Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 1999 Aug 15;519(Pt 1):71–84. doi: 10.1111/j.1469-7793.1999.0071o.x

Post-tetanic potentiation of GABAergic IPSCs in cultured rat hippocampal neurones

Kimmo Jensen *, Morten Skovgaard Jensen *, John D C Lambert *
PMCID: PMC2269478  PMID: 10432340

Abstract

  1. Dual whole-cell patch-clamp recording was used to investigate post-tetanic potentiation (PTP) of GABAergic IPSCs evoked between pairs of cultured rat hippocampal neurones. Tetanization of the presynaptic neurone at frequencies (f) ranging from 5 to 100 Hz resulted in PTP of the IPSCs. Maximum PTP had a magnitude of 51.6 % just after the stimulus train, and lasted up to 1 min. PTP was shown to be dependent on the number of stimuli in the train, but independent of f at frequencies ≥ 5 Hz.

  2. Blocking postsynaptic GABAA receptors with bicuculline during the tetanus did not affect the expression of PTP, showing that it is a presynaptic phenomenon. PTP was strongly affected by changing [Ca2+]oduring the tetanus: PTP was reduced by lowering [Ca2+]o, and increased by high [Ca2+]o.

  3. PTP was still present after presynaptic injection of BAPTA or EGTA, or following perfusion of the membrane-permeable ester EGTA-tetraacetoxymethyl ester (EGTA AM, 50 μM). On the other hand, EGTA AM blocked spontaneous, asynchronous IPSCs (asIPSCs), which were often associated with tetanic stimulation.

  4. Tetanic stimulation in the presence of 4-aminopyridine (4-AP), which promotes presynaptic Ca2+ influx, evoked sustained PTP of IPSCs in half of the neurones tested.

  5. The results indicate that PTP at inhibitory GABAergic synapses is related to the magnitude of presynaptic Ca2+ influx during the tetanic stimulation, leading to an enhanced probability of vesicle release in the post-tetanic period. The increase in [Ca2+]i occurs despite the presence of high-affinity exogenous and endogenous intracellular Ca2+ buffers. That PTP of IPSCs depends on the number, and not the frequency, of spikes in the GABAergic neurone is in accordance with a slow clearing of intracellular Ca2+ from the presynaptic terminals.

Activity-dependent changes in synaptic transmission determine the temporal behaviour and output of neuronal networks in the mammalian central nervous system (CNS) (O'Donovan & Rinzel, 1997). At single synapses, different patterns of activity can evoke frequency-dependent facilitation or depression (Dobrunz & Stevens, 1997), which in its simplest form can be studied using paired-pulse protocols in order to evoke paired-pulse facilitation (PPF) or depression (PPD) (McCarren & Alger, 1985). Changes in synaptic strength can also persist for various periods of time following the conditioning stimulation. Post-tetanic potentiation (PTP) lasts for a number of seconds after tetanization (Griffith, 1990), and may be succeeded by short-term potentiation (minutes), long-term potentiation (LTP) or long-term depression (LTD) (hours) (Alger & Teyler, 1976). All the aforementioned types of plasticity have been studied in detail at excitatory glutamatergic synapses in the CNS (Dobrunz & Stevens, 1997). However, at inhibitory GABAergic synapses, only PPD (Davies et al. 1990; Nathan & Lambert, 1991) and PPF (Tanabe & Kaneko, 1996), and LTP and LTD (McLean et al. 1996) have been described. Since PTP of GABAergic IPSPs would be an important factor for the integration of all synaptic activity (and probably for other forms of tetanus-induced synaptic plasticity), it is important to characterize the stimulus patterns required to evoke PTP of IPSCs.

The aim of the present study was, therefore, to demonstrate and characterize PTP at GABAergic hippocampal synapses in vitro. For this purpose, we made paired whole-cell recordings from cultured hippocampal neurones in the presence of glutamate receptor antagonists and stimulated the presynaptic GABAergic neurone to evoke monosynaptic GABAA receptor-mediated IPSCs. This technique also allowed us to manipulate the internal environment of the GABAergic neurone by altering the composition of the solution in the presynaptic electrode. We observed robust PTP of IPSCs which lasted up to 1 min following tetanic stimulation of the GABAergic neurone. PTP was shown to depend on the magnitude of presynaptic Ca2+ influx and internal Ca2+ buffering. Part of this work has been presented in abstract form (Jensen et al. 1998).

METHODS

Hippocampal culture preparation

Pregnant Sprague-Dawley rats were anaesthetized by pentobarbital (50 mg kg−1i.p.) at gestational day 17–18, following which the rat was killed by cutting the major arteries at the heart, in accordance with the guidelines laid down by the Danish Ministry of Justice. Fetuses were removed and decapitated, and the hippocampi were dissected free. The tissue was triturated mechanically in medium containing (mM): NaCl, 137; KCl, 5.4; Na2HPO4, 0.3; NaHCO3, 1; KH2PO4, 0.4; Hepes, 5; glucose, 30; and Phenol Red, 0.003; pH 7.3 with NaOH, and cells were plated on poly-D-lysine-coated coverslips in 35 mm Petri dishes. Plating medium consisted of minimal essential medium with Earle's salts and Glutamax-1 (Gibco) supplemented with horse serum (HS, 10 %), fetal calf serum (FCS, 10 %), penicillin (50 i.u. ml−1) and streptomycin (50 μg ml−1). Cultures were grown in 5 % CO2 and 10 % O2 at 37°C (Brewer & Cotman, 1989). Plating medium was fully replaced by 2 ml feeding medium after 1 day in vitro and thereafter 1 ml was exchanged twice weekly. Feeding medium had the same composition as plating medium except that FCS was omitted and HS was reduced to 5 %. The mitosis inhibitors 5′-fluoro-2′-deoxyuridine (FUDR, 15 μg ml−1) and uridine (35 μg ml−1) were added after 3–4 days when cultures showed a confluent background.

Electrophysiology

Coverslips with the cultured cells were placed in a stainless steel chamber with a quartz glass bottom mounted on an inverted Nikon Diaphot 200 microscope, and individual neurones were visualized through × 200 Normarski optics. The chamber was continuously perfused (1 ml min−1) with an extracellular (control) medium containing (mM): NaCl, 140; KCl, 3.5; CaCl2, 2.5; MgCl2, 2.5; glucose, 10; and Hepes, 10. pH was adjusted to 7.35 with NaOH (20–22°C). Osmolality was 305 mosmol kg−1 (regularly checked using a Wescor 5500 osmometer). Patch-clamp electrodes (tip resistance, 3–6 MΩ) were fabricated from borosilicate glass (o.d. 1.2 mm) on a Flaming-Brown P-97 puller (Sutter Instruments). Excitatory synaptic interactions between neurones were blocked by including 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 μM) and dl-2-amino-5-phosphonovaleric acid (dl-AP5, 50 μM) in the perfusion medium. GABAergic neurones were initially identified by screening single neurones by whole-cell recording and investigating whether brief stimulation (see below) was followed by an autaptic IPSC. If so, a neurone in close vicinity was then patched and tested for the presence of IPSCs synchronized at a constant latency to the presynaptic stimulation. The presynaptic electrode contained (mM): KOH, 140; EGTA, 11; CaCl2, 1; MgCl2, 2; NaCl, 15; Hepes, 10; leupeptin, 0.10; and MgATP, 2; pH adjusted to 7.3 with methanesulfonic acid, 290 mosmol kg−1. The postsynaptic electrode solution was designed to increase the driving force for Cl− and to block regenerative Na+ and K+ currents, and contained (mM): CsCl, 120; TEACl, 10; EGTA, 11; CaCl2, 1; MgSO4, 1; leupeptin, 0.10; MgATP, 4; and QX-314, 5; pH adjusted to 7.3 with CsOH, 290 mosmol kg−1. In the perforated-patch configuration, the pipette solution contained (mM): KOH, 140; KCl, 15; Hepes, 5; EGTA, 1; and freshly dissolved amphotericin B, 0.32; pH adjusted to 7.3 with methanesulfonic acid, osmolality 290 mosmol kg−1.

Stimulation protocol

Whole-cell recordings were made using Axopatch 200 and 200A amplifiers in voltage-clamp mode at a holding potential (Vh) of −70 mV, and stimulation pulse protocols were delivered by a pulse generator (Master 8, AMPI). We decided to stimulate the presynaptic neurone in the voltage-clamp configuration for two reasons. Firstly, the presynaptic neurone was clamped at −70 mV between stimuli to prevent spontaneous firing. Secondly, spikes evoked by current-clamp stimulation could have variable latency, resulting in jitter of the IPSCs. The presynaptic neurone was stimulated by stepping from −70 to 0 mV for 3 ms, which evoked a Na+-dependent action current in the soma. This was followed by an IPSC in the postsynaptic neurone with a constant latency of between 1 and 3 ms. IPSCs were blocked by tetrodotoxin (300 nM, not shown), indicating that the somatic action current led to break-away action potentials in the axon. Single control stimuli were given at 0.2 Hz, while stimulus trains were delivered at frequencies ranging from 2.5 to 100 Hz. Whole-cell currents were low-pass filtered at 10 kHz, monitored on a pen recorder (Servogor 220), digitized using an A/D converter (Instrutech VR-100 B) and stored simultaneously on a videotape recorder and a Pentium PC equipped with Clampex (pCLAMP version 6.0, Axon Instruments).

Drug application

Active substances were dissolved in water as stock solutions at 1000 times the final concentration, diluted in extracellular medium just before use and perfused through the bath (exchange time, 2–3 min). All changes in [Ca2+]o were compensated by changes in [Mg2+]o to keep the total extracellular divalent cation concentration constant at 5 mM. For experiments in which a rapid change between control medium and test solution was required, the neurones were continuously superfused from a 3-barrel gravity-feed pipette (tip opening, ∼200 μm). Trial experiments with perfusion of bicuculline (to block GABAA receptor-mediated IPSCs) and solutions with an altered Ca2+/Mg2+ ratio showed that the local medium was completely exchanged in less than 2 s. Drugs and chemicals were purchased from Sigma except CNQX and dl-AP5 (Tocris Cookson). EGTA-tetraacetoxymethyl ester (EGTA AM, Molecular Probes) was prepared as a stock solution dissolved in DMSO. When added to the control solution at the final concentration of 0.1 %, DMSO did not affect the IPSCs. Culture media were purchased from Gibco, except for FUDR, uridine and poly-D-lysine (Sigma).

Analysis

IPSC amplitudes were measured on-line using Clampex, whereas off-line analysis of amplitudes and areas of IPSCs was performed using Clampfit (pCLAMP). Occasionally, Axotape (Axon Instruments) was used to record traces for visual presentation in figures. All IPSCs evoked by a single stimulation were inspected visually and rejected if spontaneous activity disturbed the measurements. Areas of IPSCs were measured to register both early and late changes in decay kinetics. Post-tetanic potentiation (PTP) is presented as the percentage change in amplitude of a single IPSC evoked shortly (1–4 s) after the stimulus train compared with the mean of ten pre-tetanic IPSCs (peak PTP). PTP was also quantified by measuring the area under the plot of the amplitudes of the first ten post-tetanic IPSCs, which had been normalized to the pre-tetanic IPSCs (PTP area, see inset in Fig. 2A). Post-tetanic asynchronous IPSCs (asIPSCs) were quantified as the area under the trace with respect to the baseline for 1 s following the train (see Fig. 3A). Data were imported into a spreadsheet (Excel version 7.0a) where means and s.e.m. were calculated, and linear regression and Student's paired and unpaired t tests were performed. The non-parametric Kruskal-Wallis comparison of several groups was made using the statistical software package SOLO (version 6.0.4). All data are presented as means ± s.e.m. with n indicating the number of pairs of neurones tested. Changes were considered to be significant at P values less than 0.05. In some graphs, error bars have been omitted for clarity.

Figure 2. PTP of IPSCs depends on the number of stimuli in the train and not the frequency.

Figure 2

Open in a new tab

A, PTP as a function of frequency (f). Sixty-seven trials were performed in 50 pairs of neurones. Trains were applied at the frequencies shown (log scale) with the number of stimuli held constant at 80. PTP was evaluated by calculating the area under the post-tetanic curve for the first 10 post-tetanic IPSCs (shaded area in inset, subtracting values < 100 % when present). For frequencies between 5 and 80 Hz, no differences between the groups were found, the PTP medians being indicated by a cross. Stimulation at 2.5 Hz did not evoke PTP, indicating that the threshold for eliciting PTP is between 2.5 and 5 Hz. B, PTP as a function of the number of stimuli. Twenty to 160 pulses were given at 5 (a), 20 (b) and 40 Hz (c) as indicated. For each frequency, PTP was calculated for all trials and plotted against the number of stimuli (log scale). PTP was significantly and positively correlated with the number of stimuli, whereas the slopes obtained by linear regression showed no systematic changes with increasing frequency. C, cumulative data from all trials with f ranging from 5 to 100 Hz plotted against number of stimuli. The data were obtained from 116 trials in 54 pairs of neurones. The slope of the regression line was 329.2 (P < 0.05), and it intercepted the abscissa between 18 and 19 pulses. From the few data points with stimuli numbers > 100, it would seem that PTP tends to saturate.

Figure 3. Relationship between PTP and asynchronous post-tetanic IPSCs.

Figure 3

Open in a new tab

A, trace from a neurone which showed asIPSCs during and following a train of 80 Hz for 1 s. The asIPSC activity was quantified as the area under the trace for 1 s following stimulation with respect to the baseline. This area (dashed lines) was normalized to the area of the single pre-tetanic evoked IPSC (shown to the left) and was in this case 9.2. This neurone showed PTP of 82 %. B, scatter plot showing the relationship between PTP of IPSCs and the asIPSC area, measured as in A (n = 29). Trains were delivered at 80 Hz for 1 s, and the PTP area was measured as in Fig. 2A. PTP increased with the asIPSC area (linear regression, P < 0.001). Note that a number of neurones displayed a substantial asIPSC activity, but little PTP, though the reverse was not seen.

RESULTS

Monosynaptic GABAA receptor-mediated IPSCs

Monosynaptic GABAA receptor-mediated IPSCs were examined in 82 pairs of hippocampal neurones continuously perfused with CNQX (10 μM) and dl-AP5 (50 μM) to block glutamatergic excitation. The presynaptic GABAergic neurone was clamped at −70 mV and stimulated by stepping Vh from −70 to 0 mV for 3 ms, which probably evoked a break-away action potential (see Methods). This elicited short latency (1–3 ms) IPSCs in most nearby neurones. These responses were identified as GABAA receptor-mediated IPSCs, since they were blocked by bath perfusion of bicuculline (10 μM) and had a reversal potential of about +5 mV with a nearly symmetrical Cl− gradient across the cell membrane (n = 3). In spite of the inclusion of MgATP in the pipette solutions, minor rundown of IPSC amplitudes occurred. During the first 20 min of recording with low-frequency stimulation, rundown was calculated by linear regression to be 10.1 ± 4.4 % (n = 7).

Post-tetanic potentiation of IPSCs

A brief tetanization (80 Hz for 1 s) of the presynaptic GABAergic neurone resulted in potentiation of subsequent single IPSCs elicited at 0.2 Hz (Fig. 1A). This post-tetanic potentiation (PTP) reached a maximum of 51.6 ± 9.2 % (n = 19) just after the stimulus train and the duration of the potentiation was about 1 min (Fig. 1C). The rundown of test IPSCs was of no consequence for the expression of PTP, but led to a small underestimation of its amplitude. For analysis of the decay kinetics of single post-tetanic IPSCs, peak amplitudes were scaled to the pre-tetanic control IPSC and the areas measured in order to detect both early and late changes in IPSC decay kinetics. The first four post-tetanic test IPSCs showed significant enhancement in area which ranged from 8.5 ± 3.3 % for the first post-tetanic IPSC to 14.4 ± 3.6 % for the fourth IPSC (P < 0.05, n = 19). Since these changes in area were often associated with post-tetanic miniature IPSC activity, measurements of amplitudes alone were chosen to quantify the magnitude and time course of PTP.

Figure 1. Post-tetanic potentiation of IPSCs.

Figure 1

Open in a new tab

A, slow chart recording showing IPSCs elicited by single stimuli at 0.2 Hz. A brief tetanization (80 Hz for 1 s) of the presynaptic neurone was delivered at the filled bar. The first single IPSC following the train showed PTP of 40 % compared with the pre-tetanic control IPSC. The potentiation was reversible. Ba, IPSCs from another neurone on a faster time scale. Only the initial 375 ms of the train of IPSCs is shown. The amplitude of the pre-tetanic IPSC is indicated by the arrow on the IPSC recorded 4 s after the train. PTP was 38 %. Bb, IPSCs evoked before (Control) and 4 s after the train. The IPSCs have been normalized to the same amplitude and superimposed. The post-tetanic IPSC has a slightly slower decay compared with control. C, similar trials with trains of 80 Hz for 1 s were performed in 19 pairs of neurones. Amplitudes were normalized to a control IPSC, which was the mean of ten single responses preceding the train. IPSC amplitude was enhanced by 52 % just after the train and decayed back to the control level within 80 s. D, to elucidate the stimulation parameters required to evoke PTP, a variety of trains were applied to the presynaptic neurone. Results from four of these protocols are shown here. Stimulation with a 4 s train did not evoke PTP when delivered at 5 Hz (a, 20 pulses, n = 4) or 10 Hz (b, 40 pulses, n = 3), but evoked PTP when delivered at 20 Hz (c, 80 pulses, n = 9) and 80 Hz (d, 320 pulses, n = 4).

Stimulus train parameters required to evoke PTP

Since PTP depended strongly on the train parameters (see examples in Fig. 1D), we systematically examined how PTP was related to the frequency of stimulation (f). Trains fixed at 80 stimuli were applied to the presynaptic neurones at frequencies ranging from 2.5 to 80 Hz (Fig. 2A). PTP was quantified for each trial as the area under the curve described by the amplitude of ten single IPSCs following the tetanus (see inset in Fig. 2A). This method of quantification was chosen to reduce variability due to the quantal nature of transmitter release, and to include changes in the later phase of PTP. Results from 67 trials in 50 pairs of neurones disclosed no significant differences between PTP areas obtained at frequencies from 5 to 80 Hz (P > 0.50, Kruskal- Wallis non-parametric comparison of several groups). However, PTP was not observed with presynaptic stimulation at 2.5 Hz (Fig. 2A), showing that the threshold for induction of PTP lies between 2.5 and 5 Hz.

To examine whether PTP is dependent on the number of stimuli, trains consisting of 20–320 pulses given at 5–100 Hz were applied in 116 trials in 54 pairs of neurones. Examples of these results with frequencies of 5, 20 and 40 Hz are shown in Fig. 2B. The slope was calculated by applying linear regression to the plot of PTP areas against the log of the number of stimuli. At all frequencies, PTP increased significantly with the number of stimuli (P < 0.05) and the slopes showed no systematic changes with increasing frequency (5 Hz, slope = 289.1 (24 trials in 17 pairs); 20 Hz, slope = 399.6 (21 trials in 15 pairs); 40 Hz, slope = 351.7 (30 trials in 19 pairs)). All regression lines intercepted the abscissa between 17 and 20 pulses, indicating that the threshold for PTP was the same regardless of the frequency of the train (5 Hz, 18.6 pulses; 20 Hz, 19.2 pulses; 40 Hz, 17.8 pulses). Since PTP depends on the number of stimuli and is independent of f (≥ 5 Hz), a linear regression analysis was performed on the data obtained from experiments with frequencies between 5 and 100 Hz (Fig. 2C). On a semi-logarithmic plot the data could be fitted by a straight line with a slope (329.2) that was significantly different from zero (P < 0.05). Extrapolation showed a threshold for PTP of 18 pulses. Finally, from the results obtained with a high number of stimuli, it seemed that PTP tended to saturate with numbers > 100 (Fig. 2C).

Spontaneous post-tetanic IPSCs are correlated to PTP

In connection with the tetanic stimulation, we often noted an increase in the occurrence of asynchronous IPSCs (asIPSCs) which continued for 1–2 s following the train (Fig. 3A). It has previously been established that similar asynchronous activity recorded at excitatory synapses is caused by accumulation of internal Ca2+ in the presynaptic terminals (Miledi & Thies, 1971; Cummings et al. 1996). We therefore investigated whether there was a relationship between the magnitude of post-tetanic asIPSCs and PTP. In each cell, the total area of the asIPSCs with respect to the baseline was measured for the first 1000 ms following the train (Fig. 3A). This area was then normalized to the area of a single evoked pre-tetanic IPSC so that the results could be compared between cells. When no asIPSC activity was present, the normalized area included the passive decay back to baseline, which had a typical value of about 0.8 times the area of the single pre-tetanic IPSC. The asIPSC area was on average 5.8 ± 0.9 (n = 29, including neurones in which no asIPSC activity was apparently present). There was a correlation between the PTP area (measured as in Fig. 2A) and the asIPSC area (slope = 45.6, linear regression, P < 0.001), indicating that the expression of asIPSCs and PTP are linked to some extent. However, some neurones which displayed a large asIPSC area did not express any appreciable PTP (Fig. 3B).

