GSK-3β Regulates Proper Mitotic Spindle Formation in Cooperation with a Component of the γ-Tubulin Ring Complex, GCP5
2008; Elsevier BV; Volume: 283; Issue: 19 Linguagem: Inglês
10.1074/jbc.m710282200
ISSN1083-351X
AutoresNanae Izumi, Katsumi Fumoto, Shunsuke Izumi, Akira Kikuchi,
Tópico(s)Wnt/β-catenin signaling in development and cancer
ResumoGlycogen synthase kinase-3β (GSK-3β) is known to play a role in the regulation of the dynamics of microtubule networks in cells. Here we show the role of GSK-3β in the proper formation of the mitotic spindles through an interaction with GCP5, a component of the γ-tubulin ring complex (γTuRC). GCP5 bound directly to GSK-3β in vitro, and their interaction was also observed in intact cells at endogenous levels. Depletion of GCP5 dramatically reduced the GCP2 and γ-tubulin in the γTuRC fraction of sucrose density gradients and disrupted γ-tubulin localization to the spindle poles in mitotic cells. GCP5 appears to be required for the formation or stability of γTuRC and the recruitment of γ-tubulin to the spindle poles. A GSK-3 inhibitor not only led to the accumulation of γ-tubulin and GCP5 at the spindle poles but also enhanced microtubule nucleation activity at the spindle poles. Depletion of GCP5 rescued this disrupted organization of spindle poles observed in cells treated with the GSK-3 inhibitor. Furthermore, the inhibition of GSK-3 enhanced the binding of γTuRC to the centrosome isolated from mitotic cells in vitro. Our findings suggest that GSK-3β regulates the localization of γTuRC, including GCP5, to the spindle poles, thereby controlling the formation of proper mitotic spindles. Glycogen synthase kinase-3β (GSK-3β) is known to play a role in the regulation of the dynamics of microtubule networks in cells. Here we show the role of GSK-3β in the proper formation of the mitotic spindles through an interaction with GCP5, a component of the γ-tubulin ring complex (γTuRC). GCP5 bound directly to GSK-3β in vitro, and their interaction was also observed in intact cells at endogenous levels. Depletion of GCP5 dramatically reduced the GCP2 and γ-tubulin in the γTuRC fraction of sucrose density gradients and disrupted γ-tubulin localization to the spindle poles in mitotic cells. GCP5 appears to be required for the formation or stability of γTuRC and the recruitment of γ-tubulin to the spindle poles. A GSK-3 inhibitor not only led to the accumulation of γ-tubulin and GCP5 at the spindle poles but also enhanced microtubule nucleation activity at the spindle poles. Depletion of GCP5 rescued this disrupted organization of spindle poles observed in cells treated with the GSK-3 inhibitor. Furthermore, the inhibition of GSK-3 enhanced the binding of γTuRC to the centrosome isolated from mitotic cells in vitro. Our findings suggest that GSK-3β regulates the localization of γTuRC, including GCP5, to the spindle poles, thereby controlling the formation of proper mitotic spindles. Serine/threonine kinase glycogen synthase kinase-3 (GSK-3) was first described in a glycogen metabolic pathway (1Plyte S.E. Hughes K. Nikolakaki E. Pulverer B.J. Woodgett J.R. Biochim. Biophys. Acta. 1992; 1114: 147-162Crossref PubMed Scopus (342) Google Scholar). There are two GSK-3 isoforms, GSK-3α and GSK-3β, in mammalian cells, and both GSK-3 proteins regulate various physiological responses by phosphorylating many substrates, including protein synthesis, gene expression, subcellular localization of proteins, and protein degradation (2Cohen P. Frame S. Nat. Rev. Mol. Cell. Biol. 2001; 2: 769-776Crossref PubMed Scopus (1324) Google Scholar, 3Doble B.W. Woodgett J.R. J. Cell Sci. 2003; 116: 1175-1186Crossref PubMed Scopus (1793) Google Scholar). GSK-3 has been highly conserved during evolution and plays a fundamental