Previous Article |
Table of Contents
| Next Article
MEDICAL SCIENCES
DDB2 gene disruption leads to skin tumors and resistance to apoptosis after exposure to ultraviolet light but not a chemical carcinogen




Divisions of *Biochemistry and Molecular Biology and
Immunology and Cancer Research Laboratory, Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720-3202; and
Department of Dermatology, Jikei University School of Medicine, Tokyo 105-8461, Japan
Edited by Carlo M. Croce, Thomas Jefferson University, Philadelphia, PA, and approved December 22, 2003 (received for review October 9, 2003)
| Abstract |
|---|
|
|
|---|
The binding to DNA damages might suggest a direct role for DDB in nucleotide excision repair (NER) and DNA repair assays with rodent Chinese hamster ovary cell lines that lack DDB2 expression combined with ectopic overexpression of DDB2 cDNA led to the conclusion that DDB2 is a repair factor involved in global genomic repair, a subpathway of NER. However, the removal from DNA of many bulky lesions, including (6-4) photoproducts, occurs in the absence of DDB (8, 12, 13), and DDB2 is induced by UV-irradiation at a time by which NER has been completed. Moreover, the DDB2 promoter resembles that of a cell cycle-regulated gene (1, 14), and DDB2 is part of the COP9 signalosome complex (15, 16). Instead, DDB2 has been proposed to be a transcription transactivator through its stimulation of E2F1 (17), and we recently observed that DDB2 controls p53-mediated apoptosis after UV-irradiation of human diploid fibroblasts and that XP group E (XP-E) primary skin fibroblasts are abnormally resistant to killing by UV-irradiation (18). In sum, the functions of DDB are apparently varied and remain undefined.
The absence of DDB2 from Chinese hamster ovary cell lines led to the proposition that mouse models would not be appropriate for studying XP-E and related carcinogenesis (19, 20), although DDB activity for which both DDB1 and DDB2 proteins are required had been observed in mouse plasmacytoma cells (21). To elucidate any differences between mouse and human DDB2 function and to gain further insight into the role(s) of DDB2 and into the XP-E phenotype, we constructed mice with a disrupted DDB2 gene. Studies of these mice and their primary cells demonstrated that the mouse DDB2 gene is required for p53-mediated apoptosis and the suppression of skin carcinogenesis on exposure to UV-irradiation but, surprisingly, not to a chemical carcinogen.
| Materials and Methods |
|---|
|
|
|---|
Biochemical Analyses. RT-PCR analysis was performed as described (3, 18). Primers were 5'-GTTTAACCATCTCAATACCA and 5'-GTGTGAGGTGCTGTAACCA for DDB2, or 5'-GACAGAGGCAACTGAGCACC and 5'-CAAACGTCAAGACGGCCGTGTG for tubulin. Electrophoretic mobility-shift assays and Western blot analysis were as described (1, 3, 18), using 50% ammonium sulfate precipitate fractions (5). Antibodies against DDB2 (3, 18), BAX, p21CDKN1A, actin (Santa Cruz Biotechnology), MDM2 (SMP14, Santa Cruz Biotechnology; Ab-2, Oncogene Research Products), DDB1 (Zymed Laboratories), and p53 (1C12 [PDB] , Cell Signaling Technology) were used.
Mouse Fibroblasts and UV-Sensitivity Assays. MEFs (mouse embryonic fibroblasts) and neonatal fibroblasts were obtained from e13.5 embryos and p0 neonates (the day of birth), respectively, by a standard procedure. Cell-killing as measured by dye exclusion after UV-irradiation (18) or 7,12-dimethylbenz[a]anthracene (DMBA) exposure (2224) and caspase-3 assays (18) were as described.
