Keywords

1 Introduction

Tomosyn is a syntaxin-1-binding protein that we originally identified (Fujita et al. 1998). Tomosyn means tomo (friend in Japanese) of syntaxin-1. Tomosyn contains N-terminal WD40 repeats, a tail domain, and a C-terminal domain homologous to VAMP2. The C-terminal VAMP-like domain is responsible for binding to syntaxin-1 (Fujita et al. 1998) (Fig. 5.1a). Tomosyn belongs to the Lgl (lethal giant larvae) family that is conserved from yeast to human (Kagami et al. 1998; Hattendorf et al. 2007; Ashery et al. 2009) (Fig. 5.1b). The Lgl family is characterized by N-terminal fourteen WD40 repeats constituting two β-propeller structures (seven WD40 repeats constitute one β-propeller structure) (Hattendorf et al. 2007) and composed of Lgl, tomosyn, Sro7, and Sro77. Among the family members, only Lgl does not have a tail domain and a C-terminal domain homologous to VAMP2. Lgl, which was firstly identified as a tumor suppressor gene in Drosophila, has been shown to be involved in polarity formation and cell-cell adhesion in epithelial cells (Wirtz-Peitz and Knoblich 2006; Yamanaka and Ohno 2008). Sro7 and Sro77, yeast orthologues of tomosyn, have been shown to regulate exocytosis in yeast (Lehman et al. 1999). Lgl, tomosyn, and Sro7 directly interact with SNARE proteins (Fujita et al. 1998; Lehman et al. 1999; Müsch et al. 2002; Gangar et al. 2005), suggesting that the Lgl family plays roles in the SNARE-dependent vesicle trafficking. However, the common mode of action among the Lgl family members remains elucidated. Among the Lgl family members, the origins of tomosyn and Sro7 are evolutionally old (Kloepper et al. 2008). During the evolution from prokaryotes to metazoan, Lgl is thought to be generated by gene duplication of tomosyn (Kloepper et al. 2008). Therefore, elucidating modes of action for tomosyn and Sro7 is important for understanding the function of the Lgl family. In this chapter we describe, based on our findings, roles of tomosyn in the synaptic vesicle fusion phase. In the vertebrate nervous system, tomosyn-1 is expressed dominantly in the whole brain, while tomosyn-2 is expressed in the restricted area of the brain (Groffen et al. 2005). From here, we refer to tomosyn-1 as tomosyn.

Fig. 5.1
figure 1

Schematic representation of tomosyn. (a) Tomosyn is a syntaxin-1-binding protein. Tomosyn is composed of N-terminal WD40 repeats, a tail domain, and a C-terminal VAMP-like domain. Tomosyn binds to syntaxin-1 through the C-terminal VAMP-like domain. (b) Tomosyn belongs to the Lgl family. The Lgl family is characterized by the N-terminal WD40 repeats that constitute two β-propeller structures. Sro7 and Sro77 are yeast orthologues of tomosyn. Among the family members, only Lgl does not have the tail domain and the C-terminal VAMP-like domain. Phosphorylation sites are indicated with P

