Characterization of spontaneous excitatory synaptic currents in newt retinal bipolar cells

Neuroscience Letters 271 (1999) 49±52

Characterization of spontaneous excitatory synaptic currents in newt retinal bipolar cells Fusao Kawai * Department of Information Physiology, National Institute for Physiological Sciences, Okazaki 444, Japan Department of Neuroscience, University of Pennsylvania, Philadelphia, PA 19104±6058, USA Received 11 June 1999; accepted 21 June 1999

Abstract The kinetics of glutamate concentration in the synaptic cleft is an important determinant of synaptic function. To elucidate peak concentration of glutamate released from a single vesicle in the cleft, spontaneous excitatory postsynaptic currents (sEPSCs) in Off-bipolar cells from the sliced newt retina were analyzed using whole-cell patch clamp recording and the computer simulation. The sEPSCs were blocked by an AMPA/kainate (KA) antagonist, 6-cyano-7nitroquinoxaline-2,3-dione (CNQX), and prolonged by cyclothiazide. However, an N-methyl-D-aspartate (NMDA) antagonist, D-2-amino-5-phosphonopentanoic acid (D-AP5), was ineffective. These suggest that sEPSCs in Off-bipolar cells are mediated exclusively by AMPA/KA receptors. sEPSCs simulated by a detailed kinetic model of AMPA receptor best approximated the data, when peak glutamate concentration was 10 mM. Therefore, it was concluded that peak concentration of glutamate released from a single vesicle would be elevated to approximately 10 mM at the newt Off-bipolar dendrite. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: AMPA; Glutamate; Desensitization; Off-bipolar cell; Retina; Newt

The Off-bipolar cell is a second-order retinal neuron, which responds to light with slow hyperpolarization graded with stimulus intensity. Photoreceptor's synaptic terminals contain vesicles and are believed to release glutamate as their neurotransmitter [3,11]. The kinetics of glutamate concentration in the synaptic cleft between cones and bipolar cells is an important determinant of synaptic function, but remains unknown. In this study, peak glutamate concentration at the Off-bipolar dendrite was estimated from spontaneous excitatory postsynaptic currents (sEPSCs) using the whole-cell patch clamp technique and the computer simulation with a detailed kinetic model of AMPA receptor. sEPSCs are generated by spontaneous release of single vesicles [2,7,8,15,17]. Slices from adult newt retina were cut at 200 mm [7,19] and viewed on an upright microscope with differential interference contrast optics (£40 water-immersion objective). A bipolar cell was identi®ed in the slice by its position and shape (a Landolt club and an axon), which was visualized by * c/o Dr. P. Sterling, 123 Anatomy/Chemistry Bldg., The Department of Neuroscience, University of Pennsylvania, Philadelphia, PA 19104±6058, USA. Tel.: 11-215-898-9228; fax: 11610-259-1096. E-mail address: [email protected] (F. Kawai)

Lucifer yellow staining. Membrane currents were recorded in the whole-cell con®guration. Data were low-pass ®ltered (4-pole Bessel type) with a cut-off frequency of 5 kHz and then digitized at 10 kHz by an analog-to-digital interface. All experiments were performed at room temperature (23± 258C) and in room light. The control Ringer's solution contained (in mM): NaCl, 110; KCl, 3.7; CaCl2, 3; HEPES, 2; glucose, 15. The solution was adjusted with NaOH to pH 7.4. Picrotoxin (100 mM) and strychnine (1 mM) were also added to block GABAergic and/or glycinergic inputs from horizontal cells and amacrine cells. The recording pipette contained (in mM): CsCl, 119; CaCl2, 1; EGTA 5; HEPES, 10. The solution was adjusted with CsOH to pH 7.4. Pipette resistance was about 10 MV. Test substances were applied either through the bath or via pressure ejection from a `puffer' pipette. Fig. 1A shows sEPSCs in a bipolar cell at various membrane potentials in the Ringer's solution containing picrotoxin and strychinine. sEPSCs were inward at negative potentials and reversed near 0 mV, indicating that the recorded cell was Off-type. Under application of 1 mM CoCl2, 100 mM picrotoxin, and 1 mM strychinine, membrane currents induced by puffer application of 100

