large conductance Ca2+-activated K+ channel
Ca2+-activated Cl− channel
chronic hypoxic pulmonary hypertension
- E Cl
Cl− equilibrium potential
- E max
- I Cl.Ca
Ca2+-activated Cl− current
- I inst
- I ss
pulmonary arterial smooth muscle cell
time to peak
transmembrane protein 16A
voltage-dependent L-type Ca2+ channel
vascular smooth muscle cells
- • A prolonged reduced oxygen level in the lungs, as occurs in patients of many chronic lung diseases and in residents living at high altitude, causes pulmonary hypertension characterized by profound structural and functional changes in pulmonary vasculature.
- • Many of these changes are ascribed to alterations in Ca2+ homeostasis related to cation channels of pulmonary arterial smooth muscle cells.
- • Here we report the increase of an anion conductance called calcium-activated chloride channel and the expression of the channel gene TMEM16A in pulmonary arterial smooth muscle cells isolated from rats exposed to 10% oxygen for 3–4 weeks.
- • The upregulation of the chloride channel contributes to the hyper-responsiveness of pulmonary arteries to serotonin associated with pulmonary hypertension.
- • These results help us to appreciate the importance of anion channels in the pathophysiology of pulmonary hypertension, and may lead to alternative strategies for the treatment of the disease.
Abstract Chronic hypoxic pulmonary hypertension (CHPH) is associated with altered expression and function of cation channels in pulmonary arterial smooth muscle cells (PASMCs), but little is known for anion channels. The Ca2+-activated Cl− channel (CaCC), recently identified as TMEM16A, plays important roles in pulmonary vascular function. The present study sought to determine the effects of chronic hypoxia (CH) on the expression and function of CaCCs in PASMCs, and their contributions to the vascular hyperreactivity in CHPH. Male Wistar rats were exposed to room air or 10% O2 for 3–4 weeks to generate CHPH. CaCC current (ICl.Ca) elicited by caffeine-induced Ca2+ release or by depolarization at a constant high [Ca2+]i (500 or 750 nm) was significantly larger in PASMCs of CH rats compared to controls. The enhanced ICl.Ca density in CH PASMCs was unrelated to changes in amplitude of Ca2+ release, Ca2+-dependent activation, voltage-dependent properties or calcineurin-dependent modulation of CaCCs, but was associated with increased TMEM16A mRNA and protein expression. Maximal contraction induced by serotonin, an important mediator of CHPH, was potentiated in endothelium-denuded pulmonary arteries of CH rats. The enhanced contractile response was prevented by the CaCC blockers niflumic acid and T16Ainh-A01, or by the L-type Ca2+ channel antagonist nifedipine. The effects of niflumic acid and nifedipine were non-additive. Our results demonstrate for the first time that CH increases ICl.Ca density, which is attributable to an upregulation of TMEM16A expression in PASMCs. The augmented CaCC activity in PASMCs may potentiate membrane depolarization and L-type channel activation in response to vasoconstrictors and enhance pulmonary vasoreactivity in CHPH.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Chronic exposure to alveolar hypoxia, as occurs in patients of many chronic lung diseases and in residents living at high altitude, causes progressive and sustained increase in pulmonary arterial pressure, right heart hypertrophy and ultimately leads to a worsened prognosis of the underlying disease. Chronic hypoxic pulmonary hypertension (CHPH) is characterized by persistent constriction and profound remodelling of pulmonary arteries (PAs), and altered pulmonary vasoreactivity (Karamsetty et al. 1995; Stenmark et al. 2006). The pathogenesis of CHPH is a highly complex process, involving biochemical, functional and structural changes in and interactions between endothelial cells, smooth muscle cells, adventitial fibroblasts and the recruited circulating progenitor cells in PA tissue (Stenmark et al. 2006). A hallmark feature of pulmonary arterial smooth muscle cells (PASMCs) isolated from animals exposed to chronic hypoxia (CH) is the altered homeostasis of [Ca2+]i (Shimoda et al. 2000; Platoshyn et al. 2001; Lin et al. 2004), a key determinant of the contractile and proliferative state of PASMCs, and thereby enhanced pulmonary vascular tone and vascular remodelling. Yet the mechanisms underlying the alterations in [Ca2+]i are not fully elucidated.
Altered expression and function of ion channels contribute critically to the elevated [Ca2+]i in PASMCs associated with CHPH (Moudgil et al. 2006; Weir & Olschewski, 2006). Current evidence indicates that CH causes down-regulation of voltage-dependent K+ channels (Kv) and reduction in Kv currents (Platoshyn et al. 2001; Pozeg et al. 2003) as well as upregulation of store-operated, receptor-operated, and stretch-activated Ca2+-permeable non-selective cation channels in PASMCs (Lin et al. 2004; Wang et al. 2006; Ducret et al. 2010; Yang et al. 2012) resulting in the increase in voltage-dependent and voltage-independent Ca2+ influx and in resting [Ca2+]i. In contrast to the extensive studies on CH-induced cation channel remodelling (Moudgil et al. 2006; Weir & Olschewski, 2006), little is known about whether anion channels of PASMCs are affected by and contribute to CHPH; except for a report showing an increase in swelling-activated Ca2+-independent Cl− currents in PASMCs of rat exposed to hypoxia for 7 days (Liang et al. 2009). In monocrotaline (MCT)-induced PH model, several studies have suggested a possible role of Cl− channels in PH. For example, the volume-sensitive ClC3 Cl− channel was upregulated in PAs from MCT-treated rats (Dai et al. 2005). Application of Cl− channel blockers abolished the elevation in basal PA myogenic tone (Nakazawa et al. 2001) and suppressed the noradrenaline-induced PA constriction to a greater extent in MCT-treated rats (Oriowo, 2004).
