Colcemid inhibits the rejoining of the nucleotide excision repair of UVC-induced DNA damages in Chinese hamster ovary cells

Mutation Research 588 (2005) 118–128

Colcemid inhibits the rejoining of the nucleotide excision repair of UVC-induced DNA damages in Chinese hamster ovary cells Hsun Li a,1 , Tai-Wei Chang a,1 , Yi-Chi Tsai b , Shu-Fu Chu a , Yi-Yin Wu b , Bor-Show Tzang b , Chu-Bin Liao b , Yin-Chang Liu a,b,∗ a

Institute of Molecular Medicine, National Tsing-Hua University, Hsin-Chu 30043, Taiwan b Department of Life Science, National Tsing-Hua University, Hsin-Chu 30043, Taiwan

Received 8 February 2005; received in revised form 13 September 2005; accepted 16 September 2005 Available online 11 November 2005

Abstract In our previous study, we found that colcemid, an inhibitor of mitotic spindle, promotes UVC-induced apoptosis in Chinese hamster ovary cells (CHO.K1). In this study, a brief treatment of colcemid on cells after but not before UV irradiation could synergistically reduce the cell viability. Although colcemid did not affect the excision of UV-induced DNA damages such as [6–4] photoproducts or cyclobutane pyrimidine dimers, colcemid accumulated the DNA breaks when it was added to cells following UV-irradiation. This colcemid effect required nucleotide excision repair (NER) since the same accumulation of DNA breaks was barely or not detected in two NER defective strains of CHO cells, UV5 or UV24. Furthermore, the colcemid effect was not due to semi-conservative DNA replication or mitosis since the colcemid-caused accumulation of DNA breaks was also seen in non-replicating cells. Moreover, colcemid inhibited rejoining of DNA breaks accumulated by hydroxyurea/cytosine arabinoside following UV irradiation. Nevertheless, colcemid did not affect the unscheduled DNA synthesis as assayed by the incorporation of bromodeoxyuridine. Taken together, our results suggest that colcemid might inhibit the step of ligation of NER pathways. © 2005 Elsevier B.V. All rights reserved. Keywords: UVC-induced DNA damages; Colcemid; Nucleotide excision repair

1. Introduction When the genetic stability of mammalian cells is challenged by various environmental assaults such as ultraviolet light (UV), the DNA repair systems need to be turned on to counteract the deleterious effects of DNA lesions. Nucleotide excision repair is a mechanism to remove the DNA lesions induced by shortwave UV

∗ 1

Corresponding author. E-mail address: [email protected] (Y.-C. Liu). Equal contribution.

1383-5718/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.mrgentox.2005.09.005

radiation (i.e. UVB or UVC) and by numerous chemicals including chemotherapeutic agents such as cisplatin. Two major classes of DNA lesions induced by UV are cyclobutane pyrimidine dimers (CPD) and [6–4] photoproducts (6–4PP). The pathway of nucleotide excision repair of mammalian cells has been known in quite detail [1,2]. In brief, the DNA lesions are identified by DDB (XPE) or stalled RNA polymerase II, in non-transcribed or transcribed region, respectively. Then, the general transcription factor complex TFIIH, containing the XPB and XPD helicases, mediates strand separation at the site of the damage. The damage-containing sequences are removed by endonucleases XPG and ERCC1-XPF

H. Li et al. / Mutation Research 588 (2005) 118–128

complex. Finally, the gaps created by the excision step are filled by PCNA dependent DNA polymerases and the ends are sealed by ligase I. Genomic assault by evoking checkpoint pathway may arrest cell cycle progression for DNA repair or lead to apoptosis [3,4]. In our previous study, we found that the Chinese hamster ovary cells (CHO.K1) after UVCirradiation treatment failed to arrest at G1 and exhibited a certain degree of apoptosis [5]. This UV-induced apoptosis was exacerbated if the UV irradiated cells were cultured in growth medium containing colcemid, an inhibitor of mitotic spindle [5,6]. The similar if not identical effect of colcemid on DNA damage-induced apoptosis was also reported in other situation where colcemid was found to accelerate the DNA-damage induced apoptosis by actinomycin [7,8]. Although the potential of this colcemid effect in application to chemotherapy is high, the underlying mechanism of the colcemid effect remained unclear. In this report, we examined if the enhancement effect of colcemid on UV induced-cell death could be linked to the colcemid effect on repair of DNA damage. Results of the study suggest that although colcemid did not affect the excision of 6–4PP and CPD, it accumulated DNA breaks in UV irradiated cells and prevented the rejoining of DNA breaks accumulated by DNA synthesis inhibitors following UV irradiation. However, colcemid did not affect the repair DNA synthesis. We propose that colcemid may hinder the ligation step of NER. 2. Materials and methods 2.1. Cell cultures The cell lines of Chinese hamster ovary including CHO.K1 and UV5 and UV24, the two strains defective in nucleotide excision repair were originally obtained from the American Type Culture Collection (Manassas, VA). These cells were cultured at 37 ◦ C in a humidified atmosphere containing 5% CO2 in McCoy’s 5A medium (Sigma Chemicals, St. Louis, MO). The growth medium was supplemented with 10% fetal calf serum, 0.22% sodium bicarbonate, penicillin (100 U/ml), streptomycin (100 ng/ml) and 0.03% glutamine. 2.2. UV-irradiation treatment The procedures for UV irradiation were similar as those described previously [6]. Briefly, cells in the dish were washed with phosphate-buffered saline (PBS) after removal of culture medium, uncovered and exposed to UV light (254 nm) in the marked area of the tissue culture hood that had been pre-calibrated for the required dose. Doses of UV-irradiation were calibrated with a UV radiometer (UVP, San Gabriel, CA).

