DNA interstrand cross-links induced by ionizing radiation: an unsung lesion

The induction of DNA interstrand cross-links by ionizing radiation has been largely ignored in favour of studies on double-strand break formation and repair. At least part of the problem is technical; it is difficult to detect and quantify interstrand cross-links when the same agent forms both cross-links and single strand breaks because the detection of interstrand cross-links generally involves a denaturation step. Our group has studied the induction of interstrand cross-links following irradiation of DNA containing bromouracil at specific sites. We found that the formation of interstrand cross-links requires the presence of a few (3-5) mismatched bases, comprising the bromouracil. In the absence of mismatched bases, no radiation-induced cross-linking was observed; however, even in the absence of bromouracil, cross-linking still occurred, albeit at a lower efficiency. Our molecular modelling studies demonstrate that the mobility of the bases in the mismatched region is essential for the cross-linking process. Thus, our hypothesis is that ionizing radiation induces DNA interstrand cross-links in non-hybridized regions of DNA. Some obvious examples of such DNA regions are replication forks, transcription bubbles and the D-loop of telomeres. However, an abundance of studies have made it clear that there must be many single-stranded regions in the genome, such as hairpins and cruciforms. For example, alpha satellite DNA, in centromere regions of human chromosomes, forms hairpins. Thus, a variety of non-B DNA structures (hairpins, slipped DNA and tetrahelical structures) exist in the genome and should be susceptible to the formation of radiation-induced interstrand cross-links. Although interstrand cross-links have thus far been virtually ignored in radiation biology, it will be worthwhile to develop methods to detect their presence following exposure of cells to biologically relevant levels of ionizing radiation, since, on a per lesions basis, they are probably more toxic than double-strand breaks.