![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 20, Issue 8, 2218-2228, April 15, 2009
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*Laboratory of New Drug Development, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY 10021; and Astra-Zeneca, Alderley Park, Macclesfield, Cheshire SK10 4TG, United Kingdom
Submitted August 28, 2008;
Revised February 3, 2009;
Accepted February 6, 2009
Monitoring Editor: Mark J. Solomon
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). All three Aurora kinases are overexpressed in a variety of human cancers. Aurora A amplification/overexpression has been detected in human breast, bladder, colon, ovarian, and pancreatic cancers (Sakakura et al., 2001; Fraizer et al., 2004; Tong et al., 2004). Aurora B is overexpressed in cancer cells, and an increased level of Aurora B correlates with advanced stages of colorectal cancer (Katayama et al., 1999; Giet et al., 2005). It has been hypothesized that these elevated levels of Aurora kinase activity may promote tumor evolution either by providing a growth advantage or promoting genetic instability.
The expression levels, as well as the kinase activity, of all three Aurora kinases are regulated in a cell cycle–dependent manner and peak during mitosis. In addition, Aurora kinases A and B show differential subcellular localization. Aurora A is detected in proliferating cells at the centrosome, where it regulates centrosome function, and also along mitotic spindles. In contrast, Aurora B shows dynamic changes in localization during mitosis. It associates along chromosomes during prophase, inner centromeres until metaphase, and the central spindle at the cleavage furrow in anaphase and ultimately resides in the midbody during cytokinesis (Andrews et al., 2003; Carmena and Earnshaw, 2003; Meraldi et al., 2004). This localization pattern reflects the critical role Aurora B plays as a chromosome passenger protein to ensure accurate chromosome segregation and timely cytokinesis.
In fact, depletion of Aurora B by double-strand RNA interference (RNAi; oligo) leads to decreased histone H3 phosphorylation, defective chromosome condensation, failed cytokinesis, and polyploidy induction in Drosophila (Giet and Glover, 2001). Given their importance in mitosis and overexpression in human cancers, Aurora kinases have been identified as promising therapeutic targets, and considerable effort has been devoted to developing inhibitors of the kinases. ZM447439 (AstraZeneca, Alderley Park, Macclesfield, Cheshire, United Kingdom), Hesperadin (Boehringer Ingelheim, Ridgefield, CT), and VX-680 (Vertex Pharmaceuticals, Cambridge, MA) are a few of the small molecule Aurora family inhibitors that have been shown to induce an aberrant mitosis and polyploidy, depending upon the status of a p53-dependent postmitotic checkpoint (Ditchfield et al., 2003; Hauf et al., 2003; Keen and Taylor, 2004; Gizatullin et al., 2006).
The retinoblastoma (Rb) protein is a tumor suppressor that plays a pivotal role in the negative control of the cell cycle and in tumor progression. It has been shown that Rb is responsible for a major G1 checkpoint, blocking S-phase entry and cell growth. Loss of Rb functions may induce cell cycle deregulation, leading to a malignant phenotype (Giacinti and Giordano, 2006). Compromise of Rb and p53 functions leads to suppression of a postmitotic checkpoint that prevents endoreduplication after aberrant mitosis by eliciting a pseudo G1 arrest that depends on p53, p21, and Rb (Borel et al., 2002). Previous studies using small molecule inhibitors of Aurora kinases also reported the induction of polyploidy or a pseudo G1 arrest is dependent on these same pathways (Keen and Taylor, 2004).
On the basis of these findings, in this study we have made an effort to understand the roles of Rb and p53 in regulating the postmitotic checkpoint after Aurora B inhibition. We use specific small interfering RNA (siRNA) and the Aurora B small molecule inhibitor, AZD1152 HQPA, to selectively down-regulate Aurora B activity (Wilkinson et al., 2007). AZD1152 HQPA is the active metabolite of AZD1152, a selective Aurora B inhibitor (Aurora A, Ki = 1369 nM; Aurora B, Ki = 0.3 nM) now in clinical trial (Carvajal et al., 2006). AZD1152 HQPA will be herein referred to as simply AZD1152. In accord with previous studies of Aurora B inhibitors, AZD1152 induces polyploidy in tumor cells. However, in contrast to those results, AZD1152-mediated polyploidy occurs independent of p53 status. We also provide evidence that the inhibition of Aurora B results in the inhibition of Rb phosphorylation concomitant with polyploidy induction. Furthermore, we demonstrate that Aurora B regulates the postmitotic checkpoint by directly phosphorylating Rb at serine 780. Taken together, we propose that Aurora B protects the integrity of key mitotic processes, as well as prevents endoreduplication and polyploidy formation by directly regulating Rb phosphorylation status.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Clonogenic Assay
Log-phase cells were plated, in triplicate, onto 100-mm dishes at 1000 per dish and were allowed to attach for 24 h before treatment. At the end of treatment, cells were cultured in drug-free medium for 7–10 d. The resulting colonies were scored after staining with 0.01% crystal violet.
Fluorescent Microscopy
Cells were collected after drug treatment and fixed in 3% paraformaldehyde. The nuclear morphology of cells was examined under fluorescence microscope after staining with 4', 6-diamidino-2-phenylindole (DAPI) to a final concentration of 25 µg/ml. Cells were scored as apoptotic based on the presence of condensed fragmented chromatin. Multinucleated cells were defined as cells containing huge, multilobed nuclei. A minimum of 400 cells were counted for each sample and taken as a percentage of untreated cells.
