Caspase-3–Dependent Mitotic Checkpoint Inactivation by the Small-Molecule Inducers of Mitotic Slippage SU6656 and Geraldol

Published OnlineFirst March 25, 2011; DOI: 10.1158/1535-7163.MCT-10-0909

Molecular Cancer Preclinical Development Therapeutics

Jenna L. Riffell1, Reiner U. Janicke€ 2, and Michel Roberge1

Abstract Microtubule-targeting cancer drugs such as paclitaxel block cell-cycle progression at mitosis by prolonged activation of the mitotic checkpoint. Cells can spontaneously escape mitotic arrest and enter interphase without chromosome segregation by a process termed mitotic slippage that involves the degradation of cyclin B1 without mitotic checkpoint inactivation. Inducing mitotic slippage with chemicals causes cells to die after multiple rounds of DNA replication without cell division, which may enhance the antitumor activity of microtubule-targeting drugs. Here, we explore pathways leading to mitotic slippage by using SU6656 and geraldol, two recently identified chemical inducers of mitotic slippage. Mitotic slippage induced by SU6656 or geraldol was blocked by the proteasome inhibitor MG-132 and involved proteasome-dependent degradation of cyclin B1 and the mitotic checkpoint proteins budding uninhibited by benzimidazole related 1 (BubR1) and cell division cycle 20 (Cdc20) in T98G cells. Mitotic slippage and the degradation of BubR1 and Cdc20 were also inhibited by the caspase-3 and -7 inhibitor DEVD-CHO. MCF-7 cells lacking caspase-3 expression could not degrade BubR1 or undergo mitotic slippage in response to SU6656 or geraldol. Introduction of caspase-3 completely restored the ability of MCF-7 cells to degrade BubR1 and undergo mitotic slippage. However, lack of expression of caspase-3 did not affect cell death after exposure to paclitaxel, with or without mitotic slippage induction. The requirement for caspase-3 for chemically induced mitotic slippage reveals a new mechanism for mitotic exit and a link between mitosis and apoptosis that has implications for the outcome of cancer chemotherapy. Mol Cancer Ther; 10(5); 839–49. 2011 AACR.

Introduction mosome separation when even 1 kinetochore is unattached. Exposure to drugs that interfere with microtubule During cell division, genetic integrity is maintained by dynamics, such as the taxanes (4) and the Vinca alkaloids ensuring that all chromosomes are attached to microtu- (5), similarly activates the mitotic checkpoint and arrests bules emanating from both poles of the mitotic spindle cells at mitosis, effectively preventing further prolifera- before segregation of sister chromatids begins (1). This tion. process is monitored by the mitotic checkpoint, which The mitotic checkpoint acts through inhibition of the prevents initiation of anaphase until every kinetochore is anaphase-promoting complex/cyclosome (APC/C; ref. attached and tension between kinetochores of paired 6), the E3 ubiquitin ligase (7) that, when activated by sister chromatids is sufficient, ensuring biorientation cofactors cell division cycle 20 (Cdc20) or Cdh1 (8), (2). To prevent aneuploidy and ensuing genetic defects polyubiquitylates the cyclin-dependent kinase 1 (Cdk1) leading to cell death or tumorigenesis (3), the mitotic cofactor cyclin B1 (7) and the separase regulator securin checkpoint must be sufficiently sensitive to delay chro- (9), targeting them for degradation by the proteasome. This results in inactivation of Cdk1, separation of sister chromatids, and exit from mitosis. The key components Authors' Affiliations: 1Department of Biochemistry and Molecular Biol- ogy, University of British Columbia, Vancouver, British Columbia, Canada; of the mitotic checkpoint are budding uninhibited by and 2Laboratory for Molecular Radiooncology, Clinic and Policlinic for benzimidazole related 1 (BubR1), budding uninhibited Radiation Therapy and Radiooncology, Heinrich Heine Universitat€ by benzimidazole 3 (Bub3), and Cdc20, which form a Dusseldorf,€ Dusseldorf,€ Germany mitotic checkpoint complex (MCC; ref. 10). This complex Note: Supplementary data for this article are available at Molecular Cancer is the main inhibitor of APC/C activity, along with Therapeutics Online (http://mct.aacrjournals.org/). mitotic arrest dependent 2 (Mad2), which initially binds Corresponding Author: Michel Roberge, Department of Biochemistry and Molecular Biology, University of British Columbia, 2350 Health Cdc20 (11) and catalyzes its binding to BubR1 and sub- Sciences Mall, Vancouver, British Columbia, Canada V6T 1Z3. Phone: sequent formation of the MCC (12). Cdc20 is an activating 604-822-2304; Fax: 604-822-5227. E-mail: [email protected] cofactor of APC/C during mitosis (8); an active mitotic doi: 10.1158/1535-7163.MCT-10-0909 checkpoint inhibits APC/C through APC/C-dependent 2011 American Association for Cancer Research. polyubiquitylation of Cdc20 and subsequent degradation

www.aacrjournals.org 839

Riffell et al.

