Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2011 Jan 10;589(Pt 5):1159–1172. doi: 10.1113/jphysiol.2010.203224

Astrocytes control GABAergic inhibition of neurons in the mouse barrel cortex

B Benedetti 1, V Matyash 1, H Kettenmann 1
PMCID: PMC3060594  PMID: 21224221

Abstract

Astrocytes in the barrel cortex respond with a transient Ca2+ increase to neuronal stimulation and this response is restricted to the stimulated barrel field. In the present study we suppressed the astrocyte response by dialysing these cells with the Ca2+ chelator BAPTA. Electrical stimulation triggered a depolarization in stellate or pyramidal ‘regular spiking’ neurons from cortex layer 4 and 2/3 and this response was augmented in amplitude and duration after astrocytes were dialysed with BAPTA. Combined blockade of GABAA and GABAB receptors mimicked the effect of BAPTA dialysis, while glutamate receptor blockers had no effect. Moreover, the frequency of spontaneous postsynaptic currents was increased after BAPTA dialysis. Outside the range of BAPTA dialysis astrocytes responded with a Ca2+ increase, but in contrast to control, the response was no longer restricted to one barrel field. Our findings indicate that astrocytes control neuronal inhibition in the barrel cortex.

Non-technical summary

In the last few years astrocytes have come to be recognized as active elements contributing to brain activity. They sense and modulate neuronal activity, and often changes in their cytosolic Ca2+ are essential for neuron–glia interactions. In the present study we interfered with intracellular calcium signalling in astrocytes in mouse somatosensory cortex by dialysis with the calcium chelator BAPTA. Such treatment increased excitability of the nearby neurons. The effect of astrocytic calcium chelation was mimicked by pharmacological inhibition of GABA receptors, suggesting that this control is GABA mediated through a combined involvement of GABAA and GABAB receptors. This finding demonstrates a role for astrocytes in the regulation of neural inhibition in somatosensory (barrel) cortex and adds a new variant to the growing number of pathways by which astrocytes can modulate neuronal networks.

Introduction

It is well established that astrocytes respond to neuronal activity with a cytosolic Ca2+ increase. Several studies indicate that such astrocyte responses can modulate neuronal activity (for review see Halassa et al. 2007; Perea et al. 2009; Nedergaard et al. 2010), although this view was recently challenged by contrasting results on mouse lines in which astrocyte Ca2+ signalling was affected (Fiacco et al. 2007; Petravicz et al. 2008; Agulhon et al. 2010). Astrocyte activity can influence at least some forms of synaptic plasticity, such as long term potentiation (Serrano et al. 2006; Henneberger et al. 2010), short term depression (Andersson & Hanse, 2010) and number of pathological processes such as pain perception gating (Bardoni et al. 2010), hypoxia-induced network depression (Martin et al. 2007) and epilepsy (Kumaria et al. 2008).

In vivo recordings from astrocytes demonstrated that mechanical stimulation of a whisker evokes Ca2+ responses in astrocytes in the barrel cortex (Wang et al. 2006). In slices of barrel cortex, extracellular electrical stimulation evokes astrocyte calcium responses (Schipke et al. 2008) restricted to the stimulated sensory column (barrel field). A rapid neuronal Ca2+ response was followed by a slower and longer-lasting Ca2+ response in astrocytes, which depended on neuronal activity as it was completely blocked by tetrodotoxin. In the present study we used the stimulation protocol which induced the astrocytic Ca2+ responses and studied evoked neuronal activity with the patch-clamp technique. We then interfered with the astrocyte Ca2+ response and determined the impact on neuronal activity. Our data indicate that astrocytes control inhibition of barrel cortex neurons. This neuron–glia interaction probably involves both GABAA and GABAB receptors, but not glutamatergic transmission.

Methods

Ethical approval

Transgenic animals were bred in the local animal facilities and handled for experiments according to the guidelines of the European Parliament Directive for the protection of animals used for scientific purposes and the Federal Ministry of Berlin ‘Landesamt für Gesundheit und Sozialen’ (LaGeSo).

Animal preparation and Ca2+ imaging

Acute brain slices were prepared from 8- to 10-day-old NMRI mice (Charles River, Germany) or from the transgenic mouse lines eGFP/GFAP (Nolte et al. 2001) and mRFP1/GFAP (Hirrlinger et al. 2005) (kindly provided by Prof. F. Kirchhoff). Transgenic animals expressed enhanced green fluorescent protein (eGFP) or red fluorescent protein (mRFP1) under the control of the human glial fibrillary acidic protein (GFAP) promoter. After animal decapitation the brain was removed and slices of 250 μm thickness were prepared following the protocol of Agmon & Connors (1991). Before recording, slices were incubated for at least 30–45 min in artificial cerebrospinal fluid (ACSF) at room temperature. For Ca2+ imaging slices were loaded with Fluo-4-AM as described in Peters et al. (2003). Imaging and patch clamp experiments were performed with an upright microscope (Zeiss, Germany). During the experiments slices were kept in a perfusion chamber at 32–34°C with constant ACSF flow of about 5 ml min−1. Barrel fields were identified in bright field illumination (Schipke et al. 2008). Local barrel stimulation was applied through a glass electrode (tip opening about 20 μm) placed in layer 4 of the cortex within a given barrel field. The stimulus consisted of 30 voltage pulses at 4 V, 30 μA, duration of a single stimulus 1 ms, 100 Hz stimulation frequency (Schipke et al. 2008). The stimulation protocol as well as the patch-clamp recording was performed with a patch-clamp amplifier (EPC9 or EPC9/2, HEKA Elektronik, Lambrecht, Germany) and traces were acquired with a 3.0 kHz Bessel filter. For stimulation the amplifier voltage output was connected to an external stimulus isolator (NeuroLog NL 800, Digitimer Ltd, Welwyn Garden City, UK). Images for Ca2+ measurements were acquired at 1 Hz, 300 ms exposure time. Patch clamp acquisitions and imaging experiments were performed with the software TIDA (HEKA Elektronik) and Imaging Cells Easily (ICE; our own development) software or with Camware (PCO Imaging, Kelheim, Germany).

Immunohistochemistry and morphological characterization

After biocytin dialysis of single neurons, slices were fixed overnight in phosphate buffer solution of 0.1 m with 0.4% paraformaldehyde. Biocytin-filled neurons were then marked with Cy3 fluorophore conjugated to streptavidin. After fixation slices were incubated in a solution containing 0.2% Triton X-100, 2% BSA and 10% normal goat serum (NGS) in phosphate buffer at pH 7.4 for 4 h to permeabilize and block non-specific binding. Cy3 conjugated streptavidin (1:200, Covance/HISS Diagnostic GmbH, Freiburg, Germany) was diluted in 0.1 m phosphate buffer containing 0.2% Triton X-100, 2% BSA and 5% NGS. Image stacks were acquired with a confocal microscope (SPE, Leica Microsystems, Germany) and maximal intensity projections were reconstructed with ImageJ software (National Institutes of Health, Bethesda, MD, USA, http://rsb.info.nih.gov/ij/, 1997–2009).

