INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nasal polyposis (NP) is a chronic eosinophilic inflammatory disorder of the nasal and paranasal sinuses, leading to the formation of benign polyps (1). NP is frequently associated with lower airway diseases, such as asthma, which is found in approximately 50% of cases (2). Beside this finding, when systematically assessed, nonspecific bronchial hyperresponsiveness (BHR) is commonly measurable in nonasthmatic patients with NP (3, 4). The significance and the outcome of asymptomatic BHR associated with NP is unknown (5). It has been proposed that asymptomatic BHR might represent a preliminary stage before the development of asthma (6). We have reported that a subclinical bronchial inflammation consisting of eosinophils and T lymphocytes was present in patients with NP and asymptomatic BHR. Although less pronounced, this bronchial inflammation looked similar to that observed in patients with NP and asthma (7). By contrast, patients with NP without BHR and/or asthma did not feature subclinical bronchial inflammation.

Airway eosinophilic inflammation has emerged as an important contributor to the mechanisms of nonspecific BHR in asthma through the release of toxic and proinflammatory eosinophilic mediators that cause tissue damage and airway structural changes. Among the large variety of inflammatory mediators involved in asthma, two are believed to play a key role in the selective recruitment and the activation of eosinophils, namely, interleukin 5 (IL-5) and eotaxin. In vitro, IL-5 properties with respect to eosinophils include chemoattraction, transendothelial migration, cytotoxic cationic protein release, and prolongation of eosinophil survival (8). In addition, studies of animal models have demonstrated the pivotal role of IL-5 in selective eosinophil recruitment, which precedes the induction of BHR. Topical administration of IL-5, either intranasally or directly into the airways, induces a selective increase and activation of eosinophils in the airways, resulting in the development of airway hyperreactivity (9,10). Similarly, other reports have shown that neutralizing antibodies to IL-5 suppress eosinophil accumulation and subsequently abolish hyperreactivity in the lungs of experimental animals during allergic inflammation (11, 12). Eotaxin, a C-C chemokine, strongly stimulates eosinophil chemotaxis, induces an elevation in eosinophil intracellular calcium levels and the respiratory burst activity and is involved in the early phase of eosinophil recruitment after allergen challenge (13, 14). There is also substantial clinical evidence linking IL-5 and eotaxin with airway eosinophilia and BHR in asthma because increases in both IL-5 and eotaxin mRNA and protein expression were reported in the airways of patients with asthma, which correlated with eosinophil number in bronchial tissue and inversely correlated with the level of BHR (15). Finally, IL-5 and eotaxin are thought to act in cooperation to regulate lung eosinophil recruitment and eosinophil activation and the subsequent development of BHR (19).

The mechanisms underlying the pathogenesis of asymptomatic BHR are not understood. Because patients with NP and asymptomatic BHR and patients with NP and asthma share pathological similarities in terms of inflammatory cellular infiltrate, it can be hypothesized that the difference in the clinical patterns results from differences in other factors, such as the degree of cellular activation or the release of specific mediators such as IL-5 and eotaxin. To test this hypothesis, we investigated bronchial IL-5 mRNA and protein expression, eotaxin protein expression, and the state of eosinophil activation in bronchial biopsies and IL-5 in bronchial lavage (BL) fluid from patients with NP and correlated these changes to the presence of asymptomatic BHR and/or asthma.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

We studied 28 patients (17 male and 11 female patients) consecutively referred to Calmette Hospital (Lille, France) for pulmonary evaluation of noninfectious nasal polyposis. The mean ± SEM age was 41.4 ± 2.4 yr. Nasal polyps were identified in all patients by anterior rhinoscopy. Nasal symptoms associated with nasal polyps (obstruction, anosmia, sneezing, rhinorrhea, itching) were assessed. Nasal symptoms were recorded by the patients during the 2 wk before the study. Each nasal symptom was scored from 0 to 3: 0 for no symptoms, 1 for mild symptoms (just noticeable), 2 for moderate symptoms (annoying), and 3 for severe symptoms (distress), so that maximal nasal score was 15 of 15. All patients were treated with topical steroids (beclomethasone, 200 to 600 µg/d). Patients with infectious sinusitis and/or a previous history of nasal surgery were excluded from the study.