Immunofluorescence Assay
HCT 116 cells were plated in four-well chamber slides and treated with or without 50 nM AZD1152, and cells were incubated for 10 min on ice followed by incubation with 100 mM K-PIPES at pH 6.9, 10 mM EGTA, 1 mM MgCl2, and 0.2% Triton X-100 and 4% paraformaldehyde for another 10 min. The cells were then blocked in 2% goat serum followed by incubation in primary antibody as described. The following antibodies were used: monoclonal anti-tubulin (Sigma; 1:250), polyclonal anti-NuMA was a gift from Duane A. Compton (Dartmouth Medical School, Hanover, NH), rabbit polyclonal anti-phospho-ser 10 histone H3 (Cell Signaling Technology, Danvers, MA; 1:250), and anti-human CREST (1:10000) was a generous gift from B. R. Brinkley (Baylor College of Medicine, Houston, TX). DNA was labeled with DAPI at a final concentration of 25 µg/ml for 10 min.
Time-Lapse Fluorescent Microscopy
HCT116 cells expressing green fluorescent protein (GFP)-histone 2B and GFP-tubulin were developed as previously described (Tse and Schwartz, 2004). For these studies, cells were grown on 40-mm round coverslips and allowed to attach for 48 h before assembly onto a FCS2 close-system flow-observation chamber (Bioptechs, Butler, PA). The chamber was mounted onto the stage of a Zeiss Axiovert 200 M inverted microscope (Thornwood, NY) with the chamber and coverslip temperature maintained at 37°C. Drug containing media was introduced into the chamber via a peristaltic pump. Epifluorescent and phase-contrast images were acquired with a 40x objective lens, using a cooled CCD camera every 5 min, with exposure time limited to 4–5 s/image (MicroMax-1300YHS, Roper Scientific, Tucson, AZ). For each time point, images were taken with five to six different focal planes along the z-axis 2–3 µm apart. Imaging data were analyzed using MetaMorph (Universal Imaging, West Chester, PA).
For time-lapse experiments, cells were cultured in the same medium containing 20 mM HEPES without phenol red.
Flow Cytometry
Biparameter flow cytometry was performed as previously described (Motwani et al., 1999). Cells were analyzed for DNA content after propidium iodide staining (PI) and for mitotic index after labeling with the MPM-2 mAb (Upstate Biotechnology, Lake Placid, NY), Samples were analyzed on a FACScan (Becton Dickinson, Franklin Lakes, NJ) for cell cycle distribution and mitotic index (percentage of MPM-2–positive cells) using the Cell Quest software (Becton Dickinson).
siRNA Transfection
HCT 116 cells were plated on 60-mm plates and transfections using oligofectamine (Invitrogen, Carlsbad, CA) were carried out according to the manufacturer's protocol. The siRNA sequences used were Aurora B (5'-AACGCGGCACUUCACAAUUGA-3'; Lampson and Kapoor, 2005) and Aurora A (Aurora-A, 725AUGCCCUGUCUUACUGUC A743; Kufer et al., 2002) purchased from Dharmacon Research (Lafeyette, CO), and control siRNA was purchased from Santa Cruz Biotechnology (Santa Cruz, CA).
Cell Extraction, Immunoprecipitation, and Immunoblotting
Cell lysates were prepared as previously described (Nair et al., 2002). Fifty micrograms of protein were fractionated by SDS-PAGE and transferred onto Immobilon membranes (Millipore, Billerica, MA). Equal protein loading was confirmed by Amido black staining (Bio-Rad, Hercules, CA). After blocking with 5% nonfat milk, membranes were probed with primary antibodies. The following antibodies were used in this study: rabbit polyclonal and mouse monoclonal anti-Aurora B (1:1000), rabbit polyclonal Aurora A, rabbit polyclonal to p16, rabbit polyclonal to E2F1 from Abcam (Cambridge, United Kingdom), mouse monoclonal anti-Rb (1:1000) from BD Biosciences (San Jose, CA), mouse monoclonal anti-p53, cdk2, cdk4, cyclin B, cyclin E, rabbit polyclonal to Rb, mouse monoclonal to E2F1, and cyclin D from Santa Cruz Biotechnology, mouse monoclonal anti-p21 (1:1000) was from Oncogene (Cambridge, MA), mouse monoclonal anti- H2AX and anti-tubulin were from Upstate Biotechnology rabbit polyclonal-anti-phospho-serine Histone 3 (1:1000) mouse monoclonal anti-hemagglutinin (HA; 1:1000), rabbit polyclonal to phospho-Rb (Ser 807/811), Ser 780 and Ser 795, and rabbit polyclonal to phosphor-histone H3 (Ser 10) were from Cell Signaling. Bound primary antibodies were detected with horseradish peroxidase–conjugated secondary antibodies (ICN/Jackson ImmunoResearch, West Grove, PA) and visualized by enhanced chemiluminescence reagent (Amersham Pharmacia, Piscataway, NJ).
Immunoprecipitation was performed by using 500 µg to 1 mg of soluble protein. First, lysates were incubated with 2 µg of the indicated primary antibody overnight at 4°C followed by incubation with 50 µl of protein A-agarose beads (Upstate Biotechnology). Immunocomplexes were washed five times in lysis buffer, suspended in 30 µl of 2x sample buffer, and fractionated by SDS-PAGE. Western blot analysis was performed as previously described.
