Transformation products of organic contaminants â€“ Relevant risk

Transformation products of organic contaminants –Relevant risk factors?

Kathrin Fenner

Eawag, Swiss Federal Institute of Aquatic Science and Technology

Occurrence of transformation products

• Some herbicide transformation products detected in higher concentrations and more frequently than the parent pesticides in groundwater and surface water

Infotag 2009

Gilliom et al. (2005) USGS Circular 1291

Transformationproducts

5 herbicides(triazines, chloro-acetanilides)

Sinclair and Boxall (2003) Environ Sci Technol 37: 4617-4625

Parent compound ecotoxicity (EC50) (mM)

• 30% of transformation products more toxic than parent pesticide to fish, daphnids and algae

0.0001 0.01 1 100 10000

100000

1000

10

0.1

0.001

Tran

sfor

mat

ion

prod

uct e

coto

xici

ty(E

C50

) (m

M)

Transformation product10x more toxicTransformation product100x more toxic

Toxi

city

low

high lowToxicity

Transformation productequally toxic

How toxic are transformation products?

• Pesticide Directive (91/414/EEC):– Identification of relevant transformation products

mandatory– Clear guidance on assessing their relevance

• Industrial chemicals (REACh):– Identification of transformation products requested for

products > 100 t/y– No guidance on assessment

• Human and veterinary medicines (EMEA):– Consideration of environmental transformation

products subject to expert judgment

How are transformation products regulated?

• Research goals– Assess exposure to transformation products– Identify relevant transformation products

• Challenges– Which products?– Lack of analytical standards– Scarcity of experimental fate data

• Opportunities– Read-across from parent compound properties

⇒ Develop toolbox to adequately and efficiently assess transformation products

Open questions, challenges and opportunities

SBR

HR-MSMS

Fate models

Field studies

Readacross

Exposure assessmentPEC/MEC

Transformation products

Ris

k in

dica

tor

Risk assessment

Effect assessmentPNEC

Identification

Transformation product 2



Relevant transformation

product!

Identification

Strategy for assessing transformation products

• Artificial intelligence tools

• University of Minnesota Pathway Prediction System (UM-PPS)

– About 200 transformation rules– Derived from database of experimentally elucidated

biotransformation pathways (UM-BBD)– Publicly available, transparent, continuously developed– http://umbbd.msi.umn.edu/predict/index.html

Set of transformation rules

Structure-biodegradation relationships

University of Minnesota Pathway Prediction System (UM-PPS)

Structure-biodegradation relationships

Method to limit combinatorial explosion needed!

127 10 510

2nd step - # of predicted metabolites

Benzamide

bt0027bt0011

bt0012bt0013 bt0353

The challenge: Combinatorial explosionStructure-biodegradation relationships

“Known” products

• Use knowledge from experimentally elucidated pathways to learn about rule priorities

• Define rule priorities: If two rules are applicable, only apply the one more likely according to known pathways

• Find pairs of rules with clear rule priority over all known transformations

alcohol oxidation > aliphatic hydroxylation

⇒ 110 rule priorities

bt0063 > bt0036bt0022 > bt0036

“Unknown” products

>

Finding rule priorities through data miningStructure-biodegradation relationships

Formed in batch studywith activated sludge

Fenner et al., Bioinformatics, 2008

UM-PPS: Results from implementation of rule prioritiesStructure-biodegradation relationships

n Predictedreactions

OriginalPesticides 24 280With relative reasoningPesticides 24 240

• 15% reduction

-14.3%



n Predicted Known products Sensitivity Selectivityreactions predicted not

predicted(%) (%)

OriginalPesticides 24 280 43 15 74.1 15.4With relative reasoningPesticides 24 240 43 15 74.1 17.9

• 15% reduction• Sensitivity stable• Selectivity improved



Bioreactor considerations:Control pH, temperature, dissolved oxygenSludge (active biomass) concentrationSpiked compound concentrationSorption and abiotic controls

m/z

Inte

nsity

High resolution mass spectrometry

Sample preparation considerations:Sample volume – SPECompound sorption to syringeCompound sorption to filter materialSample storage

t= 0 6h 12h 1d 21d

…

Analytical method:Previously developed screening methodUM-PPS predicted TP massesIdentification with exact mass and MS/MSNo need for reference standards

Experimental setupExperimental elucidation of rule priorities

Unsubstituted or monosubstituted aromatics

2° or 3°aliphatic Aliphatic methyl

Ether AmineAmide

Aromatic methyl

Selection of pertinent rule prioritiesExperimental elucidation of rule priorities

OR3

NR2

R3O

R1

OHO

R1

NHR2

R3

NHR2

O

R1OH

R3

+

+

Hydrolysis products

Dealkylation productsOR

Case study: AmidesExperimental elucidation of rule priorities

• No transformations with priorities over amide fragments within UM-PPS

• Specific biodegradation pathway of amides remains ambiguous• Starting hypothesis: 1° and 2° amides hydrolyze rapidly, 3° amides

dealkylate

NH2

O

O

OH

NH

CH3

CH3

NHO

Cl

O

OOH

CH3 CH3N

OCH3

NHN

N

N

CH3

CH3

OOH

CH3

N

O

N

O

H

H CH3

N

Cl

N

OO

ONH

CH3

O

NH2

CH3

CH3

OCH3

Hydrolysis

Dealkylation

Other

Experimental elucidation of rule priorities

Helbling et al., in prep., 2009

NO

CH3

CH3

Cl

NO

CH3

CH3

Cl

NO CH3

CH3

N

O

CH3

CH3

CH3

NO CH3

Cl

CH3

NHO

CH3CH3

Hydrolysis

Dealkylation

Other

Experimental elucidation of rule priorities

Helbling et al., in prep., 2009

O

R1

NR2 R3

R1

[C,H], (A,a) [C,H], (A,a)

NHR2 R3

[C,H], (A,a) [C,H], (A,a)+

O R2R1(A,a) (A)

OHR1(A,a)

O R2

(A)

+

Outlook and further workStructure-biodegradation relationships

• New case study for amines and ethers

• UM-PPS mirror at ETH Zürich– More “environmental realism”– Validation and further development based on data from soil

and activated sludge simulation studies

• Feedback loop into UM-PPS

SBR

HR-MSMS

Fate models

Field studies

Readacross



Ris

k in

dica

tor

Risk assessment


Identification





product!


Mass balance model to predict concentration ratios of parent pesticide and transformation products

Gasser et al., ES&T, 2007; Kern et al., in prep., 2009

Measure for exposure to transformation product:Relative concentration in last box of river model

P TPC TP

EmissionRiver, 30 km

PCTP

Field study La Petite Glâne;Chemical analysis for 12 events and 6 pairs of parent compounds and transformation products

ToolsFinding products with relevant aquatic exposure

Application period 2nd period 3rd periodChloridazonFinding products with relevant aquatic exposure

0.01

0.10

1.00

10.00

100.00

1000.00

Azoxystrobin Atrazine Metolachlor Metamitron Sulcotrione Chloridazon

Rat

iotra

nsfo

rmat

ion

prod

uct :

par

ent p

estic

ide Application

2nd period

3rd period

Baseflow

n.d. n.d.

Atrazine Metolachlor

Summary of field study resultsFinding products with relevant aquatic exposure

AzoxystrobinAtrazine

Metolachlor

Sulcotrione

0.01

0.10

1.00

10.00

100.00

1000.00

0.01 0.10 1.00 10.00 100.00 1000.00

Rat

io T

P :

PC

Mea

sure

men

ts

Chloridazon

Metamitron

Ratio TP : PCModel predictions

Comparison model – measurementsFinding products with relevant aquatic exposure

SBR

HR-MSMS

Fate models

Field studies

Readacross



Ris

k in

dica

tor

Risk assessment


Identification





product!


Membrane-water partition coefficient(log Dmw, pH 7)

Toxi

city

(log

1/E

C50

)

Theoretical relation-ship for baseline toxicity

Effect and risk assessmentPredicting toxicity range of transformation products using read-

across

Measured, specific toxicity of parent compound

Transformationproduct 1

Transformationproduct 2

Parent compound

diuron

DCPMU

DCPU

DCA

MCPDMU

Effect and risk assessmentCase study diuron

2

3

4

5

6

7

8

2.3 2.5 2.7 2.9 3.1

log Dmw (pH 7)

log (

1/E

C50

(M))

Alg

en,

72 h DCPMU

diuron

MCPDMUDCPU DCA

0

20

40

60

80

100

1 2 3 4 5

Relative concentrations

Diuron DCPMU MCPDMU DCPU DCAConce

ntr

atio

ns/

risk

contr

ibutions

(%)

Relative risk contributions (baseline toxicity)Relative risk contributions (specific toxicity)

Risk contribution:EC50 (product)

Concentration (product)EC50 (diuron)

Concentration (diuron)

Effect and risk assessmentModeling effects, concentrations and risk contributions

• Prediction of biodegradation products– Restriction of prediction space through

data mining and targeted experiments

• Characterization of exposure to transformation products– Combination of modeling and monitoring – Importance of groundwater component!

• Procedure for risk assessment of transformation products– Prediction of relative concentrations– Prediction of toxicity range through read-

across– Targeted toxicity studies

Con

cent

ratio

n

Lipophilicity

Toxi

city DCPMU

diuron

MCPDMUDCPU DCA

Conclusions

Thank you!

KoMet team:• Susanne Kern• Judith Neuwöhner• Beate Escher, Juliane

Hollender, Heinz Singer

Swiss Federal Office for the Environment (Bafu):

• Michael Schärer• Paul Liechti, Christof Studer,

Reto Muralt, Christian Pillonel, Daniel Traber

Transformation products of organic contaminants â€“ Relevant risk

Documents