γ-Aminobutyric acid (GABA) is the major inhibitory neurotransmitter in the mammalian central nervous system and exerts its actions via both ionotropic (GABAA) channels and metabotropic (GABAB) receptors. GABAA channels are ubiquitously expressed in neuronal tissues, and in mature neurons modulate an inward chloride current resulting in neuronal inhibition due to membrane hyperpolarization. In airway smooth muscle (ASM) cells, membrane hyperpolarization favors smooth muscle relaxation. Although GABAA channels and GABAB receptors have been functionally identified on peripheral nerves in the lung, GABAA channels have never been identified on ASM itself. We detected the mRNA encoding of the GABAA α4-, α5-, β3-, δ-, γ1–3-, π-, and θ-subunits in total RNA isolated from native human and guinea pig ASM and from cultured human ASM cells. Selected immunoblots identified the GABAA α4-, α5-, β3-, and γ2-subunit proteins in native human and guinea pig ASM and cultured human ASM cells. The GABAA β3-subunit protein was immunohistochemically localized to ASM in guinea pig tracheal rings. While muscimol, a specific GABAA channel agonist, did not affect the magnitude or the time to peak contractile effect of substance P, it directly concentration dependently relaxed a tachykinin-induced contraction in guinea pig tracheal rings, which was inhibited by the GABAA-selective antagonist gabazine. Muscimol also relaxed a contraction induced by an alternative contractile agonist histamine. These results demonstrate that functional GABAA channels are expressed on ASM and suggest a novel therapeutic target for the relaxation of ASM in diseases such as asthma and chronic obstructive lung disease.
- guinea pig
- organ bath
γ-aminobutyric acid (GABA) is the major inhibitory neurotransmitter in the mammalian central nervous system (CNS). GABA acts at two distinct types of receptors, pentameric ligand-gated ionotropic GABAA channels and G protein-linked metabotropic (GABAB) receptors. GABAA channels are ubiquitously expressed in neuronal tissues, and in mature neurons modulate an inward chloride current resulting in neuronal inhibition due to membrane hyperpolarization. GABAA channels are classically pentameric protein channels composed of combinations of subunit subtypes (α1-6, β1-3, γ1-3, δ, ε, π, θ), which dictate pharmacological and gating properties of this chloride channel in the mammalian brain. Although GABAA channels are ubiquitously expressed in the CNS, their expression in nonneuronal cells has received limited study. GABAA channels in the brainstem modulate cholinergic outflow to the lung (26, 37), and both GABAA channels and GABAB receptors have been pharmacologically identified on the presynaptic side of lung postganglionic parasympathetic nerves where they inhibit cholinergic nerve activity (14, 53, 55). Although we recently described the functional expression of GABAB receptors on airway smooth muscle cells (42), GABAA channels have not been previously identified on airway smooth muscle cells. Membrane hyperpolarization of the airway smooth muscle cell is known to reduce airway smooth muscle tone (45).
In the present study, we investigated the expression of GABAA channels in native guinea pig and human airway smooth muscle and cultured human airway smooth muscle cells and assessed the ability of a selective GABAA agonist to relax tachykinin- or histamine-induced airway smooth muscle contraction in guinea pig tracheal rings.
Cells were cultured in SmGM-2 smooth muscle medium (Biowhittaker, Walkersville, MD). Lysates of human brain cerebral cortex or thalamus used as a positive protein control on immunoblots were obtained from BD Biosciences (Palo Alto, CA). Total RNA from whole human brain was purchased from Clontech (Mountain View, CA). Protease inhibitor cocktail III was purchased from EMD Biosciences (San Diego, CA). All other chemicals were obtained from Sigma (St. Louis, MO) unless otherwise stated.
