- Email: [email protected]
Synaptic mechanisms of bipolar cell terminals
Vision Research 39 (1999) 2469 – 2476
Synaptic mechanisms of bipolar cell terminals Gary Matthews * Department of Neurobiology and Beha6ior, State Uni6ersity of New York, Stony Brook, NY 11794 -5230, USA Received 8 May 1998; received in revised form 24 August 1998
Abstract Giant synaptic terminals of goldfish bipolar neurons allow direct studies of presynaptic mechanisms underlying neurotransmitter release and its modulation. Calcium influx via L-type calcium channels of the terminal triggers synaptic vesicle exocytosis, which can be monitored in isolated terminals by means of the associated changes in membrane capacitance. Information about the kinetics and calcium dependence of synaptic exocytosis has been obtained from capacitance measurements in these ribbon-type synaptic terminals. © 1999 Elsevier Science Ltd. All rights reserved. Keywords: Synaptic mechanisms; Terminals; Bipolar neurons
Studies of presynaptic mechanisms in the vertebrate central nervous system (CNS) are complicated by the small size of most CNS synaptic terminals. However, several notable exceptions allow direct electrical measurements to be made from single terminals, including giant caliciform terminals in the brainstem auditory system of the rat (Forsythe, 1994) and in the ciliary ganglion of the chick (Stanley & Atrakchi, 1990; Stanley, 1993) and large, bulbous terminals of certain types of bipolar neuron in goldfish retina (Kaneko & Tachibana, 1985; Heidelberger & Matthews, 1992). The latter are of interest in retinal neurobiology, and in this review, I will focus on our current understanding of presynaptic events that regulate neurotransmitter release in the giant terminals of goldfish bipolar neurons.
1. Calcium current of the synaptic terminal The large size (diameter:10 mm) of the synaptic terminal of some classes of bipolar cell in carp and goldfish (Stell, 1967; Ishida, Stell & Lightfoot, 1980; Saito & Kujiraoka, 1982; Sherry & Yazulla, 1993) permits direct patch-clamp recording of synaptic calcium current, either in intact bipolar cells isolated after
* Fax: + 1-516-632-6661. E-mail address: [email protected] (G. Matthews)
enzymatic digestion of goldfish retina or in single, isolated terminals. Consistent with their central role in neurotransmitter release, voltage-dependent calcium channels are located predominantly in the synaptic terminal of the bipolar cell (Tachibana & Okada, 1991; Heidelberger & Matthews, 1992), although there is also a low density of channels in the soma. The synaptic calcium current is sensitive to dihydropyridine calciumchannel agonists and antagonists, activates at a relatively positive voltage range, and shows sustained activation during prolonged depolarization (Heidelberger & Matthews, 1992). All of these characteristics are typical of L-type calcium channels (Fox, Nowycky & Tsien, 1987). The dihydropyridine-sensitive channels have been demonstrated to control glutamate release from the synaptic terminal (Tachibana, Okada, Arimura, Kobayashi & Piccolino, 1993). Thus, in goldfish bipolar cells, neurotransmitter release is controlled by L-type calcium channels.
2. Presynaptic inhibition at the bipolar-cell terminal The synaptic terminals of goldfish bipolar neurons receive extensive feedback synapses from amacrine cells (Marc, Stell, Bok & Lam, 1978; Yazulla, Studholme & Wu, 1987), which serve to modulate the synaptic output from the bipolar cells. Several functional effects of the synaptic feedback have been described, mediated by
0042-6989/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 2 - 6 9 8 9 ( 9 8 ) 0 0 2 4 9 - 1
2470
G. Matthews / Vision Research 39 (1999) 2469–2476
different neurotransmitters and neurotransmitter receptors. Many of the amacrine cells that target bipolar-cell terminals are GABAergic (Yazulla et al., 1987), and GABA acts on two distinct GABA receptors to produce presynaptic inhibition in the terminal. First, GABA activates a GABAA-like receptor and increases chloride conductance (Tachibana & Kaneko, 1987; Heidelberger & Matthews, 1991; Matthews, Ayoub & Heidelberger, 1994). Secondly, GABA produces GTPdependent inhibition of the calcium current of the bipolar-cell terminal via a GABAB-like receptor (Heidelberger & Matthews, 1991; Matthews et al., 1994). The GABA-activated chloride current hyperpolarizes the bipolar cell and thus inhibits calcium influx into the terminal via voltage-gated calcium channels (Heidelberger & Matthews, 1991; Matthews et al., 1994). GABA inhibits the calcium current by shifting its voltage dependence toward more positive potentials, thus tending to prevent activation of the current upon depolarization. Both actions of GABA in the terminal are mediated by receptors with unusual pharmacology. The receptor coupled to chloride conductance is insensitive to bicuculline and SR95531, which typically block GABAA receptors, while the action of GABA on calcium current is neither mimicked by the GABAB agonist baclofen nor blocked by the GABAB antagonist saclofen. Because of the bicuculline resistance of the GABAA-like receptors, they are commonly referred to in retinal research as GABAC receptors (Feigenspan, Wa¨ssle & Bormann, 1993; Qian & Dowling, 1993) to distinguish them from bicuculline-sensitive isoforms of the GABAA receptor. The bicuculline-resistant GABAA-like receptor probably corresponds to the retina-specific GABA receptor subunit, r1 (Cutting, Lu, O’Hara, Kasch, Montrose-Rafizader, Donovan, Shimada, Antonarakis, Guggino, Uhl & Kazazian, 1991; Shimada, Cutting & Uhl, 1992), which produces bicuculline-resistant GABA-activated chloride channels when expressed in Xenopus oocytes. Studies using conformationally locked analogs of GABA show that the bipolar-cell GABAAlike receptor prefers the trans (extended) isomer of GABA to the cis (folded) conformation (Ayoub & Matthews, 1991). In addition to GABA, other neurotransmitters affect the bipolar-cell terminal. Neuropeptides that are found in amacrine cells, including substance P and metenkephalin, produce GTP-dependent inhibition of calcium current by shifting the voltage dependence, similar to the action produced by GABA (Ayoub & Matthews, 1992). Goldfish bipolar-cell terminals also receive dopaminergic inputs (Dowling & Ehinger, 1978; Yazulla & Zucker, 1988), and dopamine was found to enhance the increase in intracellular calcium evoked in the terminal by activation of voltage-gated calcium channels (Heidelberger & Matthews, 1994). This action of do-
pamine was mimicked by forskolin and by a permeant analog of cAMP, suggesting that dopamine acts by elevating cAMP in the terminal. Taken together, the actions of GABA, dopamine, and neuropeptides demonstrate that synaptic feedback onto the bipolar cell terminal is capable of modulating transmitter release from the bipolar cell and thus the transfer of information from the outer to the inner retina.
