Fast inhibitory synaptic transmission is primarily mediated by synaptically released γ-aminobutyric acid (GABA) acting on postsynaptic GABAA receptors. GABA acting on GABAA receptors produces not only phasic but also tonic inhibitions by persistent activation of extrasynaptic receptors. However, the mechanistic characteristics of tonic inhibition in the neocortex are not well-understood. To address this, we studied pharmacologically isolated GABAA receptor–mediated currents in neocortical pyramidal neurons in rat brain slices. Bath application of bicuculline blocked miniature inhibitory postsynaptic currents (mIPSCs) and produced an outward shift in baseline holding current (Ihold). Low concentrations of SR95531, a competitive GABAA receptor antagonist, abolished mIPSCs but had no significant effect on Ihold. The benzodiazepine midazolam produced an inward shift in Ihold by augmenting tonic GABAA receptor–mediated currents, which were significantly greater in layer V neurons than in layer II/III. Single-cell reverse transcriptase–polymerase chain reaction (RT-PCR) revealed a relatively higher expressions of α1 and α5 subunit mRNA in layer V neurons. L-655708, an α5 subunit–specific inverse agonist, reduced tonic currents in layer V but not in layer II/III neurons, whereas zolpidem, an α1-subunit agonist, exerted equivalent effects in both layers. These data suggest that the α1 GABAA receptor subunit is generally involved in tonic inhibition in pyramidal neurons of the neocortex, whereas the α5 subunit is specifically involved in layer V neurons.
γ-aminobutyric acid type A (GABAA) receptors mediate fast synaptic inhibition in the mammalian central nervous system and regulate neuronal firing either by hyperpolarizing the membrane potential or by shunting excitatory inputs. Conventionally, transient activation of synaptic GABAA receptors mediates phasic inhibition, but recently it has become apparent that distinct GABAA receptors also participate in another type of inhibitory role (for review, see Mody and Pearce 2004; Semyanov et al. 2004; Farrant and Nusser 2005; Kullmann et al. 2005). This role involves mediation of tonic inhibition by the continuous activation of extrasynaptic GABAA receptors (Brickley et al. 1996; Banks and Pearce 2000; Bai et al. 2001; Hamann et al. 2002; Nusser and Mody 2002; Semyanov et al. 2003). These receptors can be activated by ambient GABA spilt over from the synaptic cleft (Rossi and Hamann 1998) and/or released nonvesicularily (Rossi et al. 2003). Experiments using gene knockout mice showed that the α6 and δ GABAA receptor subunits mediate tonic inhibition in cerebellar granule cells (Brickley et al. 2001) and dentate gyrus granule cells (DGGCs) (Stell et al. 2003), respectively. Variations in subunit composition confer considerable pharmacological differences between the synaptic and extrasynaptic GABAA receptors (Robello et al. 1999; Hutcheon et al. 2000; Bai et al. 2001; Nusser and Mody 2002). Although the existence of tonic GABAA receptor activation in adult neocortical neurons has been suggested (Salin and Prince 1996), mechanistic characterizations have not yet been undertaken.
In the present study, we demonstrate the molecular and pharmacological properties of tonic inhibition in the rat neocortex. Some of these results have been published in abstract form (Yamada et al. 2004b).
Materials and Methods
Brain Slice Preparation
All experiments conformed to the guidelines issued by Shizuoka University and Hamamatsu University School of Medicine on the ethical use of animals in experimentation, and all efforts were made to minimize the number of animals used and their suffering.
The procedures used for preparing rat brain slices containing the somatosensory neocortex were similar to those described previously (Yamada et al. 2004a). After the animals (63 Wistar rats of 4–6 weeks old) were deeply anesthetized by inhalation of halothane, they were decapitated, and brain sections were quickly placed in cold (4 °C), oxygenated, modified artificial cerebrospinal fluid (ACSF) containing (in mM) 220 sucrose, 2.5 KCl, 1.25 NaH2PO4, 12.0 MgSO4, 0.5 CaCl2·2H2O, 26.0 NaHCO3, and 30.0 glucose (330–340 mOsm). Coronal slices (400 μm) of somatosensory cortex were cut using a vibratome (VT-1000S; Leica, Bensheim, Germany). Slices were kept in oxygenated standard ACSF consisting of (in mM) 126 NaCl, 2.5 KCl, 1.25 NaH2PO4, 2.0 MgSO4, 2.0 CaCl2, 26.0 NaHCO3, and 20.0 glucose with 95% O2–5% CO2 at room temperature before recording.