Primary cultures of human tracheal smooth muscle cells were obtained from lung transplant donors in accordance with procedures approved by the University of Pennsylvania Committee on Studies Involving Human Beings as previously described (43). Cells were grown in culture medium (SmGM-2, supplemented with 5% FBS, 5 μg/ml insulin, 1 ng/ml human fibroblast growth factor, 500 pg/ml human epidermal growth factor, 30 μg/ml gentamicin, and 15 ng/ml amphotericin B; Biowhittaker, Walkersville, MD) at 37°C in an atmosphere of 5% CO2-95% air. For immunoblotting, cells were grown to confluency in T75 flasks. Cells were rinsed with cold PBS and scraped in cold PBS. Cells were pelleted (500 g, 10 min, 4°C), and cells lysed in cold lysis buffer [50 mM Tris·HCl, pH 8.0, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 150 mM NaCl, 1 mM EDTA, 1:200 dilution of protease inhibitor cocktail III (EMD Biosciences, San Diego, CA), 1 mM Na3VO4, 1 mM NaF]. Lysed cells were centrifuged (15,000 g, 15 min, 4°C), and the supernatant was subjected to protein analysis and then solubilized by heating at 95°C for 5 min in sample buffer (final concentrations: 50 mM Tris·HCl, pH6.8, 2.5% SDS, 6% glycerol, 2.5% 2-mercaptoethanol, bromophenol blue) and stored at −20°C.
Isolation of smooth muscle from human trachea and guinea pig for RNA and protein analysis.
Studies were approved by Columbia University's Institutional Review Board and deemed not human subjects research under 45 CFR 46. Human trachea came from two sources. Snap-frozen tracheas obtained at autopsy from non-asthmatic adults within 8 h of death were obtained from the National Disease Research Exchange (Philadelphia, PA). Additional tracheae were obtained from discarded regions of healthy donor lungs harvested for lung transplantation at Columbia University. Lung transplant excess tissue was transported to the laboratory on ice and submerged in cold (4°C) Krebs-Henseleit (KH) buffer (in mM: 118 NaCl, 5.6 KCl, 0.5 CaCl2, 0.24 MgSO4, 1.3 NaH2PO4, 25 NaHCO3, 5.6 glucose, pH 7.4).
Adult male guinea pigs were deeply anesthetized by intraperitoneal pentobarbital (50 mg/kg), the chest cavity was opened, and the animal exsanguinated before the dissection of the trachea. The trachea or whole brain were surgically removed and placed in cold (4°C) KH buffer.
The exteriors of either the human or guinea pig trachea were meticulously dissected free of adherent connective tissue under a dissecting microscope. Tracheas were opened longitudinally along the anterior border, the epithelium was removed, and the airway smooth muscle between the non-contiguous ends of the cartilaginous tracheal rings was dissected free. Airway smooth muscle or whole brain was homogenized (Tekmar Ultra Turrax T25 high-speed homogenizer set at top speed for 30 s) in TRIzol reagent (Ambion, Austin, TX) for total RNA extraction according to the manufacturer's recommendations or in cold (4°C) buffer [50 mM Tris, 10 mM HEPES, pH 7.4, 1 mM EDTA with a 1:200 dilution of protease inhibitor cocktail III (EMD Biosciences)] for total protein isolation. The homogenate was filtered through 125-μM Nitex mesh and centrifuged twice at 500 g for 15 min. The supernatant was transferred into new tubes and centrifuged at 50,000 g for 30 min at 4°C. The final membrane pellet was resuspended in the same buffer for protein concentration determinations and stored at −80°C.
RNA isolation and RT-PCR.
Total RNA was extracted from freshly dissected native human or guinea pig airway smooth muscle, cultured human airway smooth muscle (HASM) cells, and guinea pig whole brain using TRIzol reagent (Ambion) according to the manufacturer's recommendations. Total RNA from whole human brain (Clontech, Mountain View, CA) was used as a positive control. Using the Advantage RT-for-PCR Kit (Clontech), 1 μg of total RNA was reverse transcribed at 42°C for 1 h in 20 μl including 200 units of Moloney murine leukemia virus reverse transcriptase, 20 units of RNase inhibitor, 20 pmol oligo(dT) primer, and 0.5 mM each of dNTP mix in reaction buffer (50 mM Tris·HCl, pH8.3, 75 mM KCl, 3 mM MgCl2).