3. Capacitance measurements of exocytosis When a secretory granule fuses with the plasma membrane during exocytosis, the surface area of the secreting cell increases. The increased surface area can be estimated by monitoring the associated change in electrical capacitance of the cell, as shown schematically in Fig. 1. Sensitive techniques for monitoring capacitance in real time in single cells are well-established (Lindau & Neher, 1988; Gillis, 1995) and have been used to gather information about exocytosis in various types of secretory cell, including adrenal chromaffin cells (Neher & Marty, 1982), mast cells (Fernandez, Neher & Gomperts, 1984), pituitary cells (Thomas, Suprenant & Almers, 1990), and neuroendocrine nerve terminals (Fidler Lim, Nowycky & Bookman, 1990; Lindau, Stuenkel & Nordmann, 1992). Neurotransmitter is released at synapses by a process of exocytosis, during which transmitter-containing synaptic vesicles fuse with the membrane of the presynaptic terminal in response to calcium influx via voltage-gated calcium channels. Therefore, capacitance measurements are potentially useful at synapses as an index of neurotransmitter release. However, a difficulty for capacitance measurements from synaptic terminals is the small size—and hence the small capacitance—of the synaptic vesicles at most synapses. A synaptic vesicle with a diameter of about 30 nm, which would be typical for glutamatergic synapses in the CNS, would have an expected capacitance of about 3× 10 − 17 F (assuming that capacitance of biological membranes is approximately 10 − 14 F/mm2). The addition of such a small amount of capacitance is below the level of resolution presently attainable in whole-cell recordings, and so capacitance measurements are limited to specialized synapses at which the number of synaptic vesicles fusing with the plasma membrane during a stimulus is sufficiently large to allow detection of the corresponding increase in capacitance. In neurons, capacitance measurements have been used to study synaptic release in the giant synaptic terminals of goldfish bipolar cells (von Gersdorff & Matthews, 1994), in retinal photoreceptors (Rieke & Schwartz, 1994), and in saccular hair cells (Parsons, Lenzi, Almers & Roberts, 1994). All of these cells have distinctive presynaptic structures (synaptic ribbons or synaptic bodies), which are
G. Matthews / Vision Research 39 (1999) 2469–2476
2471
Fig. 1. Schematic diagram of the changes in membrane capacitance that accompany exocytosis and endocytosis. Electrical access to the interior of the synaptic terminal is provided by means of a patch pipette in the whole-cell recording configuration. For simplicity, the series resistance through the patch pipette and the membrane resistance of the terminal are omitted in the equivalent circuits. (A) In the resting state, a synaptic vesicle is docked at the active zone, but there is no electrical continuity between the interior of the vesicle and the extracellular space. Therefore, the capacitance observed in the patch-clamp recording is the resting membrane capacitance of the terminal, Cm. (B) Upon depolarization, voltage-gated calcium channels open, allowing influx of calcium ions (black circles) into the synaptic terminal. Calcium binds to a calcium sensor molecule, causing opening of the putative fusion pore, which is assumed to be part of the molecular assembly of the docking complex. Release of neurotransmitter commences through the open fusion pore. At this point, the vesicle capacitance, C6, becomes electrically accessible through the conductance of the fusion pore, Gp. Because of the small size of the vesicle capacitance, virtually the entire capacitance of the vesicle is immediately observable even if the fusion pore has the conductance of a single ion channel. (C). Full fusion of the vesicle membrane with the plasma membrane is thought to occur as the docking/fusion complex disassembles. The opening of the fusion pore may also be reversible, producing so-called ‘kiss-and-run’ exocytosis. (D) The added membrane, including proteins associated with the synaptic vesicle, is retrieved via endocytosis, which may by mediated via formation of a clathrin coat (as shown) or by other mechanisms.
thought to be specializations that increase the available pool of releasable vesicles at the synaptic output site. Fig. 2 shows an example of the change in capacitance elicited by depolarization in the goldfish bipolar-cell terminal. The dihydropyridine-sensitive calcium current activated by the depolarization is shown in the inset. Depolarization was followed by a rapid jump in membrane capacitance of the terminal, after which capacitance returns to baseline, typically within a few seconds after the stimulus. The rapid rise in capacitance is an
index of the amount of exocytosis triggered by activation of the calcium channels, while the return of capacitance to baseline represents endocytotic retrieval of the added membrane.
4. Depletion and replenishment of vesicle pools In bipolar-cell terminals, the capacitance jump elicited by depolarization increases with the duration of the
2472
G. Matthews / Vision Research 39 (1999) 2469–2476
depolarizing pulse up to a maximum size of approximately 150 fF, which represents the maximum size of the readily releasable pool of vesicles whose release is triggered by activation of calcium channels. This pool is released within about 200 ms during depolarization that maximally activates calcium current (i.e. depolarization to about 0 mV). The 150-fF capacitance pool can be converted into a corresponding pool of 5700 synaptic vesicles from the measured average diameter of the vesicles in goldfish bipolar-cell terminals (29 nm; von Gersdorff, Vardi, Matthews & Sterling, 1996) and assuming a membrane capacitance of 10 fF/mm2. It is interesting that reconstructions of giant terminals from electron micrographs of serial thin sections suggest that there are about 6000 vesicles tethered to synaptic ribbons in the entire terminal (von Gersdorff et al., 1996). The close agreement between the morphologically defined pool of vesicles on the ribbons and the readily releasable pool estimated from capacitance measurements suggests that the entire population of vesicles on the ribbon can be released within about 200 ms during depolarization. This implies that the average fusion rate during depolarization is therefore about 30 000 vesicles/ s for the terminal as a whole, or about 500 vesicles/s at each of the approximately 55 synaptic ribbons in the terminal. The synaptic vesicles are tethered to the ribbon of goldfish bipolar cells in about five rows on each face of the ribbon, and the vesicles in the bottom row at the base of the ribbon are in close contact with the plasma membrane of the terminal. The close proximity suggests that this subgroup of vesicles might be docked and primed for release, and if so, the docked subgroup might be released more rapidly upon depolarization than the rest of the vesicles in the higher rows on the ribbon. Capacitance measurements have indeed revealed a small pool of vesicles that are released very
Fig. 2. Change in membrane capacitance elicited by depolarization in a bipolar cell synaptic terminal. The arrow indicates the timing of the depolarizing pulse, and the calcium current activated by the depolarization is shown on expanded time scale in the inset. The rapid jump in capacitance represents exocytosis, and the return to baseline capacitance over the next few seconds represents endocytosis of the added membrane.
Fig. 3. Two kinetically distinct components of exocytosis in bipolar cell synaptic terminals. (A) Changes in capacitance (upper traces) measured in response to brief depolarization given at the arrows. There was no change in access resistance (lower traces). (B) The average capacitance change evoked in five synaptic terminals by pulses of different durations. The solid line, fitted to the first four points, represents an exponential rise with a time constant of 1.5 ms and an amplitude of 33 fF. The point at the longest duration deviates significantly from the line because of the slower component of exocytosis that is more readily apparent in panel C. (C). The slow component of exocytosis is revealed more clearly with longer pulse durations. Data points show average capacitance responses from eight terminals. The solid line represents the best-fitting sum of two exponential components, with time constants of 1.6 and 350 ms and amplitudes of 31 and 81 fF.
rapidly upon the opening of voltage-gated calcium channels (Mennerick & Matthews, 1996). The size of this pool (approximately 1100 vesicles) is in close agreement with the expected number of vesicles at the bottom of the ribbon (von Gersdorff et al., 1996). As illustrated in Fig. 3, the time constant for the depletion of the fast component was 1.5 ms, while the remainder of the releasable pool was released more slowly during depolarization, with a time constant of about 300 ms (Mennerick & Matthews, 1996). Thus, fusion of the docked vesicles at the base of the ribbon may occur extremely rapidly as calcium channels open, while the
G. Matthews / Vision Research 39 (1999) 2469–2476
vesicles in rows more distant from the plasma membrane achieve fusion more slowly during sustained depolarization. Similar fast and slow components of release have also been observed in experiments in which glutamate release from goldfish bipolar-cell terminals was directly monitored, using electrical responses of closely opposed horizontal cells to detect release (Sakaba, Tachibana, Matsui & Minami, 1997). Tachibana & Okada, 1991 also used retinal horizontal cells as detectors of glutamate release from bipolar cells and found that the ‘postsynaptic’ response activated with a delay as brief as 1 ms after the onset of depolarization in the bipolar cell. Thus, direct measurements of glutamate release independently verify the features of synaptic exocytosis revealed by capacitance measurements. As described earlier, activation of calcium current for approximately 250 ms is sufficient to completely deplete the releasable pool of vesicles in bipolar-cell terminals. If a second depolarization is given soon after a depleting stimulus, the evoked capacitance response to the second stimulus is strongly depressed (von Gersdorff & Matthews, 1997). The depression recovers with a time constant of about 8 s, and full recovery of the response to the second stimulus requires about 20 s. This provides an estimate of the time for functional recovery of the release-ready pool after a single, strong stimulus. With weaker stimuli, the amount of release produced by a single stimulus is less, and the terminal is able to sustain exocytosis at a constant rate for up to several seconds (von Gersdorff & Matthews, 1997). Thus, the amount of initial exocytosis, and hence the degree of depletion of the releasable pool, determines the ability of the bipolar cell to continue to release transmitter during sustained or repetitive stimuli.
