PKC (protein kinase C) has been known for many years to modulate regulated exocytosis in a wide variety of cell types. In neurons and neuroendocrine cells, PKC regulates several different stages of the exocytotic process, suggesting that these multiple actions of PKC are mediated by phosphorylation of distinct protein targets. In recent years, a variety of exocytotic proteins have been identified as PKC substrates, the best characterized of which are SNAP-25 (25 kDa synaptosome-associated protein) and Munc18. In the present study, we review recent evidence suggesting that site-specific phosphorylation of SNAP-25 and Munc18 by PKC regulates distinct stages of exocytosis.
Exocytosis is the fusion of secretory vesicles with the plasma membrane . Constitutive exocytosis, where fusion is apparently unregulated, is used by all cells to deliver integral membrane proteins to the plasma membrane and for the secretion of various substances. In contrast, regulated exocytosis, where fusion is triggered by an intracellular signal, is characteristic of specialized ‘professional’ secretory cells that release material only on demand, such as neurons and endocrine and exocrine cells . In neurons, depolarization of the presynaptic plasma membrane by action potentials causes an influx of Ca2+ ions into the presynaptic terminal, which triggers the fusion of neurotransmitter-containing synaptic vesicles with the plasma membrane, thus allowing the released neurotransmitter to signal to the target cell . Although the notion of Ca2+ as the key intracellular trigger for neurotransmitter release was established around 40 years ago, the downstream mechanism responsible for membrane fusion was unknown until comparatively recently. Complementary genetic , biochemical  and molecular  approaches have identified the basic membrane fusion machinery and established that this is conserved from yeast to the human brain. The core of this conserved machinery is the SNARE (soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor) complex, which in neurons comprises VAMP (vesicle-associated membrane protein)/synaptobrevin, syntaxin 1 and SNAP-25 (25 kDa synaptosome-associated protein) . All three synaptic SNARE proteins are proteolytic substrates for the clostridial neurotoxins , which are potent inhibitors of neurotransmitter release, and genetic ablation of SNAREs in yeast, flies and mice similarly blocks exocytosis . Indeed, reconstitution experiments using artificial liposomes suggest that formation of the SNARE complex may itself be the driving force for bilayer fusion . Furthermore, the recognition that two additional conserved protein families, namely the Rabs and the SM (Sec1/Munc18) proteins, are also involved in multiple vesicular transport processes has firmly established the concept of a universal mechanism of membrane fusion.
While it is clear that regulated exocytosis utilizes the same basic machinery as other membrane trafficking events, the sophisticated control that defines regulated exocytosis requires additional complexity. One obvious example is the rapid triggering of neurotransmitter release by Ca2+, which requires the presence of one or more Ca2+-binding proteins capable of regulating the conserved fusion machinery. Several Ca2+-binding proteins have been proposed to play such a role, but a variety of evidence suggests that synaptotagmin is the major Ca2+-sensor for regulated exocytosis . Another characteristic of regulated exocytosis is its acute regulation by protein phosphorylation. Many studies over the past 20 years have shown that exocytosis is modulated by protein kinases in almost all regulated secretory cell types, including neurons and neuroendocrine cells [11–14]. Hence, if Ca2+ is the near-universal trigger for exocytosis, protein phosphorylation can be thought of as a ubiquitous regulator of exocytosis. Although a wide range of serine/threonine and tyrosine kinases have been implicated, only PKA (protein kinase A) and PKC (protein kinase C) have been shown to modulate exocytosis in almost all regulated secretory cell types examined [11–14]. In the present study, we focus on recent studies that have illuminated the molecular mechanisms by which PKC acts to regulate the exocytotic machinery; for a recent review of PKA action in regulated exocytosis, see .
PKC regulates multiple stages in the exocytotic process
PKC was initially purified as a kinase that required Ca2+, phosphatidylserine and diacylglycerol for optimal activity, but was soon recognized to be a family of isoenzymes. Individual PKC isoforms are classified based on their cofactor requirements as classical (α, β and γ; activated by Ca2+ and diacylglycerol), novel (δ, ϵ, η and θ; activated by diacylglycerol) or atypical (ζ and ι/λ; activated by neither Ca2+ nor diacylglycerol) . Most of the early studies suggesting a role for PKC in the modulation of regulated exocytosis were based on the stimulation of secretion by phorbol esters, which are potent diacylglycerol analogues. Some caution is warranted when interpreting such effects, however, since it is now clear that other phorbol ester receptors exist and that some – notably Munc13 – have the ability to stimulate exocytosis in a PKC-independent manner . Nevertheless, several different pieces of evidence suggest that, in various systems, phorbol esters do indeed act mainly via PKC to regulate exocytosis. For example, the enhancement of secretion seen in response to phorbol esters in permeabilized chromaffin cells is prevented by application of PKC inhibitory peptides . Similarly, the increased neurotransmitter release induced by phorbol esters in ciliary ganglion terminals  and retinal bipolar cells  is blocked by bisindolylmaleimide PKC inhibitors, which target the ATP-binding site of PKC and so should not affect non-kinase diacylglycerol-binding proteins such as Munc13. Furthermore, the observation that addition of purified PKC enzyme enhances exocytosis in permeabilized chromaffin and PC12 cells [21,22] provides direct evidence that PKC acts downstream of Ca2+ entry to modulate the activity of the exocytotic machinery.
In more recent years, higher resolution assays of exocytosis have revealed that PKC affects multiple stages in the exocytotic process. For example, phorbol ester treatment of adrenal chromaffin cells and hippocampal neurons increases both the size and rate of refilling of the readily releasable pool of vesicles [23–25]. In chromaffin cells, phorbol esters also greatly increase the size of a small pool of vesicles that is released in response to low levels of Ca2+ ; phorbol esters have been shown to increase the Ca2+ sensitivity of neurotransmitter release at the calyx of Held [27,28], suggesting an effect on the Ca2+ sensor itself. Finally, phorbol esters have been shown to increase the rate and overall extent of exocytosis in chromaffin cells, while at the same time reducing the quantal size of individual release events [29,30]. This latter effect on quantal size is associated with faster initial release kinetics and premature termination of release, consistent with a phorbol ester-induced switch towards ‘kiss and run’ exocytosis, as has been suggested to occur in synaptosomes . Importantly, these various effects are largely blocked by bisindolylmaleimides, indicating that they are PKC-mediated. It seems highly unlikely that phosphorylation of a single protein could modulate such distinct processes, and so it appears that this sophisticated regulation of exocytosis at multiple stages is mediated by phosphorylation of multiple exocytotic proteins. Recent work has begun to identify the exocytotic PKC targets that mediate these distinct effects, as described below.
PKC substrates in exocytosis
Dozens of proteins involved in regulated exocytosis have been identified over recent years and most of these contain predicted phosphorylation sites for PKC. As a first step towards identifying the functionally important phosphoproteins that mediate the effects of protein kinases on exocytosis, various laboratories (including ours) have screened synaptic exocytotic proteins for in vitro phosphorylation by purified kinases . Although this has revealed many in vitro substrates, relatively few of these have been confirmed to be phosphorylated in cellular preparations using metabolic labelling or phospho-specific antisera. Fewer still have been shown to be functionally altered by phosphorylation in terms of their biochemical characteristics and effect on exocytosis in cells. To date, the only PKC substrates that fulfil all these criteria are SNAP-25 and Munc18.
SNAP-25 contributes two helices to the SNARE complex, with syntaxin and VAMP contributing one helix each. It is widely believed that formation of this four-helix complex mediates vesicle docking/fusion at the plasma membrane. Phosphorylation of cellular SNAP-25 has been shown to occur in response to phorbol ester application and physiological stimuli in various cell types, including PC12 cells [32,33], hippocampal organotypic cultures , pancreatic β cells  and adrenal chromaffin cells . Early studies utilizing botulinum neurotoxin cleavage suggested that the likely phosphorylation site is Ser187 , and this was subsequently confirmed using a phospho-Ser187-specific antibody . Interestingly, despite the consistent observations of phorbol ester-induced phosphorylation of cellular SNAP-25 on Ser187, recombinant SNAP-25 is a poor substrate for PKC phosphorylation on this residue in vitro  (T.J. Craig and A. Morgan, unpublished work). Consequently, very little is known about the biochemical effects of PKC phosphorylation of SNAP-25. Indeed, the only published data on this issue indicate that treatment of PC12 cell extracts with purified PKC and ATP[S] (adenosine 5′-[γ-thio]-triphosphate) reduces the amount of syntaxin co-immunoprecipitating with SNAP-25 . Although this observation suggests that phosphorylation of Ser187 reduces the affinity of SNAP-25 for syntaxin, such a mechanism is not easy to reconcile with the stimulatory effects of phorbol esters on exocytosis. Further work is needed to definitively address the biochemical consequence of Ser187 phosphorylation on SNAP-25 function. One report has claimed that PKC can also phosphorylate SNAP-25 on Thr138 , which has been characterized as an in vitro and in vivo PKA phosphorylation site [37,39]. Again, biochemical studies have provided little mechanistic insight into how phosphorylation of this residue might affect SNAP-25 function, as stoichiometric phosphorylation of Thr138 has no obvious effect on the binding of recombinant SNAP-25 to SNAREs or synaptotagmin . Recently, SNAP-23, a non-neuronal homologue of SNAP-25, has been shown to be phosphorylated by PKC in vitro and also in platelets and mast cells [40,41]. In one study, the phosphorylation sites were identified as Ser23, Thr24 and Ser161 and mutation of Ser23/Thr24 to putatively phosphomimetic aspartate residues inhibited binding of syntaxin 4 . In contrast, another study reported the phosphorylation sites to be Ser95 and Ser120, and found that mutation of these residues to either alanine (unphosphorylatable) or aspartate (putative phosphomimetic) had little effect on binding to SNAREs .
