- Email: [email protected]
Evidence for a satellite secretory pathway in neuronal dendritic spines
Brief Communication 351
Evidence for a satellite secretory pathway in neuronal dendritic spines Joseph P. Pierce*, Thomas Mayer† and J. Brian McCarthy* Long-term information storage within the brain requires the synthesis of new proteins and their use in synapse-specific modifications [1]. Recently, we demonstrated that translation sites for the local synthesis of integral membrane and secretory proteins occur within distal dendritic spines [2]. It remains unresolved, however, whether a complete secretory pathway, including Golgi and trans Golgi network-like membranes, exists near synapses for the local transport and processing of newly synthesized proteins. Here, we report evidence of a satellite secretory pathway in distal dendritic spines and distal dendrites of the mammalian brain. Membranes analogous to early (RER and ERGIC), middle (Golgi cisternae), and late (TGN) secretory pathway compartments are present within dendritic spines and in distal dendrites. Local synthesis, processing, and transport of newly translated integral membrane and secretory proteins may thus provide the molecular basis for synapsespecific modifications during long-term information storage in the brain. Addresses: * Division of Neurobiology, Department of Neurology and Neuroscience, Weill Medical College of Cornell University, 411 East 69th Street, New York, New York 10021, USA. † Cellular Biochemistry and Biophysics Program, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021, USA. Correspondence: J. Brian McCarthy E-mail: [email protected] Received: 18 December 2000 Revised: 24 January 2001 Accepted: 24 January 2001 Published: 6 March 2001 Current Biology 2001, 11:351–355 0960-9822/01/$ – see front matter 2001 Elsevier Science Ltd. All rights reserved.
Results and discussion Local protein synthesis in the dendrites of neurons versus the cell body has been proposed as a mechanism to produce proteins for local synapse-specific modifications in cases of long-lasting plasticity [1, 3]. Long-lasting synaptic plasticity requires the delivery of integral membrane and/ or secretory proteins to the postsynaptic membrane [4], and may require a local protein synthetic secretory pathway. Cytosolic, secretory, and integral membrane proteinencoding mRNAs [5], ribosomes [6], and translation/trans-
location sites [2] have been reported within vertebrate dendrites. It remains unresolved, however, whether posttranslational processing and transport of integral membrane and secretory proteins are possible in dendrites. Integral membrane proteins (such as receptors) and secretory proteins are processed in a series of subcellular membrane-enclosed compartments. Unlike proteins destined for the cytoplasm, which are translated on polyribosomes in the cytoplasm and released as soluble proteins, proteins destined for secretion or incorporation into the plasma membrane are translated by membrane-bound ribosomes on the rough endoplasmic reticulum (RER) [7]. Following synthesis at the RER, proteins move along the secretory pathway by vesicular transport through a series of compartments including the ERGIC (endoplasmic reticulumto-Golgi intermediate compartment), the cisternae (cis, medial, and trans) of the Golgi apparatus, and finally the trans Golgi network (TGN) en route to the cell surface [8]. We examined the hippocampal formation of adult rats, a brain region involved in explicit memory and widely analyzed in studies of synaptic plasticity [9]. To identify dendritic secretory pathway compartments, we analyzed the ultrastructural labeling patterns of a set of well-characterized protein markers for the secretory pathway (Figure 1) in cell body layers, and in distal dendrites within the stratum lacunosum-moleculare of hippocampal CA1 and the outer molecular layer of the dentate gyrus (see figure legends for methods). In distal dendrites, all secretory pathway markers labeled subcellular membrane-enclosed compartments, including the spine apparatus (Figure 1). Determination of membrane labeling densities (Figure 2a) revealed distinct differences in their patterns of labeling on subcellular compartments within dendrites. Giantin, ␣-mannosidase II, and Rab1b demonstrated the highest densities of labeling on compartments within spine heads. In cell bodies, these proteins are functional components of and markers for specific cisternae of the Golgi complex. Rab1b is required for the targeting/fusion of ER-derived vesicles to the cis Golgi cisternae and for transport between the cis and medial Golgi cisternae [10, 11]. The processing of newly synthesized membrane and secretory glycoproteins involves ␣-mannosidase II in the medial and trans cisternae of the Golgi complex [12, 13]. Giantin, a membrane-tethering component of the Golgi complex, serves in vesicular transport between Golgi cisternae as well as in cisternae formation [14, 15]. High densities of labeling for ERGIC53/58, TGN38, and Rab1b were found on compart-
