HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kao, G. D. Articles by Muschel, R. J. Search for Related Content PubMed PubMed Citation Articles by Kao, G. D. Articles by Muschel, R. J. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 49, 34779-34784, December 3, 1999

p34^Cdc2 Kinase Activity Is Excluded from the Nucleus during the Radiation-induced G₂ Arrest in HeLa Cells^*

Gary D. Kao, W. Gillies McKenna, and Ruth J. Muschel^§

From the Department of Radiation Oncology and  Department of Pathology and Lab Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The progression of cells from G₂ into mitosis is blocked by exposure to DNA-damaging agents such as ionizing radiation. This G₂ delay is associated with reduced cyclin B1-specific associated histone H1 kinase activity, increased inhibitory phosphorylation of p34^Cdc2, and depressed cyclin B1 levels in HeLa cells. Induction of cyclin B1 or expression of Cdc2AF, a mutant p34^Cdc2 that lacks the sites of inhibitory phosphorylation, only partially reverses the radiation-associated G₂ delay, although both maneuvers rapidly result in increased histone H1 kinase activity. To account for the persistent G₂ delay in the face of active p34^Cdc2 kinase, we determined the location of the kinase activity. Although p34^Cdc2 was active in the cytoplasm, the nuclear p34^Cdc2 was inactive. Irradiation led to nuclear accumulation of the inactive tyrosine-phosphorylated form of p34^Cdc2, whereas the active form was seen in the cytoplasm. At later times when cells had resumed cell cycle progression, nuclear kinase activity was detectable. These results give evidence of segregation of cytoplasmic and nuclear kinase activity after DNA damage that has the effect of enhancing checkpoint control. Shielding the nucleus from the potentially deleterious effects of kinase activity after DNA damage may help irradiated human cancer cells respond to irradiation.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

After exposure to ionizing radiation, eukaryotic cells arrest at various points in the cell cycle (1-3). The most widely studied of these checkpoints are those that are seen in the G₁ and G₂ phases of the cycle (4, 5). The G₁ delay induced by ionizing radiation is entirely dependent upon p53, which directs the transcription of p21waf1/cip1/sdi1 (6-8, 10). The G₂ block, however, appears to result from multiple mechanisms. Induction of p53 can lead to a G₂ arrest (9, 25); however, many malignant cells lack p53 function, as demonstrated by the absence of a G₁ arrest after radiation exposure, yet these cells almost universally retain a G₂ block. For example HeLa cells, which lack functional wild type p53, show no induction of p21 protein after exposure to ionizing radiation and no G₁ checkpoint but have a dose-dependent G₂ delay (11, 12). Neither p21 nor p53 is required for the radiation-induced G₂ delay, as cells derived from p21 or p53 knockout mice or cells in which p53 is inactivated show equivalent G₂ delays after radiation (13-16).

Other factors have been implicated as components of the G₂ block. Radiation results in transiently decreased p34^Cdc2/cyclin B1 kinase activity (17-24). The absolute levels of p34^Cdc2 protein are not altered after radiation (18, 22), and thus, the decreased activity is regulated both through levels of cyclin B1 as well as the inhibitory phosphorylation of p34^Cdc2/cyclin B complex. We have previously shown that radiation results in cyclin B1 mRNA instability and decreased cyclin B1 protein levels, which would prevent activation of p34^Cdc2 (4, 32, 40). We also found that restoration of cyclin B1 expression after radiation can only partially abrogate the G₂ delay (26). This indicates that there is a contribution of decreased cyclin B1 to the G₂ delay, but that other factors must also be at play.

The contribution of the inhibitory phosphorylations of p34^Cdc2 to the G₂ arrest has also been evaluated. Jin et al. (38) induced expression of Cdc2AF, a mutant form of Cdc2 that cannot undergo inhibitory phosphorylation. This resulted in only partial abrogation of the G₂ delay, despite markedly increased p34^Cdc2 activity. The phosphorylation of p34^Cdc2 is regulated in part by the phosphatase Cdc25C, whose activity in turn is regulated by phosphorylation. Radiation induces chk1 activity in yeast, which in turn inhibits Cdc25C (63). A number of mammalian chk1 homologs have been identified that interact with Cdc25C (43, 61). Therefore, neither the ability to phosphorylate p34^Cdc2 nor the regulation of the levels of cyclin B1 individually appears to control the G₂ delay.

