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

J. Biol. Chem., Vol. 277, Issue 46, 44220-44228, November 15, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/46/44220    most recent M206504200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Katsu, R. Articles by Hagiwara, M. Search for Related Content PubMed PubMed Citation Articles by Katsu, R. Articles by Hagiwara, M. Social Bookmarking                      What's this?

Novel SR-rich-related Protein Clasp Specifically Interacts with Inactivated Clk4 and Induces the Exon EB Inclusion of Clk^*

Rieko Katsu, Hiroshi Onogi, Kazuhiro Wada, Yasushi Kawaguchi^§, and Masatoshi Hagiwara^¶

From the Departments of  Functional Genomics and ^§ Cell Regulation, Medical Research Institute, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, Japan

Received for publication, July 1, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We identified a novel serine/arginine (SR)-rich-related protein as a binding partner of Clk4 mutant lacking kinase activity (Clk4 K189R) in the two-hybrid screen and designated it Clasp (Clk4-associating SR-related protein). Northern blot analysis revealed that Clasp mRNA was highly expressed in brain, and in situ hybridization of a mouse brain sagittal section hybridized with antisense probes revealed that both Clasp and Clk4 mRNAs were expressed in the hippocampus, the cerebellum, and the olfactory bulb. Two forms of Clasp were produced by a frameshift following alternative splicing. The staining of an HA-tagged short form of Clasp (ClaspS) showed a nucleoplasmic pattern, while the long form of Clasp (ClaspL) was localized as nuclear dots. In vitro protein interaction assay demonstrated that Clk4 K189R was bound to Clasp while wild Clk4 was not. Overexpression of ClaspL promoted accumulation of Clk4 K189R in the nuclear dots and the exon EB inclusion from CR-1 and CR-2 pre-mRNA of Clk1. These data suggest that Clasp is a binding partner of Clk4 and may be involved in the regulation of the activity of Clk kinase family.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Serine/arginine (SR)-rich proteins are essential splicing factors (1) that promote splice-site recognition (2) at an early stage of spliceosome assembly and also influence the selection of alternative splice sites (3-5). They exist in phosphorylated forms in cells, and the phosphorylation state of SR proteins appears to influence their activities in general and alternative splicing (6, 7), as well as their subnuclear localization and nuclear-cytoplasmic transport properties (7-9). Nuclear speckles are subnuclear regions where SR proteins and other splicing components are concentrated, and they are thought to represent sites of storage or assembly for splicing factors (10). SR protein-specific kinase (SRPK)¹ 1 was originally identified as a kinase of SC35 in extracts from HeLa cells (11, 12). Addition of purified SRPK1 to permeabilized cells or overexpression of SRPK1, 2, or CDC2-like kinase (Clk)1 in transfected cells result in an apparent disassembly of the nuclear speckles (11, 13-16). These results suggest that phosphorylation or hyperphosphorylation causes release of these factors from the speckles or that perhaps that the integrity of these structures is compromised.

Clk1 was initially cloned as a CDC2-like kinase (17). Clk1 has an arginine/serine (RS) domain at its N terminus and interacts with several members of the SR protein family and SR-related protein in a yeast two-hybrid screen (13, 18). Clk1 phosphorylates SR proteins and affects SR protein-dependent splicing (6). In mammals, Clk is a member of a protein kinase subfamily that contains at least four isoforms Clk1, Clk2, Clk3, and Clk4 (19, 20). mRNAs for all four Clk isoforms are alternatively spliced to produce proteins in which the kinase domain is missing (16, 21). Clk1 was independently isolated with anti-phosphotyrosine antibody screens as a prototypical dual-specific kinase, termed STY (22, 23). Clk autophosphorylates with dual specificity in vitro, and the tyrosine phosphorylation of Clk1 was proposed to be an important determinant of the kinase activity as the majority of Clk1 activity can be immunoprecipitated with antibodies against phosphotyrosine (23). In the case of Clk2, tyrosine phosphorylation of endogenous Clk2 was undetectable in Friend murine erythroleukemia cells (24). Actually, incubation of Clk1 with tyrosine phosphatases did not affect the kinase activity, while treatment of Clk1 with the serine/threonine-specific phosphatase 2A resulted in a 2- to 6-fold increase in enzymatic activity (25). The Ser-141 of Clk2 was identified as an autophosphorylation site (24), which is highly conserved among all Clk proteins. Site-directed mutation of this autophosphorylation site influenced the subnuclear localization of the kinase (24). The N-terminal domain of Clk1 containing this autophosphorylation site has been proposed to comprise a putative regulatory domain and includes a bipartite nuclear localization signal (21, 25). The N-terminal truncation of Clk1 resulted in a 45-fold increase in V_max with no change of K_m, suggesting that this domain does not contain a pseudo-substrate motif but may act to conformationally constrain the catalytic activity of enzyme (25). These data gave us an idea that Clk is associated with a regulatory protein in a phosphorylation-dependent manner. Colwill et al. (13) and Nayler et al. (20) observed that overexpressed Clk proteins were largely found in the soluble fraction of a Triton X-100 cell lysis buffer, whereas the catalytically inactive mutant kinases were found almost exclusively in the pellet, suggesting the presence of nuclear anchoring protein(s) that specifically recognize the inactive Clk. Actually Nayler et al. (26) demonstrated that Clk2 associates with nuclear scaffold attachment factor, although it is not clear whether the interaction is regulated by phosphorylation or not. We recently found that SRPK interacts with SF2/ASF only when its RS domain was not phosphorylated (27). In this paper we cloned a novel nuclear protein Clasp as a binding partner of inactivated Clk4 by a two-hybrid screen. As ClaspL could promote exon EB inclusion of Clk and increase the kinase active form of Clk in vivo, Clasp may play a regulatory role on Clk.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