Full size image

2 SNARE-Dependent Vesicle Fusion Machinery

Synaptic vesicles are transported to the active zone in the presynaptic plasma membrane where Ca2+ channels are located. Depolarization induces Ca2+ influx into the cytosol of nerve terminals through the Ca2+ channels, and this Ca2+ influx initiates the fusion of the vesicles with the plasma membrane, finally leading to exocytosis of neurotransmitters (Südhof 2004). Soluble N-ethylmaleimide-sensitive fusion protein (NSF) attachment protein (SNAP) receptors (SNAREs) are essential for the synaptic vesicle exocytosis (Sutton et al. 1998; Weber et al. 1998; Jahn and Scheller 2006; Rizo and Rosenmund 2008). Synaptic vesicles are endowed with vesicle-associated membrane protein 2 (VAMP2) as a vesicular SNARE (v-SNARE), whereas the presynaptic plasma membrane is endowed with syntaxin-1 and SNAP-25 as target SNAREs (t-SNAREs) (Fig. 5.2a). VAMP2 interacts with SNAP-25 and syntaxin-1 to form a stable SNARE complex (Trimble et al. 1988; Bennett et al. 1992; Söllner et al. 1993; Chen and Scheller 2001). The formation of the SNARE complex then brings synaptic vesicles and the plasma membrane into close apposition and provides the energy that drives the mixing of the two lipid bilayers (Weber et al. 1998; Chen and Scheller 2001; Jahn and Scheller 2006; Rizo and Rosenmund 2008). Although the previous study using the SNARE complex reconstituted on liposomes demonstrated that the SNARE complex was sufficient for membrane fusion, the fusion kinetics was very slow (Weber et al. 1998). These facts indicate in vivo existence of a special factor(s) to regulate SNARE assembly for fast membrane fusion characteristic of neurotransmitter release.

Fig. 5.2
figure 2

Regulation of the SNARE complex formation by tomosyn. (a) Synaptic vesicle fusion driven by SNARE complexes. Syntaxin-1, SNAP-25, and VAMP2 form a ternary SNARE complex, allowing the synaptic vesicle to dock at the presynaptic membrane. Subsequently, the SNARE complexes catalyze membrane fusion in a zipper-like fashion. (b) Inhibition of the SNARE complex formation by tomosyn. Tomosyn binds to syntaxin-1 and SNAP-25 through the C-terminal VAMP-like domain (VLD), thereby forming a stable tomosyn-SNARE complex. The tomosyn-SNARE complex formation inhibits the SNARE complex formation by sequestering the t-SNAREs, leading to inhibition of membrane fusion

Full size image

3 Tomosyn Regulates SNARE Complex Formation Through the C-Terminal VAMP-Like Domain

Tomosyn interacts with SNAP-25 and syntaxin-1 to form a stable tomosyn-SNARE complex through the C-terminal VAMP-like domain (Fujita et al. 1998) (Fig. 5.2b). The tomosyn-SNARE complex formation sequesters t-SNAREs on the presynaptic plasma membrane, leading to inhibition of the SNARE complex formation (Fujita et al. 1998; Hatsuzawa et al. 2003; Pobbati et al. 2004; Sakisaka et al. 2008; Yamamoto et al. 2009; Ashery et al. 2009). Consistent with the inhibitory activity on the SNARE complex formation, the genetic ablation of tomosyn in mice and C. elegans leads to enhancement of neurotransmitter release (Sakisaka et al. 2008; Gracheva et al. 2006; McEwen et al. 2006), and the overexpression in superior cervical ganglion (SCG) neurons inhibits neurotransmitter release induced by an action potential (Baba et al. 2005). Accumulating evidence suggests that tomosyn controls exocytotic efficacy of synaptic vesicles. Paired-pulse facilitation at mossy fiber synapses of hippocampi is decreased in the tomosyn-deficient mice (Sakisaka et al. 2008). In response to repetitive presynaptic action potentials, the tomosyn-overexpressing neurons show severe synaptic depression, in contrast to remarkable synaptic facilitation in control neurons (Baba et al. 2005). However, it is unclear how tomosyn controls the Ca2+-dependent exocytosis, since the C-terminal VAMP-like domain sequesters the t-SNAREs in a Ca2+-independent manner. We showed that tomosyn is directly phosphorylated by protein kinase A (PKA), which in turn reduces its interaction with syntaxin-1 and enhances the formation of the SNARE complex (Baba et al. 2005). In addition, Rho-associated serine/threonine kinase (ROCK) activated by Rho small G protein phosphorylates syntaxin-1, which in turn increases the affinity of syntaxin-1 for tomosyn and forms a stable complex with tomosyn, resulting in inhibition of the formation of the SNARE complex during neurite extension (Sakisaka et al. 2004). Thus, the inhibitory activity of the C-terminal VAMP-like domain is regulated via the well-known signal transduction pathways.

