The nucleotide excision repair (NER) system is comprised of two subpathways, i.e., transcription-coupled repair (TCR)and global genome repair (GGR). To establish the relative importance of TCR and GGR for UV effects on the skin, we have used hairless knockout mouse strains lacking either TCR (CSB −/−) or GGR(XPC −/−). In single exposure experiments, we found that CSB −/− mice have a 7–16 times higher susceptibility to sunburn than XPC −/− mice and than heterozygous (+/−) and wild-type (+/+) controls. Exposure to 80 J/m2 UV radiation (i.e., suberythemogenic in CSB −/−) on 10 consecutive days gives rise to epidermal hyperplasia in CSB −/− and XPC−/−, whereas repair-proficient controls do not show epidermal hyperplasia from these exposures. In addition, CSB −/−mice develop marked parakeratosis, whereas XPC −/−mice and controls do not. Under continued exposure to this daily dose,squamous cell carcinomas appear in CSB −/−, XPC−/−, and in the control groups, whereas only in the CSB −/− animals is a fairly high number of benign papillomas also found. The median latency time of squamous cell carcinomas (diameters ≥1 mm) is 84 days for the XPC−/− mice, 115 days for the CSB −/− mice, and 234–238 days for the heterozygous and wild-type control groups. These results indicate that GGR is more important than TCR in protection against UV-induced carcinomas of the skin but not against other UV effects such as sunburn, epidermal thickening, scaling of the stratum corneum, and development of papillomas. These results also indicate that GGR capacity may serve as a better predictor for skin cancer susceptibility than sensitivity to sunburn. The relative cancer susceptibilities of GGR- and TCR-deficient skin could well depend on the balance between an increased mutation rate and the presence (in CSB −/−) or lack (in XPC −/−) of a compensatory apoptotic response.

Non-melanoma skin cancer is the most frequently diagnosed type of cancer in the United States. In addition, non-melanoma skin cancer patients have a 20–30% increased mortality from other cancers,suggesting that skin cancer susceptibility relates to a more general type of cancer susceptibility (1). A potential underlying determinant of cancer susceptibility is the efficiency by which genotoxic damage is removed from the genome (1). The dominant types of DNA lesions responsible for induction of skin cancer are UVB (280–315 nm) radiation-induced CPD3 and 6-4PP. These types of DNA damage are removed from the genome by NER, a versatile repair system responsible for the elimination of a wide class of structurally unrelated DNA lesions (2, 3). In recent years, it has become clear that NER is comprised of two subpathways. The GGR subpathway is responsible for the removal of lesions from the entire genome. In addition, cells have developed the TCR subpathway(4, 5). In this system, the repair machinery is directed preferentially to the template strand of actively transcribed DNA to avoid that unrepaired lesions interfere with transcription by stalling RNA polymerase II (2, 3).

In humans, mutations in NER genes are the cause of the rare genetic diseases XP and CS. Complementation studies have revealed the existence of seven genes involved in XP (XPA through XPG) and two in CS (CSA and CSB; Refs.2 and 6). All XP complementation groups are defective in both GGR and TCR, with the exception of XPC and XPE, which are defective in GGR only (3, 7, 8). A specific defect in TCR is encountered in CSA and CSB (9, 10, 11). Patients within the same complementation group can vary quantitatively in their skin cancer susceptibility, because different mutations in the same gene can lead to different levels of impairment. Because it is very difficult to make well-controlled, qualitative, and quantitative comparisons in cancer susceptibility among the separate XP and CS complementation groups on basis of the available patient data, mouse models for XP and CS have been developed.

XPA(12, 13) and XPC(14, 15, 16) knockout mice show a higher susceptibility to UV-induced skin cancer than their heterozygous and wild-type littermates, and as such resemble human patients with XPA and XPC. In marked contrast to human CS patients, CSB mice are also prone to UV-and chemically induced skin cancer (11), indicating that TCR also protects against carcinogenesis. XPC and CSB mutant mice are very well suited to compare the relative contributions of GGR and TCR to the prevention of skin cancer. However,a direct quantitative comparison of skin cancer susceptibilities between rodent XPC and CSB cannot be made on basis of the published data because: (a) animals of different genetic background were used; (b) experiments were discontinued before the repair-proficient control animals contracted tumors; and (c)different irradiation schedules and different types of lamps were used. To directly compare the effects of defective GGR versus TCR,we have crossed the CSB and the XPC defect into a strain of SKH hairless mice and compared their relative susceptibilities to UV-induced skin cancer under identical experimental conditions. Here we show that GGR deficiency results in a much higher skin cancer susceptibility than TCR deficiency, whereas other UV effects such as sunburn, parakeratosis, and the development of benign papillomas appear to be primarily related to TCR deficiency.

Generation of Hairless XPC and CSB Mice.

The XPC mice (a kind gift of Dr. E. C. Friedberg, Dallas, TX) and CSB mice used in this study have been described in detail previously(11, 14). Shortly, XPC mice were generated by deletion of exon 10 of the XPC gene and in CSB mice a premature stop codon (mimicking a mutation in a CSB patient) was introduced in exon 5 of the CSB gene. The specific GGR and TCR defects in XPC and CSB mouse models have been confirmed by analysis of DNA repair characteristics in mouse embryonic fibroblasts (11, 14).