119

2.3. Cell growth analysis Same numbers of CHO cells were seeded in wells of sixwell plates for 1–2 days (to reach about 50% confluence) before treatment of UV-irradiation. After treatment, the cells were cultured in media with or without colcemid (50 ng/ml) for 6 h and then in normal media for another 30 h before cell proliferation assay using 3-(4,5-dimethylthiazolyl-2)-2,5diphenyltetrazolium bromide (MTT) kit (Sigma–Aldrich Inc.) or direct cell counts. 2.4. ELISA of UV-induced DNA damage CHO cells were exposed UV-irradiated (25 J/m2 ) and recovered in media with or without colcemid (50 ng/ml). After 6 h, the growth media of all cultures were changed to the normal medium (i.e. without colcemid). The cells were harvested by trypsinization at indicated intervals following UV-irradiation. Genomic DNA of cells was purified using a Genomic DNA extraction kit (Qiagen, Crawley, UK). 6–4PP and CPD were quantitated by an enzyme-linked immunosorbent assay (ELISA) using 64 M-2 and TDM-2 monoclonal antibodies, respectively [9]. The procedures of ELISA for determining CPD from genomic DNA [10] were followed. In brief, 96-well polyvinylchloride flat bottom microtiter plates, precoated with 0.003% protamine sulfate, were coated with sample DNA. Equal amounts of DNA (200 or 10 ng per well for 6–4PP and CPD, respectively) calculated from the absorbance at 260 nm, were coated in the wells. The binding of monoclonal antibodies to photolesions in immobilized-DNA in wells (in quadruplicate) was detected with biotinylated F(ab )2 fragment of goat anti-mouse IgG and then with streptavidin peroxidase. The absorbance of colored products derived from o-phenylene diamine was measured at 492 nm by Titertek Multiskan Plus MK (Labsystems, Helsinki, Finland). 2.5. Comet assay DNA damage was evaluated using the alkaline (pH > 13) single-cell gel electrophoresis or comet assay [11,12]. After UV and/or other treatments, cells (∼2 × 104 cells) were harvested by trysinization and resuspended in 10 ␮l of 1 × PBS and then mixed with 1% low melting agarose (LMA, 100 ␮l per 35 mm dish culture). The mixture was spread on a frosted slide pre-coated with a layer of 1% agarose. The agarose-spread was allowed to solidify on ice for 5 min before the third layer of 1% LMA (100 ␮l) was applied. Slides were dipped in a cold lysis solution (2.5 M NaCl, 100 mM EDTA, 10 mM Tris pH 10, 1% Triton X-100, 10% DMSO) at 4 ◦ C for at least 1 h. The slides were transferred to an electrophoresis box containing an alkaline solution (300 mM NaOH, 1 mM EDTA, pH 13) and kept for 20 min at 4 ◦ C. A current of 25 V (300 mA) was applied for 25 min. After electrophoresis, the slides were neutralized in 0.4 M Tris (pH 7.5) for 5 min and stained with propidium iodide (50 ␮g/ml). Average comet tail moment was scored (total 50 or more cells) under Axioplan2 Microscope (Zeiss)

