2,3-dihydro-2,2,6-trimethylbenzaldehyde

Short-term toxicity to fish

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

short-term toxicity to fish

Type of information:

experimental study

Adequacy of study:

supporting study

Study period:

From May 22, 2019 to July 29, 2019

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

test procedure in accordance with generally accepted scientific standards and described in sufficient detail

Remarks:

The method has recently been adopted as a new test guideline by the OECD; TG No. 249 Fish Cell Line Acute Toxicity: The RTgill-W1 cell line assay (adopted 14 June, 2021)

Qualifier:

equivalent or similar to guideline

Guideline:

other: The method has recently been adopted as a new test guideline by the OECD - TG No. 249 Fish Cell Line Acute Toxicity: The RTgill-W1 cell line assay

Version / remarks:

Adopted 14 June 2021

Qualifier:

according to guideline

Guideline:

other: International Organization for Standardization (ISO) : Water Quality - Determination of Acute Toxicity of Water Samples and Chemicals to a Fish Gill Cell Line (RTgill-W1), ISO21115.

Deviations:

no

GLP compliance:

no

Remarks:

Testing done in internal lab

Analytical monitoring:

yes

Details on sampling:

Preparation of Test Plates with RTgill-W1 Cells
For exposure to the test chemical, confluent cells (80 - 90% confluence) were gently washed with 2 mL Versene (Supplier no. 15040, Gibco-Invitrogen). Cells were then detached with 0.7 mL trypsine solution (0.25% in phosphate buffered saline (PBS) w/o calcium and magnesium) and trypsination was stopped by adding 10 mL L15 medium supplemented with FBS. Cell clumps were dissolved by gently pipetting the solution several times up and down and the cell suspension was centrifuged for 3 min at 1000 rpm. The cell pellet was suspended in 10 mL L15 medium with FBS and counted in a Neubauer chamber. Cell density was adjusted to 350 000 cells / mL and 1 mL was seeded in each well of a 24-well plate, except for two wells without cells to correct for background fluorescence. After 24 h of attachment, the toxicity testing was initiated by adding the test chemical dilutions in exposure medium.

Vehicle:

yes

Remarks:

Dimethylsulfoxide (DMSO) at 0.5% in Liebovitz (L15) medium

Details on test solutions:

Test Plate Dosing and Incubation
The test chemical was dissolved to a final concentration of 25.6 mg/mL in Dimethylsulfoxide (DMSO).
Serial half dilutions in DMSO were prepared resulting in final concentrations of 25.6, 12.8, 6.4, 3.2, 1.6 and 0.8 mg/mL. From each DMSO solution, a dosing solution in medium was prepared by adding 50 μL to 9.95 mL of exposure medium L15/ex (final nominal test chemical concentrations: 128, 64, 32, 16, 8, 4 mg/L). Final DMSO concentration in the test medium was 0.5%.
L15 medium was removed completely from each well of the 24-well plate which had been seeded with the cells 24 h earlier. Cells were washed once with 1 mL exposure medium L15/ex and 2.5 mL dosing solution was added to each well. Solvent controls received 2.5 mL L15/ex containing 0.5% DMSO.
Directly after addition of dosing solution, 0.5 mL of exposure solution was sampled from each well for chemical analysis. The plate, now containing 2 mL exposure solution in each well, was then covered with a plastic sealing foil.
The plate was incubated for 24 h in a BINDER Model KT 115 refrigerated incubator with thermoelectric cooling, maintained at 19°C with normal air (no CO2 enrichment) and 40% ventilation rate.

Test organisms (species):

Oncorhynchus mykiss (previous name: Salmo gairdneri)

Details on test organisms:

The rainbow trout (Oncorhynchus mykiss) RTgill-W1 cell line was obtained from the Swiss center for aquatic research EAWAG (Dübendorf, Switzerland). The cells were routinely cultured in 75 cm2 cell culture flasks containing Leibovitz (L15) medium without phenol red supplemented with Glutamine (Supplier no. 21083; Gibco-Invitrogen; Basel, Switzerland), 5% foetal bovine serum (FBS) and 0.05 mg/mL Gentamicin. The cells were split each week once with a 1:1 split. The flasks are incubated in a BINDER Model KT 115 refrigerated incubator with thermoelectric cooling, maintained at 19°C with ambient air (no CO2 enrichment) and 40% ventilation rate.

Test type:

static

Water media type:

other: Liebovitz L15 medium

Limit test:

no

Total exposure duration:

24 h

Test temperature:

19°C

Nominal and measured concentrations:

One experiment was conducted at 128, 64, 32, 16, 8, 4 mg/L based on nominal concentrations (82.18, 41.05, 20.66, 10.31, 5.10, 2.49 mg/L based on mean measured concentrations). Each concentration was tested in triplicate. A solvent control (0.5% DMSO; L15/ex) in triplicate in presence
of cells was included and compared to a cell control without solvent. Additionally, one well without cells and the highest test chemical concentration and one well without cells and no test chemical were prepared as control to correct for background fluorescence for the cell viability test.

Details on test conditions:

The plate was incubated for 24 h in a BINDER Model KT 115 refrigerated incubator with thermoelectric cooling, maintained at 19°C with normal air (no CO2 enrichment) and 40% ventilation rate.

Reference substance (positive control):

yes

Remarks:

3,4 dichloroaniline (DCA)temperature

Duration:

24 h

Dose descriptor:

EC50

Effect conc.:

42.09 mg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

test mat.

Basis for effect:

other: cytotoxicity

Key result

Duration:

96 h

Dose descriptor:

LC50

Effect conc.:

16.42 mg/L

Nominal / measured:

not specified

Conc. based on:

test mat.