To make a direct comparison of the relative contribution of GGR and TCR to UV carcinogenesis XPC (129/Sv-C57Bl/6 background) and CSB animals(129/Ola-FVB background) were crossed with HRA/SKH albino hairless mice. Hairless animals do not have to be shaved before exposure and,therefore, allow a very accurate UV dosimetry. After one more backcross with HRA/SKH mice, hairless offspring with albino eyes and heterozygous for XPC and CSB were intercrossed to generate hairless albino XPC and CSB mice. Heterozygous and wild-type littermates were used as controls. Genotyping was performed by PCR analysis of genomic DNA isolated from the tail tips (11, 14). For XPC genotyping, the following three primers were used: (a) XPCex10s 5′-ATTGCGTGCATACCTTGCAC-3′; (b) neo5′-CGCATCGCCTTCTATCGCCT-3′; and (c) XPCin10as 5′-TATCTCCTCAAACCC-TGCTC-3′. Primer a + c identifies the wild-type allele as a ±250-bp PCR fragment, and primer b + c identifies the targeted allele as a ±350-bp PCR fragment. For CSBgenotyping, the following three primers were used: (a) CSBex5s 5′-GCTGCTTATAATAATCCTCATCTCC-3′; (b) neo 5′-ATCTGCGTGTTCGAATTCGCCAATG-3′; and (c) CSBin5as 5′-GTCTTCTG-ATGACGTTAGCTATGAG-3′. Primer a + c identifies the wild-type allele as a ±200-bp PCR fragment, and primer b + c identifies the targeted allele as a ±500-bp PCR fragment. Animal experiments were approved by an ethical committee of Utrecht University, as required by Dutch law.

Single Exposure Experiments.

The dorsal skin of hairless CSB and hairless XPC mice as well as heterozygous and wild-type littermates was exposed to UVB radiation using an Hanovia Kromayer lamp equipped with a Schott-WG305 filter(exposure rate, 135 J/m2/s; 280–400 nm). This hand-held lamp allows short exposures to limited areas of the skin by placing its circular port (∼2.5 cm2) in close contact to the skin (17). For all genotypes, two male and two female mice were exposed to 1, 2, 8, 16, and 24 s of UV radiation from the Kromayer lamp. All exposures were given in duplicate on separate mice. The mice were checked independently by two experienced, but noninformed, observers for erythema and/or edema at 8,24, 36, 48, and 72 h after the exposures.

Chronic Exposure Experiments.

During the chronic exposures (that lasted ∼300 days), the animals were housed separately in macrolon type I cages. Standard mouse chow(Hope Farms RMB-H) and tap water were available ad libitum. The mice were kept at an ambient temperature of 25 ± 1oC. The room was illuminated with yellow fluorescent tubes (Philips TL40W/16) in a 12-h cycle (switched on at 6:00 a.m. and off at 6:00 p.m.). These lamps do not emit any measurable UV radiation. No daylight entered the room. In the irradiation set-up,three fluorescent American Philips F40 sunlamps were mounted above the animal compartments. A perforated metal sheet was positioned between the lamps and the animals to reduce the exposure rate at the mouse level. Fine tuning of the exposure was accomplished by electronic dimmer circuits that controlled the radiant output of the lamps. A schematic representation of the exposure set-up and the relative spectral energy output of the F40 lamps have been published earlier(18).

The groups of repair-deficient animals (CSB −/− and XPC −/−) consisted of 14 mice, with each 7 heterozygous(CSB +/− and XPC+/−) and 7 wild-type (CSB +/+ and XPC +/+) littermates as controls. The different genotypes were placed in random order under the lamps, and all animals were exposed daily to 80 J/m2 (250–400 nm) UV radiation from the Philips F40 lamps. Exposure time was 6 min/day, between 12:30 and 12:36 p.m. In addition, two CSB −/− mice, two XPC −/− mice,and controls were subjected to this dose regimen for 10 days. Mice were killed by cervical dislocation after ether anesthesia, and skin samples were routinely processed (H&E staining) for histopathological examination.

Animal Observations, Definitions, and Data Analysis.

Animals were checked weekly, and all deviations from normal skin appearance (redness, scratching, scaling, tumors, and others) were recorded. The tumor locations were mapped and numbered for each animal separately, and lesions were counted only if the observation could be confirmed during the next check-up. During the experiment, we distinguished between morphologically evident papillomas (pedunculated,protruding tumors with a “cauliflower-like” surface) and other tumors. Previous experiments have shown that macroscopic identification of papillomas corresponds very well with the histological findings, and that the “other tumors” almost exclusively consist of SCCs and their precursor lesions, actinic keratoses (18, 19, 20, 21). Tumors were also subdivided according to diameter: <1 mm, ≥1 mm, and≥2 mm. Mice were sacrificed when carrying at least 1 tumor with a diameter ≥4 mm. The day of the first exposure of an experimental group was defined as t = 0(t50 is the median tumor induction time). If a tumor was first seen at t = ti, and t = ti-1 was the previous checkup time,then the induction time was defined as(ti + ti-1)/2. Graphical representation of the prevalence versus time is based on an actuarial method described by Kaplan and Meier (22) and adapted to carcinogenesis by Peto et al.(23). This procedure computes the chance of tumor-free survival. The death-corrected tumor prevalence is then given by one minus this chance. Individual latency times of first tumors were fitted by a log-normal distribution using a maximum likelihood method for a concise description of the observations in terms of ln(t50) and the SD in ln(time-to-first-tumor/mouse).

