Hypoxia selectively upregulates cation channels and increases cytosolic [Ca 2 (cid:2) ] in pulmonary, but not coronary, arterial smooth muscle cells

RJ, Hypoxia selectively upregulates cation channels and increases cytosolic [Ca 2 (cid:2) ] in pulmonary, but not coronary, arterial smooth muscle cells. Am Physiol Physiol 2 signaling, particularly the mechanism via store-operated Ca 2 (cid:2) entry (SOCE) and receptor-operated Ca 2 (cid:2) entry (ROCE), plays a critical role in the development of acute hypoxia- induced pulmonary vasoconstriction and chronic hypoxia-induced pulmonary hypertension. This study aimed to test the hypothesis that chronic hypoxia differentially regulates the expression of proteins that mediate SOCE and ROCE [stromal interacting molecule (STIM), Orai, and canonical transient receptor potential channel TRPC6] in pulmonary (PASMC) and coronary (CASMC) artery smooth muscle cells. The resting cytosolic [Ca 2 (cid:2) ] ([Ca 2 (cid:2) ] cyt ) and the stored [Ca 2 (cid:2) ] in the sarcoplasmic reticulum were not different in CASMC and PASMC. Seahorse measurement showed a similar level of mitochondrial bioenergetics (basal respiration and ATP production) between CASMC and PASMC. Glycolysis was signiﬁcantly higher in PASMC than in CASMC. The amplitudes of cyclopiazonic acid-induced SOCE and OAG-induced ROCE in CASMC are slightly, but signif-icantly, greater than in PASMC. The frequency and the area under the curve of Ca 2 (cid:2) oscillations induced by ATP and histamine were also larger in CASMC than in PASMC. Na (cid:2) /Ca 2 (cid:2) exchanger-mediated increases in [Ca 2 (cid:2) ] cyt did not differ signiﬁcantly between CASMC and PASMC. The basal protein expression levels of STIM1/2, Orai1/2, and TRPC6 were higher in CASMC than in PASMC, but hypoxia (3% O 2 for 72 h) signiﬁcantly upregulated protein expression levels of STIM1/STIM2, Orai1/Orai2, and TRPC6 and increased the resting [Ca 2 (cid:2) ] cyt only in PASMC, but not in CASMC. The different response of essential components of store-operated and receptor- operated Ca 2 (cid:2) channels to hypoxia is a unique intrinsic property of PASMC, which is likely one of the important explanations why hypoxia causes pulmonary vasoconstriction and induces pulmonary vascular remodeling, but causes coronary vasodilation.


INTRODUCTION
The pulmonary circulation is a high-flow, low-resistance and low-pressure circulatory system that differs functionally and structurally from the systemic circulation system, e.g., the coronary circulation system (30,71). Blood in the pulmonary artery is the deoxygenated venous blood, while the blood in the coronary artery is the oxygenated (oxygen-rich) arterial blood. A unique aspect of the pulmonary circulation is the vasoconstrictor response to hypoxia, whereas systemic arteries undergo dilation (15,56,57). Hypoxic pulmonary vasoconstriction (HPV) serves as an essential physiological process to assure efficient gas exchange by optimizing ventilation-perfusion (V/Q) matching via diverting the blood from poorly ventilated regions to well-ventilated areas of the lung (17,44,61). Persistent and widespread hypoxia throughout the lung, as occurring in patients with chronic obstructive pulmonary disease and obstructive sleep apnea or in residents living at high altitude, results in hypoxic pulmonary hypertension (HPH) by inducing sustained pulmonary vasoconstriction and pulmonary vascular medial hypertrophy. HPH-associated increase in afterload would then result in right ventricular hypertrophy and right heart failure (44,49,61). In contrast, systemic blood vessels dilate in response to tissue hypoxia, and this response appears to be critical to enhance blood flow and oxygen supply to hypoxic or hypoxemic tissues. The cellular and molecular mechanisms underlying the diverse cellular responses in pulmonary and systemic (e.g., coronary) artery smooth muscle cells, however, are still not clearly understood despite the vast efforts of investigators to elucidate the mechanism.
Vasoconstriction and vasodilatation are regulated by changes in cytosolic Ca 2ϩ concentration ([Ca 2ϩ ] cyt ) in vascular smooth muscle cells. An increase in [Ca 2ϩ ] cyt in smooth muscle cells is a key factor in the initiation and maintenance of vascular contraction, and a decrease in [Ca 2ϩ ] cyt due to Ca 2ϩ extrusion or sequestration is required for vasodilation. Accumulating evidence suggests that an increase in [Ca 2ϩ ] cyt in pulmonary artery smooth muscle cells (PASMC) is a crucial trigger for pulmonary vasoconstriction and an important stimulus for pulmonary vascular remodeling by stimulating PASMC proliferation and migration (47,59,65). In PASMC, [Ca 2ϩ ] cyt is elevated by 1) Ca 2ϩ influx through Ca 2ϩ -permeable channels in the plasma membrane, 2) Ca 2ϩ release from the intracellular stores (e.g., sarcoplasmic reticulum), and 3) inward Ca 2ϩ transportation by plasmalemmal Na ϩ /Ca 2ϩ exchangers when intracellular Na ϩ concentration is raised (28,33,65,67,68). Based on the proposed models of activation, the Ca 2ϩ -permeable channels responsible for Ca 2ϩ influx are classified into three categories: 1) voltage-dependent Ca 2ϩ channels (VDCC), which are primarily opened by membrane depolarization; 2) store-operated Ca 2ϩ channels (SOCC), which are activated by active and passive depletion of Ca 2ϩ from the intracellular stores (e.g., the sarcoplasmic or endoplasmic reticulum); and 3) receptor-operated Ca 2ϩ channels (ROCC), which are activated by ligand binding to membrane receptors (23,43,47).
