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

J. Biol. Chem., Vol. 276, Issue 43, 39950-39958, October 26, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/43/39950    most recent M107807200v1 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 Zamora-Leon, S. P. Articles by Shafit-Zagardo, B. Search for Related Content PubMed PubMed Citation Articles by Zamora-Leon, S. P. Articles by Shafit-Zagardo, B. Social Bookmarking                      What's this?

Binding of Fyn to MAP-2c through an SH3 Binding Domain

REGULATION OF THE INTERACTION BY ERK2^*

S. Pilar Zamora-Leon, Gloria Lee^§, Peter Davies, and Bridget Shafit-Zagardo^¶

From the  Department of Pathology, Albert Einstein College of Medicine, Bronx, New York 10461 and the ^§ Department of Internal Medicine, The University of Iowa College of Medicine, Iowa City, Iowa 52242

Received for publication, August 14, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Microtubule-associated protein 2 (MAP-2) isoforms are developmentally expressed in the nervous system and contain a number of functional domains. Adjacent to the first repeat of the microtubule-binding domain is an RTPPKSP motif for binding SH3 domains. To identify SH3-containing proteins that interact with MAP-2, transfections, filter overlay assays, glutathione S-transferase (GST)-mediated binding assays, co-immunoprecipitations and enzyme-linked immunosorbent assays were performed. Transfections of MAP-2a, MAP-2b, and MAP-2c constructs into COS7 cells, followed by incubation of the cell lysates with SH3-GST fusion proteins, determined that the strongest interaction was between MAP-2c and the non-receptor tyrosine kinase Fyn; however, MAP-2b and MAP-2c also bound to Grb2. Co-immunoprecipitation of Fyn and MAP-2c from human fetal homogenates confirmed the interaction in vivo. MAP-2 synthetic peptides spanning the RTPPKSP motif bound to Fyn, and the interaction was regulated by phosphorylation. Co-transfections with MAP-2c and the extracellular signal-regulated kinase 2 (ERK2) demonstrated that MAP-2c is threonine/serine-phosphorylated on its RTPPKSP motif and that threonine phosphorylation abolished the MAP-2c/Fyn binding. Kinase assays and co-transfection of MAP-2c and Fyn confirmed that Fyn tyrosine kinase phosphorylates MAP-2c. Thus, the activation of signaling pathways may regulate cytoskeletal dynamics by altering the state of phosphorylation of MAP-2 by both ERK2 and Fyn kinase.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Microtubule-associated protein-2 (MAP-2)¹ belongs to the family of structural MAPs that bind microtubules (MT) and regulate cytoskeletal dynamics, the spacing between MT, dendritic elongation, and oligodendrocyte process outgrowth (1-3). A single MAP-2 gene encodes the multiple MAP-2 forms generated by alternative splicing (4-9). All MAP-2 isoforms are heat-stable phosphoproteins and are developmentally regulated in the nervous system. In rodents, MAP-2b and MAP-2c are expressed during early fetal development, followed by a decline of MAP-2c and an increase of MAP-2a (10-12). The most abundant high molecular weight MAP-2 isoforms, MAP-2a and MAP-2b, are composed of a large projection arm and a short microtubule-binding domain (MTBD). The low molecular weight isoforms MAP-2c and MAP-2d lack most of the projection arm of the high molecular weight forms, but retain the amino and carboxy portions of the molecule (13, 14). The amino terminus contains the binding domain for the regulatory subunit (RII) of cAMP-dependent protein kinase (15, 16), and the carboxy terminus contains a proline-rich region and the repeats of the MTBD (1-3). The proline-rich region consists of a stretch of PXXP motifs localized upstream of the first repeat of the MTBD. One of the PXXP repeats contains the sequence RTPPKSP, consistent with a class I ligand for the putative binding of SH3-containing proteins (17-19). Some of the proteins containing SH3 domains include the Src family of non-receptor tyrosine kinases (NRTKs), adaptor proteins such as Grb2 and Nck, the Abl tyrosine kinase, PLC, and spectrin. Among the Src family of NRTKs expressed in the CNS, Src, Fyn, and Yes are enriched in growth cones and growing processes (20-23). Fyn is abundantly expressed in neurons and oligodendrocytes, especially during process extension. In addition, Fyn is the only active NRTK in early oligodendrocyte development (23). Rat primary cultures of oligodendrocyte progenitor cells either infected with a Fyn dominant-negative vector or treated with specific inhibitors have shortened oligodendrocyte processes (23). MAP-2 isoforms expressed in neurons and oligodendrocytes contain the RTPPKSP motif, and the domain is conserved in Tau. It has been shown that Tau binds Fyn, Src, and Lck through this motif (24), suggesting an important interaction between the structural MAPs and Fyn.

To gain insight into the interaction between MAP-2 and SH3 ligands, we examined the binding of SH3-containing proteins to the RTPPKSP motif on MAP-2 isoforms and identified Fyn as a relevant interacting protein. We present biochemical evidence that Fyn can phosphorylate MAP-2c in vitro and in vivo and ERK2 can modulate the interaction by phosphorylating the RTPPKSP motif. The results depict another functional domain on MAP-2, and we speculate that this domain is capable of regulating cytoskeletal dynamics.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

cDNA Constructs-- Full-length human MAP-2c (pRC/CMV vector), full-length MAP-2b (pSV vector), full-length MAP-2a (pSV vector), and clone 900 (pcDNA3.1/His vector) cDNAs were constructed as previously described (8, 14, 25). SH3-Fyn-glutathione S-transferase (GST), SH3-Src-GST, neuronal SH3-Src-GST, SH3-Lyn-GST, SH3-Lck-GST, SH3-Crk-GST, SH3-Abl-GST, SH3-Nck-GST, SH3-PLC-GST, and SH3-spectrin-GST and full-length Grb2-GST cDNAs (pGEX vector) were obtained from Dr. Hamid Band (Brigham and Women's Hospital, Boston, MA). Full-length human ERK2 cDNA (pCEP4L vector) was obtained from Dr. Melanie Cobb (University of Texas Southwestern Medical Center, Dallas, TX). Full-length human wild-type Fyn and full-length K299M mutant Fyn cDNAs (pCMV5 vector) were obtained from Dr. Marilyn Resh (Memorial Sloan-Kettering Cancer Center, New York, NY).

Expression and Purification of GST Fusion Proteins-- The procedure was performed as previously described (26). Purification of the protein was confirmed by Western blot analysis with the GST-specific mAb DT-12.

Expression and Purification of Recombinant Human MAP-2c-- Full-length human MAP-2c (14) was cloned into the pQE-30 expression vector (Qiagen), and the construct was sequenced to verify the correct open reading frame. Induction of protein expression in bacteria and purification of 6xHis-tagged MAP-2c was performed as described by Qiagen, with some modifications. The salt concentration in the sonication buffer was increased to 500 mM NaCl. Following sonication and centrifugation the sample was boiled for 10 min and centrifuged again. The pH of the wash buffer was increased to 6.5, and the elution was done at pH 4.5. Purification of the protein was confirmed by Western blot analysis with the anti-MAP-2 mAb HM-2 (Sigma Chemical Co.).

