GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Genes to Cells (2004) 9, 631-640. doi:10.1111/j.1356-9597.2004.00752.x
© 2004 Blackwell Publishing or its licensors
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kimura, K.
Right arrow Articles by Nakanishi, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kimura, K.
Right arrow Articles by Nakanishi, S.

Hyperpolarization-activated, cyclic nucleotide-gated HCN2 cation channel forms a protein assembly with multiple neuronal scaffold proteins in distinct modes of protein–protein interaction

Kouji Kimura, Jun Kitano, Yoshiaki Nakajima and Shigetada Nakanishi*

Department of Biological Sciences, Faculty of Medicine, and Department of Molecular and System Biology, Graduate School of Biostudies, Kyoto University, Yoshida Konoe-cho, Sakyo-ku, Kyoto, 606-8501, Japan


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Hyperpolarization-activated cation currents, termed Ih, are non-uniformly distributed along dendritic arbors with current density increasing with increasing distance from the soma. The non-uniform distribution of Ih currents contributes to normalization of location-dependent variability in temporal integration of synaptic input, but the molecular basis for the graded HCN distribution remains to be investigated. The hyperpolarization-activated, cyclic nucleotide-gated cation channels (HCNs) underlie Ih currents and consist of four members (HCN1-HCN4) of the gene family in mammals. In this investigation, we report that HCN2 forms a protein assembly with tamalin, S-SCAM and Mint2 scaffold proteins, using several different approaches including immunoprecipitation of rat brain and heterologously expressing cell extracts and glutathione S-transferase pull-down assays. The PDZ domain of tamalin interacts with HCN2 at both the PDZ-binding motif and the internal carboxy-terminal tail of HCN2, whereas binding of the PDZ domain of S-SCAM occurs at the cyclic nucleotide-binding domain (CNBD) and the CNBD-downstream sequence of the carboxy-terminal tail of HCN2. A protein assembly between HCN2 and Mint2 is formed by the interaction of the munc18-interacting domain of Mint2 with the CNBD-downstream sequence of HCN2. The results demonstrate that HCN2 forms a protein complex with multiple neuronal scaffold proteins in distinct modes of protein–protein interaction.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 References
 
Hyperpolarization-activated cation currents, termed Ih, contribute to a wide range of physiological functions including cardiac and neuronal pacemaker activity, the setting of resting potentials and dedritic integration of synaptic transmission (Pape 1996; Santoro & Tibbs 1999; Robinson & Siegelbaum 2003). One of the most intriguing roles of Ih in neuronal function is the dendritic normalization of location-dependent variability in temporal integration of synaptic input (Magee 1999; Williams & Stuart 2000). The density of Ih currents increases with increasing distance from the soma to the distal dendrites of CA1 hippocampal and cortical neurones (Magee 1998; Williams & Stuart 2000). This non-uniform distribution of Ih currents is thought to be capable of effectively reducing variability in temporal summation, which results from spatial differences in input location (Magee 1999; Williams & Stuart 2000). The hyperpolarization-activated, cyclic nucleotide-gated cation channels (HCNs) underlie Ih currents and consist of four members (HCN1-HCN4) of the gene family in mammals (Santoro et al. 1998; Ludwig et al. 1998; Ishii et al. 1999). Although the biophysical properties of each channel and the subunit heteromerization of HCN channels were well studied (Ishii et al. 2001; Ulens & Tytgat 2001; Chen et al. 2001; Proenza et al. 2002; Tran et al. 2002), the molecular basis for localization and intracellular trafficking of HCN channels largely remains to be elucidated.

Multimolecular protein assembly through protein–protein interaction is important as a general mechanism of specialized functions of ion channels in neuronal and other cells (Bredt 1998; Sheng & Sala 2001). The protein assembly is built around several key scaffold proteins that contain multiple protein-interacting domains (Hata et al. 1998; Garner et al. 2000; Husi & Grant 2001). Such protein assembly contributes to distribution, trafficking and clustering of ion channels as well as coupling of ion channels to intracellular signalling cascades (Garner et al. 2000; Sheng & Sala 2001). Tamalin (also termed GRP1-associated scaffold protein) is a scaffold protein that comprises multiple protein-interacting domains (Nevrivy et al. 2000; Kitano et al. 2002, 2003). The 95-kDa postsynaptic density protein (PSD-95)/discs-large/ZO-1 (PDZ) domain of tamalin interacts with a consensus S/TSXL sequence (one-letter notation for amino acids with X representing any amino acids) present at the carboxy-terminal ends of metabotropic glutamate receptor group 1 (mGluR1 and mGluR5) and group 2 subtypes (mGluR2 and mGluR3) (Kitano et al. 2002). The carboxy-terminal ends of HCN1, 2 and 4 posses the S/XSNL sequence which shares a PDZ-binding motif of mGluRs for tamalin binding (Monteggia et al. 2000). Tamalin also forms a protein complex with several synaptic and protein-trafficking scaffold proteins including synaptic scaffolding molecule (S-SCAM), Mint2 (also termed X11-like and X11ß) and PSD-95 (Kitano et al. 2003). PSD-95 and S-SCAM play an important role in functional assembly of a synaptic macromolecular complex (Hata et al. 1998; Sheng & Sala 2001), while Mint2 is considered to be involved in protein targeting in polarized cells (Okamoto & Südhof 1997). Because the HCN channels possess a PDZ-binding motif for tamalin binding and a macromolecular protein assembly is built with several key scaffold proteins, we sought in this investigation to examine the possible protein assembly of HCNs with tamalin, S-SCAM, Mint2 and PSD-95. We report here that tamalin, S-SCAM and Mint2 are involved in a protein assembly with HCN2 through the distinct protein-interacting domains of HCN2.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 References
 
Interaction of tamalin with HCN2

Tamalin binds to the PDZ-binding motif consisting of the common S/TSXL sequence at the carboxy-terminal ends of group 1 and group 2 mGluRs (Kitano et al. 2002). Because the HCN1-HCN4 subunits share a closely related S/XS/ANL/M sequence at the carboxy-terminal ends (Fig. 1A), we first performed yeast two-hybrid assays to examine the interaction of tamalin with the last seven amino acid residues of 4 subunits of the HCN family. Among them, HCN2, but not the other three, showed positive tamalin-interacting signals in yeast two-hybrid assays (Fig. 1A).



