Download (1152Kb)

EPA Perspectives on 

Nanoinformatics: Prioritization 

Based on Potential for 

Exposure and Toxicity 

Sumit Gangwal, Ph.D. 

R. Judson, A. Wang, K. Houck, E. Cohen Hubal 

EPA National Center for Computational Toxicology (NCCT) 

The views expressed in this presentation are those of the author(s) and do not 

necessarily reflect the views or policies of the U.S. Environmental Protection Agency. 

Office of Research and Development (ORD) 

www.epa.gov 

Nanoinformatics 2010 workshop 

Nov. 4

Overview 

• Incorporating nanomaterials into ToxCast project TM 

– EPA nanomaterial research strategy 

– ToxCast project 

– Comptox toxicity and exposure research on nanomaterials 

• Databases and tools developed by CompTox 

– ACToR 

– ToxMiner 

– ExpoCast DB and exposure data curation 

– Virtual Tissue Knowledgebase (VT-KB) 


www.epa.gov 

1

EPA nanomaterial research strategy 

• Four main research themes: 

– Identifying sources, fate, transport, and 

exposure 

Comptox 

– Understanding human health and 

ecological effects to inform risk 

assessments and test methods 

– Developing risk assessment 

approaches 

– Preventing and mitigating risks 

June 2009 

http://www.epa.gov/nanoscience/files/ 

nanotech_research_strategy_final.pdf 


www.epa.gov 

2

Many nanomaterials to evaluate, 

but limited time and resources 

• Toxicity and exposure research 

is challenged to keep up with 

development of novel 

nanomaterials and applications 

• Assesment of nanomaterial 

(NM) like chemicals is typically 

case-by-case 

Case-by-case examples: 

Comprehensive Environmental 

Assessment approach; 

Life cycle assessment; 

State of Science Review 

Comprehensive Environmental 

Assessment Approach 

CNT 

Ag 

TiO 2 

ZnO 

SiO 2 

CeO 2 

• Prioritization of research and 

screening level assessment of 

NMs are needed. 

Screening level assessment: 

ToxCast TM : Bioactivity profiling + 

exposure potential 

(EPA Comptox) 


www.epa.gov 

3

EPA working on larger problem 

for chemicals 

Too Many Chemicals Too Little Data (%) 

9912 

60 

10000 

1000 

100 

10 

1 

50 

40 

30 

20 

10 

0 

IRIS TRI Pesticides 

Inerts CCL 1 & 2 HPV 

MPV 


www.epa.gov 

Acute Cancer Gentox 

Dev Tox 

Repro Tox 

…and costs too much. 

Judson et al., 2009, Environ. Health Perspect. 

4

Chemicals 

(n = 320) 

ToxCast project: Diversity of 

in vitro data from HTS assays 

• 500 fast, automated chemical screens (in vitro) generating lots of data 

• Phase 1: Screened 300+ well characterized chemicals (primarily pesticides) 

• Builds statistical and computer models to forecast potential chemical toxicity 

Assays 

(n = 467) 

Biochemical Assays 

• Protein families 

– GPCR 

– NR 

– Kinase 

– Phosphatase 

– Protease 

– Other enzyme 

– Ion channel 

– Transporter 

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 


www.epa.gov 

http://www.epa.gov/ncct/toxcast/ 

Judson et al., 2010, Environ. Health Perspect. 

• Biotransformation competent cells 

– Primary rat hepatocytes 

– Primary human hepatocytes 

• Assay formats 

– Cytotoxicity 

– Reporter gene 

– Gene expression 

– Biomarker production 

– High-content imaging for cellular 

phenotype 

5

Steps to include NMs in ToxCast 

• Classes of NM of interest: Au, Ag, CNT, TiO 2 , CeO 2 , ZnO, SiO 2 

* Initial pilot materials 

• Major steps: 

– Develop handling protocols 

• Compare protocols used in Center for Environmental 

Implications of NanoTechnology (CEINT) at Duke Univ., 

ENPRA, and Japan NIST 

– Determine concentration ranges to test 

• Select based on potential for real world human exposures 

– Characterize NMs 

• CEINT at Duke Univ. 

– Perform High-throughput screening (HTS) 

• Analyze HTS data and apply ToxCast methodology 

TEM image of TiO 2 


www.epa.gov 

6

Deposition Fraction 

Deposition Fraction 

Using Multiple-path particle dosimetry 

model to determine concentrations 

• Open-source computational MPPD modeling 

tool (Applied Research Associates) 

– Calculates human respiratory tract particle 

deposition/clearance after inputing NM aerosol 

conc. 

