Abstract
The epithelial polymeric immunoglobulin receptor/transmembrane secretory component (pIgR/SC) transports into secretions polymeric immunoglobulin A (pIgA), which is considered the first line of defense of the respiratory tract. The present study, done with quantitative immunohistochemistry, evaluated epithelial expression of secretory component (SC) and Clara cell protein (CC16) and neutrophil infiltration into the airways of eight patients with severe chronic obstructive pulmonary disease (COPD) who were undergoing lung transplantation, as compared with these processes in six nonsmoking patients with pulmonary hypertension who were used as controls and in lung specimens from five smokers without chronic bronchitis. Staining for SC was significantly decreased in the COPD patients as compared with the controls, both in large (mean optical density [MOD]: 23.4 [range: 21.1 to 27.8] versus 42.2 [range: 28.2 to 49.3], p = 0.003) and in small airways (MOD: 30.8 [range: 20.3 to 39.4] versus 41.5 [range: 39.2 to 46.2], p = 0.003). SC expression in small airways correlated strongly with functional parameters such as FEV1 (Kendall's tau (K) = 0.76, p = 0.008), FVC (K = 0.64, p = 0.03), and midexpiratory flow at 50% of VC (MEF50) (K = 0.74, p = 0.01). The reduced expression of SC in large airways correlated with neutrophil infiltration in submucosal glands (K = − 0.47, p = 0.03). Expression of CC16 in the bronchial epithelium of COPD patients was also significantly decreased as compared with that of controls, especially in small airways (MOD: 28.3 [range: 26.8 to 32.4] versus 45.8 [range: 40.7 to 56.0], p = 0.002), but no correlation was observed with lung function tests. In conclusion, this study shows that reduced expression of SC in airway epithelium is associated with airflow obstruction and neutrophil infiltration in severe COPD.
The defense mechanisms of the respiratory tract involve both cellular and humoral immune components. It is now well accepted that secretory immunoglobulin A (SIgA), the predominant Ig isotype in mucosal secretions, prevents adherence and absorption of noxious bacterial or viral agents, acting as a scavenger (through so-called “immune exclusion”). This humoral immune defense requires the expression on the basolateral surface of epithelial cells of a specific receptor for polymeric Igs (pIgA and IgM) that allows their active transport into the mucosal lumen (1). The extracellular part of the polymeric Ig receptor (pIgR) is released after cleavage at the apical pole as secretory component (SC), which exists either in the free state or bound to pIgA or IgM. This secretory immune system, which implies close cooperation between submucosal pIg-secreting plasma cells and epithelial cells, probably acts in synergy with mucociliary clearance and constitutes the first line of defense of the proximal respiratory tract (2).
Clara cell protein (CC16) is one of the most abundant respiratory tract-derived proteins (5 to 10% of total proteins recovered after bronchoalveolar lavage), being constitutively secreted by nonciliated epithelial cells, mainly bronchiolar Clara cells (3), and probably also by other nonciliated columnar epithelial cells as demonstrated by in situ hybridization (4). Putative roles of CC16 include an antiinflammatory effect mediated by inhibition of phospholipase A2 and detoxification of xenobiotics. CC16 has also been proposed to be a useful marker of airway and lung epithelial injury (5).
Chronic obstructive pulmonary disease (COPD) is the major cause of morbidity and mortality among respiratory diseases. This disorder is characterized functionally by expiratory airflow limitation that is slowly progressive and mainly irreversible (6). Among the different structural changes that occur in COPD (7) (airway wall remodeling and distortion, increased intraluminal mucus secretion), the loss of elastic recoil associated with emphysema is probably the main mechanism of irreversible airway obstruction in severe COPD. In other respects, physiologic investigations have indicated that small airways are a major site of increased resistance in COPD (8). Although cigarette smoking is well documented as the major risk factor for COPD, only a minority (15 to 20%) of heavy smokers will develop COPD, and except for α1-antitrypsin deficiency, there is no test available to predict susceptibility to the disease, which is therefore often diagnosed late in its course (9). Moreover, although neutrophilic as well as CD8+ lymphocytic infiltration have been associated with a decline in FEV1 (10), the relationship of inflammatory and immune parameters to structural changes in the airways remains unclear.
Although CC16 protein, IgA, and SC have been assessed in the fluid phase of respiratory samples, the results of most of these assessments are difficult to compare, mainly for methodologic reasons. Thus, titrations of SC and IgA vary according to their different molecular forms (11). By contrast, immunohistochemical staining of lung tissue for SC and CC16, which has not yet been quantitatively assessed in human respiratory disorders in comparison with normal epithelium, could provide a more easily interpretable profile of expression of these proteins.
The purpose of the present study was to evaluate through immunohistochemistry the bronchial expression of SC and CC16 in large (both in the surface layer and in submucosal glands) and small conducting airways of COPD patients, and its potential correlation with lung function defects and neutrophilic infiltration. We therefore stained lung tissue from COPD patients undergoing lung transplantation for SC as a marker of mucosal immunity, for CC16 as a marker of bronchial injury, and for neutrophils as a marker of inflammation. Explants from patients with pulmonary hypertension were used as patient controls. We also evaluated lung specimens from smokers without chronic bronchitis who were undergoing lung surgery for solitary tumors in order to assess a potential effect of smoking on these epithelial markers.
