This Article Abstract Full Text (PDF) All Versions of this Article: 147/8/3746    most recent Author Manuscript (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Park, J. B. Articles by Stern, J. E. Search for Related Content PubMed PubMed Citation Articles by Park, J. B. Articles by Stern, J. E.

Characterization of a Novel Tonic -Aminobutyric Acid_A Receptor-Mediated Inhibition in Magnocellular Neurosecretory Neurons and Its Modulation by Glia

Department of Psychiatry, Genome Research Institute, University of Cincinnati, Cincinnati, Ohio 45237

Address all correspondence and requests for reprints to: Javier E. Stern, Department of Psychiatry, University of Cincinnati, GRI-A Room 241, 2170 East Galbraith Road, Cincinnati, Ohio 45237. E-mail: javier.stern{at}uc.edu.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
In addition to mediating conventional quantal synaptic transmission (also known as phasic inhibition), -aminobutyric acid_A (GABA_A) receptors have been recently shown to underlie a slower, persistent form of inhibition (tonic inhibition). Using patch-clamp electrophysiology and immunohistochemistry, we addressed here whether a GABA_A receptor-mediated tonic inhibition is present in supraoptic nucleus (SON) neurosecretory neurons; identified key modulatory mechanisms, including the role of glia; and determined its functional role in controlling SON neuronal excitability. Besides blocking GABA_A-mediated inhibitory postsynaptic currents, the GABA_A receptor blockers bicuculline and picrotoxin caused an outward shift in the holding current (I_tonic), both in oxytocin and vasopressin neurons. Conversely, the high-affinity antagonist gabazine selectively blocked inhibitory postsynaptic currents. Under basal conditions, I_tonic was independent on the degree of synaptic activity but was strongly modulated by the activity GABA transporters (GATs), mostly the GAT3 isoform, found here to be localized in SON glial cells/processes. Extracellular activation of GABAergic afferents evoked a small gabazine-insensitive, bicuculline-sensitive current, which was enhanced by GAT blockade. These results suggest that I_tonic may be activated by spillover of GABA during conditions of strong and/or synchronous synaptic activity. Blockade of I_tonic increased input resistance, induced membrane depolarization and firing activity, and enhanced the input-output function of SON neurons. In summary, our results indicate that GABA_A receptors, possibly of different molecular configuration and subcellular distribution, mediate synaptic and tonic inhibition in SON neurons. The latter inhibitory modality plays a major role in modulating SON neuronal excitability, and its efficacy is modulated by the activity of glial GATs.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
-AMINOBUTYRIC ACID (GABA) is the major inhibitory neurotransmitter in the mammalian nervous system. Main GABA inhibitory actions are mediated by activation of the ubiquitous ionotropic GABA_A receptor, a chloride ion channel that belongs to the superfamily of pentameric ligand-gate ion channels (1). GABA_A receptors are known to be essential for normal brain functioning and are involved in various pathological processes, including epilepsy, anxiety, and hypertension.

Until recently GABA_A receptors underlying physiological functions were thought to be restricted to the mediation of a transient (rapid) type inhibition, resulting from the brief exposure of synaptic receptors to high concentrations of GABA. This spatially and temporally restricted inhibitory modality (also termed phasic) is thought to be critical for information processing and timing-based signaling (2). More recently, however, GABA_A receptors were found to mediate a much slower form of inhibition, which is temporally and spatially dissociated from synaptically released GABA. This modality, also termed tonic inhibition, results from the persistent activation of GABA_A receptors, often located remotely from synapses (3, 4).

GABA_A receptor-mediated tonic inhibition has been recently described in major central nervous system (CNS) areas, including the cerebellum (5), hippocampus (6), and cortex (7, 8), in which it has been shown to play a critical role in modulating network excitability. Whether GABA_A receptor-mediated tonic inhibition also plays an important role in information processing and the control of neuronal excitability within neuronal circuits involved in the maintenance of bodily homeostasis is at present unknown. To this end, we aimed here to determine whether GABA_A receptor-mediated tonic inhibition is present in the magnocellular neurosecretory system (MNS), known to play fundamental roles in reproductive and fluid balance homeostasis, both under physiological and pathological conditions.

The MNS is comprised of oxytocin (OT) and vasopressin (VP) neurons, located in the hypothalamic supraoptic (SON) and paraventricular nuclei. These neurons send projections to the neurohypophysis (9), from which OT and VP peptides are secreted into the blood stream, a process known to be tightly locked to the electrical activity of VP and OT neurons (10, 11). A large bulk of information indicates that GABA, through activation of GABA_A receptors, is a major neurotransmitter modulating SON neuronal excitability (12, 13, 14). Furthermore, GABA_A receptor-mediated inhibition in SON neurons undergoes robust structural, molecular, and functional plasticity in response to physiological challenges (15, 16, 17). Despite the wealth of information available on GABA_A receptor structure/function in the MNS, no information is available thus far on the expression and functional roles of GABA_A receptor-mediated tonic inhibition. Here we provide novel evidence supporting the presence of this inhibitory modality in SON neurons. We evaluated the relative importance of tonic vs. synaptically driven GABA_A receptor-mediated inhibition and addressed potential interactions between these two inhibitory modalities and their relative effects on input-output function and firing discharge of SON OT and VP neurons.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Experimental animals
Male Wistar rats (180–220 g) were purchased from Harlan Laboratories (Indianapolis, IN). Rats were housed in a 12-h light, 12-h dark schedule and allowed free access to food and water. All animal experimentation described in these studies were conducted in accord with accepted standards of humane animal care and adhere to the policy of the University of Cincinnati regarding the use and care of animals.

Preparation of hypothalamic slices
Patch-clamp recordings were obtained in acutely prepared coronal hypothalamic slices as previously described (18). Briefly, rats were anesthetized with pentobarbital sodium (50 mg/kg ip) and perfused through the heart with cold artificial cerebrospinal solution (ACSF) in which NaCl was replaced by an equiosmolar amount of sucrose, a procedure known to improve the viability of neurons in adult brain slices (19). Slices containing the hypothalamus were cut (200 µm) using a vibroslicer (Leica VT 100s; Leica, Bensheim, Germany) in ice-cold ACSF and placed in a holding chamber containing standard oxygenated ACSF until used. The standard ACSF consists of (in millimoles): NaCl 126; KCl 2.5; MgSO₄ 1; NaHCO₃ 26; NaH₂PO₄ 1.25; glucose 20; ascorbic acid 0.4; CaCl₂ 1; pyruvic acid 2; pH was 7.3–7.4 (290–310 mOsm/liter). The medium was saturated with 95% O₂-5% CO₂. A single slice was transferred to the recording chamber and perfused with oxygenated ACSF at a approximately 2 ml/min flow. Experiments were conducted at room temperature (22–24 C), unless otherwise indicated.

