Novel expression of GABA_A receptors on resistance arteries that modulate myogenic tone

Abstract

The clinical administration of GABAergic medications leads to hypotension which has classically been attributed to the modulation of neuronal activity in the central and peripheral nervous system. However, certain types of peripheral smooth muscle cells have been shown to express GABA_A receptors, which modulate smooth muscle tone, by the activation of these chloride channels on smooth muscle cell plasma membrane. Limited prior studies demonstrate that non-human large caliber capacitance blood vessels mounted on a wire myograph are responsive to GABA_A ligands. We questioned whether GABA_A receptors are expressed in human resistance arteries and whether they modulate myogenic tone. We demonstrate the novel expression of GABA_A subunits on vascular smooth muscle from small caliber human omental and mouse tail resistance arteries. We demonstrated that GABA_A receptors modulate both plasma membrane potential and calcium responses in primary cultured cells from human resistance arteries. Lastly, we demonstrate functional physiologic modulation of myogenic tone via GABA_A receptor activation in human and mouse arteries. Together these studies demonstrate a previously unrecognized role for GABA_A receptors in the modulation of myogenic tone in mouse and human resistance arteries.

Introduction

The prevalence of hypertension in the United States is greater than 30% in the adult population and the health care costs associated with hypertension is astronomical.[1] The direct correlation between hypertension and other comorbid states has made hypertension a risk factor for almost all leading causes of mortality[2, 3] and morbidity[4–6]. As the population of patients with hypertension increases[2], new and effective therapies are needed to treat patients who are failing current therapies for hypertension.

It is well recognized that clinically used GABAergic anesthetics are associated with hypotension due to a decreased systemic vascular resistance. Mechanisms of GABAergic anesthetic induce hypotension are thought to include peripheral autonomic ganglion inhibition,[7] central paraventricular nucleus inhibition[8] and baroreceptor inhibition.[9] However, these neuronal targets may not provide a complete mechanism of anesthetic-induced hypotension. Furthermore, there has been growing evidence that GABAergic anesthetic mediated vasorelaxation is due to direct activation of GABAA receptors on vascular smooth muscle (VSM) cells.[10–14] However, these studies were performed in large arteries or in cell cultures derived from these large vessels, neither of which are the physiological determinants of blood pressure. The primary determinants of vascular resistance are small resistance arteries. These resistance arteries have not yet been evaluated for their intrinsic GABAergic response on myogenic tone or their GABA receptor subunit expression.

GABA_A receptors are ligand-gated, inhibitory chloride channels that are well characterized in the central nervous system (CNS). While their physiology and pharmacology have been extensively characterized in the CNS, it has only recently been appreciated to be functionally expressed on many peripheral cell types, including other non-vascular smooth muscle [15]. Classic GABA_A receptors are hetero-pentamers containing a combination of 19 possible subunits (α[1–6], β[1–3], γ[1–3], δ, ε, θ, π, and ρ[1–3]). GABA_A channels containing the necessary subunits to form a functional channel have been previously described in airway smooth muscle [16]. Furthermore, activation of these GABA_A receptor channels on airway smooth muscle cells increased chloride channel flux and attenuated calcium release into the intracellular compartment, leading to a direct decrease in airway smooth muscle tension [17, 15, 18–23]. Recognizing the vasodilatory effects of GABAergic anesthetics and the growing evidence of direct GABA_A mediated smooth muscle relaxation, we hypothesize that GABA_A receptors exist on VSM cells of small resistance arteries and have significant effects on resistance arterial tone and function.

Materials and Methods

Materials:

Gabazine, NS1619, KCl was obtained from Sigma (St. Louis, MO, USA). Muscimol, picrotoxin and bicuculline were obtained from Tocris (Bristol, UK). Propofol was obtained from MP Biomedical (Santa Ana, CA, USA). The fluorescent potentiometric probe (FLIPR) membrane potential assay kit was obtained from Molecular Devices (Sunnyvale, CA, USA). Trypsin 0.05%- EDTA and FURA-2 (calcium dye) were purchased from Invitrogen (Carlsbad, CA, USA). TRIzol reagent was obtained from Ambion (Austin, TX, USA). Immunoaffinity purified primary antibodies for the GABA_A α4 subunit (LS-C15) were obtained from Lifespan Biosciences (Seattle, WA, USA).

Animals and human tissue:

All mouse protocols were approved by the Columbia University Institutional Animal Care and Use Committee. Mice were euthanized with i.p. pentobarbital. Mouse tail arteries were obtained from 5–10wk old C57/Bl6 wild type mice of both sexes. The human omental tissue protocol was reviewed by the Columbia University Institutional Review Board and were deemed not human subjects research under 45 CFR 46. Adult human omental samples were received from discarded tissue during colorectal surgeries and resistance arteries (80–120um) were dissected under a dissecting microscope. This method of procurement used deidentified samples as per IRB review and therefore medical information regarding patients is not available.