120

H. Li et al. / Mutation Research 588 (2005) 118–128

by using the comet imager software (TriTek Co., Sumerduck, VA). 2.6. Unscheduled DNA synthesis The method of UDS described previously [13] was followed with slight modification. Cells, after UV treatment, were cultured in medium containing 10 ␮M of BrdU for 1 h. Cells were fixed in methanol:acetic acid (3:1) on ice for 20 min. DNA was denatured by 4 N HCl at room temperature for 30 min. Incorporation of BrdU was revealed sequentially by antiBrdU antibody (BD Biosciences, San Jose, CA) at 1:300 dilution and Alexa fluor 594 goat anti-mouse IgG conjugate (Molecular Probes, Eugene, OR) at 1:100 dilution. Nuclear DNA was counterstained with 4 ,6-diamidino-2-phenylindole (DAPI, Molecular Probes). 2.7. DN A fragmentation assay DNA fragmentation assay was performed as described previously [14]. In brief, cells after treatment were incubated in lysis buffer (10 mM Tris–HCl, 10 mM EDTA, and 0.2% Triton X-100, at pH 7.5) for 10 min on ice. DNA in the supernatant collected from centrifugation (12,000 × g at 4 ◦ C for 30 min) of lysis mixture, was extracted with phenol and then with phenol–chloroform:isoamylalcohol (24:1), and precipitated by adding equal volume of isopropanol to the solution after the concentration of NaCl in supernatant was adjusted to 0.3 M. The DNA pellets, collected by centrifugation of the mixture, rinsed with 70% ethanol, re-suspended in TE buffer (10 mM Tris–HCl, 1 mM EDTA, at pH 7.5), were incubated with RNase A (0.6 mg/ml) at 37 ◦ C for 30 min. The DNA samples were separated by electrophoresis on 1.5% agarose gel for fragmentation assay. 2.8. Flow-cytometric analysis of apoptosis The procedures of double-staining of propidium iodide (PI) and Annexin V to were used as described previously [15]. The method is based on the differential integrity of cell membranes with living or apoptotic (or necrotic) cells, which determine the uptake (or exclusion) of PI and Annexin. In brief, cells (about 1 × 106 ), after treatment and washing twice with PBS, were incubated in buffer (10 mM Hepes/NaOH, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2 ) containing FITC-Annexin V and PI both at 1 ␮g/ml for 5 min at room temperature in dark. Flowcytometric analysis was performed with Becton Deckinson FACScan. 2.9. Statistics Data represent the mean ± standard errors. All experiments were performed independently at least three times. Statistical analyses were performed with Student’s t-test. P-values less than 0.01 were considered to be statistically significant and were marked by *.

Fig. 1. Effect of colcemid on cell growth and cell cycle distribution of UV-irradiated cells. (A) Enhancement of colcemid on the reduction of the cell growth caused by UV-irradiation. About 7.5 × 104 cells from the aliquots of a single-cell suspension were seeded to each well of sixwell plates and incubated for 1–2 days before receiving UV irradiation (25 J/m2 ) or mock treatment. Then, cells were incubated in medium with or without colcemid (0.05 ng/ml) for 6 h and for another 28 h in medium without colcemid before MTT cell proliferation assay. Viability was expressed as the relative to the control, i.e. cells received mock treatments. In case to see the effect of colcemid before UV (lane 3), the colcemid treatment was performed similarly unless colcemid as added to cells before UV-irradiation. (B) Microscopic views showing fewer mitotic cells (round shape) when colcemid was added after UV irradiation. Cells treated colcemid alone (top) or cells, after UV-irradiation, treated with colcemid (bottom) for 1 (left) or 6 h (right). (C) Flowcytometric analysis showing increase of G2/M peak in cells treated with colcemid alone but not in cells treated with colcemid after UV. G2/M peaks were marked with arrows. Time after UV was indicated (0–12 h).

H. Li et al. / Mutation Research 588 (2005) 118–128

121

Fig. 1. (Continued ).

3. Results 3.1. UV-caused cell growth inhibition is exacerbated by colcemid As mentioned earlier, colcemid enhances the UVCinduced apoptosis. The effect of colcemid could be recapitulated if the effect of colcemid on cell proliferation of UV-irradiated cells was examined as shown in Fig. 1A. The single treatments of colcemid and UV-irradiation suppressed cell growth by <10 and 50%, respectively (see lanes 2 and 4, Fig. 1A), while the combined treatment caused a significantly more cell growth inhibition (about 80%, lane 5 of Fig. 1A). In this combined treatment, treatment of colcemid was done after UV irradiation and only for a brief period, i.e. 6 h as compare to 34 h of total incubation period following UV irradiation before MTT assay. Colcemid probably did not exacerbate the growth suppression in the combined treatment via mitotic arrest. Although colcemid caused mitotic arrest in control cells (see the increase of rounded cells at 6 h, Fig. 1B, top panel), it did not exhibit similar effect in UV-irradiated cells (Fig. 1B, lower panel). The lack of increase of mitotic cells was also illustrated by the flowcytometric analysis of cell cycle distributions (Fig. 1C). The G2/M peak increased apparently in cells treated with colcemid alone during 1–12 h, while the same increase