Whole-Cell Patch-Clamp Recordings
Brain slices were transferred to a recording chamber on the stage of a microscope (ECLIPSE E600FN; Nikon, Tokyo, Japan) and continuously perfused with oxygenated standard ACSF at a flow rate of 2 mL/min and a temperature of 32 °C. Neocortical pyramidal neurons in the slices were viewed on a monitor via a 40× water-immersion objective lens with an infrared differential interference contrast filter and a charge coupled device camera (ORCA-ER C4742-95; Hamamatsu Photonics, Shizuoka, Japan). Real-time images were contrast-enhanced using Aquacosmos software (Hamamatsu Photonics).
Patch electrodes were fabricated from borosilicate capillary tubing of 1.5 mm diameter (GD-1.5; Narishige, Tokyo, Japan) using a P97 horizontal puller (Sutter, Novato, CA). The electrode resistance ranged from 3 to 4.5 MΩ. The pipette solution contained (in mM) 130 CsCl2, 2 MgCl2, 0.5 ethyleneglycol-bis(2-aminoethylether)-N,N,N′,N′-tetra acetic acid, 10 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, 3 Mg(adenosine triphosphate)2, and 0.4 guanosine triphosphate (pH 7.3 with CsOH). Membrane currents and membrane potentials were recorded using an Axopatch 200B amplifier and digitized at 5–10 kHz by means of a Digidata 1332A data-acquisition system (Axon Instruments, Sunnyvale, CA). Data were acquired by means of pClamp8 software (Axon Instruments) and stored on the hard disk for off-line analysis using Clampfit8 software (Axon Instruments). Series resistance compensation was not applied. Whole-cell recordings were made using voltage-clamp mode (Vh = −60 mV) in the presence of an α-amino-3-hydroxy-5-methyl-4-isoxazolyle propionic acid/kainate receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 μM), a N-methyl-D-aspartate receptor antagonist D(−)-2-amino-5-phosphonopentanoic acid (D-AP5, 50 μM), a GABAB receptor antagonist CGP55845 (3 μM), and tetrodotoxin (TTX, 0.3 μM). All drugs tested for synaptic responses were applied by superfusion or pressure application from a micropipette into the vicinity of the neurons being recorded. The mean holding current (Ihold) for each recording was calculated from 100 ms epochs containing no synaptic events taken every second for a 30-s period during control or drug application.
Single-Cell RT-PCR Analysis
Single-cell reverse transcriptase–polymerase chain reaction (RT-PCR) was performed as previously described (Browne et al. 2001; Yamada et al. 2004a). To harvest cytoplasm for the subsequent RT-PCR reaction, each patch-clamp pipette was filled with 10 μl standard pipette solution. After patch-clamp recording, mild suction was applied to aspirate the contents of the cell into the tip of the recording electrode, and this was then expelled into a microfuge tube. Subsequently, reverse transcription (RT) reagents were added. A 20-μl reaction contained 10 μl solution with cytoplasm, 5× RT buffer (4 μl; Invitrogen, Tokyo, Japan), 10 mM dithiothreitol, 0.5 mM each deoxynucleotide (dNTP; Amersham Biosciences, Piscataway, NJ), 5 μM random hexamer oligonucleotide (Takara Bio, Shiga, Japan), 10 U/μL reverse transcriptase (Superscript II; Invitrogen), and 2 U/μL RNase inhibitor (RNasin; Promega, Madison, WI) in a thin-walled polymerase chain reaction (PCR) tube. The mixture was incubated for 10 min at 25 °C then for 1 h at 42 °C. After the RT reaction, multiplex PCR using nested primers was performed. In the first round of PCR, outer primers for each individual GABAA subunit were employed simultaneously. These primer sequences have previously been described (Browne et al. 2001). A 50-μl PCR reaction consisted of 10 μl RT reaction products, 2× Qiagen multiplex PCR kit master mix, and 0.2 μM of each outer primer. The PCR conditions consisted of an initial 15-min preincubation at 95 °C followed by 35 cycles involving denaturation at 94 °C for 30 s, annealing at 57 °C for 1.5 min, and extension at 72 °C for 1.5 min in a thermal cycler (PC-801; ASTEC, Fukuoka, Japan). The resulting multiplex amplification product was then diluted 20-fold by DNase-free H2O (Nippon Gene, Tokyo, Japan). One microliter of the diluted first-round PCR product was used as a template for the second-round PCR. The 25-μl second-round PCR reaction consisted of 10× PCR buffer, 3 mM MgCl2, 0.2 mM of each dNTP, 1 μl of diluted first-round PCR product, and 2 U HotStarTaq DNA polymerase (QIAGEN, Hilden, Germany). In this part of the experiment, parallel PCRs containing one pair of gene-specific nested primers per reaction (0.2 μM) were performed. PCR conditions consisted of an initial 15-min preincubation at 95 °C followed by 40 cycles involving denauration at 94 °C for 30 s, annealing at 55 °C for 1 min, and extension at 72 °C for 1 min. Aliquots of the PCR products were analyzed by 2% agarose gel electrophoresis (Seakem GTG Agarose 2%; Takara Bio) and visualized using ethidium bromide. The entire product from the second-round PCR was run in parallel with known molecular weight markers (100 bp ladder; Promega). As shown in Table 1, expected sizes of the amplified GABAA subunit fragments were calculated from the published sequences (Browne et al. 2001). To confirm the identity of the amplified fragments, each product was nucleotide sequenced (ABI PRISM 310; Applied Biosystems, Foster, CA) with Bigdye (Applied Biosystems).
|Rat primer||Inner||Outer||bp (inner/outer)|
|Rat primer||Inner||Outer||bp (inner/outer)|
fw = forward, rv = reverse.
The following drugs were used: GABA, bicuculline methiodide (BMI), diazepam, picrotoxin, SR95531, THDOC (5α-pregnane-3α, 21-diol-20-one), zolpidem (Sigma, San Diego, CA), CNQX, CGP55845, D-AP5, L-655708, (Tocris, Ellisville, MO), midazolam, and TTX (Wako, Osaka, Japan).
Numerical data are presented as mean ± standard error of the mean. Unless otherwise noted, statistical significance was achieved if P < 0.05 as determined by the unpaired 2-tailed Student's t-test.
Tonic GABAA Receptor Activation in Layer V Neocortical Neurons
Whole-cell voltage-clamp recordings were made from layer V pyramidal cells in rat cortical slices. The pipette solution contained a high concentration of cesium chloride, so that GABAA receptor–mediated currents at −60 mV were inward, and most potassium currents were blocked. We estimated the tonic GABAA receptor–mediated current as the change in baseline Ihold produced by blocking GABAA receptors with 10 μM BMI (14.9 ± 3.5 pA, n = 6, Fig. 1A). The high-affinity GABAA receptor antagonist SR95531 (0.5 μM) abolished the mIPSCs but produced no significant shift in the Ihold (−0.4 ± 0.5 pA, n = 6, Fig. 1B). These currents were observed in dose-dependent manner (Fig. 1C). The apparent insensitivity of SR95531 to the tonic GABAA current could result from a higher affinity of the tonic receptors for BMI and/or the greater potency of SR95531 for synaptic receptors (Bai et al. 2001; Liang et al. 2004). Indeed, high concentrations (>10 μM) of SR95531 abolished not only the phasic mIPSCs but also the tonic currents (Fig. 1C). We next applied NO711 (20 μM), a specific GABA transporter inhibitor, to raise the extracellular GABA concentration. This led to an increase in Ihold (−18.5 ± 3.1 pA, n = 6, Fig. 1D) without apparent changes in mIPSCs amplitude (−31.1 ± 3.6 and −31.4 ± 2.8 pA), indicating that the tonic current was not a summation of mIPSCs but attributable to the continuous GABAA receptor activation by endogenous GABA.
