| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 280, Issue 42, 35108-35118, October 21, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1
2
From the
Departments of Signal Transduction Sciences and Cell Physiology, Faculty of Medicine, Kagawa University, 1750-1 Miki-cho, Kita-gun, Kagawa 761-0793, the ¶Department of Chemistry, Graduate School of Science, Tokyo Metropolitan University, Hachioji-shi, Tokyo 192-0397, and the ||Department of Biochemistry and Molecular Biology, Kanagawa Dental College, 82 Inaoka-cho, Yokosuka 238-8580, Japan
Received for publication, April 11, 2005 , and in revised form, August 15, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
, 
, 
, and 
) derived from different genes; these isoforms are ubiquitously expressed in cells and tissues and are localized primarily in the cytoplasm (11–17). Recent studies have indicated that the CaM-KI isoforms are localized in both the cytoplasm and nucleus (18) and that the -isoforms are anchored to the Golgi and plasma membranes in neurons through a CAAX motif (15, 16). These findings suggest that CaM-KI may play various roles throughout cells (i.e. from the cell membrane to the nucleus). CaM-KI is also conserved among various species, including Aspergillus nidulans (3), Schizosaccharomyces pombe (4), and Caenorhabditis elegans (19).
CaM-KI is part of a protein kinase cascade, referred to as a CaM-K cascade. Full activation of CaM-KI requires the phosphorylation of Thr177 in the activation loop by Ca2+/CaM-dependent protein kinase kinase (CaM-KK) and Ca2+/CaM binding (20–23). It has been shown that Thr177 phosphorylation results in a large increase in the affinity of CaM-KI for its substrate and that Ca2+/CaM binding is required for the release of the autoinhibitory domain from the catalytic core (19, 22, 24). Whereas a number of protein substrates for CaM-KI have been identified (e.g. synapsin I and II (11), the cystic fibrosis transmembrane conductance regulator (26), CREB (27), activating transcription factor (9), and myosin II regulatory light chain (8)), the physiological role of the CaM-KI cascade remains uncertain. To evaluate the physiological functions of the CaM-KK/CaM-KI cascade, it is clearly important to identify the substrates for CaM-KI. A recent study using phosphorylation screening of a cDNA expression library identified novel substrates such as translation initiation factor 4GII; hence, the phosphorylation screening approach may continue to identify additional physiological substrates of CaM-KI (28).
To search for CaM-KI substrates, in this study we developed a functional proteomic method using affinity chromatography with the CaM-KI catalytic domain (CD) as a ligand to partially purify CaM-KI-CD-interacting proteins; we combined this approach with a liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis for the identification of substrates. In addition to the identification of two known CaM-KI substrates (synapsin I and CREB) with this method, we identified rat Numbl and Numb, which were bound to the phosphorylated CaM-KI CD at Thr177, and demonstrated that both Numb family proteins are excellent substrates for activated CaM-KI. Furthermore, we characterized the phosphorylation of these Numb family proteins in vitro and in vivo, including their phosphorylation-dependent interactions with 14-3-3 protein.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
, wild-type mouse CaM-KIV, and wild-type rat CaM-KK were expressed and purified as previously described (31, 32). Recombinant rat CaM was expressed in the Epicurian coli BL-21 (DE3) using pET-CaM (kindly provided by Dr. Nobuhiro Hayashi, Fujita Health University, Toyoake, Japan) and was purified by phenyl-Sepharose column chromatography (33). The purified catalytic subunit of bovine cAMP-dependent protein kinase (PKA) was kindly provided by Dr. Y. Watanabe (Kagawa University, Kagawa, Japan). Rat CaM-KII holoenzyme was purified from rat forebrain. His-tagged C. elegans CREB (CRH-1) was expressed and purified as described previously (10). GST-14-3-3
was expressed in E. coli JM-109, followed by purification by glutathione-Sepharose chromatography (51). Anti-phospho-CaM-KI at Thr177 monoclonal antibody was generated as previously described (32). Anti-CaM-KI and anti-CaM-KK antibodies were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) and Transduction Laboratories, respectively. Anti--tubulin, anti-HSP-70, anti-CREB, and anti-synapsin I antibodies were obtained from Amersham Biosciences, MBL International, New England Biolabs, and Chemicon International, respectively. Anti-phospho-CREB antibody was obtained from New England Biolabs. Anti-Numbl and anti-Numb antibodies were purchased from Abcam Ltd. (Cambridge, UK) and Upstate%20Biotechnology">Upstate Biotechnology, Inc. (Lake Placid, NY), respectively. Antibodies to each 14-3-3 isoform were provided by Immuno-Biological Laboratories. Anti-phospho-Numb/Numbl monoclonal antibody was generated against the synthetic phosphopeptide corresponding to residues 295–314 (CPLEQLVRQGpSFRGFPALSQK; where pS represents phosphoserine) of rat Numbl. The phosphopeptide was conjugated with keyhole limpet hemocyanin via the N terminus cysteine and was injected into Balb/c mice as described previously (34). All other chemicals were obtained from standard commercial sources.
