David M. Reif

In Vitro Screening of Environmental Chemicals for Targeted Testing Prioritization: The ToxCast Project

Background Chemical toxicity testing is being transformed by advances in biology and computer modeling, concerns over animal use, and the thousands of environmental chemicals lacking toxicity data. The U.S. Environmental Protection Agency’s ToxCast program aims to address these concerns by screening and prioritizing chemicals for potential human toxicity using in vitro assays and in silico approaches. Objectives This project aims to evaluate the use of in vitro assays for understanding the types of molecular and pathway perturbations caused by environmental chemicals and to build initial prioritization models of in vivo toxicity. Methods We tested 309 mostly pesticide active chemicals in 467 assays across nine technologies, including high-throughput cell-free assays and cell-based assays, in multiple human primary cells and cell lines plus rat primary hepatocytes. Both individual and composite scores for effects on genes and pathways were analyzed. Results Chemicals displayed a broad spectrum of activity at the molecular and pathway levels. We saw many expected interactions, including endocrine and xenobiotic metabolism enzyme activity. Chemicals ranged in promiscuity across pathways, from no activity to affecting dozens of pathways. We found a statistically significant inverse association between the number of pathways perturbed by a chemical at low in vitro concentrations and the lowest in vivo dose at which a chemical causes toxicity. We also found associations between a small set of in vitro assays and rodent liver lesion formation. Conclusions This approach promises to provide meaningful data on the thousands of untested environmental chemicals and to guide targeted testing of environmental contaminants.

Predictive Model of Rat Reproductive Toxicity from ToxCast High Throughput Screening

The U.S. Environmental Protection Agencys ToxCast research program uses high throughput screening (HTS) for profiling bioactivity and predicting the toxicity of large numbers of chemicals. ToxCast Phase I tested 309 well-characterized chemicals in more than 500 assays for a wide range of molecular targets and cellular responses. Of the 309 environmental chemicals in Phase I, 256 were linked to high-quality rat multigeneration reproductive toxicity studies in the relational Toxicity Reference Database. Reproductive toxicants were defined here as having achieved a reproductive lowest-observed-adverse-effect level of less than 500 mg kg−1 day−1. Eight-six chemicals were identified as reproductive toxicants in the rat, and 68 of those had sufficient in vitro bioactivity to model. Each assay was assessed for univariate association with the identified reproductive toxicants. Significantly associated assays were linked to gene sets and used for the subsequent predictive modeling. Using linear discriminant analysis and fivefold cross-validation, a robust and stable predictive model was produced capable of identifying rodent reproductive toxicants with 77% ± 2% and 74% ± 5% (mean ± SEM) training and test cross-validation balanced accuracies, respectively. With a 21-chemical external validation set, the model was 76% accurate, further indicating the models potential for prioritizing the many thousands of environmental chemicals with little to no hazard information. The biological features of the model include steroidal and nonsteroidal nuclear receptors, cytochrome P450 enzyme inhibition, G protein-coupled receptors, and cell signaling pathway readouts—mechanistic information suggesting additional targeted, integrated testing strategies and potential applications of in vitro HTS to risk assessment.

Zebrafish developmental screening of the ToxCast™ Phase I chemical library

Zebrafish (Danio rerio) is an emerging toxicity screening model for both human health and ecology. As part of the Computational Toxicology Research Program of the U.S. EPA, the toxicity of the 309 ToxCast™ Phase I chemicals was assessed using a zebrafish screen for developmental toxicity. All exposures were by immersion from 6-8 h post fertilization (hpf) to 5 days post fertilization (dpf); nominal concentration range of 1 nM-80 μM. On 6 dpf larvae were assessed for death and overt structural defects. Results revealed that the majority (62%) of chemicals were toxic to the developing zebrafish; both toxicity incidence and potency was correlated with chemical class and hydrophobicity (logP); and inter-and intra-plate replicates showed good agreement. The zebrafish embryo screen, by providing an integrated model of the developing vertebrate, compliments the ToxCast assay portfolio and has the potential to provide information relative to overt and organismal toxicity.

Endocrine Profiling and Prioritization of Environmental Chemicals Using ToxCast Data

Background The prioritization of chemicals for toxicity testing is a primary goal of the U.S. Environmental Protection Agency (EPA) ToxCast™ program. Phase I of ToxCast used a battery of 467 in vitro, high-throughput screening assays to assess 309 environmental chemicals. One important mode of action leading to toxicity is endocrine disruption, and the U.S. EPA’s Endocrine Disruptor Screening Program (EDSP) has been charged with screening pesticide chemicals and environmental contaminants for their potential to affect the endocrine systems of humans and wildlife. Objective The goal of this study was to develop a flexible method to facilitate the rational prioritization of chemicals for further evaluation and demonstrate its application as a candidate decision-support tool for EDSP. Methods Focusing on estrogen, androgen, and thyroid pathways, we defined putative endocrine profiles and derived a relative rank or score for the entire ToxCast library of 309 unique chemicals. Effects on other nuclear receptors and xenobiotic metabolizing enzymes were also considered, as were pertinent chemical descriptors and pathways relevant to endocrine-mediated signaling. Results Combining multiple data sources into an overall, weight-of-evidence Toxicological Priority Index (ToxPi) score for prioritizing further chemical testing resulted in more robust conclusions than any single data source taken alone. Conclusions Incorporating data from in vitro assays, chemical descriptors, and biological pathways in this prioritization schema provided a flexible, comprehensive visualization and ranking of each chemical’s potential endocrine activity. Importantly, ToxPi profiles provide a transparent visualization of the relative contribution of all information sources to an overall priority ranking. The method developed here is readily adaptable to diverse chemical prioritization tasks.

Machine learning for detecting gene-gene interactions: a review.

Complex interactions among genes and environmental factors are known to play a role in common human disease aetiology. There is a growing body of evidence to suggest that complex interactions are ‘the norm’ and, rather than amounting to a small perturbation to classical Mendelian genetics, interactions may be the predominant effect. Traditional statistical methods are not well suited for detecting such interactions, especially when the data are high dimensional (many attributes or independent variables) or when interactions occur between more than two polymorphisms. In this review, we discuss machine-learning models and algorithms for identifying and characterising susceptibility genes in common, complex, multifactorial human diseases. We focus on the following machine-learning methods that have been used to detect gene-gene interactions: neural networks, cellular automata, random forests, and multifactor dimensionality reduction. We conclude with some ideas about how these methods and others can be integrated into a comprehensive and flexible framework for data mining and knowledge discovery in human genetics.

Incorporating Human Dosimetry and Exposure into High-Throughput In Vitro Toxicity Screening

Many chemicals in commerce today have undergone limited or no safety testing. To reduce the number of untested chemicals and prioritize limited testing resources, several governmental programs are using high-throughput in vitro screens for assessing chemical effects across multiple cellular pathways. In this study, metabolic clearance and plasma protein binding were experimentally measured for 35 ToxCast phase I chemicals. The experimental data were used to parameterize a population-based in vitro-to-in vivo extrapolation model for estimating the human oral equivalent dose necessary to produce a steady-state in vivo concentration equivalent to in vitro AC(50) (concentration at 50% of maximum activity) and LEC (lowest effective concentration) values from the ToxCast data. For 23 of the 35 chemicals, the range of oral equivalent doses for up to 398 ToxCast assays was compared with chronic aggregate human oral exposure estimates in order to assess whether significant in vitro bioactivity occurred within the range of maximum expected human oral exposure. Only 2 of the 35 chemicals, triclosan and pyrithiobac-sodium, had overlapping oral equivalent doses and estimated human oral exposures. Ranking by the potencies of the AC(50) and LEC values, these two chemicals would not have been at the top of a prioritization list. Integrating both dosimetry and human exposure information with the high-throughput toxicity screening efforts provides a better basis for making informed decisions on chemical testing priorities and regulatory attention. Importantly, these tools are necessary to move beyond hazard rankings to estimates of possible in vivo responses based on in vitro screens.

Impact of environmental chemicals on key transcription regulators and correlation to toxicity end points within EPA's ToxCast program.

Exposure to environmental chemicals adds to the burden of disease in humans and wildlife to a degree that is difficult to estimate and, thus, mitigate. The ability to assess the impact of existing chemicals for which little to no toxicity data are available or to foresee such effects during early stages of chemical development and use, and before potential exposure occurs, is a pressing need. However, the capacity of the current toxicity evaluation approaches to meet this demand is limited by low throughput and high costs. In the context of EPAs ToxCast project, we have evaluated a novel cellular biosensor system (Factorial (1) ) that enables rapid, high-content assessment of a compounds impact on gene regulatory networks. The Factorial biosensors combined libraries of cis- and trans-regulated transcription factor reporter constructs with a highly homogeneous method of detection enabling simultaneous evaluation of multiplexed transcription factor activities. Here, we demonstrate the application of the technology toward determining bioactivity profiles by quantitatively evaluating the effects of 309 environmental chemicals on 25 nuclear receptors and 48 transcription factor response elements. We demonstrate coherent transcription factor activity across nuclear receptors and their response elements and that Nrf2 activity, a marker of oxidative stress, is highly correlated to the overall promiscuity of a chemical. Additionally, as part of the ToxCast program, we identify molecular targets that associate with in vivo end points and represent modes of action that can serve as potential toxicity pathway biomarkers and inputs for predictive modeling of in vivo toxicity.