Solutions and drugs

Standard ACSF contained (in mm): NaCl (134), KCl (2.5), CaCl2 (2), MgCl2 (1.3), K2PO4 (1.25), glucose (9) and NaHCO3 (26) (for storing at room temperature prior to experiment) or (21.4) (in the recording chamber at 32–34°C). Pipette solutions for neurons contained (in mm): potassium gluconate (120), KCl (10), MgCl2 (1), EGTA (0.1), CaCl2 (0.025), Hepes (10), ATPK2 (1), GTPNa (0.2), glucose (4) (Kang et al. 1998). In the pipette solution with high [Cl−] potassium gluconate was replaced with KCl (pH 7.2). The pipette solution for astrocytes contained (in mm): potassium gluconate (90), BAPTA (40), MgCl2 (1), NaCl (8), ATP (2), GTP (0.4), Hepes (10) (pH 7.2) and in the control dialysis experiments BAPTA was replaced by potassium gluconate, (Serrano et al. 2006). Sulforhodamine (Sigma) was added at a concentration of 0.1 mg ml−1, Fluo 4-AM (Invitrogen GmbH, Darmstadt, Germany) at 10 μm and membrane impermeant Fluo 4 (Invitrogen) was added at 20 μm to the pipette solution. Drugs were bath applied at the following concentrations (in μm): LY367385 (200) (Tocris, Ellisville, MO, USA), 2-Methyl-6-(phenylethynyl) pyridine hydrochloride (MPEP) (100) (Tocris), gabazine (10) (Tocris), CGP55845 (5) (Tocris), 8-Cyclopentyl-1,3-dimethylxanthine (CPT) (4) (Sigma), strychnine (5) (Sigma). Drugs were pre-incubated for 1–2 min before stimulation. d-Amino acid oxidase (d-AAO) was used at the concentration of 17 U ml−1. For d-AAO experiments the brain slices were pre-incubated (30–90 min) at room temperature in oxygenated ACSF in which the enzyme was dissolved. The final concentration of DMSO (Sigma) in which some drugs were dissolved was tested alone and proved to have no effect on the neuronal evoked response.

Data analysis

For patch-clamp data analysis and statistics we used TIDA (HEKA Elektronik, Lambrecht, Germany), Origin (OriginLab Corp., Northampton, MA, USA), Microsoft Excel and Image-Pro PLUS 5.1 (Media Cybernetics, Inc., Bethesda, MD, USA). Data are shown as means (±s.e.m.); significance was determined with the appropriate Student's t test or with one-way ANOVA. Action potential half-width and frequency were analysed with a custom made function of the software Igor Pro (WaveMetrics, Inc., Lake Oswego, OR, USA) written and kindly provided by Dr K. van Aerde and Prof. D. Feldmeyer, (Forschungzentrum, Juelich, Germany); the action potential threshold was determined according to Kole & Stuart (2008). The stimulus-evoked depolarization in neurons was measured at time points when no action potential occurred. The integral of stimulus-evoked depolarization over time (ISTM) was determined from the baseline at the resting voltage and included also the time course of action potentials. Fluo-4 fluorescence recordings were normalized (F/F0) and filtered using a median filter. F0 was obtained by averaging 10 frames at the beginning of the recording. To determine the area occupied by Ca2+ responsive astrocytes, we compared images prior to, between and 1–3 s after stimulation. The area was composed of pixels which increased in brightness above threshold. Lateral drift of the imaged field induced by movements of the sample during the experiment was manually corrected. Ca2+ kinetics were analysed with Image Pro (Media Cybernetics).

Ca2+ chelation in astrocytes

To identify astrocytes before dialysis we used the mRFP1/GFAP mouse line with red fluorescent labelled astrocytes. The eGFP/GFAP (green) mouse line was used for dye coupling experiments where sulforhodamine (red) was added to the patch pipette solution. In all experiments we only dialysed positively identified astrocytes within layer 4 of the stimulated barrel (within 100 μm from the stimulation electrode). After establishing the whole cell recording configuration, astrocytes displayed low input resistance and linear voltage–current relationship as previously described in cortex (Schipke et al. 2008). We dialysed cells with membrane-impermeant BAPTA (40 mm) in the whole-cell configuration for 45–60 min (Serrano et al. 2006; Andersson & Hanse, 2010). In dye coupling experiments we observed that after 45–60 min dialysis, sulforhodamine spread to 12 ± 2 cells (Supplementary Fig. 1a; n= 6). In some experiments we also used biocytin in the dialysis pipette and observed a larger syncytium, similarly to the report of Serrano et al. (2006) (data not shown). In Ca2+ imaging experiments the local stimulation evoked a Ca2+ response in astrocytes as previously described (Schipke et al. 2008). In the area close to the dialysed astrocyte within the given barrel field, BAPTA dialysis resulted in a reduction in the population of astrocytes responding to stimulation. To quantify the responding cell population, we measured within a 4000 μm2 area with the injected astrocyte approximately in the centre where Fluo-4 fluorescence increased (i.e. the area occupied by responding astrocytes). This area decreased from 1600 ± 350 μm2 in control conditions to 500 ± 200 μm2 after BAPTA dialysis (n= 6). To exclude that the decreased Ca2+ response in the astrocytes was due to a dilution of the dye due to dialysis we also added membrane-impermeant Fluo-4 to our pipette solution. We found that the stimulation evoked response decreased to 18.3 ± 2.4% of control (n= 12). Notably even after BAPTA dialysis there were still few astrocytes close to the dialysed one, in which the Ca2+ response was unaltered (Supplementary Fig. 1b, sample 3). This further supports the view that the decrease in astrocytic Ca2+ signal was not due to bleaching of the samples or dilution of the Ca2+ sensor through the patch pipette.

Results

Astrocytic Ca2+ chelation leads to an enhanced stimulation-induced depolarization in neighbouring neurons

To study the influence of astrocytes on neuronal activity, we attenuated astrocyte Ca2+ signalling by BAPTA dialysis of the astrocyte network (see Methods). We used a stimulation protocol delivered via an extracellular electrode, which reliably triggers a Ca2+ increase in astrocytes under control conditions (Schipke et al. 2008). In neurons from layer 4 and 2/3, this stimulation protocol evoked a sustained depolarization by 29.8 ± 1.0 mV from an average resting potential of −68.7 ± 0.3 mV (n= 92; Fig. 1). After the stimulation, the membrane repolarized with a half-repolarization time of 0.61 ± 0.03 s. Action potentials were superimposed to this slow depolarization during stimulation, but were rarely observed during the repolarization phase (<5% of cells, data not shown). When we did a similar recording after astrocyte dialysis with BAPTA we observed a larger and longer depolarization in response to stimulation (n= 18; Fig. 1). The average depolarization was increased (P= 0.002) to 40.7 ± 2.9 mV and the half-repolarization was significantly prolonged (P= 0.0003) due to an additional plateau phase, namely to 2.50 ± 0.43 s. Moreover in 9 out of 18 of these neurons we observed 1–16 (average 4.2 ± 1.4) action potentials during the repolarization phase (Fig. 1). To further quantify the neuronal depolarization, we compared the integral of membrane depolarization over time (ISTM). In control conditions ISTM was 27 ± 2 mV s, while after astrocytic BAPTA dialysis ISTM was significantly increased (P= 0.002) to 118 ± 25 mV s (Fig. 1).

Figure 1. Astrocytic Ca2+ chelation causes an enhanced evoked response in neurons.

Figure 1

Open in a new tab

A, membrane potential (Em) recording from a neuron within the stimulated barrel. Stimulus-evoked neuronal response before (Control, left) and after dialysis of an astrocyte with BAPTA (right). The stimulation is indicated by the bar. B, mean stimulus (STM) evoked depolarization amplitude (left, STM depol), time for half-repolarization (middle, ‘half-rep’) and integral of the voltage trace (ISTM) during the evoked depolarization (right). *P= 0.002, **P < 0.001 (unpaired t test).

Bath application of tetrodotoxin (TTX, n= 6) strongly reduced the evoked depolarization to 4.3 ± 1.4 mV (17% of internal control; P= 0.0003) and ISTM to 4 ± 1 mV s (24% of the internal control, P= 8 × 10−7). No action potentials were observed in the presence of tetrodotoxin and after stimulation the membrane potential rapidly returned to the resting value (data not shown). As an additional control, astrocytes were dialysed omitting BAPTA in the pipette solution (substituted by potassium gluconate). After ∼60 min dialysis, the neuronal response to stimulation was not significantly different from control (Supplementary Fig. 2; n= 12). The evoked depolarization was 24 ± 1 mV (P= 0.07) the time of half-repolarization was 0.67 ± 0.03 s (P= 0.53) and ISTM was 21 ± 2 mV s (P= 0.27).

Selected neurons were spiny stellate and pyramidal cells

To characterize the neuronal population we studied spiking properties and morphology. We delivered depolarizing current steps of increasing intensity (step size: 10 pA) and determined the action potential frequency. The maximal action potential frequency was typically observed at about −20 mV (100 pA current injection) and amounted on average to 105 ± 8 Hz for L2/3 (n= 17) and 96 ± 10 Hz for L4 neurons (n= 15) (Fig. 2A). No significant difference in the peak frequencies was observed when comparing neurons in L4 or L2/3 (P= 0.4). The action potential half-width was 2.9 ± 0.2 ms in neurons from layer 2/3 and 2.1 ± 0.2 ms in neurons from L4. We also typically observed firing frequency adaptation as previously described (Agmon & Connors, 1992). These characteristics are thus typical for ‘regular spiking cells’ such as spiny stellate and pyramidal cells (Agmon & Connors, 1992). To reveal the morphology, biocytin was injected. In layer 4, most cells (n= 7 of 8) had the typical morphology of spiny stellate neurons (Fig. 2Ba), namely star-like branching dendrites, a round cell soma with occasional asymmetry toward the centre of the barrel and long columnar axonal projections often extending to the white matter. In layer 2/3 most neurons (n= 12 of 13) had the typical morphology of pyramidal neurons (Fig. 2Bb), namely thick apical dendrites toward the pial surface, a triangular shaped soma and long columnar axonal projection toward the white matter (White & Rock, 1980; Lubke et al. 2000). Two neurons in L4 and L2/3 could not be unequivocally indentified based on their morphology, but their firing pattern was ‘regular spiking’ (data not shown). Notably we also found a remarkable amount of coupling in spiny stellate cells in L4 (5 out of 8 experiments, Fig. 2Ca) and in pyramidal cells in L2/3 (7 out of 13 experiments Fig. 2Cb). We conclude that the morphological and physiological criteria classify the cells as excitatory neurons.

Figure 2. Selected neurons had physiological and morphological features of spiny stellate and pyramidal cells.

Figure 2

Open in a new tab

A, cells were depolarized for 500 ms with increasing current steps. Three sample traces of membrane potential recording are shown. On the right a plot describes the relation between injected current amplitude and action potential frequency. Dots in grey were obtained from the three sample recordings. B, L4 (a) and L2/3 (b) neurons dialysed with biocytin added to pipette solution. Scale bar denotes 150 μm. C, same cells as in B at higher magnification (scale bar 20 μm). Note the coupled cells in L4 (a) and L2/3 (b); the patched cell is marked by an asterisk.

Inhibiting GABAA and GABAB receptors increased the stimulation-induced depolarization of neurons

GABA is the principal inhibitory neurotransmitter in the barrel cortex (Swadlow, 2002). We studied the influence of the GABAA and GABAB receptor antagonists, gabazine and CGP55845, respectively, on the stimulation-evoked neuronal response. Gabazine (Fig. 3A) increased the stimulus-evoked depolarization (n= 9) from 26.3 ± 1.9 mV to 47.5 ± 3.2 mV (186 ± 16%; P= 6 × 10−5), while the time for half-repolarization (85 ± 17%) did not change significantly (P= 0.06). ISTM increased from 27 ± 4 mV s to 43 ± 7 mV s (162 ± 2%, P= 0.01). During GABAA receptor blockage but not in control we also recorded in 7 out of 9 neurons a variable number (1 to 7) of action potentials during the repolarization phase. CGP55845 (Fig. 3B) increased the stimulus-evoked depolarization (n= 10) from 28.4 ± 3.4 mV to 32.5 ± 3.4 mV (117 ± 6%, P= 0.003) and the half-repolarization time from 0.50 ± 0.05 s to 0.79 ± 0.15 s (154 ± 17%, P= 0.02). ISTM increased from 27 ± 7 mV s to 40 ± 7 mV s (160 ± 18%, P= 0.009). We did not observe action potentials during the repolarization phase.

Figure 3. Inhibiting GABAA and GABAB receptors increased the stimulation-induced depolarization of neurons.

Figure 3

Open in a new tab

A, membrane potential (Em) recording during stimulation (indicated by bar) in control (left trace) and in the presence of gabazine (middle trace). The graph on the right gives the mean stimulus induced depolarization (white column), half-repolarization time (grey column) and ISTM (black column) from Control (right trace) normalized to the control response as 100% (drug/Control). B and C, similarly CGP55845 (B) and a combination of gabazine and CGP55845 (C) were tested. *P < 0.05, **P < 0.01, ***P < 0.001 (paired t test of related parametric values).

Combined application of GABAA (gabazine) and GABAB receptor (CGP55845) antagonists (Fig. 3C) increased the stimulus-evoked depolarization (n= 7) from 23.4 ± 2.6 mV to 46.2 ± 8.4 mV (205 ± 21%, P= 0.0002), half-repolarization time increased from 0.51 ± 0.05 s to 0.64 ± 0.09 s (127 ± 12%P= 0.03) and ISTM from 23 ± 2 mV s to 61 ± 8 mV s (267 ± 36%P= 0.0006). During the repolarization phase we recorded action potentials (9 ± 1.6) in all neurons (Fig. 3C). We conclude that combined blockade of GABAA and GABAB receptors mimicked the changes in the neuronal response as after BAPTA dialysis of astrocytes.

Ionotropic glutamate receptor blockers have only a marginal effect on the neuronal evoked response

Glutamate is the principal neurotransmitter mediating the excitatory neuronal transmission in cortex and we tested whether antagonists of AMPA/kainate and NMDA receptors, CNQX and MK801, affect the stimulus-induced depolarization (Supplementary Fig. 3). Application of CNQX (n= 7) did not affect the evoked depolarization (P= 0.07), half-repolarization time (P= 0.2) or ISTM (P= 0.1) in comparison to controls (Supplementary Fig. 3a). The NMDA receptor blockade with MK801 (n= 8) decreased the half-repolarization time from 0.64 ± 0.1 s to 0.51 ± 0.04 s (85 ± 6%, P= 0.03), but did not affect the stimulus evoked depolarization (P= 0.1) or ISTM (P= 0.07), (Supplementary Fig. 3b). Also the combined application of CNQX and MK801 (n= 5) failed to change the stimulus-evoked depolarization (P= 0.06) slightly decreased the half-repolarization time (P= 0.03) from 0.87 ± 0.13 s to 0.69 ± 0.06 s (83 ± 6%) and ISTM (P= 0.03) from 26 ± 7 mV s to 17 ± 4 mV s (88 ± 6%) (Supplementary Fig. 3c). These results indicate that the neuronal response to stimulation is not dominated by glutamatergic transmission.

