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

This Article Full Text (PDF) All Versions of this Article: 17/11/1550 02-1003fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by WARRENER, R. Articles by GABRIELLI, B. Search for Related Content PubMed PubMed Citation Articles by WARRENER, R. Articles by GABRIELLI, B.

Tumor cell-selective cytotoxicity by targeting cell cycle checkpoints ¹

ROBYN WARRENER², HEATHER BEAMISH², ANDREW BURGESS, NIGEL J. WATERHOUSE^*, NICHOLE GILES, DAVID FAIRLIE and BRIAN GABRIELLI³

Cancer Biology Program, Centre for Immunology and Cancer Research,
Institute of Molecular Biosciences, University of Queensland, Brisbane, Queensland Australia. 4102; and
^* Cancer Immunology, Peter MacCallum Cancer Institute, Melbourne, Victoria, Australia

³Correspondence: Centre for Immunology and Cancer Research, University of Queensland, Brisbane, Queensland. 4102, Australia. E-mail: bgabrielli{at}cicr.uq.edu.au

SPECIFIC AIMS

Cell cycle checkpoints are often defective in cancers, and as the normal function of these mechanisms is to protect cells from external stresses and internal errors that would compromise the integrity of the cell, drugs that target checkpoint mechanisms should be selective for tumor cells that are defective for the drug-sensitive checkpoint and potent cytotoxins. We examined the molecular mechanism by which the new class of anticancer agents, the histone deacetylase inhibitors (HDACIs), exert their tumor-selective cytotoxicity, investigating whether this class of drugs is targeting the checkpoints that normally ensure the fidelity of mitosis and whether compromising these checkpoints results in the cell death observed.

PRINCIPAL FINDINGS

1. HDACI-induced cell death occurs after exit from mitosis
To examine the timing of cell death during the cell cycle, HeLa cell cultures synchronized using a double thymidine block release protocol were treated with the HDACI azelaic bishydroxamic acid (ABHA, 100 µg/mL) immediately after release in G1. The entry of cultures into mitosis, marked by the increased cyclin B1/cdc2 activity, was little delayed by ABHA treatment compared with untreated cultures (Fig. 1 A). The proportion of cells with subdiploid DNA content, a marker of cell death, was constant in the untreated cultures but showed a rapid increase in the ABHA-treated cultures as the cells exited mitosis, with >40% of cells with subdiploid DNA content cells by 14 h after release (Fig. 1A ). Time lapse video microscopy showed ABHA-treated cells rounding up as they entered mitosis, then attempting to undergo cytokinesis. In every case examined the cells appeared to contract in size and displayed membrane blebbing, immediately or within 1–2 h of reattaching and flattening (Fig. 1B ).

View larger version (56K): [in this window] [in a new window]   Figure 1. HDACI-induced cell death correlates with exit from mitosis. A) Synchronized HeLa cells were either untreated (squares) or treated with ABHA (triangles) immediately after synchrony release. Cells were harvested and assayed from cyclin B1 kinase activity (filled symbols) or by FACS for the % of subdiploid cells (open symbols). The cyclin B1 kinase data are the mean of duplicate determinations. Data are typical of 3 separate experiments. B) Time lapse video microscopy of synchronized HeLa cells treated in S phase with ABHA. The time after release is shown for each frame in the sequence (h:mm). In cell 1, the asterisk in the 11:45 frame shows the position of one of the daughter cells that has reattached, then displayed membrane blebbing at a time similar to the other daughter cells that failed to reattach. In cell 2, arrows indicate the mitotic cells and its daughter cells.

The cell death observed was not caspase-independent mitotic catastrophe, but cells exiting mitosis revealed the membrane blebbing typical of apoptosis, which was inhibited by addition of the general caspase inhibitor zVAD-fmk. Only cells displaying blebbing stained with Annexin V. Quantitation of Annexin V staining showed that 40% cells of cells were stained 24 h after release, corresponding to the subdiploid population. Overexpression of the antiapoptotic protein Bcl-2 also blocked cell death, further demonstrating that cell death was via an apoptotic mechanism.

