Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2000 Dec 1;529(Pt 2):373–383. doi: 10.1111/j.1469-7793.2000.00373.x

Presynaptic 5-HT3 receptor-mediated modulation of synaptic GABA release in the mechanically dissociated rat amygdala neurons

Susumu Koyama *,†, Nozomu Matsumoto *, Chiharu Kubo †, Norio Akaike *
PMCID: PMC2270199  PMID: 11101647

Abstract

  1. Nystatin-perforated patch recordings were made from mechanically dissociated basolateral amygdala neurons with preserved intact native presynaptic nerve terminals to study the mechanism of 5-HT3 receptor-mediated serotonergic modulation of GABAergic inhibition.

  2. The specific 5-HT3 agonist mCPBG (1 μM) rapidly facilitated the frequency of GABAergic miniature inhibitory postsynaptic currents (mIPSCs) and this facilitation desensitized within 1 min. Tropisetron (30 nM), a specific 5-HT3 antagonist, blocked the mCPBG effect.

  3. mCPBG augmented mIPSC amplitude. However, no direct postsynaptic serotonergic currents were evoked by mCPBG. Neither GABA-evoked current amplitude nor the kinetics of individual GABAergic mIPSCs were affected by mCPBG. Therefore, the augmentation is unlikely to be due to postsynaptic effects evoked by mCPBG. At higher concentrations mCPBG produced shorter-duration facilitation of miniature events.

  4. While mCPBG increased the mIPSC frequency in calcium-containing solution with Cd2+, this increase was absent in Ca2+-free external solution. It appears that the Ca2+ influx through voltage-dependent calcium channels was not as crucial as that through 5-HT3 receptors for synaptic GABA release.

  5. When two pulses of mCPBG (each 1 μM, 1 min) were given, the response to the second pulse elicited full recovery when the interval between pulses was at least 9 min. Protein kinase A (PKA) activation by 8-Br-cAMP (300 μM) shortened and PKA inhibition by Rp-cAMP (100 μM) prolonged the recovery time. PKA activity did not affect the time course of fast desensitization.

  6. Our results suggest that a 5-HT3-specific agonist acts on presynaptic nerve terminals facilitating synaptic GABA release without postsynaptic effects. The facilitation requires calcium influx through presynaptic 5-HT3 receptors. PKA modulates the recovery process from desensitization of presynaptic 5-HT3 receptor-mediated regulation of synaptic GABA release.

Serotonin (5-HT) receptors constitute a complex receptor group with at least seven families that are in turn further divided into at least fourteen subtypes. Among these subtypes, 5-HT3 receptors are exclusively ligand-gated ion channels (Maricq et al. 1991), and thus are functionally distinct from the large number of serotonin subtypes which couple to various GTP-binding proteins (Boess & Martin, 1994). In the brain, 5-HT3 receptors are strongly expressed within the amygdala (Steward et al. 1993) and important serotonergic afferents arise from dorsal raphe nuclei to reach amygdala neurons (Ma et al. 1991). Furthermore, fast excitatory responses mediated via 5-HT3 receptors were reported in the amygdala neurons (Sugita et al. 1992). These results suggest that 5-HT3 receptors play a significant role in the amygdala. Since 5-HT3 receptors in the central nervous system appear to be localized on presynaptic nerve terminals or fibres (Kidd et al. 1993), it would appear that they contribute to the regulation of synaptic neurotransmitter release. 5-HT3 receptor-mediated serotonergic modulation of synaptic neurotransmission was reported in various regions in the brain (Ropert & Guy, 1991; Kawa, 1994; Piguet & Galvan, 1994; Ronde & Nichols, 1998; Zhou & Hablitz, 1999). In previous studies, the postsynaptic 5-HT3 receptor function including mechanistic information about channel kinetics, ion permeability, voltage dependency and modulation by the channel phosphorylation or dephosphorylation has been extensively investigated. One of the most remarkable and common characteristics of 5-HT3 receptors is their rapid onset and desensitization during sustained activation (Yakel & Jackson, 1988; Robertson & Bevan, 1991; Yang et al. 1992; Boddeke et al. 1996; Van Hooft et al. 1998). This characteristic of 5-HT3 receptors is thought to have great importance in synaptic regulation as well as postsynaptic effects. However, precise mechanisms as to how presynaptic 5-HT3 receptors regulate synaptic neurotransmitter release until now have not been well known, especially in the amygdala.

In order to study the basis of such dynamic 5-HT3 receptor-mediated regulation of neurotransmission, we have employed several experimental strategies. First, we have mechanically dissociated amygdala to yield isolated neurons with intact functioning presynaptic nerve terminals without exposure to enzymes as is the case with more conventional approaches (Koyama et al. 1999). Second, we rapidly applied drugs using the Y-tube system that enabled the external solutions bathing these dissociated neurons to be changed within 10 ms (Murase et al. 1990). In the present study, we investigated the characteristics of presynaptic 5-HT3 receptor regulation of synaptic GABA transmission.

METHODS

Preparations

Experimental protocols were approved by the Ethics Review Committee for Animal Experimentation of Kyushu University. Wistar rats 10–14 days old were decapitated under pentobarbital anaesthesia (50 mg kg−1i.p.). The brain was quickly removed and transversely sliced at a thickness of 500 μm (DM IRB, Leica). Slices were incubated in control medium (see below) saturated with 95 % O2 and 5 % CO2 at room temperature (20–22°C) for at least 1 h. For dissociation, slices were transferred into a 35 mm culture dish (Primaria 3801, Becton Dickinson) and the basolateral region of amygdala was identified under a binocular microscope (SMZ-1, Nikon). Neurons were then mechanically dissociated using a vibrating technique that yielded neurons with attached native presynaptic nerve terminals (Koyama et al. 1999). Note that the method required no enzyme treatment. Individual neurons settled and adhered to the bottom of the dish within about 20 min following dissociation.

