We previously reported that angiotensin II stimulates an increase in nitric oxide production in pulmonary artery endothelial cells. The aims of this study were to determine which receptor subtype mediates the angiotensin II-dependent increase in nitric oxide production and to investigate the roles of the angiotensin type 1 and type 2 receptors in modulating angiotensin II-dependent vasoconstriction in pulmonary arteries. Pulmonary artery endothelial cells express both angiotensin II type 1 and type 2 receptors as assessed by RT-PCR, Western blot analysis, and flow cytometry. Treatment of the endothelial cells with PD-123319, a type 2 receptor antagonist, prevented the angiotensin II-dependent increase in nitric oxide synthase mRNA, protein levels, and nitric oxide production. In contrast, the type 1 receptor antagonist losartan enhanced nitric oxide synthase mRNA levels, protein expression, and nitric oxide production. Pretreatment of the endothelial cells with either PD-123319 or an anti-angiotensin II antibody prevented this losartan enhancement of nitric oxide production. Angiotensin II-dependent enhanced hypoxic contractions in pulmonary arteries were blocked by the type 1 receptor antagonist candesartan; however, PD-123319 enhanced hypoxic contractions in angiotensin II-treated endothelium-intact vessels. These data demonstrate that angiotensin II stimulates an increase in nitric oxide synthase mRNA, protein expression, and nitric oxide production via the type 2 receptor, whereas signaling via the type 1 receptor negatively regulates nitric oxide production in the pulmonary endothelium. This endothelial, type 2 receptor-dependent increase in nitric oxide may serve to counterbalance the angiotensin II-dependent vasoconstriction in smooth muscle cells, ultimately regulating pulmonary vascular tone.
- nitric oxide synthase
- angiotensin type 2 receptor
- pulmonary endothelium
the renin-angiotensin system (RAS) plays a major role in the control of cardiovascular, renal, and adrenal functions (15). The main effector peptide molecule of the RAS, angiotensin II (ANG II), and its metabolites elicit cellular responses through at least three receptor subtypes (15, 33, 45): type 1 (AT1), type 2 (AT2), and type 4 (AT4). Signaling pathways mediated via the AT1 receptor include stimulation of phospholipases, protein kinases, and gene transcription; calcium mobilization; and inhibition of adenylate cyclase (15). Although the AT2 receptor has been linked to inhibition of cell growth, neuronal differentiation, apoptosis, and regulation of blood pressure, there are conflicting data regarding the specific signaling pathways linked to this receptor (15, 45). For example, it has been reported that activation of the AT2 receptor can lead to both an increase (5) and decrease (27) in protein phosphatase activity. Moreover, depending on experimental conditions, the p42/p44 MAPK pathway can be either activated (19) or inhibited (24) via the AT2 receptor. Whereas cloning of the AT1 and AT2 receptors reveals that they are members of the G protein-dependent, seven transmembrane-spanning receptor family, the AT4 receptor is an insulin-regulated aminopeptidase (2). The AT4 receptor has been proposed to play a role in vasodilation (33) and in mediating natriuresis (20). Less is known about the signaling pathways linked to the AT4 receptor, although, in porcine artery endothelial cells, Chen and colleagues (11) demonstrate that ANG3–8 stimulated a phospholipase C (PLC)-, phosphatidylinositol-3 kinase-dependent increase in intracellular calcium.
We (32) and others (36, 38, 40, 43) have reported that ANG II stimulates the production of nitric oxide (NO), a potent vasodilator. NO is synthesized from l-arginine by a family of NADPH-dependent nitric oxide synthases (NOS) (3): two constitutively expressed enzymes, neuronal and endothelial (eNOS), and a cytokine-inducible isoform. Although the molecular targets of NO are varied (3), the major function of endothelial NO is to activate a soluble guanylate cyclase in underlying smooth muscle cells, leading to relaxation of the muscle.
Pulmonary circulation constricts in response to hypoxic conditions as a negative feedback mechanism for matching ventilation and perfusion (18); however, chronic hypoxia can lead to the development of pulmonary hypertension (21, 31). Several studies suggest that ANG II-enhanced contractions as well as induction of medial hypertrophy contribute to the pathogenesis of pulmonary hypertension (10, 31, 42). Chassagne and colleagues (10) have shown that, during chronic hypoxia, ANG II production is enhanced locally in the lung and that in the presence of ANG II there is a sustained hypoxic vasoconstriction (42). Furthermore, Rothman and coworkers (39) demonstrated that ANG II stimulates proliferation of smooth muscle cells from resistant pulmonary arteries and causes hypertrophy of smooth muscle cells from large conduit pulmonary arteries. Although it is known that pulmonary smooth muscle cell contraction and proliferation are mediated via the AT1 receptor, very little is known about the role that the AT2 plays in regulating pulmonary vascular tone. The present study tested the hypothesis that ANG II increases NO production via an endothelial AT2 receptor and that signaling via this AT2 receptor serves to modulate the vasoconstriction effect of ANG II in pulmonary arterial vessels.
