Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Ozone (O₃), a major component of photochemical smog, is a strong oxidant. Both in vivo and in vitro studies have documented dose-dependent effects of O₃ on the integrity of lung epithelium (1, 2). Cytotoxic effects of O₃ include epithelial necrosis, sloughing, and increased permeability of the epithelial layer. However, the possible effects of O₃ on epithelial integrins are not known despite the importance of these molecules in maintaining lung homeostasis (3).

₁ integrins play an important role in the attachment of cells to the extracellular matrix and maintenance of cell- cell interactions (4, 5). Also, ₁ integrins are regulators of cell survival, proliferation, differentiation, migration, and morphology (6). Integrins with the ₁ subunit are expressed on epithelial cells in the lung (12). Certain integrins, such as ₂₁, ₃₁, and ₉₁, are constitutively expressed in the normal lung (3). The ₂₁ and ₃₁ integrins are diffusely expressed in bronchial epithelial cells with preferential localization to basolateral cell surfaces. The integrin ₉₁ occurs predominantly in the undifferentiated basal cells of bronchial tissue. On the other hand, a subset of inducible integrins including ₅₁ are expressed only during inflammation and injury in the lung and their expression is augmented by various inflammatory mediators such as transforming growth factor ₁ (TGF-₁) (15).

The underlying factors involved in regulation of epithelial injury and repair are still incompletely defined. But it is becoming evident that integrins play a key role in these processes and that different integrins are involved in epithelial injury and repair. Treatment with hydrogen peroxide causes redistribution of the ₃₁ integrin from basolateral to apical cell surfaces and diminished binding of kidney epithelial cells to their matrix (16). The healing of injured airway epithelial cultures is prevented by using antibodies to the integrin subunits ₅ and ₁ or to fibronectin, which inhibit the binding of ₅₁ to fibronectin (9). Furthermore, the levels of integrins in epithelial cells are modulated during the early phase of airway repair. Pilewski and colleagues (17) found increased immunostaining of the _v6 integrin but no changes in the levels of ₂₁ and ₃₁ integrins during airway repair. In models of skin wound healing, migrating keratinocytes display increased levels and altered localization of ₁ integrin and several of its associated  subunits during re-epithelization of denuded areas (18).

Whereas a number of studies have shown alterations in integrins in response to oxidative injury, no studies have addressed whether environmental pollutants such as O₃ alter ₁ integrins in airway epithelial cells. The present study demonstrates that following cell loss, O₃ causes marked redistribution of the integrin ₁ subunit in those cells that resist O₃-induced detachment. Furthermore, the changes in integrin distribution caused by O₃ are not due to increased ₁ messenger RNA (mRNA) levels and ₁ protein synthesis. However, on a per-cell basis, detergent lysates from O₃-exposed cells contained more ₁ integrin than did air controls. It is possible that O₃ alters ₁ interactions with cytoskeletal components, which in turn causes increased detergent solubility of ₁ integrins. Our results show specific effects of O₃ on epithelial adhesion molecules that are involved in cell-cell and cell-matrix interactions.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents

Reagents were purchased from the following sources: Dulbecco's modified Eagle's medium:nutrient mixture F-12 (DMEM/F-12), methionine- and cysteine-free RPMI-1640 cell culture media, newborn calf serum, phosphate-buffered saline (PBS), glutamine, N-2-hydroxyethylpiperazine- N'-ethane sulfonic acid (Hepes), penicillin, streptomycin sulfate, versene, and trypsin-versene from BioWhittaker Bioproducts (Walkerville, MD); Nu-serum from Collaborative Research (Lexington, MA); fungizone and TRIzol from GIBCO BRL (Gaithersburg, MD); dispase and N-[N- (L-3-trans-carboxyoxirane-2-carbonyl)-L-leucyl]-agmatine (E-64) from Boehringer Mannheim (Indianapolis, IN); aprotonin, formaldehyde, leupeptin, phenylene diamine, and phenylmethylsulfonyl fluoride (PMSF) from Sigma (St. Louis, MO); collagen-coated Transwell culture inserts from Costar (Van Nuys, CA); Vitrogen 100 (95-98% collagen I and 2-5% collagen III) from Collagen Corp. (Fremont, CA); monoclonal antibody against the human ₁ subunit (Clone P4C10; Cat. No. 12086-013) and polyclonal antibody against the human fibronectin receptor ₅₁ (Cat. No. 12118-014) from GIBCO BRL; goat antimouse IgG (H&L) conjugated to Cy3^TM from Biological Detection Systems (Pittsburgh, PA); protein A Sepharose CL4B from Pharmacia Biotech (Piscataway, NJ); [³⁵S]protein- labeling mix containing approximately 73% L-methionine and 22% L-cysteine from DuPont (Boston, MA); and diamidino-2-phenylindole (DAPI) from Molecular Probes (Eugene, OR).

