The Morphological and Molecular Nature of Synaptic Vesicle Priming at Presynaptic Active Zones

Neuron

Article The Morphological and Molecular Nature of Synaptic Vesicle Priming at Presynaptic Active Zones Cordelia Imig,1 Sang-Won Min,2 Stefanie Krinner,1,5 Marife Arancillo,3,6 Christian Rosenmund,3 Thomas C. Su¨dhof,2 JeongSeop Rhee,1,6 Nils Brose,1,4,* and Benjamin H. Cooper1,4,* 1Department

of Molecular Neurobiology, Max Planck Institute of Experimental Medicine, 37075 Go¨ttingen, Germany of Molecular and Cellular Physiology and Howard Hughes Medical Institute, Stanford University School of Medicine, Stanford, CA 94305, USA 3Neuroscience Research Center and NeuroCure Cluster of Excellence, Charite ´ -Universita¨tsmedizin Berlin, 10117 Berlin, Germany 4Co-senior author 5Present Address: InnerEarLab, Department of Otolaryngology, University Medical Center Go ¨ ttingen, 37075 Go¨ttingen, Germany 6Present Address: Department of Pathology and Immunology, Baylor College of Medicine, Houston, TX 77030, USA *Correspondence: [email protected] (N.B.), [email protected] (B.H.C.) http://dx.doi.org/10.1016/j.neuron.2014.10.009 2Department

SUMMARY

Synaptic vesicle docking, priming, and fusion at active zones are orchestrated by a complex molecular machinery. We employed hippocampal organotypic slice cultures from mice lacking key presynaptic proteins, cryofixation, and three-dimensional electron tomography to study the mechanism of synaptic vesicle docking in the same experimental setting, with high precision, and in a near-native state. We dissected previously indistinguishable, sequential steps in synaptic vesicle active zone recruitment (tethering) and membrane attachment (docking) and found that vesicle docking requires Munc13/CAPS family priming proteins and all three neuronal SNAREs, but not Synaptotagmin-1 or Complexins. Our data indicate that membrane-attached vesicles comprise the readily releasable pool of fusion-competent vesicles and that synaptic vesicle docking, priming, and trans-SNARE complex assembly are the respective morphological, functional, and molecular manifestations of the same process, which operates downstream of vesicle tethering by active zone components.

INTRODUCTION Neuronal synaptic signaling is initiated by Ca2+-dependent fusion of synaptic vesicles (SVs) with the plasma membrane (PM). SV fusion is mediated by the PM t-SNAREs Syntaxin-1 (Stx-1) and SNAP-25, and the SV membrane v-SNARE Synaptobrevin-2 (Syb-2), which act in conjunction with the SM-protein Munc18-1. SNARE proteins are required for all cellular membrane fusion processes and contain conserved 60–70 amino acid stretches that form a stable four-a-helical SNARE complex, 416 Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc.

whose assembly is thought to provide the energy required for membrane fusion (Jahn and Scheller, 2006). Accordingly, SNAREs can induce liposome fusion in vitro, albeit with slower kinetics than neuronal Ca2+-triggered SV fusion, indicating that SNARE regulators and modulatory proteins are required upstream of the SV fusion reaction to achieve the speed and fidelity of synaptic excitation-secretion coupling (Jahn and Scheller, 2006; Rizo and Su¨dhof, 2012). The dissection of the molecular steps preceding Ca2+dependent vesicle fusion has been technically challenging. The current model poses that SVs attach, or ‘‘dock,’’ to the presynaptic active zone (AZ), and subsequently undergo a maturation process that renders them fusion competent, or ‘‘primed,’’ leading to a readily releasable pool (RRP) of SVs that can rapidly fuse upon Ca2+ stimulation (Su¨dhof, 2013). However, the underlying molecular processes are still largely enigmatic. In yeast homotypic vacuole fusion, three molecularly defined processes precede the fusion reaction, i.e., (1) an ATP-dependent ‘‘priming’’ step that is operationally different from SV ‘‘priming,’’ involves cis-SNARE complex disassembly by NSF/Sec18 and a-SNAP/Sec17, and thus provides free SNAREs to drive fusion, (2) the initiation of a loose contact between vacuole membranes by large ‘‘tethering’’ complexes, and (3) the tight ‘‘docking’’ of vacuoles via trans-SNARE complexes (Mayer and Wickner, 1997; Nichols et al., 1997; Ostrowicz et al., 2008). The accurate assessment of vesicle docking requires electron microscopy (EM) to resolve intermembrane distances in the nm range (Verhage and Sørensen, 2008). Information on proteins involved in docking of SVs and other secretory vesicles is sparse and partly inconclusive. For example, the notion that the vesicular Ca2+ sensor Synaptotagmin-1 (Syt-1) mediates vesicle docking by binding to Stx-1/SNAP-25 complexes in the PM (de Wit et al., 2009) has remained contentious, because the massive large dense-core vesicle (LDCV) docking defect in Syt-1 knockout (KO) chromaffin cells is not paralleled by a strong secretion defect (Schonn et al., 2008) and Syt-1 loss hardly affects the RRP in synapses (Geppert et al., 1994; Liu et al., 2009).

Neuron The Primed State of Synaptic Vesicles

Important reasons for these inconsistencies are that the experimental assessment of vesicle docking is technically challenging and that diverse preparations, cell types, sample fixation methods, imaging approaches, and docking definitions have been employed. Chemical fixation confounds the determination of SV docking as defined by direct PM attachment, because sample fixation and dehydration induce SV fusion (Smith and Reese, 1980), sample shrinkage, and membrane deformations (Murk et al., 2003). Further, the limited z resolution of electron micrographs from ultrathin sections (50–100 nm) impedes the identification of SV docking as SV midlines are often not contained within the section. To circumvent these problems, most classical analyses employed criteria where SVs within a certain distance of the AZ, i.e., 30 (Richmond et al., 1999), 40 (Broadie et al., 1995), or 50 nm (Hess et al., 1993), were defined as docked, assuming that SV populations within these distances are functionally homogeneous. Due to these methodological and interpretative ambiguities, past studies on vesicle docking are difficult to reconcile (Table S1 available online). Unlike SV docking, SV priming can be measured electrophysiologically by quantifying the RRP size (Rosenmund and Stevens, 1996). As with yeast vacuole fusion, SV priming and fusion are dependent on the availability of free SNAREs provided by a/b-SNAP and NSF activity (Burgalossi et al., 2010; DeBello et al., 1995; Ma et al., 2013). However, the SV priming reaction itself is executed by dedicated priming proteins of the Munc13 and CAPS families, which are not expressed in yeast but represent key determinants of the speed and fidelity of synaptic excitation-secretion coupling and are absolutely essential for neurotransmitter release (Jockusch et al., 2007; Varoqueaux et al., 2002). Several lines of evidence indicate that SV priming involves the regulation of SNARE protein conformation (Ma et al., 2011) and partial SNARE complex assembly (Walter et al., 2010; Weber et al., 2010), but the notion of stable, zippered SNARE complexes occupying an energetically favorable state prior to fusion has remained highly controversial (van den Bogaart et al., 2011; Jahn and Fasshauer, 2012; Rizo and Su¨dhof, 2012). Proteins implicated in the stabilization or clamp-like arrest of partially assembled SNARE complexes prior to fusion include the Ca2+-sensor Syt-1 (Chicka et al., 2008; Liu et al., 2009) and Complexins (CPXs) (Li et al., 2011; Tang et al., 2006), which both also facilitate SV fusion (Geppert et al., 1994; Reim et al., 2001; Xue et al., 2008; Yang et al., 2013). Based on an approach that circumvents most of the ambiguities of classical SV docking analyses (Weimer et al., 2006), we used organotypic hippocampal slice cultures from mutant mice, high-pressure freezing (HPF), freeze substitution (FS), and electron tomography (ET) of synapses to systematically characterize the molecular and morphological nature of the docked and primed SV states with nanometer accuracy. Our data indicate that SV and yeast vacuole fusion are preceded by similar and possibly conserved tethering and docking processes and that SV docking, priming, and trans-SNARE complex assembly are the respective morphological, functional, and molecular manifestations of the same process, operating downstream of SV tethering.

RESULTS Ultrastructure of Mouse Hippocampal Organotypic Slice Cultures HPF and FS yielded excellent ultrastructural preservation of hippocampal organotypic slice cultures. Well-preserved samples were characterized by a densely packed neuropil with smooth, continuous membranes exhibiting parallel bilayers (Figures 1 and S1). In tomograms, SVs that appeared in direct physical contact with the AZ were considered docked (Figures 1D1 and 1D2). Such PM-attached, or docked, SVs were assigned a 0–2 nm distance from the AZ, since our ET analysis was limited by an isotropic voxel size of 1.6 nm. Although not in direct AZ contact, some SVs appeared linked to the PM by short filaments (<5 nm length) (Figures 1E1 and 1E2), which had previously been described in cryo-ET analyses of synaptosomes and had been interpreted as assembled SNARE complexes of SVs in the RRP (Ferna´ndez-Busnadiego et al., 2010). Additionally, some SVs within 45 nm of the AZ were shown to be connected to the PM via single long tethers of unknown identity (Ferna´ndezBusnadiego et al., 2010, 2013; Siksou et al., 2007, 2009). In our analyses, closely tethered SVs (Figures 1E1 and 1E2) were not scored docked, but the number of SVs within 0–5 nm of the AZ was quantified separately (Table 1). Rarely observed SVs with an open fusion pore (Figure 1F1) or full collapse fusion events (Figure 1F2) were excluded from the analysis. Occasionally, inward PM curvature, consistent with early, ultrafast endocytotic events (Watanabe et al., 2013), was visible near AZs (Figures 1G1, 1G2, and S1D–S1F). Only one tomogram from a heterozygous SNAP-25 knockout (KO) exhibited a structure resembling bulk membrane retrieval (Figure S1G). Summaries of all ET and 2D-EM data are presented in Tables 1 and S2, respectively. Munc13 and CAPS Family Priming Proteins Are Required for SV Docking Munc13s are essential regulators of synaptic transmission. Munc13-deficient neurons lack a measurable RRP and thus show no spontaneous or evoked SV fusion (Varoqueaux et al., 2002). Studies employing HPF and FS to prepare samples for EM showed that loss of Munc13/UNC-13 proteins in neurons causes severe deficits in SV docking in C. elegans (Weimer et al., 2006) and mice (Siksou et al., 2009). We reanalyzed the role of Munc13s in SV docking and confirmed that in contrast to control (CTRL) synapses, synapses lacking Munc13s have almost no docked SVs (Figures 2A–2F). SVs in Munc13-1/2 double knockout (Munc13 DKO) synapses had the tendency to accumulate close (8–10 nm) to the AZ PM (Figure 2E), explaining why the number of PM-proximal SVs (0–40 nm) was unchanged between groups (Figure 2H). Focusing on the SV distribution within 40 nm of the AZ, we observed a significant reduction in the number of SVs within 0–5 nm in Munc13 DKO synapses, whereas SV numbers within 5–20 nm were significantly increased as compared to CTRLs (Figure 2G). In Munc13 DKO synapses, we further observed an increased SV diameter (Figure S5A) and a 28% increase in SV volume (Figure S5B), a previously unknown consequence of Munc13 loss. Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc. 417