Reversal potential and membrane conductance are similarly affected by GABA receptors blockade and astrocytic BAPTA dialysis

To characterize the stimulus-induced depolarization, we determined a current-to-voltage curve of the response in control (filled circles, n= 8), after astrocytic BAPTA dialysis (open circles, n= 8), during gabazine and CGP55845 application (filled triangles, n= 8) and after the neuron was dialysed with high chloride concentration (filled squares, High [Cl−]i= 132 mm) pipette solution (n= 7). We applied a series of voltage steps before and immediately after the stimulation ranging from −70 to +10 mV (20 mV increment, duration: 30 ms). We then determined the evoked current before and after stimulus (Fig. 4A) and the net stimulus-evoked current by subtraction (Fig. 4B‘after stimulus’–‘before stimulus’). The stimulus-evoked conductance (Fig. 4B, right) was measured from the current–voltage relation (Fig. 4B, left). The average of the stimulation-induced conductance between −10 mV and −50 mV was 15.4 ± 3.0 nS in control conditions (filled circles) and the reversal potential was close to −50 mV. After dialysis of astrocytes with BAPTA (open circles) the conductance was 6.5 ± 1.0 nS, and thus significantly decreased as compared to control (one-way ANOVA, P= 0.002). In this condition the reversal potential was close to −30 mV. Application of gabazine and CGP55845 (filled triangles) resulted in a conductance of 4.4 ± 2.0 nS with reversal potential of about 0 mV. The stimulus-evoked conductance in presence of Gabazine and CGP55845 was significantly smaller than controls (P= 0.002) but not different from the conductance after dialysis of astrocytes with BAPTA (P= 0.4). When neurons were dialysed with 132 mm intracellular chloride concentration (filled squares), their average conductance of 12.9 ± 2.3 nS (High [Cl−]i) was not significantly different from controls (P= 0.4), but the reversal potential was shifted toward positive values (about 0 mV). These results indicate that shortly after the stimulation the induced current under control conditions involves a GABA-dependent inhibitory conductance largely based on chloride and a smaller residual conductance with a reversal potential at 0 mV (see current–voltage relation in Fig. 4B, filled triangles).

Figure 4. Reversal potential and membrane conductance are similarly affected by GABA receptors blockage and astrocytic BAPTA dialysis.

Figure 4

Open in a new tab

A, currents in response to depolarizing voltage steps (−50 mV, −30 mV, −10 mV and +10 mV) from a holding potential of −70 mV were recorded in whole cell voltage clamp. Recordings were obtained before and after the stimulation with the external electrode in control conditions (upper left, •), after astrocytic BAPTA dialysis (upper right, ◯), after gabazine + CGP55845 bath application (lower left, ▴) and after dialysis of the neuron with high chloride concentration (high [Cl−]i) (lower right, ▪). Scale bar: 0.5 nA, 0.1 s. B, average current–voltage relation of the stimulus evoked conductance under the four conditions as described in A is shown at the top. Average conductance determined between −10 and −70 mV is displayed at the bottom for the four conditions described in A. **P= 0.002, one-way ANOVA.

Astrocytic Ca2+ chelation, GABA receptor blockage or dialysis with high [Cl−] reverses a stimulation induced hyperpolarization into depolarization

To elicit spontaneous action potential firing, we depolarized neurons by constant current injection. In response to our standard stimulation protocol the action potentials ceased for about 1 s and the membrane hyperpolarized by −2.2 ± 0.8 mV (n= 13). After astrocytic BAPTA dialysis (n= 10) we observed instead a depolarization by +14.0 ± 2.0 mV (P= 5.4 × 10−8, one-way ANOVA; Fig. 5A and B). Similarly in the presence of gabazine and CGP55845 (n= 5), the stimulation induced a depolarizing response of +17.6 ± 2.7 mV, which was different from control conditions (P= 1.1 × 10−7) but not significantly different from the depolarization after BAPTA dialysis (P= 0.3). Astrocyte dialysis with pipette solution in which BAPTA was omitted led as in controls (P= 0.8) to a transient hyperpolarization by −2.6 ± 1.2 mV (data not shown). After dialysis of a neuron with high [Cl−] (132 mm; see also Fig. 6) stimulation triggered a depolarization of +37.0 ± 3.8 mV before (High [Cl−]i) and of +41.0 ± 2.5 mV after astrocytic BAPTA dialysis (High [Cl−]i+ BAPTA). These two values were also not significantly different (P= 0.4). Furthermore in cells dialysed with high [Cl−] the action potential frequency was transiently decreased for about 2 s after the stimulation from 8.3 ± 1.3 Hz to 5.6 ± 1.8 Hz (n= 13). After BAPTA dialysis of neighbouring astrocytes this decrease was no longer observed: the frequency before stimulation was 10.7 ± 1.5 Hz, in the 2 s after stimulation 10.9 ± 2.1 Hz (n= 9; P= 0.4; Fig. 5, Supplementary Fig. 4).

Figure 5. Astrocytic Ca2+ chelation, GABA receptor blockage or dialysis with high [Cl−] reverses a stimulation induced hyperpolarization into depolarization.

Figure 5

Open in a new tab

A, neurons were depolarized by current injection to about −40 mV while recording membrane potential (Em). The stimulation protocol was applied as indicated by bar. In the upper three traces cells were dialysed with low [Cl−], in the bottom two with high [Cl−] in the pipette. The control response (Control) was before BAPTA dialysis, and BAPTA indicates that the neuron was close to an astrocyte dialysed with BAPTA. In the middle trace, gabazine + CGP55845 was applied. The lower traces show a recording from neurons dialysed with high [Cl−] before and after astrocyte BAPTA dialysis. In control conditions, the membrane hyperpolarized after stimulation, while in all other conditions we observed a depolarization (scale bar = 1 s). B, average hyperpolarization in control (Control) and average depolarization after astrocyte BAPTA dialysis (BAPTA), GABAA and GABAB receptor blockade (Gabazine + CGP55845), 132 mm[Cl−] inside the patched neuron before (High [Cl−]i) and after astrocytic BAPTA dialysis (High [Cl−]i+ BAPTA). **P < 0.01, ***P < 0.001, one-way ANOVA.

Figure 6. Neurons dialysed with high [Cl−] show spontaneous depolarizing events in the vicinity of a BAPTA filled astrocyte.

Figure 6

Open in a new tab

A, stimulation (indicated by bar) induced a depolarization in a neuron dialysed with high chloride concentration before (high [Cl−]i) and after astrocytic (High [Cl−]i+ BAPTA) dialysis. B, a spontaneous depolarizing event shown in A is displayed with higher time resolution. C, time histogram of spontaneous occurring depolarizations after electrical stimulation (time point 0). Note that most events occurred between 2 and 4 s after the stimulation. D, as described in Fig. 1, average stimulus (STM) evoked depolarization (left, STM depol), time for half-repolarization (middle, ‘half-rep’) and integral of the voltage trace (ISTM) during the evoked depolarization (right) in high [Cl−]i and after BAPTA dialysis.

Neurons dialysed with high [Cl−] show spontaneous depolarizing events after astrocytes were dialysed with BAPTA

Neurons dialysed with 132 mm[Cl−] through the patch pipette (n= 22) showed a large evoked response (Fig. 6A, High [Cl−]i). The membrane depolarized by 52.7 ± 2.3 mV, and the half-repolarization was 1.15 ± 0.11 s while ISTM was 103 ± 9 mV s. Stimulus evoked depolarization (P= 3 × 10−10) as well as half-repolarization (P= 1.5 × 10−5) and ISTM (P= 1.6 × 10−8) were all significantly increased when compared to the evoked response of cells dialysed with normal [Cl−] (Fig. 1A). Neurons dialysed with high [Cl−] also typically fired a burst of action potentials during the repolarization phase, after the stimulus delivery. BAPTA-dialysis of an astrocyte in the vicinity of a neuron patched with high [Cl−]i pipette solution (High [Cl−]i+ BAPTA, n= 15) did not significantly affect the parameters of the evoked response (Fig. 6D): the stimulus evoked depolarization was 55.0 ± 2.9 mV (P= 0.5), the half-repolarization lasted 1.65 ± 0.26 s (P= 0.14) and ISTM was 120 ± 19 mV s (P= 0.4). After astrocytes were dialysed with BAPTA we observed, however, a new behaviour in some neurons (8 out of 15): spontaneous depolarizing events (Fig. 6A and B) occurred after the initial, stimulus correlated depolarization with the highest probability between 2 and 4 s after the stimulation (Fig. 6C). The events had average amplitude of 22.5 ± 3.2 mV and a half-repolarization time of 0.91 ± 0.54 s. During such depolarizations (Fig. 6B) neurons fired on average 9 ± 3 action potentials. These events were never observed in control cells.

