Registration Dossier

Administrative data

Key value for chemical safety assessment

Genetic toxicity in vitro

Description of key information

Despite negative or equivocal results in in-vitro and in-vivo mammalian assays, the positive modified Ames test results for the thermally cracked gas and coker gas oils, cracked gas oils are considered to demonstrate clear genotoxic potential for cracked gas oils.

Endpoint conclusion
Endpoint conclusion:
adverse effect observed (positive)

Additional information

Additional information from genetic toxicity in vitro:

Despite negative or equivocal results in in-vitro and in-vivo mammalian assays, no conclusions can be drawn from the studies, owing to the questionable reliability of the assays and the inconsistent findings. However, the evidence from the modified Ames assay by Deininger et al. (1991) indicates clear genotoxic potential for certain cracked gas oils.

In vitro gene mutation in bacteria

A modified Ames bacterial reverse mutation assay in S. typhimurium TA 98 was conducted by Deininger et al. (1991; Klimisch score = 2). Mutagenic activity of complex aromatic hydrocarbon mixtures such as mineral oils was inadequately detected by the standard Ames assay. Consequently, modification of the assay was needed. An optimised assay was developed, in which a DMSO extract of the oil instead of the whole oil was tested. DMSO is able to extract the principle carcinogenic components (polycyclic aromatic hydrocarbons) from oils, and allows them to be tested without other ingredients interfering with the mutagenic response. As some components of oils were found to inhibit PAC metabolism, the metabolic activation system was modified by increasing the S9 concentration 8-fold and doubling of the NADP co-factor concentration. Hamster S9 instead of rat S9 is used in this assay and only the most sensitive strain of bacteria for PACs (TA98) is used. Deininger found positive results for samples of coker gas oil (CAS# 101316-59-0) and thermally cracked gas oil (CAS# 64741-82-8). Doses included 0, 1, 3, 5, 7, 10, 15, 20, 25, 40, 40, 50, or 60 µL/plate. Mutagenicity indices for relevant oils were 2.1, 3.1, 4.0, and 9.3, with and without activation, indicating a positive result. All samples had a 3-7 ring PAC content of more than 1%. These results suggest that genotoxic components may be present at significant concentrations in thermally cracked gas oils, coking gas oils and in catalytic cracked gas oils. The findings suggest that levels were lower with thermally cracked gas oils than with coker gas oils.

 

In vitro gene mutation in mammalian cells

The American Petroleum Institute (API, 1985d; Klimisch score = 2)completed a gene mutation study in the mouse lymphoma cell line L5178Y TK+/-, with and without metabolic activation, at doses of 5–80 nL/mL light catalytically cracked distillate (CAS# 64741-59-9). In this study positive findings were recorded from 10-30 nl/ml in the presence of S9. There was a clear dose response with mutant frequencies being between 3 and 5 times those of the control group. Survival was 78.5% relative to the control at the lowest concentration producing a positive result and 10% at the highest. No evidence of any increase was seen in the absence of S9.

In a further study by API (API, 1985e;Klimisch score = 3) also tested light catalytically cracked distillate with the same mouse lymphoma cell line. This test yielded ambiguous results, with and without metabolic activation, but was considered positive. In the presence of S9, a 2.5 fold increase in mutant frequency was only seen at the top concentration 0.021 µL/mL (21 nL/mL) showing only 11% survival. In the absence of S9, when examined alongside the previous study, the first trial showed an increase in mutant frequency at a concentration of 0.032 µL/mL (32 nL/mL) which resulted in only 1% survival. While it was claimed that the repeat study showed an increase in mutant frequency at a dose level of 0.03 µL/mL (30 nL/mL) with a survival of 16.5%, this appeared to be due to low solvent control values (half those of the first study). In any event these results raise questions about survival artefacts and are questionable.

 

In vitro cytogenicity in mammalian cells

Light catalytically cracked distillate was also tested in a sister chromatid exchange assay in Chinese hamster ovary (CHO) cells, both in the presence and absence of S9 (API, 1988). Marginal increases were seen without metabolic activation at 10 and 20 µg/mL in the first test, but only at 30 µg/mL in the second test. In the presence of S9, marginal increases of the same order were seen at dose levels between 10 and 80 µg/mL. These results are somewhat unreliable, due to the appearance of increases both with and without activation. 

In vivo gene mutation in mammalian cells

In a chromosome aberration study, doses of 0.2, 0.67, and 2.0 g/kg light catalytically cracked distillate (CAS# 64741-59-9) were administered to male and female Sprague-Dawley rats via intraperitoneal injection (API, 1986b; Klimisch score = 2). Bone marrow preparation and controls were the same as the above study. Negative results were obtained, with no increase in chromosomal aberrations in rat bone marrow.

In a further chromosome aberration study, doses of 0.3, 1.0, and 2.0 g/kg of light catalytically cracked distillate (CAS# 64741-59-9)were administered to male and female Sprague-Dawley rats via intraperitoneal injection (API, 1985g; Klimisch score = 2). Results were negative for genotoxicity.

 

In vivo cytogenicity in mammalian cells

In contrast, a weak positive result was reported for induction of SCE in mice following single intraperitoneal injection of a sample of light catalytically cracked distillate at doses of 340, 1700, or 3400 mg/kg body weight, four hours after implantation of a bromodeoxyuridine pellet (API, 1989c). Bone marrow cells were examined for sister-chromatid exchange (SCE) by comparison with corn oil controls. Slight increases in SCE were in both male and female mice at the mid- and high-dose levels, with the response ranging from 1.2 to 1.6 times that of the control frequency. Cyclophosphamide (a positive control) produced an increase of 5.2 times in males and 3.7 times in females. However, the positive control oil (4 g/kg) failed to produce an increase of SCEs in males and only produced a low level significant response (1.25 times the control incidence) in females. In view of the low level responses, it is questionable how much reliance can be placed on these results.

Justification for classification or non-classification

Some oil products containing relatively high concentrations of polycyclic aromatic compounds (PAC) are considered genotoxic carcinogens, and, consequently, are classified and labelled as Carcinogenic, Cat. 1B, H350 according to the EU CLP Regulation (EC No. 1272/2008). This classification as carcinogenic does not automatically imply that these substances need also to be classified as mutagenic as defined by EU CLP Regulation (EC No. 1272/2008). The EU legislation aims primarily to classify substances as mutagenic if there is evidence of producing heritable genetic damage, i. e. evidence of producing mutations that are transmitted to the progeny or evidence of producing somatic mutations in combination with evidence of the substance or relevant metabolite reaching the germ line cells in the reproductive organs. The PAC in oil products are poorly bioavailable due to their physico-chemical properties (low water solubility and high molecular weight), making it unlikely that the genotoxic constituents can reach and cause damage to germ cells (Roy, 2007; Potter, 1999). Considering their poor bioavailability, oil products which have been classified as carcinogenic do not need to be classified as mutagenic unless there is clear evidence that germ cells are affected by exposure, consistent with EU CLP Regulation (EC No. 1272/2008). For example, based on in vivo chromosome aberration assays on two cracked gas oils that were negative for genotoxicity, cracked gas oils are not classified as mutagens according to EU classification. Cracked Gas Oils are not classified as mutagenic based on the in vivo data supported by the decreased bioavailability.