did not occur in cells treated colcemid after UV irradiation (Fig. 1C). These data indicate that the enhancement of colcemid on growth inhibition of UV-irradiated cells was not due to its mitotic arrest. This suggests that the effect of colcemid treatment would be different if it was applied before UV irradiation. Indeed, when colcemid treatment was done before UV, the growth inhibition was significantly less than that when colcemid was done after UV (compare lanes 3 and 5, Fig. 1A). Thus, the enhancement effect by colcemid is apparently dependent on UV treatment. We suspected that colcemid might affect the repair of UV-induced DNA lesions. 3.2. Colcemid does not inhibit the excision of UV-induced DNA damages To see if colcemid affected the repair of UV-induced DNA damages, we performed ELISA of 6–4PP and CPD in genomic DNA of UV-irradiated cells at a period of time (0–8 h) after UV irradiation to examine the excision of UV-induced 6–4PP and CPD. The results shown in Fig. 2 indicate that colcemid had no effect on the excision of 6–4PP and CPD. By 8 h following irradiation, most of 6–4PP (>80%) had been removed by the nucleotide excision repair mechanism in cells regardless the presence of colcemid after UV treatment. In contrast, the repair of CPD was much slower. This is consistent

122

H. Li et al. / Mutation Research 588 (2005) 118–128

Fig. 2. Determination of 6–4PP and CPD after UV irradiation by using an ELISA. Cells (nearly confluent) were UV-irradiated at 25 J/m2 and then incubated in growth medium with or without colcemid. At the indicated period after irradiation, cells were harvested for genomic DNA isolation and subsequent ELISA for 6–4PP (panel A) and CPD (panel B). The ELISA data were converted to percentage of total 6–4PP or CPD remained relative to 0 h after irradiation and referred to the UVdose and DNA damage induction curve (data not shown).

with the previous reports that rodent cells are ineffective in repairing CPD. 3.3. Colcemid inhibits the rejoining following the excision of damaged DNA To know if colcemid affected the repair process after the excision of UV-induced DNA damage, we performed the comet assay to measure the DNA breaks in genome. The levels of DNA breaks in UV irradiated cells at 8 h after UV irradiation was slightly higher than control (compare lanes 1 and 3, Fig. 3A). Same was that in cells treated with colcemid alone (see lanes 1 and 2).

However, the levels of DNA breaks were much higher in cells treated with colcemid or aphidicolin following UV irradiation (compare lanes 2–5, Fig. 3A). Aphidicolin, an inhibitor of DNA polymerase ␦/␧, was used as a positive control. Thus, DNA breaks accumulated in UV irradiated cells when colcemid was present. This accumulation of DNA breaks may be due to inhibition of re-joining after nucleotide excision or DNA fragmentation due to apoptosis. To rule out the latter possibility, we performed DNA fragmentation assay as well as the apoptosis assay with cells of concern. No significant increase of DNA fragmentation or apoptosis was detected in cells treated with colcemid for 12 h after UV irradiation (see Fig. 3B and C). Thus, the accumulation of DNA breaks was not related to apoptosis. To test whether this accumulation was related to repair of UV-induced DNA damages, we performed the similar comet assay with UV5 and UV24 cells. UV5 and UV24 are known as NER defective strains of CHO cells. They were indeed demonstrated to fail to accumulate DNA breaks (indicated as comet moment) when they were treated with hydroxyurea/cytosine arabinoside (H/A) following UV treatment, whereas CHO.K1 showed apparent accumulation of DNA breaks in cells co-treated with UV and H/A (left panel, Fig. 4A). The treatment of H/A is commonly used for blocking DNA synthesis. More importantly, UV5 and UV24 showed little or no accumulation of DNA breaks when these cells were treated with colcemid after UV irradiation (right panel, Fig. 4A). In these comet assays, treatments of H/A or colcemid alone did not or only slightly increase comet moment (data not shown). Thus, the effect of colcemid on the accumulation of DNA breaks was NER dependent. To rule out the possibility that the effect of colcemid was related to semi-conservative DNA replication or mitosis, we performed the similar comet assay with quiescent cells or M-phase arrested cells. As shown in Fig. 4B, the effect of colcemid on accumulation of DNA breaks was also detected in non-replicating cells as well as in M-phase cells or in asynchronous cells (Figs. 3A and 4A). Thus, the effect of colcemid is not related to semi-conservative DNA replication or mitosis. To further test if colcemid inhibited the rejoining of DNA breaks induced by UV, we performed another comet assay in which DNA breaks were first accumulated by including H/A right after UV irradiation and then H/A was removed by replacing the growth medium with or without colcemid. In other words, cells after UV treatment were incubated with H/A for 6 h. As H/A inhibit rejoining, a large amount of DNA breaks would accumulate in cells. Then, H/A was removed to let rejoin-

