MUNC18-1 (also known as syntaxin-binding protein-1, encoded by Stxbp1) binds to syntaxin-1. Together, these proteins regulate synaptic vesicle exocytosis and have a separate role in neuronal viability. In Stxbp1 null mutant neurons, syntaxin-1 protein levels are reduced by 70%. Here, we show that dynamin-1 protein levels are reduced at least to the same extent, and transcript levels of Dnm1 (which encodes dynamin-1) are reduced by 50% in Stxbp1 null mutant brain. Several, but not all, other endocytic proteins were also found to be reduced, but to a lesser extent. The reduced dynamin-1 expression was not observed in SNAP25 null mutants or in double-null mutants of MUNC13-1 and -2 (also known as Unc13a and Unc13b, respectively), in which synaptic vesicle exocytosis is also blocked. Co-immunoprecipitation experiments demonstrated that dynamin-1 and MUNC18-1 do not bind directly. Furthermore, MUNC18-1 levels were unaltered in neurons lacking all three dynamin paralogues. Finally, overexpression of dynamin-1 was not sufficient to rescue neuronal viability in Stxbp1 null mutant neurons; thus, the reduction in dynamin-1 is not the single cause of neurodegeneration of these neurons. The reduction in levels of dynamin-1 protein and mRNA, as well as of other endocytosis proteins, in Stxbp1 null mutant neurons suggests that MUNC18-1 directly or indirectly controls expression of other presynaptic genes.

The protein machinery regulating the synaptic vesicle cycle is tightly controlled, which is necessary to maintain the intricate balance between exocytosis and endocytosis required for reliable, synchronous synaptic transmission (Südhof, 2013). One of the key exocytic proteins is the Sec1/Munc18 (SM) protein MUNC18-1 (also known as syntaxin-binding protein-1, encoded by Stxbp1), which is essential for the fusion of synaptic vesicles (Verhage et al., 2000; Toonen and Verhage, 2007). Although initial brain development occurs normally in Stxbp1 null mutant (knockout, KO) mice, despite the lack of synaptic transmission, extensive neurodegeneration is observed during later stages of neurodevelopment, starting with the structures that develop first (Verhage et al., 2000). In vitro, Stxbp1 KO neurons degenerate within a few days, before the onset of synaptogenesis (Santos et al., 2017), and degeneration is also observed in vivo after Stxbp1 depletion in mature neurons carrying conditional deletion alleles (Heeroma et al., 2004). Expression of non-cognate MUNC18 paralogues, MUNC18-2 (encoded by Stxbp2) or MUNC18-3 (encoded by Stxbp3), restores the viability of Stxbp1 KO neurons without restoring synaptic transmission (Santos et al., 2017). Thus, depletion of MUNC18-1 leads to neurodegeneration and a total arrest of the synaptic vesicle cycle.

Exocytosis and endocytosis are known to be tightly coupled, and changes in one affect the other. Alterations in neuronal activity, Ca2+ influx or synaptic vesicle exocytosis regulate the mode, extent and kinetics of endocytosis (Wu et al., 2014; Lou, 2018; Maritzen and Haucke, 2018). Several presynaptic proteins involved in regulating the synaptic vesicle cycle play a role in this coupling, including synaptotagmins (Poskanzer et al., 2003; Nicholson-Tomishima and Ryan, 2004), VAMP2 (Deák et al., 2004), SNAP25 and syntaxins (Xu et al., 2013; Zhang et al., 2013). In the reverse direction, inactivation of the endocytic protein Dynamin in the Drosophila mutant shibirets leads to depletion of synaptic vesicles and abolished synaptic transmission (Koenig et al., 1983). This raises the question of whether depletion of MUNC18-1, which leads to a complete halt of the synaptic vesicle cycle, affects other exocytic and endocytic proteins.

Previous analyses have identified differential regulation of synaptic genes in Stxbp1 null neurons (Bouwman et al., 2006). Recently, higher-resolution RNA sequencing and mass spectrometry (van Berkel et al., 2022) has revealed that a subset of endocytic proteins is downregulated in Stxbp1 null brains. Of these proteins, the reduction in dynamin-1 levels has been considered of particular interest, because neurons lacking the three main dynamin paralogues, like Stxbp1 null neurons, also degenerate (Park et al., 2013; Moro et al., 2021). Dynamin proteins play a major role in catalyzing the membrane fission step of endocytosis and intracellular trafficking (Ferguson and De Camilli, 2012; Arriagada-Diaz et al., 2022), and have been implicated in the coupling of exocytosis and endocytosis (Haucke et al., 2011).

In this study, we characterized the effect of MUNC18-1 depletion on dynamin levels and investigated a possible interdependence in the neurodegeneration observed in MUNC18-1- and dynamin-deficient neurons. We observed ∼80% reduction in dynamin protein and transcript levels in Stxbp1 KO neurons; however, MUNC18-1 levels were not reduced in dynamin KO neurons. No direct interaction was observed between dynamin-1 and MUNC18-1. Stxbp1 KO neurons showed a deficit in dynamin-dependent receptor endocytosis. Overexpression of dynamin-1 did not rescue the compromised viability of Stxbp1 KO neurons. Taken together, these findings lead us to conclude that depletion of MUNC18-1 influences the mRNA and protein expression of presynaptic genes involved in the exocytosis and endocytosis of synaptic vesicles.

Dynamin-1 protein and transcript levels are reduced in Stxbp1 KO neurons

Recently, mass spectrometry analysis has shown that a substantial number of synaptic proteins are differentially regulated in cultured Stxbp1 KO neurons (van Berkel et al., 2022). In order to investigate whether depletion of MUNC18-1 significantly affects the levels of endocytic proteins, we used the SynGO tool (Koopmans et al., 2019) to perform an enrichment analysis for all proteins that were found to be downregulated in Stxbp1 KO neurons at 2 days in vitro (DIV) in the previously published mass spectrometry dataset (van Berkel et al., 2022). Significant enrichment was found for the ontology term ‘synaptic vesicle endocytosis’ (q=5.41×10−5; Fig. 1A). We found that levels of several, but not all, endocytic proteins were significantly reduced in Stxbp1 KO neurons compared to their levels in wild-type (WT) neurons (Fig. 1B). Hence, Stxbp1 KO affects the levels of a select group of proteins involved in synaptic vesicle endocytosis, but does not cause a general downregulation of the entire endocytic machinery.

Fig. 1.

Reduction of dynamin-1 levels in Stxbp1 KO neurons at protein and transcript level. (A) SynGO sunburst plot showing enrichment analysis of all proteins found to be differentially downregulated in Stxbp1 KO neurons at DIV2 in the mass spectrometry dataset (van Berkel et al., 2022). Dashed box indicates two a priori-defined GO terms of interest shown to the right of the plot, of which one – synaptic vesicle (SV) endocytosis – showed significant enrichment. Not sign., not significant. (B) Heatmap showing the log2 fold change in Stxbp1 KO neurons at DIV2 measured for all proteins classified in the two pre-defined GO terms from panel A that were also detected in the mass spectrometry experiments. Asterisks indicate proteins with FDR-corrected P<0.05. (C) Western blot showing reduced levels of dynamin proteins (antibody used recognizes all three dynamin paralogues) in Stxbp1 KO cortex lysate, but not in SNAP25 KO cortex lysate, at E18. Actin served as loading control, MUNC18-1 (M18-1) and SNAP25 levels were assessed to confirm sample WT and KO status. (D) Dynamin levels are reduced in Stxbp1 KO cortex, as also shown in panel B, but not in MUNC13-1/2 DKO (M13-1/2 DKO) samples. Actin served as loading control. Blots shown in C and D are representative of two experiments. (E) Levels of dynamin-1, the main dynamin paralogue expressed in the brain, are reduced in the brain of two different Stxbp1 null mouse strains – Stxbp1 KO (Verhage et al., 2000) and Stxbp1 flox (Heeroma et al., 2004) – at E18. MUNC18-1 is shown to confirm WT and KO status, and γ-tubulin (g-Tubulin) serves as loading control. (F) Dynamin-1 (Dyn1) levels are lower in Stxbp1 KO brain compared to levels in WT littermate brain. Dynamin-1 protein levels were normalized to the housekeeping protein (γ-tubulin) within each lane, and then normalized to the average protein levels in the (littermate) WT condition. Data are presented as median±IQR of n=4. (G) SynGO sunburst plot showing enrichment analysis of all transcripts found to be significantly downregulated in Stxbp1 KO hippocampal tissue in the RNA sequencing dataset (van Berkel et al., 2022). The dashed box and panel on the right show the same two GO terms as highlighted in A, neither of which show significant enrichment at transcript level. (H) Log2 fold change (Stxbp1 KO over WT) in transcript levels of the three dynamin paralogues, as measured by qPCR. Data are presented as median±IQR of n=4–5. (I) Log2 fold change (F.C.) at protein level, as measured by mass spectrometry (y-axis), against the log2 fold change at RNA level, as measured by RNA sequencing (x-axis). All endocytic proteins detected at both timepoints in the proteomics dataset are plotted, as for panel B. Yellow, significantly regulated at protein level but not at RNA level; green, significantly regulated at both RNA and protein level.

Fig. 1.

Reduction of dynamin-1 levels in Stxbp1 KO neurons at protein and transcript level. (A) SynGO sunburst plot showing enrichment analysis of all proteins found to be differentially downregulated in Stxbp1 KO neurons at DIV2 in the mass spectrometry dataset (van Berkel et al., 2022). Dashed box indicates two a priori-defined GO terms of interest shown to the right of the plot, of which one – synaptic vesicle (SV) endocytosis – showed significant enrichment. Not sign., not significant. (B) Heatmap showing the log2 fold change in Stxbp1 KO neurons at DIV2 measured for all proteins classified in the two pre-defined GO terms from panel A that were also detected in the mass spectrometry experiments. Asterisks indicate proteins with FDR-corrected P<0.05. (C) Western blot showing reduced levels of dynamin proteins (antibody used recognizes all three dynamin paralogues) in Stxbp1 KO cortex lysate, but not in SNAP25 KO cortex lysate, at E18. Actin served as loading control, MUNC18-1 (M18-1) and SNAP25 levels were assessed to confirm sample WT and KO status. (D) Dynamin levels are reduced in Stxbp1 KO cortex, as also shown in panel B, but not in MUNC13-1/2 DKO (M13-1/2 DKO) samples. Actin served as loading control. Blots shown in C and D are representative of two experiments. (E) Levels of dynamin-1, the main dynamin paralogue expressed in the brain, are reduced in the brain of two different Stxbp1 null mouse strains – Stxbp1 KO (Verhage et al., 2000) and Stxbp1 flox (Heeroma et al., 2004) – at E18. MUNC18-1 is shown to confirm WT and KO status, and γ-tubulin (g-Tubulin) serves as loading control. (F) Dynamin-1 (Dyn1) levels are lower in Stxbp1 KO brain compared to levels in WT littermate brain. Dynamin-1 protein levels were normalized to the housekeeping protein (γ-tubulin) within each lane, and then normalized to the average protein levels in the (littermate) WT condition. Data are presented as median±IQR of n=4. (G) SynGO sunburst plot showing enrichment analysis of all transcripts found to be significantly downregulated in Stxbp1 KO hippocampal tissue in the RNA sequencing dataset (van Berkel et al., 2022). The dashed box and panel on the right show the same two GO terms as highlighted in A, neither of which show significant enrichment at transcript level. (H) Log2 fold change (Stxbp1 KO over WT) in transcript levels of the three dynamin paralogues, as measured by qPCR. Data are presented as median±IQR of n=4–5. (I) Log2 fold change (F.C.) at protein level, as measured by mass spectrometry (y-axis), against the log2 fold change at RNA level, as measured by RNA sequencing (x-axis). All endocytic proteins detected at both timepoints in the proteomics dataset are plotted, as for panel B. Yellow, significantly regulated at protein level but not at RNA level; green, significantly regulated at both RNA and protein level.