mGluR, glycine and purinergic receptors and serine release scarcely affect the neuronal evoked response

We also tested the impact of neuromodulators which are known to mediate neuron–astrocyte communication on the stimulus induced neuronal depolarization. The combined application of mGluR1–5 antagonists LY367385 and MPEP (n= 8) caused a slight, but significant increase in the time for half-repolarization, namely from 0.56 ± 0.13 s to 0.63 ± 0.15 s (P= 0.02), but no other significant differences in depolarization parameters (Supplementary Table. 1). Strychnine (n= 5) had only a modest inhibitory effect shortening the time of half-repolarization from 0.52 ± 0.09 s to 0.37 ± 0.05 (P= 0.04). Neither CPT (Supplementary Table 1, n= 12) nor the serine-synthetase inhibitor d-AAO (Supplementary Table 1, n= 4) induced significant differences in the neuronal evoked response. Hence none of these signalling pathways played a significant role in this neuron–astrocyte communication.

BAPTA dialysis of astrocytes eliminated locally the Ca2+ response to stimulation, but a larger population of astrocytes was activated at a distance

After 45–60 min of astrocyte dialysis (n= 19) we observed a suppression of the Ca2+ response in astrocytes due to stimulation up to ∼200 μm from the dialysed cell (see Methods and Supplementary Fig. 1). Astrocytes outside this region still responded to the stimulation with the usual Ca2+ response (Fig. 7A). Furthermore, outside the region of dialysis (Fig. 7A), particularly in layer 2/3 and 5, even more cells responded to stimulation with a Ca2+ increase including cells in neighbouring barrels (n= 14). Thus the astrocyte Ca2+ response was no longer restricted to one barrel column while the kinetic of the response was as observed in controls (1–3 s time to peak after stimulation). To quantify the increase in the responding population, we measured the area covered by responding astrocytes (for details see Methods, Fig. 7B,C). The area increased from an average of 0.16 ± 0.04 mm2 (about 10 min before dialysis) to 0.35 ± 0.06 mm2 (about 60 min after dialysis) (P= 0.01). As a control we stimulated the slice once and then repeated the stimulation 60 min later without astrocytic BAPTA dialysis (n= 5). No significant difference (P= 0.9) was found between the areas of response (0.13 ± 0.04 mm2 at the first stimulation and 0.14 ± 0.04 mm2 at the second; Fig. 7C).

Figure 7. BAPTA dialysis of astrocytes eliminated locally the Ca2+ increase to stimulation, but a larger population of astrocytes was activated at a distance.

Figure 7

Open in a new tab

A, fluorescence image at the peak of the Ca2+ response after stimulation before (Control) and after dialysis of an astrocyte with BAPTA. The borders of the barrel columns are indicated by lines. Scale bar denotes 100 μm. The fluorescence recordings were obtained from three regions of interest and are displayed in the middle. Within the region of interest (ROI) 1 the Ca2+ evoked response is suppressed, ROI 2 shows an unaffected Ca2+ response after the BAPTA dialysis, astrocytes in ROI 3 are newly recruited after BAPTA dialysis. B, spatial distribution of brightness from the figures in A ranging from low (dark grey) to medium (orange) and to high (magenta) values. C, average area covered by responding astrocytes before (Control) and 45 to 60 min of BAPTA dialysis (±s.e.m.). We also did the experiment without BAPTA dialysis. Control1 is the control, Control2 a recording after waiting for 45–60 min. **P= 0.01, one-way ANOVA.

Astrocytic Ca2+ chelation increases the spontaneous synaptic activity

To study whether the BAPTA dialysis also affects spontaneous excitatory synaptic activity, we clamped the neurons at −70 mV (Fig. 8A), close to the Cl− equilibrium potential to minimize contributions of inhibitory currents. We observed that after dialysis of astrocytes with BAPTA the frequency of spontaneous postsynaptic currents was increased from 0.3 ± 0.03 Hz in control conditions (n= 11) to 0.8 ± 0.2 Hz after BAPTA dialysis (n= 8, P= 0.01; Fig. 8B). There was no significant change of the amplitude (16 ± 2 pA, control and 13 ± 1 pA, BAPTA, P= 0.1) and decay constant (4.7 ± 2 ms, control and 7.7 ± 1 ms, BAPTA, P= 0.1). The neuronal average access resistance (Ra) was 22 ± 0.9 MΩ (17 MΩ < Ra < 30 MΩ) in control and 21.6 ± 0.6 MΩ (20 MΩ < Ra < 25 MΩ) after astrocytic BAPTA dialysis (P= 0.6). The resting membrane potential Em was also not significantly different (P= 0.4) comparing neurons in control (Em=−67.9 ± 1.5 mV) and after astrocytic dialysis (Em=−68.5 ± 0.8 mV) and we did not observe any seizure-like event.

Figure 8. Astrocytic calcium chelation increased the frequency of spontaneous excitatory post synaptic current (ePSC).

Figure 8

Open in a new tab

A, ePSCs of a voltage clamp recording with a holding potential equal to −70 mV in control (black) and after astrocytic Ca2+ chelation (grey). Single events (•) are displayed at higher magnification in the framed quadrant below. B, ePSC frequency is significantly increased after astrocytic calcium chelation (unpaired t test, *P= 0.01).

Discussion

Dialysis with BAPTA as a tool to interfere with astrocyte function

We previously found that astrocytes in the barrel cortex respond to neuronal activity with a calcium response after stimulation in layer 4. This was not a direct astrocyte response to stimulation but depended on neuronal activity since it could be blocked by tetrodotoxin (Schipke et al. 2008). In the present study we interfered with astrocyte calcium signalling by dialysis of astrocytes with a membrane-impermeant calcium chelator (BAPTA) spreading from a single cell through the patch pipette similar to that described by Serrano et al (2006) and Andersson & Hanse (2010) As others previously observed (Kang et al. 1998; Liu et al. 2004; Serrano et al. 2006; Andersson & Hanse, 2010), we found that molecules of large dimensions such as sulforhodamine and BAPTA spread to coupled astrocytes. In our hands this process extended to a network of 6–12 cells up to ∼200 μm from the patched astrocyte. We assume that this procedure did not damage neurons since their resting membrane potential or access resistance was not altered.

Astrocyte BAPTA dialysis affects excitatory neurons

The present study provides evidence that astrocytes control inhibition of excitatory neurons (pyramidal cells and spiny stellate cells) in the barrel cortex. The target cells in our study were ‘regular spiking’: this action potential pattern defines typically pyramidal neurons (Chagnac-Amitai & Connors, 1989) and spiny stellate or star-pyramidal neurons (Feldmeyer et al. 2005). In contrast, a ‘fast spiking’ pattern is commonly associated with interneurons (Bacci et al. 2002; Beierlein et al. 2003; Sun, 2009). The morphology of the neurons studied here is also compatible with previous descriptions of pyramidal cells and spiny stellate cells (White & Rock, 1980; Lubke et al. 2000; Feldmeyer et al. 2002; Lubke et al. 2003; Lubke & Feldmeyer, 2007; Fox, 2008). Cells were characterized by typical morphological features such as dendritic star-like branching, longitudinal axonal projections terminating often in white matter, and triangular (pyramids) or round (stellate) cell somata with apical dendrite in star-pyramids.