Cell Culture

Bronchial epithelial cells were obtained from healthy Macaca nemestrina killed at the Regional Primate Research Center at the University of Washington (Seattle, WA). The tissue specimens were immediately placed in DMEM/F-12 supplemented with 10% Nu-serum, 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin sulfate, 2.5 µg/ml fungizone, and 10 mM Hepes buffer and transported immediately to the laboratory. The tissue specimens were then rinsed three times in PBS containing 200 U/ ml penicillin, 200 µg/ml streptomycin sulfate, and 5 µg/ml fungizone. The upper airways were dissociated from the lung parenchyma, cut into approximately 4 × 4-mm pieces, placed in 0.4% dispase in PBS, and incubated at 37°C for 30 min or overnight at 4°C. The epithelium was separated from the underlying connective tissue and placed in trypsin/ethylenediamenetetraacetic acid (EDTA) for 10 min at 37°C prior to adding fetal calf serum (FCS) to inactivate the trypsin. The cells were pelleted at 300 × g for 5 min and resuspended in fresh media by vigorous pipetting. Cells were cultured at 0.2 × 10⁶ cells/cm² at an air interface on porous membranes precoated with 100 µg/cm² collagen in Transwell inserts. The cells were incubated in humidified air with 5% carbon dioxide at 37°C and allowed to adhere to the inserts overnight under medium, then washed with PBS to remove nonadherent cells. The cultures were then maintained at an air-cell interface by adding medium twice weekly to the cluster plate beneath the microporous membrane. This system of cell culture permitted gaseous exposures without overlying fluid (21). Previous studies in our laboratory have shown the cells to be epithelial in nature (22). Differentiated cells, including ciliated cells, in these cultures are shown with scanning electron microscopy (Figure 1).

View larger version (226K): [in this window] [in a new window]   View larger version (228K): [in this window] [in a new window]   Figure 1.   Scanning electron micrographs of bronchial epithelial cultures exposed to air (a) or O3 at 0.5 ppm (b) for 6 h and allowed to recover for 6 h prior to sample processing. Ozone exposure caused cilia damagenote the shorter and less dense cilia compared with air-exposed culturesand cell detachment in the upper layer of cells that is apparent as cell sloughing (arrows). Bar = 10 µm.

Ozone Exposure

The environmental exposure system allows parallel and simultaneous exposure of cells to either air or O₃. The present exposure level was chosen because it has been shown previously in our laboratory to alter the expression of intercellular adhesion molecule-1 (ICAM-1) (21). Cultured primate bronchial cells were exposed to either air or O₃ on platforms in identical 3-cubic-foot (ft³) exposure chambers. The exposure gas flowed into the chambers at 300 ft³/h and out via negative pressure exhaust ports adjusted to maintain pressure at 1 atm. Temperature and humidity inside the exposure chambers were continuously monitored and maintained at 37°C and 95%, respectively. Ozone was generated by passing 100% oxygen through a commercial ozonator (Model T-408; Welsbach Corp., Philadelphia, PA). The flow of O₃ was adjusted and the concentration was continuously monitored by an ultraviolet O₃ analyzer (Model 1003AH; Dasibi Corp., Glendale, CA). Both the air and O₃ exposure chambers had carbon dioxide added at 5% and were continuously monitored with a medical gas analyzer (Model LB-2; Beckman, Irvine, CA). Following exposure, cells were rinsed in PBS and then allowed to recover for 0, 6, or 24 h prior to analysis.