Sunburn Is Associated with TCR Deficiency.

Exposure of the mammalian skin to UV light results in a vascular response consisting of erythema and edema (generally known as sunburn). In humans, the just perceptible response at low UV doses peaks between 8 and 24 h after exposure and is dominated by erythema(24, 25). Probably because of differences in the cutaneous vascular density, the peak in the murine skin response occurs between 24 and 48 h after exposure, and this threshold response is dominated by edema (26, 27). In the present study, using the Kromayer lamp as UVB source, all CSB −/− mice showed erythema and edema at 48 and 72 h after exposures of 2 s and higher, whereas 1-s exposures did not result in a visible effect on the skin at any of the time points. The CSB +/− and +/+ mice showed erythema and edema at 48 and 72 h after 16- and 24-s exposures only. Hence, the MED for the CSB −/− mice is between 1 and 2 s (corresponding to 135–270 J/m2), whereas for wild-type and heterozygous mice, the MED is between 8 and 16 s (corresponding to 1080–2160 J/m2). In agreement with our previous data for another XPC mouse model (27), we did not observe a difference in response between the XPC −/−, +/−, and +/+ animals. All three genotypes had an MED between 8 and 16 s (corresponding to 1080–2160 J/m2).

We have established earlier the MED for the Kromayer lamp for hairless XPA knockout mice, which are defective in both GGR and TCR. These results parallel our present findings for the CSB mice, i.e., an MED between 1 and 2 s for XPA −/−and between 8 and 16 s for XPA +/− and +/+(18). Apparently, erythema and edema are triggered by persistent photoproducts in the template strand of transcriptionally active DNA, rather than by lesions in nontranscribed DNA.

Epidermal Hyperplasia Is Associated with TCR and GGR Deficiency.

Epidermal hyperplasia can be considered as a protective reaction against carcinogenesis, because the increase in the number of epidermal cells dilutes unrepaired DNA damage and the thickening of the stratum corneum and stratum spinosum reduces the penetration of UV radiation to the basal cell layer. On the other hand, it can also be argued that the epidermal cell proliferation leading to hyperplasia enhances the mutation rate and thus stimulates carcinogenesis(28). To establish the importance of TCR and GGR for development of epidermal hyperplasia, we sacrificed two XPC−/− mice, two CSB −/− mice, and control animals after 10 days of exposure to 80 J/m2/day (Philips F40 lamps; the daily dose used in the chronic exposure experiment). This dose corresponds to about 70% of the MED for CSB −/−mice but is <10% of the MED for XPC −/− mice and repair-proficient control animals. Fig. 1 shows representative samples of the epidermis of XPC −/−, CSB −/− and control mice after 10 days of exposure. It is evident that both TCR-deficient animals and GGR-deficient animals are prone to epidermal hyperplasia at an exposure level that does not induce a hyperplastic reaction in repair-proficient control animals. Hence, unlike erythema and edema, epidermal hyperplasia appears to be triggered by persistent photoproducts in both transcriptionally active and inactive DNA.

Parakeratosis and Papillomatosis Are Associated with TCR Deficiency.

Hairless CSB −/− mice show parakeratosis after 10 days of daily UV exposure (Fig. 1,B), which is in agreement with earlier findings in shaved CSB −/− mice (11). Under the same experimental conditions XPC −/− mice do not show parakeratosis (Fig. 1,A). The parakeratosis is macroscopically visible as scaling of the stratum corneum and is prominently present at the irradiated dorsal skin of all CSB−/− animals. In contrast, the skin of XPC −/−animals and heterozygous and wild-type controls do not show any macroscopically visible changes, besides the appearance of tumors. Fig. 2 shows representative samples of skin of CSB and of XPC mice when the first tumors appear.

During chronic UVB exposure, hairless mice only develop small numbers of papillomas, whereas during chronic UVA (315–400 nm) exposure,relatively many papillomas occur initially, before being outnumbered by SCCs (20, 21). In the present UVB experiment (with groups of 14 mice) three XPC −/− animals, four repair-proficient XPC controls, and three repair-proficient CSB controls contracted clear papillomas. In contrast, in 13 of the 14 CSB −/− animals a total of 34 papillomas were found. Evidently, the induction of papillomas is related to a TCR deficiency, rather than a GGR deficiency.

Development of SCC Is Predominantly Associated with GGR Deficiency.