role in cellular responses. For example, there are four genes, MCK1 (meiosis and centromere regulatory kinase-1), MDS1/RIM11, MRK1, and YOL128c, that encode homologs of mammalian GSK-3 in Saccharomyces cerevisiae. Mck1 stabilizes Rog1 (revertant of glycogen synthase kinase mutation protein 1) (4Andoh T. Hirata Y. Kikuchi A. Mol. Cell. Biol. 2000; 20: 6712-6720Crossref PubMed Scopus (52) Google Scholar) and stimulates gene expression by Msn2 (multicopy suppressor of SNF1 protein 2) (5Hirata Y. Andoh T. Asahara T. Kikuchi A. Mol. Biol. Cell. 2003; 14: 302-312Crossref PubMed Scopus (72) Google Scholar) in yeasts. Evidence has been accumulating showing that GSK-3 plays a role in the dynamics of microtubules (2Cohen P. Frame S. Nat. Rev. Mol. Cell. Biol. 2001; 2: 769-776Crossref PubMed Scopus (1324) Google Scholar, 6Jope R.S. Johnson G.V.W. Trends Biochem. Sci. 2004; 29: 95-102Abstract Full Text Full Text PDF PubMed Scopus (1346) Google Scholar). GSK-3, which is inactivated on the plus ends of microtubules, mediates Par6-atypical protein kinase C-dependent promotion of polarization and cell protrusion through microtubules (7Etienne-Manneville S. Hall A. Nature. 2003; 421: 753-756Crossref PubMed Scopus (719) Google Scholar). The binding of the adenomatous polyposis coli gene product (APC) to microtubules increases the stability of microtubules, and their interaction is decreased by the phosphorylation of APC by GSK-3β (8Zumbrunn J. Kinoshita K. Hyman A.A. Nathke I.S. Curr. Biol. 2001; 11: 44-49Abstract Full Text Full Text PDF PubMed Scopus (373) Google Scholar). We have found recently that GSK-3β binds and phosphorylates Bicaudal-D (BICD) (9Fumoto K. Hoogenraad C.C. Kikuchi A. EMBO J. 2006; 25: 5670-5682Crossref PubMed Scopus (71) Google Scholar). BICD is a human homologue of Drosophila Bicaudal-D (10Suter B. Romberg L.M. Steward R. Genes Dev. 1989; 3: 1957-1968Crossref PubMed Scopus (152) Google Scholar), and there are two homologues in mammals, BICD1 and BICD2 (11Hoogenraad C.C. Akhmanova A. Howell S.A. Dortland B.R. De Zeeuw C.I. Willemsen R. Visser P. Grosveld F. Galjart N. EMBO J. 2001; 20: 4041-4054Crossref PubMed Scopus (244) Google Scholar). It has been reported that BICD proteins are involved in dynein-mediated minus end-directed transport from the Golgi apparatus to the endoplasmic reticulum (12Matanis T. Akhmanova A. Wulf P. Del Nery E. Weide T. Stepanova T. Galjart N. Grosveld F. Goud B. De Zeeuw C.I. Barnekow A. Hoogenraad C.C. Nat. Cell Biol. 2002; 4: 986-992Crossref PubMed Scopus (320) Google Scholar, 13Hoogenraad C.C. Wulf P. Schiefermeier N. Stepanova T. Galjart N. Small J.V. Grosveld F. de Zeeuw C.I. Akhmanova A. EMBO J. 2003; 22: 6004-6015Crossref PubMed Scopus (168) Google Scholar). In addition to these roles, we showed that GSK-3β functions in transporting centrosomal proteins to the centrosome by stabilizing the BICD and dynein complex, resulting in the regulation of a focused microtubule organization (9Fumoto K. Hoogenraad C.C. Kikuchi A. EMBO J. 2006; 25: 5670-5682Crossref PubMed Scopus (71) Google Scholar). Thus, GSK-3 plays a role at both the plus and minus ends of microtubules to regulate elongation, anchoring, and stability of microtubules in interphase. Microtubule dynamics and organization of the microtubule network need to be controlled during mitosis in order to assemble a spindle apparatus capable of properly segregating the chromosomes. The rapidly growing and shrinking microtubules in the mitotic phase are captured and stabilized by attachment to the kinetochores, hence allowing the formation of a bipolar mitotic spindle. These changes in microtubule dynamics occur concomitantly with the phosphorylation of many proteins, through a number of mitotic kinases, including CDK1 (cyclin-dependent