View larger version (25K):
[in this window]
[in a new window]
  		Figure 1  Interaction of HCN2 with tamalin in the yeast two-hybrid system and heterologously expressing COS-7 cells. (A) Amino acid sequence alignment of the carboxy-terminal ends of group 1 and group 2 mGluRs and 4 HCN subunits; amino acids matching the consensus sequence are displayed with grey boxes; an asterisk indicates the termination of the protein sequences. Results of ß-galactosidase activity in yeast two-hybrid assays are indicated. Positive and negative interactions are shown as + and –, respectively. (B,C) COS-7 cells were transfected with FLAG-HCN2, myc-tamalin, or both (B) and FLAG-HCN2, EGFP-tamalin, or both (C). Cell lysates were immunoprecipitated with either anti-myc (lanes 5–7 in B), anti-HCN2 (lanes 5–7 in C) or control IgG (lane 8 in B and C) and immunoblotted with either anti-FLAG (B) or anti-EGFP (C); IP, immunoprecipitation. Inputs (lanes 1–4 in B and C) show 1/20 of cell lysates used for immunoprecipitation. Sizes of molecular markers are indicated on the left of each analysis.

 
The interaction of tamalin with HCN2 was examined by immunoprecipitation assay using COS-7 cells co-transfected with FLAG-tagged full-length rat HCN2 (FLAG-HCN2) and myc-tagged full-length rat tamalin (myc-tamalin) (Fig. 1B). Cell lysates were immunoprecipitated with anti-myc, followed by immunoblotting with anti-FLAG and anti-myc. Anti-myc immunoprecipitated comparable amounts of myc-tamalin regardless of co-transfection with and without FLAG-HCN2 (data not shown). FLAG-HCN2 was co-immunoprecipitated with anti-myc from co-transfected cells (Fig. 1B, lane 6) but not with non-immunized control IgG (Fig. 1B, lane 8). Immunoprecipitation of FLAG-HCN2 was not seen in cells transfected with either FLAG-HCN2 alone or myc-tamalin alone (Fig. 1B, lanes 5, 7). The interaction between tamalin and HCN2 was further confirmed by reverse immunoprecipitation analysis in which co-immunoprecipitation of tamalin was examined by immunoprecipitation with anti-HCN2. In this experiment, we initially examined cell lysates prepared from co-transfection with FLAG-HCN2 and myc-tamalin. However, anti-HCN2 antibody used for immunoprecipitation migrated at position of myc-tamalin on SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and cross-reacted with the horseradish peroxidase (HRP)-conjugated secondary antibody. We therefore fused a larger size of the tag, enhanced green fluorescent protein (EGFP), to tamalin (EGFP-tamalin) and co-transfected EGFP-tamalin with FLAG-HCN2 into COS-7 cells (Fig. 1C). Cell lysates were immunoprecipitated with anti-HCN2, followed by immunoblotting with anti-EGFP (Fig. 1C). Comparable amounts of immunoprecipitation of FLAG-HCN2 were confirmed with or without co-transfection of EGFP-tamalin when FLAG-HCN2 immunoprecipitates were immunoblotted with anti-FLAG (data not shown). EGFP-tamalin was co-immunoprecipitated with anti-HCN2 (Fig. 1C, lane 6), but this co-immunoprecipitation was not seen in cells transfected with either FLAG-HCN2 alone or EGFP-tamalin alone (Fig. 1C, lanes 5, 7), nor from co-transfected cell lysates precipitated with non-immunized control IgG (Fig. 1C, lane 8). The results indicated that tamalin and HCN2 form a protein complex in heterologously expressing mammalian cells.

Protein assembly of HCN2 with neuronal scaffold proteins in vivo and in heterologously expressing cells

Tamalin forms a protein assembly with not only mGluRs but also several neuronal scaffold proteins including Mint2, PSD-95 and S-SCAM (Kitano et al. 2003). Because ion channels and neurotransmitter receptors are often located in close proximity through interaction with scaffold proteins, we addressed whether HCN2 comprises part of a macromolecular protein assembly in vivo through interaction with neuronal scaffold proteins. Solubilized rat membrane fractions were immunoprecipitated with anti-HCN2 and immunoblotted with antibodies against Mint2, PSD-95 and S-SCAM (Fig. 2). Mint2 and S-SCAM were co-immunoprecipitated with anti-HCN2 (Fig. 2, upper and middle panels), but no such co-immunoprecipitation of PSD-95 was seen in membrane fractions (Fig. 2, lower panel).



View larger version (29K):
[in this window]
[in a new window]
  		Figure 2  Association of HCN2 with Mint2 and S-SCAM in the rat brain. Solubilized rat brain P2 membrane fractions were immunoprecipitated with either anti-HCN2 or control IgG, followed by immunoblotting with anti-Mint2 (upper panel), anti-S-SCAM (middle panel) or anti-PSD-95 (lower panel). Inputs show an equivalent amount (PSD-95) and 1/20 (Mint2 and S-SCAM) of extracts used for immunoprecipitation.

 
The interaction of HCN2 with Mint2 was examined in heterologous COS-7 cells (Fig. 3). Myc-tagged rat Mint2 (myc-Mint2) was co-transfected with either FLAG-HCN2 or HCN2 in COS-7 cells. Cell lysates were immunoprecipitated with either anti-myc or anti-HCN2 and then immunoblotted with anti-FLAG and anti-Mint2, respectively (Fig. 3A,B). Anti-myc and anti-HCN2 mutually co-immunoprecipitated FLAG-HCN2 and myc-Mint2 in lysates of co-transfected COS-7 cells, respectively (Fig. 3A,B, lane 6). This co-immunoprecipitation was not seen in cells transfected with HCN2, FLAG-HCN2 or myc-Mint2 alone, nor from co-transfected cell lysates precipitated with non-immunized control IgG (Fig. 3A,B, lanes 5, 7, 8). In experiments of Fig. 3B, higher amounts of HCN2 were immunoprecipitated with anti-HCN2 in cells co-transfected with myc-Mint2 than those transfected with HCN2 alone (data not shown), most likely reflecting the stabilization of HCN2 by coexpression with Mint2 (see Fig. 6). In contrast, comparable amounts of myc-Mint2 were immunoprecipitated with anti-myc regardless of co-transfection with or without FLAG-HCN2 in experiments of Fig. 3A (data not shown). These results indicated that HCN2 interacts with Mint2 in both the rat brain and heterologously expressing COS-7 cells.



