Endothelin-1 contributes to increased NFATc3 activation by chronic hypoxia in pulmonary arteries

Abstract

Chronic hypoxia (CH) activates the Ca²⁺-dependent transcription factor nuclear factor of activated T cells isoform c3 (NFATc3) in mouse pulmonary arteries. However, the mechanism of this response has not been explored. Since we have demonstrated that NFATc3 is required for CH-induced pulmonary arterial remodeling, establishing how CH activates NFATc3 is physiologically significant. The goal of this study was to test the hypothesis that endothelin-1 (ET-1) contributes to CH-induced NFATc3 activation. We propose that this mechanism requires increased pulmonary arterial smooth muscle cell (PASMC) intracellular Ca²⁺ concentration ([Ca²⁺]_i) and stimulation of RhoA/Rho kinase (ROK), leading to calcineurin activation and actin cytoskeleton polymerization, respectively. We found that: 1) CH increases pulmonary arterial pre-pro-ET-1 mRNA expression and lung RhoA activity; 2) inhibition of ET receptors, calcineurin, L-type Ca²⁺ channels, and ROK blunts CH-induced NFATc3 activation in isolated intrapulmonary arteries from NFAT-luciferase reporter mice; and 3) both ET-1-induced NFATc3 activation in isolated mouse pulmonary arteries ex vivo and ET-1-induced NFATc3-green fluorescence protein nuclear import in human PASMC depend on ROK and actin polymerization. This study suggests that CH increases ET-1 expression, thereby elevating PASMC [Ca²⁺]_i and RhoA/ROK activity. As previously demonstrated, elevated [Ca²⁺]_i is required to activate calcineurin, which dephosphorylates NFATc3, allowing its nuclear import. Here, we demonstrate that ROK increases actin polymerization, thus providing structural support for NFATc3 nuclear transport.

in rodents, long-term exposure to chronic hypoxia (CH) (more than 1 wk) causes pulmonary hypertension due to pulmonary vasoconstriction, arterial remodeling, and polycythemia, which ultimately results in right ventricular hypertrophy and often heart failure (64). CH and associated pulmonary hypertension also occur in patients with chronic bronchitis, emphysema, cystic fibrosis, and severe restrictive lung diseases or in high-altitude residents (43).

We have previously demonstrated that as little as 2 days of hypoxic exposure maximally activates the Ca²⁺-dependent transcription factor nuclear factor of activated T cells isoform c3 (NFATc3) in mouse pulmonary arterial smooth muscle cells (PASMC) (19, 20). This observation suggests that NFATc3 may act as an early response transcription factor. NFATc3 belongs to a family of four transcription factors (NFATc1, NFATc2, NFATc3, and NFATc4) that share the property of Ca²⁺/calcineurin-dependent nuclear translocation (reviewed in Ref. 41). In addition, we have demonstrated that NFATc3 is required for CH-induced pulmonary arterial remodeling and pulmonary hypertension in mice (19, 20). However, the mechanisms that lead to CH-induced NFATc3 activation are unknown.

Bonnet et al. (9) have demonstrated that NFATc2 is required for PASMC proliferation in patients with primary pulmonary hypertension and in monocrotaline-pulmonary hypertensive rats. In mice, the only NFAT isoform that is activated by CH is NFATc3 and mediates CH-induced PASMC hypertrophy (20).

In smooth muscle, NFATc3 nuclear accumulation and transcriptional activity are increased upon stimulation of Gq/11-coupled receptors (35, 36, 71). Increases in intracellular calcium concentration [Ca²⁺]_i lead to calcineurin activation, which dephosphorylates NFATc3 and exposes nuclear localization signals that allow for NFATc3 nuclear translocation. NFATc3 nuclear accumulation is also controlled by a series of Ser/Thr kinases such as GSK3β, CaM kinase II, and JNK1/2 (34, 36).

Interestingly, the expression of endothelin 1 (ET-1), which activates Gq/11-coupled receptors, is increased in both humans and animals with pulmonary hypertension (8, 27, 81). Furthermore, oral administration of bosentan (dual ET_A and ET_B antagonist) is currently indicated for patients with pulmonary hypertension (32) while selective ET_A receptor (ET_AR) blockers prevent CH-induced pulmonary hypertension in rats (8, 83). However, little is known about the role of NFAT in ET-1 signaling.

