INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Epidemiologic studies consistently show that asthma and rhinitis are often observed in the same patients (1). Inflammation plays a critical role in the pathogenesis of both rhinitis and asthma, and it is well known that some triggers such as allergy and aspirin are causative factors for both the diseases. Although several studies have shown that rhinitis and asthma are characterized by a similar inflammatory process (4), it is likely that its extent and evolution may differ in nasal and bronchial mucosa. In asthma several histopathologic changes of the bronchial mucosa such as epithelial shedding (10), increased thickness of the reticular basement membrane, and increased eosinophilic infiltration (10, 13) have been widely described and considered as peculiar in the inflammatory process occurring in the disease. However, it is still unknown whether, within the same asthmatic subjects suffering from allergic rhinitis, these inflammatory changes may also occur in the nasal mucosa, and whether they may also be affected by the clinical severity of the disease.

The aim of the present study was therefore to assess and compare three distinct inflammatory parameters such as eosinophilic infiltration, epithelial shedding, and the thickness of the reticular basement membrane in nasal and bronchial mucosa of the same asthmatic subjects suffering from a perennial rhinitis. Nasal and bronchial biopsies were taken using the same forceps from 15 patients who had perennial rhinitis and asthma as well as from six control subjects. Moreover, in order to evaluate whether nasal and bronchial inflammation was affected by the severity of asthma and by the administration of an oral drug, the above-mentioned inflammatory parameters were evaluated in nasal and bronchial biopsies obtained from six patients with severe corticosteroid-dependent asthma.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Fifteen asthmatics (20 to 61 yr of age; median, 59 yr; 19 male) were studied. The patients were included in the study on the basis of a clinical history of asthma as previously described in detail (4), and all had a reversibility of the airway obstruction of 12% of FEV₁ from baseline and an absolute value of 200 ml (14). The severity of asthma was characterized depending on the validated score of Aas (15). All patients had perennial rhinitis and were allergic to house dust mites. Perennial rhinitis was diagnosed as previously published on a history of perennial allergic rhinitis (16) and a positive diagnosis of immediate type hypersensitivity using allergen skin prick tests and serum-specific IgE (CAP System; Pharmacia-Upjohn, Uppsala, Sweden) (4). None of the subjects was a current or a previous smoker. Nasal, bronchial, or oral corticosteroids had been withdrawn for at least 2 mo prior to the commencement of the study and use of nasal or bronchial nedocromil sodium or cromoglycate had been stopped for at least 2 wk. No patient had received long-acting ₂-agonists during the previous month. Inhaled short-acting ₂-agonists were (if necessary) withheld 12 h prior to endoscopy. Antihistamines had been withdrawn for at least 1 wk and none of the patients had received astemizole during the previous 3 mo. This group was designated "untreated asthmatics."

Six corticosteroid-dependent asthmatics (20 to 71 yr of age; median, 59 yr; 19 male) were studied. All had perennial rhinitis and were allergic to house dust mites. Corticosteroid-dependent asthma was defined by the need for a daily oral corticosteroid treatment for the optimal control of asthma symptoms, as previously published (17). The duration of continuous oral corticosteroid treatment was at least 1 yr.

Six normal healthy volunteers 20 to 65 yr of age were used as a control group. Their pulmonary function was within normal range. They were nonallergic and had never suffered from asthma or perennial rhinitis. None of the subjects was a current or a previous smoker.

The research was carried out with the approval of the Ethics Committee of the Hôpital Arnaud de Villeneuve, Montpellier, and informed written consent of the patients.

Fiberoptic Bronchoscopy

Fiberoptic bronchoscopy was carried out as previously described (4). Airway inflammation was assessed by a visual endoscopic scoring system previously published in detail, ranging from 0 to 4 (18). A biopsy was done on a subsegmental bronchus using alligator forceps. Each bronchoscopy was carried out with oxygen and adrenaline readily available, all patients having intravenous access.

Nasal Symptom Score

The following nasal symptoms were assessed before endoscopy: nasal discharge, stuffiness, nasal itching, and sneezing. The severity of each of the five symptoms was scored from 0 to 3 according to a previous study (0, none; 1, mild; 2, moderate; 3, severe). The maximum composite score was 12.

Nasal Endoscopy

Before bronchial endoscopy, the investigator (P.C.) examined the nasal mucosa using the same fibroscope and staged the severity of the disease. There is no well-defined and accepted score for rhinitis alone, so we used the rhinosinusitis score of Lund and Kennedy (20), which studies polyps, discharge, and edema. Absence of any of these is scored 0, polyps in the middle meatus is scored 1, polyps beyond the meatus is scored 2. Clear and thin discharge is scored 1, thick and purulent discharge is scored 2. These endoscopic appearances are assessed bilaterally. The score ranges from 0 to 18.