MATERIALS AND METHODS
Bovine pulmonary artery endothelial cells (BPAECs) and bovine pulmonary artery smooth muscle cells (BPASMCs) were prepared and characterized as previously described (32). Control and treated cells were matched in each experiment for cell line, passage number (3–6), and time to monolayer confluence. BPAECs were grown to 90% confluence and made quiescent for 24 h before stimulation with ANG II.
RT-PCR analysis of the AT1 and AT2 receptor subtypes.
RNA was isolated from BPAECs using the DNA-free Kit from Ambion (Austin, TX). One microgram of total RNA was used to synthesize single-strand cDNA in a 20 μl-reaction mixture according to the protocol of Reverse Transcription System (Promega, Madison, WI). Two-microliter reaction mixtures were used for real-time PCR with the LightCycler System (LightCycler-Faststart DNA Master SYBR Green I, Roche). AT1 cDNA was amplified with the following primers, forward 5′-CAG GTG CAT TTG GCA TAG TG-3′ and reverse 5′-ATC ACC ACC AAG CTG TTT CC-3′, extending from base 361 to base 561. For the AT2 receptor the forward primer was 5′-CTT CCT CTC TGG GCA ACC TA-3′ and reverse 5′-TAA GAT GCT TGC CAG GGA TT-3′, extending from base 11 to base 207. For an internal standard, the cDNA for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was amplified with the forward primer 5′-CTC CCA ACG TGT CTG TTG TG-3′ and reverse primer 5′-CCC TGT TGC TGT AGC CAA AT-3 (from bases 35 to 306).
Total RNA was isolated from BPASMCs and BPAECs as described (32). Thirty micrograms of denatured RNA were blotted onto BrightStar TM-Plus positively charged nylon membranes (Ambion). The RNA was fixed to the membrane by baking at 80°C for 15 min and then probed with radiolabeled AT1, AT2, and GAPDH PCR products (Megaprime TM DNA Labeling System; Amersham, Piscataway, NJ).
Flow cytometry studies.
Aliquots of BPAECs (3–6 × 103) were removed from culture and fixed in 100% methanol at −20°C for 2 h. The fixed cells were incubated with a 1:100 dilution of either anti-AT1 or anti-AT2 primary antibodies (rabbit polyclonals; Santa Cruz Biotechnology, Santa Cruz, CA) for 2 h at room temperature. The secondary swine anti-rabbit antibody (Dako, Fort Collins, CO) conjugated with fluorescein isothiocyanate (FITC) was then added at a dilution of 1:20 for 30 min at room temperature. After washing, the cells were incubated with 10 μg/ml propidium iodide (PI) and 100 μg/ml RNase in phosphate-buffered saline for 20 min in the dark at room temperature. Cellular fluorescence was measured with the FACScan flow cytometer (Becton Dickinson, Franklin Lakes, NJ). The red (PI) and green (FITC) fluorescence emissions from each cell were separated and quantitated with the standard optics of the FACScan to measure the DNA content and protein expression simultaneously. The fluorescence was expressed in relative units. We calculated the percentage of cells positive for each receptor using the threshold established by the IgG controls. In each instance, utilizing the simultaneous measurement of DNA content, we calculated the percentage of positive cells only for G1 cells to avoid any differences in cell cycle distribution between samples.
Electrophoresis and Western blot analysis.
Proteins (30 μg) in HEPES buffer (50 mM HEPES, pH 7.4, containing 2 μM leupeptin, 2 μM pepstatin, 0.5 mM phenylmethylsulfonyl fluoride, and 1 mM sodium orthovanadate) were subjected to SDS-PAGE and transferred to nitrocellulose membranes. Expression of the AT1 and AT2 receptors by Western blot analysis was performed with anti-AT1 receptor and anti-AT2 receptor antibodies (Santa Cruz Biotechnology).
We detected the presence of eNOS using an anti-eNOS antibody (Transduction, Lexington, KY) followed by a donkey anti-rabbit secondary antibody conjugated to horseradish peroxidase. Peroxidase activity was determined with an ECL Western blotting kit (Amersham) and by exposure of membranes to X-ray film. The relative amount of eNOS protein was quantitated by laser densitometry and analyzed with a Hewlett Packard Scanjet IIcx. Image analysis was performed with SigmaScan/Image Software (Jandel Scientific). We monitored protein loading in each lane of the gel by probing the membranes with an anti-β-actin antibody (Sigma, St. Louis, MO). The eNOS protein level in control cells was arbitrarily set at 100%, and the levels in treated cells are shown relative to control cells.
Preparation of BPAEC membrane and cytosolic fractions.
