Dose-dependent effects of UVB-induced skin carcinogenesis in hairless p53 knockout mice

Mutation Research 571 (2005) 81–90

Dose-dependent effects of UVB-induced skin carcinogenesis in hairless p53 knockout mice Henk J. van Kranen a, ∗ , Anja Westerman a , Rob J.W. Berg b , Nicolien Kram a , Coen F. van Kreijl a , Piet W. Wester a , Frank R. de Gruijl b a

National Institute of Public Health and Environment, Laboratory of Toxicology, Pathology and Genetics, Department of Carcinogenesis Mutagenesis and Aging, P.O. Box 1, Anthonie van Leeuwenhoeklaan 9, 3720 BA, Bilthoven, The Netherlands b Leiden University Medical Center, Department of Dermatology, Wassenaarseweg 72, 2333 AL, Leiden, The Netherlands Received 30 June 2004; accepted 16 July 2004

Abstract Exposure to (solar) UVB radiation gives rise to mutations in the p53 tumor suppressor gene that appear to contribute to the earliest steps in the molecular cascade towards human and murine skin cancer. To examine in more detail the role of p53, we studied UVB-induced carcinogenesis in hairless p53 knock-out mice. The early onset of lymphomas as well as early wasting of mice interfered with the development of skin tumors in p53 null-mice. The induction of skin tumors in the hairless p53+/− mice was accomplished by daily exposure to two different UV-doses of approximately 450 J/m2 and 900 J/m2 from F40 lamps corresponding to a fraction of about 0.4 and 0.8 of the minimal edemal dose. Marked differences in skin carcinogenesis were observed between the p53+/− mice and their wild type littermates. Firstly, at 900 J/m2 , tumors developed significantly faster in the heterozygotes than in wild types, whereas at 450 J/m2 there was hardly any difference, suggesting that only at higher damage levels loss of one functional p53 allele is important. Secondly, a large portion (25%) of skin tumors in the heterozygotes were of a more malignant, poorly differentiated variety of squamous cell carcinomas, i.e. spindle cell carcinomas, a tumor type that was rarely observed in daily UV exposed wild type hairless mice. Thirdly, the p53 mutation spectrum in skin tumors in heterozygotes is quite different from that in wild types. Together these results support the notion that a point mutation in the p53 gene impacts skin carcinogenesis quite differently than allelic loss: the former is generally selected for in early stages of skin tumors in wild type mice, whereas the latter enhances tumor development only at high exposure levels (where apoptosis becomes more prevalent) and appears to increase progression (to a higher grade of malignancy) of skin tumors. © 2005 Elsevier B.V. All rights reserved. Keywords: Skin cancer; UVB-induced; Hairless mice; p53 deficient

∗

Abbreviations: UVB, ultra violet B; SCC, squamous cell carcinoma Corresponding author. Tel.: +31 30 2742182/96; fax: +31 30 2744446. E-mail address: [email protected] (H.J. van Kranen).

0027-5107/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.mrfmmm.2004.07.018

82

H.J. van Kranen et al. / Mutation Research 571 (2005) 81–90

1. Introduction It is generally recognized that the development of non-melanoma skin cancer in humans is associated with exposure to sunlight [1]. In general, (skin) tumorigenesis is considered a multi-step process driven by different genetic alterations [2]. For skin carcinomas, a major part of these alterations may be attributable to UV-light exposure [3]. Following the induction of DNA damage, several cellular processes have been identified to occur, such as accumulation of the p53 protein, over expression of certain p53-responsive genes, cell cycle arrest, DNA repair, apoptosis and immunological responses [4–6]. Especially, cell cycle arrest, apoptosis and enhanced DNA repair are important p53 mediated responses [7,8]. The p53 tumor suppressor gene is the most frequently mutated gene found in human cancers [9]. In addition, people with Li-Fraumeni syndrome, who inherit a mutation in one allele of p53, suffer from a high incidence of malignancies at early age [10]. Clearly, dysfunctional p53 contributes to tumorigenesis in a wide variety of tissues. This is best illustrated by the pivotal role of p53 in integrating numerous signals controlling cell life and death, although this signaling network is still far from completely understood [11]. Exposure to (solar) UVB (290–320 nm) has been demonstrated to give rise to UVB-fingerprint mutations in the p53 tumor suppressor gene, which appear to contribute to the earliest steps in the molecular cascade towards human and murine skin cancer [12–14]. The initiating aspect of these UV-fingerprint mutations is supported by observations that these p53 alterations are already present in microscopic clusters of cells, long before tumors with similar p53 alterations become visible [15,16]. However, sunlightassociated skin cancers have thus far not been noted to occur more frequently in persons with Li-Fraumeni syndrome [10,17]. To further examine the role of (functionally impaired) p53 in skin carcinogenesis, it would be particularly desirable to study UVB-induced carcinogenesis in hairless genetically modified mice harboring alleles with these UV-specific point mutations. Due to initial technical difficulties, mice with these socalled p53 knock-in alleles are only now being generated and phenotyped [18–20]. For initial studies on the effects of p53 dysfunction in skin carcinogenesis, we as well as others [21,22] have worked with