In a previous study, we have established the susceptibilities of hairless, totally NER-deficient XPA knockout mice and their heterozygous and wild-type littermates to UV-induced carcinogenesis(18). In the present study, we have measured the relative skin cancer susceptibility of hairless mice with a specific defect in TCR (CSB knockouts) or GGR (XPC knockouts). Histological analysis of an aselect sample of tumors (from XPC knockouts, CSB knockouts, and controls) that were macroscopically classified as “nonpapillomas” confirmed that these tumors were SCCs and their precursors, actinic keratoses (data not shown). If the prevalence of SCC (excluding benign papillomas from further quantitative analysis) is plotted on a probability scale versus time on a logarithmic scale, the points fall along a straight line, a cumulative log-normal distribution. Fig. 3 shows these plots and the log-normal distributions for XPC−/−, CSB −/−, and the control animals for both strains, fitted according to a maximum likelihood method. The optimum values for the median tumor induction time(t50), μ = ln(t50), and the SD (ς) of the fitted log-normal distributions are shown in Table 1 . The prevalence curves for XPC −/−, CSB −/−, and for the repair-proficient controls run roughly parallel (i.e., ςvalues do not differ significantly). Most susceptible to UV carcinogenesis is the XPC knockout strain, which has a median tumor induction time (t50) of 1-mm tumors of 84 days (cumulative UV dose of 6.7 kJ/m2), whereas the CSB knockout strain has a t50 of 115 days(cumulative UV dose of 9.2 kJ/m2). The appearance of skin carcinomas in the XPC controls (both +/−and +/+ mice; t50 of 238 days; cumulative UV dose of 19.0 kJ/m2) does not differ significantly from that in the CSB controls (both +/− and +/+ mice; t50 of 234 days; cumulative UV dose of 18.7 kJ/m2), as tested by the nonparametric trend analysis as described by Peto et al.(23). A separate data analysis of XPC +/− mice versus XPC+/+ mice and of CSB +/− mice versus CSB +/+mice does not show a statistically significant difference between heterozygotes and wild-type animals (data not shown). This is in line with our data published previously showing that XPA +/−mice do not have a higher susceptibility to UV carcinogenesis than XPA +/+ animals (18). In conclusion, our data demonstrate that defective GGR contributes more prominently to SCC development than defective TCR.

In a previous study, we have established the UV-induced skin cancer susceptibility of hairless XPA mice, completely deficient in NER. We have found that mice lacking functional XPA genes develop skin cancer with a latency time that is reduced by a factor 4.2 ± 0.2 compared with heterozygote and wild-type littermates (18). Here, using the same irradiation set-up and exposure times, we have compared UV carcinogenesis in XPC(GGR-deficient) and CSB (TCR-deficient) mice to assess the contributions of GGR and TCR to prevention of skin cancer. Inactivation of XPC and CSB genes appears to speed up UV-induced skin carcinogenesis by a factor of 2.8 ± 0.2(238 days/84 days) and 2.0 ± 0.2 (234 days/115 days),respectively. In the earlier experiment, 1-mm SCCs in the XPA−/− mice appeared with a t50 of 78 days (μ, 4.36 ± 0.04), which is not significantly different from the tumors appearance in XPC −/− mice in the present experiment (t50 = 84 days; μ, 4.43 ± 0.03). However, a direct comparison of t50s is difficult to achieve because in the XPA study, the tumor appearance in heterozygote and wild-type controls was much later(t50 between 313 and 331 days) than in the present study (t50 between 234 and 238 days). Because the experimental conditions (set-up and irradiation protocol) were completely identical, this difference in response of control animals is most likely attributable to a difference in genetic background (i.e., the contribution of residual 129/Sv,129/Ola, C57Bl/6, and FVB DNA in the genome of the F2 hairless mice). This finding underlines the importance of littermates as appropriate controls whereas various mutant lines are not strictly congenic. A comparison between the study with XPA mice and the present data can still be made by determining the factor by which skin carcinoma development is accelerated in the respective NER mutants. Thus, a complete NER deficiency leads to a larger increase in susceptibility to UV-induced carcinogenesis (4.2 times faster than controls) than an isolated deficiency in GGR (2.8 times faster) or TCR (2.0 times faster).

An interesting question is whether the skin cancer susceptibilities of XPC mice (GGR-deficient) and CSB mice (TCR-deficient) “add up” to the susceptibility of XPA mice (GGR + TCR-deficient). This might be an oversimplification because the fact that both XPC/CSB double knockout mice and XPA/CSB double knockout mice die before weaning (29) suggests that there is another function for the CSB protein, besides its role in NER. To answer this question properly, we are currently determining the dose-response relationships for UV carcinogenesis in the hairless XPA,XPC, and CSB mouse strains.

Although important in protection against skin cancer, GGR appears not to play a role in the development of UV-induced erythema/edema,parakeratosis, and papillomatosis. We have shown that these phenomena relate exclusively to the capacity to perform TCR. Erythema and edema occur within 48 h after a single exposure and are therefore not expected to originate from mutations in active genes. A plausible explanation, put forward by Ljungman and Zhang (30), is that persisting photolesions in the template strand of active genes block the vital process of transcription by RNA polymerase II, which triggers apoptosis. They have demonstrated that skin fibroblasts from XPA and CSB patients are highly prone to UV-induced apoptosis,whereas cells from XPC patients are not. Recently, these findings have been confirmed for the XPA, XPC, and CSB hairless mouse models.4The increased sensitivity to UV-induced apoptosis in XPA and CSB, but not in XPC, correlates well with the sensitivities to sunburn found in the present experiment, suggesting that blockage of transcription is also a key event in the development of sunburn. The same holds for the parakeratosis observed in CSB −/− mice (Figs. 1,2). In TCR-deficient animals, parakeratosis can already be found after 10 days of exposure (Fig. 1,B), whereas GGR-deficient animals do not show parakeratosis, not even at the tumor-bearing stage (Fig. 2 A). Disturbed epidermal differentiation in TCR-deficient animals might well be triggered by blockage of RNA polymerase II by persisting photolesions in transcriptionally active DNA.

