Synaptic Vesicle Reuse and Its Implications

Transcription

1 Synaptic Vesicle Reuse and Its Implications EGE T. KAVALALI Center for Basic Neuroscience and Department of Physiology, The University of Texas Southwestern Medical Center, Dallas REVIEW Presynaptic nerve terminals are exquisite vesicle trafficking machines. Neurotransmission is sustained by constant recycling of a handful of vesicles. Therefore, the rate and the pathway of vesicle trafficking can critically determine synaptic efficacy during activity. However, it is yet unclear whether synaptic vesicle recycling becomes rate limiting on a rapid time scale during physiologically relevant forms of activity in the brain. Several forms of synaptic plasticity arise from persistent alterations in the dynamics of vesicle trafficking in presynaptic terminals. What makes presynaptic forms of plasticity particularly interesting is that they not only increase or decrease the amplitude of synaptic responses but also cause frequencydependent changes in neurotransmission. In this manner, plasticity can alter the information coding in neural circuits beyond simple scaling of synaptic responses. However, studying the synaptic vesicle cycle beyond exocytosis and endocytosis has been difficult. In the past decade, several methods have been developed to infer vesicles trajectory during their cycle in the synapse. Nevertheless, several questions remain. A better understanding of the role of synaptic vesicle trafficking in neurotransmission will require novel approaches that either combine existing methods or the development of new methods to trace vesicles during their cycle. Recent evidence suggests that various presynaptic proteins involved in the synaptic function and homeostasis are either mutated or altered in their expression in several neurological and psychiatric disorders. Therefore, elucidation of the mechanisms that underlie the synaptic vesicle cycle may reveal novel therapeutic targets for brain disorders. NEUROSCIENTIST 12(1):57 66, DOI: / KEY WORDS Neurotransmission, Synaptic vesicle recycling, FM1-43, Synaptophluorin, Endocytosis, Lysosomal storage disorders, Schizophrenia Properties and the Time Course of Synaptic Vesicle Reuse Our work on presynaptic function is supported by grants from the National Institute of Mental Health (MH and MH 68437). I thank Drs Ferenc Deák, Helmut Krämer, Xinran Liu, Lisa Monteggia, Yildirim Sara, Thomas Südhof, and Tuhin Virmani for numerous invaluable discussions. Address correspondence to: Ege T. Kavalali, PhD, Center for Basic Neuroscience, U.T. Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX ( Ege.Kavalali@ UTSouthwestern.edu). Chemical synaptic transmission is a fundamental physiological process that mediates communication between neurons in addition to neuronal control of all forms of muscle contraction. During synaptic transmission, the release of neurotransmitter substances is mediated by Ca 2+ -dependent fusion of neurotransmitter-laden vesicles with a specialized protein-rich region of the axonal plasma membrane called the active zone. Following fusion, synaptic vesicles are rapidly endocytosed and recycled for reuse during subsequent bouts of presynaptic action potentials (Ceccarelli and others 1973; Heuser and Reese 1973; Sudhof 1995). This vesicle recycling process is particularly critical for small synapses of the central nervous system, as these synapses contain few vesicles (~100) and only a small fraction of these vesicles are capable of activity-dependent exocytosis (Harata and others 2001) (Fig. 1). The most conservative estimate of the rate of this membrane turnover is in the order of 60 seconds. This rapidity makes presynaptic terminals exceptionally dynamic vesicle-trafficking machines compared to other secretory systems. Recent studies suggest even faster rates for reavailability of at least some of the recycling vesicles during activity (~1 second). These findings raise the possibility that regulation of the rate of vesicle turnover may have substantial impact on short-term synaptic plasticity on a rapid time scale. This article will provide an outline of recent methods and advances as well as emerging questions on this topic. For a more detailed account of the ongoing work on the molecular mechanisms underlying the synaptic vesicle cycle, readers should refer to recent excellent reviews (Galli and Haucke 2001; Murthy and De Camilli 2003; Sudhof 2004). Multiple Pathways for Synaptic Vesicle Recycling According to the classical picture of the synaptic vesicle cycle (Fig. 2), a rise in the intrasynaptic Ca 2+ level leads to vesicle fusion and neurotransmitter release at the active zone, after which vesicles completely collapse onto the plasma membrane (Heuser and Reese 1973; Südhof 1995; Cremona and De Camilli 1997). During this process, vesicle membrane components (lipids as well as proteins) are thought to intermix with their plasma membrane counterparts. Subsequently, clathrin, through its Volume 12, Number 1, 2006 THE NEUROSCIENTIST 57 Copyright 2006 Sage Publications ISSN

2 complete collapse of the vesicle and without intermixing of synaptic vesicle membrane components with the plasma membrane (Ceccarelli and others 1973; Fesce and others 1994; Henkel and Almers 1996). In central synapses, this hypothesis is supported by reports documenting the effects of hypertonic challenge and electrical stimulation by using fluorescent lipid tracers (i.e., FM dyes) (Klingauf and others 1998; Pyle and others 2000; Stevens and Williams 2000; Aravanis and others 2003) as well as fluorescently tagged synaptic vesicle proteins (i.e., synaptophluorin, see below) (Gandhi and Stevens 2003). The actual molecular mechanism of this pathway remains to be determined. There is evidence that this rapid component of endocytosis may operate independent of clathrin (Jockusch and others 2005) and may require the function of the synaptic vesicle protein synaptobrevin, which is also critical for rapid Ca 2+ - dependent fusion (Deak and others 2004). Fig. 1. Ultrastructural analysis of vesicular horseradish peroxidase (HRP) uptake in wild-type (WT) and synaptobrevin-2 knockout (KO) synaptic terminals. Representative synapses from WT and synaptobrevin-2 KO hippocampal cultures treated with 90 mm KCl containing solution for 90 seconds. HRP was present during only the stimulus (left panels) or for an additional two minutes (right panels). The fraction of HRP-positive synaptic vesicles increased substantially (compared to WT) when an additional two-minute delay was introduced to synaptobrevin-2 KO synapses. Note that the fraction of vesicles that are labeled with HRP is low under all conditions, suggesting that only a small fraction of vesicles recycle under the strong 90 mm K + / 90 s stimulation protocol (modified from Deak and others 2004). adaptor proteins, is recruited to the membrane (typically within the periphery of the active zone) and forms coated vesicles by reclustering vesicle membrane components, which eventually bud off the plasma membrane with the help of the GTPase dynamin. Besides the plasma membrane, endocytosis of clathrin-coated vesicles may also occur from membrane infoldings or endosomal cisternae, which form upon accumulation of fused synaptic vesicles (Koenig and Ikeda 1996; Takei and others 1996; Richards and others 2000). A V-type ATPase then lowers the ph in these vesicles, and subsequently the vesicles are refilled with neurotransmitter. The reacidification and neurotransmitter filling steps are thought to be rapid, and the entire process for exocytosed vesicles to be reavailable for release occurs within 40 and 90 seconds (Betz and Bewick 1992, 1993; Ryan and Smith 1995). Recent studies provide evidence of an additional fast pathway of vesicle recycling during which synaptic vesicles retain their identity and do not intermix with the plasma membrane or endosomal compartments (Klingauf and others 1998; Pyle and others 2000; Stevens and Williams 2000; Aravanis and others 2003; Gandhi and Stevens 2003). This pathway may have characteristics of the so-called kiss-and-run mechanism in which the fusion pore opens transiently without Optical Detection of Synaptic Vesicle Recycling In the past decade, significant advances in our understanding of the dynamics of the synaptic vesicle cycle were made possible by the availability of new optical probes. The first probes were the styryl dye FM1-43 and its analogs (Betz and others 1996). FM1-43 is a partially lipophilic dye, which increases fluorescence 100-fold when it partitions into membranes, and thus it is virtually invisible in aqueous solution. The chemical structure of FM1-43 makes it difficult to cross lipid membranes, thus limiting the fluorescence labeling to outer membranes and recycling vesicles. Using this methodology, one can detect activity-dependent vesicular dye uptake and release in synaptic connections in multiple preparations including central neuronal cultures. Studies using FM dyes revealed that activity-induced dye uptake and release operates with high fidelity. Specifically, the extent of fluorescent dye endocytosed during a particular stimulation matches the magnitude of fluorescence loss due to exocytosis in response to the same stimulation. This observation suggests that synaptic vesicles recycle without intermixing with endosomal compartments, and thus the fluorescence probes do not get diluted (Murthy and Stevens 1998). However, there is also strong evidence that synaptic vesicles traverse endosomal compartments during their lifetime, probably on a much slower time scale. Endosomal trafficking is critical for long-term maintenance of synaptic vesicle trafficking, and if vesicles skip this step in repeated rounds of cycling, they may become fusion incompetent by depletion of their protein components (Wucherpfennig and others 2003). Synaptophluorin (sph) is another fluorescent probe that is commonly used to monitor exocytosis and endocytosis of synaptic vesicles (Miesenbock and others 1998; Sankaranarayanan and Ryan 2000). Synaptophluorin is a fusion construct of the synaptic vesicle protein synaptobrevin (also called vesicle-associated membrane protein or VAMP ) with a ph-sensitive EGFP at its C-terminal (located in the synaptic vesicle lumen). Synaptic vesicle 58 THE NEUROSCIENTIST Synaptic Vesicle Reuse

3 Fig. 2. Multiple pathways for synaptic vesicle recycling. In the classical picture of the synaptic vesicle cycle, a rise in Ca 2+ level leads to vesicle fusion and neurotransmitter release at the active zone, after which vesicles completely collapse onto the plasma membrane. During this process, vesicle membrane components are thought to intermix with their plasma membrane counterparts. Subsequently, clathrin, through its adaptor proteins, is recruited to the membrane (typically at the periphery of the active zone) and forms coated vesicles by reclustering vesicle membrane components, which eventually bud off the plasma membrane with the help of the GTPase dynamin. Besides the plasma membrane, endocytosis of clathrin-coated vesicles may also occur from membrane infoldings or endosomal cisternae. This classical pathway is thought to operate within the 40- and 90-second time scale. During an alternative fast (within seconds) pathway of vesicle recycling, synaptic vesicles retain their identity and do not intermix with the plasma membrane or endosomal compartments. This pathway is also referred to as the kiss and run pathway in which the fusion pore opens transiently without complete collapse of the vesicle and without intermixing of synaptic vesicle membrane components with the plasma membrane. lumen normally has an acidic ph of approximately 5.5 at which sph fluorescence is quenched. When vesicles fuse, lumenal EGFP is exposed to the extracellular ph, which results in a marked increase in its fluorescence. Studies using sph have provided valuable information on the tight coupling between exocytosis and endocytosis. One advantage of the sph probe is the molecular specificity of the fluorescence signal because the signal originates from a synaptic vesicle protein as opposed to nonspecific lipid-based fluorescence of FM dyes. Proton diffusion from a synaptic vesicle is not expected to be limited, and the time course of vesicle reacidification is thought to be fast without imposing a significant delay in the fluorescence signal during sph endocytosis. Although FM dyes and sph are commonly used to measure endocytosis and exocytosis, there are advantages and