We recruited both COPD and control patients from a population of patients undergoing lung transplantation. The indication for transplantation was based on international guidelines (12). An FEV1 < 25% predicted without reversibility, and/or a PaCO2 ⩾ 55 mm Hg (and/or pulmonary hypertension) with progressive deterioration were the main criteria for transplantation in COPD patients. Eight COPD patients, ranging in age from 42 to 60 yr (mean: 52 yr) were included. All were ex-smokers and had symptoms of chronic bronchitis, defined by cough and sputum production occuring on most days of the month for at least 3 mo per year during the 2 yr prior to the study. The patients' age, sex, and smoking indices are listed in Table 1. All COPD patients except one were treated with oral steroids (methylprednisolone, 4 to 8 mg/d), and all were receiving inhaled anticholinergic and β2-mimetic drugs before lung surgery. Controls consisted of six patients ranging in age from 17 to 55 yr (mean: 45 yr) undergoing transplantation for severe pulmonary hypertension that was either primary (two patients with the idiopathic form of disease and two patients with the anorexigenic form) or secondary to cardiovascular malformation (two patients with Eisenmenger's syndrome). None of the controls was treated with oral steroids. Severe progressive disease despite optimal medical treatment, especially with prostacyclin (New York Heart Association Classes III and IV), and especially in association with right heart failure as assessed by catheterization in primary pulmonary hypertension, was the main criterion for selection of patients with pulmonary hypertension. None of the controls had ever smoked, none had evidence of emphysema, and all had normally appearing bronchial epithelium upon optical microscopic examination.
| Patient | Age (yr) | Sex | Pack-years of smoking | |||
|---|---|---|---|---|---|---|
| COPD | ||||||
| 1 | 57 | F | 15 | |||
| 2 | 54 | M | 60 | |||
| 3 | 60 | M | 30 | |||
| 4 | 49 | F | 43 | |||
| 5 | 48 | M | 25 | |||
| 6 | 59 | M | 32 | |||
| 7 | 42 | F | 35 | |||
| 8 | 53 | M | 40 | |||
| Mean | 53 | 35* | ||||
| SEM | 2 | 4 | ||||
| Controls | ||||||
| 1 | 17 | F | 0 | |||
| 2 | 48 | M | 0 | |||
| 3 | 47 | F | 0 | |||
| 4 | 55 | M | 0 | |||
| 5 | 50 | F | 0 | |||
| 6 | 54 | M | 0 | |||
| Mean | 45 | 0 | ||||
| SEM | 5 | 0 | ||||
| Smokers | ||||||
| 1 | 69 | M | 100 | |||
| 2 | 65 | M | 15 | |||
| 3 | 42 | M | 21 | |||
| 4 | 74 | M | 40 | |||
| 5 | 44 | F | 30 | |||
| Mean | 59 | 41* | ||||
| SEM | 6 | 14 |
A third group consisted of five patients (three current and two ex-smokers) ranging in age from 42 to 74 yr (mean: 59 yr) who were recruited from a population of smokers without chronic bronchitis who were undergoing lung surgery for a solitary peripheral lung tumor. These subjects' smoking index was not significantly different from that of the COPD patients. All patients in the study were clinically stable, without evidence of lung infection at the time of surgery, and the transplant recipients and smokers had pulmonary function tests performed at 37 ± 26 d (mean ± SD) and 15 ± 4 d, respectively, before surgery. For each patient, the severity of emphysema was histologically assessed on two to six slides (mean: 3.4), using a subjective panel grading method. A score of from 0 (normal lung) to 10 (almost complete disappearance of lung parenchyma) was applied to each slide by comparison with a standard set of 10 photographs from Nagai and colleagues (13) that illustrate each grade of emphysema. A mean score was then calculated for each patient. Lung function test values, blood gas values, and emphysema score are presented for each patient in Tables 2 and 3.
| Patients | FEV1(%) | FVC (%) | FEV1/FVC (%) | MEF50(%) | FRC (%) | TLC (%) | RV (%) | Raw (%) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| COPD | ||||||||||||||||
| 1 | 12 | 32 | 37 | 3 | 160 | 111 | 216 | 365 | ||||||||
| 2 | 23 | 55 | 42 | 5 | 158 | 120 | 210 | 289 | ||||||||
| 3 | 28 | 98 | 29 | 6 | 195 | 126 | na | 226 | ||||||||
| 4 | 26 | 45 | 58 | 5 | 201 | 98 | 134 | 433 | ||||||||
| 5 | 17 | 42 | 40 | 3 | 234 | 139 | 320 | 375 | ||||||||
| 6 | 16 | 30 | 53 | 3 | 199 | 108 | 236 | 259 | ||||||||
| 7 | 22 | 75 | 29 | 5 | 252 | 165 | 336 | 340 | ||||||||
| 8 | 17 | 63 | 27 | 5 | 173 | 109 | 201 | 456 | ||||||||
| Mean | 23* | 55* | 39* | 4* | 196* | 122* | 236* | 343* | ||||||||
| SEM | 2 | 8 | 4 | 1 | 11 | 7 | 25 | 27 | ||||||||
| Controls | ||||||||||||||||
| 1 | 80 | 87 | 92 | 82 | na | 104 | 186 | 61 | ||||||||
| 2 | 95 | 96 | 99 | 98 | 100 | 101 | 94 | 88 | ||||||||
| 3 | 92 | 91 | 101 | 90 | 96 | 100 | 107 | 89 | ||||||||
| 4 | 67 | 67 | 100 | 68 | 90 | 93 | 121 | 52 | ||||||||
| 5 | 87 | 80 | 109 | 91 | 95 | 101 | 110 | 70 | ||||||||
| 6 | 70 | 73 | 96 | 63 | 94 | 86 | 138 | 127 | ||||||||