We have shown that under chronic UVB exposure, CSB −/−mice get, in addition to SCCs, a fair amount of benign papillomas, the first of which appear after about 9 weeks of exposure. This enhanced formation of papillomas is related to the TCR defect, because XPC mice develop these benign tumors at wild-type levels. We have reported recently that hairless XPA mice also develop papillomas under chronic UVB exposure, and that these tumors carry a G → A transition at codon 12 of the H-ras gene (31). We hypothesized that these mutations are caused by a UVB-induced C → T transition at this dipyrimidine site in the transcribed strand as a result of the TCR defect rather than the GGR defect in XPA mice (31). The association of papillomatosis with a TCR deficiency, as found in the present study, strongly supports this inference.

Similar to human XPC (7, 32), XPC mice are proficient in the removal of 6-4PP and CPDs from the transcribed strand of active genes by TCR (14, 27). Also in agreement with human XPC(32), the XPC mouse model is deficient in the removal of 6-4PP from inactive DNA and from the nontranscribed strand of active genes (14, 27). However, in contrast to the situation in humans, rodents are very inefficient in removing CPDs from nontranscribed DNA (33, 34). This is probably because of the lack of expression of the p48 gene, which is required for GGR of CPDs in human cells (8). As a consequence, the effect of the lack of XPC genes is negligible in removal of CPDs in rodent cells, whereas in human XPC removal of CPDs is strongly affected. The high susceptibility of XPC mice to UV carcinogenesis must, therefore, be attributable to a GGR deficiency of DNA lesions other than CPDs, such as 6-4PP. In human XPC, in addition to nonrepaired 6-4PP, nonrepaired CPDs are likely to contribute to the enhanced carcinogenesis. This is in line with the observation that 85%of skin tumors from XPC patients (35) and only 2% of UV tumors from XPC mice (36) contain CPD-derived CC →TT tandem transitions in the nontranscribed strand of the p53 tumor-suppressor gene. Thus, it is to be expected that the observed increase in UV-induced skin cancer proneness in XPC mice represents an underestimation of the susceptibility of human XPC.

The CSB mouse model is deficient in TCR of 6-4PP and CPD. Because mice have efficient GGR of 6-4PP but not of CPD, 6-4PP are removed from the transcribed strand of active genes by GGR. In human CSB, however, both 6-4PP and CPDs are removed from the transcribed strand of active genes by GGR. Hence, we expect that the skin cancer proneness in CSB mice represents an overestimation of the susceptibility of human CSB. This may explain the absence of skin tumors in CSB patients, whereas CSB mice are clearly skin cancer prone (11).

An explanation for the higher carcinogenic effect of a GGR deficiency compared with that of a TCR deficiency could lie in differential mutational activation of transcriptionally inactive cancer genes. It is conceivable that GGR removes DNA photoproducts from transcriptionally inactive proto-oncogenes, thereby preventing mutations by replication of damage-containing sequences. Thus, enhanced mutagenesis attributable to GGR deficiency can lead to increased conversion of proto-oncogenes into oncogenes, which subsequently can stimulate uncontrolled cell divisions. In addition, the importance of GGR relative to TCR can be explained by the finding that TCR deficiency leads to increased apoptosis, whereas GGR deficiency does not. In summary, stimulation or inhibition of carcinogenesis depends on the balance between mutational activation of proto-oncogenes/mutational inactivation of tumor suppressor genes on the one hand and removal of precarcinogenic cells by apoptosis on the other hand. A TCR deficiency can lead to mutational inactivation of actively transcribed tumor suppressor genes but also to increased apoptosis, the net effect of which appears to be mildly procarcinogenic, at least in rodents. A GGR deficiency can also lead to mutational inactivation of actively transcribed tumor suppressor genes(via the nontranscribed strand) as well as to mutational activation of nontranscribed proto-oncogenes, but it does not influence apoptosis. Apparently, the net effect of GGR deficiency is more severely procarcinogenic.

Sensitivity to sunburn is generally considered to be associated with susceptibility to non-melanoma skin cancer (37). The present study shows that sensitivity to sunburn is related to the capacity to perform TCR, whereas susceptibility to SCC is predominantly associated with the capacity to perform GGR. This implies that the GGR capacity may serve as a better predictor for skin cancer susceptibility than sensitivity to sunburn. Determination of MEDs and GGR capacities in non-melanoma skin cancer patients versus healthy controls will show whether this prediction, based on a study with genetically modified mice, holds true for humans.

We thank H. Sturkenboom for maintenance of the animals and A. Westerman-de Vries for genotyping of the animals. We thank Dr. E. C. Friedberg for providing the XPC mice and J. N. Bennen for critically reading the manuscript.

Fig. 1.

Representative H&E-stained sample of dorsal skin of mouse after 10 days of exposure to 80 J/m2 UV radiation from Philips F40 lamps. A, XPC −/−, showing epidermal hyperplasia. B, CSB −/−, showing epidermal hyperplasia and parakeratosis. C, repair-proficient control, showing no difference from unexposed mouse skin.

Fig. 1.

Representative H&E-stained sample of dorsal skin of mouse after 10 days of exposure to 80 J/m2 UV radiation from Philips F40 lamps. A, XPC −/−, showing epidermal hyperplasia. B, CSB −/−, showing epidermal hyperplasia and parakeratosis. C, repair-proficient control, showing no difference from unexposed mouse skin.

Close modal
Fig. 2.

Macroscopic appearance of chronically UV-exposed (80 J/m2/day; Philips F40 lamps) mouse skin at the stage in which the first tumors appear, i.e., about 9 weeks for XPC −/−(A) and about 14 weeks for CSB −/−(B).

Fig. 2.

Macroscopic appearance of chronically UV-exposed (80 J/m2/day; Philips F40 lamps) mouse skin at the stage in which the first tumors appear, i.e., about 9 weeks for XPC −/−(A) and about 14 weeks for CSB −/−(B).

Close modal
Fig. 3.

Kaplan-Meier plot for the prevalence of tumors (diameter between 1 and 2 mm) induced by daily exposure to 80 J/m2 UV radiation for XPC −/− (•), CSB −/−(○), XPC +/−, +/+ controls (▵), and CSB+/−, +/+ controls (▿).

Fig. 3.

Kaplan-Meier plot for the prevalence of tumors (diameter between 1 and 2 mm) induced by daily exposure to 80 J/m2 UV radiation for XPC −/− (•), CSB −/−(○), XPC +/−, +/+ controls (▵), and CSB+/−, +/+ controls (▿).

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Financed mainly by Grant UU97-1531 and also by Grant EUR98-1774 from the Dutch Cancer Society (K. W. F.).

3

The abbreviations used are: CPD, cyclobutane pyrimidine dimer; 6-4PP, pyrimidine [6-4] pyrimidone photoproducts;NER, nucleotide excision repair; TCR, transcription-coupled repair;GGR, global genome repair; XP, xeroderma pigmentosum; CS, Cockayne syndrome; SCC, squamous cell carcinoma; MED, minimal erythema/edema dose.

4

M. Van Oosten, H. Rebel, E. C. Friedberg, H. van Steeg, G. T. J. van der Horst, H. J. van Kranen, A. Westerman,A. A. van Zeeland, L. H. F. Mullenders, and F. R. de Gruijl. Effects of UVB irradiation on cell cycle progression and apoptosis in the epidermis of nucleotide excision repair-deficient mice, submitted for publication.

Table 1

The log-normal parameters, μ [i.e., ln(t50) in days)] and ς, derived from the prevalence for different SCC diameters