the p53 knock-out mice, lacking one or both functional copies of p53. Both heterozygous (+/−) and homozygous (−/−) p53 deficient animals develop a spectrum of spontaneous tumors, primarily lymphoid malignancies and various types of sarcomas [23,24]. The frequency of tumors is markedly enhanced either by exposing the mice to a single dose of ionizing radiation [25] or by exposing them to other genotoxic agents [26]. In a study on chemically induced skin cancer, p53+/− and p53−/− mice did not show increased susceptibility to papilloma induction, but the progression to carcinomas was enhanced compared to wild type mice [27]. Three studies investigated the impact of loss of p53 in UV carcinogenesis in shaved haired mice [21,22,28]. However, the albino hairless mouse is considered to be a more suitable model for UV-induced skin carcinogenesis. Comparable to humans, these mice develop exclusively epidermal tumors, skin carcinomas and precursor lesions (actinic keratoses, AK), under chronic UV exposure. In contrast, shaved haired mice commonly develop fibrosarcomas and tumors on the eyes. We investigated how the lack of a p53 allele would affect the kinetics and the tumor progression at different UV exposure levels in this hairless mouse model. More specifically, we ascertained

(a) whether the tumor induction rate was equally strongly increased at different levels of daily exposure (UV-induced wild type expression of p53 shows a strong UV dose dependency, and therefore the lack of a p53 allele may have a larger impact at higher dosages), (b) whether the tumors would display a different pattern of mutations in the remaining wild type p53 allele (maybe the lack of one p53 allele diminishes the developmental advantage that early tumor cells derive from acquiring a mutation in the remaining wild type p53 allele, i.e. a lowered and perhaps shifted selection of p53 mutations), (c) whether the effect would be restricted to the early stages of tumor development, or whether it would also affect the late stages of tumor progression to higher grades of malignancy (LOH of p53 is known to have this effect in chemically induced skin carcinogenesis [27]).

H.J. van Kranen et al. / Mutation Research 571 (2005) 81–90

2. Materials and methods

83

tion analysis in the p53 or ras genes (see below for details).

2.1. Mice 2.3. Immunohistochemistry p53-null mice [24] were obtained from Jackson Laboratories and crossed with albino hairless mice (HRA:SKH inbred strain, colony kept at the animal facility of the Utrecht University), to obtain hairless p53 deficient mice. Hairless p53+/− (F4) mice were intercrossed to generate sufficient numbers of heterozygous, homozygous p53 deficient mice and wild type littermates for a chronic UVB exposure experiment. Genotyping of the mice was performed by PCR analysis of DNA isolated from tail tips by standard methods. All mice (both females and males) were 6–10 weeks of age at the start of the UVB irradiation. 2.2. UVB exposure of the mice and processing of the tumors Hairless (F4 generation) p53+/+ , p53+/− (n = 28 for each genotype) and p53−/− mice (n = 15) were irradiated daily with UVB radiation. The UVB source comprised Philips F40 lamps (maximum output around 310 nm). Two exposure groups were used, one at 900 J/m2 per day corresponding to ±80% of the minimal edema/erythema dose (MED) of these mice and the other at 450 J/m2 per day (40% MED). Control (non-irradiated) groups consisted of seven mice of each of the three genotypes. Housing and irradiation setup was essentially as described previously [29]. All mice were checked weekly for the development of tumors. Mice carrying tumors ≥4 mm in size were withdrawn from the UVB exposure at least 48 h prior to the tumor isolation to circumvent p53 overexpression due acute UV irradiation [15]. Mice were sacrificed, and the tumors were carefully removed. For histology and immunohistochemical staining, one portion of the tumors was formalin fixed for 4 h, stored in 70% ethanol and paraffin embedded according to routine procedures. For histological analysis, deparaffinized 5 ␮m sections were subsequently stained with haematoxylin and eosin, and immunohistochemically for p53 (see below). The other portion of the tumor was snap frozen in liquid N2 and stored at −70 ◦ C for molecular analysis. After careful dissection of normal skin tissue to enrich for tumor tissue, genomic DNA was isolated (QIAamp Tissue Kit 250, Qiagen) for muta-

