Disodium 2-oxoglutarate

Skin sensitisation

Administrative data

Endpoint:

skin sensitisation: in vitro

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

21 March - 20 April 2017

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Data source

Reference

Reference Type:

study report

Title:

Unnamed

Year:

2017

Report date:

2017

Materials and methods

Test guideline

Qualifier:

according to guideline

Guideline:

OECD Guideline 442C (In Chemico Skin Sensitisation: Direct Peptide Reactivity Assay (DPRA))

Version / remarks:

2015

Deviations:

yes

Remarks:

The guideline states that the HPLC run sequence should be set up in order to keep the HPLC analysis time less than 30 hours. For the Cysteine samples reported, the time between sample preparation and the injection of the last sample was 35 h and 15 min.

GLP compliance:

yes (incl. QA statement)

Type of study:

direct peptide reactivity assay (DPRA)

Test material

Test material information

Constituent 1

Reference substance name:

Disodium 2-oxoglutarate

EC Number:

206-167-8

EC Name:

Disodium 2-oxoglutarate

Cas Number:

305-72-6

Molecular formula:

C5H6O5.2Na

IUPAC Name:

disodium 2-oxopentanedioate

impurity 1

Reference substance name:

Water

EC Number:

231-791-2

EC Name:

Water

Cas Number:

7732-18-5

Molecular formula:

H2O

IUPAC Name:

Oxidane

Test material form:

solid: crystalline

Remarks:

Crystalline powder

Details on test material:

Batch No: 14152300
Storage: 15 – 25 °C

In vitro test system

Details on the study design:

SUMMARY: DPRA measures the reaction of the test item with synthetic peptides containing cysteine (Ac-RFAACAA-COOH) or lysine (Ac-RFAAKAA-COOH). The custom peptides contained cysteine or lysine as the nucleophilic reaction centres and phenylalanine to facilitate HPLC detection. Test item and peptide were combined and incubated together for 24 h at room temperature. Following this incubation, the concentration of free (i.e. unreacted) peptide remaining was measured by HPLC immediately prior to the lysine peptide assay.

EXPERIMENTAL PROCEDURES

PEPTIDES:
Source: RS Synthesis
Batch:
- Cysteine: No. P161108-LC180433
- Lysine: No. P161108-LC107617
Purity:
-Cys: 95.2%
-Lys: 98.14%

BUFFERS USED:
- Phosphate buffer: ca 100 mM, pH 7.53
- Ammonium acetate buffer: ca 100 mM, pH 10.2

SOLUBILITY ASSESSMENT:
- ultrapure water was selected as the most suitable solvent for the test material

PREPARATION PEPTIDE STOCK SOLUTIONS:
- CYSTEINE: stock solution of 0.501 mg/mL (0.667 mM) in phosphate buffer
- LYSINE: stock solution of 0.518 mg/mL (0.667 mM) in ammonium acetate buffer

CYSTEINE PEPTIDE ASSAY:
-PREPARATION: test item was dissolved in ultrapure water and mixed by inversion and vortex until fully in solution. The concentration of the test solution corrected for purity, was 19.1 mg/mL (99.8 mM, 99.8% from the target). Cinnamic aldehyde was dissolved in acetonitrile with a concentration of 13.22 mg/mL (100 mM). All test item and control solutions were prepared immediately prior to use.
-PREPARATION OF THE SANDARD CURVE: Dilution buffer was prepared by mixing phosphate buffer (pH 7.5, 8 mL) with acetonitrile (2 mL). Standard 1 (STD1) was prepared by mixing peptide stock solution (1600 µL) with acetonitrile (400 µL). Serial dilutions (1:1, v/v) were prepared from this to make a standard curve (from 0.534 to 0.0167 mM). An additional sample containing only dilution buffer was included as a blank (0 mM) standard. The standard curve was analysed by HPLC immediately prior to the cysteine peptide assay.
-REFERENCE CONTROL: Acetonitrile (250 µL) was mixed with peptide stock solution (750 µL). Three replicates of this were produced for Reference Control A. Reference Control B was prepared as described for Reference Control A. Three replicates were analyzed at the beginning of the testing run, and three at the end of the testing run, to demonstrate peptide stability over the analysis time. Reference Control C samples were prepared containing the solvent that the test item was dissolved in: three replicates containing acetonitrile (250 µL) and peptide stock (750 µL) and three replicates containing ultrapure water (50 µL), acetonitrile (200 µL) and peptide stock (750 µL). These samples were included in every assay run together with the samples containing test item. They are used to verify that the solvent does not impact upon peptide stability during the assay, and to calculate percentage peptide depletion.
- PEPTIDE ASSAY METHOD: The assay contained a 1:10 molar ratio of peptide to test item. Positive control or test item (50 µL) was mixed with acetonitrile (200 µL) and the peptide solution (750 µL). The vials were mixed by vortex. Co-elution controls were prepared by mixing together acetonitrile (200 µL), phosphate buffer (750 µL) and test item (50 µL). All test items and positive control samples were prepared in triplicate. All vials were stored in the dark at ambient temperature for ca 24 h until analyzed by HPLC.