Nasal biopsies were carried out using the same forceps as used for the bronchi at the level of the inferior turbinate after the bronchial biopsy.

Biopsies

Biopsies were fixed in 10% formaldehyde (pH, 7.2) and embedded in paraffin. Tissue sections 4 µm thick were affixed to microscope slides with Le Page glue, and after dewaxing and rehydration the slides were used for immunohistochemistry. On specimens prepared similarly a monoclonal antibody EG2 was used (Pharmacia Diagnostics AB, Uppsala, Sweden) (4). This is specific for an epitope on cleaved and secreted eosinophil cationic protein. Control slides were treated with unrelated IgG₂ mouse antibody of the same IgG isotype. The reaction was revealed using the alkaline phosphatase-antialkaline phosphatase (APAAP) method (Dako LSAB; Dako, Glostrup, Denmark) following the manufacturer's instructions. Visualization of phosphatase was performed using new fuchsin as a substrate. Reticular basement membrane thickness was measured in hematoxylin-eosin-stained sections from the base of the epithelium to the outer limit of the reticular layer. Thirty measurements at predetermined intervals were taken on each biopsy section to give an average for each case, and the groups were compared. The number of measurements was determined using the method of Sullivan and coworkers (19). The percentage of the reticular basement membrane covered by intact epithelium was calculated for each biopsy by image analysis.

The evaluations were made by two independent pathologists (M.V., P.V.), unaware of the patients' data. The interinvestigator coefficient of agreement was very high (Kappa test = 0.97). All biopsy samples were coded, and sections were studied in a blinded fashion. Cellular infiltration of eosinophils, as assessed by cellular staining, was scored by the number of cells per square millimeter. The number of positively stained cells was counted in a deep zone, defined by a squared eye piece, along the length of the epithelial membrane. The number of eosinophils was evaluated in an area 100 µm beneath the epithelial reticular basement membrane. Eosinophils were measured in at least five high power fields in each biopsy. Because of the desquamation of epithelium, cells present in epithelium were not considered.

In order to assess whether epithelial shedding was an artefact, we studied episialin immunoreactivity. Episialin, the gene product of MUC-1 (20), is expressed by human bronchial epithelial cells (21). It has been shown that episialin overexpression inhibits integrin-mediated cell adhesion to extracellular matrix components (22). More recently, it has been reported that episialin can also reduce cell:cell adhesion by inhibiting E-cadherin-mediated interactions (23). Sections 4 µm thick were dewaxed and washed in Tris-buffered saline (TBS) and then incubated with bovine serum albumin (BSA) 0.5% followed by incubation with the mouse monoclonal antibody (mAb) 214D4 against episialin (kindly provided by Dr. J. Hilkens, Division of Tumor Biology and Department of Pathology, Netherlands Cancer Institute, Amsterdam) (22) at a dilution of 1:32,000. The reaction was revealed using the APAAP method. Control slides were prepared using TBS and a nonrelated antibody instead of the specific primary monoclonal antibody. For each biopsy, the immunoreactivity for episialin was expressed as the percentage of positive stained cells.

Design of the Study and Statistical Analysis

All subjects had nasal and bronchial biopsies that were analyzed for EG2 immunostaining, reticular basement membrane thickness, and epithelial shedding. Episialin immunoreactivity was studied in the nasal and bronchial biopsies of six untreated asthmatics.

Statistical analysis was carried out using nonparametric tests, and results are expressed in medians and 25-75 percentiles. Bonferroni's correction was used and only p values less than 0.02 were considered to be positive.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Demographic Characteristics of Subjects

The demographic characteristics of the subjects are presented in Table 1. Patients with untreated asthma had a variable severity of the disease (Aas score 1, three patients; Aas score 2, five patients; Aas score 3, three patients; Aas score 4, four patients).

Nasal Symptom Score and Endoscopic Assessment of Inflammation

Nasal symptom scores are reported in Table 1. All patients had symptoms or rhinitis, whereas control subjects had no or minimal nasal symptoms. None of the control subjects had an abnormal nasal or bronchial endoscopic score even though two subjects had mild nasal edema (Table 1). All asthmatics had abnormal nasal and bronchial endoscopic scores. Several patients had a severe rhinitis (score  8: six untreated asthmatics and four corticosteroid-dependent asthmatics). There was no significant difference between nasal or bronchial endoscopic scores in untreated and corticosteroid-dependent asthmatics, although patients with corticosteroid-dependent asthma tended to have a more severe score. There was no significant correlation between nasal and bronchial scores in any of the two patient groups.

