Reaction products of Dihydro-3-(tetrapropenyl)furan-2,5 dione with propane-1,2,diol

Partition coefficient

Link to relevant study record(s)

Reference

Endpoint:

partition coefficient

Type of information:

experimental study

Adequacy of study:

key study

Study period:

15 January 2016 to 07 December 2017

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 117 (Partition Coefficient (n-octanol / water), HPLC Method)

Deviations:

yes

Remarks:

The results from the two tests do not agree within ± 0.1 log units as specified in the OECD 117 guideline; however it is felt that these results are acceptable based on the chromatography obtained and the nature of the test material.

GLP compliance:

yes (incl. QA statement)

Type of method:

HPLC method

Partition coefficient type:

octanol-water

Analytical method:

high-performance liquid chromatography

Key result

Type:

log Pow

Partition coefficient:

1.15

Temp.:

25 °C

pH:

7

Key result

Type:

log Pow

Partition coefficient:

1.99

Temp.:

25 °C

pH:

7

Key result

Type:

log Pow

Partition coefficient:

2.36

Temp.:

25 °C

pH:

7

Key result

Type:

log Pow

Partition coefficient:

4.85

Remarks on result:

other: The acid: KOWWIN model

Key result

Type:

log Pow

Partition coefficient:

5.167

Remarks on result:

other: The monoester: KOWWIN model

Key result

Type:

log Pow

Partition coefficient:

10.936

Remarks on result:

other: The diester: KOWWIN model

Details on results:

Distribution Coefficient:
- The test material gave three peaks. The mean log Pow values from the two runs were 1.15, 1.99 and 2.36.
- The results from the two tests do not agree within ± 0.1 log units as specified in the OECD 117 guideline; however it is felt that these results are acceptable based on the chromatography obtained and the nature of the test material.
- Calculated Values: Calculated values for partition coefficient were estimated using the KOWWIN model:
The acid: Hydrophobic, predicted partition coefficient (log Kow) 4.8504.
The monoester: Hydrophobic, predicted partition coefficient (log Kow) 5.1671.
The diester: Partition coefficient (Kow) is difficult to measure experimentally, predicted partition coefficient (log Kow) 10.9361.

Analytical Procedure

- Due to the issues with the validation for the analytical method for water solubility it was decided, in consultation with the Sponsor, that the partition coefficient would only be done at one pH level. As the pH level selected for the water solubility test was pH 7 it was decided that this pH would also be used for the partition coefficient test. Based on this, method validation was only carried out at pH 7.

- Four validation batches were analysed as follows:

Log P Val 1: Method validation for pH 7 buffer saturated with n-octanol.

Log P Val 2: Method validation for n-octanol saturated with pH 7 buffer.

Log P Val 3: Method validation for pH 7 buffer saturated with n-octanol and n-octanol saturated with pH 7 buffer using re-dilutions of samples used for the previous two validation batches.

Log P Val 4: Method validation for pH 7 buffer saturated with n-octanol and n-octanol saturated with pH 7 buffer using freshly prepared samples.

- A successful validation could not be achieved with percent recovery and percent relative standard deviations (%RSD) falling outside the acceptance criteria at different points in each batch. As a result, the analytical method was not used and the partition coefficient was carried out using the OECD guideline 117 HPLC method which method that does not require a validated method.

- A summary of the validation results is given below but because the validation was not successful, full results of each validation batch have not been reported.

- The method was assessed against the criteria detailed in the following sections:

Specificity:

Diluting solvent and blank control solutions were generally free of any components that interfered with the analysis of the test material. However, there were a few instances where the responses were > 30 % of the responses for the limit of quantification samples as follows:

First validation (Log P Val 1) for the  pH 7 buffer saturated with n-octanol control sample had a response of 47 % of the lowest LOQ sample response for the diester.

Second validation (Log P Val 3) for the pH 7 buffer saturated with n-octanol control sample had a response of 40 % of the lowest LOQ sample response for the diester.

Third Validation (Log P Val 4) for pH 7 buffer saturated with n-octanol control sample had a response of 40 % of the lowest LOQ sample response for the acid.

Based on this it cannot be claimed that the analytical method is totally specific for the test material. However, as the analysis was done using a LC-MS/MS TOF instrument with specific ions being monitored it is more likely that the peaks in the controls were due to carryover of the test material rather than interferences from other substances.

Sensitivity:

The lowest calibration standard was 0.001 µg/mL (0.00037 µg/mL as the acid, 0.000190 µg/mL as the monoester and 0.000439 µg/mL as the diester which demonstrated sufficient sensitivity.

The limit of quantification (LOQ) was defined as the lowest validation level that gave acceptable mean recoveries and percent relative standard deviation (%RSD).

