• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Oct 29, 2002; 99(22): 14488–14493.
Published online Oct 18, 2002. doi:  10.1073/pnas.222546799
PMCID: PMC137910

The identification of vesicular glutamate transporter 3 suggests novel modes of signaling by glutamate


Quantal release of the principal excitatory neurotransmitter glutamate requires a mechanism for its transport into secretory vesicles. Within the brain, the complementary expression of vesicular glutamate transporters (VGLUTs) 1 and 2 accounts for the release of glutamate by all known excitatory neurons. We now report the identification of VGLUT3 and its expression by many cells generally considered to release a classical transmitter with properties very different from glutamate. Remarkably, subpopulations of inhibitory neurons as well as cholinergic interneurons, monoamine neurons, and glia express VGLUT3. The dendritic expression of VGLUT3 by particular neurons also indicates the potential for retrograde synaptic signaling. The distribution and subcellular location of VGLUT3 thus suggest novel modes of signaling by glutamate.

The amino acid glutamate is the principal excitatory neurotransmitter in the mammalian nervous system. Glutamate mediates the rapid synaptic signaling required for information processing (1), and changes in glutamatergic neurotransmission contribute to neural plasticity (2). However, it has been difficult to identify unambiguously the cells that release glutamate as a transmitter because this ubiquitous amino acid is a key intermediate in cellular metabolism and is required for protein synthesis. The enzyme glutaminase that produces glutamate from glutamine may be enriched in glutamatergic neurons, but is also expressed by many other cells and tissues (3–6). Plasma membrane glutamate transporters are also expressed primarily on astrocytes (7, 8). On the other hand, the most stringent criterion of neurotransmitter phenotype is exocytotic release, which in the case of classical transmitters such as glutamate requires transport into secretory vesicles.

The proteins responsible for vesicular glutamate transport eluded identification until recently, when it was recognized that a putative inorganic phosphate transporter mediates glutamate transport into synaptic vesicles (9, 10). Within the nervous system, vesicular glutamate transporter (VGLUT) isoforms 1 and 2 appear restricted to known glutamatergic neurons and exhibit a striking complementary pattern of expression at excitatory synapses (11–14), suggesting that they might define the excitatory neuronal phenotype. Indeed, heterologous expression of VGLUT1 or VGLUT2 suffices to convert inhibitory neurons to excitatory (10, 15).

Although the expression of VGLUT1 and VGLUT2 appears to account for the exocytotic release of glutamate by all known glutamatergic neurons, neither of these isoforms are expressed by a number of cells previously suggested to release glutamate by exocytosis. These cells include dopamine and serotonin neurons (16, 17), and astrocytes (18–20), raising the possibility that they express an additional isoform.

Materials and Methods

Molecular Cloning and RNA Analysis.

Using a fragment of the human VGLUT3 gene to screen a rat brain library, we identified one cDNA for VGLUT3 and isolated the remainder of the protein-coding region by PCR. For RT-PCR, a 1.2-kb fragment of VGLUT3 was amplified from random-primed first-strand cDNA by using 5′-CTGAGACCAAGATCCATAC-3′ and 5′-TATGAGTGTCCAGCAGTTCA-3′ and 35 cycles of PCR. PCR amplification with oligonucleotides 5′-GTGGGCCGCTCTAGGCACCAA-3′ and 5′-CTCTTTGATGTCACGCACGATTTC-3′ (25 cycles) generated a 540-bp control fragment of β-actin as control. In situ hybridization of brain sections from 21-day-old male rats was performed by using full-length VGLUT3 probes of high specific activity (>109 cpm/μg) (11).

Antibody Production and Immunostaining.

