PLCγ1-PKCε-IP₃R1 signaling plays an important role in hypoxia-induced calcium response in pulmonary artery smooth muscle cells

Abstract

Hypoxia-induced pulmonary vasoconstriction (HPV) is attributed to an increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i) in pulmonary artery smooth muscle cells (PASMCs). We have reported that phospholipase C-γ1 (PLCγ1) plays a significant role in the hypoxia-induced increase in [Ca²⁺]_i in PASMCs and attendant HPV. In this study, we intended to determine molecular mechanisms for hypoxic Ca²⁺ and contractile responses in PASMCs. Our data reveal that hypoxic vasoconstriction occurs in pulmonary arteries, but not in mesenteric arteries. Hypoxia caused a large increase in [Ca²⁺]_i in PASMCs, which is diminished by the PLC inhibitor U73122 and not by its inactive analog U73433. Hypoxia augments PLCγ1-dependent inositol 1,4,5-trisphosphate (IP₃) generation. Exogenous ROS, hydrogen peroxide (H₂O₂), increases PLCγ1 phosphorylation at tyrosine-783 and IP₃ production. IP₃ receptor-1 (IP₃R1) knock-down remarkably diminishes hypoxia- or H₂O₂-induced increase in [Ca²⁺]_i. Hypoxia or H₂O₂ increases the activity of IP₃Rs, which is significantly reduced in protein kinase C-ε (PKCε) knockout PASMCs. A higher PLCγ1 expression, activity, and basal [Ca²⁺]_i are found in PASMCs, but not in mesenteric artery smooth muscle cells from mice exposed to chronic hypoxia (CH) for 21 days. CH enhances H₂O₂- and ATP-induced increase in [Ca²⁺]_i in PASMCs and PLC-dependent, norepinephrine-evoked pulmonary vasoconstriction. In conclusion, acute hypoxia uniquely causes ROS-dependent PLCγ1 activation, IP₃ production, PKCε activation, IP₃R1 opening, Ca²⁺ release, and contraction in mouse PASMCs; CH enhances PASM PLCγ1 expression, activity, and function, playing an essential role in pulmonary hypertension in mice.

INTRODUCTION

Hypoxia-induced pulmonary vasoconstriction (HPV) increases pulmonary vascular resistance to direct blood to well-ventilated alveoli of the lungs to match ventilation perfusion, with the ultimate aim of proper oxygen delivery to the systemic circulation. Importantly, persistent HPV commonly results in fatal pulmonary hypertension and right ventricular heart failure (7, 10, 34, 38, 54).

HPV occurs mainly because of an increase in [Ca²⁺]_i in pulmonary artery smooth muscle cells (PASMCs). The hypoxia-induced initial increase in [Ca²⁺]_i is due to Ca²⁺ release from the sarcoplasmic reticulum (SR), and the subsequent sustained Ca²⁺ response results from Ca²⁺ influx from the extracellular source (8, 13, 15, 34, 58). Intracellular Ca²⁺ release channels, ryanodine receptors (RyRs), are activated by Ca²⁺, whereas inositol 1,4,5-trisphosphate receptors (IP₃Rs) are activated upon IP₃ binding, produced by the activity of phospholipase C (PLC). RyRs are involved in the hypoxia-induced increase in [Ca²⁺]_i in PASMCs (3, 13, 26, 63–65). In addition, hypoxia can increase IP₃ generation in pulmonary artery fibroblasts (60), which then release Ca²⁺ from the SR via IP₃Rs. Recently, we have found that PLCγ1 is activated, playing an important role in the hypoxia-induced increase in [Ca²⁺]_i in PASMCs and HPV (61). Importantly, Ca²⁺ released from IP₃Rs activates RyRs in PASMCs (62, 65), indicating a cross talk between these two Ca²⁺ release channels.

Reactive oxygen species (ROS) are important second messengers, which mediate hypoxia-induced increase in [Ca²⁺]_i in PASMCs (1, 2, 9, 23, 30, 36, 37, 41, 42, 43, 52, 55, 56) and are determinant of the heterogeneity of hypoxic responses in pulmonary artery (PA) and systemic artery (SA) smooth muscle cells (SMCs) (41, 52, 54). Mitochondria and NADPH oxidase (NOX) are the two major sources that generate ROS in PASMCs. Interestingly, hypoxia-induced mitochondrial ROS induce NOX-mediated ROS generation in PASMCs, ROS-induced ROS generation (RIRG) (41, 42). The hypoxia-induced mitochondrial ROS generation is primarily mediated by the mitochondrial electron train chain molecules before the complex III ubisemiquinone site, in which Rieske iron-sulfur protein (RISP) in the complex III is an indispensable element (25, 52). Hypoxia-induced ROS can activate protein kinase C-ε (PKCε) and, thus, increase [Ca²⁺]_i in PASMCs. Activation of PKCε induces NOX-mediated ROS generation in PASMCs (41, 42, 54). Importantly, hypoxia-induced activation of PLCγ1 is downstream to the complex III-mediated, RISP-dependent intracellular ROS generation in PASMCs (61). Thus, PLCγ1 and PKCε mediate ROS-dependent hypoxia-induced increase in [Ca²⁺]_i in PASMCs.

Various external stimuli cause an increase in [Ca²⁺]_i in a variety of cell types by producing IP₃ through activation of PLC. IP₃ then binds to IP₃Rs and causes the opening of the channel and release of Ca²⁺ from the SR (4, 5). Application of hypoxia-induced mitogenic factor results in a U73122-sensitive increase in [Ca²⁺]_i in cultured human PASMCs and IP₃ generation in cultured human PASMCs, as well as in pulmonary artery fibroblasts (14, 60). Importantly, hypoxia causes an increase in IP₃ generation that is fully abolished by pretreatment with the specific PLC inhibitor U73122, and IP₃R inhibition diminishes the hypoxic increase in [Ca²⁺]_i in PASMCs (61).

