SEARCH

SEARCH BY CITATION

Keywords:

  • Glutamate receptors;
  • Microtubules;
  • Postsynaptic density;
  • Cytoskeleton

Abstract

  1. Top of page
  2. Abstract
  3. EXPERIMENTAL PROCEDURES
  4. Generation and expression of glutathione S-transferase (GST) fusion proteins
  5. GST affinity chromatography
  6. RESULTS
  7. GST affinity chromatography
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

Abstract : The cytoplasmic C-terminal domains (CTs) of the NR1 and NR2 subunits of the NMDA receptor have been implicated in its anchoring to the subsynaptic cytoskeleton. Here, we used affinity chromatography with glutathione S-transferase-NR1-CT and -NR2B-CT fusion proteins to identify novel binding partner(s) of these NMDA receptor subunits. Upon incubation with rat brain cytosolic protein fraction, both NR1-CT and NR2B-CT, but not glutathione S-transferase, specifically bound tubulin. The respective fusion proteins also bound tubulin purified from brain, suggesting a direct interaction between the two binding partners. In tubulin polymerization assays, NR1-CT and NR2B-CT significantly decreased the rate of microtubule formation without destabilizing preformed microtubules. Moreover, only minor fractions of either fusion protein coprecipitated with the newly formed microtubules. Consistent with these findings, ultrastructural analysis of the newly formed microtubules revealed a limited association only with the CTs of the NR1 and NR2B. These data suggest a direct interaction of the NMDA receptor channel subunit CTs and tubulin dimers or soluble forms of tubulin. The efficient modulation of microtubule dynamics by the NR1 and NR2 cytoplasmic domains suggests a functional interaction of the receptor and the subsynaptic cytoskeletal network that may play a role during morphological adaptations, as observed during synaptogenesis and in adult CNS plasticity.

Many attempts at the molecular characterization of the constituents of the postsynaptic density (PSD) were made over the past decades (Banker et al., 1974 ; Therien and Mushynski, 1976 ; Matus and Jones-Taff, 1978). This specialized postsynaptic membrane is thought to contain proteins that mediate receptor targeting to the synapse, and are modulators of receptor functions and linkers to the presynaptic active zone (for review, see Kennedy, 1993). Major cytoskeletal proteins, including actin, spectrin, tubulin, and microtubule-associated proteins (MAPs), are also found within purified PSD fractions. Despite the fact that no or very few microtubules have thus far been visualized within the PSD (Gulley and Reese, 1981 ; Chicurel and Harris, 1992), tubulin represents a substantial component (up to 14%) of its total protein (Kelly and Cotman, 1978). Partial contamination from other cellular compartments is likely, but a significant fraction of tubulin does resist solubilization by detergents. Furthermore, besides the well known cytosolic pool of tubulin that generates microtubules, a tubulin pool located in or associated with plasma membranes has been characterized (Bhattacharyya and Wolff, 1975 ; Walters and Matus, 1975 ; Bernier-Valentin et al., 1983 ; for review, see Stephens, 1986). Membrane-bound and soluble tubulin show distinct biochemical characteristics, such as differences in isoelectric point and C-terminal tyrosinylation, but the main function of membrane tubulin remains unknown. In recent studies, tubulin was reported to modulate adenylate cyclase activity, as well as phospholipase C (Wang et al., 1990 ; Hatta et al., 1995 ; Popova et al., 1997). This modulation of cellular signaling by tubulin was proposed to result from a direct transfer of the labile GTP from tubulin to the α subunit of the respective G proteins (Rasenick and Wang, 1988). Furthermore, tubulin coimmunoprecipitated with Gsα solubilized from synaptic membranes, again supporting an intimate link between the two proteins (Yan et al., 1996). These studies suggest an active role for tubulin in the regulation of cellular signaling by directly modulating G protein-mediated neurotransmitter actions. The functional roles of the various forms of tubulin may thus be more diverse than originally anticipated.

Recent electrophysiological and biochemical studies have suggested a functional link between the NMDA receptor channel complex and actin filaments or microfilaments (Legendre et al., 1993 ; Rosenmund and West- brook, 1993 a, b). Indeed, rundown of the NMDA channel can be prevented by the microfilament-stabilizing drug, phalloidin, supporting a close interaction between the receptor complex and the underlying cytoskeleton (Rosenmund and Westbrook, 1993a). Recently, Wyszynski et al. (1997) have reported on a direct binding between α-actinin-2 and both NMDA receptor subunits, NR1 and NR2B, suggesting that α-actinin-2 could act as a link between the NMDA receptor channel complex and subjacent microfilaments.

A rapidly growing family of novel organizer proteins was reported recently to be enriched in PSD fractions, the MAGUKs (membrane-associated guanylate kinases). This family includes PSD-95, also named synaptic associated protein-90 (SAP-90) (Cho et al., 1992), SAP-97 (hDlg) (Muller et al., 1995), SAP-102 (Muller et al., 1996), chapsyn-110 (PSD-93) (Brenman et al., 1996bKim et al., 1996), and CASK (Hata et al., 1996). These proteins are characterized by one to three 90-amino acid repeats (the so-called PDZ domain) in the N-terminal region, a central SH3 domain, and a C-terminal guanylate kinase homology region (for further details, see recent reviews : Fanning and Anderson, 1996 ; Saras and Heldin, 1996 ; Sheng, 1996). PDZ domains have been implicated in a variety of protein-protein interactions that regulate the subcellular distribution of both receptor complexes and signaling molecules (Brenman et al., 1996a ; Kim et al., 1996 ; Brakeman et al., 1997 ; Dong et al., 1997 ; Tsunoda et al., 1997). The NR2A/B subunits of the NMDA receptor channel complex were shown to interact directly with the respective PDZ domain of PSD-95, chapsyn-110, and SAP-102 (Kornau et al., 1995 ; Kim et al., 1996 ; Lau et al., 1996). In addition to binding NMDA receptor subunits, PSD-95 and chapsyn-110 were reported to interact with the shaker-subtype voltage-gated K+ channel (Kim et al., 1995). It is interesting that upon cotransfection of either PSD-95 or chapsyn-110 in COS-7 cells, both voltage-gated K+ and NMDA channels were described to cluster as compared with a more diffuse distribution upon single transfection (Kim et al., 1995). Moreover, the phosphorylation status of serine residues on the C-terminal domain (CT) of the NR1 subunit has been shown to influence its cellular distribution pattern (Ehlers et al., 1995). Both protein-protein interactions and activation of signaling cascades thus appear to play a critical role in the subcellular localization and in the electrophysiological properties of the NMDA receptor channel complex.

Although examples of the colocalization of PSD-95/chapsyn-110 with NMDA and K+ -channel complex have been described in brain sections, as well as in cultured neurons, the relevance of these interactions in vivo is not clear. Furthermore, as the overall expression patterns of these proteins do not fully overlap, additional organizer protein(s) may be anticipated. The goal of the present study was to identify novel partner proteins of the NMDA receptor channel and to evaluate their potential role. Here, we report on the interaction between soluble forms of tubulin and the CTs of the NR1 and NR2B subunits.

