Presynaptic homeostatic plasticity (PHP) is an evolutionarily conserved form of homeostatic control that is expressed in organisms ranging from fly to human (Cull-Candy et al., 1980; Plomp et al., 1992; Davis, 2013; Wang et al., 2011), at both central and peripheral synapses (Liu and Tsien, 1995; Davis and Goodman, 1998; Burrone et al., 2002; Thiagarajan et al., 2005; Kim and Ryan, 2010; Zhao et al., 2011; Davis, 2013; Henry et al., 2012; Jakawich et al., 2010). PHP can be induced in less than ten minutes and is expressed as a dramatic increase in synaptic vesicle fusion (≥200%) at a fixed number of presynaptic release sites.

An unexplained, emergent property of PHP is the preservation of short-term release dynamics during a stimulus train, referred to here as short-term plasticity or ‘STP’ (Figure 1; see Weyhersmüller et al., 2011; Müller et al., 2012b; Müller et al., 2015; Orr et al., 2017). STP is a fundamental property of neural coding, underlying behaviorally relevant circuit-level computations (Davis and Murphey, 1994). Indeed, STP is described as being ‘…an almost necessary condition for the existence of (short-lived) activity states in the central nervous system’ (Von der Malsburg, 1986; as recently quoted in Taschenberger et al., 2016). Thus, the fact that STP is held constant during the expression of PHP may be essential to the life-long stabilization of neural circuit function and animal behavior. But, it remains fundamentally unknown how presynaptic release can be rapidly doubled at a fixed number of active zones while maintaining constant short-term release dynamics.

Figure 1 with 1 supplement see all

Download asset Open asset

Two processes within the presynaptic terminal are known to be required for the homeostatic potentiation of vesicle release: 1) an increase in presynaptic calcium influx controlled by ENaC channel insertion in the presynaptic membrane (Younger et al., 2013; Orr et al., 2017) and 2) an increase in the readily releasable pool of synaptic vesicles that requires the presynaptic scaffolding proteins RIM and RBP (Müller et al., 2012b; Davis and Müller, 2015; Müller et al., 2015). It has been proposed that the combined effects of elevated presynaptic calcium and an increased supply of release ready vesicles are sufficient to achieve PHP. But, current models have yet to address how short-term release dynamics are stabilized.

STP can be strongly influenced by the partition of the readily releasable vesicle pool into two functional subclasses: (1) docked vesicles that have a low intrinsic calcium sensitivity of release that are referred to as ‘primed’ and (2) docked vesicles that have a relatively higher intrinsic calcium sensitivity and are referred to as ‘super-primed’. During a stimulus train, super-primed vesicles will dominate release during the first few action potentials and primed vesicle will dominate subsequent release (Taschenberger et al., 2016; Lee et al., 2013; Müller et al., 2010). Thus, a synapse harboring a large proportion of super-primed vesicles will favor a high initial release rate followed by synaptic depression while a synapse harboring a small proportion of super-primed vesicles will be prone to facilitation of release, followed by subsequent synaptic depression (Taschenberger et al., 2016; Lee et al., 2013). There is both pharmacological and genetic evidence in support of this model. Genetic mutations that impair vesicle super-priming convert depression-prone, high-release probability synapses into low-release probability synapses that express short-term facilitation (Frank et al., 2006; Deák et al., 2009; He et al., 2017). The calcium sensitivity of vesicle fusion is also highly sensitive to phorbol esters (PdBu), which lower the fusion barrier to release (Taschenberger et al., 2016; Lee et al., 2013). Application of PdBu, which is considered to drive the super-priming process, dramatically potentiates vesicle release and leads to enhanced short-term synaptic depression (Wang et al., 2016; Lee et al., 2013).

Here, we provide evidence that the stabilization of STP during PHP is achieved by maintaining a constant ratio of primed to super-primed synaptic vesicles. First, we confirm that STP is precisely preserved during the expression of PHP. Second, we document a PdBu sensitive release mechanism at the Drosophila NMJ and provide evidence that the fraction of super-primed vesicles is maintained during the full extent of PHP. Third, we provide molecular insight into how a constant fraction of super-primed vesicles is maintained during the expression of PHP. We demonstrate that Unc18 has an evolutionarily conserved function during the rapid induction of PHP. We show that Unc18 function during PHP is facilitated by the activity of presynaptic RIM and, remarkably, is antagonized by presynaptic Syntaxin. Based upon these and other data, we present a new model for the homeostatic control of synaptic vesicle release that is based upon the regulated control of Unc18 levels at the presynaptic release site, acting in concert with the priming activity of RIM to stabilize STP in the presence of a homeostatic doubling presynaptic neurotransmitter release.

We begin by documenting how the short-term dynamics of presynaptic release (short-term plasticity or STP) are held constant during the expression of PHP. We rapidly induce PHP by application of sub-blocking concentrations of the glutamate receptor antagonist philanthotoxin (PhTx, 10–20 µM). PhTx causes a ~ 50% decrease in miniature excitatory postsynaptic potential (mEPSP) amplitude and induces a homeostatic increase in presynaptic vesicle release that precisely counteracts the change in mEPSP amplitude, maintaining the amplitude of action potential evoked neurotransmitter release at baseline levels (Frank et al., 2006; Davis, 2013). As shown in Figure 1A, even at elevated extracellular calcium (3.0 mM [Ca²⁺]_e), excitatory postsynaptic currents (EPSC) are precisely maintained at control levels following application of PhTx (see Müller and Davis, 2012a; see below for further quantification).

Next, we characterize the expression of STP at two concentrations of external calcium (0.75 mM and 3.0 mM [Ca²⁺]_e), doing so in the presence and absence of PhTx to induce PHP (Figure 1A–C). Note that the effects of non-linear summation prevents accurate calculation of presynaptic release at calcium concentrations in excess of approximately 0.3 mM. As such, measurement of synaptic currents in two electrode voltage clamp configuration are essential for experiments where extracellular calcium exceeds 0.3 mM. As expected (Zucker and Regehr, 2002), STP is strongly dependent on the concentration of external calcium, showing facilitation at 0.75 mM [Ca²⁺]_e and depression at 3.0 mM [Ca²⁺]_e (Figure 1A,B). Application of PhTx causes an approximate doubling of presynaptic release during PHP that is similar in magnitude to the change in release observed when comparing wild type neurotransmission at 0.75 mM [Ca²⁺]_e and 3.0 mM [Ca²⁺]_e. However, the expression of PHP occurs without a change in release dynamics (Figure 1B,C). Specifically, there is no statistically significant change in paired pulse ratio comparing the presence and absence of PhTx, and this is true at both 0.75 mM [Ca²⁺]_e where facilitation dominates (p = 0.65; Student’s t-test, two tailed) and at 3.0 mM [Ca²⁺]_e where depression dominates (Figure 1C; p = 0.45; Student’s t-test, two tailed). Thus, PHP is achieved by doubling synaptic vesicle release without altering the expression of STP, confirming prior observations in this system (Weyhersmüller et al., 2011; Müller et al., 2012b; Müller et al., 2015; Orr et al., 2017). We sought to understand how this effect could be achieved.

As outlined in the introduction, STP is strongly influenced by the partition of the readily releasable vesicle pool into two functional subclasses: (1) docked vesicles that have a low intrinsic calcium sensitivity of release that are referred to as ‘primed’ and (2) docked vesicles that have a relatively higher intrinsic calcium sensitivity and are referred to as ‘super-primed’. A synapse with a large fraction of super-primed vesicles will show synaptic depression while a synapse with a small number of super-primed vesicles will facilitate. We hypothesize that the ratio of primed to super-primed vesicles is somehow maintained during the expression of PHP, allowing for a dramatic potentiation of vesicle release without altering the dynamics of release during a stimulus train. To test this hypothesis, we used phorbol esters to probe the ratio of primed to super-primed vesicles at the Drosophila NMJ.

It is well established that phorbol esters (PdBu) decrease the energy barrier to synaptic vesicle fusion, effectively converting the docked/primed vesicle pool into a super-primed, high release probability state (Lee et al., 2013; Taschenberger et al., 2016). Thus, magnitude of PdBu-dependent potentiation of presynaptic release is proportional to the size of the pool of synaptic vesicles that reside in a docked/primed, but not super-primed state. If PHP-dependent potentiation of presynaptic release preserves the ratio of primed to super-primed vesicles, then the effects of PdBu should be the same prior to and following application of PhTx to the synapse.

