroxide, H2O2) commonly used in toxicity testing. The majority of reads mapped to protein coding genes (∼9 k for fish cells; ∼6 k for human cells), and about 50 % of differentially expressed genes were curve-fitted from which 90 % yielded gene benchmark doses. The most robust transcriptomic responses were caused by DMSO exposure, and tPOD values were 31-155 mM across the cell lines. OECD test guideline 249 (RTgill-W1 cells) recommends the use of DCA and here we calculated a tPOD of ∼5 to 76 μM. Finally, exposure of the two human cell lines to H2O2 resulted in tPOD values that ranged from 0.7 to 1.1 mM in Caco-2 cells and 5-30 μM in Hep G2 cells. The methods outlined here are designed to be performed in laboratories with basic molecular and cell culture facilities, and the performance and scalability of the TPD-seq workflow can be determined with additional case studies.

Transcriptomic response of an algal species (Raphidocelis subcapitata) exposed to 22 per- and polyfluoroalkyl substances

Per- and polyfluoroalkyl substances (PFAS) are a large class of chemicals of concern for both human and environmental health because of their ubiquitous presence in the environment, persistence, and potential toxicological effects. Despite this, ecological hazard data are limited to a small number of PFAS although there are over 4,000 identified PFAS. Traditional toxicity testing will likely be inadequate to generate necessary hazard information for risk assessment. Therefore, this study investigated the utility of using transcriptomic points of departure (tPODs) for informing PFAS algal toxicity. Raphidocelis subcapitata, a freshwater green algal species, were exposed for 24 hr in 96-well microplates to multiple concentrations of 22 different PFAS. Following exposure, RNA was extracted, and the transcriptome was evaluated by RNA sequencing followed by concentration response modeling to determine a tPOD for each PFAS. Per- and polyfluoroalkyl substance tPODs, based on measured concentrations, ranged from 0.9 µg/L for perfluorotridecanoic acid to 1 mg/L for perfluorononanoic acid. These values derived from R. subcapitata exposures were compared with published hazard benchmarks from other taxa (larval fathead minnow and Daphnia magna) and in vitro data. Although R. subcapitata was generally more sensitive to the tested PFAS than previously tested taxa and in vitro assays, the algal tPODs were, on average, three orders magnitude greater than the maximum concentrations of PFAS detected in Great Lakes tributaries. This high throughput transcriptomics assay with algae is a promising new approach method for an ecologically relevant tiered hazard evaluation strategy.

Gestational exposure to polystyrene microplastics incurred placental damage in mice: Insights into metabolic and gene expression disorders

As an emerging environmental pollutant, microplastics have attracted increasing attention to their potential health hazards. However, the current understanding about the toxicity and health implications, especially about developmental toxicity with exposure to microplastics is quite limited. In the current study, we aimed to scrutinize the deleterious effects of polystyrene microplastics (PSMPs) with different sizes (0.1 and 5 μm) on the placenta that plays crucial role in fetal development, following oral exposure during gestational stages. The results showed that two sizes of PSMPs could distribute in mouse placental tissues, and nanosized PSMPs (0.1 μm) exhibited greater capability to penetrate the placenta and deposit in the liver and brain of fetuses than microsized PSMPs (5 μm). Importantly, only 0.1 μm PSMPs induced a decrease in the junctional area, a reduction in the labyrinthine vascularization and an increase in cell apoptosis in the placenta, accompanied by fetal developmental impairments. The results of metabolome and transcriptome uncovered that 0.1 μm PSMP exposure caused changes in metabolic and gene profiles of placental tissues, across multiple pathways such as vascular supply, nutrient absorption and transportation and amino acid metabolism. Overall, our results confirmed that maternal PSMP exposure led to placental damages associated with metabolic and gene expression disorders. This study would provide new insights into the developmental impacts of microplastic consumption during gestation.

Novel Approach to Screen Endocrine-Disrupting Chemicals via Endocrine-Enhanced Reduced Human Transcriptome

Endocrine-disrupting chemicals (EDCs) can interfere with multiple pathways and trigger different modes of action. Thus, the traditional EDC in vitro screening processes often require a battery of bioassays to cover multiple target pathways. Here we developed an endocrine-enhanced reduced human transcriptome (ERHT) focused on hormone receptor signaling induced by the EDCs regulating specific genes. ERHT was developed based on 1200 prioritized genes covering 110 endocrine-related biological pathways across eight potential adverse outcomes. The ability of this approach to identify EDCs was derived from machine learning of 1068 dose-dependent transcriptome profiles and enhanced by quantifying chemical-induced critical pathway responses, and thus, it demonstrated excellent classification performance (AUC = 0.84 ± 0.03) in internal cross-validation. We ultimately applied this approach to known EDCs and inactive substances to validate the reliability of this approach. Through external validation on 210 chemicals, the extrapolation accuracy exceeded 80%, demonstrating the outstanding practical performance of this approach. Meanwhile, the pathway responses induced by the same chemical were consistent with the experimental results reported by multiple sequencing platforms, highlighting the robustness of this approach. The above results demonstrate that this approach can provide novel insights for EDCs' high-throughput screening and comprehensive toxic mechanisms through biological pathways.

Gene Set Enrichment Analysis in Zebrafish Embryos Is Susceptible to False-Positive Results in the Absence of Differentially Expressed Genes

High-throughput gene expression studies commonly employ pathway analyses to infer biological meaning from lists of differentially expressed genes (DEGs). In toxicology and pharmacology studies, treatment groups are analysed against vehicle controls to identify DEGs and altered pathways. Previously, we empirically quantified false-positive rates of DEGs in gene expression data from pools of vehicle-treated zebrafish embryos to determine appropriate study designs (sample and pool size). Here, the same data were subject to Over-Representation Analysis (ORA) and Gene Set Enrichment Analysis (GSEA) to identify false-positive enriched pathways. As expected, the number of false-positive ORA results was lowest where pool and sample sizes were largest (conditions which also generated the fewest significant DEGs). In contrast, the frequency of GSEA false-positives generated through the fast GSEA (fgsea) algorithm increased with pool and sample size and was highest for simulations that generated 0 DEGs, with ribosomal gene sets significantly enriched with the highest frequency. We describe 2 distinct mechanisms by which GSEA generated these false-positive results, both of which are most likely to generate significant gene sets under conditions where expression differences are particularly low. Finally, GSEA analyses were repeated using 1 alternative GSEA algorithm (CERNO) and 11 different ranking statistics. In almost every analysis, the number of significant results was highest where pool size was highest, with ribosome as the more frequently enriched gene set, suggesting our observations to be generalizable to different implementations of GSEA. These results from zebrafish embryos suggest caution in interpreting any GSEA results in contrasts where there are no DEGs.

Biological activity of natural 2-quinolinones

While natural products have undeniably played a crucial role in drug discovery, challenges such as limited availability and complex synthesis methods have hindered the identification of lead compounds. At the core of numerous natural and synthetic compounds, each displaying distinct biological behaviours, lies the foundational structure of 2-quinolinone. Compounds with this structural motif exhibit a broad range of effects in different tissues. Furthermore, specific members showcase therapeutic potential for various disorders. Despite the significance of these compounds, the current review literature has not provided a comprehensive overview, underscoring the essential contribution of this article in exploring their biological functions. This study examines the biological activity of selected 2-quinolinone alkaloids across diverse organisms, unveiling their potential as a source of innovative bioactive natural products.

Bioinformatic workflows for deriving transcriptomic points of departure: current status, data gaps, and research priorities

There is a pressing need to increase the efficiency and reliability of toxicological safety assessment for protecting human health and the environment. Although conventional toxicology tests rely on measuring apical changes in vertebrate models, there is increasing interest in the use of molecular information from animal and in vitro studies to inform safety assessment. One promising and pragmatic application of molecular information involves the derivation of transcriptomic points of departure (tPODs). Transcriptomic analyses provide a snapshot of global molecular changes that reflect cellular responses to stressors and progression toward disease. A tPOD identifies the dose level below which a concerted change in gene expression is not expected in a biological system in response to a chemical. A common approach to derive such a tPOD consists of modeling the dose-response behavior for each gene independently and then aggregating the gene-level data into a single tPOD. Although different implementations of this approach are possible, as discussed in this manuscript, research strongly supports the overall idea that reference doses produced using tPODs are health protective. An advantage of this approach is that tPODs can be generated in shorter term studies (e.g. days) compared with apical endpoints from conventional tests (e.g. 90-d subchronic rodent tests). Moreover, research strongly supports the idea that reference doses produced using tPODs are health protective. Given the potential application of tPODs in regulatory toxicology testing, rigorous and reproducible wet and dry laboratory methodologies for their derivation are required. This review summarizes the current state of the science regarding the study design and bioinformatics workflows for tPOD derivation. We identify standards of practice and sources of variability in tPOD generation, data gaps, and areas of uncertainty. We provide recommendations for research to address barriers and promote adoption in regulatory decision making.

Progress in toxicogenomics to protect human health

Toxicogenomics measures molecular features, such as transcripts, proteins, metabolites and epigenomic modifications, to understand and predict the toxicological effects of environmental and pharmaceutical exposures. Transcriptomics has become an integral tool in contemporary toxicology research owing to innovations in gene expression profiling that can provide mechanistic and quantitative information at scale. These data can be used to predict toxicological hazards through the use of transcriptomic biomarkers, network inference analyses, pattern-matching approaches and artificial intelligence. Furthermore, emerging approaches, such as high-throughput dose-response modelling, can leverage toxicogenomic data for human health protection even in the absence of predicting specific hazards. Finally, single-cell transcriptomics and multi-omics provide detailed insights into toxicological mechanisms. Here, we review the progress since the inception of toxicogenomics in applying transcriptomics towards toxicology testing and highlight advances that are transforming risk assessment.

Cell Painting for cytotoxicity and mode-of-action analysis in primary human hepatocytes

High-throughput, human-relevant approaches for predicting chemical toxicity are urgently needed for better decision-making in human health. Here, we apply image-based profiling (the Cell Painting assay) and two cytotoxicity assays (metabolic and membrane damage readouts) to primary human hepatocytes after exposure to eight concentrations of 1085 compounds that include pharmaceuticals, pesticides, and industrial chemicals with known liver toxicity-related outcomes. Three computational methods (CellProfiler, a Cell Painting-specific convolutional neural network, and a pretrained vision transformer) were compared to extract morphology features from single cells or entire images. We used these morphology features to predict activity in the measured cytotoxicity assays, as well as in 412 curated ToxCast assays that span cytotoxicity, cell-based, and cell-free categories. We found that the morphological profiles detect compound bioactivity at lower concentrations than standard cytotoxicity assays. In supervised analyses, they predict cytotoxicity and targeted cell-based assay readouts, but not cell-free assay readouts. We also found that the various feature extraction methods performed relatively similarly and that filtering out non-bioactive or cytotoxic concentrations did not boost supervised assay prediction performance for any assay endpoint category, although it did have a large influence on unsupervised cluster analysis. We envision that image-based profiling could serve as a key component of modern safety assessment.

Bioactivity Profiling of Chemical Mixtures for Hazard Characterization

The assessment and regulation of chemical toxicity to protect human health and the environment are done one chemical at a time and seldom at environmentally relevant concentrations. However, chemicals are found in the environment as mixtures, and their toxicity is largely unknown. Understanding the hazard posed by chemicals within the mixture is critical to enforce protective measures. Here, we demonstrate the application of bioactivity profiling of environmental water samples using the sentinel and ecotoxicology model species Daphnia to reveal the biomolecular response induced by exposure to real-world mixtures. We exposed a Daphnia strain to 30 sampled waters of the Chaobai River and measured the gene expression response profiles. Using a multiblock correlation analysis, we establish correlations between chemical mixtures identified in 30 water samples with gene expression patterns induced by these chemical mixtures. We identified 80 metabolic pathways putatively activated by mixtures of inorganic ions, heavy metals, polycyclic aromatic hydrocarbons, industrial chemicals, and a set of biocides, pesticides, and pharmacologically active substances. Our data-driven approach discovered both known bioactivity signatures with previously described modes of action and new pathways linked to undiscovered potential hazards. This study demonstrates the feasibility of reducing the complexity of real-world mixture toxicity to characterize the biomolecular effects of a defined number of chemical components based on gene expression monitoring of the sentinel species Daphnia.

Bayesian gene set benchmark dose estimation for "omic" responses

MOTIVATION

Estimating a toxic reference point using tools like the benchmark dose (BMD) is a critical step in setting policy to regulate pollution and ensure safe environments. Toxicity can be measured for different endpoints, including changes in gene expression and histopathology for various tissues, and is typically explored one gene or tissue at a time in a univariate setting that ignores correlation. In this work, we develop a multivariate estimation procedure to estimate the BMD for specified gene sets. Our approach extends the foundational univariate approach by accounting for correlation in a statistically principled way.

RESULTS

We illustrate the method using data from a 5-day rat study and Hallmark gene sets and compare to existing BMD results computed by the EPA for both gene sets and apical histopathology endpoints. In contrast to previous ad-hoc methods, our principled approach provides the needed extension to bring the foundational univariate method into the multivariate world of transcriptomics. In addition to use in a regulatory setting, our method can provide hypothesis generation when gene sets correspond to mechanistic pathways.

AVAILABILITY AND IMPLEMENTATION

BS-BMD is implemented in R and C++ and available at https://github.com/NIEHS/BS-BMD.

Outcome of IWGT workshop on transcriptomic biomarkers for genotoxicity: Key considerations for bioinformatics

As a part of the International Workshop on Genotoxicity Testing (IWGT) in 2022, a workgroup was formed to evaluate the level of validation and regulatory acceptance of transcriptomic biomarkers that identify genotoxic substances. Several such biomarkers have been developed using various molecular techniques and computational approaches. Within the IWGT workgroup on transcriptomic biomarkers, bioinformatics was a central topic of discussion, focusing on the current approaches used to process the underlying molecular data to distill a reliable predictive signal; that is, a gene set that is indicative of genotoxicity and can then be used in later studies to predict potential DNA damaging properties for uncharacterized chemicals. While early studies used microarray data, a technological shift occurred in the past decade to incorporate modern transcriptome measuring techniques such as high-throughput transcriptomics, which in turn is based on high-throughput sequencing. Herein, we present the workgroup's review of the current bioinformatic approaches to identify genes comprising transcriptomic biomarkers. Within the context of regulatory toxicology, the reproducibility of a given analysis is critical. Therefore, the workgroup provides consensus recommendations on how to facilitate sufficient reporting of experimental parameters for the analytical procedures used in a transcriptomic biomarker study, including the recommendation to develop a biomarker-specific reporting module within the OECD Omics Reporting Framework.

Signature analysis of high-throughput transcriptomics screening data for mechanistic inference and chemical grouping

High-throughput transcriptomics (HTTr) uses gene expression profiling to characterize the biological activity of chemicals in in vitro cell-based test systems. As an extension of a previous study testing 44 chemicals, HTTr was used to screen an additional 1,751 unique chemicals from the EPA's ToxCast collection in MCF7 cells using 8 concentrations and an exposure duration of 6 h. We hypothesized that concentration-response modeling of signature scores could be used to identify putative molecular targets and cluster chemicals with similar bioactivity. Clustering and enrichment analyses were conducted based on signature catalog annotations and ToxPrint chemotypes to facilitate molecular target prediction and grouping of chemicals with similar bioactivity profiles. Enrichment analysis based on signature catalog annotation identified known mechanisms of action (MeOAs) associated with well-studied chemicals and generated putative MeOAs for other active chemicals. Chemicals with predicted MeOAs included those targeting estrogen receptor (ER), glucocorticoid receptor (GR), retinoic acid receptor (RAR), the NRF2/KEAP/ARE pathway, AP-1 activation, and others. Using reference chemicals for ER modulation, the study demonstrated that HTTr in MCF7 cells was able to stratify chemicals in terms of agonist potency, distinguish ER agonists from antagonists, and cluster chemicals with similar activities as predicted by the ToxCast ER Pathway model. Uniform manifold approximation and projection (UMAP) embedding of signature-level results identified novel ER modulators with no ToxCast ER Pathway model predictions. Finally, UMAP combined with ToxPrint chemotype enrichment was used to explore the biological activity of structurally related chemicals. The study demonstrates that HTTr can be used to inform chemical risk assessment by determining in vitro points of departure, predicting chemicals' MeOA and grouping chemicals with similar bioactivity profiles.

