INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: immunohistochemistry; oxidative stress; macrophages; cigarette smoke; inflammation

Oxidative stress has been implicated in the pathogenesis of many diseases including chronic inflammatory lung disorders (1, 2). In asthma, cystic fibrosis, and bronchiectasis, raised levels of exhaled carbon monoxide (CO) have been detected, which were attributed to induction of the oxidative stress protein heme oxygenase-1 (HO-1) in the lung (3). Three genetic isoforms of heme oxygenase have been identified: the inducible form heme oxygenase-1 (HO-1), called heat shock protein (32) and the constitutive forms heme oxygenase-2 (HO-2) and heme oxygenase-3 (HO-3) (2). Heme oxygenases catalyze the degradation of heme to biliverdin, producing free iron and carbon monoxide (6). HO-2 and HO-3 are constitutive and, in animal models, are widely distributed throughout the body, with high concentration in the brain (7, 8). It has been suggested that HO-1 induction serves as a protection mechanism against oxidant-mediated cellular injury since heme degradation products have antioxidant activity (9). Several mediators are released during chronic inflammation including cytokines, reactive oxygen species (ROS), and nitric oxide (NO), which are able to induce HO-1 expression under experimental conditions (6). These lines of evidence suggest a potential induction and role of HO-1 in human diseases associated with oxidative stress. Measurements of exhaled CO in humans have been used as an indirect index of HO-1 induction in asthma (10). The two isoforms, HO-1 and HO-2, catalyze the same reaction and are expressed in bronchial biopsies of normal subjects and subjects with asthma (11). Thus, they may contribute to the levels of exhaled CO.

The role of heme oxygenases in chronic obstructive pulmonary disease (COPD) has not been studied. It is believed that the major site of airflow limitation in COPD is the peripheral airways. In this compartment of the lung, oxidative stress may occur as a consequence of cigarette smoking itself or of the associated inflammatory process. In this study we investigated the expression of HO-1 and HO-2 in peripheral lungs of 10 nonsmoking subjects, 6 healthy smokers, and 10 smokers with COPD.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Three groups of subjects undergoing lung resection for a solitary peripheral carcinoma were examined: 10 subjects with a smoking history of more than 15 pack-years and with both symptoms of chronic bronchitis (12) and fixed airway obstruction (subjects with COPD), 6 asymptomatic subjects with a smoking history of more than 15 pack-years and normal lung function, and 10 asymptomatic nonsmoking subjects with normal lung function. Pulmonary function tests were performed within the week before surgery. Fixed airway obstruction was defined as a FEV₁ less than 80% predicted (13), with a reversibility of less than 15% after inhalation of 200 µg of salbutamol. Subjects with COPD had had no exacerbations during the month preceding the study. All the subjects had been free of acute upper respiratory tract infections and none had received glucocorticoids or antibiotics within the month preceding surgery. They had negative skin tests for common allergen extracts and no past history of asthma or allergic rhinitis. The study conformed to the Declaration of Helsinki, and informed written consent was obtained for each subject.

Immunohistochemistry

Tissue blocks were taken from the subpleural parenchyma, avoiding areas involved by tumor, fixed in 4 % formaldehyde, and embedded in paraffin as described (14).

Mouse monoclonal antibody against HO-1 (anti-HO-1, Transduction Lab, Lexington, KY) and goat polyclonal antibody against HO-2 (C-20, sc-7697, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were used with the alkaline phosphatase-anti-alkaline phosphatase (APAAP) method (Dako Ltd., High Wycombe, UK). To expose the immunoreactive epitopes of cell markers, the sections in citrate buffer 5 mM at pH 6.0 were incubated in a microwave oven (Model M704; Philips, Eindhoven, The Netherlands). Negative controls for nonspecific binding incubated with secondary antibodies only were processed and revealed no signal. The slides were observed using light microscopy (Zeiss, Oberkochen, Germany) at 40× magnification in blind fashion. Immunostaining was examined in bronchial epithelium, pulmonary vessels, alveolar walls, and alveolar macrophages. To quantify HO-1 and HO-2 expression in alveolar macrophages, at least 20 high-power fields (hpf) of lung parenchyma were randomly selected for each section and at least 100 macrophages inside alveoli were evaluated. Alveolar macrophages were defined as mononuclear cells with well represented cytoplasm present in the alveolar spaces and not attached to the alveolar walls. Results were expressed as number of macrophages per hpf and as a percentage of HO-1- and HO-2-positive macrophages. The number of positively stained cells within the alveolar walls was computed using a light microscope (Leica DMLB; Leica, Cambridge, UK) connected to a video recorder linked to a computerized image system (Software: Casti Imaging SC Processing, Italy) as previously described (15).

