Expression and function of cystic fibrosis transmembrane conductance regulator in rat intrapulmonary arteries

The cystic fibrosis transmembrane conductance regulator (CFTR) gene encodes a cyclic adenosine monophosphate (cAMP)-dependent chloride channel located mainly at the apical membrane of epithelial cells. In myocytes of pulmonary arteries, numerous chloride channels have been identified and described, but not the CFTR. Thus the presence and function of the CFTR was investigated in rat intrapulmonary arteries. CFTR expression, localisation and function were analysed in cultured smooth muscle cells using Reverse transcriptase (RT)-PCR and immunoprecipitation followed by protein kinase A phosphorylation, immunolocalisation and an iodide efflux assay, respectively. The role of the CFTR in pulmonary vasoreactivity was determined in arterial rings using an organ bath system. RT-PCR and immunoprecipitation analyses, as well as the immunolocalisation study, revealed the expression of CFTR gene transcripts and protein. The iodide efflux assay showed the existence of functional cAMP-, calcium- and volume-dependent chloride channels. Furthermore, the following effects were found: 1) inhibition of forskolin/genistein-activated iodide efflux by glibenclamide, diphenylamine-2-carboxylic acid and CFTR-specific inhibitor (CFTRinh)-172; 2) activation of iodide efflux by the benzoquinolizinium derivative CFTR activators MPB-07 and MPB-91; and 3) inhibition of MPB-dependent efflux by CFTRinh-172. Finally, CFTR activators induced concentration-dependent vasorelaxation in rings preconstricted with phenylephrine, in the presence or absence of endothelium. The present results are the first to reveal functional cyclic adenosine monophosphate-regulated cystic fibrosis transmembrane conductance regulator contributing to endothelium-independent vasorelaxation in rat intrapulmonary arterial myocytes.

T he cystic fibrosis transmembrane conductance regulator (CFTR) is a cyclic adenosine monophosphate (cAMP)-dependent chloride channel expressed at the apical membrane of epithelial cells lining the tracheobronchial tree and the lumen of the digestive tract [1]. The CFTR was generally regarded as specifically expressed in epithelial cells until evidence for CFTR expression in nonepithelial tissues emerged. CFTR is expressed in cardiac muscle cells [2], brain [3] and endothelia [4], and has recently been found in tracheal smooth muscle cells (SMCs) [5] and aortic SMCs of rats and mice [6,7]. Despite this CFTR expression profile, the clinical picture of patients suffering from the CFTR-related disease cystic fibrosis appears to be unrelated to cardiac, vascular and brain dysfunction. Since truly selective activators and inhibitors are also lacking, the role of the CFTR in these tissues remains unresolved. However, the pharmacology of the CFTR has recently progressed, making available several CFTR activators, such as genistein and benzoquinolizinium derivatives [6][7][8], as well as blockers, such as the thiazolidinone compound CFTR-specific inhibitor 3-((3-trifluoromethyl)phenyl)-5-((4-carboxyphenyl) methylene)-2-thioxo-4-thiazolidinone (CFTRinh-172) [9].
Several Clconductances have been described in smooth muscle, including the pulmonary artery. Among the Clchannels expressed in SMCs, extensive studies have explored the implication of calcium-activated Clcurrents (ICl,Ca), evoked by a rise in intracellular calcium concentration ([Ca 2+ ] i ) [10]. Although the molecular nature of the ICl,Ca is still unknown, its role in resting membrane potential and vascular tone has been described for pulmonary artery [11][12][13].
Regarding the volume-sensitive Clchannel (ICl,swell), members of the Clchannel (ClC) gene family appear to be good molecular candidates for ICl,swell, and ClC-3 was recently proposed to underlie ICl,swell in rat and canine pulmonary arteries [14]. Interestingly, ClC-3 is upregulated in pulmonary hypertensive rats [14]. It is also noteworthy that the contribution of Clchannels is enhanced in basal tone and norepinephrine-induced contraction in pulmonary hypertensive rats [15,16].
A cAMP-dependent Clchannel has also been described in rat and bovine pulmonary arteries [17,18]. This channel is implicated in cAMP-induced pulmonary vasodilation, pulmonary arterial SMC (PASMC) migration and morphological changes. However, the molecular identity of such a channel is still unknown in pulmonary artery and could be related to the CFTR.
Despite the fact that the CFTR is present in aortic SMCs [6,7], and that Clchannels are important factors in the physiology as well as in the pathophysiology of the pulmonary artery, no studies have investigated the expression and functional role of the CFTR in pulmonary artery to date. Consequently, the present study focused on intrapulmonary artery (IPA) and investigated the expression of the CFTR in this vessel, and then explored cAMP-dependent iodide efflux, its pharmacology and the effect of CFTR activators on pulmonary vasorelaxation.