Affinity chromatography—GST-CaM-KI (residues 1–293, K49E) (2.5 mg) was phosphorylated with 3 µg of recombinant rat CaM-KK at 30°C for 2 h in a solution containing 50 mM HEPES (pH 7.5), 10 mM Mg(Ac)2, 1 mM CaCl2, 1 mM DTT, 2 µg CaM, and 1 mM ATP. Then either phosphorylated or unphosphorylated GST-CaM-KI (residues 1–293, K49E) (2.5 mg) or GST (1 mg) was applied to a glutathione-Sepharose column (250-µl bed volume; Amersham Biosciences), and then the columns were washed with 10 ml of Buffer A (150 mM NaCl, 50 mM Tris-HCl (pH 7.5), 1 mM DTT, 1 mM EDTA). Three fresh whole rat brains and 10 g of fresh rat liver were separately homogenized with 30 ml of Buffer A containing 1 µM microcystin LR and 0.5 µM okadaic acid (Buffer B), followed by centrifugation at 100,000 x g for 30 min. The supernatant (10 ml) was applied to each Sepharose column as described above, followed by washing with 50 ml of Buffer B. The proteins that interacted with GST-CaM-KI (residues 1–293, K49E) were eluted by adding 250 µl of Buffer A containing 30 units of PreScission protease (Amersham Biosciences) to each Sepharose resin, followed by incubation at 4 °C overnight. Twenty-five microliters of SDS-PAGE sample buffer was added to the eluate, and then each sample was stored at –30 °C until analysis by mass spectrometry or Western blotting.
GST-Numb fragment (residues 238–304, 1.5 mg) was phosphorylated with activated CaM-KI (2.1 µg) in the presence of 2 µM CaM and 2 mM ATP in a solution containing 50 mM HEPES (pH 7.5), 10 mM Mg(Ac)2, 1 mM CaCl2, 1 mM DTT. Then either phosphorylated or unphosphorylated GST-Numb fragment (residues 238–304) was applied to glutathione-Sepharose columns (250-µl bed volume), and then the columns were washed with 10 ml of Buffer A. Affinity purification of Numb-interacting proteins using the affinity resins in the presence of protein phosphatase inhibitors was performed according to essentially the same protocol as that used for the affinity purification of CaM-KI-interacting proteins described above.
Mass Spectrometry Analysis—A 30-µl sample of the eluate from the unphosphorylated or phosphorylated GST-CaM-KI (residues 1–293, K49E)-coupled glutathione-Sepharose columns described above was separated by SDS-10% PAGE and lightly stained with Coomassie Brilliant Blue. Then 16 gel slices were excised from each sample lane in the 
35–250-kDa range, followed by in-gel digestion with 10 µg/ml trypsin (Promega, Madison, WI) overnight at 37 °C (44). The digested peptides were eluted with 0.1% formic acid and were subjected to LC-MS/MS analysis. LC-MS/MS analysis was performed on a Q-Tof2 quadrupole/time-of-flight hybrid mass spectrometer (Micromass, Manchester, UK) interfaced with capillary reverse-phase liquid chromatography (Micromass CapLCTM system). A 90-min linear gradient from 5 to 45% acetonitrile in 0.1% formic acid was produced and was split at a 1:20 ratio; the gradient solution was then injected into a nano-LC column (PepMap C18, 75 µm x 150 mm; LC Packings, Sunnyvale, CA) at 25 nl/min. The eluted peptides were sprayed directly into the mass spectrometer. The MS/MS data were acquired by MassLynx software (Micromass) and converted to a single text file (containing the observed m/z of the precursor peptide, the fragment ion m/z, and intensity values) by Protein-Lynx software (Micromass). The file was analyzed with the Mascot MS/MS Ions Search (Matrix Science; available on the World Wide Web at www.matrixscience.com) to search and assign the obtained peptides to the NCBI nonredundant data base. We set the search parameters as follows: data base, NCBInr; taxonomy, all; enzyme, trypsin; fixed modifications, carbamidomethyl (C); variable modifications, oxidation (M); peptide tolerance, ±0.2 Da; MS/MS tolerance, ±0.2 Da.
LC-MS/MS was used to identify the phosphorylation site of the Numbl protein, as previously described (35). Two micrograms of phosphorylated recombinant GST-Numbl were separated by SDS-10% PAGE. The following steps were then performed as described above, with two exceptions. First, in-gel digestion was performed with 17 µg/ml chymotrypsin (Roche Applied Sciences) overnight at 25 °C; second, the search parameters were as follows: data base, GST-rat Numbl-Hisx6 (868 amino acid residues); enzyme, all; variable modification, oxidation (M) and phospho-Ser/Thr; peptide tolerance, ±0.1 Da; and MS/MS tolerance, ±0.1 Da.
