AT_1a influences GABAA-mediated inhibition through regulation of KCC2 expression

Abstract

The median preoptic nucleus (MnPO) is an integrative site involved in body fluid homeostasis, cardiovascular control, thermoregulation, and sleep homeostasis. Angiotensin II (ANG II), a neuropeptide shown to have excitatory effects on MnPO neurons, is of particular interest with regard to its role in body fluid homeostasis and cardiovascular control. The present study investigated the role of angiotensin type 1a (AT_1a) receptor activation on neuronal excitability in the MnPO. Male Sprague-Dawley rats were infused with an adeno-associated virus with an shRNA against the AT_1a receptor or a scrambled control. In vitro loose-patch voltage-clamp recordings of spontaneous action potential activity were made from labeled MnPO neurons in response to brief focal application of ANG II or the GABAA receptor agonist muscimol. Additionally, tissue punches from MnPO were taken to asses mRNA and protein expression. AT_1a receptor knockdown neurons were insensitive to ANG II and showed a marked reduction in GABAA-mediated inhibition. The reduction in GABAA-mediated inhibition was not associated with reductions in mRNA or protein expression of GABAA β-subunits. Knockdown of the AT_1a receptor was associated with a reduction in the potassium-chloride cotransporter KCC2 mRNA as well as a reduction in pS940 KCC2 protein. The impaired GABAA-mediated inhibition in AT_1a knockdown neurons was recovered by bath application of phospholipase C and protein kinase C activators. The following study indicates that AT_1a receptor activation mediates the excitability of MnPO neurons, in part, through the regulation of KCC2. The regulation of KCC2 influences the intracellular [Cl⁻] and the subsequent efficacy of GABAA-mediated currents.

INTRODUCTION

The median preoptic nucleus (MnPO) is located between two circumventricular organs, the subfornical organ (SFO) and the organum vasculosum of the lamina terminalis (OVLT), on the anterior wall of the third ventricle (30). It is involved in body fluid balance, central cardiovascular control, thermoregulation, and sleep homeostasis (4, 18, 30). The MnPO receives projections from the brain stem as well as from circumventricular organs and, on the basis of its neuroanatomy, is described as an integrative region (30). Projections from the circumventricular organs are traditionally thought to be especially important to the role of the MnPO in body fluid homeostasis and cardiovascular regulation (4, 30). It has been proposed that angiotensin II (ANG II) may act as a hormone at circumventricular organs and as a neurotransmitter in the central nervous system (14) although the existence of a brain renin angiotensin system has recently been challenged (40).

Functional studies suggest a role for angiotensin in the SFO-to-MnPO pathway. Activation of the SFO by direct injection of ANG II produces water intake that is blocked by the administration of the angiotensin antagonist, saralasin, in the MnPO (37). Similarly, electrical stimulation of the SFO produced excitation or inhibition of MnPO neurons, and the excitatory responses were blocked by iontophoresis of saralasin (38).

In vitro electrophysiological studies of MnPO neurons have shown that ANG II can produce a depolarization while reducing afterhyperpolarization and spike frequency adaptation (1). In a minority of MnPO cells, angiotensin decreases a leak potassium conductance and induces low-threshold calcium spikes, leading to rhythmic bursting (36). In vitro, electrical stimulation of the SFO produces rapid excitation and inhibition that are mediated by glutamate and γ-aminobutyric acid (GABA), respectively (24). Although they did not observe angiotensin-mediated responses resulting from SFO stimulation, the authors suggested that the peptide may produce delayed effects on MnPO neurons that were not readily apparent in their preparation (24). Recent studies suggest that endogenous angiotensin appears to be required to support the inhibitory effects of GABA on MnPO neurons (16). Inhibition of the AT_1a receptors with either saralasin or losartan was associated with a significant decrease in inhibitory currents, whereas exogenous angiotensin increased the amplitudes of muscimol currents. Although the mechanism underlying this relationship between angiotensin and GABA has not been determined, these observations are consistent with recent reports suggesting that G protein-coupled receptors (GPCRs) can regulate the activity of the chloride cotransporter KCC2 through S940 phosphorylation (29). Activity of the angiotensin type 1a (AT_1a) receptor could help stabilize KCC2 expression in the cell membrane, supporting a low intracellular chloride concentration and the inhibitory effects of GABA. In the present study, we tested the relationship between AT_1a receptors and GABA-mediated inhibition in the MnPO using a virally mediated knockdown (KD) approach.

MATERIALS AND METHODS

Animals.

Experiments were performed according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the University of North Texas Health Science Center Institutional Animal Care and Use Committee. These experiments used 6-wk-old (250–300 g) adult male Sprague-Dawley rats (Charles River Laboratory, Wilmington, MA). Rats were individually housed in temperature-controlled rooms with a 12-h:12-h light/dark cycle with the light phase lasting from 7 AM to 7 PM. Surgeries were performed using aseptic technique, and postoperative infection was prevented by subcutaneous administration of procaine penicillin G (30,000 U). A nonsteroidal anti-inflammatory drug, carprofen (Bio-Serv, Flemington, NJ; 2 mg tablet po), was given before and after surgery for pain management.

