New Science/New Paradigms:
Computational Toxicology at
the US EPA
Robert Kavlock
Director, EPAs National Center for Computational Toxicology
Office of Research and Development
National Center for Computational Toxicology
ELI 40th Anniversary
Washington, DC
October 6, 2009
Current Approach for Toxicity Testing
in vivo testing
Cancer
ReproTox
DevTox
NeuroTox
PulmonaryTox
$Millions
ImmunoTox
For a food use pesticide: up to $10m in toxicology, $1m to interpret, years to complete
Office of Research and Development
National Center for Computational Toxicology
EPA’s Need for Prioritization
Too Many Chemicals
Too Little Data (%)
60
9912
10000
50
1000
40
30
100
20
10
10
1
IRIS
Inerts
0
TRI
CCL 1 & 2
Pesticides
HPV
Acute
Cancer
Dev Tox
Repro Tox
Gentox
MPV
Office of Research and Development
Computational Toxicology Research Program
Judson, et al EHP (2009)
The Grand Challenge:
Predicting Human Toxicity
Tissues
Cellular Systems
Tissue
Dose
Molecular
Targets
Molecular
Pathways
Office of Research and Development
National Center for Computational Toxicology
Cell
Changes
Cellular
Networks
Toxicity
Administrator Jackson’s Priorities
• Reducing Greenhouse Gas Emissions
• Improving Air Quality
• Managing Chemical Risks
– “More that 30 years after Congress enacted the Toxic Substances
Control Act, it is clear that we are not doing an adequate job of
assessing and managing risks of chemicals in consumer products, the
workplace and the environment. It is now time to revise and strengthen
EPAs chemicals management and risk assessment programs”
– “…we must be sensitive to the burdens pollution has placed on
vulnerable subpopulations, including children, the elderly, the poor and
all others who are at particular risk to threats to health and the
environment. We must seek their full partnership in the greater aim of
identifying and eliminating the sources of pollution in their
neighborhoods, schools and homes.”
• Cleaning up Hazardous Waster Sites
• Protecting America’s Water
Office of Research and Development
National Center for Computational Toxicology
Source: Memo to EPA Employees, January 23, 2009
http://www.epa.gov/administrator/memotoemployees.html
“…to integrate modern computing and information technology
with molecular biology to improve Agency prioritization of
data requirements and risk assessment of chemicals”
Providing Decision Support Tools for
High-Throughput Screening, Risk Assessment and Risk
Management
Office of Research and Development
National Center for Computational Toxicology
www.epa.gov/ncct
Future of Toxicity Testing
in vitro testing
in silico analysis
Cancer
ReproTox
DevTox
NeuroTox
PulmonaryTox
ImmunoTox
$Thousands
HTS
-omics
Office of Research and Development
National Center for Computational Toxicology
Bioinformatics/
Machine Learning
www.epa.gov/ncct/toxcast
High-throughput screening (HTS):
environmental chemicals
drug development
building from the drug-discovery paradigm
target
identification
HTP in vitro
screening
targeted tests
(validation)
predictive
models
Office of Research and Development
Computational Toxicology Research Program
design
actives
target
identification
animal tests
efficacy, safety
HTP in vitro
screening
test in human
patients
unknown toxicity
(or suspected)
NAS/NRC Consultations
2007
Toxicity Testing in the 21st Century: A Vision and a Strategy
2007
Applications of Toxicogenomic Technologies to Predictive
Toxicology and Risk Assessment
2008
Phthalates and Cumulative Risk Assessment
Science and Decisions-Advancing Risk Assessment
2008
Office of Research and Development
National Center for Computational Toxicology
9
A New Paradigm:
Activation of Toxicity Pathways
Exposure
Tissue Dose
Biologic Interaction
Low Dose
Higher Dose
Higher yet
Perturbation
Normal
Biologic
Function
Biologic
Inputs
Early Cellular
Changes
Adaptive Stress
Responses
Cell
Injury
Morbidity
and
Mortality
Science 7 August 2009
009
ugust 2
EHP A
Office of Research and Development
National Center for Computational Toxicology
Nature
9
July 2
0
09
ToxCastTM Background
• 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
• Committed to stakeholder involvement and transparency
• Communities of Practice- Chemical Prioritization; Exposure
• Release of all data upon peer review publication
Office of Research and Development
National Center for Computational Toxicology
11
Prioritization Product Timeline
FY07
FY08
FY09
Proof of Concept: ToxCast
FY10
FY11
FY12
Verification/Extension
Reduce to Practice
Tox21