Basis for effect:

mortality (fish)

Remarks on result:

other: Predicted in vivo acute toxicity

Key result

Duration:

96 h

Dose descriptor:

LC50

Effect conc.:

11.7 mg/L

Nominal / measured:

not specified

Conc. based on:

test mat.

Basis for effect:

mortality (fish)

Remarks on result:

other: Predicted in vivo acute toxicity (considering regression dataset analogue information)

Details on results:

The three acceptance criteria discussed by Fischer et al. were fulfilled except for criterium. (i) The DMSO versus solvent-free control wells did not differ in raw fluorescence values by more than 10%. This criterium was just not fulfilled with 11 % reduction in control without solvent measured. Since this is only a 1% higher difference in the control without solvent measured, the test was still considered as valid. (ii) the raw fluorescence values in the cell-free control well containing the highest test chemical concentration did not vary by more than 20% from the cell-free control well containing the exposure medium only (5% variation measured, Appendix G); (iii) EC50 values of the positive control (DCA) were within the given range based on the international validation study as discussed below.
No significant (≤ 10%) dose related cytotoxicity occurred at the four lowest test concentrations (nominal concentrations 4 - 32 mg/L) at 24 h. The NOEC was thus considered to be 20.66 mg/L (nominal concentration 32 mg/L). The EC50 value based on mean measured concentration was 42.09 mg/L.

In order to predict a fish in vivo LC50 value from the in vitro EC50, the following regression equation from the study of Natsch et al. [2018] was used:

Log LC50 (fish in vivo) = (0.97 × log EC50 PB mean measured) - 0.36
n = 37, R2 = 0.94

The predicted LC50 value for fish was 16.42 mg/L.

The structural analogue (Shisolia) was part of the dataset used to generate the above regression equation [Natsch, A., et al., 2018]. In a leave-one-out analysis, performed by the authors to estimate the predictivity of the assay when challenged with a new chemical, the LC50 of Shisolia was found to be over-predicted by 1.4 (predicted LC50 value =13 mg/L, measured LC50 value = 9.3mg/L). Applying this correction factor to the predicted LC50 value for Safranal P, gives a corrected value of 11.7 mg/L.Based on the regression equation, an in vivo LC50 of 16.42 mg/L is predicted.

Results with reference substance (positive control):

The EC50 for DCA determined in two independent experiments was 38.55 and 49.74 mg/L, respectively, which is fulfilling the acceptance criteria for this positive control. Based on the results of the international validation study, the acceptance range for DCA was set as 43.6 mg/L +/- 6.1 mg/L for
metabolic activity (2.5 SD range: 28.4-58.9 mg/L).

Reported statistics and error estimates:

In order to predict a fish in vivo LC50 value from the in vitro EC50, a detailed benchmarking study on fragrance chemicals was performed (Summary appended to the study report). This led to a predictive regression equation based on the EC50 determined with PrestoBlue® (PB):
Log LC50fish in vivo = (0.97 × log EC50PB mean measured) - 0.36 (n=37, R2=0.94)
This regression equation was used to predict an in vivo LC50.

Sublethal observations / clinical signs:

The EC50 value based on mean measured concentrations in the RTgill-W1 cell assay was 42.09 mg/L. Based on a regression equation developed with 37 fragrance molecules an in vivo LC50 of 16.42 mg/L is predicted.

The median of the under-/over-prediction (fold difference between LC50 predicted by equation 1 in Appendix F and measured in vivo LC50) was 1.5-fold for the series of 37 fragrance substances tested [7] which is considered to be well within the variation in LC50 if a chemical is tested in different fish species. Furthermore, a factor of 1.5-fold is considered to be also well within the end-point variation for intra- and inter-laboratory testing using the same species. Detailed cell viability data (% viability as compared to solvent control using the PrestoBlue® assay).

Nominal conc.	128 mg/L	64 mg/L	32 mg/L	16 mg/L	8 mg/L	4 mg/L
Mean measured conc.	82.2 mg/L	41.1 mg/L	20.7 mg/L	10.3 mg/L	5.1 mg/L	2.5 mg/L
Rep 1 (% viability)	10.4	55.0	102.4	101.0	104.5	108.3
Rep 2 (% viability)	11.2	56.0	96.2	102.2	100.4	114.7
Rep 3 (% viability)	7.7	41.9	92.5	97.3	103.0	82.0
Average (% viability)	9.8	51.0	97.1	100.2	102.6	101.7
Standard deviation	1.8	7.9	5.0	2.6	2.1	17.3
Coefficient of variation (%)	18.8	15.5	5.1	2.5	2.0	17.0

Effect concentrations, LOEC and NOEC of Safranal P in the RTgill-W1 cell line.

	Effect concentrations expressed as mean measured concentration (mg/L)
	Cytotoxicity 24 h (mg/L)	Predicted in vivo acute toxicity (LC50) (mg/L)
EC50	42.09 (38.47 - 46.04)1 mg/L	16.42
LOEC	41.05 mg/L	n.a.^2
NOEC	20.66 mg/L	n.a.

1Values in brackets are 95% confidence limits

2n.a. not applicable

Validity criteria fulfilled:

yes

Conclusions:

The EC50 value based on mean measured concentrations in the RTgill-W1 cell assay was 42.09 mg/L. Based on a regression equation developed with 37 fragrance molecules an in vivo LC50 of 16.42 mg/L is predicted.
The median of the under-/over-prediction (fold difference between LC50 predicted by the regression equtaion and measured in vivo LC50) was 1.5-fold for the series of 37 fragrance substances tested which is considered to be well within the variation in LC50 if a chemical is tested in different fish species. Furthermore, a factor of 1.5-fold is considered to be also well within the end-point variation for intra- and inter-laboratory testing using the same species.

