In Vivo Recruitment of XPC to UV-induced Cyclobutane Pyrimidine Dimers by the DDB2 Gene Product
2003; Elsevier BV; Volume: 278; Issue: 47 Linguagem: Inglês
10.1074/jbc.m307254200
ISSN1083-351X
AutoresMaureen E. Fitch, Satoshi Nakajima, Akira Yasui, James M. Ford,
Tópico(s)CRISPR and Genetic Engineering
ResumoThe initial step in mammalian nucleotide excision repair (NER) of the major UV-induced photoproducts, cyclobutane pyrimidine dimers (CPDs) and 6–4 photoproducts (6–4PPs), requires lesion recognition. It is believed that the heterodimeric proteins XPC/hHR23B and UV-DDB (UV-damaged DNA binding factor, composed of the p48 and p127 subunits) perform this function in genomic DNA, but their requirement and lesion specificity in vivo remains unknown. Using repair-deficient xeroderma pigmentosum (XP)-A cells that stably express photoproduct-specific photolyases, we determined the binding characteristics of p48 and XPC to either CPDs or 6–4PPs in vivo. p48 localized to UV-irradiated sites that contained either CPDs or 6–4PPs. However, XPC localized only to UV-irradiated sites that contained 6–4PPs, suggesting that XPC does not efficiently recognize CPDs in vivo. XPC did localize to CPDs when p48 was overexpressed in the same cell, signifying that p48 activates the recruitment of XPC to CPDs and may be the initial recognition factor in the NER pathway. The initial step in mammalian nucleotide excision repair (NER) of the major UV-induced photoproducts, cyclobutane pyrimidine dimers (CPDs) and 6–4 photoproducts (6–4PPs), requires lesion recognition. It is believed that the heterodimeric proteins XPC/hHR23B and UV-DDB (UV-damaged DNA binding factor, composed of the p48 and p127 subunits) perform this function in genomic DNA, but their requirement and lesion specificity in vivo remains unknown. Using repair-deficient xeroderma pigmentosum (XP)-A cells that stably express photoproduct-specific photolyases, we determined the binding characteristics of p48 and XPC to either CPDs or 6–4PPs in vivo. p48 localized to UV-irradiated sites that contained either CPDs or 6–4PPs. However, XPC localized only to UV-irradiated sites that contained 6–4PPs, suggesting that XPC does not efficiently recognize CPDs in vivo. XPC did localize to CPDs when p48 was overexpressed in the same cell, signifying that p48 activates the recruitment of XPC to CPDs and may be the initial recognition factor in the NER pathway. The majority of DNA damage induced by ultraviolet light is caused by the transformation of adjacent pyrimidines into either cyclobutane pyrimidine dimers (CPDs), 1The abbreviations used are: CPDcyclobutane pyrimidine dimerNERnucleotide excision repair6–4PPpyrimidine (6–4) pyrimidone photoproductPBSphosphate-buffered salineXPxeroderma pigmentosumUV-DDBUV-damaged DNA binding factorGGRglobal genomic repairTCRtranscription-coupled repair. or pyrimidine (6–4) pyrimidone photoproducts (6–4PPs). In human cells, these photoproducts are repaired exclusively through the nucleotide excision repair (NER) pathway, and loss of NER leads to the skin cancer-prone syndrome xeroderma pigmentosum (XP). NER can be further subdivided into two pathways, global genomic repair (GGR), which repairs lesions found throughout the genome, and transcription-coupled repair (TCR), which removes lesions selectively from the transcribed strand of active genes (reviewed in Refs. 1de Laat W.L. Jaspers N.G. Hoeijmakers J.H. Genes Dev. 1999; 13: 768-785Crossref PubMed Scopus (938) Google Scholar and 2Ford J.M. Hanawalt P.C. Curr. Top. Microbiol. Immunol. 1997; 221: 47-70PubMed Google Scholar). TCR is believed to be activated by the recognition of a stalled RNA polymerase at the lesion. Recognition of lesions by the GGR pathway is less well understood. Mutations in the XPC and DDB2 genes, resulting in the XP-C and XP-E complementation groups, respectively, lead to a loss in GGR but not TCR, suggesting that these genes encode for the UV-damage recognition factors initiating the GGR pathway (3Venema J. van Hoffen A. Karcagi V. Natarajan A.T. van Zeeland A.A. Mullenders L.H. Mol. Cell. Biol. 1991; 11: 4128-4134Crossref PubMed Scopus (296) Google Scholar, 4Hwang B.J. Ford J.M. Hanawalt P.C. Chu G. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 424-428Crossref PubMed Scopus (509) Google Scholar). Further evidence that XPC and p48 (the protein product of the DDB2 gene) are DNA damage recognition factors comes from in vitro and in vivo experiments of the interaction of these proteins with UV-induced DNA lesions. In vitro binding studies have shown that XPC complexed with the hHR23B protein has a greater affinity for naked UV-damaged DNA than undamaged DNA and has a greater affinity for the 6–4PP than the CPD (5Sugasawa K. Ng J.M. Masutani C. Iwai S. van der Spek P.J. Eker A.P. Hanaoka F. Bootsma D. Hoeijmakers J.H. Mol. Cell. 1998; 2: 223-232Abstract Full Text Full Text PDF PubMed Scopus (753) Google Scholar, 6Batty D. Rapic'-Otrin V. Levine A.S. Wood R.D. J. Mol. Biol. 2000; 300: 275-290Crossref PubMed Scopus (186) Google Scholar, 7Hey T. Lipps G. Sugasawa K. Iwai S. Hanaoka F. Krauss G. Biochemistry. 2002; : 6583-6587Crossref PubMed Scopus (94) Google Scholar, 8Kusumoto R. Masutani C. Sugasawa K. Iwai S. Araki M. Uchida A. Mizukoshi T. Hanaoka F. Mutat. Res. 2001; 485: 219-227Crossref PubMed Scopus (104) Google Scholar). p48 and the p127 product of the DDB1 gene make up the UV-DDB complex, which has also been shown to have a much greater affinity for UV-damaged DNA than undamaged and, like XPC, shows a stronger in vitro binding affinity for the 6–4PP over the CPD (6Batty D. Rapic'-Otrin V. Levine A.S. Wood R.D. J. Mol. Biol. 2000; 300: 275-290Crossref PubMed Scopus (186) Google Scholar, 9Keeney S. Chang G.J. Linn S. J. Biol. Chem. 1993; 268: 21293-21300Abstract Full Text PDF PubMed Google Scholar, 10Fujiwara Y. Masutani C. Mizukoshi T. Kondo J. Hanaoka F. Iwai S. J. Biol. Chem. 1999; 274: 20027-20033Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar). The precise role of UV-DDB in GGR is not well understood because the repair reaction can be recapitulated in vitro in its absence (11Aboussekhra A. Biggerstaff M. Shivji M.K. Vilpo J.A. Moncollin V. Podust V.N. Protic M. Hubscher U. Egly J.M. Wood R.D. Cell. 1995; 80: 859-868Abstract Full Text PDF PubMed Scopus (760) Google Scholar), although inclusion of UV-DDB can stimulate repair 2–17-fold (12Wakasugi M. Shimizu M. Morioka H. Linn S. Nikaido O. Matsunaga T. J. Biol. Chem. 2001; 276: 15434-15440Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar, 13Wakasugi M. Kawashima A. Morioka H. Linn S. Sancar A. Mori T. Nikaido O. Matsunaga T. J. Biol. Chem. 2002; 277: 1637-1640Abstract Full Text Full Text PDF PubMed Scopus (191) Google Scholar). cyclobutane pyrimidine dimer nucleotide excision repair pyrimidine (6–4) pyrimidone photoproduct phosphate-buffered saline xeroderma pigmentosum UV-damaged DNA binding factor global genomic repair transcription-coupled repair. One caveat of these in vitro binding studies is that the DNA substrates were not bound by nucleosomes or folded into higher order chromatin structures as would be found in an intact nucleus that had sustained DNA damage. Regulation of many DNA processes, such as transcription and replication, involves chromatin remodeling; it is now clear that DNA repair also requires chromatin remodeling to access damaged nucleotides (14Meijer M. Smerdon M.J. Bioessays. 1999; 21: 596-603Crossref PubMed Scopus (55) Google Scholar, 15Thoma F. EMBO J. 1999; 18: 6585-6598Crossref PubMed Google Scholar). Binding studies performed in vivo are necessary to fully understand how these structures may be affecting damage recognition. Use of a localized UV irradiation technique whereby only parts of the nucleus are irradiated through a micropore polycarbonate filter is a powerful tool for analyzing protein localization to sites of UV irradiation in vivo. We and others have used this technique to demonstrate that p48 and XPC bind rapidly to areas that have been irradiated and that the binding of both of these factors can occur in the absence of other important repair factors, such as XPA and p53 (13Wakasugi M. Kawashima A. Morioka H. Linn S. Sancar A. Mori T. Nikaido O. Matsunaga T. J. Biol. Chem. 2002; 277: 1637-1640Abstract Full Text Full Text PDF PubMed Scopus (191) Google Scholar, 16Fitch M.E. Cross I.V. Ford J.M. Carcinogenesis. 