The canonical transient receptor potential (TRPC) channels, such as TRPC6, are the major components of ROCC responsible for receptor-operated Ca 2ϩ entry (ROCE) (29). Recent evidence suggests that stromal interacting molecule proteins (e.g., STIM1/2) and Orai1/2 channels are the major components of SOCC leading to store-operated Ca 2ϩ entry (SOCE) (9 -11). Some reports also indicate that TRPC channels contribute to forming SOCC and regulating SOCE (5,35). That Ca 2ϩ signaling plays a critical role in a wide range of cell functions underscores the importance of understanding the mechanisms involved in regulating Ca 2ϩ homeostasis in pulmonary (PASMC) and systemic or coronary arterial smooth muscle cells (CASMC) under both physiologic and pathophysiologic conditions.
Accumulating evidence suggests that increased [Ca 2ϩ ] cyt in PASMC is the major determinant contributing to pulmonary vasoconstriction induced by acute hypoxia and pulmonary vascular wall thickening (due to PASMC proliferation) induced by chronic hypoxia (28). Studies showed that the increase of [Ca 2ϩ ] cyt in PASMC exposed to acute hypoxia was due to many mechanisms including 1) membrane depolarization-induced opening of VDCC, 2) active store depletioninduced SOCE through SOCC or STIM/Orai channels, and 3) agonist-mediated ROCE through ROCC or TRPC channels (51,54). The elevated resting [Ca 2ϩ ] cyt in PASMC and the pulmonary arterial tone of chronically hypoxic rats can be reduced to the control level by antagonists of nonselective cation channels and blockers of VDCC. Chronic hypoxia not only downregulates voltage-gated K ϩ channels, leading to membrane depolarization and activation of VDCC (40), but also upregulates Cav1.2 and Cav3.2 channels to enhance Ca 2ϩ influx through L-and T-type VDCC (53). Furthermore, Ca 2ϩ entries via store-and receptor-operated channels are also crucial for the altered Ca 2ϩ homeostasis in acute hypoxic pulmonary vasoconstriction and chronic hypoxic pulmonary hypertension (29). Enhanced SOCE and ROCE are believed to be responsible for the chronic hypoxia-mediated increase in [Ca 2ϩ ] cyt in PASMC, which subsequently results in sustained pulmonary vasoconstriction and concentric pulmonary arterial wall thickening (due to enhanced PASMC proliferation and migration).
Although it induces pulmonary vasoconstriction, acute hypoxia causes systemic (e.g., coronary, cerebral and renal) vasodilation (32). Chronic hypoxia, or prolonged exposure to hypoxia, also causes different responses between pulmonary and systemic (coronary, cerebral, and mesenteric) arterial smooth muscle cells (4,7,66 (42). Thus, hypoxia may differentially regulate selected proteins that participate in forming SOCC and ROCC, leading to diverse Ca 2ϩ and contractile responses in pulmonary and systemic artery smooth muscle cells. In this study, we investigated and compared the basic expression and function of various cation channels and transporters, or the machinery of Ca 2ϩ signaling cascade in human CASMC and PASMC, and their response to hypoxia. These data provide important information for our understanding of the molecular mechanisms responsible for the hypoxiainduced increase in [Ca 2ϩ ] cyt in PASMC and the diversity of the hypoxic response between pulmonary and coronary artery smooth muscle cells. These data may also help to provide insight into developing novel therapeutic targets for pulmonary hypertension.

METHODS AND MATERIALS
Cell culture and hypoxia treatment. Human pulmonary artery smooth muscle cells (PASMC) and coronary artery smooth muscle cells (CASMC) were purchased from Lonza (Walkersville, MD). The cells were initially isolated from normal subjects who were of the same sex and race. PASMC and CASMC were cultured at 37°C under a humidified 5% CO 2 atmosphere in the smooth muscle cell growth medium (SmGM) including the smooth muscle cell basal medium (SmBM, CC-3181, Lonza) supplemented with SmGM-2 SingleQuots (CC-4149, Lonza). Cells grown at passage 5-8 were used in the experiments. For in vitro hypoxic experiments, PASMC and CASMC were cultured in an incubator equilibrated with 3% O 2 (in N 2 ), while control cells were cultured in an incubator equilibrated with room air (21% O 2 ).
[Ca 2ϩ ] cyt measurements. [Ca 2ϩ ] cyt measurements were performed as described previously (46). Briefly, human PASMC and CASMC were grown to confluence on 25-mm round glass coverslips. The cells were loaded with 4 M fura-2 acetoxymethyl ester (fura-2/AM; Invitrogen/Molecular Probes, Eugene, OR) in the dark for 60 min at room temperature (22-24°C) in normal physiological salt solution (PSS). The PSS contained 140 mM NaCl, 4.7 mM KCl, 1.8 mM CaCl 2 , 1.2 mM MgCl 2 , 10.0 mM glucose, and 10.0 mM HEPES. A coverslip containing fura-2/AM-loaded cells was placed in a recording chamber mounted on the stage of the Nikon inverted fluorescence microscope (Eclipse Ti-E; Nikon, Tokyo, Japan). The excitation wavelengths were 340 nm and 380 nm, and the emission signal at 520 nm was detected using an EM-CCD camera (Evolve; Photometrics, Tucson, AZ), a Nikon S-Plan Fluor ELWD ϫ20/0.45 objective lens and NIS Elements 3.2 software (Nikon). [Ca 2ϩ ] cyt within the region of interest (5 ϫ 5 m), which was positioned at the peripheral region of each cell, was measured as the ratio of fluorescence intensities (F 340 /F 380 ) and recorded every 2 s. [Ca 2ϩ ] cyt was calculated by the ratiometric method using the following equation: is the dissociation constant of fura-2 for Ca 2ϩ and S f2 and S b2 are the fluorescent intensities at 380-nm excitation in Ca 2ϩ -free (with EGTA) and Ca 2ϩ -saturated (using ionomycin) solutions, respectively. R is the measured fluorescence ratio, while R min and R max are the minimal and maximal ratios that were determined in cells superfused with the Ca 2ϩ -free solution (plus 5 mM EGTA) with 2 M ionomycin and the bath solution containing 11 mM CaCl 2 , respectively. In some ex-periments, Ca 2ϩ signals (R/R 0 ) were determined as the ratio of fluorescence intensities (R) divided by the average baseline ratio (R 0 ) (10,47). In Ca 2ϩ -free solution, CaCl 2 was replaced by equimolar MgCl 2 , and 0.1 mM EGTA was added to chelate residual Ca 2ϩ . All experiments for measurement of [Ca 2ϩ ] cyt were performed at room temperature (22-24°C).