To determine whether exit from mitosis was necessary for the cell death, synchronized cells were loaded with TAT-N86 cyclin B1, a nondegradable version of cyclin B1 tagged with the HIV TAT transduction sequence, to block exit from mitosis. The TAT-N86 cyclin B1 blocked cells in mitosis, as expected. Addition of TAT-N86 cyclin B1 blocked cell death by ABHA, reducing the level of apoptosis to untreated control levels. A similar reduction in cell death was observed when cultures were blocked from exiting mitosis with the proteosome inhibitor MG132.

2. ABHA disrupts the mitotic spindle checkpoint
HDACI-treated HeLa cells display an aberrant mitosis, with condensed chromosomes failing to align at the midline of the mitotic cells. Aberrant mitosis should initiate a mitotic arrest by activating the mitotic spindle checkpoint that allows spindle defects to be resolved before exit from mitosis. This checkpoint is functional in HeLa cells, but the high levels of cell death and evidence of abnormal partitioning of the DNA in surviving ABHA-treated cells suggested that the checkpoint was not operating normally. Treatment with a low dose of ABHA (10 µg/mL) resulted in a high level of aberrant mitosis; however, there was little cell death and few multinuclear cells, a marker of failed mitosis. This suggested that in low-dose drug conditions, aberrant mitosis was being resolved and cells eventually underwent normal mitotic partitioning of their chromosomes. This implied that the mitotic checkpoint was functioning under these conditions, which should be evident by a delayed transit through mitosis. Measurement of the time for synchronized cultures to transit mitosis demonstrated that low-dose ABHA-treated cultures remained 50% longer in mitosis (4.5 h) than with control cultures, which transited mitosis in 3 h. Cultures treated with 100 µg/mL ABHA transited mitosis at only a slightly reduced rate than the control untreated cells. Quantitation of the proportion of normal and aberrant mitosis showed that the high proportion of aberrant mitosis in the low-dose drug-treated cultures resolved with time to undergo apparently normal mitosis. High-dose ABHA-treated cultures showed little evidence of mitotic cells at later time points

These results pointed to high-dose ABHA treatment overcoming the mitotic checkpoint. The mitotic checkpoint can be triggered by the microtubule disrupting agent nocodazole, and this was used to determine whether high-dose ABHA treatment could bypass the mitotic arrest. Treatment of synchronized cultures with nocodazole blocked cells in mitosis with a high level of cyclin B1/cdc2 activity (Fig. 2 ). Cells treated with 100 µg/mL ABHA bypassed the nocodazole-induced arrest, with cells exiting mitosis with kinetics similar to untreated cultures (Fig. 2) . Treatment with 10 µg/mL ABHA was inefficient at bypassing the arrest. In these cultures cyclin B1/cdc2 kinase activity increased, although to a lesser extent than in nocodazole-only treated cultures, and by 24 h after synchrony release they had only 50% the cyclin B1/cdc2 kinase activity of the nocodazole-treated cells (Fig. 2B ). Washout of 100 µg/mL ABHA after the completion of S phase (7 h after synchrony release) resulted in a stable mitotic arrest (Fig. 2B ). The decline in cyclin B1/cdc2 kinase activity in 100 µg/mL ABHA continuously treated cultures was accompanied by decrease in the cyclin B1 protein levels as seen with normal mitotic exit, whereas low-dose ABHA-treated cultures retained elevated levels of cyclin B1 as in the nocodazole treated, mitotically arrested cells (Fig. 2C ).

View larger version (17K): [in this window] [in a new window]   Figure 2. High-dose ABHA overcomes the nocodazole-induced mitotic spindle checkpoint. A) Synchronized HeLa cells were either untreated (squares) or treated with nocodazole-only (dashed line), 10 µg/mL (circles), or 100 µg/mL (triangles) ABHA plus nocodazole. Samples were harvested and assayed for cyclin B1 kinase activity. The data are the mean of duplicate determinations. Data are typical of 4 separate experiments. B) Samples from an experiment similar to that in panel A but harvested 24 h after synchrony release and with cyclin B1/cdc2 kinase activity measured. Samples are from untreated control (con) or treated with nocodazole-only (noco), 10 µg/mL or 100 µg/mL ABHA, or 100 µg/mL ABHA, then washed out after S phase (w/o). Data are the mean of triplicate determinations. C) Samples from the experiment shown in panel A harvested 14 h after release were immunoblotted for cyclin B1 and cdc2 protein levels.

