Using chemical structure information to develop predictive models for in vitro toxicokinetic parameters to inform high-throughput risk-assessment

The U.S. Environmental Protection Agency (EPA), National Toxicology Program (NTP), and National Institutes of Health (NIH) Chemical Genomics Center (NCGC) have complementary research programs designed to improve chemical toxicity evaluations by developing high throughput screening (HTS) methods that evaluate the impact of environmental chemicals on key toxicity pathways. These federal partners are coordinating an extension of the EPA’s ToxCast program, the NTP’s HTS initiative, and the NCGC’s Molecular Libraries Initiative into a collaborative research program focused on identifying toxicity pathways and developing in vitro assays to characterize the ability of chemicals to perturb those pathways. The goal is to develop new paradigm for high throughput toxicity testing that collects mechanistic and quantitative data from in vitro assays measuring chemical modulation of biological processes involved in the progression to toxicity. As toxicity pathways are identified, the in vitro assays can be optimized for comparison to in vivo animal studies, and for predicting effects in humans. Subsequent computational modeling of toxicity pathway responses and appropriate chemical dosimetry will need to be developed to make these predictions relevant for human health risk assessment. This work was reviewed by EPA and approved for publication but does not necessarily reflect official Agency policy. Index Terms: Toxicogenomics, High Throughput Screening/Testing, EPA ToxCast, Chemical Risk Assessment

Download Full-text

Leveraging high-throughput screening data and conditional generative adversarial networks to advance predictive toxicology

10.1101/2020.10.02.322917 ◽

2020 ◽

Author(s):

Adrian J. Green ◽

Martin J. Mohlenkamp ◽

Jhuma Das ◽

Meenal Chaudhari ◽

Lisa Truong ◽

...

Keyword(s):

High Throughput ◽

High Throughput Screening ◽

Chemical Structure ◽

Structural Information ◽

Generative Adversarial Networks ◽

Toxicity Data ◽

Generative Adversarial Network ◽

Structure Information ◽

Adversarial Network

AbstractThere are currently 85,000 chemicals registered with the Environmental Protection Agency (EPA) under the Toxic Substances Control Act, but only a small fraction have measured toxicological data. To address this gap, high-throughput screening (HTS) methods are vital. As part of one such HTS effort, embryonic zebrafish were used to examine a suite of morphological and mortality endpoints at six concentrations from over 1,000 unique chemicals found in the ToxCast library (phase 1 and 2). We hypothesized that by using a conditional Generative Adversarial Network (cGAN) and leveraging this large set of toxicity data, plus chemical structure information, we could efficiently predict toxic outcomes of untested chemicals. CAS numbers for each chemical were used to generate textual files containing three-dimensional structural information for each chemical. Utilizing a novel method in this space, we converted the 3D structural information into a weighted set of points while retaining all information about the structure. In vivo toxicity and chemical data were used to train two neural network generators. The first used regression (Go-ZT) while the second utilized cGAN architecture (GAN-ZT) to train a generator to produce toxicity data. Our results showed that both Go-ZT and GAN-ZT models produce similar results, but the cGAN achieved a higher sensitivity (SE) value of 85.7% vs 71.4%. Conversely, Go-ZT attained higher specificity (SP), positive predictive value (PPV), and Kappa results of 67.3%, 23.4%, and 0.21 compared to 24.5%, 14.0%, and 0.03 for the cGAN, respectively. By combining both Go-ZT and GAN-ZT, our consensus model improved the SP, PPV, and Kappa, to 75.5%, 25.0%, and 0.211, respectively, resulting in an area under the receiver operating characteristic (AUROC) of 0.663. Considering their potential use as prescreening tools, these models could provide in vivo toxicity predictions and insight into untested areas of the chemical space to prioritize compounds for HT testing.SummaryA conditional Generative Adversarial Network (cGAN) can leverage a large chemical set of experimental toxicity data plus chemical structure information to predict the toxicity of untested compounds.

