Crystal structure of the amino-terminal domain of N-ethylmaleimide-sensitive fusion protein.

The cytosolic ATPase N-ethylmaleimide-sensitive fusion protein (NSF) disassembles complexes of membrane-bound proteins known as SNAREs, an activity essential for vesicular trafficking. The amino-terminal domain of NSF (NSF-N) is required for the interaction of NSF with the SNARE complex through the adaptor protein alpha-SNAP. The crystal structure of NSF-N reveals two subdomains linked by a single stretch of polypeptide. A polar interface between the two subdomains indicates that they can move with respect to one another during the catalytic cycle of NSF. Structure-based sequence alignments indicate that in addition to NSF orthologues, the p97 family of ATPases contain an amino-terminal domain of similar structure.

SNAPs and NSF: general members of the fusion apparatus.

The proteins NSF and SNAPs are known to participate in several intracellular fusion events, but their exact functions in the fusion process are unclear. Molecular studies have now shown that the ability of NSF to hydrolyse ATP is essential for membrane fusion and that this activity is regulated by SNAPs. This article reviews recent work on NSF and SNAPs, and speculates about how they interact with each other and SNAREs to promote membrane fusion.

A multisubunit particle implicated in membrane fusion.

The N-ethylmaleimide sensitive fusion protein (NSF) is required for fusion of lipid bilayers at many locations within eukaryotic cells. Binding of NSF to Golgi membranes is known to require an integral membrane receptor and one or more members of a family of related soluble NSF attachment proteins (alpha-, beta-, and gamma-SNAPs). Here we demonstrate the direct interaction of NSF, SNAPs and an integral membrane component in a detergent solubilized system. We show that NSF only binds to SNAPs in the presence of the integral receptor, resulting in the formation of a multisubunit protein complex with a sedimentation coefficient of 20S. Particle assembly reveals striking differences between members of the SNAP protein family; gamma-SNAP associates with the complex via a binding site distinct from that used by alpha- and beta-SNAPs, which are themselves equivalent, alternative subunits of the particle. Once formed, the 20S particle is subsequently able to disassemble in a process coupled to the hydrolysis of ATP. We suggest how cycles of complex assembly and disassembly could help confer specificity to the generalized NSF-dependent fusion apparatus.

N-ethylmaleimide-sensitive fusion protein: a trimeric ATPase whose hydrolysis of ATP is required for membrane fusion.

The NEM-sensitive fusion protein, NSF, together with SNAPs (soluble NSF attachment proteins) and the SNAREs (SNAP receptors), is thought to be generally used for the fusion of transport vesicles to their target membranes. NSF is a homotrimer whose polypeptide subunits are made up of three distinct domains: an amino-terminal domain (N) and two homologous ATP-binding domains (D1 and D2). Mutants of NSF were produced in which either the order or composition of the three domains were altered. These mutants could not support intra-Golgi transport, but they indicated that the D2 domain was required for trimerization of the NSF subunits. Mutations of the first ATP-binding site that affected either the binding (K266A) or hydrolysis (E329Q) of ATP completely eliminated NSF activity. The hydrolysis mutant was an effective, reversible inhibitor of Golgi transport with an IC50 of 125 ng/50 microliters assay. Mutants in the second ATP-binding site (binding, K549A; hydrolysis, D604Q) had either 14 or 42% the specific activity of the wild-type protein, respectively. Using coexpression of an inactive mutant with wild-type subunits, it was possible to produce a recombinant form of trimeric NSF that contained a mixture of subunits. The mixed NSF trimers were inactive, even when only one mutant subunit was present, suggesting that NSF action requires each of the three subunits in a concerted mechanism. These studies demonstrate that the ability of the D1 domain to hydrolyze ATP is required for NSF activity and, therefore is required for membrane fusion. The D2 domain is required for trimerization, but its ability to hydrolyze ATP is not absolutely required for NSF function.

The SNARE machinery is involved in apical plasma membrane trafficking in MDCK cells.