3. Structurally different HDACIs behave identically to ABHA
There are a number of structural classes of HDACIs: short chain fatty acids, hydroxamic acids, cyclic tetrapeptides, and benzamides. A panel of seven HDACIs was tested at doses that produced similar levels of histone hyperacetylation. All produced >90% aberrant mitosis of an identical phenotype to ABHA, efficiently escaped the nocodazole-induced mitotic spindle checkpoint arrest, and produced a similar level of cell death in response to 100 µg/mL ABHA treatment. Thus, the ability to overcome the mitotic defects and cell death is a general property of HDACIs.

4. Loss of the G2 checkpoint and disruption of mitotic checkpoint are synthetically lethal
The preceding studies suggest that the ability of HDACIs to induce cell death is due to the synthetic lethality of drug-induced aberrant mitosis, a consequence of the defective G2 checkpoint, and disruption of the mitotic spindle checkpoint. To test this hypothesis, we took advantage of the observation that addition of 100 µg/mL ABHA in G2 phase did not produce the aberrant mitotic phenotype associated with S phase addition and caused little cell death. G2 phase addition of 100 µg/mL ABHA did overcome the nocodazole arrest with slower kinetics than S phase addition of the drug. Nocodazole treatment alone resulted in 20% cell death, but this increased to almost 70% with G2 phaseABHA addition. Pretreatment of cultures with the caspase inhibitor DEVD-fmk reduced the level of cell death by 50% but had no effect on the level of cyclin B1/cdc2 kinase activity detected in these cells, indicating that the reduction in cyclin B1/cdc2 kinase activity was a consequence of mitotic exit rather than apoptotic cell death.

CONCLUSIONS AND SIGNIFICANCE

We have demonstrated that HDACIs can overcome the mitotic spindle checkpoint and cells prematurely exit mitosis without undergoing the normally tightly controlled partitioning of sister chromatids during anaphase/telophase. This is the first demonstration of a pharmacological agent with this activity. We have demonstrated two activities of HDACIs that can be resolved by their different requirements for drug dose and timing of drug addition. The first is an aberrant mitosis characterized by the failure of chromosome congression to the metaphase plate. The second effect of these drugs was the ability to escape the mitotic spindle checkpoint. The exact mechanisms by which these drugs affect these functions is not yet known, although as both chromosome movement and the spindle checkpoint involve the centromere and kinetochores, hyperacetylation of the normally hypoacetylated centromeric heterochromatin is implicated in the disruption of these mitotic functions.

The premature exit from mitosis before chromosome alignment observed with high-dose HDACI treatment results in extensive and rapid apoptosis. Disruption of the mitotic spindle checkpoint in somatic cells by targeting spindle checkpoint proteins causes premature exit from mitosis but not apoptosis. This suggests that HDACI treatment may also prime cells for apoptosis, possibly by down-regulating the expression of antiapoptotic proteins such as Bcl-2 and Bcl-xL or up-regulation of one of the proapoptotic Bcl-2 family members such as Bax.

