Phosphinylidynetrimethanol

Endpoint summary

Description of key information

In vitro / in chemico

- ARE-Nrf2 Luciferase KeratinoSens™ Test (OECD TG 442D; GLP): negative

- Human Cell Line Activation Test (h-CLAT) (OECD TG 442E; GLP): negative

- (Q)SAR prediction with the OECD QSAR Toolbox v4.2 (as surrogate for DPRA, OECD 442C): predicted mode of action provides positive indication for skin sensitisation.

In vivo

No study available; testing is not required based on the negative KeratinoSens™ and h-CLAT results (in line with the "2 out of 3" integrated testing strategy approach).

Key value for chemical safety assessment

Skin sensitisation

Link to relevant study records

Reference

Reference 1

Endpoint:

skin sensitisation: in vitro

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

Experimental phase: 06 July 2020 - 22 July 2020

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 442D (In Vitro Skin Sensitisation: ARE-Nrf2 Luciferase Test Method)

Version / remarks:

Appendix IA: In vitro Skin Sensitisation: The ARE-Nrf2 Luciferase KeratinoSens™ Test Method
(25 June 2018)

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Type of study:

other: ARE-Nrf2 luciferase KeratinoSens™ test method (migrated information)

Justification for non-LLNA method:

In vitro study should be performed first. An in vivo study shall be conducted only if in vitro/in chemico test methods described under point 8.3.1 of REACH are not applicable.

Positive control results:

The luciferase activity induction obtained with the positive control, Trans-Cinnamaldehyde, was statistically significant above the threshold of 1.5-fold in all tests.

Run / experiment:

run/experiment 1

Parameter:

other: IC30 [442D] (migrated information)

Remarks on result:

not determinable

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 1

Parameter:

other: IC50 [442D] (migrated information)

Remarks on result:

not determinable

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 2

Parameter:

other: IC30 [442D] (migrated information)

Remarks on result:

not determinable

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 2

Parameter:

other: IC50 [442D] (migrated information)

Remarks on result:

not determinable

Remarks:

no indication of skin sensitisation

For each individual test, four parallel plates were used: three replicates were used for the luciferase activity induction measurements and one was needed for theMTT cell viability assay to measure the cytotoxicity induced by the test item.

1) Test item

For the test item, twelve doses ranging from 2000.00 µM to 0.98 µM were used in both tests.

Average fold induction, significance and viability (%) values for the test item concentrations in the independent experiments are appended (cf. 'Attached background material').

The summary of the results obtained in the two independent experiments are described below.

a) First test

The test item induced no cytotoxicity (meaning there was no viability value observed below 70 %) compared to the solvent/vehicle control at any concentration in KeratinoSens™ cells. Thus, IC30 and IC50 values were not calculated.

The luciferase activity induction did not exceed the threshold of 1.5 fold compared to the respective negative control. The maximal fold induction (Imax) was 1.14 fold. Therefore, EC1.5 value could not be determined. Moreover, no dose response could be observed.

No precipitation was observed at any point during the first test.

Summary of the KeratinoSens™ results for the test item in the first test:

- Statistically significant induction over 1.5-fold: no

- Viability ≥ 70 % at lowest concentration with ≥1.5-fold: no

- EC1.5 (µg/mL): -

- Clear dose response: no

- Overall result: negative

b) Second test (repeated)

The second test when conducted for the first time, did not meet all acceptance criteria concerning the positive control due to the EC1.5 value being 6 µM which was lower than the acceptable 7 µM value. Therefore the test results were not included in the Report and the test was repeated with the same conditions.

The test item induced no cytotoxicity (meaning there was no viability value observed below 70 %) compared to the solvent/vehicle control at any concentration in KeratinoSens™ cells. Thus, IC30 and IC50 values could not be calculated.

Although the luciferase activity induction exceeded the threshold of 1.5 fold compared to the respective negative control, the EC1.5 value was over the 1000 µM threshold with the value of 1455 µM, therefore not counting as a positive result. The maximal fold induction (Imax) was 1.68 fold. An overall but not clear dose response could be observed, though.  

No precipitation was observed at any point during the second test.

Summary of the KeratinoSens™ results for the test item in the second test:

- Statistically significant induction over 1.5-fold: yes

- Viability ≥ 70 % at lowest concentration with ≥1.5-fold: yes

- EC1.5: 1455 µM (KeratinoSens™ prediction is considered positive when the the EC1.5 value is less than 1000 μM)

- Clear dose response: no

- Overall result: negative

2) Negative and positive control

The coefficient of variation (CV %) of the luminescence reading for the negative control DMSO was below 20 % in both tests (13.05 % and 13.24 % respectively).