Cell culture:

The adventitial layer was removed from human omental arteries using forceps under a dissecting microscope. The lumen was opened and the inner endothelial layer was mechanically abraded to remove the endothelium and basement membrane. The remaining muscle layer was then dissociated using a Worthington Biochemical (Lakewood, NJ, USA) Papain Dissociation System with collagenase type 4 (2 un/ml) added during the digestion step. Digestion occurred at 37°C for 1 hr under constant agitation by a rocker as per the manufacturer’s protocol, which includes an ova mucoid gradient to isolate smooth muscle from other cell types. After isolation, the cells were grown to confluence in 75-cm² flasks and 96-well black-walled clear bottom plates for fluorescent FLIPR membrane potential assays. All cells were maintained in M199 media (Gibco-Thermofisher Waltham, MA, USA) with or without additives (human recombinant fibroblast growth factor (FGF) 1ng/ml, human recombinant epidermal growth factor (EGF) 0.25ng/ml, insulin 1ug/ml, transferrin 0.55 ug/ml, selenium 0.67ng/ml and 10% fetal bovine serum (FBS) at 37°C in 95% air/5% CO₂. Prior to all cell-based assays, primary vascular smooth muscle cells were maintained in serum- and additive-free M199 media (i.e. basal M199 media only) for 24 hours prior to experimentation.

Isolation of RNA and reverse transcription polymerase chain reaction (RT-PCR):

In experiments where RNA was isolated using laser capture micro-dissection (LCMD), mouse tail and human omental resistance arteries were embedded in OCT compound followed by isopentane/dry ice freezing. Frozen sections of 6 μm thickness were made under RNase-free conditions and were placed on a single 1-mm PEN-membrane-coated slide (PALM Microlaser Technologies, Westchester, NY, USA) and process using a LCMD staining kit (Ambion AM1935). Histologic and morphologic guided laser dissection was performed, harvesting only central portions of smooth muscle to avoid contamination from other adjacent cell types using a PALM MicroBeam laser microscope. RNA was recovered from LCMD samples using an RNAqueousmicro kit (Ambion) according to manufacturer’s recommendations (approximately 3 ug total RNA was recovered per 5 million square microns). Recovered RNA was reversed transcribed into cDNA using a commercially available kit, SuperScript VILO (Invitrogen). LCMD-captured RNA (10 μl) in sample buffer was reverse transcribed using random hexamer primers at 42°C for 1 h in 20 μl, according to manufacturers’ recommendations.

Newly synthesized cDNA (5 μl) from RNA isolated by LCMD from mouse tail arterial smooth muscle cells were used in PCR using the Advantage 2 Polymerase Kit (Clontech, Mountain View, CA, USA). Sense and antisense primers (0.4 μM) were used for corresponding GABA_A channel subunits (table 1) and were designed within separate exons to span large introns to ensure that contaminating genomic DNA did not yield a false positive PCR bands at the molecular size expected for cDNA amplification. Primers targeting GAPDH (table 1) were used as a relative estimate of RNA quantity between tissue samples and to confirm RNA quality in each sample. All cDNA samples were denatured at 94°C for 10 s. Annealing and extension temperatures were all 68°C for 1 min. Each sample underwent 30–40 cycles of amplification in a PTC-200 Peltier thermal cycler (Bio-Rad, Hercules, CA, USA). PCR products were analyzed on a 5% nondenaturing polyacrylamide gel in Tris acetate, EDTA buffer. The gel was stained with ethidium bromide (Molecular Probes, Eugene, OR, USA) and visualized using a gel imager (Biospectra UVP, Upland, CA, USA) and Visionworks software (Biospectra UVP).

Table 1.

Human GABR subunit RT-PCR primer sequence

gene	Access no.	Sequence (5’ to 3’)	cDNA produc t	gDNA product
GABRA1	NM_ 000806	GAAGAGAAAGATTGGCTACTTTGTTATTCAAACAT	371	6268
		GAGCGTAAGTGTTGTTTTTCTTAATAAGAGGAT
GABRA2	NM_ 000807	TCTGCCCTAATTGAATTTGCAACTGTTAATTACTT	295	11616
		CTATTCTGGACATTCTGTCAATTTTGCTAACACTG
GABRA3	NM_ 000808	CACTTCCATCTCAAGCGAAAAATTGGCTACTTTGT	398	29255
		CCCACGATGTTGAAGGTAGTGCTGGTTTTCT
GABRA4	NM_ 000809	CAAACCGTATCAAGTGAAACCATCAAATCAAT	225	9074
		GCTTAGTGTGGTCATGGTGAGGACAGTTGTTAT
GABRA5	NM_ 000810	GCAGACGGTGGGCACTGAGAACA	138	2734
		GATAAGATCACGGTCATTATGCAGGGAAGGTA
GABRA6	NM_ 000811	CAGTGACAATATCAAAAGCTACTGAACCTTTGGAA	259	9556
		AATCCTGCAAATGCAACTGGGAAGAGAA
GABRB1	NM_ 000812	TCGCACTAGGAATCACGACGGTGCTTA	316	19063
		GAGCCACTCGTCTCATTCCGGATTT
GABRB2	NM_021911	GCTGCCAGTGCCAACAATGAGAAGA	170	4571
		TGGGGGTCCATCGTATACAGAGAGAAA
GABRB3	NM_000814	TCACAACTGTGCTGACAATGACAACCATCAAC	474	13271
		TAATTTTGAGCTGTGAAGACCTCCTCCGTAGA
GABRG1	NM_173536	CTTTCCCATGGATGAACATTCCTGTCCACTGGAATTTT	320	6253
		CAGGCACTGCATCTTTATTGATCCAAAAAGACACCC
GABRG2	NM_198904	AGGTCTCCTATGTCACAGCGATGGATCTCT	267	4022
		GACACTCATAGCCGTACTCTTCATCTCTCTCT
	NM_198903	AGGTCTCCTATGTCACAGCGATGGATCTCT	243	4022
		GACACTCATAGCCGTACTCTTCATCTCTCTCT
GABRG3	NM_033223	GACCTGGACAGATAGTCGCCTTCGATTCAACAGCACAA	393	153930
		TCTGTGGTGTTTCTGAGGCCCATGAAGTCAAACTGATA
GAPDH	NM_002046	CCAGGGCTGCTTTTAACTCTGGTAAAGTGGATA	213	432
		CATCGCCCCACTTGATTTTGGAGGGA

