Many electron spin resonance (ESR) spectra of 5,5-dimethyl-1-pyrroline N-oxide (DMPO) radical adducts from the reaction of organic hydroperoxides with heme proteins or Fe(2+) were assigned to the adducts of DMPO with peroxyl, alkoxyl, and alkyl radicals. In particular, the controversial assignment of DMPO/peroxyl radical adducts was based on the close similarity of their ESR spectra to that of the DMPO/superoxide radical adduct in conjunction with their insensitivity to superoxide dismutase, which distinguishes the peroxyl adducts from the DMPO/superoxide adduct. Although recent reports assigned the spectra suggested to be DMPO/peroxyl radical adducts to the DMPO/methoxyl adduct based on independent synthesis of the adduct and/or (17)O-labeling, (17)O-labeling is extremely expensive, and both of these assignments were still based on hyperfine coupling constants, which have not been confirmed by independent techniques. In this study, we have used online high performance liquid chromatography (HPLC or LC)/ESR, electrospray ionization-mass spectrometry (ESI-MS) and tandem mass spectrometry (MS/MS) to separate and directly characterize DMPO oxygen-centered radical adducts formed from the reaction of Fe(2+) with t-butyl or cumene hydroperoxide. In each reaction system, two DMPO oxygen-centered radical adducts were separated and detected by online LC/ESR. The first DMPO radical adduct from both systems showed identical chromatographic retention times (t(R) = 9.6 min) and hyperfine coupling constants (a(N) = 14.51 G, a(H)(beta) = 10.71 G, and a(H)(gamma) = 1.32 G). The ESI-MS and MS/MS spectra demonstrated that this radical was the DMPO/methoxyl radical adduct, not the peroxyl radical adduct as was thought at one time, although its ESR spectrum is nearly identical to that of the DMPO/superoxide radical adduct. Similarly, based on their MS/MS spectra, we verified that the second adducts (a(N) = 14.86 G and a(H)(beta) = 16.06 G in the reaction system containing t-butyl hydroperoxide and a(N) = 14.60 G and a(H)(beta) = 15.61 G in the reaction mixture containing cumene hydroperoxide), previously assigned as DMPO adducts of t-butyloxyl and cumyloxyl radical, were indeed from trapping t-butyloxyl and cumyloxyl radicals, respectively.
Guo et al. (2003) studied this question.