View larger version (35K):
[in this window]
[in a new window]
  		Figure 3  Interaction of HCN2 with Mint2 and S-SCAM in transfected COS-7 cells. COS-7 cells were transfected with myc-Mint2, FLAG-HCN2, or both (A), HCN2, myc-Mint2, or both (B), and FLAG-HCN2, myc-S-SCAM, or both (C). Cell lysates were immunoprecipitated with either anti-myc (lanes 5–7 in A), anti-HCN2 (lanes 5–7 in B and C), or control IgG (lane 8 in A–C), followed by immunoblotting with either anti-FLAG (A), anti-Mint2 (B) or anti-myc (C). Inputs (lanes 1–4) show 1/20 of cell lysates used for immunoprecipitation.

 


View larger version (41K):
[in this window]
[in a new window]
  		Figure 6  Mint2-mediated increase of HCN2 levels in COS-7 cells. COS-7 cells were transfected with a fixed amount of cDNA for (A) FLAG-HCN2 or (B) FLAG-HCN2 delC in combination with indicated amounts of myc-Mint2 cDNA; a total amount of transfected DNAs was adjusted with a vector DNA. Cell lysates were subjected to SDS-PAGE, transferred to a nitrocellulose membrane and stained with Ponceau-S. The membrane was then destained and immunoblotted with either anti-Mint2 or anti-FLAG.

 
Next, the interaction of HCN2 with S-SCAM was examined in heterologous COS-7 cells. COS-7 cells were co-transfected with myc-tagged rat S-SCAM (myc-S-SCAM) and FLAG-HCN2, and cell lysates were immunoprecipitated with anti-HCN2, followed by immunoblotting with anti-myc (Fig. 3C). Anti-HCN2 co-immunoprecipitated myc-S-SCAM (Fig. 3C, lane 6). This co-immunoprecipitation was not seen with control IgG (Fig. 3C, lane 8), nor in cells transfected with either FLAG-HCN2 or myc-S-SCAM alone (Fig. 3C, lanes 5, 7). In this experiment, comparable amounts of FLAG-HCN2 were immunoprecipitated with anti-HCN2 regardless of co-transfection with or without S-SCAM (data not shown). The results indicated that HCN2 interacts with Mint2 and S-SCAM in both the rat brain and heterologously expressing COS-7 cells.

Identification of HCN2-binding domains of tamalin, S-SCAM and Mint2

We attempted to identify HCN2-binding domains of tamalin, Mint2 and S-SCAM with glutathione S-transferase (GST) pull-down assays (Fig. 4). Tamalin was separated into two fragments covering the PDZ-containing and PDZ-lacking domains and each fused to GST (GST-N-tam and GST-C-tam, respectively) (Fig. 4A). The resultant GST fusion proteins were expressed in Escherichia coli, purified, and immobilized on glutathione-Sepharose 4B beads. FLAG-HCN2 was expressed in COS-7 cells, and supernatants of cell lysates were incubated with glutathione-Sepharose 4B beads attached with GST or the GST fusion proteins. Bound proteins were eluted and immunoblotted with anti-FLAG. This analysis showed that FLAG-HCN2 was retained on the beads by interacting with the PDZ-containing N-tam, but not so with the PDZ-lacking C-tam, nor with GST itself (Fig. 4B).



View larger version (23K):
[in this window]
[in a new window]
  		Figure 4  Identification of HCN2-interacting domains of tamalin, S-SCAM and Mint2 by GST pull-down assays. (A) Schematic structures of tamalin and truncated N-tam and C-tam mutants are shown. (B) GST, GST-N-tam and GST-C-tam were immobilized on glutathione-Sepharose 4B beads and incubated with recombinant FLAG-HCN2 expressed in COS-7 cells. Bound proteins were eluted and detected by immunoblotting with anti-FLAG. (C) Binding of GST, GST-Mint2 PDZ1 +2 and GST-S-SCAM PDZ5 to FLAG-HCN2 was analysed by GST pull-down assays as in (B). (D) The MID, PTB and PDZ1 +2 domains of Mint2 fused to GST are indicated under a schematic structure of rat Mint2. (E) GST, GST-MID, GST-PTB and GST-PDZ1 +2 were immobilized on glutathione-Sepharose 4B beads and incubated with recombinant EGFP-HCN2 expressed in COS-7 cells. Bound proteins were detected by immunoblotting with anti-EGFP. Inputs show 1/200, 1/80 and 1/40 of proteins used for pull-down assays in (B) (C) and (E), respectively.

 
Mint2 and S-SCAM possess two and six PDZ domains (Okamoto & Südhof 1997; Hirao et al. 1998, 2000). The 6th PDZ domain (PDZ5) of S-SCAM binds to a PDZ-binding motif of tamalin (Kitano et al. 2003), which is closely related to a PDZ-binding motif of HCN2. PDZ1 and PDZ2 of Mint2 and PDZ5 of S-SCAM were fused to GST (GST-Mint2 PDZ1 +2 and GST-S-SCAM PDZ5, respectively), expressed in E. coli and purified. The resultant GST fusion proteins were subjected to GST pull-down assays using FLAG-HCN2 expressed in COS-7 cells. GST-S-SCAM PDZ5 bound to FLAG-HCN2, but GST-Mint2 PDZ1 +2 had no ability to bind to FLAG-HCN2 (Fig. 4C). The result indicated that both tamalin and S-SCAM directly interact with HCN2 via their PDZ domains.