kinase 1), polo kinase, and aurora kinase (14Nigg E.A. Nat. Rev. Mol. Cell. Biol. 2001; 2: 21-32Crossref PubMed Scopus (1275) Google Scholar). It has also been reported that GSK-3 is associated with spindle microtubules and spindle poles and that the inhibition of GSK-3 induces abnormality in astral microtubule length and chromosomal alignment (15Wakefield J.G. Stephens D.J. Tavare J.M. J. Cell Sci. 2003; 116: 637-646Crossref PubMed Scopus (138) Google Scholar, 16Tighe A. Ray-Sinha A. Staples O.D. Taylor S.S. BMC Cell Biol. 2007; 8: 34-50Crossref PubMed Scopus (84) Google Scholar). However, the mechanism by which GSK-3 is involved in the regulation of mitotic spindle dynamics is not known. γ-Tubulin is essential for microtubule nucleation, and there are two γ-tubulin complexes, the γ-tubulin small complex (γTuSC) 2The abbreviations used are: γTuSC, γ-tubulin small complex; γTuRC, γ-tubulin ring complex; RNAi, RNA interference; HA, hemagglutinin; MALDI, matrix-assisted laser desorption/ionization; TOF, time of flight; PBS, phosphate-buffered saline; siRNA, small interference RNA; HEK, human embryonic kidney; EGFP, enhanced green fluorescence protein; GST, glutathione S-transferase; HA, hemagglutinin; IP, immunoprecipitation; PIPES, 1,4-piperazinediethanesulfonic acid; MS, mass spectrometry; PI, propidium iodide; DMSO, dimethyl sulfoxide. and the γ-tubulin ring complex (γTuRC) (17Wiese C. Zheng Y. Curr. Opin. Struct. Biol. 1999; 9: 250-259Crossref PubMed Scopus (97) Google Scholar). The simplest γTuSC, which is conserved among many organisms, including yeast, flies, and humans, contains γ-tubulin, Spc97p/Dgrip84/GCP2 (γ-tubulin complex protein-2), and Spc98p/Dgrip91/GCP3 and displays low microtubule nucleation activity in vitro (18Oegema K. Wiese C. Martin O.C. Milligan R.A. Iwamatsu A. Mitchison T.J. Zheng Y. J. Cell Biol. 1999; 144: 721-733Crossref PubMed Scopus (254) Google Scholar). The functions of γTuSC have been studied extensively by genetic approaches, and deletion of either corresponding gene is lethal, resulting in an accumulation of cells in mitosis (19Zhang L. Keating T.J. Wilde A. Borisy G.G. Zheng Y. J. Cell Biol. 2000; 151: 1525-1535Crossref PubMed Scopus (48) Google Scholar, 20Geissler S. Pereira G. Spang A. Knop M. Soues S. Kilmartin J. Schiebel E. EMBO J. 1996; 15: 3899-3911Crossref PubMed Scopus (158) Google Scholar). The larger complex, γTuRC, contains additional cap subunits Dgrip75/GCP4, Dgrip128/GCP5, Dgrip163/GCP6, and Dgp71WD/GCP-WD/NEDD1 (neural precursor cell-expressed, developmentally down-regulated protein 1), which hold multiple γTuSC subcomplexes together (21Wiese C. Zheng Y. J. Cell Sci. 2006; 119: 4143-4153Crossref PubMed Scopus (173) Google Scholar). Studies in Xenopus and Drosophila have shown that γTuRC associates with microtubule minus ends, possesses higher microtubule nucleation activity in vitro compared with γTuSC, and is required for the assembly of fully functional spindles (22Zheng Y. Wong M.L. Alberts B. Mitchison T. Nature. 1995; 378: 578-583Crossref PubMed Scopus (729) Google Scholar, 23Moritz M. Braunfeld M.B. Guenebaut V. Heuser J. Agard D.A. Nat. Cell Biol. 2000; 2: 365-370Crossref PubMed Scopus (236) Google Scholar, 24Verollet C. Colombie N. Daubon T. Bourbon H.M. Wright M. Raynaud-Messina B. J. Cell Biol. 2006; 172: 517-528Crossref PubMed Scopus (84) Google Scholar). In interphase, most γTuRCs are in the cytoplasm; however, at the onset of mitosis, three to five times more γ-tubulin is recruited to the centrosome (21Wiese C. Zheng Y. J. Cell Sci. 2006; 119: 4143-4153Crossref PubMed Scopus (173) Google Scholar). How γTuRC is targeted to the centrosome in a cell cycle-dependent manner is not well understood. In this study, we identified GCP5 as a novel binding