IP₃Rs are classified into three subtypes (IP₃R1, IP₃R2, and IP₃R3), encoded by the three separate genes. The functional unit of IP₃Rs is either present as homotetramers or heterotetramers. The three subtypes of IP₃Rs share 60–80% similarity in sequence (4, 5, 32, 39, 46, 47). Almost all tissues express IP₃R1, IP₃R2, and IP₃R3 to various extents; however, IP₃R1 predominates in adult vascular SMCs (47). In support, IP₃R1 mRNA is highly expressed in all arteries. It is also noted that IP₃R1 is expressed in vascular SMCs and endothelial cells, while IP₃R2 and IP₃R3 are present in endothelial cells; furthermore, weak expression of IP₃R2 is present in basilar artery SMCs (17). All three subtypes of IP₃Rs were expressed in primary cultured rat PASMCs, but their relative contributions to intracellular Ca²⁺ release are unknown (62). Thus, IP₃R1 may play a significant role in hypoxia-induced increase in [Ca²⁺]_i in PASMCs.

The opening of IP₃Rs requires the binding of IP₃; generally, each subunit of IP₃Rs binds to one IP₃ molecule. The binding of IP₃ to IP₃Rs is regulated by Ca²⁺, phosphorylation of IP₃Rs by various kinases, ATP, and various accessory proteins (4, 5, 32, 46, 47). Phosphorylation by protein kinase A (PKA), cyclic GMP dependent protein kinase G (PKG), PKC, and Ca²⁺ calmodulin-dependent protein kinase II (CaMKII) alters the channel functions in various tissues. Phosphorylation of IP₃Rs by PKA is found to enhance IP₃-induced Ca²⁺ release in hepatocytes, while inhibiting IP₃R functions in cerebellar cells. Phosphorylation of IP₃Rs by PKG, PKC, and CaMKII enhances the channel function by facilitating IP₃ binding to IP₃Rs (39). Importantly, PKC can phosphorylate IP₃Rs and enhance the binding of IP₃ to IP₃Rs, contributing to increased [Ca²⁺]_i. The phosphorylation of IP₃R1 is eight times more than IP₃R3 (49). Thus, hypoxia-induced activation of PKCε may regulate IP₃R function and, thereby, increase hypoxic responses in PASMCs.

Thus, in the current study, we sought to determine the molecular mechanisms for hypoxia-induced, PLCγ1-dependent increase in [Ca²⁺]_i in PASMCs. Further, we also investigated the possible involvement of PLCγ1 in chronic hypoxia (CH)-induced vascular remodeling, which plays an important role in the development of pulmonary hypertension.

MATERIALS AND METHODS

Reagents.

Fura-2 AM was purchased from Molecular Probes (Eugene, OR); phospho-PLCγ1 (Tyr-783) antibody was obtained from Cell Signaling Technologies; PLCγ1 antibody, smooth muscle actin antibody, and GAPDH were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The specificity and effectiveness of the antibodies against PLCγ1 and phospho-PLCγ1 have been validated in our previous studies (61). U73122, U73433, and H₂O₂ were obtained from Sigma Aldrich (St. Louis, MO); IP-One HTRF assay kit from Cisbio Bioassays (Bedford, MA). A set of mouse pLKO.1 lentiviral PLCγ1 shRNAs (TRC-Mm1.0; cat. no.: RMM4534), IP₃R1 shRNA (RMM4431-200348227), and lentiviral scrambled (nonsilencing, NS) shRNAs (cat. no. RHS4346) from Open Biosystems (Lafayette, CO). IP₃R1 (sc6093), IP₃R2 (sc7278), and IP₃R3 (sc7277) antibodies were obtained from Santa Cruz Biotechnology).

Animals.

C57/BALBc and PKCɛ^−/− mice were purchased from the Jackson Laboratory (Bar Harbor, ME); Swiss-Webster mice were purchased from Taconic (Germantown, NY). All animal experiments were approved by the Institutional Animal Care and Use Committee of Albany Medical College.

Preparation of isolated PAs and PASMCs.

PAs and PASMCs were prepared from C57BalbC and Swiss-Webster, as we described previously (25–27, 41, 42, 52, 61, 64, 65). Briefly, the heart and lungs were carefully removed from mice and placed in ice-cold physiological saline solution (PSS) containing (in mM): 130 NaCl, 1 MgSO₄, 1.8 CaCl₂, 5.4 KCl, 10 HEPES [4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid], and 10 glucose (pH 7.4). Resistance pulmonary and mesenteric arteries were then carefully dissected. To obtain PASMCs, PAs were enzymatically digested using the two-step digestion method: first in 2-ml PSS containing (in mg): 2.4 papain (Sigma; P4762) 0.4 dithioerythritol (Sigma; D8255) and 2 BSA (Sigma; A7888) for 20 min at 37°C (prewarmed for 20 min at 37°C), and then in 1.5 ml PSS (prewarmed at 37°C for 7–8 min) containing (in mg): 1.5 collagenase II (Sigma; C6885), 1.2 collagenase F (Sigma; 7926) 1.2 dithiothreitol (Sigma; D9779), and 1.5 BSA for 15 min at 37°C. The digested arteries were then incubated in ice-cold PSS containing 1% BSA on ice for 15 min with washing every 5 min. Single cells were harvested by gentle triturating using Pasteur pipettes. To examine the effects of pharmacological reagents, control experiments were carried out in the same mouse cells.

Hypoxia.

In in vitro experiments, acute hypoxia-induced responses were obtained by switching perfusion of PAs or PASMCs from a normoxic to hypoxia-inducing Krebs PSS (25–27, 41, 42, 52, 61, 64, 65). To test the effect of hypoxia on the activity of PLC, vascular smooth muscle tissues and cells were first incubated with a normoxic Krebs PSS (that was continuously aerated with a normoxic gas) for 10 min and then with a hypoxia-inducing Krebs PSS (continuously bubbled with a hypoxia-inducing gas) for 5 min. In control experiments, vascular smooth muscle tissues or cells were incubated with the normoxic Krebs PSS for 10 min, followed by continuous incubation with the normoxic Krebs PSS for 5 min. The composition of Krebs PSS was (in mM): 110 NaCl, 3.4 KCl, 2.4 CaCl₂, 0.8 MgSO₄, 25.8 NaHCO₃, 1.2 KH₂PO₄, and 5.6 glucose (pH 7.4). The normoxic PSS was made by bubbling 20% O₂, 5% CO₂, and 75% N₂, while hypoxia-inducing PSS was made by switching the gas to 1% O₂, 5% CO₂, and 94% N₂.