Of these proteins, the reduction in levels of dynamin-1 (P=0.0029 in the mass spectrometry dataset; van Berkel et al., 2022) was considered to be of particular interest, because inactivation of Dynamin leads to a loss of synaptic transmission in Drosophila (Koenig et al., 1983), and depletion of the three main dynamin paralogues causes neurodegeneration (Park et al., 2013; Moro et al., 2021). To confirm this reduction, we performed quantitative western blot analysis of WT and Stxbp1 KO mouse pups at embryonic day (E)18. To avoid indirect or non-specific effects of massive neurodegeneration occurring in the lower brain regions at E18 (Verhage et al., 2000), we used cortex tissue, which is devoid of neurodegeneration at E18. These analyses revealed a drastic reduction in dynamin protein levels (Fig. 1C). As a further validation of the mass spectrometry data, SNAP91 (also known as AP180) levels were found to be reduced in Stxbp1 KO cortex tissue, whereas levels of the endophilin SH3GL2 were unaltered (Fig. S1A). No reduction in dynamin levels was observed in cortex tissue from SNAP25 null mice (Fig. 1C). Dynamin levels were also not affected in cortex tissue of mice with double knockout (DKO) of MUNC13-1 and -2 (also known as Unc13a and Unc13b, respectively, referred to collectively here as MUNC13-1/2; Fig. 1D), which also causes a complete loss of synaptic transmission (Varoqueaux et al., 2002). Thus, the reduction in levels of dynamin-1 is not a general consequence of a lack of synaptic transmission.

Initially, an antibody was used that recognizes proteins encoded by all three dynamin genes (Dnm1, Dnm2 and Dnm3; Fig. S1B). Subsequently, selective antibodies were used for detection of dynamin-1 and dynamin-2. An antibody advertised as selective for dynamin-3 was found to also recognize dynamin-1 overexpressed in human embryonic kidney 293T (HEK293) cells (Fig. S1B). Using these selective antibodies, we observed that dynamin-1 levels were reduced by 70.6% in Stxbp1 KO mouse brains (Fig. 1E,F), whereas dynamin-2 levels were not reduced (Fig. S1C).

The Dnm1 and Stxbp1 genes are located on the same chromosome in the mouse genome. The question could be raised whether dynamin-1 reduction is a consequence of the Stxbp1 KO strategy. Removal of exons 2–6 (Verhage et al., 2000) could affect transcription of the Dnm1 locus, via potential regulatory effects of this sequence or effects on chromatin structure. However, a similar reduction (71.4%; Fig. 1E,F) was observed in a second Stxbp1 KO mouse strain that was generated using a different strategy, floxing exon 2 only (Heeroma et al., 2004). Thus, it is unlikely that the reduction in dynamin-1 is a consequence of the KO strategy.

In the same recently published study, mRNA expression levels in hippocampal tissue from WT and Stxbp1 KO mice were assessed by RNA sequencing (van Berkel et al., 2022). To analyse whether the altered protein levels are explained by changes in gene expression, we filtered this published dataset for mRNAs encoding endocytic proteins. Enrichment analysis using the SynGO tool revealed no significant enrichment for the two a priori defined endocytosis-related ontology terms (Fig. 1G). However, transcript levels of Dnm1 (encoding dynamin-1) were significantly reduced (adjusted P=6.04×10−7; van Berkel et al., 2022). Quantitative PCR (qPCR) analysis confirmed these findings (Fig. 1H). Comparisons of transcriptomic versus proteomic changes showed that only two endocytic genes were significantly reduced at both mRNA and protein level (Fig. 1I: indicated in green, versus reduction at protein level only in yellow). Thus, dynamin-1 levels are reduced at the transcript level. This raises the question of whether MUNC18-1 is present in the neuronal nucleus to regulate transcription. Thus, we investigated the localization of MUNC18-1 in WT neurons, with Stxbp1 KO neurons as a negative control. No specific MUNC18-1 signal was observed in the nucleus (Fig. S2).

Taken together, these results show that dynamin-1 RNA and protein levels are severely reduced in Stxbp1 KO brain, which does not reflect a general downregulation of the entire endocytic machinery. This reduction is not observed in SNAP25 KO nor in MUNC13-1/2 DKO neurons.

Dependence between dynamin-1 and MUNC18-1 is not reciprocal, and no co-immunoprecipitation of the two proteins is observed in heterologous cells

The reduction in dynamin-1 levels in Stxbp1 KO mouse brains raises the question of whether MUNC18-1 levels are also reduced in the absence of dynamin. Thus, primary mouse neurons were cultured that lacked one, two or all three dynamin paralogues: either dynamin-3, or both dynamin-1 and dynamin-3 were constitutively knocked out, and conditional KO of dynamin-2 was achieved through expression of Cre recombinase (Fig. 2A) (Ferguson et al., 2007; Raimondi et al., 2011; Park et al., 2013). MUNC18-1 levels were not different between control neurons (dynamin-3 KO), neurons with DKO of dynamin-1 and -3 (dynamin-1/3 DKO) or neurons with triple knockout (TKO) of dynamin-1, -2 and -3 (dynamin-1/2/3 TKO) (Fig. 2B). Thus, the dependence of dynamin-1 expression on MUNC18-1 is not reciprocal.

Fig. 2.

Connection between dynamin-1 and MUNC18-1 is not reciprocal and not explained by direct interaction. (A) Left: overview of the methodology to generate dynamin KO neurons. Neurons were plated from dynamin-3 single KO (SKO) and dynamin-1/3 DKO littermate pups at E18 and infected at DIV8 or DIV9 with either Cre or inactive Cre (ΔCre). All pups are homozygous for loxP sites allowing conditional deletion of dynamin-2, so that lentiviral infection with Cre generates dynamin-1/2/3 TKO neurons. Analyses were carried out at DIV14 or DIV15 (7 days post-infection), at which dynamin protein levels are abolished (see western blot, right; actin serves as loading control) but TKO neurons are still alive. Western blot is reproduced, with permission, from Moro et al. (2021). (B) MUNC18-1 levels are unaltered in dynamin-1 single KO (Dyn1 SKO), dynamin-1/3 DKO (Dyn1/3 DKO) or dynamin-1/2/3 TKO (Dyn1/2/3 TKO) neurons. Top: typical example of MUNC18-1 western blot with γ-tubulin (g-Tubulin) loading control. Bottom: MUNC18-1 levels normalized to γ-tubulin levels in each KO line. Horizontal bars indicate the median, n=4 experiments. (C,D) Co-immunoprecipitation (co-IP) in HEK293 cells. (C) Immunoprecipitation (IP) of MUNC18-1 (M18-1) reveals no binding to dynamin-1 (Dyn1), while expression of MUNC18-1 was confirmed and syntaxin-1 (Stx1) was detected in the pulldown. (D) IP of Dyn1, confirming Dyn1 expression and pulldown for SH3GL2–Flag, but no detection of MUNC18-1 in the IP sample. IP− lanes show the empty control. Blots shown are representative of two experiments.

Fig. 2.

Connection between dynamin-1 and MUNC18-1 is not reciprocal and not explained by direct interaction. (A) Left: overview of the methodology to generate dynamin KO neurons. Neurons were plated from dynamin-3 single KO (SKO) and dynamin-1/3 DKO littermate pups at E18 and infected at DIV8 or DIV9 with either Cre or inactive Cre (ΔCre). All pups are homozygous for loxP sites allowing conditional deletion of dynamin-2, so that lentiviral infection with Cre generates dynamin-1/2/3 TKO neurons. Analyses were carried out at DIV14 or DIV15 (7 days post-infection), at which dynamin protein levels are abolished (see western blot, right; actin serves as loading control) but TKO neurons are still alive. Western blot is reproduced, with permission, from Moro et al. (2021). (B) MUNC18-1 levels are unaltered in dynamin-1 single KO (Dyn1 SKO), dynamin-1/3 DKO (Dyn1/3 DKO) or dynamin-1/2/3 TKO (Dyn1/2/3 TKO) neurons. Top: typical example of MUNC18-1 western blot with γ-tubulin (g-Tubulin) loading control. Bottom: MUNC18-1 levels normalized to γ-tubulin levels in each KO line. Horizontal bars indicate the median, n=4 experiments. (C,D) Co-immunoprecipitation (co-IP) in HEK293 cells. (C) Immunoprecipitation (IP) of MUNC18-1 (M18-1) reveals no binding to dynamin-1 (Dyn1), while expression of MUNC18-1 was confirmed and syntaxin-1 (Stx1) was detected in the pulldown. (D) IP of Dyn1, confirming Dyn1 expression and pulldown for SH3GL2–Flag, but no detection of MUNC18-1 in the IP sample. IP− lanes show the empty control. Blots shown are representative of two experiments.

The protein levels of binding-partner proteins often depend on each other. Previous studies have shown that depletion of MUNC18-1 leads to a substantial reduction of its high affinity binding partner syntaxin-1 (herein referring to syntaxin-1A and -1B unless otherwise mentioned) at protein and RNA level (Voets et al., 2001; Bouwman et al., 2004, 2006; Toonen et al., 2005; Gulyás-Kovács et al., 2007; Santos et al., 2017). However, a direct interaction between dynamin and MUNC18-1, or between their orthologues, has not been reported so far. To assess a possible interaction between the two, HEK293 cells were transfected to express dynamin-1 and MUNC18-1, and pulldown experiments were performed. No co-precipitation was observed in pulldowns of either protein, whereas MUNC18-1 binding to syntaxin-1 (Fig. 2C) and dynamin-1 binding to SH3GL2 (Fig. 2D) could readily be observed, serving as positive controls.

Taken together, these results show a that direct binding interaction between the two proteins is not detectable in HEK293 cells and cannot contribute to an explanation of the MUNC18-1 dependence of dynamin-1 expression.