Nuclear localization of the cyclin B1/p34^Cdc2 complex has also been implicated both as a component of normal mitosis and as a component of the G₂ delay (27, 29). The nuclear localization of cyclin B1 in itself does not appear to be the sole determinant of progression into mitosis. The cell cycle delay is not abolished when cyclin B1 is introduced into the nucleus via microinjection, expression of nuclear localization signal-tagged cyclin B1 protein, or treatment with leptomycin B (which disrupts cyclin B1-CRM1 interaction) (30). The delay was, however, abolished by leptomycin B plus caffeine. We have shown that caffeine abolishes the radiation-induced suppression of cyclin B expression (31), but caffeine also has promiscuous effects on protein phosphorylation, which may also contribute to its cell cycle effects. It has been recently reported that expression of nuclear localization signal-tagged cyclin B1 and Cdc2AF together led to nuclear envelope dissolution, accelerating the normal cycle and the G₂ delay. In this case, however, normal cycling did not result but, instead, a mitotic catastrophe, leaving the question of whether this procedure resembled normal physiological processes.

We report here our efforts in investigating mechanisms underlying the G₂ delay after radiation. We find that despite cyclin B1 induction and the presence of significant levels of cyclin B1-associated kinase activity, cells remain partially blocked in G₂. We find that when cells are blocked in G₂ after radiation, the cyclin B1-associated kinase activity is localized in the cytoplasm for a substantial period of time. Nuclear kinase activity is evident only when cells exit the G₂ delay. We find that expression of both cyclin B1 and Cdc2AF together increases cytoplasmic kinase activity but is still insufficient to circumvent the cellular compartmentalization of kinase activity and completely remove the G₂ delay. These results help explain why induced expression of cyclin B1 or Cdc2AF protein were not successful in completely abrogating the G₂ delay after irradiation. These observations together highlight the presence of multiple distinct but interrelated mechanisms of regulation of cell cycle progression after exposure to DNA-damaging agents.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell Culture and Synchronization-- Experiments described here were performed with S21 cells (described in Ref. 26), which are a HeLa clone stably transfected with a dexamethasone-inducible expression vector encoding cyclin B1. Cells were grown and synchronized with sequential thymidine and aphidicolin blocks and irradiated as described previously (27). Radiation was delivered using a high dose rate cesium irradiator (12.84 Gy/min,¹ confirmed by Fricke dosimetry). Control cells were always mock-irradiated. Dexamethasone when added was at 1 µg/ml and 1 h after irradiation.

For experiments involving transfections, 10 µg of plasmid (pECE-Cdc2 (cDNA coding for coding for wild type Cdc2 protein) or pECE-Cdc2AF (p34^Cdc2 with mutations replacing the threonine 14 and tyrosine 15 with alanine and phenylalanine, respectively), both kindly provided by Dr. F. McKeon) was used for transfection of 1 × 10⁶ cells with LipofectAMINE (Life Technologies, Inc.) following the manufacturer's instructions. LipofectAMINE with the transfected plasmid was applied in serum-free medium. After 4-6 h, aphidicolin (2 µg/ml) was added in serum-containing medium for 12 h. The aphidicolin was washed out, and cells were then irradiated 3 h after the removal of aphidicolin. One h after irradiation, dexamethasone was added (1 µg/ml). In the experiments using caffeine (2 mM), it was added at the time of removal of the aphidicolin 3 h before irradiation.

Immunoblotting and Protein Expression-- For immunoblotting, 100,000 cells were harvested in 400 µl of sample buffer (10% glycerol, 2% SDS, 100 mM dithiothreitol, 50 mM Tris, pH 6.8). 30 µl of each lysate was loaded per lane and separated by electrophoresis on a 15% SDS-polyacrylamide gel before overnight transfer to a nitrocellulose membrane (Bio-Rad). The cyclin B1 (Upstate Biotechnology, Inc.), -tubulin (Sigma), or phosphotyrosine p34^Cdc2 antibodies (New England Biolabs) were used at 1:1000. Polyclonal anti-ataxia-telangiectasia protein (ATM) antibodies was kindly provided by Dr. T. Yen. Incubation with the secondary antibody, HRP-conjugated goat anti-mouse antibody, 1:2000 (Roche Molecular Biochemicals), was performed in a solution of 2.5% powdered milk in phosphate-buffered saline. Detection was performed by chemoilluminescence (ECL, Amersham Pharmacia Biotech). Densitometry was analyzed by NIH image software.

Nocodazole Trapping-- Cell cycle progression into mitosis was analyzed via nocodazole trapping of synchronized cells as described in (26, 22). Briefly, for each time point, nocodazole (0.04 µg/ml) was added to duplicate plates 2-3 h after release from the aphidicolin block. Thus the number of mitotic cells represent the cumulative total of all cells that have entered into mitosis at that time. At least 100 cells were evaluated for mitosis after staining with propidium iodide using the Zeiss Axioplan microscope.

