Tox21 Cross-Partner Projects
2. Cell Line Selection for High-Throughput Transcriptomics (HTT)
Leads: Nisha Sipes, Josh Harrill, Woody Setzer
Goal: Develop a strategy for selecting maximally diverse cell types/lines to maximally cover biological targets and pathways for high-throughput chemical screening using gene expression (i.e. transcriptomics).
- Issue: HTT is an efficient means of screening chemicals for bioactivity across a broad range of potential molecular targets. However, no single laboratory (in vitro) model will express all molecular targets or accurately model the diversity of chemical disturbances observed in different cell types of the human body. As in vitro models derived from diverse tissues and sources have different gene/protein expression patterns, they may respond differently to chemical exposures. A HTT screening panel using diverse cell lines would provide a more comprehensive understanding of chemical bioactivity than studies in any single cell type.
- Project Focus: Use computational modeling of existing transcriptomics data to select a set of “diverse” cell lines, measure transcriptomic responses of “diverse” cells following exposure to a panel of chemicals, determine the degree to which “diverse” cell lines respond differently to chemicals, and analyze the responses among pooled cell lysates.
4. In Vitro Chemical Disposition
Leads: Katie Paul Friedman, Mike Devito
Goal: Understand the impact of chemical disposition within in vitro test systems across a broad range of chemical categories and develop a computational model to predict differences between the “nominal” concentration of a chemical compared with “true” concentration in the media and cells.
- Issue: In a cell-based laboratory test (in vitro assay), the actual concentration of a chemical inside cells is likely different from the nominal concentration applied to the medium in the well of a microtiter assay plate. Mathematical models exist for predicting in vitro disposition, but very few chemicals have been evaluated for in vitro disposition. Across the Tox21 chemical library, chemical partitioning could affect the accuracy of predictions made from laboratory (in vitro) data about living (in vivo) systems, but the number of chemicals affected and to what degree are unknown.
- Project Focus: This cross-partner project includes measurement of in vitro disposition of approximately 200 chemicals, with the goal of developing model predictions for the rest of the Tox21 chemical library.
5. High-Throughput Transcriptomic Analysis
Leads: Steve Ferguson, Josh Harrill, Menghang Xia
Goal: Develop a common chemical reference dataset for interpretation of high-throughput transcriptomic screening data.
- Issue: Gene expression profiling has proven to be an invaluable tool to explore mechanisms of chemical interactions with biological systems (e.g., pharmacology, toxicology). However, these tools historically lack sufficient volume to study a broad range of chemicals or characterize concentration-response profiles necessary to identify disturbed biological-response pathways.
- Project Focus: Build a robust transcriptomic data set with hundreds of chemicals, primarily made up of data-rich reference chemicals with established links to biological-response pathways. Transcriptomic signatures will be developed to identify molecular targets and pathways that are perturbed by chemical treatment as well as the progression of perturbations as a function of chemical concentration.
7. Toxicodynamic Variability in Developmental Neurotoxicity
Leads: Mamta Behl, Alison Harrill, Syed Imam, Josh Harrill
Goal: Incorporate genetic variation into cell-based test systems to better understand potential population differences in response to chemicals that may cause toxic neurological effects.
- Issue: Genetic differences between people can have a profound effect on whether an individual is susceptible to negative health outcomes caused by a given chemical.
- Project Focus: Use neural progenitor cells derived from a highly diverse rodent population, called the Diversity Outbred, to determine variability in toxicity outcomes after exposure to known or suspected neurotoxic chemicals. The data collected has the potential to inform human health risk assessments for chemical exposures, replacing default inter-individual uncertainty factors.
8. Performance Based Validation of Alternative Test Systems and Models
Leads: Keith Houck, Richard Judson, Nicole Kleinstreuer
Goal: Develop an evaluation framework for the development of performance standards which can be used to establish confidence in alternative test systems and models.
- Issue: Validation is needed to increase the usefulness of ToxCast and Tox21 high-throughput screening (HTS) data in regulatory applications.
- Project Focus: Develop curated sets of active and inactive reference chemicals as well as known assay interference chemicals. In addition to developing a process for identifying reference chemicals, defining a process for describing the essential test method components, evaluating the assay data for accuracy and reliability, and declaring the assay “validated” is necessary. Finally, both the development of reference chemical sets and a validation process must be streamlined and fast enough to manage the tens to hundreds of assays that can help inform regulatory decisions.
9. Retrofitting Existing Tox21 High-Throughput Screening Assays with Metabolic Capability
Leads: Menghang Xia, Kristine Witt, Steve Simmons
Goal: To add xenobiotic metabolism capability to existing Tox21 assays so they provide more accurate and informative data regarding in vivo activity.
- Issue: Existing Tox21 high-throughput screening assays generally lack significant xenobiotic metabolism capability either for activation or for detoxification of tested chemicals and, as a result, testing results may not accurately reflect in vivo activity, which could lead to mischaracterization of potential hazards.
- Project Focus: 1) To add xenobiotic metabolism (XM) capability to existing Tox21 assays; 2) to use these new XM capable methods to screen the Tox21 10K compound collection to identify chemicals that are either bioactivated or detoxified by XM; and 3) to identify the enzyme(s) responsible for observable XM-mediated shifts in bioactivity.
10. Expansion of Pathway Coverage by Tox21 High-Throughput Screening Assays for Better Prediction of Adverse Drug Effects
Leads: Ruili Huang, Weida Tong, Kristine Witt, Steve Ferguson
Goal: To improve the prediction of adverse drug effects by using additional assays that can probe toxicologically important targets and pathways that are not captured in current Tox21 testing.
- Issue: Current Tox21 high-throughput screening assays cover a limited portion of the biological response space and are unlikely to capture enough toxicity mechanisms to be adequately predictive or to effectively develop comprehensive in vitro toxicity profiles.
- Project Focus: 1) Develop, optimize, and use six new assays to screen the Tox21 10K collection; and 2) improve prediction models for adverse drug effects, including but not limited to drug induced liver injury and cardiotoxicity, by incorporating data from the new assays.
12. Evaluating Thiol Reactivity of Tox21 Chemicals Using the MSTI Assay
Leads: Grace Patlewicz, Keith Houck, Menghang Xia, Steve Simmons
Goal: To generate chemical reactivity information for the Tox21 10K library and determine which chemicals are electrophiles.
- Issue: Existing Tox21 high-throughput screening assays do not characterize non-specific reactive chemicals. The ability to characterize these chemicals and determine which are electrophiles would enhance the prediction of adverse events, such as skin sensitization and liver and kidney toxicities, which may be caused by chemical electrophiles binding to cellular nucleophiles.
- Project Focus: Adapt the (E)-2-(4-mercaptostyryl)-1,3,3-trimethyl-3H-indol-1-ium thiol (MSTI), a biochemical reactivity assay, to provide semi-quantitative kinetic information, that can be used to predict toxicities mediated by electrophiles binding to cellular nucleophiles.
13. Predictive Toxicology of the Retinoid Signaling Pathway
Leads: Thomas Knudsen, Richard Judson, Ann Richard, Jocylin Pierro, Nancy Baker, Nicole Kleinstreuer, Annie Lumen, Menghang Xia, Patience Browne
Goal: To mine bioactivity profiles from the ToxCast/Tox21 portfolio for retinoid transporters, metabolism, receptors, and responsive pathways that can be formally integrated with embryological knowledge to generate data-driven models of the microphysiology of the retinoid system and provide predictive toxicological information.
- Issue: Conserved cell signaling through all-trans retinoic acid (ATRA)-dependent gene expression has been well documented and shown to have developmental effects on most tissues. Prenatal development is particularly vulnerable to genetic, pharmacological, or chemical disruption of the retinoid pathway. This is especially the case for early gestation when the fundamental body plan is established, and subsequent stages when the regional pattern of specific body segments is decoded. At least 12 assays in the ToxCast/Tox21 portfolio map to molecular targets in the retinoid signaling pathway. A preliminary analysis revealed low-/submicromolar bioactivity on one or more target assays for over 100 structurally diverse ToxCast/Tox21 chemicals (e.g., conazoles, organochlorine pesticides, organotins, retinoids, and pharma compounds) suggesting that they can be used to generate models of the retinoid system and provide predictive toxicological information.
- Project Focus: 1) Formalizing an Adverse Outcome Pathway (AOP) framework for the retinoid system; 2) mapping high-throughput screening data from relevant assays in ToxCast/Tox21 profiles to the AOP framework; and 3) building and testing computational models for quantitative disruption of ATRA signaling.
14. Investigation of Environmental Determinants of Pubertal Timing in Girls
Leads: Natalie Shaw, Menghang Xia, Kristine Witt
Goal: To identify compounds that activate or inhibit the gonadotropin-releasing hormone (GnRH) neurons that control pubertal timing in girls.
- Issue: Over the past decade, there has been a worldwide trend toward earlier breast development in girls. The rapid pace of this trend precludes a genetic etiology and points to environmental factors; however, studies attempting to identify potential triggers by measuring serum endocrine-disrupting chemicals (EDCs) in girls in relation to their pubertal timing have failed to demonstrate compelling associations. This study will be the first to examine the potential effects of EDCs on the neural components responsible for pubertal timing.
- Project Focus: 1) To screen the Tox21 10K compound collection against a human cell line genetically engineered to express the human GnRH receptor, in order to identify compounds that activate or inhibit the receptor; and 2) to confirm the biological relevance of the identified activity and determine the mechanism of action for potential GnRH receptor agonists/antagonists using physiologically relevant cellular or animal models.