Plasmid Transfections
Plasmid transfections were carried out as followed in respective cell lines. Cells were plated so as to gain 50% confluence after 24 h in 60-mm tissue culture plates. Transfections were carried out using Fugene 6 transfection reagent (Roche, Indianapolis, IN) in OPTI MEM 1 media (Invitrogen) and 2 µg of HA-tagged Rb plasmid (Addgene, Cambridge, MA), its mutant forms or HA-tagged p16 (a generous gift from Dr. Andrew Koff, Memorial Sloan Kettering Cancer Center) were used per plate. Treatments were carried out 16 h after transfection.
Luciferase Assay
Cells were plated on 24-well plates and TransLucent E2F1 reporter vector was used (Panomics, Fremont, CA). Cell extracts were analyzed for luciferase activity using the Dual Luciferase Reporter Assay (Promega, Madison, WI).
Aurora Kinase Assays
Purified active and inactive forms of Aurora B kinase were obtained from Upstate Biotechnology for in vitro kinase assays and histone H3 and full-length Rb (QED Biosciences, San Diego, CA) were used as substrates. Different Rb constructs were a gift from Dr. Nikola P. Pavletich (Memorial Sloan Kettering Cancer Center; Rubin et al., 2005).
Reactions were carried out in 30 µl kinase buffer (250 mM HEPES, 50 mM MgCl2, 50 mM β-glycerophosphate, 5 mM DTT, 12.5 mM EGTA, 0.5 mM sodium vanadate, and 5 µM sodium fluoride) containing 5 µCi of 32P-ATP, 15 µmol/L ATP at 30°C for 20 min. Products were resolved by SDS-PAGE and visualized by autoradiography. For in vivo kinase assays, Aurora B was immunoprecipitated from lysates prepared from control or AZD1152-treated HCT-116 cells. The kinase reaction was performed under the same conditions as the in vitro assay using full-length human Rb as the substrate.
Mutagenesis, Overexpression, and Purification of RbC-99
6xHis-tagged RbC-99 was mutated using Quick Change Site Directed Mutagenesis Kit (Stratagene, Cedar Creek, TX) according to the manufacturer's protocol. Mutants and wild-type RbC-99 were expressed in BL-21 (Stratagene) and purified by Ni-NTA column chromatography (Qiagen, Hilden, Germany).
MALDI Q-TOF Mass Spectra and LC-MS/MS Sequencing
MALDI Q-TOF MS (matrix-assisted laser desorption/ionization–quadrupole time-of-flight mass spectrometry) and LC-MS/MS (liquid chromatography–tandem mass spectrometry) were performed as described previously (Xu et al., 2005). Briefly, 1 µg of purified Rb construct RbC-99 (771–928) was incubated with active or inactive Aurora B kinase in 250 mM HEPES, 50 mM MgCl2, 50 mM B-glycerophosphate, 5 mM DTT, 12.5 mM EGTA, 0.5 mM sodium vanadate, 5 µM sodium fluoride 15 µmol/l ATP at 30°C for 20 min, and resolved on SDS page. The bands corresponding to RbC99 were excised and after tryptic digestion, a MALDI Q-TOF MS spectra of the tryptic peptides was acquired with a Micromass Q-TOF Ultima MALDI mass spectrometer (Waters, MA). Titanium dioxide chromatography was used to enrich the phosphopeptides in the tryptic digest and MALDI Q-TOF MS spectra was acquired on the enriched peptides and subtracted from background. The enriched peptide mixture was also sequenced using LC-ESI Q-TOF MS.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Crosio et al., 2002; Sugiyama et al., 2002). As indicated by the Western blot, 50 nM AZD1152 in HCT-116 cells completely inhibited histone H3 phosphorylation, similar to that achieved with BSi, hence validating target inhibition (Figure 1A). Immunofluorescent microscopy showed both modalities of Aurora B inhibition resulted in misaligned chromosomes (DAPI, blue) in the setting of an apparently normal bipolar spindle (
-tubulin, green) that was attached to kinetochore proteins (CREST, pink; Figure 1Bi, a and b). Figure 1Bi further shows that spindle poles and centrosomes identified using immunofluorescence probes for pole associated protein NUMA (orange, c and d) and centrosome marker, 
-tubulin (red, e and f) were intact and normal. Down-regulation of Aurora B by siRNA had no effect on the centrosome however, Aurora A down-regulation (ASi) is shown to deplete it (g and h). Despite these changes in mitosis, Aurora B inhibition did not alter the rate of mitotic entry or exit as determined by phospho MPM2 (a mitotic marker) positive cells (Figure 1Bii).
|
tubulin also confirmed that AZD1152 disrupted chromosome alignment despite normal bipolar spindle formation (Figure 1D). Conversely, Aurora A siRNA down-regulation did not result in these mitotic changes or multinucleation, instead preferentially induced apoptosis (data not shown). These observations confirm that the phenotype of cells treated with AZD1152 are consistent with that of selective Aurora B inhibition as predicted by in vitro kinase assays, reasserting the critical role of Aurora B in regulating chromosome alignment, spindle checkpoint control and cytokinesis (Hannak et al., 2001; Ducat and Zheng, 2004).