Generation and expression of glutathione S-transferase (GST) fusion proteins

  1. Top of page
  2. Abstract
  3. EXPERIMENTAL PROCEDURES
  4. Generation and expression of glutathione S-transferase (GST) fusion proteins
  5. GST affinity chromatography
  6. RESULTS
  7. GST affinity chromatography
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

As shown in a simplified diagram in Fig. 1 A, sequences corresponding to the entire CT of the NR1A subunit (amino acids 835-938, NR1-CTa), as well as to partial CTs of the NR1A (amino acids 865-938, NR1-CTb) and of the NR-2B subunit (amino acids 1,243-1,376, NR2B-CT), were amplified from the respective full-length cDNAs by PCR and subcloned into the pGEX-5X-1 expression vector (Pharmacia). All constructs were verified by sequencing. Overnight cultures of transfected E coli (JM-109 and/or BL-21 ; Novagen) were grown up to A600 of 0.5-0.6, at which time isopropyl-1-thio-β-D-galactopyranoside (1 mM) was added. After 3-4 h, bacterial cultures were pelleted, resuspended in 100 mM HEPES, 500 mM KCl, 1 mM dithiothreitol, and anti-proteases cocktail (1 μg/ml leupeptin, 1 μg/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride), and lysed by passage through a French press. Supernatants of bacterial lysates were aliquoted and kept frozen at -70°C until use. Time of protein induction, length and temperature of incubation, bacterial strains, and growth media were varied to decrease proteolysis of the various fusion proteins. However, although lower temperature (30°C) improved the yield of full-length fusion protein, significant protein degradation could not be prevented. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and western blot analysis of the bacterial lysates, with anti-GST antibody (Pharmacia), were performed according to published procedures by the manufacturer and confirmed the identity of the protein bands observed.

image

Figure 1. Polypeptide sequences and eluate analysis from NR1-CT and NR2B-CT incubated with brain fractions. A : Schematic representation of the NR1 and NR2B subunits and derived CT constructs, including their respective coding regions. Black boxes indicate hydrophobic sequence stretches. NR1-CTa includes the whole cytoplasmic CT, whereas NR1-CTb excludes the first 30 amino acids that corresponds to the so-called cassette 0 domain. NR1-CT and NR2B-CT were subcloned into the pGEX-5X expression vector and expressed in E. coli to generate the corresponding GST fusion proteins (B). Following adsorption on glutathione-Sepharose 4B, the GST fusion proteins were either eluted directly by the addition of 50 mM Tris-HCl, 10 mM glutathione, pH 8.0 (lanes 1) or incubated further with rat brain solubilized membrane (lanes 2) or cytosolic (lanes 3) protein fractions before elution. Eluted proteins were separated by SDS-PAGE and stained by Coomassie Blue. Following incubation with the brain cytosolic fraction (S2), significant amounts of tubulin were found in both NR1-CTb and NR2B-CT, but not in GST eluates (lanes 3). Molecular mass marker positions are shown (in kDa) on the left.

Download figure to PowerPoint

GST affinity chromatography

  1. Top of page
  2. Abstract
  3. EXPERIMENTAL PROCEDURES
  4. Generation and expression of glutathione S-transferase (GST) fusion proteins
  5. GST affinity chromatography
  6. RESULTS
  7. GST affinity chromatography
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

Rat brains were homogenized in 25 mM Tris-HCl, 150 mM NaCl, pH 7.4, 1 mM dithiothreitol, and anti-proteases cocktail. The supernatant following low-speed (2,000 g, 10 min) and high-speed (100,000 g, 60 min) centrifugations was kept as the cytosolic fraction (S2). The corresponding pellet was extracted in the same Tris-HCl buffer containing 1.5% (wt/vol) Triton X-100 and recentrifuged (100,000 g, 60 min) to result in the solubilized membrane fraction (S3).

The bacterial supernatants were incubated (30 min, 4°C) with glutathione-Sepharose 4B (Pharmacia). The immobilized fusion proteins were washed with 10 volumes of ice-cold phosphate-buffered saline containing 0.1% (wt/vol) Triton X-100 and either eluted from the Sepharose beads by the addition of 1 volume of 50 mM Tris-HCl, pH 8.0, 10 mM glutathione or further incubated at 4°C for 60 min with either cytosolic or solubilized membrane proteins or purified microtubule fractions (described below). CaCl2 (2 mM) was added to the Tris buffer where indicated. Eluted proteins were analyzed further by SDS-PAGE and stained by Coomassie Blue or transferred to polyvinylidene difluoride membrane (Bio-Rad). The membranes were incubated with either anti-GST (Santa Cruz or Pharmacia), anti-tubulin (Sigma), or anti-MAP-2 (gift from Dr. J. Kirsch, Frankfurt). Matched species secondary antibody coupled to horseradish peroxidase and enhanced chemiluminescence (Amersham) were used to visualize antigen-antibody complexes.

Microtubule purification and sedimentation assay

The purification of microtubules from rat brain was performed by repeated cycles of polymerization and disassembly as previously described (Vallée, 1986). In brief, brains were homogenized in PEM buffer (100 mM 1,4-piperazinediethanesulfonic acid, 1 mM EGTA, 1 mM MgCl2, pH 6.8), 1 mM dithiothreitol, anti-proteases cocktail. Tubulin polymerization was induced by the addition of 2 mM GTP (Boehringer Mannheim) and/or 20 μM paclitaxel (Sigma) to the respective S2 fraction, incubated at 37°C for 15 min, and centrifuged at room temperature (15,000 g, 20 min) through a 10% (wt/vol) sucrose-PEM buffer cushion. The pellet was then incubated on ice in fresh PEM buffer until complete depolymerization was apparent. This cycle was repeated three times and led to an enriched tubulin/MAP fraction (MT).

The isolated MT or purified tubulin fractions (Cytoskeleton, Denver, CO, U.S.A.) were incubated with the bacterial fusion proteins. The formation of microtubules was then induced by the addition of 2 mM GTP, and polymerized and free tubulin were separated by centrifugation through a sucrose-PEM buffer cushion (as described above). Supernatant and pellet fractions were analyzed further by SDS-PAGE and stained by Coomassie Blue. Western blot analysis of selected protein gels was also performed to confirm the identity of the protein bands (data not shown).

Measurements of tubulin polymerization

Aliquots of MT (110 μg) or tubulin (150 μg) were thawed in PEM buffer. Various amounts of purified GST or GST fusion protein (1-7.5 μg) and 2 mM GTP were added to the protein mixture and then transferred to a quartz cuvette (50 μl). The rate of microtubule formation was monitored by measuring the turbidity at 340 nm with a standard spectrophotometer. Measurements were performed at room temperature with MT mixtures, but at 35°C with purified tubulin fractions (without glycerol) to increase polymerization efficiency. Polymerization was allowed to proceed for 45 min or until a stable plateau was reached. In a different set of experiments, polymerization was allowed to proceed for 10 min, at which time purified GST or GST fusion proteins (5 μg) were added. The polymerization kinetics were then monitored for an additional 10 min. Selected samples were then processed further for electron microscopy (see below).