DNA Subcloning-- Mouse Clk4 cDNA was amplified by reverse transcription (RT)-PCR using a forward primer incorporating a SalI restriction site at the 5' end and a reverse primer incorporating a NotI restriction site at the 5' end, followed by subcloning into pGEM-T Easy vector (Promega). The Clk4 K189R has a lysine to arginine mutation in the ATP-binding site introduced with the in vitro Mutagenesis Kit (Takara) according to the manufacturer's instructions. To construct pME-hemagglutinin (HA)-Clk4 wild type (WT) and pME-HA-Clk4 K189R, pGEM-T-cloned vectors were digested with SalI-NotI, releasing the coding region and then ligated into SalI-NotI-digested pME-HA (28). Construction of pME-HA-SRPK2 and pME-HA-SRPK2 K108R were described previously (27). To generate pHybcIHK-Clk4 WT, pHybcIHK-Clk4 K189R, pHybcIHK-SRPK2 WT, and pHybcIHK-SRPK2 K108R, the pME-HA-ligated vectors were digested with NotI (Clks) or SacII (SRPKs), blunt-ended with Klenow (NotI) or T4-DNA polymerase (SacII), digested with XhoI (Clks) or SalI (SRPKs), and inserted into the XhoI-SalI blunt-ended vector of pHybcIHK (Invitrogen). To generate all truncated Clasp, constructs were amplified from pB42AD-clone79 obtained by two-hybrid screening with a forward primer having an EcoRI restriction site and a reverse primer containing a XhoI restriction site, digested with EcoRI-XhoI, and ligated into the EcoRI-XhoI site of pB42AD. The full-length ClaspS and ClaspL cDNAs were obtained by RT-PCR with the forward primers that have first methionine and EcoRI-SalI site and the reverse primers containing the stop codons and NotI-XhoI site using a mouse total brain cDNA as a template. The products of about 1.7 kb (ClaspS) and 2.0 kb (ClaspL) were subcloned into pGEM-T Easy vector. To generate pME-HA-, pME-glutathione S-transferase (GST)-, and pET28 (Qiagen) -Clasp constructs, pGEM-T-ClaspS and pGEM-T-ClaspL were digested with SalI-NotI and the released full-length cDNAs of ClaspS or -L were ligated into SalI-NotI digested vectors. To construct pFastBacHT-Clasp, pME-GST-ClaspS and pME-GST-ClaspL were digested with XhoI-NotI and ligated into the SalI-NotI site of pFastBacHT (Invitrogen). To generate pGEX-Clk4 WT and pGEX-Clk4 K189R, pGEM-T-Clk4 WT and pGEM-T-Clk4 K189R were digested with SalI-NotI and released cDNAs were ligated into SalI-NotI-digested pGEX5X-3 (Amersham Biosciences). The N (1-127), C-WT (128-481), and C-K189R (128-481) regions of Clk4 were amplified by PCR with a forward primer having a SalI restriction site and a reverse primer containing a NotI restriction site using pGEM-T-Clk4 WT or pGEM-T-Clk4 K189R as templates, and the PCR fragments were digested with SalI-NotI and ligated into the SalI-NotI site of pGEX5X-3. The plasmids of cytomegalovirus (CMV) CR-1 and CR-2 were kindly gifted from Dr. J. C. Bell (Ottawa). All PCR products were sequenced by Dye Terminator Cycle Sequencing Ready Reaction (PE Applied Biosystems) and ABI PRISM 310 Genetic Analyzer (PE Applied Biosystems) to confirm their sequence integrity.

Two-hybrid Screening-- Yeast strain SKY48/pLacGUS (Invitrogen) was transformed with pHybcIHK-Clk4 K189R. Expression of the construct was confirmed by Western blotting with cI antibody (Invitrogen). The resulting strain was co-transformed with Mouse Brain MATCHMAKER LexA cDNA library (Clontech) in which cDNAs were fused to the coding sequence for B42 DNA activation domain and transformants were selected by the use of appropriate media according to the manufacturer's instructions. Putative interacting clones were purified in bacteria selecting for their ampicillin resistance and retransformed into the SKY48/pLacGUS with pHybcIHK or pHybcIHK-Krev (Invitrogen) to eliminate false positives.