Cell Counts

To determine the cytotoxic effects of O₃ on epithelial cultures, cell counts were done at 24 h after exposure. Cells remaining in the epithelial layer (attached cells) were obtained from a total of six culture inserts (4.7 cm² each) by 20-min treatment with trypsin-versene. FCS was then added to inactivate the trypsin. The cells were pelleted at 300 × g for 5 min, washed once with PBS, and resuspended in 0.5 ml of PBS. A 10-µl aliquot was counted with a hematocytometer to determine cell concentrations and extrapolated to total cell number per six inserts.

Scanning Electron Microscopy

Cell cultures were rinsed with 0.16 M cacodylate buffer and fixed by immersion of the inserts in 2% glutaraldehyde in 0.1 M cacodylate buffer. After storage overnight at 4°C, the cultures were rinsed in 0.16 M cacodylate buffer, postfixed for 1 h with 1% osmium tetraoxide in 0.1 M cacodylate buffer, dehydrated in ethanol, substituted with carbon dioxide, and dried in a critical-point dryer (Autosamdri-810; Tousimis, Rockville, MD). Inserts were pressed onto 13-mm diameter circular pins or stubs (Cambridge-type specimen mounts that were covered with double-sticky tape) and the membranes were cut out of the inserts. The stubs were coated with gold palladium in a sputter coater (Desk-1; Denton, Cherry Hill, NJ) and viewed in a scanning electron microscope (JSM-35U; JOEL, Tokyo, Japan).

Immunocytochemistry

All incubations were done at room temperature and antibodies were diluted in PBS containing 1% bovine serum albumin. Cell cultures were fixed with 2% formaldehyde in 0.1 M cacodylate buffer, pH 7.2, containing 0.1 M sucrose for 20 min. After fixation, cells were rinsed twice in PBS containing Ca⁺⁺ and Mg⁺⁺ and permeabilized with 0.5% Triton X-100 for 5 min. Nonspecific binding was blocked with 10% goat serum added for 30 min prior to a 1-h incubation with human ₁ subunit antibody at a 1:1,000 dilution. Cells were washed with PBS and incubated for 1 h with goat antimouse IgG conjugated to Cy3 at a 1:2,000 dilution. The cultures were then washed once with PBS for 5 min, incubated with DAPI at 1 µg/ml in PBS for 15 min, and washed in PBS for 5 min. The membranes were cut from the inserts and mounted on glass slides in 90% glycerol containing 50 mM phenylene diamine.

Laser Cytometer Fluorescence Analysis

Immunostaining of total cellular ₁ as well as ₁ levels at the apical and basal surfaces of cells were evaluated by confocal microscopy. Each filter was scanned using an Attached Cell Analysis and Sorting (ACAS) interactive laser cytometer (ACAS 570; Meridian Instruments, Okemos, MI). The argon laser was tuned to 514 nm. Scanning was done in the confocal mode with a pinhole of 225 µm and a step size of 0.2 µm to generate multiple scans in a designated Z plane, producing a digitized confocal image of the detected fluorescence. Laser power was set at 750 mW, and the acoustoptical modulator was set at 10%.

Scanning was done in both ultraviolet (UV) and visible light modes. A 485-nm long-pass dichroic filter was placed in line with the dual detectors and no emission filters were used. Photomultiplier tube amplifier settings were chosen so that detected fluorescence was within range for individual experiments. For each condition, three to four inserts were examined and at least five different scans were obtained randomly for each insert, which allowed for the evaluation of 250 to 300 cells per condition. Air and O₃ samples were examined at 0, 6, and 24 h after exposure because previous studies of integrin redistribution reported such effects to occur at either 0 or 24 h (16, 23).

Using ACAS image analysis software, distribution of ₁ subunit was determined by quantifying the fluorescence at the apical and basal sides of cultured cells. This was done by drawing polygons on both the apical and basal sides of the cell layer to obtain average integrated fluorescence (total number of pixels per area [µm²]). Background noise fluorescence was eliminated by adjusting the image fluorescence intensity to a fixed threshold level.