The luciferase activity induction obtained with the positive control, Trans-Cinnamaldehyde was statistically significant above the threshold of 1.5 at several concentrations in both tests. The EC1.5 values of the positive control fell between 7 µM and 30 µM (9 µM in both valid tests).

The average inductions in the parallel plates for Trans Cinnamaldehyde at 64 μM were 83.16 fold and 141.72 fold in the first and second tests, respectively.

Although the luciferase activity induction in both tests was outside of the 2 – 8-fold induction range, there was a clear dose response relationship in the luciferase activity induction with increasing concentrations and corresponding fold induction rise, therefore these results were accepted as valid.

In any of the tests, there was no cytotoxicity (cell viability lower than 70 %) induced by the positive control at any of the tested concentrations.

Each above described individual test met the acceptance criteria for the negative and positive controls therefore considered valid under the conditions described.

Interpretation of results:

other: non-sensitiser based on the KeratinoSens™ results

Conclusions:

Results of two independent tests:
- THPO induced no cytotoxicity n KeratinoSens™ cells compared to the solvent/vehicle; thus, in none of the tests an IC30 or IC50 value could be determined.
- Both tests were concluded negative; in the first test the induction values of the test item did not exceed the 1.5-fold threshold, and in the second test EC1.5 value was over 1000 µM, thus corresponding to a negative response.
- Based on these results and the KeratinoSens™ prediction model, THPO is concluded negative for skin sensitization potential under the experimental conditions.

Executive summary:

In the course of this GLP study according to OECD 442D, the skin sensitization potential of the test item Tris(hydroxymethyl)phosphine oxide (THPO)  was studied using the KeratinoSens™ method (ARE-Nrf2 Luciferase Test Method).

In order to derive a prediction for the test item the results of two independent tests were used.

Since the results of the two tests were concordant, a third one was not needed in order to derive a conclusion. However, the second test needed to be repeated due to not fulfilling all validity criteria for the positive control.

The luciferase activity induction obtained with the positive control, Trans-Cinnamaldehyde, was statistically significant above the threshold of 1.5-fold in all tests.  

For the test item, twelve doses ranging from 2000.00 µM to 0.98 µM were used in both valid tests.

The test item induced no cytotoxicity (a substance is considered cytotoxic if the viability is below 70 %) in KeratinoSens™ cells compared to the solvent/vehicle. Thus, in none of the tests an IC30 or IC50 value could be determined.  

Both tests were concluded negative. In the first test the induction values of the test item did not exceed the 1.5-fold threshold, and in the second test EC1.5 value was over 1000 µM, thus corresponding to a negative response.  

Based on these results and the KeratinoSens™ prediction model, the test item THPO is concluded negative for skin sensitization potential under the experimental conditions of KeratinoSens™ method (ARE-Nrf2 Luciferase Test Method).  

Reference 2

Endpoint:

skin sensitisation: in vitro

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

Experimental phase: 28 July 2020 - 15 August 2020

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

other: OECD Guideline 442E (In Vitro Skin Sensitisation assays addressing the key event on activation of dendritic cells on the Adverse Outcome Pathway for skin sensitisation) (migrated information)

Version / remarks:

Annex I: In Vitro Skin Sensitisation: human Cell Line Activation Test (h-CLAT) (25 June 2018)

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Type of study:

other: human Cell Line Activation Test (h-CLAT) (migrated information)

Justification for non-LLNA method:

In vitro study should be performed first. An in vivo study shall be conducted only if in vitro/in chemico test methods described under point 8.3.1 of REACH are not applicable.

Specific details on test material used for the study:

- Molecular weight 140.07 g/mol
- Log Kow -4.77 (EPISuite KOWWIN (V1.68) estimation)

Details on the study design:

