The shRNAs of syntaxin-1, Munc13-1, and Munc18-1 were used as previously reported (Zhou et al., 2013; Wang et al., 2017, 2020). To knockdown the syntaxin-1a and 1b, the shRNA oligonucleotides (sense sequences: TCGACCAGAGGCAGCTGGAGATCACTTCAAGAGAGTGA TCTCCAGCTGCCTCTGGTTTTTTGGAAAT and CGCGCC CGATCATCATTTGCTGTGTGTTCAAGAGACACACAGCAA ATGATGATCATTTTTTTGGAAA) were designed to target two conserved regions, and inserted into XhoI/XbaI and AscI/RsrII sites of lentiviral vector L309, respectively. The shRNA oligonucleotides (sense sequences: GTCTGT CCACTCTCTCATCCCATGGGATGAGAGAGTGGACAGAC) were inserted into XhoI/XbaI sites of L309 to silence the Munc18-1 expression. To silence the expression of Munc13-1, the shRNA oligonucleotides (sense sequences: CCGGCCCGTGTGAAACAAAGGTTTACTCGAGTAAACCT TTGTTTCACACGGGTTTTT) were inserted into XhoI/XbaI sites of L309. Munc18-1 and its mutant sequences, syntaxin-1 and its mutant sequences, or Munc13-1 C1C2BMUN domain and its mutant sequences were inserted into the BamHI/EcoRI sites.

Rat Munc18-1 (residues 1–594), Munc18-1 mutants P335A, KKEE (K332E K333E), and rat syntaxin-1, syntaxin-1 mutants LEAA, RIAA were cloned into pGEX-KG vector that includes an N-terminal GST tag. Full-length rat synaptobrevin-2 were cloned into pGEX-6p-1 vector. Rat Mun13-1 MUN domain were constructed into pET28a vector.

HEK293T cells (CRL-11268, ATCC) were cultured in Dulbecco’s modified Eagle’s medium (Gibco), supplemented with 10% (v/v) fetal bovine serum and penicillin-streptomycin (50 and 50 μg/mL).

Neuron cultures were obtained from the cortex of newborn Kunming mice as described previously (Wang et al., 2019). Briefly, the cerebral cortex was dissected from postnatal day 0 (P0) mice, digested with 0.25% trypsin at 37^°C for 12 min, blown and sucked into single cells. The single cell suspension was obtained after passing the cell strainer. Neurons were planted on glass coverslips coated with poly-L-lysine and cultured in the medium MEM (Gibco) supplemented with 0.5% (w/v) glucose, 100 mg/L transferrin, 5% (v/v) fetal bovine serum, 2% (v/v) B27 (Gibco), and 2 mM Ara-C (Sigma). The culture medium was changed on DIV (day in vitro) 1, 4 and 9, in the manner of 500 μL out and 600 μL in. Neurons were infected with lentiviruses at DIV 5–7, and analyzed at DIV13–14.

Neurons and HEK293T cells were all grown at 37^°C, 5% CO₂ in a cell incubator (Thermo).

Lentivirus preparation from HEK293T cells as described previously. In brief, lentivirus vectors (L309 or its constructions) and three helper plasmids (pRSV-REV, pMDLg-pRRE, and pVSVG) were co-transfected by polyethylenimine (PEI) into HEK293T cells. The ratio of each component was L309: pRSV-REV: pMDLg-pRRE: pVSVG: PEI = 3: 2: 2: 1: 24. The plasmids and PEI were in the medium Opti-MEM (Gibco), incubated for 30 min at room temperature, and added into the HEK293T cells. The cell culture medium containing virus was harvested at 48 h after transfection, and then precleaned with 1,000 g centrifuge. The virus-containing medium was overlaid on a sucrose-containing buffer (50 mM Tris–HCl pH 7.4, 100 mM NaCl, 0.5 mM EDTA) at 4:1 (v/v), and centrifuged in 10,000 g for 4 h at 4^°C (Jiang et al., 2015). After centrifugation, the buffer and medium were removed. PBS buffer was added into the tube carefully and incubated in a 4^°C fridge overnight for virus resuspension. All steps were carried out under level II biosafety conditions.