Predictive Models of Prenatal Developmental Toxicity from ToxCast High-Throughput Screening Data

Environmental Protection Agencys ToxCast project is profiling the in vitro bioactivity of chemicals to assess pathway-level and cell-based signatures that correlate with observed in vivo toxicity. We hypothesized that developmental toxicity in guideline animal studies captured in the ToxRefDB database would correlate with cell-based and cell-free in vitro high-throughput screening (HTS) data to reveal meaningful mechanistic relationships and provide models identifying chemicals with the potential to cause developmental toxicity. To test this hypothesis, we built statistical associations based on HTS and in vivo developmental toxicity data from ToxRefDB. Univariate associations were used to filter HTS assays based on statistical correlation with distinct in vivo endpoint. This revealed 423 total associations with distinctly different patterns for rat (301 associations) and rabbit (122 associations) across multiple HTS assay platforms. From these associations, linear discriminant analysis with cross-validation was used to build the models. Species-specific models of predicted developmental toxicity revealed strong balanced accuracy (> 70%) and unique correlations between assay targets such as transforming growth factor beta, retinoic acid receptor, and G-protein-coupled receptor signaling in the rat and inflammatory signals, such as interleukins (IL) (IL1a and IL8) and chemokines (CCL2), in the rabbit. Species-specific toxicity endpoints were associated with one another through common Gene Ontology biological processes, such as cleft palate to urogenital defects through placenta and embryonic development. This work indicates the utility of HTS assays for developing pathway-level models predictive of developmental toxicity.

Profiling chemicals based on chronic toxicity results from the U.S. EPA ToxRef Database.

Background Thirty years of pesticide registration toxicity data have been historically stored as hardcopy and scanned documents by the U.S. Environmental Protection Agency (EPA). A significant portion of these data have now been processed into standardized and structured toxicity data within the EPA’s Toxicity Reference Database (ToxRefDB), including chronic, cancer, developmental, and reproductive studies from laboratory animals. These data are now accessible and mineable within ToxRefDB and are serving as a primary source of validation for U.S. EPA’s ToxCast research program in predictive toxicology. Objectives We profiled in vivo toxicities across 310 chemicals as a model application of ToxRefDB, meeting the need for detailed anchoring end points for development of ToxCast predictive signatures. Methods Using query and structured data-mining approaches, we generated toxicity profiles from ToxRefDB based on long-term rodent bioassays. These chronic/cancer data were analyzed for suitability as anchoring end points based on incidence, target organ, severity, potency, and significance. Results Under conditions of the bioassays, we observed pathologies for 273 of 310 chemicals, with greater preponderance (> 90%) occurring in the liver, kidney, thyroid, lung, testis, and spleen. We observed proliferative lesions for 225 chemicals, and 167 chemicals caused progression to cancer-related pathologies. Conclusions Based on incidence, severity, and potency, we selected 26 primarily tissue-specific pathology end points to uniformly classify the 310 chemicals. The resulting toxicity profile classifications demonstrate the utility of structuring legacy toxicity information and facilitating the computation of these data within ToxRefDB for ToxCast and other applications.

Multidimensional In Vivo Hazard Assessment Using Zebrafish

There are tens of thousands of man-made chemicals in the environment; the inherent safety of most of these chemicals is not known. Relevant biological platforms and new computational tools are needed to prioritize testing of chemicals with limited human health hazard information. We describe an experimental design for high-throughput characterization of multidimensional in vivo effects with the power to evaluate trends relating to commonly cited chemical predictors. We evaluated all 1060 unique U.S. EPA ToxCast phase 1 and 2 compounds using the embryonic zebrafish and found that 487 induced significant adverse biological responses. The utilization of 18 simultaneously measured endpoints means that the entire system serves as a robust biological sensor for chemical hazard. The experimental design enabled us to describe global patterns of variation across tested compounds, evaluate the concordance of the available in vitro and in vivo phase 1 data with this study, highlight specific mechanisms/value-added/novel biology related to notochord development, and demonstrate that the developmental zebrafish detects adverse responses that would be missed by less comprehensive testing strategies.