Mint2 was dissected into three segments, the amino-terminal munc18-interacting domain (MID), the middle phosphotyrosine-binding domain (PTB) and the carboxy-terminal PDZ1 +2 domains (Fig. 4D). Each segment was fused to GST, expressed in E. coli and purified. Because GST-MID migrated at position of FLAG-HCN2 in SDS-PAGE and cross-reacted with the HRP-conjugated secondary antibody, EGFP was attached to HCN2 and the resultant EGFP-HCN2 was expressed in COS-7 cells. The pull-down assay was then conducted with the GST fusion proteins and EGFP-HCN2. This analysis showed that Mint2 directly interacts with HCN2 through its MID domain (Fig. 4E).

Identification of binding domains of HCN2 for scaffold proteins

We next examined the binding domains of HCN2 for tamalin, S-SCAM and Mint2 by pull-down assays. Three deletional mutants, which lacked the last 10 amino acids, the carboxy-terminal sequence just distal from the cyclic nucleotide-binding domain (CNBD) and the sequence up to the CNBD of HCN2, were fused to FLAG and expressed in COS-7 cells: these deletional mutants were termed del10, delC and delCNBD, respectively (Fig. 5A). The wild-type and deletional mutants of HCN2 were examined for their ability to bind to GST-N-tam and GST-S-SCAM PDZ5 (Fig. 5B,C). For pull-down assays of Mint2, the fusion proteins of maltose-binding protein (MAL) were used instead of the GST fusion protein to avoid comigration of GST-MID and FLAG-HCN2. MAL was fused to the MID of Mint2 (MAL-MID), expressed in E. coli and purified. The MAL-MID was immobilized on amylose resin and incubated with FLAG-HCN2 and FLAG-HCN2 delC (Fig. 5D).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 5  Identification of scaffold protein-binding domains of HCN2 by pull-down assays. (A) Schematic structures of rat HCN2 and deletional HCN2 mutants used for pull-down assays are indicated; S1-S6, 6 transmembrane segments of HCN2. (B) GST and GST-N-tam were immobilized on glutathione-Sepharose 4B beads and incubated with recombinant FLAG-HCN2, FLAG-HCN2 del10 or FLAG-HCN2 delCNBD expressed in COS-7 cells. Bound proteins were detected by immunoblotting with anti-FLAG. (C) Binding of FLAG-HCN2 and deletional FLAG-HCN2 mutants to GST-S-SCAM PDZ5 was tested by GST pull-down assays as in (B). (D) MAL and MAL-MID were immobilized on amylose resin and incubated with recombinant FLAG-HCN2 or FLAG-HCN2 delC expressed in COS-7 cells. The ability of HCN2 and HCN2 delC to bind to MAL-MID was tested by immunoblotting with anti-FLAG. Inputs show 1/40, 1/200 and 1/200 of proteins used for pull-down assays in (B–D), respectively.

 
The binding of FLAG-HCN2 to N-tam was markedly reduced by deletion of the last 10 amino acids of HCN2 and completely lost by deletion of the carboxy-terminal sequence distal from the CNBD of HCN2 (Fig. 5B). The binding of tamalin to HCN2 thus requires not only the PDZ-binding motif but also the internal carboxy-terminal sequence of tamalin. The analysis of binding between MAL-MID and FLAG-HCN2 showed that binding was abolished in FLAG-HCN2 delC (Fig. 5D), indicating that the sequence downstream of the CNBD of HCN2 is involved in interaction of Mint2 with HCN2. Interestingly, binding of GST-S-SCAM PDZ5 to FLAG-HCN2 remained unchanged by deletion of the last 10 amino acid residues of HCN2, and this binding was significantly reduced in FLAG-HCN2 delC and completely lost in FLAG-HCN2 delCNBD (Fig. 5C). This result indicated that although S-SCAM binds to HCN2 via its PDZ domain, the internal sequence in the carboxy-terminal tail of HCN2 is involved in interaction between HCN2 and S-SCAM. Collectively, these results indicated that HCN2 directly interacts with tamalin, Mint2 and S-SCAM through the distinct protein-binding domains at the carboxy-terminal tail of HCN2.

Increase of HCN2 by Mint2 coexpression

Coexpression of the Mint2 protein in mammalian cells has been shown to stabilize the Mint2-interacting amyloid precursor protein (APP) and increase cellular contents of APP (Borg et al. 1998). Although the HCN2-binding domain of Mint2 is different from the APP-binding domain of Mint2 (Tomita et al. 1999), we addressed whether coexpression of Mint2 is capable of increasing cellular contents of HCN2 in COS-7 cells (Fig. 6). A fixed amount of FLAG-HCN2 was transfected into COS-7 cells in combination with varying amounts of myc-Mint2. Cells were lysed with SDS-PAGE sample buffer, and cell lysates were immunoblotted with anti-Mint2 and anti-FLAG (Fig. 6). No difference in an overall profile of endogenous proteins, as analysed with Ponceau-S staining, was seen by increasing amounts of transfected Mint2 (Fig. 6A). Importantly, protein levels of FLAG-HCN2 increased by increasing expression levels of Mint2 (Fig. 6A). Furthermore, when FLAG-HCN2 delC was coexpressed with varying amounts of myc-Mint2, cellular levels of HCN2 delC remained unchanged despite an increase of myc-Mint2 (Fig. 6B). These results indicated that the interaction between HCN2 and Mint2 is important for Mint2-mediated increase in cellular contents of the HCN2 protein.


