Synaptic vesicle release depends upon molecules called SNAREs, which are present on the vesicle (v-SNAREs) and the target membrane (t-SNAREs). Complementary SNAREs allow vesicles to fuse with the target membrane and release their contents (neurotransmitters) into the synaptic space. The prevailing theory of release states that these fusion events occur spontaneously at a very low rate, producing minis. An action potential causes an influx of calcium into the presynaptic cell, which activates a calcium sensor and greatly increases the probability of vesicle fusion, creating a large, synchronized release of neurotransmitter from many vesicles.
The first paper I will discuss, by Fredj and Burrone, addresses the assumption that spontaneously-released vesicles are the same as evoked vesicles. If minis are caused by the random fusion of synaptic vesicles with the cell membrane, then we would expect spontaneously-released vesicles to resemble all other vesicles, with the only difference being that they accidentally fused without receiving a calcium signal. Previous experiments have indicated that this is probably not the case (Sara et. al, 2005), but Ben Fredj and Burrone developed a new technique for labeling presynaptic vesicles that supports the notion of a separate pool of spontaneously-released vesicles. Vesicles that have released their contents are recycled by the cell, allowing them to be refilled with neurotransmitter and used again later.
The researchers created a tagged version of an internal vesicle protein called VAMP2, which they called biosyn. Fluorescently-labeled streptavidin will permanently bind to biosyn. If streptavidin is present in the synaptic space, it will label the biosyn that is exposed when vesicles fuse to the cell membrane. This allowed the researchers to visualize active, fused vesicles in a living synapse under different conditions. After verifying that biosyn is a reliable measure of spontaneous and evoked fusion events, and that the tagged protein does not interfere with normal synaptic processes, they went on to test whether spontaneous and evoked release events use the same populations of vesicles.
After testing biosyn labeling of spontaneous and evoked vesicle release, Fredj and Burron noticed that they saw only about half as much biosyn signal when examining spontaneous release, compared to evoked release. This occurred even though the vesicle populations appeared to be saturated (that is, the biosyn signal reached a maximum level after a sustained period of release, indicating that all of the available vesicles had already fused and been labeled with fluorescent streptavidin). This could mean that spontaneously-released vesicles represent a sub-population of vesicles that can undergo evoked release, or it could mean that the two types of release draw from distinct vesicle pools, with fewer spontaneous vesicle than evoked vesicles.
To distinguish between these two possible explanations, the scientists sequentially labeled evoked and spontaneous vesicles within the same cell using two different colors of fluorescent streptavidin. They describe that experiment as follows:
Synapses were first stimulated with a saturating stimulus of two 90-s depolarizations in the presence of strep488 [green streptavidin], which strongly labeled the entire recycling pool. A further depolarizing stimulus in strep594 [red streptatividin] resulted in no further labeling (or very small amounts of labeling), as all biosyn binding sites were occupied by strep488, confirming that our depolarizing stimulus mobilized all possible vesicles. On the other hand, after labeling the recycling pool with strep488, a further 15-min exposure to strep594 in conditions that would only allow spontaneous fusion events resulted in a substantial amount of labeling.
The data show that two different populations of vesicles are exposed to the fluorescent streptavidin probes under different release conditions. This prompted them to ask, "If the recycling pool of vesicles cannot account for spontaneous release, then where do these vesicles come from?" They propose that the spontaneous vesicles may come from the so-called "resting" pool. This is the name given to a previously-identified pool of vesicles that are not mobilized by neuronal activity. Further experiments by Fredj and Burrone provide evidence that this resting pool does provide the vesicles for spontaneous release.
A News and Views summary of the paper gives us more to think about:
The big question now becomes what other differences might exist between these vesicles besides the pools that they come from. Are they released from different locations in the presynaptic terminal, as suggested by a recent study? Does their protein and/or lipid composition differ? We can take some comfort in Fredj and Burrone's observation that the sizes of the evoked and spontaneous pools were highly correlated in individual axon terminals, consistent with previous studies. Further experiments will be required to identify all of the similarities and differences between these two forms of vesicle fusion and validate the continued use of spontaneous release to characterize evoked transmission.
Another paper from the same issue of the journal, by Jun Xu et al., indicates that while spontaneously-released vesicles may be drawn from a separate pool, they are using the same calcium sensor as evoked release (contrary to the belief that minis are, by definition, calcium-insensitive). They studied spontaneous (mini) and evoked inhibitory post-synaptic currents in cultured cortical neurons. While removing extracellular calcium from the culture medium depressed evoked currents more strongly than minis, the application of the membrane-permeable calcium chelator BAPTA blocked over 95% of minis. Meanwhile, applying caffeine (which increases intracellular calcium availability) increased the number of minis observed. This implies that while evoked vesicle release depends strongly on extracellular calcium influx, even minis are not calcium-independent -- some calcium must be present to trigger a spontaneous release event.