PCR was performed by adding 5 μl of newly synthesized cDNA to a 45-μl reaction mixture yielding final concentrations of 0.2 mM of each dNTP, 1× Advantage 2 Polymerase Mix, PCR buffer (Clontech), and 0.4 μM of both sense and antisense primers for corresponding GABAA subunits (Tables 1 and 2). Two-step PCR (annealing and extension at same temperature) was performed for 1 min with a PTC-200 Pelitier thermal cycler (MJ Research, Waltham, MA) for all PCR reactions at indicated temperatures (Tables 1 and 2) except for guinea pig GABAA θ for which annealing (60°C/30 s) and extension (70°C/30 s) conditions were used. PCR conditions for all reactions included an initial denaturation step at 94°C for 1 min followed by 40 cycles of denaturation (94°C for 10 s) and annealing/extension at indicated temperatures for 1 min. PCR products were electrophoresed on 5% nondenaturing polyacrylamide gel in 1× Tris, acetate, and EDTA buffer. The gel was stained with ethidium bromide (Molecular Probes, Eugene, OR), visualized using ultraviolet illumination, and analyzed using Quantity One software (BioRad, Hercules, CA).
Western blot analysis.
Membrane lysates were electrophoresed (8–10% SDS-PAGE) and transferred to PVDF membranes. The PVDF membranes were blocked for 1 h at room temperature with 5% nonfat dry milk in TBS with 0.1% Tween 20 and were then probed with antibodies directed against GABAA α4 (rabbit polyclonal 1:1,000; LS-C15, LifeSpan BioSciences, Seattle, WA), GABAA α5 (rabbit polyclonal 1:500; GTX77692, Genetex, San Antonio, TX), GABAA β2/3 (mouse monoclonal 1:250; MAB341, Chemicon, Temecula, CA), and GABAA γ2 (rabbit polyclonal 1:1,000; AB5559, Chemicon) overnight at 4°C. After washing three times, membranes were incubated for 1 h at room temperature with HRP-conjugated secondary anti-rabbit antibodies (1:5,000; 81-6120, Zymed, South San Francisco, CA) or anti-mouse antibodies (1:5,000; NA931V, Amersham Biosciences, Arlington Heights, IL). The signal from the immunoreactive bands was detected by ECL Plus (Amersham Biosciences) according to the manufacturer's recommendations and developed on film (Kodak Biomax light film; Kodak, Rochester, NY).
Guinea pig tracheal rings were fixed with 10% formalin. Tracheal rings were paraffin embedded, sectioned (5 μm), dewaxed in xylene, and rehydrated in a graded alcohol series to water. Endogenous peroxidase was blocked in 0.3% hydrogen peroxide for 30 min at room temperature. Heat-mediated antigen retrieval was performed with 10 mM sodium citrate buffer, pH 6.0, for 30 min. Sections were blocked with 10% normal serum for 30 min. After washing with Tris-buffered saline solution with 0.1% Triton X-100 (TBST), an avidin biotin blocking kit (Vector Laboratories, Peterborough, UK) was used (in 10% serum in PBS) to block endogenous biotin. Slides were rinsed with TBST and incubated overnight at 4°C in primary antibody against GABAA β2/3 (mouse, MAB341, Chemicon) at a concentration of 1:100 in 2% normal horse serum in TBST. A tracheal ring section was incubated with the appropriate isotype IgG (mouse IgG1) antibody as a negative control. Following overnight incubation at 4°C, slides were washed three times with TBST, and primary antibodies were detected using biotinylated anti-mouse antibodies (Vector Laboratories) at a concentration of 1:100. After incubation with ABC-HRP complex (Vector Laboratories) for 30 min, the antigen antibody complex was then visualized with the enzymatic reduction of 3,3-diaminobenzidine tetrahydrochloride. Sections were counterstained with hematoxylin, dried, and coverslipped using Poly-mount (Polysciences, Warrington, PA).
In vitro effects of GABAA agonist on guinea pig airway smooth muscle relaxation.
All studies were approved by Columbia University's Institutional Animal Care and Use Committee. Force measurements were performed on closed guinea pig tracheal rings suspended in organ baths as previously described (30). Briefly, Hartley male guinea pigs (∼400 g) were anesthetized with 50 mg/kg ip pentobarbital; the tracheas were removed promptly, dissected into closed rings composed of two cartilaginous rings from which mucosa, connective tissue, and epithelium were removed. Silk threads were tied to the rings such that the threads were at each end of the posterior aspect of the ring (lacking in cartilage), ∼180° from one another. One thread was attached to a fixed point at the bottom of 4-ml organ baths (Radnoti Glass Technology, Monrovia, CA), and the opposing thread was attached to a Grass FT03 force transducer (Grass-Telefactor, West Warwick, RI) coupled to a computer via BioPac hardware and Acknowledge 7.3.3 software (Biopac Systems, Goleta, CA) for continuous digital recording of muscle tension. The rings were suspended in 4-ml organ baths filled to the top with 4 ml of KH buffer solution (composition in mM: 118 NaCl, 5.6 KCl, 0.5 CaCl2, 0.2 MgSO4, 25 NaHCO3, 1.3 NaH2PO4, 5.6 d-glucose) with 10 μM indomethacin (DMSO vehicle final concentration of 0.01%) added and gassed with 95% O2 and 5% CO2 at pH 7.4 at 37°C. The rings were equilibrated at 1 g of isotonic tension for 1 h with new KH buffer added every 15 min. Rings were pretreated with 10 μM capsaicin to induce contractions and deplete endogenous nonadrenergic, noncholinergic nerves of endogenous tachykinins. Following spontaneous relaxation (∼20 min), baths and rings were washed with three buffer changes, resting tension was reset to 1 g, and all organ baths were treated with 1 μM tetrodotoxin (to block endogenous nerve activity). In experiments where substance P was the contractile agonist, the organ baths were also pretreated with 10 μM pyrilamine (to block potential effects of released histamine from resident mast cells in the preparation). In experiments where histamine was the contractile agonist, the organ baths were also pretreated with 1 μM atropine (to block histamine effects of stimulating cholinergic nerves). During the following contractile challenges, no further changes in buffer occurred, and the stated concentrations of substance P, histamine, muscimol, or gabazine represent the final concentrations of each reagent in the buffer within the organ baths.
Rings were then contracted with 1 μM substance P. Following the achievement of a stable contraction (typically 10 min), cumulatively increasing concentrations of the GABAA-selective agonist muscimol (0.1 μM-1 mM) were incrementally added to the buffer in the baths at 7-min intervals. Control-contracted rings received vehicle (KH buffer) to serve as time controls for the muscimol-induced relaxation. In separate experiments, rings were pretreated with 100 μM gabazine (GABAA-selective antagonist) 15 min before the addition of 1 μM substance P. Following the achievement of a stable contraction, a single concentration of the GABAA-selective agonist muscimol (100 μM) was added to assess effects on airway smooth muscle force.
In additional separate experiments, guinea pig tracheal rings were contracted with 10 μM histamine. Following the achievement of a stable contraction (typically 10 min), muscimol (100 μM) was added to the organ baths. Control-contracted rings received vehicle (KH buffer) to serve as time controls for the muscimol-induced relaxation.
Statistical analysis was performed using repeated measures of ANOVA, followed by Bonferroni posttest comparison using Prism 4.0 software (GraphPad, San Diego, CA). Data are presented as means ± SE; P < 0.05 was considered significant.
RT-PCR analysis of GABAA channel subunits in airway smooth muscle.
Initially, we assessed the expression of mRNA encoding GABAA channel subunits in at least three independent samples each of freshly isolated human and guinea pig airway smooth muscle and in primary cultures of human airway smooth muscle cells. Total RNA from whole brain was used as positive controls in each species. Freshly dissected native human and guinea pig airway smooth muscle expressed mRNA encoding multiple GABAA channel subunits (α4, α5, β3, γ2, γ3, δ, π, and θ) (Figs. 1 and 2). Multiple other subunits (α1, α2, α3, α6, β1, β2, γ1, ε) were not detected in freshly dissected airway smooth muscle from either human or guinea pig upper airways despite their detection in control brain RNA from both species (Figs. 1 and 2). Primary cultures of human airway smooth muscle cells expressed similar but not identical patterns of GABAA subunits. GABAA α4-, α5-, β3-, γ2-, δ-, and θ-subunits were also detected in cultured human airway smooth muscle, but GABAA γ3 and π were not detected (Fig. 1, D, E, I, K, P). Moreover, GABAA γ1, which was not detected in freshly isolated airway smooth muscle, was detected in primary cultures of airway smooth muscle (Fig. 1J). Additional semiquantitative experiments were performed for those subunits detected in fresh but not cultured human airway smooth muscle to determine if their mRNA detection was due to mRNA contamination from a minor cell type other than airway smooth muscle in these freshly dissected samples. RNA isolated from human brain and freshly dissected human airway smooth muscle contained similar quantities of mRNA encoding GABAA γ3 (Fig. 1M) and GABAA π (Fig. 1R).
Two splice variants of the human GABAA α4- and four splice variants of the human GABAA γ2-subunit proteins are expressed in human tissues (28) (Fig. 3), and we used primer sets designed to distinguish between the two α4 variants or to distinguish the γ2L variant from the other three γ2 variants (γ2S/γ2XL/γ2 variant 3) (Figs. 1 and 3 and Table 1). We detected variants 1 and 2 of the human GABAA α4-subunit in both freshly isolated and cultured human airway smooth muscle (Fig. 1D). Two variants of the GABAA γ2-subunit were detected in both human (Fig. 1K) and guinea pig (Fig. 2E) freshly dissected airway smooth muscle, and the larger variant 1 (γ2L) of the GABAA γ2-subunit was detected in primary cultures of human airway smooth muscle (Fig. 1K).
Immunoblot analysis of GABAA channel subunits in airway smooth muscle.
Lysates prepared from freshly dissected native human airway smooth muscle, freshly dissected native guinea pig airway smooth muscle, and cultured human airway smooth muscle cells were subjected to immunoblot analysis using selected specific anti-GABAA subunit antibodies (Fig. 4). Immunoreactive bands of appropriate molecular mass for GABAA α4-, α5-, β3-, and γ2-subunits were identified in airway smooth muscle. The antibody used to detect GABAA α4 is capable of identifying two splice variants of α4. A single band of the appropriate molecular mass for GABAA α4 (∼64 kDa) was identified in freshly isolated airway smooth muscle from human and guinea pig, primary cultures of human airway smooth muscle cells, and in guinea pig whole brain (Fig. 4A). The band for GABAA α4 detected in cultured HASM cells migrated at a slightly lower-molecular-mass than the other samples, suggesting the possibility that a smaller splice variant was predominantly expressed in HASM. A single band of the appropriate molecular mass for GABAA α5 (∼55 kDa) was identified in whole human brain and freshly isolated human airway smooth muscle and in a separate series of experiments in human thalamus and primary cultures of human airway smooth muscle cells (Fig. 4B). The antibody used to detect GABAA β2/3 identified a protein with an appropriate molecular mass of ∼54 kDa in airway smooth muscle freshly dissected from human or guinea pig tracheal smooth muscle and primary cultures of human airway smooth muscle cells. In guinea pig brain, an additional protein band of slightly lower-molecular-mass was also identified, as has been shown previously in brain using different anti-GABAA β3 antibodies (51, 56) including the antibody used in the present study (mouse monoclonal BD17) (20) (Fig. 4C). The antibody used to detect GABAA γ2 recognizes the cytosolic loop region of two splice variants of GABAA γ2. We resolved two immunoreactive bands of ∼45 and ∼47 kDa in whole human brain cortex and cultured human airway smooth muscle cells (Fig. 4D). In contrast, a single band corresponding to ∼47 kDa was identified in freshly dissected airway smooth muscle from human and guinea pig trachea. Whereas the larger 47-kDa band corresponds to the expected molecular mass of GABAA γ2L, the 45-kDa band could represent splice variants of GABAA γ2S (25).
Immunohistochemical detection of GABAA β3-subunit expression in guinea pig airway smooth muscle.
To confirm the localization of GABAA protein to airway smooth muscle, immunohistochemistry was performed using a monoclonal antibody that recognizes both the GABAA β2- and GABAA β3-subunit proteins in paraffin sections of guinea pig tracheal rings. Specific immunoreactivity was detected in the airway smooth muscle layer of the guinea pig trachea (indicated by brown color) with no staining in the negative control sections using a isotype-specific mouse IgG1 control primary antibody (Fig. 5). Higher magnifications (×100 oil objective) revealed homogenous staining along muscle cells consistent with cell surface expression (Fig. 5, bottom left).
Functional studies of GABAA receptor function in intact guinea pig airway rings.