0304-3940/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 9 9) 00 51 1- X

50

F. Kawai / Neuroscience Letters 271 (1999) 49±52

Fig. 1. sEPSCs in newt Off-bipolar cells are mediated by an AMPA/KA receptor. (A) Membrane currents recorded from an Off-bipolar cell at various holding potentials (Vh ˆ 240, 220, 0, 120, and 140 mV). (B) Membrane currents (Vh ˆ 250 mV) recorded from the same cell as A in the control solution (top trace) and solutions containing 100 mM CNQX (middle) or 10 mM cyclothiazide (CTZ, bottom). (C) Membrane currents (Vh ˆ 250 mV) recorded from another cell in the control solution (top trace) and 5 min after application of 1 mM dihydrokainate (DHK, bottom trace). All solutions contained 100 mM picrotoxin and 1 mM strychinine to suppress GABAergic and glycinergic inputs to the cells. Current traces are displaced arbitrarily.

mM glutamate and AMPA to the bipolar dendrite were also inward at negative potentials and reversed near 0 mV (data not shown). Fig. 2A shows amplitude histogram of sEPSCs recorded at 250 mV. The amplitude distribution was unimodal and the mean peak amplitude was 66 ^ 6 pA (n ˆ 100, four cells). To examine which types of glutamate receptor mediate sEPSCs in the Off-bipolar cells, various antagonists and a modulator of the ionotropic glutamate receptors were applied. 100 mM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), an AMPA/KA receptor antagonist, diminished sEPSCs at 250 mV (Fig. 1B). Cyclothiazide (CTZ) (10 mM), a desensitization blocker of AMPA receptors, increased the amplitude and the duration of sEPSCs (Fig. 1B). The amplitude distribution under application of CTZ was also unimodal and the mean amplitude was 112 ^ 13 pA for four cells. The decay time constant of sEPSCs in the control solution was 54 ^ 11 ms (n ˆ 100, four cells), and that in the solution containing CTZ was 87 ^ 23 ms (n ˆ 100, four cells). However, 100 mM D-2-amino-5phosphonopentanoic acid (D-AP5), an N-methyl-d-aspartate (NMDA) receptor antagonist, did not change sEPSCs signi®cantly (data not shown). Puffer application of 100 mM NMDA was also ineffective even in the Mg 21-free solution, although the compound evokes large currents in other preparations at the concentration employed [8,18]. These results suggest that sEPSCs in the Off-bipolar cells are meditated not by an NMDA receptor but by an AMPA/ KA receptor. This is consistent with previous reports using isolated Off-bipolar cell [6,13]. To investigate whether uptake or diffusion is responsible for the removal of glutamate from the synaptic cleft, the effect of dihydrokainate (DHK), an uptake blocker of gluta-

mate, on sEPSCs was examined. 1mM DHK almost diminished sEPSCs at 250 mV (Fig. 1C), suggesting that the transporter rather than diffusion is mainly responsible for the removal of glutamate from the cleft. This result is consistent with Gaal et al.'s observation [5]. In order to evaluate whether sEPSC time courses might be signi®cantly affected by electrotonic ®ltering at the bipolar dendrite, the correlation between amplitudes and decay time constants of sEPSCs was analyzed. With strong cable ®ltering, decay time constant would be expected to be slower for small amplitude events than for large events [7,16,17]. However, the decay time was not correlated signi®cantly with the peak amplitude of events (Fig. 2B). Thus, variability in amplitude cannot be attributed simply to cable ®ltering. Miniature EPSCs (mEPSCs) represent the events caused by the release of a single quantum of transmitter [4,17]. To determine the quantal content of the recorded sEPSCs, the amplitude distribution of sEPSCs in control was compared with that of mEPSCs recorded during 50 mM Cd 21 application. Cd 21 made no signi®cant difference on the amplitude distribution (data not shown). The mean amplitude was 66 ^ 6 pA (n ˆ 100) for sEPSCs and 64 ^ 7 pA (n ˆ 100) for mEPSCs. This suggests that every sEPSC in the bipolar cell represents the release of a single vesicle. When glutamate released from a single vesicle induces a mEPSC, its peak concentration at the Off-bipolar dendrite is unknown. This concentration can be estimated by analyzing the time course of sEPSCs in the bipolar cells. In the present study, the peak concentration was examined using a detailed kinetic model of AMPA receptor proposed by Raman and Trussell [9] (Fig. 3A; Table 1). Glutamate molecules, `A', bind to the receptor in a step-wise manner to form `CA2 or CA3'. Following isomerization of the liganded receptor `CA2 or CA3', the channel is converted to open-state `O1, O2 or O3', respectively. To test response kinetics of the model a simple pseudo-time course of glutamate concentration (Fig. 3B) was used as an input to the model. With various peak concentrations (3, 10, 30 mM), the decay kinetics of simulated sEPSCs were dramatically changed (Fig. 3C). The decay rate became faster with increased peak concentration.