Evoked neurotransmitter release is known to be regulated by the calcium sensor synaptotagmin. Somewhat paradoxically, genetic deletion of synaptotagmin-1 (Syt1) causes an increase in the number of minis, leading to the theory that this protein both allows evoked release and prevents spontaneous release through some sort of clamping mechanism. But as the authors note, "The clamping hypothesis ... argues against the notion that spontaneous release may be biologically meaningful, as it is difficult to imagine how an accidental byproduct of evoked release could control a physiological process. Moreover, the clamping hypothesis fails to explain why at least some mini release is Ca2+ dependent."
Xu et al. decided to better elucidate the role of Syt1 in spontaneous vesicle release. By studying minis in neurons lacking Syt1, they found that the upregulated minis in those cells were still calcium-dependent (they, too, could be blocked by BAPTA). They showed that the synapses with no Syt1 seemed to exhibit greater calcium affinity than wild-type synapses. This indicates that Syt1 is not acting merely as a clamp to block spontaneous release in normal cells, but that some other, more sensitive calcium sensor is able to produce spontaneous -- but not evoked -- vesicle release in the absence of Syt1. This was true for both excitatory and inhibitory synapses.
One explanation for this result would be that Syt1 serves as the primary calcium sensor for spontaneous release, but that an unknown sensor, normally clamped by Syt1, can lead to spontaneous release in Syt1's absence. To test this, Xu et al. generated neurons expressing mutant varieties of Syt1 with different calcium affinities. If Syt1 is indeed responsible for both evoked and spontaneous release, one would predict that changing the calcium affinity of Syt1 would create a commensurate change in the magnitude of both spontaneous and evoked release. Indeed, this is what they found. The result held true when they tested the different Syt1 mutations in brain slices as well as in cultured neurons.
The authors leave us with this conclusion:
Evoked and spontaneous neurotransmitter release are generally considered to represent distinct types of release that are differentially regulated. Their distinct natures are evidenced by the fact that spontaneous release is maintained in the presence of the sodium-channel inhibitor TTX, which abolishes action potentials and evoked release. We found, however, that despite their differential regulation, these two types of release are mechanistically identical in that they both are triggered by Ca2+ binding to Syt1. The major evidence for this conclusion rests on the three Syt1 knockin mutations that we used. We previously demonstrated that these mutations either increase Ca2+-dependent binding of Syt1 to SNARE complexes or decrease the apparent Ca2+ affinity of Syt1 binding to phospholipids. In a direct comparison of all three knockin mutations, we found that they cause a corresponding change in evoked release and a precisely equivalent change in spontaneous release. ... Moreover, our results support previous suggestions that spontaneous release is physiologically important. Ca2+ regulation generally implies a physiologically controlled function; thus, the finding that spontaneous release is controlled by Ca2+ binding to Syt1 implies a physiological role... Many neurotransmitters and neuromodulators act by increasing or decreasing presynaptical Ca2+ concentrations, suggesting that these agents may control synaptic circuits, at least in part, by regulating Syt1-dependent spontaneous release without triggering action potentials.
Of course, this article also leaves us with some burning questions. What is this second calcium sensor that is clamped by Syt1? Why is it so sensitive? What important physiological roles are being played by these highly-regulated mini release events? Clearly, more research is needed into these areas.
In conclusion, we've learned from this month's Nature Neuroscience that spontaneous release events are not the same as evoked neurotransmitter release. Although these two types of synaptic vesicles use the same major calcium sensor (and are both calcium-dependent, contrary to popular belief!), there exist separate pools of spontaneous and evoked vesicles that respond differently to intracellular calcium fluctuations, and never the twain shall meet. It will be interesting to see where we go from here, teasing apart the distinct roles that these two types of vesicles play in neurotransmission.
Fredj, N., & Burrone, J. (2009). A resting pool of vesicles is responsible for spontaneous vesicle fusion at the synapse Nature Neuroscience, 12 (6), 751-758 DOI: 10.1038/nn.2317
Xu, J., Pang, Z., Shin, O., & Südhof, T. (2009). Synaptotagmin-1 functions as a Ca2+ sensor for spontaneous release Nature Neuroscience, 12 (6), 759-766 DOI: 10.1038/nn.2320
Sara, Y., Virmani, T., Deák, F., Liu, X., & Kavalali, E. (2005) An isolated pool of vesicles recycles at rest and drives spontaneous neurotransmission. Neuron, 45 (4), 563-573 DOI: 10.1016/j.neuron.2004.12.056