The sequence of the VGLUT3 cDNA encoding amino acid 530 to the C terminus was expressed as a GST fusion protein in the BL21 strain of Escherichia coli and used to generate polyclonal rabbit antisera. Immunocytochemistry was performed by using VGLUT3 antibody at a dilution of 1:8,000 and 20 μg/ml GST-VGLUT3 for adsorption (11). Pre-embedding immuno-electron microscopy was performed by using the immunoperoxidase method (21) and VGLUT3 antibody at 1:200–1:800. For immunofluorescence, 40-μm coronal sections were immunostained with VGLUT3 antibody at 1:2,000, a mouse mAb to choline acetyltransferase (Chemicon) at 1:200, a mouse mAb to tyrosine hydroxylase (Chemicon) at 1:100, and a mouse mAb to synaptophysin (Sigma) at 1:500. The antibody deposits were detected with goat anti-rabbit IgG conjugated to Alexa red and goat anti-mouse IgG conjugated to FITC and visualized with a confocal laser microscope.

Microcultures of rat ventral tegmental area, substantia nigra pars compacta, and raphe were prepared and stained as described (17) by incubation overnight at 4°C in the rabbit polyclonal antibody to VGLUT3 (1:1,000) and mAb to serotonin (1:200, Chemicon MAB352), followed by secondary anti-rabbit antibody conjugated to Texas red and anti-mouse antibody conjugated to FITC.

Heterologous Expression, Membrane Preparation, and Transport Assay.

PC12 cells transfected with rat VGLUT3 cDNA in the IRES2-EGFP vector (CLONTECH) were selected in G418 and screened by immunofluorescence for VGLUT3. Membranes from both transfected and untransfected cells treated with 5 mM butyrate for 24 h to induce VGLUT3 expression were prepared by differential centrifugation and assayed for transport as described (11).

Results and Discussion

We searched the human genome database and identified a sequence closely related to, but distinct from, VGLUT1 and VGLUT2. The full-length rat cDNA predicts a polytopic membrane protein of 588 aa that is 78% identical to VGLUT1, 74% identical to VGLUT2, and 45% identical to the lysosomal sialic acid transporter sialin (Fig. 5 A and B, which is published as supporting information on the PNAS web site, www.pnas.org). The high level of sequence identity to VGLUT1 and VGLUT2 suggested that the predicted protein might also transport glutamate into secretory vesicles.

To determine whether the putative transporter localizes to secretory vesicles, we raised an antibody to a bacterial fusion protein containing the predicted cytoplasmic C terminus, which shows very limited similarity to the established VGLUTs. The antibody recognizes a 65-kDa protein in two stably transfected PC12 cell clones, but not in untransfected PC12 cells or after adsorption with the antigen (Fig. 6A, which is published as supporting information on the PNAS web site). Immunofluorescence shows that the protein localizes to internal membranes and processes of nerve growth factor-treated PC12 cells (Fig. 6B), similar to VGLUT1 and VGLUT2. Differential centrifugation of rat brain extracts supports the localization to synaptic vesicles in vivo (Fig. 6C). Further, the protein cofractionates with VGLUT1 and synaptophysin on a population of light membranes by glycerol velocity sedimentation (Fig. 6D), a procedure that separates synaptic vesicles from essentially all other membranes (22). Like VGLUT1 and VGLUT2, a substantial proportion of the putative VGLUT thus localizes to synaptic vesicles. However, in the cell bodies of transfected PC12 cells, the putative transporter shows minimal colocalization with the synaptic vesicle protein synaptophysin (Fig. 6B). Although the differences in localization may simply reflect heterologous expression, differential centrifugation of brain extracts shows that the novel protein sediments differently than synaptophysin and VGLUT1 (Fig. 6C).

To assess glutamate transport by the novel protein, we isolated a population of light membranes from stably transfected PC12 cell clones (9, 11). Membranes from multiple transfected cell clones accumulate significantly more 3H-glutamate than those from untransfected cells (Fig. 7A, which is published as supporting information on the PNAS web site). The activity does not require Na+, and kinetic analysis showed that the activity has a Km ≈ 1.5 mM (Fig. 7B). Like VGLUT1 and VGLUT2, it recognizes glutamate but not aspartate (Fig. 7C), exhibits a biphasic dependence on chloride (Fig. 7E), and depends to a greater extent on ΔΨ than ΔpH of the H+ electrochemical gradient (Fig. 7D) (9–12, 15, 23, 24). Other groups have reported a greater dependence on ΔpH and some recognition of aspartate (25, 26), but these discrepancies may reflect differences in the transport activity conferred by different heterologous expression systems. Thus, the novel protein transports glutamate with properties very similar to VGLUT1, VGLUT2, and native synaptic vesicles from the brain (27, 28), and we will refer to it as VGLUT3.

