Mitosis-specific Activation of LIM Motif-containing Protein Kinase and Roles of Cofilin Phosphorylation and Dephosphorylation in Mitosis
2002; Elsevier BV; Volume: 277; Issue: 24 Linguagem: Inglês
10.1074/jbc.m201444200
ISSN1083-351X
AutoresToru Amano, Noriko Kaji, Kazumasa Ohashi, Kensaku Mizuno,
Tópico(s)Hippo pathway signaling and YAP/TAZ
ResumoActin filament dynamics play a critical role in mitosis and cytokinesis. LIM motif-containing protein kinase 1 (LIMK1) regulates actin reorganization by phosphorylating and inactivating cofilin, an actin-depolymerizing and -severing protein. To examine the role of LIMK1 and cofilin during the cell cycle, we measured cell cycle-associated changes in the kinase activity of LIMK1 and in the level of cofilin phosphorylation. Using synchronized HeLa cells, we found that LIMK1 became hyperphosphorylated and activated in prometaphase and metaphase, then gradually returned to the basal level as cells entered into telophase and cytokinesis. Although Rho-associated kinase and p21-activated protein kinase phosphorylate and activate LIMK1, they are not likely to be involved in mitosis-specific activation and phosphorylation of LIMK1. Immunoblot and immunofluorescence analyses using an anti-phosphocofilin-specific antibody revealed that the level of cofilin phosphorylation, similar to levels of LIMK1 activity, increased during prometaphase and metaphase then gradually declined in telophase and cytokinesis. Ectopic expression of LIMK1 increased the level of cofilin phosphorylation throughout the cell cycle and induced the formation of multinucleate cells. These results suggest that LIMK1 is involved principally in control of mitosis-specific cofilin phosphorylation and that dephosphorylation and reactivation of cofilin at later stages of mitosis play a critical role in cytokinesis of mammalian cells. Actin filament dynamics play a critical role in mitosis and cytokinesis. LIM motif-containing protein kinase 1 (LIMK1) regulates actin reorganization by phosphorylating and inactivating cofilin, an actin-depolymerizing and -severing protein. To examine the role of LIMK1 and cofilin during the cell cycle, we measured cell cycle-associated changes in the kinase activity of LIMK1 and in the level of cofilin phosphorylation. Using synchronized HeLa cells, we found that LIMK1 became hyperphosphorylated and activated in prometaphase and metaphase, then gradually returned to the basal level as cells entered into telophase and cytokinesis. Although Rho-associated kinase and p21-activated protein kinase phosphorylate and activate LIMK1, they are not likely to be involved in mitosis-specific activation and phosphorylation of LIMK1. Immunoblot and immunofluorescence analyses using an anti-phosphocofilin-specific antibody revealed that the level of cofilin phosphorylation, similar to levels of LIMK1 activity, increased during prometaphase and metaphase then gradually declined in telophase and cytokinesis. Ectopic expression of LIMK1 increased the level of cofilin phosphorylation throughout the cell cycle and induced the formation of multinucleate cells. These results suggest that LIMK1 is involved principally in control of mitosis-specific cofilin phosphorylation and that dephosphorylation and reactivation of cofilin at later stages of mitosis play a critical role in cytokinesis of mammalian cells. actin-depolymerizing factor autoinhibitory domain 4,6-diamidino-2-phenylindole Dulbecco's modified Eagle's medium fetal calf serum LIM motif-containing protein kinase mitogen-activated protein kinase Ser-3-phosphorylated cofilin p21-activated protein kinase phosphate-buffered saline protein kinase domain Rho-associated kinase wild-type yellow fluorescent protein During the cell division cycle, the morphology