Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 1999 Jul 15;518(Pt 2):525–538. doi: 10.1111/j.1469-7793.1999.0525p.x

Presynaptic serotonergic inhibition of GABAergic synaptic transmission in mechanically dissociated rat basolateral amygdala neurons

Susumu Koyama *,†, Chiharu Kubo †, Jeong-Seop Rhee *, Norio Akaike *
PMCID: PMC2269437  PMID: 10381597

Abstract

  1. The basolateral amygdala (ABL) nuclei contribute to the process of anxiety. GABAergic transmission is critical in these nuclei and serotonergic inputs from dorsal raphe nuclei also significantly regulate GABA release. In mechanically dissociated rat ABL neurons, spontaneous miniature inhibitory postsynaptic currents (mIPSCs) arising from attached GABAergic presynaptic nerve terminals were recorded with the nystatin-perforated patch method and pharmacological isolation.

  2. 5-HT reversibly reduced the GABAergic mIPSC frequency without affecting the mean amplitude. The serotonergic effect was mimicked by the 5-HT1A specific agonist 8-OH DPAT (8-hydroxy-2-(di-n-propylamino)tetralin) and blocked by the 5-HT1A antagonist spiperone.

  3. The GTP-binding protein inhibitor N-ethylmaleimide removed the serotonergic inhibition of mIPSC frequency. In either K+-free or Ca2+-free external solution, 5-HT could inhibit mIPSC frequency.

  4. High K+ stimulation increased mIPSC frequency and 8-OH DPAT inhibited this increase even in the presence of Cd2+.

  5. Forskolin, an activator of adenylyl cyclase (AC), significantly increased synaptic GABA release frequency. Pretreatment with forskolin prevented the serotonergic inhibition of mIPSC frequency in both the standard and high K+ external solution.

  6. Ruthenium Red (RR), an agent facilitating the secretory process in a Ca2+-independent manner, increased synaptic GABA release. 5-HT also suppressed RR-facilitated mIPSC frequency.

  7. We conclude that 5-HT inhibits GABAergic mIPSCs by inactivating the AC-cAMP signal transduction pathway via a G-protein-coupled 5-HT1A receptor and this intracellular pathway directly acts on the GABA-releasing process independent of K+ and Ca2+ channels in the presynaptic nerve terminals.

Receptors for γ-aminobutyric acid (GABA), a major inhibitory neurotransmitter in the central nervous system (CNS) and the target for benzodiazepines that act on the GABAA subtype, are clinically useful for treating anxiety. The neurotransmitter 5-hydroxytryptamine (5-HT) contributes significantly to the anxiogenic process (Graeff et al. 1993; Kawahara et al. 1993; Gonzalez et al. 1996).

In the basolateral amygdaloid (ABL) nucleus, neurons are classified into two main types, large pyramidal- and small non-pyramidal-shaped ones. The pyramidal neuron is the most abundant cell type. The non-pyramidal neurons are further subdivided into stellate and cone cells (Millhouse & DeOlmos, 1983). The pyramidal neurons and GABA- containing non-pyramidal neurons have axo-somatic or axo-dendritic synapses with each other, making an inhibitory circuit (McDonald, 1985). Furthermore, serotonergic neurons in dorsal raphe nuclei send axons to the ABL (Aghajanian et al. 1973) which suppress GABAergic inhibition (Wang & Aghajanian, 1977). The mechanistic basis for this interaction between 5-HT and GABA, in particular how 5-HT modulates GABA release, is poorly understood.

The recent development of a purely mechanical dissociation of central neurons with native synaptic boutons allows both optimal voltage control and internal ionic perfusion of the postsynaptic element. Isolation of neurons with synaptic boutons occurred accidentally during the dissociation of neurons from enzymatically treated brain slices (Drewe et al. 1988; Akaike et al. 1992). In the present study, we have used the mechanical dissociation technique in order to preserve the function of the native presynaptic boutons. Using the ‘synaptic bouton preparation’, we investigated the mechanism of serotonergic modulation of GABAergic synaptic transmission.

METHODS

Preparation

Wistar rats 10 to 14 days old were decapitated under pentobarbital anaesthesia. The brain was quickly removed and transversely sliced at a thickness of 500 μm (DTK-1000, Dosaka, Kyoto, Japan). Slices were kept in the control incubation 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, NJ, USA) and the basolateral region of the amygdala was identified under a binocular microscope (SMZ-1, Nikon, Tokyo, Japan). In the fully mechanical dissociation, we used a custom-built vibration device. The device had an arm equipped with a fire-polished glass pipette which could vibrate at about 3-5 Hz (0·1-0·2 mm). The tip of the fire-polished glass pipette was lightly placed on the surface of the ABL with a micromanipulator. The tip of the glass pipette was vibrated horizontally. In about 2 min, the dissociation was complete. Slices were removed and the mechanically dissociated neurons allowed to settle and adhere to the bottom of the dish for about 20 min. Such neurons undergoing dissociation retained their original morphological features including short portions of their proximal dendrites. Pyramidal-like large cells (15-20 μm in diameter) and smaller non-pyramidal cells (< 15 μm) were optically evident.

Electrical measurements

All electrical measurements were performed using the nystatin-perforated patch recording method which allows electrical access to the cytoplasm without intracellular dialysis (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 made 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 was 5-7 MΩ. Neurons were visualized under phase contrast on an inverted microscope (Diapot, Nikon). Current and voltage were continuously monitored on an oscilloscope (Textronix 5111A, Sony, Tokyo, Japan) and recorded on paper (Recti-Horiz 8K, Nippondenki San-ei, Tokyo, Japan), and on videotapes (PCM-501 ES, Sony). Membrane currents were filtered at 1 kHz (E-3201A Dicade 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 recorded in pre-set epochs of 180-300 s (spontaneous) and 60-120 s (high K+ or Ruthenium Red (RR)) in duration, before, during and after each experimental condition. Epoch duration was set so as to include a minimum of 100 events in every count. Events were counted and analysed using DETECTiVENT software (Ankri et al. 1994) and IGOR PRO software (Wavemetrics, Lake Oswego, OR, USA). Inclusion criterion required a minimum event duration of 1·0 ms. The mIPSC analysis used cumulative probability plots. Amplitude histograms were binned in 1·5 pA intervals. Cumulative amplitude histograms were compared using the Kolmogorov- Smirnov test (P < 0·05). The time-to-peak amplitude and time course of decay of individual mIPSCs were analysed using pCLAMP software (Axon Instruments). Numerical values are provided as means ± standard error of the mean (s.e.m.). Differences in amplitude and frequency distribution were tested by Student's paired two-tailed t test.

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 bubbled with 95 % O2 and 5 % CO2. The pH was adjusted to 7·45. The standard external solution contained (mM): 150 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 10 glucose and 10 Hepes. K+-free external solution contained (mM): 145 NaCl, 5 BaCl2, 5 CsCl, 1 MgCl2, 10 glucose and 10 Hepes. External solution with high K+ contained (mM): 135 NaCl, 20 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. The pH of these external solutions was adjusted to 7·4 with Tris. In recording spontaneous mIPSCs, these solutions routinely contained 300 nM tetrodotoxin (TTX) to block voltage-dependent Na+ channels, and 1 μM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and 10 μM dl-2-amino-5-phosphovaleric acid (AP5) to block glutamatergic currents. The ionic composition of internal (patch-pipette) solution for the nystatin-perforated patch recording 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. 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 AP5, bicuculline, CNQX, EGTA, forskolin, GABA, 5-HT, N-ethylmaleimide (NEM) and nystatin from Sigma; TTX (Wako Pure Chemicals, Tokyo, Japan), glycine (Kanto Chemical, Tokyo, Japan), 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH DPAT) and spiperone (Research Biochemicals). CNQX and forskolin were dissolved in dimethyl sulfoxide (DMSO) at 10 mM as a stock solution. Drugs were applied by the ‘Y-tube system’ for rapid solution exchange.

RESULTS

Purely mechanically dissociated neurons in the ABL nucleus

After the mechanical dissociation of ABL nuclei, pyramidal-shaped large neurons (Fig. 1Aa) and ovoid-shaped relatively small neurons (Fig. 1Ab) were observed. They still retained short portions of their proximal dendrites. In the presence of 300 nM TTX and excitatory amino acid antagonists CNQX (1 μM) and AP5 (10 μM), prominent spontaneous mIPSCs were observed in dissociated ABL neurons (Fig. 1B). Mean amplitude and frequency of mIPSCs were measured under these conditions. Recordings of these mIPSCs were stable for about 100 min. At 120 min, mIPSC frequency significantly decreased to 80·5 ± 1·1 % of control (P < 0·05, n= 4) (Fig. 1C). These results indicate that presynaptic nerve terminals attached to the dissociated neurons are functional and their spontaneous activity is stable within 100 min.

Figure 1. Mechanically dissociated rat basolateral (ABL) neurons.

Figure 1

Open in a new tab

Aa, a pyramidal-shaped neuron; b, an ovoid-shaped neuron. Scale bar is 20 μm. B, spontaneous miniature inhibitory postsynaptic currents (mIPSCs) at 20, 60 and 120 min after the application of 300 nM TTX, 1 μM CNQX and 10 μM AP5 to a dissociated neuron. C, time courses of mean amplitude and frequency of the mIPSCs. All amplitudes and frequencies are normalized to those of mIPSCs at 20 min (•) after the application of TTX, CNQX and AP5. Each point is the mean of measured neurons: 20 min, n= 6; 40 min, n= 6; 60 min, n= 6; 80 min, n= 6; 100 min, n= 4; 120 min, n= 4. The vertical error bars show ±s.e.m. In this and subsequent figures, asterisk represents a statistically significant difference (P < 0·05; paired two-tailed t test).

Spontaneous GABAergic mIPSCs

Application of 100 μM bicuculline completely and reversibly blocked a substantial portion of spontaneous mIPSCs (Fig. 2A). Bicuculline (1 μM) suppressed mIPSC amplitude without affecting mIPSC frequency (Fig. 2B). Application of 100 μM GABA for 30 s evoked large inward currents and suppressed subsequent spontaneous mIPSCs for up to 1 min after the removal of GABA. At this concentration, GABA induced a maximal postsynaptic current response. GABAergic mIPSCs are also suppressed with exogenous GABA application in the hippocampus (Murphy et al. 1998). The depression of mIPSCs following removal of exogenous GABA recovered in about 2 min. However, the rat amygdala neurons have glycine receptors at somatic and proximal dendritic sites (Danober & Pape, 1998). Thus, we examined the possibility that the mIPSCs could be glycinergic. Following a similar application of 300 μM glycine, the mIPSC rates were not altered (Fig. 2C). At this concentration, glycine induced a maximal postsynaptic current response. Further, we examined the dose-response relation of bicuculline and strychinine on the mean amplitude of mIPSCs (Fig. 2D). These two agents suppressed the mIPSC amplitude in a concentration-dependent manner (IC50: 1·38 μM in bicuculline, 1·36 μM in strychinine). These results indicate that the spontaneous mIPSCs were mediated by GABAA receptors.

Figure 2. GABAergic mIPSCs.

Figure 2

Open in a new tab

A, recording of mIPSCs before, during and after the application of 100 μM bicuculline (bic). B, cumulative frequency distribution (left) and cumulative amplitude distribution (right) of mIPSCs. Concentration of bicuculline is 1 μM. C, recording of mIPSCs and responses to exogenous 100 μM GABA and 300 μM glycine. D, concentration-response curve (semilogarithmic plot) for normalized mIPSC amplitude and bicuculline or strychinine. The continuous and dashed lines represent the best fit using the Hill function. Each point is the mean of 4 neurons. Bicuculline (ratio to control): 10−8 M, 0·90 ± 0·035; 10−7 M, 0·75 ± 0·082; 10−6 M, 0·65 ± 0·076; 3 × 10−6 M, 0·32 ± 0·023; 10−5 M, 0·14 ± 0·008; 10−4 M, 0·01 ± 0·005. Strychinine (ratio to control): 10−8 M, 0·95 ± 0·071; 10−7 M, 0·75 ± 0·018; 10−6 M, 0·53 ± 0·067; 3 × 10−6 M, 0·23 ± 0·05; 10−5 M, 0·20 ± 0·036. The vertical error bars show ±s.e.m.

Serotonergic modulation of spontaneous GABAergic mIPSCs

Serotonin (1 μM) decreased the mIPSC frequency in the majority (41/57) of ABL neurons tested (Fig. 3). These 41 neurons included 29 pyramidal and 12 non-pyramidal cell types. 5-HT facilitated the mIPSC frequency in eight neurons (pyramidal, 2; non-pyramidal, 6) and had no effect in eight neurons (pyramidal, 3; non-pyramidal, 5). The 5-HT-induced mIPSC frequency decrement was reversible (Fig. 3Aa and b). Analysis of results from a single representative ABL neuron shows that 5-HT decreased the number of mIPSCs (filled area) at all amplitudes (Fig. 3B). The overall distributions of mIPSC amplitudes were similar between control and 5-HT-treated neurons (P= 0·43 with the Kolmogorov-Smirnov test, number of events 1021 for control and 635 for 5-HT). The time to peak and time course of decay were generally similar (Fig. 3B, inset). The means of 30 individual time-to-peak amplitudes were 6·64 ± 0·14 ms in control and 6·54 ± 0·11 ms in 5-HT-treated neurons (P= 0·50, n= 5). Decay time was also unaltered by 5-HT. The means of 30 individual fast time constant (τfast) values were 8·06 ± 0·31 ms in control and 8·58 ± 0·36 ms in 5-HT (P= 0·10, n= 5). The means of 30 individual τslow values were 45·2 ± 0·75 ms in control and 43·2 ± 0·76 ms in 5-HT (P= 0·06, n= 5). Figure 3C shows the cumulative probability plots. The mIPSC frequency decreased to 64·3 ± 1·5 % of control (Fig. 3Cb; P < 0·01, n= 41) but the distribution of their amplitudes did not change (101·1 ± 2·1 % of control, n= 41) (Fig. 3Ca). These results suggest that 5-HT acts at a presynaptic site.

Figure 3. Serotonergic presynaptic inhibition of mIPSCs.

Figure 3

Open in a new tab

Aa, recording of mIPSCs before, during and after the application of 1 μM 5-HT. b, time course of event frequency. The number of events in every 10 s duration is plotted. Each point is the mean of 5 neurons. B, amplitude histograms from the same neuron for control (open area) and 5-HT (filled area). Inset, individual monosynaptic mIPSCs in the same neuron with or without 5-HT are superimposed. C, cumulative amplitude distribution (a) and cumulative frequency distribution (b) of mIPSCs of the trace. Inset histograms, each column is the mean of 41 neurons. All amplitudes and frequencies are normalized to those of control mIPSCs. In this and subsequent figures, asterisks represent a statistically significant difference (**P < 0·01; paired two-tailed t test).

Subtype of 5-HT receptors

Although at least seven subtypes of 5-HT receptors are known (Boess & Martin, 1994), inhibitory synaptic transmission is modulated by presynaptic 5-HT1A receptors at many mammalian central synapses (Schmitz et al. 1995; Bijak & Misgeld, 1997). A 5-HT1A-specific agonist 8-OH DPAT (1 μM) reversibly decreased the mIPSC frequency (Fig. 4Aa and Ba) in a manner similar to 5-HT itself (Fig. 3). For example, 8-OH DPAT decreased mIPSC frequency to 66·0 ± 1·1 % of control (P < 0·01, n= 6) without affecting the mean amplitude (P= 0·44, n= 6) (Fig. 4C). The 5-HT1A antagonist spiperone (1 μM) blocked this action of 8-OH DPAT on the mIPSC frequency (Fig. 4Ab and Bb) but spiperone itself did not affect cumulative distributions of mIPSCs (data not shown). Thus, serotonin acting via a 5-HT1A receptor appears to inhibit presynaptic GABA release.

Figure 4. Subtype of a 5-HT receptor.

Figure 4

Open in a new tab

Aa, recording of mIPSCs before, during and after the application of 1 μM 8-OH DPAT. b, recording of mIPSCs with 1 μM spiperone in the same neuron. B, cumulative frequency distribution of mIPSCs with 8-OH DPAT before (a) and after (b) spiperone treatment. C, all amplitudes and frequencies are normalized to those of control mIPSCs. Each column is the mean of 6 neurons.

Effects of N-ethylmaleimide (NEM)

Most 5-HT receptor subtypes are coupled to GTP-binding protein (G-protein) (Boess & Martin, 1994). To investigate whether the 5-HT1A receptor in the GABAergic presynaptic ending is coupled to G-proteins (e.g. Gi and/or Go), we tested the effects of NEM, a sulphydryl alkylating agent (Asano & Ogasawara, 1986) (Fig. 5A). By itself, 1 μM NEM did not change either mIPSC amplitude (109·6 ± 2·7 % of control, n= 7) or mIPSC frequency (99·9 ± 8·9 % of control, n= 7) (Fig. 5Ba and C). In contrast, 5-HT-induced decrement of the mIPSC frequency (65·8 ± 4·7 % of control, P < 0·05, n= 4) was eliminated with 1 μM NEM treatment (Fig. 5Bb and C). Thus, the 5-HT1A receptor appears to be coupled to a G-protein.

Figure 5. Effects of N-ethylmaleimide (NEM).

Figure 5

Open in a new tab

A, recording of mIPSCs with 1 μM 5-HT before and after treatment with 1 μM NEM. B, cumulative frequency distribution of mIPSCs in A. a, comparison of control with NEM-treated mIPSC frequency. b, effects of 5-HT on mIPSC frequency after treatment with NEM. C, all amplitudes and frequencies are normalized to those of control mIPSCs. Each column is the mean of measured neurons: control, n= 7; 5-HT, n= 4; NEM, n= 7; NEM + 5-HT, n= 4.