Aurora B Inhibition Leads to Polyploidy Despite Activation of a p53-dependent Postmitotic Checkpoint
It has been previously proposed that under conditions of DNA damage or aberrant mitotic exit, the p53-p21 axis is induced, resulting in cyclin-dependent kinase (cdk) inhibition and hypophosphorylation of Rb, the active form of the protein that elicits a pseudo G1 arrest presumably via repression of E2F transcription factors (Lanni and Jacks, 1998; Motwani et al., 2000; Andreassen et al., 2001b). We examined the response of this checkpoint to the aberrant mitosis induced by AZD1152 in HCT-116 cells, which possess an intact p53–p21 pathway. As shown in Figure 2Ai, p53 and p21 were induced at 8 and 18 h, respectively, after AZD1152 treatment. Although cyclin-dependent kinase 2 (cdk2) protein levels were unchanged, cdk2 kinase activity immunoprecipitated from drug treated cells was completely inhibited by 24 h (Figure 2Aii). Importantly, the fastest migrating form of Rb (representing un- and hypo-phosphorylated Rb) appeared at 18 h, well before cdk2 kinase inhibition was observed (Figure 2Aii). Similarly, cdk4 immunoprecipitation-kinase assays in drug-treated cells confirmed that cdk4 kinase activity was maintained while the hypophosphorylated form of Rb appeared (Figure 2Aiii). To further rule out any role played by cdk4 in the AZD-induced polyploidy, HCT 116 cells transiently overexpressing p16 were treated with AZD1152, and the binding of Rb and E2F1 was assessed by coimmunoprecipitation assay. As shown in Figure 2Aiv p16 overexpression by itself or with treatment showed no effect in the binding of Rb to E2F1, indicating the release of E2F1 upon Aurora B inhibition is independent of p21/p16 and may be independent of p53/cdk2 pathway. These results strongly suggest that the hypophosphorylation of Rb in the context of Aurora B inhibition may occur independently of cdk inhibition.
|
), we elected to test the effect of AZD1152 in 17 different cell lines of varying p53 status (eight for wild type, two nulls, and seven mutants), including MCF-10A, a normal breast cell line. As shown in Table 1, AZD1152 induced polyploidy and hypophosphorylation of Rb (data not shown) in all the cell lines, independent of p53 status. Thus, although Aurora B inhibition does induce p53 and p21, these proteins are ineffective at preventing polyploidy in the context of selective Aurora B inhibition.
|
|
Aurora B Phosphorylates Rb In Vitro and In Vivo
Given that we observed Rb hypophosphorylation after Aurora B inhibition to occur independent of cdk inhibition, we next explored the possibility that Aurora B directly phosphorylates Rb to prevent endoreduplication after an aberrant mitosis. As shown in Figure 4A, Rb and Aurora B could be coimmunoprecipitated from HCT-116 cells. The ability of Aurora B to phosphorylate Rb was tested in an in vitro kinase assay utilizing Aurora B immunoprecipitated from control or AZD1152-treated HCT-116 cells and purified Rb as a substrate. Although immunoprecipitated Aurora B from untreated cells successfully incorporated radiolabeled 32P onto the Rb substrate, kinase activity directed at Rb was significantly decreased in drug-treated cells (Figure 4B). Further, recombinant human Aurora B directly phosphorylated full-length Rb in vitro (Figure 4C, right panel, lane 2) at levels commensurate to that observed with histone H3 (Figure 4C, left panel, lane 2). Conversely, purified inactive Aurora B failed to phosphorylate both Rb and histone H3 (Figure 4C, lane 1 in right and left panels, respectively).
|
Aurora B Phosphorylates Rb on Serine 780
To identify the specific residue(s) of Rb that are phosphorylated by Aurora B kinase, the 771–928 C-terminus construct of Rb (RbC-99) was incubated with purified inactive (lane 1) or active (lane 2, Figure 5A) Aurora B in an in vitro kinase assay. After fractionation by SDS-PAGE, the protein bands of interest were excised from the gel and digested with trypsin. The digested mixture was enriched for phosphopeptides by TiO2 method (Larsen et al., 2005) and were analyzed by MALDI-Q-TOF and by LC-MS/MS. Three peptides were identified by MALDI Q-TOF after subtracting background from control (Figure 5Aii). Six potential Aurora B phosphorylation sites (in bold) were identified in three peptides with amino acid sequences, YASTRPPTLSPIPHIPR (771–787), IYISPLKSPYKISEGLPTPTK (804–824), and MTPRSRILVSIGESFGTSEK (825–844). All putative phosphorylation sites on Rb-C99 (771–928) were mutated to alanine individually or in combination, and Aurora B in vitro kinase assays were carried out using these mutants as substrates. As shown in Figure 5B, the only mutation which consistently decreased the level of phosphorylation alone or in combination with other sites was Serine 780, a known cdk4 site. The efficiency and timing of Rb dephosphorylation at Serine 780 was examined in vivo utilizing a phosphospecific antibody directed at serine 780 (Figure 5C). A decrease in phosphorylation on serine 780 was observed at 12–18 h after treatment with AZD1152, correlating to the time course observed with the emergence of hypophosphorylated Rb. Of note, no inhibition of Rb phosphorylation at serine 807/811 was observed. Thus, serine 780 is a physiological Aurora B specific phosphorylation site.
|
|
In G1, Rb is pivotal for regulating the cellular commitment to entering S-phase and undergoing DNA synthesis by directly associating with members of the E2F family transcription factors. Binding of hypophosphorylated Rb to the E2F1 blocks expression of the E2F target genes required for the G1-S transition; hyperphosphorylation of Rb by G1 cyclin:cdks leads to dissociation of Rb from E2F, expression of G1-S genes, and DNA replication. Similarly, the Rb:E2F interaction also has been shown to be a critical regulator of the cellular commitment to endoreduplication. Therefore, we went on to examine the potential significance of this interaction for regulating DNA synthesis after Aurora B inhibition by measuring Rb:E2F1 complexes in SaOs-2 cells expressing the three Rb mutants described above. In cells expressing wild-type Rb or the S780A mutant, AZD1152 treatment resulted in decreased Rb:E2F1 complexes, consistent with a release of E2F1 and promotion of DNA synthesis necessary for polyploid cell formation. Conversely, in cells expressing the S780D mutant, AZD1152 treatment did not result in a reduction of detected Rb:E2F complexes, suggesting that phosphorylation at this site by Aurora B promotes association with E2F1. Overexpression of p16 had no effect on the binding of Rb:E2F1 further rules out any role for cdks in AZD induced E2F1 release from Rb (Figure 2Aiv).