In the determination of the effect of hypoxia on [Ca²⁺]_i in PASMCs, single cells were first perfused with the normoxic PSS for 10 min and then with the hypoxia-inducing PSS for 5 min. As a control, cells were treated identically but only perfused with normoxic PSS.

To scrutinize the effect of CH in animals, mice were placed for 21 days in a BioSpherix chamber that was continuously flushed with hypoxic air (10% O₂) generated from a mixture of room air and N₂ by a BioSpherix ProOx oxygen controller (24, 66). As control mice were exposed to normoxia (21% O₂) for 21 days. PAs and PASMCs were prepared from mice exposed to hypoxia or normoxia and used for various experiments.

Total PLC and PLCγ1 activity assay.

Total PLC activity was determined using a [³H]-phosphatidylinositol bisphosphate (PIP₂) assay (51). Twenty microliters of total lysates from PAs were mixed and incubated with 20 µl of PLC activity assay buffer (containing in mM): 100 HEPES, 6 EGTA, 160 KCl, and 1 CaCl₂ (pH 7.2) with [³H]-PIP₂ (10,000 cpm counts/reaction) for 30 min at 37°C. The reaction was stopped using 200 µl 10% TCA and 100 µl of 1% BSA. The reaction content was vortexed and centrifuged at 8,600 g for 5 min. Three hundred microliters of the supernatant was mixed with 5 ml of Formula-989 liquid scintillation cocktail (PerkinElmer) and then subjected to scintillation counting to reading [³H]-IP₃ production from [³H]-PIP₂, which reflects PLC activity.

PLCγ1 activity was assessed by examining phosphorylation of PLCγ1 at tyrosine-783 using Western blot analysis, as described previously (51), as well as in our previous publications (25, 42, 64). Fifteen-microgram proteins of cell lysates were loaded onto 8% SDS-PAGE and subsequently transferred to PVDF membranes. The membranes were incubated with primary antibodies for PLCγ1 and PLCγ1 phosphorylated on tyrosine-783 at a 1:250 dilution, and then secondary horseradish peroxidase (HRP)-conjugated antibodies at a 1:1,500 dilution. Blots were developed using an enhanced chemiluminescence detection kit (Santa Cruz Biotechnology). The intensity of the bands was quantified using multigauge software (Fujifilm).

Western blot analysis.

Western blot analysis was performed as described above (25, 42, 51, 64). For PLCγ1, 15 µg of sample proteins were loaded on 8% SDS-PAGE. Primary antibodies against PLCγ1 at a 1:250 dilution and secondary HRP-conjugated antibodies at a 1:1,500 dilution were used, respectively. For IP₃R1, 200 μg of sample proteins were loaded on 8% SDS-PAGE. Primary antibodies against IP₃R1 at a 1:200 dilution and secondary HRP-conjugated antibodies at a 1:1,000 dilution were used, respectively. Blots were developed and quantified.

Measurement of [Ca²⁺]_i.

Measurement of [Ca²⁺]_i was made using a Till fluorescence imaging system (Till Photonics, Gräfelfing, Germany) (25–27, 41, 42, 52, 64, 65). PASMCs were incubated with the fluorescent dye fura-2 AM (10 µM) for 30 min. Fura-2 was excited at 340 and 380 nm, respectively. The emitted fluorescence was detected at 510 nm. [Ca²⁺]_i was calculated using the equation: [Ca²⁺]_i  = β·K_d·(R − R_min)/(R_max − R), in which R_min represents 340/380 fluorescence ratio obtained in cells exposed to 0 Ca²⁺ with ionomycin (10 µM) and EGTA (70 mM), R_max 340/380 fluorescence ratio obtained in cells exposed to high Ca²⁺ (10 mM) with ionomycin (10 µM), R experimental values of 340/380 ratio, β R380_max/R380_min value, and K_d (a dissociation constant of fura-2) 386 nM.

Muscle tension measurement.

Muscle tension measurement was performed using a PowerLab data acquisition system (AD Instruments), as described earlier (25, 26, 64, 65). Endothelium-denuded resistant PA rings were placed in organ bath chambers that contained PSS aerated with a normoxic gas mixture (20% O₂, 5% CO₂, and 75% N₂), given a preload of 800 mg, and equilibrated for 90 min. During an equilibration period, the α-adrenergic receptor agonist phenylephrine (10 µM) was applied to induce contraction for the verification of the responsiveness of PAs. Following phenylephrine-induced contraction, ACh (10 µM) was added to confirm no relaxation (i.e., the complete removal of the endothelium). A difference between the basal tension and maximal tension following norepinephrine (NE) stimulation was calculated and normalized to tissue weight (mg/mg), which was presented as NE-induced muscle tension.

Lentiviral shRNA preparation and infection.

Lentiviral particles containing mouse pLKO.1 shRNAs specific for PLCγ1 (TRCN0000024974-75), shRNAs specific for IP₃R1 (RMM4431-200348227), and NS shRNA were produced using a classic Ca²⁺ phosphate transfection protocol modified from Icardi et al. (20) and as described earlier (61). Lentiviral packaging pCMV-dR8.2 dvpr and pVSV-G constructs were kindly provided by Dr. Peter Vincent (Center for Cardiovascular Sciences, Albany Medical College, Albany, NY). In brief, 11.1 μg of the plasmid shRNAmir vectors were cotransfected with 8.33 μg of pCMV-dR8.2 dvpr, and 5.55 μg of pVSV-G in HEK-293FT cells were seeded in a 10-cm dish. The supernatant containing the viral particles was harvested 48 and 72 h after transfection and concentrated by ultracentrifugation for 2 h at 22,000 rotations per minute in an SW28 rotor at 4°C. Target cells were then transduced accordingly with the concentrated supernatant lentiviral particles. Primary cultured PASMCs in culture medium (DMEM, 1×) with 4.5 g/l glucose, l-glutamine and sodium pyruvate medium, 10% FBS, and 1% penicillin-streptomycin) were infected with lentiviral particles for 24 h, followed by incubation with the medium in the absence of lentiviral particles for 48 h. Then, cells were incubated with puromycin (4 µM) for 48 h to kill the noninfected cells and ensure near 100% infection of cells. The infected cells were incubated with 0.3% FBS containing culture medium for 48 h. The KD efficiency of each clone was determined using Western blot analysis. The clones showing maximum KD efficiency were selected for further experiments.