protein of GSK-3. With the use of a small molecule GSK-3 inhibitor combined with an RNAi approach, we demonstrated that GSK-3 regulates the localization of GCP5 and γTuRC to the spindle poles in order to control proper mitotic spindle formation. Materials and Chemicals—HeLa S3 cells, U2OS cells, human GSK-3β cDNA, and KIAA1899 (human GCP5) cDNA were kindly provided by Dr. K. Matsumoto (Nagoya University, Nagoya, Japan), H. Saya (Keio University, Tokyo, Japan), Dr. J. R. Woodgett (Samuel Lunenfeld Research Institute, Toronto, Canada), and Kazusa DNA Research Institute (Chiba, Japan), respectively. Human GCP2 and GCP3 cDNAs and anti-GCP2 antibody were kind gifts from Dr. Y. Ono (Kobe University, Kobe, Japan). Anti-centrin 3 antibody was provided by Dr. M. Bornens (Institut Curie, Paris, France). Anti-GCP5 antibody was prepared in rabbits by immunization with recombinant full-length GCP5 protein. Anti-Myc antibody was prepared from 9E10 cells. Mouse monoclonal anti-GSK-3β and anti-EB1 antibodies were purchased from Transduction Laboratories. Mouse monoclonal anti-FLAG M2 antibody-conjugated agarose and anti-γ-tubulin antibody and a rabbit polyclonal anti-γ-tubulin antibody were from Sigma. Rabbit polyclonal anti-GFP and mouse monoclonal anti-HA (16B12) antibodies were from Invitrogen and COVANCE, respectively. GSK-3 inhibitors, SB216763 and SB415286, were from TOCRIS and Sigma, respectively. Other materials were from commercial sources. Plasmid Construction—pCGN/GSK-3β, pCGN/GSK-3β (K85R), pEGFP/GSK-3β, pEF-BOS-Myc/GSK-3β, and pGEX-4T/GSK-3β were constructed as described previously (25Ikeda S. Kishida S. Yamamoto H. Murai H. Koyama S. Kikuchi A. EMBO J. 1998; 17: 1371-1384Crossref PubMed Scopus (1113) Google Scholar, 26Tanji C. Yamamoto H. Yorioka N. Kohno N. Kikuchi K. Kikuchi A. J. Biol. Chem. 2002; 277: 36955-36961Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 27Hino S.-I. Michiue T. Asashima M. Kikuchi A. J. Biol. Chem. 2003; 278: 14066-14073Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). Standard recombinant DNA techniques were used to construct the following plasmids: pCGN/GCP5, pCGN/GCP5-(1–567), pCGN/GCP5-(568–1024), pEGFP/GCP5, and pMAL-C2/GCP5. Cell Culture—HeLa S3, U2OS, and HEK-293T cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% calf serum or 10% fetal bovine serum. HeLa cells used for large scale spinner cultures to isolate GSK-3β complexes were grown in Joklik medium supplemented with 10% calf serum. Cells were transfected with plasmid DNA using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol. When necessary, HeLa cells were incubated with 30 μm SB415286 for 16 h. Immunoaffinity Purification of GSK-3β Protein Complexes— HeLa cells were transduced with a recombinant retrovirus expressing a bicistronic mRNA coding FLAG-HA-tagged human GSK-3β linked to an interleukin-2 receptor as a surface marker. The transduced subpopulation was purified by repeated cycles of affinity cell sorting (28Nakatani Y. Ogryzko V. Methods Enzymol. 2003; 370: 430-444Crossref PubMed Scopus (232) Google Scholar). The cytoplasmic extracts were prepared as follows. Twelve liters of HeLa S3 cells (a total of ∼9 × 109 cells) were centrifuged at 2,246 × g for 8 min at 4 °C, and the cell pellet was washed once with phosphate-buffered saline (PBS). The cell pellet was resuspended with 6× cell volume of hypotonic buffer (10 mm Tris-HCl, pH 7.3, 10 mm KCl, 1.5 mm MgCl2, 1 mm phenylmethylsulfonyl fluoride, and 10 mm β-mercaptoethanol) and shaken for several min. The suspended cells were centrifuged at 980 × g for 5 min at 4 °C, and the supernatant was discarded. The pellet was resuspended again with 1× cell volume of hypotonic buffer. After 10 min of incubation on ice, the swollen cells were homogenized with a