Neuron The Primed State of Synaptic Vesicles

Figure 1. Ultrastructural Organization of Mouse Hippocampal Organotypic Slices (A–G) Electron micrographs of neuropil in organotypic slices prepared by HPF and FS. Glutamatergic spine synapses (asterisks). d, dendrite; m, mitochondrion; ast, astrocyte process; SA, spine apparatus. (D–G) SVs captured at intermediate stages of exo/ endocytosis after HPF and ET. (D1) Direct contact between the inner AZ and the outer SV phospholipid layers (docking). (D2) Occasional inward curvature of the AZ membranes toward a docked SV (arrowhead). (E1 and E2) SVs not in contact but closely tethered to the AZ (0–4 nm) by multiple, short filaments (open arrowheads). (F1 and F2) SV fusion events. (F1) A fusing SV displaying a prominent open fusion pore (arrowhead). (F2) Full collapse fusion at the AZ membrane. (G1 and G2) Endocytotic events at the periactive zone. Scale bars represent 1 mm in (A), 400 nm in (B) and (C), and 50 nm in (G2). See also Figure S1.

tic neurons lack primed SVs, with the remaining cells exhibiting an 85% reduction in the RRP size (Jockusch et al., 2007). Since only subvolumes of presynapses were reconstructed and CAPS DKO tomograms never exhibited more than 1–2 SVs per subvolume, we probably underestimated the number of synapses capable of SV docking/priming. Out of 15 Munc13 DKO tomograms, only one showed three docked SVs, which were on average still larger than docked SVs in CTRL tomograms (Figure S5C). This increase in SV size in Munc13 DKO synapses explains the reduction in the number of SVs within 100 nm of the AZ (Table 1) and the small decrease in SV cluster density (Figure S2E). Analyses of gross synaptic morphology and total SV numbers (Figures S2A–S2D) revealed no further changes, except for a small increase in the number of endosomes in Munc13 DKO terminals (Figure S2G). Given their structural and functional similarity to Munc13s, we next studied the role of CAPSs in SV docking. Despite the severe neurotransmission deficits in CAPS-1/2 DKOs (CAPS DKOs), including a dramatic decrease in the RRP size (Jockusch et al., 2007), the role of CAPSs in SV priming has been debated, mostly because loss of CAPS orthologs in C. elegans (Speese et al., 2007) and Drosophila (Renden et al., 2001) causes primarily defects in neuronal LDCV fusion. We found a dramatic reduction in the number of docked SVs in CAPS DKO synapses (Figures 2I–2N) but no changes in SV numbers within 40 or 100 nm of the AZ (Figure 2P and Table 1). Twelve out of 19 CAPS DKO synapses (62%) had no PM-attached SVs, and the remaining synapses showed an 82% reduction in the number of docked SVs (%2 docked SVs/profile) as compared to CTRLs (5 docked SVs/profile). This correlates well with the fact that 38% of CAPS DKO autap418 Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc.

These data show that CAPSs, like Munc13s, are required for SV priming (Jockusch et al., 2007) and docking. However, in contrast to Munc13 DKO synapses, CAPS DKO synapses did not accumulate SVs in proximity to the AZ, but rather exhibited a uniform SV distribution within 100 nm of the AZ (Figure 2M), with an increased number of SVs within 10–40 nm of the AZ, as compared to CTRLs (Figure 2O). Moreover, SV size (Figures S5D–S5F), SV numbers, and synaptic morphology assessed by 2D-EM (Figures S2H–S2N) of CAPS DKO synapses were similar to CTRLs, except for a slightly reduced SV cluster density. In contrast to a recent study (Sadakata et al., 2013), these data indicate that the severe SV docking and priming defects in CAPS DKO synapses are likely not due to reduced SV numbers. In summary, our data show that Munc13 and CAPS family SV priming proteins mediate SV docking at the AZ, indicating that docked SVs form the RRP of fusion-competent SVs in neurons. The SNAREs SNAP-25, Stx-1, and Syb-2 Are Required for SV Docking SV priming by Munc13s (Ma et al., 2011) and CAPSs (James et al., 2009; Parsaud et al., 2013) is thought to involve interactions with Stx-1 and other SNAREs. Ultrastructural studies using chemical fixation have failed to reveal an involvement of t-SNAREs in SV docking in mammalian neurons (Bronk et al., 2007; de Wit et al., 2006), but data obtained in C. elegans

SV Diameter

Munc13-1/2 CTRL (n = 16)

46.25 ± 0.49

Munc13-1/2 DKO (n = 15)

50.15 ± 0.72

CAPS-1/2 CTRL (n = 22)

45.19 ± 0.56

CAPS-1/2 DKO (n = 19)

45.54 ± 0.65

SNAP-25 CTRL (n = 24)

46.12 ± 0.48

SNAP-25 KO (n = 25)

52.53 ± 0.48

SNAP-25 KOD (n = 12)

52.09 ± 0.88

SNAP-25 KO

ND

(n = 13)

Numer SVs within 100 nm of AZ ***p < 0.001

Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc. 419

44.60 ± 0.39

Syntaxin-1-A KO /BYFP/YFP (n = 23)

45.31 ± 0.54

Synaptobrevin-2 CTRL (n = 25)

45.81 ± 0.45

Synaptobrevin-2 KO (n = 24)

49.07 ± 0.41

Synaptobrevin-2 KOD (n = 8)

48.65 ± 0.54

Synaptobrevin-2 KOND (n = 16)

49.28 ± 0.56

Synaptotagmin-1 CTRL (n = 23)

46.81 ± 0.55

Synaptotagmin-1 KO (n = 20)

45.41 ± 0.58

Complexin-1/2/3 CTRL (n = 19)

44.71 ± 0.62

Complexin-1/2/3 TKO (n = 25)

44.11 ± 0.41

**p = 0.0099

5.02 ± 0.38 p = 0.6834 ns

7.12 ± 0.41

***p < 0.001

6.77 ± 0.40

p = 0.0614 ns

5.45 ± 0.29

*p = 0.0216

7.11 ± 0.31

p = 0.3533 ns

6.45 ± 0.34 5.96 ± 0.44

p = 0.1772 ns

9.00 ± 0.64

p = 0.8736 ns

6.01 ± 0.32 6.86 ± 0.25

2.53 ± 0.19 2.98 ± 0.17 2.60 ± 0.17 2.58 ± 0.16

p = 0.1080 ns

3.63 ± 0.24

p = 0.1140 ns

2.39 ± 0.18 2.73 ± 0.14

1.67 ± 0.17 1.67 ± 0.20 1.13 ± 0.11 1.76 ± 0.19

***p < 0.001

1.59 ± 0.17

***p < 0.001

1.53 ± 0.24 2.24 ± 0.18 1.40 ± 0.18 1.80 ± 0.21

1.13 ± 0.13 0.50 ± 0.08

**p = 0.009

1.01 ± 0.14

***p < 0.001 ***p < 0.001 ***p < 0.001

0±0 *p = 0.0356

0.62 ± 0.11 ***p = 0.004

1.08 ± 0.14

***p < 0.001

0.36 ± 0.10 ***p < 0.001

0.97 ± 0.12

***p < 0.001

0.05 ± 0.02 *p = 0.012

1.49 ± 0.22 p = 0.1346 ns

1.20 ± 0.14

***p < 0.001

0.58 ± 0.13 *p = 0.0329

***p < 0.001

0.24 ± 0.06

0.89 ± 0.18 p = 0.5465 ns

0.98 ± 0.13

0.08 ± 0.03

1.15 ± 0.13 p = 0.7286 ns

Number SVs within 0-2 nm of AZ 0.04 ± 0.04

0.36 ± 0.09 p = 0.2840 ns

2.80 ± 0.29 *p = 0.0371

***p < 0.001

0.73 ± 0.11

2.72 ± 0.14

7.63 ± 0.50 p = 0.4094 ns

p = 0.9047 ns

2.67 ± 0.11

5.89 ± 0.21 p = 0.0873 ns

2.78 ± 0.22

1.71 ± 0.20

0.23 ± 0.07

2.69 ± 0.20

5.91 ± 0.20 p = 0.4822 ns

p = 0.3400 ns

2.95 ± 0.17 p = 0.7896 ns

7.00 ± 0.27 ***p < 0.001

2.66 ± 0.20

Number SVs within 0-5 nm of AZ 0.21 ± 0.12

2.75 ± 0.13

5.90 ± 0.37 p = 0.2975 ns

p = 0.1626 ns

2.39 ± 0.19

5.68 ± 0.24 p = 0.401 ns

2.49 ± 0.19 2.11 ± 0.18

6.02 ± 0.39

52.93 ± 0.48

Syntaxin-1 CTRL (n = 21)

6.55 ± 0.40

Number SVs within 40 nm of AZ

1.49 ± 0.20

*p = 0.0302

0.91 ± 0.16 p = 0.166 ns

0.87 ± 0.15

p = 0.2476 ns

1.12 ± 0.14

CTRL, control, KO, knockout; DKO, double knockout; TKO, triple knockout, SV, synaptic vesicle; AZ, active zone; n, number of tomograms; D, docked SVs; ND, no or few docked SVs. SV numbers within 100, 40, 0–4, and 0–2 nm of the AZ are normalized to 0.01 mm2 of AZ area. Values indicate mean ± SEM. See also Table S2.

Neuron

Sample

The Primed State of Synaptic Vesicles

Table 1. 3D Electron Tomographic Analysis of Synaptic Vesicle Docking

Neuron The Primed State of Synaptic Vesicles

Figure 2. 3D Electron Tomographic Analysis of SV Docking in Munc13-1/2 DKO and CAPS1-/2 DKO Neurons (A, C, I, and K) ET subvolumes of CTRL, Munc13 DKO, and CAPS DKO synapses. Docked SVs making AZ contact within (white arrowheads) or outside of the displayed subvolume (open arrowheads). (B, D, J, and L) 3D models of synaptic profiles including orthogonal views of the AZ (AZ, white; docked SVs, green; nonattached SVs, gray; LDCVs, beige). (E and M) Spatial distribution of SVs within 100 nm of the AZ. (F and N) Scatterplots of the mean number of docked SVs (0–2 nm of the AZ) normalized to AZ area. (G and O) Mean SV number within 0–5, 5–10, 10– 20, 20–30, 40–50, and 50–60 nm of the AZ normalized to AZ area. (H and P) Mean SV number within 0–40 nm of the AZ normalized to AZ area. Values indicate mean ± SEM; ***p < 0.001; **p < 0.01; *p < 0.05. Scale bars represent 100 nm. See also Figures S2 and S5.

indicate that Stx plays a role in SV docking (Hammarlund et al., 2007). To assess the role of the t-SNARE acceptor complex in SV docking, we analyzed a SNAP-25 KO and a Stx-1A/B KO/ hypomorph mouse line. SNAP-25 KO neurons degenerate in low-density (Washbourne et al., 2002) but survive in high-density cultures (Bronk et al., 2007). SNAP-25 KO slices appeared thinner than CTRL slices, but no obvious reduction in synapse density was observed by qualitative EM analysis. ET revealed a strong decrease in the 420 Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc.