Electrophysiology and data analysis
Electrophysiological recordings from SON neurons were made under visual control using a fixed-stage upright microscope equipped with infrared differential interference contrast video microscopy. All recordings were obtained using a Multiclamp 700A amplifier (Axon Instruments, Foster City, CA), under the voltage- or current-clamp mode. Current and voltage output were filtered at 2 kHz and digitized at 16-bit resolution (Digidata 1200; Axon Instruments) in conjunction with pClamp 9 software. Data were digitized at 10 kHz and transferred to a disk. For voltage-clamp experiments, patch pipettes (borosilicate glass, 3–7 M) were filled with a high Cl^–-containing solution, which facilitated the detection of small GABA_A receptor-mediated synaptic events (in millimoles): 140 KCl, 10 HEPES, 0.9 Mg²⁺ATP, 20 phosphocreatine (Na⁺), 0.3 Na⁺GTP, and 10 EGTA (pH 7.3) (295 mOsm/liter). For current-clamp experiments, a solution with a more physiological concentration of Cl^– was used (in millimoles): 140 K-gluconate, 10 KCl, 10 HEPES, 0.9 MgATP, 20 Na⁺, 0.3 NaGTP, and 0.2 EGTA (pH 7.3) (295 mOsm/liter). The series resistance was monitored throughout the experiments. Mean values were 10.9 ± 1.7 and 13.5 ± 1.1 M at the beginning and end of the experiments, respectively.

Synaptic (phasic) GABA_A receptor-mediated inhibition
Spontaneous (s) inhibitory postsynaptic currents (IPSCs) were recorded at a holding potential of –70 mV. Detection and analysis of phasic GABAergic synaptic activity was done using Minianalysis software (Synaptosoft, Decatur, GA), as previously described (20). Briefly, a detection threshold was set at –30 pA and 150 pAmsec for IPSC peak and area, respectively. This enabled the extraction of GABA_A-mediated IPSCs without possible contamination of glutamate-mediated excitatory postsynaptic currents (EPSCs) and/or background noise. The amplitude and area range of EPSCs was below the used detection threshold (20.43 ± 0.81 pA and 107.40 ± 10.07 pA.msec, respectively) and were blocked by glutamate receptor antagonists, as indicated below. From extracted IPSCs, the rise time, peak amplitude, and decay time constants were calculated. The mean synaptic current (I_synaptic) was calculated by multiplying the charge transfer of the averaged IPSC (Q, the integrated area under sIPSC) by the sIPSC frequency, as previously described (6). To evoke GABA IPSCs, extracellular electrical stimulation (100–350 µA, 0.1 msec, 10 pulses at 20 Hz) was applied to the region dorsolateral to the SON using a bipolar electrode made from tungsten wires (tip diameter 1–2 µm). These experiments were done in the presence of the glutamate 2-amino-3-hydroxy-5-methyl-4-isoxazol propionic acid (AMPA) and N-methyl-D-aspartate (NMDA) receptor antagonists, 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) (10 µM) and 2-amino-5-phosphonovaleric acid (APV) (100 µM), respectively. The properties of the evoked IPSCs were calculated as described above for sIPSC. The magnitude of the envelope inward current resulting from the summation of evoked IPSCs was calculated by integrating the area under the evoked curve.

Tonic GABA_A receptor-mediated inhibition
To determine whether a GABA_A receptor-mediated tonic current (I_tonic) is present in SON neurons, we assessed changes in holding current (I_holding) and baseline noise level (RMS) noise levels after bath application of the GABA_A receptor antagonists/blockers bicuculline (BIC, 20 µM), gabazine (GBZ, 1 µM), and/or picrotoxin (PIC, 100–300 µM) (5, 7, 21). The mean I_tonic amplitude was calculated as the difference in the holding current measured at three different 50-msec data segments obtained before and after a 5-min application of the GABA_A receptor blockers. RMS noise was measured using Minianalysis software (Synaptosoft, Decatur, GA), in segments of traces lacking PSCs. The charge transfer (Q_tonic) associated with I_tonic current was calculated using the equation, Q_tonic = (I_con – I_drug) x T, where I_con and I_drug are the holding current (picoamperes) during control and drug conditions, respectively, and T is recorded time. To estimate I_tonic density, its measured amplitude was normalized to the cell membrane capacitance, obtained by integrating the area under the transient capacitive phase of a 5-mV depolarizing step pulse in the voltage-clamp mode. When noted, I_tonic was also measured in ACSF containing the Na⁺ channel blocker tetrodotoxin (TTX, 0.5 µM) or the K⁺ channel blocker 4-aminopyridine (4-AP, 10–100 µM). To estimate the contribution of GABA transporters to I_tonic recordings were also obtained in the presence of the GABA transporter blockers nipecotic acid (100 µM), NO 711 (10 µM), SKF 89976A (30 µM), and ß-alanine (100 µM).

Intrinsic membrane properties and action potential discharge
To study the effects of I_synaptic and I_tonic on cell input resistance, we measured the voltage deflection evoked by a 10-pA depolarizing step pulse in the current-clamp mode (low Cl^– internal solution) in control ACSF, GBZ, and GBZ+BIC. The presence of spikes at relatively low current stimulation levels (see Fig. 10) prevented us from estimating input resistance from the slope of a current-voltage relationship.

View larger version (17K): [in this window] [in a new window]   FIG. 10. Blockade of Itonic, but not Isynaptic, enhanced the I-O function in SON neurons. A, Representative examples of action potentials evoked by depolarizing pulses of increasing amplitude in control ACSF (CTL; left panel), GBZ (1 µM, middle panel), and GBZ + BIC (20 µM, right panel). Note the increased number of evoked spikes in the latter group at each current amplitude step. Vm was held at –55 mV in all conditions. B, Mean plot of the number of evoked spikes as a function of depolarizing current amplitude in control ACSF, GBZ, and GBZ + BIC. Blockade of Itonic, but not Isynaptic, significantly shifted the I-O relationship upward and increased the steepness of the relationship (see Results). *, P < 0.05, compared with control; #, P < 0.05, compared with GBZ.

