The location of DNA photodamage within the epidermis is crucial as basal layer cells are the most likely to have carcinogenic potential. We have determined the action spectra for DNA photodamage in different human epidermal layers in situ. Previously unexposed buttock skin was irradiated with 0.5, 1, 2, and 3 minimal erythema doses of monochromatic UVR at 280, 290, 300, 310, 320, 340, and 360 nm. Punch biopsies were taken immediately after exposure and paraffin sections were prepared for immunoperoxidase staining with a monoclonal antibody against thymine dimers that were quantitated by image analysis. Dimers were measured at two basal layer regions, the mid and the upper living epidermis. The slopes of dose-response curves were used to generate four action spectra, all of which had maxima at 300 nm. Dimer action spectra between 300 and 360 nm were independent of epidermal layer, indicating comparable epidermal transmission at these wavelengths. Furthermore, we observed 300 nm-induced dimers in dermal nuclei; however, there was a marked effect of epidermal layer between 280 and 300 nm, showing relatively poor transmission of 280 and 290 nm to the basal layer. These data indicate that solar UVB (approximately 295-320 nm) is more damaging to basal cells than predicted from transmission data obtained from human epidermis ex vivo. The epidermal dimer action spectra were compared with erythema action spectra determined from the same volunteers and ultraviolet radiation sources. Overall, these spectral comparisons suggest that DNA is a major chromophore for erythema in the 280-340 nm region.

We assessed the in situ time-dependent loss of epidermal thymine dimers and 6-4 photoproducts in skin types I and II after exposure to two minimal erythema doses of solar-simulating radiation on previously unexposed buttock skin. Using quantitative image analysis, we evaluated biopsy sections stained with monoclonal antibodies. We then made comparisons, in the same volunteers, with unscheduled DNA synthesis, which is a direct marker of overall excision repair. Removal of thymine dimers was slow (half-life = 33.3 h), with high levels of lesions still present 24 h post-irradiation; some lesions were still present at 7 d. In contrast, removal of 6-4 photoproducts was rapid (half-life = 2.3 h), the decay kinetics of which correlated better with the decline in epidermal unscheduled DNA synthesis (half-life = 7.1 h). These data show that as in mouse, monkey, and in vitro models, the 6-4 photolesion is repaired preferentially in human epidermis in situ. They also raise the possibility that poor thymine dimer repair may be a feature of skin types I and II, who are more prone to skin cancer than are types III and IV. There was an inverse relationship between the onset of erythema and 6-4 photoproduct repair, suggesting that this repair process initiates erythema.