number of docked SVs in SNAP-25 KO samples, with 13 out of 25 KO synapses completely lacking docked SVs (Figures 3A–3F). Reminiscent of the Munc13 DKO phenotype, SNAP-25 KO synapses accumulated SVs in close proximity to the AZ, albeit slightly closer at 4–8 nm (Figure 3E). As a consequence, the number of SVs within 5–20 nm of the AZ was increased in SNAP-25 KO synapses, as compared to CTRLs (Figure 3G), although the overall number of SVs within 40 nm of the AZ was unchanged (Figure 3H). Moreover, the mean SV diameter in SNAP-25 KO synapses was increased to 53 nm from 46 nm in CTRL samples (Figure S5G; Table 1), causing a 48% increase in SV volume (Figure S5H). The combination of a major docking deficit and an increased SV size in SNAP-25 KO synapses explains the observed reductions in SV numbers within 100 nm (Table 1) and 30–60 nm of the AZ (Figure 3G), since the greater volume occupied by PM-proximal SVs shifts ‘‘second row’’ SVs farther from the AZ. The increased SV size also accounts for the reduced SV terminal (Figure S3D) and cluster density (Figure S3E) detected by 2D-EM analysis (Figures S3A–S3G). Additionally, the number of apparent endosomal structures in SNAP-25 KO synapses was increased (Figure S3G). To assess the role of Stx-1 in SV docking, we analyzed a knockin mouse line expressing Stx-1B fused to YFP in a

Neuron The Primed State of Synaptic Vesicles

Figure 3. 3D Electron Tomographic Analysis of SV Docking in SNAP-25 KO and Stx-1A KO/BYFP Neurons (A, C, I, and K) ET subvolumes of CTRL (A and I), SNAP-25 KO (C), and Stx-1A KO/BYFP (K) synapses. Docked SVs making AZ contact within (white arrowheads) or outside of the displayed subvolume (open arrowheads). (B, D, J, and L) 3D models of synaptic profiles including orthogonal views of the AZ (AZ, white; docked SVs, green; nonattached SVs, gray; endosomes, light green). (E and M) Spatial distribution of SVs within 100 nm of the AZ. (F and N) Scatterplots of the mean number of docked SVs (0–2 nm of the AZ) normalized to AZ area. (G and O) Mean SV number within 0–5, 5–10, 10–20, 20–30, 40–50, and 50–60 nm of the AZ normalized to AZ area. (H and P) Mean SV number within 0–40 nm of the AZ normalized to AZ area. Values indicate mean ± SEM; ***p < 0.001; **p < 0.01; *p < 0.05. Scale bars represent 100 nm. See also Figures S3 and S5.

Stx-1A KO background (Stx-1A KO/BYFP) (Arancillo et al., 2013). The expression level of YFP-tagged Stx-1B in Stx-1A KO/BYFP mice is dramatically reduced from wild-type (WT) levels, causing an 65% reduction in the RRP and changes in the SV distribution within 40 nm of the AZ as assessed by ET analysis of chemically fixed neurons (Arancillo et al., 2013). We detected a 38% reduction in the number of docked SVs in Stx-1A KO/BYFP synapses (Figures 3I–3N). No changes in the size (Figures S5L– S5N) or the spatial distribution of SVs were observed (Figures

3M, 3O, and 3P and Table 1). However, 2D-EM analyses of Stx-1A KO/BYFP synapses (Figures S3H–S3N) revealed a decrease in the SV cluster density (Arancillo et al., 2013) and a small increase in the number of endosomal structures. To complete our analysis of the main synaptic SNAREs, we investigated the role of the v-SNARE Syb-2 in SV docking. Syb-2 KO was previously shown to cause a 90% reduction in the RRP (Schoch et al., 2001), but no change in the number of docked SVs in neurons was observed (Dea´k et al., 2004). Our analysis of Syb2 KO terminals (Figures 4A–4D) revealed a severe reduction in the number of docked SVs (Figures 4E and 4F) and a prominent accumulation of SVs close (4–8 nm) to the AZ (Figure 4E), reflected by an increased number of SVs within 5– 20 nm of the AZ (Figure 4G). Syb-2 KO synapses showed no changes in the number of SVs within 40 or 100 nm of the AZ (Figure 4H and Table 1) or in the number of presynaptic SVs as quantified by 2D-EM (Figures S4A–S4D). However, PSDs in Syb-2 KO synapses were slightly longer than in CTRLs (Figure S4F), which is only partially explained by a slight but not significant increase in terminal size (CTRL 0.398 mm2 ± 0.014; Syb-2 KO 0.431 mm2 ± 0.013, p > 0.05, ns). Similar to Munc13 DKO or SNAP-25 KO synapses, Syb-2 KO synapses exhibited an increased mean SV diameter of 49 nm, as compared to 46 nm in CTRLs (Figure S5O; Table 1), causing a Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc. 421

Neuron The Primed State of Synaptic Vesicles

Figure 4. 3D Electron Tomographic Analysis of SV Docking in Syb-2 KO Neurons (A and C) ET subvolumes of CTRL (A) and Syb-2 KO (C) synapses. Docked SVs making AZ contact within (white arrowheads) or outside of the displayed subvolume (open arrowheads). (B and D) 3D models of synaptic profiles including orthogonal views of the AZ (AZ, white; docked SVs, green; nonattached SVs, gray; endosomes, light green). (E) Spatial distribution of SVs within 100 nm of the AZ. (F) Scatterplots of the mean number of docked SVs (0–2 nm of the AZ) normalized to AZ area. (G) Mean SV number within 0–5, 5–10, 10–20, 20– 30, 40–50, and 50–60 nm of the AZ normalized to AZ area. (H) Mean SV number within 0–40 nm of the AZ normalized to AZ area. Values indicate mean ± SEM; ***p < 0.001; **p < 0.01; *p < 0.05. Scale bars represent 100 nm. See also Figures S4 and S5.

25% increase in SV volume (Figure S5P). The increased SV size shifts ‘‘second row’’ SVs to a greater distance from the AZ, which accounts for the reduced number of SVs within 40–50 nm of the AZ in Syb-2 KO synapses (Figure 4G). Moreover, the number of presynaptic endocytotic structures was increased in Syb-2 KO synapses (Figure S4G). In summary, our data indicate that the three major neuronal SNAREs Stx-1, SNAP-25, and Syb-2 play key roles in SV docking to the AZ PM. Loss of Individual SNAREs Can Be Partially Compensated A comparison of individual docking profiles from Syb-2 KO tomograms (Figure 4F) revealed that 16 out of 24 Syb-2 KO synapses (66%) had few or no docked (ND) SVs (Figures 4C and 4D; Syb-2 KOND), whereas the remaining synapses had similar numbers of docked (D) SVs as CTRLs (Syb-2 KOD) (Figures 4F, 5A, and 5B, CTRL 1.08 ± 0.14; Syb-2 KOD 0.97 ± 0.12, p > 0.05, ns). SV docking was severely reduced in Syb-2 KOND in comparison to KOD synapses (Figure 5F and Table 1) with KOND synapses exhibiting a prominent SV accumulation close (4–8 nm) to the AZ (Figure 5E). Similar to Syb-2 KO samples, SNAP-25 KO neurons exhibited a substantial degree of variability in the number of docked SVs per synapse (Figure 3F), with 13 out of 25 tomograms (52%) completely lacking docked SVs (nondocked, SNAP-25 KOND; Figures 3C and 3F), whereas the rest exhibited docked SVs, albeit not in comparable numbers as CTRLs (Figures 5C, 5D, and 5M; CTRL 1.13 ± 0.13; SNAP-25 KOD 0.50 ± 0.08, p = 0.003). SV docking was severely reduced in SNAP-25 KOND in comparison to KOD synapses (Figure 5M and Table 1), with KOND synapses exhibiting a more prominent SV accumulation close (4–8 nm) to the AZ (Figure 5L). 422 Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc.

To test whether the presence of a paralog might account for the SV docking in a subset of Syb-2 KO synapses, we analyzed the localization of immunolabeled Syb-1 in organotypic slices. Confocal microscopy of WT slices showed a specific punctate labeling of Syb-2, which highly colocalized with VGLUT1 but was absent in Syb-2 KO cultures (data not shown). In CTRLs, Syb-1 labeling revealed large immunoreactive puncta (Figure 5G) that colocalized with only 8% of VGLUT1-positive synapses (Figure 5K), indicating that Syb-1 is mainly localized to nonglutamatergic synapses in WT slices. Syb-2 KO slices exhibited an increase in the number of Syb-1-positive puncta (Figures 5G and 5H; CTRL 1 ± 0.1; Syb-2 KO 3.25 ± 0.28, p < 0.001) and in the extent of colocalization with VGLUT1 (Figure 5K; CTRL 7.41% ± 1.40%; Syb-2 KO 31.25% ± 2.92%, p < 0.001), indicating Syb-1 upregulation in glutamatergic synapses in the absence of Syb-2. Moreover, the number (Figure 5I; CTRL 1 ± 0.03; Syb-2 KO 1.21 ± 0.03, p < 0.001) and size of VGLUT1immunoreactive puncta (Figure 5J; CTRL 1 ± 0.06; Syb-2 KO 1.53 ± 0.06, p < 0.001) were increased in Syb-2 KO slices. Thus, mature organotypic hippocampal Syb-2 KO slices exhibit a compensatory increase in the number of Syb-1-positive glutamatergic synapses to 31%, probably accounting for the observation that 33% of all Syb-2 KO synaptic profiles analyzed by ET still harbored docked SVs. A large percentage of SNAP-25 KO neurons degenerate in culture (Delgado-Martı´nez et al., 2007; Washbourne et al., 2002), indicating that survival of the remaining neurons probably requires an alternative, yet unknown SNARE. To test this notion, we focused on SNAP-23 since its overexpression in SNAP-25 KO neurons rescues the reduction in RRP sizes but not the impaired evoked transmitter release (Delgado-Martı´nez et al., 2007). SNAP-23 immunoreactivity often appeared in apposition

Neuron The Primed State of Synaptic Vesicles

Figure 5. Partial Compensation for the Loss of Syb-2 (A and C) ET subvolumes of synaptic profiles exhibiting docked SVs in Syb-2 KO (Syb KOD) (A) and SNAP-25 KO (SNAP-25D) (C) samples. Docked SVs making AZ contact within (white arrowheads) or outside of the displayed subvolume (open arrowheads). (B and D) 3D models of synaptic profiles including orthogonal views of the AZ (AZ, white; docked SVs, green; nonattached SVs, gray; endosomes, light green). (E and L) Spatial distribution of SVs within 100 nm of the AZ. (F and M) Mean number of docked SVs. (G and N) Immunolabeling of Syb-1 (G) and SNAP23 (N) together with VGLUT1. Colocalization indicated in white in the merged panel. (H and O) Normalized spatial density of Syb-1 (H) and SNAP-23 (O) signals. (I and P) Normalized spatial density of VGLUT1 immunoreactivity. (J and Q) Normalized mean area of VGLUT1-positive puncta. (K and R) The proportion of glutamatergic (VGLUT1) terminals in which Syb-1 (K) and SNAP23 (R) are detected. Light gray bars indicate the frequency of incidental colocalization evaluated by horizontally flipping the VGLUT1 channel. Values indicate mean + SEM; ***p < 0.001; **p < 0.01; *p < 0.05. Scale bars represent 100 nm in (A) and (C) and 5 mm in (G) and (N). See also Figure S5.

