Munc18-1 redistributes in nerve terminals in an activity- and PKC-dependent manner

Abstract

Munc18-1 is a soluble protein essential for synaptic transmission. To investigate the dynamics of endogenous Munc18-1 in neurons, we created a mouse model expressing fluorescently tagged Munc18-1 from the endogenous munc18-1 locus. We show using fluorescence recovery after photobleaching in hippocampal neurons that the majority of Munc18-1 trafficked through axons and targeted to synapses via lateral diffusion together with syntaxin-1. Munc18-1 was strongly expressed at presynaptic terminals, with individual synapses showing a large variation in expression. Axon-synapse exchange rates of Munc18-1 were high: during stimulation, Munc18-1 rapidly dispersed from synapses and reclustered within minutes. Munc18-1 reclustering was independent of syntaxin-1, but required calcium influx and protein kinase C (PKC) activity. Importantly, a PKC-insensitive Munc18-1 mutant did not recluster. We show that synaptic Munc18-1 levels correlate with synaptic strength, and that synapses that recruit more Munc18-1 after stimulation have a larger releasable vesicle pool. Hence, PKC-dependent dynamic control of Munc18-1 levels enables individual synapses to tune their output during periods of activity.

Figure 1.

Generation and confirmation of Munc18-1-Venus (M18V) mouse. (A) Generation of M18V knockin gene. Diagrams indicate WT munc18-1 gene, targeting vector, M18V-neo-knock-in gene, and Cre-recombined M18V gene. Exons are indicated by gray boxes and numbered. Black and gray horizontal bars indicate probes used for Southern blot analysis and PCR products used for genotyping, respectively. A 5-kbp section is not shown (arrowheads). LoxP, loxP sites; Venus, Venus cDNA; AvrII, restriction enzyme site; NEO, neomycin resistance gene; TK, thymidine kinase promoter; pGEM-T Easy, targeting vector. (B) Southern blot analysis of mouse tail DNA from heterozygous (+/m) and WT (+/+) mice. DNA was AvrII-digested. m, M18V-neo-knock-in gene (8.2 kbp); +, munc18-1 gene (6.3 kbp). (C) Agarose gel of PCR products from +/+, +/m, and m/m mouse DNA. M18V, munc18-1-Venus gene PCR product, 449 bp; M18, munc18-1 gene PCR product, 195 bp. (D) Western blot of brain lysate of +/+, +/m, and m/m mice stained for M18V. Actin was used as a loading control. The lines indicate that intervening lanes have been spliced out. (E) Western blot of brain region lysate of m/m and +/+ mice at E18 stained for M18V. Valosin-containing protein (VCP) was used as a loading control. (F) Quantification of M18V expression levels in +/+ and m/m brain lysate. Homogenates from +/+ and m/m brains were analyzed by SDS-PAGE (10 µg of protein per lane) and Western blotting with Munc18-1 antibodies. M18V protein levels were normalized to VCP protein levels for each mouse (n = 6). Error bars indicate mean ± SEM. (G) Hippocampal localization of M18V fluorescence in m/m (M18V homozygote) brain slice (left) and of antibody-stained Munc18-1 in WT (+/+) mice (right) compared with synaptic marker VAMP and dendritic marker MAP2. CA3, hippocampal CA region; Str, striatum; arrowheads, mossy fiber terminals of the stratum lucidum. Bars: (overview images) 100 µm; (enlarged panels) 25 µm.

Figure 2.