We first characterize the use of PdBu at the Drosophila NMJ. At 0.75 mM [Ca²⁺]_e, PdBu strongly potentiates both evoked and spontaneous vesicle fusion and converts STP from facilitation to depression (Figure 1A,D,E). Specifically, PdBu has no effect on mEPSP amplitude (Figure 1E), causes a significant increase in EPSC amplitude (Figure 1E) and a corresponding increase in quantal content (Figure 1E). We then express the effects of PdBu as a percent change compared to baseline in the absence of PdBu, observing a significant ~140% increase in release (Figure 1E, right; p<0.05). Three further effects were also quantified. First, application of PdBu causes a significant decrease in the paired-pulse ratio (Figure 1D; p<0.05). Second, we show that presynaptic release converges to a statistically similar steady state in the presence and absence of PdBu during a prolonged stimulus train (Figure 1—figure supplement 1; p>0.1, Student’s t-test, two-tailed; comparison of final three data points of stimulus train). Third, just as observed at the mammalian central synapses, we demonstrate that the effects of PdBu are dependent on the concentration of extracellular calcium (Lee et al., 2013; Taschenberger et al., 2016). When recording at elevated extracellular calcium (3 mM [Ca²⁺]_e), the effect of PdBu on EPSC amplitude is absent (Figure 1F). These data are consistent with the existence of a finite pool of docked vesicles that are uniformly accessed by action-potential induced release at elevated calcium, rendering PdBu without effect. Thus, PdBu functions comparably in Drosophila and at mammalian central synapses. And, by comparing the effects of PdBu on synaptic transmission at 0.75 mM [Ca²⁺]_e we can gain an estimate of the fraction of vesicles that exist within the primed versus super-primed state. Specifically, the magnitude of PdBu-mediated potentiation at 0.75 mM [Ca²⁺]_e is proportional to the number of vesicles that remain in the primed (not super-primed) state.

We next examined the effects of PdBu at synapses previously incubated in PhTx. Once again, at 0.75 mM [Ca²⁺]_e, PhTx causes a decrease in mEPSP amplitude, an increase in quantal content and no change in evoked EPSC amplitude (Figure 1E). When PdBu is applied after PhTx, there is no further change in mEPSP amplitude, compared to PhTx alone, as expected (Figure 1E). However, we find that application of PdBu enhances EPSC amplitudes in the presence of PhTx compared to controls with or without PhTx. The consequence is that quantal content is significantly increased compared to PdBu alone and compared to PhTx alone. Indeed, quantal content is potentiated 3-fold compared to baseline release in wild type (from approximately 200 vesicles per action potential to 600 vesicles per action potential). However, when we calculate the percent change in release caused by PdBu compared to PhTx alone, we find that release is potentiated by ~140%, the same percentage increase caused by PdBu applied to a wild type synapse (Figure 1E). Thus, the proportion of PdBu-sensitive vesicles remains constant following the induction of PHP. Since PHP in a wild type animal can be expressed without a change in short-term plasticity, we propose that the ratio of super-primed to primed vesicles remains constant during expression of PHP (Figure 1G).

Rop is required in the rapid induction of presynaptic homeostasis

In Drosophila, there is a single neuronally expressed unc-18 gene, termed Rop (Ras opposite). Throughout this paper, from this point onward, we refer to the Drosophila gene as Rop and orthologues in other species as unc-18. Prior genetic analyses isolated and characterized numerous mutations in the Rop gene, demonstrating that Rop is essential for spontaneous and evoked synaptic vesicle fusion, in agreement with data from other species (Harrison et al., 1994; Wu et al., 1998; Toonen and Verhage, 2007). We have taken advantage of previously characterized mutations in Rop to study the role of Unc-18 in presynaptic homeostatic plasticity (PHP).

We assayed PHP in two independent, heterozygous Rop loss-of-function mutants: 1) a previously characterized null allele (Rop^G27; Harrison et al., 1994; see Figure 2A) and 2) a small chromosomal deficiency that deletes the entire Rop gene locus (Df^(3L)BSC735, referred to hereafter as Df^Rop; Cook et al., 2012; Figure 2A). In both Rop^G27/+ and Df^Rop/+ heterozygous mutants, PHP is significantly suppressed compared to wild type (Figure 2B–2I). Specifically, upon application of PhTx, Rop^G27/+ mutants show an enhancement of quantal content (Figure 2H; p<0.01), but the magnitude of this enhancement is statistically significantly smaller than that observed in wild type (Figure 2I; p<0.001). Consistent with impaired homeostatic plasticity, EPSP amplitudes are significantly reduced in the presence of PhTx compared to baseline EPSP amplitudes (Figure 2C and G; p<0.05; p<0.0001). Notably, baseline release, recorded in the absence of PhTx, is unaltered in both Rop^G27/+ and Df^Rop/+ heterozygous mutants at the concentration of external calcium used in this experiment (0.3 mM [Ca²⁺]_e) (Figure 2C and G). Thus, two independently derived heterozygous loss-of-function mutants suppress PHP without an effect on baseline neurotransmission, arguing that PHP is highly sensitive to Rop gene dosage.

Figure 2

Download asset Open asset

Neuronally expressed rop is required for presynaptic homeostasis

The unc-18 gene is broadly expressed and has been shown to participate in membrane trafficking events outside the nervous system (Hata and Südhof, 1995; Riento et al., 2000; Toonen and Verhage, 2003). As such, Rop could function either pre- or postsynaptically during PHP. Therefore, we knocked down Rop expression specifically in the nervous system using the Gal4/UAS expression system. We took advantage of a previously characterized UAS-Rop-RNAi transgene (P{GD1523}v19696; Dietzl et al., 2007) to knock-down Rop specifically in neurons (c155-Gal4; Lin and Goodman, 1994). First, we demonstrate that PHP is strongly suppressed when UAS-Rop-RNAi is driven presynaptically (Figure 3A–E). Thus, Rop is necessary presynaptically for PHP. We note that presynaptic Rop knockdown also causes a decrease in baseline EPSP amplitude and quantal content by ~30% (Figure 3C; p<0.01), suggesting that presynaptic knockdown depletes Rop protein levels more substantially than that observed in the heterozygous Rop null mutant (Figure 2F–I). Indeed, when we compare wild type, Rop^G27/+ and presynaptic Rop knockdown, we find a progressively more severe decrease in mEPSP frequency that is consistent with progressively more severe depletion of Rop protein (Figure 3F). Thus, Rop knockdown can be considered a strong hypomorphic condition, supporting the conclusion that presynaptic Rop is essential for robust expression of PHP.

Figure 3

Download asset Open asset

Separable activity of rop during baseline transmission and PHP

Although PHP is selectively impaired in the Rop^G27/+ heterozygous mutant at 0.3 mM [Ca²⁺]_e, it remains formally possible that loss of Rop places a limit on the number of vesicles that can be released per action potential (quantal content) and, thereby, indirectly restricts the expression of PHP. To address this possibility, we asked whether baseline release and PHP change in parallel as a function of altered extracellular calcium in the Rop^G27/+ heterozygous mutations. Experiments were conducted at 0.3 mM [Ca²⁺]_e, 0.75 mM [Ca²⁺]_e, 1.5 mM [Ca²⁺]_e and 3.0 mM [Ca²⁺]_e. First, we find that baseline neurotransmitter release is impaired at 0.75 as well as 1.5 and 3.0 mM [Ca²⁺]_e in the Rop^G27 /+ heterozygous mutants compared to wild-type (Figure 4). But, there remains a highly cooperative relationship between extracellular calcium and vesicle release in the Rop^G27 /+ heterozygous mutants (Figure 4F). Next, we demonstrate that PHP is suppressed both at 1.5 mM (Figure 4A–C) and 3.0 mM [Ca²⁺]_e (Figure 4D–E), as evidenced by an inability of EPSCs to be restored to baseline values in the presence of PhTx (Figures 4C and 3E) (p<0.05). When we calculate the percent suppression of PHP at 0.3, 1.5, and 3.0 mM [Ca²⁺]_e we find a constant level of PHP suppression even though baseline release increases ~10 fold over this range of extracellular calcium concentrations (Figure 4F). From this we can make two conclusions. First, since release remains sensitive to changes in external calcium, a ceiling effect cannot explain the defect in PHP observed in the Rop^G27 /+ heterozygous mutants. More specifically, the increase in release in Rop^G27/+ comparing 0.3 mM [Ca²⁺]_e with 3.0 mM [Ca²⁺]_e vastly exceeds the change in vesicular release that would be expected for full expression of PHP at 0.3 mM [Ca²⁺]_e (see Figure 2). The fact that PHP is consistently inhibited by ~30%, irrespective of the concentration of external calcium, demonstrates that the magnitude of PHP expression correlates directly with the levels of Rop expression, not with the magnitude of evoked release or extracellular calcium concentration, arguing for an essential role of Rop in the mechanisms of PHP.