BPAECs were treated with 1 μM ANG II for the desired time. The cells were then homogenized in a Dounce homogenizer with 1.5 ml buffer A (20 mM Tris·HCl, pH 7.5, 2 mM EDTA, 0.5 mM EGTA, 100 μM phenylmethylsulfonyl fluoride, 1 μM leupeptin, 1 μM pepstatin, and 0.1 μM aprotinin containing 0.33 M sucrose). The homogenate was then centrifuged at 1,000 g for 10 min, 4°C. The postnuclear fraction (after 1,000 g) was then subjected to ultracentrifugation at 100,000 g for 1 h at 4°C. The supernatant, which contains cytosolic protein, was collected. The pellet was solubilized in buffer A with 1% Triton X-100, incubated on ice for 1 h, and then centrifuged at 100,000 g for 30 min at 4°C. The resulting supernatant contained solubilized membrane proteins.
Analysis of eNOS mRNA levels.
Isolation of total RNA and analyses of eNOS mRNA and 18S rRNA levels were performed as described previously (32). The eNOS mRNA level in control cells was arbitrarily set at 100%, and the levels in treated cells are shown relative to control cells.
Assay of NOS activity by quantitation of nitrite production.
Confluent BPAECs in six-well plates were cultured in DMEM containing 2% fetal bovine serum, without phenol red and antibiotics. The cells were treated with buffer, 10 μM losartan (AT1 receptor antagonist, Du Pont Merck Pharmaceutical), or 10 μM PD-123319 (AT2 receptor antagonist, Parke Davis Pharmaceutical) for 15 min before the addition of 1 μM ANG II. After 8 h, nitrate reductase was added to the culture media to convert nitrate into nitrite (NO2−). NO production was then measured as the amount of its stable metabolite, NO2−, with the Colorimetric Nitric Oxide Assay Kit (Calbiochem, San Diego, CA). Absorbance was measured at 540 nm, and NO2− concentration was determined with sodium nitrite as a standard. The amount of NO2− formed in the media was normalized to the protein content in the respective dishes.
BPAECs were pretreated with 100 μM Nω-nitro-l-arginine methyl ester hydrochloride (l-NAME), a NOS inhibitor, for 60 min before the addition of 1 μM ANG II. After 8 h the amount of NO2− in the media was determined.
Tension studies with isolated intrapulmonary arteries.
Secondary branches of bovine intrapulmonary arteries were isolated and cut into rings (3–4 mm in diameter and 3–4 mm in width). Arterial rings were mounted on wire hooks attached to force displacement transducers (Colbourn Instruments) for measurement of changes in isometric tension. The arterial rings were perfused with Krebs-bicarbonate buffer (in mM: 118 NaCl, 1.5 CaCl2, 25 NaHCO3, 1.1 MgSO4, 1.2 KH2PO4, and 5.6 glucose) at 37°C, gassed with air-5% CO2, and equilibrated for 1 h during which passive tension was adjusted to 5 g. Baseline tension of 5 g has been found to be optimum for bovine pulmonary arteries to generate maximal contraction. The arteries were treated with 120 mM K+, reequilibrated with Krebs buffer, and used in subsequent experiments.
To examine the effects of receptor antagonists on ANG II-induced acute vasoconstriction, we either left the arteries untreated or treated them with 10 μM PD-123319, 10 μM losartan, or 10 μM candesartan celextil (AT1 receptor antagonist, Astra Zeneca) for 15 min. ANG II (1 μM) was added to each of the vessels, and changes in pulmonary vessel tone were measured immediately.
For the hypoxic studies, the pulmonary arterial rings were either left untreated or treated with 1 μM ANG II. At the desired times (2, 4, 6, and 8 h following ANG II stimulation), hypoxic challenge was measured by the following procedure. The vessels were precontracted with 3 × 10−8 M U-46619, a stable thromboxane A2 analog, which provides a stronger, more reproducible, hypoxic contraction in these arteries. After steady-state tone was reached (2.5 g), the gas was changed from air/5% CO2 to 95% N2/5% CO2, hypoxic contractions were recorded for 10 min, and the tissues then reoxygenated with air/5% CO2. To examine the effects of receptor antagonists, we treated the rings with either 10 μM candesartan or 10 μM PD-123319 for 15 min before the addition of buffer or 1 μM ANG II. After 6 h, hypoxic vasoconstriction was measured. The 6-h time point was chosen based on the time frame in which ANG II had stimulated a significant increase in eNOS protein expression in BPAECs (32). Hypoxic challenge was performed in the presence of 10 μM indomethacin to rule out the contribution of vasodilator prostaglandins. To examine the role of the endothelium in modulating ANG II-dependent hypoxic vasoconstriction, we removed the endothelium by gently rubbing the lumen of the vessel, and confirmation was determined by the loss of a relaxant response to acetylcholine (10−8–10−6 M). To determine the effect of eNOS inhibition on hypoxic vasoconstriction, we treated vessels with 100 μM l-NAME 20 min before the hypoxic challenge. Data are reported as the increase in force caused by hypoxia (in the presence and absence of ANG II) above the 2.5 g U-46619-induced tone.