Time-course characterization of whole-transcriptome dynamics of HepG2/C3A spheroids and its toxicological implications

Physiologically relevant in vitro models are a priority in predictive toxicology to replace and/or reduce animal experiments. The compromised toxicant metabolism of many immortalized human liver cell lines grown as monolayers as compared to in vivo metabolism limits their physiological relevance. However, recent efforts to culture liver cells in a 3D environment, such as spheroids, to better mimic the in vivo conditions, may enhance the toxicant metabolism of human liver cell lines. In this study, we characterized the dynamic changes in the transcriptome of HepG2/C3A hepatocarcinoma cell spheroids maintained in a clinostat system (CelVivo) to gain insight into the metabolic capacity of this model as a function of spheroid size and culture time. We assessed morphological changes (size, necrotic core), cell health, and proliferation rate from initial spheroid seeding to 35 days of continuous culture in conjunction with a time-course (0, 3, 7, 10, 14, 21, 28 days) of the transcriptome (TempO-Seq, BioSpyder). The phenotypic characteristics of HepG2/C3A growing in spheroids were comparable to monolayer growth until ∼Day 12 (Day 10-14) when a significant decrease in cell doubling rate was noted which was concurrent with down-regulation of cell proliferation and cell cycle pathways over this time period. Principal component analysis of the transcriptome data suggests that the Day 3, 7, and 10 spheroids are pronouncedly different from the Day 14, 21, and 28 spheroids in support of a biological transition time point during the long-term 3D spheroid cultures. The expression of genes encoding cellular components involved in toxicant metabolism and transport rapidly increased during the early time points of spheroids to peak at Day 7 or Day 10 as compared to monolayer cultures with a gradual decrease in expression with further culture, suggesting the most metabolically responsive time window for exposure studies. Overall, we provide baseline information on the cellular and molecular characterization, with a particular focus on toxicant metabolic capacity dynamics and cell growth, of HepG2/C3A 3D spheroid cultures over time.

A new approach methodology to identify tumorigenic chemicals using short-term exposures and transcript profiling

Current methods for cancer risk assessment are resource-intensive and not feasible for most of the thousands of untested chemicals. In earlier studies, we developed a new approach methodology (NAM) to identify liver tumorigens using gene expression biomarkers and associated tumorigenic activation levels (TALs) after short-term exposures in rats. The biomarkers are used to predict the six most common rodent liver cancer molecular initiating events. In the present study, we wished to confirm that our approach could be used to identify liver tumorigens at only one time point/dose and if the approach could be applied to (targeted) RNA-Seq analyses. Male rats were exposed for 4 days by daily gavage to 15 chemicals at doses with known chronic outcomes and liver transcript profiles were generated using Affymetrix arrays. Our approach had 75% or 85% predictive accuracy using TALs derived from the TG-GATES or DrugMatrix studies, respectively. In a dataset generated from the livers of male rats exposed to 16 chemicals at up to 10 doses for 5 days, we found that our NAM coupled with targeted RNA-Seq (TempO-Seq) could be used to identify tumorigenic chemicals with predictive accuracies of up to 91%. Overall, these results demonstrate that our NAM can be applied to both microarray and (targeted) RNA-Seq data generated from short-term rat exposures to identify chemicals, their doses, and mode of action that would induce liver tumors, one of the most common endpoints in rodent bioassays.

New approach methodologies (NAMs) for the in vitro assessment of cleaning products for respiratory irritation: workshop report

The use of in vitro new approach methodologies (NAMs) to assess respiratory irritation depends on several factors, including the specifics of exposure methods and cell/tissue-based test systems. This topic was examined in the context of human health risk assessment for cleaning products at a 1-day public workshop held on 2 March 2023, organized by the American Cleaning Institute® (ACI). The goals of this workshop were to (1) review in vitro NAMs for evaluation of respiratory irritation, (2) examine different perspectives on current challenges and suggested solutions, and (3) publish a manuscript of the proceedings. Targeted sessions focused on exposure methods, in vitro cell/tissue test systems, and application to human health risk assessment. The importance of characterization of assays and development of reporting standards was noted throughout the workshop. The exposure methods session emphasized that the appropriate exposure system design depends on the purpose of the assessment. This is particularly important given the many dosimetry and technical considerations affecting relevance and translation of results to human exposure scenarios. Discussion in the in vitro cell/tissue test systems session focused on the wide variety of cell systems with varying suitability for evaluating key mechanistic steps, such as molecular initiating events (MIEs) and key events (KEs) likely present in any putative respiratory irritation adverse outcome pathway (AOP). This suggests the opportunity to further develop guidance around in vitro cell/tissue test system endpoint selection, assay design, characterization and validation, and analytics that provide information about a given assay's utility. The session on applications for human health protection emphasized using mechanistic understanding to inform the choice of test systems and integration of NAMs-derived data with other data sources (e.g., physicochemical properties, exposure information, and existing in vivo data) as the basis for in vitro to in vivo extrapolation. In addition, this group noted a need to develop procedures to align NAMs-based points of departure (PODs) and uncertainty factor selection with current human health risk assessment methods, together with consideration of elements unique to in vitro data. Current approaches are described and priorities for future characterization of in vitro NAMs to assess respiratory irritation are noted.

High-Throughput Transcriptomics Screen of ToxCast Chemicals in U-2 OS Cells

New approach methodologies (NAMs) aim to accelerate the pace of chemical risk assessment while simultaneously reducing cost and dependency on animal studies. High Throughput Transcriptomics (HTTr) is an emerging NAM in the field of chemical hazard evaluation for establishing in vitro points-of-departure and providing mechanistic insight. In the current study, 1201 test chemicals were screened for bioactivity at eight concentrations using a 24-h exposure duration in the human- derived U-2 OS osteosarcoma cell line with HTTr. Assay reproducibility was assessed using three reference chemicals that were screened on every assay plate. The resulting transcriptomics data were analyzed by aggregating signal from genes into signature scores using gene set enrichment analysis, followed by concentration-response modeling of signatures scores. Signature scores were used to predict putative mechanisms of action, and to identify biological pathway altering concentrations (BPACs). BPACs were consistent across replicates for each reference chemical, with replicate BPAC standard deviations as low as 5.6 × 10-3 μM, demonstrating the internal reproducibility of HTTr-derived potency estimates. BPACs of test chemicals showed modest agreement (R2 = 0.55) with existing phenotype altering concentrations from high throughput phenotypic profiling using Cell Painting of the same chemicals in the same cell line. Altogether, this HTTr based chemical screen contributes to an accumulating pool of publicly available transcriptomic data relevant for chemical hazard evaluation and reinforces the utility of cell based molecular profiling methods in estimating chemical potency and predicting mechanism of action across a diverse set of chemicals.

Assessment of TGx-DDI genes for genotoxicity in a comprehensive panel of chemicals

BACKGROUND

The TGx-DDI biomarker identifies transcripts specifically induced by primary DNA damage. Profiling similarity of TGx-DDI signatures can allow clustering compounds by genotoxic mechanism. This transcriptomics-based approach complements conventional toxicology testing by enhancing mechanistic resolution.

METHODS

Unsupervised hierarchical clustering and t-distributed stochastic neighbor embedding (tSNE) were utilized to assess similarity of publicly-available per- and polyfluoroalkyl substances (PFAS) and ToxCast chemicals based on TGx-DDI modulation. TempO-seq transcriptomic data after highest chemical concentrations were analyzed.

RESULTS

Clustering discriminated between genotoxic and non-genotoxic compounds while drawing similarity among chemicals with shared mechanisms. PFAS largely clustered distinctly from classical mutagens. However, dynamic range across PFAS types and durations indicated variable potential for DNA damage. tSNE visualization reinforced phenotypic groupings, with genotoxins clustering separately from non-DNA damaging agents.

DISCUSSION

Unsupervised learning approaches applied to TGx-DDI profiles effectively categorizes chemical genotoxicity potential, aiding elucidation of biological response pathways. This transcriptomics-based strategy gives further insight into the role and effect of individual TGx-DDI biomarker genes and complements existing assays by enhancing mechanistic resolution. Overall, TGx-DDI biomarker profiling holds promise for predictive safety screening.