This in vitro study was performed to assess the sensitising potential of the test item THPO by quantifying changes in the expression level of the two cell surface markers CD86 and CD54, which are associated with the process of activation of monocytes and dendritic cells. For this purpose the human monocytic leukaemia cell line THP-1 was used and treated with the test item for 24 h before evaluation. The measured expression levels were used to distinguish potential skin sensitizers from non-sensitizers. If the RFI of CD86 is equal to or greater than 150% at any tested dose (>50 % of cell viability) in at least 2 independent runs, AND/OR if the RFI of CD54 is equal to or greater than 200 % at any tested dose (>50 % of cell viability) in at least 2 independent runs, the test item prediction is considered as positive. Otherwise it is considered as a negative.
In order to conclude on the skin sensitisation potential of the test substance, a human Cell Line Activation Test (h-CLAT) comprises at least two independent and valid experiments.
When the item is tested at 5000 µg/mL in saline, 1000 µg/ml in DMSO, or highest soluble dose as the maximal test concentration instead of CV75-based dose and does not meet the positive criteria above without affecting cytotoxicity at all tested doses, the test item prediction should be considered as negative.
Since EC150 and EC200 values are just optional for test items that are found to be sensitisers, the values are calculated only in cases where a firm dose response curve can be constructed.

The h-CLAT addresses the third key event of the skin sensitisation adverse outcome pathway (AOP) that was defined by the OECD in 2012 and is a part of the AOP-based “two out of three” skin sensitisation integrated testing strategy for hazard identification (Bauch et al., 2012).
A categorization in the subcategories 1 A and 1 B is not possible. However, a positive result with the h-CLAT may be used on its own to classify a chemical into UN GHS category 1. For a confirmation of a negative result another in vitro skin sensitisation test has to be performed.

Run / experiment:

run/experiment 1

Parameter:

other: RFI CD86

Vehicle controls validity:

valid

Negative controls validity:

valid

Positive controls validity:

valid

Remarks on result:

other: RFI was lower than 150 % at all tested doses

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 1

Parameter:

other: RFI CD54

Vehicle controls validity:

valid

Negative controls validity:

valid

Positive controls validity:

valid

Remarks on result:

other: RFI was lower than 200 % at all tested doses

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 1

Parameter:

other: EC150, CD86 [442E] (migrated information)

Remarks on result:

not determinable

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 1

Parameter:

other: EC200, CD54 [442E] (migrated information)

Remarks on result:

not determinable

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 2

Parameter:

other: RFI CD86

Vehicle controls validity:

valid

Negative controls validity:

valid

Positive controls validity:

valid

Remarks on result:

other: RFI was lower than 150 % at all tested doses

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 2

Parameter:

other: RFI CD54

Vehicle controls validity:

valid

Negative controls validity:

valid

Positive controls validity:

valid

Remarks on result:

other: RFI was lower than 200 % at all tested concentrations

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 2

Parameter:

other: EC150, CD86 [442E] (migrated information)

Remarks on result:

not determinable

Remarks:

no indication of skin sensitisation

Run / experiment:

run/experiment 2

Parameter:

other: EC200, CD54 [442E] (migrated information)

Remarks on result:

not determinable

Remarks:

no indication of skin sensitisation

1) Performance checks for cells (doubling time and reactivity check)

Before the preliminary tests, the above mentioned performance checks (doubling time and reactivity check) were conducted for the cells.

The THP-1 (ATCC) master cell culture passed the reactivity check and concluded to be competent for the dose finding and main tests.

2) Preliminary tests (Dose finding assays)

Two independent runs for the dose finding assay were performed to determine the test item concentration that results in 75 % cell viability (CV75) compared to the solvent/vehicle control chosen in the trial formulation.

In the first run the highest final test item concentration on the plate was 1453.0 µg/mL and a 2 fold dilution was used when preparing the master solutions. Since no cytotoxicity was observed, in the second preliminary run the maximal allowed concentration corrected for the main constituent (HAC) was used which corresponded to the following highest final concentration on the plate 7208.0 µg/mL and a 2-fold dilution was used.

All the data generated in the preliminary test are presented in the table below: 

Dose finding test results
Name of the test item:		Tris(hydroxymethyl)phosphine oxide (THPO)
Run	Test dose (µg/mL)		11.4	22.7	45.4	90.8	181.6	363.3	726.5	1453.0
1	Viability (%)		97.71	97.75	98.15	98.06	97.82	97.66	97.75	97.97
	Test dose (µg/mL)		56.3	112.6	225.3	450.5	901.0	1802.0	3604.0	7208.0
2	Viability (%)		97.77	98.23	97.67	97.62	97.86	97.51	97.95	97.48

Since no cytotoxicity was observed, no CV75 value was determined but the corrected highest allowed concentration (HAC) was used for setting the dose range for measuring CD86 and CD54 expression in the main tests.