• Reviewed literature on NM aerosol concentrations 

in occupational settings 

– Typically < 0.1 mg/m 3 for TiO 2 , Ag, CNTs 

• Performed sensitivity analysis 

– Most important inputs: aerosol concentration, 

breathing conditions (heavy, light exercise, rest), 

aspect ratio (for CNTs) 


www.epa.gov 

ICRP 

MPPD 

Light exercise 

breathing conditions 

Integrated Science Assessment for Particulate 

Matter, Dec. 2009, EPA NCEA, Jim Brown 

7

Mass per surface area (ug/cm^2) 

Mass per surface area (ug/cm^2) 

• Ag & TiO 2 nanoparticles 

50 

NM alveolar mass retained in 

human lungs 

0.16 

45 

0.14 

40 

0.12 

35 

30 

0.1 

25 

0.08 

20 

0.06 

15 

10 

5 

1 mg/m^3 

0.1 mg/m^3 

0.04 

0.02 

1 mg/m^3 

0.1 mg/m^3 

0 

0 50 100 

0 

0 20 40 60 80 100 

Diameter (nm) 

Diameter (nm) 

• Exposure duration: Full working lifetime 

of 45 years (8 h/day, 5 days/week) 

• Exposure duration: 24 hours 


www.epa.gov 

Alveolar mass retained for a full working lifetime to 1 

mg/m 3 Similar to high-end doses (~ 100-200 ug/mL) 

typical of in vitro testing 

8

NM physicochemical characterization 

• Colloboration with Center for Environmental Implications of NanoTechnology (CEINT) at 

Duke Univ. 

All NMs 

As received (dry powder or 

suspension) 

• In stock (prepared per OECD 

protocol: sonication in water with 

2% serum) 

Selected NM medium/cell combination 

In testing mediums 

In situ (cell/tissue) 

• Size distribution, shape 

(TEM DLS Cytovita ) 

• Surface area 

(BET Calculation from DLS ) 

• Chemical composition,crystal form 

(XRD Possibly ICP-MS) 

• Rate of dissolution 

(ICP-AES 

Possibly ion specific probe) 

• Surface composition/contamination 

(TOC 

Possibly SEM+EDS) 

• Surface charge, zeta potential 

(Zetasizer ) 


www.epa.gov 

• Possibly hydrophobicity, surface redox react. 

9

Developing screening models 

Animal Study 

In vitro assay 

Toxicity 

Pathway 

Perturbation 

Connections are made 

by looking for statistical 

associations across 

many chemicals 

Requires both in vitro 

and in vivo data 

Once a model is “qualified”: 

• New chemicals (nanomaterials) can be run through assays 

• Results of assays can be used to rank chemicals (nanomaterials) 

for potential to cause toxicity 


www.epa.gov 

10

NCCT databases and tools 

• Aggregated Computational Toxicology Resource (ACToR) 

• Database to find chemical toxicity info from large number of sources. 

• ToxMiner 

• Database to house detailed data ToxCast and ToxRefDB - used for 

ToxCast analyses. 

• ExpoCast DB 

• Detailed chemical concentration by media data from observational 

exposure studies. 

• Virtual tissue Knowledge base (VT-KB) 

• Tool developed to curate literature on chemical toxicity 


www.epa.gov 

11

ACToR: Aggregated Computational 

Toxicology Resource 

http://actor.epa.gov/ 

ACToR API 

Chemical 

Chemical ID, 

Structure 

ACToR Core 

ToxRefDB 

ToxMiner 

ExpoCastDB 

Internet 

Searches 

Tabular Data, 

Links to Web 

Resources 


www.epa.gov 

In Vivo Study 

Data - OPP 

ToxCast Data – 

NCCT, ORD, 

Collaborators 

(Currently Internal) 

Exposure Data – 

NERL, NCCT 

(In Development) 

12

ACToR goals and data sources 

• Compile all publicly available information on environmental chemicals 

Category 

Count 

Data Sets 580 

Chemicals 546,956 

Assays 3,213 

Assay Components 7,221 

Data Points 6,662,296 

• EPA (OPP, OPPT, NCEA, NERL) 

