2-benzyl-4-methyl-4a,5,6,7,8,8a-hexahydro-4H-benzo[d][1,3]dioxine

Toxicity to aquatic algae and cyanobacteria

Description of key information

QSAR models, iSafeRat QSAR, key studies:

72h-ErC50 = 5.7 mg/L (95% CL: 4.8 - 67 mg/L)

72h-NOECr = 2 mg/L (95% CL: 1.6 - 2.4 mg/L)

Key value for chemical safety assessment

EC50 for freshwater algae:

5.7 mg/L

EC10 or NOEC for freshwater algae:

2 mg/L

Additional information

To assess the toxicity of the registered substance to aquatic algae, two data point are available

The first data is assessed as key study. Two Quantitative Structure-Activity Relationship (QSAR) models were used to calculate the TOXICITY TO ALGAE (72-HOUR ErC50 and NOECr) of the test item. These QSAR models have been validated to be compliant with the OECD  recommendations for QSAR modeling (OECD, 2004) and predict the endpoint values which would be expected when testing the substance under experimental conditions in a laboratory following the Guideline for Testing of Chemicals No. 201, "Freshwater Alga and Cyanobacteria, Growth Inhibition Test" (OECD, 2006), referenced as Method C.3 of Commission Regulation No. 440/2008 (European Commission, 2008). The criterions predicted were the Median Effective Concentration (ErC50), a statistically derived concentration which is expected to cause 50% inhibition of intrinsic rate of growth of the test system and the No Observed Effect Concentration (NOECr), a tested concentration which is expected to cause no effect on intrinsic rate of growth of the test system. Both criterions were determined for a period exposure of 72 hours.

The TOXICITY TO ALGAE (72-HOUR ErC50 and NOECr) was determined using iSafeRat® algEC50 and iSafeRat® algNOEC, two validated QSAR models for the Mechanism of Action (MechoA) in question (MechoA 1.1, i.e. non-polar narcosis) (Bauer et al., 2018). The QSAR models provide in silico predictions for the 72-hour ErC50 and NOECr values that can effectively be used in place of an experimentally derived results. The QSAR models are based on validated data for a training set of 40 chemicals derived from 72-hour ErC50 and 31 chemicals derived from 72-hour NOECr test on algae, for which the concentrations of the test item had been determined by chemical analyses over the test period.

The QSAR models used to achieve the study have been fully validated following the OECD recommandations (OECD, 2004). The test item falls within the applicability domain of the models and was therefore reliably predicted for its TOXICITY TO ALGAE (72-HOUR ErC50 and 72-HOUR NOECr).Therefore, this endpoint value can be considered valid for use in risk assessment and classification and labelling.

The TOXICITY TO ALGAE (72-HOUR ErC50 and NOECr) of the test item was respectively predicted as 5.7 mg/L and 2.0 mg/L.
95% confidence interval (α = 0.05) for 72h-ErC50: 4.8 – 6.7 mg/L.
95% confidence interval (α = 0.05) for 72h-NOECr: 1.6 – 2.4 mg/L.

 

The second data (phytosafe, 2012)  is assessed as supporting data point and is an experimental study. In an algal growth inhibition study performed according to OECD Guideline 201 and in compliance with GLP, the freshwater green algae species Desmodesmus subspicatus was first exposed to test item for 72 h, under semi-static conditions and constant illumination, shaking at a temperature of 21-24 °C. A range-finding study was performed using one replicate for each of the five test item treatments between 0.01-100 mg/L and the test range was selected on the basis of the results and definitive test was performed using three replicate units for each of five test item treatments and for controls. The percent inhibition for growth rate and yield was determined.

 

In range finding study no adverse effects were observed for the test item treatments up to and including 10 mg/L. The algae were totally inhibited at 100 mg/L and in definitive test The specific growth rate was regular in controls throughout the test period and standard deviation between the three replicate units did not exceed 10 % of the mean value at each sampling time and was less than 2% for the average specific growth rate per day over the entire test period and was reduced for three highest test treatments. The regression was performed using three highest test item treatment two lowest values were excluded because of pleateu phase. The coefficient of determination was 94.8 %.

 

Yield in biomass was considered as similar to the controls for the two lowest test item treatments and significantly reduced at 13.5, 20.1, and 30.0 mg/L. The regression analysis was performed using the four highest test treatments. The lowest test item treatments were excluded because of the plateau phase. Coefficient of determination for the regression analysis was 97.2 %.

 

Under the test conditions, the ErC50 (growth rate) was 22.7 mg/L (95% CL: 3.3-157.4 mg/L); and EyC50 (yield) was 17.1 (95% CL: 11.8-24.9 mg/L) of the test item to Desmodesmus subspicatus. The NOEC was 9.0 mg/L based on measured concentrations.

 

All validity criteria were fulfilled. This study is considered reliable with restrictions as a high 95% CL was determined for the ErC50 value and acetone was used as solvent in this study. Because of the potential for interaction with the test chemical resulting in an altered response in the test, solvent use should be restricted to situations where no other acceptable method of test solution preparation is available. The use of solvent is not the best method at the time being. The concentration/quantity of solvent used in the treatment solutions was 0.5 mL/L, corresponding to 395 mg/L (with a density of 0.79), which is 5 times higher than the recommended maximum level of solvent (below 0.1 mL/L; OECD No. 23) but is below the NOEC of acetone (which was reported in the ECHA disseminated dossier at 530 mg/L).