BAPTA dialysis of astrocytes led to an increase in neuronal excitability

After BAPTA dialysis we observed an increase in excitability (or decrease in inhibition) of pyramidal neurons and spiny stellate cells in the vicinity of BAPTA-dialysed astrocytes using several experimental protocols:

  1. The depolarization from resting membrane potential in response to stimulation had larger amplitude and a longer time course with additional action potentials during the repolarization phase after BAPTA dialysis.

  2. With normal chloride concentration and clamping cells to −40 mV, we observed a hyperpolarization evoked by the stimulation. This was converted to a depolarization in neurons near BAPTA-dialysed astrocytes.

  3. In neurons dialysed with high chloride we found spontaneous depolarizations near BAPTA dialysed astrocytes which were never observed under control conditions.

  4. In neurons dialysed with high chloride and depolarized to about −40 mV, we observed that stimulation led to inhibition of action potential firing under control conditions. This inhibition was not observed after astrocyte BAPTA dialysis.

  5. Under control conditions an astrocytes calcium wave is restricted to a barrel field. Inhibiting GABAA receptors leds to a widespread Ca2+ activity involving astrocytes over several barrel fields (Schipke et al. 2008). Similarly after dialysis of a small astrocytic network with BAPTA more astrocytes responded outside the dialysed area in such way that the Ca2+ response was no longer restricted within a barrel field but involved astrocytes in neighbouring barrels. We therefore assume that the recruitment of a larger astrocyte population reflects an increase in activity in the neuronal network both after local astrocyte BAPTA dialysis and after GABAA receptor blockade.

  6. Spontaneous postsynaptic currents increased after astrocytes were dialysed with BAPTA as compared to controls. The recording conditions favoured the recording of excitatory over inhibitory currents since cells were clamped to −70 mV, and thus close to the imposed Cl− equilibrium potential.

In several of our paradigms, we used a stimulation protocol which led to a large depolarization of the neurons. We chose that protocol since it reliably led to an activation of Ca2+ signalling in the astrocytes. It closely resembles a type of spontaneous neuronal activity which can be induced e.g. by sustained low Ca2+ perfusion or GABAergic inhibition (Gutnick et al. 1982; Konnerth et al. 1986; Deisz & Prince, 1987; Chagnac-Amitai & Connors, 1989; Sutor & Luhmann, 1998; Marescaux et al. 1992; Vergnes et al. 1997) and was defined as a paroxysmal depolarization shift (Mutani, 1986; Pockberger, 1990). In our present study, we only observed under one condition paroxysmal depolarizations, namely after BAPTA dialysis of astrocytes and dialysis of neurons with high Cl− containing solution. Under all other conditions, these depolarizations were only evoked by stimulation.

Inhibition of GABA receptors mimics BAPTA dialysis of astrocytes

GABAergic inhibition plays a dominant role in the control of barrel cortex excitability (Swadlow, 2002). GABAA and GABAB receptor blockade mimicked the effect of BAPTA dialysis. We found that after BAPTA dialysis both the amplitude and the duration of the stimulation evoked depolarization were increased. Blocking GABAA receptors with gabazine led only to an increase of the amplitude while blocking GABAB receptors had little effect on the amplitude but prolonged the duration of the response. In combination, GABAA and GABAB receptor blockade mimicked the effect of astrocyte BAPTA dialysis.

The involvement of GABAA receptors (ligand gated Cl− channels) is further supported by the alteration of the neuronal response when changing the chloride gradient in the neuron. Under control conditions the reversal potential of the stimulus induced currents was at about −50 mV, and thus close to the imposed Cl− equilibrium potential of about −65 mV and largely dependent on GABAA-receptors. Dialysis of neurons with high chloride concentration (ECl∼0 mV) shifted the reversal potential of such current to about 0 mV indicating that the stimulus induced current in control conditions is largely carried by chloride. Since this conductance is both sensitive to GABAA receptor blockage and affected by dialysis of the neuron with high chloride, we can conclude that the GABAA receptors are on the neuron under study, while the same cannot be stated for GABAB receptors. GABAB receptors are metabotropic receptors that can act at both, the pre- and postsynaptic membrane. The postsynaptic action is mediated by a specific family of K+ channels, the G-protein-gated inwardly rectifying K+ (GIRK/Kir3) channels (Misgeld et al. 1995; Padgett & Slesinger, 2010). The type of GIRK channel coupled to GABAB receptors can vary depending on cell type and brain region (Koyrakh et al. 2005). We found that GABAB receptor blockade did not increase the amplitude of the stimulation induced response, but prolonged the depolarization phase. We speculate that this could be due to the block of (GABAB receptor mediated) K+ channels which contribute in control conditions to the membrane repolarization.

Mechanism of the neuronal inhibition mediated by BAPTA dialysis

So far, we have not identified the mechanism by which astrocyte BAPTA dialysis leads to the increase in the excitability of excitatory neurons in the barrel cortex. It could be either a direct effect of the astrocytes or indirect, e.g. via activation of inhibitory interneurons. Lee et al. (2010) recently demonstrated that astrocytes tonically release GABA by permeation through the Bestrophin 1 anion channel. There is also evidence that astrocytes activate interneurons. Kang et al (1998) provided evidence that interneurons trigger GABAB receptor mediated Ca2+ responses in astrocytes which potentiate inhibitory postsynaptic currents in CA1 pyramidal neurons via activating interneurons. This potentiation could be abolished by astrocyte BAPTA dialysis. There are a number of other mechanisms described by which neurons and astrocytes interact. Serrano et al. (2006) provided evidence of heterosynaptic depression in hippocampus where an inhibitory neuronal network activates astrocytes, which in turn release ATP in a Ca2+-dependent fashion. ATP is then degraded to adenosine and causes the depression. Andersson & Hanse (2010) reported that a short term depression of glutamatergic transmission in CA1 pyramidal cells is abolished by dialysis of nearby astrocytes with BAPTA. Ca2+ mediated release of d-serine by astrocytes was also shown to control NMDA receptor mediated plasticity in CA1 glutamatergic synapses (Henneberger et al. 2010). We therefore tested whether these types of interactions play a role in the barrel cortex and found that neither d-serine nor adenosine nor glutamatergic transmission plays a significant role. Astrocytes are known to express GABA transporters (GATs) and they contribute up to 20% of GABA uptake from the extracellular space (Schousboe et al. 1977; Hertz et al. 1978; Schousboe, 2000). A modulation in GABA uptake or even the reverse transport of GABA through the GABA transporters might be a potential mechanism for the GABAergic inhibition mediated by astrocytes (Kozlov et al. 2006; Heja et al. 2009; Park et al. 2009). In conclusion, astrocytes control neuronal inhibition in the barrel cortex through a Ca2+-dependent mechanism potentially mediated by both GABAA and GABAB receptors.

Acknowledgments

We like to thank Prof. Dirk Feldmeyer (Forschungzentrum Juelich, Germany), Prof. Joachim Lübke (Juelich, Germany), Prof. Alexej Verkhratsky (Manchester, UK), Prof. Dr Christian Steinhäuser (Bonn, Germany), Dr Christiane Nolte (Berlin, Germany), Prof. Philip Haydon, (Boston, USA), Prof. Rosemary Grantyn (Berlin, Germany) and Prof. Rudolf Deisz (Berlin, Germany) for critical comments and helpful discussion. We thank Prof. Frank Kirchhoff (Homburg / Saar, Germany) for providing the mRFP/GFAP mouse strain, and Dr Karljin Van Aerde (Juelich, Germany) for the custom made Igor Pro function kindly provided for neuronal characterization. This work was supported by the Deutsche Forschungsgemeinschaft (SPP 1172).

