Ataxia‐Telangiectasia (A‐T) and Nijmegen breakage syndrome (NBS) are recessive genetic diseases with similar cellular phenotypes that are caused by mutations in the recently described ATM (encoding ATM) and NBS1 (encoding p95) genes, respectively. Both disorders are accompanied by immunodeficiency in a majority of patients, but the mechanism involved has as yet not beenestablished. We demonstrate that in cells from A‐T patients, the switch (S) recombination junctions are aberrant and characterized by a strong dependence on short sequence homologies and devoid of normally occurring mutations around the breakpoint. A low number of S fragments were generated in cells from NBS patients and showed only limited dependence on sequence identity and mutation frequencies were similar to those observed in normal controls. We propose that ATM and p95 are both involved in the final step(s) in class switch recombination with related, but disparate, functional roles. Thus, the general pathway involved in DNA repair also has a major influence on the immunoglobulin isotype switching process.

The sunburn reaction is the most common consequence of human exposure to ultraviolet radiation (UVR), and is mediated at least in part by interleukin-6 (IL-6). The aim of this study was to determine if DNA is a major chromophore involved in the induction of IL-6 following UV irradiation of a human epidermoid carcinoma cell line (KB), and of normal human epidermal keratinocytes. We first confirmed that IL-6 release was associated with enhanced levels of IL-6 mRNA transcripts. The wavelength dependence for IL-6 release was then investigated by irradiating the cells at defined wavelengths (254, 302, 313, 334, and 365 nm) with a monochromator. The maximum effect on IL-6 release was observed at 254 nm with only low levels of induction observed at wavelengths above 313 nm. The wavelength dependence for UV-induced IL-6 release was similar to that for DNA absorption or for the induction of cyclobutane pyrimidine dimers (CPD). To determine whether UV-induced DNA damage mediated IL-6 secretion, the role of CPD was investigated by treating keratinocytes with photosomes (photolyase encapsulated in liposomes) followed by photoreactivating light. This photoreversal procedure led to a reduction in the levels of the UVC-induced secretion of IL-6, which in normal human keratinocytes was unambiguously associated with repair of CPD. We conclude that the release of IL-6 from human keratinocytes following short-wave UVC and UVB irradiation is mediated by DNA damage and that CPD play an important role in this process.

Tanning lamps, emitting predominantly ultraviolet (UV) A, are used widely throughout the U.K. and other countries, but little is known about the long-term risks associated with their use, especially with respect to skin cancer. We have exposed normal human epidermal keratinocytes to a commercial tanning lamp and used the comet assay in association with DNA repair enzymes T4 endonuclease V and endonuclease III to investigate the relative yields of directly formed cyclobutane pyrimidine dimers (CPDs) and indirectly formed types of oxidative DNA damage. To put the risk of using tanning lamps into perspective, the sunbed used in this study (five Philips Performance 80W-R UVA tubes at a distance of 35 cm) was found to be approximately 0.7 times as potent at inducing CPDs as U.K. natural sunlight around noon on a fine summer day. This compares with a relative risk for CPD induction and erythema of 0.8 and 0.7 times, respectively, calculated from the relevant action spectra of tanning lamps and British noontime sunlight. To determine the relative contribution of UVB and UVA to the induction of CPDs and oxidative DNA damage, we modified the spectral output of the tanning lamps with a series of Schott WG UVB filters. The induction of CPDs was more dependent on the UVB component of the sunbed than oxidative types of damage. Schott WG UVB filters with 50% transmission at 305 nm reduced the yield of T4 endonuclease V sites by 42% while there was only a 17% decrease in the yield of endonuclease III sites. CPD induction was not completely abolished after irradiation through WG335 and WG345 nm filters despite there being no detectable UVB. From these data, it was estimated that, although the tanning lamps emitted only 0.8% of their total output in the UVB range, these wavelengths were responsible for the induction of over 75% of CPDs and 50% of the oxidative damage to DNA.

We have compared the induction of apoptosis and cytokine release by UVB and gamma-radiation in primary (untransformed) and in two immortalized human epithelial/keratinocyte cell lines, HaCaT and KB (KB is now known to be a subline of the ubiquitous keratin-forming tumour cell line HeLa and we therefore designate it HeLa-KB). In both the primary and the immortalized cell lines apoptosis and release of the inflammatory cytokine interleukin-6 are induced rapidly following UVB irradiation. In contrast, only the immortalized cells undergo apoptosis and release interleukin-6 after gamma-irradiation and here the onset of apoptosis and cytokine release are delayed. The same distinction between primary and immortalized cells was observed when double-strand breaks were induced with the anticancer drug mitoxantrone, which stabilizes topoisomerase II-cleavable complexes. We suggest that immortalization may sensitize keratinocytes to the apoptogenic effect of ionizing radiation or mitoxantrone by deregulating normal cell cycle checkpoints. In both human keratinocytes and fibroblasts, cell killing, as assayed by loss of colony-forming ability, is not coupled to apoptosis. Immortalization increases resistance to gamma-radiation killing but sensitizes to apoptosis. In contrast, although immortalization also sensitizes to UVB-induced apoptosis, it does not affect UVB-induced cell killing. Apoptosis unambiguously indicates death at the single cell level but clonal cell survival integrates all the cellular and genetic processes which prevent or permit a scorable clone to develop.