disadvantages to both. FM dye measurements have been instrumental in tracing vesicles during their cycle in the synapse. Using uptake and release of FM dyes, one could reach conclusions on vesicle mixing with endosomal compartments (Murthy and Stevens 1998) or on repeated use of the same vesicles during activity (Aravanis and others 2003), which would be difficult to attain using probes like sph. FM dyes report lipid cycling; sph reports the fate of synaptic vesicle protein synaptobrevin upon fusion and endocytosis. Synaptophluorin is a genetically expressed probe, which facilitates more reproducible repeated measurements on the same set of synapses (as it does not require dye uptake). However, measurement of endocytosis with sph is confounded by two factors. First, lateral diffusion of vesicular sph following fusion is commonly observed (Sankaranarayanan and Ryan 2000; Li and Murthy 2001). It is unclear whether reuptake of sph truly reflects endocytosis of vesicles or reclustering and reinternalization of synaptobrevin. It is also unknown whether endocytosis of lipid and protein components of synaptic vesicles are distinct processes or whether they are coupled. Second, the dependence of endocytosis measurements on vesicle reacidification may delay the fluorescence signal after sph internalization. There are also concerns with using FM dyes. Typically, FM dye measurements suffer from nonspecificity. FM dyes stain the surface membrane indiscriminately; thus, they are less specific to synaptic vesicles. Moreover, changes in FM dye fluorescence typically occur slowly, in the order of seconds, making it difficult to detect events in the millisecond time scale, which more closely match the time course of neurotransmission. In addition, after FM dye uptake, preparations need to be washed with dye-free solution for a significant period to reduce background fluorescence, which limits the rapid detection of synaptic vesicle reavailability after endocytosis. In summary, sph provides a molecularly specific probe to monitor the coupling between exocytosis and endocytosis, whereas FM dyes are better at tagging vesicles during their cycling within the synapses. Thus, these two optical probes complement each other in monitoring distinct aspects of the synaptic vesicle cycle. The experiments using these optical tools consistently supported the premise that synaptic vesicle exocytosis and endocytosis are tightly coupled processes. This largely kinetic coupling is also backed up by recent molecular evidence that proteins critical for exocytosis, such as synaptotagmin and synaptobrevin, are also essential for triggering endocytosis (Poskanzer and oth- Volume 12, Number 1, 2006 THE NEUROSCIENTIST 59

4 ers 2003; Deak and others 2004; Nicholson-Tomishima and Ryan 2004). Interestingly, new evidence suggests that this tight coupling between exo- and endocytosis is also valid for the case of vesicles that fuse spontaneously in the absence of stimulation (Sun and others 2002; Sara and others 2005). However, the current stage of our understanding of the synaptic vesicle recycling process leaves several important questions unanswered. Can synaptic vesicle recycling occur fast (i.e., in the order of seconds or less)? Do synaptic vesicle membranes retain their identity during the cycle, as suggested by some recent work (Murthy and Stevens 1998; Aravanis and others 2003)? More specifically, do vesicle lipid and protein components mix the plasma membrane upon fusion? Alternatively, do synaptic vesicles retain their identities upon fusion, as recently shown for the exocytosis of pancreatic zymogen granules (Thorn and others 2004)? Experimental Paradigms to Measure Synaptic Vesicle Reuse Electrophysiological measurements provide the most reliable measure of synaptic transmission with millisecond time resolution. Electrophysiological methods can be employed in conjunction with genetic manipulations or pharmacological approaches to block neurotransmitter refilling after endocytosis to assess the impact of synaptic vesicle reuse on neurotransmission. However, these measurements are somewhat limited in their ability to monitor synaptic vesicle recycling, as postsynaptic responses are not sensitive to the origin of the vesicles that give rise to particular electrical responses. In electrophysiological experiments, a fusion event can be registered regardless of whether it originated from a recycled vesicle or a vesicle that has not fused in the previous rounds of exocytosis. Therefore, to monitor synaptic vesicle reuse, it is essential to utilize methods that are sensitive to the functional history of vesicles during activity. This can be achieved by optical methods employing FM dyes, fluorescently tagged synaptic vesicle proteins such as synaptobrevin, or antibodies specific to the lumenal domain of the synaptic vesicle protein synaptotagmin. The combination of these optical methods with electrophysiological techniques is quite valuable in providing a more complete analysis of synaptic transmission. Comparison of Neurotransmitter Release and FM Dye Release Kinetics To determine the reliance of neurotransmission on synaptic vesicle endocytosis and recycling, one needs to estimate the time point when exocytosed vesicles become reavailable for release. This parameter can be estimated from the kinetic difference between the rate of FM dye destaining and the time course of neurotransmitter release from a set of synapses as long as there is good correspondence between the sources of the optical and electrophysiological signals. This method has been successfully applied to study neuromuscular junction preparations where each junction can be visually and electrically isolated and then recorded (Betz and Bewick 1992, 1993). The rationale behind these experiments stems from previous observations that during stimulation, FM dyes (particularly FM2-10, fast departitioning FM dye) can be cleared out of a fused vesicle within a second by departitioning into solution (Klingauf and others 1998), or within milliseconds by lateral diffusion in the neuronal membrane (Zenisek and others 2002). Both of these time frames are faster than the rate of fusion pore closure and endocytic retrieval (Klingauf and others 1998; Sankaranarayanan