The optical approach
Wolfhard Almers and colleagues at the Vollum Institute (Portland, OR) used evanescent wave and total internal reflection microscopy to peek <100 nm beneath the membrane of goldfish retinal bipolar neurons (1). Their images depict the intricate molecular ballet performed by synaptic vesicles as they move toward the membrane, dock, and fuse. And these techniques may explain why synapses “fatigue” (the first phase of neurotransmitter release is usually stronger than subsequent phases, unless the neuron is given time to rest and recover). Is this because the cell runs out of docked vesicles, or does something inhibit the plasma membrane from supporting further fusion? “What mechanisms mediate the transport of vesicles to active zones, and how are they regulated?” asks Almers. “What of the proteins that participate in fusion or regulation of fusion?” Because this technique allows for the direct observation of the movements of fluorescently labeled proteins, researchers can measure how soon before fusion the proteins are recruited to docking sites and how soon they are dismissed.

Almers explains that the microscope objective lens is used “both for observation and for generating the evanescent field, providing access to the cell on the other side of the objective lens for stimulation and for recording electrophysiologic measurements.” Because synaptic terminals of neurons will stick to glass, the researchers used total internal reflection microscopy, an optical technique that precisely and instantaneously measures the distance between a flat surface and a microscopic sphere. This distance is determined by illuminating the sphere with an evanescent wave and then measuring the intensity of the scattered light (2). Because of the difference in refractive indices between the sphere and the glass and the resulting nonuniform illumination by the evanescent wave passing through the sphere, the sphere scatters light in direct proportion to its distance from the flat surface.

Figure 2. Neural glow ball. As the synaptic vesicle containing the neurotransmitter (green) translocates toward the plasma membrane, it moves into the evanescent field established by the light shining on the glass coverslip. As the vesicles have been fluorescently labeled, they begin to glow and, during exocytosis, that fluorescence is transferred to the plasma membrane—a visible signal of exocytosis. (Adapted from Matthews, G. Nature 2000, 406, 835–836).

Evanescent field illumination causes vesicle material to brighten after exocytosis because the intensity of an evanescent field grows steeply with proximity to the glass–cell interface (Figure2). “In hindsight, we might have expected (but didn’t) that evanescent field illumination alone would cause vesicle material to brighten after exocytosis,” quips Almers.

It was collaborator Jeurgen Steyer who first noticed the brightness increase as he watched the exocytosis of chromaffin granules. When a cell is stuck to and secretes material against a cover glass, it is illuminated more intensely and fluoresces more strongly, making evanescent field microscopes particularly well suited for filming exocytosis.

When Almers’ group built their first evanescent wave microscope in 1995 to image larger secretory vesicles, they were unable to see the much smaller synaptic vesicles. But the release of an objective lens with an exceptionally high numerical aperture (1.65), by Olympus Corp. (Melville, NY), enabled the group to build a microscope that looks <100 nm into a cell.

The investigators described vesicles held at small, discrete “active zones” near the synaptic terminal surface, which undergo rapid exocytosis after application of an electric stimulus. This event was followed by the movement of other vesicles, held in reserve about 20 nm from the plasma membrane, to exocytotic sites and that became “release-ready” 250 ms later. “Our work opens a way to watch the supply of vesicles to sites of exocytosis,” Almers indicates. It will also provide a means to answer the question of why some vesicles are “release-ready”, while others require priming.

The biophysical approach

Figure 3. Calcium cager. DM-nitrophen is commonly used to chelate calcium ions.

Two other groups, citing the historical difficulty of imaging synaptic vesicles, have approached the synapse from a purely physiological angle, focusing directly on the involvement of calcium in neurotransmitter release. Both groups used calcium-uncaging (3) to deliver a measured amount of Ca²⁺ to the cell. This technique makes use of a photolabile calcium chelator (Figure 3), which is loaded with Ca²⁺ and maintained in darkness. After loading into a presynaptic terminal, a short, visible pulse from a flash lamp or a UV pulse from a laser leads to photolysis of the chelator and the subsequent release of Ca²⁺ into the cytoplasm. The ions are detected by fluorescent Ca²⁺-indicators that are loaded with the chelator. Although the two groups used different means of releasing the Ca²⁺, both subsequently demonstrated that synaptic transmission is remarkably more sensitive to changing in tracellular calcium concentrations ([Ca²⁺]_i) than was previously believed.