Engineering a computable epiblast for in silico modeling of developmental toxicity

Developmental hazard evaluation is an important part of assessing chemical risks during pregnancy. Toxicological outcomes from prenatal testing in pregnant animals result from complex chemical-biological interactions, and while New Approach Methods (NAMs) based on in vitro bioactivity profiles of human cells offer promising alternatives to animal testing, most of these assays lack cellular positional information, physical constraints, and regional organization of the intact embryo. Here, we engineered a fully computable model of the embryonic disc in the CompuCell3D.org modeling environment to simulate epithelial-mesenchymal transition (EMT) of epiblast cells and self-organization of mesodermal domains (chordamesoderm, paraxial, lateral plate, posterior/extraembryonic). Mesodermal fate is modeled by synthetic activity of the BMP4-NODAL-WNT signaling axis. Cell position in the epiblast determines timing with respect to EMT for 988 computational cells in the computer model. An autonomous homeobox (Hox) clock hidden in the epiblast is driven by WNT-FGF4-CDX signaling. Executing the model renders a quantitative cell-level computation of mesodermal fate and consequences of perturbation based on known biology. For example, synthetic perturbation of the control network rendered altered phenotypes (cybermorphs) mirroring some aspects of experimental mouse embryology, with electronic knockouts, under-activation (hypermorphs) or over-activation (hypermorphs) particularly affecting the size and specification of the posterior mesoderm. This foundational model is trained on embryology but capable of performing a wide variety of toxicological tasks conversing through anatomical simulation to integrate in vitro chemical bioactivity data with known embryology. It is amenable to quantitative simulation for probabilistic prediction of early developmental toxicity.

Transcriptome analysis in mouse skin after exposure to ultraviolet radiation from a canopy sunbed

Exposure to ultraviolet radiation (UV-R), from both natural and artificial tanning, heightens the risk of skin cancer by inducing molecular changes in cells and tissues. Despite established transcriptional alterations at a molecular level due to UV-R exposure, uncertainties persist regarding UV radiation characterization and subsequent genomic changes. Our study aimed to mechanistically explore dose- and time-dependent gene expression changes, that may drive short-term (e.g., sunburn) and long-term actinic (e.g., skin cancer) consequences. Using C57BL/6N mouse skin, we analyzed transcriptomic expression following exposure to five erythemally weighted UV-R doses (0, 5, 10, 20, and 40 mJ/cm2) emitted by a UV-R tanning device. At 96 h post-exposure, 5 mJ/cm2 induced 116 statistically significant differentially expressed genes (DEGs) associated with structural changes from UV-R damage. The highest number of significant gene expression changes occurred at 6 and 48 h post-exposure in the 20 and 40 mJ/cm2 dose groups. Notably, at 40 mJ/cm2, 13 DEGs related to skin barrier homeostasis were consistently perturbed across all timepoints. UV-R exposure activated pathways involving oxidative stress, P53 signaling, inflammation, biotransformation, skin barrier maintenance, and innate immunity. This in vivo study's transcriptional data offers mechanistic insights into both short-term and potential non-threshold-dependent long-term health effects of UV-R tanning.

Community-Engaged Research and the Use of Open Access ToxVal/ToxRef In Vivo Databases and New Approach Methodologies (NAM) to Address Human Health Risks From Environmental Contaminants

BACKGROUND

The paper analyzes opportunities for integrating Open access resources (Abstract Sifter, US EPA and NTP Toxicity Value and Toxicity Reference [ToxVal/ToxRefDB]) and New Approach Methodologies (NAM) integration into Community Engaged Research (CEnR).

METHODS

CompTox Chemicals Dashboard and Integrated Chemical Environment with in vivo ToxVal/ToxRef and NAMs (in vitro) databases are presented in three case studies to show how these resources could be used in Pilot Projects involving Community Engaged Research (CEnR) from the University of California, Davis, Environmental Health Sciences Center.

RESULTS

Case #1 developed a novel assay methodology for testing pesticide toxicity. Case #2 involved detection of water contaminants from wildfire ash and Case #3 involved contaminants on Tribal Lands. Abstract Sifter/ToxVal/ToxRefDB regulatory data and NAMs could be used to screen/prioritize risks from exposure to metals, PAHs and PFAS from wildfire ash leached into water and to investigate activities of environmental toxins (e.g., pesticides) on Tribal lands. Open access NAMs and computational tools can apply to detection of sensitive biological activities in potential or known adverse outcome pathways to predict points of departure (POD) for comparison with regulatory values for hazard identification. Open access Systematic Empirical Evaluation of Models or biomonitoring exposures are available for human subpopulations and can be used to determine bioactivity (POD) to exposure ratio to facilitate mitigation.

CONCLUSIONS

These resources help prioritize chemical toxicity and facilitate regulatory decisions and health protective policies that can aid stakeholders in deciding on needed research. Insights into exposure risks can aid environmental justice and health equity advocates.

Informing Hazard Identification and Risk Characterization of Environmental Chemicals by Combining Transcriptomic and Functional Data from Human-Induced Pluripotent Stem-Cell-Derived Cardiomyocytes

Environmental chemicals may contribute to the global burden of cardiovascular disease, but experimental data are lacking to determine which substances pose the greatest risk. Human-induced pluripotent stem cell (iPSC)-derived cardiomyocytes are a high-throughput cardiotoxicity model that is widely used to test drugs and chemicals; however, most studies focus on exploring electro-physiological readouts. Gene expression data may provide additional molecular insights to be used for both mechanistic interpretation and dose-response analyses. Therefore, we hypothesized that both transcriptomic and functional data in human iPSC-derived cardiomyocytes may be used as a comprehensive screening tool to identify potential cardiotoxicity hazards and risks of the chemicals. To test this hypothesis, we performed concentration-response analysis of 464 chemicals from 12 classes, including both pharmaceuticals and nonpharmaceutical substances. Functional effects (beat frequency, QT prolongation, and asystole), cytotoxicity, and whole transcriptome response were evaluated. Points of departure were derived from phenotypic and transcriptomic data, and risk characterization was performed. Overall, 244 (53%) substances were active in at least one phenotype; as expected, pharmaceuticals with known cardiac liabilities were the most active. Positive chronotropy was the functional phenotype activated by the largest number of tested chemicals. No chemical class was particularly prone to pose a potential hazard to cardiomyocytes; a varying proportion (10-44%) of substances in each class had effects on cardiomyocytes. Transcriptomic data showed that 69 (15%) substances elicited significant gene expression changes; most perturbed pathways were highly relevant to known key characteristics of human cardiotoxicants. The bioactivity-to-exposure ratios showed that phenotypic- and transcriptomic-based POD led to similar results for risk characterization. Overall, our findings demonstrate how the integrative use of in vitro transcriptomic and phenotypic data from iPSC-derived cardiomyocytes not only offers a complementary approach for hazard and risk prioritization, but also enables mechanistic interpretation of the in vitro test results to increase confidence in decision-making.

Prediction of Endocrine-Disrupting Chemicals Related to Estrogen, Androgen, and Thyroid Hormone (EAT) Modalities Using Transcriptomics Data and Machine Learning

Endocrine-disrupting chemicals (EDCs) are chemicals that can interfere with homeostatic processes. They are a major concern for public health, and they can cause adverse long-term effects such as cancer, intellectual impairment, obesity, diabetes, and male infertility. The endocrine system is a complex machinery, with the estrogen (E), androgen (A), and thyroid hormone (T) modes of action being of major importance. In this context, the availability of in silico models for the rapid detection of hazardous chemicals is an effective contribution to toxicological assessments. We developed Qualitative Gene expression Activity Relationship (QGexAR) models to predict the propensities of chemically induced disruption of EAT modalities. We gathered gene expression profiles from the LINCS database tested on two cell lines, i.e., MCF7 (breast cancer) and A549 (adenocarcinomic human alveolar basal epithelial). We optimized our prediction protocol by testing different feature selection methods and classification algorithms, including CATBoost, XGBoost, Random Forest, SVM, Logistic regression, AutoKeras, TPOT, and deep learning models. For each EAT endpoint, the final prediction was made according to a consensus prediction as a function of the best model obtained for each cell line. With the available data, we were able to develop a predictive model for estrogen receptor and androgen receptor binding and thyroid hormone receptor antagonistic effects with a consensus balanced accuracy on a validation set ranging from 0.725 to 0.840. The importance of each predictive feature was further assessed to identify known genes and suggest new genes potentially involved in the mechanisms of action of EAT perturbation.

A strategy to detect metabolic changes induced by exposure to chemicals from large sets of condition-specific metabolic models computed with enumeration techniques

BACKGROUND

The growing abundance of in vitro omics data, coupled with the necessity to reduce animal testing in the safety assessment of chemical compounds and even eliminate it in the evaluation of cosmetics, highlights the need for adequate computational methodologies. Data from omics technologies allow the exploration of a wide range of biological processes, therefore providing a better understanding of mechanisms of action (MoA) related to chemical exposure in biological systems. However, the analysis of these large datasets remains difficult due to the complexity of modulations spanning multiple biological processes.

