![[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. 18, Issue 4, 1233-1241, April 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*Institut Albert Bonniot, Centre de Recherche Institut National de la Santé et de la Recherche Médicale, Université Joseph Fourier U 823, Equipe DySAD, Université Joseph Fourier Site Santé, BP 170, F38042, Grenoble Cedex 09, France; and Equipe Phosphatase, Unité de Chimie Organique, Institut Pasteur, 75724 Paris Cedex, France
Submitted December 29, 2006;
Accepted January 8, 2007
Monitoring Editor: John Pringle
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The diversity in phosphatase functions comes from the ability of the highly conserved catalytic subunits to combine with various regulatory subunits and give rise to a large repertoire of hetero-multimeric holoenzymes (Virshup, 2000).
The protein phosphatase PP2A belongs to the serine/threonine PP2 phosphatase family that also contains PP2B (also named calcineurin) and PP2C (Lechward et al., 2001). PP2A is a hetero-trimeric complex composed of various isoforms of three subunits. The 36-kDa catalytic C subunit associates with the 65-kDa structural A subunit to form the core dimer. This dimer is able to interact with one of the several B-type regulatory proteins (B, B', and B''), which modulate the holoenzyme activity and substrate specificity (Price and Mumby, 2000; Strack et al., 2002). The A and C subunits are ubiquitously expressed in mammalian cells (Khew-Goodall and Hemmings, 1988; Hemmings et al., 1990; Mayer et al., 1990), whereas the expression of the different B subunit isoforms is tissue specific and can target the trimer to different subcellular compartments (Healy et al., 1991; Mayer et al., 1991; Zolnierowicz et al., 1994; Strack et al., 1999). Numerous functions have been described for the PP2A holoenzyme depending on the B regulatory subunit, including roles in morphogenesis, signal transduction, apoptosis, and cell-cycle progression and control (reviewed by Lechward et al., 2001). Indeed, PP2A is a key enzyme for the initiation and completion of mitosis. For example, the holoenzyme is required both for the G2/M transition (Lee, 1995; Minshull et al., 1996; Karaiskou et al., 1999) and during mitosis itself. In Xenopus eggs extracts, PP2A activity is required for bipolar spindle assembly and the maintenance of short microtubules during metaphase (Tournebize et al., 1997). PP2A has also been proposed to play a role in the metaphase-anaphase transition in both mitosis and meiosis (Chaudhuri et al., 1997; Mailhes et al., 2003). Finally, mass-spectroscopy–based proteomic analyses have revealed that PP2A is present in human interphase centrosomes (Andersen et al., 2003), which are crucial for cell division and cell-cycle progression (Rieder et al., 2001) because they contain key mitotic kinases such as Nek2 (Helps et al., 2000) and Aurora-A (Gopalan et al., 1997; Goepfert and Brinkley, 2000; Blagden and Glover, 2003; Dai et al., 2003).
The mammalian protein kinase Aurora-A is a member of a multigenic serine/threonine kinase family. The founding member of this family, Ipl1p, was identified in yeast through mutations leading to increased chromosome mis-segregation (Francisco and Chan, 1994), and in Drosophila, loss of Aurora function causes the formation of a monopolar spindle due to a centrosome-segregation defect (Glover et al., 1995). Similarly, loss of Aurora-A activity induces a monopolar spindle in Xenopus without blocking mitosis entry (Liu and Ruderman, 2006). In mammals, overexpression of Aurora-A was found in breast cancer (Sen et al., 1997), and its activity is increased in many cancers (breast, ovarian, colon, prostate, neuroblastoma, and cervical cancer cell lines) due to either gene mutation (Kallioniemi et al., 1994; Schlegel et al., 1995) or transcriptional deregulation. Aurora-A localizes to centrosomes, its expression is restricted to G2 and M phases (Gopalan et al., 1997), and its kinase activity is required for mitotic commitment (Marumoto et al., 2002; Hirota et al., 2003). Aurora-A interacts with the phosphatase PP1, which negatively regulates its activity by dephosphorylation (Katayama et al., 2001). Before mitosis entry, the Cdk1/Cyclin B1 complex inactivates PP1, leading indirectly to the activation of Aurora-A. The kinase in turn has a positive feedback on the activity of this complex, which promotes mitosis entry (Hirota et al., 2003).
