Wonchul Shin

Exocytosis and Endocytosis: Modes, Functions, and Coupling Mechanisms*

Vesicle exocytosis releases content to mediate many biological events, including synaptic transmission essential for brain functions. Following exocytosis, endocytosis is initiated to retrieve exocytosed vesicles within seconds to minutes. Decades of studies in secretory cells reveal three exocytosis modes coupled to three endocytosis modes: (a) full-collapse fusion, in which vesicles collapse into the plasma membrane, followed by classical endocytosis involving membrane invagination and vesicle reformation; (b) kiss-and-run, in which the fusion pore opens and closes; and (c) compound exocytosis, which involves exocytosis of giant vesicles formed via vesicle-vesicle fusion, followed by bulk endocytosis that retrieves giant vesicles. Here we review these exo- and endocytosis modes and their roles in regulating quantal size and synaptic strength, generating synaptic plasticity, maintaining exocytosis, and clearing release sites for vesicle replenishment. Furthermore, we highlight recent progress in understanding how vesicle endocytosis is initiated and is thus coupled to exocytosis. The emerging model is that calcium influx via voltage-dependent calcium channels at the calcium microdomain triggers endocytosis and controls endocytosis rate; calmodulin and synaptotagmin are the calcium sensors; and the exocytosis machinery, including SNARE proteins (synaptobrevin, SNAP25, and syntaxin), is needed to coinitiate endocytosis, likely to control the amount of endocytosis.

Calcium-channel number critically influences synaptic strength and plasticity at the active zone.

How synaptic-vesicle release is controlled at the basic release structure, the active zone, is poorly understood. By performing cell-attached current and capacitance recordings predominantly at single active zones in rat calyces, we found that single active zones contained 5−218 (mean, 42) calcium channels and 1−10 (mean, 5) readily releasable vesicles (RRVs) and released 0−5 vesicles during a 2-ms depolarization. Large variation in the number of calcium channels caused wide variation in release strength (measured during a 2-ms depolarization) by regulating the RRV release probability (PRRV) and the RRV number. Consequently, an action potential opened ∼1–35 (mean, ∼7) channels, resulting in different release probabilities at different active zones. As the number of calcium-channels determined PRRV, it critically influenced whether subsequent release would be facilitated or depressed. Regulating calcium channel density at active zones may thus be a major mechanism to yield synapses with different release properties and plasticity. These findings may explain large differences reported at synapses regarding release strength (release of 0, 1 or multiple vesicles), PRRV, short-term plasticity, calcium transients and the requisite calcium-channel number for triggering release.

Vesicle Pool Heterogeneity at Hippocampal Glutamate and GABA Synapses

Glutamate and GABA are the major fast excitatory and inhibitory neurotransmitters, respectively, in the CNS. Although glutamate and GABA have clearly distinct postsynaptic actions, we are just beginning to appreciate that presynaptic differences between glutamatergic and GABAergic neurons may contribute to distinct functions of these transmitter systems. We therefore probed possible differences between the functional synaptic vesicle populations of glutamatergic and GABAergic neurons. We examined superecliptic synaptopHluorin (SpH) fluorescence during 20 Hz electrical stimulation in transfected hippocampal neurons and identified the phenotype of SpH-fluorescent synapses with post hoc immunostaining. With 200 stimuli (10 s), individual glutamate synapses displayed considerably more variability in peak SpH fluorescence than GABA synapses, without a strong difference in the mean SpH fluorescence increase. This spatial heterogeneity could not be accounted for by differences in endocytosis, which was nearly constant over these short time periods across glutamate and GABA synapses. Instead, variability in vesicle exocytosis correlated with variability in total vesicle staining and in measures of the total recycling pool size. Differences were also evident using FM1-43 [N-(3-triethylammoniumpropyl)-4-(4-(dibutylamino)styryl) pyridinium dibromide] uptake. These data support the idea that the population of glutamate synapses exhibits more heterogeneity in release properties than the population of GABA synapses, possibly correlated with glutamatergic synaptic malleability.