RESULTS

To address this, we propose a strategy to reduce information overload by computing, based on transcriptomics data, a comprehensive metabolic sub-network reflecting the metabolic impact of a chemical. The proposed strategy integrates transcriptomic data to a genome scale metabolic network through enumeration of condition-specific metabolic models hence translating transcriptomics data into reaction activity probabilities. Based on these results, a graph algorithm is applied to retrieve user readable sub-networks reflecting the possible metabolic MoA (mMoA) of chemicals. This strategy has been implemented as a three-step workflow. The first step consists in building cell condition-specific models reflecting the metabolic impact of each exposure condition while taking into account the diversity of possible optimal solutions with a partial enumeration algorithm. In a second step, we address the challenge of analyzing thousands of enumerated condition-specific networks by computing differentially activated reactions (DARs) between the two sets of enumerated possible condition-specific models. Finally, in the third step, DARs are grouped into clusters of functionally interconnected metabolic reactions, representing possible mMoA, using the distance-based clustering and subnetwork extraction method. The first part of the workflow was exemplified on eight molecules selected for their known human hepatotoxic outcomes associated with specific MoAs well described in the literature and for which we retrieved primary human hepatocytes transcriptomic data in Open TG-GATEs. Then, we further applied this strategy to more precisely model and visualize associated mMoA for two of these eight molecules (amiodarone and valproic acid). The approach proved to go beyond gene-based analysis by identifying mMoA when few genes are significantly differentially expressed (2 differentially expressed genes (DEGs) for amiodarone), bringing additional information from the network topology, or when very large number of genes were differentially expressed (5709 DEGs for valproic acid). In both cases, the results of our strategy well fitted evidence from the literature regarding known MoA. Beyond these confirmations, the workflow highlighted potential other unexplored mMoA.

CONCLUSION

The proposed strategy allows toxicology experts to decipher which part of cellular metabolism is expected to be affected by the exposition to a given chemical. The approach originality resides in the combination of different metabolic modelling approaches (constraint based and graph modelling). The application to two model molecules shows the strong potential of the approach for interpretation and visual mining of complex omics in vitro data. The presented strategy is freely available as a python module ( https://pypi.org/project/manamodeller/ ) and jupyter notebooks ( https://github.com/LouisonF/MANA ).

Using Zebrafish to Screen Developmental Toxicity of Per- and Polyfluoroalkyl Substances (PFAS)

Per- and polyfluoroalkyl substances (PFAS) are found in many consumer and industrial products. While some PFAS, notably perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS), are developmentally toxic in mammals, the vast majority of PFAS have not been evaluated for developmental toxicity potential. A concentration-response study of 182 unique PFAS chemicals using the zebrafish medium-throughput, developmental vertebrate toxicity assay was conducted to investigate chemical structural identifiers for toxicity. Embryos were exposed to each PFAS compound (≤100 μM) beginning on the day of fertilization. At 6 days post-fertilization (dpf), two independent observers graded developmental landmarks for each larva (e.g., mortality, hatching, swim bladder inflation, edema, abnormal spine/tail, or craniofacial structure). Thirty percent of the PFAS were developmentally toxic, but there was no enrichment of any OECD structural category. PFOS was developmentally toxic (benchmark concentration [BMC] = 7.48 μM); however, other chemicals were more potent: perfluorooctanesulfonamide (PFOSA), N-methylperfluorooctane sulfonamide (N-MeFOSA), ((perfluorooctyl)ethyl)phosphonic acid, perfluoro-3,6,9-trioxatridecanoic acid, and perfluorohexane sulfonamide. The developmental toxicity profile for these more potent PFAS is largely unexplored in mammals and other species. Based on these zebrafish developmental toxicity results, additional screening may be warranted to understand the toxicity profile of these chemicals in other species.

Gene-environment interactions within a precision environmental health framework

Understanding the complex interplay of genetic and environmental factors in disease etiology and the role of gene-environment interactions (GEIs) across human development stages is important. We review the state of GEI research, including challenges in measuring environmental factors and advantages of GEI analysis in understanding disease mechanisms. We discuss the evolution of GEI studies from candidate gene-environment studies to genome-wide interaction studies (GWISs) and the role of multi-omics in mediating GEI effects. We review advancements in GEI analysis methods and the importance of large-scale datasets. We also address the translation of GEI findings into precision environmental health (PEH), showcasing real-world applications in healthcare and disease prevention. Additionally, we highlight societal considerations in GEI research, including environmental justice, the return of results to participants, and data privacy. Overall, we underscore the significance of GEI for disease prediction and prevention and advocate for integrating the exposome into PEH omics studies.

House dust-derived mixtures of organophosphate esters alter the phenotype, function, transcriptome, and lipidome of KGN human ovarian granulosa cells

Organophosphate esters (OPEs), used as flame retardants and plasticizers, are present ubiquitously in the environment. Previous studies suggest that exposure to OPEs is detrimental to female fertility in humans. However, no experimental information is available on the effects of OPE mixtures on ovarian granulosa cells, which play essential roles in female reproduction. We used high-content imaging to investigate the effects of environmentally relevant OPE mixtures on KGN human granulosa cell phenotypes. Perturbations to steroidogenesis were assessed using ELISA and qRT-PCR. A high-throughput transcriptomic approach, TempO-Seq, was used to identify transcriptional changes in a targeted panel of genes. Effects on lipid homeostasis were explored using a cholesterol assay and global lipidomic profiling. OPE mixtures altered multiple phenotypic features of KGN cells, with triaryl OPEs in the mixture showing higher potencies than other mixture components. The mixtures increased basal production of steroid hormones; this was mediated by significant changes in the expression of critical transcripts involved in steroidogenesis. Further, the total-OPE mixture disrupted cholesterol homeostasis and the composition of intracellular lipid droplets. Exposure to complex mixtures of OPEs, similar to those found in house dust, may adversely affect female reproductive health by altering a multitude of phenotypic and functional endpoints in granulosa cells. This study provides novel insights into the mechanisms of actions underlying the toxicity induced by OPEs and highlights the need to examine the effects of human relevant chemical mixtures.

Metabolomics Simultaneously Derives Benchmark Dose Estimates and Discovers Metabolic Biotransformations in a Rat Bioassay

Benchmark dose (BMD) modeling estimates the dose of a chemical that causes a perturbation from baseline. Transcriptional BMDs have been shown to be relatively consistent with apical end point BMDs, opening the door to using molecular BMDs to derive human health-based guidance values for chemical exposure. Metabolomics measures the responses of small-molecule endogenous metabolites to chemical exposure, complementing transcriptomics by characterizing downstream molecular phenotypes that are more closely associated with apical end points. The aim of this study was to apply BMD modeling to in vivo metabolomics data, to compare metabolic BMDs to both transcriptional and apical end point BMDs. This builds upon our previous application of transcriptomics and BMD modeling to a 5-day rat study of triphenyl phosphate (TPhP), applying metabolomics to the same archived tissues. Specifically, liver from rats exposed to five doses of TPhP was investigated using liquid chromatography-mass spectrometry and 1H nuclear magnetic resonance spectroscopy-based metabolomics. Following the application of BMDExpress2 software, 2903 endogenous metabolic features yielded viable dose-response models, confirming a perturbation to the liver metabolome. Metabolic BMD estimates were similarly sensitive to transcriptional BMDs, and more sensitive than both clinical chemistry and apical end point BMDs. Pathway analysis of the multiomics data sets revealed a major effect of TPhP exposure on cholesterol (and downstream) pathways, consistent with clinical chemistry measurements. Additionally, the transcriptomics data indicated that TPhP activated xenobiotic metabolism pathways, which was confirmed by using the underexploited capability of metabolomics to detect xenobiotic-related compounds. Eleven biotransformation products of TPhP were discovered, and their levels were highly correlated with multiple xenobiotic metabolism genes. This work provides a case study showing how metabolomics and transcriptomics can estimate mechanistically anchored points-of-departure. Furthermore, the study demonstrates how metabolomics can also discover biotransformation products, which could be of value within a regulatory setting, for example, as an enhancement of OECD Test Guideline 417 (toxicokinetics).