Northern Blot Analysis

Total cellular RNA was obtained from cultured epithelial cells solubilized at 18 h after exposure with TRIzol reagent containing phenol and guanidine isothiocyanate (24). A time point of 18 h was selected because of previous reports on the slow response of the ₁ gene in which maximal induction occurred at 12-48 h (23, 25). To isolate RNA, lysates were extracted with phenol and subsequently precipitated with isopropanol. The RNA pellet was washed twice with 75% ethanol and once with 100% ethanol, and redissolved with an appropriate volume of H₂O containing diethylpyrocarbonate at 0.1% (vol/vol).

A total of 15 µg of RNA was fractionated on 0.8% formaldehyde-agarose gels and transferred to nylon blotting membranes (Zeta Probe membranes; Bio-Rad Laboratories, Hercules, CA). Transferred RNA was cross-linked by UV irradiation and blots were stored dry at 4°C and hybridized with human complementary DNA (cDNA) probes for the fibronectin receptor ₁ subunit or elongation factor-1 (ELF). The 2.5-kb ₁ cDNA, selected from a human umbilical vein endothelial cell cDNA library in the vector lambda gt11 based on a previously characterized oligonucleotide probe (28), was kindly provided by Laurence F. Fitzgerald at the Gladstone Foundation Laboratories (San Francisco, CA). ELF cDNA, a 0.8-kb probe encoding the carboxy terminus and the 3'-untranslated region (922 through 1705), was used as a reference gene (29). Using random priming, 100 ng of ₁ or 50 ng of ELF cDNAs were prelabeled by a 10-min incubation with exonuclease free klenow enzyme (2 U) and 50 µCi of [-³²P]dCTP according to the manufacturer's protocol (Ambion, Austin, TX). The prehybridizing and hybridizing solutions contained 50% formamide, 6× standard sodium phosphate and EDTA buffer (SSPE), 5× Denhardt's solution, 0.5% sodium dodecyl sulfate (SDS), 5% dextran sulfate, and 100 µg/ml salmon testes DNA. Prelabeled ₁ and ELF probes were added to the hybridization solution at a final concentration of 1 × 10⁶ and 4 × 10⁵ cpm, respectively. Blots were hybridized for 16 h and then washed twice for 5 min each and once for 30 min in 2× SSPE and 0.1% SDS at 42°C and finally washed twice for 30 min each with 0.3× SSPE at 65°C. Autoradiographic exposures were done at 70°C with Kodak XAR-2 film (Kodak, Rochester, NY) and mRNA levels were quantified by densitometric scanning. The results are normalized to the amount of ELF mRNA.

Immunoprecipitation and SDS-Polyacrylamide Gel Electrophoresis (PAGE)

Following exposure, cells were rinsed in PBS and allowed to recover overnight in fresh DMEM/F-12 medium supplemented with 10% Nu-serum. The cultures were then incubated with methionine- and cysteine-free RPMI-1640 medium for 30 min prior to radiolabeling with 10 µCi/ml of [³⁵S]protein-labeling mix in the presence of 2% Nu- serum for 6 h at 37°C. The cells were detached from the membranes by incubation with trypsin/EDTA for 20 min at 37°C, counted, and extracted at 1 × 10⁷ per ml with 1% Triton X-100 in PBS, pH 7.4, for 30 min on ice in the presence of the protease inhibitors PMSF (1 mM), leupeptin (1 µg/ml), aprotonin (1 µg/ml), and E-64 (10 µg/ml). Samples were precleared overnight at 4°C with 40 µg/ml of protein A Sepharose CL4B prior to a 2-h incubation at 4°C with a 1:200 dilution of a monoclonal antibody against ₁ integrin subunit or a 1:100 dilution of polyclonal antiserum against ₅₁ integrin. Integrin immune complexes were isolated by the addition of 40 µg/ml of protein A Sepharose for 30 min at 4°C. Proteins were then solubilized in sample buffer (30) and separated by SDS-PAGE electrophoresis on a 4-20% Tris-glycine gradient gel (Bio-Rad) under reducing conditions. Gels were incubated with Amplify (Amersham Life Sciences, Inc., Arlington Heights, IL) for 30 min at room temperature and visualized by exposure to a BIOMAX MR imaging film (Kodak) for 2 d at 70°C.