Electrical measurement

All electrical measurements were performed using the nystatin-perforated patch recording method, allowing the intracellular signalling system to remain as intact as possible (Akaike & Harata, 1994). All voltage clamp recordings were made at a holding potential of −70 mV (CEZ-2300, Nihon Kohden, Tokyo, Japan). Patch pipettes were pulled from borosilicate capillary glass (1.5 mm o.d., 0.9 mm i.d.; G-1.5, Narishige, Tokyo, Japan) in two stages on a vertical pipette puller (PB-7, Narishige). The resistance of the recording electrode filled with pipette solution was 5–7 MΩ. Neurons were visualized under phase contrast on an inverted microscope (Diaphot, Nikon). Current and voltage were continuously monitored on an oscilloscope (Textronix 5111A, Sony) and recorded on a paper chart recorder (Recti-Horiz 8K, Nippomdenki San-ei, Tokyo, Japan) as well as on videotape (PCM-501 ES, Sony). Membrane currents were filtered at 1 kHz (E-3201A Decade Filter, NF Electronic Instruments, Tokyo, Japan) and digitized at 4 kHz. All experiments were performed at room temperature.

Data analysis

Miniature inhibitory postsynaptic currents (mIPSCs) were collected in pre-set epochs before, during and after each experimental condition. Inclusion criteria required a minimum event duration of 1 ms and a minimum event amplitude of 4 pA. Current events were counted and analysed using DETECTiVENT software (Ankri et al. 1994) and IGOR pro software (Wavemetrics, Lake Oswego, OR, USA). The amplitude of each mIPSC was measured from the initial point, not from the baseline, to the peak of the mIPSC (Fig. 2C). All traces were visually checked before analysing further to avoid measuring noises that could not be deleted by automated analysis. Numerical data are presented as the means ±s.e.m. Differences in amplitude and frequency distributions of mIPSCs were checked by non-parametric analysis (Kolmogorov-Smirnov (KS) test) with P < 0.05 been considered significant. Measured mean amplitude and event numbers of mIPSCs were normalized to those of control, and statistically analysed by Student’s two-tailed t test. All statistical analyses were performed using StatView software (SAS Institute Inc., Cary, NC, USA). Kinetics of individual mIPSCs, such as time to peak and decay time constant, were measured using Clampfit software (Axon Instruments, Foster City, CA, USA).

Figure 2. Analysis of mCPBG effects on mIPSCs.

Figure 2

Open in a new tab

A, amplitude histograms obtained from the same neuron as in Fig. 1A before (a), during (b) and after application of mCPBG (c). Bin width is 2.0 pA. B, cumulative probability plots of the amplitude distribution of mIPSCs before, during and after the application of mCPBG. C, schematic drawing of amplitude measurements. The amplitude was measured by peak value subtracted by value at onset time.

Solutions

The ionic composition of the incubation medium was (mM): 124 NaCl, 5 KCl, 1.2 KH2PO4, 24 NaHCO3, 2.4 CaCl2, 1.3 MgSO4 and 10 glucose; and bubbled with 95 % O2 and 5 % CO2. The standard external solution was composed of (mM): 150 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 10 glucose and 10 Hepes. Ca2+-free external solution contained (mM): 150 NaCl, 5 KCl, 3 MgCl2, 10 glucose, 10 Hepes and 2 EGTA. Cd2+-containing external solution was standard external solution with 100 μM CdCl2 added. The pH of these external solutions were adjusted to 7.4 with Tris-OH. In order to isolate spontaneous IPSCs, external solutions routinely contained 1 μM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and 10 μM dl-2-amino-5-phosphovaleric acid (dl-AP5) to block glutamatergic currents. To record miniature IPSCs, 300 nM tetrodotoxin (TTX) was added to block voltage-dependent Na+ channels. The ionic composition of the internal (patch-pipette) solution for the nystatin-perforated recordings was (mM): 20 N-methyl-D-glucamine methanesulfonate, 20 caesium methanesulfonate, 5 MgCl2, 100 CsCl and 10 Hepes with pH adjusted to 7.2 with Tris-OH. Nystatin was dissolved in acidified methanol at 10 mg ml−1. This stock solution was diluted with the internal solution just before use to a final concentration of 100–200 μg ml−1.

Drugs

Drugs used in the present study included adenosine 3′-5′-cyclic monophosphothioate Rp-isomer (Rp-cAMP), dl-AP5, bicuculline, 8-bromoadenosine 3′-5′-cyclic monophosphate (8-Br-cAMP), CNQX, EGTA, 1-(m-chlorophenyl)-biguanide (mCPBG) and nystatin from Sigma; TTX from Wako Pure Chemicals, Tokyo, Japan; and 3-tropanyl-indole-3-carboxylate hydrochloride (tropisetron) from Research Biochemicals.