Office of Research and Development
Computational Toxicology Research Program
Component of the
Transformation in Toxicology
High Throughput Screening
Information Technology and Management
Pathway Based
Testing
Molecular, Cellular and Systems Biology
Office of Research and Development
National Center for Computational Toxicology
Component of the
Transformation in Toxicology
High Throughput Screening
Information Technology and Management
Pathway Based
Testing
Molecular, Cellular and Systems Biology
Office of Research and Development
National Center for Computational Toxicology
Chemical Classes in
ToxCast_320 (Phase I)
309 Unique Structures
Replicates for QC
291 Pesticide Actives
14 HPVs
54/67 Proposed Tier 1
Endocrine Disruption
Screening Program
Office of Research and Development
National Center for Computational Toxicology
Misc (<4 members)
CHLORINE
ORGANOPHOSPHORUS
AMIDE
ESTER
ETHER
PYRIDINE
FLUORINE
CARBOXYLIC ACID
PHENOXY
KETONE
TRIAZINE
CARBAMATE
PHOSPHOROTHIOATE
PYRIMIDINE
BENZENE
ORGANOCHLORINE
AMINE
PYRETHROID
SULFONYLUREA
TRIAZOLE
UREA
IMIDAZOLE
NITRILE
ALCOHOL
CYCLO
PHOSPHORODITHIOATE
THIOCARBAMATE
ANILINE
THIAZOLE
DINITROANILINE
OXAZOLE
PHOSPHATE
IMINE
NITRO
PHENOL
PHTHALIMIDE
PYRAZOLE
15
SULFONAMIDE
ToxCast In vitro HTS Assays
Biochemical Assays
• Protein families
–
–
–
–
–
–
–
–
GPCR
NR
Kinase
Phosphatase
Protease
Other enzyme
Ion channel
Transporter
• Assay formats
– Radioligand binding
– Enzyme activity
– Co-activator recruitment
467 Endpoints
Office of Research and Development
National Center for Computational Toxicology
Cellular Assays
• Cell lines
– HepG2 human hepatoblastoma
– A549 human lung carcinoma
– HEK 293 human embryonic kidney
• Primary cells
–
–
–
–
–
–
Human endothelial cells
Human monocytes
Human keratinocytes
Human fibroblasts
Human proximal tubule kidney cells
Human small airway epithelial cells
• Biotransformation competent cells
– Primary rat hepatocytes
– Primary human hepatocytes
• Assay formats
–
–
–
–
–
Cytotoxicity
Reporter gene
Gene expression
Biomarker production
High-content imaging for cellular phenotype
ToxCast Phase I Assay Results
(n=624 measurements)
Cell Free HTS
Multiplexed TF
Human BioMap
HCS
qNPAs
XMEs
Impedance
Genotoxicity
Office of Research and Development
National Center for Computational Toxicology
>64,000 dose-response curves
Judson et al, EHP, submitted
Component of the
Transformation in Toxicology
High Throughput Screening
Information Technology and Management
Pathway Based
Testing
Molecular, Cellular and Systems Biology
Office of Research and Development
National Center for Computational Toxicology
ToxRefDB Endpoint Coverage
data evaluation records
ToxRefDB
CHRONIC/CANCER (CHR)
Martin et al. (2008) Environ Hlth Persp
doi:10.1289/ehp.0800074
MULTIGENERATION REPRODUCTIVE (MGR)
Martin et al. (2009) Toxicol Sci
doi: 10.1093/toxsci/kfp080
PRENATAL DEVELOPMENTAL (DEV)
Knudsen et al. (2009) Reprod Toxicol
doi: 10.1016/j.reprotox.2009.03.016
A = Rat
Office of Research and Development
National Center for Computational Toxicology
B = Mouse
SOURCE: Matt Martin, NCCT, 2009
C = Rabbit
Chemicals
Digitizing Legacy in Vivo Data in ToxRefDB
Chronic/Cancer
Multigenation
Developmental
Martin et al 2009a,b
Knudsen et al 2009
Office of Research and Development
National Center for Computational Toxicology
30 years and more than $2B worth of data
Component of the
Transformation in Toxicology
High Throughput Screening
Information Technology and Management
Pathway Based
Testing
Molecular, Cellular and Systems Biology
Office of Research and Development
National Center for Computational Toxicology
Pathway Mapping
Biologically Multiplexed
Activity Profiling (BioMAP)
Multiplex Transcription
Reporter Assay
Cell-based HTS Assays
Cell-free HTS Assays
High Content Cell Imaging
Assays
Office of Research and Development
National Center for Computational Toxicology
ToxCast Assays Mapped to Disease Pathways
Node Size
51-100
31-50
21-30
11-20
4-10
1
Office of Research and Development
National Center for Computational Toxicology
ToxCast Hazard-Based Prioritization
ToxScore = f(Chemical properties + In vitro assays + Pathways + Dosimetry)
(II) Systemic
Chemical properties (e.g. benzene fragment)
Cancer (I)
In vitro assays (e.g. p53 activation)
Pathways (e.g. DNA damage repair)
Dosimetry (e.g. biotransformation)
Features for each toxicity sector are selected for specific
prioritizations; different chemical properties, assays, pathways
and dosimetry for specific types of toxicity testing.