Statistics

Data analysis was performed with the statistical package Statview (BrainPower Inc., Calabasas, CA). The unpaired Student's t test was used to analyze the data and P values less than 0.05 were considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Scanning Electron Microscopy

Epithelial cultures were grown at air-liquid interface to obtain differentiated cultures as indicated by the presence of ciliated cells (Figure 1). Cultured primate bronchial cells formed a pseudostratified phenotype of two and occasionally three cell layers. Ozone exposure caused focal cell detachment and significant damage to ciliary structures (Figure 1). The detachment process involved the upper layer of cells, resulting in a flatter monolayer of epithelial cells. Cell counts showed a 60 ± 3% cell loss at 24 h following O₃ exposure (Figure 2).

View larger version (44K): [in this window] [in a new window]   Figure 2.   Effects of O3 on cell number in bronchial epithelial cultures. Cultures were exposed to air or O3 at 0.5 ppm for 6 h and allowed to recover for 24 h. For each exposure condition, cells were obtained from a total of six culture inserts (4.7 cm2 each) and counted with a hematocytometer (n = 4 experiments); *P < 0.05, O3 versus air. Cell loss was observed in O3-exposed cultures.

Immunofluorescence Microscopy

The effect of O₃ on ₁ integrin subunit in the remaining attached cells was examined to define changes of ₁ integrins in epithelial cultures subsequent to cell injury and loss. Figure 3 shows computer-generated, two-dimensional images of cell cultures derived from laser cytometer fluorescence analysis. The color scale shows different levels of ₁ integrin subunit immunostaining; with low staining intensity indicated by violet and higher levels indicated by red and white. In the air-exposed cultures, diffuse staining of ₁ subunit appeared in the cytoplasm, at cell-cell contacts, and was greater at the apical side of the culture (Figure 3). In contrast, O₃-exposed cells showed higher levels of cellular ₁ immunoreactivity, as indicated by the presence of red and white colors. The increased ₁ immunostaining was localized at the apical surfaces and, to a lesser extent, at cell-cell contacts of exposed cells. The O₃-induced increase in ₁ immunostaining was also time-dependent with 23 ± 5% and 66 ± 6% increases at 6 and 24 h, respectively (Figure 4). Following exposure to O₃, there were no observable changes in ₁ immunostaining at the basal surface of cultured cells at any time point. Furthermore, apical and basal levels of ₁ immunoreactivity did not change after air exposure at any time (Figure 4).

View larger version (44K): [in this window] [in a new window]   Figure 3.   ACAS images of cultured bronchial epithelial cells exposed to either air or O3 at 0.5 ppm for 6 h and allowed to recover for 24 h prior to immunocytochemical analysis of the 1 integrin subunit. The primary antibody is omitted in the negative control sample. The color scale represents different immunostaining levels of the 1 subunit. Low staining intensity is indicated by violet while greater levels are indicated by red and white. The terms "apical" and "basal" describe the sides of cultured cells relative to the insert membrane. Air-exposed cultures showed diffuse cytoplasmic staining of the 1 subunit. Ozone-treated cells displayed distinct increases of 1 immunoreactivity, particularly at cell contacts and at the apical surface. Nonspecific autofluorescence of the insert membrane is shown in the negative control sample. Bar = 10 µm.

View larger version (13K): [in this window] [in a new window]   Figure 4.   Effects of O3 on localization of 1 subunit in bronchial epithelial cells as determined by ACAS analysis. The levels of 1 subunit were quantified at the basal and apical portions of cultured cells at 0, 6, and 24 h following exposure to either air or O3 at 0.5 ppm for 6 h. An increase in the apical localization of 1 integrins was seen at 6 and 24 h following O3 exposure (n = 3 experiments); *P < 0.05, O3 versus air.