Figure 4

Download asset Open asset

Rop dependent vesicle priming correlates with expression of PHP

Unc-18 has been implicated in several different stages of the release process, including vesicle docking, priming, fusion pore formation and regulation of the readily releasable pool (RRP) of synaptic vesicles (Weimer et al., 2003; Toonen et al., 2006; Gulyás-Kovács et al., 2007; Fisher et al., 2001; Toonen and Verhage, 2007). Since we observe a significant disruption of PHP in the heterozygous Rop^G27/+ mutant background, we sought to define the parameters of presynaptic release that are particularly sensitive to the heterozygous Rop^G27/+ mutant, highlighting those actions of Rop that best correlate with impaired PHP. First, we observe a decrease in the frequency of spontaneous mEPSP events in Rop mutants (Figure 3F). mEPSP frequency is decreased by 36% and 59% compared to wild-type in Rop^G27/+ and UAS-Rop-RNAi, respectively (Figure 3F). This is consistent with prior reports in Drosophila and other systems (Wu et al., 1998; Toonen et al., 2006; Patzke et al., 2015).

It has previously been shown that a homeostatic modulation of the readily releasable pool (RRP) of vesicles is required for the expression of PHP (Müller et al., 2012b). RRP size can be estimated through quantification of the cumulative EPSC amplitude during a high frequency stimulus train (60 HZ, 30 stimuli) at elevated extracellular calcium (1.5 mM; Figure 5) and back extrapolation according to published protocols (Schneggenburger et al., 1999; Müller et al., 2015). We demonstrate that the cumulative EPSC amplitude in the Rop^G27/+ heterezogous mutant is unaltered at baseline compared to wild-type (Figure 5B). Then, we demonstrate that the Rop^G27/+ mutants show normal modulation of the RRP following application of PhTx, as demonstrated the maintenance of the cumulative EPSC amplitude in the presence and absence of PhTx, which diminishes the amplitude of underlying unitary release events by ~50% (Figure 5B). Thus, at the Drosophila NMJ, the RRP is not sensitive to Rop haplo-insufficiency and Rop is not limiting for the homeostatic potentiation of the RRP.

Figure 5

Download asset Open asset

It remains apparent, however, that the initial EPSC of the stimulus train in Rop^G27/+ mutants, recorded in the presence of PhTx, is smaller than that observed in wild type (Figure 5A–B see also Figure 3C for quantification). Thus, the homeostatic potentiation of the initial EPSC amplitude is disrupted in the Rop^G27/+ mutants, whereas the homeostatic potentiation of the RRP is normal. This is quantified by dividing the initial EPSC amplitude by the cumulative EPSC amplitude during the stimulus train, a parameter referred to as P_train (Figure 5C). It is apparent that release dynamics are altered as a consequence of decreased P_train in the Rop^G27/+ mutant in the presence of PhTx. Super-priming should contribute significantly to vesicle release in response to the first action potential of a stimulus train. Thus, one explanation for this result is that the super-primed vesicle pool has not been appropriately expanded in the Rop^G27/+ mutant. During the stimulus train, elevated intra-terminal calcium could overcome Rop^G27/+ haploinsufficiency by driving the normal priming process, leading to expansion of the RRP in the presence of PhTx. These data argue that Rop may be essential for the expansion of the super-primed vesicle pool during expression of PHP. It is worth noting that this is the first example of a significant alteration in presynaptic release dynamics that is specific to the induction of PHP.

A genetic interaction of rop and RIM during PHP

There has been considerable progress identifying presynaptic proteins that are necessary for presynaptic homeostatic plasticity (Davis, 2013; Müller et al., 2015). These genes include the active zone associated scaffolding proteins RIM (Rab3 Interacting Molecule) and RBP (RIM Binding Protein), both of which are components of a proposed molecular priming pathway within the presynaptic nerve terminal (Südhof, 2012a). If Rop is integrally involved in the mechanisms of PHP within the presynaptic terminal, potentially acting in the priming process during PHP, then we might expect genetic interactions with rim and RBP.

Genetic interactions were performed by assaying heterozygous, null mutations alone and in comparison to the effects observed in a double heterozygous condition (Frank et al., 2009; Müller et al., 2015). We first assayed baseline release. At 0.3 mM [Ca²⁺]_e, baseline release was normal in both Rop^G27/+ and rim¹⁰³/+ (Figure 6B). However, release (quantal content; 6C) was diminished by nearly 50% in the double heterozygous condition (WT QC = 34.8 ± 0.8; Rop^G27/+ QC = 30.1±1.6; rim¹⁰³/+ QC = 32.3±1.4; Rop^G27/rim¹⁰³ QC = 18.7 ± 1.3) (Figure 6C). This is a very strong genetic interaction for baseline release, even by comparison with previously published genetic interactions of other genes with rim (Wang et al., 2016; Orr et al., 2017; Hauswirth et al., 2018).

Figure 6

Download asset Open asset

Next, we assessed PHP. Upon application of PhTx, a heterozygous null mutation in rim (rim¹⁰³/+) causes a suppression of PHP (Figure 6D). This suppression is similar in magnitude to that observed in the Rop^G27/+ heterozygous null mutation (Figure 6D). However, when we examine a double heterozygous mutant with Rop^G27 /+ placed in trans to rim¹⁰³ /+, PHP is completely blocked (Figure 6C,D). To underscore the robustness of this genetic interaction, we plot the relationship between mEPSP amplitude and quantal content for individual recordings (Figure 6E). Each data point represents the average mEPSP and quantal content for a single NMJ recording (see also Hauswirth et al., 2018). In wild type, data can be fit with a line representing the homeostatic process. We also present dotted lines that encompass 95% of all of the data points in the wild type graph. This fit and 95% data interval are used to compare data distributions to other mutant backgrounds. In both of the single heterozygous mutants (Rop^G27 /+ and rim¹⁰³ /+) the data points lie below the best-fit line for wild type, but are largely retained within the 95% interval, consistent with a minor suppression of PHP when all data points are averaged (Figure 6C,D). However, in the double heterozygous animal (Rop^G27/rim¹⁰³) it is clear that PHP fails and the majority of data points in the presence of PhTx reside outside the interval that contains 95% of all wild type data points. More specifically, for heterozygous null Rop and rim recordings in the presence of PhTx, 25% and 16% of these individual data points in the presence of PhTx fall outside of the lines encompassing 95% of wild-type data (Figure 6E). In the double heterozygous mutants, 87% of PhTx data fall outside of the 95% confidence intervals (Figure 6E). This genetic interaction, referring specifically to PHP expression, is also conserved at elevated calcium levels (1.5 and 3.0 mM calcium; data not shown).

At baseline, there is a strong synergistic interaction between Rop and RIM, suggesting that both are functioning to control the same process relevant to vesicle release, perhaps synaptic vesicle priming (Gulyás-Kovács et al., 2007; Koushika et al., 2001). It remains unclear whether rop and rim function in the same genetic pathway based on this genetic interaction. The data are equally compatible with parallel pathways converging on the mechanism of PHP. Regardless, our data underscore that Rop has an essential function during the process of PHP. We consider this a particularly important line of reasoning, since it is not feasible to eliminate Rop and assess a block in PHP.

Next, we sought to understand the specificity of the genetic interaction between Rop and rim. To do so, we performed a series of additional genetic interactions with other genes previously implicated in the presynaptic synaptic vesicle priming process. First, we asked whether Rop shows a genetic interaction with RIM Binding Protein (RBP). RBP biochemically interacts with RIM and is thought to form an extended presynaptic scaffold (Wang et al., 2000; Liu et al., 2011). In Drosophila, RBP is also essential for both baseline release and PHP (Müller et al., 2015). But, the mechanism by which RBP participates in PHP is distinct from the mechanism by which RIM participates in PHP (Müller et al., 2015). The heterozygous null mutation in rbp/+ has no effect on baseline transmission at 0.3 mM external calcium, consistent with prior observations (Müller et al., 2015) (Supplementary file 1). We then demonstrate that PHP is also normal in the Rop^G27/rbpSTOP^STOP1 double heterozygous mutant (Figure 6F). This is evidence that there is specificity to the rop interaction with rim during PHP. It is worth noting that PHP is highly variable in the Rop^G27/rbpSTOP^STOP1 double heterozygous mutant, preventing us from making any additional conclusions regarding whether PHP is similar to, or different from wild type.