Biopsies

The number of EG2-positive cells in control subjects and in corticosteroid-dependent asthmatics (Table 2 and Figures 1 and 2) was very low and not statistically different between nasal and bronchial biopsies. In untreated asthmatics, EG2-positive cells were always found in nasal and bronchial biopsies, and their number in both tissues was significantly increased by comparison with control subjects or corticosteroid-dependent subjects (p < 0.001; Mann-Whitney U test). In untreated asthmatics, EG2-positive cell number was higher in bronchial biopsies than in nasal biopsies (p = 0.002; Wilcoxon's W test). In addition, in both nasal and bronchial biopsies of untreated asthmatics many EG2-positive cells were degranulated, as shown by the presence of an extracellular immunostaining because of the eosinophil cationic protein (ECP) release by activated eosinophils.

View larger version (104K): [in this window] [in a new window]   Figure 1.   EG2 immunoreactivity of a nasal (a and c) and a bronchial biopsy (b and d ) taken from the same control (a and b) and untreated asthmatic subject (c and d ). (a and b) In the control subject, epithelium is not shed in the nasal and bronchial biopsy and the EG2 immunoreactivity is absent. (c and d ) In the untreated asthmatic, in the bronchial biopsy epithelium is shed, whereas in the nasal biopsy it is almost intact. Eosinophils (EG2 immunoreactivity) are similar in numbers and in microscopic features. The thickness of the reticular basement membrane is increased in the bronchial biopsy. Original magnification: ×400.

View larger version (23K): [in this window] [in a new window]   Figure 2.   Results of EG2 immunohistochemistry in nasal and bronchial biopsies. Closed circles indicate untreated; open circles indicate corticosteroid (CS)-dependent. Statistical analysis by Wilcoxon's W test.

Basement reticular layer thickness (Table 2 and Figure 3) was similar in nasal and bronchial biopsies from control subjects. In untreated and corticosteroid-dependent asthmatics, the basement reticular layer thickness was significantly increased in bronchial but not in nasal biopsies (p = 0.001; Wilcoxon's W test). In addition, in nasal and bronchial biopsies obtained from corticosteroid-dependent and untreated asthmatics the thickness of the basement reticular layer was significantly increased compared with that in control subjects (nasal biopsies, p < 0.004 and p < 0.008, respectively; bronchial biopsies, p = 0.008 and p < 0.001, respectively; Mann-Whitney U test).

View larger version (25K): [in this window] [in a new window]   Figure 3.   Results of the thickness of the reticular basement membrane in nasal and bronchial biopsies. Closed circles indicate untreated; open circles indicate corticosteriod (CS)-dependent. Statistical analysis by Wilcoxon's W test.

The extent of epithelium shedding in nasal biopsies was not statistically different in the three study groups (Table 2). Epithelial shedding was minimal in bronchi of control subjects and corticosteroid-dependent asthmatics. However, in bronchial biopsies obtained from untreated asthmatics, epithelium shedding was significantly greater than in control subjects or in corticosteroid-dependent asthmatics (p < 0.005; Mann-Whitney U test) (Table 2 and Figures 1 and 4). In addition, in untreated asthmatics, epithelium shedding was significantly increased in bronchial biopsies compared with that in nasal biopsies (p < 0.006; Wilcoxon's W test). Episialin expression was low in nasal epithelial cells and was greatly increased in bronchial epithelial cells in all six untreated asthmatics (Figure 5).

View larger version (25K): [in this window] [in a new window]   Figure 4.   Results of epithelial shedding in nasal and bronchial biopsies. The figure shows the percentage of reticular basement membrane (RBM) covered by intact epithelium. Closed circles indicate untreated; open circles indicate corticosteriod (CS)-dependent. Statistical analysis by Wilcoxon's W test.

View larger version (98K): [in this window] [in a new window]   Figure 5.   Episialin immunoreactivity of nasal and bronchial biopsies from the same untreated asthmatic patient. (a) Episialin immunoreactivity in nasal mucosa. (b) Bronchial biopsy incubated with an irrelevant mAb (internal control). (c) Episialin immunoreactivity in bronchial mucosa. Original magnification: ×400.

There was no significant correlation between the nasal symptom score and inflammatory markers or between the endoscopic score and inflammatory markers.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study has shown that although in control and corticosteroid-dependent asthmatic subjects the eosinophilic infiltration, the thickness of the reticular layer of the basement membrane, and the epithelial shedding do not differ between the bronchial and the nasal mucosa, in untreated asthmatic subjects these three distinct features of bronchial inflammation are significantly enhanced in bronchial mucosa when compared with those in the nasal mucosa. In addition, this study has demonstrated that epithelium shedding may be associated with the expression of the antiadhesion molecule episialin. Finally, in corticosteroid-dependent asthmatic patients, airway inflammation appears to be characterized by a very low number of eosinophils and by a low extent of epithelial shedding.

The clinical characterization of patients with asthma and rhinitis is always a matter of debate. However, in the present study the clinical phenotyping of the subjects was carefully done using the recommended criteria for the definition of asthma, rhinitis, and IgE-mediated allergy. In order to have a relatively homogeneous group of patients, we purposely studied patients with clinical manifestations of both rhinitis and asthma, with nasal and bronchial endoscopic abnormalities, assessed using published scoring systems (3, 18), and allergic to house dust mites only. Most of these patients were also allergic to other allergens. Thus, the results observed cannot be referred to asthmatics without clinical symptoms of rhinitis. The severity of asthma was assessed by the clinical score of Aas, which has been used and validated consistently in our studies for many years. Some untreated asthmatics had a severe form of the disease, although they did not receive any anti-inflammatory treatment, but these patients were seen for the first time in our clinic and were studied thoroughly before appropriate treatment was started.

Eosinophilic inflammation represents a paramount feature of asthma (5) and several studies have shown an increased number of activated eosinophils in the airways of symptomatic asthmatics (4, 24, 25). However, there was no study examining whether these cells could concomitantly infiltrate both the bronchial and the nasal mucosa of asthmatics with perennial rhinitis. In this study we have shown that although in untreated asthmatics eosinophilic infiltration is high in nasal and bronchial mucosa, in bronchial mucosa their numbers are significantly increased when compared with those in the nose. In addition, in untreated asthmatics, eosinophils appear to be activated in both the nasal and bronchial mucosa. It seems, therefore, that nose and bronchi show similar eosinophilic inflammation. Although these cells are usually located beneath the reticular basement membrane and are in an activated state, they can also infiltrate the epithelium layer. However, because of the extensive epithelial shedding in the bronchi of untreated asthmatics, we were unable to count the number of eosinophils in the epithelium. Oral corticosteroids were found to ablate eosinophilic inflammation in both the nose and the bronchi showing that an orally administered drug acts on both tissues.

The pseudo-thickening of the reticular basement membrane is another typical feature of the asthmatic bronchus (26). In many studies, thickening has not been related to bronchial hyperreactivity or to asthma severity (10, 13). Here we show that in nasal and bronchial biopsies obtained from corticosteroid-dependent and untreated asthmatics the thickness of the basement reticular layer is significantly increased by comparison with control subjects. In addition, we have demonstrated for the first time that in asthmatics the thickness of the basement reticular layer is significantly greater in the bronchial than in the nasal one. These data confirm that airway inflammation is almost invariably associated with an increased thickness of the reticular basement membrane regardless of the clinical severity of asthma and suggest that, at least in untreated asthmatics, there are "factors" inducing an increased thickness of the reticular layer of the basement membrane in the bronchi (26, 27) that are not present in the nose. Among these factors an important role may be played by the activation of mesenchymal cells such as fibroblasts and myofibroblasts, as well as by the proteases and antiproteases balance controlling the extracellular matrix turnover, which may differ in the nose and bronchi.

Epithelial shedding, another prominent histopathologic feature of airway inflammation in asthma, has been found to be related to the severity of the disease (10, 12, 28). In vitro and animal studies have shown the deleterious effect of the eosinophil granule-derived proteins ECP and major basic protein (MBP), which are cytotoxic for the bronchial epithelium (29, 30) and affect cilia. In this study, bronchial epithelial shedding was seen only in untreated asthmatics. On the other hand, nasal epithelial cells were not shed, even in untreated asthmatics. Although eosinophils are in greater number in the bronchial mucosa than in the nasal mucosa, it seems unlikely that eosinophils represent the only mechanism involved. However, we do not have enough evidence on the activation of the cells that may significantly differ in the two tissues. This study has suggested, however, that the mechanisms regulating inflammation are different in the bronchi from those in the nose. It has been suggested that shedding of the epithelium can be caused by plasma exudates leaking through vascular endothelial gaps (31) as a consequence of the release of several inflammatory mediators such as eosinophil granule proteins, oxygen free radicals, tumor necrosis factor-alpha (TNF-) (32), or metalloproteases from epithelial cells (33) or macrophages (34). Furthermore, the increased epithelial fragility and shedding may also be due to a weakened attachment of superficial epithelial cells to basal cells or to their reticular basement membrane; this probably reflects a disturbance in cell:cell adhesion (35). To better understand this point we have evaluated the expression of the antiadhesion molecule episialin, which has been shown to inhibit integrin-mediated cell adhesion to extracellular matrix components causing a dramatic reduction in cell adhesiveness (22). Episialin is mainly responsible for the inhibition of E-cadherin-mediated cell:cell interactions (22). A higher expression of episialin was found in the bronchial epithelium than in the nasal one, suggesting that other mechanisms than a simple eosinophilic inflammation may be involved in bronchial epithelial shedding.

In conclusion, several concomitant and differential histopathologic changes occur in the bronchial and nasal mucosa of asthmatic patients with perennial rhinitis, suggesting that the nasal and bronchial mucosa behave differently and are not identical.