Cysteine residues of SNAP-25 are required for SNARE disassembly and exocytosis, but not for membrane targeting.

The release of neurotransmitter at a synapse occurs via the regulated fusion of synaptic vesicles with the plasma membrane. The fusion of the two lipid bilayers is mediated by a protein complex that includes the plasma membrane target soluble N-ethylmaleimide-sensitive fusion protein (NSF) attachment protein (SNAP) receptors (t-SNAREs), syntaxin 1A and synaptosome-associated protein of 25 kDa (SNAP-25), and the vesicle SNARE (v-SNARE), vesicle-associated membrane protein (VAMP). Whereas syntaxin 1A and VAMP are tethered to the membrane by a C-terminal transmembrane domain, SNAP-25 has been suggested to be anchored to the membrane via four palmitoylated cysteine residues. We demonstrate that the cysteine residues of SNAP-25 are not required for membrane localization when syntaxin 1A is present. Analysis of the 7 S and 20 S complexes formed by mutants that lack cysteine residues demonstrates that the cysteines are required for efficient SNARE complex dissociation. Furthermore, these mutants are unable to support exocytosis, as demonstrated by a PC12 cell secretion assay. We hypothesize that syntaxin 1A serves to direct newly synthesized SNAP-25 through the Golgi transport pathway to the axons and synapses, and that palmitoylation of cysteine residues is not required for targeting, but to optimize interactions required for SNARE complex dissociation.

SNAP-25 with mutations in the zero layer supports normal membrane fusion kinetics.

Considerable data support the idea that intracellular membrane fusion involves a conserved machinery containing the SNARE proteins. SNAREs assembled in vitro form a stable 4-helix bundle and it has been suggested that formation of this complex provides the driving force for bilayer fusion. We have tested this possibility in assays of exocytosis in cells expressing a botulinum neurotoxin E (BoNT/E)-resistant mutant of SNAP-25 in which additional disruptive mutations have been introduced. Single or double mutations of glutamine to glutamate or to arginine in the central zero layer residues of SNAP-25 did not impair the extent, time course or Ca2+-dependency of exocytosis in PC12 cells. Using adrenal chromaffin cells, we found that exocytosis could be reconstituted in cells transfected to express BoNT/E. A double Q-->E mutation did not prevent reconstitution and the kinetics of single granule release events were indistinguishable from control cells. This shows a high level of tolerance of changes in the zero layer indicating that the conservation of these residues is not due to an essential requirement in vesicle docking or fusion and suggests that formation of a fully stable SNARE complex may not be required to drive membrane fusion.

Botulinum neurotoxin E-insensitive mutants of SNAP-25 fail to bind VAMP but support exocytosis.

Neurotransmitter release from synaptic vesicles is mediated by complex machinery, which includes the v- and t-SNAP receptors (SNAREs), vesicle-associated membrane protein (VAMP), synaptotagmin, syntaxin, and synaptosome-associated protein of 25 kDa (SNAP-25). They are essential for neurotransmitter exocytosis because they are the proteolytic substrates of the clostridial neurotoxins tetanus neurotoxin and botulinum neurotoxins (BoNTs), which cause tetanus and botulism, respectively. Specifically, SNAP-25 is cleaved by both BoNT/A and E at separate sites within the COOH-terminus. We now demonstrate, using toxin-insensitive mutants of SNAP-25, that these two toxins differ in their specificity for the cleavage site. Following modification within the COOH-terminus, the mutants completely resistant to BoNT/E do not bind VAMP but were still able to form a sodium dodecyl sulfate-resistant complex with VAMP and syntaxin. Furthermore, these mutants retain function in vivo, conferring BoNT/E-resistant exocytosis to transfected PC12 cells. These data provide information on structural requirements within the C-terminal domain of SNAP-25 for its function in exocytosis and raise doubts about the significance of in vitro binary interactions for the in vivo functions of synaptic protein complexes.

On the action of botulinum neurotoxins A and E at cholinergic terminals.