Protein preparation and Western blotting. Human PASMC and CASMC were lysed in RIPA buffer (Thermo Scientific, Rockford, IL) supplemented with a protease inhibitor cocktail (Roche, Mannheim, Germany). The cell lysate was centrifuged at 13,000 rpm for 10 min at 4°C, and the supernatants were used as sample protein. An equal amount of proteins (30 g) from each cell type was separated on SDS-polyacrylamide gels and transferred to nitrocellulose membranes (Bio-Rad, Munich, Germany). After blocking with 5% skim milk in Trisbuffered saline supplemented with 0.1% Tween 20 for 1 h at room temperature, the membranes were incubated at 4°C overnight with primary antibody against STIM1 (  (18), and TRPC6 (diluted 1:1,000; catalog no. ACC-120, Alomone) (27). The membranes were then incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies (Cell Signaling Technology, Beverly, MA) at 1:5,000 dilution for 30 min at room temperature. The protein level was normalized to ␤-actin (1:5,000; Santa Cruz Biotechnology, Santa Cruz, CA). The gel bands were visualized with Amersham ECL prime Western blotting detection reagent (GE Healthcare, Little Chalfont, UK), and then the band density was quantified with ImageJ software (National Institutes of Health, Bethesda, MD).
Immunofluorescence. Human PASMC and CAMSC were seeded onto 14-mm coverslips at a density of 2 ϫ 10 4 cells per coverslip. After the hypoxia treatment, the cells were washed twice with PBS and fixed with a 4% formaldehyde solution for 20 min at room temperature. Cells were then washed twice with PBS and permeabilized with 0.5% Triton X-100 in PBS for 10 min. Unspecific binding was reduced by blocking with 5% bovine serum albumin (BSA; Sigma) in PBS for 60 min. Cells were incubated overnight at 4°C with a mouse monoclonal anti-Orai1 antibody (1:100, catalog no. ALM-025, Alomone) (34) and rabbit polyclonal anti-STIM2 antibody (1:200; catalog no. ab59342, Abcam, Cambridge, MA) (3). After incubation, cells were extensively washed with 1% BSA in PBS for 10 min 3 times. The cells were then incubated with secondary antibodies with Hoechst (1:1,000, Thermo Scientific) in 1% BSA in PBS for 60 min at room temperature. A goat anti-mouse IgG highly cross-adsorbed secondary antibody conjugated with Alexa Fluor 594 (1:1,000, Thermo Scientific) was used to display Orai1 fluorescence image, and donkey anti-rabbit IgG highly cross-adsorbed secondary antibody conjugated with Alexa Fluor 488 (1:1,000, Thermo Scientific) was used to display STIM2 fluorescence image. Coverslips were then mounted with Prolong Diamond Antifade mountant (Molecular Probes), and images were taken on a Nikon Eclipse Ti-E inverted fluorescence microscope. All images were acquired under identical imaging conditions using identical acquisition parameters. Quantitative analysis of fluorescence was performed with Image-Pro Plus 5.0 (Media Cybernetics, Silver Spring, MD). The integrated optical density (IOD) was calculated.
Analysis of mitochondrial bioenergetics and cellular glycolysis. The XF24 Analyzer (Seahorse Biosciences, North Billerica, MA) was used to measure O 2 consumption rate (mitochondrial stress test) and extracellular acidification rate (glycolysis stress test) according to the manufacturer's instructions. All the drugs used in the study were from the XF Cell Mito Stress Test Kit and XF Cell Glycolysis Stress Test Kit (Seahorse Biosciences). CASMC and PASMC were seeded into XF24 Cell Culture Microplates (Seahorse Biosciences) at their previously optimized density of 30,000 cells per well (0.275 cm 2 ) and allowed to adhere overnight. The cells were incubated at 37°C in a CO 2 -free XF prep station 60 min before the Seahorse assay to allow the cells to equilibrate with the assay medium. For the mitochondrial stress test, drugs were injected in the following order: oligomycin (an ATP synthase inhibitor), FCCP (carbonyl cyanide-4-trifluoromethoxy phenylhydrazone, an electron transport chain accelerator), and rotenoneantimycin A combination (mitochondrial Complex 1 and III inhibitors). Normalization was performed by determining the protein concentration. ATP-linked oxygen consumption was determined as the difference between the baseline oxygen consumption rate (OCR) and the post-oligomycin minimum OCR. For the glycolysis stress test, CASMC and PASMC were glucose-starved in XF assay medium in a CO 2 -free XF prep station at 37°C for 60 min and then sequentially injected with glucose, oligomycin, and 2-deoxy-D-glucose (2-DG; glycolysis inhibitor). Glycolytic capacity was determined as the difference between the acidification rate before the addition of D-glucose and that obtained in the presence of oligomycin.
Drugs and chemicals. 1-Oleoyl-2-acetyl-sn-glycerol (OAG) was purchased from Cayman Chemical (Ann Arbor, MI) and was supplied as a solution in acetonitrile. Adenosine 5-triphosphate disodium salt hydrate (ATP) and histamine were prepared as concentrated stock solutions in distilled water. Cyclopiazonic acid (CPA) was prepared as a concentrated stock solution in dimethyl sulfoxide (DMSO). Ryanodine was purchased from Abcam and soluble in ethanol. All stocks were aliquoted and kept at Ϫ20°C and were diluted into PSS on the day of use. All reagents used were obtained from Sigma-Aldrich unless mentioned otherwise.
Statistical analysis. Data are expressed as means Ϯ SE. Statistical significance between two or among multiple groups was performed using Student's t-test or one-way analysis of variance followed by Tukey's multiple comparison post hoc test, respectively. Differences were considered to be significant when P Ͻ 0.05.

Resting [Ca 2ϩ ] cyt is comparable in CASMC and PASMC.
We first compared the resting [Ca 2ϩ ] cyt level between human CASMC and PASMC. As shown in Fig. 1, A and B, the resting [Ca 2ϩ ] cyt was not significantly different between CASMC (85.07 Ϯ 0.71 nM) and PASMC (89.80 Ϯ 0.94 nM) cultured under the same conditions. The histogram of the resting [Ca 2ϩ ] cyt in CASMC and PASMC showed a similar distribution with almost overlapped fitting curves in both cell types (Fig. 1A).