IP₃ production assay.

Production of IP₃ was measured by analyzing the production of inositol monophosphate (IP₁, downstream metabolite of IP₃, which is stable in the presence of LiCl) using IP-One HTRF assay (48), according to the manufacturer’s protocol. Primary cultured PASMCs were incubated in assay buffer containing LiCl to inhibit IP₁ degradation for 1.5 h at 37°C. Eu³⁺-labeled anti-IP₁ antibody (acceptor) was added, followed by IP₁-d2 conjugate (donor). After incubation for 1 h at room temperature, the reaction plate wells were excited at 343 nm on a FlexStation-III Microplate Reader (Molecular Devices), and emitted light was measured at 620 nm and 665 nm. The time-resolved fluorescence resonance energy transfer (ratio of 665 nm/620 nm), which is inversely proportional to the IP₁ amount, was used to determine IP₃ production.

[³H]-IP₃ binding assay.

The [³H]-IP₃ binding assay was performed, modified as previously described (27, 31). Briefly, cell lysate (90μg protein) was incubated with [³H]-IP₃ at 5–25nM at 37°C for 3h in a binding assay solution (300μl) containing 25mM Tris, 50mM HEPES, a protease inhibitor mix, and 100μM CaCl₂ (pH 7.4). The binding mixture was diluted with ice-cold washing buffer (3ml) containing 25mM Tris (pH 8.0) and 250mM KCl and immediately filtered through Millipore Membrane filters presoaked with washing buffer. The filters were washed three times with ice-cold washing buffer (5ml). The radioactivity associated with the filters was determined by liquid scintillation counting. Nonspecific binding was determined in the presence of 50μM unlabeled IP₃. All binding assays were executed in duplicate.

Statistical analysis.

All of the data was presented as box and whisker plots with a vertical line to mark the median vertical. The comparison between two groups was made using Student’s t-test. Experiments comprising three or more groups were compared using one-way ANOVA followed by Tukey’s multiple-comparison post hoc analysis. Values at P < 0.05 were considered statistically significant. Wherever used, numbers in parentheses indicate the number of cells isolated and tested from four or more animals.

RESULTS

PLCγ1 may play a role in the hypoxia-induced increase in [Ca²⁺]_i in PASMCs.

It is generally agreed that hypoxia causes vasoconstriction in PAs, but not SAs; thus, we first used a muscle tension measurement method to examine the effect of hypoxia on contraction in isolated PA and MA rings. As shown in Fig. 1, A and B, we observed that hypoxia produced vasoconstriction in PA rings, whereas there was no contraction in MA rings.

Fig. 1.

Phospholipase C-γ1 (PLCγ1) play an important role in hypoxia-induced pulmonary vasoconstriction (HPV). A: pulmonary artery (PA) and mesenteric artery (MA) rings were exposed to a physiological saline solution (PSS) bubbled with a normoxic gas (21% O₂, 5% CO₂, and the balance with N₂) for 10 min and then a hypoxic gas (1% O₂, 5% CO₂, and the balance with N₂) for 30 min. B: graph summarizes the tension generated in response to hypoxia in PAs and MAs (n = 4; *P < 0.05 compared with PA). C: representative Western blots of PLCγ1 expression in pulmonary artery smooth muscle cells (PASMCs) and mesenteric artery smooth muscle cells (MASMCs). D: graph summarizes the quantification of Western blots of PLCγ1 from three independent experiments. *P < 0.05 compared with MASMCs. E: representative traces show a hypoxia-induced increase in [Ca²⁺]_i in PASMCs preincubated without or with U73122 (1 µM) or U73433 (1 µM) for 30 min. F: quantification of the hypoxic increase in [Ca²⁺]_i in PASMCs untreated and treated with U73122 or U73433. Data were obtained from at least four different animals. *P < 0.05 compared with control.

The aforementioned results, together with our previous report that PLCγ1 plays an important role in HPV (61), inspired us to assess the potential role of PLCγ1 in the heterogeneity of hypoxic responses. The results indicated that expression of PLCγ1 was significantly higher in PASMCs compared with that in mesenteric artery smooth muscle cells (MASMCs) (Fig. 1, C and D). We also found a hypoxia-induced increase in [Ca²⁺]_i in PASMCs was inhibited by the PLC inhibitor U73122 (1 µM) and was unaffected by its inactive analog U73433 (1 µM, Fig. 1, E and F).

Exogenous ROS cause PLCγ1 activation and IP₃ production in PASMCs.

We have reported that hypoxia-induced increase in [Ca²⁺]_i is associated with ROS-dependent increase in the activity of PLCγ1 in PASMCs (61). In view of this previous finding, we investigated whether application of exogenous ROS, H₂O₂, may cause an increase in the activity of PLCγ1 in PASMCs, assessed by determining phosphorylation of PLCγ1 at tyrosine-783, as described previously (16, 18, 53). The results revealed that hypoxia and H₂O₂ both significantly increased the activity of PLCγ1 (Fig. 2, A and B).

Fig. 2.

Hypoxia and H₂O₂ both increase PLCγ1 activity and inositol monophosphate-1 (IP₁) production in PASMCs. A: Western blots of PLCγ1 phosphorylated at tyrosine 783 (pTyr-783) and total PLCγ1 in PASMCs exposed to normoxia, hypoxia, untreated control, and H₂O₂ (500 µM, 10 min). B: bar graph summarizes the ratio of pTyr-783 expression level relative to total PLCγ1 protein expression level from three separate experiments. *P < 0.05 compared with respective control. C: IP₁ production in cells untreated and treated with H₂O₂ in various concentrations for 10 min. Data were obtained from three separate experiments. *P < 0.05 compared with untreated group.

Furthermore, our data indicated that H₂O₂ produced a concentration-dependent increase in IP₃ production, as determined by measuring its downstream metabolite IP₁ production, in PASMCs (Fig. 2C).

PLCγ1 mediates hypoxia- and H₂O₂-induced increase in IP₃ production in PASMCs.