Matrix	LOQ for the acid (µg/mL)	LOQ for the monoester (µg/mL)	LOQ for the diester (µg/mL)
pH 7 buffer saturated with n-octanol	Could not be defined as none of the validation batches passed for this constituent	0.190	0.439
n-octanol saturated with pH 7 buffer	979.81 (mean concentration of 5 samples)	501.79 (mean concentration of 5 samples)	1121.56* (mean concentration of 5 samples)

Linearity:

Response of the LC-MS/MS TOF system to the test material was linear with a 1/X weighting, over the range 0.001 to 0.1 μg/mL (0.000371 to 0.0371 µg/mL as the acid, 0.000190 to 0.0190 µg/mL as the monoester and 0.000439 to 0.0439 µg/mL as the diester, with correlation coefficients of > 0.99. However it should be noted that for three of the validation batches (Log P Val 1, Log P Val 2 and Log P Val 3)  the calibration failed for the diester (correlation coefficients of 0.91076, 0.90273 and 0.96639, respectively).

Accuracy (Recovery):

The accuracy mean recovery values varied between batches with no one batch showing acceptable mean recoveries at all validation levels for all three constituents.

The summary of the acceptability of the accuracy mean recovery results is as follows:

Matrix	Validation batch	Constituent	Acceptable mean recoveries
			Low level	Mid-level	High level
pH 7 buffer saturated with n-octanol	Log P Val 1	Acid	No	No	No
		Monoester	No	No	Yes
		Diester	No*	No*	No*
	Log P Val 3	Acid	No	No	No
		Monoester	Yes	No	Yes
		Diester	No*	No*	Yes*
	Log P Val 4	Aid	No	Yes	No
		Monoester	Yes	Yes	No
		Diester	Yes	Yes	Yes
n-Octanol saturated with pH 7 buffer	Log P Val 2	Acid	Yes	No	No
		Monoester	Yes	No	No
		Diester	Yes*	Yes*	Yes*
	Log P Val 3	Acid	Yes	No	Yes
		Monoester	No	No	No
		Diester	No*	No*	No*
	Log P Val 4	Acid	Yes	Yes	Yes
		Monoester	No	No	Yes
		Diester	Yes	Yes	Yes

 

Precision (Repeatability):

The precision %RSD values varied between batches with no one batch showing acceptable recoveries at all validation levels for all three constituents.

The summary of the acceptability of the precision %RSD results is as follows:

 

Matrix	Validation batch	Constituent	Acceptable %RSD
			Low level	Mid-level	High level
pH 7 buffer saturated with n-octanol
	Log P Val 1	Acid	Yes	Yes	Yes
		Monoester	Yes	Yes	Yes
		Diester	No*	No*	No*
	Log P Val 3	Acid	Yes	Yes	Yes
		Monoester	Yes	Yes	Yes
		Diester	No*	No*	Yes*
	Log P Val 4	Acid	No	No	No
		Monoester	Yes	No	No
		Diester	Yes	Yes	Yes
n-Octanol saturated with pH 7 buffer	Log P Val 2	Acid	Yes	Yes	Yes
		Monoester	Yes	Yes	Yes
		Diester	No*	Yes*	No*
	Log P Val 3	Acid	Yes	Yes	Yes
		Monoester	Yes	Yes	Yes
		Diester	Yes*	Yes*	Yes*
	Log P Val 4	Acid	Yes	Yes	Yes
		Monoester	Yes	Yes	Yes
		Diester	Yes	Yes	Yes

*The calibration lines for the diester failed (correlation coefficients < 0.99) for batches Log P Val 1, Log P Val 2 and Log P Val 3.

 

- Due to the poor validation results for this method it was considered not possible to successfully validate the analytical method for partition coefficient. This meant that the shake flask and slow stir methods for partition coefficient could not be used. As a consequence so, the OECD 117 HPLC method was used instead.

Interpreting Results

The KOWWIN model uses a "fragment constant" method to predict the partition coefficient (Kow usually reported as the logarithm log Kow). In the "fragment constant" method, a molecule is divided into fragments (atoms or larger functional groups) and the assigned coefficient values for each fragment are added to give the Kow estimate, which is reported as log Kow.

The predicted partition coefficients are interpreted as follows:

 

For log Kow< 1 – the substance is classified as highly soluble in water (hydrophilic).

For log Kow> 4 – the substance is classified as not very soluble in water (hydrophobic).

For log Kow> 8 – the substance is classified as not readily bioavailable.

For log Kow> 10 – the substance is classified as not bioavailable – partitioning properties are difficult to measure experimentally.

The EPI suite has no statistical test that can be used to judge the reliability of the predicted values. The KOWWIN model is a fragment approach so a surrogate method to judge the reliability of the predictions is to check that all the substructures in the compound of interest are represented in the calculations. If all the substructures are in the model domain, and are accounted for, then this increases the confidence in the predictions and the model result may be assumed to provide relevant information.