In normal cells, Aurora-A activity is at least partially controlled by degradation through the APC-ubiquitin-proteasome pathway (Honda et al., 2000). Recently, Littlepage and Ruderman (2002) have proposed a degradation model in Xenopus. According to this model, Aurora-A degradation is inhibited by phosphorylation of serine 53, which would be sustained until the end of mitosis. Thus, Aurora-A degradation would depend on a phosphatase activity distinct from PP1. On the basis of the similarities in the localization and functions of PP2A and Aurora-A during mitosis, we hypothesized that the two proteins may interact at this stage, suggesting a possible role for PP2A in controlling the Aurora-A phosphorylation and/or degradation pathway. We present here data suggesting that the direct dephosphorylation of serine 51 (the human counterpart of serine 53 in Xenopus) by PP2A allows Aurora-A degradation at the end of mitosis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MEM) with 10% heat-inactivated bovine serum except where indicated. HeLa cells were grown in DMEM with 10% heat-inactivated bovine serum. For synchronization, exponentially growing cells were blocked in S phase by adding 5 mM thymidine. After 16 h, cells were washed three times in phosphate-buffered saline (PBS) and released in fresh medium for 3 h to allow complete progression through S and G2 phases. Cells were then incubated with 35 ng/ml nocodazole (Sigma-Aldrich, L'isle d'Abeau, France) for 3 h and then washed three times with PBS and reinoculated into fresh medium to allow commitment to mitosis. To compare the abilities of cell lines to proliferate, 2 x 104 cells of each line were seeded in six-well plates and counted every day for 6 d.
Flow Cytometry
Cell cycle analysis was performed using flow cytometry (FACS) of propidium iodide–stained cells. Transfected or control cells (roughly 2 x 106) were incubated in DMEM without serum and supplemented with 1% saponin for 30 min. After a short centrifugation, the cells were resuspended in DMEM containing 1% fetal calf serum, 0.1% saponin, and 50 µg/ml propidium iodide (Sigma-Aldrich). The red fluorescence was analyzed using a FACScan flow cytometer and the CellQuest analysis software (Becton-Dickinson, Le Pont de Claix, France). The percentages of G0/G1 and G2/M cells were calculated relative to total in-cycle cells, and SDs were obtained from three independent experiments.
Immunofluorescence
Synchronized cells were grown on coverslips coated with 0.2 mg/ml poly-D-lysine (Sigma-Aldrich), washed twice in TBS (20 mM Tris, 137 mM NaCl, pH 7.6), and fixed in TBS/4% formaldehyde for 4 min at room temperature. After three washes with TBS, the cells were fixed again by incubation in methanol for 5 min at –20°C and washed three times in TBS. Nonspecific binding was blocked by a 45-min incubation in filtered TBS supplemented with 0.1% Tween 20 and 3% BSA. The coverslips were then incubated overnight at 4°C with mixed primary antibodies diluted in 0.1% Tween-TBS, washed three times in 0.1% Tween-TBS, and incubated for 3 h with mixed secondary antibodies diluted in 0.1% Tween-TBS. Finally, the coverslips were washed three times in 0.1% Tween-TBS, once in TBS, and then mounted in Mowiol 4-88 mounting medium (Roth Sochiel, Lauterbourg, France) containing 10 µg/ml 4'6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich). All primary antibodies were used at a dilution of 1:500. The primary mAb IAK1 (anti-Aurora-A) was from BD Biosciences (le Pont de Claix, France; Cat. No. 610939); rabbit polyclonal anti-Aurora-A was from Calbiochem (VWR International, Fontenay-sous-Bois, France; Cat. No. PC742-100); the mouse monoclonal (clone 1D6), rabbit polyclonal anti-PP2A-C were from Upstate (Chemicon International, Chandlers Ford, UK; Cat. Nos. 05-421 and 06-222); and the rabbit polyclonal anti-
-tubulin was from Santa Cruz Biotechnology (Santa Cruz, CA; Cat. No. SC10732). All secondary antibodies were purchased from Molecular Probes (Eugene, OR) and used at a dilution of 1:1500. They were labeled either with Alexa Fluor 546 or Alexa Fluor 488. Microscopic images were obtained using a Zeiss LSM510 confocal microscope (Thornwood, NY) with a Plan-neofluar 40x ph3 oil objective, NA 1.3.
