Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Environmental fate & pathways

Bioaccumulation: aquatic / sediment

Currently viewing:

Administrative data

Link to relevant study record(s)

Referenceopen allclose all

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
Not applicable
Reliability:
3 (not reliable)
Rationale for reliability incl. deficiencies:
results derived from a valid (Q)SAR model, but not (completely) falling into its applicability domain, with adequate and reliable documentation / justification
Remarks:
The chemical is out of the interpolation stuctural space.
Justification for type of information:
1. SOFTWARE
OASIS Catalogic v5.12.1

2. MODEL (incl. version number)
BCF base-line model v.02.09

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
CC1(C)CCCC2(C)C1CCC1(C)C2CCO1
Log Kow: 5.09 (measured, according to OECD 123)

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
See attached QMRF

5. APPLICABILITY DOMAIN
See attached QPRF

6. ADEQUACY OF THE RESULT
See attached QPRF
Qualifier:
according to guideline
Guideline:
other: REACH guidance on QSARs R.6, May 2008
Deviations:
no
Principles of method if other than guideline:
BCF calculated using BCF base-line model.
See attached QMRF and QPRF.
GLP compliance:
no
Radiolabelling:
no
Details on sampling:
Not applicable
Details on preparation of test solutions, spiked fish food or sediment:
Not applicable
Details on test organisms:
Not applicable
Test type:
other: QSAR
Water / sediment media type:
not specified
Hardness:
Not applicable
Test temperature:
Not applicable
pH:
Not applicable
Dissolved oxygen:
Not applicable
TOC:
Not applicable
Salinity:
Not applicable
Details on test conditions:
Not applicable
Nominal and measured concentrations:
Not applicable
Reference substance (positive control):
no
Details on estimation of bioconcentration:
Not applicable
Key result
Type:
BCF
Value:
4 265 L/kg
Basis:
whole body w.w.
Remarks on result:
other: BCF corrected with mitigating factors
Type:
BCF
Value:
6 026 L/kg
Basis:
whole body w.w.
Remarks on result:
other:
Remarks:
BCF max, not accounting for mitigating factors
Details on kinetic parameters:
Not applicable
Metabolites:
Not applicable
Results with reference substance (positive control):
Not applicable
Details on results:
See details in "Any other information on results incl. tables".
The predicted value from the BCFMAX model determines practically the maximum possible bioconcentration factor (BCFMAX) if the only factor determining the water/organism equilibrium is lipophilicity. This outcome overestimates the bioconcentration potential for most of the chemicals because different chemical and organism properties (called mitigating factors) could reduce significantly the accumulation of substances, such as metabolism, ionization, molecular size, etc. 
Therefore the predictions to be retained are the ones issued from the log BCF corrected at 3.63 ± 0.0866 L/kg wet, corresponding to a BCF corrected at 4265 ± 1.2 L/kg wet.
Reported statistics:
Not applicable

BCF Baseline-model:

log BCF corrected = 3.63 ± 0.0866 L/kg wet, corresponding to a BCF corrected at 4265 +/- 1.2 L/kg wet

 

Concomitant related predictions with BCF model:

log BCFmax = 3.780 L/kg wet, corresponding to a BCF max at 6026 L/kg wet (not accounting for mitigating factors).

 

relative mitigating effect of Acids= 0.000

relative mitigating effect of Metabolism= 0.000

relative mitigating effect of Phenols= 0.000

relative mitigating effect of Size= 0.997

relative mitigating effect of Water solubility= 0.001964

DiamMax Min value = 10.425 Å

DiamMax Max value = 11.430 Å

DiamMax Average = 11.111 Å

 

Validity criteria fulfilled:
yes
Conclusions:
The BCF for the tests substance, using the BCF base-line model v.02.09 available in OASIS Catalogic v5.12.1, is determined at 4 265 L/kg using the experimental log Kow value at 5.09, and using mitigating factors that account for the reduction of the bioaccumulation potential of the chemical substance. However, the outcome of the model for the test substance is not considered to be suitable for the assessment of the bioaccumulation potential. This is due to the fact that the chemical is out of the structural domain (ca. 12% of the atom centered fragments are unknown). No regulatory conclusion can be provided.
Executive summary:

The bioconcentration factor (BCF) of the substance was evaluated with BCF base-line model v.02.09 available in OASIS Catalogic v5.12.1, which fulfilled all OECD principles. The substance fulfils the general property requirements and can be considered within the parametric domain of the model, the substance is in the mechanistic domain of the model but is out of the interpolation structural space. The BCF for the test substance is determined at 4 265 L/kg using the experimental log Kow value at 5.09, and using mitigating factors that account for the reduction of the bioaccumulation potential of the chemical substance. However, the outcome of the model for the test substance is not considered to be suitable for the assessment of the bioaccumulation potential. This is due to the fact that the chemical is out of the structural domain (ca. 12% of the atom centered fragments are unknown). No regulatory conclusion can be provided.

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
Not applicable
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
results derived from a valid (Q)SAR model and falling into its applicability domain, with adequate and reliable documentation / justification
Remarks:
The value is not an experimental result, however the QSAR model is recommended by the ECHA guidance document on information requirements, is well documented with regard to validation parameters according to OECD principles. Moreover, the substance is fully characterised and within the applicability domain.
Justification for type of information:
1. SOFTWARE
EPISUITE v4.1

2. MODEL (incl. version number)
BCFBAF v3.01

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
CC1(C)CCCC2(C)C1CCC1(C)C2CCO1
log Kow = 5.09 (measured, according to OECD 123)

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
See attached QMRF
- Defined endpoint: Bioconcentration Factor (BCF) and Bioaccumulation Factor (BAF).
- Unambiguous algorithm: BCFBAF v3.01 contains two methods for estimating BCFs. In this endpoint study record, one method is presented.
This method has been expanded to calculate both BCF as well as BAF for three general trophic levels of fish. This method is based on the multiple-linear regression-derived equation which is used by the BCFBAF program to estimate the kM Biotransformation Half-Life.
- Defined domain of applicability: The estimation domain for BCF is based on the number of instances given for each correction factor in any of the training set compounds, and the minimum and maximum values for molecular weight and log Kow.
- Appropriate measures of goodness-of-fit and robustness and predictivity: Arnot-Gobas BCF and BAF statistics for the kM biotransformation half-life of the training dataset are; Number = 421 / r² = 0.821 / std deviation = 0.494 / avg deviation = 0.383. For the kM validation dataset; Number = 211 / r² = 0.734 / std deviation = 0.602 / avg deviation = 0.446.
- Mechanistic interpretation: no data

5. APPLICABILITY DOMAIN
See attached QPRF

6. ADEQUACY OF THE RESULT
See attached QPRF
Qualifier:
according to guideline
Guideline:
other: REACH guidance on QSARs R.6, May 2008
Deviations:
no
Principles of method if other than guideline:
BCF calculated using Arnot-Gobas method.
See attached QMRF and QPRF (the QMRF is only available for the former version of the QSAR model).
GLP compliance:
no
Specific details on test material used for the study:
No additional information
Radiolabelling:
no
Details on sampling:
Not applicable
Details on preparation of test solutions, spiked fish food or sediment:
Not applicable
Details on test organisms:
Not applicable
Test type:
other: QSAR
Water / sediment media type:
not specified
Hardness:
Not applicable
Test temperature:
Not applicable
pH:
Not applicable
Dissolved oxygen:
Not applicable
TOC:
Not applicable
Salinity:
Not applicable
Details on test conditions:
Not applicable
Nominal and measured concentrations:
Not applicable
Reference substance (positive control):
no
Details on estimation of bioconcentration:
Not applicable
Key result
Type:
BCF
Value:
>= 2 152 - <= 2 563 L/kg
Basis:
whole body w.w.
Remarks on result:
other: Arnot-Gobas Method, including a biotransformation rate estimates. Depending on the trophic level (upper to lower trophic).
Details on kinetic parameters:
Not applicable
Metabolites:
Not applicable
Results with reference substance (positive control):
Not applicable
Details on results:
Arnot-Gobas Method: It's more relevant to consider the result including biotransformation rate estimates instead of the result assuming a biotransformation rate of zero.
Reported statistics:
Not applicable

TYPE

NUM

LOG BIOTRANSFORMATION FRAGMENT DESCRIPTION

COEFF

VALUE

Frag

Frag

Frag

Frag

Frag

Frag

L Kow

MolWt

Const

3

1

4

7

2

1

*

*

*

Carbon with 4 single bonds & no hydrogens

Aliphatic ether [C-O-C]

Methyl [-CH3]

-CH2- [cyclic]

-CH- [cyclic]

Number of fused acyclic rings

Log Kow = 5.09 (user-entered)

Molecular Weight Parameter

Equation Constant

-0.2984

-0.0232

0.2451

0.0963

0.0126

0.6477

0.3073

-0.8953

-0.0232

0.9804

0.6738

0.0252

0.6477

1.5644

-0.6062

-1.5371

RESULT

RESULT

NOTE

Log Bio Half-life (days)

Bio Half-life (days)

Bio Half-life Normalized to 10 g fish at 15°C

0.8296

6.755

Biotransformation rate constant: kM (rate constant) = 0.1026 / day (10 gram fish).

Arnot-Gobas BCFBAF Methods (including biotransformation rate estimates):

Estimated Log BCF (upper trophic) = 3.333 (BCF = 2152 L/Kg wet-wt).

Estimated Log BCF (mid trophic) = 3.399 (BCF = 2505 L/Kg wet-wt).

Estimated Log BCF (lower trophic) = 3.409 (BCF = 2563 L/Kg wet-wt).

Arnot-Gobas BCFBAF Methods (assuming a biotransformation rate of zero):

Estimated Log BCF (upper trophic) = 3.960 (BCF = 9131 L/Kg wet-wt).

Validity criteria fulfilled:
yes
Conclusions:
The BCF for the substance using the Arnot-Gobas method is comprised between 2 152 and 2 563 L/Kg wet-wt, using the experimental log Kow value at 5.09, and including biotransformation rate estimate.
Executive summary:

The bioconcentration factor (BCF) of the substance was evaluated with BCFBAF model v3.01 from EPI Suite v4.1, which fulfilled all OECD principles. The substance is fully characterised towards the applicability domain. The BCF for the substance using the Arnot-Gobas method is comprised between 2 152 and 2 563 L/Kg wet-wt, using the experimental log Kow value at 5.09, and including biotransformation rate estimate.

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
Not applicable
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
results derived from a valid (Q)SAR model and falling into its applicability domain, with adequate and reliable documentation / justification
Remarks:
The value is not an experimental result, however the QSAR model is recommended by the ECHA guidance document on information requirements, is well documented with regard to validation parameters according to OECD principles. Moreover, the substance is fully characterised and within the applicability domain.
Justification for type of information:
1. SOFTWARE
EPISUITE v4.1

2. MODEL (incl. version number)
BCFBAF v3.01

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
CC1(C)CCCC2(C)C1CCC1(C)C2CCO1
log Kow = 5.09 (measured, according to OECD 123)

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
See attached QMRF
- Defined endpoint: Bioconcentration Factor (BCF) and Bioaccumulation Factor (BAF).
- Unambiguous algorithm: BCFBAF v3.01 contains two methods for estimating BCFs. In this endpoint study record, one method is presented. This method is based on the original BCFWIN model and classifies a compound as either ionic or non-ionic. Ionic compounds include carboxylic acids, sulfonic acids and salts of sulfonic acids, and charged nitrogen compounds (nitrogen with a +5 valence such as quaternary ammonium compounds). All other compounds are classified as non-ionic.
- Defined domain of applicability: The estimation domain for BCF is based on the minimum and maximum values for molecular weight and log Kow, and the correction factors if needed.
- Appropriate measures of goodness-of-fit and robustness and predictivity: The following gives statistical information for the BCF training and validation datasets (includes ionic and non-ionic compounds). Number = 527 / r² = 0.833 / std deviation = 0.502 / avg deviation = 0.382. For the BCF validation dataset; Number = 158 / r² = 0.82 / std deviation = 0.59 / avg deviation = 0.46.
- Mechanistic interpretation: no data

5. APPLICABILITY DOMAIN
See attached QPRF

6. ADEQUACY OF THE RESULT
See attached QPRF
Qualifier:
according to guideline
Guideline:
other: REACH guidance on QSARs R.6, May 2008
Deviations:
no
Principles of method if other than guideline:
BCF calculated using regression-based estimate method.
See attached QMRF and QPRF (the QMRF is only available for the former version of the QSAR model).
GLP compliance:
no
Specific details on test material used for the study:
No additional information
Radiolabelling:
no
Details on sampling:
Not applicable
Details on preparation of test solutions, spiked fish food or sediment:
Not applicable
Details on test organisms:
Not applicable
Test type:
other: QSAR
Water / sediment media type:
not specified
Hardness:
Not applicable
Test temperature:
Not applicable
pH:
Not applicable
Dissolved oxygen:
Not applicable
TOC:
Not applicable
Salinity:
Not applicable
Details on test conditions:
Not applicable
Nominal and measured concentrations:
Not applicable
Reference substance (positive control):
no
Details on estimation of bioconcentration:
Not applicable
Key result
Type:
BCF
Value:
1 060 L/kg
Basis:
whole body w.w.
Remarks on result:
other: BCF result using the experimental log Kow value
Details on kinetic parameters:
Not applicable
Metabolites:
Not applicable
Results with reference substance (positive control):
Not applicable
Details on results:
See "Any other information on results incl. tables".
Reported statistics:
Not applicable

Log Kow (estimated) : 4.76

Log Kow (experimental): not available from database

Log Kow used by BCF estimates: 5.09 (user entered)

Equation Used to Make BCF estimate:

Log BCF = 0.6598 log Kow - 0.333 + Correction

Correction(s) value: no applicable correction factors

Estimated Log BCF = 3.025 (BCF = 1 060 L/kg wet-wt)

Validity criteria fulfilled:
yes
Conclusions:
The BCF for the substance using the regression-based estimate method is determined at 1 060 L/Kg wet-wt based on the experimental log Kow value, at 5.09.
Executive summary:

The bioconcentration factor (BCF) of the substance was evaluated with BCFBAF model v3.01 from EPI Suite v4.1, which fulfilled all OECD principles. The substance is fully characterised towards the applicability domain. The BCF for the substance using the regression-based estimate method is determined at 1 060 L/Kg wet-wt based on the experimental log Kow value, at 5.09.

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
Not applicable
Reliability:
4 (not assignable)
Rationale for reliability incl. deficiencies:
documentation insufficient for assessment
Remarks:
BCF value is determined using calculation method from the Technical Guidance Document in a weight of evidence approach.
Justification for type of information:
1. SOFTWARE
None. Technical Guidance Document (2003)

2. MODEL (incl. version number)
None.

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
log Kow = 5.09 (measured, according to OECD 123)

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
- Defined endpoint: Bioconcentration Factor (BCF).
- Unambiguous algorithm: For a substance with a log Kow of 2-6, the following linear relationship can be used as developed by Veith et al. (1979): log BCF (fish) = 0.85*log Kow - 0.70. For a substance with a log Kow higher than 6, a parabolic equation can be used: log BCF (fish) = -0.20*log Kow² + 2.74*log Kow - 4.72.
- Defined domain of applicability: The model applicability domain consists of 2 parts: organic chemicals with log Kow of 2-6 and organic chemicals with log Kow > 6.
- Appropriate measures of goodness-of-fit and robustness and predictivity: n (number of data) = 55; R² (correlation coefficient) = 0.90
- Mechanistic interpretation: No data

5. APPLICABILITY DOMAIN
See details in the Endpoint Study Record

6. ADEQUACY OF THE RESULT
See details in the Endpoint Study Record
Qualifier:
no guideline required
Principles of method if other than guideline:
Calculations based on structure activity.
GLP compliance:
no
Remarks:
(not applicable)
Specific details on test material used for the study:
No additional information
Radiolabelling:
no
Details on sampling:
not applicable
Details on preparation of test solutions, spiked fish food or sediment:
not applicable
Test organisms (species):
no data
Details on test organisms:
No data
Route of exposure:
other: QSAR
Test type:
other: QSAR
Water / sediment media type:
not specified
Hardness:
not applicable
Test temperature:
not applicable
pH:
not applicable
Dissolved oxygen:
not applicable
TOC:
not applicable
Salinity:
not applicable
Details on test conditions:
not applicable
Nominal and measured concentrations:
not applicable
Reference substance (positive control):
not required
Remarks:
QSAR
Details on estimation of bioconcentration:
not applicable
Type:
other: log BCF
Value:
3.627 dimensionless
Remarks on result:
other: TGD, part. II, eq. (74) (logKow = 5.09)
Key result
Type:
BCF
Value:
4 232 L/kg
Basis:
whole body w.w.
Remarks on result:
other: TGD, part. II, eq. (74) (logKow = 5.09)
Details on kinetic parameters:
no data
Metabolites:
no data
Results with reference substance (positive control):
no data
Details on results:
no data
Reported statistics:
no data

Equation used:

log BCF (fish) = 0.85*log Kow - 0.70

Validity criteria fulfilled:
yes
Conclusions:
The BCF value of the substance using the TGD part. II eq. (74) is 4 232 L/kg.
Executive summary:

The BCF value of the substance has been assessed by using the TGD part. II eq. (74). The result is shown below:

- TGD part. II eq. (74) approach, log BCF = 3.6265, BCF (L/kg) = 4 232

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
Not applicable
Reliability:
3 (not reliable)
Rationale for reliability incl. deficiencies:
results derived from a valid (Q)SAR model, but not (completely) falling into its applicability domain, with adequate and reliable documentation / justification
Remarks:
The predicted compound could be out of the Applicability Domain of the model
Justification for type of information:
1. SOFTWARE
VEGA QSAR MODELS v1.2.1

2. MODELS (incl. version number)
BCF model (CAESAR) 2.1.14

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
O1CCC2C1(C)CCC3C(C)(C)CCCC23(C)

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
See attached QMRF

You can find complete details on the model and on how to read results in the proper model's guide, available on-line at www.vega-qsar.eu or directly in the VegaNIC application:
QSAR model for fish BCF, based on a Radial Basis Function neural network. The model extends the original CAESAR BCF model 1.0, full reference to the model: C. Zhao, E. Boriani, A. Chana, A.Roncaglioni, E. Benfenati "A new hybrid system of QSAR models for predicting bioconcentration factors (BCF)", Chemosphere 73 (2008) 1701–1707. The original model was developed inside the CAESAR Project (http://www.caesar-project.eu/).

5. APPLICABILITY DOMAIN
See attached VEGA predictions report

6. ADEQUACY OF THE RESULT
See attached VEGA predictions report
Qualifier:
according to guideline
Guideline:
other: REACH guidance on QSARs R.6, May 2008
Deviations:
no
Principles of method if other than guideline:
The model is recognised as scientifically valid (using the OECD principles)
See attached QMRF.
GLP compliance:
no
Specific details on test material used for the study:
Measured log Kow value was not taken into account as input of this in silico prediction.
Radiolabelling:
no
Details on sampling:
Not applicable
Details on preparation of test solutions, spiked fish food or sediment:
Not applicable
Details on test organisms:
Not applicable
Test type:
other: QSAR
Water / sediment media type:
not specified
Hardness:
Not applicable
Test temperature:
Not applicable
pH:
Not applicable
Dissolved oxygen:
Not applicable
TOC:
Not applicable
Salinity:
Not applicable
Details on test conditions:
Not applicable
Nominal and measured concentrations:
Not applicable
Reference substance (positive control):
no
Details on estimation of bioconcentration:
Not applicable
Key result
Type:
BCF
Value:
153 L/kg
Basis:
not specified
Remarks on result:
other: The predicted compound could be out of the applicability domain of the model
Details on kinetic parameters:
Not applicable
Metabolites:
Not applicable
Results with reference substance (positive control):
Not applicable
Details on results:
Prediction is 2.18 log(L/kg), but the result shows some critical aspects, which require to be checked:
- only moderately similar compounds with known experimental value in the training set have been found
- accuracy of prediction for similar molecules found in the training set is not optimal
- some similar molecules found in the training set have experimental values that disagree with the predicted value
- the maximum error in prediction of similar molecules found in the training set has a moderate value, considering the experimental variability
Reported statistics:
Not applicable

Compound SMILES: O1CCC2C1(C)CCC3C(C)(C)CCCC23(C)

Experimental value [log(L/kg)]: -

Predicted BCF [log(L/kg)]: 2.18

Predicted BCF [L/kg]: 153

Predicted BCF from sub-model 1 (HM) [log(L/kg)]: 2.15

Predicted BCF from sub-model 2 (GA) [log(L/kg)]: 2.47

Predicted LogP (MLogP): 4.12

Structural alerts: -

Reliability: the predicted compound could be out of the Applicability Domain of the model

Remarks: none

Validity criteria fulfilled:
not specified
Conclusions:
The BCF for the test substance, using the BCF model CAESAR v2.1.14, is determined at 153 L/kg. The result shows some critical aspects which required to be checked. No regulatory conclusion can be provided.
Executive summary:

The bioconcentration factor (BCF) of the substance was evaluated with BCF model CAESAR v2.1.14, available using VEGA QSAR models platform, which fulfilled all OECD principles. BCF prediction is 153 L/kg but the result shows some critical aspects, which require to be checked:

- only moderately similar compounds with known experimental value in the training set have been found

- accuracy of prediction for similar molecules found in the training set is not optimal

- some similar molecules found in the training set have experimental values that disagree with the predicted value

- the maximum error in prediction of similar molecules found in the training set has a moderate value, considering the experimental variability.

Based on this prediction, no regulatory conclusion can be provided.

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
Not applicable
Reliability:
3 (not reliable)
Rationale for reliability incl. deficiencies:
results derived from a valid (Q)SAR model, but not (completely) falling into its applicability domain, and documentation / justification is limited
Remarks:
the predicted compound is outside the Applicability Domain of the model
Justification for type of information:
1. SOFTWARE
VEGA QSAR MODELS v1.2.1

2. MODELS (incl. version number)
BCF model (KNN/Read-Across) 1.1.0

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
O1CCC2C1(C)CCC3C(C)(C)CCCC23(C)

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
You can find complete details on the model and on how to read results in the proper model's guide, available on-line at www.vega-qsar.eu or directly in the VegaNIC application: KNN (Read-Across) model for fish BCF.

5. APPLICABILITY DOMAIN
See attached VEGA predictions report

6. ADEQUACY OF THE RESULT
See attached VEGA predictions report
Qualifier:
according to guideline
Guideline:
other: REACH guidance on QSARs R.6, May 2008
Deviations:
no
GLP compliance:
no
Specific details on test material used for the study:
Measured log Kow value was not taken into account as input of this in silico prediction.
Radiolabelling:
no
Details on sampling:
Not applicable
Details on preparation of test solutions, spiked fish food or sediment:
Not applicable
Details on test organisms:
Not applicable
Test type:
other: QSAR
Water / sediment media type:
not specified
Hardness:
Not applicable
Test temperature:
Not applicable
pH:
Not applicable
Dissolved oxygen:
Not applicable
TOC:
Not applicable
Salinity:
Not applicable
Details on test conditions:
Not applicable
Nominal and measured concentrations:
Not applicable
Reference substance (positive control):
no
Details on estimation of bioconcentration:
Not applicable
Key result
Type:
BCF
Value:
66 L/kg
Basis:
not specified
Remarks on result:
other: The predicted compound is outside the Applicability Domain of the model
Details on kinetic parameters:
Not applicable
Metabolites:
Not applicable
Results with reference substance (positive control):
Not applicable
Details on results:
Prediction is 1.82 log(L/kg), but the result may be not reliable. A check of the information given in the following section should be done, paying particular attention to the following issues:
- accuracy of prediction for similar molecules found in the training set is not optimal
- some similar molecules found in the training set have experimental values that disagree with the predicted value
- the maximum error in prediction of similar molecules found in the training set has a high value, considering the experimental variability
Reported statistics:
Not applicable

Compound SMILES: O1CCC2C1(C)CCC3C(C)(C)CCCC23(C)

Experimental value [log(L/kg)]: -

Predicted BCF [log(L/kg)]: 1.82

Molecules used for prediction: 4

Reliability: the predicted compound is outside the Applicability Domain of the model

Remarks: none

Validity criteria fulfilled:
not specified
Conclusions:
The BCF for the test substance, using the BCF model (KNN/Read-Across) 1.1.0, is determined at 66 L/kg, but the result may be not reliable. Therefore, no regulatory conclusion can be provided.
Executive summary:

The bioconcentration factor (BCF) of the substance was evaluated with BCF model (KNN/Read-Across) 1.1.0, available using VEGA QSAR models platform, which fulfilled all OECD principles. BCF prediction is 66 L/kg but the result may be not reliable. A check of the information given in the following section should be done, paying particular attention to the following issues:

- accuracy of prediction for similar molecules found in the training set is not optimal

- some similar molecules found in the training set have experimental values that disagree with the predicted value

- the maximum error in prediction of similar molecules found in the training set has a high value, considering the experimental variability

Based on this prediction, no regulatory conclusion can be provided.

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
Not applicable
Reliability:
3 (not reliable)
Rationale for reliability incl. deficiencies:
results derived from a valid (Q)SAR model, but not (completely) falling into its applicability domain, and documentation / justification is limited
Remarks:
the predicted compound is outside the Applicability Domain of the model
Justification for type of information:
1. SOFTWARE
VEGA QSAR MODELS v1.2.1

2. MODELS (incl. version number)
BCF model (Meylan) 1.0.3

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
O1CCC2C1(C)CCC3C(C)(C)CCCC23(C)

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
You can find complete details on the model and on how to read results in the proper model's guide, available on-line at www.vega-qsar.eu or directly in the VegaNIC application:
QSAR model for fish BCF, based on Meylan approach, as implemented in EPI Suite. Full reference to this model can be found in the EPI Suite help (http://www.epa.gov/oppt/exposure/pubs/episuite.htm) and in the original paper from Meylan: Meylan W.M., Howard P.H., Boethling R.S. et al. Improved Method for Estimating Bioconcentration / Bioaccumulation Factor from Octanol/Water Partition Coefficient. 1999, Environ. Toxicol. Chem. 18(4): 664-672. Model developed inside the VEGA platform.