Molecular identification of multiple subunits of GABAA receptors led us to question whether functional GABAA channels could modulate airway smooth muscle tone. Guinea pig tracheal rings suspended in organ baths and pretreated with capsaicin, tetrodotoxin, and pyrilamine demonstrated muscimol-induced relaxation in a concentration-dependent manner (0.1–100 μM) of a substance P-induced (1 μM) contraction (Fig. 6). Under identical conditions, the pretreatment of rings with the GABAA antagonist gabazine (100 μM) significantly attenuated the relaxation induced by 100 μM muscimol (Fig. 7). The pretreatment of guinea pig tracheal rings with 100 μM gabazine did not influence either the rate at which peak contraction with substance P was achieved (672 ± 30 s vs. 654 ± 43 s, in the absence or presence of gabazine pretreatment, respectively, n = 6) or the absolute magnitude of the peak contraction achieved with substance P (0.73 ± 0.10 g vs. 0.68 ± 0.11, in the absence or presence of gabazine pretreatment, respectively, n = 6).
In separate studies, the ability of muscimol to relax a different contractile agonist was assessed in guinea pig rings. Following a stable increase in muscle force achieved with 10 μM histamine, 100 μM muscimol significantly relaxed smooth muscle force in tracheal rings measured at 10 min (Fig. 8).
The primary findings of the present study include the first demonstration of GABAA subunit mRNA and protein expression in human and guinea pig airway smooth muscle cells and the ability of a GABAA-selective agonist to relax a tachykinin- or histamine-induced contraction in airway smooth muscle.
GABA receptors include both ionotropic (GABAA channels) as well as metabotropic (GABAB) receptors. We have recently demonstrated the expression and functional coupling of GABAB receptors to the Gi protein in human airway smooth muscle cells (42). The GABAA receptors are pentameric ligand-gated ion channels that conduct chloride currents, which in mature neurons results in hyperpolarization of the cell membrane impeding the effect of depolarizing (i.e., stimulatory) signals. GABAA receptors are classically pentamers composed of combinations of subunit subtypes (α1-6, β1-3, γ1-3, δ, ε, π, θ) that dictate pharmacological and gating properties of this chloride channel in the mammalian brain (8, 29, 34, 35). This diversity is further extended by splice variants for multiple subunit subtypes (6, 32). Despite the enormous combinational possibilities, most GABAA receptors in the CNS are composed of α-, β-, and γ-subunits in a 2:2:1 ratio (4, 12).
GABA receptors are ubiquitously expressed in the central nervous system, and the modulation of neuronal activity by GABA has been extensively studied. GABA receptors are also expressed in the peripheral nervous system where they also serve an inhibitory function (10, 57). In contrast, the native expression of GABA receptors in nonneuronal cells has received limited study. Initial attempts to survey the expression of GABA receptors outside the central nervous system relied on RT-PCR analysis of RNA isolated from whole peripheral organs (1, 9, 17, 41). Although these studies suggested ubiquitous expression of many GABA receptor subunits, it is unknown what specific cellular components of these tissues were expressing GABA receptor subunits. It is possible that these studies identified GABA receptor subunits expressed in peripheral nerves contained within these organs. More specific expression of GABA receptor subunits were identified in neuroendocrine cells including pancreatic β (54, 59), pituitary (48), and adrenal cells (44). Indirect pharmacological evidence suggested GABA receptor expression in vascular (22), bladder (19), uterine (2, 17, 21), and gut (47) smooth muscle in addition to the identification of GABA receptors in the peripheral nerves that innervate these tissues (5, 15, 46, 50). The expression of GABA receptors in these smooth muscles of gut, bladder, vascular, and uterine smooth muscle was inferred from pharmacological responses as opposed to a direct molecular identification of GABA receptors within the smooth muscle (2, 17, 19, 21, 22, 39, 47). Subsequently, subunits of GABA receptors have been demonstrated in heart (16), uterus (27, 40), kidney (49), liver (18, 52), and fibroblasts (24).
GABAA channels in the brainstem modulate cholinergic outflow to the lung (26, 37), and both GABAA channels and GABAB receptors have been pharmacologically identified on the presynaptic side of lung postganglionic parasympathetic nerves where they inhibit cholinergic nerve activity (14, 53, 55). Conversely, GABA receptors have never been described in airway smooth muscle itself. It has been known for some time that GABAB-specific agents decrease electrically field-stimulated airway constriction by modulating acetylcholine release from parasympathetic nerves (13, 14, 33, 53, 55). This is mediated by a presynaptic inhibition of acetylcholine release by GABAB receptors.
Previous studies have discounted a role for GABA receptors in the direct modulation of airway smooth muscle function, but these studies have typically been performed with GABA rather than a GABAA-selective agonist (14, 53, 55). Simultaneous activation of smooth muscle GABAA and GABAB receptors may not allow for relaxation effects of GABAA receptors to be elucidated. Two previous studies have failed to show an effect of the GABAA-selective agonist muscimol on basal tone (14, 53, 53) or acetylcholine-induced contraction (14) in guinea pig airway rings. However, these studies were limited to a single contractile agonist (acetylcholine) and more importantly did not address the potential ability of muscimol to relax contracted tissue as opposed to muscimol's ability to impair an initial contraction.
In neural tissues, activation of the GABAA receptor in mature neurons results in an inward chloride current and membrane hyperpolarization. In airway smooth muscle cells, membrane hyperpolarization is one mechanism that contributes to decreased intracellular calcium and decreased muscle tone (31). Hyperpolarization of the airway smooth muscle cell membrane by GABAA-mediated chloride entry would be expected to contribute to a decrease in airway tone by reducing intracellular calcium, reducing the Ca2+/calmodulin-dependent activation of myosin light chain (MLC) kinase, resulting in reduced phosphorylation of MLC, reduced interactions of actin and myosin, and less muscle contraction.
The current study demonstrates that airway smooth muscle cells express a limited repertoire of GABAA subunits and that there is remarkable agreement in subunit expression between the species studied (guinea pig and human). Identification of mRNA in freshly isolated human or guinea pig airway smooth muscle ensures that native expression of subunits is being examined. However, even with careful dissection, freshly isolated tissue is contaminated by other cell types that could give rise to cDNA products during sensitive RT-PCR amplification. Therefore, in all studies, we also evaluated the expression of GABAA subunits in established primary cultures of human airway smooth muscle cells that are homogenous populations of airway smooth muscle cells. Despite their cellular purity, cultured cell systems can result in an altered phenotype of protein expression. Despite these limitations, there was remarkable agreement but some differences between the expression of GABAA subunits in freshly isolated vs. cultured human airway smooth muscle. We detected mRNA encoding the α4-, α5-, β3-, γ2-, δ-, and θ-subunits in all sources of airway smooth muscle studied, including human and guinea pig freshly isolated airway smooth muscle and from cultured human airway smooth muscle cells. Additionally, multiple mRNAs encoding splice variants of the GABAA α4-subunit were detected in freshly isolated vs. cultured airway smooth muscle.
In contrast, the expression of the γ1-, γ3-, and π-subunits differed between freshly isolated and cultured airway smooth muscle. The GABAA γ3- and θ-subunits were expressed in freshly isolated airway smooth muscle from both human and guinea pig, but their expression did not persist in cultured human airway smooth muscle cells. One interpretation of these findings could be that the detection of these subunits was as a result of amplification from a nonmuscle cell type in the dissected tissue. To evaluate this possibility, equal amounts of total RNA from human brain (an abundant source of GABAA γ3 and θ mRNA) and freshly dissected human airway smooth muscle were analyzed in parallel during increasing numbers of PCR cycles. If the GABAA γ3 or θ cDNA products were arising from a minor cellular contaminant in the airway smooth muscle dissection, it would be expected that the cDNA product would be very small relative to the cDNA product arising from human brain where GABAA γ3 and θ mRNA expression is extensive. This was done to make a relative, not an absolute, comparison between mRNA expression in brain vs. airway smooth muscle. Indeed, the relative band intensity of the cDNA product was equivalent from freshly dissected human airway smooth muscle and the brain control, suggesting that the mRNA expression arising from freshly dissected human and guinea pig airway smooth muscle was truly arising from the predominant cell type in the dissected tissue, smooth muscle. The decreased expression of the GABAA γ3- and θ-subunits likely represents a phenotypic switch of these particular subunits during cell culture. In contrast, the GABAA γ1-subunit was not identified in freshly isolated guinea pig or human airway smooth muscle but was readily detected in cultured airway smooth muscle cells. This raises the interesting possibility that there is a phenotypic switch from the GABAA γ3- to the GABAA γ1-subunit in cultured human airway smooth muscle cells, the functional consequence of which is unknown.