Staining for p53 expression was performed on deparaffinized 5 ␮m sections of the embedded tumors with the CM-5 rabbit polyclonal antibody, raised against the mouse p53 protein (kindly provided by Dr. David Lane) as described previously [15]. For the differentiation of epithelial and mesenchymal cells, paraffin sections of selected tumors were preincubated in trypsin, and stained immunohistochemically for vimentin and cytokeratin according to routine procedures. Primary antibodies were rabbitantihuman cytokeratin (DAKO), rabbit-antihuman vimentin (E-YLab) both cross-reacting with the murine proteins; as secondary antibody peroxidase-labelled swine-anti rabbit (DAKO) was used. Sections were counterstained with hematoxylin. 2.4. Detection of p53 gene mutations To increase the sensitivity of detecting mutant p53 alleles within heterogeneous tumor samples, the DGGE technique [30,31] was applied for the exons 5, 6, 7 and 8 in two rounds of amplification. In the first round, the genomic fragment of p53 encompassing a part of intron 4 up till intron 8 was amplified. For the second round, nested exon-specific primers were used, one of which contained a 40 bp 5 -GC-clamp. The composition and design of the GC-clamped primers is essentially as described previously [32]. An overview is presented in Table 1. Exons 5, 6, 7 and 8 of the p53 gene of DNA isolated from skin tumors were individually amplified and subsequently sequenced using a procedure essentially as described previously [33]. The cycle-sequencing reactions were performed with Thermo-Sequenase (Amersham) according to the manufacturers protocol. 2.5. Detection of ras gene mutations Codon 12 of the H-ras gene was analyzed for the presence of activating point mutations using a dot–blot procedure as described previously [33]; two oligomers were used for the UVB-inducible mutations, GGA to either AGA or GAA on the non-transcribed strand. Fil-

84

H.J. van Kranen et al. / Mutation Research 571 (2005) 81–90

Table 1 PCR primers and DGGE parameters for mutation analysis of p53 p53 exon

PCR product

Denaturing gradient (%)

Primer sequences

5

375 bp

40–70

Sense: 5 -GC-clamp-CTTCATTAGTTCCCCACCTTG-3 Antisense: 5 -AGAGCAAGAATAAGTCAGAA-3

6

313

45–65

Sense; 5 -GCCTGTGGGGTTAGGACTGG-3 Antisense:5 -GC-clamp-CTCAGGAGGGTGAGGCAAACG-3

7

243

45–65

Sense:5 -TGTAGTGAGGTAGGGAGCGACTT-3 Antisense: 5 -GC-clamp-CTGGGGAAGAAACAGGCTAA-3

8

322

40–70

Sense; 5 -GC-clamp-CTAGTTTACACACAGTCAGGATG-3 Antisense; 5 -GGCTCCTCCGCCTCCTTGGT-3

GC-clamp: 5 -CGCCCGCCGCGCCCCGCGCCCGTCCCGCCGCCCCCGCCC.

ters were applied, using a 1 h prehybridisation step at 56 ◦ C, followed by O/N hybridisation with 10 pmol of 3 -DIG end labeled allele specific oligonucleotide (final volume 5 ml). For detection, an anti-DIG–AP conjugate antibody with CPD-Star as a substrate (both Boehringer) were used, according to the manufacturers protocol.

3. Results 3.1. Tumor induction and dose–response We did not attain a reliable tumor response in the hairless p53−/− mice for the following reasons. The overall ratio of genotypes (103 mice, p53+/+ :p53+/− :p53−/− = 36:47:17) of the hairless p53 mice was similar to that reported in the literature [34]. This severely interfered with our intended study design because 13 p53−/− mice could be allocated to the 900 J/m2 per day exposure group, but only 2 to the 450 J/m2 per day group. In addition, most p53−/− mice died before or during the onset of the first skin tumor. Pathology reports of these animals were indicative for malignant lymphomas, which have been reported to occur spontaneously in p53-null mice at an early age [23,24]. Importantly, the condition of the p53−/− mice was rapidly detoriating which manifestly hampered the development (outgrowth) of the skin tumors. Together this resulted in far too few p53−/− mice in which UVBinduced skin tumors could effectively develop. These complications did not occur in the p53+/− mice. In the course of the experiment, all p53+/− and p53+/+ mice contracted multiple skin tumors. A dose

dependent difference in tumor response was observed between the p53+/− mice and their wild type littermates. At 900 J/m2 , tumors developed significantly faster in the p53+/− mice compared to the wild types, whereas at 450 J/m2 the difference was less pronounced (prevalence in Fig. 1A). Also with regard to the tumor yield, a clear difference between the p53+/− mice and the wild types at 900 J/m2 was observed during the first 15 weeks of UVB irradiation (Fig. 1B, a shift in time by approximately 20% to reach one tumor per animal), which again was less in the 450 J/m2 dose group (about a 10% shift in time) (Fig. 1B). There was also a remarkable difference in tumor histology between wild type mice and p53 heterozygous mice for both UV doses. A substantial part (∼25%) of the skin tumors in the p53 heterozygotes were characterized as spindle cell carcinomas (a poorly differentiated, more malignant type of the squamous cell carcinoma), which are very infrequently (∼ < 5%) observed among UVB-induced tumors in wild type hairless mice (see Fig. 2). 3.2. p53 mutation spectrum In total 77 tumors were analyzed for the presence of mutations in the exons 5–8 of the p53 gene. The majority of the tumors selected, originated from the p53 heterozygous mice; the mutation spectrum in wild type mice was extensively characterized in an earlier study [13]. The distribution among different genotypes and the two UVB doses is presented in Table 2. The observed mutation frequencies of p53 in wild type mice of respectively 100% and 83%, as shown in Table 3, are higher than previously reported [13]

H.J. van Kranen et al. / Mutation Research 571 (2005) 81–90

85

Fig. 1. Comparison of tumor induction among p53+/− and p53+/+ hairless mice. (A) Prevalence curve (probability scale) of UVB-induced squamous cell carcinomas (>1 mm) vs. time (log scale). (B) Tumor yield of all tumors vs. time (log–log plot).

and most likely due to the higher sensitivity for mutant alleles of the DGGE pre-screening technique applied. In Fig. 3, two examples, one for exon 7 and one for exon 8, of typical DGGE gels are presented. Sample numbers 19 (Fig. 3A) and 1 (Fig. 3B) are typical homoduplex bands from wild type p53, whereas the mutants are characterized by additional heteroduplex and/or homoduplex bands. From Table 2, it is clear that the majority of mutations are C > T transitions located on the non-transcribed strand. The summary presented in Table 3 shows a decrease in the percentage of (C > T) mutations in the p53+/− mice irradiated with 450 J/m2

when compared with wild type littermates. The number of mutations in exons 5, 6 and 8 is dropping by almost a factor of two. Only in 66% of the tumors analyzed from the p53 heterozygotes irradiated with 450 J/m2 per day p53 mutations were observed and, in contrast to the other groups, no codon 267 hotspot mutations were detected. 3.3. Ras mutations As found earlier in wild type hairless mice, none of the 77 tumors analyzed for p53 mutations showed

86

H.J. van Kranen et al. / Mutation Research 571 (2005) 81–90

Fig. 2. Hematoxylin stained sections of (A) a spindle cell carcinoma and (B) a squamous cell carcinoma.

a UV-related point mutation [33] in codon 12 of H-ras.

4. Discussion Our results show that the impact of p53 ko heterozygosity on the rate of tumor induction is clearly dependent on the level of UV exposure: at 0.8 MED/day there is a much faster tumor induction in the heterozygous mice than in the wild type mice, but at 0.4 MED/day this effect is less. In the heterozygous mice, there is still a preponderance of tumors with UV-related mutations in the remaining wild type p53 allele. However, we

did find indications of a shift in the mutation spectrum when compared to the tumors in the wild type mice, especially with the low UV exposure (0.4 MED/day) where we observed a complete lack of 267 hotspot mutations. Besides speeding up the tumor development at an early stage, the lack of a p53 allele also clearly enhanced late stage progression of the carcinomas to higher grades of malignancy, in which the tumors became less differentiated and their cells became spindle shaped. Hence, this study in albino hairless mice underscores both the importance of the level of p53 gene dosage necessary for protection against UVB-induced skin carcinogenesis in conjunction with the level of DNA damage needed. These results correlate well with the demonstration that p53+/− mice exhibit markedly reduced sunburn cell formation in the epidermis after UV irradiation compared to wild type mice [35]. Also Friedberg and co-workers are eluding to a more prominent role for p53 in initiating apoptosis after DNA damage compared to arresting normal proliferation based on their work with Xpc deficient mice [36]. The major role of p53 for the induction of apoptosis in the skin has also been confirmed with more recent in vivo studies [37]. Also a correlation with the results presented on similar experiments in haired mice is observed especially with regard to the tumor types induced [22]. These authors also report an increased frequency of mesenchymal tumors like sarcomas and fibrosarcomas. Interestingly, this tumor type has very infrequently been observed in UVB carcinogenesis in wild type hairless mice, but chemically induced squamous cell carcinomas progress to this state after losing their wild type p53 allele [38]. Remarkably, a dose–response for UVB was hardly observed in the experiments of Jiang and co-workers. They applied even higher UVB doses (15 kJ/(m2 week) and 39 kJ/(m2 week)) compared to our UV doses (3.2 kJ/(m2 week) and 6.4 kJ/(m2 week)). Our data show that at 900 J/m2 tumors developed significantly faster in the heterozygotes than in wild types, whereas at 450 J/m2 the difference was much smaller, suggesting that only at higher damage levels loss of one functional p53 allele is important. Support for this view comes from several papers dealing with the effects of p53 haploinsufficiency [39,40], which was recently elegantly discussed by Peter Hohenstein [41]. The observed difference at 450 J/(m2 day) between wild type