Tonic Currents Were Significantly Enhanced by a Benzodiazepine in Layer V Neocortical Neurons
We next tested whether the tonic GABAA current in cortical neurons is also sensitive to benzodiazepines, as recently reported in cultured hippocampal neurons (Bai et al. 2001). As shown in Figure 2A, application of midazolam produced an inward current, indicating an enhancement of the tonic current by midazolam. The concentration–response relationship for the shifts in Ihold indicated a half-maximal enhancement of 22.4 nM in layer V neurons (Fig. 2B). The increases in the tonic currents by midazolam were blocked by BMI but not by low concentration of SR95531 (Fig. 2C), indicating augmentation of the tonic GABAA receptor–mediated currents. Another benzodiazepine, diazepam, also increased tonic currents (data not shown).
Next, we examined whether the tonic current was specifically present in layer V neurons. When the midazolam (1 μM)-induced Ihold shifts (in pA) were compared between the cortical layers (1-way analysis of variance followed by Student–Newman–Keuls test), the shift in layer V (22.6 ± 2.5, n = 6) was significantly larger than in layer II/III (11.9 ± 2.5, n = 6, P < 0.05), layer IV (12.9 ± 2.2, n = 6, P < 0.05), and layer VI neurons (10.0 ± 3.0, n = 5, P < 0.01).
GABAA Receptor Subunit mRNA Expression in Layer II/III and Layer V Pyramidal Neurons
To reveal the relationship between GABAA receptor subunits and tonic currents, we performed single-cell RT-PCR on mRNAs from 6 individual GABAA receptor subunits (α1–5 and δ) expressed in the cerebral cortex (Pirker et al. 2000). Recordings of BMI (50 μM) modulation of Ihold were successful in 12 layer V and 12 layer II/III pyramidal cells, which were used for GABAA receptor subunit analysis (Fig. 3A). The BMI-induced inhibition of tonic currents in layer V neurons (18.8 ± 1.9 pA) was significantly larger than those in layer II/III neurons (10.0 ± 1.1 pA, P < 0.001). The tonic current density (ΔIhold divided by cell capacitance) was also significantly larger in layer V neurons (1.1 ± 0.1 pA/pF) than in layer II/III neurons (0.8 ± 0.1 pA/pF, P < 0.01) (Fig. 3B). The incidence of GABAA receptor subunit mRNA expression in both layers is illustrated in Figure 3C, which indicates that mRNA expression levels of the α1 and α5 subunits were significantly higher in layer V neurons than in layer II/III neurons.
Pharmacological Differences of Tonic Currents between Layer V and Layer II/III Neurons
To clarify the subunits responsible for the laminar differences in the tonic current density, we next tested the effects of subunit-specific agonists in each layer. L-655708, an α5 subunit–specific inverse agonist, reduced the tonic current density in layer V (0.9 ± 0.1 pA/pF, n = 5) but not in layer II/III neurons (0.01 ± 0.08 pA/pF, n = 5, Fig. 4A). In contrast, zolpidem, an α1-subunit agonist, produced equivalent tonic current density in layer II/III (−1.1 ± 0.02 pA/pF, n = 4) and layer V (−1.3 ± 0.1 pA/pF, n = 4) pyramidal neurons (Fig. 4B). THDOC, a δ subunit–specific agonist, was not effective in either layer II/III (−0.08 ± 0.1 pA/pF, n = 4) or layer V neurons (−0.09 ± 0.1 pA/pF, n = 4, Fig. 4C). On the other hand, THDOC increased the tonic conductance in the DGGCs (0.6 ± 0.1 pA/pF, n = 3, Fig. 4C) in consistent with the previous report (Stell et al. 2003).
We investigated the GABAA receptor subunits that generate tonic inhibitory currents in the rat neocortex. Layer V pyramidal neurons had larger tonic GABAA receptor–mediated currents than any other layers. Using single-cell RT-PCR in combination with pharmacological approaches, we found that tonic currents are mediated via α1 and α5 subunit–containing GABAA receptors in layer V pyramidal neurons and via α1 subunit–containing GABAA receptors in layer II/III pyramidal cells.
