Oxidative DNA damage mediated by copper(II), iron(II) and nickel(II) fenton reactions: evidence for site-specific mechanisms in the formation of double-strand breaks, 8-hydroxydeoxyguanosine and putative intrastrand cross-links.

The role of metal ion-DNA interactions in the Fenton reaction-mediated formation of putative intrastrand cross-links, 8-hydroxydeoxyguanosine (8-OHdG) and single- and double-strand breaks was investigated. Salmon sperm DNA and pBluescript K+ plasmid were incubated with hydrogen peroxide and either copper(II), iron(II), or nickel(II), which differ in both their affinity for DNA and in the spectrum of oxidative DNA damage they induce in Fenton reactions. EDTA was included in these incubations according to two different strategies; the first (strategy 1) in which DNA and metal ions were mixed prior to the addition of EDTA, the second (strategy 2) in which EDTA and metal ions were mixed prior to the addition of DNA. The formation of the putative intrastrand cross-links, monitored by 32P-postlabelling, was not affected by the addition of between 10 microM and 5 mM EDTA to the copper(II) Fenton reaction according to strategy 1. In contrast, the level of cross-links declined significantly upon inclusion of 20 microM EDTA and above when added according to strategy 2. Similarly, formation of these lesions declined in the iron(II) Fenton reaction more dramatically upon addition of 5 mM EDTA when added according to strategy 2 compared to strategy 1, while the yield of cross-links formed in the nickel(II) Fenton reaction declined equally with both strategies with up to 25 mM EDTA. The formation of single- and double-strand breaks was investigated in plasmid DNA by agarose gel electrophoresis and subsequent densitometry. The formation of linear DNA in the iron(II) Fenton reaction decreased dramatically upon inclusion of EDTA according to strategy 2, while no such decline was observed using strategy 1. In contrast, the formation of linear DNA in the copper(II) Fenton reaction decreased upon inclusion of EDTA according to both strategies. A decrease in the formation of open-circular DNA was also observed upon inclusion of EDTA according to both strategies; however this decrease occurred at a lower EDTA concentration in strategy 2 (100 microM) compared to strategy 1 (200 microM), and the level of open-circular DNA reached a lower level (8. 5% compared to 24.2%). The nickel(II) Fenton reaction generated only open-circular DNA, and this was completely inhibited upon addition of 25 microM EDTA according to both strategies. There was less formation of 8-OHdG in the copper(II) and iron(II) Fenton reactions when EDTA was added according to strategy 2 than according to strategy 1. These results suggest that a site-specific mechanism is involved in the formation of double-strand breaks and, to a lesser extent, 8-OHdG and the putative intrastrand cross-links, while the formation of single-strand breaks is more likely to involve generation of hydroxyl radicals in solution. Copyright 1999 Elsevier Science B.V.