H.J. van Kranen et al. / Mutation Research 571 (2005) 81–90

87

Table 2 Overview of p53 mutations as detected by direct sequencing Genotypemice

UVB dose (J/m2 )

Exon

Codon

Sequence

# of mutations

Aa. change

p53+/+

450

5 5 6 6 6 6 6 7 8

146 152 184 184 188 207 218 242 267

CCa → TTa Cgc → Tgc cCt → cTt cCt → cGt cTt → cCt Cgc → Tgc GaG → AaA Cgc → Tgc Cgt → Tgt

1 1 2 1 1 1 1 1 4

Pro → Leu Arg → Cys Pro → Leu Pro → Arg Leu → Pro Arg → Cys Glu → Lys Arg → Cys Arg → Cys

p53+/−

450

5 5 5 6 6 6 7 7 8 8 8 8

127 130 146 184 185 207 235 236 272 272 272 277

cTa → cCa Cag → Tag CCa → TTa cCt → cTt cCc → cTc Cgc → Tgc tCc → tTc tGc → tTc Cct → Act cCt → cAt CCt → ATt Cgt → Tgt

1 1 2 1 1 2 2 1 1 1 1 1

Leu → Pro Gln → Stop Pro → Leu Pro → Leu Pro → Leu Arg → Cys Ser → Phe Cys → Phe Pro → Thr Pro → His Pro → Ile Arg → Cys

p53+/+

900

5 6 6 6 8 8 8

173 184 207 209 267 272 272

Cat → Tat cCt → cTt Cgc → Tgc aGc → aAc Cgt → Tgt Cct → Tct cCt → cTt

1 1 2 1 2 1 1

His → Tyr Pro → Leu Arg → Cys Ser → Asn Arg → Cys Pro → Ser Pro → Leu

p53+/−

900

5 5 6 8 8

123 153 207 267 272

Ccc → Tcc gCc → gTc Cgc → Tgc Cgt → Tgt cCt → cTt

1 1 2 3 3

Pro → Ser Ala → Val Arg → Cys Arg → Cys Pro → Leu

mice and p53 heterozygotes in mutation spectrum, together with hardly any difference in tumor latency and yield, could implicate that selection for mutations in other genes is more important for skin tumor development in the p53 heterozygotes. The number of tumors harboring point mutations in their DNA binding domain (DBD) region was higher in the present study than previously reported [13,33]. This is most likely due to the improved sensitivity of detecting mutant alleles by the DGGE pre-screening method. Although the absolute numbers are small, it is remarkable that differences seem to exist between the distribution of the two hotspot mutations in codon 267 and codon 272 in the remaining p53 allele between the different groups.

Especially, the absence of codon 267 mutations in the 450 J/m2 irradiated p53+/− mice is noteworthy, maybe indicating different biological effects of the two hotspot mutations. These aspects can certainly be experimentally addressed more adequately when transgenic mice with the specific knock-in alleles containing these point mutations become available in the near future. In an earlier paper [42], we reported on a strongly speeded up induction of early p53-mutant patches in the heterozygous p53 ko mice, i.e. they developed microscopic clusters of cells with mutant-p53 much earlier than their wild type counterpart under daily exposure. We found this difference between the heterozygous and wild type to be strong both at 0.8 MED/day and

88

H.J. van Kranen et al. / Mutation Research 571 (2005) 81–90

Table 3 Tumors analyzed for p53 mutations by direct sequencing and DGGE p53 genotype

UVB (J/m2 )

% CM-5 positive tumors

Number of tumors analyzed

Number of p53 mutant tumors per exon (%) Exon 5 sequenceDGGE; total (%)

Exon 6 sequenceDGGE; total (%)

Exon 7 sequenceDGGE; total (%)

Exon 8 sequenceDGGE; total (%)

+/+

450

80

12

2–3; 4/12 (33)

5–7; 7/12 (58)

1–2; 2/12 (17)