Intracellular labeling and immunohistochemical identification of recorded cells
For intracellular labeling, biocytin (0.2%) was added to the internal solution and neurons were filled by diffusion. After recordings, slices were fixed in 4% paraformaldehyde and 0.2% picric acid dissolved in 0.15 M phosphate buffer (pH 7.3). Sections were then incubated in streptaviding-cy3 (Vector Laboratories, Burlingame, CA) to reveal the recorded neuron, followed by a conventional double-immunohistochemical procedure to label for OT and VP immunoreactivities, as previously described (18). Briefly, slices were incubated 3 d in a cocktail of primary antibodies containing a polyclonal rabbit anti-VP and a polyclonal guinea pig anti-OT (Bachem, Torrance CA; 1:50,000 and 1:200,000, respectively). The reaction was followed by 2 d incubation in the presence of fluorescently labeled secondary antibodies [donkey antirabbit fluorescein isothiocyanate (FITC)-labeled and donkey anti-guinea pig Cy3-labeled antibodies; Jackson ImmunoResearch Laboratories Inc., West Grove PA; 1:800 and 1:50, respectively]. All antibodies were diluted with PBS containing 0.5% Triton X-100. Fluorescent signals of each of the three channels were captured using a confocal microscope (Zeiss LSM 510; Carl Zeiss, Göttingen, Germany). Fluorescent signal cross-talk among the channels was avoided by setting image-acquisition parameters with individually labeled sections. Images were then superimposed to determine signal colocalization (Adobe Photoshop, Adobe Systems, Mountain View, CA).

Immunohistochemistry in thin hypothalamic sections obtained from fixed brains
Single- and double-immunohistochemical fluorescent reactions were used to study the expression of the GABA transporters GAT1 and GAT3 in the SON and their possible colocalization with GABAergic terminals (glutamic decarboxylase, GAD67 immunoreactive) and/or astroglial cell/processes [glial fibrillary acid protein (GFAP) immunoreactive]. For these studies, a group of rats (n = 4) were deeply anesthetized with sodium pentobarbital (100 mg/kg ip) and perfused transcardially with 0.01 M PBS (150 ml) followed by 4% paraformaldehyde (500 ml). Brains were postfixed in 4% paraformaldehyde for 4 h at 4 C. Fixed brains were cryoprotected at 4 C with 0.1 m PBS containing 30% sucrose for a minimum of 48 h. Sections (25 µm) were then cut using a cryostat and incubated in a solution of 0.01 M PBS with 0.01% Triton X-100 and 10% normal goat serum for 1 h.

For single-immunofluorescence reactions, sections were incubated for 48 h in the presence of a polyclonal rabbit anti GAT1 or GAT3 primary antibodies (Chemicon International, Temecula CA; 1:100 dilution). For double-immunofluorescence reactions, sections were incubated for 48 h in a mix of primary antibodies that included the polyclonal rabbit anti-GAT1 or GAT3 (as described above) in conjunction with either a monoclonal mouse anti-GFAP (Chemicon International; 1:10,000 dilution) or a monoclonal mouse anti GAD67 (Chemicon International; 1:5,000 dilution) antibodies. Incubation in primary antibodies was followed by a 4-h incubation in secondary antibodies (donkey antirabbit FITC labeled together with a donkey antimouse Cy5 labeled, 1:200 and 1:50 dilutions, respectively). All antibodies were diluted with PBS containing 0.1% Triton X-100. All secondary antibodies were obtained from Jackson ImmunoResearch Laboratories). Control experiments were performed by omitting primary or secondary antibodies.

Histological sections were examined with a Zeiss LSM 510 confocal microscope system. Confocal images of six consecutive optical focal planes were obtained (1 µm thick each), and a projection image of the focal planes was used for display purposes. The argon-krypton and HeNe lasers were used to excite the FITC (488 nm) and Cy5 fluorophores (633 nm). Fluorescent signal cross-talk between the two channels was avoided by setting image-acquisition parameters with individually labeled sections. To determine colocalization of various fluorescent markers, confocal images obtained from the different fluorophores were merged. A positive colocalization was considered by the appearance of yellow (red + green) profiles in the merged image. Furthermore, an identifiable structure needed to be clearly discernible in each image before merging. Figures were composed using Adobe Photoshop (Adobe Systems).

Statistical analysis
Numerical data are presented as means ± SEM. Paired Student’s t test was used to compare the effects of a drug treatment on synaptic and/or tonic currents. Unpaired Student’s t tests were used when comparing results obtained from OT and VP neurons. ANOVA repeated measures (ANOVA-RM), followed by Tukey’s post hoc tests, as noted throughout the text.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Synaptic and tonic GABA_A-mediated inhibitory modalities are present in SON neurons
Electrophysiological recordings were obtained from a total of 139 SON neurons. As previously described (22), SON neurons were found to be under a tonic barrage of GABA_A receptor-mediated IPSCs. GABA_A IPSCs occurred at a mean frequency of 2.06 ± 0.30 Hz and were blocked by the GABA_A receptor antagonists BIC (20 µM), GBZ (1 µM), or PIC (100–300 µM) (Fig. 1).

View larger version (52K): [in this window] [in a new window]   FIG. 1. Antagonist sensitivity of GABAA receptor-mediated Itonic and Isynaptic currents in SON neurons. A, Representative example showing that in addition to blocking fast IPSCs (large arrow), BIC induced an outward shift in Iholding and reduced RMS (A1). Mean changes in Iholding and RMS induced by BIC are summarized in A2 (n = 21). B, Representative example showing similar effects evoked by the Cl– channel blocker PIC (B1). Mean changes in Iholding and RMS induced by PIC are summarized in B2 (n = 6). C, The high-affinity GABAA receptor antagonist GBZ blocked fast IPSCs but failed to induce changes in Iholding or RMS (C1). An outward shift in Iholding and a reduction in RMS were still evoked by further addition of BIC (C2) (same cell as in C1). *, P < 0.05, compared with their respective controls. Small arrows point to EPSCs observed in the presence of GABAA receptor blockers.

Differently from BIC, low micromolar concentration of GBZ (1 µM), a high-affinity GABA_A receptor competitive antagonist, selectively blocked synaptic activity, without inducing a significant shift in I_holding (control: –46.1 ± 3.5 pA. GBZ: –49.6 ± 4.7 pA) or the baseline RMS noise (control: 4.1 ± 0.2, GBZ: 3.9 ± 0.1) (n = 9, P > 0.1 in both cases; see also example in Fig. 1C). Similar results were observed at higher GBZ concentrations (30–300 µM, I_holding: control, –58.4 ± 9.2; GBZ, –57.1 ± 8.8; RMS noise: control, 4.2 ± 0.5, GBZ, 3.8 ± 0.3, n = 6, P > 0.5 in both cases). When BIC was applied in the presence of GBZ, the outward shift in holding current was still observed, although its magnitude was smaller than that evoked in control ACSF (5.98 ± 1.15 pA), suggesting a competitive interaction between the two GABA_A blockers (23). Thus, the consecutive use of a low micromolar concentration of GBZ and BIC enables the pharmacological dissection between synaptic and tonic GABA_A-mediated components in SON neurons.

In a subset of experiments (n = 8), recordings were performed in the presence of the AMPA and NMDA glutamate receptor blockers CNQX (10 µM) and AP-5 (100 µM). The amplitude of I_tonic under this condition was not significantly different from that recorded in the absence of these antagonists (control: 20.20 ± 2.54 pA; CNQX+APV: 22.20 ± 3.49 pA, P > 0.05). Based on this finding and the fact that glutamate currents did not interfere with the detection of the relatively larger GABA currents, further experiments were done without blocking glutamate receptors.