5. APPLICABILITY DOMAIN
See attached VEGA predictions report

6. ADEQUACY OF THE RESULT
See attached VEGA predictions report
Qualifier:
according to guideline
Guideline:
other: REACH guidance on QSARs R.6, May 2008
Deviations:
no
GLP compliance:
no
Specific details on test material used for the study:
Measured log Kow value was not taken into account as input of this in silico prediction.
Radiolabelling:
no
Details on sampling:
Not applicable
Details on preparation of test solutions, spiked fish food or sediment:
Not applicable
Details on test organisms:
Not applicable
Test type:
other: QSAR
Water / sediment media type:
not specified
Hardness:
Not applicable
Test temperature:
Not applicable
pH:
Not applicable
Dissolved oxygen:
Not applicable
TOC:
Not applicable
Salinity:
Not applicable
Details on test conditions:
Not applicable
Nominal and measured concentrations:
Not applicable
Reference substance (positive control):
no
Details on estimation of bioconcentration:
Not applicable
Key result
Type:
BCF
Value:
1 806 L/kg
Basis:
not specified
Remarks on result:
other: The predicted compound is outside the Applicability Domain of the model
Details on kinetic parameters:
Not applicable
Metabolites:
Not applicable
Results with reference substance (positive control):
Not applicable
Details on results:
Prediction is 3.26 log(L/kg), but the result may be not reliable. A check of the information given in the following section should be done, paying particular attention to the following issues:
- only moderately similar compounds with known experimental value in the training set have been found
- similar molecules found in the training set have experimental values that disagree with the predicted value
- the maximum error in prediction of similar molecules found in the training set has a moderate value, considering the experimental variability
- reliability of logP value used by the model is not adequate
Reported statistics:
Not applicable

Compound SMILES: O1CCC2C1(C)CCC3C(C)(C)CCCC23(C)

Experimental value [log(L/kg)]: -

Predicted BCF [log(L/kg)]: 3.26

Predicted BCF [L/kg]: 1806

Predicted LogP (Meylan/Kowwin): 5.44

Predicted LogP reliability: Low

MW: 235.06

Ionic compound: no

Reliability: the predicted compound is outside the Applicability Domain of the model

Remarks: none

Validity criteria fulfilled:
not specified
Conclusions:
The BCF for the test substance, using the BCF model Meylan 1.0.3, is determined at 1 806 L/kg, but the result may be not reliable. Therefore, no regulatory conclusion can be provided.
Executive summary:

The bioconcentration factor (BCF) of the substance was evaluated with BCF model Meylan v1.0.3, available using VEGA QSAR models platform, which fulfilled all OECD principles. BCF prediction is 1 806 L/kg but the result may be not reliable. A check of the information given in the following section should be done, paying particular attention to the following issues:

- only moderately similar compounds with known experimental value in the training set have been found

- similar molecules found in the training set have experimental values that disagree with the predicted value

- the maximum error in prediction of similar molecules found in the training set has a moderate value, considering the experimentalvariability

- reliability of logP value used by the model is not adequate

Based on this prediction, no regulatory conclusion can be provided.

Endpoint:
bioaccumulation in aquatic species: invertebrate
Type of information:
experimental study
Adequacy of study:
weight of evidence
Study period:
From 16 December 2016 to 7 Mars 2017
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Remarks:
Non-GLP study
Qualifier:
no guideline available
Principles of method if other than guideline:
An aqueous exposure study (one test concentration) with non-radiolabeled test substance on Hyalella azteca was carried out under flow-through conditions at 25 +/- 2°C. The objective of the study was to determine the steady-state and kinetic bioconcentration factors (BCFss and BCFk) of the test substance in Hyalella azteca.
The study included the following steps:
- One experimental tank was stocked with around 1600 adult, male H. azteca.
- The amphipods were exposed to a constant concentration (±20% of the mean) of the test item in a flow-through system (uptake phase). This was ensured by measuring the concentration of the test substance in water (Cw). The steady-state bioconcentration factor (BCFss) was calculated as the quotient of concentration of the test substance in Hyalella Ch and the time-weighted average (TWA) Cw, when no further increase of tissue concentrations was observed. A steady-state is reached, when analyses of Ch made on samples taken at three successive sampling points are within ± 20% of each other.
- Following the uptake period, the remaining amphipods were transferred to uncontaminated water and further samples collected for analysis of tissue concentrations of the test item (depuration phase). The decrease of the test substance by time was calculated by fitting an exponential function (Ch(t) = Ch(t0)* e(-k2*t)) to the concentrations measured in Hyalella, resulting in the depuration rate constant (k2).
- The data of Ch/ TWA Cw and k2 were used to calculate the uptake rate constant (k1) by nonlinear regression.
- The kinetic bioconcentration factor BCFk was calculated by using the relation BCFk = k1/k2.
- Evaluation of depuration-time to t1/2 and t95%.
GLP compliance:
no
Remarks:
Non-GLP study performed in Weight of Evidence approach for the determination of the bioaccumulation of the registered substance. The use of non-vertebrate species Hyalella azteca is an alternative method to replace the use of fish for BCF testing.
Specific details on test material used for the study:
No additional information
Radiolabelling:
no
Details on sampling:
- Sampling schedule: see Table 5.3.1/1 in "Any other information on materials and methods, incl. tables".

ANALYSIS OF WATER SAMPLES:
Two water samples of 40 mL were taken from the test basin at each sampling time. Sample preparation: 7 mL of Ethylacetate was added to the sample; 50 µL of internal standard spike solution (11.38 mg/L) was added; the samples were then shaken and centrifuged; 1 mL of the organic phase was transferred into a GC sample vial and directly measured via GC-MSD.

ANALYSIS OF HYALELLA SAMPLES:
The samples were extracted as follows:
- A porcelain mortar and a wire were cooled in liquid Nitrogen
- The sample vial with the Hyalella (~49-61 mg sample weight) was dipped into liquid nitrogen
- The mortar was taken out of the liquid Nitrogen container (with some liquid N2 inside)
- The Hyalellas were transferred into the mortar with the wire and ground with the pestle
- Mortar and pestle were put onto a heating plate (at ca. 75°C)
- 2 mL of Dichloromethane/Acetone (1:1 V/V) was added
- When the sample was completely thawed (2-3 min) it was stirred and the pestle washed with 1 mL of Dichloromethane/Acetone (1:1 V/V)
- 50 μL of internal standard spike solution (400 μg/L)was added
- The sample was transferred completely to a 8 mL vial while continuously stirring the sample.
- The mortar was washed with four times one milliliter of Dichloromethane/Acetone, transferring the washing solution into the 8 mL vial as well.
- A small amount of Na2SO4 (2 spatula tips worth) were added and the solution thoroughly mixed on a Vortex.
- The samples were then centrifuged for 5 min at ca. 4000 rpm.
- Samples were frozen for further preparation and analysis.
The cleanup of the Hyalella samples was performed via SPE as follows:
- The complete extract was transferred to a Zymark vessel
- The fluid was concentrated to 0.5 mL at 40 °C
- The SPE cartridges were preconditioned with two times 4 mL of Dichloromethane/Acetone and two times 4 mL of Hexane
- The catridges were dried by applying a vacuum
- The sample was pipetted into the cartridge and the Zymark vessel washed with 250 μL of Hexane and the fluid added to the sample
- When the sample was absorbed into the SPE material it was eluted with two times 4 mL of Dichloromethane
- By applying a vacuum all fluid was transferred to the receiving Zymark vessel in the vacuum chamber
- In the Zymark vessel the solution was concentrated to 0.5 mL again
- The 0.5 mL sample was then transferred to a GC vial and evaporated to dryness
- 250 μL of Toluene were pipetted into the vial
- The samples were measured by GC-MSD.
Vehicle:
yes
Details on preparation of test solutions, spiked fish food or sediment:
PREPARATION OF TEST MEDIA AND NOMINAL TEST CONCENTRATION
A stock solution of the test substance in ethanol with a concentration of approximately 6.55 mg/ml was prepared. For a basic solution, 1000 μL of the stock solution were filled into separate 10L-brown glass bottles with screw caps. The main amount of ethanol was reduced by flushing with nitrogen and the bottles were filled with Cu-free water up to total volume 10 L. The solution was stirred overnight (approximately 14 h) using a magnetic stirrer. New basic solutions were prepared on a daily basis. The stirred basic solution served as application reservoir. A glass capillary outlet to a membrane pump was integrated in screw top of the reservoir. The diluted aqueous solution from the reservoir was pumped with a flow rate of approximately 4.5 mL/min into a mixing chamber with magnetic stirring. Through a second inlet of the mixing chamber, fresh copper-free water was added to reach a total flow rate of 109 mL/min. The test media flew continuously into the test basins (25 L volume filled with approximately 20 L of test media). For the daily preparation of the application reservoir (10 L fresh aqueous basic solutions) clean bottles were used. The nominal concentration of the test media was thus 27 μg/L in order to achieve the intended test concentration of 20-30 μg/L. The test concentration was selected based on the acute toxicity data of the test item analogue for invertebrates given in the ECHA dossier. With an EC50 (48h) value of 1.8 mg/L, the selected test item concentration is by a factor of around 60-90 below the acute toxicity concentration of the analogue. The suitability of the test concentration was further confirmed by a 48h-acute toxicity pre-test with the test item carried out on Hyalella azteca. Nominal concentrations of 3, 10, 30, and 100 μg/L were tested against a control group. Each test group and the control consisted of 4 replicates with each 5 animals. No mortalities were observed during the test.
Test organisms (species):
Hyalella azteca
Details on test organisms:
- Justification of the use: Hyalella azteca is commonly used in North America as test species for ecotoxicity studies. Only healthy amphipods free from observable diseases and abnormalities were used in this study.
- Specification: Hyalella azteca
- Size: adult amphipods, age > 2 months
- Sex: male
- Source: laboratory bred. The amphipods were originally obtained from Freds Haustierzoo, Berliner Strasse 37, 51063 Köln
- Loading: 1600 amphipods per tank (25 L)
- Food: Daily feeding ad libitum with algae aggregates (Desmodesmus subspicatus). Food residues (algae) were completely removed from the experimental tank after feeding.
Route of exposure:
aqueous
Justification for method:
aqueous exposure method used for following reason: determine the steady-state and kinetic bioconcentration factors (BCFss and BCFk) of the test substance in Hyalella azteca.
Test type:
flow-through
Water / sediment media type:
other: Purified drinking water was used according to the OECD Guideline 305
Total exposure / uptake duration:
144 h
Total depuration duration:
144 h
Hardness:
Total hardness = 6.5 mmol/L
Test temperature:
Mean temperature was 24.9 ± 0.2°C
pH:
The pH average of the test medium was 7.82 ± 0.06
Dissolved oxygen:
oxygen saturation was between 96% and 107% throughout the study.
TOC:
No data
Salinity:
Not applicable
Conductivity:
No data
Details on test conditions:
TEST PERFORMANCE
The exposure period started on December 16, 2016. A group of 1600 amphipods was exposed in a flow-through system at a water temperature of 25 +/- 2°C. The exposure (uptake) phase lasted for 144 h (6 days) and was followed by a depuration period of 144 h (6 days). A 25 L glass aquarium (no lid) filled with 20 L of test solution was used as test basin. A continuous flow of test solution (uptake phase) or water (depuration phase) equivalent to 7.8 volume replacements per day was maintained throughout the test using a metering pump system at a rate of 6.5 L/h. The water in the test vessel was aerated via a glass capillary. Amphipods were fed daily ad libitum with algae aggregates (Desmodesmus subspicatus) using the filter disk method. Disks and attached feed residues were completely removed from the experimental tank after feeding.
Observations were made throughout the test period on the behavior of the amphipods. Temperature, pH and the oxygen concentrations in the test vessel were measured daily. The Ammonium, Nitrate and Nitrite parameters were measured at the beginning and the end of uptake and depuration phase. The light/dark cycle was adjusted to 16/8 hours. During the uptake phase, the amphipods were continuously exposed to nominal concentration of 27 μg/L of the test substance. Thereafter the amphipods were transferred into a new aquarium served with clean dilution water for the depuration phase. The concentration of the test substance in the amphipods and the water was determined throughout the uptake and elimination period. Hyalella and water samples were collected according to the schedule (Table 5.3.1/1). For the uptake and depuration phase 7 and 5 Hyalella sampling dates were scheduled, respectively. Samples of 3 times 20 amphipods were removed from the test vessel, rinsed in dilution water, blotted dry, weighed and immediately frozen at -20°C. The pooled Hyalella samples (n=3) were analyzed for the test item by GC/MS. Directly before the start of exposure and at each Hyalella sampling date during the uptake phase, an adequate amount of test solution was collected from the test vessel and analyzed for the test item. An additional water sample was taken from the test vessel at the start of the depuration period (after complete change of the test system). As the test item could not be detected anymore, water measurements were not required for the remainder of the depuration phase.

