| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Centre de recherche en sciences neurologiques and Département de Physiologie, Université de Montréal, Montreal, Quebec H3C 3J7, Canada
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Ouardouz, Mohamed and Jean-Claude Lacaille. Properties of unitary IPSCs in hippocampal pyramidal cells originating from different types of interneurons in young rats. J. Neurophysiol. 77: 1939-1949, 1997. Whole cell recordings were used in hippocampal slices of young rats to examine unitary inhibitory postsynaptic currents (uIPSCs) evoked in CA1 pyramidal cells at room temperature. Loose cell-attached stimulation was applied to activate single interneurons of different subtypes located in stratum oriens (OR), near stratum pyramidale (PYR), and at the border of stratum radiatum and lacunosum-moleculare (LM). uIPSCs evoked by stimulation of PYR and OR interneurons had similar onset latency, rise time, peak amplitude, and decay. In contrast, uIPSCs elicited by activation of LM interneurons were significantly smaller in amplitude and had a slower time course. The mean reversal potential of uIPSCs was 
53.1 ± 2.1 (SE) mV during recordings with intracellular solution containing potassium gluconate. With the use of recording solution containing the potassium channel blocker cesium, the reversal potential of uIPSCs was not significantly different (58.5 ± 2.6 mV), suggesting that these synaptic currents were not mediated by potassium conductances. Bath application of the 
-aminobutyric acid-A (GABAA) receptor antagonist bicuculline (25 µM) reversibly blocked uIPSCs evoked by stimulation of all interneuron subtypes. In bicuculline, the mean peak amplitude of uIPSCs recorded with potassium gluconate was reduced to 3.5 ± 4.4% of control (n = 7). Similarly, with cesium methanesulfonate, the mean amplitude in bicuculline was 2.9 ± 3.1% of control (n = 13). Application of the GABAB receptor antagonist CGP 55845A (5 µM) resulted in a significant and reversible increase in the mean amplitude of uIPSCs recorded with cesium-containing intracellular solution. Thus uIPSCs from all cell types appeared under tonic presynaptic inhibition by GABAB receptors. Paired stimulation of individual interneurons at 100- to 200-ms intervals did not result in paired pulse depression of uIPSCs. For individual responses, a significant negative correlation was observed between the amplitude of the first and second uIPSCs. A significant paired pulse facilitation (154.0 ± 8.0%) was observed when the first uIPSC was smaller than the mean of all first uIPSCs. A small, but not significant, paired pulse depression (90.8 ± 4.0%) was found when the first uIPSC was larger than the mean of all first uIPSCs. Our results indicate that these different subtypes of hippocampal interneurons generate Cl
-mediated GABAA uIPSCs. uIPSCs originating from different types of interneurons may have heterogeneous properties and may be subject to tonic presynaptic inhibition via heterosynaptic GABAB receptors. These results suggest a specialization of function for inhibitory interneurons and point to complex presynaptic modulation of interneuron function.
Granule cells of the dentate gyrus and pyramidal cells of Ammon's horn make up a well-defined excitatory circuit in the hippocampus (Lorente de No 1934; Ramon y Cajal 1911 Slices
Hippocampal slices (300 µm thick) were prepared from young (14-21 days) male Sprague-Dawley rats as previously described (Ouardouz and Lacaille 1995 Recording and stimulation
Whole cell recordings were made from CA1 pyramidal cells with the use of patch electrodes (5-10 M uIPSCs
CA1 pyramidal cells and interneurons were visually distinguished in rat hippocampal slices by the location and morphology of their somata, and their dendritic arborizations (Morin et al. 1996
Properties of uIPSCs
uIPSCs evoked in pyramidal cells differed depending on the type of interneuron stimulated. In initial recordings with intracellular solution containing potassium gluconate, uIPSCs evoked by stimulation of OR (n = 4) and PYR (n = 2) interneurons were generally similar, but those elicited by activation of LM interneurons (n = 5) were smaller in amplitude and had slower time courses (data not shown). These properties were then characterized in cells held at +20 mV with the use of cesium methanesulfonate intracellular solution to block K+ conductances and improve space clamp (Fig. 2). The mean peak amplitude was not significantly different for averaged uIPSCs evoked by stimulation of OR (78.9 ± 19.1 pA; range: 14.6-208.9 pA; n = 11) and PYR (112.3 ± 18.6 pA; range: 49.0-288.0 pA; n = 18) interneurons (Fig. 2; P > 0.05). However, the peak amplitude of uIPSCs evoked from LM interneurons (33.2 ± 8.4 pA; range: 10.3-79.9 pA; n = 8) was significantly smaller than responses elicited by stimulation of interneurons in PYR (P < 0.05) and OR (P < 0.05, 1-tailed test; Fig. 2). The mean onset latency was similar for uIPSCs elicited by activation of interneurons in OR (4.8 ± 0.4 ms; range: 2.9-8.8 ms) and PYR (4.3 ± 0.2 ms; range: 2.3-6.8 ms; P > 0.05), but significantly longer for uIPSCs from LM cells (6.9 ± 0.4 ms; range: 5.1-9.7 ms; P < 0.05). uIPSCs evoked by stimulation of OR and PYR interneurons showed a similar mean rise time (5.9 ± 1.1 ms and 4.1 ± 0.8 ms, respectively; P > 0.05). The mean rise time was slower for uIPSCs from LM cells (12.0 ± 1.8 ms; P < 0.05; Fig. 2). The decay of averaged uIPSCs was well fitted with a single exponential for all cell types. The mean decay time constant of uIPSCs was 69.7 ± 11.8 ms for interneurons in LM, 54.4 ± 7.1 ms for those of OR, and 43.4 ± 3.4 ms for those of PYR. The difference in mean decay time constant was significant only between responses from PYR and LM interneurons (P < 0.05; Fig. 2). A positive but nonsignificant correlation was observed between decay time constant and rise time for uIPSCs evoked from all cell types (Fig. 2; P > 0.05).
Current-voltage relation
uIPSCs evoked from different interneurons all showed a similar equilibrium potential. With the use of intracellular solutions with potassium gluconate, the mean reversal potential (Erev) of uIPSCs of all cell types was
GABAA uIPSCs
uIPSCs evoked by different interneuron subtypes were all mediated by GABAA receptors. During bath application of 25 µM bicuculline, the mean peak amplitude of uIPSCs, recorded with potassium gluconate intracellular solutions, was blocked up to 3.5 ± 4.4% of control (n = 7, P < 0.05). uIPSCs were antagonized for all presynaptic cell types tested (n = 2 PYR, 2 OA and 3 LM cells). Similarly, with the use of cesium methanesulfonate intracellular solutions, bicuculline reversibly blocked uIPSCs evoked from all cell types (Fig. 4; mean amplitude in bicuculline 2.9 ± 3.1% of control, n = 6 PYR, 4 OA, and 3 LM cells; P < 0.05).
Presynaptic GABAB inhibition
uIPSCs evoked by stimulation of LM cells, and recorded with potassium gluconate pipette solutions, were larger in amplitude (129.5 ± 3.4% of control, n = 2) during bath application of the GABAB receptor antagonist CGP 55845A (5 µM). The effects of CGP 55845A were further characterized with the use of cesium methanesulfonate-filled electrodes. The mean peak amplitude of averaged uIPSCs was significantly and reversibly increased to 141.6 ± 12.8% of control (n = 10, P < 0.05) in CGP 55845A (Fig. 6). More individual uIPSCs of large amplitude were seen in CGP 55845A than in control ACSF (Fig. 7). For individual cell types, the mean amplitude of uIPSCs was significantly increased in CGP 55845A for three of four PYR, three of four OR, and two of two LM interneurons tested. The effect of the GABAB receptor antagonist on the amplitude distribution of uIPSCs was examined (Fig. 7). The cumulative probability distribution of the peak amplitude of uIPSCs was significantly shifted toward larger values in the presence of CGP 55845A compared with control (Fig. 7, Kolmogorov-Smirnov test, P < 0.05). Such a significant shift was seen in 6 of 10 presynaptic cells tested (n = 2 of 4 PYR, 3 of 4 OR, and 1 of 2 LM interneurons). In the presence of CGP 55845A, the amplitudes of the larger uIPSCs were greater than in control (Fig. 7B).
Paired pulse stimulation
Because monosynaptic IPSCs in CA1 pyramidal cells show paired pulse GABAB depression (Davies et al. 1990
The principal findings of the present study are that activation of individual interneurons in PYR, OR, and LM layers of the CA1 region evoked uIPSCs in pyramidal cells that were blocked by the GABAA antagonist bicuculline, were potentiated by the GABAB antagonist CGP 55845A, showed a reversal potential near GABAA postsynaptic mechanisms originating from different interneurons
GABAA postsynaptic mechanisms observed at unitary synapses of interneurons are consistent with previous reports on PYR interneurons (basket and axoaxonic cells) (Buhl et al. 1994a GABAA inhibition by LM interneurons
Stellate cells in LM are not activated by collaterals of pyramidal cells, and therefore may only provide feed-forward dendritic inhibition (Lacaille and Schwartzkroin 1988b Properties of uIPSCs
The similar properties of uIPSCs originating from PYR and OR interneurons suggest homogeneous synaptic mechanisms for these cells despite large variations in their postsynaptic target sites on pyramidal cells. PYR interneurons contact pyramidal cell somata and proximal dendrites (basket cells, Buhl et al. 1994a Properties of uIPSCs originating from LM interneurons
The reduced amplitude and slower kinetics of uIPSCs originating from LM cells suggest that postsynaptic mechanisms may be different between interneurons. Several reasons could explain these uIPSC differences. First, they could be due to dendritic filtering in poorly clamped dendrites (Spruston et al. 1993 Presynaptic inhibition of uIPSCs
The potentiation of uIPSCs by the GABAB antagonist CGP 55845A (Davies et al. 1993
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
). In addition, interneurons form complex local networks within the hippocampus (Amaral 1978; Buhl et al. 1994a; Lacaille et al. 1989; Lorente de No 1934; Ramon y Cajal 1911; Sik et al. 1995). Most interneurons are inhibitory, using -aminobutyric acid (GABA) as neurotransmitter (Ribak et al. 1978; Woodson et al. 1989), and they critically offset the excitatory activity in principal cells (Andersen et al. 1964a,b; Kandel et al. 1961; Traub et al. 1987).
; Sivilotti and Nistri 1990). Synaptically released GABA opens postsynaptic GABAA receptors/channels permeable to Cl, to produce a rapid postsynaptic response (Alger and Nicoll 1982a,b). Postsynaptic GABAA mechanisms may be heterogeneous, because selective activation of subsets of inhibitory afferents with the use of minimal stimulation elicits inhibitory postsynaptic currents with different physiological and pharmacological properties (Lambert and Wilson 1993; Pearce 1993). GABA also binds to postsynaptic GABAB receptors, leading to G protein activation and K+ channel opening, to generate a slow postsynaptic response (Dutar and Nicoll 1988a,b; Newberry and Nicoll 1985). Multiple types of inhibitory interneurons have been identified in the CA1 region on the basis of their morphology, membrane properties, and local circuit interactions (Buhl et al. 1994a; Kawaguchi and Hama 1987; Lacaille and Schwartzkroin 1988a,b; Lacaille et al. 1987; McBain et al. 1994; Schwartzkroin and Mathers 1978). Furthermore, these interneuron subtypes are segregated in distinct layers in the CA1 region. The precise role of these various subtypes of interneurons in generating GABA inhibition of pyramidal cells remains largely unknown. Direct stimulation of individual interneurons in or near the pyramidal cell layer elicits GABAA inhibitory postsynaptic potentials (IPSPs) in pyramidal cells (Buhl et al. 1994a; Miles 1990). The synaptic mechanisms involved in the inhibition of pyramidal cells by other subtypes of interneurons remains to be demonstrated, although some evidence suggests that specific subpopulations of interneurons may be responsible for GABAB inhibition (Lacaille and Schwartzkroin 1988b; Lacaille et al. 1989; Samulack and Lacaille 1993; Segal 1990; Williams and Lacaille 1992).
). Subtypes of interneurons, located near the pyramidal cell layer (stratum pyramidale, PYR), in distal apical (stratum lacunosum-moleculare/radiatum border, LM) and basal (stratum oriens, OR) dendritic layers, were selectively activated. These different subtypes of hippocampal interneurons were found to produce unitary inhibitory postsynaptic currents (uIPSCs) in CA1 pyramidal cells via GABAA receptors and Cl channels, but with heterogeneous properties. In addition, uIPSCs appeared under tonic presynaptic inhibition via heterosynaptic GABAB receptors.
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
; Williams et al. 1994). Slices were maintained at room temperature (22-24°C) in a container filled with artificial cerebrospinal fluid (ACSF) containing (in mM) 124 NaCl, 5 KCl, 1.25 NaH2PO4, 2 MgSO4, 2 CaCl2, 26 NaHCO3, and 10 D-glucose saturated with 95% O2-5% CO2. After an initial incubation period of 1 h, a slice was placed in a recording chamber and maintained submerged during continuous perfusion at 3-4 ml/min with oxygenated ACSF (Edwards et al. 1989). The recording chamber was mounted on an upright microscope (Zeiss Axioskop) equipped with a long-range water-immersion objective (×40), Nomarski optics, and a near infrared charge-coupled device camera (Cohu 6500) to allow visual identification of various subtypes of interneurons and pyramidal cells in the CA1 area.
) filled with either of two solutions containing (in mM) 1) 140 potassium gluconate, 5 NaCl, 2 MgCl2, 10 N-2-hydroxyethylpiperazine-N
-2-ethanesulfonic acid (HEPES), 0.5 ethylene glyco-bis(
-aminoethyl ether)-N,N,N,N-tetraacetic acid (EGTA), 2 adenosine 5-triphosphate (ATP)-Tris, and 0.4 guanosine 5-triphosphate (GTP)-Tris, pH adjusted to 7.3 with KOH; or 2) 140 cesium methanesulfonate, 5 NaCl, 1 MgCl2, 10 HEPES, 1 EGTA, 2 ATP-Tris and 0.4 GTP-Tris, pH adjusted to 7.3 with CsOH. Voltage-clamp recordings were made with the use of an Axopatch 1D amplifier (Axon Instruments) with low-pass filtering at 10 kHz (3 dB). Series resistance was compensated to 80% and regularly monitored for constancythroughout the experiment. The mean series resistance was 18.1 ±1.1 (SE) M (n = 64). Recordings were digitized at 22 kHz for storage on a video cassette recorder (Neurocorder DR-886), and also analyzed with a PC-DOS microcomputer-based data acquisition system (TL-125 and pClamp, Axon Instruments). For stimulation of individual interneurons, a patch pipette (5-10 M) filled with ACSF was first positioned on the soma of a visually identified interneuron and gentle suction was applied (Edwards et al. 1990). In this loose cell-attached configuration, constant current stimulation (1 ms, 0-8 µA, 0.2 Hz) was applied with a stimulus isolation unit (WPI), in the presence of the glutamate receptor antagonists 20 µM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, RBI) and 50 µM (±)2-amino-5-phosphonopentanoic acid (AP-5, RBI). In some control experiments, biphasic monosynaptic IPSPs were recorded from pyramidal cells in slices taken from mature rats (175 g) with the use of potassium gluconate containing electrodes. Monopolar electrical stimulations (0-200 µA, 0.05 ms) were delivered with a tungsten microelectrode positioned in dendritic layers and synaptic responses were monitored in current clamp.
16 traces). Individual traces with superimposed spontaneous IPSCs or with failures were rejected from analysis. The peak amplitude was taken as the difference in mean current between a 1-ms time window before the stimulus artifact and an 0.5-ms window at the peak of the averaged IPSC. The decay time constant of uIPSCs was determined by exponential fitting (pClamp). In all cases, visual inspection indicated that the decay of averaged uIPSCs was well fitted by a single exponential. Group measures were expressed as means ± SE. Statistical differences between means of groups was determined with Student's t-tests (significance level P < 0.05). For amplitude distributions of uIPSCs, statistical differences were assessed with Kolmogorov-Smirnov tests. The GABA antagonists bicuculline methiodide (Sigma, 25 µM) and CGP 55845A (Ciba-Geigy, 5 µM) were bath applied in ACSF.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
; Williams et al. 1994). A patch pipette containing extracellular solution was positioned in the loose cell-attached configuration onto an interneuron for stimulation, and whole cell voltage-clamp recordings were obtained from a CA1 pyramidal cell in the presence of the glutamate receptor antagonists 20 µM CNQX and 50 µM AP-5 to eliminate excitatory synaptic activity (Fig. 1; see METHODS). With the pyramidal cell held at +20 mV, the stimulus intensity was increased until an IPSC was evoked in an all-or-none fashion (Fig. 1, B and C). Further increases in stimulation intensity did not result in increases in IPSC amplitude, suggesting that a single presynaptic cell had been recruited. These IPSCs were referred to as uIPSCs. uIPSCs evoked by all interneurons fluctuated in amplitude, and occasionally showed failure of transmission. In some cases, particularly for LM interneurons, uIPSCs disappeared abruptly after a few minutes as a result of cell damage; these recordings were not included in the study. With the use of this approach, 64 synaptic connections were identified between pyramidal cells and interneurons in PYR (n = 22), OR (n = 18), and LM (n = 24).
