| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Rudolf-Buchheim-Institute of Pharmacology, Justus-Liebig-University Giessen, Germany;
Department of Pharmacology and Centre for Vascular Health, Monash University, Melbourne, Australia; and
University of Giessen Lung Center, Medical Clinic II/V, Giessen, Germany
1Correspondence: Department of Pharmacology & Centre for Vascular Health, Monash University, Wellington Rd., Bldg. 13e, Rm. 116, Melbourne, Victoria 3800, Australia. E-mail: harald.schmidt{at}med.monash.edu.au
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: vascular disease • PH • lung perfusate
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1/β1) and a subsequent increase in intracellular cyclic GMP (cGMP) levels and plasma spillover (1
, 2)
. cGMP leads to an activation of cGMP-dependent protein kinase (cGK I), modulation of its target proteins, and inhibition of contraction (3)
. Moreover, an impairment of the vasodilatory NO-cGMP signaling pathway was suggested to underlie the pathogenesis of PH (4
, 5)
. Consequently the major therapeutic breakthrough of cGMP-specific phosphodiesterase 5 (PDE 5) inhibitors such as sildenafil to reduce PH in both clinically and appropriate animal models such as hypoxic PH (HPH) was attributed to a mechanism-based augmentation of a pathophysiologically attenuated NO-cGMP pathway (6
7
8
9
10)
. This concept received further indirect support from the recently observed effectiveness of NO-independent sGC activators, BAY 56–2667 and BAY 41–2272, in PH (11
12)
. In this study we wanted to elucidate in a mouse model of early and established HPH whether these beneficial effects of sildenafil are indeed mechanism based with respect to attenuated NO-cGMP signaling in hypoxia or a merely symptomatic cGMP-mediated vasodilatation unrelated to the pathomechanism of HPH.
Single components of NO-cGMP signaling have been reported to be altered; however, results vary and a comprehensive study is missing. For example, NO synthase (NOS) activity was found to be unchanged (13)
, decreased (14)
, and most frequently up-regulated (15
, 16)
, including NOS II and III protein and mRNA in rat (13
, 15
16
17
18
19
20)
or mouse lungs (21)
after 2 days to 3 wk of hypoxia vs. normoxia. In the isolated porcine artery, increased NOS III expression was already found after 2 h of hypoxic incubation (22)
. Furthermore, expression and activity of sGC 1/β1, the main receptor of NO, was shown to be decreased (23)
, but also unchanged (17
, 24)
and even increased (21
, 25)
in crude lung homogenate of mice and rats exposed to hypoxia vs. normoxia for 1 to 4 wk.
However, even an increased production of NO in HPH may not necessarily result in increased NO bioavailability and NO signaling. Reactive oxygen species (ROS), such as superoxide, which are generated during hypoxia, can scavenge NO to prevent sGC-cGMP signaling, and form the powerful oxidant peroxynitrite (ONOO–), leading to lipid peroxidation and nitration of cellular proteins (26)
. Increased nitrotyrosine staining has been found in rat pulmonary vasculature after 2 days (20)
, and in isolated porcine pulmonary artery after only 2 h of hypoxia (22)
.
Even under conditions of elevated cGMP production, counter-regulated expression and activity of PDEs, mainly PDE 5, could still lead to decreased NO-cGMP signaling via elevated cGMP breakdown. In support of this, expression levels of PDE 1 and PDE 5 are high in the normal lung of rats (27
, 28)
, and further increased PDE mRNA, protein (8
, 29)
, and activity (30)
have been observed in pulmonary arteries of rats exposed to 2 wk of hypoxia.
Direct measurements of exhalted NO as well as tissue and perfusate levels of NO (31
32
33
34
35)
and cGMP (10
, 13
, 36
37
38
39)
in hypoxic lung and cell cultures yielded contradictory results depending on the animal model used and on the severity and duration of hypoxic ventilation. However, tissue and perfusate cGMP levels may not reliably depict NO-cGMP signaling if the cGMP target protein, cGMP-dependent protein kinase I (cGK I), was affected. This kinase plays a key role in mediating NO-cGMP induced vascular relaxation (40)
. Unfortunately, there is a paucity of information pertaining to changes in cGK I, an increase (41)
has been reported, while pulmonary vasodilation to cGMP analogs is attenuated (24)
, after chronic hypoxia.