Group<1 mm≥1 mm≥2 mm
XPC−/−    
t50  59  84 111 
μ 4.07 ± 0.03 4.43 ± 0.03 4.71 ± 0.03 
ς 0.12 ± 0.02 0.10 ± 0.02 0.10 ± 0.02 
CSB−/−    
t50  98 115 133 
μ 4.59 ± 0.05 4.75 ± 0.05 4.89 ± 0.05 
ς 0.19 ± 0.04 0.18 ± 0.04 0.18 ± 0.04 
XPC+/− and+/+    
t50 217 238 271 
μ 5.38 ± 0.05 5.47 ± 0.04 5.60 ± 0.04 
ς 0.18 ± 0.03 0.14 ± 0.03 0.15 ± 0.03 
CSB+/− and+/+    
t50 210 234 261 
μ 5.35 ± 0.03 5.45 ± 0.02 5.56 ± 0.03 
ς 0.11 ± 0.02 0.08 ± 0.02 0.12 ± 0.02 
Group<1 mm≥1 mm≥2 mm
XPC−/−    
t50  59  84 111 
μ 4.07 ± 0.03 4.43 ± 0.03 4.71 ± 0.03 
ς 0.12 ± 0.02 0.10 ± 0.02 0.10 ± 0.02 
CSB−/−    
t50  98 115 133 
μ 4.59 ± 0.05 4.75 ± 0.05 4.89 ± 0.05 
ς 0.19 ± 0.04 0.18 ± 0.04 0.18 ± 0.04 
XPC+/− and+/+    
t50 217 238 271 
μ 5.38 ± 0.05 5.47 ± 0.04 5.60 ± 0.04 
ς 0.18 ± 0.03 0.14 ± 0.03 0.15 ± 0.03 
CSB+/− and+/+    
t50 210 234 261 
μ 5.35 ± 0.03 5.45 ± 0.02 5.56 ± 0.03 
ς 0.11 ± 0.02 0.08 ± 0.02 0.12 ± 0.02 
1
Kahn H. S., Tatham L. M., Patel A. V., Thun M. J., Heath C. W. Increased cancer mortality following a history of nonmelanoma skin cancer.
J. Am. Med. Assoc.
,
280
:
910
-912,  
1998
.
2
Ma L., Hoeijmakers J. H. J., van der Eb A. J. Mammalian nucleotide excision repair.
Biochem. Biophys. Acta
,
1242
:
137
-164,  
1995
.
3
Lindahl T., Wood R. D. Quality control by DNA repair.
Science (Washington DC)
,
286
:
1897
-1905,  
1999
.
4
Bohr V. A., Smit C. A., Okumoto D. S., Hanawalt P. C. DNA repair in an active gene: removal of pyrimidine dimers from the DHFR gene of CHO cells is much more efficient than in the genome overall.
Cell
,
40
:
359
-369,  
1985
.
5
Mellon I., Spivak G., Hanawalt P. C. Selective removal of transcription-blocking DNA damage from the transcribed strand of the mammalian DHFR gene.
Cell
,
51
:
241
-249,  
1987
.
6
Bootsma D. The genetic defect in DNA repair deficiency syndromes.
Eur. J. Cancer
,
29A
:
1482
-1488,  
1993
.
7
Venema J., van Hoffen A., Natarajan A. T., van Zeeland A. A., Mullenders L. H. F. The residual repair capacity of xeroderma pigmentosum complementation group C fibroblasts is highly specific for transcriptionally active DNA.
Nucleic Acids Res.
,
18
:
443
-448,  
1990
.
8
Hwang B. J., Ford J. M., Hanawalt P. C., Chu G. Expression of the p48 xeroderma pigmentosum gene is p53-dependent and is involved in global genomic repair.
Proc. Natl. Acad. Sci. USA
,
96
:
424
-428,  
1999
.
9
Troelstra C., van Gool A., de Wit J., Vermeulen W., Bootsma D., Hoeijmakers J. H. J. ERCC6, a member of a subfamily of putative helicases, is involved in Cockayne’s syndrome and preferential repair of active genes.
Cell
,
71
:
939
-953,  
1992
.
10
Henning K. A., Li L., Iyer N., McDaniel L. D., Reagan M. S., Legerski R., Schultz R. A., Stefanini M., Lehmann A. R., Mayne L. V. The Cockayne syndrome group A gene encodes a WD repeat protein that interacts with CSB protein and a subunit of RNA polymerase II TFIIH.
Cell
,
82
:
555
-564,  
1995
.
11
Van der Horst G. T. J., van Steeg H., Berg R. J. W., van Gool A. J., de Wit J., Weeda G., Morreau H., Beems R. B., van Kreijl C. F., de Gruijl F. R., Bootsma D., Hoeijmakers J. H. J. Defective transcription-coupled repair in Cockayne Syndrome B mice is associated with skin cancer susceptibility.
Cell
,
89
:
425
-435,  
1997
.
12
De Vries A., Van Oostrom C. T. M., Hofhuis F. M. A., Dortant P. M., Berg R. J. W., De Gruijl F. R., Wester P. W., Van Kreijl C. F., Capel P. J. A., Van Steeg H., Verbeek S. J. Increased susceptibility to ultraviolet-B and carcinogens of mice lacking the DNA excision repair gene XPA.
Nature (Lond.)
,
377
:
169
-173,  
1995
.
13
Nakane H., Takeuchi S., Yuba S., Saijo M., Nakatsu Y., Murai H., Nakatsuru Y., Ishikawa T., Hirota S., Kitamura Y., Kato Y., Tsunoda Y., Miyauchi H., Horio T., Tokunaga T., Matsunaga T., Nikaido O., Nishimune Y., Okada Y., Tanaka K. High incidence of ultraviolet-B- or chemical carcinogen-induced skin tumours in mice lacking the xeroderma pigmentosum group A gene.
Nature (Lond.)
,
377
:
165
-169,  
1995
.
14
Cheo D. L., Ruven H. J., Meira L. B., Hammer R. E., Burns D. K., Tappe N. J., van Zeeland A. A., Mullenders L. J., Friedberg E. C. Characterization of defective nucleotide excision repair in XPC mutant mice.
Mutat. Res.
,
374
:
1
-9,  
1997
.
15
Sands A. T., Abuin A., Sanchez A., Conti C. J., Bradley A. High susceptibility to ultraviolet induced carcinogenesis in mice lacking XPC.
Nature (Lond.)
,
377
:
162
-165,  
1995
.
16
Cheo D. L., Meira L. B., Hammer R. E., Burns D. K., Doughty A. T. B., Friedberg E. C. Synergistic interactions between XPC and p53 mutations in double-mutant mice: neural tube abnormalities and accelerated UV radiation-induced skin cancer.
Curr. Biol.
,
6
:
1691
-1694,  
1996
.
17
Sontag Y., Garssen J., de Gruijl F. R., van der Leun J. C., van Vloten W. A., van Loveren H. Ultraviolet radiation-induced impairment of the early initiating and the late effector phases of contact hypersensitivity to picrylchloride: regulation by different mechanisms.
J. Investig. Dermatol.
,
102
:
923
-927,  
1994
.
18
Berg R. J. W., de Vries A., van Steeg H., de Gruijl F. R. Relative susceptibilities of XPA knockout mice and their heterozygous and wildtype littermates to UVB-induced skin cancer.
Cancer Res.
,
57
:
581
-584,  
1997
.
19
De Gruijl F. R., van der Meer J. B., van der Leun J. C. Dose-time dependency of tumor formation by chronic UV exposure.
Photochem. Photobiol.
,
37
:
53
-67,  
1983
.
20
Kelfkens G., de Gruijl F. R., van der Leun J. C. Tumourigenesis by short-wave ultraviolet-A: papillomas versus squamous cell carcinomas.
Carcinogenesis (Lond.)
,
12
:
1377
-1382,  
1991
.
21
Berg R. J. W., de Gruijl F. R., van der Leun J. C. Interaction between ultraviolet-A and ultraviolet-B radiations in skin cancer induction in hairless mice.
Cancer Res.
,
53
:
4212
-4217,  
1993
.
22
Kaplan E. L., Meier P. Non-parametric estimation from incomplete observations.
J. Am. Statist. Assoc.
,
53
:
203
-224,  
1958
.
23
Peto, R., Pike, M. C., Day, M. E., Gray, R. G., Lee, P. H., Parish, S., Peto, J., Richards, S., and Wahrendorf, J. Guidelines for simple sensitivity test for carcinogenic effects in long-term animal experiments. In: IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals in Humans, Suppl. 2, pp. 311–426. Lyon, France: IARC, 1980.
24
Van der Leun J. C. Delayed pigmentation and ultraviolet erythema.
Photochem. Photobiol.
,
4
:
459
-464,  
1965
.
25
Andersen P. H., Abrams K., Bjerring P., Maibach H. A time-correlation study of ultraviolet B-induced erythema measured by reflectance spectroscopy and laser Doppler flowmetry.
Photodermatol. Photoimmunol. Photomed.
,
8
:
123
-128,  
1991
.
26
Trevithick J. R., Xiong H., Lee S., Shum D. T., Sanford E., Karlik S. J., Norley C., Dilworth G. R. Topical tocopherol acetate reduces post-UVB, sunburn-associated erythema, edema, and skin sensitivity in hairless mice.
J. Investig. Dermatol.
,
296
:
575
-582,  
1992
.
27
Berg R. J. W., Ruven H. J. T., Sands A. T., de Gruijl F. R., Mullenders L. H. F. Defective global genome repair in XPC mice is associated with skin cancer susceptibility but not with sensitivity to UVB induced erythema and edema.
J. Investig. Dermatol.
,
110
:
405
-409,  
1998
.
28
De Gruijl F. R., Berg R. J. W. In situ molecular dosimetry and tumor risk: UV-induced DNA damage and tumor latency time.
Photochem. Photobiol.
,
68
:
555
-560,  
1998
.
29
Van Gool A. J., van der Horst G. T. J., Citterio E., Hoeijmakers J. H. J. Cockayne syndrome: defective repair of transcription?.
EMBO J.
,
16
:
4155
-4162,  
1997
.
30
Ljungman M., Zhang F. Blockage of RNA polymerase as a possible trigger for U.
V. light induced apoptosis. Oncogene
,
13
:
823
-831,  
1996
.
31
De Vries A., Berg R. J. W., Wijnhoven S., Westerman A., Wester P. W., van Kreijl C. F., Capel P. J. A., de Gruijl F. R., van Kranen H. J., van Steeg H. XPA-deficiency in hairless mice causes a shift in skin tumor types and mutational target genes after exposure to low doses of U.
V.B. Oncogene
,
16
:
2205
-2212,  
1998
.
32
Van Hoffen A., Venema J., Meschini R., van Zeeland A. A., Mullenders L. H. F. Transcription-coupled repair removes both cyclobutane pyrimidine dimers and 6-4 photoproducts with equal efficiency and in a sequential way from transcribed DNA in xeroderma pigmentosum group C fibroblasts.
EMBO J.
,
14
:
360
-367,  
1995
.
33
Ruven H. J. T., Berg R. J. W., Seelen C. M. J., Dekkers J. A. J. M., Lohman P. H. M., Mullenders L. H. F., van Zeeland A. A. Ultraviolet-induced cyclobutane pyrimidine dimers are selectively removed from transcriptionally active genes in the epidermis of the hairless mouse.
Cancer Res.
,
53
:
1642
-1645,  
1993
.
34
Ruven H. J. T., Seelen C. M. J., Lohman P. H. M., van Kranen H. J., van Zeeland A. A., Mullenders L. H. F. Strand-specific removal of cyclobutane pyrimidine dimers from the p53 gene in the epidermis of UVB-irradiated hairless mice.
Oncogene
,
9
:
3427
-3434,  
1994
.
35
Giglia G., Dumaz N., Drougard C., Avril M-F., Daya-Grosjean L., Sarasin A. p53 mutations in skin and internal tumors of xeroderma pigmentosum patients belonging to complementation group C.
Cancer Res.
,
58
:
4402
-4409,  
1998
.
36
Ananthaswamy H. N., Ouhtit A., Evans R. L., Gorny A., Khaskina P., Sands A. T., Conti C. J. Persistence of p53 mutations and resistance of keratinocytes to apoptosis are associated with the increased susceptibility of mice lacking the XPC gene to UV carcinogenesis.
Oncogene
,
18
:
7395
-7398,  
1999
.
37
Naylor M. F. Erythema, skin cancer risk, and sunscreens.
Arch. Dermatol.
,
133
:
373
-375,  
1997
.