Study Protocol

The study protocol was approved by a local ethics committee and informed consent was obtained from all the patients. Subjects attended the laboratory on two different occasions, 2 d apart. On the first visit, subject characteristics were documented by a general questionnaire on upper and lower respiratory symptoms, history of atopy, asthma, and medications required for nasal polyposis and/or asthma. Skin prick tests for common inhalant allergens (house dust mite, animal danders, grass pollen, tree pollen, and molds) and spirometric measurements were performed. Serum total IgE was measured by radioimmunosorbent test (Pharmacia Diagnostics, Uppsala, Sweden). On the second visit, subjects underwent methacholine inhalation challenge and a fiberoptic bronchoscopy for bronchial lavage and bronchial biopsies (within 3 h of methacholine challenge). Three patients had positive skin tests to seasonal allergens; they underwent the methacholine inhalation challenge and the fiberoptic bronchoscopy during the period without pollen exposure. Previous reports have demonstrated that methacholine inhalation does not alter the airway cellularity in healthy subjects and in subjects with asthma (23).

Definitions

Asthma was defined according to the criteria suggested by the American Thoracic Society (ATS) (24). The asthma severity was assessed according to the Aas scoring system (from score 1 for very mild forms to score 5 for an incapacitating disease) (25). Exclusion criteria for patients with asthma included lower respiratory tract infection in the previous 8 wk, current smoking habit, asthmatic exacerbation in the preceding 8 wk, and treatment with inhaled or oral corticosteroids in the 4 mo before the study. Asymptomatic bronchial hyperresponsiveness (BHR) was defined as a methacholine PC₂₀  16 mg/ml in the absence of symptoms suggestive of asthma (intermittent wheeze, chest tightness, dyspnea, recurrent cough) in subjects who did not report a previous history of childhood asthma and who never required any asthma medication (26). Atopy was defined as the presence of more than one positive skin prick reaction to the common inhalant allergens, that is, wheal diameter at 15 min more than 3 mm larger than the negative control.

Pulmonary Function Tests and Methacholine Inhalation Test

All measurements were made in the sitting position. FEV₁, FVC, and maximal forced midexpiratory flow (FEF_25-75) were obtained from flow-volume curves, using a spirometer (Medgraphics, St. Paul, MN). All measurements were recorded before fiberoptic bronchoscopy. The largest values of FVC, FEV₁ and FEF_25-75 from the first three technically satisfactory forced expirations were selected. All data were expressed in absolute values and as a percentage of predicted normal values (27).

Methacholine challenge was performed sequentially with a DeVilbiss 646 nebulizer (output of 0.13 ml/min) (26). Patients inhaled first, for 2 min of tidal breathing, the diluent solution followed by doubling concentrations of methacholine at intervals of 5 min, starting with 0.125 mg/ml and ending with 16 mg/ml. The procedure was ended when a 20% fall in FEV₁ from the postdiluent value was obtained or when a methacholine concentration of 8 mg/ml was reached. The lowest, technically satisfactory FEV₁ value obtained between 90 s and 3 min was used in the analysis to calculate the dose-response curve. Dose- response curves were constructed by plotting FEV₁ against increasing concentration of methacholine as a log scale. The provocative concentration (PC₂₀) was defined as the methacholine concentration necessary to decrease FEV₁ by 20% from baseline values. BHR was defined as a PC₂₀  16 mg/ml.

Assessment of Airway Inflammation

Fiberoptic bronchoscopy. The bronchoscopy procedure was performed at the same time of day (11:00 AM). Bronchoscopy was performed according to published guidelines (28). After local anesthesia of the throat, larynx, and bronchi with 2% lidocaine, a flexible bronchoscope (OES 10 fiberscope; Olympus, Melville, NY) was introduced into the bronchial tree and gently wedged into a segmental bronchus of the right middle lobe. One 50-ml aliquot of normal saline (37° C) was instilled and aspirated with a syringe via the bronchoscope channel. This is referred to as the bronchial lavage (BL). Bronchial biopsies were taken from segmental divisions of the main bronchi after completion of the lavage. During bronchoscopy, oxygen was readily available and the patient had an intravenous infusion to provide venous access. Satisfactory endobronchial biopsies and BL fluid were obtained from all patients.