Eight final concentrations (µg/mL) were used for the test item tested in the main tests; these are (nominal concentrations):

             HAC (7200.0 μg/mL);

1/1.2   × HAC (6000.0 μg/mL);

1/1.2² × HAC (5000.0 μg/mL);

1/1.2³ × HAC (4166.7 μg/mL);

1/1.2⁴ × HAC (3472.2 μg/mL);

1/1.2⁵ × HAC (2893.5 μg/mL);

1/1.2⁶ × HAC (2411.3 μg/mL);

and 1/1.2⁷ × HAC (2009.4 μg/mL).

3) Main tests (CD86 and CD54 expression)

The CD86/54 expression was measured right after the dose finding assays, using the same batch of THP-1 cells.

For CD86/CD54 expression measurement, the test item was tested in two independent runs to derive a single prediction (positive/negative). Each independent run was performed on a different day and from a different batch/flask of cells. Both runs met the acceptance criteria therefore considered valid and these runs were used in the prediction model.

The relative fluorescence intensity (RFI) was calculated for the test item, positive and negative controls for each concentration in each run for both surface markers, using the geometric mean fluorescence intensities.

a) Negative and positive control

The positive control gave expected results for both markers, meaning that the RFI values of both CD86 and CD54 expression was over the positive criteria (CD86 ≥ 150 % and CD54 ≥ 200 %) and the respective cell viabilities were more than 50 % in each valid run.

The DMSO controls had negative outcomes compared to the medium control for both markers in all valid runs, meaning that the RFI values of CD86 and CD54 marker expression was never over the positive criteria. The cell viabilities of medium and DMSO controls were higher than 90 % in all independent, valid runs (taken cell viabilities of the IgG1 isotypic control).

For medium and DMSO controls, the MFI ratio of both CD86 and CD54 to isotype control was over 105 % in both valid runs.

	sample	concentration	RFI		viability (%) ‑ IgG1
			CD86	CD54	IgG
Exposure date:   14 August 2020	Medium	-	100	100	97.61
	DMSO	0.2%	101	99	97.79
	DNCB	4.6 μg/mL	546	489	89.74
Exposure date:   15 August 2020	Medium	-	100	100	97.53
	DMSO	0.2 %	100	102	97.67
	DNCB	4.4 μg/mL	434	322	91.77

b) Test item

The increase in CD86 marker expression (RFI) was lower than 150 % at all tested doses (with > 50 % of cell viability) compared to the respective negative controls in each independent valid run. Based on the two valid negative results, CD86 marker expression was concluded to be concordantly negative. Therefore, Effective concentration for CD86 expression (EC150) was not determined.

The increase in CD54 marker expression (RFI) was lower than 200 % at all tested concentrations (with > 50 % of cell viability) compared to the respective negative controls in both independent valid runs. Based on the concordant results of the two individual runs for CD54 expression, prediction was concluded as negative. Therefore, Effective concentration for CD54 expression (EC200) was not determined.

Since both markers (CD86 and CD54) gave negative results in both independent valid runs, the overall outcome of the study was negative.

The geometric mean of the fluorescent intensities (MFI) for all the concentrations of the test item tested in each run and the relative fluorescent intensities for CD86 and CD54 and cell viabilities are appended ("Attached background material").

c) Optional EC150 and EC200 values for sensitisers

Since the study was concluded to be negative, no effective concentrations (EC150 and EC200) were determined.

Interpretation of results:

other: non-sensitiser based on the h-CLAT results

Conclusions:

Results of two independent tests:
- Obtained CV75 value: highest allowed concentration [7200.0 µg/mL; test item, taking into account the content of the main constituent THPO]
- h-CLAT result for CD86: negative
- h-CLAT result for CD54: negative
- h-CLAT result obtained: non-sensitiser

Executive summary:

In the course of this GLP study according to OECD 442E, the skin sensitisation potential of Tris(hydroxymethyl)phosphine oxide (THPO) was examined.

The extent of cytotoxicity induced on THP-1 cells by the test item was studied in two dose finding tests. Since no cytotoxicity was observed, no CV75 value was determined. Since the tested concentrations during the study were corrected by a correction factor (1.44) for the test substance, based on the results of the two dose finding tests, eight doses between 7200.0 µg/mL – 2009.4 µg/mL (nominal concentrations of the test item due to recalculation to the main constituent, THPO) were used for the main test in two independent runs and both runs were concluded valid.  

The increase in CD86 marker expression (RFI) was lower than 150 % at all tested doses (with > 50 % of cell viability) compared to the respective negative controls in all valid runs. Based on the two valid negative results, CD86 marker expression was concluded to be negative.

