| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
1 CNRS, UMR 8126 and 2 CNRS, UMR 8125, Institut Gustave Roussy, Villejuif, France
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Chemicals.
Demecolcine and roscovitine (Sigma Chemical Co., St. Louis, MO) were resuspended in dimethylsulfoxide to a stock concentration of 2.6 mmol/L and 75 mmol/L, respectively, and used at a 1:10,000 dilution and 1:1,000 dilution, respectively. CLQ (Sigma) was resuspended in water at the stock concentration of 40 mg/mL and used at the indicated dilutions. TSA (Sigma) was resuspended in EtOH to a stock concentration of 20 mmol/L and used at the indicated dilutions.
Radiation Treatment.
Cells were irradiated in a culture medium containing demecolcine (Dmc) at room temperature with 50 Gy (D1) or 100 Gy (HeLa) using a 137Cs gamma source at a dose rate of 1.95 Gy/min.
Antibodies.
Rabbit anti-BLM antiserum 1343 was generated and used as described previously (2)
. Monoclonal antibodies C-17, anti-cdc2, and GNS1 anti–cyclin B1 were from Santa Cruz Biotechnology (Santa Cruz, CA) and were used at a 1:500 dilution. Monoclonal antibodies Ki-S1 anti–topoisomerase II
and MAB3408 anti–ß-tubulin were from Chemicon International, Inc. (Temecula, CA) and were used at a final concentration of 1 µg/mL and at a 1:400 dilution, respectively. Monoclonal antibody antihistone H2AX phosphorylated at serine 139 (05–636; Upstate Biotechnology, Lake Placid, NY) was used at a final concentration of 0.5 µg/mL. Monoclonal anti–poly(ADP-ribose) polymerase-1 (PARP-1) antibody (Ab-2; Calbiochem, Darmstadt, Germany) was used at a final concentration of 1 µg/mL. The antibody against histone H3 phosphorylated at serine 10 (06–570; Upstate Biotechnology) was used at final concentrations of 0.5 µg/mL for Western blot analysis and 5 µg/mL for immunofluorescence. Goat antirabbit IgG antiserum conjugated to peroxidase (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) and goat antimouse IgG antiserum conjugated to peroxidase (Pierce, Rockford, IL) were used at a 1:5000 dilution. Goat antirabbit IgG antiserum conjugated to Alexa Fluor-568 and goat antimouse IgG antiserum conjugated to Alexa Fluor-488 were from Molecular Probes, Inc. (Eugene, OR) and were used at a 1:400 dilution.
Flow Cytometry Analysis.
Cells were harvested (floating Dmc-arrested cells) or scraped (TSA-, CLQ-, or roscovitine-treated Dmc-arrested cells) and then washed in PBS. Cell cycle analysis was performed as described previously (2)
.
Western Blot Analysis.
Cells were lysed in either 1% SDS in water for 5 minutes at 95°C or in 8 mol/L urea buffer [8 mol/L urea, 150 mmol/L ß-mercaptoethanol, and 50 mmol/L Tris (pH 7.4) in water] and then sonicated. The equivalent of 5 x 105 cells was subjected to immunoblot analysis as described previously (2)
. Alternatively, cells were lysed in immunoprecipitation buffer (see Kinase Assay).
NP40 Extraction.
Cells were washed with PBS and then subjected to NP40 extraction as described in detail previously (3)
.
Phosphatase Treatment.
Cells were washed with PBS and subjected to NP40 extraction. Proteins (80 µg) were resuspended in phosphatase buffer and incubated for 2 hours at 30°C with 1200 units of 
protein phosphatase in the presence of MnCl2, according to the manufacturer’s instructions (New England Biolabs, Beverly, MA).
Kinase Assay.
Cells were lysed in immunoprecipitation buffer [250 mmol/L NaCl, 1.5% NP40, 0.5 mmol/L EDTA, 10 mmol/L sodium orthovanadate, 10 mmol/L NaF, 40 mmol/L Tris (pH 7.4), 10 mmol/L sodium PPI, 1 mmol/L phenylmethylsulfonylfluoride, 5 mmol/L DTT, 1 µg/mL leupeptin, 1 µg/mL pepstatin, and 1 µg/mL aprotinin]. Cell lysates were incubated by stirring gently with monoclonal anti-cdc2 antibody (1:100 dilution) overnight at 4°C. Immunocomplexes bound to protein G-Sepharose were collected by centrifugation and washed once in immunoprecipitation buffer and three times in kinase assay buffer [50 mmol/L Tris (pH 7.5), 10 mmol/L MgCl2, 1 mmol/L EDTA, and 1 mmol/L DTT]. The beads then were incubated with 1 µg of histone H1, 50 µmol/L of cold ATP, and 14 µCi of 3000 Ci/mmol [
-32P]ATP (MP Biomedicals Germany GmbH, Eschwege, Germany) in the kinase assay buffer in a total volume of 30 µL for 30 minutes at 30°C with shaking. The reaction was stopped by the addition of 2x sample buffer. The reaction mixture then was heated at 90°C for 5 minutes. Proteins were resolved by 10% SDS-PAGE before transferred onto membrane. The membrane was analyzed by autoradiography, and the amount of cdc2 or cyclin B1 protein in immunoprecipitates was determined by probing the membrane with anti-cdc2 or anti–cyclin B1 antibodies, respectively.