2473
much lower ( 1 mM) concentration of calcium achieved in the terminal as a whole during activation of calcium current. Although dialysis experiments give a qualitative indication of the level of calcium necessary to drive exocytosis, the slowness of dialysis via a whole-cell patch pipette does not allow conclusions about the calcium dependence of the rate of exocytosis. This type of information was obtained instead in experiments employing photorelease of caged calcium from calciumloaded DM-nitrophen (Heidelberger, Heinemann, Neher & Matthews, 1994). Fig. 4 illustrates sample results of such an experiment. With a bright flash of uncaging light, which elevated calcium concentration to 100 mM in the example of Fig. 4, capacitance increased rapidly after the flash. With dimmer flashes, which elevated calcium to lower levels, the rate of rise of the capacitance after the flash was slower, and at calcium concentrations below about 10–20 mM, no capacitance change was observed. By repeating such photolysis experiments at a variety of calcium concentrations, the relation between calcium concentration and the rate of exocytosis was obtained. The relation could be modeled assuming that synaptic vesicle fusion was triggered by cooperative binding of calcium to four binding sites, followed by a rapid fusion step. The overall half-saturating calcium concentration in the model was 190 mM, and the maximal rate constant of
5. Calcium dependence of exocytosis The ability to monitor exocytosis by means of capacitance changes in the bipolar-cell terminal offers the opportunity for direct study of the calcium-dependence of neurotransmitter release. One method of estimating the calcium dependence of exocytosis is to monitor changes in membrane capacitance while dialyzing the cell with solutions containing buffered levels of elevated calcium concentration (Augustine & Neher, 1992). In bipolar-cell synaptic terminals, dialysis with 1.4 mM calcium ( 10 times the resting level) failed to elicit an increase in membrane capacitance, but when calcium concentration was increased to50 mM, capacitance did increase during dialysis (von Gersdorff & Matthews, 1994). This suggests that exocytosis is triggered in bipolar-cell terminals by the high concentration of calcium expected within the microdomain of elevated calcium near open calcium channels (Chad & Eckert, 1984; Fogelson & Zucker, 1985), and not by the
Fig. 4. Capacitance change in a bipolar cell synaptic terminal evoked by photolytic release of caged calcium. The timing of the flash of uncaging light is shown by the arrow. The upper trace shows the capacitance response, with the initial rise shown on a faster time scale in the inset. The rise in calcium resulting from the release of caged calcium is shown in the lower trace on a slow time scale.
2474
G. Matthews / Vision Research 39 (1999) 2469–2476
about 3000 s − 1 was achieved as calcium concentration approached several hundred mM (Heidelberger et al., 1994; Heidelberger, 1998).
6. Calcium action potentials in bipolar cells Capacitance measurements suggest that large-terminal bipolar cells of goldfish retina are capable of very rapid release of a finite pool of neurotransmitter, at least when stimulated with depolarizations that are sufficient to maximally activate the voltage-dependent calcium channels of the synaptic terminal. However, the question arises whether such depolarizations would occur naturally in bipolar neurons. Intracellular recordings in intact retina suggest that they do. Saito, Kondo and Toyoda (1979) showed that light flashes elicited a rapid depolarization in on-type rod bipolar cells that peaked at a membrane potential of approximately − 10 to −15 mV. This spike-like depolarization lasted about 200 ms and is therefore comparable to the stimuli used in capacitance measurements to activate rapid exocytosis. The waveform of the intracellularly recorded light response of large-terminal, on-type bipolar cells (Saito et al., 1979; Saito & Kujiraoka, 1982) may in part reflect calcium action potentials triggered in response to synaptic depolarization. Large-terminal bipolar cells isolated from goldfish retina are electrically excitable and produce robust calcium action potentials (see Fig. 5) that originate from the voltage-gated calcium channels of the synaptic terminal (Zenisek & Matthews, 1998). The amplitude and duration of the calcium action potential are similar to the light responses observed by Saito et al. (1979) and are sufficient to release most of the pool of 6000 readily releasable vesicles inferred from capacitance measurements. Thus, an active response to depolarization, resulting from voltage-gated calcium channels in the synaptic terminal, may serve to amplify the depolarizing response to illumination. If such calcium action potentials do occur in the intact retina, then neurotransmitter release from the rod bipolar cell would be large, rapid, and transient. It is important to emphasize, however, that action potentials, and thus rapid and transient synaptic signaling, might occur only under restricted conditions (for example, in the highly darkadapted state). With smaller, graded depolarizations to illumination, capacitance measurements suggest that neurotransmitter release would occur at a slower rate but would be more sustained.
7. Conclusions Capacitance measurements provide a direct view of exocytosis in the giant synaptic terminals of goldfish bipolar neurons. New information has resulted about the
Fig. 5. Calcium action potentials are triggered by depolarization in goldfish bipolar cells. Each set of four traces shows the voltage responses recorded in current clamp in response to the indicated levels of depolarizing current, injected at the time indicated by the pulse below the upper set of traces. Depolarization above a threshold level evoked a regenerative action potential that was abolished by addition of 50 mM cadmium to the external solution (middle set of traces).
dynamics of releasable vesicle pools and about the calcium dependence of exocytosis at these ribbon synapses. The bipolar-cell synaptic terminal is in a central position to determine the efficacy of synaptic transfer between the outer and inner retina and is a prominent target for synaptic feedback in the retina. Several actions of this feedback on bipolar-cell terminals have already been identified. It seems likely that vesicle pool size, pool refilling rates, and other aspects of the release machinery of the synapse might be additional targets for neurotransmitters and neuromodulators. If so, capacitance measurements should prove valuable in unraveling these modulatory actions. The retina contains many different kinds of bipolar cell, and the conclusions summarized here are based on studies of a single type, selected because of its large synaptic terminal. Other types of bipolar neurons might differ in important physiological and pharmacological ways from the rod-dominant, on-type cells on which we have focused. For instance, cone-driven bipolar cells might not require the high amplification provided by calcium action potentials and so might lack this feature. Also, our results emphasize the transient nature of transmitter release from the large-terminal bipolar cells, but the retina must possess bipolar cells that are capa-
G. Matthews / Vision Research 39 (1999) 2469–2476
ble of sustained transmitter release to signal steady illumination levels. Clearly, characterization of the synaptic release properties of these other classes of bipolar cell will be required for a more complete understanding of retinal neurophysiology. References Augustine, G. J., & Neher, E. (1992). Calcium requirements for secretion in bovine chromaffin cells. Journal of Physiology (London), 450, 247 – 271. Ayoub, G. S., & Matthews, G. (1991). Conformational specificity of GABA binding to the presynaptic GABAA receptor. Neuroreport, 2, 809 – 811. Ayoub, G. S., & Matthews, G. (1992). Substance P modulates calcium current in retinal bipolar neurons. Vision Neuroscience, 8, 539 – 544. Chad, J. E., & Eckert, R. (1984). Calcium domains associated with individual channels can account for anomalous voltage relations of Ca-dependent responses. Biophysical Journal, 45, 993–999. Cutting, G. R., Lu, L., O’Hara, B. F., Kasch, L. M., MontroseRafizader, C., Donovan, D. M., Shimada, S., Antonarakis, S. E., Guggino, W. B., Uhl, G. R., & Kazazian, H. H. (1991). Cloning of the g-aminobutyric acid (GABA) r1 cDNA: a GABA receptor subunit highly expressed in the retina. Proceedings of the National Academy of the USA, 88, 2673–2677. Dowling, J. E., & Ehinger, B. (1978). The interplexiform cell system. I. Synapses of the dopaminergic neurons of the goldfish retina. Proceedings of the Royal Society of London Series B, 201, 7–26. Feigenspan, A., Wa¨ssle, H., & Bormann, J. (1993). Pharmacology of GABA receptor Cl − channels in rat retinal bipolar cells. Nature, 361, 159 – 162. Fernandez, J. M., Neher, E., & Gomperts, B. D. (1984). Capacitance measurements reveal stepwise fusion events in degranulating mast cells. Nature, 312, 453–455. Fidler Lim, N., Nowycky, M. C., & Bookman, R. J. (1990). Direct measurement of exocytosis and calcium currents in single vertebrate nerve terminals. Nature, 344, 449–451. Fogelson, A. L., & Zucker, R. S. (1985). Presynaptic calcium diffusion from various arrays of single channels. Biophysical Journal, 48, 1003 – 1017. Forsythe, I. D. (1994). Direct patch recording from identified presynaptic terminals mediateing glutamatergic EPSCs in the rat CNS, in vitro. Journal of Physiology (London), 479, 381–387. Fox, A. P., Nowycky, M. C., & Tsien, R. W. (1987). Kinetic and pharmacological properties distinguishing three types of calcium currents in chick sensory neurons. Journal of Physiology (London), 394, 149 – 172. Gillis, K. D. (1995). Techniques for membrane capacitance measurements. In B. Sakmann, & E. Neher, Single-Channel Recording (2, pp. 155 – 198). New York: Plenum. Heidelberger, R., & Matthews, G. (1991). Inhibition of calcium influx and calcium current by g-aminobutyric acid in single synaptic terminals. Proceedings of the National Academy of Science of the USA, 88, 7135–7139. Heidelberger, R., & Matthews, G. (1992). Calcium influx and calcium current in single synaptic terminals of goldfish retinal bipolar neurons. Journal of Physiology (London), 447, 235–256. Heidelberger, R., & Matthews, G. (1994). Dopamine enhances Ca2 + responses in synaptic terminals of retinal bipolar neurons. Neuroreport, 5, 729 – 732. Heidelberger, R., Heinemann, C., Neher, E., & Matthews, G. (1994). Calcium dependence of the rate of exocytosis in a synaptic terminal. Nature, 371, 513–515.
