| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
Prevents G2 Progression to Mitotic Catastrophe in Human Colon Carcinoma Cells after DNA Damage, But p21WAF1 Induces Stable G1 Arrest in Resulting Tetraploid Cells1
Institut de Biologie Structurale Jean-Pierre Ebel (CEA-CNRS), 38027 Grenoble Cedex 1, France
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, which are both transcriptional products of p53, have been reported to play a role in the G2 DNA damage checkpoint in mammalian cells. Human colon carcinoma cells, isogenic except for the presence or absence of either p21WAF1 or 14-3-3 (T. A. Chan et al., Genes Dev., 14: 1584–1588, 2000), are useful models for analysis of the role of these proteins in checkpoint control. Here, we have examined mitotic behavior within a single cell cycle after DNA damage in these cell lines. Our results show that p21WAF1, but not 14-3-3, imposes a significant G2 delay after DNA damage. After G2 delay, we found that all isogenic cells, including those competent for both p21WAF1 and 14-3-3, adapt to the DNA damage checkpoint and progress into mitosis, where they undergo incomplete chromosome segregation and reenter G1 with a tetraploid DNA content. Strikingly, our results show that p21WAF1, but not 14-3-3, activates a checkpoint in response to DNA damage that prevents continued cycling of the tetraploid cells that result from a mitotic catastrophe characterized by failure to complete cell division. These results demonstrate that a tetraploid DNA content is not a reliable criterion to establish that arrest occurs in G2. Also, the DNA damage checkpoint mediated by p53-dependent induction of p21WAF1 assures neither G2 arrest nor DNA repair sufficient to enable accurate chromosome segregation in human colon carcinoma cells. We conclude that p21WAF1, but not 14-3-3, has a unique role in the induction of G1 arrest in tetraploid cells that results from mitotic catastrophe after DNA damage. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
p53 induces the transcription of several proteins that regulate cell cycle progression. Among these proteins, both p21WAF1 and 14-3-3 appear to be induced in a p53-dependent manner after DNA damage (7, 8, 9)
. p21WAF1 is important to G1 arrest (10
, 11) , and both p21WAF1 and 14-3-3 have been reported to be important for G2 arrest after DNA damage (6
, 12)
. p21WAF1 (13, 14, 15, 16)
is one of a group of proteins, termed cyclin-dependent kinase inhibitors that bind to and inhibit the CDKs,3
a family of protein kinases that drive the cell cycle (reviewed in Ref. 17
). p21WAF1 appears to induce cell cycle arrest after DNA damage by directly inhibiting the activity of CDKs (14, 15, 16)
, whereas 14-3-3 has been reported to act by cytoplasmic sequestration of the cyclin B-cdc2 complex that normally induces mitosis upon nuclear entry (9)
. Furthermore, by ensuring G2 arrest after DNA damage, 14-3-3 has been reported to prevent a mitotic catastrophe that induces cell death (9)
. The different activities of p21WAF1 and 14-3-3 in inducing G2 arrest have been reported to act synergistically to prevent cell death after DNA damage (18)
.
Here we have used two-dimensional flow cytometry for both DNA content and the mitotic marker MPM-2 (19)
to critically examine the rate of mitotic entry after DNA damage of isogenic human colon carcinoma cells that vary only by the presence or absence of the genes for either p21WAF1 (20)
or 14-3-3 (9)
. By examining mitotic entry within the first cell cycle subsequent to the DNA damage event, we were able to distinguish premitotic responses to DNA damage from postmitotic responses. Surprisingly, we have found that control human colon carcinoma (HCT116) cells, which are competent to induce both p21WAF1 and 14-3-3, undergo adaptation to the G2 DNA damage checkpoint, a phenomenon demonstrated previously in yeast (21, 22, 23)
, and enter mitosis after DNA damage. In mitosis the cells undergo a mitotic catastrophe in which they fail in chromosome segregation and become tetraploid, a result reported previously only for HCT116 cells deficient for p21WAF1 (6)
. This finding demonstrates that the functions of p53, including induction of its transcriptional products p21WAF1 and 14-3-3, are not sufficient to sustain stable G2 arrest in human colon carcinoma cells, nor are they sufficient to ensure repair of DNA damage so that chromosome segregation occurs normally.
It is clear from our results that DNA ploidy alone cannot be used to assay for G2 arrest, although this has been standard in past studies. Cells with 4N DNA content could be either in G2, or in mitosis, or in G1 after a failed mitosis. Although each of the isogenic human colon carcinoma cell lines used in this study, regardless of p21WAF1 or 14-3-3 status, proceed through G2 and an abortive mitosis after DNA damage, p21WAF1-competent cells enter mitosis more slowly than p21WAF1-/- cells. Thus, p21WAF1 imposes a G2 delay, but not a G2 arrest, in response to DNA damage. In contrast, 14-3-3 status has no evident inhibitory effect on progression past G2 after DNA damage. Furthermore, we found that, although insufficient to prevent mitotic entry and catastrophe after DNA damage, p21WAF1, but not 14-3-3, is required to prevent rereplication of DNA after tetraploidization.
Our results have important implications for chemotherapy of tumors using DNA-damaging agents. HCT116 cells that are deficient for p21WAF1 are sensitized to DNA damage-induced apoptosis (24) . Because we found that HCT116 cells are not competent for prolonged arrest in G2, it appears that arrest of tetraploid cells in G1 may be an important component of response to DNA damage. Additional studies to elucidate other components of the G1 tetraploidy checkpoint provoked by DNA damage should define other targets and genetic determinants important to successful chemotherapy.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
+/+) and its p21WAF1-/- and 14-3-3-/- derivatives were obtained from Dr. B. Vogelstein (Johns Hopkins University, Baltimore, MD). Cells were grown as monolayers as described previously (20)
.