to, but not overlapping with, VGLUT1 signals in both genotypes, indicating a potential postsynaptic localization, as has been described previously (Suh et al., 2010). The number of SNAP23-positive puncta (Figures 5N and 5O; CTRL 1 ± 0.09; SNAP25 KO 1.37 ± 0.13, p = 0.230) and the number (Figure 5P; CTRL 1 ± 0.06; SNAP-25 KO 1.60 ± 0.04, p < 0.001) and size of VGLUT1-immunoreactive puncta (Figure 5Q; CTRL 1 ± 0.05; SNAP-25 KO 1.90 ± 0.07, p < 0.001) were increased in SNAP25 KO slices. Colocalization of SNAP-23 and VGLUT-1 signals, though only rarely observed, increased in SNAP-25 KO slices (Figures 5N and 5R; CTRL 5.78% ± 0.78%; SNAP-25 KO 13.61 ± 1.81, p < 0.001). However, the comparable increase in incidental colocalization in SNAP-25 KO slices (see Supplemental Experimental Procedures) indicates that the apparent increase in association between SNAP-23 and VGLUT1 in SNAP-25 KO slices is related to a generalized increase in labeling densities rather than to specific SNAP-23 upregulation in glutamatergic synapses. It is therefore unlikely that SNAP-23

compensates substantially for the loss of SNAP-25 in neurons. SNAP-29 and SNAP-47 (Holt et al., 2006; Steegmaier et al., 1998), other neuronal SNAP-25 paralogs, are similarly unlikely to significantly compensate for a loss of SNAP25, because SNAP-29 has been proposed to inhibit synaptic transmission (Pan et al., 2005), cultured excitatory SNAP-29 KO neurons do not show deficits in basic synaptic transmission (N. Sivakumar, G. Wieser, S. Go¨bbels, N.B., and J.-S.R., unpublished data), and SNAP-47 is mainly present on intracellular membranes and SVs (Holt et al., 2006). Given that Stx-1A/B DKO neurons degenerate in culture, it is unlikely that another Syntaxin paralog is playing a significant compensatory role. Thus, residual SV docking in Stx-1A KO/BYFP synapses is probably mediated by the remaining Stx-1B. In summary, the presence of a small pool of docked SVs in the absence of the major synaptic SNAREs is probably attributable to a compensatory effect by other SNARE paralogs. Consistent with a recent report on the roles of Syb-1 and Syb-2 in presynaptic function (Zimmermann et al., 2014), such a compensatory effect appears to operate in Syb-2 KO synapses, where an upregulation of Syb-1 can rescue SV docking in a subset of glutamatergic synapses. The SNAREs capable of compensating for the loss of SNAP-25 remain unknown, although it cannot be excluded that some SV docking occurs in the absence of any Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc. 423

Neuron The Primed State of Synaptic Vesicles

SNAP-25-like SNARE or, based on the present data, that SNAP23, SNAP-29, and SNAP-47 play minor roles in this context. Synapses Lacking Munc13, SNAP-25, or Syb-2 Exhibit Larger SVs We observed increased SV diameters in Munc13 DKO and in SNAP-25 and Syb-2 (Figure S5) KO synapses. Moreover, SVs appeared similarly enlarged in Syb-2 KOD and KOND synapses (Figures S5R and S5S) and in SNAP-25 KOD and KOND synapses (Figures S5J and S5K). The finding that docked SVs were also larger in Munc13 DKO, Syb-2 KO, and SNAP-25 KO synapses (Figure S5C, S5I, and S5Q) indicates that increased SV size is not causally linked to the observed docking phenotypes. The altered SV size in SNAP-25 and Syb-2 KO synapses may reflect perturbed SV recycling due to aberrant endocytosis (Zhang et al., 2013). In support of this notion, increased SV sizes are also seen upon knockdown (KD) or blockade of the endocytotic Syb-2 adaptor AP180 (Koo et al., 2011; Morgan et al., 1999; Zhang et al., 1998). The 28% increase in SV volume observed in Munc13 DKO synapses is more difficult to explain, but one possible reason could be that the constant block of SV fusion in these synapses prevents vesicular proteins (i.e., Syt-1, Syb2) required for endocytosis from reaching the PM (Diril et al., 2006; Koo et al., 2011). However, synaptosomes lacking RIM1a, a major Munc13 binding partner at the AZ, exhibit a similar increase in SV volumes (Ferna´ndez-Busnadiego et al., 2013) as Munc13 DKO synapses, yet RIM1a KO synapses are capable of SV fusion (Schoch et al., 2002; Andrews-Zwilling et al., 2006; Han et al., 2011; Deng et al., 2011). These data on RIM1a, along with our finding that CAPS DKO synapses have normally sized SVs despite severely impaired SV docking and synaptic transmission, indicate that the increased SV size in Munc13 DKO synapses cannot solely be attributable to the complete block of synaptic transmission. Syt-1 Is Not a Major SV Docking Protein The role of the exocytotic Ca2+ sensor Syt-1 in SV docking and priming in mammalian neurons is unclear. The massive LDCV docking defect seen upon Syt-1 deletion in chromaffin cells is paralleled by only minor defects in secretion (de Wit et al., 2009), indicating that LDCV docking in chromaffin cells may not be directly related to LDCV priming and fusion. To examine the role of Syt-1 in SV docking, we extended our analysis to Syt-1 KO mice. Docked SV numbers were reduced in Syt-1 KO synapses by 39% (Figures 6A–6F). In contrast to all other analyzed mutants, Syt-1 KO synapses exhibited an 23% decrease in SV number within 40 nm of the AZ membrane (Figure 6H) and lacked the accumulation of SVs close to the AZ seen in Munc13, SNAP-25, and Syb-2-deficient synapses (Figure 6E). A detailed analysis of SV distributions revealed that the reduction of SV numbers within 40 nm of the AZ in Syt-1 KO synapses is caused by a significant loss of SVs within 0–5 nm of the AZ (Figure 6G and Table 1). SV terminal and cluster density were also reduced in Syt-1 KO synapses (Figures S6A–S6G). Considering that SV numbers per synapse were unchanged (Figure S6C), the reduced SV densities probably reflect the slightly larger terminal area sampled in Syt-1 KO synapses (CTRL: 0.335 mm2 ± 0.012; Syt-1 KO: 0.373 mm2 ± 424 Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc.

0.014, p < 0.05). SV size (Figures S5T and S5U), PSD length (Figure S6F), and the number of endosomal structures (Figure S6G) were comparable between groups. Thus, Syt-1 KO neurons exhibit a small decrease in the number of SVs near the AZ, with a consequent reduction in the number of docked SVs, indicating a minor and indirect role of Syt-1 in SV docking. These findings are in line with the fact that the RRP size in cultured hippocampal neurons is only minimally affected upon Syt-1 KO (Geppert et al., 1994; Liu et al., 2009), but they do not corroborate the observation that Syt-1 KO perturbs SV docking in C. elegans and Drosophila synapses (Jorgensen et al., 1995; Reist et al., 1998; Yu et al., 2013). To test whether an upregulation of Syt-1 paralogs might compensate for the loss of Syt-1 and thus account for the lack of an SV docking deficit in Syt-1 KO neurons, we focused on Syt-4, Syt-7, and Syt-11, which are coexpressed with Syt-1 in the hippocampus (Bacaj et al., 2013). We performed a quantitative western blot analysis of homogenates of organotypic Syt-1 KO and CTRL slices to assess potential changes in protein expression of the respective Syt isoforms. All tested Syt isoforms exhibited slight but nonsignificant trends toward increased expression levels in Syt-1 KO slices (Syt-4 - CTRL 1 ± 0.16, n = 3; Syt-1 KO 1.29 ± 0.13, n = 3; p = 0.22; Syt-7 - CTRL 1 ± 0.17, n = 3; Syt-1 KO 1.29 ± 0.06, n = 3; p = 0.17; Syt-11 CTRL 1 ± 0.24, n = 3; Syt-1 KO 1.62 ± 0.19, n = 3; p = 0.12), although the respective mRNA levels are not changed in cultured Syt-1 KO neurons (Bacaj et al., 2013). Based on these findings, it cannot be excluded that alternative Syt isoforms compensate for Syt-1 loss as regards its role in SV docking. However, such a scenario is difficult to reconcile with the facts that Syt-1 KO has only an indirect effect on SV docking (Figures 6, S5, and S6; Table 1), that Syt-1 is by far the most abundantly expressed Syt isoform on SVs and in presynaptic terminals (Takamori et al., 2006), and that multiple lines of evidence indicate nonredundant characteristics and functions of the Syt isoforms we studied (Bacaj et al., 2013; Dai et al., 2004; Geppert et al., 1994; Maximov et al., 2008; von Poser et al., 1997; Schonn et al., 2008). CPXs Have a Postdocking Role in the SV Cycle Murine CPXs were initially proposed to facilitate Ca2+-triggered SV fusion at a postpriming step, since evoked release and SV release probability are drastically reduced in cultured hippocampal neurons from CPX-1/2/3 TKO (CPX TKO) mice, with no changes in RRP or the number of docked SVs (Reim et al., 2001; Xue et al., 2008). Conflicting evidence, mainly based on biochemical studies in vitro and genetic studies in C. elegans and Drosophila, indicates that CPXs might also play an inhibitory role by acting as fusion clamps that prevent spontaneous SV fusion (Giraudo et al., 2006; Huntwork and Littleton, 2007; Wragg et al., 2013). Recent experiments involving KD of CPXs in cortical neurons even indicate a role for CPXs in SV priming (Yang et al., 2013). CPX TKO slices exhibited no changes in the number of docked, and therefore possibly primed, SVs (Figures 6I–6N), in SV size (Figures S5W–S5Y), or in SV distribution (Figures 6M, 6O, and 6P), except for a small increase in SV numbers within 100 nm of the AZ (Table 1). 2D-EM analyses (Figures

Neuron The Primed State of Synaptic Vesicles

Figure 6. 3D Electron Tomographic Analysis of SV Docking in Syt-1 KO and CPX-1/2/3 TKO Neurons (A, C, I, and K) ET subvolumes of CTRL, Syt-1 KO, and CPX TKO synapses. Docked SVs making AZ contact within (white arrowheads) or outside of the displayed subvolume (open arrowheads). (B, D, J, and L) 3D models of synaptic profiles including orthogonal views of the AZ (AZ, white; docked SVs, green; non-attached SVs, gray; endosomes, light green; LDCVs, beige). (E and M) Spatial distribution of SVs within 100 nm of the AZ. (F and N) Scatterplots of the mean number of docked SVs (0–2 nm of the AZ) normalized to AZ area. (G and O) Mean SV number within 0–5, 5–10, 10–20, 20–30, 40–50, and 50–60 nm of the AZ normalized to AZ area. (H and P) Mean SV number within 0-40 nm of the AZ normalized to AZ area. Values indicate mean ± SEM; ***p < 0.001; **p < 0.01; *p < 0.05. Scale bars represent 100 nm. See also Figures S5 and S6.

S6H–S6N) revealed no changes in SV numbers or synaptic morphology. Since total presynaptic and docked SV populations are unaffected upon CPX TKO, we interpret our data to indicate that CPXs act downstream of SV docking. Differential Roles of SNAREs and SV Priming Proteins in Synaptic LDCV Secretion Since all proteins analyzed in the present study have also been implicated in chromaffin cell LDCV fusion (Ashery et al., 2000;

Borisovska et al., 2005; Cai et al., 2008; Liu et al., 2010; Speidel et al., 2005; de Wit et al., 2006, 2009), we quantified presynaptic LDCVs in the respective mouse mutants (Table S3). In SNARE mutant tomograms, we observed LDCVs close to but not in contact with the AZ (Figures 7A–7F). The number of LDCVs within 200 nm of the AZ was increased in Syb-2 KO, SNAP-25 KO (n.s.), and Stx1A KO/BYFP mutants (Figure 7G). Presynaptic LDCV accumulation in SNARE mutants was also evident by 2D-EM (Figure 7H). These data support the notion that all three SNAREs are required for neuronal LDCV fusion. In Munc13 DKO, CAPS DKO, Syt-1 KO, and CPX TKO synapses, the number of LDCVs within 200 nm of the AZ was unchanged (Figures 7G and 7H). Although in line with previous data (Jockusch et al., 2007), our finding of unchanged presynaptic LDCV numbers in murine CAPS DKO neurons is in conflict with a recent study indicating a loss of LDCVs in CAPS-1-deficient mouse neurons (Sadakata et al., 2013). Moreover, C. elegans and Drosophila presynapses accumulate undocked LDCVs in the absence of their respective CAPS homologs (Hammarlund et al., 2008; Renden et al., 2001). Despite evidence demonstrating an involvement of Munc13s in synaptic but not extrasynaptic LDCV secretion (van de Bospoort et al., 2012), we demonstrate that LDCVs do not accumulate in synapses in the absence of Munc13s. Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc. 425

Neuron The Primed State of Synaptic Vesicles

Figure 7. SNARE Mutants Accumulate LDCVs in Presynaptic Terminals (A–F) LDCVs (asterisks) accumulate within 40 nm but do not dock to the AZ in Syb-2 KO, SNAP-25 KO, and Stx-1A KO/1BYFP synapses. (G) ET analysis of mean LDCV number within 200 nm of the AZ per synaptic profile (n = number of tomograms). (H) 2D-EM analysis of mean LDCV number per synaptic profile (n = number of synapses). Values indicate mean + SEM; **p < 0.01; *p < 0.0. Scale bar represents 100 nm in (F). See also Table S3.

DISCUSSION Membrane Attachment as the Morphological Manifestation of SV Priming To analyze synaptic ultrastructure by HPF, FS, and ET in mouse mutants that lack key components of the SV cycle and exhibit perinatally lethal phenotypes, we employed a hippocampal organotypic slice culture system, which yielded well-vitrified samples with minimally perturbed ultrastructure (Figure 1). In synaptic subvolumes from 200-nm-thick sections, we observed 5–6 docked SVs per CTRL profile (Figures 1, 2, 3, 4, and 6). Since the mean AZ area sampled by ET was 0.05 mm2, i.e., approximately half the area of a complete AZ (Ferna´ndez-Busnadiego et al., 2013), our data indicate an average of 10–12 docked SVs per hippocampal synapse. This is in line with previous studies, describing 3–4 docked SVs per WT spine synapse in 100-nm-thick hippocampal sections (Siksou et al., 2009), and with estimates of 10 AZ-proximal SVs in brain synapses from aldehyde-fixed samples (Schikorski and Stevens, 1997). Consistent with previous findings (Siksou et al., 2009; Weimer et al., 2006), deletion of Munc13 family priming proteins, which completely eliminates the RRP (Varo426 Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc.

queaux et al., 2002), results in an almost complete loss of docked SVs (Figures 2A–2G). Further, we report a hitherto undetected SV docking deficit in CAPS DKO synapses (Figures 2I–2O), which also correlates closely with corresponding RRP reductions in CAPS DKO neurons (Jockusch et al., 2007) (Table S4 and Figure S7). Taken together, our findings imply that membrane-attached, docked SVs comprise the RRP of primed SVs, whereas SVs further removed from the AZ do not. In support of this conclusion, the number of docked SVs in CTRL synapses (10–12) correlates well with calculated synaptic RRP sizes in cultured hippocampal neurons (Stevens and Tsujimoto, 1995) and stimulation of cultured neurons causes a specific decrease in the number of docked SVs at AZs, as shown in a recent study combining optogenetics and rapid HPF fixation (Watanabe et al., 2013). trans-SNARE Complex Assembly as the Molecular Basis of SV Docking and Priming The molecular basis of SV docking and priming and the involvement of the SNARE complex and regulatory proteins like Syt-1 and CPXs in these processes have been the focus of intense debate. Many hypotheses pose that the primed SV state involves at least partially assembled trans-SNARE complexes (Rizo and Su¨dhof, 2012). This view has been challenged, based on the arguments that the capture of trans-SNARE complexes in vitro is difficult and that the biochemical and thermodynamic characteristics of SNARE complex assembly and SNARE-mediated fusion cannot easily be reconciled with the notion of the existence of stable, partially, or fully assembled SNARE complexes prior to Ca2+-triggered SV fusion (Jahn and Fasshauer, 2012). SNARE motifs of individual SNAREs in opposing membranes start to interact at a distance of 8 nm, SNARE zippering arrests at a distance of 2–4 nm (Li et al., 2007), with 2 nm corresponding to the fully assembled SNARE bundle (Sutton et al., 1998), and Fo¨rster resonance energy transfer between t-SNAREs and v-SNAREs during liposome fusion starts at

Neuron The Primed State of Synaptic Vesicles

intermembrane distances below 5 nm (van den Bogaart et al., 2011). We used ET to resolve intermembrane distances with nm precision in tomograms with an isotropic voxel size of 1.6 nm and found that all three neuronal SNAREs are required for SV docking at a distance that is well below the maximum limit for trans-SNARE complex engagement (Figures 3 and 4). The reductions in the number of docked SVs in individual SNARE mutant synapses (Figures 3 and 4) correlate well with corresponding changes in RRP sizes (Arancillo et al., 2013; Bronk et al., 2007; Schoch et al., 2001) (Table S4 and Figure S7). These findings indicate that SV docking and priming involve at least the partial assembly of SNARE complexes. That SV docking and priming are only incompletely abolished in Syb-2 or SNAP-25 KO neurons can be explained by the presence of SNARE paralogs (Bronk et al., 2007; Zimmermann et al., 2014), as indicated by the upregulation of Syb-1 expression in glutamatergic Syb2 KO synapses (Figures 4O–4S). The remaining docked SVs in Stx-1A KO/BYFP synapses can be explained by the residual expression of Stx-1BYFP in the mutant. Syt-1 and CPXs Are Dispensable for SV Docking and Priming Syt-1 KO synapses exhibit decreased SV numbers within 40 nm of the AZ and a moderate reduction in the number of docked SVs (Figures 6A–6H). These findings are congruent with the fact that Syt-1 KO has only minor effects on the RRP (Geppert et al., 1994; Liu et al., 2009) and indicate an indirect effect of Syt-1 loss on SV docking. This may involve altered SV recycling in the absence of Syt-1 (Jorgensen et al., 1995), a deficit in the tethering of SVs to the AZ that can potentially disrupt the refilling of the RRP of docked SVs, which would corroborate the effect of Syt-1 on liposome clustering (Arac¸ et al., 2006; Seven et al., 2013), or an increased rate of spontaneous SV fusion (Kerr et al., 2008), which might contribute to the partial overall depletion of SVs and to the reduction in the number of docked SVs in Syt-1 KO synapses. CPX TKO synapses show no signs of altered SV distribution or docking (Figures 6I–6P), in line with earlier studies indicating unaltered RRP sizes in CPX-deficient neurons (Reim et al., 2001; Xue et al., 2008). This indicates a postdocking/priming role of CPXs. Despite the similarities observed between C. elegans and mouse synapses that point toward an evolutionarily conserved mechanism of SV release (e.g., for Unc-13/ Munc13 or Unc-64/Syntaxin-1), our data on the role of CPXs in SV docking highlight species differences in the involvement of regulatory proteins. For instance, CPX-1 KO in C. elegans has severe effects on the numbers of total and docked SVs at neuromuscular junctions, possibly induced by an increase in spontaneous fusion events and consequent depletion of SV pools (Hobson et al., 2011). In contrast, CPX-1 KO in Drosophila results in an increase in spontaneous release with no effect on the number of total and docked SVs at the neuromuscular junction (Huntwork and Littleton, 2007; Jorquera et al., 2012). We found that loss of all murine CPXs expressed in the hippocampus affects neither SV docking nor the total numbers of SVs. This is in line with previous data showing that cultured mouse hippocampal neurons of CPX TKO mice do not exhibit an increase, but rather a small decrease, in mEPSC frequency (Xue et al., 2008).

In summary, our data support the notion that in mouse hippocampal synapses, CPX facilitates SV fusion in a postdocking/ priming step. Such a role might involve the stabilization of partially assembled SNARE complexes in the docked/primed SV state. This would explain why the SV release probability, which is thought to be affected by the number of available SNARE complexes per SV (Arancillo et al., 2013), is strongly reduced in the absence of CPXs (Xue et al., 2008). However, different methods of triggering RRP release and different cell culture models have led to different results regarding the effects of CPX loss on the RRP (e.g., Xue et al., 2008 versus Yang et al., 2013). In view of these findings, it was proposed that CPX activity conveys a highly fusogenic ‘‘superprimed’’ SV state (Yang et al., 2013), which, according to the present data, is not reflected by a different SV docking state but potentially becomes manifest at the level of SV release probability. Tethering Steps Preceding SV Docking and Priming Cryo-ET analyses of synaptosomal ultrastructure identified filaments tethering SVs at a distance of 10 nm from the AZ (Ferna´ndez-Busnadiego et al., 2010) and led to the postulate that members of the RIM family of AZ scaffold and regulatory proteins are involved in this process (Ferna´ndez-Busnadiego et al., 2013). The notion that AZ scaffold proteins contribute to SV tethering is further supported by classical EM studies showing defects in the AZ recruitment (i.e., tethering) of SVs in aldehyde-fixed synapses lacking RIMs (Kaeser et al., 2011) or Piccolo and Bassoon (Mukherjee et al., 2010). In the present study, nondocked SVs accumulated at distances of 8–10 nm from the AZ in synapses lacking Munc13s (Figures 2E and 2G) and 4–8 nm from the AZ in synapses deficient for SNAP-25 (Figures 3E and 3G) or Syb-2 (Figures 4E and 4G). In contrast, synapses lacking CAPSs (Figure 2M), Syt-1 (Figure 6E), or synapses with reduced Stx-1 levels (Figure 3M) do not accumulate nondocked SVs near the AZ. These findings indicate an SV tethering process that precedes SV docking/priming, that does not require Munc13s, Syb-2, or SNAP-25, and that might involve CAPSs and Stx-1. Indeed, loss of UNC-18/Munc18-1, an essential regulator of Stx-1 function in SV fusion, and loss of the Stx-1 homolog UNC-64 in C. elegans have both been associated with a decreased number of nondocked SVs near AZs, i.e., in SV tethering (Gracheva et al., 2010). We therefore propose that SVs are recruited to the AZ with the involvement of AZ components and become tethered in close AZ proximity (10 nm). Munc13s, CAPSs, and/or yet unknown interacting proteins might then enable SVs to approach to an 6 nm distance from the AZ prior to trans-SNARE interactions for final docking/priming. The capturing of SVs for final membrane attachment by Munc13s might be mediated via the Munc13/RIM/Rab3A interaction (Dulubova et al., 2005), and Munc13s and CAPSs might then initiate or accelerate SNARE complex assembly via their interactions with Stx-1 (Khodthong et al., 2011; Ma et al., 2011, 2013; Parsaud et al., 2013). Our analyses of SV docking in neurons are in conflict with the prevalent model for LDCV docking in chromaffin cells, where Syt-1 interactions with t-SNARE acceptor complexes are thought to mediate LDCV membrane attachment independently of Syb-2 (Borisovska et al., 2005; Gerber et al., 2008; de Wit et al., 2006, 2009). However, the corresponding LDCV Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc. 427

