J Biol Chem, Vol. 274, Issue 39, 27463-27466, September 24, 1999

Synamon, a Novel Neuronal Protein Interacting with Synapse-associated Protein 90/Postsynaptic Density-95-associated Protein^*

Ikuko Yao, Yutaka Hata, Kazuyo Hirao, Maki Deguchi, Nobuyuki Ide, Masakazu Takeuchi, and Yoshimi Takai^§¶

From the  Takai Biotimer Project, Exploratory Research for Advanced Technology, Japan Science and Technology Corporation, c/o JCR Pharmaceuticals Co. Ltd., 2-2-10 Murotani, Nishi-ku, Kobe 651-2241, Japan and ^§ Department of Molecular Biology and Biochemistry, Osaka University Graduate School of Medicine, Faculty of Medicine, Suita 565-0871, Japan

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Guanylate kinase-associated protein (GKAP)/SAP90/PSD-95-associated protein (SAPAP)/DLG-associated protein (DAP) is a protein of the postsynaptic density (PSD), and binds to the guanylate kinase domain of PSD-95/synapse-associated protein (SAP) 90 and synaptic scaffolding molecule. GKAP/SAPAP/DAP recruits PSD-95/SAP90 and its interacting protein, brain-enriched guanylate kinase-interacting protein, into the Triton X-100-insoluble fraction in transfected cells, suggesting that GKAP/SAPAP/DAP may link several PSD components to the Triton X-100-insoluble structures in the PSD. We have identified here a novel neuronal GKAP/SAPAP/DAP-binding protein and named it synamon. Synamon has seven ankyrin repeats at the NH₂ terminus followed by one src homology 3 domain and one PSD-95/Dlg-A/ZO-1 domain, and several proline-rich regions at the carboxyl terminus. Synamon interacts with the COOH-terminal region of GKAP/SAPAP/DAP via the middle region containing a PSD-95/Dlg-A/ZO-1 domain. Synamon was coimmunoprecipitated with SAPAP from rat crude synaptosomes and colocalized with SAPAP in primary cultured rat hippocampal neurons. Because synamon is composed of various protein-interacting modules, it may also interact with proteins other than GKAP/SAPAP/DAP to organize the architecture of the PSD.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

The postsynaptic density (PSD)¹ is a submembranous structure at the postsynaptic membrane mainly at the excitatory synapses (for review, see Refs. 1-3). The neurotransmitter receptors are assembled and fixed at the PSD, and several molecules implicated in the synaptic plasticity are also enriched. PSD-95/synapse-associated protein (SAP) 90 is involved in the molecular organization of these components of the PSD (Refs. 4 and 5; for review, see Refs. 6-11) and essential for learning and memory (12). PSD-95/SAP90 is a member of membrane-associated guanylate kinases (for review, see Ref. 13), and binds to guanylate kinase-associated protein (GKAP)/SAP90/PSD-95-associated protein (SAPAP)/DLG-associated protein (DAP) and brain-enriched guanylate kinase-interacting protein (BEGAIN) via the guanylate kinase domain (14-17). In this paper, we have described GKAP/SAPAP/DAP as SAPAP just for simplicity. SAPAP is Triton X-100-insoluble in both the brain homogenate and the transfected cells. It links PSD-95/SAP90 to the Triton X-100-insoluble structures of transfected Chinese hamster ovary cells (17). BEGAIN is also recruited by SAPAP into the Triton X-100-insoluble fraction with PSD-95/SAP90 in Chinese hamster ovary cells (17). SAPAP interacts with another neuronal membrane-associated guanylate kinase, synaptic scaffolding molecule (S-SCAM), which also provides scaffolds for the components of synaptic junctions (18). Thereby, SAPAP is a key protein in the architecture of the PSD. The detergent-insolubility of SAPAP is conferred by its NH₂-terminal and middle regions. The middle region contains 5 repeats of 14 amino acids (aa) and is involved in the interaction with PSD-95/SAP90 (14-16). A proline-rich region that fits to the consensus motif for the binding of the src homology 3 (SH3) domain exists immediately after the middle region (19). The carboxyl-terminal region is Triton X-100-soluble and well conserved among the four isoforms of SAPAP, suggesting that it may have some function. We have searched for a SAPAP-binding protein, identified a novel neuronal protein, which binds to the carboxyl-terminal region of SAPAP, and named it synamon.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Yeast Two-hybrid Screening and cDNA Cloning-- Rat brain yeast two-hybrid library was constructed using pVP16 vector and screened (20). Rat brain cDNA libraries (Stratagene and CLONTECH) were screened with the [-³²P]dCTP-labeled random-primed probes (20).

