Killing cells by targeting mitosis

Cell cycle deregulation is a common feature of human cancer. Tumor cells accumulate mutations that result in unscheduled proliferation, genomic instability and chromosomal instability. Several therapeutic strategies have been proposed for targeting the cell division cycle in cancer. Whereas inhibiting the initial phases of the cell cycle is likely to generate viable quiescent cells, targeting mitosis offers several possibilities for killing cancer cells. Microtubule poisons have proved efficacy in the clinic against a broad range of malignancies, and novel targeted strategies are now evaluating the inhibition of critical activities, such as cyclin-dependent kinase 1, Aurora or Polo kinases or spindle kinesins. Abrogation of the mitotic checkpoint or targeting the energetic or proteotoxic stress of aneuploid or chromosomally instable cells may also provide further benefits by inducing lethal levels of instability. Although cancer cells may display different responses to these treatments, recent data suggest that targeting mitotic exit by inhibiting the anaphase-promoting complex generates metaphase cells that invariably die in mitosis. As the efficacy of cell-cycle targeting approaches has been limited so far, further understanding of the molecular pathways modulating mitotic cell death will be required to move forward these new proposals to the clinic.     

Plk1 Inhibition Causes Post-Mitotic DNA Damage and Senescence in a Range of Human Tumor Cell Lines

Plk1 is a checkpoint protein whose role spans all of mitosis and includes DNA repair, and is highly conserved in eukaryotes from yeast to man. Consistent with this wide array of functions for Plk1, the cellular consequences of Plk1 disruption are diverse, spanning delays in mitotic entry, mitotic spindle abnormalities, and transient mitotic arrest leading to mitotic slippage and failures in cytokinesis. In this work, we present the in vitro and in vivo consequences of Plk1 inhibition in cancer cells using potent, selective small-molecule Plk1 inhibitors and Plk1 genetic knock-down approaches. We demonstrate for the first time that cellular senescence is the predominant outcome of Plk1 inhibition in some cancer cell lines, whereas in other cancer cell lines the dominant outcome appears to be apoptosis, as has been reported in the literature. We also demonstrate strong induction of DNA double-strand breaks in all six lines examined (as assayed by γH2AX), which occurs either during mitotic arrest or mitotic-exit, and may be linked to the downstream induction of senescence. Taken together, our findings expand the view of Plk1 inhibition, demonstrating the occurrence of a non-apoptotic outcome in some settings. Our findings are also consistent with the possibility that mitotic arrest observed as a result of Plk1 inhibition is at least partially due to the presence of unrepaired double-strand breaks in mitosis. These novel findings may lead to alternative strategies for the development of novel therapeutic agents targeting Plk1, in the selection of biomarkers, patient populations, combination partners and dosing regimens.     

Cell death in leukemia: passenger protein regulation by topoisomerase inhibitors

Etoposide is a potent inducer of mitotic catastrophe; a type of cell death resulting from aberrant mitosis. It is important in p53 negative cells where p53 dependent apoptosis and events at the G1 and G2 cell cycle checkpoints are compromised. Passenger proteins regulate many aspects of mitosis and siRNA interference or direct inhibition of Aurora B kinase results in mitotic catastrophe. However, there is little available data of clinical relevance in leukaemia models. Here, in p53 negative K562 myeloid leukemia cells, etoposide-induced mitotic catastrophe is shown to be time and/or concentration dependent. Survivin and Aurora remained bound to chromosomes. Survivin and Aurora were also associated with Cdk1 and were shown to form complexes, which in pull down experiments, included INCENP. There was no evidence of Aurora B kinase suppression. These data suggests etoposide will complement Aurora B kinase inhibitors currently in clinical trials for cancer.     

