ORD’s Computational Toxicology Research Program Tox Methods Nov 2005/2 Dix Session 1.pdf• Build upon examples where mode/mechanism of action has already, or is being, employed

ORD’sComputational Toxicology

Research Program

17 November, 2005Int’l Society of Regulatory Toxicology & Pharmacology

David DixNational Center for Computational Toxicology

US Environmental Protection AgencyResearch Triangle Park, NC

Enabling “Omic” Technologies Leading to aSystems Approach and the Premise of

Computational Toxicology

DNA

mRNA

Protein

Metabolites

Transcription

Translation

Metabolism

Cell Biology

Enabling HTS TechnologiesDeveloped in the Search for

Bioactive Compounds

• 96 to 384 to 1536 robotics revolution

• Compound/chemical libraries >106

• Assay development and miniaturization

• Computational tools for management andanalysis of large volumes of data

Reasons for Building a ComputationalToxicology Program: Regulatory

Challenges Facing EPA











Program Development

FY02

FY03

FY04

FY05

Initializing

Congressional redirection

EDC Proof of Concepts

Building Foundation

Design Team

Framework document

SAB and BOSC reviews

RTP Workshop

Bibliographic Inventory

Website

STAR HTPS RFA

Implementing

CTISC

Proof of Concepts Expansion

Internal competitions

STAR Systems Biology RFA

Institutionalizing

National Center formed

BOSC Site Visit

Prioritization Initiative: ToxCast Concept

FY06

Operating

STAR Informatic Centers

Implementation Plan

ToxCast Development

Program Development: A Framework forORD’s Computational Toxicology ResearchThemes: Technology-based, hypothesis-driven

effort to increase the soundness ofEPA risk assessments

Capacity to prioritize, screen andevaluate chemicals by enhancing thepredictive understanding of toxicitypathways or potential

Success: Ability to produce faster and more

accurate risk assessments for lesscost, and to classify chemicals bypotential to affect molecular andbiochemical pathways of concern

www.epa.gov/comptox

EnvironmentalRelease

EnvironmentalConcentration

Exposure Concentrations

Target Organ Dose

Toxicity Pathway

Adverse Outcome

Fate/TransportModels/Data

ExposureModels/Data

PBPKModels/Data

BBDRModels/Data

SystemsModels/Data

The NCCT isproviding

linkages alongthe source to

outcomecontinuum

Program Development: Formation andStaffing of a National Center for

Computational Toxicology

Link

ages

Prioritization

QRA

NCCT

I. Information TechnologyA RichardTBN (Title 42)TBN

II. Prioritization and ScreeningD DixR Kavlock S Little M Martin (intern)M Pasquinelli*J RabinowitzTBN

III. Biological ModelsH BartonJ Blancato

M Breen*R ConollyTBN (Title 42)TBNTBN*TBN*

IV. Cumulative RiskE HubalW SetzerTBN*

Administrative Support Karen Dean, Program Analyst

Sandra Roberts, Exec Secretary

Program Development: Implementing aCross-ORD Research Portfolio

Link

ages

Prioritization

QRA

Metabolic Simulator

DSSTox

Diesel Particles

Microbial Metagenomics

Amphibian Systems Model

Fish ToxicogenomicsASTER

Fish Proteomics

Conazole MOA

Children’s HealthPulmonary Biomarkers

Fish metabonomics

Pellston, SOT and NCEA Workshops

STAR HTPS RFA

STAR Systems Models

STARInformatics

Centers

(UNC and UMDJ)

HPG Axis Model

H295R AssayIconix Contract

ER Binding DataMolecular DockingER/AR Scale upToxCast

CompToxResearch Program

NCCT

Computational/Omic Research

ORDResearch

National Center for ComputationalToxicology (NCCT) is part of a larger whole…

ToxCast– A Prioritization Concept• Assumptions

Prioritization/Categorization is needed Prioritization is not equivalent to screening Need broad coverage of potential outcomes Outcomes mediated by chemical-biological interactions There is no current model Technological advances can be employed (e.g., HTS) Cost is a factor in acceptance

• Pharmaceutical experience is helpful, but caveats Focused on targets Accepts a high false negative rate “Activity” levels higher than for environmental chemicals

• Build upon examples where mode/mechanism of action hasalready, or is being, employed in hazard or risk assessment

Linkages

Prioritization

QRA

NCCTLinkages

Prioritization

QRA

NCCT

ToxCast Information Domains

……pKaLogP

Physical-Chemical Indicators

Chem 1

Chem 2

Chem 3

Chem 4

.

.

.

Example: the Persistent, Bioaccumulative,and Toxic (PBT) Profiler


……pKaLogP


……PredictedReceptorDocking

SAR

Bio-Computational Indicators

Chem 1

Chem 2

Chem 3

Chem 4

.

.

.

http://www.ibmc.msk.ru/PASS/

SAR: Prediction of ActivitySpectra for Substances (PASS)

Rabinowitz et al, Unpublished

Clustering of Potential Steroid Receptor Ligands Basedon Predicted Binding to Nuclear Receptors


……pKaLogP


……MolecularReceptor

Docking #1PASS


……Kinase

InhibitionGPCRs

Biochemical Based IndicatorsChem 1

Chem 2

Chem 3

Chem 4

.

.

.