Mouse GABR subunit RT-PCR primer sequence

gene	Access no.	Sequence (5’ to 3’)	cDNA product	gDNA product
GABRA1	NM_010250	CTATGGACAGCCCTCCCAAGATGAACTTA	147	13427
		GTGACGAAAATGTCGGTCTTCACTTCAGTTA
GABRA2	NM_008066	GAACAGAGAATCGGTGCCAGCAAGA	175	32654
		TGCAAATTCAATTAGGGCAGAGAACACAA
GABRA3	NM_008067	GGTTCATAGCCGTCTGTTATGCCTTTGTA	158	10348
		GTTTTTCTTTGTTGGAGCTGCTGGTGTTT
GABRA4	NM_010251	CCTGTGCTGAAGGAGAAACACACAGAA	210	51841
		GAATGGATTTGGACTGGAAGCTAAGTG A
GABRA5	NM_176942	TGCTATGCATTTGTCTTCTCTGCTCTGATT	182	4859
		GGAGGATGGGTCAGCTTTCCAGTTGTAA
GABRA6	NM_008068	GTCTGAATCCCTGCAAGCAGAGATTGTT	138	7544
		TTTAAGATGGGCGTTCTACTGAGGGCT TT
GABRB1	NM_008069	TGTCGCCAGCATAGACATGGTCTCGGAA	138	168664
		TCAGCTACTCTGTTGTCAAGGGTCAGAT
GABRB2	NM_008070	GGGTGCCTGACACCTACTTCCTGAATGATA	175	41666
		CAGTTTTGTTCATCCAGTGGATACCGCCTT
GABRB3	NM_008071	TACCACCGTGCTCACCATGACAACCATCAAC	475	8373
		TTTGATTTTGAGCTGTGAAGACCTCCTCCGT
GABRD	NM_008072	CCAGTTCACTATCACCAGTTACCGCTTCAC	441	1292
		GCGTTCCTCACATCCATCTCTGCCCTT
GABRE	NM_017369	CACATGCTCAATTTTCCAATGGATTCTCACTCTT	377	5545
		AGCGTGGCCATGGTGAGCACAGAACTGAC
GABRG1	NM_010252	GGAATACGGAACCTTGCATTATTTTACTAGCAACA	175	20154
		CAAACACTGGTAGCCATAATCATCTTCCCCTT
GABRG2	NM_008073	GACGCTGTGGATTCTGCTCCTGCTAT	121	23665
		CTCTGGAACTTTTGGAGTCAACACCCAT
GABRG3	NM_008074	GCCAGCTGCAACTGCATAACTTCCCT	156	209406
		AGTCAAACTGATAGAGCCGCCATGAT
GABRP	NM_146017	GAGAACCTGCATTGGAGTGACAACGGTGTTA	320	1771
		AGTTGACATTGTCACCAGAGATTTCAATGCT
GABRQ	NM_020488	CTGTTCCCTGGATCTGCAAAAATTCCCTATGGAC	346	1191
		TACCCTGGCTGCAGAGGAATCATAGTTCATCCAA
GAPDH	NM_008084	CCGTAGACAAAATGGTGAAGGTCGGTGTGAA	120	1954
		CAATGAAGGGGTCGTTGATGGCAACAAT

Immunohistochemistry for GABA_A α4 subunit in human omental artery:

Omental artery tissue samples were dissected and fixed with 5% formalin for 4 hrs and stored in 70% ethanol. Following serial dehydration through a graded alcohol series, fixed tissues were then embedded in paraffin blocks and serial 6 um sections were placed on a glass microscope slide. The slide was then dewaxed and incubated in hydrogen peroxide (30% in PBS) for 30 min. After washing and permeablization with 0.1% triton-X, the slide was placed in 10% goat serum in PBS for two hours at room temperature. Serum was removed and the slide was incubated with an immunoaffinity purified GABA_A α4 subunit-specific antibody (rabbit) (LS-C15, LifeSpan BioSciences, Seattle, WA, USA) (1:100) in PBS, overnight at 4ºC. A serial section on the same slide was incubated under identical conditions except that the primary antibody was omitted. The slide was again washed and incubated with biotinylated goat anti-rabbit IgG (1:80) in 10%BSA/PBS for 1 hr at room temperature. Biotinylated antibodies were then stained and visualized using Vectastain (Burlingame, CA, USA) ABC kit.

Membrane potential measurements using fluorescent potentiometric dye:

Human vascular smooth muscle cells were grown to 100% confluency in 96-well black-walled clear-bottom plates. Cells were made serum- and growth factor-free for 24 h prior to the assay (in basal M199 medium). On the day of the assay, cells were washed with Kreb’s buffer [140 mM NaCl / 4.7 mM KCl / 2.5 mM CaCl₂ / 1.2 mM MgCl₂ / 11 mM HEPES / 10 mM D-glucose, pH 7.4]. FLIPR membrane potential blue dye was dissolved in Krebs buffer at the concentration of 1 vial (~125 mg) in 100 ml. Cells were incubated in 100 μl of 50% dye for 20 min in a humidified 37°C cell culture incubator (95% air/5% CO₂). Fluorescence was repetitively measured at 2 s intervals in a prewarmed (37°C) Flex Station 3 UV spectrophotometer (Molecular Devices) using an excitation wavelength of 530 nm, an emission wavelength of 565 nm and a cutoff filter of 550 nm. NS1619 100 μM (potassium channel opener causing hyperpolarization), 40 mM KCl (depolarizing agent), 10uM propofol or 0.1% DMSO control was added at a single concentration using the automated injection feature of the spectrophotometer during continuous fluorescent readings. In some experiments bicuculline was added to the wells manually 5min before propofol as a pretreatment.

Intracellular calcium measurements using a fluorescent ratiometric calcium indicator:

Primary cultures of human omental artery smooth muscle cells were grown to full confluence in black-walled, clear-bottomed 96-well plates. Cells were made serum- and growth factor-free for 24 h prior to the assay (in basal M199 medium). Cells were washed with 100 μl (per well) of Hanks’ balanced salt solution (HBSS) and then loaded with HBSS (100 μl/well) containing 2.5 μM Fura-2 AM and then incubated for 30 min at 37°C. Wells were then washed with HBSS and an additional incubation step was performed for 30 min at 37°C (95% air, 5% CO₂). Cells were then pretreated for 2 min with muscimol 1–100 μM, propofol 0.1–10uM or vehicle (0.1% DMSO) before the addition of 50nM endothelin-1 by use of the automatic injection feature of the FlexStation 3 microplate reader (Molecular Devices, Sunnyvale, CA) during continuous fluorescent measurements (excitation 340/380 nm and emission 510 nm). Real-time changes in intracellular Ca²⁺ are reported as RFU (relative fluorescence units).

Electrophysiologic recordings of vascular smooth muscle using an automated patch clamp system:

Human vascular smooth muscle cells were dispersed into Kreb’s buffer at a concentration of 5 million cells/ml and cells were made serum- and growth factor-free for 24 h prior to the assay (in basal M199 medium). Cells were loaded into a Syncropatch; 96 well automated patch clamp system (Nanion Munich, Germany) within a disposable loading plate with a medium resistance pore size. Cells were then placed in to the automated patch system and a whole cell configuration with a transient use of seal enhancement buffer. Cells were continuously analyzed for the quality of the seal resistance and capacitance. Cells were then placed under – 80 mV voltage clamp and treated with 100uM muscimol. The currents were recorded in 4 or 10 sec sweeps at a sample rate of 20kHz. Pretreatments were programmed into the automation, where 50uM picrotoxin was added in the external solution or low chloride solution replaced the internal solution. Treatments were performed on cells that initially responded to muscimol and had seals within >200 MOhms. (Internal solution in mM- KCl 110, NaCl 10, EGTA 20, HEPES10 (pH 7.2) External solution in mM- NaCl 140, KCl 4, MgCl₂ 1, CaCl₂ 2, Glucose 5, Hepes 10 (pH 7.4) Seal enhancer solution in mM- NaCl 80, KCl 3, MgCl₂ 10 CaCl₂, Hepes 10 (pH 7.4). Low chloride internal solution in mM- K gluconate 110, NaCl 10, EGTA 20, HEPES 10 (pH 7.2)

Resistance artery myogenic tone studies:

Arteries from mouse tail or human omentum were dissected, leaving adventitia and endothelium intact. Mouse arteries of less than 110 microns in diameter were used and human arteries less than 300 microns in diameter were used. Vessels were cannulated with glass micro pipettes (75–100um). The pipettes were mounted within a chamber that was continuously perfused with buffered solution ( buffer used in human studies in mM: 118 NaCl, 5.6 KCl, 2.5 CaCl₂, 2.4 MgSO₄, 1.3 NaH₂PO₄, 25 NaHCO₃, and 5.6 glucose, pH 7.4) ( buffer used in murine studies in mM: 115 NaCl, 2.5 KCl, 2.5 CaCl₂, 2.4 MgSO₄, 1.3 NaH₂PO₄, 25 NaHCO₃, and 5.6 D-glucose: pH 7.4.) that was oxygenate (95% O₂/ 5% CO₂) and temperature controlled (37°C). The micropipettes were connected to a peristaltic pump manually controlled to provide a constant intraluminal pressure. The wall diameter was monitored by video edge detection, measuring smooth muscle isotonic contraction or relaxation (Living systems St. Albans, VT, USA). The arteries were exposed to 3 cycles of an acute rapid increase in intraluminal pressures from 0 mmHg and held for two minutes at a higher intraluminal pressure; 90 mmHg for mice and 120 mmHg for human at which time the vessel diameter was measured. Once a control myogenic contraction was established, the resistance arteries were exposed to GABA_A modulators and re-challenged for 3 cycles with increasing luminal pressure (0–90 mmHg for mice and 0–120 mmHg for human). The 3^rd and 6^th myogenic challenge under control or experimental conditions were compared. Time (buffer) and vehicle controls (DMSO) were performed to verify the consistency between 3^rd and 6^th myogenic tone challenge. GABA_A modulators included muscimol (100uM; specific GABA_A agonist), gabazine (100uM; specific GABA_A antagonist), propofol (0.1uM IV anesthetic with GABA_A agonist effects) in 0.1% DMSO. Myogenic tone was measured at maximal constriction within the five minutes of increased intraluminal pressure. Studies were performed with 100uM gabazine alone to demonstrate the effects of GABA_A antagonism on baseline tone in mouse tail arteries. At an intraluminal pressure of 90mmHg, diameters before and after 5 minutes of treatment of gabazine were digitally collected. The pressures 90 mmHg and 120 mmHg for myogenic tone measurements were chosen based on two reasons. One, these pressures have been demonstrated in multiple prior studies to cause a large magnitude stretch-induced vessel contraction (myogenic tone)[24–27]. Secondly, intraluminal pressures in these higher ranges will achieve the force generation that occurs during hypertension, allowing for an assessment as to whether GABA_A receptors have a significant effect on myogenic tone within hypertension ranges.

Statistical analyses:

Each experimental permutation included intra-experimental vehicle controls. In the fluorescence membrane potential, calcium indicator and electrophysiology experiments, we employed one-way ANOVA with Bonferroni post-test comparisons between appropriate groups. In the myogenic tone studies, a 2-tailed paired Student’s t test was employed. Data were presented as means ± SEM; p < 0.05 in all cases was considered significant.

Results

RT-PCR survey of GABA_A subunits in human and mouse resistance arterial vascular smooth muscle:

To evaluate the expression profile of GABA_A receptor subunits in resistance arteries, mouse tail arterial smooth muscle samples and human omental arterial samples were analyzed for the expression of mRNAs encoding GABA_A subunits. (Figure 1) Laser capture microdissection (LCMD) was used to histologically identify and selectively capture vascular smooth muscle (Figure 1A). From these target tissue samples, RNA was extracted for RT-PCR. The PCR results demonstrate the expression of mRNA encoding the α3, α4, β2, γ2 and δ GABA_A subunits in mouse tail arterial (MTA) smooth muscle (Figure 1B) (n=3) and α4, β3, and δ GABA_A subunits in human omental artery (HOA) smooth muscle (Figure 1C) (n=4). All negative control water blank samples (no input RNA) did not yield any PCR products (data not shown) and all positive control brain (mouse and human, respectively) samples yielded PCR products of the expected molecular size. RT-PCR products using primers against GAPDH demonstrate the relative loading amounts of RNA from vascular smooth samples and brain samples in both human and mouse experiments (Figure 1B and C). GABA_A subunits α1, α2, α5, α6, β1, β3, γ1 and γ3 were not detected in RNA isolated from either mouse tail or human omental arteries (data not shown) despite detecting their expected cDNA products in the brain positive control tissues. To confirm the specificity of GABA_A subunit expression in resistance arterial vascular smooth muscle and to demonstrate the protein expression of the GABA_A α4 subunit, immunohistochemistry on human omental resistance arteries was performed. A longitudinally sectioned omental artery demonstrating dense brown staining representing GABA_A α4 protein expression was evident in the smooth muscle layer (Figure 1D) (n=3). The control serial section with primary antibody omitted was devoid of significant brown staining. (n=number of mice or patients)

Figure 1:

(A) Representative images of OCT embedded mouse tail artery sections. The smooth muscle layer was morphologically identified before (left image) RNA was captured from vascular smooth muscle cells using laser capture microdissection (LCMD). The area excised is indicated by the red arrow in the right image. (B and C) Representative gel images of RT-PCR products from LCMD-excised vascular smooth muscle from mouse tail arteries (MTA) (B) or human omental artery (HOA) (C) using primers targeting mRNA encoding different GABA_A subunits. RNA extracted from whole brain from mouse and human, respectively, served as positive controls. (B and C). Primers targeting GAPDH mRNA demonstrates the relative mRNA input from brain versus vascular smooth muscle. (D) A representative light microscopic image of a longitudinal section of human omental artery labeled with a primary rabbit antibody targeted against human GABA_A receptor α4 subunit (brown staining) (left panel). The vascular smooth muscle layer is denoted by a star. A control serial section of human omental artery in which the primary antibody was omitted during immunohistochemistry is shown (right panel). Black bar indicates 100 microns.

GABA_A receptor modulation of plasma membrane potential in arterial smooth muscle cells:

To demonstrate the function of GABA_A channels in primary vascular smooth muscle cells, cells were loaded with the FLIPR potentiometric dye. Propofol (100nM), an intravenous anesthetic with known GABA_A activity, induced a depolarizing change in membrane potential demonstrated by an increase in FLIPR fluorescence (**p<0.01 compared to 0.1% DMSO control; n=4) (Figure 2A and B). Furthermore, the change in membrane potential induced by propofol was inhibited by pretreatment with a GABA_A specific inhibitor, bicuculline (10uM)) (*p<0.05 compared to 100nM propofol; n=5), indicating that the change in membrane potential induced by propofol was via GABA_A receptor activation. KCl (40mM) was used as a positive control for depolarization (**p<0.01 compared to 0.1% DMSO control; n=8). NS1619 (potassium channel activator) (10uM) was used as a hyperpolarization control (*p<0.05 compared to 0.1% DMSO control; n=8) (Figure 2A and B). The data was analyzed by comparing the absolute maximal fluorescence change compared to the DMSO vehicle control.

Figure 2:

(A) Aggregate tracing of the change in fluorescence in primary cell cultures of human omental vascular smooth muscle cells loaded with potentiometric dye (FLIPR). Fluorescence was continuously measured as 40mM KCl, 100nM propofol, 100uM NS1619 or 0.1% DMSO was added. (B) Graphical representation of RFU changes during treatments of airway smooth muscle loaded with FLIPR dye. Propofol at 100nM demonstrated a 36 ± 3.2 ∆RFU (mean ± SEM) (**p<0.01 compared to control; n=4), which was significantly inhibited by 10uM of bicuculline (GABA_A specific inhibitor) (18 ± 1.3 ∆RFU (mean ± SEM)) (*p<0.05 compared to 100nM propofol; n=3). 40mM KCl induced a 53 ± 4.1 ∆RFU (mean ± SEM) (**p<0.01 compared to control; n=7). As a hyperpolarization control, 10uM NS1619 (potassium channel activator) was used and demonstrated a −27 ± 2.7 ∆RFU (mean ± SEM) (*p<0.05 compared to control; n=7). A vehicle control, 0.1% DMSO vehicle demonstrated a −2.5 ± 0.66 ∆RFU (mean ± SEM); n=7. (C) A representative tracing of intracellular calcium fluorescence in freshly dispersed human omental vascular smooth muscle cells pre-incubated with calcium ratiometric dye (Fura-2). Cells were stimulated with 50nM endothelin-1 at 50sec. In cells pretreated with vehicle (0.1% DMSO) for 2 minutes, endothelin-1 induced an acute increase in calcium-mediated fluorescence that was attenuated by pretreatment with either 100nM propofol or 100uM muscimol pretreatment. (D) Graphical representation of ratiometric fluorescence values of Fura-2 loaded human omental resistance arterial smooth muscle cells. Endothelin-1 in the presence of 0.1 % DMSO vehicle demonstrated a ratiometric increase of 0.38 ± 0.06 ∆RFU (mean ± SEM) (ex340/380) (n=3). Propofol at 10uM, 1uM and 0.1uM decreased endothelin responses to 0.15 ±0.10, 0.09 ± 0.01 and 0.15 ± 0.03 ∆RFU (mean ± SEM) (ex340/380), respectively (*p<0.05, **p<0.01 and *p<0.05, respectively, when compared to endothelin-1 alone; n=4). Muscimol at 100uM, 10uM and 1uM decreased endothelin-1 response to 0.11 ± 0.01, 0.11 ± 0.01 and 0.14 ± 0.01 ∆RFU (mean ± SEM) (ex340/380), respectively, (**p<0.01, **p<0.01 and *p<0.05 respectively when compared to endothelin-1 alone; n=4).

GABA_A receptor modulation of intracellular calcium in arterial smooth muscle cells:

To demonstrate the effects of GABA_A receptor activation on intracellular calcium responses in cultured human resistance arterial smooth muscle cells, cells were pretreated with the GABAergic anesthetic propofol or the classic GABA_A agonist muscimol before a subsequent challenge with endothelin-1, a known vasoconstrictor. Cells were treated with propofol (0.1uM10uM) or muscimol (1uM −100uM) for 2 minutes before treatment with endothelin-1 (50nM). Endothelin-1 alone in the presence of 0.1 % DMSO vehicle demonstrated a ratiometric increase of 0.38 ±0.06 ∆RFU (mean ± SEM) (n=3) indicating an increase in the intracellular concentration of calcium (Figures 2C and 2D). Pretreatment with propofol at 10uM, 1uM and 0.1uM decreased endothelin-1-induced calcium responses (p<0.05, p<0.01 and p<0.05 respectively compared to endothelin-1 alone; n=4) (n=4) (Figure 2D). The classic GABA_A receptor ligand, muscimol, at 100uM, 10uM and 1uM decreased the endothelin-1 calcium responses. (p<0.01, p<0.01, p<0.05, respectively, compared to endothelin-1 alone; n=4) (Figure 2D).

Electrophysiologic recordings of human arterial smooth muscle cells using an automated patch clamp system:

To quantify the electrophysiologic responses of GABA_A channel activation in human omental resistance arterial smooth muscle cells, electrical current recordings were performed using an automated patch clamp system. Primary cultured cells of resistance arterial smooth muscle demonstrated inward current responses to 100uM muscimol (Figure 3) (n=6). A muscimol response was demonstrated in 10% of cells patched in culture. To demonstrate the specificity of the muscimol response, the muscimol-induced currents were inhibited by the GABA_A specific inhibitor picrotoxin (p<0.05 compared to muscimol alone; n=3) (Figure 3A and C). Furthermore, to demonstrate that the GABA_A response was indeed a chloride current, the internal solution was changed to a low chloride composition (by replacing KCl with K gluconate) reducing the internal chloride concentration from 120mM to 10mM. This resulted in an equalization of the electrochemical gradient for chloride and a loss of muscimol induced current (p<0.05 compared to muscimol alone; n=3) (Figure 3B and C).

Figure 3:

Representative electrophysiologic recordings of human omental vascular cells using an automated patch clamp system. Cells were held in a voltage clamp configuration with a holding potential at −80mV. (A) Picrotoxin (50uM) inhibited muscimol (100uM)-induced currents, and (B) decreasing the chloride internal concentration from 120mM to 10mM also inhibited muscimolinduced currents. (C) Graphical representation of current induced by muscimol (669 ± 149 pA (mean ± SEM; n=6). Muscimol currents were significantly reduced in the presence of picrotoxin (94.7 ± 12.7 pA (mean ± SEM) (*p<0.05 compared to muscimol; n=3) and also reduced by low internal chloride concentrations (104 ± 26.2 pA (mean ± SEM) (*p<0.05 when compared to muscimol; n=3).

GABA_A ligands decreased myogenic tone in mouse tail resistance artery:

In order to study the physiologic effects of GABA_A activation in intact vessels, resistance artery myogenic tone studies were performed using pressure myography. Under control conditions (buffer alone), mouse tail resistance arteries demonstrated myogenic tone when challenged with an increase in intraluminal pressure from 0–90 mmHg (Figure 4A). The addition of propofol (0.1uM) to the buffer superfusing the arterial segment, decreased myogenic tone, resulting in an increased dilation (p<0.05) (n=3) (Figures 4A and 4B). The inhibition of myogenic tone by propofol was reversed by the addition of 100uM gabazine, a GABA_A specific inhibitor (p>0.05) (n=3) resulting in myogenic tone that was not significantly different from control (ns n=3) (Figures 4A and 4C), indicating that the propofol inhibition of myogenic tone is GABA_A mediated.