Time-Lapse Imaging of Neuroblastoma Cells to Determine Cell Fate upon Gene Knockdown

Neuroblastoma is the most common extra-cranial solid tumor of early childhood. Standard therapies are not effective in case of poor prognosis and chemotherapy resistance. To improve drug therapy, it is imperative to discover new targets that play a substantial role in tumorigenesis of neuroblastoma. The mitotic machinery is an attractive target for therapeutic interventions and inhibitors can be developed to target mitotic entry, spindle apparatus, spindle activation checkpoint, and mitotic exit. We present an elaborate analysis pipeline to determine cancer specific therapeutic targets by first performing a focused gene expression analysis to select genes followed by a gene knockdown screening assay of live cells. We interrogated gene expression studies of neuroblastoma tumors and selected 240 genes relevant for tumorigenesis and cell cycle. With these genes we performed time-lapse screening of gene knockdowns in neuroblastoma cells. We classified cellular phenotypes and used the temporal context of the perturbation effect to determine the sequence of events, particularly the mitotic entry preceding cell death. Based upon this phenotype kinetics from the gene knockdown screening, we inferred dynamic gene functions in mitosis and cell proliferation. We identified six genes (DLGAP5, DSCC1, SMO, SNRPD1, SSBP1, and UBE2C) with a vital role in mitosis and these are promising therapeutic targets for neuroblastoma. Images and movies of every time point of all screened genes are available at https://ichip.bioquant.uni-heidelberg.de.     

Mitosis as an anti-cancer drug target

Suppression of cell proliferation by targeting mitosis is one potential cancer intervention. A number of existing chemotherapy drugs disrupt mitosis by targeting microtubule dynamics. While efficacious, these drugs have limitations, i.e. neuropathy, unpredictability and development of resistance. In order to overcome these issues, a great deal of effort has been spent exploring novel mitotic targets including Polo-like kinase 1, Aurora kinases, Mps1, Cenp-E and KSP/Eg5. Here we summarize the latest developments in the discovery and clinical evaluation of new mitotic drug targets.     

Mitochondrial biogenesis in epithelial cancer cells promotes breast cancer tumor growth and confers autophagy resistance

Here, we set out to test the novel hypothesis that increased mitochondrial biogenesis in epithelial cancer cells would “fuel” enhanced tumor growth. For this purpose, we generated MDA-MB-231 cells (a triple-negative human breast cancer cell line) overexpressing PGC-1α and MitoNEET, which are established molecules that drive mitochondrial biogenesis and increased mitochondrial oxidative phosphorylation (OXPHOS). Interestingly, both PGC-1α and MitoNEET increased the abundance of OXPHOS protein complexes, conferred autophagy resistance under conditions of starvation and increased tumor growth by up to ~3-fold. However, this increase in tumor growth was independent of neo-angiogenesis, as assessed by immunostaining and quantitation of vessel density using CD31 antibodies. Quantitatively similar increases in tumor growth were also observed by overexpression of PGC-1β and POLRMT in MDA-MB-231 cells, which are also responsible for mediating increased mitochondrial biogenesis. Thus, we propose that increased mitochondrial “power” in epithelial cancer cells oncogenically promotes tumor growth by conferring autophagy resistance. As such, PGC-1α, PGC-1β, mitoNEET and POLRMT should all be considered as tumor promoters or “metabolic oncogenes.” Our results are consistent with numerous previous clinical studies showing that metformin (a weak mitochondrial “poison”) prevents the onset of nearly all types of human cancers in diabetic patients. Therefore, metformin (a complex I inhibitor) and other mitochondrial inhibitors should be developed as novel anticancer therapies, targeting mitochondrial metabolism in cancer cells.     

Dual inhibition of Kif15 by oxindole and quinazolinedione chemical probes

The mitotic spindle is a microtubule-based machine that segregates a replicated set of chromosomes during cell division. Many cancer drugs alter or disrupt the microtubules that form the mitotic spindle. Microtubule-dependent molecular motors that function during mitosis are logical alternative mitotic targets for drug development. Eg5 (Kinesin-5) and Kif15 (Kinesin-12), in particular, are an attractive pair of motor proteins, as they work in concert to drive centrosome separation and promote spindle bipolarity. Furthermore, we hypothesize that the clinical failure of Eg5 inhibitors may be (in part) due to compensation by Kif15. In order to test this idea, we screened a small library of kinase inhibitors and identified GW108X, an oxindole that inhibits Kif15 in vitro. We show that GW108X has a distinct mechanism of action compared with a commercially available Kif15 inhibitor, Kif15-IN-1 and may serve as a lead with which to further develop Kif15 inhibitors as clinically relevant agents.     

Forcing dividing cancer cells to die; low‐dose drug combinations to prevent spindle pole clustering

Mitosis, under the control of the microtubule-based mitotic spindle, is an attractive target for anti-cancer treatments, as cancer cells undergo frequent and uncontrolled cell divisions. Microtubule targeting agents that disrupt mitosis or single molecule inhibitors of mitotic kinases or microtubule motors kill cancer cells with a high efficacy. These treatments have, nevertheless, severe disadvantages: they also target frequently dividing healthy tissues, such as the haematopoietic system, and they often lose their efficacy due to primary or acquired resistance mechanisms. An alternative target that has emerged in dividing cancer cells is their ability to “cluster” the poles of the mitotic spindle into a bipolar configuration. This mechanism is necessary for the specific survival of cancer cells that tend to form multipolar spindles due to the frequent presence of abnormal centrosome numbers or other spindle defects. Here we discuss the recent development of combinatorial treatments targeting spindle pole clustering that specifically target cancer cells bearing aberrant centrosome numbers and that have the potential to avoid resistance mechanism due their combinatorial nature.

     

Cyclin-dependent kinase 1-mediated phosphorylation of protein kinase N1 promotes anchorage-independent growth and migration

Protein kinase N1 (PKN1) is a member of the protein kinase C superfamily. Aberrations of PKN1 kinase activity are involved in several human pathological processes, including cancer. We found that PKN family proteins (PKN1/2/3) are phosphorylated in response to antitubulin drug-induced mitotic arrest. We identified cyclin-dependent kinase 1 (CDK1) as the corresponding kinase for PKN protein phosphorylation. CDK1 phosphorylates PKN1 at S533, S537, S562, and S916 in vitro and in cells during drug-induced mitotic arrest. Immunofluorescence staining further confirmed that PKN1 phosphorylation occurs during normal mitosis in a CDK1-dependent manner. Knockdown of PKN1 significantly inhibited anchorage-independent growth and migration without affecting proliferation in multiple cancer cell lines. We further showed that mitotic phosphorylation is essential for PKN1's oncogenic function, as the non-phosphorylatable mutant PKN1-4A failed to rescue anchorage-independent growth and migration in PKN1-knockdown cells. Thus, our findings reveal a novel regulatory mechanism for PKN1 in mitosis and its role in tumorigenesis.     

Metformin Treatment Does Not Inhibit Growth of Pancreatic Cancer Patient-Derived Xenografts

There is currently tremendous interest in developing anti-cancer therapeutics targeting cell signaling pathways important for both cancer cell metabolism and growth. Several epidemiological studies have shown that diabetic patients taking metformin have a decreased incidence of pancreatic cancer. This has prompted efforts to evaluate metformin, a drug with negligible toxicity, as a therapeutic modality in pancreatic cancer. Preclinical studies in cell line xenografts and one study in patient-derived xenograft (PDX) models were promising, while recently published clinical trials showed no benefit to adding metformin to combination therapy regimens for locally advanced and metastatic pancreatic cancer. PDX models in which patient tumors are directly engrafted into immunocompromised mice have been shown to be excellent preclinical models for biomarker discovery and therapeutic development. We evaluated the response of four PDX tumor lines to metformin treatment and found that all four of our PDX lines were resistant to metformin. We found that the mechanisms of resistance may occur through lack of sustained activation of adenosine monophosphate-activated protein kinase (AMPK) or downstream reactivation of the mammalian target of rapamycin (mTOR). Moreover, combined treatment with metformin and mTOR inhibitors failed to improve responses in cell lines, which further indicates that metformin alone or in combination with mTOR inhibitors will be ineffective in patients, and that resistance to metformin may occur through multiple pathways. Further studies are required to better understand these mechanisms of resistance and inform potential combination therapies with metformin and existing or novel therapeutics.     

Mitochondrial biogenesis in epithelial cancer cells promotes breast cancer tumor growth and confers autophagy resistance