Electron microscopy

Following tubulin assembly, microtubule mixtures, with or without fusion protein, were adsorbed on carbon-coated copper grids and fixed with either 0.5% (wt/vol) glutaraldehyde or dithiobis(succinimidylpropionate) (DSP ; Pierce) for negative staining or immunostaining, respectively. After several rinses in distilled water, samples were either stained with a few drops of 1% (wt/vol) uranyl acetate solution or processed further for immunolabeling. Incubations with either anti-GST (Santa Cruz), anti-tubulin (Sigma), or gold-conjugated goat antimouse IgG (5 nm ; Sigma) antibody were performed in phosphate-buffered saline. Samples were further fixed with 0.5% (wt/vol) glutaraldehyde and negatively stained by the addition of 1% (wt/vol) uranyl acetate solution. The grids were then examined with a Zeiss EM-10 electron microscope. Even though the use of glutaraldehyde revealed a better preservation of the fine microtubular structure as compared with DSP, no significant staining with any of the antibodies used was observed.

GST affinity chromatography

  1. Top of page
  2. Abstract
  3. EXPERIMENTAL PROCEDURES
  4. Generation and expression of glutathione S-transferase (GST) fusion proteins
  5. GST affinity chromatography
  6. RESULTS
  7. GST affinity chromatography
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

Recombinant GST-NR1-CT, GST-NR2B-CT, and GST were generated in E. coli with the use of the pGEX expression vector and purified from bacterial lysates through binding to glutathione-Sepharose 4B. Purified fusion proteins are shown in Fig. 1B (lanes 1) ; all protein bands were recognized by the anti-GST antibody upon western blot analysis. Significant bacterial proteolysis could not be prevented during the purification of these proteins (see Experimental Procedures). Nevertheless, columns constructed with the purified constructs allowed for specific affinity purification from brain extracts. Eluates from the respective fusion proteins incubated with either solubilized membrane protein or cytosolic fractions from rat brain are shown in Fig. 1 B (lanes 2 and 3). An additional protein band migrating at ~50 kDa was obvious in eluates from both NR1-CTb and NR2B-CT that had been exposed to the cytosolic protein fraction. GST eluates or the bacterial extract per se did not reveal a band of this apparent molecular mass. Upon protein sequencing, this band was identified as a mixture of α- and β-tubulin (kindly performed by Dr. Lottspeich, Munich, Germany). Its nature was further confirmed by immunostaining with a monoclonal anti-tubulin antibody (data not shown, but see Fig. 2 B).

image

Figure 2. Tubulin and MAP-2 elution profiles. NR1-CTb and NR2B-CT immobilized on glutathione-Sepharose 4B were incubated with the rat brain cytosolic protein fraction (S2). Bound proteins were eluted by increasing salt concentration [25 mM Tris-HCl, 200, 400, or 600 mM NaCl, pH 7.4 (lanes 1-3)], followed by 50 mM Tris-HCl, 10 mM glutathione, pH 8.0 (lanes 4). Eluates were analyzed by SDS-PAGE and stained by Coomassie Blue (A). B : Fusion proteins were incubated with the cytosolic brain fraction in the presence or absence of 5 mM ATP. The glutathione eluates were then transferred to polyvinylidene difluoride membrane and incubated with an antibody mixture against tubulin and MAP-2 (B). Bound tubulin was essentially insensitive to the increasing concentration of salt in eluting buffer (A), as well as to the presence of ATP in the cytosolic brain fractions (B). In contrast, bound MAP-2 was eluted with low concentrations of salt (starting at 200 mM ; A). Moreover, addition of ATP to the brain fraction prevented the binding of MAP-2 to the fusion proteins and/or to tubulin that is bound to the fusion proteins (B).

Download figure to PowerPoint

In some eluates, an additional high molecular mass band (>200 kDa) also became apparent after Coomassie staining (Fig. 2 A). This band corresponds to several minor bands on a lower percentage gel (5% ; data not shown). One of these bands was positively recognized by a polyclonal anti-MAP-2 antibody (Fig. 2 B). PSD-95 and its homologues were apparently absent in these eluates, as none of the constructs contained the described sequence essential for this interaction (Kornau et al., 1995). High salt concentrations (up to 600 mM NaCl) did not decrease significantly the amount of tubulin bound to either NR1-CTb or NR2B-CT fusion proteins (Fig. 2 A). In contrast, MAP-2 was released by high salt concentrations (starting at 200 mM ; Fig. 2 A). Moreover, addition of ATP (5 mM) to the cytosolic brain fraction prevented the binding of MAP-2 to the immobilized fusion proteins and/or to tubulin bound to the fusion proteins, as revealed by western blot analysis (Fig. 2 B). Tubulin binding to both NR1-CTb and NR2B-CT was not affected by the presence of ATP (Fig. 2 B).

Significant amounts of tubulin from purified tubulin fractions bound to the NR1-CTb and NR2B-CT fusion proteins, but not to GST (Fig. 3). This binding was increased considerably upon the addition of CaCl2 (2 mM) to the incubation buffer (Fig. 3). Tubulin binding is thus most prominent when the fusion proteins are incubated with brain cytosolic fractions, but is also significant with purified tubulin fractions, especially in the presence of calcium. Recombinant GST-NR1-CTa was also generated to include the first 30 amino acid domain of the NR1-CT, which contains a high-affinity calmodulin binding site, as recently shown in a different study (Ehlers et al., 1996). Purified calmodulin bound significantly to NR1-CTa, but not to GST or NR2B-CT, nor did it compete with the tubulin binding to the NR1-CTa (data not shown).

image

Figure 3. Binding of purified tubulin to NR1-CT and NR2B-CT. Immobilized fusion proteins were incubated with purified tubulin (50 μg) in the absence (-) or presence (+) of 2 mM CaCl2. Eluted proteins were analyzed by SDS-PAGE and stained by Coomassie Blue. Tubulin bound most efficiently to the fusion proteins, but not to GST, upon addition of CaCl2 to the Tris buffer (25 mM Tris-HCl, 150 mM NaCl, pH 7.4). Experiments were carried out at 4°C throughout. Molecular mass marker positions are shown (in kDa) on the left.