Salin and Prince (1996) reported that both tonic and phasic GABAA receptor activation occurs in neocortical pyramidal neurons. We also observed GABAA receptor–mediated tonic currents in all layers. These currents were pharmacologically distinct from phasic currents because mIPSCs but not the tonic currents were abolished by low concentrations of SR95531. This result is consistent with those previously reported in the hippocampus (Bai et al. 2001; Stell and Mody 2002). SR95531 competes with GABA bindings (Hamann et al. 2002), thus the lower sensitivity of tonic currents to this agent suggests a relatively higher affinity for GABA binding of the receptors mediating tonic currents. Blocking GABA uptake also revealed comparable tonic currents, confirming that these were due to the persistent activation of GABAA receptors by endogenous GABA. Benzodiazepines do not directly activate native GABAA receptors in the absence of GABA, but potentiate GABA-evoked channel opening by increasing agonist affinity (Lavoie and Twyman 1996). In agreement with this, the application of midazolam produced an inward current in a dose-dependent manner. These characteristics were consistent with those identified by binding assays (Johnston 1996) and those described previously (Rogers et al. 1994; Bai et al. 2001).
GABAA receptors containing the α1, α2, α3, or α5 subunits are sensitive to benzodiazepine modulation, whereas those containing the α4 and α6 (which are not expressed in the neocortex; Pirker et al. 2000) subunits are insensitive (for review see, Mohler et al. 2002; Fritschy and Brunig 2003). In our experiments, midazolam evoked an increase in the amplitude of tonic currents in all layers, thus the α subunits may be involved in tonic inhibition in rat neocortical neurons. It has been reported that there is a higher expression of the α5 subunit in the deep layers compared with superficial layers (Sur et al. 1999; Pirker et al. 2000; Li et al. 2001). Consistent with this, high expression levels of the α5 subunit were identified in layer V pyramidal neurons but not layer II/III pyramidal cells. In addition, application of an α5 subunit–specific inverse agonist, L-655708, reduced tonic currents in layer V but not in layer II/III pyramidal neurons. The correlation between the pharmacological properties and mRNA expression profiles suggests that α5 subunit–containing receptors mediate the tonic GABAergic inhibition in layer V pyramidal neurons.
We also tested the effect of zolpidem, an α1-subunit modulator, and THDOC, a δ subunit–specific agonist, on tonic currents in layers V and II/III neurons. The δ subunit–containing GABAA receptors have been shown to mediate tonic inhibition in DGGCs (Stell et al. 2003), and its expression was also detected in neocortical neurons (Fig. 3). However, under our recording conditions, an increase in tonic conductance was not produced by THDOC at the concentration that could augment tonic currents in the dentate gyrus (Fig. 4C). Thus, the contribution, if any, of the δ subunit to the tonic current in the neocortex might not be significant. The zolpidem-sensitive GABAA receptor–mediated tonic currents have also been shown in hippocampal interneurons (Semyanov et al. 2003). In contrast to THDOC, zolpidem increased tonic currents in layers II/III and V neurons. Although the α1-subunit mRNA expression levels were higher in layer V than in layer II/III neurons, tonic conductance by zolpidem was the same in both layers. Considering that L-655708 was not effective and expression of the α5 subunit was almost negligible in layer II/III neurons, the tonic GABAA current in layer II/III neurons was attributable to the α1 subunit, whereas in layer V neurons, it was attributable to both the α1 and α5 subunits. Because the layer V neurons had significantly larger tonic currents than any other layer, the α5 subunit may be responsible for the unique tonic inhibition observed in layer V neurons. The α5 GABAA receptor subunit is located extrasynaptically at the base of the spines and on the adjacent shaft of the dendrites of the CA1 hippocampal neurons (Brunig et al. 2002), implying an involvement in the modulation of dendritic excitability and the efficacy of excitatory inputs. Thus, the α5 GABAA receptor subunit may play a critical role in rapidly and massively counterbalancing ongoing excitation (Caraiscos et al. 2004).
In conclusion, neocortical tonic inhibition is molecularly and pharmacologically distinct from tonic inhibition in the dentate gyrus and cerebellum. Neocortical tonic inhibition may increase resting membrane conductance and therefore may play an important role in preventing the development of hyperexcitability, modulating excitatory synaptic events, and controlling the rate and patterns of the spike discharge. Particularly, tonic inhibition in layer V neurons is unique in terms of amplitude possibly due to α5-subunit expression, which might thus be physiologically important in the modulation of cortical output.
This work was supported by Grant-in-Aid for Scientific Research #15590207 (JY), #16390058 (AF) from the Japan Society for the Promotion of Science and by the Research Grant (16A-3) for Nervous and Mental Disorders from the Ministry of Health, Labor, and Welfare (AF). Conflict of interest: None declared.