Mismatch repair protein Msh2 contributes to UVB-induced cell cycle arrest in epidermal and cultured mouse keratinocytes

DNA Repair 4 (2005) 81–89

Mismatch repair protein Msh2 contributes to UVB-induced cell cycle arrest in epidermal and cultured mouse keratinocytes Marijke van Oostena,1,2 , Gerdine J Stoutb,1 , Claude Backendorfc , Heggert Rebelb , Niels de Winda , Firouz Darroudia , Henk J van Kranend , Frank R de Gruijlb , Leon HF Mullendersa,∗ a

d

Department of Toxicogenetics, Leiden University Medical Center, Wassenaarseweg 72, 2333 AL Leiden, The Netherlands b Department of Dermatology, Leiden University Medical Center, Leiden, The Netherlands c Laboratory of Molecular Genetics, Leiden Institute of Chemistry, Leiden, The Netherlands Laboratory of Toxicology Pathology and Genetics, National Institute of Public Health and Environment, Bilthoven, The Netherlands Received 10 June 2004; accepted 18 August 2004 Available online 27 September 2004

Abstract Nucleotide excision repair (NER), cell cycle regulation and apoptosis are major defence mechanisms against the carcinogenic effects of UVB radiation. NER eliminates UVB-induced DNA photolesions via two subpathways: global genome repair (GGR) and transcription-coupled repair (TCR). In a previous study, we found UVB-induced accumulation of tetraploid (4N) keratinocytes in the epidermis of Xpc−/− mice (no GGR), but not in Xpa−/− (no TCR and no GGR) or in wild-type (WT) mice. We inferred that this arrest in Xpc−/− mice is caused by erroneous replication past photolesions, leading to ‘compound lesions’ known to be recognised by mismatch repair (MMR). MMR-induced futile cycles of breakage and resynthesis at sites of compound lesions may then sustain a cell cycle arrest. The present experiments with Xpc−/− Msh2−/− mice and derived keratinocytes show that the MMR protein Msh2 indeed plays a role in the generation of the UVB-induced arrested cells: a Msh2-deficiency lowered significantly the percentage of arrested cells in vivo (40–50%) and in vitro (30–40%). Analysis of calyculin A (CA)-induced premature chromosome condensation (PCC) of cultured Xpc−/− keratinocytes showed that the delayed arrest occurred in late S phase rather than in G2 -phase. Taken together, the results indicate that in mouse epidermis and cultured keratinocytes, the MMR protein Msh2 plays a role in the UVB-induced S-phase arrest. This indicates that MMR plays a role in the UVB-induced S-phase arrest. Alternatively, Msh2 may have a more direct signalling function. © 2004 Elsevier B.V. All rights reserved. Keywords: Cell cycle; Mismatch repair; Xeroderma pigmentosum; UV-light; Epidermis

1. Introduction Nucleotide excision repair (NER) is the principal pathway for the elimination of ultraviolet light (UV)-induced photolesions (e.g., cyclobutane pyrimidine dimers (CPD) and ∗

Corresponding author. Tel.: +31 71 5276126; fax: +31 71 5276173. E-mail address: [email protected] (L.H. Mullenders). 1 Both authors contributed equally to the work. 2 Present address: Laboratory for Vaccine-Preventable Diseases, National Institute of Public Health and Environment, Bilthoven, The Netherlands. 1568-7864/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.dnarep.2004.08.008

pyrimidine (6-4) pyrimidone photoproducts (6-4PP)). NER consists of two subpathways: the global genome repair pathway (GGR) and a specialised pathway termed transcriptioncoupled repair (TCR). The latter pathway removes DNA lesions from the transcribed strand of transcriptionally active genes. The importance of NER to counteract UVB-induced toxic and carcinogenic effects in the skin is demonstrated by the human genetic disorder xeroderma pigmentosum (XP). This disorder is clinically characterised by a high sensitivity to sunburn at relatively low levels of exposure [1] and by increased risk of developing skin cancer at sun-exposed

82

M. van Oosten et al. / DNA Repair 4 (2005) 81–89

areas [2]. Cell fusion studies have revealed the existence of seven XP complementation groups (A through G) [3]. Cells from XP complementation groups display an overall NER defect (i.e. impaired GGR and TCR) with the exception of cells from XPC and XPE patients that are GGR defective only [4,5]. To elucidate the role of NER pathways in alleviating UVB-induced acute effects (erythema and edema) and longterm effects (skin cancer), NER-deficient mouse models have proved to be valuable tools. Murine epidermis shows a similar phenotype as their human counterparts with respect to susceptibility to UVB-induced sunburn [6–10] and to UVBinduced non-melanoma skin cancer [6,7,11–13]. Besides NER, cell cycle arrest and apoptosis counteract the detrimental effects of UVB-induced photolesions in the epidermis of the mouse [14,15]. In addition, interactions between the various defence systems have been identified in numerous studies [16–19]. In a former study [18], we showed the impact of deficiencies in GGR and TCR on cell cycle progression and apoptosis of keratinocytes in the epidermis of UVB-exposed mice. UVB-irradiated Xpa−/− mice (defective in TCR and GGR) displayed severe inhibition of S-phase progression at UVB doses that did not affect cell cycle progression in wild-type (WT) mice. In contrast, epidermal cells of UVBexposed Xpc−/− mice (expressing TCR only) were capable to progress through S-phase indistinguishably from keratinocytes of UVB-irradiated WT mice. However, whereas no G2 arrest was observed in WT mice, UVB-irradiated Xpc−/− mice apparently accumulated epidermal cells in the G2 -phase of the cell cycle beginning at 24 h after exposure. We hypothesized that the difference in cell cycle arrest between UVBirradiated WT and Xpc−/− mice is due to error-prone DNA replication past photolesions in the GGR-deficient Xpc−/− mice, producing mismatches opposite photolesions. Mismatches that result from DNA replication errors are primarily eliminated by the mismatch repair (MMR) pathway. There is experimental evidence [20,21] that compound lesions are recognised by the MMR system, giving rise to futile cycles of breakage and resynthesis as described for compound lesions harbouring alkylating damage [22]. This process may induce cell cycle arrest and apoptosis. In WT mice, UVB radiation induces expression of Msh2 in the epidermis, consistent with the recognition of UVB-induced lesions [23]. In this study, we addressed the question whether MMR plays a role in effectuating and/or maintaining the UVBinduced cell cycle arrest and in preventing apoptosis in Xpc−/− keratinocytes in vivo and in vitro. For this purpose, we generated Xpc−/− Msh2−/− double-knockout (hairless) mouse models and derived keratinocyte cell cultures from these mice. Our results demonstrate that the MMR protein Msh2 reduces the percentage of arrested cells by half. Although other factors remain to be identified, we have shown that Msh2 plays a crucial role in UVB-induced cell cycle arrest both in the epidermis and in cultured keratinocytes.