by the proteasome (12). BubR1 binds to and inhibits both Materials and Methods Cdc20 (13) and APC/C itself (14), acting as a pseudosub- strate inhibitor that, depending on acetylation status, can Cell culture and chemicals be actively degraded by APC/CCdc20 (15). The role of T98G cells, obtained from the American Type Culture Bub3 in the MCC is unclear, although in fission yeast it is Collection (ATCC; characterized by short tandem repeat involved in MCC localization (16). Other components of analysis) and used within 6 months of resuscitation, were the mitotic checkpoint include the kinases Bub1, mono- maintained in Dulbecco’s Modified Eagle’s Medium polar spindle 1 (Mps1), and Aurora B (2). (Invitrogen) supplemented with 10% FBS (Gibco). Caspases have well-characterized apoptotic functions, MCF-7 cell lines, obtained from the ATCC and stably but caspase-3 and caspase-7 have both recently been transfected with empty vector (pcDNA) or caspase-3 observed to play a role, yet to be defined, in mitotic (casp3), were maintained in RPMI (Invitrogen) supple- progression (17, 18). Their activities are tightly regulated mented with 10% FBS and 10 mmol/L HEPES, pH 7.3 and must be restrained during mitotic stress to prevent (Invitrogen). Paclitaxel was obtained from USB, SU6656, extensive cell death, most notably through survivin, and MG-132 from Sigma, geraldol from Chromadex, and which inhibits caspase activation during mitotic arrest cell-permeable DEVD-CHO from Enzo Life Sciences. and functions as part of the mitotic checkpoint machinery (19). Slippage induction assay Mitotic checkpoint activation during an unperturbed T98G cells at 75% confluency were treated with 30 mitosis provides sufficient time for microtubule attach- nmol/L paclitaxel, or MCF-7 cells were treated with 50 ment, preventing aneuploidy (20) and increasing cell nmol/L paclitaxel, for 20 hours at 37 C, and mitotic cells survival (21). However, long-term activation of the were harvested by shake-off, counted using a hemacyt- mitotic checkpoint during exposure to antimitotic ometer, seeded in a 96-well plate (PerkinElmer View- agents can be problematic because chromosome con- plate) at 5,000 cells per well, and treated with chemicals as densation hinders RNA transcription (22). With time, an indicated for 4 hours at 37 C. Unattached mitotic cells imbalance between new protein production and protein were then aspirated and discarded while attached, degradation may cause the levels of proteins essential to slipped cells were fixed in 3% paraformaldehyde maintain mitotic arrest to fall, triggering mitotic slip- (EMD) in PBS for 15 minutes at room temperature, and page. Also termed mitotic checkpoint adaptation, mito- stained with Hoechst 33342 (Invitrogen) in PBS for 10 tic slippage occurs when cells exit mitosis without minutes at room temperature. Five fields per well were chromosome segregation or cell division (20, 23) and counted by a Cellomics ArrayScan VTI automated fluor- results from slow APC/CCdc20- and proteasome-depen- escence imager (ThermoFisher) by using a 10Â objective. dent degradation of cyclin B1 in the presence of an Individual nuclei of slipped cells were detected and active mitotic checkpoint (24, 25). Cells that have under- counted using the Cellomics Target Activation Analysis gone mitotic slippage enter a G1-like state with decon- Program. In all figures, mitotic slippage was expressed as densed chromosomes that form multiple micronuclei a percentage of the cells seeded in each well (26). (23), allowing resumption of transcription and other cellular processes. Immunoblotting Our group and others have identified chemicals that Cells were washed in PBS and lysed for 5 minutes on stimulate mitotic slippage and observed that slipped ice in lysis buffer containing 20 mmol/L Tris-HCl cells typically undergo at least 1 round of DNA replica- (Fisher), pH 7.5, 150 mmol/L NaCl (Fisher), 1 mmol/L tion without subsequent cell division but that, even- EDTA (Sigma), 1 mmol/L EGTA (Sigma), 1% Triton tually, all cells that undergo mitotic slippage die (26– X-100 (LabChem Inc.), 2.5 mmol/L sodium pyropho- 30). Known chemical inducers of mitotic slippage sphate (Fisher), 1 mmol/L b-glycerol phosphate (Sigma), include CDK1 inhibitors (roscovitine, RO3066; ref. 1 mmol/L sodium orthovanadate (Sigma), and 1Â pro- 28), histone deacetylase complex inhibitors (SBHA, tease inhibitor cocktail (Roche). Lysates were spun at SAHA, sodium butyrate, trichostatin A; refs. 31, 32), 15,000 Â g for 15 minutes, and supernatants were and Aurora inhibitors [ZM447439 (33), MLN8054 (34), removed and assayed for protein concentration by using Go6976€ (29), OM137 (27), and fisetin (30); Supplemen- the Bradford assay (Sigma). Sample concentration was tary Fig. S1]. We previously identified SU6656 and equalized and diluted in 50 mmol/L Tris-HCl (Fisher), geraldol as chemical inducers of mitotic slippage that pH 6.8, 2% SDS (Fisher), 0.1% bromophenol blue (Sigma), increased cell killing after induction of mitotic arrest by and 10% glycerol (Fisher), run on a 12% acrylamide (Bio- microtubule-targeting agents. This study investigates Rad) gel, and stained with Coommassie Brilliant Blue to how chemicals can modulate mitotic slippage, reveals verify equal protein loading or transferred to a polyvi- a mechanism for mitotic slippage that is different from nylidene difluoride membrane (Millipore Immobilon-P). that described for spontaneous mitotic slippage, and The membrane was blocked in 5% milk (Nestle) in TBS shows how interplay between pathways associated with containing 0.1% Tween-20 (TBS-T; MP Biomedicals) for mitosis and apoptosis can contribute to the outcome of 30 minutes and incubated overnight at 4 C with primary antimitotic cancer treatments. antibody in 5% milk in TBS. Membranes were then

840 Mol Cancer Ther; 10(5) May 2011 Molecular Cancer Therapeutics

Caspase-3–Dependent Chemical Induction of Mitotic Slippage

washed 2 Â 10 minutes in TBS-T, incubated at room ing that mitotic slippage induced by both chemicals temperature with secondary antibody in 5% milk for 1 requires proteasomal protein degradation. Both chemi- hour, washed 3 Â 10 minutes in TBS-T, and imaged by cals caused complete disappearance of cyclin B1, as chemiluminescence (Millipore Immobilon Western). determined by immunoblotting, and this degradation Antibodies used were mouse a-cyclin B1 (1:100; BD was entirely prevented by coincubation with MG-132 Pharmingen), mouse a-BubR1 (1:1,000; BD Transduc- (Fig. 1B). Therefore, induction of mitotic slippage by these tion), mouse a-Mad2 (1:500; Santa Cruz), mouse chemicals is similar to normal exit from mitosis and a-p55CDC/Cdc20 (1:1,000; Santa Cruz), mouse a-Mps1 spontaneous slippage with respect to proteasome depen- (1:500; Abcam), goat a-mouse horseradish peroxidase dence and cyclin B1 degradation. (1:10,000), and goat a-rabbit peroxidase conjugate (1:10,000). Induction of mitotic slippage involves proteasome- dependent degradation of mitotic checkpoint In vitro kinase assays proteins SU6656 or geraldol were incubated for 20 to 30 minutes Although spontaneous mitotic slippage occurs via the at room temperature with 20 to 40 nmol/L active kinase, degradation of cyclin B1 by APC/CCdc20, it does not 0.2 mg/mL myelin basic protein (Aurora kinases), or 0.4 involve inactivation of the mitotic checkpoint (24, 25). mg/mL synthetic Src substrate (KVEKIGEGTYGVVYK) Using immunoblotting, the effects of SU6656 and geral- and 50 mmol/L 33P-ATP in kinase assay buffer containing dol on the cellular levels of the main mitotic checkpoint 25 mmol/L MOPS, pH 7.2, 12.5 mmol/L b-glycerol mediators Cdc20, BubR1, Mad2, and Mps1 were exam- phosphate, 25 mmol/L MgCl2, 5 mmol/L EGTA, 2 ined. Levels of these proteins increased in cells arrested in mmol/L EDTA, and 0.25 mmol/L DTT (Aurora kinases) mitosis (Fig. 1B), and subsequent exposure to SU6656 and or 25 mmol/L MOPS, pH 7.2, 12.5 mmol/L b-glycerol geraldol caused complete degradation of BubR1 and phosphate, 20 mmol/L MgCl2, 25 mmol/L MnCl2,5 Cdc20 and a sizeable reduction in Mps1 levels. The mmol/L EGTA, 2 mmol/L EDTA, and 0.25 mmol/L degradation of these 3 proteins was prevented by coin- DTT (Src). Ten microliters of this reaction mixture was cubation with MG-132 (Fig. 1B). Exposure to SU6656, but then spotted on a phosphocellulose Multiscreen plate and not geraldol, decreased cellular levels of Mad2 and this washed 3 Â 15 minutes in 1% phosphoric acid. Scintilla- depletion was not proteasome dependent (Fig. 1B). tion fluid was added and the radioactivity on the plate In T98G cells, BubR1 is not degraded during comple- was counted using a Trilux scintillation counter against a tion of mitosis (Supplementary Fig. S2), indicating that its control incubated without substrate. degradation in cells induced to undergo mitotic slippage by SU6656 or geraldol is not simply a normal conse- Results quence of exiting mitosis. Depletion of BubR1 or Mps1 has previously been shown to be sufficient to inactivate Induction of mitotic slippage by SU6656 and the mitotic checkpoint (14, 35, 36). Therefore, our results geraldol requires proteasomal activity imply that SU6656 and geraldol induce mitotic slippage A previous screening effort by our group identified through degradation of mitotic checkpoint proteins. SU6656 and geraldol as chemicals capable of inducing mitotic slippage (26). When cells spontaneously slip out SU6656 and geraldol do not induce the degradation of mitotic arrest, as well as during normal exit from of cyclin B1, BubR1, or Cdc20 in interphase cells mitosis, proteasomal activity is required for the degrada- To determine whether SU6656 and geraldol induce the tion of cyclin B1 (24, 25). First, to determine whether proteasome-dependent degradation of cyclin B1, BubR1, escape from mitotic arrest induced by SU6656 or geraldol and Cdc20 in interphase and in mitotic cells, proliferating similarly requires proteasome activity, mitotic slippage T98G cells, which comprise 98% interphase cells, were was examined in the presence of the proteasome inhibitor exposed to 5 mmol/l SU6656 or 5 mmol/L geraldol for 4 MG-132. T98G cells at 75% confluency were arrested in hours and analyzed by immunoblotting. No decrease in mitosis by exposure to 30 nmol/L paclitaxel for 20 hours, the levels of cyclin B1, BubR1, or Cdc20 was observed harvested via shake-off, and seeded in 96-well plates. The (Fig. 1C); rather, a slight increase in the levels of all 3 cells were exposed to the chemical inducers of mitotic proteins was observed. Simultaneous treatment with 20 slippage SU6656 (5 mmol/L) or geraldol (5 mmol/L) in the mmol/L MG-132 had no additional effect (Fig. 1C). These presence of various concentrations of MG-132 for 4 hours results indicate that an active mitotic checkpoint is and in the continued presence of paclitaxel. Residual required for SU6656- and geraldol-induced selective unattached mitotic cells were removed and the attached, degradation of mitotic checkpoint components. slipped cells were fixed, stained with Hoechst 33342, and quantified using automated fluorescence microscopy Mitotic checkpoint inactivation and slippage (26). In the absence of MG-132, SU6656 and geraldol induction by SU6656 and geraldol require caspase-3 induced 60% to 90% of mitotic cells to undergo mitotic Caspases have been implicated in mitotic progression slippage. MG-132 reduced the proportion of slipped cells (18), and, in particular, BubR1 is reportedly degraded by in a concentration-dependent manner (Fig. 1A), indicat- the effector caspase-3 during exit from mitosis (17),