Ca2+-activated Cl− channels (CaCCs) are widely expressed in many cell types and exert a variety of important functions, including the regulation of vascular tone under physiological conditions (Large & Wang, 1996; Hartzell et al. 2005). The recent discovery of TMEM16A and TMEM16B, two members of the mammalian TMEM16 family, as CaCC subunits (Caputo et al. 2008; Schroeder et al. 2008; Yang et al. 2008) has led to the identification of TMEM16A as the main components of CaCCs in rat PASMCs (Manoury et al. 2010). Because of active intracellular Cl− accumulation, the Cl− equilibrium potential (ECl) has a more positive value than the resting potential of vascular smooth muscle cells (VSMCs), including PASMCs (Chipperfield & Harper, 2000). Activation of CaCCs by agonist-induced or spontaneous Ca2+ release from the sarcoplasmic reticulum (SR) (Large & Wang, 1996) elicits Cl− efflux, leading to membrane depolarization (Klockner & Isenberg, 1991; Bakhramov et al. 1996; Yuan, 1997), activation of voltage-dependent Ca2+ channels (VDCCs), as well as increase in [Ca2+]i and vascular tone (Criddle et al. 1997; Guibert et al. 1997; Wang et al. 1997; Yuan, 1997; Lamb & Barna, 1998). Indeed, the involvement of CaCCs in the agonist-induced PA constriction has been well documented in normoxic animals (Guibert et al. 1997; Yuan, 1997). No information is available, however, on whether CH alters the expression and activity of CaCCs in PA smooth muscle; and whether these channels contribute to the enhanced pulmonary vasoconstriction associated with CHPH. The present study aimed to address these issues with the use of a well-characterized chronic hypoxic rat model of pulmonary hypertension.
All animal procedures conform to the Guide for the Care and Use of Laboratory Animals published by the United States National Institutes of Health and were approved by the Johns Hopkins University Animal Care and Use Committee, as well as conforming to UK regulations, as described in Drummond (2009).
The CHPH was generated by placing male Wistar rats (150–200 g) in a hypoxic chamber and exposed to normobaric hypoxia (10 ± 0.5% O2) for 3–4 weeks, as previously described and characterized in our laboratory (Shimoda et al. 1999, 2000; Lin et al. 2004). The O2 level inside the chamber was servo-controlled and continuously monitored using PRO:OX 110 gas analysers (Reming Bioinstruments Co., Redfield, NY, USA). Rats were exposed to room air for 10 min twice a week to clean the cages, and to replenish food and water supplies. Age-matched animals housed in room air were used as normoxic controls.
Rats were anaesthetized with sodium pentobarbital (130 mg kg−1, i.p.). Lungs were removed quickly and immersed in ice-cold Hepes-buffered salt solution (HBSS) containing (in mm) 130 NaCl, 5 KCl, 1.2 MgCl2, 1.5 CaCl2, 10 Hepes and 10 glucose (pH 7.2 with NaOH). The third generation of intralobar PAs (outer diameter: 300–800 μm) were dissected out and cleaned of connective tissue. After removal of endothelium by rubbing the luminal surface with a cotton swab, the arteries were digested at 37°C for 20 min in 20 μmol-Ca2+ HBSS containing type I collagenase (1750 U ml−1), papain (9.6 U ml−1), bovine serum albumin (2 mg ml−1) and DTT (1 mmol l−1). After washing with nominal Ca2+-free HBSS, the PASMCs were mechanically dispersed and kept at 4°C in the Ca2+-free HBSS for use within 6 h after isolation. For some experiments, the cells were kept in Ham's F-12 medium (with 0.5% of fetal calf serum) overnight before use under 4% O2–5% CO2 and 21% O2–5% CO2 (37°C) for cells isolated from hypoxic and normoxic rats, respectively.