The phosphorylation state of a substrate protein of cGK I, the vasodilator-stimulated phosphoprotein (VASP), was established as an important biomarker for the overall effectiveness of NO-cGMP signaling (42
43
44)
, and phospho-VASP levels have not been determined in the setting of HPH.
In the present study we aimed to comprehensively quantify how key pulmonary enzymes of the NO-cGMP signaling cascade are affected in a mouse model of early and established HPH and to investigate the feasibility of phospho-VASP to reflect the overall effectiveness of pulmonary NO-cGMP signaling. Finally, these data were correlated with the effects of acute sildenafil treatment on pulmonary vascular resistance and perfusate cGMP levels in isolated perfused and ventilated lungs.
Our results show that PDE 5 inhibition does not target a pathomechanism per se but represents a symptomatic, yet effective, approach in HPH. These data further elucidate the rationale behind sildenafil or sGC activators as "mimetics" of NO-cGMP signaling.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
24°C) and lighting (12/12 h light/dark cycle) with free access to food and water. Mice were placed in a normobaric chamber and exposed to hypoxia (10% O2) or normoxia for 5 or 21 days as described (45)
. All animal experiments were performed with permission from the local authorities in Regierungspräsidium, Giessen, Germany.
Isolated perfused lungs
Mouse lungs were isolated, perfused, and ventilated as described before (46)
. Briefly, 5 or 21 days after exposure to normobaric normoxia or hypoxia, mice were anesthetized with ketamine/xylazine (i.p.) and anticoagulated with heparin (1000 U/kg) by i.v. injection. Animals were placed in a water-jacketed chamber (Type 839, Hugo Sachs Elektronik, March-Hugstetten, Germany) for temperature control. After intubation via a tracheostomy, mice were ventilated with room air (positive pressure ventilation) with a 250 µl tidal volume, 90 breaths/min, and 2 cmH2O positive end-expiratory pressure using a Minivent Type 845 (Hugo Sachs Elektronik). Midsternal thoracotomy was followed by insertion of catheters into the pulmonary artery and left atrium. Using a peristaltic pump (ISM834A V2.10, Ismatec, Glattbrugg, Switzerland), buffer perfusion via the pulmonary artery was started at 4°C and a flow of 0.2 ml/min. In parallel with the onset of artificial perfusion, ventilation was changed from room air to a premixed gas (21.0% O2, 5.3% CO2, balanced with N2). For perfusion, Krebs-Henseleit buffer (Serag-Wiessner, Naila, Germany) containing 120 mmol/L NaCl, 4.3 mmol/L KCl, 1.1 mmol/L KH2PO4, 2.4 mmol/L CaCl2, 1.3 mmol/L MgCl2, and 13.3 mmol/L glucose as well as 5% (w/v) hydroxyethylamylopectin (mol wt 200,000) was used. The pH was adjusted to 7.37–7.40 with NaHCO3. After rinsing the lungs with 20 ml buffer, the perfusion circuit was closed for recirculation (total system volume: 13 ml, time set at zero) and left atrial pressure was set at 2.0 mm Hg. Meanwhile, the flow was slowly increased from 0.2 to 2 ml/min and the entire system was heated to 37°C. Pressures in the pulmonary artery were registered via small-diameter catheters. Pulmonary artery pressure (PAP) directly reflects pulmonary vascular resistance in this setup as the lungs were perfused at constant flow. In hypoxic and normoxic animals, sildenafil (100 nM) was added to the perfusate at time zero. The cGMP levels of perfusate samples were measured after 5, 30, 60 and 90 min with a radioimmunoassay.
Tissue homogenization
After exposure to hypoxia or normoxia for 5 or 21 days and, in the case of the perfused lung, after perfusion for 90 min, the mice were sacrificed. Lungs were removed and rinsed with saline. For Western blot analysis, tissue powder was homogenized in 5 volumes of Roti-Load (Roth, Karlsruhe, Germany) and heated to 95°C for 15 min. After centrifugation (2000 g), the supernatant was retained for Western blot. For qualitative Western blots, additional tissues (rat and bovine lung) were prepared as described for mouse lung. For activity assays of cGMP-specific PDE and sGC as well as cGMP tissue and perfusate level measurement, tissue powder was homogenized in a glass-glass homogenizer at 4°C in 5 volumes of homogenization buffer [25 mM TEA (pH 7.5), 1 mM ethylenediaminetetraacetic acid (EDTA) (pH 7.4), 5 mM dithiotreitol (DDT), 50 mM NaCI, 10% glycerin, proteinase inhibitor [Roche, Nutley, NJ, USA]. After centrifugation (2000 g), the supernatant was used for the activity assays and cGMP measurement. Protein concentrations were determined according to Lowry et al. (47)
.