Next, we asked whether rim genetically interacts with Unc-13 during baseline neurotransmitter release and PHP. Munc-13 is known to biochemically interact with RIM (Betz et al., 2001). Alone, the dunc-13^P84200/+ heterozygous null mutation has no effect on baseline transmission or PHP (Supplementary file 1). Remarkably, the double heterozygous condition of rim and dunc-13 also has no effect on baseline transmission or PHP (Supplementary file 1 and Figure 6G). A similar set of findings is observed when we tested a double heterozygous condition of rim with rbp (Supplementary file 1 and Figure 6H). While baseline transmission is decreased in the double heterozygous combination of rim/+ and rbp/+, we find that PHP is normal, confirming previously published data (Müller et al., 2015). It is somewhat surprising that the rim/+; unc13/+ double heterozygous mutant has significantly more PHP than observed in rim/+ alone, perhaps relating to the functional importance of the RIM-Unc13 biochemical interaction during PHP. Regardless, when taken together, our results underscore the specificity and importance of the genetic interaction between rop and rim and the relevance of rop to the mechanisms of PHP.

Loss of rop and rim limits the super-primed vesicle pool

We note that heterozygous Rop^G27/+ mutants have a defect in PHP that is apparent on the first action potential of a stimulus train, but the homeostatic expansion of the RRP is apparent upon further stimulation (Figure 5). Both Rop and Rim are well-established molecular players that contribute to synaptic vesicle priming (Südhof, 2012a; Cook et al., 2012). We returned to the use of PdBu to assess whether the block of PHP in the Rop^G27/rim¹⁰³ double heterozygous mutant is correlated with a deficit in vesicle super-priming. As outlined above, the magnitude of PdBu-dependent potentiation in response to a single action potential should reflect the balance of primed to super-primed vesicles. A large PdBu effect argues for a smaller pool of super-primed vesicle pool. We compared wild type to the Rop^G27/rim¹⁰³ double heterozygous mutant, recording in the presence and absence of PdBu (Figure 7). Experiments were performed at 0.75 mM [Ca²⁺]_e, a condition in which PdBu potentiates wild type synapses by ~165% (Figure 7A–D). Next, we demonstrate that application of PdBu to each heterozygous mutant alone, either Rop^G27/+ or rim¹⁰³/+, is identical to wild type (Figure 7C,D; p>0.1 ANOVA). Then, we demonstrate that PdBu has a dramatically increased effect size when applied to the Rop^G27/rim¹⁰³ double heterozygous mutant, potentiating release by more than 350%, a dramatically increased effect size compared to application of PdBu to wild type and single heterozygous controls (Figure 7A–D, p<0.01; ANOVA). These data argue that the Rop^G27/rim¹⁰³ double heterozygous mutant limits the size of the super-primed vesicle pool, thereby impairing vesicle release in response to a single action potential, an effect that is strongly correlated with a block in the expression of PHP. Two other observations should be noted. First, the Rop^G27/rim¹⁰³ double heterozygous mutant recorded at 0.75 mM [Ca²⁺]_e has a greater effect on baseline release than observed at 0.3 mM [Ca²⁺]_e. The basis for this effect is unknown. Second, we show that the enhanced effects of PdBu on the Rop^G27/rim¹⁰³ double heterozygous mutant is consistent with the potentiation of release probability and an associated change in paired pulse ratio (Figure 7C). This simply confirms that PdBu is behaving as expected when applied to the Rop^G27/rim¹⁰³ double heterozygous mutant. Again, this type of synergistic genetic interaction is generally taken as evidence that two genes participate in the same process. It is not possible to make any conclusion regarding whether these genes are acting in a linear signaling pathway or in parallel signaling pathways. But, based on formal genetic argument, we are able to conclude that they likely converge to control the same processes at the presynaptic terminal, both the PdBu sensitive vesicle pool and PHP.

Figure 7

Download asset Open asset

Loss of syx1A rescues PHP in the heterozygous rop mutant background

Unc-18 is a well-established syntaxin binding protein (Hata et al., 1993; Pevsner et al., 1994; Halachmi et al., 1995). The Unc-18 interaction with Syntaxin is complex, including progressive interactions with closed and open conformations of Syntaxin. Ultimately, the Unc-18 interaction with open Syntaxin is thought to catalyze SNARE assembly, greatly decreasing the energy barrier to calcium-driven vesicle fusion. Indeed, Unc-18 binding to open Syntaxin is believed to be a prerequisite for efficient synaptic vesicle fusion (Dulubova et al., 2007; Deák et al., 2009). Based on the genetic interactions reported above, we expected that loss of syx1A would strongly enhance the loss of function phenotype of the heterozygous Rop^G27/+ mutant, resulting in diminished vesicle release and a block of PHP.

We acquired a previously published null mutation in syntaxin1A and examined baseline neurotransmitter release in a heterozygous syx1A mutant (syx1A^Δ229; Schulze et al., 1995). There is no change in baseline mEPSP amplitude, EPSP amplitude or quantal content in syx1A^Δ229/+ assayed at 0.3 mM [Ca²⁺]_e (Figure 8A–D). Thus, syntaxin is not haplo-insufficient for baseline release. Remarkably, the same is true for the double heterozygous condition, combining syx1A^Δ229/+ with Rop^G27/+. Baseline transmission is normal compared to wild type (Figure 8D). In other systems, it is estimated that Syntaxin1A is present in ~5 fold excess compared to the levels of synaptic Unc-18 (Graham et al., 2004). This could explain the lack of a genetic interaction at baseline.

Figure 8

Download asset Open asset

Next, we examined PHP. A heterozygous null mutation in syx1A (syx1A^Δ229/+) has normal PHP (Figure 8E), again consistent with a possible excess of Syntaxin protein at the release site. The heterozygous null mutation in Rop (Rop^G27/+) suppresses homeostatic potentiation (Figure 8E), confirming experiments presented earlier in this study. Remarkably, when we place Rop^G27 in trans to syx1A^Δ229 (double heterozygous condition), we find that homeostasis is fully expressed (Figure 8E). Impaired PHP caused by the Rop^G27/+ mutation is completely rescued to wild type levels. More specifically, in the double heterozygous mutants, EPSP amplitudes fully compensate in the presence of PhTx (Figure 8C) and there is a wild type level enhancement of quantal content (Figure 8E) in the presence of PhTx. Finally, the rescue of the Rop^G27/+ mutation by syx1A^Δ229/+ is not restricted to PHP. The rate of spontaneous vesicle fusion is also restored to wild type levels, underscoring the validity of this genetic rescue (Figure 8F). Note that there is, as yet, no indication that mEPSP frequency is directly responsible for the induction of PHP (Frank et al., 2006; Frank et al., 2009; Harris et al., 2015; Goold and Davis, 2007). Since we are examining heterozygous, null mutations, the most parsimonious conclusion is that Syntaxin normally functions to restrict the action of Rop that is required for PHP. Reducing the level of Syntaxin relieves a restriction on Rop activity and restores full expression of PHP. To provide further evidence for this surprising finding, we sought to disrupt the physical interaction of Rop and Syntaxin and assess whether this might also rescue the expression of PHP in the Rop^G27/+ mutant background.

Evidence that Syx1A restrains rop from participating in PHP

We performed an in vitro binding assay to confirm the biochemical interaction between Rop and Syntaxin1A. Recombinant wild-type Rop protein binds strongly to recombinant Syntaxin1A (GST-syx1A^ΔC) (Figure 9B–C) (K_D = 0.4 μM). Next, we surveyed previously characterized point mutations in the Rop gene, searching for candidate mutations that reside near the conserved Syntaxin binding interface (Figure 9A). Unc-18 binds to Syntaxin at two sites, an N-terminal region that interacts with the N-terminal peptide of Syntaxin and a helical region that represents a larger interaction surface. The Rop^G11 mutant harbors a point mutation (Asp 45→ Asn; Harrison et al., 1994) that resides within or directly adjacent to a predicted helical Syntaxin binding interface (Misura et al., 2000). Here, we demonstrate that this mutation completely abolishes in vitro binding between recombinant Rop^G11 mutant protein and recombinant GST-syx1A^ΔC protein (Figure 9B–C) (K_D = 36.1 μM).