Examination of bronchial lavage fluid. BL fluid was filtered through sterile surgical gauze to remove mucus. The cells from BL fluid were pelleted by centrifugation at 200 × g for 10 min at 4° C. The supernatant was decanted and stored at  70° C for later analysis. The cells were resuspended in RPMI 1640 and kept on ice until processed. Lavage fluid (10 µl) was used to determine the total cell count, using a standard hemacytometer. Aliquots of the BL (200 µl) were spun into a cytocentrifuge and stained with eosin/methylene blue (RAL, Paris, France) for differential cell counts. At least 400 cells on each slide were read by two investigators blinded to the clinical details of the patients. Another two cytospins were stained with toluidine blue and 1,500 cells were counted for metachromatic cells. Average cell differentials of the two investigators are reported. Macrophages, neutrophils, lymphocytes, eosinophils, mast cells, and epithelial cells were enumerated and results are given both in percentage and number of cells per milliliter of fluid recovered. Eosinophil cationic protein (ECP) was assayed in BL supernatants in duplicate with specific and sensitive radioimmunoassays (Pharmacia, St Quentin Yvelines, France). Histamine and interleukin 5 were measured in bronchial lavage fluid supernatant by enzyme-linked immunosorbent assay (ELISA) (Immunotech, Marseille, France). Results are expressed as nanograms per milliliter, nanomolar units, and picograms per milliliter for ECP, histamine, and IL-5, respectively. Lower limits of detection were, respectively, 2 ng/ml, 0.5 nM, and 1 pg/ml for ECP, histamine, and IL-5, respectively.

Biopsy sampling and processing. One biopsy was immediately snap frozen in liquid nitrogen for immunohistochemistry analysis. A set of specimens was immediately fixed in freshly prepared 4% paraformaldehyde in phosphate-buffered saline (PBS), and then dehydrated and further processed for paraffin embedding. Another set of specimens was immediately fixed in 1% glutaraldehyde in cacodylate buffer for 1 h at room temperature, washed three times in cacodylate buffer, and further processed for immunoelectron microscopy.

Cytokine protein immunochemistry. Frozen sections were saturated in 1% bovine serum albumin in PBS and incubated for 1 h at 4° C with a primary antibody directed against IL-5 (polyclonal rabbit antibody from Genzyme, Cambridge, MA) at a dilution of 1:250, against eotaxin (R&D Systems, Abingdon, UK) at a dilution of 1:250, against eosinophil peroxidase (EPO; Oncogene Sciences, Uniondale, NY) at a dilution of 1:100, against tryptase (monoclonal mouse anti-human mast cell tryptase; Dako, Copenhagen, Denmark) at a dilution of 1:50, and against human T cell CD3 (monoclonal mouse antibody; Dako). The binding was detected with streptavidin complex (LSAB2 kit, alkaline phosphatase; Dako). Controls included omission of the first antibody and substitution of the first antibody with a nonspecific antibody. For each section, the numbers of positively stained cells were counted in at least three nonoverlapping high-power fields until the available area from the reticular basement membrane to a depth of 125 µm was covered. At least three sections were stained for each cellular marker. The results were expressed as the number of cells per millimeter of basement membrane. To avoid observer bias, the slides were coded before analysis and read blind.

In situ hybridization. Cryostat sections (10 µm) were cut from paraformaldehyde (PFA)-fixed biopsies and air dried overnight at 37° C. Sections were hybridized with antisense probes (complementary to mRNA) for IL-5. Sense probe (sequence identical to mRNA) for IL-5 was used as negative control for each biospy. At least three sections were hybridized with the probes. The method has been described previously (16). As an additional negative control, a separate set of sections was pretreated with RNase A solution at 37° C for 30 min before the prehybridization step. Hybridization was subsequently performed with labeled cRNA probes as described above. No hybridization signals were observed in any of the controls, thus confirming the specificity of the results obtained with antisense cRNA probes. Specific hybridization was recognized as clear dense deposits of silver grains in the emulsion overlying cells in the tissue preparations. Cells were identified as dense, discrete, well-circumscribed areas of silver grains. The results were expressed as the number of cells per millimeter of basement membrane.