Western blot analysis
Proteins (10 µg/lane for determination of sGC 1/β1, NOS I, NOS II, NOS III, cGK I; 50 µg/lane for determination of nitrotyrosine, VASP, P-VASP, PDE 5) were separated via SDS-PAGE (7.5%) and transferred to a nitrocellulose membrane (Bio-Rad, Hercules, CA, USA). Subsequently, proteins were blocked for 1 h at room temperature with 3% or 5% fat-free milk in TBST (20 mM Tris-HCI (pH 7.5), 150 mM NaCI, 0.1% (g/v) Tween-20) to determine sGC 1/β1, NOS I, NOS II, NOS III, cGK I, PDE 5, P-VASP, and nitrotyrosine; for determination of VASP, 1% bovine hemoglobin in PBSTT (140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4,0.05% Tween-20, 0.3% Triton X-100) was used as blocking reagent. Nitrocellulose membranes were then incubated overnight at 4°C with antibodies diluted in a blocking reagent and directed against NOS I (1:1000), NOS II (1:2000), NOS III (1:3000), sGC 1 (1:5000), sGC β1(1:3000); in the case of commercial antiserum (Cayman Labs, Ann Arbor, MI, USA; Cat. #160894), against sGC 1/β1 (1:1000), cGK I (1:2000), P-VASPSer-239 (1:1000), VASP (1:2500); nitrotyrosine (1:1000, Alexis Corporation, Lausen, Switzerland; 1:1000, Calbiochem, Darmstadt, Germany; 1:4000, monoclonal, J. S. Beckman; 1:5000, L. O. Uttenthal, Madrid, Spain), followed by a goat anti-mouse or anti-rabbit HRP-labeled secondary antibody (1:2000). Protein signals were detected using enhanced chemiluminescence detection reagent (ECL) and quantified with a densitometer using Kodak imaging software.
sGC activity assay
Basal or NO-stimulated sGC activity in lung homogenates was measured as the formation of cGMP at 37°C during 10 min in a total incubation volume of 100 µl containing 50 mM triethanolamine (TEA) -HCL (pH 7.4), 3 mM MgCl2, 3 mM reduced glutathione (GSH), 1 mM 3-isobutyl-1-methylxanthine (IBMX), 100 µM zaprinast, 5 mM creatine phosphate, 0.25 mg/ml creatine kinase, and 500 µM GTP. The reaction was started by simultaneous addition of 10 µl of the crude lung homogenate (1 µg protein/µl) together with either 10 µl of NaOH (5 mM) or 10 µl of the NO donor, DEA-NO (300 µM). The reaction was stopped by boiling for 10 min at 95°C. In preliminary experiments, linear cGMP formation during the incubation period was verified (data not shown). The amount of cGMP was subsequently determined by a commercial enzyme immunoassay.
PDE activity assay
The overall and sildenafil (100 nM) -inhibited activity of cGMP-specific phosphodiesterases (PDE 5) in lung homogenates were measured with minor modifications, as described previously (48)
. cGMP breakdown at 37°C was determined over 10 min in a total incubation volume of 100 µl containing 10 µl crude lung homogenate (1 µg protein/µl) and 90 µl of reaction buffer (20 mM Tris, 20 mM imidazol/HCI (pH 7.5), 3 mM MgCl2, 15 mM magnesium acetate, 250 nM calmodulin, 250 µM Ca2Cl, 5 µM cGMP). The reaction was started by addition of 10 µl water or a 100 nM sildenafil solution and 10 µl crude lung homogenate to the reaction buffer, then stopped by heating to 95°C for 10 min. In preliminary experiments, linear cGMP breakdown during the incubation period was verified (data not shown). The amount of remaining cGMP was determined using a commercial enzyme immunoassay.