Tissue preparation
Male Wistar rats (aged 8-10 weeks) (Janvier, Le Genest-Saint-Isle, France) were stunned and then killed by cervical dislocation according to the local animal care and use committee (agreement number AP2/11/2005 from the regional ethics committee of Aquitaine/Poitou-Charentes, France). The heart and lungs were removed and placed in Krebs-Henseleit (KH) solution, which comprised 118.4 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl 2 , 1.2 mM MgSO 4 , 1.2 mM KH 2 PO 4 , 25 mM NaHCO 3 and 11.1 mM D-glucose (pH 7.4), saturated with 15% oxygen/5% carbon dioxide/80% nitrogen. IPA of the first order from the left lung was dissected free from surrounding connective tissues in KH solution.

Cell culture
The entire heart-lung preparation was rapidly removed and rinsed in culture medium (Dulbecco's modified Eagle medium/HEPES (pH 7.3) supplemented with 1% penicillin-streptomycin, 1% sodium pyruvate and 1% nonessential amino acids). IPA was dissected in culture medium under sterile conditions and cut into several pieces (1-2 mm 2 ). SMCs were obtained from these explants and cultured as previously described [5]. All cells from these explants were immunostained using the monoclonal antibody anti-smooth muscle aactin (Sigma-Aldrich, Saint-Quentin Fallavier, France), whereas they were negatively labelled with the endothelial nitric oxide synthase antibody (BD Transduction Laboratories, Le Pont-de-Claix, France; data not shown), demonstrating the presence of a population of SMCs.

Isometric tension measurements
The effect of CFTR activators on mechanical activity were measured using IPA rings (1.5-2.5 mm long) as previously reported [20]. In brief, KH solution, saturated with 15% oxygen/ 5% carbon dioxide/80% nitrogen, was used. Mechanical properties were assessed using an organ bath and transducer systems (EMKA Technologie, Paris, France), coupled to IOX software (EMKA Technologie) in order to facilitate data acquisition and analysis. As determined in preliminary experiments, tissues were set at optimal length by equilibration against a passive load of 0.8 g. IPA rings were then washed and precontracted with 0.3 mM phenylephrine (PHE) in order to test the relaxant properties of CFTR activators (10-chloro-6hydroxybenzo[c]quinolizinium chloride (MPB-07), 5-butyl-10chloro-6-hydroxybenzo[c]quinolizinium chloride (MPB-91) and 10-fluoro-6-hydroxybenzo[c]quinolizinium chloride (MPB-80); 3-300 mM, prepared as described previously [21]) by constructing a cumulative concentration-response curve. Some experiments were performed in endothelium-denuded rings. Endothelium was removed by perfusing the lumen of the vessels with a solution containing 0.3% 3-((3-cholamidopropyl) diethylammonio)-1-propane sulphonate (CHAPS), followed by washout with the drug-free solution, as previously described [22]. The effect of CHAPS was confirmed by the absence of relaxation with 10 mM carbamylcholine of 0.3 mM PHE-induced precontraction. All experiments were performed at 37uC.
Iodide efflux CFTR Clchannel activity was assayed by measuring the rate of 125 I efflux from cultured cells as previously described [6,7]. Cells were washed with efflux buffer, which comprised 136.9 mM NaCl, 5. In order to stimulate volume-sensitive Cltransport, the osmolarity of the efflux buffer was reduced from 300 to 150 mOsm?L -1 . Cells incubated in efflux buffer containing Na 125 I (New England Nuclear, Boston, MA, USA; 37 kBq?mL -1 ) for 1 h at 37uC were then washed with efflux medium in the presence or absence of 0.3 mM PHE to remove extracellular 125 I. Loss of intracellular 125 I was determined by removing the medium with the efflux buffer every minute for up to 8 min. The first three aliquots were used to establish a stable baseline in efflux buffer alone. Medium containing the appropriate drug was used for the remaining aliquots. The fraction of the initial intracellular 125 I lost at each time-point was determined, and time-dependent rates of 125 I efflux were calculated thus: where 125 It is the intracellular 125 I concentration at time t, and t 1 and t 2 are successive time-points. Curves were constructed by plotting the rate of 125 I efflux versus time. All comparisons were based on the maximal time-dependent rate (k; in min -1 ), excluding the points used to establish the baseline (kpeakkbasal).

Statistical analysis
Data are presented as the mean¡SEM of n observations, or the number of rings for the tension recordings. Sets of data were compared using ANOVA or an unpaired t-test. Differences were considered significant when p,0.05.

RESULTS
Expression of CFTR in IPA CFTR expression in IPA was revealed using three experimental approaches. First, the presence of CFTR mRNA was detected by reverse transcriptase-PCR ( fig. 1a) in intestine, an organ known to express CFTR and used as a positive control for expression of the CFTR (lane 1), and PASMCs (lane 3), but not in rat skeletal muscle (lane 2). Secondly, immunoprecipitation using the anti-CFTR antibody followed by in vitro PKA phosphorylation analysis demonstrated that, as in the CHO cell line used as a positive control, mature CFTR was phosphorylated in vitro by PKA in cultured PASMCs (fig. 1b). The major CFTR form (band C) was a 175-kDa protein, as determined using molecular mass standards. Controls with nonimmune mouse IgG are also shown (lanes 2 and 4). Thirdly, an immunocytochemical approach was used to determine the presence and location of the CFTR in PASMCs. Cells were stained with the anti-CFTR C-terminus monoclonal antibody and anti-smooth muscle actin antibody, whereas no staining could be detected in control experiments ( fig. 2). Taken together, these results demonstrate that the CFTR is endogenously expressed in IPA SMCs and can be detected as PKA-phosphorylated mature protein.
Analysis of chloride transport in cultured myocytes from IPA In order to verify that the iodide efflux method can be applied to IPA cultured cells, the cells were first exposed to hypoosmotic bath solution (150 mOsm?L -1 ) to stimulate the endogenous ICl,swell ( fig. 3a). The cAMP agonists vasoactive intestinal peptide (500 nM) and forskolin (LC laboratory, PKC Pharmaceuticals, Inc., Woburn, MA, USA; 10 mM) plus the isoflavone genistein (30 mM) and agents that raise [Ca 2+ ] i , such as the Ca 2+ ionophore A23187 (1 mM) and ATP (100 mM), significantly stimulated iodide efflux (p,0.001; n54), demonstrating the functional presence of multiple Cltransporters dependent on cAMP, calcium and/or cell volume in rat IPA ( fig. 3).

Inhibitors of cAMP-dependent chloride transport in cultured intrapulmonary arterial myocytes
Since the CFTR is activated by cAMP, it was hypothesised that the cAMP-dependent forskolin/genistein-activated Cltransport would be supported by the CFTR. Glibenclamide and diphenylamine-2-carboxylic acid (DPC) are two nonspecific inhibitors of Clchannels, including the CFTR [23], the stilbene derivative 4,49-diisothiocyanatostilbene-2,29-disulphonic acid (DIDS) is a nonspecific blocker of Clchannels but does not inhibit the CFTR from the extracellular side of the plasma membrane [23], and calixarene is an inhibitor of outwardly rectifying Clchannels but not of the CFTR [23,24]. Forskolin/ genistein-dependent iodide efflux was fully inhibited by 100 mM glibenclamide and 500 mM DPC, but by neither 100 nM calixarene nor 500 mM DIDS (n54 for each; fig. 4,). CFTRinh-172 (Calbiochem, VWR International, Fontenay-sous-Bois, France; 10 mM), a specific CFTR blocker [9], also strongly inhibited the iodide efflux response to forskolin/genistein ( fig. 4), suggesting that the CFTR is likely to be responsible for the forskolin/ genistein-activated Cltransport in cultured IPA cells.

Role of CFTR in intrapulmonary arterial vasorelaxation
Since cAMP-related agonists induce pulmonary vasorelaxation [25,26], the effect of CFTR activators were investigated in rat IPA rings preconstricted with 0.3 mM PHE. MPB-07 and -91 are good specific CFTR activators, whereas MPB-80 is a very poor CFTR activator but a useful control [6,7,21]. Unlike MPB-80, MPB-07 and MPB-91 induced strong vasorelaxation in IPA rings whether endothelium-denuded or not (figs 5 and 6b and figs 5 and 6a, respectively). In the presence of 500 mM DIDS, a ICl,Ca blocker [11] that had no effect on the cAMP-dependent Cltransport ( fig. 4) Taken together, the present results show that activation of the CFTR induces endothelium-independent vasorelaxation in rat IPAs, suggesting a potential and unexpected role of the CFTR in pulmonary vascular tone.

DISCUSSION
The current report presents evidence that a functional CFTR is endogenously expressed in IPA smooth muscle. First, it was shown that agonists of the cAMP pathway, such as forskolin, stimulated CFTR Clchannel activity. Secondly, the pharmacology of the CFTR is very similar to that of the epithelial CFTR, in terms of both activation (using MPB derivatives and genistein) and inhibition (using glibenclamide, DPC and CFTRinh-172). Moreover, the structural and pharmacological specificity of MPBs (i.e. the different activity of MPB-80, MPB-07 and MPB-91) are similar in IPA cells from rat and mouse aortas [6,7] and in epithelia [21]. Thirdly, activation of the  CFTR in IPA leads to endothelium-independent vasorelaxation. The present study is the first to show the presence and function of CFTR channels in primary cultured PASMCs and their implications for pulmonary vasorelaxation.
The presence of ICl,Ca and ICl,swell in rat PASMCs was also confirmed, and the use of cultured IPA SMCs to study Cltransport consequently validated.
The presence of functional cAMP-dependent Clchannels has previously been suggested in bovine pulmonary arterial smooth muscle by the use of rather nonspecific Clchannel blockers, such as phenylanthranilic acid, 9-anthracene carboxylic acid, 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB) and DIDS, on cAMP-dependent responses induced by 5-hydroxytryptamine [17,27]. The molecular origin of this channel has not been identified, although the cAMP-related 5hydroxytryptamine responses were inhibited by phenylanthranilic acid and 9-anthracene carboxylic acid but not by DIDS and NPPB. The CFTR is insensitive to DIDS [23], and this result was confirmed in the present study, suggesting that the channel observed in rat IPA may be similar to that observed in bovine pulmonary arterial smooth muscle [17]. The pharmacological profile of CFTR activation in IPA using MPB activators (MPB-91.MPB-07.MPB-80) studied with the iodide efflux and isometric tension techniques (in the present report) is similar to that previously obtained in rat or mouse aortic myocytes [6,7], epithelial cells [21] and tracheal myocytes [5]. Moreover, endothelium removal did not influence pulmonary arterial vasorelaxation in response to MPBs, a result in agreement with that previously reported for rat and mouse aorta [6,7]. Finally, the activation of cAMP-and MPBdependent iodide efflux was fully inhibited by the CFTRspecific inhibitor CFTRinh-172 in IPA, as in epithelial cells [9] and aortic SMCs [6]. All of these results contributed to the conclusion that the CFTR, as a cAMP-dependent Clchannel expressed in PASMCs, is involved in endothelium-independent pulmonary vasorelaxation.  In epithelial cells, cAMP agonists stimulate transepithelial Cltransport via phosphorylation and opening of apical CFTR channels [1,28]. In SMCs from the systemic circulation, i.e. aortas [6,7], or pulmonary circulation (present study), activation of the CFTR was evidenced after precontraction of the muscle cells. These observations are consistent with previous studies showing that cAMP is implicated in the relaxation of vascular SMCs in response to vasodilators such as b-adrenergic agonists or vasoactive intestinal peptide [25,26]. Since the current study demonstrated the presence of the CFTR in IPA, it may explain, at least in part, why an increase in cAMP concentration would induce pulmonary vasorelaxation via activation of the CFTR.
It is noteworthy that the present study was conducted in IPAs, which thus exhibit a functional CFTR. It is tempting to hypothesise that altered CFTR function in pulmonary arteries may be linked to the development of pulmonary hypertension since: 1) IPAs, compared with extrapulmonary arteries, are particularly sensitive to hypoxia and thus strongly involved in the pathogenesis of pulmonary hypertension; 2) patients affected by the genetic disease cystic fibrosis also develop some respiratory diseases, eventually leading to pulmonary hypertension; and 3) some Clchannels are involved in pulmonary hypertension (see Introduction section). Although this is beyond the scope of the present study, further investigations are required to explore the expression and function of the CFTR in pulmonary arteries from pulmonary hypertensive animals.
In summary, the current report presents the first time direct evidence of functional cystic fibrosis transmembrane conductance regulator expression in rat intrapulmonary arterial primary cultured smooth muscle cells. In a more integrated model, such as intrapulmonary arterial rings, activation of the cystic fibrosis transmembrane conductance regulator induces endothelium-independent vasorelaxation, indicating a potential role of cystic fibrosis transmembrane conductance regulator in the regulation of pulmonary vascular tone. An interesting new therapeutic development could arise from the discovery that cystic fibrosis transmembrane conductance regulator activators are able to relax pulmonary arterial smooth muscle cells, thus being potential anti-hypertensive agents.