Actin dynamics provides membrane tension to merge fusing vesicles into the plasma membrane

Vesicle fusion is executed via formation of an Ω-shaped structure (Ω-profile), followed by closure (kiss-and-run) or merging of the Ω-profile into the plasma membrane (full fusion). Although Ω-profile closure limits release but recycles vesicles economically, Ω-profile merging facilitates release but couples to classical endocytosis for recycling. Despite its crucial role in determining exocytosis/endocytosis modes, how Ω-profile merging is mediated is poorly understood in endocrine cells and neurons containing small ∼30–300 nm vesicles. Here, using confocal and super-resolution STED imaging, force measurements, pharmacology and gene knockout, we show that dynamic assembly of filamentous actin, involving ATP hydrolysis, N-WASP and formin, mediates Ω-profile merging by providing sufficient plasma membrane tension to shrink the Ω-profile in neuroendocrine chromaffin cells containing ∼300 nm vesicles. Actin-directed compounds also induce Ω-profile accumulation at lamprey synaptic active zones, suggesting that actin may mediate Ω-profile merging at synapses. These results uncover molecular and biophysical mechanisms underlying Ω-profile merging.

Post-fusion structural changes and their roles in exocytosis and endocytosis of dense-core vesicles

Vesicle fusion with the plasma membrane generates an Ω-shaped membrane profile. Its pore is thought to dilate until flattening (full-collapse), followed by classical endocytosis to retrieve vesicles. Alternatively, the pore may close (kiss-and-run), but the triggering mechanisms and its endocytic roles remain poorly understood. Here, using confocal and STED imaging of dense-core vesicles, we find that fusion-generated Ω-profiles may enlarge or shrink while maintaining vesicular membrane proteins. Closure of fusion-generated Ω-profiles, which produces various sizes of vesicles, is the dominant mechanism mediating rapid and slow endocytosis within ~1–30 s. Strong calcium influx triggers dynamin-mediated closure. Weak calcium influx does not promote closure, but facilitates the merging of Ω-profiles with the plasma membrane via shrinking rather than full-collapse. These results establish a model, termed Ω-exo-endocytosis, in which the fusion-generated Ω-profile may shrink to merge with the plasma membrane, change in size, or change in size then close in response to calcium, which is the main mechanism to retrieve dense-core vesicles.

Hemi-fused structure mediates and controls fusion and fission in live cells

Membrane fusion and fission are vital for eukaryotic life. For three decades, it has been proposed that fusion is mediated by fusion between the proximal leaflets of two bilayers (hemi-fusion) to produce a hemi-fused structure, followed by fusion between the distal leaflets, whereas fission is via hemi-fission, which also produces a hemi-fused structure, followed by full fission. This hypothesis remained unsupported owing to the lack of observation of hemi-fusion or hemi-fission in live cells. A competing fusion hypothesis involving protein-lined pore formation has also been proposed. Here we report the observation of a hemi-fused Ω-shaped structure in live neuroendocrine chromaffin cells and pancreatic β-cells, visualized using confocal and super-resolution stimulated emission depletion microscopy. This structure is generated from fusion pore opening or closure (fission) at the plasma membrane. Unexpectedly, the transition to full fusion or fission is determined by competition between fusion and calcium/dynamin-dependent fission mechanisms, and is notably slow (seconds to tens of seconds) in a substantial fraction of the events. These results provide key missing evidence in support of the hemi-fusion and hemi-fission hypothesis in live cells, and reveal the hemi-fused intermediate as a key structure controlling fusion and fission, as fusion and fission mechanisms compete to determine the transition to fusion or fission.