2475
Heidelberger, R. (1998). Adenosine triphosphate and the late steps in calcium-dependent exocytosis at a ribbon synapse. Journal of General Physiology, 111, 225 – 241. Ishida, A. T., Stell, W. K., & Lightfoot, D. A. (1980). Rod and cone inputs to bipolar cells in goldfish retina. Journal of Comparati6e Neurology, 191, 315 – 335. Kaneko, A., & Tachibana, M. (1985). A voltage-clamp analysis of membrane currents in solitary bipolar cells dissociated from Carassius auratus. Journal of Physiology (London), 385, 131–152. Lindau, M., & Neher, E. (1988). Patch-clamp techniques for timeresolved capacitance measurements in single cells. Pflu¨gers Arch., 411, 137 – 146. Lindau, M., Stuenkel, E. L., & Nordmann, J. J. (1992). Depolarization, intracellular calcium and exocytosis in single vertebrate nerve endings. Biophysics Journal, 61, 19 – 30. Marc, R. E., Stell, W. K., Bok, D., & Lam, D. M. K. (1978). GABAergic pathways in the goldfish retina. Journal of Comparati6e Neurology, 182, 221 – 246. Matthews, G., Ayoub, G. S., & Heidelberger, R. (1994). Presynaptic inhibition by GABA is mediated via two distinct GABA receptors with novel pharmacology. Journal of Neuroscience, 14, 1079 – 1090. Mennerick, S., & Matthews, G. (1996). Ultrafast exocytosis elicited by calcium current in synaptic terminals of retinal bipolar neurons. Neuron, 17, 1241 – 1249. Neher, E., & Marty, A. (1982). Discrete changes of cell membrane capacitance observed under conditions of enhanced secretion in bovine adrenal chromaffin cells. Proceedings of the National Academy of Science of the USA, 79, 6712 – 6716. Parsons, T. D., Lenzi, D., Almers, W., & Roberts, W. M. (1994). Calcium-triggered exocytosis and endocytosis in an isolated presynaptic cell: capacitance measurements in saccular hair cells. Neuron, 13, 875 – 883. Qian, H., & Dowling, J. E. (1993). Novel GABA responses from rod-driven retinal horizontal cells. Nature, 361, 162 – 164. Rieke, F., & Schwartz, E. A. (1994). A cGMP-gated current can control exocytosis at cone synapses. Neuron, 13, 863 – 873. Saito, T., Kondo, H., & Toyoda, J.-I. (1979). Ionic mechanisms of two types of ON-center bipolar cells in the carp retina. Journal of General Physiology, 73, 73 – 90. Saito, T., & Kujiraoka, T. (1982). Physiological and morphological identification of two types of on-center bipolar cells in the carp retina. Journal of Comparati6e Neurology, 205, 161 – 170. Sakaba, T., Tachibana, M., Matsui, K., & Minami, N. (1997). Two components of transmitter release in retinal bipolar cells: exocytosis and mobilization of synaptic vesicles. Neuroscience Research, 27, 357 – 370. Sherry, D. M., & Yazulla, S. (1993). Goldfish bipolar cells and axon terminal patterns: a golgi study. Journal of Comparati6e Neurology, 329, 188 – 200. Shimada, S., Cutting, G., & Uhl, G. R. (1992). g-Aminobutyric acid A or C receptor? g-Aminobutyric acid r1 receptor RNA induces bicuculline-, barbiturate-, and benzodiazepine-insensitive gaminobutyric acid responses in Xenopus oocytes. Molecular Pharmacology, 41, 683 – 687. Stanley, E. F., & Atrakchi, A. H. (1990). Calcium currents recorded from a vertebrate presynaptic nerve terminal are resistant to the dihydropyridine nifedipine. Proceedings of the National Academy of Science of the USA, 87, 9683 – 9687. Stanley, E. F. (1993). Single calcium channels and acetylcholine release at a presynaptic nerve terminal. Neuron, 11, 1007–1011. Stell, W. K. (1967). Structure and relationships of horizontal cells and photoreceptor-bipolar synaptic complexes in goldfish retina. American Journal of Anatomy, 121, 401 – 424. Tachibana, M., & Kaneko, A. (1987). g-Aminobutyric acid exerts a local inhibitory action on the axon terminal of bipolar cells: evidence for negative feedback from amacrine cells. Proceedings of the National Academy of Science of the USA, 84, 3501 –3505.
2476
G. Matthews / Vision Research 39 (1999) 2469–2476
Tachibana, M., & Okada, T. (1991). Release of endogenous excitatory amino acids from ON-type bipolar cells isolated from the goldfish retina. Journal of Neuroscience, 11, 2199–2208. Tachibana, M., Okada, T., Arimura, T., Kobayashi, K., & Piccolino, M. (1993). Dihydropyridine-sensitive calcium current mediates neurotransmitter release from bipolar cells of the goldfish retina. Journal of Neuroscience, 13, 2898–2909. Thomas, P., Suprenant, A., & Almers, W. (1990). Cytosolic Ca2 + , exocytosis, and endocytosis in single melanotrophs of the rat pituitary. Neuron, 5, 723–733. von Gersdorff, H., & Matthews, G. (1994). Dynamics of synaptic vesicle fusion and membrane retrieval in synaptic terminals. Nature, 367, 735 – 739. von Gersdorff, H., Vardi, E., Matthews, G., & Sterling, P. (1996).
.
Evidence that vesicles on the synaptic ribbon of retinal bipolar neurons can be rapidly released. Neuron, 16, 1221– 1227. von Gersdorff, H., & Matthews, G. (1997). Depletion and replenishment of vesicle pods at a ribbon-type synaptic terminal. Journal of Neuroscience, 17, 1919 – 1927. Yazulla, S., Studholme, K. M., & Wu, J.-Y. (1987). GABAergic input to the synaptic terminals of Mb-1 bipolar cells in the goldfish retina. Brain Research, 411, 400 – 405. Yazulla, S., & Zucker, C. L. (1988). Synaptic organization of dopaminergic interplexiform cells in the goldfish retina. Vision Neuroscience, 1, 13 – 29. Zenisek, D., & Matthews, G. (1998). Calcium action potentials in retinal bipolar neurons. Vision Neuroscience, 15, 69 – 75.