2003; 24: 843-850Crossref PubMed Scopus (67) Google Scholar, 17Volker M. Mone M.J. Karmakar P. van Hoffen A. Schul W. Vermeulen W. Hoeijmakers J.H. van Driel R. van Zeeland A.A. Mullenders L.H. Mol. Cell. 2001; 8: 213-224Abstract Full Text Full Text PDF PubMed Scopus (660) Google Scholar). Although UV photoproducts are repaired solely by the NER pathway in human cells, many other organisms possess an alternative pathway for the repair of individual UV photoproducts by the use of single enzymes called photolyases (18Sancar G.B. Mut. Res. 1990; 236: 147-160Crossref PubMed Scopus (159) Google Scholar). Once bound specifically to the lesion, these enzymes use energy from the visible spectrum of light to reverse the UV photoproduct without any incision or religation event, a process termed photoreactivation. Photolyases specific for either the CPD or the 6–4PP have been described from bacteria, lower eukaryotes, plants, and even marsupials (19Yasui A. Eker A.P. Yasuhira S. Yajima H. Kobayashi T. Takao M. Oikawa A. EMBO J. 1994; 13: 6143-6151Crossref PubMed Scopus (181) Google Scholar). UV irradiation induces substantially more CPDs than 6–4PPs, yet the repair, or consequence of having unrepaired lesions, of either of the individual photoproducts in DNA is unclear. Therefore, generation of an experimental system with defined UV photoproducts is of great interest for understanding the physiologic response to UV irradiation. Photolyases have recently been used in mammalian systems to address the contribution of the individual photoproducts to the mutagenic effects of UV-B (20You Y.H. Lee D.H. Yoon J.H. Nakajima S. Yasui A. Pfeifer G.P. J. Biol. Chem. 2001; 276: 44688-44694Abstract Full Text Full Text PDF PubMed Scopus (240) Google Scholar), the apoptotic response of HeLa cells to UV-C (21Chigancas V. Miyaji E.N. Muotri A.R. de Fatima J.J. Amarante-Mendes G.P. Yasui A. Menck C.F. Cancer Res. 2000; 60: 2458-2463PubMed Google Scholar), and also UV resistance in whole animals (22Schul W. Jans J. Rijksen Y.M.A. Klemann K.H.M. Eker A.P.M. de Wit J. Nikaido O. Nakajima S. Yasui A. Hoeijmakers J.H.J. van der Horst G.T.J. EMBO J. 2002; 21: 4719Crossref PubMed Scopus (84) Google Scholar). These studies have all suggested that the majority of the cellular response to UV is attributable to CPDs. Understanding the mechanism of recognition of CPDs versus 6–4PPs in the repair process is therefore of great importance. We have used NER-deficient XP-A cells that stably express either a heterologous CPD-specific photolyase or a 6–4PP-specific photolyase to address the specific roles in vivo of p48 and XPC in recognition of the individual photoproducts after UV irradiation. Antibodies—For immunoblots, rabbit polyclonal anti-photolyase antibodies were used at 1:2000 and horseradish peroxidase-conjugated donkey anti-rabbit at 1:5000 (Pierce) for chemiluminescent detection (anti-photolyase antibodies were gifts from Dr. Andre P. M. Eker, Erasmus University, Rotterdam, The Netherlands). 50 μg of protein was loaded on a 12% SDS-PAGE for analysis of the photolyases. Primary antibodies for immunofluorescence were mouse monoclonal anti-CPD at 1:1500 (TDM2), mouse monoclonal anti-6–4PP at 1:400 (64M-2) (gifts from Toshio Mori, Nara Medical University, Nara, Japan) (23Mori T. Nakane M. Hattori T. Matsunaga T. Ihara M. Nikaido O. Photochem. Photobiol. 