GABA_A receptor-mediated tonic inhibition was observed in all recorded neurons. However, to determine whether the magnitude of I_tonic varied between SON cell types, in a few instances, SON neurons were immunohistochemically identified as either OT (n = 6) or VP (n = 6), and the amplitude and density of I_tonic were compared between the two cell types. As summarized in Fig. 2, no significant differences were observed in any of these parameters between OT and VP neurons (P > 0.13).

View larger version (40K): [in this window] [in a new window]   FIG. 2. GABAA receptor-mediated tonic inhibition is present in both VP and OT SON neurons. A, Examples of GABAA Itonic in immunoidentified VP (A1–A3) and OT (A4–A6) neurons. The recorded neurons were visualized with Cy3-conjugated streptavidin (arrowheads in A1 and A4) and were positively labeled with VP immunoreactivity, visualized with FITC-conjugated secondary antibody (A2) or OT immunoreactivity, and visualized with Cy5-conjugated secondary antibody (A5). Both neurons displayed negative immunoreactivity to the complementary peptide (results not shown). An outward shift in Iholding was evoked by BIC (20 µM, thick line in A3 and A6) in both neurons. B, Summary data showing no significant differences in the Itonic amplitude (left panel) or density (right panel) between SON cell types (n = 6 in each group, P > 0.8 in both cases).

View larger version (32K): [in this window] [in a new window]   FIG. 3. GABAA receptor-mediated tonic inhibition is not dependent on activation of glycine receptors. A, Representative example showing the lack of changes in Iholding and RMS after blockade of glycine receptors with 1 µM strychnine (STR). Summary data are shown in lower bar graphs (n = 5). B, Representative example showing the lack of changes in Iholding and RMS after a hypoosmotic stimulus (10% decrease in ACSF osmolarity). The summary data are shown in the lower bar graphs (n = 7). CTL, Control. *, P < 0.05 vs. control.

View larger version (25K): [in this window] [in a new window]   FIG. 4. GABAA receptor-mediated tonic inhibition is not dependent on action potential-evoked GABA release. A, The lack of correlation between Itonic and IPSC frequency (left panel, R2 = 0.04) and Itonic and IPSC amplitude (right panel, R2 = 0.28) indicates that Itonic is not dependent on the degree of ongoing synaptic GABAergic activity. B, Representative examples (upper panels) of the effects of TTX and BIC on synaptic and tonic GABAA receptor-mediated currents. Note that whereas TTX decreased IPSC frequency, no changes in RMS or Iholding were observed. In the presence of TTX, BIC still induced an outward shift in Iholding and decreased RMS. The mean data are summarized in the lower bar graphs. CTL, Control. *, P < 0.05, compared with control (n = 6).

To determine whether a condition of enhanced ongoing synaptic activity affected I_tonic, recordings were obtained in the presence of the K⁺ channel blocker 4-AP (100 µM), known to increase neurotransmitter release of ongoing synaptic activity (28). Results are summarized in Fig. 5. Bath application of 4-AP significantly increased IPSC frequency and amplitude (by 400 and 65%, respectively). In addition, an inward shift in I_holding (16.42 ± 4.47 pA) along with an increase in RMS (from 3.88 ± 0.26 pA to 4.60 ± 0.45 pA) was observed in the presence of 4-AP (n = 11). If the 4-AP-induced shift in I_holding were due to activation of I_tonic, addition of BIC would be expected to block both the 4-AP-induced shift as well as the basal I_tonic, resulting in an outward shift of I_holding beyond baseline. Whereas an outward shift in I_holding was indeed evoked by BIC in the presence of 4-AP (15.6 ± 3.8, n = 11), this shift failed to go beyond baseline (Fig. 5A). In fact, the magnitude of the BIC-induced shift in the presence of 4-AP was not different from the one observed in control conditions (P > 0.4), suggesting that the BIC-induced shift in the presence of 4-AP was due only to blockade of basal I_tonic, without affecting the 4-AP-induced inward shift. This is also supported by the fact that the 4-AP-induced change in I_holding was still observed when slices were preincubated with BIC (13.94 ± 1.92, n = 4), Thus, these results suggest that a mechanism other than activation of GABA_A receptors mediated the 4-AP-induced increased conductance. This was not further explored in the present study.

View larger version (33K): [in this window] [in a new window]   FIG. 5. GABAA receptor-mediated tonic inhibition is not affected by increasing ongoing synaptic GABAA receptor-mediated activity. A, Representative traces from an SON neuron showing that the K+ channel blocker 4-AP (100 µM) induced a robust increase in sIPSCs frequency. Note that an inward shift in Iholding was also evident. Addition of BIC blocked sIPSCs and diminished Iholding. Note, however, that an outward shift in Iholding beyond control conditions was not evoked by BIC. B, Representative example showing that addition of 4-AP in the presence of BIC was still able to evoke an inward shift in Iholding. C, Summary data depicting the 4-AP-induced changes in sIPSC (n = 11). D, Summary data depicting changes in Iholding in the presence of 4-AP (100 µM) and the GABAA antagonists BIC (20 µM) and GBZ (1 µM). Note that the inward shift in Iholding induced by 4-AP was not blocked GBZ and was still induced in the presence of BIC (cross-hatched bar). Whereas BIC application in the presence of 4-AP diminished Iholding, an actual outward shift beyond control condition was not observed. In fact, the percent change in Iholding induced by BIC in the presence of 4-AP (n = 5) was in the same range as that observed in control ACSF (see Results). CTL, Control. *, P < 0.05, compared with their respective controls; #, P < 0.05, compared with 4-AP+BIC.

View larger version (29K): [in this window] [in a new window]   FIG. 6. Synchronous activation of SON GABAergic terminals activates Itonic. A, Extracellular stimulation (150 µA, 10 pulses at 20 Hz) in the dorsolateral SON in control ACSF (black trace) evoked IPSCs (arrows) that successively summed up, inducing a large envelope inward current. In the presence of GBZ (1 µM, gray trace), the magnitude of the evoked envelope was robust, though not completely blocked. Further addition of BIC (20 µM) completely blocked the evoked response. The inset shows the evoke response in the presence of GBZ at an expanded scale. B, Summary data of the area of the evoked response in control ACSF, GBZ, and GBZ+BIC in the presence (filled bars, n = 8) and absence (empty bars, n = 5) of the GAT blocker NA (100 µM). Note the larger GBZ-insensitive component in the presence of NA. *, P < 0.05 when compared with their respective controls. #, P < 0.05 when compared with their respective group without NA. C, The second IPSCs evoked in the train shown in A (black: control ACSF, gray: GBZ) are displayed at an expanded time scale. Note the slower IPSC rising kinetics in the presence of GBZ. D, Summary data of the rising slopes of evoked IPSCs in control ACSF in the presence of GBZ (n = 7), *, P < 0.05, compared with its respective control.

