| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 281, Issue 6, 3669-3678, February 10, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1
3
From the
Departments of Psychiatry and Medicine, Brain Research Centre, University of British Columbia, Vancouver, British Columbia V6T 1Z3, Canada and ¶Department of Surgery, Vanderbilt University School of Medicine, Nashville, Tennessee 37232-2733
Received for publication, October 31, 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-amino-3-hydroxy-5-methyl-4-isoxazole propionate-type glutamate receptor subunit GluR1. In developing hippocampal neurons, expression of the tail domain of myosin Vb, but not myosin Va, enhanced GluR1 accumulation in the soma and reduced its surface expression. These changes were accompanied by reduced GluR1 clustering and diminished frequency of excitatory but not inhibitory synaptic currents. Similar effects were observed upon expression of full-length myosin Vb lacking a C-terminal region required for binding to the small GTPase Rab11. In contrast, mutant myosin Vb did not change the localization of several other neurotransmitter receptors, including the glutamate receptor subunit NR1. These results reveal a novel mechanism for the transport of a specific glutamate receptor subunit in neurons mediated by a member of the myosin V family. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Three known members of the myosin V family have been detected in brain extracts. The most studied member, myosin Va, is widely expressed in the brain (5). Dilute mice, which possess mutation in the myosin Va gene, suffer from impaired melanosome transport and severe seizures and die within 2-3 weeks after birth (6). These observations suggest that alteration in the transport of important yet unknown cargos contributed to the observed defects in neuronal function. In neurons, myosin Va is enriched at the postsynaptic density (PSD)4 of excitatory synapses (7) and associates with the scaffolding guanylate kinase domain-associated protein through interaction with the dynein light chain (8). The association of guanylate kinase domain-associated protein with the postsynaptic density protein-95 (PSD-95), a protein involved in glutamate receptor clustering, may functionally couple these proteins to myosin Va (1, 9).
Myosin Vb and myosin Vc are two additional members of the myosin V family that are also expressed in the brain; however, their exact localization in neurons remains unclear (10, 11). All of these motors share 
42% identity and contain a conserved N-terminal motor domain followed by a coiled-coil region and a globular C-terminal tail. The globular tail of class V myosins contains the cargo binding domain (2, 12). Expression of truncated forms of various members of the myosin V family lacking the N-terminal motor domain but containing the globular tail domain leads to a dominant-negative phenotype in different cultured cell systems (11, 13-17). This dominant negative approach revealed an important role of myosin Va in the control of melanosome transport in melanocytes (18). A similar strategy has been used to delineate the involvement of myosin Vb in recycling of several receptors in non-neuronal cells. These include the M4 subtype of muscarinic acetylcholine receptors in PC12 cells (15), transferrin and polymeric IgA receptors in HeLa and Madin-Darby canine kidney cells (11, 13, 14), and the CXC chemokine receptor 2 in immune cells (16).
Members of the Rab family of small GTPases have recently emerged as potential mediators of vesicle transport by members of the myosin V family (4, 19, 20). In melanocytes, myosin Va associates with melanosomes through interaction with a receptor complex containing Rab27a and melanophilin/Slac2a (21-25). Association of myosin Vb with Rab 11a has also been reported to regulate plasma membrane recycling (14). In contrast, the association of myosin Vc with Rab8-positive endosomes is involved in transferrin trafficking in HeLa cells (11). These observations indicate that coupling of individual members of the myosin V family to a particular subset of endosomal proteins may control the specificity of cargo transported by these motors.
Although these studies revealed some of the mechanisms underlying myosin V-mediated trafficking, the identity of cargo transported by these proteins in neurons remains unknown. Also unclear is whether various members of the myosin V family share similar mechanisms for delivering cargo to specific subcellular locations. In this study we show that myosin Vb is widely distributed in the brain and that it is expressed in several neuronal populations. In brain tissue myosin Vb associates with the -amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA)-type glutamate receptor subunit GluR1. Expression of the tail region of myosin Vb reduced surface expression and clustering of GluR1. These observations correlated with a decrease in the frequency of excitatory currents. Using myosin V mutants lacking regions required for coupling to Rab11, we further report that the effects of myosin Vb on GluR1 require coupling to Rab11. Taken together, these results uncover a novel role for myosin Vb in trafficking of a specific subunit of the AMPA-type excitatory neurotransmitter receptors in neurons.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cDNA Cloning and Mutagenesis—The cDNA encoding rat MyoVa was a gift from Dr. P. Bridgman (Washington University, St. Louis, MO). GFP-tagged mutant myosin Va (MyoVa CT) was generated by subcloning GFP in-frame with amino acids 1005-1830 of MyoVa in pCDNA3.1 (Invitrogen). The truncated form of myosin Vb (MyoVb CT) was generated by subcloning GFP in-frame with amino acids 1221-1846 of rat myosin Vb into pCR3.1 vector (Invitrogen). The truncated form of myosin Vb lacking the Rab11 interacting region was generated by deleting the last 49 amino acids at the C terminus of the GFP-tagged myosin Vb tail mutant (MyoVb CT 
Rab11; 1221-1797). The generation of full-length myosin Vb construct fused to GFP was described earlier (14). Full-length myosin Vb lacking the Rab11 interacting region was generated by deleting 15 amino acids corresponding to 1797-1811 (MyoVb FL Rab11; 1-1796, 1812-1846) using the QuikChange site-directed mutagenesis kit (Stratagene). The generation of hemagglutinin-tagged wild type GluR1 and flag-tagged full-length BERP was previously described (26, 27). The generated constructs were verified by sequencing.
Cell Culture and Transfections—Dissociated primary neuronal cultures were prepared from the hippocampi of E18/E19 rats. Briefly, hippocampi were dissociated by papain enzymatic digestion. Cells were coated on poly-D-lysine (Sigma)-treated coverslips. Cultures were maintained in neurobasal media (Invitrogen) supplemented with B27, penicillin, streptomycin, and L-glutamine as described elsewhere (28). COS-7 or HEK-293 cells were grown in Dulbecco's modified Eagle's medium or minimum essential medium (Invitrogen) containing 10% fetal bovine serum and penicillin and streptomycin. PC12 cells were cultured in RPMI 1640 medium (Invitrogen) containing 2 mM L-glutamine and penicillin and streptomycin. Transfections were performed using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol or calcium phosphate method (Clontech) as previously described (29). Neurons were transfected at least 3 days before fixation and immunostaining. For COS-7, PC12 and HEK cells, immunostaining or biochemistry were performed 24 to 48 h post-transfection.
Immunocytochemistry—Coverslips were fixed at room temperature for 10 min in 2% paraformaldehyde or in -20°C methanol when staining for synaptic proteins. Cells were washed 3 times with phosphate-buffered saline containing 0.3% Triton-X-100 before and after each antibody incubation. The following primary antibodies were used (immunoreactivity and dilution as indicated): GluR1 (rabbit, 1:1000; Upstate%20Biotechnology">Upstate Biotechnology); GluR2/3 (rabbit, 1:1000; Chemicon International); 
-amino butyric acid-type A 2 (rabbit, 1:2000; Alomone); Rab11 (rabbit, 1:500; Zymed Laboratories Inc. and mouse, 1:200; BD Transduction Laboratories); calnexin (rabbit, 1:200; Sigma); Rab5 (mouse, 1:200; BD Transduction Laboratories); NR1 (mouse, 1:1000; Synaptic Systems); GluR2 (mouse, 1:500; MAB397; Chemicon); PSD-95 (mouse, 1:200; Affinity Bioreagents); synaptophysin (mouse, 1:500; Pharmingen); microtubule-associated protein (MAP-2) (mouse, 1:500; BD Pharmingen); GM130 (mouse, 1:200; Transduction Laboratories); hemagglutinin (mouse, 1:1000; Upstate%20Biotechnology">Upstate Biotechnology); acetylcholine receptor 4 (guinea pig, 1:1000; Chemicon); GFP (guinea pig, 1:1000; custom made by Affinity Bioreagents). All antibody incubations were performed in blocking solution containing 2% normal goat serum for 1 h at room temperature or overnight at 4 °C. Cells were then incubated 1 h at room temperature in blocking solution containing the appropriate Cy3- or Alexa-conjugated secondary antibody (1:200; Jackson ImmunoResearch; 1:1000; Molecular Probes). Coverslips were then mounted on slides (Frost Plus; Fisher) with Fluoromount-G (Southern Biotechnology Associates, Birmingham, AL). GluR1 surface labeling experiments were carried out using antibodies directed against an extracellular region of GluR1 under non-permeabilized conditions (rabbit: 1:15; Oncogene Research) as previously described (30).
Imaging and Analysis—Images of neurons and cell lines used in this study were taken using a 63x objective affixed to a Zeiss Axiovert M200 motorized microscope and AxioVision software. For analysis of total and surface cluster density (number), the images were analyzed in Northern Eclipse (Empix Imaging, Missassauga, Canada) by using custom written software routines as described elsewhere (31). Briefly, dendrites of cells of interest were manually outlined using fluorescent signal. Puncta were defined as sites of intensities at least 1.5 times the dendritic background. The average puncta number per dendrite length in transfected cells was compared with either GFP-transfected or untransfected cells present in the same field. For changes in perinuclear accumulation in neuronal and non-neuronal cells, a minimum of 30 transfected cells per group was analyzed from at least 3 independent experiments. Image J 1.33u software (Wayne Rasband, National Institutes of Health) was used to represent graphically the fluorescence intensity patterns per µm. Statistical analysis were performed using Student's t test.
Immunoprecipitation, Western Blotting, and Subcellular Fractionation—For immunoprecipitation, whole brains from adult or postnatal day 16-19 Wistar rats were quickly removed. Brain tissue was homogenized in TEEN buffer (50 mM Tris-HCl, pH 7.4, 1 mM EDTA, 1 mM EGTA, 150 mMNaCl) supplemented with 10 mM ATP, 10 mM MgCl2, 2.5 mM Na3Va4, 1 mM phenylmethylsulfonyl fluoride (Sigma), and 1 protease inhibitor mixture tablet/10 ml (Roche Applied Science). For myosin Vb immunoprecipitation, cells were lysed in TEEN by the addition of 0.1% SDS, 0.8% Triton X-100. For GluR1 immunoprecipitation, cells were lysed in TEEN containing 0.5% deoxycholate and 1% Nonidet P-40. After rotation for 1 h at 4 °C, insoluble material was removed by centrifugation at 50,000 rpm for 30 min at 4 °C. Samples were then incubated for 1 h at 4 °C with 5µg of anti-myosin Vb or anti-GluR1 polyclonal antibodies raised against glutathione S-transferase fusion proteins of coiled-coil region of myosin Vb and carboxyl tail of GluR1, respectively. After the addition of 40 µl of protein A-Sepharose 4 Fast Flow beads (Amersham Biosciences), samples were incubated at 4 °C for 1 h or overnight. Immunoprecipitates were washed 3 times with TEEN buffer containing 1% Triton X-100. Samples were boiled in SDS-PAGE sample buffer with 10%
-mercaptoethanol for 3 min and analyzed by SDS-PAGE. Western blots signals were detected with Odyssey machine (Li-Cor) as previously described (30) or ECL (Amersham Biosciences). Co-immunoprecipitations from HEK-293 cells were performed in lysis buffer containing 1% Triton X-100, and samples were processed as described above. For subcellular fractionation, cerebral cortices from seven adult rats were homogenized and fractionated as previously described (32). 10 µg of each fraction was loaded on gel (P1, debris and nuclei; S1, postnuclear supernatant; P2, crude synaptosomal fraction; S2, small compartments; P3, microsomal pellet; S3, soluble protein fraction; LP1, synaptosomal membrane-enriched fraction; LS1, supernatant; LP2, synaptic vesicle-enriched fraction; LS2, presynaptic cytosol).
|
-amino butyric acid-type A receptor-mediated miniature postsynaptic currents (-60 mV) to isolate mEPSCs or at the reversal potential for ionotropic glutamate receptor-mediated miniature PSCs (+10 mV) to isolate mIPSCs. Recorded mEPSCs and mIPSCs were antagonized completely by the ionotropic glutamate receptor antagonist cyano-7-nitroquinoxaline-2, 3-dione (Sigma-Aldrich) and the -amino butyric acid-type A receptor antagonist bicuculline (Sigma-Aldrich), respectively (data not shown). Recordings were low pass-filtered at 2 kHz, sampled at 10 kHz, and stored in a computer by using CLAMPEX 8.0 (Axon Instruments). |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
200 kDa was detected in homogenates obtained from various brain regions of juvenile P17 (postnatal day 17) and adult rats (Fig. 1B). Myosin Vb was mainly detected in neurons in the cortex, hippocampus, septum, striatum, midbrain, and in specific neuronal subpopulations in the brain stem and cerebellum (Fig. 1C and data not shown). Intense myosin Vb immunoreactivity was also found in the posterior dentate gyrus of the hippocampus and pyramidal neurons in somatosensory cortex (Fig. 1D). In both adult neurons in brain slices and hippocampal neurons in culture, myosin Vb showed a prominent punctate staining in the perinuclear region and dendrites (Fig. 1D and Supplemental Fig. 1). Myosin Vb staining was also more intense in GAD-65-positive interneurons present in the hippocampus and cortex (data not shown).
|
Expression of a Mutant Form of Myosin Vb Alters the Distribution of Specific Glutamate Receptor Subunits—It is well established that the tail domains of myosin Va-Vc are critical for cargo transport (2, 12). Truncated forms expressing the tail domain of either myosin Va or myosin Vb accumulate in a perinuclear vesicular compartment that co-localizes with recycling endosomal markers and alters cargo transport in non-neuronal cells (13, 15, 18). To assess their role in neuronal protein trafficking, we transfected DIV 9 hippocampal neurons with GFP fusion constructs containing the tail domain of either myosin Va (MyoVa CT) (33) or myosin Vb (MyoVb CT) (10). 3-4 days post-transfection, neurons were fixed, and the distribution of several neuronal proteins was assessed using immunocytochemistry. Remarkably, neurons expressing MyoVb CT but not MyoVa CT showed a striking perinuclear accumulation of the AMPA-type glutamate receptor subunit GluR1 (Fig. 2, A-C). In contrast, no significant change in the accumulation of several other neurotransmitter receptors was observed, including the N-methyl-D-aspartic acid-type glutamate receptor subunits NR1 and NR2B, -amino butyric acid-type A 2, and the nicotinic acetylcholine receptor 4 (Fig. 2, D-G and J). Moreover, MyoVb CT did not alter the distribution of microtubule-associated protein MAP-2 or synaptophysin, suggesting a specific involvement of myosin Vb on GluR1 trafficking from a site emanating from the perinuclear region (Fig. 2, H and J). Consistent with previous studies, the truncated form of myosin Vb did not induce a gross rearrangement of internal organelles such as the Golgi and endoplasmic reticulum (ER) (Supplemental Fig. 2) (11, 14). Moreover, the perinuclear compartment containing GluR1 and myosin Vb mutant did not colocalize with either the ER marker calnexin or with the Golgi marker GM130, indicating that the accumulation of GluR1 with MyoVb CT is not due to aberrant trapping of GluR1 in these compartments.
Altered GluR1 localization suggests that myosin Vb exists in a protein complex containing GluR1. To assess this, myosin Vb was immunoprecipitated from brain tissue lysates, and possible binding partners were probed for using Western blotting. This analysis revealed that GluR1 co-immunoprecipitates with myosin Vb; however, neither NR1 nor the presynaptic protein synaptotagmin I was detected (Fig. 2K). Conversely, myosin Vb was also detected in GluR1 immunoprecipitates (Fig. 2L). To further characterize the relation between myosin Vb and GluR1 in developing hippocampal neurons, we next assessed GluR1 localization in DIV 7 and DIV 14 neurons at synaptic and non-synaptic sites (Supplemental Fig. 1). GluR1 was weakly clustered at DIV 7 neurons. Moreover, synaptophysin-positive puncta mainly lacked GluR1. This pattern resembles the distribution of myosin Vb in these young neurons. In DIV 14 neurons, GluR1 showed an accumulation at synaptic sites, which was more prominent than myosin Vb at the same age. However, at this stage significant amounts of both proteins were still detectable in the soma and the dendrites at non-synaptic sites. Subcellular fractionation analysis of extracts obtained from adult brain tissue revealed the presence of myosin Vb in the crude synaptosomal membranes (P2) and synaptosomal membrane-enriched fraction (LP1) that are enriched in GluR1 (Fig. 2M). This is consistent with the detection of both myosin Vb and GluR1 in the dendritic spines of mature (DIV 28) neurons. However, myosin Vb was not restricted to these fractions, but was also enriched in the soluble synaptosomal extracts (LS1 and LS2). This wide subcellular distribution resembles the one previously reported for myosin Va (34). These results suggest that in the adult brain, myosin Vb exists in a complex containing GluR1 in both synaptic and non-synaptic sites.
Surprisingly, expression of MyoVb CT had no effects on GluR1 localization in several heterologous expression systems tested, including COS-7 cells, HEK-293 cells, and the neuronal-like cell line PC12 (Supplemental Fig. 3 and data not shown). The inability of MyoVb CT to disrupt GluR1 localization in these cell type suggests a lack of direct association. To further assess this possibility, HEK-293 were co-transfected with either the C-terminal tail or full-length myosin Vb fused to GFP and hemagglutinin-tagged GluR1 or FLAG-tagged BERP, a known binding partner of myosin Vb. Consistent with previous studies, myosin Vb constructs co-immunoprecipitated BERP (Supplemental Fig. 4) (27). However, GluR1 was absent from myosin Vb immunoprecipitates. These results indicate that myosin Vb association with GluR1 may require an adaptor molecule exclusively expressed in neurons. Alternatively, differences in the vesicular machinery involved in sorting of GluR1 in heterologous cells and neurons may have contributed to the differential effects of mutant myosin Vb on GluR1.
Next, we analyzed whether enhanced perinuclear accumulation of GluR1 in neurons expressing mutant myosin Vb results in aberrant GluR1 trafficking and/or sorting to the plasma membrane. For this analysis, DIV 9 neurons were transfected with GFP or truncated forms of either myosin Va or Vb. Four days later, neurons were incubated with antibodies that recognize the extracellular domain of GluR1, fixed and incubated with a fluorescently conjugated secondary antibody under non-permeabilized conditions. Quantitative analysis of the number of GluR1 puncta per dendrite length in transfected cells compared with untransfected controls revealed reduced amounts of GluR1 (0.48 ± 0.05-fold) on the surface of neurons expressing MyoVb CT. In contrast, no change in GluR1 surface localization was observed in cells expressing either MyoVa CT (1.0 ± 0.2-fold) or GFP alone (1.0 ± 0.1-fold) (Fig. 3). These results show that expression of the mutant form of myosin Vb in neurons hinders the delivery of GluR1 to the cell surface.
To further characterize the role of myosin Vb in AMPA receptor trafficking, we also analyzed whether altered surface expression of GluR1 results in reduced clustering of total GluR1 at the synapse. For this analysis, DIV 9 neurons were transfected with either full-length myosin Vb or the truncated forms of myosin Va and Vb. Cells were fixed 4 days later, permeabilized, and stained for GluR1 and the synaptic marker synaptophysin (Syn). Quantitative analysis showed that expression of MyoVb CT but not full-length myosin Vb results in a 35% ± 6% decrease in the total number of GluR1 clusters (Fig. 4, A-D). The reduction of GluR1 clustering at sites positive for synaptophysin indicates a reduction in GluR1 accumulation at synapses (Fig. 4E). Expression of MyoVb CT, however, did not alter clustering of the N-methyl-D-aspartic acid receptor subunit NR1 (1.1 ± 0.2-fold) (Fig. 4, F-G). Moreover, the effects on GluR1 clustering were specific to MyoVb CT since the expression of MyoVa CT had no significant effect on the clustering of either GluR1 or NR1 (0.92 ± 0.07- and 0.95 ± 0.07-fold, respectively) (Fig. 4, D and G).
|
|
Myosin Vb-mediated Effects on GluR1 Involves Coupling to Rab11— Recent studies in non-neuronal cells showed that myosin Vb associates with the recycling endosome protein Rab11 and that this protein is involved myosin Vb-mediated cargo transport (13-15, 39, 40). In neurons, endogenous myosin Vb and Rab11 show partially overlapping colocalization in puncta present in the soma and dendrites (Supplemental Fig. 5). Thus, a plausible mechanism for vesicular transport of GluR1 in neurons may involve coupling of myosin Vb to Rab11. Consistent with previous studies, expression of the tail region of myosin Vb resulted in enhanced perinuclear accumulation of endogenous Rab11 in COS-7 cells, PC12 cells, and in cultured neurons (Supplemental Fig. 3 and Fig. 6, A and B). In contrast, no change in the distribution of early endosome-associated proteins, including Rab5 and the early endosome-associated protein 1 (EEA1), was observed upon MyoVb CT expression (Fig. 6, C and D, and data not shown).
|
Rab11) consisting of the C-terminal globular tail but lacking amino acids 1797-1846, which contains the region required for interaction with Rab11 (14). This construct is expected to neither associate with actin nor with Rab11. In transfected neurons, MyoVb CT Rab11 was detected in the soma and dendrites and was not restricted to the perinuclear region. Overexpression of this mutant form did not alter the localization of endogenous Rab11 when expressed in COS-7 cells or in neurons (Supplemental Fig. 6) or resulted in any significant change in GluR1 clustering (0.90 ± 0.08-fold) (Fig. 6, F and I). We also examined whether expression of the full-length version of myosin Vb lacking the region required for interaction with Rab11 (amino acids 1797-1811; MyoVb FL Rab11) (14) interferes with GluR1 trafficking (14). Because this mutant contains the actin binding motor domain, it is expected to associate and move along actin filaments; however, because it lacks the Rab11 binding site, such a mutant may interfere with cargo transport that requires coupling of Rab11 to myosin Vb. Unlike truncated form of myosin Vb, this mutant was detected in both the soma and dendrites of transfected neurons, and its localization was not restricted to the perinuclear region. Neurons transfected with MyoVb FL Rab11 exhibited a decrease in both total GluR1 clustering (0.68 ± 0.05-fold) and surface expression (0.73 ± 0.05-fold) (Fig. 6, H and I, and Fig. 7, C and D). These results support the notion that coupling of myosin Vb to Rab11 is involved in GluR1 trafficking. Finally, to directly assess the role of Rab11 in this process, we expressed a dominant negative form of Rab11 (Rab11-S25N) in DIV 9 neurons and analyzed its effect on GluR1 and myosin Vb localization. Overexpression of Rab11-S25N in hippocampal neurons moderately enhanced perinuclear accumulation of GluR1 and myosin Vb (Supplemental Fig. 7). Further analysis revealed that expression of Rab11-S25N results in a significant decrease in surface expression of GluR1 (0.71 ± 0.08-fold) (Fig. 7D). This suggests that blocking Rab11 function interferes with recycling of GluR1 in an endocytic compartment in the soma/dendrites and hampers its reinsertion at the plasma membrane. Consistent with this, recent findings showed that Rab11 controls recycling of AMPA receptors (41). Taken together, the findings provide further evidence that both myosin Vb and Rab11 cooperate in the regulation of GluR1 trafficking.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our analysis shows that myosin Vb is present in several neuronal populations in the brain. In neurons, the enrichment of endogenous myosin Vb in the perinuclear region suggests that myosin Vb is involved in regulating constitutive transport of cargo from a vesicular compartment in the soma. Notably, myosin Vb staining was absent from synaptic sites in young hippocampal neurons but was present in spines in mature neurons, suggesting that myosin Vb may also mediate cargo transport at synaptic sites at later stages of neuronal development. The presence of myosin Vb at dendritic spines of adult neurons, sites where myosin Va is also enriched (7, 43), suggests that both motor proteins may control some aspects of glutamate receptor trafficking at the synapse.
|
AMPA receptor subunits are thought to be synthesized and assembled in the ER and directed to the Golgi for subsequent post-translational modifications (47). A surprising finding depicted from our analysis is the effect of mutant myosin Vb on the clustering of the AMPA receptor subunit GluR1 but not GluR2. Also, GluR2 was absent from myosin Vb immunoprecipitates. This suggests that myosin Vb preferentially regulates a specific vesicular pool containing GluR1 but not GluR2. Biochemical analysis of AMPA receptor complexes from CA1/CA2 hippocampal pyramidal neurons revealed that 8% of the total AMPA receptor complexes contains homomeric GluR1 (48). Moreover, recent evidence suggests the existence of a pool of homomeric GluR1 in cultured hippocampal neurons (49). However, considering that a small amount of GluR1 is homomeric in pyramidal hippocampal neurons, it was surprising to see a noticeable change in the localization of AMPA receptors. Taking into consideration that changes in AMPA receptor localization was always analyzed 3-4 days after expression of mutant forms of myosin Vb, this may have accounted for the observed accumulation of GluR1 in the perinuclear region and its reduced surface expression. The partial reduction rather than total loss of GluR1 from the cell surface is consistent with this notion. Another possibility is that myosin Vb was selectively immunoprecipitated from a neuronal population containing homomeric GluR1. Indeed, the cortex has a large population of interneurons expressing homomeric GluR1. Although it is unlikely that we have selectively immunoprecipitated myosin Vb from a specific neuronal population mainly containing homomeric GluR1, this possibility cannot be excluded.
|
Several lines of evidence presented in this study suggest that coupling of myosin Vb to Rab11 is involved in GluR1 transport. First, expression of a truncated form of myosin Vb lacking the N-terminal motor domain but containing the globular tail domain interferes with both Rab11 and GluR1 trafficking. However, further deletion of the C-terminal region that harbors Rab11 binding site eliminates the dominant negative effects of this mutant on both Rab11 and GluR1. Second, results obtained from expression of full-length myosin Vb lacking amino acids required for binding to Rab11 further support this notion. This mutant contains the actin binding motor domain, and thus, it is expected to associate and move on actin filaments. Still, the failure of this mutant to associate with Rab11 may interfere with myosin Vb-associated cargo transport. Indeed, expression of this mutant resulted in a significant decrease in GluR1 clustering and surface expression. Third, expression of a dominant negative form of Rab11 (Rab11-S25N) redistributes GluR1 and myosin Vb in the soma and reduces surface expression of GluR1.
However, it remains unclear why myosin Vb-induced changes in GluR1 localization were only restricted to neuronal cells. This result was unexpected, since Rab11 localization was altered in both neuronal and non-neuronal cells. It is possible that GluR1 is differentially sorted to various populations of transport vesicles and that coupling to Rab11-positive endosomes only occurs in neurons. Nevertheless, we were unable to detect Rab11 in immunoprecipitates of both GluR1 and myosin Vb obtained from brain extracts (not shown). Taken together, these results suggest that interaction of myosin Vb with GluR1 is mediated by a neuron-specific adaptor protein rather than through direct association with Rab11. The presence of GluR1 in myosin Vb immunoprecipitates derived from brain lysates, but not from heterologous expression systems, lends further support to this concept. In neurons, putative adaptor proteins may include PDZ (PSD-95, Discs large, Zona occludens 1)-containing proteins, which have been implicated in regulating sorting of glutamate receptors (1, 51, 55-57). Coupling of myosin Vb to a neuronal protein complex containing the endosome-associated protein hrs, actinin-4, and BERP may have also contributed to the specific modulation of GluR1 trafficking observed in neurons (58). Despite the lack of a clear mechanism on how GluR1 is coupled to myosin Vb and Rab11, we believe that the substantial evidence presented here demonstrate that the link between myosin Vb and Rab11 is critical for GluR1 trafficking.
In summary, our data reveals a selective action of myosin Vb on the trafficking of specific neuronal proteins. However, it should be noted that only a subset of proteins have been examined in this study, and thus, one cannot exclude the possibility that myosin Vb regulates trafficking of many other neuronal proteins. Further work will be required to define other cargo regulated by this motor. Nevertheless, the robust effects observed on GluR1 localization indicate that this motor is critical for AMPAR receptor trafficking.
|
FOOTNOTES |
|---|
Research and the Michael Smith Foundation for Health Research (to A. E.-H. and Y. T. W.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 
The on-line version of this article (available at http://www.jbc.org) contains Supplemental Figs. S1-S7 and Appendix 1.
1 Supported by a University of British Columbia graduate fellowship and the Michael Smith Foundation for Health Research.
2 Supported by National Institutes of Health Grant DK48370.
3 Supported by funds from the EJLB foundation. To whom correspondence should be addressed. Tel.: 604-822-7526; E-mail: alaa{at}interchange.ubc.ca.
4 The abbreviations used are: PSD, postsynaptic density; GFP, green fluorescent protein; AMPA, -amino-3-hydroxy-5-methyl-4-isoxazole propionate; mEPSC, miniature excitatory postsynaptic current; mIPSC, miniature inhibitory postsynaptic current; ER, endoplasmic reticulum; DIV, days in vitro; HEK cells, human embryonic kidney cells; Syn, synaptophysin; BERP, brain-expressed ring finger protein.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