Reagents
Antibodies
NOS I, NOS II, and NOS III (Transduction Laboratories, Lexington, KY, USA); self-made sGC 1 (residues 634–647) and sGC β1 (residues 593–614) antibodies affinity-purified against synthetic peptide sequences corresponding to human sGC as described previously (49)
; sGC antiserum (Cat. #160894, Cayman Chemicals, USA); cGK I (kindly donated by S. Lohmann, Germany (50)
; VASP M4 (Immunoglobe, Himmelstadt, Germany); P-VASP (monoclonal antibody 16 C2 directed against the phosphopeptide sequence RKVpS(239)KQE, which represents the VASP phosphorylation site at serine 239 (51)
; nitrotyrosine [Alexis Corporation, Calbiochem; kindly donated by L. O. Uttenthal (polyclonal; ref. 52
) and J. S. Beckman (monoclonal; ref. 53
)]; PDE 5 (Calbiochem); actin (Oncogene Research Products, Darmstadt, Germany); and polyclonal anti-rabbit antibodies conjugated to horseradish peroxidase (Dako, Hamburg, Germany).
Materials
ECL immunodetection kit, Hybond ECL nitrocellulose membrane and low and high molecular weight-SDS calibration kit, and glutathione Sepharose 4B were from Amersham Pharmacia Biotech (Freiburg, Germany); enzyme immunoassay was from Assay Designs Inc. (Ann Arbor, MI, USA); glycerin, KH2PO4, magnesium acetate, MgCl2, Na2HPO4, and Triton X-100 were from Merck (Darmstadt, Germany); NaCI, Roti-Load sample buffer, Rotiphorese (30% acrylamid, 0.8% bisacrylamid in H2O), SDS Tris were from Roth GmbH (Karlsruhe, Germany); IBMX, zaprinast, and DEA-NO were from Alexis; and DDT, APS, and Ponceau S were from Serva Feinbiochemica, Heidelberg, Germany). All other chemicals were of the highest purity grade available and were obtained from Sigma (Deisenhofen, Germany). Water was de-ionized to 18 M
/cm (Milli-Q; Millipore, Eschborn, Germany). Purified recombinant human sGC was obtained as described earlier (45)
, with minor modifications. Briefly, crude supernatant and crude particulate fractions of Sf9 cells expressing a GST-sGC1/sGCβ1, protein were separated by centrifugation (20.000 g) for 15 min at 4°C. The crude supernatant fraction was incubated with glutathione Sepharose 4B for 1 h at 25°C, then washed three times with lysis buffer containing 25 mM TEA (pH 7.8), 75 mM NaCI, 1 mM EDTA, 5 mM DDT, 1 pM leupeptin, and 0.5 µg ml soybean trypsin inhibitor. For proteolytic cleavage of the GST fusion tag, glutathione Sepharose 4B was incubated with thrombin (up to 80 U/ml).
Statistics
Data are presented as means ± SE. Statistical differences between the means were analyzed using the Student’s unpaired t test. For multiple comparisons, 1-way analysis of variance (ANOVA) was used with a Bonferroni post hoc test. Both analyses were carried out using MS EXCEL Software (Windows USA, 2001) and PRISM Graph Pad (3.0, Graph Pad Software, San Diego, CA, USA). Probabilities of <5% (P<0.05) were considered to be statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A). Protein expression of NOS II and NOS III were significantly up-regulated in mice exposed to 5 days (222.2±33.8% of norm, P<0.05; 192.0±15.9% of norm, respectively, P<0.01) or 21 days (180.0±6.6% of norm, P<0.05; 179.5±10.2% of norm, respectively, P<0.01) hypoxia vs. normoxia (Fig. 1D, E
). Despite an increase in NOS expression after hypoxic exposure, NO bioavailability may be reduced due to increased formation of ROS and scavenging of NO to form ONOO–, which may result in increased nitration of cellular proteins.
|
Nitration of proteins in lungs of mice exposed to hypoxia is unaltered or even attenuated in lungs of mice exposed to 5 or 21 days of hypoxia
To determine tyrosine nitration of cellular proteins, the immunoreactive signals of four commercial nitrotyrosine antibodies were compared. Whereas nitrated bovine serum albumin was detected by all of the antibodies, albeit with different affinities, only one nitrotyrosine antibody (52)
showed two immunoreactive signals at 70 kDa and 38 kDa; all other antibodies tested gave only weak signals in the lungs of normoxic mice (Fig. 1C
). The immunoreactive signal at 40 kDa was not altered by exposure to 5 or 21 days of hypoxia vs. normoxia (99.9±8.8% of norm; 91.3±14.7% of norm), but the immunoreactive signal at 70 kDa was slightly but significantly decreased after 5 and 21 days (76.6±6.9% of norm; 87.2±4.2, P<0.05) (Fig. 1D, E
). Finding no evidence of impairment at the level of NOS expression or NO bioavailability, we next sought to determine whether dysfunction occurred at the level of the receptor for NO, sGC 1/β1.
Expression and activity of sGC 1/β1 in lungs of mice exposed to hypoxia is unaltered
We initially compared the specificity of our subunit specific sGC 1/β1 antibodies (49)
with a commercially available sGC antibody (21
, 25)
against purified human sGC 1/β1 in rat, mouse, and bovine lung (Fig. 1B
). The commercial antibodies and ours both produced an immunoreactive signal for the β1 subunit at 70 kDa in all preparations tested. In contrast, the commercial antiserum yielded to an immunoreactive signal for the subunit at 90 kDa in all tissues tested, with a lower SDS mobility as the signal detected by our subunit specific sGC 1 antibody at 80 kDa. These findings with the commercial antibody were apparent even in bovine lung, where a smaller 1 subunit than in other tissues has been reported (54)
. The commercial antibody determined only one immunoreactive signal in human sGC standard (Fig. 1B
). Using our subunit specific sGC 1/β1 antibodies (49)
, protein expression of sGC 1/β1 was unchanged in lungs of mice exposed to 5 days or 21 days (5 days: 1, 108.3±10.1% of norm; β1, 127.2±13.3% of norm; 21 days: 1, 115.3±7.8% of norm; β1, 108.3±10.1% of norm) of hypoxia vs. normoxia (Fig. 1D, E
). Basal sGC 1/β1 activity of crude lung homogenate taken from mice exposed to 5 or 21 days of normoxia (90.2±14.5 pmol/mg proteinxmin; 93.7±4.3 pmol/mg proteinxmin) was not significantly different from that measured in hypoxic lungs (88.6±1.9 pmol/mg proteinxmin; 98.4±12.1 pmol/mg proteinxmin). DEA-NO (300 µM) increased sGC activity by 40-fold in lungs taken from normoxic (5 days: 3735.7±316.3 pmol/mg proteinxmin; 21 days: 4692.6±172.5 pmol/mg proteinxmin) as well as hypoxic animals (5 days: 3904.1±153.2 pmol/mg proteinxmin; 21 days: 4679.1±339.9 pmol/mg proteinxmin) (Fig. 2
A, B). Without evidence of an impairment of the NO-cGMP signaling at the level of NOS and sGC 1/β1, we hypothesized that hypoxic-induced elevation of pulmonary vascular resistance may be explained by increased cGMP degradation by PDE. To test this, we determined the protein expression of PDE 5 and the activity of cGMP-specific phosphodiesterases in lungs of mice exposed to hypoxia for 5 and 21 days vs. normoxia.
|
Expression and activity of phosphodiesterase PDE 5 in lungs of mice exposed to hypoxia is unaltered or even decreased in lungs of mice exposed to hypoxia
Determination of immunoreactive signals in lungs of mice with a commercial PDE 5 antibody used by Murray et al. in lung homogenates of rat (29)
yielded at least four signals (Fig. 1A
). The SDS mobility of one signal was 93 kDa, consistent with the reported molecular weight of PDE 5 in lungs of rat (8
, 29)
. Surprisingly, quantification of the 93 kDa signal showed no alteration in hypoxic vs. normoxic-treated animals after 5 days (95.5±8.9% of norm) but a significant decrease after 21 days (73.3±6.9% of norm, P<0.05) (Fig. 1D, E
). Calcium stimulated overall activity of cGMP-hydrolyzing phosphodiesterases was 3000 pmol cGMP/mg protein x min and was not changed in lungs of hypoxic-treated animals (5 days: 3757.2±105.7 pmol/mgxmin; 21 days: 3363.7±205.3 pmol/mgxmin) vs. normoxic animals (5 days: 3399.6±196.3 pmol/mgxmin, 21 days: 2691.2±290.3 pmol/mgxmin) (Fig. 2C, D
). We showed that 70% of the overall activity of cGMP-specific PDEs is inhibited by sildenafil and therefore likely to represent PDE 5 activity. However, we found no alteration of sildenafil-inhibited activity in lungs of mice exposed to 5 or 21 days of hypoxia (5 days: 2622.6±186.7 pmol/mgxmin; 21 days: 2190.5±231.8 pmol/mg) vs. normoxia (5 days: 2439.0±241.7 pmol/mgxmin; 21 days: 1633.8±264.0 pmol/mg). Finding unaltered sGC and PDE expression with unaltered or, in the case of PDE 5, even decreased activity in hypoxic conditions, we sought to clarify whether the increased NO expression results in increased cGMP levels by measuring perfusate cGMP levels in perfused mice lung after exposure to hypoxia. We elucidated the effectiveness of the NO-cGMP signaling by determining the phosphorylation of VASP on Serin 239 by cGK I as an overall marker of pathway effectiveness.
Increased cGMP levels lead to increased VASP phosphorylation via activation of cGK I
Accumulation of perfusate levels of cGMP were augmented in mice lungs exposed to 5 or 21 days of hypoxia (5 days: 1.73±0.34 µM at 30 min, 3.40±0.56 µM at 60 min; 6.13±0.81 µM at 90 min; 21 days: 0.77±0.30 µM at 30 min, 2.95±1.84 µM at 60 min; 3.85±0.81 µM at 90 min) vs. normoxia (5 days:1.02±0.16 µM at 30 min, 1.80±0.25 µM at 60 min; 2.84±0.51 at 90 min; 21 days: 0.81±0.48 µM at 30 min, 0.79±0.30 µM at 60 min; 1.71±0.5 at 90 min) (Fig. 3
C, D). However, this did not reach statistical significance. Without alteration in cGK I expression after 5 or 21 days of hypoxia vs. normoxia (95.4±9.1% of norm; 82.0±5.3% of norm) (Fig. 1)
, this results in increased phosphorylation of VASP with statistical significance after exposure to 21 days of hypoxia vs. normoxia (5 days: 136.8±21.4% of norm; 21 days: 349.0±84.5% of norm) (Fig. 1D, E
). Because of interfering unspecific signals of mouse immunglobulins with the secondary antibody, we were only able to quantify the immunoreactive signal at 46 kDa for P-VASP (Fig. 1A
). Alteration of whole VASP content in crude lung homogenate was excluded by determination of protein expression of VASP (Fig. 1)
. Given that our results indicate up-regulated NO-cGMP signaling in hypoxic conditions, we wanted to test whether acute sildenafil treatment targets NO-cGMP signaling by measuring perfusate cGMP and VASP phosphorylation in perfused mice lungs after exposure to hypoxia with the addition of sildenafil to the perfusate. We also measured pulmonary artery pressure to show the physiological relevance.
|
Acute sildenafil treatment after exposure to hypoxia further increased cGMP perfusate levels and VASP phosphorylation, resulting in attenuated HPH after 5 days but not after 21 days of hypoxia
Addition of the PDE 5 inhibitor sildenafil (100 nM) to the perfusate significantly further elevated cGMP in both hypoxic (5 days, 21 days) and normoxic animals, yet to a greater extent in hypoxic (5 days: 5.48±1.75 µM at 30 min, 17.60±7.86 µM at 60 min, 17.58±3.79 µM at 90 min; 21 days: 6.12±3.67 µM at 30 min, 12.70±10.57 µM at 60 min, 15.50±9.28 µM at 90 min) vs. normoxic mice (5 days:1.75±0.30 µM at 30 min, 4.08±1.19 µM at 60 min; 6.20±2.39 µM at 90 min; 21 days:1.30±0.71 µM at 30 min, 1.34±0.43 µM at 60 min; 1.50±0.57 µM at 90 min) (Fig. 3C, D
). Whereas sildenafil treatment is followed by an increase in cGMP perfusate levels after 5 and 21 days of hypoxia and, as a consequence, phosphorylation of VASP is elevated by sildenafil treatment in lungs of animals exposed to 21 days of hypoxia vs. normoxia (335.84±36.4 of norm, P<0.01) (Fig. 3E
), PAP was reduced to that of normoxic-treated animals (PAP mean of normoxic animals 10 mmHg) after 5 (98.92±4.83% of norm at 5 min, 100.05±12.32% of norm at 30 min, 88.61±4.78% of norm at 60 min, 88.57±5.72% of norm at 90 min) but not after 21 days (116.05±5.92% of norm at 5 min, 109.89±5.11% of norm at 30 min, 108.04±4.88% of norm at 60 min, 106.65±4.63% of norm at 90 min) of hypoxia (Fig. 3A, B
).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The method of determining protein expression and activity of all key enzymes in whole-lung homogenate provides a concise overview of the effects of hypoxia on NO-cGMP signaling. However, small subcellular changes that might be important cannot be excluded. We were able to determine increased NOS II and NOS III expression consistent with many earlier reports in rats and mice using whole-lung homogenate (17
18
19)
; however, Murata et al. found unchanged expression of NOS III, with decreased activity in isolated pulmonary artery of rats exposed to hypoxia (14)
. Our finding of unchanged PDE 5 activity with significantly decreased PDE 5 protein expression after 21 days of hypoxia implies that small changes previously reported in PDE 5 activity and protein expression in distinct areas of lung (8
, 29
, 30)
could be of interest. Whereas Sebkhi et al. (8)
found a significant increase in PDE 5 protein expression in small pulmonary resistance vessles, Murray et al. (29)
and MacLean et al. (30)
detected a significant increase in PDE 5 protein expression and activity in large (but not small) pulmonary vessels in lungs of rats exposed to 2 wk of hypoxia. Nevertheless, given that small resistance pulmonary arteries are the major site in the pathogenesis of HPH compared to larger vessels, the consequence of an increase in PDE 5 activity in large pulmonary arteries remains to be determined. Although we found unaltered overall activity of cGMP PDEs in the lung, other isoenzymes of PDE might be of interest in the setting of pulmonary hypertension. This is suggested by the recent report by Schermuly et al., who found increased expression in the lungs of hypoxic mice (55)
.
Discrepancies in published results concerning effects of hypoxia on NO-cGMP signaling in lungs of mice could also be due to the different duration of hypoxic exposure. Most studies using mice (9
, 10
, 21)
exposed the animals for 2 or 3 wk to hypoxia, consistent with most widespread studies in rats (11
, 12
, 24)
.
Demiryück et al. could show that NOS III is up-regulated in rat lungs after exposure to hypoxia for 2 days (20)
. Additional NOS III expression was shown to be increased in isolated porcine pulmonary arteries already after 2 h of exposure to hypoxia (22)
. These data suggest a fast response of NO-cGMP signaling to hypoxia, whereas development of hypoxia-induced pulmonary hypertension reaches a plateau in vivo after 3 wk of hypoxia in rats (8)
. In a model of the early phase of HPH in which mice were maintained under normobaric hypoxia (10% O2) for 5 days, we showed that NO-cGMP signaling is up-regulated comparable to longer durations of hypoxia; HPH is completely reversible at this stage of disease. This time point might be interesting when investigating the mechanism of acute hypoxic pulmonary vasoconstriction before irreversible HPH has occurred. Acute sildenafil treatment could not reduce PAP in animals exposed for 21 days to hypoxia, suggesting that acute sidenafil treatment reduces PAP only in the early stages of HPH; acute sildenafil treatment after full establishment of HPH does not, at least in the perfused lung model of mice. To our knowledge, only one report about rats (by Pauvert et al.) shows different results in perfused lung after acute sildenafil treatment in rats (56)
. Results in isolated perfused lungs may differ from those in intact animals because pulmonary vascular tone is very low due to the lack of vasoconstrictors present in blood in vivo, such as endothelin-1 or thromboxane.
Other important discrepancies between our results and previous reports concerning expression of sGC and nitration of tyrosine are likely due to specificity issues of the different antibodies used. Care should be should be taken when using commercial nitryrosine and sGC antibodies. A previous report concerning sGC activity showed increased basal activity in hypoxic conditions (25)
. An alteration in the basal production of cGMP may reflect changes in both sGC and particulate GC (pGC) activity, the latter of which is activated by intracellular ANP but not by NO. Therefore, we determined NO-stimulated sGC activity and found no alteration induced by hypoxia. Sildenafil may also enhance the pGC-cGMP pathway in PH, as its ability to reduce pulmonary vascular remodeling is blunted in ANP knockout mice (10)
and synergistic effects of combined sildenafil and ANP treatment have been observed (57)
. However, a role for pGC in pulmonary hypertension has not been suggested.
To our knowledge, increased posphorylation of VASP as a marker for the net result of the NO-cGMP pathway is reported for the first time in lungs of mice exposed to hypoxia. Quantification of specific immunoreactive signal is limited to the signal at 46 kDa because of unspecific signals of the secondary antibody against mouse immunoglobulins, with an SDS mobility at 50 kDa. This biomarker may have clinical applications similar to its use in platelet therapeutic monitoring (44)
. We cannot exclude the possibility that sildenafil targets other pathophysiological mechanisms at the onset and in established HPH. This raises the question as to which mechanisms might be important for reversible acute hypoxic vasoconstriction and irreversible chronic HPH that counteract the NO-cGMP signaling.
An increasing number of reports suggest that NADPH oxidase (NOX) isoforms for NOX 1 and 4 are important sources of ROS that are involved in vascular remodeling and regulation of vascular tone (58
59
60)
. Notably, Martyn et al. reported that Nox 4 produces predominately hydrogen peroxide and not superoxide (61)
. Therefore, alterations in Nox 4 protein expression or activity may not necessarily lead to superoxide, which is required to scavenge NO, or form ONOO– and subsequently lead to tyrosine nitration, but are signals into H2O2-dependent genes important for inflammatory or vascular remodeling processes. Furthermore, the cAMP pathway is reported to be attenuated in hypoxia (10
, 62
, 63)
. Thus, given that cGMP can impair cAMP-specific phosphodiesterases (e.g., PDE1, PDE3) (64)
, the ability of sildenafil to elevate cGMP may also lead to a decreased breakdown of cAMP. However, we found no alteration of VASP phosphorylation on Serin157 as a marker for activation of cAMP-dependent kinases.
Besides the mechanism of acute hypoxic vasoconstriction, we think it is of great importance to further elucidate the mechanisms of pulmonary remodeling that appear to be the main cause for developing the disease of HPH. The mechanisms that trigger pulmonary remodeling might be completely different from the pathophysiology of acute hypoxic pulmonary vasoconstriction, as previously suggested (65)
. The number of studies concerning the remodeling process increases and reveals promising options for therapy with for the PDGF pathway (66)
, angiotensin-converting enzyme system (67)
, or rho-kinases as target (68)
. Notably, sildenafil was recently shown to influence rho-kinase signaling by preventing RhoA expression, which further raises the hypothesis that sildenafil may have targets besides the augmentation of NO-cGMP signaling (69)
.
Taken together, our results demonstrate that the NO cGMP pathway is up-regulated in HPH, counteracting hypoxic vasoconstrictive mechanisms. Acute sildenafil treatment alleviates HPH via a compensatory up-regulation of NO-cGMP signaling and leads to reduced HPH as long as pulmonary remodeling is not fully established. Although sildenafil offers an exciting new therapeutic strategy in treating pulmonary hypertension, the initial vasoconstrictive mechanism and the ongoing pulmonary remodeling have to be further investigated and represent promising targets of future treatments such as sGC activators and stimulators (11
, 12)
. Even our comprehensive study cannot exclude the possibility that sildenafil may also indirectly enhance the pGC-cGMP pathway in PH, as its ability to reduce pulmonary vascular remodeling is blunted in ANP knockout mice (10)
, and synergistic effects of combined sildenafil and ANP treatment have been observed (55)
. Other mechanisms of pulmonary vasoconstriction and remodeling in humans may include endothelin-1, which is already targeted by the receptor antagonist bosentan (70)
, serotonin, and vasoactive intestinal peptide (71
, 72)
.
|
ACKNOWLEDGMENTS |
|---|
Received for publication December 28, 2006. Accepted for publication July 5, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A.-C. Peyter, V. Muehlethaler, L. Liaudet, M. Marino, S. Di Bernardo, G. Diaceri, and J.-F. Tolsa Muscarinic receptor M1 and phosphodiesterase 1 are key determinants in pulmonary vascular dysfunction following perinatal hypoxia in mice Am J Physiol Lung Cell Mol Physiol, July 1, 2008; 295(1): L201 - L213. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and cross-reactive immunoreactive signals of NOS II, NOS III, nitrotyrosine (NT p70, p38), sGC (
, 
) or with addition of 100 nM sildenafil (•, 
).Values are means of PAP ± SE, expressed as % of normoxic control (A, B) or means ± SE of perfusate cGMP in [µM] (n=4–7) (C, D). E) Effect of sildenafil treatment on VASP phosphorylation of animals exposed 21 days to hypoxia. Shown is the densitometric quantification of VASP46, VASP46 phosphorylation, and VASP50-to-VASP46 ratio in lungs of animals exposed to 21 days to hypoxia (filled bars) vs. normoxia (open bars) after perfusion for 90 min with addition of sildenafil (+sildenafil, 100 nm) to perfusate vs. untreated animals. Values are means ± SE of the specific antibody/antiactin immunoreactive signal ratios of 6 or 7 animals expressed as % of normoxic control. Asterisks indicate statistically significant differences (*P<0.05, significance compared to normoxic animals;