Effect concentrations expressed as mean measured concentration (mg/L)
Cytotoxicity 24 h (mg/L) "Predicted in vivo acute toxicity (LC50) (mg/L)"
EC50 42.09 (38.47 - 46.04)1 mg/L 16.42
LOEC 41.05 mg/L n.a.2
NOEC 20.66 mg/L n.a.

1Values in brackets are 95% confidence limits
2n.a. not applicable

The structural analogue (Shisolia) was part of the dataset used to generate the regression equation [Natsch, A., et al., 2018]. In a leave-one-out analysis, performed by the authors to estimate the predictivity of the assay when challenged with a new chemical, the LC50 of Shisolia was found to be over-predicted by 1.4 (predicted LC50 value =13 mg/L, measured LC50 value = 9.3mg/L). Applying this correction factor to the predicted LC50 value for Safranal P, gives a corrected value of 11.7 mg/L.

Executive summary:

The objective of the study was to determine the effects of Safranal P on the viability of the RTgill-W1 cell line to predict the fish acute toxicity. This cell line from rainbow trout (Oncorhynchus mykiss) can be used to assess toxicity of chemicals to the gills which are the organ of fish in most direct contact to environmental chemicals. The close correlation of this in vitro assay has already been demonstrated for predicting acute fish toxicity of chemicals with a narcotic mode of action. Further, a benchmarking study has been performed on 38 fragrance chemicals, covering a broad range of physico-chemical properties and diverse chemistries, and, possessing good quality in vivo fish toxicity studies. The regression equation resulting from this study was used to predict the acute fish toxicity for Safranal P.

The RTgill-W1 cells were seeded in 24-well plates and exposed in minimal medium to different Safranal P concentrations and controls without chemical in triplicate for 24 h at 19°C. GC-analysis of the test chemical concentrations in media was conducted at the start (0 h) and the end (24 h) of exposure. Cell viability was determined at 24 h using metabolic activity as endpoint (PrestoBlue® assay).

One experiment with the RTgill-W1 assay was performed using six different test concentrations. Concentrations ranged from 4 – 128 mg/L based on nominal concentrations (measured concentrations ranging from 3.67 – 122.01 mg/L at the start and 1.32 – 42.35 mg/L at the end of the exposure).

The mean measured test chemical concentrations were 82.18, 41.05, 20.66, 10.31, 5.10, 2.49 mg/L (nominal concentrations 128, 64, 32, 16, 8, 4 mg/L). The effect concentrations (EC50) are based on mean measured concentrations.

In order to predict a fish in vivo LC50 value from the in vitro EC50, a detailed benchmarking study on 38 fragrance chemicals was performed. This led to a predictive regression equation based on the EC50 determined with PrestoBlue® (PB): Log LC50fish in vivo = (0.97 × log EC50PB mean measured) - 0.36

This regression equation was used to predict an in vivo LC50.

The EC50 values and the corresponding NOEC and LOEC values are shown in the following Table, along with the extrapolated in vivo LC50 values.

	Effect concentrations expressed as mean measured concentration (mg/L)
	Cytotoxicity 24 h (mg/L)	Predicted in vivo acute toxicity (LC50) (mg/L)
EC50	42.09 (38.47 - 46.04) mg/L	16.42
LOEC	41.05 mg/L	n.a.
NOEC	20.66 mg/L	n.a.

Values in brackets are 95% confidence limits

n.a. not applicable

The concentrations at which no significant (≤10%) dose related cytotoxicity occurred were ≤20.66 mg/L (nominal 32 mg/L). The NOEC was thus considered to be 20.66 mg/L. After 24 hours, the EC50 value based on mean measured concentration was 42.09 mg/L. Based on the regression equation derived from a training set of 37 fragrance molecules, an in vivo LC50 of 16.42 mg/L is predicted.

Conclusion: The EC50 value for Safranal P was 42.09 mg/L based on mean measured concentrations, and the predicted LC50 value for fish was 16.42 mg/L.

Reference 2

Endpoint:

short-term toxicity to fish

Type of information:

experimental study

Adequacy of study:

supporting study

Study period:

From October 03, 2011 to October 28, 2011

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Justification for type of information:

To support the analogue read-across justification for the daphnia and algal inhibition endpoints.

Qualifier:

according to guideline

Guideline:

OECD Guideline 203 (Fish, Acute Toxicity Test)

Deviations:

no

Qualifier:

according to guideline

Guideline:

EU Method C.1 (Acute Toxicity for Fish)

Deviations:

no

Qualifier:

according to guideline

Guideline:

EPA OPPTS 850.1075 (Freshwater and Saltwater Fish Acute Toxicity Test)

Deviations:

no

Qualifier:

according to guideline

Guideline:

other: Commission Regulation (EC) No 440/2008 of 30 May 2008 laying down test methods pursuant to Regulation (EC) No 1907/2006 of the European Parliament and of the Council on the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), C.

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Remarks:

swissmedic, decision: 19-November-2010

Analytical monitoring:

yes

Details on sampling:

For the determination of the actual test item concentrations in this semi-static test, duplicate samples were taken from each treatment just before the start of the first and of the last renewal periods (Days 0 and 3). At the end of these renewal periods (Days 1 and 4), additional samples were taken for determination of the stability of the test item during the renewal periods of 24 hours.

However, from the two highest nominal test concentrations of 32 and 70 mg/L, the last samples were taken after 24 hours, since all fish were dead at that observation time at these concentrations.

All samples were taken from the approximate center of each aquarium without mixing the test medium, and were stored deep-frozen (at about -20 °C) immediately after sampling. Based on a pre-experiment (non GLP ), the test item was stable under the storage conditions.

The concentrations of the test item GR-50-0091 were determined in one of the duplicate test medium samples from the nominal test concentrations of 7.0, 15 and 32 mg/L from all sampling times.

The samples taken from the lowest and highest nominal test concentration of 3.2 and 70 mg/L were not analyzed, since these test concentrations were below the 96 hour NOEC and LC100, respectively. These test concentrations were therefore not considered as being a relevant part of the concentration-effect curve.

From the control samples, only one of the duplicate samples was analyzed per sampling time.

The analytical procedure and results are described in the attached Appendix I - Analytical Investigations.

Vehicle:

no

Details on test solutions:

The preparation of the test media was performed as far as possible in the dark to avoid photolytic degradation of the test item.

Due to the large volume of the test media, the highest test concentration of 70 mg/L was prepared by completely dissolving 759.9 mg/L of test item into 10850 mL of test water. For the test media of the lower test concentrations a concentrated stock solution of nominal 100 mg/L was freshly prepared by completely dissolving 1128 mg of test item in 11280 mL of test water under intense stirring for 3 hours at room temperature in the dark in a completely filled and tightly closed stirring vessel. The stock solution showed no trace of precipitate or surface floating oily film, and, thus, the test material was assumed to be completely dissolved. Adequate volumes of this intensively stirred stock solution were added to the test water in the test vessels and were intensively mixed to prepare the test media with the test concentrations as stated above.

The test media were freshly prepared just before introduction of the fish (= start of the exposure) and before each test medium renewal.

Test organisms (species):

Danio rerio (previous name: Brachydanio rerio)

Details on test organisms:

The study was performed with zebra fish (Brachydanio rerio). The fish were obtained from a breeding culture at Harlan Laboratories. No medication was applied during holding and acclimatization. Prior to test start, the test fish were acclimated for one week to the test water and temperature.

During holding and acclimatization until two days before the start of the test, the fish were fed with a commercial fish diet (Tetra Min® Hauptfutter, supplied by TETRA-Werke, 49304 Melle / Germany). During the last two weeks prior to the test, no fish died in the test fish batch and all fish were healthy. The age of the tst fish was approximately

From the acclimated test fish batch, 10 fish were measured at the start of the test. The mean body length of the fish was 3.0 ± 0.18 cm (Mean ± SD), the mean body wet weight was 0.20 ± 0.05 g (Mean ± SD). These measured fish were not introduced for the test.


The test method and test species are recommended by the international test guidelines.

Test type:

semi-static

Water media type:

freshwater

Limit test:

no

Total exposure duration:

96 h

Hardness:

1.25 mmol/L

Test temperature:

The water temperature was 22°C during the test (see attached Table 4).

pH:

The pH values in all treatments ranged from 6.8 to 7.0 (see attached Table 2).

Dissolved oxygen:

The oxygen concentration was at least 7.4 mg/L (see attached Table 3), corresponding to at least 84% oxygen saturation.

Salinity:

according to OECD test guideline.

Nominal and measured concentrations:

The nominal test item concentrations tested were 3.2, 7.0, 15, 32 and 70 mg/L. Additionally, a control group was tested in parallel.

In the analyzed test medium samples from the start of the test medium renewal periods the measured test item concentrations ranged between 90 and 95% of the nominal values. This shows the correct preparation of the test media. At the end of the test medium renewal periods the test concentrations had slightly decreased to 79 to 88% of the nominal values.

The mean concentration of each renewal period was calculated as geometric mean of the test item concentrations measured at the start and the end of the respective test medium renewal. From the values obtained, the mean measured test item concentration during the test period was calculated as arithmetic mean.

Details on test conditions:

Reconstituted test water was used in the study. It consisted of analytical grade salts dissolved in purified water. For further information on the test water please see section " Any other information on materials and methods incl. tables" balow.

Since the test item was considered to be volatile (according to the results of a pre-experiment without GLP), the test was performed in a closed system. One 10-liter Schott flask was used for each treatment. The Erlenmeyer flasks were nearly completely filled to keep the air-space in the test flasks as small as possible and were tightly sealed.

The test vessels were labeled with the study number and all necessary additional information to ensure unique identification.

The water temperature in the test vessels was maintained at 22 °C. The test vessels were not aerated during the test period.

Since the photolytic stability of the test item was unkown, the test was performed under reduced light conditions to avoid photolytic degradation of the test item.

The test duration was 96 hours and the fish were not fed during the test.

The following nominal concentrations of GR-50-0091 were tested: 3.2, 7.0, 15, 32 and 70 mg/L. Additionally, a control (test water without test item) was tested in parallel.

The selection of the test concentrations was based on the results of a range-finding test (non-GLP).

According to the range finding test, the test item was not stable during the test period of 96 hours. Therefore, a semi-static test procedure was chosen: the test media were renewed daily to keep the concentrations of GR 50-0091 in the test media as constant as possible during the test period of 96 hours.

During this semi-static test, the test fish were transferred daily by carefully netting into a new test vessel with freshly prepared test medium of the corresponding treatment .

At the start of the exposure, 7 fish were introduced into each test vessel in a random order. The loading rate was 0.14 g fish wet weight per liter test medium. Thus, the requirement of a loading rate not exceeding 0.8 g fish/L was fulfilled.