Dounce homogenizer. The homogenate was centrifuged at 2,385 × g for 15 min at 4 °C, and the resulting supernatant (cytoplasmic extracts) was supplemented with one-tenth volume of 10× buffer (300 mm Tris-HCl, pH 7.3, 1.5 m KCl, 30 mm MgCl2) and incubated for 1 h at 4 °C. The extract was ultracentrifuged at 90,800 × g for 1 h at 4 °C. The obtained cytoplasmic extract was immediately frozen in liquid nitrogen before storage at –80 °C. The GSK-3β-containing complexes in the cytoplasmic extracts (10 ml, 100 mg of total protein) were incubated with anti-FLAG M2 antibody-conjugated agarose (500 μl of beads). As a control, mock purification was performed from nontransduced HeLa cells through all purification steps. After an extensive wash with IP buffer (20 mm Tris-HCl, pH 8.0, 100 mm KCl, 5 mm MgCl2, 0.2 mm EDTA, 10% glycerol, 0.1% Tween 20, 1 mm β-mercaptoethanol, and a protease inhibitor mix (1 mm phenylmethylsulfonyl fluoride, 20 μg/ml leupeptin, 20 μg/ml aprotinin, and 10 μg/ml pepstatin A)), the bound protein complexes were eluted from M2-agarose by incubation for 30 min with the 0.1 mg/ml FLAG peptide (Sigma) in 1 ml of IP buffer. The eluates were further purified by immunoprecipitation with anti-HA (12CA5) antibody conjugated to protein A-Sepharose (General Electric) (500 μl of beads). After washing with IP buffer, the bound complexes were eluted with 500 μl of 100 mm glycine-HCl, pH 2.5. Mass Spectrometry Analysis—The GSK-3β complex purified from cytoplasmic extracts of HeLa cell was concentrated by StrataClean Resin (Stratagene). Resin-bound proteins were eluted with Laemmli buffer. The sample was separated by SDS-PAGE and silver-stained. Each band was excised from the SDS-polyacrylamide gel, digested with trypsin, and analyzed by mass spectrometry. The digest was lyophilized and dissolved in 50 μl of 0.1% trifluoroacetic acid solution, desalted using ZipTip μC18, and subjected to MALDI-TOF MS analysis. All MALDI-TOF mass spectra were obtained in the reflector positive mode using α-cyano-4-hydroxycinnamic acid (saturated solution in 50% acetonitrile with 0.1% trifluoroacetic acid) as the matrix. Analytes were prepared by mixing 0.5 μl of peptide sample with 0.5 μl of matrix solution on a MALDI plate and allowed to air-dry at room temperature in a hood before being inserted into the spectrometer. External calibration was conducted using ACTH-(1–7) and bradykinin. Peptides were identified using the Mascot search program (Matrix Science) to perform theoretical trypsin digests. All visible bands were analyzed, but some bands were not hit by the Mascot search program search. In addition to that, there were bands excised as a single band that consisted of multiple molecules. Only the consistent results through three independent experiments were described. Complex Formation Assay—To show the interaction between endogenous GCP5 and GSK-3β, HeLa S3 cells were washed once with PBS and lysed in IP buffer. The lysates were immunoprecipitated with anti-GCP5 antibody, and the immunoprecipitate was probed with the indicated antibodies. To show the complex formation of overexpressed proteins, the indicated proteins were expressed in HEK-293T cells. The cells were washed once with Hepes-buffered saline (50 mm HEPES-NaOH, pH 7.4, 150 mm NaCl) and lysed in lysis buffer (50 mm HEPES-NaOH, pH 7.4, 150 mm NaCl, 1 mm dithiothreitol, 1 mm EDTA, 0.5% Triton X-100, and a protease inhibitor mix). The lysates were centrifuged at 16,000 × g for 10 min at 4 °C, and then the obtained supernatants were immunoprecipitated with the indicated antibodies. The experiments were repeated at least three times, and representative data are shown. Immunocytochemistry—HeLa S3 and U2OS cells were fixed with 4% paraformaldehyde for 10 min and permeabilized with