Cell Fractionation-- Cell fractionation was performed via a modification of methods as described by Schreiber et al. (33). Briefly, 1 million cells grown on 10-cm tissue culture dishes (Fisher) were harvested by scraping with a rubber policeman. Cell pellets were resuspended in 200 µl of hypotonic buffer (10 mM HEPES, pH7.9, 10 mM KCL, 0.1 mM EDTA, 0.1 mM EGTA) incubated at 4 °C for 10 min, and Nonidet P-40 added to a final concentration of 0.5%. Samples were vortexed ×10 s and centrifuged at 800 rpm × 5 min to collect the cytoplasmic fraction. The nuclear pellet was resuspended and incubated in 200 µl of nuclear extract buffer (20 mM HEPES, pH 7.9, 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 20% glycerol) for 30 min before centrifugation at 2000 rpm × 10 min to collect the nuclear fraction.

Histone H1 Kinase Assay-- Immunoprecipitation of cyclin B1-specific histone H1 kinase (p34^Cdc2) activity was performed using anti-cyclin B1 antibody (Upstate Biotechnology) as described by Pines and Hunter (34). Immunoprecipitated samples were preincubated with HB buffer (25 mM MOPS, pH 7.2, 60 mM -glycerophosphate, 15 mM p-nitrophenylphosphate, 15 mM MgCl₂, 15 mM EGTA, 1 mM dithiothreitol, 0.1 mM sodium vanadate, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, 20 µg/ml leupeptin, 40 µg/ml aprotonin) at 37 °C for 5 min. 10 µl of KIN buffer (1 mg/ml histone H1 (Roche Molecular Biochemicals), 200 µM -labeled [³²P]ATP in HB buffer) was then added and incubated for a further 20 min. Finally, 2× lysis buffer was added, and samples were boiled for 5 min and separated by 12% SDS/polyacrylamide gel electrophoresis and detected by autoradiography. Densitometry was performed with NIH Image software.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of Induction of Cyclin B1 on p34^Cdc2 Activity in Irradiated Cells-- To ask whether the depression of cyclin B1 seen after irradiation in HeLa cells contributes to the G₂ delay, we had previously developed a HeLa cell clone (S21) carrying a dexamethasone-inducible vector for cyclin B1 (26). Dexamethasone itself affected neither cyclin B1 levels nor the cell cycle delay in the absence of the expression vector. Treatment of stable clones carrying the inducible vector with dexamethasone leads to elevated levels of cyclin B1 as soon as 2 h after addition. Induction of cyclin B1 in unirradiated cells had no discernable effect on the timing of G₂, whereas restoration of cyclin B1 levels in irradiated cells shortened but did not eliminate the G₂ delay.

To investigate the effect of induction of cyclin B1 on p34^Cdc2 kinase activity, we synchronized the S21 cells with sequential thymidine and aphidicolin blocks (see "Materials and Methods"), and 2 h after release from the aphidicolin block, irradiated with 3 Gy. One h after irradiation, dexamethasone was added to induce cyclin B1. Irradiation markedly diminished cyclin B1-immunoprecipitable p34^Cdc2 histone H1 kinase activity (Fig. 1A, compare irradiated versus unirradiated samples, 11 (lane 4 versus lane 12) and 13 (lane 6 versus lane 13) h after release from the aphidicolin block). The addition of dexamethasone increased cyclin B1 levels (Fig. 1A) and restored cyclin B1-p34^Cdc2 kinase activity (lane 3). Despite the restoration of cyclin B1-p34^Cdc2 kinase activity, however, cell cycle progression was only partially accelerated. At the 15-h time point, most of the unirradiated control cells (79%) had entered mitosis, compared with only 2% of the irradiated cells and 19% of the irradiated cells with induced cyclin B1. These cumulative mitotic counts (Fig. 1B) indicated that the cell cycle delay after irradiation was diminished but not abolished by the induced expression of cyclin B1, despite overall levels of cyclin B1-p34^Cdc2 kinase activity that approximated levels in unirradiated control cells.

View larger version (39K): [in this window] [in a new window]   Fig. 1.   p34Cdc2 activity and cell cycle progression after irradiation (IR) and induced expression of cyclin B1. S21 cells containing the dexamethasone (Dex)-inducible cyclin B1 expression vector were synchronized sequentially with thymidine and aphidicolin, released from the aphidicolin block, and irradiated 3 h after release (3 Gy) at a time when 90% of the cells were in S phase. Dexamethasone or control media was added 1 h after irradiation. Cells were harvested at the indicated times after release from the aphidicolin block (time = 0 h) and analyzed for cyclin B1-p34Cdc2 histone H1 kinase activity, cyclin B1 protein, and cell cycle progression. The cyclin B1 immunoblot was stripped and reprobed for tubulin protein as a loading control. A shows the autoradiograph of the phosphorylated histone H1 indicating p34Cdc2 kinase activity and the immunoblots of cyclin B1 and tubulin. B shows the cumulative mitotic counts performed in the presence of nocodazole to prevent cells from re-entering G1 ("nocodazole trapping"). Con, control.

