INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Arachidonic acid metabolites, when triggered by allergens, trauma, or other injury, have been implicated as mediators of inflammation. Consisting largely of leukotrienes and prostaglandins, they have received much attention in the pathogenesis of asthma owing to their potent bronchomotor properties. The cyclooxygenase pathway represents the arm of the arachidonate metabolism that generates prostaglandins, thromboxane A₂, and prostacyclins. Prostaglandin H synthase (PGHS), also referred to as cyclooxygenase, performs the rate-limiting steps in the biosynthesis of prostanoids. Two isoforms of the enzyme have now been characterized; PGHS-1 is constitutively expressed whereas PGHS-2 is induced by inflammatory stimuli. The products of these two isoforms of enzyme depend on the cell type in which it is present. PGHS-1 is expressed in most mammalian tissues and likely plays a cellular housekeeping role, as suggested by the cytotoxicity of drugs that inhibit this enzyme. Conversely, PGHS-2 is expressed in many cell types, including airway epithelial and smooth muscle cells, after stimulation by various cytokines, mitogens, and lipopolysaccharide (1, 2). Subsequent studies have shown that expression of PGHS-2 appears to be constitutive in lung-derived epithelia, while the reverse holds true in bronchial smooth muscle cells, where PGHS-1 is predominant (3). There have been two reports describing the expression of PGHS in the airway of patients with asthma, one showing no difference between control subjects and patients with asthma and the other showing significant induction of PGHS-2 in patients with asthma compared with control subjects (4, 5). Both studies were carried out on bronchial biopsies, using semiquantitative techniques to assess the level of PGHS expression.

There has been a resurgence of studies on induced sputum. The use of induced sputum provides several advantages: it is safe, reliable, and quantifiable and can be used repeatedly in patients with mild and severe asthma. In this study, we employed immunohistochemistry and immunocytochemistry to determine the level of PGHS isoform expression in bronchial biopsies and induced sputum from patients with asthma and control subjects. We also assessed PGHS protein expression in induced sputum of patients with chronic obstructive pulmonary disease (COPD). We hypothesize that under the inflammatory conditions of asthma, which favors the induction of PGHS-2 and detectable increases of prostaglandins, airway PGHS-2 is upregulated.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients from Whom Sputum Was Collected

This study was approved by the Ethics Committee of the Montreal General Hospital (Montreal, PQ, Canada). Six subjects with atopic asthma, six control subjects with chronic obstructive pulmonary disease (COPD), and seven age-matched controls without pulmonary disease were included in the study (Table 1). The diagnosis of asthma was defined according to the criteria of the American Thoracic Society (6). Subjects were nonsmokers, or ex-smokers who had stopped smoking for at least 12 mo, and had smoked less than 5 pack-years in their lifetime. Subjects with asthma were classified clinically by their physicians as having mild or moderate asthma and all were using ₂- agonists on an as needed basis only, and/or were using < 1,000 µg/d of inhaled glucocorticoid or other antiinflammatory medications (sodium cromoglycate, nedocromil, ketotifen), which were discontinued at least 30 d before the study. Subjects using > 1,000 µg/d of inhaled glucocorticoid, oral glucocorticoid, theophylline, long-acting ₂-agonists, leukotriene antagonists, or antihistamines within the 3 mo before the study were excluded. None of the patients with patients with COPD were atopic. All patients with COPD were smokers of 47-73 pack-years (median, 62.5 pack-years). None of the study subjects had a history of respiratory tract infection within the previous 6 wk or received immunotherapy within the previous 12 mo. Blood was drawn for complete blood count and differential, total serum IgE, and coagulation studies. On another test day, subjects had sputum induced in the laboratory. A methacholine inhalation test (FEV₁ > 70%) preceded the induction of sputum by at least 48 h. FEV₁ was within a 10% difference on the two test days. Salbutamol (200 µg) was inhaled to inhibit possible bronchoconstriction during sputum induction. Increasing concentrations (3, 4, and 5%) of hypertonic saline generated by a DeVilbiss ultrasonic nebulizer (Ultra-Neb 99; Sunrise Medical, Somerset, PA) was administered. The procedure was interrupted every 2 min to measure FEV₁. Subjects were asked to rinse their mouth, swallow water, and blow their nose to minimize contamination with saliva and postnasal discharge. They were also instructed to cough sputum into a sterile container whenever they felt that sputum might be present. These procedures were repeated sequentially for 8-min periods at each concentration unless a fall in FEV₁ > 10% occurred, which resulted in termination of the procedure. Sputum was treated by adding 2 ml of Hanks' balanced salt solution (HBSS) containing 0.1% dithiothreitol (DTT), vortexed for up to 10 min. When homogeneous, samples were further diluted with 2 ml of phosphate-buffered solution (PBS) to stop the action of DTT and again vortexed briefly. The suspension was centrifuged at 300 × g for 5 min. The cell pellet was resuspended in PBS and a total cell count of leukocytes and cell viability were determined. The cell suspension was placed into the cups of a Shandon III cytocentrifuge (Shandon Southern Instruments, Sewickley, PA) and several coded cytospins were air dried and stained with Wright stain or fixed in 4% paraformaldehyde for 30 min, washed twice for 5 min with PBS, kept at 37° C overnight, and then stored at 80° C until further use. Sputum samples containing less than 20% contaminating squamous cells were considered suitable for analysis.

Patients in the Bronchial Biopsy Study

Patients with asthma (n = 9; five males; mean age 36.4 ± 4.8 yr) were included in the study on the basis of a compatible clinical history (FEV₁ 80.4 ± 8.1, FVC 88.3 ± 5.2) and either reversible airflow limitation (increase in FEV₁ of 15% or more with bronchodilators) or increased airway responsiveness to methacholine (PC₂₀ [provocative concentration of methacholine causing a 20% fall in FEV₁] < 8 mg/ml). Eight of the patients with asthma were atopic and all were nonsmokers. Of the nine patients with asthma, five were treated with inhaled corticosteroids (400 to 800 µg/d) and ₂-agonist (200 to 400 µg/d), and four patients were treated with ₂-agonist alone. Normal control subjects (n = 6; four males, mean age 26 ± 0.9; yr) were all nonsmokers and four were atopic. Pulmonary functions of these subjects were normal (FEV₁ 102.3 ± 5.52, FVC 100.3 ± 5.07) and the PC₂₀ was > 8 mg/ ml. Fiberoptic bronchoscopy and bronchial biopsies were carried out according to National Institutes of Health/American Thoracic Society (NIH/ATS) guidelines. Subjects included in this portion of the study were independent of those included in the sputum study. The tissues were fixed in 2% paraformaldehyde and placed in embedding medium. Cryostat sections were then cut (10 µm) and placed on poly-L-lysine-coated slide for immunohistochemical analysis.

Immunocytochemistry and Immunohistochemistry

Sputum and bronchial biopsies were immunostained with polyclonal antibodies against PGHS-1 and PGHS-2, which were used in the previous studies (7). A modification of the avidin-biotin-peroxidase complex method was used as described previously (8). Sections were immersed in 2% hydrogen peroxide for 1 h to block endogenous peroxidase activity. The sections were incubated with 10% normal serum to reduce background, followed by incubation with the first layer antiserum overnight at 4° C. The sections were incubated with biotin-conjugated goat anti-rabbit IgG for 45 min. The avidin-biotin-peroxidase complex was added to the sections for 45 min. Immunostaining was visualized by immersion in diaminobenzidine and hydrogen peroxide and counterstained with the nuclear stain hematoxylin. Negative control experiments included the preabsorption of the first layer antisera with their respective antigen and the use of nonimmune serum instead of the first layer antisera. As for immunocytochemistry, cells were graded as positive or negative. The number of positively immunostained cells in each sputum was assessed in six high-power fields under a Zeiss (Thornwood, NY) microscope (×100) and expressed as a percentage of total cell counts. This analysis was performed by two investigators in a blinded fashion. When there was a disagreement between the two observers on the final score, the mean of the two values was registered. In biopsy specimens, the percentage of immunostained surface area was assessed in six high-power fields for each specimen, using a computerized image-analysis system (Image-Pro Plus; Media Cybernetics, Silver Spring, MD).

Statistical Analysis

Data obtained in the analysis of immunocyto- and immunohistochemistry were designated as means ± standard error of the mean, and were compared by one-way factorial analysis of variance. A p value of less than 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Sputum

Immunostained sputum samples revealed a significant difference in PGHS expression between samples obtained from unaffected control subjects compared with patients with asthma (Figure 1A-D). All unaffected control subject samples showed minimal expression of PGHS-1 (5.8 ± 2.9%; Figure 1A) and PGHS-2 (2.6 ± 0.8%; Figure 1B), localized mainly to macrophages and to a lesser extent in neutrophils. In samples from patients with COPD, the percentage of immunostained cells for PGHS-2 (35.6 ± 3.8%) was significantly higher than in control subjects (p < 0.001). Most positively stained cells were macrophages. Patients with asthma showed a significantly higher number of immunostained cells for PGHS-2 (41.1 ± 5.6%, p = 0.001; Figures 1D and 2B) than did control subjects. Although patients with asthma had a greater percentage of immunostained cells for PGHS-1 (20.7 ± 7.3%) than did unaffected control subjects, this difference was not significant (p = 0.293, Figures 1C and 2A). Patients with asthma had a higher percentage of immunostained cells than did patients with COPD; however, this difference did not reach significance. (Immunostained cells consisted mainly of eosinophils, neutrophils, and macrophages. Cylindrical epithelial cells were also stained for both enzymes.) The percentage of cells immunostained for each enzyme is shown in Table 1. Differential cell counts revealed that the percentages of macrophages, neutrophils, and eosinophils in the control samples were 81.3 ± 5.1, 18.5 ± 5.0, and 0.3 ± 0.3%, respectively; 16.0 ± 7.5% of macrophages, 9.3 ± 5.8% of neutrophils, and 0% of eosinophils were immunostained for PGHS-2. Percentages of macrophages, neutrophils, and eosinophils in the asthma samples were 77.2 ± 3.7, 19.8 ± 3.2, and 3.0 ± 1.0%, respectively; 53.1 ± 5.7% of macrophages, 39.0 ± 5.5% of neutrophils, and 39.5 ± 30.7% of eosinophils were immunostained for PGHS-2. The percentages of macrophages and neutrophils in the COPD samples were 88.5 ± 3.4 and 11.6 ± 3.3%, respectively; 27.2 ± 7.4% of macrophages and 27.7 ± 9.4% of neutrophils were immunostained for PGHS-2. One control patient had a PC₂₀ of 8.22 mg/ml. Interestingly, his sputum showed a significant number of eosinophils, neutrophils, and macrophages immunostained for PGHS-1 and PGHS-2 (Table 1).

View larger version (114K): [in this window] [in a new window]   Figure 1.   (A-D) Sputum samples immunostained for PGHS-1 and PGHS-2. Samples from control subjects showed minimal immunoexpression for both PGHS-1 (A) and PGHS-2 (B). Staining was seen in a few macrophages. In contrast, samples from patients with asthma showed strong immunoreactivity for both PGHS-1 (C ) and PGHS-2 (D) in macrophages, eosinophils, and neutrophils. (E-H ) Bronchial biopsies immunostained for PGHS-1 and PGHS-2. (E ) An unaffected control subject showing moderate diffuse immunostaining for PGHS-2 in the airway epithelium, and moderate immunostaining in the inflammatory cells and vascular endothelium. (G) A biopsy from an unaffected control subject showing diffuse moderate immunostaining for PGHS-1 in the airway epithelium and vascular endothelium. (H ) A negative control; section immunostained with nonimmune serum.

Bronchial Biopsy

Bronchial biopsies from unaffected control subjects immunostained for PGHS-2 demonstrated moderate diffuse immunostaining in the columnar epithelium and diffuse weak immunostaining in the basal epithelium, vascular endothelium, and smooth muscle cells. Bronchoscopies from patients with asthma showed diffuse strong immunostaining for PGHS-2 in the denuded columnar epithelium. The basal epithelial cells and inflammatory cells demonstrated moderate PGHS-2 immunoexpression, while the vascular endothelium and smooth muscle cells showed diffuse weak immunostaining. The percentage of immunoreactive area for PGHS-2 was greater in patients with asthma (45.0 ± 6.9%) than in control subjects (30.4 ± 5.2%); however, the difference did not reach statistical significance. Corticosteroid therapy in our small patient population did not appear to affect PGHS immunoexpression. In unaffected control biopsies, there was strong diffuse immunoexpression of PGHS-1 in the columnar epithelial cells, with focal staining in the basal epithelium, and diffuse moderate immunostaining in the underlying smooth muscle cells and vascular endothelium. Bronchial biopsies from patients with asthma and immunostained for PGHS-1 demonstrated diffuse strong immunostaining in the denuded columnar epithelium, diffuse weak immunoreactivity in the underlying basal epithelium and vascular endothelium, and moderate staining in the smooth muscle layer and inflammatory cells. Negative control experiments revealed no immunostaining.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates the expression of PGHS-1 and PGHS-2 immunoreactivities in cells obtained by induced sputum from patients with asthma, patients with COPD, and unaffected controls, and in the bronchial tissue of patients with asthma and control subjects. The data demonstrated constitutive expression of PGHS-1 in the epithelium, smooth muscle cells, and vascular endothelium of unaffected controls and patients with asthma, except for staining of inflammatory cells, which was mainly seen in the latter. In contrast, bronchial biopsies of patients with asthma showed more abundant PGHS-2 in the airway epithelium and inflammatory cells than did those of unaffected control subjects. The percentage and total number of cells immunostained for PGHS-2 in the induced sputum of patients with asthma or COPD was significantly greater than that in control subjects. These findings demonstrate the constitutive expression of both enzymes in airways of control subjects, and significant induction of PGHS-2 in the airway epithelium and inflammatory cells of patients with asthma, which may contribute to alterations in bronchial tone and homeostasis.