(III) Reproductive
Developmental (IV)
C
I
P
E
1
1
1
1
ToxScore = ∑ w c ∗ chemProp c + ∑ w i ∗ assay i + ∑ w p ∗ pathway p + ∑ w e ∗ dosimetry e
Example ToxScores
prioritizing chemicals
for Cancer, Systemic,
Reproductive or
Developmental testing,
respectively.
ChemicalB
ChemicalA
(I)
Office of Research and Development
National Center for Computational Toxicology
(II)
ChemicalC
(III)
ChemicalD
(IV)
ToxScores of six selected chemicals
along the distribution for all 309
ToxCast Phase-I chemicals
Bisphenol A
highest
Linuron
Perfluorooctane
sulfonic acid
Priority (relative rank)
Symclosene
Rotenone
lowest
Tebuthiuron
Office of Research and Development
lowest
National Center for Computational Toxicology
highest
ToxScore
Tox21 Community
10K
2.8K
1.4K
2008
2006
2004
2005
Tox21 Chemical
Library
2007
2010
2009
FDA
EU JRC
Office of Research and Development
National Center for Computational Toxicology
Expanding ToxCast Chemical Space to
Meet EPA Program Needs
450
400
Phase II
Phase I
300
250
200
150
100
50
OPPTS
OPPTS
OW
OW
OSWER
OSWER
Regions
Regions
FDA
FDA
OECD
OECD
EU
EU
Ph
ar
m
a
N
an
o
PC
C
L
G
re
en
PV
M
H
PV
ic
ro
bi
al
s
A
nt
im
In
er
ts
0
A
ct
iv
es
# chemicals
350
FY2012: thousands of chemicals in Phase III
Office of Research and Development
National Center for Computational Toxicology
Future Chemical Prioritizations
Priority = f(Chemical Descriptors + ToxCast + V-Tissues + Exposure)
Cancer
ToxCast
HO
O
HO
O
O
N
N
S
S
Cl
O Cl
O
Cl
Cl
Cl
Cl
O
O
Cl
Cl
O
O
S
O H3C
P
H3C
SNH O CH3
OO
P
H3C
CH3
NH
O
O
NH
O
N
N
N
Chemical
ChemicalStructures
Structures
N
CH3
NH
NH
NH
O
CH3
O
O
CH3
O
ToxRefDB
CH3
ExpoCast
Production/
Import
Volumes
Source/Stressor
Formation
EPA IUR
EPA TRI
Chemicals in
Consumer
Products
DOE GHG
EU ESIS
Product
Usage
Information
Transport/Fate
Environmental
Releases
EPA HPVIS
EPA Pesticide
Usage Data
Production/
Process
Information
UK Pharmaceutical Usage
Household
Products DB
Cosmetic
Voluntary
Reg. DB
Environmental Environmental
Transformation Fate Simulator
EPA HPVIS
ATSDR Tox
Profiles
ECOTOX DB
DEA NFLIS
M
NR
Signaling
Environmental
Concentration
Exposure
Indoor Air
Monitoring
Data
Human
Exposure
Monitoring
DOE IndoorAir
NHEXAS
CTEPP
NHEXAS
EUROPA PestDiet
Outdoor Air
Monitoring
Data
Activity
Patterns
Information
Cell Cycle
Xenobiotic
Metabolism
CESAR
EPA NATA
EPA AIRS/AFS
Exposure
Limits
EPA NHAPS
EPA CHAD
Human
Biological
Monitoring
Data
Proliferation
G0
G2
G1
AIHA WEEL
OSHA PEL
UN IPCC GHG
MAPK
Signaling
NIOSH NOES
S
Lipid
Metabolism
Mitochondrial
Function
CDC NHANES
Necrosis
Apoptosis
NHEXAS
CTEPP
Office of Research and Development
National Center for Computational Toxicology
Oxidative
Stress
DNA Damage
DNA Damage
Sensing
Repair
Virtual Tissues
ToxMiner
Cl
Developmental
H3C
Cl
Reproductive
From Research to Practice …
Providing High Throughput Decision Support Tools for Screening and
Assessing Chemical Exposure, Hazard, and Risk
FY09
FY10
FY11
FY12
FY13
FY14
Informatics: DSSTox, ToxRefDB, ACToR, MetaPath, ExposureDB, VT‐KB
Prioritization: ToxCast, Tox21, ExpoCast, V‐Tissues, ToxMiner
Risk Assessment…
Office of Research and Development
National Center for Computational Toxicology
29