TEST MEDIUM / WATER PARAMETERS
Purified drinking water was used according to the OECD Guideline 305. The purification included filtration with charcoal, aeration and passage through a lime stone column; pH, NPOC, hardness and further characteristics of a representative measurement are reported.
Nominal and measured concentrations:
- Nominal concentration of 27 μg/L
- Measured concentration (TWA) was 25.3 ± 0.56 μg/L.

The test substance concentration in water (Cw) during the uptake phase was evaluated as time weighted average (TWA) water concentration. TWA Cw values were computed by weighing the mean test item concentration in water of two sampling time points with the time of exposure between the respective sampling time points (in days). From these weighted concentrations an overall mean and standard deviation was calculated to derive the TWA Cw ±SD.
Reference substance (positive control):
no
Details on estimation of bioconcentration:
The uptake and depuration curves of the test substance were obtained by plotting the concentration of the parent compound measured in Hyalella collected during the study (Ch) against time.

STEADY-STATE BIOCONCENTRATION FACTOR:
The steady-state bioconcentration factor (BCFss) was calculated as the ratio of the mean measured parent concentration in Hyalella at steady-state and the time weighted average concentration in water during the uptake phase:

DEPURATION RATE CONSTANT:
The depuration rate constant (k2) was calculated using the parent concentrations measured in Hyalella during the depuration:
A linear regression of the ln-transformed Ch data versus time was performed and k2 derived as the slope of the regression line.

UPTAKE RATE CONSTANT:
The uptake rate constant (k1) was calculated by non-linear regression analysis of the ratios Ch/TWA Cw against time during the uptake phase and including the depuration rate (k2) fitted before. The fitted model assumes an attenuation of uptake by simultaneous depuration, increasing with increasing Ch up to equilibrium between uptake and depuration.
Whether the uptake and depuration patterns could be sufficiently described by first order kinetics was checked by visual inspection of the curves when plotted against the measured sample point data.

KINETIC BIOCONCENTRATION FACTOR:
The kinetic bioconcentration factor (BCFk) was calculated as the ratio of the two first order kinetic rate constants.

GRAVIMETRIC LIPID DETERMINATION USING THE SMEDES EXTRACTION METHOD:
Additional amphipods collected at the onset and the end of the uptake period (each time 3x10) as well as at the end of the depuration period (2x10+4) were extracted by the lipid extraction method originally described by Smedes and recommended by OECD 305 for fish lipid determination.
Lipid content:
1.555 %
Time point:
other: Mean lipid content, calculated as the average of 0 and 144 h.
Remarks on result:
other:
Remarks:
The lipid content of the test animals was determined at the beginning and the end of the uptake period as well as the end of the depuration period. The mean lipid content of H. azteca measured at the end of depuration (288 h) was significantly lower than the values obtained after 0 and 144 h. The average lipid content of H. azteca of 1.555%, which was used for lipid normalization, was therefore calculated only based on the values obtained for 0 and 144 h. See table 5.3.1/3.
Conc. / dose:
25.3 µg/L
Temp.:
24.9 °C
pH:
7.82
Type:
BCF
Value:
326 L/kg
Basis:
total lipid content
Time of plateau:
24 h
Calculation basis:
steady state
Conc. / dose:
25.3 µg/L
Temp.:
24.9 °C
pH:
7.82
Type:
BCF
Value:
322 L/kg
Basis:
total lipid content
Time of plateau:
24 h
Calculation basis:
kinetic
Key result
Conc. / dose:
25.3 µg/L
Temp.:
24.9 °C
pH:
7.82
Type:
BCF
Value:
1 048 L/kg
Basis:
normalised lipid fraction
Remarks:
(5%)
Time of plateau:
24 h
Calculation basis:
steady state
Conc. / dose:
25.3 µg/L
Temp.:
24.9 °C
pH:
7.82
Type:
BCF
Value:
1 036 L/kg
Basis:
normalised lipid fraction
Remarks:
(5%)
Time of plateau:
24 h
Calculation basis:
kinetic
Key result
Elimination:
yes
Parameter:
DT50
Depuration time (DT):
0.15 d
Remarks on result:
other: = 3.6 hours
Key result
Elimination:
yes
Parameter:
DT90
Depuration time (DT):
0.69 d
Remarks on result:
other: rapid uptake and elimination kinetics of the test substance in Hyalella
Key result
Rate constant:
overall uptake rate constant (L kg-1 d-1)
Value:
1 377
Key result
Rate constant:
overall depuration rate constant (d-1)
Value:
4.27
Details on kinetic parameters:
Steady-state concentration (Ch, ss) was reached within 24 hours of exposure during the uptake phase (six successive sampling points within ± 20% of each other).
The steady-state BCF (BCFss) was calculated as 326 based on the Ch, ss of 8.25 mg/kg and the TWA Cw of 25.3 μg/L.
Based on the trajectory of the tissue concentrations, a kinetic concentration factor (BCFk) of 322 L/kg was estimated. The uptake and depuration rate constants k1 and k2 were calculated with 1377 L/kg d and 4.27 1/d, respectively. Due to the unexpected rapid uptake and elimination kinetics of the test substance in Hyalella, only one sampling time point prior to, and two data points after attainment of steady-state could be used to calculate k1 and k2. A visual inspection of goodness of fit of the curves to the measured data points and verification if first order kinetics apply, in order to allow for an evaluation of the reliability of the estimated kinetic rate constants was thus not possible. However, since the calculated BCFk is in good accordance with the BCFss value, the estimates of k1 and k2 can be considered reliable.
The calculated depuration half-life (t1/2) was approximately 3.6 hours (0.15 d) and the time required for 95% depuration (t95%) was reached after 0.69 days.
Metabolites:
Not measured
Results with reference substance (positive control):
Not applicable
Details on results:
- Results of chemical analyses of the test substance in H. azteca: Hyalella tissue concentrations increased rapidly at the beginning of the uptake phase reaching a mean concentration level of around 8.54 mg/kg already after the second sampling time point (24 h of exposure). Animals collected at the end of the uptake period (144 h of exposure) showed a comparable mean substance concentration of around 8.04 mg/kg. In-between these two
sampling time points of the uptake phase a slight increase and subsequent decrease of mean tissue concentrations was observed. However, apparent differences between the mean concentrations measured after 24 h and 144 h were not significant (p= 0.161) as shown by one way analysis of variance test performed including all data points between 24 h and 144h using SigmaStat 3.5, Systat Software, Inc.. The steady-state concentration in Hyalella (Ch,ss) of 8.25 ± 0.83 mg/kg (%SD 10.0) was therefore calculated as the average of the mean tissue concentrations measured between 24 h and 144 h of uptake. Hyalella tissue concentrations decreased rapidly immediately after stopping the exposure. Animals collected at the first sampling time point of the depuration period (144h+10h= 154h) showed a mean substance concentration of around 1.18 mg/kg which is equivalent to 85.3% depuration. The bioaccumulated test substance was almost completely eliminated after 24 hours of depuration (0.110 mg/kg, 98.6%). Thereafter, measured values were < LOQ until the end of the
depuration phase.

Table 5.3.1/2: Measured concentrations of the test substance in water (Cw) during the uptake phase and calculated TWA (time-weighted average) Cw

Time (h)

Cw test substance (µg/L)

4

28

48

71

96

119

143

24.9

25.7

25.3

24.3

25.1

26.0

25.9

TWA Cw

SD TWA Cw

%SD TWA Cw

25.3

0.56

2.20

Table 5.3.1/3: Measured mean concentrations of the test substance in Hyalella (Ch) during the uptake phase and calculated steady-state concentration (Ch,ss)

Time (h)

Ch test substance (mg/kg)

10

24

48

72

96

120

144

6.36

8.54

9.19

9.05

7.50

7.16

8.04

Ch,ss

SD Ch,ss

%SD Ch,ss

8.25*

0.83*

10.0*

* calculated based on values from 24 h to 144 h

Table 5.3.1/3: Mean lipid content (% of FW) of H. azteca at 0, 144 and 288 h

Time (h)

Mean lipid content (% FW)

0

1.549

144

1.561

288

0.971

Mean

1.555*

* calculated as the average of 0 and 144 h

Validity criteria fulfilled:
not applicable
Conclusions:
The bioaccumulation potential of the test substance was investigated in a flow-through test on H. azteca. The results show that the test substance is bioaccumulating in H. Azteca with a BCFss of 326 L/kg and a BCFss L of 1048 L/kg. The values obtained for the kinetic BCF (BCFk 322 L/kg and BCFk L 1036 L/kg) were in good agreement with the results for the steady-state BCF. Total analytic concentrations in the water phase were measured by GC/MS. Optimal experimental conditions were applied as shown by the results of the water analysis (< 20% deviation of the measured concentrations from the TWA Cw throughout the uptake phase). During the entire study high oxygen concentrations and only very low concentrations of nitrogen metabolites (nitrite, ammonia) far below their toxicity level were measured. A group of around 1600 animals was sufficient to run the present BCF study (single concentration, 12 Hyalella sampling time points) with H. Azteca. A toxic effect of the test substance to Hyalella is therefore not indicated. This is further supported by the results of an acute toxicity test carried out with Hyalella prior to this study showing that a toxicity effect from the low test concentrations applied in the present study is not expected. Reproduction of test animals during the experimental period may lead to depuration caused by the release of neonates. Therefore only male animals were used.
Executive summary:

An aqueous exposure bioconcentration study with non-radiolabeled test substance on Hyalella azteca was carried out under flow-through conditions at 25 +/- 2°C. The objective of the study was to determine the steady-state and kinetic bioconcentration factors (BCFss and BCFk) of the test substance in Hyalella azteca.

Hyalella were exposed to a nominal concentration of 27 μg/L during the uptake phase lasting for 144 h (6 days) followed by a depuration period of 144 h (6 days). Hyalella samples were taken at the beginning and after 10, 24, 48, 72, 96, 120, and 144 h of exposure, as well as after 10, 24, 48, 96 and 144 h of depuration to determine the trajectory of tissue concentrations in Hyalella by GC/MS. Additional animals were sampled at the beginning of exposure, as well as at the beginning and end of depuration to determine the lipid content of Hyalella. Throughout the study, test concentrations in water, as well as physical and chemical water parameters were determined to monitor the exposure conditions.

The time weighted average (TWA) concentration of the test substance in water during the uptake phase was 25.3 ± 0.56 μg/L. With a deviation of 2.20 % from the TWA Cw, sufficiently stable exposure conditions (< 20% deviation of the measured concentrations from the TWA Cw throughout the uptake phase) were applied. During the entire study high oxygen concentrations and only very low concentrations of nitrogen metabolites (nitrite, ammonia) far below their toxicity level were measured. Hyalella tissue concentrations increased rapidly at the beginning of the uptake phase reaching steady-state within 24 h of exposure. The steady-state concentration in Hyalella (Ch,ss) was calculated as 8.25 ± 0.83 mg/kg (%SD 10.0). Based on the Ch,ss and the TWA Cw, a BCFss of 326 L/kg was derived for Hyalella. The lipid normalized BCF (BCFss L) for Hyalella was calculated with 1048 L/kg using the average lipid content obtained for the animals of 1.555%. For the uptake and depuration rate constants (k1 and k2, respectively) values of 1377 L/kg d and 4.27 1/d were calculated, respectively. The values obtained for the kinetic BCF in Hyalella (BCFk 322 L/kg and BCFk L 1036 L/kg) were in good agreement with the results for the steady-state BCF.

A toxic effect of the test substance to Hyalella during the present study was not indicated.