Method

Stock solutions were prepared by weighing out the standards and dissolving in methanol as follows:

Reference Substance	Weight of Reference Substance Used (mg)	Final Volume of Solution in Methanol (mL)	Final Concentration, Corrected for Purity (µg/mL)
Formamide	212.78	10	21129.05
Acetanilide	41.45	10	4136.71
Benzonitrile	70.21	10	7018.19
Methyl benzoate	53.88	10	5382.61
Ethyl benzoate	59.57	10	5951.04
Bromobenzene	69.63	10	6949.07
Naphthalene	17.61	10	1759.24
Phenanthrene	13.90	10	1363.59

 

These solutions were then diluted further to give working solutions:

Reference Substance	Concentration of Stock Solution (µg/mL)	Volume of Stock Solution Used (mL)	Final Volume in Mobile Phase Diluent (mL)	Final Concentration, Corrected for Purity (µg/mL)
Formamide	21129.05	0.10	10	211.79
Acetanilide	4136.71	0.05	10	20.68
Benzonitrile	7018.19	0.05	10	35.09
Methyl benzoate	5382.61	0.05	10	26.91
Ethyl benzoate	5951.04	0.10	10	59.51
Bromobenzene	6949.07	0.05	10	34.75
Naphthalene	1759.24	0.10	10	17.59
Phenanthrene	1363.59	0.25	10	34.09

 

The amounts of the reference substance and test substance injected on the HPLC-UV system were as follows:

 

Reference Substance	Concentration of Working Solution used (µg/mL)	Volume of Solution Injected (µL)	Amount of Reference Substance Injected (µg)
Formamide	211.79	10	2.11
Acetanilide	20.68	10	0.21
Benzonitrile	35.09	10	0.35
Methyl benzoate	26.91	10	0.27
Ethyl benzoate	59.51	10	0.60
Bromobenzene	34.75	10	0.35
Naphthalene	17.59	10	0.18
Phenanthrene	34.09	10	0.34
Test material	394.20	10	3.942

 

Conclusions:

Under the conditions of this study, the test material gave three peaks. The mean log Pow values from the two runs were 1.15, 1.99 and 2.36.
Calculated values for partition coefficient were estimated using the KOWWIN model:
The acid: Hydrophobic, predicted partition coefficient (log Kow) 4.8504.
The monoester: Hydrophobic, predicted partition coefficient (log Kow) 5.1671.
The diester: Partition coefficient (Kow) is difficult to measure experimentally, predicted partition coefficient (log Kow) 10.9361.

Executive summary:

The partition coefficient of the test material was investigated in accordance with the standardised guideline OECD 117, under GLP conditions.

Due to issues with getting acceptable validation results for the analytical procedure, in aqueous pH buffers saturated with n-octanol and n-octanol saturated with aqueous pH 7 buffers, a suitable validated analytical method could not be obtained. As a result, the partition coefficient could not be carried out using the shake flask method so testing was done following the HPLC method as specified in OECD guideline 117. Testing was done at pH 7 only.

Under the conditions of this study, the test material gave three peaks. The mean log Pow values from the two runs were 1.15, 1.99 and 2.36. The results from the two tests do not agree within ± 0.1 log units as specified in the OECD 117 guideline; however it is felt that these results are acceptable based on the chromatography obtained and the nature of the test material.

Calculated values for partition coefficient were estimated using the KOWWIN model:

The acid: Hydrophobic, predicted partition coefficient (log Kow) 4.8504.

The monoester: Hydrophobic, predicted partition coefficient (log Kow) 5.1671.

The diester: Partition coefficient (Kow) is difficult to measure experimentally, predicted partition coefficient (log Kow) 10.9361.

Description of key information

Under the conditions of this study, the test material gave three peaks. The mean log Pow values from the two runs were 1.15, 1.99 and 2.36.

For the chemical safety assessment, a weighted average log Pow of 1.84 (at 25 °C) was derived, based on the constituents’ experimentally derived log Pow values and their proportion (%) in the substance (see table below).

Constituent	Percentage in substance	Log Pow	Log Pow component
The acid	37.1	1.15	0.43
The monoester	19	1.99	0.38
The diester	43.9	2.36	1.04
Weighted average Log Pow Component			1.84

 

Key value for chemical safety assessment

Log Kow (Log Pow):

1.84

at the temperature of:

25 °C

Additional information

The partition coefficient of the test material was investigated in accordance with the standardised guideline OECD 117, under GLP conditions. The study was awarded a reliability score of 1 in accordance with the criteria set forth by Klimisch et al. (1997).

Due to issues with getting acceptable validation results for the analytical procedure, in aqueous pH buffers saturated with n-octanol and n-octanol saturated with aqueous pH 7 buffers, a suitable validated analytical method could not be obtained. As a result, the partition coefficient could not be carried out using the shake flask method so testing was done following the HPLC method as specified in OECD guideline 117. Testing was done at pH 7 only.

Under the conditions of this study, the test material gave three peaks. The mean log Pow values from the two runs were 1.15, 1.99 and 2.36. The results from the two tests do not agree within ± 0.1 log units as specified in the OECD 117 guideline; however it is felt that these results are acceptable based on the chromatography obtained and the nature of the test material.