Myrtus communis, ext.

Toxicity to aquatic algae and cyanobacteria

Description of key information

Myrtle oil is a Natural Complex Substance (UVCB) with a well-defined composition for which the relative percentage and its reported variation of each constituent is known. The acute toxicity to aquatic algae property has been investigated using an in-house calculation method that mimics an OECD 201 study and guideline for Testing of Chemicals No. 23 (i.e. WAF conditions). The acute toxicity to aquatic algae was determined using a calculation method for Mode of Action 1 (non-polar narcotics) for each individual constituent present in the mixture. This algorithm is based on a QSAR model which has been validated to be compliant with the OECD recommendations for QSAR modeling (OECD, 2004). Further to this the lethal loading rate of the WAF is determined by using a series of calculation steps using phase equilibrium thermodynamics first to determine the analyzable fraction (the concentration which should be analyzable in a WAF study) and then excluding the non-bioavailable fraction of the remaining constituents. The remaining bioavailable fraction corresponds to the lethal loading value of the mixture. This approach has been validated using data derived from 48-hour ErL50 tests on algae (OECD 201 study and the OECD guidance document on toxicity testing for difficult substances and mixtures No. 23, i.e. WAF conditions). Two theoretical compositions have been investigated, the “typical” composition proposed by the Lead registrant and a theoretical worst case composition that maximizes the concentration of the most toxic constituents. The 72-h ErL50 was 5.1 mg test material/L for the typical composition of Myrtle oil, and The 72-h ErL50 was 4.2 mg test material/L for the theoretical worst case composition. ErL50 were based on growth rate.  

Key value for chemical safety assessment

EC50 for freshwater algae:

4.2 mg/L

Additional information

Myrtle oil is a Natural Complex Substance (UVCB) with a well-defined composition for which the relative percentage of each constituent is known. Therefore, it has been decided that the ecotoxicity of Myrtle oil will be derived from knowledge of the constituents; constituents approach. The mixture ecotoxicity properties may be derived from the ecotoxicity of the individual constituents (table 1) using the CLP additivity calculation approach.

However,CLP additivity approach is calculated on the basis that all the substances are at their maximum solubility andit has been observed that CLP additivity calculations for mode of action 1 compounds are unreasonably conservative when compared to classic WAF studies. This has been proved in a number of cases for natural complex substances. Indeed, natural extract compositions are a mixture of hydrophilic alcohol molecules and hydrophobic terpene molecules. Therefore, when a WAF is performed most of substances fully dissolved in the aqueous phase are the hydrophilic fraction while the hydrophobic fraction (the more toxic elements for MOA 1 substances) may be below their water solubility value.

The acute toxicity to algae property for Myrtle oil has been investigated using an in-house calculation method (iSafeRat® WAF module for mixture toxicity calculation) that mimics an OECD 201 study and guideline for Testing of Chemicals No. 23 (i.e. WAF conditions). Two theoretical compositions have been investigated, the “typical” composition proposed by the Lead registrant and a theoretical worst case composition that maximizes the concentration of the most toxic constituents.The acute toxicity to algae was determined using a calculation method for Mode of Action 1 (non-polar narcotics) for each individual constituent present in the mixture.

 

The first step of the iSafeRat mixture toxicity calculation employs phase equilibrium thermodynamics in order to determine the concentrations of each constituent within the WAF. This fraction equates to the analysable fraction of a WAF study. In the calculation the second step is to remove this non-bioavailable fraction. Within the WAF, the constituents also partition between themselves further reducing the bioavailable fraction and thus the toxicity of the mixture compared to the individual constituents.These two reasons explain why ecotoxicity values from WAF studies are always higher for non-polar narcotic mixtures than the calculated values from CLP additivity calculation.

 

The final step is to determine the truly bioavailable fraction of the WAF per constituent. The ErC50s of each constituent are already known from literature or predicted using the iSafeRat® QSAR model (table 1). Each value has been included as a supporting study in the IUCLID. An additivity approach (based on Chemical Activity of each constituent) is used in order to calculate the Effective Loading rate of the WAF.

 

Table1Myrtle oil, constituents data used to derive Myrtle oil Ecotoxicity value.

constituent	typical composition (%)	worst case concentration (%)	72h-algae ErC50 (mg/L)	72h-algae ErC50 reference
α-pinene	26.00	35.00	> 0.494	experimental data
cineol 1,8	29.00	25.00	> 74	experimental data
myrtenyl acetate	13.00	20.00	7.19 [4.67 – 11.08]	iSafeRat® prediction
limonene	11.00	15.00	1.13 [0.72 – 1.76]	iSafeRat® prediction
linalool	4.60	2.20	60.40 [44.21 – 82.52]	iSafeRat® prediction
terpineol	3.50	2.00	66.85 [48.61 – 91.92]	iSafeRat® prediction
para-cymene	0.40	0.40	1.70 [1.13 – 2.55]	iSafeRat® prediction
β-pinene	0.40	0.40	0.45 [0.26 – 0.78]	iSafeRat® prediction

 

Using this approach, the 72-h ErL50 for aquatic algae was 5.1 mg test material/L for the typical composition of Myrtle oil, and the 72-h ErL50 was 4.2 mg test material/L for the theoretical worst case composition. These ErL50 were based on growth rate.

Validation document of iSafeRat® is attached to the dossier.

 

Based on the results of this study, Myrtle oil would not be classified as acutely toxic to aquatic organisms in accordance with the classification of the CLP.

This toxicity prediction has been validated and is considered acceptable to fulfill the aquatic algae toxicity endpoint.