Figure 4:

(A) Representative vessel diameter tracings recorded during pressure myography in mouse tail resistance arteries. Pressure was raised from 0–90 mmHg which induced an initial dilation which induced a contraction (myogenic tone) as demonstrated in the control (buffer solution alone) tracing. Propofol pretreatment inhibited myogenic tone resulting in increased dilation. The GABA_A antagonist gabazine antagonized the propofol dilatory effects resulting in a control level of myogenic tone. (B): Quantitative analysis of myogenic tone expressed as a change in vessel diameter. In control studies a 0–90 mmHg intraluminal pressure challenge resulted in a mean (±SEM) diameter change of 26 ± 7.4 um, while propofol pretreatments blocked myogenic tone resulting in vasodilation and a larger change in vessel diameter (50 ± 5.8 um) (mean ± SEM) (*p<0.05; n=3). (C) Co-pretreatment with propofol and the GABA_A antagonist gabazine resulted in a change in vessel diameter (30 ± 5.8 um) (mean ± SEM) that was not different from control (31 ± 10 um) which indicated that the propofol inhibition of myogenic tone is GABA_A mediated. (ns. not significant; n=3). (D) Representative vessel diameter tracings during pressure myography in mouse tail artery (0–90 mmHg). Muscimol pretreatment inhibited myogenic tone and increased vessel dilation. (E) Quantitative analysis of myogenic tone under control conditions demonstrated a vessel diameter change of 22 ± 5.6 um (mean ± SEM). The GABA_A agonist muscimol inhibited myogenic tone resulting in an increased vessel diameter of 33 ± 7.5 um (mean ± SEM) (*p<0.05; n=3). (F) Results demonstrating no significant difference (not significant; n=3) between 0.1 % DMSO 19 ± 3.6 um (mean ± SEM) and buffered solution alone control 16 ± 2.8 um (mean ± SEM) on myogenic tone generation. (G) Demonstrating the effects of gabazine alone on baseline tone at 90mmHg. Gabazine treatment demonstrated no significant change to baseline tone 94.8 ± 11.4 um (mean ± SEM) at 90mmHg (ns n=3) when compared to buffered solution alone control 95.1 ± 10.5 um (mean ± SEM).

To further prove that GABA_A activation reduces myogenic tone, the GABA_A agonist muscimol (100uM) was added to the buffer superfusing the mouse tail arteries before the change in the intraluminal pressure from 0–90 mmHg. Muscimol inhibited myogenic tone, resulting in an increased diameter change, which was significantly dilated compared to the control (buffered solution alone) (Figure 4D and 4E) (n=3). Time controls were performed to demonstrate a maintenance of myogenic responses between the 3^rd rise in intraluminal pressure and the 6^th rise in intraluminal pressure. No significant differences were found between 3^rd pressure cycle and the 6^th pressure cycle in presence of 0.1% DMSO vehicle control (Figure 4F). A gabazine control was performed by measuring the diameter change at 90mmHg before and after 100uM gabazine was added. There was no significant change in the plateau phase of the myogenic response after gabazine treatment (Figure 4G).

The GABA_A agonist muscimol decreased myogenic tone in human omental resistance arteries:

To determine whether the effects of GABA_A activation demonstrated in mouse arteries are also present in human resistance arteries, myogenic tone studies were performed with freshly isolated human omental resistance arteries. Myogenic tone under control conditions (buffered solution alone) demonstrated myogenic function following an increase in intraluminal pressure from 0–120 mmHg. Pretreatment of human vessels with muscimol (100uM) inhibited myogenic tone resulting in an increased vessel dilation (p<0.05; n=3) (Figure 5A and 5B).

Figure 5:

(A) Representative diameter tracings recorded during pressure myography in human omental resistance arteries. Intraluminal pressure was increased from 0–120 mmHg inducing an initial dilation, which induces a control contraction (myogenic tone). Muscimol, a GABA_A specific agonist pretreatment inhibited myogenic tone and increased vessel dilation. (B): Quantitative analysis of myogenic tone under control conditions demonstrated a vessel diameter change of 38 ± 0.7 um (mean ± SEM). The GABA_A agonist muscimol inhibited myogenic tone resulting in an increased vessel 42 ± 0.7 um (mean ± SEM). (*p<0.05, n=3).

Discussion

The primary finding of the present study are (1) GABA_A subunits are expressed on smooth muscle cells from small caliber murine vessels and human resistance arteries; (2) targeting of GABA_A receptors on human resistance arterial smooth muscle modulates plasma membrane potential; and (3) myogenic tone is inhibited by GABA_A agonists in whole tissue models. In vascular smooth muscle (VSM), myogenic tone is a contractile mechanism that is directly controlled by membrane potential and depolarization. When certain small caliber arteries are faced with increased intraluminal pressure, stretch induced cation channels depolarize the vascular smooth muscle leading to activation of voltage-dependent calcium channels and TRP channels, resulting in calcium influx and a concomitant increase in intrinsic vascular smooth muscle tone. [28–30] Studies have demonstrated that the amount of depolarization in vascular smooth muscle directly correlates to the amount of calcium influx and the myogenic tone generated.[29] Given this established relationship between membrane depolarization and myogenic tone, we utilized an ex-vivo myogenic tone model as the physiologic correlate to explore the influences of GABAA channel activation on voltage dependent vascular smooth muscle tone. These studies describe for the first time GABAA-mediated decreases of myogenic tone in human vessels.