Viral vectors and injections.

Viral vectors used in these experiments were purchased from GENEDETECT (Auckland, New Zealand). The adeno-associated viruses (AAV) (rAAV1/2) contained either a small hairpin RNA sequence targeted to the AT_1a receptor (AT_1a KD, GD1009-RV Rat AT_1aR RNAi) or a scrambled (Scr, rAAV1/2-U6-SCR.shRNA-terminator-CAG-EGFP-WPRE-BGH-polyA) sequence. Viruses were used undiluted at a titer of 1.1 × 10¹² genomic particles/ml. Both viruses expressed green fluorescent protein (GFP) to allow identification of transfected neurons.

Microinjections.

Rats were anesthetized with 2% isoflurane, and their scalps were shaved and disinfected with alcohol and iodine. Each rat was placed in a stereotaxic head frame (David Kopf, Tujunga, CA). To ensure accurate injections, skulls were leveled at two cranial suture landmarks bregma and lambda. The injector was angled 8° from medial to lateral, and the injection coordinates used for the MnPO were 0 mm anterior, 0.9 mm lateral, and 6.7 mm ventral from bregma (33). After a burr hole was drilled at the site of injection, a 30-gauge steel injector was lowered to the MnPO, and 200–300 nl of AAV was delivered at a rate of 200 nl/min. The injector was connected to a Hamilton 5-µl syringe (no. 84851; Hamilton, Reno, NV) by calibrated polyethylene tubing that was used to determine the injection volume. The injector remained in place for 5 min to allow for absorption and then was slowly withdrawn. Gel foam was packed in to the opening in the cranium. Absorbable antibiotic suture was used to close the incision site and minimize postsurgical infection.

Slice preparation.

Two weeks after the microinjection, rats were anesthetized with isoflurane (2%) and decapitated. Coronal slices (300 µm) containing the MnPO were cut using a Microslicer DTK Zero 1 (Ted Pella, Redding, CA) in ice-cold (0–1°C), oxygenated (95% O₂-5% CO₂) cutting solution consisting of the following (in mM): 3.00 KCl, 1.00 MgCl₂·6H₂O, 2.00 CaCl₂, 2.00 MgSO₄, 1.25 NaH₂PO₄, 26.00 NaHCO₃, 10.00 d-glucose, and 206.00 sucrose (300 mosM, pH 7.4). Slices were incubated at room temperature in oxygenated (95% O₂-5% CO₂) artificial cerebrospinal fluid (aCSF) containing the following (in mM): 126.00 NaCl, 3.00 KCl, 2.00 CaCl₂, 2.00 MgSO₄, 1.25 NaH₂PO₄, 26.00 NaHCO₃, and 10.00 d-glucose (300 mosM, pH 7.4) for a minimum of 1 h before recording.

Electrophysiology.

Slices containing the MnPO were transferred to a submersion recording chamber, where they were superfused with aCSF (31 ± 1°C). Slices were visualized using an upright epifluorescent microscope (BX50WI; Olympus, Waltham, MA) with differential interference contrast optics. Loose-patch voltage-clamp recordings were obtained using borosilicate glass micropipettes (1–3 MΩ) containing aCSF as the internal solution. Action potentials were amplified and digitized using Multiclamp 200B and Digidata 1440A, respectively (Molecular Devices, San Jose, CA). Signals were filtered at 2 kHz and digitized at 10 kHz. Recordings from MnPO neurons were made by targeting enhanced GFP-expressing neurons in slices prepared from rats injected with the shRNA against AT_1a or the scrambled control construct (Fig. 1A).

Fig. 1.

Angiotensin 1a receptor knockdown (AT_1a KD). A: digital image of green fluorescent protein (GFP) fluorescence and recording electrode from slice containing the median preoptic nucleus (MnPO) (scale bar = 20 µm). Loose-patch voltage-clamp recordings were made from GFP-positive MnPO neurons. B: results from qRT-PCR studies showing that AT_1a KD (n = 7) reduced AT_1a mRNA expression compared with uninjected controls (n = 6) and scrambled sequence (Scr)-injected rats (n = 7). **P < 0.01.

Baseline action potential firing was recorded for 5 min. Drugs were then focally applied for 30 s (10 psi) using a Pico spritzer (WPI, Sarasota, FL) with a patch pipette containing the drug placed 150–200 μm upstream of the recording electrode followed by an additional 10 min of recording. In some cases, the AT_1a antagonist losartan was bath applied followed by focal application of ANG II (100 nM). In these instances, bath application of losartan occurred before baseline recordings and was maintained throughout the recording procedure. Compounds used were the AT_1a agonist ANG II (Sigma-Aldrich, St. Louis, MO), the AT_1a antagonist losartan (10 µM; Tocris, Bristol, UK), the GABAA agonist muscimol (100 µM; Sigma-Aldrich), and the GABAA antagonist bicuculline methobromide (10 µM; Sigma-Aldrich). ANG II, losartan, and bicuculline were dissolved in aCSF. Stock solutions of muscimol (100 mM) were first dissolved in 0.05 M HCl then diluted to a final concentration in aCSF (100 µM). The KCC2 blocker VU 0240511, the PKC activator phorbol 12,13-dibutyrate (PDBu), and the PLC activator m-3M3FBS were purchased from Tocris and were first dissolved in DMSO (Sigma-Aldrich) then diluted to a final concentration of 10 µM (< 0.1% DMSO) in aCSF for bath application. Parameter measured for action potential firing was spike count separated into 30-s bins.

qRT-PCR.

Punches containing the MnPO (2–3 per rat) were collected in PARIS Kit (Invitrogen, Carlsbad, CA) buffer, and RNA was purified according to manufacturer instructions. The purified RNA underwent amplification using the epicenter TargetAmp 2-Round AminoallylaRNA Amplification Kit 1.0 and purified using the RNeasy MinElute Cleanup Kit (Madison, WI). Quality and purity of the RNA were verified spectrophotemetrically on a Nanodrop 2000c (ThermoScientific, Wilmington, DE). RNA was used in the experiment when the 260- to 280-nm wavelength absorbance ratio was above 1.8. Forty nanograms of the resulting purified RNA were used for reverse transcriptase PCR via the Sensiscript Reverse Transcription Kit as previously described (7, 11, 31, 32, 34).

The cDNA that was produced from reverse transcription was used for quantitative PCR. Individual qPCR reactions consisting of 1.8 µl of cDNA, 1.2 µl of a primer mix, 7.5 µl of iQ SYBR Green Supermix, and 4.5 μl RNase-free water were performed in a 96-well plate. A Bio-Rad iQTM5 iCycler 191 system (Bio-Rad Laboratories, Hercules, CA) and thermocycler were used to perform the qPCR. Each reaction was performed in duplicate, and relative fluorescence unit data from the iCycler were analyzed according to the 2^-∆∆Ct method in accordance with our previous publications (7, 11, 31, 32). Specific primer sequences are shown in Table 1.

Table 1.

Forward and reverse primer sequences used in qRT-PCR experiments

	Sequence
AT1a
Forward	5′-AGCCTGCGTCTTGTTTTGAG-3′
Reverse	5′-GCTGCCCTGGCTTCTGTC-3′
GABAA-β1
Forward	5′-CGTGTTCCTGGCTCTACTGG-3′
Reverse	5′-TCAGTGGTTTGCGGTACTGG-3′
GABAA-β2
Forward	5′-GGACAATCGAGTGGCAGACC-3′
Reverse	5′-GTGGATACCGCCTTAGGTCC-3′
GABAA-β2
Forward	5′-ATGGAACAGTGCTGTACGGG-3′
Reverse	5′-ACCTGTGGCGAAGACAACAT-3′
KCC2
Forward	5′-CCATGGCTTTGATGTCTGTGCCAA-3′
Reverse	5′-ACTCATCACAGGTGGCATTGAGGA-3′
S18
Forward	5′-CAGAAGGACGTGAAGGATGG-3′
Reverse	5′-CAGTGGTCTTGGTGTGCTGA-3′

AT1_a, angiotensin type 1a; GABA, γ-aminobutyric acid.

Western Blot.

Two weeks after microinjection of the AAV vectors, rats were deeply anesthetized with 2% isoflurane and then decapitated. Brains were removed and cut into 1-mm coronal sections using a commercially available brain matrix. Punches containing the MnPO (2–3 per rat) were collected in a RIPA lysis buffer containing DTT, chelators, and protease inhibitors using a blunt-tip 23-gauge needle and stored at −80°C until protein isolation and Western blots were performed. Total lysate (20–25 µg) was separated via electrophoresis on 15% acrylamide SDS gel (no. 465–1084; Bio-Rad) in Tris-glycine buffer at 100 V for 1–2 h. Samples were then transferred to a PVDF membrane (Immobilon-P; EMD-Millipore, Billerica, MA) in Tris-glycine buffer (25.0 mM Tris, 192.0 mM glycine, 0.1% SDS; pH 8.3) containing 20% methanol (vol/vol) at 50 V for 2 h at 4°C. Membranes were blocked with 5% BSA in Tris-buffered solution with Tween 20 (TBST) (25 nM Tris base, 125 mM NaCl, 0.1% Tween 20) for 30 min at room temperature. Blots were incubated in primary antibodies for Ser940 pKCC2 (1:500 rabbit polyclonal; 612-401-E15; Rockland Immunochemicals, Limerick, PA), KCC2 (1:300 rabbit polyclonal; 07-432; Millipore Sigma, Burlington, MA), GABAA-β2 (1:1,000 rabbit polyclonal; AB5561; Millipore Sigma), GABAA-β3 (1:1,000 mouse monoclonal; ab98968; Abcam, Cambridge, UK), or β-actin (1:2,000 mouse monoclonal; A5441; Sigma-Aldrich) overnight at 4°C. Blots were washed with TBST and incubated in species-appropriate secondary antibodies (1:2,000) for 2 h at room temperature. Proteins were detected by chemiluminescence using a Supersignal West Femto Maximum Sensitivity kit (Thermo Scientific, Waltham, MA) and imaged with a G-Box (Syngene, Frederick, MD). Membranes were then washed in TBST and stripped with Restore PLUS stripping buffer (Thermo Scientific) before being reprobed. Separate blots were used for analysis of KCC2 and GABAA β-subunit protein expression. Densitometry was analyzed using ImageJ, and quantification of protein was normalized to β-actin.

Statistics.

Data from loose-patch recordings were analyzed for group- and time-dependent effects using two-way repeated-measures ANOVA or three-way ANOVA followed by Holm-Sidak post hoc tests. Mann-Whitney U-test were used to make baseline and area under the curve comparisons. Western blot and qRT-PCR data were analyzed by one-way ANOVA and Bonferroni t-tests. Analyses were performed using SigmaPlot software (v. 12.0; Systat Software, San Jose, CA).

RESULTS

AT_1a KD results in reduced AT_1a message compared with control.

To confirm the efficacy of AT_1a KD to reduce AT_1a mRNA, qRT-PCR was used to measure AT_1a mRNA in MnPO punches from AT_1a KD and Scr shRNA-injected rats. Message from AT_1a KD rats was 33% reduced compared with control rats using 2^-∆∆Ct analysis (Fig. 1B). One-way ANOVA shows that there was a significant effect of treatment [F(2,17) = 17.46, P < 0.001]. The AT_1a KD was effective at reducing AT_1a mRNA compared with samples from uninjected control rats (P < 0.001) and Scr-treated rats (P < 0.001). Control and Scr AT_1a mRNA were not different from each other (P = 0.54).

AT_1a KD impairs ANG II-dependent excitability in MnPO neurons.

The activity of MnPO neurons was monitored using in vitro loose-patch recordings to assess the effect of ANG II on spontaneous action potential firing in slices prepared from rats injected with either Scr or AT_1a KD (Fig. 2, A–C). Neurons from rats that received Scr injections exhibited a mean basal firing rate of 1.4 ± 0.3 Hz (n = 12 cells from 6 rats). In contrast, neurons from rats that received the AT_1a KD showed a higher basal firing rate (2.6 ± 0.6 Hz, n = 12 cells from 3 rats) compared with that observed in the Scr-injected rats (Fig. 2D). Mann-Whitney U-test showed that the basal firing rate of AT_1a KD neurons was significantly higher than Scr neurons (P < 0.05, Fig. 2E).

Fig. 2.

Angiotensin responses in median preoptic nucleus (MnPO) neurons. A: representative raw loose-patch recordings of the effects of focally applied angiotensin II (ANG II) (100 nM, shaded bars) on the activity of green fluorescent protein (GFP)-labeled MnPO neurons in slices from scrambled sequence (Scr)-injected rats (left) and angiotensin type 1a knockdown (AT_1a KD)-injected rats (right). B and C: firing rate histograms of the recordings in A showing that ANG II (shaded bar) increases spontaneous activity of Scr-injected (B) but not AT_1a KD (C) MnPO neurons. D: summary of firing rate data shows that ANG II-mediated increases in activity seen in Scr (n = 12) MnPO neurons are blocked by bath application of losartan (10 µM, n = 12) and AT_1a KD (n = 12). E: AT_1a KD is associated with an increase in baseline spontaneous activity. F: summary data showing that ANG II application had no significant effect on the activity of neurons from AT_1a KD rats but induced a significant threefold increase from baseline in spontaneous activity in neurons from Scr rats. Losartan blocked the ANG II-mediated increase in activity in neurons from Scr rats. G: integrated area in response to ANG II was significantly greater in Scr compared with AT_1a KD and Scr + Losartan. *P < 0.05. AUC, area under the curve.

In Scr neurons, focal application of ANG II resulted in a transient increase in firing rate (9 of 12 neurons) that peaked within 2 min and persisted for 10 min. Bath application of the AT_1a antagonist losartan did not significantly alter the basal firing rate of MnPO neurons in vitro (1.5 ± 0.2 Hz, n = 12 cells from 3 rats), but losartan did block the responses to ANG II (Fig. 2F). In rats that received the AT_1a KD, ANG II failed to produce a change in firing rate (Fig. 2F). Two-way repeated-measures ANOVA on normalized firing rate shows that there was a significant effect of treatment [F(2,33) = 4.00, P < 0.05)] time [F(19,627) = 3.16, P < 0.001], and an interaction between treatment and time [F(38,627) = 1.44, P < 0.05] following application of ANG II. Holm-Sidak post hoc test shows that firing rate of AT_1a KD neurons was significantly reduced in response to ANG II application compared with Scr neurons (P < 0.05). The percentage change in firing rate of AT_1a KD neurons was smaller than Scr neurons 2 min following ANG II application and was no different from baseline activity. The integrated area of the ANG II response was calculated as the magnitude of the change in firing rate compared with the mean basal firing over the 10-min period following ANG II application (Fig. 2G). The greatest average integrated area was observed in the Scr group and was significantly larger than the AT_1a KD group and the Scr + losartan group (Mann-Whitney U-test, P < 0.05). There was no significant difference between the AT_1a KD group and the Scr + losartan group.

GABAA-mediated inhibition is impaired in AT_1a KD in loose-patch recordings.

To assess the effects of AT_1a KD on GABAA receptor function, we applied muscimol (100 µM) to the slices for 30 s using a Pico spritzer. The baseline firing rates for MnPO neurons in these experiments were as follows: Scr 0.94 ± 0.21 Hz (n = 10 cells from 3 rats), AT_1a KD 1.79 ± 0.53 Hz (n = 10 cells from 3 rats), and unlabeled 1.78 ± 0.36 Hz (n = 11 cells from 3 rats). MnPO neurons injected with the Scr exhibited a pronounced reduction in spontaneous action potential firing upon application of muscimol (Fig. 3, A and B). The muscimol-dependent reduction in spontaneous activity persisted for 9 min following termination of muscimol application. Unlabeled MnPO neurons (i.e., those that did not express GFP and hence did not express the shRNA) also showed a robust inhibition in response to muscimol that persisted for the 10-min duration of the recording. The magnitude and time course of the muscimol effect were not significantly different between Scr (n = 10 cells from 3 rats) and unlabeled (n = 11 cells from 3 rats) MnPO neurons (Fig. 3C). Muscimol did not alter firing rate of labeled cells from slices prepared from rats injected with AT_1a KD compared with their baseline firing. Two-way repeated-measures ANOVA of normalized firing rates showed that, although there was no significant effect of treatment [F(2,28) = 1.82, P = 0.18], there was a time-dependent effect [F(19,532) = 5.85, P < 0.001] and an interaction between treatment and time [F(38,532) = 1.50, P < 0.05]. AT_1a KD did reduce the initial inhibitory effect of muscimol during the first 1.5 min following application compared with Scr or unlabeled controls (Holm-Sidak post hoc tests, both P < 0.01). Furthermore, bath application of the GABAA antagonist bicuculline (10 µM) significantly increased basal activity in Scr (paired t-test, P < 0.01, n = 13 cells from 3 rats) but did not alter basal activity in AT_1a KD (paired t-test, P = 0.67, n = 9 cells from 3 rats). This indicates that AT_1a KD significantly attenuated the inhibitory effects of GABAA receptor activation.

Fig. 3.

Effects of angiotensin type 1a knockdown (AT_1a KD) on muscimol-mediated inhibition. AT_1a KD significantly reduced the inhibitory effects of focal applications of muscimol (100 µM, shaded bar) on the firing rate (A) and percentage of baseline (B) of median preoptic nucleus (MnPO) neurons from AT_1a KD-injected rats (n = 10) compared with neuron from slices from scrambled sequence (Scr)-injected rats (n = 10) and unlabeled control (n = 11). C: data from B expressed as integrated area under the curve (AUC). D: bicuculline increases basal firing in Scr neurons (n = 13) but not AT_1a KD neurons (n = 9). *P < 0.05, **P < 0.01.

KCC2 and AT_1a interaction.

Because we observed reduced GABAA-mediated inhibition associated with AT_1a KD, we tested for changes in expression of GABAA receptor subunits by qRT-PCR and Western blot analysis (Fig. 4A). We used qRT-PCR to compare the expression in β-subunit mRNA for the GABAA receptor in AT_1a KD to Scr-injected rats. We failed to detect a difference in β-subunit mRNA between AT_1a KD and Scr rats for all three subunits (β1 P = 0.71, β2 P = 0.12, β3 P = 0.62). We also investigated whether there was reduction in GABAA β-subunit protein expression despite the failure to detect a significant AT_1a KD-mediated reduction in β-subunit mRNA (Fig. 4, B and C). We failed to detect expression of the β1-subunit and failed to detect differences in β3-subunit protein expression between MnPO samples from AT_1a KD- and Scr-treated rats. There was a trend toward a reduction in β2-subunit expression in the AT_1a KD compared with Scr, but this was not significant (P = 0.07).

Fig. 4.

Angiotensin type 1a knockdown (AT_1a KD) had no effect on γ-aminobutyric acid (GABA)A β-subunit expression. A: AT_1a KD (n = 7) was not associated with a significant reduction in β-subunit mRNA for GABAA receptors compared with scrambled sequence (Scr) (n = 7). B: similarly, AT_1a KD (n = 3) was not associated with a significant reduction in β-subunit protein for GABAA receptors compared with Scr (n = 3). C: Western blots for the β-subunit expression with β-actin loading controls.

Because we failed to detect decreases in GABAA β-subunit mRNA and protein expression associated with AT_1a KD, we tested for changes in chloride transporter expression to determine whether changes in the Cl⁻ gradient might underlie the impaired GABAA-mediated inhibition in the AT_1a KD (Fig. 5A). Analysis of KCC2 mRNA expression showed that AT_1a KD resulted in a reduction in KCC2 mRNA compared with the Scr (P = 0.034).

Fig. 5.

Angiotensin type 1a knockdown (AT_1a KD) reduced KCC2 mRNA and S940 phosphorylation in the median preoptic nucleus (MnPO). A: KCC2 mRNA was reduced in AT_1a KD (n = 7) compared with scrambled sequence (Scr) (n = 7). B: there was no difference in total KCC2 protein between AT_1a KD (n = 3) and Scr (n = 4). C: there was a reduction in pS940 KCC2 in AT_1a KD. D: there was a reduction in the ratio of pS940 KCC2/total KCC2 compared with Scr. E: Western blots show expression of total KCC2 and pKCC2 in AT_1a KD- and Scr-treated tissue. *P < 0.05.

We also investigated whether the reduction in KCC2 mRNA was accompanied by a reduction in KCC2 protein expression. The reduction in KCC2 mRNA was not associated with a reduction in total KCC2 protein expression (P = 0.54, Fig. 5B). However, AT_1a KD was associated with a reduction in pS940 KCC2 protein (P = 0.013, Fig. 5C). The ratio of pKCC2/β-actin was reduced in AT_1a KD rats (n = 4, Fig. 5D). The reduction in pS940 KCC2 observed in AT_1a KD was significantly reduced [1-way ANOVA: F(1,5) = 14.210, P < 0.05] compared with Scr-injected rats (n = 3).

To further test the role of KCC2 in GABAA-mediated inhibition, we conducted electrophysiological experiments to determine the effects of pharmacological blockade of KCC2 on GABAA-mediated inhibition using loose-patch recording (Fig. 6A). As before, cells from AT_1a KD rats demonstrated impaired GABAA-mediated inhibition compared with Scr (Holm-Sidak post hoc, P < 0.001). In cells from Scr rats, bath application of the KCC2 blocker VU 0240551 impaired GABAA-mediated inhibition compared with GABAA-mediated inhibition in aCSF (Holm-Sidak post hoc, P < 0.001). The impaired GABAA-mediated inhibition observed in cells from the AT_1a KD rats was not reduced further in the presence of VU 0240551 (Holm-Sidak post hoc, P = 0.25). As before, there was a reduction in the magnitude of the integrated area in AT_1a KD-treated rats compared with Scr. Bath application of the KCC2 blocker did not further reduce integrated area in AT_1a KD rats but did reduce the magnitude of the integrated area in Scr rats (Fig. 6B). There was a trend toward blockade of KCC2 increasing the basal firing of Scr neurons (Fig. 6C) that did not reach significance (t-test, P = 0.06). Additionally, blockade of KCC2 did not alter the basal firing of AT_1a KD neurons (t-test, P = 0.89). This suggests that the AT_1a KD-mediated attenuation of GABAA-mediated inhibition involves an altered function of KCC2.

Fig. 6.

Effects of KCC2 blockade on muscimol inhibition in median preoptic nucleus (MnPO) neurons. A: γ-aminobutyric acid (GABA)A-mediated inhibition produced by muscimol (100 µM) seen in the scrambled sequence (Scr) (n = 9) MnPO neurons was reduced by KCC2 blockade with VU 0240551 (10 µM). The impaired GABAA-mediated inhibition seen in the angiotensin type 1a knockdown (AT_1a KD) (n = 8) was not further reduced by bath application of a KCC2 blocker. B: integrated area under the curve (AUC) for percentage of baseline measures was reduced in all AT_1a KD and in Scr MnPO neurons treated with a KCC2 blocker. *P < 0.05. C: KCC2 inhibition with VU 0240551 did not influence firing rate.

KCC2 has been shown to require phosphorylation of S940 to be stably expressed in the cell membrane, and phosphorylation of this residue is maintained by serine/threonine kinases via PKC and PLC. To further investigate the mechanisms linking AT_1a to KCC2 function, we looked to see whether we could recover function of KCC2 and subsequently GABAA-mediated inhibition in AT_1a KD (Fig. 7A) rats by directly activating PKC/PLC (downstream effectors of AT_1a activation). There was a significant effect of drug application on GABAA-mediated inhibition [F(2,27) = 3.99, P < 0.05]. As shown before, the inhibitory effects of muscimol were reduced significantly in neurons from slices prepared from AT_1a KD rats (Fig. 7A). Application of PDBu, a PKC activator, recovered GABAA-mediated inhibition in the AT_1a KD (P < 0.05). Additionally, activation of PLC using m-3M3FBS recovered GABAA-mediated inhibition (P < 0.05) but was not different in recovering GABAA-mediated inhibition in AT_1a KD from PDBu (P = 0.75). However, application of either PKC or PLC activators failed to alter GABAA-mediated inhibition in cells from Scr rats (Fig. 7B). The magnitude of integrated area was only reduced in MnPO neurons from AT_1a KD rats treated with aCSF (Fig. 7C). Bath application of PKC or PLC activators did not alter the basal firing rate of AT_1a KD or Scr neurons (data not shown).

Fig. 7.

Effects of pharmacological activation of phospholipase C (PLC) or protein kinase C (PKC) on muscimol-mediated inhibition of neurons from scrambled sequence (Scr)- and angiotensin type 1a knockdown (AT_1a KD)-injected rats. A: impaired γ-aminobutyric acid (GABA)A-mediated inhibition in response to muscimol application (100 µM, shaded bar) was observed in AT_1a KD median preoptic nucleus (MnPO) neurons (n = 10). Bath application of the PLC activator m-3M3FBS (10 µM; n = 8) or the PKC activator phorbol 12,13-dibutyrate (PDBu) (10 µM; n = 12) recovered GABAA-mediated inhibition in AT_1a KD neurons in MnPO. B: GABAA-mediated inhibition seen in Scr MnPO neurons (n = 10) was not altered by bath application of PLC (n = 13) or PKC (n = 9) activators. C: integrated area under the curve (AUC) in response to muscimol application was not reduced in AT_1a KD MnPO treated with a PLC or a PKC activator. *P < 0.05 compared with Scr treated with artificial cerebrospinal fluid (aCSF).

DISCUSSION

Focal application of ANG II increased the spontaneous firing in MnPO neurons consistent with previous reports demonstrating the excitatory effects of ANG II in the MnPO (1, 36). Virally mediated AT_1a KD reduced mRNA expression for the receptor in the MnPO and blocked the excitatory effects of exogenous ANG II application. The AT_1a KD was as effective at blocking the excitatory effects of ANG II as the AT_1a antagonist losartan. However, AT_1a KD was also associated with an increase in basal firing, whereas AT_1a inhibition with losartan was not associated with an increase in basal firing. Although losartan did not alter spontaneous firing in MnPO neurons in the above loose-patch voltage-clamp recordings, it has been shown to reduce inhibitory currents in whole cell voltage-clamp recordings (16).

The AT_1a KD-mediated increases in basal activity were due to reductions in GABAA-mediated inhibition. In the present experiment, the GABAA agonist muscimol induced a robust inhibition in spontaneous activity of Scr-injected MnPO neurons. In neurons from AT_1a KD rats, muscimol-mediated inhibition was greatly reduced. However, this reduced response to muscimol associated with AT_1a KD in the MnPO was not associated with significant reductions in GABAA β-subunit expression of mRNA or protein (although there was a trend toward a reduction in β2-subunit mRNA). This suggests that the effects of AT_1a KD on GABAA responses do not involve a direct effect on the receptor.

We next tested for possible changes in Cl⁻ homeostasis, as it has been shown that GPCR regulates the membrane expression of KCC2 (29). Inhibition of KCC2, a K⁺/Cl⁻ cotransporter involved in maintaining low intracellular Cl⁻ concentrations, reduced the inhibitory effects of GABAA receptor activation in cells recorded in slices from rats injected with the control vector. In slices from rats treated with shRNA against AT_1a, KCC2 inhibition did not influence GABAA-mediated inhibition, suggesting that the KCC2 activity was already compromised. This finding is consistent with a change in KCC2 expression and/or function associated with AT_1a KD. As shown above, AT_1a KD was associated with a reduction in KCC2 mRNA, and reductions in transcription have been shown to reduce GABAA currents (21). Reductions in mRNA do not necessarily translate to reductions in functional protein. In fact, AT_1a KD did not reduce total KCC2 protein expression in the MnPO. However, there was a reduction in S940 phosphorylation of KCC2 in the MnPO of AT_1a KD-treated rats. Phosphorylation of KCC2, specifically at S940, has been shown to be important for stabilizing membrane expression of KCC2, and it is PKC dependent (25, 26). Although it has not been demonstrated that PLC directly interacts with KCC2 to modify phosphorylation, PLC does upregulate diacylglycerol- and inositol triphosphate-dependent increases in intracellular calcium, both of which increase the activity of PKC. Other GPCR systems have been shown to modulate KCC2 function. Activation of group 1 mGluRs, a receptor system that upregulates PKC function, has been shown to mediate GABA inhibition through the regulation of KCC2 (3). The present study suggests that chronic inhibition of AT_1a function is associated with reduced GABAA inhibition through a decrease in KCC2 function.

Activation of AT_1a receptors increases PLC and PKC activity (5). Bath application of PLC and PKC activators was effective in recovering GABAA-mediated inhibition of spontaneous activity in cells transfected with shRNA against AT_1a. Although PLC/PKC activation in Src controls did not enhance GABAA-mediated inhibition of spontaneous activity, it is still possible that there was an increase in the magnitude of inhibitory currents. The recording methods in these experiments were limited to measurements of firing rate and, therefore, subject to floor effects. Another possibility is that constitutive activity of PLC/PKC pathway is high enough that phosphorylation of KCC2 cannot be increased further.

The above experiments implicate a role for AT_1a receptors in the modulation of GABAA-mediated inhibition. This permissive AT_1a receptor effect appears to be mediated by PLC/PKC-dependent phosphorylation of KCC2. This suggests that angiotensin in the MnPO may play an important modulatory role that supports the inhibitory effects of GABA. Maintaining the inhibitory effects of GABA could be important in maintaining the balance between inhibition and excitation in this nucleus. Changes in the excitatory/inhibitory balance within MnPO neurons could influence body fluid homeostasis, central cardiovascular control, thermoregulation, and sleep homeostasis. Chronic intermittent hypoxia (CIH), an animal model simulating the hypoxemia associated with sleep apnea, is associated with a sustained increase in blood pressure that persists even during periods of normoxia (13, 22). Neuronal activity (as measured by FosB staining) in the MnPO is increased following CIH (22). Additionally, intracerebroventricular infusions of losartan block CIH-induced hypertension as well as increases in FosB staining (23). This suggests a role of AT_1a receptor activation in the pathogenesis of CIH-induced hypertension. Further studies are required to determine the role of AT_1a receptors in CIH-induced hypertension and whether that role involves AT_1a-mediated changes in Cl⁻ homeostasis.

It has been shown that the SFO utilizes ANG II as a neurotransmitter in its projections to the paraventricular nucleus (PVN) (2, 28) and may also utilize ANG II as a neurotransmitter in projections to the MnPO. ANG II-dependent activation of AT_1a receptors in the PVN has been shown to be excitatory (8, 27). Activation of preautonomic PVN neurons via AT_1a receptor stimulation can thus enhance blood pressure through projections to the rostral ventrolateral medulla (10, 35). In much the same way, ANG II-mediated activation of MnPO could influence blood pressure through projections to the PVN (15, 19, 39).

The central effects of ANG II on blood pressure are not static and can exhibit plasticity. Chronic peripheral administration of a nonpressor dose of ANG II has been shown to sensitize subsequent pressor doses of ANG II and enhance hypertension through central ANG II receptors (42). The induction of this sensitization to ANG II has also been demonstrated with chronic administration of aldosterone (43) or a high-salt diet (9). Although acute exposure to elevated levels of ANG II can enhance neuronal excitability and transiently increase blood pressure, chronic elevations may sensitize the neuronal responses to ANG II.

The results from the present study suggest that the AT_1a receptor contributes to two seemingly opposing effects in MnPO neurons. The excitatory effects of ANG II lead to increased activity and appear to occur on an acute time scale. In contrast, constitutive activity of the AT_1a may be necessary for the inhibitory effects of GABA. The net effect of AT_1a receptors in MnPO neurons may be to contribute to synaptic homeostasis or maintain a balance between inhibition and excitation, needed for the MnPO to play its role in several different organismal homeostatic systems. Changes in angiotensin signaling could alter synaptic homeostasis in the MnPO, which could have functional consequences for behavioral and physiological systems that are regulated by this region. Different patterns of ANG II administration can produce desensitization (12) or sensitization (17, 20) of AT_1a-mediated responses, and these changes in sensitivity may involve different signaling pathways (20, 41). The functional consequences of desensitization or sensitization of AT_1a receptors in MnPO could depend on how these different electrophysiological responses are integrated to influence the activity of downstream targets.

Perspectives and Significance

The present study demonstrates a novel interaction between AT_1a receptors and GABAA inhibition in MnPO neurons. Chronic AT_1a receptor knockdown appears to decrease GABAA inhibition not through changes in GABAA receptor efficacy but rather through the regulation of the K⁺/Cl⁻ cotransport system. Here, loss of AT_1a receptor function disrupts the chloride homeostasis and reduces GABAA-mediated inhibition. It has been previously shown that acute pharmacological inhibition of AT1 receptors can influence GABAA receptors without influencing chloride transport although this affect appeared to be limited to sodium-sensitive cells (16). It could be that chronic vs. acute inhibition of AT_1a may involve different mechanisms that regulate GABAA inhibition.

The MnPO is involved in body fluid homeostasis and cardiovascular control as well as thermoregulation and sleep (30). The interaction between AT_1a and GABAA receptors may be involved in maintaining synaptic homeostasis through control of the excitatory/inhibitory balance of MnPO neurons. Changes in this excitatory/inhibitory balance could impair the regulatory function of the MnPO. For example, dysregulation of excitatory/inhibitory balance in MnPO neurons may occur at physiological or pathophysiological states featuring AT_1a overactivation. Although the present study did not address the effects of AT_1a overactivation, as might be observed in hypertension, future studies would benefit from such investigations of the role of elevated AT_1a activation on GABAA-mediated inhibition.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grant P01-HL-088052.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

G.E.F. and J.T.C. conceived and designed research; G.E.F., K.B., M.E.B., and J.T.L. performed experiments; G.E.F., K.B., M.E.B., and J.T.L. analyzed data; G.E.F., K.B., M.E.B., J.T.L., and J.T.C. interpreted results of experiments; G.E.F., K.B., and J.T.C. prepared figures; G.E.F. and J.T.C. drafted manuscript; G.E.F. and J.T.C. edited and revised manuscript; G.E.F., K.B., and J.T.C. approved final version of manuscript.