and Ryan 2001). Therefore, recycled vesicles would not contain significant amounts of FM dye that could be detected by further destaining; in contrast, the same vesicles would be refilled with neurotransmitter following endocytosis, which could then give rise to further synaptic responses. This difference between the two reporters of vesicle mobilization should result in a deviation between the kinetics of FM dye destaining and neurotransmitter release at the time when recycled vesicles start to be reused. The application of this method to central synapse preparations such as hippocampal cultures has already yielded valuable information (Sara and others 2002). However, the estimation of the rate of synaptic vesicle reuse using this approach requires careful consideration of factors such as partial release of FM1-43 upon fusion, due to its slow departitioning rate (Klingauf and others 1998; Aravanis and others 2003; Richards and others 2005), which may confound the interpretation of the results. Optical Monitoring of Rapid Uptake and Release of FM Dyes Another common means to quantify synaptic vesicle reuse is by detecting release of FM dyes rapidly after their uptake into the synapse (Ryan and Smith 1995). In this experimental paradigm, FM dye is applied during and/or immediately after stimulation to label endocytosing vesicles. The dye-containing solution is then swiftly washed out to prevent further dye uptake. This dye application stains a number of vesicles depending on the stimulation strength as well as the duration of dye presence. Application of a second stimulation at gradually delayed time points (typically 0 to 120 seconds) causes dye loss due to the reavailability of labeled vesicles. However, the proximity of stimulations inducing dye uptake and release makes it difficult to image actual dye loss during the second stimulation because of the highfluorescence background originating from residual dye. This shortcoming is usually circumvented by monitoring the remaining detectable staining at a later time point (typically ~5 minutes) after complete dye wash out. The releasable fluorescence trapped in the synapse gradually decreases following stimulations applied at longer intervals or for longer durations after dye wash out. The time course of this decrease in fluorescence provides an estimate of the time constant of vesicular reuse, which is the time it takes for an endocytosed vesicle to become release-ready again. This is a very sensitive method to 60 THE NEUROSCIENTIST Synaptic Vesicle Reuse

5 detect vesicle reuse. As stated above, the major caveat is the background fluorescence; although usually not sufficient to restain vesicles during stimulation, it makes fast imaging nearly impossible. Because of this problem, the vesicle reuse kinetics is typically reconstructed from a population of synapses rather than from repeated measurements at individual presynaptic terminals. Detection of Synaptic Vesicle Trafficking Using Molecular Epitopes with the Vesicle Lumen Antibodies generated against the lumenal domain of the synaptic vesicle protein synaptotagmin-1 have been powerful tools to visualize vesicle trafficking at presynaptic terminals. These antibodies can specifically label endocytosing synaptic vesicles under the same conditions as FM dyes albeit with limited kinetic resolution (Matteoli and others 1992; Malgaroli and others 1995). Nevertheless, these antibodies have been instrumental in investigating the molecular composition of synaptic vesicles during recycling. In a recent study, we used an antibody against the lumenal domain of synaptotagmin-1 and tested whether spontaneously endocytosed synaptic vesicles recycle and then become reavailable for fusion and thus take up a second probe (Sara and others 2005). To test this possibility, we exposed cultures to the antibody for 15 minutes in the presence of tetrodotoxin (TTX). After wash out of the primary antibody (>15 minutes), we incubated cultures with the fluorescently conjugated secondary antibody (in the presence of TTX), which only recognizes the primary antibody once it is exposed to the extracellular medium. This experiment resulted in significant punctate immunolabeling of hippocampal cultures without permeabilization (Sara and others 2005). When we introduced the secondary antibody through permeabilization after the spontaneous uptake of the primary antibody, the amount and the pattern of fluorescence labeling was comparable to the one seen when the secondary antibody was introduced by uptake without permeabilization. To examine the amount of synaptotagmin-1 epitope available on the surface membrane at any given time during spontaneous vesicle recycling, we labeled cultures with the secondary antibody after fixation but without permeabilization. This maneuver resulted in substantially diminished fluorescence labeling, which may be partly due to nonspecific reactivity of the secondary antibody. This finding suggests that most of the fluorescence in earlier experiments originated from internalized synaptotagmin-1. The reavailability of the primary synaptotagmin-1 antibody for interaction with the fluorescent secondary after a 15-minute wash period indicates that synaptic vesicles are spontaneously recycled with minimal loss of their molecular identity. Experimental paradigms using molecular recognition of synaptic vesicle associated epitopes would be extremely informative if they could be modified to detect activity-dependent synaptic vesicle turnover on a rapid time scale. Such an experimental approach would help us determine if the vesicle protein composition is conserved or compromised under stimulation. It would also be greatly beneficial if new antibodies against the lumenal domains of other synaptic vesicle proteins were available. A more comprehensive picture of synaptic vesicle protein trafficking could be obtained using these tools, which are complementary to rapid fluorescent and/or electrophysiological measurements. Reacidification of Synaptic Vesicles Synaptophluorin-based measurements can also report if synaptic vesicles are reused on a rapid time scale. Recently, Murthy and colleagues designed an experiment using bafilomycin A to inhibit the vacuolar ATPase hence reacidification