Cells were replated at subconfluency for 24 h prior to any treatment. Irradiation was delivered by a 137Cs 
-irradiator at about 2 Gy/min. Adriamycin was obtained from Sigma Chemical Co. (St. Louis, MO) and was prepared as a 1 mg/ml stock solution in sterilized H2O and then kept at 4°C. VP-16 and nocodazole were obtained from Sigma Chemical Co. and were prepared as 25 mg/ml (VP-16) and 1 mg/ml (nocodazole) stock solutions in DMSO that were kept frozen until used.
To specifically study the effect of DNA damage upon the G2-M transition, cells grown as monolayers were treated 16 h with 2 mM hydroxyurea (Sigma Chemical Co.) prepared from a frozen 200 mM stock in culture medium. Cells were washed twice with drug-free medium and were then treated to induce DNA damage. Cells were, in addition, treated with 0.5 µg/ml nocodazole to accumulate all cells that passed through to mitosis. Samples were analyzed 20 h after release from hydroxyurea, a time when controls released from hydroxyurea had proceeded through and exited mitosis.
Flow Cytometric Analysis.
For quantification of mitosis, cells were analyzed by two-dimensional flow cytometry using MPM-2, a mitotic marker (19)
, and propidium iodide, a marker of DNA content. Cells were fixed with 2% paraformaldehyde in PBS for 20 min, permeabilized for 3 min with 0.2% Triton X-100 in PBS, and then labeled with MPM-2 antibodies and FITC-conjugated goat antimouse IgG secondary antibodies (Jackson Laboratories, West Grove, PA), followed by incubation with propidium iodide as described previously (25)
.
Data were collected using a FACScan flow cytometer (Becton Dickinson, San Jose, CA). For each sample, 10,000 events were collected, aggregated cells were gated out, and MPM-2 positive cells were quantitated using CellQuest software (Becton Dickinson). Cell cycle distribution was determined from histograms of DNA content using ModFit LT software (Verity Software House, Inc., Topsham, ME). Similar results were obtained by counting MPM-2-positive cells by immunofluorescence microscopy.
Immunofluorescence Microscopy.
For immunofluorescence microscopy, cells were grown on 12-mm diameter glass coverslips for 24 h prior to drug treatment. Cells were fixed and permeabilized as described above for flow cytometry. Incubation with primary and secondary antibodies and washes were as described previously (25)
. Counterstain, when used, was with 0.5 µg/ml propidium iodide in PBS for 5 min.
The following antibodies were used for indirect immunofluorescence microscopy. MPM-2 mouse monoclonal antibody (Dako, Carpinteria, CA) was used at a 100-fold dilution; anti-ß-tubulin ascites antibody from Sigma Chemical Co. (TUB 2.1) was used at a 400-fold dilution; and rabbit antihuman lamin B antiserum (26) , obtained from J-C. Courvalin (Institut Jacques Monod, Paris), was used at a 300-fold dilution. Secondary antibodies, all from The Jackson Laboratory, included FITC-conjugated affinity purified goat antimouse and antirabbit IgG antibodies applied at 2.5 µg/ml.
Samples were observed using an Optiphot II microscope (Nikon, Inc., Melville, NY) attached to a MRC-600 laser scanning confocal apparatus (Bio-Rad Microscience Division, Herts, United Kingdom). Images were treated with Adobe Photoshop and printed on an Epson Stylus color 900 printer (Seiko Epson Corp.).
Immunoprecipitation and Histone H1 Kinase Assays.
Lysates were prepared in 50 mM Tris-HCl (pH 7.4), 250 mM NaCl, 5 mM EGTA, 0.1% NP40, 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 1.0 mM phenylmethylsulfonyl fluoride for 30 min on ice. Lysate supernatants were then collected by centrifugation at 13,000 x g for 5 min at 4°C.
To preclear cell extracts of proteins nonspecifically binding to Sepharose beads, 50 µl of each extract were incubated for 30 min at 4°C, with agitation, in a 1:1 slurry with protein A-Sepharose 4B beads (Sigma Chemical Co.) in lysis buffer. Forty µg of each cleared extract were then incubated with 4 µl of anti-p21WAF1 antibody (C19; Santa Cruz Biotechnology, Santa Cruz Biotechnology, CA), 3 µl of anti-cyclin A antiserum (27) , 100 µl of anti-cyclin B hybridoma supernatant (GNS-1), 4 µl of anti-cdk2 (27) , or 4 µl of rabbit anti-cdc2 antiserum (25) for 1 h at 4°C with agitation. Fifty µl of protein A-Sepharose slurry was then added, and the extract was further incubated for 1 h at 4°C. The resulting immune complex was washed four times in lysis buffer.
Kinase assays to measure cdk2 or cdc2 activity were performed by immunoprecipitation of cdk2 or cdc2 as described above. After washes with lysis buffer, the resulting immune complex was washed once with kinase buffer composed of 50 mM Tris (pH 7.4), 10 mM MgCl2, 1 mM DTT, and 0.1 mg/ml BSA. The pellet was resuspended in 50 µl of kinase buffer containing 1 µg of histone H1 (Boehringer), 30 µM ATP, and 5 µCi of [-32P]ATP. The H1 kinase reaction and resolution by SDS-PAGE was as described previously (25)
. Autoradiographs were prepared by exposure to Hyperfilm-MP (Amersham, Arlington Heights, IL).
Immunoblotting.