Double immunostaining was performed to determine the cell type of staining positively for HO-1 and HO-2. CD68 (macrophages) and surfactant protein (SP)-B (type 2 pneumocytes)-positive cells were studied. Slides were double stained with antibodies anti-CD68 (goat polyclonal SC7083, Santa Cruz Biotechnology, Heidelberg, Germany) or anti-SP-B (polyclonal rabbit antibody, a gift from Prof. A. Baritussio, University of Padova, Italy) (16). HO-1 primary antibody was detected using the EnVision+ method (K4000, Dako) developed with diaminobenzidine. Slides were then incubated with anti-CD68 and bound antibody was labeled with biotinylated donkey anti-goat immunoglobulins (SC2042, Santa Cruz Biotechnology) followed by avidin-biotin complex and Fast Red staining (Dako). SP-B primary antibody was detected with peroxidase-conjugated swine anti-rabbit immunoglobulins (P399, Dako) developed with diaminobenzidine. The second primary antibody (HO-2) was then added and detected with biotinylated donkey anti-goat immunoglobulins (SC2042) followed by avidin-biotin complex and Fast Red staining (Dako). Control slides were obtained by performing double staining in which one of the primary antibodies was not added.

Statistical Analysis

Differences between groups were analyzed using the analysis of variance (ANOVA) for clinical data, and the Kruskall-Wallis test for histological data. The Mann-Whitney U test was carried out after the Kruskall-Wallis test when appropriate. Correlation coefficients were calculated using Spearman's rank method. At least three replicate measurements of immunostained slides were performed by the same observer in 10 randomly selected slides to assess the intraobserver reproducibility (17). The intraclass correlation coefficient was 0.90. Probability values of p  0.05 were accepted as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Table 1 shows the characteristics of the subjects examined. The three groups of subjects were similar with regard to age and to Pa_O2 and Pa_CO2 values. There was no significant difference in the smoking history between subjects with COPD and with normal lung function. As expected from the selection criteria, subjects with COPD had a significantly lower value of FEV₁ (percentage of predicted) and FEV₁/FVC ratio as compared with smokers with normal lung function (p < 0.05) and nonsmokers (p < 0.05). In subjects with COPD, whose FEV₁ ranged from 54 to 79% predicted, the average response to bronchodilator was 5%.

HO-1 immunostaining was observed mainly in cells of alveolar spaces (Figure 1A, 1D), whereas it was negative in bronchial epithelium (Figure 1B) and vessels (Figure 1E). By immunohistochemistry, 88% (interquartile range 82-92) of the cells counted as alveolar macrophages were CD68 positive (Figure 2B). The majority of HO-1-positive cells in alveolar spaces were CD68 positive (median 89% [83-98]) (Figure 2A). In five of the 26 subjects, occasional HO-1-positive cells were detected in alveolar wall or interstitial tissue. The median (interquartile range) numbers of alveolar macrophages counted were 114 (103-148) in nonsmokers, 112 (106-171) in smokers with normal lung function, and 160 (123-269) in smokers with COPD. Smokers with COPD showed an increased number of alveolar macrophages per hpf as compared with nonsmokers (p < 0.05) (Table 2). Analyzing the three groups together, there was an inverse correlation between the number of alveolar macrophages and FEV₁/FVC (Rho = 0.50, p < 0.02). The percentages of HO-1-positive alveolar macrophages was significantly increased in subjects with COPD (p < 0.02) as compared with nonsmoking subjects, whereas no difference was observed between smokers with or without COPD. No differences in the expression of HO-1 on alveolar macrophages were observed when we compared current smokers (36.4 [10.9-54.8]% of positive macrophages; n = 8) with ex-smokers (33.6 [17-44.9]%; n = 8). No correlations were found between expression of HO-1 and clinical or functional parameters of the subjects.