Identification of mRNA encoding multiple GABAA subunits in airway smooth muscle led us to investigate GABAA subunit protein expression by immunoblotting and immunohistochemistry where suitable antibodies were available. In agreement with our mRNA analysis, the GABAA α4 protein was identified in airway smooth muscle from human and guinea pig airway smooth muscle, cultured airway smooth muscle cells, and the positive control guinea pig brain. The GABAA α5 protein was identified in freshly dissected and primary cultures of human airway smooth muscle and in human brain thalamus controls. In addition to a protein with an appropriate molecular mass of 54 kDa, the antibody used to detect GABAA β2/3 also identified a protein band of slightly lower molecular mass. Similar results with multiple different anti-GABAA β3 antibodies (51, 56), including the antibody used in the present study (mouse monoclonal BD17) (20), have been obtained in brain yet the identity of this smaller immunoreactive band is unknown (20, 51, 56). The antibody used to detect GABAA γ2 recognizes the cytosolic loop region of two splice variants of GABAA γ2. The long splice variant (γ2L) differs from the short isoform (γ2S) by the presence of the eight-amino acid insert LLRMFSFK, which bears the PKC phosphorylation site, Ser343 (36, 38, 58). We resolved two immunoreactive bands of ∼45 and ∼47 kDa in whole human brain cortex and cultured human airway smooth muscle cells. In contrast, a single band corresponding to ∼47 kDa was identified in freshly dissected airway smooth muscle from human and guinea pig trachea. Whereas the larger 47-kDa band corresponds to the expected molecular mass of GABAA γ2L, the 45-kDa band could represent different splice variants of GABAA γ2S (25, 28). The smooth muscle-specific expression of GABAA channels was further confirmed by the immunohistochemical detection of the GABAA β2/β3-subunit in guinea pig airway smooth muscle.
The molecular identification of GABAA subunits in airway smooth muscle cells led us to question whether these subunits formed functional channels. GABAA subunits in neuronal cells are activated by their endogenous ligand GABA as well as by synthetic selective ligands such as muscimol or 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-01 (THIP). Activation of neuronal GABAA channels in mature neurons results in an inward chloride current and hyperpolarization of the cell membrane. Hyperpolarization of the plasma membrane in airway smooth muscle cells is a component of airway muscle relaxation and is believed to be a partial mechanism by which classic relaxant agonists (i.e., β-adrenoceptor agonists) mediate their effects. Therefore, we questioned whether a GABAA agonist could potentiate the relaxation of intact airway smooth muscle. The spontaneous relaxation of tachykinin-contracted guinea pig airway smooth muscle was potentiated by the GABAA-selective agonist muscimol in a concentration-dependent manner, and this effect was attenuated by the GABAA-selective antagonist gabazine. However, 100 μM gabazine pretreatment did not completely block the effect of 100 μM muscimol, and there may be multiple reasons for this. It is possible that targets other than the GABAA receptor are involved in the muscimol effect; however, to date, no other nonspecific targets of muscimol have been identified in any tissue. Second, the efficiency with which gabazine penetrates tracheal rings suspended in organ baths is unknown, and its stability over time and the relative potency of gabazine (100 μM) to that of muscimol (100 μM) is unknown in this tissue. Moreover, there is evidence that gabazine may have selectivity for tonic but not phasic GABAA receptor activation (3, 7). Indeed, in this manuscript, we have identified the expression of the subunits in airway smooth muscle known to be required for tonic GABAA inhibition [α4 and δ in the thalamus (11) as well as α5 and δ in the cerebellum (23)] as well as subunits classically associated with phasic GABAA activation. Therefore, it is possible that our partial blockade by gabazine represents gabazine blockade of a subset of channels in our tissue that modulate tonic but not those that modulate phasic GABAA activity. This hypothesis will require further studies.
In summary, we demonstrate for the first time the molecular expression of multiple GABAA subunits in human and guinea pig airway smooth muscle. We demonstrate that a selective GABAA agonist can relax intact airway smooth muscle contracted with two different contractile agonists (substance P or histamine). These studies suggest a novel pharmacological target for the relaxation of airway smooth muscle.
This work was supported by National Institutes of Health Grants GM-065281 (C. W. Emala) and NS-45718 (J. Yang).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society