Morphology and synaptic transmission of Munc18-1-Venus (M18V) hippocampal neurons is indistinguishable from WT neurons. Shown are the characterization of neuron morphology (A–D) and electrophysiology (E–K) of WT (gray) and M18V (green) hippocampal autapse cultures. (A) Dendrite length was measured using MAP2 staining at 1–2 DIV, 3–4 DIV, and 7–8 DIV (M-W test: WT 1–2 DIV, n = 26 cells; M18V 1–2 DIV, n = 44 cells, P = 0.049; WT 3–4 DIV, n = 27 cells; M18V 3–4 DIV, 27 cells, P > 0.05; WT 7–8 DIV, n = 26 cells; M18V 7–8 DIV, 32 cells, P > 0.05). (B) Axon length measured using a cocktail of Smi-312 and Ankyrin G staining at 1–2 DIV, 3–4 DIV, and 7–8 DIV (M-W: WT 1–2 DIV, n = 26 cells; M18V 1–2 DIV, n = 44 cells, P > 0.05; WT 3–4 DIV, n = 27 cells; M18V 3–4 DIV, n = 27 cells, P > 0.05; WT 7–8 DIV, n = 26 cells; M18V 7–8 DIV, n = 32 cells, P > 0.05). (C) Dendrite length measured using MAP2 staining at DIV 14 (M-W test: WT, n = 32 cells; M18V, n = 27 cells, P > 0.05). (D) Synapse number per cell measured using synapsin staining at 14 DIV (unpaired t test: WT, n = 13 cells; M18V, n = 13 cells, P > 0.05). (E) Single evoked EPSC amplitudes (WT, 6.4 ± 0.9 nA, n = 22; M18V, 6.8 ± 1.0 nA, n = 24; M-W test, P > 0.05). (F) Normalized short-term plasticity (STP) curves with 5 EPSCs in 1 Hz (left), 10 Hz (middle), and 50 Hz (right) interstimulus intervals (ISI). (insets) Paired pulse ratios (PPRs; WT 1 Hz, 0.9180 ± 0.025, n = 14; M18V 1 Hz, 0.921 ± 0.017, n = 18; WT 10 Hz, 0.769 ± 0.049, n = 12; M18V 10 Hz 0.804 ± 0.041, n = 19; WT 50 Hz, 0.557 ± 0.044, n = 15; M18V 50 Hz, 0.514 ± 0.046, n = 19; M-W test, P > 0.05 for all). (G) Example traces for STP experiments at 50 Hz (top) and 1 Hz (bottom). (H) Frequency (left) and amplitude (right) of spontaneous miniature events (mEPSC; WT frequency, 13.1 ± 2.1 Hz, n = 19; M18V frequency, 13.5 ± 2.2 Hz, n = 20; WT amplitude, 27.4 ± 1.6 pA; M18V amplitude, 28.9 ± 1.4 pA; M-W test, P > 0.05). (I) Cumulative EPSC amplitudes in a 100 AP, 40 Hz train with readily releasable pool size estimation via steady-state back extrapolation (inset: pool size in WT, 41.9 ± 0.1 nA, n = 13; M18V, 42.1 ± 0.2 nA, n = 14; M-W test, P > 0.05). (J) RRP recovery after train depletion (using 100 AP at 40 Hz) and a test pulse with varying ISIs (Recovery tau: WT, 665 ms, n = 20; M18V, 659 ms, n = 15; M-W test, P > 0.05). (K) Example traces for spontaneous miniature EPSCs (top) and 100 AP, 40 Hz stimulus train with a recovery test pulse (ISI 400 ms; bottom). Error bars indicate mean ± SEM.

Figure 3.

Munc18-1-Venus dynamics in axons depend on syntaxin-1. (A) Distal axon of WT neuron expressing GFP (green) used in FRAP experiments. Inverted greyscale images show time points before (pre) and immediately after photobleaching (t = 0 s), and during fluorescence recovery (post t = 2 s, post t = 34 s, and post t = 158 s). Arrowheads indicate the region of photobleaching. (B) Fluorescence image (green) and inverted greyscale images (same time points as in A) showing FRAP of a distal axon of WT neuron expressing syntaxin-1-EYFP (Stx-YFP). Note the difference in speed and the extent of recovery between soluble GFP and membrane-bound Stx-YFP. (C) Distal axon from a Munc18-1-Venus (M18V) neuron shown in differential interference contrast (greyscale), a fluorescence image (green), and a combined image (composite). Inverted greyscale images show time points before (pre) and immediately after bleaching (t = 0 s), and during fluorescence recovery (post t = 34 s and post t = 158 s). (D) Kymograph (top) and FRAP analysis (bottom) of the axon in C. The kymograph shows fluorescence recovery from both sides of the unbleached area. Fluorescence intensity was measured in the middle of the bleached area (broken lines) and normalized to prebleach and t = 0 s (see Materials and methods). (E) FRAP analysis of axonal M18V, Stx-YFP, and GFP. There were no statistically significant differences between M18V and Stx-YFP at t = 10 s, 30 s, and 160 s (M-W test with FDR [6] corrections: M18V, n = 7 cells, n = 15 axons, bleach length = 23.7 ± 1.3 µm; Stx-YFP, n = 16 field of views, n = 16 axons, bleach length = 25.2 ± 1.0 µm; GFP, n = 9 field of views, n = 9 axons, bleach length = 40.4 ± 2.9 µm). (F) FRAP analysis of axonal M18V (see E) and M18V + BoNT/C. (inset) Statistical significance (M-W test with FDR [6] corrections: ***, P < 0.001; **, P < 0.01; M18V + BoNT/C, n = 7 cells, n = 19 axons, bleach length = 25.7 ± 1.1 µm). Bars, 5 µm.