www.aacrjournals.org Mol Cancer Ther; 10(5) May 2011 841

Riffell et al.

A B –++ +++ Paclitaxel 100 – – + + – – SU6656 SU6656 Figure 1. Mitotic slippage occurs – – – – + + Geraldol 80 Geraldol through proteasome-dependent ––– +–+ MG-132 degradation of mitotic checkpoint 60 proteins. A, T98G cells arrested in Cyclin B1 mitosis by 30 nmol/L paclitaxel were harvested by shake-off, 40 BubR1 seeded in 96-well plates, and exposed to 5 mmol/L SU6656 or 5 % Cells slipped Cdc20 mmol/L geraldol simultaneously 20 with 0 to 75 mmol/L MG-132. After Mad2 4 hours, the attached, slipped 0 cells were fixed, stained, and 051015 20 Mps1 quantified using an automated Concentration MG-132 (µmol/L) fluorescence imager. Error bars represent 95% CIs. B, mitotic T98G cells were harvested C –+ +– – SU6656 by shake-off and exposed to 5 mmol/L SU6656 or 5 mmol/L –– –+ + Geraldol geraldol without or with 20 mmol/L –– +– + MG-132 MG-132 for 4 hours. Lysates were immunoblotted for the indicated Cyclin B1 proteins. C, cycling T98G cells were exposed to 5 mmol/L SU6656 BubR1 or 5 mmol/L geraldol for 4 hours without or with 20 mmol/L MG- 132, lysed, and immunoblotted for Cdc20 the indicated proteins.

although we did not observe this effect in T98G cells in MCF-7 cells could be due to lack of caspase-3 activity. (Supplementary Fig. S2). To determine whether mitotic Indeed, SU6656 and geraldol did not induce mitotic slippage induced by SU6656 and geraldol involves cas- slippage in MCF-7 cells stably transfected with an empty pase-3, paclitaxel-arrested T98G cells harvested via shake- expression vector (MCF-7pcDNA; Fig. 2C). However, off were exposed to 5 mmol/L SU6656 or 5 mmol/L stable transfection of CASP-3 cDNA into MCF-7 cells geraldol concurrently with cell-permeable DEVD-CHO, (MCF-7casp3), which results in expression of procas- an inhibitor of caspases-3 and -7, for 4 hours. DEVD- pase-3 (37), was sufficient to enable cells to undergo CHO prevented induction of mitotic slippage at 50 to robust mitotic slippage in the presence of SU6656 or 100 mmol/L (Fig. 2A), indicating a requirement for caspase geraldol (Fig. 2C). Therefore, caspase-3 is required for activity in mitotic slippage induction by SU6656 and induction of mitotic slippage by these chemicals. geraldol. Immunoblotting of paclitaxel-arrested T98G We next asked whether there were differences in the cells exposed to SU6656 or geraldol together with 50 degradation of cyclin B1, BubR1, and Cdc20 in MCF- mmol/L DEVD-CHO revealed that degradation of BubR1 7pcDNA and MCF-7casp3 cells during exposure to and Cdc20 is caspase dependent (Fig. 2B), in addition to SU6656 and geraldol. Paclitaxel-arrested MCF-7pcDNA being proteasome dependent (Fig. 1B). In contrast, cyclin and MCF-7casp3 cells were harvested via shake-off, B1 degradation during induction of mitotic slippage is not exposed to 5 mmol/L SU6656 or 5 mmol/L geraldol for dependent on caspase-3 or caspase-7 (Fig. 2B), as cotreat- 4 hours, and analyzed by immunoblotting (Fig. 2D). ment with DEVD-CHO did not prevent cyclin B1 degra- Cyclin B1 was completely degraded in both cell lines dation in response to SU6656 or geraldol. on exposure to either SU6656 or geraldol. Because MCF- We previously observed that MCF-7 cells do not 7pcDNA cells do not undergo mitotic slippage under undergo mitotic slippage, spontaneous (26) or induced these conditions whereas MCF-7casp3 cells do, this result by SU6656 or geraldol (not shown). MCF-7 cells do not implies that cyclin B1 degradation is not sufficient to express caspase-3 because of a deletion within exon 3 of induce mitotic slippage. Cdc20 was degraded in both the CASP-3 gene that results in the introduction of a MCF-7pcDNA and MCF-7casp3 cells. Interestingly, premature stop codon that completely abrogates transla- BubR1 was degraded only in MCF-7casp3 cells tion of the CASP-3 mRNA (37). This observation was (Fig. 2D), implying that BubR1 is degraded in a cas- used to determine whether caspase-3 is required for pase-3–dependent manner during mitotic slippage and mitotic slippage induction by SU6656 and geraldol; the that its degradation is required for mitotic slippage inability of these chemicals to stimulate mitotic slippage induction. The observation by Kim and colleagues that

