Secretory carrier membrane proteins 31-35 define a common protein composition among secretory carrier membranes.

A novel compositional overlap between membranes of exocrine and endocrine granules, synaptic vesicles, and a liver Golgi fraction has been identified using a monoclonal antibody (SG7C12) raised against parotid secretion granule membranes. This antibody binds secretory carrier membrane proteins with apparent Mr 31,000, 33,000 and 35,000 (designated SCAMPs 31, 33, 35). The proteins are nonglycosylated integral membrane components, and the epitope recognized by SG7C12 is on the cytoplasmic side of the granule membrane. SCAMP 33 is found in all secretory carrier membranes studied so far while SCAMP 35 is found in exocrine and certain endocrine granules and liver Golgi membranes and SCAMP31 only in exocrine granules. They are not related to other similar-sized proteins that have been studied previously in relation to vesicular transport and secretion. Immunocytochemical staining shows that these SCAMPs are highly concentrated in the apical cytoplasm of exocrine cells. Antigens are present not only on exocrine granules and synaptic vesicles but also on other smooth membrane vesicles of exocrine and neural origin as revealed by immunolocalization in subcellular fractions and immunoadsorption to antibody-coated magnetic beads. The wide tissue distribution and localization to secretory carriers and related membranes suggest that SCAMPs 31-35 may be essential components in vesicle-mediated transport/secretion.     

A new type of coated vesicular carrier that appears not to contain clathrin: its possible role in protein transport within the Golgi stack

Isolated Golgi membranes incubated in the presence of ATP and a cytosolic protein fraction form a population of coated buds or vesicles from the Golgi cisternae. The coats do not have the characteristic hexagonal-pentagonal basketwork of clathrin, and do not react with anti-clathrin polyclonal antibody. The conditions that produce these apparently nonclathrin-coated buds also reconstitute protein transport between compartments of the Golgi stack. The membrane of the buds contains the glycoprotein in transit through these Golgi stacks (VSV-encoded G protein). This suggests that protein transport through the Golgi stack is mediated by a new type of coated vesicle that does not contain clathrin. The concentration of G protein in the coated buds reflects the local concentration of G protein in the cisternae, raising the possibility that the Golgi coated vesicles may be "bulk" membrane carriers.     

Is cytochrome P-450 transported from the endoplasmic reticulum to the Golgi apparatus in rat hepatocytes?

The Golgi apparatus mediates intracellular transport of not only secretory and lysosomal proteins but also membrane proteins. As a typical marker membrane protein for endoplasmic reticulum (ER) of rat hepatocytes, we have selected phenobarbital (PB)-inducible cytochrome P- 450 (P-450[PB]) and investigated whether P-450(PB) is transported to the Golgi apparatus or not by combining biochemical and quantitative ferritin immunoelectron microscopic techniques. We found that P-450(PB) was not detectable on the membrane of Golgi cisternae either when P-450 was maximally induced by phenobarbital treatment or when P-450 content in the microsomes rapidly decreased after cessation of the treatment. The P-450 detected biochemically in the Golgi subcellular fraction can be explained by the contamination of the microsomal vesicles derived from fragmented ER membranes to the Golgi fraction. We conclude that when the transfer vesicles are formed by budding on the transitional elements of ER, P-450 is completely excluded from such regions and is not transported to the Golgi apparatus, and only the membrane proteins destined for the Golgi apparatus, plasma membranes, or lysosomes are selectively collected and transported.     

Identification of the Plasma Membrane Proteolipid Protein as a Constituent of Brain Coated Vesicles and Synaptic Plasma Membrane

We have analyzed brain coated vesicles and synaptic plasma membrane for the presence of the plasma membrane proteolipid protein. Coated vesicles were isolated from calf brain gray matter with a final purification on Sephacryl S-1000 and reisolated twice by chromatography to ensure homogeneity. Fractions were analyzed by gel electrophoresis, immunoblotting for clathrin heavy chain, and by electron microscopy. Using an immunoblotting assay we were able to demonstrate the presence of the plasma membrane proteolipid protein in these coated vesicles at a significant level (i.e., approximately 1% of the bilayer protein of these vesicles). Reisolation of coated vesicles did not diminish the concentration of the protein in this fraction. Removal of the clathrin coat proteins or exposure of the coated vesicles to 0.1 M Na2CO3 showed that the plasma membrane proteolipid protein is not removed during uncoating and lysis but is intrinsic to the membrane bilayer of these vesicles. These studies demonstrate that plasma membrane proteolipid protein represents a significant amount of the bilayer protein of coated vesicles, suggesting that these vesicles may be a transport vehicle for the intracellular movement of the plasma membrane proteolipid protein. Isolation of synaptic plasma membranes proteolipid adult rat brain and estimation of the plasma membrane proteolipid protein content using the immunoblotting method confirmed earlier studies that show this protein is present in this membrane fraction at high levels as well (approximately 1-2%). The level of this protein in the synaptic plasma membrane suggests that the synaptic plasma membrane is one major site to which these vesicles may be targeted or from which the protein is being retrieved.     

