Factors effecting the efficiency of purine analogues as selective agents for mutants of mammalian cells induced by ionising radiation

In the Chines hamster cell line V79-4, the frequencies of cells selected for their resistance to purine analogues do not always reflect the true frequencies of resistant mutants. The frequency of cells resistant to 8-azaguanine varied widely, especially when different sources of serum were used in the selective medium. Even with the more efficient analogue, 6-thioguanine, small colonies arose in the selective medium at a frequency which was strongly dependent upon analogue concentration and viable cell seeding density. These colonies were shown to have a phenotype which was indistinguishable from wild type. Hence with irradiated cells, where the viability of the cell population is reduced to an extent varying with the dose and the interval allowed for mutant expression, the counting of all colonies arising in selective medium can lead to spuriously variable, and sometimes very high, 'mutation frequencies'. Although the frequency of wild type colonies selected in thioguanine was diminished by the use of high concentrations of the analogue, a loss of induced mutants also occurred at these concentrations. Further, the V79-4 line contained two distinct types of mutant with different levels of hypoxanthine-guanine phosphoribosyl transferase (HGPRT) activity, and only one of these types (HGPRT-negative mutants) increased in frequency with radiation dose. These results can account for many of the anomalies encountered in previous studies with purine analogues as selective agents, and show that some care has to be taken to characterize the mutants selected by resistance to purine analogues before meaningful dose-response relationships can be established