Decreased haem oxygenase-1 and increased inducible nitric oxide synthase in the lung of severe COPD patients

Oxidant/antioxidant imbalance has been implicated in thepathogenesis of chronic obstructive pulmonary disease (COPD) 1. Exogenous oxidants are contained in cigarette smoke, the major risk factor for COPD, and are produced bythe inflammatory cells that characterise the pathology ofCOPD 2. Oxidative stress induces a cascade signalling mechanism, which activates transcription factors leading to upregulation of genes coding for pro-inflammatory and anti-inflammatory cytokines and enzymes 1. This study focuses on two of these enzymes, haem oxygenase (HO) and inducible nitric oxide synthase (iNOS). The three isoforms of HO, inducible HO-1, constitutive HO-2 and HO-3, catalyse the degradation of haem to biliverdin, which is then reduced to bilirubin, producing free iron and carbon monoxide (CO) 3–4. HO has been suggested to have important antioxidant capacity and significance in protecting lung cells against oxidant exposure. HO-1 was demonstrated to be induced by a variety of injurious stimuli, including reactive oxygen species (ROS) 3. In an earlier study, the present authors showed thatchronic smokers exhibit increased expression of HO-1 inalveolar spaces and HO-2 in alveolar walls of the lung 5.

Nitric oxide (NO) synthase may play a role in the pathogenesis and progression of COPD, since NO has direct toxic effects and also reacts with superoxide and generates toxic peroxynitrite in the cells 6. Some studies found that iNOSwas more expressed in interstitial lung disease 7. Therole of this enzyme in the pathogenesis and progression of COPD is controversial 8, 9. In addition, most studies examined iNOS and its products in induced sputum, while it is believed that the peripheral airways are the major site ofairflow limitation in COPD. Investigating the mechanisms involved in the progression of COPD may prove usefulin the identification of subgroups of subjects prone torapid decline in lung function and targeting therapeutic intervention.

The current study analysed the expression of enzymes withantioxidant properties, as HO-1 and HO-2, and pro-oxidant properties, as iNOS, in the peripheral lung of patients with COPD and severe impairment of lung function in comparison to subjects with comparable smoking history, but with no impairment of lung function. The involvement of NO-derived oxidants was evaluated by immunostaining ofnitrotyrosine adducts to proteins. The immunoreactivity of HO, iNOS and nitrotyrosine wasquantified in alveolar macrophages and in alveolar walls, and theresults were correlated with indexes of lung function.

Methods

Results

Table 1⇑ shows the characteristics of the subjects examined. The two groups of subjects were similar with regard to age and smoking history. Severe COPD exhibited lower values ofarterial oxygen tension (p<0.001), CO diffusing capacity (p<0.001) and diffusion coefficient (p<0.01). According to theGlobal initiative for Chronic Obstructive Lung Disease (GOLD) classification 16, in the group of severe COPD sixpatients were in stage IIB and five patients in stage III.

The counts of alveolar macrophages in severe COPD (median 154 (interquartile range 143–174)) were not significantly different from those in control smokers (157 (142–204)). The majority of cells in alveolar spaces counted as alveolar macrophages were CD68-positive (70.2% (55.4–83.8)). HO-1 immunostaining was observed mainly in cells of alveolar spaces, whereas it was negative in bronchial epithelium and vessels (fig. 1⇓). Most of the HO-1-positive cells in alveolar spaces were CD68-positive (82.6% (72.4–88.9)). The percentage of HO-1-positive alveolar macrophages was significantly lower in subjects with severe COPD (26.0% (12.3–32.6)) compared with control smokers (44.1% (34.7–57.5); p<0.01; table 2⇓, fig. 2⇓). The percentages of iNOS+ (fig. 2⇓) and N-Tyr+ macrophages were not different between the two groups (table 2⇓).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.—

Immunohistochemistry study of human lung surgical specimens. a) Alveolar spaces stained for haem oxygenase (HO)-1 in a control smoker. b) Alveolar spaces stained for HO-1 in a severe chronic obstructive pulmonary disease (COPD) patient. c) Alveolar wall stained for inducible nitric oxide synthase (iNOS) in a control smoker. d) Alveolar wall stained for iNOS in a severe COPD subject. Formalin-fixed paraffin-embedded sections were processed for immunohistochemistry using antihuman HO-1 antibody and alkaline phosphatase-anti-alkaline phosphatase (APAAP) method developed with fast red (a, b) or antihuman iNOS antibodies and Envision secondary antibodies conjugated with peroxidase developed with diamino benzydine (c, d). All sections are counterstained with haematoxylin.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.—

Comparison of individual values of inducible nitric oxide synthase (iNOS) and haem oxygenase (HO)-1-positive alveolar macrophages between control smokers (•) and severe chronic obstructive pulmonary disease (▴). **: p<0.01.