Ultrastructural study. Specimens rinsed in cacodylate buffer were dehydrated, fixed in osmium tetroxide, and embedded in Lowicryl K4M (Polysciences, Warrington, PA). On semithin sections, areas containing inflammatory cells were selected. The pathological analysis focused on fine cellular alterations, that is, modifications of intracytoplasmic granules, presence of extracellular granules, and/or nuclear lysis with cytoplasmic damage.

Statistical Analysis

Group data are expressed as means ± SEM. Most of the data were not normally distributed. For these data, statistical analysis was performed by the nonparametric Kruskall-Wallis test for differences among groups. When this test indicated a significant difference, each pairing was examined by means of the Mann-Whitney U test. For normally distributed data (FEV₁ %pred, FVC %pred, FEV₁/FVC, and FEF_25-75 %pred), analysis of variance (ANOVA) was applied to examine differences between the groups. When this test indicated a significant difference, each pairing was examined by Student t test. Differences in sex, atopy, and aspirin intolerance were examined by the Fisher exact test. The correlations were investigated with the Spearman rank correlation coefficient test. A probability value of less than 0.05 was taken as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Clinical and Physiological Data

We studied 28 patients with nasal polyposis (NP). Clinical characteristics of the patients are shown in Table 1. According to clinical history and bronchial response to methacholine inhalation challenge, three groups of patients were identified: 11 patients did not exhibit BHR (Group A), 8 patients with NP had asymptomatic BHR (Group B), and 9 patients with NP exhibited BHR associated with asthma (Group C). In Group C, severity of asthma was mild; patients with asthma were classified by Aas score 1 (n = 4) or 2 (n = 5). Nasal score was similar in the three groups. Pulmonary function characteristics of the three groups are reported in Table 2. Patients in Group C exhibited a lower FEV₁ (liters), FEF_25-75, and FEV₁/FVC ratio than patients in Group A. FEV₁ (liters) and FEF_25-75 were lower in Group C than in Group B. Methacholine PC₂₀ ranged between 0.25 and 16 mg/ml in Group B and between 0.04 and 9.6 mg/ml in Group C. 

Cellular Characteristics of Bronchial Biopsies and Bronchial Lavages

Data concerning the characteristics of the cellular infiltrate of the bronchial biopsies and the cellular profile of BL fluid were reported for 20 of the 28 patients in our previous study (7). Pathological analysis of bronchial biopsies demonstrated that the number of EPO⁺ eosinophils and the number of CD3⁺ T cells in bronchial submucosa in Group C and in Group B were significantly increased, as compared with Group A (Table 3). In addition, the number of eosinophils and T cells was significantly higher in Group C than in Group B. The number of mast cells was similar in the three groups.

The total and differential cell counts in the BL fluid are reported in Table 3. There was no difference between groups in BL fluid recovery (data not shown). The mean percentage of eosinophils in Group B and in Group C was significantly greater than in Group A. The percentage of macrophages, neutrophils, lymphocytes, mast cells, and epithelial cells in each group was not significantly different. Although this difference did not reach significance, ECP levels tended to be higher in BL fluid in Group C than in Group A and in Group B. Histamine did not differ between the groups.

Eosinophils Morphological Analysis

Morphological analysis of eosinophils revealed that the majority of eosinophils in Group C were activated because EPO immunostaining was located both within eosinophil cytoplasm and extracellularly (Figure 1, right). By contrast, eosinophil morphology was not altered in Group B with EPO immunostaining strictly located within eosinophil cytoplasm (Figure 1, left). Ultrastructural analysis of the bronchial biopsies in Group C showed numerous eosinophils associated with lymphocytes in bronchial submucosa. The fine structure of eosinophils was altered. Nuclear lysis associated with cytoplasmic damage was frequently observed. Most intracytoplasmic granules had an inverted density of the central core and numerous free extracellular eosinophil granules showed similar alterations. By contrast, eosinophil structure in Group B was not altered; nearly no eosinophil nuclear lysis was observed in bronchial biopsies from Group B. 