Download figure to PowerPoint

Tubulin polymerization assay

To address the potential functional significance of the interaction between tubulin and the cytoplasmic domain of the NMDA receptor subunits, we performed coprecipitation assays and monitored the rate of microtubule formation in the presence or absence of the fusion proteins. Figure 4 shows the SDS-PAGE analysis of the recovered pellet and supernatant fractions following polymerization of purified tubulin (Fig. 4 A) or MT fractions (Fig. 4 B), upon addition of 2 mM GTP. In both sets of experiments, most of the fusion proteins remained in the supernatant fractions. Only a minor fraction of the NR1-CTa and NR2B-CT coprecipitated with the polymerized tubulin, suggesting a preferential binding of the fusion proteins to soluble forms of tubulin as compared with polymerized tubulin.

image

Figure 4. NR1-CT and NR2B-CT interact with soluble tubulin. Fusion proteins were incubated with either purified tubulin (A) or purified tubulin plus MAPs (B). Tubulin polymerization was induced by the addition of GTP. Newly formed microtubules were separated from soluble forms of tubulin by centrifugation through a sucrose cushion. Pellet (P) and supernatant (S) fractions were analyzed by SDS-PAGE and stained by Coomassie Blue. All fusion proteins remained largely in the supernatant fraction. Only a minor fraction coprecipitated with the newly formed microtubules. These findings suggest a higher affinity of NR1-CT and NR2B-CT for soluble tubulin compared with when it is incorporated into a microtubular structure.

Download figure to PowerPoint

Turbidity measurements (at 340 nm) allow for monitoring of the rate of microtubule formation. Here, tubulin polymerization was initiated by the addition of GTP (2 mM) to either purified tubulin or MT mixtures. It is interesting that the initial rate of microtubule formation was strongly decreased in the presence of 2-7.5 μg of either fusion protein (Fig. 5 A, B, D, and E). This decrease was observed for both tubulin and MT purified fractions. Moreover, similar concentrations of the NR1-CT and NR2B-CT did not significantly destabilize or induce significant depolymerization of microtubules (Fig. 5) C). GST per se had no significant effect at the various concentrations tested (Fig. 5 A and data not shown). The protein ratio of tubulin and the fusion proteins required for this effect was high (insets in Fig. 5 A, B, D, and E), suggesting that the CTs of the NMDA receptor subunits may interfere with the initiation or nucleation step of the in vitro microtubule formation.

image

Figure 5. Dose-dependent inhibition of tubulin polymerization by NR1-CT and NR2B-CT. Purified mixtures of MAPs and tubulin (A and B) or purified tubulin (D and E) were incubated with increasing amounts of either NR1-CTb or NR2B-CT fusion protein. GTP was added, and polymerization kinetics were monitored by measuring the turbidity at 340 nm. A : Polymerization profiles of MT mixture alone (•), MT plus GST (2.5 μg, ▴), and MT plus NR1-CTb (1.5 μg, ▪ ; 2.5 μg, ♦ ; 5 μg, ▾). B : Comparison of the profile of MT mixture alone (♦) with those of MT plus NR2B-CT (0.6 μg, • ; 1.8 μg, ▴ ; 6 μg, ▪). C : Polymerization kinetics were monitored for 10 min, at which time point NR1-CT (5 μg, •), NR2-CT (5 μg, ▴), or GST (5 μg, ▪) was added (arrow) to the MT mixture. D : Polymerization profiles of purified tubulin (•) and tubulin plus NR1-CTb (1.5 μg, ▪ ; 7.5 μg, ♦). E : Comparison of purified tubulin mixture (♦) with tubulin mixture plus NR2B-CT (1 μg, • ; 3 μg, ▴). Insets : Gel images revealing the compositions of the respective protein mixtures.

Download figure to PowerPoint

Electron microscopy

Samples containing microtubules formed in the presence of 5-6 μg of either fusion protein were analyzed further by electron microscopy. Highways of microtubules were observed in all samples (Fig. 6). After reaching equilibrium, certain tubulin dimers, as visualized by immunostaining, were found between the microtubular structures or as part of an intermediate structure, too small to be visible by negative staining (arrowheads in Fig. 6 B). Furthermore, the NR1-CT and NR2B-CT, as revealed by a GST antibody, were found in close proximity or bound to microtubules (arrows in Fig. 6 D and F), but also as “free” proteins in the microtubule mixture (arrowheads in Fig. 6 D and F). The free form possibly corresponds to CTs that are bound to smaller microtubular structures, which are not revealed by uranyl acetate contrasting. Finally, fusion proteins were also observed close to an apparently dynamic region of the microtubule like at the ends, where microtubules are elongating or disassembling (asterisk in Fig. 6 D) or at apparently linear substructures that may correspond to partially polymerized microtubules (asterisk in Fig. 6 F). In contrast to the NR1-CT and NR2B-CT, very few GST were found in close proximity or between the microtubules (arrowhead in Fig. 6 H). The secondary antibody did not bind nonspecifically to the carbon-coated films, and the fusion proteins, as well as GST per se, did not form aggregates under the experimental conditions used (data not shown). The electron microscopic analysis thus supports a very limited association of the NMDA receptor subunits with microtubules.

image

Figure 6. Electron microscopy analysis of newly formed microtubules. Photomicrographs of polymerized tubulin in the absence of fusion protein (A and B) are compared with microtubules formed in the presence of NR1-CT (5 μg ; C and D), NR2B-CT (6 μg ; E and F), or GST (5 μg ; G and H). Newly formed microtubules were fixed with either 0.5% glutaraldehyde or DSP and immediately stained with 1% uranyl acetate solution (A, C, E, and G) or further processed for immunolabeling (B, D, F, and H), respectively. Microtubule mixtures were first incubated with anti-tubulin (B) or anti-GST (D, F, and H) antibody, followed by an incubation with a gold-conjugated goat anti-mouse IgG (5 nm) antibody. The samples were then further fixed with 0.5% glutaraldehyde and negatively stained by the addition of 1% uranyl acetate solution. Arrows emphasize labeled tubulin moieties (B) or fusion proteins (D and F) incorporated or in close vicinity to the microtubules, whereas arrowheads point to tubulin yet to be incorporated (B) or to fusion proteins or GST that is more distant from the microtubular structures (D, F, and H). Asterisks indicate apparent dynamic areas of the microtubule (D and F).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. EXPERIMENTAL PROCEDURES
  4. Generation and expression of glutathione S-transferase (GST) fusion proteins
  5. GST affinity chromatography
  6. RESULTS
  7. GST affinity chromatography
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

Our data reveal an interaction between soluble forms of tubulin and the NR1 and NR2B CTs of the NMDA receptor subunits. This interaction decreased the rate of tubulin polymerization without significantly destabilizing preformed microtubules. Moreover, as shown by coprecipitation assays and electron microscopy, little binding was observed with newly formed microtubules, suggesting that tubulin dimers or soluble intermediate microtubular structures preferentialy interact with NMDA receptor subunits.