Synaptic mechanisms of bipolar cell terminals Gary Matthews * Department of Neurobiology and Beha6ior, State Uni6ersity of New York, Stony Brook, NY 11794 -5230, USA Received 8 May 1998; received in revised form 24 August 1998
Abstract Giant synaptic terminals of goldfish bipolar neurons allow direct studies of presynaptic mechanisms underlying neurotransmitter release and its modulation. Calcium influx via L-type calcium channels of the terminal triggers synaptic vesicle exocytosis, which can be monitored in isolated terminals by means of the associated changes in membrane capacitance. Information about the kinetics and calcium dependence of synaptic exocytosis has been obtained from capacitance measurements in these ribbon-type synaptic terminals. © 1999 Elsevier Science Ltd. All rights reserved. Keywords: Synaptic mechanisms; Terminals; Bipolar neurons
Studies of presynaptic mechanisms in the vertebrate central nervous system (CNS) are complicated by the small size of most CNS synaptic terminals. However, several notable exceptions allow direct electrical measurements to be made from single terminals, including giant caliciform terminals in the brainstem auditory system of the rat (Forsythe, 1994) and in the ciliary ganglion of the chick (Stanley & Atrakchi, 1990; Stanley, 1993) and large, bulbous terminals of certain types of bipolar neuron in goldfish retina (Kaneko & Tachibana, 1985; Heidelberger & Matthews, 1992). The latter are of interest in retinal neurobiology, and in this review, I will focus on our current understanding of presynaptic events that regulate neurotransmitter release in the giant terminals of goldfish bipolar neurons.
1. Calcium current of the synaptic terminal The large size (diameter:10 mm) of the synaptic terminal of some classes of bipolar cell in carp and goldfish (Stell, 1967; Ishida, Stell & Lightfoot, 1980; Saito & Kujiraoka, 1982; Sherry & Yazulla, 1993) permits direct patch-clamp recording of synaptic calcium current, either in intact bipolar cells isolated after
* Fax: + 1-516-632-6661. E-mail address: [email protected] (G. Matthews)
enzymatic digestion of goldfish retina or in single, isolated terminals. Consistent with their central role in neurotransmitter release, voltage-dependent calcium channels are located predominantly in the synaptic terminal of the bipolar cell (Tachibana & Okada, 1991; Heidelberger & Matthews, 1992), although there is also a low density of channels in the soma. The synaptic calcium current is sensitive to dihydropyridine calciumchannel agonists and antagonists, activates at a relatively positive voltage range, and shows sustained activation during prolonged depolarization (Heidelberger & Matthews, 1992). All of these characteristics are typical of L-type calcium channels (Fox, Nowycky & Tsien, 1987). The dihydropyridine-sensitive channels have been demonstrated to control glutamate release from the synaptic terminal (Tachibana, Okada, Arimura, Kobayashi & Piccolino, 1993). Thus, in goldfish bipolar cells, neurotransmitter release is controlled by L-type calcium channels.
2. Presynaptic inhibition at the bipolar-cell terminal The synaptic terminals of goldfish bipolar neurons receive extensive feedback synapses from amacrine cells (Marc, Stell, Bok & Lam, 1978; Yazulla, Studholme & Wu, 1987), which serve to modulate the synaptic output from the bipolar cells. Several functional effects of the synaptic feedback have been described, mediated by
0042-6989/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 2 - 6 9 8 9 ( 9 8 ) 0 0 2 4 9 - 1
2470
G. Matthews / Vision Research 39 (1999) 2469–2476
different neurotransmitters and neurotransmitter receptors. Many of the amacrine cells that target bipolar-cell terminals are GABAergic (Yazulla et al., 1987), and GABA acts on two distinct GABA receptors to produce presynaptic inhibition in the terminal. First, GABA activates a GABAA-like receptor and increases chloride conductance (Tachibana & Kaneko, 1987; Heidelberger & Matthews, 1991; Matthews, Ayoub & Heidelberger, 1994). Secondly, GABA produces GTPdependent inhibition of the calcium current of the bipolar-cell terminal via a GABAB-like receptor (Heidelberger & Matthews, 1991; Matthews et al., 1994). The GABA-activated chloride current hyperpolarizes the bipolar cell and thus inhibits calcium influx into the terminal via voltage-gated calcium channels (Heidelberger & Matthews, 1991; Matthews et al., 1994). GABA inhibits the calcium current by shifting its voltage dependence toward more positive potentials, thus tending to prevent activation of the current upon depolarization. Both actions of GABA in the terminal are mediated by receptors with unusual pharmacology. The receptor coupled to chloride conductance is insensitive to bicuculline and SR95531, which typically block GABAA receptors, while the action of GABA on calcium current is neither mimicked by the GABAB agonist baclofen nor blocked by the GABAB antagonist saclofen. Because of the bicuculline resistance of the GABAA-like receptors, they are commonly referred to in retinal research as GABAC receptors (Feigenspan, Wa¨ssle & Bormann, 1993; Qian & Dowling, 1993) to distinguish them from bicuculline-sensitive isoforms of the GABAA receptor. The bicuculline-resistant GABAA-like receptor probably corresponds to the retina-specific GABA receptor subunit, r1 (Cutting, Lu, O’Hara, Kasch, Montrose-Rafizader, Donovan, Shimada, Antonarakis, Guggino, Uhl & Kazazian, 1991; Shimada, Cutting & Uhl, 1992), which produces bicuculline-resistant GABA-activated chloride channels when expressed in Xenopus oocytes. Studies using conformationally locked analogs of GABA show that the bipolar-cell GABAAlike receptor prefers the trans (extended) isomer of GABA to the cis (folded) conformation (Ayoub & Matthews, 1991). In addition to GABA, other neurotransmitters affect the bipolar-cell terminal. Neuropeptides that are found in amacrine cells, including substance P and metenkephalin, produce GTP-dependent inhibition of calcium current by shifting the voltage dependence, similar to the action produced by GABA (Ayoub & Matthews, 1992). Goldfish bipolar-cell terminals also receive dopaminergic inputs (Dowling & Ehinger, 1978; Yazulla & Zucker, 1988), and dopamine was found to enhance the increase in intracellular calcium evoked in the terminal by activation of voltage-gated calcium channels (Heidelberger & Matthews, 1994). This action of do-
pamine was mimicked by forskolin and by a permeant analog of cAMP, suggesting that dopamine acts by elevating cAMP in the terminal. Taken together, the actions of GABA, dopamine, and neuropeptides demonstrate that synaptic feedback onto the bipolar cell terminal is capable of modulating transmitter release from the bipolar cell and thus the transfer of information from the outer to the inner retina.
