Registration Dossier

Ecotoxicological information

Long-term toxicity to fish

Currently viewing:

Administrative data

Link to relevant study record(s)

Reference
Endpoint:
fish early-life stage toxicity
Type of information:
experimental study
Adequacy of study:
key study
Study period:
2017-09-22 to 2018-04-16
Reliability:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
guideline study
Qualifier:
according to guideline
Guideline:
OECD Guideline 210 (Fish, Early-Life Stage Toxicity Test)
Version / remarks:
adopted July 26, 2013
Deviations:
no
Qualifier:
according to guideline
Guideline:
other: OECD Series on Testing and Assessment, No. 23, "Guidance Document on Aquatic Toxicity Testing of Difficult Substances and Mixtures"
Version / remarks:
December 15, 2000
GLP compliance:
yes (incl. QA statement)
Specific details on test material used for the study:
Biodegradability: Readily biodegradable, see IUCLID section 4.2.1;
Water solubility: 0.078 μg/L (Sonntag 2017), see IUCLID section 4.8;
Storage Conditions at Test Facility: At 20 ± 5 °C, in the dark.
Analytical monitoring:
yes
Details on sampling:
One sample from the freshly prepared stock solutions and one sample from each replicate (aquaria) of the test media of all test concentrations and the controls was taken prior to the initiation of the test. Afterwards once per week at day 0 (=start of the test), 3, 10, 17, 24 and 30 days post hatch, one sample from each replicate (aquaria) from the test media of all test concentrations and the controls was taken. In addition, one sample from the freshly prepared stock solutions was taken at day 0 (=start of the test), 4, 11, 18 and 25 days post hatch. All test medium samples were taken from the approximate centre of the aquaria. One aliquot (apart from stock solution samples in organic solvent) of the samples was diluted by a factor of 2 with acetonitrile. Another aliquot remained undiluted until sample preparation, which was performed directly after sampling.

Storage: All undiluted samples were extracted with the organic solvent n-hexane directly after sampling. An aliquot of the n-hexane extracts as well as all samples diluted with acetonitrile were stored in a freezer (≤ - 20 °C), protected from light until analysis was performed. Afterwards the samples were again stored deep frozen (< -20 °C) and were kept stored up to the date of the final report.

Analyses: The concentrations of the test item Matrilox LP101M were analysed in each of the undiluted test media samples from all test concentrations, and in each of the undiluted control samples, from all sampling times. In addition, the test item was analysed in the stock solution samples taken at day 0 and 4 days post hatch. The samples diluted with acetonitrile were not analysed.
Vehicle:
yes
Remarks:
N,N-Dimethylformamide (DMF) at 20 µL/L
Details on test solutions:
Test Concentrations (nominal) definitive test:
0.12, 0.08, 0.05 μg test item/L (each in 20 μL DMF/L);
Controls:
Control: In the control, test water was used without addition of the solvent or the test item.
Solvent Control: In the solvent control, test water was used with the solvent 20 μL DMF/L but without the test item.

Dosage of Test Item:
A concentrated stock solution of 300 mg test item/L DMF was prepared by dissolving:
15.0 mg test item in 50 mL DMF or
15.4 mg test item in 51.3 mL DMF or
16.0 mg test item in 53.3 mL DMF
by intensive stirring for 1 - 2 min.

Stock solutions for each test concentration were prepared with 6000, 4000 and 2500 μg test item/L DMF. In order to reach the requested concentrations of 0.12, 0.08, 0.05 μg test item/L the stock solutions of test item were dosed with 0.278 μL/min (± 10%) and mixed with 13.9 mL/min (± 10%) of test water.
For the control 13.9 mL/min (± 10%) of test water was carried.
For the solvent control 0.278 μL/min (± 10%) DMF were mixed with 13.9 mL/min (± 10%) of test water.

The dosage of the test item stock solution and solvent was performed with a syringe pump. The accuracy of the dosage was checked before equilibration phase and after the exposure period (end of biological part). Accordingly, over all groups and replicates, a minimum flow of 0.254 µL/min (91% of nominal) and a maximum flow of 0.278 µL/min (100% of nominal) could be demonstrated and such, correct dosing could be confirmed. An operation checkout of the syringe pumps was performed twice a week during exposure phase.
The dosage of the test water was performed with a tube pump. The accuracy of the dosage was checked 9, 6, 2 and 1 days prior the initiation of the experiment. After insertion of the eggs checks on accuracy were performed twice a week. Accordingly, over all groups and replicates, a minimum flow of 13 mL/min (94% of nominal) and a maximum flow of 15 mL/min (108% of nominal) could be demonstrated and such, correct dilution water flow could be confirmed.
The stock solutions and the test water were pumped in mixing vessels (Erlenmeyer glass flask, one per replicate) with a constant flow rate by syringe pumps (test item stock solutions or flexible-tube pumps (test water), respectively. In these mixing vessels, the application stock solutions and the test water were continuously mixed using a magnetic stirrer. Nominal test concentrations of 0.12, 0.08, and 0.05 μg test item/L did result. The mixing vessels and the aquaria were connected by a tube (Polytetrafluoroethylene (PTFE)).
The stock solutions were renewed every 3 - 4 days.

