INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Bronchial asthma is defined as a chronic inflammatory disorder of the airways with variable airflow obstruction and nonspecific bronchial hyperresponsiveness (NSBH) (1). Pathological examination of asthmatic lungs shows the reconstruction of airway structures, called airway remodeling, including phenomena such as the shedding of bronchial epithelium, goblet cell hyperplasia, thickening of the basement membrane (subepithelial fibrosis) and submucosal tissue, and smooth muscle hyperplasia and hypertrophy (2). This airway remodeling has been speculated to be responsible for NSBH and irreversible airflow obstruction in "chronic asthma" and to be one of the factors that make the treatment of such patients difficult (10). The mechanisms by which chronic airway inflammation in bronchial asthma promotes airway remodeling have not been clarified yet. In idiopathic pulmonary fibrosis, platelet-derived growth factor (PDGF) and insulin-like growth factor I (IGF-I) are considered to have important roles in the remodeling of the alveolar structure, such as in interstitial fibrosis and smooth muscle hyperplasia (11). However, PDGF was reported to make little contribution to the structural changes in asthmatic airways (12).

Epidermal growth factor (EGF) was discovered in 1962 by Cohen as a substance that initiated premature eyelid opening and incisor eruption when injected into the neonatal mouse (13). EGF was characterized as being 53 amino acids long. It is initially synthesized as a prepro-EGF molecule consisting of approximately 1,200 amino acids, which is then processed to EGF through a pro-EGF molecule stage. The EGF receptor is a single polypeptide chain of 1,186 amino acids and has one transmembranous portion. The cytoplasmic domain of the EGF receptor performs as an EGF-regulated tyrosine kinase and is thought to initiate several pathways for signal transduction (14). EGF was initially thought to have a stimulatory effect on epithelial proliferation. In more recent studies, EGF has been revealed to modify airway smooth muscle responses directly (15, 16) and to stimulate smooth muscle proliferation (17).

On the basis of this background, we hypothesized that EGF might play an important role in the pathophysiology of bronchial asthma, especially in airway remodeling resulting in goblet cell and bronchial gland hyperplasia, as well as in smooth muscle hypertrophy and hyperplasia. To examine this hypothesis we studied the expression of EGF and EGF receptor immunoreactivity in human asthmatic airways by immunohistochemical methods.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Airway specimens from seven patients with asthma and eight control subjects were examined. The diagnosis of bronchial asthma was based on the criteria of the American Thoracic Society (18) and the international consensus report on the diagnosis and management of asthma (1). Clinical characteristics of the patients with asthma are shown in Table 1. Atopic predisposition was confirmed by the serum allergen-specific IgE test against common airborne antigens. Three of the patients with asthma died of fatal asthma attacks and four had received surgical lung resection because of solitary lung cancer. The control group was matched with the asthma group by age, sex, and smoking status. All control subjects died of nonpulmonary illnesses and were confirmed to have no pulmonary complications by microscopic examination (Table 2).

Tissue Specimens

We sampled the tissue specimens from three different portions of the airway of each subject: large bronchi, about 1 cm in diameter; small bronchi, about 3 mm in diameter; and peripheral airways. All specimens were fixed with 10% buffered formaldehyde and embedded in paraffin. The blocked tissues were processed to 5-µm sections and put on glass slides. One of these specimens was stained with hematoxylin and eosin for routine pathological evaluation and the others were stained by immunohistochemical methods.

Immunohistochemistry

Before immunohistochemical staining, the sections were deparaffined and rehydrated in xylene and graduated dilutions of ethanol. Immunohistochemistry was performed by the avidin-biotin-peroxidase complex (ABC) method (V ectastain ABC-Elite kit; Vector Laboratories, Burlingame, CA). For EGF immunostaining, rabbit polyclonal anti- human EGF antibody (Wakunaga, Tokyo, Japan) (19) was used. For EGF receptor immunostaining, mouse monoclonal anti-human EGF receptor antibodies (clone C11 for the extracellular domain and clone F4 for the cytoplasmic domain; Cambridge Research Biochemicals, Northwich, UK) (20) were used. Distribution of peroxidase was revealed by incubating the sections in a solution containing 3,3'-diaminobenzidine tetrahydrochloride (DAB). For the morphometry of the bronchial smooth muscle, we used mouse monoclonal anti--smooth muscle actin antibody (clone 1A4; Dako, Glostrup, Denmark) and the avidin- biotin-alkaline phosphatase complex (ABC-AP) method (Vectastain ABC-AP kit; Vector Laboratories). Vector red (Vector Laboratories), which is highly fluorescent, was used as the substrate for alkaline phosphatase. The antibody concentrations, the incubation time, and the other conditions were determined according to the instructions of the supplier and the results of preliminary experiments carried out by the authors. As positive controls for EGF and EGF receptor immunostaining, specimens of the human submaxillary salivary gland were used. For negative controls, nonspecific IgG (rabbit or mouse) or a buffer, instead of the first antibody, was used.

Quantification

To avoid observer bias, all specimens were randomly coded before analysis. In EGF and EGF receptor immunostaining, the staining intensity was semiquantified and expressed as negative (), no clear stain; questionably positive (±), slightly stronger intensity than background stain; definitely positive (+), almost the same intensity as positive controls; and strongly positive (++), more intensity than positive controls (21). For statistical analysis, negative () and questionably positive (±) were considered as negative, and definitely positive (+) and strongly positive (++) were considered positive. According to this classification, we discriminated the staining intensities for bronchial epithelium, basement membrane, smooth muscle bundles, and glands.

For morphometrical analysis of the bronchial smooth muscle, we used a fluorescence microscope and a computerized digital image analyzer (XL-10; Olympus, Tokyo, Japan). At first, the length and thickness of the bronchial basement membrane were measured. The outline of the bronchial smooth muscle was then traced and the area of the smooth muscle was measured under the fluorescence microscope, using the image analyzer. The thickness of the smooth muscle was calculated by dividing its area (SM) by the length of the basement membrane (BM).

Statistical Analysis

The differences in the intensity of the immunohistochemical staining were estimated by the chi-square test or Fisher's exact probability test, as appropriate. The differences in the thickness of the bronchial smooth muscle and basement membrane between patients with asthma and controls were examined by the Mann-Whitney U test. A p value less than 0.05 was considered to be significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Morphological Examination (Hematoxylin and Eosin Staining)

Compared with the control airways, the asthmatic airways showed mild shedding of bronchial epithelial cells, thickening of basement membrane (subepithelial fibrosis), infiltration by inflammatory cells including eosinophils and mononuclear cells, and increases in the numbers of goblet cells and bronchial glands. The airways of subjects who died of asthma attacks (Table 1; patients 1 and 4) showed remarkable narrowing, mucous plugs containing eosinophils, and peeled bronchial epithelial cells in the lumen.

Immunohistochemical Expression of EGF and EGF Receptor

In control airways, EGF immunoreactivity was weakly expressed in bronchial serous glandular cells (Figure 1A), epithelial cells, and few smooth muscle bundles. In asthmatic airways, EGF staining was strongly expressed in epithelial cells, bronchial glandular cells, smooth muscle bundles, and basement membrane (Figure 1B). The levels of expression of EGF immunoreactivity in the control and the asthmatic airways are shown in Table 3. In airway epithelium, the level of expression of EGF immunoreactivity in the peripheral airways was significantly higher in asthmatic airways compared with controls (p < 0.03). In large and small airways, there was no significant difference in the level of expression between asthmatic and control airways. In bronchial glands, the level of EGF staining in asthmatic airways was higher in small bronchi (p < 0.001) than in controls. In airway smooth muscle, the level of EGF staining in asthmatic airways was significantly higher in large (p < 0.03) and small bronchi (p < 0.001) than in controls. Immunohistochemical staining was also observed in the basement membrane of asthmatic airways but not in control airways. In individual total airways, the levels of expression of EGF immunoreactivity in the patients with asthma were significantly higher in glands and smooth muscle than in the controls.

View larger version (118K): [in this window] [in a new window]   Figure 1.   Immunohistochemical staining of EGF in human airways. In the control airway (A), only a few serous glandular cells were weakly stained. However, in the asthmatic airway (B), epithelial cells, basement membrane, glandular cells (arrow), and smooth muscle bundles (Sm) expressed EGF immunoreactivity. Bar = 100 µM.

EGF receptor immunoreactivity was observed only in the bronchial epithelium and glands of control airways (Figure 2A). In asthmatic airways, clear EGF receptor staining was observed in the bronchial smooth muscle and basement membrane in addition to bronchial epithelium and glands (Figure 2B). Table 4 presents the level of expression of EGF receptor immunoreactivity in control and asthmatic airways. The level of epithelial EGF receptor immunoreactivity in asthmatic airways was significantly higher in the small bronchi (p = 0.01) than in corresponding samples from control airways. There was no significant difference in the level of expression in large bronchi and peripheral airways between the two groups. The level of smooth muscle EGF receptor staining in asthmatic airways was significantly higher in the small bronchi (p = 0.01) than in small bronchi from control airways. It was also higher in the large bronchi and peripheral airways of patients with asthma than in controls, but not significantly so. EGF receptor immunoreactivity was also observed frequently in basement membrane from asthmatic airways but rarely in basement membrane from control airways. Basement membrane EGF receptor immunoreactivity in asthmatic airways was positive both for the anti-extracellular domain antibody (clone C11) and for the anti-cytoplasmic domain antibody (clone F4). In individual total airways, the levels of EGF receptor immunoreactivity in the patients with asthma were significantly higher in glands and smooth muscle compared with the controls.

View larger version (122K): [in this window] [in a new window]   Figure 2.   Immunohistochemical staining of EGF receptor in the human airways. In the control airway (A), glandular cells were strongly stained (arrow). In the asthmatic airway (B), epithelial cells, basement membrane (arrow), smooth muscle bundles (Sm), and glandular cells (curved arrow) were stained. Bar = 100 µm.

Quantitative Analysis of Basement Membrane and Airway Smooth Muscles

To examine the structural changes in the subjects of this study, we measured the thickness of the bronchial smooth muscle (Figure 3) and basement membrane. The thickness of smooth muscle of the asthmatic and control airways was (respectively) 1.97 ± 0.09 and 1.51 ± 0.05 in large bronchi, 1.51 ± 0.04 and 1.27 ± 0.11 in small bronchi, and 0.79 ± 0.10 and 0.82 ± 0.07 in peripheral airways (converted to logarithms; mean ± SEM). The thickness of smooth muscle of asthmatic airways was markedly increased in the large (p < 0.001) and small bronchi (p = 0.05) compared with controls. The thickness of the bronchial basement membrane in the large bronchi of the asthmatic airways was also significantly greater than that of the control airways (9.11 ± 0.77 and 4.62 ± 0.97 µm, respectively; p = 0.002).

View larger version (50K): [in this window] [in a new window]   Figure 3.   The thickness of bronchial smooth muscle (area of smooth muscle divided by length of basement membrane) in the three portions of airways. Error bars represent means ± SEM. SM = area of smooth muscle; BM = length of basement membrane; NS = not significant.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Bronchial asthma is recognized to be a disease characterized by transient bronchial smooth muscle contraction evoked by a variety of stimuli. It has become widely recognized that the essential pathophysiology of bronchial asthma is chronic airway inflammation in which a number of inflammatory cells such as eosinophils, mast cells, and lymphocytes play a part (1). On the other hand, bronchial asthma has been known to cause pathological changes in airways such as shedding and damage to the airway epithelium, goblet cell and bronchial gland hyperplasia, thickening and hyalinization of the reticular layer beneath the basal lamina of epithelial basement membrane, and an increase in the bronchial smooth muscle mass (2, 3, 6). These pathological changes, called airway remodeling, have been speculated to cause the airway hyperresponsiveness and irreversible airflow limitation seen in chronic asthma and to be a novel target for the development of anti-asthma drugs (22). The mechanisms by which chronic airway inflammation in bronchial asthma promotes airway remodeling are still unclear. Some chemical mediators such as endothelin 1, histamine, leukotriene C₄, and thromboxane A₂ are also reported to have mitogenetic or modulatory effects on smooth muscle proliferation and are possible candidates for the substance promoting airway remodeling (22). Other candidates include various growth factors. As another pathological state in which proliferation of smooth muscle cells is observed, the mechanism of vascular remodeling in atherosclerosis has been well examined. Now it is known that PDGF, transforming growth factor  (TGF-), and fibroblast growth factor  (FGF-) play important roles in proliferation of vascular smooth muscle cells and the formation of atherosclerotic plaques (23). However, in asthmatic airways, PDGF is expressed only in macrophages, the number of PDGF-positive cells is similar to that in nonasthmatic airways (12), and TGF-₁ expression shows no clear difference between the airways of subjects with asthma and those of smokers with and without chronic obstructive pulmonary disease (COPD) (24). These studies suggest that PDGF and TGF-₁ might not play an important role in the airway remodeling observed in bronchial asthma.

