Registration Dossier

Administrative data

Key value for chemical safety assessment

Genetic toxicity in vitro

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (negative)

Genetic toxicity in vivo

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (negative)

Additional information

Additional information from genetic toxicity in vivo:

The discussion includes a summary of the data as well as additional genotoxicity modeling conducted to demonstrate that read across is appropriate and genotoxicity is not expected.

In vitro studies:

Ames test - In this key study for in vitro genetic toxicity (Machadoet al, 1985, report number: SOCAL 2321) there was no guideline specified, however it was considered to be comparable to OECD Guideline 471 (Bacterial Reverse Mutation Assay). The study was conducted in line with GLP. A reliability rating of 1 according to the criteria of Klimisch, 1997.

Two additional Ames tests are included (Lawlor, 1997 and Machado et al., 1986) for surrogate test materials to cover additional strains not tested in the key study (Machado et al., 1985).

Mouse lymphoma- In this key study for in vitro genetic toxicity (Winingeret al, 1985, report number: SOCAL 2322) there was no guideline specified, however it was considered to be comparable to OECD Guideline 476 (In vitro Mammalian Cell Gene Mutation Test). The study was conducted in line with GLP. A reliability rating of 1 according to the criteria of Klimisch, 1997.

In vivo study:

In the key study for in vivo genetic toxicity (Ivett, 1997, Corning Hazleton report number: 17865-0-455CO) the study was conducted according to OECD Guideline 474 (Mammalian Erythrocyte Micronucleus Test). The study was conducted in line with GLP.

The reliability rating for this study is 1, however this is being used as read across from a supporting substance (Phenol, tetrapropenyl-, sulfurized, calcium salts CAS No. 220794 -90 -1) as there was no available data to fulfil this endpoint for the test material and so the reliability rating will be reduced to 2, according to the criteria of Klimisch, 1997.

The following information is taken into account for any hazard / risk assessment:

There are 2 key endpoints required for in vitro genetic toxicity:

Ames test - The test material was tested in the histidine-deficient strains of Salmonella typhimurium TA98, TA100, and TA102 and in the tryptophan-deficient strain of Escherichia coli WP2 uvrA at dose levels of 0.033 to 3.33 mg/plate with and without metabolic activation provided by Aroclor-induced rat liver S-9. The test material was suspended in 25% Pluronic 127 (w/w in ethanol). The suspension was miscible with the top agar. It was not cytotoxic at any concentration tested.

Under the conditions tested, the test material was not mutagenic to TA98, TA100, TA102, or E. coli WP2 uvrA.

Mouse lymphoma - The test material was tested in the L5178Y TK+/- Mutagenicity Screen with and without S-9 metabolic activation. The cultures with activation were tested at concentrations ranging from 75 μg/ml to 275 μg/ml; cultures without activation were tested at concentrations ranging from 60 μg/ml to 110 μg/ml.

The results indicated that the test material did not induce a significant increase in the mutant frequencies of cultures tested either with or without metabolic activation. Under the conditions tested the test material was not mutagenic in this screen.

In vivo:

Based on the results of the dose selection study, the maximum tolerated dose was estimated as >5000 mg/kg.

The test article did not induce a statistically significant increase in micro-nuclei in bone marrow polychromatic erythrocytes under the conditions of this assay and is considered negative in the mouse micronucleus assay.

To further evaluate the genotoxicity potential as well as justify the applied read across and determine whether calcium carbonate overbasing impacts genotoxicity, OASIS TIMES v.2.27.15.146 was used to predict whether chromosomal aberrations may occur. The suitability of OASIS TIMES was reviewed in a JRC Scientific and Technical Report (Serafimova, Gatnik, and Worth, 2010) where it was found to have advantages over other predictive models for genotoxicity as it incorporates metabolism.

 

To start, UVCB G Graph 1.0 was used to create a Generic SMILES to incorporate all variations possible in the UVCB nature EC 272-234-3/CAS 220794-90-1 – this substance is an alkyl phenate sulfide with no overbasing. Over 1500 isomers were predicted and a filter was used to reduce the number to a manageable but representative amount of structures to be predicted by TIMES. The filter option selected was Molecular Weight with 3 intervals in order to separate the isomers based on the amount of sulfur bridging (1-3). Five members were randomly selected from each distribution group for a total of 15 constituents for final modeling. 

 

The difference between the registered substance (overbased) and the substance used for read across is the degree of calcium carbonate overbasing. Based on the principles in “Specific Rules of Ionic Characteristics of Metal Salts of Organic Chemicals” found in the OASIS software, dissociation of the carbonate group could occur based on the large differences in electronegativity resulting from the ionic bond between the calcium and oxygen ions. Based on the principle that dissociation will occur at the Ca-O bond, both the registered substance and read-across substance would be expected to dissociate to the same phenol, thiobis substance. Therefore, TIMES was modeled with and without hydrolysis selected to mimic dissociation and non-dissociation structures, respectively.

 

Running the model without dissociation (calcium carbonate present) resulted in negative predictions for the parent structures. However, the corresponding parent structures with the calcium dissociated yielded metabolites that were positive in situ for chromosomal aberration due to aromatic ring hydroxylation and the potential formation of hydroquinones and catechols and subsequent oxidation to benzoquinones. However, all of the predicted positive metabolites are below 0.05 for probability to obtain and probability to be. Therefore, the metabolites would not be predicted to be available to interact with DNA. This is consistent with the negative micronucleus study for CAS 122384-85-4 and transformation to genotoxic species does not appear to be relevant in an animal model. 

 

As stated, TIMES was run with hydrolysis turned off to show the prediction with the carbonate group attached. This resulted in negative predictions for both parents and metabolites. This illustrates that dissociation of the calcium carbonate is required for transformation to genotoxic metabolites in situ. This is likely due to the calcium carbonate preventing transformation to a quinone species. Therefore, read across from low overbased to high overbased substances could be considered conservative

Justification for classification or non-classification

The results for the key parameters chosen for genetic toxicity were negative and so the criteria set out in Directive 67/548/EEC and also Regulation (EC) no 1272/2008 do not apply, therefore classification for genetic toxicity was not considered to be necessary.