1991; 54: 225-232Crossref PubMed Scopus (395) Google Scholar), rabbit polyclonal anti-photolyase antibodies at 1:500, and mouse anti-V5 fluorescein isothiocyanate conjugated at 1:500 (Invitrogen). Secondary antibodies were Alexa Fluor 594 goat anti-mouse and Alexa Fluor 488 goat anti-rabbit, both used at 1:500 (Molecular Probes). Immunoslotblot detection of photoproducts was performed as previously described (24Fitch M.E. Cross I.V. Turner S.J. Adimoolam S. Lin C.X. Williams K.G. Ford J.M. DNA Repair (Amst.). 2003; : 819-826Crossref PubMed Scopus (84) Google Scholar). Cell Lines—All cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 2 mm glutamine, antibiotics, and 500 μg/ml G418 (for maintenance of transgenic photolyase genes) and were incubated at 37 °C and 5% CO2. XP12ROSV, an SV40-transformed cell line derived from an XP-A patient, is mutated in the XPA gene and possesses no NER activity (25Satokata I. Tanaka K. Miura N. Narita M. Mimaki T. Satoh Y. Kondo S. Okada Y. Mutat. Res. 1992; 273: 193-202Crossref PubMed Scopus (73) Google Scholar). For expression of photolyase genes in the XP12ROSV cell line, a cDNA of the CPD photolyase gene (CPDphr) derived from rat kangaroo Potorous tridactylis (19Yasui A. Eker A.P. Yasuhira S. Yajima H. Kobayashi T. Takao M. Oikawa A. EMBO J. 1994; 13: 6143-6151Crossref PubMed Scopus (181) Google Scholar) and a cDNA of the 6–4PP photolyase gene (6–4phr) derived from plant Arabidopsis thaliana (26Nakajima S. Sugiyama M. Iwai S. Hitomi K. Otoshi E. Kim S.T. Jiang C.Z. Todo T. Britt A.B. Yamamoto K. Nucleic Acids Res. 1998; : 634-638Google Scholar) were used. 2S. Nakajima and A. Yasui, manuscript in preparation. The cDNA of each gene was introduced into the vector pCY4B, which contains the cytomegalovirus enhancer, chicken β-actin promoter, and rabbit poly(A) signal (27Niwa H. Yamamura K. Miyazaki J. Gene. 1991; 108: 193-199Crossref PubMed Scopus (4667) Google Scholar). CPD-3 cells were further subcloned by single cell dilution. UV Irradiation—For photoreactivation, cells were irradiated at the indicated dose of UV-C using a germicidal lamp (predominantly 254 nm). After UV-C exposure, cells were incubated in Dulbecco's modified Eagle's medium without phenol red and exposed to photoreactivating 360-nm UV-A light from two bulbs (Sankyo Denki, F15T8BLB 15W; PGC Scientific) from above. To block shorter wavelengths, the black lights were filtered through one 5-mm borosilicate glass plate. The distance from the bulbs to the cells was ∼15 cm. For local UV irradiation, cells were grown overnight on glass coverslips. Prior to irradiation, the media were aspirated, and the cells were washed in PBS. For every experiment using localized irradiation, a 3-μm isopore polycarbonate filter (Millipore) presoaked in PBS was placed over the cells, and the cells were irradiated through the filter with 200 J/m2 of UV-C from a germicidal lamp calibrated to deliver 10 J/m2/s. The membrane was removed, and the cells were photoreactivated for 2 h under UV-A blacklight. Immunofluorescence—Cells were grown as indicated on coverslips in a 35-mm dish, washed in PBS, then fixed by 2% formaldehyde in 0.2% Triton X-100/PBS for 10 min on ice. Cells were washed three times in PBS, and then the DNA was denatured by incubation in 2 n HCl for 5 min at 37 °C. Cells were incubated in 20% fetal bovine serum in washing buffer (0.1% Triton X-100 in PBS) for 30 min at room temperature to block nonspecific binding. Primary and secondary antibodies were made up in 1% bovine serum albumin in washing buffer and incubated for 45 min at room temperature. After each antibody step, cells were washed three times for 5 min in washing buffer. When staining for both CPDs or 6–4PPs and the V5 epitope-tagged proteins, a second blocking step of 5 μg/ml mouse IgG (Sigma) was added for 30 min after the CPD and the goat-anti mouse antibodies had been incubated to block non-specific interactions between them and the V5 antibody. Anti-V5 fluorescein isothiocyanate-conjugated antibody was added after the IgG step and incubated for 45 min at room temperature. Coverslips were mounted in VectaShield with 4,6-diamidino-2-phenylindole (DAPI; Vector Laboratories). Images were captured by a Nikon Eclipse E800 microscope (with ×60 oil objective) using an RT Slider CCD camera (Spot Diagnostic), analyzed by Spot RT 3.0 software (Spot Diagnostic), and further adjusted in Adobe PhotoShop 6.0. Characterization of Photolyase-expressing Cell Lines—XP-A cell lines were established that stably express either the CPD photolyase from the rat kangaroo Potorous tridactylis or the 6–4PP photolyase from Arabidopsis thaliana. The advantage of using XP-A cells completely deficient in repair of both CPDs and 6–4PPs (28Lommel L. Hanawalt P.C. Mol. Cell. Biol. 1993; 13: 970-976Crossref PubMed Scopus (49) Google Scholar, 29Mitchell D.L. Haipek C.A. Clarkson J.M. Mutat. Res. 1985; 143: 109-112Crossref PubMed Scopus (246) Google Scholar) is that any removal of a photoproduct occurs solely through the action of the photolyase and not through the NER pathway. Fig. 1A is an immunoblot demonstrating the specific expression of each photolyase in XP-A cells; CPD-3 cells express the CPD photolyase, and 6–4 cells express the 6–4PP photolyase. We examined the kinetics of photoreactivation by first irradiating the photolyase-expressing cells with 20 J/m2 of UV-C and then exposing the cells for varying times to the reactivating energy found in UV-A light. Substantial photoreactivation was observed with one hour of UV-A treatment as measured using monoclonal antibodies to the specific photoproducts (Fig. 1B). By two hours, ∼83% of CPDs had been photoreversed from CPD-3 cells. Increasing lengths of UV-A treatment did not cause significantly more reversal of CPDs (data not shown). Repair of 6–4PPs was much more efficient in the 6–4 cells, with 98% of the 6–4PPs reversed following 2 h of UV-A treatment (Fig. 1B). It is not clear why 6–4PPs were repaired so much more efficiently, but one factor may be that 6–4PPs typically occur outside of chromatinized regions and so may be more accessible to the enzyme (30Smerdon M.J. Conconi A. Prog. Nucleic Acids Res. Mol. Biol. 1999; 62: 227-255Crossref PubMed Scopus (119) Google Scholar). Previous work (31Kosmoski J.V. Smerdon M.J. Biochemistry. 1999; 38: 9485-9494Crossref PubMed Scopus (48) Google Scholar, 32Schieferstein U. Thoma F. EMBO J. 1998; 17: 306Crossref PubMed Scopus (45) Google Scholar) has demonstrated that nucleosomes inhibit the activity of photolyase on CPDs, suggesting that the residual CPDs seen in the UV-A-treated CPD-3 cells may reside within nucleosomal structures. Exposure of the cells to UV-A light did not induce more CPDs or 6–4PPs, as indicated by the lack of change in the band intensities of the vector-only expressing cells at the indicated time points. Notably, 6–4PPs also diminished in the CPD-3 cells after two hours of UV-A treatment to ∼30% of original levels, possibly because of some cross-reactivity between the CPD photolyase and the 6–4PP. Localization of p48 and XPC to DNA Damage after Photoreactivation of CPDs or 6–4PPs—Having established the characteristics of each photolyase clone, we utilized these cells to determine the binding properties of p48 or XPC to the individual photoproducts in vivo. For the study of each protein, we transiently transfected the photolyase-expressing XP-A cells with either a DDB2 or XPC cDNA that had an additional V5 epitope tag to aid in detection and then used indirect immunofluorescence to visualize both the protein and photoproduct of interest. Irradiation was performed using the localized introduction of lesions through a micropore polycarbonate filter and colocalization of nuclear proteins to sites of DNA damage determined using specific monoclonal antibodies to CPDs and 6–4PPs (23Mori T. Nakane M. Hattori T. Matsunaga T. Ihara M. Nikaido O. Photochem. Photobiol. 1991; 54: 225-232Crossref PubMed Scopus (395) Google Scholar). Photoreactivation under UV-A light was carried out for 2 h because this was sufficient to remove nearly all of the 6–4PPs in the 6–4 photolyase-expressing cells and the majority of CPDs in the CPD-3 photolyase-expressing cells (Fig. 1B). Fig. 2 shows representative images of p48 and XPC binding to lesions induced by 200 J/m2 UV-C through a 3-μm polycarbonate filter and then subjected to 2 h of UV-A reactivating light. We have previously used the local UV irradiation assay to demonstrate that a dose of 200 J/m2 through a 3-μm filter was able to activate the tumor suppressor protein p53 and that this dose induced irradiated sites that efficiently bound both p48 and XPC (16Fitch M.E. Cross I.V. Ford J.M. Carcinogenesis. 2003; 24: 843-850Crossref PubMed Scopus (67) Google Scholar). In CPD-3 cells, where the majority of CPDs have been removed by the CPD-specific photolyase, p48 clearly localized to sites that contain 6–4PPs (Fig. 2A). This localization pattern was observed in 100 of 105 cells examined visually (95.2%). One caveat to this finding is that because of the small number of photolyase-resistant CPDs remaining in these irradiated sites, we have not definitively determined if p48 bound only to 6–4PPs or if the residual CPDs are the cause of the binding pattern seen. In 6–4 cells that express the 6–4PP-specific photolyase, p48 localized to sites that contained CPDs, as shown in Fig. 2B. This colocalization pattern was observed in 88 of 112 cells examined visually (78.5%). The binding of p48 to the irradiated sites in the 6–4 cells after 2 h of reactivation appears to be specific to CPDs because the vast majority of 6–4PPs had been removed by the photolyase at this time point (Fig. 1B). There were not as many cells with the specific colocalization pattern of p48 and CPDs in 6–4 cells as there were for p48 and 6–4PPs in the CPD-3 cells (78.5 versus 95.2%), suggesting that CPDs may not be as strong a recognition signal as the 6–4PP for p48 in vivo. This would be consistent with the in vitro findings that UV-DDB has a higher affinity for 6–4PPs over CPDs (6Batty D. Rapic'-Otrin V. Levine A.S. Wood R.D. J. Mol. Biol. 2000; 300: 275-290Crossref PubMed Scopus (186) Google Scholar, 9Keeney S. Chang G.J. Linn S. J. Biol. Chem. 1993; 268: 21293-21300Abstract Full Text PDF PubMed Google Scholar). The lower percentage of localization observed in the 6–4 photolyase-expressing cells also suggests that the high percentage of colocalization of p48 to irradiated sites observed in the CPD-3 cells was indeed due to p48 recognizing the 6–4PPs and not only the residual CPDs. The XPC binding pattern was also determined in the photolyase-expressing XP-A cells. Fig. 2C is a representative image of XPC binding to irradiated sites that contained 6–4PPs in CPD-3 cells 2 h after UV-C irradiation and reactivation with UV-A. The colocalization of XPC to 6–4PP sites was observed in 99 of 108 cells (91.7%), suggesting that XPC is proficient at recognizing the 6–4PP in vivo. In contrast, 2 h after reactivation XPC did not colocalize to irradiated sites that contained only CPDs in 6–4 cells, as seen in Fig. 2D. Only 9 of 110 cells examined (8.2%) demonstrated an appreciable amount of colocalization of XPC to CPD-containing sites. The majority of 6–4PPs have been removed from these sites. Thus it can be inferred that the remaining CPDs were not able to efficiently recruit enough XPC to these sites to be seen by indirect immunofluorescence. The lack of binding to CPDs in the 6–4 cells further demonstrated that it was the 6–4PPs that were recognized by XPC in the CPD-3 cells and not the residual CPDs that were not removed by the CPD photolyase. This in vivo data is again consistent with in vitro binding studies that demonstrated that XPC has a much greater affinity for the 6–4PP than the CPD (6Batty D. Rapic'-Otrin V. Levine A.S. Wood R.D. J. Mol. Biol. 2000; 300: 275-290Crossref PubMed Scopus (186) Google Scholar, 7Hey T. Lipps G. Sugasawa K. Iwai S. Hanaoka F. Krauss G. Biochemistry. 2002; : 6583-6587Crossref PubMed Scopus (94) Google Scholar, 8Kusumoto R. Masutani C. Sugasawa K. Iwai S. Araki M. Uchida A. Mizukoshi T. Hanaoka F. Mutat. Res. 2001; 485: 219-227Crossref PubMed Scopus (104) Google Scholar). This is the first in vivo demonstration that p48 is a more efficient recognition factor for CPDs than XPC (78.5% positive colocalization verses 8.2%). p48 Activates XPC Binding to CPD-containing Sites—The different binding patterns observed between p48 and XPC led to the hypothesis that p48 may stimulate XPC binding to CPDs in vivo, thereby explaining why XP-E cells mutant for DDB2 are predominantly deficient in the GGR of CPDs, more than 6–4PPs (4Hwang B.J. Ford J.M. Hanawalt P.C. Chu G. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 424-428Crossref PubMed Scopus (509) Google Scholar). We have previously observed that p48 can stimulate XPC binding to UV-irradiated sites that contained both CPDs and 6–4PPs (16Fitch M.E. Cross I.V. Ford J.M. Carcinogenesis. 2003; 24: 843-850Crossref PubMed Scopus (67) Google Scholar). Fig. 3 is a representative image of 6–4 cells that were transfected with both XPC and p48, irradiated with 200 J/m2 UV-C, and then photoreactivated with 2 h of UV-A to remove the 6–4PPs. In comparison to Fig. 2D where no detectable XPC binding in 6–4 cells was seen after photoreactivation, XPC strongly colocalized to irradiated sites that contain only CPDs in cells that overexpressed p48 (Fig. 3A). This pattern was observed in 100 of 115 cells examined (87.0%). This clearly demonstrates that p48 stimulates XPC binding to CPDs in vivo and provides the mechanism for the previously observed enhancement of XPC binding by p48. One caveat to our transfection studies is that these conditions are not the same as the photoreversal conditions used to determine the efficiency of the 6–4 photolyase in Fig. 1 because we have overexpressed ectopic p48 and XPC, which potentially could interfere with the action of the 6–4 photolyase. We therefore also stained for 6–4PPs in the transfected 6–4 cells after2hof photoreactivating UV-A in the presence of overexpressed p48 and XPC. Fig. 3B shows two representative cells from a transfection experiment with both XPC and p48 transgenes as in Fig. 3A, only now stained for 6–4PPs and XPC. XPC can be seen localizing to areas that are presumably irradiated sites that contain CPDs, but there are no discernable 6–4PPs and the XPC-expressing cell has the same background staining pattern as the cell below it, which is not expressing the transgenic XPC and p48. These results were verified in whole cell irradiation experiments (20 J/m2, 2-h photoreactivation); again we did not observe any residual 6–4PPs in cells that had overexpressed p48 or XPC alone or in combination over cells that did not express the transgenes. In conclusion, we do not believe that p48 or XPC interferes with the action of the 6–4 photolyase and that the results we observe are due to p48 activating the binding of XPC to CPDs. Table I summarizes the colocalization data of p48 and XPC in the photolyase-expressing cells.Table IColocalization data of p48 and XPC in photolyase-expressing cellsCell lineTransfected cDNAPercent remaining CPDsPercent remaining 6-4PPsNo. of transfected cells countedNo. of cells with colocalizationPercentCPD-3DDB21730105100ap48 colocalization.95.2CPD-3XPC173010899bXPC colocalization.91.76-4DDB298211288ap48 colocalization.78.66-4XPC9821109bXPC colocalization.8.26-4XPC + DDB2982115100bXPC colocalization.87.0a p48 colocalization.b XPC colocalization. Open table in a new tab We have used XP-A cells that express specific DNA photolyases to define the in vivo binding properties of p48 and XPC. p48 bound to both lesion types, with a slight preference for the 6–4PP. XPC, in contrast, showed a very strong binding preference for the 6–4PP over the CPD. Our data corroborate in vitro studies of XPC and p48 binding preferences (6Batty D. Rapic'-Otrin V. Levine A.S. Wood R.D. J. Mol. Biol. 2000; 300: 275-290Crossref PubMed Scopus (186) Google Scholar, 7Hey T. Lipps G. Sugasawa K. Iwai S. Hanaoka F. Krauss G. Biochemistry. 