Northern Analysis

Since the observed increases in apical immunostaining of ₁ integrin subunit occurred between 6 and 24 h (Figure 4), ₁ mRNA levels were analyzed at 18 h after exposure to determine whether O₃ modulates integrins at the transcriptional level (Figures 5a and 5b). A representative blot hybridized with ³²P-labeled cDNA probes for the ₁ subunit of the fibronectin receptor ₅₁ and ELF is shown in Figure 5a. ₁ cDNA recognized a single band of 4.2 kb in size (28). Ethidium staining of 28S and 18S ribosomal RNA and measurements of ELF mRNA were done to confirm equal loading of samples and to allow proper treatment comparisons. Thus, the mean densitometric results of ₁ mRNA were normalized to the values obtained for ELF as shown in Figure 5b. Ozone did not increase the levels of ₁ mRNA in epithelial cells at 18 h after exposure and mRNA levels of ELF were unchanged. Therefore, O₃ does not appear to regulate ₁ integrins at the transcriptional level.

View larger version (43K): [in this window] [in a new window]   View larger version (49K): [in this window] [in a new window]   Figure 5.   (a) Northern blot analysis of 1 mRNA in bronchial epithelial cultures. Epithelial cells were exposed to either air or 0.5 ppm O3 for 6 h and allowed to recover for an additional 18 h prior to analysis as described in MATERIALS AND METHODS. Equal amounts (15 µg) of cellular RNA were separated by formaldehyde-agarose gel electrophoresis and band levels of 1 and ELF mRNA were determined by Northern blotting. There was no detectable increase in 1 mRNA levels after O3 exposure. (b) Effects of O3 exposure at 0.5 ppm for 6 h on 1 transcript levels in bronchial epithelial cells. Results from four different experiments are expressed as mean ± SE of IOD (Integrated Optical Density) units. The data were normalized to the values of ELF mRNA. O3 did not upregulate 1 transcript levels.

Immunoprecipitation

Immunoprecipitation analysis of newly synthesized ₁ subunit was performed at 24 h following O₃ exposure to determine whether redistribution of ₁ integrin subunit is accompanied by increased synthesis of the ₁ protein. Cultured cells were allowed to recover for 18 h after exposure and then radiolabeled with a mixture of [³⁵S]L-methionine and [³⁵S]L-cysteine for an additional 6 h to examine synthesis of ₁ integrin protein during the period in which ₁ redistribution occurred. Epithelial cells that remained attached to their matrix were trypsinized off the inserts and extracted with Triton X-100. Newly synthesized ₁ subunit was isolated by immunoprecipitation analysis of the lysates using antibodies specific for the integrin ₁ subunit and the ₅₁ integrin receptor. The immunoprecipitation results were normalized to either cell numbers or radioactivity counts. Figure 6a (left panel) shows the ₅ and ₁ subunit proteins of the integrin receptor after air and O₃ exposures as obtained by immunoprecipitation of detergent-soluble proteins from equal numbers of cells. This analysis demonstrated (1) the presence of the ₅₁ integrin complex in bronchial epithelial cultures and (2) that O₃ increased detergent solubility of the ₅₁ integrin complex since lysates from O₃-exposed cells contained greater amounts of both the ₅ and ₁ subunits compared with air controls. To determine the effects of O₃ on the levels of detergent-soluble ₁ subunit, the mean results of immunoprecipitation experiments using antibodies specific to either the ₁ subunit (panel I) or the ₅₁ integrin (panel II) are shown in Figure 6b. At 24 h after exposure, O₃ increased the detergent-soluble amounts of total cellular ₁ subunit and the ₁ subunit of the ₅₁ integrin by 90 ± 15% and 60 ± 10%, respectively. Additional experiments were then done to determine whether these observed effects were due to increased protein synthesis of ₁ integrins. Figure 6a (right panel) shows the  and ₁ subunit proteins immunoprecipitated with an antibody against total cellular ₁ subunit from detergent lysates containing equal radioactivity counts. Levels of newly synthesized ₁ integrin subunit and its associated  subunit were not different in air- and O₃-exposed cells. These results suggest that O₃ does not alter the synthesis of ₁ integrins in bronchial epithelial cells. Taken together, it appears that O₃ increases detergent extractibility of ₁ integrins, resulting in the detection of greater amounts of ₁ integrins in detergent lysates of O₃-exposed cells compared with air controls, rather than a true increase in the synthesis of ₁ integrins.