H. Li et al. / Mutation Research 588 (2005) 118–128

123

Fig. 3. Increase of DNA breaks in cells treated with colcemid after UV-irradiation. (A) Comet assay. CHO.K1 cells were UV-irradiated (25 J/m2 ) and then incubated with or without colcemid (0.05 ␮g/ml) or aphidicolin (15 ␮M) for 8 h before comet assay. The * marks are for p < 0.01 with vs. without colcemid (or aphidicolin). (B) DNA fragmentation test and (C) flow-cytometric analysis of cell-cycle distribution showing no significant DNA ladders or apoptosis in cells treated with colcemid (0.05 ␮g/ml) for 12 h after UV irradiation (25 J/m2 ). Symbols in (C): L, live cells; A, cells in early apoptosis; D, dead cells.

ing resume. Thus, if there was no other interference, the DNA breaks would decline. This was indicated by the curve without colcemid shown in Fig. 5 (curve of open circles). Without colcemid, the comet moment declined markedly within 8 h after removal of H/A. However, in the presence of colcemid the comet moment only declined slightly during the same period of time (Fig. 5, curve of close circles). The results shown in Fig. 5 clearly indicate that colcemid inhibited the rejoining of DNA breaks induced by UV in CHO.K1 cells. Taken together, colcemid may inhibit NER by inhibiting the step(s) of gap-filling and/or ligation.

3.4. Colcemid does not inhibit the gap-filling of nucleotide excision repair To know if colcemid affected the step of gap-filling of nucleotide excision repair pathways, we performed the unscheduled DNA synthesis (UDS) assay. The incorporation of thymidine analogue, bromodeoxyuridine (BrdU) during the gap-filling in UV irradiated cells was examined under fluorescence microscope. The immunostaining of UDS showed punctate spots in nuclei, in contrast to the uniformly and intensive staining found in ordinary DNA synthesis. The nuclei with punc-

124

H. Li et al. / Mutation Research 588 (2005) 118–128

Fig. 4. Accumulation of UV-induced DNA breaks in cells treated with colcemid post UV-irradiation was related to NER but not cell cycle. (A) Little or no accumulation of DNA breaks in UV24 or UV5 cells treated with colcemid after UV. (Left panel) Comet assay to show that UV24 and UV5 were NER defective. Comet assay was done in cells treated with or without H/A (hydroxyurea at 10 mM and cytosine ␤-d-arabinofuranoside at 100 ␮M) for 4 h after UV irradiation (25 J/m2 ). (Right panel) Accumulation of DNA breaks in CHO.K1 but not in UV24 or UV5 cells treated with colcemid (0.05 ␮g/ml) for 12 h after UV(25 J/m2 ). (B) Accumulation of DNA breaks in Go or mitotic CHO.K1 cells treated with colcemid after UV. To obtain Go cells or M-phase cells, cells at 80–90% confluence were serum-starved for 2–3 days or incubated with colcemid (0.5 ␮g/ml) for 24 h, respectively, before use. Cells were largely at Go or M-phase, respectively, according to the histograms of flow-cytometric analysis (data not shown). Comet assay was done similarly as in (A, right panel). Cells were treated with colcemid (0.05 ␮g/ml) for 12 h after UV irradiation (25 J/m2 ) before comet assay. The * marks are for p < 0.01 with vs. without colcemid.

tate immunostaining were only present in UV-irradiated cells as compared to the untreated cells (Fig. 6 and data not shown). Also, UV-irradiated UV24 cells did not show the punctate immunostaining (data not shown). Aphidicolin, the DNA polymerase inhibitor almost completely

suppressed the UDS, whereas colcemid did not (Fig. 6). In terms of the percent of UDS nuclei, there is no significant difference if colcemid was present or not after UV irradiation (24.8 ± 4.2% versus 21.0 ± 3.8%). Thus, colcemid did not affect the step of gap-filling in NER.

H. Li et al. / Mutation Research 588 (2005) 118–128

Fig. 5. Inhibition of rejoining of DNA breaks by colcemid. CHO.K1 cells of overnight culture were treated H/A (same as in Fig. 4) for 6 h following UV irradiation (25 J/m2 ), then H/A was removed from the growth medium and the cells were incubated in the medium with (curve of close circle) or without (curve of open circle) colcemid (0.05 ␮g/ml) for the indicated time before comet assay.

The data shown in Fig. 6 was done with M-phase cells, and similar results were obtained with quiescent cells or asynchronous cells (data not shown). In summary of the present data of UDS study and the previous results

Fig. 6. Unscheduled DNA synthesis in cells treated with or without colcemid after UV. Cells were seeded to 3.5 cm-dishes at same density and cultured for 1–2 days before incubated with colcemid for 24 h to become arrested at M-phase. The M-phase cells were irradiated with UV (25 J/m2 ) or mock-treated. Following UV treatment, cells were cultured in BrdU (10 ␮M) containing medium with or without colcemid (0.5 ␮g/ml) or aphidicolin (30 ␮M) for 1–2 h before immunostaining procedures. Nuclei were counterstained with DAPI (blue).