| Mean | 82 | 82 | 100 | 82 | 95 | 97 | 126 | 81 | ||||||||
| SEM | 5 | 4 | 2 | 5 | 1 | 2 | 12 | 10 | ||||||||
| Smokers | ||||||||||||||||
| 1 | 107 | 107 | 100 | 73 | 95 | 110 | 90 | 51 | ||||||||
| 2 | 97 | 96 | 101 | 89 | 102 | 95 | 101 | 41 | ||||||||
| 3 | 107 | 117 | 91 | 92 | 94 | 111 | 96 | 38 | ||||||||
| 4 | 101 | 98 | 103 | 97 | 90 | 90 | 89 | 34 | ||||||||
| 5 | 93 | 109 | 85 | 66 | 93 | 103 | 87 | 51 | ||||||||
| Mean | 101 | 105 | 96 | 83 | 95 | 102 | 93 | 43 | ||||||||
| SEM | 2 | 3 | 3 | 5 | 1 | 3 | 2 | 3 |
| Patient | Dl CO(%) | PaO2 (mm Hg) | PaCO2 (mm Hg) | Emphysema Score | ||||
|---|---|---|---|---|---|---|---|---|
| COPD | ||||||||
| 1 | 34 | na* | na* | 5 | ||||
| 2 | 25 | 63 | 40 | 7 | ||||
| 3 | 30 | 75 | 38 | 5 | ||||
| 4 | 16 | 55 | 47 | 7 | ||||
| 5 | 19 | 51 | 52 | 4 | ||||
| 6 | 14 | 64 | 43 | 3.5 | ||||
| 7 | 18 | 59 | 43 | 5.5 | ||||
| 8 | 34 | 57 | 42 | 5.7 | ||||
| Mean | 24* | 61* | 44* | 5* | ||||
| SEM | 3 | 3 | 2 | 1 | ||||
| Controls | ||||||||
| 1 | 60 | 71 | 32 | 0 | ||||
| 2 | 93 | 68 | 39 | 0 | ||||
| 3 | 87 | 60 | 28 | 0 | ||||
| 4 | 63 | 66 | 31 | 1 | ||||
| 5 | 88 | 70 | 35 | 0 | ||||
| 6 | 102 | 67 | 40 | 0 | ||||
| Mean | 82 | 67 | 34 | 0 | ||||
| SEM | 6 | 1 | 2 | 0 | ||||
| Smokers | ||||||||
| 1 | 88 | na | na | 0 | ||||
| 2 | 90 | na | na | 1 | ||||
| 3 | 82 | na | na | 0 | ||||
| 4 | 91 | na | na | 0 | ||||
| 5 | 43 | na | na | 0 | ||||
| Mean | 79 | na | na | 0.5 | ||||
| SEM | 6 | na | na | 0 |
Formaldehyde, glass slides, mounting medium, and Tris were purchased from Vel (Leuven, Belgium). Other reagents and their sources were hydrogen peroxide (H2O2) (Merck, Darmstadt, Germany); bovine serum albumin (BSA) and 3,3′-diaminobenzidine tetrahydrochloride (DAB; ICN, Costa Mesa, CA); and streptavidin–horseradish peroxidase conjugate (Sigma, St. Louis, MO). Affinity-purified, biotinylated rabbit antigoat IgG, goat antirabbit IgG, and goat antimouse IgG antibodies, as well as rabbit polyclonal antihuman CC16 antibody and mouse monoclonal IgG1κ antihuman neutrophil elastase antibody (clone NP57) were purchased from Dako (Glostrup, Denmark). Goat antihuman SC antiserum was prepared in our laboratory as previously reported (14). Two other polyclonal antibodies against human SC (hSC) were used in order to confirm the specificity of the staining: a goat anti-hSC IgG fraction and a rabbit affinity-purified anti-hSC antibody. These three different anti-SC preparations gave similar results. Normal goat serum (NGS), normal rabbit serum (NRS), and mouse IgG1κ isotype, were used as controls for their respective primary antibodies.
After surgical removal, the whole-lung explants were immediatly immersed in 4% formaldehyde in phosphate-buffered saline at pH 7.4 under a constant inflation pressure of 30 cm H2O for at least 24 h. Random lung tissue blocks (from two to six for each patient; mean: 3.4) were sampled and examined at a magnification of ×25 for histologic grading of emphysema according to Nagai and colleagues (13). Samples for immunohistology were obtained from central and peripheral areas of each lobe of one (single transplantation) or both (bipulmonary transplantation) lungs and were embedded under vacuum in paraffin. Lung specimens from smokers were sampled at locations away from the tumor site, in the same lobe (lobectomy) or in two or three lobes of the same lung (right or left pneumonectomy, respectively) obtained at surgery. Serial sections of 5 μm thickness were cut from paraffin blocks, spread on polylysine-coated glass slides, and dried at 40° C for at least 24 h.
The slides were then processed for immunostaining, each step of the procedure being followed by washing with Tris-buffered saline (pH 7.4). After disembedding and rehydration of the specimen, endogenous peroxidases were inhibited by incubation in 0.03% (vol/vol) H2O2 in water for 30 min, and the slides were treated with 1% (wt/vol) BSA in Tris-buffered saline for 30 min to neutralize remnant reactive aldehyde groups originating from fixation. Slides were then incubated overnight at 4° C with goat anti-SC antiserum, rabbit polyclonal anti-CC16 antibody (1:500 dilution for each), or mouse anti-neutrophil elastase antibody (1:50). Control sections were treated with NGS, NRS, or mouse IgG1 isotype at the same dilution. The secondary antibody, biotinylated rabbit antigoat IgG, goat antirabbit IgG, or goat antimouse IgG (1:500), was applied in 10% (wt/vol) defatted dry milk for 30 min. The reaction was amplified with streptavidin–peroxidase conjugate (1:500) in Tris-buffered saline containing 1% BSA for 30 min, and color was developed by incubation with 0.6 mg/ml diaminobenzidine in 0.03% H2O2 for 10 min. After the reaction was stopped by washing in water, slides were counterstained with Mayer's hemalum and mounted with coverslips in Eukitt's medium.
