Chinese hamster ovary cell assays for mutation and chromosome damage: data from non-carcinogens.

In vitro genotoxicity tests are employed to screen chemicals for their capability to cause various DNA and chromosomal alterations, and the results are often used to predict their potential for carcinogenicity. However, there is controversy regarding the apparent low specificity of some in vitro genotoxicity assays, which result in a high false positive rate. Since we use and rely upon in vitro assays for risk assessment and prediction of carcinogenicity, this specificity issue is of serious concern to us. Hence, we selected ten compounds deemed non-carcinogens in the literature to test for the induction of gene mutation and chromosomal damage using the Chinese hamster ovary cell/hypoxanthine-guanine phosphoribosyl transferase (CHO/HGPRT) mutation assay performed concurrently with a CHO micronucleus assay. The chemical exposures for the two end-points were done simultaneously. The protocol for the two end-points was developed using the carcinogens N-methyl-N'-nitro-N-nitrosoguanidine, 3-methylcholanthrene, cyclophosphamide and 7,12-dimethylbenzanthracene. The non-carcinogens chosen were 4-nitro-o-phenylenediamine, p-phenylenediamine dihydrochloride, 3-nitropropionic acid, dichlorvos, 2-(chloromethyl)pyridine, N-(1-naphthyl)ethylenediamine 2HCl, O-anthranilic acid, 4-nitroanthranilic acid, anilazine and triphenyltin hydroxide. Each of these chemicals had been reported positive in the Ames test and/or the mouse lymphoma TK+/- mutation assay. In addition, eight of them were also reported positive in in vitro assays for chromosome aberrations and/or sister chromatid exchange (SCE). We found four of the ten chemicals negative for gene mutation and micronucleus induction without and with activation in the CHO/HGPRT mutation and CHO micronucleus assays. However, one of these four chemicals may be a potential carcinogen according to other carcinogenicity reviewers. Four other chemicals that induced only micronuclei were negative for gene mutation. Dichlorvos was positive for gene mutation and micronucleus induction without and with activation. This chemical has been shown recently to cause various tumors in rodents. One of the non-carcinogens was positive in the micronucleus test and equivocally positive in the mutation test. These results indicate that the CHO/HGPRT mutation assay may provide more relevant results than the CHO micronucleus assay, the mouse lymphoma mutation assay, or in vitro SCE and chromosome aberration assays when screening chemicals for potential carcinogenicity.