Tumorigenesis Studies. For UV-B treatment, shaved dorsal skin of 8-week-old mice was irradiated for 5 days per week at a dose of 2,500 J·m2 per day by using a narrow-band UV-B TL-01 lamp (311 nm; Phillips). Individual mice were irradiated until either skin tumors appeared or for a maximum of 20 weeks. For treatment with DMBA, 10 µg of DMBA in 0.1 ml of acetone was applied to shaved dorsal skin once per week for 20 weeks. Mice were killed if profoundly ill or if external tumors exceeded 2 cm in diameter and these mice were scored as a death in the tumor incidence KaplanMeier analyses. Statistical significance was measured by using the log-rank test. For histology, skin tumors were fixed with 10% neutral-buffered formalin and stained with hematoxylin/eosin by conventional methods.
| Results |
|---|
|
|
|---|
|
DDB2/ mice were viable and fertile, and heterozygote crosses produced the expected Mendelian proportions among 488 offspring: 128 (26%) +/+, 236 (48%) +/, and 125 (26%) /. Eight weeks after birth, DDB2/ mice were somewhat smaller than their DDB2+/+ or DDB2+/ littermates {males, 24.49 ± 0.28 g (n = 51) [+/+], 24.13 ± 0.29 g (n = 82) [+/], 22.28 ± 0.40 g (n = 49) [/]; females, 19.44 ± 0.26 g (n = 52) [+/+], 19.26 ± 0.24 g (n = 76) [+/], 18.14 ± 0.24 g (n = 47) [/]}.
DDB2/ MEFs Are Abnormally Resistant to Killing After UV-Irradiation but Not After DMBA Exposure. To determine the effect of the DDB2 disruption on susceptibility to killing by UV-irradiation or chemical carcinogens, MEFs (e13.5) were treated with UV-C (254 nm) or DMBA, a chemical carcinogen that forms bulky DNA adducts. As predicted from human XP-E diploid fibroblasts (18), MEFs from the DDB2/ mice were more resistant than those from DDB2+/+ or DDB2+/ mice to killing by UV-irradiation as measured by dye exclusion (Fig. 2 A and B). By contrast, all three cell types showed equal killing by the bulky chemical carcinogen (Fig. 2 C and D). Colony-forming assays could not be used for the MEF studies because of the extremely low plating efficiency and clonogenic ability of MEFs (2325). (In any case, colony-forming assays do not measure specifically early events in killing processes, notably apoptosis.)
|
Nontransformed fibroblasts from XP-E patients are resistant to UV-irradiation because they have an abnormally low or no apoptotic response to the radiation (18). To clarify the effect of the mouse DDB2 gene disruption on p53-mediated apoptosis, we examined in MEFs the activity of caspase-3, an effector of the apoptosis pathway. Indeed, whereas the basal levels of caspase-3 were not significantly different between the DDB2/ and DDB2+/+ MEFs, the induction of caspase-3 activity by UV-irradiation was greatly impaired in the DDB2/ cells (Fig. 2E).
The amounts of several other proteins involved in p53-mediated responses to UV-irradiation were also examined in extracts from neonatal cells from the mouse strains (Fig. 3A). As determined by immunoblots, the basal and induced amounts of p53 and its downstream proteins, MDM2, BAX, and p21CDKN1A, were generally reduced in the DDB2/ cells (Fig. 3A). (The BAX induction was underestimated in the DDB2+/+ cells, but not the DDB2/ cells because of the absence of the lysed apoptotic cells from the DDB2+/+ sample.) Therefore, diminished p53-mediated apoptosis could explain the resistance of the mouse DDB2/ cells to killing by UV-irradiation.
|
As seen in Fig. 1E, mouse DDB2 protein is difficult to detect, because available antibodies to DDB2 are each directed against human DDB2 and poorly cross-react with mouse DDB2. Therefore, levels of DDB2 protein in the mouse cells were estimated from damaged DNA binding assays (Fig. 3B). [In human diploid fibroblasts, levels of DDB2 protein are limiting for the DDB heterodimer and paralleled damaged DNA binding activity (1, 3).] Consistent with the observations in normal human cells (1, 18), DDB activity decreased on irradiation and then reappeared after DNA repair was largely completed in the mouse DDB2+/+ cells (Fig. 3B). However, no DDB activity was detectable in DDB2/ cells before or after UV-irradiation (Fig. 3B). As with human cells (18), mouse DDB1 protein amounts were not affected by UV-irradiation (Fig. 3A).