Our previous report has shown that hypoxia leads to PLC-dependent IP₃ production in PASMCs, and PLCγ1 regulates the hypoxia-induced increase in [Ca²⁺]_i in PASMCs (61). This suggests a possible role of PLCγ1 in hypoxia-induced IP₃ production. To address this possibility, we infected PASMCs with lentiviral particles, enclosing specific PLCγ1 or NS shRNAs, and the shRNA-infected PASMCs were used to assay the effect of PLCγ1 knockdown (KD) on hypoxic IP₃ production. Unsurprisingly, corresponding to its effect on PLCγ1 expression, PLCγ1 KD significantly attenuated the hypoxia-induced increase in IP₁ production (Fig. 3), demonstrating the inhibitory effect of PLCγ1 KD on IP₃ production in PASMCs. In contrast, NS shRNA did not affect either PLCγ1 expression or hypoxic IP₁ production. In addition, the basal IP₁ levels were also reduced by PLCγ1 KD.

Fig. 3.

PLCγ1 knock-down (KD) abolishes hypoxia- and H₂O₂-induced increase in IP₁ production in pulmonary artery smooth muscle cells PASMCs. A: summary of the hypoxia-induced IP₁ production in PASMCs. Cells were uninfected (control) or infected with lentiviral particles containing nonsilencing (NS) short hairpin (sh)RNAs or shRNAs for PLCγ1. Data were obtained from three different experiments. *P < 0.05 compared with respective normoxia group. #P < 0.05 compared with normoxia. ^ωP < 0.05 compared with normoxia. B: H₂O₂ (500 µM) induced IP₁ production in PASMCs infected with NS shRNA or PLCγ1 shRNAs. Data were obtained from three different experiments *P < 0.05 compared with NS shRNA. #P < 0.05 compared with NS shRNA.

We have also shown that H₂O₂-induced increase in IP₁ production was significantly reduced in PLCγ1 KD cells. Thus, PLCγ1 mediates the basal and hypoxia- as well as H₂O₂-induced increase in IP₃ production in PASMCs.

IP₃R1 serves as an essential downstream molecule in hypoxia-induced increase in [Ca²⁺]_i in PASMCs.

Acute hypoxia enhances IP₃ production in PASMCs; IP₃ can activate IP₃Rs and release Ca²⁺ from the SR. Among all three isoforms of IP₃Rs, IP₃R1 is predominantly present in PASMCs (45, 47). Thus, we thought that IP₃R1 might be involved in the hypoxia-induced Ca²⁺ responses as well. To investigate our hypothesis, we created lentiviral particle containing shRNAs targeting IP₃R1. These shRNAs were able to knock down IP₃R1 protein levels in PASMCs (Fig. 4, A and B). To substantiate the specificity of IP₃R1 KD, we measured the levels of IP₃R2 and IP₃R3 in IP₃R1 KD cells. Our data revealed that levels of IP₃R2 and IP₃R3 were unaffected in PASCMCs infected with shRNA targeting IP₃R1 (Fig. 4, C–E). All of these findings also validate the specificity and effectiveness of IP₃R1 antibodies.

Fig. 4.

Ca²⁺ release from IP₃ receptor 1 (R1) is involved in hypoxia- and H₂O₂-induced increase in [Ca²⁺]_i in PASMCs. A: Western blot of IP₃R1 in PASMCs infected with lentiviral particles containing shRNAs targeting IP₃R1 and NS shRNAs. B: graph summarizes the quantification of Western blots of IP₃R1 from three independent experiments. C: Western blots of the expression of IP₃R2 and IP₃R3 in PASMCs infected with NS shRNA and IP₃R1 shRNA. D: summary of the quantification of IP₃R2 protein expression. E: graph summarizes the average IP₃R3 expression. F: effect of IP₃R1 gene KD on hypoxia-induced induced increase in [Ca²⁺]_i. G: effect of IP₃R1 KD on H₂O₂ (500 µM)-induced increase in [Ca²⁺]_i. Cells were uninfected (control) or infected with lentiviral particles containing NS shRNAs and shRNAs specific for IP₃R1. *P < 0.05 compared with control.

Importantly, the hypoxia-produced increase in [Ca²⁺]_i was significantly attenuated in IP₃R1 KD PASMCs, whereas cells infected with the NS shRNAs showed a similar increase in [Ca²⁺]_i to that in control (uninfected) cells (Fig. 4F). In agreement with the hypoxia-induced response, H₂O₂ (500 µM) induced a much smaller increase in [Ca²⁺]_i in IP₃R1 KD PASMCs (Fig. 4G). These data clearly point toward the involvement of IP₃R1 in the hypoxia- and H₂O₂-induced increase in [Ca²⁺]_i in PASMCs.

PKCε regulates hypoxia- and H₂O₂-induced binding affinity or activity of IP₃Rs in PASMCs.

We have shown that mitochondrial ROS cause PKCε activation, which is involved in the hypoxia-induced increase in [Ca²⁺]_i and NOX-dependent ROS production in PASMCs (41, 42). It is also known that PKC can phosphorylate IP₃Rs to enhance their activity, and the PKC-mediated phosphorylation of IP₃R1 is eight times higher than that of IP₃R3 (49). Thus, we have speculated that hypoxia-induced activation of PKCε might phosphorylate IP₃Rs, thereby increasing the binding of IP₃ to its receptors and inducing Ca²⁺ release in PASMCs. To investigate this possibility, we compared the IP₃ binding in PASMCs from control (wild-type) and PKCε gene knockout (PKCε^−/−) mice. As expected, we have found that hypoxia caused a large increase in the binding of IP₃ to IP₃Rs in PASMCs from control mice, but a much smaller increase in cells from PKCε^−/− mice (Fig. 5A).

Fig. 5.

PKCε regulates hypoxia- and H₂O₂-induced increase in IP₃R binding affinity or activity in PASMCs. Radiolabeled [³H] IP₃ binding assay was conducted in lysates of PAMSCs to test the activity of IP₃Rs. A: effect of acute hypoxia on the binding of IP₃Rs in control (wild-type, WT) and PKCε^−/− PASMCs. B: effect of H₂O₂ (500 μM) on [³H] IP₃ binding of IP₃Rs in control and PKCε^−/− PASMCs. Data were obtained from three independent experiments. *P < 0.05 compared with control.

In addition, H₂O₂ (500 µM) increased the binding of IP₃ in control PASMCs, and the effect of H₂O₂ was significantly reduced in PKCε^−/− cells (Fig. 5B). Thus, the hypoxia-induced activation of PKCε may phosphorylate IP₃Rs, thereby increasing IP₃ binding, IP₃R activity, Ca²⁺ release, and ultimately [Ca²⁺]_i in PASMCs.

Mice exposed to CH show an increase in PLCγ1 activity in PASMCs and a decrease in MASMCs.

Acute hypoxia activates PLCγ1 through the mitochondrial ROS-dependent signaling mechanism (61) and, thus, increases [Ca²⁺]_i via the IP₃ production-mediated, IP₃R1-based Ca²⁺ release process (Figs. 3 and 4C). The increase in [Ca²⁺]_i in PASMCs may make a significant contribution to the development of HPV (Fig. 1). This unique response, if persisted, together with pulmonary arterial remodeling, ultimately leads to pulmonary hypertension (19, 50). Accordingly, we wanted to study the role of PLCγ1 in hypoxic pulmonary hypertension. In these experiments, mice were exposed to 10% or 21% O₂ for 21 days. As shown in Fig. 6A, PLC activity was significantly higher in PASM tissues from mice exposed to CH than from control (normoxic) animals; in contrast, PLC activity was lower in MASM tissues from hypoxic mice.

Fig. 6.

Mice exposed to chronic hypoxia (CH) show higher basal PLC activity and [Ca²⁺]_i in PASMCs than that in MASMCs. A: lysates from PAs and MAs from mice exposed to CH for 3 wk were subjected to PLC activity assay. Data were obtained from four independent experiments. The data were obtained by taking the average of the test values in the normoxia group, dividing the individual value from each test in the normoxia and hypoxia groups by the averaged value from the normoxia group, and finally taking the average of the calculated values in the normoxia and hypoxia groups (B) to provide basal intracellular [Ca²⁺]_i in PASMCs and MASMCs. Data were obtained from four animals. *P < 0.05 compared with normoxia.

In accordance with the altered PLC activity, the resting [Ca²⁺]_i was higher in PASMCs from hypoxic mice, while it was lower in MASMCs from hypoxic mice (Fig. 6B).

To further determine the role of PLCγ1 in hypoxic pulmonary hypertension, we isolated PASMCs from normoxic and hypoxic mice. The cells were cultured and then subjected to Western blot analysis. The results indicated that PLCγ1 expression levels were higher in PASMCs from hypoxic mice compared with normoxic mice (Fig. 7A).

Fig. 7.

Mice exposed CH showed higher expression and basal activity of PLCγ1 in PASMCs than in MASMCs. A: lysates from PASMCs of mice exposed to normoxia and CH were subjected to Western blot analysis. A, top: representative blots of PLCγ1 and GAPDH. A, bottom: quantification of PLCγ1 expression levels in PASMCs from five animals exposed to normoxia and three animals to CH. B, top: representative Western blots of protein expression of PLCγ1 phosphorylated at tyrosine 783 (pTyr-783) and total PLCγ1 in PASMCs from normoxia and CH mice. B, bottom: summary of the fold increase in the ratio of pTyr-783 expression level relative to total PLCγ1 protein expression level in PASMCs. Data were obtained from four mice exposed to normoxia and three mice exposed to CH. *P < 0.05 compared with normoxia.

Further, we examined PLCγ1 activation by assessing its Tyr-783 phosphorylation. The data revealed that the basal activity of PLCγ1 was higher in PASMCs from hypoxic mice; thus, Tyr-783 phosphorylation was significantly increased. (Fig. 7B). The higher expression and activity of PLCγ1 in PASMCs point toward its possible involvement in hypoxic pulmonary hypertension.

PLC-dependent agonist-induced pulmonary vasoconstriction is largely augmented in mice exposed to CH.

Hypoxic pulmonary hypertension is well characterized by pulmonary vasoconstriction and remodeling; these two major cellular responses are primarily mediated by the increased [Ca²⁺]_i in PASMCs (28, 44, 58, 59). Concordantly, the purinergic receptor agonist ATP (10 µM) induced a larger increase in [Ca²⁺]_i in PASMCs from mice exposed to CH relative to normoxic mice (Fig. 8A). Similarly, H₂O₂ (500 µM) produced a similar result.

Fig. 8.

ATP and H₂O₂-induced increase in [Ca²⁺]_i in PASMCs and PLC-dependent norepinephrine (NE)-induced contraction in PAs are greatly enhanced in mice exposed to CH. A: effect of ATP (10 µM) on [Ca²⁺]_i in PASMCs from normoxic and CH mice. B: effect of H₂O₂ (500 µM) on [Ca²⁺]_i in PASMCs from normoxic and CH mice. Cells were obtained from at least four different animals. *P < 0.05 compared with normoxia. C: summary of NE-induced contraction in PA rings from normoxia and CH mice treated with and without U73122 (10 µM). Data were obtained from three normoxic and four CH mice. *P < 0.05 compared with normoxia, and #P < 0.05 compared with CH.

Application of the α-adrenergic receptor agonist NE induced a contraction-dependent vasoconstriction in PAs (Fig. 8C). This agonist-induced pulmonary vasoconstriction was remarkably augmented in mice exposed to CH. More interestingly, the hypoxic augmentation of agonist-evoked pulmonary vasoconstriction was almost eliminated by pretreatment with the PLC inhibitor U73122 (10 µM) for 10 min. Notably, treatment with U73122 did not affect agonist-induced pulmonary vasoconstriction in PAs from normoxic mice. Thus, PLC, particularly PLCγ1, may mediate the enhanced pulmonary vasoconstriction in mice following CH.

DISCUSSION

HPV and associated pulmonary hypertension are mainly associated with the increased [Ca²⁺]_i in PASMCs, and the underlying signaling mechanisms are still being energetically explored. Recently, we have reported that acute hypoxia causes a significant increase in the activity of PLCγ1, which plays an important role in the hypoxia-induced increase in [Ca²⁺]_i and attendant contraction in PASMCs. The hypoxia-induced increase in the activity of PLCγ1 is secondary to the increased mitochondrial ROS production primarily through RISP-based ROS generation system in the complex III. The role of PLCγ1 in mediating hypoxia-induced Ca²⁺ and contractile responses in PASMCs are a result of the increased PLC-reliant IP₃ production, IP₃Rs opening, and SR Ca²⁺ release (61). However, the downstream molecular players that are involved in the mechanisms for PLCγ1-dependent hypoxic increase in [Ca²⁺]_i in PASMCs are undetermined. In the present study, we have confirmed that PLCγ1 mediates the hypoxia-induced increase in IP₃ production in PASMCs. Importantly, we have found that IP₃R1 is involved in the hypoxia-induced increase in [Ca²⁺]_i in PASMCs. Notably, PKCε regulates the hypoxia-induced increase in the activity of IP₃Rs in PASMCs. In addition, our data also indicate that PLCγ1 may play a vital role in the CH-induced pulmonary hypertension.

Hypoxia causes vasoconstriction in PAs and produces relaxation in SAs (1, 2, 12). We have observed HPV in isolated PAs from mice, while MAs show no contractile responses to hypoxia (Fig. 1, A and B). The heterogeneity of these hypoxic responses in PAs and SAs has been attributed primarily to a large increase in [Ca²⁺]_i in PASMCs, but not in mesenteric and other SA SMCs. Importantly, the hypoxia-induced increase in [Ca²⁺]_i in PASMCs is dependent on the production of mitochondrial superoxide (O₂^·−), H₂O₂, and other ROS molecules in PASMCs. The hypoxia-induced mitochondrial ROS-dependent activation of PKCε has been observed in PASMCs. Overall, the generation of ROS in response to hypoxia is an important factor that mediates the heterogeneity of hypoxia responses in the pulmonary and systemic vasculature (41, 52, 54). In accordance, acute hypoxia results in a significant increase in the activity of total PLC in PAs and not in MAs (61). To test the potential role of PLCγ1 in the heterogeneity of hypoxia responses in PAs and Mas, we determined the expression levels of PLCγ1 in PASMCs and MASMCs. Out data reveal that PLCγ1 expression in PASMCs is significantly higher compared with that in MASMCs (Fig. 1, C and D). Interestingly, the hypoxia-induced increase in the activity of PLCγ1 is dependent on RISP-dependent mitochondrial ROS generation in PASMCs (61). Thus, the differential ROS-dependent activation of PLCγ1 in PAs during hypoxia exposure may play an essential role in the heterogeneity of hypoxia responses in PAs and SAs. Consistent with our previous report (61), our current study has discovered that hypoxia stimulation increases [Ca²⁺]_i in PASMCs. This increase in [Ca²⁺]_i is diminished upon PLC inhibition with U3122 and is unaffected by the inactive U73122 analog U73433 (Fig. 1, E and F).

Hypoxia-induced increase in [Ca²⁺]_i in PASMCs is attributed to an increase in intracellular ROS produced by mitochondria and NOX (54). Mitochondrial O₂^·− is generated through the electron transport chain (ETC) and then rapidly converted to H₂O₂ by manganese superoxide dismutase. H₂O₂ is a more diffusible and active ROS. H₂O₂, with O₂^·− and other ROS, activates various intracellular signaling molecules and mediators. The hypoxic generation of mitochondrial ROS is primarily mediated by the ETC molecules before the complex III ubisemiquinone site, in which RISP in the complex III is an indispensable element (25, 52). Interestingly, the hypoxic increase in RISP-dependent mitochondrial ROS may cause activation of PKCε and subsequently NOX, leading to further ROS generation, termed RIRG (41, 42). In addition, RISP-dependent mitochondrial ROS activate PLCγ1 and, thereby, regulate the hypoxic Ca²⁺ response in PASMCs (61). Thus, the newly discovered RIRG and PLCγ1 activation may play a significant role in hypoxic Ca²⁺ and contractile responses in PASMCs. Furthermore, we have reported that the H₂O₂-induced increase in [Ca²⁺]_i in PASMCs is dependent on PLCγ1 (61). In agreement with this previous report, the current study has revealed that hypoxia and H₂O₂ both cause PLCγ1 activation (Fig. 2, A and B). H₂O₂ also increases IP₁ production (Fig. 2C). All of the current data support the report that the blockade of IP₃Rs inhibits the H₂O₂-induced increase in [Ca²⁺]_i in PASMCs and contraction in PAs (61).

IP₃, a product of PLC, opens IP₃Rs on the SR and subsequently induces Ca²⁺ release, which mediates various cellular responses in almost all types of cells. Hypoxia causes a large increase in the activity of PLCγ1, and PLCγ1 KD brings about a large reduction in the hypoxia-induced increase in [Ca²⁺]_i in PASMCs (61). In accord with our expectation, the data obtained in this study demonstrate that PLCγ1 mediates the production of IP₃ in response to acute hypoxia in PASMCs and also maintains the basal IP₃ production in PASMCs (Fig. 3A). In addition, H₂O₂-induced IP₁ production is also dependent on the activity of PLCγ1 in PASMCs (Fig. 3B).

We have reported that inhibition of IP₃Rs reduces the hypoxia-induced increase in [Ca²⁺]_i in PASMCs (61). Three subtypes of IP₃Rs (IP₃R1, IP₃R2, and IP₃R3) are expressed in mammalian cells and encoded by three separate genes. IP₃R1, IP₃R2, and IP₃R3 are present to various extents in every type of cells, but IP₃R1 predominates in vascular SMCs (17, 45–47). Considering this fact, we determined the role of IP₃R1 in the hypoxia-induced increase in [Ca²⁺]_i in PASMCs using the gene KD approach. Transfection of lentiviral shRNAs specific for IP₃R1, but not NS shRNAs, greatly knocked down its protein expression in PASMCs (Fig. 4, A and B). Importantly, our data reveal that IP₃R1 KD significantly diminishes the hypoxia-induced increase in [Ca²⁺]_i in PASMCs (Fig. 4F) and also H₂O₂-induced response (Fig. 4G). These findings are consistent with the general notion that IP₃R1 expression and activity predominate in vascular SMCs (17, 45–47) and also demonstrate that IP₃R1 plays an important role in the hypoxia- and H₂O₂-induced increase in [Ca²⁺]_i in PASMCs. It should also be noted that further investigations are needed to exclude the potential role of IP₃R2 and/or IP₃R3 in the hypoxia-induced [Ca²⁺]_i response in PASMCs.

IP₃Rs are opened as a result of IP₃ binding; on the other hand, phosphorylation of IP₃Rs can regulate its binding by IP₃, opening, and function (46). For instance, PKC phosphorylates IP₃Rs, enhances IP₃ binding, and increases Ca²⁺ release. Interestingly, PKC causes the phosphorylation of IP₃R1 eight times more than that of IP₃R3 (49). We and other scientists have disclosed that PKCε is involved in the hypoxia-induced increase in [Ca²⁺]_i and contraction in PASMCs (41), as well as HPV in isolated lungs (29). In the current study, we have found that IP₃R1 mediates the hypoxia-induced increase in [Ca²⁺]_i in PASMCs (Fig. 4). Considering the aforementioned facts, we speculate that PKCε may regulate IP₃Rs channel activation and, thus, contribute to the hypoxia-induced increase in [Ca²⁺]_i in PASMCs. Indeed, we have found that hypoxia enhances the binding of IP₃ to IP₃Rs (Fig. 5A). Furthermore, the hypoxia-induced increase in IP₃ binding is significantly reduced in PASMCs from PKCε^−/− mice. In addition, H₂O₂ also enhances IP₃ binding, which is also significantly inhibited in PASMCs from PKCε^−/− mice (Fig. 5B). Thus, the hypoxia-induced activation of PKCε may increase the activity of IP₃R1, increasing IP₃ binding and subsequently [Ca²⁺]_i in PASMCs.

Our previous report (61) and the current findings reveal the importance of PLCγ1 in the hypoxia-induced increase in [Ca²⁺]_i in PASMCs. The increase in [Ca²⁺]_i in PASMCs is important for HPV, which serves as a vital physiological process in the lungs (8, 11, 34, 54, 58). Instead, persistent HPV, together with pulmonary vascular remodeling, leads to the development of pulmonary hypertension and right ventricular heart failure (10, 31). We have shown that hypoxia activates PLC to increase [Ca²⁺]_i in PASMCs and to cause HPV (61). As a natural and important extension of our previous findings, the current study has found that the basal PLC activity is much higher in PAs from mice exposed to CH for 21 days (Fig. 6A). The higher PLC activity is related very well to the higher basal [Ca²⁺]_i in PASMCs (Fig. 6B) and higher vascular tone (Fig. 8C). Similarly, the correlation of the higher pulmonary artery pressure with higher vascular tone in PA has been reported (35, 57). Thus, the higher PLC activity corresponds well to the higher basal tone. In addition, we have found that the basal PLC activity and [Ca²⁺]_i are lower in MASMCs from CH mice (Fig. 6, A and B). In accordance, we have also observed that PLCγ1 expression and activity are significantly higher in PASMCs from CH mice (Fig. 7, A and B). CH is known to be associated with variable degrees of the proliferation of SMCs or smooth muscle-like cells (33), and the hypoxia-induced increase in [Ca²⁺]_i functions as an important signaling player to cause pulmonary vasoconstriction and remodeling (40). In particular, pulmonary hypertension is characterized by the increased PA tone due to the elevated [Ca²⁺]_i in PASMCs (21, 22, 28, 44, 58, 59). Thus, pulmonary vasoconstriction and remodeling may use the overlapping signaling processes in mediating their vasoconstrictive and structural responses. Consistent with the higher PLCγ1 activity, we have further observed that PAs from CH mice show the hypercontractility to the major vascular neurotransmitter NE, and this hypercontraction is blocked by the PLC inhibitor U73122 at 100 nM and 1 µM concentrations (Fig. 8C). Equally intriguingly, H₂O₂ and ATP both induce a larger increase in [Ca²⁺]_i in PASMCs from CH mice. Application of ATP stimulates purinergic P2_X and P2_Y receptors. The former receptors are the Ca²⁺-permeable ligand-gated ion channels and can allow Ca²⁺ influx upon stimulation. Activation of P2_Y receptors can activate PLC and produce IP₃ and DAG (6). Conceivably, CH may perhaps affect these two receptor-dependent signaling processes in coordination with PLCγ1 to mediate the development of pulmonary hypertension.

On the basis of the current findings with our previous report (61), we present a schematic diagram, as shown in Fig. 9, to conclude that hypoxia increases RISP-dependent mitochondrial ROS production, activates PLCγ1, causes IP₃ production, opens IP₃R, and induces Ca²⁺ release in PASMCs. On the other hand, hypoxia-induced mitochondrial ROS activate PKCε, enhance IP₃R1 activity, IP₃ binding, and induce further Ca²⁺ release. Altogether, these two signaling pathways contribute toward the hypoxia-induced increase in [Ca²⁺]_i in PASMCs, HPV, and pulmonary hypertension.

Fig. 9.

A schematic diagram of the signaling mechanisms for hypoxia-induced PLCγ1- and PKCε-dependent increase in [Ca²⁺]_i in PASMCs and HPV. Hypoxia causes an increase in RISP-dependent mitochondrial ROS production, which causes PLCγ1 activation, IP₃ production, IP₃R1 opening, and Ca²⁺ release. RISP-dependent mitochondrial ROS may also activate PKCε, phosphorylate IP₃R1, enhance IP₃ binding, induce Ca²⁺ release in PASMCs. Taken together, PLCγ1 and PKCε can synergistically contribute to the hypoxic increase in [Ca²⁺]_i in PASMCs, HPV, and pulmonary hypertension.

GRANTS

This work was supported by an American Heart Association Established Investigator Award 0340160N (to Y.-X. Wang) and Scientist Development Grant 0630236N (to Y.-M. Zheng) and National Institutes of Health Grants R01HL64043, HL064043-S1, HL075190, and HL108232 (to Y.-X. Wang).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

V.R.Y., L.M., L.J., Y.-M.Z., and Y.-X.W. conceived and designed research; V.R.Y., T.S., L.M., and L.J. performed experiments; V.R.Y., T.S., L.M., L.J., and Y.-M.Z. analyzed data; V.R.Y., T.S., L.M., L.J., Y.-M.Z., and Y.-X.W. interpreted results of experiments; V.R.Y., T.S., L.M., and L.J. prepared figures; V.R.Y. and T.S. drafted manuscript; V.R.Y., T.S., Y.-M.Z., and Y.-X.W. edited and revised manuscript; Y.-M.Z. and Y.-X.W. approved final version of manuscript.