Intracellular mechanisms of 5-HT action

At least three possible mechanisms can be suggested for this G-protein-coupled presynaptic inhibition of GABA release by serotonin. First, activation of K+ channels could hyperpolarize the presynaptic membranes resulting in decreased synaptic release (Umemiya & Berger, 1995; Bijak & Misgeld, 1997). Second, inhibition of Ca2+ influx into the nerve terminal by closing of voltage-dependent Ca2+ channels could decrease exocytosis (Umemiya & Berger, 1995). Third, inhibition of an adenylyl cyclase (AC)-cAMP signal transduction pathway could decrease the neurotransmitter release mechanism (Dumuis et al. 1988). The following experiments were performed to investigate the possible contributions of these three mechanisms.

Firstly, if modulation of presynaptic K+ channels is responsible for the serotonergic decrease of mIPSC frequency, then K+-free external solution ought to increase this decrement. We examined the effects of K+-free solution on mIPSCs (Fig. 6A). K+-free conditions significantly increased mIPSC frequency to 292 ± 43 % of control (P < 0·05, n= 5) but did not alter mIPSC amplitude (98·8 ± 9·6 % of control, n= 5) (Fig. 6Ba and C). In the K+-free external solution, 5-HT decreased mIPSC frequency to 59·0 ± 7·5 % of K+-free control (P < 0·05, n= 4) but had no effect on the mean amplitudes (P= 0·28, n= 4) (Fig. 6Bb and C). Thus, presynaptic K+ conductance does not appear to contribute to the inhibitory effect of 5-HT on mIPSC frequency.

Figure 6. Effects of K+-free external solution.

Figure 6

Open in a new tab

A, recording of mIPSCs before, during and after the application of 1 μM 5-HT in K+-free external solution. B, cumulative frequency distribution of mIPSCs in A. a, comparison of mIPSC frequency in standard external solution with K+-free external solution. b, effects of 5-HT on mIPSC frequency in K+-free external solution. C, all amplitudes and frequencies are normalized to those of mIPSCs in standard external solution. Each column is the mean of measured neurons: normal solution, n= 5; K+-free, n= 5; K+-free + 5-HT, n= 4.

Secondly, if presynaptic voltage-dependent Ca2+ channels contribute to the serotonergic modulation of presynaptic GABA release, then the 5-HT-induced frequency decrement should disappear in Ca2+-free external solutions (Fig. 7A). The exclusion of Ca2+ ions from the external solution decreased mIPSC frequency to 57·3 ± 7·2 % of control (P < 0·05, n= 4) without affecting the distribution of their amplitudes (86·5 ± 5·9 % of control, n= 4) (Fig. 7Ba and C). In this Ca2+-free external solution, 5-HT still decreased reversibly the mIPSC frequency to 70·8 ± 2·2 % of the Ca2+-free control (P < 0·01, n= 4) without affecting the mean amplitude (P= 0·80, n= 4) (Fig. 7Bb and C). This result suggests that the closing of voltage-dependent Ca2+ channels by 5-HT does not participate in the inhibitory modulation of 5-HT on synaptic GABA release.

Figure 7. Effects of Ca2+-free external solution.

Figure 7

Open in a new tab

A, recording of mIPSCs before, during and after the application of 1 μM 5-HT in Ca2+-free external solution. B, cumulative frequency distribution of mIPSCs in the upper trace. a, comparison of mIPSC frequency in standard external solution with Ca2+-free external solution. b, effects of 5-HT on mIPSC frequency in Ca2+-free external solution. C, all amplitudes and frequencies are normalized to those of control mIPSCs in standard external solution. Each column is the mean of 4 neurons.

While the voltage control in these isolated neurons is direct, the precise membrane potential of the presynaptic nerve terminals is uncertain and more difficult to manipulate. A rapid increase in the extracellular K+ concentration ([K+]o) could depolarize the presynaptic membrane and open voltage-dependent Ca2+ channels with subsequent increases in exocytosis at the nerve terminals (Akaike et al. 1992). The increase in [K+]o-facilitated mIPSC frequency varied in a concentration-dependent manner but did not change mIPSC amplitude (Fig. 8A). For example, 20 mM [K+]o significantly increased mIPSC frequency to 1126 ± 172 % of the 5 mM [K+]o level (P < 0·01, n= 6) without affecting the mean amplitude (109 ± 14·1 % of 5 mM [K+]o, P= 0·7, n= 6). The effect on mIPSC frequency evoked by high K+ solution at 20 mM [K+]o was rapidly reversed. Therefore, 20 mM [K+]o was used in the following experiments as the high K+ stimulus. Figure 8Ba shows the recording of mIPSCs during a 3 min application of high K+ solution with or without 5-HT. Interestingly, the decrease of mIPSC frequency induced by 5-HT was transiently stronger in the first minute (Fig. 8Bb1) than in the second minute (Fig. 8Bb2). In the first minute, 5-HT decreased the mIPSC frequency to 65·8 ± 4·9 % of high K+ control (P < 0·01, n= 6) (Fig. 8Bb1; inset graph) without affecting mIPSC amplitude (109·2 ± 5·5 % of high K+ control, n= 6).

Figure 8. Serotonergic modulation of high K+-induced mIPSCs.

Figure 8

Open in a new tab

A, concentration-response curve (semilogarithmic plot) for [K+]o and normalized mIPSC amplitude or frequency. All amplitudes and frequencies of mIPSCs are normalized to those of mIPSCs at 5 mM [K+]o (•). Each point is the mean of measured neurons: 0·1 mM, n= 6; 0·3 mM, n= 4; 0·5 mM, n= 6; 1 mM, n= 6; 3 mM, n= 3; 5 mM, n= 7; 10 mM, n= 4; 20 mM, n= 6; 30 mM, n= 4. Ba, recording of mIPSCs by 3 min application of high K+ (20 mM) solution for control and 1 μM 5-HT in the same neuron: 1 and 1′, the first minute; 2 and 2′, the following 2 min. b, cumulative frequency distributions of the traces in a in the first minute (b1) and the following 2 min (b2). Inset histogram, all frequencies are normalized to those of high K+ control mIPSCs. Each column is the mean of 6 neurons.

Cadmium (Cd2+) effectively blocks voltage-dependent Ca2+ channels. We first examined the effects of 100 μM Cd2+ on mIPSCs (Fig. 9Aa). The presence of 100 μM Cd2+ decreased mIPSC amplitude to 59·7 ± 4·7 % of control (P < 0·01, n= 6) and mIPSC frequency to 63·7 ± 5·1 % of control (P < 0·01, n= 6) (Fig. 9C). This result suggests that Cd2+ blocked voltage-dependent Ca2+ channels in nerve terminals, decreasing synaptic GABA transmission, and antagonized the postsynaptic GABA response. We further examined the effects of 100 μM Cd2+ on high K+-induced mIPSCs for 1 min duration (Fig. 9Ab). Cd2+ decreased the high K+-induced mIPSC amplitude to 56·5 ± 6·1 % of the high K+ control (P < 0·05, n= 6) and the high K+-induced mIPSC frequency to 9·9 ± 1·7 % of the high K+ control (P < 0·01, n= 6) (Fig. 9Ba and C). This result suggests that voltage-dependent Ca2+ channels in nerve terminals are critical for increases in mIPSC frequency with high K+ stimulation. As serotonergic inhibition was prominent in the first minute of high K+ stimulation (Fig. 8B), we checked whether voltage-dependent Ca2+ channels could contribute during this period. When the voltage-dependent Ca2+ channels in the nerve terminals were blocked with Cd2+, 8-OH DPAT (1 μM) still reduced high K+-induced mIPSC frequency to 52·8 ± 4·0 % of the high K+ control (P < 0·01, n= 6) without affecting the mean amplitude (P= 0·65, n= 6) (Fig. 9Bb and C), suggesting that the blockade of Ca2+ influx via voltage-dependent Ca2+ channels is not the major mechanism responsible for the serotonergic inhibition of synaptic GABA releases.

Figure 9. Effects of Cd2+.

Figure 9

Open in a new tab

Aa, recording of mIPSCs before, during and after the application of 100 μM Cd2+. b, recording of high K+ (20 mM)-induced mIPSCs before, during and after the application of 100 μM Cd2+. In the presence of Cd2+, high K+-induced mIPSCs with or without 1 μM 8-OH DPAT were recorded. B, cumulative amplitude and frequency distributions of high K+-induced mIPSCs. a, comparison of control with Cd2+-treated high K+-induced mIPSCs. b, effects of 8-OH DPAT on high K+-induced mIPSCs in the presence of Cd2+. C, all amplitudes and frequencies are normalized to those of control mIPSCs. Each column is the mean of 6 neurons.

Thirdly, since 5-HT1A is likely to be coupled to a G-protein, we investigated whether the AC-cAMP signal transduction pathway is involved (Fig. 10A). Activation of AC with forskolin (10 μM) (Seamon et al. 1981) significantly increased mIPSC frequency to 172 ± 12 % of control (P < 0·01, n= 10) without affecting the mean mIPSC amplitude (117·1 ± 7·8 % of control, n= 10), consistent with actions at a presynaptic site (Fig. 10Ba and C). In each neuron, 5-HT reduced mIPSC frequency before forskolin treatment (58·4 ± 3·6 % of control, P < 0·01, n= 6), but this action was prevented by 10 μM forskolin (Fig. 10Bb and C).

Figure 10. Serotonergic modulation of mIPSCs with forskolin.

Figure 10

Open in a new tab

A, recording of mIPSCs with 1 μM 5-HT before and after 10 μM forskolin treatment. B, cumulative frequency distribution of mIPSCs in A. a, comparison of control with folskolin-treated mIPSC frequency. b, effects of 5-HT on mIPSC frequency after forskolin treatment. C, all amplitudes and frequencies are normalized to those of control mIPSCs. Each column is the mean of measured neurons: control, n= 10; 5-HT, n= 6; forskolin, n= 10; forskolin + 5-HT, n= 6.

The involvement of AC in serotonergic inhibition of GABA release was examined further during presynaptic depolarization (Fig. 11A). High K+ application (1 min) rapidly increased the rate of mIPSCs. Forskolin did not significantly alter high K+-induced mIPSC frequency (108 ± 11 % of control, n= 6) or amplitude (107 ± 3·4 % of control, n= 6) (Fig. 11Ba and C). As before, 5-HT inhibited high K+-induced mIPSC frequency to 74·0 ± 1·5 % (P < 0·01, n= 4), while the inhibition was removed by forskolin (Fig. 11Bb and C). Thus, serotonin appears to be negatively coupled to the AC-cAMP signal transduction pathway.

Figure 11. Serotonergic modulation of high K+-induced mIPSCs with forskolin.

Figure 11

Open in a new tab

A, recording of mIPSCs during 1 min application of 20 mM high K+ solution with or without 1 μM 5-HT before and after treatment with forskolin. B, cumulative frequency distribution of high K+-induced mIPSCs in A. a, comparison of control with folskolin-treated high K+-induced mIPSC frequency. b, effects of 5-HT on high K+-induced mIPSC frequency after forskolin treatment. C, all amplitudes and frequencies are normalized to those of control mIPSCs in high K+ solution. Each column is the mean of measured neurons: high K+, n= 6; high K++ 5-HT, n= 4; high K+ (forskolin), n= 6; high K+ (forskolin) + 5-HT, n= 4.

Effect of Ruthenium Red (RR)

RR is believed to act on plasma membranes to facilitate neurosecretion in a Ca2+-independent manner (Trudeau et al. 1996a). Using this pharmacological tool, we examined whether 5-HT could modulate RR-enhanced GABA release. RR (300 nM) rapidly and reversibly enhanced mIPSCs (Fig. 12A). RR (100 nM) did not facilitate mIPSCs. We avoided higher RR concentrations, since preliminary experiments showed that neurons treated with 1-3 μM RR produced a prolonged and enhanced mIPSC frequency for more than 20-50 min after the removal of RR. RR (300 nM) significantly increased mIPSC frequency to 170 ± 13 % of control (P < 0·05, n= 4) but had no effect on the distribution of their amplitude (88·8 ± 7·4 % of control, n= 4), consistent with a presynaptic site of RR action (Fig. 12Ba and C). As before, 5-HT decreased mIPSC frequency to 67·8 ± 0·9 % of control (P < 0·01, n= 4) and also reduced the RR-facilitated mIPSC frequency to 62·7 ± 3·2 % (P < 0·05, n= 4) without affecting the mean amplitude (P= 0·47, n= 4) (Fig. 12Bb and C).