During CH there is an increase in ET-1-mediated vasoconstriction in isolated pulmonary arteries, which involves increased PASMC [Ca²⁺]_i through inhibition of voltage-gated potassium (K_V) channels, membrane depolarization, and influx of Ca²⁺ through L-type Ca²⁺ channels (52). In addition, ET-1 increases Ca²⁺ sensitivity (increased muscle tension under constant-Ca²⁺ conditions) (79). RhoA is a member of a subfamily of the Ras superfamily of monomeric G proteins that activates Rho kinase (ROK), mediating ET-1-induced Ca²⁺ sensitization (79) and cytoskeleton reorganization (stress-fiber formation) (40, 42, 72). The cytoskeleton of smooth muscle cells is a filamentous network consisting largely of filamentous actin (F-actin), and to a lesser extent, monomeric actin (G-actin). RhoA/ROK is a signaling pathway that also appears to regulate NFAT activity (80). We have shown that ROK inhibition prevents intermittent hypoxia-induced NFATc3 activation in mouse mesenteric arteries both in vivo and ex vivo (17). However, the mechanism by which RhoA/ROK activates NFATc3 is still unclear.

Therefore, we hypothesized that ET-1 contributes to CH-induced NFATc3 activation through a mechanism that requires increased PASMC [Ca²⁺]_i and stimulation of RhoA/ROK leading to calcineurin activation and actin cytoskeleton polymerization, respectively. Consistent with this hypothesis, we found that: 1) CH increases pulmonary arterial pre-pro-ET-1 mRNA expression and lung RhoA activity; 2) inhibition of ET_AR, calcineurin, L-type Ca²⁺ channels, and ROK blunts CH-induced activation of NFATc3; and 3) ET-1-induced NFATc3 nuclear translocation in isolated mouse pulmonary arteries and human PASMC (HPASMC) depends on ROK and actin polymerization.

METHODS

All protocols employed in this study were reviewed and approved by the Institutional Animal Care and Use Committee of the University of New Mexico, School of Medicine (Albuquerque, NM).

Animals.

Adult male 9x-NFAT-luciferase reporter (NFAT-luc), NFATc3 knockout (KO), and BalB/C wild-type (WT) mice (20–25 g) were used. NFAT-luc mice were provided by Dr. Jeffery D. Molkentin (Department of Pediatrics, Children's Hospital Medical Center, Cincinnati, Ohio) (11). NFATc3 KO mice were a gift from Dr. Laurie Glimcher (Harvard University) (63). Heterozygous mice were bred to obtain age-matched WT (Balb/C) and KO.

Chronic hypoxia exposure.

Animals designated for exposure to CH were housed in a hypobaric chamber with barometric pressure maintained at ∼380 Torr for 2, 7, or 21 days. Control animals were housed at ambient barometric pressure (normoxia, N, ∼630 Torr). All animals were maintained on a 12:12-h light-dark cycle.

Animal treatments.

All treatments were started 1 day before CH exposure, followed by 2 days during CH. NFAT-luc mice were treated as follows: 1) bosentan (dual ET receptor antagonist; 30 mg·kg⁻¹·day⁻¹; gift from Actelion Pharmaceuticals, Switzerland) was administered in the drinking water; 2) cyclosporine A (CsA, calcineurin inhibitor; 25 mg·kg⁻¹·day⁻¹; Calbiochem) or Cremophor EL (vehicle, Sigma) was injected subcutaneously; 3) PD155080 (selective ET_AR antagonist; 50 μg·kg⁻¹·day⁻¹; College of Pharmacy, University of New Mexico) or placebo was administered in oral dough pellets (17, 84); 4) diltiazem (L-type Ca²⁺ channel inhibitor; 100 mg·kg⁻¹·day⁻¹; Sigma) or saline (vehicle) was administered subcutaneously via osmotic pumps (Alzet); or 5) the ROK inhibitors fasudil (30 mg·kg⁻¹·day⁻¹), HA 1152 (1 mg·kg⁻¹·day⁻¹) (both from LC Laboratories), or saline (vehicle) was administered subcutaneously via osmotic pumps.

Luciferase activity.

Intrapulmonary arteries were isolated from NFAT-luc mice and lysed (Promega buffer). Luciferase activity was measured using a Luciferase Assay System kit (Promega), and light was detected with a luminometer (TD20/20, Turner). Protein content was determined by the Bradford method (Bio-Rad) or bicinchoninic acid protein assay reagent (Pierce) and used to normalize luciferase activity.