Data are expressed as means ± SE, where n refers to the number of experiments conducted with different cell preparations or animals. Receptor antagonist studies were analyzed by one-way ANOVA with Tukey’s posttest. All other statistical analysis was performed by a Student’s t-test for paired data between respective control and experimental groups. Statistical significance was accepted at P < 0.05.
BPAECs express both AT1 and AT2 receptors.
Confirmation of AT1 and AT2 receptor expression on BPAECs was performed by RT-PCR analysis, Western blot analysis, and flow cytometry. First, expression of the AT1 and AT2 receptor subtype mRNAs in BPAECs was examined by RT-PCR analysis (Fig. 1A) followed by dot-blot analysis with the radiolabeled PCR products (Fig. 1B). The expected sizes for the PCR products are 201 base pairs for AT1 receptor, 197 base pairs for the AT2 receptor, and 274 base pairs for GAPDH. Dot-blot hybridization analysis (Fig. 1B) with the radiolabeled PCR products demonstrated that BPAECs express both the AT1 and AT2 receptor messages, whereas BPASMCs express only the AT1 receptor mRNA.
To demonstrate that BPAECs translate the receptor mRNAs into proteins, we investigated expression of the AT1 (50 kDa) and AT2 (50 kDa) receptor proteins by Western blot analysis (Fig. 2A). BPAECs express the AT1 receptor, the AT2 receptor, and β-actin (42 kDa), whereas BPASMCs express only the AT1 receptor protein and α-smooth muscle actin (42 kDa). In the presence of the AT2 receptor-competing peptide, the 50-kDa molecular mass band recognized by the anti-AT2 receptor antibody was completely blocked, whereas the competing peptide had no effect on the 50-kDa molecular mass band recognized by the anti-AT1 receptor antibody.
To determine the percentage of BPAECs that express both the AT1 and AT2 receptors on their plasma membranes, we detected the levels of expression of AT1 and AT2 receptors immunocytochemically on intact individual cells by multiparameter flow cytometry in conjunction with cellular DNA content. Flow cytometry studies (Fig. 2B) demonstrated that >96% of BPAECs express both AT1 and AT2 receptors on their plasma membranes (representative data from three different cell preparations). To show specificity of the ANG II receptor antibodies, we coincubated the AT1 receptor antigen peptide (5 μg) with the AT1 and AT2 receptor antibodies. In the presence of the competing peptide, the detection of AT1 expression was completely blocked, whereas >97% of the cells remained positive for the AT2 receptor.
Dose-response curve, kinetics, and subcellular location of ANG II-stimulated eNOS protein expression.
We have previously reported that ANG II stimulates a significant increase in eNOS protein expression at 8 h (32). In the present study, we further characterized the effects of ANG II on eNOS protein expression. ANG II stimulated a dose-dependent increase in eNOS protein expression at 8 h. An increase in ANG II-induced protein expression was seen at a concentration as low as 1 nM ANG II (2.23 ± 0.37-fold), with 1 μM giving maximal stimulation (4.5 ± 0.92-fold; n = 7, P < 0.05; Fig. 3A) in BPAECs. However, at higher ANG II concentrations (10 μM), eNOS protein expression was increased only 2.10 ± 0.55-fold over unstimulated controls. If the BPAECs were pretreated with losartan, the 10 μM ANG II-dependent increase in eNOS protein expression was similar to that seen with 1 μM ANG II alone (data not shown). To demonstrate that the stability of ANG II is not critical to the response of the BPAECs, we removed the ANG II-containing medium at 15 min and then added fresh serum-free medium to the cells. After 8 h, BPAECs produced a significant increase in eNOS protein expression. This suggests that once ANG II is added to the cell culture media, ANG II binds to its receptor(s) on BPAECs and the signaling pathways linked to the receptors are immediately activated. Because 1 μM ANG II gave maximal stimulation, this concentration was chosen for succeeding experiments.
We next examined the subcellular location of this ANG II-dependent increase in eNOS expression. As determined by Western blot analysis, ANG II (1 μM) stimulated a time-dependent increase in eNOS protein expression in both the cytosolic and membrane-enriched fractions (Fig. 3B). However, there is a greater increase in the membrane fraction consistent with its localization in the plasmalemmal caveolae (41). Presence of caveolae in the Triton X-100 fraction was confirmed by Western blot analyses with an anti-caveolin antibody (Santa Cruz Biotechnology, data not shown).
Effects of ANG1–7 and ANG3–8 on eNOS protein expression.