Myogenic tone is a property specific to small resistance arteries and increased myogenic tone is associated with an overall increased contractility of the VSM. When the luminal pressure increases, a myogenic response is initiated, which leads to vessel constriction to maintain a constant flow through the arteries. During pathologic states, such as hypertension, resistance arteries demonstrate an increase in myogenic tone in the cerebral and skeletal muscle vasculature.[31–33] However, the distinct pathophysiologic role of myogenic tone in hypertension is not well defined. Not all animal models of hypertension consistently show enhanced myogenic tone. In the rat model of sodium intake-induced hypertension, there seems to be a decrease in myogenic tone.[34] In contrast, in the spontaneously hypertensive rat (SHR) model, myogenic tone is increased in certain vascular beds.[31–33] The mechanisms that allow the cardiovascular system to compensate and modulate their myogenic tone may have influence over normal physiology and pathophysiology. By demonstrating the effects of GABAA activation on myogenic tone, we implicate a possible mechanism of endogenous vascular modulation of myogenic tone since GABA is circulating in blood and many peripheral cells are capable of synthesizing GABA. The effect of GABAA on vascular biology will require further studies in order to fully determine the role that GABA plays in vascular physiology and the potential therapeutic value of GABAergic targeting. In the current study, resistance vessels from human omentum were used as a model of myogenic tone. In order to make further conclusions on blood pressure regulation and the pathogenesis or treatment of hypertension, studies on other vascular tissue from other regions (i.e. mesenteric and renal) will need to be performed, recovered from patients carefully clinically phenotyped with regard to their cardiovascular status.

Since GABAA ligand selectivity is largely determined by the α subunit composition of the GABAA receptor, we performed RT-PCR analysis using laser capture microdissection (LCMD) to obtain discrete areas of smooth muscle from mouse tail and human omental resistance arteries. We detected mRNA encoding the α3, α4, β2, γ2 and δ GABAA subunits in vascular smooth muscle, while mRNA encoding for other GABAA subunits were not detected (α1, α2, x5, x6, β1, β3, γ1 and γ3). These results demonstrate an expression profile of GABAA α, β and δ/γ subunits in a sufficient repertoire needed to form a classic functional GABAA receptor. Data from RT-PCR analysis revealed that the specific α subunit composition of VSM GABAA (α4,α3) receptors did not include the predominant subunits found in the CNS (α1,α2), but rather an expression profile more consistent with extrasynaptic/peripheral GABAA receptors.[35] Extrasynaptic GABAA channels are characteristically responsive to lower concentrations of GABA and display slower desensitization properties (collectively termed “tonic”).[36] This current study describes the novel mRNA expression profile of the GABAA receptor subunits on vascular smooth muscle. Further studies need to be performed to further demonstrate the protein expression profile to fully correlate with the mRNA expression profiles and to understand the possible subunit stoichiometry of these expressed subunits.

Primary VSM cells loaded with a potentiometric fluorescent indicator demonstrated an increased fluorescence in the presence of propofol, a widely used clinical GABAergic anesthetic. Propofol had a greater depolarization effect on VSM than a classic GABAA agonist, muscimol. Previous studies have demonstrated that propofol causes a variety of molecular and physiologic responses in vascular smooth muscle cells other than GABAA activation, such as increases in smooth muscle TRP channel activation and modulation of vascular smooth muscle calcium at higher concentrations. [37, 38] To further specify and quantify the electrophysiologic response of GABAA activation on resistance artery physiology, intracellular current recordings were performed using the GABAA specific agonist muscimol on individual vascular smooth muscle cells. This demonstrated an inward current at negative membrane potential, and a loss of that current in the presence of a GABAA specific inhibitor or a decreased chloride gradient (10mM chloride in internal solution).

Only 10% of the smooth muscle had responses to GABA agonist exposure in the patch clamp system, while global measurements of the potentiometric indicator in a monolayer of cells revealed a more robust response. This is likely due to the potentiometric indicator allowing for the measurement of fluorescence changes within the entire field of cells in a 96 well plate, while the automated patch system measures a single cell in a dispersed culture. Cells grown to confluence in culture allows for cell to cell interactions such as gap junction formation. This GABA responsive subpopulation may also give rise to the possibility that there is differential expression within the vascular smooth muscle itself.

We further demonstrate that GABAA activators decreased the calcium responses elicited by endothelin. We demonstrate that at a membrane potential of −80mV an inward current is elicited by a GABAA receptor agonist. With an electrochemical equilibrium that is estimated at −40mV in systemic arteries[39], a tonic chloride flux through GABAA receptors would prevent the full activation of membrane potential gated channels, normally around 20mV, and inhibit calcium entry through voltage regulated calcium channels. Further studies demonstrating in situ tissue recordings will be needed to fully elucidate the mechanism of GABAA mediated loss of myogenic tone.

In summary, we describe the subunit expression profile of GABAA receptors in resistance arterial vascular smooth muscle cells. Furthermore we demonstrate that the GABAA α4 subunit protein is contained within the VSM layer of the resistance artery. We elicit a functional GABAA channel response by demonstrating the ability of the receptor to cause a membrane potential change when activated by an GABAA specific agonist. We demonstrate a physiologic response to the membrane potential changes caused by the GABAA activation in ex vivo myogenic tone studies in human and mouse resistance arteries. These studies indicate that functional GABAA receptors exist on resistance arteries and may be a viable therapeutic target for the treatment of hypertension.