Quantification of immunostaining was done according to a previously published methodology (15). For each slide, staining for SC and CC16 was studied in 10 different areas of well-preserved bronchial epithelium (defined by the presence of both basal and columnar cells without detachment from the basement membrane) in both the surface layer and submucosal glands, including areas of COPD epithelium showing hyperplasia and metaplasia (Figure 1). From two to eight different airways per slide, and a total of eight to 20 large airways and 10 to 30 small airways per patient, were evaluated. The length of epithelium analyzed for each patient was 9.35 ± 2.47 mm (mean ± SEM) in large and 9.60 ± 3.38 mm in small airways for COPD patients; 8.03 ± 2.86 mm and 8.32 ± 2.59 mm, respectively, for controls; and 7.54 ± 2.50 mm and 7.71 ± 2.47 mm, respectively, for smokers. No large airway was available for one control. For neutrophils, staining was studied in 10 areas of epithelium (intraepithelial neutrophils), 10 areas of lamina propria (defined as the tissue present between the basement membrane and the smooth-muscle layer), and 10 areas of submucosal glands (defined as acini and interstitium between acini) in each section in which these structures were available and preserved. From three to 11 slides were studied for each patient. Computer-assisted quantification of the staining in the selected areas was done with the SAMBA 2005 system (Alcatel TITN, Grenoble, France) equipped with a color camera and an optical microscope (BX 50; Olympus, Tokyo, Japan), using a final magnification of ×400.
Fig. 1. Computer-assisted analysis of immunostaining. Microscopic magnification of the surface epithelium in a large airway from a control patient showing brown staining by Clara cell protein (anti-CC16) antibody (original magnification: ×400). The broken black line, outlined by a cursor on the computer screen, defines an area of bronchial epithelium for quantification of staining.
[More] [Minimize]The definition of large and small airways was based on histologic criteria. Large airways were defined as bronchi with cartilage and/or submucosal glands, and small airways as membranous bronchioles without cartilage or glands (and not yet alveolated), and with an internal diameter < 2 mm, as previously described (16).
Results were expressed through two different staining indexes: mean optical density (MOD) and labeling index (LI). MOD (arbitrary units [a.u.]) represents the mean staining intensity of the considered area of bronchial epithelium. LI (%) represents the relative percentage of surface staining (i.e., stained surface as a percentage of the total selected surface), and therefore represents the relative number of positive cells. All staining values were corrected by substracting values obtained with control slides.
Data considered for statistical analysis were the staining indices for the different slides (from three to 11 slides) from each patient. Because parametric conditions were not satisfied after checking the equality of variance by means of Levene's test and fitting to a normal distribution by means of the chi-square test of goodness of fit was not found, nonparametric tests were used. Results are thus expressed as medians and ranges. Two types of comparisons were made. First, a mean value was calculated for each patient from the staining indices obtained with the different slides (Figure 4), and between-group comparisons (COPD patients versus controls) were made with these data, using the Mann–Whitney U test. Second, within-group comparisons (large versus small airways) were made for paired data from the same slide (in cases in which both large and small airways were present), using Wilcoxon's matched pairs test. Rank correlations were tested with the nonparametric Kendall's test, and values of p < 0.05 were considered significant. All statistical analyses were done with Statistica software (Statsoft, Tulsa, OK).
Fig. 4. Intensity (mean optical density [MOD]) of SC expression in large airways (A), small airways (B), and submucosal glands (C) from controls (CT; left panel ), smokers (SM; middle panel ), and COPD patients (right panel ). Values for the six controls, five smokers, and eight COPD patients represent means (solid squares) ± SEM (open rectangles) and SD (bars) for from three to 11 slides, each assessed for 10 different epithelial areas.
[More] [Minimize]The validity and reproducibility of the analytical method used in the study were determined by considering the mean coefficient of variation for three repeated measurements made by the same observer (which ranged from 3% to 9%) and the interobserver correlation coefficient (which varied from 0.92 to 0.98). These coefficients were obtained for the different immunostaining results when 10 areas per slide were quantified and did not improve significantly with the use of more than 10 areas.
The pattern of SC staining was rather homogeneous, with little variability from one field to another in the same airway and little variability from one airway to another in the same slide, with respect to airway size (large versus small airways). As previously reported (17), SC-positive cells were found to correspond generally to nonciliated glandular (mucous and serous) cells, both in the surface and in the submucosal glandular epithelium, and more weakly to ciliated cells (Figure 2). In contrast, and in accord with the previously reported presence of “clusters” of CC16-expressing cells (18), staining for CC16 was very patchy. The CC16-positive cells were found to correspond to thin, nonciliated cells defined as Clara cells (3). However, in large airways, CC16-positive cells were also represented by goblet cells (Figure 3).
Fig. 2. Secretory component (SC) expression in airways from control and chronic obstructive pulmonary disease (COPD) patients. SC immunostaining in large (A to D) and small ( E to H ) airways from a COPD patient (C to D and G to H ) versus a control patient (A and B and E and F ) (A, C, E, and G, original magnification: ×40) (B, D, F, and H, original magnification: ×400). Bronchial epithelium from the control patient exhibits intense and homogeneous staining, both in the surface layer and in submucosal glands. By contrast, SC immunostaining is strongly reduced and heterogeneous in the airway epithelium of the COPD patient.