2002; : 6583-6587Crossref PubMed Scopus (94) Google Scholar, 8Kusumoto R. Masutani C. Sugasawa K. Iwai S. Araki M. Uchida A. Mizukoshi T. Hanaoka F. Mutat. Res. 2001; 485: 219-227Crossref PubMed Scopus (104) Google Scholar, 9Keeney S. Chang G.J. Linn S. J. Biol. Chem. 1993; 268: 21293-21300Abstract Full Text PDF PubMed Google Scholar, 10Fujiwara Y. Masutani C. Mizukoshi T. Kondo J. Hanaoka F. Iwai S. J. Biol. Chem. 1999; 274: 20027-20033Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar). Overexpression of p48 was able to dramatically alter the binding properties of XPC so that significant colocalization of XPC was observed to CPD-only-containing sites. Loss of p48 in XP-E cells causes a decrease in repair of CPDs but does not greatly affect the repair of 6–4PPs (4Hwang B.J. Ford J.M. Hanawalt P.C. Chu G. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 424-428Crossref PubMed Scopus (509) Google Scholar). The ability of XPC to readily recognize 6–4PPs but not CPDs in vivo correlates with the phenotype of XP-E cells and demonstrates in vivo that XPC can function to recognize 6–4PPs. When p48 is present at higher levels, as with overexpression in our study or through upregulation by p53 both at basal levels and after DNA damage (4Hwang B.J. Ford J.M. Hanawalt P.C. Chu G. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 424-428Crossref PubMed Scopus (509) Google Scholar), p48 is able to recruit XPC to CPDs and therefore stimulate repair of these lesions. This also explains recent observations by Wang et al. (33Wang Q. Zhu Q. Wani M.A. Wani G. Chen J. Wani A.A. DNA Repair (Amst). 2003; 2: 483-499Crossref PubMed Scopus (83) Google Scholar) describing a p53-dependent effect on binding of XPC and TFIIH to UV-irradiated sites, where they speculated that regulation of a p53-dependent gene may affect the recruitment of XPC to irradiated sites, and specifically to CPDs. This is the first demonstration in vivo of the mechanism by which p48 stimulates DNA repair through the NER pathway. There is some repair of CPDs in XP-E cells, and this may occur through the recognition of CPDs by XPC when other chromatin remodeling processes occur, such as those related to transcription or DNA replication. The in vitro binding studies of XPC to CPDs would suggest that there is some affinity between the two, yet the fact that we did not observe significant binding of XPC to CPDs when p48 was not overexpressed suggests that the amount of XPC binding at any one time was not enough to allow detection by our technique. Although p48 clearly enhances the binding of XPC to CPDs, we do not know the mechanism for this stimulation. p48 has recently been shown to interact with the COP9 signalosome (34Groisman R. Polanowska J. Kuraoka I. Sawada J. Saijo M. Drapkin R. Kisselev A.F. Tanaka K. Nakatani Y. Cell. 2003; 113: 357-367Abstract Full Text Full Text PDF PubMed Scopus (602) Google Scholar), a complex that has ubiquitin ligase activity. This would suggest a mechanism whereby p48 mediates ubiquitin ligation of substrates around CPDs, possibly including histones, which may lead to nucleosome unfolding and thereby allow access of XPC and the remaining components of the NER machinery to CPDs. p48 is itself degraded rapidly after UV irradiation via the ubiquitin-mediated proteasome (24Fitch M.E. Cross I.V. Turner S.J. Adimoolam S. Lin C.X. Williams K.G. Ford J.M. DNA Repair (Amst.). 2003; : 819-826Crossref PubMed Scopus (84) Google Scholar, 35Rapic-Otrin V. McLenigan M.P. Bisi D.C. Gonzalez M. Levine A.S. Nucleic Acids Res. 2002; 30: 2588-2598Crossref PubMed Scopus (143) Google Scholar); XPC stability is also regulated through the proteasome (36Ng J.M.Y. Vermeulen W. van der Horst G.T.J. Bergink S. Sugasawa K. Vrieling H. Hoeijmakers J.H.J. Genes Dev. 2003; : 260003Google Scholar). Clearly, regulation of ubiquitin pathways is an important feature of NER and an exciting new avenue to explore in the understanding of NER.
Referência(s)