of vesicles upon endocytosis (Li and others 2005). This maneuver prevents protondependent requenching of sph fluorescence after reinternalization and restricts the change in sph fluorescence during stimulation to exocytosis. Therefore, vesicles that have undergone exocytosis and endocytosis cannot be requenched, whereas vesicles that did not recycle are quenched. Using this setting, they showed identical fluorescence changes in repeated rounds of sph fusion, suggesting that recently endocytosed vesicles (which are already brighter) are not immediately reused. Short-term Synaptic Plasticity and Synaptic Vesicle Reuse What is the physiological significance of synaptic vesicle trafficking? Is it solely necessary to maintain fusion competence of synaptic vesicles and structural homeostasis of synapses in the long term? Alternatively, can alterations in synaptic vesicle trafficking affect frequency dependence of synaptic responses during shortterm synaptic plasticity (in the order of seconds or less)? During sustained activity, most synapses show depression characterized by a rapid decrease in the postsynaptic responses down to a plateau level. Multiple mechanisms contribute to this rapid depression, including a decrease in vesicle fusion probability, inactivation of voltagegated Ca 2+ channels, or use-dependent inhibition of release machinery by presynaptic receptors. One of the underlying causes of this depression is thought to be a rapid reduction in the number of vesicles available for release in the synapse. The vesicles available for release can be supplied from two sources. One source is vesicle replenishment from a set of reserve vesicles. The second source is the reuse of vesicles that have undergone exocytosis and endocytosis. If the synaptic vesicle reuse is fast enough to replenish vesicles during a brief burst of action potentials, then it can play a substantial role in regulating the rate of synaptic depression. The most compelling evidence for the contribution of recycled vesicles to synaptic transmission during activity comes from the analysis of Drosophila and mouse mutants of proteins involved in endocytosis. Several studies using the Drosophila temperature-sensitive dynamin mutant shibire have provided strong support for a relationship between vesicle recycling and synaptic Volume 12, Number 1, 2006 THE NEUROSCIENTIST 61

6 release during sustained stimulation. Dynamin is a GTPase that serves to pinch vesicles off the membrane following recruitment of clathrin and its adaptor proteins onto vesicles fused with the plasma membrane. Using this system, Koenig and Ikeda showed that vesicle recycling can occur through two pathways, one at the active zone of the Drosophila neuromuscular junction and the other in its vicinity (Koenig and Ikeda 1996, 1999). The active zone and nonactive zone pathways were shown to populate different pools of vesicles at different time scales, with the active zone pathway repopulating the readily releasable pool (RRP) within 1 minute of moving from the nonpermissive (no dynamin function) to the permissive temperature (dynamin is functional), whereas reserve pool (RP) replenishment from the slower pathway occurred on the order of 25 minutes. Recent experiments in Drosophila neuromuscular junction compared the rate of synaptic depression between wild-type and shibire mutant flies. In this study, synapses from shibire mutant flies at nonpermissive temperatures rapidly depressed without a plateau phase in response to high-frequency stimulation. The kinetic difference between this rate of depression and the depression observed with functional dynamin revealed a recycling rate of one to two vesicles per second per active zone (Delgado and others 2000), a rate considerably faster than previous estimates and in line with estimations from hippocampal synapses (Sara and others 2002). The premise that synaptic vesicle recycling contributes to the maintenance of synaptic transmission on a rapid time scale is also supported by other molecular perturbations at the synapse. For instance, disruption of dynamin SH3 domain interactions (Shupliakov and others 1997); genetic impairment of synaptojanin 1, an abundant presynaptic molecule that functions as a polyphosphoinositide phosphatase (Cremona and others 1999; Luthi and others 2001); or expression of differentially spliced isoforms of synaptotagmin 7 (Virmani and others 2003) all led to frequency-dependent changes in the rate of shortterm synaptic depression. Therefore, molecular manipulations of the synaptic vesicle recycling machinery not only help us uncover vesicle trafficking mechanisms but also provide an extremely valuable setting to study the kinetics and physiological significance of synaptic vesicle reuse during synaptic activity. Synaptic Vesicle Pools To better visualize the pathways underlying synaptic vesicle recycling, it is important to pay particular attention to the functional heterogeneity of synaptic vesicles within a synapse. Vesicles in a CNS terminal can be divided into two pools (Fig. 3). The first pool contains a relatively small fraction (5% 10%) of vesicles close to release sites. These vesicles are generally thought to be release ready because they can be released by rapid uncaging of intrasynaptic Ca 2+ (Schneggenburger and others 1999), a 10-ms Ca 2+ -current pulse (Wu and Borst 1999), a brief high-frequency train of action potentials (Murthy and Stevens 1999), or hypertonic stimulation (Rosenmund and Stevens 1996). This release-ready pool of vesicles is referred to as the immediately releasable pool or the readily releasable pool (RRP). RRP vesicles are usually considered to be in a morphologically docked state, although not all morphologically docked vesicles are necessarily release competent at any given time (Schikorski and Stevens 2001). In addition to the morphological docking, a priming step is required to make vesicles fully release competent (Jahn and others 2003). A secondary pool of vesicles, the reserve pool (RP), is spatially distant from the release sites and constantly replaces the vesicles in the RRP that have been exocytosed. The rate of replenishment of RRP vesicles from the reserve pool is a critical parameter that determines the response of synapses