LYSINE PEPTIDE ASSAY:
-PREPARATION: test item was dissolved in ultrapure water and mixed by inversion and vortex until fully in solution. The concentration of the test solution corrected for purity, was 19.1 mg/mL (99.7 mM, 99.7% from the target). Cinnamic aldehyde was dissolved in acetonitrile with concentration of 13.22 mg/mL (100 mM). All tets item and control solutions were prepared immediately prior to use.
- PREPARATION OF THE STANDARD CURVE: Dilution buffer was prepared by mixing ammonium acetate buffer (pH 10.20, 8 mL) with acetonitrile (2 mL). Standard 1 (STD1) was prepared by mixing peptide stock solution (1600 µL) with acetonitrile (400 µL). Serial dilutions (1:1, v/v) were prepared from this to make a standard curve (from 0.534 to 0.0167 mM). An additional sample containing only dilution buffer was included as a blank (0 mM) standard. The standard curve was analyzed by HPLC.
- REFERENCE CONTROL: like for cysteine
- PEPTIDE ASSAY METHOD: The assay contained a 1:50 molar ratio of peptide to test item. Cinnamic aldehyde or test item (250 µL) were mixed with peptide solution (750 µL). The vials were mixed by inversion and vortex. Co-elution controls were prepared by mixing together ammonium acetate buffer (750 µL) and test item (250 µL). All vials were stored in the dark at ambient temperature for ca 24 h until analysed by HPLC.

CHROMATOGRAPHIC AND DETECTOR PARAMETERS
- Column: Phenomenex Luna C18 (2) (2 x 100 mm, 3 µm)
- Run Time: 20 min
- Mobile Phase Conditions: Mobile Phase A: trifluoracetic acid (0.1%, v/v) in Milli-Q H2O
Mobile Phase B: trifluoracetic acid (0.085%, v/v) in acetonitrile
- Flow Rate: 0.35 mL/min
- Column Temperature: 30°C
- Auto Sampler Temperature: Room temperature
- Injection Volume: 7 µL
- UV Wavelength: 220 nm
- HPLC Gradient: see below

CALCULATIONS:
The concentration of peptide remaining in each sample following incubation was calculated from integrated peak area, with reference to the peptide standard curve. Percent peptide depletion was calculated from the following formula:
Peptide Depletion (%) = 1 – ( Peak Area (Sample) / Mean Peak Area (Reference Control C)) x 100

Results and discussion

Positive control results:

The mean depletion value for the positive control was 70.1% showing a high reactivity (Sensitizer)

In vitro / in chemico

Results

Key result

Run / experiment:

other: DPRA cysteine and lysine prediction model

Parameter:

other: %peptide depletion (mean value)

Value:

4.6

Vehicle controls validity:

valid

Negative controls validity:

not applicable

Positive controls validity:

valid

Remarks:

Cinnamic aldehyde

Remarks on result:

other:

Remarks:

Minimal reactivity (Non-Sensitizer)

Other effects / acceptance of results:

No co-elution of the test item with either peptide was observed.

SYSTEM SUITABILITY FOR THE CYSTEINE ASSAY
The calibration linearity, r2, for the cysteine standard curve was 0.9955. This met the acceptance criteria for r2 which was >0.990.

The mean peptide concentration of Reference Control A was 0.608 ± 0.003 mM (mean ± SD). The calculated peptide concentration in the Reference Control C samples was 0.599 ± 0.004 mM (acetonitrile) and 0.576 ± 0.005 mM (ultrapure water). These controls did not meet the acceptance criteria (0.5 ± 0.05 mM).

For the six Reference Control B and three Reference Control C samples in acetonitrile, the coefficient of variation (CV) was 1.7% (acceptance criteria for CV was <15%).

The mean percentage peptide depletion value of the three replicates for cinnamic aldehyde fell within the lower bound and upper bound values of 60.8% and 100.0% for cysteine, with a peptide depletion value of 84.6 ± 0.1% (mean ± SD).

Finally, the standard deviation of replicate test item samples was <14.9% for a-Ketoglutarate, di-Na (actual SD was 0.9%).