Botulinum neurotoxins type A and E (BoNT/A and /E) are metalloproteases with a unique specificity for SNAP-25 (synaptosomal-associated protein of 25 kDa), an essential protein component of the neuroexocytotic machinery. It was proposed that this specificity is based on the recognition of a nine-residue sequence, termed SNARE motif, which is common to the other two SNARE proteins: VAMP (vesicle-associated membrane protein) and syntaxin, the only known substrates of the other six clostridial neurotoxins. Here we report on recent studies which provide evidence for the involvement of the SNARE motif present in SNAP-25 in its interaction with BoNT/A and /E by following the kinetics of proteolysis of SNAP-25 mutants deleted of SNARE motifs. We show that a single copy of the motif is sufficient for BoNT/A and /E to recognise SNAP-25. While the copy of the motif proximal to the cleavage site is clearly involved in recognition, in its absence, other more distant copies of the motif are able to support proteolysis. We also report on studies of poisoning human neuromuscular junctions with either BoNT/A or BoNT/E and describe the unexpected finding that the time of recovery of function after poisoning is much shorter in the case of type E with respect to type A intoxication. These data are discussed in terms of the different sites of action of the two toxins within SNAP-25.

In vitro biological activity and toxicity of tetanus and botulinum neurotoxins.

Tetanus and botulinum neurotoxins are the most potent toxins known and cause tetanus and botulism, respectively. They are zinc-endopeptidases acting in the cytosol, where they cleave SNARE proteins. Here, we report on the assay of their metalloproteolytic activity in vitro on recombinant SNARE proteins. We also describe the assay of their activity in nerve cells in culture using antibodies specific for the SNARE proteins. Together with recent reports from other laboratories, these results show that the toxicity of these powerful neurotoxins can be appropriately assayed in vitro, thus reducing considerably the number of animals currently used in the evaluation of the toxicity of tetanus toxoid vaccine and of the botulinum neurotoxins to be used for human therapy.

Botulinum neurotoxin types A and E require the SNARE motif in SNAP-25 for proteolysis.

Botulinum neurotoxins type A and E (BoNT/A and BoNT/E) are metalloproteases with a unique specificity for SNAP-25 (synaptosome-associated protein of 25 kDa), an essential protein component of the neuroexocytotic machinery. It has been suggested that this specificity is directed through the recognition of a nine residue sequence, termed SNARE motif, that is common to the other two SNARE proteins: VAMP (vesicle-associated membrane protein) and syntaxin, the only known substrates of the other six clostridial neurotoxins. Here we analyse the involvement of the four copies of the SNARE motif present in SNAP-25 in its interaction with BoNT/A and BoNT/E by following the kinetics of proteolysis of SNAP-25 mutants deleted of SNARE motifs. We show that a single copy of the motif is sufficient for BoNT/A and BoNT/E to recognise SNAP-25. While the copy of the motif proximal to the cleavage site is clearly involved in recognition, in its absence, other more distant copies of the motif are able to support proteolysis. Also, a non-neuronal isoform of SNAP-25, Syndet, is shown to be sensitive to BoNT/E, but not BoNT/A, whilst the SNAP-25 isoforms from Torpedo marmorata and Drosophila melanogaster were demonstrated not to be substrates of these metalloproteases.

Vesicle-associated membrane protein-2 (synaptobrevin-2) forms a complex with synaptophysin.

Vesicle-associated membrane protein (VAMP) (or synaptobrevin), a type II membrane protein of small synaptic vesicles, is essential for neuroexocytosis because its proteolysis by tetanus and botulinum neurotoxins types B, D, F and G blocks neurotransmitter release. The addition of cross-linking reagents to isolated small synaptic vesicles induces the formation of 30 and 50 kDa complexes containing the isoform 2 of VAMP (VAMP-2). Whereas the 30 kDa band is a VAMP-2 homodimer, the 50 kDa species results from the cross-linking of VAMP-2 with synaptophysin. This heterodimer also forms in detergent-solubilized vesicles and involves the N-terminal part of VAMP-2. The implications of the existence of a synaptophysin-VAMP-2 complex in the processes of vesicle docking and fusion with the presynaptic membrane are discussed.