2. Material and methods 2.1. Generation of mice The generation of Msh2-deficient mice has been described previously [24]. This mouse strain was crossed with albino hairless mice (SKH-1) and the offspring was crossed back with SKH-1. This progeny was inter-crossed to obtain hairless WT, Msh2-deficient and heterozygous littermates. The heterozygous and WT littermates were used as controls for the Msh2-deficient mice. Hairless Xpc-deficient mice [18] were crossed with hairless Msh2-deficient mice (see above) to obtain hairless mice heterozygous for both the Xpc gene and the Msh2 gene. These mice were subsequently crossed with hairless Xpc-deficient mice. The appropriate offspring (Xpc-deficient and heterozygous for the Msh2 gene) was crossbred to generate hairless Xpc-deficient mice being Msh2-proficient (Xpc−/− Msh2+/+ ), Msh2-deficient (Xpc−/− Msh2−/− ) or heterozygous for the Msh2 gene (Xpc−/− Msh2+/− ). Animals were kept individually with standard mouse chow and water available ad libitum in a room illuminated with yellow light (no measurable UV) in a 12-h cycle. 2.2. UV dosimetry and radiation regimens Accurate comparisons between our present study and previously reported data [18] require equal UVB radiation regimens in all experiments. The irradiation set-up (with Philips TL12 lamps) used in the present experiments was identical to that described previously [13,18]. Permission for the experiments was granted by the ethical committee of Animal Wellfare of Utrecht University. 2.3. Determination of UV sensitivity Hairless Msh2−/− , Msh2+/− and Msh2+/+ mice were exposed to UVB radiation using a Hanovia Kromayer lamp equipped with a Schott-WG305 filter (output = 135 J/m2 s; 290–400 nm) as described [25]. Mice were exposed to 8, 12, 16, 20, 24 and 28 s of UVB light from the Kromayer lamp. All exposures were given in duplicate on separate mice. The mice were checked for edema and/or erythema 48 h after exposure; the sensitivity was assessed by the minimum exposure that resulted in just visible skin reactions. 2.4. Isolation of epidermis and keratinocytes Mice were irradiated with a single UV dose from TL12 lamps (500 or 2000 J/m2 ) and sacrificed at various times after irradiation. One hour before sacrifice, mice received an i.p. injection of 5-bromo-deoxyuridine (BrdUrd) (5 mg in 300 ␮l phosphate buffered saline (PBS), pH 7.4). The centre dorsal skin (0.5 cm × 0.5 cm) was snap frozen in liquid nitrogen. In addition, two strips of mid dorsal skin were excised. Epidermis and dermis were separated by overnight

M. van Oosten et al. / DNA Repair 4 (2005) 81–89

thermolysine/Protease X (Sigma) treatment at 4 ◦ C. Epidermal cell suspensions were obtained by treatment with trypsine (0.25%, 3 h, 4 ◦ C) followed by ultrasonic vibration in 5% NBCS in PBS on ice. Cells were fixated in 70% ethanol and stored at −20 ◦ C. 2.5. Establishment of mouse keratinocyte cell lines Mouse keratinocytes were isolated from epidermal sheets isolated from newborn mice and used to establish permanent cell lines described in detail elsewhere (manuscript in preparation). In short, skins were separated from newborns and epidermis and dermis were separated by overnight trypsine treatment (0.25%, 4 ◦ C). Epidermis was minced with forceps to release the cells. Cell cultures were maintained under low calcium conditions (0.05 mM) in EMEM supplemented with 8% foetal calf serum [26,27] and 1 ng/ml keratinocyte growth factor (KGF). Low-calcium conditions ensure that the keratinocytes will not undergo terminal differentiation. For cell cycle analysis, 2 × 106 keratinocytes were plated per P50 petridish and cells were grown to an areal coverage of approximately 70%. Prior to UVB radiation, the medium was discarded and keratinocytes were exposed in PBS to 50 J/m2 UV from TL12 lamps. Subsequently, PBS was removed and fresh medium was added. One hour before harvesting the cells, 100 ␮M BrdUrd and 10 ␮M FdU were added to the cultures to label cells in S-phase. After harvesting, cells were washed, fixated in ethanol, and stored at −20 ◦ C. 2.6. Determination of apoptotic keratinocytes in skin biopsies Apoptotic keratinocytes in the epidermis were identified as described previously [18]. The extent of apoptosis was quantified as the number of active caspase 3 positive cells per arbitrary unit length along the epidermis, i.e. approximately 33 basal cells. 2.7. Dual BrdUrd/DNA flow cytometric cell cycle analysis The simultaneous analysis of BrdUrd incorporation and DNA content using flow cytometry was performed as described previously [28]. Data analysis was performed as described before [29] using Cell Quest software (Becton Dickinson) and WinMDI 2.8 (free software). 2.8. Premature chromosome condensation analysis The technique of premature chromosome condensation (PCC) in principle enables preparation of chromosome spreads independent of cell cycle stage and with minimal culturing artefacts. To induce PCC, methods have been introduced that involve inhibitors of types I and 2A protein phosphatases such as calyculin A (CA). Analysis of chromosomes in inter-phase keratinocytes was performed after

83

calyculin A-induced PCC [30,31]. Briefly, 72 h after UV exposure, 120 nM calyculin A (Sigma) was added to the cultures and the cells were incubated for 1.5–2 h, harvested and treated with pre-warmed 0.075 M KCl for 20 min at 37 ◦ C. Cells were fixed with methanolic/acetic acid (4/1, v/v), dropped on precleaned slides and air-dried. For analysis, preparations were stained with an aqueous Giemsa solution. The morphology of observed PCCs (univalent in G1 , pulverised in S-phase and bivalent in G2 ) represented the position of cells at specific stages of the cell cycle.

3. Results 3.1. UVB sensitivity of Msh2 knockout mice Msh2+/+ , Msh2+/− and Msh2−/− were exposed for 8, 12, 16, 20, 24 and 28 s to the low-cut filtered Hanovia Kromayer lamp, and monitored at 48 h after exposure. No clear edema was seen in any of the genotypes with exposures up to 2700 J/m2 . Exposures of 3240 and 3780 J/m2 elicited edema in all three genotypes tested. As no statistically significant differences were observed between the genotypes, Msh2−/− mice appear to have a minimal erythema dose (MED) comparable to that of Msh2 heterozygous and WT mice. 3.2. Apoptosis At a dose of 500 J/m2 UVB, small numbers of apoptotic cells were identified in the epidermis (Fig. 1a). Comparable numbers were observed in our previous study [18]. Small, but significant, differences were observed between Xpc−/− Msh2−/− and Msh2−/− or Msh2+/+ mice at 6 h after irradiation (p < 0.01) and between Xpc−/− Msh2−/− and Msh2−/− mice at 24 h after irradiation (p < 0.05). In our previous study, the differences between the WT and Xpc−/− genotypes at this dose were somewhat less pronounced. Importantly, absence of Msh2 has no influence whether or not in a Xpc−/− background. This is further highlighted at higher dose at which the apoptotic response is much more pronounced. A dose of 2000 J/m2 induced apoptosis in the epidermis of Msh2−/− mice and this was not significantly different from apoptosis in WT or Msh2+/− mice (Fig. 1b). 3.3. Cell cycle progression in UVB-irradiated mouse epidermis The percentage of keratinocytes in the different phases of the cell cycle was determined by simultaneous BrdUrd/DNA flow-cytometric analysis. In unirradiated animals, no significant differences in distribution of cells over the different phases of the cell cycle were observed between Msh2+/+ Xpc−/− , Msh2+/− Xpc−/− and Msh2−/− Xpc−/− mice. A representative distribution is shown in Fig. 2a with approximately 92% of the cells in G1 /G0 -phase, 4% in G2 /M-phase and 4% in S-phase. This distribution is comparable to that of

84

M. van Oosten et al. / DNA Repair 4 (2005) 81–89

Fig. 1. UVB-induced apoptosis in the mouse epidermis. Mice were exposed to a single dose of UVB and killed at various times after irradiation (n ≥ 2 at each time point). (a) Apoptotic cells/arbitrary unit (a.u.) in the epidermis of WT (closed bars), Msh2−/− (crossed-hatched bars), Xpc−/− Msh2+/+ (hatched bars), Xpc−/− Msh2−/− (open bars) exposed to 500 J/m2 UVB; (b) number of apoptotic cells/a.u. in the epidermis of WT (closed bars), Msh2+/− (hatched bars), Msh2−/− (open bars) exposed to 2000 J/m2 UVB. Data are means ± S.E.M. (n ≥ 3) or variation (n = 2; t = 168 h).

SKH-1 hairless mice [28] and hairless NER-deficient mice [18]. Exposure of mice to UVB radiation resulted in a change in the distribution of cells over the different phases of the cell cycle. Since WT and mice heterozygous for Msh2 displayed similar cell cycle profiles after UVB irradiation, we combined the data from these mice in the calculations. The same held for Xpc−/− and Xpc−/− Msh2+/− mice. Fig. 3 shows the percentage of cells in the S-phase of the cell cycle as function of time after a single exposure to 500 J/m2 UVB. The percentage of BrdUrd-positive S-phase cells is transiently increased in WT, Msh2−/− ,Xpc−/− mice and Msh2−/− Xpc−/− doubleknockout mice (Fig. 3a). No significant differences in the percentage of BrdUrd-positive S-phase cells were observed between Msh2-proficient and Msh2-deficient mice at any of the time points tested. The epidermis of UVB-irradiated NER-deficient mice harbours cells with a DNA content between 2N and 4N that lack BrdUrd uptake (Fig. 2b). These BrdUrd-negative ‘quiescent’ S-phase (QS) cells [32] were previously observed in Csb−/− and Xpa−/− mice already after a UVB-dose of 40 J/m2 , whereas in Xpc−/− mice, a UVB dose of 250 J/m2

was required to generate these QS cells [18]. In contrast, even at UVB doses up to 2000 J/m2 , these cells are not manifest in WT cells. Similar to WT mice, Msh2−/− mice did not appear to accumulate QS cells, but these cells did accumulate in the Xpc−/− Msh2−/− mice as well as in Xpc−/− mice (Fig. 3b). However, at 72 h after irradiation, the percentage of QS cells was significantly higher in Xpc−/− mice than in Xpc−/− Msh2−/− mice (p < 0.01; n ≥ 4) indicating that Msh2 contributes to the accumulation of QS cells in the epidermis of Xpc−/− mice. Fig. 4 shows the percentage of cells in the G2 -phase (4N DNA content and BrdUrd negative) at various time points after UVB irradiation. Quantitatively, the accumulation of cells in the G2 fraction in Xpc−/− mice is very similar to results in a previous study [18] demonstrating high reproducibility of the experiments. We note here that no such accumulation of cells was observed in Msh2-deficient or WT littermates exposed to 500 J/m2 UV. Likewise to Xpc−/− mice, we observed an accumulation of G2 cells in Xpc−/− Msh2−/− mice; however, at 72 h after irradiation, this accumulation was significantly less than in Msh2-proficient Xpc−/− mice (p < 0.01; n ≥

Fig. 2. Bivariate dot-plots showing the distribution of the log green fluorescence of the FITC anti-BrdUrd staining (DNA-synthesis, y-axis) vs. the red fluorescence of the PI staining (DNA content, x-axis) of keratinocytes from Xpc−/− Msh2−/− mice. Mice were (a) unexposed or (b) exposed to a single dose of UVB (500 J/m2 ) 48 h prior to cell isolation. Cells were in vivo labelled with BrdUrd during the last hour before epidermal cell isolation.

M. van Oosten et al. / DNA Repair 4 (2005) 81–89

85

Fig. 4. Accumulation of cells in G2 -phase after UVB–irradiation. Mice were exposed to 500 J/m2 UVB, keratinocytes were isolated at various time points after exposure and the percentage of cells in the different phases of the cell cycle was determined. The percentage of cells in G2 derived from WT (circles), Msh2−/− (triangles up), Xpc−/− Msh2+/+ (squares) and Xpc−/− Msh2−/− (triangles down) mice are given. Data are means ± S.E.M. (n ≥ 3) or variation (n = 2; t = 168 h).

Fig. 3. Accumulation of S-phase cells after UVB–irradiation. Mice were exposed to 500 J/m2 UVB radiation and epidermal cells were isolated at various time points after UVB exposure. One hour before cell isolation, mice received a single dose of BrdUrd to identify S-phase cells. (a) The percentage of S-phase cells from UVB-irradiated WT (circles), Msh2−/− (triangles up), Xpc−/− Msh2+/+ (squares) and Xpc−/− Msh2−/− (triangles down) mice; (b) the percentage of BrdUrd negative S-phase cells from WT (circles), Msh2−/− (triangles up), Xpc−/− Msh2+/+ (squares) and Xpc−/− Msh2−/− (triangles down) mice. Data are means ± S.E.M. (n ≥ 3) or variation (n = 2; t = 168 h).

4). The approximately 40% reduction indicates that Msh2 affects the UVB-induced cell cycle arrest in keratinocytes from Xpc-deficient mice. 3.4. Cell cycle progression in UVB-irradiated cultured mouse keratinocytes To confirm the in vivo results and to further investigate the effects of Msh2 deficiency on UVB-induced cell cycle arrest

in Xpc−/− keratinocytes, we performed cell cycle analysis of established cultures of mouse keratinocytes (Fig. 5). Consistent with the in vivo experiments, UVB-exposed Xpc−/− keratinocytes displayed a cell cycle arrest (Fig. 5a) reaching a percentage of 37% cells in G2 versus approximately 5% in WT under the semi-confluent conditions of cell growth (see Section 2). In the UVB-irradiated Xpc−/− Msh2−/− keratinocytes, we observed a substantially lower percentage of G2 -arrested cells similar to the difference observed in vivo (32% reduction compared to Xpc−/− keratinocytes). Moreover, Msh2-deficiency also affected the UV-induced generation of QS cells: whereas Xpc−/− keratinocytes exposed to 50 J/m2 UV light showed 28% of QS cells at 72 h after exposure, Xpc−/− Msh2−/− accumulated only 16% of QS cells (a reduction of about 40%) (Fig. 5b). Using flow cytometry analysis, the G2 fraction is poorly separated from QS cells in late S-phase (with DNA content just under or equal to 4N). To understand the mechanism by which Msh2 contributes to the cell cycle arrest observed in UVB-exposed Xpc−/− keratinocytes, it is essential to

Fig. 5. Percentages of G2 - (a) and BrdUrd-negative S-phase cells (b) after single UVB exposure. WT (circles), Xpc−/− (triangles) and Xpc−/− Msh2−/− (squares) keratinocytes were exposed to 50 J/m2 UVB in vitro. One hour before isolation, cells were labelled with BrdUrd/FdUrd and subsequently prepared for flow cytometric analysis. Unexposed cells of each genotype served as controls at all time points. Percentages of S and G2 cells of unexposed cells were comparable to UV-exposed WT cells.

86

M. van Oosten et al. / DNA Repair 4 (2005) 81–89

Fig. 6. PCC analysis of UVB-irradiated Xpc−/− keratinocytes: (a) chromosomal morphology of S- (left) and G2 -phase (right) PCC, (b) percent S and G2 cells in Xpc−/− and WT keratinocytes 72 h after UVB-exposure (50 J/m2 ) and controls (unexposed cells). Bars represent unexposed (black) and irradiated (hatched bars) WT keratinocytes and unexposed (crossed-hatched bars) and irradiated (clear bars) Xpc−/− keratinocytes. (Note that G1 cells are underrepresented due to a lack of PCC in these cells.)

know the positioning of the arrested cells in the cell cycle. PCC analysis of UVB-irradiated keratinocytes revealed two different types of chromosomal pattern (Fig. 6a): fragmented (smashed) chromosomes and intact chromosomes with paired and condensed chromatids corresponding to Sphase and G2 cells, respectively. Quantification of these patterns in Xpc−/− and WT keratinocytes at 72 h after UV exposure revealed that the large majority of the arrested cells were actually not G2 cells, but rather represented (mid and) late S-phase cells (Fig. 6b).

4. Discussion In a previous study [18], we assessed cell cycle arrest of epidermal keratinocytes in UVB-irradiated NER-deficient mice. Most notably, the accumulation of tetraploid (4N) cells (termed G2 -arrested cells) was observed in epidermal cells from Xpc−/− mice exposed to a broad dose range of UVB radiation but not in those from Csb−/− , Xpa−/− or WT mice. In the current study, we show that UVB-irradiated Xpc−/− keratinocytes cultured in vitro mimic the cell cycle arrest observed in mice. In addition, we show here that the arrested Xpc−/− cells in culture do not reside in the G2 -phase of the cell cycle but rather represent non-replicating cells blocked in late S-phase. This finding is consistent with the observations by Hittelman and co-workers [33,34] who found that UV-

irradiated human XP cells exhibited impaired chromosome condensation most likely due to incomplete DNA replication at sites of UV-induced DNA damage. From the in vitro data, we assume that the fraction of arrested 4N cells in vivo in fact represents arrested S-phase cells rather than G2 cells. Owing to their TCR proficiency, epidermal keratinocytes of UVB-irradiated Xpc−/− mice can escape apoptosis and enter S-phase in the presence of persistent DNA damage in non-transcribed DNA [18]. The gradual accumulation of QS cells and a prominent accumulation of tetraploid late S-phase cells in UV-exposed Xpc−/− keratinocytes (in vivo and in vitro) indicate that unrepaired photolesions are responsible for inducing the S-phase arrest. Surprisingly, comparison of Msh2-deficient Xpc−/− with their Msh2-proficient Xpc−/− littermates disclosed that the MMR protein Msh2 is responsible for about half of the S-phase arrest at 72 h after UVB irradiation in the mouse skin. A similar observation was made for cultured keratinocytes. Taken together, this indicates that accumulation of S-phase Xpc−/− keratinocytes after UVB exposure is in part caused by Msh2 and that other factors (remain to be identified) also contribute to the observed Sphase arrest. Since Xpc−/− keratinocytes with UVB-induced DNA damage are not effectively arrested in S-phase in the absence of Msh2, an increased number of cells with DNA lesions will continue to cycle. This could lead to an increased number of cells with fixed mutations and thus an increased risk for development of skin cancer. Indeed, Xpc−/− Msh2−/− mice appeared to be more prone to UVB-induced skin cancer than Xpc−/− mice [35]. These results let the authors to conclude that an unidentified interaction between MMR and NER operates during the cells’ response to UV radiation [35]. Our present data show that this interaction involves cell cycle checkpoints. It is unlikely that the partial alleviation of the S-phase arrest represents an effect of MMR on TCR as mouse dermal fibroblasts deficient in Msh2 display normal TCR [36]. Moreover, Msh2−/− mice irradiated with a single dose of UVB performed a level of CPD repair identical to WT mice (unpublished results) and consistent with functional TCR, the MED value, levels of UV-induced apoptosis and cell cycle progression of epidermal keratinocytes in UVB-exposed Msh2−/− and WT mice were indistinguishable. In addition, Msh2−/− Xpa−/− mice are more prone to UV-induced skin cancer than either single mutant alone [37]. Since Xpa−/− mice are TCR deficient, the effect of Msh2 deficiency could not be mediated by impairment of TCR in these mice. The finding that UVB-irradiated Xpc−/− Msh2−/− cells proceed through early S-phase prior to arresting suggests that replication of damaged DNA is instrumental to the observed (mid and) late S-phase arrest. Replicative bypass of DNA lesions (translesion synthesis, TLS) during S-phase is mediated by specialised DNA polymerases capable to extend replication beyond DNA lesions either in an error-free or error-prone way. In humans, several DNA polymerases are known that promote synthesis past UV-induced photolesions, most notably DNA polymerases pol␩ and pol␨ [38,39].

M. van Oosten et al. / DNA Repair 4 (2005) 81–89

In case of error-prone TLS, incorrect bases may be incorporated opposite lesions giving rise to so-called compound lesions. These lesions might be recognised by the heterodimeric Msh2–Msh6 mismatch repair protein as suggested by results of in vitro binding experiments [40,41] and as a consequence, futile cycles of breakage and DNA synthesis might be induced [42], analogous to MNNG-induced alkylation damage [22]. The presence of chromosomes carrying single stranded gaps or breaks might provide the signal for the S-phase arrest (QS and 4N cells) in Xpc−/− mice and cultured keratinocytes. In addition to putative MMR-induced futile cycles of breakage and resynthesis underlying S-phase arrest, other mechanisms involving MMR might give rise to S-phase arrest of UVB-irradiated Xpc−/− keratinocytes. Brown et al. [19] reported that the mechanistic basis for MMR-dependent activation of ionising radiation-induced S-phase checkpoint is the damage-enhanced interaction of Msh2 with Chk2. Impairment of this system in cells deficient for MMR leads to radioresistant DNA synthesis. It is attractive to speculate that analogous to ionising radiation, Msh2 maintains the UVBinduced S-phase arrest by facilitating activation of Chk2 phosphorylation and degradation of CDC25A. This would imply that actual MMR per se is not necessarily required for the Msh2-dependent cell cycle arrest in XPC-deficient cells, but that the decreased arrest relates to a more direct signalling function of Msh2, e.g., binding but not necessarily futile cycling. Recently, Lin et al. observed a dissociation between DNA repair and DNA damage response functions of Msh2 in (cells of) mice with an Msh2 point mutation in their DNA [43], indicating that indeed, Msh2 can exert functions independent of MMR. Interestingly, the S-phase arrest in UVB-irradiated Xpc−/− mice is a transient event and arrested cells disappear between 72 and 168 h after irradiation. Our data demonstrate that the process underlying the abrogation of cell cycle arrest at later time points after irradiation does not require functional Msh2. In contrast to mice, cultured Xpc−/− keratinocytes exhibited a persistent cell cycle arrest after UVB irradiation (for a period up to 4 days). A possible explanation for the difference between mouse epidermis and cultured keratinocytes might be related to epidermal differentiation, a process by which arrested cells can be removed from the proliferating pool. In the cultured cells, this process is significantly retarded due to the low calcium culture conditions [26]. The reduced number of arrested cells in UVB-irradiated Xpc−/− Msh2−/− keratinocytes compared to Xpc−/− keratinocytes indicates that in the absence of functional MMR, a fraction of UVB-damaged cells passes through the G2 -phase into mitosis and continues to cycle resulting in enhanced survival after UV treatment. There is however no consensus as to whether Msh2+/+ cells are more sensitive to UV-induced cytotoxic effects than Msh2−/− cells. Whereas some findings [44,45,37] support a greater clonal survival of Msh2-deficient cells after UV exposure than cells with functional Msh2, others [45,46] found no difference in survival after UV irradiation. Moreover, down regulation of MMR may decrease UV

87

sensitivity of XPA-deficient skin cancer cells [47]. Our pilot studies show enhanced clonal survival of UVB-exposed Xpc−/− Msh2−/− when compared to Xpc−/− keratinocytes (results not shown). In summary, our experiments indicate that the absence of an apoptotic response [18] allows epidermal cells in UVexposed Xpc−/− mice to enter S-phase and to continue replication in the presence of persistent DNA lesions. However, the gradual accumulation of ‘quiescent’ S-phase cells and a prominent arrest of tetraploid cells in UV-exposed Xpc−/− keratinocytes (in vivo and in vitro) indicate that even in the presence of TLS polymerases, the unrepaired photolesions provide severe blocks to the replication machinery. Comparison of Xpc−/− Msh2−/− mice with Xpc−/− Msh2+/−,+/+ littermates showed that Msh2 was in part responsible for the S-phase arrest. The nature of S-phase cells blocked or not blocked by Msh2 and the implications of failure in this checkpoint clearly need to be investigated further.

Acknowledgement This work was supported by the Dutch Cancer Society Grant UU 97-1531.

References [1] J.E. Cleaver, K.H. Kraemer, Xeroderma pigmentosum, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic Basis of Inherited Disease, McGraw-Hill, New York, 1989, pp. 2949–2971. [2] K.H. Kraemer, M.L. Myung, J. Scotto, DNA repair protects against cutaneous and internal neoplasia; evidence from xerpderma pigmentosum, Arch. Dermatol. 123 (1987) 241–250. [3] D. Bootsma, The genetic defect in DNA repair deficiency syndromes, Eur. J. Cancer 29A (1993) 1482–1488. [4] J. Venema, A. van Hoffen, V. Karcagi, A.T. Natarajan, A.A. van Zeeland, L.H. Mullenders, Xeroderma pigmentosum complementation group C cells remove pyrimidine dimers selectively from the transcribed of active genes, Mol. Cell. Biol. 11 (1991) 4128–4134. [5] B.J. Hwang, J.M. Ford, P.C. Hanawalt, G. Chu, Expression of the p48 xeroderma pigementosum gene is p53-dependent and is involved in global genomic repair, Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 424–428. [6] A. de Vries, C.T. van Oostrom, F.M. Hofhuis, P.M. Dortant, R.J. Berg, F.R. de Gruijl, P.W. Wester, C.F. van Kreijl, P.J. Capel, H. van Steeg, Increased susceptibility to ultraviolet-B and carcinogens of mice lacking the DNA excision repair gene Xpa, Nature 377 (1995) 169–173. [7] H. Nakane, S. Takeuchi, S. Yuba, M. Saijo, Y. Nakatsu, H. Murai, Y. Nakatsuru, T. Ishikawa, S. Hirota, Y. Kitamura, High incidence of ultraviolet-B- or chemical-carcinogen-induced skin tumours in mice lacking the xeroderma pigmentosum group A gene, Nature 377 (1995) 165–168. [8] D.L. Cheo, H.J.T. Ruven, L.B. Meira, R.E. Hammer, D.K. Burns, N.J. Tappe, A.A. van Zeeland, L.H.F. Mullenders, E.C. Friedberg, Characterisation of defective nucleotide excision repair in XPC mutant mice, Mutat. Res. 374 (1997) 1–9. [9] R.J. Berg, H.J. Ruven, A.T. Sands, F.R. de Gruijl, L.H.F. Mullenders, Defective global genome repair in XPC mice is associated with skin cancer susceptibility but not with sensitivity to UVB-

88

[10]

[11]

[12]

[13]

[14]

[15]

[16] [17] [18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

M. van Oosten et al. / DNA Repair 4 (2005) 81–89 induced erythema and edema, J. Invest. Dermatol. 110 (1998) 405– 409. G.T. van der Horst, H. van Steeg, R.J. Berg, A.J. van Gool, J. de Wit, G. Weeda, H. Morreau, R.B. Beems, C.F. van Kreijl, F.R. de Gruijl, D. Bootsma, J.H. Hoeijmakers, Defective transcription-coupled repair in Cockayne syndrome B mice is associated with skin cancer predisposition, Cell 89 (1997) 425–435. A.T. Sands, A. Abuin, A. Sanchez, C.J. Conti, A. Bradley, High susceptibility to ultraviolet-induced carcinogenesis in mice lacking XPC, Nature 377 (1995) 162–165. D.L. Cheo, L.B. Meira, D.E. Hammer, D.K. Burns, A.T.B. Doughty, E.C. Friedberg, Synergistic interaction between XPC and p53 mutations in double-mutant mice: neural tube abnormalities and accelerated UV radiation-induced skin cancer, Curr. Biol. 6 (1996) 1691–1694. R.J. Berg, A. de Vries, H. van Steeg, F.R. de Gruijl, Relative susceptibilities of XPA knockout mice and their heterozygous and wildtype littermates to UVB-induced skin cancer, Cancer Res. 57 (1997) 581–584. L.L. Hill, A. Ouhtit, S.M. Loughlin, M.L. Kripke, H.N. Ananthaswamy, L.B. Owen-Schaub, Fas ligand: a sensor for DNA damage critical in skin cancer etiology, Science 285 (1999) 898– 900. A. Ziegler, A.S. Jonason, D.J. Leffell, J.A. Simon, H.W. Sharma, J. Kimmelman, L. Remington, T. Jacks, D.E. Brash, Sunburn and p53 in the onset of skin cancer, Nature 372 (1994) 773–776. J.H.J. Hoeijmakers, Genome maintenance mechanisms for preventing cancer, Nature 411 (2001) 366–374. K.D. Kolodner, C.D. Putnam, K. Myung, Maintenance of genome stability in Saccharomyces cerevisiae, Science 297 (2002) 552–557. M. van Oosten, H. Rebel, E.C. Friedberg, H. van Steeg, G.T.J. van der Horst, H.J. van Kranen, A. Westerman, A.A. van Zeeland, L.H.F. Mullenders, F.R. de Gruijl, Differential role of transcription-coupled repair in UVB-induced G2 arrest and apoptosis in mouse epidermis, Proc. Natl. Acad. Sci. U.S.A. 97 (2000) 11268–11273. K.D. Brown, A. Rathi, R. Kamath, D.I. Beardsley, Q. Zhan, J.L. Mannino, R. Baskaran, The mismatch repair system is required for S-phase checkpoint activation, Nat. Genet. 33 (2003) 80–84. W.Y. Feng, E.H. Lee, J.B. Hays, Recombinagenic processing of UVlight photoproducts in nonreplicating phage DNA by the Escherichia coli methyl-directed mismatch repair system, Genetics 129 (1991) 1007–1020. W.Y. Feng, J.B. Hays, DNA structures generated during recombination initiated by mismatch repair of UV-irradiated nonreplicating phage DNA in Escherichia coli: requirements for helicase, exonucleases, and RecF and RecBCD functions, Genetics 140 (1995) 1175–1186. P. Karran, M. Bignami, Self-destruction and tolerance in resistance of mammalian cells to alkylation damage, Nucleic Acids Res. 20 (1992) 2933–2940. Y.P. Lu, Y.-R. Lou, P. Yen, D. Mitchell, M.T. Huang, A.H. Conney, Time course of early adaptive responses to ultraviolet B light in the epidermis of SKH-1 mice, Cancer Res. 59 (1999) 4591–4602. R. Smits, N. Hofland, W. Edelmann, M. Geugien, S. JagmohanChangur, C. Albuquerque, C. Breukel, R. Kucherlapati, M.F. Kielman, R. Fodde, Somatic Apc mutations are selected upon their capacity to inactivate the beta-catenin downregulating activity, Genes Chromosomes Cancer 29 (2000) 229–239. R.J.W. Berg, H. Rebel, G.T.J. van der Horst, H.J. van Kranen, L.H.F. Mullenders, W.A. van Vloten, F.R. de Gruijl, Impact of global genome repair versus transcription-coupled repair on ultraviolet carcinogenesis in hairless mice, Cancer Res. 60 (2000) 2858–2863. H. Hennings, D. Michael, M.D. Cheng, P. Steinert, K. Holbrook, S.H. Yuspa, Calcium regulation of growth and differentiation of mouse epidermal cells in culture, Cell 19 (1980) 245–254. H. Hennings, in: I.M. Leigh, F.M. Watt (Eds.), Keratinocyte Methods, Cambridge University Press, Cambridge, 1994, pp. 21–23.

