XPA-Deficiency in Hairless Mice Causes a Shift in Skin Tumor Types
Total Page:16
File Type:pdf, Size:1020Kb
Oncogene (1998) 16, 2205 ± 2212 1998 Stockton Press All rights reserved 0950 ± 9232/98 $12.00 http://www.stockton-press.co.uk/onc XPA-de®ciency in hairless mice causes a shift in skin tumor types and mutational target genes after exposure to low doses of U.V.B. Annemieke de Vries1,3,5, Rob JW Berg2, Susan Wijnhoven3, Anja Westerman3, Piet W Wester4, Coen F van Kreijl3, Peter JA Capel1, Frank R de Gruijl2, Henk J van Kranen3 and Harry van Steeg3 Departments of 1Immunology, 2Dermatology, University of Utrecht, Heidelberglaan 100, 3584 CX Utrecht; 3National Institute of Public Health and the Environment, Laboratory of Health Eects Research, Department of Carcinogenesis, Mutagenesis and Genetics; 4Laboratory of Pathology and Immunobiology, PO Box 1, 3720 BA Bilthoven, The Netherlands Xeroderma pigmentosum (XP) patients with a defect in arrest, DNA repair, apoptosis and immunological the nucleotide excision repair gene XPA, develop tumors responses (Mukhtar and Elmets, 1996; Kraemer, with a high frequency on sun-exposed areas of the skin. 1997). The central role of DNA damage in skin Here we describe that hairless XPA-de®cient mice also carcinogenesis and the importance of an ecient develop skin tumors with a short latency time and a DNA repair mechanism to eliminate the U.V.-induced 100% prevalence after daily exposure to low doses of DNA damage, is best illustrated by the human U.V.B. Surprisingly and in contrast to U.V.B.-exposed heritable disease xeroderma pigmentosum (XP) repair pro®cient hairless mice who mainly develop (Cleaver and Kraemer, 1995). Seven complementation squamous cell carcinomas, the XPA-de®cient mice groups exist in XP (XP-A to XP-G), each caused by a developed papillomas with a high frequency (31%) at a defect in a dierent gene involved in nucleotide U.V. dose of 32 J/m2 daily. At the highest daily dose of excision repair (NER). Due to the inability to repair 80 J/m2 mainly squamous cell carcinomas (56%) and U.V.-induced lesions, cells isolated from XP patients only 10% of papillomas were found in XPA-de®cient are highly sensitive to U.V. radiation. Furthermore, XP hairless mice. p53 gene mutations were examined in patients develop skin tumors with an extremely high exons 5, 7 and 8 and were detected in only 3 out of 37 of frequency (41000-fold increase compared to normal these skin tumors, whereas in tumors of control U.V.B.- individuals). The tumors comprise, like in the general irradiated wild type littermates this frequency was higher population, mainly basal cell carcinomas (BCC) and (45%) and more in line with our previous data. squamous cell carcinomas (SCC) at sun-exposed areas Strikingly, a high incidence of activating ras gene of the skin, and melanomas, which are more randomly mutations were observed in U.V.B.-induced papillomas distributed (Cleaver and Kraemer, 1995). (in 11 out of 14 tumors analysed). In only two out of 14 Only two of the genetic events that occur in human squamous cell carcinomas we found similar ras gene skin carcinogenesis are known in some detail; i.e. mutations. The observed shift from squamous cell alterations in the p53 tumor suppressor gene and carcinomas in wild type hairless mice to papillomas in although found at a much lower frequency, in ras XPA-de®cient hairless mice, and a corresponding shift in oncogenes (Van der Schroe et al., 1990; Ananthas- mutated cancer genes in these tumors, provide new clues wamy and Pierceall, 1990; Campbell et al., 1993). In on the pathogenesis of chemically- versus U.V.B.-induced addition, locations for several putative tumor suppres- skin carcinogenesis. sor genes have recently been identi®ed by LOH analysis (Rees, 1994). Mutations in the p53 gene have Keywords: XPA; UVB; p53; ras; papilloma; squamous frequently been found in human non-melanoma skin cell carcinoma tumors (90% in SCCs and 50% in BCCs) (Brash et al., 1991; Ziegler et al., 1993; Daya-Grosjean et al., 1995; Nataraj et al., 1995; Matsumura et al., 1996; Kraemer, 1997). In the p53 gene, C?T transitions and CC?TT Introduction tandem transitions are almost exclusively observed at dipyrimidine sites. These mutations are thought to It is generally accepted that the development of non- display the unique ®ngerprint of DNA damage caused melanoma skin cancer in humans is correlated with by U.V.B.-irradiation (Bredberg et al., 1986; Dorado et exposure to sunlight (IARC Monographs, 1992). Skin al., 1991; Yagi et al., 1991). In skin tumors isolated tumor development is a multistep process initiated from XP patients, the relative occurrence of these when genomic DNA is damaged by U.V.