p34^Cdc2 Kinase Activity in Cytoplasmic and Nuclear Fractions-- To understand why cyclin B1-p34^Cdc2 kinase activity might be ineffective in abolishing the G₂ block, we measured the kinase activity after separation into cytoplasmic and nuclear fractions (Fig. 2). S21 cells were synchronized and harvested either at the G₁/S interface, in G₂, unirradiated in G₂, or blocked in metaphase with nocodazole. Verification of the separation between the nuclear and the cytoplasmic fractions was achieved by immunoblotting parallel samples with anti--tubulin (a cytoplasmic protein) or anti-ataxia-telangiectasia protein (a nuclear protein) (36). Nuclear and cytoplasmic compartments will be mixed in mitosis due to the dissolution of the nuclear envelope, and mitotic cells arrested by nocodazole had both markers in each fraction. There was little contamination of the nuclear fraction with cytoplasm, although trace nuclear material was found in the cytoplasm under the conditions used. After irradiation, p34^Cdc2 kinase activity was detectable only in the cytoplasm of the irradiated cells, not in the nucleus, and was diminished in amount compared with that of the unirradiated cells. In unirradiated G₂ cells, p34^Cdc2 kinase activity was detectable in both the cytoplasm and nucleus, whereas activity was not detected in either compartment in G₁/S cells.

View larger version (45K): [in this window] [in a new window]   Fig. 2.   Cytoplasmic and nuclear p34Cdc2 kinase activity after irradiation. S21 cells were synchronized with thymidine and aphidicolin sequential blocks. Cells were harvested at the G1/S interface (G1/S) immediately after the final removal of aphidicolin. Irradiated (IR) cells were synchronized and irradiated (IR, 10 Gy) 3 h after release from aphidicolin and harvested 12 h after release. Unirradiated G2 cells (G2) were synchronized, mock-irradiated, and harvested 12 h after release when the majority were in G2. Cells were placed into nocodazole (Noco) treatment for 12 h after release from the thymidine block at the same time as aphidicolin would have been applied. All treatments were scheduled in order that all cells were harvested and analyzed for kinase activity and protein at the same time. All samples were fractionated into cytoplasmic and nuclear fractions using the methods of Schreiber et al. (33). Parallel samples were assayed for p34Cdc2 kinase activity and immunoblotting. Samples were probed for tubulin and ATM protein, as respectively cytoplasmic and nuclear markers.

We evaluated the effect of cyclin B1 induction on kinase activity in irradiated cells (Fig. 3). At each time point after irradiation, parallel samples were harvested for analysis of kinase activity after fractionation into cytoplasmic and nuclear components and analysis of cell cycle progression. Cyclin B1-p34^Cdc2 kinase activity was reduced in irradiated cells compared with the unirradiated cells. Kinase activity in the nucleus was not seen in the irradiated cells (Fig. 3A, compare lanes 5 and 7). At 15 h, cytoplasmic cyclin B1-p34^Cdc2 kinase activity in the irradiated cells was equivalent to that in unirradiated cells, but nuclear kinase activity was still depressed. Induction of cyclin B1 resulted in levels of cytoplasmic p34^Cdc2 kinase activity that exceeded the levels in unirradiated control cells, yet the nuclear p34^Cdc2 kinase activity was still depressed (compare activity in lanes 3 and 5). Fig. 3B shows in graphical form the densitometry values of the p34^Cdc2 kinase assays shown in Fig. 3A, showing the dissociation of cytoplasmic from nuclear kinase activity after irradiation and showing that nuclear, not cytoplasmic, kinase activity correlated with progression into mitosis.

View larger version (37K): [in this window] [in a new window]   Fig. 3.   Effect of cyclin B1 induction on cytoplasmic (Cyto) and nuclear (Nuc) p34Cdc2 kinase activity. S21 cells were synchronized and irradiated (IR) as described in Fig. 1. Dexamethasone (Dex) was added at the time of irradiation. Synchronized cells harvested at the indicated times after release were fractionated into cytoplasmic and nuclear fractions and analyzed for cyclin B1-p34Cdc2 activity, whereas duplicate samples were analyzed for cell cycle progression. A shows the autoradiograph of phosphorylated histone H1. B shows densitometry values of autoradiographs of the kinase activity and the proportion of cells progressing into mitosis measured by nocodazole.