Reference substance (positive control):

no

Key result

Duration:

96 h

Dose descriptor:

LC50

Effect conc.:

9.2 mg/L

Nominal / measured:

meas. (geom. mean)

Conc. based on:

test mat.

Basis for effect:

mortality (fish)

Remarks on result:

other: 95%CL: 6.1 - 14 mg/L

Duration:

96 h

Dose descriptor:

LC100

Effect conc.:

14 mg/L

Nominal / measured:

meas. (geom. mean)

Conc. based on:

test mat.

Basis for effect:

mortality (fish)

Duration:

96 h

Dose descriptor:

LOEC

Effect conc.:

14 mg/L

Nominal / measured:

meas. (geom. mean)

Conc. based on:

test mat.

Basis for effect:

mortality (fish)

Duration:

96 h

Dose descriptor:

NOEC

Effect conc.:

6.1 mg/L

Nominal / measured:

meas. (geom. mean)

Conc. based on:

test mat.

Basis for effect:

mortality (fish)

Details on results:

In the analyzed test medium samples from the start of the test medium renewal periods the measured test item concentrations ranged between 90 and 95% of the nominal values (see analytical results and Table 3 in Appendix I). This shows the correct preparation of the test media. At the end of the test medium renewal periods the test concentrations had slightly decreased to 79 to 88% of the nominal values.

The mean concentration of each renewal period was calculated as geometric mean of the test item concentrations measured at the start and the end of the respective test medium renewal. From the values obtained, the mean measured test item concentration during the test period was calculated as the arithmetic mean.

For the calculation formula of mean measured test item concentration please see section "Any other information on results incl. tables", below.

The biological test results after 96 hours exposure were related to the mean measured test concentrations.

In the control and at the test concentrations up to and including 6.1 mg/L, all fish survived until the end of the test and no visible abnormalities were observed in the test fish. At the next higher test concentration of 14 mg/L, all test fish showed severe visible abnormalities, and, therefore they were humanely killed before the end of the test for the animal welfare reasons. At the two highest test concentrations all fish were dead already after 24 hours of exposure.

Thus, the 96 hour NOEC and LC0 of GR-50-0091 to zebra fish were both determined to be 6.1 mg/L (mean measured concentration). The 96 hour LOEC was 14 mg/L.

The 96 hour LC50 of GR-50-0091 was calculated to be 9.2 mg/L with a 95% confidence interval of 6.1 14 mg/L. The 96 hour LC100 was 14 mg/L. Thus, the concentration effect relationship was relatively steep.

No remarkable observations were made concerning the appearance of the test media. All test media were clear solutions throughout the entire test duration.

The pH values in all treatments ranged from 6.8 to 7.0 (see attached Table 2). The oxygen concentration was at least 7.4 mg/L (see attached Table 3), corresponding to at least 84% oxygen saturation. The water temperature was 22°C during the test (see attached Table 4).

The test was considered to be valid, since no mortality in the control was observed and the validity criterion of at least 60% oxygen saturation was fulfilled.

Reported statistics and error estimates:

The test fish were observed for mortality and visible abnormalities ( see attached Table 1 for evaluation criteria) after approximately 3, 24, 48, 72 and 96 hours test duration. Dead fish were removed at least once daily and discarded.

Test fish, which were found in a moribund condition, were humanely killed before test end due to the reasons of animal welfare (see section General Information / Animal Welfare). Criteria of moribund condition of the test fish are stated in the Harlan Laboratories Standard Operating Procedure and are based on recommendations of the Swiss Federal Veterinary Office (Schweizer Bundesamt für Veterinärwesen). For calculation of the LC50 value, the sacrificed fish were taken into account as dead fish.

The LC50 and the 95%-confidence interval at the observation interval after 48 hours was calculated by Moving Average Interpolation.

The LC50 at the observation intervals after 24, 72 and 96 hours could not be calculated by Moving Average Interpolation or regression analysis due to a steep concentration-effect relationship. Instead, the LC50 values were determined as a geometric mean value of the two consecutive test concentrations with 0 and 100% mortality, and the 95% confidence limits for the LC50 as the test concentrations with 0 and 100% mortality.

The NOEC, LOEC, LC0 and LC100 were determined directly from the raw data.

Sublethal observations / clinical signs:

The Table below shows the calculated mean measured concentrations:

 

Nominal test concentration	Mean measured concentration of the test item
(mg/L)	(mg/L)	(% of nominal)
3.2	---	---
7.0	6.1	87
15	14	93
32	28	88
70	---	---

---:      not analyzed, since below NOEC or above LC100

 

Validity criteria fulfilled:

yes

Conclusions:

This acute fish toxicity study is for an analogue substance (Shisolia). It has been included in the registration dossier of 2,6,6-trimethylcyclohexa-1,3-dienecarbaldehyde to support the analogue read-across justification for the daphnia and algal inhibition endpoints. The determined LC50 value for the analogue substance, Shisolia, is 9.2 mg/L. The determined LC50 value for the registered substance (see key study) is 4.3mg/L.

The similar acute fish LC50 values support the analogue read-across hypothesis that based on the identical aquatic toxicity mode of action classes identified for the main constituents of the source and target substance and the similar hydrophobicity, the estimated strength of effects for the target substance (Safranal) is expected to be in the same range as that for the source substance (Shisolia).

Executive summary:

The acute toxicity of the test item GR-50-0091 to zebra fish (Brachydanio rerio) was determined in a 96‑hour semi-static test with daily test medium renewal according to the EU Commission Directive 92/69/, Part C.1, the Commission Regulation (EC) No 440/2008, Part C.1 and the OECD Guideline for Testing of Chemicals No. 203 (1992), as well as according to the OPPTS Guideline No. 850.1075 (Public Draft, April 1996).

 

Since the test item was considered to be volatile, the test was performed in a closed system. The test was performed under reduced light conditions to avoid photolytic degradation of the test item, in the event that the test substance may be susceptile to aqueous photolysis.

 

The nominal test item concentrations tested were 3.2, 7.0, 15, 32 and 70 mg/L. Additionally, a control group was tested in parallel.

 

In the analyzed test medium samples from the start of the test medium renewal periods the measured test item concentrations ranged between 90 and 95% of the nominal values. This shows the correct preparation of the test media. At the end of the test medium renewal periods the test concentrations had slightly decreased to 79 to 88% of the nominal values.

 

The mean concentration of each renewal period was calculated as geometric mean of the test item concentrations measured at the start and the end of the respective test medium renewal. From the values obtained, the mean measured test item concentration during the test period was calculated as arithmetic mean.

 

Nominal test concentration	Mean measured concentration of the test item
(mg/L)	(mg/L)	(% of nominal)
3.2	---	---
7.0	6.1	87
15	14	93
32	28	88
70	---	---

---:      not analyzed, since below NOEC or above LC100

 

 

In the control and at the test concentrations up to and including 6.1 mg/L, all fish survived until the end of the test and no visible abnormalities were observed in the test fish. At the next higher test concentration of 14 mg/L, all test fish showed severe visible abnormalities, and, therefore they were humanely killed before the end of the test for the animal welfare reasons. At the two highest test concentrations all fish were dead already after 24 hours of exposure.

 

The biological test results after 96 hours exposure were related to the mean measured test concentrations:

 

– 96-hour LC50:		9.2	mg/L
	(95% confidence interval:	6.1‑14	mg/L)
– 96-hour LC0:		6.1	mg/L
– 96-hour LC100:		14	mg/L
– 96-hour NOEC:		6.1	mg/L
– 96-hour LOEC:		14	mg/L

 

 

The test was considered to be valid, since no mortality in the control was observed and the validity criterion of at least 60% oxygen saturation was fulfilled.

Reference 3

Endpoint:

short-term toxicity to fish

Type of information:

experimental study

Adequacy of study:

key study

Study period:

From September 17, 2019 to December 03, 2019

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 203 (Fish, Acute Toxicity Test)

GLP compliance:

yes (incl. QA statement)

Analytical monitoring:

yes

Remarks:

HPLC-UV

Details on sampling:

The 10 mL test samples and control samples were completely diluted with 2.5 mL methanol.

Vehicle:

no

Test organisms (species):

Danio rerio (previous name: Brachydanio rerio)

Details on test organisms:

The study was performed with zebrafish (Danio rerio). The fish were obtained from a breeding culture at IES Laboratories. During the last three weeks prior to the test, no fish died in the test fish batch and all fish were healthy. No medication was applied during holding and acclimatization. Prior to test start, the test fish were acclimated for at least one week to the test water and temperature.
During holding and acclimatization until one day before the start of the test, the fish were fed with a commercial fish diet (Tetra Min® Hauptfutter, supplied by TETRA GmbH, 49324 Melle / Germany).
From the acclimated test fish batch, a representative sub-batch of 10 fish was measured at the start of the test. The total body length (measured from the mouth to the end of the tail fin) of the fish was between 1.8 and 2.0 cm (with a mean of 1.96 cm) and the body wet weight was between 0.04 and 0.06 g (with a mean of 0.05 g). The measured fish were not used for the test.
The test method and the test species are recommended by the test guidelines. All fish used for the test item treatments were humanely sacrificed at the end of the test according to the relevant IES SOP and the Swiss Federal Food Safety and Veterinary Office guideline 3.01 “Skilled and animal welfare compliant killing of laboratory animals”.

Test type:

semi-static

Water media type:

other: Reconstituted test water was used in the study.

Limit test:

no

Total exposure duration:

96 h

Test temperature:

The water temperature was measured to be 21 °C during the test period.

pH:

The pH values in the test media and the control ranged from 7.0 to 7.5 during the test period.

Dissolved oxygen:

The dissolved oxygen concentration was in the range of 8.5 to 8.7 mg/L at the start of the test media renewal periods and slightly decreased to 8.0 – 8.5 mg/L (i.e. 89 – 95 % saturation) in the closed system during the 24-hour renewal periods. Therefore, the oxygen concentration never dropped below 60 % of the air saturation value.

Nominal and measured concentrations:

The following nominal concentrations of Safranal P were tested in the full test: 1.5, 2.7, 4.9, 8.9 and 16 mg/L, giving a spacing factor of 1.8.

Details on test conditions:

At the start of the test, seven fish were introduced into each test vessel in a random order. The loading rate of the fish was 0.06 g fish wet weight per liter medium. After 24 hours, the fish were carefully transferred into clean test bottles (of the same filling volume) with freshly prepared test medium.

Reference substance (positive control):

no

Key result

Duration:

96 h

Dose descriptor:

LC50

Effect conc.:

4.3 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Duration:

96 h

Dose descriptor:

LC0

Effect conc.:

2.7 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Duration:

96 h

Dose descriptor:

LC100

Effect conc.:

8.9 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Duration:

96 h

Dose descriptor:

NOEC

Effect conc.:

2.7 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Key result

Duration:

96 h

Dose descriptor:

LOEC

Effect conc.:

4.9 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Details on results:

In the control and the two lower tested concentrations of 1.5 and 2.7 mg/L, no toxic effects were observed in the fish until the end of the test. At the next higher test concentration of 4.9 mg/L two of seven fish died at 48 hours of exposure and finally only two fish survived at the end of the test. For the test concentration of 8.9 mg/L hypoactivity (i.e. abnormal swimming behavior) was observed during the second renewal period, followed by the death of six fish at 48 hours of exposure and of the seventh fish before the last renewal. At the highest test concentrations of 16 mg/L all fish were dead after 24 hours of exposure.

The actual test item concentrations in the test media were directly measured in samples taken at the start and end of each renewal period. Between 93 and 98 % of the nominal Safranal P values were measured in the freshly prepared test media at the start of the four test medium renewal periods demonstrating the correct preparation of the test media. The measured test item concentrations in the aged test media at the end of the test medium renewal periods of 24 hours were between 86 and 96 % of the nominal concentrations. Therefore, by applying a semi-static test design the exposure concentration could be maintained as constant as possible during the total exposure period of 96 hours. In consequence, the test results are based on nominal concentrations.

Validity criteria fulfilled:

yes

Remarks:

The test was considered valid as no fish died in the control during the 96-hour test period. The oxygen concentration in the test media and the control did not drop below 60 % of the air saturation during the test.

Conclusions:

The test item Safranal P had acute toxic effects on juvenile zebrafish (Danio rerio) under the tested conditions (i.e. semi-static test with 24 h renewal periods in a closed system). Based on nominal concentrations, the 96-hour LC50 for fish was calculated to be 4.3 mg/L.
This valid test was performed according to the OECD Guideline for the Testing of Chemicals, No. 203, Fish, Acute Toxicity Test, 2019.

Executive summary:

The acute toxicity of the test item Safranal P to juvenile fish of the species zebrafish (Danio rerio) was determined in a 96-hour test according to the OECD Guideline for the Testing of Chemicals, No. 203, Fish, Acute Toxicity Test, 2019.

First a limit test at the nominal test item concentration of 17 mg/L was performed to demonstrate that the test organism fish is not the most sensitive test organism to Safranal P compared to algae and daphnia. The test method is based on the threshold approach as developed for chemical substances at the European Commission’s joint Research Centre.

Since mortality was observed in the limit test, according to the guideline, a full test with five test concentrations was performed. The following nominal test item concentrations were tested in parallel to a control without test item: 1.5, 2.7, 4.9, 8.9 and 16 mg Safranal P/L, giving a spacing factor of 1.8.

The test concentrations and the test design were selected based on the results of a range finding test and in agreement with the Sponsor. A semi-static test design with test medium renewals after 24 hours was used to keep the concentration of the test item as constant as possible during the test period of 96 hours of the full test. Further, in order to counteract a possible test item loss by volatility, the test was performed in a closed system using 5-liter glass bottles completely filled (without headspace) with test medium that were tightly sealed with Teflon stoppers.

The test design was based on the OECD Guidance Document No. 23 on Aqueous-Phase Aquatic Toxicity Testing of Difficult Test Chemicals, 2019.

The actual test item concentrations in the test media were directly measured in samples taken at the start and end of each renewal period. Between 93 and 98 % of the nominal Safranal P values were measured in the freshly prepared test media at the start of the four test medium renewal periods demonstrating the correct preparation of the test media.

The measured test item concentrations in the aged test media at the end of the test medium renewal periods of 24 hours were between 86 and 96 % of the nominal concentrations. Therefore, by applying a semi-static test design the exposure concentration could be maintained as constant as possible during the total exposure period of 96 hours. In consequence, the test results are based on nominal concentrations.

The biological results after 96 hours of exposure to Safranal P are summarized in the table below.

Endpoints after 96 Hours Test Period	Based on Nominal Concentrations of Safranal P [mg/L]
96-hour LC50	1.3
95 %-confidence limits	3.5-5.3
96-hour LC0	2.7
96-hour LC100	8.9
96-hour NOEC	2.7
96-hour LOEC	4.9

Validity Criteria: The test was considered valid as no fish died in the control during the 96-hour test period. The oxygen concentration in the test media and the control did not drop below 60 % of the air saturation during the test.

Description of key information

Based on a valid short-term fish study on the registered substance (test item, Safranal P), the 96h LC50 for 2,6,6 -trimethylcyclohexa-1,3-dienecarbaldehyde was determined to be 4.3mg/L. A gill cell-line assay was also performed on Safranal P, which gave a predicted LC50 of 16.4 mg/L (11.7 mg/L, considering analogue information). The LC50 value of 4.3 mg/L has been chosen as the key value because it is the most conservative and from the in vivo OECD 203 study.

Key value for chemical safety assessment

Fresh water fish

Fresh water fish

Effect concentration:

4.3 mg/L

Additional information

The target substance 2,6,6-trimethylcyclohexa-1,3-dienecarbaldehyde is a mono-constituent substance (EC 204-133-7, CAS 116-26-7). Two commercial qualities are covered by the REACH lead registration, Safranal and Safranal P. The only difference between the two qualities is the amount of the additive, alpha-tocopherol (EC 233-466-0, CAS 10191-41-0), that is intentionally added to stabilise the substance. The level of this additive is 9-14% in Safranal and 1-2% in Safranal P. The latter was used by the lead registrant in aquatic toxicity studies because of the lower alpha-tocopherol level and hence higher purity of mono-constituent substance. Two studies have been performed using Safranal P to address the short term toxicity to fish; Acute Toxicity to Zebra fish according to OECD No 203 (KEY study, Eisner 2019) and a RTgill-WI cell-line assay for predicting Fish Acute Toxicity (SUPPORTING study, Laue 2019). The measured and predicted LC50 values were 4.3 and 16.4 mg/L (11.7 mg/L, considering analogue information). The LC50 value of 4.3 mg/L has been chosen as the key value because it is the most conservative and from the in vivo OECD 203 study. However, the prediction from the cell line assay demonstrates the potential utility of this alternative method in avoiding animal testing in other substance registrations. The details of both these studies are summarised further below.

A valid acute fish study for the closely related analogue substance (Shisolia) is also available. This has been included in the registration dossier of 2,6,6-trimethylcyclohexa-1,3-dienecarbaldehyde to support the analogue read-across justification for the daphnia and algal inhibition endpoints. The determined LC50 value for the analogue substance, Shisolia, is 9.2 mg/L. Details are available in a supporting end-point study record but not provided here in the end-point summary as they have not been taken into account in the assessment of the acute toxicity to fish of the registered substance.

ACUTE TOXICITY STUDY

The acute toxicity of the test item Safranal P to juvenile fish of the species zebrafish (Danio rerio) was determined in a 96-hour test according to the OECD Guideline for the Testing of Chemicals, No. 203, Fish, Acute Toxicity Test, 2019. First a limit test at the nominal test item concentration of 17 mg/L was performed to demonstrate that the test organism fish is not the most sensitive test organism to Safranal P compared to algae and daphnia (EC50 values of 15 and 19 mg/L based on read-across, which was the only information available to the Lead Registrant at the time of placing the acute fish test). The test method is based on the threshold approach as developed for chemical substances at the European Commission’s Joint Research Centre. Since mortality was observed in the limit test, according to the guideline, a full test with five test concentrations was performed. The following nominal test item concentrations were tested in parallel to a control without test item: 1.5, 2.7, 4.9, 8.9 and 16 mg Safranal P/L, giving a spacing factor of 1.8.

A semi-static test design with test medium renewals after 24 hours was used to keep the concentration of the test item as constant as possible during the test period of 96 hours. Further, in order to counteract a possible test item loss by volatility, the test was performed in a closed system using 5-liter glass bottles completely filled (without headspace) with test medium that were tightly sealed with Teflon stoppers.

The actual test item concentrations in the test media were directly measured in samples taken at the start and end of each renewal period. Between 93 and 98 % of the nominal Safranal P values were measured in the freshly prepared test media at the start of the four test medium renewal periods demonstrating the correct preparation of the test media. The measured test item concentrations in the aged test media at the end of the test medium renewal periods of 24 hours were between 86 and 96 % of the nominal concentrations. Therefore, by applying a semi-static test design the exposure concentration could be maintained as constant as possible during the total exposure period of 96 hours. In consequence, the test results are based on nominal concentrations. The 96 -hour LC50 was determined to be 4.3mg/L (95% confirnce limits of 3.5 - 5.3 mg/L).

IN VITRO CELL LINE ASSAY

The objective of the study was to determine the effects of Safranal P on the viability of the RTgill-W1 cell line to predict the fish acute toxicity. This cell line from rainbow trout (Oncorhynchus mykiss) can be used to assess toxicity of chemicals to the gills which are the organ of fish in most direct contact to environmental chemicals. The close correlation of this in vitro assay has already been demonstrated for predicting acute fish toxicity of chemicals with a narcotic mode of action [Tanneberger, K., et al, 2013]. Further, a benchmarking study has been performed on 38 fragrance chemicals, covering a broad range of physico-chemical properties and diverse chemistries, and, possessing good quality in vivo fish toxicity studies [Natsch, A., et al., 2018].

The RTgill-W1 cells were seeded in 24-well plates and exposed in minimal medium to different Safranal P concentrations and controls without chemical in triplicate for 24 h at 19°C. GC-analysis of the test chemical concentrations in media was conducted at the start (0 h) and the end (24 h) of exposure. Cell viability was determined at 24 h using metabolic activity as endpoint (PrestoBlue® assay). One experiment with the RTgill-W1 assay was performed using six different test concentrations. Concentrations ranged from 4 – 128 mg/L based on nominal concentrations (measured concentrations ranging from 3.67 – 122.01 mg/L at the start and 1.32 – 42.35 mg/L at the end of the exposure). The mean measured test chemical concentrations were 82.18, 41.05, 20.66, 10.31, 5.10, 2.49 mg/L (nominal concentrations 128, 64, 32, 16, 8, 4 mg/L). The effect concentrations (EC50) are based on mean measured concentrations. The EC50 was determined to be 42.09 mg/L (95% confidence limits, 38.47 - 46.04 mg/L).

In order to predict a fish in vivo LC50 value from the in vitro EC50, the following regression equation from the study of Natsch et al. [2018] was used: Log LC50 (fish in vivo) = (0.97 × log EC50 PB mean measured) - 0.36; n = 37, R2 = 0.94. The predicted LC50 value for fish was 16.42 mg/L.

The structural analogue (Shisolia) was part of the dataset used to generate the above regression equation [Natsch, A., et al., 2018]. In a leave-one-out analysis, performed by the authors to estimate the predictivity of the assay when challenged with a new chemical, the LC50 of Shisolia was found to be over-predicted by 1.4 (predicted LC50 value =13 mg/L, measured LC50 value = 9.3mg/L). Applying this correction factor to the predicted LC50 value for Safranal P, gives a corrected value of 11.7 mg/L.

REFERENCES

Tanneberger, K., et al., Predicting fish acute toxicity using a fish gill cell line-based toxicity assay. Environ. Sci. Technol., 2013. 47(2): p. 1110-9.

Natsch, A., et al., Accurate prediction of acute fish toxicity of fragrance chemicals with the RTgill-W1 cell assay. Environ. Toxicol. Chem., 2018. 37(3): p. 931-941.