- Email: [email protected]
The endocytic machinery in nerve terminals surrounds sites of exocytosis
Brief Communication
1411
The endocytic machinery in nerve terminals surrounds sites of exocytosis Jack Roos and Regis B. Kelly In most models of endocytosis, the endocytic machinery is recruited from the cytoplasm by cytoplasmic tails of the plasma membrane proteins that are to be internalized. This does not appear to be true at synapses where the endocytic machinery required for synaptic vesicle recycling is localized to membraneassociated ‘hot spots’ [1,2]. In Drosophila neuromuscular junctions, the multi-domain protein Dap160 is also localized to hot spots [3] and has some characteristics expected of an anchoring protein. Anchoring the endocytic machinery to the plasma membrane might help contribute to the remarkable speed of synaptic vesicle recycling [4]. Here, we report that the endocytic machinery surrounds sites that are believed to be sites of exocytosis. We propose that the radial distribution of the synaptic vesicle recycling machinery already present on the plasma membrane in unstimulated nerve terminals is a fundamental unit of pre-synaptic organization and allows the nerve terminal to extract maximum recycling efficiency out of conventional endocytic machinery.
neuromuscular junctions using a deconvolution microscope. The images acquired were refined by applying deconvolution algorithms to reduce out-of-focus haze that is generally present in confocal sections. The images of single sections of a synaptic bouton revealed that the endocytic machinery had holes of almost constant diameter that give the terminal a ‘honeycomb’ appearance (Figure 1). Comparison to post-synaptic receptor clusters
Address: Department of Biochemistry and Biophysics and the Hormone Research Institute, University of California, San Francisco, San Francisco, California 94143-0534, USA.
The number and dimensions of the holes in the distribution of both Dap160 and dynamin (data not shown) were reminiscent of the size of active zones measured at Drosophila neuromuscular junctions [12]. They were also similar to the number and size of the clusters of postsynaptic glutamate receptors at the Drosophila neuromuscular junction. The distribution of the postsynaptic glutamate receptor, which is a postsynaptic marker for the presynaptic active zone, has been studied by both confocal and electron microscopy in Drosophila third instar larvae using a Myc-epitope-tagged form of the receptor [13]. The receptors cluster in oval discs about 1 µm in length on the post-synaptic membrane opposite synaptic vesicles docked at presynaptic active zones. To determine whether there is any relationship between the sites of endocytosis
Correspondence: Regis B. Kelly E-mail: [email protected]
Figure 1
Received: 12 August 1999 Revised: 21 October 1999 Accepted: 21 October 1999 Published: 22 November 1999 Current Biology 1999, 9:1411–1414 0960-9822/99/$ – see front matter © 1999 Elsevier Science Ltd. All rights reserved.
Results and discussion Concentration of the endocytic machinery into a honeycomb matrix
Dynamin [5,6] and AP-2 [2] are required for synaptic vesicle recycling in Drosophila. Both proteins were shown by confocal immunofluorescence to co-localize with Dap160 in third instar larval neuromuscular junctions (J.R. and R.B.K., unpublished observations). Dap160 was the first member of a family of proteins now called the intersectins [7–10] that have two Eps15 homology domains, multiple Src-homology 3 (SH3) groups and are involved in the formation of endocytic clathrin-coated vesicles [11]. We examined the localization of proteins involved in endocytosis in unstimulated nerve terminals of Drosophila
Current Biology
Dap160 is restricted to varicosities in a honeycomb pattern in Drosophila neuromuscular junctions. Wild-type third instar larvae were dissected and stained for Dap160 and visualized by DeltaVision microscopy. Images were then processed using deconvolution algorithms. Individual Z sections of synaptic boutons on muscle 4 are shown. Similar distributions have been seen for boutons present on muscle 6 and throughout the larval neuromuscular system. The scale bar represents 2.5 µm.
1412
Current Biology Vol 9 No 23
Figure 2
Figure 3 (a)
(a)
GluRB
GluRB
Dap160
Merge
Dap160 (b)
Merge Current Biology
(b)
(c)
GluRB Dap160
GluRB Dynamin Current Biology
Active zones are surrounded by an endocytic honeycomb in resting nerve terminals. MHC-DGluRIIB–Myc larvae were dissected and stained for post-synaptic glutamate receptors (GluRB; green staining) using an anti-Myc monoclonal antibody (9E10). The larvae were also stained with antibodies against either (a,b) Dap160 (red) or (c) dynamin (red). The scale bar represents 5 µm in (a) and 2.5 µm in (b,c).
and the post-synaptic clusters of receptors, the distribution of Dap160 was examined in strains expressing the Mycepitope-tagged glutamate receptor (DGluRIIB–Myc). The Myc-epitope-tagged glutamate receptors clustered into oval rings within each nerve terminal (Figure 2a). Dap160 and dynamin did not overlap with the receptor clusters but instead filled the membrane space surrounding them (Figure 2b,c). Previous analysis of the Drosophila neuromuscular junctions revealed that larger nerve terminals have more active zones per bouton than smaller terminals [14,15]. For boutons of all sizes, the glutamate receptor clusters were always about 1 µm in length and had a minimal separation of about 0.5 µm, which is consistent with previous measurements [12]. The dimensions of the endocytosis zones seen here in unstimulated nerve terminals are likely to be too large to correspond to single coated vesicles recycling synaptic vesicle proteins.. To estimate how a single, 50–100 nm clathrin-coated vesicle might appear, we compared the dimensions of the Dap160 honeycomb with that of
Dap160 and the endocytic honeycomb are coupled to glutamatereceptor clusters in vesicle-depleted nerve terminals. Male third instar larvae of shibirets1; DGluRIIB–Myc mutants were dissected and then incubated at 34°C in 60 mM K+-saline for 5 min. Preparations were fixed and stained for DGluRIIB–Myc (GluRB) using monoclonal antibody 9E10, and with an antibody against Dap160. (a) Individual confocal Z sections reveal that Dap160 is concentrated at unique sites on the plasma membrane that are adjacent to, but do not co-localize with, GluRB staining. (b) Confocal projections of individual Z sections in vesicle-depleted shibirets1; DGluRIIB–Myc male synapses after incubation at the non-permissive temperature. Although the plasma membrane contains additional membrane derived from synaptic vesicles, the endocytic honeycomb remains tightly tethered to clusters of glutamate receptors. The scale bar represents 5 µm.
fluorescent beads of known diameter by pixel counting. The width of the endocytic honeycomb was measured to be about 400 nm in diameter (data not shown), which is larger than the diameter of a clathrin-coated vesicle (about 100 nm). In addition, the concentrations of endocytic machinery observed were not radially symmetrical, as is seen for single clathrin-coated vesicles [16]. Effect of vesicle exocytosis on the endocytic machinery
In some models of endocytosis, the cargo proteins on the plasma membrane recruit the endocytic machinery to the plasma membrane (for reviews see [17,18]). The cargo for the nerve terminal endocytic machinery is the complement of synaptic vesicle proteins. Because there are very few cargo proteins on the plasma membrane in resting terminals, the honeycomb distribution is unlikely to be determined by the presence of cargo molecules. The synaptic vesicle cargo can be redistributed to the plasma membrane when neuromuscular junctions of shibirets1 larvae are stimulated [1,3,5,19,20], but synaptic vesicle uptake and recyling is blocked at the non-permissive temperature. To determine whether the organization of the recycling machinery is modified by redistributing cargo from synaptic vesicles to the plasma membrane, the distribution of Dap160 was examined by confocal microscopy in Drosophila shibirets1 strains possessing the DGluRIIB–Myc
Brief Communication
Figure 4
1413
Figure 5 (a)
Csp
(a)
Dap160
Glutamate receptor clusters
Endocytosis zone Synaptic Active vesicles zone
Wild type (b) (b)
shibire
Current Biology
Dap160 and cysteine string protein (csp) are both enriched in hot spots on Drosophila nerve terminal membranes. (a) Wild-type third instar larvae were dissected and stained with antibodies against cysteine string protein (csp) and Dap160 and visualized by confocal microscopy. (b) Third instar larvae from shibirets1 mutants were dissected, incubated at 34°C for 5 min to deplete the nerve terminal complement of synaptic vesicles, fixed and stained for cysteine string protein and Dap160 and visualized by confocal microscopy. The scale bar represents 4 µm in (a) and 2.5 µm in (b).
transgene. Dap160 was recovered in puncta or ‘hot spots’ in stimulated nerve terminals at the non-permissive temperature as previously described [1,3]. When the distribution of the Myc-epitope-tagged glutamate receptors was compared to the distribution of Dap160 on the nerve terminal membrane, again, no significant overlap was found, despite the massive redistribution of synaptic vesicles. This was true for individual sections (Figure 3a) or when several optical sections were combined (Figure 3b). Thus, it seems likely that, in nerve terminals, the endocytic machinery is clustered into zones independent of the location of the endocytic cargo. After transfer of the synaptic vesicle cargo to the plasma membrane by stimulating shibirets1 flies at the non-permissive temperature, the plasma membrane distribution of these synaptic vesicle antigens was not uniform. A significant fraction of the synaptic vesicle marker, cysteine string protein (csp), colocalized with the hot spots of Dap 160 (Figure 4). The colocalization would be expected if the concentrations of endocytic machinery are indeed sites of endocytosis. The data from Drosophila neuromuscular junctions suggest the concept of the recycling zone as a unit of presynaptic organization (Figure 5). The model is based on a number of reasonable assumptions. First, on the basis of published electron micrographs [13], that active zones of exocytosis lie opposite to post-synaptic receptor clusters and are so physically close that the receptor clusters can be taken as markers of active zones. Unfortunately, no presynaptic
Current Biology
A model of the organization of the recycling machinery in a Drosophila synaptic bouton. (a) Postsynaptic receptors (blue) cluster in apposition to active zones at which synaptic vesicles (red) are docked. Synaptic vesicles in the reserve pool, mitochondria and cytoplasmic elements are omitted. The spaces between active zones are filled with the endocytic machinery (green). (b) A three-dimensional image demonstrates that the proteins of each synaptic vesicle will not have far to migrate before encountering the surrounding endocytic machinery.
active zone marker is currently available in Drosophila. A second assumption is that sites enriched in endocytic machinery are also sites of endocytosis. The data from stimulated shibirets1 mutants are consistent with this assumption (Figure 4). A synaptic organization in which each exocytic active zone has a ‘recapture–surround’ zone of endocytic machinery has some satisfying features. It offers an explanation why, when nerve terminals have more than one active zone, as at the frog neuromuscular junction [21,22] or mossy fiber synapses in the hippocampus [23], the active zones are separated by a minimum distance of about 1 µm. The recapture–surround hypothesis also suggests an explanation of the remarkable constancy of active zone size, from Drosophila to mammals. Active zones might be limited to one micron or so to allow synaptic vesicle proteins to diffuse to the recycling machinery within a second or two after fusion, thus allowing rapid recycling. If all the calcium channels are in the active zone, the spacing we describe could also explain why, on the basis of calcium diffusion models, exocytosis occurs at calcium concentrations in the range of 100 µM whereas the endocytic machinery responds to calcium concentrations that are ten times smaller [4,24]. In conclusion, the large size and convenience of the Drosophila larval neuromuscular junction have revealed an
1414
Current Biology Vol 9 No 23
unexpected feature of nerve terminals. Regions that are predicted to be active zones of exocytosis seem to be embedded in zones of endocytosis. We propose that an active zone of exocytosis surrounded by an annulus of endocytic machinery be considered a recycling zone. Nerve terminals would grow by adding more recycling zones. The separation between active zones, measured to be at least 0.5 µm, might reflect a need for surrounding the active zone with sufficient endocytic machinery to ensure efficient recapture of synaptic vesicle proteins.
Materials and methods Flies Flies (Or-R, shibirets1, MHC-DGluRIIB–Myc) were raised on standard fly food and kept at 18°C. Third instar larval neuromuscular junctions from MHC-DGluRIIB–Myc flies were prepared as described [1]. Results of the complementary distribution of Dap160 and DGluRIIB–Myc were examined on over 25 individual boutons on muscles 6/7 in hemi-segments A2 and A3 and over 20 individual boutons on muscle 4, as well as other neuromuscular synapses throughout the larval preparation. These results were obtained from dissections of six different larval preparations. Fly shibirets1; MHC-DGluRIIB–Myc lines were generated by crossing shibirets1 females to MHC-DGluRIIB–Myc males. Six third instar male progeny were then dissected and analyzed at the non-permissive temperature as described [1]. The distributions of Dap160 and DGluRIIB–Myc were compared on muscles 6/7 and muscle 4 as described for unstimulated MHC-DGluRIIB–Myc flies.
Antibodies Dap160 polyclonal antibodies (Ab1704) [3] and dynamin polyclonal antibodies (Ab2074) [1] were affinity purified as described. Monoclonal antibody 9E10, which recognizes the Myc epitope was the generous gift of Dr Gary Herman (UCSF). Monoclonal antibody against the cysteine string protein was the generous gift of Konrad Zinsmaier (U Penn).
Microscopy Confocal images were acquired on a Leica NT system as described [3]. Optically acquired fluorescent images were generated using a DeltaVision workstation. Z sections were acquired at 0.1 µm increments. Images from individual boutons were processed by deconvolution algorithms using DeltaVision software.
Acknowledgements Thanks to Victor Faundez and Graeme Davis (UCSF) for helpful discussions throughout this work, to Gary Herman (UCSF) for anti-Myc (mAb 9E10) antibodies and to Konrad Zinsmaier (U Penn) for anti-csp antibodies, to Aaron DiAntonio and Corey Goodman (UC/Berkeley) for the MHCDGluRIIB–Myc transgenic flies, to Pat O’Farrell (UCSF) for use of his DeltaVision workstation and to John Sedat and Brigitte Hansen (UCSF) for the fluorescently conjugated beads used to determine the dimensions of the endocytic honeycomb. This work was supported by Postdoctoral Fellowship PF-4314 (JR) from the American Cancer Society and National Institutes of Health Grant NS15927 (RBK).
References 1. Estes PS, Roos J, van der Bliek A, Kelly RB, Krishnan KS, Ramaswami M: Traffic of dynamin within individual Drosophila synaptic boutons relative to compartment-specific markers. J Neurosci 1996, 16:5443-5456. 2. Gonzalez-Gaitan M, Jackle H: Role of Drosophila alpha-adaptin in presynaptic vesicle recycling. Cell 1997, 88:767-776. 3. Roos J, Kelly RB: Dap160, a neural-specific Eps15 homology and multiple SH3 domain-containing protein that interacts with Drosophila dynamin. J Biol Chem 1998, 273:19108-19119. 4. von Gersdorff H, Matthews G: Electrophysiology of synaptic vesicle cycling. Annu Rev Physiol 1999, 61:725-752.
5. Kosaka T, Ikeda K: Possible temperature-dependent blockage of synaptic vesicle recycling induced by a single gene mutation in Drosophila. J Neurobiol 1983, 14:207-225. 6. van der Bliek AM, Meyerowitz EM: Dynamin-like protein encoded by the Drosophila shibire gene associated with vesicular traffic. Nature 1991, 351:411-414. 7. Yamabhai M, Hoffman NG, Hardison NL, McPherson PS, Castagnoli L, Cesareni G, et al.: Intersectin, a novel adaptor protein with two Eps15 homology and five Src homology 3 domains. J Biol Chem 1998, 273:31401-31407. 8. Guipponi M, Scott HS, Chen H, Schebesta A, Rossier C, Antonarakis SE: Two isoforms of a human intersectin (ITSN) protein are produced by brain-specific alternative splicing in a stop codon. Genomics 1998, 53:369-376. 9. Sengar AS, Wang W, Bishay J, Cohen S, Egan SE: The EH and SH3 domain Ese proteins regulate endocytosis by linking to dynamin and Eps15. EMBO J 1999, 18:1159-1171. 10. Okamoto M, Schoch S, Sudhof TC: EHSH1/intersectin, a protein that contains EH and SH3 domains and binds to dynamin and SNAP-25. A protein connection between exocytosis and endocytosis. J Biol Chem 1999, 274:18446-18454. 11. Simpson F, Hussain NK, Qualmann B, Kelly RB, Kay BK, McPherson PS, et al.: SH3-domain-containing proteins function at distinct steps in clathrin-coated vesicle formation. Nat Cell Biol 1999, 1:119-124. 12. Meinertzhagen IA, Govind CK, Stewart BA, Carter JM, Atwood HL: Regulated spacing of synapses and presynaptic active zones at larval neuromuscular junctions in different genotypes of the flies Drosophila and Sarcophaga. J Comp Neurol 1998, 393:482-492. 13. Petersen SA, Fetter RD, Noordermeer JN, Goodman CS, DiAntonio A: Genetic analysis of glutamate receptors in Drosophila reveals a retrograde signal regulating presynaptic transmitter release. Neuron 1997, 19:1237-1248. 14. Atwood HL, Govind CK, Wu C-F: Differential ultrastructure of synaptic terminals on ventral longitudinal abdominal muscles in Drosophila larvae. J Neurobiol 1993, 24:1008-1024. 15. Stewart BA, Schuster CM, Goodman CS, Atwood HL: Homeostasis of synaptic transmission in Drosophila with genetically altered nerve terminal morphology. J Neurosci 1996, 16:3877-3886. 16. Gaidarov I, Santini F, Warren RA, Keen JH: Spatial control of coated pit dynamics in living cells. Nat Cell Biol 1999, 1:1-7. 17. Schmid SL: Clathrin-coated vesicle formation and protein sorting: an integrated process. Annu Rev Biochem 1997, 66:511-548. 18. Lewis DA, Mellman I: Sorting out adaptors. Biochim Biophys Acta 1998, 1401:129-145. 19. Ramaswami M, Krishnan KS, Kelly RB: Intermediates in synaptic vesicle recycling revealed by optical imaging of Drosophila neuromuscular junctions. Neuron 1994, 13:363-375. 20. van de Goor J, Ramaswami M, Kelly RB: Redistribution of synaptic vesicles and their proteins in temperature-sensitive shibirets1 mutant Drosophila. Proc Natl Acad Sci USA 1995, 92:5739-5743. 21. Heuser JE, Reese TS: Evidence for recycling of synaptic vesicle membrane during transmitter release at the frog neuromuscular junction. J Cell Biol 1973, 57:315-344. 22. Miller TM, Heuser JE: Endocytosis of synaptic vesicle membrane at the frog neuromuscular junction. J Cell Biol 1984, 98:685-698. 23. tom Dieck S, Sanmarti-Vila L, Langnaese K, Richter K, Kindler S, Soyke A, et al.: Bassoon, a novel zinc-finger CAG/glutaminerepeat protein selectively localized at the active zone of presynaptic nerve terminals. J Cell Biol 1998, 142:499-509. 24. Neher E: Vesicle pools and Ca2+ microdomains: new tools for understanding their roles in neurotransmitter release. Neuron 1998, 20:389-399.
1411
The endocytic machinery in nerve terminals surrounds sites of exocytosis Jack Roos and Regis B. Kelly In most models of endocytosis, the endocytic machinery is recruited from the cytoplasm by cytoplasmic tails of the plasma membrane proteins that are to be internalized. This does not appear to be true at synapses where the endocytic machinery required for synaptic vesicle recycling is localized to membraneassociated ‘hot spots’ [1,2]. In Drosophila neuromuscular junctions, the multi-domain protein Dap160 is also localized to hot spots [3] and has some characteristics expected of an anchoring protein. Anchoring the endocytic machinery to the plasma membrane might help contribute to the remarkable speed of synaptic vesicle recycling [4]. Here, we report that the endocytic machinery surrounds sites that are believed to be sites of exocytosis. We propose that the radial distribution of the synaptic vesicle recycling machinery already present on the plasma membrane in unstimulated nerve terminals is a fundamental unit of pre-synaptic organization and allows the nerve terminal to extract maximum recycling efficiency out of conventional endocytic machinery.
neuromuscular junctions using a deconvolution microscope. The images acquired were refined by applying deconvolution algorithms to reduce out-of-focus haze that is generally present in confocal sections. The images of single sections of a synaptic bouton revealed that the endocytic machinery had holes of almost constant diameter that give the terminal a ‘honeycomb’ appearance (Figure 1). Comparison to post-synaptic receptor clusters
Address: Department of Biochemistry and Biophysics and the Hormone Research Institute, University of California, San Francisco, San Francisco, California 94143-0534, USA.
The number and dimensions of the holes in the distribution of both Dap160 and dynamin (data not shown) were reminiscent of the size of active zones measured at Drosophila neuromuscular junctions [12]. They were also similar to the number and size of the clusters of postsynaptic glutamate receptors at the Drosophila neuromuscular junction. The distribution of the postsynaptic glutamate receptor, which is a postsynaptic marker for the presynaptic active zone, has been studied by both confocal and electron microscopy in Drosophila third instar larvae using a Myc-epitope-tagged form of the receptor [13]. The receptors cluster in oval discs about 1 µm in length on the post-synaptic membrane opposite synaptic vesicles docked at presynaptic active zones. To determine whether there is any relationship between the sites of endocytosis
Correspondence: Regis B. Kelly E-mail: [email protected]
Figure 1
Received: 12 August 1999 Revised: 21 October 1999 Accepted: 21 October 1999 Published: 22 November 1999 Current Biology 1999, 9:1411–1414 0960-9822/99/$ – see front matter © 1999 Elsevier Science Ltd. All rights reserved.
Results and discussion Concentration of the endocytic machinery into a honeycomb matrix
Dynamin [5,6] and AP-2 [2] are required for synaptic vesicle recycling in Drosophila. Both proteins were shown by confocal immunofluorescence to co-localize with Dap160 in third instar larval neuromuscular junctions (J.R. and R.B.K., unpublished observations). Dap160 was the first member of a family of proteins now called the intersectins [7–10] that have two Eps15 homology domains, multiple Src-homology 3 (SH3) groups and are involved in the formation of endocytic clathrin-coated vesicles [11]. We examined the localization of proteins involved in endocytosis in unstimulated nerve terminals of Drosophila
Current Biology
Dap160 is restricted to varicosities in a honeycomb pattern in Drosophila neuromuscular junctions. Wild-type third instar larvae were dissected and stained for Dap160 and visualized by DeltaVision microscopy. Images were then processed using deconvolution algorithms. Individual Z sections of synaptic boutons on muscle 4 are shown. Similar distributions have been seen for boutons present on muscle 6 and throughout the larval neuromuscular system. The scale bar represents 2.5 µm.
1412
Current Biology Vol 9 No 23
Figure 2
Figure 3 (a)
(a)
GluRB
GluRB
Dap160
Merge
Dap160 (b)
Merge Current Biology
(b)
(c)
GluRB Dap160
GluRB Dynamin Current Biology
Active zones are surrounded by an endocytic honeycomb in resting nerve terminals. MHC-DGluRIIB–Myc larvae were dissected and stained for post-synaptic glutamate receptors (GluRB; green staining) using an anti-Myc monoclonal antibody (9E10). The larvae were also stained with antibodies against either (a,b) Dap160 (red) or (c) dynamin (red). The scale bar represents 5 µm in (a) and 2.5 µm in (b,c).
and the post-synaptic clusters of receptors, the distribution of Dap160 was examined in strains expressing the Mycepitope-tagged glutamate receptor (DGluRIIB–Myc). The Myc-epitope-tagged glutamate receptors clustered into oval rings within each nerve terminal (Figure 2a). Dap160 and dynamin did not overlap with the receptor clusters but instead filled the membrane space surrounding them (Figure 2b,c). Previous analysis of the Drosophila neuromuscular junctions revealed that larger nerve terminals have more active zones per bouton than smaller terminals [14,15]. For boutons of all sizes, the glutamate receptor clusters were always about 1 µm in length and had a minimal separation of about 0.5 µm, which is consistent with previous measurements [12]. The dimensions of the endocytosis zones seen here in unstimulated nerve terminals are likely to be too large to correspond to single coated vesicles recycling synaptic vesicle proteins.. To estimate how a single, 50–100 nm clathrin-coated vesicle might appear, we compared the dimensions of the Dap160 honeycomb with that of
Dap160 and the endocytic honeycomb are coupled to glutamatereceptor clusters in vesicle-depleted nerve terminals. Male third instar larvae of shibirets1; DGluRIIB–Myc mutants were dissected and then incubated at 34°C in 60 mM K+-saline for 5 min. Preparations were fixed and stained for DGluRIIB–Myc (GluRB) using monoclonal antibody 9E10, and with an antibody against Dap160. (a) Individual confocal Z sections reveal that Dap160 is concentrated at unique sites on the plasma membrane that are adjacent to, but do not co-localize with, GluRB staining. (b) Confocal projections of individual Z sections in vesicle-depleted shibirets1; DGluRIIB–Myc male synapses after incubation at the non-permissive temperature. Although the plasma membrane contains additional membrane derived from synaptic vesicles, the endocytic honeycomb remains tightly tethered to clusters of glutamate receptors. The scale bar represents 5 µm.
fluorescent beads of known diameter by pixel counting. The width of the endocytic honeycomb was measured to be about 400 nm in diameter (data not shown), which is larger than the diameter of a clathrin-coated vesicle (about 100 nm). In addition, the concentrations of endocytic machinery observed were not radially symmetrical, as is seen for single clathrin-coated vesicles [16]. Effect of vesicle exocytosis on the endocytic machinery
In some models of endocytosis, the cargo proteins on the plasma membrane recruit the endocytic machinery to the plasma membrane (for reviews see [17,18]). The cargo for the nerve terminal endocytic machinery is the complement of synaptic vesicle proteins. Because there are very few cargo proteins on the plasma membrane in resting terminals, the honeycomb distribution is unlikely to be determined by the presence of cargo molecules. The synaptic vesicle cargo can be redistributed to the plasma membrane when neuromuscular junctions of shibirets1 larvae are stimulated [1,3,5,19,20], but synaptic vesicle uptake and recyling is blocked at the non-permissive temperature. To determine whether the organization of the recycling machinery is modified by redistributing cargo from synaptic vesicles to the plasma membrane, the distribution of Dap160 was examined by confocal microscopy in Drosophila shibirets1 strains possessing the DGluRIIB–Myc
Brief Communication
Figure 4
1413
Figure 5 (a)
Csp
(a)
Dap160
Glutamate receptor clusters
Endocytosis zone Synaptic Active vesicles zone
Wild type (b) (b)
shibire
Current Biology
Dap160 and cysteine string protein (csp) are both enriched in hot spots on Drosophila nerve terminal membranes. (a) Wild-type third instar larvae were dissected and stained with antibodies against cysteine string protein (csp) and Dap160 and visualized by confocal microscopy. (b) Third instar larvae from shibirets1 mutants were dissected, incubated at 34°C for 5 min to deplete the nerve terminal complement of synaptic vesicles, fixed and stained for cysteine string protein and Dap160 and visualized by confocal microscopy. The scale bar represents 4 µm in (a) and 2.5 µm in (b).
transgene. Dap160 was recovered in puncta or ‘hot spots’ in stimulated nerve terminals at the non-permissive temperature as previously described [1,3]. When the distribution of the Myc-epitope-tagged glutamate receptors was compared to the distribution of Dap160 on the nerve terminal membrane, again, no significant overlap was found, despite the massive redistribution of synaptic vesicles. This was true for individual sections (Figure 3a) or when several optical sections were combined (Figure 3b). Thus, it seems likely that, in nerve terminals, the endocytic machinery is clustered into zones independent of the location of the endocytic cargo. After transfer of the synaptic vesicle cargo to the plasma membrane by stimulating shibirets1 flies at the non-permissive temperature, the plasma membrane distribution of these synaptic vesicle antigens was not uniform. A significant fraction of the synaptic vesicle marker, cysteine string protein (csp), colocalized with the hot spots of Dap 160 (Figure 4). The colocalization would be expected if the concentrations of endocytic machinery are indeed sites of endocytosis. The data from Drosophila neuromuscular junctions suggest the concept of the recycling zone as a unit of presynaptic organization (Figure 5). The model is based on a number of reasonable assumptions. First, on the basis of published electron micrographs [13], that active zones of exocytosis lie opposite to post-synaptic receptor clusters and are so physically close that the receptor clusters can be taken as markers of active zones. Unfortunately, no presynaptic
Current Biology
A model of the organization of the recycling machinery in a Drosophila synaptic bouton. (a) Postsynaptic receptors (blue) cluster in apposition to active zones at which synaptic vesicles (red) are docked. Synaptic vesicles in the reserve pool, mitochondria and cytoplasmic elements are omitted. The spaces between active zones are filled with the endocytic machinery (green). (b) A three-dimensional image demonstrates that the proteins of each synaptic vesicle will not have far to migrate before encountering the surrounding endocytic machinery.
active zone marker is currently available in Drosophila. A second assumption is that sites enriched in endocytic machinery are also sites of endocytosis. The data from stimulated shibirets1 mutants are consistent with this assumption (Figure 4). A synaptic organization in which each exocytic active zone has a ‘recapture–surround’ zone of endocytic machinery has some satisfying features. It offers an explanation why, when nerve terminals have more than one active zone, as at the frog neuromuscular junction [21,22] or mossy fiber synapses in the hippocampus [23], the active zones are separated by a minimum distance of about 1 µm. The recapture–surround hypothesis also suggests an explanation of the remarkable constancy of active zone size, from Drosophila to mammals. Active zones might be limited to one micron or so to allow synaptic vesicle proteins to diffuse to the recycling machinery within a second or two after fusion, thus allowing rapid recycling. If all the calcium channels are in the active zone, the spacing we describe could also explain why, on the basis of calcium diffusion models, exocytosis occurs at calcium concentrations in the range of 100 µM whereas the endocytic machinery responds to calcium concentrations that are ten times smaller [4,24]. In conclusion, the large size and convenience of the Drosophila larval neuromuscular junction have revealed an
1414
Current Biology Vol 9 No 23
unexpected feature of nerve terminals. Regions that are predicted to be active zones of exocytosis seem to be embedded in zones of endocytosis. We propose that an active zone of exocytosis surrounded by an annulus of endocytic machinery be considered a recycling zone. Nerve terminals would grow by adding more recycling zones. The separation between active zones, measured to be at least 0.5 µm, might reflect a need for surrounding the active zone with sufficient endocytic machinery to ensure efficient recapture of synaptic vesicle proteins.
Materials and methods Flies Flies (Or-R, shibirets1, MHC-DGluRIIB–Myc) were raised on standard fly food and kept at 18°C. Third instar larval neuromuscular junctions from MHC-DGluRIIB–Myc flies were prepared as described [1]. Results of the complementary distribution of Dap160 and DGluRIIB–Myc were examined on over 25 individual boutons on muscles 6/7 in hemi-segments A2 and A3 and over 20 individual boutons on muscle 4, as well as other neuromuscular synapses throughout the larval preparation. These results were obtained from dissections of six different larval preparations. Fly shibirets1; MHC-DGluRIIB–Myc lines were generated by crossing shibirets1 females to MHC-DGluRIIB–Myc males. Six third instar male progeny were then dissected and analyzed at the non-permissive temperature as described [1]. The distributions of Dap160 and DGluRIIB–Myc were compared on muscles 6/7 and muscle 4 as described for unstimulated MHC-DGluRIIB–Myc flies.
Antibodies Dap160 polyclonal antibodies (Ab1704) [3] and dynamin polyclonal antibodies (Ab2074) [1] were affinity purified as described. Monoclonal antibody 9E10, which recognizes the Myc epitope was the generous gift of Dr Gary Herman (UCSF). Monoclonal antibody against the cysteine string protein was the generous gift of Konrad Zinsmaier (U Penn).
Microscopy Confocal images were acquired on a Leica NT system as described [3]. Optically acquired fluorescent images were generated using a DeltaVision workstation. Z sections were acquired at 0.1 µm increments. Images from individual boutons were processed by deconvolution algorithms using DeltaVision software.
Acknowledgements Thanks to Victor Faundez and Graeme Davis (UCSF) for helpful discussions throughout this work, to Gary Herman (UCSF) for anti-Myc (mAb 9E10) antibodies and to Konrad Zinsmaier (U Penn) for anti-csp antibodies, to Aaron DiAntonio and Corey Goodman (UC/Berkeley) for the MHCDGluRIIB–Myc transgenic flies, to Pat O’Farrell (UCSF) for use of his DeltaVision workstation and to John Sedat and Brigitte Hansen (UCSF) for the fluorescently conjugated beads used to determine the dimensions of the endocytic honeycomb. This work was supported by Postdoctoral Fellowship PF-4314 (JR) from the American Cancer Society and National Institutes of Health Grant NS15927 (RBK).
References 1. Estes PS, Roos J, van der Bliek A, Kelly RB, Krishnan KS, Ramaswami M: Traffic of dynamin within individual Drosophila synaptic boutons relative to compartment-specific markers. J Neurosci 1996, 16:5443-5456. 2. Gonzalez-Gaitan M, Jackle H: Role of Drosophila alpha-adaptin in presynaptic vesicle recycling. Cell 1997, 88:767-776. 3. Roos J, Kelly RB: Dap160, a neural-specific Eps15 homology and multiple SH3 domain-containing protein that interacts with Drosophila dynamin. J Biol Chem 1998, 273:19108-19119. 4. von Gersdorff H, Matthews G: Electrophysiology of synaptic vesicle cycling. Annu Rev Physiol 1999, 61:725-752.
5. Kosaka T, Ikeda K: Possible temperature-dependent blockage of synaptic vesicle recycling induced by a single gene mutation in Drosophila. J Neurobiol 1983, 14:207-225. 6. van der Bliek AM, Meyerowitz EM: Dynamin-like protein encoded by the Drosophila shibire gene associated with vesicular traffic. Nature 1991, 351:411-414. 7. Yamabhai M, Hoffman NG, Hardison NL, McPherson PS, Castagnoli L, Cesareni G, et al.: Intersectin, a novel adaptor protein with two Eps15 homology and five Src homology 3 domains. J Biol Chem 1998, 273:31401-31407. 8. Guipponi M, Scott HS, Chen H, Schebesta A, Rossier C, Antonarakis SE: Two isoforms of a human intersectin (ITSN) protein are produced by brain-specific alternative splicing in a stop codon. Genomics 1998, 53:369-376. 9. Sengar AS, Wang W, Bishay J, Cohen S, Egan SE: The EH and SH3 domain Ese proteins regulate endocytosis by linking to dynamin and Eps15. EMBO J 1999, 18:1159-1171. 10. Okamoto M, Schoch S, Sudhof TC: EHSH1/intersectin, a protein that contains EH and SH3 domains and binds to dynamin and SNAP-25. A protein connection between exocytosis and endocytosis. J Biol Chem 1999, 274:18446-18454. 11. Simpson F, Hussain NK, Qualmann B, Kelly RB, Kay BK, McPherson PS, et al.: SH3-domain-containing proteins function at distinct steps in clathrin-coated vesicle formation. Nat Cell Biol 1999, 1:119-124. 12. Meinertzhagen IA, Govind CK, Stewart BA, Carter JM, Atwood HL: Regulated spacing of synapses and presynaptic active zones at larval neuromuscular junctions in different genotypes of the flies Drosophila and Sarcophaga. J Comp Neurol 1998, 393:482-492. 13. Petersen SA, Fetter RD, Noordermeer JN, Goodman CS, DiAntonio A: Genetic analysis of glutamate receptors in Drosophila reveals a retrograde signal regulating presynaptic transmitter release. Neuron 1997, 19:1237-1248. 14. Atwood HL, Govind CK, Wu C-F: Differential ultrastructure of synaptic terminals on ventral longitudinal abdominal muscles in Drosophila larvae. J Neurobiol 1993, 24:1008-1024. 15. Stewart BA, Schuster CM, Goodman CS, Atwood HL: Homeostasis of synaptic transmission in Drosophila with genetically altered nerve terminal morphology. J Neurosci 1996, 16:3877-3886. 16. Gaidarov I, Santini F, Warren RA, Keen JH: Spatial control of coated pit dynamics in living cells. Nat Cell Biol 1999, 1:1-7. 17. Schmid SL: Clathrin-coated vesicle formation and protein sorting: an integrated process. Annu Rev Biochem 1997, 66:511-548. 18. Lewis DA, Mellman I: Sorting out adaptors. Biochim Biophys Acta 1998, 1401:129-145. 19. Ramaswami M, Krishnan KS, Kelly RB: Intermediates in synaptic vesicle recycling revealed by optical imaging of Drosophila neuromuscular junctions. Neuron 1994, 13:363-375. 20. van de Goor J, Ramaswami M, Kelly RB: Redistribution of synaptic vesicles and their proteins in temperature-sensitive shibirets1 mutant Drosophila. Proc Natl Acad Sci USA 1995, 92:5739-5743. 21. Heuser JE, Reese TS: Evidence for recycling of synaptic vesicle membrane during transmitter release at the frog neuromuscular junction. J Cell Biol 1973, 57:315-344. 22. Miller TM, Heuser JE: Endocytosis of synaptic vesicle membrane at the frog neuromuscular junction. J Cell Biol 1984, 98:685-698. 23. tom Dieck S, Sanmarti-Vila L, Langnaese K, Richter K, Kindler S, Soyke A, et al.: Bassoon, a novel zinc-finger CAG/glutaminerepeat protein selectively localized at the active zone of presynaptic nerve terminals. J Cell Biol 1998, 142:499-509. 24. Neher E: Vesicle pools and Ca2+ microdomains: new tools for understanding their roles in neurotransmitter release. Neuron 1998, 20:389-399.