Stxbp1 KO neurons show a selective defect in receptor-mediated endocytosis

Dynamin proteins are required for a range of endocytic processes (Ferguson and De Camilli, 2012). Therefore, we asked whether the reduction in dynamin levels leads to impairments in known dynamin-dependent cellular processes. To assess this, the receptor-mediated uptake of two fluorescently labelled substrates, transferrin (Tfn–488; Fig. 3A) and cholera toxin B subunit (CTB–488; Fig. 3E) was compared. Transferrin uptake, via endocytosis of the transferrin receptor (TfR), requires dynamin (Herskovits et al., 1993; Vallis et al., 1999; Marks et al., 2001; Song et al., 2004), whereas cholera toxin B subunit has been shown to be endocytosed in a dynamin-independent manner as well (Shogomori and Futerman, 2001; Torgersen et al., 2001; Massol et al., 2004). Following ligand incubation, fluorescent particles were visible in somata and neurites of WT and Stxbp1 KO neurons (Fig. 3B for Tfn–488; Fig. 3F for CTB–488). A similar number of Tfn–488 particles were detected in all neurons that met the analysis criteria (Fig. 3C). However, in Stxbp1 KO neurons, these particles were significantly less intense (Fig. 3D). This suggests that fluorescently labelled transferrin binds receptors to an equal extent in WT and Stxbp1 KO neurons, but the lower fluorescence intensity indicates that ligand-bound receptors are not endocytosed to the same extent in Stxbp1 KO neurons. Half of all Stxbp1 KO neurons did not show sufficient Tfn–488 signal for analysis, whereas almost all WT neurons did (Fig. 3I). In contrast, both the number of fluorescently labelled CTB–488 puncta and their intensity (Fig. 3F–H) were not significantly different between WT and Stxbp1 KO neurons. Reduced Tfn–488 uptake could alternatively be explained by reduced expression of TfRs. However, mass spectrometry data indicate no difference in expression of the TfR (P=0.63 at DIV2; dataset of van Berkel et al., 2022).

Fig. 3.

Stxbp1 KO neurons show a selective defect in receptor-mediated endocytosis. (A) Schematic overview of fluorescently labelled transferrin (Tfn–488) uptake. NB−, unsupplemented Neurobasal medium; PFA, paraformaldehyde. (B) Typical examples of DIV3 WT and Stxbp1 KO neurons after incubation with Tfn–488. The neuronal marker MAP2 is shown in merged images on the right. (C) Density of Tfn–488 puncta is not altered in Stxbp1 KO neurons (WT median, 0.157; WT IQR, 0.133–0.207; KO median, 0.136; KO IQR, 0.109–0.200). P=0.51. (D) Tfn–488 normalized particle intensity (a.u., arbitrary units) is significantly reduced in Stxbp1 KO neurons (WT median, 0.977; WT IQR, 0.850–1.159; KO median, 0.753; KO IQR, 0.836–0.583). P=0.002. (E) Schematic overview of CTB–488 uptake experiment. (F) Typical examples of DIV3 WT and Stxbp1 KO neurons following incubation with CTB–488. The neuronal marker MAP2 is shown in merged images on the right. (G) Density of CTB–488 particles is not different between WT and Stxbp1 KO neurons (WT median, 0.292; WT IQR, 0.262–0.335; KO median, 0.296; KO IQR, 0.235–0.364). P=0.64. (H) CTB–488 particle intensity is not altered in Stxbp1 KO neurons (WT median, 1.06; WT IQR, 0.872–1.167; KO median, 0.899; KO IQR, 0.748–1.056). P=0.14. Boxplots in C,D,G,H show median, IQR and Tukey whiskers, with individual datapoints plotted. Statistical comparisons used unpaired two-tailed Student's t-tests (**P<0.01; n.s., not significant). Sample sizes are indicated as n (number of neurons)/N (number of culture batches) in C and G (and also apply to D and H, respectively). Scale bars: 10 µm. (I) Overview of the number of Tfn-positive and Tfn-negative cells observed, split by replicate experiment (for example, 10 Tfn-positive neurons in culture batch 1, 9 in culture batch 2 and 9 in culture batch 3). Only neurons counted as Tfn-positive were included for the analyses shown in C and D.

Fig. 3.

Stxbp1 KO neurons show a selective defect in receptor-mediated endocytosis. (A) Schematic overview of fluorescently labelled transferrin (Tfn–488) uptake. NB−, unsupplemented Neurobasal medium; PFA, paraformaldehyde. (B) Typical examples of DIV3 WT and Stxbp1 KO neurons after incubation with Tfn–488. The neuronal marker MAP2 is shown in merged images on the right. (C) Density of Tfn–488 puncta is not altered in Stxbp1 KO neurons (WT median, 0.157; WT IQR, 0.133–0.207; KO median, 0.136; KO IQR, 0.109–0.200). P=0.51. (D) Tfn–488 normalized particle intensity (a.u., arbitrary units) is significantly reduced in Stxbp1 KO neurons (WT median, 0.977; WT IQR, 0.850–1.159; KO median, 0.753; KO IQR, 0.836–0.583). P=0.002. (E) Schematic overview of CTB–488 uptake experiment. (F) Typical examples of DIV3 WT and Stxbp1 KO neurons following incubation with CTB–488. The neuronal marker MAP2 is shown in merged images on the right. (G) Density of CTB–488 particles is not different between WT and Stxbp1 KO neurons (WT median, 0.292; WT IQR, 0.262–0.335; KO median, 0.296; KO IQR, 0.235–0.364). P=0.64. (H) CTB–488 particle intensity is not altered in Stxbp1 KO neurons (WT median, 1.06; WT IQR, 0.872–1.167; KO median, 0.899; KO IQR, 0.748–1.056). P=0.14. Boxplots in C,D,G,H show median, IQR and Tukey whiskers, with individual datapoints plotted. Statistical comparisons used unpaired two-tailed Student's t-tests (**P<0.01; n.s., not significant). Sample sizes are indicated as n (number of neurons)/N (number of culture batches) in C and G (and also apply to D and H, respectively). Scale bars: 10 µm. (I) Overview of the number of Tfn-positive and Tfn-negative cells observed, split by replicate experiment (for example, 10 Tfn-positive neurons in culture batch 1, 9 in culture batch 2 and 9 in culture batch 3). Only neurons counted as Tfn-positive were included for the analyses shown in C and D.

To confirm the dynamin-dependency of different aspects of Tfn–488 and CTB–488 uptake in neurons, two control experiments were performed. First, dynamin activity was acutely inhibited in WT neurons at DIV3 by application of dynamin-inhibitory Dyngo4a (Fig. S3A). Similar to the uptake by Stxbp1 KO neurons, the intensity of Tfn–488 puncta was reduced following application of Dyngo4a (Fig. S3B) whereas the puncta number was unchanged (Fig. S3C). Secondly, Tfn–488 uptake was assessed in dynamin KO neurons (Fig. S3D–F). A significant reduction in Tfn–488 puncta intensity (Fig. S3E) and number (Fig. S3F) was observed in dynamin-1/2/3 TKO neurons compared to control (dynamin-3 KO) neurons. In contrast, CTB–488 uptake was not altered by Dyngo4a incubation (Fig. S3G–I) and was even enhanced in dynamin-1/2/3 TKO neurons (Fig. S3J–L). Taken together, these results show that Stxbp1 KO neurons exhibit a selective impairment in dynamin-dependent function, as indicated by impairment of Tfn–488 endocytosis but not of CTB–488 endocytosis, which is dynamin independent.

Reduced dynamin levels do not explain neurodegeneration and altered Golgi morphology of Stxbp1 KO neurons

The reduction in dynamin-1 protein levels might contribute to the degeneration of Stxbp1 KO neurons. Expression of WT dynamin-1 in Stxbp1 KO neurons did not improve viability compared to an empty vector expressing mCherry only (Fig. 4A,B). In contrast, as a positive control, expression of MUNC18-3 did rescue viability, as reported previously (Santos et al., 2017). In line with this, depletion of all three dynamin paralogues, but not depletion of dynamin-3 or dynamin-1 alone, leads to degeneration (Fig. 4C; Park et al., 2013; Moro et al., 2021).

Fig. 4.

Reduced dynamin levels do not underlie cardinal Stxbp1 KO features. (A) Typical example images showing mCherry expression in a randomly selected field of view of Stxbp1 KO neurons expressing the indicated proteins at DIV2 (top row) or DIV4 (bottom row). Scale bar: 20 µm. (B) Expression of MUNC18-3 (M18-3) in Stxbp1 KO neurons efficiently rescues degeneration compared to negative control (infection with empty vector). Overexpression of WT dynamin-1 (Dyn1) or GTPase-deficient K44A dynamin-1 does not improve viability compared to negative control. Data are presented as the mean±s.d. of n=3 experiments. (C) Viability of dynamin-1/2/3 TKO neurons declines within days after infection with active Cre, whereas dynamin-3 single KO (SKO) and dynamin-1/3 DKO neurons have uncompromised viability. Data are presented as the mean±s.e.m. of n=3 experiments. #P<0.01, *P<0.05, **P<0.01 (one-way ANOVA with Dunn–Sidak test). Panel reproduced, with permission, from Moro et al. (2021). (D) Typical examples of Golgi morphology (GM130 marker, bottom; MAP2 is shown as a neuronal marker, top) in WT, dynamin-3 KO (DynSKO), dynamin-1/3 DKO (DynDKO) and dynamin-1/2/3 TKO (DynTKO) neurons. Scale bar: 10 µm. (E) Normalized Golgi size is not different between the four genotypes (WT median, 0.31; WT IQR, 0.27–0.37; SKO median, 0.46; SKO IQR, 0.28–0.45; DKO median, 0.29; DKO IQR, 0.23–0.36; TKO median, 0.31; TKO IQR, 0.21–0.40). P=0.082 (n.s., not significant; Kruskal–Wallis test). (F) Golgi shape, as measured by the ‘Roundness’ parameter in Fiji, is not altered between the four genotypes (WT median, 0.49; WT IQR, 0.34–0.65; SKO median, 0.46; SKO IQR, 0.27–0.69; DKO median, 0.39; DKO IQR, 0.22–0.69; TKO median, 0.61, TKO IQR, 0.38–0.67). P=0.37 (n.s., not significant; Kruskal–Wallis test). Sample sizes in E and F are indicated as n (number of neurons)/N (number of culture batches). Boxplots show median, IQR and Tukey whiskers, with individual datapoints plotted.

Fig. 4.