3. Capacitance measurements of exocytosis When a secretory granule fuses with the plasma membrane during exocytosis, the surface area of the secreting cell increases. The increased surface area can be estimated by monitoring the associated change in electrical capacitance of the cell, as shown schematically in Fig. 1. Sensitive techniques for monitoring capacitance in real time in single cells are well-established (Lindau & Neher, 1988; Gillis, 1995) and have been used to gather information about exocytosis in various types of secretory cell, including adrenal chromaffin cells (Neher & Marty, 1982), mast cells (Fernandez, Neher & Gomperts, 1984), pituitary cells (Thomas, Suprenant & Almers, 1990), and neuroendocrine nerve terminals (Fidler Lim, Nowycky & Bookman, 1990; Lindau, Stuenkel & Nordmann, 1992). Neurotransmitter is released at synapses by a process of exocytosis, during which transmitter-containing synaptic vesicles fuse with the membrane of the presynaptic terminal in response to calcium influx via voltage-gated calcium channels. Therefore, capacitance measurements are potentially useful at synapses as an index of neurotransmitter release. However, a difficulty for capacitance measurements from synaptic terminals is the small size—and hence the small capacitance—of the synaptic vesicles at most synapses. A synaptic vesicle with a diameter of about 30 nm, which would be typical for glutamatergic synapses in the CNS, would have an expected capacitance of about 3× 10 − 17 F (assuming that capacitance of biological membranes is approximately 10 − 14 F/mm2). The addition of such a small amount of capacitance is below the level of resolution presently attainable in whole-cell recordings, and so capacitance measurements are limited to specialized synapses at which the number of synaptic vesicles fusing with the plasma membrane during a stimulus is sufficiently large to allow detection of the corresponding increase in capacitance. In neurons, capacitance measurements have been used to study synaptic release in the giant synaptic terminals of goldfish bipolar cells (von Gersdorff & Matthews, 1994), in retinal photoreceptors (Rieke & Schwartz, 1994), and in saccular hair cells (Parsons, Lenzi, Almers & Roberts, 1994). All of these cells have distinctive presynaptic structures (synaptic ribbons or synaptic bodies), which are
G. Matthews / Vision Research 39 (1999) 2469–2476
2471
Fig. 1. Schematic diagram of the changes in membrane capacitance that accompany exocytosis and endocytosis. Electrical access to the interior of the synaptic terminal is provided by means of a patch pipette in the whole-cell recording configuration. For simplicity, the series resistance through the patch pipette and the membrane resistance of the terminal are omitted in the equivalent circuits. (A) In the resting state, a synaptic vesicle is docked at the active zone, but there is no electrical continuity between the interior of the vesicle and the extracellular space. Therefore, the capacitance observed in the patch-clamp recording is the resting membrane capacitance of the terminal, Cm. (B) Upon depolarization, voltage-gated calcium channels open, allowing influx of calcium ions (black circles) into the synaptic terminal. Calcium binds to a calcium sensor molecule, causing opening of the putative fusion pore, which is assumed to be part of the molecular assembly of the docking complex. Release of neurotransmitter commences through the open fusion pore. At this point, the vesicle capacitance, C6, becomes electrically accessible through the conductance of the fusion pore, Gp. Because of the small size of the vesicle capacitance, virtually the entire capacitance of the vesicle is immediately observable even if the fusion pore has the conductance of a single ion channel. (C). Full fusion of the vesicle membrane with the plasma membrane is thought to occur as the docking/fusion complex disassembles. The opening of the fusion pore may also be reversible, producing so-called ‘kiss-and-run’ exocytosis. (D) The added membrane, including proteins associated with the synaptic vesicle, is retrieved via endocytosis, which may by mediated via formation of a clathrin coat (as shown) or by other mechanisms.
thought to be specializations that increase the available pool of releasable vesicles at the synaptic output site. Fig. 2 shows an example of the change in capacitance elicited by depolarization in the goldfish bipolar-cell terminal. The dihydropyridine-sensitive calcium current activated by the depolarization is shown in the inset. Depolarization was followed by a rapid jump in membrane capacitance of the terminal, after which capacitance returns to baseline, typically within a few seconds after the stimulus. The rapid rise in capacitance is an
index of the amount of exocytosis triggered by activation of the calcium channels, while the return of capacitance to baseline represents endocytotic retrieval of the added membrane.
4. Depletion and replenishment of vesicle pools In bipolar-cell terminals, the capacitance jump elicited by depolarization increases with the duration of the
2472
G. Matthews / Vision Research 39 (1999) 2469–2476
depolarizing pulse up to a maximum size of approximately 150 fF, which represents the maximum size of the readily releasable pool of vesicles whose release is triggered by activation of calcium channels. This pool is released within about 200 ms during depolarization that maximally activates calcium current (i.e. depolarization to about 0 mV). The 150-fF capacitance pool can be converted into a corresponding pool of 5700 synaptic vesicles from the measured average diameter of the vesicles in goldfish bipolar-cell terminals (29 nm; von Gersdorff, Vardi, Matthews & Sterling, 1996) and assuming a membrane capacitance of 10 fF/mm2. It is interesting that reconstructions of giant terminals from electron micrographs of serial thin sections suggest that there are about 6000 vesicles tethered to synaptic ribbons in the entire terminal (von Gersdorff et al., 1996). The close agreement between the morphologically defined pool of vesicles on the ribbons and the readily releasable pool estimated from capacitance measurements suggests that the entire population of vesicles on the ribbon can be released within about 200 ms during depolarization. This implies that the average fusion rate during depolarization is therefore about 30 000 vesicles/ s for the terminal as a whole, or about 500 vesicles/s at each of the approximately 55 synaptic ribbons in the terminal. The synaptic vesicles are tethered to the ribbon of goldfish bipolar cells in about five rows on each face of the ribbon, and the vesicles in the bottom row at the base of the ribbon are in close contact with the plasma membrane of the terminal. The close proximity suggests that this subgroup of vesicles might be docked and primed for release, and if so, the docked subgroup might be released more rapidly upon depolarization than the rest of the vesicles in the higher rows on the ribbon. Capacitance measurements have indeed revealed a small pool of vesicles that are released very
Fig. 2. Change in membrane capacitance elicited by depolarization in a bipolar cell synaptic terminal. The arrow indicates the timing of the depolarizing pulse, and the calcium current activated by the depolarization is shown on expanded time scale in the inset. The rapid jump in capacitance represents exocytosis, and the return to baseline capacitance over the next few seconds represents endocytosis of the added membrane.
Fig. 3. Two kinetically distinct components of exocytosis in bipolar cell synaptic terminals. (A) Changes in capacitance (upper traces) measured in response to brief depolarization given at the arrows. There was no change in access resistance (lower traces). (B) The average capacitance change evoked in five synaptic terminals by pulses of different durations. The solid line, fitted to the first four points, represents an exponential rise with a time constant of 1.5 ms and an amplitude of 33 fF. The point at the longest duration deviates significantly from the line because of the slower component of exocytosis that is more readily apparent in panel C. (C). The slow component of exocytosis is revealed more clearly with longer pulse durations. Data points show average capacitance responses from eight terminals. The solid line represents the best-fitting sum of two exponential components, with time constants of 1.6 and 350 ms and amplitudes of 31 and 81 fF.
rapidly upon the opening of voltage-gated calcium channels (Mennerick & Matthews, 1996). The size of this pool (approximately 1100 vesicles) is in close agreement with the expected number of vesicles at the bottom of the ribbon (von Gersdorff et al., 1996). As illustrated in Fig. 3, the time constant for the depletion of the fast component was 1.5 ms, while the remainder of the releasable pool was released more slowly during depolarization, with a time constant of about 300 ms (Mennerick & Matthews, 1996). Thus, fusion of the docked vesicles at the base of the ribbon may occur extremely rapidly as calcium channels open, while the
G. Matthews / Vision Research 39 (1999) 2469–2476
vesicles in rows more distant from the plasma membrane achieve fusion more slowly during sustained depolarization. Similar fast and slow components of release have also been observed in experiments in which glutamate release from goldfish bipolar-cell terminals was directly monitored, using electrical responses of closely opposed horizontal cells to detect release (Sakaba, Tachibana, Matsui & Minami, 1997). Tachibana & Okada, 1991 also used retinal horizontal cells as detectors of glutamate release from bipolar cells and found that the ‘postsynaptic’ response activated with a delay as brief as 1 ms after the onset of depolarization in the bipolar cell. Thus, direct measurements of glutamate release independently verify the features of synaptic exocytosis revealed by capacitance measurements. As described earlier, activation of calcium current for approximately 250 ms is sufficient to completely deplete the releasable pool of vesicles in bipolar-cell terminals. If a second depolarization is given soon after a depleting stimulus, the evoked capacitance response to the second stimulus is strongly depressed (von Gersdorff & Matthews, 1997). The depression recovers with a time constant of about 8 s, and full recovery of the response to the second stimulus requires about 20 s. This provides an estimate of the time for functional recovery of the release-ready pool after a single, strong stimulus. With weaker stimuli, the amount of release produced by a single stimulus is less, and the terminal is able to sustain exocytosis at a constant rate for up to several seconds (von Gersdorff & Matthews, 1997). Thus, the amount of initial exocytosis, and hence the degree of depletion of the releasable pool, determines the ability of the bipolar cell to continue to release transmitter during sustained or repetitive stimuli.