RESULTS

Facilitation of the frequency of GABAergic miniature inhibitory postsynaptic currents (mIPSCs) via 5-HT3 receptors

All mechanically dissociated basolateral amygdala neurons examined exhibited spontaneous IPSCs in the presence of CNQX and dl-AP5 at a holding potential of −70 mV. mIPSCs were recorded after the application of TTX (300 nM; Fig. 1A). In a previous study of ours, the mIPSCs were mediated by GABAA receptors (Koyama et al. 1999). The specific 5-HT3 agonist mCPBG (Kilpatrick et al. 1987), applied at 1 μM, rapidly but transiently increased the GABAergic mIPSC frequency (Fig. 1A). A specific 5-HT3 antagonist, tropisetron (30 nM), blocked this facilitation (Fig. 1B), suggesting that the facilitation of mIPSC frequency was mediated by 5-HT3 receptors. To evaluate the duration of the transient facilitation, miniature synaptic events were plotted before, during and after the application of 1 μM mCPBG (Fig. 1C). A sustained 3 min application of mCPBG revealed a rapid increase of miniature synaptic events within the first 1 min and a quick return to the control level afterwards. Therefore, we estimated the effective duration of mCPBG (1 μM) as 1 min. mCPBG increased mIPSC frequency to 226.1 ± 14.1 % of control (P < 0.01, n = 32).

Figure 1. Modulation of GABAergic miniature inhibitory postsynaptic currents (mIPSCs) via 5-HT3 receptors.

Figure 1

Open in a new tab

A, spontaneous IPSCs in the presence of CNQX and dl-AP5. mIPSCs with further addition of TTX. Transient facilitation of mIPSC frequency by mCPBG was observed. B, responses of mIPSCs to mCPBG before, during, and after the treatment of tropisetron. C, time course of the number of miniature synaptic events before, during and after the application of mCPBG. Events in every 10 s duration are normalized to the averaged synaptic events in every 10 s duration from before and after controls. Each point is the mean of 10 neurons. The vertical bars show ±s.e.m.

Augmentation of the amplitude of mIPSCs via 5-HT3 receptors

In addition to the change of mIPSC frequency, mCPBG also augmented mIPSC amplitude. mCPBG at 1 μM increased the mean amplitude of mIPSCs to 118.2 ± 4.3 % of control (P < 0.01, n = 32). Amplitude histograms were obtained from the same neuron as in Fig. 1A (Fig. 2A). mIPSC amplitudes for controls before and after addition of mCPBG ranged from 0 to 148 pA, and from 0 to 163 pA, respectively (Fig. 2Aa and c). In contrast, mCPBG at 1 μM elicited a wider range of mIPSC amplitude distribution with a maximal amplitude of 264 pA (Fig. 2Ab). This observation was also supported in the cumulative probability plots. mCPBG shifted the curve toward the right (Fig. 2B).

No postsynaptic effects by mCPBG

Since the amygdala neurons were reported to show postsynaptic 5-HT3 receptor-mediated responses (Sugita et al. 1992), we investigated whether postsynaptic responses were evoked by mCPBG. A submaximal dose of bicuculline (10 μM) completely eliminated the GABAergic mIPSCs (Fig. 3Aa). In this condition, no postsynaptic serotonergic currents were evoked by mCPBG in the same neuron in which mCPBG prominently increased the GABAergic synaptic events (Fig. 3Ab). Similar results were obtained in seven other neurons. To exclude possible mCPBG effects on postsynaptic GABAA receptors, we simultaneously applied GABA (10 μM) and mCPBG (1 μM) to the neuron (Fig. 3B). mCPBG had no effect on GABA-evoked postsynaptic currents. Similar results were obtained in five other neurons. Finally, the kinetics of 40 individual miniature synaptic currents before, during and after application of mCPBG from the same neuron was analysed. There were no differences between them in the time to peak and the decay time (Fig. 3C). All these findings suggest that mCPBG did not affect postsynaptic GABAA receptors.

Figure 3. No postsynaptic effects evoked by mCPBG.

Figure 3

Open in a new tab

Aa, complete elimination of mIPSCs by bicuculline. Responses of mIPSCs to mCPBG before (b) and during (c) bicuculline application. B, simultaneous application of GABA and mCPBG. C, kinetics of individual mIPSCs before, during and after the application of mCPBG. Representative miniature events before, during and after the application of mCPBG are superimposed. Scale bar is 200 ms. Time to peak (▪): before, 1.48 ± 0.11 ms; mCPBG, 1.47 ± 0.07 ms; after, 1.42 ± 0.10 ms. Decay time fitted to a single exponential function (□): before, 16.50 ± 1.23 ms; mCPBG, 16.18 ± 0.80 ms; after, 16.83 ± 1.06 ms. Each column is the mean of 40 events. The vertical bars show +s.e.m.

Concentration effects of mCPBG on mIPSCs

We next investigated effects of mCPBG concentration on mIPSCs. While mCPBG at 10 nM did not facilitate mIPSC frequency, at higher concentrations it did exhibit facilitation (Fig. 4A). There was a tendency for mCPBG at higher concentrations to elicit prompter desensitization of the facilitation of miniature synaptic events (Fig. 4Ab-d). In addition, mCPBG at higher concentrations augmented mIPSC amplitude more (Fig. 4A, insets). While mCPBG at 100 nM had no effect on the amplitude distribution of mIPSCs (KS test, P = 0.37, 149 events for control and 209 events), mCPBG at 1 μM shifted the curve toward the right (KS test, P < 0.05, 148 events for control and 273 events). mCPBG at 10 μM also shifted the curve rightward (KS test, P < 0.001, 147 events for control and 165 events). Therefore a higher mCPBG concentration evoked a shorter duration of facilitated miniature events accompanied by current augmentation.

Figure 4. Effects of mCPBG concentration on mIPSC amplitude and frequency.