We have investigated the controversial involvement of components of the SNARE (soluble N-ethyl maleimide-sensitive factor [NSF] attachment protein [SNAP] receptor) machinery in membrane traffic to the apical plasma membrane of polarized epithelial (MDCK) cells. Overexpression of syntaxin 3, but not of syntaxins 2 or 4, caused an inhibition of TGN to apical transport and apical recycling, and leads to an accumulation of small vesicles underneath the apical plasma membrane. All other tested transport steps were unaffected by syntaxin 3 overexpression. Botulinum neurotoxin E, which cleaves SNAP-23, and antibodies against alpha-SNAP inhibit both TGN to apical and basolateral transport in a reconstituted in vitro system. In contrast, we find no evidence for an involvement of N-ethyl maleimide-sensitive factor in TGN to apical transport, whereas basolateral transport is NSF-dependent. We conclude that syntaxin 3, SNAP-23, and alpha-SNAP are involved in apical membrane fusion. These results demonstrate that vesicle fusion with the apical plasma membrane does not use a mechanism that is entirely unrelated to other cellular membrane fusion events, but uses isoforms of components of the SNARE machinery, which suggests that they play a role in providing specificity to polarized membrane traffic.

Intracellular membrane fusion.

Protein trafficking and membrane assembly are accomplished in eukaryotes by the specific targeting and fusion of vesicles. In this review we describe some of the molecules implicated as components of the fusion apparatus, and evidence that suggests the same factors are recruited for a variety of intracellular fusion events.

N-ethylmaleimide sensitive factor (NSF) structure and function.

Our understanding of the molecular mechanisms of membrane trafficking advanced at a rapid rate during the 1990s. As one of the initial protein components of the trafficking machinery to be identified, N-ethylmaleimide sensitive factor (NSF) has served as a reference point in many of these recent studies. This hexameric ATPase is essential for most of the membrane-trafficking events in a cell. Initially, due to its ATPase activity, NSF was thought to be the motor that drove membrane fusion. Subsequent studies have shown that NSF actually plays the role of a chaperone by activating SNAP receptor proteins (SNAREs) so that they can participate in membrane fusion. In this review we will examine the initial characterization of NSF, its role in membrane fusion events, and what new structural information can tell us about NSF's mechanism of action.

The effects of SNAP/SNARE complexes on the ATPase of NSF.

The ATPase of the N-ethylmaleimide sensitive factor (NSF) appears to be central to the events that culminate in vesicle-target membrane fusion. Complexes containing different combinations of NSF, alpha-SNAP, Vamp-2 (synaptobrevin 2), syntaxin 1, and SNAP-25 were reconstituted and then tested for their effect on the ATPase of NSF. While NSF interacts with all alpha-SNAP-containing complexes, only the alpha-SNAP/t-SNARE complex significantly stimulated ATPase activity. This stimulation was dependent on increasing SNAP/t-SNARE complex and was saturable. The apparent stimulation of ATPase activity is due to a 10-fold increase in initial hydrolysis rate. Complex containing both v- and t-SNAREs bound significantly more alpha-SNAP but did not stimulate the ATPase of NSF.

Organization of the secretory machinery in the rodent brain: distribution of the t-SNAREs, SNAP-25 and SNAP-23.

Vesicular transport events appear to be facilitated by the VAMP/synaptobrevin family of membrane proteins in the vesicle (v-SNAREs) and a heterodimeric complex of syntaxin and SNAP-23/25 family members in the target membrane (t-SNAREs). In this manuscript we examine the tissue distribution and composition of the heterodimeric t-SNARE complexes in adult rodent brain. Analysis of protein extracts from brain regions shows that SNAP-25, syntaxin 1, and 4 are broadly distributed, while SNAP-23, syntaxin 3, and 7 show distinct patterns of expression. Further immunohistochemistry and fractionation studies show that while SNAP-25 is enriched in axons and nerve terminals, SNAP-23 is concentrated in cell bodies. Both SNAP-23 and SNAP-25 associate with the plasma membrane and can be metabolically labeled with [(3)H] palmitate in AtT-20 cells. Anti-SNAP-25 antibodies co-immunoprecipitate t-SNARE heterodimers from brain extracts that predominantly contain syntaxin 1 and 2. Contrary to results from in vitro binding assays, SNAP-23 was found predominantly associated with syntaxin 3. These observations suggest that t-SNARE, heterodimer composition is governed more by SNARE expression and localization than by simple protein-protein affinity.

Influence of serotonin on the kinetics of vesicular release.

The mechanisms by which synaptic vesicles are transported and primed to fuse with the presynaptic membrane are important to all chemical synapses. Processes of signal transduction that affect vesicular dynamics, such as the second-messenger cascades induced by neuromodulators, are more readily addressed in assessable synaptic preparations of neuromuscular junctions in the crayfish. We assessed the effects of serotonin (5-HT) through the analysis of the latency jitter and the quantal parameters: n and p in the opener muscle of the walking leg in crayfish. There is an increase in the size of the postsynaptic currents due to more vesicles being released. Quantal analysis reveals a presynaptic mechanism by an increase in the number of vesicles being released. Latency measures show more events occur with a short latency in the presence of 5-HT. No effect on the frequency or size of spontaneous release was detected. Thus, the influence of 5-HT is presynaptic, leading to a release of more vesicles at a faster rate.