Reduced dynamin levels do not underlie cardinal Stxbp1 KO features. (A) Typical example images showing mCherry expression in a randomly selected field of view of Stxbp1 KO neurons expressing the indicated proteins at DIV2 (top row) or DIV4 (bottom row). Scale bar: 20 µm. (B) Expression of MUNC18-3 (M18-3) in Stxbp1 KO neurons efficiently rescues degeneration compared to negative control (infection with empty vector). Overexpression of WT dynamin-1 (Dyn1) or GTPase-deficient K44A dynamin-1 does not improve viability compared to negative control. Data are presented as the mean±s.d. of n=3 experiments. (C) Viability of dynamin-1/2/3 TKO neurons declines within days after infection with active Cre, whereas dynamin-3 single KO (SKO) and dynamin-1/3 DKO neurons have uncompromised viability. Data are presented as the mean±s.e.m. of n=3 experiments. #P<0.01, *P<0.05, **P<0.01 (one-way ANOVA with Dunn–Sidak test). Panel reproduced, with permission, from Moro et al. (2021). (D) Typical examples of Golgi morphology (GM130 marker, bottom; MAP2 is shown as a neuronal marker, top) in WT, dynamin-3 KO (DynSKO), dynamin-1/3 DKO (DynDKO) and dynamin-1/2/3 TKO (DynTKO) neurons. Scale bar: 10 µm. (E) Normalized Golgi size is not different between the four genotypes (WT median, 0.31; WT IQR, 0.27–0.37; SKO median, 0.46; SKO IQR, 0.28–0.45; DKO median, 0.29; DKO IQR, 0.23–0.36; TKO median, 0.31; TKO IQR, 0.21–0.40). P=0.082 (n.s., not significant; Kruskal–Wallis test). (F) Golgi shape, as measured by the ‘Roundness’ parameter in Fiji, is not altered between the four genotypes (WT median, 0.49; WT IQR, 0.34–0.65; SKO median, 0.46; SKO IQR, 0.27–0.69; DKO median, 0.39; DKO IQR, 0.22–0.69; TKO median, 0.61, TKO IQR, 0.38–0.67). P=0.37 (n.s., not significant; Kruskal–Wallis test). Sample sizes in E and F are indicated as n (number of neurons)/N (number of culture batches). Boxplots show median, IQR and Tukey whiskers, with individual datapoints plotted.

The WT dynamin-1 expression vector used here yielded functional dynamin-1, which has previously been shown to rescue the viability of dynamin-1/2/3 TKO neurons and dense core vesicle release (Moro et al., 2021). Further confirming functionality of this construct, synaptophysin-pHluorin (SypHy) analysis showed that overexpression of dynamin-1 rescued synaptic vesicle endocytosis compared to that in dynamin-1/2/3 TKO neurons (Fig. S4). Compared to dynamin-3 KO neurons (controls), dynamin-1/2/3 TKO neurons showed an increase in surface expression of fluorescent SypHy and a lack of synaptic vesicle endocytosis following stimulation, as expected. Expression of dynamin-1 in the TKO neurons restored these phenotypes towards control levels for most parameters (Fig. S4B–H). Hence, the reduction in dynamin-1 protein levels does not explain neuronal cell death in Stxbp1 KO neurons.

Expression of a dominant-negative dynamin-1 mutant (K44A, which has abolished GTPase activity) has previously been shown to delay degeneration of neurons after BoNT/C-mediated ablation of Syntaxin-1 (Peng et al., 2013). Expression of this construct also did not improve viability of Stxbp1 KO neurons (Fig. 4A,B).

Aside from the compromised viability, one of the defining phenotypes of Stxbp1 KO neurons is an abnormal Golgi morphology (Santos et al., 2017; van Berkel et al., 2021). In non-neuronal cells, dysfunctional dynamin-2 causes Golgi abnormalities (Cao et al., 2000). Therefore, Golgi morphology was assessed in dynamin KO neurons (Fig. 4D). No abnormalities were observed in Golgi size (normalized to soma; Fig. 4E) or Golgi shape (Fig. 4F). In line with this, overexpression of WT dynamin-1 did not rescue the Golgi abnormalities of Stxbp1 KO neurons (Fig. S5). Taken together, our results show that reduced levels of dynamin do not explain the Golgi aberrations or neurodegeneration observed for Stxbp1 KO neurons.

Reduction in dynamin-1 levels is not a general consequence of lack of synaptic transmission

Stxbp1 KO neurons show differential expression of a number of endocytic proteins, and previous analysis of mass spectrometry data has demonstrated that a substantial number of non-endocytic proteins are also differentially regulated (van Berkel et al., 2022). Previous transcriptomic analysis has indicated that a broader subset of synapse-specific genes is downregulated in Stxbp1 KO neurons (Bouwman et al., 2006). Initially, this downregulation was suggested to be the result of the lack of synaptic activity (Bouwman et al., 2006). However, in other genotypes that have no evoked synaptic activity (SNAP25 KO), or no synaptic transmission at all (MUNC13-1/2 DKO), no similar reduction in synaptic protein levels is observed (Washbourne et al., 2002; Varoqueaux et al., 2002). In line with the lack of a general effect on synaptic protein levels, we showed that dynamin levels are not reduced in either SNAP25 KO or MUNC13-1/2 DKO neurons. In contrast, a previous study of syntaxin-1A and -1B DKO mouse brains has found that expression of several synaptic proteins is reduced (Vardar et al., 2016), but dynamin levels were not assessed in that study. Taken together, the reduced levels of dynamin-1 RNA and protein in Stxbp1 KO neurons are unlikely to be a mere consequence of the KO strategy used, nor a generic result of blocked exocytosis, but rather an effect that specifically ensues from the absence of MUNC18-1.

Defects in receptor endocytosis of Stxbp1 KO neurons are not caused by dynamin-1 reduction alone

Control experiments demonstrated that only KO of all three dynamin paralogues, or acute general dynamin inhibition, causes a deficit in Tfn–488 uptake, whereas Stxbp1 KO neurons only showed a reduction in dynamin-1 and -3, but still expressed dynamin-2. The lack of dynamin-3 alone might already affect Tfn–488 uptake, but this seems unlikely, as dynamin-3 represents only a minor fraction of all dynamin proteins and single KO of dynamin-3 does not reveal any phenotype (Raimondi et al., 2011; Moro et al., 2021). Alternatively, considering that dynamin paralogue expression and localization changes over the course of development (Cook et al., 1996; Gray et al., 2003), the discrepancy might reflect developmental changes in dependency of neuronal processes on dynamin paralogues. The dependency of receptor endocytosis-specific dynamin paralogues could only be assessed in mature neurons due to the conditional KO system needed to generate dynamin-1/2/3 TKO neurons. Definitive evidence that the deficit in Tfn–488 uptake of Stxbp1 KO neurons is due to the reduced levels of dynamin would require assessing uptake after overexpression of dynamin-1.

Dynamins are not only required at the cell surface, but also play a role in intracellular trafficking routes (Praefcke and McMahon, 2004; Ferguson and De Camilli, 2012) and Golgi function (Cao et al., 2000). Since dynamin KO neurons did not recapitulate the reported Golgi abnormalities of Stxbp1 KO neurons (Santos et al., 2017; van Berkel et al., 2021), it is unlikely that the reduction in dynamin-1 directly underlies this Stxbp1 KO phenotype. However, retrograde trafficking of both cholera toxin B subunit and the TrkB receptor (also known as NTRK2) after their initial endocytosis is altered in Stxbp1 KO neurons (van Berkel et al., 2021). Possibly, reduced levels of dynamin interfere with intracellular budding events. However, membrane budding events, both at the plasma membrane and intracellularly, require a wider range of endocytic proteins. For instance, transferrin uptake involves the entire clathrin-mediated endocytosis machinery (Mayle et al., 2012). Thus, it is conceivable that the reduction in levels of other endocytic proteins observed in Stxbp1 KO neurons contributes, alongside the reduction in dynamin-1, to the observed endocytic defects.

Reduced levels of dynamin-1 are not the single cause of Stxbp1 KO neuron degeneration

Overexpression of dynamin-1 or dominant-negative dynamin-1 did not improve viability of Stxbp1 KO neurons, suggesting that the reduced levels of dynamin-1 do not directly correlate to the rapid degeneration in Stxbp1 KO neurons. In line with this, neurons lacking dynamin-1 and dynamin-3, which most closely reflects the situation in Stxbp1 KO neurons, do not show compromised viability (Raimondi et al., 2011; Moro et al., 2021). However, the possibility cannot be excluded that loss of dynamin-1 contributes to some extent to neuronal death, as it is required for a range of cellular processes aside from its canonical role in synaptic vesicle endocytosis (Ferguson et al., 2007; Raimondi et al., 2011).

At the plasma membrane, several receptors require endocytosis in order to activate intracellular signalling pathways, of which the TfR tested in this study is one example (Ferguson and De Camilli, 2012; Cosker and Segal, 2014). For example, neurotrophic signalling through BDNF has been shown to require dynamin-mediated endocytosis to promote neuronal survival and outgrowth (Zheng et al., 2008). Application of trophic factors – insulin or BDNF – delays degeneration of Stxbp1 KO neurons in vitro (Heeroma et al., 2004). Thus, a deficit in receptor endocytosis caused by reduced levels of endocytic proteins possibly contributes to the cell death. However, eventually, increased trophic support fails to prevent degeneration (Heeroma et al., 2004), suggesting that reduced neurotrophic signalling is not the primary cause of degeneration.

Taken together, the reduction in dynamin levels and other endocytic proteins is not the single cause of degeneration of Stxbp1 KO neurons, but the reduced levels might negatively affect intracellular trafficking and signalling pathways, and thus contribute to the compromised viability of these neurons.

MUNC18-1-dependent dynamin-1 and syntaxin-1 expression might be regulated through a shared mechanism

In the absence of MUNC18-1, syntaxin-1 – the canonical binding partner of MUNC18-1 – is the most severely affected protein identified so far. Syntaxin-1 levels are reduced by 70% and the protein is mistargeted in Stxbp1 KO chromaffin cells (Voets et al., 2001; Gulyás-Kovács et al., 2007) and neurons (Toonen et al., 2005; Santos et al., 2017). Moreover, syntaxin-1 mRNA levels are reduced in Stxbp1 KO neurons (Bouwman et al., 2006), as was observed in the present study for dynamin-1 transcript levels.

Although no direct interaction was detected between MUNC18-1 and dynamin-1, the yeast dynamin orthologue Vps1 does bind to the syntaxin orthologue Vam3 (Alpadi et al., 2013), and syntaxin-1 has been found to interact with dynamin-1 in rat synaptosomes and chromaffin cells (Galas et al., 2000; Zhang et al., 2014), unlike SNAP25 (Galas et al., 2000). In line with this, we found that dynamin levels were unaltered in SNAP25 KO brains. The absence of dynamins has been found to impair the recruitment of syntaxin-1 by SNAP25, and the integration syntaxin-1 into SNARE complexes, but does not affect the total syntaxin-1 expression levels in primary neurons (Moro et al., 2021). Hence, this potential connection would place dynamin-1 protein regulation downstream of mistargeting of its interaction partner syntaxin-1.