Electrophysiological recording was performed in whole-cell patch clamping mode by using HEKA EPC10 amplifier as described previously (Gong et al., 2016). Micropipettes were pulled from borosilicate glass capillary tubes (Word Precision Instruments) by using P-97 puller (Sutter instruments). The micropipette solution contained 120 mM CsCl, 10 mM HEPES, 10 mM EGTA, 0.3 mM Na-GTP, 3 mM Mg-ATP. Adjust the pH to 7.2–7.4 with CsOH, and adjust the osmotic pressure with dd H₂O to about 305. Add 5 mM QX-314 to the micropipette solution before use for the evoke release recording. The bath solution contained 140 mM NaCl, 5 mM KCl, 2 mM MgCl₂, 2 mM CaCl₂, 10 mM HEPES, and 10 mM glucose. Adjust the pH to 7.2–7.4 with NaOH, and adjust the osmotic pressure with dd H₂O to about 315. Inhibitory postsynaptic currents (IPSCs) were isolated by adding the AMPA and NMDA receptor blockers CNQX (20 μM) and AP-5 (50 μM) to the bath solution. Evoked IPSC was recorded in the stimulus pulse (90 μA) mini IPSCs were recorded in the presence of tetrodotoxin (1 μM) for action-potential blocking. Sucrose IPSC was measured with a 0.5 M sucrose application in the bath solution. Electrophysiological data analyzed by Clampfit 10 (Molecular devices).

All recombinant proteins were expressed in Escherichia coli BL21 (DE3) and purified as described previously (Wang et al., 2020). In short, the cells were grown to the OD₆₀₀ ≈ 0.8 at 37^°C and induced by 0.5 mM IPTG (isopropyl-β-thiogalactoside) at 16 or 20^°C. Cultured cells were harvested by centrifugation at 5,000 g, 4^°C for 20 min, resuspended by PBS, and lysed with high pressure crusher in the presence of 1 mM PMSF (phenylmethanesulfonyl fluoride), 5 mM DTT (1,4-dithio-DL-threitol), and 1 mM EDTA (pH 8.0), then centrifugated at 31,000 g. For the GST-fused proteins, the supernatants were collected and mixed with glutathione Sepharose 4B (GE Healthcare) affinity media. Thrombin (Sigma) was used for removing GST fusion tag. For hexa-histidine-tagged proteins, the supernatants were collected and mixed with Nickel-NTA agarose (Qiagen) affinity media, and eluted with buffer containing 300 mM imidazole. The eluted proteins were further purified by size-exclusion chromatography. All recombinant proteins were expressed in Escherichia coli BL21 (DE3) and purified as described previously (Wang et al., 2020). In short, the cells were grown to the OD600 ≈ 0.8 at 37^°C and induced by 0.5 mM IPTG (isopropyl-β-thiogalactoside) at 16 or 20^°C. Cultured cells were harvested by centrifugation at 5,000 g, 4^°C for 20 min. For GST-fused proteins, cells were resuspended by PBS (pH 7.4) supplemented with 1 mM PMSF (phenylmethanesulfonyl fluoride), 5 mM DTT (1,4-dithio-DL-threitol), and 1 mM EDTA (pH 8.0), lysed with high pressure crusher and then centrifugated at 31,000 g. the supernatants were collected and mixed with glutathione Sepharose 4B (GE Healthcare) affinity media. Thrombin (Sigma) was used for removing GST tag. For hexa-histidine-tagged proteins, cells were resuspended with Tris–HCl buffer (50 mM Tris–HCl pH 8.0, 150 mM NaCl, 10% glycerol) containing 5 mM imidazol. Cells were lysed and centrifugated as GST-fused proteins. The supernatants were incubated with Nickel-NTA agarose (Qiagen) affinity media. Hexa-histidine-fused proteins bound to agarose affinity media were eluted with Tris–HCl buffer buffer containing 300 mM imidazole. Both eluted proteins were further purified by size-exclusion chromatography.