Filter Overlay Assays-- Varying amounts (1-30 µg) of purified GST fusion proteins from Fyn, Src, and full-length Grb2 and GST protein without an insert were loaded in 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). After protein transfer, the nitrocellulose membranes were incubated with bacterially expressed human MAP-2c (10 µg/ml in 5% milk in 1× TBS (140 mM NaCl, 1 mM Tris-HCl) for 2 h at room temperature (RT), extensively washed with 1× TBS and 0.1% Tween-20, incubated with the anti-MAP-2 mAb HM-2 (1:500; 2 h at RT), followed by IgG₁-goat anti-mouse horseradish peroxidase conjugate (HRP) (1:10,000; 1 h at RT) and visualization by enhanced chemiluminescence (ECL, Amersham Pharmacia Biotech). The reverse experiment, where the bacterially expressed human MAP-2c was loaded on a gel, transferred to nitrocellulose, and incubated with the different GST fusion proteins was also performed.

Cell Cultures, Transient Transfections, and Protein Extractions-- COS7 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin at 37 °C in an 8% humidified CO₂ atmosphere. For transient transfections, COS7 cells were plated at a density of 1 × 10⁶ cells/100-mm dish the day before transfection. 5-10 µg of the respective cDNAs were mixed with LipofectAMINE (Life Technologies, Inc.) in serum-free medium as indicated in the manufacture's protocol. After 5 h the medium was replaced with complete medium. 24-48 h later the transfections were stopped. For total protein extraction, COS7-transfected cells were washed in 1× TBS, followed by addition of extraction buffer (1 mM Tris-HCl, pH 7.4, 140 mM NaCl, 2 mM EGTA, 2 µg/ml leupeptin, 4 µg/ml pepstatin, 5 mM sodium pyrophosphate, 30 mM -glycerolphosphate, 30 mM sodium fluoride, 0.1 mM AEBSF). All steps were performed on ice. The cells were scraped and centrifuged at 1,200 rpm in an Eppendorf centrifuge at 4 °C. Supernatants were discarded, and the pellets were resuspended in extraction buffer and placed at 70 °C for at least 30 min. The samples were thawed on ice, homogenized, and centrifuged at 14,000 rpm in an Eppendorf centrifuge for 5 min at 4 °C. Supernatants were saved, and the pellets were resuspended and reprocessed as described. Following centrifugation the supernatants were combined and protein quantification was performed (Bio-Rad protein assay). For heat-stable protein extraction, an aliquot of the total proteins was taken and NaCl (0.34 M final) and -mercaptoethanol (0.28 M final) were added. The samples were boiled for 5 min, and centrifuged at 14,000 rpm in an Eppendorf centrifuge for 5 min at 4 °C. Supernatants were saved, and protein concentration was quantified.

GST-mediated Binding Assays following Transient Transfection-- COS7 cells were lysed on ice in 1 mM Tris-HCl, pH 7.4, 140 mM NaCl, 2 mM EGTA, 5 mM sodium pyrophosphate, 30 mM -glycerolphosphate, 30 mM sodium fluoride, 2 µg/ml leupeptin, 4 µg/ml pepstatin, 0.1 mM AEBSF, and 0.5% Triton X-100. The homogenates were cleared by centrifugation (12,000 rpm in an Eppendorf centrifuge for 1 min), and the supernatants were incubated overnight at 4 °C with GST fusion proteins bound to glutathione-agarose beads. Subsequently, samples were washed 5× in extraction buffer, and 0.5% Triton X-100, pellets were resuspended in 1× sample and boiled, and the supernatants were separated in 10% SDS-PAGE.

Source of Human Fetal Tissue-- Fetal brain and spinal cord were obtained from elective terminations of pregnancies in the second trimester of gestation performed at affiliated hospitals of Albert Einstein College of Medicine. All studies were approved by the Clinical Investigation and the Health and Hospital Corporation of the City of New York. The brain and spinal cord used in this study were obtained from aborted fetuses of females without significant medical problems.

Preparation of Human Fetal Brain and Spinal Cord Homogenates-- Human fetal brain tissue (19 gestational weeks) was homogenized in extraction buffer at three times weight/volume and centrifuged at 22,900 × g for 20 min at 4 °C. The supernatant was saved, and the protein content was quantified. The brain protein homogenates were aliquoted and stored at 80 °C.

Human fetal spinal cord tissue (20 gestational weeks) was dissected from the meninges and homogenized in extraction buffer as described for brain. The sample was centrifuged at 12,000 rpm in an Eppendorf centrifuge for 20 min at 4 °C. The supernatant was retained, and protein quantification was performed.

GST-mediated Binding Assays Using Protein Homogenates from Fetal Nervous Tissue-- Protein homogenates from human fetal brain were thawed on ice and centrifuged at 250,000 × g for 25 min at 4 °C to eliminate association of microtubule-associated proteins with tubulin. 1 mg of the supernatant protein was mixed with extraction buffer and 0.5% Triton X-100 to a final volume of 1 ml and incubated with the GST fusion proteins as previously described. For the pull-down assays performed with the agarose-conjugated bacterially expressed and purified Fyn protein (Santa Cruz Biotechnology, sc-4040), the volume was adjusted to 1 ml with immunoprecipitation buffer (IP, 20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2.5 mM EDTA, 2 µg/ml leupeptin, 4 µg/ml pepstatin, 5 mM sodium pyrophosphate, 30 mM -glycerolphosphate, 30 mM sodium fluoride, 0.1 mM AEBSF, 0.5% Triton X-100). The samples were pre-cleared with 20 µl of resuspended protein-G PLUS-agarose (Santa Cruz Biotechnology) by incubation for 30 min at 4 °C followed by centrifugation at 5,000 rpm in an Eppendorf centrifuge for 5 min at 4 °C. The supernatants were collected and incubated with either 10 µg of purified Fyn protein coupled to agarose or 10 µl of resuspended protein-G PLUS-agarose as a control, and rocked overnight at 4 °C. Following centrifugation the pellets were washed five times in the immunoprecipitation buffer, resuspended in 1× sample buffer, and boiled for 5 min, and the supernatants were separated in a 10% SDS-PAGE.

One milligram of human fetal spinal cord protein homogenate was adjusted to 1 ml with IP buffer and processed similar to the brain homogenates with 10 µg of purified Fyn protein coupled to agarose or 10 µl of resuspended protein-G PLUS-agarose.

Immunoprecipitations and Co-immunoprecipitations-- For immunoprecipitations, transiently co-transfected COS7 cells were lysed on ice in IP buffer (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2 mM EGTA, 5 mM sodium pyrophosphate, 30 mM -glycerolphosphate, 30 mM sodium fluoride, 0.1 mM AEBSF, 1 mM orthovanadate, 2 µg/ml leupeptin, 4 µg/ml pepstatin, and 0.5% Triton X-100), and all subsequent steps were performed at 4 °C. Cellular debris was pelleted by centrifugation at 10,000 rpm in an Eppendorf centrifuge for 10 min; the supernatants were precleared with 1 µg of the control IgG_2b antibody plus 20 µl of resuspended protein-G PLUS-agarose (Santa Cruz Biotechnology), gently mixed for 30 min, and centrifuged at 5,000 rpm for 5 min. The supernatants were immunoprecipitated with either 10 µg of the mAb 4G10, that recognizes phosphotyrosine residues, or a control IgG_2b antibody, and mixed on a Nutator for at least 4 h. 20 µl of resuspended protein-G PLUS-agarose was added to the samples and rocked overnight. Following centrifugation, the pellets were washed five times in IP buffer, resuspended in 1× sample buffer, boiled for 5 min, and centrifuged, and the supernatants were separated in a 10% SDS-PAGE.