Figure 9

Download asset Open asset

We next assayed baseline release in the Rop^G11 mutant, placed in trans to either a deficiency that removes the Rop gene locus, or the Rop^G27 null allele. To our surprise, both allelic combinations are viable to the third instar stage and we observe robust neurotransmitter release (Figure 9D). Specifically, we find that EPSC amplitudes are decreased by ~45% on average in both the Rop^G11/Df^Rop and Rop^G11/Rop^G27 allelic combinations compared to wild type (WT EPSC = 217.4 nA±7.3; Rop^G27/Rop^G11 EPSC = 134.2 nA±12.2; p<0.01; Df^Rop/Rop^G11 EPSC = 156.3 nA±17.2; p<0.01) (Figure 9F). This effect is more severe than the defect observed in the Rop^G11/+ heterozygous condition, consistent with loss of Rop function in both the Rop^G11/Df^Rop and Rop^G11/Rop^G27 allelic combinations. But, the presence of synchronous release is very surprising, since there is no wild type Rop protein at the synapse. From these data we conclude that the Rop^G11 mutation does not completely block all interactions between Rop and Syntaxin in vivo. The in vitro binding assay is predicted to test the binding of Rop to the closed conformation of Syntaxin. We speculate, based on work in other systems, that Rop^G11 could be localized to the SNARE complex through other molecular interactions, in vivo. Once at the release site, Rop^G11 mutant protein might still interact with the open conformation of Syntaxin and facilitate SNARE-mediated fusion.

Next, we assayed PHP in the Rop^G11/+ as well as the Rop^G11/Df^Rop and Rop^G11/Rop^G27 allelic combinations. In the presence of PhTx, EPSC amplitudes are restored to baseline values in the Rop^G11/+ mutant, indicative of fully functional PHP (Figure 9D–F). Remarkably, PHP is also fully expressed in both the Rop^G11/Df^Rop and Rop^G11/Rop^G27 allelic combinations (Figure 9D–F). Thus, the presence of the Rop^G11 allele fully rescues PHP in the presence of the Rop^G27/+ as well as the Df^Rop/+ alleles. While surprising, these data are entirely consistent with the observation that a heterozygous null mutation in syntaxin1A rescues PHP in the Rop^G27 mutant background (Figure 8). Taken together, our data are consistent with an emerging model in which Syx1A regulates the availability of Rop to participate in PHP (see discussion).

We have advanced our understanding of presynaptic homeostatic plasticity in several important ways. First, we demonstrate that rop (Unc-18) is essential for PHP, significantly extending prior evidence presented at the mouse NMJ (Sons et al., 2003). Thus, we argue that rop (Unc-18) is an evolutionarily conserved component of PHP signaling, enabling expression of PHP at the NMJ of Drosophila (this paper) and mice (Sons et al., 2003). Indeed, Unc-18 is the first molecular signaling component demonstrated to have a conserved, required function during PHP in both systems. Fly and mouse are separated by nearly 500 million years of evolution, suggesting that the molecular mechanisms of PHP may be as ancient as mechanisms that achieve action-potential induced, calcium-dependent, neurotransmitter release.

Second, our data demonstrate that Rop is limiting for the expression of PHP. Prior work at the mouse NMJ was the first to provide evidence that Unc-18 might participate in PHP (Sons et al., 2003). However, the prior work only included an analysis of Unc-18/+ heterozygous null mutant animals and, ultimately, could not rule out the formal possibility that Unc-18 was essential for baseline release and, as a secondary consequence, limited the expression of PHP. We provide several lines of evidence supporting the conclusion that Unc-18 has an activity that is necessary for PHP. For example, neurotransmitter release in the heterozygous Rop^G27 /+ mutant remains highly sensitive to changes in extracellular calcium. Yet, across a 10-fold range of extracellular calcium, PHP is suppressed by a constant fraction of ~30%. We also pursued a series of genetic interactions and provide evidence for a strong, specific, genetic interaction of Rop with rim, placing Unc-18 within a known PHP signaling framework.

Third, Unc-18 is the first integral component of the synaptic vesicle fusion apparatus to be linked to the expression of PHP. As such, our data provide a reasonable endpoint for the presynaptic homeostatic signaling system. In recent years, trans-synaptic signaling molecules have been shown to be required for PHP including Semaphorin/Plexin signaling (Orr et al., 2017), innate immune signaling (Harris et al., 2015) and signaling from the synaptic matrix (Wang et al., 2016). Many of these signaling systems interact with the presynaptic scaffolding protein RIM (Harris et al., 2015; Wang et al., 2016; Orr et al., 2017; Hauswirth et al., 2018). But, it was previously unknown what molecular mechanisms reside downstream of RIM, since RIM is a molecular scaffolding protein. We cannot formally conclude, based on our genetic data, that Rop functions downstream of RIM. However, it is clear that Rop is in a position to directly modulate the fusion apparatus and, as such, is very likely to mediate signaling that is localized to the active zone by the RIM-dependent cytomatrix.

The rescue of PHP by loss of Syntaxin or by disruption of Rop-Syntaxin binding is a surprise, but one that can be understood when placed in the context of work previously documented in other systems. A recent single molecule imaging study demonstrates that the majority of Unc-18 may reside outside of active zone with limited mobility (Smyth et al., 2013). It has also been shown that Syntaxin is present in large excess compared to Unc-18, and Syntaxin protein is broadly distributed beyond sites of synaptic vesicle fusion (Broadie et al., 1995; Graham et al., 2004). Thus, the majority of Unc-18 protein could interact with Syntaxin in the peri-active zone, presumably binding a closed Syntaxin conformation. Accordingly, Unc-18 would be in equilibrium, moving between a peri-active zone reservoir and fusion competent vesicles at the active zone. In this way, Syntaxin could restrict the amount of Unc18 available for participation in synaptic vesicle fusion at the release site. Thus, when we remove one copy of the syntaxin gene, or diminish the binding of Unc-18 to the closed confirmation of Syntaxin, we might be shifting the distribution of Unc-18 toward the release site and achieve a rescue PHP. This model is consistent with data in other systems, demonstrating that the levels of Unc18 at the release site can influence vesicle release rate. For example, over-expression of unc-18 in mice is sufficient to potentiate vesicle release (Voets et al., 2001), albeit to a limited extent (Toonen and Verhage, 2003). Thus, we propose that the expression of PHP involves an as yet unknown signaling event that mobilizes Unc-18 to the active zone where it is sufficient to promote vesicle priming as a final stage necessary for the full expression of PHP.

Data generate during this study are included in the manuscript and supporting files and tables.

1. 1. Aravamudan B
   2. Fergestad T
   3. Davis WS
   4. Rodesch CK
   5. Broadie K

   (1999) Drosophila UNC-13 is essential for synaptic transmission

   Nature Neuroscience 2:965–971.

   https://doi.org/10.1038/14764

   • PubMed
   • Google Scholar
2. 1. Betz A
   2. Thakur P
   3. Junge HJ
   4. Ashery U
   5. Rhee JS
   6. Scheuss V
   7. Rosenmund C
   8. Rettig J
   9. Brose N

   (2001) Functional interaction of the active zone proteins Munc13-1 and RIM1 in synaptic vesicle priming

   Neuron 30:183–196.

   https://doi.org/10.1016/S0896-6273(01)00272-0

   • PubMed
   • Google Scholar
3. 1. Broadie K
   2. Prokop A
   3. Bellen HJ
   4. O'Kane CJ
   5. Schulze KL
   6. Sweeney ST

   (1995) Syntaxin and synaptobrevin function downstream of vesicle docking in drosophila

   Neuron 15:663–673.

   https://doi.org/10.1016/0896-6273(95)90154-X

   • PubMed
   • Google Scholar
4. 1. Burrone J
   2. O'Byrne M
   3. Murthy VN

   (2002) Multiple forms of synaptic plasticity triggered by selective suppression of activity in individual neurons

   Nature 420:414–418.