View larger version (92K): [in this window] [in a new window]   Figure 1.   Representative EPO immunoreactivity of the bronchial submucosa (Indirect immunophosphatase ×400). Bronchial biopsy from a representative patient with nasal polyposis and asymptomatic BHR. EPO immunostaining is strictly located within eosinophil cytoplasm (left). Bronchial biopsy from a representative patient with nasal polyposis and asthma (right). EPO immunostaining is located both within eosinophil cytoplasm and extracellularly.

Interleukin 5 Expression in Bronchial Biopsies and IL-5 Levels in Bronchial Lavages

Immunostaining for IL-5 showed an increased number of IL-5-positive cells in bronchial submucosa in Group C (16.7 ± 1.6 cells/mm) and in Group B (9.4 ± 1.5 cells/mm) as compared with Group A (0.9 ± 0.5 cells/mm (p < 0.0005) (Figure 2A). The number of IL-5-positive cells was also higher in Group C than in Group B (p < 0.01) (Figures 2A and 3).

View larger version (25K): [in this window] [in a new window]   Figure 2.   Bronchial IL-5 and eotaxin expression in patients with nasal polyposis. (A) Number of IL-5-immunoreactive cells in the bronchial submucosa of patients with NP without BHR (n = 11), patients with NP and asymptomatic BHR (n = 8), and patients with NP and asthma (n = 9). Results are expressed as number of cells per millimeter of basement membrane. (B) Number of IL-5 mRNA-positive cells by in situ hybridization in the bronchial submucosa of patients with NP without BHR (n = 4), patients with NP and asymptomatic BHR (n = 5), and patients with NP and asthma (n = 5). Results are expressed as number of cells per millimeter of basement membrane. (C) Number of eotaxin-immunoreactive cells in bronchial submucosa submucosa of patients with NP without BHR (n = 9), patients with NP and asymptomatic BHR (n = 5), and patients with NP and asthma (n = 7). Results are expressed as number of cells per millimeter of basement membrane. (D) IL-5 levels in bronchial lavages of patients with nasal polyposis without BHR (n = 11), patients with NP and asymptomatic BHR (n = 8), and patients with NP and asthma (n = 9). NP = nasal polyposis; BHR = bronchial hyperresponsiveness. Probability value of Mann-Whitney U test: *p < 0.0005; p < 0.01; p < 0.05.

View larger version (94K): [in this window] [in a new window]   Figure 3.   Representative IL-5 immunoreactivity of the bronchial submucosa in patients with nasal polyposis. Bronchial biopsy from a patient with nasal polyposis without BHR, immunostained with anti-IL-5 antibody, showing essentially no cellular staining (left; Indirect immunophosphatase ×100). Bronchial biopsy from a patient with nasal polyposis and asymptomatic BHR (right), showing broad distribution of immunostaining of submucosal cells (Indirect immunophosphatase ×400).

By in situ hybridization, the number of IL-5 mRNA-positive cells was higher in Group C (15.6 ± 1.6 cells/mm) than in Group B (6.8 ± 1.4 cells/mm) and in Group A (3.5 ± 0.6 cells/ mm) (p < 0.01) (Figure 2B). The difference for IL-5 mRNA-positive cells between Group B and Group A did not reach significance.

Levels of IL-5 in BL fluid were increased only in BL fluid from Group C as compared with BL fluid from Groups A and B (Figure 2D).

Eotaxin Expression in Bronchial Biopsies

Immunostaining for eotaxin (Figure 2C) revealed an increased number of eotaxin-positive cells in the bronchial submucosa of patients in Group C (13.9 ± 3.8 cells/mm) as compared with Group A (0.1 ± 0.1 cells/mm) (p < 0.001) and Group B (2.5 ± 1.9 cells/mm)(p < 0.01).