For co-immunoprecipitation studies, frozen human fetal brain homogenates were thawed on ice and centrifuged at 250,000 × g for 25 min at 4 °C, and all subsequent steps were performed at 4 °C. One milligram of the supernatant protein was adjusted to 1 ml with IP buffer, without orthovanadate, and 0.5% Triton X-100 final. The samples were pre-cleared with either 1 µg of an irrelevant IgG₁ control antibody or 1 µg of rabbit serum, plus 20 µl of resuspended protein-G PLUS-agarose. The homogenates were gently mixed for 30 min and centrifuged at 5,000 rpm for 5 min. The supernatants were incubated with either 10 µg of the anti-MAP-2 mAb HM-2 and an irrelevant IgG₁ antibody as a control, or with the dual specific Fyn/Src polyclonal antibody and rabbit serum. The samples were mixed on a Nutator for 2 h, and then 20 µl of resuspended protein-G PLUS-agarose was added to the samples and rocked overnight. Following centrifugation, the pellets were washed five times in the IP buffer, resuspended in 1× sample buffer, and boiled, and the supernatants were separated in a 10% SDS-PAGE.

Western Blot Analysis and Antibodies-- Following 10% SDS-PAGE (27), the proteins were transferred to nitrocellulose (28) and incubated with 5% non-fat dry milk in 1× TBS. After blocking, the membranes were incubated with the respective primary antibodies followed by HRP-conjugated secondary antibodies. Visualization was by enhanced chemiluminescent (Amersham Pharmacia Biotech or Pierce). When the Fyn mAb was used to determine if Fyn was co-immunoprecipitated with MAP-2, the membrane was first incubated with the secondary antibody to eliminate the reactivity with the IgGs that will interfere with the detection of Fyn, followed by the primary and secondary antibodies as usual.

Antibodies-- anti-MAP-2 mAb HM-2 (IgG₁, Sigma) recognizes an amino-terminal epitope of all MAP-2 isoforms. Tau46 (IgG₁, Zymed Laboratories Inc.) recognizes the carboxyl terminus of all MAP-2 isoforms and Tau. The monoclonal antibodies CP9 (phosphothreonine) (29), MC6 (phosphoserine) (30, 31), and DT-12 (anti-GST) were provided by Dr. Peter Davies. Anti-Fyn mAb was from Transduction Laboratories (F19720, IgG_2b). The dual specific Fyn/Src polyclonal antibody (anti-Src sc-18) was from Santa Cruz Biotechnology. The sheep anti-human pp60^c-src polyclonal antibody was from Chemicon International, Inc. The tyrosine-phosphorylated mAb 4G10 (IgG_2b) was from Upstate Biotechnology Inc. HRP-conjugated secondary antibodies were from Southern Biotechnology Associates, Inc.

Mutagenesis-- Mutant forms of human MAP-2c were made by polymerase chain reaction cloning as described under the QuikChange site-directed mutagenesis kit instruction manual (Stratagene). The primers used to mutate residues (underlined) within the RTPPKSP motif on MAP-2 are: for the Thr/Ala mutation: GTCGCCATCATACGTGCTCCTCCAAAATCTCCTG (sense), CAGGAGATTTTGGAGGAGCACGTATGATGGCGAC (antisense); for the Ser/Ala mutation: CGTACTCCTCCAAAAGCTCCTGCGACTACTCCC (sense), GGGAGTAGTCGCAGGAGCTTTTGGAGGAGTACG (antisense). The amplification protocol consisted of 18 cycles of the following conditions: denaturation for 0.3 min at 95 °C (the first cycle was 0.6 min), primer annealing for 1 min at 55 °C, chain extension for 20 min at 68 °C, and an extra 20 min for the last cycle. Both constructs were sequenced to verify the correct open reading frames and the mutation sites.

ELISA-- The MAP-2, Tau, and the irrelevant synthetic peptides (Table I) were synthesized with a biotin tag at the amino terminus and bound to neutravidin-coated ELISA plates at 20 ng/well in 2% bovine serum albumin, for 1 h at RT. Excess peptide was washed off and the plates were incubated overnight at 4 °C with either SH3-Fyn GST fusion protein or GST without insert in 2% bovine serum albumin at 0.25, 0.5, 1, and 2 µg/well, respectively. Binding was detected by incubation with the anti-GST mAb DT-12 (1:100; 1.5 h), followed by goat anti-mouse IgG₁-HRP (1:500; 1 h) and horseradish peroxidase substrates A and B (ABTS substrate, Bio-Rad, 30 min). Optical densities were recorded at 405 nm with an ELISA reader (TECAN). The experiments were performed in triplicate and repeated three times. Two-way analysis of variance statistical analysis was performed.

In Vitro Kinase Assay-- Bacterially expressed human 6xHis-tagged MAP-2c (1.8 µg) was incubated with 2.5 units of Fyn tyrosine kinase (Upstate Biotechnology Inc.) according to manufacture's protocol using a reaction volume of 20 µl and incubation at 37 °C for 90 min; the reaction was performed without radioactivity. Kinase reactions were immunoblotted with the phosphotyrosine mAb 4G10.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Human MAP-2 Interacts Specifically with the SH3 Domain of Fyn and Full-length Grb2 in GST-mediated Binding Assays and Blot Overlay Assays-- To examine the MAP-2 isoforms that interact with the SH3 domain of the NRTKs Fyn and Src and the adaptor protein Grb2, cDNAs to human MAP-2c, MAP-2b, MAP-2a, and clone 900, consisting of the last 900 bp of MAP-2, were transiently transfected into COS7 cells. Cell lysates were incubated with the SH3-Fyn, SH3-Src, or full-length Grb2-GST fusion proteins coupled to glutathione-agarose beads. Immunoblotting with the anti-MAP-2 mAb HM-2 indicated that MAP-2a did not interact with any of the GST fusion proteins (Fig. 1A, top panel). MAP-2b showed a stronger interaction with Grb2 than Fyn (Fig. 1B, top panel), whereas MAP-2c showed a very strong interaction with Fyn, an interaction with Grb2, and a minimal to no interaction with Src (Fig. 1C, top panel). Clone 900 interacted strongly with Fyn and Grb2-GST fusion proteins (Fig. 1D, top panel), demonstrating that the interaction between MAP-2 and the fusion proteins occurs within a region encompassing the last 900 bp of MAP-2. This region of the molecule contains a putative class I SH3 binding domain, the RTPPKSP motif, and the repeats of the MTBD. To confirm that the lanes contained equal amounts of fusion proteins, the blots were stripped and reincubated with the mAb DT-12, an antibody specific to the GST region of the fusion protein (Fig. 1, A-D, lower panels). In addition to examining Fyn, Src, and Grb2, interaction of MAP-2 with SH3-GST fusion protein from neuronal Src, Lyn, Lck, Crk, Abl, Nck, PLC, and spectrin were examined, and all the interactions were negative (data not shown).

