Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Mechanical injury or exposure of the airway epithelium to airborne pollutants, e.g., cigarette smoke or NO₂, is frequently accompanied by a wave of cell proliferation. This activity initially leads to a proliferative and metaplastic epithelium that subsequently returns to a nearly normal, pseudostratified morphology within 5 to 10 d (1, 2). Many studies have evaluated the proliferative events and morphologic changes during the regeneration following epithelial injury (3), but little is known about mechanisms that determine selection of hyperplastic cells which remain to form the remodeled epithelium. Cell number must be strictly controlled during the healing process of bronchial epithelia to eliminate certain cells and to retain others that will form the remodeled epithelium. Disruption of these controls may lead to sustained hyperplasia (e.g., increased numbers of mucous cells in asthmatics) or may allow hyperplastic cells to continue proliferating, and further selection could favor changes that lead to autonomous growth and neoplasia.

Programmed cell death, also called apoptosis, is a primary mechanism in the precise regulation of cell numbers in tissue homeostasis (4, 5). Apoptosis is a genetically controlled active cellular process. In recent years, there has been an avalanche of information on the genes involved in the regulation of apoptosis (6). The importance of apoptosis in neoplastic transformation was noted after the observation that the most common change in human B-cell follicular lymphoma is a translocation that juxtaposed the Bcl-2 gene with the immunoglobulin heavy-chain gene (7). Subsequently, it has been shown that the Bcl-2 protein does not stimulate proliferation, but rather confers survival upon cells (8) by blocking apoptotic cell death.

The Bcl-2 protein is found within various cell lineages and tissues that are all characterized by apoptotic cell turnover. Hockenbery and colleagues (9) have reported that Bcl-2 is expressed in lymphoid cells and several epithelial tissues, as well. Bcl-2 is localized in glandular epithelia that are regulated by hyperplasia or involution, usually in response to hormonal stimuli, e.g., the breast duct epithelium, the thyroid epithelium, and the basal cuboidal epithelium of the prostate gland. In complex, organized epithelia, Bcl-2 expression is restricted to the basal layers, such as in the basal layers of epidermis and the lower crypt of small and large intestines (9, 10). It has been suggested that this pattern of expression assists the survival of stem cells while preventing the overaccumulation of differentiated cells (10). Based on these reports, we investigated whether Bcl-2 and related proteins are involved in the repair process of airway epithelia following toxicant-induced hyperplasia/metaplasia.

Exposure of the respiratory tract to ambient levels of ozone or endotoxin induces mucous cell metaplasia (MCM) in the nasal or pulmonary airway epithelia of F344/N rats (11, 12). After exposure of F344/N rats to ozone, the number of epithelial cells/mm basal lamina of the nasal transitional epithelium increases briefly, then decreases to initial levels (13). Disruption of mechanisms that control resolution of MCM may be responsible for the sustained presence of mucous cells and excessive mucus secretion in patients with chronic airway diseases (e.g., asthma, chronic bronchitis, and cystic fibrosis). Therefore, the purpose of the present investigation was to determine whether the regulator of apoptosis, Bcl-2, plays a role in the development and resolution of transient MCM. The present study shows that Bcl-2 is expressed in metaplastic mucous cells but not in endogenous mucous cells lining respiratory epithelia, and that high levels of Bcl-2 can be detected in metaplastic mucous cells that were induced by ozone or endotoxin in surface epithelia lining the nasal or bronchial airways, respectively.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Exposure to Ozone and Histopathology

At least three male F344/N rats (Harlan Sprague-Dawley Inc., Indianapolis, IN) for each group were exposed to filtered air or to 0.5 ppm of ozone 8 h/d between 4:30 P.M. and 12:30 A.M. for 1, 3, and 6 mo. Another group of rats was kept for 13 wk in filtered air after exposure to ozone for 3 mo. The exposure method has been described (14). Briefly, ozone was generated by an OREC Model O3VI-O ozonizer (Ozone Research & Equipment Corp. Phoenix, AZ). Dilution air was mixed with ozone to bring the total airflow through the exposure chamber to approximately 15 air exchanges/h. The ozone concentration within the chamber was held at target levels by adjusting the amount of ozone delivered to the chamber by computer-controlled mass flow valves. Chamber temperature and relative humidity were maintained between 21-25°C and 40-70%, respectively.

Rats were exsanguinated via the left ventricle after deep anesthesia with 5% halothane in oxygen. After death, the head of each rat was removed, and the nasal cavity was flushed with 10 ml zinc formalin through the nasopharyngeal orifice. The lower jaw, skin, and musculature were removed, and the head was immersed in a large volume of zinc formalin for at least 24 h. Following fixation, the heads were decalcified in 13% formic acid for 4 d and rinsed in tap water for at least 4 h. A tissue block was removed from the anterior nasal cavity by making two transverse cuts perpendicular to the hard palate immediately posterior to the upper incisor teeth. The tissue block was imbedded in paraffin, and 5-µm-thick sections were cut from the anterior surface.

Immunohistochemistry

Endogenous peroxidase activity was blocked by incubating sections in 0.3% hydrogen peroxide/methanol for 30 min. Slides were washed in deionized water; all subsequent washes were in 0.05% Brij/Dulbecco's phosphate-buffered saline (pH 7.4). Bcl-2 was unmasked by microwaving slides while immersed in antigen retrieval solution (BioGenex, San Ramon, CA) for 15 min (700 W). Slides were first incubated in 1% normal goat serum in 2% bovine serum albumin/0.1% Triton X-100, the primary antibody (polyclonal rabbit antimouse Bcl-2 from PharMingen, San Diego, CA; and Santa Cruz Biotechnology, Santa Cruz, CA), or normal rabbit serum at 1:1,000 dilution was applied. After an overnight incubation at room temperature, the immunoreaction was visualized using the peroxidase substrate diaminobenzidine. Epithelial cells with mucosubstances were detected by staining tissue sections with alcian blue (15).

Quantitation of MCM and Bcl-2-positive Cells

The total number of mucous cells lining the midseptum, lateral wall, and nasoturbinates and maxilloturbinates were counted. The surface epithelium lining these regions was designated as shown in Figure 1. Mucous cells were identified on tissue sections from control and ozone-exposed rats by alcian blue staining and the percentage of Bcl-2-positive mucous cells determined.

View larger version (53K): [in this window] [in a new window]   Figure 1.   Low-power light micrograph of the nasal septum of a rat. The three regions of transitional epithelia lining the lateral wall (LW) and the nasoturbinates (NT) and maxilloturbinates (MT) and the respiratory epithelium of the midseptum (MS) are shown, where the number of mucous cells was determined after alcian blue staining of tissue sections.

Western Blot Analysis

Protein was extracted from nasal epithelia of rats exposed to ozone or air for 1 mo. Nasal epithelia were removed from the midseptum and nasal turbinates and homogenized in RIPA buffer (10 mM Tris, pH 7.4), 150 mM NaCl, 1% Triton X-100, 1% deoxycholate, 0.1% sodium docecyl sulfate, 5 mM ethylenediamenetetraacetic acid supplemented with the protease inhibitors, phenylmethylsulfonyl fluoride (1 mM), pepstatin (10 µg/ml), aprotinin (2 µg/ml), and benzamidine (2 µg/ml). The protein concentration was determined using the BCA assay kit (Pierce, Rockford, IL); 120 µg protein from control and ozone-exposed rats was loaded on each lane. Western blotting was carried out as described earlier (16), and filters were stained with Ponceau S to confirm that equivalent amounts of protein had been loaded on each lane. The Bcl-2 antibody was used at 1:1,000 dilution.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Immunohistochemistry

Harkema and Hotchkiss (13) found that after exposure to ozone, the respiratory epithelium of the midseptum did not show a detectable difference from the epithelium of unexposed animals. However, transitional epithelia of the nasal and maxillary turbinates change substantially (13). In the present study, transitional epithelia lining the nasoturbinates and maxilloturbinates of air-exposed control rats were negative for the Bcl-2 protein (Figure 2A). Numerous metaplastic mucous cells were formed in these epithelia due to ozone exposure, and many of these mucous cells showed intensive immunostaining for Bcl-2 (Figure 2B). Mucous cells of the respiratory epithelium of the midseptum showed only a few Bcl-2-postive cells after exposure to ozone. These results were reproducible when using polyclonal antibodies that were raised to different Bcl-2-specific peptides.

View larger version (140K): [in this window] [in a new window]   Figure 2.   Induction of MCM and Bcl-2 expression in transitional epithelia lining the nasoturbinates and maxilloturbinates of F344/ N rats. Immunoreaction of the Bcl-2 antibody was detected using diaminobenzidine and shows brown. Tissue sections were also stained with alcian blue to identify metaplastic mucous cells. While no expression can be detected in the transitional epithelia in rats exposed to air only (A), many ozone-induced metaplastic mucous cells show intensive immunostaining for Bcl-2 (B).

Western Blot Analysis

The Bcl-2 protein (28 kD) was detected in protein extracts from nasal epithelia of rats exposed to ozone for 1 mo, but not in control rats (Figure 3). This analysis shows that the immunoreaction of the antibody is specific to the 28-kD Bcl-2 protein, and confirms results obtained by immunohistochemistry that ozone induces Bcl-2 expression in rat nasal epithelia. The protein extract was prepared from the nasal epithelia that included the basal cells. A 41-kD protein that is present in control and ozone-exposed rats cross-reacted with the Bcl-2 antibody. This protein may be contained in the basal cells and may be at least partly responsible for the immunostaining in the transitional epithelium (Figures 2A and 2B).

View larger version (36K): [in this window] [in a new window]   Figure 3.   Detection of Bcl-2 by Western blot analysis in nasal epithelia of rats exposed to 0.5 ppm ozone for 1 mo. Protein was extracted from the nasal epithelia removed from the septum and turbinates of rats exposed to ozone for 1 mo, and 120 µg was analyzed by Western blotting using Bcl-2 antibodies at 1:1,000 dilution. The Bcl-2 protein was detected in extracts from ozone- exposed rats (lanes 1-4) but not from control rats (lane C).

Quantification of Bcl-2-positive Cells

The total number of metaplastic mucous cells increased from 0 to 102, 78, and 172 after 1 mo of ozone exposure in the transitional epithelia lining the lateral walls and the nasoturbinates and maxilloturbinates, respectively (Figure 4). After 3 and 6 mo the number of mucous cells lining the lateral walls and the nasal turbinates increased further to about 200 and 150, respectively. No significant change was observed in the epithelia lining the maxillary turbinates (Figure 4). After a recovery period of 13 wk, the number of metaplastic mucous cells in these epithelia decreased to 0-10 in rats exposed for 3 mo. In contrast to these dramatic changes in these epithelia, the number of mucous cells lining the midseptum did not change significantly during exposures to ozone for 1, 3, and 6 mo and after the recovery period (Figure 4).

View larger version (38K): [in this window] [in a new window]   Figure 4.   Changes in the total number of mucous cells in epithelia lining the midseptum, lateral wall, and nasoturbinates and maxilloturbinates. Alcian blue-positive cells in areas lining the nasal epithelia designated in MATERIALS AND METHODS were counted. Tissue sections were used from at least three rats each exposed to air as control or to ozone for 1, 3, and 6 mo and from 3-mo exposed rats maintained in filtered air for a recovery period of 13 wk. Error bars denote the standard deviation from the mean.

The percentage of Bcl-2-positive mucous cells lining the midseptum, lateral wall, and nasoturbinates and maxilloturbinates in tissue sections from three rats exposed to ozone was determined. Only 3 to 7% of mucous cells lining the midseptum were Bcl-2 positive 1 mo after ozone exposure, and 14% of these cells were Bcl-2 positive after a 6-mo exposure to ozone (Figure 5). In the transitional epithelia of the lateral wall and the nasoturbinates and maxilloturbinates, 35 to 55% of the alcian blue-positive cells immunoreacted to Bcl-2 antibodies in rats exposed to ozone for 1 mo. After a 3- and a 6-mo exposure, only about 10 to 18% of these cells were Bcl-2 positive (Figure 5). Bcl-2 reactivity of metaplastic mucous cells decreased to 0 to 2% in the midseptum, the lateral wall, and nasal turbinates after a recovery period of 13 wk in rats exposed to ozone for 3 mo. In these rats, about 8% of the remaining mucous cells lining the maxillary turbinates showed Bcl-2 reactivity (Figure 5). Cells with a morphology that is characteristic of apoptosis were not detected in epithelia from either normal or ozone-exposed rats by light microscopy.

View larger version (30K): [in this window] [in a new window]   Figure 5.   The percentage of Bcl-2-positive mucous cells in nasal epithelia of rats exposed to air or to ozone for 1, 3, and 6 mo and in rats maintained in filtered air for a recovery period of 13 wk after a 3-mo ozone exposure. Endogenous mucous cells lining the midseptum show no increase in Bcl-2 staining after 1 and 3 mo ozone exposure; a significant increase in Bcl-2 immunostaining is seen after 6 mo exposure. Error bars denote the standard deviation from the mean. After 1 mo exposure, about 35 to 55% of metaplastic mucous cells of the transitional epithelia lining the lateral wall and the nasoturbinates and maxilloturbinates were Bcl-2 positive. After 3 and 6 mo, only 12-18% were Bcl-2 positive. After the recovery from ozone exposure, no Bcl-2 reactivity was observed in mucous cells except in the mucous cells lining the maxillary turbinates.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present study provides evidence that Bcl-2, a regulator of apoptosis, may be involved in the healing process of airway epithelia after irritant-induced airway cell metaplasias. Bcl-2 was expressed in ozone-induced metaplastic mucous cells lining the epithelia of the nasal airways. After remodeling, the number of metaplastic mucous cells and the percentage of Bcl-2-expressing cells decreased. These observations suggest that Bcl-2 may be part of a biochemical mechanism involved in the reorganization process in the healing epithelium.

The Bcl-2 protein was found in the cytosolic areas of the mucous cells. It has been shown that large amounts of agranular endoplasmic reticulum and numerous mitochondria are present in the apical portion of mucous cells of the rat (17). Therefore, our observations agree with previous reports that have localized Bcl-2 as an integral mitochondrial membrane protein or to other subcellular compartments, such as endoplasmic reticulum and nuclear membrane (18).

The bronchial epithelia of control rats showed no immunostaining for Bcl-2 in the present study. Earlier reports found that in humans, basal cells of normal bronchial epithelial mucosa are positive for Bcl-2, with the more differentiated cells being negative (19, 20). However, in a recent study, Vignola and associates (21) analyzed bronchial biopsies obtained from asthmatics, chronic bronchitics, and control subjects, and detected Bcl-2 expression in basal cells from asthmatics and chronic bronchitics but not from control subjects. The difference in results may stem from the fact that the first reports did not differentiate between normal subjects and those with bronchial disorders. The observations by Vignola and coworkers correspond with our studies showing no Bcl-2 expression in basal cells of rat bronchial epithelia.

It is interesting that only a few Bcl-2-expressing mucous cells were detected in the respiratory epithelium of the midseptum, even after a 3-mo exposure to ozone. Both the metaplastic mucous cells lining the nasoturbinates and maxilloturbinates and the endogenous mucous cells lining the respiratory epithelium synthesize the same type of apomucin, MUC 5 (personal observation), but the pattern of Bcl-2 expression indicates differences in the two types of mucous cells. In the present study, the number of Bcl-2-positive mucous cells of the midseptum increased after a 6-mo ozone exposure, indicating that after long exposures these mucous cells are altered. Walker and colleagues (22) observed in some cases that occasional well-differentiated columnar cells in histologically normal human tracheal epithelium express Bcl-2. Walker and associates mentioned that these cells are unlikely to reflect an early event in the transformation to malignancy; however, no speculation is given as to why these cells express Bcl-2. Our studies suggest that exposure of respiratory epithelia to toxins, such as ozone, for long periods of time may give rise to mucous cells that resemble metaplastic cells.

Metaplastic mucous cells are heterogeneous in their Bcl-2 expression. Some cells contain high levels, whereas others express low levels or no Bcl-2. Previous reports (19) suggest that Bcl-2 expression is associated mainly with activated and undifferentiated cells that are undergoing terminal differentiation and need protection from apoptosis. In complex, organized epithelia, such as the small and large intestines and the skin, loss of Bcl-2 expression correlates with differentiation, loss of proliferative capacity, and reduction of remaining life span (9, 19). Certain cells must be discarded during the resolution of the airway epithelia, so it is likely that Bcl-2-overexpressing cells may be retained to form remodeled epithelium. Antibodies to Bcl-x showed a similar expression pattern in these epithelia (personal observation), supporting the hypothesis that this family of apoptotic regulators is involved in the remodeling process.

Heterogeneity in Bcl-2 expression indicates that selected cells may be designated to remain and constitute the healed epithelium by overexpressing this protein. It is unclear what factors could be involved in inducing Bcl-2 expression in certain cells. Perhaps basal cells within the airway epithelium signal the "abnormal" condition to the metaplastic cells and initiate the repair process as soon as the insult to the epithelium is removed. During the process of differentiation, these epithelial cells may be intrinsically programmed to live longer than neighboring cells or may express certain receptor proteins that signal induction of Bcl-2 in response to inflammatory cytokines. The presence of Bcl-2 in only a portion of metaplastic mucous cells clearly argues for the importance of regulatory mechanisms that induce or suppress the Bcl-2 gene expression. The percentage of Bcl-2-positive metaplastic mucous cells is higher in rats exposed to ozone for 1 mo than for 3 or 6 mo. The reduction in Bcl-2 expression may indicate that after longer exposures to the insult, basal cells accommodate the new, "long-term" condition. Possibly, the cytokine concentrations in the epithelium may be decreased after longer exposures and result in reduction of Bcl-2 expression.

Several studies indicate that the morphologic response to ozone involves epithelial cell injury along the entire respiratory tract, resulting in cell loss and replacement (23). Airway epithelial cell loss after ozone exposure occurs within 2-4 h (24). In the present study, time points during the period of metaplastic tissue development and resolution were examined for apoptosis; however, no apoptotic cells were found by light microscopy. It is possible that these cells are eliminated very rapidly or may not involve the known morphologic changes characterized by membrane blebbing, cytoplasmic and nuclear condensation, and the formation of apoptotic bodies (24). In a study that compared bronchial epithelia of asthmatic and normal subjects, Vignola and coworkers (21) found that the number of apoptotic cells was lower in asthmatics than in control subjects and correlated with the number of Bcl-2-positive cells. They concluded from their studies that disregulation of apoptosis can play a key role in the pathogenesis of asthma. These studies support the hypothesis that the Bcl-2-positive cells in rat bronchial epithelia may be differentially refractory to the induction of apoptosis.

Identical patterns of Bcl-2 expression in metaplastic mucous cells have been observed in the bronchial epithelia of rats exposed to cigarette smoke or intratracheally instilled with endotoxin (personal observation). The fact that Bcl-2 is expressed in metaplastic mucous cells regardless of the inducing agent and location of airway epithelia suggests that Bcl-2 plays a ubiquitous role in the development or sustenance of such metaplasias.

In stable bronchial asthma, the mucosa in the airways is characterized by a decrease in the number of ciliated cells, and an increase in the number of mucous cells along with MCM (25). Excess mucous secretion in the lumen (26), combined with impaired mucociliary transport and reduced clearance, lead to the accumulation of mucus, to the development of mucous plugs, and to airflow obstruction (25). Because the disregulation of Bcl-2 expression could be at least partially responsible for the sustained MCM, understanding the role of Bcl-2 in the development and maintenance of hyperplastic, mucus-secreting cells could lead to prevention or new therapeutics for diseases associated with mucus overproduction and secretion.

Furthermore, Bcl-2 is absent in normal, well-differentiated epithelial cells of the human lung, but is present in 28% of non-small-cell lung cancers (20). Furthermore, in dysplastic bronchial epithelium, 66% of the epithelial cells throughout the full epithelial thickness are Bcl-2 positive (22). This aberrant pattern of Bcl-2 expression correlates with an increasing grade of dysplasia. Future studies will determine whether persistent Bcl-2 expression would allow metaplastic mucous cells to accumulate abnormally and possibly render them more susceptible to transformation.