Therefore, effective concentration for CD86 expression (EC150) was not determined.

The increase in CD54 marker expression (RFI) was lower than 200 % at all tested doses (with > 50 % of cell viability) compared to the respective negative controls in all valid runs. Based on the two valid negative results, CD54 marker expression was concluded to be negative.

Therefore, effective concentration for CD54 expression (EC200) was not determined.

Since the CD86 and CD54 markers gave negative results in all independent runs, the overall h-CLAT prediction was concluded to be negative, as well.

Based on these results and the h-CLAT prediction model, the test item THPO is concluded negative for skin sensitisation potential under the experimental conditions of human Cell Line Activation Test.  

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (not sensitising)

Additional information:

Based on the OECD 'Guidance Document on the Reporting of Defined Approaches and Individual Information Sources to be Used Within Integrated Approaches to Testing and Assessment (IATA) for Skin Sensitisation' Series on Testing & Assessment No. 256 (ENV/JM/MONO(2016)29 and ENV/JM/MONO(2016)29/ANN1 and ENV/JM/MONO(2016)29/ANN2) an Adverse Outcome Pathway-based "2 out of 3" integrated testing strategy approach was applied to determine the skin sensitization potential of the substance Tris(hydroxymethyl)phosphine oxide (THPO).

This defined approach describes an integrated testing strategy (ITS) for the identification of the skin sensitisation hazard of a substance primarily for the purposes of classification and labelling without the use of animal testing. The three in-vitro assays used in the “2 out of 3” defined approach target different key events of the skin sensitization adverse outcome pathway (AOP): key event #1 is the assessment of protein binding in chemico (OECD 442 C: DPRA), key event #2, the trigger of an antioxidant/inflammatory response in keratinocytes (OECD 442D: KeratinoSens™) and key event #3, the activation of dendritic cells (OECD 442E: h-CLAT).

The prediction model entails that two concordant results obtained from methods addressing different steps of first three key events of the AOP, determine the final classification.

In silico prediction was performed with the OECD QSAR Toolbox v4.2 as surrogate for the DPRA (key event #1); endpoint specific profilers (Protein binding alerts for skin sensitization by OASIS v1.6 (2017), Protein binding alerts for skin sensitization according to GHS v1.1 (2017)) and autoxidation simulation was applied to assess the potential of protein-binding, a crucial mode of action to induce skin sensitisation. The (Q)SAR prediction is given a Klimisch score of 2 (reliable with restrictions) as the results are derived from a valid (Q)SAR model with adequate and reliable documentation/justification, but not (completely) falling into the applicability domain.

Based on the results of the autoxidation simulation, phosphates and phosphonates are formed by autoxidation. These structures are associated with protein binding by nucleophilic substitution at the sp3 carbon atom. This predicted mode of action provides positive indication for skin sensitisation.

The (Q)SAR prediction suggests the potential for skin sensitization, however, it is not consistent with the in vitro results of key event #2 and #3:

In the KeratinoSens™ assay (ARE-Nrf2 Luciferase Test Method, key event #2), a negative result was obtained in two independent tests. In the first test the induction values of the test item did not exceed the 1.5-fold threshold, and in the second test EC1.5 value was over 1000 µM, thus corresponding to a negative response. Since the results of the two tests were concordant, a third one was not needed in order to derive a conclusion. The study was assigned a Klimisch score of 1.

Additionally, in the human cell line activate test (h-CLAT, key event #3), the overall h-CLAT prediction was concluded to be negative since the CD86 and CD54 markers gave negative results in all independent runs:

- The increase in CD86 marker expression (RFI) was lower than 150 % at all tested doses (with > 50 % of cell viability) compared to the respective negative controls in all valid runs;

- The increase in CD54 marker expression (RFI) was lower than 200 % at all tested doses (with > 50 % of cell viability) compared to the respective negative controls in all valid runs.

The study was given a Klimisch score of 1.

In conclusion, in line with the "2 out of 3" integrated testing strategy approach, the substance THPO is a non-sensitiser and does not need to be classified.

Respiratory sensitisation

Endpoint conclusion

Endpoint conclusion:

no study available

Justification for classification or non-classification

Tris(hydroxymethyl)phosphine oxide (THPO) was evaluated for its skin sensitization potential using the "2 out of 3" integrated testing strategy approach.

Based on the negative KeratinoSens™ and h-CLAT results, it is concluded that THPO does not need to be classified according to Regulation (EC) No 1272/2008.