Fig. 2. The amplitude and decay time constant distributions for the sEPSCs (Vh ˆ 250 mV). (A) Histogram of the sEPSC amplitude. Bath solution contained 100 mM picrotoxin and 1 mM strychinine. (B) Comparison of sEPSC decay time constant with amplitude. Decay time was plotted against the amplitude.

F. Kawai / Neuroscience Letters 271 (1999) 49±52

51

Fig. 3. The decay kinetics of sEPSC recorded in the control solution best approximated by the simulated sEPSC response to a peak glutamate concentration of 10 mM. (A) Markov model for an AMPA receptor [9]. `C', `O', and `D' show close, open, and desensitized states respectively, and `A' a glutamate molecule. Subscripts indicate the number of glutamate molecules bound to a receptor. (B) A normalized pseudo-time course of glutamate concentration. A decay time constant was 50 ms. (C) Simulated sEPSC responses to the pseudotime course of glutamate (shown in (B)) at various peak concentrations: 3 mM (solid line), 10 mM (dotted line), and 30 mM (dashed line). Simulated sEPSCs are equal to the total open probability (O1 1 O2 1 O3). Note that decay kinetics of simulated sEPSCs depends on peak glutamate concentration. (D) An averaged waveform of the 10 sEPSCs recorded in the solution containing CTZ. Its amplitude was normalized, and used for an input (i.e. glutamate concentration) to the AMPA receptor model. (E,F) Simulated sEPSC responses to glutamate at various peak concentrations and a normalized sEPSC recorded in the control solution (thick line). The control sEPSC is the average of 10 records. Dotted, solid and dashed lines in (E) are 1, 3 and 10 mM, respectively. Solid and dotted lines in (F) are 30 and 100 mM, respectively.

Since CTZ blocks the desensitization of AMPA receptors, one might expect that the time course of sEPSCs recorded in the solution containing CTZ is similar to that of the glutamate concentration at the Off-bipolar dendrite. Thus, the time course of sEPSC recorded in the CTZ solution was used as an input (i.e. the time course of glutamate concentration) to the AMPA receptor model. Because the decay kinetics of the AMPA receptor model depends on peak glutamate concentration (Fig. 3C), the sEPSC was normalized by its amplitude, and its peak value was adjusted to each concentration (1±100 mM). Fig. 3E,F show simulated sEPSC responses to the glutamate inputs of various peak concentrations. Both sEPSCs under the control (thick line in Fig. 3E,F) and CTZ (Fig. 3D) conditions are the average of 10 records near the mean amplitude in the sEPSC distribution. The decay time course of sEPSC recorded in the control solution was ®tted best, when peak concentration was 10 mM. The same value was estimated from the averages of 10 sEPSCs within the distributions of both low (10% minimum) and high (10% maximum) amplitudes, indicating that the estimation of glutamate concentration is robust. These results suggest that peak glutamate concentration for sEPSCs at the Off-bipolar dendrite would be approximately 10 mM. This value is roughly comparable with EC50 (18 mM) of AMPA receptors for horizontal cells in the salamander retina [5]. In contrast, the present value (~10 mM) is much