Integrating gene expression and splicing dynamics across dose-response oxidative modulators

Toxicological risk assessment increasingly utilizes transcriptomics to derive point of departure (POD) and modes of action (MOA) for chemicals. One essential biological process that allows a single gene to generate several different RNA isoforms is called alternative splicing. To comprehensively assess the role of splicing dysregulation in toxicological evaluation and elucidate its potential as a complementary endpoint, we performed RNA-seq on A549 cells treated with five oxidative stress modulators across a wide dose range. Differential gene expression (DGE) showed limited pathway enrichment except at high concentrations. However, alternative splicing analysis revealed variable intron retention events affecting diverse pathways for all chemicals in the absence of significant expression changes. For instance, diazinon elicited negligible gene expression changes but progressive increase in the number of intron retention events, suggesting splicing alterations precede expression responses. Benchmark dose modeling of intron retention data highlighted relevant pathways overlooked by expression analysis. Systematic integration of splicing datasets should be a useful addition to the toxicogenomic toolkit. Combining both modalities paint a more complete picture of transcriptomic dose-responses. Overall, evaluating intron retention dynamics afforded by toxicogenomics may provide biomarkers that can enhance chemical risk assessment and regulatory decision making. This work highlights splicing-aware toxicogenomics as a possible additional tool for examining cellular responses.

A 50-gene biomarker identifies estrogen receptor-modulating chemicals in a microarray compendium

High throughput transcriptomics (HTTr) profiling has the potential to rapidly and comprehensively identify molecular targets of environmental chemicals that can be linked to adverse outcomes. We describe here the construction and characterization of a 50-gene expression biomarker designed to identify estrogen receptor (ER) active chemicals in HTTr datasets. Using microarray comparisons, the genes in the biomarker were identified as those that exhibited consistent directional changes when ER was activated (4 ER agonists; 4 ESR1 gene constitutively active mutants) and opposite directional changes when ER was suppressed (4 antagonist treatments; 4 ESR1 knockdown experiments). The biomarker was evaluated as a predictive tool using the Running Fisher algorithm by comparison to annotated gene expression microarray datasets including those evaluating the transcriptional effects of hormones and chemicals in MCF-7 cells. Depending on the reference dataset used, the biomarker had a predictive accuracy for activation of up to 96%. To demonstrate applicability for HTTr data analysis, the biomarker was used to identify ER activators in a set of 15 chemicals that are considered potential bisphenol A (BPA) alternatives examined at up to 10 concentrations in MCF-7 cells and analyzed by full-genome TempO-Seq. Using benchmark dose (BMD) modeling, the biomarker genes stratified the ER potency of BPA alternatives consistent with previous studies. These results demonstrate that the ER biomarker can be used to accurately identify ER activators in transcript profile data derived from MCF-7 cells.

Comparing Transcriptomic Points of Departure to Apical Effect Concentrations For Larval Fathead Minnow Exposed to Chemicals with Four Different Modes Of Action

It is postulated that below a transcriptomic-based point of departure, adverse effects are unlikely to occur, thereby providing a chemical concentration to use in screening level hazard assessment. The present study extends previous work describing a high-throughput fathead minnow assay that can provide full transcriptomic data after exposure to a test chemical. One-day post-hatch fathead minnows were exposed to ten concentrations of three representatives of four chemical modes of action: organophosphates, ecdysone receptor agonists, plant photosystem II inhibitors, and estrogen receptor agonists for 24 h. Concentration response modeling was performed on whole body gene expression data from each exposure, using measured chemical concentrations when available. Transcriptomic points of departure in larval fathead minnow were lower than apical effect concentrations across fish species but not always lower than toxic effect concentrations in other aquatic taxa like crustaceans and insects. The point of departure was highly dependent on measured chemical concentration which were often lower than the nominal concentration. Differentially expressed genes between chemicals within modes of action were compared and often showed statistically significant overlap. In addition, reproducibility between identical exposures using a positive control chemical (CuSO4) and variability associated with the transcriptomic point of departure using in silico sampling were considered. Results extend a transcriptomic-compatible fathead minnow high-throughput assay for possible use in ecological hazard screening.

Empirical Characterization of False Discovery Rates of Differentially Expressed Genes and Transcriptomic Benchmark Concentrations in Zebrafish Embryos

High-throughput transcriptomics (HTTr) is increasingly applied to zebrafish embryos to survey the toxicological effects of environmental chemicals. Before the adoption of this approach in regulatory testing, it is essential to characterize background noise in order to guide experimental designs. We thus empirically quantified the HTTr false discovery rate (FDR) across different embryo pool sizes, sample sizes, and concentration groups for toxicology studies. We exposed zebrafish embryos to 0.1% dimethyl sulfoxide (DMSO) for 5 days. Pools of 1, 5, 10, and 20 embryos were created (n = 24 samples for each pool size). Samples were sequenced on the TempO-Seq platform and then randomly assigned to mock treatment groups before differentially expressed gene (DEG), pathway, and benchmark concentration (BMC) analyses. Given that all samples were treated with DMSO, any significant DEGs, pathways, or BMCs are false positives. As expected, we found decreasing FDRs for DEG and pathway analyses with increasing pool and sample sizes. Similarly, FDRs for BMC analyses decreased with increasing pool size and concentration groups, with more stringent BMC premodel filtering reducing BMC FDRs. Our study provides foundational data for determining appropriate experiment designs for regulatory toxicity testing with HTTr in zebrafish embryos.

A Comparison of In Vitro Points of Departure with Human Blood Levels for Per- and Polyfluoroalkyl Substances (PFAS)

Per- and polyfluoroalkyl substances (PFAS) are widely used, and their fluorinated state contributes to unique uses and stability but also long half-lives in the environment and humans. PFAS have been shown to be toxic, leading to immunosuppression, cancer, and other adverse health outcomes. Only a small fraction of the PFAS in commerce have been evaluated for toxicity using in vivo tests, which leads to a need to prioritize which compounds to examine further. Here, we demonstrate a prioritization approach that combines human biomonitoring data (blood concentrations) with bioactivity data (concentrations at which bioactivity is observed in vitro) for 31 PFAS. The in vitro data are taken from a battery of cell-based assays, mostly run on human cells. The result is a Bioactive Concentration to Blood Concentration Ratio (BCBCR), similar to a margin of exposure (MoE). Chemicals with low BCBCR values could then be prioritized for further risk assessment. Using this method, two of the PFAS, PFOA (Perfluorooctanoic Acid) and PFOS (Perfluorooctane Sulfonic Acid), have BCBCR values < 1 for some populations. An additional 9 PFAS have BCBCR values < 100 for some populations. This study shows a promising approach to screening level risk assessments of compounds such as PFAS that are long-lived in humans and other species.

Unlocking the Power of Transcriptomic Biomarkers in Qualitative and Quantitative Genotoxicity Assessment of Chemicals

To modernize genotoxicity assessment and reduce reliance on experimental animals, new approach methodologies (NAMs) that provide human-relevant dose-response data are needed. Two transcriptomic biomarkers, GENOMARK and TGx-DDI, have shown a high classification accuracy for genotoxicity. As these biomarkers were extracted from different training sets, we investigated whether combining the two biomarkers in a human-derived metabolically competent cell line (i.e., HepaRG) provides complementary information for the classification of genotoxic hazard identification and potency ranking. First, the applicability of GENOMARK to TempO-Seq, a high-throughput transcriptomic technology, was evaluated. HepaRG cells were exposed for 72 h to increasing concentrations of 10 chemicals (i.e., eight known in vivo genotoxicants and two in vivo nongenotoxicants). Gene expression data were generated using the TempO-Seq technology. We found a prediction performance of 100%, confirming the applicability of GENOMARK to TempO-Seq. Classification using TGx-DDI was then compared to GENOMARK. For the chemicals identified as genotoxic, benchmark concentration modeling was conducted to perform potency ranking. The high concordance observed for both hazard classification and potency ranking by GENOMARK and TGx-DDI highlights the value of integrating these NAMs in a weight of evidence evaluation of genotoxicity.