Correlations

In bronchial biopsies, the number of IL-5 protein⁺ cells and the number of IL-5 mRNA⁺ cells were positively correlated (p = 0.005, r = 0.79) when considering the 28 patients (Table 4). Positive correlations were also found between the number of IL-5 protein⁺ cells, IL-5 mRNA⁺ cells, eotaxin⁺ cells, and the number of CD3⁺ T cells and EPO⁺ eosinophils when considering the 28 patients. Taking only patients of groups B and C, similar significant correlations were found, except for IL-5 protein⁺ cells and FEF_25-75. There was a negative correlation between IL-5 mRNA⁺ cells and PC₂₀ (p < 0.05, r =  0.70) (Figure 4) in patients of Groups B and C. 

View larger version (20K): [in this window] [in a new window]   Figure 4.   Correlations between IL-5 (protein and mRNA)positive cells and bronchial hyperresponsiveness (methacholine PC20) in patients with nasal polyposis and asymptomatic BHR (Group B) and in patients with nasal polyposis and asthma (Group C). Correlations were conducted with the Spearman rank correlation test.

In bronchial lavages, levels of IL-5 positively correlated with the percentage of eosinophils (p < 0.05, r = 0.43), EPO⁺ eosinophils (p < 0.05, r = 0.43), CD3⁺ T cells (p < 0.05, r = 0.48), eotaxin⁺ cells (p = 0.0015, r = 0.71), IL-5 protein⁺ cells (p < 0.01, r = 0.51) in bronchial biopsies when considering the 28 patients. Taking only patients of Groups B and C, IL-5 levels in BL fluid positively correlated with IL-5 protein⁺ cells in bronchial biopsies (p < 0.05, r = 0.63).

No significant correlation was found when each group was examined separately.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, we have demonstrated that IL-5 expression was enhanced in lower airways of both patients with NP and asymptomatic BHR and patients with NP and asthma, but not in patients with NP without BHR. On the other hand, increased levels of IL-5 in BL fluid, increased expression of eotaxin, and activation of eosinophils were evidenced only in lower airways of patients with NP and asthma but not in patients with NP and asymptomatic BHR. These findings suggest that (1) IL-5 contributes to bronchial eosinophilia in patients with NP and asymptomatic BHR and in patients with NP and asthma, and (2) IL-5 cooperates with eotaxin to induce eosinophil activation in patients with NP and asthma.

The mechanisms underlying the pathogenesis of asymptomatic BHR are not fully understood. Studies have provided evidence in favor of the presence of a subclinical bronchial eosinophilic process associated with BHR in upper airway diseases such as allergic rhinitis, nonallergic rhinitis with eosinophilic syndrome (NARES), and NP (7, 29, 30). The clinical relevance of the presence of asymptomatic BHR is unknown. It has been proposed that it could represent a risk factor for the development of subsequent asthma (31, 32). No data concerning the role of IL-5 and eotaxin in the development of asymptomatic BHR are available. Thus, it was our aim to examine the bronchial expression of IL-5 and eotaxin in lower airways of subjects with NP. Our findings of increased IL-5 bronchial expression in subjects with asymptomatic BHR and subjects with asthma but not in subjects without BHR, which inversely correlated with the degree of BHR, suggest that IL-5 is involved in the occurrence of (a)symptomatic BHR associated with NP. Considering the known properties of IL-5 in eosinophil recruitment, the increased IL-5 expression may account for the eosinophilic airway inflammation in the airways of subjects with asymptomatic BHR. It has been previously demonstrated that T lymphocytes are not capable of storing cytokine protein and of accumulating sufficient protein to be detectable by immunostaining (33). By contrast, T lymphocytes were shown to be the main source of IL-5 mRNA by in situ hybridization and may secrete large amounts of IL-5 in vitro. In the present study, IL-5 mRNA expression and IL-5 immunoreactivity strongly correlated with eosinophil and T lymphocyte counts in the bronchial submucosa, suggesting that T lymphocytes and eosinophils represent the main sources of IL-5 in lower airways of patients with NP. Mast cells may account for a substantial secretion of IL-5. However, only a few mast cells were found in bronchial biopsies from patients with NP and IL-5 mRNA expression and IL-5 immunoreactivity did not significantly correlate with mast cell count. These findings suggest that mast cells are not central to IL-5 expression or to the inflammatory process evidenced in the lower airways of patients with NP and BHR.

