Linear induction of DNA double-strand breakage with X-ray dose, as determined from DNA fragment size distribution.

Pulsed-field gel electrophoresis has been applied to separate DNA from mouse L1210 cells exposed to X-ray doses of 1 to 50 Gy. Simultaneous separation of marker chromosomes in the range 0.1 to 12.6 Mbp allowed calculation of the size distribution of the radiation-induced fragments. The distribution was consistent with a random induction of double-strand breaks (DSBs). A theoretical relationship between the size distribution of such fragments and the average number of induced breaks was used to calculate the yield and dose response. The DNA distribution was determined by both radiolabeling and fluorescence staining. Two independent methods were used to evaluate the radiation-induced yield of DSBs, both assuming that all DNA is broken at random. In the first method we compared the theoretical and experimental fraction of DNA that is below a given size limit. By this method we estimated the yield to be 0.006-0.007 DSB/Gy per million base pairs using the radiolabel and 0.004-0.008 DSB/Gy per million base pairs by fluorescence staining. The dose response was linear in both cases. In the second method we looked only at the size distribution in the resolving part of the gel and compared it to the theoretical distribution. By this method a value of approximately 0.012 DSB/Gy/Mbp was found, using fluorescence as a measure of DNA distribution. In a normal diploid mammalian genome of size 6000 Mbp, this is equivalent to a yield of 25-50 DSBs/Gy or 70 DSBs/Gy, respectively. The second approach, which looks only at the smaller fragments, may overestimate the yield, while the first approach suffers from uncertainties about the fraction of DNA irreversibly trapped in the well. The assay has the capacity to detect a dose of less than 1 Gy.