Op 18/stathmin and its close homologue, SCG10, have been identified recently as soluble factors that interfere with the formation of microtubules Stein et al., 1988 ; Belmont et al., 1996). SCG10 is neuron-specific, and its expression is temporally regulated with highest levels in developing CNS. It is interesting that, similar to the present findings with the NR1-CT and NR2B-CT, Op 18/stathmin and SCG10 were also shown to interact with tubulin dimers and to decrease dramatically, in a dose-dependent manner, the polymerization rate of tubulin (Belmont and Mitchison, 1996 ; Riederer et al., 1997). Furthermore, as observed for the cytoplasmic domains of the NMDA receptor subunits, SCG10 did not coprecipitate significantly with the newly formed microtubules, but remained largely in the supernatant fraction. It thus appears that all these proteins preferentially interact with the soluble forms of tubulin as compared with tubulin incorporated into highly structured microtubules. The very steep dose-response curve reported for SCG10, with concentrations ranging between 1 and 5 μM for minimal to maximal effect in decreasing the tubulin polymerization rate, respectively, is another characteristic shared by this family of proteins and the CTs of the NMDA receptor subunits. This atypical concentration dependence may indicate that these proteins predominantly interfere with the initiation or nucleation step of microtubule formation. The high ratio between the interacting proteins (tubulin/NMDA receptor CTs ; see insets in Fig. 5) further supports an effect on the nucleation versus elongation of microtubules. Finally, sequence comparisons between SCG10 and NR1-CT or NR2B-CT revealed 35-46% similarity (HUSAR sequence comparison program). However, in contrast to SCG10, the CTs of the NMDA receptor subunits did not significantly destabilize preformed microtubules. It remains to be determined to what extent the structural similarities underlie the similar modulation of tubulin polymerization and tubulin binding by these two protein families.

The synaptic plasma membrane protein syntaxin was also shown recently to bind tubulin (Fujiwara et al., 1997). A homologous tubulin binding sequence, such as found in MAPs, was found between amino acids 89 and 106 of syntaxin. We compared the amino acid sequence of the CTs of the NMDA receptor subunits, but could not identify regions of high homology to any of the known tubulin consensus binding motifs. However, the tubulin domain involved in this interaction is likely to be different from the ones that bind to known MAPs, as the former is affected by the incorporation of tubulin dimers into microtubules, whereas the latter are not (Paschal et al., 1989 ; Pedrotti et al., 1994).

The finding that the CTs of the NMDA receptor subunits bind preferentially to soluble forms of tubulin is supported further by the observation that tubulin bound more efficiently to the respective fusion proteins in the presence of millimolar concentrations of calcium. Chelation of calcium in the polymerization buffer is critical to allow for efficient tubulin polymerization (Weisenberg, 1972 ; Berkowitz and Wolff, 1981). Calcium can modulate the formation of microtubules in various ways, i.e., directly by interacting with tubulin or indirectly by modulating proteases and kinase/phosphatase activities (Berkowitz and Wolff, 1981 ; Keith et al., 1983 ; Job et al., 1985). However, the effects of calcium are most prominent when the nucleation conditions of microtubules are not optimal, i.e., at low temperatures and/or tubulin protein concentrations (Berkowitz and Wolff, 1981). Concentrations of calcium as low as micromolar were reported to be sufficient to decrease tubulin polymerization significantly under these conditions. In contrast, higher concentrations (high micromolar range) are usually required under biochemical conditions that closely mimic a physiological environment. A direct calcium effect on tubulin polymerization in a cellular context is thus unlikely (Keith et al., 1983). The observed increase in tubulin bound to the CTs of the NMDA receptor subunits in the presence of calcium (Fig. 3) may thus reflect a calcium-sensitive interaction between the two binding partners or, alternatively, that the calcium inhibition of tubulin polymerization leads to an increased availability of tubulin for binding. According to the data presented here, the latter hypothesis is more likely.

The morphology of synaptic spines is influenced by the dynamic status of the underlying cytoskeletal network, which in turn may be modified by synaptic activity (Harris and Kater, 1994 ; Kaech et al., 1996 ; Chen and Tonegawa, 1997). Over 30 proteins have been identified thus far in the subsynaptic network, including the well characterized structural proteins actin, myosin, spectrin, tubulin, MAPs, and dynamin (Kelly and Cotman, 1978 ; Walsh and Kuruc, 1992 ; Kennedy, 1993). Excitatory neurotransmission was shown to modulate the growth and morphology of neurons (Mattson and Kater, 1989 ; Bigot et al., 1991 ; Rashid and Cambray-Deakin, 1992). A close link between glutamate receptors and the subsynaptic network is thus anticipated. Thus far, most effects generated after glutamate release were attributed to local calcium influx. This calcium increase may be sufficient to destabilize the subsynaptic lattice by disrupting bonds formed by microfilament cross-linking proteins, like α-actinin, as well as by increasing the proteolysis of stabilizing proteins, such as brain spectrin (Baudry et al., 1981 ; Siman and Noszek, 1988 ; for review, see van Rossum, 1998). Moreover, an intimate link between NMDA receptor activation and the dynamics of microtubules is supported by recent reports. Halpain and colleagues have observed that calcium influx through NMDA receptors caused a significant decrease of MAP-2 phosphorylation via the activation of calcineurin (phosphatase 2B) (Halpain and Greengard, 1990 ; Quinlan and Halpain, 1996). MAP-2 is well known to stabilize microtubular structures, but its phosphorylation status is highly critical for its affinity to microtubules (Hernández et al., 1987 ; Brugg and Matus, 1991). Indeed, a decrease in MAP-2 phosphorylation led to an increased microtubular stability (Illenberger et al., 1996). Therefore, it is likely that NMDA receptor channel activation may indirectly affect the stability of microtubules by modulating the phosphorylation of various proteins, such as MAP-2. In agreement, the addition of excitatory amino acids to cortical and spinal cord neuronal cultures results in a rearrangement of MAP-2, tau, and tubulin immunoreactivity (Bigot and Hunt, 1990, 1991 ; Bigot et al., 1991). An increased MAP-2 binding to microtubules, as well as an increased resistance to the depolymerizing drug nocodazole, suggested a more stable microtubular network in the treated neurons as compared with controls (Bigot and Hunt, 1990, 1991 ; Bigot et al., 1991).

Our data are consistent with the view that the NMDA receptor channel complex does not modulate the dynamic status of the underlying cytoskeleton only through calcium entry, but also via direct protein-protein interactions. Wyszynski et al. (1997) recently reported an interaction between NMDA receptor subunits and α-actinin-2, which supports a close link between the glutamate channel complex and microfilaments. Our data suggest a direct interaction between soluble forms of tubulin and the NMDA receptor. This is in accordance with the fact that both tubulin and MAPs were consistently purified from PSD fractions, but no or very few microtubular structures were described within the dendritic spine (Westrum et al., 1980 ; Gulley and Reese, 1981 ; Bernhardt and Matus, 1984 ; Morales and Fifkova, 1989), except in more complex spine structures, such as described by Chicurel and Harris (1992). Tubulin is thus assumed to be in a dimeric or soluble form within the dendritic spine (Fig. 7). The intracellular domains of the NMDA receptor subunits may thus act as one of several local factors favoring the maintenance of a soluble pool of tubulin moieties, within the PSD, by buffering the absolute amount of tubulin available for the initiation of polymerization. Upon excitatory neurotransmission and localized calcium entry, the subjacent synaptic lattice is destabilized and allows for plastic rearrangements. Depending on the strength and duration of the synaptic activation, local concentrations of tubulin may rise and allow for polymerization to occur. The newly formed microtubule may be stabilized further by increased MAP-2 binding (Fig. 7). In support of such a hypothesis, the formation of long and stable bundles was reported following dissolution of the microfilament network in a MAP-2 transfected human hepatoma cell line (Edson et al., 1993). Thus far, few groups have observed fragmented microtubules in close proximity to the PSD or mature microtubules within the spines (Fifkova and Van Harreveld, 1977 ; Westrum et al., 1980 ; Gray et al., 1982). This may very well be due to the fact that only fixation methods best suited to preserve microtubular structures can give positive observations. Furthermore, the stability of microtubules is variable, depending on the cellular compartment in which they are localized (Gray et al., 1982). According to our model, the microtubule(s) in spines should be highly dynamic and dependent on the activation status of a particular spine at a given time. This high turnover rate may contribute to the difficulty in observing such structures in vivo (Gray et al., 1982). It is thus tempting to speculate that the CTs of the NMDA receptor subunits may constitute one of the many local factors that modulate the formation of this highly dynamic population of microtubules. Proteins, such as calmodulin and MAPs, likely further contribute to the modulation of tubulin polymerization equilibrium (Lee and Wolff, 1984 ; Kotani et al., 1985 ; Sobel et al., 1989).

image

Figure 7. Cytoskeletal gating of spine activity. The dendritic spine at a resting state is depicted with a relatively rigid subsynaptic lattice composed of a large variety of proteins. Emphasis is given to the anchoring of glutamate receptor channels [α-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) and NMDA] and G protein-coupled receptor (e.g., metabotropic), as well as of K+ channels via the cross-linking abilities of the MAGUK protein family. Right under and intermingled with the PSD fraction is a protein mesh composed essentially of microfilament and microfilament-regulating proteins. Tubulin and MAPs are also localized within the PSD fraction. Direct links between tubulin dimers or soluble forms of tubulin and the NMDA receptor channel complex, as well as between α-actinin and NMDA receptor subunits, are also shown. Upon synaptic activation, intracellular calcium concentrations increase, which in turn modulate a complex cascade of molecular events. Here is shown a dissolution of the cytoskeletal subsynaptic lattice that may allow for local concentrations of tubulin dimers to increase and polymerization to proceed. The newly formed microtubule(s) may be stabilized further by the high-affinity binding of MAPs. The NMDA subunits, as well as other local factors, such as calmodulin and MAPs, would play a critical role in maintaining the fragile dynamic equilibrium between polymerization and depolymerization of this highly labile population of microtubules in dendritic spines.

Download figure to PowerPoint

In conclusion, our data are consistent with the view that, in addition to forming an ionic channel, the NMDA receptor subunits may regulate spine morphology by modulating the local dynamics of tubulin structures. Future studies shall address whether the interactions described here contribute to the morphological adaptation elicited upon glutamatergic neurotransmission.

Acknowledgements

  1. Top of page
  2. Abstract
  3. EXPERIMENTAL PROCEDURES
  4. Generation and expression of glutathione S-transferase (GST) fusion proteins
  5. GST affinity chromatography
  6. RESULTS
  7. GST affinity chromatography
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

This work was supported by the Human Frontier Science Program (LT-174/95) and Fonds der Chemischen Industrie. D.V.R. currently holds a fellowship from the Medical Research Council of Canada. The cDNAs of the NMDA receptor subunits were generously provided by Prof. S. Nakanishi and Prof. M. Mishina. The authors also wish to acknowledge the technical expertise of W. Hofer in the handling of the electron microscope samples.

REFERENCES

  1. Top of page
  2. Abstract
  3. EXPERIMENTAL PROCEDURES
  4. Generation and expression of glutathione S-transferase (GST) fusion proteins
  5. GST affinity chromatography
  6. RESULTS
  7. GST affinity chromatography
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES
  • 1
    Banker G., Churchill L. & Cotman C.W. (1974) Proteins of the postsynaptic density.J. Cell Biol. 63,456465.
  • 2
    Baudry M., Bundman M.C., Smith E.K. & Lynch G. (1981) Micromolar calcium stimulates proteolysis and glutamate binding in rat brain synaptic membranes.Science 212,937938.
  • 3
    Belmont L.D. & Mitchison T.J. (1996) Identification of a protein that interacts with tubulin dimers and increases the catastrophe rate of microtubules.Cell 84,623631.
  • 4
    Belmont L., Mitchison T. & Deacon H.W. (1996) Catastrophic revelations about Op18/stathmin.Trends Biochem. Sci. 21,197198.
  • 5
    Berkowitz S.A. & Wolff J. (1981) Intrinsic calcium sensitivity of tubulin polymerization : the contributions of temperature, tubulin concentration, and associated proteins.J. Biol. Chem. 256,1121611223.
  • 6
    Bernhardt R. & Matus A. (1984) Light and electron microscopic studies of the distribution of microtubule-associated protein 2 in rat brain : a difference between dendritic and axonal cytoskeletons.J. Comp. Neurol. 226,203221.
  • 7
    Bernier-Valentin F., Aunis D. & Rousset B. (1983) Evidence for tubulin-binding sites on cellular membranes, plasma membranes, mitochondrial membranes, and secretory granule membranes.J. Cell Biol. 97,209216.
  • 8
    Bhattacharyya B. & Wolff J. (1975) Membrane-bound tubulin in brain and thyroid tissue.J. Biol. Chem. 250,76397646.
  • 9
    Bigot D. & Hunt S.P. (1990) Effect of excitatory amino acids on microtubule-associated proteins in cultured cortical and spinal neurones.Neurosci. Lett. 111,275280.
  • 10
    Bigot D. & Hunt S.P. (1991) The effects of quisqualate and nocodazole on the organization of MAP2 and neurofilaments in spinal cord neurons in vitro.Neurosci. Lett. 131,2126.
  • 11
    Bigot D., Matus A. & Hunt S.P. (1991) Reorganization of the cytoskeleton in rat neurons following stimulation with excitatory amino acids in vitro.Eur. J. Neurosci. 3,551558.
  • 12
    Brakeman P.R., Lanahan A.A., O'Brien R., Roche K., Barnes C.A., Huganir R.L. & Worley P.F. (1997) Homer : a protein that selectively binds metabotropic glutamate receptors.Nature 386,284288.
  • 13
    Brenman J., Chao D.S., Gee S.H., McGee A.W., Craven S.E., Santillano D.R., Wu Z., Huang F., Xia H., Peters M.F., Froehner S.C. & Bredt D.S. (1996 a) Interaction of nitric oxide synthase with the postsynaptic density protein psd-95 and alpha 1-syntrophin mediated by pdz domains. Cell 84,757767.
  • 14
    Brenman J.E., Christopherson K.S., Craven S.E., McGee A.W. & Bredt D.S. (1996 b) Cloning and characterization of postsynaptic density 93, a nitric oxide synthase interacting protein. J. Neurosci. 16,74077415.
  • 15
    Brugg B. & Matus A. (1991) Phosphorylation determines the binding of microtubule-associated protein 2 (MAP2) to microtubules in living cells.J. Cell Biol. 114,735743.
  • 16
    Chen C. & Tonegawa S. (1997) Molecular genetic analysis of synaptic plasticity, activity-dependent neural development, learning, and memory in the mammalian brain.Annu. Rev. Neurosci. 20,157184.
  • 17
    Chicurel M.E. & Harris K.M. (1992) Three-dimensional analysis of the structure and composition of CA3 branched dendritic spines and their synaptic relationships with mossy fiber boutons in the rat hippocampus.J. Comp. Neurol. 325,169182.
  • 18
    Cho K.O., Hunt C.A. & Kennedy M.B. (1992) The rat brain postsynaptic density fraction contains a homolog of the Drosophila discs-large tumor suppressor protein.Neuron 9,929942.
  • 19
    Dong H., O'Brien R.J., Fung E.T., Lanahan A.A., Worley P.F. & Huganir R.L. (1997) GRIP : a synaptic PDZ domain-containing protein that interacts with AMPA receptors.Nature 386,279284.
  • 20
    Edson K., Weisshaar B. & Matus A. (1993) Actin depolymerisation induces process formation on MAP2-transfected non-neuronal cells.Development 117,689700.
  • 21
    Ehlers M.D., Tingley W.G. & Huganir R.L. (1995) Regulated subcellular distribution of the NR1 subunit of the NMDA receptor.Science 269,17341737.
  • 22
    Ehlers M.D., Zhang S., Bernhadt J.P. & Huganir R.L. (1996) Inactivation of NMDA receptors by direct interaction of calmodulin with the NR1 subunit.Cell 84,745755.
  • 23
    Fanning A.S. & Anderson J.M. (1996) Protein-protein interactions : PDZ domain networks.Curr. Biol. 6,13851388.
  • 24
    Fifkova E. & Van Harreveld A. (1977) Long-lasting morphological changes in dendritic spines of dentate granular cells following stimulation of the entorhinal area.J. Neurocytol. 6,211230.
  • 25
    Fujiwara T., Yamamori T., Yamaguchi K. & Akagawa K. (1997) Interaction of hpc-1/syntaxin 1a with the cytoskeletal protein, tubulin.Biochem. Biophys. Res. Commun. 231,352355.
  • 26
    Gray E.G., Westrum L.E., Burgoyne R.D. & Barron J. (1982) Synaptic organisation and neuron microtubule distribution.Cell Tissue Res. 226,579588.
  • 27
    Gulley R.L. & Reese T.S. (1981) Cytoskeletal organization at the postsynaptic complex.J. Cell Biol. 91,298302.
  • 28
    Halpain S. & Greengard P. (1990) Activation of NMDA receptors induces rapid dephosphorylation of the cytoskeletal protein MAP2.Neuron 5,237246.
  • 29
    Harris K.M. & Kater S.B. (1994) Dendritic spines : cellular specializations imparting both stability and flexibility to synaptic function.Annu. Rev. Neurosci. 17,341371.
  • 30
    Hata Y., Butz S. & Sudhof T.C. (1996) Cask : a novel dlg/psd95 homolog with an N-terminal calmodulin-dependent protein kinase domain identified by interaction with neurexins.J. Neurosci. 16,24882494.
  • 31
    Hatta S., Ozawa H., Saito T. & Ohshika H. (1995) Participation of tubulin in the stimulatory regulation of adenylyl cyclase in rat cerebral cortex membranes.J. Neurochem. 64,13431350.
  • 32
    Hernandez M.A., Wandosell F. & Avila J. (1987) Localization of the phosphorylation sites for different kinases in the microtubule-associated protein MAP2.J. Neurochem. 48,8493.
  • 33
    Illenberger S., Drewes G., Trinczek B., Biernat J., Meyer H.E., Olmsted J.B., Mandelkow E.M. & Mandelkow E. (1996) Phosphorylation of microtubule-associated proteins MAP2 and MAP4 by the protein kinase p110mark : phosphorylation sites and regulation of microtubule dynamics.J. Biol. Chem. 271,1083410843.
  • 34
    Job D., Pabion M. & Margolis R.L. (1985) Generation of microtubule stability subclasses by microtubule-associated proteins : implications for the microtubule “dynamic instability” model.J. Cell Biol. 101,16801689.
  • 35
    Kaech S., Ludin B. & Matus A. (1996) Cytoskeletal plasticity in cells expressing neuronal microtubule-associated proteins.Neuron 17,11891199.
  • 36
    Keith C., DiPaola M., Maxfield F.R. & Shelanski M.L. (1983) Microinjection of Ca++-calmodulin causes a localized depolymerization of microtubules. J. Cell Biol. 97,19181924.
  • 37
    Kelly P.T. & Cotman C.W. (1978) Characterization of tubulin and actin and identification of a distinct postsynaptic density polypeptide.J. Cell Biol. 79,173183.
  • 38
    Kennedy M.B. (1993) The postsynaptic density.Curr. Opin. Neurobiol. 3,732737.
  • 39
    Kim E., Niethammer M., Rothschild A., Jan Y.N. & Sheng M. (1995) Clustering of shaker-type K+ channels by interaction with a family of membrane-associated guanylate kinases.Nature 378,8588.
  • 40
    Kim E., Cho K.O., Rothschild A. & Sheng M. (1996) Heteromultimerization and NMDA receptor-clustering activity of chapsyn-110, a member of the PSD-95 family of proteins.Neuron 17,103113.
  • 41
    Kornau H.C., Schenker L.T., Kennedy M.B. & Seeburg P.H. (1995) Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD-95.Science 269,17371740.
  • 42
    Kotani S., Nishida E., Kumagai H. & Sakai H. (1985) Calmodulin inhibits interaction of actin with MAP2 and tau, two major microtubule-associated proteins.J. Biol. Chem. 260,1077910783.
  • 43
    Lau L.F., Mammen A., Ehlers M.D., Kindler S., Chung W.J., Garner C.C. & Huganir R.L. (1996) Interaction of the N-methyl-D-aspartate receptor complex with a novel synapse-associated protein, SAP102. J. Biol. Chem. 271,2162221628.
  • 44
    Lee Y.C. & Wolff J. (1984) Calmodulin binds to both microtubule-associated protein 2 and τ proteins.J. Biol. Chem. 259,12261230.
  • 45
    Legendre P., Rosenmund C. & Westbrook G.L. (1993) Inactivation of NMDA channels in cultured hippocampal neurons by intracellular calcium.J. Neurosci. 13,674684.
  • 46
    Mattson M.P. & Kater S.B. (1989) Excitatory and inhibitory neurotransmitters in the generation and degeneration of hippocampal neuroarchitecture.Brain Res. 478,337348.
  • 47
    Matus A.I. & Jones-Taff D.H. (1978) Morphology and molecular composition of isolated postsynaptic junctional structures.Proc. R. Soc. Lond. 203,135151.
  • 48
    Morales M. & Fifkova E. (1989) Distribution of MAP2 in dendritic spines and its colocalization with actin.Cell Tissue Res. 256,447456.
  • 49
    Muller B.M., Kistner U., Veh R.W., Cases-Langhoff C., Becker B., Gundelfinger E.D. & Garner C.C. (1995) Molecular characterization and spatial distribution of SAP97, a novel presynaptic protein homologous to SAP90 and the Drosophila discs-large tumor suppressor protein.J. Neurosci. 15,23542366.
  • 50
    Muller B.M., Kistner U., Kindler S., Chung W.J., Kuhlendahl S., Fenster S.D., Lau L.F., Veh R.W., Huganir R.L., Gundelfinger E.D. & Garner C.C. (1996) SAP102, a novel postsynaptic protein that interacts with NMDA receptor complexesin vivo. Neuron 17,255265.
  • 51
    Paschal B., Obar R.A. & Vallee R.B. (1989) Interaction of brain cytoplasmic dynein and MAP2 with a common sequence at the C terminus of tubulin.Nature 342,569572.
  • 52
    Pedrotti B., Colombo R. & Islam K. (1994) Interactions of microtubule-associated protein MAP2 with unpolymerized and polymerized tubulin and actin using a 96-well microtiter plate solid-phase immunoassay.Biochemistry 33,87988806.
  • 53
    Popova J.S., Garrison J.C., Rhee S.G. & Rasenick M.M. (1997) Tubulin, Gq, and phosphatidylinositol 4,5-bisphosphate interact to regulate phospholipase Cb1 signaling.J. Biol. Chem. 272,67606765.
  • 54
    Quinlan E.M. & Halpain S. (1996) Postsynaptic mechanisms for bidirectional control of MAP2 phosphorylation by glutamate receptors.Neuron 16,357368.
  • 55
    Rasenick M.M. & Wang N. (1988) Exchange of guanine nucleotides between tubulin and GTP-binding proteins that regulate adenylate cyclase : cytoskeletal modification of neuronal signal transduction.J. Neurochem. 51,300311.
  • 56
    Rashid N.A. & Cambray-Deakin M.A. (1992) N-Methyl-D-aspartate effects on the growth, morphology and cytoskeleton of individual neurons in vitro. Dev. Brain Res. 67,301308.
  • 57
    Riederer B.M., Pellier V., Antonsson B., Di Paolo G., Stimpson S.A., Lutjens R., Catsicas S. & Grenningloh G. (1997) Regulation of microtubule dynamics by the neuronal growth-associated protein SCG10.Proc. Natl. Acad. Sci. USA 94,741745.
  • 58
    Rosenmund C. & Westbrook G.L. (1993 a) Calcium-induced actin depolymerization reduces NMDA channel activity. Neuron 10,805814.
  • 59
    Rosenmund C. & Westbrook G.L. (1993 b) Rundown of N-methyl-D-aspartate channels during whole-cell recording in rat hippocampal neurons : role of Ca2+ and ATP.J. Physiol. (Lond.) 470,705729.
  • 60
    Saras J. & Heldin C. -H. (1996) PDZ domains bind carboxy-terminal sequences of target proteins.Trends Biochem. Sci. 21,455458.
  • 61
    Sheng M. (1996) PDZs and receptor/channel clustering : rounding up the latest suspects.Neuron 17,575578.
  • 62
    Siman R. & Noszek J.C. (1988) Excitatory amino acids activate calpain I and induce structural protein breakdown in vivo.Neuron 1,279287.
  • 63
    Sobel A., Boutterin M., Beretta L., Chneiweiss H., Doye V. & Peyro-Saint-Paul H. (1989) Intracellular substrates for extracellular signaling : characterization of a ubiquitous, neuron-enriched phosphoprotein (stathmin).J. Biol. Chem. 264,37653772.
  • 64
    Stein R., Mori N., Matthews K., Lo L.C. & Anderson D.J. (1988) The NGF-inducible SCG10 mRNA encodes a novel membrane-bound protein present in growth cones and abundant in developing neurons.Neuron 1,463476.
  • 65
    Stephens R.E. (1986) Membrane tubulin.Biol. Cell 57,95110.
  • 66
    Therein H.M. & Mushynski W.E. (1976) Isolation of synaptic junctional complexes of high structural integrity from rat brain.J. Cell Biol. 71,807822.
  • 67
    Tsunoda S., Sierralta J., Sun Y., Bodner R., Suzuki E., Becker A., Socolich M. & Zuker C.S. (1997) A multivalent PDZ-domain protein assembles signalling complexes in a G-protein-coupled cascade.Nature 388,243249.
  • 68
    Vallée R.B. (1986) Reversible assembly purification of microtubules without assembly-promoting agents and further purification of tubulin, microtubule-associated proteins, and map fragments.Methods Enzymol. 134,89104.
  • 69
    Van Rossam D. (1998) Modulation of the cytoskeletal architecture by calcium, in Integrative Aspects of Calcium Signalling (Verkhratsky A. and Toescu E., eds), pp. 177196. Plenum Press, New York.
  • 70
    Walsh M.J. & Kuruc N. (1992) The postsynaptic density : constituent and associated proteins characterized by electrophoresis, immunoblotting, and peptide sequencing.J. Neurochem.59,667678.
  • 71
    Walters B.B. & Matus A.I. (1975) Tubulin in postsynaptic junctional lattice.Nature 257,496498.
  • 72
    Wang N., Yan K. & Rasenick M.M. (1990) Tubulin binds specifically to the signal-transducing proteins, Gsa and Gial.J. Biol. Chem. 265,12391242.
  • 73
    Weisenberg R.C. (1972) Microtubule formation in vitro in solutions containing low calcium concentrations.Science 177,11041105.
  • 74
    Westrum L.E., Jones D.H., Gray E.G. & Barron J. (1980) Microtubules, dendritic spines and spine apparatuses.Cell Tissue Res. 208,171181.
  • 75
    Wyszynski M., Lin J., Rao A., Nigh E., Beggs A.H., Craig A.M. & Sheng M. (1997) Competitive binding of α-actinin and calmodulin to the NMDA receptor. Nature 385,439442.
  • 76
    Yan K., Greene E., Belga F. & Rasenick M.M. (1996) Synaptic membrane G proteins are complexed with tubulin in situ.J. Neurochem. 66,14891495.