2473
much lower ( 1 mM) concentration of calcium achieved in the terminal as a whole during activation of calcium current. Although dialysis experiments give a qualitative indication of the level of calcium necessary to drive exocytosis, the slowness of dialysis via a whole-cell patch pipette does not allow conclusions about the calcium dependence of the rate of exocytosis. This type of information was obtained instead in experiments employing photorelease of caged calcium from calciumloaded DM-nitrophen (Heidelberger, Heinemann, Neher & Matthews, 1994). Fig. 4 illustrates sample results of such an experiment. With a bright flash of uncaging light, which elevated calcium concentration to 100 mM in the example of Fig. 4, capacitance increased rapidly after the flash. With dimmer flashes, which elevated calcium to lower levels, the rate of rise of the capacitance after the flash was slower, and at calcium concentrations below about 10–20 mM, no capacitance change was observed. By repeating such photolysis experiments at a variety of calcium concentrations, the relation between calcium concentration and the rate of exocytosis was obtained. The relation could be modeled assuming that synaptic vesicle fusion was triggered by cooperative binding of calcium to four binding sites, followed by a rapid fusion step. The overall half-saturating calcium concentration in the model was 190 mM, and the maximal rate constant of
5. Calcium dependence of exocytosis The ability to monitor exocytosis by means of capacitance changes in the bipolar-cell terminal offers the opportunity for direct study of the calcium-dependence of neurotransmitter release. One method of estimating the calcium dependence of exocytosis is to monitor changes in membrane capacitance while dialyzing the cell with solutions containing buffered levels of elevated calcium concentration (Augustine & Neher, 1992). In bipolar-cell synaptic terminals, dialysis with 1.4 mM calcium ( 10 times the resting level) failed to elicit an increase in membrane capacitance, but when calcium concentration was increased to50 mM, capacitance did increase during dialysis (von Gersdorff & Matthews, 1994). This suggests that exocytosis is triggered in bipolar-cell terminals by the high concentration of calcium expected within the microdomain of elevated calcium near open calcium channels (Chad & Eckert, 1984; Fogelson & Zucker, 1985), and not by the
Fig. 4. Capacitance change in a bipolar cell synaptic terminal evoked by photolytic release of caged calcium. The timing of the flash of uncaging light is shown by the arrow. The upper trace shows the capacitance response, with the initial rise shown on a faster time scale in the inset. The rise in calcium resulting from the release of caged calcium is shown in the lower trace on a slow time scale.
2474
G. Matthews / Vision Research 39 (1999) 2469–2476
about 3000 s − 1 was achieved as calcium concentration approached several hundred mM (Heidelberger et al., 1994; Heidelberger, 1998).
6. Calcium action potentials in bipolar cells Capacitance measurements suggest that large-terminal bipolar cells of goldfish retina are capable of very rapid release of a finite pool of neurotransmitter, at least when stimulated with depolarizations that are sufficient to maximally activate the voltage-dependent calcium channels of the synaptic terminal. However, the question arises whether such depolarizations would occur naturally in bipolar neurons. Intracellular recordings in intact retina suggest that they do. Saito, Kondo and Toyoda (1979) showed that light flashes elicited a rapid depolarization in on-type rod bipolar cells that peaked at a membrane potential of approximately − 10 to −15 mV. This spike-like depolarization lasted about 200 ms and is therefore comparable to the stimuli used in capacitance measurements to activate rapid exocytosis. The waveform of the intracellularly recorded light response of large-terminal, on-type bipolar cells (Saito et al., 1979; Saito & Kujiraoka, 1982) may in part reflect calcium action potentials triggered in response to synaptic depolarization. Large-terminal bipolar cells isolated from goldfish retina are electrically excitable and produce robust calcium action potentials (see Fig. 5) that originate from the voltage-gated calcium channels of the synaptic terminal (Zenisek & Matthews, 1998). The amplitude and duration of the calcium action potential are similar to the light responses observed by Saito et al. (1979) and are sufficient to release most of the pool of 6000 readily releasable vesicles inferred from capacitance measurements. Thus, an active response to depolarization, resulting from voltage-gated calcium channels in the synaptic terminal, may serve to amplify the depolarizing response to illumination. If such calcium action potentials do occur in the intact retina, then neurotransmitter release from the rod bipolar cell would be large, rapid, and transient. It is important to emphasize, however, that action potentials, and thus rapid and transient synaptic signaling, might occur only under restricted conditions (for example, in the highly darkadapted state). With smaller, graded depolarizations to illumination, capacitance measurements suggest that neurotransmitter release would occur at a slower rate but would be more sustained.
7. Conclusions Capacitance measurements provide a direct view of exocytosis in the giant synaptic terminals of goldfish bipolar neurons. New information has resulted about the
Fig. 5. Calcium action potentials are triggered by depolarization in goldfish bipolar cells. Each set of four traces shows the voltage responses recorded in current clamp in response to the indicated levels of depolarizing current, injected at the time indicated by the pulse below the upper set of traces. Depolarization above a threshold level evoked a regenerative action potential that was abolished by addition of 50 mM cadmium to the external solution (middle set of traces).
dynamics of releasable vesicle pools and about the calcium dependence of exocytosis at these ribbon synapses. The bipolar-cell synaptic terminal is in a central position to determine the efficacy of synaptic transfer between the outer and inner retina and is a prominent target for synaptic feedback in the retina. Several actions of this feedback on bipolar-cell terminals have already been identified. It seems likely that vesicle pool size, pool refilling rates, and other aspects of the release machinery of the synapse might be additional targets for neurotransmitters and neuromodulators. If so, capacitance measurements should prove valuable in unraveling these modulatory actions. The retina contains many different kinds of bipolar cell, and the conclusions summarized here are based on studies of a single type, selected because of its large synaptic terminal. Other types of bipolar neurons might differ in important physiological and pharmacological ways from the rod-dominant, on-type cells on which we have focused. For instance, cone-driven bipolar cells might not require the high amplification provided by calcium action potentials and so might lack this feature. Also, our results emphasize the transient nature of transmitter release from the large-terminal bipolar cells, but the retina must possess bipolar cells that are capa-
G. Matthews / Vision Research 39 (1999) 2469–2476
ble of sustained transmitter release to signal steady illumination levels. Clearly, characterization of the synaptic release properties of these other classes of bipolar cell will be required for a more complete understanding of retinal neurophysiology. References Augustine, G. J., & Neher, E. (1992). Calcium requirements for secretion in bovine chromaffin cells. Journal of Physiology (London), 450, 247 – 271. Ayoub, G. S., & Matthews, G. (1991). Conformational specificity of GABA binding to the presynaptic GABAA receptor. Neuroreport, 2, 809 – 811. Ayoub, G. S., & Matthews, G. (1992). Substance P modulates calcium current in retinal bipolar neurons. Vision Neuroscience, 8, 539 – 544. Chad, J. E., & Eckert, R. (1984). Calcium domains associated with individual channels can account for anomalous voltage relations of Ca-dependent responses. Biophysical Journal, 45, 993–999. Cutting, G. R., Lu, L., O’Hara, B. F., Kasch, L. M., MontroseRafizader, C., Donovan, D. M., Shimada, S., Antonarakis, S. E., Guggino, W. B., Uhl, G. R., & Kazazian, H. H. (1991). Cloning of the g-aminobutyric acid (GABA) r1 cDNA: a GABA receptor subunit highly expressed in the retina. Proceedings of the National Academy of the USA, 88, 2673–2677. Dowling, J. E., & Ehinger, B. (1978). The interplexiform cell system. I. Synapses of the dopaminergic neurons of the goldfish retina. Proceedings of the Royal Society of London Series B, 201, 7–26. Feigenspan, A., Wa¨ssle, H., & Bormann, J. (1993). Pharmacology of GABA receptor Cl − channels in rat retinal bipolar cells. Nature, 361, 159 – 162. Fernandez, J. M., Neher, E., & Gomperts, B. D. (1984). Capacitance measurements reveal stepwise fusion events in degranulating mast cells. Nature, 312, 453–455. Fidler Lim, N., Nowycky, M. C., & Bookman, R. J. (1990). Direct measurement of exocytosis and calcium currents in single vertebrate nerve terminals. Nature, 344, 449–451. Fogelson, A. L., & Zucker, R. S. (1985). Presynaptic calcium diffusion from various arrays of single channels. Biophysical Journal, 48, 1003 – 1017. Forsythe, I. D. (1994). Direct patch recording from identified presynaptic terminals mediateing glutamatergic EPSCs in the rat CNS, in vitro. Journal of Physiology (London), 479, 381–387. Fox, A. P., Nowycky, M. C., & Tsien, R. W. (1987). Kinetic and pharmacological properties distinguishing three types of calcium currents in chick sensory neurons. Journal of Physiology (London), 394, 149 – 172. Gillis, K. D. (1995). Techniques for membrane capacitance measurements. In B. Sakmann, & E. Neher, Single-Channel Recording (2, pp. 155 – 198). New York: Plenum. Heidelberger, R., & Matthews, G. (1991). Inhibition of calcium influx and calcium current by g-aminobutyric acid in single synaptic terminals. Proceedings of the National Academy of Science of the USA, 88, 7135–7139. Heidelberger, R., & Matthews, G. (1992). Calcium influx and calcium current in single synaptic terminals of goldfish retinal bipolar neurons. Journal of Physiology (London), 447, 235–256. Heidelberger, R., & Matthews, G. (1994). Dopamine enhances Ca2 + responses in synaptic terminals of retinal bipolar neurons. Neuroreport, 5, 729 – 732. Heidelberger, R., Heinemann, C., Neher, E., & Matthews, G. (1994). Calcium dependence of the rate of exocytosis in a synaptic terminal. Nature, 371, 513–515.