4a –6; 6/12 (50)

12/12 (100)b

58

29

4–4; 5/29 (17)

4–6; 6/29 (21)

3–4; 4/29 (14)

4c –8; 5/24 (21)

19/29 (66)d

83

12

1–3; 3/12 (25)

4–5; 5/12 (42)

0–0; 0/12 (0)

4h –9; 9/29 (31)

10/12 (83)e

0–5; 5/24 (21)

6f –12;

20/24 (83)g

+/− +/+

900

+/− a b c d e f g h

64

24

2–7; 7/24 (29)

2–8; 8/24 (33)

12/24 (50)

Total p53 mutant tumors (%)

All codon 267. Five tumors with >1/multiple mutation(s). Number of codon 267, 3× 272 + 1× 277. Three tumors with >1/multiple mutation(s). Four tumors with >1/multiple mutation(s). 3× codon 267 + 3× codon 272. Eight tumors with >1/multiple mutation(s). 2× codon 267 + 2× codon 272.

Fig. 3. Example of the DGGE analysis: (A) samples of amplicons of exon 7 and (B) of exon 8.

0.4 MED/day. The present observation that the difference in subsequent tumor development between p53+/− and p53+/+ mice is more pronounced at the high exposure levels is therefore remarkable. Apparently, the lacking p53 allele has an impact on the development from a microscopic p53-mutant patch to a tumor. This might be attributable to the putative mechanism that p53-mutant cells carry a lower probability to become apoptotic than their wild type counterparts, and this difference should be more pronounced at high exposure levels, and is proportional to the difference in the number functional p53 alleles in the cells [35] If the daily UV exposure is discontinued, the cells with mutant p53 loose their developmental edge over their neighboring cells and run the risk of being expelled from the basal epidermal layer. In summary, marked differences in skin carcinogenesis were observed between the p53+/− mice and their wild type littermates. At 900 J/m2 , tumors developed significantly faster in the heterozygotes than in wild types, whereas at 450 J/m2 there was almost no difference. A large portion of skin tumors in the heterozygotes were of a more malignant, poorly differentiated variety of squamous cell carcinomas, i.e. the spindle cell carcinomas. Finally, the data show that the p53 mutation spectrum in skin tumors in heterozygotes is quite different from that in the wild types. Together, these results support the notion that a point mutation in

H.J. van Kranen et al. / Mutation Research 571 (2005) 81–90

the p53 gene impacts skin carcinogenesis quite differently than allelic loss: the former is generally selected for in early stages of skin tumors in wild type mice, whereas the latter enhances tumor development only at high exposure levels (where apoptosis becomes more prevalent) and appears to increase progression (to a higher grade of malignancy) of skin tumors.

[14]

[15]

Acknowledgements [16]

We thank T. Hesp and H. Sturkeboom for excellent biotechnical support and Heggert Rebel for assistance with the analysis of the tumor data.

[17] [18]

References [1] IARC, World Health Organization WHO, Solar and ultraviolet radiation, Lyon, IARC monographs on the evaluation of carcinogenic risks of chemicals to humans 55 (1992). [2] D. Hanahan, R.A. Weinberg, The hallmarks of cancer, Cell 100 (January (1)) (2000) 57–70. [3] F.R. de Gruijl, H.J. van Kranen, L.H. Mullenders, UV-induced DNA damage, repair, mutations and oncogenic pathways in skin cancer, J. Photochem. Photobiol. B 63 (1–3) (2001) 19–27. [4] B.B. Zhou, S.J. Elledge, The DNA damage response: putting checkpoints in perspective, Nature 408 (November (6811)) (2000) 433–439. [5] H. Mukhtar, C.A. Elmets, Photocarcinogenesis: mechanisms, models and human health implications, Photochem. Photobiol. 63 (4) (1996) 356–357. [6] K.H. Kraemer, Sunlight and skin cancer: another link revealed, Proc. Natl. Acad. Sci. U.S.A. 94 (1997) 11–14. [7] D.P. Lane, Cancer p53, guardian of the genome, Nature 358 (1992) 15–16 (news; comment). [8] A.J. Levine, p53, the cellular gatekeeper for growth and division, Cell 88 (1997) 323–331. [9] M. Hollstein, D. Sidransky, B. Vogelstein, C.C. Harris, p53 mutations in human cancers, Science 253 (1991) 49–53. [10] D. Malkin, F.P. Li, L.C. Strong, J.F.J. Fraumeni, C.E. Nelson, D.H. Kim, J. Kassel, M.A. Gryka, F.Z. Bischoff, M.A. Tainsky, et al., Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms, Science 250 (1990) 1233–1238 (see comments). [11] B. Vogelstein, D. Lane, A.J. Levine, Surfing the p53 network, Nature 408 (November (6810)) (2000) 307–310 (news, in process citation). [12] D.E. Brash, J.A. Rudolph, J.A. Simon, A. Lin, G.J. McKenna, H.P. Baden, A.J. Halparin, J. Ponten, A role for sunlight in skin cancer: UV-induced p53 mutations in squamous cell carcinoma, Proc. Natl. Acad. Sci. U.S.A. 88 (1991) 10124–10128. [13] N. Dumaz, H.J. van Kranen, A. de Vries, R.J. Berg, P.W. Wester, C.F. van Kreijl, A. Sarasin, L. Daya-Grosjean, F.R. de Gruijl,

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

89

The role of UV-B light in skin carcinogenesis through the analysis of p53 mutations in squamous cell carcinomas of hairless mice, Carcinogenesis 18 (1997) 897–904. L. Daya-Grosjean, N. Dumaz, A. Sarasin, The specificity of p53 mutation spectra in sunlight induced human cancers, J. Photochem. Photobiol. B 28 (1995) 115–124. R.J. Berg, H.J. van Kranen, H.G. Rebel, A. de Vries, W.A. van Vloten, C.F. van Kreijl, J.C. van der Leun, F.R. de Gruijl, Early p53 alterations in mouse skin carcinogenesis by UVB radiation: immunohistochemical detection of mutant p53 protein in clusters of preneoplastic epidermal cells, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 274–278. A.S. Jonason, S. Kunala, G.J. Price, R.J. Restifo, H.M. Spinelli, J.A. Persing, D.J. Leffell, R.E. Tarone, D.E. Brash, Frequent clones of p53-mutated keratinocytes in normal human skin, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 14025–14029. D. Malkin, p53 and the Li-Fraumeni syndrome, Bba, Rev. Cancer 1198 (1994) 197–213. G. Liu, J.M. Parant, G. Lang, P. Chau, A. Chavez-Reyes, A.K. El Naggar, A. Multani, S. Chang, G. Lozano, Chromosome stability, in the absence of apoptosis, is critical for suppression of tumorigenesis in Trp53 mutant mice, Nat. Genet. 36 (1) (2004) 63–68. G.S. Jimenez, M. Nister, J.M. Stommel, M. Beeche, E.A. Barcarse, X.Q. Zhang, S. O’Gorman, G.M. Wahl, A transactivationdeficient mouse model provides insights into Trp53 regulation and function, Nat. Genet. 26 (1) (2000) 37–43. A. de Vries, E.R. Flores, B. Miranda, H.M. Hsieh, C.T. van Oostrom, J. Sage, T. Jacks, Targeted point mutations of p53 lead to dominant-negative inhibition of wild-type p53 function, Proc. Natl. Acad. Sci. U.S.A. 99 (March (5)) (2002) 2948–2953. G. Li, V. Tron, V. Ho, Induction of squamous cell carcinoma in p53-deficient mice after ultraviolet irradiation, J. Invest. Dermatol. 110 (1) (1998) 72–75. W. Jiang, H.N. Ananthaswamy, H.K. Muller, M.L. Kripke, p53 protects against skin cancer induction by UV-B radiation, Oncogene 18 (July (29)) (1999) 4247–4253. L.A. Donehower, M. Harvey, B.L. Slagle, M.J. McArthur, C.A.J. Montgomery, J.S. Butel, A. Bradley, Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours, Nature 356 (1992) 215–221. T. Jacks, L. Remington, B.O. Williams, E.M. Schmitt, S. Halachmi, R.T. Bronson, R.A. Weinberg, Tumor spectrum analysis in p53-mutant mice, Curr. Biol. 4 (1994) 1–7. C.J. Kemp, T. Wheldon, A. Balmain, p53-deficient mice are extremely susceptible to radiation-induced tumorigenesis, Nat. Genet. 8 (1994) 66–69. R.W. Tennant, J.E. French, J.W. Spalding, Identifying chemical carcinogens and assessing potential risk in short-term bioassays using transgenic mouse models, Environ. Health Perspect. 103 (1995) 942–950. C.J. Kemp, L.A. Donehower, A. Bradley, A. Balmain, Reduction of p53 gene dosage does not increase initiation or promotion but enhances malignant progression of chemically induced skin tumors, Cell 74 (1993) 813–822. A.M. Reis, D.L. Cheo, L.B. Meira, M.S. Greenblatt, J.P. Bond, D. Nahari, E.C. Friedberg, Genotype-specific Trp53 mutational

90

[29]

[30]

[31]

[32]

[33]

[34]

[35]

H.J. van Kranen et al. / Mutation Research 571 (2005) 81–90 analysis in ultraviolet B radiation-induced skin cancers in Xpc and Xpc Trp53 mutant mice, Cancer Res. 60 (March (6)) (2000) 1571–1579. R.J.W. Berg, F.R. de Gruijl, J.C. van der Leun, Interaction between ultraviolet-A and ultraviolet-B radiations in skin cancer induction in hairless mice, Cancer Res. 53 (1993) 4212– 4217. S.G. Fischer, L.S. Lerman, DNA fragments differing by single base-pair substitutions are separated in denaturing gradient gels: correspondence with melting theory, Proc. Natl. Acad. Sci. U.S.A. 80 (6) (1983) 1579–1583. E.S. Abrams, S.E. Murdaugh, L.S. Lerman, Comprehensive detection of single base changes in human genomic DNA using denaturing gradient gel electrophoresis and a GC clamp, Genomics 7 (4) (1990) 463–475. R. Smits, A. Kartheuser, S. Jagmohan-Changur, V. Leblanc, C. Breukel, A. de Vries, H. van Kranen, J.H. van Krieken, S. Williamson, W. Edelmann, R. Kucherlapati, P.M. Khan, R. Fodde, Loss of Apc and the entire chromosome 18 but absence of mutations at the Ras and Tp53 genes in intestinal tumors from Apc1638N, a mouse model for Apc-driven carcinogenesis, Carcinogenesis 18 (2) (1997) 321–327. H.J. van Kranen, F.R. de Gruijl, A. de Vries, Y. Sontag, P.W. Wester, H.C.M. Senden, E. Rozemuller, C.F. van Kreijl, Frequent p53 alterations but low incidence of ras mutations in UV-B-induced skin tumors of hairless mice, Carcinogenesis 16 (1995) 1141–1147. V.P. Sah, L.D. Attardi, G.J. Mulligan, B.O. Williams, R.T. Bronson, T. Jacks, A subset of p53-deficient embryos exhibit exencephaly, Nat. Genet. 10 (1995) 175–180. A. Ziegler, A.S. Jonason, D.J. Leffell, J.A. Simon, H.W. Sharma, J. Kimmelman, L. Remington, T. Jacks, D.E. Brash, Sunburn and p53 in the onset of skin cancer, Nature 372 (1994) 773–776.

[36] D.L. Cheo, L.B. Meira, R.E. Hammer, D.K. Burns, A.T.B. Doughty, E.C. Friedberg, Synergistic interactions between XPC and p53 mutations in double-mutant mice: neural tube abnormalities and accelerated UV radiation-induced skin cancer, Curr. Biol. 6 (1996) 1691–1694. [37] D.E. Brash, N.M. Wikonkal, E. Remenyik, G.T. van der Horst, E.C. Friedberg, D.L. Cheo, H. van Steeg, A. Westerman, H.J. van Kranen, The DNA damage signal for Mdm2 regulation, Trp53 induction, and sunburn cell formation in vivo originates from actively transcribed genes, J. Invest. Dermatol. 117 (5) (2001) 1234–1240. [38] A. Balmain, C.J. Kemp, P.A. Burns, A.B. Stoler, D.J. Fowlis, R.J. Akhurst, Functional loss of tumour suppressor genes in multistage chemical carcinogenesis, Princess Takamatsu Symp. 22 (1991) 97–108. [39] S. Venkatachalam, Y.P. Shi, S.N. Jones, H. Vogel, A. Bradley, D. Pinkel, L.A. Donehower, Retention of wild-type p53 in tumors from p53 heterozygous mice: reduction of p53 dosage can promote cancer formation, EMBO J. 17 (August (16)) (1998) 4657–4667. [40] J.E. French, G.D. Lacks, C. Trempus, J.K. Dunnick, J. Foley, J. Mahler, R.R. Tice, R.W. Tennant, Loss of heterozygosity frequency at the Trp53 locus in p53-deficient (+/−) mouse tumors is carcinogen-and tissue-dependent, Carcinogenesis 22 (1) (2001) 99–106. [41] P. Hohenstein, Tumour suppressor genes-one hit can be enough, PLoS. Biol. 2 (2) (2004) E40. [42] H. Rebel, L.O. Mosnier, R.J. Berg, A. Westerman-de Vries, H. van Steeg, H.J. van Kranen, F.R. de Gruijl, Early p53-positive foci as indicators of tumor risk in ultraviolet-exposed hairless mice: kinetics of induction, effects of DNA repair deficiency, and p53 heterozygosity, Cancer Res. 61 (February (3)) (2001) 977–983.