Absence of 2'-deoxyoxanosine and presence of abasic sites in DNA exposed to nitric oxide at controlled physiological concentrations.

Abstract

Nitric oxide (NO(*)) is a physiologically important molecule at low concentrations, while high levels have been implicated in the pathophysiology of diseases associated with chronic inflammation, such as cancer. While an extensive study in vitro suggests that oxidative and nitrosative reactions dominate the complicated chemistry of NO(*)-mediated genotoxicity, neither the spectrum of DNA lesions nor their consequences in vivo have been rigorously defined. We have approached this problem with a major effort to define the spectrum of nitrosative DNA lesions produced by NO(*)-derived reactive nitrogen species under biological conditions. Plasmid pUC19 DNA was exposed to steady state concentrations of 1.3 microM NO(*) and 190 microM O(2) (calculated steady state concentrations of 40 fM N(2)O(3) and 3 pM NO(2)(*) in the bulk solution) in a recently developed reactor that avoids the undesired gas phase chemistry of NO(*) and approximates the conditions at sites of inflammation in tissues. The resulting spectrum of nitrosatively induced abasic sites and nucleobase deamination products was defined using plasmid topoisomer analysis and a novel LC/MS assay, respectively. With a limit of detection of 100 fmol and a sensitivity of 6 lesions per 10(7) nt in 50 microg of DNA, the LC/MS analysis revealed that 2'-deoxyxanthosine (dX), 2'-deoxyinosine (dI), and 2'-deoxyuridine (dU) were formed at nearly identical rates (k = 1.2 x 10(5) M(-1) s(-1)) to the extent of approximately 80 lesions per 10(6) nt after 12 h exposure to NO(*) in the reactor. While reactions with HNO(2) resulted in the formation of high levels of 2'-deoxyoxanosine (dO), one of two products arising from deamination of dG, dO, was not detected in 500 microg of DNA exposed to NO(*) in the reactor for up to 24 h (<6 lesions per 10(8) nt). This result leads to the prediction that dO will not be present at significant levels in inflamed tissues. Another important observation was the NO(*)-induced production of abasic sites, which likely arise by nitrosative depurination reactions, to the extent of approximately 10 per 10(6) nt after 12 h of exposure to NO(*) in the reactor. In conjunction with other studies of nitrosatively induced dG-dG cross-links, these results lead to the prediction of the following spectrum of nitrosative DNA lesions in inflamed tissues: approximately 2% dG-dG cross-links, 4-6% abasic sites, and 25-35% each of dX, dI, and dU.