2475
Heidelberger, R. (1998). Adenosine triphosphate and the late steps in calcium-dependent exocytosis at a ribbon synapse. Journal of General Physiology, 111, 225 – 241. Ishida, A. T., Stell, W. K., & Lightfoot, D. A. (1980). Rod and cone inputs to bipolar cells in goldfish retina. Journal of Comparati6e Neurology, 191, 315 – 335. Kaneko, A., & Tachibana, M. (1985). A voltage-clamp analysis of membrane currents in solitary bipolar cells dissociated from Carassius auratus. Journal of Physiology (London), 385, 131–152. Lindau, M., & Neher, E. (1988). Patch-clamp techniques for timeresolved capacitance measurements in single cells. Pflu¨gers Arch., 411, 137 – 146. Lindau, M., Stuenkel, E. L., & Nordmann, J. J. (1992). Depolarization, intracellular calcium and exocytosis in single vertebrate nerve endings. Biophysics Journal, 61, 19 – 30. Marc, R. E., Stell, W. K., Bok, D., & Lam, D. M. K. (1978). GABAergic pathways in the goldfish retina. Journal of Comparati6e Neurology, 182, 221 – 246. Matthews, G., Ayoub, G. S., & Heidelberger, R. (1994). Presynaptic inhibition by GABA is mediated via two distinct GABA receptors with novel pharmacology. Journal of Neuroscience, 14, 1079 – 1090. Mennerick, S., & Matthews, G. (1996). Ultrafast exocytosis elicited by calcium current in synaptic terminals of retinal bipolar neurons. Neuron, 17, 1241 – 1249. Neher, E., & Marty, A. (1982). Discrete changes of cell membrane capacitance observed under conditions of enhanced secretion in bovine adrenal chromaffin cells. Proceedings of the National Academy of Science of the USA, 79, 6712 – 6716. Parsons, T. D., Lenzi, D., Almers, W., & Roberts, W. M. (1994). Calcium-triggered exocytosis and endocytosis in an isolated presynaptic cell: capacitance measurements in saccular hair cells. Neuron, 13, 875 – 883. Qian, H., & Dowling, J. E. (1993). Novel GABA responses from rod-driven retinal horizontal cells. Nature, 361, 162 – 164. Rieke, F., & Schwartz, E. A. (1994). A cGMP-gated current can control exocytosis at cone synapses. Neuron, 13, 863 – 873. Saito, T., Kondo, H., & Toyoda, J.-I. (1979). Ionic mechanisms of two types of ON-center bipolar cells in the carp retina. Journal of General Physiology, 73, 73 – 90. Saito, T., & Kujiraoka, T. (1982). Physiological and morphological identification of two types of on-center bipolar cells in the carp retina. Journal of Comparati6e Neurology, 205, 161 – 170. Sakaba, T., Tachibana, M., Matsui, K., & Minami, N. (1997). Two components of transmitter release in retinal bipolar cells: exocytosis and mobilization of synaptic vesicles. Neuroscience Research, 27, 357 – 370. Sherry, D. M., & Yazulla, S. (1993). Goldfish bipolar cells and axon terminal patterns: a golgi study. Journal of Comparati6e Neurology, 329, 188 – 200. Shimada, S., Cutting, G., & Uhl, G. R. (1992). g-Aminobutyric acid A or C receptor? g-Aminobutyric acid r1 receptor RNA induces bicuculline-, barbiturate-, and benzodiazepine-insensitive gaminobutyric acid responses in Xenopus oocytes. Molecular Pharmacology, 41, 683 – 687. Stanley, E. F., & Atrakchi, A. H. (1990). Calcium currents recorded from a vertebrate presynaptic nerve terminal are resistant to the dihydropyridine nifedipine. Proceedings of the National Academy of Science of the USA, 87, 9683 – 9687. Stanley, E. F. (1993). Single calcium channels and acetylcholine release at a presynaptic nerve terminal. Neuron, 11, 1007–1011. Stell, W. K. (1967). Structure and relationships of horizontal cells and photoreceptor-bipolar synaptic complexes in goldfish retina. American Journal of Anatomy, 121, 401 – 424. Tachibana, M., & Kaneko, A. (1987). g-Aminobutyric acid exerts a local inhibitory action on the axon terminal of bipolar cells: evidence for negative feedback from amacrine cells. Proceedings of the National Academy of Science of the USA, 84, 3501 –3505.
2476
G. Matthews / Vision Research 39 (1999) 2469–2476
Tachibana, M., & Okada, T. (1991). Release of endogenous excitatory amino acids from ON-type bipolar cells isolated from the goldfish retina. Journal of Neuroscience, 11, 2199–2208. Tachibana, M., Okada, T., Arimura, T., Kobayashi, K., & Piccolino, M. (1993). Dihydropyridine-sensitive calcium current mediates neurotransmitter release from bipolar cells of the goldfish retina. Journal of Neuroscience, 13, 2898–2909. Thomas, P., Suprenant, A., & Almers, W. (1990). Cytosolic Ca2 + , exocytosis, and endocytosis in single melanotrophs of the rat pituitary. Neuron, 5, 723–733. von Gersdorff, H., & Matthews, G. (1994). Dynamics of synaptic vesicle fusion and membrane retrieval in synaptic terminals. Nature, 367, 735 – 739. von Gersdorff, H., Vardi, E., Matthews, G., & Sterling, P. (1996).
.
Evidence that vesicles on the synaptic ribbon of retinal bipolar neurons can be rapidly released. Neuron, 16, 1221– 1227. von Gersdorff, H., & Matthews, G. (1997). Depletion and replenishment of vesicle pods at a ribbon-type synaptic terminal. Journal of Neuroscience, 17, 1919 – 1927. Yazulla, S., Studholme, K. M., & Wu, J.-Y. (1987). GABAergic input to the synaptic terminals of Mb-1 bipolar cells in the goldfish retina. Brain Research, 411, 400 – 405. Yazulla, S., & Zucker, C. L. (1988). Synaptic organization of dopaminergic interplexiform cells in the goldfish retina. Vision Neuroscience, 1, 13 – 29. Zenisek, D., & Matthews, G. (1998). Calcium action potentials in retinal bipolar neurons. Vision Neuroscience, 15, 69 – 75.