Registration Dossier

Administrative data

Key value for chemical safety assessment

Additional information

There is a relatively large database on genetic toxicity studies of gasoline covering both in vivo and in vitro tests. Blended gasoline was not mutagenic, either with or without metabolic activation, in in vitro test systems, including Salmonella typhimurium, Saccharamyces cerevisiae, a mammalian cell line (L5178Y), and human lymphoblastoid cells, or in in vivo test systems, including rat micronucleus and rat chromosome aberration assays.

Naphtha streams produced mixed results in in vitro gene mutation assays but negative results in vivo. For example, light straight run naphtha (CAS 64741-46-4) increased the number of revertant colonies in a Salmonella assay, but other low boiling point naphtha streams, sweetened naphtha (CAS 64741-87-3) and light alkylate naphtha (CAS 64741-66-8), produced no evidence of chromosomal damage in vivo. Where there were positive findings in mouse lymphoma tests, these tended to be associated with the higher boiling cracked streams, suggesting that there may have been small amounts of mutagenic constituents present in these streams. The positive results with the higher boiling fractions may not be directly relevant to human health risk assessment since these higher boiling constituents are unlikely to be present in the vapor to which humans would normally be exposed.

It also seems unlikely that genetic toxicity contributes to the tumorigenic response, as the mechanistic data are more consistent with non-genotoxic processes. Further, the organs in which in vivo mutagenic responses were found were not those in which tumors developed (see further discussion in section 7.7 Carcinogenicity). Finally, the validity of some of the older mouse lymphoma tests has been questioned (Caldwell J. 1993. Perspective on the usefulness of the mouse lymphoma assay as an indicator of a genotoxic carcinogen: ten compounds which are positive in the mouse lymphoma assay but are not genotoxic carcinogens. Teratogenesis, Carcinogenesis and Mutagenesis 13:185 -190), and the possibility that the positive results in these studies are not actually evidence of mutagenic activity must be considered.

Short description of key information:
Genetic toxicity - in vitro gene mutation: not genotoxic in bacterial cells (OECD TG 471)
Genetic toxicity - in vitro gene mutation: not genotoxic in mammalian cells (OECD TG 476)
Genetic toxicity - in vivo micronucleus assay: not genotoxic in rats (OECD 474)
Genetic toxicity - in vivo chromosome aberration: not genotoxic in rats (OECD TG 475)

Endpoint Conclusion:

Justification for classification or non-classification

Many of the “standard” assays, i.e., Salmonella, mouse lymphoma, in vivo cytogenetics, and the dominant lethal test of gasoline were conducted prior to the publication of regulatory protocols or the promulgation of good laboratory practice guidelines. Nevertheless, they followed the recommendations of the developers of the various assays, and were, therefore, consistent with the guidelines that were subsequently developed. There are no obvious errors or omissions in the various tests. The micronucleus test of gasoline as well as the studies of the naphtha blending stocks were conducted more recently and were in accordance with regulatory guidelines and procedures. Thus the data can be used without reservations for regulatory purposes. The remaining data come from published articles and have been independently peer-reviewed. The available information goes far beyond the minimum data requirements; accordingly, no additional testing for genetic toxicity is recommended. Further, the data indicate that classification of gasoline and/or naphtha blending stocks is not warranted.

It should be noted that, although the data do not support classification of gasoline per se for genotoxic potential, there is a regulatory requirement to classify as genotoxic gasoline and naphtha streams containing >0.1% benzene.