Critical to SNARE protein function in neurotransmission are the accessory proteins, soluble N-ethylmaleimide-sensitive factor (NSF) attachment protein (SNAP), and NSF, that play a role in activation of the SNAREs for membrane fusion. In this report, we demonstrate the depolarization-induced, calcium-dependent phosphorylation of NSF in rat synaptosomes. Phosphorylation of NSF is coincident with neurotransmitter release and requires an influx of external calcium. Phosphoamino acid analysis of the radiolabeled NSF indicates a role for a serine/threonine-specific kinase. Synaptosomal phosphorylation of NSF is stimulated by phorbol esters and is inhibited by staurosporine, chelerythrine, bisindolylmaleimide I, calphostin C, and Ro31-8220 but not the calmodulin kinase II inhibitor, Kn-93, suggesting a role for protein kinase C (PKC). Indeed, NSF is phosphorylated by PKC in vitro at Ser-237 of the catalytic D1 domain. Mutation of this residue to glutamic acid or to alanine eliminates in vitro phosphorylation. Molecular modeling studies suggest that Ser-237 is adjacent to an inter-subunit interface at a position where its phosphorylation could affect NSF activity. Consistently, mutation of Ser-237 to Glu, to mimic phosphorylation, results in a hexameric form of NSF that does not bind to SNAP-SNARE complexes, whereas the S237A mutant does form complex. These data suggest a negative regulatory role for PKC phosphorylation of NSF.
The molecular mechanisms of neurotransmitter (NT)1 release have been the subject of much attention in recent years resulting in an evolving model known as the SNARE hypothesis (1-3). The SNARE hypothesis holds that membrane proteins in the vesicle (v-SNAREs, e.g. synaptobrevins) bind to a heterodimer in the target membrane (t-SNAREs, heterodimers of syntaxins and SNAP-25-like proteins). v-and t-SNAREs bind to form a 7 S complex (1, 4), composed of a bundle of four parallel, coiled-coil domains (5-8) that, through reconstitution studies, has been demonstrated to be minimally required for bilayer fusion (9). The N-ethylmaleimide-sensitive factor (NSF) and the soluble NSF attachment proteins (SNAPs, not to be confused with SNAP-25) affect the composition and structure of the SNARE complex. SNAPs act as adapters and are required for binding of NSF to the 7 S complex. ATP hydrolysis by NSF causes the resulting 20 S complex to disassemble into monomeric SNAREs (4, 10), a step required for vesicle trafficking (11-13). Detailed kinetic experiments have placed one role of NSF and SNAPs at steps prior to v-/t-SNARE binding (14 -16), suggesting a model in which NSF disassembles cis 7 S complexes that exist in the same bilayer (17-19). The resulting monomeric SNAREs can then form trans 7 S complexes that span the opposing bilayers of the vesicle and target membrane. In this manner, NSF acts as a chaperone to activate or "prime" the SNARE proteins for subsequent trans complex formation and membrane fusion.Studies by Schweizer et al. (20), using NSF-derived peptides microinjected pre-synaptically, showed that ...