PNAS January 11, 2005 vol. 102 no. 2 261–266CHEMISTRY PHARMACOLOGY

1576 compoundsX 92 assays

Azole cluster bybiospectra similarity


……pKaLogP



Docking #1PASS


……HMG-CoAreductase

GPCRs

Biochemical Based Indicators

……PC12H4IIE

Cellular Based Indicators

Chem 1

Chem 2

Chem 3

Chem 4

.

.

.

Example is a reporter vector is used to monitor transactivation of PPAR.

http://www.panomics.com

Cell Based Assays Monitoring Effects ofChemical-Nuclear Receptor Interactions

High Content Screening

Requiresautomated imageanalysis tools forhigher throughputscreening

http://www.cellomics.com/


……pKaLogP



Docking #1PASS



GPCRs


……H295RH4IIE


……H295RPrimary

Heptatocytes

In Vitro Omics Indicators

Chem 1

Chem 2

Chem 3

Chem 4

.

.

.

NCI-60 Project(Scherf et al, Nature Genetics 24:236 (2000))

1376 genes in 60 cell lines

118 test compounds

Zhang et al. 2005. Quantitative RT-PCR Methods for Evaluating Toxicant-Induced Effectson Steroidogenesis Using the H295R Cell Line. Environ. Sci. Technol. 39:2777-2785.

H295R Steroidogenesis and Gene Expression


……pKaLogP



Docking #1PASS



GPCRs


……H295RH4IIE


……StemCells

PrimaryHeptatocytes


……Signature 2Signature 1

In Vivo Omics Indicators

Chem 1

Chem 2

Chem 3

Chem 4

.

.

.

http://www.altheatech.com


……pKaLogP



Docking #1PASS



GPCRs


……H295RH4IIE


……StemCells

PrimaryHeptatocytes


……Signature 2Signature 1

In Vivo Omics IndicatorsBin ….

….

….

Bin 3

Bin 2

Bin 1

ChemicalGrouping

Increasing Biological Relevance

Increasing Cost

MLSCN Center at Columbia University

Emory Chemistry-Biology Center in the MLSCN

Southern Research Molecular Libraries Screening Center (SRMLSC)

San Diego Chemical Library Screening Center

Scripps Research Institute Molecular Screening Center

New Mexico Molecular Libraries Screening Center

The Penn Center for Molecular Discovery

University of Pittsburgh Molecular Libraries Screening Center

Vanderbilt Screening Center for GPCRs, Ion Channels, and Transporters

NIH Chemical Genomics Center (NCGC)

Molecular Libraries Screening Centers Network(MLSCN) Complementary HTS Assays

GPCRs EnzymesProteases

protein-protein interactions

High Content Screening

High-throughput Microscopy

NMR-based methods

Protein misfolding and degradation Complementary

HTS

HTS Flow Cytometry

Protein Kinases

Ion channelsTransporters

Kalypsys

BSL-3

Courtesy of Chris Austin

HTS Assay Target Classesin 64 Applications in Response to Assay Solicitation PAR-05-060

Enzyme

Ion channel

Nuclear Receptor

Cell- based assay

Other

GPCR

Enzymetransferase, polymerase,

proteases, kinase

GPCR

Cell-based assayCytotoxicity/Viability assay, HCS, neurite outgrowth

Other

Ion Channel 3%

Nuclear Receptor 2%

Yeast-based assay, Zebrafish, Protein-protein

29%

16%27%

23%

Absorbance, Fluorescence, FP, FRET, FLIPR, AlphaScreen , InCell 3000, Arrayscan Cellomics, Flow Cytometry, and Microarray.


MLSCN Operation

Investigators

HTS AssayApplications

Peerreview

CandidateBiomodulatorOptimization

Chemistry,Probes

Selectionby PT,Assigned toCenter

Assay Optimizationand Screening

SmallMoleculeRepository(SMR)

Screening DataCourtesy of Chris Austin

NIH Chemical Genomics Center

• Launched June 2004

• Lab operational Feb 2005 9800 Medical Center Drive,

Rockville

Co-housed with ImagingProbe Development Center(ML6)

• http://www.ncgc.nih.gov/

Discovering the PathwayBetween Chemistry and Biology


Results from NCGC in PubChem

ToxCast- Prioritization Issues

• Consensus building Briefings

• ORD, EPA Program Offices, NTP, NIH MLI, ACC, CropLife Presentations at meetings

• Toxicology Forum, World Congress on Alternatives, ISRTP,Keystone, SOT, TestSmart

• Need Proof of Concept demonstration Chemical selection process Assay selection process Chemical management

• Lack of metabolism in many potential assays• Coverage of developmental susceptibility

• Use of the NAS Signaling Pathways??• Key partnerships: NTP, NCGC…

Toxcast- Potential Outcomes

• Ability to categorize or prioritize chemicals Tool box of indicators across information domains Cost effective approach for assessing potential to be biologically

active agents Potential targeting of outcomes of concern

• Flexibility Adaptable to technological advances Refinement of key indicators with experience

• Development of predictive models as database enlarges• Involvement with Green Chemistry initiatives• Possible foundation for more effective and efficient use of

animals in testing

Linkages

Prioritization

QRA

NCCTLinkages

Prioritization

QRA

NCCT

Documents

ORD’s Computational Toxicology Research Program Tox Methods Nov 2005/2 Dix Session 1.pdf• Build upon examples where mode/mechanism of action has already, or is being, employed