Basal metabolism between CASMC and PASMC. In CASMC and PASMC, OCR was measured and compared with estimate mitochondrial bioenergetics. Our data identified a similar level of basal respiration between CASMC and PASMC (Fig. 1C). In addition, there was no significant difference in the decrease in OCR between CASMC and PASMC exposed to oligomycin (Fig. 1C). Extracellular acidification rates (ECAR) were monitored to reflect glycolysis. There was no difference in the glycolytic rate between CASMC and PASMC in the absence of substrate (Fig. 1D). However, the addition of D-glucose caused a marked increase in ECAR in PASMC that was significantly greater than the increase in CASMC (Fig. 1, D and E). The oligomycin-mediated increase in ECAR was also significantly greater in PASMC than in CASMC (Fig. 1D). The maximal glycolytic capacity, determined as the difference in ECAR between ECAR obtained in the presence of oligomycin and that after the addition of 2-deoxy-D-glucose (2-DG), was significantly higher in PASMC than in CASMC (Fig. 1E, right). The glycolytic inhibitor, 2-deoxy-D-glucose, reduced the acidification rate in both CASMC and PASMC to the same extent (Fig. 1D).
The changes in [Ca 2ϩ ] cyt due to CPA-induced SOCE and OAG-induced ROCE in CASMC and PASMC. We first measured and compared the amplitude of increases in [Ca 2ϩ ] cyt due to SOCE in CASMC and PASMC. In the absence of extracellular Ca 2ϩ (0Ca), blockade of the sarcoplasmic reticulum (SR)/endoplasmic reticulum (ER) Ca 2ϩ -pump (SERCA) with cyclopiazonic acid (CPA) ( ] cyt due to SOCE (upon extracellular Ca 2ϩ restoration) in CASMC was slightly, but statistically significantly, greater than in PASMC (Fig. 2, A and Ba). We also determined the maximum change of [Ca 2ϩ ] cyt in R/R 0 , which displayed a similar trend (Fig. 2Bb).
Extracellular application of the membrane-permeable diacylglycerol (DAG) analogue, 1-oleoyl-2-acetyl-sn-glycerol (OAG, 100 M), can directly activate ROCC and induce ROCE. The amplitude of OAG-induced ROCE in CASMC was significantly higher compared with PASMC, with R/R 0 also exhibiting a similar trend (Fig. 2, Bb and Dd). Furthermore, the area under the curve (AUC) of the OAG-mediated increase in [Ca 2ϩ ] cyt , which represents the total amount of Ca 2ϩ that enters the cells, was significantly greater in CASMC than in PASMC (Fig. 2, C and D). The proportion of responsive cells in CASMC was also higher than in PASMC (Fig. 2D). These results indicate that the amplitudes of CPA-induced SOCE and the OAG-induced ROCE in CASMC were greater than in PASMC.

ATP-induced increase in [Ca 2ϩ
] cyt is greater in CASMC than in PASMC. ROCE was also evaluated by changes in [Ca 2ϩ ] cyt upon exposure to the receptor agonist ATP. ATP binds to purinergic receptors, stimulating phospholipase C to produce IP 3 and DAG. DAG is a known second messenger that directly opens ROCC to induce ROCE, while IP 3 activates the IP 3 receptors in the SR, thus producing Ca 2ϩ release and store depletion causing SOCE. Of the responding cells, we observed three patterns of ATP-mediated increases in [Ca 2ϩ ] cyt in CASMC and PASMC: 1) a transient increase (pattern 1), 2) a transient increase followed by a plateau or sustained increase (pattern 2), and 3) an oscillatory increase (pattern 3) (Fig. 3A). The amplitude of ATP-induced increases in [Ca 2ϩ ] cyt showed no significant difference between CASMC and PASMC, while the AUC and frequency of Ca 2ϩ oscillations induced by ATP were more significant in CASMC than in PASMC (Fig. 3B). and PASMC (67%) exhibited pattern 2 increases in [Ca 2ϩ ] cyt in response to extracellular application of ATP (Fig. 3C); the proportion of responsive cells in CASMC was also higher than in PASMC (Fig. 3D). The dose-response curves of ATPinduced increases in [Ca 2ϩ ] cyt showed significant differences between CASMC and PASMC (Fig. 3E). In CASMC, ATP increased [Ca 2ϩ ] cyt at the dose ranging from 10 Ϫ8 to 10 Ϫ7 M and the estimated EC 50 was~0.2 M (Fig. 3E). In PASMC, ATP increased [Ca 2ϩ ] cyt at the concentration ranging from 10 Ϫ6 to 10 Ϫ5 M with an estimated EC 50 of 15 M (Fig. 3E).
These data indicate that CASMC are more sensitive to ATP than PASMC.
Histamine-induced increase in [Ca 2ϩ ] cyt in CASMC and PASMC. ROCE was also evaluated by changes in [Ca 2ϩ ] cyt upon exposure to the receptor agonist histamine. Similar to the ATP-mediated changes, there are three patterns of the increases in [Ca 2ϩ ] cyt in response to histamine in CASMC and PASMC: 1) a transient increase (pattern 1), 2) a transient increase followed by a plateau or sustained increase (pattern 2), and 3) an oscillatory increase (pattern 3) (Fig. 3F). The histamine-mediated increase in [Ca 2ϩ ] cyt (in terms of both amplitude and AUC) in CASMC is greater than in PASMC (Fig. 3G). More CASMC (72%) exhibited pattern 2 increases in [Ca 2ϩ ] cyt in response to extracellular application of histamine than PASMC (48%) (Fig. 3H); the percentage of responsive cells in CASMC (97%) was also higher than in PASMC (80%) (Fig. 3I). These results imply that CASMC are more sensitive to histamine than PASMC.
