(5'S)-8,5'-Cyclo-2'-deoxyadenosine (S-cdA), which arises from the reaction of the hydroxyl radical (*OH) with 2'-deoxyadenosine in DNA, is a lesion comprising a base-sugar linkage that distorts the DNA backbone. This structure impedes transcription and blocks polymerase action. Further, a single S-cdA lesion in the TATA box reduces gene expression. Considering the ability of S-cdA to disrupt DNA structure, which is likely associated with increased cancer risk, we determined S-cdA concentrations in the DNA of stroma, epithelium, and myoepithelium from normal breast tissues using liquid chromatography/mass spectrometry (LC/MS). We also identified differences in the base and backbone structures using Fourier transform-infrared (FT-IR) spectroscopy. LC/MS revealed that the lowest concentration of S-cdA in the stroma (0.04 +/- 0.02 lesions/10(6) bases) occurred in women ages 17 to 30. The highest concentration (0.13 +/- 0.07 lesions/10(6) bases) was found in women 33 to 46. FT-IR spectroscopy showed significant base and backbone differences in the stromal DNA between the women under 30 and those over 50. These findings imply that distortions in the geometry of the helix increase with age, reaching significant proportions in older women. No differences were found in the S-cdA concentrations between the three cell types, suggesting that the *OH attack on the base structure may be essentially random. Initial insight is provided on changes in DNA structure that potentially affect gene expression and increase breast cancer risk.

Reciprocal interactions between the stroma and epithelium are considered to be intimately associated with the development of breast cancer. In studies of whole breast tissues, a keen interest exists in the occurrence of the mutagenic DNA lesions 8-hydroxy-2'-deoxyguanosine and 8-hydroxy-2'-deoxyadenosine. However, there is an apparent lack of information on the presence of these lesions in the DNA of the stroma, epithelium, and myoepithelium, despite the fact that these oxidation products may significantly influence reciprocal interactions between these cell types implicated in carcinogenesis. We report age-related increases in concentrations of both lesions in the stromal DNA, which occur roughly commensurate with the known rise in breast cancer incidence between 30 and 40 years of age. However, no further increases in these concentrations occurred in the older women. Plots of lesion concentrations revealed an uneven distribution, with some younger women having relatively high concentrations and some older women having relatively low concentrations. This finding implies that while increased age is a probable factor in lesion accumulations, other factors may also be influential [e.g., cellular concentrations of reactive oxygen species (ROS)]. Distinct differences were found between the base and backbone structures of the stromal DNA from younger women (ages 17 -30), compared to older women (ages 50 -62). In addition, comparisons of matched stromal, epithelial, and myoepithelial DNA (from the same individual) showed no differences in DNA damage, suggesting a random attack by the hydroxyl radical on all three groups. Collectively, the findings imply that the structural changes in DNA described may potentially disrupt normal reciprocal interactions between the cell types, thus increasing breast cancer risk.