   https://doi.org/10.1038/nature01242

   • PubMed
   • Google Scholar
5. 1. Cook RK
   2. Christensen SJ
   3. Deal JA
   4. Coburn RA
   5. Deal ME
   6. Gresens JM
   7. Kaufman TC
   8. Cook KR

   (2012) The generation of chromosomal deletions to provide extensive coverage and subdivision of the drosophila melanogaster genome

   Genome Biology 13:R21.

   https://doi.org/10.1186/gb-2012-13-3-r21

   • PubMed
   • Google Scholar
6. 1. Cull-Candy SG
   2. Miledi R
   3. Trautmann A
   4. Uchitel OD

   (1980) On the release of transmitter at normal, myasthenia gravis and myasthenic syndrome affected human end-plates

   The Journal of Physiology 299:621–638.

   https://doi.org/10.1113/jphysiol.1980.sp013145

   • PubMed
   • Google Scholar
7. 1. Davis GW
   2. Goodman CS

   (1998) Synapse-specific control of synaptic efficacy at the terminals of a single neuron

   Nature 392:82–86.

   https://doi.org/10.1038/32176

   • PubMed
   • Google Scholar
8. 1. Davis GW
   2. Murphey RK

   (1994) Retrograde signaling and the development of transmitter release properties in the invertebrate nervous system

   Journal of Neurobiology 25:740–756.

   https://doi.org/10.1002/neu.480250612

   • PubMed
   • Google Scholar
9. 1. Davis GW

   (2013) Homeostatic signaling and the stabilization of neural function

   Neuron 80:718–728.

   https://doi.org/10.1016/j.neuron.2013.09.044

   • PubMed
   • Google Scholar
10. 1. Davis GW
    2. Müller M

    (2015) Homeostatic control of presynaptic neurotransmitter release

    Annual Review of Physiology 77:251–270.

    https://doi.org/10.1146/annurev-physiol-021014-071740

    • PubMed
    • Google Scholar
11. 1. Deák F
    2. Xu Y
    3. Chang WP
    4. Dulubova I
    5. Khvotchev M
    6. Liu X
    7. Südhof TC
    8. Rizo J

    (2009) Munc18-1 binding to the neuronal SNARE complex controls synaptic vesicle priming

    The Journal of Cell Biology 184:751–764.

    https://doi.org/10.1083/jcb.200812026

    • PubMed
    • Google Scholar
12. 1. Dietzl G
    2. Chen D
    3. Schnorrer F
    4. Su KC
    5. Barinova Y
    6. Fellner M
    7. Gasser B
    8. Kinsey K
    9. Oppel S
    10. Scheiblauer S
    11. Couto A
    12. Marra V
    13. Keleman K
    14. Dickson BJ

    (2007) A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila

    Nature 448:151–156.

    https://doi.org/10.1038/nature05954

    • PubMed
    • Google Scholar
13. 1. Dulubova I
    2. Khvotchev M
    3. Liu S
    4. Huryeva I
    5. Südhof TC
    6. Rizo J

    (2007) Munc18-1 binds directly to the neuronal SNARE complex

    PNAS 104:2697–2702.

    https://doi.org/10.1073/pnas.0611318104

    • PubMed
    • Google Scholar
14. 1. Fisher RJ
    2. Pevsner J
    3. Burgoyne RD

    (2001) Control of fusion pore dynamics during exocytosis by Munc18

    Science 291:875–878.

    https://doi.org/10.1126/science.291.5505.875

    • PubMed
    • Google Scholar
15. 1. Frank CA
    2. Kennedy MJ
    3. Goold CP
    4. Marek KW
    5. Davis GW

    (2006) Mechanisms underlying the rapid induction and sustained expression of synaptic homeostasis

    Neuron 52:663–677.

    https://doi.org/10.1016/j.neuron.2006.09.029

    • PubMed
    • Google Scholar
16. 1. Frank CA
    2. Pielage J
    3. Davis GW

    (2009) A presynaptic homeostatic signaling system composed of the Eph receptor, ephexin, Cdc42, and CaV2.1 calcium channels

    Neuron 61:556–569.

    https://doi.org/10.1016/j.neuron.2008.12.028

    • PubMed
    • Google Scholar
17. 1. Genc O
    2. Kochubey O
    3. Toonen RF
    4. Verhage M
    5. Schneggenburger R

    (2014) Munc18-1 is a dynamically regulated PKC target during short-term enhancement of transmitter release

    eLife 3:e01715.

    https://doi.org/10.7554/eLife.01715

    • PubMed
    • Google Scholar
18. 1. Goold CP
    2. Davis GW

    (2007) The BMP ligand Gbb gates the expression of synaptic homeostasis independent of synaptic growth control

    Neuron 56:109–123.

    https://doi.org/10.1016/j.neuron.2007.08.006

    • PubMed
    • Google Scholar
19. 1. Graham ME
    2. Barclay JW
    3. Burgoyne RD

    (2004) Syntaxin/Munc18 interactions in the late events during vesicle fusion and release in exocytosis

    Journal of Biological Chemistry 279:32751–32760.

    https://doi.org/10.1074/jbc.M400827200

    • PubMed
    • Google Scholar
20. 1. Gulyás-Kovács A
    2. de Wit H
    3. Milosevic I
    4. Kochubey O
    5. Toonen R
    6. Klingauf J
    7. Verhage M
    8. Sørensen JB

    (2007) Munc18-1: sequential interactions with the fusion machinery stimulate vesicle docking and priming

    Journal of Neuroscience 27:8676–8686.

    https://doi.org/10.1523/JNEUROSCI.0658-07.2007

    • PubMed
    • Google Scholar
21. 1. Halachmi N
    2. Feldman M
    3. Kimchie Z
    4. Lev Z

    (1995)

    Rop and Ras2, members of the Sec1 and Ras families, are localized in the outer membranes of labyrinthine channels and vesicles of Drosophila nephrocyte, the Garland cell

    European Journal of Cell Biology 67:275–283.

    • PubMed
    • Google Scholar
22. 1. Harris N
    2. Braiser DJ
    3. Dickman DK
    4. Fetter RD
    5. Tong A
    6. Davis GW

    (2015) The Innate Immune Receptor PGRP-LC Controls Presynaptic Homeostatic Plasticity

    Neuron 88:1157–1164.

    https://doi.org/10.1016/j.neuron.2015.10.049

    • PubMed
    • Google Scholar
23. 1. Harrison SD
    2. Broadie K
    3. van de Goor J
    4. Rubin GM

    (1994) Mutations in the Drosophila Rop gene suggest a function in general secretion and synaptic transmission

    Neuron 13:555–566.

    https://doi.org/10.1016/0896-6273(94)90025-6

    • PubMed
    • Google Scholar
24. 1. Hata Y
    2. Slaughter CA
    3. Südhof TC

    (1993) Synaptic vesicle fusion complex contains unc-18 homologue bound to syntaxin

    Nature 366:347–351.

    https://doi.org/10.1038/366347a0

    • PubMed
    • Google Scholar
25. 1. Hata Y
    2. Südhof TC

    (1995) A novel ubiquitous form of Munc-18 interacts with multiple syntaxins. Use of the yeast two-hybrid system to study interactions between proteins involved in membrane traffic

    The Journal of Biological Chemistry 270:13022–13028.

    https://doi.org/10.1074/jbc.270.22.13022

    • PubMed
    • Google Scholar
26. 1. Hauswirth AG
    2. Ford KJ
    3. Wang T
    4. Fetter RD
    5. Tong A
    6. Davis GW

    (2018) A postsynaptic PI3K-cII dependent signaling controller for presynaptic homeostatic plasticity

    eLife 7:e31535.

    https://doi.org/10.7554/eLife.31535

    • PubMed
    • Google Scholar
27. 1. He E
    2. Wierda K
    3. van Westen R
    4. Broeke JH
    5. Toonen RF
    6. Cornelisse LN
    7. Verhage M

    (2017) Munc13-1 and Munc18-1 together prevent NSF-dependent de-priming of synaptic vesicles

    Nature Communications 8:15915.

    https://doi.org/10.1038/ncomms15915

    • PubMed
    • Google Scholar
28. 1. Henry FE
    2. McCartney AJ
    3. Neely R
    4. Perez AS
    5. Carruthers CJ
    6. Stuenkel EL
    7. Inoki K
    8. Sutton MA

    (2012) Retrograde changes in presynaptic function driven by dendritic mTORC1

    Journal of Neuroscience 32:17128–17142.

    https://doi.org/10.1523/JNEUROSCI.2149-12.2012

    • PubMed
    • Google Scholar
29. 1. Jakawich SK
    2. Nasser HB
    3. Strong MJ
    4. McCartney AJ
    5. Perez AS
    6. Rakesh N
    7. Carruthers CJ
    8. Sutton MA

    (2010) Local presynaptic activity gates homeostatic changes in presynaptic function driven by dendritic BDNF synthesis

    Neuron 68:1143–1158.

    https://doi.org/10.1016/j.neuron.2010.11.034

    • PubMed
    • Google Scholar
30. 1. Kaeser PS
    2. Regehr WG

    (2014) Molecular mechanisms for synchronous, asynchronous, and spontaneous neurotransmitter release

    Annual Review of Physiology 76:333–363.

    https://doi.org/10.1146/annurev-physiol-021113-170338

    • PubMed
    • Google Scholar
31. 1. Kim SH
    2. Ryan TA

    (2010) CDK5 serves as a major control point in neurotransmitter release

    Neuron 67:797–809.

    https://doi.org/10.1016/j.neuron.2010.08.003

    • PubMed
    • Google Scholar
32. 1. Koushika SP
    2. Richmond JE
    3. Hadwiger G
    4. Weimer RM
    5. Jorgensen EM
    6. Nonet ML

    (2001) A post-docking role for active zone protein Rim

    Nature Neuroscience 4:997–1005.

    https://doi.org/10.1038/nn732

    • PubMed
    • Google Scholar
33. 1. Lee JS
    2. Ho WK
    3. Neher E
    4. Lee SH

    (2013) Superpriming of synaptic vesicles after their recruitment to the readily releasable pool

    PNAS 110:15079–15084.

    https://doi.org/10.1073/pnas.1314427110

    • PubMed
    • Google Scholar
34. 1. Lin DM
    2. Goodman CS

    (1994) Ectopic and increased expression of Fasciclin II alters motoneuron growth cone guidance

    Neuron 13:507–523.

    https://doi.org/10.1016/0896-6273(94)90022-1

    • PubMed
    • Google Scholar
35. 1. Liu G
    2. Tsien RW

    (1995) Properties of synaptic transmission at single hippocampal synaptic boutons

    Nature 375:404–408.

    https://doi.org/10.1038/375404a0

    • PubMed
    • Google Scholar
36. 1. Liu KS
    2. Siebert M
    3. Mertel S
    4. Knoche E
    5. Wegener S
    6. Wichmann C
    7. Matkovic T
    8. Muhammad K
    9. Depner H
    10. Mettke C
    11. Bückers J
    12. Hell SW
    13. Müller M
    14. Davis GW
    15. Schmitz D
    16. Sigrist SJ

    (2011) RIM-binding protein, a central part of the active zone, is essential for neurotransmitter release

    Science 334:1565–1569.

    https://doi.org/10.1126/science.1212991

    • PubMed
    • Google Scholar
37. 1. Misura KM
    2. Scheller RH
    3. Weis WI

    (2000) Three-dimensional structure of the neuronal-Sec1-syntaxin 1a complex

    Nature 404:355–362.

    https://doi.org/10.1038/35006120

    • PubMed
    • Google Scholar
38. 1. Müller M
    2. Goutman JD
    3. Kochubey O
    4. Schneggenburger R

    (2010) Interaction between facilitation and depression at a large CNS synapse reveals mechanisms of short-term plasticity

    Journal of Neuroscience 30:2007–2016.

    https://doi.org/10.1523/JNEUROSCI.4378-09.2010

    • PubMed
    • Google Scholar
39. 1. Müller M
    2. Davis GW

    (2012a) Transsynaptic control of presynaptic Ca²⁺ influx achieves homeostatic potentiation of neurotransmitter release

    Current Biology 22:1102–1108.

    https://doi.org/10.1016/j.cub.2012.04.018

    • PubMed
    • Google Scholar
40. 1. Müller M
    2. Liu KS
    3. Sigrist SJ
    4. Davis GW

    (2012b) RIM controls homeostatic plasticity through modulation of the readily-releasable vesicle pool

    Journal of Neuroscience 32:16574–16585.

    https://doi.org/10.1523/JNEUROSCI.0981-12.2012

    • PubMed
    • Google Scholar
41. 1. Müller M
    2. Genç Ö
    3. Davis GW

    (2015) RIM-binding protein links synaptic homeostasis to the stabilization and replenishment of high release probability vesicles

    Neuron 85:1056–1069.

    https://doi.org/10.1016/j.neuron.2015.01.024

    • PubMed
    • Google Scholar
42. 1. Orr BO
    2. Fetter RD
    3. Davis GW

    (2017) Retrograde semaphorin-plexin signalling drives homeostatic synaptic plasticity

    Nature 550:109–113.

    https://doi.org/10.1038/nature24017

    • PubMed
    • Google Scholar
43. 1. Park S
    2. Bin NR
    3. Yu B
    4. Wong R
    5. Sitarska E
    6. Sugita K
    7. Ma K
    8. Xu J
    9. Tien CW
    10. Algouneh A
    11. Turlova E
    12. Wang S
    13. Siriya P
    14. Shahid W
    15. Kalia L
    16. Feng ZP
    17. Monnier PP
    18. Sun HS
    19. Zhen M
    20. Gao S
    21. Rizo J
    22. Sugita S

    (2017) UNC-18 and Tomosyn Antagonistically Control Synaptic Vesicle Priming Downstream of UNC-13 in Caenorhabditis elegans

    The Journal of Neuroscience 37:8797–8815.

    https://doi.org/10.1523/JNEUROSCI.0338-17.2017

    • PubMed
    • Google Scholar
44. 1. Patzke C
    2. Han Y
    3. Covy J
    4. Yi F
    5. Maxeiner S
    6. Wernig M
    7. Südhof TC

    (2015) Analysis of conditional heterozygous STXBP1 mutations in human neurons

    Journal of Clinical Investigation 125:3560–3571.

    https://doi.org/10.1172/JCI78612

    • PubMed
    • Google Scholar
45. 1. Pevsner J
    2. Hsu SC
    3. Scheller RH

    (1994) n-Sec1: a neural-specific syntaxin-binding protein

    PNAS 91:1445–1449.

    https://doi.org/10.1073/pnas.91.4.1445

    • PubMed
    • Google Scholar
46. 1. Plomp JJ
    2. van Kempen GT
    3. Molenaar PC

    (1992) Adaptation of quantal content to decreased postsynaptic sensitivity at single endplates in alpha-bungarotoxin-treated rats

    The Journal of Physiology 458:487–499.

    https://doi.org/10.1113/jphysiol.1992.sp019429

    • PubMed
    • Google Scholar
47. 1. Pouille F
    2. Scanziani M

    (2004) Routing of spike series by dynamic circuits in the hippocampus

    Nature 429:717–723.

    https://doi.org/10.1038/nature02615

    • PubMed
    • Google Scholar
48. 1. Riento K
    2. Kauppi M
    3. Keranen S
    4. Olkkonen VM

    (2000) Munc18-2, a functional partner of syntaxin 3, controls apical membrane trafficking in epithelial cells

    Journal of Biological Chemistry 275:13476–13483.

    https://doi.org/10.1074/jbc.275.18.13476

    • PubMed
    • Google Scholar
49. 1. Schlüter OM
    2. Basu J
    3. Südhof TC
    4. Rosenmund C

    (2006) Rab3 superprimes synaptic vesicles for release: implications for short-term synaptic plasticity

    Journal of Neuroscience 26:1239–1246.

    https://doi.org/10.1523/JNEUROSCI.3553-05.2006

    • PubMed
    • Google Scholar
50. 1. Schneggenburger R
    2. Meyer AC
    3. Neher E

    (1999) Released fraction and total size of a pool of immediately available transmitter quanta at a calyx synapse

    Neuron 23:399–409.

    https://doi.org/10.1016/S0896-6273(00)80789-8

    • PubMed
    • Google Scholar
51. 1. Schulze KL
    2. Broadie K
    3. Perin MS
    4. Bellen HJ

    (1995) Genetic and electrophysiological studies of Drosophila syntaxin-1A demonstrate its role in nonneuronal secretion and neurotransmission

    Cell 80:311–320.