Alternatively, considering that both dynamin-1 and syntaxin-1 expression are regulated by MUNC18-1 at the mRNA level, both might be a target of the same as-yet-unknown MUNC18-1-dependent regulatory pathway, possibly as part of a larger group of synapse-specific mRNAs found to be reduced in the absence of MUNC18-1 (Bouwman et al., 2006; van Berkel et al., 2022). One explanation would be that MUNC18-1 acts as a transcriptional regulator. Although this hypothesis has hardly been explored, MUNC18-1 has been reported to bind double-stranded DNA, possess putative nuclear import and export signals, and localize to some extent to neuronal nuclei (Sharma et al., 2005). Alternatively, the absence of MUNC18-1 might indirectly alter transcription regulatory pathways. MUNC18-1 is a substrate of ERK (Schmitz et al., 2016) and Cdk5 kinase activities, and has been found to colocalize with Cdk5 in the nucleus (Sharma et al., 2005). Cdk5 is known to extensively influence transcription in neurons in an activity-dependent manner (Kim and Ryan, 2010; Liang et al., 2015). Moreover, some studies link MUNC18-1 and its interacting proteins Mint1 and Mint2 (also known as APBA1 and APBA2, respectively) to modulation of APP processing (Ho et al., 2002; Hill et al., 2003; Sakurai et al., 2008), which by affecting APP cleavage, can regulate transcription (Cao and Südhof, 1999; Biederer et al., 2002). However, we did not observe MUNC18-1 signal in the nucleus, suggesting that a direct DNA-binding role for MUNC18-1 is not the most likely scenario. Aside from possibly regulating transcription, MUNC18-1 may be involved in regulating the stability and/or transport of mRNA. One study has shown that the Drosophila MUNC18-1 orthologue Rop is necessary for localization of specific mRNAs during oogenesis (Ruden et al., 2000). Thus, there are some potential links between MUNC18-1 and transcriptional regulation, either by direct modulation of DNA or mRNA, or through interaction with other proteins known to have a transcription regulatory role. In summary, the absence of MUNC18-1 leads to reduced expression of a set of key presynaptic proteins at transcript and protein level, which suggests that MUNC18-1 triggers a regulatory mechanism with the capacity to regulate expression at the transcriptional level.

Animals

All animals were housed and bred according to institutional and Dutch Animal Ethical Committee regulations (DEC-FGA 11-03). The generation of mouse lines has been described previously: Stxbp1 KO (Verhage et al., 2000), Stxbp1lox/lox (Heeroma et al., 2004), SNAP25 KO (Washbourne et al., 2002), MUNC13-1/2 DKO (Varoqueaux et al., 2002), and dynamin-1/3 KO with dynamin-2lox/lox (Ferguson et al., 2007; Raimondi et al., 2011; Park et al., 2013). On embryonic day 18 (E18), embryos were obtained by caesarean section of pregnant females from timed breeding. For the mass spectrometry and RNA sequencing experiment datasets (van Berkel et al., 2022), WT littermates served as control for Stxbp1 KO. For all other experiments, WT mice were used from separate (homozygous WT) litters from the same background as Stxbp1 KO mice.

Neuronal cultures

Cortices of E18 mouse embryos were dissected in Hank's balanced salt solution (Sigma) and digested in 0.25% trypsin (Life Technologies) for 20 min at 37°C. Neurons were then triturated, resuspended and plated in Neurobasal medium supplemented with 2% B-27 (Invitrogen), 1.8% HEPES, 0.25% GlutaMAX (Invitrogen), and 0.1% penicillin-streptomycin (Invitrogen). Neurons were plated in 12-well plates at a density of 25,000–80,000 cells per well on glass coverslips on a confluent layer of rat glia prepared from newborn P0/P1 Wistar rat pups (Santos et al., 2017). For western blot and mass spectrometry analyses, neurons were plated without glia layer directly on poly-L-ornithine- and laminin-coated 6-well plates at a density of 3×105–1×106 cells per well. For cultures that were kept until 14 days post-plating for analysis (dynamin KO neuron experiments), AraC (2 μM, Sigma) was added 24 h after lentiviral infection to prevent glial overgrowth, and Neurobasal medium was refreshed (50%) every 7 days.

Constructs and lentiviruses

Lentiviral particles were produced as described previously (Naldini et al., 1996; Kovačević et al., 2018). Dynamin-1/3 DKO neurons were infected on DIV7–9 with Cre-expressing lentivirus to remove the floxed dynamin-2 alleles and generate dynamin1/2/3 TKO neurons by DIV14–16. As a control, expression of inactive ΔCre was used on dynamin-3 single KO and dynamin-1/3 DKO neurons.

For the experiments assessing the effect of dynamin overexpression on viability, Stxbp1 KO neurons were infected at DIV0 with a construct to drive EGFP expression to visualize neuronal morphology and either WT dynamin-1, K44A dynamin-1 or MUNC18-3, all with an IRES2-mCherry-NLS signal to control for infection efficiency. As negative control, an empty vector expressing only mCherry-NLS was used. The constructs expressing WT or K44A dynamin-1 were derived from Addgene vectors 34681 (K44A-Dynamin1-pEGFP) and 34680 (WT-Dynamin1-pEGFP).

Western blots

Cultured HEK293 cells or neurons were collected in ice-cold PBS with protease inhibitors (Sigma) and spun down at 12,000 g at 4°C. PBS was removed and 1× SDS loading buffer was added (10% glycerol, 2% SDS, 0.26 M β-mercaptoethanol, 60 mM Tris) followed by heating the sample for several minutes at 90°C. Brain tissue was first dissociated by shear force and, following the heating step, further homogenized using an insulin syringe. Samples were run on a 10% SDS–PAGE gel [33.2% acrylamide/bis-acrylamide (Bio-Connect; 30% w/v); 24.9% 1.5 M Tris-HCl (pH 8.8); 0.13% SDS (VWR Chemicals); 0.07% APS (AppliChem) and 0.007% TEMED (VWR Chemicals)] and transferred onto PVDF membranes. Following blocking in PBS supplemented with 0.1% Tween-20 (PBS-Tween), 2% skim milk powder (Merck) and 0.5% FBS (Life Technologies #10270106), primary antibodies were incubated PBS-Tween overnight at 4°C. After three washes in PBS-Tween, a secondary antibody conjugated to alkaline phosphatase (1:1000; alkaline phosphatase AffiniPure goat anti-mouse or goat anti-rabbit IgG H+L; Jackson ImmunoResearch) for 1 h at room temperature. The secondary antibody was visualized using AttoPhos AP fluorescent substrate (Promega), and membranes were scanned with a Fuji Image FLA-5000 Reader with ImageReader FLA5000 software version 2.0. Alternatively, for quantification of protein levels, samples were transferred onto nitrocellulose membranes and blocked in PBS-Tween with 2% bovine serum albumin (BSA; Fisher Emergo). IRDye® 680LT-anti mouse IgG or IRDye® 680CW-anti rabbit IgG secondary antibodies were used (1:5000; LI-COR Biosciences). The membranes were scanned on an Odyssey CLx Imaging system and quantified with Image Studio Lite software (LI-COR Biotechnology). Analysis of western blots was performed in Image Studio Lite software (v. 5.2.5), using background subtraction option ‘Median’. The signal of the protein of interest was normalized to the housekeeping protein in the same lane, and then normalized to the control condition (WT for Fig. 1E, dynamin-3 KO for Fig. 2B). Primary antibodies used: anti-dynamin-1 (Fisher Emergo, PA1-660; 1:500), anti-dynamin-2 (Abcam, AB3457; 1:1000), anti-dynamin-3 (Synaptic Systems, 115-302; 1:1000), monoclonal anti-γ-tubulin (clone GTU-88; Sigma-Aldrich, T5326; 1:1000), polyclonal anti-MUNC18-1 (non-commercial polyclonal 2701, Santos et al., 2017; 1:1000) anti-pan-dynamins (non-commercial; available upon request, 1:1000), anti-actin (Chemicon, MAB1501, 1:10000), anti-SNAP25 (Covance, SMI-81R; 1:5000), anti-AP180 (Synaptic Systems, 155-022; 1:1000), anti-alpha-tubullin (Synaptic Systems, 302-211; 1:1000) and anti-SH3GL2 (Synaptic Systems, 159-002; 1:1000). Images of full western blots are shown in Fig. S6.

HEK293 cell culture and co-immunoprecipitation experiment

HEK293 cells were cultured in DMEM/F12 medium with L-glutamine, 10% FCS, 1% NEAA and 1% penicillin-streptomycin (all Gibco) and transfected 1 day after plating using Ca2+ phosphate transfection, as previously described (Brouwer et al., 2019). For the verification of dynamin antibodies (Fig. S1), the following constructs were used: pSyn-Dynamin1-IRES2-NLSmCherry (human sequence, derived from Addgene vector 34680), pCMV-Dynamin1-IRES2-NLSmCherry, pCMV-Dynamin2-IRES2-NLSmCherry or pCMV-Dynamin3-IRES2-NLSmCherry (mouse sequences, extracted from a library; see Moro et al., 2021). Expression of mouse dynamin-1 protein was unsuccessful, and subsequent sequencing of the construct indicated the sequence was incomplete.

For the co-immunoprecipitation experiments, HEK293 cells were co-transfected with constructs expressing mouse SH3GL2-Dest-3×Flag, MUNC18-1-IRES2-EGFP, syntaxin1-IRES2-EGFP and dynamin-1-IRES2-NLS-mCherry (derived from Addgene 34680). Co-IP was performed as described previously (Brouwer et al., 2019). Pulldown was performed using the following antibodies: polyclonal anti-Munc18-1 (1 μl; rabbit; non-commercial polyclonal 2701), anti-dynamin-1 (1 μl; rabbit; Fisher Emergo, PA1-660), monoclonal anti-syntaxin-1 (1 μl; mouse; Sigma, S0664). Western blots were performed as described above. Resulting blot membranes were consecutively stained using monoclonal anti-MUNC18-1 (1:1000; mouse; Transduction Labs, #610336) together with monoclonal anti-syntaxin-1 (1:1000; mouse; Sigma S0664) or anti-Flag (1:1000; mouse; Sigma F1804), then with anti-dynamin-1 (1:1000; rabbit; Fisher Emergo, PA1-660). Images of full western blots are shown in Fig. S6.