   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 References
 
The present investigation has indicated that HCN2 interacts with multiple neuronal scaffold proteins through its distinct protein-binding domains. The protein assembly built around HCN2 with Mint2 and S-SCAM was verified by co-immunoprecipitation analysis of rat brain membrane fractions and heterologously expressing cell lysates. The direct evidence indicating a protein assembly between tamalin and HCN2 in vivo remained to be elucidated, due to no availability of a high-affinity tamalin antibody to be used for immunoprecipitation of brain membrane fractions. However, the yeast two-hybrid and heterologous expression analysis as well as in vitro pull-down assays indicated that tamalin and HCN2 form a protein complex through the direct protein–protein interaction between these two proteins. Tamalin binds to HCN2 via its PDZ domain and this binding occurs at not only the PDZ-binding motif but also the internal carboxy-terminal tail of HCN2. This mode of binding is similar to that of the interaction between tamalin and mGluR1, in which both the PDZ-binding motif and the internal carboxy-terminal tail of mGluR1 are required for a maximal binding of the two proteins (Kitano et al. 2002). The PDZ domain of S-SCAM also binds to HCN2 but this binding is directed by interaction with the internal CNBD and CNBD-downstream sequences of the carboxy-terminal tail of HCN2. The MID domain of Mint2 interacts with the CNBD-downstream region in the carboxy-terminal tail of HCN2. HCN2 thus forms a protein complex with multiple scaffold proteins in distinct modes of protein–protein interaction.

Neuronal scaffold proteins are involved not only in organizing the complex protein lattice in the dendrites and axons of neuronal cells but also in enabling channel and receptor proteins to be transported from intracellular compartments to plasma membrane (Bredt 1998; Sheng & Sala 2001). Tamalin serves as a key element that contributes to multimolecular protein assembly in neuronal cells (Kitano et al. 2002, 2003). It forms a protein complex with several postsynaptic receptors and scaffold proteins including group 1 mGluRs, PSD-95, S-SCAM and SAP90/PSD-95-associated proteins (SAPAPs) (Kitano et al. 2002, 2003). It also associates with protein-trafficking scaffold proteins such as cytohesins, Mint2 and CASK (Kitano et al. 2002, 2003). Tamalin has been shown to promote intracellular trafficking and cell surface expression of group 1 mGluRs in COS-7 cells and cultured hippocampal neurones (Kitano et al. 2002). S-SCAM is widely distributed to both dendrites and axons of cultured neurones and is enriched in the membrane fraction of the brain (Hirao et al. 1998; Mok et al. 2002). S-SCAM interacts with membrane proteins including the Kv1.4 potassium channel, the NMDA receptor NR2 subunits, neuroligin (Hirao et al. 1998) and ß1-adrenergic receptor (ß1AR) (Xu et al. 2001). It also associates with intracellular signalling molecules including a GDP/GTP exchange factor for Rap 1 (Ohtsuka et al. 1999), protein tyrosine phosphatase (Wu et al. 2000) and ß-catenin (Kawajiri et al. 2000; Nishimura et al. 2002). S-SCAM has been shown to enhance agonist-induced internalization of ß1AR and promote association of ß1AR with ß-catenin (Xu et al. 2001). Mints are enriched at the Golgi apparatus but also distributed throughout axons and dendrites (Tomita et al. 1999; Nakajima et al. 2001). Mints bind to munc-18, a protein essential for synaptic vesicle exocytosis, and CASK involved in targeting and localization of synaptic membrane proteins (Butz et al. 1998). Mint1 knockout mice showed impairments of {gamma}-aminobutyric acid release in hippocampal interneurones and methamphetamine-induced dopamine release in striatal neurones (Mori et al. 2002; Ho et al. 2003). In C. elegans, the ortholog of the Mint family protein, LIN-10, is required for postsynaptic localization of the glutamate receptor GLR1 in nematode neurones (Rongo et al. 1998). In addition, Mints interact with APP and presenilins and increase the cellular levels of APP in transfected cells (Sastre et al. 1998; Lau et al. 2000). All three scaffold proteins identified as HCN2-interacting proteins thus possess multiple functions involved in synaptic organization and protein trafficking in neuronal cells.

One of the important characteristics of Ih is its regulation by cyclic nucleotide binding (Robinson & Siegelbaum 2003). A number of neurotransmitters have been shown to regulate Ih in different neurones through either enhancing or diminishing cAMP levels (Pape 1996). This regulation of Ih gating is mediated by direct binding of cAMP to the CNBD of HCNs. cGMP also binds to the CNBD and enhances channel activities, although the affinity for cGMP is much lower than that for cAMP (Robinson & Siegelbaum 2003). In addition, It has been reported that the activity of Ih is regulated by protein phosphorylation and dephosphorylation (Robinson & Siegelbaum 2003). The HCN2-interacting scaffold proteins can bind to the CNBD or its adjacent region of HCN2 and recruit intracellular signalling molecules. The protein assembly of the scaffold proteins around the CNBD region of HCN2 may thus be able to regulate the interaction of cyclic nucleotides or phosphorylation/dephosphorylation of HCN2. The characteristic feature of Ih currents is a graded distribution of these currents along long dendritic arbors of neuronal cells, thus contributing to normalization of distance-dependent variability in temporal integration of synaptic inputs. The identification of multiple HCN2-interacting scaffold proteins reported here will help to study mechanisms underlying localization, trafficking and signal transduction of HCN channels in neuronal cells.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
References
 
Cloning of rat HCN2 cDNA and DNA constructs

The HCN2 cDNA was cloned from a rat forebrain library as previously described (Moriyoshi et al. 1991). The 256-bp cDNA fragment of mouse HCN2 (GenBank AJ225122 [GenBank] , nucleotide residues, 2372–2627) was used as a hybridization probe. The 2633-bp HCN2 cDNA sequence covering the entire protein-coding region and its surrounding regions was determined on both strands (GenBank AB164197 [GenBank] ). The mammalian expression vectors containing rat HCN2, FLAG-HCN2 and EGFP-HCN2 were constructed by inserting the rat HCN2 cDNA into the mammalian expression vectors of pCI (Promega), pCMV-Tag2B (Stratagene) and pEGFPC2 (BD Biosciences Clontech). Three deletional mutants of HCN2 (HCN2 del10, HCN2 delC and HCN2 delCNBD) were constructed by PCR techniques and inserted into pCMV-Tag2B. Myc-Mint2, myc-tamalin and myc-S-SCAM were prepared as previously described (Kitano et al. 2003). EGFP-tamalin was constructed by isolating an appropriate restriction fragment of myc-tamalin and inserting it into pEGFPC2. The GST fusion proteins containing the MID domain, the PTB domain and two PDZ domains of Mint2 (GST-MID, GST-PTB and GST-PDZ1 +2, respectively), the GST fusion proteins covering the PDZ-containing domain (GST-N-tam) and the PDZ-lacking domain of tamalin (GST-C-tam) and the GST fusion protein containing the 6th PDZ domain of S-SCAM (GST-S-SCAM PDZ5) were prepared as previously described (Kitano et al. 2003). MAL-MID was constructed by inserting an appropriate fragment of GST-MID into pMAL-c2X (New England BioLabs). Proper in-frame insertions and the absence of any sequence errors of all PCR products were confirmed by DNA sequencing.

