Suche  |   Sitemap  |   Impressum  |   Kontakt  |    deutsch   
Max-Planck-Institut für Experimentelle Medizin

2.         Presynaptic Function and the Regulation of Neurotransmitter Release


In contrast to most other secretory processes, synaptic neurotransmitter release is restricted to designated release sites. These so called active zones are electron dense regions of the presynaptic plasma membrane at which the final steps of synaptic vesicle exocytosis take place with extreme spatial and temporal accuracy. Typically, a presynaptic bouton contains hundreds of vesicles that cluster in close proximity of the active zone. A subpopulation of vesicles in this cluster is tethered specifically to the plasma membrane at the active zone. These tethered vesicles have to mature to a fusion-competent, primed state before an increase of the intracellular Ca2+ concentration can trigger their fusion. Usually, only a fraction of tethered vesicles in any given synapse is primed to fusion competence (Wojcik and Brose, 2007) (Figure 2). The size of this readily releasable vesicle pool determines synaptic release probability and signaling capacity. The second main focus of our research concerns the molecular mechanisms of neurotransmitter release, with an emphasis on active zone function.

Figure 2: A Molecular Model of SNARE Regulatory Processes in Synaptic Vesicle Fusion.
On the plasma membrane, the SNARE proteins Syntaxin-1 and SNAP-25 may form a heterodimeric docking platform (A). The Syntaxin-1/SNAP-25 heterodimers act as acceptors for vesicular Synaptobrevin-2 in the formation of the SNARE complex (B). Munc13s, CAPSs, Munc18s, and RIMs are likely candidates in enabling SNARE complex formation, which in addition may be regulated by Ca2+. Complexin binds the fully assembled SNARE complex and may stabilize it in a highly fusogenic (and, as indicated, possibly even hemifused) state (C). In this state, the hemifused complex may be clamped and thereby prevented from complete fusion. Upon arrival of an action potential, Ca2+ influx, and binding of Ca2+ to the C2A and C2B domains of Synaptotagmin-1, Synaptotagmin-1 may displace Complexin from the SNARE complex, initailly leading to the formation of a fusion pore (D), and subsequently to full fusion and highly synchronous transmitter release (E). Deformation of the membranes induced by partial insertion of the Synaptotagmin-1 C2 domains into the phospholipid layers and induced by electrostatic changes in the C2B domain, which can interact simultaneously with the vesicular and the plasma membrane, may be an important factor in triggering membrane fusion. In the absence of Complexin, SNARE complexes may fully assemble in a spontaneous fashion and thus cause spontaneous fusion pore opening (C') and fusion (E'). In the presence of Ca2+, this latter process may be responsible for asynchronous transmitter release (Wojcik and Brose, 2007; Brose, 2008b).

Munc13 Proteins


Substantial research efforts with respect to presynaptic acive zone function continue to focus on Munc13s, which we showed to be essential synaptic vesicle priming proteins of active zones (Augustin et al., 1999; Brose et al., 2000; Varoqueaux et al., 2002; Rosenmund et al., 2003) (Figure 2). They are regulated by diacylglycerol (DAG) and Ca2+/Calmodulin and thereby determine synaptic short-term plasticity (Rhee et al., 2002; Rosenmund et al., 2002; Junge et al., 2004).

A study performed in collaboration with the group of C. Rosenmund (Houston, USA) showed that diacylglycerol binding to Munc13-1 regulates the energy barrier for vesicle fusion (Basu et al., 2007). In a collaboration with the group of U. Ashery (Tel Aviv, Israel), Munc13 function in chromaffin cells was shown to also be regulated by Ca2+/Calmodulin (Zikich et al., 2008). We further characterized the Ca2+/Calmodulin binding to Munc13s using biochemical tools. Corresponding data indicate that all Munc13 isoforms can form a complex with Calmodulin already at low Ca2+ concentrations just above resting levels, underscoring the Ca2+ sensor/effector function of this interaction in short-term synaptic plasticity phenomena (Dimova et al., 2006).

After the discovery of the role of Munc13s in synaptic vesicle priming, we embarked on collaborative projects with T. Xu (Peking, China) and H. Gaysano (Toronto, Canada), who examined the role of Munc13s in insulin secretion. These studies showed that Munc13-1 is required for sustained release of insulin from pancreatic b-cells (Kwan et al., 2006; Kang et al., 2006; see also Stevens et al., 2005, for a corresponding study on chromaffin cells).

Apart from second messengers, Munc13-1 and ubMunc13-2 are also regulated by other active zone proteins, the most important one of which is RIM. We showed previously that Munc13-1 or ubMunc13-2 and RIM interact functionally (Betz et al., 2001). Disruption of this interaction causes a loss of fusion competent synaptic vesicles, creating a hypomorphic phenocopy of Munc13 deficient neurons. We postulated that the Munc13/RIM1 interaction may create a functional link between synaptic vesicle tethering and priming or regulate the priming reaction itself, thereby determining the number of fusion competent vesicles. In a recent study, we discovered that RIM binding regulates the synaptic anchoring of Munc13-1 and ubMunc13-2 (Andrews-Zwilling et al., 2006).

With respect to the analysis of the trafficking of Munc13 proteins to synapses, we generated of knock-in mutant mice that express from the respective endogenous genetic loci Munc13s tagged at their C-termini with different fluorescent proteins. In the case of the Munc13-1-EGFP knock-ins, we have completed and published the general phenotypic analysis. In collaboration with N. Ziv (Haifa, Israel) and C. Rosenmund (Houston, USA), we found that the tagged Munc13-1-EGFP behaves exactly like its wild type counterpart. Using neurons from the knock-in mice, we found in FRAP experiments in vivo that the total active zone pool of Munc13-1 in presynaptic active zones turns over within 50 min (Kalla et al., 2006). We will further use these mice to study Munc13 trafficking, active zone generation, and protein turnover at synapses under control and plasticity conditions (e.g. during mossy fiber LTP) in vivo. These types of knock-ins are very useful tools in the field of presynaptic physiology and will support our research at multiple levels.

Direct fluorescence of Munc13-1-EGFP in a sagittal brain section from a knock-in mutant mouse (Kalla et al., 2006).

CAPS Proteins


CAPS proteins (CAPS-1 and CAPS-2) are distant relatives of Munc13s and were thought to be specifically and selectively involved in large dense-core vesicle exocytosis. To our surprise, we found in collaboration with the group of J.-S. Rhee in our institue that CAPS-1 and CAPS-2 are essential components of the synaptic vesicle priming machinery, and not specifically involved in large dense-core vesicle exocytosis. CAPS-deficient neurons contain no or very few fusion competent synaptic vesicles, which causes a selective impairment of fast phasic transmitter release. Increases in the intracellular Ca2+ levels can transiently revert this defect. Our findings demonstrate that CAPS proteins generate and maintain a highly fusion competent synaptic vesicle pool that supports phasic Ca2+ triggered release of transmitters (Jockusch et al., 2007) (Figure 2).

Concerning the role of CAPS-1 and CAPS-2 in dense-core vesicle release, we studied insulin secretion in CAPS2-/- and CAPS1+/--CAPS2-/- mice in collaboration with the group of P. Rorsmann (Oxford, UK, and Lund, Sweden). We found that these mutants, despite having increased insulin sensitivity, are glucose intolerant, and that this effect is attributable to a marked reduction of glucose-induced insulin secretion. This correlates with diminished Ca2+-dependent exocytosis, a reduction in the size of the morphologically docked vesicle pool, a decrease in the readily releasable pool of secretory vesicles, slowed granule priming, and suppression of second-phase (but not first phase) insulin secretion. In b cells of CAPS1+/--CAPS2-/- mice, the lowered insulin content and granule numbers were associated with an increase in lysosome numbers and lysosomal enzyme activity. We conclude that although CAPS proteins are not required for Ca2+-dependent dense-core vesicle exocytosis to proceed, they exert a modulatory effect on insulin granule priming, exocytosis, and stability (Speidel et al., 2008; see also Speidel et al., 2005).

CAPS1 (green) and GAD65 (red) in a glomerulus of the cerebellar granule cell layer (Jockusch et al., 2007).



Ca2+-triggered synaptic vesicle exocytosis is the fastest and most tightly regulated membrane fusion reaction known. It is executed by the SNARE complex, which is formed by the vesicle protein Synaptobrevin-2 and the plasma membrane proteins Syntaxin-1 and SNAP-25. The molecular basis of the exquisite regulation of synaptic vesicle fusion is still only partially understood. In the past, numerous studies led to the discovery of proteins, which are involved in the regulation of the synaptic vesicle fusion process (Reim et al., 2001; Brose 2008a and 2008b). Among them are the Complexins, which are small charged proteins that bind very rapidly and with high affinity to the SNARE complex. So far, four Complexin isoforms have been identified. Interestingtly, they show different expression patterns. Whereas Complexins-I, -II and -III are expressed in brain and retina, Complexin-IV is only present at retinal ribbon synapses (Reim et al., 2005).

Complexins in the retina (Reim et al., 2005).

To stringently define the role of Complexins in exocytosis of conventional synapses in the mammalian central nervous system, we genetically eliminated all Complexins expressed in mouse brain by generating Complexin-I/II/III triple knock-out mice. We examined cultured hippocampal glutamatergic and GABAergic neurons as well as acute brainstem slices, and found that in the tested synapses removal of Complexins reduces fast Ca2+-triggered release without enhancing spontaneous release. In contrast to Drosophila neuromuscular junctions, where Complexin appears to act as a clamp and inhibits spontaneous release, our findings provide compelling evidence that Complexins are positive facilitators of the synaptic vesicle fusion machinery (Xue et al., 2008; see also Reim et al., 2001) (Figure 2).

Additional studies were performed on the role of SNARE binding in Complexin function. Using cosedimetation assays and electrophysiological rescue experiments we showed that SNARE complex binding of Complexin I via its central a-helix is necessary but unexpectedly not sufficient for its key function in promoting neurotransmitter release. An accessory a-helix N-terminal of the SNARE complex binding region plays an inhibitory role in fast synaptic exocytosis, while its N-terminally adjacent sequences facilitate Ca2+-triggered release even in the absence of the Ca2+ sensor Synaptotagmin-1. Our results indicate that distinct functional domains of Complexins differentially regulate synaptic exocytosis, and that via the interplay between these domains Complexins play a crucial role in fine-tuning Ca2+-triggered fast neurotransmitter release (Xue et al., 2007).

© 2006, Max-Planck-Gesellschaft     realized by vokativ