4 Tomosyn Regulates SNARE Complex Formation Through the N-Terminal WD40 Repeats

The N-terminal WD40 repeats of tomosyn are also responsible for potent inhibition of neurotransmitter release. It has been demonstrated that catecholamine secretion is potently inhibited in chromaffin cells by overexpressing the N-terminal WD40 repeats (Yizhar et al. 2007). Intriguingly, the inhibitory activity of the N-terminal WD40 repeats in the chromaffin cells depends on Ca2+ concentration (Yizhar et al. 2004, 2007). We have also demonstrated that acetylcholine release from SCG neurons in long-term culture is potently inhibited by microinjecting the tomosyn fragment encompassing the N-terminal WD40 repeats (Sakisaka et al. 2008). Similarly to the inhibitory activity of the N-terminal WD40 repeats in the chromaffin cells, the inhibitory activity of tomosyn in the SCG neurons is influenced by Ca2+ concentration (Baba et al. 2005). In C. elegans, tomosyn associates with synaptic vesicles through the N-terminal WD40 repeats (McEwen et al. 2006), raising the possibility that the N-terminal WD40 repeats may negatively regulate the function of synaptic vesicle in a Ca2+-dependent manner. While we have shown that tomosyn oligomerizes the SNARE complex through the N-terminal WD40 repeats (Sakisaka et al. 2008), this does not account for the Ca2+-dependent inhibitory activity of the N-terminal WD40 repeats since the oligomerization takes place in a Ca2+-independent manner. Therefore, the N-terminal WD40 repeats are expected to functionally interact with a Ca2+-responsive protein(s) involved in the regulation of the synaptic vesicle fusion.

Indeed, we have demonstrated that tomosyn directly binds to synaptotagmin-1, a synaptic vesicle protein with two C2 domains that both bind to Ca2+, through the N-terminal WD40 repeats in a Ca2+-dependent manner (Yamamoto et al. 2010a). Synaptotagmin-1 underlies Ca2+ responsiveness in the neurotransmitter release (Geppert et al. 1994; Sutton et al. 1995; Shao et al. 1998; Fukuda et al. 1999; Augustine 2001; Fernandez et al. 2001; Chapman 2008). Upon Ca2+ binding, synaptotagmin-1 induces positive curvature of the presynaptic membrane by inserting the hydrophobic loops in the C2 domains into the presynaptic membrane, thereby catalyzing fast synaptic vesicle fusion in cooperation with the SNARE complex (Martens et al. 2007; Stein et al. 2007; Xue et al. 2008; Hui et al. 2009) (Fig. 5.3a). Importantly, the Ca2+-dependent binding between tomosyn and synaptotagmin-1 impairs the synaptotagmin-1 catalysis (Yamamoto et al. 2010a) (Fig. 5.3b), indicating that tomosyn negatively regulates the synaptotagmin-1-mediated step of Ca2+-dependent neurotransmitter release through the N-terminal WD40 repeats. Furthermore, the Ca2+-dependent binding enhances the activity of the C-terminal VAMP-like domain of tomosyn to sequester t-SNAREs (Yamamoto et al. 2010a) (Fig. 5.3b). These findings raise an attractive possibility that the interplay between tomosyn and synaptotagmin-1 underlies the inhibitory control of Ca2+-dependent neurotransmitter release. In response to a rise in Ca2+ concentration, synaptotagmin-1 on the synaptic vesicle catches tomosyn and inactivates its own catalysis for membrane fusion. Simultaneously, the synaptotagmin-1-tomosyn complex enhances sequestering of t-SNAREs on the presynaptic membrane and blocks the SNARE assembly. Eventually, the Ca2+-dependent synaptotagmin-1-tomosyn-SNARE complex formation will ensure inactivation of the fusion machineries on both the donor and target membranes and thereby inhibit priming of the synaptic vesicles. Synaptotagmin-1, cooperating with tomosyn on synaptic vesicles, would act as an alternative Ca2+ sensor to negatively control exocytotic efficacy of the synaptic vesicles, sensing Ca2+ concentration change near Ca2+ channel clusters.