Here, we set out to test the novel hypothesis that increased mitochondrial biogenesis in epithelial cancer cells would “fuel” enhanced tumor growth. For this purpose, we generated MDA-MB-231 cells (a triple-negative human breast cancer cell line) overexpressing PGC-1α and MitoNEET, which are established molecules that drive mitochondrial biogenesis and increased mitochondrial oxidative phosphorylation (OXPHOS). Interestingly, both PGC-1α and MitoNEET increased the abundance of OXPHOS protein complexes, conferred autophagy resistance under conditions of starvation and increased tumor growth by up to ~3-fold. However, this increase in tumor growth was independent of neo-angiogenesis, as assessed by immunostaining and quantitation of vessel density using CD31 antibodies. Quantitatively similar increases in tumor growth were also observed by overexpression of PGC-1β and POLRMT in MDA-MB-231 cells, which are also responsible for mediating increased mitochondrial biogenesis. Thus, we propose that increased mitochondrial “power” in epithelial cancer cells oncogenically promotes tumor growth by conferring autophagy resistance. As such, PGC-1α, PGC-1β, mitoNEET and POLRMT should all be considered as tumor promoters or “metabolic oncogenes.” Our results are consistent with numerous previous clinical studies showing that metformin (a weak mitochondrial “poison”) prevents the onset of nearly all types of human cancers in diabetic patients. Therefore, metformin (a complex I inhibitor) and other mitochondrial inhibitors should be developed as novel anticancer therapies, targeting mitochondrial metabolism in cancer cells.     

Fate of primary cells at the G1/S boundary after polo-like kinase 1 inhibition by SBE13

Human polo-like kinase 1 (Plk1), a key regulator of mitosis, is over-expressed in various human tumors. It is a negative prognostic factor for cancer patients and a measure for the aggressiveness of a tumor. Thus, targeting Plk1 might be a promising approach for cancer therapy. Plk1 inhibitors represent attractive tools for cancer research and for the mechanistic investigation of checkpoint control. Here, we show the impact of Plk1 inhibition on cell cycle regulation in primary cells. After treatment with SBE13 the G1/S checkpoint was intact, indicated by reduced pRb, resulting in slower cell cycle progression but overall cell proliferation was not significantly impaired. Thus, we provide strong evidence that SBE13 leaves checkpoint control in primary cells unaffected making it a remarkable future anti-cancer therapeutic.     

Establishment of the active catalytic domain of human PDGFRbeta tyrosine kinase-based ELISA assay for inhibitor screening

Tyrosine kinases are emerging as frequent targets of primary oncogenic events and therefore represent an optimal focus of therapeutic intervention. In an effort towards therapeutic PDGFR inactivation, we expressed the catalytic domain of PDGFRbeta as a soluble active kinase using Bac-to-Bac expression system, and studied the correlations between PDGFRbeta activity and enzyme concentration, ATP concentration, substrate concentration and divalent cation type. And a convenient, effective and non-radioactive ELISA screening model is then established for identification of the potential inhibitors targeting PDGFRbeta kinase. Of 500 RTK target-based compounds, TKI-30 was identified as a small molecule potential inhibitor of PDGFRbeta (IC(50)=0.34 microM). Further studies indicated that TKI-30 blocked PDGF-BB-induced autophosphorylation of PDGFRbeta in a dose-dependent manner in Swiss 3T3 cells and human umbilical vein smooth muscle cells (HUVSMCs). Moreover, it dose-dependently suppressed the PDGF-BB-induced proliferation in HUVSMCs and tube formation of HUVEC. Our data collectively indicated that PDGFRbeta-based ELISA assay is a new method available for screening inhibitors targeting PDGFRbeta kinase and TKI-30 is a potential novel anti-cancer agent worthy of being further investigated.     

Non-mitotic functions of polo-like kinases in cancer cells

Inhibitors of mitotic protein kinases are currently being developed as non-neurotoxic alternatives of microtubule-targeting agents (taxanes, vinca alkaloids) which provide a substantial survival benefit for patients afflicted with different types of solid tumors. Among the mitotic kinases, the cyclin-dependent kinases, the Aurora kinases, the kinesin spindle protein and Polo-like kinases (PLKs) have emerged as attractive targets of cancer therapeutics.The functions of mammalian PLK1-5 are traditionally linked to the regulation of the cell cycle and to the stress response. Especially the key role of PLK1 and PLK4 in cellular growth and proliferation, their overexpression in multiple types of human cancer and their druggability, make them appealing targets for cancer therapy. Inhibitors for PLK1 and PLK4 are currently being tested in multiple cancer trials. The clinical success of microtubule-targeting agents is attributed not solely to the induction of a mitotic arrest in cancer cells, but also to non-mitotic effects like targeting intracellular trafficking on microtubules. This raises the question whether new cancer targets like PLK1 and PLK4 regulate critical non-mitotic functions in tumor cells. In this article we summarize the important roles of PLK1-5 for the regulation of non-mitotic signaling. Due to these functions it is conceivable that inhibitors for PLK1 or PLK4 can target interphase cells, which underscores their attractive potential as cancer drug targets. Moreover, we also describe the contribution of the tumor-suppressors PLK2, PLK3 and PLK5 to cancer cell signaling outside of mitosis. These observations highlight the urgent need to develop highly specific ATP-competitive inhibitors for PLK4 and for PLK1 like the 3^rd generation PLK-inhibitor Onvansertib to prevent the inhibition of tumor-suppressor PLKs in- and outside of mitosis. The remarkable feature of PLKs to encompass a unique druggable domain, the polo-box-domain (PBD) that can be found only in PLKs offers the opportunity for the development of inhibitors that target PLKs exclusively. Beyond the development of mono-specific ATP-competitive PLK inhibitors, the PBD as drug target will support the design of new drugs that eradicate cancer cells based on the mitotic and non-mitotic function of PLK1 and PLK4.     