to repetitive stimulation. The number of vesicles contained in the RRP is a critical parameter that regulates the probability of release, which is defined as the probability that a presynaptic action potential can result in an exocytotic event. Therefore, the rate and pathways by which vesicles in the RRP are replenished are also a crucial determinant of presynaptic efficacy and of several forms of short- and long-term synaptic plasticity. In addition, several lines of evidence support the presence of the nonrecycling pool of vesicles in the synapse. Mechanisms that can render this resting (or dormant ) pool functional remain to be determined (Sudhof 2000; Harata and others 2001). Within the synapse, neurotransmitter release is restricted to active zones. The presynaptic active zone is precisely aligned with the postsynaptic neurotransmitter receptors and the postsynaptic density. The fusion of synaptic vesicles with these electron-dense regions of the presynaptic plasma membrane is spatially and temporally regulated. After docking at the active zone, synaptic vesicles undergo a series of priming reactions to mature to a fusion-competent state. Recent evidence suggests that docking and priming reactions can occur within 300 ms (Zenisek and others 2000). At this point, the influx of Ca 2+ ions through voltage-gated Ca 2+ channels in response to action potentials triggers rapid exocytosis of fusion-competent vesicles. The initial release of vesicles from the RRP or docked-primed pool, and subsequent replenishment and release from the reserve pool, results in biphasic release kinetics, with a fast-release phase corresponding to release from the RRP and a slower release phase due to the mobilization and release of vesicles from the reserve pool. Such differences in mobilization rates from different pools have been observed in a number of secretory systems, both containing active zones (e.g., central synapses, neuromuscular junction) and lacking active zones (e.g., adrenal chromaffin cells) (Neher 1998). In contrast to this prevailing view on the functional architecture of the synapse, a recent study in the frog neuromuscular junction showed that the docked vesicle pool might not correspond to the RRP and some vesicles that are not docked and a distance away from the release site may have a priority to fuse over vesicles that appear to be docked (Rizzoli and Betz 2004). These findings question whether the anatomical location of vesicles 62 THE NEUROSCIENTIST Synaptic Vesicle Reuse

7 Fig. 3. Current hypotheses regarding synaptic vesicle pool hierarchy. The figure depicts alternative models that can account for the differential availability of synaptic vesicles for fusion. In the classical model, the functional allocation of vesicles into pools corresponds to their relative proximity to the active zone or release sites on the plasma membrane. Current evidence, however, challenges this classical picture and suggests either that vesicle pools possess differential mobility to access the release site or that vesicles have an intrinsic molecular heterogeneity, which makes them more or less fusion competent. in the synapse (i.e., their distance to the release site) could explain the differences in their availability for fusion. Two alternative models may help us reconcile the functional observation of vesicle pools with a physical picture. The first model proposes a vesicleselective molecular cage restricting the mobility of some vesicles but allowing others to be freely mobile during stimulation (Rizzoli and Betz 2005). Such a molecular cage may be composed of synapsin-dependent attachment of synaptic vesicles to the actin cytoskeleton. This model is also consistent with the recent observation that the restricted mobility of synaptic vesicles could be modulated by phosphorylation as well as the actin cytoskeleton (Jordan and others 2005). The second model requires a molecular diversity among synaptic vesicles, which can account for the observed differences in fusion Volume 12, Number 1, 2006 THE NEUROSCIENTIST 63

8 competence. According to this model, all vesicles can be equally mobile, but only few have the full molecular complement that enables them to fuse with high probability. Especially, vesicles that have gone through the synaptic vesicle cycle a couple of times may gradually lose their full complement of essential proteins and may become less competent for fusion. Taken together with the assumption that vesicles preserve their identity during exo-endocytosis, this vesicle heterogeneity model can also explain why in several experimental preparations vesicles were found to recycle back to their pool of origin (Pyle and others 2000; Richards and others 2003). Two recent studies support diversity in fusion competence and thus in the molecular composition of vesicles. The first study examined fusion propensities of chromaffin cell dense core granules tagged with a fluorescent probe that changes color over time (as vesicles age). This setting revealed a strong relationship with the fusion probability of vesicles and their relative age. Old vesicles became gradually release incompetent, whereas fresh vesicles were more competent for release (Duncan and others 2003). The second study, performed in hippocampal synapses in culture, showed that vesicles that fuse spontaneously in the absence of extrinsic stimulation originate from a distinct pool than those that fuse in response to presynaptic action potentials (Sara and others 2005). Interestingly, vesicles that respond to activity had a far lower propensity to fuse spontaneously compared to vesicles that carry out spontaneous neurotransmission. The results of these two studies are consistent with the hypothesis that the vesicle population in a synapse is biochemically heterogeneous. The source of this heterogeneity can be due to multiple parameters including the age of vesicles, their use history, or possibly their distinct biogenic origins. Future studies with better molecular tools and an improved ability to identify differentially labeled vesicles at the ultrastructural level will be instrumental in addressing questions on vesicle heterogeneity in the synapse. Synaptic Vesicle Recycling as a Substrate for Neurological and