The increase in [Ca 2ϩ ] cyt due to Na ϩ /Ca 2ϩ exchange in CASMC and PASMC. The activity of Na ϩ /Ca 2ϩ exchangers (NCX) in the plasma membrane depends predominantly on the transmembrane Na ϩ gradient. Removal of extracellular Na ϩ reverses the transmembrane Na ϩ gradient to favor Na ϩ extru-sion and Ca 2ϩ entry via the reverse mode of Na ϩ /Ca 2ϩ exchange. As shown in Fig. 4A, removal of extracellular Na ϩ using an equimolar concentration NMDG ϩ resulted in two patterns of increases in [Ca 2ϩ ] cyt : 1) a transient increase (pattern 1) and 2) an oscillatory increase (pattern 3). The amplitude of the transient increase in [Ca 2ϩ ] cyt (pattern 1) and the frequency of the oscillatory increase in [Ca 2ϩ ] cyt (pattern 3) due to the inward transportation of Ca 2ϩ through NCX were not different between CASMC and PASMC; however, the AUC, an indicator of the total amount of Ca 2ϩ entering the cell, of the NCX-associated inward transportation of Ca 2ϩ in CASMC was significantly larger than in PASMC (Fig. 4B). More CASMC (80%) exhibited Ca 2ϩ oscillations (pattern 3) when extracellular Na ϩ was removed than PASMC (60%) (Fig. 4C).
The local increase in [Ca 2ϩ ] cyt due to inward Ca 2ϩ transportation via the reverse mode of NCX may trigger the Ca 2ϩinduced Ca 2ϩ release (CICR) by activating ryanodine receptors (RyR) on the SR membrane. Therefore, the increase in [Ca 2ϩ ] cyt by removal of extracellular Na ϩ is composed of both inward Ca 2ϩ transportation due to NCX and CICR. To specifically determine and compare the increases in [Ca 2ϩ ] cyt via inward Ca 2ϩ transportation through NCX, we repeated the experiments described above in the presence of 50 M ryanodine (which inhibits RyR at this dose). As shown in Fig. 4, E-H, the amplitude and AUC of the NCX-associated transient increases in [Ca 2ϩ ] cyt and the frequency of the NCX-associated oscillatory increases in [Ca 2ϩ ] cyt were comparable in CASMC and PASMC (Fig. 4, E and F). The percentage of cells exhibiting pattern 1 (transient increase) and pattern 3 (oscillatory increase) changes of [Ca 2ϩ ] cyt is also similar in CASMC and PASMC (Fig. 4G); the percentage of responsive cells, however, was higher in CASMC than in PASMC (Fig. 4H).   (Fig. 2), our Western blot experiments showed that protein expression levels of STIM1/STIM2, Orai1/Orai2 and TRPC6 were significantly higher in CASMC than in PASMC (Fig. 5,  A and B). These results indicate that the basal expression of the proteins participating in the formation of SOCC (e.g., STIM1/2 and Orai1/2) and ROCC (e.g., TRPC6) is higher in CASMC than in PASMC. The Western blot experimental data are consistent with the functional data showing that the amplitudes of CPA-induced SOCE and OAG-induced ROCE are greater in CASMC than in PASMC.

C509
DIFFERENT RESPONSE OF PASMC AND CASMC TO HYPOXIA in PASMC, but not in CASMC (Fig. 5, C and D). These data show that exposure to hypoxia selectively increases expression of proteins involved in SOCE and ROCE in PASMC, but not in CASMC. The hypoxia-induced upregulation of STIM1/2, Orai1/2, and TRPC6 may be one of the unique characteristics of PASMC, in comparison to CASMC. Effect of prolonged hypoxia on STIM2 and Orai1 immunofluorescence in CASMC and PASMC. We also determined and compared the protein expression levels of STIM2 and Orai1 using immunofluorescence assay. The fluorescence intensities of STIM2 and Orai1 are shown in Fig. 6, A and B, which was consistent with the Western blot results shown in Fig. 5, C and D. Chronic or prolonged hypoxia significantly increased the protein expression level of STIM2 and Orai1 in PASMC, but there was no significant difference in the fluorescence intensity of STIM2 and Orai1 between normoxic and hypoxic CASMC. As shown in Fig. 6C, line scans also showed a marked increase in the fluorescence intensities of STIM2 and Orai1 in PASMC exposed to hypoxia (in comparison to normoxic control PASMC), but not in CASMC exposed to hypoxia (in comparison to normoxic control CASMC). The line scan region is indicated in Fig. 6A. STIM2 is a Ca 2ϩ sensor expressed in the SR/ER membrane that is mainly involved in the regulation of the resting [Ca 2ϩ ] cyt and ER Ca 2ϩ levels (8). Previous studies have suggested that compared with STIM1, STIM2 is more sensitive to small changes in the ER [Ca 2ϩ ] due to its lower affinity for Ca 2ϩ (6,38). Ong et al. (38) have demonstrated that STIM2 promotes STIM1 clustering and recruits STIM1 to ER-PM junctions at low stimulus intensities when ER Ca 2ϩ depletion is minimal, thus enhancing interaction with Orai1 and activating SOCE, suggesting that STIM2 is a critical protein that determines STIM1 clustering. Moreover, Orai1 is the main and essential pore-forming subunit of SOCC (41 (Fig. 7, A and  B). There was no shift in the histogram showing the resting [Ca 2ϩ ] cyt distribution in hypoxic CASMC compared with normoxic CASMC (Fig. 7A). In contrast, the basal [Ca 2ϩ ] cyt in PASMC was significantly increased by exposure to hypoxia (Fig. 7, C and D). The histogram showing the resting [Ca 2ϩ ] cyt distribution in hypoxic PASMC was shifted to the right in   (Fig. 7,  C and D). These results indicate that hypoxia selectively increases the resting [Ca 2ϩ ] cyt in PASMC, but not in CASMC.