DDB2/ and DDB2+/ Mice Have Dramatically Enhanced Skin Tumorigenesis After Exposure to UV Light but Not After Exposure to DMBA. To establish whether DDB2/ mice are susceptible to developing skin cancers, mice with shaved back skin were treated with UV-B (311 nm). No significant changes in the skin (e.g., sunburn) were found in DDB2+/+, DDB2+/, or DDB2/ mice after 4 days of exposure to 2,500 J·m2 per day. Mice were also irradiated with 2,500 J·m2 of UV-B five times per week for 20 weeks or until tumors developed. DDB2/ mice started to develop skin tumors 13 weeks after the start of treatment, whereas both DDB2+/+ and DDB2+/ mice first showed tumors after 19 weeks (Fig. 4A). Once started, however, tumor development proceeded at a much greater rate in the DDB2+/ mice than in the DDB2+/+ mice. There were also significant differences in the number of UV-induced tumors among the three genotypes (Fig. 4B). For example, there were 0.08, 0.3, or 13 tumors per mouse for DDB2+/+, DDB2+/, or DDB2/ mice, respectively, 25 weeks after the start of treatment.
|
When the shaved backs were exposed to two DMBA treatments of 10 µg per week, no significant changes in the skin were noted. After treatment of 10 µg of DMBA per week for 20 weeks, however, tumor development started after 19, 16, or 17 weeks for DDB2+/+, DDB2+/, or DDB2/ mice, respectively (Fig. 4C). Moreover, the pattern of the development was not significantly different among the three genotypes (Fig. 4C) and the DMBA treatment resulted in no significant difference in the number of tumors among the three genotypes (Fig. 4D).
All tumors appeared only on UV-B or DMBA exposed skin areas and none appeared internally. The tumors were predominantly squamous cell carcinomas after UV exposure (Fig. 4 E and F) or papillomas after DMBA exposure (Fig. 4G) among all three genotypes. [The papillomas would histopathologically be classified in humans as trichilemmal tumors (proliferating trichilemmal cysts or trichilemmal horns) rather than papillomas.]
| Discussion |
|---|
|
|
|---|
The apparent conservation of DDB2 in mice and the development of UV-induced tumors in DDB2/ animals strongly support the validity of mouse models for studying XP-E and DDB2 function. Importantly, the DDB2/ (XPE) mice did not exhibit any abnormality in DMBA-induced carcinogenesis, whereas mouse models that are defective in NER genes such as XPA (23, 28), XPD (29), or CSB (24) show abnormal skin cancer predisposition after either UV-irradiation or exposure to DMBA, presumably because these mice are defective in proteins directly involved in NER and cannot normally repair bulky DNA adducts in general. By contrast, TP53/ mice do not increase initiation or promotion of DMBA-induced skin tumors (30), strongly supporting the proposal that DDB2 is involved in controlling p53 levels and/or its activity after UV-irradiation. We are currently crossing p53/ and DDB2/ mice to further study the DDB2p53 interactions.
Does DDB2 affect responses to DNA damages brought about by agents other than UV-irradiation? The mismatch repair proteins hMLH1, hMSH2, hMSH6, and hPMS2 have been implicated in p53-dependent and p53-independent apoptosis induced by alkylating agents, suggesting that these have dual roles of DNA repair and apoptotic signaling (refs. 3136; reviewed in ref. 37). Conversely, DDB2 mRNA appears to be up-regulated in response to
-radiation (38). How mismatch repair proteins, DDB2, or other proteins might each regulate p53 in response to DNA damage are clearly questions that warrant intense study.
As a final note, the enhanced tumorigenesis after UV-irradiation of the DDB2+/ mice even at the subsunburn doses used also emphasizes not only the utility of the DDB2 mouse model but also the need to carefully assess susceptibilities to cancer of individuals heterozygous for XP-E or other genes involved in signaling or repair in response to DNA damages. Moreover, the apparent normality of DDB2/ mice in development observed here and the late onset of tumors in XP-E individuals (2, 3) could indicate that XP-E is often not diagnosed and might be more common in the human population then has been appreciated. In this regard, we look forward to data from the Environmental Genome Project of the National Institute of Environmental Health Sciences (www.genome.utah.edu/genesnps) in which correlations of single-nucleotide polymorphisms of some 200 human genes including DDB2 are being matched against susceptibility to environmental diseases.
| Acknowledgements |
|---|
| Footnotes |
|---|
Abbreviations: DMBA, 7,12-dimethylbenz[a]anthracene; ES, embryonic stem; MEF, mouse embryonic fibroblast; NER, nucleotide excision repair; XP, xeroderma pigmentosum; XP-E, XP group E.