Glutamate receptor antagonists inhibit calpain-mediated cytoskeletal proteolysis in focal cerebral ischemia.

Excitatory amino acids may promote microtubular proteolysis observed in ischemic neuronal degeneration by calcium-mediated activation of calpain, a neutral protease. We tested this hypothesis in an animal model of focal cerebral ischemia without reperfusion. Spontaneously hypertensive rats were treated with 2, 3-dihydroxy-6-nitro-7-sulfamoyl-benzo-(F)quinoxaline (NBQX), a competitive antagonist of the neuronal receptor for alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), or cis-4-[phosphono-methyl]-2-piperidine carboxylic acid (CGS 19755), a competitive antagonist of the N-methyl-d-aspartate (NMDA) receptor. After treatment, all animals were subjected to permanent occlusion of the middle cerebral artery for 6 or 24 h. Infarct volumes measured in animals pretreated with CGS 19755 after 24 h of ischemia were significantly smaller than those quantified in ischemic controls. Rats pretreated with NBQX showed partial amelioration of cytoskeletal injury with preserved immunolabeling of microtubule-associated protein 2 (MAP 2) at 6 and 24 h and reduced accumulation of calpain-cleaved spectrin byproducts only at 6 h. Prevention of cytoskeletal damage was more effective after pretreatment with CGS 19755, as shown by retention of MAP 2 immunolabeling and significant restriction of calpain activity at both 6 and 24 h. Preserved immunolabeling of tau protein was observed at 6 and 24 h only in animals pretreated with CGS 19755. Western analysis performed on ischemic cortex taken from controls or rats pretreated with either NBQX or CGS 19755 suggested that loss of tau protein immunoreactivity was caused by dephosphorylation, rather than proteolysis. These results demonstrate a crucial link between excitotoxic neurotransmission, microtubular proteolysis, and neuronal degeneration in focal cerebral ischemia.

Increased hyaluronic acid is associated with dermal delayed-type hypersensitivity.

Rabbits sensitized subcutaneously with heat-killed bacilli Calmette-Guerin (BCG) and challenged intradermally with heat-killed BCG or purified protein derivative (PPD) demonstrated classical dermal delayed-type hypersensitivity which peaked two days postchallenge. Animals challenged with BCG developed dermal granulomas as measured by induration and gross observation. Challenge with either PPD or BCG resulted in increased levels of dermal hyaluronic acid (HA) by two days postchallenge. Dermal HA returned to normal levels by seven days postchallenge regardless of the challenge antigen. These results indicated that increased HA is associated with dermal delayed-type sensitivity, but increased HA is not associated with dermal granulomatous hypersensitivity. These results are in contrast to previously reported work which indicates that increased HA is associated with both pulmonary delayed hypersensitivity and pulmonary granulomatous hypersensitivity.

Reduction of vesicle-associated membrane protein 2 expression leads to a kindling-resistant phenotype in a murine model of epilepsy.

Our previous work has correlated permanent alterations in the rat neurosecretory machinery with epileptogenesis. Such findings highlighted the need for a greater understanding of the molecular mechanisms underlying epilepsy so that novel therapeutic regimens can be designed. To this end, we examined kindling in transgenic mice with a defined reduction of a key element of the neurosecretory machinery: the v-SNARE (vesicle-bound SNAP [soluble NSF attachment protein] receptor), synaptobrevin/vesicle-associated membrane protein 2 (VAMP2). Initial analysis of biochemical markers, which previously displayed kindling-dependent alterations in rat hippocampal synaptosomes, showed similar trends in both wild-type and VAMP2(+/-) mice, demonstrating that kindled rat and mouse models are comparable. This report focuses on the effects that a ~50% reduction of synaptosomal VAMP2 has on the progression of electrical kindling and on glutamate release in hippocampal subregions. Our studies show that epileptogenesis is dramatically attenuated in VAMP2(+/-) mice, requiring both higher current and more stimulations to reach a fully kindled state (two successive Racine stage 5 seizures). Progression through the five identifiable Racine stages was slower and more variable in the VAMP2(+/-) animals compared with the almost linear progression seen in wild-type littermates. Consistent with the expected effects of reducing a major neuronal v-SNARE, glutamate-selective, microelectrode array (MEA) measurements in specific hippocampal subregions of VAMP2(+/-) mice showed significant reductions in potassium-evoked glutamate release. Taken together these studies demonstrate that manipulating the levels of the neurosecretory machinery not only affects neurotransmitter release but also mitigates kindling-induced epileptogenesis.