Because published studies (6, 9, 33, 35) suggest that it may be another peptide product of the RAS such as ANG1–7 or ANG3–8 that is responsible for the vasodilatory effect of ANG II, we examined whether these ANG II fragments could stimulate NO production in BPAECs. Treating BPAECs with either ANG1–7 or ANG3–8 (1 μM) for 2–8 h did not stimulate a significant increase in eNOS protein expression (data not shown).
ANG II-dependent increase in NO production is mediated via the AT2 receptor.
To determine which receptor subtype mediates the ANG II-dependent increase in NO production in BPAECs, the effects of AT1 and AT2 receptor antagonists on eNOS mRNA, protein, and NO2− production were determined. BPAECs were pretreated for 15 min with either 10 μM losartan (an AT1 antagonist), 10 μM PD-123319 (an AT2 antagonist), or both followed by stimulation with 1 μM ANG II for 6 h (32). Losartan alone (2.5 ± 0.4-fold vs. control; n = 5, P < 0.05), as well as in the presence of ANG II (3.2 ± 0.95-fold vs. control; n = 5, P < 0.01), stimulated an increase in eNOS mRNA expression (Fig. 4A). In contrast, PD-123319 alone had no effect on eNOS mRNA levels; however, it did prevent the ANG II (1 μM)-dependent increase in eNOS mRNA levels (1.5 ± 0.35-fold vs. 2.6 ± 0.32-fold increase, respectively; Fig. 4A). The losartan induction of eNOS mRNA expression, alone and in the presence of ANG II, was prevented when the cells were also pretreated with PD-123319 (Fig. 4A).
PD-123319 (10 μM) also prevented the ANG II (1 μM)-dependent increase in eNOS protein expression at 8 h (2.74 ± 0.94-fold; n = 6, P < 0.05, Fig. 4B). Similar to its effects on eNOS mRNA levels, losartan stimulated an increase in eNOS protein expression alone and in the presence of ANG II (approximately fourfold, Fig. 4B). Additionally, when BPAECs were pretreated for 15 min with a different AT1 receptor antagonist, candesartan (10 μM), there was also an increase in eNOS protein expression at 8 h (3.8 ± 0.25-fold increase over basal; n = 3, P < 0.05). When BPAECs were pretreated with both losartan and PD-123319, there was not an increase in eNOS protein expression in either control or ANG II-treated cells (Fig. 4B). Treatment of BPAECs with increasing concentrations of the AT4 receptor antagonist Nor1,Leu3 ANG IV (0.1–10 μM, Pacific Northwest Technology) did not block the ANG II-dependent increase in eNOS protein expression at 8 h (data not shown).
There was an approximate twofold increase in NO2− production in ANG II-stimulated BPAECs compared with control (97.3 ± 10 vs. 43.4 ± 5.7 nmol/mg protein, respectively; n = 6, P < 0.05; Fig. 4C). When BPAECs were treated with losartan alone, there was a significant increase in NO production (84.4 ± 17.7 nmol/mg protein; n = 6, P < 0.05) that was further enhanced in the presence of ANG II (128.5 ± 25.5 nmol/mg protein; n = 6, P < 0.01). Pretreatment of BPAECs with PD-123319 blocked the ANG II-dependent NO2− accumulation (22.7 ± 11.6 nmol/mg protein) as well as prevented the losartan-induction (alone and in the presence of ANG II) of NO2− accumulation at 8 h (Fig. 4C).
Because the losartan enhancement of eNOS protein expression was prevented by PD-123319, we wanted to determine whether the losartan-dependent increase in eNOS protein expression involves endogenous synthesis of ANG II that, in turn, may selectively stimulate the AT2 receptor. Preincubation of BPAECs with an anti-ANG II antibody (10 μg/ml, Santa Cruz Biotechnology) for 5 min prevented both the ANG II- and losartan-dependent increase in eNOS protein expression at 8 h (Fig. 5). Control IgG had no effect on eNOS protein expression.
l-NAME, an analog of arginine that inhibits NO production, decreased basal production of NO and also prevented the ANG II-dependent increase in NO2− accumulation at 8 h (97.3 ± 10 vs. 0.5 ± 0.43 nmol/mg protein; n = 6, P < 0.01).
Effect of dithiothreitol on eNOS protein expression.
Because it is known that ANG II receptor subtypes have differential sensitivity to dithiothreitol (DTT; Ref. 12), we used this as a second approach to investigate the roles of the receptors in mediating eNOS protein expression. ANG II-dependent stimulation of eNOS protein expression at 8 h (2.90 ± 0.33-fold; n = 6, Fig. 6) was not affected by increasing concentrations of DTT from 0.1 to 2 mM, consistent with this event being mediated via the AT2 receptor. In addition, DTT increased eNOS protein expression in a dose-dependent manner (Fig. 6) in BPAECs with significant increases seen with 1 and 2 mM DTT. Given that DTT preferentially inactivates the AT1 receptor, these data support the losartan data, suggesting that the AT1 receptor may tonically inhibit eNOS protein expression.