The key finding of this study is the molecular basis for the tumor-selective toxicity of HDACIs. We previously demonstrated that resistance to killing by these drugs is related to an intact G2 checkpoint response. The aberrant mitosis observed in checkpoint defective cells treated with HDACI during S phase may be related to loss of the checkpoint. However, the aberrant mitosis is not sufficient to cause cell death, as both low-dose ABHA and removal of high-dose ABHA before mitosis caused high levels of aberrant mitosis but little cell death. Likewise, disruption of the mitotic spindle checkpoint by HDACIs in itself is not sufficient to cause cell death, but dysfunction of the G2 and mitotic checkpoint is synthetically lethal. Our data demonstrate that the selective cytotoxicity of HDACIs is the result of the failure of chromosome congression in mitosis and failure of the mitotic checkpoint to stop these cells from exiting mitosis without correctly segregating their chromosomes, which signals apoptosis. In cells with an intact HDACI-sensitive G2 checkpoint, disruption of the spindle checkpoint would not be an issue as these cell do not enter mitosis (Fig. 3 ). Thus, HDACIs are one of the first examples of potentially useful chemotherapeutic drugs that specifically kill tumor cells by targeting cell cycle checkpoint controls. This provides not only the specificity of action, but also the toxicity for these drugs. Elucidation of how these drugs cause the aberrant mitosis, how it is connected with loss of the G2 checkpoint, and how they overcome the mitotic spindle checkpoint will provide new targets for developing more specific and potent anticancer drugs.

View larger version (14K): [in this window] [in a new window]   Figure 3. Disruption of two checkpoints underlies the selective cytotoxicity of HDACIs. Resistant cells with an intact HDACI-sensitive checkpoint arrest in G2 with HDACI treatment. In cells with a defective checkpoint, cells transit through G2 in to an aberrant mitosis, then prematurely exit mitosis due to disruption of the mitotic checkpoint caused by HDACI treatment. These cells either die by apoptosis or survive as multinuclear cells.

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-1003fje; doi: 10.1096/fj.02-1003fje

² These authors contributed equally to this work.

This article has been cited by other articles:

				G. Eot-Houllier, G. Fulcrand, Y. Watanabe, L. Magnaghi-Jaulin, and C. Jaulin Histone deacetylase 3 is required for centromeric H3K4 deacetylation and sister chromatid cohesion Genes & Dev., October 1, 2008; 22(19): 2639 - 2644. [Abstract] [Full Text] [PDF]

				Y. Dai, S. Chen, C. A. Venditti, X.-Y. Pei, T. K. Nguyen, P. Dent, and S. Grant Vorinostat synergistically potentiates MK-0457 lethality in chronic myelogenous leukemia cells sensitive and resistant to imatinib mesylate Blood, August 1, 2008; 112(3): 793 - 804. [Abstract] [Full Text] [PDF]

				A. M. Sargeant, R. C. Rengel, S. K. Kulp, R. D. Klein, S. K. Clinton, Y.-C. Wang, and C.-S. Chen OSU-HDAC42, a Histone Deacetylase Inhibitor, Blocks Prostate Tumor Progression in the Transgenic Adenocarcinoma of the Mouse Prostate Model Cancer Res., May 15, 2008; 68(10): 3999 - 4009. [Abstract] [Full Text] [PDF]

				M. Malinen, A. Saramaki, A. Ropponen, T. Degenhardt, S. Vaisanen, and C. Carlberg Distinct HDACs regulate the transcriptional response of human cyclin-dependent kinase inhibitor genes to trichostatin A and 1{alpha},25-dihydroxyvitamin D3 Nucleic Acids Res., January 17, 2008; 36(1): 121 - 132. [Abstract] [Full Text] [PDF]

				M. T. Epping, L. Wang, J. A. Plumb, M. Lieb, H. Gronemeyer, R. Brown, and R. Bernards A functional genetic screen identifies retinoic acid signaling as a target of histone deacetylase inhibitors PNAS, November 6, 2007; 104(45): 17777 - 17782. [Abstract] [Full Text] [PDF]

				L. Magnaghi-Jaulin, G. Eot-Houllier, G. Fulcrand, and C. Jaulin Histone Deacetylase Inhibitors Induce Premature Sister Chromatid Separation and Override the Mitotic Spindle Assembly Checkpoint Cancer Res., July 1, 2007; 67(13): 6360 - 6367. [Abstract] [Full Text] [PDF]

				T. Choudhuri, S. C. Verma, K. Lan, M. Murakami, and E. S. Robertson The ATM/ATR Signaling Effector Chk2 Is Targeted by Epstein-Barr Virus Nuclear Antigen 3C To Release the G2/M Cell Cycle Block J. Virol., June 15, 2007; 81(12): 6718 - 6730. [Abstract] [Full Text] [PDF]