To further delineate the functional significance of these findings, E2F1 promoter activity was measured via luciferase assay (cells were cotransfected with vector containing E2F1 consensus sequence fused to firefly luciferase gene and Rb). In cells expressing the wild type or S780A mutant, AZD1152 treatment resulted in increased E2F1 promoter activity, a requisite event for DNA synthesis (Figure 6C). In contrast, the S780D phospho-mimetic Rb mutant blocked AZD1152-mediated enhancement of E2F1 promoter activity (Figure 6C). This result is further confirmed in the expression level of cyclin E, an E2F1 target gene (Figure 6B). Thus, we conclude that Aurora B phosphorylation of Rb at serine 780 after an aberrant mitosis negatively regulates endoreduplication by promoting Rb binding to E2F1, thereby preventing E2F1 promoter activation and subsequent polyploid cell formation.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) and VX-680 (Aurora A, Ki = 0.6 nM; Aurora B, Ki = 18 nM; Aurora C, Ki = 4.6 nM; Harrington et al., 2004; Carvajal et al., 2006). Whereas the phenotype of Aurora B inhibition predominates in cells treated with these drugs, the impact upon other Aurora family members cannot be completely discounted. Conversely, AZD1152 possesses profound selectivity for Aurora B (Aurora A, Ki = 1369 nM; Aurora B, Ki = 0.3 nM). The pharmacological effects of AZD1152 on cells closely mimics those observed with Aurora B knockdown via siRNA: decreased histone H3 phosphorylation at serine 10 and override of the mitotic checkpoint, leading to aberrant mitosis characterized by chromosomal mis-segregation, failed cytokinesis, mitotic exit, and endoreduplication/polyploidy (Figure 1). Hence, AZD1152 is a uniquely selective tool that can be used to study the biological role of Aurora B in human tumor cells.
Utilizing both this reagent and an siRNA approach, we addressed the question of how Aurora B inhibition results in polyploidy. The p53 tumor suppressor has been implicated as the regulator of a "postmitotic checkpoint" that functions to block endoreduplication after exit from failed mitoses caused by mitotic spindle poisons (Lanni and Jacks, 1998; Motwani et al., 2000; Andreassen et al., 2001a). It has been suggested that p53 also prevents endoreduplication after Aurora inhibition. Studies with combined Aurora A/B inhibitors ZM44739 and VX-680 reported that mutant p53 promoted increased percentages of polyploidy relative to cells with intact p53 function (Ditchfield et al., 2003; Hauf et al., 2003; Harrington et al., 2004; Gizatullin et al., 2006). However, selective Aurora B inhibition with AZD1152 demonstrated no such dependence on the p53 pathway: comparable levels of polyploidy were observed in HCT-116 cells (p53 wild type, 87% polyploidy) relative to isogenic HCT-116 p53 null (73% polyploidy) and HCT-116 p21 null (80% polyploidy) cells (Figure 2B and Supplemental Movie 1). An extensive survey involving AZD1152 treatment of cell lines varying in p53 status (n = 8, wild type; n = 9, mutant/null for p53; Table 1) confirmed that although a range of polyploidy was observed (30–87%), rates of polyploidy induction did not correlate with p53 status. Taken together, the data demonstrates that p53 status does not predict the cellular propensity to undergo polyploidy induction with selective Aurora B inhibition.
p53 is thought to prevent endoreduplication and elicit cell cycle arrest through p21 induction, leading to inhibition of G1 cdk activity and Rb hypophosphorylation (Lanni and Jacks, 1998; Motwani et al., 2000; Andreassen et al., 2001b). Indeed, down-regulation of cdk 2 activity and the appearance of the fastest migrating form of Rb on immunoblot (representing both un- and hypo-phosphorylated Rb; Ezhevsky et al., 2001) has been correlated with exit from an aberrant mitosis into a pseudo G1 phase (Lanni and Jacks, 1998; Motwani et al., 1999, 2000; Chan et al., 2007). Rb-negative cells are more susceptible to polyploidy induction in response to a variety of stimuli, including treatment with mitotic spindle poisons (Di Leonardo et al., 1997), exogenous p21 expression (Niculescu et al., 1998), and ionizing radiation (Avni et al., 2003), consistent with a model in which hypophosphorylated Rb inhibits endoreduplication and polyploidy.
Importantly, however, we observed that polyploidy was induced in HCT-116 cells despite increases in p53 and p21 protein levels (Figure 2A), implying that Aurora B inhibition overcomes a potential p53-mediated arrest. In fact, polyploidy was induced in the presence of hypophosphorylated Rb (Figure 2B). Furthermore, decreased polyploidy was observed with Aurora B inhibition in cells that lack Rb compared with cells with intact Rb (Figure 3C), suggesting that Rb may functionally promote polyploidy in this cellular context. Introduction of wild-type Rb into Rb null cells enhanced AZD1152-induced polyploidy concomitant with the appearance of hypophosphorylated protein, whereas transfection of siRNA targeting Rb led to decreased polyploidy in Rb wild-type cells (Figure 3D). The importance of active Aurora B in inhibiting cells from entering endoreduplication is obvious from the reduction of nocodazole-induced polyploidy in Mad2-haplosufficient HCT-116 cells overexpressing Aurora B (Figure 3B). Taken together, these initial observations suggest that in the context of Aurora B inhibition Rb promotes DNA endoreduplication and polyploidy after an aberrant mitosis.