Pulmonary artery vessel tension studies.
The data presented thus far demonstrate that the ANG II-dependent increase in NO production is mediated via the AT2 receptor in BPAECs. The aim of the next set of experiments was to investigate whether this endothelial AT2 receptor plays a physiological role in the effects of ANG II on pulmonary vascular tone. Previously published studies (42, 45) demonstrated that the AT2 receptor is linked to acute activation of NOS, so we first investigated whether the AT2 receptor plays a role in mediating ANG II-dependent acute vasoconstriction in isolated pulmonary arteries. To achieve this goal, we treated pulmonary arteries with losartan, candesartan, or PD-123319 for 15 min. ANG II was then added to the vessels, and contraction was measured immediately. Losartan inhibited ∼55% of the ANG II-dependent acute vasoconstriction (3.92 ± 0.4 vs. 8.6 ± 0.74 g; n = 8, P < 0.05), whereas candesartan completely blocked the response (Fig. 7). In contrast, PD-123319 had no effect on ANG II-induced vasoconstriction (7.2 ± 1.8 vs. 8.6 ± 0.74 g, n = 8). These data confirm that acute ANG II-dependent vasoconstriction is mediated via the AT1 receptor and that signaling via the AT2 receptor does not modulate acute responses to ANG II in pulmonary arteries.
Because ANG II causes a sustained enhancement of hypoxic pulmonary vasoconstriction (HPV) (42), we then chose to examine the role of the AT2 receptor in regulating this ANG II response. To confirm that the effects of ANG II were due to the peptide and/or receptor antagonist and not due to a change in vessel reactivity over time, hypoxic vasoconstriction in control and ANG II-treated vessels for 2 to 8 h was examined. In arterial rings treated with the cyclooxygenase inhibitor indomethacin alone (control vessels), HPV remained relatively stable over an 8-h period, ranging from 2.76 ± 0.46 g at 2 h to 2.00 ± 0.66 g at 8 h (Fig. 8A). Furthermore, in the ANG II-treated vessels, hypoxic vasoconstriction began to increase at 2 h (4.23 ± 0.42 g) and remained significantly enhanced up to 8 h (3.78 ± 1.17 g; n = 12, P < 0.05 at each time point).
We then examined the roles of the AT1 and AT2 receptors in modulating hypoxic vasoconstriction at 6 h, a time at which there is a significant ANG II-dependent increase in eNOS protein expression (see Fig. 3B). To perform this study, we pretreated the vessels with either candesartan or PD-123319 for 15 min before stimulation with ANG II. After 6 h, hypoxic vasoconstriction was determined. The AT1 receptor antagonist candesartan completely blocked the ANG II-dependent increase in hypoxic vasoconstriction (4.85 ± 1.32 g vs. 2.64 ± 0.76 g; n = 12, P < 0.05; Fig. 8B). Candesartan was used in this part of the study because it completely inhibited acute ANG II-dependent contraction, whereas losartan inhibited only 55% of the response (see Fig. 7). In contrast, treatment of the vessels with the AT2 receptor antagonist PD-123319 augmented hypoxic vasoconstriction in the ANG II-treated rings (6.93 ± 2.94 g) compared with those treated with ANG II alone (4.85 ± 1.32 g; n = 12, P < 0.05; Fig. 8B). If vessels were pretreated with both candesartan and PD-123319, ANG II did not enhance hypoxic contractions (data not shown). When the endothelium was removed from the pulmonary arteries, PD-123319 no longer enhanced hypoxic vasoconstriction (Fig. 8B; n = 3), suggesting that AT2 receptor-dependent enhancement of hypoxic vasoconstriction requires the presence of intact endothelium. ANG II enhanced hypoxic vasoconstriction in the endothelium-denuded vessels, but it did not reach statistical significance.
Treatment of the pulmonary vessels with 100 μM l-NAME for 20 min before hypoxic challenge stimulated an approximately two- to threefold increase in pulmonary contraction compared with control vessels (5.77 ± 1.5 vs. 2.42 ± 0.38 g, n = 6). However, when the vessels had been pretreated with 100 μM l-NAME, there was no difference in the hypoxic vasoconstriction between the control and ANG II-treated vessels.
Because ANG1–7 and ANG3–8 have been shown previously to elicit vasodilatory responses in the lungs (6, 33, 35), we examined the effects of these metabolites on both acute vessel responses and on the ANG II-enhanced HPV. Neither ANG3–8 nor ANG1–7 induced a vasodilatory response in pulmonary arterial vessels (Fig. 7). In addition, ANG3–8 did not affect the hypoxic contraction of the vessels at 6 h, either alone (2.42 ± 0.38 g, n = 6) or in the presence of ANG II (4.12 ± 0.83 g, n = 6). Furthermore, Nor1,Leu3 ANG IV, an AT4 receptor antagonist, did not affect the ANG II-dependent acute vasoconstriction, nor did it affect the ANG II-dependent enhanced hypoxic vasoconstriction at 6 h (data not shown).