PBS containing 0.5% Triton X-100 for 10 min. Alternatively, the cultured cells were fixed with 100% methanol at –20 °C for 5 min and blocked in 2% bovine serum albumin in PBS for 10 min. To measure the accumulation area of γ-tubulin and GCP5 at the spindle poles, each staining area at the spindle pole was measured manually using LSM510 software. For microtubule nucleation activity measurement, quantification of fluorescence intensity of EB1 staining at the spindle poles (1 μm2 circular area) was performed. To show colocalization of two proteins, more than 50 cells were observed carefully, and representative images are shown in the figures. RNAi—The following double-stranded RNA oligonucleotides were synthesized using a CUGA in vitro small interference RNA (siRNA) synthesis kit (Nippon Gene), and the targeted sequences were as follows: human GCP5, 5′-GGAACATCATGTGGTCCATCA-3′; control (scrambled siRNA), 5′-CAGTCGCGTTTGCGACTGG-3′. HeLa S3 cells were transfected with each siRNA at 100 nm using Oligofectamine (Invitrogen), and the cells were used for experiments at 72 h post-transfection. Sucrose Density Gradient Centrifugation Analysis—HeLa S3 cell extracts were prepared accordingly (29Murphy S.M. Urbani L. Stearns T. J. Cell Biol. 1998; 141: 663-674Crossref PubMed Scopus (180) Google Scholar) with slight modifications. In brief, confluent cells (10-cm diameter dish) were lysed in 0.4 ml of HEPES-S buffer (50 mm HEPES-NaOH, pH 7.4, 100 mm NaCl, 1 mm EGTA, 1 mm MgCl2, 1 mm β-mercaptoethanol, and a protease inhibitor mix). The cell extract was homogenized by passing 30 times through a 29½-gauge needle. After centrifugation for 10 min at 12,000 × g at 4 °C, the supernatant was loaded onto a 3.3-ml 5–40% discontinuous sucrose gradient consisting of 0.66 ml each of 5, 10, 20, 30, and 40% sucrose in HEPES-S buffer. The gradient was then centrifuged in a RPS56T-swing (Hitachi, Tokyo, Japan) rotor for 4 h at 314,000 × g at 4 °C. Fractions were collected from top to bottom (14 fractions) and analyzed by immunoblotting. Centrosome Isolation—Centrosome was prepared from mitotic HeLa S3 cells according to the methods described previously (9Fumoto K. Hoogenraad C.C. Kikuchi A. EMBO J. 2006; 25: 5670-5682Crossref PubMed Scopus (71) Google Scholar, 30Mitchison T.J. Kirschner M.W. Methods Enzymol. 1986; 134: 261-268Crossref PubMed Scopus (79) Google Scholar). To obtain mitotic cells, HeLa cells were treated with double thymidine blocks with the addition of 2 mm thymidine. After the second thymidine block, cells were released for 6 h before adding 50 ng/ml nocodazole. At 6–8 h after the addition of nocodazole, cells arrested in mitosis were harvested by shake-off. After centrifugation of HeLa S3 cell lysates by a discontinuous gradient consisting of 1 ml of 40%, 1 ml of 50%, and 1.6 ml of 70% sucrose solutions, fractions (200 μl/each fraction) were collected from top to bottom. The centrosome fractions were determined by immunoblotting with anti-γ-tubulin and anti-GCP5 antibodies. The fractions that contained centrosome were then prepared accordingly as described (31Okuda M. Horn H.F. Tarapore P. Tokuyama Y. Smulian A.G. Chan P.K. Knudsen E.S. Hofmann I.A. Snyder J.D. Bove K.E. Fukasawa K. Cell. 2000; 103: 127-140Abstract Full Text Full Text PDF PubMed Scopus (572) Google Scholar). Centrosome fractions (400 μl) were diluted with 5× volume of PIPES buffer (10 mm PIPES-NaOH, pH 7.2) and centrifuged for 15 min at 24,000 × g at 4 °C. Pelleted centrosome was then resuspended in 200 μl of PIPES buffer. Purification of HA-GCP5/γTuRC—Forty-eight hours after transfection, lysates of HEK-293T cells expressing HA-GCP5 (three 10-cm diameter dishes) were prepared to purify HA-GCP5/γTuRC as described (32Murphy S.M. Preble A.M. Patel U.K. O'Connell