We went on to investigate how this unique Rb function may be regulated. Rb phosphorylation status and function has been classically attributed to cyclin:cdk activity (Weinberg, 1995). Although both cdk 2 and cdk 4 activity did decrease with Aurora B inhibition, Rb hypophosphorylation preceded this inhibition of kinase activity (Figure 2Ai). We hypothesized that Aurora B itself was the endogenous kinase specifically responsible for regulating Rb in this context. In fact, we found that Aurora B directly phosphorylates the C terminus of Rb in vitro, and utilizing mass spectrometry, we identified serine 780 as an Aurora B phosphorylation site. We confirmed that phosphorylation at serine 780 also decreases in vivo with the same kinetics as the appearance of hypophosphorylated Rb in response to AZD1152 treatment (Figure 5C). Although introduction of an Rb wild-type construct increased AZD1152-induced polyploidy in Rb-negative SaOs2 cells, an Rb mutant with serine 780 changed to a phospho-mimetic aspartate (S780D) prevented Rb-mediated polyploidy in response to Aurora B inhibition (Figure 6A). This effect was further confirmed by the increased binding of the S780D to E2F1 and a decreased E2F1 promoter activity in S780D-overexpressed cells upon Aurora B inhibition (Figure 6, B and C). Thus, we conclude that serine 780 is an endogenous Aurora B phosphorylation site on Rb that impacts regulation of endoreduplication after exit from an aberrant mitosis.
These findings collectively suggest that the functional role of Rb in regulating the commitment to DNA replication differs depending upon the cellular context. It is well established that hypophosphorylated Rb is the functional form of the tumor suppressor that inhibits E2F transactivation, resulting in G1 arrest (Weinberg, 1995; Lam and La Thangue, 1994). However, we found that, although expression of wild-type Rb enhanced the G1 peak in asynchronous cells, in the context of Aurora B inhibition Rb actually promoted cell cycle progression with polyploidy induction (Figure 6). These results suggest that in the context of unperturbed cell cycles and normal mitoses, ectopic Rb expression delays G1 progression, but after aberrant mitoses in the setting of Aurora B inhibition, Rb plays a functionally distinct role by promoting endoreduplication. In fact, endoreduplication in the context of active Rb has been previously reported: cells expressing the phosphorylation resistant murine 
cdk Rb mutant (mutated at 10 of the 16 potential cdk phosphorylation site), which constitutively inhibits E2F transactivation, surprisingly undergo endoreduplication after a prolonged G1 arrest (Lukas et al., 1999). Additionally, Rb expression has been reported to be required for cell cycle progression as well. The proliferative advantage of the constitutively active H-RasV12 mutant is dependent on the presence of functional Rb; cells harboring the Ras mutant proliferate more slowly when Rb is down-regulated via RNAi (Williams et al., 2006). Although the mechanism(s) by which Rb may execute these functions has yet to be elucidated, one hypothesis that has been posed is that Rb-mediated suppression of other pocket protein family members' expression is required for cell cycle progression (Williams et al., 2006).
Although this question clearly is worthy of further investigation, our data does indicate that that when Rb is present, Aurora B phosphorylation leads to stabilization of Rb:E2F1 complexes and inhibition of endoreduplication, consistent with an Aurora B role in negatively regulating DNA synthesis after an aberrant mitosis with failed cytokinesis (Figure 6C). Though the addition of phosphates on Rb in the G1-S phase transition is generally associated with cell cycle progression, it has been previously demonstrated that the activation of cyclin D–associated kinase activity in the G0-G1 transition also leads to phosphorylation of Rb that stabilizes the Rb:E2F association, leading to repression of E2F target genes (Ezhevsky et al., 2001). Our data reaffirms that the functional significance of Rb phosphorylation at a specific phospho-acceptor site has to be interpreted in the cellular context in which it occurs. After failed mitosis with aberrant mitotic exit, persistent Aurora B activity in pseudo G1 acts as an intrinsic cellular checkpoint to prevent inappropriate DNA replication and polyploid cell formation by directly phosphorylating of Rb to repress expression of E2F1 target genes. This is a novel role for Aurora B outside of mitosis to prevent cellular progression toward aberrant cell cycles that may be relevant to other physiological scenarios of mitotic failure.
Lastly, our findings also hold therapeutic implications for the future development of Aurora B targeted therapy as tumors with differing Rb status would be predicted to have distinct responses. Preliminary work suggests that Rb negative tumor cells undergo increased apoptosis in the context of decreased polyploidy (data not shown). Such findings will be critical for eventually delineating susceptibility and resistance mechanism(s) of Aurora B targeted therapy that would need to be incorporated in future clinical testing.
|
Footnotes |
|---|
Address correspondence to: Gary K. Schwartz (schwartg{at}mskcc.org)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Andreassen, P. R., Lohez, O. D., Lacroix, F. B., and Margolis, R. L. (2001b). Tetraploid state induces p53-dependent arrest of nontransformed mammalian cells in G1. Mol. Biol. Cell 12, 1315–1328.