Immunocytochemistry and confocal microscopy

Neurons were fixed with 3.7% paraformaldehyde for 20 min at room temperature, then washed in PBS three times and permeabilized for 5 min with 0.5% Triton X-100 (Fisher Emergo). Non-specific binding sites were blocked by incubation with 2% normal goat serum and 0.1% Triton X-100 in PBS for 30 min at room temperature. Primary antibodies were incubated for 2 h at room temperature. The following primary antibodies were used: polyclonal anti-MUNC18-1 (non-commercial polyclonal 2701; 1:1000), polyclonal anti-MAP2 (Abcam, ab5392; 1:250–1:500), monoclonal anti-GM130 (BD Transduction Laboratories, 610822; 1:1000), anti-dynamin-1 (Fisher Emergo, PA1-660; 1:250), monoclonal anti-mCherry (Signalway Antibody, T515; 1:1000), polyclonal anti-GFP (Synaptic Systems, 132005; 1:500). After three washes in PBS, secondary antibodies were incubated for 1 h at room temperature. Secondary antibodies were Alexa Fluor-conjugated antibodies (1:1000; Invitrogen). Coverslips were mounted in Mowiol and imaged on either a Zeiss 510 Meta confocal microscope with 40× objective (1.3 NA) or alternatively on a Nikon Ti-Eclipse confocal microscope with 40× oil-immersion objective (1.3 NA) controlled by NIS-Elements 4.30 software. Z-stack images were acquired of individual neurons after selection based on the morphological channel. For quantification of neuronal viability (Fig. 4C), imaging software was used to image 25 random fields of view per coverslip to allow for unbiased cell counting.

Endocytosis assays

To assess the uptake of cholera toxin B subunit (CTB), CTB conjugated to Alexa Fluor 488 (CTB–488; Thermo Fisher Scientific; 100 ng/ml) was added to the medium of the neurons and incubated for 15 min at 37°C, 5% CO2. Following two quick washes with pre-warmed supplemented Neurobasal medium, cells were fixed with 3.7% paraformaldehyde solution followed by immunostaining for MAP2 (see ‘Immunocytochemistry and confocal microscopy’ section). To assess the uptake of transferrin (Tfn), Tfn conjugated to Alexa Fluor 488 (Tfn–488; Life Technologies) was used. Standard neuronal culture medium was supplemented with B-27 (Invitrogen), which contains holo-transferrin. To promote neuronal uptake of fluorescently conjugated Tfn ligand, neuronal medium was replaced with unsupplemented Neurobasal medium (Neurobasal−) during 2 h prior to incubation of Tfn–488. 15 μg/ml Tfn–488 was incubated for 15 min at 37°C, 5% CO2. Following two quick washes with pre-warmed Neurobasal− medium, cells were fixed with 3.7% paraformaldehyde solution followed by immunostaining for MAP2. To assess the effect of acute dynamin inhibition on CTB and Tfn uptake, the procedures described above were preceded by a 5-min incubation with either Dyngo4a (30 μM; Abcam) or DMSO (control; Sigma-Aldrich) in the medium prior to incubation with CTB–488 or Tfn–488.

Image analysis

For quantification of fluorescence intensity and morphological features, z-stacks were collapsed to maximum intensity projections and background subtraction was performed using Fiji software (https://fiji.sc/). SynD software (Schmitz et al., 2011) was used to generate a morphology mask subsequently used to measure fluorescence intensity using Fiji software. For analysis of CTB and Tfn uptake, background subtraction was performed using Gaussian and top-hat (closing at half the particle size) filtering steps. Particles were then detected using Renyi Entropy thresholding (Tfn–488 analysis) or Moments thresholding (for CTB–488 analysis). Segmented particles were then selected on the basis of size, from the minimum particle size up to three times the square of the particle size (in pixels). Fig. 4C is reproduced, with permission, from Moro et al. (2021), where protein levels and survival of dynamin single, double and triple KO neurons were reported previously (Moro et al., 2021). To assess survival of Stxbp1 KO neurons (Fig. 4A,B), neurons of which the soma was located within the random field of view were counted, and the presence of mCherry signal (indicative of successful lentiviral infection) was assessed. Infection efficiency was found to be 100% for all lentiviral particles used. Survival is expressed as a ratio between the number of neurons after 4 days in vitro (DIV4; after onset of degeneration of non-rescued Stxbp1 KO neurons) over the number of neurons at 2 days in vitro (DIV2; prior to onset of neurodegeneration).

RNA sequencing and qPCR

Experimental and analysis procedures for the RNA sequencing data are described in detail by van Berkel et al. (2022). For qPCR experiments, RNA was isolated as for RNA sequencing, and cDNA synthesis was performed using the SensiFast cDNA Synthesis Kit (GC Biotech) according to the manufacturer's instructions. Then, cDNA was quantified using SensiFAST SYBR No-ROX (Bioline) in a LightCycler 480 (Roche Life Sciences) using the program: 5 min incubation at 95°C (4.8°C/s ramp rate), followed by 50 cycles of 10 s at 95°C (4.8°C/s), 20 s at 60°C (2.4°C/s) and 1 s at 72°C (4.8°C/s). For each sample, 1 μl of cDNA sample was used per well with 10 μM primers (forward and reverse): dynamin-1, 5′-CCCTTTCGAGCTGGTCAAGA-3′, 5′-GGCCATGTCTGGGGTAAACA-3′; dynamin-2, 5′-AGGAACTGCATCCTTGGGAC-3′, 5′-TCCTGGACCCCAGAAAGAAC-3′; dynamin-3, 5′-CATACCGACCTTCCGCACAT-3′, 5′-CGGACTGAATGGTGGCGTTA-3′. Samples were quantified in duplicate, average values were used and normalized to geometric mean of two housekeeping genes (EEF, EEF1A1; 18S, Rn18s).

Analysis of mass spectrometry data

Sample collection and analysis of the mass spectrometry data was published previously (van Berkel et al., 2022). Enrichment analysis was performed using the SynGO tool (Koopmans et al., 2019), using all proteins that were significantly downregulated at DIV2, with all proteins detected in the full dataset as background. Enrichment was assessed for the Gene Ontology (GO) term ‘synaptic vesicle cycle’ (GO: 0099504) and its child terms ‘synaptic vesicle endocytosis’ (GO: 0048488), ‘synaptic vesicle endosomal processing’ (GO: 0099532) and ‘regulation of synaptic vesicle cycle’ (GO: 0098693).

To assess the regulation of endocytic proteins, a list of the core endocytic machinery was compiled by integrating the lists of core endocytic machinery components provided in review articles from the leading experts in the field (McMahon and Boucrot, 2011; Saheki and De Camilli, 2012). If not already listed, gene paralogues were also included in this list, since paralogue expression may be affected if the canonical paralogue is altered, and this may be of interest. The heatmap (Fig. 1B) shows the log2 fold changes for all proteins that were on this list and detected in the proteomics experiment. Asterisks within the heatmap indicate a false discovery rate (FDR)-adjusted P-value <0.05. Where P-values are reported for single protein comparisons, FDR-adjusted P-values are used.

Live-cell imaging

Live-cell imaging was performed following previously described procedures (Moro et al., 2020). Experiments were performed at room temperature, while superfusing Tyrode's solution. Images were acquired on an Axiovert II microscope (Zeiss, Oberkochen, Germany) with a 40× oil objective (1.3 NA). Time-lapse recordings were acquired using Metamorph 6 and an EM-CCD camera, with acquisition frequency of 2 Hz. The imaging protocol included 10 s of baseline recording, 5 s of acid wash in Tyrode's solution (2 mM CaCl2, 2.5 mM KCl, 119 mM NaCl, 2 mM MgCl2, 30 mM glucose, and 25 mM MES at pH 5.5), followed by a recovery phase of 15 s. Subsequently, electrical field stimulation was applied by an A-385 stimulus isolator (WPI) controlled by a Master 8 (AMPI), delivering 30 mA pulses of 1 ms duration for 5 s at 40 Hz, followed by 1 min of recovery time. Then, an additional 5 s of Tyrode's (pH 5.5) superfusion step was performed, and a final 5 s perfusion with modified Tyrode's solution containing NH4Cl (2 mM CaCl2, 2.5 mM KCl, 119 mM NaCl, 2 mM MgCl2, 30 mM glucose, 25 mM HEPES and 50 mM NH4Cl, at pH 7.4) delivered by gravity flow through a capillary placed above the cell.

Experimental design and statistical analysis

Statistical analysis and graphing was performed using GraphPad Prism versions 8 and 9. Parametric tests were used when assumptions of homoscedasticity and normality were met, otherwise, non-parametric tests were used. All statistical tests were two-tailed, and an error probability level of P<0.05 was used. If a statistical comparison was performed, this is indicated on the graph and details of the test are included in the legend. Data are represented as single datapoints and/or boxplots with Tukey whiskers, showing median and interquartile range (IQR). Summary statistics, P-values and details of the statistical tests used are reported in the figure legends.

The authors thank Joke Wortel for breeding mutant mice and Joost Hoetjes for genotyping. We thank Robbert Zalm for help with cloning and production of viral particles, and Frank den Oudsten, Desiree Schut and Lisa Laan for preparation of glia feeder plates.

Author contributions

Conceptualization: H.C.A.L., R.F.T., M.V.; Methodology: H.C.A.L., A.M., I.S.; Formal analysis: H.C.A.L., A.M.; Investigation: H.C.A.L., A.M., I.S.; Data curation: H.C.A.L.; Writing - original draft: H.C.A.L., R.F.T., M.V.; Writing - review & editing: H.C.A.L., A.M., I.S., R.F.T., M.V.; Visualization: H.C.A.L., A.M.; Supervision: R.F.T., M.V.; Project administration: H.C.A.L.; Funding acquisition: M.V.

Funding

This work was supported by a European Research Council (ERC) Advanced grant (322966) of the European Union (to M.V.), COSYN (Comorbidity and Synapse Biology in Clinically Overlapping Psychiatric Disorders; Horizon 2020 Framework Programme of the European Union under RIA grant agreement 667301 to M.V.) and Lundbeckfonden (R277-2018-802 to M.V.). Open access funding provided by Vrije Universiteit Amsterdam. Deposited in PMC for immediate release.

