The primary purpose of this research is to examine the hypothesis that reactive oxygen species generated by estrogen quinonoids are the main source for the formation of aldehydic DNA lesions (ADL) in genomic DNA. ADL induced by quinonoid metabolites of 17beta-estradiol (E2), e.g. 4-hydroxyestradiol (4-OH-E2), 2-hydroxyestradiol (2-OH-E2), estrogen-3,4-quinones (E2-3,4-Q) and estrogen- 2,3-quinone (E2-2,3-Q), were investigated in calf thymus DNA (CT-DNA) under physiological conditions. The abasic sites resulting from the spontaneous depurination-depyrimidination of the modified bases and the aldehydic base and sugar lesions resulting from the oxidative damage to deoxyribose moieties in the DNA molecules were measured by an aldehyde reactive probe and were estimated as the number of ADL per 106 nucleotides. With the addition of NADPH (100 micro M) and Cu(II) (20 micro M), nanomolar levels (100 nM) of 4-OH-E2 and 2-OH-E2 induced approximately 10-fold increases in the number of ADL over control (P<0.001). In parallel, increases in 8-oxoguanine were detected in DNA exposed to 4-OH-E2 and 2-OH-E2 (100 nM) plus Cu(II) and NADPH. Further investigation indicated that the ADL induced by estrogen catechols plus Cu(II) and NADPH were causally involved in the formation of hydrogen peroxide and Cu(I). Both E2-2,3-Q and E2-3,4-Q alone induced a 2-fold increase in the number of ADL over control (P<0.05) in CT-DNA at high concentrations (1 mM). Neither neutral thermal hydrolysis nor lower ionic strength of the reaction medium induced further increases in the number of ADL in E2-3,4-Q-modified CT-DNA. Conversely, with the inclusion of Cu(II) and NADPH, both E2-3,4-Q and E2-2,3-Q (1 micro M) induced parallel formation of DNA single strand breaks and approximately 20-fold increases in the number of ADL over control (P < 0.001). The data also demonstrated that the ADL induced by estrogen quinones with and without the presence of Cu(II) and NADPH contain 69 and 78% putrescine-excisable ADL in CT-DNA, respectively. Additionally, results of the ADL cleavage assay indicate that the ADL induced by estrogen quinones plus Cu(II) and NADPH in CT-DNA were predominantly T7 exonuclease-excisable (50%) and exonuclease III- excisable (20%) ADL, whereas the intact ADL, and other ADL accounted for 5 and 25%, respectively. These results suggest that the ADL induced by estrogen quinones in CT-DNA are derived from oxidative events rather than depurination/depyrimidination of labile estrogen quinone-DNA adducts. Overall, our results are at variance with the idea that depurination of estrogen quinone-DNA adducts is the major source for the formation of ADL in genomic DNA. We hypothesize that in addition to DNA adducts and oxidized bases, the ADL induced by estrogen quinonoid-mediated oxidative stress may play a role in estrogen-induced carcinogenicity.