Figure 4.

Syntaxin-1 binding is the major determinant of an immobile Munc18-1 pool at synapses. (A) Munc18-1-Venus (M18V) axon expressing synapsin-mCherry. Overlay of M18V and synapsin-mCherry fluorescence (Composite) shows Munc18-1-Venus expression at presynaptic sites. Bar, 4 µm. (B) Normalized fluorescence intensities of Stx-YFP and mVenus in WT neurons and M18V measured from the center of synapsin-mCherry positive puncta up to 4 µm into the axon (left; see scale bar in A). Intensity was normalized to axon intensity at 4 µm. Right, synapse (0 µm)-to-axon ratio (4 µm) of M18V, Stx-YFP, and mVenus (K-W test: ***, P < 0.001; n.s., P > 0.05; M18V, n = 10 field of views, n = 72 synapses; Stx-YFP, n = 15 field of views, n = 58 synapses; mVenus, n = 12 field of views, n = 52 synapses). (C) Diagram illustrating FRAP experiments shown in D–G. M18V neurons (green) expressing synapsin-mCherry as a synapse marker (not depicted) were cultured together with WT neurons (gray) in a 1:50 ratio. Fluorescence in synapses was photobleached (arrowheads and box) at t = 0 s, and fluorescence recovery was followed over time (FRAP, right). (D) M18V axon with three synapses (arrowheads) identified by synapsin-mCherry. Inverted greyscale images show time points before (pre), immediately after photobleaching (bleach, open arrowheads), and during fluorescence recovery (post t = 22 s and post t = 123 s). Bottom right, normalized intensity of bleached synapse over time. Bar, 5 µm. (E) FRAP analysis of synaptic M18V, GFP, and Stx-YFP. (inset) Statistical significance (M-W test with FDR [6] corrections: n.s., P > 0.05; *, P < 0.05; M18V, n = 14 field of views, n = 37 synapses; Stx-YFP, n = 9 field of views, n = 23 synapses; GFP, n = 9 field of views, n = 28 synapses). The broken line indicates a biexponential fit of M18V recovery with a fast mobile fraction of 0.32 and a tau of 4.6 s, and a slow mobile fraction of 0.36 and a tau of 59.8 s. (F) FRAP analysis of synaptic M18V (see E) and M18V + BoNT/C. (inset) Statistical significance (M-W test with FDR [6] corrections: **, P < 0.01; *, P < 0.05; M18V + BoNT/C, n = 6 cells, n = 16 synapses). (G) Mobile fraction of synaptic M18V compared with M18V + BoNT/C calculated from F at t = 180 s (M-W test: ***, P < 0.001). Error bars indicate mean ± SEM.

Figure 5.

Activity increases Munc18-1 and Syntaxin-1 mobility at synapses. (A) FRAP analysis of Munc18-1-Venus (M18V) in control synapses and synapses stimulated with 600 AP at 20 Hz starting at t = −15 s (black bar). (inset) Statistical significance (M-W test with FDR [3] corrections: n.s., P > 0.05; *, P < 0.05; M18V, n = 8 field of views, n = 19 synapses; M18V + stimulation, n = 8 field of views, n = 16 synapses). (B) Mobile fraction of synaptic M18V, control compared with stimulation calculated from A at t = 160 s (M-W test: **, P < 0.01). (C) FRAP analysis of Stx-YFP in control synapses and synapses stimulated with 600 AP at 20 Hz starting at t = −15 s (black bar). (inset) Statistical significance (M-W test with FDR [3] corrections: n.s., P > 0.05; Stx-YFP, n = 6 field of views, n = 16 synapses; Stx-YFP + stimulation, n = 6 field of views, n = 13 synapses). (D) Mobile fraction of synaptic Stx-YFP control compared with stimulation calculated from C at t = 160 s (M-W test: *, P < 0.05). Error bars indicate mean ± SEM.

Figure 6.