Neuron The Primed State of Synaptic Vesicles

studies mainly employed chemical fixation and 2D-EM, where reliable detection of membrane attachment is more difficult than with ET due to the limited z resolution. In these studies, LDCVs not only failed to dock to the PM in Syt-1, Stx-1, or SNAP-25 KO chromaffin cells but also exhibited a change in the cumulative vesicle distribution within 100 nm of the PM (de Wit et al., 2006, 2009). This raises the possibility that the described docking deficit reflects an inability of LDCVs to be recruited and tethered to the PM upstream of final membrane attachment. Indeed, a recent study using HPF for EM of LDCV docking in PC12 cells demonstrated that proteolytic cleavage of Syb-2 causes a decrease in the number of docked LDCVs with no major changes in the number of membrane-proximal and possibly tethered vesicles (Wu et al., 2012). These data support a sequential model, according to which LDCVs are recruited or tethered to the PM, which—at least in chromaffin cells—might be primarily mediated by Syt-1 interactions with the t-SNARE acceptor complex and then become docked in a process that requires all three SNAREs, similar to the model that we propose here for the molecular steps preceding SV release in neurons. Such a Syt-1/t-SNARE-mediated tethering step, although apparently crucial for LDCV membrane recruitment in chromaffin cells, may be less critical in synapses due to the dense AZ protein network, which might employ alternative or additional mechanisms and proteins (e.g., RIMs [Ferna´ndez-Busnadiego et al., 2013; Kaeser et al., 2011], Bassoon, and Piccolo [Mukherjee et al., 2010]) to recruit SVs in proximity to the AZ.

blindly using ImageJ (National Institutes of Health). For details, see Supplemental Experimental Procedures. Statistics Statistical analyses were performed with GraphPad Prism (Version 6.00; GraphPad). To test for normality, we used the Kolmogorov-Smirnov test. Normally distributed data sets were compared by unpaired Student’s t tests. For non-Gaussian data sets, the Mann-Whitney U test was used. Data are presented as mean and SEM (mean ± SEM). SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, seven figures, and six tables and can be found with this article online at http://dx.doi.org/10.1016/j.neuron.2014.10.009. AUTHOR CONTRIBUTIONS C.I., J.-S.R., N.B., and B.C. designed research; C.I., and B.C. performed research; S.-W.M., M.A., C.R., and T.C.S. provided unpublished mouse lines; C.I., S.K., and B.C. analyzed data; C.I., J.-S.R., N.B., and B.C. wrote the paper. ACKNOWLEDGMENTS This work was supported by the Max Planck Society (N.B.), the German Research Foundation (SFB889/B1, J.-S.R., N.B.), and the European Commission (EUROSPIN and SynSys, N.B.). C.I. was a student of the Go¨ttingen PhD Program and IMPRS ‘Neurosciences’ and the Go¨ttingen Graduate School for Neurosciences and Molecular Biosciences (GGNB) (DFG grant GSC 226/ 1). We thank D. Riedel, B. Altas, and F. Benseler for discussions and advice, I. Thanha¨user, D. Schwerdtfeger, and C. Harenberg for excellent technical support, Y.-J. Wu, U. Teichmann, and M. Loos for the provision of mice, and the MPIEM animal facility for mouse husbandry.

EXPERIMENTAL PROCEDURES Mouse Lines and Hippocampal Organotypic Slice Cultures Tables S5 and S6 list mouse lines, genotypes, and animal numbers used for 2D-EM and 3D-EM. Organotypic hippocampal slice cultures from mutants and CTRL littermates were prepared according to the interface method (Stoppini et al., 1991) and cultured as published (Kerr et al., 2008). For details, see Supplemental Experimental Procedures. Immunohistochemistry, Confocal Microscopy, and Image Processing and Analysis Organotypic hippocampal slices were processed for immunohistochemistry as published (Yoon et al., 2011). Images were acquired with a Leica TCSSP5 confocal microscope. Image processing and colocalization analyses were performed using ImageJ (National Institutes of Health). For details, also on the antibodies used, see Supplemental Experimental Procedures. Western Blot Analysis Antibodies and methods used for western blotting are described in Supplemental Experimental Procedures. Protein levels on western blots for Syt-4, Syt-7, and Syt-11 were determined by using the LI-COR Odyssey scanner and software (LI-COR Biosciences). Expression levels were normalized using Stx-1A/B as loading control. HPF, FS, EM, and ET Hippocampal slices were frozen in cryoprotectant (20% BSA in culture medium, 340 mOsm) using a HPM100 (Leica) HPF device and cryosubstituted in an EM AFS2 (Leica). For 2D analyses of synaptic morphology, electron micrographs were acquired with a transmission electron microscope (Zeiss LEO 912-Omega) operating at 80 kV. For 3D-ET analysis of SV docking, 200-nmthick sections were imaged in a JEM-2100 transmission electron microscope (JEOL) operating at 200 kV. Tomograms were reconstructed from tilt series using the IMOD package (Kremer et al., 1996). Quantifications were performed

428 Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc.

Accepted: October 2, 2014 Published: October 22, 2014 REFERENCES Andrews-Zwilling, Y.S., Kawabe, H., Reim, K., Varoqueaux, F., and Brose, N. (2006). Binding to Rab3A-interacting molecule RIM regulates the presynaptic recruitment of Munc13-1 and ubMunc13-2. J. Biol. Chem. 281, 19720–19731. Arac¸, D., Chen, X., Khant, H.A., Ubach, J., Ludtke, S.J., Kikkawa, M., Johnson, A.E., Chiu, W., Su¨dhof, T.C., and Rizo, J. (2006). Close membrane-membrane proximity induced by Ca(2+)-dependent multivalent binding of synaptotagmin1 to phospholipids. Nat. Struct. Mol. Biol. 13, 209–217. Arancillo, M., Min, S.-W., Gerber, S., Mu¨nster-Wandowski, A., Wu, Y.-J., Herman, M., Trimbuch, T., Rah, J.-C., Ahnert-Hilger, G., Riedel, D., et al. (2013). Titration of Syntaxin1 in mammalian synapses reveals multiple roles in vesicle docking, priming, and release probability. J. Neurosci. 33, 16698– 16714. Ashery, U., Varoqueaux, F., Voets, T., Betz, A., Thakur, P., Koch, H., Neher, E., Brose, N., and Rettig, J. (2000). Munc13-1 acts as a priming factor for large dense-core vesicles in bovine chromaffin cells. EMBO J. 19, 3586–3596. Bacaj, T., Wu, D., Yang, X., Morishita, W., Zhou, P., Xu, W., Malenka, R.C., and Su¨dhof, T.C. (2013). Synaptotagmin-1 and synaptotagmin-7 trigger synchronous and asynchronous phases of neurotransmitter release. Neuron 80, 947–959. Borisovska, M., Zhao, Y., Tsytsyura, Y., Glyvuk, N., Takamori, S., Matti, U., Rettig, J., Su¨dhof, T., and Bruns, D. (2005). v-SNAREs control exocytosis of vesicles from priming to fusion. EMBO J. 24, 2114–2126. Broadie, K., Prokop, A., Bellen, H.J., O’Kane, C.J., Schulze, K.L., and Sweeney, S.T. (1995). Syntaxin and synaptobrevin function downstream of vesicle docking in Drosophila. Neuron 15, 663–673.

Neuron The Primed State of Synaptic Vesicles

Bronk, P., Dea´k, F., Wilson, M.C., Liu, X., Su¨dhof, T.C., and Kavalali, E.T. (2007). Differential effects of SNAP-25 deletion on Ca2+ -dependent and Ca2+ -independent neurotransmission. J. Neurophysiol. 98, 794–806.

Hammarlund, M., Watanabe, S., Schuske, K., and Jorgensen, E.M. (2008). CAPS and syntaxin dock dense core vesicles to the plasma membrane in neurons. J. Cell Biol. 180, 483–491.

Burgalossi, A., Jung, S., Meyer, G., Jockusch, W.J., Jahn, O., Taschenberger, H., O’Connor, V.M., Nishiki, T., Takahashi, M., Brose, N., and Rhee, J.S. (2010). SNARE protein recycling by aSNAP and bSNAP supports synaptic vesicle priming. Neuron 68, 473–487.

Han, Y., Kaeser, P.S., Su¨dhof, T.C., and Schneggenburger, R. (2011). RIM determines Ca2+ channel density and vesicle docking at the presynaptic active zone. Neuron 69, 304–316.

Cai, H., Reim, K., Varoqueaux, F., Tapechum, S., Hill, K., Sørensen, J.B., Brose, N., and Chow, R.H. (2008). Complexin II plays a positive role in Ca2+-triggered exocytosis by facilitating vesicle priming. Proc. Natl. Acad. Sci. USA 105, 19538–19543. Chicka, M.C., Hui, E., Liu, H., and Chapman, E.R. (2008). Synaptotagmin arrests the SNARE complex before triggering fast, efficient membrane fusion in response to Ca2+. Nat. Struct. Mol. Biol. 15, 827–835. Dai, H., Shin, O.-H., Machius, M., Tomchick, D.R., Su¨dhof, T.C., and Rizo, J. (2004). Structural basis for the evolutionary inactivation of Ca2+ binding to synaptotagmin 4. Nat. Struct. Mol. Biol. 11, 844–849.

Hess, S.D., Doroshenko, P.A., and Augustine, G.J. (1993). A functional role for GTP-binding proteins in synaptic vesicle cycling. Science 259, 1169–1172. Hobson, R.J., Liu, Q., Watanabe, S., and Jorgensen, E.M. (2011). Complexin maintains vesicles in the primed state in C. elegans. Curr. Biol. 21, 106–113. Holt, M., Varoqueaux, F., Wiederhold, K., Takamori, S., Urlaub, H., Fasshauer, D., and Jahn, R. (2006). Identification of SNAP-47, a novel Qbc-SNARE with ubiquitous expression. J. Biol. Chem. 281, 17076–17083. Huntwork, S., and Littleton, J.T. (2007). A complexin fusion clamp regulates spontaneous neurotransmitter release and synaptic growth. Nat. Neurosci. 10, 1235–1237.

de Wit, H., Cornelisse, L.N., Toonen, R.F.G., and Verhage, M. (2006). Docking of secretory vesicles is syntaxin dependent. PLoS ONE 1, e126.

