When oxygen availability is reduced, as it is the case of hypoxia, thin-walled pulmonary arteries constrict (Michelakis, 2003), increasing pulmonary arterial pressure (Coggins and Bloch, 2007). Concerning the airways, there is growing evidence showing that hypoxia reduces bronchial calibre, indicating bronchoconstriction (Dagg et al., 1997). However, there are some factors such as nitric oxide, which can mediate to maintain pulmonary blood flow and bronchial tone in oxygen-deficiency situations.
Nitric oxide (NO) is a vasodilator and bronchodilator molecule with a critical role in numerous physiologic and inflammatory processes in the lung (Shaul et al., 1994; Nagasaka et al., 2008). NO is generated from L-arginine via NO synthases (NOS). Three NOS isoforms have been well characterized: endothelial NOS (eNOS), inducible NOS (iNOS), and neuronal NOS (nNOS) (Förstermann et al., 1991). eNOS and nNOS are constitutively expressed, while iNOS transcription can be activated by stimuli such as endogenous mediators (chemokines and cytokines) and exogenous factors (hypoxia) (Ricciardolo et al., 2004).
Two main sites of NO production have been identified in the lung, the vasculature and the airways (Kobzik et al., 1993; Sherman et al., 1999; MacRitchie et al., 2001). The main isoform in vascular endothelium is eNOS. In the airways, all three isoforms have been detected in bronchial epithelium (Kobzik et al., 1993; Shaul et al., 1995). Nevertheless, it is important to mention that nNOS does not have an important function in either hypoxic or normoxic rat lung (Le Cras et al., 1996; Fagan et al., 1999; Shirai et al., 2003).
When NO is supplied in excess, as in hypoxia/reoxygenation situations, part of it released from NOS can rapidly react with superoxide to form peroxynitrite, an oxidizing agent (Pryor and Squadrito, 2005). This specie may cause lipid peroxidation, apoptosis, alterations in DNA, and protein nitration and oxidation, all of which can lead to profound cellular disturbances (Szabo, 1996). Particularly, one of the targets of peroxynitrite is tyrosine; the oxidizing attack to these residues in proteins forms the stable product 3-nitro-L-tyrosine (nitrotyrosine), which can be used as a marker of the potentially cytotoxic effect of NO (Beckman and Koppenol, 1996).
Previous studies of NOS expression in hypoxic lung have yielded conflicting results (Fagan et al., 2001; Zulueta et al., 2002). Indeed, changes in the NO/NOS system as well as the role of NOS in hypoxia/reoxygenation situations still remain controversial, probably due to the severity of the hypoxia model, the enormous intrinsic complication of the NO/NOS regulation system, and the species and tissue utilised.
In this context, we have made a detailed study of the NO/NOS system, including some parameters of cell and tissue damage such as nitrated protein expression and apoptosis, at seven different reoxygenation times (0, 2, 12, 24, 48, 72 h, and 5 days) after a situation of acute hypobaric hypoxia. The study of this reoxygenation period is a quite novel way to deal with these types of studies.
The study was performed on mature adult (4–5-months old) male albino Wistar rats kept under standard conditions of light and temperature and allowed ad libitum access to food and water. All the experiments were conducted according to E.U. guidelines on the use of animals for biochemical research (86/609/EU) and to the Bioethical Committee of the University of Jaén.
The acute hypobaric hypoxia (AHH) was carried out as previously published (Lopez-Ramos et al., 2005; Martínez-Romero et al., 2006). Briefly, for 30 min, rats were exposured to hypobaric hypoxia (9142 m, 30.000 ft). For getting AHH conditions, the experimental chamber was decompressed to 225 mmHg, resulting in a 48 mmHg O2 partial pressure. After the AHH period, animals were kept at atmospheric pressure for different reoxygenation times (0, 2, 12, 24, 48, 72 h, and 5 days). Adult animals maintained in the chamber for 30 min under normobaric normoxic conditions were used as controls.
A total of 40 albino Wistar rats were used for the biochemical experiments (five animals per experimental group). After the corresponding reoxygenation times, the rats were sacrificed by cervical dislocation and the lungs were immediately removed, rinsed in saline solution and stored at −80°C until used. Another 40 rats were used for histological techniques (five animals per experimental group). The rats were anaesthetized with Ketolar (Parke Davis, 1 mL/250 g total weight) and perfused in each reoxygenation time. Then the lungs were removed, rinsed in saline solution and fixed with paraformaldehyde 4% (4 h for the NADPH-diaphorase histochemistry and 18 h for immunohystochemistry).
