The p53 tumor suppressor protein is important for many cellular responses to DNA damage in mammalian cells, but its role in regulating DNA repair in human keratinocytes is undefined. We compared the nucleotide excision repair (NER) response of human fibroblasts and keratinocytes deficient in p53. Fibroblasts expressing human papillomavirus 16 E6 oncoprotein had impaired repair of UV radiation–induced cyclobutane pyrimidine dimers in association with reduced levels of p53 and XPC, which is involved in DNA damage recognition. In contrast, keratinocytes expressing E6 alone or concurrently with the E7 oncoprotein, while possessing reduced levels of p53 but normal levels of XPC, continued to repair pyrimidine dimers as efficiently as control cells with normal p53 levels. Despite preservation of DNA repair, E6 and E6/E7 keratinocytes were hypersensitive to UV radiation. E6 fibroblasts exhibited markedly reduced basal and induced levels of mRNA encoding DDB2, another protein implicated in early events in global NER. In contrast, E6 or E6/E7 keratinocytes possessed basal DDB2 mRNA levels that were not significantly altered relative to control cells, although little induction occurred following UV radiation. Intact global NER was also confirmed in SCC25 cells possessing inactivating mutations in p53 as well as in cells treated with pifithrin-α, a chemical inhibitor of p53 that decreased sensitivity of cells to UV radiation. Collectively, these results indicate that human keratinocytes, unlike fibroblasts, do not require p53 to maintain basal global NER activity, but p53 may still be important in mediating inducible responses following DNA damage.

The major UV radiation–induced DNA photoproducts, cyclobutane pyrimidine dimers (CPD) and pyrimidine(6-4)pyrimidone photoproducts (6-4 PP), are removed from the human genome by nucleotide excision repair (NER). The central importance of NER in protecting the genome is exemplified by the disease, xeroderma pigmentosum, in which impaired NER by mutations in any one of seven NER proteins (XPA through XPG) is typically associated with an increased risk of skin cancer (1). NER is initiated following recognition of damaged DNA by two distinct mechanisms: In global genomic repair (GGR), the heterodimeric proteins, XPC/hHR23B and DDB1/DDB2, are capable of binding to DNA lesions throughout the genome. Although DDB1/DDB2 is not essential for NER in cell-free systems (2, 3), growing evidence indicates that this protein complex is important for damage recognition in cells (48). In contrast, transcription-coupled repair (TCR) is initiated by RNA polymerase II stalled at a lesion on the transcribed strand, thereby resulting in preferential repair of lesions from the transcribed strand of active genes (9).

In addition to its well-known roles in cell cycle arrest and apoptosis, the tumor suppressor, p53, can be a critical regulator of NER. Evidence has accumulated that p53 is essential for both basal and induced GGR responses to certain types of DNA lesions, including CPD (1014). The human papillomavirus type 16 (HPV16) E6 oncoprotein, which leads to ubiquitin-mediated degradation of p53 and inhibits p53 acetylation necessary to transactivate genes in chromatin (15), is associated with loss of GGR, further supporting a role for p53 in maintaining basal GGR activity (13, 1619). In the nonkeratinocyte cell types that have been examined, the role of p53 in regulating GGR seems to be mediated in part by its ability to transactivate the expression of the genes encoding the NER proteins, DDB2 and XPC (4, 20).

Although keratinocytes derived from p53−/− mice have also been reported to have defective removal of both 6-4 PP and CPD (21, 22), the role of p53 in human keratinocytes is unclear. Human oral keratinocytes infected with whole HPV16 exhibited reduced TCR and GGR, but not in association with reduced basal p53 levels, and the defects in NER were not attributable to the expression of E6 or to the lack of p53 (23). In addition, the clinical phenotype of the cancer-prone Li-Fraumeni syndrome (LFS) suggests that p53 may not play the same role in NER in human keratinocytes that it does in other cell types. Individuals with LFS most frequently possess germ-line mutations in one copy of the p53 gene, and loss of heterozygosity in cells predisposes these individuals to a variety of malignancies (24). LFS-derived fibroblasts with mutations in both alleles of p53 exhibit impaired GGR but enhanced resistance to UV radiation–induced apoptosis that could predispose cells to accumulate mutagenic lesions and yet survive (25). Although a variety of malignancies do occur, LFS is not associated with an increased risk of nonmelanoma skin cancers (24), raising the possibility that keratinocytes possess p53-independent mechanisms for responding to genotoxic stress, including regulation of GGR.

Recently, we have observed that differentiating keratinocytes lose p53 expression following the onset of differentiation yet preserve GGR of UV radiation–induced lesions, suggesting that keratinocytes may differ from other cell types in the regulation of GGR (26). To define the role of p53 in NER in human epidermal keratinocytes more systematically, we examined GGR as well as the expression of the DDB2 and XPC genes in both human fibroblasts and keratinocytes made deficient in p53 by several mechanisms. In contrast to dermal fibroblasts in which E6 expression impairs GGR of CPD, keratinocytes made deficient in p53 preserve GGR, and basal levels of XPC and DDB2 gene expression remain intact, although much of the inducible response is lost. The results indicate that human keratinocytes possess p53-independent mechanisms for maintaining GGR and gene expression of the key proteins involved.

Cells. Normal human keratinocytes were isolated from neonatal foreskins using standard techniques (27) and cultured in serum-free medium (medium 154CF with human keratinocyte growth supplement, Cascade Biologics, Portland, OR) containing 0.07 mmol/L Ca2+ and used within two passages. Human dermal fibroblasts derived from neonatal foreskin and infected with amphotropic retroviruses containing vector LXSN with or without the HPV16 E6 gene (a gift from D. Galloway, Fred Hutchinson Cancer Research Center, Seattle, WA) were grown in DMEM supplemented with 10% fetal bovine serum (FBS) and used within 10 passages (28). Human epidermal keratinocytes derived from neonatal foreskin and infected with retroviruses containing vector LXSN with or without the HPV16 E6 (a gift from G. Disbrow and R. Schlegel, Georgetown University Medical Center, Washington, DC) were grown in keratinocyte serum-free medium supplemented with 5 ng/mL epidermal growth factor and 5 μg/mL bovine pituitary extract (Invitrogen, Carlsbad, CA) and used within 10 passages (29). Keratinocytes expressing LXSN retroviral vector containing both HPV16 E6 and E7 genes (CRL-2309, American Type Culture Collection, Manassas, VA) were grown in keratinocyte serum-free medium supplemented with 35 ng/mL epidermal growth factor and 5 μg/mL bovine pituitary extract. SCC25 cells (American Type Culture Collection) were grown in 1:1 Ham's F-12:DMEM supplemented with 10% FBS.

Treatment with pifithrin-α. Pifithrin-α (Sigma, St. Louis, MO) was dissolved in DMSO and added to cells at a final concentration of 20 μmol/L for 1 hour before irradiation and during subsequent incubations. Control cells were treated with an equal volume of DMSO alone.

Irradiation. Cells in 10-cm culture dishes with the lids removed were placed in 1 mL PBS and irradiated from above. The source was a germicidal lamp emitting predominantly at 254 nm with a fluence rate of 170 mW/m2 as measured with a radiometer (IL1400A, International Light, Inc., Newburyport, MA). Following irradiation, cells were either harvested or allowed to incubate in their original conditioned medium.

Survival assay. Cells were irradiated with doses ranging from 0 to 40 J/m2 and grown for 7 days. Cells were then incubated with 0.5 mg/mL thiazolyl blue or 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma) for 4 hours and washed in PBS. DMSO was added for 5 minutes at room temperature to dissolve the precipitate and absorbance was measured at 540 nm (Titertek, Huntsville, AL).

Global genomic repair immunoassay. GGR was assayed as described previously (26). Briefly, cells were prelabeled with [methyl-3H]thymidine (30), washed with PBS, irradiated with 10 J/m2 UV radiation, and either lysed immediately or allowed to repair for up to 24 hours. DNA was prepared as described previously (26, 31). DNA was quantified fluorometrically using Hoechst 33258 (Bio-Rad, Hercules, CA; ref. 32) and damage was probed with an immunoblot assay (11) using monoclonal antibodies against CPD (TDM-2) and 6-4 PP (6-4M-2; gifts from Prof. T. Mori, Nara Medical University, Nara, Japan; ref. 33). The blot was then incubated with secondary antibody conjugated to horseradish peroxidase, visualized with enhanced chemiluminescence (ECL; Amersham, Piscataway, NJ), and quantified using a PhosphorImager (GS-363 Molecular Imager, Bio-Rad). Individual DNA samples were then excised from the blot and subjected to scintillation counting to normalize the chemiluminescent signal for equivalent amounts of unreplicated DNA. Results presented are the average of at least three independent experiments.

Western immunoblotting. Cells were harvested by scraping and gently pelleted by centrifugation. Protein was extracted by incubation with lysis buffer and quantified, and extracts containing equivalent amounts of total protein were separated by SDS-PAGE and electroeluted onto nitrocellulose membranes as described previously (26). Membrane blots were incubated with primary antibodies for 1 hour followed by secondary antibodies conjugated to horseradish peroxidase for 1 hour and detected by ECL. Monoclonal antibodies were obtained commercially for p53 (DO-1, 1:500, Santa Cruz Biotechnology, Santa Cruz, CA), XPC (3.26, 1:1,000, GeneTex, San Antonio, TX), and α-tubulin (B-5-1-2, 1:10,000, Sigma).

Quantitative real-time reverse transcription-PCR. mRNA levels were assayed as described previously (26). Briefly, total cellular RNA was prepared using RNeasy spin columns (Qiagen, Union City, CA). Reverse transcription was done using a TaqMan kit (Applied Biosystems, Foster City, CA) following the manufacturer's instructions. Quantitative real-time PCR was done using SYBR Green PCR Master Mix (Applied Biosystems) and 125 nmol/L of each primer in a volume of 20 μL/replicate. The PCR primers (Oligos Etc., Wilsonville, OR) for glyceraldehyde-3-phosphate dehydrogenase (GAPDH), peptidyl prolyl isomerase (PPIA), XPA, XPC, and DDB2 have been described previously (26). To permit quantitative comparison of levels for each mRNA, a set of total keratinocyte or fibroblast RNA ranging from 0 to 200 ng was simultaneously reverse transcribed and amplified using each primer set for every experiment to generate a standard curve used to perform relative quantitation of the experimental samples. Amplification using an ABI 7900HT cycler (Applied Biosystems) occurred as follows: 50.0°C for 2 minutes, 95°C for 10 minutes and then 40 cycles of 95°C for 15 seconds and 60°C for 1 minute.