of cells is altered. In animal cells in culture, flat and adherent cells in interphase become spherical and weakly adherent in mitosis. At later stages of mitosis, cortical actin filaments are reorganized and recruited to the cleavage furrow, and an actomyosin-based contractile ring is formed around the equator of dividing cells, constricted to lead to cell cleavage, and disappears at the end of cytokinesis. Actin filament dynamics, reorganization, and redistribution play a principal role in these processes (1Fishkind D.J. Wang Y.L. Curr. Opin. Cell Biol. 1995; 7: 23-31Crossref PubMed Scopus (178) Google Scholar, 2Robinson D.N. Spudich J.A. Trends Cell Biol. 2000; 10: 228-237Abstract Full Text Full Text PDF PubMed Scopus (152) Google Scholar, 3Prokopenko S.N. Saint R. Bellen H.J. J. Cell Biol. 2000; 148: 843-848Crossref PubMed Scopus (60) Google Scholar). Previous studies revealed the involvement of various actin-binding proteins and signaling proteins that regulate actin filaments in mitotic processes and cytokinesis (1Fishkind D.J. Wang Y.L. Curr. Opin. Cell Biol. 1995; 7: 23-31Crossref PubMed Scopus (178) Google Scholar, 2Robinson D.N. Spudich J.A. Trends Cell Biol. 2000; 10: 228-237Abstract Full Text Full Text PDF PubMed Scopus (152) Google Scholar, 3Prokopenko S.N. Saint R. Bellen H.J. J. Cell Biol. 2000; 148: 843-848Crossref PubMed Scopus (60) Google Scholar), but mechanisms by which cells coordinately change shapes in response to cell cycle cues remain largely uncharacterized. Cofilin and its close relative, actin-depolymerizing factor (ADF),1 bind to actin monomers and filaments and play a critical role in regulating actin filament dynamics and reorganization by stimulating depolymerization and severance of actin filaments (4Moon A. Drubin D.G. Mol. Biol. Cell. 1995; 6: 1423-1431Crossref PubMed Scopus (226) Google Scholar, 5Bamburg J.R. Annu. Rev. Cell Dev. Biol. 1999; 15: 185-230Crossref PubMed Scopus (842) Google Scholar, 6Pantaloni D., Le Clainche C. Carlier M.-F. Science. 2001; 292: 1502-1506Crossref PubMed Scopus (559) Google Scholar). The activity of cofilin/ADF is reversibly regulated by phosphorylation and dephosphorylation at Ser-3, with the phosphorylated form being inactive (7Agnew B.J. Minamide L.S. Bamburg J.R. J. Biol. Chem. 1995; 270: 17582-17587Abstract Full Text Full Text PDF PubMed Scopus (317) Google Scholar, 8Moriyama K. Iida K. Yahara I. Genes Cells. 1996; 1: 73-86Crossref PubMed Scopus (316) Google Scholar). LIM kinase 1 (LIMK1) and LIM kinase 2 (LIMK2) (9Mizuno K. Okano I. Ohashi K. Nunoue K. Kuma K. Miyata T. Nakamura T. Oncogene. 1994; 9: 1605-1612PubMed Google Scholar, 10Okano I. Hiraoka J. Otera H. Nunoue K. Ohashi K. Iwashita S. Hirai M. Mizuno K. J. Biol. Chem. 1995; 270: 31321-31330Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar) phosphorylate cofilin/ADF specifically at Ser-3 and thereby inhibit actin binding, depolymerizing, and severing activities of cofilin/ADF (11Yang N. Higuchi O. Ohashi K. Nagata K. Wada A. Kangawa K. Nishida E. Mizuno K. Nature. 1998; 393: 809-812Crossref PubMed Scopus (1068) Google Scholar, 12Arber S. Barbayannis F.A. Hanser H. Schneider C. Stanyon C.A. Bernard O. Caroni P. Nature. 1998; 393: 805-809Crossref PubMed Scopus (1158) Google Scholar). LIM kinases are activated by Rho family small GTPases, Rac, Rho, and Cdc42, this activation being mediated by downstream effector protein kinases, such as p21-activated protein kinase (PAK) and Rho-associated kinase (ROCK), by phosphorylation of Thr-508 of LIMK1 or Thr-505 of LIMK2 (11Yang N. Higuchi O. Ohashi K. Nagata K. Wada A. Kangawa K. Nishida E. Mizuno K. Nature. 1998; 393: 809-812Crossref PubMed Scopus (1068) Google Scholar, 12Arber S. Barbayannis F.A. Hanser H. Schneider C. Stanyon C.A. Bernard O. Caroni P. Nature. 