Native gel was prepared with polyacrylamide without SDS. To obtain the Munc18-1/syntaxin-1 complex, purified Munc18-1 and syntaxin-1 were incubated together in the ratio of 1.2:1 at 4^°C overnight. Then, the Munc18-1/syntaxin-1 (3 μM/2.5 μM) complex was mixed with synaptobrevin-2 (10 μM) and SNAP-25 (10 μM), with or without Munc13-1 MUN domain (30 μM) at 30^°C for 2 h. Then samples were mixed with 5 × loading buffer and electrophoresis was carried out at 4^°C for 2 h in 15% native gel. Gels were stained with Coomassie Brilliant Blue.

Purified synaptobrevin-2 (residues 29–93, with mutation of S61C) and SNAP-25 (with native cysteines mutated to serine and mutation of Q197C) were mixed and labeled with donor-dye BODIPY FL-maleimide (Molecular Probes) and acceptor dye tetramethylrhodamine-5-maleimide. The ratio of protein and dye was 1:5. The mixture rotated overnight at 4^°C avoiding light in buffer containing 25 mM HEPES (pH 7.4), 150 mM KCl, and 10% (v/v) glycerol. Excess dyes were removed by PD-10 desalting columns (GE Healthcare) after addition of 10 mM DTT. To obtain the Munc18-1/syntaxin-1 or its mutant complex, Munc18-1, or its mutants P335A/KKEE and syntaxin-1 or its mutants LEAA/RIAA were pre-incubated with a ratio of 1.5:1 at 4^°C overnight. The labeled synaptobrevin-2 (2 μM) and SNAP-25 (10 μM) were mixed with Munc18-1/syntaxin-1 (5 μM) or its mutant complex (5 μM) and Munc13-1 MUN domain (30 μM). All ensemble FRET experiments were performed by a QM 40 spectrofluorometer (Photon Technology Incorporated) in a 1-cm quartz cuvette at 30^°C. The excitation and emission wavelength were 485 and 513 nm. FRET efficiency was calculated by the formula: E = (F₀-F_obs)/F₀ × 100% (E: FRET efficiency; F₀: the initial fluorescent intensity; F_obs, the observed fluorescent intensity).

Prism 6.01 (GraphPad) was used for statistical analysis, all of which are described in figure legends.

In our previous in vitro studies, we showed that the opening of synatxin-1 linker region and the extension of Munc18-1 domain 3a are driven by the interaction of the Munc13-1 MUN domain with the Munc18-1/syntaxin-1 complex (Wang et al., 2017, 2020). To verify this in vivo, we employed the crucial mutations that were previously identified (Figure 1), including N1128A/F1131A in Munc13-1 MUN domain (Yang et al., 2015), R151A/I155A and L165A/E166A in the syntaxin-1 linker region (Dulubova et al., 1999; Wang et al., 2017), and K332E/K333E and P335A in the Munc18-1 domain 3a (Han et al., 2014; Parisotto et al., 2014; Wang et al., 2020). The NF sequence in the MUN domain and RI sequence in the syntaxin-1 linker region correspond to the essential binding sites responsible for the MUN domain targeting to the syntaxin-1 linker region (Figure 1C). The syntaxin-1 LEAA mutation helps to open the syntaxin-1 linker region (Wang et al., 2017) and the Munc18-1 P335A mutation promotes the extension of domain 3a (Parisotto et al., 2014).

We used a knockdown (KD)-rescue approach in cultured mouse cortical neurons. The endogenous expression of Munc13-1 and syntaxin-1 (syntaxin-1a and 1b) was suppressed by lentivirus-delivered short hairpin RNAs (shRNAs), as previously reported (Zhou et al., 2013; Wang et al., 2017). We then monitored synaptic exocytosis with the expression of the main functional domain of Munc13-1, C1C2BMUN, or its NFAA mutant, as well as syntaxin-1a or its LEAA mutant. Expression of C1C2BMUN and syntaxin-1a rescued both the frequency of spontaneous mini IPSCs and the amplitude and charge transfer of action potential-evoked IPSCs (Figures 2A,B). In contrast, the NFAA mutant of C1C2BMUN cannot rescue any of these parameters (Figures 2A,B), implying that the NFAA mutant fails to support synaptic vesicle exocytosis, consistent with previous results (Wang et al., 2017). Interestingly, the expression of the syntaxin-1 LEAA mutant rescued the defects caused by the C1C2BMUN NFAA mutation, suggesting that the opening of the syntaxin-1 linker region bypasses the role of the Munc13-1 NF sequence in both spontaneous and evoked release. In addition, the mini IPSC amplitude was unchanged under all conditions (Figure 2A), ruling out major postsynaptic effects of these mutant proteins.