View larger version (23K):
[in a new window]
FIG. 1.
All-or-none recruitment of inhibitory postsynaptic currents (IPSCs) by stimulation of individual interneurons. A: diagram of experimental procedures. Under visual guidance, whole cell recordings were obtained from a CA1 pyramidal cell while an individual interneuron was stimulated extracellularly with a loose cell-attached patch electrode containing artificial cerebrospinal fluid (ACSF). B: superimposed responses, evoked by stimulation of a stratum oriens (OR) interneuron at increasing intensities (0-3 µA, 
), show the all-or-none recruitment of unitary IPSCs (uIPSCs) in a pyramidal cell held at +20 mV. C: graph of uIPSC peak amplitude vs. stimulus intensity. IPSCs were evoked only at stimulation intensities >1 µA and did not increase further in amplitude with increases in stimulation intensity (1.5-3 µA). Vhold, holding potential.
View larger version (28K):
[in a new window]
FIG. 2.
Properties of uIPSCs evoked by stimulation of different interneuron subtypes. A-C: representative examples of uIPSCs (Vhold = 20 mV) evoked by stimulation (0.2 Hz) of interneurons near stratum pyramidale (PYR, A), in OR (B), and near the stratum radiatum and lacunosum-moleculare border (LM, C). Superimposed single traces (left) and averaged uIPSCs (right) were similar for responses evoked by activation of PYR (A) and OR (B) cells, but were smaller in amplitude and had a slower time course in responses elicited by stimulation of LM cells (C). D-F: summary histograms, for all cells tested, of mean ± SE peak amplitude (D), rise time (E), and decay time constant (F) of averaged uIPSCs evoked from different interneuron subtypes. Asterisks: significant differences in amplitude, rise time, and decay of uIPSCs between cell types. G: graph of decay time constant as a function of rise time for uIPSCs of all cell types. The positive correlation between decay and rise time was not statistically significant.
53.1 ± 2.1 mV (range: 45.4 to 59.7 mV, n = 9). For each type of interneurons tested, mean Erevs were 54.0 ± 4.0 mV for PYR (n = 2), 55.0 ± 5.0 mV for OR (n = 3) and 48.8 ± 2.4 mV for LM (n = 4) cells. With the use of intracellular solutions with cesium methanesulfonate to block potassium conductances, the mean Erev of uIPSCs for all cell types was not significantly different (58.5 ± 2.6 mV, range: 45.3 to 70.4 mV, n = 13, P > 0.05) (Fig. 3). With this recording solution, the mean Erevs of uIPSCs for each cell type were 58.3 ± 12.5 mV for PYR (n = 9), 57.5 ± 12.5 mV for OR (n = 2), and 62.5 ± 2.5 mV for LM (n = 2) cells. These Erev values were generally more depolarized than the calculated chloride equilibrium potential in our experimental conditions (68 mV for potassium gluconate solution and 74 mV for cesium methanesulfonate solution). However, the lack of significant difference between the mean Erev with the use of these two recording solutions suggested that potassium conductances were not involved in these uIPSCs. The current-voltage relation of uIPSCs was nonlinear (Fig. 3) and showed outward rectification in 14 of 22 cell tested. The kinetics of uIPSCs appeared voltage sensitive (Fig. 3). The mean rise time of uIPSCs for all cell types was not significantly different at 20 mV (4.5 ± 0.9 ms) and +40 mV (3.8 ± 0.8 ms, n = 10; P > 0.05). The mean decay time constant was significantly slower at +40 mV (53.9 ± 7.2 ms) than 20 mV (39.8 ± 7.2 ms, n = 13, P < 0.05).
View larger version (18K):
[in a new window]
FIG. 3.
Current-voltage relation of uIPSCs. A: superimposed averaged uIPSCs evoked by stimulation of a PYR cell at holding potentials between 80 and +30 mV, recorded with cesium methanesulfonate solution. B: plot of peak amplitude vs. holding potential of averaged uIPSCs shown in A. The reversal potential was close to the Cl equilibrium potential. C and D: graphs of mean decay time constant (C) and rise time (D) vs. membrane potential for pooled data (n = 13 and 10 cells, respectively). Decay time constant was significantly increased with depolarization (C; asterisk indicates significant statistical difference, P < 0.05), but rise time did not significantly change (D).
View larger version (19K):
[in a new window]
FIG. 4.
Block of uIPSCs by a -aminobutyric acid-A (GABAA) receptor antagonist. A-C: superimposed single traces (left, number of traces indicated in parentheses) and averaged uIPSCs (right) evoked from a PYR interneuron and recorded at +20 mV with the use of a cesium methanesulfonate-containing electrode. uIPSCs were completely blocked by bicuculline (B) and the antagonism was reversible (C).