The SNARE Proteins SNAP25 and Synaptobrevin Are Involved in Endocytosis at Hippocampal Synapses

SNAP25, an essential component of the soluble NSF (N-ethylmaleimide-sensitive factor) attachment protein receptor (SNARE) complex that mediates exocytosis, is not considered to play a role in endocytosis, which couples to exocytosis by retrieving a similar amount of exocytosed vesicles. By knocking down SNAP25 and imaging slow endocytosis at a conventional synapse, the rat cultured hippocampal synapse, we found that SNAP25 is involved in slow, clathrin-dependent endocytosis. With similar techniques, we found that not only SNAP25, but also synaptobrevin is involved in slow endocytosis. These results provide the first evidence showing the dual role of SNAP25 and synaptobrevin in both exocytosis and slow endocytosis at conventional synapses. Such a dual role may contribute to mediate the coupling between exocytosis and clathrin-dependent endocytosis at conventional synapses, a mechanism critical for the maintenance of synaptic transmission and the normal structure of nerve terminals.

Voltage-Dependent Calcium Channels at the Plasma Membrane, but Not Vesicular Channels, Couple Exocytosis to Endocytosis

Although calcium influx triggers endocytosis at many synapses and non-neuronal secretory cells, the identity of the calcium channel is unclear. The plasma membrane voltage-dependent calcium channel (VDCC) is a candidate, and it was recently proposed that exocytosis transiently inserts vesicular calcium channels at the plasma membrane, thus triggering endocytosis and coupling it to exocytosis, a mechanism suggested to be conserved from sea urchin to human. Here, we report that the vesicular membrane, when inserted into the plasma membrane upon exocytosis, does not generate a calcium current or calcium increase at a mammalian nerve terminal. Instead, VDCCs at the plasma membrane, including the P/Q-type, provide the calcium influx to trigger rapid and slow endocytosis and, thus, couple endocytosis to exocytosis. These findings call for reconsideration of the vesicular calcium channel hypothesis. They are likely to apply to many synapses and non-neuronal cells in which VDCCs control exocytosis, and exocytosis is coupled to endocytosis.

Most Vesicles in a Central Nerve Terminal Participate in Recycling

Studies over the last decade using FM dyes to label vesicles at many terminals, including the calyx-type nerve terminal, led to a well accepted “principle” that only a small fraction of vesicles (∼5–20%) participate in recycling under physiological conditions. This principle imposes a large challenge in maintaining synaptic transmission during repetitive firing, because the small recycling pool may limit the number of available vesicles for release and nerve terminals would have to distinguish the recycling pool from the reserve pool and keep reserve pool vesicles from being used. By recording the presynaptic capacitance changes and the postsynaptic EPSC at rat calyx of Held synapses in the absence or presence of transmitter glutamate in nerve terminals, we developed a new method to count functional recycling vesicles. We found that essentially all vesicles in calyces participated in recycling, challenging the small-recycling-pool principle established by FM dye labeling. Nerve terminals may use all available vesicles to maximize their ability in maintaining synaptic transmission during repetitive firing.

Closure of Pre-Existing Ω Profiles Mediates Compensatory Endocytosis, Endocytosis Overshoot and Bulk Endocytosis

In neurons and endocrine cells, depolarization induces exocytosis, whereas endocytosis retrieves exocytosed vesicles to maintain secretion. It is generally thought that after depolarization, endocytosis must involve three steps, the Ω-profile formation, vesicular membrane protein recruitment, and Ω-profile closure. Here we report a novel form of endocytosis that involves only one step after depolarization, the pore closure of pre-existing Ω-profiles, in neuroendocrine chromaffin cells. Pre-existing Ω-profiles were fusion-generated Ω-profiles from previous rounds of exocytosis that did not collapse, but retained vesicular membrane proteins. Their pore closure was triggered by calcium influx during subsequent depolarization and was mediated by dynamin, which generated vesicles within seconds without the need for recruiting vesicular proteins. Such closures substantially mediated compensatory endocytosis, endocytosis overshoot and bulk endocytosis as widely observed in neurons and endocrine cells. We conclude that pre-existing Ω-profile closure is a major mode of endocytosis enabling efficient vesicle retrieval in secretory cells.