Reactive oxygen species (ROS) are associated not only with initiation, but also with promotion and progression in the multistage carcinogenesis model. In the present review, we will focus on the involvement of ROS in skin carcinogenesis, especially that induced by ultraviolet (UV) radiation. UV-specific DNA damage has been well studied thus far. However, recent reports have revealed the previously unknown participation of oxidative stress in UV-induced skin carcinogenesis. Indeed, in addition to transition-type mutations at dipyrimidine sites, G:C to T:A transversions, which may be induced by the presence of 8-oxoguanine during DNA replication, are frequently observed in the ras oncogene and p53 tumor suppressor gene in human skin cancers of sun-exposed areas and in UV-induced mouse skin cancers. Recent studies have shown that not only UV-B, but also UV-A is involved in UV-induced carcinogenesis. A wide variety of biological phenomena other than direct influence by UV, such as inflammatory and immunological responses and oxidative modifications of DNA and proteins, appear to play roles in UV-induced skin carcinogenesis. Furthermore, it has become clear that genetic diseases such as xeroderma pigmentosum show deficient repair of oxidatively modified DNA lesions. The involvement of ROS in skin carcinogeneisis caused by arsenic and chemical carcinogens will also be discussed.

8-Hydroxy-2'-deoxyguanosine (8-OHdG) is a mutation-prone (G:C to T:A transversion) DNA base-modified product generated by reactive oxygen species or photodynamic action. G:C to T:A transversions are observed in the p53 and ras genes of UVB-induced skin cancers of mice and in squamous and basal cell carcinomas of human skin exposed to sunlight. In the current study, 8-OHdG formation was evaluated in the epidermis of hairless mice after repeated exposure to UVB, and possible mechanisms involved were studied. Exposure of hairless mice to either 3.4 [2 minimal erythema dose (MED)] or 16.8 (10 MED) kJ/m2 of UVB three times a week for 2 wk induced a 2.5- or 6.1-fold increase, respectively, in the levels of 8-OHdG in DNA, compared to the unexposed controls. An immunohistochemical method using a monoclonal antibody specific for 8-OHdG showed stronger and more extensive staining in the nuclei of UV-irradiated epidermal cells than in those of nonirradiated cells. Western blots probed with antibodies against 4-hydroxy-2-nonenal-modified proteins confirmed the involvement of reactive oxygen species in the epidermal damage induced by chronic UVB exposure. 3-Nitro-L-tyrosine was detected in western blots in a concentration-dependent manner, suggesting that peroxynitrite derived from the reaction of nitric oxide and superoxide, both of which were probably released from inflammatory cells, was involved in modifying the DNA bases. Therefore, the formation of 8-OHdG after UVB exposure appears to be regulated by at least three pathways: photodynamic action, lipid peroxidation, and inflammation and may play a role in sunlight-induced skin carcinogenesis.

Tissue from 54 histologically-identified basal cell carcinomas of the skin was obtained at surgery and assayed using a combination of functional and immunochemical procedures for matrix metalloproteinases (MMPs) with collagenolytic activity and for MMPs with gelatinolytic activity. Collagenolytic enzymes included MMP-1 (interstitial collagenase), MMP-8 (neutrophil collagenase) and MMP-13 (collagenase-3). Gelatinolytic enzymes included MMP-2 (72-kDa gelatinase A/type IV collagenase) and MMP-9 (92-kDa gelatinase B/type IV collagenase). Inhibitors of MMP activity including tissue inhibitor of metalloproteinases-1 and -2 (TIMP-1 and TIMP-2) were also assessed. All three collagenases and both gelatinases were detected immunochemically. MMP-1 appeared to be responsible for most of the functional collagenolytic activity while gelatinolytic activity reflected both MMP-2 and MMP-9. MMP inhibitor activity was also present, and appeared, based on immunochemical procedures, to reflect the presence of TIMP-1 but not TIMP-2. As a group, tumours identified as having aggressive-growth histologic patterns were not distinguishable from basal cell carcinomas with less aggressive-growth histologic patterns. In normal skin, the same MMPs were detected by immunochemical means. However, only low to undetectable levels of collagenolytic and gelatinolytic activities were present. In contrast, MMP inhibitor activity was comparable to that seen in tumour tissue. In previous studies we have shown that exposure of normal skin to epidermal growth factor in organ culture induces MMP up-regulation and activation. This treatment concomitantly induces stromal invasion by the epithelium (Varani et al (1995) Am J Pathol146: 210–217; Zeigler et al (1996 b) Invasion Metastasis16: 11–18). Taken together with these previous data, the present findings allow us to conclude that the same profile of MMP/MMP inhibitors that is associated with stromal invasion in the organ culture model is expressed endogenously in basal cell carcinomas of skin. © 2000 Cancer Research Campaign

Photoaging is significantly different from chronological aging in both clinical and histological appearance. It has been suggested that oxidative stress, generated by ultraviolet radiation (UVR), leads to photoaging over a long period. The presence of 8-OHdG, and oxidatively modified proteins such as 4-hydroxy-2-nonenal-modified protein, 3-L-nitro-tyrosine and N(-epsilon) (carboxymethyl)lysine in UV-exposed skin specimens, supports this theory. The pathophysiology of photoaging of the skin caused by chronic inflammation after UVR is reviewed and discussed, with a focus on oxidative stress.

No abstract