73.5 ± 6.5 mV and a slow component with a null potential of 82.5 ± 5.0 mV (n = 2, Fig. 5A). In the presence of bicuculline (25 µM), only slow monosynaptic IPSPs (Fig. 5B) were present, showing a reversal potential of 79.3 ± 3.7 mV (n = 3).
View larger version (15K):
[in a new window]
FIG. 5.
Compound monosynaptic inhibitory postsynaptic potentials (IPSPs) in CA1 pyramidal cells of mature rats. A: in the presence of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and (±)2-amino-5-phosphonopentanoic acid (AP-5), electrical stimulation of inhibitory fibers elicited biphasic IPSPs consisting of an early ( 
) and a late (
) component. The early IPSP reversed between 52 and 86 mV, whereas the late IPSP became null near 86 mV. B: with the addition of 25 µM bicuculline, only the late IPSP remained, which reversed between 86 and 96 mV. Each trace is an average of 3 responses.
View larger version (20K):
[in a new window]
FIG. 6.
Tonic presynaptic inhibition of uIPSCs by GABAB receptors. A-C: single traces (left) and averaged uIPSCs (right) evoked by stimulation of a PYR interneuron and recorded with cesium methanesulfonate intracellular solution. After bath application of 5 µM CGP 55845A, individual and averaged uIPSCs were larger in amplitude (B) compared with control (A). The effect of the antagonist was partially reversible after 20 min of washout (C).
View larger version (15K):
[in a new window]
FIG. 7.
Amplitude distribution of uIPSCs in CGP 55845A. A: cumulative probability distribution of uIPSC amplitude in control ( 
) and in CGP 55845A (- - -). Number of events was 70 and 56, respectively, in each condition. The distribution was significantly shifted toward higher values in the GABAB antagonist. B: histograms of uIPSC amplitude distribution in control (top) and CGP 55845A (bottom). Although smaller events appeared similar in amplitude in both conditions, the amplitude of larger uIPSCs was greater in CGP 55845A (>100 pA) than in control (<100 pA).
), we examined whether uIPSCs displayed paired pulse inhibition at interstimulus intervals of 100 and 200 ms during recordings with cesium methanesulfonate intracellular solution. No significant differences were found between the amplitude of first and second averaged uIPSCs or between ensembles of first and second uIPSCs (Fig. 8A) for all presynaptic cells tested (n = 5 PYR and 4 OR cells). The difference between the mean amplitude of averaged uIPSC1 (127.2 ± 24.9 pA) and uIPSC2 (135.0 ± 26.9 pA) was not significant (P > 0.05). However, because at excitatory synapses, the amplitudes of the second responses evoked by paired pulse stimulation are dependent on the probability of transmitter release during the first responses, the paired pulse ratio of individual responses (uIPSC2/uIPSC1 × 100) was examined as a function of the amplitude of uIPSC1 (Debanne et al. 1996). In all cells tested, a significant negative correlation was found between paired pulse ratio and uIPSC1 amplitude, which was well fitted by a hyperbolic function [y = (a/x) + b] (Debanne et al. 1996) (Fig. 8, C and D). When the quantal content of the first respone was low (i.e., amplitude of uIPSC1 < mean amplitude of all uIPSCs1), the amplitude of uIPSC2 was larger than that of uIPSC1 (Fig. 8, B and C). For the nine cells examined, when uIPSC1 was smaller than the mean of all uIPSCs1, the mean paired pulse ratio was 154.0 ± 8.0%. In addition, the mean amplitude of uIPSC2 for this group of responses (167.05 ± 25.1 pA) was significantly larger than the mean amplitude of all uIPSCs1 (138.7 ± 22.4 pA). Thus analysis of individual responses revealed paired pulse facilitation when the quantal content of the first response was low. There was a tendency toward paired pulse inhibition when the quantal content of the first response was high, but this was not significant (Fig. 8D). For all cells, the mean paired pulse ratio was 90.8 ± 4.0% when the amplitude of uIPSC1 was larger than the mean amplitude of all uIPSCs1. But, for this group of responses, the mean amplitude of uIPSC2 (142.8 ± 24.4 pA) was not significantly different from the mean amplitude of all uIPSCs1, indicating an absence of paired pulse inhibition in these conditions.
View larger version (23K):
[in a new window]
FIG. 8.
Paired pulse stimulation. A: paired stimulations were delivered at 100-ms interval to an OR interneuron. A subset of individual uIPSCs (n = 14 traces) is superimposed at left, and the average uIPSCs for the total series (n = 64 traces) are displayed at right. In the average response, the amplitudes of the 1st and 2nd uIPSCs were similar. The mean peak amplitude of ensembles of uIPSCs1 and uIPSCs2 were not significantly different (P > 0.05). B-D: analysis of single events evoked by paired pulse stimulation showed an inverse relation between paired pulse ratio and the amplitude of uIPSC1. B: in single events, when the 1st uIPSC was large, the 2nd uIPSC was generally similar (left trace); but when the 1st uIPSC was small, the 2nd uIPSC was larger (right trace). C: graph of the paired-pulse ratio (uIPSC2/uIPSC1 × 100) vs. the amplitude of the 1st uIPSC for all events of the cell illustrated in A and B. Paired pulse facilitation can be clearly seen when uIPSC1 was small. The relation was well described by a hyperbolic function [y = (a/x) + b], with a = 10944.88 and b = 33.16 (coefficient of correlation, r = 0.98). D: summary of paired-pulse ratio relation for all cells examined (n = 9). The amplitudes of uIPSCs1 have been normalized to the largest response for each cell. Means ± SE of paired-pulse ratios were taken for bins of 10% of normalized uIPSC1 amplitude. Although paired pulse facilitation was found when uIPSC1 was small, no significant evidence of paired pulse inhibition was observed.
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
55 mV, and were unaltered by the K+ conductance blocker cesium. Thus all unitary responses were mediated via GABAA receptors and Cl channels. uIPSCs originating from LM interneurons were smaller in amplitude and had a slower time course. Therefore synaptic mechanisms originating from different interneurons may be heterogeneous. Finally, paired pulse stimulation of interneurons resulted in facilitation, but not in depression, of uIPSCs, suggesting that tonic presynaptic GABAB inhibition of unitary responses was heterosynaptic.
; Miles 1990). Our results provide evidence of GABAA postsynaptic mechanisms at synapses made by OR and LM interneurons. Vertical (Lacaille et al. 1987) and horizontal (Maccaferri and McBain 1995) cells in OR are involved in recurrent inhibition of pyramidal cells. Thus the GABAA inhibition found to originate from OR interneurons provides direct evidence that recurrent inhibition may solely involve GABAA postsynaptic mechanisms (Alger and Nicoll 1982a; Newberry and Nicoll 1984; Samulack and Lacaille 1993). Because axons of OR and LM interneurons contact pyramidal cell dendrites extensively (Kunkel et al. 1988; Lacaille et al. 1987; Lacaille and Schwartzkroin 1988a; McBain et al. 1994; Sik et al. 1995), a large number of interneurons will provide significant dendritic GABAA inhibition of pyramidal cells, in addition to the well-characterized somatic inhibition (Andersen et al. 1964a,b; Buhl et al. 1994a).
). Our observations of GABAA uIPSCs originating from LM interneurons indicate that, in addition to feed-forward GABAB inhibition (Alger and Nicoll 1982a; Newberry and Nicoll 1984), some feed-forward inhibitory pathways to pyramidal cells may involve solely GABAA postsynaptic mechanisms. Because stellate cells also project todentate gyrus and CA3 area (Kunkel et al. 1988; Williams et al. 1994), the synchronization they provide across hippocampal regions may also involve GABAA mechanisms.
). However, many factors may account for these differences. First, some LM cells did not tolerate the present stimulation procedure. Thus different presynaptic cells may have been sampled. Second, animals of different ages were used (juvenile vs. mature). Regional differences have been observed in the development of postsynaptic GABAB responses. Postsynaptic GABAB receptors became functional ~11 days postnatally in rat somatosensory and visual cortex (Luhmann and Prince 1991), and ~6 days postnatally in rat CA3 pyramidal cells (Gaiarsa et al. 1995), but only between 12-30 days postnatally in rabbit CA1 pyramidal cells (Janigro and Schwartzkroin 1988). Thus, in the present study, the absence of postsynaptic GABAB components could be related to their late maturation in CA1 pyramidal cells. Third, different methods of stimulation, single-cell versus local glutamate, were used. Indeed, strong electrical stimulation evokesGABAB responses, whereas weak stimulation, or spontaneous activation of single cells or synapses, elicits GABAA responses (Dutar and Nicoll 1988a,b; Otis and Mody 1992a). Also, inhibition of GABA uptake can recruitGABAB components in evoked IPSPs, but not in miniature events (Thompson and Gähwiler 1992). Thus GABAB receptor activation may require a large release of GABA and extrasynaptic diffusion, or the release of a cofactor (Mody et al. 1994). Specialized interneurons responsible for GABAB inhibition have, however, been reported (Segal 1990; Sugita et al. 1992).
; Sik et al. 1995), axon initial segments (axoaxonic cells, Buhl et al. 1994a,b), or basal and apical dendrites (bistratified cells, Buhl et al. 1994a). In contrast, OR interneurons contact basal and apical dendrites of pyramidal cells (vertical cells, Lacaille et al. 1987; Sik et al. 1995), or specifically their distal apical dendrites (horizontal cells, McBain et al. 1994). The similar kinetics of unitary synaptic currents from PYR and OR interneurons observed in the present study are in contrast to the fast and slow time course of synaptic potentials generated by interneurons making somatic and dendritic synapses that were reported in current-clamp recordings (Buhl et al. 1994a). This would suggest that the different time course of synaptic potentials originating from dendritic and somatic synapses (Buhl et al. 1994a) may be due to electrotonic filtering of more distant dendritic synapses and not to differences in underlying synaptic currents. Finally, in the present study, the observed reversal potentials of uIPSCs (about 55 mV) were more depolarized than the calculated chloride equilibrium potential (about 70 mV). Because GABA responses generated in dendrites of CA1 pyramidal cells can be depolarizing (Andersen et al. 1980), the more depolarized reversal potential of uIPSCs might reflect the involvement of depolarizing dendritic GABA synapses in these responses.
), because LM cell axons contact mostly pyramidal cell apical dendrites (stellate cells, Kunkel et al. 1988; Lacaille and Schwartzkroin 1988a; Misgeld and Frotscher 1986). However, the K+ conductance blocker cesium was used to improve voltage control. Also, the similar reversal potential of uIPSCs and the nonsignificant correlation between rise time and decay are suggestive of appropriate dendritic voltage control (Hestrin et al. 1990; but see Spruston et al. 1993). More importantly, because the electrotonic location of some OR interneuron synapses may be equally or more distant from pyramidal cell somata than LM cell synapses (Lacaille et al. 1987; McBain et al. 1994; Sik et al. 1995), and because uIPSCs from OR interneurons displayed faster kinetics and larger amplitude than their LM interneuron counterparts, poor voltage-clamp control cannot explain totally the differences in kinetics and properties. Identification of synaptic contact sites by individual interneurons combined with uIPSC characterization should help resolve this issue.
). Additionally, skewed distributions of miniature IPSC decay rate, suggestive of distinct underlying populations, have been observed in hippocampal granule cells (Otis and Mody 1992b). Our results are consistent with two such distinct GABAA conductances, but also suggest that PYR and OR interneurons may be involved in fast GABAA responses and LM cells in slow responses. Single-channel recordings and rapid applications of GABA in hippocampal and cerebellar granule cells have uncovered channels with fast and slow kinetics, compatible with biexponentially decaying GABAA IPSCs (Edwards et al. 1990; Puia et al. 1994). However, uIPSC decay was monoexponential and lacking a fast component in the present study. If the slower kinetics of uIPSCs of LM interneurons reflect different receptor mechanisms, then these slower receptor mechanisms could contribute, in addition to electrotonic factors affecting distant dendritic synapses, to the slow time course and attenuation of synaptic potentials generated by LM interneurons (Lacaille and Schwartzkroin 1988b).
). Synaptic currents with slower kinetics, originating from LM interneurons, may effectively inhibit slower NMDA-mediated excitatory events (Davies et al. 1991; Hestrin et al. 1990) or local Ca2+ responses (Masukawa and Prince 1984; Miyakawa et al. 1992).
) in the presence of intracellular cesium suggested tonic presynaptic GABAB inhibition. Tonic GABAB inhibition of uIPSPs originating from basket cells has been reported (Buhl et al. 1995). However, our results ruled out postsynaptic sites of actions and further suggest that tonic presynaptic GABAB inhibition of GABA release may be generalized to many types of interneurons.
). This discrepancy may be due to the different stimulation protocols used. Others have also reported that fast IPSCs evoked by minimal stimulation in PYR do not show paired pulse depression within the100- to 200-ms range, but that slow IPSCs evoked by minimal stimulation in stratum lacunosum-moleculare displayGABAB-dependent paired pulse depression (Pearce et al. 1995). In the present study, only uIPSCs evoked by stimulation of interneurons in PYR and OR were tested with paired pulse stimulation. These uIPSCs may therefore correspond to the fast IPSCs that do not show paired pulse inhibition (Pearce 1993; Pearce et al. 1995). GABAB-independent paired pulse depression has also been described for IPSCs in hippocampus (Lambert and Wilson 1994; Pearce et al. 1995; Wilcox and Dichter 1994). This paired pulse depression is dependent on the probability of release during the first response (Lambert and Wilson 1994; Wilcox and Dichter 1994). Manipulations that reduced release probability decreased paired pulse depression and resulted in paired pulse facilitation (Lambert and Wilson 1994; Wilcox and Dichter 1994). In the present work, no significant paired pulse inhibition was observed. Individual inhibitory interneurons in the CA1 area can make many synaptic contacts (up to 12) with a single pyramidal cell (Buhl et al. 1994a), and these may not all be active each time a presynaptic cell is stimulated. In our experimental conditions, release probability for individual interneurons may have been low, because uIPSCs showed large fluctuations in amplitude with occasional transmission failures. In addition, the presence of tonic presynaptic GABAB inhibition likely contributed to reducing release probability. Thus differences in release probability under the various experimental conditions of each study may explain the discrepancy between our results and those of others (Lambert and Wilson 1994; Wilcox and Dichter 1994). Paired pulse stimulation of individual interneurons during manipulations that increase probability of release should help resolve this issue.
; Yoon and Rothman 1991). Tonic presynaptic GABAB inhibition was not observed in some cells (see also Buhl et al. 1995), and presynaptic GABAB inhibition may be selective for distinct subsets of inhibitory fibers (Lambert and Wilson 1993; Pearce et al. 1995). Thus heterosynaptic GABAB mechanisms may inhibit GABA release only in subsets of interneurons. The functional consequences of such complex presynaptic modulation of inhibitory cells remain unclear, yet it points to a tremendous flexibility of local inhibitory processes in the hippocampus.
| |
ACKNOWLEDGEMENTS |
|---|
The authors thank Ciba-Geigy for the generous gift of CGP 55845A.
This work was supported by grants from the Medical Research Council of Canada (MT-10848 to J.-C. Lacaille) and the Fonds pour la Formation de Chercheurs et l'Aide à la Recherche (FCARÉquipe de Recherche). J.-C. Lacaille is a senior scholar of the Fonds de la Recherche en Santé du Québec and a member of the FCARGroupe de Recherche sur le Système Nerveux Central (GRSNC). M. Ouardouz was supported by a postdoctoral fellowship from the FCAR (GRSNC) and the Savoy Foundation.
| |
FOOTNOTES |
|---|
Address for reprint requests: J.-C. Lacaille, Dept. de Physiologie-124, Université de Montréal, C.P. 6128, Succ. Centre-ville, Montreal, Quebec H3C 3J7, Canada.
Received 14 May 1996; accepted in final form 5 December 1996.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
New Vistas,
edited by
V. Chan-Palay,
and C. Kohler
. New York: Liss, 1989, p. 287-305 currents in Purkinje versus granule neurons in rat cerebellar slices.
Neuron
12: 117-126, 1994.[Medline]This article has been cited by other articles:
|
|
 |
|
  C. J. Price, R. Scott, D. A. Rusakov, and M. Capogna GABAB Receptor Modulation of Feedforward Inhibition through Hippocampal Neurogliaform Cells J. Neurosci., July 2, 2008; 28(27): 6974 - 6982. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. K. Andrasfalvy and I. Mody Differences between the scaling of miniature IPSCs and EPSCs recorded in the dendrites of CA1 mouse pyramidal neurons J. Physiol., October 1, 2006; 576(1): 191 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Price, B. Cauli, E. R. Kovacs, A. Kulik, B. Lambolez, R. Shigemoto, and M. Capogna Neurogliaform Neurons Form a Novel Inhibitory Network in the Hippocampal CA1 Area J. Neurosci., July 20, 2005; 25(29): 6775 - 6786. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. M. Baker, P. S. Pennefather, B. A. Orser, and F. K. Skinner Disruption of Coherent Oscillations in Inhibitory Networks With Anesthetics: Role of GABAA Receptor Desensitization J Neurophysiol, November 1, 2002; 88(5): 2821 - 2833. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Xiang, J. R. Huguenard, and D. A. Prince Synaptic Inhibition of Pyramidal Cells Evoked by Different Interneuronal Subtypes in Layer V of Rat Visual Cortex J Neurophysiol, August 1, 2002; 88(2): 740 - 750. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. V. Buhler and T. V. Dunwiddie alpha 7 Nicotinic Acetylcholine Receptors on GABAergic Interneurons Evoke Dendritic and Somatic Inhibition of Hippocampal Neurons J Neurophysiol, January 1, 2002; 87(1): 548 - 557. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kim and B. E. Alger Random Response Fluctuations Lead to Spurious Paired-Pulse Facilitation J. Neurosci., December 15, 2001; 21(24): 9608 - 9618. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L A Martin, D-S Wei, and B E Alger Heterogeneous susceptibility of GABAA receptor-mediated IPSCs to depolarization-induced suppression of inhibition in rat hippocampus J. Physiol., May 1, 2001; 532(3): 685 - 700. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Bertrand and J.-C. Lacaille Unitary synaptic currents between lacunosum-moleculare interneurones and pyramidal cells in rat hippocampus J. Physiol., April 15, 2001; 532(2): 369 - 384. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Chery and Y. De Koninck GABAB Receptors Are the First Target of Released GABA at Lamina I Inhibitory Synapses in the Adult Rat Spinal Cord J Neurophysiol, August 1, 2000; 84(2): 1006 - 1011. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. White, M. I. Banks, R. A. Pearce, and N. J. Kopell Networks of interneurons with fast and slow gamma -aminobutyric acid type A (GABAA) kinetics provide substrate for mixed gamma-theta rhythm PNAS, June 23, 2000; (2000) 100124097. [Abstract] [Full Text]
|
|
|
|
|
|
G. Maccaferri, J David, B Roberts, P. Szucs, C. A Cottingham, and P. Somogyi Cell surface domain specific postsynaptic currents evoked by identified GABAergic neurones in rat hippocampus in vitro J. Physiol., April 1, 2000; 524(1): 91 - 116. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Jiang, S. Sun, M. Nedergaard, and J. Kang Paired-pulse modulation at individual GABAergic synapses in rat hippocampus J. Physiol., March 1, 2000; 523(2): 425 - 439. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Dvorak-Carbone and E. M. Schuman Patterned Activity in Stratum Lacunosum Moleculare Inhibits CA1 Pyramidal Neuron Firing J Neurophysiol, December 1, 1999; 82(6): 3213 - 3222. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Chapman and J.-C. Lacaille Cholinergic Induction of Theta-Frequency Oscillations in Hippocampal Inhibitory Interneurons and Pacing of Pyramidal Cell Firing J. Neurosci., October 1, 1999; 19(19): 8637 - 8645. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Bracci, M. Vreugdenhil, S. P. Hack, and J. G. R. Jefferys On the Synchronizing Mechanisms of Tetanically Induced Hippocampal Oscillations J. Neurosci., September 15, 1999; 19(18): 8104 - 8113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Wierenga and W. J. Wadman Miniature Inhibitory Postsynaptic Currents in CA1 Pyramidal Neurons After Kindling Epileptogenesis J Neurophysiol, September 1, 1999; 82(3): 1352 - 1362. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Chapman and J.-C. Lacaille Intrinsic Theta-Frequency Membrane Potential Oscillations in Hippocampal CA1 Interneurons of Stratum Lacunosum-Moleculare J Neurophysiol, March 1, 1999; 81(3): 1296 - 1307. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Woodhall, C. E. Gee, R. Robitaille, and J.-C. Lacaille Membrane Potential and Intracellular Ca2+ Oscillations Activated by mGluRs in Hippocampal Stratum Oriens/Alveus Interneurons J Neurophysiol, January 1, 1999; 81(1): 371 - 382. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Pham, S. Nurse, and J.-C. Lacaille Distinct GABAB Actions Via Synaptic and Extrasynaptic Receptors in Rat Hippocampus In Vitro J Neurophysiol, July 1, 1998; 80(1): 297 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. I. Banks, T.-B. Li, and R. A. Pearce The Synaptic Basis of GABAA,slow J. Neurosci., February 15, 1998; 18(4): 1305 - 1317. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Vida, K. Halasy, C. Szinyei, P. Somogyi, and E. H Buhl Unitary IPSPs evoked by interneurons at the stratum radiatum-stratum lacunosum-moleculare border in the CA1 area of the rat hippocampus in vitro J. Physiol., February 1, 1998; 506(3): 755 - 773. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. White, M. I. Banks, R. A. Pearce, and N. J. Kopell Networks of interneurons with fast and slow gamma -aminobutyric acid type A (GABAA) kinetics provide substrate for mixed gamma-theta rhythm PNAS, July 5, 2000; 97(14): 8128 - 8133. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