125

Fig. 7. Synergistic effect of colcemid on cell growth suppression due to UV irradiation or cisplatin but not X-ray irradiation. Same numbers of cells were seeded and grown to 50% confluency before UV-irradiation (25 J/m2 ; lanes 3 and 4) or X-ray irradiation (16 Gy; lanes 5 and 6) or cisplatin (12.5 ␮M; lanes 7 and 8). The cells were recovered in media with or without colcemid (0.05 ␮g/ml) for 6 h and then continually grown in media without colcemid for another 30 h. Then, the cells were harvested for cell counts. Data were expressed as the relative to the control (lane 1). The * marks are for p < 0.01 with vs. without colcemid.

from comet assay, we propose that colcemid inhibits the ligation of UV induced NER in CHO.K1 cells (Fig. 8). 3.5. Colcemid enhances the cell growth inhibition (or cell death) by UV or cisplatin but not X-ray To know if the effect of colcemid is specific to nucleotide excision repair as well as to link the previous mechanistic study to the potential application in chemotherapy, we examined the effect of colcemid on cell growth of cells treated with cisplatin. Cisplatin, known as an anticancer drug, causes DNA damages which also trigger the nucleotide excision mechanism in cells for DNA repair. As a comparison, we also studied the effect of colcemid on X-ray irradiated cells. X-ray causes DNA strand breaks and involves repair mechanisms different from nucleotide excision repair. For this study, the dose of each specific treatment that causes about 50% growth inhibition was used. We found that 25 J/m2 of UV, 16 Gy of X-ray and 12.5 ␮M cisplatin, respectively, caused 50% of growth inhibition in CHO.K1 cells (data not shown). The results of this study shown in Fig. 7 indicate that colcemid enhanced the growth-inhibition by UV-irradiation or cisplatin but not X-ray irradiation (see lanes 3–8, Fig. 7). We also examined the effect of colcemid on cell death. The results

126

H. Li et al. / Mutation Research 588 (2005) 118–128

Fig. 8. Model of nucleotide excision repair and the plausible process affected by colcemid. Colcemid probably inhibits the ligation step.

indicate that the cell death caused by UV or cisplatin but not X-ray was enhanced by colcemid (data not shown). Thus, the effect of colcemid seems to specific to NER. 4. Discussion In our previous study, we found that UVC-induced apoptosis in CHO.K1 cells and the cell death was exacerbated by colcemid [5,6]. Colcemid makes the cells more sensitive to UV-irradiation. In this study, we linked this effect of colcemid to the repair of UV-induced DNA damages. First, we found a brief treatment of colcemid had a synergistic effect on growth inhibition by UV (Fig. 1A). Consistent with our previous observation that mitotic cells are not more UV sensitive than cells of other phases [6], the present study indicates that the effect of colcemid was not related to the mitotic arrest. No significant increase of M-phase cells was detected in cells treated with colcemid post UV-irradiation (Fig. 1B and C). As reported previously by others, UV-irradiation could delay cell progression of S phase in the hamster cells [16], and this delay was likely to put off the mitotic arrest by colcemid. Furthermore, our data indicate that the effect of colcemid was obtained only when colcemid was applied after the UV treatment (Fig. 1A), suggesting that the colcemid effect relied upon UV. Thus, we suspected the effect of colcemid might be on the repair of DNA damages. We examined the exci-

sion of UV-induced 6–4PP and CPD, and found that colcemid had no effect on nucleotide excision (Fig. 2). However, colcemid accumulated the DNA breaks when it was added to cells following UV-irradiation (Fig. 3A); the accumulation of DNA breaks was not the result of apoptosis due to DNA fragmentation (Fig. 3B and C). Furthermore, the effect of colcemid required NER since it was not found with NER defective cells (Fig. 4A). The effect of colcemid was not related to regular DNA replication or cell-cycle since the effect was detected in nonreplicating cells (Fig. 4B). More importantly, colcemid inhibited the rejoining of DNA breaks accumulated by hydroxyurea/cytosine arabinoside following UV irradiation (Fig. 5). These results, in conjunction with the observation that colcemid did not affect the unscheduled DNA synthesis (Fig. 6), suggest that colcemid may inhibit the ligation step of NER (Fig. 8). In consistence with the above conclusion, we made an observation from an in-vitro experiment: DNA was isolated from the cells with care to avoid breakage. The DNA samples were then treated with terminal transferase. If the DNA sample had free 3 ends, the ends would be labeled by terminal transferase with digoxigenin-ddUTP. We found that DNA from cells co-treated with UV and colcemid had more free 3 ends than that from cells treated UV alone. In this experiment, DNA from the untreated cells was only barely labeled. More importantly, the 3 ends of DNA from cells co-treated with UV and colcemid was greatly reduced if the DNA was pre-treated with T4 ligase, suggesting that the ends could be ligated. Thus, colcemid indeed affected ligation. Our data indicate that colcemid exacerbated the growth inhibition or cell death caused by UV or cisplatin but not X-ray (Fig. 7). We related this to the effect of colcemid on NER because both UV and cisplatin but not X-ray involving NER. The ligation step of NER is specifically carried out by ligase I. At this moment, we do not know how colcemid might inhibit the activity of ligase I. Nevertheless, there are reports indicate that DNA repair proteins such as rad21 of fission yeast, murine DNA glycosylase NEIL2 and human DNA polymerase ␤ are microtubule-binding proteins [17,18]. It has been known that microtubules involve not only the spindle necessary for mitosis but also the trafficking of macromolecules [18,19]. The plausible microtubule-binding of DNA repair proteins may have the advantages of the active regulation as well as preventing the dilution of the activity of the repair proteins. As we found that the effect of colcemid upon UV-induced cell death was only detected at 50 ng/ml or higher and the 50 ng/ml is enough to cause M-phase arrest (data not shown), we speculate that the effect of colcemid was related to its