lower than previous data from another preparation [1]. Clements et al. [1] showed that peak glutamate concentration would be approximately 1 mM at hippocampal synapses, which are conventional (non-ribbon) synapses. Their value is a hundred times higher than the one in this study. Rao-Mirotznik et al. [10] estimated that the peak concentration at mammalian `On-type' bipolar dendrites would reach 0.5±1 mM, which is also much higher than the present estimate. The reasons for the discrepancy are not clear. One possibility might be due to the difference of the synaptic structure between the present preparation and theirs. In mammalian retina, the invaginating (On-type) bipolar cell dendrites end as the central elements in the cone terminal invaginations, whereas the ¯at (Off-type) bipolar cell dendrites terminate at basal junctions away from the invaginations [3,11]. In contrast, there is variability in receptor terminal organization in cold-blooded vertebrates. In the gold®sh retina, Stell et al. [14] reported that On-bipolar cells make a ribbon contact with cones, whereas Off-bipolar cells make a nonribbon contact. However, Saito et al. [12] reported that many Off-bipolar cells in the carp retina make a ribbon contact with cones and rods, although some Off-cells make a non-ribbon contact with cones. The origin of a large variance in sEPSC amplitude remains to be established. There are two obvious models that could account for the large variation in the sEPSC

52

F. Kawai / Neuroscience Letters 271 (1999) 49±52

Table 1 A description of an AMPA receptor model proposed by Raman and Trussell [9] (shown in Fig. 3A) a

[2]

d C/d t ˆ d CA/d t ˆ

[3]

23 * k1 * C * A 1 k21 * CA 21 * (k21 1 2 * k * A 1 kn) * CA 1 3 * k1 * C * A 1 2 * k22 * CA2 1 k2n * DA d CA2/d t ˆ 21 * (2 * k21 1 b1 1 b2 1 km) * CA2 1 2 * k2 * CA * A 1 a1 * O1 1 a2 * O2 1 k2m * DA2 d CA3/d t ˆ 21 * (3 * k23 1 b3) * CA3 1 k3 * DA2 * A 1 a3 * O3 d O1/d t ˆ 21 * a1 * O1 1 b1 * CA2 d O1/d t ˆ 21 * a2 * O2 1 b2 * CA2 d O3/d t ˆ 21 * a3 * O3 1 b3 * CA3 d DA/d t ˆ 21 * (k2n 1 2 * k2d * A) * DA 1 kn * CA 1 2 * k22d * DA2 d DA2/d t ˆ 21 * (2 * k22d 1 k2m 1 k3 * A) * DA2 1 2 * k2d * DA * A 1 km * CA2 1 3 * k23 * CA3 Open probability ˆ O1 1 O2 1 O3 k1 ˆ 10 7 k21 ˆ 300 k2 ˆ 10 7 k22 ˆ 3 * 10 5 7 k3 ˆ 10 k23 ˆ 220 k2d ˆ 10 7 k22d ˆ 519 k2n ˆ 300 kn ˆ 1000 km ˆ 2.7 * 10 4 k2m ˆ 14 a1 ˆ 3000 b1 ˆ 6 * 10 4 a2 ˆ 350 b2 ˆ 3000 b3 ˆ 6 a3 ˆ 2000

[4] [5]

[6] [7]

[8]

[9]

[10]

a

`C', `O', and `D' show close, open, and desensitized states, respectively, and `A' a glutamate concentration. Each rate constant (k, a, b) is equal to their parameters.