Preparation of Recombinant Proteins-- GST-Clk4 and its derivatives were expressed in Escherichia coli and purified as described before (28). For preparation of His-GST-tagged Clasp, we used the Bac-to-Bac Expression System (Invitrogen). DH10Bac, the host strain of E. coli was transformed with the plasmid of pFastBacHT-ClaspS or pFastBacHT-ClaspL, and the recombinant bacmid was prepared. Sf9 cells were cultured in Sf-900II serum-free medium (Invitrogen) supplemented with 10% fatal bovine serum (FBS) and transfected with recombinant bacmid. After 6 days, the viral supernatant was collected, and the virus was amplified twice. For protein preparation, Sf9 cells were infected, lysed at 3 days postinfection, harvested, resuspended in buffer A (20 mM Tris-HCl, pH 8.0, 300 mM NaCl, 10 mM NaF, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1% Triton X-100, 10 µg/ml aprotinin, 5 µg/ml leupeptin, 10 mM imidazole), and rotated for 30 min at 4 °C. Insoluble material was collected by centrifugation and solubilized in buffer B (8 M urea, 10 mM Tris-HCl, pH 8.0, 200 mM NaCl, 10 mM imidazole), sonicated, and subsequently rotated 2 h at room temperature. After centrifugation, the supernatant was incubated with nickel-nitrilotriacetic acid-agarose (Qiagen) for 6 h at room temperature. The resin was washed five times with lysis buffer B. The resin-bound protein was eluted with elution buffer (250 mM imidazole in lysis buffer B). For renaturation of Clasp protein, the eluted protein was dialyzed in 500 ml of buffer C (4 M urea, 10 mM Tris-HCl, pH 8.0, 200 mM NaCl, 1 mM PMSF) for 2 h. Subsequently 5 liters of buffer D (20 mM Tris-HCl, pH 8.0, 100 mM KCl, 0.2 mM EDTA, 5% glycerol, 1 mM dithiothreitol (DTT), 0.5 mM PMSF) was added at a speed of 4 ml/min. Finally, the protein was dialyzed against buffer D. All steps in the renaturation of Clasp proteins were performed at 4 °C. Any aggregated material formed during the dialysis was removed by centrifugation. The clarified supernatant was stored at 80 °C until required.

Northern Hybridization Analysis and in Situ Hybridization-- Clasp full-length cDNA probe or -actin probe were labeled with radioactive [-³²P]dCTP using the multiprime DNA labeling system from Amersham Biosciences. A mouse Multiple Tissue Northern blot (Clontech) membrane was used. Hybridization was carried out overnight at 42 °C in hybridization mixture (6× SSC (1× SSC is 150 mM NaCl plus 15 mM sodium citrate), 50% formamide, 1% SDS, 1× Denhardt's solution, 10% dextran sulfate, 100 µg of denatured salmon sperm DNA per ml). The membrane was washed three times at room temperature in washing buffer (0.1× SSC, 0.1% SDS), and the hybridized transcripts were observed with a BAS2000 image analyzer (Fuji Film). For in situ hybridization, three male mice (ICR, 4 weeks after birth) were used in this study. Clasp and Clk4 full-length cDNA were used to make sense and antisense riboprobes. In situ hybridization was carried out as described with a few modifications (29). Briefly, frozen sections (10-µm thick) were fixed in 3% paraformaldehyde in phosphate buffered saline (PBS) and acetylated. The sections were dehydrated in an ascending ethanol series and air-dried. The ³⁵S-labeled probes (5× 10⁵ dpm/glass) were dissolved in a buffer containing 10 mM Tris-HCl, pH 8.0, 30 mM NaCl, 12 mM EDTA, 10 mM DTT, 50% formamide, 10% dextran, 1× Denhardt's solution and 0.5 mg/ml yeast tRNA. 100 µl of probe solution was applied to each slide for 16 h at 65 °C. Washing was conducted in 2× SSPE containing 50% formamide twice for 30 min each time and then in 0.2× SSPE, 30 min each for two times all at 65 °C before finally being dehydrated again. The sections were exposed to x-ray films (Kodak) for 2 days, the films were developed, and then slides were exposed to NTB-2 (Kodak) emulsion for 4-5 weeks. After development, sections were counterstained with crystal violet and coverslipped. For image analysis of hybridized sagittal brain sections, x-ray film autoradiograms and brain sections developed with silver grains were imaged with a charge-coupled device camera (DIAGNOSTIC instruments) coupled to a microscope (Leica DM RXA2).

Immunofluorescence Assay-- COS-7 cells were maintained in Dulbecco's modified Eagle medium (Sigma) supplemented with 10% FBS. For transfection, COS-7 cells were grown on coverslips and transfected with mammalian expression vectors. After culturing for 24 h, cells were washed in PBS twice, fixed with 4% paraformaldehyde in PBS for 10 min at room temperature, and washed twice in PBS. The fixed cells were permeabilized with the 0.2% Triton X-100 in PBS for 3 min at room temperature and washed four times in PBS. The cells were incubated in blocking solution (10% FBS in PBS) for 2 h at room temperature. The coverslips were incubated with 25 µg/ml of monoclonal anti-HA antibody (12CA5, Roche Molecular Biochemicals) in blocking solution overnight at 4 °C. After washing five times in PBS, the coverslips were incubated with 1:50 dilution of goat anti-mouse IgG Texas Red-Conjugated (Southern Biotechnology Associates, Inc.) in blocking solution for 4 h at room temperature. The stained coverslips were washed four times in PBS, then dried and mounted with PermaFluor (Lipahw Immunon). The images were visualized by confocal microscopy (Radiance, BioRad).