Psychiatric Illness Subtle alterations in synaptic vesicle trafficking may directly or indirectly contribute to pathophysiology associated with several neurological and psychiatric diseases. Modifications in presynaptic vesicle trafficking can cause frequency-dependent changes in chemical neurotransmission and thus alter the information coding in neural circuits. Furthermore, with respect to longterm plasticity, pathological expression of a presynaptic protein could affect not only presynaptic long-term plasticity but also, by altering glutamate release patterns, indirectly alter postsynaptic long-term plasticity. Therefore, a closer look at presynaptic dysfunction as a potential cause of neurological symptoms or cognitive deficits associated with these brain disorders is warranted. Presynaptic machinery may also prove to be a viable therapeutic target because small changes in a neurotransmitter release pattern in response to activity may counter the symptoms without substantial interference with postsynaptic receptor pharmacology. Studies examining potential alterations in synaptic transmission in brain disorders have started to analyze rodent models of various diseases. For instance, we have recently analyzed a mouse model of a severe form of neuronal ceroid lipofuscinosis (NCL). NCLs are a newly recognized set of lysosomal storage disorders that eventually trigger significant neurodegeneration. In our analysis of neuronal cultures that lack the enzyme responsible for the infantile form of neuronal ceroid lipofuscinosis (palmitoyl protein thioesterase-1 or PPT1), we found a progressive reduction in synaptic vesicle number per synapse. This reduction in vesicle number had a discernable impact on neurotransmission, which could be related to the key features of the disease, such as myoclonus and seizures (Virmani and others 2005). For the analysis for complex psychiatric diseases such as schizophrenia, however, it is important to investigate molecular changes in the levels of multiple synaptic proteins in postmortem human tissue to arrive at a testable hypothesis on the role of synaptic dysfunction. If synaptic vesicle recycling is rate-limiting during physiologically relevant forms of activity in the brain, then this process may underlie pathological conditions in the schizophrenic brain, which may in part originate from malfunction in glutamatergic signaling in multiple brain regions (Tamminga 1998). This hypothesis is consistent with microarray analysis performed on brains from postmortem schizophrenic patients, which revealed a selective alteration in presynaptic proteins that include N-ethylmaleimide-sensitive factor (NSF), synapsin-2, and synaptojanin (Mirnics and others 2000; Mirnics and others 2001). Interestingly, these proteins are involved in recycling synaptic vesicles rather than driving synaptic vesicle fusion reaction per se. Synapsin-2 is thought to be crucial for vesicle recruitment and maintenance of synaptic vesicle pools in the terminals, synaptojanin is a critical protein for uncoating of clathrin-coated vesicles endocytosed from the plasma membrane, and NSF is a molecular chaperone required for recycling of soluble NSF attachment protein receptors (SNAREs) (critical for the vesicle fusion process) after fusion. Clearly, investigation of the synaptic basis for neurological and mental illness is still at its infancy. However, with the emergence of new mouse models for disorders such as other lysosomal storage diseases or more complex neuropsychiatric disorders, we will be able to delineate the synaptic abnormalities that are associated with these diseases. This endeavor will significantly benefit from a better understanding of the mechanisms that underlie synaptic vesicle trafficking. References Aravanis AM, Pyle JL, Tsien RW Single synaptic vesicles fusing transiently and successively without loss of identity. Nature 423: THE NEUROSCIENTIST Synaptic Vesicle Reuse

9 Betz WJ, Bewick GS Optical analysis of synaptic vesicle recycling at the frog neuromuscular junction. Science 255: Betz WJ, Bewick GS Optical monitoring of transmitter release and synaptic vesicle recycling at the frog neuromuscular junction. J Physiol 460: Betz WJ, Mao F, Smith CB Imaging exocytosis and endocytosis. Curr Opin Neurobiol 6: Ceccarelli B, Hurlbut WP, Mauro A Turnover of transmitter and synaptic vesicles at the frog neuromuscular junction. J Cell Biol 57: Cremona O, De Camilli P Synaptic vesicle endocytosis. Curr Opin Neurobiol 7: Cremona O, Di Paolo G, Wenk MR, Luthi A, Kim WT, Takei K, and others Essential role of phosphoinositide metabolism in synaptic vesicle recycling. Cell 99: Deak F, Schoch S, Liu X, Sudhof TC, Kavalali ET Synaptobrevin is essential for fast synaptic-vesicle endocytosis. Nat Cell Biol 6: Delgado R, Maureira C, Oliva C, Kidokoro Y, Labarca P Size of vesicle pools, rates of mobilization, and recycling at neuromuscular synapses of a Drosophila mutant, shibire. Neuron 28: Duncan RR, Greaves J, Wiegand UK, Matskevich I, Bodammer G, Apps DK, and others Functional and spatial segregation of secretory vesicle pools according to vesicle age. Nature 422: Fesce R, Grohovaz F, Valtorta F, Meldolesi J Neurotransmitter release: fusion or kiss-and-run? Trends Cell Biol 4:1 4. Galli T, Haucke V Cycling of synaptic vesicles: how far? How fast! Sci STKE 2001:RE1. Gandhi SP, Stevens CF Three modes of synaptic vesicular recycling revealed by single-vesicle imaging. Nature 423: Harata N, Pyle JL, Aravanis AM, Mozhayeva M, Kavalali ET, Tsien RW Limited numbers of recycling vesicles in small CNS nerve terminals: implications for neural signaling and vesicular cycling. Trends Neurosci 24: Henkel AW, Almers W Fast steps in exocytosis and endocytosis studied by capacitance measurements in endocrine cells. Curr Opin Neurobiol 6: Heuser JE, Reese TS Evidence for recycling of synaptic vesicle membrane during transmitter release at the frog neuromuscular junction. J Cell Biol 57: Jahn R, Lang T, Sudhof TC Membrane fusion. Cell 112: Jockusch WJ, Praefcke GJ, McMahon HT, Lagnado L Clathrindependent and clathrin-independent retrieval of synaptic vesicles in retinal bipolar cells. Neuron 46: Jordan R, Lemke EA, Klingauf J Visualization of synaptic vesicle movement in intact synaptic