Synaptic Munc18-1-Venus acutely disperses upon stimulation. (A) Fluorescence and greyscale images of a Munc18-1-Venus (M18V) axon (right) expressing synapsin-mCherry (left). Greyscale images show the same axon before (pre) and during stimulation (stim) with 600 AP at 20 Hz. A linescan (bottom) along the axon compares fluorescence intensity in synapses marked with arrowheads of synapsin-mCherry before (pre, black broken line) with during (stim, red solid line) stimulation and intensity of M18V before (pre, black dashed line) with during (stim, green solid line) stimulation. Bar, 5 µm. (B) Relative intensity changes (ΔF/F0) of M18V, Stx-YFP, mVenus, and synapsin-mCherry (inset) at synapses during stimulation starting at t = 0 s (20 Hz, black bar). M18V and synapsin-mCherry, n = 38 field of views, n = 929 synapses; Stx-YFP, n = 17 field of views, n = 202 synapses; mVenus, n = 20 field of views, n = 305 synapses. (C) Relative intensity changes at t = 10 s of M18V, Stx-YFP, and mVenus calculated from B (K-W test: ***, P < 0.001). (D) Relative intensity changes (ΔF/F0) of M18V, M18V + BoNT/C, and synapsin-mCherry (inset) in synapses during stimulation starting at t = 0 s (20 Hz, black bar). M18V and synapsin-mCherry, n = 15 cells, n = 533 synapses; M18V + BoNT/C, n = 8 cells, n = 290 synapses. (E) Relative intensity changes of M18V and M18V + BoNT/C at t = 4 s calculated from D (M-W test: n.s., P > 0.05). Error bars indicate mean ± SEM. The horizontal broken line indicates no change.

Figure 7.

Synaptic activity increases Munc18-1-Venus mobility and recruits Munc18-1 to synapses independent of syntaxin-1. (A) Fluorescence and greyscale images of Munc18-1-Venus (M18V) axons expressing synapsin-mCherry. Greyscale images show the same axon before (pre), during (t = 12 s), and after stimulation (t = 198 s, M18V only) with 600 AP at 20 Hz (start t = 0 s). Bar, 5 µm. (B) Relative M18V intensity changes (ΔF/F0) in synapses marked by arrowheads in A. Black bars indicate the period of stimulation. Some synapses, after initial dispersion during stimulation, increase fluorescence above initial levels (broken line), whereas others remain low after initial dispersion or do not change during measurement. (C) Relative M18V intensity changes of all synapses (green line, n = 23 field of views, n = 510 synapses) compared with subsets (gray lines) of synapses with net positive change (Δ+, 46.3% of all synapses) and synapses with net negative change (Δ−, 37.8%) at t = 160 s. (D) Relative Stx-YFP intensity changes of all synapses (blue line, n = 10 field of views, n = 125 synapses) and of subsets (gray line) of synapses with net positive change (Δ+, 22.4% of all synapses) or a net negative change (Δ−, 36.0%) at t = 160 s. (E) Relative M18V intensity changes of all synapses (green, n = 5 cells, n = 553 synapses) and M18V ± BoNT/C synapses (yellow, n = 5 cells, n = 290 synapses). (inset) Percentage of synapses with net positive change (−BoNT/C, 38.6%; +BoNT/C, 44.1%) and synapses with net negative change (−BoNT/C, 31.9%; +BoNT/C, 19.3%) at t =160 s (Pearson χ² test: ***, P < 0.001). (F) Relative M18V intensity change at t = 10 s (left) and t = 160 s (right) at control synapses (green) and synapses expressing BoNT/C (yellow) calculated from E (M-W test with FDR [2] corrections: ***, P < 0.001). (G) Relative M18V and synapsin-mCherry (inset) intensity changes of control synapses (green, n = 6 cells, n = 250 synapses) and synapses superfused with calcium channel blockers to prevent calcium influx (brown, n = 6 cells, n = 252 synapses). Synapsin-mCherry intensity at t = 30 s, control versus calcium channel blockers, M-W test: ***, P < 0.001). (H) Relative M18V intensity change at t = 10 s (left) and t = 160 s (right) at control synapses (green) and synapses superfused with calcium channel blockers (brown) calculated from G (M-W test with FDR [2] corrections: n.s., P > 0.05; ***, P < 0.001). Note that although average dispersion (t = 10 s) in control synapses is lower compared with E, Fig. S5 shows that dispersion in Δ+ and Δ− control synapses is significantly higher than in synapses with calcium channel blockers. The horizontal broken line indicates no change.

Figure 8.

PKC phosphorylation of Munc18-1 is necessary for synaptic recruitment. (A) Relative Munc18-1-Venus (M18V) intensity changes (ΔF/F0) at control synapses (green, n = 9 field of views, n = 57 synapses) and synapses superfused with 1 µM PMA (blue, n = 6 field of views, n = 52 synapses) for 30 s at t = 0 s (black bar). (B) Relative M18V intensity change at t = 10 s (left) and t = 160 s (right) at control synapses and synapses with PMA calculated from A (M-W test with FDR [2] corrections: n.s., P > 0.05). (C) Relative M18V intensity changes after stimulation with 600 AP at 20 Hz starting at t = 0 s (black bar) at control synapses (green, n = 2 field of views, n = 43 synapses) and synapses with 1 µM PMA (blue, n = 7 field of views, n = 112 synapses) for >2 min (bath application). (D) Relative M18V intensity change at t = 10 s (left) and t = 160 s (right) at control synapses and synapses with PMA calculated from C (M-W test with FDR [2] corrections: n.s., P > 0.05; *, P < 0.05). (E) Relative M18V intensity changes after stimulation with 600 AP at 20 Hz starting at t = 0 s (black bar) at control synapses (green, n = 17 field of views, n = 465 synapses) and synapses with 1 µM Ro 31-8220 (purple, n = 11 field of views, n = 181 synapses). (F) Relative M18V intensity change at t = 10 s (left) and t = 160 s (right) at control synapses (green) and synapses with Ro 31-8220 (purple) calculated from E (M-W test with FDR [2] corrections: ***, P < 0.001). (G) Relative M18-GFP (green, n = 5 cells, n = 140 synapses) and M18_PKCi-GFP (orange, n = 3 cells, n = 84 synapses) intensity changes at synapses. (H) Relative M18-GFP and M18_PKCi-GFP intensity change at t = 10 s (left) and t = 160 s (right) at synapses calculated from G (M-W test with FDR [2] corrections: n.s., P > 0.05; **, P < 0.01). Error bars indicate mean ± SEM. The horizontal broken line indicates no change.

Figure 9.

Dynamic Munc18-1-Venus levels correlate with release of synaptic vesicles. (A) Experimental design to probe synaptic release before and after stimulation that induces synaptic Munc18-1-Venus (M18V) dispersion and reclustering. (1) Baseline M18V fluorescence in synapses of naive M18V neurons. (2) First FM4-64 loading using 30 AP at 1 Hz in same synapses. Stim: 600 AP at 20 Hz (high frequency) or 2 Hz (low frequency, does not induce M18V dispersion). (3) FM4-64 unloading and M18V reclustering after 600 AP at 20 Hz. (4) Second FM4-64 loading using 30 AP at 1 Hz. Synapses were followed over time using synapsin-CFP (not depicted). (B) Greyscale and fluorescence images showing four stages of the experiment to probe synaptic release (see A) with FM4-64 loading (middle) in a M18V neuron (top). Composite (bottom) shows colocalization of M18V with FM4-64 puncta (arrowheads). Bar, 5 µm. (C) Initial M18V fluorescence (see A, 1) plotted against the first FM4-64 loading (see A, 2) of individual synapses (Spearman’s correlation r_s = 0.27, P < 0.001; n = 11 field of views, n = 232 synapses). (D) Change in FM4-64 loading (between the first and second load) in synapses with a net positive and net negative reclustering of Munc18-1-Venus after 600 AP at 2 Hz or 20 Hz stimulation (t test: n.s., P > 0.05; **, P = 0.006; 2 Hz, n = 4 field of views, n = 146 synapses; 20 Hz, n = 7 field of views, n = 195 synapses). Error bars indicate mean ± SEM.

Figure 10.

Munc18-1 transport to and behavior in the synapse. In mature neurons, Munc18-1 (shown in green) is transported to synapses via lateral diffusion with membrane-bound syntaxin-1 (shown in light gray, 1). At the synapse, Munc18-1 interacts with syntaxin-1 (individually or in microdomains) or dissociates and interacts with other binding partners (X; 2). Calcium influx triggers dissociation from binding partners (3) and temporarily increases free Munc18-1 concentrations in the synapse (4). After stimulation, PKC controls reclustering of Munc18-1 (5).