842 Mol Cancer Ther; 10(5) May 2011 Molecular Cancer Therapeutics

Caspase-3–Dependent Chemical Induction of Mitotic Slippage

Figure 2. Mitotic checkpoint AB inactivation but not cyclin B1 degradation occurs through 100 –+++++ Paclitaxel – caspase-3 dependent cleavage ––++–– SU6656 of BubR1. A, T98G cells were 80 SU6656 arrested in mitosis by 30 nmol/L Geraldol ––––++ Geraldol paclitaxel, harvested by shake-off, 60 –––+–+ DEVD-CHO and seeded in 96-well plates. After exposure to 5 mmol/L SU6656 or 5 Cyclin B1 mmol/L geraldol simultaneously 40 with 0 to 100 mmol/L Ac-DEVD- % Cells slipped CHO for 4 hours, the slipped cells 20 BubR1 were stained with Hoechst 33342 and quantified by automated fluorescence microscopy. Error 0 Cdc20 bars represent 95% CIs. B, mitotic 04020 8060 100 120 T98G cells were harvested by Concentration DEVD-CHO (µmol/L) shake-off and incubated with 5 mmol/L SU6656 or 5 mmol/L CD geraldol without or with 50 mmol/L MCF-7pcDNA MCF-7casp3 MCF-7pcDNA SU6656 DEVD-CHO for 4 hours. Lysates 100 MCF-7pcDNA geraldol were immunoblotted for the MCF-7casp3 SU6656 + + + +++++++ Paclitaxel indicated proteins. C, MCF-7 cells MCF-7casp3 geraldol ++ ++ –––––– stably transfected with empty 80 SU6656 vector (MCF-7pcDNA) or –––++ ––– ++ Geraldol caspase-3 (MCF-7casp3) were arrested in mitosis by 50 nmol/L 60 –––––– ++++ MG-132 paclitaxel, harvested by shake-off, and seeded in 96-well plates. The 40 Cyclin B1 cells were exposed to 0 to

m % Cells slipped BubR1 15 mol/L SU6656 or geraldol for 20 4 hours, stained with Hoechst 33342, and quantified using an Cdc20 0 automated fluorescence imager. 0246810121416 Error bars represent 95% CIs. D, Concentration (µmol/L) mitotic MCF-7pcDNA and MCF- 7casp3 cells were harvested by E shake-off and incubated with 5 Interphase mmol/L SU6656 or 5 mmol/L 100 Mitotic 100 geraldol without or with 20 mmol/L Slipped MG-132 for 4 hours. Lysates were 80 80 immunoblotted for the indicated proteins. E, MCF-7pcDNA or 60 60 MCF-7casp3 cells were exposed to 100 nmol/L paclitaxel for up to

% Cells 40 % Cells 40 28 hours, and nuclei were fixed and stained with Hoechst 33342. The total number of cells was 20 20 quantified using automated fluorescence microscopy, and the 0 0 images were visually inspected to 0 5 10 15 20 25 30 0 5 10 15 20 25 30 determine the proportion of Time (h) Time (h) slipped and mitotic cells at each MCF-7pcDNA MCF-7casp3 time.

caspase-3 can directly cleave BubR1 during mitotic exit proteasome dependent, but the degradation of cyclin (17) suggests that caspase-3 may degrade BubR1 directly B1 was not (Fig. 2D). Taken together, these results indi- during chemically induced mitotic slippage. cate that SU6656 and geraldol stimulate the degradation Although mitotic slippage induction by SU6656 and of BubR1 by caspase-3, inactivating the mitotic check- geraldol in MCF-7casp3 cells is proteasome dependent point and resulting in mitotic slippage. (data not shown), BubR1 degradation in MCF-7casp3 Approximately 20% of MCF-7pcDNA cells underwent cells was not inhibited by MG-132 (Fig. 2D), further mitotic slippage in the absence of SU6656 and geraldol indicating that it is caspase-3 and not the proteasome (Fig. 2C), indicating that spontaneous mitotic slippage that is required for the degradation of BubR1 during does not require caspase-3. To extend this observation, induction of mitotic slippage. Cdc20 degradation was mitotic arrest and slippage in MCF-7pcDNA and

www.aacrjournals.org Mol Cancer Ther; 10(5) May 2011 843

Riffell et al.

A 15 1 106 DMSO 5 µmol/L geraldol 8 105

10 6 105

4 105

% cells slipped 5

TG3 fluorescence (A.U.) 2 105

0 0 0 50 100 150 200 0 50 100 150 200 Time (min) Time (min)

B SU6656 Geraldol

U M 15 30 60 120 180 15 30 60 120 180 Time (min)

Cyclin B1

BubR1

Cdc20

Figure 3. Timeline of mitotic checkpoint inactivation and slippage. A, T98G cells arrested in mitosis by 30 nmol/L paclitaxel were harvested by shake-off, seeded in 96-well plates, and incubated with DMSO or 5 mmol/L geraldol for 15 minutes to 3 hours. Slipped cells were quantified after staining with Hoechst (left) or with mouse TG3 antibody against mitotically phosphorylated nucleolin (right). The proportion of slipped cells was lower than usually observed because of the numerous washes during immunofluorescent staining that removed many attached, slipped cells. Error bars represent 95% CIs. B, cycling T98G cells (U) were arrested in mitosis by exposure to 30 nmol/L paclitaxel and harvested by shake-off. Mitotic cells (M) were incubated with 5 mmol/L SU6656 or 5 mmol/L geraldol for 15 minutes to 3 hours and lysates were immunoblotted for the indicated proteins.