Fig. 5.3
figure 3

Regulation of the Ca2+-dependent synaptic vesicle fusion by tomosyn. (a) Synaptotagmin-1 is a Ca2+ sensor that cooperates with the SNARE complex toward membrane fusion. Upon Ca2+ influx, Ca2+ binds and activates synaptotagmin-1. Activated synaptotagmin-1 bends the presynaptic membrane, thereby facilitating the SNARE complex-driven membrane fusion. (b) Tomosyn acts as an alternative Ca2+ sensor that negatively regulates the synaptic vesicle fusion. Upon Ca2+ influx, tomosyn binds and inactivates Ca2+-bound synaptotagmin-1 through the N-terminal WD40 repeats. The synaptotagmin-1 binding enhances sequestration of t-SNAREs by the C-terminal VAMP-like domain (VLD) of tomosyn. Eventually, membrane fusion is potently inhibited in a Ca2+-dependent manner

Full size image

5 The Tail Domain Regulates the Activity of Tomosyn

We have demonstrated that the tail domain of tomosyn acts as a regulatory domain for the C-terminal VAMP-like domain (Yamamoto et al. 2009, 2010b). The tail domain can directly bind to either the C-terminal VAMP-like domain or the N-terminal WD40 repeats (Yamamoto et al. 2009). The binding of the tail domain to the C-terminal VAMP-like domain represses the activity of the C-terminal VAMP-like domain to inhibit the SNARE complex formation, which is restored by the binding of the tail domain to the N-terminal WD40 repeats (Yamamoto et al. 2009). Therefore, tomosyn will be in a state of equilibrium between two conformational states upon the tail domain binding. In one conformational state (state I), the tail domain binds to the N-terminal WD40 repeats, leading to exposure of the C-terminal VAMP-like domain. The exposed C-terminal VAMP-like domain efficiently forms the tomosyn-SNARE complex, resulting in the potent inhibition of the SNARE complex formation. In the other conformational state (state II), the tail domain masks the C-terminal VAMP-like domain and thereby blocks the tomosyn-SNARE complex formation (Yamamoto et al. 2009) or enables VAMP2 to displace tomosyn from the tomosyn-SNARE complex (Yamamoto et al. 2010b), resulting in loss of the inhibition of the SNARE complex formation. In support of the idea of equilibrium, full-length tomosyn moderately inhibited the SNARE-driven membrane fusion relative to the tomosyn fragment encompassing only the C-terminal VAMP-like domain in vitro (Yamamoto et al. 2010b), suggesting that full-length tomosyn exists in both states. Given that full-length tomosyn inhibited the SNARE-driven membrane fusion more than the tomosyn fragment encompassing both the tail domain and the C-terminal VAMP-like domain did in vitro (Yamamoto et al. 2010b), the state I may be a dominant state. What drives the conformational change from state I to state II? We previously reported that PKA phosphorylates tomosyn, resulting in reducing the binding of the C-terminal VAMP-like domain to syntaxin-1 (Baba et al. 2005). Therefore, PKA might be a possible regulator to drive the conformational change. The physiological meaning of the structural regulation of tomosyn upon the tail binding still remains elusive. We have generated tomosyn-deficient mice, characterized them electrophysiologically, and revealed that the tomosyn-deficient mice lacked short-term potentiation (Sakisaka et al. 2008). Therefore, the structural regulation of tomosyn may be important for the short-term memory. Further studies will be needed for understanding the regulation of the conformational change of tomosyn.