Download Full-text

Potency ranking of per- and polyfluoroalkyl substances using high-throughput transcriptomic analysis of human liver spheroids

Toxicological Sciences ◽

10.1093/toxsci/kfab102 ◽

2021 ◽

Author(s):

A J F Reardon ◽

A Rowan-Carroll ◽

S S Ferguson ◽

K Leingartner ◽

R Gagne ◽

...

Keyword(s):

Gene Expression ◽

Risk Assessment ◽

High Throughput ◽

Human Liver ◽

Human Health Risk ◽

Carbon Chain Length ◽

Human Liver Cell ◽

Polyfluoroalkyl Substances ◽

Potency Ranking

Abstract Per- and polyfluoroalkyl substances (PFAS) are some of the most prominent organic contaminants in human blood. Although the toxicological implications of human exposure to perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) are well established, data on lesser-understood PFAS are limited. New approach methodologies (NAMs) that apply bioinformatic tools to high-throughput data are being increasingly considered to inform risk assessment for data-poor chemicals. The aim of this study was to compare the potencies (i.e., benchmark concentrations: BMCs) of PFAS in primary human liver microtissues (3D spheroids) using high-throughput transcriptional profiling. Gene expression changes were measured using TempO-seq, a templated, multiplexed RNA-sequencing platform. Spheroids were exposed for 1 or 10 days to increasing concentrations of 23 PFAS in three subgroups: carboxylates (PFCAs), sulfonates (PFSAs), and fluorotelomers and sulfonamides. PFCAs and PFSAs exhibited trends toward increased transcriptional potency with carbon chain-length. Specifically, longer-chain compounds (7 to 10 carbons) were more likely to induce changes in gene expression and have lower transcriptional BMCs. The combined high-throughput transcriptomic and bioinformatic analyses support the capability of NAMs to efficiently assess the effects of PFAS in liver microtissues. The data enable potency ranking of PFAS for human liver cell spheroid cytotoxicity and transcriptional changes, and assessment of in vitro transcriptomic points of departure. These data improve our understanding of the possible health effects of PFAS and will be used to inform read-across for human health risk assessment.

Download Full-text

Translating Neurobehavioral Toxicity Across Species From Zebrafish to Rats to Humans: Implications for Risk Assessment

Frontiers in Toxicology ◽

10.3389/ftox.2021.629229 ◽

2021 ◽

Vol 3 ◽

Author(s):

Charles V. Vorhees ◽

Michael T. Williams ◽

Andrew B. Hawkey ◽

Edward D. Levin

Keyword(s):

Risk Assessment ◽

Animal Models ◽

Experimental Animal ◽

High Throughput ◽

Clinical Studies ◽

Drugs Of Abuse ◽

Toxicant Exposure ◽

Essential Information ◽

Experimental Animal Models

There is a spectrum of approaches to neurotoxicological science from high-throughput in vitro cell-based assays, through a variety of experimental animal models to human epidemiological and clinical studies. Each level of analysis has its own advantages and limitations. Experimental animal models give essential information for neurobehavioral toxicology, providing cause-and-effect information regarding risks of neurobehavioral dysfunction caused by toxicant exposure. Human epidemiological and clinical studies give the closest information to characterizing human risk, but without randomized treatment of subjects to different toxicant doses can only give information about association between toxicant exposure and neurobehavioral impairment. In vitro methods give much needed high throughput for many chemicals and mixtures but cannot provide information about toxicant impacts on behavioral function. Crucial to the utility of experimental animal model studies is cross-species translation. This is vital for both risk assessment and mechanistic determination. Interspecies extrapolation is important to characterize from experimental animal models to humans and between different experimental animal models. This article reviews the literature concerning extrapolation of neurobehavioral toxicology from established rat models to humans and from zebrafish a newer experimental model to rats. The functions covered include locomotor activity, emotion, and cognition and the neurotoxicants covered include pesticides, metals, drugs of abuse, flame retardants and polycyclic aromatic hydrocarbons. With more complete understanding of the strengths and limitations of interspecies translation, we can better use animal models to protect humans from neurobehavioral toxicity.

Download Full-text