Jahn, R., and Fasshauer, D. (2012). Molecular machines governing exocytosis of synaptic vesicles. Nature 490, 201–207.

de Wit, H., Walter, A.M., Milosevic, I., Gulya´s-Kova´cs, A., Riedel, D., Sørensen, J.B., and Verhage, M. (2009). Synaptotagmin-1 docks secretory vesicles to syntaxin-1/SNAP-25 acceptor complexes. Cell 138, 935–946.

Jahn, R., and Scheller, R.H. (2006). SNAREs—engines for membrane fusion. Nat. Rev. Mol. Cell Biol. 7, 631–643.

Dea´k, F., Schoch, S., Liu, X., Su¨dhof, T.C., and Kavalali, E.T. (2004). Synaptobrevin is essential for fast synaptic-vesicle endocytosis. Nat. Cell Biol. 6, 1102–1108. DeBello, W.M., O’Connor, V., Dresbach, T., Whiteheart, S.W., Wang, S.S., Schweizer, F.E., Betz, H., Rothman, J.E., and Augustine, G.J. (1995). SNAPmediated protein-protein interactions essential for neurotransmitter release. Nature 373, 626–630. Delgado-Martı´nez, I., Nehring, R.B., and Sørensen, J.B. (2007). Differential abilities of SNAP-25 homologs to support neuronal function. J. Neurosci. 27, 9380–9391. Deng, L., Kaeser, P.S., Xu, W., and Su¨dhof, T.C. (2011). RIM proteins activate vesicle priming by reversing autoinhibitory homodimerization of Munc13. Neuron 69, 317–331. Diril, M.K., Wienisch, M., Jung, N., Klingauf, J., and Haucke, V. (2006). Stonin 2 is an AP-2-dependent endocytic sorting adaptor for synaptotagmin internalization and recycling. Dev. Cell 10, 233–244. Dulubova, I., Lou, X., Lu, J., Huryeva, I., Alam, A., Schneggenburger, R., Su¨dhof, T.C., and Rizo, J. (2005). A Munc13/RIM/Rab3 tripartite complex: from priming to plasticity? EMBO J. 24, 2839–2850. Ferna´ndez-Busnadiego, R., Zuber, B., Maurer, U.E., Cyrklaff, M., Baumeister, W., and Lucic, V. (2010). Quantitative analysis of the native presynaptic cytomatrix by cryoelectron tomography. J. Cell Biol. 188, 145–156.

James, D.J., Kowalchyk, J., Daily, N., Petrie, M., and Martin, T.F.J. (2009). CAPS drives trans-SNARE complex formation and membrane fusion through syntaxin interactions. Proc. Natl. Acad. Sci. USA 106, 17308–17313. Jockusch, W.J., Speidel, D., Sigler, A., Sørensen, J.B., Varoqueaux, F., Rhee, J.-S., and Brose, N. (2007). CAPS-1 and CAPS-2 are essential synaptic vesicle priming proteins. Cell 131, 796–808. Jorgensen, E.M., Hartwieg, E., Schuske, K., Nonet, M.L., Jin, Y., and Horvitz, H.R. (1995). Defective recycling of synaptic vesicles in synaptotagmin mutants of Caenorhabditis elegans. Nature 378, 196–199. Jorquera, R.A., Huntwork-Rodriguez, S., Akbergenova, Y., Cho, R.W., and Littleton, J.T. (2012). Complexin controls spontaneous and evoked neurotransmitter release by regulating the timing and properties of synaptotagmin activity. J. Neurosci. 32, 18234–18245. Kaeser, P.S., Deng, L., Wang, Y., Dulubova, I., Liu, X., Rizo, J., and Su¨dhof, T.C. (2011). RIM proteins tether Ca2+ channels to presynaptic active zones via a direct PDZ-domain interaction. Cell 144, 282–295. Kerr, A.M., Reisinger, E., and Jonas, P. (2008). Differential dependence of phasic transmitter release on synaptotagmin 1 at GABAergic and glutamatergic hippocampal synapses. Proc. Natl. Acad. Sci. USA 105, 15581–15586. Khodthong, C., Kabachinski, G., James, D.J., and Martin, T.F.J. (2011). Munc13 homology domain-1 in CAPS/UNC31 mediates SNARE binding required for priming vesicle exocytosis. Cell Metab. 14, 254–263.

Ferna´ndez-Busnadiego, R., Asano, S., Oprisoreanu, A.-M., Sakata, E., Doengi, M., Kochovski, Z., Zu¨rner, M., Stein, V., Schoch, S., Baumeister, W., , V. (2013). Cryo-electron tomography reveals a critical role of and Lucic RIM1a in synaptic vesicle tethering. J. Cell Biol. 201, 725–740.

Koo, S.J., Markovic, S., Puchkov, D., Mahrenholz, C.C., Beceren-Braun, F., Maritzen, T., Dernedde, J., Volkmer, R., Oschkinat, H., and Haucke, V. (2011). SNARE motif-mediated sorting of synaptobrevin by the endocytic adaptors clathrin assembly lymphoid myeloid leukemia (CALM) and AP180 at synapses. Proc. Natl. Acad. Sci. USA 108, 13540–13545.

Geppert, M., Goda, Y., Hammer, R.E., Li, C., Rosahl, T.W., Stevens, C.F., and Su¨dhof, T.C. (1994). Synaptotagmin I: a major Ca2+ sensor for transmitter release at a central synapse. Cell 79, 717–727.

Kremer, J.R., Mastronarde, D.N., and McIntosh, J.R. (1996). Computer visualization of three-dimensional image data using IMOD. J. Struct. Biol. 116, 71–76.

Gerber, S.H., Rah, J.-C., Min, S.-W., Liu, X., de Wit, H., Dulubova, I., Meyer, A.C., Rizo, J., Arancillo, M., Hammer, R.E., et al. (2008). Conformational switch of syntaxin-1 controls synaptic vesicle fusion. Science 321, 1507–1510.

Li, F., Pincet, F., Perez, E., Eng, W.S., Melia, T.J., Rothman, J.E., and Tareste, D. (2007). Energetics and dynamics of SNAREpin folding across lipid bilayers. Nat. Struct. Mol. Biol. 14, 890–896.

Giraudo, C.G., Eng, W.S., Melia, T.J., and Rothman, J.E. (2006). A clamping mechanism involved in SNARE-dependent exocytosis. Science 313, 676–680.

Li, F., Pincet, F., Perez, E., Giraudo, C.G., Tareste, D., and Rothman, J.E. (2011). Complexin activates and clamps SNAREpins by a common mechanism involving an intermediate energetic state. Nat. Struct. Mol. Biol. 18, 941–946.

Gracheva, E.O., Maryon, E.B., Berthelot-Grosjean, M., and Richmond, J.E. (2010). Differential regulation of synaptic vesicle tethering and docking by UNC-18 and TOM-1. Front. Synaptic Neurosci. 2, 141. Hammarlund, M., Palfreyman, M.T., Watanabe, S., Olsen, S., and Jorgensen, E.M. (2007). Open syntaxin docks synaptic vesicles. PLoS Biol. 5, e198.

Liu, H., Dean, C., Arthur, C.P., Dong, M., and Chapman, E.R. (2009). Autapses and networks of hippocampal neurons exhibit distinct synaptic transmission phenotypes in the absence of synaptotagmin I. J. Neurosci. 29, 7395–7403.

Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc. 429

Neuron The Primed State of Synaptic Vesicles

Liu, Y., Schirra, C., Edelmann, L., Matti, U., Rhee, J., Hof, D., Bruns, D., Brose, N., Rieger, H., Stevens, D.R., and Rettig, J. (2010). Two distinct secretory vesicle-priming steps in adrenal chromaffin cells. J. Cell Biol. 190, 1067–1077.

Schoch, S., Dea´k, F., Ko¨nigstorfer, A., Mozhayeva, M., Sara, Y., Su¨dhof, T.C., and Kavalali, E.T. (2001). SNARE function analyzed in synaptobrevin/VAMP knockout mice. Science 294, 1117–1122.

Ma, C., Li, W., Xu, Y., and Rizo, J. (2011). Munc13 mediates the transition from the closed syntaxin-Munc18 complex to the SNARE complex. Nat. Struct. Mol. Biol. 18, 542–549.

Schoch, S., Castillo, P.E., Jo, T., Mukherjee, K., Geppert, M., Wang, Y., Schmitz, F., Malenka, R.C., and Su¨dhof, T.C. (2002). RIM1alpha forms a protein scaffold for regulating neurotransmitter release at the active zone. Nature 415, 321–326.

Ma, C., Su, L., Seven, A.B., Xu, Y., and Rizo, J. (2013). Reconstitution of the vital functions of Munc18 and Munc13 in neurotransmitter release. Science 339, 421–425. Maximov, A., Lao, Y., Li, H., Chen, X., Rizo, J., Sørensen, J.B., and Su¨dhof, T.C. (2008). Genetic analysis of synaptotagmin-7 function in synaptic vesicle exocytosis. Proc. Natl. Acad. Sci. USA 105, 3986–3991. Mayer, A., and Wickner, W. (1997). Docking of yeast vacuoles is catalyzed by the Ras-like GTPase Ypt7p after symmetric priming by Sec18p (NSF). J. Cell Biol. 136, 307–317. Morgan, J.R., Zhao, X., Womack, M., Prasad, K., Augustine, G.J., and Lafer, E.M. (1999). A role for the clathrin assembly domain of AP180 in synaptic vesicle endocytosis. J. Neurosci. 19, 10201–10212. Mukherjee, K., Yang, X., Gerber, S.H., Kwon, H.-B., Ho, A., Castillo, P.E., Liu, X., and Su¨dhof, T.C. (2010). Piccolo and bassoon maintain synaptic vesicle clustering without directly participating in vesicle exocytosis. Proc. Natl. Acad. Sci. USA 107, 6504–6509. Murk, J.L., Posthuma, G., Koster, A.J., Geuze, H.J., Verkleij, A.J., Kleijmeer, M.J., and Humbel, B.M. (2003). Influence of aldehyde fixation on the morphology of endosomes and lysosomes: quantitative analysis and electron tomography. J. Microsc. 212, 81–90. Nichols, B.J., Ungermann, C., Pelham, H.R., Wickner, W.T., and Haas, A. (1997). Homotypic vacuolar fusion mediated by t- and v-SNAREs. Nature 387, 199–202. Ostrowicz, C.W., Meiringer, C.T.A., and Ungermann, C. (2008). Yeast vacuole fusion: a model system for eukaryotic endomembrane dynamics. Autophagy 4, 5–19. Pan, P.-Y., Cai, Q., Lin, L., Lu, P.-H., Duan, S., and Sheng, Z.-H. (2005). SNAP29-mediated modulation of synaptic transmission in cultured hippocampal neurons. J. Biol. Chem. 280, 25769–25779. Parsaud, L., Li, L., Jung, C.H., Park, S., Saw, N.M.N., Park, S., Kim, M.Y., and Sugita, S. (2013). Calcium-dependent activator protein for secretion 1 (CAPS1) binds to syntaxin-1 in a distinct mode from Munc13-1. J. Biol. Chem. 288, 23050–23063. Reim, K., Mansour, M., Varoqueaux, F., McMahon, H.T., Su¨dhof, T.C., Brose, N., and Rosenmund, C. (2001). Complexins regulate a late step in Ca2+dependent neurotransmitter release. Cell 104, 71–81. Reist, N.E., Buchanan, J., Li, J., DiAntonio, A., Buxton, E.M., and Schwarz, T.L. (1998). Morphologically docked synaptic vesicles are reduced in synaptotagmin mutants of Drosophila. J. Neurosci. 18, 7662–7673. Renden, R., Berwin, B., Davis, W., Ann, K., Chin, C.T., Kreber, R., Ganetzky, B., Martin, T.F., and Broadie, K. (2001). Drosophila CAPS is an essential gene that regulates dense-core vesicle release and synaptic vesicle fusion. Neuron 31, 421–437. Richmond, J.E., Davis, W.S., and Jorgensen, E.M. (1999). UNC-13 is required for synaptic vesicle fusion in C. elegans. Nat. Neurosci. 2, 959–964. Rizo, J., and Su¨dhof, T.C. (2012). The membrane fusion enigma: SNAREs, Sec1/Munc18 proteins, and their accomplices—guilty as charged? Annu. Rev. Cell Dev. Biol. 28, 279–308. Rosenmund, C., and Stevens, C.F. (1996). Definition of the readily releasable pool of vesicles at hippocampal synapses. Neuron 16, 1197–1207. Sadakata, T., Kakegawa, W., Shinoda, Y., Hosono, M., Katoh-Semba, R., Sekine, Y., Sato, Y., Tanaka, M., Iwasato, T., Itohara, S., et al. (2013). CAPS1 deficiency perturbs dense-core vesicle trafficking and Golgi structure and reduces presynaptic release probability in the mouse brain. J. Neurosci. 33, 17326–17334. Schikorski, T., and Stevens, C.F. (1997). Quantitative ultrastructural analysis of hippocampal excitatory synapses. J. Neurosci. 17, 5858–5867.