6 Tomosyn Regulates the Readily Releasable Pool Size

Evidence is accumulating that tomosyn regulates the readily releasable pool (RRP) size in response to repetitive presynaptic activity. Tomosyn-deficient mice show reduced paired-pulse facilitation in hippocampi (Sakisaka et al. 2008), suggesting that, without tomosyn, the first action potential depletes synaptic vesicles in the RRP. Tomosyn-overexpressing presynaptic SCG neurons evoke smaller excitatory postsynaptic potentials (EPSPs) but cannot respond to following repetitive action potentials, thus inducing severe synaptic depression (Baba et al. 2005). In addition, synchronization for repeated transmitter release was lost under tomosyn loss-function by point mutations in the N-terminal WD40 repeats (Baba et al. 2005). The SCG neuron in culture forms synapses with many varicosities wrapping the cell soma (Baba et al. 2005; Ma et al. 2009); therefore, the RRP size is relatively large, 84–180 synaptic vesicles (Ma et al. 2009; unpublished data), calculated from the depletion of the RRP with a train of high-frequency action potentials. The averaged EPSP amplitude is ≈20 mV (Baba et al. 2005; Ma et al. 2009), suggesting that the number of releasable vesicles in response to an action potential is well controlled. From the averaged EPSP integral, we estimate that neurotransmitters are released from ≈50 synaptic vesicles in the GFP-overexpressing or non-transfected SCG neurons (Baba et al. 2005; Ma et al. 2009). However, only 14 synaptic vesicles in tomosyn-overexpressing neurons (unpublished data) and 12–20 synaptic vesicles in the tomosyn mutant–overexpressing neurons (Baba et al. 2005) can be exocytosed in response to an action potential. Therefore, tomosyn is a key molecule to determine the size of the RRP.

The size of the RRP in the central nervous system is extremely small (1–2 % of the total number of vesicles) (Schikorski and Stevens 2001; Sakaba et al. 2002; Rizzoli and Betz 2004). The vast majority of the synaptic vesicles are reserved at the presynaptic nerve terminals despite Ca2+ influx (Südhof 2000; Rizzoli and Betz 2005), and a subset of them is accordingly mobilized to prevent depletion of the RRP (Südhof 2000; Harata et al. 2001; Rizzoli and Betz 2005), resulting in maintaining of the RRP size. As mentioned above, tomosyn is the molecule that inactivates synaptotagmin-1 in response to Ca2+ influx and thereby perturbs the SNARE machinery activation (Yamamoto et al. 2010a) (Figs. 5.2b and 5.3b). Interestingly, the presynaptic interplay between tomosyn and synaptotagmin-1 controls the EPSP shape in the falling phase, enabling the neurons to respond to high-frequency action potentials (unpublished data). This result raises the possibility that the RRP size, i.e., exocytotic efficacy of release-ready synaptic vesicles, might be determined by the interaction between tomosyn and synaptotagmin-1 during arrivals of consecutive neuronal signals to the presynaptic terminal. The Ca2+-dependent synaptotagmin-1-tomosyn-SNARE complex formation as depicted in Fig. 5.3b may ensure inactivation of the fusion machineries on both the donor and target membranes under high Ca2+ concentration accumulating with repetitive Ca2+ influxes to reserve the synaptic vesicles, leading to maintaining of the RRP size. By laser photolysis of caged calcium in a rat calyx of Held synapse, a rise in Ca2+ concentration to 1–2 μM readily evoked release (Bollmann et al. 2000; Felmy et al. 2003). Brief local Ca2+ rise to 10–25 μM is sufficient to achieve the amount and the kinetics of the physiological transmitter release (Schneggenburger and Neher 2005). An increase to >30 μM depleted the RRP in <0.5 ms (Bollmann et al. 2000). In addition, 40 μM is the peak concentration for synaptic vesicles at the release site during an action potential (half-width approximately 0.4 ms; Meinrenken et al. 2002). Therefore, in our scenario, the synaptotagmin-1-tomosyn-SNARE complex will be maximally formed in the high-range concentration (10–50 μM) of local Ca2+ rise that depletes the RRP (Meinrenken et al. 2002), in order to negatively control exocytotic efficacy of synaptic vesicle in a late phase of transmitter release preventing the RRP depletion. Severe synaptic depression induced by repetitive action potentials in tomosyn-overexpressing neurons (Baba et al. 2005) supports that the synaptotagmin-1-tomosyn-SNARE complex strongly retains low efficacy of synaptic vesicles for exocytosis in the accumulated Ca2+ rise. By contrast, our in vitro Ca2+ titration analysis shows that the synaptotagmin-1-tomosyn-SNARE complex formation is not saturated at 10–50 μM and proceeds more as the Ca2+ concentration increases (unpublished data). However, this result does not debate on the physiological relevance of the synaptotagmin-1-tomosyn-SNARE complex formation. The in vitro biochemical reactions do not reconstitute the local Ca2+ rise as seen in neurons. It has been reported that a specific membrane lipid, PI(4,5)P2, increases Ca2+ affinity of synaptotagmin-1 (Radhakrishnan et al. 2009). Therefore, to address biochemically the precise dependency on Ca2+ concentration for the synaptotagmin-1-tomosyn-SNARE complex formation, it will be required to develop more physiologically relevant assay reflecting the local Ca2+ rise and the membranous environment.