Figure 4

Open in a new tab

A, time courses of the number of miniature synaptic events before and during mCPBG application at various concentrations. Number of events in every 2 s period are plotted. Insets are the cumulative amplitude distribution plots obtained from the same neurons. B, relationship between mCPBG concentration and mIPSC frequency. All mCPBG-induced mIPSC frequencies are normalized to those of control. Normalized ratio: 1 nM, 102.5 ± 7.7; 10 nM, 108.3 ± 8.4; 100 nM, 176.0 ± 25.4; 1 μM, 227 ± 16.5; 10 μM, 182 ± 3.2. Each point is the mean of 4 neurons. The vertical bars show ±s.e.m. Asterisks represent the statistically significant differences (**P < 0.01, Student’s two-tailed t test).

mCPBG increased mIPSC frequency in a dose-dependent manner, but paradoxically, mCPBG at 10 μM facilitated mIPSC frequency less than at 1 μM (Fig. 4B). There are two possibilities for the interpretation of this observation. First, the solution exchange rate (10 ms) could be slower than entry into desensitization producing an artificial reduction in the peak response. Second, the summation of events without any delay can produce a single large event, thus showing apparent reduction in event frequency while peak amplitude is augmented. In the following experiments, we used 1 μM mCPBG because this concentration prominently and stably increased event numbers of the miniature synaptic currents.

Ca2+ permeability of presynaptic 5-HT3 receptors

An increase in Ca2+ influx into nerve terminals is necessary for the acceleration of synaptic neurotransmission and this mechanism could also play an important role in the facilitation of synaptic GABA release via presynaptic 5-HT3 receptors in the present study. 5-HT3 receptors are considered to be relatively non-selective cation channels (Maricq et al. 1991). In a previous study, cations permeating through 5-HT3 receptors depolarized the membrane, opened voltage-dependent Ca2+ channels (VDCCs), and then increased Ca2+ influx (Takenouchi & Munekata, 1998), suggesting a significant cooperation between 5-HT3 receptors and VDCCs. Therefore, we investigated whether this cooperation was essential for the mechanism of facilitation of synaptic GABA release by 5-HT3 activation. In the presence of 100 μM Cd2+ to block VDCCs, mIPSC frequency decreased (by 62.6 ± 6.7 % of control, P < 0.05, n = 5) and the mean amplitude of mIPSCs was attenuated (66.2 ± 3.5 % of control, P < 0.05, n = 5). In this condition, mCPBG still facilitated mIPSC frequency (P < 0.05, n = 4) (Fig. 5A, left and 5Ba, inset). The cumulative probability plots of the distribution of mIPSC inter-event interval show that mCPBG shifted the curve to the left (KS test, P < 0.01, 72 events for control and 120 events for mCPBG). In Ca2+-free external solution, mIPSC frequency decreased (55.6 ± 7.8 % of control, P < 0.05, n = 5) while the mean amplitude of mIPSCs was not affected (89.6 ± 7.1 % of control, n = 5). Under these conditions, mCPBG failed to facilitate mIPSC frequency (Fig. 5A, right and 5Bb, inset graph). The cumulative probability plots show that mCPBG had no effect on the distribution of mIPSC inter-event interval in Ca2+-free external solution (KS test, P > 1.00, 143 events for control and 159 events for mCPBG). These results suggest that the cooperation between 5-HT3 receptors and VDCCs was not a primary trigger and that Ca2+ influx through presynaptic 5-HT3 receptors mediated the increase in GABA release from nerve terminals.

Figure 5. Ca2+ permeability of presynaptic 5-HT3 receptors.

Figure 5

Open in a new tab

A, responses of mIPSCs to mCPBG in the presence of Cd2+ (left) or in Ca2+-free external solution (right) from the same neuron. B, cumulative probability plots of the inter-event interval of mIPSCs before, during and after application of mCPBG in the presence of Cd2+(a) and in Ca2+-free external solution (b), from the upper traces. Inset graphs, all mIPSC frequencies are normalized to those of control. Each column is the mean of 4–5 neurons. The vertical bars show +s.e.m. Asterisk represents the statistically significant differences (*P < 0.05, Student’s two-tailed t test).

Fast desensitization of presynaptic 5-HT3 receptor regulation of synaptic GABA release

Fast desensitization of 5-HT3 receptors by the sustained application of their agonists has commonly been reported (Yakel & Jackson, 1988; Robertson & Bevan, 1991; Yang et al. 1992; Boddeke et al. 1996; Van Hooft et al. 1998) and the desensitization of 5-HT3 responses is accelerated by cAMP-dependent protein kinase (Yakel & Jackson, 1988). These characteristics of 5-HT3 receptors could reflect synaptic modulation regulated by intracellular second messengers. Therefore, we tested whether protein kinase A (PKA) could affect the fast desensitization of presynaptic 5-HT3 receptor modulation of synaptic GABA release, using a specific PKA inhibitor, Rp-cAMP (Van Haastert et al. 1984), and a specific PKA activator, 8-Br-cAMP (Fletcher et al. 1986). First, we examined the effects of the PKA activator and inhibitor on mIPSCs. In the presence of Rp-cAMP (100 μM) for 30 min, mIPSC amplitude was 102.4 ± 13.3 % of control (n = 6) and mIPSC frequency was 114.2 ± 5.7 % of control (n = 6). In the presence of 8-Br-cAMP (300 μM) for 20 min, mIPSC amplitude did not change (106.3 ± 12.8 % of control, n = 5) but mIPSC frequency increased (201.3 ± 27.7 % of control, P < 0.05, n = 5). These results suggest that PKA activation accelerated synaptic GABA transmission.