As regards the identification of Numb-binding proteins, the protein bands of interest were excised from the gel, and in-gel digestion with trypsin (Promega) was performed as described above. The resulting peptides were desalted using a ZipTip µC18 (Millipore Corp.) according to the manufacturer's protocol. Peptides were eluted with 1 µl of matrix (-cyano-4-hydroxycinnamic acid) prepared in 50% acetonitrile, 0.1% trifluoroacetic acid, and then the peptides were spotted onto a stainless steel target plate. The protein digests were analyzed by MALDI-TOF MS using the Voyager DE-STR system (Applied Biosystems). Calibration was performed using the autodigestion peaks of trypsin (1045.5642 and 2211.1046). The observed m/z values were submitted to MS-Fit (available on the World Wide Web at prospector.ucsf.edu/ucs-fhtml4.0/msfit.htm) for mass fingerprint searching. We set the search parameters as follows: data base, NCBInr; digest, trypsin; maximum number of missed cleavages, 1; Cys modified by, carbamidomethylation; possible modifications mode, none; minimum number of peptides required to match, 4; mass tolerance, 25 ppm.
Cloning and Construction of Rat Numbl and Numb cDNAs—Rat Numbl cDNA (accession number AB210107 [GenBank] ) was obtained by reverse transcriptase-mediated PCR with Pyrobest DNA polymerase (Takara, Tokyo, Japan) using rat brain cDNA (Invitrogen) as a template (sense primer 5'-CGTCAGATCGAGCCGCCGCCACCACAGCAG-3', derived from the 5'-untranslated region in the cDNA sequence in the data base (XP_218360), and antisense primer 5'-GGCAAGCACAGCATTGGCAGTGAACACAGC-3', derived from the genomic sequence in rat chromosome 1). The initial PCR was followed by a second PCR, using sense primer 5'-GGTCTAGAGATGTCCCGCAGCGCGGCGGCC-3' and antisense primer 5'-GGCTCGAGCTACAGTTCAATCTCGAAGGTC-3' (where underlines indicate nucleotide sequences of restriction sites). Rat Numb cDNA (accession number AB210108 [GenBank] ) was obtained by reverse transcriptase-mediated PCR with Pyrobest DNA polymerase (Takara) using rat brain cDNA (Invitrogen) as a template (sense primer, 5'-CTGTGTCTCCAGGTTGTAAAAGTTAAC-3'; antisense primer, 5'-CCTTGCCTAAGGACAGAAAGAACCATC-3'; both primers were from the genomic sequence in rat chromosome 6). This latter PCR was followed by a second PCR, using sense primer 5'-GGTCTAGACATGAACAAACTACGGCAGAGTTTC-3' and antisense primer 5'-CCCTCGAGCTAAAGCTCTATTTCAAATGCC-3', both of which were derived from the data base (XP_234394 [GenBank] ). The PCR fragments were subcloned into the XbaI/XhoI site of the pGEX-KG-PreS vector. To insert a His6 tag at the C-terminal end of GST-Numbl and also at that of Numb, we inserted annealed oligonucleotides (5'-GAAGTTCTGTTCCAGGGGCCCGAGCACCACCACCACCACCACTGAC-3' and 5'-TCGAGTCAGTGGTGGTGGTGGTGGTGCTCGGGCCCCTGGAACAGAACTTC-3') encoding EVLFQGPEHHHHHH after Leu617 in GST-Numbl (pGEX-KG-PreS-Numbl-His6) or after Leu592 in GST-Numb (pGEX-KG-PreS-Numb-His6). The S304A mutant of pGEX-KG-PreS-Numbl-His6 and the S264A mutant of pGEX-KG-PreS-Numbl-His6 were created by site-directed mutagenesis (GeneEditorTM; Promega) using each mutagenic oligonucleotide. An expression plasmid for GST-fused Numb 238–304 was constructed by PCR using sense primer 5'-GGTCTAGAGACTGCTTCTTTAGAGATGAAC-3' and antisense primer 5'-GGATCCACGCGGAACCAGTGTGTTTTTTATTGGGAA-3', followed by ligation into the XbaI/XhoI sites of pGEX-KG-PreS together with the annealed oligonucleotides encoding the His6 tag, as described above. The C-terminal FLAG-tagged Numbl (pME-Numbl-FLAG) was constructed by PCR using wild type or a S304A mutant Numbl cDNA as a template and antisense primer (5'-CCTCTAGACTACTTATCGTCGTCATCCTTGTAATCCAGTTCAATCTCGAAGGTCTTCTGC-3') containing a region encoding the FLAG epitope, and then the PCR fragment was subcloned into pME18s vector (DNAX Research Institute, Inc.). The nucleotide sequences of all constructs used in this study were confirmed by an ABI377 automated sequencer (PE Biosystems, Foster City, CA).