Effect of Expression of Cdc2AF and Cyclin B1 Protein-- Cdc2AF is a mutant p34^Cdc2 protein that lacks the inhibitory phosphorylation sites at positions 14 and 15 (37-39). We irradiated synchronized cells that had been transfected with an expression vector for Cdc2AF. As before, radiation resulted in markedly decreased cytoplasmic cyclin B1-p34^Cdc2 activity (Fig. 4A, lanes 1 and 2) and no detectable nuclear cyclin B1-p34^Cdc2 activity in the control cells. The induced expression of either cyclin B1 or Cdc2AF led to increased cytoplasmic kinase activity but did not effect an increase in nuclear activity (lanes 3 and 5). The induced expression of cyclin B1 and Cdc2AF together led to a greater increase of cyclin B1-p34^Cdc2 activity than either alone, but still there was no increase in nuclear activity (lane 4). Transfection of a vector coding for wild type p34^Cdc2 protein did not increase the cytoplasmic or the nuclear cyclin B1-p34^Cdc2 activity. Dexamethasone alone had a negligible effect on nuclear cyclin B1-p34^Cdc2 activity, consistent with its lack of effect on cell cycle progression (Ref. 26 and data not shown).

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Compartmentalization of p34Cdc2 kinase activity after transfection of Cdc2AF or wild type Cdc2 and induction of cyclin B1 expression. Cells were transfected with pECE Cdc2AF (CDC2AF) or pECE wild type Cdc2 (CDC2) or mock-transfected as described under "Materials and Methods." 4-6 h after the initiation of transfection, aphidicolin (2 µg/ml) was applied for 12 h to partially synchronize the cells and then washed out. Irradiation (IR, 3 Gy) was performed 3 h after release from aphidicolin, and samples were harvested 6 h after irradiation. Caffeine was added immediately before irradiation (Caff), whereas dexamethasone (Dex) was added 1 h after irradiation. Cells were fractionated, harvested, and assayed for p34Cdc2 kinase activity as described previously. All samples were harvested at the same time. The treatment groups are as indicated to the right of the autoradiographs. A and B show different experiments. Cyto, cytoplasmic; Nuc, nuclear.

Localization and Expression of Tyrosine 15 Phosphorylated p34^Cdc2-- We determined the phosphorylation status of p34^Cdc2 in the nucleus and cytoplasm using an antibody specific for the phosphotyrosine 15 form of p34^Cdc2, which has minimal cross-reactivity with unphosphorylated p34^Cdc2. Using this antibody, it has previously been reported that exposure of yeast to DNA-damaging agents results in accumulation of tyrosine-phosphorylated p34^Cdc2 (45). Synchronized HeLa cells were irradiated with 5 Gy, fractionated into cytoplasmic and nuclear components, and assessed for tyrosine 15 p34^Cdc2 protein at 7 h, when the control unirradiated cells would be in G₂ (7 h) and when virtually all irradiated cells would have accumulated in the G₂ block, whereas unirradiated cells would have completed mitosis (24 h) (Fig. 5). Irradiation led to marked nuclear accumulation of tyrosine 15-phosphorylated p34^Cdc2 protein (Fig. 5A). At 24 h, unirradiated control cells have already passed through mitosis and show slight accumulation of tyrosine 15-phosphorylated p34^Cdc2. In contrast, irradiated cells show marked nuclear accumulation of tyrosine 15 phosphorylated p34^Cdc2.

View larger version (46K): [in this window] [in a new window]   Fig. 5.   Accumulation and localization of tyrosine 15-phosphorylated p34Cdc2 after DNA damage. Synchronized HeLa cells were prepared as described in Fig. 1. Cells were treated with DNA-damaging agents 3 h after release from aphidicolin. Control cells (Con) were mock-treated. In A cells were irradiated (IR) with 5 Gy. Both irradiated and unirradiated control cells were harvested at the indicated times and separated into nuclear (Nuc) and cytoplasmic (Cyto) fractions, followed by immunoblotting with antiphosphotyrosine 15 antibody. In B, synchronized cells were irradiated or treated with etoposide (Etop) 3 h after release. Etoposide treatment consisted of 20 µg/ml of etoposide for 20 min before being washed off thoroughly with phosphate-buffered saline. Caffeine (Caff, 2 mg/ml final concentration) was added at the time of radiation or etoposide treatment and was present in the media throughout the remainder of the experiment. Nocodazole (Noco) was used at 20 ng/ml final concentration and was added 3 h after release as well. anti-Tyr-Pi, anti-phosphotyrosine 15 antibody. In C, under conditions identical to B, the cell cycle progression into mitosis was measured using nocodazole trapping as described in the preceding figures.

Caffeine abrogates the G₂ delay resulting from irradiation (31) and leads to increased nuclear p34^Cdc2 activity (Fig. 4B, lanes 5 and 6). We examined the location of tyrosine 15-phosphorylated p34^Cdc2 after caffeine treatment. Synchronized HeLa cells were treated with radiation or etoposide with or without caffeine and harvested 16 h after release from the G₁/S block. Harvested samples were fractionated into cytoplasmic and nuclear components and probed for phosphotyrosine 15 p34^Cdc2 protein (Fig. 5B). Nocodazole-treated cells, serving as controls, as expected showed no tyrosine 15-phosphorylated p34^Cdc2 protein. In contrast, both radiation or etoposide treatment led to marked nuclear accumulation and minimal cytoplasmic accumulation of tyrosine 15-phosphorylated p34^Cdc2. Caffeine substantially reduced tyrosine 15-phosphorylated p34^Cdc2 protein accumulation after radiation and etoposide treatment. The blot was then stripped and reprobed for total (both phosphorylated and unphosphorylated) p34^Cdc2 protein. This showed that total p34^Cdc2 protein was comparable in amounts in the cytoplasm and nucleus under all treatment conditions, suggesting that there was no marked redistribution of protein. Progression into mitosis was measured for cells under the treatment conditions described in Fig. 5B, and the results are shown in Fig. 5C. In each case, inhibition of progression into mitosis was associated with high levels of nuclear tyrosine-phosphorylated p34^Cdc2.

Radiation Dose-dependent Accumulation of Tyrosine 15-phosphorylated p34^Cdc2 Protein after DNA Damage-- The dose dependence of nuclear accumulation of tyrosine 15-phosphorylated p34^Cdc2 protein after irradiation was determined. Synchronized cells were mocked-irradiated or irradiated with 2, 5, or 10 Gy (Fig. 6). All cells were harvested at 15 h after release from the G₁/S block. Cells treated with 10 Gy showed the greatest accumulation, whereas cells treated with 2 Gy showed less but still an appreciable accumulation of phosphotyrosine 15 p34^Cdc2 protein.

View larger version (32K): [in this window] [in a new window]   Fig. 6.   Radiation dose-dependent accumulation of tyrosine 15-phosphorylated p34Cdc2 protein after DNA damage. HeLa cells were grown, and lysates were prepared as described in Fig. 5. Synchronized cells were mock-irradiated (No IR) or irradiated (IR) with 2, 5, or 10 Gy 3 h after release. Cells treated with nocodazole (20 µg/ml) (Noc) 3 h after release served as the negative control. All cells were harvested at the same time, 15 h after release. The nuclear fractions were isolated for immunoblotting.

In summary, treatment of HeLa cells by radiation leads to nuclear accumulation of tyrosine 15-phosphorylated p34^Cdc2 concomitant with G₂ delay, and caffeine abrogates both the cell cycle delay as well as the accumulation of tyrosine 15-phosphorylated p34^Cdc2.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In this report, we showed that the induced expression of either cyclin B1 or Cdc2AF increased p34^Cdc2 kinase activity after irradiation, but that the restoration of activity failed to result in cell cycle progression. The barrier appears to lie in localization of the activity within the cell. The increased p34^Cdc2 activity was initially confined to the cytoplasm. These results indicated that exclusion of p34^Cdc2 kinase activity from the nucleus may contribute to the cell cycle delay after radiation in addition to contributing to the availability of cyclin B1 and phosphorylation of p34^Cdc2.

These results are in accord with and give further insight into a number of previous observations. In the report of Jin et al. (27), expression of Cdc2AF in HeLa cells only partially reversed the G₂ delay after irradiation, despite the rapid induction of substantial p34^Cdc2 kinase activity to amounts greater than those in unirradiated cells. In this report, we confirm that p34^Cdc2 kinase activity is increased by Cdc2AF expression, but we found that the increased activity was not in the nucleus. The delay in the appearance of p34^Cdc2 activity in the nucleus or repression of nuclear p34^Cdc2 kinase activity may account for the residual G₂ delay. These results are quite similar to results obtained by restoring cyclin B1 protein in irradiated HeLa cells. Increased cyclin B1 resulted in increased total p34^Cdc2 activity, yet the p34^Cdc2 activity was also initially restricted to the cytoplasm, consistent with the observation that increased cyclin B1 only partially abrogated the G₂ block in HeLa cells. p34^Cdc2 activity is reduced by radiation, but even high levels of p34^Cdc2 cytoplasmic activity are not sufficient to eliminate the G₂ block.

Sluder et al. (41) have found evidence of cytoplasmic and nuclear compartmentalization of p34^Cdc2 kinase activity in sea urchin zygotes. Pronuclear fusion in these cells is normally marked by the synchronous dissolution of both the sperm and egg nuclei. However, pretreatment of sperm or egg nuclei with ultraviolet light or psoralen to damage the DNA prevented nuclear membrane dissolution of the treated nuclei despite high levels of cytoplasmic p34^Cdc2 kinase activity. Similarly, treatment of HeLa cells with a dose of etoposide, which leads to profound cell cycle block, led to cytoplasmic accumulation of levels of p34^Cdc2 histone H1 kinase activity comparable with control untreated cells (42) without cell cycle progression. Lock (18) also found high levels of p34^Cdc2 activity in HeLa cells treated with etoposide at a time when they remain blocked from cell cycle progression.

Heald et al. (37) found evidence of cytoplasmic-nuclear compartmentalization in baby hamster kidney cells. Transfection of the catalytic domain of Wee1, which is known to phosphorylate and thus inhibit p34^Cdc2, inhibited the onset of mitosis but, paradoxically, led to 5- to 20-fold increase in histone H1 kinase activity. The increased kinase activity appeared to be localized to the cytoplasm, suggesting that the nucleus was protected from the inappropriately high kinase activity (37). These observations, as well as our results, do not exclude the possibility that DNA damage may raise the threshold level of cytoplasmic p34^Cdc2 activity that is required to dissolve the nuclear membrane and initiate mitosis or the possibility of a DNA damage-inducible inhibitor.

The regulation of nuclear p34^Cdc2 activity could occur at the level of cyclin B1, p34^Cdc2 phosphorylation, or both or possibly involve an inhibitor. Expression of Gadd45, a damage-inducible gene, may inhibit cyclin B1-p34^Cdc2 activity (48). Exclusion of cyclin B1 from the nucleus could prevent nuclear p34^Cdc2 activity. Cyclin B1 contains a sequence, the cytoplasmic retention signal (CRS), which confers a degree of cell cycle-dependent localization (Pines and Hunter (34)). The CRS has been shown to be a CRM-1-dependent nuclear export signal. Cyclin B1 is localized in the cytoplasm until just before the onset of mitosis, when retention in the nucleus is accompanied by phosphorylation and inactivation of the nuclear export signal. Nuclear cyclin B1 becomes evident as cells progress from G₂ to mitosis in normally cycling cells, accompanied by phosphorylation of the cytoplasmic retention signal (29). Li et al. (29) showed that although phosphorylation of cyclin B1 is not required for its binding to p34^Cdc2 or induction of p34^Cdc2 kinase activity, cyclin B1 appears to be compartmentalized through the normal cell cycle, and regulation of that compartmentalization by phosphorylation may contribute to the radiation-induced G₂ delay and localization of p34^Cdc2 activity.

What is the role of nuclear localization of cyclin B1 in regulating the G₂ delay induced by DNA damage? Jin et al. (38) recently reported that the expression of Cdc2AF together with the transfection of a vector expressing cyclin B1 in which a nuclear localization signal has been placed (to counteract the CRS) completely abrogated the G₂ delay after irradiation, whereas neither the cyclin B1 nuclear localization signal nor Cdc2AF by itself had such an effect. However, although the cells exited G₂, this was not a normal cell cycle progression but instead resulted in mitotic catastrophe. Toyoshima et al. (30) have found that microinjection of HeLa cells treated with etoposide with mutant cyclin B1 protein that contains a defective CRS and hence remains localized in the nucleus still manifest a profound G₂ delay. These data indicate that translocation of cyclin B1 into the nucleus is not sufficient to effect the transition into mitosis. As suggested by Jin et al. (38), it is plausible that the nuclear localization signal cyclin B1 bypasses the compartmentalization of p34^Cdc2 kinase activity by driving cytoplasmic kinase activity into the nucleus. Since cyclin B1 localization is controlled by its phosphorylation, it remains an intriguing possibility that cyclin B1-directed kinases regulate cyclin B-p34^Cdc2 localization and are modulated by checkpoint pathways (44). Our finding of significant levels of cyclin B1-p34^Cdc2 activity persisting in the cytoplasm after exposure to DNA-damaging agents underscore the complexity of this regulation. Thus cyclin B1 availability and location potentially affect the G₂ delay. In addition, p34^Cdc2 activity is also inhibited after irradiation. It has been previously reported that DNA damage leads to tyrosine phosphorylation of p34^Cdc2 (18, 24, 45, 54-58). DNA damage has been reported to up-regulate Wee1 kinase activity, a kinase that leads to p34^Cdc2 inactivation by tyrosine phosphorylation (45, 46). However, other reports have found that radiation does not have an effect on Wee1 kinase (47), suggesting that the regulation of Wee1 kinase activity may be cell type-specific or that other inhibitory kinases may exist.

