This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jans, J.
Right arrow Articles by van der Horst, G. T. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jans, J.
Right arrow Articles by van der Horst, G. T. J.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, November 2006, p. 8515-8526, Vol. 26, No. 22
0270-7306/06/$08.00+0     doi:10.1128/MCB.00807-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Differential Role of Basal Keratinocytes in UV-Induced Immunosuppression and Skin Cancer{triangledown}

Judith Jans ,1,{dagger},{ddagger} George A. Garinis,1,{ddagger} Wouter Schul,1,§ Adri van Oudenaren,2 Michael Moorhouse,3 Marcel Smid,4 Yurda-Gul Sert,1 Albertina van der Velde,1 Yvonne Rijksen,1 Frank R. de Gruijl,5 Peter J. van der Spek,3 Akira Yasui,6 Jan H. J. Hoeijmakers,1 Pieter J. M. Leenen,2 and Gijsbertus T. J. van der Horst1*

MGC, Department of Cell Biology and Genetics, Center for Biomedical Genetics, Erasmus University Medical Center, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands,1 Department of Immunology, Erasmus University Medical Center, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands,2 Department of Bioinformatics, Erasmus University Medical Center, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands,3 Department of Medical Oncology, Josephine Nefkens Institute, Erasmus University Medical Center, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands,4 Department of Dermatology, Leiden University Medical Center, Sylvius Laboratory, 2300 RA Leiden, The Netherlands,5 Department of Molecular Genetics, Institute of Development, Aging and Cancer, Tohoku University, Sendai, 980-8575, Japan6

Received 8 May 2006/ Returned for modification 12 June 2006/ Accepted 30 August 2006


arrow
ABSTRACT
 
Cyclobutane pyrimidine dimers (CPDs) and 6-4 photoproducts (6-4PPs) comprise major UV-induced photolesions. If left unrepaired, these lesions can induce mutations and skin cancer, which is facilitated by UV-induced immunosuppression. Yet the contribution of lesion and cell type specificity to the harmful biological effects of UV exposure remains currently unclear. Using a series of photolyase-transgenic mice to ubiquitously remove either CPDs or 6-4PPs from all cells in the mouse skin or selectively from basal keratinocytes, we show that the majority of UV-induced acute effects to require the presence of CPDs in basal keratinocytes in the mouse skin. At the fundamental level of gene expression, CPDs induce the expression of genes associated with repair and recombinational processing of DNA damage, as well as apoptosis and a response to stress. At the organismal level, photolyase-mediated removal of CPDs, but not 6-4PPs, from the genome of only basal keratinocytes substantially diminishes the incidence of skin tumors; however, it does not affect the UVB-mediated immunosuppression. Taken together, these findings reveal a differential role of basal keratinocytes in these processes, providing novel insights into the skin's acute and chronic responses to UV in a lesion- and cell-type-specific manner.


arrow
INTRODUCTION
 
Exposure to UV light has undesired health consequences with increasing impact; apart from acute effects (e.g., sunburn), skin tumors are considered a major threat, demonstrated by their increasing incidence in white populations, due to altered life style and the erosion of the protecting ozone layer (1, 43). Besides the ability of RNA (20, 21) and proteins (5) to absorb light at the UV wavelength, our recent findings have unequivocally pointed to DNA as the biologically most relevant target of UV radiation (12, 22, 35). UV induces the formation of major dimer configurations, namely the cyclobutane pyrimidine dimers (CPDs) and 6-4 photoproducts (6-4PPs), generated by covalent bonds between two adjacent pyrimidines, that interfere with biological processes (e.g., transcription and replication) critical for cell viability (28).

To recognize and remove effectively the wide range of hazardous DNA lesions, mammalian cells employ a resourceful battery of DNA repair systems (10, 11, 15). For instance, nucleotide excision repair (NER) removes UV-induced DNA damage as well as numerous other helix-distorting lesions (15). NER is divided in two subpathways: global genome NER (GG-NER) and transcription-coupled NER (TC-NER) that differ primarily in the way damage is recognized (2, 27, 42). In GG-NER, the protein complex XPC/hHR23B probes the complete genome for deformation of the DNA double helix (39). However, whereas this complex readily recognizes the highly helix-distorting 6-4PPs, thereby allowing their fast removal, it poorly recognizes and removes the mildly distorting CPDs (2, 28). In humans, CPD recognition is enhanced in potentially genotoxic conditions due to p53-dependent upregulation of the p48 subunit of the DNA damage binding protein DDB (18). Rodents, however, lack the p53-responsive element in the p48 gene and are, therefore, unable to repair CPDs by GG-NER (18, 19). In TC-NER, damage recognition is initiated when an elongating RNA polymerase II is stalled upon transcription-blocking lesions (e.g., CPDs and 6-4PPs) on the template strand of active genes. This initial step in TC-NER requires the action of CSB and CSA proteins. Subsequently, the XPB and XPD helicases of the 10-subunit transcription factor TFIIH unwind the helix surrounding the lesion. XPA verifies the damage whereas RPA stabilizes the complex. Next, a single-strand DNA fragment of ~30 nucleotides long flanking the damage is excised by XPG and XPF/ERCC1 endonucleases. Finally, DNA synthesis of the excised strand resumes to fill the gap followed by ligation of the nick (8, 10, 15, 44).

The indispensable task of NER in removing (UV-induced) DNA lesions is highlighted by three clinically and genetically heterogeneous human syndromes that carry defects in NER-associated genes and that are all sensitive to UV exposure: xeroderma pigmentosum (XP), Cockayne syndrome (CS), and trichothiodystrophy (7). Particularly, xeroderma pigmentosum patients are characterized by a >1,000-fold increased susceptibility to sunlight-induced skin cancer (4).

Even so, many organisms are equipped with yet another mechanism for repair of UV-induced DNA lesions, named photoreactivation (PR) (46). Unlike NER, PR is carried out by photolyases, monomeric enzymes that specifically recognize and repair either CPDs or 6-4PPs by damage reversal, thereby obviating the need for excision and DNA resynthesis. Photolyases require visible light as a source of energy to split the pyrimidine dimer and revert distorted base conformations back to their original state. However, despite their strong evolutionary conservation in many organisms (ranging from bacteria to marsupials), photolyase enzymes are absent in placental mammals. Therefore, rodents and humans rely solely on the complex and, for CPDs, the less efficient NER system (46).

To unravel the contribution of the individual classes of photolesions (i.e., CPDs versus 6-4PPs) to the deleterious outcome of UV exposure in the skin, we have previously generated transgenic mice that ubiquitously express either the Potorous tridactylus CPD photolyase (ß-act-CPD-PL) or the Arabidopsis thaliana 6-4PP photolyase (ß-act-6-4PP-PL) from the ß-actin promoter (35). Light-dependent removal of CPDs and/or 6-4PPs from the skin provided ample evidence that CPDs, rather than 6-4PPs, are mostly responsible for the majority of the adverse effects upon UV exposure, including sunburn, mutagenesis, and skin cancer (21, 33). However, the ß-act-CPD-PL and ß-act-6-4PP-PL mouse models provided little information regarding the contribution of distinct cell types to the biological processes associated UV exposure.

In this regard, the basal keratinocyte is particularly prone to oncogenic transformation, as it proliferates vigorously while it is burdened with the task to respond continuously to environmental triggers. Thus, we generated mice expressing the P. tridactylus CPD photolyase enzyme from the basal keratinocyte-specific promoter keratin-14 (K14) (33). The K14-CPD-photolyase (K14-CPD-PL) transgenic mice allowed rapid, light-dependent removal of CPDs from basal keratinocytes only, whereas DNA lesions in all other cell types (e.g., fibroblasts and more differentiated keratinocytes) could only be removed by the substantially slower NER. Besides the UV-mediated mutagenic effects, the suppression of the immune system subsequent to UV exposure contributes substantially to the development of skin cancer (41). Here, the K14-6-4PP-PL mice complete the set of available tools, allowing us to address lesion and cell type specificity in relation to systemic immunosuppression and skin cancer upon exposure to UVB irradiation.


arrow
MATERIALS AND METHODS
 
Generation of K14-photolyase transgenic mice. The construct for the generation of K14-6-4PP-PL transgenic mice was cloned in the vector pSP72 (Promega) and contained the human keratin-14 promoter (2.3-kb PCR fragment, generated using forward primer 5'-AAGCTTATATTCCATGCTAG-3' and reverse primer 5'-GGATCCTGAGTGAAGAGAAGG-3') followed by the A. thaliana 6-4PP-PL cDNA. At the 3' end, exon 2 (the last 22 bp), intron 2, exon 3, and the 3' untranslated region (including the polyadenylation signal) of the human ß-globin gene were inserted. The expression constructs were excised from the plasmid using SalI, separated from vector DNA by agarose gel electrophoresis, isolated from the gel with a GeneClean II kit (Bio 101), and further purified using Elutip-D minicolumns (Schleicher and Schuell, Germany). The fragment was dissolved in injection buffer (10 mM Tris-HCl, pH 7.5, 0.08 mM EDTA) and injected into the pronucleus of fertilized eggs derived from FVB/N intercrosses as described previously (16).