MCF-7casp3 cells were examined during exposure to geraldol for up to 3 hours while cell attachment and TG3 paclitaxel. Cells were exposed to 100 nmol/L paclitaxel fluorescence were measured (Fig. 3A). TG3 recognizes for up to 28 hours and the proportion of interphase, nucleolin phosphorylated by Cdk1/cyclin B1 and is a mitotic, and slipped cells was determined (Fig. 2E). In marker for mitosis (38). A significant proportion of cells both cell lines, mitotic and slipped cells accumulated over began to attach after 2 hours of exposure to geraldol, and time as the proportion of interphase cells declined the proportion of attached cells continued to increase (Fig. 2E). After 24 hours, the proportion of slipped cells over time (Fig. 3A, left). TG3 fluorescence decreased became greater than that of mitotic cells. The kinetics of appreciably within 60 minutes of exposure to geraldol accumulation of mitotic cells and slipped cells were very and continued to decrease over time (Fig. 3A, right). TG3 similar in MCF-7pcDNA and MCF-7casp3 cells (Fig. 2E), fluorescence during exposure to SU6656 could not be confirming that caspase-3 is not required for mitotic measured because of autofluorescence of the compound. arrest or spontaneous slippage in response to paclitaxel. The timing of degradation of cyclin B1, BubR1, and Cdc20 Therefore, although caspase-3 is not required for sponta- during mitotic slippage was also examined. Paclitaxel- neous mitotic slippage in response to antimitotic agents, arrested cells were exposed to 5 mmol/L SU6656 or 5 it is absolutely required for mitotic slippage induction by mmol/L geraldol for 15 minutes to 3 hours. Cyclin B1 SU6656 and geraldol. disappeared completely within 30 minutes (Fig. 3B). BubR1 and Cdc20 were partially degraded within 15 Mitotic slippage correlates temporally with minutes of exposure and almost completely degraded degradation of BubR1 and Cdc20 after about 2 hours (Fig. 3B), around the time when cells To examine the timing of chemically induced exit from began to attach and lose nucleolin phosphorylation, con- mitosis, paclitaxel-arrested mitotic cells were harvested sistent with a requirement for degradation of BubR1 and by shake-off, seeded in 96-well plates, and exposed to Cdc20 for mitotic slippage.

844 Mol Cancer Ther; 10(5) May 2011 Molecular Cancer Therapeutics

Caspase-3–Dependent Chemical Induction of Mitotic Slippage

A SU6656 Geraldol

0.1 µmol/L 100 1 µmol/L 100 10 µmol/L 80 80

60 60

Figure 4. Inhibition of Aurora A 40 40 and Aurora B by SU6656 and % Activity % Activity geraldol. A, SU6656 and geraldol in vitro were assayed for inhibition 20 20 of Aurora A and Aurora B as described in Materials and Methods. B, MCF-7 cells were 0 0 arrested in mitosis by 50 nmol/L Aurora A Aurora B Aurora A Aurora B paclitaxel for 20 hours, harvested by shake-off, seeded in 96-well plates, and incubated with 0.1 to m B 20 mol/L ZM447439 for 4 hours. 100 Cells were fixed, stained with Hoechst 33342, and imaged by automated fluorescence 80 microscopy. Error bars represent 95% CIs. 60

40 % Cells slipped 20

0 051015 2520 Concentration ZM447439 (µmol/L)

Inhibition of the Aurora kinases by SU6656 and inhibited more potently. The intracellular effects of geraldol SU6656 and geraldol were compared with those of The Aurora kinases play complex roles in metaphase ZM447439, a well-characterized Aurora B inhibitor that arrest and anaphase initiation, including chromosome induces mitotic slippage (33). ZM447439 induces 50% to congression and interkinetochore tension sensing (39, 60% of mitotic MCF-7 cells to undergo mitotic slippage 40). Inhibition of Aurora A or Aurora B in mitotic cells (Fig. 4B), whereas SU6656 and geraldol require the intro- results in mitotic slippage (33, 34). SU6656 was designed duction of caspase-3 to induce mitotic slippage in MCF-7 as a Src family kinase inhibitor (41) but has since been cells (Fig. 2C). Thus, mitotic slippage induction through reported to inhibit Aurora B in vitro (42, 43). Geraldol has inhibition of Aurora B does not seem to require caspase-3 no known biological activity, but fisetin, a closely struc- activation. Therefore, although SU6656 and geraldol may turally related flavonoid that induces mitotic slippage stimulate mitotic slippage in part by inhibition of Aurora less potently than geraldol (Supplementary Fig. S3), has B, these compounds probably have additional activities. also been reported to inhibit Aurora B (30). Geraldol was assayed for in vitro inhibition of a panel of kinases Cell survival after mitotic slippage is not affected by including Src, Aurora A, and Aurora B (Supplementary caspase-3 Table S1) and showed significant inhibition of Aurora A Given the role that caspase-3 plays in apoptosis and in and Aurora B but not Src. SU6656 and geraldol were then mitosis (44, 45), induction of mitotic slippage may affect assayed at 0.1 to 10 mmol/L for inhibition of Aurora A the survival of cells lacking and expressing caspase-3 and Aurora B kinase activity (Fig. 4A). Both compounds differently. MCF-pcDNA and MCF-7casp3 cells inhibited Aurora A and Aurora B, although Aurora B was arrested at mitosis with paclitaxel were harvested via

www.aacrjournals.org Mol Cancer Ther; 10(5) May 2011 845

Riffell et al.

MCF-7pcDNA MCF-7casp3 A 120 120 Figure 5. Dependence of the 100 100 DMSO outcome of spontaneous and SU6656 induced mitotic slippage on 80 80 Geraldol caspase-3. A, slipped cells, MCF- Slipped 60 60 7pcDNA and MCF-7casp3 cells cells were exposed to 50 nmol/L 40 40 paclitaxel for 20 hours, harvested % Cells remaining % Cells remaining 20 20 by shake-off, and seeded in 96- well plates. After exposure to 0 0 051015 051015 0.1% DMSO, 5 mmol/L SU6656, or Time (d) Time (d) 5 mmol/L geraldol for 4 hours, B 1,200 1,200 unattached (mitotic) cells were removed and adherent (slipped) 1,000 1,000 DMSO cells were allowed to grow in fresh Paclitaxel + DMSO 800 800 Paclitaxel + SU6656 culture medium for up to 14 days Paclitaxel + geraldol All before staining with Hoechst and 600 600 cells quantification as a proportion of 400 400 mitotic cells by automated fluorescence microscopy. B and % Cells remaining % Cells remaining 200 200 C, all cells and viable cells, MCF- 7pcDNA or MCF-7casp3 cells in 0 0 02468 1210 02468 1210 96-well plates were exposed to Time (d) Time (d) 0.1% DMSO or 50 nmol/L C 1,000 1,000 paclitaxel for 20 hours and then DMSO 0.1% DMSO, 5 mmol/L SU6656, or 800 800 Paclitaxel + DMSO 5 mmol/L geraldol for a further 4 Paclitaxel + SU6656 Paclitaxel + geraldol hours. Drugs were washed away 600 600 Viable and the cells were allowed to grow cells in fresh culture medium for up to 400 400 14 day before staining with