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
experimental study
Adequacy of study:
weight of evidence
Study period:
Not specified
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Remarks:
Non-GLP study
Qualifier:
equivalent or similar to guideline
Guideline:
other: draft document of the new OECD TG for the S9 assay.
Deviations:
not specified
Qualifier:
equivalent or similar to guideline
Guideline:
other: DRAFT OECD TEST GUIDELINE Determination of in vitro intrinsic clearance using rainbow trout liver S9 sub-cellular fraction (RT-S9)
Deviations:
not specified
Principles of method if other than guideline:
A major determinant of chemical bioaccumulation in fish is the extent to which the chemicals undergo biotransformation. Fish liver based in vitro systems provide a reliable tool to determine chemical biotransformation rates in vitro. Incorporating such in vitro data into existing bioaccumulation prediction models substantially improves their performance as shown by several studies. In the present study liver S9 sub-cellular fraction of rainbow trout (Oncorhynchus mykiss) (RTL-S9) was used to determine the in vitro intrinsic clearance rate (CL in vitro,int) of the test substance. The determined parameter was used for an extrapolation of the whole animal in vivo biotransformation rate (Kmet) which was in turn incorporated into an existing mass-balance model for fish bioconcentration (BCF) prediction for the test substance.
The study included the following steps:
• Preparation and characterization of enzymatically active (live) S9 und enzymatically inactive (heat-inactivated, HI) S9 material
• Performance of preliminary experiments to evaluate the concentration-dependence of the enzymatic activity, to assess the kinetics of depletion, and to optimize the sampling schedule.
• Performance of independent substrate depletion assays using active and enzymatically inactive S9 material to determine the CL in vitro,int of the test substance according to the methods described in the draft of the new OECD test guideline “Determination of in vitro intrinsic clearance using rainbow trout liver S9 sub-cellular fraction (RTL-S9)”.
• Plot of the log10-transformed substrate concentrations against time for each experimental run to demonstrate a log-linear decline (high R2 value).
• Determination of a first-order elimination rate constant, Ke (1/h) and CL in vitro,int (mL/h/mg protein) for each experimental run.
• Determination of the mean CL in vitro,int and variability of CL in vitro,int (%SD) between independent in vitro runs for the test substance.
• Extrapolation of the whole animal biotransformation rate (Kmet) and modelling of the BCF, assuming two different binding dynamics of the test substance (expressed by two scenarios, through modifying the binding factor Fu).
GLP compliance:
no
Remarks:
Non-GLP study performed in Weight of Evidence approach for the determination of the bioaccumulation of the registered substance. In vitro studies are alternative methods to replace the use of animals.
Specific details on test material used for the study:
No additional information
Radiolabelling:
no
Details on sampling:
SAMPLE ANALYSIS
After receiving the samples from the biology department they were evaporated until dryness and resolved with 100 μL of toluene and directly measured via GC-MSD.
Vehicle:
yes
Details on preparation of test solutions, spiked fish food or sediment:
TEST CHEMICAL CONCENTRATION AND TEST DESIGN
The main experiment was performed with a chemical reaction concentration of 0.316 μM (C2 of the pre-study, 74.7 μg/L), selected based on the results from the preliminary experiments. For the test design a multiple vial approach was used, which was based on test design 2 “Independent replicates tests using multiple sample vials – Option 1: with cofactor mastermix” described in Annex 6 of the draft of the cited S9 assay OECD TG.
Incubations were carried out in multiple sample vials containing 200 μL total volume RTL-S9 reaction mixture (containing RTL-S9 fraction suspension, potassium phosphate buffer, alamethicin and cofactors master mix). Stopping solution (containing the internal standard) was added directly to each sample vial at the various time points. Replicates (e.g. triplicates) were carried out as independent tests (run1-3), using fresh solutions of RTL-S9 fraction suspension (i.e. active RTL-S9 plus test chemical, inactive HI-RTL-S9 plus test chemical) and the separate spiking solutions of the test chemical. Glass amber vials (4 mL, 13 mm diameter, caps including Teflon septum) were used as reaction vials, to account for a potential adsorption of the test chemical to the vessel walls and a decline of the test chemical due to photolysis.

PREPARATION OF TEST CHEMICAL SPIKING SOLUTION, BUFFER, COFACTORS AND STOPPING SOLUTION
A stock solution of the test substance in acetonitrile (ACN) was prepared at a concentration of 63.2 μM (14.99 mg/L) and stored at around 4°C. The stock solution was used to create the test chemical spiking solution with 12.64 μM (2.99 mg/L) by 1:5 dilution with the potassium phosphate (KPO4) reaction buffer. Separate spiking solutions for each in vitro run were prepared freshly from the stock solution on the day of the experiment. The stopping solution was prepared by dissolving the internal standard o-terphenyl in hexane at a concentration of 5 μg/L and was stored at around 4°C until use.
The 100 mM potassium phosphate (KPO4) reaction buffer was prepared by mixing together 88 mL of 100 mM potassium phosphate dibasic (K2HPO4) with 12 mL of 100 mM potassium phosphate monobasic (KH2PO4). For all potassium phosphate recipes ultrapure water was used. The pH of the reaction buffer was adjusted to 7.8 at 15°C using additional potassium phosphate dibasic (base) or potassium phosphate monobasic (acid). The buffer was stored at 4°C until usage and the pH re-adjusted on the day of use.
Alamethicin stock solution (10 mg/mL) was prepared in methanol by adding 0.5 mL methanol and 5 mg of alamethicin to the vial. The vial was recapped, carefully vortexed and the alamethicin stock solution stored as 25 μL aliquots in 1.5 mL microcentrifuge tubes at -20°C until use. alamethicin working solution in KPO4 buffer was prepared freshly on the day of the experiment and separately for each independent run by diluting one 25 μL aliquot of 10 mg/mL stock solution with 975 μL KPO4 buffer for a final concentration of 250 μg/mL.
The three cofactors β-nicotinamide adenine dinucleotide 2’-phosphate, tetrasodium salt (NADPH), uridine 5’ -diphosphoglucuronic acid, trisodium salt (UDPGA), and L-glutathione (GSH) were pre-weighed prior to the experiment and stored at -20°C. On the day of the experiment just prior to the start and separately for each run, 20 mM (16.67 mg/mL) NADPH, 20 mM (12.93 mg/mL) UDPGA, and 50 mM (15.37 mg/mL) GSH cofactor solutions were prepared freshly by dissolving in ice-cold KPO4 buffer. The solutions were stored on ice until use.
A concentrated solution of 3’-phosphoadenosine 5’-phosphosulfate (PAPS) at 10 mM was prepared (considering the purity of anhydrous acid and the water content) prior to the experiment and stored frozen as 50 μL aliquots. The 50 μL aliquot of pre-frozen 10 mM PAPS solution was diluted with 450 μL of KPO4 buffer to make a 1 mM PAPS solution. The dilution step was done on the day of the experiment immediately before addition of the master mix.
A master mix of cofactors was prepared just prior to the pre-incubation step on ice as follows
• 500 μL of 20 mM NADPH
• 500 μL of 20 mM UDPGA
• 500 μL of 50 mM GSH
• 500 μL of 1 mM PAPS
Everything was mixed well and stored on ice for the least amount of time before preparing the reaction mixtures.
Test organisms (species):
Oncorhynchus mykiss (previous name: Salmo gairdneri)
Details on test organisms:
Rainbow trout liver S9-sub-cellular fraction (RTL-S9)

PREPARATION OF RTL-S9
The RTL-S9 material used for the experiments in the present study (lot S9-01-2017) was prepared in-house according to the methods published by Johanning et al. Rainbow trout (Oncorhynchus mykiss; Teleostei, Salmonidae) used for the present study were obtained from Fischzucht Störk (Bad Saulgau, Germany) at a size of 2-4 cm. The fish were raised to their final body weight in the Fraunhofer IME husbandry, Schmallenberg, in charcoal-filtered and dechlorinated tap water at a temperature of 14-16°C. Feeding was carried out daily on working days at a feed ratio corresponding to the size of the fish and using a commercially available trout feed (Milkivit-type F-2P B40). A day/night light regime of 16/8 hours was applied. Specimens selected for the experiment demonstrated a healthy behavior and were fasted 24 hours prior to liver donation. Only female rainbow trout with a body weight of 247.19 to 447.35 g and immature oocytes (GSI<0.5) were used. The livers of 5 animals were pooled for the S9 fraction isolation to compensate a possible variability in the biotransformation potential of individual fish. The isolation procedure included an in situ liver perfusion step prior to tissue homogenization, to free the liver from blood and thus ensure high quality isolation results. Cells were homogenized using a glass rod prior centrifugation at 12000g at 4°C for 20 min to gather the S9 supernatant. The freshly prepared and pooled RTL-S9 was aliquoted (150 μL, 500 μL, 700 μL or 1400 μL) to 1.5 mL microcentrifuge tubes and placed in a -80°C freezer for storage.

CHARACTERISATION OF RTL-S9
- Determination of RTL-S9 protein content:
Two RTL-S9 samples (1x 150 μL and 1x 500 μL) were thawed in an ice-water bath and placed on ice until the ice crystals were dissolved. The protein content was determined using the Pierce BCA protein assay kit following the manufacturer’s instructions and using 96-well flat-bottom plates and a microplate reader. For each S9 sample, the extinction of two different dilutions (1:100 and 1:250) was measured twice. The protein content of the given RTL-S9 lot (S9-01-2017) was determined as the average of the protein content determined for the two S9 samples. The mean protein content of RTL-S9 lot S9-01-2017 calculated in this way was 27.29 mg/mL.
- Determination of the EROD activity of RTL-S9:
The isolated RTL-S9 lot was evaluated for the ability to catalyze phase I biotransformation reactions (CYP1A activity) by means of an Ethoxycoumarin O-deethylation (EROD) assay. Therefore, one RTL-S9 sample (1x 500 μL) was thawed in an ice-water bath. The EROD assay was carried out in triplicates in 100 mM KPO4 buffer (pH 7.8) at a S9-protein concentration of 1 mg/mL. The total incubation volume was 500 μL. The final NADPH concentration in the incubation system was 1 mM. Ethoxyresorufin was spiked at 0.5 μM using acetone as carrier. The final solvent concentration in the incubation system was 1%. Methanol (600 μL) was used to terminate the reactions after 30 minutes of incubation at 15°C. In addition to the triplicates of active samples (RTL-S9, KPO4, NADPH, acetone carrier, ethoxyresorufin), duplicates of matrix blanks (RTL-S9, KPO4, NADPH, acetone carrier but no ethoxyresorufin) and duplicates of inactive samples (HI-RTL-S9, KPO4, NADPH, acetone carrier, ethoxyresorufin) were run in parallel as controls. After centrifugation of the stopped samples for 5 min at 3000 x g and 4°C, the supernatants were subjected to fluorescence analysis at an excitation wavelength of 550 nm and an emission wavelength of 580 nm in 96-well plates. The EROD activity of the samples was determined by measuring the fluorescence against a resorufin standard curve. The EROD activity of the given RTL-S9 lot (S9-01-2017) was determined as the average of the EROD activities determined for the triplicates of the active sample. The mean EROD activity of 0.605 pmol/min/mg protein determined for the RTL-S9 lot used in this study was within the range of activities usually determined for rainbow trout liver S9 sub-cellular fraction.
- Preparation of heat-inactivated RTL-S9 (HI-RTL-S9):
RTL-S9 was enzymatically inactivated by heat-inactivation to provide negative control material for the substrate depletion test. The negative control was used to distinguish between enzymatic biotransformation and abiotic decrease by adsorption to the reaction vial, volatilization, and abiotic degradation. The HI-RTL-S9 was pre-diluted with KPO4 buffer (pH 7.8) to a protein concentration of 10 mg/mL. The heat-inactivated RTL-S9 (HI-RTL-S9) was then prepared by heating active RTL-S9 at 100°C (in boiling water) in a capped vial for 10 min. A larger volume of pre-diluted HI-RTL-S9 was prepared in this way and stored in aliquots at -20°C until the day of the experiment.

PREPARATION OF RTL-S9 SUSPENSION AND REACTION MIXTURES
One tube of active RTL-S9 and one tube of pre-diluted inactive HI-RTL-S9 were thawed in an ice-water bath. The active RTL-S9 (300 μL with 27.29 mg protein/mL) was diluted with 519 μL KPO4 buffer to a protein content of 10 mg/mL. Everything was mixed well and stored on ice. Thawing of the S9 material was carried out separately for each independent run within approximately 45 min before dosing. For each independent experimental run in the multiple vial approach, eight vials with active RTL-S9 and eight vials with enzymatically inactive HI-RTL-S9 were used.
The reaction mixtures for the RTL-S9 and HI-RTL-S9 were initially prepared in bulk by adding 800 μL of KPO4 buffer pH 7.8 to each of two vials (4 ml amber glass). The diluted (10 mg/mL) active RTL-S9 or enzymatically inactive HI-RTL-S9 (200 μL each) was added to the corresponding vial. Then, 100 μL of the alamethicin working solution was added to each of the vials and the vials pre-incubated on ice for 15 min. During the pre-incubation step, the cofactors master mix was prepared as described above. The master mix was re-vortexed and 800 μL were transferred to each of the pre-incubated reaction mixture vials. Each vial was gently swirled until thoroughly mixed.
Finally, the 16 reaction vials for each experimental run were prepared by adding 200 μL of the final active RTL-S9 or enzymatically inactive HI-RTL-S9 reaction mixture to the corresponding set of eight 4 mL amber glass vials. All reaction vials were placed into a shaking incubator and pre-incubated at the test temperature of 15°C for 10 minutes with gentle shaking (110 rpm).
Route of exposure:
aqueous
Justification for method:
other: Liver S9 sub-cellular fraction of rainbow trout was used to determine the in vitro intrinsic clearance rate of the test substance. The in vitro rates were extrapolated to the whole animal, to obtain the in vivo biotransformation rate, and then BCF.
Test type:
static
Water / sediment media type:
not specified
Total exposure / uptake duration:
40 min
Hardness:
No data
Test temperature:
15°C
pH:
7.8
Dissolved oxygen:
No data
TOC:
No data
Salinity:
Not applicable
Conductivity:
No data
Details on test conditions:
INCUBATION WITH TEST CHEMICAL AND STOPPING OF REACTION
The reaction is initiated by dosing 5 μL of the the test substance spiking solution directly into the suspension of each reaction vial. The total amount of solvent in the incubation mixture was thus 0.5%. The vials were swirled to distribute the chemical and tightly capped to prevent evaporation of the chemical. For sampling at the specified time points, 2000 μL of ice-cold stopping solution containing the internal standard was added directly into the respective reaction vials. To ensure thorough mixture of the sample, pipetting up and down in the solvent three times was performed. Collected samples were immediately stored at 4°C. Directly after completion of sampling, all samples were vortexed for 10 min and then stored at -20°C until further sample preparation and analysis. In order to prevent any reaction in the 0 min time point samples, for these reaction vials the stopping solution was dosed and mixed thoroughly with the reaction mixture prior to the addition of the test chemical. The pre-incubation time for the 0 min time point reaction vials was approximately only 6-8 min.