Alpadi
,
K.
,
Kulkarni
,
A.
,
Namjoshi
,
S.
,
Srinivasan
,
S.
,
Sippel
,
K. H.
,
Ayscough
,
K.
,
Zieger
,
M.
,
Schmidt
,
A.
,
Mayer
,
A.
,
Evangelista
,
M.
et al. 
(
2013
).
Dynamin-SNARE interactions control trans-SNARE formation in intracellular membrane fusion
.
Nat. Commun.
4
,
1704
-
1708
.
Arriagada-Diaz
,
J.
,
Prado-Vega
,
L.
,
Cárdenas
,
A. M.
,
Ardiles
,
A. O.
and
Gonzalez-Jamett
,
A. M.
(
2022
).
Dynamin superfamily at pre- and postsynapses : master regulators of synaptic transmission and plasticity in health and disease
.
Neuroscientist
28
,
41
-
58
.
Biederer
,
T.
,
Cao
,
X.
,
Südhof
,
T. C.
and
Liu
,
X.
(
2002
).
Regulation of APP-dependent transcription complexes by Mint/X11s: differential functions of Mint isoforms
.
J. Neurosci.
22
,
7340
-
7351
.
Bouwman
,
J.
,
Maia
,
A. S.
,
Camoletto
,
P. G.
,
Posthuma
,
G.
,
Roubos
,
E. W.
,
Oorschot
,
V. M. J.
,
Klumperman
,
J.
and
Verhage
,
M.
(
2004
).
Quantification of synapse formation and maintenance in vivo in the absence of synaptic release
.
Neuroscience
126
,
115
-
126
.
Bouwman
,
J.
,
Spijker
,
S.
,
Schut
,
D.
,
WäChtler
,
B.
,
Ylstra
,
B.
,
Smit
,
A. B.
and
Verhage
,
M.
(
2006
).
Reduced expression of neuropeptide genes in a genome-wide screen of a secretion-deficient mouse
.
J. Neurochem.
99
,
84
-
96
.
Brouwer
,
M.
,
Farzana
,
F.
,
Koopmans
,
F.
,
Chen
,
N.
,
Brunner
,
J. W.
,
Oldani
,
S.
,
Li
,
K. W.
,
Van Weering
,
J. R. T.
,
Smit
,
A. B.
,
Toonen
,
R. F.
et al. 
(
2019
).
SALM1 controls synapse development by promoting F–actin/PIP2–dependent Neurexin clustering
.
EMBO J.
38
,
e101289
.
Cao
,
X.
and
Südhof
,
T. C.
(
1999
).
A transcriptively active complex of APP with Fe65 and histone acetyltransferase Tip60
.
Science
293
,
115
-
120
.
Cao
,
H.
,
Thompson
,
H. M.
,
Krueger
,
E. W.
and
Mcniven
,
M. A.
(
2000
).
Disruption of Golgi structure and function in mammalian cells expressing a mutant dynamin
.
J. Cell Sci.
113
,
1993
-
2002
.
Cook
,
T.
,
Mesa
,
K.
and
Urrutia
,
R.
(
1996
).
Three dynamin-encoding genes are differentially expressed in developing rat brain
.
J. Neurochem.
67
,
927
-
931
.
Cosker
,
K. E.
and
Segal
,
R. A.
(
2014
).
Neuronal signaling through endocytosis
.
Cold Spring Harbor Perspect. Biol.
6
,
a020669.
Deák
,
F.
,
Schoch
,
S.
,
Liu
,
X.
,
Südhof
,
T. C.
and
Kavalali
,
E. T.
(
2004
).
Synaptobrevin is essential for fast synaptic-vesicle endocytosis
.
Nat. Cell Biol.
6
,
1102
-
1108
.
Ferguson
,
S. M.
and
De Camilli
,
P.
(
2012
).
Dynamin, a membrane-remodelling GTPase
.
Nat. Rev. Mol. Cell Biol.
13
,
75
-
88
.
Ferguson
,
S. M.
,
Brasnjo
,
G.
,
Hayashi
,
M.
,
Wölfel
,
M.
,
Collesi
,
C.
,
Giovedi
,
S.
,
Raimondi
,
A.
,
Gong
,
L.-W.
,
Ariel
,
P.
,
Paradise
,
S.
et al. 
(
2007
).
A selective activity-dependent requirement for dynamin 1 in synaptic vesicle endocytosis
.
Science
316
,
570
-
574
.
Galas
,
M. C.
,
Chasserot-Golaz
,
S.
,
Dirrig-Grosch
,
S.
and
Bader
,
M.-F.
(
2000
).
Presence of dynamin-syntaxin complexes associated with secretory granules in adrenal chromaffin cells
.
J. Neurochem.
75
,
1511
-
1519
.
Gray
,
N. W.
,
Fourgeaud
,
L.
,
Huang
,
B.
,
Chen
,
J.
,
Cao
,
H.
,
Oswald
,
B. J.
,
Hémar
,
A.
and
Mcniven
,
M. A.
(
2003
).
Dynamin 3 is a component of the postsynapse, where it interacts with mGluR5 and homer
.
Curr. Biol.
13
,
510
-
515
.
Gulyás-Kovács
,
A.
,
De Wit
,
H.
,
Milosevic
,
I.
,
Kochubey
,
O.
,
Toonen
,
R.
,
Klingauf
,
J.
,
Verhage
,
M.
and
Sorensen
,
J. B.
(
2007
).
Munc18-1: sequential interactions with the fusion machinery stimulate vesicle docking and priming
.
J. Neurosci.
27
,
8676
-
8686
.
Haucke
,
V.
,
Neher
,
E.
and
Sigrist
,
S. J.
(
2011
).
Protein scaffolds in the coupling of synaptic exocytosis and endocytosis
.
Nat. Rev. Neurosci.
12
,
127
-
138
.
Heeroma
,
J. H.
,
Roelandse
,
M.
,
Wierda
,
K.
,
Van Aerde
,
K. I.
,
Toonen
,
R. F. G.
,
Hensbroek
,
R. A.
,
Brussaard
,
A.
,
Matus
,
A.
and
Verhage
,
M.
(
2004
).
Trophic support delays but not prevent cell-intrinsic degeneration of neurons deficient for munc18-1
.
Eur. J. Neurosci.
20
,
623
-
634
.
Herskovits
,
J. S.
,
Burgess
,
C. C.
,
Obar
,
R. A.
and
Vallee
,
R. B.
(
1993
).
Effects of mutant rat dynamin on endocytosis
.
J. Cell Biol.
122
,
565
-
578
.
Hill
,
K.
,
Li
,
Y.
,
Bennett
,
M.
,
Mckay
,
M.
,
Zhu
,
X.
,
Shern
,
J.
,
Torre
,
E.
,
Lah
,
J. J.
,
Levey
,
A. I.
and
Kahn
,
R. A.
(
2003
).
Munc18 interacting proteins: ADP-ribosylation factor-dependent coat proteins that regulate the traffic of β-Alzheimer's precursor protein
.
J. Biol. Chem.
278
,
36032
-
36040
.
Ho
,
C. S.
,
Marinescu
,
V.
,
Steinhilb
,
M. L.
,
Gaut
,
J. R.
,
Turner
,
R. S.
and
Stuenkel
,
E. L.
(
2002
).
Synergistic effects of Munc18a and X11 proteins on amyloid precursor protein metabolism
.
J. Biol. Chem.
277
,
27021
-
27028
.
Kim
,
S. H.
and
Ryan
,
T. A.
(
2010
).
CDK5 serves as a major control point in neurotransmitter release
.
Neuron
67
,
797
-
809
.
Koenig
,
J. H.
,
Saito
,
K.
and
Ikeda
,
K.
(
1983
).
Reversible control of synaptic transmission in a single gene mutant of Drosophila melanogaster
.
J. Cell Biol.
96
,
1517
-
1522
.
Koopmans
,
F.
,
Van Nierop
,
P.
,
Andres-Alonso
,
M.
,
Byrnes
,
A.
,
Cijsouw
,
T.
,
Coba
,
M. P.
,
Cornelisse
,
L. N.
,
Farrell
,
R. J.
,
Goldschmidt
,
H. L.
,
Howrigan
,
D. P.
et al. 
(
2019
).
SynGO: an evidence-based, expert-curated knowledge base for the synapse
.
Neuron
103
,
217
-
234.e4
.
Kovačević
,
J.
,
Maroteaux
,
G.
,
Schut
,
D.
,
Loos
,
M.
,
Dubey
,
M.
,
Pitsch
,
J.
,
Remmelink
,
E.
,
Koopmans
,
B.
,
Crowley
,
J.
,
Cornelisse
,
L. N.
et al. 
(
2018
).
Protein instability, haploinsufficiency, and cortical hyper-excitability underlie STXBP1-encephalopathy
.
Brain
141
,
1350
-
1374
.
Liang
,
Z.
,
Ye
,
T.
,
Zhou
,
X.
,
Lai
,
K.-O.
,
Fu
,
A. K. Y.
and
Ip
,
N. Y.
(
2015
).
Cdk5 regulates activity-dependent gene expression and dendrite development
.
J. Neurosci.
35
,
15127
-
15134
.
Lou
,
X.
(
2018
).
Sensing exocytosis and triggering endocytosis at synapses: synaptic vesicle exocytosis–endocytosis coupling
.
Front. Cell. Neurosci.
12
,
66
.
Maritzen
,
T.
and
Haucke
,
V.
(
2018
).
Coupling of exocytosis and endocytosis at the presynaptic active zone
.
Neurosci. Res.
127
,
45
-
52
.
Marks
,
B.
,
Stowell
,
M. H. B.
,
Vallis
,
Y.
,
Mills
,
I. G.
,
Gibson
,
A.
,
Hopkins
,
C. R.
and
Mcmahon
,
H. T.
(
2001
).
GTPase activity of dynamin and resulting conformation change are essential for endocytosis
.
Nature
410
,
231
-
235
.
Massol
,
R. H.
,
Sardon
,
T.
,
Zimmerman
,
T.
,
Wittmann
,
T.
,
Pepperkok
,
R.
,
Karsenti
,
E.
and
Vernos
,
I.
(
2004
).
Characterization of the TPX2 domains involved in microtubule nucleation and spindle assembly in xenopus nucleation around chromatin and functions in a network of other molecules, some of which also are regulated by
.
Mol. Biol. Cell
15
,
5318
-
5328
.
Mayle
,
K. M.
,
Le
,
A. M.
and
Kamei
,
D. T.
(
2012
).
The intracellular trafficking pathway of transferrin
.
Biochim. Biophys. Acta Gen. Sub.
1820
,
264
-
281
.
Mcmahon
,
H. T.
and
Boucrot
,
E.
(
2011
).
Molecular mechanism and physiological functions of clathrin-mediated endocytosis
.
Nat. Rev. Mol. Cell Biol.
12
,
517
-
533
.
Moro
,
A.
,
Van Woerden
,
G. M.
,
Toonen
,
R. F.
and
Verhage
,
M.
(
2020
).
Camkii controls neuromodulation via neuropeptide gene expression and axonal targeting of neuropeptide vesicles
.
PLoS Biol.
18
,
e3000826
.
Moro
,
A.
,
van Nifterick
,
A.
,
Toonen
,
R. F.
and
Verhage
,
M.
(
2021
).
Dynamin controls neuropeptide secretion by organizing dense-core vesicle fusion sites