On the other hand, the comparison between subjects with NP and asymptomatic BHR and subjects with NP and asthma disclosed definite discrepancies despite their sharing similarities in terms of bronchial cellular infiltrate. First, the number of IL-5 immunoreactive and IL-5 mRNA-positive cells was significantly lower in subjects with asymptomatic BHR than in subjects with asthma. Second, the levels of IL-5 in the BL fluid of subjects with asymptomatic BHR was low, close to that observed in subjects without BHR, whereas subjects with asthma exhibited increased amounts of IL-5 in bronchial lavage. Third, the number of eotaxin-positive cells was lower in the airways of subjects with asymptomatic BHR. Fourth, activated eosinophils were mostly evidenced by EPO immunostaining and ultrastructural examination in the airways of patients with asthma whereas most tissue eosinophils were in a resting state in patients with asymptomatic BHR. All together, these findings demonstrate that, although close, asymptomatic BHR and symptomatic asthma are in fact two distinct entities. It seems that eosinophils that accumulate into the lower airways of NP subjects with asymptomatic BHR are not activated and do not release proinflammatory mediators.

The reason why eosinophils accumulate but are not activated in the airways of subjects with asymptomatic BHR may be due to the weak expression of eotaxin in the airways of subjects with asymptomatic BHR, in comparison with subjects with asthma. Eotaxin has been shown to be a potent activator of eosinophil activation (13, 14). Moreover, the cooperative role between IL-5 and eotaxin in the recruitment and activation of eosinophils has been expanded in more recent studies. The authors demonstrated that the effects of locally injected eotaxin on eosinophil recruitment into the skin of mice or guinea pigs was amplified by the intraveinous administration of IL-5 (19,21). Furthermore, intranasal or subcutaneous application of eotaxin resulted in recruitment of eosinophils into the skin and lungs only in IL-5 transgenic mice, which constitutively expressed elevated levels of IL-5, and not in wild-type mice (20). Mould and coworkers reported that gene tranfer of IL-5 and/or eotaxin to the lungs of naive mice induced a selective airway eosinophilia, which was amplified by the coexpression of both IL-5 and eotaxin (22). However, for eosinophil degranulation and BHR to occur, inhalation of the antigen was required, supporting the fact that eosinophils alone are not sufficient to induce BHR. The authors further demonstrated that eosinophil degranulation and the development of BHR were CD4⁺ T cell dependent. Taking into account these data, we may hypothesize that eotaxin is needed to cooperate with IL-5 to induce both recruitment and activation of eosinophils in the airways of subjects with NP and asthma. The role of T lymphocytes, which may provide additional signals for eosinophil degranulation, remains to be investigated.

One can hypothesize that the nasal disease may, at least in part, determine the occurrence of bronchial inflammatory modifications in patients with NP. There are numerous reports of significantly increased levels of IL-5 in nasal lavage or within the stroma of patients suffering from NP, suggesting that IL-5 and eotaxin are involved in selective recruitment and prolonged survival of eosinophils in nasal polyps (34). Some authors have proposed that the release of proinflammatory mediators at the nasal level could result in the occurrence of concomitant chronic upper and lower airway disorders in patients with NP. Conversely, Braunstahl and coworkers demonstrated in patients with allergic rhinitis that, after segmental bronchial provocation, allergic response was widespread and included systemic eosinophilia and nasal inflammation (eosinophil and IL-5), supporting the link between the nose and the bronchi (38). Further work on concomitant assessment of nasal and bronchial products from subjects with NP is needed to clarify whether or not nasal cytokine production regulates bronchial inflammatory processes.

In summary, this study provides evidence of enhanced bronchial IL-5 expression in subjects with NP and asymptomatic BHR and in subjects with NP and asthma, suggesting that IL-5 plays a role in the pathogenesis of (a)symptomatic BHR associated with NP. The differences between subjects with NP and asymptomatic BHR and subjects with NP and asthma result from differences both in the number and degree of activation of recruited inflammatory cells, from eotaxin expression, and from the amount of IL-5 secretion in the lower airways. Further studies are needed to assess whether asymptomatic BHR associated with NP translates into overtly symptomatic asthma.