The data have been accepted despite the Reference Control A and Reference Control C samples not meeting acceptance criteria. This was due to an error in the preparation of the standard curve, leading to the top two concentrations being erroneous (lower than expected). As the top sample in the standard curve was the only sample with a higher nominal concentration than the reference controls, this could not be excluded from the calculations. The standard curve is used to calculate the concentration of peptide present in the Reference Control samples whereas peptide depletion is calculated from the peak area of the appropriate Reference Control C samples. The function of Reference Control A is to demonstrate the accuracy (or otherwise) of the standard curve, which it did. The function of the Reference Control C samples is to demonstrate peptide stability in the solvents. The peptide was shown to be stable from the consistent high (and historically acceptable) peak area values measured for all reference control samples.

Therefore, the concentration of peptide present in the standards does not have an impact on the calculation of peptide depletion, and furthermore all other acceptance criteria have been met, including a correct prediction for the positive control. The results are accepted and there is no impact on the outcome of the cysteine assay.

SYSTEM SUITABILITY FOR THE LYSINE ASSAY
The calibration linearity, r2, for the lysine standard curve was 1.0000. This met the acceptance criteria for r2 which was >0.990.

The mean peptide concentration of Reference Control A was 0.504 ± 0.001 mM (mean ± SD). The mean peptide concentration of Reference Control B analysed prior to the testing run was 0.504 ± 0.002 mM. The mean peptide concentration of Reference Control B analysed after the testing run was 0.493 ± 0.003 mM. The calculated peptide concentration in the Reference Control C samples was 0.502 ± 0.002 mM (acetonitrile), 0.499 ± 0.006 mM (ultrapure water). These samples met the acceptance criteria of 0.5 ± 0.05 mM. In addition, for the six Reference Controls B and three Reference Control C in acetonitrile, the CV was 1.1% (acceptance criteria for CV was <15%).

The mean percentage peptide depletion value of the three replicates for cinnamic aldehyde fell within the lower bound and upper bound values of 40.2% and 69.0% for lysine, with the SD <11.6%. The actual percentage peptide depletion value reported for cinnamic aldehyde was 55.6% ± 4.3% (mean ± SD). Finally, the standard deviation of replicate test item samples was <11.6% for the test item (actual SD was 1.1%).

PROTOCOL DEVIATIONS
-Prior to HPLC analysis, the samples should be inspected visually. This was not done. Samples were to be inspected for the presence of precipitate. This will only impact upon the test item samples, as all other samples were prepared according to the prescribed method, and do not have solubility concerns. Solubility of the test item was assessed previously, and not found to be an issue. Therefore, precipitates are very unlikely, and there is no impact on study integrity.
-The time between sample preparation and the last injection of a sequence should not exceed 35 h. For the Cysteine samples reported, the time between sample preparation and the injection of the last sample was 35 h and 15 min. The ECVAM protocol actually allows a far greater time window than this. In addition, the final sample was one of the replicates of Reference Control B, the purpose of which is to confirm the stability of the peptide over the analysis time. The CV of these samples met the acceptance criteria, therefore the peptide was stable over the duration of testing. There is no impact on study integrity.
- Details the acceptance criteria for the tests. In the cysteine assay these criteria were not met for the peptide concentration in Reference Control C and Reference Control A however, the data was accepted. The peptide concentration in the reference control samples was calculated from the Standard Curve. There was a suspected preparation error in the standard curve, leading to the top two concentrations being erroneous (lower than expected). Standard 2 was excluded due to being an outlier from the linear relationship. As the top sample in the standard curve was the only sample with a higher nominal concentration than the reference controls, this could not be excluded from the calculations. The standard curve is used to calculate the concentration of peptide present in the Reference Control samples whereas peptide depletion is calculated from the peak area of the appropriate Reference Control C samples. The function of Reference Control A is to demonstrate the accuracy (or otherwise) of the standard curve, which it did. The function of the Reference Control C samples is to demonstrate peptide stability in the solvents. The peptide was shown to be stable from the consistent high (and historically acceptable) peak area values measured for all reference control samples. Therefore, the concentration of peptide present in the standards does not have an impact on the calculation of peptide depletion, and furthermore all other acceptance criteria have been met, including a correct prediction for the positive control. The results are accepted and there is no impact on the outcome of the cysteine assay.

DEMONSTRATION OF TECHNICAL PROFICIENCY
Prior to use, Charles River Laboratories demonstrated technical proficiency in the DPRA test, using the panel of proficiency chemicals listed in OECD 442C (Toner, F, 2015).