H. Li et al. / Mutation Research 588 (2005) 118–128

anti-microtubule activity. Colcemid may either affect the recruitment of ligase I or prevent the departure of the gap-filling proteins such as DNA polymerase ␦ and ␧, PCNA, RFC to allow the ligase I to access the sites to seal the ends. We are currently investigating the possibility by using the UV-micropore irradiation developed by Mori and co-workers [20,21]. On the other hand, the effect of colcemid may be subjected to the modulation by other factors such as p21(waf1/cip1), the general inhibitor of cyclin dependent kinases. Previously, we have reported that p21 is not detected in UV-irradiated CHO.K1 cells and the ectopic expression of human p21 gene was able to suppress the UV-induced apoptosis even in the presence of colcemid [6,22]. In accordance with this, the inhibitory effect of colcemid on rejoining of DNA breaks accumulated by H/A was not detected in UV-irradiated mouse fibroblast cells (NIH3T3) (data not shown). In contrast to CHO.K1 cells, NIH3T3 cells express p21 and the p21 expression was greatly elevated by UV-irradiation; colcemid had no synergistic effect on UV-induced cell death in NIH3T3 cells. Thus, we consider that the effect of colcemid would not take place if p21 was present. Whether p21 acted through its effect of cell cycle arrest or its direct involvement in NER or remains unclear. In conclusion, colcemid inhibited the rejoining of nucleotide excision repair in UV-irradiated CHO-K1 cells. This inhibition is probably through the ligation step of NER. The effect of colcemid may be exploited for cancer chemotherapy involving nucleotide excision repair such as cisplatin. Acknowledgements This work was supported by grants from the National Science Council of Taiwan (NSC 92-23 11-B-007013) and National Research Program for Genomic Medicine project (NSC 92-3112-B-007-004). We thank Drs. T. Mori and T. Iwamoto (Nara Medical University, Kashihara, Nara, Japan) for the antibodies (64 M-2; TDM-2) and their technical assistance for measuring the 6–4PP and CPD with ELISA. We also thank Dr. K.Y. Jan (Academia Sinica, Taipei) for his critical review and valuable suggestions. References [1] D. Subramanian, B.S. Rosenstein, M.T. Muller, Ultravioletinduced DNA damage stimulates topoisomerase I-DNA complex formation in vivo: possible relationship with DNA repair, Cancer Res. 58 (1998) 976–984. [2] S. Tornaletti, G.P. Pfeifer, Slow repair of pyrimidine dimers at p53 mutation hotspots in skin cancer, Science 263 (1994) 1436–1438.

127

[3] A. Sancar, L.A. Lindsey-Boltz, K. Unsal-Kacmaz, S. Linn, Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints, Annu. Rev. Biochem. 73 (2004) 39–85. [4] C.J. Norbury, B. Zhivotovsky, DNA damage-induced apoptosis, Oncogene 23 (2004) 2797–2808. [5] B.S. Tzang, Y.C. Lai, Y.C. Liu, UV-induced but P53 independent apoptotic death in CHO.K1 cells is promoted by M phase inhibitors, In Vitro Cell Dev. Biol. Anim. 35 (1999) 17–18. [6] B.S. Tzang, Y.J. Chiang, H.C. Lan, C.B. Liao, Y.C. Liu, Tuning up or down the UV-induced apoptosis in Chinese hamster ovary cells with cell cycle inhibitors, Photochem. Photobiol. 75 (2002) 662–667. [7] X. Shang, Y. Shiono, Y. Fujita, S. Oka, Y. Yamazaki, Synergistic enhancement of apoptosis by DNA- and cytoskeleton-damaging agents: a basis for combination chemotherapy of cancer, Anticancer Res. 21 (2001) 2585–2589. [8] M. Tsuruga, Y. Dang, Y. Shiono, S. Oka, Y. Yamazaki, Differential effects of vitamin E and three hydrophilic antioxidants on the actinomycin D-induced and colcemid-accelerated apoptosis in human leukemia CMK-7 cell line, Mol. Cell Biochem. 250 (2003) 131–137. [9] T. Mori, M. Nakane, T. Hattori, T. Matsunaga, M. Ihara, O. Nikaido, 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. 54 (1991) 225–232. [10] K. Imoto, N. Kobayashi, S. Katsumi, Y. Nishiwaki, T.A. Iwamoto, A. Yamamoto, Y. Yamashina, T. Shirai, S. Miyagawa, Y. Dohi, S. Sugiura, T. Mori, The total amount of DNA damage determines ultraviolet-radiation-induced cytotoxicity after uniformor localized irradiation of human cells, J. Invest. Dermatol. 119 (2002) 1177–1182. [11] N.P. Singh, M.T. McCoy, R.R. Tice, E.L. Schneider, A simple technique for quantitation of low levels of DNA damage in individual cells, Exp. Cell Res. 175 (1988) 184–191. [12] R.R. Tice, E. Agurell, D. Anderson, B. Burlinson, A. Hartmann, H. Kobayashi, Y. Miyamae, E. Rojas, J.C. Ryu, Y.F. Sasaki, Single cell gel/comet assay: guidelines for in vitro and in vivo genetic toxicology testing, Environ. Mol. Mutagen. 35 (2000) 206–221. [13] A. Nakagawa, N. Kobayashi, T. Muramatsu, Y. Yamashina, T. Shirai, M.W. Hashimoto, M. Ikenaga, T. Mori, Three-dimensional visualization of ultraviolet-induced DNA damage and its repair in human cell nuclei, J. Invest. Dermatol. 110 (1998) 143–148. [14] S. Kondo, G.H. Barnett, H. Hara, T. Morimura, J. Takeuchi, MDM2 protein confers the resistance of a human glioblastoma cell line to cisplatin-induced apoptosis, Oncogene 10 (1995) 2001–2006. [15] I. Vermes, C. Haanen, H. Steffens-Nakken, C. Reutelingsperger, A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V, J. Immunol. Methods 184 (1995) 39–51. [16] D.K. Orren, L.N. Petersen, V.A. Bohr, Persistent DNA damage inhibits S-phase and G2 progression, and results in apoptosis, Mol. Biol. Cell 8 (1997) 1129–1142. [17] K. Tatebayashi, J. Kato, H. Ikeda, Isolation of a Schizosaccharomyces pombe rad21ts mutant that is aberrant in chromosome segregation, microtubule function, DNA repair and sensitive to hydroxyurea: possible involvement of Rad21 in ubiquitinmediated proteolysis, Genetics 148 (1998) 49–57. [18] K.A. Conlon, H. Miller, T.A. Rosenquist, D.O. Zharkov, M. Berrios, The murine DNA glycosylase NEIL2 (mNEIL2) and

128

H. Li et al. / Mutation Research 588 (2005) 118–128

human DNA polymerase beta bind microtubules in situ and in vitro, DNA Repair (Amst.) 4 (2005) 419–431. [19] P. Giannakakou, D.L. Sackett, Y. Ward, K.R. Webster, M.V. Blagosklonny, T. Fojo, p53 is associated with cellular microtubules and is transported to the nucleus by dynein, Nat. Cell Biol. 2 (2000) 709–717. [20] S. Katsumi, N. Kobayashi, K. Imoto, A. Nakagawa, Y. Yamashina, T. Muramatsu, T. Shirai, S. Miyagawa, S. Sugiura, F. Hanaoka, T. Matsunaga, O. Nikaido, T. Mori, In situ visualization of ultraviolet-light-induced DNA damage repair in locally irradi-

ated human fibroblasts, J. Invest. Dermatol. 117 (2001) 1156– 1161. [21] M. Wakasugi, A. Kawashima, H. Morioka, S. Linn, A. Sancar, T. Mori, O. Nikaido, T. Matsunaga, DDB accumulates at DNA damage sites immediately after UV irradiation and directly stimulates nucleotide excision repair, J. Biol. Chem. 277 (2002) 1637–1640. [22] B.S. Tzang, Y.C. Lai, M. Hsu, H.W. Chang, C.C. Chang, P.C. Huang, Y.C. Liu, Function and sequence analyses of tumor suppressor gene p53 of CHO.K1 cells, DNA Cell Biol. 18 (1999) 315–321.