The yields of unrepairable single-and double-strand breaks in the DNA of x-irradiated Chinese hamster cells were measured by low-speed neutral and alkaline sucrose density gradient sedimentation in order to investigate the relation between these lesions and reproductive death. After maximal single-strand rejoining, at alldoses, the number of residual single-strand breaks was twice the number of residual double-strand breaks. Both double-strand and unrepairable single-strand breaks were proportional to the square of absorbed dose, in the range 10-50 krad. No (6, 8, 20-22, 25, 26).The technical difficulties inherent in measurements of SSBs and, particularly, DSBs in mammalian DNA, using velocity sedimentation, have been to some degree resolved by many recent studies (e.g., refs. 11, 21, and 22 The sedimentation velocity relation for determining DNA molecular weight was calibrated using intact strands of Bacillus subtilis bacteriophage SP50 DNA (duplex molecular weight 108). DNA strand breaks were quantified using a computer simulation of random breakage (27) applied to the molecular weight distributions from control samples (irradiated to 10.8 krad and incubated, see below).
RESULTS AND DISCUSSIONAbsence of double-strand rejoining Three pairs of sedimentation profiles are shown in Fig. 1; each pair consists of profiles from an alkaline and a neutral gradient sedimentation of cells scraped from the same dish. The three pairs are controls (see below), cells irradiated to 54 krad and held at 00 until lysis, and cells incubated for 210 min at 370 after 54 krad. The latter case represents a separate experiment, shown here for convenient comparison, which is possible because the positions of the profiles are reproducible between experiments. The lines shown are the best fit, to profiles for irradiated cells, from computer simulation of random breakage applied to control DNA (27). The close agreement between lines and data points indicates that profiles for irradiated cells correspond to random breakage products. That is, both single-and double-strand breakage, and single-strand rejoining appear to be random (28).The profiles from irradiated cells show that, although 95% of the single-strand breaks were repaired, neither the number nor the distribution of double-strand breaks was significantly different after incubation. Nature of unrepairable single-strand breaks In the above experiment, centrifugation times were such that molecular weights from the neutral and alkaline gradients were closely in two to one correspondence for any sedimentation distance in the central region of the gradients (see upper and lower scales, Fig. 1). It is thus evident that the single-strand profile from incubated cells corresponds accurately to halves of the double-strand fragments from either incubated or unincubated cells. This observation was confirmed by a more detailed comparison. Thus, after incuba-809
