SNAP-29: a general SNARE protein that inhibits SNARE disassembly and is implicated in synaptic transmission

Abstract

Using the yeast two-hybrid system with syntaxin-1A as bait, we isolated soluble NSF attachment protein (SNAP)-29 from a human brain cDNA library. Synaptosomal fractionation and immunocytochemical staining of hippocampal neurons in culture showed that SNAP-29 is present at synapses and is predominantly associated with synaptic vesicles. The interaction of SNAP-29 with syntaxin-1 was further confirmed with immunoprecipitation analysis. Binding competition studies with SNAP-29 demonstrated that it could compete with alpha-SNAP for binding to synaptic SNAP receptors (SNAREs) and consequently inhibit disassembly of the SNARE complex. Introduction of SNAP-29 into presynaptic superior cervical ganglion neurons in culture significantly inhibited synaptic transmission in an activity-dependent manner. Although SNAP-29 has been suggested to be a general SNARE component in membrane trafficking, our findings suggest that it may function as a regulator of SNARE complex disassembly and modulate the process of postfusion recycling of the SNARE components.

Figure 1

Subcellular distribution of SNAP-29 in synaptosomal fractions. (A) Crude synaptosomes (SS) were separated into fractions enriched in presynaptic cytosol, synaptic vesicles, and synaptosomal plasma membrane by sucrose gradient centrifugation. Ten micrograms of synaptosomal fractions on SDS/PAGE was sequentially immunoblotted with antibodies, as indicated, after stripping between antibody applications. (B) Synaptic vesicles purified by anti-synaptotagmin-conjugated affinity beads (anti-tagmin 1) were analyzed by SDS/PAGE and sequential immunoblotting with antibodies as indicated. Normal IgG-conjugated beads (IgG) were applied as a control.

Figure 2

SNAP-29 is present in the synapses of cultured hippocampal neurons. Hippocampal cultures were immunostained with antibodies against SNAP-29 (green) (A) and synaptophysin (red) (B). Double punctate staining on the neuronal processes is marked by arrowheads (C).

Figure 3

Association of SNAP-29 with syntaxin-1 and SNARE complex. (A) Immunoprecipitation of syntaxin-1A with SNAP-29 from cotransfected HEK293 T cells. The syntaxin-1A–SNAP-29 complex was immunoprecipitated from cell extracts using anti-SNAP-29 antiserum (anti-29) followed by immunoblot with antibody to syntaxin-1. Normal rabbit IgG (IgG) was used as a control. (B) Immunoprecipitation of syntaxin-1 with SNAP-29 from solubilized rat brain homogenates. Syntaxin-1 was immunoprecipitated by anti-SNAP-29 antiserum and by anti-VAMP, but not by preimmune serum (preserum) or normal rabbit IgG (IgG). A 10% volume of brain homogenate (b.h.) was loaded (right lane). (C) Immobilized GST–SNAP-29 (GST-29) or GST alone on Sepharose beads was incubated with 100 μg of rat brain homogenates at 4°C for 3 h. Native SNARE complex captured was subjected to SDS/PAGE at temperatures as indicated and then analyzed by immunoblotting with the antibodies against syntaxin-1, SNAP-25, and VAMP-2. (Lower) Immunoblotting on the same membrane with anti-GST antibody.

Figure 4

SNAP-29 competes with α-SNAP for binding to SNAREs. (A) Immobilized GST–syntaxin-1A (stx-1A) or GST–stx-1A–SNAP-25 heterodimers (stx 1A-SNAP-25) and ternary SNARE complex (SNARE) on glutathione-Sepharose were incubated with 1 μg of recombinant α-SNAP in the presence (5 μg) or absence of His-SNAP-29. Bound α-SNAP was visualized by anti-α-SNAP antibody. (B) Concentration-dependent competition of SNAP-29 with α-SNAP for binding to immobilized GST–stx-1A. (C) Immobilized GST–syntaxin-1A was incubated with ≈1 μmol of His-tagged SNAP-25 and VAMP-2 to form recombinant SNARE complex in vitro, and the beads were then washed extensively. The preformed SNARE complex was then incubated with His-tagged α-SNAP (0.16 μmol) in the presence or absence of 0.8 μmol of His-SNAP-29 or His-SNAP-25 as indicated. Protein complexes were analyzed by SDS/PAGE and visualized by sequential immunoblotting. (D) The relative levels of α-SNAP binding to the SNARE complex were calculated based on a linear standard curve of α-SNAP aliquots by using National Institutes of Health image scanning. Semiquantitative analysis revealed that normalized percentage of α-SNAP binding to SNARE complex in the presence of 0.8 μmol of His-SNAP-29 (n = 5) or His-SNAP-25 (n = 3) relative to control in the absence of both SNAP-29 and SNAP-25. Data are expressed as mean ± SEM. Result marked as * (P < 0.01) was considered significantly different.

Figure 5

SNAP-29 reduces disassembly of preformed SNARE core complex. (A) Recombinant SNARE core complex was constituted by incubation of GST–VAMP-2 (1 μg) bound glutathione-Sepharose beads with purified syntaxin-1A and SNAP-25 (5 μg each) at 4°C for 3 h. (B) Native SNARE complexes were isolated from solubilized synaptosomes (50 μg) by anti-VAMP-2 antibody. After washing to remove unbound SNARE proteins, purified His-SNAP-29 (10 μg) was added into the preformed SNARE complexes and further incubated at room temperature for 1 h. SNARE core complexes were disassembled by addition of recombinant NSF and α-SNAP (2 μg each) in buffer with 2 mM MgCl₂/2.5 mM ATP/20 mM Tris, pH 8.4/100 mM NaCl for 30 min at 30°C. As a control, the ATPase activity of NSF was abolished by chelating MgCl₂ with 10 mM EDTA. All samples were heated at either 80°C or 100°C for 5 min and analyzed by SDS/PAGE and sequential immunoblotting with antibodies against SNAP-25 and His tag.

Figure 6

A synaptic activity-dependent effect of SNAP-29 on synaptic transmission of cultured SCGNs. (A) A typical trace of EPSPs with SNAP-29 injection into the presynaptic neuron. The pipette concentration of the protein was 170 μM. The presynaptic neuron was stimulated every 20 s (0.05 Hz). EPSPs from one representative experiment recorded 3 min before injection and 42 and 78 min after injection are illustrated. (B) A typical trace of EPSPs with 2.5 mM SNAP-25 (196) peptide injection. EPSPs from one representative experiment recorded 3 min before injection and 33 and 64 min after injection are illustrated. (C) Normalized EPSP amplitudes with SNAP-29 injection were averaged and plotted from experiments with presynaptic stimulation once every 20 s (0.05 Hz) (n = 6; ◊) or every 5 s (0.2 Hz) (n = 4; □). As a control, heat-denatured full-length SNAP-29 did not show any significant effects for 60 min after injection at a stimulation frequency of 0.2 Hz (n = 4; ●). (D) Normalized EPSP amplitudes with 2.5 mM SNAP-25 (196) or SNAP-25 (146) peptide injection were averaged and plotted from experiments with presynaptic stimulation as indicated. (E) Normalized EPSP amplitudes with 170 μM SNAP-29 injection at 0.02-Hz stimulation (●) and with 2.5 mM SNAP-25 (196) peptide injection at 0.01-Hz stimulation (○). (F) Normalized EPSP amplitudes with coinjection of SNAP-29 (170 μM) and α-SNAP (43 or 68 μM; □ or ◊; n = 4) at 0.2-Hz stimulation. As a control, α-SNAP (68 μM) was injected at 0.05-Hz stimulation (n = 7; ●).