    https://doi.org/10.1016/0092-8674(95)90414-X

    • PubMed
    • Google Scholar
52. 1. Smyth AM
    2. Yang L
    3. Martin KJ
    4. Hamilton C
    5. Lu W
    6. Cousin MA
    7. Rickman C
    8. Duncan RR

    (2013) Munc18-1 protein molecules move between membrane molecular depots distinct from vesicle docking sites

    Journal of Biological Chemistry 288:5102–5113.

    https://doi.org/10.1074/jbc.M112.407585

    • PubMed
    • Google Scholar
53. 1. Sons MS
    2. Verhage M
    3. Plomp JJ

    (2003) Role of Munc18-1 in synaptic plasticity at the myasthenic neuromuscular junction

    Annals of the New York Academy of Sciences 998:404–406.

    https://doi.org/10.1196/annals.1254.052

    • PubMed
    • Google Scholar
54. 1. Südhof TC

    (2012a) Calcium control of neurotransmitter release

    Cold Spring Harbor Perspectives in Biology 4:a011353.

    https://doi.org/10.1101/cshperspect.a011353

    • PubMed
    • Google Scholar
55. 1. Südhof TC

    (2012b) The presynaptic active zone

    Neuron 75:11–25.

    https://doi.org/10.1016/j.neuron.2012.06.012

    • PubMed
    • Google Scholar
56. 1. Taschenberger H
    2. Woehler A
    3. Neher E

    (2016) Superpriming of synaptic vesicles as a common basis for intersynapse variability and modulation of synaptic strength

    PNAS 113:E4548–E4557.

    https://doi.org/10.1073/pnas.1606383113

    • PubMed
    • Google Scholar
57. 1. Thiagarajan TC
    2. Lindskog M
    3. Tsien RW

    (2005) Adaptation to synaptic inactivity in hippocampal neurons

    Neuron 47:725–737.

    https://doi.org/10.1016/j.neuron.2005.06.037

    • PubMed
    • Google Scholar
58. 1. Toonen RF
    2. Verhage M

    (2003) Vesicle trafficking: pleasure and pain from SM genes

    Trends in Cell Biology 13:177–186.

    https://doi.org/10.1016/S0962-8924(03)00031-X

    • PubMed
    • Google Scholar
59. 1. Toonen RF
    2. Wierda K
    3. Sons MS
    4. de Wit H
    5. Cornelisse LN
    6. Brussaard A
    7. Plomp JJ
    8. Verhage M

    (2006) Munc18-1 expression levels control synapse recovery by regulating readily releasable pool size

    PNAS 103:18332–18337.

    https://doi.org/10.1073/pnas.0608507103

    • PubMed
    • Google Scholar
60. 1. Toonen RF
    2. Verhage M

    (2007) Munc18-1 in secretion: lonely Munc joins SNARE team and takes control

    Trends in Neurosciences 30:564–572.

    https://doi.org/10.1016/j.tins.2007.08.008

    • PubMed
    • Google Scholar
61. 1. Verhage M
    2. Maia AS
    3. Plomp JJ
    4. Brussaard AB
    5. Heeroma JH
    6. Vermeer H
    7. Toonen RF
    8. Hammer RE
    9. van den Berg TK
    10. Missler M
    11. Geuze HJ
    12. Südhof TC

    (2000) Synaptic assembly of the brain in the absence of neurotransmitter secretion

    Science 287:864–869.

    https://doi.org/10.1126/science.287.5454.864

    • PubMed
    • Google Scholar
62. 1. Voets T
    2. Toonen RF
    3. Brian EC
    4. de Wit H
    5. Moser T
    6. Rettig J
    7. Südhof TC
    8. Neher E
    9. Verhage M

    (2001) Munc18-1 promotes large dense-core vesicle docking

    Neuron 31:581–592.

    https://doi.org/10.1016/S0896-6273(01)00391-9

    • PubMed
    • Google Scholar
63. Book

    1. Von der Malsburg C

    (1986)

    Disordered Systems and Biological Organization

    In: Bienenstock E, editors. Statistical Coding and Short-Term Synaptic Plasticity: A Scheme for Knowledge Representation in the Brain, 20. Springer. pp. 247–272.

    • Google Scholar
64. 1. Wang Y
    2. Sugita S
    3. Sudhof TC

    (2000) The RIM/NIM family of neuronal C2 domain proteins. Interactions with Rab3 and a new class of Src homology 3 domain proteins

    The Journal of biological chemistry 275:20033–20044.

    https://doi.org/10.1074/jbc.M909008199

    • PubMed
    • Google Scholar
65. 1. Wang HL
    2. Zhang Z
    3. Hintze M
    4. Chen L

    (2011) Decrease in calcium concentration triggers neuronal retinoic acid synthesis during homeostatic synaptic plasticity

    Journal of Neuroscience 31:17764–17771.

    https://doi.org/10.1523/JNEUROSCI.3964-11.2011

    • PubMed
    • Google Scholar
66. 1. Wang T
    2. Jones RT
    3. Whippen JM
    4. Davis GW

    (2016) α2δ-3 Is Required for Rapid Transsynaptic Homeostatic Signaling

    Cell Reports 16:2875–2888.

    https://doi.org/10.1016/j.celrep.2016.08.030

    • PubMed
    • Google Scholar
67. 1. Wang X
    2. Rich MM

    (2018) Homeostatic synaptic plasticity at the neuromuscular junction in myasthenia gravis

    Annals of the New York Academy of Sciences 1412:170–177.

    https://doi.org/10.1111/nyas.13472

    • PubMed
    • Google Scholar
68. 1. Weimer RM
    2. Richmond JE
    3. Davis WS
    4. Hadwiger G
    5. Nonet ML
    6. Jorgensen EM

    (2003) Defects in synaptic vesicle docking in unc-18 mutants

    Nature Neuroscience 6:1023–1030.

    https://doi.org/10.1038/nn1118

    • PubMed
    • Google Scholar
69. 1. Weyhersmüller A
    2. Hallermann S
    3. Wagner N
    4. Eilers J

    (2011) Rapid active zone remodeling during synaptic plasticity

    Journal of Neuroscience 31:6041–6052.

    https://doi.org/10.1523/JNEUROSCI.6698-10.2011

    • PubMed
    • Google Scholar
70. 1. Wu MN
    2. Littleton JT
    3. Bhat MA
    4. Prokop A
    5. Bellen HJ

    (1998) ROP, the Drosophila Sec1 homolog, interacts with syntaxin and regulates neurotransmitter release in a dosage-dependent manner

    The EMBO Journal 17:127–139.

    https://doi.org/10.1093/emboj/17.1.127

    • PubMed
    • Google Scholar
71. 1. Younger MA
    2. Müller M
    3. Tong A
    4. Pym EC
    5. Davis GW

    (2013) A presynaptic ENaC channel drives homeostatic plasticity

    Neuron 79:1183–1196.

    https://doi.org/10.1016/j.neuron.2013.06.048

    • PubMed
    • Google Scholar
72. 1. Zhao C
    2. Dreosti E
    3. Lagnado L

    (2011) Homeostatic synaptic plasticity through changes in presynaptic calcium influx

    Journal of Neuroscience 31:7492–7496.

    https://doi.org/10.1523/JNEUROSCI.6636-10.2011

    • PubMed
    • Google Scholar
73. 1. Zucker RS
    2. Regehr WG

    (2002) Short-term synaptic plasticity

    Annual Review of Physiology 64:355–405.

    https://doi.org/10.1146/annurev.physiol.64.092501.114547

    • PubMed
    • Google Scholar

Author details

1. Jennifer M Ortega

   Department of Biochemistry and Biophysics, Kavli Institute for Fundamental Neuroscience, University of California San Francisco, San Francisco, California

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

2. Özgür Genç

   Department of Biochemistry and Biophysics, Kavli Institute for Fundamental Neuroscience, University of California San Francisco, San Francisco, California

   Contribution

   Data curation, Formal analysis, Writing—review and editing

   Competing interests

   No competing interests declared

3. Graeme W Davis

   Department of Biochemistry and Biophysics, Kavli Institute for Fundamental Neuroscience, University of California San Francisco, San Francisco, California

   Contribution

   Conceptualization, Funding acquisition, Methodology, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   graeme.davis@ucsf.edu

   Competing interests

   Reviewing editor, eLife

   "This ORCID iD identifies the author of this article:" 0000-0003-1355-8401

• 2,772

  views

• 455

  downloads

• 33

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.