% Cells remaining 200 % Cells remaining 200 Hoechst and quantification (all cells) or analysis of cell viability by 0 0 the MTT assay (viable cells). Error 02468 1210 02468 1210 bars represent 95% CIs. Time (d) Time (d)

shake-off and exposed to dimethyl sulfoxide (DMSO), We previously reported that, after undergoing mitotic SU6656, or geraldol for 4 hours. The unattached mitotic slippage, cells remained metabolically active for up to cells were removed and the attached, slipped cells were several days and underwent 1 or more rounds of DNA cultured in the absence of any drugs for up to 14 days replication without cell division before undergoing apop- while cell numbers were determined by automated tosis (26). MCF-7pcDNA and MCF-7casp3 cells were fluorescence microscopy (Fig. 5A). Extensive cell death treated as before and metabolic activity was examined occurred in both cell lines such that 14 days after mitotic using the MTT assay. The metabolic activity of treated slippage, less than 20% of the initial number of mitotic cells in both cell lines increased during the first 3 days to cells remained (Fig. 5A). Therefore, caspase-3 expres- roughly the same extent as untreated cells (Fig. 5C), sion does not seem to play a major role in cell survival although untreated cells proliferated rapidly during that after mitotic slippage. time and there was a minimal increase in the number of The fate of the entire cell population after exposure to treated cells (Fig. 5B). Metabolic activity reached a pla- paclitaxel and SU6656 or geraldol was also examined. teau after 3 days and decreased considerably after 7 days, MCF-7pcDNA and MCF-7casp3 cells were exposed to a response not altered by caspase-3 expression. This 50 nmol/L paclitaxel for 20 hours and 0.1% DMSO, result indicates that, for 3 days after treatment with 5 mmol/L SU6656, or 5 mmol/L geraldol was added for paclitaxel without or with SU6656 or geraldol, little or a further 4 hours before both drugs were washed away. no cell proliferation or death took place, but the cells The cells were then allowed to grow in fresh cell culture continued to grow in size. Cell growth was arrested medium for up to 10 days before staining and quantifica- between 3 and 7 days before extensive cell death took tion. Initially, a small increase in cell number was place after day 7. observed, indicating that some cells recovered and were able to divide (Fig. 5B). However, extensive cell death Discussion began to occur 5 days following drug treatment and the majority of cells died before day 10 (Fig. 5B). Caspase-3 This study aimed to better understand pathways lead- expression did not alter this response. ing to mitotic slippage through the use of chemicals.

846 Mol Cancer Ther; 10(5) May 2011 Molecular Cancer Therapeutics

Caspase-3–Dependent Chemical Induction of Mitotic Slippage

SU6656 and geraldol, 2 compounds found to stimulate Choi and colleagues observed that BubR1 deacetylation mitotic slippage in cells exposed to a microtubule-target- at metaphase results in abrogation of its anaphase inhibi- ing agent (26), induce the proteasome-dependent degra- tion effects and in its degradation by APC/CCdc20 (15). dation of cyclin B1 as occurs during exit from mitosis Cotreatment of mitotic cells with SU6656 or geraldol and (Fig. 1B; ref. 46). However, these chemicals inactivate the the deacetylase inhibitor trichostatin A did not prevent mitotic checkpoint through the proteasome-dependent chemical induction of mitotic slippage (data not shown), degradation of BubR1 (Fig. 1B) that is sufficient to com- indicating that SU6656 and geraldol do not induce pre- promise the mitotic checkpoint (35, 36, 47). This effect mature deacetylation of BubR1. Although a small protea- occurs only in mitotic cells (Fig. 1C), and BubR1 is not some-dependent decrease in BubR1 was observed in degraded during completion of mitosis in T98G cells MCF-7pcDNA cells in response to SU6656 and geraldol (Supplementary Fig. S2). These results suggest that, (Fig. 2D), this is not sufficient for extensive mitotic slip- rather than accelerating spontaneous mitotic slippage, page to occur and is probably due to some deacetylation SU6656 and geraldol activate an alternate pathway lead- and proteasome-dependent degradation of BubR1. Kim ing to mitotic slippage through BubR1 degradation. and colleagues observed cleavage of BubR1 by caspase-3 Examination of the timing of mitotic checkpoint inac- during mitosis, which also led to exit from mitosis (17). tivation and slippage revealed that mitotic slippage, BubR1 is degraded in a caspase-3- but not proteasome- defined in this experiment by cell attachment and loss dependent manner in MCF-7 cells (Fig. 2D), indicating of the mitotic phosphoepitope recognized by the TG3 that caspase-3 initiates mitotic slippage through cleavage antibody, begins to occur 2 hours following exposure to of BubR1. geraldol whereas cyclin B1 degradation is complete 30 However, caspase-3 does not seem to be required for minutes after drug treatment (Fig. 3A and B). BubR1 and cell death following paclitaxel treatment (Fig. 5). Cdc20 degradation occurs more slowly (Fig. 3B) and Although caspase-3 is required for DNA fragmentation approximately correlate with mitotic slippage, indicating during apoptosis (37), cells that lack caspase-3 can never- a possible requirement for checkpoint inactivation prior theless undergo apoptosis. Other cell death pathways to mitotic slippage, even in the absence of Cdk1/cyclin B1 may also be involved in the fate of cells following expo- activity. In agreement with this observation, cyclin B1 is sure to an antimitotic agent: for instance, necrosis and also completely degraded in MCF-7 cells in response to autophagy have both been implicated in cell death fol- SU6656 and geraldol, although mitotic slippage cannot be lowing antimitotic therapy (49, 50). induced (Fig. 2B). It is not known whether the degrada- Spontaneous mitotic slippage has been described tion of other APC/C substrates or dephosphorylation of to occur through slow ubiquitylation of cyclin B1 by mitotic checkpoint kinase substrates might be involved in APC/CCdc20 and subsequent proteasome-dependent mitotic slippage. degradation despite mitotic checkpoint activity (24, 25). Caspase-3 is upregulated during mitosis (18) and has We propose a model (Fig. 6) for induced mitotic slippage, been shown to increase the formation of micronuclei in wherein decreased Cdk1 activity due to slow cyclin B1 response to antimitotic agents (48), but a role for caspase- 3 in mitosis remains undefined. This study reveals a novel role for caspase-3: mitotic slippage induced by SU6656 and geraldol is caspase-3 dependent. Cotreat- Cyclin B1 ment of mitotic cells with SU6656 or geraldol and the Induced Spontaneous caspase-3 and -7 inhibitor DEVD-CHO prevented mitotic mitotic slippage Cdk1 mitotic slippage slippage and checkpoint inactivation (Fig. 2A and B). P Although introduction of caspase-3 into MCF-7 cells does CTaspase-9 caspase-9 Cdk1 Inactive not affect the frequency of spontaneous mitotic slippage Phosphatase in response to paclitaxel (Fig. 2E), mitotic slippage can be inactive active induced by SU6656 and geraldol in MCF-7 cells only Cyclin B1 when exogenous caspase-3 is expressed (Fig. 2C), indi- Ub cating that caspase-3 is absolutely required for mitotic procaspase-3 Caspase-3 Ub Ub slippage induction by SU6656 and geraldol. This requirement for caspase-3 is likely due to its role in inactivation of the mitotic checkpoint; mitotic slippage Bub3 APC/C and BubR1 degradation occur in MCF-7 cells with active Mad2 Cdc20 caspase-3 but not in MCF-7 cells lacking caspase-3 BubR1 (Fig. 2C and D). BubR1 degradation is sufficient to inac- APC/C BubR1 tivate the mitotic checkpoint (14, 35, 36), and furthermore, Cdc20 depletion of BubR1 by mutation (47), gene knockdown, or, recently, in response to chemicals (36) has been implicated in the development of polyploidy. Several factors can influence the degradation of BubR1 during mitosis. Figure 6. Model for spontaneous and induced mitotic slippage.