boutons using fluorescence fluctuation spectroscopy. Biophys J 89: Klingauf J, Kavalali ET, Tsien RW Kinetics and regulation of fast endocytosis at hippocampal synapses. Nature 394: Koenig JH, Ikeda K Synaptic vesicles have two distinct recycling pathways. J Cell Biol 135: Koenig JH, Ikeda K Contribution of active zone subpopulation of vesicles to evoked and spontaneous release. J Neurophysiol 81: Li Z, Burrone J, Tyler WJ, Hartman KN, Albeanu DF, Murthy VN Synaptic vesicle recycling studied in transgenic mice expressing synaptophluorin. Proc Natl Acad Sci U S A 102: Li Z, Murthy VN Visualizing postendocytic traffic of synaptic vesicles at hippocampal synapses. Neuron 31: Luthi A, Di Paolo G, Cremona O, Daniell L, De Camilli P, McCormick DA Synaptojanin 1 contributes to maintaining the stability of GABAergic transmission in primary cultures of cortical neurons. J Neurosci 21: Malgaroli A, Ting AE, Wendland B, Bergamaschi A, Villa A, Tsien RW, and others Presynaptic component of long-term potentiation visualized at individual hippocampal synapses. Science 268: Matteoli M, Takei K, Perin MS, Sudhof TC, De Camilli P Exoendocytotic recycling of synaptic vesicles in developing processes of cultured hippocampal neurons. J Cell Biol 117: Miesenbock G, De Angelis DA, Rothman JE Visualizing secretion and synaptic transmission with ph-sensitive green fluorescent proteins. Nature 394: Mirnics K, Middleton FA, Lewis DA, Levitt P Analysis of complex brain disorders with gene expression microarrays: schizophrenia as a disease of the synapse. Trends Neurosci 24: Mirnics K, Middleton FA, Marquez A, Lewis DA, Levitt P Molecular characterization of schizophrenia viewed by microarray analysis of gene expression in prefrontal cortex. Neuron 28: Murthy VN, De Camilli P Cell biology of the presynaptic terminal. Annu Rev Neurosci 26: Murthy VN, Stevens CF Synaptic vesicles retain their identity through the endocytic cycle. Nature 392: Murthy VN, Stevens CF Reversal of synaptic vesicle docking at central synapses. Nat Neurosci 2: Neher E Vesicle pools and Ca2+ microdomains: new tools for understanding their roles in neurotransmitter release. Neuron 20: Nicholson-Tomishima K, Ryan TA Kinetic efficiency of endocytosis at mammalian CNS synapses requires synaptotagmin I. Proc Natl Acad Sci U S A 101: Poskanzer KE, Marek KW, Sweeney ST, Davis GW Synaptotagmin I is necessary for compensatory synaptic vesicle endocytosis in vivo. Nature 426: Pyle JL, Kavalali ET, Piedras-Renteria ES, Tsien RW Rapid reuse of readily releasable pool vesicles at hippocampal synapses. Neuron 28: Richards DA, Bai J, Chapman ER Two modes of exocytosis at hippocampal synapses revealed by rate of FM1-43 efflux from individual vesicles. J Cell Biol 168: Richards DA, Guatimosim C, Betz WJ Two endocytic recycling routes selectively fill two vesicle pools in frog motor nerve terminals. Neuron 27: Richards DA, Guatimosim C, Rizzoli SO, Betz WJ Synaptic vesicle pools at the frog neuromuscular junction. Neuron 39: Rizzoli SO, Betz WJ The structural organization of the readily releasable pool of synaptic vesicles. Science 303: Rizzoli SO, Betz WJ Synaptic vesicle pools. Nat Rev Neurosci 6: Rosenmund C, Stevens CF Definition of the readily releasable pool of vesicles at hippocampal synapses. Neuron 16: Ryan TA, Smith SJ Vesicle pool mobilization during action potential firing at hippocampal synapses. Neuron 14: Sankaranarayanan S, Ryan TA Real-time measurements of vesicle-snare recycling in synapses of the central nervous system. Nat Cell Biol 2: Sankaranarayanan S, Ryan TA Calcium accelerates endocytosis of vsnares at hippocampal synapses. Nat Neurosci 4: Sara Y, Mozhayeva MG, Liu X, Kavalali ET Fast vesicle recycling supports neurotransmission during sustained stimulation at hippocampal synapses. J Neurosci 22: Sara Y, Virmani T, Deak F, Liu X, Kavalali ET An isolated pool of vesicles recycles at rest and drives spontaneous neurotransmission. Neuron 45: Schikorski T, Stevens CF Morphological correlates of functionally defined synaptic vesicle populations. Nat Neurosci 4: Schneggenburger R, Meyer AC, Neher E Released fraction and total size of a pool of immediately available transmitter quanta at a calyx synapse. Neuron 23: Shupliakov O, Low P, Grabs D, Gad H, Chen H, David C, and others Synaptic vesicle endocytosis impaired by disruption of dynamin-sh3 domain interactions. Science 276: Stevens CF, Williams JH Kiss and run exocytosis at hippocampal synapses. Proc Natl Acad Sci U S A 97: Südhof TC The synaptic vesicle cycle: a cascade of protein-protein interactions. Nature 375: Sudhof TC The synaptic vesicle cycle revisited. Neuron 28: Sudhof TC The synaptic vesicle cycle. Annu Rev Neurosci 27: Volume 12, Number 1, 2006 THE NEUROSCIENTIST 65

10 Sun JY, Wu XS, Wu LG Single and multiple vesicle fusion induce different rates of endocytosis at a central synapse. Nature 417: Takei K, Mundigl O, Daniell L, De Camilli P The synaptic vesicle cycle: a single vesicle budding step involving clathrin and dynamin. J Cell Biol 133: Tamminga CA Schizophrenia and glutamatergic transmission. Crit Rev Neurobiol 12: Thorn P, Fogarty KE, Parker I Zymogen granule exocytosis is characterized by long fusion pore openings and preservation of vesicle lipid identity. Proc Natl Acad Sci U S A 101: Virmani T, Gupta P, Liu X, Kavalali ET, Hofmann SL Progressively reduced synaptic vesicle pool size in cultured neurons derived from neuronal ceroid lipofuscinosis-1 knockout mice. Neurobiol Dis (in press). Virmani T, Han W, Liu X, Sudhof TC, Kavalali ET Synaptotagmin 7 splice variants differentially regulate synaptic vesicle recycling. EMBO J 22: Wu LG, Borst JG The reduced release probability of releasable vesicles during recovery from short-term synaptic depression. Neuron 23: Wucherpfennig T, Wilsch-Brauninger M, Gonzalez-Gaitan M Role of Drosophila Rab5 during endosomal trafficking at the synapse and evoked neurotransmitter release. J Cell Biol 161: Zenisek D, Steyer JA, Almers W Transport, capture and exocytosis of single synaptic vesicles at active zones. Nature 406: Zenisek D, Steyer JA, Feldman ME, Almers W A membrane marker leaves synaptic vesicles in milliseconds after exocytosis in retinal bipolar cells. Neuron 35: THE NEUROSCIENTIST Synaptic Vesicle Reuse