Perforated and conventional whole-cell patch clamp techniques were used to record ICl.Ca evoked by caffeine-induced Ca2+ release and by pipette solution containing high Ca2+, respectively. Membrane currents were sampled at 5–10 kHz and filtered at 2–5 kHz using an Axopatch 200B amplifier driven by pCLAMP 10 software (Molecular Devices, Sunnyvale, CA, USA). Contamination by other currents was minimized by replacing K+ ions with Cs+ and by adding TEA chloride in both pipette and bath solutions. For perforated patch clamp experiments, the pipette solution contained (in mm): 125 caesium methanesulfonate, 20 CsCl, and 10 Hepes (pH 7.2 adjusted with CsOH). Freshly prepared amphotericin B was added to the pipette solution (300 μg ml−1) before experiments. For conventional whole-cell patch clamp experiments, the pipette solution contained (mm): 110 CsCl, 20 TEA-Cl, 2 MgATP, 10 EGTA, 5 Hepes and 0.16 MgCl2 (pH 7.2 adjusted with CsOH). CaCl2 at 8.095 or 7.475 mm was also included to set the free Ca2+ concentration at 750 or 500 nm, respectively, as estimated using the MaxC program (Stanford University). The current–voltage (I–V) relationships, constructed by depolarizing the cell from a holding potential (Vh) of −70 mV to voltages ranging from −60 to +120 mV in 20 mV increments, were assessed every 3–5 min after whole-cell formation. The current amplitude usually increased progressively and stabilized after 15–30 min of intracellular dialysis with the Ca2+-buffered pipette solution. The bath solution for both types of experiments contained (in mm): 135 NaCl, 5.4 CsCl, 1 MgCl2, 1 CaCl2, 0.33 NaH2PO4, 5 TEA-Cl, 10 Hepes and 10 glucose (pH 7.35 adjusted with NaOH). The calculated ECl is −50.8 mV for experiments using amphotericin B-perforated patch clamp method, and −0.5 mV for conventional whole-cell patch clamp experiments. All experiments were performed at room temperature.
Caffeine-induced Ca2+ transients and ICl.Ca were recorded simultaneously from PASMCs loaded with fluo-3 or indo 1 acetoxymethyl ester (fluo-3/AM or indo-1/AM). The extracellular dye was washed after incubation of cells with 5 μmol l−1 of either indicator for 30–45 min at room temperature. The cells were then superfused with bath solution for ∼20 min to allow for complete de-esterization of intracellular dye. The intracellular fluo-3 was excited at 480 nm and the emitted fluorescence detected at 535 nm. The amplitude of Ca2+ signals was expressed as the ratio of fluorescence change (F) over the resting fluorescence (F0). The intracellular indo-1 was excited at 365 nm and the emitted fluorescence detected at 405 and 485 nm. The ratio of fluorescence emitted at 405 nm over that at 485 nm (R405/485) was used as an index of [Ca2+]i.
RT-PCR and quantitative real-time PCR
Total RNA was extracted from endothelium-denuded PAs, and first-strand cDNA was synthesized. Gene-specific primers were used to amplify TMEM16A (5′-TGACGAGGATACCAAAATCCA-3′ and 5′-CGGGTCTCACTGATGTGGT-3′) and 18S rRNA (5′-CGGCTACCACATCCAAGGAA-3′ and 5′-AGCTGG AATTACCGCGGC-3′). Real-time PCR reactions were performed with SYBR Green PCR Master Mix on an iQ5 Real-time PCR detection System. For the internal control, 18S rRNA was used. Standard curves for each gene were generated for each real-time PCR reaction. The absolute copy number of mRNA for each gene was determined by interpolation of their respective standard curves with the threshold cycle value of each sample. The TMEM16A mRNA levels were expressed as their respective copy number relative to that of 18S rRNA for each sample obtained in the same run.
Endothelium-denuded PAs were homogenized and resuspended in ice-cold lysis buffer containing 50 mm Tris-Cl (pH 7.4), 150 mm NaCl, 1% deoxycholic acid, 0.1% SDS, 0.5% NP-40 and protease inhibitor cocktail (Roche, Mannheim, Germany). The homogenate was centrifuged at 4°C at 1000 g for 5 min, the supernatant was collected, and the total amounts of proteins were quantified with bicinchoninic acid assay. The proteins (5 μg) were separated on SDS-polyacrylamide gels (10%), transferred onto polyvinylidene difluoride membranes, immunoblotted with a goat polyclonal antibody against TMEM16A (sc-69343, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) at 1:200 dilution and visualized with Pierce ECL (Thermo Scientific). Competition studies using blocking peptide supplied with the antibody were performed according to manufacturer's instruction to verify the specificity of results. Actin, used as internal control, was immunodetected using a goat polyclonal antibody (sc-1615, Santa Cruz Biotechnology, Inc.) for each experiment. The intensity of interested bands was quantified using the Gel Logic 200 image system (Kodak, New Haven, CT, USA).
Isometric tension measurements in PAs
Intralobar PAs (o.d.: 300–800 μm) were isolated, cleaned of connective tissue and cut into ring segments 4 mm in length. Endothelium was disrupted by gently rubbing the lumen with a small wooden stick. The PA rings were suspended between two stainless steel stirrups and mounted in organ chambers filled with modified Krebs solution containing (in mm): 118.3 NaCl, 4.7 KCl, 1.2 MgSO4, 25 NaHCO3, 11.1 glucose, 1.2 KH2PO4 and 2 CaCl2. The solution was gassed with 16% O2–5% CO2 to maintain a pH of 7.4 at 37°C. The isometric tension was measured using a strain gauge (Grass Instruments, Quincy, MA, USA) connected to a PowerLab amplifier driven by the Chart software (ADInstruments Inc., Colorado Springs, CO, USA). After equilibrium under a resting tension of 0.8 g for 60 min, the arteries were exposed to 80 mm KCl 3 times with 30 min of washout to establish viability and maximum contraction. Serotonin concentration-dependent responses were assessed by adding the agonist cumulatively (0.5 log unit increments) to the bath in the presence of the vehicle solution (DMSO), or different inhibitors added 15 min before application of serotonin. The concentration–response curves obtained in the presence DMSO were used as controls. In some cases as indicated, the artery rings were first challenged with a maximum dose of serotonin (10−4 M), and the effects of different drugs were tested when the contractile response reached the plateau. At the end of experiments, the PA ring was exposed to phenylephrine (3 × 10−7 m) followed by acetylcholine (10−6 m) to verify complete disruption of endothelium. The rings that relaxed with acetylcholine or did not contract with KCl were discarded. The isometric tension response to serotonin was normalized to the maximal tension generated by 80 mM of KCl.