[28] A. de Laat, E.D. Kroon, F.R. de Gruijl, Cell cycle effects and concomitant p53 expression in hairless human skin after longwave UVA (365 nm) irradiation: a comparison with UVB irradiation, Photochem. Photobiol. 65 (1997) 730–735. [29] A. de Laat, M. van Tilburg, J.C. van der Leun, W.A. van Vloten, F.R. de Gruijl, Cell cycle kinetics following UVA irradiation in comparison to UVB and UVC irradiation, Photochem. Photobiol. 63 (1996) 492–497. [30] E. Gotoh, Y. Asakawa, H. Kosaka, Inhibition of protein serine threonine phosphatase directly induces premature chromosome condensation in mammalian cells, Biomed. Res. 16 (1995) 63–68. [31] V. Bezrookove, R. Smits, G. Moeslein, R. Fodde, H.J. Tanke, A.K. Raap, F. Darroudi, Premature chromosome condensation revisited: a novel chemical approach permits efficient cytogenetic analysis of cancers, Genes Chromosomes Cancer 38 (2003) 177–186. [32] F. Zolzer, C. Struben, K. Knuhmann, C. Streffer, H. Sack, Quiescent S-phase cells as indicators of extreme physiological conditions in human tumor xenografts, Int. J. Radiat. Oncol. Biol. Phys. 45 (1999) 1019–1024. [33] W.N. Hittelman, M. Pollard, Visualization of chromatin events associated with repair of ultraviolet light-induced damage by premature chromosome condensation, Carcinogenesis 5 (1984) 1277– 1285. [34] W.N. Hittelman, Visualization of chromatin events during DNA excision repair in XP cells: deficiency in localized but not generalized chromatin events, Carcinogenesis 7 (1986) 1975–1980. [35] L.B. Meira, D.L. Cheo, A.M. Reis, N. Claij, D.K. Burns, H. te Riele, E.C. Friedberg, Mice defective in the mismatch repair gene Msh2 show increased predisposition to UVB radiation-induced skin cancer, DNA Repair (Amst.) 1 (2002) 929–934. [36] E. Sonneveld, H. Vrieling, L.H.F. Mullenders, A. van Hoffen, Mouse mismatch repair gene Msh2 is not essential for transcription-coupled repair of UVB-induced cyclobutane pyrimidine dimers, Oncogene 20 (2001) 538–841. [37] M. Yoshino, Y. Nakatsu, H. te Riele, S. Hirota, Y. Kitamura, K. Tanaka, Additive roles of XPA and MSH2 genes in UVB-induced skin tumorigenesis in mice, DNA Repair (Amst.) 1 (2002) 935–940. [38] S.-L. Yu, R.E. Johnson, S. Prakash, L. Prakash, Requirement of DNA polymerase eta for error-free bypass of UV-induced CC and TC photoproducts, Mol. Cell. Biol. 21 (2001) 185– 188. [39] R.E. Johnson, M.T. Washington, L. Haracska, S. Prakash, L. Prakash, Eukariotic polymerases iota and zeta act sequentially to bypass DNA lesions, Nature 406 (2000) 1015–1019. [40] D. Mu, M. Tursun, D.R. Duckett, J.T. Drummond, P. Modrich, A. Sancar, Recognition and repair of compound DNA lesions (base damage and mismatch) by human mismatch repair and excision repair systems, Mol. Cell. Biol 17 (1997) 760–769. [41] H. Wang, C.W. Lawrence, G.M. Li, J.B. Hays, Specific binding of human MSH2.MSH6 mismatch repair protein heterodimers to DNA incorporating thymine- or uracil-containing UVlight photoproducts opposite mismatched bases, J. Biol. Chem. 274 (1999) 16894– 16900. [42] M.T. Hawn, A. Umar, J.M. Carethers, G. Marra, T.A. Kunkel, C.R. Boland, M. Koi, Evidence for a connection between mismatch repair system and G2 cell cycle checkpoint, Cancer Res. 55 (1995) 3721–3725. [43] D.P. Lin, Y. Wang, S.J. Scherer, A.B. Clark, K. Yang, E. Avdievich, B. Jin, U. Werling, T. Parris, N. Kurihara, A. Umar, R. Kucherlapati, M. Lipkin, T.A. Kunkel, W. Edelmann, An Msh2 point mutation uncouples DNA mismatch repair and apoptosis, Cancer Res. 64 (2004) 517–522. [44] I. Mellon, D.K. Rajpal, M. Koi, C.R. Boland, G.N. Champe, Transcription-coupled repair deficiency and mutations in human mismatch repair genes, Science 272 (1996) 557–560. [45] A.H. Reitmair, R. Risley, R.G. Bristow, T. Wilson, A. Ganesh, A. Jang, J. Peacock, S. Benchimol, R.P. Hill, T.W. Mak, R. Fishel,

M. van Oosten et al. / DNA Repair 4 (2005) 81–89 M. Meuth, Mutator phenotype in Msh2-deficient murine embryonic fibroblasts, Cancer Res. 57 (1997) 3765–3771. [46] K. Nara, F. Nagashima, A. Yasui, Highly elevated ultraviolet-induced mutation frequency in isolated Chinese hamster cell lines defective in nucleotide excision repair and mismatch repair proteins, Cancer Res. 61 (2001) 50–52.

89

[47] M. Ichikawa, H. Nakane, G. Marra, C. Corti, J. Jiricny, M. Fitch, J.M. Ford, M. Ikejima, T. Shimada, M. Yoshino, S. Takeuchi, Y. Nakatsu, K. Tanaka, Decreased UV sensitivity, mismatch repair activity and abnormal cell cycle checkpoints in skin cancer cell lines derived from UVB-irradiated XPA-deficient mice, Mutat. Res. 459 (2000) 285–298.