K.L. Dias D.P. Moritz M. Agard D. Stults J.T. Stearns T. Mol. Biol. Cell. 2001; 12: 3340-3352Crossref PubMed Scopus (154) Google Scholar). The lysates were immunoprecipitated with anti-HA (12CA5) antibody-conjugated Sepharose (300 μl of beads), and the beads were washed three times with HEPES-A buffer (50 mm HEPES-NaOH, pH 7.4, 150 mm NaCl, 1 mm dithiothreitol, 1 mm EGTA, 1 mm MgCl2, 0.25 mm GTP, and a protease inhibitor mix) containing 0.5% Triton X-100, once with HEPES-A buffer containing 250 mm NaCl instead of 150 mm NaCl, and once with HEPES-A buffer. The complex was eluted by incubation with 300 μl of 1 mg/ml HA peptide (Roche Applied Science) in HEPES-A buffer for 3 h at 4 °C. The presence of HA-GCP5 and γ-tubulin complex in the eluates was confirmed by immunoblotting with anti-GCP5, anti-GCP2, and anti-γ-tubulin antibodies. Co-sedimentation Assay of GCP5/γTuRC and Centrosome— An In vitro sedimentation assay was performed by mixing equal volumes (10 μl each) of centrosome fractions (0.8 μg) and purified HA-GCP5/γTuRC (35 ng). To inhibit GSK-3 activity, SB415286 was added in the mixture at the beginning of incubation. After incubation for 30 min at 30 °C, the mixture was separated into the pellet and supernatant by centrifugation at 20,000 × g for 10 min at 4 °C. The pellet was washed once with PIPES buffer and resuspended in Laemmli buffer. The samples were subjected to SDS-PAGE, followed by immunoblotting with the indicated antibodies. Statistical Analysis—All of the chromosome alignment counts, GCP5 and γ-tubulin area measurements, and microtubule nucleation activity quantifications were performed for at least three independent experiments. At least 20 cells in each of prophase, prometaphase, and metaphase were analyzed in a single experiment. Five hundred cells were analyzed in a single chromosome alignment count. The results shown indicate means ± S.D. Statistical analysis was performed using StatView software (SAS Institute Inc.). An unpaired t test with a p value of <0.05 was used to determine statistical significance. Identification of GCP5 as a GSK-3-binding Protein—To understand how GSK-3 is involved in mitotic spindle formation, GSK-3-binding proteins were purified from cytoplasmic extracts of HeLa cells stably expressing FLAG and HA epitope-tagged GSK-3β. As a first step, the proteins were purified by affinity chromatography with anti-FLAG antibody-conjugated agarose, and the bound polypeptides were eluted with the FLAG peptide. As a control, we performed a mock purification from cytoplasmic extracts prepared from nontransduced HeLa cells. The FLAG affinity-purified materials were further purified by immunoaffinity chromatography with an anti-HA antibody. After the second affinity purification, these proteins were subjected to protein identification by mass spectrometry. Twelve proteins were identified (Fig. 1, A and B). Among them, we were interested in GCP5, because it is one of the subunits comprising γTuRC (32Murphy S.M. Preble A.M. Patel U.K. O'Connell K.L. Dias D.P. Moritz M. Agard D. Stults J.T. Stearns T. Mol. Biol. Cell. 2001; 12: 3340-3352Crossref PubMed Scopus (154) Google Scholar). GCP5 belongs to the GCP family (GCP2 to -6) and has two conserved regions, called Spc (spindle pole body component), in common with other members of this family (Fig. 1C). GSK-3β and GCP5 formed a complex at their endogenous levels in HeLa S3 cells when GCP5 was immunoprecipitated with a specific antibody (Fig. 2A). γ-Tubulin was also observed in the immunocomplexes (Fig. 2A). When HA-GCP5 and Myc-GSK-3β were co-expressed in HEK-293T cells, HA-GCP5 was observed in the Myc-GSK-3β immunocomplexes (Fig. 2B). Reciprocally, enhanced green fluorescence protein (EGFP)-fused GSK-3β (EGFP-GSK-3β) was immunoprecipitated