Yeast two-hybrid assay

The oligonucleotides that encoded the last 7 amino acid residues and a stop codon of rat HCN1-4 were cloned into a bait plasmid, pAS2-1 (BD Biosciences Clontech). The pACT2 prey plasmid containing the full-length rat tamalin was co-transfected with the above bait plasmids into yeast Y190 cells (BD Biosciences Clontech) and ß-galactosidase reporter gene assays were performed as previously described (Kitano et al. 2002).

Immunoprecipitation

Adult rat whole brains were homogenized and fractionated into cytosols and P1 and P2 membrane fractions as previously described (Kitano et al. 2002). The P2 membrane fraction was solubilized in 50 mM Tris HCl, pH 7.4 containing 1% sodium deoxycholate at 36 °C for 30 min and dialysed against 50 mM Tris HCl, pH 7.4 and 0.1% Triton X-100 overnight. For immunoprecipitation analysis of the heterologously expressing system, the expression vectors were transfected into COS-7 cells on a six-well plate by LipofectAMINE (Invitrogen). Cells were lysed 40 h after transfection with phosphate-buffered saline (PBS) (137 mM NaCl, 2.68 mM KCl, 8.1 mM Na2HPO4 and 1.47 mM KH2PO4) containing 1% Triton X-100 and protease inhibitor mixture Complete (Roche Diagnostics). Supernatants were prepared by centrifugation of cell lysates at 10 000 g at 4 °C for 10 min. The solubilized P2 membrane fractions and the supernatants of COS-7 cell lysates were incubated with antibody (1 µg) and attached to protein A Sepharose. Immunoprecipitates were washed three times and dissolved into 2 x SDS-PAGE loading buffer, followed by immunoblotting with the primary antibody. Immunoblots were reacted with the HRP-conjugated secondary antibody against mouse or rabbit IgG and detected with HRP reaction (Kitano et al. 2002). The primary antibodies used for immunoprecipitaion and immunoblotting were obtained from the following sources: anti-HCN2 (Alomone), anti-myc (Santa Cruz, Cell Signal), anti-FLAG (Sigma) and anti-EGFP (Molecular Probes). Other antibodies and reagents were obtained as previously described (Kitano et al. 2003).

Pull-down assay

GST and MAL fusion proteins were expressed in E. coli and purified by glutathione-Sepharose 4B beads (Amersham Biosciences) and amylose resin (New England Biolabs), respectively. The purified proteins were dialysed against PBS at 4 °C. GST and MAL fusion proteins were immobilized on glutathione-Sepharose 4B beads and amylose resin, respectively, and incubated with supernatants containing recombinant FLAG-HCN2, EGFP-HCN2 or FLAG-HCN2 mutants, all of which were transiently expressed in COS-7 cells. The beads and resin were washed 3 times in PBS containing 1% Triton X-100 and eluted with SDS-PAGE loading buffer, followed by immunoblotting.


   Acknowledgements
 
We thank Dr Yutaka Hata and Dr Yoshimi Takai for providing anti-S-SCAM antibody and an expression vector for S-SCAM and Dr Takahiro Ishii for valuable advice. This work was supported in part by research grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan. K.K. was a fellow of the Japan Society for the Promotion of Science.


   Footnotes
 
Communicated by: Shuh Narumiya

* Correspondence: E-mail: snakanis{at}phy.med.kyoto-u.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Borg, J.-P., Yang, Y., De Taddéo-Borg, M., Margolis, B. & Turner, R.S. (1998) The X11{alpha} protein slows cellular amyloid precursor protein processing and reduces Aß40 and Aß42 secretion. J. Biol. Chem. 273, 14761–14766.[Abstract/Free Full Text]

Bredt, D.S. (1998) Sorting out genes that regulate epithelial and neuronal polarity. Cell 94, 691–694.[CrossRef][Medline]

Butz, S., Okamoto, M. & Südhof, T.C. (1998) A tripartite protein complex with the potential to couple synaptic vesicle exocytosis to cell adhesion in brain. Cell 94, 773–782.[CrossRef][Medline]

Chen, S., Wang, J. & Siegelbaum, S.A. (2001) Properties of hyperpolarization-activated pacemaker current defined by coassembly of HCN1 and HCN2 subunits and basal modulation by cyclic nucleotide. J. General Physiol. 117, 491–503.[Abstract/Free Full Text]

Garner, C.C., Nash, J. & Huganir, R.L. (2000) PDZ domains in synapse assembly and signalling. Trends Cell Biol. 10, 274–280.[CrossRef][Medline]

Hata, Y., Nakanishi, H. & Takai, Y. (1998) Synaptic PDZ domain-containing proteins. Neurosci. Res. 32, 1–7.[CrossRef][Medline]

Hirao, K., Hata, Y., Ide, N., et al. (1998) A novel multiple PDZ domain-containing molecule interacting with N-methyl-D-aspartate receptors and neuronal cell adhesion proteins. J. Biol. Chem. 273, 21105–21110.[Abstract/Free Full Text]

Hirao, K., Hata, Y., Yao, I., et al. (2000) Three isoforms of synaptic scaffolding molecule and their characterization. Multimerization between the isoforms and their interaction with N-methyl-D-aspartate receptors and SAP90/PSD-95-associated protein. J. Biol. Chem. 275, 2966–2972.[Abstract/Free Full Text]

Ho, A., Morishita, W., Hammer, R.E., Malenka, R.C. & Südhof, T.C. (2003) A role for Mints in transmitter release: Mint 1 knockout mice exhibit impaired GABAergic synaptic transmission. Proc. Natl. Acad. Sci. USA 100, 1409–1414.[Abstract/Free Full Text]

Husi, H. & Grant, S.G.N. (2001) Proteomics of the nervous system. Trends Neurosci. 24, 259–266.[CrossRef][Medline]

Ishii, T.M., Takano, M. & Ohmori, H. (2001) Determinants of activation kinetics in mammalian hyperpolarization-activated cation channels. J. Physiol. 537, 93–100.

Ishii, T.M., Takano, M., Xie, L.-H., Noma, A. & Ohmori, H. (1999) Molecular characterization of the hyperpolarization-activated cation channel in rabbit heart sinoatrial node. J. Biol. Chem. 274, 12835–12839.[Abstract/Free Full Text]

Kawajiri, A., Itoh, N., Fukata, M., et al. (2000) Identification of a novel ß-catenin-interacting protein. Biochem. Biophys. Res. Commun. 273, 712–717.[CrossRef][Medline]

Kitano, J., Kimura, K., Yamazaki, Y., et al. (2002) Tamalin, a PDZ domain-containing protein, links a protein complex formation of group 1 metabotropic glutamate receptors and the guanine nucleotide exchange factor cytohesins. J. Neurosci. 22, 1280–1289.[Abstract/Free Full Text]

Kitano, J., Yamazaki, Y., Kimura, K., Masukado, T., Nakajima, Y. & Nakanishi, S. (2003) Tamalin is a scaffold protein that interacts with multiple neuronal proteins in distinct modes of protein–protein association. J. Biol. Chem. 278, 14762–14768.[Abstract/Free Full Text]

Lau, K.-F., McLoughlin, D.M., Standen, C. & Miller, C.C.J. (2000) X11{alpha} and X11ß interact with presenilin-1 via their PDZ domains. Mol. Cell. Neurosci. 16, 557–565.[CrossRef][Medline]

Ludwig, A., Zong, X., Jeglitsch, M., Hofmann, F. & Biel, M. (1998) A family of hyperpolarization-activated mammalian cation channels. Nature 393, 587–591.[CrossRef][Medline]

Magee, J.C. (1998) Dendritic hyperpolarization-activated currents modify the integrative properties of hippocampal CA1 pyramidal neurons. J. Neurosci. 18, 7613–7624.[Abstract/Free Full Text]

Magee, J.C. (1999) Dendritic Ih normalizes temporal summation in hippocampal CA1 neurons. Nature Neurosci. 2, 508–514.[CrossRef][Medline]

Mok, H., Shin, H., Kim, S., Lee, J.-R., Yoon, J. & Kim, E. (2002) Association of the kinesin superfamily motor protein KIF1B{alpha} with postsynaptic density-95 (PSD-95), synapse-associated protein-97, and synaptic scaffolding molecule PSD-95/Discs Large/Zona Occludens-1 proteins. J. Neurosci. 22, 5253–5258.[Abstract/Free Full Text]

Monteggia, L.M., Eisch, A.J., Tang, M.D., Kaczmarek, L.K. & Nestler, E.J. (2000) Cloning and localization of the hyperpolarization-activated cyclic nucleotide-gated channel family in rat brain. Brain Res. Mol. Brain Res. 81, 129–139.[Medline]

Mori, A., Okuyama, K., Horie, M., et al. (2002) Alteration of methamphetamine-induced striatal dopamine release in mint-1 knockout mice. Neurosci. Res. 43, 251–257.[CrossRef][Medline]

Moriyoshi, K., Masu, M., Ishii, T., Shigemoto, R., Mizuno, N. & Nakanishi, S. (1991) Molecular cloning and characterization of the rat NMDA receptor. Nature 354, 31–37.[CrossRef][Medline]

Nakajima, Y., Okamoto, M., Nishimura, H., et al. (2001) Neuronal expression of mint1 and mint2, novel multimodular proteins, in adult murine brain. Brain Res. Mol. Brain Res. 92, 27–42.[Medline]

Nevrivy, D.J., Peterson, V.J., Avram, D., et al. (2000) Interaction of GRASP, a protein encoded by a novel retinoic acid-induced gene, with members of the cytohesin family of guanine nucleotide exchange factors. J. Biol. Chem. 275, 16827–16836.[Abstract/Free Full Text]

Nishimura, W., Yao, I., Iida, J., Tanaka, N. & Hata, Y. (2002) Interaction of synaptic scaffolding molecule and ß-catenin. J. Neurosci. 22, 757–765.[Abstract/Free Full Text]

Ohtsuka, T., Hata, Y., Ide, N., et al. (1999) nRap GEP: a novel neural GDP/GTP expange protein for Rap1 small G protein that interacts with synaptic scaffolding molecule (S-SCAM). Biochem. Biophys. Res. Commun. 265, 38–44.[CrossRef][Medline]

Okamoto, M. & Südhof, T.C. (1997) Mints, Munc18-interacting proteins in synaptic vesicle exocytosis. J. Biol. Chem. 272, 31459–31464.[Abstract/Free Full Text]

Pape, H.-C. (1996) Queer current and pacemaker: The hyperpolarization-activated cation current in neurons. Annu. Rev. Physiol. 58, 299–327.[CrossRef][Medline]

Proenza, C., Tran, N., Angoli, D., Zahynacz, K., Balcar, P. & Accili, E.A. (2002) Different roles for the cyclic nucleotide binding domain and amino terminus in assembly and expression of hyperpolarization-activated, cyclic nucleotide-gated channels. J. Biol. Chem. 277, 29634–29642.[Abstract/Free Full Text]

Robinson, R.B. & Siegelbaum, S.A. (2003) Hyperpolarization-activated cation currents: From molecules to physiological function. Annu. Rev. Physiol. 65, 453–480.[CrossRef][Medline]