Preparation of Chromosome Spreads.
Roscovitine-treated Dmc-arrested HeLa cells were scraped and washed in PBS. Cells then were treated with hypotonic solution (0.075 mmol/L KCl in water) for 15 minutes at 37°C and fixed in three changes of fixative solution (3:1, methanol/acetic acid; v/v). Chromosome preparations were dropped onto glass coverslips, exposed to steam (30 seconds), and air-dried (1 hour). After washing in PBS, chromosomes were stained by incubating with 1 mg/mL of 4,6-diamidino-2-phenylindole solution for 5 minutes (Sigma), followed by a wash in water. Confocal fluorescent images were collected using a Leica TCS confocal system (Leica, Wetzlar, Germany). For each sample, chromosomes from 45 mitotic cells were counted.
Immunofluorescence Labeling.
Cells were transferred onto glass coverslips and fixed in 3% formaldehyde in PBS (20 minutes, room temperature). All of the subsequent steps were performed at room temperature. Cells were rinsed in PBS, incubated for 10 minutes in 50 mmol/L NH4Cl in PBS, rinsed in PBS, and permeabilized with 0.5% Triton X-100 for 5 minutes. After three washes in PBS, cells were blocked for 1 hour with 10% fetal calf serum in PBS. The slides then were incubated with primary antibodies (antihistone H3 phosphorylated at serine 10 and anti–ß-tubulin) diluted in PBS containing 2.5% bovine serum albumin (1 hour). After being washed three times with PBS, the slides were incubated with the secondary antibodies (goat antirabbit conjugated with Alexa Fluor-568 and goat antimouse conjugated with Alexa Fluor-488) for 30 minutes at room temperature in the dark. After washing in PBS, nuclear DNA was stained by incubating with TO-PRO3 solution (dilution, 1:1000; Molecular Probe) for 10 minutes. After two washes in PBS and one wash in water, coverslips were mounted on glass slides (mounting media was purchased from Molecular Probe). Confocal fluorescent images were collected using a Zeiss LSM 510 confocal system (Zeiss, Oberkochen, Germany).
|
RESULTS AND DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
and a time course experiment (data not shown) of the effect of these chromatin-modifying drugs on the BLM migration shift and subcellular localization, which were expected to be sensitive to cdc2 inactivation (3)
. BLM phosphorylation was reversed completely at the highest TSA concentrations (Fig. 1A
, Lanes 6 and 7) but only partially reversed by the highest CLQ concentration (Fig. 1B
, Lanes 5 and 6), migrating to a position between BLM in unsynchronized cells (Lane 1) and BLM in mitosis-arrested cells (Lane 2). This was caused by the partial dephosphorylation of mitotic BLM by CLQ treatment because phosphatase treatment of the protein extract from CLQ-treated Dmc-arrested cells resulted in the recovery of a band that migrated similarly to BLM from untreated exponentially growing cells or mitotic BLM treated with phosphatase (Fig. 1C)
.
|
, top). We observed that TSA treatment of Dmc-arrested cells, and to a lesser extent CLQ treatment, induced apoptosis (sub-G1 population; Fig. 1D and E
). Phosphorylated BLM from nontreated Dmc-arrested HeLa cells was extracted completely in the NP40-soluble fraction, whereas dephosphorylated BLM from TSA-, CLQ-, or roscovitine-treated Dmc-arrested HeLa cells was found only in the NP40-insoluble fraction (Fig. 1D–F
, bottom). As a control, the same membranes were reprobed with specific antibodies for the topoisomerase II, which has a similar molecular mass to BLM and is the major nonhistone protein component of the chromosome scaffold (15)
. The same results were obtained using MCF7 or U2OS cells (data not shown). These results show that mitotic BLM is dephosphorylated and recruited to the NP40-insoluble fraction in response to roscovitine, known to inhibit cdc2 kinase, and in response to TSA and CLQ. One possible explanation for these results is that TSA and CLQ treatment of Dmc-arrested cells may alter cdc2 kinase activity.
