Chromosome aberrations can occur by secondary duced by metabolic poisons, 2,4-dinitrophenol, (unmechanism(s) associated with cytotoxicity, induced couples oxidative phosphorylation), and sodium ioby chemicals that do not attack DNA. Aberrations doacetate, (NaI; blocks ATP production). Five of the are formed from DNA double-strand breaks, and chemicals that induced aberrations in CHO cells DSBs are known to be induced by nonmutagenic were tested in human TK6 cells and four were posi-(Ames test negative) noncarcinogens at toxic levels tive, the fifth being equivocal. Stable aberrations [Storer et al. (1996): Mutat Res 368:59 -101]. (translocations) were induced in human cells by Here, 8 of 12 of these chemicals caused aberra-NaI. Clearly, chemicals can give ''false-positive'' tions in CHO cells at cytotoxic doses, and often only results in the chromosome aberration assay at cytowhen cell counts (survival) at 20 hr approached toxic levels, though cytotoxicity does not always Β°50% of controls. Five of eight noncarcinogens produce aberrations, so that further information (2,4,-dichlorophenol, dithiocarb, menthol, phthalic (e.g., DNA reactivity) is needed to determine anhydride, and ethionamide) and one of two equiv-whether a result is a ''false-positive.'' Primary DNAocal carcinogens (bisphenol A) caused aberrations, damaging chemicals such as alkylators are also cyusually over a narrow dose range with steeply in-totoxic, but give strong increases in aberrations creasing cytotoxicity. Phthalic anhydride and ethi-without marked initial toxicity by the measures used onamide were positive only at doses with precipi-here, although the aberrations they induce do retate. Phenformin was negative even at toxic doses duce long-term survival in colony-forming assays. and ephedrine and phenylephrine were negative Environ. Mol. Mutagen. 31:316 -326, 1998 and gave little toxicity. Aberrations were also in-α§ 1998 Wiley-Liss, Inc.