The phosphorylation of p34^Cdc2 is also controlled by the phosphatase Cdc25C, which activates p34^Cdc2 by removing the inhibitory phosphorylation (49). Cdc25C in turn may be inhibited by Chk1, a protein kinase that is activated by DNA damage (28, 50-52). It has recently been reported that ionizing radiation results in exclusion of Cdc25 from the nucleus in fission yeast, mediated by association of Cdc25 protein with Rad24 protein, the latter of which acts as a nuclear export sequence (53). It remains to be seen whether similar effects are noted in human cells. Nonetheless, the exclusion of both cyclin B1 and Cdc25 protein from the nucleus after exposure to DNA-damaging agents underscores the central importance of nuclear localization of factors necessary for cell cycle progression in mediating progression through the G₂ delay after DNA damage.

We noted that tyrosine-phosphorylated p34^Cdc2 is predominantly localized to the nucleus. It is possible that nuclear localization of p34^Cdc2 (with or without its cyclin B1 partner) may first require tyrosine 15 or threonine 14 phosphorylation. Nuclear localization signals have frequently been found to be regulated by phosphorylation. Boulikas (59) found that most nonmembrane serine/threonine protein kinases, including Cdc2, contain karyophilic and acidic clusters of amino acids and has proposed that this may serve as a weak nuclear localization signal, perhaps in conjunction with transporter proteins. The inability of Cdc2AF to completely abolish the cell cycle delay after radiation may be related to delayed entry into the nucleus; we are currently investigating this possibility.

Other cyclins that are present in substantial quantities during the G₂ to M transition include cyclin A and B2. Cyclin A is constitutively nuclear (60). The CRS is conserved between cyclin B1 and B2, but there are significant differences in their localization patterns. Human cyclin B2 appears to colocalize with the Golgi apparatus and remains in the cytoplasm even as cells enter prophase and cyclin B1 precipitously relocates into the nucleus (34, 62). The roles cyclin A-p34^Cdc2 and cyclin B2-p34^Cdc2 play in the checkpoint response remain to be determined.

DNA damage results in cell cycle block and regulation of p34^Cdc2 activity in a wide range of cell types. Several mechanisms have now been identified for this block, including cyclin B1 levels, p34^Cdc2 phosphorylation, and localization of p34^Cdc2 activity as distinct but overlapping mechanisms that result in cell cycle block after DNA damage. In this report, we find that induction of cyclin B1, introduction of the constitutively active form of cyc2AF, or both only partially reverse the radiation-induced G₂ delay. Our findings that p34^Cdc2 activity is compartmentalized help explain the inability of these maneuvers to completely abrogate the block. Cyclin B1 expression can markedly stimulate the kinase activity, but that activity is confined to the cytoplasmic compartment. Taken together, these results suggest that the increased kinase activity is in itself ineffectual in initiating mitosis and that altered compartmentalization of p34^Cdc2 activity may contribute to the cell cycle delays following exposure to ionizing radiation in HeLa cells. These results suggest a revised model of cell cycle progression in which cyclin B1-p34^Cdc2 kinase activity is initially diminished but also initially confined to the cytoplasm after exposure of HeLa cells to DNA-damaging agents, including during the G₂ delay. The portion of p34^Cdc2 that is nuclear is maintained in an inactive state by tyrosine phosphorylation. Progression into mitosis occurs when nuclear activity becomes evident. Further work will be required to fully characterize the mechanisms that regulate this activity.

	ACKNOWLEDGEMENTS

We thank Karen Chang and Yi Cheng for superb technical assistance and Dr. Frank McKeon for the gift of the pECE-Cdc2 and pECE-Cdc2AF plasmids.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants GM 47439 and CA 751138 (to R. J. M.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ To whom correspondence should be addressed: Rm. 269 John Morgan Bld., Dept. of Pathology, 36 and Hamilton Walk, University of Pennsylvania, Philadelphia, PA 19104. Tel.: 215-898-8401; Fax: 215-573-4243; E-mail: muschel@mail.med.upenn.edu.

	ABBREVIATIONS

The abbreviations used are: Gy, gray; MOPS, 3-(N-morpholino)propanesulfonic acid; CRS, cytoplasmic retention signal.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				C.-C. Chang, J. D. Heller, J. Kuo, and R. C. C. Huang Tetra-O-methyl nordihydroguaiaretic acid induces growth arrest and cellular apoptosis by inhibiting Cdc2 and survivin expression PNAS, September 7, 2004; 101(36): 13239 - 13244. [Abstract] [Full Text] [PDF]

				L. Fletcher, Y. Cheng, and R. J. Muschel Abolishment of the Tyr-15 Inhibitory Phosphorylation Site on cdc2 Reduces the Radiation-induced G2 Delay, Revealing a Potential Checkpoint in Early Mitosis Cancer Res., January 1, 2002; 62(1): 241 - 250. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kao, G. D. Articles by Muschel, R. J. Search for Related Content PubMed PubMed Citation Articles by Kao, G. D. Articles by Muschel, R. J. Social Bookmarking                      What's this?