PRELIMINARY EXPERIMENTS
Preliminary studies (preliminary experiment 1 and 2) were conducted to establish specific reaction conditions needed to reliably measure in vitro intrinsic clearance rates of the test substance in RTL-S9. This included the testing of range finding conditions (e.g. substrate concentration and incubation time), the optimization of sampling time points, the verification of the test chemical stability in buffer (HI-RTL-S9 negative control experiment), and the verification of the suitability of the test system (multiple vial approach with incubation in closed vials). The primary goal of conducting preliminary experiments was to determine reaction conditions that result in first-order depletion kinetics. For the reaction to result in first-order depletion kinetics, it needs to be ensured that the substrate starting concentration is lower than the Michaelis-Menten affinity constant (Km). In order to evaluate the concentration-dependence of the reaction, three different concentrations were tested during the preliminary studies. The lowest value of 0.1 μM (C1, Low Value) was selected as low as possible considering the LOQ of the test substance and high enough to allow for at least 50% depletion during the in vitro assay. The highest value was selected as 1 μM (C3, High Value) as the spread between the high and low concentration in preliminary experiments should be at least one order of magnitude. The third concentration of 0.316 μM (C3, Mid Value) was selected based on the high and low value and calculated as Mid Value = √(High/Low). The three values were distributed well on a log scale.

Nominal and measured concentrations:
0.316 μM
Reference substance (positive control):
no
Details on estimation of bioconcentration:
DETERMINATION OF IN VITRO INTRINSIC CELARANCE RATES
Measured substrate concentrations were log10-transformed and plotted against time for each replicate. Obvious outliers identified by visual inspection of the plot were removed from the depletion curves and excluded from further calculations. A first-order elimination rate constant, Ke (1/h) was determined as -2.3 × slope of the log-linear decline. To obtain the in vitro intrinsic clearance (CL in vitro,int) the following formula was applied, while the protein concentration (CS9) mixture was set to 1mg/mL:
CL in vitro,int = Ke /CS9

IN VITRO-IN VIVO EXTRAPOLATION AND BCF PREDICTION
In vitro – in vivo extrapolation of Kmet and mass balance BCF model prediction were carried out in accordance to Nichols and as recommended by the DRAFT OECD TEST GUIDELINE Determination of in vitro intrinsic clearance using rainbow trout liver S9 sub-cellular fraction (RT-S9). Two different binding assumptions (i.e. Fu modelled and Fu = 1.0) were considered for BCF calculation to estimate upper and lower limits of hepatic clearance.
Remarks on result:
other: Not applicable
Key result
Conc. / dose:
0.316 other: µM
Temp.:
15 °C
pH:
7.8
Type:
BCF
Value:
1 869.3 L/kg
Basis:
not specified
Calculation basis:
kinetic
Remarks on result:
other: Fu modelled
Key result
Conc. / dose:
0.316 other: µM
Temp.:
15 °C
pH:
7.8
Type:
BCF
Value:
284.8 L/kg
Basis:
not specified
Calculation basis:
kinetic
Remarks on result:
other: Fu = 1.0
Remarks on result:
other: Not applicable
Details on kinetic parameters:
See "Details on results" and Tables in "Any other information on results incl. tables".
Metabolites:
Not measured
Results with reference substance (positive control):
Not applicable
Details on results:
RESULTS FROM THE MAIN TEST AND CALCULATION OF IN VITRO INTRINSIC CLEARANCE RATES
During the main test, three independent in vitro runs were performed at 0.316 μM over an incubation period of 40 min. Each experimental run with active RTL-S9 was carried out in parallel to the negative control HI-RTL-S9. A log-linear decline with R2 values of > 0.90 was demonstrated for each of the three active RTL-S9 runs. For the negative HI-RTL-S9 controls, with 17.07% no significant loss according to the draft of the OECD TG (<20% of the active) was observed. The calculation of the depletion rate was based on 8 data points for run I and III. For run II, 6 data points could be used for the rate calculation since 2 data points had to be excluded after being identified as outliers. According to the draft of the new OECD TG for the S9 assay, the depletion rate should be determined from the linear portion of the curve with a minimum of 4 data points for a reliable estimate of CL in vitro,int. For the three in vitro runs, a mean CL in vitro,int of 1.113 mL/h/mg protein was calculated. The rates ranged from 0.962 to 1.284 mL/h/mg protein for run II and III, respectively. The %SD of the mean CL in vitro,int was calculated with 11.89%.

IN VITRO-IN VIVO EXTRAPOLATION AND BCF PREDICTION
Depending on the binding assumption applied in the BCF calculation, in vitro BCF estimates of 1869.3 L/kg (Fu modelled) and 284.8 L/kg (Fu = 1.0) were obtained for the test substance.

See Tables in "Any other information on results incl. tables"
Reported statistics:
None

Table 5.3.1/2: Calculated in vitro intrinsic clearance rate (CL in vitro,int) for active RTL-S9

Experiment

Slope of linear depletion curve

Elimination rate constant, Ke (1/h)

CL in vitro,int (mL/h/mg protein)

Run I

Run II

Run III

Mean

SD

%SD

0.4759

0.4181

0.5584

/

/

/

1.095

0.962

1.284

/

/

/

1.095

0.962

1.284

1.113

0.132

11.89

Table 5.3.1/3: Calculated in vitro intrinsic clearance rate (CL in vitro,int) for inactive HI-RTL-S9

Experiment

Slope of linear depletion curve

Elimination rate constant, Ke (1/h)

CL in vitro,int (mL/h/mg protein)

Run I

Run II

Run III

Mean

SD

%SD

0.0587

0.0841

0.1051

/

/

/

0.1350

0.1935

0.2418

/

/

/

0.135

0.193

0.242

0.190

0.044

23.98

% of active rate              /                            /                                   17.07

Validity criteria fulfilled:
not applicable
Conclusions:
Two different binding assumptions were considered for BCF calculation to estimate upper and lower limits of hepatic clearance. Depending on the binding assumption applied in the BCF calculation, in vitro BCF estimates of 1869.3 l/kg (Fu modelled) and 284.8 l/kg (Fu = 1.0) were obtained for the test substance.
Executive summary:

In the present study, liver S9 sub-cellular fraction of rainbow trout (Oncorhynchus mykiss) (RTL-S9) was used to determine the in vitro intrinsic clearance rate (CL in vitro,int) of the test substance. A “multiple vial approach” (i.e. separate reactions for each sampling time point) was selected as test system for the in vitro incubations, due to the high hydrophobicity and volatility of the test chemical. During two preliminary experiments, specific reaction conditions were optimized to reliably measure in vitro intrinsic clearance rates of the test substance in RTL-S9. Based on the results of the preliminary experiments, the optimal test concentration, incubation period, and sampling time points for the main test were identified.

The main test was carried out with 3 independent in vitro runs according to test design 2 suggested in the draft of the new OECD TG for the S9 assay. The obtained mean CL in vitro,int of the test substance in active RTL-S9 was 1.113 mL/h/mg protein with a %SD of 11.89%. As required for a reliable determination of CL in vitro,int in RTL-S9, the experiments were carried under reaction conditions that result in first order kinetics. The log10 -transformed substrate concentrations plotted against time for each in vitro run demonstrated a log-linear decline with high R2 values (> 0.90). The depletion rates were determined with 6 to 8 data points satisfying the requirement of a minimum of 4 data points for CL in vitro,int calculation in the draft OECD TG. All measured concentrations used for CL in vitro,int calculation of the main test were above the LOQ of 9.11 μg/L. No significant loss according to the draft of the cited OECD TG (<20% of the active) was observed for the negative controls. The observed depletion of the test substance in active RTL-S9 could thus be attributed to enzymatic biotransformation and other processes like abiotic decrease of the test substance by adsorption to the reaction vial, volatilization, or abiotic degradation.

Despite using the same S9 -lot, the mean CL in vitro,int of the test substance in active RTL-S9 obtained from the main test was lower compared to the rates determined for the same test concentration and incubation period during the preliminary experiments (1.399 and 1.569 mL/h/mg protein, respectively). The observed difference between the results from the preliminary experiments and the main test (i.e. 16.94 %SD), however, was still within the expected variability of CL in vitro,int for fish liver based in vitro assays. As shown by Fay et al. in an inter-laboratory comparison, the variability of CL in vitro,int for the same chemical determined in cryopreserved trout hepatocytes can range up to 30 % within the same laboratory despite using the same cell lot. The RTL-S9 lot used in the present study was prepared in-house. As recommended, the livers of five animals were pooled for S9 -isolation in order to reduce the variability of metabolic activity originating from individual differences in the metabolic potential. Prior to the use in this study, the RTL-S9 lot was evaluated for the ability to catalyze phase I metabolic enzyme reactions by means of the EROD assay. The pH of the potassium phosphate buffer used for the incubations was adjusted to 7.8±0.1. The protein content in the reaction mix was 1 mg/mL and thus in the range of 0.25 -2 mg/mL recommended for the S9 -assay. Finally, the incubation temperature was adjusted to the acclimation temperature of the donor fish, i.e. 15 °C. As demonstrated above, all criteria for the in vitro test to be valid according to the draft of the new OECD TG were met in the present study. The calculated CL in vitro,int of the test substance in RTL-S9 is therefore considered reliable.

In vitro BCF values of 1869.3 and 284.8 were estimated for the test substance. The different estimates can be explained according to the OECD Draft Guidance document as follows:

“For some hydrophobic chemicals, there was a poor correlation between empirical BCFs and BCFs predicted using the full modelled binding assumption (i.e., fu = fu,P/fu,HEP or fu,S9). Instead, there was better agreement using the binding assumption fu=1.0, especially for slowly metabolized chemicals. These observations suggest that hepatic clearance values predicted using the full binding assumption under-predict true levels of in vivo clearance resulting in overestimation of measured BCFs. A systematic bias toward under-prediction of in vivo hepatic clearance by in vitro systems (hepatocytes, microsomes) derived from human liver tissue has been noted by several authors.”