Construction of Expression Vectors-- Various expression vectors were constructed by conventional molecular biology techniques and polymerase chain reaction method using pBTM116, pCMV Myc, pClneo Myc, and pGex4T-1 (Amersham Pharmacia Biotech). pBTM116 SAPAP2-1 contains full-length SAPAP2. The following constructs contain the following aa residues of SAPAP1: pCMV Myc SAPAP1-1, 1-992; pCMV Myc SAPAP1-2, 1-477; pCMV Myc SAPAP1-6, 478-992; pCMV Myc SAPAP1-8, 568-992; pClneo Myc SAPAP1-8, 800-992; and pClneo Myc SAPAP1-22, 800-985. The following constructs contain the following aa residues of synamon: pGex4T-1 synamon-16, 1977-2091; pGex4T-1 synamon-18, 616-1110; and pGex4T-1 synamon-19, 863-1337.

Antibodies-- Rabbit polyclonal antibodies were raised against the product of pGex4T-1 synamon-16, which is the COOH-terminal region of synamon. The anti-SAPAP antibody was described previously (15). The mouse monoclonal anti-Myc-tag antibody was obtained from American Type Culture Collection. The rhodamine-conjugated and fluorescein isothiocyanate-conjugated second antibodies for dual labeling were purchased from Chemicon. The monoclonal anti-NMDA receptor 1 antibody was a gift of Dr. Nils Brose (Max Planck Institute, Goettingen, Germany).

Preparation of COS Cell Extracts-- COS cells were cultured in Dulbecco's modified Eagle medium with 10% fetal bovine serum under 10% CO₂ at 37 °C and transfected with various Myc-tagged constructs using the DEAE-dextran method (15). COS cells of two 10-cm plates were homogenized in 0.6 ml of 20 mM Hepes/NaOH, pH 8.0, containing 6 M urea and 1% (w/v) Triton X-100, and centrifuged at 100,000 × g for 15 min. The supernatant was dialyzed against 2 liters of 20 mM Hepes/NaOH, pH 8.0, containing 100 mM NaCl overnight, then centrifuged at 100,000 × g for 15 min to remove the debris, and used as the COS cell extracts.

In Vitro Binding Experiment Using Glutathione S-transferase (GST) Fusion Proteins and COS Cell Extracts-- 0.5-ml aliquots of the extracts of COS cells expressing various Myc-tagged constructs of SAPAP1 were incubated with 800 pmol of various GST fusion proteins fixed on 20 µl of glutathione-Sepharose 4B beads. After the beads were washed with Buffer A (20 mM Hepes/NaOH, pH 8.0, containing 100 mM NaCl and 1% (w/v) Triton X-100), the proteins on the beads were detected by immunoblotting using the anti-Myc antibody.

Coimmunoprecipitation-- The urea/detergent extracts of rat crude synaptosomes were prepared as described (17). 4-ml aliquots of the extracts were incubated with 30 µl of the anti-SAPAP serum or the preimmune serum fixed on protein G-Sepharose Fast Flow beads. After the beads were washed three times with Buffer A, the proteins on the beads were detected by immunoblotting using the anti-synamon antibody.

Miscellaneous Procedures-- Subcellular fractionation of rat brain, primary cultures of rat hippocampal neurons, immunocytostaining, SDS-polyacrylamide gel electrophoresis, and protein determination were performed as described (15). Northern and Western blottings were performed using multiple tissue Northern blots (CLONTECH) and ECL reagents (Amersham Pharmacia Biotech), respectively.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

We performed the yeast two-hybrid screening of a rat brain cDNA library using a bait containing full-length SAPAP2 (pBTM116 1305-1). We obtained 6 positive independent clones from 8 × 10⁵ clones; 4 clones were PSD-95/SAP90, and the remaining two clones, pPrey 3102 and 3103, were encoded novel proteins. pPrey 3102 and pPrey 3103 interacted with not only SAPAP2 but also SAPAP1, -3, and -4 in the yeasts (data not shown). We started the cDNA library screenings to obtain the full-length sequences of these two genes to find that they are alternative splicing isoforms. We obtained the presumptive full-length coding sequence of synamon through the conventional hybridization screening and polymerase chain reaction (Fig. 1). The protein, synamon, was composed of 2,158 aa. The NH₂ terminus had 7 ankyrin repeats followed by one SH3 and one PDZ domain. pPrey 3103 contained the residues 404-902 of synamon with an insert of 9 aa (Ile-Leu-Ile-Asp-Gly-Ile-Asp-Ser-Gly) between the residues 644 and 645 (at the triangle in Fig. 1). Several independent clones contained the same insert at the same position. pPrey 3102 contained the residues 646-776 of synamon preceded by an additional 50 aa, which were unique for pPrey 3102. The constructs used in this study were prepared from the clone without the insert. Synamon had the homology to Caenorhabditis elegans C33B4.3, which had 6 ankyrin repeats and one PDZ domain without an SH3 domain. The software TopPred in the simple modular architecture research tool predicted the residues 1793-1813 as a transmembrane domain (21), but another software, SOSUI in GenomeNet WWW server, predicted a soluble protein. Synamon has several proline-rich regions in the COOH-terminal region (boxed in Fig. 1).

View larger version (78K): [in this window] [in a new window]   Fig. 1.   Sequence of synamon. Residues are shown in single-letter codes. The residues of the ankyrin repeats are double-underlined. The residues of the SH3 and PDZ domains are shown on gray and black backgrounds, respectively. Proline-rich regions are boxed. The region contained in both of pPrey 3102 and pPrey 3103 from the yeast two-hybrid screening is underlined. The triangle indicates the position of the insert in pPrey 3103.

We confirmed the binding of synamon to SAPAP in the crude synaptosomes. Synamon was coimmunoprecipitated with SAPAP from the rat crude synaptosomes (Fig. 2A). Next, we determined which region of SAPAP was involved in the interaction with synamon. Both pPrey 3102 and pPrey 3103 contained the region around the PDZ domain of synamon (underlined in Fig. 1), and this region was conceivably the SAPAP-interacting region of synamon. We prepared two GST fusion proteins of synamon, GST-synamon-18 and -19, and tested the interactions with various Myc-tagged constructs of SAPAP1 (Fig. 2B). GST-synamon-18 covered the putative SAPAP-interacting region around the PDZ domain, whereas GST-synamon-19 did not contain it. GST-Synamon-18 interacted with full-length SAPAP1 and its COOH-terminal region, but not with the NH₂-terminal region (Fig. 2C). The deletion of COOH-terminal 7 amino acids of SAPAP1 abolished the interaction with GST-synamon-18. GST-synamon-19 did not interact with any Myc-tagged constructs of SAPAP1. The GST-synamon-18 and -19 did not cover the whole sequence of synamon, and we could not exclude the possibility that some untested region of synamon binds to the NH₂-terminal and/or middle regions of SAPAP1. The current result, however, indicates that the synamon-interacting region of SAPAP1 is the COOH-terminal region and distinct from the PSD-95/SAP90-interacting middle region.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Interaction of synamon with SAPAP. Panel A, coimmunoprecipitation of synamon with SAPAP from rat crude synaptosomes. The urea/detergent extracts of rat crude synaptosomes were incubated with either the anti-SAPAP serum or the preimmune serum fixed on protein G-Sepharose beads, and the proteins attached to the beads were immunoblotted with the anti-synamon antibody. Lane 1, the original extract before the incubation; lane 2, the immunoprecipitate with the preimmune serum; and lane 3, the immunoprecipitate with the anti-SAPAP serum. Panel B, schematic description of various Myc-tagged constructs of SAPAP1. The middle region containing 5 repeats of 14 aa is shown by a box. Panel C, the interaction of synamon with the COOH-terminal region of SAPAP1. The extracts of COS cells transfected with various Myc-tagged constructs of SAPAP1 were incubated with either GST-synamon-19 or GST-synamon-18 fixed on the glutathione-Sepharose 4B beads, and the proteins attached to the beads were immunoblotted with the anti-Myc antibody. Lanes 1, 4, 7, 10, 13, and 16, the original extract before the incubation; lanes 2, 5, 8, 11, 14, and 17, with the GST-synamon-19; lanes 3, 6, 9, 12, 15, and 18, with the GST-synamon-18; lanes 1, 2, and 3, pCMV Myc SAPAP1-1; lanes 4, 5, and 6, pCMV Myc SAPAP1-2; lanes 7, 8, and 9, pCMV Myc SAPAP1-6; lanes 10, 11, and 12, pCMV Myc SAPAP1-8; lanes 13, 14, and 15, pClneo Myc SAPAP1-8; and lanes 16, 17, and 18, pClneo Myc SAPAP1-22.

Northern blot analysis revealed a 9.0-kilobase message only in brain (Fig. 3A). No message was detected in heart, spleen, lung, liver, kidney, skeletal muscles, or testis. Western blot analysis showed a signal with a molecular mass of 288 kDa only in brain (Fig. 3B). The signal of a molecular mass of 182 kDa was also detected. Because the calculated molecular weight of synamon is 225,508, the 288-kDa protein is likely to be synamon and the 182-kDa protein may be a degradation product or a short isoform. In rat hippocampal neurons, synamon was localized in the cell body and the dendrites (Fig. 4A). On the dendrites, synamon was colocalized with NMDA receptor 1 (Fig. 4A). Because SAPAP was colocalized with NMDA receptor 1 (Fig. 4B), synamon was considered to be colocalized with SAPAP in rat hippocampal neurons.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Tissue distribution of synamon. Panel A, Northern blot analysis. A blot with 2 µg of mRNA from rat tissues was hybridized with a uniformly labeled probe prepared from pPrey 3103 and exposed at 80 °C for 2 days. Lane 1, heart; lane 2, brain; lane 3, spleen; lane 4, lung; lane 5, liver; lane 6, skeletal muscle; lane 7, kidney; and lane 8, testis. Panel B, Western blot analysis. Equal aliquots of the homogenates of rat tissues (25 µg of protein each) were immunoblotted with the anti-synamon antibody. Each lane contains the same tissue as each lane in Panel A.

View larger version (76K): [in this window] [in a new window]   Fig. 4.   Subcellular localization of synamon in neurons. Panel A, synamon and NMDA receptor 1 in rat primary cultured hippocampal neurons. Rat hippocampal neurons were double-stained with the polyclonal rabbit anti-synamon and the monoclonal mouse NMDA receptor 1 antibodies. a, synamon; and b, NMDA receptor 1. Panel B, SAPAP and NMDA receptor 1 in rat primary cultured hippocampal neurons. Rat hippocampal neurons were double-stained with the polyclonal rabbit anti-SAPAP and the monoclonal mouse NMDA receptor 1 antibodies. a, SAPAP; and b, NMDA receptor 1. The bar indicates 10 µm.

In this paper, we have identified a novel neuronal protein interacting with GKAP/SAPAP/DAP, synamon, by the yeast two-hybrid screening. Synamon was coimmunoprecipitated with GKAP/SAPAP/DAP from rat crude synaptosomes. Synamon was not remarkably enriched in the PSD fraction and distributed in the soma and neurites of rat primary cultured hippocampal neurons. However, at dendrites, synamon was colocalized with NMDA receptors and GKAP/SAPAP/DAP. These findings suggest that synamon interacts with GKAP/SAPAP/DAP in vivo. Synamon has a characteristic molecular structure composed of ankyrin repeats, an SH3 domain, a PDZ domain, and proline-rich sequences. Synamon interacts with the COOH-terminal region of GKAP/SAPAP/DAP via its middle region containing a PDZ domain. Although ligands for domains other than the PDZ domain of synamon need to be clarified, synamon may interact with unidentified molecules and link them to GKAP/SAPAP/DAP to form the architecture of the PSD.

	Note Added in Proof

We submitted the sequence of synamon to the GenBank^TM in November, 1998 and corrected a sequencing error in May, 1999. After the submission of this work to the Journal of Biological Chemistry, one group reported the same protein as a protein interacting with GKAP/SAPAP/DAP, cortactin, and Homer and named it Shank (Naisbitt, S., Kim, E., Tu, J. C., Xiao, B., Sala, C., Valtschanoff, J., Weinberg, R. J., Worley, P. F., and Sheng, M. (1999) Neuron 23, 569-582 and Tu, J. C., Xiao, B., Naisbitt, S., Yuan, J. P., Petralia, R. S., Brakeman, P., and Doan, A., Aakalu, V., K., Lanahan, A. A., Sheng, M., and Worley, P. F. (1999) Neuron 23, 583-592). Another group directly submitted a gene named SPANK-1 to the GenBank^TM (AF159016) (Tobaben, S., Sudhof, T. C., and Stahl, B.). Shank 1a, synamon, and SPANK-1 are identical. The protein named ProSAP1/CortBP1 was identified as a synaptic cortactin-binding protein (Boeckers, T. M., Kruetz, M. R., Winter, C., Zuschratter, W., Smalla, K.-H., Sanmarti-Vila, L., Wex, H., Langnaese, K., Bockmann, J., Garner, C. C., and Gundelfinger, E. D. (1999) J. Neurosci. 19, 6506-6518). The authors compared the sequences of ProSAP1/CortBP1 and synamon and discussed that synamon was an isoform of ProSAP1/CortBP1. Although the authors used the sequence of synamon that we submitted originally and that had a sequencing error, their conclusion was correct, and Shank la/synamon/SPANK-1 is an isoform of ProSAP1/CortBP1.

	FOOTNOTES

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

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EMBL Data Bank with accession number(s) AF102855.

^¶ To whom correspondence should be addressed. Tel: 81-6-6879-3410; Fax: 81-6-6879-3419; E-mail: ytakai@molbio.med.osak-u.ac.jp.

	ABBREVIATIONS

The abbreviations used are: PSD, postsynaptic density; SAP90, synapse-associated protein 90; GKAP, guanylate kinase-associated protein; SAPAP, SAP90/PSD-95-associated protein; DAP, DLG-associated protein; BEGAIN, brain-enriched guanylate kinase-interacting protein; S-SCAM, synaptic scaffolding molecule; aa, amino acid; SH3, src homology 3; NMDA, N-methyl-D-aspartate; GST, glutathione S-transferase.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES