|
|
|
|
|
|
|
||
Article |
Address correspondence to Melitta Schachner, Zentrum für Molekulare Neurobiologie, Universität Hamburg, Martinistrasse 52, D-20246 Hamburg, Germany. Tel.: 49-40-42803-6246. Fax: 49-40-42803-6248. E-mail: melitta.schachner{at}zmnh.uni-hamburg.de
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
IßI) binds to the intracellular domain of NCAM180, and isolated spectrin subunits bind to both NCAM180 and NCAM140, as does the ßI spectrin fragment encompassing second and third spectrin repeats (ßI2–3). In NCAM120-transfected cells, ßI spectrin is detectable predominantly in lipid rafts. Treatment of cells with methyl-ß-cyclodextrin disrupts the NCAM120–spectrin complex, implicating lipid rafts as a platform linking NCAM120 and spectrin. NCAM140/NCAM180–ßI spectrin complexes do not depend on raft integrity and are located both in rafts and raft-free membrane domains. PKCß2 forms detergent-insoluble complexes with NCAM140/NCAM180 and spectrin. Activation of NCAM enhances the formation of NCAM140/NCAM180–spectrin–PKCß2 complexes and results in their redistribution to lipid rafts. The complex is disrupted by the expression of dominant-negative ßI2–3, which impairs binding of spectrin to NCAM, implicating spectrin as the bridge between PKCß2 and NCAM140 or NCAM180. Redistribution of PKCß2 to NCAM–spectrin complexes is also blocked by a specific fibroblast growth factor receptor inhibitor. Furthermore, transfection with ßI2–3 inhibits NCAM-induced neurite outgrowth, showing that formation of the NCAM–spectrin–PKCß2 complex is necessary for NCAM-mediated neurite outgrowth.
Key Words: NCAM; spectrin; PKC; neurons; outgrowth
The online version of this article includes supplemental material.
* Abbreviations used in this paper: FGFR, fibroblast growth factor receptor; MCD, methyl-ß-cyclodextrin; NCAM, neural cell adhesion molecule.
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Typically, cell surface transmembrane receptors involved in cell–cell and cell–matrix recognition interact with the cytoskeleton. The integrin receptors that bind to the extracellular matrix organize complex ensembles of interacting cytoskeletal proteins such as talin, tensin, vinculin, actin, and -actinin (Fields and Itoh, 1996; Juliano, 2002). The classical cadherins bind both actin and spectrin via -catenin, a cytoplasmic adaptor protein (Rimm et al., 1995; Gumbiner, 2000; Pradhan et al., 2001). The immunoglobulin superfamily recognition molecule L1 and other L1 family members, such as CHL1, neurofascin, NgCAM, NrCAM, and neuroglian, bind to ankyrin, a large adaptor protein with binding sites for spectrin (De Matteis and Morrow, 2000; Bennett and Baines, 2001). Another example is neural cell adhesion molecule (NCAM)* 180, the largest major isoform of NCAM, which interacts with spectrin (Pollerberg et al., 1986, 1987).
Spectrin is now recognized as a ubiquitous scaffolding protein that acts in conjunction with a variety of adaptor proteins to organize membrane microdomains on both the plasma membrane as well as on intracellular organelles. Spectrin can also link membranes and membrane protein complexes to filamentous actin or to microtubule transport motors (dynein–dynactin and some kinesins) (for reviews see Hirokawa, 1998; De Matteis and Morrow, 2000). The functional unit of spectrin is an ,ß heterodimer, although homopolymeric forms exist (Bloch and Morrow, 1989). Two spectrin genes encode -type subunits, and five genes encode the ß spectrins. Spectrin–membrane interactions are mediated by both protein–protein and protein–lipid interactions. The most studied interactions operate through ankyrin. These join spectrin to a variety of membrane receptors and channels, including Na,K-ATPase, voltage-gated Na+ channel, and tyrosine phosphate phosphatase CD45. Ankyrin-independent association with proteins, such as complexes of cadherin–catenin and NCAM180–spectrin, as well as with cortical actin, offers additional pathways of membrane interaction. Finally, many ß spectrins contain a pleckstrin homology domain. This domain mediates a direct interaction of spectrin with phosphatidylinositol 4,5-bisphosphate (PtdInsP2) and other acidic phospholipids.
Different ß spectrin isoforms play specific roles in the formation of unique membrane microdomains. For example, whereas ßII spectrin, the most widely distributed isoform, tends to be fairly uniformly distributed over the plasmalemma of neurons and other cells, ßI spectrin sorts to specific organelles (De Matteis and Morrow, 2000) and to organized plasma membrane domains, such as the motor end plate of skeletal muscle (Bloch and Morrow, 1989), the postsynaptic density of cerebellar neurons (Malchiodi-Albedi et al., 1993), or to CD45-rich patches in T cells (Pradhan and Morrow, 2002). ßI spectrin transcripts are now recognized in many nonerythroid cells, including neurons, lymphocytes, and epithelial cells. ßIII spectrin is another widely expressed spectrin associated with intracellular organelles and the plasma membrane. Given the close association of spectrin with membrane-associated receptor clusters in muscle, neurons, and lymphocytes, it is likely that spectrin may trap or stabilize proteins at specific loci in neural membranes. However, little is known of the mechanisms that guide spectrin to membrane domains, or the consequences of its participation in cell interactions.
In the present study, and given our earlier observations that NCAM180 associates with spectrin in the brain (Pollerberg et al., 1986, 1987), we have hypothesized that NCAM might initiate the segregation of spectrin to localized membrane microdomains. We have chosen here to study ßI spectrin, as it is the isoform most often associated with discrete receptor and organelle compartments, and is the form most prominent in postsynaptic densities. We find evidence for a direct interaction of the cytoplasmic domain of NCAM180 and NCAM140 with this spectrin and their colocalization in discrete membrane clusters. Building on observations that spectrin binds activated PKC (Rodriguez et al., 1999) and that PKC mediates NCAM-dependent neurite outgrowth (Kolkova et al., 2000a), we show that PKCß2, an isoform enriched in neurons, interacts with NCAM140 and NCAM180 via spectrin in a fibroblast growth factor receptor (FGFR)–dependent manner. Activation of NCAM results in a redistribution of the NCAM–ßI spectrin–PKCß2 complex to lipid-enriched microdomains to mediate neurite outgrowth.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Previous studies have demonstrated an interaction of NCAM180 with spectrin (Pollerberg et al., 1986, 1987). To extend this analysis, CHO cells and hippocampal neurons from an NCAM-deficient mouse were transfected with NCAM120, NCAM140, or NCAM180. Immunofluorescence analysis revealed that all three NCAM isoforms colocalized with spectrin, both in CHO cells and in neurons (Fig. 1, A and B). Cells transfected with NCAM (versus GFP alone) also accumulated more spectrin (Fig. 1, C and D). This was also observed in the brains of wild-type versus NCAM-deficient mice (Fig. 1 E). As spectrin is stabilized when incorporated into a detergent-resistant membrane cytoskeleton (Molitoris et al., 1996), we examined the impact of NCAM expression on spectrin's detergent solubility. In CHO cells expressing any of the three major NCAM isoforms, the 0.1% Triton X-100–insoluble fraction was enriched in spectrin, whereas there was no effect on the detergent-soluble fraction (see Fig. S2, available at http://www.jcb.org/cgi/content/full/jcb.200303020/DC1). We conclude that all major NCAM isoforms promote spectrin's incorporation into a detergent-insoluble membrane skeleton.
|
|
IßI) was measured by semiquantitative assay. Interestingly, the intact spectrin dimer bound selectively to IC180, but not IC140 (Fig. 3 A). However, when the subunits of the erythrocyte spectrin heterodimer were dissociated, both the I and ßI chains bound not only IC180, but also IC140 (Fig. 3, B and C). There was no binding of either subunit to BSA or to the intracellular domains of a different cell adhesion molecule, a close homologue of L1 (CHL1) (unpublished data). To verify whether binding of IC180 and IC140 to ßI spectrin monomers was physiologically relevant, we analyzed the oligomeric state of ßI spectrin bound to brain membranes, using Western blotting of membrane fractions after PAGE under nondenaturing conditions (Fig. 3 E). Two immunoreactive spectrin bands were evident: a complex of 
520 kD, presumably representing spectrin heterodimers and possibly ßI spectrin homodimers (Bloch and Morrow, 1989), and more importantly, a single band at 240 kD, representing ßI spectrin monomers. This membrane-bound ßI spectrin monomeric pool amounted to 13% of the total membrane-associated spectrin.
|
I subunit. This region also exhibits homology with -actinin (Byers et al., 1989; Winkelmann et al., 1990). Thus, although the binding to the isolated I spectrin was not likely to be biologically meaningful, given the paucity of I transcripts in either epithelial cells or neurons, the fact that this subunit could bind suggested that the biologically important ligand site might reside within homologous sequences in the NH2-terminal repeats of ßI spectrin. To verify that NCAM bound in a biologically significant way to this region of ßI spectrin, a series of spectrin truncation mutants were expressed in hippocampal neurons and in CHO cells. Neurons transfected with a GFP fusion construct encoding the NH2-terminal region of ßI spectrin and its first five repeat units (GFP–ßIN-5) demonstrated tight colocalization of NCAM with the GFP-labeled fragment (see Fig. S3, available at http://www.jcb.org/cgi/content/full/jcb.200303020/DC1), indicating that the ßIN-5 fragment bound the NCAM complex in vivo. This was confirmed when ßIN-5 was coimmunoprecipitated with NCAM in CHO cells after they were cotransfected with FLAG epitope–labeled ßIN-5 (FLAG–ßIN-5) and either NCAM140 or NCAM180 (Fig. 3 G). A larger spectrin fragment containing ßIN-5 fused to the ankyrin-binding region (repeats 14–15, ßIN-5,14,15) also coimmunoprecipitated with NCAM140 and NCAM180 (unpublished data). Spectrin ßI fragments encompassing repeats 1–2, 3–5, or 14–15 alone, or the actin- and dynactin-binding domain alone within the NH2 terminus, did not precipitate with NCAM (Fig. 3 G), whereas the ßI spectrin fragment containing 2–3 spectrin repeats coimmunoprecipitated with NCAM140 and NCAM180. We conclude that the 2–3 homologous repeat units of ßI spectrin are necessary and sufficient to bind to the intracellular domains of NCAM140 and NCAM180 in living cells.
Lipid rafts are necessary for the interaction of NCAM120 with ßI spectrin
GPI-anchored NCAM120 is confined mainly to lipid rafts that are insoluble in cold 1% Triton (Kramer et al., 1999; He and Meiri, 2002). The coincidence of spectrin with NCAM120 (Fig. 1) suggested that spectrin also associates with lipid rafts, as it does in erythrocytes (Salzer and Prohaska, 2001). CHO cells and NCAM-/- neurons were thus transfected with NCAM120, NCAM140, or NCAM180 and treated with 5 mM methyl-ß-cyclodextrin (MCD), which disintegrates lipid rafts. The relationship of spectrin to NCAM was then examined by immunofluorescence and coimmunoprecipitation (Fig. 4). MCD did not disturb the plasma membrane association of spectrin with NCAM140 or NCAM180. After MCD treatment, NCAM140 precipitated 74 ± 7.32% (n = 3) of the amount of spectrin that coprecipitated with NCAM180 (set to 100%). This level was similar to the amount precipitated in cells not treated with MCD. In contrast, when rafts were dispersed by MCD, the association of NCAM120 with spectrin was lost. In NCAM120-transfected CHO cells treated with MCD, the spectrin distribution internalized and shifted to a more perinuclear appearance (Fig. 4 A), presumably reflecting its association with intracellular organelles (De Matteis and Morrow, 2001). We conclude that NCAM120 interacts indirectly with spectrin through lipid rafts.
|
|
2 spectrin interacts with activated PKCß2 (Rodriguez et al., 1999). To investigate whether spectrin may link PKCß2 to NCAM, the distribution of these proteins was examined in hippocampal neurons (Fig. 6 A). NCAM colocalized with PKCß2 along neurites and in growth cones, where clusters of NCAM overlapped with intensely labeled accumulations of PKCß2. Similar results were also obtained with CHO cells and NCAM-/- neurons transfected with NCAM120, NCAM140, or NCAM180 (unpublished data). Clustering of NCAM by the application of antibodies to all NCAM isoforms to live neurons redistributed PKCß2 to NCAM clusters (Fig. 6 A). When neurons were extracted with 1% Triton X-100, 74.6 ± 1.3% (n = 11 neurons, 84 neurites) of the total PKCß2 immunofluorescence remained associated with NCAM clusters, confirming their coassociation in a detergent-insoluble complex (Fig. 6 A). Moreover, PKCß2 was coimmunoprecipitated with NCAM from mouse brain homogenates (Fig. 6 B). When precipitation was performed with antibodies against PKCß2, only NCAM140 and NCAM180 were coimmunoprecipitated (Fig. 6 C). Also in CHO cells, PKCß2 coimmunoprecipitated with NCAM140 and NCAM180. NCAM140 precipitated 58.7 ± 8.92% (n = 3) of the amount of PKCß2 that coprecipitated with NCAM180. PKCß2 did not coimmunoprecipitate with NCAM120 (Fig. 6 D). Given that spectrin and NCAM120 do coprecipitate (Fig. 2), the lack of PKCß2 in the spectrin–NCAM120 raft complex suggested that there may be competing interactions that can favor the release of activated PCKß2 from spectrin.
|
|
|
NCAM-mediated activation of PKC occurs via the FGFR (Kolkova et al., 2000a). Inhibition of the FGFR with the specific inhibitor PD173074 blocked redistribution of PKCß2 to detergent-insoluble NCAM clusters in response to NCAM activation (Fig. 8 C). It also significantly inhibited coprecipitation of PKCß2 with NCAM140/NCAM180 from CHO cells (Fig. 8 D). By contrast, inhibition of the FGFR did not affect the ability of NCAM to recruit spectrin and redistribute to lipid rafts (Fig. 8, C and D).
Formation of a complex between PKCß2, spectrin, and NCAM is implicated in NCAM-mediated neurite outgrowth
To investigate whether the association of PKCß2 with NCAM140 and NCAM180 was functionally important for NCAM-mediated outgrowth, we compared neurite length in hippocampal neurons from wild -type mice cotransfected with GFP and ßIN-5 or ßI2–3 with neurons transfected with GFP alone. Labeling of neurons transfected with GFP–ßIN-5 by NCAM antibodies under nonpermeabilizing conditions confirmed that ßIN-5 did not disturb delivery of NCAM to the plasma membrane (Fig. S3, A). In ßIN-5-transfected neurons, NCAM was distributed along neurites and accumulated in growth cones in a manner similar to nontransfected cells. GFP–ßIN-5 colocalized with NCAM and also accumulated in growth cones. When grown on a poly-L-lysine substrate, GFP- or GFP + ßIN-5–transfected neurons extended neurites equally well. Stimulation with soluble NCAM-Fc led to a significant increase in the neurite lengths of the GFP-transfected neurons, whereas the GFP + ßIN-5– or GFP + ßI2–3–transfected neurons were unresponsive (Fig. S3, B). Conversely, neurite outgrowth in response to NCAM-Fc stimulation in neurons transfected with spectrin ßIN-2 or ßIN,3–5 was not inhibited, suggesting that the ßIN-5 and ßI2–3 effect is due to the disruption of the NCAM–ßI spectrin–PKCß2 complex. We also compared NCAM-Fc– and L1-Fc–mediated neurite outgrowth from neurons transfected with the ßI2–3 spectrin construct. L1-mediated neurite outgrowth depends on the activation of pp60c-src, p21rac, and phosphatidylinositol 3-kinase, which differs from the NCAM-mediated signal transduction cascade (Ignelzi et al., 1994; Beggs et al., 1997; Schmid et al., 2000). Transfection with ßI2–3 did not inhibit L1-mediated neurite outgrowth (Fig. S3, B). We conclude that the association of PKCß2 with NCAM140 and NCAM180 via spectrin is implicated in NCAM-mediated neurite outgrowth.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
We have previously shown that FGFR activation in nonraft membrane domains, followed by the recruitment of NCAM140 to lipid rafts via palmitoylation, is necessary for neurite outgrowth (Niethammer et al., 2002). The present results establish that the spectrin-mediated recruitment of activated PKCß2 to lipid rafts is also a step along this pathway. Our concept of how this process works is presented in Fig. 9. Earlier work has established that PKC can be activated by diacylglycerol (Inoue et al., 1977; Kishimoto et al., 1980) via the FGFR (Walsh and Doherty, 1997; Kolkova et al., 2000a). We envision that the NCAM-mediated activation of the FGFR occurs independently of lipid rafts and induces the formation of a spectrin microdomain enriched in activated PKCß2 that binds to spectrin's pleckstrin homology domain. Subsequent palmitoylation of NCAM140 transfers the NCAM–spectrin–PKCß2 complex to a lipid raft. It is also possible that direct palmitoylation of spectrin may contribute to this process (Das et al., 1997). Once in the lipid raft, the NCAM–spectrin–PKCß2 complex encounters growth-associated and cytoskeletal control molecules that include GAP43, CAP23, and MARCKS (Laux et al., 2000; He and Meiri, 2002). These proteins, recruited to lipid rafts via myristolylation, are major substrates of PKC, and at least GAP43 is required for NCAM-stimulated neurite outgrowth (Meiri et al., 1998). Thus, the recruitment of the NCAM–spectrin–PKCß2 complex to lipid rafts may initiate the activation of downstream cytoskeletal organizers that contribute to neurite outgrowth.
|
It is interesting, in this respect, that NCAM180 coimmunoprecipitates spectrin with higher efficiency than NCAM140 or NCAM120. This phenomenon could account for our previous observation that NCAM180, but not NCAM140, coisolated with the spectrin by immunoaffinity chromatography (Pollerberg et al., 1986, 1987, note the purification procedure for NCAM140 and NCAM120 in these studies, which consisted of consecutive isolation of NCAM180 from the immunoaffinity-purified L1 fraction and of NCAM140 and NCAM120 from the residual fraction, thus possibly accounting for a depletion of spectrin in the latter fraction). The ability of spectrin to bind as the ß heterodimer to the recombinant intracellular domains of NCAM180, but not of NCAM140, has been noted previously (Pollerberg et al., 1986, 1987). We attribute little significance to the in vitro binding of the isolated I spectrin subunit, as this isoform of spectrin is either nonexistent or present at extremely low levels in either neurons or epithelial cells. However, both heterodimeric, and to a lesser extent homopolymeric, ßI spectrins exist in neurons and in other cells, often as specialized membrane complexes, such as with the secretory pathway, at the postsynaptic density, or with lymphocyte receptors (Bloch and Morrow, 1989; De Matteis and Morrow, 2000). As demonstrated here, 13% of the total membrane-bound ßI spectrin exists in a monomeric state. We thus believe that ßI spectrin could play a role in the control of NCAM-mediated signaling.
The question arises how NCAM120, the GPI-linked isoform of NCAM, associates with spectrin. It cannot do so directly, as it lacks an intracellular domain, yet it colocalizes and coimmunoprecipitates with spectrin in both neurons and in transfected CHO cells. Both NCAM120 and spectrin associate with lipid rafts, and this association is disturbed by the dispersal of rafts after treatment with MCD. We infer that a direct association with acidic lipids, possibly in rafts, may be mediated by spectrin's pleckstrin homology domain, as detected in other studies (for reviews see De Matteis and Morrow, 2000; Muresan et al., 2001). As PKCß2 binds to the pleckstrin homology domain and as this domain is occupied by acidic lipids in rafts, the lack of this kinase in the NCAM120 immunoprecipitates can be accounted for. The functional role of spectrin's association with NCAM120 is interesting in view of its predominant expression by astrocytes. NCAM120 is involved in signal pathways regulating astrocyte proliferation via glucocorticoid receptors (Krushel et al., 1998). These pathways are different from the signals regulating neurite outgrowth in neurons (Krushel et al., 1998). The intracellular glucocorticoid receptor binds to the heat shock protein and chaperone hsp70 (Morishima et al., 2000), which in turn binds to spectrin (Di et al., 1995). Whether spectrin also provides a platform for glucocorticoid receptors and whether NCAM120 modulates this interaction are intriguing issues for further studies.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Cultures and transfection of hippocampal neurons and CHO cells
Cultures of hippocampal neurons were prepared from 1–3-d-old C57BL/6J mice or from NCAM-deficient (NCAM-/-) mice (Cremer et al., 1994) inbred for at least nine generations onto the C57BL/6J background. Neurons were grown in 10% horse serum on glass coverslips coated with poly-L-lysine (100 µg/ml) in conjunction with laminin (20 µg/ml) (Dityatev et al., 2000). Transfection of hippocampal neurons and neurite outgrowth quantification were performed as previously described (Niethammer et al., 2002). CHO cells were maintained in Glasgow modified Eagle's medium containing 10% fetal calf serum. Cells were transfected using Lipofectamine Plus reagent (Invitrogen) following the manufacturer's instruction.
Fluorescence labeling
Indirect immunofluorescence staining of fixed cells was performed as previously described (Dityatev et al., 2000). Clustering of NCAM was induced by incubating live cells for 15 min (5% CO2 at 37°C) with NCAM antibodies, visualized with secondary antibodies applied for 5 min. To visualize cholesterol-enriched microdomains, fluorescein-conjugated cholera toxin B subunit (8 µg/ml) was applied to formaldehyde-fixed cells for 30 min at room temperature. Images were acquired using a confocal laser-scanning microscope (LSM510; Carl Zeiss MicroImaging, Inc.).
Colocalization analysis
For colocalization analysis, we defined an NCAM cluster as an accumulation of NCAM labeling with a mean intensity at least 30% higher than background. NCAM clusters were automatically outlined using the threshold function of the Scion Image software. Within the outlined areas, the mean intensities of NCAM, spectrin, PKCß2, and GM1 labeling associated with an NCAM cluster were measured. The same threshold was used for all groups. In nontransfected neurons, GM1 clusters were outlined using the same procedure. To determine the total amount of spectrin or PKCß2, neurites were manually outlined, and the total fluorescence of spectrin or PKCß2 along the neurites was measured. Colocalization profiles were plotted using LSM510 software.
Detergent extraction, cholesterol depletion, and cytoskeleton disruption
Detergent extraction followed Ledesma et al. (1998). Cells washed in PBS, pH 7.3, were incubated for 1 min in cold microtubule-stabilizing buffer (MSB; 2 mM MgCl2, 10 mM EGTA, 60 mM Pipes, pH 7.0) and extracted 8 min on ice with 1% Triton X-100 in MSB. After washing, cells were fixed with cold 4% formaldehyde. To deplete cholesterol from lipid rafts, cultures were incubated for 15 min at 37°C with 5 mM MCD (Sigma-Aldrich) in culture medium. To disrupt microtubules, cell cultures were incubated with vincristine (5 µM) for 5 h before fixation (Allison et al., 2000). For disruption of actin filaments, cultures were incubated in 5 µM latrunculin B for 24 h before fixation (Allison et al., 1998).
Protein purification
The IßI spectrin dimers were purified from erythrocyte ghosts (Shotton, 1998). Cells were washed three times in 15 volumes of 155 mM NaCl and then three times in 15 volumes of 155 mM sodium phosphate buffer, pH 7.6. Erythrocytes were lysed for 20 min in 15 volumes of 10 mM sodium phosphate buffer, pH 7.6. Erythrocyte ghosts were collected by centrifugation at 30,000 g for 10 min at 4°C, washed in ice-cold extraction buffer (1 mM EDTA, 10 µM DTT, pH 9.5), sonicated for 30 s, incubated for 60 min at 37°C with occasional mixing, and centrifuged at 230,000 g for 60 min at 4°C. The supernatant containing IßI spectrin dimers and actin was further purified by gel filtration chromatography on Sepharose 4B in 1 mM Tris-HCl buffer, pH 8.0, containing 100 µM EDTA, 10 µM DTT, and 0.02% NaN3.
The isolated spectrin I and ßI subunits of spectrin were purified according to Davis and Bennett (1983). The IßI spectrin dimers were dialyzed against 10 mM sodium phosphate buffer, pH 6.3, containing 7 M urea, 10 mM glycine, 0.05% Tween 20, 1 mM DTT, and applied to a hydroxylapatite column (Sigma-Aldrich). I subunit of spectrin was eluted by 80 mM sodium phosphate buffer, pH 6.3, containing 7 M urea, 10 mM glycine, 0.05% Tween 20, and 1 mM DTT. ßI spectrin was eluted by 250 mM sodium phosphate buffer, pH 7.1, containing 7 M urea, 10 mM glycine, 0.05% Tween 20, and 1 mM DTT. Subunits were dialyzed against the 10 mM sodium phosphate buffer, pH 7.4, containing 1 M NaBr, 6 M urea, 10 mM glycine, 0.05% Tween 20, 1 mM DTT, with subsequent dialysis against the same buffer without urea. Aggregated complexes of ßI spectrin were removed by gel filtration on Sepharose 4B.
DNA constructs
Rat NCAM140 and rat NCAM180/pcDNA3 were a gift of Patricia Maness (University of North Carolina, Durham, NC). Rat NCAM120 (gift of Elisabeth Bock, University of Copenhagen, Copenhagen, Denmark) was subcloned into the pcDNA3 vector (Invitrogen) by two EcoRI sites. The eGFP plasmid was from CLONTECH Laboratories, Inc. Spectrin ßI full-length DNA was used as template to produce fragments encoding ßI spectrin domains that were cloned in pcDNA3 vector (Invitrogen) and verified by DNA sequencing (Devarajan et al., 1997; De Matteis and Morrow, 2001).
Production of intracellular domains of NCAM140 and NCAM180
The BamHI sites were introduced at the 5' and 3' ends of the cDNAs encoding the NCAM140 or NCAM180 intracellular domains and were cloned in frame into the BamHI site of pQE30 (QIAGEN). Proteins were expressed in Escherichia coli strain M15 and purified on Ni-NTA-agarose (QIAGEN) according to the manufacturer's instructions.
ELISA protein ligand-binding assay
Intracellular domains of NCAM180 or NCAM140 (IC180 and IC140; 0.025–50 nM) were immobilized overnight on 96-well polyvinyl chloride plates (Nunc) in PBS. Wells were then blocked for 1.5 h with PBS containing 1% BSA and incubated for 1.5 h at RT with IßI spectrin dimers or I and ßI spectrin subunits diluted in PBS with 0.05% Tween 20 (PBS-T) and 1% BSA. Plates were washed three times with PBS-T and incubated for 1.5 h at room temperature with polyclonal antibodies against human erythrocyte spectrin diluted 1:4,000 in PBS-T containing 1% BSA. After washing with PBS-T, wells were incubated with peroxidase-coupled secondary antibodies in PBS-T containing 1% BSA, washed three times, and developed with 0.1% ABTS (Roche Diagnostics) in 100 mM acetate buffer, pH 5.0. The reaction was stopped with 100 mM NaF. The OD was measured at 405 nm.
Coimmunoprecipitation
Transiently transfected CHO cells were washed two times with ice-cold PBS, lysed 30 min on ice with RIPA buffer (50 mM Tris-HCl buffer, pH 7.5, containing 150 mM NaCl, 1% Nonidet P-40, 1 mM Na2P2O7, 1 mM NaF, 1 mM EDTA, 2 mM NaVO4, 0.1 mM PMSF, and protease inhibitor cocktail from Roche Diagnostics), and centrifuged for 15 min at 20,000 g at 4°C. Supernatants were cleared with protein A/G–agarose beads (Santa Cruz Biotechnology, Inc.) (3 h at 4°C) and incubated with NCAM Pab or control Ig (1.5 h, 4°C), followed by precipitation with protein A/G–agarose beads (1 h, 4°C). The beads were washed three times with RIPA buffer and analyzed by immunoblotting.
Detergent fractionation of spectrin from cells
Fractionation of spectrin into 0.1% Triton X-100–soluble and –insoluble material was performed according to Molitoris et al. (1996). In brief, CHO cells were washed twice with ice-cold PBS, collected with a rubber policeman, solubilized, and extracted for 5 min at 4°C with PHEM buffer (60 mM Pipes, 25 mM Hepes, 10 mM EGTA, 2 mM MgCl2, pH 6.9, containing 0.1% Triton X-100, 0.1 mM PMSF, 0.1 mM DTT). Cells were then centrifuged for 10 min at 48,000 g at 4°C. Supernatants and pellets were analyzed by immunoblotting.
Subcellular fractionation and isolation of lipid-enriched microdomains
Membrane fractions were isolated from brains of adult mice (Kleene et al., 2001), and rafts were prepared according to Brown and Rose (1992) with modifications. Membranes were lysed on ice for 20 min in four volumes of ice-cold 1% Triton X-100 in TBS and mixed with an equal volume of 80% sucrose in 0.2 M Na2CO3. A 10–30% linear sucrose gradient was layered on top of the lysate and centrifuged for 18 h at 230,000 g at 4°C (SW55 Ti rotor; Beckman Coulter). Raft fractions were collected as previously described (He and Meiri, 2002). Nerve growth cones were isolated as previously described (Pfenninger et al., 1983).
PKC assay and immunoblotting
PKC activity was measured using the PepTag assay for nonradioactive detection according to the manufacturer's instructions (Promega). Proteins separated by SDS-PAGE (8%) or PAGE (3–6%) (nondenaturing conditions) were electroblotted onto nitrocellulose transfer membrane (PROTRAN; Schleicher & Schuell) for 3 h at 250 mA. Immunoblots were incubated with the appropriate primary antibodies followed by incubation with peroxidase-labeled secondary antibodies and visualized using Super Signal West Pico reagents (Pierce Chemical Co.) on BIOMAX film (Sigma-Aldrich). Molecular weight markers were prestained protein standards from Bio-Rad Laboratories.
Online supplemental material
The supplemental material (available at http://www.jcb.org/cgi/content/full/jcb.200303020/DC1) includes images showing colocalization of NCAM and spectrin in detergent-insoluble clusters in neurons (Fig. S1), Western blots showing NCAM-mediated formation of a detergent-insoluble spectrin cytoskeleton (Fig. S2), and neurite outgrowth data showing the involvement of NCAM association with PKCß2 via spectrin in NCAM-mediated neurite outgrowth (Fig. S3).
|
Acknowledgments |
|---|
This work was supported by Deutsche Forschungsgemeinschaft (M. Schachner), and Zonta Club Hamburg-Alster (I. Leshchyns'ka) and in part by grants from the National Institutes of Health (J.S. Morrow).
Submitted: 4 March 2003
Revised: 24 March 2003
Accepted: 24 March 2003
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Aarts, L.H., P. Verkade, J.J. van Dalen, A.J. van Rozen, W.H. Gispen, L.H. Schrama, and P. Schotman. 1999. B-50/GAP-43 potentiates cytoskeletal reorganization in raft domains. Mol. Cell. Neurosci. 14:85–97.[CrossRef][Medline]
Allison, D.W., V.I. Gelfand, I. Spector, and A.M. Craig. 1998. Role of actin in anchoring postsynaptic receptors in cultured hippocampal neurons: differential attachment of NMDA versus AMPA receptors. J. Neurosci. 18:2423–2436.
Allison, D.W., A.S. Chervin, V.I. Gelfand, and A.M. Craig. 2000. Postsynaptic scaffolds of excitatory and inhibitory synapses in hippocampal neurons: maintenance of core components independent of actin filaments and microtubules. J. Neurosci. 20:4545–4554.
Beggs, H.E., S.C. Baragona, J.J. Hemperly, and P.F. Maness. 1997. NCAM140 interacts with the focal adhesion kinase p125(fak) and the SRC-related tyrosine kinase p59(fyn). J. Biol. Chem. 272:8310–8319.
Bhat, S., and D.H. Silberberg. 1986. Oligodendrocyte cell adhesion molecules are related to neural cell adhesion molecule (N-CAM). J. Neurosci. 6:3348–3354.[Abstract]
Bennett, V., and A.J. Baines. 2001. Spectrin and ankyrin-based pathways: metazoan inventions for integrating cells into tissues. Physiol. Rev. 81:1353–1392.
Bloch, R.J., and J.S. Morrow. 1989. An unusual ß-spectrin associated with clustered acetylcholine receptors. J. Cell Biol. 108:481–493.
Brown, D.A., and J.K. Rose. 1992. Sorting of GPI-anchored proteins to glycolipid-enriched membrane subdomains during transport to the apical cell surface. Cell. 68:533–544.[CrossRef][Medline]
Byers, T.J., A. Husain-Chishti, R.R. Dubreuil, D. Branton, and L.S. Goldstein. 1989. Protein sequence similarity of the amino-terminal domain of Drosophila ß spectrin to actinin and dystrophin. J. Cell Biol. 109:1633–1641.
Cremer, H., R. Lange, A. Christoph, M. Plomann, G. Vopper, J. Roes, R. Brown, S. Baldwin, P. Kraemer, S. Scheff, et al. 1994. Inactivation of the N-CAM gene in mice results in size reduction of the olfactory bulb and deficits in spatial learning. Nature. 367:455–459.[CrossRef][Medline]
Das, A.K., B. Dasgupta, R. Bhattacharya, and J. Basu. 1997. Purification and biochemical characterization of a protein-palmitoyl acyltransferase from human erythrocytes. J. Biol. Chem. 272:11021–11025.
Davis, J., and V. Bennett. 1983. Brain spectrin. Isolation of subunits and formation of hybrids with erythrocyte spectrin subunits. J. Biol. Chem. 258:7757–7766.
De Matteis, M.A., and J.S. Morrow. 2000. Spectrin tethers and mesh in the biosynthetic pathway. J. Cell Sci. 113:2331–2343.[Abstract]
De Matteis, M.A., and J.S. Morrow. 2001. ADP ribosylation factor (ARF) as a regulator of spectrin assembly at the Golgi complex. Regulators and Effectors of Small GTPases, part F. Vol. 329. W. Balch, editor. Academic Press, New York. 405–416.
Devarajan, P., P. Stabach, A.M. De Matteis, and J.S. Morrow. 1997. Na, K-ATPase transport from endoplasmic reticulum to Golgi requires the Golgi spectrin-ankyrin G119 skeleton in Madin Darby canine kidney cells. Proc. Natl. Acad. Sci. USA. 94:10711–10716.
Di, Y.P., E. Repasky, A. Laszlo, S. Calderwood, and J. Subjeck. 1995. Hsp70 translocates into a cytoplasmic aggregate during lymphocyte activation. J. Cell. Physiol. 165:228–238.[CrossRef][Medline]
Dityatev, A., G. Dityateva, and M. Schachner. 2000. Synaptic strength as a function of post- versus presynaptic expression of the neural cell adhesion molecule NCAM. Neuron. 26:207–217.[CrossRef][Medline]
Fields, R.D., and K. Itoh. 1996. Neural cell adhesion molecules in activity-dependent development and synaptic plasticity. Trends Neurosci. 19:473–480.[CrossRef][Medline]
Gennarini, G., M. Hirn, H. Deagostini-Bazin, and C. Goridis. 1984. Studies on the transmembrane disposition of the neural cell adhesion molecule N-CAM. The use of liposome-inserted radioiodinated N-CAM to study its transbilayer orientation. Eur. J. Biochem. 142:65–73.[Medline]
Gumbiner, B.M. 2000. Regulation of cadherin adhesive activity. J. Cell Biol. 148:399–404.
He, Q., and K.F. Meiri. 2002. Isolation and characterization of detergent-resistant microdomains responsive to NCAM-mediated signaling from growth cones. Mol. Cell. Neurosci. 19:18–31.[CrossRef][Medline]
Hirokawa, N. 1998. Kinesin and dynein superfamily proteins and the mechanism of organelle transport. Science. 279:519–526.
Ignelzi, M.A., Jr., D.R. Miller, P. Soriano, and P.F. Maness. 1994. Impaired neurite outgrowth of src-minus cerebellar neurons on the cell adhesion molecule L1. Neuron. 12:873–884.