DISCUSSION
The results from this study demonstrate that: 1) The basal [Ca 2ϩ ] cyt was comparable in CASMC and PASMC under resting conditions; 2) the mitochondrial bioenergetics (e.g., basal respiration and ATP production) were also similar between CASMC and PASMC; 3) glycolysis was significantly higher in PASMC than in CASMC; 4) SOCE/ROCE is slightly, but significantly, greater in CASMC than in PASMC under normoxic conditions, which was due apparently to a higher expression of STIM1/2, Orai1/2, and TRPC6 in CASMC than in PASMC; 5) the increase in [Ca 2ϩ ] cyt by inward Ca 2ϩ transportation via Na ϩ /Ca 2ϩ exchangers was comparable in CASMC and PASMC, while the rise in [Ca 2ϩ ] cyt due to Ca 2ϩ -induced Ca 2ϩ release via RyR was greater in CASMC than in PASMC; and 6) most interestingly, hypoxia had a negligible effect on protein expression of SOCC/ROCC (STIM1/STIM2, Orai1/ Orai2, and TRPC6) and the resting [Ca 2ϩ ] cyt in CASMC, but significantly upregulated SOCC/ROCC and increased the resting [Ca 2ϩ ] cyt in PASMC. Collectively, these data imply that it is a unique intrinsic characteristic or feature for PASMC to respond to hypoxia by upregulating cation channels responsible for SOCE/ROCE. These data also provide one of the important explanations why hypoxia causes pulmonary vasoconstriction and induces pulmonary vascular remodeling, but causes coronary vasodilation. The pulmonary circulation system is very different in many aspects from the systemic (e.g., coronary) circulation system.
The pulmonary circulation is a high-flow, low-resistance, and low-pressure system; the pulmonary arterial pressure is typically about one-fifth to one-sixth of the systemic arterial pressure. The primary function of the pulmonary artery is to deliver the deoxygenated venous blood to the lung for gas exchange or reoxygenation of the venous blood. The primary function of the coronary artery is to deliver the oxygenated arterial blood to cardiomyocytes (17,50,71). Thus, tissue hypoxia causes coronary vasodilation to increase blood flow to provide more oxygen and nutrients to heart muscle for maintaining heart function, whereas alveolar hypoxia causes pulmonary vasoconstriction to maintain an optimal ventilation/ perfusion match for maximal oxygenation of the venous blood (15,56,57). Prolonged alveolar hypoxia, in patients with chronic obstructive lung disease or inhabitants living at high altitude, can cause sustained pulmonary vasoconstriction and pulmonary vascular remodeling, and ultimately pulmonary hypertension. Thus, the mechanisms by which hypoxia induces pulmonary vasoconstriction and coronary vasodilation are therefore an active area of research. There are, however, still no adequate explanations for the mechanisms responsible for hypoxia-induced pulmonary vasoconstriction and coronary vasodilation.
An increase in [Ca 2ϩ ] cyt in PASMC due to Ca 2ϩ entry through Ca 2ϩ channels and/or transporters in the plasma membrane is a major trigger for pulmonary vasoconstriction and an important stimulus for PASMC migration and proliferation that contribute to pulmonary vascular remodeling. A sustained increase in the resting [Ca 2ϩ ] cyt in PASMC is one of the indispensable prerequisites for enhancing PASMC proliferation in patients with idiopathic pulmonary arterial hypertension and animals with hypoxia-induced pulmonary hypertension. Recent evidence suggests that STIM1-mediated recruitment of Orai1/2 forms SOCC in the plasma membrane and induces SOCE. Active depletion of intracellular Ca 2ϩ stores, such as the IP 3 -sensitive SR, by agonists or growth factors that stimulate IP 3 synthesis, or passive depletion of the SR by inhibiting the SR Ca 2ϩ pump with CPA, promotes STIM1 to polymerize and then translocate to the puncta area, where STIM1 interacts with Orai1 and facilitates to form tetrameric Orai1 channels to induce SOCE (9,11,14,36). STIM2 shares similar structure with STIM1, but the sensitivity of STIM2 to changes of intracellularly stored [Ca 2ϩ ] in the SR is much greater than that of STIM1 (8,70). Therefore, a very small decrease in [Ca 2ϩ ] in the SR is sufficient to activate STIM2 and ultimately stimulate Ca 2ϩ entry through Orai1/2 channels. It has therefore been proposed that STIM1 is mainly involved in the regulation of store depletion-mediated Ca 2ϩ entry, while STIM2 is predominantly involved in regulating the resting [Ca 2ϩ ] cyt (8). TRPC6 forms homotetrameric or heterotetrameric cation channels that are activated by DAG upon ligand-mediated activation of G protein-coupled receptors (GPCR) and tyrosine kinase receptors (21,24). TRPC6-formed ROCCs can also be opened by the membrane-permeable DAG analogue, OAG, or by other signaling messengers involved in the G protein signal pathway (29). Knockdown of TRPC6 with siRNA decreased DAG-induced cation influx in rat PASMC (29). ATP and histamine can bind to the related GPCR to activate phospholipase C to produce DAG and IP 3 , which act synergistically to promote Ca 2ϩ entry through ROCC and SOCC. In the present study, we determined the basic expression and function of various cation channels and transporters, or the machinery of Ca 2ϩ signaling cascade, in human CASMC and PASMC. The data showed that the amplitudes of CPA-induced SOCE and OAG-induced ROCE in CASMC are slightly, but significantly, greater than in PASMC. The frequency and the area under the curve (AUC) of Ca 2ϩ oscillations induced by ATP were larger in CASMC than in PASMC. Histamine-induced increase in [Ca 2ϩ ] cyt (both amplitude and AUC) was also greater in CASMC than in PASMC. The greater increases in [Ca 2ϩ ] cyt to Ca 2ϩ entry through ROCC and SOCC are likely due to the slightly higher expression of TRPC6 (ROCC) and STIM1/ Orai1/2 (SOCC) in CASMC than in PASMC. The results from this study also demonstrated that human CASMC and PASMC functionally express Na ϩ /Ca 2ϩ exchangers. The increase in [Ca 2ϩ ] cyt due to the inward transportation of Ca 2ϩ through Na ϩ /Ca 2ϩ exchangers is not significantly different between CASMC and PASMC under the resting conditions. Intriguingly, one of the major differences between CASMC and PASMC is their response to hypoxia: prolonged hypoxia had a negligible effect on protein expression of SOCC (STIM1/ Orai1/2) and ROCC (TRPC6) or the resting [Ca 2ϩ ] cyt in CASMC, but it significantly upregulated SOCC and ROCC and increased the resting [Ca 2ϩ ] cyt in PASMC. A wealth of data has been accumulated suggesting that STIM1 and Orai1/2 contribute to form SOCC responsible for SOCE and are upregulated in PASMC from patients with idiopathic pulmonary arterial hypertension and animals with experimental pulmonary hypertension, in comparison to controls (9,11,46). Our in vitro data showing that hypoxia increases the resting [Ca 2ϩ ] cyt in PASMC, but not in CASMC, are consistent with the previous reports demonstrating the increased basal [Ca 2ϩ ] cyt in hypoxic rat PASMC (29).