7 Conclusions and Perspectives

The crystal structure of the N-terminal WD40 repeats of Sro7, the yeast orthologue of tomosyn, has been solved (Hattendorf et al. 2007). Based on the solved structure, Sro7 is suggested to bind to Sec9, a yeast counterpart of SNAP-25, through the N-terminal WD40 repeats and thereby inhibits the SNARE complex formation. While Sec9 binds to Sro7 through both the N-terminal region and the SNARE motifs, the N-terminal region of Sec9 is not conserved in mammalian SNAP-25 (Hattendorf et al. 2007). As far as we examined, the N-terminal WD40 repeats of tomosyn had no inhibitory activity on the SNARE complex formation. Therefore, the inhibitory activity of the N-terminal WD40 repeats of Sro7 on the SNARE complex formation may not be evolutionally conserved. On the other hand, the binding of tomosyn to synaptotagmin-1 is in good agreement with the association of tomosyn with the synaptic vesicles in C. elegans (McEwen et al. 2006), raising a possibility that the activity of the N-terminal WD40 repeats to inhibit the synaptotagmin-1 function might be evolutionally conserved between nematodes and mammals. The N-terminal WD40 repeats also have the activity to oligomerize the SNARE complex (Sakisaka et al. 2008). However, it remains to be elucidated whether the inhibitory activity on the synaptotagmin-1 function and the oligomerization activity on the SNARE complex are mutually exclusive or compatible. Tomosyn adopts two conformational states upon reciprocal intramolecular bindings of the tail domain (Yamamoto et al. 2009). In one conformational state where the tail domain binds to the N-terminal WD40 repeats, tomosyn potently inhibits the SNARE complex formation through the C-terminal VAMP-like domain. In the other conformational state where the tail domain binds to the C-terminal VAMP-like domain, the inhibitory activity of the C-terminal VAMP-like domain is decreased. The binding of synaptotagmin-1 to the N-terminal WD40 repeats enhances the tomosyn-SNARE complex formation through the C-terminal VAMP-like domain (Yamamoto et al. 2010a). Therefore, synaptotagmin-1 binding may stabilize the former conformational state of tomosyn, leading to enhancement of the tomosyn-SNARE complex formation. Future structural studies of full-length tomosyn, the synaptotagmin-1-tomosyn complex, and the synaptotagmin-1-tomosyn-SNARE complex will be required to address these concerns.