Vector Constructions, Transfection, and RNA Interference Experiments
A full-length human Aurora-A cDNA in the pEF6/V5-His TOPO TA plasmid (Clontech-Takara Bio Europe, Saint-Germain-en-Laye, France) was provided by Dr. F. Hans (Grenoble, France). For site-directed mutagenesis, full-length hAurA was amplified by PCR and cloned into the pEGFP-N1 vector (Invitrogen, Cergy Pontoise, France). Mutations were then generated in the Aurora-A-GFP fusion protein using the QuickChange Site-Directed Mutagenesis Kit (Stratagene Europe, Amsterdam, The Netherlands) with the primers indicated in Table 1.
|
), as shown in Table 1 along with the control mismatch primers. Because no PP2A-C RNAi experiments have been described previously in mammalian cells, we designed a sequence likely to knock down both the and 
isoforms of PP2A-C by aligning the PP2A-C and PP2A-C coding sequences with the RNAi oligonucleotide sequence previously described for Drosophila. We chose a sequence that matches both PP2A-C isoforms and designed the primers and control mismatch primers described in Table 1. Electroporations were performed following Protocol T820 provided for the BTX ElectroSquarePorator (Harvard Apparatus SARL, Les Ulis Cedex, France). HeLa cells were harvested using a mixture of trypsin and 0.5 mM EDTA, washed twice in PBS, and resuspended in PBS at 5 x 106 cells/ml. DNA, 20 µg, was added to a 400-µl sample in a BTX Disposable Cuvette Model 640 with a 4-mm gap. Electroporation was performed using square pulses of 150 V and 99-ms duration. Two pulses were performed. Electroporated cells were immediately transferred to a 24-well plate and incubated for 24 h at 37°C in fresh DMEM supplemented with 10% serum. Stably silenced cell lines [AUR(RNAi) and PP2A(RNAi)] and control lines [AUR(Mm.RNAi) and PP2A(Mm.RNAi)] were selected using Zeocin (Invitrogen) at a final concentration of 1 mg/ml. For transient transfections, HeLa cells in 6-well plates were transfected using ExGen 500 transfection reagent (Euromedex, Souffelweyersheim, France) following the manufacturer's instructions.
For transient RNAi experiments, chemically synthesized RNAs were purchased from Eurogentec (Angers, France) in the desalted, preannealed duplex form. The interfering sequences were identical to those used in the pSuper experiments (see above). One day before transfection, cells were plated in six-well plates and grown at 37°C to 50% confluence in DMEM with 10% serum and without antibiotics. Oligofectamine (Invitrogen) was used for transfections following the manufacturer's instructions. Briefly, a transfection mixture (200 µl) containing siRNA (200 pmol) and Oligofectamine (4 µl) in serum-reduced Opti-MEM (Invitrogen) was added to each well. Each well was subsequently filled to 1 ml with serum-free DMEM. Cells were incubated in this transfection mixture for 4 h and then cultured further in 2 ml of DMEM supplemented with 10% serum. Two rounds of transfection were performed before synchronization and Western blot analysis.
Okadaic Acid Treatment
Okadaic acid (Sigma-Aldrich) was dissolved in DMSO to a final concentration of 1 µM and added to synchronized cells to a final concentration of 1 nM for at least 15 min but never longer than 2 h. After incubation, the cells were washed three times in PBS and prepared for Western blot analysis or immunofluorescence.
Western Blot Analyses
Synchronized cells were lysed in the presence of protease inhibitors either in 25 mM HEPES, pH 7.2, 150 mM NaCl, 5 mM MgCl2, 1% Triton X-100, or in 25 mM HEPES, pH 7.2, 150 mM NaCl, 5 mM MgCl2, 1% SDS, 1% NP40. Protein quantification was performed using the microBCA method (PerBio Science, Brebières, France). Laemmli denaturing sample buffer (25 µl) was added, and a final volume containing 100 µg of protein was heated for 15 min at 90°C. Proteins were loaded on a 10% SDS gel, run for 1 h at 100 V, and transferred for 1 h at 100 V to a PVDF membrane. Membranes were blocked for 45 min in TBS containing 5% fat free dried milk, 0.1% Tween 20 and then incubated overnight at 4°C in TBS/0.1% Tween 20 containing monoclonal IAK1 antibody diluted 1:250, monoclonal AC-40 anti-actin antibody (Sigma Aldrich; Cat. No. A3853) diluted 1:100, or polyclonal or monoclonal anti-PP2A-C antibodies (see above) diluted 1:1000. Signals were detected using an horseradish peroxide–conjugated goat anti-mouse IgG or goat anti-rabbit IgG antibody and the ECL chemiluminescence procedure (Amersham Biosciences, Piscataway, NJ). Western blot quantifications were performed using ImageMaser VDS-CL and the 1D gel-analysis procedure of the ImageQuant software (Amersham Biosciences) and normalized against actin. The standard deviations were obtained from three to five independent experiments.