Transgenic animals were identified by Southern blot analysis of genomic tail DNA, using the 6-4PP-PL cDNA as a probe. To estimate the number of integrated copies, equal amounts of genomic DNA from transgenic mice were subjected to Southern blot analysis. As a standard, we used equal amounts of genomic tail DNA supplemented with 0, 10, 30, or 100 pg of the corresponding 6-4PP-PL expression construct. The hybridization signal obtained with the 6-4PP-PL cDNA probe was quantified using a Molecular Dynamics PhosphorImager and ImageQuant software. After comparison of signal intensities, the transgene copy number was estimated using the supplemented reference samples. Routine genotyping of mice was performed by PCR analysis. The primer set 5'-GCACGATTCAGCAAGCAAGG-3' (forward primer) and 5'-CGGTACCTCTACCTATTTGAGTTTTG-3' (reverse primer) was used to amplify a 200-bp fragment of the 6-4PP-PL coding region. Experiments were performed on mice in a mixed FVB/C57BL/6J background, except for the carcinogenesis and microarray experiments, in which animals were further crossed with hairless HRA/SKH mice. The generation of ß-act-CPD-PL and ß-act-6-4PP-PL transgenic mice, as well as genotyping procedures, has been described elsewhere (22, 35).

As required by Dutch law, the Dutch Ministry of Agriculture, Nature and Food Quality approved the generation of genetically modified mice. An independent Animal Ethical Committee (Dutch equivalent of the IACUC) approved all animal studies.

RNA isolation and reverse transcriptase PCR (RT-PCR). Mouse skin RNA was isolated, and cDNA synthesis was performed using Superscript II RNase H reverse transcriptase (Life Technologies) according to the protocol of the supplier. A PCR was performed on the cDNA using a forward primer in the photolyase transgene (5'-GCACGATTCAGCAAGCAAGG-3') and a reverse primer in exon 3 of the ß-globin gene (5'-TGGACAGCAAGAAAGCGAG-3'). The presence of introns in the ß-globin moiety of the photolyase transgenes allows discrimination between the cDNA-derived PCR product and possible genomic DNA contamination.

Photoreactivation in mouse cells and skin. Cells were grown on coverslips and washed with phosphate-buffered saline, exposed to 20 J/m2 UV-C (Philips TUV germicidal lamp), and subsequently kept in Hank's buffer (137 mM NaCl, 5.4 mM KCl, 4.4 mM KH2PO4, 0.33 mM Na2HPO4, 1.3 mM CaCl2, 0.81 mM MgSO4, 4.2 mM NaHCO3, 1 g/liter glucose, pH 7.4). Photoreactivation was performed by exposing cells for 1 h to light from four white fluorescent tubes (Philips TLD 18W/54) at a distance of 15 cm and shielded by a 5-mm glass filter. Nonphotoreactivated cells were given the same treatment except that dishes were covered with two layers of aluminum foil and put under the same fluorescent lamps. Immunocytochemical staining of CPDs and 6-4PPs using the antibodies TDM2 and 64M2, respectively, was performed as described previously (35).

Mice were anesthetized, and hairs were removed from a small area on the back of the animal. One third of the hairless area was covered with black nonadhesive tape, and the remaining area was exposed to the light of two Philips TL-12 (40 W) tubes emitting UVB light. Typically, 1 minimal erythemal dose (MED) was obtained with an exposure of 2 min. Subsequently, half of the UV-exposed area was covered with tape, and mice were exposed for 3 h to the light of 4 white fluorescent tubes (GE Lightning Polylux XL F36W/840) filtered through 5 mm of glass. Then, mice were sacrificed, and skin samples were taken from the unexposed area, the UV-irradiated area that was covered, and the UV-irradiated area that was exposed to PR light. Skin sections were stained as described previously with antibodies TDM2 or 64M2 recognizing CPDs or 6-4PPs, respectively (35).

Microarray experimental design, total RNA isolation, cDNA labeling, hybridization, and data extraction. Two groups of hairless mice (three mice per group) were irradiated with 1 MED of UVB and PR for 3 h (group 1) or kept in the dark (group 2). Skin samples were isolated at 8 h after UV irradiation as described above. Total RNA was isolated as previously described (12), and RNA concentration and quality were assessed by spectrophotometry and by the use of an Agilent 2100 Bioanalyzer. Differentially labeled cDNA was prepared from RNA from either the UV-irradiated, PR-light-treated skin or the UV-irradiated, non-light-treated skin; it was mixed (per time point) and hybridized to cDNA microarrays representing ~15,000 mouse genes (obtained from The Netherlands Cancer Institute). Labeling and hybridization protocols were performed as previously described. Dye incorporation bias was avoided by reverse labeling all samples during the total RNA reverse transcription. A pair of self-hybridizations (dye reversed) preceded each batch of six microarrays. cDNA microarrays were scanned using a laser confocal scanner (Scanarray Express HT; Perkin Elmer Inc.). Data were extracted by means of the Imagene software package version 5.0 (Biodiscovery Inc., California). Detailed information on labeling and hybridization protocols and the microarray platform can be found at http://microarrays.nki.nl/download/index.html.

Data processing and analysis. Hierarchical clustering, principal component analysis, self-organizing maps, K-clustering, analysis of variance, gene ontology classification, and network analysis were performed by the Spotfire Decision Site software package 7.2, version 10.0 (Spotfire Inc., Massachusetts), the Gene Ontology (GO) Mapper application (38), and Ingenuity Pathways Analysis software (www.ingenuity.com) as previously described (12). Significant overrepresentation of GO-classified biological processes was assessed by comparing the number of pertinent genes in a given biological process to the total number of the relevant genes printed on the NIA 15K cDNA microarray for that biological process (Fisher exact test, P ≤ 0.05; false detection rate of ≤0.1) using the publicly accessible software Ease (17).

Quantitative real-time PCR evaluation. Quantitative real-time PCR was performed with the DNA engine Opticon according to the instructions of the manufacturer (MJ Research). For quantitation of cDNA, primer pairs for Rad51 (forward, 5'-TAC ATT GAC ACC GAG GGC AC-3'; reverse, 5'-CTG ACG CTT GGT AAA GGA GC-3'), Mcl-1 (forward, 5'-GAT GGC GTA ACA AAC TGG G-3'; reverse, 5'-GGA AGA ACT CCA CAA ACC C-3'), Sumo (forward, 5'-GCC AGT GAT GTG AAG AGA CC-3'; reverse, 5'-GGT GGG TTC TGA AAG TGG AG-3'), Hspcb (forward, 5'-AGA GCC TCA CCA ATG ACT GG-3'; reverse, 5'-ATG ATG AAC ACA CGG CGG A-3'), Maged (forward, 5'-AGA ATG CCA CCA CAA AGG G-3'; reverse, 5'-ACT GGG AGA CTG AGG GAA AT-3'), Trp53 (forward, 5'-AAC TAT GGC TTC CAC CTG G-3'; reverse, 5'-GCT GTG ACT TCT TGT AGA TGG-3'), Gpx3 (forward, 5'-TCT ACG AGT ATG GAG CCC TCA-3'; reverse, 5'-GCC CAG AAT GAC CAA GCC AA-3'), Ddb1 (forward, 5'-GCC AGT CAA AGA GGT GGG AA-3'; reverse, 5'-AAT GAT GCC AGT CTC CGA GG-3'), Sod1 (forward, 5'-GGG ACA ATA CAC AAG GCT GT-3'; reverse, 5'-GCC AAT GAT GGA ATG CTC TC-3'), and Hprt-1 (forward, 5'-GGC AAC ATC AAC AGG ACT CC-3'; reverse, 5'-CGA AGT GTT GGA TAC AGG CC-3') were designed to generate intron-spanning products of 180 to 210 bp. Hypoxanthine guanine phosphoribosyltransferase 1 (Hprt-1) mRNA was used as an external standard. For data analysis, the second derivative maximum method was applied: E1gene of interest{Delta}CP (control – sample)/Ehprt-1{Delta}CP (control – sample), where E is efficiency.

Apoptosis. For detection of apoptotic cells in the skin, we used a terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) assay (Fluorescein Apoptosis Detection System; Promega). Depilated areas on the back of mice were exposed to UV and PR light as described above and subsequently kept in the dark. Skin samples, taken 40 h after UV exposure, were fixed overnight in 4% paraformaldehyde, washed in phosphate-buffered saline, and embedded in paraffin. Skin sections (5 µm) were deparafinized and incubated as described by the manufacturer.