Prior to the initiation of the test, the dosing system was calibrated through the use of appropriate analysis techniques (this part was not performed according to the GLP-Regulations and is excluded from
the Statement of Compliance in the final report, but the raw data of these tests are archived under the project number of the present study).

Appearance of the Test Item in Test Medium:
The appearance of the test item in the test medium was observed once every day in all test concentrations.
Test organisms (species):
Danio rerio (previous name: Brachydanio rerio)
Details on test organisms:
Species: Zebrafish (Danio rerio)
Age: Freshly fertilised eggs (before beginning of gastrulation)
Origin: The test eggs were obtained from the brood stock of the ibacon GmbH
Test type:
flow-through
Water media type:
freshwater
Limit test:
no
Total exposure duration:
34 d
Remarks on exposure duration:
Exposure duration: until 30 days post hatch, corresponding to 34 days after insertion of eggs, i.e. total exposure duration
Post exposure observation period:
No
Hardness:
124.6 to 160.2 mg CaCO3/L
Test temperature:
25.1 to 26.4 °C
pH:
7.6 to 8.0
Dissolved oxygen:
96 to 102 % of the air saturation value
Salinity:
not applicable
Conductivity:
Deionised water (conductivity <= 10 µS/cm) was used for preparation of the test medium (ISO medium according to OECD 203).
Nominal and measured concentrations:
0.12, 0.08, 0.05 μg test item/L (each in 20 μL DMF/L), corresponding to the following mean measured concentrations of the test item, corrected for recovery of fortified samples:
0.0600, 0.0328 and 0.0243 μg test item/L (each in 20 μL DMF/L).
Based on results from the preliminary test and the determined water solubility of 0.078 µg/L, no toxicity of the test item up to the saturation concentration in water was expected. Therefore, it was decided that an extended limit test as foreseen according to OECD 210, paragraph 22 would be most appropriate in this case. Using a lower spacing factor of 1.5 and testing three test item concentrations resulted in actually applied nominal test item concentrations of 0.12 µg test item/L (water solubility times a safety factor of 1.5 as a reasonable compromise between trying to avoid a relevant non-dissolved fraction on the one hand and minimizing the risk of failing a saturated solution in case of analytical recoveries somewhat below the nominal concentration), 0.08, and 0.05 µg test item/L. While reducing the normally required 5 test item concentrations to three, analytical monitoring was extended beyond the minimum required according to OECD 210 (weekly, at least in one replicate per concentration level), in that regularly all four replicates per concentration level were analysed. For details on considerations regarding test design see IUCLID section "Any other information on materials and methods incl. tables".


Details on test conditions:
Plese see IUCLID section "Any other information on materials and methods incl. tables".
Reference substance (positive control):
no
Key result
Duration:
34 d
Dose descriptor:
NOEC
Effect conc.:
>= 0.06 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Remarks:
completely dissolved
Basis for effect:
number hatched
Remarks on result:
other: corresponding to the saturation level in test medium
Key result
Duration:
34 d
Dose descriptor:
LOEC
Effect conc.:
> 0.06 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Remarks:
completely dissolved
Basis for effect:
number hatched
Remarks on result:
other: corresponding to the saturation level in test medium
Key result
Duration:
34 d
Dose descriptor:
NOEC
Effect conc.:
>= 0.06 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Remarks:
completely dissoved
Basis for effect:
mortality
Remarks:
post hatch survival
Remarks on result:
other: corresponding to the saturation level in test medium
Key result
Duration:
34 d
Dose descriptor:
LOEC
Effect conc.:
> 0.06 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Remarks:
completely dissolved
Basis for effect:
mortality
Remarks:
post hatch survival
Remarks on result:
other: corresponding to the saturation level in test medium
Key result
Duration:
34 d
Dose descriptor:
NOEC
Effect conc.:
>= 0.06 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Remarks:
completely dissolved
Basis for effect:
length
Remarks on result:
other: corresponding to the saturation level in test medium
Key result
Duration:
34 d
Dose descriptor:
LOEC
Effect conc.:
> 0.06 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Remarks:
completely dissolved
Basis for effect:
length
Remarks on result:
other: corresponding to the saturation level in test medium
Key result
Duration:
34 d
Dose descriptor:
NOEC
Effect conc.:
>= 0.06 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Remarks:
completely dissolved
Basis for effect:
weight
Remarks:
wet as well as dry weight
Remarks on result:
other: corresponding to the saturation level in test medium
Key result
Duration:
34 d
Dose descriptor:
LOEC
Effect conc.:
> 0.06 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Remarks:
completely dissolved
Basis for effect:
weight
Remarks:
wet as well as dry weight
Remarks on result:
other: corresponding to the saturation level in test medium
Details on results:
Up to the saturation level in test medium (corresponding to the mean measured concentration of 0.060 µg test item /L) not any toxic effect of the test item relative to controls could be observed. While analytically determined mean measured concentration over all replicates for the highest nominal concentration of 0.12 µg/L test item was somewhat below the determined solubility in test medium (0.078 µg/L), the determined arithmetic mean measured concentration for replicate 3 over all days (0.072 µg/L) corresponds to 92% of the water solubility, while arithmetic mean measured concentration for replicate 3 up to and including day 24 post hatch (28 days of exposure) (0.084 µg/L) is even slightly above the saturation level of the test item in ISO medium (107% of the water solubility). No distinct differences to other replicates of this test concentration were obvious, neither for dry / wet weight or fish length, nor for hatching success or post hatch survival. This corroborates that up to the saturation level in water, no effects of the test item on fish early life stages occurred.
For further details, please see IUCLID sections "Any other information on results incl. tables" as well as "Overall remarks, attachments".

Results with reference substance (positive control):
Not applicable
Reported statistics and error estimates:
Please see IUCLID section "Any other information on materials and methods incl. tables".

Validity Criteria of the Study

Control Survival:       

In the control the survival of fertilised eggs was 100% being > 70 % (validity criterion). The post hatch survival was 96 % being > 75 % (validity criterion). The overall mortality (including coagulated eggs, dead embryos and dead larvae was 4 %.

Thus the validity criterion for survival in the control at the end of the study was met.

Water Temperature:

The water temperature differed not more than ±1.5°C between the test vessels or between the days at any time during the test.

Oxygen Concentration:       

The dissolved oxygen concentration in the test media did not fall below 60% of air saturation value during the test.

Analytical measurement:       

The test concentrations were verified at test start in all replicates (aquaria) and at least once a week afterwards throughout the test until test end. Since the measured concentrations did not remain within 80-120% of the nominal concentration, the effect concentrations were expressed as arithmetic mean concentrations.

Biological Results       

Embryonic Stage at Test Start:       

8 -16 cell stage;

Number of Coagulated Eggs:       

No eggs coagulated in any treatment group.

Hatching Time and Success:       

First larvae hatched from eggs 2 days after insertion in the control and all test item treatment groups. > 90 % of eggs were hatched at day 4 after insertion (start of post hatch period). Hatching was completed 5 days after insertion of eggs in the control and all test item treatment groups. The LOEC for hatchability was determined to be > 0.0600 µg test item/L, and the NOEC was ≥ 0.0600 µg test item/L (100% hatching success in all control and treatment groups).

Mortality and Sublethal effects:      

No test item related sublethal effects occurred. At 0.0328 µg test item /L one fish tumbled and one fish was found mainly on the bottom. However, these abnormalities were in the range of biological variance.

During the test several larvae died in all treatment groups and the controls. The survival in all treatment groups was comparable to the pooled control and therefore within the biological variance. For details, see Table 1 below. The calculated LOEC for post hatch survival was determined to be > 0.0600 µg test item/L, and the NOEC was ≥ 0.0600 µg test item/L.

Table 1: Post hatch survival of zebrafish (Danio rerio) exposed to Matrilox LP101M

mean measured concentration

Hatching succsess

Post hatch Survival

overall mortality

[µg/L]

[%]

[%]

[%]*

Control

100

96.0

4.0

Solvent Control

100

97.0

3.0

0.0243

100

96.0

4.0

0.0328

100

95.0

5.0

0.0600

100

93.0

7.0

*incl. coagulated eggs, dead fish embryos and dead fish larvae

Values refer to corrected (for recovery of analytical method, see text below for details) mean measured test concentrations.

Length:       

Mean total length for the control, solvent control and all test item concentration fish 30 days post hatch was between 11.7 and 12.95 mm. The calculated LOEC for body length was determined to be > 0.0600 µg test item/L, and the NOEC was ≥ 0.0600 µg test item/L.