We next analyzed vesicle priming by characterizing the size of the readily releasable pool of vesicles induced by the application of a hypertonic sucrose solution (Rosenmund and Stevens, 1996). The significant decrease in the sucrose-induced charge transfer observed in neurons expressing the C1C2BMUN NFAA mutant was rescued by expressing the syntaxin-1 LEAA mutant but not by expressing wild-type (WT) syntaxin-1 (Figure 2C). Therefore, these in vivo data indicate that the open form of syntaxin-1 linker region (syntaxin-1 LEAA) can compensate the function of the Munc13-1 NF sequence in priming of synaptic vesicles.

We thought to verify the above results by using a previously established native-PAGE assay that can detect the transition from the Munc18-1/syntaxin-1 complex to the SNARE complex catalyzed by the MUN domain (Wang et al., 2020). Consistent with previous results, the transition to the SNARE complex assembly from the Munc18-1/syntaxin-1 complex was promoted by the MUN domain but not by the MUN NFAA mutant in the presence of synaptobrevin-2 and SNAP-25 (Figure 3A). However, when the Munc18-1/syntaxin-1 LEAA mutant was used, the transition can be promoted the MUN NFAA mutant (Figure 3A), suggesting that the open form of syntaxin-1 compensate the function of the NF sequence in SNARE assembly.

We further characterized the function of the syntaxin-1 LE sequence using an ensemble FRET assay (Wang et al., 2017). The FRET experiments demonstrated that the NFAA mutation abrogated the MUN-catalyzed transition to the SNARE complex from the Munc18-1/syntaxin-1 complex (Figures 3B,C). Likewise, the syntaxin-1 LEAA mutant promoted the transition to the SNARE complex by rescuing the defects caused by the NFAA mutation (Figures 3B,C).

Collectively, our data confirmed that the MUN domain is crucial for synaptic vesicle release due to the functional importance of its NF sequence in the transition from the Munc18-1/syntaxin-1 complex to the SNARE complex. Moreover, our results showed that the open form of the syntaxin-1 linker region was equal to the function of the Munc13-1 NF sequence, emphasizing that syntaxin-1 linker region opening is driven by the Munc13-1 MUN interaction with Munc18-1/syntaxin-1.

Two conformational changes occur in the Munc18-1/syntaxin-1 complex upon its activation by Munc13-1. However, the relationship between the two conformational changes under physiological conditions has not been fully understood. Our previous in vitro studies have shown that the opening of syntaxin-1 linker region driven by the MUN domain leads to the extension of Munc18-1 domain 3a (Wang et al., 2020). The MUN domain was unable to open the linker region when syntaxin-1 contains the RIAA mutation, as the RI sequence mediates interaction with the MUN NF pocket (Wang et al., 2017). To further figure out the action order of the two conformational changes, we suppressed the endogenous expression of Munc18-1 and syntaxin-1 (syntaxin-1a and 1b) using lentivirus-delivered shRNAs in mouse cortical neurons. We found that the expression of the syntaxin-1 RIAA mutant decreased both the frequency of mini IPSCs and the amplitude and charge transfer of evoked IPSCs (Figures 4A,B), which is consistent with previous findings (Wang et al., 2017). Interestingly, this defect was rescued by the expression of Munc18-1 P335A, but not by Munc18-1 WT (Figures 4A,B). These results imply that Munc18-1 P335A bypasses the opening of syntaxin-1 linker region in synaptic vesicle exocytosis.

In addition, we analyzed vesicle priming, as accessed by measurements of the readily releasable vesicle pool (Figure 4C). As before, the decrease in sucrose-induced charge transfer in neurons expressing the RIAA mutant was rescued by expressing Munc18-1 P335A but not Munc18-1 WT (Figure 4C), indicating that Munc18-1 P335A could overcome the defect of opening of sytanxin-1 linker region in the priming of synaptic vesicles. Thus, our results suggest that the extension of domain 3a actions downstream of syntaxin-1 linker region opening. Our data confirm that the extension of Munc18-1 domain 3a results from of the interaction of the Munc18-1/syntaxin-1 complex with Munc13-1 MUN domain.

To corroborate the above results, we used native-PAGE and ensemble FRET assays to analyses the MUN-catalyzed transition from the Munc18-1/syntaxin-1 complex to the SNARE complex. Consistent to previous results, the syntaxin-1 RIAA mutant abolished SNARE complex formation in the presence of MUN (Figure 5 and Supplementary Figure 1), confirming that RI residues are crucial for the MUN-catalyzed transition to the SNARE complex. However, the SNARE complex formed when Munc18-1 P335A was complexed with the syntaxin-1 RIAA mutant (Figure 5 and Supplementary Figure 1), indicating that the defect caused by syntaxin-1 RIAA was rescued by Munc18-1 P335A. Our results suggest that the extension of domain 3a bypasses the requirement for opening of syntaxin-1 linker region in SNARE complex assembly.

In summary, the above results suggest that the extension of Munc18-1 domain 3a and the opening of the syntaxin-1 linker region are caused by the interaction between MUN and Munc18-1/syntaxin-1. The extension of domain 3a is a subsequent event following the opening of syntaxin-1 linker region.

Synaptic exocytosis and SNARE complex assembly can occur when Munc18-1 domain 3a is extended (P335A mutant, gain-of-function) even when the syntaxin-1 linker region adopts a closed conformation (RIAA mutant, loss-of-function). However, when the syntaxin-1 linker region is in a constitutive opened state (LEAA, gain-of-function), it is unclear whether the state of Munc18-1 domain 3a influences synaptic exocytosis. According to previous results, the KKEE mutation in domain 3a abolishes the vesicle exocytosis because of its inability to interact with the H3 domain of syntaxin-1 which impedes the Munc18-1 function in templating SNARE complex assembly. Hence, we use the Munc18-1 KKEE mutant and the syntaxin-1 LEAA mutant to address this issue.

The endogenous expression of Munc18-1 and syntaxin-1 (syntaxin-1a and 1b) was suppressed by lentivirus-delivered shRNAs in mouse cortical neurons, as described above. The Munc18-1 KKEE mutant, which abolished the template function of domain 3a, was unable to rescue the decrease in the mini IPSC frequency, the evoked IPSC amplitude, or the evoked charge transfer. Overexpression of the syntaxin-1 LEAA mutant rescued the decrease in the mini IPSC frequency and evoked IPSC amplitude and charge transfer caused by Munc18-1 KKEE (Figures 6A,B). We also analyzed vesicle priming (Figure 6C). Likewise, the decrease in sucrose-evoked charge transfer caused by the Munc18-1 KKEE mutant was rescued by expressing the syntaxin-1 LEAA mutant, but not by syntaxin-1 WT (Figure 6C), indicating that the opening of the syntaxin-1 linker region promotes vesicle priming in a way that is not dependent on binding of domain 3a to the syntaxin-1 H3, which is not expected. To address this in vitro, we further conducted a SNARE complex assembly experiment using FRET assay. The KKEE mutant of Munc18-1 abolished the transition of Munc18-1/syntaxin-1 to the SNARE complex catalyzed by MUN (Figure 7). However, the syntaxin-1 LEAA mutant promoted the formation of the SNARE complex by rescuing the defects caused by the KKEE mutation (Figure 7). A plausible explanation is that the syntaxin-1 LEAA combined with the Munc18-1 KKEE mutation exerts a strong effect leading to the escape of the syntaxin-1 H3 from Munc18-1 clamping. In this circumstance, the H3 assembles with synaptobrevin-2 and SNAP-25 to form the SNARE complex in a manner independent of Munc18-1 and Munc13-1.