using high throughput assays and tools for endocrine screening · incorporating vald dlidated...
TRANSCRIPT
U.S. Environmental Protection AgencyEndocrine Disruptor Screening Program
Using High Throughput Assays and Using High Throughput Assays and
p g g
Computational Tools for Endocrine ScreeningComputational Tools for Endocrine Screening
Patience Browne, Ph.D.Senior Scientist, Office of Science Coordination and Policy
Patience Browne, Ph.D.Senior Scientist, Office of Science Coordination and Policy, y
Office of Chemical Safety and Pollution [email protected]
, yOffice of Chemical Safety and Pollution Prevention
*does not necessarily reflect EPA policy*does not necessarily reflect EPA policy
September 18, 2015September 18, 2015
Screening – Tier 1Tier 1 Screening Battery
Endocrine Pathway
E + E ‐ A + A ‐HPGAxis
HPTAxis
In vitro
ER Bi diER Binding ■ ■
ERα Transcriptional Activation* ■
AR Binding ■ ■
Steroidogenesis H295R* ■ ■ ■ ■
Aromatase Recombinant ■
In vivo
Uterotrophic* ■
Hershberger* ■ ■
Pubertal Male ■ ■ ■ ■ ■ ■
Pubertal Female ■ ■ ■ ■ ■ ■
Amphibian Metamorphosis* ■
Fish Short‐term ReproductionFish Short‐term Reproduction(male & female)*
■ ■ ■ ■ ■
*OECD harmonized guidelines
EDSP List 2
Evolution of EDSP
EDSP Chemical
EDSP List 2107 Chemicals
Universe10,000 chemicals(FIFRA & SDWA)
EDSP List 167 Chemicals
Based on current pace it could take decades to screen all 10,000 chemicals in EDSP Universe
Use predictive models and high throughput assasy to rapidly screen chemicals for potential bioactivity and exposure
EDSP Goals
Use computational tools and models in the EDSP framework to:1. Contribute to the weight of evidence evaluation of a chemical’s g
potential bioactivity2. Provide alternative data for specific endpoints in the EDSP Tier 1
battery
Ultimately, these goals are common to the estrogen, androgen and thyroid pathways, however, estrogen bioactivity is the most mature model and is used to demonstrate the proposed approach.
4
Estrogen Receptor Bioactivity Model
ER bioactivity modelER bioactivity model• 18 HTS assays
Detect receptor interaction at various Detect receptor interaction at various points along signaling pathway Use a variety of technologies
• Capable of distinguishing “true” activity from cytotoxicity
Values range from 0 to 1• ER agonists/antagonists
EDSP Tier 1 ER Agonist Assays vs. ToxCast ER Agonist Assays
ToxCast High Throughput Screening ER assaysEDSP Tier 1 assaysAssay Name Biological Process Organism TissueNVS_NR_bER receptor binding bovine uterusNVS_NR_hER receptor binding human NANVS_NR_mERa receptor binding mouse NAOT_ER_ERaERa_0480 protein complementation human kidney
Assay Name Biological Process Organism Tissue
RUC ER binding receptor binding rat uterus
OT_ER_ERaERa_1440 protein complementation human kidneyOT_ER_ERaERb_0480 protein complementation human kidneyOT_ER_ERaERb_1440 protein complementation human kidneyOT_ER_ERbERb_0480 protein complementation human kidneyOT_ER_ERbERb_1440 protein complementation human kidneyOT_ERa_EREGFP_0120 gene expression human cervixOT_ERa_EREGFP_0480 gene expression human cervixATG_ERa_TRANS_up mRNA induction human liverATG_ERE_CIS_up mRNA induction human liverTox21_ERa_BLA_Agonist_ratio gene expression human kidneyT 21 ER LUC BG1 A i i h
ERTA gene expression human Ovary
Cell proliferation/ Tox21_ERa_LUC_BG1_Agonist gene expression human ovaryACEA_T47D_80hr_Positive cell proliferation human breast
Uterotrophic Cell proliferation/organ response
Rodent uterus
ER ModelER Agonism only
(agonism and antagonism)
Judson et al. 2015. Tox Sci
“Integrated Model of Chemical Perturbations of a Biological Pathwayg y
Using 18 In Vitro High Throughput Screening Assays for the Estrogen Receptor"
Kleinstreuer et al. 2015; EHP“A Curated Database of Rodent
Uterotrophic Bioactivity"
Browne et al. 2015; ES&T
Federal Register Notice, June 2015“EPA Solicits Comments on Use of High‐Throughput
Assays and Computational Tools in Endocrine Disruptor Screening Program”
http://www.regulations.gov/#!documentDetail;D=EPA‐HQ‐OPPT‐p g g ;2015‐0305‐0001.