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
experimental study
Adequacy of study:
weight of evidence
Study period:
From 21 August 2008 to 22 October 2008
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
guideline study with acceptable restrictions
Remarks:
This study was performed according to OECD Guideline 305 without GLP statement. All validity criteria were fulfilled but restrictions should be taken into account. The BCF result was not expressed as normalised to a fish with 5% lipid content, therefore the BCF was recalculated. In addition, this results seems to be overestimated because the BCF was based upon total radiolabelled residues without taking into account the potential of degradation of the test substance and the total radioactive residues in fish were not measured between 28 days (maximum accumulation in fish) and 35 days (total depuration).
Qualifier:
according to guideline
Guideline:
OECD Guideline 305 (Bioconcentration: Flow-through Fish Test)
Version / remarks:
, 14 June 1996
Deviations:
no
Principles of method if other than guideline:
Not applicable
GLP compliance:
no
Remarks:
This study was performed according to OECD Guideline 305 without GLP statement.
Specific details on test material used for the study:
- Name of test material (as cited in study report): ST 11 C 08
- Water solubility: 1.88 mg/L
- Radiochemical purity (if radiolabelling): 99.6%
- Specific activity (if radiolabelling): 5.27 MBq/mg (142.6 µCi/mg)
- Locations of the label (if radiolabelling): 14C-ST 11 C 08
- Expiration date of radiochemical substance (if radiolabelling): The radiochemical purity was determined before use. Therefore no expiry date was needed.
- Stability under test conditions: The initial radiochemical purity of 14C-ST 11 C 08 as determined by HPLC was 97.4%. The radiochemical purities of the stock solution (stock solution 2) and the administration solution were found to be 96.0% and 97.4%, respectively. An aliquot of stock solution 2 was analysed by HPLC after about 8 weeks storage at -20°C. The purity was found to be 95.0%. These results proved that the test item was sufficiently stable under test conditions.
Radiolabelling:
yes
Details on sampling:
- Sampling intervals/frequency for test organisms and sample storage conditions: See table 5.3.1/1 in "Any other information on materials and methods incl. tables". On each sampling occasion, four fish were collected randomly from each exposure tank, rinsed with water, sacrificed in 1.5% (v/v) 2-phenoxy-ethanol in purified water and blotted dry. Four fish were analysed immediately following sacrifice. At one time interval (after 28 days of accumulation) eight fish were sampled and stored at about -20 °C for possible additional analyses. Additionally, ten fish were sampled for lipid determination.
- Sampling intervals/frequency for test medium samples: See table 5.3.1/1 in "Any other information on materials and methods incl. tables". Daily during the uptake phase and at selected time points during the depuration phase, duplicate samples of appropriate volumes (10 mL) were removed from the treated tank and as far as appropriate from the corresponding mixture chambers. Radioactivity was directly determined. Additionally at the last time interval, duplicate samples of 500 mL were taken and stored at -20 °C for possible additional analyses.
- Details on sampling and analysis of test organisms and test media samples (e.g. sample preparation, analytical methods): Duplicate samples of 10 mL water were directly analysed for total radioactivity by LSC. Four fish were individually analysed for total radioactivity at each sampling point. The fish were weighed and solubilized at 40 °C for 48 - 96 hours with the tissue solubilizer Solvable (Perkin Elmer, about 1 mL solubilizer per 100 mg fish weight). Thereafter, duplicate solubilized subsamples (corresponding to 100 or 200 mg) were measured by LSC. The 2 x 5 fish sampled on day 28 were used to determine the fish lipid/wet weight ratio at the end of the accumulation period by means of chloroform/methanol. The additional eight fish sampled on day 28 were weighed and stored at -20 °C. However, as the average BCF value was below 1000, these fish were not further analysed and remained stored at -20 °C.
Vehicle:
yes
Details on preparation of test solutions, spiked fish food or sediment:
PREPARATION AND APPLICATION OF TEST SOLUTION (especially for difficult test substances)
Based on the target concentration of 10 μg/L and based on a total volume of 7750 L (assuming a flow-through volume of 250 L per day during 3 days pre-treatment, and 28 days treatment), an amount of 78 mg 14C-ST 11 C 08 at a specific radioactivity of 0.527 MBq/mg was needed. Including an appropriate excess, 110 mg at 0.527 MBq/mg were prepared (approx. 60 MBq).
- Stock solution 1: The total amount of the available material (14.42 mg at 5.27 MBq/mg) was dissolved in 10 mL ethanol. The radioactivity was determined and an amount of 15.60 mg was measured.
- Stock solution 2: Based on the required amount of about 11 mg 14C-ST 11 C 08 at 5.27 MBq/mg, 7.20 mL of stock solution 1 were added to and 99.0573 mg unlabelled test item in a volumetric flask (weighed amount: 102.65 unlabelled material with a purity of 96.5 %). The volume was filled up to 100 mL with ethanol. Using liquid scintillation counting (LSC), the amount was determined to be 109.9675 mg at a specific activity 0.5229 MBq/mg. The final concentration in stock solution 2 was 1.0997 mg/mL.
- Preparation of application solution: Aliquot of the stock solution 2 were diluted in ethanol. Before initiating the test, level in the treated tank was adjusted by adding 0.68 mL of stock solution 2 to the tank. Throughout the entire test, the system was equilibrated with tap water at a flow-through volume of 250 L/day, to which 14C-ST 11 C 08 was added from the application solutions at a rate of 25 mL/24 h to achieve final target concentrations of 10 μg/L.
Test organisms (species):
Oncorhynchus mykiss (previous name: Salmo gairdneri)
Details on test organisms:
TEST ORGANISM
- Common name: rainbow trout
- Strain: no data
- Source: Forellenzucht Hohler, Zeiningen, Switzerland.
- Age at study initiation (mean and range, SD): no data
- Length at study initiation (lenght definition, mean, range and SD): no data
- Weight at study initiation (mean and range, SD): 2.5-3.5g
- Weight at termination (mean and range, SD): no data
- Method of breeding: no data
- Health status: good
- Description of housing/holding area: no data
- Feeding during test: yes
The fish were fed daily (H.U. Hofmann AG, HOKOVIT, “Forellenfutter”, diet of known lipid and total protein content), based on about 3% of the average fish body weight during acclimation and during the study, taking into account increasing body weights and the decreasing number of fish per sampling interval.

ACCLIMATION
- Acclimation period: at least two weeks
- Acclimation conditions (same as test or not): same as test
Route of exposure:
aqueous
Test type:
flow-through
Water / sediment media type:
natural water: freshwater
Total exposure / uptake duration:
28 d
Total depuration duration:
14 d
Hardness:
Water hardness measured once during the test (on day 6 of exposure) was 12 °d (21.36 °f) or 2.136 mmol/L for the treated tank.
Test temperature:
Temperature was monitored from day 0 to day 28 and measurements ranged from 13.0 – 15.4°C.
pH:
pH was monitored from day 0 to day 28 and measurements ranged from 7.9 - 8.3.
Dissolved oxygen:
Oxygen concentration was monitored from day 0 to day 28 and measurements ranged from 6.7 – 9.2 mg/L.
TOC:
TOC values were not measured within this study. However, historical TOC values for tank water at the Test Facility are < 2 mg C/L.
Salinity:
Not applicable
Details on test conditions:
TEST SYSTEM
- Test vessel: One aerated and temperature controlled tank containing 75 L. The tank was dosed with the 14C-labelled test item from an application solution with a flow-through volume of 250 L/24 hours (i.e. about 3 times the tank volume), delivered via a Hamilton dispenser unit into a mixing flask where the application solution was pre-diluted with tap water (Figure 6). Before placing the fish in the tank, the flow-through system was run for 2-3 times 24 hours with labelled test item (i.e. about 7-10 tank volumes) to ensure a stabilized concentration of the test item at the onset of the uptake phase.
- No. of organisms per vessel: 60
- No. of vessels per concentration (replicates): not applicable
- No. of vessels per control / vehicle control (replicates): not applicable
- Biomass loading rate: 0.55 g/L/day, based on a daily flow-through volume of 250 L.

TEST MEDIUM / WATER PARAMETERS
- Source/preparation of dilution water: Local tap water (non chlorinated well water of drinking water quality), reduced for total hardness by ion exchange

OTHER TEST CONDITIONS
- Adjustment of pH: no
- Photoperiod: 16 hours light
- Light intensity: approximately 300-400 lux

RANGE-FINDING / PRELIMINARY STUDY
None
Nominal and measured concentrations:
- Nominal concentration: 10 µg/L
- Average exposure concentration: 9.46 µg/L. See table 5.3.1/2 in "Any other information on results incl. tables"
Reference substance (positive control):
no
Details on estimation of bioconcentration:
Not applicable
Lipid content:
76.1 other: mg/g fish
Time point:
end of exposure
Remarks on result:
other: none
Lipid content:
78.4 other: mg/g fish
Time point:
end of exposure
Remarks on result:
other: none
Type:
BCF
Value:
864 dimensionless
Basis:
other: average of total radioactivity in parent equivalents in the fish during exposure related to the concentration of total radioactivity in the water.
Time of plateau:
3 d
Calculation basis:
steady state
Remarks on result:
other: Conc.in environment / dose:9.46 µg eq/L
Type:
BCF
Value:
11 189 dimensionless
Basis:
total lipid content
Time of plateau:
3 d
Calculation basis:
steady state
Remarks on result:
other: Conc.in environment / dose:9.46 µg eq/L
Key result
Type:
BCF
Value:
559.4 dimensionless
Basis:
normalised lipid fraction
Time of plateau:
3 d
Calculation basis:
steady state
Remarks on result:
other: recalculated
Remarks:
Conc.in environment / dose:9.46 µg eq/L
Key result
Elimination:
yes
Parameter:
DT50
Depuration time (DT):
2 d
Details on kinetic parameters:
No data
Metabolites:
Not measured
Results with reference substance (positive control):
Not applicable
Details on results:
- Mortality of test organisms: none
- Behavioural abnormalities: none
- Observations on body length and weight: no data
- Other biological observations: none
- Organ specific bioaccumulation: no data
- Bound residues forming a plateau: no data
- Mortality and/or behavioural abnormalities of control: no data
- Loss of test substance during test period: See table 5.3.1/3 in "Any other information on results incl. tables". See Figure in "Illustration".
- Results with vehicle control: not applicable
Reported statistics:
None

Table 5.3.1/2: Actual concentration of total radioactivity of 14C-ST 11 C 08 in the exposure water during accumulation period.

Time interval (days)

Exposure tank (mean of duplicates)

comments

Dose (µg eq/L)

-3

09:30 a.m.

10.08

-3

11:30 a.m.

9.18

-3

01:30 p.m.

9.20

-2

07:15 a.m.

6.59

-2

09:30 a.m.

10.0

-2

01:30 p.m.

9.17

-2

15:00 p.m.

8.56

-1

08:30 a.m.

7.51

-1

11:00 a.m.

10.01

-1

13:00 p.m.

11.29

-1

15:00 p.m.

12.03

0

Before adding fish

11.18

0

After adding fish

9.49

0

 

7.92

1

 

8.13

1

 

8.65

2

 

9.16

3

 

10.08

4

 

10.49

5

 

10.41

6

 

9.90

7

 

9.68

8

 

10.00

9

 

10.04

10

 

9.58

11

 

9.95

12

 

9.86

13

 

10.55

14

 

9.51

15

 

9.43

16

 

9.13

17

 

9.34

18

 

10.34

19

 

9.64

20

 

9.73

21

 

8.91

22

 

8.91

23

 

8.91

24

 

8.97

25

 

8.84

26

 

9.01

27

 

9.15

28

 

9.03

Mean

(Day 1 to 28)

9.46

± Standard Deviation

 

0.65

Table 5.3.1/3: Concentration of radioactivity in the whole fish during the test at an average dose level of 9.46 µg/L

Phase

Time interval

Concentration of radioactivity in fish

Days after the onset of accumulation

Days after the onset of depuration

Dose

µg eq/g

%*

Accumulation period

3

5

7

14

21

28

-

-

-

-

-

-

7.498

8.082

7.353

8.089

8.094

9.932

-

-

-

-

-

-

Mean

± Standard Deviation

8.175

0.392

-

-

Depuration period

35

42

7

14

0.740

0.326

9.1

4.0

* Residual radioactivity in fish in percent of the average concentration in fish during exposure.

Validity criteria fulfilled:
yes
Conclusions:
The BCF value of 560 (recalculated, expressed as normalised to a fish with 5% lipid content) and the depuration half-life of 2.0 days indicate that the test substance did not bioaccumulate in the rainbow trout. In addition, this results seems to be overestimated because the BCF was based upon total radiolabelled residues without taking into account the potential of degradation of the test substance and the total radioactive residues in fish were not measured between 28 days (maximum accumulation in fish) and 35 days (total depuration).
Executive summary:

This study was performed according to OECD Guideline 305 without GLP statement, to assess the bioconcentration and depuration characteristics of the test substance in the rainbow trout under dynamic flow through system.

The fish were continuously exposed to 14C-ST 11 C 08 at an average dose level of 9.46 μg eq/L for 28 days. After the exposure, the fish were transferred to flowing untreated water and the depuration of radioactivity was followed for further 14 days. Temperature, pH and oxygen concentration were monitored from day 0 to day 42 and were within acceptable limits; measurements ranged from 13.0-15.4 °C, 7.9-8.3 and 6.7-9.2 mg/L, respectively.

The plateau levels could be determined as average concentration level in the fish. The radioactive residues in fish during exposure amounted on average to 8.175 ± 0.392 μg eq/g. The radioactive residues during depuration decreased to 0.740 μg eq/g on day 35 and to 0.326 μg eq/g on day 42.

Depuration half-life calculations was performed based on the concentration of radioactivity in fish on day 28 (exposure) and the values measured during 14 days of depuration. For the dose level selected, a depuration half-life of 2.0 days was calculated (correlation R² = 0.999).

For the dose level selected, based on the radioactivity levels in fish after exposure to 14C-ST 11 C 08 at an average dose level of 9.46 μg eq/L, the average steady state bioconcentration factor (BCF) amounted to 864 ± 41.

BCFlipid based on the lipid content measured in representative fish amounted to 11189.

All validity criteria were fulfilled but restrictions should be taken into account. The BCF result was not expressed as normalised to a fish with 5% lipid content, therefore the BCF was recalculated, at 560. In addition, this results seems to be overestimated because the BCF was based upon total radiolabelled residues without taking into account the potential of degradation of the test substance and the total radioactive residues in fish were not measured between 28 days (maximum accumulation in fish) and 35 days (total depuration).

In conclusion, the BCF value of 560 indicated that the test substance did not bioaccumulate in the rainbow trout.

Endpoint:
bioaccumulation in aquatic species: fish
Type of information:
read-across from supporting substance (structural analogue or surrogate)
Adequacy of study:
key study
Justification for type of information:
REPORTING FORMAT FOR THE ANALOGUE APPROACH
[further information is included as attachment to Iuclid section 13]

1. HYPOTHESIS FOR THE ANALOGUE APPROACH
This read-across approach is based on the hypothesis that the source and target substances have similar physico-chemical and environmental fate properties because of their structural similarity.