Figure 12. Serotonergic modulation of Ruthenium Red (RR)-facilitated mIPSCs.

Figure 12

Open in a new tab

A, recording of mIPSCs during 3 min application of 300 nM RR with or without 1 μM 5-HT. B, cumulative frequency distribution from traces in A. a, comparison of control with RR-facilitated mIPSC frequency. b, effects of 5-HT on RR-facilitated mIPSC frequency. C, all amplitudes and frequencies are normalized to those of control mIPSCs. Each column is the mean of 4 neurons.

DISCUSSION

Intracellular signal pathway of 5-HT

5-HT1A receptors commonly couple to the GTP-binding protein (G-protein). Activation of these receptors leads to inhibition of adenylyl cyclase (AC). Alternatively, 5-HT is known to directly modulate K+ channels via G-protein (Oh et al. 1995). In the present study, any action of 5-HT on postsynaptic K+ channels was prevented by Cs+ being present in the internal solution. Moreover, K+ channels did not appear to contribute to the presynaptic serotonergic modulation of GABAergic IPSCs (Fig. 6). This observation is consistent with results showing that presynaptic K+ channels in the hippocampus were not affected by 5-HT (Schmitz et al. 1995). In contrast, however, increases of K+ conductance by 5-HT may contribute to presynaptic transmission in the brainstem and the dendate gyrus (Umemiya & Berger, 1995; Bijak & Misgeld, 1997). Thus, 5-HT modulation of presynaptic K+ channels might vary with brain region.

Voltage-dependent Ca2+ channels are also directly modulated with 5-HT via G-protein (Koike et al. 1994; Rhee et al. 1996) and those channels trigger neurotransmitter release. In the present study, a Ca2+-independent mechanism modulating GABAergic transmission (Fig. 7) is consistent with studies in the hippocampus (Scanziani et al. 1992; Thompson et al. 1993). Activation of the AC-cAMP signal transduction pathway directly facilitates the presynaptic neurotransmitter release in rat central neurons (Chavez-Noriega & Stevens, 1994; Capogna et al. 1995; Trudeau et al. 1996b,Chen & Regehr, 1997). Numerous proteins on the surface of a synaptic vesicle (SV) and the inside of the presynaptic membrane contribute to synaptic vesicle docking and fusion with the presynaptic membrane. Several of these proteins are putative sites for phosphorylation by protein kinases (Greengard et al. 1993; Sudhof, 1995). Phosphorylation of these sites may cause the mobilization of SVs from the cytoskeleton and the docking of SVs at the release site or their subsequent fusion with the plasma membrane. Considering these findings and our results (Figs 10 and 11), we suggest that inhibition of AC-cAMP signal transduction with 5-HT might lead to dephosphorylation of proteins on SVs in nerve terminals, causing decreases in synaptic GABA transmission.

Action site of the serotonergic signal transduction pathway

Neurotransmitter release is regulated by the time-dependent mechanisms of the SV cycle within a nerve terminal. This cycle can be divided into three main processes: exocytosis with neurotransmitter release, endocytosis of the empty vesicle and regeneration of fresh vesicles. Of these three processes, exocytosis is the key event where many protein-protein interactions occur between SVs and plasma membrane and these may be modulated by Ca2+ and enzymes (Greengard et al. 1993; Sudhof, 1995). To investigate exocytotic processing, RR is a useful tool (Trudeau et al. 1996a). RR was first used to block mitochondrial Ca2+ transport (Moore, 1971). In the present study, RR facilitated synaptic GABA release (Fig. 12). The possible mechanisms are thought to be: (1) increased mitochondrial Ca2+ release, (2) opening of voltage-dependent Ca2+ channels accelerating Ca2+ influx, (3) increased Ca2+ release from intracellular stores, and (4) direct activation of the exocytotic apparatus. The first mechanism described above does not appear to contribute to release, since RR acted immediately and the recovery was relatively fast after washing with the standard external solution. Other studies suggest that mitochondria do not contribute to the short-term Ca2+ buffering (McBurney & Neering, 1987). The second and third mechanisms are also unlikely, since RR blocks voltage-dependent Ca2+ channels (Gomis et al. 1994; Hamilton & Lundy, 1995; Trudeau et al. 1996a) as well as ryanodine-activated Ca2+ release and IP3-induced Ca2+ release (Vites & Pappano, 1992; Ma, 1993). The fourth mechanism is therefore most likely. RR binds to the membrane phospholipids and reduces the energy barrier for exocytosis (Voelker & Smejtek, 1996). RR is also thought to act on synaptobrevin, enhancing SV binding to the plasma membrane and the fusion/exocytosis process (Trudeau et al. 1996a). Thus, RR might directly accelerate the exocytotic phase of GABA release and 5-HT may suppress this process.

Quantitative time-lapse fluorescence imaging techniques suggest that the entire SV cycle requires approximately 1 min (Betz & Bewick, 1992; Ryan et al. 1993). According to the analysis of single synaptic boutons, it takes less than 15 s from membrane depolarization to the end of exocytosis. The time from endocytosis to repriming is within 30 s (Ryan et al. 1993). In Fig. 8B, the depolarization of the presynaptic membranes opened voltage-dependent Ca2+ channels and presumably increased Ca2+ influx causing the enhancement of synaptic GABA release during the 3 min application of high K+ solution. The inhibitory effect of 5-HT on the enhanced GABA release is stronger at the beginning compared with the end of the stimulus period (Fig. 8Bb). There are two possibilities to explain these observations. First, more SVs might be available for exocytosis in the first minute but then become depleted in the subsequent minute. Second, the high K+ stimulus might depolarize almost all the presynaptic terminals simultaneously and synchronize their exocytotic process in the first minute. In the subsequent phase, however, the presynaptic terminals might begin to behave according to their individual vesicle recycling times resulting in a loss of this initial synchrony. The results in Figs 8 and 12 suggest that the site of 5-HT action may be the exocytotic process in the SV cycling process.