[More] [Minimize]
Fig. 3. CC16 expression in airways from control and COPD patients, shown in the same order as for SC immunostaining in Figure 2. CC16-expressing cells corresponded especially to Clara and goblet cells in small and large airways, respectively. Staining for CC16 is especially reduced in epithelium of small airways from COPD patient.
[More] [Minimize]In normal control epithelium there was no significant difference between SC expression in large and small airways (Figure 5). Thus, SC expression appeared to be homogeneous along the normal bronchial tree. By contrast, SC expression in the airways of COPD patients was more specifically reduced in large than in small airways, as reflected both by MOD (MOD in large airways: 23.4 [range: 21.1 to 27.8] versus MOD in small airways: 30.8 [range: 20.3 to 39.4]; p = 0.003) and LI (LI in large airways: 16.4 [range: 5.0 to 40.7], versus LI in small airways: 48.0 [range: 23.7 to 57.0], p = 0.002) (Figure 5).
Fig. 5. SC and CC16 staining indices in bronchial epithelium. Reduction of SC expression in large and small airways and in submucosal glands from COPD patients versus controls is given in terms of MOD (p = 0.003 for all) (A) and LI (p = 0.003, p = 0.002, p = 0.002, respectively) (B). Both MOD and LI for SC expression were more strongly reduced in large than in small airways of COPD patients (p = 0.003, p = 0.002 respectively) than of controls. CC16 expression was also reduced in COPD patients versus controls, both in terms of MOD (C ) and LI (D), especially in small airways and in glands.
[More] [Minimize]In agreement with the preferential bronchiolar localization of Clara cells, the intensity (MOD) of CC16 immunoreactivity was stronger in small than in large airways from controls (MOD in large airways: 37.6 [range: 32.5 to 46.7] versus MOD in small airways: 45.8 [range: 40.7 to 56.0], p = 0.001), but the relative number of positive cells (LI) was not significantly different between small and large airways from these patients. However, the pattern in COPD was different, since the strongest CC16 expression occurred in large airways, both in terms of MOD (MOD in large airways: 31.9 [range: 28.7 to 38.1] versus MOD in small airways: 28.3 [range: 26.8 to 32.4], p = 0.001) and LI (LI in large airways: 72.8 [range: 27.9 to 89.3] versus LI in small airways: 65.9 [range: 20.3 to 87.6], p = 0.002) (Figure 5).
As shown in Figure 5 (see also Figures 2 and 3), the intensity of SC expression (MOD) was strongly reduced in the bronchial epithelium of COPD patients as compared with controls, both in large (MOD: 23.4 [range: 21.1 to 27.8] versus 42.2 [range: 28.2 to 49.3], respectively; p = 0.003) and in small airways (MOD: 30.8 [range: 20.3 to 39.4] versus 41.5 [range: 39.2 to 46.2], respectively; p = 0.003). The relative number of SC-positive epithelial cells was also significantly decreased in COPD patients as compared with controls, both in large (LI: 16.4 [range: 5.0 to 40.7] versus 88.4 [range: 59.1 to 91.9], respectively; p = 0.003) and small airways (LI: 48.0 [range: 23.7 to 57.0] versus 77.7 [range: 67.9 to 87.6], respectively; p = 0.002). The same pattern was observed in the submucosal glands (MOD: 27.8 [range: 23.7 to 30.0] versus 40.4 [range: 37.5 to 44.5], respectively, p = 0.003; and LI: 63.6 [range: 55.5 to 68.1] versus 86.8 [range: 84.7 to 91.3], COPD patients versus controls, respectively, p = 0.002).
Despite some overlap between values for COPD patients and controls, CC16 immunostaining was significantly reduced in COPD patients' airways. In particular, the MOD for CC16 in COPD was especially reduced in small airways (MOD: 28.3 [range: 26.8 to 32.4] versus 45.8 [range: 40.7 to 56.0], p = 0.002) and submucosal glands (MOD: 26.6 [range: 24.7 to 28.6] versus 34.6 [range: 28.9 to 44.6], p = 0.003) (Figure 5).
SC expression was not significantly different in smokers than in nonsmoking control patients (see Figure 4). In contrast, the MOD for CC16 expression was significantly decreased in smokers as compared with controls, both in large (MOD: 29.8 [range: 27.4 to 33.7] versus 37.6 [range: 32.6 to 46.8], p = 0.02) and small airways (MOD: 34.2 [range: 31.3 to 38.3] versus 46.8 [range: 40.7 to 55.9], p = 0.006). In the submucosal glands, both the MOD and the LI for CC16 expression were significantly decreased in smokers as compared with controls (MOD: 27.6 [range: 26.4 to 33.2] versus 34.6 [range: 28.9 to 44.6], p = 0.03; and LI: 68.4 [range: 67.0 to 71.4] versus 80.8 [range: 76.0 to 95.6], p = 0.009).