Quantitative Real-Time Polymerase Chain Reaction
Total lung RNA, isolated using Trizol Reagent according to the manufacturer's instructions, was used for cDNA synthesis with Superscript reverse transcriptase, also according to the manufacturer's instructions. Quantitative real-time polymerase chain reaction (RT-PCR) was carried out in the Stratagene Mx3005PTM thermal cycler. Hypoxia inducible factor-1ß (HIF-1ß) was used as the reference gene, as it is constitutively expressed under normoxic and hypoxic situations (Semenza, 2001; Sharp et al., 2004). Primers used in RT-PCR to assess the expression of eNOS, iNOS and HIF-1ß are described in Table 1.
Table 1. Primers used in RT-PCR to assess the expression of endothelial and inducible nitric oxide synthases (eNOS, iNOS), and the endogenous reference hypoxia inducible factor-1β (HIF-1β) in rat lung homogenates
Amplification conditions consist of initial denaturing (95°C, 10 min), followed by 45 cycles of: denaturing (95°C, 30 s), primer annealing (60°C, 30 s), elongation (72°C, 30 s); and later 95°C (1 min), 55°C (30 s) and 95°C (30 s). Experiments were performed in triplicates, and the values were used to calculate the ratio of NOS to HIF-1ß, with a value of 1 used as the control (calibrator). Relative expression of NOS was calculated by the 2[−ΔΔC(T)] method (Li et al., 2006).
Western Blot Analysis
For Western blot analysis, equal amounts of denatured lung total-protein extracts were loaded and separated in 7.5% SDS-polyacrylamide gel. Proteins in the gel were transferred to a PVDF membrane (Amersham Pharmacia Biotech, Buckinghamshire, UK) and then blocked. Monoclonal antibodies to eNOS (1/800, BD Transduction Laboratories), iNOS (1/700, BD Transduction Laboratories), and α-tubulin (Sigma, St. Louis, MO, USA), as internal control, were used to detect the respective proteins. Polyclonal antibody rabbit anti-3-nitro-L-tyrosine A4 (1:3000, gift from Professor J. Rodrigo from CSIC Cajal Institute of Madrid) was used to detect nitrated proteins. Antibody reaction was revealed with chemiluminescence detection procedures according to the manufacturer's recommendations (ECL kit, Amersham Corp., Buckinghamshire, UK).
NOS have been shown to have NADPH-diaphorase activity, as evidenced by colocalization and coprecipitation of NADPH-diaphorase and NOS activity (Snyder, 1992). In fact, NADPH-diaphorase histochemistry has been widely used as an indirect way to determine in situ NOS activity (Kugler et al., 1994; Roufail et al., 1995; Moraes et al., 2001). Lung sections, 40-μm thickness, were cut using a cryostat (2800 Frigocut E, Reichert-Jung Vienna, Austria). The lungs were included in O.C.T. medium (Sakura), and then frozen in 2-metilbutane with liquid nitrogen. Free-floating sections were incubated for 4 h in PBS containing 0.1% Triton X-100. After several washes in 0.1 M Tris-HCl pH 7.4 buffer, they were incubated in the dark, for 45 min at 37°C, in 0.1 M Tris-HCl, pH 7.4, containing 1 mM β-NADPH and 2 mM NBT (in 70% dimethylformamide). The sections were then washed twice with 0.1 M Tris-HCl, pH 7.4, quickly dehydrated in a graded ethanol series, cleared, and mounted in DPX (Fluka, Madrid, Spain).
Lungs were fixed with paraformaldehyde 4% for 18 h, and then embedded in paraffin (Paraplast Extra, Tyco). Sections were incubated for 30 min with 10% serum of the animal in which the secondary antibody was produced. Then, they were first incubated with diluted monoclonal anti-iNOS (1:200) and anti-eNOS (1:150) antibodies (Transduction Laboratories), and polyclonal rabbit anti-3-nitro-L-tyrosine A4 antibody (1:500, gift from Cajal Institut of Madrid) in PBS overnight at 4°C, and later with biotinylated secondary antibodies (Pierce) followed by peroxidase-linked ABC. The peroxidase activity was demonstrated following the nickel-enhanced diamino-benzidine procedure (Shu et al., 1988). Sections were mounted on slides, dehydrated, and covered using DPX. Controls for background staining were performed by replacing the primary antibody with PBS.
The reaction of NO with ozone results in the emission of light, and this light (emitted in proportion to the NO concentration) is the basis of one of the most accurate NO assays available (Fontijn et al., 1997; Laitinen et al., 1993). NO production has been indirectly quantified by nitrate/nitrite and S-nitrose compounds (NOx) measurement using an ozone chemiluminescence-based method. For this technique, lungs were homogenized in PBS with protease inhibitors. Homogenates were then sonicated, centrifuged, and deproteinized with NaOH 0.8 N and ZnSO4 16% solutions. The total amount of NOx was determined by a modification (Lopez-Ramos et al., 2005) of the procedure described by Braman and Hendrix (1989) using a NO analyser (NOA™ 280i Sievers Instruments). NOx concentrations were calculated by comparison with standard solutions of sodium nitrate. Final NOx values were referred to the total protein concentration in the initial extracts.
TUNEL Assay for Assessment of Apoptotic Cell Death
Terminal deoxynucleotidyl transferase [TdT]-mediated desoxyuridinetriphosphate [dUTP] nick end-labelling (TUNEL) is a technique to estimate apoptosis in tissue sections. The protocol was performed in sections obtained from lungs embedded in paraffin according to the manufacturer's recommendations (TdT-FragELTM DNA Fragmentation Detection Kit, Calbiochem). Deionized water was used instead of TdT enzyme as a negative control. Apoptotic bodies were stained brown.
Data were expressed as mean ± SD (standard deviation). The statistical treatment to evaluate significant differences between groups was performed with SPSS 15.0 software. The data followed a normal distribution (tested with Kolmogorov-Smirnov test) and the principle of homoscedasticity of variances (tested with Levene test), and were tested by one-way ANOVA. The statistical signification degree was established by applying Bonferroni test to compare differences between means. The statistically significant differences versus the control group were expressed as *P < 0.05; **P < 0.02; ***P < 0.01; ****P < 0.001.
NOS mRNA Expression
The quantitative analysis of eNOS mRNA expression (Fig. 1A) showed a significant increase at 0 h of reoxygenation (P < 0.001), which persisted until 2 h posthypoxia (P < 0.001). Afterwards, mRNA levels fell to basal levels. iNOS mRNA levels (Fig. 1B) significantly rose after the AHH episode from 2 h (2 h; P < 0.001; 12 h; P < 0.05) to 24 h posthypoxia (P < 0.001). For the rest of the reoxygenation period, mRNA levels descended to the level found in the control group.
NOS Isoforms Expression
The expression of eNOS protein was detected as a 140-kDa band (Fig. 2A), which rose significantly immediately after AHH (0 h; P < 0.001). This early rise was followed by a fall at the final reoxygenation times, reaching basal levels. The iNOS expression was detected as a 130-kDa protein band in all the experimental groups (Fig. 2B). The protein level rose from 48 h of reoxygenation, and values increased progressively until 5 days (48 h, 72 h, 5 days; P < 0.001).
In the microphotographs (Fig. 3), the staining showing in situ NOS activity was found in all the experimental groups, mainly in endothelial cells of the blood vessels that irrigate the pulmonary parenchyma and in bronchiolar epithelial cells. At 0 h posthypoxia, it was possible to detect an increase in the NADPH-diaphorase reaction intensity both in endothelial and in epithelial cells with respect to the control group. It was not until 48 h of reoxygenation that the staining again became more intense and it continued high until the 5 days after the AHH episode. We have shown here only the most representative images, corresponding to 4 from the 8 experimental groups.
eNOS and iNOS immunohistochemistry
In all the experimental groups, eNOS immunoreactivity was detected in bronchiolar epithelium and vascular endothelial cells (Fig. 4). The eNOS-positive staining was detected more intense in these two cell types when AHH ended (0 h of reoxygenation). Afterwards (from 2 h to 5 days), the staining intensity did not show detectable differences in the histological sections compared to the control group.
As can be observed in Fig. 5, iNOS immunorreactivity was found in bronchiolar epithelial cells, endothelial cells, and vascular smooth muscle cells. The intensity of iNOS-positive staining was higher in the late reoxygenation times (from 48 h to 5 days) when compared with the control group.
We have shown here only the most representative images for eNOS and iNOS location, corresponding to 4 from the 8 experimental groups.
Table 2 shows the determinations of nitrate/nitrite and other S-nitrose compounds (NOx) in the different experimental groups. As can be seen, NOx levels underwent an early rise at 0 h posthypoxia (P < 0.001), and afterwards dipped to reach values similar to those corresponding to the normobaric normoxic control.
Table 2. Nitrate, nitrite, and other S-nitrose compounds (NOx) in rat lung homogenates
NOx level (μmol/mg protein)
Experimental groups: Control, and 0, 2, 12, 24, 48, 72 h, and 5 days post-hypoxia. Data are the mean ± SD (standard deviation) of five determinations (n = 5). The statistically significant differences versus the control group were expressed as *P < 0.05; **P < 0.02; ***P < 0.01; ****P < 0.001.
280.06 ± 65.96
548.48 ± 55.23****
253.17 ± 65.72
271.63 ± 42.88
262.91 ± 64.35
250.57 ± 57.59
290.52 ± 8.41
221.40 ± 80.17
Three nitrotyrosine immunoreactive bands, corresponding to proteins of 126, 112, and 70-kDa were detected in all the experimental groups (Fig. 6). The quantitative evaluation of bulk-nitrated proteins increased significantly from 48 h to 5 days posthypoxia (48 h, 72 h, 5 days; P < 0.001).
In all the experimental groups, the nitrotyrosine immunoreactivity was detected mainly in bronchiolar epithelial cells and some cells of the vascular endothelium (Fig. 7). Corroborating the previous result, the nitrotyrosine-positive staining was detected more intense from 48 h to 5 days posthypoxia in these cell types with regard to the control group. We have shown here only the most representative images, corresponding to 4 from the 8 experimental groups.
The TUNEL reaction, which identifies apoptotic cells, showed a low apoptosis level at the early reoxygenation times in the hypoxic lung. However, it is possible to identify an increased apoptosis level from 48 h to 5 days posthypoxia (Fig. 8). We have shown here only the most representative images, corresponding to 4 from the 8 experimental groups.
This is one of the few studies that describe the time-course changes in the NO/NOS system during the reoxygenation period after an acute hypobaric hypoxic (AHH) process in the rat lung.
Our results have shown an increase in eNOS mRNA and protein expression immediately after AHH, coinciding with a rise in in situ NOS activity, eNOS immunostaining, and NO level, indirectly quantified by nitrate/nitrite and S-nitrose compounds (NOx). From these data, we may assume that eNOS contributes to the increase in NOx level at 0 h of reoxygenation, as an immediate response to hypoxia. This NO could produce bronchodilation, and pulmonary vasodilation, avoiding vasoconstriction and protecting the lungs against pulmonary hypertension, as has been long proposed (Brashers et al., 1988). In fact, hypoxia has been shown to modulate eNOS expression (Liao et al., 1995; Le Cras et al., 1996). The higher eNOS protein and mRNA levels after chronic hypoxia has been already reported in previous works (Le Cras et al., 1996, 1998; Quinlan et al., 2000), however, this work is one of the few ones using a model of AHH. One of the most accepted mechanisms by which hypoxia could increase eNOS expression is based on the fact that hypoxia can activate the transcription of a reporter gene under control of the eNOS 5′-flanking region (Arnet et al., 1996). In addition, Hoffman et al. (2001) suggested that activation of the redox-sensitive activator protein-1 transcription factor during hypoxia might be responsible for the increase in eNOS expression. On the other hand, the augmented eNOS expression coincides with a high staining intensity in eNOS-positive structures and increased in situ NOS activity, however, there are no changes in eNOS cellular location. In this work, eNOS has been found to be located in endothelial and bronchiolar epithelial cells, coinciding with previous descriptions in a chronic hypoxia model (Xue et al., 1994; Le Cras et al., 1996). Also in this study, in situ NOS activity was detected by NADPH-diaphorase staining in the same location as eNOS, in agreement with other authors (Xue et al., 1994). Moreover, we have found both in situ NOS activity and eNOS positive staining in the endothelium of pulmonary vessels, not only in the hypoxic but also in the normoxic lung. This result contrasts with some reports that failed to detect eNOS in the endothelium of control arteries (Shirai et al., 2003), and also with other authors who did not show in situ NOS expression in the endothelium of normoxic pulmonary vessels (Xue et al., 1994).
Alternatively, iNOS mRNA levels rose from 2 h to 24 h of reoxygenation, while iNOS protein expression increased later between 48 h and 5 days posthypoxia. This no-correlation between mRNA and protein expression coincides with the results of Chibana et al. (2008), who also found that iNOS mRNA and protein were not correlated in bronchial epithelial cells. This no-correlation may be because some time it could be necessary for iNOS protein translation. In addition, it would be possible for part of the increased iNOS mRNA in early reoxygenation times to be stabile enough to prompt an iNOS protein rise some hours later from this increase. In this regard, the antisense strand corresponding to the 3′-untranslated region (3′UTR) of iNOS mRNA is transcribed from the iNOS gene. This natural antisense transcript interacts with iNOS mRNA, causing its stabilization (Matsui et al., 2008). To conclude, the iNOS boost detected in the last reoxygenation times may suggest that this isoform could be involved in a late response to hypoxia.
Curiously, it is not surprising that our results showed an iNOS baseline expression in the normoxic lung. In this sense, it has been previously reported a “constitutive” expression of iNOS at mRNA and protein level in human respiratory epithelial cells (Kobzik et al., 1993; Guo et al., 1995). Nevertheless, changes in iNOS mRNA and protein expression in the hypoxic lung are controversial in the literature. Some works have reported an increase in iNOS mRNA (Le Cras et al., 1996; Fagan et al., 2001), as well as a higher protein level (Le Cras et al., 1996; Quinlan et al., 1998) in the rat lung following hypoxia, whereas others have indicated that hypoxia alone does not induce the expression of iNOS mRNA in endothelial cells (Zulueta et al., 2002). The most accepted hypotheses trying to explain how hypoxia increases iNOS expression implies the hypoxia inducible factor-1 (HIF-1), a central component of the oxygen sensing system that coordinates cellular response in decreased oxygen availability conditions (Jung et al., 2000). iNOS promoter contains a hypoxic responsive element, and in hypoxic situations HIF-1 could bind this element producing an increase in iNOS transcription (Palmer et al., 1998; Brüne and Zhou, 2003). On the other hand, at late reoxygenation times (from 48 h to 5 days), we have detected greater in situ NOS activity in the vascular endothelium and bronchiolar epithelium, probably as a consequence of the iNOS upregulation during this period. These results agree with our iNOS immunostaining data: iNOS was found to be augmented at late reoxygenation times in endothelial and vascular smooth muscle cells, as well as in bronchiolar epithelial cells.
In this study, the nitrotyrosine-modified protein expression showed the same expression pattern as the inducible isoform, increasing from 48 h to 5 days posthypoxia. Moreover, these results were corroborated with the nitrated-modified protein location data. These results suggest that the iNOS-derived NO could be involved in the protein nitration process. Curiously, although both iNOS and nitrotyrosine-modified protein expression augmented in the late reoxygenation times, no increased NOx levels have been found at those times in our AHH model, probably because the reoxygenation period causes further NO consumption through its interaction with superoxide to produce peroxynitrite (Pryor and Squadrito, 1995).
Our results have also suggested that our AHH model produces cell damage as it is indicated by the use of the TUNEL assay, indicator of apoptosis. In this study, an increased apoptotic level were detected from 48 h and 5 days posthypoxia, coinciding with the increase in iNOS expression, and suggesting that iNOS-derived NO could be involved in the apoptotic events occurring in the hypoxic lung. In this sense, it has been previously reported that there is a relationship between NO and apoptosis, and it has become clear that NO acts as a second messenger activating a number of cytokines and inducing apoptosis (Saugstad, 2000).
All these findings indicate that hypoxia/reoxygenation processes alter NO concentration and NOS expression in the lung. We have reported in this study that two responses to hypoxia/reoxygenation occur in this organ: an immediate physiological response mediated by eNOS; and a late pathological one, mediated by iNOS. In addition, we have identified a pattern of NOS upregulation: eNOS and iNOS are time-dependent upregulated in the reoxygenation period. We suggest that just after the hypoxic stimuli, NO is produced mainly by eNOS in order to ensure a sufficient amount to guarantee pulmonary vasodilation and bronchodilation, thereby avoiding vasoconstriction. In this scenario, we propose that the upregulation of iNOS at later reoxygenation times could be deleterious, provoking protein nitration and apoptosis.
The authors wish to thank to Mr. Rafael Lomas for his statistic assistance, and Mr. David Nesbitt for his editorial help with the English version of the manuscript.