Rongo, C., Whitfield, C.W., Rodal, A., Kim, S.K. & Kaplan, J.M. (1998) LIN-10 is a shared component of the polarized protein localization pathways in neurons and epithelia. Cell 94, 751–759.[CrossRef][Medline]

Santoro, B., Liu, D.T., Yao, H., et al. (1998) Identification of a gene encoding a hyperpolarization-activated pacemaker channel of brain. Cell 93, 717–729.[CrossRef][Medline]

Santoro, B. & Tibbs, G.R. (1999) The HCN gene family: Molecular basis of the hyperpolarization-activated pacemaker channels. Ann. NY Acad. Sci. 868, 741–764.[CrossRef][Medline]

Sastre, M., Turner, R.S. & Levey, E. (1998) X11 interaction with ß-amyloid precursor protein modulates its cellular stabilization and reduces amyloid ß-protein secretion. J. Biol. Chem. 273, 22351–22357.[Abstract/Free Full Text]

Sheng, M. & Sala, C. (2001) PDZ domains and the organization of supramolecular complexes. Annu. Rev. Neurosci. 24, 1–29.[CrossRef][Medline]

Tomita, S., Ozaki, T., Taru, H., et al. (1999) Interaction of a neuron-specific protein containing PDZ domains with Alzheimer's amyloid precursor protein. J. Biol. Chem. 274, 2243–2254.[Abstract/Free Full Text]

Tran, N., Proenza, C., Macri, V., et al. (2002) A conserved domain in the NH2 terminus important for assembly and functional expression of pacemaker channels. J. Biol. Chem. 277, 43588–43592.[Abstract/Free Full Text]

Ulens, C. & Tytgat, J. (2001) Functional heteromerization of HCN1 and HCN2 pacemaker channels. J. Biol. Chem. 276, 6069–6072.[Abstract/Free Full Text]

Williams, S.R. & Stuart, G.J. (2000) Site independence of EPSP time course is mediated by dendritic Ih in neocortical pyramidal neurons. J. Neurophysiol. 83, 3177–3182.[Abstract/Free Full Text]

Wu, X., Hepner, K., Castelino-Prabhu, S., et al. (2000) Evidence for regulation of the PTEN tumor suppressor by a membrane-localized multi-PDZ domain containing scaffold protein MAGI-2. Proc. Natl. Acad. Sci. USA 97, 4233–4238.[Abstract/Free Full Text]

Xu, J., Paquet, M., Lau, A.G., Wood, J.D., Roos, C.A. & Hall, R.A. (2001) ß1–adrenergic receptor association with the synaptic scaffolding protein membrane-associated guanylate kinase inverted-2 (MAGI-2). Differential regulation of receptor internalization by MAGI-2 and PSD-95. J. Biol. Chem. 276, 41310–41317.[Abstract/Free Full Text]

Received: 24 February 2004
Accepted: 13 April 2004




This article has been cited by other articles:


Home page
 Physiol. Rev.Home page
M. Biel, C. Wahl-Schott, S. Michalakis, and X. Zong
Hyperpolarization-Activated Cation Channels: From Genes to Function
Physiol Rev, July 1, 2009; 89(3): 847 - 885.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Cell Physiol.Home page
L. D. Ridgway, E. Y. Kim, and S. E. Dryer
MAGI-1 interacts with Slo1 channel proteins and suppresses Slo1 expression on the cell surface
Am J Physiol Cell Physiol, July 1, 2009; 297(1): C55 - C65.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
Y. Sano, V. G. Ornthanalai, K. Yamada, C. Homma, H. Suzuki, T. Suzuki, N. P. Murphy, and S. Itohara
X11-Like Protein Deficiency Is Associated with Impaired Conflict Resolution in Mice
J. Neurosci., May 6, 2009; 29(18): 5884 - 5896.
[Abstract] [Full Text] [PDF]

Home page
 Physiol. GenomicsHome page
H. A. Jackson, C. R. Marshall, and E. A. Accili
Evolution and structural diversification of hyperpolarization-activated cyclic nucleotide-gated channel genes
Physiol Genomics, May 11, 2007; 29(3): 231 - 245.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
K. J. Fogle, A. K. Lyashchenko, H. K. Turbendian, and G. R. Tibbs
HCN Pacemaker Channel Activation Is Controlled by Acidic Lipids Downstream of Diacylglycerol Kinase and Phospholipase A2
J. Neurosci., March 14, 2007; 27(11): 2802 - 2814.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
D. Boassa, W. D. Stamer, and A. J. Yool
Ion channel function of aquaporin-1 natively expressed in choroid plexus.
J. Neurosci., July 26, 2006; 26(30): 7811 - 7819.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
F. Deng, M. G. Price, C. F. Davis, M. Mori, and D. L. Burgess
Stargazin and other transmembrane AMPA receptor regulating proteins interact with synaptic scaffolding protein MAGI-2 in brain.
J. Neurosci., July 26, 2006; 26(30): 7875 - 7884.
[Abstract] [Full Text] [PDF]

Home page
 J. Physiol.Home page
T. A Simeone, J. M Rho, and T. Z Baram
Single channel properties of hyperpolarization-activated cation currents in acutely dissociated rat hippocampal neurones
J. Physiol., October 15, 2005; 568(2): 371 - 380.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
X. Zong, C. Eckert, H. Yuan, C. Wahl-Schott, H. Abicht, L. Fang, R. Li, P. Mistrik, A. Gerstner, B. Much, et al.
A Novel Mechanism of Modulation of Hyperpolarization-activated Cyclic Nucleotide-gated Channels by Src Kinase
J. Biol. Chem., October 7, 2005; 280(40): 34224 - 34232.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
C. Wahl-Schott, L. Baumann, X. Zong, and M. Biel
An Arginine Residue in the Pore Region Is a Key Determinant of Chloride Dependence in Cardiac Pacemaker Channels
J. Biol. Chem., April 8, 2005; 280(14): 13694 - 13700.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kimura, K.
Right arrow Articles by Nakanishi, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kimura, K.
Right arrow Articles by Nakanishi, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS