Download Glutamate Controls the Induction of GABA

Glutamate Controls the Induction of GABA-Mediated Giant
Depolarizing Potentials Through AMPA Receptors in Neonatal Rat
Hippocampal Slices
SONIA BOLEA,1 ELENA AVIGNONE,2 NICOLA BERRETTA,2 JUAN V. SANCHEZ-ANDRES,1 AND
ENRICO CHERUBINI2
1
Departamento de Fisiologia, Instituto de Bioingenieria, Universidad Miguel Hernandez, Campus de San Juan, 03550 San
Juan, Alicante, Spain; and 2Neuroscience Program and Istituto Nazionale Fisica della Materia Unit, International School
for Advanced Studies, 34014 Trieste, Italy
INTRODUCTION
A peculiar characteristic of the hippocampus of neonatal rats is
the presence of spontaneous network-driven GABA-mediated oscillatory events, the so-called giant depolarizing potentials (GDPs)
(Ben Ari et al. 1989; Xie et al. 1994). These appear to be
synchronous within the entire hippocampus (Menendez de la
Prida et al. 1998; Strata et al. 1997) and are generated by GABA
(acting on GABAA receptors) that at this early developmental
stage is depolarizing and excitatory (Cherubini et al. 1991).
GABA-induced depolarization represents a general feature of
postnatal development (Ben Ari et al. 1989; Chen et al. 1996;
Cherubini et al. 1991; Kaneda et al. 1996; Owens et al. 1996;
Rohrbough and Spitzer 1996; Serafini et al. 1995; Wu et al. 1992).
GABA, through its depolarizing action, activates voltagedependent calcium channels, thereby increasing the levels of
[Ca21]i (Leinekugel et al. 1995). Therefore GDPs are strictly
associated with spontaneous calcium oscillations occurring in
groups of neighboring cells (Garaschuk et al. 1998). These
highly correlated calcium signals are thought to be essential for
consolidation of synaptic connections and development of the
adult neuronal network (Shatz 1990). However, GDPs need a
glutamatergic drive for their expression. In previous studies
from this (Strata et al. 1995) and other laboratories (Ben Ari et
al. 1989; Gaiarsa et al. 1991) it has been shown that GDPs can
be modulated by activation of both ionotropic and metabotropic glutamate receptors. More recently, it was suggested that
synchronization is determined by cooperation of excitatory
GABAergic connections between interneurons and glutamatergic connections presumably from pyramidal cells to interneurons (Khazipov et al. 1997). Hence in interneurons a glutamatergic component of the GDP could be revealed after
intracellular blockade of GABAA receptors with an internal
solution containing fluoride (Khazipov et al. 1997). In particular, it was suggested that the main contribution to GDP
generation is provided by N-methyl-D-aspartate (NMDA) receptors that would be activated after GABA-induced depolarization, whereas AMPA receptors would play only a minor role
(Ben Ari et al. 1997; Leinekugel et al. 1997; McLean et al. 1995).
The current experiments were undertaken to further elucidate the role of glutamatergic receptors in GDP generation. In
contrast to previously published data (Khazipov et al. 1997;
Leinekugel et al. 1997) it was found that AMPA receptors play
a major role in neuronal synchronization.
Part of this work was reported in preliminary form (Bolea et
al. 1998).
METHODS
Slice preparation
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Experiments were performed on hippocampal slices obtained from
postnatal P1–P6 Wistar rats (P0 is the day of birth) according to the
0022-3077/99 $5.00 Copyright © 1999 The American Physiological Society
2095
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 15, 2017
Bolea, Sonia, Elena Avignone, Nicola Berretta, Juan V. SanchezAndres, and Enrico Cherubini. Glutamate controls the induction of
GABA-mediated giant depolarizing potentials through AMPA receptors in neonatal rat hippocampal slices. J. Neurophysiol. 81: 2095–2102,
1999. Giant depolarizing potentials (GDPs) are generated by the
interplay of the depolarizing action of GABA and glutamate. In this
study, single and dual whole cell recordings (in current-clamp configuration) were performed from CA3 pyramidal cells in hippocampal
slices obtained from postnatal (P) days P1- to P6-old rats to evaluate
the role of ionotropic glutamate receptors in GDP generation. Superfusion of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) (10 – 40
mM) completely blocked GDPs. However, in the presence of CNQX,
it was still possible to re-induce the appearance of GDPs with GABA
(20 mM) or (RS)-a-amino-3-hydroxy-5-methyl-4-isoxadepropionate
(AMPA) (5 mM). This effect was prevented by the more potent and
selective AMPA receptor antagonist GYKI 53655 (50 –100 mM). In
the presence of GYKI 53655, both kainic or domoic acid (0.1–1 mM)
were unable to induce GDPs. In contrast, bath application of
D-(2)-2-amino-5-phosphonopentanoic acid (50 mM) or (1)-3-(2carboxy-piperazin-4-yl)-propyl-L-phosphonic acid (20 mM) produced
only a 37 6 9% (SE) and 36 6 11% reduction in GDPs frequency,
respectively. Cyclothiazide, a selective blocker of AMPA receptor
desensitization, increased GDP frequency by 76 6 14%. Experiments
were also performed with an intracellular solution containing KF to
block GABAA receptor-mediated responses. In these conditions, a
glutamatergic component of GDP was revealed. GDPs could still be
recorded synchronous with those detected simultaneously with KClfilled electrodes, although their amplitude was smaller. Similar results
were found in pair recordings obtained from minislices containing
only a small portion of the CA3 area. These data suggest that GDP
generation requires activation of AMPA receptors by local release of
glutamate from recurrent collaterals.
2096
BOLEA, AVIGNONE, BERRETTA, SANCHEZ-ANDRES, AND CHERUBINI
nous in pair of CA3 recordings, 3) they are reversibly blocked
by a high-magnesium, low-calcium solution, and 4) they are
abolished by TTX. These events are GABAA mediated because
they are reversibly blocked by bicuculline. In a first set of
experiments, the role of ionotropic glutamate receptors in GDP
generation was reexamined, taking also advantage of new
pharmacological tools. Then, in a second series of experiments,
the conductances activated during GDPs in pyramidal cells
were studied after intracellular blockade of GABAA-receptor–
evoked responses with fluoride.
Patch-clamp whole cell recordings
AMPA receptor activation is necessary for GDP induction
Spontaneous or evoked GDPs were recorded in the whole cell
configuration of the patch-clamp technique (current-clamp mode)
with a standard amplifier (Axoclamp 2B, Axon Instruments; Foster
City, CA). Patch electrodes had a resistance of 3– 6 MV when filled
with an intracellular solution containing (in mM) 140 KCl or KF, 1
MgCl2, 1 NaCl, 1 EGTA, 5 HEPES, and 2 K2ATP; pH was adjusted
to 7.3 with KOH. GDPs were evoked with bipolar twisted NiCrinsulated electrodes (50 mm OD) placed in the hilus. Pulses of
variable amplitude (5–15 V, 100-ms duration) were applied. Membrane input resistance was measured from the amplitude of small
hyperpolarizations (300-ms duration) evoked by passing current
pulses across the cell membrane.
To test the role of AMPA receptors in GDPs induction, the
selective AMPA– kainate receptor antagonist CNQX (10 – 40
mM) was applied (n 5 11). As reported earlier (Gaiarsa et al.
1991), this compound completely blocked spontaneous GDPs
(data not shown), even in rats as young as P2. A full recovery
was obtained 10 –15 min after wash. However, in the presence
of CNQX, it was still possible to re-induce the appearance of
spontaneous GDPs with GABA (20 mM) or AMPA (5 mM)
and to evoke GDPs by focal stimulation. The effect of AMPA
was usually associated to a 5- to 12-mV membrane depolarization. Both GDPs as well as membrane depolarization disappeared when AMPA was applied in the presence of bicuculline (10 mM, data not shown), suggesting that both phenomena
were due to the release of GABA after an incomplete block of
AMPA receptors with CNQX. Similar results were obtained
with kynurenic acid (2 mM, n 5 8), DNQX (20 mM, n 5 2),
and GYKI 52466 (20 –100 mM, n 5 6). The possibility to
synchronize again the network with AMPA suggests that
AMPA receptors were not completely blocked by the selective
antagonists used. To check whether this hypothesis was correct, a more potent and selective AMPA antagonist, GYKI
53655, was tested (Lerma et al. 1997). GYKI 53655 (50 –100
mM) blocked spontaneous GDPs as well as those evoked by
electrical stimulation of the hilus. Moreover, AMPA (5 mM) or
GABA (20 mM) application were unable to re-induce GDPs
(Fig. 1). These data suggest that the glutamatergic drive via
AMPA receptors is essential for GDP generation.
To further examine the role of AMPA receptors in GDPs
induction, experiments were performed in the presence of
cyclothiazide, a selective blocker of AMPA receptor desensitization (Partin et al. 1993). In the presence of cyclothiazide
(20 mM) GDPs frequency increased by 76 6 14% (n 5 6, Fig.
2). Kainate receptors were not apparently involved in GDPs
induction because in the presence of GYKI 53655 both kainate
and domoic acid (0.1–1 mM) failed to induce GDPs (n 5 3,
data not shown).
In contrast to AMPA receptor antagonists, the selective
NMDA receptor antagonist D-AP5 (50 mM) applied in the bath
for 10 –15 min produced only a reduction of GDP frequency
(Fig. 3). Occasionally a complete blockade was observed during the first minutes of drug application. Later on, in the
continuous presence of D-AP5, GDP frequency slowly returned
toward control values. On average a 35% reduction in GDP
frequency was measured along the time of drug application.
Similar results were obtained with CPP (20 mM). No significant changes in the shape of GDPs were observed during
application of CPP or D-AP5. These data suggest that, although
NMDA receptors may contribute to GDP generation, they are
Drugs
Drugs were dissolved in ACSF and applied in the bath by changing
the superfusion solution to one that differed only in its content of
drug(s). The ratio of flow rate to bath volume ensured complete
exchange within 1 min. Drugs used were (RS)-a-amino-3-hydroxy5-methyl-4-isoxadepropionate (AMPA), (1)-3-(2-carboxy-piperazin4-yl)-propyl-1-phosphonic acid (CPP), or D-(2)-2-amino-5-phosphonopentanoic acid (D-AP5), 6-cyano-7-nitroquinoxaline-2,3-dione
(CNQX), 6,7-dinitroquinoxaline-2,3(1H,4H)-dione (DNQX), domoic
acid, kainic acid, GABA, and bicuculline (all purchased from Tocris
Cookson; Bristol, UK); kynurenic acid, TTX, and 6-chloro-3,4dihydro-3-[2-norbornen-5-yl]22H-1,2– 4benzothiadiazine-7-sulfonamide
1,1-dioxide (cyclothiazide; from Sigma; Milano, Italy); GYKI 52466
(from RBI); and GYKI 53655 (from Lilly; Indianapolis, IN). If not
otherwise stated, data are expressed as means 6 SE.
Data acquisition and analysis
Data of spontaneous and evoked GDPs were stored on a magnetic
tape and transferred to a computer after digitization with an A/D
converter (Digidata 1200). Data acquisition was done with pClamp
(Axon Instruments; Foster City, CA), and the amplitude and frequency of GDPs were analyzed with Axoscope (Axon Instruments;
Foster City, CA).
RESULTS
Stable whole cell recordings (in current-clamp configuration) lasting .30 min were obtained from 120 CA3 hippocampal pyramidal cells in slices from P1- to P6-old rats that
exhibited spontaneous GDPs. They occurred at the frequency
of 5.2 6 0.43 GDPs/min and consisted of large (30 –50 mV)
depolarizing potentials lasting 400 –700 ms, which triggered
action potentials, followed by an afterhyperpolarization. In
previous experiments (Ben Ari et al. 1989), it was reported that
GDPs are network-driven events generated by a large population of neurons firing synchronously because 1) their frequency
is independent of the membrane potential, 2) they are synchro-
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 15, 2017
methods already described (Strata et al. 1995). Briefly, animals were
decapitated after being anesthetized with intraperitoneal injection of
urethan (2 g/kg). The brain was quickly removed from the skull, and
the hippocampi were dissected free. Transverse 600-mm thick slices
were cut with a tissue chopper and maintained at room temperature
(22–24 °C) in oxygenated artificial cerebrospinal fluid (ACSF) containing (in mM) 126 NaCl, 3.5 KCl, 1.2 NaH2PO4, 1.3 MgCl2, 2
CaCl2, 25 NaHCO3, and 11 glucose, pH 7.3, saturated with 95%
O2-5% CO2. After incubation in ACSF for $1 h, an individual slice
was transferred to a submerged recording chamber, continuously
superfused at 33–34°C with oxygenated ACSF at a rate of 3 ml/min.
GDPS AND GLUTAMATE RECEPTORS
not essential for their induction. However, in the presence of
CNQX, D-AP5 was able to prevent evoked GDPs or re-induction of spontaneous oscillations by GABA and AMPA (data
not shown), indicating that in particular conditions also NMDA
receptors may contribute to network synchronization.
Intracellular blockade of GABAA receptor with fluoride
reveals an AMPA-mediated component of GDPs
In a previous work aimed at understanding synchronization
of interneuronal network (Khazipov et al. 1997), whole cell
recordings were made from interneurons dialyzed with the
poorly permeable anion F2, which produces a block of
GABAA receptors (Bormann et al. 1987). We used a similar
approach to see whether in CA3 pyramidal cells glutamate
receptors are activated during GDPs. To this purpose we tested
1) GDP reversal potential immediately and 30 – 40 min after a
full whole cell access was achieved with a F2 containing
solution and 2) the pharmacological sensitivity of GDPs to
GABAA and ionotropic NMDA and non-NMDA glutamate
receptor antagonists. Immediately (3–5 min) after breaking
into whole cell configuration, the reversal of evoked and spontaneous GDPs was 245.7 6 6.1 mV (n 5 28) and 240.8 6 6.2
mV (n 5 11), respectively (Fig. 4, A and B). Plots of amplitude–voltage relationships were always linear. The slope of the
regression lines through different data points was 20.58 6
0.19 (n 5 28) for evoked and 20.67 6 0.14 (n 5 11) for
spontaneous GDPs. After 40-min dialysis with an intracellular
F2 containing solution a complete block of GABA-evoked
responses in the presence of TTX (1 mM) was observed (Fig.
4C). The block of GABAA receptors was associated with a
significant shift of the reversal toward more positive values.
The reversal of the evoked and spontaneous GDPs was
214.1 6 11.4 mV (n 5 17) and 215.8 6 13.1 mV (n 5 6),
respectively (Fig. 4, A and B). Changes in the reversal were
associated with a reduction in the slope of the regression lines,
indicating a reduction of GDP amplitude (this was 20.29 6
0.12, n 5 17, for evoked and 20.31 6 0.13, n 5 6, for
spontaneous GDPs). In the majority of cells the amplitude–
voltage relationship was linear, whereas in four cases a rectification at more negative potentials was observed. The shift of
the reversal potential toward more positive values as well as
the reduction of the slope in the amplitude–voltage relationship
suggest the presence of a different component unmasked after
intracellular blockade of GABAA receptors with F2. To understand the nature of this component, spontaneous and evoked
GDPs were recorded at different membrane potentials in the
FIG. 2. Cyclothiazide increases the frequency of GDPs. Representative tracing
from a CA3 pyramidal cell recorded at P4 in
control condition (A) and during bath application of cyclothiazide (B). C: GDP frequency of the cell shown in A and B is
plotted against time. Each column represents
the number of GDPs recorded in 1 min.
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 15, 2017
FIG. 1. (RS)-a-Amino-3-hydroxy-5-methyl-4-isoxadepropionate (AMPA)
receptor activation is essential for Giant depolarizing potential (GDP) induction. A–D: tracings from the same cell (recorded at P5). Bath application of the
selective AMPA receptor antagonist GYKI 53655 (arrow) readily blocked
spontaneous and evoked GDPs. In the presence of GYKI 53655, electrical
stimulation failed to evoke GDPs (see inset above the trace on an expanded
time scale). In the presence of GYKY 53655, both AMPA (bar, B) and GABA
(bar, C) failed to re-induce GDPs; 30 min after washing GYKI, spontaneous
GDPs reappeared; GDP frequency increased during application of GABA (20
mM, bar).
2097
2098
BOLEA, AVIGNONE, BERRETTA, SANCHEZ-ANDRES, AND CHERUBINI
FIG. 3. N-Methyl-D-aspartate (NMDA) receptor activation is not essential for GDP generation. A: representative tracing from a CA3
pyramidal cell recorded at P4 in control condition and during bath application of D-AP5
(bar). B: 2 GDPs from the recording in A are
shown on an expanded time scale. C: GDP
frequency of the cell shown in A is plotted
against time. Each column represents the number of GDPs recorded in 1 min. D: normalized
GDP frequency recorded during application of
D-AP5 (50 mM) and CPP (20 mM). Bars indicate SE; n refers to the number of experiments
et al. 1998). In early postnatal days, when the present experiments were performed, mossy fibers were not completely developed (Gaarskjaer 1986) and therefore their contribution to
GDPs induction appears to be modest. Other sources of glutamate are the associative-commisural fibers coming from the
controlateral hippocampus as well as from collaterals of principal cells. To examine whether GDPs can be generated locally
by these fibers, in three slices, a small part of the CA3 region
was isolated with a knife cut from the rest of the hippocampus,
and pair recordings were performed from two CA3 adjacent
pyramidal cells with two intrapipette solutions containing KF
and KCl, respectively. In these conditions GDPs with characteristics similar to those already reported could be still recorded
synchronously in the two cells (Fig. 7). This suggests that a
local population of pyramidal cells and interneurons is sufficient to generate GDPs and that collaterals of principal cells are
probably the main source of glutamatergic drive.
Source of glutamatergic drive needed to synchronize
GABAergic interneurons
AMPA receptor activation is necessary for GDP induction
Glutamate may be released from mossy fibers, known to
make synaptic contacts with the proximal dendrites of CA3
pyramidal neurons and with GABAergic interneurons (Acsády
DISCUSSION
In line with a previous report (Ben Ari et al. 1997), the
current experiments clearly show that GDPs are generated by
the synergistic action of glutamate and GABA. However, in
contrast with earlier findings on interneurons (Khazipov et al.
1997; Leinekugel et al. 1997), they demonstrate that the release
of GABA is triggered by glutamate acting mainly on AMPA
type of receptors and that the ionotropic glutamatergic component of GDPs, revealed after blockade of GABAA-evoked
responses with intracellular F2, is predominantly of the nonNMDA type.
AMPA receptor activation was necessary for GDP induction
as proved by the observation that AMPA receptor antagonists
were fully effective in blocking GDPs, whereas NMDA receptor antagonists only produced a transient reduction in their
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 15, 2017
presence of AMPA and NMDA receptor antagonists. Bath
application of CNQX (10 –20 mM) completely abolished the
residual component of spontaneous GDPs and in 8 of 11
completely blocked the evoked GDPs. D-AP5 (50 mM) reduced
the residual component only in 3 of 11 (Fig. 5, A–C). In the
presence of D-AP5 and CNQX GDPs could not be evoked. It
could be that the glutamatergic component of the GDP, which
remains after blockade of GABAA receptor, is amplified as the
result of the interference of intracellular F2 with phosphatases
(Andrews and Babior 1984), enzymes that regulate receptor
desensitization (Yakel 1997). To test this hypothesis, the amplitude of AMPA responses recorded in TTX (1 mM) with a F2
containing intrapipette solution was compared with that of
responses recorded with a KCl solution. As shown in Fig. 5D,
in F2, AMPA responses were not significantly (P . 0.08)
different from those recorded in KCl.