Effects of Cd2+

Cd2+ (100 μM) decreased both spontaneous synaptic GABA transmission and postsynaptic GABA currents (Fig. 9). Divalent cations such as Cd2+ and Zn2+ directly antagonize the GABAA response (Nakagawa et al. 1991). Both 100 μM Cd2+ and Ca2+-free conditions decreased mIPSC frequency to about 60 % of control (Figs 7 and 9). This suggests that voltage-dependent Ca2+ channels and Ca2+ influx from outside of plasma membranes in nerve terminals contribute to generate mIPSC. In the case of high K+-induced mIPSCs, 100 μM Cd2+ also decreased current amplitude in a similar manner in mIPSCs (Fig. 9), indicating postsynaptic inhibition of GABA current with Cd2+.

Acknowledgments

The authors wish to thank Dr H. Yawo for advice concerning the 5-HT modulatory effect on presynaptic K+ and Ca2+ channels, and Dr M. Andresen and Dr P. Sah for kind critical reading and comments. This study was supported by Grants-in-Aid for Scientific Research to J.-S. R. (No. 10770010) and N. A. (Nos 10044301 and 10470009) from the Ministry of Education, Science and Culture, Japan.

References

  1. Aghajanian GK, Kuhar MJ, Roth RH. Serotonin-containing neuronal perikarya and terminals: Differential effects of p-chlorophenylalanine. Brain Research. 1973;54:85–101. doi: 10.1016/0006-8993(73)90036-x. [DOI] [PubMed] [Google Scholar]
  2. 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]
  3. Akaike N, Harata N, Ueno S, Tateishi N. GABAergic synaptic current in dissociated nucleus basalis of Meynert neurons of the rat. Brain Research. 1992;570:102–108. doi: 10.1016/0006-8993(92)90569-u. [DOI] [PubMed] [Google Scholar]
  4. 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]
  5. Asano T, Ogasawara N. Uncoupling of gamma-aminobutyric acid B receptors from GTP-binding proteins by N-ethylmaleimide: effect of N-ethylmaleimide on purified GTP-binding proteins. Molecular Pharmacology. 1986;29:224–249. [PubMed] [Google Scholar]
  6. Betz WJ, Bewick GS. Optical analysis of synaptic vesicle recycling at the frog neuromuscular junction. Science. 1992;255:200–203. doi: 10.1126/science.1553547. [DOI] [PubMed] [Google Scholar]
  7. Bijak M, Misgeld U. Effects of serotonin through serotonin 1A and serotonin 4 receptors on inhibition in the guinea-pig dendate gyrus in vitro. Neuroscience. 1997;78:1017–1026. doi: 10.1016/s0306-4522(96)00666-5. [DOI] [PubMed] [Google Scholar]
  8. 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]
  9. Capogna M, Gahwiler BH, Thompson SM. Presynaptic enhancement of inhibitory synaptic transmission by protein kinase A and C in the rat hippocampus in vitro. Journal of Neuroscience. 1995;15:1249–1260. doi: 10.1523/JNEUROSCI.15-02-01249.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chavez-Noriega LE, Stevens CF. Increased transmitter release at excitatory synapses produced by direct activation of adenylate cyclase in rat hippocampal slices. Journal of Neuroscience. 1994;14:310–317. doi: 10.1523/JNEUROSCI.14-01-00310.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chen C, Regehr WG. The mechanism of cAMP-mediated enhancement at a cerebellar synapse. Journal of Neuroscience. 1997;17:8687–8694. doi: 10.1523/JNEUROSCI.17-22-08687.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Danober L, Pape H-C. Strychinine-sensitive glycine responses in neurons of the lateral amygdala: an electrophysiological and immunocytochemical characterization. Neuroscience. 1998;85:427–441. doi: 10.1016/s0306-4522(97)00648-9. [DOI] [PubMed] [Google Scholar]
  13. Drewe JA, Childs GV, Kunze DL. Synaptic transmission between dissociated adult mammalian neurons and attached synaptic boutons. Science. 1988;241:1810–1813. doi: 10.1126/science.2459774. [DOI] [PubMed] [Google Scholar]
  14. Dumuis A, Sebben M, Bockaert J. Pharmacology of 5-hydroxytryptamine-1A receptors which inhibit cAMP production in hippocampal and cortical neurons in primary culture. Molecular Pharmacology. 1988;33:178–186. [PubMed] [Google Scholar]
  15. Gomis A, Gutierrez LM, Sala F, Viniegra S, Reig JA. Ruthenium red inhibits selectively chromaffin cell calcium channels. Biochemical Pharmacology. 1994;47:225–231. doi: 10.1016/0006-2952(94)90010-8. [DOI] [PubMed] [Google Scholar]
  16. Gonzalez LE, Andrews N, File SE. 5-HT1A and benzodiazepine receptors in the basolateral amygdala modulate anxiety in the social interaction test, but not in the elevated plus-maze. Brain Research. 1996;732:145–153. doi: 10.1016/0006-8993(96)00517-3. [DOI] [PubMed] [Google Scholar]
  17. Graeff FG, Silveira MC, Nogueira RL, Audi EA, Oliveira RM. Role of the amygdala and periaqueductal gray in anxiety and panic. Behavioral Brain Research. 1993;58:123–131. doi: 10.1016/0166-4328(93)90097-a. [DOI] [PubMed] [Google Scholar]
  18. Greengard P, Valtorta F, Czernik AJ, Benfenanti F. Synaptic vesicle phosphoproteins and regulation of synaptic function. Science. 1993;259:780–785. doi: 10.1126/science.8430330. [DOI] [PubMed] [Google Scholar]
  19. Hamilton MG, Lundy PM. Effect of ruthenium red on voltage-sensitive Ca2+ channels. Journal of Pharmacology and Experimental Therapeutics. 1995;273:940–947. [PubMed] [Google Scholar]
  20. Kawahara H, Yoshida M, Yokoo H, Nishi M, Tanaka M. Psychological stress increases serotonin release in the rat amygdala and prefrontal cortex assessed by in vivo microdialysis. Neuroscience Letters. 1993;162:81–84. doi: 10.1016/0304-3940(93)90565-3. [DOI] [PubMed] [Google Scholar]
  21. Koike H, Saito H, Matsuki N. 5-HT1A receptor-mediated inhibition of N-type calcium current in acutely isolated ventromedial hypothalamic neuronal cells. Neuroscience Research. 1994;19:161–166. doi: 10.1016/0168-0102(94)90139-2. [DOI] [PubMed] [Google Scholar]
  22. Ma J. Block by ruthenium red of the ryanodine-activated calcium release channel of skeletal muscle. Journal of General Physiology. 1993;102:1031–1056. doi: 10.1085/jgp.102.6.1031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. McBurney RN, Neering IR. Neuronal calcium homeostasis. Trends in Neurosciences. 1987;10:164–169. [Google Scholar]
  24. McDonald AJ. Immunohistochemical identification of gamma-aminobutyric acid-containing neurons in the rat basolateral amygdala. Neuroscience Letters. 1985;53:203–207. doi: 10.1016/0304-3940(85)90186-7. [DOI] [PubMed] [Google Scholar]
  25. Millhouse OE, DeOlmos J. Neuronal configurations in lateral and basolateral amygdala. Neuroscience. 1983;10:1269–1300. doi: 10.1016/0306-4522(83)90112-4. [DOI] [PubMed] [Google Scholar]
  26. Moore CL. Specific inhibition of mitochondrial Ca2+ transport by ruthenium red. Biochemical and Biophysical Research Communications. 1971;42:298–305. doi: 10.1016/0006-291x(71)90102-1. [DOI] [PubMed] [Google Scholar]
  27. Murphy DD, Cole NB, Greenberger V, Segel M. Estradiol increases dendritic spine density by reducing GABA neurotransmission in hippocampal neurons. Journal of Neuroscience. 1998;18:2550–2559. doi: 10.1523/JNEUROSCI.18-07-02550.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Nakagawa T, Wakamori M, Shirasaki T, Nakaye T, Akaike N. γ-Aminobutyric acid-induced response in acutely isolated nucleus solitarii neurons of the rat. American Journal of Physiology. 1991;260:C745–749. doi: 10.1152/ajpcell.1991.260.4.C745. [DOI] [PubMed] [Google Scholar]
  29. Oh U, Ho YK, Kim D. Modulation of the serotonin-activated K+ channel by G protein subunits and nucleotides in rat hippocampal neurons. Journal of Membrane Biology. 1995;147:241–253. doi: 10.1007/BF00234522. [DOI] [PubMed] [Google Scholar]
  30. Rhee JS, Ishibashi H, Akaike N. Serotonin modulates high-voltage-activated Ca2+ channels in rat ventromedial hypothalamic neurons. Neuropharmacology. 1996;35:1093–1100. doi: 10.1016/s0028-3908(96)00052-4. [DOI] [PubMed] [Google Scholar]
  31. Ryan TA, Reuter H, Wendland B, Schweizer FE, Tsien RW, Smith SJ. The kinetics of synaptic vesicle recycling measured at single presynaptic boutons. Neuron. 1993;11:713–724. doi: 10.1016/0896-6273(93)90081-2. [DOI] [PubMed] [Google Scholar]
  32. Scanziani M, Capogna M, Gahwiler BH, Thompson SM. Presynaptic inhibition of miniature excitatory synaptic currents by baclofen and adenosine in the hippocampus. Neuron. 1992;9:919–927. doi: 10.1016/0896-6273(92)90244-8. [DOI] [PubMed] [Google Scholar]
  33. Schmitz D, Empson RM, Heinemann U. Serotonin reduces inhibition via 5-HT1A receptors in area CA1 of rat hippocampal slices in vitro. Journal of Neuroscience. 1995;15:7217–7225. doi: 10.1523/JNEUROSCI.15-11-07217.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Seamon KB, Padgett W, Daly JW. Forskolin: unique diterpene activator of adenylate cyclase in membranes and in intact cells. Proceedings of the National Academy of Sciences of the USA. 1981;78:3363–3367. doi: 10.1073/pnas.78.6.3363. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sudhof TC. The synaptic vesicle cycle: a cascade of protein- protein interactions. Nature. 1995;375:645–653. doi: 10.1038/375645a0. [DOI] [PubMed] [Google Scholar]
  36. Thompson SM, Capogna M, Scanziani M. Presynaptic inhibition in the hippocampus. Trends in Neurosciences. 1993;16:222–227. doi: 10.1016/0166-2236(93)90160-n. [DOI] [PubMed] [Google Scholar]
  37. Trudeau LE, Doyle RT, Emery DG, Haydon PG. Calcium-independent activation of the secretory apparatus by ruthenium red in hippocampal neurons: a new tool to assess modulation of presynaptic function. Journal of Neuroscience. 1996a;16:46–54. doi: 10.1523/JNEUROSCI.16-01-00046.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Trudeau LE, Emery DG, Haydon PG. Direct modulation of the secretory machinery underlies PKA-dependent synaptic facilitation in hippocampal neurons. Neuron. 1996b;17:789–797. doi: 10.1016/s0896-6273(00)80210-x. [DOI] [PubMed] [Google Scholar]
  39. Umemiya M, Berger AJ. Presynaptic inhibition by serotonin of glycinergic inhibitory synaptic currents in the rat brain stem. Journal of Neurophysiology. 1995;73:1192–1201. doi: 10.1152/jn.1995.73.3.1192. [DOI] [PubMed] [Google Scholar]
  40. Vites AM, Pappano AJ. Ruthenium red selectively prevents Ins(1,4,5)P3- but not caffeine-gated calcium release in avian atrium. American Journal of Physiology. 1992;262:H268–277. doi: 10.1152/ajpheart.1992.262.1.H268. [DOI] [PubMed] [Google Scholar]
  41. Voelker D, Smejtek P. Adsorption of ruthenium red to phospholipid membranes. Biophysical Journal. 1996;70:818–830. doi: 10.1016/S0006-3495(96)79621-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wang RY, Aghajanian GK. Inhibition of neurons in the amygdala by dorsal raphe stimulation: mediation through a direct serotonergic pathway. Brain Research. 1977;120:85–102. doi: 10.1016/0006-8993(77)90499-1. [DOI] [PubMed] [Google Scholar]

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

RESOURCES