Transcriptomics-Based Points of Departure for Daphnia magna Exposed to 18 Per- and Polyfluoroalkyl Substances

Per- and polyfluoroalkyl substances (PFAS) represent a large group of contaminants of concern based on their widespread use, environmental persistence, and potential toxicity. Many traditional models for estimating toxicity, bioaccumulation, and other toxicological properties are not well suited for PFAS. Consequently, there is a need to generate hazard information for PFAS in an efficient and cost-effective manner. In the present study, Daphnia magna were exposed to multiple concentrations of 22 different PFAS for 24 h in a 96-well plate format. Following exposure, whole-body RNA was extracted and extracts, each representing five exposed individuals, were subjected to RNA sequencing. Following analytical measurements to verify PFAS exposure concentrations and quality control on processed cDNA libraries for sequencing, concentration-response modeling was applied to the data sets for 18 of the tested compounds, and the concentration at which a concerted molecular response occurred (transcriptomic point of departure; tPOD) was calculated. The tPODs, based on measured concentrations of PFAS, generally ranged from 0.03 to 0.58 µM (9.9-350 µg/L; interquartile range). In most cases, these concentrations were two orders of magnitude lower than similarly calculated tPODs for human cell lines exposed to PFAS. They were also lower than apical effect concentrations reported for seven PFAS for which some crustacean or invertebrate toxicity data were available, although there were a few exceptions. Despite being lower than most other available hazard benchmarks, D. magna tPODs were, on average, four orders of magnitude greater than the maximum aqueous concentrations of PFAS measured in Great Lakes tributaries. Overall, this high-throughput transcriptomics assay with D. magna holds promise as a component of a tiered hazard evaluation strategy employing new approach methodologies. Environ Toxicol Chem 2024;00:1-16. © 2024 SETAC. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.

A Comparison of Machine Learning Approaches for predicting Hepatotoxicity potential using Chemical Structure and Targeted Transcriptomic Data

Animal toxicity testing is time and resource intensive, making it difficult to keep pace with the number of substances requiring assessment. Machine learning (ML) models that use chemical structure information and high-throughput experimental data can be helpful in predicting potential toxicity . However, much of the toxicity data used to train ML models is biased with an unequal balance of positives and negatives primarily since substances selected for in vivo testing are expected to elicit some toxicity effect. To investigate the impact this bias had on predictive performance, various sampling approaches were used to balance in vivo toxicity data as part of a supervised ML workflow to predict hepatotoxicity outcomes from chemical structure and/or targeted transcriptomic data. From the chronic, subchronic, developmental, multigenerational reproductive, and subacute repeat-dose testing toxicity outcomes with a minimum of 50 positive and 50 negative substances, 18 different study-toxicity outcome combinations were evaluated in up to 7 ML models. These included Artificial Neural Networks, Random Forests, Bernouilli Naïve Bayes, Gradient Boosting, and Support Vector classification algorithms which were compared with a local approach, Generalised Read-Across (GenRA), a similarity-weighted k-Nearest Neighbour (k-NN) method. The mean CV F1 performance for unbalanced data across all classifiers and descriptors for chronic liver effects was 0.735 (0.0395 SD). Mean CV F1 performance dropped to 0.639 (0.073 SD) with over-sampling approaches though the poorer performance of KNN approaches in some cases contributed to the observed decrease (mean CV F1 performance excluding KNN was 0.697 (0.072 SD)). With under-sampling approaches, the mean CV F1 was 0.523 (0.083 SD). For developmental liver effects, the mean CV F1 performance was much lower with 0.089 (0.111 SD) for unbalanced approaches and 0.149 (0.084 SD) for under-sampling. Over-sampling approaches led to an increase in mean CV F1 performance (0.234, (0.107 SD)) for developmental liver toxicity. Model performance was found to be dependent on dataset, model type, balancing approach and feature selection. Accordingly tailoring ML workflows for predicting toxicity should consider class imbalance and rely on simpler classifiers first.

Towards Precision Ecotoxicology: Leveraging Evolutionary Conservation of Pharmaceutical and Personal Care Product Targets to Understand Adverse Outcomes Across Species and Life Stages

Translation of environmental science to the practice aims to protect biodiversity and ecosystem services, and our future ability to do so relies on the development of a precision ecotoxicology approach wherein we leverage the genetics and informatics of species to better understand and manage the risks of global pollution. A little over a decade ago, a workshop focusing on the risks of pharmaceuticals and personal care products (PPCPs) in the environment identified a priority research question, "What can be learned about the evolutionary conservation of PPCP targets across species and life stages in the context of potential adverse outcomes and effects?" We review the activities in this area over the past decade, consider prospects of more recent developments, and identify future research needs to develop next-generation approaches for PPCPs and other global chemicals and waste challenges. Environ Toxicol Chem 2024;43:526-536. © 2023 SETAC. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.

Hazard and risk characterization of 56 structurally diverse PFAS using a targeted battery of broad coverage assays using six human cell types

Per- and poly-fluoroalkyl substances (PFAS) are extensively used in commerce leading to their prevalence in the environment. Due to their chemical stability, PFAS are considered to be persistent and bioaccumulative; they are frequently detected in both the environment and humans. Because of this, PFAS as a class (composed of hundreds to thousands of chemicals) are contaminants of very high concern. Little information is available for the vast majority of PFAS, and regulatory agencies lack safety data to determine whether exposure limits or restrictions are needed. Cell-based assays are a pragmatic approach to inform decision-makers on potential health hazards; therefore, we hypothesized that a targeted battery of human in vitro assays can be used to determine whether there are structure-bioactivity relationships for PFAS, and to characterize potential risks by comparing bioactivity (points of departure) to exposure estimates. We tested 56 PFAS from 8 structure-based subclasses in concentration response (0.1-100 μM) using six human cell types selected from target organs with suggested adverse effects of PFAS - human induced pluripotent stem cell (iPSC)-derived hepatocytes, neurons, and cardiomyocytes, primary human hepatocytes, endothelial and HepG2 cells. While many compounds were without effect; certain PFAS demonstrated cell-specific activity highlighting the necessity of using a compendium of in vitro models to identify potential hazards. No class-specific groupings were evident except for some chain length- and structure-related trends. In addition, margins of exposure (MOE) were derived using empirical and predicted exposure data. Conservative MOE calculations showed that most tested PFAS had a MOE in the 1-100 range; ∼20% of PFAS had MOE<1, providing tiered priorities for further studies. Overall, we show that a compendium of human cell-based models can be used to derive bioactivity estimates for a range of PFAS, enabling comparisons with human biomonitoring data. Furthermore, we emphasize that establishing structure-bioactivity relationships may be challenging for the tested PFAS.

Derivation of Transcriptomics-Based Points of Departure for 20 Per- or Polyfluoroalkyl Substances Using a Larval Fathead Minnow (Pimephales promelas) Reduced Transcriptome Assay

Traditional toxicity testing has been unable to keep pace with the introduction of new chemicals into commerce. Consequently, there are limited or no toxicity data for many chemicals to which fish and wildlife may be exposed. Per- and polyfluoroalkyl substances (PFAS) are emblematic of this issue in that ecological hazards of most PFAS remain uncharacterized. The present study employed a high-throughput assay to identify the concentration at which 20 PFAS, with diverse properties, elicited a concerted gene expression response (termed a transcriptomics-based point of departure [tPOD]) in larval fathead minnows (Pimephales promelas; 5-6 days postfertilization) exposed for 24 h. Based on a reduced transcriptome approach that measured whole-body expression of 1832 genes, the median tPOD for the 20 PFAS tested was 10 µM. Longer-chain carboxylic acids (12-13 C-F); an eight-C-F dialcohol, N-alkyl sulfonamide; and telomer sulfonic acid were among the most potent PFAS, eliciting gene expression responses at concentrations <1 µM. With a few exceptions, larval fathead minnow tPODs were concordant with those based on whole-transcriptome response in human cell lines. However, larval fathead minnow tPODs were often greater than those for Daphnia magna exposed to the same PFAS. The tPODs overlapped concentrations at which other sublethal effects have been reported in fish (available for 10 PFAS). Nonetheless, fathead minnow tPODs were orders of magnitude higher than aqueous PFAS concentrations detected in tributaries of the North American Great Lakes, suggesting a substantial margin of safety. Overall, results broadly support the use of a fathead minnow larval transcriptomics assay to derive screening-level potency estimates for use in ecological risk-based prioritization. Environ Toxicol Chem 2024;00:1-16. © 2024 SETAC. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.