EPA is planning to incorporate scientific advancements and new tools l d d h h h h d lincorporating validated high‐throughput assays and a computational
model as an alternative for some of the current assays in the EDSP Tier 1 battery.
Currently, EPA has partial screening results for over 1,800 chemicals that have been evaluated using the high‐throughput assays and computational model for the estrogen receptor pathway.p g p p y
Performance Based Approach to Establish Scientific ConfidenceScientific Confidence
Reference chemical set that includes a range of e e e ce c e ca set t at c udes a a ge ostructures and potencies that are accurately detected• in vitro reference chemicals• In vivo reference chemicalsN h d d i h h d New methods compared with current methods• Bioactivity model versus Tier 1 results
Evaluated by independent, external peer review• FIFRA Scientific Advisory Panel meetings• FIFRA Scientific Advisory Panel meetings
ER bioactivity model: in vitro reference chemicals40 i it ER f h i l ith 40 in vitro ER reference chemicals with independently confirmed activity (OECD 2012)
# True Pos 26 (25)# True Neg 11 (11)# False Pos 1 (0)# False Neg 2 (2)Accuracy 0.93 (0.95)Sensitivity 0.93 (0.93)S ifi i 0 92 (1 0)Specificity 0.92 (1.0)
Judson et al. 2015, Tox SciBrowne et al. 2015, ES&T
ER bioactivity model: in vivo reference chemicals
43 in vivo ER reference chemicals with independently confirmed activity (Kleinstreuer et al. 2015)• “Active” >1 POS result• Active >1 POS result “Inactive” >1 NEG result and no POS result “equivocal” > 1 POS result and >1 NEG result
# True Pos 29 (29)# True Neg 8 (8)# False Pos 5 (1)# False Pos 5 (1)# False Neg 1 (1)Accuracy 0.86 (0.95)Sensitivity 0.97 (0.97)y ( )Specificity 0.67 (0.89)
Browne et al. 2015, ES&TKleinstreuer et al., 2015, EHP
ER bioactivity model versus Tier 1 ER model performs as well or better than existing methodsER model performs as well or better than existing methods Model evaluated with 45 reference chemicals
• T1 ER binding: 23 (35% were not were not consistent with expected outcome)• T1 ERTA: 12• T1 UT: 7
ER model in 100% agreement with Tier 1 ER, ERTA, and Uterotrophic results for List 1 chemicals (very low or no ER activity)
ER model may be more sensitive than Tier 1 assays due to redundancyER model may be more sensitive than Tier 1 assays due to redundancy
ER Agonist Activity
Path Forward Determine how well existing models predict intact animal results• Comparison to other Tier 1 endpoints• Additional Tier 1 assay substitution?
U dditi l t ti l t l t d l Use additional computational tools to develop Adverse Outcome Pathway models for estrogen, androgen, and thyroid hormones• Integrate more assays• Integrate more key events
Expand reference chemicals with defined potencies Expand reference chemicals with defined potencies for performance based test guidelines incorporating computational tools• Curation of high quality in vivo data from peer reviewed literature
EDSP Tier 1 AR Assays vs. ToxCast AR Assays
ToxCast High Throughput Screening ER assaysEDSP Tier 1 assaysAssay Name Biological Process Organism Tissue
NVS_NR_hAR receptor binding human NANVS_NR_rAR receptor binding rat prostateOT_AR_ARSRC1_0480 protein stabilization human kidneyOT_AR_ARSRC1_0960 protein stabilization human kidney
EDSP Tier 1 assaysAssay Name Biological Process Organism Tissue
RPC AR binding receptor binding rat prostate
ATG_AR_TRANS_up transcription factor human liverTox21_AR_BLA_Agonist_ratio gene expression human kidneyTox21_AR_LUC_MDAKB2_Agonist gene expression human breastTox21_AR_BLA_Antagonist_ratio gene expression human kidneyTox21_AR_LUC_MDAKB2_Antagonist gene expression human breast
Hershberger organ response Rodent AST
AR Model( i / i )
AR Agonism/Antagonism
(agonistm/antagonism)
Evolution of Screening and Testing in the EDSPEDSP Tier 1 Battery of Assays High Throughput Assays and Computationaly y
(current)g g p y p