Immunoprecipitation and Peptide Competition
HeLa cells were lysed in TBS supplemented either with 1% Triton X-100 or with 1% NP40 and 1% SDS. Protein-G–coupled beads were incubated for 2 h with 3% BSA in PBS at 4°C. A volume of total lysate containing 500 µg of protein was precleared by incubation with 100 µl of protein G/BSA beads for 2 h and then with a nonrelevant rabbit immunoglobulin G (Jackson ImmunoResearch Laboratories Europe, Newmarket, United Kingdom; Cat. No. 005-000-003) for 1 h. The mixture was centrifuged for 2 min at 1600 rpm after each step. The cleared supernatant was then incubated for 2 h with either 8 µg of the polyclonal anti-Aurora-A antibody (see above) or 4 µg of the polyclonal anti-PP2A-C antibody. Protein G–coupled beads, 100 µl, blocked with BSA (see above) were then added, and the mixture was rocked gently overnight at 4°C and then centrifuged for 2 min at 1600 rpm. The immunoprecipitated complex was then washed three times in the lysis buffer. For Western blot analysis, samples of the initial total lysate and of the immunoprecipitation supernatant, each containing 100 µg of total protein, were heated at 90°C in Laemmli sample buffer and loaded onto the gel. The nonspecific precipitate and specific immunoprecipitate were each resuspended in Laemmli sample buffer to a final volume of 20 µl, heated to 90°C, and loaded on the gel.
For the peptide-competition experiment, two peptides were used: a scrambled-sequence peptide (H2N-NSQRRDLSVCP-COOH) and a specific hAURA-D51 peptide (H2N-RVLCPDNSSQR-COOH), which mimicked the phosphorylated form of the Ser-51–containing peptide. The HeLa cell lysate was precleared as described above and split into three fractions containing 500 µg of total protein apiece. The first fraction was incubated only with the PP2A-C polyclonal antibody, whereas the second and third fractions were incubated for 6 h at 4°C with 100 µM of either scrambled-sequence peptide or hAURA-D51 peptide, respectively, before immunoprecipitation of PP2A-C as described above.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-tubulin in metaphase and anaphase cells (Figure 1B), but it occupied a somewhat more restricted area than Aurora A and a somewhat broader area than tubulin. The latter results show that PP2A-C is not strictly restricted to centrosomes at the cell poles.
|
|
proposed that Aurora-A degradation depends on dephosphorylation of serine 53 in the A box of the Xenopus protein, suggesting that PP2A-C may interact with a region that encompasses the phospho-serine in this position (serine 51 of the human protein). To test this hypothesis, we performed a competition experiment using an 11-amino acid peptide, hAURAD51, that corresponded to the A box region but with Asp instead of Ser at position 51 in order to mimic the phospho-peptide. A control peptide contained the same amino acids but in a scrambled sequence. Immunoprecipitation of PP2A-C was performed as before but in the presence of these peptides in order to characterize their ability to compete with the endogenous Aurora-A and prevent its association with PP2A-C. As expected, Aurora A was coimmunoprecipitated with PP2A-C in the absence of added peptide or in the presence of the scrambled peptide (Figure 2B). However, peptide hAURAD51 fully inhibited the coimmunoprecipitation (Figure 2B), indicating that the peptide competes efficiently with endogenous Aurora A for interaction with the phosphatase. Thus, the interaction of Aurora A and PP2A-C appears to be direct and requires a region that encompasses the A box. To determine whether the interaction between PP2A-C and Aurora-A indeed depends on the phosphorylation of serine 51, we generated constructs expressing an Aurora-A-GFP fusion protein and mutant forms of this protein in which S51 was substituted either by aspartic acid (AurAS51D- GFP) or by alanine (AurAS51A-GFP). These constructs were transiently transfected into HeLa cells, and protein expression was monitored by fluorescence microscopy (not shown) and by Western blotting (Figure 2C, left). The Western blots showed that AurA-GFP and AurAS51A-GFP were expressed at comparable levels, that AurAS51D-GFP was expressed at a somewhat higher level, and that the expression of endogenous Aurora-A and PP2A-C appeared to be unaffected. Coimmunoprecipitation experiments using PP2A-C–specific antibody revealed that AurA-GFP and endogenous Aurora- A were both efficiently coimmunoprecipitated, that the S51A mutation weakened the interaction of AurA-GFP with the phosphatase, and, conversely, that the S51D mutation strengthened the interaction sufficiently that the binding of the endogenous Aurora-A to PP2A-C was efficiently competed (Figure 2C, right). Taken together, our data indicate that Aurora-A and PP2A-C interact directly and that phosphorylation of serine 51 of the A box promotes this interaction.