In the presence of Rp-cAMP, mCPBG also transiently facilitated mIPSC frequency (Fig. 6Aa). The miniature synaptic events from the peak facilitation are plotted in Fig. 6Ab. Before and during Rp-cAMP treatment, the increased event numbers returned to control level within 20 s. In the presence of 8-Br-cAMP, mCPBG transiently facilitated mIPSC frequency (Fig. 6Ba). Before and during 8-Br-cAMP treatment, the increased event numbers returned to control level within 20 s (Fig. 6Bb). These results suggest that PKA did not regulate the fast desensitization of presynaptic 5-HT3 receptor modulation of synaptic GABA release.

Figure 6. Fast desensitization of presynaptic 5-HT3 receptor-mediated regulation.

Figure 6

Open in a new tab

Aa, responses of mIPSCs to mCPBG before and during treatment with Rp-cAMP. Ab, time course of the number of miniature synaptic events from the peak of the facilitation (^, control; ▪, Rp-cAMP treatment). Number of events in every 2 s period are normalized to those of the averaged synaptic events in 2 s periods from control mIPSCs. Each point is the mean of 6 neurons. Ba, responses of mIPSCs to mCPBG before and during the treatment with 8-Br-cAMP. Bb, time course of the number of miniature synaptic events from the peak of the facilitation (^, control; ▴, 8-Br-cAMP treatment). Each point is similarly plotted as the experiment using Rp-cAMP. Each point is the mean of 5 neurons. The vertical bars show ±s.e.m.

Recovery of presynaptic 5-HT3 receptor regulation of synaptic GABA release

5-HT3 receptor-mediated responses are decreased by the repetitive application of serotonin. If the interval between application of agonists was long enough, 5-HT3 receptor responses recovered (Boddeke et al. 1996). In addition, intracellular second messengers are known to regulate this recovery process (Robertson & Bevan, 1991; Boddeke et al. 1996). Therefore we examined the recovery time of presynaptic 5-HT3 receptor-mediated facilitation of synaptic GABA release. Two mCPBG pulses applied at 9 min intervals or longer allowed full recovery from presynaptic 5-HT3 receptor desensitization (Fig. 7Aa and B). We next examined whether manipulation of PKA activity substantially affected the recovery time. In the presence of Rp-cAMP (100 μM), the interval between two mCPBG pulses needed for full recovery was markedly prolonged by up to 29 min (Fig. 7Ab and B). In contrast, 8-Br-cAMP (300 μM) treatment shortened recovery times so that only about 2 min was enough for full recovery (Fig. 7Ac and B). These results suggest that the phosphorylation state with PKA determined the recovery of 5-HT3 receptor-mediated presynaptic modulation and that PKA inhibition prolonged and PKA activation shortened the recovery time.

Figure 7. Recovery of presynaptic 5-HT3 receptor-mediated regulation.

Figure 7

Open in a new tab

A, responses of mIPSCs to a two-pulse application of mCPBG at 1 μM for 1 min for each pulse. The traces show responses of mIPSC frequency to the first (P1) and second (P2) mCPBG pulses at various inter-pulse intervals in control (a) and in the presence of Rp-cAMP (b) and 8-Br-cAMP (c). B, relationship between the interval between two mCPBG pulses and the recovery ratio. The responses of mIPSC frequency to the second mCPBG pulse (P2) are normalized to those to the first pulse (P1). Control (^): 1 min, 53.8 ± 3.7; 2 min, 58.5 ± 5.9; 4 min, 75.0 ± 8.6; 6.5 min, 96.0 ± 8.2; 9 min, 99.8 ± 3.3; 11 min, 100.3 ± 7.9; 19 min, 100.4 ± 3.4; 29 min, 98.6 ± 5.8. In the presence of Rp-cAMP (▪): 9 min, 58.0 ± 5.0; 19 min, 58.2 ± 3.1; 29 min, 96.7 ± 11.7. In the presence of 8-Br-cAMP (▴): 1 min, 70.0 ± 4.0; 2 min, 104.0 ± 3.1; 4 min, 104.0 ± 13.2; 6.5 min, 100.0 ± 7.9; 9 min, 106.0 ± 17.0. Each point is the mean of 4–6 neurons. The vertical bars show ±s.e.m. Asterisks represent the statistically significant differences (*P < 0.05, **P < 0.01, Student’s two-tailed t test).

DISCUSSION

In the present study, the mechanism of 5-HT3 receptor-mediated serotonergic modulation of GABAergic inhibition was investigated using the mechanically dissociated basolateral amygdala neurons with functioning native GABAergic nerve terminals. Activation of presynaptic 5-HT3 receptors transiently facilitated mIPSC frequency. Higher concentrations of mCPBG exhibited a shorter duration of this synaptic facilitation together with augmentation of mIPSC amplitude. However, a specific 5-HT3 agonist did not affect postsynaptic GABAA receptors. Ca2+ influx into nerve terminals through presynaptic 5-HT3 receptors was required for the facilitation of miniature synaptic GABA release. The facilitation of mIPSC frequency by mCPBG desensitized within 1 min and this desensitization process was not affected by protein kinase A (PKA). More than 9 min was enough for full recovery of 5-HT3 receptor-mediated presynaptic modulation. This recovery process was shortened by activation of PKA and prolonged by inhibition of PKA.