Differences in the protein composition of bovine retinal rod outer segment disk and plasma membranes isolated by a ricin-gold-dextran density perturbation method

The plasma membrane and disk membranes of bovine retinal rod outer segments (ROS) have been purified by a novel density-gradient perturbation method for analysis of their protein compositions. Purified ROS were treated with neuraminidase to expose galactose residues on plasma membrane-specific glycoproteins and labeled with ricin-gold-dextran particles. After the ROS were lysed in hypotonic buffer, the plasma membrane was dissociated from the disks by either mild trypsin digestion or prolonged exposure to low ionic strength buffer. The dense ricin-gold-dextran-labeled plasma membrane was separated from disks by sucrose gradient centrifugation. Electron microscopy was used to follow this fractionation procedure. The dense red pellet primarily consisted of inverted plasma membrane vesicles containing gold particles; the membrane fraction of density 1.13 g/cc consisted of unlabeled intact disks and vesicles. Ricin-binding studies indicated that the plasma membrane from trypsin-treated ROS was purified between 10-15-fold. The protein composition of plasma membranes and disks was significantly different as analyzed by SDS gels and Western blots labeled with lectins and monoclonal antibodies. ROS plasma membrane exhibited three major proteins of 36 (rhodopsin), 38, and 52 kD, three ricin-binding glycoproteins of 230, 160, and 110 kD, and numerous minor proteins in the range of 14-270 kD. In disk membranes rhodopsin appeared as the only major protein. A 220-kD concanavalin A-binding glycoprotein and peripherin, a rim-specific protein, were also present along with minor proteins of 43 and 57-63 kD. Radioimmune assays indicated that the ROS plasma membrane contained about half as much rhodopsin as disk membranes.     

Effect of packing density on rhodopsin stability and function in polyunsaturated membranes

Rod outer segment disk membranes are densely packed with rhodopsin. The recent notion of raft or microdomain structures in disk membranes suggests that the local density of rhodopsin in disk membranes could be much higher than the average density corresponding to the lipid/protein ratio. Little is known about the effect of high packing density of rhodopsin on the structure and function of rhodopsin and lipid membranes. Here we examined the role of rhodopsin packing density on membrane dynamic properties, membrane acyl chain packing, and the structural stability and function of rhodopsin using a combination of biophysical and biochemical techniques. We reconstituted rhodopsin into large unilamellar vesicles consisting of polyunsaturated 18:0,22:6n3PC, which approximates the polyunsaturated nature of phospholipids in disk membranes, with rhodopsin/lipid ratios ranging from 1:422 to 1:40. Our results showed that increased rhodopsin packing density led to reduced membrane dynamics revealed by the fluorescent probe 1,6-diphenyl-1,3,5-hexatriene, increased phospholipid acyl chain packing, and reduced rhodopsin activation, yet it had minimal impact on the structural stability of rhodopsin. These observations imply that densely packed rhodopsin may impede the diffusion and conformational changes of rhodopsin, which could reduce the speed of visual transduction.     

Rab11 mediates post-Golgi trafficking of rhodopsin to the photosensitive apical membrane of Drosophila photoreceptors

In developing Drosophila photoreceptors, rhodopsin is trafficked to the rhabdomere, a specialized domain within the apical membrane surface. Rab11, a small GTPase implicated in membrane traffic, immunolocalizes to the trans-Golgi network, cytoplasmic vesicles and tubules, and the base of rhabdomeres. One hour after release from the endoplasmic reticulum, rhodopsin colocalizes with Rab11 in vesicles at the base of the rhabdomere. When Rab11 activity is reduced by three different genetic procedures, rhabdomere morphogenesis is inhibited and rhodopsin-bearing vesicles proliferate within the cytosol. Rab11 activity is also essential for development of MVB endosomal compartments; this is probably a secondary consequence of impaired rhabdomere development. Furthermore, Rab11 is required for transport of TRP, another rhabdomeric protein, and for development of specialized membrane structures within Garland cells. These results establish a role for Rab11 in the post-Golgi transport of rhodopsin and of other proteins to the rhabdomeric membranes of photoreceptors, and in analogous transport processes in other cells.     