Ralf Schneggenburger and Erwin Neher of Max Planck Institute for Biophysical Chemistry (Göttingen, Germany) used calcium-uncaging to explore the details of the calcium dependence of synaptic transmission rates in a large central nervous system synapse, the calyx of Held (4). It has been generally assumed that [Ca²⁺]_i in presynaptic neurons must reach at least 100 µM near the sites of vesicle fusion, but Schneggenburger’s team was able to show that step-wise elevations in [Ca²⁺]_i to only 10 mM induced fast transmitter release, effectively releasing 80% of the pool of available vesicles in less than 3 ms. Biophysical measurements and subsequent calculations showed that the calcium sensors were far from saturated, making fast synaptic transmission very sensitive to regulation by changing [Ca²⁺]_i.

“At the synapse, two successive stimuli will, in most cases, not evoke similarly large amounts of neurotransmitter release. Rather, the second release event will either be larger (“facilitated”) or smaller(“depressed”). This short-term plasticity has a major influence on how a postsynaptic neuron can integrate the information flow from a presynaptic neuron, and thus, how information processing takes place in neuronal networks,” Schneggenburger explains. He further stressed that “without understanding the mechanism of neurotransmitter release at the molecular level, it will be hard to understand how neurotransmitter release is regulated.”

Johann Bollmann and colleagues from Max Planck Institute for Medical Research (Heidelberg, Germany) and Swammerdam Institute for Life Sciences (Amsterdam) used calcium-uncaging with laser photolysis to measure the calcium sensitivity of glutamate release in a calyx-type terminal, concluding that a rise in [Ca²⁺]_i to just 1 µM readily evoked glutamate release, and an increase greater than 30 µM depleted the releasable vesicle pool within 0.5 ms (5). This suggested that the sensitivity of these neurons to neurotransmitter release was greater than previously thought and predicted that the calcium sensor, which triggers the two waves of neurotransmitter release, may be the same. Thus, the delayed release is a consequence of delayed triggering rather than the triggering of a different sensor with a higher affinity.

Thanks to calcium uncaging, more is understood about the sensitivity of neurotransmitter release to intracellular calcium levels. For example, explains Schneggenburger, “since we now know the range of relevant [Ca²⁺]_i, it will be easier for molecular biologists to identify the Ca²⁺-binding proteins that initiate the fusion reaction upon Ca²⁺-binding.”

Calcium uncaging also lends itself to other intriguing questions about the modulation of synaptic vesicle fusion. Because the amplitude of the Ca²⁺ signal produced by calcium uncaging is now known, investigators can apply stimuli and determine whether the relationship between [Ca²⁺] and neurotransmitter release changes. One question that might be answered is whether synaptic facilitation—the increase in neurotransmitter release sometimes observed with repeated neuronal stimulation—is caused by an increase in Ca²⁺ directly dictating neurotransmitter release, or whether the Ca²⁺ acts on yet another sensor to modulate this release.

Writing in the series Science for All in the late 1800s (6), Andrew Wilson talks of “exercising our ‘senses’—whatever these may prove to be.” Wouldn’t he and his contemporaries have enjoyed fiddling with these new tools?

References

1. Zenisek, D.; Steyer, J. A.; Almers, W. Nature 2000, 406, 849–854.
2. Evanscent Waves. www.andrew.cmu.edu/user/dcprieve/Evanescent%20waves.htm and Total Internal Reflection Microscopy. www.andrew.cmu.edu/user/dcpreive/TIRM.htm. (accessed Oct 2000).
3. Zucker, R. S. Cell Calcium 1992, 13, 29–40.
4. Schneggenburger, R.; Neher, E. Nature 2000, 406, 889–893.
5. Bollmann, J. H.; Sakmann, B; Borst, J.G.G. Science 2000, 289, 953–956.
6. Wilson, A. Nerves or No Nerves? Or, The Art of Feeling. In Science for All; R. Brown, Ed.; Cassell, Petter and Galpin: London, 1889; pp 174–180.ne.