In Vitro Binding Assay-- For in vitro binding assays with GST fusion proteins, 3 µg of the GST fusion proteins were incubated with 30 µl of glutathione-Sepharose beads in binding buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.2 mM EDTA, 1 mM DTT, 1 mM PMSF, 1% Triton X-100) for 2 h at 4 °C. All the beads were washed three times with the binding buffer. In vitro translated His-T7-tagged proteins were prepared in a rabbit reticulocyte-coupled transcription and translation system (Promega) using either ClaspS or ClaspL in pET28 with T7 RNA polymerase. The 30 µl of in vitro translated His-T7-tagged proteins were added to glutathione-Sepharose-bound GST fusion proteins in 1 ml of binding buffer and rotated overnight at 4 °C. All the beads were washed three times in a binding buffer, and the bound proteins were eluted in 2× SDS sample buffer before separation by 8.5% SDS-PAGE. As a standard, one-twelfth of the His-T7-tagged translated products before pull-down were run in parallel on the gel. Bound proteins were analyzed by immunoblotting with anti-T7 monoclonal antibody (Novagen). The blot was stripped and then reprobed with anti-GST polyclonal antibody (Santa Cruz) to examine GST fusion proteins. Western blot analysis was performed by using the enhanced chemiluminescence detection system (Amersham Biosciences).

In Vitro Kinase Assay-- In the protein kinase assays employed in these studies we utilized 0.5 µg of bacterially expressed recombinant Clk4 in the presence of 0.5 µg of baculovirus-expressed recombinant Clasp as a substrate. The reactions were carried out for 30 min at 30 °C in 40 ml of kinase buffer (40 mM HEPES-KOH, pH 7.8, 10 mM MgCl₂, 2 mM DTT, 1 mM NaVO4, 1 mM EGTA, 12 mM -glycerophosphate) containing 10 µCi of [-³²P]ATP. The reaction was stopped by boiling with an equal volume of 2× SDS sample buffer, and the proteins were separated on 8.5% SDS-PAGE. The gel was stained with Coomassie Brilliant Blue, dried, and subjected to autography.

Alternative Splicing Assay-- CMV CR-1 and CMV CR-2 plasmids were introduced into COS-7 cells (60-mm diameter dish) with indicated amounts of expression vectors of Clk4, ClaspL, or ClaspS using LipofectAMINE (Invitrogen) as per the manufacture's instructions. Cells were harvested at 24 h posttransfection, and total RNA and protein were extracted using ISOGEN (Nippon Gene). The 5 µg of RNA was used for RT and 1/50 (100 ng) was used for PCR amplification with the primers as reported by Duncan et al. (30). PCR conditions, including the number of cycles and template concentrations, were optimized to maintain the linearity during amplification. PCR products were separated on 1.5% agarose gel and stained with ethidium bromide. Simultaneously prepared protein samples were separated on 9% SDS-PAGE. Bound proteins were analyzed by immunoblotting with anti-Myc monoclonal antibody (9E10, Santa Cruz), anti-HA polyclonal antibody (MBL) and anti-actin polyclonal antibody (Sigma).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Clk4 Interacts with a Novel Protein That Has an Arginine/Serine Dipeptide Repeat-- To search for regulatory proteins of Clk, yeast two-hybrid screen (31) was performed. Full-length cDNA of mouse Clk4 K189R was fused to a cDNA encoding the cI repressor DNA binding domain to generate the pHybcIHK-Clk4 K189R construct as a bait of this two-hybrid screen. Ten million colonies containing both Clk4 and mouse brain cDNA library plasmids were screened for stimulation of the reporter LYS2 gene. To eliminate false positives, it was confirmed that the encoded proteins did not stimulate the reporter for themselves. The reporter was activated when the yeast strain SKY48/pLacGUS was transformed with one of the positive cDNA clones (clone79) and pHybcIHK-Clk4 K189R, though it was not stimulated with Clk4 WT or SRPK2 (another SR protein kinase as described, Kuroyanagi et al.) (14), indicating that the protein encoded by clone 79 specifically interacts with Clk4 K189R (Fig. 1A). The clone 79 was sequenced, and the resulting sequence partially overlapped with registered cDNAs (DDBL/EMBL/GenBank^TM accession no. NM016680 and AF042799). Comparing with the registered cDNAs, clone 79 seemed to lack the 127 amino acids of the N terminus. As this protein has two characteristic arginine/serine dipeptide repeat domains (RS1 and RS2 in Fig. 1B), we designated it Clasp (Clk4-associating SR-related protein). To identify the domain required for the binding to Clk4 K189R, we prepared its truncated mutants and examined the interaction with Clk4 K189R in the two-hybrid assay (Fig. 1B). Clk4 K189R interacted with clone 79 (Clasp 128-573), Clasp 249-573, and Clasp 321-573 but did not interact with Clasp 354-573 and Clasp 354-527, indicating that both regions of amino acids 321-354 and 527-273 of Clasp are required for the interaction.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Interaction of Clasp with Clk4 in the yeast two-hybrid. A, interaction between clone 79 (Clasp) and Clk4 or its KR mutant. The transformants were selected on His-Trp-plate (left) and interactants were subsequently selected on His-Trp-Lys-galactose/raffinose plate (right). B, schematic map of the interaction domain of clone 79 (Clasp) in yeast. Deletion clones were created at the indicated amino acid regions as described under "Experimental Procedures." The minimum region of Clasp required for interaction with mClk4 K189R was mapped in the two-hybrid analysis. RS, arginine/serine repeats domain. A, alanine repeat region.