.
Sci. Adv.
7
,
eabf0659
. .
Naldini
,
L.
,
Gallay
,
P.
,
Ory
,
D.
,
Mulligan
,
R.
,
Gage
,
F. H.
,
Verma
,
I. M.
and
Trono
,
D.
(
1996
).
In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector
.
Science
272
,
263
-
267
.
Nicholson-Tomishima
,
K.
and
Ryan
,
T. A.
(
2004
).
Kinetic efficiency of endocytosis at mammalian CNS synapses requires synaptotagmin I
.
Proc. Natl. Acad. Sci. USA
101
,
16648
-
16652
.
Park
,
R. J.
,
Shen
,
H.
,
Liu
,
L.
,
Liu
,
X.
,
Ferguson
,
S. M.
and
De Camilli
,
P.
(
2013
).
Dynamin triple knockout cells reveal off target effects of commonly used dynamin inhibitors
.
J. Cell Sci.
126
,
5305
-
5312
.
Peng
,
L.
,
Liu
,
H.
,
Ruan
,
H.
,
Tepp
,
W. H.
,
Stoothoff
,
W. H.
,
Brown
,
R. H.
,
Johnson
,
E. A.
,
Yao
,
W. D.
,
Zhang
,
S. C.
and
Dong
,
M.
(
2013
).
Cytotoxicity of botulinum neurotoxins reveals a direct role of syntaxin 1 and SNAP-25 in neuronal survival
.
Nat. Commun.
77
,
1472
.
Poskanzer
,
K. E.
,
Marek
,
K. W.
,
Sweeney
,
S. T.
and
Davis
,
G. W.
(
2003
).
Synaptotagmin I is necessary for compensatory synaptic vesicle endocytosis in vivo
.
Nature
426
,
559
-
563
.
Praefcke
,
G. J. K.
and
Mcmahon
,
H. T.
(
2004
).
The dynamin superfamily: Universal membrane tubulation and fission molecules?
Nat. Rev. Mol. Cell Biol.
5
,
133
-
147
.
Raimondi
,
A.
,
Ferguson
,
S. M.
,
Lou
,
X.
,
Armbruster
,
M.
,
Paradise
,
S.
,
Giovedi
,
S.
,
Messa
,
M.
,
Kono
,
N.
,
Takasaki
,
J.
,
Cappello
,
V.
et al. 
(
2011
).
Overlapping role of dynamin isoforms in synaptic vesicle endocytosis
.
Neuron
70
,
1100
-
1114
.
Ruden
,
D. M.
,
Sollars
,
V.
,
Wang
,
X.
,
Mori
,
D.
,
Alterman
,
M.
and
Lu
,
X.
(
2000
).
Membrane fusion proteins are required for oskar mRNA localization in the Drosophila egg chamber
.
Dev. Biol.
218
,
314
-
325
.
Saheki
,
Y.
and
De Camilli
,
P.
(
2012
).
Synaptic vesicle endocytosis
.
Cold Spring Harbor Perspect. Biol.
4
,
a005645
.
Sakurai
,
T.
,
Kaneko
,
K.
,
Okuno
,
M.
,
Wada
,
K.
,
Kashiyama
,
T.
,
Shimizu
,
H.
,
Akagi
,
T.
,
Hashikawa
,
T.
and
Nukina
,
N.
(
2008
).
Membrane microdomain switching: a regulatory mechanism of amyloid precursor protein processing
.
J. Cell Biol.
183
,
339
-
352
.
Santos
,
T.
,
Wierda
,
K.
,
Broeke
,
J. H.
,
Toonen
,
R. F.
and
Verhage
,
M.
(
2017
).
Early Golgi abnormalities and neurodegeneration upon loss of presynaptic proteins
.
J. Neurosci.
37
,
4525
-
4539
.
Schmitz
,
S. K.
,
Hjorth
,
J. J. J.
,
Joemai
,
R. M. S.
,
Wijntjes
,
R.
,
Eijgenraam
,
S.
,
De Bruijn
,
P.
,
Georgiou
,
C.
,
De Jong
,
A. P. H.
,
Van Ooyen
,
A.
,
Verhage
,
M.
et al. 
(
2011
).
Automated analysis of neuronal morphology, synapse number and synaptic recruitment
.
J. Neurosci. Methods
195
,
185
-
193
.
Schmitz
,
S. K.
,
King
,
C.
,
Kortleven
,
C.
,
Huson
,
V.
,
Kroon
,
T.
,
Kevenaar
,
J. T.
,
Schut
,
D.
,
Saarloos
,
I.
,
Hoetjes
,
J. P.
,
Wit
,
H.
et al. 
(
2016
).
Presynaptic inhibition upon CB 1 or mGlu 2 / 3 receptor activation requires ERK / MAPK phosphorylation of Munc 18 - 1
.
EMBO J.
35
,
1236
-
1250
.
Sharma
,
V. M.
,
Shareef
,
M.
,
Bhaskar
,
K.
,
Kalidas
,
S.
,
Shetty
,
P.
,
Christopher
,
R.
,
Ramamohan
,
Y.
,
Pant
,
H.
and
Shetty
,
T.
(
2005
).
Nuclear localization of Munc18-1 (p67) in the adult rat brain and PC12 cells
.
Neurochem. Int.
47
,
225
-
234
.
Shogomori
,
H.
and
Futerman
,
A. H.
(
2001
).
Cholera toxin is found in detergent-insoluble rafts/domains at the cell surface of hippocampal neurons but is internalized via a raft-independent mechanism
.
J. Biol. Chem.
276
,
9182
-
9188
.
Song
,
B. D.
,
Yarar
,
D.
and
Schmid
,
S. L.
(
2004
).
An assembly-incompetent mutant establishes a requirement for Dynamin self-assembly in clathrin-mediated endocytosis in vivo
.
Mol. Biol. Cell
15
,
2243
-
2252
.
Südhof
,
T. C.
(
2013
).
Neurotransmitter release: the last millisecond in the life of a synaptic vesicle
.
Neuron
80
,
675
-
690
.
Toonen
,
R. F. G.
and
Verhage
,
M.
(
2007
).
Munc18-1 in secretion: lonely Munc joins SNARE team and takes control
.
Trends Neurosci.
30
,
564
-
572
.
Toonen
,
R. F. G.
,
De Vries
,
K. J.
,
Zalm
,
R.
,
Südhof
,
T. C.
and
Verhage
,
M.
(
2005
).
Munc18-1 stabilizes syntaxin 1, but is not essential for syntaxin 1 targeting and SNARE complex formation
.
J. Neurochem.
93
,
1393
-
1400
.
Torgersen
,
M. L.
,
Skretting
,
G.
,
Van Deurs
,
B.
and
Sandvig
,
K.
(
2001
).
Internalization of cholera toxin by different endocytic mechanisms
.
J. Cell Sci.
114
,
3737
-
3747
.
Vallis
,
Y.
,
Wigge
,
P.
,
Marks
,
B.
,
Evans
,
P. R.
and
Mcmahon
,
H. T.
(
1999
).
Importance of the pleckstrin homology domain of dynamin in clathrin-mediated endocytosis
.
Curr. Biol.
9
,
257
-
263
.
Van Berkel
,
A. A.
,
Santos
,
T. C.
,
Shaweis
,
H.
,
Weering
,
J. R. T.
,
Toonen
,
R. F.
and
Verhage
,
M.
(
2021
).
Loss of MUNC18-1 leads to retrograde transport defects in neurons
.
J. Neurochem.
157
,
450
-
466
.
Van Berkel
,
A. A.
,
Koopmans
,
F.
,
Gonzalez-Lozano
,
M. A.
,
Lammertse
,
H.
,
Feringa
,
F.
,
Bryois
,
J.
,
Sullivan
,
P. F.
,
Smit
,
A. B.
,
Toonen
,
R. F.
and
Verhage
,
M.
(
2022
).
Dysregulation of synaptic and developmental transcriptomic/proteomic profiles upon depletion of MUNC18-1
.
eNeuro
.
ENEURO.0186-22.2022
.
Vardar
,
G.
,
Chang
,
S.
,
Arancillo
,
M.
,
Wu
,
Y.-J.
,
Trimbuch
,
T.
and
Rosenmund
,
C.
(
2016
).
Distinct functions of syntaxin-1 in neuronal maintenance, synaptic vesicle docking, and fusion in mouse neurons
.
J. Neurosci.
36
,
7911
-
7924
.
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
.
Verhage
,
M.
,
Maia
,
A. S.
,
Plomp
,
J. J.
,
Brussaard
,
A. B.
,
Heeroma
,
J. H.
,
Vermeer
,
H.
,
Toonen
,
R. F.
,
Hammer
,
R. E.
,
Van Den
,
T. K.
,
Berg
,
H.
et al. 
(
2000
).
Synaptic assembly of the brain in the absence of neurotransmitter secretion
.
Science
287
,
864
-
869
.
Voets
,
T.
,
Toonen
,
R. F.
,
Brian
,
E. C.
,
De Wit
,
H.
,
Moser
,
T.
,
Rettig
,
J.
,
Südhof
,
T. C.
,
Neher
,
E.
and
Verhage
,
M.
(
2001
).
Munc18-1 promotes large dense-core vesicle docking
.
Neuron
31
,
581
-
592
.
Washbourne
,
P.
,
Thompson
,
P. M.
,
Carta
,
M.
,
Costa
,
E. T.
,
Mathews
,
J. R.
,
Lopez-Benditó
,
G.
,
Molnár
,
Z.
,
Becher
,
M. W.
,
Valenzuela
,
C. F.
,
Partridge
,
L. D.
et al. 
(
2002
).
Genetic ablation of the t-SNARE SNAP-25 distinguishes mechanisms of neuroexocytosis
.
Nat. Neurosci.
5
,
19
-
26
.
Wu
,
L. G.
,
Hamid
,
E.
,
Shin
,
W.
and
Chiang
,
H.-C.
(
2014
).
Exocytosis and endocytosis: modes, functions, and coupling mechanisms*
.
Annu. Rev. Physiol.
76
,
301
-
331
.
Xu
,
J.
,
Luo
,
F.
,
Zhang
,
Z.
,
Xue
,
L.
,
Wu
,
X.-S.
,
Chiang
,
H.-C.
,
Shin
,
W.
and
Wu
,
L.-G.
(
2013
).
SNARE proteins synaptobrevin, SNAP-25, and syntaxin are involved in rapid and slow endocytosis at synapses
.
Cell Rep.
3
,
1414
-
1421
.
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
.
Zhang
,
C.
,
Omran
,
A. G.
,
He
,
F.
,
Deng
,
X.
,
Wu
,
L.
,
Peng
,
J.
and
Yin
,
F.
(
2014
).
Screening and identification of dynamin-1 interacting proteins in rat brain synaptosomes
.
Brain Res.
1543
,
17
-
27
.
Zheng
,
J.
,
Shen
,
W.-H.
,
Lu
,
T.-J.
,
Zhou
,
Y.
,
Chen
,
Q.
,
Wang
,
Z.
,
Xiang
,
T.
,
Zhu
,
Y.-C.
,
Zhang
,
C.
,
Duan
,
S.
et al. 
(
2008
).
Clathrin-dependent endocytosis is required for TrkB-dependent Akt-mediated neuronal protection and dendritic growth
.
J. Biol. Chem.
283
,
13280
-
13288
.

Competing interests

The authors declare no competing or financial interests.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.

Supplementary information