www.aacrjournals.org Mol Cancer Ther; 10(5) May 2011 847

Riffell et al.

depletion, combined with phosphatase activity, results in outcome of antimitotic cancer therapy both through loss of the mitosis-specific Cdk1/cyclin B1 inhibitory mitotic slippage and through apoptosis. phosphorylation on caspase-9 (45). Active caspase-9 may then cleave procaspase-3, and activated caspase-3 Disclosure of Potential Conflict of Interest can cleave BubR1, resulting in inactivation of the mitotic No potential conflicts of interest were disclosed. checkpoint and activation of APC/CCdc20. This event would lead to further ubiquitylation and degradation Acknowledgments of cyclin B1, combining to force exit from mitosis through mitotic slippage. We thank Connie Kim for her help in characterizing analogues of In summary, these results show that the chemical geraldol and Peter Davies for the TG3 antibody. inducers of mitotic slippage SU6656 and geraldol cause proteasome- and caspase-dependent inactivation of the Grant Support mitotic checkpoint, in contrast to the accepted model for This work was supported by Canadian Breast Cancer Foundation grant spontaneous mitotic slippage. Caspase-3 is required for (M. Roberge), Michael Smith Foundation for Health Research Junior Graduate mitotic slippage induction and checkpoint inactivation Scholarship (J.L. Riffell), and Deutsche Forschungsgemeinschaft (SFB 728/B1) through degradation of BubR1, although not for cell grant (R.U. J€anicke). The costs of publication of this article were defrayed in part by the death in response to antimitotic agents. We propose a payment of page charges. This article must therefore be hereby marked model for induced mitotic slippage that includes an advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate important role for caspases in modulation of mitotic this fact. arrest. In response to the cellular stress presented by a Received September 30, 2010; revised February 1, 2011; accepted March prolonged mitotic arrest, caspases may contribute to the 11, 2011; published OnlineFirst March 25, 2011.

References 1. O’Connell CB, Khodjakov AL. Cooperative mechanisms of mitotic 15. Choi E, Choe H, Min J, Choi JY, Kim J, Lee H. BubR1 acetylation at spindle formation. J Cell Sci 2007;120:1717–22. prometaphase is required for modulating APC/C activity and timing of 2. Musacchio A, Salmon ED. The spindle-assembly checkpoint in space mitosis. EMBO J 2009;28:2077–89. and time. Nat Rev Mol Cell Biol 2007;8:379–93. 16. Vanoosthuyse V, Meadows JC, Van Der Sar SJ, Millar JB, Hardwick 3. Ganem NJ, Storchova Z, Pellman D. Tetraploidy, aneuploidy and KG. Bub3p facilitates spindle checkpoint silencing in fission yeast. cancer. Curr Opin Genet Dev 2007;17:157–62. Mol Biol Cell 2009;20:5096–105. 4. Schiff PB, Fant J, Horwitz SB. Promotion of microtubule assembly in 17. Kim M, Murphy K, Liu F, Parker SE, Dowling ML, Baff W, et al. vitro by taxol. Nature 1979;277:665–7. Caspase-mediated specific cleavage of BubR1 is a determinant of 5. Jordan MA, Thrower D, Wilson L. Mechanism of inhibition of cell mitotic progression. Mol Cell Biol 2005;25:9232–48. proliferation by Vinca alkaloids. Cancer Res 1991;51:2212–22. 18. Hashimoto T, Yamauchi L, Hunter T, Kikkawa U, Kamada S. Possible 6. Fang G, Yu H, Kirschner MW. Direct binding of CDC20 protein family involvement of caspase-7 in cell cycle progression at mitosis. Genes members activates the anaphase-promoting complex in mitosis and Cells 2008;13:609–21. G1. Mol Cell 1998;2:163–71. 19. Peterson D, Lee J, Lei XC, Forrest WF, Davis DP, Jackson PK, et al. A 7. King RW, Peters JM, Tugendreich S, Rolfe M, Hieter P, Kirschner MW. chemosensitization screen identifies TP53RK, a kinase that restrains A 20S complex containing CDC27 and CDC16 catalyzes the mitosis- apoptosis after mitotic stress. Cancer Res 2010;70:6325–35. specific conjugation of ubiquitin to cyclin B. Cell 1995;81:279–88. 20. Andreassen PR, Martineau SN, Margolis RL. Chemical induction of 8. Visintin R, Prinz S, Amon A. CDC20 and CDH1: a family of substrate- mitotic checkpoint override in mammalian cells results in aneuploidy specific activators of APC-dependent proteolysis. Science 1997; following a transient tetraploid state. Mutat Res 1996;372:181– 278:460–3. 94. 9. Cohen-Fix O, Peters JM, Kirschner MW, Koshland D. Anaphase 21. Jin L, Williamson A, Banerjee S, Philipp I, Rape M. Mechanism of initiation in Saccharomyces cerevisiae is controlled by the APC- ubiquitin-chain formation by the human anaphase-promoting com- dependent degradation of the anaphase inhibitor Pds1p. Genes plex. Cell 2008;133:653–65. Dev 1996;10:3081–93. 22. Blagosklonny MV. Mitotic arrest and cell fate: why and how mitotic 10. Sudakin V, Chan GK, Yen TJ. Checkpoint inhibition of the APC/C in inhibition of transcription drives mutually exclusive events. Cell Cycle HeLa cells is mediated by a complex of BUBR1, BUB3, CDC20, and 2007;6:70–4. MAD2. J Cell Biol 2001;154:925–36. 23. Elhajouji A, Cunha M, Kirsch-Volders M. Spindle poisons can induce 11. Kallio M, Weinstein J, Daum JR, Burke DJ, Gorbsky GJ. Mammalian polyploidy by mitotic slippage and micronucleate mononucleates in p55CDC mediates association of the spindle checkpoint protein the cytokinesis-block assay. Mutagenesis 1998;13:193–8. Mad2 with the cyclosome/anaphase-promoting complex, and is 24. Brito DA, Rieder CL. Mitotic checkpoint slippage in humans occurs via involved in regulating anaphase onset and late mitotic events. J Cell cyclin B destruction in the presence of an active checkpoint. Curr Biol Biol 1998;141:1393–406. 2006;16:1194–200. 12. Nilsson J, Yekezare M, Minshull J, Pines J. The APC/C maintains the 25. Lee J, Kim JA, Margolis RL, Fotedar R. Substrate degradation by the spindle assembly checkpoint by targeting Cdc20 for destruction. Nat anaphase promoting complex occurs during mitotic slippage. Cell Cell Biol 2008;10:1411–20. Cycle 2010;9:1792–801. 13. Wu H, Lan Z, Li W, Wu S, Weinstein J, Sakamoto KM, et al. p55CDC/ 26. Riffell JL, Zimmerman C, Khong A, McHardy LM, Roberge M. Effects hCDC20 is associated with BUBR1 and may be a downstream target of chemical manipulation of mitotic arrest and slippage on cancer cell of the spindle checkpoint kinase. Oncogene 2000;19:4557–62. survival and proliferation. Cell Cycle 2009;8:3025–38. 14. Chan GK, Jablonski SA, Sudakin V, Hittle JC, Yen TJ. Human BUBR1 27. DeMoe JH, Santaguida S, Daum JR, Musacchio A, Gorbsky GJ. A is a mitotic checkpoint kinase that monitors CENP-E functions at high throughput, whole cell screen for small molecule inhibitors of the kinetochores and binds the cyclosome/APC. J Cell Biol 1999;146: mitotic spindle checkpoint identifies OM137, a novel Aurora kinase 941–54. inhibitor. Cancer Res 2009;69:1509–16.

848 Mol Cancer Ther; 10(5) May 2011 Molecular Cancer Therapeutics

Caspase-3–Dependent Chemical Induction of Mitotic Slippage

28. Chan YW, Ma HT, Wong W, Ho CC, On KF, Poon RY. CDK1 inhibitors 39. Kim Y, Holland AJ, Lan W, Cleveland DW. Aurora kinases and protein antagonize the immediate apoptosis triggered by spindle disruption phosphatase 1 mediate chromosome congression through regulation but promote apoptosis following the subsequent rereplication and of CENP-E. Cell 2010;142:444–55. abnormal mitosis. Cell Cycle 2008;7:1449–61. 40. Liu D, Vader G, Vromans MJ, Lampson MA, Lens SM. Sensing 29. Stolz A, Vogel C, Schneider V, Ertych N, Kienitz A, Yu H, et al. chromosome bi-orientation by spatial separation of aurora B kinase Pharmacologic abrogation of the mitotic spindle checkpoint by an from kinetochore substrates. Science 2009;323:1350–3. indolocarbazole discovered by cellular screening efficiently kills can- 41. Blake RA, Broome MA, Liu X, Wu J, Gishizky M, Sun L, et al. SU6656, a cer cells. Cancer Res 2009;69:3874–83. selective src family kinase inhibitor, used to probe growth factor 30. Salmela AL, Pouwels J, Varis A, Kukkonen AM, Toivonen P, Halonen signaling. Mol Cell Biol 2000;20:9018–27. PK, et al. Dietary flavonoid fisetin induces a forced exit from mitosis by 42. Bain J, McLauchlan H, Elliott M, Cohen P. The specificities of protein targeting the mitotic spindle checkpoint. Carcinogenesis 2009;30: kinase inhibitors: an update. Biochem J 2003;371:199–204. 1032–40. 43. Bain J, Plater L, Elliott M, Shpiro N, Hastie CJ, McLauchlan H, et al. 31. Stevens FE, Beamish H, Warrener R, Gabrielli B. Histone deacetylase The selectivity of protein kinase inhibitors: a further update. Biochem J inhibitors induce mitotic slippage. Oncogene 2008;27:1345–54. 2007;408:297–315. 32. Noh EJ, Lim DS, Jeong G, Lee JS. An HDAC inhibitor, trichostatin A, 44. Tao W, South VJ, Zhang Y, Davide JP, Farrell L, Kohl NE, et al. induces a delay at G2/M transition, slippage of spindle checkpoint, Induction of apoptosis by an inhibitor of the mitotic kinesin KSP and cell death in a transcription-dependent manner. Biochem Biophys requires both activation of the spindle assembly checkpoint and Res Commun 2009;378:326–31. mitotic slippage. Cancer Cell 2005;8:49–59. 33. Ditchfield C, Johnson VL, Tighe A, Ellston R, Haworth C, Johnson T, 45. Allan LA, Clarke PR. Phosphorylation of caspase-9 by CDK1/cyclin B1 et al. Aurora B couples chromosome alignment with anaphase by protects mitotic cells against apoptosis. Mol Cell 2007;26:301–10. targeting BubR1, Mad2, and Cenp-E to kinetochores. J Cell Biol 46. Gascoigne KE, Taylor SS. Cancer cells display profound intra- and 2003;161:267–80. interline variation following prolonged exposure to antimitotic drugs. 34. Wysong DR, Chakravarty A, Hoar K, Ecsedy JA. The inhibition of Cancer Cell 2008;14:111–22. Aurora A abrogates the mitotic delay induced by microtubule perturb- 47. Shin HJ, Baek KH, Jeon AH, Park MT, Lee SJ, Kang CM, et al. Dual ing agents. Cell Cycle 2009;8:876–88. roles of human BubR1, a mitotic checkpoint kinase, in the monitoring 35. Dai W, Wang Q, Liu T, Swamy M, Fang Y, Xie S, et al. Slippage of of chromosomal instability. Cancer Cell 2003;4:483–97. mitotic arrest and enhanced tumor development in mice with BubR1 48. Decordier I, Cundari E, Kirsch-Volders M. Survival of aneuploid, haploinsufficiency. Cancer Res 2004;64:440–5. micronucleated and/or polyploid cells: crosstalk between ploidy con- 36. Tovar C, Higgins B, Deo D, Kolinsky K, Liu JJ, Heimbrook DC, et al. trol and apoptosis. Mutat Res 2008;651:30–9. Small-molecule inducer of cancer cell polyploidy promotes apoptosis 49. Yeung TK, Germond C, Chen X, Wang Z. The mode of action of taxol: or senescence: implications for therapy. Cell Cycle 2010;9:3364– apoptosis at low concentration and necrosis at high concentration. 75. Biochem Biophys Res Commun 1999;263:398–404. 37. Janicke RU, Sprengart ML, Wati MR, Porter AG. Caspase-3 is required 50. Gorka M, Daniewski WM, Gajkowska B, Lusakowska E, Godlewski for DNA fragmentation and morphological changes associated with MM, Motyl T. Autophagy is the dominant type of programmed cell apoptosis. J Biol Chem 1998;273:9357–60. death in breast cancer MCF-7 cells exposed to AGS 115 and EFDAC, 38. Anderson HJ, de Jong G, Vincent I, Roberge M. Flow cytometry of new sesquiterpene analogs of paclitaxel. Anticancer Drugs 2005; mitotic cells. Exp Cell Res 1998;238:498–502. 16:777–88.

www.aacrjournals.org Mol Cancer Ther; 10(5) May 2011 849

Caspase-3−Dependent Mitotic Checkpoint Inactivation by the Small-Molecule Inducers of Mitotic Slippage SU6656 and Geraldol

Jenna L. Riffell, Reiner U. Jänicke and Michel Roberge

Mol Cancer Ther 2011;10:839-849. Published OnlineFirst March 25, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-10-0909

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2011/03/24/1535-7163.MCT-10-0909.DC1

Cited articles This article cites 50 articles, 18 of which you can access for free at: http://mct.aacrjournals.org/content/10/5/839.full#ref-list-1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/10/5/839. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.