The Level of Mitotic BLM Phosphorylation Reflects the Level of cdc2 Kinase Activity.
We next immunoprecipitated cdc2 from protein extracts from untreated or TSA-, CLQ-, or roscovitine-treated Dmc-arrested HeLa cells. The cdc2 kinase activity was measured by following histone H1 phosphorylation (ref. 16
; Fig. 2A
, top). The amount of cdc2 immunoprecipitated in the kinase assay was checked by Western blot analysis using an anti-cdc2 antibody (Fig. 2A
, middle). The reversion of mitotic BLM phosphorylation also was checked by Western blot analysis (data not shown). As expected, cdc2 kinase activity was stabilized in Dmc-arrested cells and inhibited in roscovitine-treated Dmc-arrested cells. Furthermore, treatment of Dmc-arrested cells with TSA completely inhibited cdc2 kinase activity, whereas treatment with CLQ partially inhibited cdc2 kinase activity (Fig. 2A
, top). We also found that the amount of cyclin B1 that coimmunoprecipitated with cdc2 was slightly lower in CLQ-treated cells and much lower in TSA- or roscovitine-treated cells than in Dmc-arrested cells (Fig. 2A
, bottom). The same pattern of cyclin B1 production was observed in the corresponding total protein extracts (data not shown), indicating that the total level of cyclin B1 was much lower in TSA- or roscovitine-treated cells than in untreated Dmc-arrested cells. The partial inactivation of cdc2 (Fig. 2A)
correlates with the partial dephosphorylation of mitotic BLM (Figs. 1B
and 2B
) in CLQ-treated Dmc-arrested cells. These results indicate that the level of mitotic BLM phosphorylation reflects the level of cdc2 kinase activity.
|
-H2AX; ref. 17
; Fig. 2B
). Large amounts of -H2AX were present in mitotic cellular extracts treated by ionizing radiation known to inhibit cdc2 kinase activity (ref. 3
; Supplemental Data) and used as a positive control, and in TSA-treated mitotic cellular extracts (Fig. 2B)
. Small amounts of -H2AX were detected in CLQ-treated mitotic cellular extracts. No -H2AX was detected in untreated mitotic cellular extracts (negative control) or in roscovitine-treated cellular extracts (Fig. 2B)
. Roscovitine inhibits cdc2 by competing with its ATP-binding site (18)
and was not expected to generate DNA double-strand breaks. These results suggest that partial inactivation of cdc2 by CLQ and total inactivation of cdc2 by TSA probably are caused by the presence of DNA double-strand breaks. To investigate this point further, we studied the effects of increasing TSA concentrations (5, 10, and 20 µmol/L) on induction of -H2AX and cdc2 kinase activity (Fig. 2C)
. The higher the TSA concentrations the more -H2AX and the dephosphorylated form of BLM were detected in the corresponding total protein extracts (Fig. 2C
, top). In contrast, increasing TSA concentrations were associated with a decreasing level of cyclin B1 (Fig. 2C
, top). Moreover, increasing TSA concentration was correlated with increasing inhibition of cdc2 kinase activity, whereas similar amounts of cdc2 were immunoprecipitated from these extracts at all of the concentrations (Fig. 2C
, bottom). These results show a strict correlation between TSA concentrations, the level of -H2AX accumulation, and the level of cdc2 activity, indicating that numerous DNA breaks must occur during the spindle assembly checkpoint to inactivate cdc2 completely.
DNA damage during the spindle assembly checkpoint prevents exit from mitosis (3
, 19)
. To determine whether the activation of the mitotic checkpoint in response to DNA damage depends on cdc2 activity, we analyzed the effect of TSA, CLQ, and roscovitine on exit from mitosis. We carried out experiments with nonmalignant Epstein-Barr virus–transformed D1 lymphoblastoid cells, which display a mitotic DNA damage checkpoint in response to ionizing radiation (3)
. Dmc-treated D1 lymphoblastoid cells (54% of cells in G2/M), untreated or treated with ionizing radiation (as control), TSA, CLQ, or roscovitine, were released into drug-free medium and harvested and fixed 4 and 8 hours later. Eight hours after release, D1 cells that had been treated only with Dmc had exited mitosis and entered the next G1 phase (13% of cells in G2/M). Conversely, at this time point, a large proportion of cells exposed to ionizing radiation, TSA, or roscovitine remained arrested with a 4N DNA content (31%, 19%, or 34%, respectively; Fig. 2D
). CLQ-treated cells exited mitosis and entered the next G1 phase as fast as untreated cells (13% of cells in G2/M 8 hours after the release; Fig. 2D
). Roscovitine treatment of Dmc-arrested HeLa cells and their release into a drug-free medium further confirmed that inactivation of cdc2 is sufficient to block mitotic exit (Fig. 2E)
. The same experiments were conducted using purvalanol A, another selective inhibitor of cdc2 kinase activity, and the same block of mitotic exit was observed in treated HeLa cells (Supplemental Data). Hence, the cdc2 kinase must be completely inactivated during the spindle assembly checkpoint to stop cell cycle progression because CLQ partially inactivated cdc2 in mitotic cells (Fig. 2A)
and did not prevent mitotic exit (Fig. 2D)
, whereas ionizing radiation (3)
, TSA, and roscovitine completely inactivated cdc2 (Fig. 2A)
and blocked exit from mitosis (Fig. 2D and E)
. These results indicate that the mitotic cdc2 kinase inactivation is a key event in the establishment of the mitotic DNA damage checkpoint.
In Response to Mitotic cdc2 Inactivation, the M/G1 Transition Is Delayed in D1 Nonmalignant Cells, whereas HeLa Tumor Cells Exit Mitosis without Dividing, Resulting in a Mitotic Catastrophe.
The strong decrease in cyclin B1 concentration following TSA or roscovitine treatment of mitosis-arrested HeLa cells suggested that cells moved to a tetraploid G1-like state (20)
following cdc2 inactivation. This is not consistent with previous results showing that no BLM can be detected in G1 (2)
. To investigate this apparent contradiction, we analyzed the cell cycle progression of Dmc-arrested D1 or HeLa cells, left untreated or treated with roscovitine, after their exit from the mitotic block. Cells then were released into a fresh medium without drugs, were harvested 4, 8, 16, 24, 48, 72, and 120 hours later and then every 2 days for 1 week, and were analyzed for DNA content by flow cytometry (Fig. 3A and B)
. Roscovitine-treated lymphoblastoid cells left G2/M and entered G1 later than untreated cells; they exhibited a normal cell cycle distribution 3 days after the release instead of 16 hours (Fig. 3A)
. The percentage of sub-G1 in untreated and roscovitine-treated lymphoblastoid cells was similar and did not exceed 23% (Fig. 3A)
. Identical results were obtained using Epstein-Barr virus–transformed lymphoblastoid BS cells GM3403D (data not shown).
|
. The proportion of apoptotic cells (sub-G1) increased over time, reaching 65% 5 days after the release. The remaining cells displayed a normal cell cycle distribution 14 days after Dmc removal (Fig. 3B)
. Roscovitine-treated Dmc-arrested HeLa cells still were arrested in G2/M 8 hours after drug removal. A population exhibiting >4N DNA content was detectable 16 hours after release (19%) and progressed to 8N (24%) by 24 hours. Forty-eight hours after release, all of the cells were apoptotic, and no live cells were detected after 5 days (Fig. 3B)
. These results suggest that a subpopulation of roscovitine-treated Dmc-arrested HeLa cells exited mitosis without dividing and progressed into a second S phase 16 to 24 hours after their release into a drug-free medium. We confirmed this by counting the number of chromosomes in mitotic spreads: 8 hours after the release, roscovitine-treated mitotic cells had 65 ± 6 chromosomes, whereas 24 hours after release they had 126 ± 14 chromosomes (Fig. 3C)
. We next used Western blot analysis to analyze the presence of BLM, PARP, cyclin B1, cdc2, and serine 10–phosphorylated histone H3 (P-H3) in Dmc-arrested cells left untreated or treated with roscovitine and released into a drug-free medium at the indicated times (Fig. 3D)
. After the release of untreated Dmc-arrested cells from the mitotic block, BLM was progressively dephosphorylated, and the levels of cyclin B1 and P-H3 decreased (Fig. 3D
, right). Increasing amounts of cleaved forms of BLM (
BLM) and PARP (PARP), characteristic of apoptosis (21)
, were detected between 4 and 48 hours after release (Fig. 3D
, right). BLM and PARP still were detected at 120 hours but to a lesser extent. The cdc2 levels remained constant. These results indicate that after the release from the Dmc-induced spindle assembly checkpoint, cells progressively exit mitosis. Some of these cells become apoptotic, and the others divide normally, which is in agreement with the DNA content data (Fig. 3B)
. Analysis of roscovitine-treated Dmc-arrested HeLa cells showed that at the end of roscovitine treatment, BLM phosphorylation was reversed and that the concentrations of cyclin B1 and P-H3 decreased sharply, indicating that cells were in an interphase-like state (Fig. 3D
, left). Up to 48 hours after release from the cdc2 inactivation block, cyclin B1 progressively accumulated. The mitotic phosphorylated form of BLM and P-H3 were detectable 24 hours after release (Fig. 3D
, middle). These results indicate that between 16 and 24 hours after the release of roscovitine-treated Dmc-arrested HeLa cells into a drug-free medium, cdc2/cyclin B1 is reactivated, and chromosomes are condensed in a subpopulation of cells, probably those with an 8N DNA content detected by flow cytometry (Fig. 3B)
and observed in mitotic spreads (Fig. 3C)
. The amounts of the cleaved apoptotic forms of BLM (BLM) and PARP (PARP) detected increased between 4 and 48 hours after release. One hundred twenty hours after release, BLM levels decreased sharply, and a cleaved form of cdc2 was detected (Fig. 3D
, middle). Coimmunostaining of cells with anti–ß-tubulin and anti–P-H3 antibodies 16 and 24 hours after their exit from either the spindle assembly checkpoint or the cdc2 inactivation block showed that Dmc treatment alone gave rise to mitotic cells with condensed chromosomes and bipolar mitotic spindles and to interphase cells (no apoptotic cells were detected; Fig. 3E and E
'). Conversely, sequential treatment with Dmc and roscovitine resulted in mitotic cells with multipolar spindles and in large cells with several micronuclei containing uncondensed chromosomes that reflect a mitotic catastrophe (ref. 22
; Fig. 3F and F
'). These results indicate that despite strong BLM expression, roscovitine-treated Dmc-arrested HeLa cells are in a G1 tetraploid-like state as they progress to the S phase (23)
. These cells then try to initiate mitosis by reactivating cdc2 (detected through BLM phosphorylation), condensing chromosomes, and forming spindles and thus undergo mitotic catastrophe. In conclusion, the strict correlation between mitotic BLM phosphorylation and cdc2 activity shows that mitotic cdc2 inactivation plays a major role in the establishment of the mitotic DNA damage checkpoint. Inactivation of cdc2 during the spindle assembly checkpoint stops the cell cycle from progressing. This blockade can take place independently of DNA damage but also is inducible by a high level of DNA double-strand breaks. In nonmalignant cells, mitotic cdc2 inactivation delays M/G1 transition without affecting cell survival. In contrast, tumor cells reinitiate a cell cycle without dividing, resulting in a mitotic catastrophe. This is in agreement with the finding that down-regulation of cdc2 in a human fibrosarcoma cell line is associated with extensive DNA rereplication and apoptosis, indicating a role for cdc2 in preventing premature initiation of S phase (24) , as shown in fission yeast (25) . The efficiency of sequential treatment with Dmc and roscovitine in killing tumor cells does not require microtubule stabilization and continued drug exposure as reported for tumor growth suppression by ablating a cdc2 survival checkpoint (26) , opening the way for new antitumoral protocols.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: Supplementary data for this article can be found at Cancer Research Online (http://cancerres.aacrjournals.org). Present address of E. Bayart, O. Grigorieva, R. Onclercq-Delic, and M. Amor-Guéret: Institut Curie-CNRS UMR 2027, Bâtiment 110, Centre Universitaire, 91405 Orsay Cedex, France. Present address of S. Leibovitch: INRA-CNRS UMR 866, 2 place Pierre Viala, 34060 Montpellier Cedex 1, France.
Requests for reprints: Mounira Amor-Guéret, Institut Curie-CNRS UMR 2027, Bâtiment 110, Centre Universitaire, 91405 Orsay Cedex, France. E-mail: mounira.amor{at}curie.u-psud.fr
Received 5/ 7/04. Revised 8/12/04. Accepted 10/10/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. Magnaghi-Jaulin, G. Eot-Houllier, G. Fulcrand, and C. Jaulin Histone Deacetylase Inhibitors Induce Premature Sister Chromatid Separation and Override the Mitotic Spindle Assembly Checkpoint Cancer Res., July 1, 2007; 67(13): 6360 - 6367. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. I. Shapiro Cyclin-Dependent Kinase Pathways As Targets for Cancer Treatment J. Clin. Oncol., April 10, 2006; 24(11): 1770 - 1783. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
