32P-postlabeling analysis of DNA adducts persisting for up to 42 weeks in the skin, epidermis and dermis of mice treated topically with 7,12-dimethylbenz[a]anthracene.

The initial and persistent levels of 7,12-dimethylbenz[a]-anthracene (DMBA)-DNA adducts in mouse skin, epidermis and dermis after topical carcinogen application were studied by 32P-postlabeling assay. In the major experiment, a single dose of 1.2 mumol of the carcinogen was applied to the shaved backs of adult female BALB/cANN mice, and DNA was isolated from epidermis and dermis, respectively, 24 h and 1, 2, 3, 4, 8, 16, 24, 36 and 42 weeks later. Total binding at 24 h was approximately 34 and approximately 28 adducts in 10(7) normal nucleotides for epidermal and dermal DNA, respectively. (One adduct in 10(7) nucleotides equals 0.3 fmol adduct/microgram DNA.) While initial binding was higher in epidermal DNA, the adducts were approximately 10 times more persistent in dermal DNA: at 42 weeks, total binding levels were approximately 0.17 and approximately 1.7 adducts in 10(7) nucleotides for epidermis and dermis, respectively. To quantitate low levels of DMBA-DNA adducts, 32P-postlabeling assays were run in the presence of a limiting amount of carrier-free [gamma-32P]ATP; this was found to favor labeling of the adducts, thereby leading to a 20- to 100-fold enhancement of the method's sensitivity for individual adducts. One of the three major DMBA-DNA adducts was more persistent than were the others; the level of this adduct remained constant at approximately 60% of the total in epidermal and dermal DNA during the last 18 weeks of the 42-week observation period. Since a [3H]thymidine-labeling experiment showed a normal epidermal DNA turnover 40 weeks after DMBA treatment, it was concluded that the bulk of the persistent adducts was present in subpopulations of dormant cells. We have hypothesized that such cells, in the absence of a promoting stimulus, are incapable of division because of the adduction and/or mutation of genes critical for growth (proto-oncogenes), and may thus correspond to the 'latent tumor cells', as defined by Berenblum and Shubik in their classical analysis of the attributes of tumor initiation and promotion.