• FDA, NIH, CDC, OSHA, USDA 

• States and other countries 

• Universities 

• NGOs 

• Companies 

• Make data available for downloading, data mining 

– Available through data.gov 

– Entire DB can be downloaded and installed locally 

• Make it easy to see data gaps 

– Provides resource for EPA testing programs 

• Make it widely used 

– over 2000 regular users 


www.epa.gov 

Store NM physicochemical 

properties

Nanomaterial identity 

Nanomaterials require the same types of naming 

conventions 

• Common names 

• Systematic names 

• Computable representation of structure 

• Open source CASRN-like “code” for linking data from 

many sources 


www.epa.gov 

14

ToxMiner – ToxCast Data 

• Links 

– Chemicals 

– Assays 

– Genes 

– Pathways 

– Endpoints 

• Allows data analyses 

– Statistical associations (R-script) 

– Biologically driven data mining 


www.epa.gov 

Store in vitro HTS assay data 

on NMs

ToxCast assay data workflow 

Data on FTP site 

(from collaborators, 

contractors, etc.) 

Intranet wiki: 

log files, directory links, version 

stamps 

Data 

import 

“Raw” 

data 

Processing 

Processing 

scripts 

“Primary” data 

(standardized conc.- 

response format) 

Feedback & 

refinement 

Custom analysis 

(AC 50 fits, etc.) 

Local storage 

(working directory) 

EPA/NCCT 

servers 

“Fit” data (AC 50 /LEC) 

Network backup 

Network share 

(limited access) 

EPA 

servers 


www.epa.gov 

ToxMiner

ExpoCast DB 

• Data from NERL studies 

1) American Health Home 

Survey 

2) HUD Child Care Center 

Survey (“CCC” ) 

3) CTEPP – NC 

4) CTEPP – OH 

• Full raw data sets available to 

download 

• Browse data capability 

• By study name, chemical list, 

media list 

• Descriptive statistics 

capabilities 

generic_chemical 

ACToR 

CASRN 

Name 

Location_CV 

N 

Sources 

N 

1 

N 

1 Measure 

(Chemical) 

N 

Location 

N 

N 

1 

Study_CV 

N 

N 

N 

N 

N 

N 

N 

N 

N 

Sample 

N 

1 

Study 

N 

1 

N 

N 

N 

Laboratory 

method 

N 

1 

N 

N 

N 

Medium 

(wipes, urine, air, 

soil) 

Subject 

N 

N 

N 

Technique / 

sampling method 

N 

N 

N 

N 

Exposure 

Taxonomy 

(ACToR) 


www.epa.gov 

N 

1 

N 

Conceptual 

Store exposure aerosol concentrations 

of NM in occupational settings 

Data Model 

17

Pilot curation of 

exposure data into CTD 

Exposure 

(Time, 

Location) 

Chemicals 

Target 

(Individual) 

chemical-gene 

interactions 

chemical-disease 

relationships 

Genes 

gene-disease 

relationships 

Diseases 

functional annotations 


www.epa.gov 

pathway data 

18

Summary 

• Results of nanomaterial ToxCast screening and physicochemical 

characterization will be publicly accessible through ACToR 

• For chemicals, informatics infrastructure is in place for: 

– Capturing chemical identity 

– Capturing in vitro and in vivo data 

– Measuring and modeling biotransformation / metabolism 

– Building statistical and biologically-based models 

– Prioritizing chemicals for targeted testing 

– Dealing with 10 4 to 10 6 chemicals 

• Challenges for nanomaterials 

– Material identity 

– Quantification of imaging characterization results 


www.epa.gov 

19

Acknowledgments 

EPA National Center for Computational 

Toxicology (NCCT), ORD 

Elaine Cohen Hubal, Shad Mosher 

Amy Wang, Keith Houck 

Richard Judson, Ann Richard 

David Reif 

David Dix 

Robert Kavlock 

EPA ORD 

Peter Egeghy 

James Brown 

Stephanie Padilla 

Academia, Industry and others 

G. Oberdörster; Univ. of Rochester 

E. Rogers; Univ. of Massachusetts-Lowell 

M. Weisner, C. Matson, R. Badireddy, S. Marinakos, R. Giulio, M. Arnold; CIENT at Duke Univ. 

Lang Tran, Christoph Klein; ENPRA (Risk Assessment of Engineered Nanoparticles) 

O. Price and B. Asgharian; Applied Research Associates, Inc 


www.epa.gov 

This work was reviewed by EPA and approved for 

presentation but does not necessarily reflect Agency policy

Download (1152Kb)

Create successful ePaper yourself

Delete template?

Save as template?