No postsynaptic effects evoked by mCPBG

Postsynaptic 5-HT3 receptor-mediated fast responses were reported by Sugita et al. (1992) in the amygdala. In our study, in contrast, mCPBG (1 μM) evoked no direct postsynaptic responses (Fig. 3A). If postsynaptic responses were evoked by activation of 5-HT3 receptors, they would be recorded as inward currents under our conditions. It is unlikely that the concentration of mCPBG of 1 μM was too low to evoke postsynaptic responses because mCPBG at concentrations higher than 100 nM evoked a response in presynaptic terminals on the dissociated neurons. While 5-HT3 receptors generally show voltage dependency, inward currents were recorded at a holding potential of −70 mV with the intra- and extracellular solutions used in this study. This discrepancy could be explained by differences between the nuclei, the basolateral and the lateral amygdalas, or the histological specificity of postsynaptic neurons, as it has been reported previously that only some types of neurons exhibited postsynaptic serotonergic currents via 5-HT3 receptors (Kawa, 1994; Zhou & Hablitz, 1999). Sugita et al. (1992) also reported that mCPBG, a more selective 5-HT3 agonist than 5-HT, did not evoke a postsynaptic response. Thus, there is a possibility that other subtypes of serotonin receptors participated in the postsynaptic responses in the amygdala. In the central nervous system, 5-HT3 receptors are located on nerve terminals (Kidd et al. 1993). Furthermore, 5-HT3 receptors are expressed on GABAergic interneurons in the amygdala (Morales & Bloom, 1997). Therefore, it is likely that 5-HT3 receptors are located on GABAergic nerve terminals in the basolateral amygdala, modulating synaptic GABA transmission.

Facilitation of synaptic GABA release by presynaptic 5-HT3 receptor activation

In addition to producing no direct postsynaptic effects, mCPBG did not exhibit any modulation of postsynaptic GABAA receptors (Fig. 3B and C), suggesting a presynaptic action site of the 5-HT3 agonist. However, mCPBG at 1 μM significantly augmented mIPSC amplitude as well as dramatically increasing mIPSC frequency. During synaptic facilitation, we measured the mIPSC peak amplitude minus the current value at the onset time (Fig. 2C). While our measurement is likely to underestimate the current amplitude during a synaptic burst, mCPBG still produced a statistically significant increase. When the concentration of 5-HT3 agonist increased, the duration of the synaptic facilitation became shorter and, at the same time, the amplitude of miniature currents were more augmented (Fig. 4). This augmentation of mIPSC amplitude could be explained if the 5-HT3 agonist at a higher concentration tends to activate multiple nerve terminals simultaneously to release GABA, resulting in a summation of miniature postsynaptic currents.

Ca2+ permeability of presynaptic 5-HT3 receptors

In a previous calcium-imaging study, Ca2+ influx through VDCCs activated by cation influx through 5-HT3 receptors increased intracellular Ca2+ concentration, suggesting that this is significant for the acceleration of neurotransmission (Takenouchi & Munekata, 1998). However, in the present study, this coupling between 5-HT3 receptors and VDCCs was not primary but Ca2+ influx through presynaptic 5-HT3 receptors itself evoked the increase in the miniature synaptic events (Fig. 5). From the viewpoint of the receptor structure, the Ca2+ permeability of 5-HT3 receptors depends on the combination of their receptor subunits, so that a homomeric 5-HT3 receptors elicit relatively high Ca2+ permeability compared with heteromeric ones (Brown et al. 1998; Davies et al. 1999). Therefore, presynaptic 5-HT3 receptors in the basolateral amygdala might have homomeric receptor structure. Since neuronal 5-HT3 receptors are thought to have low Ca2+ permeability, 5-HT3 receptors on nerve terminals might be different from postsynaptic ones as Ronde & Nichols (1998) suggested. While Ca2+ permeation through 5-HT3 receptors interacts with the calcium/calmodulin-dependent process (Riad et al. 1994), it is unclear whether Ca2+ influx through the 5-HT3 receptors directly evokes exocytosis of synaptic vesicles or whether it requires the contribution of further intracellular signal pathways.

Fast desensitization and recovery of presynaptic 5-HT3 receptor regulation

Combining our experimental approach of rapid Y-tube drug application with patch recording from these mechanically dissociated neurons offers a high-resolution method to assay modulation by presynaptic 5-HT3 receptors. We precisely analysed the facilitation of mIPSC frequency by the agonist under several experimental conditions and examined the regulation of the fast desensitization by PKA. Yakel & Jackson (1988) first proposed that the desensitization of 5-HT3 receptor-mediated currents was accelerated by PKA phosphorylation, while Yang et al. (1992) suggested that PKA exhibited no effects on this process. In the present study, the time course of modulation of synaptic GABA release by presynaptic 5-HT3 receptors was independent of phosphorylation by PKA (Fig. 6).

Ligand-gated ion channels, such as those linked to nicotinic, GABAA or NMDA receptors, have unique channel recovery times (Jones & Westbrook, 1996). In cultured neuroblastoma cells, the half-time of recovery of the 5-HT3 receptor response was 2.6 ± 0.12 min and blockade of dephosphorylation of 5-HT3 receptors accelerated recovery (Boddeke et al. 1996). In addition, inhibition of protein kinases was reported to decrease 5-HT3 responses to repeated agonist applications (Robertson & Bevan, 1991). In the present study, 5-HT3 receptor responses recovered quickly so that an interval of more than 9 min was sufficient for complete recovery (Fig. 7) and phosphorylation by PKA accelerated the recovery of presynaptic 5-HT3 receptor responses. While exact PKA phosphorylation sites in nerve terminals were not identified by our studies, phosphorylation sites are found between the M3 and M4 membrane-penetrating regions of amino acid sequences of 5-HT3 receptors (Maricq et al. 1991) and these regions of presynaptic 5-HT3 receptors are likely to be targets. In other ligand-gated ion channels, phosphorylation of niconinic acetylcholine receptors regulates recovery time without affecting fast desensitization (Paradiso & Brehm, 1998). Therefore, phosphorylation of presynaptic 5-HT3 receptors could specifically regulate recovery rates and this might contribute to the plasticity regulating GABA release.