The MOD for neutrophil infiltration was significantly increased in mucosal tissue from COPD patients as compared with controls, in both the surface epithelium and lamina propria of large airways (MOD: 11.3 [range: 8.3 to 16.4] versus MOD: 1.1 [range: 0.0 to 4.1], respectively, p = 0.002; and MOD: 9.6 [range: 4.4 to 12.0] versus MOD: 2.5 [range: 0.0 to 6.4], respectively, p = 0.003), as well as in the submucosal glands (MOD: 15.4 [range: 3.7 to 19.2] versus MOD: 0.7 [range: 0.0 to 4.1], respectively, p = 0.003) and in the epithelium and lamina propria of small airways (MOD: 15.1 [range: 7.7 to 22.5] versus MOD: 2.7 [range: 0.0 to 5.3], respectively, p = 0.002; and MOD: 21.9 [range: 11.0 to 30.5] versus MOD: 2.1 [range: 0.0 to 4.7], respectively, p = 0.002). The relative number of neutrophils was also significantly higher in COPD patient's airways than in those of controls, both in the epithelium and in the lamina propria of large airways (LI: 2.9 [range: 2.2 to 6.0] versus LI: 0.003 [range: 0.0 to 0.1], respectively, p = 0.002; and LI: 2.2 [range: 1.3 to 3.5] versus LI = 0.1 [range: 0.0 to 0.2], respectively, p = 0.002 ), as well as in glands (LI = 4.3 [range: 0.3 to 6.8] versus LI: 0.005 [range: 0.0 to 0.2], respectively, p = 0.002), and in the epithelium and lamina propria of small airways (LI = 6.0 [range: 2.1 to 11.3] versus LI: 0.001 [range: 0.0 to 0.2], respectively, p = 0.002; and LI: 9.5 [range: 4.4 to 15.6] versus LI: 0.2 [range: 0.0 to 0.5], respectively, p = 0.002) (Figure 7).
Fig. 7. Neutrophils (brown staining) in the epithelium of a large airway and in glands from control (A and B, respectively) and COPD (C and D, respectively) patients (original magnification: ×400). Neutrophils were consistently present in the airway mucosa of COPD patients but were rarely observed in controls.
[More] [Minimize]No significant correlation was found between CC16 expression and clinical data (age or smoking history), functional data, or histologic emphysema score.
In contrast, SC expression in small airways (but not in large airways) correlated strongly with functional parameters of obstruction, such as FEV1 (Kendall's tau [K] = 0.76, p = 0.008 for MOD; K = 0.62, p = 0.03 for LI), FVC (K = 0.64, p = 0.02 for MOD; p = NS for LI), and midexpiratory flow at 50% of VC (MEF50) (K = 0.74, p = 0.01 for MOD; K = 0.65, p = 0.02 for LI) (Figure 6). There was no correlation with other functional parameters (static volumes, diffusing capacity, blood gases) or clinical data (age, smoking index) or with the morphologic emphysema score.
Fig. 6. Functional parameters of obstruction (FEV1, FVC, MEF50) plotted against the MOD for SC expression in small airways of COPD patients. Significant correlation was found between SC expression and lung function parameters (K-Tau = Kendall-Tau).
[More] [Minimize]The MOD for SC expression in large airways was significantly (negatively) correlated with the MOD for neutrophils in submucosal glands (K = −0.47, p = 0.03). No correlation was observed in other bronchial compartments, or between CC16 expression and neutrophils.
In COPD, bronchial and parenchymal damage is thought to lead to defects in airflow and gas exchange. Despite important epithelial changes observed in bronchi from COPD patients (7), only indirect data suggest a decreased production of epithelial-derived proteins such as SC and CC16. The present study was designed to evaluate, by computer-assisted quantification of immunostaining, epithelial SC and CC16 expression in airways from patients with severe COPD, as compared with those of controls, and the potential correlation of this expression with functional parameters of ventilation and gas exchange.
In these patients with severe COPD, both total SC expression (MOD) and the relative number of SC-positive cells (LI) were drastically reduced in both large (in the surface layer and in submucosal glands) and small airways in comparison with those of controls with pulmonary hypertension. This decreased staining was particularly striking in terms of LI, without overlap between the two patient groups. Selection of the COPD population recruited from among patients undergoing lung transplantation, and therefore in a terminal phase of their disease, may account for the large differences between COPD patients and controls observed in this study. Our results also demonstrate a reduction of immunostaining for CC16 in the bronchial epithelium of patients with severe COPD as compared with controls, although to a lesser extent than for SC. This is in agreement with previous reports of lower levels of fluid-phase CC16 in bronchoalveolar lavage fluid (BALF) or sputum (5), and decreased epithelial CC16 immunoreactivity in bronchial biopsies (19) from smokers, but with an overlap between smokers and nonsmokers. In addition, the decreased expression of CC16 in COPD is more striking in small than in large airways. According to recent data (20), CC16-positive cells in large airways appear to correspond to goblet cells, suggesting that this relative preservation of CC16 expression in large airways of COPD patients could be related to goblet cell hyperplasia, which is more intense in these airways.
The decline of SC immunoreactivity in severe COPD is in agreement with the previously described decreased level of SC in BALF from 20% of cigarette smokers (21). In lung fibrosis and sarcoidosis, decreased free SC in BALF has also been described, in association with increased monomeric and polymeric IgA (22). In contrast, other authors have reported an increase in SC and IgA in sputum from infected patients with COPD (23) or bronchiectasis (24), and in BALF from patients with stable asthma (25). Comparisons between these studies are difficult because of the different assay methods used and disease severity of the patient populations, as well as the potential role of infection. Quantitative immunohistochemistry for SC has allowed us to propose that secretory IgA (SIgA)-dependent immunity is clearly impaired in bronchi from patients with severe COPD. Moreover, despite need for the consideration that cancer could influence SC production in nonneoplasic and contralateral lobes (5), the decline in SC expression in COPD does not seem to be a direct effect of smoking, since the level of expression of SC in smokers (without chronic bronchitis) is not significantly different from that in nonsmokers with pulmonary hypertension. The decline in SC and CC16 in COPD is not related only to metaplasia, since nonmetaplastic areas also exhibit a consistently decreased expression of these proteins. Furthermore, since all (except one) of the COPD patients in our study were treated with low-dose steroids, in contrast to the controls, we cannot exclude the possibility that this treatment influenced SC (or CC16) expression.
SIgA is thought to prevent chronic mucosal inflammation by inhibiting adherence of antigens and microorganisms to the surface epithelium (1, 2), and by downregulating proinflammatory processes, such as the respiratory burst in monocytes and neutrophils (26). The concept of protective secretory immunity applied to COPD may account for some major features of this disorder, and our observation that expression of SC was decreased in the airway epithelium of COPD patients would support this concept. Thus, a deficiency in SIgA may lead to bacterial colonization and infection, which are typical complications and aggravating factors of COPD (6, 9). The invading bacteria may initiate inflammatory processes, either directly, through the release of bacterial mediators such as endotoxin, and/or indirectly, through the release of cytokines and chemokines by epithelial and mononuclear cells (27), leading to recruitment and activation of inflammatory cells, such as neutrophils, in the airway wall and lumen. Progression of the disease may be related to a vicious cycle initiated by bacteria and/or by neutrophils, which can degrade IgA through specific proteases (1) or through elastase (28), respectively. Several authors have reported an inflammatory infiltration of the bronchial mucosa in COPD patients (7, 29-32), and a correlation between neutrophils as well as CD8+ lymphocytes and parameters of airway obstruction such as FEV1. However, the precise role of inflammation is more controversial in COPD than in asthma (33). Impairment of secretory immunity could provide a link between structural changes and inflammation. Our correlation between neutrophils in submucosal glands and decreased SC expression in bronchi from COPD patients supports this hypothesis, as well as the concept that bronchial glands represent a particular compartment, as suggested by others (31).
The strong correlations found in our study between expression of SC/pIgR and plethysmographic parameters of airway obstruction (such as FEV1, FVC, and MEF50) in COPD are to our knowledge the first correlations to be shown between a noninflammatory biologic marker and airflow limitation. If the narrow range of functional defects in our COPD patients limits the conclusions about correlations, the absence of correlation with other functional parameters or with smoking history suggests a potential pathophysiologic relevance of our observations. In addition, the correlations were restricted to small airways, as opposed to large airways, and this supports previous studies (34) suggesting an important role of small airways in the pathogenesis and progression of COPD, as well as in the airflow limitation in severe COPD. That these correlations especially concern the MOD values for SC staining supports the concept that total SC production, rather than the relative number of SC-producing cells, better reflects relevant changes in secretory immunity.
Our study specifically evaluated patients with severe COPD undergoing lung transplantation allowing us to test a large number of both large and small airways, and thus to compare the epithelial secretory profile between these different airways. Future studies should include assessment of SC immunostaining in lung specimens from patients with less severe COPD.
In summary, the present study reports an impairment of SC expression by the bronchial epithelium of patients with severe COPD undergoing lung transplantation. The most striking feature of this was decreased staining of SC in both large (in surface and in submucosal glands) and small airways of these patients, whereas CC16 expression was reduced to a lesser extent and did not correlate with defects in lung function. Moreover, decreased expression of SC correlated with functional parameters of airway obstruction and with neutrophil infiltration. The mechanisms related to these correlations remain to be elucidated. In this context, we plan to study recently obtained SC-deficient mice (35) for their susceptibility to COPD.
The authors thank P. Thurion for his excellent technical assistance.
Supported by grant FRSM 3.4590.99 and by the Fonds National de la Recherche Scientifique, Belgium.
| 1. | Brandtzaeg P, Jahnsen FL, Farstad INImmune functions and immunopathology of the mucosa of the upper respiratory pathways. Acta Otolaryngol1161996149159 |
| 2. | Underdown BJ, Schiff JMImmunoglobulin A: strategic defence at the mucosal surface. Annu Rev Immunol41986389417 |
| 3. | Gail DB, Lenfant CJMCells of the lung: biology and clinical implications. Am J Respir Cell Mol Biol1271982366387 |
| 4. | Peri A, Cordella-Miele E, Miele L, Mukherjee ABTissue-specific expression of the gene coding for human Clara cell 10-kD protein, a phospholipase A2-inhibitory protein. J Clin Invest92199320992109 |
| 5. | Bernard A, Marchandise FX, Depelchin S, Lauwerys R, Sibille YClara cell protein in serum and bronchoalveolar lavage of healthy subjects and patients with pulmonary diseases. Eur Respir J5199212311238 |
| 6. | American Thoracic Society. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am Rev Respir Dis 1986;136:225–244. |
| 7. | Jeffery PK1991. Morphology of the airway wall in asthma and in chronic obstructive pulmonary disease. Am Rev Respir Dis143199111521158 |