[11] [12]

amplitude [4,15]. Firstly, all sEPSC amplitude variation is due to variation in the amount of transmitter released from a single vesicle. Alternatively, all sEPSC amplitude variation is produced solely by variation in the number of postsynaptic receptor. Although the present analysis cannot distinguish these two models, the robust estimation of glutamate concentration over the entire range of the amplitude distribution of sEPSCs might support that the latter is the case. Collectively, the estimated peak glutamate concentration (10 mM) at the Off-bipolar dendrite is roughly comparable with the EC50 (~20 mM) of glutamate for AMPA receptors [5]. This result suggests that glutamate released from a single or a few vesicles can activate AMPA receptors near EC50 in the Off-bipolar cells. I thank Drs. Akimichi Kaneko, Peter Sterling, Robert Smith, Noga Vardi, Michael Freed and Jonathan Demb for their critical reading of the manuscript. [1] Clements, J.D., Lester, R.A., Tong, G., Jahr, C.E. and West-

[13] [14] [15]

[16]

[17]

[18] [19]

brook, G.L., The time course of glutamate in the synaptic cleft. Science, 258 (1992) 1498±1501. Diamond, J.S. and Jahr, C.E., Transporters buffer synaptically released glutamate on a submillisecond time scale. J. Neurosci., 17 (1997) 4672±4687. Dowling, J.E., The Retina, The Belknap Press of Harvard University Press, Cambridge, 1987. Frerking, M., Borges, S. and Wilson, M., Variation in GABA mini amplitude is the consequence of variation in transmitter concentration. Neuron, 15 (1995) 885±895. Gaal, L., Roska, B., Picaud, S.A., Wu, S.M., Marc, R. and Werblin, F.S., Postsynaptic response kinetics are controlled by a glutamate transporter at cone photoreceptors. J. Neurophysiol., 79 (1998) 190±196. Gilbertson, T.A., Scobey, R. and Wilson, M., Permeation of calcium ions through non-NMDA glutamate channels in retinal bipolar cells. Science, 251 (1991) 1613±1615. Matsui, K., Hosoi, N. and Tachibana, M., Excitatory synaptic transmission in the inner retina: paired recording of bipolar cells and neurons of the ganglion cell layer. J. Neurosci., 18 (1998) 4500±4510. Mittman, S., Taylor, W.R. and Copenhagen, D.R., Concomitant activation of two types of glutamate receptor mediates excitation of salamander retinal ganglion cells. J. Physiol., 428 (1990) 175±197. Raman, I.M. and Trussell, L.O., Concentration-jump analysis of voltage-dependent conductances activated by glutamate and kainate in neurons of the avian cochlear nucleus. Biophys. J., 69 (1995) 1868±1879. Rao-Mirotznik, R., Buchsbaum, G. and Sterling, P., Transmitter concentration at a three-dimensional synapse. J. Neurophysiol., 80 (1998) 3163±3172. Rodieck, R.W., The First Step in Seeing, Sinauer Associates, Sunderland, 1998. Saito, T., Kujiraoka, T., Yonaha, T. and Chino, Y., Reexamination of photoreceptor-bipolar connectivity patterns in carp retina: HRP-EM and Golgi-EM studies. J. Comp. Neurol., 236 (1985) 141±160. Sasaki, T. and Kaneko, A., l-Glutamate-induced responses in OFF-type bipolar cells of the cat retina. Vis. Res., 36 (1996) 787±795. Stell, W.K., Ishida, A.T. and Lightfoot, D.O., Structural basis for on-and off-center responses in retinal bipolar cells. Science, 198 (1977) 1269±1271. Taylor, W.R., Chen, E. and Copenhagen, D.R., Characterization of spontaneous excitatory synaptic currents in salamander retinal ganglion cells. J. Physiol., 486 (1995) 207± 221. Taylor, W.R., Mittman, S. and Copenhagen, D.R., Passive electrical cable properties and synaptic excitation of tiger salamander retinal ganglion cells. Vis. Neurosci., 13 (1996) 979±990. Tian, N., Hwang, T.N. and Copenhagen, D.R., Analysis of excitatory and inhibitory spontaneous synaptic activity in mouse retinal ganglion cells. J. Neurophysiol., 80 (1998) 1327±1340. Tong, G. and Jahr, C.E., Multivesicular release from excitatory synapses of cultured hippocampal neurons. Neuron, 12 (1994) 51±59. Werblin, F.S., Transmission along and between rods in the tiger salamander retina. J. Physiol., 280 (1978) 449±470.