Any other information on results incl. tables

Test Item	% Peptide Depletion Cysteine (Mean  ± SD)	% Peptide Depletion Lysine (Mean ± SD)	Mean of Cysteine and Lysine	DPRA Classification (Cysteine and Lysine Prediction Model)
Test Material	6.4 ±0.9	2.8 ± 1.1	4.6	Minimal Reactivity (Non-Sensitizer)
Positive control	84.6 ±0.1	55.6 ±4.3	70.1	High Reactivity (Sensitizer)

Using the cysteine and lysine prediction model (see Table below) the test material was categorised as minimally reactive and a non-sensitiser.

Mean depletion values (Cys Lys)	Mean Depletion values (cys only)	Reactivity classification	DPRA Prediction
<6.38 %	<13.89%	Minimal	Non Sensitizer
6.38 -22.62%	13.89 -23.09%	Low	Sensitizer
22.62 -42.47%	23.09%-98.24%	Moderate	Sensitizer
>42.47	>98.24%	High	Sensitizer

Applicant's summary and conclusion

Interpretation of results:

other: minimally reactive: non-sensitizer

Remarks:

Study will be used for classificatin in combination with other studies (Weight of Evidence)

Conclusions:

In conclusion, according to the DPRA cysteine and lysine prediction model alpha-Ketoglutarate, di-Na (CAS 305-72-6) was classified as minimally reactive and was, therefore, a non-sensitiser.

Executive summary:

Skin sensitisation is a type IV (delayed) hypersensitivity reaction that results from the interaction of a sensitising agent with host proteins to form an immunogenic complex.

Small molecules that can interact with proteins in this way are referred to as haptens, and are generally not immunogenic in isolation. Hapten-modified proteins are recognised as foreign by antigen presenting cells, leading to T-cell activation and localised inflammation at the site of all subsequent exposures to the hapten.

Most skin sensitising agents are electrophiles, i.e. will accept an electron pair from a nucleophile to form a covalent bond.  The amino acids cysteine and lysine are thought to be the nucleophiles most frequently modified in proteins during sensitisation, and the ability of small molecules to react with these amino acids forms the basis of the Direct Peptide Reactivity Assay (DPRA).

The objective of this study was to assess the peptide binding capability of test material using synthetic cysteine and lysine peptides and to classify the test material to one of the four reactivity classes leading to a DPRA prediction according to the following prediction model.

Mean depletion values (Cys Lys)	Mean Depletion values (cys only)	Reactivity classification	DPRA Prediction
<6.38 %	<13.89%	Minimal	Non Sensitizer
6.38 -22.62%	13.89 -23.09%	Low	Sensitizer
22.62 -42.47%	23.09%-98.24%	Moderate	Sensitizer
>42.47	>98.24%	High	Sensitizer

The reaction of the test item with synthetic peptides containing cysteine (Ac-RFAACAA-COOH) or lysine (Ac-RFAAKAA-COOH) was performed.  The custom peptides contained cysteine or lysine as the nucelophilic reaction centres and phenylalanine to facilitate detection by HPLC analysis.

For each peptide assay, the test item was prepared at a concentration of 100 mM in ultrapure water as determined from the solubility experiment conducted under Charles River Study No. 799765 (Vinall, J, 2017).  The test item and peptides were combined and incubated together for ca 24 h at room temperature. Following this incubation, the concentration of free (i.e. unreacted) peptide remaining was measured by HPLC.  From the results obtained, a reactivity class was assigned and a DPRA prediction was made according to the above criteria.

The results obtained are presented in the following table:

Test Item	% Peptide Depletion Cysteine	% Peptide Depletion Lysine	DPRA Classification
Test Material	6.4  ± 0.9	2.8  ± 1.1	Minimal Reactivity (Non-Sensitizer)
Cinnamic Aldehyde (positive control)	84.6  ± 0.1	55.6 ± 4.3	Moderate Reactivity (Sensitizer)

The acceptance criteria for the test were fulfilled, with the exception that for the cysteine test, the peptide concentration calculated in the Reference Control A and Reference Control C samples was slightly high, due to an error in the preparation of the standard curve, leading to the top two concentrations being erroneous (lower than expected).  This does not have an impact on the calculation of peptide depletion, and furthermore all other acceptance criteria have been met, including a correct prediction for the positive control, therefore the results are accepted and there is no impact on the outcome of the cysteine assay.

In conclusion, according to the DPRA cysteine and lysine prediction model, alpha-Ketoglutarate, di-Na (CAS 305-72-6) was classified as minimally reactive and was, therefore, a non-sensitiser.