Semiquantitative RT-PCR used to detect the low levels of VGLUT3 mRNA in vivo shows expression in the liver and kidney as well as the brain (Fig. 5C). By in situ hybridization, VGLUT3 antisense RNA labels multiple, discrete brain regions in a pattern very different from sense RNA (Fig. (Fig.11 AC) and that of VGLUT1 and VGLUT2 (11). We observe specific hybridization for VGLUT3 mRNA in the cerebral cortex, amygdala, hippocampus, and hypothalamus, all of which contain glutamatergic cell populations. However, VGLUT3 also labels the striatum (Fig. (Fig.11A), a region that does not contain the cell bodies of any known glutamate neurons. Dark-field microscopy of emulsion autoradiographs showed that layer II of the parietal cortex labels more strongly for VGLUT3 mRNA than other cortical layers (Fig. (Fig.11D). In contrast, VGLUT1 mRNA is expressed at high levels by multiple layers (Fig. (Fig.11E) and VGLUT2 transcripts are highly enriched in layer IV (Fig. (Fig.11F). In the hippocampus, moderate levels of VGLUT3 expression were observed by principal cells in the pyramidal and dentate gyrus granule cell layers (Fig. (Fig.11 G and H). Surprisingly, scattered cells in stratum radiatum of CA1–3 contain very high levels of VGLUT3 mRNA, suggesting expression by interneurons generally considered to be inhibitory. In addition, the substantia nigra pars compacta and ventral tegmental area contain moderate levels of VGLUT3 mRNA (Fig. (Fig.11I and Fig. 8, which is published as supporting information on the PNAS web site), suggesting expression by dopamine neurons. Further, cells in certain raphe nuclei label strongly for VGLUT3 (Fig. (Fig.11J), suggesting expression by a subset of serotonergic neurons. In the cerebellum, cells in the granular layer express VGLUT3 (Fig. (Fig.11K). Granule cells are excitatory, but additional labeling occurs in the molecular and Purkinje cell layers, again suggesting expression by cells that are not traditionally considered glutamatergic.

Fig 1.
Discrete cell populations express VGLUT3 mRNA. (A) In situ hybridization with 35S-labeled VGLUT3 antisense RNA shows strong labeling of a superficial layer in the cortex, including the piriform cortex (Pir), and structures in the caudate-putamen (CPu). ...

To characterize the distribution of VGLUT3 protein, we immunostained rat brain sections. In the hippocampus, punctate labeling was observed in the pyramidal cell and dentate gyrus granule cell layers, as well as at the border between stratum radiatum and lacunosum-moleculare (Fig. (Fig.22A). At high resolution, the punctate immunoreactivity surrounds unstained pyramidal cell bodies and proximal dendrites in CA1–3 (Fig. (Fig.22C), suggesting that VGLUT3, despite a presumed role in excitatory neurotransmission, might be expressed at inhibitory synapses made by interneurons. Supporting this possibility, stratum radiatum contains scattered cells labeled for VGLUT3 that resemble interneurons (Fig. (Fig.22E), consistent with the intense expression of VGLUT3 mRNA by scattered cells in this layer (Fig. (Fig.11 G and H). VGLUT3 puncta also contact unlabeled cell bodies and dendrites in stratum radiatum (Fig. (Fig.22D). Because the labeling of puncta around pyramidal cells strongly resembles the distribution of γ-aminobutyric acid (GABA)ergic terminals labeled for the biosynthetic enzyme glutamic acid decarboxylase (GAD) and the vesicular GABA transporter (29), we double-stained sections for GAD. Immunofluorescence shows that VGLUT3 colocalizes with GAD in a subset of GAD-positive cell bodies and processes (Fig. (Fig.33 AC). Although we observed expression of VGLUT3 mRNA by hippocampal pyramidal and dentate gyrus granule cells, we did not detect VGLUT3 protein at their terminals. These neurons have been observed to express mRNAs without the corresponding proteins (30).

Fig 2.
Immunocytochemical localization of VGLUT3 to nerve terminals, inteneuron cell bodies, and glia. (A) In the hippocampus, the VGLUT3 antibody labels puncta in the pyramidal and granule cell layers, and at the border between stratum radiatum and lacunosum-moleculare. ...
Fig 3.
GABAergic, cholinergic, and aminergic neurons express VGLUT3. (AC) Scattered interneurons in stratum radiatum (Rad) of the hippocampus double stain for both VGLUT3 (A) and GAD (B and C). A subset of GAD-immunoreactive processes within the pyramidal ...

In the neocortex, cell bodies resembling interneurons also label for VGLUT3 (Fig. (Fig.22G). These and scattered unstained cells, particularly in layer II, are heavily decorated by VGLUT3-immunoreactive puncta (Fig. (Fig.22H), a pattern that is observed in many different brain regions. As in the hippocampus, the location of VGLUT3-immunoreactive synapses on cell bodies and dendrites resembles that of vesicular GABA transporter and strongly suggests expression at inhibitory synapses.

In retina, the VGLUT3 antibody labels a discrete set of synapses within the inner plexiform layer (Fig. (Fig.33D). VGLUT3 also localizes to cell bodies of the inner nuclear layer, indicating expression by amacrine cells. Amacrine cells are generally considered inhibitory neurons that release either glycine or GABA (31), but they also express a wide range of other classical transmitters including acetylcholine (ACh) and dopamine (32, 33). However, cell bodies and processes expressing VGLUT3 do not colocalize with any other markers, including choline acetyltransferase for cholinergic neurons and tyrosine hydroxylase for catecholamine neurons (data not shown). They also fail to express GAD, which is expressed by a large proportion of all amacrine cells (Fig. (Fig.33 E and F). These results indicate that VGLUT3 defines a previously unidentified subset of amacrine cells that may be capable of exocytotic glutamate release and thus excitatory.

The striatum contains VGLUT3-immunoreactive cell bodies (Fig. (Fig.22F) that presumably correspond to the striatal cells expressing VGLUT3 mRNA by in situ hybridization (Fig. (Fig.11A). To assess possible expression by cholinergic interneurons, we double-stained for choline acetyltransferase. Although we observed no colocalization in basal forebrain, cortex, hippocampus, or motor nuclei, the VGLUT3 antibody specifically labels most cholinergic cell bodies in the dorsal striatum (Fig. (Fig.33 GI), suggesting that these cells signal through glutamate as well as ACh. Previous work has focused on the neuromodulatory actions of ACh released by these cells (34), but the expression of VGLUT3 indicates the potential for rapid synaptic signaling.

In light of the VGLUT3 expression in substantia nigra pars compacta, ventral tegmental area, and raphe nuclei observed by in situ hybridization, we double-stained brain sections for the catecholamine marker tyrosine hydroxylase and the plasma membrane serotonin transporter, but observed no colocalization in the cortex, hippocampus, or brainstem (data not shown). However, previous work has raised the possibility of segregated release sites for monoamines and glutamate (17, 35). To confirm expression by serotonin neurons, we examined microcultures from the raphe, where all of the processes in an island derive from a single cell. We found extensive colocalization of VGLUT3 with either serotonin or its biosynthetic enzyme tryptophan hydroxylase in >50% of serotonergic neurons (Fig. (Fig.33 JL), supporting the physiological significance of previous studies demonstrating glutamate release by monoamine neurons in culture (16, 17). Remarkably, individual axonal processes of isolated raphe neurons were often observed to contain either VGLUT3 or serotonin, supporting previous suggestions that dopamine neurons segregate monoamine and glutamate markers (17, 35).

To extend these observations to the ultrastructural level, we performed immuno-electron microscopy. In hippocampal pyramidal and granule cell layers as well as throughout neocortex, immunoperoxidase staining for VGLUT3 labels a substantial number of nerve terminals. All of the labeled terminals form symmetric rather than asymmetric synapses (Fig. (Fig.44 A and B), consistent with the expression of VGLUT3 at inhibitory synapses suggested by light microscopy. Further, the labeled terminals make contact with the shaft of proximal dendrites, a location characteristic of inhibitory synapses. The results thus suggest release of glutamate at what otherwise resembles an inhibitory synapse. Although it remains possible that these cells include a subset of excitatory interneurons (36) that do not release GABA, the colocalization of VGLUT3 with GAD, the expression at symmetric synapses, and the termination on cell bodies and dendritic shafts strongly suggest release of both glutamate and GABA by the same cells. However, only a subset of symmetric synapses in hippocampus and neocortex appear to stain for VGLUT3.

Fig 4.
Ultrastructural localization of VGLUT3 to inhibitory nerve endings, dendrites, and astrocytic processes. (A) A VGLUT3-immunoreactive terminal (T) makes symmetric synapses (demarcated by arrowheads) onto two unlabeled pyramidal cells (p) in area CA1 of ...

We also observed that VGLUT3 localizes to dendrites in multiple brain regions including the striatum (Fig. (Fig.44 CG). The staining can extend throughout the dendritic tree (Fig. (Fig.44G) and typically involves internal membranous structures located near synaptic contacts, which are often asymmetric and unlabeled (Fig. (Fig.44 C and EG). This subcellular localization raises the possibility of a role for VGLUT3 in retrograde synaptic signaling by glutamate. Previous work in the olfactory bulb has indeed demonstrated robust dendrodendritic signaling between excitatory mitral cells and inhibitory granule cells (37, 38), and we detected VGLUT3 expression by processes with morphological features of dendrites in the external plexiform layer of the olfactory bulb (Fig. 9, which is published as supporting information on the PNAS web site), which contains these synapses. Recent work has also identified retrograde signaling by exocytotic glutamate release from pyramidal cell dendrites at inhibitory synapses in the cortex (39).

VGLUT3 further localizes to a subset of astrocytes throughout the brain, by both light microscopy (Fig. (Fig.22 I and J) and immuno-electron microscopy (Fig. (Fig.44 H and I). Although the astrocyte labeling is less intense than the neuronal, and most convincing in white matter tracts, it nonetheless occurs in a large proportion of the cells. Perivascular staining of glial endfeet also occurs (Fig. (Fig.44I) but is not the predominant feature (Fig. (Fig.44H). These results warrant a consideration of the possibility that astrocytes release glutamate by exocytosis in vivo as well as in culture (18–20).

The results show that a number of cell populations not traditionally considered to release glutamate by exocytosis express a unique VGLUT. In these cells, VGLUT3 could have a number of physiological roles. For example, glutamate uptake might help to load secretory vesicles with a different transmitter. Indeed, previous work has indicated synergism between the vesicular transport of another anion, ATP, and monoamines (40). Alternatively, VGLUT3 could serve to create an intracellular store that buffers cytoplasmic glutamate. Expression in liver and kidney might be considered to support these alternatives to a role for VGLUT3 in glutamate release. However, glutamate absorption by the kidney requires export across the basolateral surface of renal epithelial cells, and nitrogen metabolism by the liver involves glutamate transfer between hepatocytes (41, 42), raising the possibility that VGLUT3 contributes to the exocytotic release or at least the efflux of glutamate by these non-neural cells. Further, heterologous expression of VGLUT1 and VGLUT2 suffices to confer exocytotic glutamate release (10, 15), and VGLUT3 transports glutamate with properties very similar to VGLUT1 and VGLUT2. We thus presume that VGLUT3 contributes to the exocytotic release of glutamate.

In summary, the expression of VGLUT3 provides a mechanism for the exocytotic release of glutamate by cholinergic, monoaminergic, and GABAergic neurons as well as astrocytes. Because glutamate generally mediates fast neurotransmission with the potential for many forms of plasticity whereas ACh and monoamines in particular often act as neuromodulators, the release of glutamate and ACh by striatal interneurons and of glutamate and monoamines by dopamine and serotonin neurons has profound implications for the role of these cells in motor control, reward pathways, and neuropsychiatric disease (43). The potential release of glutamate by GABAergic neurons is particularly surprising because these transmitters are usually considered to have opposing functions in, respectively, excitation and inhibition. In addition to its expression by cells not traditionally considered glutamatergic, VGLUT3 localizes to perikarya and dendrites as well as the nerve terminal, suggesting a nonclassical role in retrograde synaptic signaling. In contrast, VGLUT1 and VGLUT2 appear to be essentially restricted to the nerve terminal. VGLUT3 thus differs dramatically in its membrane trafficking from VGLUT1 and VGLUT2.

Supplementary Material

Supporting Figures:


We thank D. Fortin for her assistance with the confocal microscopy, the members of the Edwards laboratory for helpful discussions, and the Norwegian Research Council, the European Union (to F.A.C., J.S.-M., and T.Q.), the Parkinson's Disease Foundation (to D.S.), and the National Institutes of Health (to R.T.F., J.J., R.J.R., D.R.C., and R.H.E.) for their support.


  • VGLUT, vesicular glutamate transporter
  • GAD, glutamic acid decarboxylase
  • GABA, γ-aminobutyric acid
  • ACh, acetylcholine


1. Augustin I., Rosenmund, C., Sudhof, T. C. & Brose, N. (1999) Nature 400, 457-461. [PubMed]
2. Kandel E. M. (2001) Science 294, 1030-1038. [PubMed]
3. Svenneby G., Kvamme, E., Schousboe, A. & Torgner, I. A. (1987) Neurochem. Int. 10, 79-82. [PubMed]
4. Kaneko T., Kang, Y. & Mizuno, N. (1995) J. Neurosci. 15, 8362-8377. [PubMed]
5. Kvamme E., Torgner, I. A. & Roberg, B. (2001) J. Neurosci. Res. 66, 951-958. [PubMed]
6. Curthoys N. P. (2001) J. Nutr. 131, 2491-2495. [PubMed]
7. Rothstein J. D., Martin, L., Levey, A. I., Dykes-Hoberg, M., Jin, L., Wu, D., Nash, N. & Kuncl, R. W. (1994) Neuron 13, 713-725. [PubMed]
8. Chaudhry F. A., Lehre, K. P., vanLookeren Campagne, M., Otterson, O. P., Danbolt, N. C. & Storm-Mathisen, J. (1995) Neuron 15, 711-720. [PubMed]
9. Bellocchio E. E., Reimer, R. J., Fremeau, R. T. J. & Edwards, R. H. (2000) Science 289, 957-960. [PubMed]
10. Takamori S., Rhee, J. S., Rosenmund, C. & Jahn, R. (2000) Nature 407, 189-194. [PubMed]
11. Fremeau R. T., Troyer, M. D., Pahner, I., Nygaard, G. O., Tran, C. H., Reimer, R. J., Bellocchio, E. E., Fortin, D., Storm-Mathisen, J. & Edwards, R. H. (2001) Neuron 31, 247-260. [PubMed]
12. Herzog E., Bellenchi, G. C., Gras, C., Bernard, V., Ravassard, P., Bedet, C., Gasnier, B., Giros, B. & El Mestikaway, S. (2001) J. Neurosci. 21, RC181. [PubMed]
13. Varoqui H., Schafer, M. K.-H., Zhu, H., Weihe, E. & Erickson, J. D. (2002) J. Neurosci. 22, 142-155. [PubMed]
14. Kaneko T., Fujiyama, F. & Hioki, H. (2002) J. Comp. Neurol. 444, 39-62. [PubMed]
15. Takamori S., Rhee, J. S., Rosenmund, C. & Jahn, R. (2001) J. Neurosci. 21, RC182. [PubMed]
16. Johnson M. D. (1994) Neuron 12, 443-442. [PubMed]
17. Sulzer D., Joyce, M. P., Lin, L., Geldwert, D., Haber, S. N., Hattori, T. & Rayport, S. (1998) J. Neurosci. 18, 4588-4602. [PubMed]
18. Bezzi P., Carmignoto, G., Pasti, L., Vesce, S., Rossi, D., Rizzini, B. L., Pozzan, T. & Volterra, A. (1998) Nature 391, 281-285. [PubMed]
19. Newman E. A. & Zahs, K. R. (1998) J. Neurosci. 18, 4022-4028. [PMC free article] [PubMed]
20. Araque A., Li, N., Doyle, R. T. & Haydon, P. G. (2000) J. Neurosci. 20, 666-673. [PubMed]
21. Chaudhry F. A., Reimer, R. J., Krizaj, D., Barber, D., Storm-Mathisen, J., Copenhagen, D. R. & Edwards, R. H. (1999) Cell 99, 769-780. [PubMed]
22. Clift-O'Grady L., Linstedt, A. D., Lowe, A. W., Grote, E. & Kelly, R. B. (1990) J. Cell Biol. 110, 1693-1703. [PMC free article] [PubMed]
23. Wolosker H., de Souza, D. O. & de Meis, L. (1996) J. Biol. Chem. 271, 11726-11731. [PubMed]
24. Bai L., Xu, H., Collins, J. F. & Ghishan, F. K. (2001) J. Biol. Chem. 276, 36764-36769. [PubMed]
25. Gras C., Herzog, E., Bellenchi, G. C., Bernard, V., Ravassard, P., Pohl, M., Gasnier, B., Giros, B. & El Mestikawy, S. (2002) J. Neurosci. 22, 5442-5451. [PubMed]
26. Takamori S., Malherbe, P., Broger, C. & Jahn, R. (2002) EMBO Rep. 3, 798-803. [PMC free article] [PubMed]
27. Maycox P. R., Deckwerth, T., Hell, J. W. & Jahn, R. (1988) J. Biol. Chem. 263, 15423-15428. [PubMed]
28. Carlson M. D., Kish, P. E. & Ueda, T. (1989a) J. Biol. Chem. 264, 7369-7376. [PubMed]
29. Chaudhry F. A., Reimer, R. J., Bellocchio, E. E., Danbolt, N. C., Osen, K. K., Edwards, R. H. & Storm-Mathisen, J. (1998) J. Neurosci. 18, 9733-9750. [PubMed]
30. Torp R., Hoover, F., Danbolt, N. C., Storm-Mathisen, J. & Ottersen, O. P. (1997) Anat. Embryol. 195, 317-326. [PubMed]
31. Vardi N. & Auerbach, P. (1995) J. Comp. Neurol. 351, 374-384. [PubMed]
32. O'Malley D. M., Sandell, J. H. & Masland, R. H. (1992) J. Neurosci. 12, 1394-1408. [PubMed]
33. Casini G. & Brecha, N. C. (1992) J. Comp. Neurol. 326, 283-301. [PubMed]
34. Kawaguchi Y., Wilson, C. J., Augood, S. J. & Emson, P. C. (1995) Trends Neurosci. 18, 527-535. [PubMed]
35. Hattori T., Takada, M., Moriizumi, T. & van der Kooy, D. (1991) J. Comp. Neurol. 309, 391-401. [PubMed]
36. Soriano E. & Frotscher, M. (1993) J. Comp. Neurol. 15, 435-448. [PubMed]
37. Isaacson J. S. & Strowbridge, B. W. (1998) Neuron 20, 749-761. [PubMed]
38. Didier A., Carleton, A., Bjaalie, J. G., Vincent, J.-D., Ottersen, O. P., Storm-Mathisen, J. & Lledo, P.-M. (2001) Proc. Natl. Acad. Sci. USA 98, 6441-6446. [PMC free article] [PubMed]
39. Zilberter Y. (2000) J. Physiol. (London) 528, 489-496. [PMC free article] [PubMed]
40. Bankston L. A. & Guidotti, G. (1996) J. Biol. Chem. 271, 17132-17138. [PubMed]
41. Haussinger D. (1990) Biochem. J. 267, 281-290. [PMC free article] [PubMed]
42. Kanai Y. & Hediger, M. A. (1992) Nature 360, 467-471. [PubMed]
43. Nicola S. M., Surmeier, J. & Malenka, R. C. (2000) Annu. Rev. Neurosci. 23, 185-215. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...