| 8. | Hogg J, Macklem PT, Thurlbeck WMSite and nature of airway obstruction in chronic obstructive lung disease. N Engl J Med178196813551369 |
| 9. | European Respiratory Society. Consensus statement: optimal assesment and management of chronic obstructive pulmonary disease (COPD). Eur Respir J 1995;8:1398–1420. |
| 10. | Di Stefano A, Capelli A, Lusuardi M, Balbo P, Vecchio C, Maestrelli P, Mapp CE, Fabbri LM, Donner CF, Saetta MSeverity of airflow limitation is associated with severity of airflow inflammation in smokers. Am J Respir Crit Care Med158199812771285 |
| 11. | Delacroix DL, Dive C, Rambaud JC, Vaerman JP. IgA subclasses in various secretions and in serum. Immunology 1982;47:383–385. |
| 12. | American Thoracic Society. International guidelines for the selection of lung transplant candidates. Am J Respir Crit Care Med 1998;158:335– 339. |
| 13. | Nagai A, Yamawaki I, Thurlbeck WM, Takizawa TAssessment of lung parenchymal destruction by using routine histologic tissue sections. Am Rev Respir Dis1391989313319 |
| 14. | Delacroix DL, Vaerman JPA solid phase, direct competition, radioimmunoassay for quantification of secretory IgA in human serum. J Immunol Methods401980345358 |
| 15. | Camby I, Nagy N, Lopes MB, Schafer BW, Maurage CA, Ruchoux MM, Murman P, Pochet R, Heizman CW, Brotchi J, et al.. Supratentorial pilocytic astrocytomas, astrocytomas, anaplastic astrocytomas and glioblastomas are characterized by a differential expression of S100 proteins. Brain Pathol91999119 |
| 16. | Mullen JBM, Wright JL, Wiggs BR, Paré PD, Hogg JCStructure of central airways in current smokers and ex-smokers with and without mucus hypersecretion: relationship to lung function. Thorax421987843848 |
| 17. | Goodman MR, Link DW, Brown WR, Nakane PKUltrastructural evidence of transport of secretory IgA across bronchial epithelium. Am Rev Respir Dis1231981115119 |
| 18. | Lumsden AB, McLean A, Lamb DGoblet and Clara cells of human distal airways: evidence for smoking induced changes in their numbers. Thorax391984844849 |
| 19. | Shijubo N, Yamagushi Y, Shibuya Y, Morita Y, Hirasawa M, Okutani R, Kawai T, Abe SSerum and BAL Clara cell 10 kDa protein (CC10) levels and CC10–positive bronchiolar cells are decreased in smokers. Eur Respir J10199711081114 |
| 20. | Boers JE, Ambergen AW, Thunnissen FBJMNumber and proliferation of Clara cells in normal human airway epithelium. Am J Respir Crit Care Med159199915851591 |
| 21. | Merrill WW, Goodenberger D, Strober W, Matthay RA, Naegel GP, Reynolds HYFree secretory component and other proteins in human lung lavage. Am Rev Respir Dis1221980156161 |
| 22. | Delacroix DL, Marchandise FX, Francis C, Sibille YAlpha-2-macroglobulin, monomeric and polymeric immunoglobulin A, and immunoglobulion M in bronchoalveolar lavage. Am Rev Respir Dis1321985829835 |
| 23. | Stockley RA, Afford SC, Burnett DAssessment of 7S and 11S immunoglobulin A in sputum. Am Rev Respir Dis1221980959965 |
| 24. | Burnett D, Hill SL, Bradwell AR, Stockley RAIgA subclasses in sputum from patients with bronchiectasis. Respir Med841990123127 |
| 25. | Van de Graaf EA, Out TA, Kobesen A, Jansen HMLactoferrin and secretory IgA in the bronchoalveolar lavage fluid from patients with a stable asthma. Lung1691991275283 |
| 26. | Wolf HM, Vogel E, Fischer MB, Rengs H, Schwarz HP, Eibl MMInhibition of receptor-dependent and receptor-independent generation of the repiratory burst in human neutrophils and monocytes by human serum IgA. Pediatr Res361994235243 |
| 27. | Koyama S, Rennard SR, Leikauf GD, Shoji S, Von Essen S, Claassen L, Robbins RAEndotoxin stimulates bronchial epithelial cells to release chemotactic factors for neutrophils. J Immunol147199142934301 |
| 28. | Niederman MS, Merrill WW, Polomski LM, Reynolds HY, Gee JBLInfluence of sputum IgA and elastase on tracheal cell bacterial adherence. Am Rev Respir Dis1331985255260 |
| 29. | Cosio MG, Hale KA, Niewoehner DEMorphologic and morphometric effects of prolonged cigarette smoking on the small airways. Am Rev Respir Dis1221980265271 |
| 30. | Bosken CH, Hards J, Gatter K, Hogg JCharacterization of the inflammatory reaction in the peripheral airways of cigarette smokers using immunocytochemistry. Am Rev Respir Dis1451992911917 |
| 31. | Saetta M, Turato G, Facchini FM, Corbino L, Lucchini RE, Casoni G, Maestrelli P, Mapp CE, Ciaccia A, Fabbri LMInflammatory cells in the bronchial glands of smokers with chronic bronchitis. Am J Respir Crit Care Med156199716331639 |
| 32. | Hill AT, Bayley D, Stockley RAThe interrelationship of sputum inflammatory markers in patients with chronic bronchitis. Am J Respir Crit Care Med1601999893898 |
| 33. | Culpitt SV, Maziak W, Loukidis S, Nightingale JA, Matthews JL, Barnes PJEffect of high dose inhaled steroids on cells, cytokines, and proteases in induced sputum in chronic obstructive pulmonary disease. Am J Respir Crit Care Med160199916351639 |
| 34. | Cosio M, Ghezzo H, Hogg JC, Corbin R, Loveland M, Dosman J, Macklem PTThe relationship between structural changes in small airways and pulmonary-function tests. N Engl J Med298197712771281 |
| 35. | Johansen F-E, Pekna M, Natvig Norderhaus I, Haneberg B, Hietela MA, Krajci P, Betsholtz C, Brandtzaeg P. Absence of epithelial immunoglobulin A transport, with increased mucosal leakiness, in polymeric immunoglobulin receptor/secretory component-deficient mice. J Exp Med 1999;190:915–921. |