VAMP8/endobrevin is overexpressed in hyperreactive human platelets: suggested role for platelet microRNA.

BACKGROUND: Variation in platelet reactivity contributes to disorders of hemostasis and thrombosis, but the molecular mechanisms are not well understood. OBJECTIVES: To discover associations between interindividual platelet variability and the responsible platelet genes, and to begin to define the molecular mechanisms altering platelet gene expression. SUBJECTS/METHODS: Two hundred and eighty-eight healthy subjects were phenotyped for platelet responsiveness. Platelet RNA from subjects demonstrating hyperreactivity (n=18) and hyporeactivity (n=11) was used to screen the human transcriptome. RESULTS: Distinctly different mRNA profiles were observed between subjects with differing platelet reactivity. Increased levels of mRNA for VAMP8/endobrevin, a critical v-SNARE involved in platelet granule secretion, were associated with platelet hyperreactivity (Q=0.0275). Validation studies of microarray results showed 4.8-fold higher mean VAMP8 mRNA levels in hyperreactive than hyporeactive platelets (P=0.0023). VAMP8 protein levels varied 13-fold among platelets from these normal subjects, and were 2.5-fold higher in hyperreactive platelets (P=0.05). Among our cohort of 288 subjects, a VAMP8 single-nucleotide polymorphism (rs1010) was associated with platelet reactivity in an age-dependent manner (P<0.003). MicroRNA-96 was predicted to bind to the 3'-untranslated regionof VAMP8 mRNA and was detected in platelets. Overexpression of microRNA-96 in VAMP8-expressing cell lines caused a dose-dependent decrease in VAMP8 protein and mRNA, suggesting a role in VAMP8 mRNA degradation. CONCLUSIONS: These findings support a role for VAMP8/endobrevin in the heterogeneity of platelet reactivity, and suggest a role for microRNA-96 in the regulation of VAMP8 expression.

Incorporation of [3H]ethanolamine into a single cytosolic protein in a cell free system: ethanolaminylation of EF-1 alpha in vitro.

Ethanolamine containing modifications of cytosolic proteins have only been described for elongation factor 1 alpha (EF-1 alpha) which contains two ethanolamine-phosphoglycerol moieties at Glu 301 and 374 (Whiteheart et al. (1989) J. Biol. Chem. 264, 14334-14341 and Dever et al. (1989) J. Biol. Chem. 264, 20518-20525). In this report, we describe a cell-free, cytosolic extract which specifically incorporates [3H]ethanolamine into a single cytosolic protein with properties identical to EF-1 alpha. The incorporation reaction is dependent on time and extract and is independent of any membrane-bound components. The single in vitro-radiolabeled protein is modified on two sites and chromatographic analysis of pronase-digested peptides is consistent with the addition of an unmodified ethanolamine. Ethanolaminylation does not require divalent cations or ATP but is inhibited by N-ethylmaleimide and stimulated by reducing agents (beta-mercaptoethanol and dithiothreitol), indicating the requirement for free sulfhydryls. The nucleophile, hydroxylamine, at low concentrations, greatly inhibits the incorporation reaction, indicating the importance of an electrophilic center. This cytosolic extract appears to be able to carry out only the initial step in the addition of the ethanolamine-phosphoglycerol moieties to EF-1 alpha, and subsequent addition of the phosphoglycerol moiety appears to require membrane components.

SNAP-23 is a target for calpain cleavage in activated platelets.

The role of calpain in platelet function is generally associated with aggregation and clot retraction. In this report, data are presented to show that one component of the platelet secretory machinery, SNAP-23, is specifically cleaved by calpain in activated cells. Other proteins of the membrane fusion machinery, e.g. syntaxins 2 and 4 and alpha-SNAP, are not affected. In vitro studies, using permeabilized platelets, demonstrate that cleavage is time- and calcium-dependent. Analysis of SNAP-23 cleavage products suggests that the calpain cleavage site(s) is in the C-terminal third of the molecule potentially between the cysteine-rich acyl attachment sites and the C-terminal coiled-coil domain. The time course of cleavage is most consistent with late calpain-mediated events such as pp60(c-src) cleavage, but not early events such as protein-tyrosine phosphatase-1B activation. SNAP-23 cleavage is inhibited by calpeptin, calpastatin, calpain inhibitor IV, and E-64d, but not by caspase 3 inhibitor III or cathepsin inhibitor I. When tested for their effect on secretion, none of the calpain-specific inhibitors significantly affected release of soluble components from any of the three platelet granule storage pools. These results indicate that SNAP-23 cleavage occurs after granule release and therefore may play a role in affecting granule membrane exteriorization. This is consistent with the ultrastructural morphology of calpeptin-treated platelets after activation.