Pair recordings performed from adjacent CA3 pyramidal
cells (n 5 21) with two different intrapipette solutions showed
that GDPs recorded with KF and KCl intrapipette solutions
were still synchronous in the two cells, confirming their combined glutamatergic and GABAergic nature (Fig. 6). However,
a clear difference in the shape of GDPs recorded with KF or
KCl was observed. As depicted in Fig. 6, in KF GDPs were
smaller in amplitude and exhibited a slower rising phase that
often did not reach the threshold for action potential generation. Moreover, when GABA (20 mM) was applied in the
presence of kynurenic acid (2 mM), known to abolish spontaneous GDPs (Strata et al. 1995), this amino acid was able to
re-induce GDPs only in the cell recorded with KCl electrode
(Fig. 6).
GDPS AND GLUTAMATE RECEPTORS
2099
frequency (see also Gaiarsa et al. 1991). In agreement with the
current data, it has been recently shown, with fluorometric
calcium imaging techniques, that CNQX is able to completely
block spontaneous oscillatory calcium transients, associated to
GDPs, recorded in the CA1 region of the hippocampus during
early postnatal development (Garaschuk et al. 1998). In previous studies a complete block of GDPs with NMDA receptors
antagonists was reported. The same block was occasionally
observed only during the first 2–3 min of drug application. At
an early stage of development, at least in the CA1 area,
glutamatergic synapses containing only NMDA receptors were
described (Durand et al. 1996). However, the observations that
at P2 it is already possible to block the entire network with
CNQX and to increase GDPs frequency with cyclothiazide
indicate that also AMPA receptors are present and functional at
this early developmental period. The contribution of NMDA
receptors to network synchronization is suggested by the experiments in which GDPs evoked by GABA or AMPA in the
presence of CNQX were abolished by addition of D-AP5 or
CPP. However, it should be stressed that in these cases AMPA
responses were not completely abolished, probably because of
AMPA receptors localized on basket cells, not entirely blocked
by CNQX. Therefore we suggest that functional AMPA receptors would provide the depolarization sufficient to remove the
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 15, 2017
FIG. 4. Intracellular blockade of GABA receptors with fluoride reveals a novel component of evoked GDPs. A: evoked GDPs
(arrows) recorded at different membrane potentials (at P4) with a patch pipette containing KF, 1 min (left) and 40 min (right) after
breaking into the whole cell configuration. B: plot of GDPs (shown in A) amplitude vs. membrane potentials. Note the shift to the
right of the reversal potential and the change in the slope of the amplitude-voltage plot, 40 min after dialyzing the cell with fluoride.
C: pairs of recordings from adjacent CA3 pyramidal cells patched with pipettes containing KCl and KF. Application of GABA (bar)
in the presence of TTX (1 mM) induced a membrane depolarization associated with an increase in membrane conductance only in
the cell recorded with KCl. Downward deflections are electrotonic potentials evoked by injection of steady current pulses of the
same amplitude (250-ms duration).
2100
BOLEA, AVIGNONE, BERRETTA, SANCHEZ-ANDRES, AND CHERUBINI
magnesium block and to activate NMDA receptor channels
present on interneurons. Further evidence in favor of the involvement of AMPA receptor in GDP induction is given by the
experiment with GYKI 53655, a more-selective AMPA receptor antagonist (Lerma et al. 1997). In these cases, the impossibility to re-activate the network with GABA or AMPA suggest that network synchronization may involve AMPA
receptors localized on GABAergic interneurons, having, in
comparison with principal cells, distinct biophysical and pharmacological properties. Striking differences in the functional
characteristics of AMPA receptors in basket and principal cells
were reported (Isa et al. 1996; Koh et al. 1995; for review see
Freund and Buzsaki 1996). These consist of larger singlechannel conductance, rapid desensitization kinetics (see also
Livsey et al. 1993), and in the majority of the cases a higher
permeability to calcium (Burnashev et al. 1995; Koh et al.
1995). This latter issue appears to be particularly interesting in
view of the role that this divalent cation has in synaptic
plasticity processes during development. Thus the rapid transient increase in calcium flux through AMPA receptor channel
at rest would temporally summate to that entering the cell
through GABA-mediated activation of voltage-dependent cal-
FIG. 6. GDPs component revealed after intracellular blockade of GABAA-mediated responses with fluoride is glutamatergic. A: pairs
of recordings from adjacent pyramidal cells (at
P5) patched with intrapipette solutions containing either KCl or KF. GDPs are synchronous in the 2 recordings. However, GDPs recorded with KF were smaller in amplitude and
exhibited slower kinetics. Two GDPs (marked
with an *) are superimposed and shown on the
right on an expanded time scale. B: bath application of the broad spectrum NMDA and
non-NMDA antagonist kynurenic acid abolished GDPs that could be reinduced by bath
application of GABA (bar) only on the cell
patched with KCl.