Andrews, P. D., Knatko, E., Moore, W. J., and Swedlow, J. R. (2003). Mitotic mechanics: the auroras come into view. Curr. Opin. Cell Biol 15, 672–683.[CrossRef][Medline]
Avni, D., Yang, H., Martelli, F., Hofmann, F., ElShamy, W. M., Ganesan, S., Scully, R., and Livingston, D. M. (2003). Active localization of the retinoblastoma protein in chromatin and its response to S phase DNA damage. Mol. Cell 12, 735–746.[CrossRef][Medline]
Borel, F., Lohez, O. D., Lacroix, F. B., and Margolis, R. L. (2002). Multiple centrosomes arise from tetraploidy checkpoint failure and mitotic centrosome clusters in p53 and RB pocket protein-compromised cells. Proc. Natl. Acad. Sci. USA 99, 9819–9824.
Carmena, M., and Earnshaw, W. C. (2003). The cellular geography of aurora kinases. Nat. Rev. Mol. Cell Biol 4, 842–854.[CrossRef][Medline]
Carvajal, R. D., Tse, A., and Schwartz, G. K. (2006). Aurora kinases: new targets for cancer therapy. Clin. Cancer Res 12, 6869–6875.
Chan, F. et al. (2007). Mechanism of action of the Aurora kinase inhibitor CCT129202 and in vivo quantification of biological activity. Mol. Cancer Ther 6, 3147–3157.
Crosio, C., Fimia, G. M., Loury, R., Kimura, M., Okano, Y., Zhou, H., Sen, S., Allis, C. D., and Sassone-Corsi, P. (2002). Mitotic phosphorylation of histone H 3, spatio-temporal regulation by mammalian Aurora kinases. Mol. Cell. Biol 22, 874–885.
Di Leonardo, A., Khan, S. H., Linke, S. P., Greco, V., Seidita, G., and Wahl, G. M. (1997). DNA rereplication in the presence of mitotic spindle inhibitors in human and mouse fibroblasts lacking either p53 or pRb function. Cancer Res 57, 1013–1019.
Ditchfield, C., Johnson, V. L., Tighe, A., Ellston, R., Haworth, C., Johnson, T., Mortlock, A., Keen, N., and Taylor, S. S. (2003). Aurora B couples chromosome alignment with anaphase by targeting BubR1, Mad2, and Cenp-E to kinetochores. J. Cell Biol 161, 267–280.
Ducat, D., and Zheng, Y. (2004). Aurora kinases in spindle assembly and chromosome segregation. Exp. Cell Res 301, 60–67.[CrossRef][Medline]
Ezhevsky, S. A., Ho, A., Becker-Hapak, M., Davis, P. K., and Dowdy, S. F. (2001). Differential regulation of retinoblastoma tumor suppressor protein by G(1) cyclin-dependent kinase complexes in vivo. Mol. Cell. Biol 21, 4773–4784.
Fraizer, G. C., Diaz, M. F., Lee, I. L., Grossman, H. B., and Sen, S. (2004). Aurora-A/STK15/BTAK enhances chromosomal instability in bladder cancer cells. Int. J. Oncol 25, 1631–1639.[Medline]
Giacinti, C., and Giordano, A. (2006). RB and cell cycle progression. Oncogene 25, 5220–5227.[CrossRef][Medline]
Giet, R., and Glover, D. M. (2001). Drosophila aurora B kinase is required for histone H3 phosphorylation and condensin recruitment during chromosome condensation and to organize the central spindle during cytokinesis. J. Cell Biol 152, 669–682.
Giet, R., Petretti, C., and Prigent, C. (2005). Aurora kinases, aneuploidy and cancer, a coincidence or a real link? Trends Cell Biol 15, 241–250.[CrossRef][Medline]
Gizatullin, F., Yao, Y., Kung, V., Harding, M. W., Loda, M., and Shapiro, G. I. (2006). The Aurora kinase inhibitor VX-680 induces endoreduplication and apoptosis preferentially in cells with compromised p53-dependent postmitotic checkpoint function. Cancer Res 66, 7668–7677.
Hannak, E., Kirkham, M., Hyman, A. A., and Oegema, K. (2001). Aurora-A kinase is required for centrosome maturation in Caenorhabditis elegans. J. Cell Biol 155, 1109–1116.
Harrington, E. A. et al. (2004). VX-680, a potent and selective small-molecule inhibitor of the Aurora kinases, suppresses tumor growth in vivo. Nat. Med 10, 262–267.[CrossRef][Medline]
Hauf, S., Cole, R. W., LaTerra, S., Zimmer, C., Schnapp, G., Walter, R., Heckel, A., van Meel, J., Rieder, C. L., and Peters, J. M. (2003). The small molecule Hesperadin reveals a role for Aurora B in correcting kinetochore-microtubule attachment and in maintaining the spindle assembly checkpoint. J. Cell Biol 161, 281–294.
Hsu, J. Y. et al. (2000). Mitotic phosphorylation of histone H3 is governed by Ipl1/aurora kinase and Glc7/PP1 phosphatase in budding yeast and nematodes. Cell 102, 279–291.[CrossRef][Medline]
Katayama, H., Ota, T., Jisaki, F., Ueda, Y., Tanaka, T., Odashima, S., Suzuki, F., Terada, Y., and Tatsuka, M. (1999). Mitotic kinase expression and colorectal cancer progression. J. Natl. Cancer Inst 91, 1160–1162.
Keen, N., and Taylor, S. (2004). Aurora-kinase inhibitors as anticancer agents. Nat. Rev. Cancer 4, 927–936.[CrossRef][Medline]
Kufer, T. A., Sillje, H. H., Korner, R., Gruss, O. J., Meraldi, P., and Nigg, E. A. (2002). Human TPX2 is required for targeting Aurora-A kinase to the spindle. J. Cell Biol 158, 617–623.
Lam, E. W., and La Thangue, N. B. (1994). DP and E2F proteins: coordinating transcription with cell cycle progression. Curr. Opin. Cell Biol 6, 859–866.[CrossRef][Medline]
Lampson, M. A., and Kapoor, T. M. (2005). The human mitotic checkpoint protein BubR1 regulates chromosome-spindle attachments. Nat. Cell Biol 7, 93–98.[CrossRef][Medline]
Lanni, J. S., and Jacks, T. (1998). Characterization of the p53-dependent postmitotic checkpoint following spindle disruption. Mol. Cell. Biol 18, 1055–1064.
Larsen, M. R., Thingholm, T. E., Jensen, O. N., Roepstorff, P., and Jorgensen, T. J. (2005). Highly selective enrichment of phosphorylated peptides from peptide mixtures using titanium dioxide microcolumns. Mol Cell Proteom 4, 873–886.
Lukas, J., Sorensen, C. S., Lukas, C., Santoni-Rugiu, E., and Bartek, J. (1999). p16INK4a, but not constitutively active pRb, can impose a sustained G1 arrest: molecular mechanisms and implications for oncogenesis. Oncogene 18, 3930–3935.[CrossRef][Medline]
Meraldi, P., Honda, R., and Nigg, E. A. (2004). Aurora kinases link chromosome segregation and cell division to cancer susceptibility. Curr. Opin. Genet. Dev 14, 29–36.[CrossRef][Medline]
Motwani, M., Delohery, T. M., and Schwartz, G. K. (1999). Sequential dependent enhancement of caspase activation and apoptosis by flavopiridol on paclitaxel-treated human gastric and breast cancer cells. Clin. Cancer Res 5, 1876–1883.
Motwani, M., Li, X., and Schwartz, G. K. (2000). Flavopiridol, a cyclin-dependent kinase inhibitor, prevents spindle inhibitor-induced endoreduplication in human cancer cells. Clin. Cancer Res 6, 924–932.
Nair, J. S., DaFonseca, C. J., Tjernberg, A., Sun, W., Darnell, J. E., Jr, Chait, B. T., and Zhang, J. J. (2002). Requirement of Ca2+ and CaMKII for Stat1 Ser-727 phosphorylation in response to IFN-gamma. Proc. Natl. Acad. Sci. USA 99, 5971–5976.
Niculescu, A. B., 3rd, Chen, X., Smeets, M., Hengst, L., Prives, C., and Reed, S. I. (1998). Effects of p21(Cip1/Waf1) at both the G1/S and the G2/M cell cycle transitions: pRb is a critical determinant in blocking DNA replication and in preventing endoreduplication. Mol. Cell. Biol 18, 629–643.
Nigg, E. A. (2001). Mitotic kinases as regulators of cell division and its checkpoints. Nat. Rev 2, 21–32.[CrossRef]
Rubin, S. M., Gall, A. L., Zheng, N., and Pavletich, N. P. (2005). Structure of the Rb C-terminal domain bound to E2F1-DP1, a mechanism for phosphorylation-induced E2F release. Cell 123, 1093–1106.[CrossRef][Medline]
Sakakura, C. et al. (2001). Tumour-amplified kinase BTAK is amplified and overexpressed in gastric cancers with possible involvement in aneuploid formation. Br. J. Cancer 84, 824–831.[CrossRef][Medline]
Sugiyama, K., Sugiura, K., Hara, T., Sugimoto, K., Shima, H., Honda, K., Furukawa, K., Yamashita, S., and Urano, T. (2002). Aurora-B associated protein phosphatases as negative regulators of kinase activation. Oncogene 21, 3103–3111.[CrossRef][Medline]
Tong, T. et al. (2004). Overexpression of Aurora-A contributes to malignant development of human esophageal squamous cell carcinoma. Clin. Cancer Res 10, 7304–7310.
Tse, A., and Schwartz, G. K. (2004). Potentiation of cytotoxicity of topoisomerase I poison by concurrent and sequential treatment with the checkpoint inhibitor UCN-01 involves disparate mechanisms resulting in either p53-independent clonogenic suppression or p53-dependent mitotic catastrophe. Cancer Res 64, 6635–6644.
Weinberg, R. A. (1995). The retinoblastoma protein and cell cycle control. Cell 81, 323–330.[CrossRef][Medline]
Wilkinson, R. W. et al. (2007). AZD1152, a selective inhibitor of Aurora B kinase, inhibits human tumor xenograft growth by inducing apoptosis. Clin. Cancer Res 13, 3682–3688.
Williams, J. P., Stewart, T., Li, B., Mulloy, R., Dimova, D., and Classon, M. (2006). The retinoblastoma protein is required for Ras-induced oncogenic transformation. Mol. Cell. Biol 26, 1170–1182.
Xu, C. F., Lu, Y., Ma, J., Mohammadi, M., and Neubert, T. A. (2005). Identification of phosphopeptides by MALDI Q-TOF MS in positive and negative ion modes after methyl esterification. Mol. Cell Proteom 4, 809–818.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
D) of Rb-C99. Only those constructs containing a 780 S