Glossary

Abbreviations

Cy3

cyanine dye 3

d-AAO

d-amino acid oxidase

eGFP

enhanced green fluorescent protein

GFAP

glial fibrillary acidic protein

GAT

GABA transporter

ISTM

integral of stimulus evoked depolarization

mRFP

murine red fluorescent protein

STM depol

stimulus evoked depolarization

Author contributions

All authors have taken part in the conception and design of the experiments, collection, analysis and interpretation of data, and drafting the article or revising it critically for important intellectual content. The authors have no conflict of interest.

Supplementary material

Supplementary Fig.1

Supplementary Fig.2

Supplementary Fig.3

Supplementary Fig.4

Supplementary table.1

tjp0589-1159-SD1.pdf (262.6KB, pdf)

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors

References

  1. Agmon A, Connors BW. Thalamocortical responses of mouse somatosensory (barrel) cortex in vitro. Neuroscience. 1991;41:365–379. doi: 10.1016/0306-4522(91)90333-j. [DOI] [PubMed] [Google Scholar]
  2. Agmon A, Connors BW. Correlation between intrinsic firing patterns and thalamocortical synaptic responses of neurons in mouse barrel cortex. J Neurosci. 1992;12:319–329. doi: 10.1523/JNEUROSCI.12-01-00319.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Agulhon C, Fiacco TA, McCarthy KD. Hippocampal short- and long-term plasticity are not modulated by astrocyte Ca2+ signaling. Science. 2010;327:1250–1254. doi: 10.1126/science.1184821. [DOI] [PubMed] [Google Scholar]
  4. Andersson M, Hanse E. Astrocytes impose postburst depression of release probability at hippocampal glutamate synapses. J Neurosci. 2010;30:5776–5780. doi: 10.1523/JNEUROSCI.3957-09.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bacci A, Sancini G, Verderio C, Armano S, Pravettoni E, Fesce R, Franceschetti S, Matteoli M. Block of glutamate-glutamine cycle between astrocytes and neurons inhibits epileptiform activity in hippocampus. J Neurophysiol. 2002;88:2302–2310. doi: 10.1152/jn.00665.2001. [DOI] [PubMed] [Google Scholar]
  6. Bardoni R, Ghirri A, Zonta M, Betelli C, Vitale G, Ruggieri V, Sandrini M, Carmignoto G. Glutamate-mediated astrocyte-to-neuron signalling in the rat dorsal horn. J Physiol. 2010;588:831–846. doi: 10.1113/jphysiol.2009.180570. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Beierlein M, Gibson JR, Connors BW. Two dynamically distinct inhibitory networks in layer 4 of the neocortex. J Neurophysiol. 2003;90:2987–3000. doi: 10.1152/jn.00283.2003. [DOI] [PubMed] [Google Scholar]
  8. Chagnac-Amitai Y, Connors BW. Synchronized excitation and inhibition driven by intrinsically bursting neurons in neocortex. J Neurophysiol. 1989;62:1149–1162. doi: 10.1152/jn.1989.62.5.1149. [DOI] [PubMed] [Google Scholar]
  9. Deisz RA, Prince DA. Effect of D890 on membrane properties of neocortical neurons. Brain Res. 1987;422:63–73. doi: 10.1016/0006-8993(87)90540-3. [DOI] [PubMed] [Google Scholar]
  10. Feldmeyer D, Lubke J, Silver RA, Sakmann B. Synaptic connections between layer 4 spiny neurone-layer 2/3 pyramidal cell pairs in juvenile rat barrel cortex: physiology and anatomy of interlaminar signalling within a cortical column. J Physiol. 2002;538:803–822. doi: 10.1113/jphysiol.2001.012959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Feldmeyer D, Roth A, Sakmann B. Monosynaptic connections between pairs of spiny stellate cells in layer 4 and pyramidal cells in layer 5A indicate that lemniscal and paralemniscal afferent pathways converge in the infragranular somatosensory cortex. J Neurosci. 2005;25:3423–3431. doi: 10.1523/JNEUROSCI.5227-04.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fiacco TA, Agulhon C, Taves SR, Petravicz J, Casper KB, Dong X, Chen J, McCarthy KD. Selective stimulation of astrocyte calcium in situ does not affect neuronal excitatory synaptic activity. Neuron. 2007;54:611–626. doi: 10.1016/j.neuron.2007.04.032. [DOI] [PubMed] [Google Scholar]
  13. Fox K. Barrel Cortex. Cambridge, UK: Cambridge University Press; 2008. [Google Scholar]
  14. Gutnick MJ, Connors BW, Prince DA. Mechanisms of neocortical epileptogenesis in vitro. J Neurophysiol. 1982;48:1321–1335. doi: 10.1152/jn.1982.48.6.1321. [DOI] [PubMed] [Google Scholar]
  15. Halassa MM, Fellin T, Takano H, Dong JH, Haydon PG. Synaptic islands defined by the territory of a single astrocyte. J Neurosci. 2007;27:6473–6477. doi: 10.1523/JNEUROSCI.1419-07.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Heja L, Barabas P, Nyitrai G, Kekesi KA, Lasztoczi B, Toke O, Tarkanyi G, Madsen K, Schousboe A, Dobolyi A, Palkovits M, Kardos J. Glutamate uptake triggers transporter-mediated GABA release from astrocytes. PLoS One. 2009;4:e7153. doi: 10.1371/journal.pone.0007153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Henneberger C, Papouin T, Oliet SH, Rusakov DA. Long-term potentiation depends on release of D-serine from astrocytes. Nature. 2010;463:232–236. doi: 10.1038/nature08673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hertz L, Wu PH, Schousboe A. Evidence for net uptake of GABA into mouse astrocytes in primary cultures – its sodium dependence and potassium independence. Neurochem Res. 1978;3:313–323. doi: 10.1007/BF00965577. [DOI] [PubMed] [Google Scholar]
  19. Hirrlinger PG, Scheller A, Braun C, Quintela-Schneider M, Fuss B, Hirrlinger J, Kirchhoff F. Expression of reef coral fluorescent proteins in the central nervous system of transgenic mice. Mol Cell Neurosci. 2005;30:291–303. doi: 10.1016/j.mcn.2005.08.011. [DOI] [PubMed] [Google Scholar]
  20. Kang J, Jiang L, Goldman SA, Nedergaard M. Astrocyte-mediated potentiation of inhibitory synaptic transmission. Nat Neurosci. 1998;1:683–692. doi: 10.1038/3684. [DOI] [PubMed] [Google Scholar]
  21. Kole MH, Stuart GJ. Is action potential threshold lowest in the axon? Nat Neurosci. 2008;11:1253–1255. doi: 10.1038/nn.2203. [DOI] [PubMed] [Google Scholar]
  22. Konnerth A, Heinemann U, Yaari Y. Nonsynaptic epileptogenesis in the mammalian hippocampus in vitro. I. Development of seizurelike activity in low extracellular calcium. J Neurophysiol. 1986;56:409–423. doi: 10.1152/jn.1986.56.2.409. [DOI] [PubMed] [Google Scholar]