In the present study, we provide experimental evidence that the AT2 receptor is linked to an increase in NO production in pulmonary endothelial cells and that signaling via the endothelial AT2 receptor serves to attenuate the ANG II-dependent enhanced hypoxic vasoconstriction in pulmonary arteries. The presence of both AT1 and AT2 receptors on BPAECs was confirmed by a variety of techniques including RT-PCR, Western blot analysis, and flow cytometry. Although the AT1 receptor has been identified in most vascular tissues, there is a discrepancy in the literature regarding AT2 receptor expression. Originally the AT2 receptor was thought to be expressed mainly in fetal tissue; however, numerous studies have now reported the expression of the AT2 receptor in adult adrenal tissue, kidney, brain, and heart (15). In BPAECs, ANG II induced an approximately two- to threefold increase in the levels of eNOS mRNA and protein, consistent with that seen with many other stimuli (49). The dose-response studies demonstrated that ANG II at concentrations (10 nM) seen in hypertensive states (37) can stimulate an increase in eNOS protein expression. Furthermore, emerging data have demonstrated that ANG II can be generated at tissue sites, suggesting that at times local concentrations of ANG II (14) can be higher than the concentrations measured in the plasma.
Although there are published studies (38, 40) demonstrating that ANG II increases NO production in multiple cell types, there is disagreement as to the receptor subtype linked to this increase in NO. Moreover, ANG II has been shown to stimulate NO production both by increasing eNOS gene expression (32, 38) and by activating eNOS enzyme activity (40). Data obtained with the specific AT1, AT2, and AT4 receptor antagonists, as well as the studies examining the sensitivity to DTT, demonstrated that ANG II stimulates an increase in eNOS mRNA, which, in turn, leads to an increase in eNOS protein and NO production in BPAECs, via the AT2 receptor. In agreement with our studies, Ritter and colleagues (38) found that ANG II increases eNOS protein expression via an AT2 receptor, calcineurin-NF-AT pathway in cardiomyocytes. On the other hand, Siragy and Carey (43) found that activation of the AT2 receptor leads to acute production of NO via a bradykinin-dependent pathway.
Our results are in contrast to those that demonstrate that ANG II increases NO production by activating eNOS enzyme via the AT1 receptor (8, 40). In fact, treatment of BPAECs with the AT1 receptor antagonist losartan or inactivation of the AT1 receptor with DTT resulted in increased NO production, suggesting that AT1 receptor-linked signaling pathway(s) functions to inhibit NO production. Stimulation of the AT1 receptor can lead to Gαq/11-mediated activation of PLC, generating diacylglycerol and inositol (1,4,5)trisphosphate (IP3). Diacylglycerol can then activate protein kinase C (PKC), whereas IP3 leads to an increase in cytosolic calcium. Previously published studies (7, 16, 29, 31a, 38) suggest that components of this AT1 receptor-linked pathway may regulate NO production; however, the data are not consistent. Although Ohara and colleagues (31a) demonstrated that PKC decreases NOS mRNA expression, Li et al. (29) demonstrated that phorbol esters stimulate an increase in eNOS expression in endothelial cells. Furthermore, there is evidence supporting a role for calcium in both upregulation (7, 38) and downregulation (16) of NOS expression. Preliminary data from our lab (Zhao X, Li J, Li X, and Olson S, unpublished observations) suggest that an AT1/PLC/Ca2+/PKC-dependent pathway downregulates eNOS protein expression in BPAECs.
Both an anti-ANG II antibody and an AT2 receptor antagonist blocked the losartan-dependent increase in eNOS protein expression in BPAECs. Collectively, these data suggest that the mechanism by which an AT1 receptor antagonist leads to an increase in eNOS protein expression may involve endogenous production of ANG II, which, in turn, selectively stimulates the AT2 receptor. In addition, in the presence of the AT1 receptor antagonist, there may be some shunting of ANG II from the AT1 to the AT2 receptor. Our results are consistent with Thai et al. (47), who found that AT1 receptor blockade enhances vasorelaxation in heart failure by an AT2 receptor-mediated increase in NO bioavailability. In addition, Klingbeil and colleagues (28) found that in a group of hypertensive patients blockade of the AT1 receptor with valsartan improves basal production and release of NO. Consequently, the beneficial effects of AT1 receptor antagonists may include both inhibition of vascular smooth muscle cells (VSMCs) vasoconstriction and increased NO production by endothelial cells. Although our data demonstrate that the AT1 receptor is linked to decreased NO production, they also show that a functional AT2 receptor is required for the increased NO production, supporting an important role for the AT2 receptor in the therapeutic effects of AT1 receptor antagonists.
A number of reports (6, 9, 33, 35) have shown that angiotensin fragments such as ANG1–7 and ANG3–8 may mediate some of the effects of ANG II in the cardiovascular system. Previous studies have shown that these peptides elicit a vasodilatory response in the lung (6, 33, 35); however, our studies do not support a role for these metabolites in modulating pulmonary vascular tone. Neither ANG1–7 nor ANG3–8 stimulated an increase in eNOS protein expression in BPAECs, nor did they elicit a vasodilatory response in pulmonary arterial vessels. This discrepancy may be due to different species and cell types, different experimental conditions and ANG II concentrations, as well as different lengths of time of exposure to ANG II.
Several studies suggest that ANG II may be an important modulator of hypoxia-dependent pulmonary hypertension (10, 31, 39, 42); nevertheless, very little is known about the ANG II receptors and the mechanism by which they regulate pulmonary vasoconstriction and vessel remodeling. Recently, Chassagne and colleagues (10) demonstrated that chronic hypoxia induces a transient increase in AT1 and AT2 receptors in the rat lung. Although they were able to demonstrate that the vasoconstrictive response to ANG II was due mainly to activation of the AT1 receptor, they were unable to identify a role for the AT2 receptor in their hypoxic model. Our data suggest that preexposure to ANG II causes a time-dependent enhancement of hypoxic contractions that may be regulated by both AT1 and AT2 receptors. ANG II enhancement of hypoxic vasoconstriction is mediated via the AT1 receptor, whereas signaling via an endothelial AT2 receptor serves to modulate this response possibly through the production of a vasodilator. Therefore, impairment of the synthesis of NO via AT2 receptors could contribute to the development of HPV. Several other studies have indicated a role for AT2 receptor subtype in blood pressure regulation using AT2 receptor phosphorothiolated antisense oligonucleotides (30), transgenic mice lacking AT2 receptor (23, 25), and transgenic mice overexpressing the AT2 receptor in VSMCs (48).
The endothelium produces many vasoactive compounds known to regulate pulmonary vascular tone (1); however, the role of the endothelium in hypoxic vasoconstriction remains controversial. Even though ANG II enhanced hypoxic vasoconstriction in bovine pulmonary arteries when the endothelium was removed, it did not reach statistical significance. Nonetheless, endothelial denudation did prevent the PD-123319 enhancement of ANG II-induced hypoxic contractions in these vessels, suggesting that ANG II stimulates the production of both endothelium-derived contracting and relaxing factors that can ultimately affect pulmonary vascular tone. Additionally, the cell experiments support the isolated pulmonary vessel studies, suggesting that the AT2 receptor-linked vasodilator may be endothelium-derived NO, as pulmonary VSMCs do not express the AT2 receptor.
The concentration of l-NAME (100 μM) that significantly inhibited NO production in BPAECs augmented hypoxic contractions compared with untreated pulmonary arterial vessels. In support of our results, other investigators have shown that decreased production of NO and/or decreased bioavailability to NO enhances acute hypoxic vasoconstriction in pulmonary arteries (4, 13, 44). Furthermore, Fagan and colleagues (17) demonstrated that NO is an important modulator of the pulmonary vasculature response to mild hypoxia in homozygous and heterozygous eNOS-null mice. However, in contrast to our results, studies have shown that blocking eNOS activity had no effect (22, 26) or even suppressed (46) hypoxic vasoconstriction. In our study, when the pulmonary arteries were treated with l-NAME, there was no significant difference in the hypoxic vasoconstriction between control and ANG II-treated vessels. One possible explanation is that removal of an endogenous vasodilator resulted in maximal contraction under hypoxic conditions and that the vessels were not able to further constrict in response to ANG II. We propose that blocking NO synthesis by l-NAME or blocking the AT2 receptor-dependent increase in NO potentiates hypoxic vasoconstriction, suggesting the NO opposes vasoconstriction in the presence of increased tone.
The results of the present study, together with our previous findings (32), demonstrate that ANG II stimulates an increase in eNOS mRNA and protein expression, as well as an increase in NO production via an AT2 receptor, and that signaling via the AT1 receptor appears to negatively regulate NO production in the pulmonary endothelium. Furthermore, we have demonstrated that the ANG II enhancement of hypoxic contractions in pulmonary arteries is mediated via the AT1 receptor, whereas signaling via an endothelial AT2 receptor serves to limit the severity of the contractions. We propose that this AT2 receptor-dependent increase in eNOS may provide a protective mechanism in the pulmonary circulation when challenged by elevated levels of ANG II, such as that seen during hypoxic conditions and in renin-dependent systemic hypertension. The signaling pathways linked to this AT2 receptor-dependent increase and AT1 receptor-dependent decrease in NO production are currently being investigated.
This research was supported by National Heart, Lung, and Blood Institute Grant HL-63182.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society