View larger version (82K): [in this window] [in a new window]   View larger version (34K): [in this window] [in a new window]   Figure 6.   (a) Immunoprecipitation analysis of 1 integrin subunit in bronchial epithelial cultures. Cultures were exposed to either air or 0.5 ppm O3 for 6 h and immunoprecipitation analysis was done at 24 h after exposure as described in MATERIALS AND METHODS. Left panel  shows the 5 and 1 subunit proteins as obtained by immunoprecipitation analysis of detergent-soluble proteins from equal numbers of cells with polyclonal antiserum against the 51 integrin. Right panel  shows the  and 1 subunit proteins immunoprecipitated with an antibody against total cellular 1 subunit from detergent lysates containing equal radioactivity counts. Detergent-soluble levels of both the  and  subunits were higher in O3-exposed cells and the change was not due to increased synthesis of 1 integrins. (b) Effects of O3 exposure at 0.5 ppm for 6 h on detergent-soluble 1 levels in bronchial epithelial cells. The relative amounts of the 1 subunit in lysates from equal numbers of air- and O3-exposed cells were determined by immunoprecipitation analysis using antibodies against the 1 subunit (panel I; n = 3 experiments) and against 51 (panel II; n = 4 experiments). Results are expressed as mean ± SE of IOD units. Ozone exposure elevated extractible levels of total 1 subunit and the 1 subunit specific to the 51 integrin; *P < 0.05, O3 versus air.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study was designed to determine the modulatory effects of O₃ on ₁ integrins in cultured bronchial epithelial cells. Our results show that O₃ causes marked cell loss and altered localization of the ₁ integrin subunit to the apical surfaces of the remaining attached cells. The redistribution of the ₁ subunit after O₃ exposure appears to be part of a remodeling response of the remaining epithelial cells following O₃-induced injury. Following in vivo epithelial damage in the trachea, basal cells migrate and proliferate to ensure re-epithelization of denuded areas (31). Migration, spreading, and proliferation of epithelial cells are all known to be dependent on ₁ integrins, and altered distribution of ₁ integrins occurs during wound healing of airway epithelial cultures (9, 32). Therefore, the observed changes of ₁ integrins following O₃ exposure appear to be involved in the recovery of the epithelial cell layer.

Ozone disrupts tissue morphology and causes epithelial shedding (2). Whether epithelial tissue is capable of recovery is dependent on the level and duration of exposure. In a study of the kinetics of cell death following a single exposure to 0.5 ppm O₃, time-dependent necrosis and sloughing of nasal epithelial cells were found at 4, 6, and 8 h after exposure (33). Epithelial cell numbers decreased 17% at 8 h and recovered to 10% at 20 h after exposure. Recovery of epithelial cell numeric density occurred at later time points and was associated with increased epithelial cell proliferation. A similar temporal incidence of epithelial necrosis and proliferation was observed in rhesus monkeys exposed to 0.64 and 0.96 ppm O₃ (34). In all these studies, the reversible nature of O₃-induced injury was associated with regeneration of epithelium, which is dependent on the ability of the remaining epithelial cells to spread, migrate, proliferate, and differentiate to regenerate the pseudostratified columnar epithelium.

The observed increase of ₁ integrins at the cellular apical surface following O₃ exposure (Figure 3) is consistent with the apical localization of ₁ integrins displayed by migrating airway epithelial cells (9). Epithelial cells maintain a dynamic expression of integrins and the cellular distribution of integrins is indicative of certain epithelial cell functions. Resting epithelial cells express ₁ integrins at basolateral sides of the cell (9). Basolateral distribution of the ₁ integrins serves to anchor cells to their matrix and to maintain cell-cell interactions (5). On the other hand, migration and proliferation of epithelial cells require a lesser degree of cell-matrix interaction, which might explain the apical redistribution of integrins.

The redistribution of the ₁ subunit that we observed was not accompanied by increases of ₁ mRNA levels (Figure 5b) or ₁ protein synthesis (Figure 6a). These results are in agreement with previous reports that oxidative treatment causes redistribution of integrins from basal to apical cell surfaces without altering total cellular levels of integrins (16, 37). Gailit and colleagues (16) reported that the expression of ₃, ₄, and _v integrin subunits in BS-C-1 epithelial cells is not altered by exposure to hydrogen peroxide. Furthermore, hydrogen peroxide treatment failed to increase the total cellular amounts of ₁ integrin subunit in human umbilical vein endothelial cells (37). However, disruption of focal adhesion sites and redistribution of these integrin subunits from basal to apical surfaces occur following treatment with hydrogen peroxide (16, 37). The redistribution of integrins results in decreased cell adhesion and is associated with altered cellular morphology.

The two studies mentioned above were performed with cells cultured on plastic surfaces, which inhibits differentiation of epithelial cells (38) and alters cellular distribution and expression of integrins (39). In comparison, our studies were done with epithelial cells cultured at an air-cell interface on collagen-coated membranes, which facilitates cellular differentiation. Under these culture conditions, ₁ integrins were expressed diffusely in the cytoplasm and preferentially at cellular contacts, similar to those seen in vivo (3).

The expression of ₁-containing integrin heterodimers is usually regulated by altering transcription and synthesis of the associated  subunits (23, 25). The ₁ integrin subunit is constitutively expressed in excess and thus it is upregulated to a lesser degree than the associated  subunits. Sheppard and associates (23) studied the transcriptional regulation of the ₁ integrin subunit by TGF-₁ in guinea-pig airway epithelial cells. Prolonged TGF-₁ treatment for 12, 24, and 48 h failed to increase ₁ mRNA. However, TGF-₁ stimulated a 6-fold increase in mRNA levels of the ₅ subunit at 12 h and remained at elevated levels at 48 h (23). The cell-surface expression of ₁-containing heterodimers was enhanced with TGF-₁ stimulation for 24 to 72 h, which shows the slow kinetics of ₁ integrin synthesis and surface expression. The effects of TGF-₁ on ₁ mRNA appears to be cell-specific because studies with human smooth muscle cells and fibroblasts show that TGF-₁ treatment slowly increased ₁ mRNA levels, which were maximal at 12-48 h after the onset of treatment (25). In comparison, a more prominent effect of TGF-₁ on several  integrin subunits was evident at earlier time points (23, 25).

Our results indicate that O₃ increased detergent extractibility of cellular ₁ integrins (Figure 6b). A significant proportion of integrins and related proteins such as ICAM-1 associate with the cytoskeleton (40, 41). Differential extractibility of integrins results from altering the interactions of integrins and cytoskeleton components. Cytochalasins disrupt the cytoskeleton, resulting in increased detergent extractibility of integrins and diminished cell adhesion (41, 42). Similarly, ozone may cause disorganization of the cytoskeleton and thereby enhance integrin extractibility.

The delayed effect of O₃ on epithelial integrins, seen at 24 h rather than immediately following exposure, suggests that the mechanism by which O₃ alters cellular ₁ integrins might be through the release of inflammatory mediators. One hypothesis is that O₃ affects epithelial integrins through released chemokines, which have an autocrine activity. In respiratory epithelial cultures, enhanced cytokine production occurs after O₃ exposure (21, 43) and cytokines are known to upregulate the expression of integrins (15). Whether these events occur in vivo and could be modulated by mediators released from other lung cells is not yet determined.

The use of primary cultures of airway epithelial cells has been a valuable tool in studying the mechanisms underlying lung diseases and inflammation (44). Airway epithelial cells cultured at an air-liquid interface have growth and differentiation characteristics of in vivo epithelial cells, and form a pseudostratified epithelial layer similar to that seen in vivo (45). The major cell types of the epithelial layer are basal, ciliated, and goblet cells. These cells exhibit different morphologic, biochemical, and functional properties. The highly proliferative basal cells are believed to be precursors of the more differentiated ciliated and goblet cells (46). Basal cells in the human epidermis have been shown to express higher levels of ₁ integrins (47, 48). Whether the observed effects of O₃ on ₁ integrins in bronchial cell cultures are specific to a certain cell type is not yet determined.

In conclusion, O₃ alters cellular localization of ₁ integrin subunit in cultured bronchial epithelial cells, which is preceded by marked cell loss. Integrins appear to play an essential role in the regeneration of the lung epithelial cell layer following O₃-induced injury.