2. SOURCE AND TARGET CHEMICAL(S) (INCLUDING INFORMATION ON PURITY AND IMPURITIES)
Both the target and the source substances are multi-constituents. They are structurally related, as stereoisomers: the source substance is the racemic form (±) of the 9bR* isomer, while the target substance is the racemic form (±) of the 9bS* isomer. Some impurities were reported above 1.0% but none of these impurities may contribute to the classification of the substances. Therefore, it is concluded that the impurities will not affect the validity of the read-across.

3. ANALOGUE APPROACH JUSTIFICATION
A read-across approach for the octanol-water partition coefficient (log Kow) of the target substance was used with the source substance. Indeed, the partition coefficient should not be significantly impacted, with log Kow expected >5. Moreover, as the target substance is more soluble than the source substance, it could be anticipated that the log Kow would also be lower for the target substance and therefore the read-across is a worst-case. In the same way, the bioaccumulation potential of the target substance would be lower than the source substance. Hence, it's considered suitable and scientifically justified (worst case) to read-across the bioaccumulation study performed on the source substance to fill the bioaccumulation endpoint of the target substance.

4. DATA MATRIX
See attached document in Iuclid section 13.
Reason / purpose for cross-reference:
read-across source
Reason / purpose for cross-reference:
read-across: supporting information
Remarks:
Read-across justification document
GLP compliance:
no
Remarks:
Non-GLP study performed in Weight of Evidence approach for the determination of the bioaccumulation of the registered substance.
Type:
BCF
Value:
864 dimensionless
Basis:
other: average of total radioactivity in parent equivalents in the fish during exposure related to the concentration of total radioactivity in the water.
Time of plateau:
3 d
Calculation basis:
steady state
Remarks on result:
other: Conc.in environment / dose:9.46 µg eq/L
Type:
BCF
Value:
11 189 dimensionless
Basis:
total lipid content
Time of plateau:
3 d
Calculation basis:
steady state
Remarks on result:
other: Conc.in environment / dose:9.46 µg eq/L
Key result
Type:
BCF
Value:
559.4 dimensionless
Basis:
normalised lipid fraction
Time of plateau:
3 d
Calculation basis:
steady state
Remarks on result:
other: recalculated
Remarks:
Conc.in environment / dose:9.46 µg eq/L
Key result
Elimination:
yes
Parameter:
DT50
Depuration time (DT):
2 d
Conclusions:
The BCF value of 560 (recalculated, expressed as normalised to a fish with 5% lipid content) and the depuration half-life of 2.0 days indicate that the source substance (and the target substance by read-across) did not bioaccumulate in the rainbow trout. In addition, this results seems to be overestimated because the BCF was based upon total radiolabelled residues without taking into account the potential of degradation of the test substance and the total radioactive residues in fish were not measured between 28 days (maximum accumulation in fish) and 35 days (total depuration).
Executive summary:

The Bioaccumulation endpoint is a standard information requirement of Annex IX of the REACH Regultion but is needed in the present dossier to complete the PBT Assessment of the target substance. No bioaccumulation test is available on the target substance, therefore good quality data for a related source substance have been read across for this endpoint. The target and the source substances are Cycloalkane ethers multi-constituents. They are structurally related, in that the source substance is the racemic form (±) of the 9bR* isomer while the target substance is the racemic form (±) of the 9bS* isomer. Some impurities were reported above 1.0% but none of these impurities may contribute to the classification of the substances. In addition to the composition/structural similarity, a read-across approach for the octanol-water partition coefficient (log Kow) of the target substance was used with this source substance. Indeed, the partition coefficient should not be significantly impacted, with log Kow expected >5. Moreover, as the target substance is more soluble than the source substance, it could be anticipated that the log Kow would also be lower for the target substance and therefore is a worst-case value. In the same way, it could be anticipated that the bioaccumulation potential of the target substance would be lower than the source substance. Hence, it's considered suitable and scientifically justified (worst case) to read-across the bioaccumulation study performed on the source substance to fill the bioaccumulation endpoint of the target substance.  

The experimental study was performed according to OECD Guideline 305 without GLP statement, to assess the bioconcentration and depuration characteristics of the source substance in the rainbow trout under dynamic flow through system.

The fish were continuously exposed to 14C-ST 11 C 08 at an average dose level of 9.46 μg eq/L for 28 days. After the exposure, the fish were transferred to flowing untreated water and the depuration of radioactivity was followed for further 14 days. Temperature, pH and oxygen concentration were monitored from day 0 to day 42 and were within acceptable limits; measurements ranged from 13.0-15.4 °C, 7.9-8.3 and 6.7-9.2 mg/L, respectively.

The plateau levels could be determined as average concentration level in the fish. The radioactive residues in fish during exposure amounted on average to 8.175 ± 0.392 μg eq/g. The radioactive residues during depuration decreased to 0.740 μg eq/g on day 35 and to 0.326 μg eq/g on day 42.

Depuration half-life calculations was performed based on the concentration of radioactivity in fish on day 28 (exposure) and the values measured during 14 days of depuration. For the dose level selected, a depuration half-life of 2.0 days was calculated (correlation R² = 0.999).

For the dose level selected, based on the radioactivity levels in fish after exposure to 14C-ST 11 C 08 at an average dose level of 9.46 μg eq/L, the average steady state bioconcentration factor (BCF) amounted to 864 ± 41.

BCFlipid based on the lipid content measured in representative fish amounted to 11189.

All validity criteria were fulfilled but restrictions should be taken into account. The BCF result was not expressed as normalised to a fish with 5% lipid content, therefore the BCF was recalculated, at 560. In addition, this results seems to be overestimated because the BCF was based upon total radiolabelled residues without taking into account the potential of degradation of the test substance and the total radioactive residues in fish were not measured between 28 days (maximum accumulation in fish) and 35 days (total depuration).

In conclusion, the BCF value of 560 indicated that the test substance did not bioaccumulate in the rainbow trout.

Description of key information

Weight of Evidence approach with an Integreated Testing Strategy:

The extensive bioaccumulation assessment on the registered substance confirms that the substance is not bioaccumulative in aquatic organisms.

Key value for chemical safety assessment

BCF (aquatic species):
560 dimensionless

Additional information

The Bioaccumulation endpoint is a standard information requirement of Annex IX of the REACH Regultion. However, as the registered substance is not readily biodegradable and there is no sufficient degradation after 100 days to prove the non-persistence according to PBT/vPvB criteria, more investigations on the bioaccumulation potential of the registered substance in aquatic organisms were needed in the present dossier (Annex VII) to ensure the non-vPvB properties of the registered substance.

In this way, an Integrated Testing Strategy with three Tiers has been developed and presented in a Weight of Evidence approach: in silico predictions (initial tier), alternative innovative approach for regulatory and pipeline screening (medium tier) and fish in vivo study (high tier).

The initial tier is presented in two steps:

- Step 1: Physico-chemical determination.

The first step is the determination of the log Kow value. The slow-stir study performed on a source substance was retained as key study with an experimental log Kow value at 5.09. As the log Kow is greater than 4.5, the registered substance was considered as a potential B and/or vB substance according to PBT criteria.

- Step 2: Computer models estimate BCF.

The experimental log Kow value was used in different computer models to predict the bioconcentration factor (BCF). Five QSAR models and one calculation method were used and the predictions were presented in the table below.

QSAR models or calculation method

Retained BCF value

Klimisch score

Comments

TGD part II

4232

Rel.4

Calculation method

BCFBAF model v3.01, Regression-based estimate method (EPI Suite v4.1)

1060

Rel.2

The substance is fully characterised and within the applicability domain

BCFBAF model v3.01, Arnot-Gobas method (EPI Suite v4.1)

2152 - 2563

Rel.2

The substance is fully characterised and within the applicability domain

BCF base-line model v02.09 (OASIS Catalogic v5.12.1)

4265

Rel.3

The substance is out of the interpolation stuctural space.

BCF CAESAR v2.1.14 model (VEGA QSAR models platform)

153

Rel.3

The predicted compound could be out of the Applicability Domain of the model

BCF Meylan v1.0.3 model (VEGA QSAR models platform)

1806

Rel.3

The predicted compound is outside the Applicability Domain of the model

BCF KNN/Read-Across 1.1.0 model (VEGA QSAR models platform)

66

Rel.3

The predicted compound is outside the Applicability Domain of the model

The estimated BCF values for the registered substance were comprised between 66 and 4232, calculated using the experimental log Kow value at 5.09. Based on these BCF predictions (highest values), the registered substance seems to be bioaccumulable (BCF > 2000) but not very bioaccumulable (BCF < 5000) according to PBT criteria.

The medium tier relates to alternative approach, developped by Regulators together with industry scientists and academia, which use no or less animals.

The first study (Fraunhofer, 2017) is a bioconcentration test with an invertebrate species, Hyalella atzeca. Hyalella is a freshwater amphipod and therefore is an alternative to animal test with fish. Fraunhofer is promoting this approach as a replacement of the OECD Guideline 305. In this study, H. atzeca were exposed, under flow-through conditions, to non-radiolabeled registered substance during the uptake and depuration periods (6 days update phase / 6 days depuration phase). The results show that the registered substance is bioaccumulating in H. Azteca with a BCFss (steady state) of 326 L/kg, corresponding to a normalised BCFss to 5% lipid content of 1048 L/kg.

The second study (Fraunhofer, 2018) is a biotransformation in vitro assay. A major determinant of chemical bioaccumulation in fish is the extent to which the chemicals undergo biotransformation. Fish liver based in vitro systems provide a reliable tool to determine chemical biotransformation rates in vitro. Incorporating such in vitro data into existing bioaccumulation prediction models substantially improves their performance. In this study, liver S9 sub-cellular fraction of rainbow trout (Oncorhynchus mykiss) was used to determine the in vitro intrinsic clearance rate of the registered substance (1.113 mL/h/mg protein). The determined parameter was used for an extrapolation of the whole animal in vivo biotransformation rate which was in turn incorporated into an existing mass-balance model for fish bioconcentration (BCF) prediction for the registered substance. In vitro BCF values of 1869 and 285 were estimated for the registered substance, assuming two different binding dynamics of the substance.

Based on these BCFs values (comprised between 285 and 1869), the registered substance is not bioaccumulable (BCF < 2000) according to PBT criteria and is well metabolised in fish S9 liver system.

Finally, the high tier relates to a more definitive stage: the fish in vivo study.

No in vivo bioaccumulation test on fish is available on the registered (target) substance, but a good quality data is available on a source substance. The target and the source substances are Cycloalkane ethers multi-constituents. They are structurally related, in that the source substance is the racemic form (±) of the 9bR* isomer while the target substance is the racemic form (±) of the 9bS* isomer. Some impurities were reported above 1.0% but none of these impurities may contribute to the classification of the substances. In addition to the composition/structural similarity, a read-across approach for the octanol-water partition coefficient (log Kow) of the target substance was used with the source substance. Indeed, the partition coefficient should not be significantly impacted, with log Kow expected >5. Moreover, as the target substance is more soluble than the source substance, it could be anticipated that the log Kow would also be lower for the target substance and therefore is a worst-case value. In the same way, it could be anticipated that the bioaccumulation potential of the target substance would be lower than the source substance. Hence, it's considered suitable and scientifically justified (worst case) to read-across the bioaccumulation study performed on the source substance to fill the bioaccumulation endpoint of the target substance.  

This study (Harlan, 2009) was performed according to OECD Guideline 305 without GLP statement, to assess the bioconcentration and depuration characteristics of the source substance in the rainbow trout under dynamic flow through system. The fish were continuously exposed to 14C-test substance for 28 days. After the exposure, the fish were transferred to flowing untreated water and the depuration of radioactivity was followed for further 14 days. All validity criteria were fulfilled but restrictions should be taken into account. The BCF value was determined at 864 in the study report. This value was not expressed as normalised to a fish with 5% lipid content, therefore the BCF was recalculated, at 560. In addition, this results seems to be overestimated because the BCF was based upon total radiolabelled residues without taking into account the potential of metabolisation of the test substance and the total radioactive residues in fish were not measured between 28 days (maximum accumulation in fish) and 35 days (total depuration). In conclusion, the BCF value of 560 indicated that the test substance did not bioaccumulate in the rainbow trout.

In addition, histopathological results from toxicological studies on this source substance showed liver hypertrophy consistent with the increased metabolism associated with detoxification*.

The extensive bioaccumulation assessment of the registered substance confirms that the substance is not bioaccumulable in aquatic organisms, and therefore is not vPvB.

The refined predicted BCF (determined in the medium tier with the biotransformation in vitro assay) is in the same order of magnitude as the fish in vivo BCF. Also, based on this weight of evidence approach, we can confirm that the in silico predictions are over-conservative.

* see REACH registration dossier of (±)-(3aR*,5aS*,9aS*,9bR*)-3a,6,6,9a-Tetramethyldodecahydronaphtho[2,1-b]furan (source chemical): Endpoint Repeated dose toxicity.