with HA-GCP5 (Fig. 2B). In addition, EGFP-GSK-3β also formed a complex with HA-GCP2 and HA-GCP3, which are the other components of γTuRC (Fig. 2B). In vitro binding studies using recombinant proteins demonstrated that GST-GSK-3β binds directly to MBP-GCP5 at a Kd value of ∼125 nm (Fig. 2C). Both wild-type and a kinase-inactive mutant (K85R) of GSK-3β formed a complex with EGFP-GCP5 (supplemental Fig. S1). These results indicate that GCP5 binds directly to GSK-3β in intact cells and that the kinase activity of GSK-3 is not necessary for the formation of the complex. To identify which region of GCP5 is important for binding to GSK-3β, two deletion mutants of HA-GCP5 were expressed in HEK-293T cells (Fig. 2D). Myc-GSK-3β formed a complex with GCP5-N (GCP5-(1–567)), the N-terminal region of GCP5, but interacted little with HA-GCP5-C (GCP5-(568–1024)), the C-terminal region of GCP5 (Fig. 2D). HA-GCP5-C formed a complex with γ-tubulin slightly, and HA-GCP5-N did not. These results suggest that GCP5 has different sites for interaction with GSK-3β and γ-tubulin. Consistent with these observations, HA-GCP5 formed a complex with γ-tubulin and Myc-GSK-3β (Fig. 2D). Therefore, it is likely that GSK-3β is associated with the γ-tubulin complex. Roles of GCP5 in the Recruitment of γ-Tubulin to the Spindle Poles—Since it has been shown that GCP5 is localized with γ-tubulin as γTuRC to the centrosome (32Murphy S.M. Preble A.M. Patel U.K. O'Connell K.L. Dias D.P. Moritz M. Agard D. Stults J.T. Stearns T. Mol. Biol. Cell. 2001; 12: 3340-3352Crossref PubMed Scopus (154) Google Scholar), we examined the role of GCP5 in the formation of γTuRC, to determine the physiological relevance of the function of a complex between GSK-3β and GCP5. When the lysates of HeLa S3 cells were fractionated by sucrose density gradient centrifugation, γ-tubulin, GCP2, and GCP5 co-sedimented in fraction 11 (Fig. 3A). γTuRC has been shown to have a size of ∼32 S (32Murphy S.M. Preble A.M. Patel U.K. O'Connell K.L. Dias D.P. Moritz M. Agard D. Stults J.T. Stearns T. Mol. Biol. Cell. 2001; 12: 3340-3352Crossref PubMed Scopus (154) Google Scholar, 33Haren L. Remy M.H. Bazin I. Callebaut I. Wright M. Merdes A. J. Cell Biol. 2006; 172: 505-515Crossref PubMed Scopus (219) Google Scholar); thus, the heavy fraction (fraction 11 in Fig. 3A) in the HeLa S3 cell lysates could correspond to γTuRC. GCP2 and γ-tubulin were also present in light fractions (fraction 5), which presumably correspond to γTuSC (18Oegema K. Wiese C. Martin O.C. Milligan R.A. Iwamatsu A. Mitchison T.J. Zheng Y. J. Cell Biol. 1999; 144: 721-733Crossref PubMed Scopus (254) Google Scholar, 34Moritz M. Zheng Y. Alberts B.M. Oegema K. J. Cell Biol. 1998; 142: 775-786Crossref PubMed Scopus (203) Google Scholar). The light fractional entity of GCP5 (fraction 4) was presumed to correspond to monomeric GCP5 (35Vogt N. Koch I. Schwarz H. Schnorrer F. Nusslein-Volhard C. Development. 2006; 133: 3963-3972Crossref PubMed Scopus (49) Google Scholar). GCP5, GCP2, and γ-tubulin in fraction 3 may be breakdown products of γTuRC during sucrose density gradient centrifugation (33Haren L. Remy M.H. Bazin I. Callebaut I. Wright M. Merdes A. J. Cell Biol. 2006; 172: 505-515Crossref PubMed Scopus (219) Google Scholar). When GCP5 was depleted by RNAi, endogenous GCP5 was reduced to less than 20% (Fig. 3A). This depletion was specific to GCP5 as determined by examining the protein levels of GCP2, β-tubulin, γ-tubulin, GSK-3β, and β-actin (Fig. 3A). When extracts from GCP5-depleted cells were subjected to the sucrose density gradient centrifugation, the amounts of γ-tubulin and GCP2 present in the γTuRC fraction were greatly reduced (Fig. 3A). Since the total amounts of γ-tubulin and GCP-2 were not changed by depl
Referência(s)