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 15, 2017
FIG. 5. A NMDA and non-NMDA ionotropic glutamatergic component can be revealed in the evoked GDPs after block of
GABAA-mediated responses with intracellular
fluoride. A: GDPs evoked after 30 min breaking into whole cell configuration, in control
condition, and in the presence of D-AP5 (50
mM) and CNQX (20 mM). Membrane potential 280 mV. B: plot of GDPs area vs. membrane potential in control conditions () and in
the presence of D-AP5 (ƒ). C: GDPs evoked in
another neuron in the same experimental conditions as in A. Note the presence of a D-AP5–
sensitive component. Arrows indicate the time
of stimulation; membrane potential ranging
from 277 to 265 mV. D: responses to increasing concentrations of AMPA recorded in
the presence of TTX (1 mM) with KCl (E) or
KF (●). Each symbol is the mean of 3–7 responses obtained at P3-P6. Bars represent the
SE (when not visible they are within the symbols).
GDPS AND GLUTAMATE RECEPTORS
cium channels. As suggested by recombinant studies (Burnashev et al. 1992), different subunit composition may underline
differences in function. Interestingly, in a recent study on
hippocampal slices obtained from 16- to 20-day-old rats, it has
been shown that only interneurons activated by the mossy
fibers bear polyamine toxin (philanthotoxin)-sensitive AMPA
receptors, presumably calcium permeable (Tóth and McBain
1998). Further support to the idea that glutamate triggers the
release of GABA from GABAergic interneurons through
AMPA receptor subtypes is given by the experiments in which
during superfusion of CNQX, AMPA-evoked membrane depolarization and associated increase in frequency of GDPs are
blocked by bicuculline. Moreover the double-recording experiments (with KCl or KF), in which in the presence of kynurenic
acid it was not possible to induce GDPs in the cell recorded
with KF, strongly indicate that AMPA receptors on pyramidal
cells are blocked. The involvement of kainate receptors in GDP
generation can be excluded on the basis of the impossibility to
synchronize the network when kainic or domoic acid are applied in the presence of GYKI 53655.
AMPA receptor-mediated component of GDPs could be
revealed after intracellular blockage of GABAA receptor
with fluoride
After dialyzing principal cells with F2 (see also Khazipov et
al. 1997) in concomitance with a complete block of GABAAinduced responses, a residual component with slower kinetics
can be revealed. This component appears to be glutamatergic
in origin, as suggested by the shift in the reversal potential
toward more positive values and by the change in the slope of
the amplitude–voltage plot, obtained after dialyzing the cell
with F2 for $30 – 40 min. The observed shift in reversal was
similar to that already reported for GDPs in interneurons
(Khazipov et al. 1997) and was associated, in the majority of
the cases, with a linear amplitude–voltage relationship. This
indicates the activation of a non-NMDA receptor type, unmasked after intracellular blockade of GABA. Further support
in favor of activation of AMPA receptor type in evoked GDPs
is given by pharmacological experiments in which, in 8 out of
11 cases, CNQX was able to completely block the residual
component. In this respect our data differ from those reported
earlier by Khazipov et al. (1997) and Leinekugel et al. (1997).
However, in those reports, the assumption that the residual part
of GDPs (which remains after intracellular dialysis with F2) is
NMDA dependent was based only on the characteristic voltage
dependency of the responses. No pharmacological tools were
used to dissect out the different glutamatergic components. On
the other hand, the lack of effect of CNQX in blocking spikes
recorded in cell-attached configuration and supposed to represent interneuronal GDPs reported in those experiments (Khazipov et al. 1997; Leinekugel et al. 1997) has, as pointed out by
the same authors, strong limitations. In the current experiments, the similarity in the dose-response curves for AMPA,
obtained with intracellular solutions containing KCl and F2,
respectively, allows to exclude the possibility that the residual
AMPA-mediated component of GDPs is modified as the result
of the interference of F2 with some phosphorylation processes
(Andrews and Babior 1984).
Source of glutamatergic drive needed to synchronize
GABAergic interneurons
As already shown by Khazipov et al. (1997), Garaschuk et
al. (1998), and Menendez de la Prida et al. (1998), GDPs were
still present in small CA3 islands isolated from the rest of the
hippocampus, implying that a local circuit comprehensive of a
relatively small population of principal cells and interneurons
was sufficient to generate them. In these cases, the glutamatergic input presumably originates from collaterals of principal
cells because the mossy fibers are disconnected from the granule cells of the dentate gyrus. Also the glutamatergic component revealed in pyramidal cells recorded with KF is presumably due to glutamate release from recurrent collaterals. In
summary, these data show that GDP generation requires activation of AMPA receptors by local release of glutamate from
recurrent collaterals. Other glutamatergic components can also
be involved but do not seem to be essential.
This work was partially supported by a grant from Consiglio Nazionale delle
Ricerche, Ministero dell’Universita’ e Ricerca Scientifica e Tecnologica to E.
Cherubini and by a grant from the Fondo de Investigacion Sanitaria to J. V.