Function of 5-HT3 receptors in the amygdala

Direct injection of 5-HT3 receptor agonists and antagonists into the amygdala increased and decreased, respectively, aversive-like behaviour in animals (Costall et al. 1989). Thus, 5-HT3 receptors participate in the regulation of anxiolytic responses. In the present study, 5-HT3 receptors significantly facilitated synaptic GABA release from GABAergic presynaptic nerve terminals. Since GABA is an inhibitory neurotransmitter in the amygdala, these mechanisms of presynaptic 5-HT3 receptors contribute to the regulation of anxiety. Furthermore, PKA seems to regulate serotonergic modulation of synaptic GABA release via presynaptic 5-HT3 receptors, contributing to the fear-conditioning models in regulating long-lasting GABAergic synaptic efficacy.

In our previous study, activation of presynaptic 5-HT1A receptors continuously inhibited synaptic GABA release in the basolateral amygdala (Koyama et al. 1999). 5-HT3 and 5-HT1A receptors show opposite modes of presynaptic modulation and exhibit quite different time courses, since 5-HT3 receptors are ligand-gated ion channels and 5-HT1A receptors couple to GTP-binding proteins. In addition to various actions of other subtypes of serotonin receptors, presynaptic modulation via a combination of 5-HT3 and 5-HT1A receptors could further increase the variety of responses to a single endogenous agonist, serotonin. The variations of serotonergic synaptic modulation, especially in the amygdala, could generate subtle differences in the emotional and behavioural processes of the animal.

Acknowledgments

The authors thank Dr M. C. Andresen (Department of Physiology and Pharmacology, Oregon Health Science University, Portland, OR, USA) for his kind and critical reading of our manuscript and helpful suggestions. This study was supported by Kyushu University Interdisciplinary Programs in Education and Projects in Research Development and The Japan Health Sciences Foundation (Research on Brain Sciences) for N. Akaike.