View larger version (133K): [in this window] [in a new window]   Figure 1.   Immunohistochemis-try study of human lung surgical specimens. Formalin-fixed paraffin-embedded sections were processed for immunohistochemistry using anti-human HO-1 and HO-2 antibodies and the alkaline phosphatase-anti-alkaline phosphatase (APAAP) method developed with Fast Red. All sections are counterstained with hematoxylin. (A) Alveolar spaces stained for HO-1 in nonsmoking subject. (B) Alveolar space and bronchiolar epithelium stained for HO-1 in a subject with COPD. (C  ) Alveolar wall stained for HO-2 in a nonsmoking subject. (D) Alveolar spaces stained for HO-1 in a subject with COPD. (E  ) Alveolar space and a vessel stained for HO-1 in a subject with COPD. (F ) Alveolar wall stained for HO-2 in a subject with COPD. Original magnification: A, C, D, E, F ×40; B ×25.

View larger version (141K): [in this window] [in a new window]   Figure 2.   Human lung sections that have been doubly stained for heme oxygenase (HO)-1 as brown staining and CD68 (macrophages) as red staining (A); or for HO-2 as red staining and surfactant protein (SP)-B ( Type 2 cells) as brown staining (D). There is double staining for CD68 and HO-1 (A, arrows), and for HO-2 and SP-B (D, arrows). In control sections of double staining one of the primary antibodies was not added and single staining was detected: HO-1+/CD68 (B), HO-1/CD68+ (C ), HO-2+/ SP-B (E ), HO-2/SP-B+ (F ). Original magnification: ×40.

Lower percentages of alveolar macrophages exhibited positive staining for HO-2 and no significant differences were observed between the three groups of subjects (Table 2). Immunostaining for HO-2 was present in the alveolar wall, in smooth muscle and in cells of adventitia of pulmonary arteries, and in few cells of bronchial wall (Figure 1C and 1F). The number of HO-2-positive cells in the alveolar wall was significantly increased in smokers with and without COPD as compared with nonsmokers (Table 3) and was inversely related to the age at which subjects started smoking (Rho = 0.73, p < 0.01). When the three groups were analyzed together, HO-2-positive cells in alveolar wall correlated inversely with FEV₁/FVC (Rho = 0.58, p < 0.005). Double immunostaining indicated that the majority of HO-2-positive cells in the alveolar wall were type 2 pneumocytes (median 95% [92-97], whereas the remaining were CD68 positive (Figure 2D).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study we demonstrated that inducible HO-1 and constitutive HO-2 are detectable in human lung tissue, but their cell localization is different. HO-1 predominates over HO-2 in alveolar macrophages, whereas HO-2 predominates in lung parenchyma. HO-1 and HO-2 expression is increased in smokers, but no significant differences were observed between healthy smokers and patients with COPD. This is the first study that examined the expression of heme oxygenases 1 and 2 in situ in human peripheral lung tissue.

HO-1 is induced by heme in physiological conditions and it is expressed at high concentrations in the spleen and in the liver, where senescent erythrocytes are sequestered and destroyed. HO-1 may be induced by a variety of nonheme products, such as reactive oxygen species (ROS), endotoxin, inflammatory cytokines, and nitric oxide (6). The functional significance of HO-1 induction is not well understood, as HO-1 activity can result in either cell protection or cell injury depending on the experimental setting (18). This enzyme is believed to play an important role in maintaining cellular homeostasis in response to oxidant injury. Rats pretreated with hemoglobin, a potent inducer of HO-1, are less susceptible to oxidant-mediated endotoxic shock and to hyperoxic lung injury (19, 20). Taken together, these observations suggest that HO-1 may have an important role in protecting tissue from oxidative stress. The lung is a major target organ for injury by ROS or endotoxin and the cell types that have the potential to express the inducible HO-1 have been studied in vitro or in animal models in vivo. Scant basal levels of HO-1 were observed by immunohistochemistry in alveolar and bronchiolar epithelium of rats in vivo (21, 22), whereas HO-1 labeling of unstimulated lungs was detected in macrophages only (23). A suggestion that macrophages are a source of HO-1 protein in humans was made by the study of Horvath and coworkers (10) in which expression of HO-1 was detected in adherent cells obtained from induced sputum by Western blot analysis. This was confirmed by Lim and coworkers (11) in bronchoalveolar lavage macrophages from subjects with asthma and normal subjects by immunohistochemistry. Our data are consistent with the results obtained in animal and human models and confirm that alveolar macrophages are the major source of HO-1 in peripheral lung. In contrast with the findings in bronchial biopsies (11) no immunoreactivity for either HO-1 or HO-2 was detected in bronchiolar epithelium. It is possible that the expression of these enzymes is compartmentalized on epithelial cells of central airways. There are, however, differences between the studies in the processing of biopsies and lung specimens and in the type of primary antibodies, which may lead to different immunoreactivity. In addition, the subjects studied by Lim and coworkers (11) had no lung cancer and likely were much younger than ours.

The constitutive isoform HO-2 is found in the central and peripheral nervous system and is thought to generate carbon monoxide, which functions as an intra- and extracellular neurotransmitter. In the respiratory system, immunoreactivity for HO-2 was detected in ganglion nerve cell bodies localized to the trachea and bronchi in humans (24). In peripheral airways and parenchyma, where parasympathetic ganglia are not represented, we observed that HO-2 immunoreactivity was widely distributed in alveolar spaces and lung tissue. The observation that the noninducible isoform HO-2 is increased in lung parenchyma of smokers might indicate that recruitment or proliferation of cells that express constitutive HO-2 occurs in this lung compartment as a consequence of cigarette smoking (25, 26). The role of HO-2 in COPD has not been investigated. Experiments in mice suggest that HO-2 provides a protective function in antioxidant defense over the effects of HO-1 (27). Upregulation of the HO-2 gene by glucocorticoids has been observed in neonatal rat brain (28) and dexamethasone enhanced HO-2 protein in human primary epithelial cells (29). Whether the increase in HO-2 expression in subjects with poorer lung function represents an adverse role or a more active lung defense against injury might be relevant for the action of steroids in COPD therapy.

The reason ex-smokers still exhibited increased expression of HO-1 remains undetermined. It has been reported that pathological changes in the peripheral lung may occur in smokers despite normal lung function (30) and that airway inflammation may persist after smoking cessation (31, 32). The median interval since our ex-smokers stopped smoking was 6.5 yr (IQR 3.5-11.5), that is, intermediate between those in the ex-smokers studied by Mullen and coworkers (31), median 3.5 yr, and Turato and coworkers (32), median 13 yr. It is therefore conceivable that an underlining inflammation may be present in our ex-smokers and that products of inflammatory cells, such as oxygen radicals and cytokines, may upregulate HO-1 even in the absence of the direct stimulus of cigarette smoke.

Although the present data did not show differences in the percentages of macrophages expressing HO between healthy smokers and patients with COPD, we cannot exclude a role for HO-1 and HO-2 in smoking-induced pulmonary diseases for several reasons. In fact, subjects who started smoking at a younger age or had poorer lung function exhibited more HO-2-positive cells in the alveolar wall. In addition, the number of macrophages in lung sections from patients with COPD appeared to be increased when compared with smokers with normal lung function. Therefore, the overall amount of induced HO-1 in the alveolar spaces of patients with the disease could be higher. We need to be cautious with this interpretation because the degree of shrinking of lung tissue blocks may vary among subjects and interfere with a precise quantification of cells per hpf. The induction of HO-1 after appropriate stimulus has a kinetic of hours (21). Cross-sectional studies of patients in a stable condition might fail to detect a transient increase in the enzyme. Because elevated HO-1 gene expression was observed in animal models during chronic hypoxia (33), HO-1 might play a role in patients with severe COPD who have reduced arterial oxygen at variance with patients with relatively mild COPD that we studied. Recent data suggest that polymorphism in the HO-1 gene promoter may reduce the inducibility of HO-1 by reactive oxygen species, thereby resulting in different susceptibility to damage by cigarette smoke among individuals (34). We indeed observed a different intensity of HO-1 staining among alveolar macrophages. However, the immunohistochemical technique is not suitable for quantitative determination of the enzyme.

In conclusion, we showed that both inducible HO-1 and constitutive HO-2 are detectable in human lung, but exhibit a different distribution. In smokers, HO-1 immunoreactivity is prevalent in alveolar spaces, whereas HO-2 immunoreactivity is prevalent in alveolar walls. The results suggest that oxidative stress due to cigarette smoke increases the number of lung cells expressing HO-1 and HO-2, whereas the role of this enzyme in the pathogenesis of COPD remains uncertain.