Is I_tonic dependent on the activity of neuronal and/or glial GATs?
In other neuronal populations, including hippocampal interneurons and cerebellar granule cells (8, 31), I_tonic was found to be influenced by ambient GABA concentration, known to be tightly controlled by the activity of GATs (4, 32). To determine whether this was the case in SON neurons and to shed light into the possible GAT isoforms and cell types involved, we combined pharmacological and immunohistochemical approaches. Results are summarized in Table 1 and Fig. 7. Bath application of the nonselective GAT blocker NA (100 µM) caused a large inward shift in I_holding and an increase in RMS, effects that were blocked by BIC. These results suggest that blockade of GATs leads to rapid accumulation of GABA in the extracellular space and further activation of extrasynaptic GABA_A receptors. Similar effects were observed when the potent and relative specific GAT3 blocker ß-alanine (100 µM) (33, 34, 35, 36, 37) was used. On the other hand, significantly smaller changes, or no changes in I_holding or RMS, were observed when the specific GAT1 blockers NO711 (10 µM) or SKF 89976A (30 µM) (38) were used. These results suggest that the activity of GABA transporters, in particular the GAT3 isoform, strongly influences the magnitude of I_tonic in SON neurons.

View larger version (32K): [in this window] [in a new window]   FIG. 7. Effect of GABA transporter blockade on Itonic. A, Bath application of the nonselective GAT blocker NA (100 µM) induced a strong inward shift in Iholding, an effect that was blocked by BIC (20 µM). B, Mean data summarizing the effects of NA (n = 5), the GAT1 selective blockers NO711 (n = 4) and SKF 89976A (SKF, n = 5), and the GAT3 selective blocker ß-alanine (ß-ala, n = 5). Note that whereas both NA and ß-Ala induced a robust and significant inward shift in Iholding (Idrug), much smaller or statistically insignificant effects were evoked by the GAT1 blockers. *, P < 0.05 and **, P < 0.01 when compared with their respective controls; #, P < 0.05 and ##, P < 0.01 when compared with NA.

Because the temperature at which these experiments were performed was lower than physiological temperature, it could be possible that I_tonic and its associated RMS represented a recording artifact or resulted from a less effective clearance of neurotransmitter. To rule out this possibility, we explored whether I_tonic and the effects of GAT blockade persisted at more physiological temperatures. Increasing the recording temperature from approximately 23 C to approximately 35 C (n = 5) did not affect the magnitude of I_tonic (control: 20.20 ± 2.54 pA; high temperature: 19.8 ± 2.7 pA, P > 0.5) or RMS (control: 4.19 ± 0.27; high temperature: 4.03 ± 0.68, P > 0.5). Similarly, the nonselective GAT blocker NA (100 µM, n = 5) still induced a significant shift in I_holding and RMS, effects that were not statistically different from those evoked at room temperature (I_holding shift: control: –122.52 ± 20.43 pA; high temperature: –85.6 ± 11.7 pA, n = 5, P > 0.5; RMS change: control: 7.53 ± 1.13; high temperature: 6.42 ± 1.35, P > 0.5). Our results indicate that I_tonic recorded at room temperature is not the consequence of impaired GABA uptake, results that are in agreement with recent findings obtained from cortical interneurons (8) and cerebellar granule cells (40).

Immunoreactivity for GAT1 and GAT3 isoforms in the SON
GAT1 and GAT3 have been shown to be predominantly but not exclusively located in neurons and astrocytes, respectively (41, 42, 43). Consistent with the above pharmacological experiments, we found a robust and weak GAT3 and GAT1 immunoreactivities in the SON, respectively (Fig. 8). Robust GAT3ir was observed in the neuropil, mostly located in astroglial-like processes surrounding SON somata. In fact, double-immunofluorescence studies showed a high degree of colocalization between GAT3 and GFAP immunoreactivities (Fig. 8, A and B). On the other hand, no colocalization between GAT3 and the GABAergic marker GAD67 was observed (results not shown).

View larger version (58K): [in this window] [in a new window]   FIG. 8. GAT3 and GAT1 immunoreactivities colocalize with GFAP and GAD67 immunoreactivities in the SON and perinuclear zone, respectively. A, Confocal photomicrograph displaying GAT3ir (green) and GFAPir (red) in the SON. Note the high degree of colocalization (yellow). B, GAT3ir (B1) and GFAPir (B2) of a representative SON neuron from A (asterisk) is shown at higher magnification to better depict partial colocalization of the two signals (B3). C, Confocal photomicrograph displaying GAT1ir (green) and GAD67ir (red) in the SON. Note the very low abundance of GAT1ir within the SON proper. A higher density of both GAT1 and GAD67 immunoreactivities is observed in the perinuclear zone (area between arrows), in which both signals were found to partially colocalize (yellow). Vertical and horizontal arrows in A point to dorsal and medial aspects of the SON, respectively. Scale bars, A and B, 15 µm; C,10 µm.

Relative contribution of synaptic and tonic GABA_A receptor-mediated inhibition in SON neurons
To determine the relative contribution of synaptic and tonic GABA_A receptor-mediated inhibition in SON neurons, we estimated and compared the mean current and charge transfer carried by these two inhibitory modalities (see Materials and Methods). Despite their large peak amplitude and due to their transient nature and rapid kinetics, the average IPSC had an overall small charge transfer (2.99 ± 0.28 pC, n = 13). At a frequency of approximately 2 Hz, IPSC charge transfer resulted in a mean synaptic current of 5.62 ± 0.62 pA, which was significantly smaller that the mean I_tonic (20.20 ± 2.54 pA, P < 0.01). Similar results were observed in identified VP and OT neurons (the mean I_tonic was 3.0 times and 3.8 times larger than I_synaptic, respectively, P < 0.05, n = 6 in both cases). Thus, our results indicate that most (80%) of the total GABA_A receptor-mediated charge transfer (synaptic + tonic) in SON neurons is carried by the tonic modality.

I_tonic restrains firing activity in SON neurons
To determine the relative contribution of I_synaptic and I_tonic on SON Vm and firing activity, experiments were performed in the current-clamp mode. Recorded neurons were either spontaneously active (four of 11), or firing activity was evoked by small DC current injection. The sequential application of GBZ and GBZ+BIC was used to differentiate between the roles of I_synaptic and I_tonic in modulating these parameters (see Materials and Methods). Representative examples are shown in Fig. 9, A and B, and mean values are summarized in Fig. 9C. Bath application of GBZ alone, used at a concentration of 1 µM (n = 11) or 30 µM (n = 4), failed to significantly change Vm (+2 mV) or firing discharge (+0.1 Hz) (P > 0.5 in both cases). On the other hand, additional application of BIC significantly depolarized SON cells by approximately 5 mV (P < 0.01, n = 11) and increased firing rate by approximately 5 times (P < 0.05, n = 7, see Fig. 9C). In three of four separate cases in which drugs were applied to initially silent SON neurons, coapplication of GBZ+BIC, but not GBZ alone, evoked firing activity in the recorded cells (see example in Fig. 9A). Similar results were obtained when the Cl^– channel blocker PIC was used (n = 5, Fig. 9, B and C).

View larger version (38K): [in this window] [in a new window]   FIG. 9. Blockade of Itonic but not Isynaptic induces membrane depolarization and firing activity in SON neurons. A, Representative example showing that blockade of Isynaptic (GBZ, 1 µM) failed to induce membrane depolarization and/or firing activity in this silent SON neuron. On the other hand, subsequent blockade of Itonic (GBZ+BIC, 20 µM) efficiently induced firing activity in the same neuron. B, Representative example showing PIC-induced increased firing activity in another SON neuron. C, Summary data depicting changes in Vm and firing discharge induced by GBZ (n = 7), GBZ + BIC (n = 7), and PIC (n = 5). *, P < 0.05, compared with their respective controls. Neurons shown in A and B were held at –55 and –50 mV, respectively.

Further supporting a functionally relevant role of GAT activity on I_tonic and SON excitability, we found that both NA (n = 5) or ß-alanine (n = 6) (100 µM each) decreased the firing activity of SON neurons by 63.1 ± 14.3 and 57.4 ± 8.1%, respectively (P < 0.05 and P < 0.01 when compared with baseline, respectively, Fig. 11). These effects were blocked by preincubation of sliced in BIC (not shown, n = 4 and 3, for NA and ß-alanine, respectively).

View larger version (36K): [in this window] [in a new window]   FIG. 11. Blockade of GATs decreased the firing activity of SON neurons. A, Plot of mean firing frequency over time (bin = 10 sec) in a representative SON neuron. Bath application of the nonselective GAT blocker NA transiently diminished the firing activity in the recorded cell. B, Representative traces obtained from segments a–c in A are depicted. The shown neuron was initially held at approximately –50 mV. C, Mean data summarizing the decrease firing rate in SON neurons evoked by NA (n = 5) and ß-alanine (ß-Ala) (n = 6). CTL, Control. *, P < 0.05 vs. control.

Role of excitatory glutamate transmission in mediating tonic GABA_A receptor-mediated excitation in SON neurons
To determine whether increased SON neuronal excitability after blockade of I_tonic required an intact ionotropic receptor-mediated excitatory glutamate synaptic input, experiments were repeated in the presence of the AMPA and NMDA glutamate receptor antagonists CNQX (10 µM) and APV (100 µM), respectively. Recorded neurons were either spontaneously active (three of five) or firing activity was evoked by small DC current injection. Results are summarized in Fig. 12. Application of ionotropic glutamate receptor antagonists completely blocked SON firing discharge. Thus, firing activity was evoked again using DC current injection before the effects of GABA_A receptor blockers were tested. In the presence of glutamate receptor antagonists, GBZ alone (n = 5) did not change SON firing discharge (+13.7 ± 5.0%, P > 0.05). On the other hand, a significantly larger increase in firing discharge (358.6 ± 163.3%, P < 0.05) was induced when GBZ was coapplied with BIC. Even though the BIC-induced effect in the presence of ionotropic glutamate antagonists was smaller than that evoked in control conditions (479.3 ± 175.0%), this difference did not reach statistical significance (P = 0.5) Altogether these results suggest that blockade of I_tonic per se is sufficient to increase neuronal excitability and firing discharge of SON neurons, even in the absence of an active glutamatergic input.

View larger version (32K): [in this window] [in a new window]   FIG. 12. Blockade of Itonic increases firing activity in SON neurons in the presence of ionotropic glutamate receptor antagonists. A, Plot of mean firing frequency over time (bin = 10 sec) in a representative SON neuron. The neuron was initially silent (a) (Vm –60 mV), and firing activity was induced through DC current injection (b). Bath application of the NMDA and AMPA receptor antagonists AP5 (100 µM) and CNQX (10 µM), respectively, blocked firing activity in the recorded cell (c). DC current injection was applied to restore firing activity (d). No changes in firing activity were observed in the presence of GBZ (1 µM, e). On the other hand, further addition of BIC (20 µM) resulted in an increased firing discharge. B, Representative traces obtained from segments a–f in A are depicted. C, Summary data (n = 5) showing that in the presence of ionotropic glutamate receptor antagonists, blockade of Itonic, but not Isynaptic, increased firing activity in SON neurons.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In recent years, it has become increasingly clear that in addition to mediating conventional quantal synaptic transmission, GABA_A receptors also underlie a persistent tonic form of inhibition. These multiple modalities of GABA_A receptor-mediated inhibition have been demonstrated in various brain regions, including the cerebellum (5), cortex (7), and hippocampus (6, 8). Accumulating evidence indicates that GABA_A receptors that mediate both types of inhibition have distinct biophysical and pharmacological properties as well as different subcellular distributions (see Refs. 3, 4 for recent reviews on this subject). In this sense, there is general consensus that GABA_A receptors mediating synaptic inhibition are clustered at postsynaptic sites, have a relatively low affinity for GABA, and are activated by brief exposure to high concentration of neurotransmitter (49). On the other hand, GABA_A receptors mediating tonic inhibition appear to be located at peri- and/or extrasynaptic sites, display high affinity for GABA, and are activated by low ambient concentration of transmitter in the extracellular space (21). Importantly, these two types of GABA_A receptors are differentially modulated by clinically relevant compounds, including benzodiazepines (23), neurosteroids (50), and alcohol (51).

GABA_A receptor-mediated tonic inhibition in SON neurons
The presence of a GABA_A receptor-mediated tonic conductance in SON neurons is supported by our findings showing that two GABA_A channel blockers, BIC and PIC, induced an outward shift in the holding current. This effect was accompanied by a decrease in background noise, consistent with block of stochastic ion channel openings (5, 40). Interestingly, whereas BIC and PIC blocked both synaptic and tonic GABA_A receptor-mediated inhibition, a low micromolar concentration of the competitive, high-affinity antagonist GBZ selectively blocked synaptic inhibition. In this sense, the efficacy of nonsaturating concentrations of a competitive antagonist with binding and off-rate properties such as GBZ is expected to be dependent on the affinity of the receptor for the endogenous ligand (52). Thus, and as previously shown in hippocampal neurons (53), the lack of sensitivity of I_tonic to GBZ in SON neurons suggests that GABA_A receptors mediating I_tonic in SON neurons have a higher affinity for GABA than those mediating I_synaptic.