-light, and not distinctive U.V.-induced CC?TT transitions is even properly repaired. Following the induction of DNA more frequent in the p53 gene (Sato et al., 1993; damage, several cellular processes have been identi®ed Dumaz et al., 1994). to occur, such as accumulation of the p53 protein, The genetic analysis of mouse skin tumors induced overexpression of certain p53-regulated genes, cell cycle in a controlled way by U.V.B.-irradiation has provided additional evidence for the causal involvement of p53 gene mutations. Several laboratories have analysed skin Correspondence: H van Steeg tumors (mainly SCCs) induced by U.V.B. in dierent 5 Current address: Center for Cancer Research, MIT, Cambridge mouse strains for mutations in the p53 gene (Kress et Massachusetts 02139, USA Received 26 August 1997; revised 24 November 1997; accepted 25 al., 1992; Kanjilal et al., 1993; Van Kranen et al., November 1997 1995). Dierences in frequencies of p53 gene mutations U.V.B.-induced skin tumorigenesis in hairless XPA-deficient mice A de Vries et al 2206 between mouse genetic backgrounds were detected. However, our recently compiled mutational spectrum in the p53 gene from U.V.B.-induced hairless mouse skin tumors, resembles to a great extent the spectrum observed in human skin tumors (Dumaz et al., 1997). In hairless mice, the majority of the detected mutations in the p53 gene were also C?T transitions, or (to a lesser extent) CC?TT double transitions. Most of the p53 mutations in these mice were found at dipyrimidine sites in the non-transcribed strand, which is in line with the observation that DNA repair of cyclobutane pyrimidine dimers on this strand is in-ecient in hairless mice (Ruven et al., 1994). We have recently generated mice with a defect in one of the genes involved in NER, i.e. the XPA gene (De Vries et al., 1995). These XPA-de®cient mice display a skin cancer proneness that is highly comparable to that observed in human XP-A (De Vries et al., 1995; Berg et al., 1997). To enable and facilitate the analysis of the eects of U.V.B. irradiation in more detail, the XPA- de®cient mice were crossbred with hairless (HRA : SKH) mice. HRA : SKH mice have been extensively studied in U.V.-exposure experiments, and therefore, the relationship between U.V.B. irradiation and the carcinogenic response is well established (Ruven et al., 1994; De Gruijl and Forbes, 1995; Van Kranen et al., 1995; Dumaz et al., 1997). Figure 1 Incidence and latency time of U.V.B.-induced skin In the present study we show that, unexpectedly, tumors of hairless XPA-de®cient mice. & 0 J/m2, * 32 J/m2, ~ 80 J/m2. Once mice had developed a tumor 52 mm in size, they hairless XPA-de®cient mice develop next to the usually 2 observed actinic keratoses (AK) and SCCs, high were counted. One animal exposed to 32 J/m /day died of unknown reasons before the ®rst skin tumor in this exposure numbers of papillomas on their skin after exposure group was detected, and was, therefore, excluded from the to low levels of U.V.B. Moreover, papillomas were experiment much more frequently found in the lowest dose group (32 J/m2/day) than in the highest dose group (80 J/m2/ day). Surprisingly, we found in these skin papillomas dose of 32 J/m2/day developed skin tumors (accumu- of hairless XPA-de®cient mice a high frequency (11 out lated dose of 3.0 kJ/m2, 1 out of 13). At this time of 14) of ras gene mutations, resembling one of the point, 57% of the XPA7/7 mice exposed to 80 J/m2/day genetic changes observed in chemically-induced mouse were tumor bearing. It is apparent from Figure 1, that skin papillomas (Brown et al., 1990; 1995). In contrast tumor development on the skin of XPA-de®cient mice to the carcinomas found in repair pro®cient hairless occurred in a dose-related manner. After 129 (80 J/m2/ mice, the tumors found in XPA-de®cient mice (both in day) and 144 days (32 J/m2/day), all XPA-de®cient papillomas and carcinomas) displayed a very low mice had developed one or more tumors. At these time incidence of mutations in the conserved domains of points, the accumulated doses were 10.3 kJ/m2 and the p53 gene. These results will be discussed in relation 4.6 kJ/m2 for the two exposure groups, respectively. In to the present views on the pathogenesis of human and comparison, in previous studies (De Gruijl and Forbes, murine skin cancer. 1995; Van Kranen et al., 1995) using much higher daily doses (i.e. 1300 J/m2/day), we showed that a 100% prevalence of skin tumors was reached in wild-type HRA : SKH mice after exposure to an accumulated Results dose of app. 160 kJ/m2 of U.V.B.