1998; 393: 805-809Crossref PubMed Scopus (1158) Google Scholar, 13Chernoff J. Nat. Cell Biol. 1999; 1: E115-E117Crossref PubMed Scopus (12) Google Scholar, 14Sumi T. Matsumoto K. Takai Y. Nakamura T. J. Cell Biol. 1999; 147: 1519-1532Crossref PubMed Scopus (314) Google Scholar, 15Edwards D.C. Sanders L.C. Bokoch G.M. Gill G.N. Nat. Cell Biol. 1999; 1: 253-259Crossref PubMed Scopus (846) Google Scholar, 16Maekawa M. Ishizaki T. Boku S. Watanabe N. Fujita A. Iwamatsu A. Obinata T. Ohashi K. Mizuno K. Narumiya S. Science. 1999; 285: 895-898Crossref PubMed Scopus (1288) Google Scholar, 17Ohashi K. Nagata K. Maekawa M. Ishizaki T. Narumiya S. Mizuno K. J. Biol. Chem. 2000; 275: 3577-3582Abstract Full Text Full Text PDF PubMed Scopus (421) Google Scholar, 18Amano T. Tanabe K. Eto T. Narumiya S. Mizuno K. Biochem. J. 2001; 354: 149-159Crossref PubMed Scopus (137) Google Scholar). Previous studies indicated that cofilin phosphorylation by LIM kinases is a critical signaling event in a variety of stimulus-induced cell responses, including growth factor-induced lamellipodium formation, lysophophatidic acid-induced stress fiber formation, chemokine-induced T cell chemotaxis, semaphorin 3A-induced neuronal growth cone collapse, and pathogenic bacteriumListeria-induced phagocytosis (11Yang N. Higuchi O. Ohashi K. Nagata K. Wada A. Kangawa K. Nishida E. Mizuno K. Nature. 1998; 393: 809-812Crossref PubMed Scopus (1068) Google Scholar, 16Maekawa M. Ishizaki T. Boku S. Watanabe N. Fujita A. Iwamatsu A. Obinata T. Ohashi K. Mizuno K. Narumiya S. Science. 1999; 285: 895-898Crossref PubMed Scopus (1288) Google Scholar, 19Zebda N. Bernard O. Bailly M. Welti S. Lawrence D.S. Condeelis J.S. J. Cell Biol. 2000; 151: 1119-1128Crossref PubMed Scopus (170) Google Scholar, 20Nishita M. Aizawa H. Mizuno K. Mol. Cell. Biol. 2002; 22: 774-783Crossref PubMed Scopus (117) Google Scholar, 21Aizawa H. Wakatsuki S. Ishii A. Moriyama K. Sasaki Y. Ohashi K. Sekine-Aizawa Y. Sehara-Fujisawa A. Mizuno K. Goshima Y. Yahara I. Nat. Neurosci. 2001; 4: 367-373Crossref PubMed Scopus (297) Google Scholar, 22Bierne H. Gouin E. Roux P. Caroni P. Yin H.L. Cossart P. J. Cell Biol. 2001; 155: 101-112Crossref PubMed Scopus (155) Google Scholar). In addition to functions in actin cytoskeletal remodeling in interphase cells, cofilin/ADF seems to be involved in processes related to mitosis and cytokinesis. Cofilin/ADF is concentrated at the cleavage furrow and the midbody during cytokinesis of cultured mammalian cells (23Nagaoka R. Abe H. Kusano K. Obinata T. Cell Motil. Cytoskeleton. 1995; 30: 1-7Crossref PubMed Scopus (60) Google Scholar) and cleavage of Xenopus fertilized eggs (24Abe H. Obinata T. Minamide L.S. Bamburg J.R. J. Cell Biol. 1996; 132: 871-885Crossref PubMed Scopus (165) Google Scholar). Injection of the antibody that inhibits the activity of cofilin/ADF intoXenopus blastomeres blocks the cleavage of the blastomeres (24Abe H. Obinata T. Minamide L.S. Bamburg J.R. J. Cell Biol. 1996; 132: 871-885Crossref PubMed Scopus (165) Google Scholar). In addition, Drosophila twinstar mutants, in which expression of the twinstar gene encoding aDrosophila cofilin/ADF ortholog is repressed, exhibit frequent failures in cytokinesis in larval neuroblasts and in testicular meiotic cells (25Gunsalus K.C. Bonaccorsi S. Williams E. Verni F. Gatti M. Goldberg M.L. J. Cell Biol. 1995; 131: 1243-1259Crossref PubMed Scopus (253) Google Scholar). These observations suggest that cofilin/ADF plays a critical role in cytokinesis. Furthermore, injection into Xenopus blastomeres of a nonphosphorylatable (constitutively active) form of cofilin/ADF blocks cytokinesis, but injection of wild-type cofilin/ADF that can be phosphorylated has no apparent effect (24Abe H. Obinata T. Minamide L.S. Bamburg J.R. J. Cell Biol. 1996; 132: 871-885Crossref PubMed Scopus (165) Google Scholar), thus indicating that excess activity of cofilin/ADF also prevents cytokinesis, and the proper control of cofilin/ADF activity by phosphorylation and dephosphorylation is important for progression of mitosis. We thus assumed that LIM kinases play a role in mitosis and cytokinesis by phosphorylating and regulating the activity of cofilin/ADF. We have now examined changes in the kinase activity of LIMK1 and the level of cofilin phosphorylation during the cell cycle, and we found that LIMK1 becomes activated and hyperphosphorylated in the early stages of mitosis and then gradually reverts to basal levels in late stages. Mitotic activation of LIMK1 is induced by a mechanism distinct from that of Thr-508 phosphorylation by ROCK or PAK. We also noted cell cycle-associated changes in the level of cofilin phosphorylation which are similar to those seen with LIMK1 activity. Ectopic expression of LIMK1 induces increases in the level of cofilin phosphorylation throughout the cell cycle and leads to the formation of multinucleate cells. We propose that LIMK1 plays an important role in regulating cofilin/ADF activity during mitosis and that cofilin dephosphorylation in the late stages of mitosis is critical for cytokinesis. To generate a pMYC-C1 vector containing the Myc epitope sequence (EQKLISEEDL), a pEYFP-C1 vector (CLONTECH, Palo Alto, CA) was digested withNheI and BglII, and the oligonucleotides coding for Myc epitope peptide were inserted in place of the cDNA for yellow fluorescent protein (YFP). Expression plasmids coding for N-terminally Myc-tagged LIMK1 and LIMK2 were constructed by inserting full-length human LIMK1 and LIMK2 cDNAs (9Mizuno K. Okano I. Ohashi K. Nunoue K. Kuma K. Miyata T. Nakamura T. Oncogene. 1994; 9: 1605-1612PubMed Google Scholar, 10Okano I. Hiraoka J. Otera H. Nunoue K. Ohashi K. Iwashita S. Hirai M. Mizuno K. J. Biol. Chem. 1995; 270: 31321-31330Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar), respectively, into the BglII and SacII sites of the pMYC-C1 vector. Plasmids coding for Myc-LIMK1(D460A) and Myc-LIMK1(T508V), in which Asp-460 and The-508 in Myc-LIMK1 were replaced by Ala and Val, respectively, were constructed, using a site-directed mutagenesis kit (CLONTECH). To construct the plasmid coding for Myc-LIMK1(PK) containing amino acid residues 267–647, the cDNA fragment was amplified by PCR, using primers 5′-CTAGCTCGCCACCATGGGATACCCATACGATGTTCCAGATTACGCTGGATCCAGATCTGGGCCTGAGACCAGCCCC-3′ and 5′-TCCCGCGGAGGAATCTGG-3′. The PCR-amplified fragments were digested with BglII and SacII and ligated into the BglII and SacII sites of the pMYC-C1 vector. The plasmid coding for LIMK1-YFP was constructed by doubly inserting the PCR-amplified YFP cDNA into the XbaI and SalI sites of pMYC-C1-Myc-LIMK1. To construct the plasmid coding for PAK-AI (an autoinhibitory domain of PAK3, amino acid residues 78–146), the cDNA fragment was PCR amplified, using primers 5′-CAGAGCGGCCGCATACGATTCATGTGGGGTTT-3′ and 5′-GTATGCGGCCGCACTTTTATCTCCTGATGTAA-3′, and PAK3 cDNA as a template (provided by Dr. H. Sumimoto, Kyushu University, Fukuoka, Japan), digested with NotI, and subcloned into the NotI site of pCAG vector (17Ohashi K. Nagata K. Maekawa M. Ishizaki T. Narumiya S. Mizuno K. J. Biol. Chem. 2000; 275: 3577-3582Abstract Full Text Full Text PDF PubMed Scopus (421) Google Scholar). The authenticity of plasmids was confirmed by nucleotide sequence analysis. HeLa cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS). Cells were synchronized at beginning of the S phase using the double thymidine block method (26Hirota T. Morisaki T. Nishiyama Y. Marumoto T. Tada K. Hara T. Masuko