Sialyltransferases as specific cell surface probes of terminal and penultimate saccharide structures on living cells.

Rat liver beta-galactoside alpha-2,6-sialyltransferase and Vibrio cholerae sialidase were used, in conjunction with CMP-N-acetyl-[3H]neuraminic acid, to probe the glycoconjugate distribution, sialylation state, and level of penultimate Gal beta 1-4GlcNAc residues on the surfaces of murine thymic lymphocytes. We report a detailed characterization of this sialyltransferase-mediated labeling system. Exogenous sialylation of intact cells is dependent on transferase, sugar nucleotide donor, cell number, and incubation time. Additionally, we have demonstrated that the system labeling the cell surface is noncytotoxic and nonmetabolic and is interacting with the entire cell population. Analysis of the exosialylated structures indicates that the sialyltransferase specifically produces an alpha 2-6 linkage on N-linked oligosaccharides. Using this labeling system, we have probed the cell surface saccharide structures of murine thymocytes and demonstrated that most Gal beta 1-4GlcNAc residues are sialylated in the native state. However, one antigen, T200 (Ly-5), is strikingly undersialylated when compared to other cell surface glycoproteins (e.g., Thy 1.2). Upon analysis of exogenously sialylated oligosaccharides, labeled sialic acid was found almost exclusively on monosialylated structures with the remainder on bisialylated oligosaccharides. This suggests that the purified sialyltransferase is very precise in its recognition of oligosaccharides present on the surface of living thymic lymphocytes. This paper illustrates the combined uses of specific glycosidases and glycosyltransferases and how they can be employed in the detailed study of selected cell surface saccharide structures on living nucleated cells.

Intracellular localization of SNAP-23 to endosomal compartments.

We have reexamined the intracellular localization of the ubiquitously expressed target membrane SNAP receptor (t-SNARE), SNAP-23. While SNAP-23 appears on the plasma membrane, in the cell types examined there is a significant pool associated with endosomal compartments. Immuno-staining and expression of green fluorescent protein-tagged SNAP-23, show that it has a punctate, perinuclear localization in HepG2 and HT4 cells. This distribution overlaps significantly with transferrin receptor and slightly with the late endosome/lysosomal protein LAMP-1. The localization of SNAP-23 changes as HepG2 cells polarize. Initially it is concentrated at sites of cell-cell contact and then almost exclusively to the apical (or bile canalicular) domain of the cell. These data are consistent with a role for SNAP-23 in both endosome-plasma membrane trafficking as well as endosome-endosome transport.

Identification of a cellubrevin/vesicle associated membrane protein 3 homologue in human platelets.

Several studies suggest membrane trafficking events are mediated by integral, membrane proteins from both transport-vesicle and target membranes, called v- and t-SNAREs (SNAp REceptors), respectively. Previous experiments using antibodies to synaptobrevin/vesicle associated membrane protein (VAMP) 1, 2, or rat cellubrevin failed to detect these v-SNAREs in human platelets, although membrane proteins from these cells could support 20S complex formation. To identify v-SNAREs in platelets, we used a polymerase chain reaction (PCR) approach with degenerate primers to amplify potential VAMP-like v-SNAREs. A cDNA encoding a novel v-SNARE was isolated from a human megakaryocyte cDNA library. Termed human cellubrevin (Hceb), this protein has greater than 93% identity with human VAMP 1, 2, and rat cellubrevin over the conserved core region, but has a unique N-terminal domain. Northern blot analysis showed that the 2. 5-kB mRNA encoding Hceb is expressed in every human tissue tested. Hceb from detergent-solubilized platelet membranes, participated in alpha-SNAP-dependent 20S complex formation and adenosine triphosphate (ATP)-dependent disassembly, showing that Hceb can act as a v-SNARE in platelets. Immunofluorescence microscopy, using an anti-Hceb antibody showed a punctate, intracellular staining pattern in platelets, megakaryocytes, and HEK-293 cells. This same pattern was observed in surface-activated platelets even though all dense core and most alpha-granule contents had been released. These data suggest that Hceb may reside on a platelet organelle that is not primarily involved in the exocytic pathway.