Concerning the mean length, no distinct differences were observed when comparing fish exposed to the three test item concentrations with fish from the pooled control. In the highest nominal test concentration of 0.0600 µg test item/L, fish of replicate 3 had the highest average length although they were exposed to the highest arithmetic mean concentration compared to the other 3 replicates within this test concentration. For details on biological results, see Table B-1, IUCLID section "Overall remarks, attachments"; for details on analytical data see Table 2 below (analytical summary) and Table-A-1, IUCLID section "Overall remarks, attachments" for analytical details on the highest test item concentration of nominal 0.12 µg/L.

Wet Weight:       

Mean wet weight per fish for the control was 13.3 mg; the mean wet weight per fish in the solvent control was 14.6 mg. The mean wet weight of larvae in the test item treatment groups ranged between 14.9 mg and 15.5 mg. The calculated LOEC for body wet weight was determined to be > 0.0600 µg test item/L, and the NOEC was ≥ 0.0600 µg test item/L.

Comparing the mean wet weight of fish exposed to the three test item concentration with fish from the negative and the solvent control, no distinct differences were obvious. In the highest nominal test concentration of 0.0600 µg test item/L, fish of replicate 3 had the highest average wet weight although they were exposed to the highest arithmetic mean concentration compared to the other 3 replicates within this test concentration. For details on biological results, see Table B-1, IUCLID section "Overall remarks, attachments"; for details on analytical data see Table 2 below (analytical summary) and Table-A-1, IUCLID section "Overall remarks, attachments" for analytical details on the highest test item concentration of nominal 0.12 µg/L.

Dry Weight:       

Mean dry weight per fish for the control was 2.8 mg; the mean dry weight for fish in the solvent control was 3.2 mg. The mean wet weight of larvae in the test item treatment groups ranged between 3.2 and 3.4 mg. The calculated LOEC for body dry weight was determined to be > 0.0600 µg test item/L, and the NOEC was ≥ 0.0600 µg test item/L.

Concerning the mean dry weight, also no distinct differences were observed between fish exposed to the three test item concentrations and fish from the pooled control. In the highest nominal test concentration of 0.0600 µg test item/L, fish of replicate 3 had the second highest average dry weight although they were exposed to the highest arithmetic mean concentration compared to the other three replicates within this test concentration. For details on biological results, see Table B-1, IUCLID section "Overall remarks, attachments"; for details on analytical data see Table 2 below (analytical summary) and Table-A-1, IUCLID section "Overall remarks, attachments" for analytical details on the highest test item concentration of nominal 0.12 µg/L.

Appearance of the Test Item in Test Medium:       

There were no remarkable observations.

Analytical Results

Limit of Detection:

0.004 μg test item/L;

Limit of Quantification:

0.05 μg test item/L after enrichment by factor 2.5 (corresponding to fortification level of nominal 0.02 μg test item/L); Mean recovery: 103% (n = 4, RSD 6%, see Table A-3)

Mean Recovery in the Test

Samples:

39% (n = 70, RSD 60%); for details on recovery for the highest nominal test item concentration of 0.12 µg/L, see Table A-1, IUCLID section "Overall remarks, attachments".

Mean Recovery in the DMF stock solutions:

81% (n = 6, RSD 21%); For details see Table A-2, IUCLID section "Overall remarks, attachments".

Table 2. Summary of Analytical Results

Sample Description [µg test item/L]

% of nominal1

RSD [%]

n

mean measured concentration
[µg test item/L]2

corrected mean measured concentration
[µg test item/L]3

Solvent Control

n.a.

n.a.

23

n.a.

n.a.

Control

n.a.

n.a.

24

n.a.

n.a.

0.05

41

61

22

0.0204

0.0243

0.08

34

53

23

0.0276

0.0328

0.12

42

57

23

0.0504

0.0600

0.12 R3

50

59

6

0.0602

0.0717

0.12 R3 (w/o 30 dph)

59

24

5

0.0703

0.0836

(1) mean value of all measured samples per treatment group

(2) The results represent rounded values

(3) Mean measured concentration were corrected by a factor of 1.19 (100%/84%) with respect to the average recoveries achieved by the applied method for the fortified samples (84%.)

RSD: relative standard deviation per treatment group; n: number of analysed samples; n.a.: not applicable

R3: Replicate 3 (aquarium 3) of the test concentration of nominal 0.12 µg test item/L (w/o 30 dph: determined concentration of 30 days post hatch not considered)

Validity Criteria of the Analytical Part

According to the study plan, the method was validated according to GLP based on the criteria set forth by SANCO/3029 (SANCO/3029/99 rev.4 11/07/00).

Specificity:

No significant (< 30%) interference of total peak area for the target analyte was found.

Linearity:

Calibration Range:

0.02 – 0.45 μg test item/L (first analysis)

0.05 – 0.45 μg test item/L (second and third analysis)

0.035 – 0.45 μg test item/L (fourth analysis)

0.025 – 0.35 μg test item/L (fifth analysis)

Linearity of Response:

Correlation of peak area of different standard solutions with their corresponding concentrations, using a linear regression.

Regression Coefficients: r = 0.9936 (at least)

Example of Calibration Curve:

y = 1158390 * x – 15401

Accuracy and Precision:

Mean Recovery Rates in the Fortified Samples: 84% (n = 17, RSD 16%). The values found for the precision (RSD) and for the accuracy (mean recovery rate) are acceptable (for details see Table A-3, IUCLID section "Overall remarks, attachments").

Due to recovery values of 84% for the applied method, final arithmetic mean concentrations of the test item concentrations were corrected by factor 1.19 (100%/84%) to accurately reflect the performance of the applied method and to avoid underestimation of the determined test concentrations (see Table 2 above).

Conclusion: The validity criteria for the analytical method have been met.

Discussion       

Analytical results:       

The test item recoveries in all test concentrations showed a high fluctuation after test start with a decreasing tendency towards test end (please see Table A-1, IUCLID section "Overall remarks, attachments" for details on highest nominal concentration of 0.12 µg/L). Since the dosage system of the test item worked correctly and almost nominal recoveries were determined in selected DMF stock solutions (see Table A-2, IUCLID section "Overall remarks, attachments"), the application of the test item was done appropriately.

The fluctuating concentrations are likely an effect of the test items high log Kow and its property of being highly biodegradable. With continuing test duration, increasing amounts of biomass and microorganism are existent in the aquaria due to growing fish, excrements and fish food. The high log Kow promotes adsorption of the test item on biomass surface. Thus, the low recovery of the test item is likely due to the presence of biomass and probably caused by adsorption as well as biodegradation losses.

These factors seemed to have a strong negative influence on analytical test item recovery despite of the fact that all test item concentrations were renewed fivefold per day and the aquaria were kept as clean as possible (as low surplus food as possible and regular cleaning of the aquaria) without stressing the fish. Higher test water renewal rates would have been disadvantageous for fish hatching and survival, especially at the presence of organic solvents such as DMF, which was shown in pre-experiments at the testing facility. Therefore, the fivefold-renewal rate was a reasonable compromise to keep as much test item as possible available on the one side and to have optimal conditions for the fish on the other side.

Biological results:       

The arithmetic mean measured concentration of 0.0600 µg test item/L determined for the highest test concentration of nominal 0.12 µg test item/L was slightly below the determined solubility of the test item of 0.078 µg test item/L (Sonntag 2017). Nevertheless, analytical dose verification confirmed test item availability for all test concentrations throughout the test, especially during the critical hatching phase (please see details for highest nominal test item concentration of 0.12 µg/L in Table A-1, IUCLID section "Overall remarks, attachments"). Emerging biological effects are not expected for concentrations above 0.0600 µg test item/L up to the determined solubility of 0.078 µg test item/L. This expectation is underlined by replicate 3 of the highest test concentration of nominal 0.12 µg test item/L (see Table 2 above). Here, the highest arithmetic mean concentration of 0.0717 µg test item/L (0.0836 µg test item/L until 24 days post hatch) was determined which corresponds to 92% of the test item solubility (107% considering only determined concentrations up to and including 24 days post hatch). No distinct differences to other replicates of this test concentration were obvious, neither for dry / wet weight or fish length (see Table B-1, IUCLID section "Overall remarks, attachments"), nor for hatching success or post hatch survival. Therefore, it seems reasonable to state that NOEC and LOEC values for hatching success, mortality, wet and dry weight as well as length are very likely to be ≥ 0.078 µg test item/L corresponding to the water solubility of the test item.

Conclusion

Up to the saturation level in test medium (corresponding to the mean measured concentration of 0.060 µg test item /L) not any toxic effect of the test item relative to controls could be observed:

The NOEC for hatching success was determined to be ≥ 0.0600 µg test item/L and the LOEC > 0.0600 µg test item/L.

Based on the test results, the NOEC for 30 days post hatch (DPH) survival was calculated  to be ≥ 0.0600 µg test item/L and the LOEC > 0.0600 µg test item/L.

For body length, the NOEC was calculated to be ≥ 0.0600 µg test item/L and the LOEC was calculated to be > 0.0600 µg test item/L.

For body wet weight, the NOEC was calculated to be ≥ 0.0600 µg test item/L and the LOEC was calculated to be > 0.0600 µg test item/L.

For body dry weight, the NOEC was calculated to be ≥ 0.0600 µg test item/L and the LOEC was calculated to be > 0.0600 µg test item/L.

No effect on mortality, body length, body dry and body wet weight occurred. Therefore no LCx or ECx values were determined.

In the analytical part, prior to initiation of the exposure period and during the test, the test item concentrations were determined in regular intervals to characterise exposure. All reported results refer to arithmetic mean concentrations, since the test item concentrations were not within ± 20% of the nominal concentrations during the test.

References:

SANCO/3029/99 rev.4 11/07/00: Residues: Guidance for generating and reporting methods of analysis in support of pre-registration data requirements for Annex II (part A; Section 4) and Annex III (part A; Section 5) of directive 91/414

Validity criteria fulfilled:
yes
Conclusions:
Up to the saturation level in test medium (corresponding to the mean measured concentration of 0.060 µg test item /L) not any toxic effect of the test item on Danio rerio early life stages could be
observed relative to controls (OECD 210; GLP; 2018):
NOEC (30 d post hatch; overall survival, length, body dry/wet weight) >= saturation level in water;
LOEC (30 d post hatch; overall survival, length, body dry/wet weight) > saturation level in water.
Executive summary:

The purpose of this study was to evaluate the toxicity of the test item Matrilox LP101M (Trimethylolpropane trinonanoate) to the early-life stages of fish. For this purpose, fertilised eggs of Zebrafish (Danio rerio) were exposed in a flow-through test to aqueous test media containing the test item at various concentrations under defined conditions. The test duration was until 30 days after hatching. The recorded effects were mortality, hatching, growth, weight and sublethal effects of the fish.

The method used is recommended by the test guidelines, and also Zebrafish is one of the fish species recommended by the international test guidelines of the OECD and EC.

The purpose of the analytical part of this study was to verify the concentrations of the test item in the test medium. The test was performed according to the following guidelines, compliant with GLP:

-       OECD Guideline for Testing of Chemicals, Section 2, No. 210 "Fish, Early-life Stage Toxicity Test", adopted July 26, 2013.

-       OECD Series on Testing and Assessment, No. 23, "Guidance Document on Aquatic Toxicity Testing of Difficult Substances and Mixtures", December 15, 2000

This study encompassed 3 treatment groups (3 concentrations of the test item), a control and a solvent control, with 4 replicates each containing 25 freshly fertilized eggs. The eggs and larvae were observed daily for sublethal effects and mortality. Dead larvae were removed at least once daily and discarded. The test item concentrations in the test water taken two days before insertion of eggs, at start (insertion of eggs) and after 3, 10, 17, 24 and 30 days post hatch were analysed.

To determine the most appropriate test item concentrations for the definite test, a range-finding test was performed. Further, the solubility of the test item Matrilox LP101M in the test water (ISO-medium) was determined via a separately conducted GLP-study to a value of 0.078 µg/L (Sonntag, F, 2017). Based on results from the preliminary test and the determined water solubility of 0.078 µg/L, no toxicity of the test item up to the saturation concentration in water was expected. Both, OECD 210 as well as OECD No. 23 emphasize that it is highly important to conduct aquatic toxicity testing on truly dissolved test item solutions only, i.e. up to the saturation level in water, but not beyond. Accordingly, the determined saturation level in ISO medium of 0.078 µg/L was used as the upper limit for aquatic toxicity testing on fish, multiplied with a safety factor of 1.5 as a reasonable compromise between trying to avoid a relevant non-dissolved fraction on the one hand and minimizing the risk of failing a saturated solution in case of analytical recoveries somewhat below the nominal concentration. This safety factor was based on close to nominal recoveries in the range-finding test. As such, an upper nominal test item concentration of (rounded) 0.12 µg/L was derived. Testing five test item concentrations with a spacing factor of 3.2 as recommended according to OECD 210 would have resulted in testing concentrations pronouncedly below a nominal test item concentration of 0.05 µg/L and thus would have been out of scope of the analytical quantification method used for verification of test item concentrations during the course of the study. Therefore, it was decided that an extended limit test as foreseen according to OECD 210, paragraph 22 would be most appropriate in this case. Using a lower spacing factor of 1.5 and three test item concentrations resulted in actually applied nominal test item concentrations of 0.12, 0.08, 0.05 µg test item/L. While reducing the normally required 5 test item concentrations to three, analytical monitoring was extended beyond the minimum required according to OECD 210 (weekly, at least in one replicate per concentration level), in that regularly all four replicates per concentration level were analysed.

The quantification of the test item Matrilox LP101M was performed using liquid-liquid extraction with n-Hexane followed by the analysis via liquid chromatography (HPLC) with MS/MS detection. Two days before the start of the test (equilibration check of the flow through system), the mean test recoveries of the nominal test concentrations varied between 75 and 99% (all test concentrations considered). At the start of the test (day of egg insertion = DAI0), test item recoveries varied between 38 and 94% (all test concentrations considered). The further determined recoveries throughout the test until test end varied between 8 and 97% with a decreasing tendency towards test end. The fluctuating concentrations are likely an effect of the high log Kow of the test item and its property of being highly biodegradable. Due to deviation of analytically determined concentrations by more than ± 20% from nominal, all (no)effect concentrations are given based on arithmetic mean measured values.

Results:

Up to the saturation level in test medium (corresponding to the mean measured concentration of 0.060 µg test item /L) not any toxic effect of the test item relative to controls could be observed:

- The NOEC for hatching success was determined to be ≥ 0.0600 µg test item/L and the LOEC > 0.0600 µg test item/L.

- Based on the test results, the NOEC for 30 days post hatch (DPH) survival was calculated  to be ≥ 0.0600 µg test item/L and the LOEC > 0.0600 µg test item/L.

- For body length, the NOEC was calculated to be ≥ 0.0600 µg test item/L and the LOEC was calculated to be > 0.0600 µg test item/L.

- For body wet weight, the NOEC was calculated to be ≥ 0.0600 µg test item/L and the LOEC was calculated to be > 0.0600 µg test item/L.

- For body dry weight, the NOEC was calculated to be ≥ 0.0600 µg test item/L and the LOEC was calculated to be > 0.0600 µg test item/L.

No effect on mortality, body length, body dry and body wet weight occurred. Therefore no LCx or ECx values were determined.

While analytically determined mean measured concentration over all replicates for the highest nominal concentration (0.12 µg/L) of test item was somewhat below the determined solubility for the test medium (0.078 µg/L), the determined arithmetic mean measured concentration for replicate 3 over all days (0.072 µg/L) corresponds to 92% of the water solubility, while arithmetic mean measured concentration for replicate 3 up to and including day 24 post hatch (28 days of exposure) (0.084 µg/L) was even slightly above the saturation level of the test item in ISO medium (107% of the water solubility). The time span till day 24 post hatch covers the most sensitive develpmental phase. No distinct differences to other replicates of this test concentration were obvious, neither for dry / wet weight or fish length, nor for hatching success or post hatch survival. This corroborates that up to the saturation level in water, no effects of the test item on fish early life stages occurred.

Description of key information

Up to the saturation level in test medium (corresponding to the mean measured concentration of 0.060 µg test item /L) not any toxic effect of the test item on Danio rerio early life stages could be

observed relative to controls (OECD 210; GLP; 2018):

NOEC (30 d post hatch; overall survival, length, body dry/wet weight) >= saturation level in water;

LOEC (30 d post hatch; overall survival, length, body dry/wet weight) > saturation level in water.

Because NOEC / LOEC values are solely determined by the very low water solubility of the test item (no toxicity observed up to the water solubility limit) and are thus "greater than" values, no key values for chemical safety assessment are given below, because no hazard was identified.

Key value for chemical safety assessment

Additional information

One reliable key study is available for the endpoint long-term toxicity to fish.

The purpose of this study was to evaluate the toxicity of the test item Matrilox LP101M (Trimethylolpropane trinonanoate) to the early-life stages of fish. For this purpose, fertilised eggs of Zebrafish (Danio rerio) were exposed in a flow-through test to aqueous test media containing the test item at various concentrations under defined conditions. The test duration was until 30 days after hatching. The recorded effects were mortality, hatching, growth, weight and sublethal effects of the fish.

The method used is recommended by the test guidelines, and also Zebrafish is one of the fish species recommended by the international test guidelines of the OECD and EC.

The purpose of the analytical part of this study was to verify the concentrations of the test item in the test medium. The test was performed according to the following guidelines, compliant with GLP:

-       OECD Guideline for Testing of Chemicals, Section 2, No. 210 "Fish, Early-life Stage Toxicity Test", adopted July 26, 2013.

-       OECD Series on Testing and Assessment, No. 23, "Guidance Document on Aquatic Toxicity Testing of Difficult Substances and Mixtures", December 15, 2000

This study encompassed 3 treatment groups (3 concentrations of the test item), a control and a solvent control, with 4 replicates each containing 25 freshly fertilized eggs. The eggs and larvae were observed daily for sublethal effects and mortality. Dead larvae were removed at least once daily and discarded. The test item concentrations in the test water taken two days before insertion of eggs, at start (insertion of eggs) and after 3, 10, 17, 24 and 30 days post hatch were analysed.

To determine the most appropriate test item concentrations for the definite test, a range-finding test was performed. Further, the solubility of the test item Matrilox LP101M in the test water (ISO-medium) was determined via a separately conducted GLP-study to a value of 0.078 µg/L (Sonntag, F, 2017). Based on results from the preliminary test and the determined water solubility of 0.078 µg/L, no toxicity of the test item up to the saturation concentration in water was expected. Both, OECD 210 as well as OECD No. 23 emphasize that it is highly important to conduct aquatic toxicity testing on truly dissolved test item solutions only, i.e. up to the saturation level in water, but not beyond. Accordingly, the determined saturation level in ISO medium of 0.078 µg/L was used as the upper limit for aquatic toxicity testing on fish, multiplied with a safety factor of 1.5 as a reasonable compromise between trying to avoid a relevant non-dissolved fraction on the one hand and minimizing the risk of failing a saturated solution in case of analytical recoveries somewhat below the nominal concentration. This safety factor was based on close to nominal recoveries in the range-finding test. As such, an upper nominal test item concentration of (rounded) 0.12 µg/L was derived. Testing five test item concentrations with a spacing factor of 3.2 as recommended according to OECD 210 would have resulted in testing concentrations pronouncedly below a nominal test item concentration of 0.05 µg/L and thus would have been out of scope of the analytical quantification method used for verification of test item concentrations during the course of the study. Therefore, it was decided that an extended limit test as foreseen according to OECD 210, paragraph 22 would be most appropriate in this case. Using a lower spacing factor of 1.5 and three test item concentrations resulted in actually applied nominal test item concentrations of 0.12, 0.08, 0.05 µg test item/L. While reducing the normally required 5 test item concentrations to three, analytical monitoring was extended beyond the minimum required according to OECD 210 (weekly, at least in one replicate per concentration level), in that regularly all four replicates per concentration level were analysed.

The quantification of the test item Matrilox LP101M was performed using liquid-liquid extraction with n-Hexane followed by the analysis via liquid chromatography (HPLC) with MS/MS detection. Two days before the start of the test (equilibration check of the flow through system), the mean test recoveries of the nominal test concentrations varied between 75 and 99% (all test concentrations considered). At the start of the test (day of egg insertion = DAI0), test item recoveries varied between 38 and 94% (all test concentrations considered). The further determined recoveries throughout the test until test end varied between 8 and 97% with a decreasing tendency towards test end. The fluctuating concentrations are likely an effect of the high log Kow of the test item and its property of being highly biodegradable. Due to deviation of analytically determined concentrations by more than ± 20% from nominal, all (no)effect concentrations are given based on arithmetic mean measured values.

Results:

Up to the saturation level in test medium (corresponding to the mean measured concentration of 0.060 µg test item /L) not any toxic effect of the test item relative to controls could be observed:

- The NOEC for hatching success was determined to be ≥ 0.0600 µg test item/L and the LOEC > 0.0600 µg test item/L.

- Based on the test results, the NOEC for 30 days post hatch (DPH) survival was calculated  to be ≥ 0.0600 µg test item/L and the LOEC > 0.0600 µg test item/L.

- For body length, the NOEC was calculated to be ≥ 0.0600 µg test item/L and the LOEC was calculated to be > 0.0600 µg test item/L.

- For body wet weight, the NOEC was calculated to be ≥ 0.0600 µg test item/L and the LOEC was calculated to be > 0.0600 µg test item/L.

- For body dry weight, the NOEC was calculated to be ≥ 0.0600 µg test item/L and the LOEC was calculated to be > 0.0600 µg test item/L.

No effect on mortality, body length, body dry and body wet weight occurred. Therefore no LCx or ECx values were determined.

While analytically determined mean measured concentration over all replicates for the highest nominal concentration (0.12 µg/L) of test item was somewhat below the determined solubility for the test medium (0.078 µg/L), the determined arithmetic mean measured concentration for replicate 3 over all days (0.072 µg/L) corresponds to 92% of the water solubility, while arithmetic mean measured concentration for replicate 3 up to and including day 24 post hatch (28 days of exposure) (0.084 µg/L) was even slightly above the saturation level of the test item in ISO medium (107% of the water solubility). The time span till day 24 post hatch covers the most sensitive develpmental phase. No distinct differences to other replicates of this test concentration were obvious, neither for dry / wet weight or fish length, nor for hatching success or post hatch survival. This corroborates that up to the saturation level in water, no effects of the test item on fish early life stages occurred.