Clasp Has Two Alternative Spliced Forms-- In the course of the sequencing of full-length cDNA of Clasp, we obtained two independent clones that differ in the presence or absence of four bases insertion (the portion between two triangles in Fig. 2A), which are predicted to have differential C-terminal ends due to the frameshift. The clone 79 obtained in our yeast two-hybrid screen was corresponding to the short form (ClaspS). The long form Clasp (ClaspL) has an additional RS domain (RS3, striped in Fig. 2B) on its C terminus. Schematic diagrams of the ClaspS and ClaspL are shown in Fig. 2B. Because the nucleotide sequences of ClaspS and ClaspL at the N terminus are identical, they seemed to be driven from one gene by an aberrant alternative splicing. Because two isoforms of Clasp proteins had characteristic RS repeat motifs, we next introduced the HA-tagged expression vectors of ClaspS and ClaspL into COS-7 cells and checked the subcellular distribution of these proteins under a confocal microscope. As shown in Fig. 2C, HA-tagged ClaspS was located in the nucleoplasm, while HA-tagged ClaspL was concentrated in the nuclear dots with the clear margins. Characteristic amino acid sequences of ClaspL at their C terminus may affect the intranuclear localization of the protein.

View larger version (53K): [in this window] [in a new window]   Fig. 2.   Sequence and schematic illustration of alternative splicing products of Clasp and their intracellular localization. A, cDNA and amino acid sequences of Clasp isoforms. Amino acid sequences of short form (ClaspS) and long form (ClaspL) produced by alternative splicing are respectively shown under the cDNA sequences. Alternative spliced sites are indicated by triangles (left for long form and right for short form of Clasp) and asterisks indicate the positions of stop codon. The RS domains are indicated by light gray shading, and SR or RS dipeptide repeats are shown in boldface type. Alanine repeat region is indicated by dark gray shading. The polyadenylation signal is boxed. Nucleotides and amino acids are numbered on the left and right, respectively. B, domain structures of ClaspS (upper) and ClaspL (lower) proteins. The RS motifs are striped, and alanine repeat regions are shown by gray shading. C, subcellular localization of ClaspS and ClaspL. COS-7 cells were transiently transfected with pME-HA-ClaspS (left) or pME-HA-ClaspL (right) to observe the localization of two isoforms. These cells were fixed 24 h after transfection, and localization of HA-ClaspS or HA-ClaspL were visualized by indirect immunofluorescent staining with the anti-HA monoclonal antibody followed by Texas red-conjugated secondary antibody.

Clasp Is Highly Expressed in Brain and Testis-- We next examined the expression of Clasp mRNAs by Northern blot using a Mouse Multiple Tissue Northern blot membrane (Clontech). As shown in Fig. 3A, upper panel, major transcripts with approximate sizes of 2.7 and 2.5 kb were highly expressed in brain and moderately expressed in lung and liver. Smaller transcripts (2.2 and 1.2 kb) were specifically expressed in testis. In addition, two longer transcripts (~5 and 6 kb) were visible in heart, brain, and testis. These suggest to us that more transcripts than those coding ClaspS and ClaspL are transcribed from the Clasp gene. The same blot was reprobed with -actin as a control to verify that equivalent amounts of RNA were loaded (Fig. 3A, lower panel). As the expression patterns of 2.7- and 2.5-kb transcripts seems to correspond to that of Clk4 (20), we further examined the cell type-specific expression of Clasp and Clk4 by in situ hybridization. Sagittal section of a mouse brain hybridized with antisense probes revealed that both Clasp and Clk4 mRNAs were highly expressed in the hippocampus, the cerebellum, and the olfactory balb (Fig. 3B, a for Clasp, b for Clk4). Moreover, on the crystal violet staining, Clasp and Clk4 mRNA were expressed in neuron-like cells with high-volume nuclei in the hippocampus CA1 region (Fig. 3B, c for Clasp, d for Clk4) and cerebellum (Fig. 3B, e for Clasp, f for Clk4), supporting our hypothesis that Clasp is a binding partner of Clk4.