Cells were detached by trypsinization, pooled with nonattached cells, centrifuged, and washed with PBS. Lysates were prepared in 50 mM Tris-HCl (pH 7.4), 250 mM NaCl, and 5 mM EGTA for 30 min on ice. Lysates were then resolved on 12% polyacrylamide gels, and gel-separated proteins were then transferred to nitrocellulose sheets using a semidry blotting apparatus, except for detection of 14-3-3 (with polyclonal goat anti-14-3-3 antibody from Santa Cruz Biotechnology, CA), where proteins were transferred to Immobilon (Millipore, Bedford, MA). Membranes were blocked with 5% nonfat milk, incubated overnight with primary antibodies, washed, and then incubated with horseradish peroxidase-conjugated goat antirabbit IgG secondary antibodies, as described previously (25)
, with the exception that horseradish peroxidase-conjugated donkey antigoat IgG secondary antibodies (Santa Cruz Biotechnology) were used for the detection of 14-3-3. Protein-antibody complex was detected by enhanced chemiluminescence (Pierce, Rockford, IL).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
in the G2 DNA damage checkpoint, we used isogenic human colon carcinoma (HCT116) cells differing only in the presence or absence of p21WAF1 or 14-3-3. In cells deficient for either p21WAF1 or 14-3-3, the endogenous genes had been deleted by homologous recombination (9
, 20)
. We monitored cell cycle progression by flow cytometry and used a two-dimensional assay (25)
to distinguish 4N cells that were mitotic from those in either G2 or tetraploid G1 (also with 4N DNA content). This assay used MPM-2, a specific marker for mitosis (19)
, and the DNA marker propidium iodide. In previous reports, HCT116 cells were examined at 24 or 48 h after -irradiation, time points that were substantially past the initial entry into G2 after DNA damage (6
, 9)
. To focus on the G2 to M transition, we have examined cells at 15 h after -irradiation, when they were still actively passing through G2-M. Representative results are shown in Fig. 1
. By 15 h, nearly all cells had progressed to G2 after 12 Gy -irradiation, as indicated by the accumulation of cells with a 4N DNA content in the presence of nocodazole (77 and 79%, respectively, of p21+/+ and p21-/- cells; Fig. 1
). At shorter time points only a subset of the cell population had accumulated with a 4N DNA content (data not shown), indicating that cells were actively progressing into G2 at the 15-h time point. Interestingly, both control (wild-type for both p21WAF1 and 14-3-3) and p21-/- cells had entered mitosis by 15 h after -irradiation, as indicated by an accumulation of cells with elevated MPM-2 signals in both populations. However, p21-/- cells showed a greater mitotic population than did control parental cells (26.1% versus 11.7%; Fig. 1A
). Both controls and p21-/- cells had similar cell cycle times, because 78.3 and 66.3%, respectively, accumulated in mitosis after treatment with 0.5 µg/ml nocodazole for 15 h (Fig. 1A)
.
|
-irradiation (Fig. 1B)
. Cells that contained condensed chromosomes, as determined by staining with propidium iodide, also displayed a strong MPM-2 signal. Interphase cells, which were discerned by the presence of intact nuclei, had no visible MPM-2 signal. The presence of an elevated MPM-2 signal is thus restricted to mitotic cells (see also Ref. 25
) and serves as a reliable basis for quantifying mitotic entry after DNA damage.
We determined the effect of DNA damage on progression into mitosis of cells wild-type for both p21WAF1 and 14-3-3, relative to p21-/- or 14-3-3-/- cells. For this, we compared the mitotic indices attained by exposure to nocodazole for 15 h in each cell type (Fig. 2)
. In each case, the mitotic index for DNA damaged cells was measured relative to the index obtained for the same cells without DNA damage. The results of representative experiments are shown in Fig. 2
. Experiments were repeated an average of three times, each time with comparable results.
|
-irradiation (3–12 Gy), Adriamycin (0.05–0.2 µg/ml), and VP-16 (0.2–5 µg/ml; Fig. 2A
).
Because control HCT116 cells do show some arrest in G1 in response to DNA damage (Fig. 1A
; Ref. 20
), we next tested the role of p21WAF1 at G2 independent from its role in the G1 checkpoint by first arresting cells in early S-phase with HU. Cells were then released from HU and were treated with either ionizing radiation, Adriamycin, or VP-16 over a range of doses (Fig. 2B)
. Rates of entry into mitosis were determined by comparing cells exposed to DNA damage then accumulated in mitosis with nocodazole to the same cells exposed to nocodazole alone upon release from HU. Similar to results obtained with unsynchronized cells, we found that p21-/- cells entered mitosis at approximately twice the rate of the parental control cells after DNA damage. As determined by flow cytometry, p21+/+ and p21-/- cells progressed through S-phase at indistinguishable rates after synchronization in early S-phase with hydroxyurea and subsequent exposure to DNA-damaging agents over the full range of doses tested (data not shown). Thus, we conclude that p21+/+ cells display a longer delay in G2 attributable to DNA damage than do p21-/- cells.
In surprising contrast to p21-/- cells, 14-3-3-/- cells reproducibly showed a slightly diminished rate of mitotic entry relative to control cells (wild-type for p21WAF1 and 14-3-3) after -irradiation or treatment with either Adriamycin or VP-16 after release from HU (Fig. 2C)
. These results show that parental HCT116 cells, competent for both p21WAF1 and 14-3-3, do enter mitosis after DNA damage and thus delay, but do not arrest, in G2. Furthermore, p21WAF1, but not 14-3-3, appears to be artially responsible for mediating the G2 delay.
As shown by immunoblotting, DNA damage by each of the agents used here (-irradiation, Adriamycin, and VP-16) induced elevated p21WAF1 levels in control cells relative to untreated controls (Fig. 2D)
. Cells were treated with DNA-damaging agents for 15 h, a time at which control cells accumulated in G2-M (see Fig. 1
). Thus, DNA damage indeed augmented levels of p21WAF1 protein in G2-M, but this induction did not yield an absolute G2 arrest. In addition, increasing the dosage of -irradiation, Adriamycin, or of VP-16 did not further augment the induction of p21WAF1. Because p53 mediates the induction of both p21WAF1 (7
, 8)
and 14-3-3 (12)
, the finding that the degree of G2 delay was dose dependent (Fig. 2)
for parental cells, as well as for p21WAF1-deficient cells, requires that other factors are involved in mediating G2 delay and suggests that these may be p53 independent.
p21WAF1 inhibits cell cycle progression by binding to and inhibiting cyclin-dependent kinases (17)
. To determine the target of p21WAF1 in DNA damage-induced delay in G2, kinase activities of cdk2 or cdc2 immunoprecipitates were measured by radioassay (Fig. 3A)
. cdk2 activity was strongly inhibited in p21+/+ cells treated with Adriamycin, -irradiation, or VP-16 relative to nocodazole-arrested mitotic cells. In contrast, the activity of cdk2 was not suppressed in p21-/- cells after DNA damage induced by Adriamycin, -irradiation, or VP-16, supporting a role for p21WAF1 in the inhibition of cdk2-cyclin A during the G2 delay induced by DNA damage. Although p21WAF1 does not bind cdc2 after DNA damage (Fig. 3B)
, we found that cdc2 protein kinase activity levels in p21+/+ cells were inhibited relative to those in p21-/- cells (Fig. 3A)
. On the basis of quantitative gel scans, we found that cdc2 activity was suppressed 2.0–2.4-fold in p21+/+ cells relative to p21-/- cells accumulated in mitosis with nocodazole after DNA damage. As expected, because cdc2 activation coincides with mitotic entry (28)
, this result mirrors the decreased mitotic index in p21+/+ cells, relative to p21-/- cells, after DNA damage (see Fig. 2A
).
|
, but p21WAF1 was not detected in immunoblots of the same immunoprecipitates (Fig. 3B)
.
Evidence presented above suggested that control cells competent for the induction of both p21WAF1 and 14-3-3 nonetheless enter mitosis, by the criterion of an elevated MPM-2 signal. To verify that control (p21+/+, 14-3-3+/+) cells indeed enter mitosis after DNA damage, we performed immunofluorescence microscopy to assay for the presence of a mitotic spindle and chromosome condensation (Fig. 4A)
. An example of a control (p21+/+, 14-3-3+/+) cell that had progressed to metaphase with a mitotic spindle and condensed chromosomes 24 h after -irradiation (12 Gy) is shown in Fig. 4A
. However, as shown by incomplete chromosome segregation in a cell that had initiated cytokinesis and by the reformation of a lamin envelope around incompletely segregated chromosomes in another example, control cells could not successfully complete reductional division after -irradiation (Fig. 4B)
. By using nocodazole to induce arrest in mitosis, it is evident that control (p21+/+, 14-3-3+/+) cells progress past G2 in a time-dependent manner after either 6 or 12 Gy irradiation (Fig. 4C)
. By 24 h, 52 and 32% of cells had arrested in mitosis after 6 and 12 Gy irradiation, respectively, and subsequent treatment with 0.5 µg/ml nocodazole. Accumulation in mitosis is slower in p21+/+ cells than in p21-/- cells after ionizing radiation (Fig. 4C)
, consistent with a p21WAF1-mediated G2 delay in response to DNA damage.
|
-irradiation (9 Gy), control cells (p21+/+, 14-3-3+/+) displayed a high degree of micronucleation, whereas untreated control cells remained mononucleate (Fig. 5A)
. Note the presence of a recently divided pair of untreated parental (p21+/+, 14-3-3+/+) HCT116 cells, which do not possess a chromosome bridge (indicated by an asterisk), as determined by the two daughter nuclei having discrete nuclear envelopes detected with anti-lamin B antibody.
|
-irradiation over a range of doses (3–12 Gy; data not shown). Micronucleated cells accumulated with time (Fig. 5B)
after -irradiation (12 Gy). In the additional presence of nocodazole, micronucleated cells did not form (Fig. 5B)
. Because nocodazole effectively inhibits mitotic progression and arrests cells in mitosis (see Fig. 1
), these results demonstrate that micronucleation after DNA damage involves exit from mitosis to interphase after abortive mitotic progression (29
, 30)
. Control (p21+/+, 14-3-3+/+) HCT116 cells and those deficient for either p21WAF1 or 14-3-3 all formed binucleate cells with a chromatin bridge, as detected by immunofluorescence microscopy with lamin antibodies and staining of DNA with propidium iodide, after exposure to -irradiation and release from mitotic synchronization with nocodazole (Fig. 5C)
. This demonstrates that the different cells, regardless of their p21WAF1 or 14-3-3 status, progress into mitosis after DNA damage and undergo a mitotic catastrophe, characterized by exit from mitosis without completing chromosome segregation or cytokinesis.
Histograms of DNA content show that at 36 h after -irradiation, parental HCT116 cells (p21+/+, 14-3-3+/+) had largely 4N DNA content (Fig. 6A)
. The control cells had thus become tetraploid after incomplete chromosome segregation induced by DNA damage and arrested in G1 without rereplication of their DNA. By contrast, p21-/- cells rereplicated their DNA and became largely 8N after DNA damage. Unlike p21-/- cells, 14-3-3-/- cells remained 4N. Thus, p21WAF1, but not 14-3-3, is required for the arrest of cells with tetraploid G1 status as a result of DNA damage.
|
increased beginning at 16 h after -irradiation (9 Gy) and remained at high levels through 36 h after -irradiation (Fig. 6B)
. p21WAF1 and 14-3-3 were absent in p21-/- and 14-3-3-/- cells, respectively, as expected, but p21WAF1 and 14-3-3 could each be induced by DNA damage in the absence of the other. Cyclin B is a marker for G2/mitosis (31)
, and cyclin B levels peaked at 16 h after -irradiation, as expected given the accumulation of cells at G2/mitosis at this time point. Cyclin B disappeared in control and 14-3-3-/- cells 24 and 36 h after irradiation, confirming that these cells had exited mitosis. By contrast, cyclin E, which is expressed at late G1/early S-phase (32
, 33)
, was present at high levels 24 and 36 h after -irradiation, consistent with arrest of the tetraploid cells in G1. Both cyclins B and E were present at intermediate levels in p21-/- cells at later times, as expected of cells which continued to cycle.
Given that control (p21+/+, 14-3-3+/+) and 14-3-3-/- cells stably arrest in tetraploid G1 after failure to arrest in G2 and passage through an abortive mitosis (Fig. 6A)
, we next tested whether the arrest was simply attributable to the detection of DNA damage arising in the previous cell cycle (Fig. 7)
, or whether mitotic failure was an important component of the induction of tetraploid G1 arrest. To address this question, we examined the effect of irradiating cells in mitosis on their subsequent arrest in G1. For this, mitotic cells were generated by treatment with nocodazole for 8 h, -irradiated (12 Gy), and then harvested by selective detachment (Fig. 7A, left)
. Cells treated in parallel were then replated after irradiation and released from nocodazole for 29 h (Fig. 7A, right)
and then fixed in preparation for flow cytometry.
|
). The 4N cell population was in tetraploid G1, having exited mitosis within 5 h of release from nocodazole, as verified by loss of MPM-2 signal (data not shown). The generation of a postmitotic 4N population demonstrates that DNA damage in mitosis largely interfered with chromosome segregation and cytokinesis so that the majority of cells became tetraploid G1.
In contrast, if similarly synchronized cells were -irradiated (12 Gy) in diploid G1 5 h after release from nocodazole, they failed to arrest in G1 but instead accumulated as a 4N population (68% of cells had progressed to G2-M or further) within 24 h after irradiation (Fig. 7B, right)
. As a control, if such cells (irradiated 5 h after nocodazole release) were maintained in HU subsequent to DNA damage, they maintained a stable 2N DNA content (Fig. 7B, left)
, demonstrating that they had properly exited mitosis and divided but had not entered S-phase prior to induction of DNA damage.
Taken together, our results demonstrate that DNA damage at G1 or G2 is not alone sufficient to induce a stable cell cycle arrest, but that the combination of DNA damage and abortive mitosis does induce a stable p21WAF1-dependent arrest in a tetraploid G1 state.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Is Partially Responsible for G2 Delay in Response to DNA Damage. upon mitotic entry within a single cell cycle (15 h) after DNA damage in interphase. We found that HCT116 human colon carcinoma cells progress into and through mitosis with evident genomic damage, as demonstrated by their inability to complete chromosome segregation, irrespective of whether the genes for either p21WAF1 or 14-3-3 are present. Because we show here that p21WAF1 competent cells subsequently arrest in tetraploid G1 with 4N DNA content (see Fig. 6
), we conclude that analysis of the roles of p21WAF1 and 14-3-3 in the G2 checkpoint cannot be made simply upon the basis of DNA content. Instead, such an analysis additionally requires examining the competence for progression from G2 to mitosis within the first cell cycle after DNA damage.
For this purpose, we have used a biparameter flow cytometric assay with MPM-2 antibody to quantitate entry into mitosis. With this approach, we found that the absence of p21WAF1, but not 14-3-3, increases the rate of mitotic entry, shortening the G2 delay in response to DNA damage. Both p21WAF1 and 14-3-3 are induced in a p53-dependent manner after DNA damage (7
, 8 , 12)
. Although we also found that they are induced after DNA damage under our experimental conditions, our results clearly show their presence is not sufficient to mediate sustained G2 arrest. Because p21-/- cells still exhibit a DNA damage-induced delay in G2, it is apparent that p21WAF1 is only partially responsible for the observed G2 delay in human colon carcinoma cells in response to DNA damage.
With respect to inhibitory mechanisms independent of p53 (34)
, a variety of DNA-damaging agents, including -radiation and VP-16, induce inhibitory phosphorylation of cdc2 at T14 and Y15 (35, 36, 37)
, which in turn mediates G2 delay after DNA damage in human cells. Such inhibitory phosphorylation of cdc2 likely participates in the G2 delay induced by DNA damage in both p21-/- and 14-3-3-/- HCT116 cells.
Adaptation to the G2 DNA damage checkpoint has been described in yeast (21, 22, 23)
. It appears that HCT116 cells similarly adapt to G2 arrest in response to DNA damage induced by various agents. We found that p21WAF1 delays but does not prevent G2 adaptation, whereas 14-3-3 is without effect. Galgoczy and Toczyski (23)
demonstrated recently that adaptation precedes genomic instability induced by DNA damage in yeast. Chromosome bridging during mitosis in HCT116 cells (Fig. 5)
, including those that are p21WAF1 competent, suggests that failure of p21WAF1 to prevent adaptation similarly results in genomic damage in HCT116 cells.
Our findings indicate that p21WAF1 binds both specifically to cdk2-cyclin A and inhibits its activity after late cell cycle DNA damage. In the normal cell cycle, both cyclin A levels and cdk2 activity peak at G2 (38) . Although microinjection of antibodies directed against cyclin A induces G2 arrest (38) , the role of cdk2-cyclin A at G2 has not been well defined. Given that cdk2-cyclin A appears to be a specific target of p21WAF1 after DNA damage in G2, we propose that suppression of cdk2 activity at G2 is required for implementation of the G2 DNA damage checkpoint by p21WAF1. However, it is evident from our results that inhibition of cdk2-cyclin A activity by p21WAF1 can mediate only a transient G2 delay and is not sufficient to induce stable G2 arrest in HCT116 cells after DNA damage. Although there have been reports that p21WAF1 binds to cyclin B-cdc2 (39 , 40) , we have found no detectable association of p21WAF1 with cyclin B or cdc2 after late cell cycle DNA damage in HCT116 cells. This result is consistent with reports that p21WAF1 is associated with only a small fraction of cyclin-complexed cdc2 after UV-induced DNA damage (41) , and that p21WAF1 binds with lower affinity to cdc2-cyclin B in vitro than to either cyclin A or cdk2 (42) .
Chan et al. (9)
reported that the absence of 14-3-3 suppressed G2 arrest after DNA damage and caused death upon entry into mitosis. We found, in contrast, that the absence of 14-3-3 does not diminish the period of delay in G2 but actually appears to extend it to a small but reproducible extent. Furthermore, we found that cell death in 14-3-3-/- cells does not occur in mitosis but instead occurs in tetraploid G1 (see below). We conclude that 14-3-3 does not play a role in the G2 delay in response to DNA damage.
Furthermore, it was reported (9)
that 14-3-3 is required to prevent mitotic catastrophe, defined as cell death induced during inappropriate mitotic entry after DNA damage. Here we find that all HCT116 human colon carcinoma cells tested, even those containing p21WAF1 and 14-3-3, enter mitosis and undergo chromosome missegregation and exit mitosis after DNA damage. Cell death as reported by Chan et al. (9)
was evident at 32 or more h after DNA damage, which we have demonstrated to be substantially longer than the time course of the entry and exit from mitosis of both control and 14-3-3-/- cells after DNA damage. Thus, it is evident that it is not simply entry into mitosis or chromosome missegregation that induces cell death in human colon carcinoma cells after DNA damage. We conclude that the presence of neither p21WAF1 nor 14-3-3 is sufficient to prevent mitotic catastrophe. It is important to note that we define mitotic catastrophe as it was originally defined in yeast (43)
, as a severe chromosome missegregation event.
In yeast, mitotic catastrophe is accompanied by cell death (43)
. In contrast, in this work we found that cell death does not follow immediately after and is not necessarily linked to chromosome missegregation events. Control cells (p21+/+, 14-3-3+/+) do not show substantial cell death after DNA damage (data not shown and Ref. 24
), although they do undergo chromosome missegregation (Fig. 5)
. Thus, any role of 14-3-3 in protecting against cell death after DNA damage must take place in G1 after a mitotic catastrophe and is not attributable to preventing improper mitotic entry, as reported previously (9)
. Our results thus suggest that death principally occurs in 14-3-3-/- cells that are arrested in tetraploid G1 after DNA damage.
p21WAF1 but not 14-3-3 Is Required for the Arrest of Tetraploid Cells in G1 after DNA Damage.
Unless deficient for p21WAF1, HCT116 cells arrest as tetraploid cells in a G1 end state after DNA damage. The G1 state is defined by the criteria of micronucleation, absence of cyclin B, and presence of cyclin E. Micronucleation of cells, as found here, indicates that cells have passed through mitosis without completing cell division (29
, 30)
. As expected of micronucleation attributable to mitotic failure, nocodazole treatment, which maintains cells in mitosis, also suppresses formation of micronuclei after DNA damage.
The means by which cells fail in mitosis is evident. Our results (Fig. 5)
show that the cells initiate but do not complete cleavage because of the presence of unseparated chromosomes bridging the cleavage furrow. Such a result has been reported for HCT116 cells deficient for p21WAF1 (6)
but not for HCT116 parental cells, which are wild-type for p21WAF1 (as well as 14-3-3). These results are similar to those observed when cleavage is blocked by the inhibition of topoisomerase II-dependent DNA decatenation required for chromosome separation (44)
. As further evidence that HCT116 cells that are wild-type for p21WAF1 and 14-3-3 or wild-type for p21WAF1 but deficient for 14-3-3 have proceeded past mitosis and arrested in G1, cyclin B is absent 24 h after DNA damage. Cyclin B forms a heterodimer with cdc2 and is required for cdc2 activity in mitosis. It is normally maximally present during G2 and mitosis (31)
, then is specifically destroyed by proteolysis during the metaphase to anaphase transition (45)
, and is absent in G1.
A recent report showed that in HCT116 cells, among other cell lines, both cyclin B and cdc2 are down-regulated after DNA damage with the interpretation that this suppression was necessary for G2 arrest (46) . It should be noted that the absence of DNA rereplication, and not mitotic entry, was used as the criterion of arrest in G2. Given our demonstration of cyclin B suppression in micronucleated HCT116 cells after DNA damage, we conclude that the loss of cyclin B arises instead from arrest of tetraploid cells in G1 after a mitotic catastrophe induced by DNA damage.
Another cell cycle marker, cyclin E, is maximally expressed in G1 (32 , 33) and is absent in G2 and mitosis. In our results, cyclin E becomes detectable 24 h after DNA damage and is maintained at high levels at subsequent time points, again supporting the likelihood of G1 arrest.
The arrest of tetraploid cells after DNA damage is p21WAF1 dependent. In this respect, it is like other G1 checkpoints, following G1 DNA damage (10
, 11)
, or mitotic exit in the absence of a functioning mitotic spindle (47
, 48)
. Although yeast similarly show an adaptation to the G2 DNA damage checkpoint, the consequences are different in HCT116 cells because of the presence of the p21WAF1-dependent arrest of tetraploid cells in G1. The chromosome bridging seen in all HCT116 cell lines observed, including parental cells (p21+/+, 14-3-3+/+) and their p21-/- and 14-3-3-/- derivatives, is indicative of genomic damage and is also a likely source of further genomic damage. But, unlike yeast which continue to proliferate after adaptation and damage to the genome, p21WAF1-competent HCT116 cells arrest in tetraploid G1, thereby preventing the replication of damaged cells.
We found, in surprising contrast, that neither the G2 DNA damage checkpoint nor G1 arrest of tetraploid cells after DNA damage is dependent on 14-3-3. Because 14-3-3 is induced in a p53-dependent manner by DNA damage (12)
, it is reasonable to ask what the true role of 14-3-3 might be, if not to mediate cell cycle arrest. One attractive possibility is that 14-3-3 plays a role in preventing apoptosis after DNA damage. HCT116 cells with 14-3-3 deleted die by apoptosis at time points after DNA damage later than shown in this report, when they must be arrested in tetraploid G1 (Ref. 9
and data not shown). The possibility that 14-3-3 plays such a role is supported by the recent finding that a dominant-negative mutant of another 14-3-3 isoform, 14-3-3
, induces apoptosis after various stimuli, including UV irradiation (49)
. It is interesting to note that such a putative role for 14-3-3 in suppressing apoptosis after mitotic catastrophe would appear to parallel a function reported previously for survivin (50)
, a BIR motif containing mitotic passenger protein (51
, 52) .
Implications for Tumor Chemotherapy.
It is unlikely that the tetraploid G1 arrest of HCT116 cells after DNA damage and mitotic catastrophe is entirely attributable to DNA damage incurred in the previous cell cycle, because DNA damage in either G1 or G2 does not by itself lead to a stable arrest in HCT116 cells. As we have shown recently (53)
, the tetraploid state arising from a failure to undergo cell cleavage can itself generate a G1 delay or arrest in p53 competent cells. In this context, it is important to note that the tetraploid state itself appears to induce a stable G1 arrest in p21-competent HCT116 cells that become tetraploid in response to inhibitors of mitotic spindle assembly (48)
. In G1, HCT116 cells that have undergone mitotic catastrophe after DNA damage may exhibit a relatively stable arrest attributable to the tetraploid state of the cell. Such a tetraploidy checkpoint might function to prevent progression to gross aneuploidy in cells that are tetraploid and that have persistent DNA damage after a DNA damage-induced mitotic catastrophe (53)
.
The goal of chemotherapy or radiotherapy is to selectively kill tumor cells. A large number of agents that are used clinically induce DNA damage (54) . It has been shown that p21WAF1-deficient HCT116 human colon cancer cells are more sensitive to DNA damage both in culture and in xenografts than p21WAF1-competent cells (24 , 55 , 56) . Our results suggest that the p21WAF1-dependent G1 arrest of tetraploid cells after DNA damage may be a critical determinant of this sensitivity. Further characterization of components of the tetraploid G1 arrest after DNA damage may both elucidate new targets for therapy and define further genetic determinants of successful therapy.
|
ACKNOWLEDGMENTS |
|---|
. |
FOOTNOTES |
|---|
1 This work was supported by a grant from La Ligue Nationale Contre le Cancer ("Laboratoire Labelisée"). 
2 To whom requests for reprints should be addressed, at Institut de Biologie Structurale Jean-Pierre Ebel, 41 rue Jules Horowitz, 38027 Grenoble Cedex 1, France. Phone: (33)-4-38-78-96-16; Fax: (33)-4-38-78-54-94; E-mail: margolis{at}ibs.fr
3 The abbreviations used are: CDK, cyclin-dependent kinase; VP-16, etoposide; HU, hydroxyurea.
Received 3/ 5/01. Accepted 8/ 8/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
is required to prevent mitotic catastrophe after DNA damage. Nature (Lond.), 401: 616-620, 1999.[Medline]
This article has been cited by other articles:
|
|
 |
|
  Z. HAN, K. DIMAS, X. TIAN, Y. WANG, H. HEMMI, K. YAMADA, N. KATO, P. PANTAZIS, R. J. RAMANUJAM, S. ANANT, et al. 14-3-3{sigma}-Dependent Resistance to Cisplatin Anticancer Res, June 1, 2009; 29(6): 2009 - 2014. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Nair, A. L. Ho, A. N. Tse, J. Coward, H. Cheema, G. Ambrosini, N. Keen, and G. K. Schwartz Aurora B Kinase Regulates the Postmitotic Endoreduplication Checkpoint via Phosphorylation of the Retinoblastoma Protein at Serine 780 Mol. Biol. Cell, April 15, 2009; 20(8): 2218 - 2228. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Lee, J. A. Kim, V. Barbier, A. Fotedar, and R. Fotedar DNA Damage Triggers p21WAF1-dependent Emi1 Down-Regulation That Maintains G2 Arrest Mol. Biol. Cell, April 1, 2009; 20(7): 1891 - 1902. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bataller, C. Mendez, J. A. Salas, and J. Portugal Mithramycin SK modulates polyploidy and cell death in colon carcinoma cells Mol. Cancer Ther., September 1, 2008; 7(9): 2988 - 2997. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Qiao, X. Yang, K. Meyer, and A. Friedl Glypican-1 Regulates Anaphase Promoting Complex/Cyclosome Substrates and Cell Cycle Progression in Endothelial Cells Mol. Biol. Cell, July 1, 2008; 19(7): 2789 - 2801. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Vigneron, E. Gamelin, and O. Coqueret The EGFR-STAT3 Oncogenic Pathway Up-regulates the Eme1 Endonuclease to Reduce DNA Damage after Topoisomerase I Inhibition Cancer Res., February 1, 2008; 68(3): 815 - 825. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. Levesque, A. A. Fanous, A. Poh, and A. Eastman Defective p53 signaling in p53 wild-type tumors attenuates p21waf1 induction and cyclin B repression rendering them sensitive to Chk1 inhibitors that abrogate DNA damage-induced S and G2 arrest Mol. Cancer Ther., February 1, 2008; 7(2): 252 - 262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Stevens, G. Liu, S. W. Bremer, K. J. Ye, W. Xu, J. Xu, Y. Sun, G. S. Wu, S. Savasan, S. A. Krawetz, et al. Mitotic Cell Death by Chromosome Fragmentation Cancer Res., August 15, 2007; 67(16): 7686 - 7694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Levesque and A. Eastman p53-based cancer therapies: is defective p53 the Achilles heel of the tumor? Carcinogenesis, January 1, 2007; 28(1): 13 - 20. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Vigneron, J. Cherier, B. Barre, E. Gamelin, and O. Coqueret The Cell Cycle Inhibitor p21waf1 Binds to the myc and cdc25A Promoters upon DNA Damage and Induces Transcriptional Repression J. Biol. Chem., November 17, 2006; 281(46): 34742 - 34750. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Churikov, C. Wei, and C. M. Price Vertebrate POT1 Restricts G-Overhang Length and Prevents Activation of a Telomeric DNA Damage Checkpoint but Is Dispensable for Overhang Protection. Mol. Cell. Biol., September 1, 2006; 26(18): 6971 - 6982. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Blank, Y. Lerenthal, L. Mittelman, and Y. Shiloh Condensin I recruitment and uneven chromatin condensation precede mitotic cell death in response to DNA damage J. Cell Biol., July 17, 2006; 174(2): 195 - 206. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. B. Affar, F. Gay, Y. Shi, H. Liu, M. Huarte, S. Wu, T. Collins, E. Li, and Y. Shi Essential dosage-dependent functions of the transcription factor yin yang 1 in late embryonic development and cell cycle progression. Mol. Cell. Biol., May 1, 2006; 26(9): 3565 - 3581. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Klochendler-Yeivin, E. Picarsky, and M. Yaniv Increased DNA Damage Sensitivity and Apoptosis in Cells Lacking the Snf5/Ini1 Subunit of the SWI/SNF Chromatin Remodeling Complex. Mol. Cell. Biol., April 1, 2006; 26(7): 2661 - 2674. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Bhonde, M.-L. Hanski, J. Budczies, M. Cao, B. Gillissen, D. Moorthy, F. Simonetta, H. Scherubl, M. Truss, C. Hagemeier, et al. DNA Damage-induced Expression of p53 Suppresses Mitotic Checkpoint Kinase hMps1: THE LACK OF THIS SUPPRESSION IN p53MUT CELLS CONTRIBUTES TO APOPTOSIS J. Biol. Chem., March 31, 2006; 281(13): 8675 - 8685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Vigneron, I. B. Roninson, E. Gamelin, and O. Coqueret Src Inhibits Adriamycin-Induced Senescence and G2 Checkpoint Arrest by Blocking the Induction of p21waf1 Cancer Res., October 1, 2005; 65(19): 8927 - 8935. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Huang, T. Tran, L. Zhang, R. Hatcher, and P. Zhang DNA damage-induced mitotic catastrophe is mediated by the Chk1-dependent mitotic exit DNA damage checkpoint PNAS, January 25, 2005; 102(4): 1065 - 1070. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. E. Bosco, C. N. Mayhew, R. F. Hennigan, J. Sage, T. Jacks, and E. S. Knudsen RB signaling prevents replication-dependent DNA double-strand breaks following genotoxic insult Nucleic Acids Res., January 2, 2004; 32(1): 25 - 34. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Scott, J. D. Earle, and P. H. Gumerlock Functional p53 Increases Prostate Cancer Cell Survival After Exposure to Fractionated Doses of Ionizing Radiation Cancer Res., November 1, 2003; 63(21): 7190 - 7196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P.H. Chow, W. Y. Siu, T. K. Fung, W. M. Chan, A. Lau, T. Arooz, C.-P. Ng, K. Yamashita, and R. Y.C. Poon DNA Damage during the Spindle-Assembly Checkpoint Degrades CDC25A, Inhibits Cyclin-CDC2 Complexes, and Reverses Cells to Interphase Mol. Biol. Cell, October 1, 2003; 14(10): 3989 - 4002. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. T. Barrett, D. Pritchard, C. Palanca-Wessels, J. Anderson, B. J. Reid, and P. S. Rabinovitch Molecular Phenotype of Spontaneously Arising 4N (G2-Tetraploid) Intermediates of Neoplastic Progression in Barrett's Esophagus Cancer Res., July 15, 2003; 63(14): 4211 - 4217. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. D. Lohez, C. Reynaud, F. Borel, P. R. Andreassen, and R. L. Margolis Arrest of mammalian fibroblasts in G1 in response to actin inhibition is dependent on retinoblastoma pocket proteins but not on p53 J. Cell Biol., April 14, 2003; 161(1): 67 - 77. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Borel, O. D. Lohez, F. B. Lacroix, and R. L. Margolis Multiple centrosomes arise from tetraploidy checkpoint failure and mitotic centrosome clusters in p53 and RB pocket protein-compromised cells PNAS, July 23, 2002; 99(15): 9819 - 9824. [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 |