Quantitative real-time polymerase chain reaction.

Isolated intrapulmonary arteries from NFAT-luc (exposed to normoxia, 2, 7, and 21 days of CH), NFATc3 WT and KO (exposed to normoxia and 2 days CH) mice were stored in RNAlater (Ambion). Total RNA was isolated using the RNeasy Mini Kit (Qiagen) and reverse transcribed to cDNA using a high capacity reverse transcription kit (Applied Biosystems). For real time detection of pre-pro-ET-1 transcripts, SYBR Green Master Mix (Applied Biosystems) was used as previously described (18). The normalized gene expression method (2^−ΔΔCT) for relative quantification of gene expression was used (53).

RhoA activity.

RhoA activity was determined in whole lung extracts using the G-LISA Small G-protein Activation Assay from Cytoskeleton. RhoA activity could not be detected with this assay in isolated intrapulmonary arteries due to limited sensitivity of the assay. An alternative is the GTP-bound RhoA pull-down assay (cytoskeleton); however, this assay requires more protein that could be extracted from these small arteries.

NFATc3 nuclear import.

HPASMC (Invitrogen) were grown on polylysine-coated plates in Media 231 with smooth muscle growth supplement (Invitrogen) at 37°C in 5% CO₂ with controlled humidity. Cells were used up to passage 5. HPASMC were electroporated with NFATc3-green fluorescent protein (GFP) expression vector using Nucleofector (Lonza). The vector was created by Dr. F. McKeon (Harvard University, Cambridge, MA) and kindly provided by Dr. L. F. Santana (Washington State University, Seattle, WA).

NFATc3-GFP expressing HPASMC were seeded on microscope coverslips and cultured for at least 48 h on Media 231 with smooth muscle differentiation supplement (Invitrogen). At the time of experimentation, the coverslip was mounted on a heating plate chamber and maintained in HEPES-PSS at ∼37°C during the experiment. Nuclear GFP fluorescence was monitored using a Nikon Diaphot 300 at ×200 magnification. Individual HPASMC were imaged once every 30 s for 30 min. All cells were incubated for 5 min with a myosin light chain peptide inhibitor (Peptide 18, 10⁻⁶ M, Calbiochem) to reduce cell shrinkage in response to constrictors. Cells were treated with the CRM1 exportin inhibitor leptomycin B (4 × 10⁻⁸ M, LC Laboratories) to prevent NFATc3-GFP nuclear export. Immediately, cells were treated with vehicle or ET-1 (10⁻⁸ M, Sigma) in the presence or absence of the ROK inhibitor fasudil (10⁻⁶ M, LC Laboratories); the actin polymerization inhibitor cytochalasin B (10⁻⁶ M, Sigma), or the well-established actin stabilizing agent (14) jasplakinolide (0.05 μM, Sigma). Images were captured using Andor IQ 1.9 software. Nuclear fluorescence (F) was background corrected and expressed as fold change from baseline nuclear fluorescence (F/F₀) (Metamorph Universal Imaging software).

[Ca²⁺]_i measurement.

HPASMC were loaded with a mix of fura-2 AM (2 μM)-pluronic acid (0.02%) in HEPES PSS solution for 15 min at room temperature, followed by 20 min wash with HEPES PSS at 37°C. The ratio of 340/380 nm fura-2 AM fluorescence was monitored every 300 ms for 17 min using a NIKON Diaphot 300 at ×200 magnification. Images were captured using Andor IQ 1.9 software. Cells were treated with ET-1 in the presence or absence of leptomycin B, fasudil, or cytochalasin B. Fluorescence at 340 and 380 nm was background corrected, and the ratio was expressed as percent fold change from baseline.

Immunofluorescence confocal microscopy.

NFATc3-GFP expressing HPASMC were incubated with vehicle or ET-1 for 5 min at 37°C in the presence or absence of fasudil or cytochalasin B. Cells were formaldehyde fixed (4% in phosphate-buffered saline, PBS), permeabilized, and blocked for nonspecific binding. Primary antibody (rabbit polyclonal anti-NFATc3, 1:100; Santa Cruz) was prepared in 0.2% gelatin in PBS and applied overnight at 4°C. Secondary antibody (anti-rabbit Cy5; Jackson Immunoresearch Laboratories) was prepared in 0.2% gelatin in PBS and applied for 1 h together with 6.6 × 10⁻⁶ M Alexa Fluor 488 phallodine (Invitrogen) and 0.64 × 10⁻⁶ M Alexa Fluor 594 DNase I (Invitrogen) at room temperature. Nuclei were stained using Hoechst 33342 (1:2,000 in PBS; Invitrogen). Cells were imaged with a ×63 objective on a META Zeiss 510 laser scanning confocal microscope. Specificity of immune staining was confirmed by the absence of fluorescence in cells incubated with primary or secondary antibodies alone. NFATc3-F-actin (phallodine) colocalization was determined using Image J software (NIH). The software was programmed so that individual pixels appear white if F-actin (green) stain and NFATc3 (red) stain are colocalized, and the number of white pixels was determined as reported previously (29). F-actin and G-actin (DNase I) integrated intensity (arbitrary units, AU) were measured in the threshold individual channels using Metamorph and used to calculate F/G actin content. The same threshold was applied to every image.

Statistics.

Results are expressed as means ± SE. Statistical significance was tested at the 95% (P < 0.05) confidence level using one- or two-way repeated measures ANOVA. NFATc3-GFP nuclear import curves were obtained by nonlinear regression fit to a one-phase association equation.

RESULTS

CH-induced NFATc3 activation requires ET-1 and ROK.

Pre-pro-ET-1 mRNA expression was significantly increased in pulmonary arteries from mice exposed to CH from 2 to 21 days (Fig. 1A). In addition, pre-pro-ET-1 mRNA levels were similarly increased in NFATc3 WT and KO mice at 2 days of CH (Fig. 1B) demonstrating that CH-induced increases in pre-pro-ET-1 mRNA are independent of NFATc3.

Fig. 1.

Chronic hypoxia (CH) increases intrapulmonary arterial pre-pro-endothelin-1 (ET)-1 mRNA independently of nuclear factor of activated T cells isoform c3 (NFATc3). Pre-pro-ET-1 transcripts were measured in isolated intrapulmonary arteries from mice exposed to normoxia, CH (2, 7, or 21 days) (A), and from NFATc3 wild-type (WT) and knockout (KO) mice exposed to normoxia and 2 days of CH (B). The results were expressed as fold change from normoxia. Data are means ± SE; n = 6 animals/group. *P < 0.05 vs. normoxia.

To determine whether the increased expression of ET-1 leads to increased intrapulmonary arterial NFATc3 transcriptional activity in CH (20), the dual ET-1 receptor blocker bosentan or the selective ET_AR blocker PD155080 was administered to NFAT-luc mice while exposed to CH for 2 days. Both bosentan and PD155080 significantly attenuated CH-induced NFATc3 transcriptional activation in isolated intrapulmonary arteries, shown as luciferase reporter activity, with PD155080 being significantly more effective (Fig. 2A).

Fig. 2.

Mechanisms of NFAT activation by CH. Bosentan or PD155080 (A), diltiazem (B), and cyclosporin A (C) were administered to mice exposed to CH for 2 days, and luciferase activity normalized to total protein was measured in isolated intrapulmonary arteries. The results were expressed as fold change from normoxia (dotted line). Data are means ± SE; n = 6 animals/group. *P < 0.05 vs. normoxia; #P < 0.05 vs. CH vehicle; †P < 0.05 vs. CH bosentan.

Increased [Ca²⁺]_i is required for NFATc3 dephosphorylation and activation by calcineurin (34–36, 41). Accordingly, we found that administration of the L-type Ca²⁺ channel inhibitor diltiazem significantly attenuated CH-induced NFATc3 activation (Fig. 2B).

As expected, calcineurin inhibition with CsA prevented NFATc3 activation by CH (Fig. 2C). Administration of Cremophor EL, the vehicle for CsA, significantly reduced the effect of CH on NFATc3 activation compared with the vehicle for diltiazem and bosentan (saline in osmotic pump and drinking water, respectively) (Fig. 2, A–C). Regardless of the effect of Cremophor EL, CsA completely inhibited the effect of CH on NFATc3 activation.

The RhoA/ROK pathway has been shown to be upregulated in CH (46, 79). Since we hypothesized that this pathway is implicated in ET-1-induced NFATc3 nuclear translocation, RhoA activity was compared between normoxia and CH exposed mice. Figure 3A shows that RhoA activity was increased in lungs from CH-exposed mice. To determine whether RhoA activation is downstream of ET_AR stimulation, mice were treated with bosentan or PD155080 during CH exposure. Indeed, ET_AR activation was required for CH-induced increases in RhoA activity as shown in Fig. 3A. One limitation of our study is that we were not able to measure RhoA activity in isolated pulmonary arteries due to the limited sensitivity of the assay, which is consistent with the lack of reports measuring RhoA activity in isolated mouse small arteries.

Fig. 3.

CH-induced NFATc3 activation in a RhoA/ROK-dependent manner. A: RhoA activity was measured in whole lung lysate from mice exposed to normoxia or 2 days of CH and treated with bosentan or PD155080; B: fasudil and HA 1152 were administered to mice exposed to CH for 2 days, and luciferase activity normalized by total protein was measured in isolated intrapulmonary arteries. The results were expressed as fold change from normoxia (dotted line). Data are means ± SE; n = 6 animals/group. *P < 0.06 vs. normoxia; #P < 0.05 vs. CH vehicle; †P < 0.05 vs. CH fasudil.

Also consistent with our hypothesis, we found that administration of the ROK inhibitor fasudil [IC₅₀ 1.9 × 10⁻⁶ M (66)], significantly attenuated CH-induced NFATc3 activation in intrapulmonary arteries from NFAT-luc mice (Fig. 3B). In addition, a more potent ROK inhibitor HA 1152 [IC50 7.9 × 10⁻⁸ M (66)] prevented NFATc3 activation by CH (Fig. 3B). Together, these results suggest that CH increases ET-1 levels leading to activation of NFATc3 in intrapulmonary arteries via increased [Ca²⁺]_i, calcineurin, and RhoA/ROK activity.

ET-1-induced NFATc3 nuclear import in HPASMC requires ROK and actin polymerization.

To determine the mechanism by which ROK mediates ET-1-induced NFATc3 activation, we performed studies in cultured HPASMC.

Figure 4 shows that 5 min treatment with ET-1 significantly increased the F/G actin ratio in HPASMC. This increase was prevented by the ROK inhibitor fasudil. As expected, cytochalasin B, which blocks addition of actin monomer (G-actin) to filamentous actin (F-actin), significantly reduced the F/G actin ratio and prevented ET-1-induced increases in F-actin content (Fig. 4).

Fig. 4.

ET-1 increases F-actin through ROK. Human pulmonary arterial smooth muscle cells (HPASMC) were treated with ET-1 ± fasudil or cytochalasin B, and F/G actin content was determined by inmunofluorescence confocal microscopy. The results were expressed as F/G actin fluorescence intensity (AU). Data are means ± SE; n = 7 cells/group. *P < 0.05 vs. control, #P < 0.05 vs. ET.

GFP nuclear fluorescence accumulation was tracked in NFATc3-GFP-expressing HPASMC after stimulation with ET-1 in the presence or absence of the ROK inhibitor fasudil. Figure 5A shows that fasudil significantly attenuated ET-1-induced increases in NFATc3-GFP nuclear import. In addition, cytochalasin B, which decreases F-actin content, inhibited ET-1-induced NFATc3 nuclear import (Fig. 5B). Interestingly, jasplakinolide, an actin stabilizing agent (14), also inhibited the effect of ET-1 on NFATc3 nuclear import (Fig. 5C). Fasudil, cytochalasin B, and jasplakinolide had no effect when applied without ET-1 (data not shown).

Fig. 5.

ET-1-induced NFATc3 nuclear import is inhibited by actin cytoskeleton depolymerization and stabilization. HPASMC were treated with ET-1 ± fasudil (A), cytochalasin B (B), or ± jasplakinolide (C). NFATc3-green fluorescent protein nuclear fluorescence was monitored over time. After background subtraction, nuclear fluorescence (F) was divided by basal fluorescence at time 0 (F₀). Data are means ± SE; n = 7 cells/group. *P < 0.05 vs. control, #P < 0.05 vs. ET.

To demonstrate that ROK inhibition and derangement of actin cytoskeleton do not reduce ET-1-induced increases in [Ca²⁺]_i, the effect of each inhibitor was measured on HPASMC loaded with the Ca²⁺ indicator fura-2 AM loaded. As expected, ET-1 significantly increased [Ca²⁺]_i, showing an initial peak followed by a plateau (Fig. 6A). Neither fasudil, cytochalasin B, leptomycin B, nor jasplakinolide significantly affected ET-1-induced increases in [Ca²⁺]_i (Fig. 6, B and C).

Fig. 6.

ET-1-induced increases in intracellular Ca²⁺ concentration ([Ca²⁺]_i) are not affected by ROK inhibition or actin cytoskeleton depolymerization. Fura-2 AM-loaded HPASMC were treated with ET-1 ± leptomycin B, fasudil, or cytochalasin B. Ratio of 340/380 fluorescence was background subtracted and expressed as percent change from baseline. Representative trace (A), average peak (B), and plateau of ET-1-induced [Ca²⁺]_i increase (C) are shown. Data are means ± SE; n = 7 cells/group. No significant difference was observed between groups.

To determine how ET-1-induced F-actin polymerization increases NFATc3 nuclear import, immunofluorescence colocalization of F-actin and NFATc3 was performed in HPASMC treated with ET-1 in the presence or absence of fasudil and cytochalasin B. Figure 7, A and B, shows that ET-1 increased the colocalization of F-actin and NFATc3, which was prevented by fasudil and cytochalasin B. Neither fasudil nor cytochalasin B had any significant effect when applied without ET-1.

Fig. 7.

ET-1 increases F-actin and NFATc3 colocalization. HPASMC were treated with ET-1 ± fasudil or cytochalasin B, fixed and F-actin and NFATc3 were fluorescently labeled. A: representative confocal images of F-actin (green) and NFATc3 (red). White pixels indicate F-actin and NFATc3 colocalization. B: summary of results showing percent F-actin and NFATc3 colocalization. Data are means ± SE; n = 15 cells/group. *P < 0.05 vs. control, #P < 0.05 vs. ET.

ET-1-mediated NFATc3 transcriptional activation in isolated mouse pulmonary arteries requires ROK and actin polymerization.

To test whether ET-1 activates NFATc3 ex vivo in the absence of hypoxia or transmural pressure, isolated intrapulmonary arteries from NFAT-luc mice were incubated with ET-1. Figure 8 shows that ET-1 induced a fourfold increase in luciferase activity; i.e., NFATc3 transcriptional activity, over basal. Pretreatment with the ROK inhibitor fasudil and the actin polymerization inhibitor cytochalasin B prevented ET-1-induced NFATc3 activation (Fig. 8). Neither fasudil nor cytochalasin B had any significant effect when applied without ET-1.

Fig. 8.

ET-1-induced increases in NFAT-dependent transcription in pulmonary arteries depend on ROK activation and actin cytoskeleton polymerization. Intrapulmonary arteries were isolated from NFAT-luciferase mice and treated with ET-1 ± fasudil or cytochalasin B. Luciferase activity was measured, normalized to total protein, and expressed as fold change from control (dotted line). Data are means ± SE; n = 6 arteries from 6 mice/group. *P < 0.05 vs. control, #P < 0.05 vs. ET.

DISCUSSION

In this study, we hypothesized that CH activates NFATc3 in PASMC through ET-1-mediated Ca²⁺ mobilization and stimulation of RhoA/ROK, leading to calcineurin activation and F-actin polymerization. In agreement with this hypothesis, we found that: 1) CH increases pulmonary arterial pre-pro-ET-1 mRNA expression and lung RhoA activity; 2) inhibition of ET_AR, calcineurin, L-type Ca²⁺ channels, and ROK blunts CH-induced increase in NFATc3 transcriptional activity; and 3) ET-1-induced NFATc3 nuclear translocation in HPASMC and isolated mouse pulmonary arteries depends on ROK-mediated F-actin polymerization.

CH-induced NFATc3 activation requires ET-1 and ROK.

Our present results show that CH increases ET-1 expression in mouse pulmonary arteries are consistent with the reported role for ET-1 as an important mediator of CH-induced pulmonary hypertension in both animals and humans (8, 27, 32, 81, 83). In addition, we have demonstrated that CH-induced upregulation of ET-1 is independent of NFATc3. It is not likely that other NFATc isoforms contribute to CH-induced ET-1 expression because we have previously demonstrated that NFATc3 is the only isoform activated by CH in mouse pulmonary arteries (20). This previous finding also supports our current focus on the study on NFATc3.

This is the first study to demonstrate that in vivo inhibition of ET_AR attenuates CH-induced NFATc3 activation in intrapulmonary arteries. These findings are in agreement with previous reports showing that ET-1 is a potent activator of NFAT (17, 48, 61).

ET-1 causes release of Ca²⁺ from intracellular stores, protein kinase C-dependent inhibition of voltage-gated K⁺ (K_v) channels, and subsequent membrane depolarization and Ca²⁺ influx through L-type Ca²⁺ channels (69, 70). In addition, other Ca²⁺ entry pathways have been shown to contribute to global changes in Ca²⁺ in pulmonary hypertension (1). Consistently, administration of the L-type Ca²⁺ channel inhibitor diltiazem significantly attenuated but did not abolish CH-induced NFAT activation. Although, it is well established that increased [Ca²⁺]_i via influx through L-type Ca²⁺ channels and Ca²⁺ release from intracellular stores are required for calcineurin activation followed by NFAT dephosphorylation (34–36, 41), it has not been previously demonstrated in vivo. This study demonstrates a role for Ca²⁺ in CH-induced NFATc3 activation in vivo. In addition, our results are consistent with the demonstrated central role of calcineurin in NFAT regulation since CsA abrogated CH-induced NFATc3 activation.

It is well described that ET-1 binding to ET_AR activates RhoA/ROK pathway (7, 33, 55). Therefore, we explored the role of RhoA/ROK signaling as an additional pathway that could participate in CH-induced NFATc3 activation. Our results showing elevated RhoA levels in lungs from CH mice are in accordance with previous reports demonstrating that RhoA/ROK signaling is upregulated by hypoxia (13, 45, 46) and the implication of this pathway in the development of pulmonary hypertension (2, 3, 12, 25, 31, 56, 57, 79).

Our results also suggest that ET_AR are activated in CH due to the elevated ET-1 levels leading to increased RhoA activity because both a dual ET receptor and a selective ET_AR antagonist prevent CH-induced increases in RhoA activity. Subsequent elevations in PASMC [Ca²⁺]_i and RhoA lead to NFATc3 nuclear translocation and activation. This conclusion is supported by our data showing that ET receptor, L-type Ca²⁺ channels, calcineurin, and ROK inhibitors attenuate and/or prevent CH-induced increases in NFATc3 transcriptional activity. However, additional factors besides ET-1 might be contributing to CH-induced NFATc3 activation because the ET_AR receptor antagonist completely prevented CH-induced RhoA activation but did not completely prevent NFATc3 activation; it caused ∼75% attenuation. Therefore, our study does not exclude the contribution of other factors such as serotonin, angiotensin II, or reactive oxygen species (ROS), which have been shown to be elevated by CH (4, 15, 23, 24, 28, 37, 46, 68) and are known activators of NFAT (5, 30, 37, 47, 49, 60, 73, 76, 78). Of interest, several studies have shown cross talk between ET-1 and angiotensin II. In particular, administration of an angiotensin converting enzyme inhibitor reduces elevated plasma concentrations of ET-1 (54). In addition, angiotensin II increases ET-1 synthesis via elevated ROS in human arteries in culture, and there is a significant association between elevated angiotensin II levels, increased oxidative stress, and increased ET-1 concentrations (50). Recently, it has been shown that 20-wk-old mice with SM22α-targeted overexpression of the serotonin transporter are pulmonary hypertensive (37). NFATc2 is activated in PASMC from these mice. However, calcineurin/NFAT inhibition with CsA administration does not reverse the pulmonary hypertension (37). Thus the potential role of angiotensin II, serotonin, and ROS in CH-induced NFATc3 activation remains to be explored.

CH causes pulmonary vasoconstriction leading to elevated pulmonary pressure. It is possible to argue that in vivo administration of inhibitors of ET receptors, L-type Ca²⁺ channels, and calcineurin prevents CH-induced vasoconstriction and increases in pulmonary pressure since the effectiveness of these drugs in lowering pulmonary pressure in rodents and humans has been previously demonstrated, and the site of action is primarily the vasculature (10, 17, 18, 21, 22, 44, 51, 75, 82). Elevated intrapulmonary pressure could be a mechanical stimulus that contributes to increased PASMC [Ca²⁺]_i leading to NFATc3 activation (36, 58). However, ex vivo application of ET-1 to isolated intrapulmonary arteries in the absence of intravascular pressure also increased NFATc3 activity through ROK activation. Therefore, upregulation of ET-1 (and possibly other vasoactive factors), rather than elevated pulmonary arterial pressure, is most likely the stimulus that leads to NFATc3 activation in response to CH.

ET-1-induced RhoA/ROK-mediated actin polymerization leads to NFATc3 nuclear translocation.

RhoA/ROK signaling has been previously implicated in NFAT activation by our laboratory and others (17, 80). The mechanism has not been explored; however, it has been shown that RhoG, a RhoA-related small GTPase, potentiates NFAT-dependent transcription correlating with its capacity to increase actin polymerization (77). This finding supports the possibility that NFAT-dependent transcription is an actin-dependent process. Along these lines, ET-1 via RhoA/ROK increases F-actin content (i.e., induces cytoskeleton actin polymerization) in airway smooth muscle cells and preadipocytes (42, 72). Consistent with these reported findings, we found that in HPASMC, ET-1 increased F-actin content and NFATc3 nuclear import, both of which were ROK dependent.

The increase in the proportion of F-actin that occurs in response to the stimulation of smooth muscle cells and the essential role of stimulus-induced actin polymerization and cytoskeletal dynamics in the generation of mechanical tension has been convincingly documented in vascular smooth muscle (reviewed in Refs. 39, 72, 74). In addition, actin cytoskeleton polymerization has been implicated in the transport of transcription factors to the nucleus (26). Actin cytoskeleton dynamics are regulated by many signaling pathways. In general, receptor activation and/or integrin ligation activates protein kinases and/or small GTPases, which in turn regulate the functional status of the actin regulatory proteins and eventually cause actin filament assembly or structural reorganization (reviewed in Ref. 74). In particular, it has been reported that nuclear import of nuclear factor κB (NF-κB; which, along with NFAT, is a member of the Rel family of transcription factors) requires activation of RhoA/ROK (26). Stimulation of RhoA/ROK enhances LIM kinase activity, leading to phosphorylation and inactivation of cofilin (actin regulatory protein) (6, 26). When cofilin is phosphorylated, it no longer binds to actin, thus facilitating actin polymerization (6) and NF-κB transport to the nucleus (26). A recent study demonstrates that ET-1, in a ROK-dependent manner, increases phospho-cofilin-to-cofilin ratio in pulmonary artery rings and cultured PASMC (16). Hypoxia also has been shown to increase phophocofilin levels in PASMC (59). Our present study shows that both destabilization and stabilization of the actin cytoskeleton inhibit NFATc3 nuclear transport. This finding suggests that NFATc3 nuclear transport depends on dynamic reorganization of the actin cytoskeleton and is similar to the mechanism of NF-κB nuclear transport (26). Contrary to our findings in HPASMC, disruption of actin polymerization in T cells results in prolonged [Ca²⁺]_i elevation and persistent NFAT nuclear accumulation due to decreased Ca²⁺ export (67).

In conclusion, this study demonstrates that CH increases ET-1 expression leading to ET_AR activation, elevated PASMC [Ca²⁺]_i, and stimulation of RhoA/ROK activity. As previously demonstrated, elevated [Ca²⁺]_i activates calcineurin, which dephosphorylates NFATc3, exposing nuclear localization signals (reviewed in Refs. 41 and 65). In addition, ROK increases actin polymerization providing the structural support for NFATc3 nuclear transport.

Considering that NFATc3 is required for CH-induced pulmonary hypertension in mice (20), future studies should address whether the signaling pathway demonstrated in this study for NFATc3 activation applies to other animal models of pulmonary hypertension, primary pulmonary hypertension, and/or to other NFATc isoforms. Arguing against a role for the ET-1/ETR/RhoA/ROK pathway in NFATc2 activation, however, is evidence that NFATc2, but not NFATc3, is activated in PASMC from primary pulmonary hypertensive patients (9) treated with ET receptor antagonists. Nevertheless, much evidence suggests that RhoA/ROK signaling plays an important role in the pathogenesis of various experimental models of pulmonary hypertension, including those resulting from CH, monocrotaline, bleomycin, shunt, and vascular endothelial growth factor receptor inhibition combined with CH exposure (2, 3, 38, 57, 62, 85), as well as in patients with pulmonary arterial hypertension (31). The role of NFATc3 activation in the pathogenesis of pulmonary hypertension in patients with chronic bronchitis, emphysema, cystic fibrosis, and severe restrictive lung diseases or in high-altitude residents should be explored. Therefore, our study highlights the importance of exploring whether NFATc3 pathway inhibition could have potential clinical relevance to prevent and/or reverse the development of pulmonary hypertension and its associated vascular pathology.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grant R01-HL-088151 and supplement from American Recovery And Reinvestment Act of 2009.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).