In this study, we tested the hypothesis that EGF plays a role in the airway remodeling of bronchial asthma by examining human asthmatic airways histologically and immunohistochemically. Airway smooth muscle and basement membrane were remarkably thickened in asthmatic airways compared with control airways. The EGF immunoreactivities were expressed strongly in epithelial cells, bronchial glandular cells, and airway smooth muscle bundles of asthmatic airways. In addition, EGF receptor immunoreactivity was also observed in these cells of the asthmatic airways. In control airways, EGF and EGF receptor immunoreactivity was weak and their incidences were low. EGF has potent growth activity for mesenchymal cells, including smooth muscle (17). In the development and growth of rat lung, EGF, TGF-, and the EGF receptor are expressed in epithelial and smooth muscle cells of bronchioles and bronchi (25). In humans, Kasselberg and coworkers found that fetal trachea has immunoreactive EGF-containing cells (26). Kajikawa and coworkers also reported that EGF immunoreactivity was observed in bronchial epithelial cells of the adult human lung (27). These findings suggest the importance of EGF in the growth and repair of airway epithelium and smooth muscle. In addition, EGF is known to have a direct contractile effect on vascular and gastric smooth muscle (17). According to our observations, EGF also has a potent airway smooth muscle contractile effect and EGF induces contraction through the cascade of tyrosine kinase activation, phospholipase-2 activation, arachidonic acid release, and leukotriene (LT) production (15). In addition, Wang and associates observed that the LTC₄ and LTD₄ inhibitor suppresses the airway remodeling observed in a repeatedly antigen-exposed rat asthma model (28). As EGF repairs the epithelial damage and is used as a clinical drug for wound healing, it is possible that the expression of EGF and EGF receptor immunoreactivities in the asthmatic airways simply reflects the repair of damage incurred by asthmatic inflammation. However, overhealing of the damage is speculated to be one of the mechanisms of airway remodeling in asthmatic airways. According to these considerations, the results of our study suggest the possible contribution of EGF to be remodeling of asthmatic airways. However, our results did not eliminate the possibility that other growth factors, such as insulin-like growth factor (IGF), may work on the airway remodeling.

Interestingly, EGF and EGF receptor immunoreactivities were also observed in thickened epithelial basement membrane (or subepithelial fibrosis) of asthmatic airways. Our preliminary study revealed that PDGF-BB and PDGF receptor immunoreactivity was not observed in the basement membrane (data not shown). Some growth factors are reported to have affinity for basement membrane and their immunoreactivities are demonstrated in it (29). The results of our study suggested that EGF also had affinity for the thickened basement membrane of asthmatic airways, and there seems to be a possibility that this deposit of EGF could contribute to airway remodeling. However, it seemed peculiar that the EGF receptor was also observed in basement membrane, which consists of noncellular extracellular matrix. One possible explanation is that shedding of the EGF receptor occurs in asthmatic airways and that shed receptors bind nonspecifically to basement membrane and/or specifically to EGF in the basement membrane. Another possibility is that of an antibody cross-reaction between the EGF receptor and the basement membrane component. To examine these possibilities, we stained the same specimens by the same method with two monoclonal antibodies that recognize extracellular and cytoplasmic domains of the EGF receptor independently (clone C11 and F4, respectively). The patterns of staining with these two antibodies were the same. In addition, the fact that clear staining with anti-EGF and EGF receptor antibodies was rarely observed in control airways weakens the possibility that these antibodies cross-react with an original component of basement membrane. On the basis of these considerations, although it is not possible to eliminate the possibility of antibody cross-reaction completely, it is suggested that whole-molecule EGF receptor exists in the basement membrane. It might be speculated that the cell destruction incurred during chronic inflammation causes the release of EGF receptors from EGF receptor-bearing cells, resulting in the binding of EGF receptors to the basement membrane. However, it seems difficult to find a reasonable explanation simply on the basis of the results of our study, and further investigation is required to understand this phenomenon.

In summary, we examined the expression of EGF and EGF receptor immunoreactivity in the airways of patients with asthma and control patients and found more intense expression of both molecules in the asthmatic airways than in the control airways. These results suggested a possible contribution of EGF to the pathophysiology of bronchial asthma, including airway remodeling.