Small Molecule R1498 as a Well-Tolerated and Orally Active Kinase Inhibitor for Hepatocellular Carcinoma and Gastric Cancer Treatment via Targeting Angiogenesis and Mitosis Pathways

Protein kinases play important roles in tumor development and progression. Lots of kinase inhibitors have entered into market and show promising clinical benefits. Here we report the discovery of a novel small molecule, well-tolerated, orally active kinase inhibitor, R1498, majorly targeting both angiogenic and mitotic pathways for the treatment of hepatocellular carcinoma (HCC) and gastric cancer (GC). A series of biochemical and cell-based assays indicated that the target kinase cluster of R1498 included Aurora kinases and VEGFR2 et al. R1498 showed moderate in vitro growth inhibition on a panel of tumor cells with IC50 of micromole range. The in vivo anti-tumor efficacy of R1498 was evaluated on a panel of GC and HCC xenografts in a parallel comparison with another multikinase inhibitor sorafenib. R1498 demonstrated superior efficacy and toxicity profile over sorafenib in all test models with >80% tumor growth inhibition and tumor regression in some xenogratfts. The therapeutic potential of R1498 was also highlighted by its efficacy on three human GC primary tumor derived xenograft models with 10–30% tumor regression rate. R1498 was shown to actively inhibit the Aurora A activity in vivo, and decrease the vascularization in tumors. Furthermore, R1498 presented good in vivo exposure and therapeutic window in the pharmacokinetic and dose range finding studies. Theses evidences indicate that R1498 is a potent, well-tolerated, orally active multitarget kinase inhibitor with a unique antiangiogenic and antiproliferative profile, and provide strong confidence for further development for HCC and GC therapy.     

A Role for Mitogen-activated Protein Kinase in the Spindle Assembly Checkpoint in XTC Cells

The spindle assembly checkpoint prevents cells whose spindles are defective or chromosomes are misaligned from initiating anaphase and leaving mitosis. Studies of Xenopus egg extracts have implicated the Erk2 mitogen-activated protein kinase (MAP kinase) in this checkpoint. Other studies have suggested that MAP kinases might be important for normal mitotic progression. Here we have investigated whether MAP kinase function is required for mitotic progression or the spindle assembly checkpoint in vivo in Xenopus tadpole cells (XTC). We determined that Erk1 and/or Erk2 are present in the mitotic spindle during prometaphase and metaphase, consistent with the idea that MAP kinase might regulate or monitor the status of the spindle. Next, we microinjected purified recombinant XCL100, a Xenopus MAP kinase phosphatase, into XTC cells in various stages of mitosis to interfere with MAP kinase activation. We found that mitotic progression was unaffected by the phosphatase. However, XCL100 rendered the cells unable to remain arrested in mitosis after treatment with nocodazole. Cells injected with phosphatase at prometaphase or metaphase exited mitosis in the presence of nocodazole—the chromosomes decondensed and the nuclear envelope re-formed—whereas cells injected with buffer or a catalytically inactive XCL100 mutant protein remained arrested in mitosis. Coinjection of constitutively active MAP kinase kinase-1, which opposes XCL100's effects on MAP kinase, antagonized the effects of XCL100. Since the only known targets of MAP kinase kinase-1 are Erk1 and Erk2, these findings argue that MAP kinase function is required for the spindle assembly checkpoint in XTC cells.     