N. Inagaki M. Hatakeyama K. Saya H. J. Cell Biol. 2000; 149: 1073-1086Crossref PubMed Scopus (176) Google Scholar, 27Kimura K. Tsuji T. Takada Y. Miki T. Narumiya S. J. Biol. Chem. 2000; 275: 17233-17236Abstract Full Text Full Text PDF PubMed Scopus (145) Google Scholar). In brief, cells were cultured in DMEM containing 10% FCS and 1 mm thymidine for 24 h, incubated in fresh DMEM containing 10% FCS for 8 h, and again cultured in thymidine-containing medium for 16 h. For cell cycle progression, cells synchronized at the beginning of S phase were then released for 6–12 h in fresh DMEM containing 10% FCS. To synchronize cells at mitosis, cells were synchronized at the beginning of S phase by a single thymidine block and cultured in fresh DMEM containing 10% FCS for 6 h and then cultured for 6 h in the presence of 100 ng/ml nocodazole. For transient transfection experiments, HeLa cells were transfected using the LipofectAMINE method (Invitrogen), following the manufacturer's protocol, cultured in the thymidine-containing medium for 24 h, and released for 6 h in fresh DMEM containing 10% FCS. Nocodazole, a microtubule-depolymerizing agent, was added to the medium, and the culture was continued for another 6 h. Mitotic cells that rounded up and adhered weakly on the culture dish were collected selectively by the mechanical shake-off procedure, whereas the attached interphase cells were harvested by scraping them off. To prepare cells synchronized at later stages of mitosis, mitotic cells collected by nocodazole arrest were washed with phosphate-buffered saline (PBS) to remove nocodazole, suspended in DMEM containing 10% FCS, plated into poly-l-lysine-coated culture dishes or coverslips then incubated at 37 °C to allow for cell cycle progression. At 0, 45, 90, 180 min after release from nocodazole arrest, the cells were harvested and subjected to analyses. For flow cytometry, cells were trypsinized, fixed with 70% methanol, stained with propidium iodide, and analyzed by a flow cytometer EPICS XL-MCL (Beckman Coulter). COS-7 cells were cultured in DMEM supplemented with 10% FCS. Cells were transiently transfected using LipofectAMINE. Three h after transfection, the medium was changed to normal cultivation medium. After an additional 36 h culture, cells were harvested and subjected to analyses. Rabbit anti-LIMK1 antibody (C-10) was raised against the C-terminal peptide of LIMK1, as described (10Okano I. Hiraoka J. Otera H. Nunoue K. Ohashi K. Iwashita S. Hirai M. Mizuno K. J. Biol. Chem. 1995; 270: 31321-31330Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar). An anti-cofilin antibody (COF-1) was prepared by immunizing rabbits with His6-cofilin, which was purified from lysates ofEscherichia coli expressing it, using as nickel-nitrilotriacetic acid-agarose (Qiagen, Valencia, CA). Rabbit anti-P-cofilin antibody specific to the Ser-3-phosphorylated forms of cofilin and ADF was raised against the phosphopeptide acetyl-A(pS)GVAVSDC and purified, as described (28Toshima J. Toshima J.Y. Amano T. Yang N. Narumiya S. Mizuno K. Mol. Biol. Cell. 2001; 12: 1131-1145Crossref PubMed Scopus (222) Google Scholar). Anti-Myc epitope monoclonal antibody (9E10) and anti-HA epitope monoclonal antibody (12CA5) were purchased from Roche Molecular Biochemicals. Cells were washed three times with ice-cold PBS, suspended in the lysis buffer (50 mmTris-HCl, pH 7.5, 150 mm NaCl, 1% Nonidet P-40, 1 mm dithiothreitol, 10 mm NaF, 20 mmβ-glycerophosphate, 1 mm phenylmethylsulfonyl fluoride, 10 μg/ml leupeptin, and 2 μg/ml pepstatin A), and incubated on ice for 30 min. After centrifugation, protein concentrations in cell lysates were determined in a Micro BCA protein assay (Pierce), and equal amounts of protein were precleared with protein A-Sepharose (Amersham Biosciences) at 4 °C for 1 h. The supernatants were incubated overnight at 4 °C with anti-LIMK1 (C-10) antibody or 9E10 anti-Myc antibody, and protein A-Sepharose. After centrifugation, the immunoprecipitates were washed three times with lysis buffer and used for in vitro kinase reaction and immunoblot analysis. For immunoblot analysis, cell lysates or immunoprecipitated proteins were separated on SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Bio-Rad). The membrane was blocked overnight with 4% nonfat dry milk in PBS containing 0.05% Tween 20 and incubated for 2 h at room temperature with primary antibody diluted in PBS containing 1% nonfat dry milk and 0.05% Tween 20. After washing in PBS containing 0.05% Tween 20, the membrane was incubated with horseradish peroxidase-conjugated donkey anti-rabbit IgG or sheep anti-mouse IgG (Amersham Biosciences). Immunoreactive protein bands were visualized using an ECL reagent (Amersham Biosciences). The immunoprecipitates were washed three times with lysis buffer and then three times with kinase buffer (20 mm Hepes-NaOH, pH 7.2, 5 mmMgCl2, 5 mm MnCl2, 1 mmdithiothreitol, 10 mm NaF, 20 mmβ-glycerophosphate, 1 mm phenylmethylsulfonyl fluoride, 10 μg/ml leupeptin, and 2 μg/ml pepstatin A) and incubated for 1 h at 30 °C in 30 μl of kinase buffer containing 50 μm ATP, 5 μCi of [γ-32P]ATP (3,000 Ci/mmol, Amersham Biosciences) and 2 μg of His6-cofilin. The reaction mixture was solubilized in Laemmli's sample buffer (50 mm Tris-HCl, pH 6.8, 10% glycerol, 1 mmdithiothreitol, 1% SDS, 0.002% bromphenol blue) for 5 min at 95 °C, and aliquots were separated on SDS-PAGE, using 15 and 8% gels. Proteins were transferred onto polyvinylidene difluoride membranes. The membrane from a 15% gel was subjected to autoradiography to measure 32P-labeled cofilin, using the BAS1800 Bio-Image Analyzer (Fuji Film, Tokyo, Japan), and Amido Black staining. The membrane from the 8% gel was analyzed by immunoblotting with the C-10 anti-LIMK1 antibody or the 9E10 anti-Myc antibody to detect LIMK. The kinase activity was normalized by dividing the radioactivity incorporated into cofilin by the immunoreactive density of LIMK estimated using a densitometer. Anti-LIMK1 immunoprecipitates were washed three times with lysis buffer without phosphatase inhibitors and then three times with phosphatase buffer (1 mmMgSO4 and 100 mm Tris-HCl, pH 8.0) and treated for 1 h at 25 °C with 20 units of calf intestinal alkaline phosphatase (Takara Biochemicals, Tokyo) in 30 μl of phosphatase buffer, then the immunoprecipitates were washed three times with lysis buffer containing phosphatase inhibitors and three times with kinase buffer and subjected to in vitro kinase reaction, as described above. HeLa cells were fixed in 4% paraformaldehyde in PBS for 20 min and permeabilized with absolute methanol for 10 min at −20 °C. After blocking with 1% bovine serum albumin in PBS for 30 min, cells were stained with anti-P-cofilin antibody followed by staining with rhodamine-conjugated anti-rabbit IgG (Chemicon). 4,6-Diamidino-2-phenylindole (DAPI, Molecular Probes) was used for DNA staining. After washing with PBS, coverslips were mounted on a glass slide and images were obtained using a Leica DMLB fluorescence microscope. To investigate the role of LIMK1 during the cell cycle, we first examined the kinase activity of LIMK1 in different stages of the cell cycle. For cell cycle analysis, HeLa cells were synchronized at the beginning of S phase by double thymidine block. At different times after release from double thymidine block, cells were lysed, and endogenous LIMK1 was immunoprecipitated and subjected to in vitro kinase assay, using His6-cofilin as a substrate. Cell cycle progression was monitored by flow cytometry (Fig.1A) and DAPI staining (data not shown). As shown in Fig. 1B, LIMK1 prepared from cells at 9 h after release from the thymidine block, when about 40% of the cells were in mitotic phase (as measured by DAPI staining), had 2.6-fold higher kinase activity than that of LIMK1 from cells at 6 h after release, when about 90% of the cells were in S phase (as determined using flow cytometry). At 12 h after release, about 80% of the cells had gone through cytokinesis and entered the G1 phase, and here the kinase activity of LIMK1 reverted to the level seen in LIMK1 in cells at 6 h. Immunoblot analysis revealed that slow migrating bands of LIMK1 appeared at 9 h, although the total amount of LIMK1 remained unchanged throughout the cell cycle. The gel migration shift and activation of LIMK1 became prominent when cells were more stringently synchronized at the early mitotic phase by treatment with the microtubule-depolymerizing agent, nocodazole (Fig.1C). LIMK1 from cells in the early mitotic phase had a 5.8-fold higher kinase activity than in the interphase. Mitotic LIMK1 significantly retarded mobility on gel electrophoresis compared with interphase LIMK1 (Fig. 1C). To determine whether the activation and mobility shift of mitotic LIMK1 were related to the phosphorylation, LIMK1 was treated with calf intestinal alkaline phosphatase. This treatment abrogated the mobility shift and activation of mitotic LIMK1 (Fig. 1D). Effects of phosphatase were nil when phosphatase inhibitors were present (data not shown). These results suggest that the slow migrating bands are the phosphorylated forms of LIMK1 and that LIMK1 is specifically activated and phosphorylated in the mitotic phase. To search for the mechanism of mitosis-specific activation and gel mobility shift of LIMK1, Myc-tagged wild-type LIMK1 and its kinase-inactive mutant LIMK1(D460A), in which the catalytic residue Asp-460 is replaced by alanine, were transiently expressed in HeLa cells, and these cells were synchronized at the mitotic phase, using nocodazole. As shown in Fig.2A, the transiently expressed Myc-LIMK1 in mitotic cells showed a gel mobility shift and a 6.4-fold increase in the kinase activity compared with Myc-LIMK1 in interphase cells. Thus, the transiently expressed Myc-LIMK1 behaves similarly to endogenous LIMK1. In both interphase and mitotic cells, Myc-LIMK1(D460A) immunoprecipitated from HeLa cells exhibited no cofilin phosphorylating activity, which suggests that Myc-LIMK1 immunoprecipitates were probably not contaminated with other cofilin-phosphorylating kinases. Similar to Myc-LIMK1, Myc-LIMK1(D460A) was retarded on gel electrophoresis, which means that autophosphorylation of LIMK1 may not be involved in mobility shift of mitotic LIMK1. Myc-LIMK2 transiently expressed in HeLa cells exhibited a 1.8-fold increase in the kinase activity in mitotic phase, but the gel mobility shift of Myc-LIMK2 was not visible after cells had been treated with nocodazole (Fig. 2B). Thus, it is likely that LIMK2 is at least in part involved in cofilin phosphorylation during mitosis. To determine the region involved in the mitosis-specific phosphorylation and activation of LIMK1, an expression plasmid coding for an N-terminally truncated mutant of LIMK1, Myc-LIMK1(PK) (Fig.2C), was constructed and transfected into HeLa cells. As reported (29Nagata K. Ohashi K. Yang N. Mizuno K. Biochem. J. 1999; 343: 99-105Crossref PubMed Google Scholar), in interphase cells, LIMK1(PK) exhibited significant increases in kinase activity compared with the activity of wild-type LIMK1 (Fig. 2D). Although the LIMK1(PK) mutant was activated further by ROCK in in vitro experiments (17Ohashi K. Nagata K. Maekawa M. Ishizaki T. Narumiya S. Mizuno K. J. Biol. Chem. 2000; 275: 3577-3582Abstract