Transfection of phosphorylation site mutants and subsequent analysis of exocytosis is a powerful approach to understand the cellular consequences of site-specific phosphorylation. This method has been applied to SNAP-25 in several different cell types. In both insulin-secreting cells  and hippocampal pyramidal neurons , neither alanine nor aspartate/glutamate mutants had any significant effect on exocytosis, arguing against a major role for SNAP-25 phosphorylation in the stimulation seen by phorbol esters in these cell types. It should be noted, however, that assays of overall exocytosis may not reveal specific actions of SNAP-25 phosphorylation on individual stages of the exocytotic process. Indeed, patch-clamp capacitance analysis of chromaffin cells expressing putative phosphomimetic Ser187→Glu/Asp SNAP-25 mutants revealed a stimulatory effect on the refilling of the readily releasable pool after stimulation, while unphosphorylatable Ser187→Ala/Cys mutants inhibited the rate of refilling . This suggests that one effect of PKC activation, namely the increased rate of releasable pool refilling, occurs via phosphorylation of Ser187 of SNAP-25. Such an effect would likely be missed in the study on hippocampal neurons, where single action potentials were used to release from the readily releasable pool and refilling kinetics are not thought to make a significant contribution . Indeed, phosphomimetic Ser187 mutants do not increase the size of the readily releasable pool in chromaffin cells either , indicating that the stimulatory action of PKC on the size of the releasable pool must be mediated by a different mechanism. This could potentially also occur via SNAP-25 phosphorylation on the Thr138 site, as mutation of this site affects the size of the releasable pools in chromaffin cells , although SNAP-25-independent mechanisms are equally likely. Transfection of both unphosphorylatable and putative phosphomimetic Ser95/Ser120 mutants of SNAP-23 has recently been shown to decrease overall exocytosis in mast cell lines, as assayed via release of co-transfected human growth hormone . One interpretation of this finding is that SNAP-23 phosphorylation/dephosphorylation affects post-fusion SNARE recycling . Alternatively, it may simply be that Ser95 and Ser120 are required for normal SNAP-23 function and that mutation of these residues partially disables the protein, resulting in the observed inhibition of exocytosis. Clearly, more work is needed to characterize fully the effect of SNAP-25/23 phosphorylation on the various stages of exocytosis, and in particular, to shed light on the underlying molecular mechanisms involved.
SM proteins, like the SNAREs, are involved in all intracellular membrane fusion events. Indeed, sec1 mutants in yeast are defective in constitutive exocytosis, unc-18 mutants in Caenorhabditis elegans exhibit severely impaired neurotransmission and Munc18-1 knockout mice display a complete block in synaptic vesicle exocytosis . SM proteins from various species bind tightly to syntaxin homologues involved in the same membrane fusion step, suggesting an important role for this interaction in SM protein function, although recent results suggest that syntaxin-independent effects may be equally important. The mammalian SM proteins involved in exocytosis are Munc18-1, -2 and -3 (or a, b and c). Munc18-1 and -3 have been shown to be phosphorylated in intact cells in response to secretory stimuli and phorbol ester treatment [30,44–46]. Munc18-1 is phosphorylated in vitro by PKC on Ser306 and Ser313 , but using phospho-specific antisera to these sites, only phosphorylation on Ser313 has so far been confirmed to occur in vivo . Ser313 is phosphorylated in both intact and permeabilized chromaffin cells upon stimulation with secretagogues, and also in cortical synaptosomes in response to depolarization or activation of metabotropic glutamate receptors . Phosphorylation of Munc18-1 by PKC reduces the binding affinity of Munc18-1 for syntaxin 1A , and mutation of both Ser306 and Ser313 to glutamate mimics this effect both in in vitro binding assays  and in cells . These studies provide direct evidence of the phosphomimetic nature of the mutations, and thereby validate the use of the mutant constructs in functional assays of exocytosis.
Expression of the phosphomimetic Munc18-1 Ser306/Ser313→Glu mutant was found to mimic one of the effects of phorbol esters in chromaffin cells as assayed by amperometry, namely the accelerated kinetics of catecholamine release and termination that results in reduced quantal size . Importantly, this effect of phorbol esters was abrogated in cells expressing the phosphomimetic construct, indicating that this effect is directly mediated via PKC phosphorylation of Munc18-1 on Ser313. Another effect of phorbol esters, an increase in the overall number of exocytotic events, was not mimicked by expressing phosphomimetic Munc18-1, suggesting that another PKC substrate must be involved in this effect of PKC. This could conceivably be SNAP-25, in view of the stimulatory effect of phosphomimetic Ser187 constructs on the slow phase of release in chromaffin cells . However, it is likely that PKC-induced rearrangements of the actin cytoskeleton are involved in such gross increases in exocytosis, with many actin binding/severing proteins and also 14-3-3 proteins being potential PKC effectors . Although it is clear that phosphorylation by PKC reduces the affinity of Munc18-1 for syntaxin, it is not clear whether it is solely the reduced syntaxin binding that is responsible for the effect on release kinetics and quantal size or whether altered binding to other proteins is involved. Since some functions of Munc18-1 appear to be independent of high-affinity syntaxin binding , further work is required to define fully the molecular mechanism by which PKC phosphorylation of Munc18 alters release kinetics and quantal size.
Conclusions and future perspectives
Studies in the last few years have begun to identify how phosphorylation of distinct proteins at defined sites contributes to the regulation of various stages of exocytosis by PKC. It now seems that the effect of PKC of increasing the rate of refilling of releasable pools is mediated by phosphorylation of SNAP-25 on Ser187, while the effect on release kinetics and quantal size is mediated by phosphorylation of Munc18-1 on Ser313 (and possibly also Ser306). Although some mechanistic information is available, it is still too early to propose detailed models of the molecular mechanisms by which phosphorylation of these residues leads to the observed changes to exocytosis. Further work in this regard is clearly needed, especially for SNAP-25. One major unresolved question is the extent to which these findings, based on studies of dense core granule exocytosis in chromaffin cells, apply to synaptic vesicle exocytosis. This is an important issue, since similar effects on recruitment into the readily releasable pool and on quantal size would be predicted to result in altered synaptic transmission and plasticity. Rigorous examination of the effects of the phosphomimetic constructs described above in neuronal cells using a range of stimulation methods is required to address this question. Finally, although some pieces of the jigsaw are fitting into place, others remain unclear. For example, the substrates involved in the PKC-mediated increased Ca2+ sensitivity of release in neurons and other cells remain unknown. Synaptotagmin, the likely Ca2+ sensor for exocytosis, is an attractive candidate here as it is an in vitro PKC substrate and has also been shown to be phosphorylated in cells. Various other exocytotic proteins (e.g. rabphilin) have been well characterized in terms of their phosphorylation by PKC, but no data are available on whether this has a functional effect on the various stages of exocytosis that are modulated by phorbol esters. Further use of the phosphomimetic mutant approach, coupled with biochemical analysis, should shed further light on the molecular mechanisms by which PKC regulates exocytosis.
Cellular Information Processing: A Focus Topic at BioScience2005, held at SECC Glasgow, U.K., 17–21 July 2005. Edited by F. Antoni (Edinburgh, U.K.), C. Cooper (Essex, U.K.), M. Cousin (Edinburgh, U.K.), A. Morgan (Liverpool, U.K.), M. Murphy (Cambridge, U.K.), S. Pyne (Strathclyde, U.K.) and M. Wakelam (Birmingham, U.K.).