Purification of GST-Numbl and GST-Numb—cDNAs carrying GST-fused rat Numbl, Numb, and Numb fragment (pGEX-KG-PreS-Numbl-His6, pGEX-KG-PreS-Numb-His6, and pGEX-KG-PreS-Numb 238–304-His6), including S304A (Numbl) and S264A (Numb) mutants, were introduced into E. coli JM-109, and expression of the recombinant proteins was induced by the addition of 1 mM isopropyl--D-thiogalactopyranoside. An E. coli pellet containing GST fusion protein was lysed with PBS, followed by purification using glutathione-Sepharose column chromatography, as described in the manufacturer's protocol. Further purification was then carried out by using Ni2+-nitrilotriacetic acid-agarose column chromatography (Qiagen) to obtain GST-fused proteins containing full-length Numbl and Numb. Full-length Numb and Numbl were prepared by the cleavage of GST-Numb and GST-Numbl with PreScission protease.
Phosphorylation of Rat Numbl and Numb in Vitro—Purified recombinant CaM-KI (0.1 mg/ml) or CaM-KIV (0.1 mg/ml) was incubated with or without CaM-KK (3 µg/ml) at 30 °C for 20 min in a solution containing 50 mM HEPES (pH 7.5), 10 mM Mg(Ac)2, 1 mM DTT, 2 mM CaCl2, 10 µM CaM, and 200 µM of ATP. The reaction was initiated by the addition of ATP and terminated by a 10-fold dilution with ice-cold 50 mM HEPES (pH 7.5), 2 mg/ml bovine serum albumin, 10% ethylene glycol, and 2 mM EDTA. Small aliquots of CaM-KK-treated (activated) and CaM-KK-untreated (unactivated) CaM-KI or CaM-KIV were stored at –80 °C until used. Purified GST-Numbl, GST-Numb, Numbl, or Numb was incubated without or with unactivated or activated CaM-KI (10 ng) or other protein kinases at 30 °C for the indicated periods in a solution containing 50 mM HEPES (pH 7.5), 10 mM Mg(Ac)2, 1 mM DTT, 2 mM CaCl2, 5 µM CaM, and either 200 µM [-32P]ATP (1000 cpm/pmol) or 200µM ATP for mass spectrometry analysis or Western blotting. Ca2+/CaM was omitted for PKA phosphorylation. The reaction was terminated by the addition of SDS-PAGE sample buffer. The samples were then subjected to SDS-7.5% PAGE followed by either Western blot analysis, autoradiography, or quantification of 32P incorporation into the GST-Numbl by Cerenkov counting of the excised gels.
Phosphorylation of Endogenous Numb in COS-7 Cells—COS-7 cells were maintained in Dulbeco's modified Eagle's medium containing 10% fetal bovine serum. The cells were subcultured in 6-well dishes and then were further cultured in serum-free medium for 22 h, and then the cells were treated with or without 1 µM ionomycin for the indicated period of time. Stimulation was terminated by the addition of 150 µl of SDS-PAGE sample buffer, and then the whole cell lysates were heated at 100 °C for 10 min. After centrifugation, 20 µl of each sample was subjected to SDS-10% PAGE followed by Western blot analysis using antiphospho-Numb/Numbl antibody.
Preparation of Rat Tissue Extract—Rat tissue samples were homogenized with 5 volumes of extraction buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 1 mM DTT, 0.2 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin, and 10 µg/ml trypsin inhibitor), and then the samples were centrifuged at 4 °C. SDS-PAGE sample buffer was added to the supernatant, and each sample was stored at –30 °C until used for the Western blot analysis.
Expression of FLAG-tagged Numbl in COS-7 Cells—Transfection of pME-Numbl-FLAG into COS-7 cells was carried out using Lipofectamine reagent (Invitrogen) with 10 µg of the plasmid DNA according to the manufacturer's instruction. After 40 h of incubation, the cells were extracted, and then the cell extract was subjected to the pull-down assay as described below.
Pull-down Assay—Purified GST-14-3-3 (10 µg) was incubated with either phosphorylated or unphosphorylated Numb/Numbl or with COS-7 cell extract containing overexpressed FLAG-tagged Numbl in a solution containing 150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 1 mM DTT, 1mM EDTA, 1 mM EGTA, 1% Nonidet P-40, 0.2 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin and 10 µg/ml trypsin inhibitor, 1 µM microcystin LR, and 0.5 µM okadaic acid (Buffer C) for 2 h at 4°C or room temperature (for recombinant Numb/Numbl), and then the samples were incubated with 40 µl of glutathione-Sepharose (50% slurry) for 1 h. The glutathione-Sepharose resins were washed five times with buffer C, and then SDS-PAGE sample buffer was added to the resin, followed by either SDS-PAGE or Western blot analysis.
Dephosphorylation of Numbl—GST-Numbl was phosphorylated by activated CaM-KI at 30 °C for 30 min as described above, followed by purification using glutathione-Sepharose column chromatography. Phosphorylated GST-Numbl (1 µg) was either left untreated or incubated with 0.025 units of protein phosphatase 2A (Upstate%20Biotechnology">Upstate Biotechnology, Inc., Lake Placid, NY) at 30 °C for the indicated period of time in a solution containing 50 mM Tris-HCl (pH 7.5), 5 mM MgCl2, and 1 mM DTT in the presence of either GST (1.5 µg) or GST-14-3-3 (3.2 µg). The reaction was terminated by the addition of SDS-PAGE sample buffer, and the samples were then subjected to SDS-7.5% PAGE, followed by Western blot analysis using either anti-phospho-Numb/Numbl antibody or anti-Numbl antibody.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
Gly-Pro) for PreScission protease between GST and the CaM-KI CD, which led to the specific elution of the CaM-KI CD-interacting proteins from the affinity matrix by protease treatment. GST-fused CaM-KI CD was fully phosphorylated with CaM-KK in the presence of Ca2+/CaM for 2 h, as was recently demonstrated (32), and then phosphorylated CaM-KI CD was coupled with glutathione-Sepharose resin (Fig. 1A, c). As controls, we prepared two glutathione-Sepharose columns, one coupled with GST alone (Fig. 1A, a) and the other with unphosphorylated GST-fused CaM-KI CD (Fig. 1A, b). An equal volume of rat brain extract was applied to each column in the presence of protein phosphatase inhibitors (microcystin LR and okadaic acid), followed by extensive washing of each column (Fig. 1B). CaM-KI CD-interacting proteins were eluted by incubation with 30 units of PreScission protease at 4 °C overnight. The same volume of eluate from each column was subjected to SDS-PAGE, followed by either protein staining (Fig. 2A, left) or Western blot analysis with anti-phospho-CaM-KI antibody (Fig. 2A, right). The Thr177-phosphorylated CaM-KI CD (lane c) and the unphosphorylated CaM-KI CD (lane b) were cleaved from GST and were eluted from each column in approximately the same amounts, together with the CaM-KI CD-associated proteins with a molecular weight range of 40–250 kDa (Fig. 2A, left). The proteins eluted from the columns, together with the cleaved CaM-KI CD, appeared to interact directly with CaM-KI CD, since we did not detect any proteins with a molecular mass higher than 30 kDa in the eluate from the GST-coupled glutathione-Sepharose (Fig. 2A, left, lane a). Next, to identify the CaM-KI CD-interacting proteins, we excised 16 gel slices in the 35–250-kDa range from the SDS-polyacrylamide gel, in which the eluates from either Thr177-phosphorylated or unphosphorylated CaM-KI CD-coupled glutathione-Sepharose were separated. The slices were then subjected to in-gel digestion with trypsin. The digested peptides eluted from each slice were subjected to LC-MS/MS analysis in order to identify the proteins by searching the NCBI data base. Among the proteins we detected from the Thr177-phosphorylated and unphosphorylated CaM-KI CD-coupled glutathione-Sepharose samples, 100 were found to occur in both samples. These proteins included -tubulin and HSP-70, which have been found to possess the assigned peptides in similar numbers (data not shown). We confirmed by Western blot analysis that approximately the same amounts of -tubulin (Fig. 2B, left) and HSP-70 (Fig. 2B, right) were eluted from both Thr177-phosphorylated and unphosphorylated CaM-KI-CD-coupled resins, thus indicating that both proteins interacted with CaM-KI CD, at least in a Thr177 phosphorylation-independent manner. However, we detected a large amount of synapsin I in the eluate from Thr177-phosphorylated CaM-KI CD-coupled glutathione-Sepharose (the assigned peptides covered 48% of the total sequence of rat synapsin I), in contrast to that detected in the eluate from unphosphorylated CaM-KI CD-coupled resin (the assigned peptides covered 9% of the total sequence of rat synapsin I) (data not shown). Western blot analysis with anti-synapsin I antibody clearly indicated that CaM-KK-mediated Thr177 phosphorylation significantly increased the amount of synapsin I available to interact with CaM-KI CD (Fig. 2C, right). In addition, we detected two peptides derived from rat CREB (residues 136–150, KILNDLSSDAPGVPR; residues 137–150, ILNDLSSDAPGVPR), but only in the eluate from Thr177-phosphorylated CaM-KI CD-coupled glutathione-Sepharose. This finding is consistent with the results of the Western blot analysis using anti-CREB antibody (Fig. 2C, left), namely, CREB was detected only in the sample from Thr177-phosphorylated CaM-KI CD-coupled resin (lane c). Synapsin I and CREB have been shown to be phosphorylated in vitro by CaM-KI at Ser9 (P-site 1) and Ser133, respectively (27, 36).
