INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Sensory neuropeptides are present in human airway nerves, beneath and within the epithelium, around blood vessels and submucosal glands, and within the bronchial smooth-muscle layer (1). Cholinergic nerves constitute the predominant bronchoconstrictor pathway. On the other hand, there is no direct functional adrenergic supply to human airways. The role of the nonadrenergic, noncholinergic (NANC) system in asthma is still poorly understood. A study has shown that immunoreactivity for vasoactive intestinal peptide (VIP) was absent in the airways of patients who died from an exacerbation of asthma (6), but this study has not been confirmed in a small number of mild asthmatic patients (7). Substance P (SP) and tachykinins may be increased in mucosal biopsy specimens from asthmatic patients (8), leading to an increased constrictor mechanism. Levels of SP are increased in sputum and bronchoalveolar lavage fluid (BALF) of asthmatic subjects, and plasma concentrations of neuropeptides may vary during exacerbations of asthma (9). Asthmatic subjects are hyperresponsive to neurokinin A and SP. Thus, the imbalance between sensory neuropeptides has been proposed as playing an important role in the mechanisms of asthma (10, 11), but neuropeptide distribution remains to be elucidated in vivo in asthmatic individuals with severe disease and receiving various treatments.

Although no direct study has confirmed the implication of the cholinergic system in chronic bronchitis, its relative importance was suggested by the efficacy of anticholinergic agents in this condition (12). The role of sensory neuropeptides in chronic bronchitis is far less understood.

We conducted a study to compare the immunoreactivity of various sensory neuropeptides in mucosal biopsies of 16 control subjects, 49 asthma patients, and 13 chronic bronchitis patients. We studied the group of patients with chronic bronchitis in parallel with the other two groups to appreciate the specificity of neuropeptide distribution in asthma. We studied asthmatic patients with disease of variable severity, including 16 patients with the most severe form of asthma, requiring long-term oral corticosteroid treatment, since it has been suggested that neuropeptide immunoreactivity may be impaired only in such patients. We selected the neuropeptides VIP, tachykinins, neuropeptide Y (NPY), and calcitonin-gene-related peptide (CGRP) because of their putative role in asthma.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Forty-nine asthmatic individuals, aged 18 to 71 yr (median and 25th to 75th percentiles: 37 yr and 24 to 54 yr, respectively) were studied. Asthma was defined as previously (13), and all patients had reversible airways disease. The severity of asthma varied from mild to severe (Aas score: 1 to 5) (15), and the patient's FEV₁ ranged from 34% to 100% of predicted values (median and 25th to 75th percentiles: 74%, and 63 to 99%, respectively). None of these subjects was a smoker. None had had any bronchial infection during the previous month. Thirty-three patients had not received inhaled corticosteroids for at least 1 mo and oral corticosteroids for at least 3 mo prior to the study. Sixteen patients were taking high-dose inhaled steroids (1.5 to 2 mg/d of beclomethasone or equivalent) and long-term oral corticosteroids (prednisone equivalent daily dose: 5 to 65 mg/d [median and 25th to 75th percentiles: 25 mg and 17 to 30 mg, respectively]). The duration of their treatment ranged from 1 to 10 yr (median and 25th to 75th percentiles: 3.5 yr and 1.5 to 5.5 yr, respectively).

Thirteen patients with chronic bronchitis, aged 44 to 74 yr (median and 25th to 75th percentiles: 57 yr and 46.7 to 64 yr, respectively) were studied. Chronic bronchitis was defined as previously described in detail according to the criteria of the American Thoracic Society (14). Patients were excluded if they showed a history of allergic diseases, wheezing, an improvement in the FEV₁ of more than 10% after inhalation of 200 µg of salbutamol, or if they had had a bronchial infection during the month preceding the study. FEV₁ ranged from 43% to 90% of predicted values (median and 25th to 75th percentiles: 71% and 59 to 81%, respectively). All of these patients were smokers and did not receive any treatment. All were current smokers at the time of the study.

Sixteen healthy subjects, aged 21 to 62 yr (median and 25th to 75th percentiles: 29.5 yr and 24.5 to 49 yr, respectively) were used as a control group. None of these subjects was a smoker. None had had any bronchial infection during the previous month.

Informed consent was obtained from all subjects prior to the study, and the study was approved by the ethics committee of our hospital.

Investigations

Assessment of the severity of asthma. The severity of asthma was assessed according to the score of Aas (15), which is used to grade chronic asthma from very mild forms (score of 1) to incapacitating disease requiring medications (score of 5). The grading is based on events occurring within the course of a year, and combines symptoms (number and duration of asthma episodes, total duration of symptoms, presence or absence of symptom-free intervals between attacks) and the requirements for medications. It does not take into account the pulmonary function of the patient. Patients receiving antiinflammatory treatment are classified as having a score of 5. In addition, the Aas score is correlated with FEV₁ (13), eosinophilic inflammation (13), and the activation of inflammatory cells pertinent to airways inflammation (16, 17).

A pulmonary function test was done with the same equipment (Pneumoscreen; E. Jaeger Laboratories, Würzburg, Germany) used for all subjects, FEV₁ was measured at entry and at the end of the study. All patients had a pulmonary function test at least 4 h after the last dose of any bronchodilator. Predicted values were measured according to Knudson and colleagues (18).

Assessment of the severity of chronic bronchitis. Chronic obstructive pulmonary disease (COPD) was defined as a disorder characterized by abnormal tests of expiratory flow, and although there is no definitive cutoff limit, patients with FEV₁ under 70% predicted were classified as having COPD (14, 19).

Immunohistochemistry. Fiberoptic bronchoscopy was performed as previously described (13). Biopsy specimens were fixed immediately by immersion in modified Bouin's fixative for 3 to 4 h and then washed for 16 h in phosphate-buffered saline (PBS: 0.01 M phosphate buffer, pH 7.4; 0.15 M NaCl, pH 7.4) containing 15% sucrose and 0.01% sodium azide, and stored at 4° C. Frozen blocks were prepared by embedding the biopsy specimen in mounting medium (ornithyl carbamyltransferase [OCT] compound; Miles Laboratories, UK) on a cork disk and freezing it in melting isopentane previously cooled in liquid nitrogen. Frozen sections (7 µm thick) were collected on poly- L-lysine-coated slides, dried for 1 to 2 h at room temperature, and stained according to a modified immunofluorescence technique as described previously (20). Briefly, the slides were immersed in PBS containing 0.2% (vol/vol) Triton X-100 (BDH, Poole, UK) for 60 min, and were then transferred to a 3% solution of Pontamine Sky Blue (BDH) in PBS for 30 min to reduce background staining and produce a counterstain (21). Three sections were stained per antiserum. Sections were then incubated with diluted primary antisera, the neuronal marker PGP 9.5 (22, 23), and neuropeptides (Table 1) for 16 to 24 h at 4° C, and were washed in three changes (10 min each) of PBS. Primary antisera were reapplied for 4 h at room temperature, and the sections were washed again in PBS (thrice, for 5 min each) and incubated with fluorescein isothiocyanate (FITC)-conjugated sheep antiserum to rabbit IgG (Sigma, Poole, UK) for 60 min at room temperature. After washing again in PBS, slides were coverslipped, using PBS:glycerol (9:1 vol/vol) as a mountant, and were viewed and photographed in a fluorescence microscope (AH-2; Olympus, Tokyo, Japan). Immunostaining controls included controls without the primary antiserum and with its replacement with antiserum absorbed for 16 h with homologous antigen. It seems unlikely that peptide degradation occurred after biopsies were taken, as the fixation process we used is rapid. We have not tested the use of protease inhibitors, but animal studies have suggested that delay in fixation does not decrease neuropeptide immunoreactivity (20).

Sections were screened and graded by two observers (D.S. and A.M.-H.), both of whom were unaware of the specimen type. For each antigen, visual assessments of the density of immunoreactive nerve fibers were graded from zero to four in airway epithelium, submucosa, and smooth muscle. The gradings represent the density of nerves within and around each structure, and therefore give an approximation of the relative densities of nerves among cases. This method has been used previously (7, 24), its reproducibility confirmed, and, inter- or intra-observer disparities were nonsignificant. The reproducibility of nerve density between multiple biopsies from the same patient was not tested in the present study but previous studies have suggested that the correlation is high (34).

Statistical Analysis

Statistical analyses were done with nonparametric tests. The Mann- Whitney U test with Bonferroni's correction was used to compare the different groups.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients' Characteristics

The demographic characteristics of the subjects enrolled in the study are presented in Table 2. The mean ages of asthmatic patients, chronic bronchitis patients, and normal subjects were significantly different. The pulmonary function of normal subjects was within the normal range. Patients with asthma had a significant decrease in FEV₁ (22% to 108% of predicted values, median percent of predicted) by comparison with normal subjects and chronic bronchitis patients. Patients with asthma had a variable severity of the disease (Aas score 1: eight patients; Aas score 2: 16 patients; Aas score 3: six patients; Aas score 4: three patients; and Aas score 5: 16 patients). All patients with a clinical score of 5 had very severe asthma and were being treated with oral corticosteroids. Six patients with chronic bronchitis had an FEV₁ below 70% predicted.

The bronchoscopy procedure was well-tolerated, and only three asthmatic subjects had some chest tightness and required a nebulized salbutamol following the procedure.

Biopsy-specimen Characteristics

The bronchial structure was analyzable in most of the subjects. Epithelium was completely absent in two asthmatic subjects and two normal subjects, but in none of the chronic bronchitis patients. The smooth-muscle layer was not seen in its integrity in the biopsy specimens of 31% of the asthmatic subjects, 50% of the normal controls, and 54% of the chronic bronchitis patients. The submucosal glands were found in the biopsies of 31% of the asthmatic subjects, 31% of the controls, and 69% of the chronic bronchitis patients. The numbers of bronchial vessels and endocrine cells observed were too low for any interpretable analysis. Patients with asthma had a thickened basement membrane and some shedding of the epithelium. Some patients with chronic bronchitis had a metaplastic epithelium.

Distribution of Nerves

There was no significant difference in the immunoreactivity of the general neuronal marker PGP 9.5 among the three groups of subjects within the epithelium or in the smooth muscle area (Figures 1 and 2). In the asthmatic subjects there was no difference in PGP 9.5 immunoreactivity according to Aas score.

View larger version (112K): [in this window] [in a new window]   Figure 1.   Immunostaining in sections of endobronchial biopsies from nonasthmatic control (a, c, e) and asthmatic patients (b, d, f  ). Nerves in the smooth muscle are immunostained for the general neural marker PGP 9.5 to show all types of nerves (a, b), and for the neuropeptides VIP (c, d ) and NPY (e, f  ). All original magnifications: ×120.

View larger version (34K): [in this window] [in a new window]   Figure 2.   Results for sensory neuropeptide immunoreactivity expressed as means ± SD. Statistical analysis was done with the Mann-Whitney U test with Bonferroni's correction; only p < 0.017 is considered significant.

Immunoreactivity for VIP was decreased near the smooth muscle area of asthmatic as compared with normal subjects, but the difference was no longer significant when Bonferonni's correction was applied (Figures 1-3). There was no difference between the immunoreactivity observed in the smooth muscle of normal subjects and that of chronic bronchitis patients. In asthma, there was no difference in VIP immunoreactivity according to the Aas score. Corticosteroid-dependent asthmatic subjects had no decrease in VIP immunoreactivity by comparison with patients with milder asthma.

View larger version (19K): [in this window] [in a new window]   Figure 3.   VIP immunoreactivity in smooth muscle from asthma patients: 1-2 and 3- 4 = Aas score of asthma; CS = corticosteroid- dependent asthma.

Immunoreactivity for NPY was significantly decreased near the smooth-muscle area of asthmatic and chronic bronchitis patients as compared with normal subjects (Figures 1 and 2) (p < 0.006 for asthma and p < 0.005 for chronic bronchitis, Mann-Whitney U test). In asthma patients there was no difference in NPY immunoreactivity according to the Aas score. Corticosteroid-dependent asthmatic subjects had a similar decrease in NPY immunoreactivity by comparison with patients with milder diseases.

Immunoreactivity for the other peptides was not different at the different sites of biopsy in the three groups of subjects (Figure 2). Nerves positive for SP and CGRP were rarely found in the biopsy specimens. The numbers of nerves positive for the C-flanking peptide of neuropeptide tyrosine (C-PON) were similar in asthmatic subjects, normal subjects, and chronic bronchitis patients. In asthmatic subjects there was no difference in NPY immunoreactivity according to the Aas score.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, sensory peptide immunoreactivity was not found to differ greatly in mucosal biopsy specimens of normal subjects, asthma patients, and chronic bronchitis patients. Differences observed were subtle, but we found a trend toward a decrease in VIP immunoreactivity in the airway smooth muscle of asthma patients. Innervation of smooth muscle by NPY-positive nerves in airways of both asthma and chronic bronchitis patients was significantly decreased. These findings were not related to the severity of asthma or chronic bronchitis in either group of patients.

The methods used in the present study for assessing patients and collecting specimens have been largely validated by our group. Because it had been found that patients who died from asthma presented some abnormalities in sensory neuropeptide immunoreactivity (6), we studied a large number of asthma patients with disease ranging from very mild to the most severe form of asthma (corticosteroid-dependent asthma). These patients had received at least 1 yr of regular oral corticosteroid treatment, and some may be described as having been corticosteroid-resistant (25). However, definitions of asthma are not clear enough to have classified these patients perfectly, and the purpose of the study was not the classification of asthma but to study a group of patients with very severe asthma.

The study of bronchial biopsy specimens has improved the understanding of the pathophysiology of asthma, but this approach suffers from some defects that may be of importance in the analysis of this study. Bronchoscopy does not make it possible to study the lower airways where the major defects in asthma may exist. The size of the samples provided by bronchoscopy is small, and all bronchial structures cannot be examined in all specimens. In the present study, however, epithelium was present in most samples, and the smooth muscle was seen in about half of the samples. The small size of the biopsy specimens might be a problem however, as far as quantification of nerves or any other small, heterogeneously distributed structures is concerned, but their use offers two main advantages over larger pieces of tissue obtained postmortem or at surgery. First, the tissue is obtained rapidly, with virtually no ischemia; second, it is fixed very quickly because of its small size and the smaller chance of delays caused by fixative penetration. Consequently, bronchial biopsy reflects the situation in vivo.

Regarding sample size, we have studied the difference in numerical values for nerve density in biopsy specimens and in the airways from which they were obtained, using endobronchial biopsy forceps to sample the carinas of airways from pneumonectomy specimens. We found no difference in the tissue density (immunostained nerve as a percentage of tissue area) of nerves immunoreactive for the general neural marker PGP 9.5 or (in testing for nerves present at lower density) VIP. This shows that biopsy provides an adequate sample for determining nerve density in a given airway; we did not test the relationship of one airway to another (Springall and colleagues, manuscript in preparation).

Numerous neuropeptides have been found in animal and human airways, and their precise localization has been reported (2, 26). They are selectively distributed with specific localizations such as epithelium, smooth muscle, and glands. Some are colocalized, and the distribution and numbers of these neuropeptides are different at different stages of life and during development (2).

An imbalance between the different components of the nonadrenergic, noncholinergic system (NANC) peptidergic system was proposed as underlying some of the mechanisms of bronchial diseases such as asthma and chronic bronchitis (10, 11). SP and neurokinin A (NKA) are potent endogenous bronchoconstrictors, whereas VIP is a potent endogenous bronchodilator. Tachykinins can also induce mucus secretion and plasma extravasation, and have important proinflammatory effects, such as the chemoattraction of eosinophils and neutrophils, adhesion of neutrophils, and stimulation of lymphocytes, macrophages, and mast cells. The tachykinins interact with targets in the airways through specific tachykinin receptors. The NK1 and NK2 receptors have been characterized in human airways both pharmacologically and by cloning. On many occasions, the bronchoactive effects of the tachykinins are masked by their degradation at or near the site of their release (27). Although it appears that the human asthmatic lung may be an environment in which the effects of neuropeptides can be amplified, the role of neuropeptides in the pathogenesis of airway obstruction remains speculative (28). Moreover, the localization of sensory neuropeptides had never been studied in depth in asthma, with only small numbers of subjects tested and studies done on patients at the far extremes of the disease, with either mild asthma (7, 8) or at autopsy (6).

In the present study, VIP immunoreactivity was decreased both in asthma and chronic bronchitis, but the results did not reach significance and there was no correlation with the severity of asthma. The discrepancy between the findings in our study and that of Ollerenshaw and colleagues (6) may be explained at least in part by a rapid degradation of VIP following the death of their patients due to the release of large amounts of peptidases within the patients' inflamed airways. Recently, VIP-like immunoreactivity was found to be similar in the trachea and parenchyma of both asthmatic and nonasthmatic subjects (29). The data gathered in the present study contribute to confirming asthma as a heterogeneous disease with various abnormalities, with a partial loss of VIP-containing nerves as a possible characteristic of some patients. In our study, we failed to find some characteristics for asthmatic subjects lacking VIP. Moreover, some chronic bronchitis patients exhibited an absence of VIP immunoreactivity similar to that of asthma patients, suggesting that if some asthma patients have a VIP deficiency, the same deficiency applies in chronic bronchitis. Moreover, it is unknown whether the VIP deficiency is a cause or a result of the disease. Treatment did not affect the expression and distribution of VIP, as had previously been reported by Ollerenshaw and colleagues (6).

The results of the present study did not confirm that immunoreactivity for bronchoconstrictive neuropeptides (SP, NPY) is increased in asthma or chronic bronchitis. Recently it was observed that measured SP-like immunoreactivity was decreased in tracheal tissue of asthmatic subjects studied at autopsy (29). On the other hand, the release of substance P may be increased in asthma without any increase in immunoreactivity, as has recently been shown (30). In the present study, NPY immunoreactivity was found to be significantly decreased in both asthma and chronic bronchitis. These findings are not in accord with those in studies that have shown an increase in the plasma level of NPY in asthmatic subjects after an exacerbation (9), but NPY may be released into blood from other sites in addition to bronchial nerves. The relation between the number of NPY-containing nerves and the amount of NPY being released is unknown. A relation between a lack of immunoreactivity for NPY within the airways and high plasma levels of NPY is questionable, and the turn over of NPY between the bronchi and the blood is unknown. The normal breakdown of neuropeptides involves peptidases, high levels of which can be found within inflamed airways (27), and neutral endopeptidases have been found to specifically inactivate tachykinins. NPY is one of the most common neuropeptides in humans. It may be involved in the regulation of airway tone (31), and was found to inhibit human tracheal gland-cell secretion (32), to increase pulmonary vascular permeability, and to inhibit neurogenic inflammation by prejunctional inhibition of neuropeptide release from airway sensory-nerve terminals (33). NPY may exert some protective effect in the airways by reducing vascular leakage, as was demonstrated in the nasal mucosa. A relative decrease in NPY, as seen in the biopsy specimens of the patients with chronic bronchitis and asthma in our study, may therefore be partly responsible for the increase in microvascular leakage and hypersecretion that are hallmarks of these diseases.

Our findings confirm a trend toward a reduction in VIP-containing nerves in bronchial biopsy specimens from asthma and chronic bronchitis patients. On the other hand, we did not confirm an imbalance between the bronchodilatory and excitatory NANC systems, finding a concomitant decrease in NPY- containing nerves, whereas the numbers of SP- and CGRP-containing nerves were unaffected by these diseases. These findings show that there are differences between the airways of patients with asthma and those with chronic cough, since the latter have abnormal intraepithelial airway nerves containing increased quantities of CGRP (34). In the present study, most of our patients had coughs, as expected, and we could therefore not discriminate between patients according to their CGRP expression.

Although differences in sensory neuropeptide immunoreactivity were not observed in the large number of highly selected patients in the present study, it is not possible to conclude that sensory neuropeptides are not involved in asthma or chronic bronchitis, since functional abnormalities may occur either through differences in the release of neuropeptides or through differences in their degradation.