Hyperplasia. Mice were anesthetized and an area on the back was depilated by plucking. Mice were exposed to 1 MED UVB and PR light for four consecutive days. One week after the start of the experiment, mice were sacrificed, and 8-µm skin sections were obtained. Sections were further processed and stained with hematoxylin and eosin.

Immunosuppression. Systemic immunosuppression was determined as before (3) with slight modifications. The animals were shaven on the back 1 day prior to UV treatment. For irradiation, a Bluepoint 2 source (Hönle, München, Germany) was used. Two circular areas (6.3 cm2 in total) were irradiated with 0.5 MED (2,000 J/m2) each day for five consecutive days. The mice were skin sensitized 4 days after the last UV exposure by topical application of 5% picryl chloride (PCl) to nonirradiated shaved abdomen, chest, and feet. Four days after sensitization, both ears of mice were challenged with 0.8% PCl in olive oil. At 24 h after challenge, duplicate ear measurements were performed with an engineer's micrometer (Mitutoyo Digimatic 293561; Veenendaal, The Netherlands). The responses were statistically evaluated using a two-tailed Student's t test.

Skin carcinogenesis. Hairless photolyase mice and their wild-type (wt) littermates aged 8 to 12 weeks were exposed daily to 500 J/m2 UVB (Philips TL-12 tubes) followed by 3 h of PR light (GE Lightning Polylux XL F36W/840 lamps). Mice were followed in time and thoroughly screened weekly for the occurrence of skin abnormalities. Typically, carcinomas in wt animals were expected to occur after 3 months of treatment. Mice were sacrificed when tumors of >4 mm occurred. Biopsies of tumors were taken and processed for routine hematoxylin-eosin staining. Graphical representation of the prevalence versus time is based on an actuarial method described by Kaplan and Meier (reviewed in reference 45) and adapted to carcinogenesis by Peto et al. (33).


arrow
RESULTS
 
Generation of keratin-14-photolyase transgenic mouse lines. The generation of the K14-CPD-photolyase mouse lines was described previously (35). To obtain mice expressing the 6-4PP-PL transgene in basal keratinocytes, a construct was generated containing the A. thaliana 6-4PP-PL cDNA, preceded by the human K14 promoter. To enhance mRNA stability, a part of the human ß-globin gene (i.e., exons 2 and 3, intron 2, the 3' untranslated region, and the polyadenylation signal) was cloned behind the photolyase cDNA (Fig. 1A). The photolyase encoded by the A. thaliana 6-4PP-PL cDNA specifically repairs 6-4PPs, leaving the CPDs unrepaired, as shown by PR experiments in UV-exposed dermal fibroblasts from ß-actin-6-4PP-PL mice (Fig. 1B).


Figure 1
View larger version (42K):
[in this window]
[in a new window]
  		FIG. 1. Generation of K14-6-4PP photolyase transgenic mice. (A) Expression construct for the generation of K14-6-4PP-photolyase transgenic mice, containing the human K14 promoter, the A. thaliana 6-4PP-photolyase cDNA, and human genomic ß-globin sequences, including exons 2 and 3, intron 2, the 3' untranslated region, and the polyadenylation signal. Arrows indicate the position of the primers used for the RT-PCR experiment. (B) Photoreactivation of 6-4PPs in cultured transgenic fibroblasts. Induction of CPD and 6-4PP lesions in cultured MDFs from 6-4PP-PL transgenic mice by 20 J/m2 of UVC light and subsequent exposure to photoreactivating light for 1 h. Photolesions were detected by immunofluorescent labeling using CPD- or 6-4PP-specific antibodies and fluorescein isothiocyanate-conjugated secondary antibodies. (C) RT-PCR analysis of RNA from skin extracts of K14-6-4PP photolyase transgenic mice results in a 300-bp band. (D) 6-4PP lesions in the depilated dorsal skin of K14-6-4PP photolyase mice following exposure to 1 MED of UVB light and without (middle panel) or with (bottom panel) subsequent exposure to PR light. Photolesions were detected using 6-4PP-specific antibodies and horseradish peroxidase-conjugated secondary antibodies. Diaminobenzidine was used as substrate. Nuclei are visualized by methyl green staining. Note the uniform high density of nuclear labeling in the non-PR tissue, in contrast to the heterogeneous and lighter nuclear labeling of the PR tissue.

Oocyte injections resulted in several independent K14-6-4PP-PL mouse lines. The selected mouse line contained ~25 copies of the transgene, as determined by Southern blot analysis (data not shown). RT-PCR analysis of total skin RNA showed the presence of a 300-bp fragment, indicative of proper expression and splicing of the transgene (Fig. 1C).

Photoreactivation of the mouse skin. The transgenically expressed photolyase enzyme is expected to allow light-dependent removal of DNA lesions in a subset of epidermal keratinocytes only. Previously, we showed that PR in K14-CPD-PL mice indeed resulted in CPD removal in basal keratinocytes specifically. To investigate light-dependent removal of 6-4PPs, we applied an immunohistochemical assay on K14-6-4PP-PL mice, using an antibody specifically recognizing 6-4PPs (64M2). One-third of a depilated area on the back of mice was covered, while the remaining part was exposed to 1 MED of UVB. Next, half of the UV-exposed area was covered, while the remaining part of the skin was exposed to PR light for 3 h. Skin biopsies were taken, and sections were further processed for immunohistochemical staining. As expected, the non-UV-irradiated skin did not show any DNA lesions (Fig. 1D, No UV). Similar to our previous experiments, upon UV exposure, both CPDs (see reference 35) and 6-4PPs (this work) are induced in epidermis and upper dermis (Fig. 1D, UV 3 h dark). Importantly, upon exposure to PR light for 3 h, 6-4PPs are rapidly removed from a subset of cells in the epidermis only, indicating that the enzyme is indeed functional in basal keratinocytes (Fig. 1D, 3 h PR light). Together, these findings validate the K14-6-4PP-PL mice for the investigation of the role of basal keratinocytes in the response of the skin to UV.


Figure 3
View larger version (54K):
[in this window]
[in a new window]
  		FIG. 3. Transcriptome analysis in K14-CPD-PL mice. (A) Tree graph representation of the similarity between significant gene expression profiles (analysis of variance, P ≤ 0.05; ≥1.5-fold change) of irradiated non-PR (M1 to M3) and PR (M4 to M6) skin samples compared to non-UV-irradiated control skin samples (C1 and C2). Note the clustering of all samples into two main groups correlating to exposure to PR (light) or not (dark) subsequent to UV irradiation. (B) Principle component analysis of irradiated PR and non-PR skin samples. Each sphere, colored according to PR status, represents a skin tissue sample that is positioned in a reconstructed three-dimensional gene space so that proximity between spheres represents similarity between gene expression profiles of the corresponding matrix points. Note that mice are segregated according to the UV irradiation treatment and PR status. (C) Heat map representation of PR-dependent, significant gene expression profiles of genes involved in stress-related responses including DNA damage repair, apoptosis, and oxidative stress. Changes in relative expression are represented by red (upregulated) and green (downregulated); black indicates no change compared to nonirradiated controls. (D) Verification of microarray data by quantitative real-time PCR. Relative changes in expression (n-fold) indicate the relative average expression levels of indicated genes in irradiated K14-CPD-PL mouse skin (n = 3) treated with PR (or not) compared to those from nonirradiated, non-PR-treated k14-CPD-PL mouse skin samples.

Acute effects upon exposure to UVB-light. (i) Apoptosis of basal keratinocytes requires the presence of CPDs. Exposure of the skin to UV light leads to apoptosis of keratinocytes and, upon chronic treatment, to epidermal hyperplasia, detected as thickening of the epidermis. Recently, we have shown that removal of CPDs, rather than 6-4PPs, from the entire skin is sufficient to prevent initiation of the apoptotic response (22), an effect that was also observed when CPDs were removed from basal keratinocytes only (35). We now also investigated the effect of enhanced removal of 6-4PPs in basal keratinocytes on apoptotic events; K14-6-4PP-PL mice and K14-CPD-PL mice (used as a control) were exposed to a single dose of UVB (1 MED), followed by PR light for 3 h, and compared to UV-irradiated K14-photolyase mice that were kept in the dark. Forty hours after UVB-exposure, skin biopsies were taken, and skin sections were further processed for the TUNEL assay, thereby allowing the detection of apoptotic cells. As shown in Fig. 2A, preferential elimination of 6-4PPs from basal keratinocytes did not have a significant impact on the apoptotic response (Fig. 2A). This is fully consistent with previous observations where the levels of apoptosis were not significantly altered upon PR of 6-4PPs in the ubiquitously expressing ß-act-6-4PP-PL (22).


Figure 2
View larger version (57K):
[in this window]
[in a new window]
  		FIG. 2. Effect of photoreactivation of photoproducts from basal keratinocytes on UVB-induced acute responses. (A) Apoptotic response in photolyase mice. Photolyase mice were exposed to 1 MED UVB, followed by exposure to PR light for 3 h or darkness. Apoptosis was measured 40 h after UV exposure by a TUNEL assay. (B) Hyperplasia in photolyase mice. Mice were exposed to 1 MED UVB for four subsequent days, followed by exposure to PR light (3 h) or darkness. Four days after the last exposure, mice were sacrificed, and skin sections were stained with hematoxylin and eosin.

(ii) CPDs in basal keratinocytes are responsible for UV-induced hyperplasia. To study the effect of distinct DNA lesions in basal keratinocytes on epidermal hyperplasia, K14-CPD-PL and K14-6-4PP-PL mice were exposed to UVB (1 MED) for four consecutive days. Every UV treatment was followed by exposure to PR light for 3 h, while UV-irradiated, non-PR light-treated transgenic mice were kept in the dark. Three days after the last UV exposure, mice were sacrificed, and skin samples were stained with hematoxylin and eosin. As expected, a clear induction of epidermal hyperplasia was observed in mice that were only exposed to UVB (Fig. 2B). However, after exposure of K14-CPD photolyase mice to PR light, i.e., upon removal of CPDs from basal keratinocytes only, only a minor thickening of the epidermis could be observed, indicating that epidermal hyperplasia was nearly absent. In marked contrast, light-dependent enhanced removal of 6-4PPs from the basal keratinocytes of K14-6-4PP-PL mice did not significantly alter the skin hyperplasia observed in mouse skin after UVB. These findings identify the basal keratinocyte as the major cell type responsible for UV-induced hyperplasia, and, in line with previous studies in ß-act photolyase mouse models (21), point towards CPDs as the critical lesion involved in this process.

(iii) Photoreactivation of CPDs in basal keratinocytes profoundly alters the transcriptional response to UV in whole skin. The above findings stress the importance of CPD lesions, rather than 6-4PPs, for a wide range of biological processes. Previously, we showed that genome-wide changes in transcription required the continuous presence of UV-induced CPD photolesions in in vitro cultured transgenic CPD-PL mouse dermal fibroblasts (MDFs) in a time-, dose-, and light-dependent manner (12). However, cultured cells are continuously under physiological stress, as they are in a state of continuous proliferation and have adapted to culture conditions. To assess the physiological impact of persisting CPD photolesions in basal keratinocytes on the transcriptional response, we used a functional genomics approach in K14-CPD-PL mice following exposure of mouse skin to irradiation. Total RNA was isolated from whole skin samples of UVB-irradiated (1 MED) K14-CPD-PL hairless mice (n = 3) either photoreactivated or not. Animals were sacrificed 8 h post-UV exposure, a relevant time point that allows the expression differences between the photoreactivated and nonphotoreactivated mice to evolve (12). In a dye-swap approach, differentially labeled cDNA prepared from RNA from either the UV-irradiated, PR light-treated skin or the UV-irradiated, non-light-treated skin was mixed and hybridized to cDNA microarrays representing ~15,000 mouse genes.

An unsupervised hierarchical clustering of the genes that varied significantly (P ≤ 0.05 and >1.5-fold change) between the UV-irradiated, PR- and non-PR-treated skin samples demonstrated the profound effect of PR (and by inference, of removal of CPDs from basal keratinocytes) on the transcriptional response to UVB in the skin. Of note, all samples clustered primarily into two main groups. This clustering was determined predominantly by PR status, suggesting that PR of CPDs in basal keratinocytes comprises a dominant determinant of gene expression in the total skin (Fig. 3A).

To confirm the results of the hierarchical clustering, as well as to reduce the dimensionality of the data set and improve the visualization of meaningful variables between samples, we employed a principle component analysis, as previously described (12). Herein, each sphere represents one sample that is positioned in a reconstructed three-dimensional "gene space" so that proximity between spheres demonstrates the degree of similarity of expression profiles between samples (Fig. 3B). In agreement with the previously demonstrated tree graph, all spheres representing the PR, non-PR, and untreated control samples segregated into distinct groups on the basis of their PR status, thus establishing the prime significance of PR and our ability to measure it on the transcriptional level using an in vivo experimental model system.

(iv) The presence of a single type of UV-induced photolesions (i.e., CPDs) in basal keratinocytes impinges on a wide range of biological processes in vivo. To avoid data preselection and potential introduction of bias, we employed an unbiased approach to unveil the significantly overrepresented biological processes and underlying networks employed by basal keratinocytes to sustain key cellular functions upon UVB irradiation and subsequent PR (or not). For this, all significant genes responsible for clustering the PR and non-PR samples into separate groups were assembled according to the GO classification system and integrated in the complete tree of "Physiological Processes (GO:0007582)" using a previously developed three-dimensional interactive visualization system (http://www2.eur.nl/fgg/ch1/k14network/). Each illustrated GO term was then scored for its relative overrepresentation by examining separately the number of up- and downregulated genes with respect to the total population of genes printed on our cDNA platform for that GO term. Further insight was gained by employing the commercially available Ingenuity database (www.ingenuity.com) to identify all the significantly transcribed genes whose gene products participated in overlapping and distinct networks. These networks were then examined for their statistical significance and explored further by listing the most significantly relevant functions associated with combinations of genes that participated in each of these networks (see "network analysis" in interactive visualizations at http://www2.eur.nl/fgg/ch1/k14network/).

With exposure to 1 MED of UVB in the absence of PR, the response to oxidative stress and DNA damage along with the protein and DNA metabolism and biosynthesis was prominently present among the cellular pathways that were significantly overrepresented (Fig. 3C) (a detailed list is available at http://www2.eur.nl/fgg/ch1/k14network/). Subsequent analysis revealed a number of gene networks involved in numerous vital cellular processes, ranging from cell growth and maintenance to immune responses. This suggests that the presence of CPDs (as the most critical type of UV-induced photolesion) in a subset of keratinocytes is sufficient to trigger a broad spectrum of transcriptional responses underlying a plethora of biological processes in the mouse skin.

In keratinocytes, various genotoxic stresses that damage the DNA have previously been shown to result in the activation of genes involved in cell cycle checkpoints, leading to diverse cellular responses including cell cycle arrest, repair of DNA damage, and programmed cell death (25). Consistent with the time frame (8 h post-UV) and dose employed in this study, a number of genes associated with DNA repair (e.g., Rad51{uparrow}, Ubl1{downarrow}, Ddb1{uparrow}, and Fen1{uparrow}, where{uparrow} and {downarrow} indicate up- and dowregulation, respectively) were significantly modulated upon the presence of CPDs in basal keratinocytes, suggesting the essential role of these cells in provoking the DNA damage response in the UV-irradiated mouse skin. Interestingly, despite the fact that the majority of nondividing cells are expected to mask this response in the mouse skin compared to the cultured cells, the transcriptional regulation of the human recA homologue Rad51 is in agreement with our previous results suggesting that the presence of unrepaired CPDs during DNA replication may induce the formation of DNA breaks and/or homologous recombination (12).

Coupled to this response, a number of genes involved in histone metabolism and chromatin modification were significantly upregulated in the absence of PR (e.g., Hmgn1{uparrow}, Hdac2{uparrow}, H1f0{uparrow}, Hira{uparrow}, and Smarcb1{uparrow}), supporting previous findings that showed human keratinocytes to induce the expression of genes coding for specific histones and histone modification proteins upon UVB irradiation (25). Interestingly, the latter could also be associated with an enhanced proliferative response for a subset of epidermal cells that could eventually result in the epidermal hyperplasia, known to occur at later stages upon UVB irradiation.

Studies in cultured human keratinocytes in vitro (26, 34) and in the hairless murine skin in vivo (32, 37) have provided compelling evidence that exposure of the skin to UV irradiation stimulates the skin's enzymatic antioxidant defense system. In agreement, the presence of CPD lesions provoked the transcriptional upregulation of two major enzymatic antioxidants (e.g., Sod1{uparrow} and Gpx3{uparrow}) known to confer cellular resistance to oxidative stress (34, 36), suggesting a tissue-specific antioxidant response to UVB-mediated phototoxicity mediated by CPD lesions. The PR-dependent nature of this response indicates that the presence of CPD lesions in basal keratinocytes is sufficient for activation of this defense system. Of note, the significant overrepresentation of the response to oxidative stress was only part of the identified, broader response to stress, including genes associated with the response to heat (e.g., Hspa8{uparrow} and Hspcb{uparrow}), inflammation, and stress-induced signaling (e.g., Nfatc4{uparrow}, Csf3r{downarrow}, and Map2k3{downarrow}).

Even so, irreparable DNA damage, or insufficient repair in time, is expected to trigger the death of cells carrying excess DNA damage or in specific stages of differentiation through the activation of apoptotic responses. In this study, several proapoptotic genes exhibited significant changes in expression at 8 h post-UV treatment, though not unidirectionally, presumably due to the heterogeneity in the response between cells (e.g., Elmo2{downarrow}, Ddx4L{downarrow}, Pdcd5{uparrow}, Pdcd8{uparrow}, Pdcd11{downarrow}, Casp6{downarrow}, Trim35{uparrow}, Cfdp1{uparrow}, App1{uparrow}, Mcl1{uparrow}, and Tpt1{uparrow}). Since increased apoptosis was demonstrated in basal keratinocytes 40 h after the exposure of the skin to UVB irradiation (Fig. 2), these data suggest the early presence of both pro- and antiapoptotic responses, whose final outcome may be largely dependent on the extent of DNA damage, the efficiency of DNA repair itself, the stage of differentiation, and cell cycle phase.

Finally, in order for the cell to respond rapidly to growth conditions and insults induced by UVB irradiation, it needs to cover its metabolic needs through the tight regulation of several biosynthetic pathways. In agreement with this, the distinguishable overrepresentation of metabolic genes (evidenced at various GO levels in this data set) was for the most part attributable to the upregulation of several ribosomal protein genes (e.g., L3{uparrow}, L4{uparrow}, L5{uparrow}, L6{uparrow}, L10{uparrow}, S3a{uparrow}, S4{uparrow}, S5{uparrow}, S7{uparrow}, and S8{uparrow}) along with genes involved in translational elongation (e.g., Eef1g{uparrow}, Eef2{uparrow}, Eef1a1{uparrow}, and Elf2s3y{uparrow}), highlighting the role of ribosomal biogenesis and protein biosynthesis in the economy of the cell.

Examining the mRNA levels of several genes associated with the DNA damage response and other biological processes (Mcl1, Rad51, Sumo, Hpsb, Maged, Trp53, Gpx3, Ddb1, and Sod1) in the skin of UV-irradiated, PR-treated (or not) K14-CPD-PL mice validated the accuracy of the microarray data (Fig. 3D).

Chronic effects upon exposure to UVB-light. (i)Removal of CPDs from basal keratinocytes results in substantial protection from skin carcinomas. CPD lesions are an important trigger for UV-induced skin cancer, as mice ubiquitously expressing a CPD-PL transgene are protected substantially from the formation of these tumors, in strong contrast to mice ubiquitously expressing a 6-4PP-PL transgene (22). Whereas basal keratinocytes are considered a crucial cell type involved in the formation of UV-induced skin tumors, it is still unknown whether systemic effects may also play a significant role in skin neoplasia. To address directly whether distinct types of DNA lesions in keratinocytes specifically play an important role in skin carcinogenesis, we subjected hairless K14-CPD-PL and K14-6-4PP-PL animals to a single daily dose of UVB (1 MED), immediately followed by exposure to PR light for 3 h, to specifically remove either CPD or 6-4PPs from the basal keratinocytes, using hairless wt mice as a control. The fraction of tumor-free wt control animals declined shortly after the initiation of UV treatment (Fig. 4A). Correspondingly, the average number of tumors (mostly squamous cell carcinomas and occasionally papillomas) per mouse increased considerably (Fig. 4B). Preferential removal of 6-4PPs from basal keratinocytes did not significantly reduce cancer incidence, while removal of CPDs or both CPDs and 6-4PPs from basal keratinocytes yielded a very strong cancer protection (Kaplan-Meier analysis, P < 0.01). Whereas by week 16 all 11 wt animals had one or more tumors, by that time only 1/21 K14-CPD-PL or K14-CPD/6-4PP-PL transgenic mice had acquired the first tumor. From these data we conclude that apparently rapid removal of CPDs, rather than 6-4PPs from basal keratinocytes only, is sufficient to protect against carcinogenic events, thus providing compelling evidence for both the relevance of basal keratinocytes as well as of UV-induced CPD lesions in skin carcinogenesis.


Figure 4
View larger version (21K):
[in this window]
[in a new window]
  		FIG. 4. Effect of photoreactivation of photoproducts from basal keratinocytes on skin carcinomas. Photolyase mice (for K14-CPD-PL, n = 10; K14-6-4PP-PL, n = 9; K14-CPD/6-4PP-PL, n = 11) and their wt littermates (n = 11) received daily UVB treatments (500 J/m2) followed by 3 h of PR light. (A) Kaplan-Meier plot showing the fraction of tumor-bearing mice in time after the first UV treatment. K14-CPD-PL and K14-CPD/6-4PP-PL animals remain tumor-free longer than wt littermates (P < 0.01). (B) The average number of squamous cell carcinomas per mouse in time after the first UV treatment.

(ii) Removal of CPDs from the total mouse skin, but not from basal keratinocytes alone, abrogates UV-mediated immunosuppression. Photocarcinogenesis does not merely depend on the induction of mutations; failing surveillance by the immune system plays a pivotal role in this process as well. It has long been demonstrated that UV-induced skin tumors are highly immunogenic and are immediately rejected when transplanted onto syngeneic mice, unless recipient animals are exposed to subcarcinogenic doses of UVB light prior to transplantation (9). Thus, UV radiation exerts immunosuppressive effects, allowing skin tumors to persist. Among the potential chromophores that are thought to be responsible for eliciting a suppressive response upon UV are urocanic acid (UCA) and DNA. UCA is formed in the stratum corneum by deamination of histidine. Upon absorption of UV light, the naturally occurring trans form isomerizes to into cis UCA that is known to act as a mediator of the UV-induced immunosuppression (14, 31). The dramatic skin cancer protection observed upon fast removal of CPDs may, therefore, be caused by both a reduction in the mutation load and an alteration of the immune response (22). Of note, this comprehensive set of transgenic photolyase mice now enables a thorough investigation of the role of DNA lesions in the UV-mediated immunosuppressive response.

First, we studied the systemic immune response in the ubiquitously expressing ß-act-CPD-PL and ß-act-6-4PP-PL mice. Mice were exposed to five daily UVB treatments, applied at two spots on the shaved back. After every UV exposure, mice were exposed to PR light for 3 h or else kept in the dark. Four days after the last UVB exposure, all mice were sensitized by application of picryl chloride (PCl) on their shaved bellies. Four days later, the ear thickness was measured (time zero), and ears were challenged with PCl. Ear thickness was measured again 24 h after the challenge, and the immune response of each mouse was expressed as a percentage of ear swelling. To exclude that exposure to PR light might affect the immune response, we first studied the suppression of the immune response in mice that did not express the photolyase transgene (Fig. 5A). As expected, exposure to UVB, but importantly not to PR light alone, resulted in a reduced immune response, confirming the immunosuppressive effects of UVB irradiation and the use of PR as a valid tool to study the contribution of DNA lesions to immunosuppression. Exposure of ß-act-CPD-PL mice to UVB irradiation resulted in a clear suppression of the immune system (Fig. 5B). Strikingly, ubiquitous photolyase-mediated removal of CPDs abolished completely the immunosuppressive effects of UV light, revealing a critical role for these lesions in the response. In contrast, exposing irradiated ß-act-6-4PP-PL mice to PR light did not alter the immunosuppression (Fig. 5C).


Figure 5
View larger version (17K):
[in this window]
[in a new window]
  		FIG. 5. Systemic immunosuppression in wt and ß-act-photolyase mice. wt (A), ß-act-CPD-PL (B), and ß-act-6-4PP-PL (C) mice were irradiated with 0.5 MED UVB for five consecutive days. The mice were skin-sensitized 4 days after the last UV exposure by topical application of PCl to nonirradiated shaved abdomen, chest, and feet. Four days after sensitization, both ears of the mice were challenged with 0.8% PCl in olive oil. At 24 h after the challenge, duplicate ear measures were performed with an engineer's micrometer. Thirty-four animals per genotype were used. Error bars indicate the standard errors of the means.

Keratinocytes are thought to play an important role in modulation of the immune response by releasing various cytokines upon exposure to UV. To examine whether the presence of DNA damage in other cells than the basal keratinocytes plays a similarly critical role in induction of immune suppression as it does in acute sunburn, hyperplasia, and carcinogenesis, we studied the systemic immunosuppression upon UV exposure in K14 photolyase mice. Figure 6 shows that UV exposure of both K14-6-4PP or K14-CPD PL mice resulted in suppression of the systemic immune response, comparable to the suppression in wt mice. As such, K14-CPD PL mice remarkably contrast with ß-act-CPD-PL mice, in which ubiquitous removal of CPDs eliminates immunosuppression. These findings suggest that, whereas CPDs in basal keratinocytes comprise a major trigger for UV-induced sunburn and carcinogenesis, DNA lesions (and in particular CPDs) in other cell types are primarily responsible for immunosuppression.


Figure 6
View larger version (20K):
[in this window]
[in a new window]
  		FIG. 6. Systemic immunosuppression in K14-photolyase mice. K14-CPD-PL (A) and K14-6-4PP-PL (B) mice were irradiated with UVB for five consecutive days. The mice were skin sensitized 4 days after the last UV exposure by topical application of PCl to the nonirradiated shaved abdomen, chest, and feet. Four days after sensitization, both ears of the mice were challenged with 0.8% PCl in olive oil. At 24 h after the challenge, duplicate ear measures were performed with an engineer's micrometer. Thirty-four animals per genotype were used. Error bars indicate the standard errors of the means.


arrow
DISCUSSION
 
Photolyase transgenic mice. The generation of ubiquitously expressing ß-act-CPD and ß-act-6-4PP photolyase transgenic mice, along with basal keratinocyte-specific K14-CPD photolyase transgenic mice has been of great value to delineate the role of distinct DNA lesions in acute UV effects (i.e., sunburn) and skin cancer (22, 35). To further improve our knowledge on lesion and cell type specificity in relation to photocarcinogenesis, we now have generated mice expressing the 6-4PP-photolyase from the basal keratinocyte-specific promoter K14. Exposure of UV-irradiated K14-6-4PP-PL mice to visible light confirmed the functionality and specificity of the transgene, as it allowed efficient repair of the 6-4 PP lesions in the basal keratinocytes only, whereas such lesions persisted in all other cell types (Fig. 1D).

Role of basal keratinocytes in apoptosis and hyperplasia. Using the K14-6-4PP-PL mouse model, we have shown that removal of 6-4PPs from basal keratinocytes had no effect on either apoptosis (sunburn) or hyperplasia, consistent with our previous findings with ubiquitously expressing 6-4PP-PL transgenic mice (21). In contrast, removal of CPDs from basal keratinocytes only substantially decreased the incidence of cell death (33) (Fig. 2A), as well as the proliferative response resulting in hyperplasia (this study). A similar observation was made after ubiquitous removal of CPDs from the mouse skin. Taken together, these findings indicate a central role of basal keratinocytes and CPD lesions in the onset of these UV-induced acute skin effects. These findings also corroborate the observation that in rodent cells CPDs—in contrast to 6-4PPs—are hardly recognized and virtually not repaired by GG-NER. Thus, whereas an additional mechanism for the repair of 6-4PPs does not significantly improve their UV resistance, providing mice with the CPD-photolyase enzyme proves beneficial.

Photoreactivation-dependent transcriptional responses of the UVB-exposed K14-CPD-PL mouse skin. Using cultured MDFs, PR was previously shown to impact, at the transcriptional level, biologically significant cellular UV-induced responses (12) and, thus, was expected to affect gene expression considerably in our in vivo experimental model system. Importantly, hierarchical clustering as well as principal component analysis grouped the significantly transcribed genes primarily by PR status (Fig. 3A and B), thereby establishing CPDs in basal keratinocytes as a major determinant of gene expression changes in the UV-exposed mouse skin. The fact that the transcriptional response to CPDs is time dependent, with different kinetics depending on the UV dose, thoroughly explains why, in this study, the PR samples were still remote from the controls in the three-dimensional space at this relatively early time point. In this light, the fact that the systemic immunosuppression is maintained in PR UV-treated CPD-PL mice, and as such dissociates from the acute inflammation and carcinogenesis, is highly interesting.

Exposure of the K14-CPD-PL mouse skin to UV revealed several PR-dependent, and thus CPD-provoked biological processes (Fig. 3C). In cultured cells, DNA photoproducts can stall replication, thereby generating a more toxic lesion, the double-strand break (12). In this study, we identified genes involved in DNA recombination and the repair of DNA breaks (e.g., Rad51 and Ubl1), which suggests that, rather than CPDs themselves, CPD-dependent replication products (e.g., stalled replication forks and DNA breaks) in basal keratinocytes are likely to contribute to UV effects in the skin, which is in perfect agreement with our in vitro data (12). There has been a long-standing question whether damage to DNA can affect nucleosomal stability, thereby altering lesion accessibility (40). For instance, exposure of human epidermal keratinocytes to UVB induces the expression of genes involved in histone synthesis (6). We now have shown that CPDs are a major determinant in upregulating the expression of genes involved in the modification and maintenance of nucleosomal structures. Additionally, the upregulation of genes associated with ribosomal biosynthesis, translational elongation, and histone synthesis could all reflect the onset of a proliferative response, underlying the observed hyperplasia and/or counteracting cell loss due to apoptosis. None of these genes was upregulated upon photoreactivation of CPDs in the mouse skin, signifying the autonomous and predominant nature of the basal UV-induced cell response to CPDs. Both the proapoptotic response at the transcriptional level and the subsequent induction of cell death (40 h after UV irradiation) in irradiated non-PR mouse skin underscore the destructive potential of CPDs (or CPD-dependent replication products), revealing that their sole presence in basal keratinocytes suffices to trigger apoptosis. In addition to the data presented here, a number of additional genes were shown to demonstrate PR-dependent expression. These annotated data sets are available for interactive querying and hypothesis-driven analyses at http://www2.eur.nl/fgg/ch1/k14network/.

Role of basal keratinocytes in skin cancer. Previously, we have shown that ubiquitous removal of CPDs from the skin provides a high level of protection from cancer initiation, pointing to this lesion as the major trigger for photocarcinogenesis (21) To examine whether the presence of UV-induced photolesions in basal keratinocytes constitutes the primary cause of skin cancer, we investigated the incidence of skin tumors in K14-CPD and K14-6-4PP photolyase mice upon exposure to UVB and subsequent specific removal of lesions by PR. Strikingly, fast removal of CPDs from basal keratinocytes dramatically decreased the incidence of skin cancer in K14-CPD-PL mice (Fig. 4B) to a level similar to that observed when CPDs were ubiquitously removed from the skin. This finding points to basal keratinocytes as the primary cell type for the tumorigenesis observed in the irradiated mouse skin, thereby limiting the impact of systemic mechanisms. Even more, since PR in keratinocytes can prevent apoptosis and keratinocyte-associated carcinogenesis, an interesting follow-up experiment would be to expose K14-CPD-PL/K14-6-4PP-PL double transgenic mice to the maximum tolerable dose of UV (which in the absence of photoreactivation likely would be lethal) and find out whether melanomas develop.

DNA damage and immunosuppression. The relevance of UV-mediated immunosuppressive effects in carcinogenesis was first demonstrated by Fisher and Kripke, who showed that transplantation of UV-induced skin tumors to syngeneic mice can result in tumor rejection (9). If, however, recipient mice had received subcarcinogenic doses of UV light, rejection did not take place, indicating that UV light allows persistence of skin tumors by suppressing the immune system. However, the mechanism underlying this process is not yet clear. Both UCA and DNA damage can act as chromophores responsible for immunosuppression (24, 31, 41). Importantly, treating mice with HindIII-containing liposomes was previously shown to induce immune suppression (30). These data in combination with the earlier data published by Kripke and coworkers (24) indicate that DNA damage, either as double-strand breaks (HindIII) or pyrimidine dimer formation, is sufficient to activate immune suppression. Utilizing the ubiquitously expressing ß-act-photolyase mice, it is now possible to investigate to what extent specific UV-induced DNA lesions affect the immune system by removing either CPDs or 6-4PPs in a light-dependent manner. In contrast to 6-4PPs, ubiquitous removal of CPDs abolished UV-induced immunosuppression, suggesting that photoisomerization of trans UCA alone is not sufficient to induce the suppressive effects; CPD lesions are a prerequisite. In contrast to CPDs, repair of 6-4PPs by NER is fast and adequate, providing mice with a system that further enhances removal of 6-4PPs and does not significantly affect the biological outcome. However, providing mice with CPD-photolyase, thereby allowing fast removal of the most abundant UV lesion, does affect the biological outcome and appears beneficial to the animal. Thus, these results confirm and expand on the earlier findings (24, 30) suggesting that DNA damage (strand breaks, pyrimidine dimers, or 6-4 photoproducts) induces immune suppression but that only pyrimidine dimer formation is involved in photocarcinogenesis.

To investigate to what extent 6-4PPs are capable of eliciting immunosuppressive effects, one could breed photolyase mice with available NER-deficient mice. For instance, complete NER abrogation, (e.g., Xpa knockout mice), strongly increases the susceptibility to UV-induced systemic immunosuppression (13, 29). However, whereas in mice lacking only TC-NER (e.g., Csb mutant mice) or GG-NER (e.g., XPC mice) systemic immunosuppression is substantially reduced, local immunosuppression correlates with local Langerhans cell depletion and occurs at lower UV doses in TC-NER-deficient mice (Csb and Xpa mice) (23). In contrast, the UV dose required to suppress the immune system in XPC mice is similar to the dose required in wt mice. Thus, TC-NER deficiency likely enhances local immunosuppression. Furthermore, only a total NER defect results in increased systemic immunosuppression. Crossing our photolyase transgenic mice to NER-deficient mice will elucidate the role of 6-4PPs in this process and indicate whether their presence on transcribed versus nontranscribed regions leads to differential effects.

Role of basal keratinocytes in immunosuppression. Next, we sought to examine the relevant cell types involved in the immune suppression induced by UV. For example, Langerhans cells migrate towards the draining lymph nodes in irradiated skin, shifting the Th1/Th2 balance towards a Th2 response (41). Using the K14-driven photolyase mice, it was possible to remove DNA lesions from basal keratinocytes, whereas damage remained in, e.g., Langerhans cells and differentiated keratinocytes. Removal of DNA lesions from basal keratinocytes did not abolish systemic immunosuppression, suggesting that DNA lesions in other cell types are sufficient to induce this response. This discrepancy between systemic immune suppression and the acute UV effects related to DNA damage in basal keratinocytes (i.e., sunburn and hyperplasia) is in line with previous findings in Csb mice (12). These mice demonstrate UV sensitivity with acute epithelial effects, yet their systemic immune suppression profile is comparable to that of wt mice. To unravel the cell type(s) responsible for UV-induced immunosuppression, transgenic mice have been generated in which the photolyase transgene is expressed specifically in other skin cell types, e.g., Langerhans cells.


arrow
ACKNOWLEDGMENTS
 
We thank O. Nikaido for providing us with the TDM2 and 64M2 antibodies. Also, we thank Jun-Ichi Miyazaki (Osaka University Medical School, Osaka, Japan) for providing us with the pCY4B vector and J. Garssen for the supply of picryl chloride.

This work was supported by the Dutch Cancer Foundation (EUR 98-1774 and EMCR 2002-2701), the Interuniversitary Research Institute for Radiopathology and Radioprotection (IRS grant 7.3.5), the Association for International Cancer Research (AICR 98-259 and AICR 03-128), and the Japanese Ministry of Education, Science and Culture (MONBUSHO grant 10044231).


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: MGC, Department of Cell Biology and Genetics, Center for Biomedical Genetics, Erasmus University Medical Center, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. Phone: 31 10 4087455. Fax: 31 10 4089468. E-mail: g.vanderhorst{at}erasmusmc.nl. Back

{triangledown} Published ahead of print on 11 September 2006. Back

{dagger} J.J. and G.A.G. contributed equally to this work. Back

{ddagger} Present address: Department of Molecular and Cell Biology, University of California at Berkeley, 125 Koshland Hall, Berkeley, Calif. Back

§ Present address: Novartis, Institute of Tropical Disease, Singapore. Back


arrow
REFERENCES
 
    1
  1. Armstrong, B. K., and A. Kricker. 2001. The epidemiology of UV induced skin cancer. J. Photochem. Photobiol. B 63:8-18.[CrossRef][Medline]
    2. 2
  2. Bohr, V. A., C. A. Smith, D. S. Okumoto, and P. C. Hanawalt. 1985. 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.[CrossRef][Medline]
    3. 3
  3. Boonstra, A., A. van Oudenaren, M. Baert, H. van Steeg, P. J. Leenen, G. T. van der Horst, J. H. Hoeijmakers, H. F. Savelkoul, and J. Garssen. 2001. Differential ultraviolet-B-induced immunomodulation in XPA, XPC, and CSB DNA repair-deficient mice. J. Investig. Dermatol. 117:141-146.[CrossRef][Medline]
    4. 4
  4. Bootsma, D., K. H. Kraemer, J. E. Cleaver, and J. H. J. Hoeijmakers. 2001. Nucleotide excision repair syndromes: xeroderma pigmentosum, Cockayne syndrome and trichothiodystrophy, p. 677-703. In C. R. Scriver, A. L. Beaudet, W. S. Sly, and D. Valle (ed.), The metabolic and molecular bases of inherited disease. McGraw-Hill, New York, N.Y.
    5. 5
  5. Coffer, P. J., B. M. Burgering, M. P. Peppelenbosch, J. L. Bos, and W. Kruijer. 1995. UV activation of receptor tyrosine kinase activity. Oncogene 11:561-569.[Medline]
    6. 6
  6. Dazard, J. E., H. Gal, N. Amariglio, G. Rechavi, E. Domany, and D. Givol. 2003. Genome-wide comparison of human keratinocyte and squamous cell carcinoma responses to UVB irradiation: implications for skin and epithelial cancer. Oncogene 22:2993-3006.[CrossRef][Medline]
    7. 7
  7. de Boer, J., and J. H. Hoeijmakers. 2000. Nucleotide excision repair and human syndromes. Carcinogenesis 21:453-460.[Abstract/Free Full Text]
    8. 8
  8. de Laat, W. L., N. G. Jaspers, and J. H. Hoeijmakers. 1999. Molecular mechanism of nucleotide excision repair. Genes Dev. 13:768-785.[Free Full Text]
    9. 9
  9. Fisher, M. S., and M. L. Kripke. 1977. Systemic alteration induced in mice by ultraviolet light irradiation and its relationship to ultraviolet carcinogenesis. Proc. Natl. Acad. Sci. USA 74:1688-1692.[Abstract/Free Full Text]
    10. 10
  10. Friedberg, E., G. C. Walker, and W. Siede (ed.). 1995. DNA repair and mutagenesis. ASM Press, Washington, D.C.
    11. 11
  11. Friedberg, E. C. 2003. DNA damage and repair. Nature 421:436-440.[CrossRef][Medline]
    12. 12
  12. Garinis, G. A., J. R. Mitchell, M. J. Moorhouse, K. Hanada, H. de Waard, D. Vandeputte, J. Jans, K. Brand, M. Smid, P. J. van der Spek, J. H. Hoeijmakers, R. Kanaar, and G. T. van der Horst. 2005. Transcriptome analysis reveals cyclobutane pyrimidine dimers as a major source of UV-induced DNA breaks. EMBO J. 24:3952-3962.[CrossRef][Medline]
    13. 13
  13. Garssen, J., H. van Steeg, F. de Gruijl, J. de Boer, G. T. van der Horst, H. van Kranen, H. van Loveren, M. van Dijk, A. Fluitman, G. Weeda, and J. H. Hoeijmakers. 2000. Transcription-coupled and global genome repair differentially influence UV-B-induced acute skin effects and systemic immunosuppression. J. Immunol. 164:6199-6205.[Abstract/Free Full Text]
    14. 14
  14. Gruner, S., H. Oesterwitz, H. Stoppe, W. Henke, R. Eckert, and N. Sonnichsen. 1992. Cis-urocanic acid as a mediator of ultraviolet-light-induced immunosuppression. Semin. Hematol. 29:102-107.[Medline]
    15. 15
  15. Hoeijmakers, J. H. 2001. Genome maintenance mechanisms for preventing cancer. Nature 411:366-374.[CrossRef][Medline]
    16. 16
  16. Hogan, B., R. Beddington, F. Costantini, and E. Lacy. 1994. Manipulating the mouse embryo: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
    17. 17
  17. Hosack, D. A., G. J. Dennis, B. T. Sherman, H. C. Lane, and R. A. Lempicki. 2003. Identifying biological themes within lists of genes with EASE. Genome Biol. 4:R70. [Online.] http://genomebiology.com/2003/4/10/R70.[CrossRef][Medline]
    18. 18
  18. Hwang, B. J., J. M. Ford, P. C. Hanawalt, and G. Chu. 1999. 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.[Abstract/Free Full Text]
    19. 19
  19. Hwang, B. J., S. Toering, U. Francke, and G. Chu. 1998. p48 Activates a UV-damaged-DNA binding factor and is defective in xeroderma pigmentosum group E cells that lack binding activity. Mol. Cell. Biol. 18:4391-4399.[Abstract/Free Full Text]
    20. 20
  20. Iordanov, M. S., D. Pribnow, J. L. Magun, T. H. Dinh, J. A. Pearson, S. L. Chen, and B. E. Magun. 1997. Ribotoxic stress response: activation of the stress-activated protein kinase JNK1 by inhibitors of the peptidyl transferase reaction and by sequence-specific RNA damage to the alpha-sarcin/ricin loop in the 28S rRNA. Mol. Cell. Biol. 17:3373-3381.[Abstract]
    21. 21
  21. Iordanov, M. S., D. Pribnow, J. L. Magun, T. H. Dinh, J. A. Pearson, and B. E. Magun. 1998. Ultraviolet radiation triggers the ribotoxic stress response in mammalian cells. J. Biol. Chem. 273:15794-15803.[Abstract/Free Full Text]
    22. 22
  22. Jans, J., W. Schul, Y. G. Sert, Y. Rijksen, H. Rebel, A. P. Eker, S. Nakajima, H. van Steeg, F. R. de Gruijl, A. Yasui, J. H. Hoeijmakers, and G. T. van der Horst. 2005. Powerful skin cancer protection by a CPD-photolyase transgene. Curr. Biol. 15:105-115.[CrossRef][Medline]
    23. 23
  23. Kolgen, W., H. van Steeg, G. T. van der Horst, J. H. Hoeijmakers, W. A. van Vloten, F. R. de Gruijl, and J. Garssen. 2003. Association of transcription-coupled repair but not global genome repair with ultraviolet-B-induced Langerhans cell depletion and local immunosuppression. J. Investig. Dermatol. 121:751-756.[CrossRef][Medline]
    24. 24
  24. Kripke, M. L., P. A. Cox, L. G. Alas, and D. B. Yarosh. 1992. Pyrimidine dimers in DNA initiate systemic immunosuppression in UV-irradiated mice. Proc. Natl. Acad. Sci. USA 89:7516-7520.[Abstract/Free Full Text]
    25. 25
  25. Kulms, D., and T. Schwarz. 2000. Molecular mechanisms of UV-induced apoptosis. Photodermatol. Photoimmunol. Photomed. 16:195-201.[CrossRef][Medline]
    26. 26
  26. Leccia, M. T., M. J. Richard, F. Joanny-Crisci, and J. C. Beani. 1998. UV-A1 cytotoxicity and antioxidant defence in keratinocytes and fibroblasts. Eur. J. Dermatol. 8:478-482.[Medline]
    27. 27
  27. Mellon, I., G. Spivak, and P. C. Hanawalt. 1987. Selective removal of transcription-blocking DNA damage from the transcribed strand of the mammalian DHFR gene. Cell 51:241-249.[CrossRef][Medline]
    28. 28
  28. Mitchell, D. L. 1988. The relative cytotoxicity of (6-4) photoproducts and cyclobutane dimers in mammalian cells. Photochem. Photobiol. 48:51-57.[Medline]
    29. 29
  29. Miyauchi-Hashimoto, H., K. Tanaka, and T. Horio. 1996. Enhanced inflammation and immunosuppression by ultraviolet radiation in xeroderma pigmentosum group A (XPA) model mice. J. Investig. Dermatol. 107:343-348.[CrossRef][Medline]
    30. 30
  30. Nishigori, C., D. Yarosh, A. O'Connor, V. K. Shreedhar, S. E. Ullrich, P. Cox, and M. L. Kripke. 1998. HindIII liposomes suppress delayed-type hypersensitivity responses in vivo and induce epidermal IL-10 in vitro. J. Immunol. 161:2684-2691.[Abstract/Free Full Text]
    31. 31
  31. Noonan, F. P., and E. C. De Fabo. 1992. Immunosuppression by ultraviolet B radiation: initiation by urocanic acid. Immunol. Today 13:250-254.[CrossRef][Medline]
    32. 32
  32. Okada, K., Y. Takahashi, K. Ohnishi, O. Ishikawa, and Y. Miyachi. 1994. Time-dependent effect of chronic UV irradiation on superoxide dismutase and catalase activity in hairless mice skin. J. Dermatol. Sci. 8:183-186.[CrossRef][Medline]
    33. 33
  33. Peto, R., M. C. Pike, N. E. Day, R. G. Gray, P. N. Lee, S. Parish, J. Peto, S. Richards, and J. Wahrendorf. 1980. Guidelines for simple, sensitive significance tests for carcinogenic effects in long-term animal experiments. IARC Monogr. Eval. Carcinog. Risk Chem. Hum. 2(Suppl.):311-426.
    34. 34
  34. Sasaki, H., H. Akamatsu, and T. Horio. 1997. Effects of a single exposure to UVB radiation on the activities and protein levels of copper-zinc and manganese superoxide dismutase in cultured human keratinocytes. Photochem. Photobiol. 65:707-713.[Medline]
    35. 35
  35. Schul, W., J. Jans, Y. M. Rijksen, K. H. Klemann, A. P. Eker, J. de Wit, O. Nikaido, S. Nakajima, A. Yasui, J. H. Hoeijmakers, and G. T. van der Horst. 2002. Enhanced repair of cyclobutane pyrimidine dimers and improved UV resistance in photolyase transgenic mice. EMBO J. 21:4719-4729.[CrossRef][Medline]
    36. 36
  36. Shindo, Y., E. Witt, D. Han, B. Tzeng, T. Aziz, L. Nguyen, and L. Packer. 1994. Recovery of antioxidants and reduction in lipid hydroperoxides in murine epidermis and dermis after acute ultraviolet radiation exposure. Photodermatol. Photoimmunol. Photomed. 10:183-191.[Medline]
    37. 37
  37. Shindo, Y., E. Witt, and L. Packer. 1993. Antioxidant defense mechanisms in murine epidermis and dermis and their responses to ultraviolet light. J. Investig. Dermatol. 100:260-265.[CrossRef][Medline]
    38. 38
  38. Smid, M., and L. C. Dorssers. 2004. GO-Mapper: functional analysis of gene expression data using the expression level as a score to evaluate gene ontology terms. Bioinformatics 20:2618-2625.[Abstract/Free Full Text]
    39. 39
  39. Sugasawa, K., T. Okamoto, Y. Shimizu, C. Masutani, S. Iwai, and F. Hanaoka. 2001. A multistep damage recognition mechanism for global genomic nucleotide excision repair. Genes Dev. 15:507-521.[Abstract/Free Full Text]
    40. 40
  40. Thoma, F. 1999. Light and dark in chromatin repair: repair of UV-induced DNA lesions by photolyase and nucleotide excision repair. EMBO J. 18:6585-6598.[CrossRef][Medline]
    41. 41
  41. Ullrich, S. E. 2005. Mechanisms underlying UV-induced immune suppression. Mutat. Res. 571:185-205.[Medline]
    42. 42
  42. van Hoffen, A., A. T. Natarajan, L. V. Mayne, A. A. van Zeeland, L. H. Mullenders, and J. Venema. 1993. Deficient repair of the transcribed strand of active genes in Cockayne's syndrome cells. Nucleic Acids Res. 21:5890-5895.[Abstract/Free Full Text]
    43. 43
  43. Wassberg, C., M. Thorn, A. M. Johansson, R. Bergstrom, B. Berne, and U. Ringborg. 2001. Increasing incidence rates of squamous cell carcinoma of the skin in Sweden. Acta Derm. Venereol. 81:268-272.[CrossRef][Medline]
    44. 44
  44. Wood, R. D. 1996. DNA repair in eukaryotes. Annu. Rev. Biochem. 65:135-167.[CrossRef][Medline]
    45. 45
  45. Wu, M. C., S. Hunsberger, and D. Zucker. 1994. Testing for differences in changes in the presence of censoring: parametric and nonparametric methods. Stat. Med. 13:635-646.[Medline]
    46. 46
  46. Yasui, A., and A. P. M. Eker. 1997. DNA photolyases, p. 9-32. In J. A. Nickoloff and M. F. Hoekstra (ed.), DNA damage and repair: biochemistry, genetics and cell biology, vol. 2. Humana Press, Totowa, N.Y.

Molecular and Cellular Biology, November 2006, p. 8515-8526, Vol. 26, No. 22
0270-7306/06/$08.00+0     doi:10.1128/MCB.00807-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Black, A. T., Gray, J. P., Shakarjian, M. P., Laskin, D. L., Heck, D. E., Laskin, J. D. (2008). Distinct effects of ultraviolet B light on antioxidant expression in undifferentiated and differentiated mouse keratinocytes. Carcinogenesis 29: 219-225 [Abstract] [Full Text]  
  • Tamanini, F., Chaves, I., Bajek, M. I., van der Horst, G. T. J. (2007). Structure Function Analysis of Mammalian Cryptochromes. Cold Spring Harb Symp Quant Biol 72: 133-139 [Abstract]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jans, J.
Right arrow Articles by van der Horst, G. T. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jans, J.
Right arrow Articles by van der Horst, G. T. J.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»