Sanchez-Andres. S. Bolea was supported by a fellowship from the Ministerio
de Educacion y Cultura. N. Berretta was supported by a fellowship from
Novartis Pharma. S. Bolea and E. Avignone contributed equally to this work.
Present address of N. Berretta: IRCCS, S. Lucia, Via Ardeatina 306, 00178
Roma, Italy.
Address for reprint requests: E. Cherubini, International School for Advanced Studies, Via Beirut 2– 4, 34014 Trieste, Italy.
Received 9 October 1998; accepted in final form 19 January 1999.
REFERENCES
ACSÁDY, L., KAMONDI, A., Sı́K, A., FREUND, T., AND BUZSÁKI, G. GABAergic
cells are the major postsynaptic targets of mossy fibers in the rat hippocampus. J. Neurosci. 18: 3386 –3403, 1998.
ANDREWS, P. C. AND BABIOR, B. M. Phosphorylation of cytosolic proteins by
resting and activated human neutrophils. Blood 64: 883– 890, 1984.
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 15, 2017
FIG. 7. GDPs are still present in minislices of the CA3 region. A: schematic
diagram showing a minislice of the CA3 region disconnected with a knife cut
from the rest of the hippocampus at P5. B: pairs of recordings from adjacent
pyramidal cells (patched with KCl and KF intrapipette solutions) are still
synchronous. In this case, the glutamatergic component of the GDP detected
with KF can originate only from a local circuit.
2101
2102
BOLEA, AVIGNONE, BERRETTA, SANCHEZ-ANDRES, AND CHERUBINI
LEINEKUGEL, X., TSEEB, V., BEN ARI, Y., AND BREGESTOVSKI, P. Synaptic
GABAA activation induces Ca11 rise in pyramidal cells and interneurones
from rat neonatal hippocampal slice. J. Physiol. (Lond.) 487: 319 –329,
1995.
LERMA, J., MORALES, M., AND VICENTE, M. A. Glutamate receptors of the
kainate type and synaptic transmission. Trends Neurosci. 20: 9 –12, 1997.
LIVSEY, C. T., COSTA, E., AND VICINI, S. Glutamate-activated currents in
outside-out patches from spiny versus aspiny hilar neurons of rat hippocampal slices. J. Neurosci. 13: 5324 –5333, 1993.
MCLEAN, H. A., ROVIRA, C., BEN ARI, Y., AND GAIARSA, J. L. NMDAdependent GABAA-mediated polysynaptic potentials in the neonatal rat
hippocampal CA3 region. Eur. J. Neurosci. 7: 1442–1448, 1995.
MENENDEZ DE LA PRIDA, L., BOLEA, S., AND SANCHEZ-ANDRES, J. V. Origin of
the synchronized network activity in the rabbit developing hippocampus.
Eur. J. Neurosci. 10: 899 –906, 1998.
OWENS, D. F., BOYCE, L. H., DAVIS, M.B.E., AND KRIEGSTEIN, A. R. Excitatory
GABA responses in embryonic and neonatal cortical slices demonstrated by
gramicidin perforated-patch recordings and calcium imaging. J. Neurosci.
16: 6414 – 6423, 1996.
PARTIN, K. M., PATNEAU, D. K., WINTERS, C. A., MAYER, M. L., AND BUONANNO, A. Selective modulation of desensitization at AMPA versus kainate
receptors by cyclothiazide and concanavalin A. Neuron 11: 1069 –1082,
1993.
ROHRBOUGH, J. AND SPITZER, N. C. Regulation of intracellular Cl2 levels by
Na1-dependent Cl2 cotransport distinguishes depolarizing from hyperpoalrizing GABAA receptor-mediated responses in spinal neurons. J. Neurosci.
16: 82–91, 1996.
SERAFINI, R., VALEYEV, A. Y., BARKER, J. L., AND POULTER, M. O. Depolarizing GABA-activated Cl2 channels in embryonic rat spinal and olfactory
bulb cells. J. Physiol. (Lond.) 488: 371–386, 1995.
SHATZ, C. J. Impulse activity and the patterning of connections during CNS
development. Neuron 5: 745–756, 1990.
STRATA, F., ATZORI, M., MOLNAR, M., UGOLINI, G., TEMPIA, F., AND CHERUBINI, E. A pacemaker current in dye-coupled hilar interneurons contributes
to the generation of giant GABAergic potentials in developing hippocampus. J. Neurosci. 17: 1435–1446, 1997.
STRATA, F., SCIANCALEPORE, M., AND CHERUBINI, E. cAMP-dependent modulation of Giant Depolarizing Potentials by metabotropic glutamate receptors
in the rat hippocampus. J. Physiol. (Lond.) 489: 115–125, 1995.
TÓTH, K. AND MCBAIN, C. J. Afferent-specific innervation of two distinct
AMPA receptor subtypes on single hippocampal interneurons. Nature Neurosci. 7: 572–578, 1998.