Risk-based prioritization of PFAS using phenotypic and transcriptomic data from human induced pluripotent stem cell-derived hepatocytes and cardiomyocytes

Per- and polyfluoroalkyl substances (PFAS) are chemicals with important applications; they are persistent in the environment and may pose human health hazards. Regulatory agencies are con­sidering restrictions and bans of PFAS; however, little data exists for informed decisions. Several prioritization strategies were proposed for evaluation of potential hazards of PFAS. Structure-based grouping could expedite the selection of PFAS for testing; still, the hypothesis that structure-effect relationships exist for PFAS requires confirmation. We tested 26 structurally diverse PFAS from 8 groups using human induced pluripotent stem cell-derived hepatocytes and cardiomyocytes, and tested concentration-response effects on cell function and gene expression. Few phenotypic effects were observed in hepatocytes, but negative chronotropy was observed in cardiomyocytes for 8 PFAS. Substance- and cell type-dependent transcriptomic changes were more prominent but lacked substantial group-specific effects. In hepatocytes, we found upregulation of stress-related and extracellular matrix organization pathways, and down-regulation of fat metabolism. In car­diomyocytes, contractility-related pathways were most affected. We derived phenotypic and transcriptomic points of departure and compared them to predicted PFAS exposures. Conservative estimates for bioactivity and exposure were used to derive a bioactivity-to-exposure ratio (BER) for each PFAS; 23 of 26 PFAS had BER > 1. Overall, these data suggest that structure-based PFAS grouping may not be sufficient to predict their biological effects. Testing of individual PFAS may be needed for scientifically-supported decision-making. Our proposed strategy of using two human cell types and considering phenotypic and transcriptomic effects, combined with dose-response analysis and calculation of BER, may be used for PFAS prioritization.

Combination of computational new approach methodologies for enhancing evidence of biological pathway conservation across species

The ability to predict which chemicals are of concern for environmental safety is dependent, in part, on the ability to extrapolate chemical effects across many species. This work investigated the complementary use of two computational new approach methodologies to support cross-species predictions of chemical susceptibility: the US Environmental Protection Agency Sequence Alignment to Predict Across Species Susceptibility (SeqAPASS) tool and Unilever's recently developed Genes to Pathways - Species Conservation Analysis (G2P-SCAN) tool. These stand-alone tools rely on existing biological knowledge to help understand chemical susceptibility and biological pathway conservation across species. The utility and challenges of these combined computational approaches were demonstrated using case examples focused on chemical interactions with peroxisome proliferator activated receptor alpha (PPARα), estrogen receptor 1 (ESR1), and gamma-aminobutyric acid type A receptor subunit alpha (GABRA1). Overall, the biological pathway information enhanced the weight of evidence to support cross-species susceptibility predictions. Through comparisons of relevant molecular and functional data gleaned from adverse outcome pathways (AOPs) to mapped biological pathways, it was possible to gain a toxicological context for various chemical-protein interactions. The information gained through this computational approach could ultimately inform chemical safety assessments by enhancing cross-species predictions of chemical susceptibility. It could also help fulfill a core objective of the AOP framework by potentially expanding the biologically plausible taxonomic domain of applicability of relevant AOPs.

Investigating open access new approach methods (NAM) to assess biological points of departure: A case study with 4 neurotoxic pesticides

Open access new approach methods (NAM) in the US EPA ToxCast program and NTP Integrated Chemical Environment (ICE) were used to investigate activities of four neurotoxic pesticides: endosulfan, fipronil, propyzamide and carbaryl. Concordance of in vivo regulatory points of departure (POD) adjusted for interspecies extrapolation (AdjPOD) to modelled human Administered Equivalent Dose (AEDHuman) was assessed using 3-compartment or Adult/Fetal PBTK in vitro to in vivo extrapolation. Model inputs were from Tier 1 (High throughput transcriptomics: HTTr, high throughput phenotypic profiling: HTPP) and Tier 2 (single target: ToxCast) assays. HTTr identified gene expression signatures associated with potential neurotoxicity for endosulfan, propyzamide and carbaryl in non-neuronal MCF-7 and HepaRG cells. The HTPP assay in U-2 OS cells detected potent effects on DNA endpoints for endosulfan and carbaryl, and mitochondria with fipronil (propyzamide was inactive). The most potent ToxCast assays were concordant with specific components of each chemical mode of action (MOA). Predictive adult IVIVE models produced fold differences (FD) < 10 between the AEDHuman and the measured in vivo AdjPOD. The 3-compartment model was concordant (i.e., smallest FD) for endosulfan, fipronil and carbaryl, and PBTK was concordant for propyzamide. The most potent AEDHuman predictions for each chemical showed HTTr, HTPP and ToxCast were mainly concordant with in vivo AdjPODs but assays were less concordant with MOAs. This was likely due to the cell types used for testing and/or lack of metabolic capabilities and pathways available in vivo. The Fetal PBTK model had larger FDs than adult models and was less predictive overall.

High-Throughput Transcriptomics of Water Extracts Detects Reductions in Biological Activity with Water Treatment Processes

The presence of numerous chemical contaminants from industrial, agricultural, and pharmaceutical sources in water supplies poses a potential risk to human and ecological health. Current chemical analyses suffer from limitations, including chemical coverage and high cost, and broad-coverage in vitro assays such as transcriptomics may further improve water quality monitoring by assessing a large range of possible effects. Here, we used high-throughput transcriptomics to assess the activity induced by field-derived water extracts in MCF7 breast carcinoma cells. Wastewater and surface water extracts induced the largest changes in expression among cell proliferation-related genes and neurological, estrogenic, and antibiotic pathways, whereas drinking and reclaimed water extracts that underwent advanced treatment showed substantially reduced bioactivity on both gene and pathway levels. Importantly, reclaimed water extracts induced fewer changes in gene expression than laboratory blanks, which reinforces previous conclusions based on targeted assays and improves confidence in bioassay-based monitoring of water quality.

Exploring the effects of experimental parameters and data modeling approaches on in vitro transcriptomic point-of-departure estimates

Multiple new approach methods (NAMs) are being developed to rapidly screen large numbers of chemicals to aid in hazard evaluation and risk assessments. High-throughput transcriptomics (HTTr) in human cell lines has been proposed as a first-tier screening approach for determining the types of bioactivity a chemical can cause (activation of specific targets vs. generalized cell stress) and for calculating transcriptional points of departure (tPODs) based on changes in gene expression. In the present study, we examine a range of computational methods to calculate tPODs from HTTr data, using six data sets in which MCF7 cells cultured in two different media formulations were treated with a panel of 44 chemicals for 3 different exposure durations (6, 12, 24 hr). The tPOD calculation methods use data at the level of individual genes and gene set signatures, and compare data processed using the ToxCast Pipeline 2 (tcplfit2), BMDExpress and PLIER (Pathway Level Information ExtractoR). Methods were evaluated by comparing to in vitro PODs from a validated set of high-throughput screening (HTS) assays for a set of estrogenic compounds. Key findings include: (1) for a given chemical and set of experimental conditions, tPODs calculated by different methods can vary by several orders of magnitude; (2) tPODs are at least as sensitive to computational methods as to experimental conditions; (3) in comparison to an external reference set of PODs, some methods give generally higher values, principally PLIER and BMDExpress; and (4) the tPODs from HTTr in this one cell type are mostly higher than the overall PODs from a broad battery of targeted in vitro ToxCast assays, reflecting the need to test chemicals in multiple cell types and readout technologies for in vitro hazard screening.

The phosphoproteome is a first responder in tiered cellular adaptation to chemical stress followed by proteomics and transcriptomics alteration

Next-generation risk assessment (NGRA) for environmental chemicals involves a weight of evidence (WoE) framework integrating a suite of new approach methodologies (NAMs) based on points of departure (PoD) obtained from in vitro assays. Among existing NAMs, the omic-based technologies are of particular importance based on the premise that any apical endpoint change indicative of impaired health must be underpinned by some alterations at the omics level, such as transcriptome, proteome, metabolome, epigenome and genome. Transcriptomic assay plays a leading role in providing relatively conservative PoDs compared with apical endpoints. However, it is unclear whether and how parameters measured with other omics techniques predict the cellular response to chemical perturbations, especially at exposure levels below the transcriptomically defined PoD. Multi-omics coverage may provide additional sensitive or confirmative biomarkers to complement and reduce the uncertainty in safety decisions made using targeted and transcriptomics assays. In the present study, we conducted multi-omics studies of transcriptomics, proteomics and phosphoproteomics on two prototype compounds, coumarin and 2,4-dichlorophenoxyacetic acid (2,4-D), with multiple chemical concentrations and time points, to understand the sensitivity of the three omics techniques in response to chemically-induced changes in HepG2. We demonstrated that, phosphoproteomics alterations occur not only earlier in time, but also more sensitive to lower concentrations than proteomics and transcriptomics when the HepG2 cells were exposed to various chemical treatments. The phosphoproteomics changes appear to approach maximum when the transcriptomics alterations begin to initiate. Therefore, it is proximal to the very early effects induced by chemical exposure. We concluded that phosphoproteomics can be utilized to provide a more complete coverage of chemical-induced cellular alteration and supplement transcriptomics-based health safety decision making.