References

  1. Akaike N, Harata N. Nystatin perforated patch recording and its applications to analysis of intracellular mechanisms. Japanese Journal of Physiology. 1994;44:433–473. doi: 10.2170/jjphysiol.44.433. [DOI] [PubMed] [Google Scholar]
  2. Ankri N, Legendre P, Faber DS, Korn H. Automatic detection of spontaneous synaptic responses in central neurons. Journal of Neuroscience Methods. 1994;52:87–100. doi: 10.1016/0165-0270(94)90060-4. [DOI] [PubMed] [Google Scholar]
  3. Boddeke HWGM, Meigel I, Boeijinga P, Arbuckle J, Docherty RJ. Modulation by calcineurin of 5-HT3 receptor function in NG 108–15 neuroblastoma × glioma cells. British Journal of Pharmacology. 1996;118:1836–1840. doi: 10.1111/j.1476-5381.1996.tb15611.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Boess FG, Martin IL. Molecular biology of 5-HT receptors. Neuropharmacology. 1994;33:275–317. doi: 10.1016/0028-3908(94)90059-0. [DOI] [PubMed] [Google Scholar]
  5. Brown AM, Hope AG, Lambert JJ, Peters JA. Ion permeation and conduction in a human recombinant 5-HT3 receptor subunit (h5-HT3A) The Journal of Physiology. 1998;507:653–665. doi: 10.1111/j.1469-7793.1998.653bs.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Costall B, Kelly ME, Naylor RJ, Onaivi ES, Tyers MB. Neuroanatomical sites of action of 5-HT3 receptor agonists and antagonists for alteration of aversive behaviour in the mouse. British Journal of Pharmacology. 1989;96:325–332. doi: 10.1111/j.1476-5381.1989.tb11821.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Davies PA, Pistis M, Hanna MC, Peters JA, Lambert JJ, Hales TG, Kirkness EF. The 5-HT3B subunit is a major determinant of serotonin-receptor function. Nature. 1999;397:359–363. doi: 10.1038/16941. [DOI] [PubMed] [Google Scholar]
  8. Fletcher WH, Van Patten SM, Cheng H-C, Walsh DA. Cytochemical identification of the regulatory subunit of the cAMP-dependent protein kinase by use of fluorescently labeled catalytic subunit. Journal of Biological Chemistry. 1986;261:5504–5513. [PubMed] [Google Scholar]
  9. Jones MV, Westbrook GL. The impact of receptor desensitization on fast synaptic transmission. Trends in Neurosciences. 1996;19:96–101. doi: 10.1016/s0166-2236(96)80037-3. [DOI] [PubMed] [Google Scholar]
  10. Kawa K. Distribution and functional properties of 5-HT3 receptors in the rat hippocampal dendate gyrus: a patch-clamp study. Journal of Neurophysiology. 1994;71:1935–1947. doi: 10.1152/jn.1994.71.5.1935. [DOI] [PubMed] [Google Scholar]
  11. Kidd EJ, Laporte AM, Langlois X, Fattaccini C-M, Doyen C, Lombard MC, Gozlan H, Hamon M. 5-HT3 receptors in the rat central nervous system are mainly located on nerve fibers and terminals. Brain Research. 1993;612:289–298. doi: 10.1016/0006-8993(93)91674-h. [DOI] [PubMed] [Google Scholar]
  12. Kilpatrick GJ, Jones BJ, Tyers MB. Identification and distribution of 5-HT3 receptors in rat brain using radioligand binding. Nature. 1987;330:746–748. doi: 10.1038/330746a0. [DOI] [PubMed] [Google Scholar]
  13. Koyama S, Kubo C, Rhee J-S, Akaike N. Presynaptic serotonergic inhibition of GABAergic synaptic transmission in mechanically dissociated rat basolateral amygdala neurons. The Journal of Physiology. 1999;518:525–538. doi: 10.1111/j.1469-7793.1999.0525p.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Ma QP, Yin GF, Ai MK, Han JS. Serotonergic projections from the nucleus raphe dorsalis to the amygdala in the rat. Neuroscience Letters. 1991;134:21–24. doi: 10.1016/0304-3940(91)90499-j. [DOI] [PubMed] [Google Scholar]
  15. Maricq AV, Peterson AS, Brake AJ, Myers RM, Julius D. Primary structure and functional expression of the 5-HT3 receptor, a serotonin-gated ion channel. Science. 1991;254:432–437. doi: 10.1126/science.1718042. [DOI] [PubMed] [Google Scholar]
  16. Morales M, Bloom FE. The 5-HT3 receptor is present in different subpopulations of GABAergic neurons in the rat telencephalon. Journal of Neuroscience. 1997;17:3157–3167. doi: 10.1523/JNEUROSCI.17-09-03157.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Murase K, Randic M, Shirasaki T, Nakagawa T, Akaike N. Serotonin suppresses N-methyl-D-aspartate responses in acutely isolated rat dorsal horn neurons. Brain Research. 1990;525:84–91. doi: 10.1016/0006-8993(90)91323-9. [DOI] [PubMed] [Google Scholar]
  18. Paradiso K, Brehm P. Long-term desensitization of nicotinic acetylcholine receptors is regulated via protein kinase A-mediated phosphorylation. Journal of Neuroscience. 1998;18:9227–9237. doi: 10.1523/JNEUROSCI.18-22-09227.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Piguet P, Galvan M. Transient and long-lasting actions of 5-HT on rat dendate gyrus neurons in vitro. The Journal of Physiology. 1994;481:629–639. doi: 10.1113/jphysiol.1994.sp020469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Riad M, Hamon M, Emerit MB. Pharmacological evidence for the involvement of calcium/calmoduline in serotonin 5-HT3 receptor-mediated cation permeability in NG 108–15 cells. Journal of Neurochemistry. 1994;62:2224–2232. doi: 10.1046/j.1471-4159.1994.62062224.x. [DOI] [PubMed] [Google Scholar]
  21. Robertson B, Bevan S. Properties of 5-hydroxytryptamine 3 receptor-gated currents in adult rat dorsal root ganglion neurones. British Journal of Pharmacology. 1991;102:272–276. doi: 10.1111/j.1476-5381.1991.tb12165.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Ronde P, Nichols RA. High calcium permeability of serotonin 5-HT3 receptors on presynaptic nerve terminals from rat striatum. Journal of Neurochemistry. 1998;70:1094–1103. doi: 10.1046/j.1471-4159.1998.70031094.x. [DOI] [PubMed] [Google Scholar]
  23. Ropert N, Guy N. Serotonin facilitates GABAergic transmission in the CA1 region of rat hippocampus in vitro. The Journal of Physiology. 1991;441:121–136. doi: 10.1113/jphysiol.1991.sp018742. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Steward LJ, West KE, Kilpatrick GJ, Barnes NM. Labeling of 5-HT3 receptor recognition sites in the rat brain using the agonist radioligand [3H]meta-chlorophenylbiguanide. European Journal of Pharmacology. 1993;243:13–18. doi: 10.1016/0014-2999(93)90161-a. [DOI] [PubMed] [Google Scholar]
  25. Sugita S, Shen K-Z, North RA. 5-Hydroxytryptamine is a fast excitatory transmitter at 5-HT3 receptors in rat amygdala. Neuron. 1992;8:199–203. doi: 10.1016/0896-6273(92)90121-s. [DOI] [PubMed] [Google Scholar]
  26. Takenouchi T, Munekata E. Serotonin increases cytoplasmic Ca2+ concentration in PC12h cells: effect of tachykinin peptides. Neuroscience Letters. 1998;246:141–144. doi: 10.1016/s0304-3940(98)00253-5. [DOI] [PubMed] [Google Scholar]
  27. Van Haastert PJM, Driel RV, Jastorff B, Baraniak J, Stec WJ, De Wit RJW. Competitive cAMP antagonists for cAMP-receptor proteins. Journal of Biological Chemistry. 1984;259:10020–10024. [PubMed] [Google Scholar]
  28. van Hooft JA, Spier AD, Yakel JL, Lummis SCR, Vijverberg HPM. Promiscuous coassembly of serotonin 5-HT3 and nicotinic α4 receptor subunits into Ca2+-permeable ion channels. Proceedings of the National Academy of Sciences of the USA. 1998;95:11456–11461. doi: 10.1073/pnas.95.19.11456. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Yakel JL, Jackson MB. 5-HT3 receptors mediate rapid responses in cultured hippocampus and a clonal cell line. Neuron. 1988;1:615–621. doi: 10.1016/0896-6273(88)90111-0. [DOI] [PubMed] [Google Scholar]
  30. Yang J, Mathie A, Hille B. 5-HT3 receptor channels in dissociated rat superior cervical ganglion neurons. The Journal of Physiology. 1992;448:237–256. doi: 10.1113/jphysiol.1992.sp019039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Zhou F-M, Hablitz JJ. Activation of serotonin receptors modulates synaptic transmission in rat cerebral cortex. Journal of Neurophysiology. 1999;82:2989–2999. doi: 10.1152/jn.1999.82.6.2989. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES