INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Asthma is characterized by a chronic inflammatory disease of the airways that is associated with increased expression of inflammatory mediators, including proinflammatory cytokines such as tumor necrosis factor alpha (TNF-), interleukins (IL) such as IL-1, IL-4, and IL-5, prostaglandins, reactive oxygen species (ROS), and inducible nitric oxide synthase (1). Oxidative stress has been implicated in the pathogenesis of asthma, and increased release of ROS such as superoxide radicals (O₂) and hydrogen peroxide (H₂O₂) has been reported in exhaled condensates (2, 3) and from circulating monocytes and neutrophils, bronchoalveolar lavage cells of patients with asthma (4, 5). ROS are generated from normal cellular respiration and aerobic metabolism, from a respiratory burst of inflammatory cells, or from exogenous oxidants, and can cause cellular damage by oxidizing nucleic acids, proteins, and membrane lipids (6). Oxidative stress can activate the transcription factor, nuclear factor kappa-B (NF-B) (7, 8), which is increased in expression in epithelial cells of patients with asthma (9). Increased NF-B expression and DNA binding in asthma may underlie the increased expression of several inflammatory proteins such as iNOS, GM-CSF, eotaxin, and RANTES in asthmatic airway epithelium (10).

Heme oxygenase (HO) catalyzes the rate-limiting step in the oxidative degradation of heme to bilirubin, and two isoforms, HO-1 and HO-2, of the enzyme have been found (14). Although HO-2 is constitutively expressed, HO-1 is inducible by many agents that lead to oxidant stress, including hydrogen peroxide and heavy metals (15, 16). It has been proposed that HO-1 plays an important role in cellular protection against oxidant injury (17). Previous studies have indicated that HO activity may be increased in the airways of patients with asthma, as levels of exhaled carbon monoxide (CO), a product of heme metabolism, have been reported (18, 19). Expression of HO-1 in macrophages obtained by induced sputum were found to be elevated in patients with asthma (19). However, the expression of the two isoforms of HO in the airway submucosa of patients with asthma is not yet known.

We therefore investigated the expression of HO-1 and HO-2 in bronchial biopsies obtained from patients with mild asthma compared with that of subjects without asthma. We hypothesized that the expression of HO-1, in contrast to that of HO-2, would be enhanced in asthma. In addition, because patients with asthma on inhaled corticosteroid therapy demonstrate low to normal values of CO in exhaled air (18, 19), we also examined the effect of inhaled corticosteroid therapy on the expression of HO-1 and HO-2 in patients with asthma in a separate study.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Ten and 16 (total 26) subjects with mild stable asthma receiving treatment with only the inhaled ₂-adrenergic agonist aerosol, albuterol, for intermittent relief of wheeze were recruited for two separate protocols (see below). In the first protocol examining HO expression in mild asthma, 10 nonatopic control subjects without asthma were also recruited for comparison. All patients with asthma demonstrated a > 15% improvement in forced expiratory volume (FEV₁) following 200 µg of albuterol and airway hyperresponsiveness to methacholine with a provocative concentration of methacholine producing a 20% fall in FEV₁ (PC₂₀) of < 4 mg/ml. All patients were atopic as defined by two or more positive skin prick tests to common allergens. None of the patients with asthma had received oral or inhaled corticosteroids for the preceding 12 mo, or any other treatment apart from inhaled ₂-agonists. Current smokers or ex-smokers of more than five pack-years and patients with FEV₁ less than 80% predicted were excluded. All controls had negative skin prick test, no airway hyperresponsiveness to methacholine (> 64 mg/ml), had normal lung function, and were nonsmokers. The study was approved by the Royal Brompton Hospital Ethics Committee, and all patients gave their informed consent to participate.

Protocols

Protocol 1: Expression of HO-1 and HO-2 in asthma. Fiberoptic bronchoscopy was performed in 10 patients with mild asthma and 10 nonatopic volunteers without asthma, and the expression of the HO isoenzymes in the bronchial biopsies was studied by immunohistochemistry and Western blotting. We also examined the expression of these isoenzymes in alveolar macrophages obtained by bronchoalveolar lavage.

Protocol 2: Effect of inhaled corticosteroids on expression of HO-1 and HO-2. To determine the effect of inhaled glucocorticoid treatment on the expression of HO and the production of CO, we performed a double-blind, randomized, parallel, and placebo-controlled study in 16 untreated patients with mild asthma. Spirometry, airway responsiveness, and exhaled NO and CO measurements were taken at the start of the study, followed by fiberoptic bronchoscopy. The patients were then randomized to receive either inhaled budesonide (800 µg twice daily from a multiple dose dry powder inhaler, Turbuhaler) or a matched placebo for 4 wk. There were eight patients with asthma in each group. The measurements, together with fiberoptic bronchoscopy, were repeated at the end of the treatment period. Biopsies were studied for the expression of HO isoenzymes using immunofluorescence microscopy.

Lung Function and Challenge Tests

Baseline spirometric parameters were recorded from the best of three attempts using a dry wedge spirometer (Vitalograph, Buckingham, UK). Spirometric measurements and provocation tests were performed before the beginning of each treatment period and at Day 26 of the treatment period. All patients had been previously instructed to abstain from using inhaled ₂-agonists and from caffeine-containing foods for 12 h before the test. After completing a baseline spirometric measurement, a standardized bronchial provocation protocol was performed using nebulized buffered isotonic methacholine solution. After an initial nebulized saline challenge, doubling concentrations of methacholine starting from 0.06 mg/ml were administered via a dosimeter (MEFAR, Bovezzo, Italy). Spirometric values were recorded 2 min following administration of each concentration until a > 20% fall in FEV₁ was achieved. The PC₂₀ value was obtained by linear interpolation of each concentration-response curve.

Nitric Oxide Measurements

Exhaled NO was measured by a chemiluminescence analyzer (Model LR2000; Logan Research, Rochester, UK), with sensitivity from 1 ppb to 100 ppm of NO, accuracy ± 0.5 ppb and response time of < 2 s to 90% of full scale. In addition, the analyzer also measured CO₂, expiration flow, and pressure, and the exhaled volume in real time. The analyzer was fitted with a biofeedback display unit to provide visual guidance for the subject to maintain pressure and exhalation flow within a certain range (3.0 ± 0.4 mm Hg and 5-6 L/min for end-exhaled NO measurements). The sampling rate was 250 mL/min for all measurements. The analyzer was calibrated daily using NO-free certified compressed air to set absolute zero and then a certified concentration of NO in nitrogen of 90 and 500 ppb (BOC Special Gases; Surrey Research Park, Guildford, UK) and certified 5% CO₂ (BOC). Ambient air NO levels were recorded and the absolute zero was adjusted prior to all measurements. Subjects exhaling slowly from total lung capacity over 20-30 s with exhalation flow of 5-6 L/min NO was sampled from a side-arm attached to the mouthpiece. The mean value of the last 100 measurements, acquired with a 0.04 s interval, was taken from the point corresponding to the plateau of the end-exhaled measurement (CO₂ reading 5-6%), and representing the lower respiratory tract sample. Results of the analyses were computed and graphically displayed on a plot of NO and CO₂ concentrations, pressure, and flow against time.

Carbon Monoxide Measurements

Exhaled CO was measured by a modified electrochemical analyzer (EC50-MICRO Smokerlyzer CO monitor; Bedfont Scientific Ltd, Herts, UK) sensitive to CO from 0 to 1,000 ppm (by volume) with a resolution of 1 ppm. Prior to the start of the study, the instrument was calibrated with a gaseous mixture containing 50 ppm of CO in accordance with the manufacturer's instructions. Ambient CO levels in the room were recorded prior to the subject readings. All subjects were then asked to exhale fully, inhale deeply, and hold their breath for 20 s before exhaling rapidly into a disposable mouthpiece as previously described (19). Three successive recordings were made and the mean value was obtained.

Fiberoptic Bronchoscopy

Subjects attended the bronchoscopy suite at 8.30 A.M. after having fasted from midnight and were pretreated with atropine (0.6 mg intravenously) and midazolam (5-10 mg intravenously). Oxygen (3 L/min) was administered via nasal prongs throughout the procedure and oxygen saturation was monitored with a digital oximeter. Using local anesthesia with lidocaine (4%) to the upper airways and larynx, a fiberoptic bronchoscope (Olympus BF10 Key-Med, Herts, UK) was passed through the nasal passages into the trachea. Bronchoalveolar lavage was performed from the right middle lobe using warmed 0.9% NaCl with four successive aliquots of 60 ml. Multiple mucosal biopsies were taken from segmental and subsegmental bronchi.

Bronchial mucosal biopsies were immediately placed in Optimal Cutting Temperature (OCT) embedding media, then snap-frozen in isopentane precooled with liquid nitrogen and stored at 70° C. Sections (6 µm) were cut on a cryostat and placed on poly-L-lysine-coated microscope slides (Sigma, Poole, UK). The slides were air-dried for 30 min then wrapped in aluminum foil and stored at 70° C prior to immunostaining. Cytospin preparations of bronchoalveolar lavage cells were performed on poly-L-lysine microscope slides and the slides were air-dried and stored at 70° C wrapped in aluminium foil until further use.

Immunostaining of Eosinophils

Frozen sections were allowed to warm to room temperature and fixed in acetone at 4° C for 10 min, and air-dried for 10 min. Endogenous peroxidase activity was blocked by incubating the slides in 0.3% (wt/ vol) hydrogen peroxide in 70% methanol for 30 min. To block for nonspecific binding sites, the sections were incubated with horse serum. To stain for eosinophils, a mouse monoclonal anti-human major basic protein antibody, BMK-13 (Monosan, Uden, Netherlands; 1:50, 1 h at room temperature), was used. Following the primary antibody, a biotinylated mouse immunoglobulin (1:50) followed by avidin- biotin complex (Vectra Stain kit; Vectra Laboratories, Peterborough, UK) were added. Diaminobenzidine was used for 5 mins and the slides were counterstained with hematoxylin and mounted on mounting medium (DPX; Raymond Lamb, London, UK).

Counts of positive cells were made on all sections. The number of positive cells was expressed as the number per field. At least four fields at ×400 magnification were examined. One field was defined as a length of 175 µm and for the subepithelium, an area of 175 × 175 µm². All counts were made by two experienced observers unaware of the origin of the sections.

Immunohistochemistry for HO-1 and HO-2

Immunohistochemistry for HO-1 and HO-2 was performed by two variant methods for the two protocols.

For Protocol 1 (comparative study), the biopsies were fixed in acetone. Then the biopsies were washed with phosphate-buffered saline (PBS) containing 3% hydrogen peroxide with 0.02% sodium peroxide. Immunostaining procedures were performed using the Vectra Stain kit (Vectra Laboratories). Nonspecific labeling was blocked by coating with normal goat serum for 20 min at room temperature. After washing in PBS, the tissues were incubated with a rabbit polyclonal anti-HO-1 antibody (Affiniti; Mamhead, Exeter, UK, diluted 1:50 in the preincubation solution) or with a rabbit polyclonal anti-HO-2 antibody (Affiniti; diluted 1:100) at room temperature for 1 h. After incubation and repeated washing steps with PBS, the sections were subsequently incubated with biotinylated goat anti-rabbit IgG (1:200) for 1 h at room temperature. The slides were washed and then avidin-biotin complex was applied for 30 min. Secondary antiserum was detected with 3,3' diaminobenzidine (Sigma, Poole Dorset, UK). Normal rabbit IgG was used as negative control. Sections were counterstained with hematoxylin and mounted with mounting medium (DPX). The immunoreactivity of the HO isoenzymes was graded semiquantitatively from 0 to 4, with 0 representing no staining, grade 1 representing focal staining, and grades 2, 3, and 4 representing diffuse weak, moderate, and strong staining, respectively. All analyses, including immunohistochemical grading, were performed in a blind fashion by two investigators.

Cytospins of BAL cells were fixed in chilled acetone, followed by immunostaining with anti-HO-1 and anti-HO-2 antibodies as described above.

For Protocol 2 (effect of inhaled steroids), we used a different method of fixation. The biopsies were fixed with 4% phosphate-buffered paraformaldehyde, washed repeatedly with 0.1 M phosphate buffer, and stored overnight in cold (4° C) 18% sucrose containing 0.1 M phosphate buffer. Tissues were mounted on filter paper in OCT compound and frozen in liquid nitrogen. For immunohistochemistry the fixed biopsies were cut in a cryostat to 8-µm sections, mounted on gelatin-coated glass slides, and air-dried for 2 h. After rinsing in PBS, nonspecific labeling was blocked by coating with preincubation serum (0.1 M phosphate buffer containing 1% bovine serum albumin and 10% normal swine serum) for 1 h at room temperature. After washing in PBS the tissues were incubated with a rabbit polyclonal anti-HO-1 antibody (Affiniti; diluted 1:400 in the preincubation solution) at 4° C overnight or with a rabbit polyclonal anti-HO-2 antibody (StressGen, Victoria, BC, Canada; 1:1,500). After overnight incubation and repeated washing steps with PBS, the sections were subsequently incubated with biotinylated goat anti-rabbit IgG (Amersham, Braunschweig, Germany; 1:200) for 1 h at room temperature. Secondary antiserum was detected with a Streptavidin-Texas Red conjugate (Amersham; 1:50). Specificity of the antibody reaction was verified in parallel sections that were incubated either with the primary antiserum that had been preabsorbed with recombinant HO-1 or HO-2 protein (StressGen; 20 µg protein/ml diluted in antiserum) or only the secondary antibodies. Slides were coverslipped in carbonate-buffered glycerol (pH 8;6) and viewed using epifluorescence microscopy.

Double-staining Immunohistochemistry

Biopsies obtained from Protocol 1 were also subjected to double immunostaining in order to determine the submucosal cell type staining positively for HO-1 or HO-2. We studied CD68 (macrophage) and CD3 (T cell) positive cells. Slides were stained with the anti-HO-1 or anti-HO-2 antibody, as described for Protocol 1. Secondary antiserum was detected using diaminobenzidine. Sections were rinsed in tris-buffered saline (TBS), and incubated with normal horse serum from Vectra Stain kit (Vectra Laboratories) for 20 min. Slides were then incubated with anti-CD68 (Dako, Ely, UK; 1:50) or anti-CD3 (Dako; 1:50) monoclonal antibodies for 1 h. After further washing with TBS, slides were incubated with biotinylated horse anti-mouse immunoglobulin G (IgG) at 1:300 for 45 min, followed by avidin-biotin complex (Vectra Laboratories; 1:300, 30 min). Slides were developed with Fast Red (Sigma) for 5-7 min, and mounted using an aqueous mountant.

Western Blot Analysis for HO-1 Expression

Whole-cell extracts were prepared from bronchial biopsies. In brief, bronchial biopsies were resuspended in 500 µl of reporter lysis buffer 1 × (Promega, Southampton, UK), frozen to 70° C, and rewarmed. After centrifugation at 12,000 × g for 10 min at 4° C, protein concentration was measured by the Bradford method according to the directions of the manufacturer (Bio-Rad, Hemel Hempstead, UK). At least 10 µg/lane of whole-cell extracts was subjected to a 10% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, and transferred to nitrocellulose filter (Amersham Pharmacia Biotech, Amersham, UK) by blotting. Filters were blocked for 1 h at room temperature in PBS, 0.2% Tween 20, 5% nonfat dry milk. The filters were then incubated with rabbit anti-human HO-1 (Affiniti) for 11 h at room temperature in PBS with 0.2% Tween 20, and 5% nonfat dry milk. Filters were washed three times in PBS with 0.2% Tween 20 and after incubating with anti-rabbit antibody conjugated to horseradish peroxidase (Dako) in PBS, 0.2% Tween 20 and 5% nonfat dry milk. After further three washes in PBS and 0.2% Tween 20, visualization of the immunocomplexes was performed using the Enhanced chemiluminescence (ECL) kit as recommended by the manufacturer (Amersham Pharmacia Biotech, UK). Using the same blots, the expression of the housekeeping protein, -actin, was measured using an anti--actin antibody (Santa Cruz Biotech, Santa Cruz, CA), and the results were expressed as a ratio of the expression of HO-1 to that of -actin.

Data Analysis

Data are reported as mean ± SEM. Comparison of data between normal subjects and subjects with asthma were performed using unpaired t test, while the effect of placebo or corticosteroid treatment was determined using a Student's paired t test. A p value less than 0.05 was taken as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Localization of HO-1 and HO-2 Expression

In Protocol 1, HO-1 and HO-2 were expressed in the airways of both patients with mild asthma as well as control subjects. Expression was observed particularly in the airway epithelium (Figure 1). There were no significant differences in the amount of staining between patients with mild asthma and control subjects; for HO-1, the scoring was 1.82 ± 0.23 and 2.20 ± 0.37 in normal subjects and patients with asthma, respectively, and for HO-2, 2.02 ± 0.18 and 1.75 ± 0.25. There was significant staining for HO-1 and HO-2 in alveolar macrophages obtained from normal subjects and patients with asthma to an equal extent (Figures 1g and 1h). In control sections or cytospins in which the primary antibody was omitted, there was no positive staining (Figures 1e and 1i). Double staining revealed that HO-1 and HO-2 were expressed in CD68⁺ macrophages located in the subepithelium, but not in CD3⁺ T cells (Figure 2).

View larger version (114K): [in this window] [in a new window]   Figure 1.   High-power photomicrographs of bronchial biopsy sections that have been stained for heme oxygenase-1 (HO-1; a, b, c ) or for heme oxygenase-2 (HO-2, d, f  ) from normal subjects (a, b, d ) or subjects with asthma (c, f  ). There is intense brown staining for both heme oxygenases in the airway epithelium and some submucosal cells, with no apparent differences between normal subjects and subjects with asthma. (e) A control section with the primary antibody not added. (g, h, and i ) Brown staining of HO-1 (g) and HO-2 (h) in alveolar macrophages from an asthmatic subject. (i ) A control cytospin with the primary antibody not added. Scale bars represent 50 µm.

View larger version (131K): [in this window] [in a new window]   Figure 2.   High-power photomicrographs of bronchial biopsy sections that have been doubly-stained for heme oxygenase-1 (HO-1; a, b) or for heme oxygenase-2 (HO-2; c, d ) as brown staining, and for either CD68+ macrophages (a, c) or CD3+ T cells (b, d ) as red staining. The epithelium stains positive brown for HO-1 and HO-2. There is double staining for CD68+ cells for both HO-1 and HO-2 (arrows) with both brown and red staining in these cells, but not for CD3+ T cells, which demonstrates only red staining. Scale bars represent 50 µm.

In Protocol 2, in biopsies obtained prior to treatment with corticosteroids, immunoreactivity for HO-1 was localized to the respiratory epithelium and to single cells in the mucosal layer (Figure 3). The immunoreactivity was intense and of a nongranular type. No particular association to the cellular membrane or to specific intracellular organelle pattern was observed. The immunoreactive single cells underneath the respiratory epithelium displayed an intense staining. Due to their morphology, size, and distribution, these cells most likely represent macrophages. Despite the differences in immunohistochemical techniques used, the distribution of HO isoenzymes was similar in both protocols. Immunofluorescence was abolished after incubation with preabsorbed serum or when primary or secondary antibodies were omitted (Figure 3).

View larger version (100K): [in this window] [in a new window]   Figure 3.   Localization of heme oxygenase-1 (b, c, f   ) and -2 (a, e) immunoreactivity in biopsies from subjects with asthma before (a, b, c) and after (e, f   ) steroid treatment using immunofluorescence. HO-1 immunoreactivity is localized to bronchial epithelium and subepithelial cells (probably macrophages) before (b, c) and after (f   ) treatment. HO-2 reactivity is also displayed to epithelium and subepithelial cells before (a) and after (e) treatment. Control sections stained for HO-1 (d ) and HO-2 (g) in biopsies before steroid treatment in the absence of the antigen peptide do not show specific immunoreactivity. A high-power photomicrograph is shown in the inset (c). Scale bar represents 50 µm.

Western blot analysis of the bronchial biopsies showed that there was significant expression of HO-1 in normal subjects and subjects with asthma with no differences between the groups (Figure 4).

View larger version (24K): [in this window] [in a new window]   Figure 4.   Immunodetection by Western analysis of heme oxygenase-1 from bronchial biopsies of three normal subjects and three subjects with asthma, demonstrated as a specific 32-kD band (arrow). The -actin band is also shown.

Effect of Inhaled Steroids on Inflammatory Markers

Treatment with inhaled budesonide led to a significant reduction in exhaled NO compared with the effect of placebo (6.19 ± 6.72 versus 16.28 ± 2.69 ppb NO; p < 0.05) (Figure 5), but there were no significant differences in levels of exhaled CO following steroid treatment (3.28 ± 0.39 versus 3.17 ± 0.36 ppm CO). There was a decrease in the number of mucosal eosinophils following inhaled steroid (from 2.03 ± 0.50 to 0.50 ± 0.33 eosinophils/HPF, p < 0.05). There was a decrease in airway hyperresponsiveness following steroid treatment, as shown by an increase in log PC₂₀ FEV₁ of  0.41 ± 0.21 to 0.55 ± 0.37 (p < 0.05).

View larger version (25K): [in this window] [in a new window]   Figure 5.   Effect of inhaled budesonide for 1 mo on eosinophil counts in the airway submucosa measured as major basic protein (MBP+) positive cells (A), exhaled nitric oxide (B), exhaled carbon monoxide, (C ) and bronchial responsiveness (D) measured as log PC20. Data shown as mean ± SEM for eight subjects with asthma. Significance relates to comparison before and after treatment with budesonide. *p < 0.05; **p < 0.01.

Effect of Inhaled Steroids on Expression of HO-1 and HO-2

Biopsies taken from patients with asthma before and after treatment with corticosteroids displayed a similar pattern of distribution and intensity of HO-1 immunoreactivity with strong staining of respiratory epithelium and macrophages (Figure 3). Similarly, immunohistochemical staining for HO-2 revealed the same degree of specific staining in the bronchial epithelium in patients before and after inhaled steroid treatment (Figure 3).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have shown that both isoenzymes of heme oxygenase, HO-1 and HO-2, are expressed in the airways of normal subjects and subjects with asthma in equal amounts, and that their expression in asthma airways is not modulated by pretreatment with corticosteroids, despite evidence for an antiinflammatory effect. The expression of both isoenzymes was observed in resident cells, particularly in the airway epithelium, and also in CD68⁺ macrophages. The high basal expression of these isoenzymes was not accompanied by an increase in exhaled CO measurements of patients with asthma, although exhaled NO levels were significantly elevated.

Studies in cells in vitro have demonstrated that HO-2 is the constitutive form of HO, whereas HO-1 is the inducible form that is not usually expressed in unstimulated cells (14). HO-1 expression can be induced by several stimuli, particularly those that increase oxidant stress, including ultraviolet irridiation, superoxide and hydroxyl radicals, hydrogen peroxide, and sodium arsenite (15, 16, 20). In addition, lipopolysaccharide and hyperoxia can up-regulate the activity of HO-1 (23, 24). The high endogenous expression of HO-1 present in the airways, particularly in the epithelium, may result from the oxidant burden to the airways. A particularly high expression in the lining airway epithelium supports this suggestion because the epithelium is exposed to daily oxidant stresses such as environmental pollutants, including ozone, nitrogen dioxide, and particulates, which have potent oxidizing properties (25, 26). Although there is evidence for increased oxidative stress in asthma (2, 4, 5), we found no difference in the expression of HO-1 and HO-2 in bronchial biopsies using immunohistochemical methods and Western blot analysis. It is possible that the expression of these enzymes was already maximal, or that our techniques are not sensitive enough to quantify differences at these high levels of basal expression.

The functional significance of HO-1 induction following oxidative stress is not known. HO-1 may be provide cellular protection against oxidant injury (17). In an A549 epithelial cell line, transfection with HO-1 was cytoprotective against oxidant injury (27), and HO-1-deficient cells were more sensitive to cytotoxic injury following hemin and cadmium exposure (28). Basal expression of HO-1 has been described in hamster fibroblasts and this constitutive HO-1 expression was associated with resistance to hyperoxia (29). Similarly, there is basal expression of HO-1 in neonatal rat lung, which could be further increased following hyperoxia (30).

A glucocorticoid response element (GRE) is present in the HO-1 promoter, and gene expression of HO-1 can be inhibited by corticosteroids in vitro (31, 32). In contrast, a GRE identified in the HO-2 promoter results in increased gene and protein expression following dexamethasone treatment (33). Despite this, our in vivo studies indicate that neither isoenzyme is modulated, at least at the protein level, by treatment with corticosteroids that have demonstrable antiinflammatory effects in patients with asthma. In contrast, inhaled corticosteroids decrease the expression of the inducible form of nitric oxide synthase in the airways epithelium of patients with asthma (34), an effect that is reflected by a significant decrease in the raised levels of exhaled NO observed in the present study. Interestingly, HO-1 can be induced by NO in vitro (35), an effect that has been attributed to an interaction of NO with superoxide-producing peroxynitrite, which then stimulates HO-1 (36). A reduction in NO production and in proinflammatory cytokine expression should lead to a reduction in HO-1 activity, but there was neither a decrease in HO-1 immunoreactivity nor a decrease in the levels of exhaled CO, indicating that changes in NO may not modulate HO-1 expression.

The levels of exhaled CO measured in our patients with asthma in the inhaled corticosteroid study were not significantly different from those reported in normal subjects (19), which contrasts with the significantly raised levels observed in previous reports (18, 19). Of eight patients in the corticosteroid study, two had levels above the range found in normal individuals. The patients that we studied had mild asthma and were not usually symptomatic. It is possible that higher levels of exhaled CO are seen in patients with more severe asthma. We found no significant correlation between the degree of immunostaining with either the anti-HO-1 or anti-HO-2 antibody and the level of exhaled CO. We also explored the possibility that this may reflect the degree of inflammation within the airways but found no significant relationship between the submucosal eosinophil counts in the bronchial biopsies and exhaled CO. The kinetics and source of CO in exhaled breath are not yet elucidated. However, it is possible that the high level of expression of both HO-1 and HO-2 in both normal subjects and patients with asthma we observed is already maximal.

In summary, this is the first study in human subjects that examines the expression of HO isoenzymes in the airway submucosa. We found a large expression of the HO-1 and HO-2 isoenzymes in both patients with asthma and normal subjects, particularly in the airway epithelium and in submucosal macrophages, with no significant differences between the two groups. This suggests that there may be active intrinsic antioxidant mechanisms to protect against environmental agents that induce oxidative stress. There was no regulation following treatment with inhaled corticosteroids, despite demonstrable antiinflammatory effect. Our study implicates hemeoxygenases as a potentially important antioxidant defence mechanism in the airways.