Characterization of hydantoin products from one-electron oxidation of 8-oxo-7,8-dihydroguanosine in a nucleoside model

Use of one-electron oxidants such as Na2IrCl6 to oxidize 8-oxo-7,8-dihydro-2 ' -deoxyguanosine (OG) residues in oligodeoxynucleotides was previously shown to lead to predominant formation of a base lesion of mass M - 10 compared to starting material [Duarte et al. (1999) Nucleic Acids Res. 27, 596-502]. To thoroughly characterize the structure of this lesion, the oxidation of the nucleoside 9-N-(2 ' ,3 ' ,5 ' -tri-O-acetyl-beta -D-erythro-pentanosyl)-8-oxo-7,8-dihydroguanine with one-electron oxidants at pH 2-4 was used as a model for duplex DNA oxidation of OG residues. H-1 NMR and H,H COSY NMR studies in CD3OD along with LC-ESI-MS/MS fragmentation analysis are consistent with the assignment of the M - 10 species as a mixture of two pH-dependent equilibrating isomers, a guanidinohydantoin (Gh) and an iminoallantoin (Ia) nucleoside, both present as mixtures of epimers at the C5 position of the hydantoin ring, i.e., four total isomers are formed. The Gh/Ia mixture is formed from hydration and decarboxylation of the initially formed intermediate 5-hydroxy-8-oxo-7,8-dihydroguanosine, a species that is also produced by four-electron oxidation (e.g., singlet oxygen) of guanosine. The product mixture can be further oxidized to a species designated Ia(ox), a hydrolytically unstable material at pH 7 that has been characterized by ESI-MS and H-1 NMR. Competition studies with 8-oxo-7,8-dihydroadenosine placed the redox potential of Gh/Ia at about 1.0 V vs NHE. These studies provide important information concerning the structures of lesions obtained when OG, a hot spot for oxidative damage, serves as a hole trap in long-range electron-transfer studies.