Induced sputum has increasingly been used as an investigative tool in studying the pathogenesis of asthma and COPD. It provides a safe, reliable, and rapid way of determining cellular events in mild or severe obstructive lung diseases. Cell profiles in the sputum of patients with asthma and control subjects have been well characterized by studies that showed them to be comparable to bronchial biopsies and bronchoalveolar lavage in monitoring the presence and severity of airway inflammation in asthma (9). In our study, the induction of sputum was performed in a group of patients with mild asthma who were treated only with bronchodilators to determine the level of PGHS induction. We also collected induced sputum from a group of patients with COPD and a group of normal control subjects to assess the level of PGHS in asthma compared with other airway inflammatory diseases and with control subjects. We clearly demonstrate induction of PGHS-2 in asthma, and to a lesser extent in COPD, which suggests a role for this enzyme in a number of inflammatory lung diseases, and particularly in asthma. The findings of increased PGHS-2 immunoreactivity in COPD is not surprising because a number of PGHS-2 mediators of induction (e.g., cytokines, chemokines, and growth factors) are common to both COPD and asthma.

Studies have reported the expression of both PGHS-1 and PGHS-2 in airways of patients with asthma. Demoly and coworkers (4) reported no difference in the qualitative and quantitative scores of PGHS between control subjects and patients with asthma or chronic bronchitis. Conversely, Sousa and colleagues (5) reported increased expression of PGHS-2 but not PGHS-1 in the epithelium and submucosa of patients with asthma compared with control subjects. Our biopsy study demonstrates expression of both enzymes in airways of control subjects and patients with asthma, with the latter showing greater intensity for PGHS-2 in the epithelium and inflammatory cells. Bronchial biopsies provide useful information on the type of antigen-producing cells; however, it is often difficult to perform an accurate quantitative assessment of the number of immunoreactive cells. We therefore used immunocytochemistry on induced sputum to clearly identify the percentage of immunoreactive cells in the study subjects.

An important difference between this and previous studies is that none of the patients participating in the sputum study had received inhaled or oral glucocorticoid treatment within the 30 d before the study, ruling out any possible effect of steroid therapy on the expression of PGHS. Although five of the nine patients included in the biopsy study were receiving an inhaled steroid, it was not possible to detect any significant difference in PGHS-1 or PGHS-2 immunoreactivities between those receiving the therapy and those not receiving the therapy. The effect of inhaled corticosteroid therapy on PGHS expression needs to be addressed in a double-blind placebo- controlled study.

The present findings of constitutive expression of both isoforms of PGHS in the airway epithelium are supported by several previous reports. PGHS-1 and PGHS-2 have both been detected in human airway epithelial and smooth muscle cells, and PGHS-2 can be induced in response to cytokines and lipopolysaccharide (LPS) treatment (1). A morphology study has demonstrated constitutive expression of PGHS-1 and PGHS-2 in the airway epithelium of normal rats (10). Moreover, release of arachidonic acid metabolites is known to be elevated in patients with asthma (11). It appears, therefore, that induction of PGHS-2 as shown in this study, and increased PGHS-2-derived metabolites may contribute to the enhanced inflammatory response in asthma.

In conclusion, by using specific PGHS-1 and PGHS-2 antisera we demonstrated induction of PGHS-2 in the airways of patients with asthma and patients with COPD compared with unaffected control subjects without asthma. With the ever- increasing number of specific PGHS-2 inhibitors available for experimental use, the role that each isoform may play in asthma and other inflammatory airway diseases needs be assessed in future studies.