View larger version (49K): [in this window] [in a new window]   Fig. 3.   Expression of Clasp mRNA in tissues. A, Northern blot analysis of Clasp. The mouse Multiple Tissue Northern blot (Clontech) membrane was hybridized with a cDNA probe of Clasp isoforms (upper panel). The same filter was rehybridized with a -actin control probe (lower panel). Molecular size markers are shown on the left. B, Clasp and Clk4 expression pattern in the mouse brain. X-ray film autoradiograms negative image views of sagittal brain sections hybridized with 35S-radiolabeled riboprobes for Clasp (a) and Clk4 (b). White label is the mRNA signal. Scale bar in a for a and b = 1 mm. Higher magnification images of hippocampus CA1 region (c, Clasp; d, Clk4) and cerebellum (e, Clasp; f, Clk4) under the blight-field microscope. Scale bar in c for c-f = 20 µm. Cb, cerebellum; Ctx, cortex; Gr, cerebellum granule cells; Hp, hippocumcus; Mol, molecular layer of cerebellum; Ob, Olfactory bulb; P, Purkinje cells; Str, striatum; Tha, thalamus.

Clasp Directly Binds to Clk4 K189R-- To verify the two-hybrid interactions and the binding domain of Clk4 K189R to Clasp biochemically, we carried out an in vitro pull-down assay. In vitro translated His-T7-tagged ClaspS and ClaspL proteins were incubated with recombinant GST or GST-Clk4 and its truncated proteins (see Fig. 4A), respectively, and pulled down with glutathione-Sepharose beads. In this assay, ClaspS and ClaspL interacted efficiently with Clk4 K189R (Fig. 4B, lanes 4 and 11) but did not interact with Clk4 WT (lanes 3 and 10) as observed in the two-hybrid analysis. Deletion analysis of Clk4 revealed that Clasp was bound to the catalytic domain of the inactivated kinase (lanes 7 and 14).

View larger version (45K): [in this window] [in a new window]   Fig. 4.   Direct interaction of Clasp with Clk4. A, schematic map of Clk4 mutants used for the in vitro binding assay. B, identification of the binding domain of Clasp to Clk4. GST (lanes 2 and 9), GST-fused Clk4 WT (lanes 3 and 10), Clk4 K189R (lanes 4 and 11), Clk4 N (lanes 5 and 12), Clk4 C-WT (lanes 6 and 13), or Clk4 C-K189R (lanes 7 and 14) protein was incubated with glutathione-Sepharose beads. After washing two times, in vitro transcribed/translated His-T7-tagged ClaspS (lanes 2-7) or ClaspL (lanes 9-14) were added and incubated. After washing, the proteins retained on glutathione-Sepharose beads and one-twelfth of the input Clasp proteins (lanes 1 and 8) were resolved by 8.5% SDS-PAGE followed by immunoblotting using anti-T7 antibody (upper panel). The blot was subsequently stripped and reblotted with anti-GST antibody to confirm the bound GST-tagged protein (lower panel).

Clk4 Can Phosphorylate Clasp in Vitro-- In the two-hybrid assay and the in vitro pull-down assay, we suggested that Clasp is a binding partner of Clk4. Next we examined whether Clasp was phosphorylated by Clk4 in vitro. In the presence of ATP, both ClaspS and ClaspL prepared from the baculovirus-infected cells were phosphorylated by GST-Clk4 WT (Fig. 5, lanes 5 and 8) but not by Clk4 K189R (lanes 6 and 9) or GST alone (lanes 4 and 7), indicating that Clasp is one of the putative substrates of Clk4.

View larger version (52K): [in this window] [in a new window]   Fig. 5.   Phosphorylation of Clasp by Clk4. His-GST (lanes 1-3), His-GST-ClaspS (lanes 4-6), or His-GST-ClaspL (lanes 7-9) protein was incubated for 30 min at 30 °C with [-32P]ATP in kinase buffer in the presence of the GST-Clk4 WT (lanes 2, 5, and 8), GST-Clk4 K189R (lanes 3, 6, and 9), or GST (lanes 1, 4, and 7), respectively. The reaction was terminated, resolved by 8.5% SDS-PAGE, and visualized by an imaging analyzer (upper panel). The amounts of proteins (0.5 µg each) in the reactions were monitored by CBB staining (lower panel).

Subnuclear Localization of ClaspL Is Altered by the Clk4 Activity-- It has been reported that the catalytically inactive form of Clk co-localized with endogenous SR proteins in nuclear speckles and that overexpression of the wild type Clk caused redistribution of SR proteins in the nucleus (13). Therefore, we examined the effect of overexpressed Clk4 on the localization of Clasp. Green fluorescent protein (GFP)-Clk4 fluorescence was observed as a diffuse pattern (Fig. 6A) and the mutant, lacking the kinase activity, GFP-Clk4 K189R exhibited the speckled pattern (Fig. 6B) in the nuclei of COS-7 cells same as previously reported on Clk1 (13, 20). When GFP-Clk4 K189R was overexpressed in COS-7 cells with HA-tagged ClaspL, GFP fluorescence was overlapped with anti-HA staining, showing typical nuclear dots (Fig. 6, F-H) as observed (Fig. 2C, right panel). Overexpression of GFP-Clk4 dramatically changed the staining pattern of ClaspL (Fig. 5, C-E), indicating that ClaspL was disassembled from nuclear dots and subsequently redistributed by the kinase activity of Clk4.