|
|
-32P]ATP at 30 °C for 5 min without or with either unactivated or activated CaM-KI (10 ng) having a CaM-KK phosphorylation at Thr177, only the activated CaM-KI efficiently phosphorylated GST-Numbl (Fig. 4A). This finding was in good agreement with results demonstrating that Numbl can directly interact only with Thr177-phosphorylated CaM-KI CD and not with unphosphorylated CaM-KI CD (Fig. 3B). To determine the site(s) of Numbl that are phosphorylated by activated CaM-KI, we analyzed phosphorylated GST-Numbl (2 µg) as described above by in-gel digestion with chymotrypsin, followed by LC-MS/MS analysis. We obtained peptide sequences that covered 78% of the entire amino acid sequence of Numbl (data not shown). Among these peptides, we detected a single phosphopeptide corresponding to residues 297–305 in Numbl. LC-MS/MS analysis also revealed a single phosphorylation site at Ser304 in the peptide (Fig. 4B). To confirm the phosphorylation of Numbl at Ser304 by activated CaM-KI, we used a GST-Numbl S304A mutant in a phosphorylation assay (Fig. 4A, right lane). It was observed that 32P incorporation into GST-Numbl S304A was significantly lower than that into the wild-type Numbl, although weak residual phosphorylation was observed in the mutant. These results indicate that the Ser304 residue in rat Numbl is a primary site for phosphorylation by CaM-KK-activated CaM-KI in vitro. The Ser304 in rat Numbl is conserved in mouse and human counterparts as well as in Numb (Ser264), another Numb family protein (Fig. 3A). Next, we kinetically analyzed rat Numbl phosphorylation (Fig. 4C). A time course experiment on the phosphorylation of GST-Numbl (2 µg) by activated CaM-KI (10 ng) revealed that rat Numbl was rapidly (t
= 1–2 min) and stoichiometrically (0.9 mol of Pi incorporation into 1 mol of GST-Numbl) phosphorylated under these conditions.
Identification of Numb from Rat Liver as a CaM-KI Substrate—In a separate experiment employing this proteomic approach as applied to rat liver extract, four peptides derived from rat Numb (Fig. 3A, bold-faced sequences in rat Numb) were detected; however, these peptides were only detected in the eluate from Thr177-phosphorylated CaM-KI CD-coupled glutathione-Sepharose. This finding, which was supported by the results of Western blot analysis using anti-Numb antibody (Fig. 5A, right), was also similar to what we observed with Numbl from rat brain extract (Fig. 3B). We detected two immunoreactive bands with molecular masses of 75 and 65 kDa, which was consistent with a previous report in which four mouse Numb isoforms having molecular masses of 65, 66, 71, and 72 kDa were generated by alternative splicing of the Numb mRNA (43). CREB from rat liver was also detected only in the eluate from Thr177-phosphorylated CaM-KI CD-coupled resin (Fig. 5A, left panel). These results suggest that both Numbl and Numb are potential substrates for activated CaM-KI. Therefore, we cloned Numb cDNA (1779 bp; accession number AB210108
[GenBank]
) by reverse transcription-PCR, which revealed that this cDNA encoded a p65 isoform composed of 592 amino acid residues (Fig. 3A) (43) and then produced recombinant GST-fused Numb protein. Similar to the results of the phosphorylation experiment with rat Numbl (Fig. 4A), GST-fused Numb was significantly phosphorylated, but only by activated CaM-KI, and 32P-incorporation into rat Numb was largely reduced by an Ala mutation at Ser264 (Fig. 5B).
|
Since multifunctional CaM-Ks (including CaM-KI, CaM-KII, and CaM-KIV) have been shown to phosphorylate similar consensus sequences, we examined whether two other CaM-Ks (CaM-KII and CaM-KIV) were capable of phosphorylating the CaM-KI-sites of Numbl (Ser304) and Numb (Ser264). We also investigated the phosphorylation of Numbl by PKA. As a control experiment (Fig. 6C, left), we confirmed that all of the kinases tested were able to phosphorylate the Ser29 residue of C. elegans CREB protein (CRH-1) (10). In addition to activated CaM-KI, we found that CaM-KII and activated CaM-KIV by CaM-KK phosphorylation were capable of phosphorylating the Ser304 residue of Numbl as well as the Ser264 residue of Numb; however, PKA did not phosphorylate these residues (Fig. 6C, right two panels).