View larger version (62K): [in this window] [in a new window]   Fig. 1.   The last 900 bp of MAP-2c are responsible for the interaction with the SH3 domain of Fyn and full-length Grb2. Following transient transfection of MAP-2a (A), MAP-2b (B), MAP-2c (C), and clone 900 cDNAs (D) into COS7 cells, cell lysates were incubated with SH3-Fyn, SH3-Src or full-length Grb2-GST fusion proteins or GST protein without insert. Nitrocellulose membranes were incubated with the anti-MAP-2 mAb HM-2 (upper panels, A-C) or the Tau46 mAb, which recognizes the carboxyl terminus of MAP-2 (upper panel E). The lower panels (A-D) show incubation with mAb DT-12 and demonstrate that the amounts of fusion proteins were equal. All the blots contained total cell homogenates prior to pull-down (homog).

To further examine the interaction of MAP-2c with Fyn, Src, and Grb2, bacterially expressed human MAP-2c was incubated with varying amounts of nitrocellulose-bound GST fusion proteins expressing SH3-Fyn, SH3-Src, and full-length Grb2. As shown in Fig. 2, MAP-2c bound in a concentration-dependent manner to all of the fusion proteins; however, MAP-2c binding to Fyn was much stronger than to Src or Grb2. Minimal to no binding was observed with GST protein without insert. The reverse experiment, where varying amounts of bacterially expressed human MAP-2c were bound to nitrocellulose membranes and incubated with the respective GST fusion proteins, gave the same results (data not shown).

View larger version (58K): [in this window] [in a new window]   Fig. 2.   Bacterially expressed human MAP-2c interacts with the SH3 domain of Fyn and Src and full-length Grb2 in filter overlay assays. Nitrocellulose containing varying amounts of SH3-Fyn, SH3-Src, and full-length Grb2-GST fusion proteins were incubated with bacterially expressed human MAP-2c followed by incubation with the anti-MAP-2 mAb HM-2. Fyn and Src panels, lanes 1-5: 1 µg, 4 µg, 7.5 µg, 15 µg, and 30 µg of the respective fusion protein; lanes 6-9, GST protein without insert: 30 µg, 15 µg, 7.5 µg, and 4 µg. Grb2 panel, lanes 1-4: 1.25 µg, 2.5 µg, 5 µg, and 10 µg; lanes 5-7: GST protein without insert: 2.5 µg, 5 µg, and 10 µg.

GST-mediated binding assays were performed with human fetal brain and the SH3 domains of Fyn, Src, and full-length Grb2-GST fusion proteins to examine the interactions between the GST fusion proteins with the brain endogenous MAPs. As shown in Fig. 3A, MAP-2c was the predominant MAP-2 form interacting with SH3-Fyn and Grb2; MAP-2c did not efficiently interact with SH3-Src. Consistent with the data shown in Fig. 1B, MAP-2b interacted with Grb2. To confirm that the lanes contained equal amounts of fusion proteins, the blots were stripped and reincubated with the GST-specific DT-12 mAb (Fig. 3A, lower panel). As a result of the strong Fyn and MAP-2c binding, we focused our studies on this interaction.

View larger version (49K): [in this window] [in a new window]   Fig. 3.   MAP-2c interacts with Fyn and Grb2 in human fetal brain GST-mediated binding assays. One milligram of human fetal brain protein homogenate was incubated with, A, SH3-Fyn, SH3-Src, full-length Grb2-GST fusion proteins or GST protein without insert; and B, Fyn protein coupled to agarose, or protein G PLUS-agarose (control) as described under "Experimental Procedures." Homogenates prior to pull-down (homog) and the precipitates were separated in 10% SDS-PAGE and transferred to nitrocellulose membranes. The blots were incubated with the anti-MAP-2 mAb HM-2 (A, upper panel and B) or with mAb DT-12 (A, lower panel).

To confirm the MAP-2c interaction observed with the Fyn-SH3 domain, purified Fyn protein coupled to agarose was incubated with human fetal brain homogenate. As shown in Fig. 3B, only MAP-2c was present in the proteins associating with Fyn, further supporting an interaction between these proteins in the human fetal CNS.

MAP-2 Co-immunoprecipitates with Fyn from Human Fetal Brain and Spinal Cord Homogenates-- To explore an in vivo interaction between MAP-2c and Fyn or Src in brain we used the anti-MAP-2 HM-2 mAb and determined that Fyn was co-immunoprecipitated (Fig. 4A, upper panel). A Src-specific polyclonal antibody (Chemicon) did not detect Src, indicating that Src did not co-immunoprecipitate with MAP-2c (data not shown). As expected, incubation of the membrane with the anti-MAP-2 mAb confirmed the presence of MAP-2 in the pellet (Fig. 4A, lower panel). The MAP-2c/Fyn interaction was further confirmed by immunoprecipitations performed with a carboxy-terminal polyclonal antibody that recognizes Fyn and Src (Santa Cruz Biotechnology). As shown in Fig. 4B, upper panel, this dual-specific antibody immunoprecipitated MAP-2c and Fyn (determined with a specific mAb that does not cross-react with Src; Fig. 4B, lower panel). Fyn antibodies that recognized epitopes at the amino terminus of the protein failed to immunoprecipitate MAP-2 from human fetal brain homogenates (data not shown), and we speculate that the binding of MAP-2 to Fyn might prevent the binding of these amino-terminal-specific antibodies to Fyn. The results and the data observed with the dual-specific Fyn/Src antibody and the HM-2 mAb suggest that the MAP-2c interaction is with Fyn and not with Src and supports an Fyn/MAP-2c interaction in human fetal brain.

View larger version (37K): [in this window] [in a new window]   Fig. 4.   Fyn co-immunoprecipitates with MAP-2c from fetal human brain and spinal cord homogenates. One milligram of total brain protein homogenate (A and B) or spinal cord protein homogenate (C and D) was incubated with either 10 µg of the anti-MAP-2 mAb HM-2 (A and C), 10 µg of the Fyn/Src dual polyclonal antibody (B), or 10 µg of Fyn protein coupled to agarose (D), as described under "Experimental Procedures." Total protein homogenates prior to immunoprecipitations (homog), and pellets following immunoprecipitation were separated in 10% SDS-PAGE. A and C, IP with the anti-MAP-2 mAb HM-2 or an irrelevant control mAb; the nitrocellulose membrane was incubated with a Fyn mAb. The lower blot shows the membrane incubated with the mAb HM-2; B, IP with the Fyn/Src polyclonal antibody or rabbit serum (control); the nitrocellulose membrane was incubated with the anti-MAP-2 mAb HM-2. The lower blot shows the membrane incubated with a Fyn mAb. D, GST-mediated binding assays with Fyn protein coupled to agarose or protein G PLUS-agarose (control); the nitrocellulose membrane was incubated with the anti-MAP-2 mAb HM-2.

In human fetal spinal cord homogenate, the interaction between MAP-2c and Fyn was also detected following immunoprecipitation with the anti-MAP-2 mAb HM-2 (Fig. 4C) or association with the Fyn protein coupled to agarose (Fig. 4D). This further supports an interaction between Fyn and MAP-2c in vivo.

