Suggestions concerning the relationship between mutant frequency and mutation rate at the hprt locus in human...

Mutant frequency is defined as the proportion of mutant cells in a population and is readily estimated. It should be distinguished from mutation rate, which relates to the rate at which mutation events arise, and is generally expressed as events per cell division. Since one mutation event may give rise to one or many mutant cells, depending on the generation in which it has arisen, the relationship of mutant frequency to the underlying mutation rate is complex. A large number of estimates of mutant frequency at the hprt locus in human lymphocytes are available, from our two laboratories among others. From our two extensive data sets, we have determined median hprt mutant frequencies of different age groups and used the method of Lea and Coulson (J. Genet., 49, 1949, 264-285) to attempt to estimate the underlying mutation rate at this locus. It is in principle possible to obtain estimates of mutation rate from the mutant frequency in newborns, from the increase in mutant frequency with age, and from the difference between the upper and lower quartile mutant frequencies. We discuss reasons for the discrepancies between these estimates and argue that the best estimate can probably be obtained from the increase in mutant frequency with age. We arrive at an estimate of mutation rate to 6-thioguanine resistance at the hprt locus of about 5 x 10(-7) mutation events per nominal cell division.