The effect of changing [Ca2+]o on PTP

Next we investigated the effect of manipulating presynaptic Ca2+ influx occurring during the tetanus on the magnitude of PTP. Using local pipette perfusion, the extracellular Ca2+ concentration could be changed in about 1 s (see Methods). Perfusion of the test solution was started immediately after the last pre-tetanic single IPSC, and was switched back to control solution after the stimulus train (Fig. 4A). Perfusion of nominally Ca2+-free solution ([Mg2+], 5.0 mM) completely blocked IPSCs during the train (40 Hz for 2 s), and abolished PTP (PTP area was 34.4 ± 139 in Ca2+-free solution, n = 3, P > 0.05 when tested against zero, not shown). On varying [Ca2+]o between 1.2 and 4.0 mM, we found that the amplitude of the first IPSC after the train (evoked in 2.5 mM Ca2+) was positively correlated to [Ca2+]oduring the train. PTP of the first post-tetanic IPSC following 1.2 mM Ca2+ was −2.4 ± 5.9 % (P < 0.05, n = 6) (which increased to 29.1 ± 13 % for the second IPSC), 42.6 ± 16 % in the continuous presence of 2.5 mM Ca2+, and 63.6 ± 21 % following 4.0 mM Ca2+ (P < 0.05, n = 5). PTP areas (measured as in Fig. 2A) also increased with [Ca2+]o and were 163 ± 46 following 1.2 mM Ca2+, 278 ± 56 during 2.5 mM Ca2+, and 337 ± 90 following 4.0 mM Ca2+ (Fig. 4B).

Figure 4. Changing [Ca2+]o during the train modulates PTP.

Figure 4

Open in a new tab

A, representative traces showing the effect of changing [Ca2+]o during tetanic stimulation. The middle record shows control responses in 2.5 mM Ca2+ throughout the recording. Solutions containing 1.2 mM Ca2+ (upper record) and 4.0 mM Ca2+ (lower record) were applied 1–2 s before a tetanus at 40 Hz for 2 s was delivered, and application was terminated shortly afterwards. Pre- and post-tetanic single IPSCs evoked in 2.5 mM Ca2+ are shown on either side of the trains. During the train, IPSCs showed tetanic depression which was reduced in 1.2 mM Ca2+ and enhanced in 4.0 mM Ca2+. PTP of the IPSCs showed a similar dependence on the level of extracellular Ca2+ present during the train. This is illustrated to the right, where the post-tetanic responses elicited after returning to 2.5 mM Ca2+ are superimposed (amplified and displaced for clarity). B, graph showing PTP area as a function of [Ca2+]o perfused during the train. PTP was calculated as shown in the inset in Fig. 2A. PTP was absent when a nominally Ca2+-free solution was perfused during the train, and increased with increasing [Ca2+]o. Data were obtained from 22 trials in 10 pairs of neurones. Individual n values are indicated. C, representative traces showing PTP evoked in control solution by a train of 80 Hz for 1 s (upper trace), and with bicuculline (10 μM) present for the duration of the train (lower trace). The presence of bicuculline during the train completely blocked the postsynaptic response and asynchronous activity, but did not affect the induction of PTP.

PTP does not depend on activation of postsynaptic GABAA receptors

To test whether PTP is dependent on activation of postsynaptic GABAA receptors, bicuculline (10 μM) was applied from the local perfusion pipette for the duration of the stimulation train (Fig. 4C) (n = 2). While bicuculline completely abolished postsynaptic responses during the train (Fig. 4C), PTP was not different from that evoked in control solution. This indicates that PTP occurs at a presynaptic locus.

The effect of presynaptic intracellular BAPTA on PTP

Although EGTA and BAPTA have similar affinities for Ca2+, the on-rate for chelation by BAPTA (108 M−1 s−1) is two orders of magnitude faster than for EGTA (106 M−1 s−1) (Deisseroth et al. 1996). It has been proposed that calcium ions associate with the secretory machinery at a rate intermediate between that of chelation by EGTA and BAPTA (Heidelberger et al. 1994). This leads to a depression of IPSCs by BAPTA (Spigelman et al. 1996).

One mechanism for PTP could be that calcium ions bind to a trigger with fast forward kinetics. To test the ability of BAPTA to affect PTP, we injected BAPTA via the presynaptic electrode. A presynaptic electrode solution containing 11 mM BAPTA, instead of 11 mM EGTA, was used in five experiments (Fig. 5). Four to six minutes after establishing whole-cell recordings from the presynaptic neurones, IPSCs were depressed by 26.1 ± 6.6 % compared with just after membrane rupture. In the presence of BAPTA, absolute IPSC amplitudes progressively increased during train stimulation at 80 Hz for 1 s (tetanic facilitation, Fig. 5A). BAPTA did not block asIPSC activity associated with the stimulus train. Peak PTP was dramatically enhanced with BAPTA (241 ± 87 %, P < 0.05, unpaired t test) compared with EGTA-containing electrodes (51.6 ± 9.2 %, Fig. 5B). PTP area was 661 ± 277 with BAPTA which is nearly three times greater than with EGTA (224 ± 38). These data show that the relative PTP increases when the pre-tetanic probability of release is lowered by using a more rapid Ca2+ chelator, and that the PTP process probably does not depend on a Ca2+ sensor with a fast on-rate for Ca2+.

Figure 5. Presynaptic injection of BAPTA does not block PTP.

Figure 5

Open in a new tab

A, representative traces showing the effect of using BAPTA (11 mM) instead of EGTA (11 mM) in the presynaptic intracellular solution. During the presynaptic tetanus of 80 Hz for 1 s, the summed IPSCs progressively increased in amplitude during the first part of the train, which contrasts with the tetanic depression seen with EGTA-containing electrodes (e.g. Fig. 4A and C). Tetanic stimulation was followed by substantial spontaneous activity, which lasted for about 1 s. The first post-tetanic IPSC (right) was enhanced by 567 % compared with the pre-tetanic control. B, peak PTP with BAPTA (▪) was 341 %, which was significantly larger (* P < 0.05, n = 5) than for EGTA-filled neurones (▴, n = 19).