Chemicals and drugs
Fluo-3/AM and Indo-1/AM (Molecular Probes) stocks were prepared in 20% pluronic acid F-127 (Molecular Probes) solution prepared with DMSO. The stock solution of serotonin (Sigma) was prepared in minipore water. Niflumic acid (Sigma), nifedipine (Research Biochemicals International) and T16Ainh-A01 (Asinex) stock solutions were prepared in DMSO at concentrations at least 1000 times of that used in the experiments.
Student's t tests or analyses of variance were performed to evaluate the statistical significance of differences between two means, multiple means or two means at multiple test voltages as indicated. Tukey's test as a post hoc analysis was performed for multiple pairwise comparisons. Pooled data were expressed as means ± standard error of mean.
Caffeine-induced ICl.Ca density is enhanced in PASMCs of CH rats
To investigate whether CH alters CaCC activity in rat intralobar PASMCs, we first characterized the ICl.Ca evoked by caffeine-induced Ca2+ release from the SR in fluo-3/AM-loaded normoxic PASMCs using perforated patch clamp technique. Rapid application of 10 mm caffeine evoked a robust outward current accompanied by a large Ca2+ signal in a cell held at 0 mV (Fig. 1A). A second caffeine application immediately following the first did not induce further Ca2+ release or detectable current, suggesting that the caffeine-activated current is Ca2+-dependent. When sufficient time (5–10 min) was allowed for refilling SR with Ca2+, repetitive caffeine application (5 mm) at different holding potentials evoked Ca2+ transients of similar amplitude accompanied by a membrane current changing progressively from a large inward current at −80 mV to a large outward current at 0 mV, with no significant current being recorded at −35 mV (Fig. 1B). The current–voltage (I–V) relation exhibited linear voltage dependence (Fig. 1B). The current reversed at −49.8 ± 0.51 mV (n= 6 cells) after correction for junction potential, a value similar to the predicted ECl (−50.8 mV) set for these experiments. The Cl− selectivity of this current was further verified by changing the transmembrane Cl− gradient. When [Cl−]o was lowered from 150 to 20 mm (equal to [Cl−]i), the outward current recorded at 0 mV was abolished despite a comparable caffeine-induced Ca2+ transient (Fig. 1C). The caffeine-induced current was also inhibited by niflumic acid (100 μm, Fig. 1D). Taken together, these results are consistent with ICl.Ca evoked by caffeine-induced Ca2+ release in PASMCs.
The effects of CH on caffeine-induced ICl.Ca were examined at 0 and −80 mV. Figure 2A and B shows typical caffeine-induced ICl.Ca and Ca2+ signal obtained from single normoxic and hypoxic PASMCs. The ICl.Ca density in CH cells was significantly higher than that of normoxic PASMCs, without apparent difference in the peak amplitude of Ca2+ release between the two groups. The magnitude of increase in ICl.Ca density in CH PASMCs was similar in both inward and outward direction (Fig. 2C). To confirm that the increase in ICl.Ca density in CH PASMCs was not due to changes in the amplitude of caffeine-induced Ca2+ release, ICl.Ca and Ca2+ transient elicited by rapid caffeine application were simultaneously measured in the ratiometric Ca2+ indicator indo-1/AM-loaded PASMCs held at 0 mV. As shown in Fig. 2D, while the ICl.Ca density was significantly increased in CH cells compared to normoxic cells, the caffeine-induced changes in indo-1 ratio (ΔR405/485), reflecting the amplitude of Ca2+ transients, were not different between two groups. These results reveal for the first time that CH enhances ICl.Ca activated by Ca2+ release from the ryanodine receptor-gated Ca2+ stores in PASMCs, and this increased current density is not due to an increase in the amplitude of Ca2+ release under our experimental conditions.
Effects of CH on Ca2+-dependent properties of caffeine-induced ICl.Ca
To examine whether Ca2+-dependent properties of caffeine-induced ICl.Ca were altered by CH, the kinetics of simultaneously recorded Ca2+ transient (by fluo-3) and ICl.Ca evoked by caffeine were analysed. The activation rates of ICl.Ca and Ca2+ release, measured as time to peak (TTP) of the signals, were not different between normoxic and hypoxic groups at either 0 or −80 mV (Fig. 3A). The speed of activation of ICl.Ca correlated strongly with that of caffeine-induced Ca2+ signals in both normoxic and hypoxic cells at both 0 and −80 mV, with the slope of linear regression derived from CH cells slightly reduced at 0 mV (Fig. 3B). The positive correlation of TTP of ICl.Ca and Ca2+ transients confirms the Ca2+-dependent activation of ICl.Ca, and the absence of change in the relationship in the two groups of cells indicate that the Ca2+ dependence and time course of CaCC activation were unaltered by CH.
In normoxic PASMCs, the current exhibited faster deactivation at −80 mV than at 0 mV, while the relaxation kinetics of concomitant Ca2+ signals was not different at both voltages (Fig. 3C). No correlation was found between the ICl.Ca deactivation and Ca2+ transient relaxation kinetics at either 0 or −80 mV (Fig. 3D). Similar faster deactivation of ICl.Ca at the negative voltage was found in hypoxic cells. Compared to the normoxic control, however, the deactivation of ICl.Ca was significantly slower at −80 mV accompanied by a slower relaxation of Ca2+ signals (Fig. 3C). A similar tendency in ICl.Ca deactivation was also observed at 0 mV (P= 0.057). Plotting the 50% deactivation time of ICl.Ca against the 50% relaxation time of Ca2+ transients revealed a fair correlation of these two processes in hypoxic cells (R= 0.723, P= 0.012 at 0 mV; and R= 0.703, P= 0.023 at −80 mV) (Fig. 3D). These results suggest that the relaxation rate of Ca2+ transients is decreased in hypoxic PASMCs, which could be responsible for the slower decline of caffeine-induced ICl.Ca observed in these cells.
Effects of CH on voltage-dependent properties of ICl.Ca in PASMCs
To further investigate whether the voltage-dependent properties of CaCCs were altered by CH and contributed to enhanced ICl.Ca, we examined the current evoked by depolarization at a high [Ca2+]i (750 nm) set by a Ca2+-buffered pipette solution under conventional whole-cell configuration in freshly isolated cells. Depolarization evoked slowly activating outward currents followed by inward tail currents upon repolarization in both normoxic and hypoxic PASMCs. The current amplitude was significantly higher in CH cells (Fig. 4A). Niflumic acid (100 μm) almost completely eliminated the time-dependent outward currents and tail currents in both cell types. The I–V relationships of steady-state currents (Iss), measured at the end of 1 s pulses, exhibited a strong outward rectification (Fig. 4B, left). Despite a marked increase in the current density in CH cells, the ratio of current amplitude measured at +60 mV over that at −60 mV (i.e. rectification index) was not different between normoxic and hypoxic groups (8.7 ± 0.5, n= 10 vs. 8.2 ± 0.8, n= 10, P > 0.05 by Student's t test). The instantaneous current (Iinst) measured at the end of depolarization-induced capacitance transients, reflecting the portion of channels activated at the holding potential, was also elevated in CH cells (Fig. 4B, middle). The ratio of Iinst/ISS was not, however, altered by CH over the examined voltage range (Fig. 4B, right). The activation time constants, obtained by fitting the outward current developed during depolarization to a monoexponential function, were similar between the two groups and independent of membrane potential at voltages between +40 and +120 mV (Fig. 4C). These results suggest that the activation kinetics of ICl.Ca during sustained depolarization, which is sensitive to Ca2+ (Kuruma & Hartzell, 2000), remained unchanged in hypoxic cells. The steady-state activation curves of ICl.Ca of normoxic and hypoxic PASMCs were generated by fitting the chord conductance measured at the end of 1 s pulses and plotted against the depolarizing voltages (Fig. 4D). CH significantly augmented the maximal conductance (2.06 ± 0.32 vs. 0.87 ± 0.07 nS pF−1, P < 0.01 by Student's t test), but did not affect the V1/2 (141.2 ± 12.7 vs. 153.7 ± 12.2 mV, P > 0.05). These results indicate that CH enhanced the ICl.Ca density without altering the rectification of current, and voltage-dependent activation of the channels.
To investigate whether the slower deactivation of caffeine-induced ICl.Ca in CH cells was entirely due to the slower relaxation of Ca2+ transients, the deactivation properties of ICl.Ca evoked at constant high [Ca2+]i (750 nm) were examined. Figure 4E shows the tail currents elicited by repolarization to different voltages from a strong depolarization in a normoxic and a CH cell. The decay of tail currents accelerated as the repolarizing voltage became more negative in both cell types, a well-known property of classical ICl.Ca (Kuruma & Hartzell, 2000). The time constants of deactivation were slightly, but significantly greater in hypoxic PASMCs at all voltages examined (Fig. 4E, right). These results revealed a genuine alteration in ICl.Ca deactivation kinetics (independent of Ca2+ changes) in hypoxic PASMCs.
Calcineurin inhibitor cyclosporine A did not affect CH-induced increase in ICl.Ca
Several recent studies suggest that the CaCC activity in vascular smooth muscle cells is modulated by Ca2+–calmodulin-dependent kinase II (CaMKII) and the Ca2+-dependent phosphatase calcineurin. Inhibition of CaMKII or intracellular dialysis of constitutively active calcineurin enhanced ICl.Ca and altered the activation and deactivation kinetics (Ledoux et al. 2003; Greenwood et al. 2004). Moreover, CH-induced increased activation of nuclear factor of activated T cells, which can be prevented by calcineurin inhibitor cyclosporine A, has been reported in mouse PASMCs, suggesting increased calcineurin activity in CHPH (de Frutos et al. 2011). We therefore examined whether the enhanced ICl.Ca density and slower channel deactivation observed in CH cells are due to altered channel function by calcineurin. Application of 10 μm cyclosporine A in both bath and pipette solutions did not alter significantly the density of ICl.Ca, the I–V relation or activation kinetics, but accelerated the deactivation in both normoxic and hypoxic PASMCs dialysed with 500 nm of [Ca2+]i (Fig. 5). These results suggest that the CH-induced alterations in ICl.Ca were likely not the result of a change in channel modulation mediated by calcineurin.
Increased TMEM16A mRNA and protein expression in PAs of chronically hypoxic rats
To determine whether the enhanced ICl.Ca density observed in CH PASMCs resulted from the upregulation of CaCC channel expression, TMEM16A mRNA and protein expression in endothelium-denuded PAs were examined. Quantitative RT-PCR analysis revealed a significant increase in the TMEM16A mRNA levels in PA samples from CH rats (Fig. 6A). Western blot analysis detected a band of ∼114 kDa (the predicted size of TMEM16A) in PA protein lysates from both normoxia and CH rats. The signal was much stronger in the hypoxic PA samples (Fig. 6B). The specificity of the antibody against TMEM16A was verified with the blocking peptide, which significantly reduced the intensity of the 114 kDa band. Pooled data showed a significant increase in the TMEM16A protein level in hypoxic PAs (Fig. 6C). These results indicate that CH enhances the TMEM16A expression in PAs, and that the CH-induced increase in TMEM16A expression occurs at both mRNA and protein levels.
Contribution of upregulated CaCCs to enhanced PA contractile response to serotonin in CH rats
To explore the pathophysiological function of enhanced CaCC activity in CHPH, we examined the contribution of CaCCs to the contractile response of PAs isolated from normoxic and CH rats to serotonin, a vasoconstrictor and mitogen that plays an important role in the pathogenesis of CHPH (MacLean & Dempsie, 2009). The maximal serotonin-induced contraction (Emax, relative to KCl-induced contraction) was enhanced in CH PAs compared to normoxic controls with an increased sensitivity to the agonist (EC50= 4.68 ± 0.65 vs. 11.20 ± 0.84 μm, P < 0.05, Student's t test), consistent with previous reports (Fig. 7A) (MacLean et al. 1996). The amplitude of KCl-induced PA contraction was not different between normoxic and CH groups (0.813 ± 0.026 g, n= 66 vs. 0.808 ± 0.025 g, n= 83, P > 0.05, Student's t test). The CaCC blocker niflumic acid (30 μm, a concentration showing no effect on KCl-induced PA contraction; Yuan, 1997) or the VDCC antagonist nifedipine (1 μm) attenuated the contractile response to serotonin to a similar extent (Fig. 7B) without affecting the EC50 (Fig. 7C) in both normoxic and CH rats. Combined application of niflumic acid and nifedipine did not produce additive inhibitory effects. Furthermore, the enhanced serotonin-induced maximal contraction observed in PAs of CH rats disappeared in the presence of these agents when applied individually or simultaneously (Fig. 7D). The inhibition of serotonin response by niflumic acid was not due to the stimulation of Ca2+-activated K+ channels (BKca), since the pretreatment of PA rings with 1 mm TEA did not alter the magnitude of constriction induced by serotonin (100 μm), nor the inhibitory effects of niflumic acid in both normoxic and CH rats (Fig. 7E). Application of T16Ainh-A01, a newly identified TMEM16A channel-specific inhibitor that is structurally different from niflumic acid (Namkung et al. 2011), also eliminated the increase in serotonin-induced contraction in CH PAs (Fig. 7F). Collectively, these results suggest that the augmented CaCC activity found in CH PASMCs contributes to the enhanced pulmonary vasoconstriction in response to serotonin observed in CH rats via the activation of VDCCs.
The development of CHPH significantly worsens the prognosis of patients suffering from many chronic lung diseases. Better understanding of the underlying pathophysiological processes may lead to novel therapies for more effective management of this clinical syndrome. Membrane depolarization and increased [Ca2+]i in PASMCs of CH animals (Shimoda et al. 2000; Bonnet et al. 2001; Platoshyn et al. 2001; Lin et al. 2004; Rodat et al. 2007) are known to play a pivotal role in the induction of PASMC proliferation and the increase of PA tone associated with CHPH (Platoshyn et al. 2000; Moudgil et al. 2006; Cogolludo et al. 2007). Extensive studies on the role of cation channels in CHPH have reached the current consensus that the altered Ca2+ homeostasis in PASMCs associated with CHPH is the result of down-regulation of Kv channels and upregulation of Ca2+ permeable cation channels caused by CH (Moudgil et al. 2006; Weir & Olschewski, 2006). In the present study, we have demonstrated for the first time that the ICl.Ca activity and TMEM16A expression are enhanced in PA smooth muscle of CH rats. The increased CaCC activity, as a result of upregulation of TMEM16A expression, contributes to the enhanced pulmonary vasoreactivity to agonist associated with CHPH through a mechanism involving the activation of voltage-dependent Ca2+ influx. A possible role for CaCCs in mediating the ex vivo hypoxia-induced increase in resting [Ca2+]i in PASMCs has also been proposed by Yang et al. (2006) based on their observation that the presence of Cl− channel blockers niflumic acid or indaryloxyacetic acid in culture medium prevented the increase in [Ca2+]i and proliferation of PASMCs exposed to hypoxia for 48 h.
CH-induced increased ICl.Ca in PASMCs is attributable to an upregulation of TMEM16A expression
Our results show that a rapid rise in the [Ca2+]i caused by caffeine-induced Ca2+ release, and steady state elevation of [Ca2+]i by intracellular dialysis of buffered Ca2+ solution both activated robust ICl.Ca in rat PASMCs. This current shares similar biophysical and pharmacological properties, such as voltage and Ca2+ dependence of activation, voltage-dependent deactivation, selectivity to Cl−, and sensitivity to niflumic acid with the previously reported classical ICl.Ca (Large & Wang, 1996; Leblanc et al. 2005). This current also bears the signature feature of the classical ICl.Ca that the current–voltage relation is different when recorded at different [Ca2+]i (Kuruma & Hartzell, 2000; Xiao et al. 2011), e.g. the strong outward rectification of ICl.Ca steady-state I–V relationship at submicromolar [Ca2+]i (750 nm), and the linear relationship at high [Ca2+]i caused by caffeine-induced Ca2+ release observed in this study. The ICl.Ca evoked by caffeine or by buffered Ca2+ solution was significantly increased in PASMCs of CH rats. We did not observe, however, a significant change in time-independent baseline Cl− currents, as reported previously in PASMCs of rats exposed to 1 week of hypoxia (Liang et al. 2009), in the CH myocytes under our recording conditions.
Detailed analyses suggest that the CH-induced increase in ICl.Ca density is not due to an increase in Ca2+ sensitivity of the channel or alterations in voltage- and Ca2+-dependent activation. First, our results showed that the rectification index of ICl.Ca was not different between normoxic and hypoxic myocytes at a given [Ca2+]i, and the voltage dependence of steady-state activation was unchanged by CH despite a marked increase in the maximal conductance of the channel. This is incompatible with an increase in CaCC Ca2+ sensitivity, which is known to associate with a reduction in outward rectification of ICl.Ca and a shift of steady-state activation curve to the negative potentials (Kuruma & Hartzell, 2000; Xiao et al. 2011). Second, the activation kinetics of ICl.Ca during sustained depolarizations, which is voltage independent and Ca2+ sensitive (Kuruma & Hartzell, 2000), remained unchanged in CH cells as demonstrated by similar activation time constants and ratios of Iinst/Iss found in two groups of cells over a wide range of voltages. Third, CH did not affect the Ca2+ dependence of ICl.Ca activation, as correlations between the rate of caffeine-induced rise in [Ca2+]i and the rate of ICl.Ca activation were similar in normoxic and hypoxic cells. Fourth, inhibition of calcineurin by cyclosporine A had no effects on the ICl.Ca density despite the accelerated channel deactivation in both normoxic and CH myocytes. The lack of effects of calcineurin inhibition on ICl.Ca density in normoxic PASMCs is inconsistent with previous reports (Greenwood et al. 2004), and it is currently unclear whether this is due to a difference in the basal phosphorylation states of the channels, or in the species (rat vs. rabbit) or PA regions (intralobar vs. extralobar) used in different studies. Finally, analyses of TMEM16A expression revealed a significant increase in mRNA and protein levels in PAs from CH rats. Taken together, it seems reasonable to suggest that the CH-induced increase in ICl.Ca density in PASMCs is mainly attributable to the upregulation of CaCC protein TMEM16A expression by CH.
Another observation in this study is that CH alters the deactivation kinetics of ICl.Ca in PASMCs, and the underlying mechanism is not entirely clear. The slower decay of caffeine-induced ICl.Ca observed in hypoxic cells may be due in part to a slower relaxation of Ca2+ transients because sustained increase in [Ca2+] slows down CaCC deactivation due to a reduced rate of dissociation of Ca2+ from the channel (Kuruma & Hartzell, 2000; Xiao et al. 2011). Slower relaxation of caffeine-induced Ca2+ transients has also been reported by others in PASMCs of CH rats and attributed to impaired SR Ca2+ reuptake (Bonnet et al. 2001). However, similar observation in CH PASMCs at constant [Ca2+]i suggests other additional mechanisms might be involved. The finding that cyclosporine A treatment accelerated the deactivation of ICl.Ca to the same extent in normoxic and CH cells rules out the possibility that this alteration in kinetics was mediated by calcineurin. Regardless the underlying mechanism, the slower deactivation of CaCCs together with slower relaxation of agonist-induced Ca2+ signals would favour sustained membrane depolarization in PASMCs and may contribute to maintaining the elevation in vascular tone under the influence of vasoconstrictors in CH rats.
Enhanced CaCC activity contributes to potentiated agonist-induced pulmonary vasoconstriction in CHPH
Altered pulmonary vasoreactivity to vasoactive agents is one of the salient features of CHPH (Karamsetty et al. 1995; MacLean et al. 1996; Shimoda et al. 2000; Cogolludo et al. 2007). Enhanced PA constriction in response to vasoconstrictors such as serotonin and endothelin-1 has been well documented in CH animals (Karamsetty et al. 1995; MacLean et al. 1996; Shimoda et al. 2000). The contractile response evoked by agonist is primarily mediated by a rise in [Ca2+]i resulting from increased Ca2+ influx and SR Ca2+ release, as well as an increase in myofilament Ca2+ sensitivity. In contrast to the dominant role of SR Ca2+ release in providing Ca2+ for striated muscle contraction, Ca2+ released from SR in arterial SMCs plays an important regulatory role in controlling membrane potential by targeting Ca2+-sensitive ion channels on the plasma membrane (Thorneloe & Nelson, 2005). CaCCs and BKCa channels are two major types of Ca2+-activated channels in VSMCs including PASMCs. Activation of CaCC depolarizes the membrane potential, leading to the activation of VDCCs, increase in [Ca2+]i and contraction (positive feedback), whereas activation of BKCa channels results in opposite effects leading to cell relaxation (negative feedback) (Wellman & Nelson, 2003). It is well documented that CaCCs in PASMCs can be activated by spontaneous or agonist-evoked Ca2+ release from the SR at physiological membrane potentials, and their activation contributes to agonist-induced PA constriction by means of the positive feedback mechanism (Large & Wang, 1996; Guibert et al. 1997; Yuan, 1997). Increased TMEM16A expression and ICl.Ca density in PASMCs of CH rats found in the present study may lead to more robust membrane depolarization for Ca2+ influx via VDCCs and enhance vascular reactivity to agonists. Consistent with this hypothesis, our study demonstrated that two structurally different CaCC blockers, niflumic acid and T16Ainh-A01, or VDCC antagonist nifedipine attenuated the contractile response of PAs to serotonin in both normoxic and CH rats, and abolished the enhanced pulmonary vasoconstriction to serotonin observed in CH rats. Moreover, combined use of niflumic acid and nifedipine did not produce further inhibition.
The markedly enhanced PA vasoconstriction to serotonin with significant increase in both efficacy and sensitivity observed in CH rats in this study is consistent with previous reports (MacLean et al. 1996; Homma et al. 2007; Rodat et al. 2007). It has been proposed that the enhanced responsiveness to serotonin is due to the increase in the activity of 5-HT1B or 5-HT2B receptors (MacLean et al. 1996; MacLean & Dempsie, 2009) and reactive oxygen species production (Liu et al. 2006), as well as related to the enhanced activation of RhoA/Rho kinase (Homma et al. 2007) and voltage-independent Ca2+ influx (Rodat et al. 2007). The inhibitory effect of niflumic acid on the efficacy but not on the sensitivity of the serotonin-induced response observed in the present study suggests that CaCC is a major downstream effector in the serotonin signalling pathway for enhancing PA contraction in CH rats. Similar conclusion has been reached in pulmonary hypertensive rats associated with hyperthyroidism (Oriowo et al. 2011).
Our finding that pretreatment of PA rings with TEA did not alter the magnitude of their contractile response to serotonin, which is consistent with previous report (Yuan, 1997), suggests that BKCa channels may not play a significant role in modulating the vasoreactivity of intralobar PAs to serotonin. Furthermore, the observation that the inhibitory effect of niflumic acid on serotonin-induced contractile response was unaltered by TEA in both normoxic and CH PAs indicates that the suppression of contraction by niflumic acid resulted from CaCC inhibition rather than stimulation of the BKCa channels.
In conclusion, CaCC activity and expression are upregulated in pulmonary arterial smooth muscle associated with CHPH. Upregulation of CaCCs, in conjunction with well-documented elevations in circulating and local levels of vasoconstrictors and increased agonist receptors in PASMCs (Karamsetty et al. 1995; Stenmark et al. 2006; MacLean & Dempsie, 2009) may provide a powerful mechanism for increasing and sustaining the elevated vascular tone in CHPH. Further studies on the role of CaCCs in the development and regulation of CH-induced pulmonary vasoconstriction and remodelling should add insight into our understanding of CHPH pathophysiology with ultimate goal of developing new therapeutic strategies.
H.S. and J.S.K.S.: conception and design of the experiments; H.S., Y.X., O.P., and X.R.Y.: collection, analysis and interpretation of data; H.S. and J.S.K.S.: drafting the article and/or revising it critically for important intellectual content. The authors have no conflicts of interest to declare. All authors approved the final version for publication.
This work is supported by the National Institutes of Health (HL-071835 and HL-075134 to J.S.K.S.); and the American Heart Association (11SDG7440069 to H.S.).