Model Tier 1 Battery AlternativesEstrogen Receptor (ER) Binding ER Model (alternative)
Estrogen Receptor Transactivation (ERTA) ER Model (alternative)Uterotrophic ER Model (alternative)Uterotrophic ER Model (alternative)
Androgen Receptor (AR) Binding AR Model (Future)Hershberger AR Model (Future)Aromatase STR Model (Future)
St id i (STR) STR M d l (F t )Steroidogenesis (STR) STR Model (Future)Female Rat Pubertal ER, STR , THY Models (Future)Male Rat Pubertal AR, STR , THY Models (Future)
Fish Short Term Reproduction ER, AR, STR Models (Future)Amphibian Metamorphosis THY Model (Future)
EDSP Tier 2 Tests High Throughput Assays and ComputationalModel Tier 2 Battery Alternatives
Rat 2‐gen/EOGRT ER, AR, STR, THY(Future)g / , , , ( )MEOGRT ER, AR, STR (Future)LAGDA THY (Future)Quail ER, AR, STR, THY (Future)
ER = estrogen receptor; AR = androgen receptor; STR = steroidogenesis; THY = thyroid
Metabolism
• Metabolite prediction software Generation of likely metabolites f fi h Ph I bi t f ti t h P450
Parent chemicals run Parent + 1o metabolites + 2ometabolites run through ER
Bioactivity of parent‐metabolite pairs evaluated
for five human Phase I biotransformation cytochrome P450 enzymes (cyps)
through metabolism predictor software•Phase I metabolism only•CYP dependent•Parent‐metabolite pairs identified
metabolites run through ER QSAR•List QSAR models – trained on 1800 ToxCast ER results•“Reference chemicals” of parent compounds with known estrogenic
metabolite pairs evaluated•QSAR results from several models normalized and average•Relative predicted estrogen agonist bioactivity examined
in literature used to validate metabolites used to validate
Parent ChemicalParent Chemical
Primary Metabolite(s)
Secondary Metabolite(s)X
QSAR ER model Prediction of Activity of i b li
Parent compound Estrogenic Metabolite
Estrogenic Metabolites Methoxychlor Mono-hydroxymethoxychlor
HPTEMestranol 17α-ethynylestradiolTrans-stilbene Trans-4-hydroxystilbene
Trans-4,4'-dihydroxystilbeneAzobenzene 4-HydroxyazobenzeneDiphenyl 4-Hydroxydiphenyl
3-Hydroxydiphenyl3-Hydroxydiphenyl4,4'-Dihydroxydiphenyl
Diphenylmethane 4-Hydroxydiphenylmethane4,4'-Dihydroxydiphenylmethane
2,2-Diphenylpropane 2-(4-Hydroxyphenyl)-2-phenylpropane4,4'-Dihydroxydiphenylpropane
Permethrin 3-Phenoxybenzylalcohol3-(4’-Hydroxyphenoxy)-benzyl alcohol3-Phenoxybenzaldehyde
Cypermethrin 3-PhenoxybenzaldehydeBenzo[a]pyrene 3-Hydroxybenzo[a]pyrene
9-Hydroxybenzo[a]pyrene8-Hydroxybenzo[a]pyrene7-Hydroxybenzo[a]pyrene
Benz[a]anthracene 3-Hydroxybenz[a]anthracene4-Hydroxybenz[a]anthracene4-Hydroxybenz[a]anthracene10-Hydroxybenz[a]anthracene
Chrysene 1-Hydroxychrysene2-Hydroxychrysene3-Hydroxychrysene
Fluorene 2-HydroxyfluorenePhenanthrene 2-Phenanthrol
9-PhenanthrolPyrene 1-HydroxypyreneNaphthalene 1-Naphthol
2-NaphtholFluoranthene 3-HydroxyfluorantheneBenzophenone 4-HydroxybenzophenoneBenzophenone 3 2,4-Dihydroxybenzophenone
2,3,4-Trihydroxybenzophenone Metabolites failed to be predicted are indicated in red
Summary Pivot to using high throughput and computational methods in EDSP
Endocrine pathway models will continue to be revised Endocrine pathway models will continue to be revised and improved as more data are available (ER, AR, thyroid…)
id bi i i di i f h d f h i l• Provides bioactivity predictions for thousands of chemicals Allows resources to be focused on chemicals more likely to have endocrine effectsy• List 1 chemicals have limited estrogen and/or androgen receptor‐mediated bioactivity
• Prioritizes chemicals based on bioactivity (and exposure)Prioritizes chemicals based on bioactivity (and exposure)• Provides alternative to current Tier 1 screening
Multi‐century project becomes multi‐year
Acknowledgements
EPA OSCP• Kristan Markey• Kristan Markey• Caroline Pinto• David Dix• Jane Robbins
EPA ORD• Richard JudsonRichard Judson• Rusty Thomas• John WambaughNIEHS/NICEATM NIEHS/NICEATM• Warren Casey• Nicole Kleinstreuer (NICEATM/ILS)( / )