Immunostaining for HO-2 was present in alveolar macrophages, in the alveolar wall, in smooth muscle and in cells of adventitia of pulmonary arteries, and in a few cells of bronchial wall. Percentages of HO-2 immunostaining of alveolar macrophages were similar in the two groups (COPD 75.4% (54.2–88.1) versus controls 67.9% (35.9–75.6)). The number ofHO-2-positive cells in the alveolar wall was significantly higher (p<0.01) in severe COPD (18.0 cells·mm⁻¹ (12.7–22.8)) compared to control smokers (11.3 cells·mm⁻¹ (10.6–12.2)) (table 3⇓, fig. 3⇓). Double immunostaining indicated that the HO-2-positive cells in the alveolar wall were type 2 pneumocytes (92.3% (88.6–93.9)). Conversely, 97% (93–98%) of type 2 pneumocytes were HO-2-positive.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.—

Comparison of individual values of inducible nitric oxide synthase (iNOS) and haem oxygenase (HO)-2-positive cells in alveolar wall between control smokers (•) and severe chronic obstructive pulmonary disease (▴). **: p<0.01; ***: p<0.001.

iNOS staining was present in the alveolar wall and alveolar macrophages, in smooth muscle, in adventitia of pulmonary arteries and bronchial wall. The number of iNOS-positive cells in the alveolar wall was significantly increased in severe COPD (13.9 cells·mm⁻¹ (11.4–19.2)) with respect of control smokers (4.2 cells·mm⁻¹ (2.3–5.9); p<0.001; table 3⇑, fig. 3⇑). By double immunostaining, most of iNOS-positive cells showed immunoreactivity for SP-B (85.7% (63.9–87.8)), indicating that they were type 2 pneumocytes. The extent of N-Tyr staining in alveolar wall tended to be greater in severe COPD than in control smokers, but the difference was not statistically significant (p=0.07, table 3⇑). Considering the two groups together, there was a positive, linear correlation between immunoreactivity for iNOS and N-Tyr in alveolar walls (rho=0.54, p<0.05). A similar relationship was observed between iNOS+ macrophages and N-Tyr+ macrophages (rho=0.41, p=0.06). iNOS+ cells in alveolar wall were inversely related toHO-1-positive macrophages (rho=−0.44, p<0.05).

Immunostaining for N-Tyr in alveolar wall inversely correlated with FEV₁/forced vital capacity (FVC) (rho=−0.62, p=0.05) in control smokers and showed the same trend in the group of severe COPD (rho=−0.60, p=0.07). No significant relationships were detected between immunostaining for HO-1, HO-2, iNOS, and lung function parameters or smoking history in the two groups of patients.

Discussion

The present study found that in severe COPD there was areduced expression of HO-1 in alveolar macrophages, whereas the numbers of type 2 pneumocytes expressing iNOSand HO-2 was increased. The rate of protein nitration in the lung tissue was directly related with the expression of iNOS and was associated with lower values of FEV₁/FVC ratio.

Oxidative stress and imbalance of host defence mechanisms have been implicated in the pathogenesis of COPD 1. Oxidants, either inhaled or produced by inflammatory cells, induce inflammatory responses in the lung via signalling mechanisms. Redox-sensitive transcription factors, such as nuclear factor-κB and activator protein 1, have been shown tobe activated in epithelial cells and inflammatory cells during oxidative stress, leading to upregulation of a number of anti- and pro-inflammatory genes, including those coding for HO-1 and iNOS 1. The functional significance ofinduction of HO-1 and iNOS is not well understood, as their activity can result in either protection or injury depending on experimental setting 17, 18.

Excessive production of NO via iNOS can cause adverse effects in lung tissue by reactions with superoxide anion and oxygen, leading to development of reactive nitrogen species 19. Through its products CO, iron and bilirubin, HO is believed to have a role in cytoprotection against oxidant injury 3. The two HO, HO-1 and HO-2, have differential roles and can only partially replace each other 20. In a previous study, the present authors found a different localisation of the two isoforms in the lung of healthy subjects and smokers, with HO-1 predominating in macrophages and HO-2 in lung parenchyma, particularly in type 2 pneumocytes 5. A similar distribution of HO-1 and HO-2 was detected insevere COPD patients, indicating that progression of inflammatory process and remodelling do not alter the localisation of these enzymes.

Expression of HO-1 was most prominent in alveolar macrophages of patients with sarcoidosis and interstitial pneumonia, whereas HO-1 immunoreactivity was not detectable or was very low in bronchiolar epithelium, alveolar epithelium and endothelium 21. This is in agreement with the HO-1 localisation in the samples of peripheral lung in the present study and contrasts with the finding of HO-1-positive bronchial epithelial cells in larger airways in asthma 22. Taken together, these observations suggest that expression of HO-1 in bronchial epithelial cells is dependent on the site of the bronchial tree examined.

The current authors previously showed that HO-1 and HO-2 expression was increased in smokers compared with nonsmokers, but no differences were found between healthy smokers and smokers with mild COPD 5. By studying more severe COPD patients, including those with a high degree ofparenchymal destruction, the current authors observed decreased HO-1 expression in alveolar macrophages and increased number of HO-2+ type 2 pneumocytes. As recently described, HO-1 gene promoter polymorphisms may reduce the inducibility of HO-1 by ROS 23, thereby resulting in different susceptibility to damage by cigarette smoke. The current findings confirm a relationship between HO-1 diminution and COPD severity. While smokers exhibited higher HO-1 levels, with respect to nonsmokers, thereby suggesting active defence mechanisms against smoke injury, patients with severe COPD may have an impaired defence against oxidative stress, although their smoking history was not different from control smokers. Altered expression of HO might also be a consequence of the disease. Since HO-2 is a constitutive enzyme and as a result is expressed in alveolar wall exclusively by type 2 pneumocytes, its increase may reflect the increase in number of these cells in the lung. In fact,lung cell proliferation may be induced by cigarette smoke and, in spite of destructive process, alveolar wall thickening may be detected in severe COPD 24, 25. The role of the constitutive form of HO in the lung remains undetermined. Based on the present authors' results, HO-2 could prevent excessive protein nitration favoured by iNOS products.

This is the first study on iNOS in peripheral lung tissue ofpatients with COPD. The expression of iNOS in human lung has been documented previously in normal airway epithelium, asthma, interstitial pneumonias and granulomatous lung diseases 7, 26–28. The current study's results on immunoreactivity for iNOS are consistent with previous reports that show a wide distribution of the enzyme in lung cells, including epithelial, vascular and inflammatory cells 29,30. By examining alveolar macrophages, the current study confirms previous findings of similar expression of iNOS in sputum macrophages in subjects with COPD and healthy subjects 8, 9. However, in the parenchymal compartment of the lung, the current study detected higher iNOS expression in type 2 pneumocytes of severe COPD. This compartmentalisation of iNOS in human lung is in agreement with observations made in rats 31. Rat type 2 pneumocytes expressed iNOS and released similar levels of NO as alveolar macrophages, although macrophages are generally considered the primary cellular mediator of host defence. The current results are consistent with the hypothesis that type 2 pneumocytes play a role as the first line of defence, together with epithelium and macrophages, against lung injury 32, and that their function may be critical for the lung tissue being destroyed by emphysema 33.

Excess of NO, produced by iNOS, in the presence of superoxide anion promotes nitration of tyrosine residues on proteins. Tyrosine nitration is heavily dependent on both theconcentration of NO and the molecular environment 19.Data on the role of iNOS products, as NO, are rather contradictory in the literature. Higher, lower or similar levels of exhaled NO have been shown in COPD subjects compared with controls and either positive or negative correlations of exhaled NO with FEV₁ have been reported 8, 34, 35. Nitrotyrosine-positive cells were more frequent in sputum of COPD patients in respect of healthy controls 8. The present authors were able to confirm the latter finding in the alveolar wall, but not in alveolar macrophages in severe COPD. In addition, the present authors' data suggest that nitration of proteins in lung tissue may have an adverse effect on lung function. The substantial differences between the two studies make it difficult to compare the results. In fact, the severity ofCOPD was milder in the study of Ichinose et al. 8 and the compartment of respiratory tract examined was different. Moreover, immunohistochemical technique is not suitable foraccurate quantitative evaluation of nitrated proteins.

The inverse relationship between HO-1 and iNOS in the lung might be explained by regulatory interaction between thetwo enzymes. HO is able to downregulate iNOS by degrading haem, a component of the enzyme, and by the inhibitory action of CO 36. A reduction of HO-1 may increase haem availability for synthesis and optimal activity of iNOS and enhance formation of pro-oxidant byproducts, that is, NO derived from iNOS activation. Conversely, a stabilising effect of NO on HO-1 messenger ribonucleic acid has been shown invitro 37, which would upregulate HO-1. In addition to NO, hypoxia and tumour necrosis factor-α may be present inCOPD as HO-1 inducers. However, the current results suggest that severe patients are unable to respond to these stimuli by increasing expression of HO-1.

A potential bias in this study performed on surgically resected specimens is that all control smokers had lung cancerwhereas nine out of 11 severe COPD had not, and the presence of cancer itself may have influenced the results. There are no studies on HO-1 expression and lung cancer, while upregulation of iNOS was reported in patients with primary lung cancer 38. If this is the case, the differences iniNOS expression would be even greater between two bettermatched groups of patients. Another potential bias is that, in LVRS, the surgeon selects the worst part of the lung, which might exaggerate pathological findings with respect to functional classification of the disease. No obvious differences were detected in the amount of tissue available for histological analysis (length of alveolar walls and number of alveolar macrophages) and in the counts of immunoreactivity betweenthe two severe COPD patients with lung cancer and the remaining nine who had LVRS. However, the need for alveolar wall integrity for reliable cell counts might have prevented the evaluation of some parts of tissue from LVRS not suitable for analysis.

In conclusion, the current study showed that two enzymes that are inducible in response to oxidative stress in chronic obstructive pulmonary disease localise in different cell types of peripheral lung. Immunoreactivity for haem oxygenase-1 is mainly detected in alveolar macrophages, whereas immunoreactivity for inducible nitric oxide synthase is localised in type 2 pneumocytes. The results suggest that these enzymes may have a different role in the lung defence and that imbalance between haem oxygenase-1 and inducible nitric oxide synthase may be linked with the development of severeimpairment in chronic obstructive pulmonary disease patients.