Two supercoiled (SC), double-stranded DNAs, pBR 322 and pUC 19, have been subjected to oxidative stress using horseradish peroxidase (HRP) and HRP + tert-butyl hydroperoxide (BOOH). HRP alone causes single-strand cleavage of these SC DNAs. Strand cleavage is enhanced substantially in the presence of commercial BOOH (which contains H 2 O 2 ) but is, at best, only very slightly enhanced in the presence of pure BOOH. In the HRP/pure BOOH system, the DNA single-strand-scission which does occur appears to be due to a direct action of oxidized HRP. It is not due to tert-butylperoxyl radicals because strand-scision is not even retarded by 10 mM Trolox, an outstanding water-soluble trap for peroxyl radicals. The present results are congruent with our earlier conclusion [Paul, T., et al. (2000) Biochemistry 39, 4129] that neutral alkylperoxyl radicals produce little or no direct single-strand-scission in SC DNAs.It was recently demonstrated in this laboratory that direct single-strand cleavage of (double-strand) supercoiled DNA by water-soluble alkylperoxyl radicals (1) 1 at 37°C was relatively facile when the peroxyl carried a positive charge but was generally below the level of detection when the peroxyl was uncharged or carried a negative charge (2). For example, with the positively charged peroxyl radical, (H 2 N) 2 + CC(CH 3 ) 2 OO • ( + ROO • ), 2 and the supercoiled (SC) plasmid DNA, pBR 322, it was found that ca. 50% of the SC DNA suffered a singlestrand break to afford relaxed (R) DNA at a + ROO • /bp ratio of 0.2 (2). Strand cleavage by a different positively changed peroxyl occurred with a similar efficiency (2). In contrast, the neutral, water-soluble peroxyl radical, HOCH 2 CH 2 NHC(O)C(CH 3 ) 2 OO • , generally produced no detectable strand-scission (i.e., no detectable R DNA) at a ROO • /bp ratio as high as 5:1 (2, 3), 3 and the negatively charged peroxyl radical, -O 3 SCH 2 CH 2 C(CH 3 )(CN)OO • , generally produced no strand-scission even at a ROO • / bp ratio of 24:1 (2). Essentially identical results were obtained with another SC DNA, pUC 19 (2). The uniquely strong DNA strand cleaving abilities of the positively charged peroxyl radicals was attributed to their Coulombic attraction to the negatively charged SC DNA polyanion (2).After these null-results on the neutral ROO • /DNA reaction were published (2), a paper appeared which caused us surprise and concern because it claimed that neutral alkylperoxyl radicals "generated in situ" from five different alkyl hydroperoxides and three different peroxidases generally produced single-strand breaks in pBR 322 (4). This strand-scission was not a consequence of the DNA having been subjected to very much higher ROO • /bp ratios than we had employed. 4 We were therefore prompted to investigate one of the peroxidase/alkyl hydroperoxide systems. We chose horseradish peroxidase (HRP) and tert-butyl hydroperoxide (BOOH) because both are readily available commercially and because the HRP/ BOOH pair had been reported to have "a substantial DNA-cleaving activity"(4). 5...