View larger version (44K): [in this window] [in a new window]   Fig. 6.   Colocalization of Clk4 and Clasp in vivo. COS-7 cells were transfected with the expression vector encoding HA-tagged ClaspL together with the vector encoding GFP-tagged Clk4 WT or K189R. After 24 h incubation, the cells were fixed, immunostained with anti-HA monoclonal antibody and Texas red-conjugated secondary antibody, and then observed under confocal microscopy. The images were superimposed. The scale bar represents 10 µm. Shown are GFP-Clk4 WT only (A), GFP-Clk4 K189R only (B), GFP-Clk4 WT (C) plus HA-ClaspL (D) and merged (E), GFP-Clk4 K189R (F), plus HA-ClaspL (G) and merged (H).

ClaspL Promotes Exon EB Inclusion of Clk1 Pre-mRNA in Vivo-- It was shown previously that the expression of Clk1 itself was regulated through alternative splicing of the Clk1 pre-mRNA, yielding mRNAs encoding catalytically active and truncated inactive polypeptides (Clk1 and Clk1^T, respectively) (30). Because the disassemble of spliceosomal proteins induced by Clk in subnuclear regions affects alternative splicing patterns (30), we next examined the effect of Clasp overexpression on the CMV CR1 minigene (Fig. 7A, upper panel) expression. It contains two introns flanking alternative spliced exon EB and the ratio of exon inclusion (EB⁺)/exon skipping (EB) transcripts is regulated by the kinase activity of Clk1 (30). Overexpression of catalytically active Clk4 with this reporter favors skipping of exon EB (Fig. 7B, lane 3) as described using the Clk1 expression vector (30). Next we examined the possibility that ClaspL can affect the alternative splicing of Clk1. To test this idea, a fixed amount of CMV CR-1 was transfected along with increasing amounts of ClaspL expression vector, and we observed an increase in inclusion of exon EB (Fig. 7B, a-d, lanes 7-9). In contrast, increasing amounts of ClaspS expression vector did not alter the EB⁺/EB ratio (lanes 4-6). Because CR-1 reporter construct retains the coding capacity to produce truncated Clk1, we confirmed the effect of Clasp on CR-2 minigene (Fig. 7A, lower panel), which contains a translational stop codon preventing the production of any Clk-related proteins (30), and essentially the same results were obtained (Fig. 7C, a-c).

View larger version (43K): [in this window] [in a new window]   Fig. 7.   ClaspL promotes exon inclusion of Clk1 pre-mRNA in vivo. A, schematic representation of CR1 and CR2 minigenes. The primers used for amplification by RT-PCR are indicated. B, a, pattern of CR-1 alternative splicing upon cotransfection of ClaspS and ClaspL expression vectors in COS-7 cells. Lane 1, cells with filler DNA; lanes 2-9, 1 µg of CMV CR-1 plus 5 µg of pME-HA (lane 2), pME-HA-Clk4 WT (lane 3), 0.2 (lanes 4 and 7), 1.0 (lanes 5 and 8), 5.0 (lanes 6 and 9) µg of HA-ClaspS (lanes 4-6), HA-ClaspL (lanes 7-9), respectively. Positions of molecular size standards are indicated on the left of the panel. b, ratios for two alternative forms (EB+ and EB) of CR-1 mRNA shown in a. Lanes are as described for a. c, detection of Myc-tagged protein expressed from CR-1 reporter minigene by anti-Myc monoclonal antibody immunoblot analysis. Lanes are as described for a. Positions of molecular mass markers (in kilodaltons) are on the left of the panel. d, the blot was subsequently stripped and reblotted with anti-HA polyclonal antibody to confirm the expressed HA-tagged protein. e, the blot was subsequently stripped and reblotted with anti--actin polyclonal antibody to confirm equal loadings of proteins. C, pattern of CR-2 alternative splicing same as B.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Clasp cloned in our two-hybrid screening using Clk4 K189R as a bait has RS repeat motifs. As any RNA-recognition motif is not found in it, Clasp is not a member of SR protein family. Colwill et al. cloned SR proteins including hnRNP G, RNP S1, X16, SF2, and SRp75, from the unstimulated mouse T-cell cDNA library with Clk1 as a bait in the two-hybrid assay (13). In their two-hybrid assay system, kinase inactive Clk1 bound these SR proteins with similar efficiency as wild type Clk1, and kinase domain alone did not interact with these proteins (13). In our two-hybrid screening condition, no SR proteins have been cloned as a binding partner of Clk4. In the GenBank^TM data search, we found a human cDNA sequence named suppressor-of-white-apricot (SWAP) 2 (DDBL/EMBL/GenBank^TM accession no. NM007056), which is homologous to Clasp cDNA. Though SWAP protein also has RS repeat motif, total homology of Clasp and SWAP are 26% in the amino acids level, and Clasp has no "surp module" described in SWAP protein (32, 33).