SOCE is one of the Ca 2ϩ entry pathways responsible for the elevated basal [Ca 2ϩ ] cyt in chronically hypoxic PASMC (29). There is growing evidence supporting the pivotal role of TRPC6 in acute hypoxia-induced pulmonary vasoconstriction (13,58) and chronic hypoxia-induced pulmonary hypertension (29). Upregulation of TRPC6 and enhanced Ca 2ϩ entry through ROCC have also been found in PASMC exposed to chronic hypoxia; knockdown of TRPC6 with siRNA attenuates the hypoxia-induced enhancement of ROCE (29,39). TRPC6 mediates agonist-induced ROCE in vascular smooth muscle cells (20,25). The TRPC6-formed ROCC appears to be expressed in systemic (e.g., coronary) vascular smooth muscle cells (19,56); however, the protein expression cannot be upregulated by hypoxia. This is likely one of the reasons why hypoxia fails to increase [Ca 2ϩ ] cyt and result in contractile responses in smooth muscle from the systemic circulation system (e.g., coronary arterial smooth muscle). These data also explain why hypoxia causes pulmonary vasoconstriction and coronary (or systemic) vasodilation and, furthermore, why prolonged hypoxia only causes pulmonary hypertension. Our previous data indicated that that Ca 2ϩ -sensing receptor (CaSR) might play a significant pathogenic role in the development and progression of sustained pulmonary vasoconstriction and concentric pulmonary arterial wall thickening in patients with idiopathic pulmonary arterial hypertension (IPAH) and animals with experimental pulmonary hypertension (63,64). The mRNA and protein expression level of CaSR was upregulated, while the extracellular Ca 2ϩ -induced increase in [Ca 2ϩ ] cyt was enhanced, in PASMC from patients with IPAH in comparison to normal controls (63). Importantly, CaSR functionally interacted with TRPC6 to mediate extracellular Ca 2ϩ -induced Ca 2ϩ influx and increase in [Ca 2ϩ ] cyt in IPAH-PASMC (52). Enhanced Ca 2ϩ entry due to upregulated expression and enhanced function of CaSR contributes to enhancing Ca 2ϩ signaling and activating Ca 2ϩ /CaM-sensitive transcription factors associated with PASMC proliferation and migration. It is therefore important to study the potential role of CaSR (and its functional interaction with STIM1/2, Orai1/2, and TRPC6) in the differential effects of hypoxia on Ca 2ϩ signaling in PASMC and CASMC.
The mechanisms underlying differential responses to hypoxia in CASMC and PASMC, regarding protein expression of SOCC (STIM1/Orai1/2) and ROCC (TRPC6), are unknown. Activation of Notch signaling can potentially mediate transcriptional upregulation of genes encoding SOCC (e.g., TRPC6, and Orai1/2) in human PASMC, thus enhancing SOCE (65). Our recent study showed that hypoxia activates Notch signaling and upregulates TRPC6 channels; genetic deletion of the TRPC6 gene, trpc6 (trpc6 Ϫ/Ϫ ) significantly attenuates acute hypoxiainduced pulmonary vasoconstriction and chronic hypoxia-induced pulmonary hypertension. These data imply that Notch signaling may transcriptionally regulate TRPC6 and may hence exert a critical role in pulmonary hypertension (44). It is, however, unknown whether hypoxia or prolonged hypoxia activates Notch signaling in systemic arterial smooth muscle cells like CASMC. Furthermore, hypoxia increases hypoxiainducible factors 1␣ (HIF-1␣) and 2␣ (HIF-2␣) in both systemic and pulmonary arterial smooth muscle cells (2,26). The downstream targets of HIF-1␣ and HIF-2␣ may vary among different types of cells. In PASMC, hypoxia-induced upregulation of TRPC (e.g., TRPC1 and TRPC6) channels can be mediated by HIF-1␣ and/or HIF-2␣, hypoxia-sensitive transcription factors that are known to be upregulated by increased superoxide as well (55,60). Further studies are needed to define the different mechanisms responsible for hypoxia-me-  diated transcriptional, posttranscriptional, and/or translational regulation of ROCC (e.g., STIM1 and Orai1/2) and SOCC (e.g., TRPC6/TRPV1) channel genes in CASMC and PASMC. Although effects of hypoxia on Ca 2ϩ signaling and contractile response in the two types of vascular smooth muscle cells (CASMC and PASMC) have been described, the precise mechanisms involved in hypoxia (or oxygen) sensing and its downstream effectors (dependent or independent of HIF-1␣/HIF-2␣) are not well known. Substantial evidence has accumulated demonstrating that reactive oxygen species (ROS) can influence vascular smooth muscle [Ca 2ϩ ] cyt (43,56) and hypoxia can differentially regulate the production of cytoplasmic and mitochondrial ROS including superoxide in systemic and pulmonary vascular cells (1,57). A recent report indicates that chronic hypoxia increases H 2 O 2 generation in PASMC, and H 2 O 2 is the pivotal regulator that ultimately acts to promote the expression and interaction of SOCC and function as the functional regulator in response to hypoxia (10). In contrast, hypoxia inhibits the effect of oxidants (including H 2 O 2 ) and increases cellular reducing capacity, thereby reducing Ca 2ϩ influx in systemic arteries (15). Wu et al. (60) have shown that chronic hypoxia differently regulates the generation of ROS in human pulmonary and coronary artery smooth muscle cells, with PASMC exhibiting an increase in ROS, whereas CASMC is showing a decrease. As one of the downstream effectors of ROS, ion channels including ROCC and SOCC are highly sensitive to the cellular redox status (16,37); hence, chronic hypoxia may divergently regulate SOCC/ROCC by affecting intracellular ROS generation, which leads to diverse Ca 2ϩ and contractile responses in pulmonary and coronary artery smooth muscles. Mitochondrial-derived ROS may also contribute to the different responses in these two types of vascular smooth muscle cells (CASMC vs. PASMC). Mitochondria in PASMC exhibit lower respiratory rates and higher rates of ROS and H 2 O 2 production compared with the mitochondria in smooth muscle cells of systemic arteries (31). Accordingly, a recent study shows that a specific subunit 4 isoform of the mitochondrial complex IV, the cytochrome-c oxidase subunit 4 isoform 2 (Cox4i2), which is preferentially expressed in lungs, plays an important role in hypoxia-induced superoxide release and acute hypoxic pulmonary vasoconstriction (45). Mitochondria in PASMC also possess unique functional and metabolic properties that are different from systemic vascular smooth muscle cells (72).