Co-transfection of MAP-2c and ERK2 Results in Threonine and Serine Phosphorylation of the RTPPKSP Motif on MAP-2c-- Previous studies have determined that MAP-2 is a substrate for ERK2 and that T/P and S/P are minimal phosphorylation sites for this MAP kinase. To determine if ERK2 phosphorylates the threonine and/or serine residues on the RTPPKSP motif on MAP-2c, we used the mAbs CP9 (29) and MC6 (30, 31), which recognize, respectively, either phosphothreonine or phosphoserine within the RTPPKSP motif. COS7 cells were transiently co-transfected with ERK2 and wild-type MAP-2c (wt). As shown in Fig. 5A, a minimum level of immunoreactivity with mAb CP9 or MC6 was observed, however, co-transfection of wt MAP-2c with ERK2 results in a dramatic increase in phosphorylation on both threonine and serine residues. Phosphorylation was greatly diminished in the presence of the MEK inhibitor U0126 (10 µM; Promega). Cells homogenates from non-transfected COS7 cells were immunoblotted and shown to be negative for CP9, MC6, and HM-2 immunoreactivity (data not shown).

View larger version (25K): [in this window] [in a new window]   Fig. 5.   Co-transfection of MAP-2c and ERK2 results in MAP-2c phosphorylation at the RTPPKSP motif. COS7 cells were transiently transfected with wt MAP-2c or MAP-2c mutated either at threonine (Thr/Ala) or serine (Ser/Ala) within the RTPPKSP motif, with or without ERK2. The MEK inhibitor U0126 was added to the medium 24 h following transfections. Heat-stable (hs; 3 µg each) or total protein homogenates (tp; 8 µg each) were extracted and separated by gel electrophoresis. The nitrocellulose membranes contain proteins derived from wt MAP-2c co-transfections (A), mutant (Thr/Ala)-MAP-2c co-transfections (B), and mutant (Ser/Ala)-MAP-2c co-transfections (C). All immunoblots were incubated with the anti-MAP-2 mAb HM-2, CP9, MC6, or ERK2 mAb. The threonine and the serine within the RTPPKSP domain are highly phosphorylated on wt MAP-2c when co-transfected with ERK2.

To confirm that CP9 was specific for phosphothreonine, and MC6 was specific for phosphoserine, the MAP-2c mutated at either the threonine residue (Thr/Ala) or the serine residue (Ser/Ala) within the RTPPKSP motif, were tested for immunoreactivity following co-transfection with ERK2. As shown in Fig. 5, B and C, the Thr/Ala MAP-2c mutant was negative with CP9, whereas the Ser/Ala MAP-2c mutant was negative for MC6 immunoreactivity. This demonstrates the specificity of these mAbs for the specific phosphoepitopes within the RTPPKSP motif of MAP-2c.

Surprisingly, we observed that when the threonine was mutated (Thr/Ala-MAP-2c), no phosphorylation on serine was detected by MC6, suggesting that threonine phosphorylation is needed prior to serine phosphorylation. In addition, we detected an increase in CP9 immunoreactivity when the neighboring serine residue of the MAP-2c RTPPKSP motif was mutated to an alanine (Ser/Ala-MAP-2c), suggesting that the loss of serine phosphorylation enhances phosphorylation on threonine. In the presence of U0126, we observed decreased phosphorylation on threonine but not total elimination of kinase activity, suggesting that the mutation may generate a novel sequence (RTPPKAP) that is recognized by another kinase in COS7 cells. If the activity of this kinase is additive to that of ERK2, it may be responsible for the enhanced phosphorylation of the Ser/Ala-MAP-2c protein.

Phosphorylation of MAP-2c by ERK2 Abolishes the Interaction with Fyn-- To determine if ERK2 phosphorylation within the RTPPKSP motif on MAP-2c affects the Fyn/MAP-2c interaction, transient transfections of MAP-2c with or without ERK2, followed by incubation of the cell lysates with SH3-Fyn-GST fusion protein or GST protein without insert were performed. As shown in Fig. 6 (upper panel), the MAP-2c pulled down with SH3-Fyn was not phosphorylated on either threonine or serine within the RTPPKSP motif, because the immunoreactivities of the pellets were negative for the mAbs CP9 and MC6. The results suggest that phosphorylation within these residues blocks the binding of Fyn to MAP-2c. Stripping of the membrane followed by incubation with Tau46 (middle panel), an mAb that recognizes the carboxy terminus of the protein, shows that MAP-2c was pulled down in all the cases. Incubation with the GST mAb (lower panel) indicated equal amounts of fusion proteins. Cell lysates prior to the GST-mediated binding assays, derived from MAP-2c and ERK2 co-transfections, were positive for CP9 and MC6 immunoreactivity, as well as for the anti-MAP-2 mAb HM-2 and ERK2 mAb (data not shown).

View larger version (33K): [in this window] [in a new window]   Fig. 6.   Threonine phosphorylated MAP-2c is not pulled-down with SH3-Fyn-GST fusion protein. Following transient transfection into COS7 cells of MAP-2c with or without ERK2, cell lysates were incubated with 40 µg of SH3-Fyn-GST or GST protein without insert. The pellets were separated in 10% SDS-PAGE. After protein transfer, the nitrocellulose membranes were incubated with the mAbs HM-2, CP9, or MC6. There is no interaction between Fyn and MAP-2c phosphorylated in either threonine or serine residues within the RTPPKSP motif (upper panel). After stripping, the nitrocellulose blots were incubated with Tau46 (middle panel). Equal amounts of fusion proteins were determined with the mAb DT-12 (lower panel).

In Human Fetal Development the RTPPKSP Motif of MAP-2c Is Phosphorylated on Threonine but Not on Serine-- To determine if the threonine and serine residues within the RTPPKSP motif are phosphorylated in vivo, immunoblots containing heat-stable human fetal brain homogenates were incubated with CP9 and MC6. The anti-MAP-2 mAbs HM-2 and Tau46, an antibody that recognized both MAP-2c and Tau native sequences, were used to compare the ratio of phosphorylation to the total amount of MAP-2c protein. As shown in Fig. 7A, a fraction of the total MAP-2c was phosphorylated on threonine whereas on serine no phosphorylation was observed. By contrast, Tau, shown to migrate at about 50 kDa, was phosphorylated on both serine and threonine residues in vivo (Fig. 7A).

View larger version (27K): [in this window] [in a new window]   Fig. 7.   Human fetal brain MAP-2c phosphorylated on threonine within the RTPPKSP motif does not interact with Fyn. A, 5 µg of heat-stable proteins from human fetal brain homogenates were separated in 10% SDS-PAGE, transferred to a nitrocellulose membrane, and incubated with HM-2, CP9, MC6, and Tau46 mAbs. B, GST-mediated binding assays with the Fyn protein coupled to agarose or protein-G PLUS-agarose (control). The nitrocellulose membranes were incubated with HM-2, CP9, or MC6 mAbs. MAP-2c phosphorylated on threonine is not pulled-down with Fyn.

Phosphorylation of Threonine within the RTPPKSP Motif of Human Fetal MAP-2c Blocks the Interaction with Fyn-- To determine whether threonine phosphorylation on MAP-2c interferes with Fyn binding in vivo, Fyn protein coupled to agarose was used to pull down MAP-2c from human fetal brain homogenates and was examined for CP9 and MC6 immunoreactivity. MAP-2c was pulled down with the Fyn protein (HM-2 immunoreactivity) but was CP9- and MC6-negative (Fig. 7B). We have considered that, because only a subfraction of the total brain MAP-2c is threonine-phosphorylated within the RTPPKSP motif, any threonine-phosphorylated MAP-2c that might be in the pellet following pull-down assays with Fyn could be below the level of detection. Nevertheless, the results agree with the GST-mediated binding assays of MAP-2c and SH3-Fyn-GST fusion protein where the CP9 and MC6 immunoreactivity were also negative (Fig. 6) and suggest that phosphorylation of the threonine (RTPPKSP) blocks the interaction between MAP-2c and Fyn in vivo.

Fyn Interacts with the RTPPKSP Motif on MAP-2c-- To confirm whether the putative class I SH3-binding motif on MAP-2c binds to Fyn and to determine the essential amino acids required for the interaction, MAP-2 synthetic peptides spanning the RTPPKSP motif were used in ELISAs. The various peptides and their mutations are listed in Table I. The RP peptide (SEKKVAIIRTPPKSPAT) depicts the wt MAP-2 sequence. Additional peptides were synthesized with a mutation in the arginine (AP peptide), the second proline (RA peptide), or both the arginine and proline (AA peptide); all these residues were replaced with an alanine. Other mutations outside of this domain but not affecting the consensus RTPPKSP motif were also made (peptide 02). Because we observed threonine phosphorylation in vivo we generated the synthetic peptide 03, containing a phosphothreonine. Tau shares a common RTPPKSP motif and was shown previously to bind to Fyn; therefore, we generated the synthetic peptide 813 consisting of the Tau motif, and peptide 1085, which contains a phosphothreonine residue within the motif. An irrelevant Tau peptide (935) was used as a control to assess background or basal binding in the context of the assay.

View larger version (15K): [in this window] [in a new window]   Fig. 8.   The SH3 domain of Fyn binds to the RTPPKSP-binding motif on MAP-2: phosphorylation at threonine or mutation at arginine dramatically reduces the binding. Synthetic peptides coupled to neutravidin-coated ELISA plates at 20 ng/well and incubated with either SH3-Fyn GST fusion protein or GST without insert at 0.25, 0.5, 1, and 2 µg/well, respectively. Binding of Fyn to the synthetic peptides was detected by incubation with an anti-GST mAb, followed by goat anti-mouse IgG1-HRP and ABTS substrate. Optical densities were recorded at 405 nm. The data are presented as the relative reading of Fyn-GST minus GST only. Peptides RP, RA, and 02 showed binding to Fyn, whereas binding to peptides AP, 03, and 935 was not apparent; peptide AA showed only a very reduced but not significant interaction at 2 µg/well. The Tau synthetic peptide 813 also bound Fyn, and the threonine phosphorylation (1085 peptide), as the threonine-phosphorylated MAP-2 peptide (03), blocked the binding. The results are representative of three independent experiments.

As shown in Fig. 8, the ELISA analysis demonstrated that the SH3 domain of Fyn interacts with the RTPPKSP motif of MAP-2 (peptides RP and 02). Also, Fyn interacted with Tau (peptide 813), although the binding observed for MAP-2 appeared stronger than with the Tau peptide. Note that the irrelevant peptide 935 does not bind to Fyn. The AP peptide containing the Arg to Ala mutation abolished the interaction between MAP-2 and Fyn, confirming the importance of the arginine residue to classify MAP-2 as a class I ligand for the SH3 domain of Fyn. Statistical analysis of the interaction between MAP-2 and Fyn indicated that there is no difference in the affinity of Fyn to RP, 02, and 813 (p values > 0.1). There is also no statistical difference in the binding with RA (p value > 0.1), indicating that the mutation of the second Pro of the RTPPKSP motif does not play a crucial role in the interaction between MAP-2 and Fyn. As expected, the double-mutant (AA peptide) inhibited the binding of MAP-2 to Fyn. In addition, the phosphorylation of threonine (peptide 03) dramatically reduced the binding of Fyn to MAP-2. Furthermore, when Tau is phosphorylated at this site there is reduced binding to Fyn (peptide 1085) indicating that this is a potential means of regulating the binding of Fyn to MAP-2 and Tau. These data are consistent with the co-transfection and brain GST-mediated binding assays (Figs. 6 and 7B), which demonstrated that Fyn binding to MAP-2c is abolished in the presence of a phosphothreonine in the context of the RTPPKSP motif.

MAP-2c Is Tyrosine Phosphorylated by Fyn-- Considering that Fyn is a tyrosine kinase, we sought to determine if the binding of Fyn to MAP-2c leads to tyrosine phosphorylation of MAP-2c. An in vitro kinase assay containing purified bacterially expressed human MAP-2c and active Fyn kinase was performed and demonstrated that MAP-2c was tyrosine-phosphorylated by the Fyn kinase. The observed doublet (Fig. 9A, lane 2) is the result of protein degradation, because immunoblotting with Tau46 gave the same pattern.

View larger version (29K): [in this window] [in a new window]   Fig. 9.   MAP-2c is tyrosine-phosphorylated by Fyn. A, in vitro kinase assay showing that MAP-2c is tyrosine-phosphorylated by active Fyn kinase. Bacterially expressed human MAP-2c (1.8 µg) was incubated with active Fyn kinase (2.5 units; lane 2). The blot shows an immunoreactive doublet band with the phosphotyrosine mAb 4G10 only when active Fyn kinase was incubated with MAP-2c (lane 2). The doublets are MAP-2-positive and result from proteolysis (data not shown). B, Fyn tyrosine phosphorylates MAP-2c in co-transfected COS7 cells. The cells were transiently co-transfected with MAP-2c and wt Fyn (lanes 5 and 6) or MAP-2c and K299M Fyn (lanes 7 and 8). After 24 h the cells were lysed and immunoprecipitated with the phosphotyrosine mAb 4G10 (lanes 2 and 3) or an irrelevant control mAb (lanes 1 and 4). Homogenates prior to pull-down (homog) and the precipitates were separated in 10% SDS-PAGE and transferred to a nitrocellulose membrane and incubated with the anti-MAP-2 HM-2 mAb. A positive band was detected in the pellet only when MAP-2c was co-transfected with wt Fyn (lane 3).

To determine if Fyn can phosphorylate MAP-2c in a cellular environment, COS7 cells were co-transfected with MAP-2c and either wt Fyn or the K299M mutant Fyn, which is inactive due to a point mutation in the ATP binding pocket. Cell lysates were immunoprecipitated with the phosphotyrosine mAb 4G10, or an irrelevant IgG_2b mAb. As shown in Fig. 9B, MAP-2c was immunoprecipitated only when co-transfected with wt Fyn, demonstrating that the Fyn tyrosine kinase phosphorylates MAP-2c in a transfected cell system.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In this study we have demonstrated a novel interaction between the RTPPKSP motif on MAP-2c and the SH3 domain of Fyn and determined that Fyn is able to tyrosine-phosphorylate MAP-2c. We provide evidence that these two proteins bind in vivo during human fetal development, and the MAP-2c/Fyn interaction is regulated by ERK2 phosphorylation of the threonine within the RTPPKSP motif, a previously unreported site of ERK2 phosphorylation on MAP-2.

We explored the association of MAP-2a, MAP-2b, and MAP-2c with different SH3-containing proteins and determined that MAP-2c is the predominant isoform to interact with the SH3 domain of Fyn. The adaptor protein Grb2 (full-length) also bound MAP-2c but the association was weaker than the MAP-2c/Fyn interaction (Fig. 1C). By contrast, Grb2 bound to MAP-2b better than Fyn (Fig. 1B), and by immunoprecipitation Grb2 bound MAP-2b more efficiently than MAP-2c (data not shown), suggesting that, although these domains are conserved on MAP-2, different MAP-2 isoforms can preferentially bind distinct SH3-containing proteins. MAP-2b is expressed throughout development, and its levels remain fairly constant in the adult rat brain (1). The fact that MAP-2b binds more efficiently to the adaptor protein Grb2 suggest that this isoform serves as a scaffold protein (32), linking the MT network with signal transduction. No interaction was observed between MAP-2a and Fyn, Grb2, or Src indicating that there is a functional difference between MAP-2a and the other MAP-2 isoforms. For the interaction with Fyn, the functional difference is not evident from the primary structure, because all MAP-2 forms contain the motif and yet MAP-2a does not interact with Fyn. Although an interaction between rat MAP-2c and Src has been reported, in our hands the interaction with Src and human MAP-2c was minimal. Also, the reported interaction was not within the RTPPKSP motif but within the repeats of the MTBD (32).

The fact that MAP-2c was the predominant MAP-2 isoform to bind Fyn and the only one to co-immunoprecipitate with Fyn suggests that there is a developmentally important role for this interaction. MAP-2c is expressed early in development and in neuronal cell bodies and dendrites (11). In human fetal spinal cord MAP-2c is also expressed in the axon (33). MAP-2c is down-regulated in the mature CNS during synaptogenesis. Its continued expression in photosensitive cells of the adult retina and in the olfactory system, suggests that MAP-2c may allow for flexibility and rearrangement in neural cells (34-37). MAP-2c is a poor promoter of MT polymerization, possibly allowing for neural plasticity (38). Fyn is highly expressed in brain and spinal cord throughout development and is expressed early in neurons and glia subpopulations (20, 21, 23, 39), including oligodendrocyte progenitors and differentiated oligodendrocytes (23). Fyn localizes to the cell body and processes and is the only active NRTK in early oligodendrocytes (23). Fyn dominant-negative studies and cultures derived from Fyn-deficient mice exhibit reductions in the length of axons, dendrites, and oligodendrocyte processes (23, 40). Also, the extent of myelination is decreased in fyn-deficient mice (41, 42).

In this study we demonstrated that MAP-2c is tyrosine-phosphorylated both in vitro and in a transfected cell system by Fyn (Fig. 9, A and B). MAP-2c has five putative tyrosine-phosphorylatable residues, and future studies will define the site(s) phosphorylated by Fyn. The stimulus responsible for inducing Fyn to bind to and tyrosine-phosphorylate MAP-2c is not known. It has been shown that Src does not phosphorylate MAP-2c (32), and this is consistent with its poor binding efficiency in our GST-mediated binding assays (Fig. 1) and the inability of an Src-specific antibody to co-immunoprecipitate MAP-2c in vivo (data not shown). Although it was reported that Src could co-immunoprecipitate rat MAP-2 in vivo, a Src/Fyn dual-specific antibody was used and an interaction between MAP-2 and Fyn was not investigated. In addition, using an MAP-2-specific antibody, Src was not shown to co-immunoprecipitate with MAP-2 (32). Fyn kinase may associate with MAP-2c to regulate cytoskeletal dynamics. Furthermore, whether tyrosine-phosphorylated MAP-2c affects its interaction with MTs is not known. Fyn has been shown to bind tubulin via its SH2 domain (43); however, it is not known if it can simultaneously bind to MAP-2c via its SH3 domain.

By using MAP-2 synthetic peptides spanning the RTPPKSP motif in an ELISA, we demonstrated that MAP-2c specifically binds the SH3 domain of Fyn; the MTBD is not required for the interaction. Also, we demonstrated that the arginine to alanine mutation (AP peptide) blocked the binding, consistent with the documented importance of arginine in the binding of the SH3 domain of Fyn (Fig. 8) (17-19). In addition, phosphorylation on the threonine residue (peptide 03) of the RTPPKSP domain abolished the binding (Fig. 8). Tau binds to the SH3 domain of Fyn, Src, and Lck, and the interaction is mediated through the proline-rich motif (24) that shares the same amino acids present in MAP-2, although the flanking amino acids residues vary (Table I, peptide 813 versus RP). Fyn bound more efficiently to the non-phosphorylated 02 MAP-2 synthetic peptide (bar value equals 0.438) than the non-phosphorylated Tau synthetic peptide (bar value equals 0.381).

Because the peptide data demonstrated that phosphorylation on threonine abolishes Fyn's ability to bind to MAP-2c, we generated two constructs containing an alanine substitution for either the threonine or serine within the RTPPKSP motif of MAP-2c and examined whether the threonine or serine residues were required for Fyn binding. When the Thr/Ala and Ser/Ala MAP-2c mutant constructs were used to transfect COS7 cells and GST-mediated binding assays were performed with the SH3-Fyn-GST fusion protein, both constructs were able to interact with the SH3 domain of Fyn (data not shown). This indicates that the mutation of the threonine or serine within the RTPPKSP motif is not essential for the binding and, consequently, not as critical as the threonine phosphorylation, where the binding is abolished.

Although our data do not directly address the question of whether the RTPPKSP motif is the sole SH3 interacting motif in MAP-2c that Fyn binds, it is the only sequence to contain a consensus class I SH3-binding motif; MAP-2c does not contain a consensus class II motif (17). We have determined that ERK2 phosphorylates this motif and this phosphorylation inhibits the binding of Fyn to MAP-2c. Thus, we infer that this is the major Fyn binding site, and the peptide analysis supports this assumption.

It is known that SH3 domains play important roles in protein-protein interactions and are found in a variety of signaling proteins (44, 45). It has been shown that SH3 domains are involved in the control of cell morphology (46, 47) and targeting of proteins to specific subcellular localizations (48). In addition, SH3 domains bind with moderate affinity to proline-rich motifs (17), implying that the binding proteins can exchange easily. This suggests that the binding of MAP-2c with Fyn could be a short-lived interaction. The association of Fyn to the RTPPKSP motif of MAP-2c depicts an additional functional domain on MAP-2c.

MAP-2 is a substrate of several protein kinases, and the majority of the phosphorylation sites are conserved from rodent to human (1-3). Kinases known to phosphorylate MAP-2 include cAMP-dependent protein kinase A (49), protein kinase C (50, 51), GSK3 (52, 53), and the extracellular signal-regulated kinase ERK2 (52, 54, 55). Highly expressed in brain (56), ERK2 has been shown to phosphorylate MAP-2 in vitro, and some specific sites have been identified (52, 54, 55); in addition, some studies indicate that ERK2 phosphorylates MAP-2 in vivo (55, 57). Here we report the previously uncharacterized phosphorylation of the threonine and serine residues within the RTPPKSP motif on MAP-2c; the phosphorylation is mediated by ERK2 and is blocked by the addition of the MEK inhibitor U0126 in MAP-2c and ERK2 co-transfected COS7 cells (Fig. 5). Using the phosphorylated-specific mAbs CP9 and MC6, we determined that mutating threonine to alanine eliminated phosphorylation of both the threonine and serine residues, indicating that threonine phosphorylation is required for phosphorylation of the neighboring serine residue (Fig. 5B). By contrast, the serine to alanine mutation enhanced the phosphorylation on the neighboring threonine, because CP9 immunoreactivity was increased relative to the wild-type MAP-2c (Fig. 5, compare C with A). We considered whether a change in the conformation of MAP-2c as a result of the serine to alanine substitution might have caused the threonine to be more readily accessible for ERK2 phosphorylation. Surprisingly, when the serine to alanine MAP-2c mutant and ERK2 were co-transfected in the presence of U0126, there was only a slight decrease in threonine phosphorylation (Fig. 5C). This suggests that the mutation generates a sequence recognized by another kinase in the context of the RTPPKAP motif.

Immunoblots of human fetal brain homogenates determined that Tau is heavily phosphorylated on the threonine and serine residues within the RTPPKSP motif, whereas only a subfraction of MAP-2c is threonine-phosphorylated in vivo; no serine phosphorylation within the motif on MAP-2c was detected with MC6 (Fig. 7A). Pull-down results from MAP-2c and ERK2 co-transfected COS7 cells in addition to ELISA analyses with the phosphopeptide (03) support the conclusion that threonine phosphorylation blocks the interaction between MAP-2c and Fyn. In that way, the development of neural cells could be regulated by the localization of Fyn and MAP-2c as well as ERK2 phosphorylation. Although it is known that threonine and serine phosphorylation of MAP-2 or Tau alters binding to the MT (49, 58-60), we have no evidence that threonine or serine phosphorylation within the RTPPKSP motif of MAP-2c alters microtubule dynamics. Also, it is possible that MAP-2c bound to Fyn can bind to microtubules.

The involvement of MAP-2 in the propagation of intracellular signaling through protein-protein interactions is an area that is not fully characterized. MAP-2 has a number of binding domains that regulates its function, including the binding domain for the regulatory subunit (RII) of cAMP-dependent protein kinase A (15, 16), an actin-binding domain (61), the three to four repeats of the MTBD, and now the RTPPKSP motif characteristic of a class I ligand for the SH3 domain of Fyn (17-19). We postulate that the binding of Fyn to MAP-2c activates a novel intracellular signaling pathway that regulates cytoskeletal dynamics. The Fyn/MAP-2c interaction would be fine-tuned by ERK2 phosphorylation of MAP-2c, leading to an "on-off" cycling of this protein interaction.

	ACKNOWLEDGEMENTS

We thank Dr. Marilyn Resh for the gifts of the human Fyn wild-type and mutant constructs, Dr. Melanie Cobb for the gift of the human ERK2 construct, Dr. Hamid Band for the SH3 fusion constructs, Kathleen O'Guin for her help in the purification of plasmids and fusion proteins, and Dr. Brad Polous and Dr. Karen Weidenheim and the Einstein Human fetal tissue bank for material. Data in this paper are from the thesis of S. P. Zamora-Leon to be submitted in partial fulfillment of the requirements for the Degree of Doctor of Philosophy in the Sue Golding Graduate Division of Medical Sciences, Albert Einstein College of Medicine, Yeshiva University.

	FOOTNOTES

^* This work was supported by National Multiple Sclerosis Society Grant RG3020 and National Institutes of Health Grant NS38102 (to B. S. Z.) and by National Institute of Mental Health Grants 38623 (to P. D.) and NS32100 (to G. L.).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.

^¶ To whom correspondence should be addressed: Dept. of Pathology, Albert Einstein College of Medicine, Bronx, NY 10461. Tel.: 718-430-2189; Fax: 718-430-8541; E-mail: zagardo@aecom.yu.edu.

Published, JBC Papers in Press, August 23, 2001, DOI 10.1074/jbc.M107807200

	ABBREVIATIONS

The abbreviations used are: MAP-2, microtubule-associated protein 2; MT, microtubule; MTBD, microtubule-binding domain; ERK2, extracellular signal-regulated kinase 2; GSK3, glycogen synthase kinase 3; GST, glutathione S-transferase; His-tag-MAP-2c, histidine-tagged MAP-2c; IP, immunoprecipitation; mAb, monoclonal antibody; NRTK, non-receptor tyrosine kinase; RT, room temperature; wt, wild-type; bp, base pair(s); PLC, phospholipase C; CNS, central nervous system; CMV, cytomegalovirus; PAGE, polyacrylamide gel electrophoresis; HRP, horseradish peroxidase; AEBSF, 4-(2-aminoethyl)benzenesulfonylfluoride hydrochloride; ELISA, enzyme-linked immunosorbent assay; MEK, mitogen-activated protein kinase/extracellular signal-regulated kinase kinase.

	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:

				C. H. Reynolds, C. J. Garwood, S. Wray, C. Price, S. Kellie, T. Perera, M. Zvelebil, A. Yang, P. W. Sheppard, I. M. Varndell, et al. Phosphorylation Regulates Tau Interactions with Src Homology 3 Domains of Phosphatidylinositol 3-Kinase, Phospholipase C{gamma}1, Grb2, and Src Family Kinases J. Biol. Chem., June 27, 2008; 283(26): 18177 - 18186. [Abstract] [Full Text] [PDF]

				V. M. Sharma, J. M. Litersky, K. Bhaskar, and G. Lee Tau impacts on growth-factor-stimulated actin remodeling J. Cell Sci., March 1, 2007; 120(5): 748 - 757. [Abstract] [Full Text] [PDF]

				K. Bhaskar, S.-H. Yen, and G. Lee Disease-related Modifications in Tau Affect the Interaction between Fyn and Tau J. Biol. Chem., October 21, 2005; 280(42): 35119 - 35125. [Abstract] [Full Text] [PDF]

				S. P. Zamora-Leon, A. Bresnick, J. M. Backer, and B. Shafit-Zagardo Fyn Phosphorylates Human MAP-2c on Tyrosine 67 J. Biol. Chem., January 21, 2005; 280(3): 1962 - 1970. [Abstract] [Full Text] [PDF]

				M. Lewitzky, M. Harkiolaki, M.-C. Domart, E. Y. Jones, and S. M. Feller Mona/Gads SH3C Binding to Hematopoietic Progenitor Kinase 1 (HPK1) Combines an Atypical SH3 Binding Motif, R/KXXK, with a Classical PXXP Motif Embedded in a Polyproline Type II (PPII) Helix J. Biol. Chem., July 2, 2004; 279(27): 28724 - 28732. [Abstract] [Full Text] [PDF]

				G. Lee, R. Thangavel, V. M. Sharma, J. M. Litersky, K. Bhaskar, S. M. Fang, L. H. Do, A. Andreadis, G. Van Hoesen, and H. Ksiezak-Reding Phosphorylation of Tau by Fyn: Implications for Alzheimer's Disease J. Neurosci., March 3, 2004; 24(9): 2304 - 2312. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/43/39950    most recent M107807200v1 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 Zamora-Leon, S. P. Articles by Shafit-Zagardo, B. Search for Related Content PubMed PubMed Citation Articles by Zamora-Leon, S. P. Articles by Shafit-Zagardo, B. Social Bookmarking                      What's this?