WU, W., ZISKIND-CONHAIM, L., AND SWEET, M. A. Early development of
glycine- and GABA-mediated synapses in rat spinal cord. J. Neurosci. 12:
3935–3945, 1992.
XIE, X., HIDER, R., AND SMART, T. G. Modulation of GABA-mediated synaptic
transmission by endogenous zinc in the immature hippocampus in vitro.
J. Physiol. (Lond.) 478: 75– 86, 1994.
YAKEL, J. L. Calcineurin regulation of synaptic function: from ion channels to
transmitter release and gene transcription. Trends Pharmacol. Sci. 18: 124 –
134, 1997.
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 15, 2017
BEN ARI, Y., CHERUBINI, E., CORRADETTI, R., AND GAIARSA, J. L. Giant
synaptic potentials in immature rat CA3 hippocampal neurones. J. Physiol.
(Lond.) 416: 303–325, 1989.
BEN ARI, Y., KHAZIPOV, R., LEINEKUGEL, X., CAILLARD, O., AND GAIARSA, J. L.
GABAA, NMDA and AMPA receptors: a developmentally regulated ‘ménage à trois’. Trends Neurosci. 20: 523–529, 1997.
BOLEA, S., AVIGNONE, E., BERRETTA, N., SANCHEZ-ANDRES, J. V., AND CHERUBINI, E. Non-NMDA receptor activation is essential for generating giant
depolarising potentials (GDPs) in neonatal rat hippocampal slices.
J. Physiol. (Lond.) 509: 69, 1998.
BORMANN, J., HAMILL, O. P., AND SAKMANN, B. Mechanism of anion permeation through channels gated by glycine and g-aminobutyric acid in mouse
cultured spinal neurones. J. Physiol. (Lond.) 385: 243–286, 1987.
BURNASHEV, N., MONYER, H., SEEBURG, P. H., AND SAKMANN, B. Divalent ion
permeability of AMPA receptor channels is dominated by the edited form of
a single subunit. Neuron 8: 189 –198, 1992.
BURNASHEV, N., ZHOU, Z., NEHER, E., AND SAKMANN, B. Fractional calcium
currents through recombinant GluR channels of the NMDA, AMPA and
kainate recptor subtypes. J. Physiol. (Lond.) 485: 403– 418, 1995.
CHEN, G., TROMBLEY, P., AND VAN DEN POL, A. N. Excitatory actions of GABA
in developing hypothalamic neurons. J. Physiol. (Lond.) 494: 451– 464,
1996.
CHERUBINI, E., GAIARSA, J. L., AND BEN ARI, Y. GABA: an excitatory transmitter in early postnatal life. Trends Neurosci. 14: 515–519, 1991.
DURAND, G. M., KOVALCHUK, Y., AND KONNERTH, A. Long-term potentiation
and functional synapse induction in developing hippocampus. Nature 381:
71–75, 1996.
FREUND, T. F. AND BUZSAKI, G. Interneurons of the hippocampus. Hippocampus 6: 347– 470, 1996.
GAARSKJAER, F. B. The organization and development of the hippocampal
mossy fiber system. Brain Res. Rev. 11: 335–357, 1986.
GAIARSA, J. L., CORRADETTI, R., CHERUBINI, E., AND BEN ARI, Y. Modulation
of GABA-mediated synaptic potentials by glutamatergic agonists in neonatal CA3 rat hippocampal neurons. Eur. J. Neurosci. 3: 301–309, 1991.
GARASCHUK, O., HANSE, E., AND KONNERTH, A. Developmental profile and
synaptic origin of early network oscillations in the CA1 region of rat
neonatal hippocampus. J. Physiol. (Lond.) 507: 219 –236, 1998.
ISA, T., ITAZAWA, S., IINO, M., TSUZUKI, K., AND OZAWA, S. Distribution of
neurones expressing inwardly rectifying and Ca21-permeable AMPA receptors in rat hippocampal slices. J. Physiol. (Lond.) 491: 719 –733, 1996.
KANEDA, M., FARRANT, M., AND CULL-CANDY, S. G. Whole-cell and single
channel currents activated by GABA and glycine in granule cells of the rat
cerebellum. J. Physiol. (Lond.) 485: 419 – 435, 1996.
KHAZIPOV, R., LEINEKUGEL, X., KHALILOV, I., GAIARSA, J. L., AND BEN ARI, Y.
Synchronization of GABAergic interneuronal network in CA3 subfield of
neonatal rat hippocampal slices. J. Physiol. (Lond.) 498: 763–772, 1997.
KOH, D. S., GEIGER, J.R.P., JONAS, P., AND SAKMANN, B. Ca21-permeable
AMPA and NMDA receptor channels in basket cells of rat hippocampal
dentate gyrus. J. Physiol. (Lond.) 485: 383– 402, 1995.
LEINEKUGEL, X., MEDINA, I., KHALILOV, I., BEN ARI, Y., AND KHAZIPOV, R. Ca21
oscillations mediated by the synergistic excitatory actions of GABAA and
NMDA receptors in the neonatal hippocampus. Neuron 18: 243–255, 1997.