To whom correspondence may be addressed. E-mail: slinn{at}socrates.berkeley.edu or toshiki{at}uclink4.berkeley.edu.
| References |
|---|
|
|
|---|
This article has been cited by other articles in HighWire Press-hosted journals:
![]() |
K. Sugasawa Xeroderma pigmentosum genes: functions inside and outside DNA repair Carcinogenesis, March 1, 2008; 29(3): 455 - 465. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Stoyanova, T. Yoon, D. Kopanja, M. B. Mokyr, and P. Raychaudhuri The Xeroderma Pigmentosum Group E Gene Product DDB2 Activates Nucleotide Excision Repair by Regulating the Level of p21Waf1/Cip1 Mol. Cell. Biol., January 1, 2008; 28(1): 177 - 187. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Itoh, S. Iwashita, M. B. Cohen, D. K. Meyerholz, and S. Linn Ddb2 is a haploinsufficient tumor suppressor and controls spontaneous germ cell apoptosis Hum. Mol. Genet., July 1, 2007; 16(13): 1578 - 1586. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Yang, D. Curley, M. W. Bosenberg, and H. Tsao Loss of Xeroderma Pigmentosum C (Xpc) Enhances Melanoma Photocarcinogenesis in Ink4a-Arf-Deficient Mice Cancer Res., June 15, 2007; 67(12): 5649 - 5657. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. M. Al Khateeb and D. F. Schroeder DDB2, DDB1A and DET1 Exhibit Complex Interactions During Arabidopsis Development Genetics, May 1, 2007; 176(1): 231 - 242. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Cang, J. Zhang, S. A. Nicholas, A. L. Kim, P. Zhou, and S. P. Goff DDB1 is essential for genomic stability in developing epidermis PNAS, February 20, 2007; 104(8): 2733 - 2737. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Hu, M. Shao, J. Yuan, L. Xu, F. Wang, Y. Wang, W. Yuan, J. Qian, H. Ma, Y. Wang, et al. Polymorphisms in DNA damage binding protein 2 (DDB2) and susceptibility of primary lung cancer in the Chinese: a case-control study Carcinogenesis, July 1, 2006; 27(7): 1475 - 1480. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Kulaksiz, J. T. Reardon, and A. Sancar Xeroderma Pigmentosum Complementation Group E Protein (XPE/DDB2): Purification of Various Complexes of XPE and Analyses of Their Damaged DNA Binding and Putative DNA Repair Properties Mol. Cell. Biol., November 15, 2005; 25(22): 9784 - 9792. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Alekseev, H. Kool, H. Rebel, M. Fousteri, J. Moser, C. Backendorf, F. R. de Gruijl, H. Vrieling, and L. H.F. Mullenders Enhanced DDB2 Expression Protects Mice from Carcinogenic Effects of Chronic UV-B Irradiation Cancer Res., November 15, 2005; 65(22): 10298 - 10306. [Abstract] [Full Text] [PDF] |
||||
![]() |
Q.-E. Wang, Q. Zhu, G. Wani, M. A. El-Mahdy, J. Li, and A. A. Wani DNA repair factor XPC is modified by SUMO-1 and ubiquitin following UV irradiation Nucleic Acids Res., July 19, 2005; 33(13): 4023 - 4034. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Holmberg, O. Fleck, H. A. Hansen, C. Liu, R. Slaaby, A. M. Carr, and O. Nielsen Ddb1 controls genome stability and meiosis in fission yeast Genes & Dev., April 1, 2005; 19(7): 853 - 862. [Abstract] [Full Text] [PDF] |
||||
![]() |
C.-L. Sun and C. C.-K. Chao Cross-Resistance to Death Ligand-Induced Apoptosis in Cisplatin-Selected HeLa Cells Associated with Overexpression of DDB2 and Subsequent Induction of cFLIP Mol. Pharmacol., April 1, 2005; 67(4): 1307 - 1314. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||