In vitro binding study, recombinant Clasp protein was bound only to the inactive mutant protein of Clk4 in good accordance with the two-hybrid study. As Clk4 autophosphorylates with dual specificity in vitro, it is possible that Clasp is released from the autophosphorylated Clk4 by recognizing the structural change of the kinase. Loss of the kinase activity may stabilize the interaction as we observed in the case of interaction between SF2/ASF and SRPK (27). Although the phosphorylation state of Clk4 in mammalian cells has not been investigated, endogenous Clk4 may exist as an unphosphorylated form like Clk2 as reported by Nayler et al. (20). If so, Clasp should be able to interact with endogenous Clk4 in the mammalian cells. All together, a possible role of ClaspL in vivo is to trap the unphosphorylated form of Clk4 and keep it in nuclear dots. In the splicing assay in vivo, ClaspL promoted exon inclusion of the CR1 minigene. It has been reported that Clk has four family members (Clk1-4) in human and mouse, and its exon/intron boundaries in these four genes seem to be well conserved. So it is conceivable that ClaspL has the same effect on exon inclusion of other Clk members. In this splicing assay, it is uncertain that ClaspL directly binds Clk and influences kinase activity. In the in vitro kinase assay, Clasp protein did not affect the kinase activity of recombinant Clk1 (data not shown).

The biological functions of mammalian Clks are unknown, although a possible role in PC12 differentiation has been suggested (34). Clk homologues have been isolated from a number of organisms: Saccharomyces cerevisiae KNS1 (35) and Arabidopsis thaliana AFC1, AFC2, and AFC3 (36). AFC1 complements mutations in the yeast mitogen-activated protein kinase KSS1 and FUS3 mutation of darkener of apricot (Doa) (37). These kinases have a conserved amino acid motif EHLAMMERILG in the kinase subdomain X, which has led these kinases to be dubbed LAMMER kinases (19.38). The only known LAMMER kinase in Drosophila melanogaster is DOA (38). Doa mutations were isolated in screens for dosage-sensitive modifiers of white^apricot (w^a) (39). Doa protein is required for segmentation and development of the nervous system, and Doa mutations are almost invariably recessive-lethal (40). The organization and development of pigment cells, bristles, and photoreceptors are also affected and roughened eyes were observed in various mutant classes, suggesting that Doa function is critical to Drosophila eye development (38). Recently Du et al. (41) showed that mutations in Doa locus disrupt sex-specific splicing of doublesex pre-mRNA by affecting the phosphorylation, altering sexual differentiation, and interacting synergistically with SR proteins transformer (TRA) and TRA2. Thus it will be highly possible that the kinase activity of Doa should be regulated depending the developing stage. In C. elegans, the phosphorylation state of SR proteins is dependent on the developmental stages (42) though the responsive kinase(s) have not been identified. Identification of Clasp may give us a clue to clarify the regulatory mechanisms of Clk signal pathway in vivo.

	ACKNOWLEDGEMENTS

We are grateful to laboratory members for discussion. We thank Dr. J. C. Bell (Ottawa Regional Cancer Center) for CR-1 and CR-2 minigene plasmids. We are also grateful for Dr. T. Mizuno for helpful advice on the yeast experiments.

	FOOTNOTES

^* This work was supported by Research for the Future Program.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EBI Data Bank with accession number(s) AB080582 and AB080583.

^¶ To whom correspondence should be addressed. Tel.: 81-35803-5836; Fax: 81-35803-5853; E-mail: m.hagiwara.end@mri.tmd.ac.jp.

Published, JBC Papers in Press, August 6, 2002, DOI 10.1074/jbc.M206504200

	ABBREVIATIONS

The abbreviations used are: SRPK, SR-protein specific kinase; Clk, CDC2-like kinase; RT, reverse transcription; HA, hemagglutinin; WT, wild type; GST, glutathione S-transferase; CMV, cytomegalovirus; FBS, fetal bovine serum; PMSF, phenylmethylsulfonyl fluoride; DTT, dithiothreitol; PBS, phosphate-buffered saline; GFP, greeen fluorescent protein; SWAP, suppressor-of-white-apricot; Doa, darkener of apricot.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J.-i. Yomoda, M. Muraki, N. Kataoka, T. Hosoya, M. Suzuki, M. Hagiwara, and H. Kimura Combination of Clk family kinase and SRp75 modulates alternative splicing of Adenovirus E1A. Genes Cells, March 1, 2008; 13(3): 233 - 244. [Abstract] [Full Text] [PDF]

				A. H. Mangs, H. J.L. Speirs, C. Goy, D. J. Adams, M. A. Markus, and B. J. Morris XE7: A novel splicing factor that interacts with ASF/SF2 and ZNF265 Nucleic Acids Res., October 18, 2006; 34(17): 4976 - 4986. [Abstract] [Full Text] [PDF]

				M. Muraki, B. Ohkawara, T. Hosoya, H. Onogi, J. Koizumi, T. Koizumi, K. Sumi, J.-i. Yomoda, M. V. Murray, H. Kimura, et al. Manipulation of Alternative Splicing by a Newly Developed Inhibitor of Clks J. Biol. Chem., June 4, 2004; 279(23): 24246 - 24254. [Abstract] [Full Text] [PDF]

				J. Prasad and J. L. Manley Regulation and Substrate Specificity of the SR Protein Kinase Clk/Sty Mol. Cell. Biol., June 15, 2003; 23(12): 4139 - 4149. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/46/44220    most recent M206504200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Katsu, R. Articles by Hagiwara, M. Search for Related Content PubMed PubMed Citation Articles by Katsu, R. Articles by Hagiwara, M. Social Bookmarking                      What's this?