Release of the phophodiesterase activator by cyclic AMP-dependent ATP:protein phosphotransferase from subcellular fractions of rat brain

The subcellular distribution of the endogenous phosphodiesterase activator and its release from membranes by a cyclic AMP-dependent ATP:protein phosphotransferase was studied in fractions and subfractions of rat brain homogenate. These fractions were obtained by differential centrifugation and sucrose density gradient; their identity was ascertained by electron microscopy and specific enzyme markers.In the subcellular particulate fractions, the concentration of activator is highest in the microsomal fraction, followed by the mictochondrial and nuclear fractions. Gradient centrifugation of the main mitochondrial subfraction revealed that activator was concentrated in those fractions containing mainly synaptic membranes.Activator was released from membranes by a cyclic AMP-dependent phosphorylation of membrane protein. The release of activator occurred mainly from the mitochondrial subfractions containing synaptic membranes and synaptic vesicles.The data support the view that a release of activator from membranes may be important in normalizing the elevated concentration of cyclic AMP following persistent transsynaptic activation of adenylate cyclase.     

Localization of transferrin receptors and insulin-like growth factor II receptors in vesicles from 3T3-L1 adipocytes that contain intracellular glucose transporters

Transferrin receptors in detergent extracts of subcellular membrane fractions prepared from 3T3-L1 adipocytes were measured by a binding assay. There was a small but significant increase (1.2-fold) in the amount of receptor in a crude plasma membrane fraction and a 40% decrease in the number of transferrin receptors in microsomal membranes prepared from insulin-treated cells, when compared with corresponding fractions from control cells. Intracellular vesicles containing insulin-responsive glucose transporters (GT) have been isolated by immunoadsorption from the microsomal fraction (Biber, J. W., and G. E. Lienhard. 1986. J. Biol. Chem. 261:16180-16184). All of the transferrin receptors in this fraction were localized in these vesicles; however, because the GT vesicles contain approximately 30-fold fewer transferrin receptors than GT, on the average only one vesicle in three contains a transferrin receptor. The binding of 125I-pentamannose 6-phosphate BSA to 3T3-L1 adipocytes at 4 degrees C was used to monitor surface insulin-like growth factor II (IGF-II)/mannose 6-phosphate receptors. Exposure of cells to insulin at 37 degrees C for 5 min resulted in a 2.5-4.5-fold increase in surface receptors. There was a corresponding 20% decrease in the amount of IGF-II receptors in the microsomal membranes prepared from insulin-treated cells, as assayed by immunoblotting. Moreover, the IGF-II receptors and GT were located in the same intracellular vesicles, since antibodies to the carboxyterminal peptide of either protein immunoadsorbed vesicles containing 70-95% of both proteins initially present in the microsomal fraction. In conjunction with other studies, these results indicate that in 3T3-L1 adipocytes, three membrane proteins (the GT, the transferrin receptor, and the IGF-II receptor) respond similarly to insulin, by redistributing to the surface from intracellular compartment(s) in which they are colocalized.     

Cholesterol and vesicular stomatitis virus G protein take separate routes from the endoplasmic reticulum to the plasma membrane

Transport of newly synthesized cholesterol and vesicular stomatitis virus G protein from the endoplasmic reticulum to the plasma membrane is interrupted by incubation at 15 degrees C. Under this condition the newly synthesized molecules accumulate in both the endoplasmic reticulum (ER) and a subcellular vesicle fraction of low density called the lipid-rich vesicle fraction. The material in the lipid-rich vesicle fraction appears to be a post-ER intermediate in the transport process to the plasma membrane (PM). Although both newly synthesized cholesterol and G protein accumulate in this intermediate compartment at 15 degrees C, suggesting cotransport, treatment with Brefeldin A does not affect cholesterol transport to the PM, whereas it strongly inhibits G protein transport. We conclude that cholesterol and G protein leave the ER in separate vesicles, the cholesterol containing vesicles bypass the Golgi apparatus and proceed to the PM, whereas G protein containing vesicles follow the well documented Golgi route to the cell surface.     

Release of the phosphodiesterase activator by cyclic AMP-dependent ATP:protein phosphotransferase from subcellular fractions of rat brain.

The subcellular distribution of the endogenous phosphodiesterase activator and its release from membranes by a cyclic AMP-dependent ATP:protein phosphotransferase was studied in fractions and subfractions of rat brain homogenate. These fractions were obtained by differential centrifugation and sucrose density gradient; their identity was ascertained by electron microscopy and specific enzyme markers. In the subcellular particulate fractions, the concentration of activator is highest in the microsomal fraction, followed by the mitochondrial and nuclear fractions. Gradient centrifugation of the main mitochondrial subfraction revealed that activator was concentrated in those fractions containing mainly synaptic membranes. Activator was releasted from membranes by a cyclic AMP-dependent phosphorylation of membrane protein. The release of activator occurred mainly from the mitochondrial subfractions containing synaptic membranes and synaptic vesicles. The data support the view that a release of activator from membranes may be important in normalizing the elevated concentration of cyclic AMP following persistent transsynaptic activation of adenylate cyclase.     

Characteristics of endoplasmic reticulum-derived transport vesicles

We have isolated vesicles that mediate protein transport from the ER to Golgi membranes in perforated yeast. These vesicles, which form de novo during in vitro incubations, carry lumenal and membrane proteins that include core-glycosylated pro-alpha-factor, Bet1, Sec22, and Bos1, but not ER-resident Kar2 or Sec61 proteins. Thus, lumenal and membrane proteins in the ER are sorted prior to transport vesicle scission. Inhibition of Ypt1p-function, which prevents newly formed vesicles from docking to cis-Golgi membranes, was used to block transport. Vesicles that accumulate are competent for fusion with cis-Golgi membranes, but not with ER membranes, and thus are functionally committed to vectorial transport. A 900-fold enrichment was developed using differential centrifugation and a series of velocity and equilibrium density gradients. Electron microscopic analysis shows a uniform population of 60 nm vesicles that lack peripheral protein coats. Quantitative Western blot analysis indicates that protein markers of cytosol and cellular membranes are depleted throughout the purification, whereas the synaptobrevin-like Bet1, Sec22, and Bos1 proteins are highly enriched. Uncoated ER-derived transport vesicles (ERV) contain twelve major proteins that associate tightly with the membrane. The ERV proteins may represent abundant cargo and additional targeting molecules.     

The Endoplasmic Reticulum of Mung Bean Cotyledons: ROLE IN THE ACCUMULATION OF HYDROLASES IN PROTEIN BODIES DURING SEEDLING GROWTH

The subcellular localization of two hydrolases (ribonuclease and vicilin peptidohydrolase) which are synthesized de novo in the cotyledons of mung bean seedlings was studied. Earlier experiments had shown that both enzymes accumulate in the protein bodies in the course of seedling growth. Two methods to fractionate subcellular organelles were used to demonstrate that a significant proportion of the enzymes is organelle-associated. This proportion is highest (up to 50% for vicilin peptidohydrolase and 15% for ribonuclease) when synthesis of the enzymes has just started. Evidence obtained with isopycnic sucrose gradients indicates that both hydrolases are associated with membranes rich in NADH-cytochrome c reductase, a marker enzyme for the endoplasmic reticulum (ER). The hydrolases band with the NADH-cytochrome c reductase under conditions where the ribosomes remain attached or are detached from the ER-derived vesicles. Treatment of the ER-derived vesicles with Triton X-100 shows that vicilin peptidohydrolase and vesicle membranes can be physically separated without dissolving the membranes, indicating that the proteinase is soluble within the vesicles. These data support the conclusion that the ER is involved in the transport of ribonuclease and proteinase to the protein bodies.     

Low molecular weight GTP-binding proteins are associated with neuronal organelles involved in rapid axonal transport and exocytosis

Recent evidence suggests that low molecular weight GTP-binding proteins may play important roles in a variety of membrane transport processes. In order to address the question of whether these proteins are involved in transport processes in the nerve axon, we have assessed their presence in rapid transport membranes from rabbit optic nerve. We report the characterization of a group of low molecular weight GTP-binding proteins which are constituents of rapid transport vesicles. Although these proteins are components of rapid transport vesicles, they are apparently not major rapidly transported species. They are localized in cytosolic as well as in membrane fractions of axons, and the membrane-associated form behaves as an integral membrane protein(s). These proteins are also found in association with a variety of vesicular and organellar components of neurons including coated vesicles, synaptic vesicles, synaptic plasma membranes, and mitochondria. We discuss the possible roles of these proteins in rapid axonal transport and exocytosis.     

Comparison of the properties of active Ca2+ transport by the islet-cell endoplasmic reticulum and plasma membrane

The properties of active or ATP-dependent calcium transport by islet-cell endoplasmic reticulum and plasma membrane-enriched subcellular fractions were directly compared. These studies indicate that the active calcium transport systems of the two membranes are fundamentally distinct. In contrast to calcium uptake by the endoplasmic reticulum-enriched fraction, calcium uptake by islet-cell plasma membrane-enriched vesicles exhibited a different pH optimum, was not sustained by oxalate, and showed an approximate 30-fold greater affinity for ionized calcium. A similar difference in affinity for calcium was exhibited by the Ca²⁺-stimulated ATPase activities which are associated with these islet-cell subcellular fractions. Consistent with the effects of calmodulin on calcium transport, calmodulin stimulated Ca²⁺-ATPase in the plasma membranes, but did not increase calcium-stimulated ATPase activity in the endoplasmic reticulum membranes. The physiological significance of the differences observed in calcium transport by the endoplasmic reticulum and plasma membrane fractions relative to the regulation of insulin secretion by the islets of Langerhans is discussed.     

ADP ribosylation factor and a 14-kD polypeptide are associated with heparan sulfate-carrying post-trans-Golgi network secretory vesicles in rat hepatocytes

Constitutive secretory vesicles carrying heparan sulfate proteoglycan (HSPG) were identified in isolated rat hepatocytes by pulse-chase experiments with [35S]sulfate and purified by velocity-controlled sucrose gradient centrifugation followed by equilibrium density centrifugation in Nycodenz. Using this procedure, the vesicles were separated from plasma membranes, Golgi, trans-Golgi network (TGN), ER, endosomes, lysosomes, transcytotic vesicles, and mitochondria. The diameter of these vesicles was approximately 100-200 nm as determined by electron microscopy. A typical coat structure as described for intra- Golgi transport vesicles or clathrin-coated vesicles could not be seen, and the vesicles were not associated with the coat protein beta-COP. Furthermore, the vesicles appear to represent a low density compartment (1.05-1.06 g/ml). Other constitutively secreted proteins (rat serum albumin, apolipoprotein E, and fibrinogen) could not be detected in purified HSPG-carrying vesicles, but banded in the denser fractions of the Nycodenz gradient. Moreover, during pulse-chase labeling with [35S]methionine, labeled albumin did not appear in the post-TGN vesicle fraction carrying HSPGs. These findings indicate sorting of HSPGs and albumin into different types of constitutive secretory vesicles in hepatocytes. Two proteins were found to be tightly associated with the membranes of the HSPG carrying vesicles: a member of the ADP ribosylation factor family of small guanine nucleotide-binding proteins and an unknown 14-kD peripheral membrane protein (VAPP14). Concerning the secretory pathway, we conclude from these results that ADP ribosylation factor proteins are not only involved in vesicular transport from the ER via the Golgi to the TGN, but also in vesicular transport from the TGN to the plasma membrane.     

Synaptobrevin: an integral membrane protein of 18,000 daltons present in small synaptic vesicles of rat brain.

A protein with an apparent mol. wt of 18,000 daltons (synaptobrevin) was identified in synaptic vesicles from rat brain. Some of its properties were studied using monoclonal and polyclonal antibodies. Synaptobrevin is an integral membrane protein with an isoelectric point of approximately 6.6. During subcellular fractionation, synaptobrevin followed the distribution of small synaptic vesicles, with the highest enrichment in the purified vesicle fraction. Immunogold electron microscopy of subcellular particles revealed that synaptobrevin is localized in nerve endings where it is concentrated in the membranes of virtually all small synaptic vesicles. No significant labeling was observed on the membranes of peptide-containing large dense core vesicles. In agreement with these results, synaptobrevin immunoreactivity has a widespread distribution in nerve terminal-containing regions of the central and peripheral nervous system as shown by light microscopy immunocytochemistry. Outside the nervous system, synaptobrevin immunoreactivity was found in endocrine cells and cell lines (endocrine pancreas, adrenal medulla, PC12 cells, insulinoma cells) but not in other cell types, for example smooth muscle, skeletal muscle and exocrine pancreas. Thus, the distribution of synaptobrevin is similar to that of synaptophysin, a well-characterized membrane protein of small vesicles in neurons and endocrine cells.     

The t-SNAREs syntaxin 1 and SNAP-25 are present on organelles that participate in synaptic vesicle recycling

Syntaxin 1 and synaptosome-associated protein of 25 kD (SNAP-25) are neuronal plasmalemma proteins that appear to be essential for exocytosis of synaptic vesicles (SVs). Both proteins form a complex with synaptobrevin, an intrinsic membrane protein of SVs. This binding is thought to be responsible for vesicle docking and apparently precedes membrane fusion. According to the current concept, syntaxin 1 and SNAP-25 are members of larger protein families, collectively designated as target-SNAP receptors (t-SNAREs), whose specific localization to subcellular membranes define where transport vesicles bind and fuse. Here we demonstrate that major pools of syntaxin 1 and SNAP-25 recycle with SVs. Both proteins cofractionate with SVs and clathrin-coated vesicles upon subcellular fractionation. Using recombinant proteins as standards for quantitation, we found that syntaxin 1 and SNAP-25 each comprise approximately 3% of the total protein in highly purified SVs. Thus, both proteins are significant components of SVs although less abundant than synaptobrevin (8.7% of the total protein). Immunoisolation of vesicles using synaptophysin and syntaxin specific antibodies revealed that most SVs contain syntaxin 1. The widespread distribution of both syntaxin 1 and SNAP-25 on SVs was further confirmed by immunogold electron microscopy. Botulinum neurotoxin C1, a toxin that blocks exocytosis by proteolyzing syntaxin 1, preferentially cleaves vesicular syntaxin 1. We conclude that t- SNAREs participate in SV recycling in what may be functionally distinct forms.     

Ultrastructural localization of vesicle-associated membrane protein(s) to specialized membrane structures in human pericytes, vascular smooth muscle cells, endothelial cells, neutrophils, and eosinophils.

Vesicle-associated membrane proteins (VAMPs) are important to the trafficking of vesicles between membrane-bound intracytoplasmic organelles, in the facilitation of neurosecretion, and in constitutive and regulated secretion in non-neuronal cells. We used a pre-embedding ultrastructural immunonanogold method to localize VAMPs to subcellular sites in human cells of five lineages known to have cytoplasmic vesicles that may function in vesicular transport. We found VAMPs localized to caveolae in pericytes, vascular smooth muscle cells, and endothelial cells of venules, to the vesiculo-vacuolar organelle, recently defined in venular endothelial cells, to the vesicle-rich intergranular cytoplasm and secretory granule membranes of neutrophils, and to perigranular cytoplasmic secretory vesicles and secretory granule membranes in eosinophils. These specific localizations in five human vascular and granulocyte lineages support the notion that VAMPs have vesicle-associated functions in these cells.     

Biochemical and immunochemical analysis of rat brain dynamin interaction with microtubules and organelles in vivo and in vitro

We have purified a 100-kD rat brain protein that has microtubule cross- linking activity in vitro, and have determined that it is dynamin, a putative microtubule-associated motility protein. We find that dynamin appears to be specific to neuronal tissue where it is present in both soluble and particulate tissue fractions. In the cytosol it is abundant, representing as much as 1.5% of the total extractable protein. Dynamin appears to be in particulate material due to association with a distinct subcellular membrane fraction. Surprisingly, by immunofluorescence analysis of PC12 cells we find that dynamin is distributed uniformly throughout the cytoplasm with no apparent microtubule association in either interphase, mitotic, or taxol-treated cells. Upon nerve growth factor (NGF) induction of PC12 cell differentiation into neurons, dynamin levels increase approximately twofold. In the cell body, the distribution of dynamin again remains clearly distinct from that of tubulin, and in axons, where microtubules are numerous and ordered into bundles, dynamin staining is sparse and punctate. On the other hand, in the most distal domain of growth cones, where there are relatively few microtubules, dynamin is particularly abundant. The dynamin staining of neurites is abolished by extraction of the cells with detergent under conditions that preserve microtubules, suggesting that dynamin in neurites is associated with membranes. We conclude that dynamin is a neuronal protein that is specifically associated with as yet unidentified vesicles. It is possible, but unproven, that it may link vesicles to microtubules for transport in differentiated axons.