Human keratinocytes expressing human papillomavirus 16 E6 are hypersensitive to UV radiation. p53 plays a central role in UV radiation–mediated apoptosis and its loss is expected to affect cell survival following genotoxic stress. Because HPV16 E6 is known to mediate p53 degradation, we determined the consequences of HPV16 E6 expression in keratinocytes on survival following UV radiation. Keratinocytes with empty control vector LXSN or expressing either E6 or both E6 and E7 were exposed to increasing doses of UV radiation and assayed for survival 1 week later by measuring the ability of cells to metabolize thiazolyl blue. As shown in Fig. 1, keratinocytes expressing E6 were hypersensitive to UV radiation relative to LXSN controls. The expression of E7 in addition to E6 did not sensitize keratinocytes to UV radiation more than expression of E6 alone. Thus, although expression of E6 would be expected to reduce p53 (see below) and possibly reduce p53-mediated apoptosis, keratinocytes expressing E6 exhibited reduced survival in response to UV radiation.

Figure 1.

UV sensitivity in keratinocytes infected with LXSN (•) or with vectors expressing E6 (▪) or E6 and E7 (▴). Cells were irradiated with 0 to 40 J/m2 of UV radiation and allowed to grow for 7 days before analysis of metabolic activity by the thiazolyl blue (MTT) assay. Activity was normalized to unirradiated cells for each cell type. Points, average of triplicate measurements for each cell type.

Figure 1.

UV sensitivity in keratinocytes infected with LXSN (•) or with vectors expressing E6 (▪) or E6 and E7 (▴). Cells were irradiated with 0 to 40 J/m2 of UV radiation and allowed to grow for 7 days before analysis of metabolic activity by the thiazolyl blue (MTT) assay. Activity was normalized to unirradiated cells for each cell type. Points, average of triplicate measurements for each cell type.

Close modal

Human papillomavirus 16 E6 oncoprotein impairs global genomic repair in fibroblasts but not in keratinocytes. To further define the response of keratinocytes expressing HPV16 E6, GGR in E6 keratinocytes was compared with E6 fibroblasts relative to the LXSN vector control using an immunoblot assay with monoclonal antibodies directed against either 6-4 PP or CPD (Fig. 2). Following UV radiation, control LXSN fibroblasts removed nearly all 6-4 PP by 2 hours and ∼60% of CPD by 24 hours (Fig. 2A and B). Fibroblasts expressing HPV16 E6, however, were less efficient in GGR of CPD by 8 hours and removed only 25% of CPD by 24 hours. GGR of 6-4 PP was unaffected by E6 expression in fibroblasts. These results corroborate that fibroblasts made deficient in p53 due to HPV16 E6 preferentially lose GGR of CPD (16, 18).

Figure 2.

GGR in fibroblasts and keratinocytes infected with either LXSN or vectors expressing E6 or E6/E7. Fibroblasts (A and B) or keratinocytes (C and D) treated with LXSN vector (•) or expressing E6 (▪) or E6 and E7 (▴) were irradiated with 10 J/m2 UV radiation and allowed to repair up to 24 hours. Cells were harvested and DNA was analyzed by an immunoblot assay using monoclonal antibodies to either 6-4 PP (A and C) or CPD (B and D). Points, averages of two or three independent biological experiments.

Figure 2.

GGR in fibroblasts and keratinocytes infected with either LXSN or vectors expressing E6 or E6/E7. Fibroblasts (A and B) or keratinocytes (C and D) treated with LXSN vector (•) or expressing E6 (▪) or E6 and E7 (▴) were irradiated with 10 J/m2 UV radiation and allowed to repair up to 24 hours. Cells were harvested and DNA was analyzed by an immunoblot assay using monoclonal antibodies to either 6-4 PP (A and C) or CPD (B and D). Points, averages of two or three independent biological experiments.

Close modal

In contrast, GGR in keratinocytes was much less sensitive to expression of HPV16 E6 (Fig. 2C and D). Control LXSN keratinocytes exhibited GGR of 6-4 PP and CPD that was similar in kinetics and extent to that observed in fibroblasts. E6 keratinocytes were also not significantly different from LXSN cells in their repair of 6-4 PP at all times. Although E6 keratinocytes were slightly slower than control cells in repair of CPD from 0 to 8 hours, they were indistinguishable from control LXSN keratinocytes by 24 hours. Similarly, keratinocytes simultaneously expressing both HPV16 E6 and E7 oncoproteins removed 6-4 PP and CPD as efficiently as control LXSN cells at all times, suggesting that neither E6 nor E7 affects GGR in keratinocytes.

Expression of p53 and XPC in cells expressing human papillomavirus 16 E6 oncoprotein. To ensure that the GGR phenotype observed in the fibroblasts and keratinocytes was actually occurring in the p53 backgrounds expected for LXSN and E6 cells, we assayed the p53 expression of the LXSN, E6, and E6/E7 cells. Following exposure to 10 J/m2 UV radiation, both fibroblasts and keratinocytes that expressed only the control LXSN vector exhibited detectable levels of p53 at baseline as well as a several-fold induction by 24 hours (Fig. 3) comparable with those observed in normal fibroblasts and keratinocytes (data not shown). Both fibroblasts and keratinocytes expressing HPV16 E6 possessed basal levels of p53 that were markedly reduced, and although there was some induction following UV radiation, these induced levels at 24 hours were similar to or below basal levels in control LXSN cells. Keratinocytes that expressed both HPV16 E6 and E7 exhibited dramatically reduced levels of p53 that were detectable only on lengthy exposures of the Western blot to film (data not shown). These results confirmed that expression of HPV16 E6 results in substantially reduced p53 levels in both fibroblasts and keratinocytes.

Figure 3.

Expression of p53 and XPC. A, fibroblasts infected with either the LXSN retroviral vector alone or with vector expressing HPV16 E6. B, keratinocytes infected with either LXSN alone or with vectors expressing HPV16 E6 or both E6 and E7. Cells were irradiated with 10 J/m2 UV radiation and harvested immediately afterward or at 24 hours. Equivalent amounts of total protein were analyzed by Western blotting for p53, XPC or tubulin (used as a loading control). Numbers below bands, quantities relative to the initial amount of the particular protein.

Figure 3.

Expression of p53 and XPC. A, fibroblasts infected with either the LXSN retroviral vector alone or with vector expressing HPV16 E6. B, keratinocytes infected with either LXSN alone or with vectors expressing HPV16 E6 or both E6 and E7. Cells were irradiated with 10 J/m2 UV radiation and harvested immediately afterward or at 24 hours. Equivalent amounts of total protein were analyzed by Western blotting for p53, XPC or tubulin (used as a loading control). Numbers below bands, quantities relative to the initial amount of the particular protein.

Close modal

Because XPC has been reported to be under transcriptional control of p53, XPC levels were also examined (20). Basal levels of XPC protein seemed to be slightly reduced in E6 fibroblasts relative to LXSN fibroblasts, but XPC was still induced following UV radiation. In contrast, keratinocytes had slightly elevated basal levels of XPC protein in the presence of E6 or both E6 and E7. However, whereas LXSN keratinocytes displayed modest induction of XPC protein following UV radiation, neither E6 nor E6/E7 keratinocytes induced XPC.

mRNA encoding nucleotide excision repair proteins in human papillomavirus 16 E6 fibroblasts and keratinocytes. Because p53 has been reported to regulate GGR by transactivating certain downstream genes, such as DDB2 and XPC (4, 20), we examined levels of DDB2 and XPC mRNA in fibroblasts and keratinocytes using quantitative reverse transcription-PCR (RT-PCR; Fig. 4). mRNA encoding XPA, which has not been reported to be induced in cells following UV radiation, as well as GAPDH and PPIA, served as loading controls and did not vary significantly in LXSN and E6 fibroblasts or in LXSN and E6 keratinocytes following UV radiation. In LXSN fibroblasts (Fig. 4A), both XPC and DDB2 mRNA were induced by 24 hours following UV radiation. XPC mRNA levels were sustained, whereas DDB2 levels returned to near-basal levels by 48 hours. E6 fibroblasts had normal basal levels but diminished induction of XPC mRNA at all times. Basal levels of DDB2 mRNA were reduced 5-fold in E6 fibroblasts (P < 0.01), and although induction of this mRNA did occur, the induced levels at 24 and 48 hours were still relatively low and did not exceed the basal levels found in control LXSN cells.

Figure 4.

Levels of mRNA encoding NER proteins following UV radiation. Fibroblasts expressing LXSN or E6 (A) and keratinocytes expressing LXSN, E6, or E6 and E7 (B) were irradiated with 10 J/m2 UV radiation and harvested for total RNA over 48 hours. mRNA was reverse transcribed and the products were amplified and detected using quantitative real-time PCR to measure relative levels of XPA, XPC, and DDB2 as well as GAPDH and PPIA. Results have been normalized to the initial level of a particular mRNA in the LXSN-expressing cells. Columns, averages of two or three independent biological experiments with the exception for data from E6/E7 keratinocytes, which represent samples from a single biological experiment assayed in triplicate.

Figure 4.

Levels of mRNA encoding NER proteins following UV radiation. Fibroblasts expressing LXSN or E6 (A) and keratinocytes expressing LXSN, E6, or E6 and E7 (B) were irradiated with 10 J/m2 UV radiation and harvested for total RNA over 48 hours. mRNA was reverse transcribed and the products were amplified and detected using quantitative real-time PCR to measure relative levels of XPA, XPC, and DDB2 as well as GAPDH and PPIA. Results have been normalized to the initial level of a particular mRNA in the LXSN-expressing cells. Columns, averages of two or three independent biological experiments with the exception for data from E6/E7 keratinocytes, which represent samples from a single biological experiment assayed in triplicate.

Close modal

LXSN keratinocytes behaved similarly to LXSN fibroblasts, exhibiting XPC and DDB2 mRNA induction following UV radiation within 24 hours (Fig. 4B). Similar to E6 fibroblasts, E6 keratinocytes also had normal basal levels and diminished induction of XPC mRNA at 24 hours, although by 48 hours induced levels were similar to LXSN cells. In contrast to E6 fibroblasts, however, E6 keratinocytes possessed basal levels of DDB2 mRNA that were not significantly different from those observed in LXSN cells, although there was also no significant induction of this mRNA following UV radiation. In E6/E7 keratinocytes, as with E6 keratinocytes, induction of XPC mRNA was below levels observed in LXSN cells; although, unlike E6 keratinocytes, E6/E7 keratinocytes also exhibited diminished basal XPC mRNA levels. As in E6 keratinocytes, basal DDB2 mRNA levels were unaffected by HPV16 E6 and E7, and levels were minimally induced following UV radiation. E6/E7 keratinocytes also displayed a modest induction of XPA mRNA following UV radiation that was not seen in LXSN or E6 keratinocytes.

Global genomic repair in other types of keratinocytes deficient in p53. Because the HPV E6 and E7 oncoproteins have several effects on keratinocytes (34), it was desirable to examine GGR activity in keratinocytes that were depleted of p53 function by independent mechanisms to ensure that preservation of GGR in E6-expressing keratinocytes was not due to other effects of E6. First, we examined SCC25 cells, derived from an oral squamous cell carcinoma, which possess mutations in p53 that drastically reduce the level of p53 mRNA and protein (35). We confirmed that, in contrast to normal human epidermal keratinocytes, SCC25 cells did not possess detectable levels of p53 at baseline or by 24 hours following UV radiation, and we observed that although XPC levels were near normal at baseline they actually decreased slightly by 24 hours following UV radiation (data not shown). Nevertheless, SCC25 cells were as proficient as normal human keratinocytes in removing 6-4 PP and CPD by 24 hours (data not shown).

Second, to further confirm that p53 is not essential to GGR, normal human keratinocytes were treated with an organic chemical inhibitor of p53, pifithrin-α (36). Relative to diluent-treated controls, application of pifithrin-α increased survival following UV radiation from 65% to 87% at 5 J/m2, 55% to 83% at 10 J/m2, and 48% to 89% at 15 J/m2, consistent with its reported inhibition of p53-mediated apoptosis (36). However, pifithrin-α had no detectable effect on GGR of either 6-4 PP or CPD at 8 or 24 hours relative to cells that were treated with diluent alone (data not shown).

Work from several laboratories employing a variety of nonkeratinocyte cells has provided growing evidence that p53 is necessary for normal GGR of CPD and certain other DNA lesions (1113, 16, 17, 19, 25). Corroborating this work, we observed that skin fibroblasts expressing HPV16 E6 and reduced p53 were impaired in GGR of CPD although not of 6-4 PP. Others have observed that E6 expression is also associated with a modest reduction in repair of 6-4 PP, and this difference from our results may reflect individual variations among the specific cell lines employed (16).

In contrast, human keratinocytes depleted of p53 by three independent mechanisms—E6-mediated p53 degradation, genetic mutation in the p53 gene, and chemical inhibition—all repaired both 6-4 PP and CPD over 24 hours as efficiently as control cells expressing normal p53 levels. Keratinocytes expressing E6 with or without E7 had reductions in p53 levels of the same or greater magnitude as those seen in fibroblasts and yet exhibited normal GGR kinetics. SCC25 keratinocytes, derived from a human oral squamous cell carcinoma, harbor a 2-bp deletion mutation in the p53 gene that renders these cells markedly deficient in p53 transcripts as well as in the p53 protein (35). We observed that although these cells have undetectable p53 levels they exhibit normal repair of 6-4 PP as well as normal repair of CPD by 24 hours. Finally, we examined cells treated with pifithrin-α, a reversible chemical inhibitor of p53. Although pifithrin-α suppressed UV radiation–induced cytotoxicity in keratinocytes relative to control cells, as expected for a p53 inhibitor (36), it had no effect on GGR of either 6-4 PP or CPD.

It is unlikely that the differences between our results in keratinocytes and those in nonkeratinocyte cells can be trivially explained by the use of differing media for different types of cells. For example, although the media used to grow keratinocytes and fibroblasts differed significantly, LXSN keratinocytes and LXSN fibroblasts had nearly identical GGR kinetics for both 6-4PP and CPD, as did normal keratinocytes and fibroblasts (data not shown). Further, even among the different keratinocyte cell lines used, significant differences in media existed, such as in the concentration of epidermal growth factor, and yet all keratinocyte lines produced quantitatively similar results when made deficient in p53. Taken together, the results from E6 and E6/E7 keratinocytes, SCC25 cells, and pifithrin-treated keratinocytes indicate that p53 is not essential for maintaining basal GGR activity in human keratinocytes and that the roles of p53 in apoptosis and in GGR are separable in keratinocytes. These results are also consistent with prior work in which we have observed preservation of GGR in differentiating keratinocytes in spite of a significant deficiency of p53 in these cells relative to undifferentiated cells (26).

These results in human keratinocytes differ from the reduced repair of UVB-induced CPD and 6-4 PP that has been reported for murine keratinocytes that were genetically p53 null (21, 22). It is possible that murine keratinocytes regulate GGR differently from human keratinocytes. For example, in comparison to human cells, rodent cells generally exhibit poor GGR that has been attributed to the lack of a functional p53 response element that allows p53-mediated transactivation of the DDB2 gene (5, 37). An additional explanation may be that keratinocytes regulate GGR in a p53-independent manner when irradiated with the 254 nm irradiation used in our study, whereas UVB radiation, used in the prior study of murine keratinocytes (21, 22), may be associated with p53-dependent GGR. Our results also differ from those obtained with human oral keratinocytes that were infected with whole HPV16 virus and that showed a decrease in both TCR and GGR relative to uninfected cells following 2.5 J/m2 254 nm radiation (23). However, in addition to the much lower UV radiation dose employed, the HPV-infected cells used in that prior report actually possessed normal basal levels of p53 and, when treated with a proteasome inhibitor, did not respond with elevated p53 levels as expected if E6-mediated degradation of p53 had occurred. Therefore, that report concluded that loss of NER activity in HPV16-infected cells could not be attributed to E6. It is possible that the observed loss of NER in HPV16-infected cells resulted from whole viral infection disrupting cellular physiology more globally than occurs with expression of the E6 oncoprotein alone.

Although the depletion of p53 might be expected to increase cellular resistance to UV radiation–induced apoptosis normally mediated by p53, both E6 and E6/E7 keratinocytes were hypersensitive to UV radiation relative to normal cells. Similar results have been observed in E6/E7 keratinocytes exposed to UV radiation (38) as well as E6 fibroblasts (16, 28) and E6 MCF-7 breast carcinoma cells (39) that were exposed to UV radiation and other DNA-damaging agents. These results differ from our chemical inhibition of p53 by pifithrin-α as well as reports by others that loss of p53 function is associated with reduced apoptosis after DNA damage (25, 36, 40). It may be that E6 and E7 have additional biological effects beyond p53 that affect survival following DNA damage (34). In any case, regardless of the mechanism by which cells in our study were depleted of p53 and their survival phenotype, it is reassuring that GGR following UV radiation consistently seemed to be largely unaffected.

The E7 oncoprotein leads to inhibition of Rb tumor suppressor function, and its presence along with the E6 protein also did not diminish GGR activity in keratinocytes. Our result that GGR is preserved in E6/E7 keratinocytes is consistent with other work showing that Rb does not play an essential role in GGR of bulky, non-UV radiation–mediated DNA damage (13). Although Rb has been implicated in control of GGR (17), it is possible that our differing results may be due to the differing cell types or UV radiation sources as has been shown for the role of p53 in GGR and TCR (14).

In E6 fibroblasts, loss of p53 was associated with a modest reduction of basal XPC protein levels, although induction of both proteins following UV radiation was still observed, comparable with prior results (20). These results are consistent with a model in which p53 transcriptionally activates key GGR proteins, and both XPC and DDB2 genes possess p53 response elements (20, 37, 41). The small levels of p53 remaining in the presence of E6, while resulting in significant basal down-regulation of the XPC gene, are apparently sufficient to transactivate the gene to some degree following UV radiation. In contrast, both E6 and E6/E7 keratinocytes, while having significant loss of p53, possessed basal levels of XPC protein that were actually slightly greater than those in control cells. However, these cells had little induction of XPC by 24 hours following UV radiation. These basal levels of XPC may be sufficient to mediate the normal GGR kinetics over 24 hours following UV radiation that were observed in keratinocytes, although this time may be too short to detect an effect of abrogation of induced XPC levels on GGR.

To further explore the consequences of p53 deficiency on genes involved in GGR, we examined mRNA levels of XPC and DDB2 in cells expressing HPV16 E6. In E6 fibroblasts, deficient basal and induced DDB2 mRNA levels corroborated results obtained in p53-deficient fibroblasts previously (4, 20). In contrast, both E6 and E6/E7 keratinocytes, in which p53 was significantly reduced, exhibited DDB2 mRNA levels that were not significantly different from those of control LXSN cells. Similar to XPC protein levels, there was modest if any induction of XPC and DDB2 mRNA in E6 and E6/E7 keratinocytes. These results suggest that basal expression of the DDB2 gene is strongly dependent on p53 in fibroblasts but not in keratinocytes. However, induction of both XPC and DDB2 genes to levels seen in LXSN controls seems to depend on p53 expression in all cells that were examined.

It is unclear why changes in mRNA and protein levels for XPC in E6 and E6/E7 cells did not always correlate with each other. Levels of XPC protein seem to be regulated in a complex manner both at the level of synthesis (20) and in proteasomal degradation (42, 43). Keratinocytes expressing E6 and E7 oncoproteins have significant alterations in numerous genes involved in translation as well as in protein degradation (34), and it is possible that expression of E6 and E7 result in increased steady-state levels of XPC even when mRNA levels are low relative to control cells. Collectively, the results from quantitative RT-PCR of XPC and DDB2 mRNA as well as from Western blotting of XPC indicate that basal expression of these genes is dissociated from p53 expression in human keratinocytes. This dissociation may explain why GGR in keratinocytes is largely intact in spite of reduced or absent p53 levels, in contrast to the situation in fibroblasts and other nonkeratinocyte cell types. It is possible these genes and hence GGR activity in keratinocytes are coregulated by proteins, such as the p53 homologues p63 or p73, whose expression is restricted to certain cells, including epidermal keratinocytes, and whose functions have yet to be fully defined (44).

These results also suggest one explanation for the lack of increased incidence of nonmelanoma skin cancers in LFS: Loss of p53 in keratinocytes in LFS individuals does not necessarily result in a defect in GGR of UV radiation–induced DNA lesions in these cells, although its loss in other noncutaneous tissues is associated with a repair deficiency that further predisposes affected individuals to accumulate mutagenic DNA lesions that may eventually result in internal malignancies. We suggest that keratinocytes may be different if not unique in this ability to mediate repair of UV radiation–induced lesions in the absence or reduction of p53.

Grant support: VA Medical Research Service.

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

We thank K. Yeh and K. Carrion for technical assistance, S. Pennypacker for assistance with cell culture, D. Galloway for generously providing fibroblasts expressing E6, G. Disbrow and R. Schlegel for generously providing keratinocytes expressing E6, T. Mori for monoclonal antibodies, P. Hanawalt and J. Ford for helpful discussions, and C. Largman for critical reading of the article. The real-time PCR was done using the Molecular Core Facilities of the San Francisco VA Medical Center and Northern California Institute for Research and Education.

1
Kraemer KH, Lee MM, Scotto J. Xeroderma pigmentosum.
Arch Dermatol
1987
;
123
:
241
–50.
2
Wood RD, Biggerstaff M, Shivji MKK. Detection and measurement of nucleotide excision repair synthesis by mammalian cell extracts in vitro.
Methods Enzymol
1995
;
7
:
163
–75.
3
Reardon JT, Sancar A. Recognition and repair of the cyclobutane thymine dimer, a major cause of skin cancers, by the human excision nuclease.
Genes Dev
2003
;
17
:
2539
–51.
4
Hwang BJ, Ford JM, Hanawalt PC, Chu G. Expression of the p48 xeroderma pigmentosum gene is p53-dependent and is involved in global genomic repair.
Proc Natl Acad Sci U S A
1999
;
96
:
424
–8.
5
Tang JY, Hwang BJ, Ford JM, Hanawalt PC, Chu G. Xeroderma pigmentosum p48 gene enhances global genomic repair and suppresses UV-induced mutagenesis.
Mol Cell
2000
;
5
:
737
–44.
6
Fitch ME, Nakajima S, Yasui A, Ford JM. In vivo recruitment of XPC to UV-induced cyclobutane pyrimidine dimers by the DDB2 gene product.
J Biol Chem
2003
;
278
:
46906
–10.
7
Wang Q, Zhu Q, Wani G, Chen J, Wani AA. UV radiation-induced XPC translocation within chromatin is mediated by damaged-DNA binding protein, DDB2.
Carcinogenesis
2004
;
25
:
1033
–43.
8
Sugasawa K, Okuda Y, Saijo M, et al. UV-induced ubiquitylation of XPC protein mediated by UV-DDB-ubiquitin ligase complex.
Cell
2005
;
121
:
387
–400.
9
Mellon I, Spivak G, Hanawalt PC. Selective removal of transcription-blocking DNA damage from the transcribed strand of the mammalian DHFR gene.
Cell
1987
;
51
:
241
–9.
10
Smith ML, Chen IT, Zhan Q, O'Connor PM, Fornace AJ. Involvement of the p53 tumor suppressor in repair of U.V.-type DNA damage.
Oncogene
1995
;
10
:
1053
–9.
11
Ford JM, Hanawalt PC. Expression of wild-type p53 is required for efficient global genomic nucleotide excision repair in UV-irradiated human fibroblasts.
J Biol Chem
1997
;
272
:
28078
–80.
12
Lloyd DR, Hanawalt PC. p53-dependent global genomic repair of benzo[a]pyrene-7,8-diol-9,10-epoxide adducts in human cells.
Cancer Res
2000
;
60
:
517
–21.
13
Wani MA, El-Mahdy MA, Hamada FM, et al. Efficient repair of bulky anti-BPDE DNA adducts from non-transcribed DNA strand requires functional p53 but not p21(waf1/cip1) and pRb.
Mutat Res
2002
;
515
:
13
–25.
14
Mathonnet G, Leger C, Desnoyers J, Drouin R, Therrien JP, Drobetsky EA. UV wavelength-dependent regulation of transcription-coupled nucleotide excision repair in p53-deficient human cells.
Proc Natl Acad Sci U S A
2003
;
100
:
7219
–24.
15
Thomas MC, Chiang CM. E6 oncoprotein represses p53-dependent gene activation via inhibition of protein acetylation independently of inducing p53 degradation.
Mol Cell
2004
;
17
:
251
–64.
16
Ford JM, Baron EL, Hanawalt PC. Human fibroblasts expressing the human papillomavirus E6 gene are deficient in global genomic nucleotide excision repair and sensitive to ultraviolet irradiation.
Cancer Res
1998
;
58
:
599
–603.
17
Therrien J-P, Drouin R, Baril C, Drobetsky EA. Human cells compromised for p53 function exhibit defective global and transcription-coupled nucleotide excision repair, whereas cells compromised for pRb function are defective only in global repair.
Proc Natl Acad Sci U S A
1999
;
96
:
15038
–43.
18
Zhu Q, Wani MA, El-Mahdy M, Wani AA. Decreased DNA repair efficiency by loss or disruption of p53 function preferentially affects removal of cyclobutane pyrimidine dimers from non-transcribed strand and slow repair sites in transcribed strand.
J Biol Chem
2000
;
275
:
11492
–7.
19
El-Mahdy MA, Hamada FM, Wani MA, Zhu Q, Wani AA. p53-degradation by HPV-16 E6 preferentially affects the removal of cyclobutane pyrimidine dimers from non-transcribed strand and sensitizes mammary epithelial cells to UV-radiation.
Mutat Res
2000
;
459
:
135
–45.
20
Adimoolam S, Ford JM. p53 and DNA damage-inducible expression of the xeroderma pigmentosum group C gene.
Proc Natl Acad Sci U S A
2002
;
99
:
12985
–90.
21
Li G, Mitchell DL, Ho VC, Reed JC, Tron VA. Decreased DNA repair but normal apoptosis in ultraviolet-irradiated skin of p53 transgenic mice.
Am J Pathol
1996
;
148
:
1113
–23.
22
Li G, Ho VC, Mitchell DL, Trotter MJ, Tron VA. Differentiation-dependent p53 regulation of nucleotide excision repair in keratinocytes.
Am J Pathol
1997
;
150
:
1457
–764.
23
Rey O, Lee S, Park NH. Impaired nucleotide excision repair in UV-irradiated human oral keratinocytes immortalized with type 16 human papillomavirus genome.
Oncogene
1999
;
18
:
6997
–7001.
24
Olivier M, Goldgar DE, Sodha N, et al. Li-Fraumeni and related syndromes: correlation between tumor type, family structure and TP53 genotype.
Cancer Res
2003
;
63
:
6643
–50.
25
Ford JM, Hanawalt PC. Li-Fraumeni syndrome fibroblasts homozygous for p53 mutations are deficient in global DNA repair but exhibit normal transcription-coupled repair and enhanced UV resistance.
Proc Natl Acad Sci U S A
1995
;
92
:
88766
–880.
26
Oh DH, Yeh K. Differentiating human keratinocytes are deficient in p53 but retain global nucleotide excision repair following ultraviolet radiation. DNA Repair. Epub 2005 Jul 22.
27
Pillai S, Bikle DD, Hincenbergs M, Elias PM. Biochemical and morphological characterization of growth and differentiation of normal human neonatal keratinocytes in a serum-free medium.
J Cell Physiol
1988
;
134
:
229
–37.
28
Hawkins DS, Demers GW, Galloway DA. Inactivation of p53 enhances sensitivity to multiple chemotherapeutic agents.
Cancer Res
1996
;
56
:
892
–8.
29
Pei XF, Sherman L, Sun Y-L, Schlegel R. HPV16 E7 protein bypasses keratinocyte growth inhibition by serum and calcium.
Carcinogenesis
1998
;
19
:
1481
–6.
30
Spivak G, Hanawalt PC. Determination of damage and repair in specific DNA sequences.
Methods Enzymol
1995
;
7
:
147
–61.
31
Vos JMH. Analysis of psoralen monoadducts and interstrand crosslinks in defined genomic sequences. In: Friedberg EC, Hanawalt PC. DNA repair: a laboratory manual of research procedures. Vol. 3. New York: Marcel Dekker; 1988. pp. 367–98.
32
Sambrook J, Russel DW. Fluorometric quantitation of DNA using Hoechst 33258. In: Molecular cloning: a laboratory manual. 3rd ed. Vol. 3. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 2001. pp. A8.22–3.
33
Mori T, Nakane M, Hattori T, Matsunaga T, Ihara M, Nikaido O. Simultaneous establishment of monoclonal antibodies specific for either cyclobutane pyrimidine dimer or (6-4) photoproduct from the same mouse immunized with ultraviolet-irradiated DNA.
Photochem Photobiol
1991
;
54
:
225
–32.
34
Garner-Hamrick PA, Fostel JM, Chien W-M, et al. Global effects of human papillomavirus type 18 E6/E7 in an organotypic keratinocyte culture system.
J Virol
2004
;
78
:
9041
–50.
35
Min BY, Baek JH, Shin KH, Gujuluva CN, Cherrick HM, Park NH. Inactivation of the p53 gene by either mutation or HPV infection is extremely frequent in human oral squamous cell carcinoma cell lines.
Oral Oncol Eur J Cancer
1994
;
30
:
338
–45.
36
Komarov PG, Komarova EA, Kondratov RV, et al. A chemical inhibitor of p53 that protects mice from the side effects of cancer therapy.
Science
1999
;
285
:
1733
–7.
37
Tan T, Chu G. p53 binds and activates the xeroderma pigmentosum DDB2 gene in humans but not mice.
Mol Cell Biol
2002
;
22
:
3247
–54.
38
Simbulan-Rosenthal CM, Velena A, Veldman T, Schlegel R, Rosenthal DS. HPV-16 E6/7 immortalization sensitizes human keratinocytes to ultraviolet B by altering the pathway from caspase-8 to caspase-9-dependent apoptosis.
J Biol Chem
2002
;
277
:
24709
–16.
39
Fan S, Smith ML, Rivet DJ, et al. Disruption of p53 function sensitizes breast cancer MCF-7 cells to cisplatin and pentoxyfylline.
Cancer Res
1995
;
55
:
1649
–54.
40
Fan S, El-Deiry W, Bae I, et al. p53 gene mutations are associated with decreased sensitivity of human lymphoma cells to DNA damaging agents.
Cancer Res
1994
;
54
:
5824
–30.
41
Itoh T, O'Shea C, Linn S. Impaired regulation of tumor suppressor p53 caused by mutations in the xeroderma pigmentosum DDB2 gene: mutual regulator interactions between p48DDB2 and p53.
Mol Cell Biol
2003
;
23
:
7540
–53.
42
Ng JMY, Vermeulen W, Van der Horst GTJ, et al. A novel regulation mechanism of DNA repair by damage-induced and RAD23-dependent stabilization of xeroderma pigmentosum group C protein.
Genes Dev
2003
;
17
:
1630
–45.
43
Lommel L, Ortolan T, Chen L, Madura K, Sweder KS. Proteolysis of a nucleotide excision repair protein by the 26S proteasome.
Curr Genet
2002
;
42
:
9
–20.
44
Yang A, Kaghad M, Caput D, McKeon F. On the shoulders of giants: p63, p73 and the rise of p53.
Trends Genet
2002
;
18
:
90
–5.