In this study, we also determined and compared oxygen consumption profile for oxidative phosphorylation in CASMC and PASMC. No significant difference between CASMC and PASMC in terms of the mitochondrial respiration was observed; both basal respiration and the oxygen consumed for ATP production were comparable between CASMC and PASMC under resting conditions. The glycolysis, however, was higher in PASMC than in CASMC, suggesting that differences in metabolic profiles exist between PASMC and CASMC. Accumulating evidence suggests that mitochondria in PASMC appear structurally and functionally distinct from those in systemic arteries (12,31). Michelakis et al. (31) showed that, at baseline, lung mitochondria are less active and have lower respiratory rates and higher rates of ROS and H 2 O 2 production than kidney mitochondria. Recently, Zhu et al. (72) provided compelling evidence that transplantation of mitochondria derived from femoral artery smooth muscle cells inhibits acute hypoxia-induced pulmonary vasoconstriction, attenuates chronic hypoxia-mediated pulmonary vascular remodeling, and thus prevents the development of pulmonary hypertension and reverses the established pulmonary hypertension in rats exposed to chronic hypoxia. In the present study, however, we examined CASMC and PASMC under resting conditions. It is possible that, when the cells are exposed to hypoxia or stimulated by other cellular stressors, the mitochondrial function and/or oxygen consumption profile maybe different.
On the basis of data from our current study and others, it is plausible to speculate that glycolytic capacity might at least in part explain the distinct responses of PASMC and CASMC to hypoxia. It should be noted that our data do not prove a direct causal relationship between the mitochondrial diversity and differentially regulating protein expression (and function) of SOCC and ROCC. However, the results from this study are promising and provide clear clues for future in-depth study. Previous studies also implicate that hypoxia markedly increases mitochondrial ROS production in pulmonary, but not in systemic, vascular system. The increased mitochondrial ROS may result in Ca 2ϩ entry through TRPC channels and increase [Ca 2ϩ ] cyt in PASMC. In contrast, hypoxia has a negligible effect on mitochondrial ROS production to affect the expression and/or function of SOCC/ROCC in systemic vascular smooth muscle cells (56). These data direct us to hypothesize that mitochondrial ROS and glycolytic capacity may be responsible for differentially regulating protein expression (and function) of SOCC and ROCC in CASMC and PASMC, which contribute, at least in part, to the distinction of coronary and pulmonary arterial smooth muscle cells regarding their response to hypoxia.
One potential limitation of the present study is the use of cultured PASMC and CASMC, which may exhibit alterations compared with freshly isolated cells. Thus, caution should be used in drawing conclusions based on experiments using cultured cells as the study subject, as they may have different molecular features compared with freshly isolated cells. More studies using freshly isolated human tissue (when possible) are required to further prove and clarify the conclusions from the present study. Additionally, the previous study showed that modifying culture conditions could preserve the differentiated state of smooth muscle cells (22), which would be a better way to investigate cultured smooth muscle cells when freshly isolated smooth muscle cells are not suitable.
Furthermore, differential regulation of the IP 3 and IP 3 receptors (IP 3 R) through allosteric modifiers, for example, or Ca 2ϩ feedback mechanisms, may cause the cells (e.g., CASMC and PASMC) to respond differently to the same agonist at the same concentration and duration. In CASMC and PASMC, we categorized the agonist-induced changes of [Ca 2ϩ ] cyt into three patterns: pattern 1, a transient increase in [Ca 2ϩ ] cyt ; pattern 2, a transient increase in [Ca 2ϩ ] cyt followed by a plateau or sustained increase in [Ca 2ϩ ] cyt ; and pattern 3, an oscillatory increase in [Ca 2ϩ ] cyt . The kinetics and/or pattern of agonistinduced changes of [Ca 2ϩ ] cyt depend largely on IP 3 synthesis and IP 3 R activity, and IP 3 interaction with IP 3 R. Therefore, differential regulation of IP 3 and IP 3 R through endogenous intracellular regulators would be a central contributor to the three patterns of [Ca 2ϩ ] cyt changes identified in CASMC and PASMC. More experiments on IP 3 synthesis, IP 3 -IP 3 R inter-action, and IP 3 R regulation in CASMC and PASMC are needed to define whether IP 3 and its downstream signaling are the central arbiter of signaling properties involving the three patterns and of the differences between CASMC and PASMC.
An increase in [Ca 2ϩ ] cyt due to Ca 2ϩ entry through upregulated SOCC and ROCC may play an important pathogenic role in the development and progression of pulmonary vascular remodeling and vasoconstriction in pulmonary hypertension. The hypoxia-mediated upregulation of STIM1/2, Orai1/2, and TRPC6 in PASMC may reflect an intrinsic pathophysiological response unique to the pulmonary vasculature, which contributes to sustained pulmonary vasoconstriction and concentric pulmonary arterial wall thickening in patients and animals with hypoxia-induced pulmonary hypertension. Hypoxia-mediated upregulation of STIM1, Orai1/2, and TRPC6 is, however, not observed in CASMC so that hypoxia may result in coronary vasodilation via different mechanisms. Of note, a majority of previously reported studies are performed in animals or using animal cells; this study provides compelling evidence that hypoxia differentially regulates protein expression of ROCC and SOCC in coronary and pulmonary arterial smooth muscle cells, which may help improve our understanding of the molecular mechanisms involved in the functional distinction of the pulmonary and coronary vasculature.