Nineteen fractions of whole condensate of smoke from the University of Kentucky Reference Cigarette IRI were tested for mutagenicity in vitro using a bacterial indicator system. As little as 25 mug of the active fractions were mutagenic toward histidine-requiring Salmonella typhimurium TA1538, if the condensates were incubated in the presence of rat or human liver homogenates of lung were relatively inactive. Homogenates from livers of rats that had been treated with 3-methylcholanthrene

more »

... ed condensates to mutagens more efficiently than did liver homogenates from man or from normal or phenobarbital-treated rats. Use of homogenates from animals treated with 3-methylcholanthrene gave much more reproducible results in smoke fraction assays because larger numbers of revertants were obtained, and dose-response curves were linear over the range 25 to 250 mug condensate. The linear dose-response curves permitted quantitative comparison of the various fractions. The mutagenicity per mg of basic fractions of whole smoke condensate is very high and that of neutral polycyclic hydrocarbons is very low. Because of the exquisite preferential sensitivity of the TA1538 test system to polycyclic amines and insensitively to alkyl polycyclics, there is a poor quantitative correlation between mutagenicity and carcinogenicity, as measured by skin painting or in vitro cell transformation. There is substantial evidence that many carcinogens are mutagens but that most of these compounds require metabolism before they are biologically active. If further development improves the sensitivity of the bacterial testing system to mutagenic derivatives of alkyl polycyclic and heteropolycyclic hydrocarbons, it may provide a convenient, rapid, quantitative, and inexpensive bioassay for the detection of potentially carcinogenic substances in tobacco smoke condensates.