A similar selective pharmacological sensitivity of I_synaptic to low concentrations of GBZ has been previously found in hippocampal neurons (8, 21, 23, 53). On the other hand, GBZ was reported to block both I_tonic and I_synaptic in cerebellar and dentate gyrus granule cells (6, 54, 55, 56). Because GABA_A receptor affinity for its ligand is known to be critically dependent on receptor subunit composition (57), differences in the molecular configuration of GABA_A receptors underlying tonic inhibition among these neurons (4) may account for the reported differences sensitivity fir GBZ blockade.

Similarly, differences between GABA_A receptors underlying synaptic and tonic inhibition could be explained, at least in part, by distinct receptor subunit composition. Association of -subunit with 1, 2 or 3, and ß2/3 is the predominant configuration of synaptic GABA_A receptors. On the other hand, extrasynaptic GABA_A receptors underlying tonic inhibition have been shown to contain the 5-, 6-, and/or -subunits (5, 40, 50, 58). Interestingly, recent work by Pirker et al. (59) demonstrated that in addition to expressing the 1-, 2-, ß2-, and -subunits (mostly localized to dendritic compartments), SON somata were found to be enriched in the 5-, ß1-, and -subunits, supporting the expression in SON neurons of GABA_A receptor subunits previously shown to mediate I_tonic in other neuronal types. Clearly, future studies are needed to determine the specific molecular configuration of GABA_A receptors underlying tonic inhibition in the SON.

Sources of GABA underlying tonic inhibition in SON neurons
The precise mechanisms that underlie and modulate tonic GABAergic inhibition are still not well understood. Both vesicular, e.g. summation of overlapping IPSCs (7, 60) or spillover of transmitter released from neighboring synapses (30), and nonvesicular mechanisms yet to be identified (31, 40) have been shown to act as sources of GABA-mediating tonic inhibition. Several lines of evidence from our studies indicate that under basal conditions, a nonvesicular source of GABA most likely mediates tonic inhibition in SON neurons. First, the selective block of synaptic activity by GBZ argues against the summation of unresolved events as a factor contributing to tonic inhibition. Second, tonic inhibition in SON neurons, at least under basal conditions, is not mediated or dependent on action potential-evoked GABA release. Furthermore, doubling the degree of synaptic activity by increasing neurotransmitter release probability with 4-AP (28) was not accompanied by an enhanced degree of GABA_A receptor-mediated tonic inhibition. Thus, under basal conditions, tonic GABA_A receptor-mediated inhibition in SON neurons appears to operate independently from synaptically released GABA. This in turn further supports the notion that I_tonic in this neuronal population results from the detection of low extracellular GABA concentrations by high-affinity, nonsensitizing receptors (21).

Whereas vesicular GABA does not seem to contribute to I_tonic under basal synaptic activity, our data suggest that strong and/or synchronous activation of GABAergic terminals results in spillover of GABA, capable of activating GABA_A receptors underlying I_tonic. This is supported by the presence of a GBZ-insensitive, BIC-sensitive component on the synaptically evoked response. As expected, this evoked component was enhanced by GABA transporter blockade, a condition known to facilitate GABA spillover and diffusion out of the cleft (29, 30).

Similar to synaptic IPSCs (GBZ sensitive), activation of GABA_A receptors underlying I_tonic (GBZ insensitive) during extracellular stimulation resulted in transient IPSC-like responses that summed to form an envelope response. In addition to being significantly smaller in amplitude, the IPSC-like events were also significantly slower in their rising (but not decaying) phase. Because the kinetics of activation of GABA_A receptors depend on agonist concentration (61, 62, 63), it is likely that the slower kinetics of the GBZ-insensitive IPSC-like events resulted from the lower GABA concentration reaching these receptors after release and spillover of GABA (30). Alternatively, the slower events could result from dendritic filtration of distal synaptic events. This is unlikely though because these synaptic events would also be expected to be blocked by GBZ and their decay kinetics to be affected by dendritic filtering. Interestingly, our data suggest that even a single stimulation pulse was sufficient to evoke a transient, GBZ-insensitive BIC-sensitive response. This is in agreement with work by Rossi and Hamann (30) showing that single evoked IPSCs in cerebellar granule cells display fast and slow-rising components, the latter mediated by activation of 6-containing extrasynaptic GABA_A receptors. It will be important in future studies to address the influence of various patters and degree of synaptic activation on spillover activation of I_tonic as well as its functional significance.

Does taurine contribute to I_tonic in SON neurons?
Recent studies have shown that endogenous release of the amino acid taurine from SON astroglia induced a Cl^–-mediated conductance, resulting in the inhibition of SON neurons and the concomitant VP release from neurohypophysial terminals (24, 25, 26, 27, 64). Release of taurine from astroglial cells is stimulated during hypoosmotic conditions (26, 27), a mechanism proposed to contribute to osmoregulation of VP release from SON neurons (65). Taurine is known to act as an efficient agonist on glycine receptors (26, 66), although at higher concentrations, e.g. with a much lower affinity and potency, it can also activate GABA_A receptors (26). Still, whereas taurine can potentially activate both receptors, the former have been shown to be the predominant mediators of taurine inhibitory effects in the SON. For example, at up to a 1-mM concentration, taurine induced a strychnine-sensitive, GBZ-insensitive current in SON neurons (26). In addition, taurine inhibition of K⁺-induced increase in intracellular calcium and VP release in neurohypophysial nerve terminals were blocked by strychnine but not GBZ (25). Finally, even at high concentrations, taurine hyperpolarized SON neurons in an explant preparation, an effect completely blocked by strychnine (14).

Interestingly, Song and Hatton (64) reported in a recent study a taurine-evoked increase in basal hormone release from the neurohypophysis, an effect that involved activation of both GABA_A and strychnine-insensitive glycine receptors in OT and VP neurons, respectively. In this study, however, a relatively high concentration of taurine (10 mM) was used.

Whereas taurine is highly concentrated in astroglial cells (up to a few tens of millimolar), the actual level found in the SON extracellular fluid under basal conditions is unknown. Previous estimates in other systems indicate this to be at most a few tens of micromolar (67, 68), a concentration that would activate glycine but not GABA_A receptors in the SON (26). In fact, activation of glycine receptors by basal levels of taurine is supported by a recent in vivo study showing that strychnine increased the ongoing firing discharge of VP neurons, an effect that was enhanced during hypoosmotic conditions (65).

Based on these data, we evaluated the potential contribution of taurine to the I_tonic described in this work. Our results showing a lack of an effect of strychnine on I_holding and RMS, as well as the lack of an increase in I_holding after a hypoosmotic stimulation, argue against a glycine-mediated, taurine contribution to I_tonic. Because the GABA_A receptor antagonists are unable to discriminate between the contribution of GABA and/or taurine as endogenous ligands, a role for a GABA_A-mediated taurine effect cannot be completely ruled out. However, due to the much higher affinity and potency of taurine for glycine rather than GABA_A receptors, it is highly unlikely that in the absence of a glycine receptor-mediated effect, taurine would evoke GABA_A receptor-mediated response.

The activity of glial GABA transporters efficiently modulates tonic inhibition
GATs are responsible for removing GABA from the extracellular space, thus playing a key role in controlling local ambient extracellular GABA concentration. In agreement with previous reports in other CNS neuronal types, (6, 8, 31, 40), we found that blockade of GAT activity increased the strength of I_tonic, inhibiting in turn the firing discharge of SON neurons. Based on the use of relatively specific pharmacological agents, we found these actions to be most likely mediated by the GAT3 but not the GAT1 transporter isoform. There is general consensus that these two isoforms are differentially distributed within specific cellular compartments in the CNS, with GAT1 and GAT3 predominantly, but not exclusively, located in neurons and astrocytes, respectively (see Refs. 41, 42, 43 for recent reviews). A key role for the GAT3 isoform in the SON is further supported by our immunohistochemical results showing robust and weak GAT3 and GAT1ir, respectively, in close apposition to SON somata. Moreover, our results showing colocalization of GAT3ir and GFAP-positive astrocytic processes further support a major role for SON astrocytes in the control of GABA_A I_tonic efficacy.

It is well established that SON synapses and neuronal elements are tightly enwrapped by thin glial processes (69, 70). Thus, through potent and efficient neurotransmitter uptake mechanisms, these processes are strategically positioned to influence the efficacy of synaptic transmission. This is supported by recent work showing that high-affinity glial glutamate transporters modulate glutamate synaptic efficacy and intersynaptic cross-talk within the SON (71, 72). Thus, our present results supporting a strong influence of glia transporters on the efficacy of tonic GABA_A inhibition, along with previous observations (71), underscore a pivotal role for SON glia in regulating the balance between excitatory and inhibitory signals modulating SON neuronal excitability. Because the degree of glial synaptic/neuronal coverage in the SON is known to dynamically change in an activity-dependent manner (73, 74), it will be important in future studies to determine the impact of such neuronal-glial structural remodeling on GABA_A I_tonic efficacy. Interestingly, a similar differential contribution of GAT1 and GAT3 to the control of GABA_A I_tonic was found in cerebellar granule cells (31, 40), which along with their synaptic inputs within the glomeruli are all ensheathed within glial coats. Thus, the structure of the local neuronal-synaptic-glial microenvironment is likely to influence the lifetime and concentration of GABA in the extracellular space and, in turn, its ability to activate extrasynaptic GABA_A receptors underlying I_tonic.

Finally, glial cells could also contribute to modulation of tonic inhibition through the active release of GABA (75). However, the fact that that GAT blockade within the SON always resulted in an increased magnitude of GABA_A I_tonic argues against this possibility, at least under basal conditions.

Functional role of GABA_A receptor-mediated tonic inhibition
Our findings indicate that I_tonic in SON neurons accounts for approximately 80% of the total GABA_A-receptor mediated current. Thus, this inhibitory modality is expected to have a major impact on SON neuronal excitability. Accordingly, we found that blockade of I_tonic increased SON input resistance, induced membrane depolarization, and increased firing discharge. In addition, the upward shift and increased slope in the I-O function after blockade of I_tonic indicates that this inhibitory modality induces a multiplicative gain change in this function (76, 77), having thus an important impact on SON neuronal sensitivity to incoming excitatory and/or inhibitory synaptic inputs. The fact that these effects were evoked by both BIC and PIC argues against the possibility that BIC-mediated changes in firing activity were due to potential effects of methiodide (the BIC derivative used in this study) on Ca²⁺-dependent K⁺ currents (78). Blockade of I_synaptic, on the other hand, failed to induce significant changes in any of these parameters. Thus, it is likely that these two inhibitory modalities contribute differently to information processing in SON neurons. Whereas I_synaptic likely mediates a transient and specific point-to-point inhibition, our data indicate that I_tonic is an important determinant of overall SON neuronal excitability. In addition, by having a direct impact on membrane input resistance, and thus membrane time constant, I_tonic may affect synaptic efficacy and integration in these neurons (77). Our results showing that the excitatory effect induced by blockade of I_tonic was still observed in the presence of ionotropic glutamate receptor antagonists suggest, nevertheless, that modulation of I_tonic per se, without necessarily increasing the efficacy of excitatory inputs, can directly affect intrinsic properties and firing output of SON neurons.

Finally, numerous studies reported that relevant neurotransmitters/modulators in the SON, such as nitric oxide (79), affect neuronal excitability by altering a synaptic GABA tone. Importantly, most of these studies were based on the use of BIC to block GABAergic activity. Because our results indicate that both synaptic and tonic inhibitory modalities are blocked by BIC, a contribution of a nonsynaptic tone (I_tonic) to the reported actions of these neuromodulators cannot be ruled out.

In summary, results from the present work indicate that GABA_A receptors, possibly of different molecular configuration and subcellular distribution, mediate two different modalities of inhibition in SON neurons: the conventional fast synaptic inhibition (also known as phasic mode) and a novel, sustained form of inhibition (tonic mode), which operates on a much slower time scale, is strongly influenced by the activity of glial GATs and plays a key role in modulating SON neuronal excitability and firing discharge.

	Footnotes

 
This work was supported by National Institutes of Health Grants RO1 HL68725 (to J.E.S.) and KRF-2004-013-E00001 (to J.B.P.).

Disclosure statement: J.B.P., S.S., and J.E.S. have nothing to disclose.

First Published Online May 4, 2006

Abbreviations: ACSF, Artificial cerebrospinal solution; AMPA, 2-amino-3-hydroxy-5-methyl-4-isoxazol propionic acid; ANOVA-RM, ANOVA repeated measures; 4-AP, 4-aminopyridine; APV, 2-amino-5-phosphonovaleric acid; BIC, bicuculline; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; CNS, central nervous system; FITC, fluorescein isothiocyanate; GABA, -aminobutyric acid; GAT, GABA transporter; GBZ, gabazine; GFAP, glial fibrillary acid protein; I_holding, holding current; I-O, input-output; IPSC, inhibitory postsynaptic current; I_synaptic, synaptic current; I_tonic, tonic current; MNS, magnocellular neurosecretory system; NA, nipecotic acid; Na⁺, phosphocreatine; NMDA, N-methyl-D-aspartate; OT, oxytocin; PIC, picrotoxin; RMS, baseline noise level; s, spontaneous; SON, supraoptic nucleus; TTX, tetrodotoxin; Vm, membrane potential; VP, vasopressin.

Received February 17, 2006.

Accepted for publication April 24, 2006.

	References

Top Abstract Introduction Materials and Methods Results Discussion