				B. Gabrielli, Y. Q. Chau, N. Giles, A. Harding, F. Stevens, and H. Beamish Caffeine Promotes Apoptosis in Mitotic Spindle Checkpoint-arrested Cells J. Biol. Chem., March 9, 2007; 282(10): 6954 - 6964. [Abstract] [Full Text] [PDF]

				L. Geng, K. C. Cuneo, A. Fu, T. Tu, P. W. Atadja, and D. E. Hallahan Histone Deacetylase (HDAC) Inhibitor LBH589 Increases Duration of {gamma}-H2AX Foci and Confines HDAC4 to the Cytoplasm in Irradiated Non-Small Cell Lung Cancer Cancer Res., December 1, 2006; 66(23): 11298 - 11304. [Abstract] [Full Text] [PDF]

				L. R. Saunders and E. Verdin Ornithine decarboxylase activity in tumor cell lines correlates with sensitivity to cell death induced by histone deacetylase inhibitors. Mol. Cancer Ther., November 1, 2006; 5(11): 2777 - 2785. [Abstract] [Full Text] [PDF]

				Y. Li, G. D. Kao, B. A. Garcia, J. Shabanowitz, D. F. Hunt, J. Qin, C. Phelan, and M. A. Lazar A novel histone deacetylase pathway regulates mitosis by modulating Aurora B kinase activity Genes & Dev., September 15, 2006; 20(18): 2566 - 2579. [Abstract] [Full Text] [PDF]

				B. G. Gabrielli Histone Deacetylase Inhibitors Deliver Their Therapeutic Benefit by Targeting the Cell Cycle Am. Assoc. Cancer Res. Educ. Book, April 1, 2006; 2006(1): 318 - 321. [Full Text] [PDF]

				R. R. Rosato, J. A. Almenara, S. C. Maggio, P. Atadja, R. Craig, J. Vrana, P. Dent, and S. Grant Potentiation of the lethality of the histone deacetylase inhibitor LAQ824 by the cyclin-dependent kinase inhibitor roscovitine in human leukemia cells Mol. Cancer Ther., November 1, 2005; 4(11): 1772 - 1785. [Abstract] [Full Text] [PDF]

				M. R. Acharya, A. Sparreboom, J. Venitz, and W. D. Figg Rational Development of Histone Deacetylase Inhibitors as Anticancer Agents: A Review Mol. Pharmacol., October 1, 2005; 68(4): 917 - 932. [Abstract] [Full Text] [PDF]

				W.-S. Xu, G. Perez, L. Ngo, C.-Y. Gui, and P. A. Marks Induction of Polyploidy by Histone Deacetylase Inhibitor: A Pathway for Antitumor Effects Cancer Res., September 1, 2005; 65(17): 7832 - 7839. [Abstract] [Full Text] [PDF]

				S. Doi, H. Soda, M. Oka, J. Tsurutani, T. Kitazaki, Y. Nakamura, M. Fukuda, Y. Yamada, S. Kamihira, and S. Kohno The histone deacetylase inhibitor FR901228 induces caspase-dependent apoptosis via the mitochondrial pathway in small cell lung cancer cells Mol. Cancer Ther., November 1, 2004; 3(11): 1397 - 1402. [Abstract] [Full Text] [PDF]

				J.-H. Park, Y. Jung, T. Y. Kim, S. G. Kim, H.-S. Jong, J. W. Lee, D.-K. Kim, J.-S. Lee, N. K. Kim, T.-Y. Kim, et al. Class I Histone Deacetylase-Selective Novel Synthetic Inhibitors Potently Inhibit Human Tumor Proliferation Clin. Cancer Res., August 1, 2004; 10(15): 5271 - 5281. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 17/11/1550 02-1003fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by WARRENER, R. Articles by GABRIELLI, B. Search for Related Content PubMed PubMed Citation Articles by WARRENER, R. Articles by GABRIELLI, B.