Office of Research and Development National Center for Computational Toxicology Keith Houck, PhD ISRPT 2009 Endocrine Workshop 9-10 Sept 2009 What Can ToxCast Tell Us About Endocrine Disruption? This work was reviewed by EPA and approved for presentation but does not necessarily reflect official Agency policy. Mention of trade names or commercial products does not constitute endorsement or recommendation by EPA for use.

Office of Research and Development National Center for Computational Toxicology 1 What Can ToxCast Tell Us About Endocrine Disruption? Probably not much but our focus is chemical prioritization for targeted testing

Office of Research and Development National Center for Computational Toxicology 22 Too Many ChemicalsToo High a Cost Cancer DevTox NeuroTox ReproTox ImmunoTox PulmonaryTox Millions $ Our Challenge: 11,000 90,000 …and not enough data. Judson, et al EHP 117(5): (2009)

Office of Research and Development National Center for Computational Toxicology 3 ToxCast TM Background  Research program of EPA’s National Center for Computational Toxicology  Addresses chemical screening and prioritization needs for pesticidal inerts, anti-microbials, CCLs, HPVs and MPVs  Comprehensive use of HTS technologies to generate biological fingerprints and predictive signatures  Coordinated with NTP and NHGRI/NCGC via Tox21  Committed to stakeholder involvement and public release of data  Communities of Practice- Chemical Prioritization; Exposure  NCCT website-  ACToR- Aggregated Computational Toxicology Resource 

Office of Research and Development National Center for Computational Toxicology 4 MOA Classes with > 3 chemicals Acetylcholine esterase inhibitors conazole fungicides Sodium channel modulators pyrethroid ester insecticides organothiophosphate acaricides dinitroaniline herbicides pyridine herbicides thiocarbamate herbicides imidazolinone herbicides organophosphate insecticides phenyl organothiophosphate insecticides aliphatic organothiophosphate insecticides amide herbicides aromatic fungicides chloroacetanilide herbicides chlorotriazine herbicides growth inhibitors organophosphate acaricides oxime carbamate insecticides phenylurea herbicides pyrethroid ester acaricides strobilurin fungicides unclassified acaricides unclassified herbicides Misc 309 unique structures Replicates for QC 8 metabolites 291 total pesticide actives 273 registered pesticide actives 22 pesticide inerts 33 antimicrobials 54 Tier 1 EDSP 23 IUR 13 HPV 73 OW PCCL 11 CCL1 10 CCL2 25 CCL3 ToxCast_320 Phase I Chemicals

Office of Research and Development National Center for Computational Toxicology 5 EDSP Chemicals in ToxCast 320 Select Other Chemicals in ToxCast 320

Office of Research and Development National Center for Computational Toxicology 6 Types of Endocrine Assays Tested In ToxCast –Radioligand Receptor Binding (NRs) Single conc at 25  M followed by 7-point concentration response for actives –Multiplexed Reporter Gene Assay (Attagene) Trans: GAL4-Ligand Binding Domain NR Assays Cis: NR Response Element Assays 5-point concentration-response up to 100  M (unless limited by cytotoxicity) –Nuclear Receptor Coactivator Recruitment (NRs) Single conc at 25  M followed by 7-point concentration response for actives –CYP Enzyme Activity Assays Single conc at 10  M followed by 7-point concentration response for actives –NR Target Gene Assays (in primary human hepatocytes) 5-point concentration-response up to 40  M

Office of Research and Development National Center for Computational Toxicology 7 Endocrine Related Assays

Office of Research and Development National Center for Computational Toxicology 8 Other Endocrine Related Assays

Office of Research and Development National Center for Computational Toxicology 9 Other Endocrine Related Assays

Office of Research and Development National Center for Computational Toxicology 10 Example NR Binding Assay Actives hER  hAR hTR 

Office of Research and Development National Center for Computational Toxicology 11 Overview of Assay Results Generally found expected actives for ER  and AR binding assays Little activity for Thyroid Receptor Some species and/or assay sensitivity differences seen, e.g. low sensitivity of rat AR binding assay Not always agreement between binding and cellular reporter assays Many CYP inhibitors including those from expected chemical classes, i.e. conazoles and organophosphates Potency mostly in the micromolar to tens of micromolar range; few nanomolar actives

Office of Research and Development National Center for Computational Toxicology 12 Tox21 Tripartite Collaboration: EPA/NTP/NCGC Quantitative HTS Assays at NCGC Accepting Nominations for Assays from All Sources Chemical Libraries Being Screened: –EPA Library of 1462 Substances Contains Toxcast 309 Contains many NR reference compounds, e.g. pharmacological tools and drugs Many different environmental chemicals, e.g. bisphenols, phthalates, PFAAs –1408 substances in NTP library also screened –Expanding to 10,000 this year including all human pharmaceuticals

Office of Research and Development National Center for Computational Toxicology 13 Tox21 Assays Nuclear Receptor Cellular Reporter Gene Assays –GAL4-Ligand Binding Domain System--principally ligand detection –Receptors screened in agonist and antagonist modes –11-15 concentrations (highest 92  M) antagonist mode data interpretation confounded by cytotoxicity –Receptors screened: Human –ERa, AR, TRa, VDR, MR, RXRa, PPARa, PPARb, PPARg, PXR, (AhR) Rat –PXR

Office of Research and Development National Center for Computational Toxicology 14 Tox21 Results EPA Library Actives: Agonists

Office of Research and Development National Center for Computational Toxicology 15 Tox21 EDSP Chemicals Agonist (All)

Office of Research and Development National Center for Computational Toxicology 16 Tox21 EPA Library Antagonist Actives

Office of Research and Development National Center for Computational Toxicology 17 Partial Agonist Examples: ER 

Office of Research and Development National Center for Computational Toxicology 18 Importance of Biotransformation ER  radioligand binding assay ER  cellular (HEK293) transactivation assay

Office of Research and Development National Center for Computational Toxicology 19 Cellular Assays Biochemical Assays Toxicology Endpoints Physical chemical properties Data Normalization In vivo/In vitro Correlation Genomic Signatures In silico Predictions

Office of Research and Development National Center for Computational Toxicology 20 ToxRefDB website:

Office of Research and Development National Center for Computational Toxicology 21 Reproductive and Endocrine Organ Toxicity Endpoints from ToxRefDB

Office of Research and Development National Center for Computational Toxicology 22 Using the Data for Prioritization In vitro assays (ToxCast) In vivo endpoints (ToxRefDB) Chemical properties (descriptors) Pathways (endocrine)

Office of Research and Development National Center for Computational Toxicology 23 Using the Data for Prioritization Perfluorooctane sulfonic acid Symclosene Linuron FenitrothionBisphenol A

Office of Research and Development National Center for Computational Toxicology 24 Summary of Preliminary Findings Suite of assays useful in prioritization –endocrine-active chemicals score higher across assays –Individual assays may not be sufficient on their own Moderate concordance across different assay types for same target –Differences in biotransformation between assays –Assay-specific false positives/false negatives –Error –Differences in biology we don’t understand Sensitivity range of different assay types presents a challenge to interpret biological significance

ACKNOWLEDGEMENTS:  NCCT/ORD/USEPA  Matt Martin  David Dix  Richard Judson  David Reif  Ann Richards  Robert Kavlock  NCGC/NIH  Menghang Xia  Ruili Huang  Chris Austin