PP2A-C Triggers Aurora-A Degradation
Constructs expressing RNAi sequences directed against Aurora-A or PP2A-C, or control mismatch sequences, were generated as described in Material and Methods, and stable cell lines were selected after transfection. Quantitative Western blot analysis of these cell lines (Figure 3A) revealed that 
75% of Aurora-A was depleted in AUR(RNAi) cells and 60% of PP2A-C was depleted in PP2A(RNAi) cells, whereas the mismatch constructs had no effect, confirming the specificity of the RNAi. Consistent with the hypothesized role of PP2A in Aurora-A degradation, expression of the kinase was increased by 30% upon depletion of PP2A-C (Figure 3A). Interestingly, PP2A-C expression was also slightly diminished upon Aurora-A knockdown. We also performed parallel experiments using transient transfection with siRNA constructs (see Materials and Methods), obtaining identical results (Supplementary Figure S1).
|
48 h for HeLa cells and the two mismatch-construct lines and 96 h for the AUR(RNAi) and PP2A(RNAi) lines. Thus, cell proliferation is dramatically slowed by knockdown of these enzymes. However, flow cytometry analyses revealed similar cell cycle distributions (30% G2/M cells and 60% G0/G1 cells) in each cell population (Figure 3C). Thus, the slower proliferation observed upon PP2A-C or Aurora-A knockdown appears to be due to a general cell cycle delay rather than to an inability of the cells to progress through a particular phase of the cell cycle, in good agreement with other recent data (Liu and Ruderman, 2006). Thus, the increased level of Aurora-A in PP2A(RNAi) cells probably results from impaired degradation due to the knockdown of phosphatase activity.
Aurora-A/PP2A-C Interaction Is Required for Localization of Both Enzymes to the Cell Poles
To gain insight into the functional significance of the interaction between Aurora-A and PP2A-C, we analyzed the subcellular localizations of these proteins in AUR(RNAi) and PP2A(RNAi) cells. On Aurora-A knockdown, PP2A-C no longer accumulated at the cell poles during prometaphase, metaphase, and anaphase (Figure 4A). Centrosomes were still present and could be observed by -tubulin staining (Figure 4A), but they formed monopolar spindles typical of Aurora-A inhibition or depletion, as observed also by others (Liu and Ruderman, 2006). Thus, the recruitment of PP2A-C to the cell poles of normal mitotic HeLa cells depends on Aurora-A. Reciprocally, Aurora-A was not only stabilized as expected in PP2A(RNAi) cells, but it was also delocalized throughout the cells (Figure 4B), although centrosomes were present at both cell poles, as observed by -tubulin staining.
|
). In these experiments, synchronized cells were treated with 1 nM OA after release from nocodazole arrest in order to avoid any blockage of mitosis entry due to inhibition of protein phosphatase 1, and general cytotoxicity effects were avoided by restricting treatments to 75 min. Samples collected at different times after the release from arrest (time 0) were analyzed by quantitative Western blotting (Figure 5A, top, and B). Within 15 min, the total amount of Aurora-A increased by 20% in treated cells; it then remained stable from 15 to 75 min, whereas it decreased in control cells. Stable knockdown of PP2A-C by RNAi had a similar effect (Figure 5A, bottom, and B), suggesting strongly that the OA effect was due mainly to inhibition of PP2A.
|
; Anand et al., 2003).
Littlepage and Ruderman (2002) have shown that either substitution of serine 53 of the A box by a residue mimicking phosphorylation or a single point mutation in the degradation box (D box: RxxLxxVxE) is sufficient to stabilize Xenopus Aurora-A in vitro. The former mutation presumably prevents the unmasking of the D box by dephosphorylation of the A Box. As the R residue of the D box is involved in recognition by the APC/Cdh1 complex, its alteration also prevents degradation. It is likely that the mechanism observed in vitro in Xenopus is general and also takes place in mammalian cells in vivo. To address this question, we transiently transfected CHO cells with constructs expressing either wild-type Aurora-A-GFP or one of two mutant versions, AurA(S51D)-GFP and AurA(R371D)-GFP, and compared the susceptibilities of these proteins to degradation. Consistent with the results obtained in the Xenopus studies, both Aurora-A mutants were present at much higher levels than the wild-type protein throughout the cell cycle in the CHO cells, as judged by the GFP fluorescence intensity (Figure 6, A–C). Because the efficiencies of transfection were similar in all cases (35%), we could also compare the protein levels by Western blotting, which yielded similar results (Figure 6D).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). In particular, Aurora-A is activated at the beginning of mitosis by the LIM protein Ajuba, which promotes Aurora-A autophosphorylation (Hirota et al., 2003). Before this activation stage, Aurora-A is maintained in an inactive, unphosphorylated state by the protein phosphatase PP1 (Katayama et al., 2001). Recent work in Xenopus oocytes also emphasizes that Aurora-A activation depends on kinase rather than phosphatase activities (Maton et al., 2005). In this report, we have shown by immunofluorescence microscopy that Aurora-A and the catalytic subunit of PP2A are colocalized at the cell poles during mitosis and by immunoprecipitation that the two proteins interact within the same complex. These results are consistent with a recent proteomic analysis of interphase human-centrosome components, which reported the presence of PP2A in this organelle (Andersen et al., 2003). Peptide-competition experiments strongly suggest that the interaction between PP2A and Aurora-A is direct and involves at least a small peptide sequence around phosphoserine 51 in the A degradation box. This conclusion is supported by the observation that the Aurora-A S51D point mutation, which mimics phosphorylation, increases PP2A-C affinity for the kinase. RNAi knockdown of either Aurora-A or PP2A-C indicated that the interaction between these proteins is required to target both proteins to the cell poles. Dephosphorylation of Aurora-A serine 51 by PP2A also appears to be required for the degradation of Aurora-A at the end of mitosis, as shown by the stabilization of Aurora-A by OA treatment at a concentration that preferentially inactivates PP2A, RNAi silencing of the two PP2A-C subunits in HeLa cells, or the S51D mutation. Taken together, our results support the hypothesis that the interaction between Aurora-A and PP2A is physiologically significant, and is involved in the degradation of the kinase at the end of mitosis.
Silencing either Aurora-A or PP2A resulted in a decrease in cell proliferation. Because both Aurora-A and PP2A are involved in mitosis entry, metaphase alignment, and cytokinesis, it is not easy to ascribe these defects specifically to one of these enzymes. However, our results may provide a molecular basis to explain the similar phenotypes described in the literature whenever either Aurora-A or PP2A-C expression is deregulated (Goepfert et al., 2002; Chen et al., 2004; Meraldi et al., 2004). If both proteins are segregated within the same complex, overexpression of one of the partners should result in the delocalization of that protein. PP2A inhibition by either OA treatment or RNAi stabilized but also delocalized the kinase throughout the cell. Thus, phosphorylation of serine 51 within the A box may maintain a structural conformation not only unfavorable for recognition by the proteasome complex but also essential for kinase localization. It is possible that the centrosomal localization of Aurora-A may be required for the degradation process by allowing the interaction of the kinase with some mediator(s) of proteolysis.
Finally, although it is well established that Aurora-A kinase is a crucial regulator of mitosis, the role of PP2A has remained unclear. Our data clearly show that PP2A is a key regulator of Aurora-A degradation and generalize the mechanism suggested by the pioneering studies of Ruderman's group in Xenopus.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
The online version of this article contains supplemental material at MBC Online (http://www.molbiolcell.org). 
Address correspondence to: Marc R. Block (marc.block{at}ujf-grenoble.fr)
Abbreviations used: MEM, Eagle's medium with alpha modification; DAPI, 4',6-diamidino-2-phenylindole; FACS, fluorescence-activated cell sorter; PP2A, protein phosphatase 2A; RNAi, RNA interference; siRNA, small interfering RNA.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Andersen, J. S., Wilkinson, C. J., Mayor, T., Mortensen, P., Nigg, E. A., Mann, M. (2003). Proteomic characterization of the human centrosome by protein correlation profiling. Nature 426, 570–574.[CrossRef][Medline]
Bialojan, C. and Takai, A. (1988). Inhibitory effect of a marine-sponge toxin, okadaic acid, on protein phosphatases. Specificity and kinetics. Biochem. J 256, 283–290.
Bischoff, J. R. and Plowman, G. D. (1999). The Aurora/Ipl1p kinase family: regulators of chromosome segregation and cytokinesis. Trends Cell Biol 9, 454–459.[CrossRef][Medline]
Blagden, S. P. and Glover, D. M. (2003). Polar expeditions—provisioning the centrosome for mitosis. Nat. Cell Biol 5, 505–511.[CrossRef][Medline]
Chaudhuri, S. K., Ghosh, S., Paweletz, N., Schroeter, D. (1997). Effects of low concentrations of okadaic acid in HeLa cells. Indian J. Exp. Biol 35, 1044–1054.[Medline]
Chen, W., Possemato, R., Campbell, K. T., Plattner, C. A., Pallas, D. C., Hahn, W. C. (2004). Identification of specific PP2A complexes involved in human cell transformation. Cancer Cell 5, 127–136.[CrossRef][Medline]
Dai, W., Huang, X., Ruan, Q. (2003). Polo-like kinases in cell cycle checkpoint control. Front. Biosci 8, d1128–1133.[Medline]
Eyers, P. A. and Maller, J. L. (2003). Regulating the regulators: Aurora A activation and mitosis. Cell Cycle 2, 287–289.[Medline]
Francisco, L. and Chan, C. S. (1994). Regulation of yeast chromosome segregation by Ipl1 protein kinase and type 1 protein phosphatase. Cell. Mol. Biol. Res 40, 207–213.[Medline]
Glover, D. M., Leibowitz, M. H., McLean, D. A., Parry, H. (1995). Mutations in aurora prevent centrosome separation leading to the formation of monopolar spindles. Cell 81, 95–105.[CrossRef][Medline]
Goepfert, T. M., Adigun, Y. E., Zhong, L., Gay, J., Medina, D., Brinkley, W. R. (2002). Centrosome amplification and overexpression of aurora A are early events in rat mammary carcinogenesis. Cancer Res 62, 4115–4122.
Goepfert, T. M. and Brinkley, B. R. (2000). The centrosome-associated Aurora/Ipl-like kinase family. Curr. Top. Dev. Biol 49, 331–342.[Medline]
Gopalan, G., Chan, C. S., Donovan, P. J. (1997). A novel mammalian, mitotic spindle-associated kinase is related to yeast and fly chromosome segregation regulators. J. Cell Biol 138, 643–656.
Healy, A. M., Zolnierowicz, S., Stapleton, A. E., Goebl, M., DePaoli-Roach, A. A., Pringle, J. R. (1991). CDC55, a Saccharomyces cerevisiae gene involved in cellular morphogenesis: identification, characterization, and homology to the B subunit of mammalian type 2A protein phosphatase. Mol. Cell. Biol 11, 5767–5780.
Helps, N. R., Luo, X., Barker, H. M., Cohen, P. T. (2000). NIMA-related kinase 2 (Nek2), a cell-cycle-regulated protein kinase localized to centrosomes, is complexed to protein phosphatase 1. Biochem. J 349, 509–518.[CrossRef][Medline]
Hemmings, B. A., Adams-Pearson, C., Maurer, F., Muller, P., Goris, J., Merlevede, W., Hofsteenge, J., Stone, S. R. (1990). - and -forms of the 65-kDa subunit of protein phosphatase 2A have a similar 39 amino acid repeating structure. Biochemistry 29, 3166–3173.[CrossRef][Medline]
Hirota, T., Kunitoku, N., Sasayama, T., Marumoto, T., Zhang, D., Nitta, M., Hatakeyama, K., Saya, H. (2003). Aurora-A and an interacting activator, the LIM protein Ajuba, are required for mitotic commitment in human cells. Cell 114, 585–598.[CrossRef][Medline]
Honda, K., Mihara, H., Kato, Y., Yamaguchi, A., Tanaka, H., Yasuda, H., Furukawa, K., Urano, T. (2000). Degradation of human Aurora2 protein kinase by the anaphase-promoting complex-ubiquitin-proteasome pathway. Oncogene 19, 2812–2819.[CrossRef][Medline]
Kallioniemi, A., Kallioniemi, O. P., Piper, J., Tanner, M., Stokke, T., Chen, L., Smith, H. S., Pinkel, D., Gray, J. W., Waldman, F. M. (1994). Detection and mapping of amplified DNA sequences in breast cancer by comparative genomic hybridization. Proc. Natl. Acad. Sci. USA 91, 2156–2160.
Karaiskou, A., Jessus, C., Brassac, T., Ozon, R. (1999). Phosphatase 2A and polo kinase, two antagonistic regulators of cdc25 activation and MPF auto-amplification. J. Cell Sci 112, 3747–3756.[Abstract]
Katayama, H., Zhou, H., Li, Q., Tatsuka, M., Sen, S. (2001). Interaction and feedback regulation between STK15/BTAK/Aurora-A kinase and protein phosphatase 1 through mitotic cell division cycle. J. Biol. Chem 276, 46219–46224.
Khew-Goodall, Y. and Hemmings, B. A. (1988). Tissue-specific expression of mRNAs encoding - and -catalytic subunits of protein phosphatase 2A. FEBS Lett 238, 265–268.[CrossRef][Medline]
Lechward, K., Awotunde, O. S., Swiatek, W., Muszynska, G. (2001). Protein phosphatase 2A: variety of forms and diversity of functions. Acta Biochim. Pol 48, 921–933.[Medline]
Lee, T. H. (1995). The role of protein phosphatase type-2A in the Xenopus cell cycle: initiation of the G2/M transition. Semin. Cancer Biol 6, 203–209.[CrossRef][Medline]
Littlepage, L. E. and Ruderman, J. V. (2002). Identification of a new APC/C recognition domain, the A box, which is required for the Cdh1-dependent destruction of the kinase Aurora-A during mitotic exit. Genes Dev 16, 2274–2285.
Liu, Q. and Ruderman, J. V. (2006). Aurora A, mitotic entry, and spindle bipolarity. Proc. Natl. Acad. Sci. USA 103, 5811–5816.
Mailhes, J. B., Hilliard, C., Fuseler, J. W., London, S. N. (2003). Okadaic acid, an inhibitor of protein phosphatase 1 and 2A, induces premature separation of sister chromatids during meiosis I and aneuploidy in mouse oocytes in vitro. Chromosome Res 11, 619–631.[CrossRef][Medline]
Marumoto, T., et al. (2002). Roles of aurora-A kinase in mitotic entry and G2 checkpoint in mammalian cells. Genes Cells 7, 1173–1182.[Abstract]
Maton, G., Lorca, T., Girault, J. A., Ozon, R., Jessus, C. (2005). Differential regulation of Cdc2 and Aurora-A in Xenopus oocytes: a crucial role of phosphatase 2A. J. Cell Sci 118, 2485–2494.
Mayer, R. E., Hendrix, P., Cron, P., Matthies, R., Stone, S. R., Goris, J., Merlevede, W., Hofsteenge, J., Hemmings, B. A. (1991). Structure of the 55-kDa regulatory subunit of protein phosphatase 2A: evidence for a neuronal-specific isoform. Biochemistry 30, 3589–3597.[CrossRef][Medline]
Mayer, R. E., Khew-Goodall, Y., Stone, S. R., Hemmings, B. A. (1990). Expression and organization of protein phosphatase 2A catalytic subunit genes. Adv. Second Messenger Phosphoprotein Res 24, 236–241.[Medline]
Meraldi, P., Honda, R., Nigg, E. A. (2004). Aurora kinases link chromosome segregation and cell division to cancer susceptibility. Curr. Opin. Genet. Dev 14, 29–36.[CrossRef][Medline]
Minshull, J., Straight, A., Rudner, A. D., Dernburg, A. F., Belmont, A., Murray, A. W. (1996). Protein phosphatase 2A regulates MPF activity and sister chromatid cohesion in budding yeast. Curr. Biol 6, 1609–1620.[CrossRef][Medline]
Price, N. E. and Mumby, M. C. (2000). Effects of regulatory subunits on the kinetics of protein phosphatase 2A. Biochemistry 39, 11312–11318.[CrossRef][Medline]
Rieder, C. L., Faruki, S., Khodjakov, A. (2001). The centrosome in vertebrates: more than a microtubule-organizing center. Trends Cell Biol 11, 413–419.[CrossRef][Medline]
Schlegel, J., Stumm, G., Scherthan, H., Bocker, T., Zirngibl, H., Ruschoff, J., Hofstadter, F. (1995). Comparative genomic in situ hybridization of colon carcinomas with replication error. Cancer Res 55, 6002–6005.
Sen, S., Zhou, H., White, R. A. (1997). A putative serine/threonine kinase encoding gene BTAK on chromosome 20q13 is amplified and overexpressed in human breast cancer cell lines. Oncogene 14, 2195–2200.[CrossRef][Medline]
Strack, S., Chang, D., Zaucha, J. A., Colbran, R. J., Wadzinski, B. E. (1999). Cloning and characterization of B
, a novel regulatory subunit of protein phosphatase 2A. FEBS Lett 460, 462–466.[CrossRef][Medline]
Strack, S., Ruediger, R., Walter, G., Dagda, R. K., Barwacz, C. A., Cribbs, J. T. (2002). Protein phosphatase 2A holoenzyme assembly: identification of contacts between B-family regulatory and scaffolding A subunits. J. Biol. Chem 277, 20750–20755.
Tournebize, R., Andersen, S. S., Verde, F., Doree, M., Karsenti, E., Hyman, A. A. (1997). Distinct roles of PP1 and PP2A-like phosphatases in control of microtubule dynamics during mitosis. EMBO J 16, 5537–5549.[CrossRef][Medline]
Virshup, D. M. (2000). Protein phosphatase 2A: a panoply of enzymes. Cur. Opin. Cell. Biol 12, 180–185.[CrossRef][Medline]
Zolnierowicz, S. and Bollen, M. (2000). Protein phosphorylation and protein phosphatases. De Panne, Belgium, September 19–24, 1999. EMBO J 19, 483–488.[CrossRef][Medline]
Zolnierowicz, S., Csortos, C., Bondor, J., Verin, A., Mumby, M. C., DePaoli-Roach, A. A. (1994). Diversity in the regulatory B-subunits of protein phosphatase 2A: identification of a novel isoform highly expressed in brain. Biochemistry 33, 11858–11867.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