|
|
75 and 65 kDa in these cells (Fig. 7B, lane c); this result was similar to that observed with partially purified Numb isoforms from rat liver extract (Fig. 5A, right, lane c). We also detected CaM-KK and CaM-KI in COS-7 cells (Fig. 7B, lanes a and b). We stimulated COS-7 cells with 1 µM ionomycin for various periods of time after 22-h serum starvation, and then the whole cell lysates were analyzed by Western blotting using the anti-phospho-Numb/Numbl antibody (Fig. 7, B (middle) and C). The immunoreactivity of the 65-kDa protein, which was possibly a phosphorylated form of the Numb isoform, was detected in the sample without stimulation. The phosphorylation of the 65- and 75-kDa Numb isoforms was rapidly induced (2.5-fold) in response to ionomycin treatment. Phosphate incorporation peaked within 2–5 min and then gradually decreased to the basal level within 15 min; this decrease might have been due to the enhanced dephosphorylation of Numb. We obtained a similar result with HeLa cells, indicating that endogenous Numb is phosphorylated at Ser264 and that phosphorylation is induced in response to intracellular Ca2+ mobilization in intact cells.
CaM-KI Phosphorylation of Numb Family Proteins Regulates Interaction with 14-3-3 Proteins—To further our understanding of the functional consequences of Numb/Numbl phosphorylation, we attempted to identify the proteins that interact with Numb/Numbl in a phosphorylation-dependent manner. To purify the Numb-interacting proteins from rat liver extract, we used as the affinity ligands an unphosphorylated GST-fused Numb fragment (residues 238–304; underlined sequence in Fig. 3A) and a GST-fused Numb fragment phosphorylated at Ser364 by activated CaM-KI. The Numb-interacting proteins were eluted by treatment of the affinity resins with PreScission protease to cleave the Numb fragment from GST according to essentially the same protocol as that used for the affinity purification of CaM-KI-interacting proteins, followed by SDS-PAGE analysis (Fig. 8A, left panel) and Western blotting with anti-phospho-Numb/Numbl antibody (Fig. 8A, right panel). As shown in Fig. 8A, proteins with molecular masses of 32 and 30 kDa were purified specifically with phosphorylated GST-Numb fragment (lane b). Mass fingerprinting identified those Numb-binding proteins as 14-3-3
(32-kDa band), 14-3-3
, and 14-3-3 (30-kDa bands), respectively, indicating that the 14-3-3 isoforms were capable of interacting with the Numb fragment (residues 238–304) in a phosphorylation-dependent manner. To examine whether the Numb fragment preferentially binds to any specific 14-3-3 isoform(s), we analyzed purified samples by Western blotting using isoform-specific anti-14-3-3 antibodies (Fig. 8B). Here we found that all of the 14-3-3 isoforms (, , , 
, , and ) tested interacted with the Ser264-phosphorylated Numb fragment. We then investigated whether the full-length Numb protein could interact with 14-3-3 protein in a phosphorylation-dependent manner by using a pull-down assay. The GST-14-3-3 was incubated with recombinant Numb with or without activated CaM-KI phosphorylation, and then the samples were subjected to the pull-down assay with glutathione-Sepharose, followed by Western blot analysis using anti-Numb antibody (Fig. 8C). As a result, the phosphorylated form of Numb was found to specifically interact with GST-14-3-3. We then examined the phosphorylation-dependent interaction of Numbl with 14-3-3 protein by pull-down assay using recombinant Numbl expressed in E. coli (Fig. 9A) and COS-7 cells extract containing overexpressed FLAG-tagged Numbl (Fig. 9B). The results clearly demonstrated that the 14-3-3 protein interacted with phosphorylated Numbl at Ser304 but not with unphosphorylated Numbl or with the S304A mutant treated with activated CaM-KI (Fig. 9A). These findings were confirmed with C-terminal FLAG-tagged Numbl expressed in COS-7 cells. A certain population of FLAG-tagged Numbl was phosphorylated at Ser304 in transfected COS-7 cells (Fig. 9B, top two panels), which was consistent with our observation regarding the phosphorylation of endogenous Numb in nonstimulated COS-7 cells (Fig. 7B). We found that the phosphorylated form of FLAG-tagged wild-type Numbl was specifically pulled down with GST-14-3-3, whereas no interaction of the Numbl S304A mutant with GST-14-3-3 was observed (Fig. 9B, bottom two panels), thus indicating that the phosphorylation of Ser304 is critical for the association of Numbl with 14-3-3 protein. Finally, we tested whether the interaction with 14-3-3 protein might regulate dephosphorylation of Numbl. With activated CaM-KI-phosphorylated GST-Numbl as substrate, protein phosphatase 2A gave robust dephosphorylation of Ser304, and this dephosphorylation was completely blocked by the presence of GST-14-3-3 (Fig. 9C).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
The primary sequences surrounding the CaM-KI phosphorylation sites of both Numb family proteins (Leu-Val-Arg-Gln-Gly-Ser(P)304-Phe-Arg-Gly-Phe in Numbl and Leu-Ala-Arg-Gln-Gly-Ser(P)264-Phe-Arg-Gly-Phe in Numb) match completely with an optimal consensus sequence for substrate recognition of CaM-KI (Hyd-Xaa-Arg-Xaa-Xaa-Ser(P)/Thr(P)-Xaa-Xaa-Xaa-Hyd, where Hyd represents a hydrophobic amino acid residue), as determined by synthetic peptides (25). This is consistent with the notion that Numb family proteins are the down-stream targets of the Ca2+/CaM-dependent signaling cascade, and CaM-KK/CaM-KI is implicated in this pathway. However, we also found that two other multifunctional CaM-Ks (CaM-KII and CaM-KIV) were capable of phosphorylating the CaM-KI phosphorylation sites in both Numbl and Numb in vitro. This observation suggested the possibility that multiple CaM kinases might be involved in enhanced Numb phosphorylation in response to the Ca2+ mobilization observed in COS-7 cells; this possibility will need to be addressed in the future.
Numb family proteins are known to structurally resemble an adaptor or scaffold protein that contains a PTB domain (42), a proline-rich carboxyl terminal region containing several putative Src homology 3 domain-binding sites (45), and an Eps15 homology domain-binding motif (46). Although the middle portions surrounding the CaM-KI phosphorylation sites (residues 288–347 in rat Numbl and residues 248–307 in rat Numb) (Fig. 3A) exhibit high homology in Numb and Numbl (87% identity), the function of this region has not yet been elucidated. Determination of phosphorylation in this conserved region in vitro and in vivo suggested that the middle portions of Numb/Numbl might exert similar function(s), such as protein/protein interaction, which could in turn be regulated by phosphorylation. Indeed, we demonstrated that 14-3-3 proteins were capable of interacting with Numb and Numbl in a phosphorylation-dependent manner at Ser264 and Ser304, respectively. Based on our results, it appears that one of the effects of 14-3-3 protein binding to phosphorylated Numb family proteins may be to block subsequent dephosphorylation, resulting in the maintenance of Numb/Numbl in a phosphorylated state. The 14-3-3-interaction region containing the CaM-KI phosphorylation site is located adjacent to the PTB domain of Numb/Numbl, which has been shown to be required for Notch1 ubiquitination and down-regulation of Notch1 nuclear activity (56). A recent study has shown that Numb interacts with collapsing mediator protein-2 through the PTB domain, resulting in the regulation of Numb-mediated endocytosis at the growth cone; moreover, both Numb and collapsing mediator protein-2 have been shown to co-localize in the central region of axonal growth cones in hippocampal neurons (57). It has already been demonstrated that the interaction of 14-3-3 protein with its target proteins results in the regulation of subcellular localization, catalytic activity, or protein-protein interaction of the target proteins (52–55). Therefore, the phosphorylation-dependent interaction with 14-3-3 protein may affect the function of the PTB domain of Numb family proteins. Also, the CaM-KK/CaM-KI cascade has been shown to regulate growth cone motility (7). Thus, this kinase cascade may be integrated into the Numb-mediated pathways involved in axonal growth. Since the physiological consequences of the phosphorylation-dependent interaction of Numb/Numbl with 14-3-3 proteins remain to be examined, future experiments specifically designed to explore this issue will be necessary in order to better understand the regulatory mechanisms of the Numb family proteins.
|
FOOTNOTES |
|---|
The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EBI Data Bank with accession number(s) AB210107
[GenBank]
(rat Numbl) and AB210108
[GenBank]
(rat Numb). 
2 Present address: Dept. of Pharmacology, Emory University School of Medicine, Rollins Research Center, Room 5172, 1510 Clifton Rd., Atlanta, GA 30322.
1 To whom correspondence should be addressed: Dept. of Signal Transduction Sciences, Faculty of Medicine, Kagawa University, 1750-1 Miki-cho, Kita-gun, Kagawa 761-0793, Japan. Tel./Fax: 81-87-891-2368; E-mail: tokumit{at}med.kagawa-u.ac.jp.
3 The abbreviations used are: CaM-K, Ca2+/CaM-dependent protein kinase; CaM-KK, CaM-K kinase; CaM, calmodulin; GST, glutathione S-transferase; DTT, dithiothreitol; CD, catalytic domain; CREB, cAMP-response element-binding protein; PKA, cAMP-dependent protein kinase, MS, mass spectrometry; LC, liquid chromatography; MS/MS, tandem mass spectrometry; PTB, phosphotyrosine binding.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