352 Current Biology Vol 11 No 5
Figure 1 Evidence for a secretory pathway in distal dendritic spines and dendrites. (a) Illustration of the known distribution of secretory pathway markers for the RER, ERGIC, Golgi, and TGN. These proteins serve in the intracellular transport, processing, and sorting of proteins within these post-RER secretory pathway compartments (see text). Compartments within neuronal cell bodies of the hippocampal formation that label for TGN38 (b), ERGIC53/58 (c), ␣-mannosidase II (d), and giantin (e). (f) ERGIC53/58 immunogold particle (arrowhead) labeling a large diameter compartment within a dendrite. (g) Rab1b immunogold particles (arrowheads) labeling a compartment within a spine head, and in the cytoplasm. (h) ␣-mannosidase II immunogold particle (arrowhead) labeling the spine apparatus. (i) Giantin immunogold particles (arrowheads) labeling the spine apparatus. (j) Three Rab6 immunogold particles (arrowheads) on a large diameter compartment within a dendrite. (k) A TGN38 immunogold particle (arrowhead) labeling a complex subcellular compartment within a dendrite. Immunogold particle labeling of a compartment within a spine for ␣-mannosidase II (l), and of a compartment at the base of a spine for Rab6 (m). Acrolein/paraformaldehyde-perfused tissue [21] from three adult male SpragueDawley rats was examined using an osmium/ ferrocyanide fixation method [2, 22] and preembedding silver-enhanced immunogold electron microscopy [21], for precise subcellular localization and optimal subcellular membrane preservation. Antibodies against ERGIC53/58 (Sigma), Rab1b and Rab6 [23], giantin (gift of Dr. H.P. Hauri, Basel, Switzerland), ␣-mannosidase II (gift of Dr. B. Burke, Calgary, AB), and TGN38 (Transduction Lab, Lexington, KY) were used at 1:100. D, dendrite; T, presynaptic terminal; S, spine. All methods were approved by the Institutional Animal Care and Use Committee. The (a–e) scale bar represents 0.5 m. For (f–m), the scale bar represents 0.25 m.
ments within spine heads and on large diameter (⬎100 nm) compartments within dendrites. ERGIC53/58 cycles between the ER, ERGIC, and cis Golgi cisternae, where it functions as a sorting receptor for proteins in the early secretory pathway and is required for ER-to-Golgi vesicular transport of newly synthesized proteins [16, 17], while TGN38 serves in the sorting of proteins into vesicles and mediates the localization of proteins to the TGN [18]. Rab6, which functions in the targeting and fusion of vesi-
cles from the medial Golgi through the TGN [19, 20], densely labeled large diameter compartments within dendrites. Additionally, immunogold particles directed against TGN38 and ␣-mannosidase II labeled the dendritic plasma membrane most densely. This is consistent with TGN38 cycling between the TGN and the cell surface [18], and small quantities of ␣-mannosidase II at the cell surface [13]. The distributions of these proteins imply that membrane-enclosed compartments within spines can
Brief Communication 353
Figure 2
Subcellular labeling patterns of secretory pathway proteins in distal dendritic spines and dendrites. (a) Subcellular membrane labeling densities within dendritic spines and dendrites (immunogold particles/ 1000 m) subdivided by membrane type. Quantitation of preembedding immunogold labeling in both dendritic and cell body regions was done as previously described [2, 24, 25]. All silver-enhanced immunogold particles were counted in randomly selected fields (dendritic regions: 12,100 m2 per antibody; cell body regions: 4200 m2 per antibody) along the plastic/tissue interface of thin sections (two per antibody). Subcellular membrane labeling density was defined as the total particle count per membrane type divided by the product of the area examined
and SV, the surface density of the membrane type, as measured from random micrographs with the NIH Image program. The number of gold particles examined per antibody was as follows: ERGIC53/58, 348; Rab1b, 229; ␣-mann II, 222; giantin, 414; Rab6, 363; and TGN38, 315. (b) Consistency of dendritic subcellular membrane labeling with known distributions. The dendrogram illustrates the results of the hierarchical cluster analysis of subcellular membrane labeling densities in dendrites for secretory pathway markers; the clusters were formed using the average linkage method (between groups).
be involved in posttranslational protein processing and transport, while large diameter compartments within dendrites may serve in local vesicular targeting and the sorting of transport vesicles. Comparable labeling densities were observed on somal subcellular membranes (tubulovesicular versus stacked structures: giantin, 38 versus 62 particles/1000 m; ␣-mannosidase II, 38 versus 44; Rab1b, 48 versus 30; ERGIC53/58, 38 versus 37; TGN38, 48 versus 0; Rab6, 40 versus 27), where it is known that these antigens are concentrated.
that the labeling patterns of dendritic membranes covaried in a manner analogous to their known distributions in the cell body secretory system. The resulting dendrogram disclosed three main groups (Figure 2b). Consistent with Rab1b, giantin, and ERGIC53/58 coexisting on cisternae of the Golgi complex, the distributions of these antigens in distal spines and dendrites clustered early. TGN38 and ␣-mannosidase II clustered at a comparably early point, while Rab6 constituted a third group. Primary distinguishing criteria for this pattern of clustering (as determined using k-means cluster analysis and assuming three groups) were the labeling densities on large diameter subcellular
Hierarchical cluster analysis provided additional evidence
354 Current Biology Vol 11 No 5
Figure 3
Acknowledgements We wish to thank Drs. Teresa A. Milner, Martin Wiedmann, and Michael J. Caplan for their support and assistance.
References
The proposed model of the distribution of protein biosynthetic compartments based upon the most prominent labeling densities for marker proteins within distal dendritic spines and dendrites. Early secretory pathway compartments (RER and ERGIC) are found throughout dendritic spines and within dendrites; middle compartments (Golgi cisternae) are predominately found in spine heads, while late compartments (later Golgi and TGN) are predominately found in dendrites. PSD, postsynaptic density.
compartments and the plasma membrane, factors that, as noted, are important in distinguishing Rab6, TGN38, and ␣-mannosidase II labeling. Active protein synthesis is required for long-term memory and long-lasting modifications of synaptic strength. Individual (or small groups of) synapses on the same neuron are selectively modified during the long-lasting phases of synaptic plasticity. This activity-dependent modification of select synapses is widely believed to underlie longterm information storage in the brain [1]. Protein synthesis and processing in dendrites has been proposed to generate proteins for rapid synapse-specific modification in cases of long-lasting plasticity [1, 3, 5]. Our findings imply the existence of a satellite membrane system near synapses, analogous to the somatic RER-ERGIC-Golgi-TGN (Figure 3). Vesicular transport (and thus protein processing) could occur quickly within this system, because of the short distances between compartments. A linear path (toward the synapse) may therefore not be essential. This satellite system in adult dendritic spines may only be needed under specific circumstances, such as synapse modification in response synaptic activity. Thus, a satellite secretory system could provide rapid and spatially localized delivery of proteins to specific synapses, and could be the molecular basis for long-lasting synaptic plasticity.
1. Schuman EM: Synapse specificity and long-term information storage. Neuron 1997, 18:339-342. 2. Pierce JP, van Leyen K, McCarthy JB: Translocation machinery for synthesis of integral membrane and secretory proteins in dendritic spines. Nat Neurosci 2000, 3:311-313. 3. Wells DG, Richter JD, Fallon JR: Molecular mechanisms for activity-regulated protein synthesis in the synapto-dendritic compartment. Curr Opin Neurobiol 2000, 10:132-137. 4. Lledo PM, Zhang X, Sudhof TC, Malenka RC, Nicoll RA: Postsynaptic membrane fusion and long-term synaptic plasticity. Science 1998, 279:399-403. 5. Kiebler MA, DesGroseillers L: Molecular insights into mRNA transport and local translation in the mammalian nervous system. Neuron 2000, 25:19-28. 6. Bodian D: A suggestive relationship of nerve cell RNA with specific synaptic sites. Proc Natl Acad Sci USA 1965, 53: 418-425. 7. Blobel G, Walter P, Chang CN, Goldman BM, Erickson AH, Lingappa VR: Translocation of proteins across membranes: the signal hypothesis and beyond. Symp Soc Exp Biol 1979, 33:9-36. 8. Pfeffer SR, Rothman JE: Biosynthetic protein transport and sorting by the endoplasmic reticulum and Golgi. Annu Rev Biochem 1987, 56:829-852. 9. Bailey CH, Bartsch D, Kandel ER: Toward a molecular definition of long-term memory storage. Proc Natl Acad Sci USA 1996, 93:13445-13452. 10. Nuoffer C, Davidson HW, Matteson J, Meinkoth J, Balch WE: A GDPbound of rab1 inhibits protein export from the endoplasmic reticulum and transport between Golgi compartments. J Cell Biol 1994, 125:225-237. 11. Griffiths G, Ericsson M, Krijnse-Locker J, Nilsson T, Goud B, Soling HD, et al.: Localization of the Lys, Asp, Glu, Leu tetrapeptide receptor to the Golgi complex and the intermediate compartment in mammalian cells. J Cell Biol 1994, 127:15571574. 12. Herscovics A: Importance of glycosidases in mammalian glycoprotein biosynthesis. Biochim Biophys Acta 1999, 1473:96-107. 13. Velasco A, Hendricks,L, Moremen KW, Tulsiani DR, Touster O, Farquhar MG: Cell type-dependent variations in the subcellular distribution of ␣-mannosidase I and II. J Cell Biol 1993, 122:39-51. 14. Linstedt AD, Jesch SA, Mehta A, Lee TH, Garcia-Mata R, Nelson DS, et al.: Binding relationships of membrane tethering components. The giantin N terminus and the GM130 N terminus compete for binding to the p115 C terminus. J Biol Chem 2000, 275:10196-10201. 15. Seemann J, Jokitalo EJ, Warren G: The role of the tethering proteins p115 and GM130 in transport through the Golgi apparatus in vivo. Mol Biol Cell 2000, 11:635-645. 16. Tisdale EJ, Plutner H, Matteson J, Balch WE: p53/58 binds COPI and is required for selective transport through the early secretory pathway. J Cell Biol 1997, 137:581-593. 17. Hauri HP, Kappeler F, Andersson H, Appenzeller C: ERGIC-53 and traffic in the secretory pathway. J Cell Sci 2000, 113: 587-596. 18. Banting G, Ponnambalam S: TGN38 and its orthologues: roles in post-TGN vesicle formation and maintenance of TGN morphology. Biochim Biophys Acta 1997, 1355:209-217. 19. Antony C, Cibert C, Geraud G, Santa Maria A, Maro B, Mayau V, et al.: The small GTP-binding protein rab6p is distributed from medial Golgi to the trans-Golgi network as determined by a confocal microscopic approach. J Cell Sci 1992, 103: 785-796. 20. Mayer T, Touchot N, Elazar Z: Transport between cis and medial Golgi cisternae requires the function of the Ras-related protein Rab6. J Biol Chem 1996, 271:16097-16103. 21. Chan J, Aoki C, Pickel VM: Optimization of differential immunogold-silver and peroxidase labeling with maintenance of ultrastructure in brain sections before plastic embedding. J Neurosci Methods 1990, 33:113-127. 22. Spacek J, Harris KM: Three-dimensional organization of
Brief Communication 355
smooth endoplasmic reticulum in hippocampal CA1 dendrites and dendritic spines of the immature and mature rat. J Neurosci 1997, 17:190-203. 23. Elazar Z, Mayer T, Rothman JE: Removal of Rab GTP-binding proteins from Golgi membranes by GDP dissociation inhibitor inhibits inter-cisternal transport in the Golgi stacks. J Biol Chem 1994, 269:794-797. 24. Rodriguez JJ, Pickel VM: Enhancement of N-methyl-D-aspartate (NMDA) immunoreactivity in residual dendritic spines in the caudate-putamen nucleus after chronic haloperidol administration. Synapse 1999, 33:289-303. 25. Pickel VM, Chan J, Sesack SR: Cellular basis for interactions between catecholaminergic afferents and neurons containing Leu-enkephalin-like immunoreactivity in rat caudate-putamen nuclei. J Neurosci Res 1992, 31:212-230.
Evidence for a satellite secretory pathway in neuronal dendritic spines Joseph P. Pierce*, Thomas Mayer† and J. Brian McCarthy* Long-term information storage within the brain requires the synthesis of new proteins and their use in synapse-specific modifications [1]. Recently, we demonstrated that translation sites for the local synthesis of integral membrane and secretory proteins occur within distal dendritic spines [2]. It remains unresolved, however, whether a complete secretory pathway, including Golgi and trans Golgi network-like membranes, exists near synapses for the local transport and processing of newly synthesized proteins. Here, we report evidence of a satellite secretory pathway in distal dendritic spines and distal dendrites of the mammalian brain. Membranes analogous to early (RER and ERGIC), middle (Golgi cisternae), and late (TGN) secretory pathway compartments are present within dendritic spines and in distal dendrites. Local synthesis, processing, and transport of newly translated integral membrane and secretory proteins may thus provide the molecular basis for synapsespecific modifications during long-term information storage in the brain. Addresses: * Division of Neurobiology, Department of Neurology and Neuroscience, Weill Medical College of Cornell University, 411 East 69th Street, New York, New York 10021, USA. † Cellular Biochemistry and Biophysics Program, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021, USA. Correspondence: J. Brian McCarthy E-mail: [email protected] Received: 18 December 2000 Revised: 24 January 2001 Accepted: 24 January 2001 Published: 6 March 2001 Current Biology 2001, 11:351–355 0960-9822/01/$ – see front matter 2001 Elsevier Science Ltd. All rights reserved.
Results and discussion Local protein synthesis in the dendrites of neurons versus the cell body has been proposed as a mechanism to produce proteins for local synapse-specific modifications in cases of long-lasting plasticity [1, 3]. Long-lasting synaptic plasticity requires the delivery of integral membrane and/ or secretory proteins to the postsynaptic membrane [4], and may require a local protein synthetic secretory pathway. Cytosolic, secretory, and integral membrane proteinencoding mRNAs [5], ribosomes [6], and translation/trans-
location sites [2] have been reported within vertebrate dendrites. It remains unresolved, however, whether posttranslational processing and transport of integral membrane and secretory proteins are possible in dendrites. Integral membrane proteins (such as receptors) and secretory proteins are processed in a series of subcellular membrane-enclosed compartments. Unlike proteins destined for the cytoplasm, which are translated on polyribosomes in the cytoplasm and released as soluble proteins, proteins destined for secretion or incorporation into the plasma membrane are translated by membrane-bound ribosomes on the rough endoplasmic reticulum (RER) [7]. Following synthesis at the RER, proteins move along the secretory pathway by vesicular transport through a series of compartments including the ERGIC (endoplasmic reticulumto-Golgi intermediate compartment), the cisternae (cis, medial, and trans) of the Golgi apparatus, and finally the trans Golgi network (TGN) en route to the cell surface [8]. We examined the hippocampal formation of adult rats, a brain region involved in explicit memory and widely analyzed in studies of synaptic plasticity [9]. To identify dendritic secretory pathway compartments, we analyzed the ultrastructural labeling patterns of a set of well-characterized protein markers for the secretory pathway (Figure 1) in cell body layers, and in distal dendrites within the stratum lacunosum-moleculare of hippocampal CA1 and the outer molecular layer of the dentate gyrus (see figure legends for methods). In distal dendrites, all secretory pathway markers labeled subcellular membrane-enclosed compartments, including the spine apparatus (Figure 1). Determination of membrane labeling densities (Figure 2a) revealed distinct differences in their patterns of labeling on subcellular compartments within dendrites. Giantin, ␣-mannosidase II, and Rab1b demonstrated the highest densities of labeling on compartments within spine heads. In cell bodies, these proteins are functional components of and markers for specific cisternae of the Golgi complex. Rab1b is required for the targeting/fusion of ER-derived vesicles to the cis Golgi cisternae and for transport between the cis and medial Golgi cisternae [10, 11]. The processing of newly synthesized membrane and secretory glycoproteins involves ␣-mannosidase II in the medial and trans cisternae of the Golgi complex [12, 13]. Giantin, a membrane-tethering component of the Golgi complex, serves in vesicular transport between Golgi cisternae as well as in cisternae formation [14, 15]. High densities of labeling for ERGIC53/58, TGN38, and Rab1b were found on compart-
352 Current Biology Vol 11 No 5
Figure 1 Evidence for a secretory pathway in distal dendritic spines and dendrites. (a) Illustration of the known distribution of secretory pathway markers for the RER, ERGIC, Golgi, and TGN. These proteins serve in the intracellular transport, processing, and sorting of proteins within these post-RER secretory pathway compartments (see text). Compartments within neuronal cell bodies of the hippocampal formation that label for TGN38 (b), ERGIC53/58 (c), ␣-mannosidase II (d), and giantin (e). (f) ERGIC53/58 immunogold particle (arrowhead) labeling a large diameter compartment within a dendrite. (g) Rab1b immunogold particles (arrowheads) labeling a compartment within a spine head, and in the cytoplasm. (h) ␣-mannosidase II immunogold particle (arrowhead) labeling the spine apparatus. (i) Giantin immunogold particles (arrowheads) labeling the spine apparatus. (j) Three Rab6 immunogold particles (arrowheads) on a large diameter compartment within a dendrite. (k) A TGN38 immunogold particle (arrowhead) labeling a complex subcellular compartment within a dendrite. Immunogold particle labeling of a compartment within a spine for ␣-mannosidase II (l), and of a compartment at the base of a spine for Rab6 (m). Acrolein/paraformaldehyde-perfused tissue [21] from three adult male SpragueDawley rats was examined using an osmium/ ferrocyanide fixation method [2, 22] and preembedding silver-enhanced immunogold electron microscopy [21], for precise subcellular localization and optimal subcellular membrane preservation. Antibodies against ERGIC53/58 (Sigma), Rab1b and Rab6 [23], giantin (gift of Dr. H.P. Hauri, Basel, Switzerland), ␣-mannosidase II (gift of Dr. B. Burke, Calgary, AB), and TGN38 (Transduction Lab, Lexington, KY) were used at 1:100. D, dendrite; T, presynaptic terminal; S, spine. All methods were approved by the Institutional Animal Care and Use Committee. The (a–e) scale bar represents 0.5 m. For (f–m), the scale bar represents 0.25 m.
ments within spine heads and on large diameter (⬎100 nm) compartments within dendrites. ERGIC53/58 cycles between the ER, ERGIC, and cis Golgi cisternae, where it functions as a sorting receptor for proteins in the early secretory pathway and is required for ER-to-Golgi vesicular transport of newly synthesized proteins [16, 17], while TGN38 serves in the sorting of proteins into vesicles and mediates the localization of proteins to the TGN [18]. Rab6, which functions in the targeting and fusion of vesi-
cles from the medial Golgi through the TGN [19, 20], densely labeled large diameter compartments within dendrites. Additionally, immunogold particles directed against TGN38 and ␣-mannosidase II labeled the dendritic plasma membrane most densely. This is consistent with TGN38 cycling between the TGN and the cell surface [18], and small quantities of ␣-mannosidase II at the cell surface [13]. The distributions of these proteins imply that membrane-enclosed compartments within spines can
Brief Communication 353
Figure 2
Subcellular labeling patterns of secretory pathway proteins in distal dendritic spines and dendrites. (a) Subcellular membrane labeling densities within dendritic spines and dendrites (immunogold particles/ 1000 m) subdivided by membrane type. Quantitation of preembedding immunogold labeling in both dendritic and cell body regions was done as previously described [2, 24, 25]. All silver-enhanced immunogold particles were counted in randomly selected fields (dendritic regions: 12,100 m2 per antibody; cell body regions: 4200 m2 per antibody) along the plastic/tissue interface of thin sections (two per antibody). Subcellular membrane labeling density was defined as the total particle count per membrane type divided by the product of the area examined
and SV, the surface density of the membrane type, as measured from random micrographs with the NIH Image program. The number of gold particles examined per antibody was as follows: ERGIC53/58, 348; Rab1b, 229; ␣-mann II, 222; giantin, 414; Rab6, 363; and TGN38, 315. (b) Consistency of dendritic subcellular membrane labeling with known distributions. The dendrogram illustrates the results of the hierarchical cluster analysis of subcellular membrane labeling densities in dendrites for secretory pathway markers; the clusters were formed using the average linkage method (between groups).
be involved in posttranslational protein processing and transport, while large diameter compartments within dendrites may serve in local vesicular targeting and the sorting of transport vesicles. Comparable labeling densities were observed on somal subcellular membranes (tubulovesicular versus stacked structures: giantin, 38 versus 62 particles/1000 m; ␣-mannosidase II, 38 versus 44; Rab1b, 48 versus 30; ERGIC53/58, 38 versus 37; TGN38, 48 versus 0; Rab6, 40 versus 27), where it is known that these antigens are concentrated.
that the labeling patterns of dendritic membranes covaried in a manner analogous to their known distributions in the cell body secretory system. The resulting dendrogram disclosed three main groups (Figure 2b). Consistent with Rab1b, giantin, and ERGIC53/58 coexisting on cisternae of the Golgi complex, the distributions of these antigens in distal spines and dendrites clustered early. TGN38 and ␣-mannosidase II clustered at a comparably early point, while Rab6 constituted a third group. Primary distinguishing criteria for this pattern of clustering (as determined using k-means cluster analysis and assuming three groups) were the labeling densities on large diameter subcellular
Hierarchical cluster analysis provided additional evidence
354 Current Biology Vol 11 No 5
Figure 3
Acknowledgements We wish to thank Drs. Teresa A. Milner, Martin Wiedmann, and Michael J. Caplan for their support and assistance.
References
The proposed model of the distribution of protein biosynthetic compartments based upon the most prominent labeling densities for marker proteins within distal dendritic spines and dendrites. Early secretory pathway compartments (RER and ERGIC) are found throughout dendritic spines and within dendrites; middle compartments (Golgi cisternae) are predominately found in spine heads, while late compartments (later Golgi and TGN) are predominately found in dendrites. PSD, postsynaptic density.
compartments and the plasma membrane, factors that, as noted, are important in distinguishing Rab6, TGN38, and ␣-mannosidase II labeling. Active protein synthesis is required for long-term memory and long-lasting modifications of synaptic strength. Individual (or small groups of) synapses on the same neuron are selectively modified during the long-lasting phases of synaptic plasticity. This activity-dependent modification of select synapses is widely believed to underlie longterm information storage in the brain [1]. Protein synthesis and processing in dendrites has been proposed to generate proteins for rapid synapse-specific modification in cases of long-lasting plasticity [1, 3, 5]. Our findings imply the existence of a satellite membrane system near synapses, analogous to the somatic RER-ERGIC-Golgi-TGN (Figure 3). Vesicular transport (and thus protein processing) could occur quickly within this system, because of the short distances between compartments. A linear path (toward the synapse) may therefore not be essential. This satellite system in adult dendritic spines may only be needed under specific circumstances, such as synapse modification in response synaptic activity. Thus, a satellite secretory system could provide rapid and spatially localized delivery of proteins to specific synapses, and could be the molecular basis for long-lasting synaptic plasticity.
1. Schuman EM: Synapse specificity and long-term information storage. Neuron 1997, 18:339-342. 2. Pierce JP, van Leyen K, McCarthy JB: Translocation machinery for synthesis of integral membrane and secretory proteins in dendritic spines. Nat Neurosci 2000, 3:311-313. 3. Wells DG, Richter JD, Fallon JR: Molecular mechanisms for activity-regulated protein synthesis in the synapto-dendritic compartment. Curr Opin Neurobiol 2000, 10:132-137. 4. Lledo PM, Zhang X, Sudhof TC, Malenka RC, Nicoll RA: Postsynaptic membrane fusion and long-term synaptic plasticity. Science 1998, 279:399-403. 5. Kiebler MA, DesGroseillers L: Molecular insights into mRNA transport and local translation in the mammalian nervous system. Neuron 2000, 25:19-28. 6. Bodian D: A suggestive relationship of nerve cell RNA with specific synaptic sites. Proc Natl Acad Sci USA 1965, 53: 418-425. 7. Blobel G, Walter P, Chang CN, Goldman BM, Erickson AH, Lingappa VR: Translocation of proteins across membranes: the signal hypothesis and beyond. Symp Soc Exp Biol 1979, 33:9-36. 8. Pfeffer SR, Rothman JE: Biosynthetic protein transport and sorting by the endoplasmic reticulum and Golgi. Annu Rev Biochem 1987, 56:829-852. 9. Bailey CH, Bartsch D, Kandel ER: Toward a molecular definition of long-term memory storage. Proc Natl Acad Sci USA 1996, 93:13445-13452. 10. Nuoffer C, Davidson HW, Matteson J, Meinkoth J, Balch WE: A GDPbound of rab1 inhibits protein export from the endoplasmic reticulum and transport between Golgi compartments. J Cell Biol 1994, 125:225-237. 11. Griffiths G, Ericsson M, Krijnse-Locker J, Nilsson T, Goud B, Soling HD, et al.: Localization of the Lys, Asp, Glu, Leu tetrapeptide receptor to the Golgi complex and the intermediate compartment in mammalian cells. J Cell Biol 1994, 127:15571574. 12. Herscovics A: Importance of glycosidases in mammalian glycoprotein biosynthesis. Biochim Biophys Acta 1999, 1473:96-107. 13. Velasco A, Hendricks,L, Moremen KW, Tulsiani DR, Touster O, Farquhar MG: Cell type-dependent variations in the subcellular distribution of ␣-mannosidase I and II. J Cell Biol 1993, 122:39-51. 14. Linstedt AD, Jesch SA, Mehta A, Lee TH, Garcia-Mata R, Nelson DS, et al.: Binding relationships of membrane tethering components. The giantin N terminus and the GM130 N terminus compete for binding to the p115 C terminus. J Biol Chem 2000, 275:10196-10201. 15. Seemann J, Jokitalo EJ, Warren G: The role of the tethering proteins p115 and GM130 in transport through the Golgi apparatus in vivo. Mol Biol Cell 2000, 11:635-645. 16. Tisdale EJ, Plutner H, Matteson J, Balch WE: p53/58 binds COPI and is required for selective transport through the early secretory pathway. J Cell Biol 1997, 137:581-593. 17. Hauri HP, Kappeler F, Andersson H, Appenzeller C: ERGIC-53 and traffic in the secretory pathway. J Cell Sci 2000, 113: 587-596. 18. Banting G, Ponnambalam S: TGN38 and its orthologues: roles in post-TGN vesicle formation and maintenance of TGN morphology. Biochim Biophys Acta 1997, 1355:209-217. 19. Antony C, Cibert C, Geraud G, Santa Maria A, Maro B, Mayau V, et al.: The small GTP-binding protein rab6p is distributed from medial Golgi to the trans-Golgi network as determined by a confocal microscopic approach. J Cell Sci 1992, 103: 785-796. 20. Mayer T, Touchot N, Elazar Z: Transport between cis and medial Golgi cisternae requires the function of the Ras-related protein Rab6. J Biol Chem 1996, 271:16097-16103. 21. Chan J, Aoki C, Pickel VM: Optimization of differential immunogold-silver and peroxidase labeling with maintenance of ultrastructure in brain sections before plastic embedding. J Neurosci Methods 1990, 33:113-127. 22. Spacek J, Harris KM: Three-dimensional organization of
Brief Communication 355
smooth endoplasmic reticulum in hippocampal CA1 dendrites and dendritic spines of the immature and mature rat. J Neurosci 1997, 17:190-203. 23. Elazar Z, Mayer T, Rothman JE: Removal of Rab GTP-binding proteins from Golgi membranes by GDP dissociation inhibitor inhibits inter-cisternal transport in the Golgi stacks. J Biol Chem 1994, 269:794-797. 24. Rodriguez JJ, Pickel VM: Enhancement of N-methyl-D-aspartate (NMDA) immunoreactivity in residual dendritic spines in the caudate-putamen nucleus after chronic haloperidol administration. Synapse 1999, 33:289-303. 25. Pickel VM, Chan J, Sesack SR: Cellular basis for interactions between catecholaminergic afferents and neurons containing Leu-enkephalin-like immunoreactivity in rat caudate-putamen nuclei. J Neurosci Res 1992, 31:212-230.