Effect of serine/threonine protein kinases and protein phosphatases inhibitors on mitosis progression in a synchronized tobacco BY-2 culture

In order to investigate the role of various serine/threonine protein kinases and protein phosphatases in the regulation of mitosis progression in plant cells, the influence of cyclin(olomoucine) and Ca²⁺/calmodulin-dependent (W7) protein kinases inhibitors, as well as protein kinase C inhibitors (H7 and staurosporine), and a protein phosphatases inhibitor (okadaic acid) on mitosis progression in synchronized tobacco BY-2 cells has been studied. It was found that BY-2 culture treatment with inhibitors of cyclin-dependent protein kinases and protein kinase C caused a prophase delay, reduced the mitotic index, and displaced the mitotic peak as compared with control cells. Inhibition of Ca²⁺/calmodulin-dependent protein kinases enhanced the cells entry into prophase and delayed their exit from mitosis. Meanwhile inhibition of serine/threonine protein phosphatases insignificantly enhances synchronized BY-2 cells entering into all phases of mitosis.     

USP7 and Daxx regulate mitosis progression and taxane sensitivity by affecting stability of Aurora-A kinase

A large number of patients are resistant to taxane-based chemotherapy. Functional mitotic checkpoints are essential for taxane sensitivity. Thus, mitotic regulators are potential markers for therapy response and could be targeted for anticancer therapy. In this study, we identified a novel function of ubiquitin (Ub)-specific processing protease-7 (USP7) that interacts and cooperates with protein death domain-associated protein (Daxx) in the regulation of mitosis and taxane resistance. Depletion of USP7 impairs mitotic progression, stabilizes cyclin B and reduces stability of the mitotic E3 Ub ligase, checkpoint with forkhead and Ring-finger (CHFR). Consequently, cells with depleted USP7 accumulate Aurora-A kinase, a CHFR substrate, thus elevating multipolar mitoses. We further show that these effects are independent of the USP7 substrate p53. Thus, USP7 and Daxx are necessary to regulate proper execution of mitosis, partially via regulation of CHFR and Aurora-A kinase stability. Results from colony formation assay, in silico analysis across the NCI60 platform and in breast cancer patients suggest that USP7 levels inversely correlate with response to taxanes, pointing at the USP7 protein as a potential predictive factor for taxane response in cancer patients. In addition, we demonstrated that inhibition of Aurora-A attenuates USP7-mediated taxane resistance, suggesting that combinatorial drug regimens of Taxol and Aurora-A inhibitors may improve the outcome of chemotherapy response in cancer patients resistant to taxane treatment. Finally, our study offers novel insights on USP7 inhibition as cancer therapy.     

GSK-3 inhibitors induce chromosome instability

Background

Several mechanisms operate during mitosis to ensure accurate chromosome segregation. However, during tumour evolution these mechanisms go awry resulting in chromosome instability. While several lines of evidence suggest that mutations in adenomatous polyposis coli (APC) may promote chromosome instability, at least in colon cancer, the underlying mechanisms remain unclear. Here, we turn our attention to GSK-3 – a protein kinase, which in concert with APC, targets β-catenin for proteolysis – and ask whether GSK-3 is required for accurate chromosome segregation.

Results

To probe the role of GSK-3 in mitosis, we inhibited GSK-3 kinase activity in cells using a panel of small molecule inhibitors, including SB-415286, AR-A014418, 1-Azakenpaullone and CHIR99021. Analysis of synchronised HeLa cells shows that GSK-3 inhibitors do not prevent G1/S progression or cell division. They do, however, significantly delay mitotic exit, largely because inhibitor-treated cells have difficulty aligning all their chromosomes. Although bipolar spindles form and the majority of chromosomes biorient, one or more chromosomes often remain mono-oriented near the spindle poles. Despite a prolonged mitotic delay, anaphase frequently initiates without the last chromosome aligning, resulting in chromosome non-disjunction. To rule out the possibility of "off-target" effects, we also used RNA interference to selectively repress GSK-3β. Cells deficient for GSK-3β exhibit a similar chromosome alignment defect, with chromosomes clustered near the spindle poles. GSK-3β repression also results in cells accumulating micronuclei, a hallmark of chromosome missegregation.

Conclusion

Thus, not only do our observations indicate a role for GSK-3 in accurate chromosome segregation, but they also raise the possibility that, if used as therapeutic agents, GSK-3 inhibitors may induce unwanted side effects by inducing chromosome instability.