The effect of EGTA AM on asIPSCs and PTP

Activity-dependent release of synaptic vesicles relies on the influx of Ca2+ through voltage-dependent Ca2+ channels (VDCCs), and is enhanced when the ambient level of free Ca2+ within the terminal is elevated (Heidelberger et al. 1994). To gain insight into the Ca2+-dependent mechanisms underlying asIPSCs and PTP, we manipulated the presynaptic Ca2+-buffering capacity using EGTA. Firstly, the effect of endogenous buffers was investigated by making recordings in the perforated-patch mode whilst delivering trains of 80 Hz for 1 s to the presynaptic GABAergic neurone (n = 3). In these experiments, asIPSCs were observed during and following the train (Fig. 6Aa), while peak PTP was 50.9 ± 6.0 %. In another population of neurones, whole-cell recordings were made from the presynaptic GABAergic neurone using electrodes containing 11 mM EGTA (control solution). These neurones showed PTP of 58.8 ± 11 %, and asIPSC activity of a similar magnitude to that seen with perforated-patch recordings (n = 29, Fig. 6Ab). In five of the whole-cell recordings, the cell-permeable ester EGTA AM (50 μM) was perfused after recording control responses. EGTA AM enters the terminals by uptake and ‘trapping’, and EGTA accumulates intracellularly following cleavage by unspecific esterases (Cummings et al. 1996). Ten minutes after the start of perfusion of EGTA AM, the asIPSC area was significantly smaller (P < 0.05, unpaired t test, Fig. 6Ac and B), and was only marginally greater than the ratio of 0.8 when no asIPSC activity was present. This shows that the rise in [Ca2+]i after stimulation at 80 Hz for 1 s did not exceed the threshold for asynchronous vesicle release in the presence of EGTA AM. The differential effect of EGTA (11 mM) injected via the presynaptic neurone, and of perfusion of EGTA AM (50 μM), on asIPSCs (not blocking and blocking asIPSCs, respectively), indicates that the boutons only attained a low concentration of EGTA using presynaptic injection.

Figure 6. EGTA AM inhibits asIPSCs, but not PTP of IPSCs.

Figure 6

Open in a new tab

A, representative traces from three experiments reflecting different levels of Ca2+ buffering in the presynaptic GABAergic neurone. Aa, recording in the perforated-patch mode, so as not to introduce any exogenous Ca2+ chelators. Stimulation was accompanied and followed by asIPSC activity and PTP of the evoked IPSC. Ab, whole-cell recording using an intracellular solution containing 11 mM EGTA gave similar results to the perforated-patch recording. Ac, as in b but with additional extracellular perfusion with 50 μM EGTA AM for 10 min. This abolished the asIPSCs, but did not block PTP of the evoked IPSC. Trains were elicited at 80 Hz for 1 s. B, histogram summarizing the effect of the three different treatments shown in A on the asIPSC area measured as in Fig. 3A. EGTA AM significantly depressed the asIPSC area (* P < 0.05) compared with the perforated-patch (n = 3) and whole-cell recordings (n = 29). C, graph showing that EGTA AM (50 μM) had no effect on whole-cell recordings of PTP. Tetanic stimulation at 80 Hz for 1 s was delivered to the presynaptic neurones in control solution (▴) and again after 10 min of extracellular perfusion of 50 μM EGTA AM (□). In contrast to the profound effect on asIPSCs, EGTA AM had no effect on either the amplitude or the time course of PTP of IPSCs (P > 0.05, n = 5, error bars have been omitted for clarity).

EGTA AM had no effect on PTP (Fig. 6C). Ten minutes after addition of EGTA AM, PTP had increased slightly (Fig. 6C). PTP amplitude increased from 76.2 ± 43.7 to 84.4 ± 38.4 % (P > 0.05), while PTP area increased from 244 ± 97.2 to 281 ± 123 (P > 0.05). Prolonged exposure (for up to 30 min) of the neurones to EGTA AM did not induce further changes in PTP.

The effect of 4-AP on PTP of IPSCs

4-AP blocks the A-type K+ channel, which reduces repolarization following the action potential and will increase presynaptic influx of Ca2+ during tetanic stimulation. The effect of 4-AP on PTP was tested in seven experiments. In each pair of neurones, PTP was elicited in control solution and subsequently in the presence of 4-AP (20–50 μM). In three pairs of neurones, 4-AP enhanced the IPSC baseline level by 14.3 ± 12 %, and tetanization caused a sustained potentiation of IPSC amplitude for at least 4 min (Fig. 7B). In one prolonged recording (Fig. 7A), potentiation was observed for 55 min. After washout of 4-AP, IPSCs were still potentiated. In the remaining four pairs of neurones, perfusion of 4-AP exerted a much stronger effect on the IPSC baseline level, which was enhanced by 114 ± 95 %. On the background of this strongly potentiated single response, PTP was 39.4 ± 29 % compared with 128 ± 18 % for the control (P < 0.05, paired t test). In these neurones, the potentiation in 4-AP declined to pre-stimulation levels over the course of 1 min and the effects of 4-AP were reversible on washing.

Figure 7. Sustained PTP of IPSCs induced by tetanization in the presence of 4-AP.

Figure 7

Open in a new tab

A, representative traces showing the effect of 4-AP on PTP. Single IPSCs were elicited at 0.2 Hz in the presence of 4-AP (20 μM). Post-tetanic responses are shown for 0, 20 and 40 min, where the potentiation was 64, 40 and 34 %, respectively. Recording was continued for 55 min, at which point the IPSC was still potentiated by 21 %. B, sustained PTP as shown in A was observed in 3 out of 7 pairs of neurones tested after tetanization in 4-AP (20–50 μM). Results pooled from these three pairs are shown in the graph. Trains of 80 Hz for 1 s were applied in control solution (⋄) and subsequently in the presence of 4-AP (20–50 μM) (♦). In 4-AP, baseline IPSC amplitudes were enhanced by 14.3 % (to the dashed line). Immediately after tetanization, peak PTP in 4-AP was 70 % (with respect to the pre-tetanic level), compared with 53 % for the control. In 4-AP, however, IPSCs showed sustained potentiation (potentiation after 1 min was 43 %; after 4 min it was 39 %, as averaged from four consecutive responses). The two curves were significantly different at both 1 and 4 min (P < 0.05).

DISCUSSION

Post-tetanic potentiation at GABAergic synapses

This report represents the first characterization of PTP at inhibitory GABAergic synapses in the hippocampus. In previous studies which focused on LTP of isolated IPSPs, sampling frequencies were generally too low to detect the presence of PTP (Xie et al. 1995; Komatsu, 1996) or PTP was superimposed upon different degrees of sustained potentiation of the IPSCs (Glaum & Brooks, 1996). Qualitatively similar PTP has been demonstrated at cholinergic inhibitory synapses in invertebrate neurones (Gardner, 1986), where PTP was 80 % and lasted about 2 min.

Mechanism of PTP

PTP at the neuromuscular junction was originally ascribed to the accumulation of free intracellular Ca2+ in the terminals as a consequence of tetanic activity (‘residual Ca2 +’; Katz & Miledi, 1968; Rosenthal, 1969). This explanation for PTP has subsequently been substantiated at the crayfish neuromuscular junction by the use of Ca2+ imaging (Delaney & Tank, 1994), application of modulators of intracellular Ca2+ stores (Tang & Zucker, 1997) and flash photolysis of Ca2+ chelators (Kamiya & Zucker, 1994), and at hippocampal excitatory synapses using Ca2+ imaging (Regehr et al. 1994). During the tetanus, Ca2+ will enter through VDCCs and bind to the Ca2+ sensor at the secretory apparatus, as well as to other mobile and immobile Ca2+ buffers in the terminal (Neher, 1998). Despite this buffering, [Ca2+]i is raised following the tetanus (Cummings et al. 1996), which increases the probability of vesicle release.

In our experiments, PTP was still present when the postsynaptic GABAA receptors were blocked during the stimulus train, indicating that PTP is a presynaptic phenomenon (Fig. 4C). PTP was positively correlated to the amount of extracellular Ca2+ present during the stimulus train, which would be in accordance with a Ca2+-dependent increase in the probability of vesicle release during the post-tetanic period. PTP increased when the basal probability of release was lowered by presynaptic injection of BAPTA. The action of BAPTA on the probability of release is likely to be caused by a direct effect on the secretory apparatus (see Results). Inhibition of action potential conduction is unlikely to be involved, since millimolar concentrations of the BAPTA derivatives fura-2 and fluo-3 do not cause distal block of action potential propagation in cultured cortical or hippocampal neurones (MacKenzie et al. 1996; MacKenzie & Murphy, 1998).

PTP was positively correlated to the magnitude of post-tetanic asynchronous release (Fig. 3), again underlining the association between Ca2+ accumulation and the PTP process. Nevertheless, the occurrence of asIPSCs and PTP were not inextricably linked. Thus, in one-third of the neurones showing marked asIPSC activity, only marginal PTP was induced (Fig. 3). Whether this reflects differences in the Ca2+ affinities of the sensors for PTP and asIPSCs in subsets of GABAergic neurones, or is due to a difference in the Ca2+-buffering/-clearing mechanisms, cannot be determined from the present results.

Internal Ca2+ buffering

Asynchronous transmitter release is blocked by EGTA AM at both excitatory (Cummings et al. 1996) and inhibitory synapses (Rumpel & Behrends, 1999) and therefore probably requires diffusion of Ca2+ through the cytosol. This contrasts with synchronous release which is usually unaffected by EGTA AM, partly because the Ca2+ channels are thought to be in close association with the docked vesicles (Sheng et al. 1996; Mochida et al. 1996), and the binding rate of Ca2+ by EGTA is much slower than that by the vesicle-associated Ca2+ sensor. However, in giant calyx-type synapses in the rat brainstem (Forsythe, 1994), high concentrations of EGTA injected directly into the presynaptic terminal are able to block synchronous release (Borst & Sakmann, 1996), suggesting that Ca2+ entry through a number of dispersed calcium channels is needed to release a vesicle at this particular synapse.

In the present report, we show that asynchronous release is not affected by a relatively high concentration (11 mM) of either EGTA or BAPTA in the presynaptic electrode. While it is likely that BAPTA and EGTA diffuse right out to the terminals, the concentration is probably much lower than in the presynaptic pipette. For a given concentration of EGTA and BAPTA, the total Ca2+-buffering capacity is similar (Spigelman et al. 1996). Since asynchronous release depends on the ambient level of intracellular Ca2+, it would not be expected to be differentially affected by EGTA or BAPTA (compare Figs 5A and 6Ab). When using EGTA AM, EGTA would be concentrated in the neurone and probably reach a much higher level in the terminals than can be attained using the patch electrode. This would further limit the increase in [Ca2+]i and diffusion of Ca2+ through the cytosol, and block asynchronous vesicle release. On the other hand, PTP was not blocked by EGTA AM, indicating that Ca2+ chelation by EGTA was insufficient to prevent activation of the sensor. PTP was actually enhanced by presynaptic injection of BAPTA. This is mainly because BAPTA reduces the basal release and therefore the size of the pre-tetanic IPSC. Moreover, the Ca2+ that enters during the stimulus train will bind to BAPTA and less will be available to buffer the transient rise in [Ca2+]i evoked by a single stimulation during the post-tetanic period. Synaptotagmin I is thought to be the sensor and trigger for vesicle release, and binding of four calcium ions is probably required to fully activate the release process (Geppert et al. 1994). Binding of Ca2+ to the first site (which has a suggested dissociation constant of 100 nM), would prime the synaptotagmin I-vesicle complex which may be the mechanism underlying PTP (Bertram et al. 1996), whereas asynchronous release probably requires substantially higher Ca2+ concentrations (Heidelberger et al. 1994).

Induction of PTP

Ca2+ entry associated with 20 presynaptic spikes was the threshold for evoking PTP. Above this threshold, PTP increased with every extra spike up to a maximum of around 100 spikes. PTP could not be induced when the GABAergic neurone was stimulated at 2.5 Hz, but developed fully when the frequency was increased to 5 Hz. This shows that the spatial and temporal summation of [Ca2+]i was not sufficient for generation of PTP at inter-pulse intervals greater than 200 ms. The Ca2+-clearing mechanisms therefore require about 300 ms to keep the level of intracellular Ca2+ below the threshold for PTP. Calcium ions are cleared from the cytoplasm by Na+-Ca2+ exchange (Reuter & Porzig, 1995) and active extrusion by Ca2+-dependent ATPases. It has recently been suggested that the low-affinity/high-capacity Na+-Ca2+ exchanger mediates the initial rapid clearing of Ca2+, while the high-affinity/low-capacity plasma membrane Ca2+-ATPase ensures that [Ca2+]i in the terminals is ultimately restored to its resting level (Regehr, 1997). If significant amounts of Na+ accumulate intracellularly as a result of spiking and exchange with Ca2+, activity of the Na+-Ca2+ exchanger is depressed (or even reversed), and clearing of Ca2+ is delayed (Bouron & Reuter, 1996). Ca2+ clearing would also be prolonged if Ca2+ loading of mitochondria occurred during the tetanus (Tang & Zucker, 1997).

We found that PTP at cultured GABAergic synapses (recorded at 20–22 ° C) is shorter in duration than at hippocampal excitatory synapses (5 min at 22 ° C; Tong et al. 1996) or at the neuromuscular junction (5–10 min at 20°C; Rosenthal, 1969). This could indicate that GABAergic boutons have an intrinsically stronger Ca2+-buffering/-clearing system. Various Ca2+-binding proteins, including parvalbumin and calretinin, are selectively located in some GABAergic neurones (Freund & Buzsáki, 1996), and may play an important role in buffering free [Ca2+]i in GABAergic terminals.

Sustained potentiation of IPSCs

Long-term plasticity of GABAergic responses (Xie et al. 1995) has recently been termed LTDGABAA and LTPGABAA (McLean et al. 1996). In about half the neurones studied here, tetanic stimulation in the presence of 4-AP resulted in a sustained potentiation of the IPSC, which persisted after washout of 4-AP. 4-AP enhances presynaptic Ca2+ influx, and sustained potentiation of IPSCs may require a specific level of intracellular Ca2+ to be achieved during stimulation, as has been suggested for excitatory glutamatergic synapses (Stanton, 1996). While it has been suggested that sustained potentiation of GABAergic IPSCs occurs at a presynaptic locus (Glaum & Brooks, 1996), it has also been shown that LTPGABAA requires postsynaptic Ca2+ influx, which is triggered by GABAA receptor-mediated depolarization (McLean et al. 1996). In our cultured hippocampal neurones, a significant postsynaptic change in [Ca2+]i is unlikely to occur during GABAA receptor-mediated activity, since the boutons are located close to the soma (Benson & Cohen, 1996; K. Jensen, unpublished observation), which was clamped at a hyperpolarized membrane potential with an electrode containing EGTA. We therefore consider it most likely that presynaptic mechanisms are involved in the expression of sustained potentiation of IPSCs, and that subtle grading of the Ca2+ levels in the GABAergic boutons could switch the transient potentiation, which characterizes PTP, into a sustained potentiation.

Functional consequences

PTP of IPSPs will participate in the integration of synaptic inputs, and modulate postsynaptic behaviour. PTP of IPSPs would also participate in synaptic plasticity at excitatory synapses such as LTP and LTD, which depend strongly on the frequency and pattern of stimulation (Larson et al. 1986; Bear & Malenka, 1994; Stanton, 1996). Accordingly, activation of GABAergic interneurones at even relatively low frequencies (≥ 5 Hz) will evoke PTP of IPSPs. Finally, PTP of IPSPs could be important in pathophysiological states such as epileptic activity, where high-frequency firing in GABAergic interneurones during ictal activity would evoke PTP of the IPSPs, thereby facilitating cessation of the seizure activity.

Conclusions

We have examined the consequences of tetanus-induced Ca2+ influx in presynaptic GABAergic boutons in hippocampal neurones. The rise in [Ca2+]i probably peaks just after the tetanic stimulation, where substantial asynchronous GABA release occurred. During the following minute, the GABAergic IPSCs displayed PTP which resulted from an increase in the probability of release in response to a single stimulation. For PTP to develop, the GABAergic neurones must be activated at least 20 times at 5 Hz or above. PTP of IPSCs could be converted into a sustained potentiation when transient K+ channels were blocked by 4-AP, which would increase stimulus-induced Ca2+ entry into the terminals. This suggests that the level of presynaptic Ca2+ entry may determine whether the synaptic enhancement is transient or sustained.

Acknowledgments

We are grateful to the Danish Medical Research Council and Aarhus Universitets Forsknings Fond for financial support, and thank Kirsten Kandborg for preparation of the cultures and Sys Kristensen for technical assistance.

References

  1. Alger BE, Teyler TJ. Long-term and short-term plasticity in the CA1, CA3, and dentate regions of the rat hippocampal slice. Brain Research. 1976;110:463–480. doi: 10.1016/0006-8993(76)90858-1. [DOI] [PubMed] [Google Scholar]
  2. Bear MF, Malenka RC. Synaptic plasticity: LTP and LTD. Current Opinion in Neurobiology. 1994;4:389–399. doi: 10.1016/0959-4388(94)90101-5. [DOI] [PubMed] [Google Scholar]
  3. Benson DL, Cohen PA. Activity-independent segregation of excitatory and inhibitory synaptic terminals in cultured hippocampal neurons. Journal of Neuroscience. 1996;16:6424–6432. doi: 10.1523/JNEUROSCI.16-20-06424.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bertram R, Sherman A, Stanley EF. Single-domain/bound calcium hypothesis of transmitter release and facilitation. Journal of Neurophysiology. 1996;75:1919–1931. doi: 10.1152/jn.1996.75.5.1919. [DOI] [PubMed] [Google Scholar]
  5. Borst JGG, Sakmann B. Calcium influx and transmitter release in a fast CNS synapse. Nature. 1996;383:431–434. doi: 10.1038/383431a0. [DOI] [PubMed] [Google Scholar]
  6. Bouron A, Reuter H. A role of intracellular Na+ in the regulation of synaptic transmission and turnover of the vesicular pool in cultured hippocampal cells. Neuron. 1996;17:969–978. doi: 10.1016/s0896-6273(00)80227-5. [DOI] [PubMed] [Google Scholar]
  7. Brewer GJ, Cotman CW. Survival and growth of hippocampal neurons in defined medium at low density: Advantages of a sandwich culture technique or low oxygen. Brain Research. 1989;494:65–74. doi: 10.1016/0006-8993(89)90144-3. [DOI] [PubMed] [Google Scholar]
  8. Cummings DD, Wilcox KS, Dichter M A. Calcium-dependent paired-pulse facilitation of miniature EPSC frequency accompanies depression of EPSCs at hippocampal synapses in culture. Journal of Neuroscience. 1996;16:5312–5323. doi: 10.1523/JNEUROSCI.16-17-05312.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Davies CH, Davies SN, Collingridge GL. Paired-pulse depression of monosynaptic GABA-mediated inhibitory postsynaptic responses in rat hippocampus. The Journal of Physiology. 1990;424:513–531. doi: 10.1113/jphysiol.1990.sp018080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Deisseroth K, Bito H, Tsien RW. Signaling from synapse to nucleus: postsynaptic CREB phosphorylation during multiple forms of hippocampal synaptic plasticity. Neuron. 1996;16:89–101. doi: 10.1016/s0896-6273(00)80026-4. [DOI] [PubMed] [Google Scholar]
  11. Delaney KR, Tank DW. A quantitative measurement of the dependence of short-term synaptic enhancement on presynaptic residual calcium. Journal of Neuroscience. 1994;14:5885–5902. doi: 10.1523/JNEUROSCI.14-10-05885.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dobrunz LE, Stevens CF. Heterogeneity of release probability, facilitation, and depletion of central synapses. Neuron. 1997;18:995–1008. doi: 10.1016/s0896-6273(00)80338-4. [DOI] [PubMed] [Google Scholar]
  13. Forsythe ID. Direct patch recording from identified presynaptic terminals mediating glutamatergic EPSCs in the rat CNS, in vitro. The Journal of Physiology. 1994;479:381–388. doi: 10.1113/jphysiol.1994.sp020303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Freund TF, Buzsáki G. Interneurons of the hippocampus. Hippocampus. 1996;6:347–470. doi: 10.1002/(SICI)1098-1063(1996)6:4<347::AID-HIPO1>3.0.CO;2-I. [DOI] [PubMed] [Google Scholar]
  15. Gardner D. Variations in amplitude and time course of inhibitory postsynaptic currents. Journal of Neurophysiology. 1986;56:1424–1438. doi: 10.1152/jn.1986.56.5.1424. [DOI] [PubMed] [Google Scholar]
  16. Geppert M, Goda Y, Hammer RE, Li C, Rosahl TW, Stevens CF. Synaptotagmin I: A major Ca2+ sensor for transmitter release at a central synapse. Cell. 1994;79:717–727. doi: 10.1016/0092-8674(94)90556-8. [DOI] [PubMed] [Google Scholar]
  17. Glaum SR, Brooks PA. Tetanus-induced sustained potentiation of monosynaptic inhibitory transmission in the rat medulla: evidence for a presynaptic locus. Journal of Neurophysiology. 1996;76:30–38. doi: 10.1152/jn.1996.76.1.30. [DOI] [PubMed] [Google Scholar]
  18. Griffith WH. Voltage-clamp analysis of posttetanic potentiation of the mossy fiber to CA3 synapse in hippocampus. Journal of Neurophysiology. 1990;63:491–501. doi: 10.1152/jn.1990.63.3.491. [DOI] [PubMed] [Google Scholar]
  19. Heidelberger R, Heinemann C, Neher E, Matthews G. Calcium dependence of the exocytosis in a synaptic terminal. Nature. 1994;371:513–515. doi: 10.1038/371513a0. [DOI] [PubMed] [Google Scholar]
  20. Jensen K, Jensen MS, Lambert JDC. Frequency-dependent modulation of transmitter release in cultured GABAergic neurones. European Journal of Neuroscience. 1998;10(suppl. 10):217. [Google Scholar]
  21. Kamiya H, Zucker RS. Residual Ca2+ and short-term synaptic plasticity. Nature. 1994;371:603–606. doi: 10.1038/371603a0. [DOI] [PubMed] [Google Scholar]
  22. Katz B, Miledi R. The role of calcium in neuromuscular facilitation. The Journal of Physiology. 1968;195:481–492. doi: 10.1113/jphysiol.1968.sp008469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Komatsu Y. GABAB receptors, monoamine receptors, and postsynaptic inositol trisphosphate-induced Ca2+ release are involved in the induction of long-term potentiation at visual cortical inhibitory synapses. Journal of Neuroscience. 1996;16:6342–6352. doi: 10.1523/JNEUROSCI.16-20-06342.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Larson J, Wong D, Lynch G. Patterned stimulation at the theta frequency is optimal for the induction of hippocampal long-term potentiation. Brain Research. 1986;368:347–350. doi: 10.1016/0006-8993(86)90579-2. [DOI] [PubMed] [Google Scholar]
  25. McCarren M, Alger BE. Use-dependent depression of IPSPs in rat hippocampal pyramidal cells in vitro. Journal of Neurophysiology. 1985;53:557–571. doi: 10.1152/jn.1985.53.2.557. [DOI] [PubMed] [Google Scholar]
  26. MacKenzie PJ, Murphy TH. High safety factor for action potential conduction along axons but not dendrites of cultured hippocampal and cortical neurons. Journal of Neurophysiology. 1998;80:2089–2101. doi: 10.1152/jn.1998.80.4.2089. [DOI] [PubMed] [Google Scholar]
  27. MacKenzie PJ, Umemiya M, Murphy TH. Ca2+ imaging of CNS axons in culture indicates reliable coupling between single action potentials and distal functional release sites. Neuron. 1996;16:783–795. doi: 10.1016/s0896-6273(00)80098-7. [DOI] [PubMed] [Google Scholar]
  28. McLean HA, Caillard O, Ben-Ari Y, Gaiarsa JL. Bidirectional plasticity expressed by GABAergic synapses in the neonatal rat hippocampus. The Journal of Physiology. 1996;496:471–477. doi: 10.1113/jphysiol.1996.sp021699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Miledi R, Thies R. Tetanic and post-tetanic rise in frequency of miniature end-plate potentials in low-calcium solutions. The Journal of Physiology. 1971;212:245–257. doi: 10.1113/jphysiol.1971.sp009320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mochida S, Sheng ZH, Baker C, Kobayashi H, Catterall WA. Inhibition of neurotransmission by peptides containing the synaptic protein interaction site of N-type Ca2+ channels. Neuron. 1996;17:781–788. doi: 10.1016/s0896-6273(00)80209-3. [DOI] [PubMed] [Google Scholar]
  31. Nathan T, Lambert JDC. Depression of the fast IPSP underlies paired-pulse facilitation in area CA1 of the rat hippocampus. Journal of Neurophysiology. 1991;66:1704–1715. doi: 10.1152/jn.1991.66.5.1704. [DOI] [PubMed] [Google Scholar]
  32. Neher E. Vesicle pools and Ca2+ microdomains: New tools for understanding their roles in neurotransmitter release. Neuron. 1998;20:389–399. doi: 10.1016/s0896-6273(00)80983-6. [DOI] [PubMed] [Google Scholar]
  33. O'Donovan MJ, Rinzel J. Synaptic depression: a dynamic regulator of synaptic communication with varied functional roles. Trends in Neurosciences. 1997;20:431–433. doi: 10.1016/s0166-2236(97)01124-7. [DOI] [PubMed] [Google Scholar]
  34. Regehr WG. Interplay between sodium and calcium dynamics in granule cell presynaptic terminals. Biophysical Journal. 1997;73:2476–2488. doi: 10.1016/S0006-3495(97)78276-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Regehr WG, Delaney KR, Tank DW. The role of presynaptic calcium in short-term enhancement at the hippocampal mossy fiber synapse. Journal of Neuroscience. 1994;14:523–537. doi: 10.1523/JNEUROSCI.14-02-00523.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Reuter H, Porzig H. Localization and functional significance of the Na+/Ca2+ exchanger in presynaptic boutons of hippocampal cells in culture. Neuron. 1995;15:1077–1084. doi: 10.1016/0896-6273(95)90096-9. [DOI] [PubMed] [Google Scholar]
  37. Rosenthal J. Post-tetanic potentiation at the neuromuscular junction of the frog. The Journal of Physiology. 1969;203:121–133. doi: 10.1113/jphysiol.1969.sp008854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Rumpel E, Behrends JC. Sr2+-dependent asynchronous evoked transmission at rat striatal inhibitory synapses in vitro. The Journal of Physiology. 1999;514:447–458. doi: 10.1111/j.1469-7793.1999.447ae.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sheng ZH, Rettig L, Cook T, Catterall WA. Calcium-dependent interaction of N-type calcium channels with the synaptic core complex. Nature. 1996;379:451–454. doi: 10.1038/379451a0. [DOI] [PubMed] [Google Scholar]
  40. Spigelman I, Tymianski M, Wallace CM, Carlen PL, Velumian AA. Modulation of hippocampal synaptic transmission by low concentrations of cell-permeant Ca2+ chelators: effects of Ca2+ affinity, chelator structure and binding kinetics. Neuroscience. 1996;75:559–572. doi: 10.1016/0306-4522(96)00283-7. [DOI] [PubMed] [Google Scholar]
  41. Stanton PK. LTD, LTP, and the sliding threshold for long-term synaptic plasticity. Hippocampus. 1996;6:35–42. doi: 10.1002/(SICI)1098-1063(1996)6:1<35::AID-HIPO7>3.0.CO;2-6. [DOI] [PubMed] [Google Scholar]
  42. Tanabe M, Kaneko T. Paired pulse facilitation of GABAergic IPSCs in ventral horn neurons in neonatal rat spinal cord. Brain Research. 1996;716:101–106. doi: 10.1016/0006-8993(96)00051-0. [DOI] [PubMed] [Google Scholar]
  43. Tang YG, Zucker RS. Mitochondrial involvement in post-tetanic potentiation of synaptic transmission. Neuron. 1997;18:483–491. doi: 10.1016/s0896-6273(00)81248-9. [DOI] [PubMed] [Google Scholar]
  44. Tong G, Malenka RC, Nicoll RA. Long-term potentiation in cultures of single hippocampal granule cells: A presynaptic form of plasticity. Neuron. 1996;16:1147–1157. doi: 10.1016/s0896-6273(00)80141-5. [DOI] [PubMed] [Google Scholar]
  45. Xie Z, Yip S, Morishita W, Sastry BR. Tetanus-induced potentiation of inhibitory postsynaptic potentials in hippocampal CA1 neurons. Canadian The Journal of Physiology and Pharmacology. 1995;73:1706–1713. doi: 10.1139/y95-734. [DOI] [PubMed] [Google Scholar]

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

RESOURCES