  23. Koyrakh L, Lujan R, Colon J, Karschin C, Kurachi Y, Karschin A, Wickman K. Molecular and Cellular diversity of neuronal G-protein-gated potassium channels. J Neurosci. 2005;25:11468–11478. doi: 10.1523/JNEUROSCI.3484-05.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kozlov AS, Angulo MC, Audinat E, Charpak S. Target cell-specific modulation of neuronal activity by astrocytes. Proc Natl Acad Sci U S A. 2006;103:10058–10063. doi: 10.1073/pnas.0603741103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kumaria A, Tolias CM, Burnstock G. ATP signalling in epilepsy. Purinergic Signal. 2008;4:339–346. doi: 10.1007/s11302-008-9115-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Liu QS, Xu Q, Kang J, Nedergaard M. Astrocyte activation of presynaptic metabotropic glutamate receptors modulates hippocampal inhibitory synaptic transmission. Neuron Glia Biol. 2004;1:307–316. doi: 10.1017/S1740925X05000190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lubke J, Egger V, Sakmann B, Feldmeyer D. Columnar organization of dendrites and axons of single and synaptically coupled excitatory spiny neurons in layer 4 of the rat barrel cortex. J Neurosci. 2000;20:5300–5311. doi: 10.1523/JNEUROSCI.20-14-05300.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lubke J, Feldmeyer D. Excitatory signal flow and connectivity in a cortical column: focus on barrel cortex. Brain Struct Funct. 2007;212:3–17. doi: 10.1007/s00429-007-0144-2. [DOI] [PubMed] [Google Scholar]
  29. Lubke J, Roth A, Feldmeyer D, Sakmann B. Morphometric analysis of the columnar innervation domain of neurons connecting layer 4 and layer 2/3 of juvenile rat barrel cortex. Cereb Cortex. 2003;13:1051–1063. doi: 10.1093/cercor/13.10.1051. [DOI] [PubMed] [Google Scholar]
  30. Marescaux C, Vergnes M, Liu Z, Depaulis A, Bernasconi R. GABAB receptor involvement in the control of genetic absence seizures in rats. Epilepsy Res Suppl. 1992;9:131–138. [PubMed] [Google Scholar]
  31. Martin ED, Fernandez M, Perea G, Pascual O, Haydon PG, Araque A, Cena V. Adenosine released by astrocytes contributes to hypoxia-induced modulation of synaptic transmission. Glia. 2007;55:36–45. doi: 10.1002/glia.20431. [DOI] [PubMed] [Google Scholar]
  32. Misgeld U, Bijak M, Jarolimek W. A physiological role for GABAB receptors and the effects of baclofen in the mammalian central nervous system. Prog Neurobiol. 1995;46:423–462. doi: 10.1016/0301-0082(95)00012-k. [DOI] [PubMed] [Google Scholar]
  33. Mutani R. Neurophysiological mechanisms underlying epileptogenesis. Funct Neurol. 1986;1:385–389. [PubMed] [Google Scholar]
  34. Nedergaard M, Rodriguez JJ, Verkhratsky A. Glial calcium and diseases of the nervous system. Cell Calcium. 2010;47:140–149. doi: 10.1016/j.ceca.2009.11.010. [DOI] [PubMed] [Google Scholar]
  35. Nolte C, Matyash M, Pivneva T, Schipke CG, Ohlemeyer C, Hanisch UK, Kirchhoff F, Kettenmann H. GFAP promoter-controlled EGFP-expressing transgenic mice: a tool to visualize astrocytes and astrogliosis in living brain tissue. Glia. 2001;33:72–86. [PubMed] [Google Scholar]
  36. Padgett CL, Slesinger PA. GABAB receptor coupling to G-proteins and ion channels. Adv Pharmacol. 2010;58:123–147. doi: 10.1016/S1054-3589(10)58006-2. [DOI] [PubMed] [Google Scholar]
  37. Park JB, Jo JY, Zheng H, Patel KP, Stern JE. Regulation of tonic GABA inhibitory function, presympathetic neuronal activity and sympathetic outflow from the paraventricular nucleus by astroglial GABA transporters. J Physiol. 2009;587:4645–4660. doi: 10.1113/jphysiol.2009.173435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Perea G, Navarrete M, Araque A. Tripartite synapses: astrocytes process and control synaptic information. Trends Neurosci. 2009;32:421–431. doi: 10.1016/j.tins.2009.05.001. [DOI] [PubMed] [Google Scholar]
  39. Peters O, Schipke CG, Hashimoto Y, Kettenmann H. Different mechanisms promote astrocyte Ca2+ waves and spreading depression in the mouse neocortex. J Neurosci. 2003;23:9888–9896. doi: 10.1523/JNEUROSCI.23-30-09888.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Petravicz J, Fiacco TA, McCarthy KD. Loss of IP3 receptor-dependent Ca2+ increases in hippocampal astrocytes does not affect baseline CA1 pyramidal neuron synaptic activity. J Neurosci. 2008;28:4967–4973. doi: 10.1523/JNEUROSCI.5572-07.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Pockberger H. [Focal epilepsy: generation and spreading mechanisms in experimental conditions] Wien Klin Wochenschr. 1990;102:201–205. [PubMed] [Google Scholar]
  42. Schipke CG, Haas B, Kettenmann H. Astrocytes discriminate and selectively respond to the activity of a subpopulation of neurons within the barrel cortex. Cereb Cortex. 2008;18:2450–2459. doi: 10.1093/cercor/bhn009. [DOI] [PubMed] [Google Scholar]
  43. Schousboe A. Pharmacological and functional characterization of astrocytic GABA transport: a short review. Neurochem Res. 2000;25:1241–1244. doi: 10.1023/a:1007692012048. [DOI] [PubMed] [Google Scholar]
  44. Schousboe A, Svenneby G, Hertz L. Uptake and metabolism of glutamate in astrocytes cultured from dissociated mouse brain hemispheres. J Neurochem. 1977;29:999–1005. doi: 10.1111/j.1471-4159.1977.tb06503.x. [DOI] [PubMed] [Google Scholar]
  45. Serrano A, Haddjeri N, Lacaille JC, Robitaille R. GABAergic network activation of glial cells underlies hippocampal heterosynaptic depression. J Neurosci. 2006;26:5370–5382. doi: 10.1523/JNEUROSCI.5255-05.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sun QQ. A novel role of dendritic gap junction and mechanisms underlying its interaction with thalamocortical conductance in fast spiking inhibitory neurons. BMC Neurosci. 2009;10:131. doi: 10.1186/1471-2202-10-131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sutor B, Luhmann HJ. Involvement of GABA(B) receptors in convulsant-induced epileptiform activity in rat neocortex in vitro. Eur J Neurosci. 1998;10:3417–3427. doi: 10.1046/j.1460-9568.1998.00351.x. [DOI] [PubMed] [Google Scholar]
  48. Swadlow HA. Thalamocortical control of feed-forward inhibition in awake somatosensory ‘barrel’ cortex. Philos Trans R Soc Lond B Biol Sci. 2002;357:1717–1727. doi: 10.1098/rstb.2002.1156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Vergnes M, Boehrer A, Simler S, Bernasconi R, Marescaux C. Opposite effects of GABAB receptor antagonists on absences and convulsive seizures. Eur J Pharmacol. 1997;332:245–255. doi: 10.1016/s0014-2999(97)01085-6. [DOI] [PubMed] [Google Scholar]
  50. Wang X, Lou N, Xu Q, Tian GF, Peng WG, Han X, Kang J, Takano T, Nedergaard M. Astrocytic Ca2+ signaling evoked by sensory stimulation in vivo. Nat Neurosci. 2006;9:816–823. doi: 10.1038/nn1703. [DOI] [PubMed] [Google Scholar]
  51. White EL, Rock MP. Three-dimensional aspects and synaptic relationships of a Golgi-impregnated spiny stellate cell reconstructed from serial thin sections. J Neurocytol. 1980;9:615–636. doi: 10.1007/BF01205029. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

tjp0589-1159-SD1.pdf (262.6KB, pdf)

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

RESOURCES