430 Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc.

Schonn, J.-S., Maximov, A., Lao, Y., Su¨dhof, T.C., and Sørensen, J.B. (2008). Synaptotagmin-1 and -7 are functionally overlapping Ca2+ sensors for exocytosis in adrenal chromaffin cells. Proc. Natl. Acad. Sci. USA 105, 3998–4003. Seven, A.B., Brewer, K.D., Shi, L., Jiang, Q.-X., and Rizo, J. (2013). Prevalent mechanism of membrane bridging by synaptotagmin-1. Proc. Natl. Acad. Sci. USA 110, E3243–E3252. Siksou, L., Rostaing, P., Lechaire, J.-P., Boudier, T., Ohtsuka, T., Fejtova´, A., Kao, H.-T., Greengard, P., Gundelfinger, E.D., Triller, A., and Marty, S. (2007). Three-dimensional architecture of presynaptic terminal cytomatrix. J. Neurosci. 27, 6868–6877. Siksou, L., Varoqueaux, F., Pascual, O., Triller, A., Brose, N., and Marty, S. (2009). A common molecular basis for membrane docking and functional priming of synaptic vesicles. Eur. J. Neurosci. 30, 49–56. Smith, J.E., and Reese, T.S. (1980). Use of aldehyde fixatives to determine the rate of synaptic transmitter release. J. Exp. Biol. 89, 19–29. Speese, S., Petrie, M., Schuske, K., Ailion, M., Ann, K., Iwasaki, K., Jorgensen, E.M., and Martin, T.F.J. (2007). UNC-31 (CAPS) is required for dense-core vesicle but not synaptic vesicle exocytosis in Caenorhabditis elegans. J. Neurosci. 27, 6150–6162. Speidel, D., Bruederle, C.E., Enk, C., Voets, T., Varoqueaux, F., Reim, K., Becherer, U., Fornai, F., Ruggieri, S., Holighaus, Y., et al. (2005). CAPS1 regulates catecholamine loading of large dense-core vesicles. Neuron 46, 75–88. Steegmaier, M., Yang, B., Yoo, J.S., Huang, B., Shen, M., Yu, S., Luo, Y., and Scheller, R.H. (1998). Three novel proteins of the syntaxin/SNAP-25 family. J. Biol. Chem. 273, 34171–34179. Stevens, C.F., and Tsujimoto, T. (1995). Estimates for the pool size of releasable quanta at a single central synapse and for the time required to refill the pool. Proc. Natl. Acad. Sci. USA 92, 846–849. Stoppini, L., Buchs, P.-A., and Muller, D. (1991). A simple method for organotypic cultures of nervous tissue. J. Neurosci. Methods 37, 173–182. Su¨dhof, T.C. (2013). Neurotransmitter release: the last millisecond in the life of a synaptic vesicle. Neuron 80, 675–690. Suh, Y.H., Terashima, A., Petralia, R.S., Wenthold, R.J., Isaac, J.T.R., Roche, K.W., and Roche, P.A. (2010). A neuronal role for SNAP-23 in postsynaptic glutamate receptor trafficking. Nat. Neurosci. 13, 338–343. Sutton, R.B., Fasshauer, D., Jahn, R., and Brunger, A.T. (1998). Crystal structure of a SNARE complex involved in synaptic exocytosis at 2.4 A resolution. Nature 395, 347–353. Takamori, S., Holt, M., Stenius, K., Lemke, E.A., Grønborg, M., Riedel, D., Urlaub, H., Schenck, S., Bru¨gger, B., Ringler, P., et al. (2006). Molecular anatomy of a trafficking organelle. Cell 127, 831–846. Tang, J., Maximov, A., Shin, O.-H., Dai, H., Rizo, J., and Su¨dhof, T.C. (2006). A complexin/synaptotagmin 1 switch controls fast synaptic vesicle exocytosis. Cell 126, 1175–1187. van de Bospoort, R., Farina, M., Schmitz, S.K., de Jong, A., de Wit, H., Verhage, M., and Toonen, R.F. (2012). Munc13 controls the location and efficiency of dense-core vesicle release in neurons. J. Cell Biol. 199, 883–891. van den Bogaart, G., Thutupalli, S., Risselada, J.H., Meyenberg, K., Holt, M., Riedel, D., Diederichsen, U., Herminghaus, S., Grubmu¨ller, H., and Jahn, R. (2011). Synaptotagmin-1 may be a distance regulator acting upstream of SNARE nucleation. Nat. Struct. Mol. Biol. 18, 805–812. Varoqueaux, F., Sigler, A., Rhee, J.-S., Brose, N., Enk, C., Reim, K., and Rosenmund, C. (2002). Total arrest of spontaneous and evoked synaptic transmission but normal synaptogenesis in the absence of Munc13-mediated vesicle priming. Proc. Natl. Acad. Sci. USA 99, 9037–9042.

Neuron The Primed State of Synaptic Vesicles

Verhage, M., and Sørensen, J.B. (2008). Vesicle docking in regulated exocytosis. Traffic 9, 1414–1424. von Poser, C., Ichtchenko, K., Shao, X., Rizo, J., and Su¨dhof, T.C. (1997). The evolutionary pressure to inactivate. A subclass of synaptotagmins with an amino acid substitution that abolishes Ca2+ binding. J. Biol. Chem. 272, 14314–14319. Walter, A.M., Wiederhold, K., Bruns, D., Fasshauer, D., and Sørensen, J.B. (2010). Synaptobrevin N-terminally bound to syntaxin-SNAP-25 defines the primed vesicle state in regulated exocytosis. J. Cell Biol. 188, 401–413. Washbourne, P., Thompson, P.M., Carta, M., Costa, E.T., Mathews, J.R., Lopez-Bendito´, G., Molna´r, Z., Becher, M.W., Valenzuela, C.F., Partridge, L.D., and Wilson, M.C. (2002). Genetic ablation of the t-SNARE SNAP-25 distinguishes mechanisms of neuroexocytosis. Nat. Neurosci. 5, 19–26.

Wu, Y., Gu, Y., Morphew, M.K., Yao, J., Yeh, F.L., Dong, M., and Chapman, E.R. (2012). All three components of the neuronal SNARE complex contribute to secretory vesicle docking. J. Cell Biol. 198, 323–330. Xue, M., Stradomska, A., Chen, H., Brose, N., Zhang, W., Rosenmund, C., and Reim, K. (2008). Complexins facilitate neurotransmitter release at excitatory and inhibitory synapses in mammalian central nervous system. Proc. Natl. Acad. Sci. USA 105, 7875–7880. Yang, X., Cao, P., and Su¨dhof, T.C. (2013). Deconstructing complexin function in activating and clamping Ca2+-triggered exocytosis by comparing knockout and knockdown phenotypes. Proc. Natl. Acad. Sci. USA 110, 20777–20782. Yoon, J.J., Suri, V., Nicholson, L.F.B., and Green, C.R. (2011). Improved Immunohistochemical Protocol for Organotypic Brain Slice Cultures. Microsc. Anal. 25, 5–7.

Watanabe, S., Rost, B.R., Camacho-Pe´rez, M., Davis, M.W., So¨hl-Kielczynski, B., Rosenmund, C., and Jorgensen, E.M. (2013). Ultrafast endocytosis at mouse hippocampal synapses. Nature 504, 242–247.

Yu, S.-C., Klosterman, S.M., Martin, A.A., Gracheva, E.O., and Richmond, J.E. (2013). Differential roles for snapin and synaptotagmin in the synaptic vesicle cycle. PLoS ONE 8, e57842.

Weber, J.P., Reim, K., and Sørensen, J.B. (2010). Opposing functions of two sub-domains of the SNARE-complex in neurotransmission. EMBO J. 29, 2477–2490.

Zhang, B., Koh, Y.H., Beckstead, R.B., Budnik, V., Ganetzky, B., and Bellen, H.J. (1998). Synaptic vesicle size and number are regulated by a clathrin adaptor protein required for endocytosis. Neuron 21, 1465–1475.

Weimer, R.M., Gracheva, E.O., Meyrignac, O., Miller, K.G., Richmond, J.E., and Bessereau, J.-L. (2006). UNC-13 and UNC-10/rim localize synaptic vesicles to specific membrane domains. J. Neurosci. 26, 8040–8047.

Zhang, Z., Wang, D., Sun, T., Xu, J., Chiang, H.-C., Shin, W., and Wu, L.-G. (2013). The SNARE proteins SNAP25 and synaptobrevin are involved in endocytosis at hippocampal synapses. J. Neurosci. 33, 9169–9175.

Wragg, R.T., Snead, D., Dong, Y., Ramlall, T.F., Menon, I., Bai, J., Eliezer, D., and Dittman, J.S. (2013). Synaptic vesicles position complexin to block spontaneous fusion. Neuron 77, 323–334.

Zimmermann, J., Trimbuch, T., and Rosenmund, C. (2014). Synaptobrevin 1 mediates vesicle priming and evoked release in a subpopulation of hippocampal neurons. J. Neurophysiol. 112, 1559–1565.

Neuron 84, 416–431, October 22, 2014 ª2014 Elsevier Inc. 431