Full Text Full Text PDF PubMed Scopus (421) Google Scholar), it was not activated further in mitotic phase, and a gel mobility shift of mitotic LIMK1(PK) mutant was never observed (Fig. 2D). Thus, the N-terminal region of LIMK1 containing LIM and PDZ domains may be involved in the mitosis-specific mobility shift (phosphorylation) and activation of LIMK1. We attempted to show the gel mobility shift of the N-terminal fragment of LIMK1 in mitosis, but it was not successful because the protein band of the fragment expressed in HeLa cells was not detectable in mitotic cell lysates, probably because of its instability. Previous studies revealed that LIMK1 is activated by protein kinases, ROCK and PAK, through phosphorylation at Thr-508 within the kinase domain of LIMK1 (15Edwards D.C. Sanders L.C. Bokoch G.M. Gill G.N. Nat. Cell Biol. 1999; 1: 253-259Crossref PubMed Scopus (846) Google Scholar, 16Maekawa M. Ishizaki T. Boku S. Watanabe N. Fujita A. Iwamatsu A. Obinata T. Ohashi K. Mizuno K. Narumiya S. Science. 1999; 285: 895-898Crossref PubMed Scopus (1288) Google Scholar, 17Ohashi K. Nagata K. Maekawa M. Ishizaki T. Narumiya S. Mizuno K. J. Biol. Chem. 2000; 275: 3577-3582Abstract Full Text Full Text PDF PubMed Scopus (421) Google Scholar). We therefore asked whether ROCK and/or PAK is involved in the activation and mobility shift of LIMK1 in mitosis. To examine the role of ROCK in this regard, HeLa cells were treated with Y-27632, a specific inhibitor of ROCK (30Uehata M. Ishizaki T. Satoh H. Ono T. Kawahara T. Morishita T. Tamakawa H. Yamagami K. Inui J. Maekawa M. Narumiya S. Nature. 1997; 389: 990-994Crossref PubMed Scopus (2545) Google Scholar), and the activity and gel mobility of mitotic LIMK1 were analyzed using in vitro kinase assay and immunoblotting. Gel retardation of mitotic LIMK1 was observed in the presence of Y-27632 (Fig. 3A).In vitro kinase assay revealed that the kinase activity of mitotic LIMK1 was only faintly reduced by Y-27632 treatment (Fig.3A). Thus, a protein kinase(s) other than ROCK seems to be involved in the mitosis-specific mobility shift and activation of LIMK1. We next examined the role of PAK in mitosis-specific activation and mobility shift of LIMK1. As an inhibitor for PAK, we used an autoinhibitory domain of PAK (PAK-AI). As reported (15Edwards D.C. Sanders L.C. Bokoch G.M. Gill G.N. Nat. Cell Biol. 1999; 1: 253-259Crossref PubMed Scopus (846) Google Scholar), expression of PAK-AI inhibited the activation of LIMK1 induced by RacV12, a constitutively active form of Rac (Fig. 3B), thereby indicating that PAK-AI has the potential to inhibit PAK activity in cultured cells. Because the coexpression of PAK-AI and LIMK1 into HeLa cells did not interfere with mobility shift or activation of LIMK1 in mitosis (Fig. 3C), PAK does not appear to have a role in this regard. To determine further whether phosphorylation of Thr-508 by ROCK, PAK, or other protein kinases is involved in the mitotic mobility shift and activation of LIMK1, we expressed in HeLa cells LIMK1(T508V), in which Thr-508 is replaced by nonphosphorylatable valine, and analyzed its gel mobility and kinase activity in mitotic phase. Similar to wild-type LIMK1, LIMK1(T508V) in mitosis migrated slowly on the gel (Fig.3D), thereby clearly indicating that phosphorylation at Thr-508 is not related to the mobility shift of mitotic LIMK1 and that phosphorylation of residue(s) other than Thr-508 is responsible for the gel retardation. As reported (31Edwards D.C. Gill G.N. J. Biol. Chem. 1999; 274: 11352-11361Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar), replacing Thr-508 by Val significantly reduced the kinase activity of LIMK1 (Fig.3D). However, kinase activity of the LIMK1(T508V) mutant increased about 3-fol
Referência(s)