DNA strand breaks produced by oxidative stress in mammalian cells exhibit 3'-phosphoglycolate termini

In recent years two mechanisms have been proposed for the production of DNA strand breaks in cells undergoing oxidative stress: (i) DNA attack by OH radical, produced by Fenton reaction catalyzed by DNA-bound iron; and (ii) DNA attack by calcium-activated nucleases, due to the increase of cytosolic and nuclear calcium induced by oxidative stress. We set out to investigate the participation of the former mechanism by detecting and quantifying 3'-phosphoglycolate, a 3' DNA terminus known to be formed by OH radical attack to the deoxyribose moiety, followed by sugar ring rupture and DNA strand rupture. These structures were found in DNA of monkey kidney cells exposed to hydrogen peroxide, iron nitrilotriacetate or ascorbate, all species known to favor a cellular pro-oxidant status. The method employed to measure 3' phosphoglycolate was the 32P-postlabeling assay. Repair time course experiments showed that it takes 10 h for 3'-phosphoglycolate to be removed from DNA. It was found that the DNA of both control cells and cells exposed to hydrogen peroxide had a very poor capacity of supporting in vitro DNA synthesis, catalyzed by DNA polymerase I. If the DNA was previously incubated with exonuclease III, an enzyme able to expose 3'-OH primers by removal of 3'-phosphoglycolate and 3'-phosphate termini the in vitro synthesis was substantially increased. This result shows that either of these termini are present at the break and that 3'-hydroxyl termini are virtually absent. At least 25% of the strand breaks exhibited 3'-phosphoglycolate termini as determined by the 32P-postlabeling assay, but due to the characteristic of the method this percentage is likely to be higher. These results favor the hypothesis that an oxidative agent generated by Fenton reaction is responsible for DNA strand breakage in cells undergoing oxidative stress.