Negative impact of DEP exposure on human airway epithelial cell ...

Am J Physiol Lung Cell Mol Physiol 284: L119–L132, 2003. First published August 23, 2002; 10.1152/ajplung.00039.2002.

Negative impact of DEP exposure on human airway epithelial cell adhesion, stiffness, and repair BLANDINE DOORNAERT,* VALERIE LEBLOND,* STEPHANE GALIACY, GABRIEL GRAS, EMMANUELLE PLANUS, VALERIE LAURENT, DANIEL ISABEY, AND CHANTAL LAFUMA Faculte´ de Me´decine, Faculte´ des Sciences Universite´ Paris XII, Institut National de la Sante´ et de la Recherche Me´dicale, Unite´ 492 de Physiopathologie et The´rapeutique Respiratoire, 94010 Cre´teil; Faculte´ de Me´decine, Laboratoire d’Enzymologie et de Chimie des Proteínes, 37032 Tours Cedex; and De´partement des Sciences du Vivant, Service de ´ Neurovirologie, Commissariat a` l’Energie Atomique, 92265 Fontenay aux Roses, France Submitted 23 January 2002; accepted in final form 9 August 2002

Doornaert, Blandine, Valerie Leblond, Stephane Galiacy, Gabriel Gras, Emmanuelle Planus, Valerie Laurent, Daniel Isabey, and Chantal Lafuma. Negative impact of DEP exposure on human airway epithelial cell adhesion, stiffness, and repair. Am J Physiol Lung Cell Mol Physiol 284: L119–L132, 2003. First published August 23, 2002; 10.1152/ajplung.00039.2002.—Epidemiological and experimental studies suggest that diesel exhaust particles (DEPs) may be associated with increased respiratory mortality and morbidity. Several recent studies have also shown that DEPs increase the production of inflammatory cytokines by human bronchial epithelium (HBE) cells in vitro. The present study investigates the effects of DEPs on the interaction of l-HBE cells (16HBE14o-) with the cell and matrix microenvironment based on evaluation of integrin-type cell/ matrix ligand expression, cytoskeleton (CSK) stiffness, and matrix remodeling via matrix metalloproteinase (MMP)-1, MMP-2, and MMP-9 expression. The results showed that DEP exposure induced: 1) a net dose-dependent decrease in CSK stiffness through actin fibers, 2) a concomitant specific reduction of both ␣3- and ␤1-integrin subunits extensively expressed on the HBE cell surface, 3) a decrease in the level of CD44, which is a major HBE cell-cell and HBE cell-matrix adhesion molecule; and 4) an isolated decrease in MMP-1 expression without any change in tissue inhibitor of matrix metalloproteinase (TIMP)-1 or TIMP-2 tissue inhibitors. Restrictive modulation of cell-matrix interaction, cell-cell connection, CSK stiffness, and fibrillary collagen remodeling results in a decreased wound closure capacity and an increased deadhesion capacity. In conclusion, on the basis of these results, we can propose that, in addition to their ability to increase the production of inflammatory cytokines, DEPs could also alter the links between actin CSK and the extracellular matrix, suggesting that they might facilitate HBE cell detachment in vivo. diesel exhaust particle

air pollution has been observed over the last 15 years, characterized by high concentrations of atmospheric hydrocarbons, nitrogen oxides,

A PROGRESSIVE CHANGE IN

* B. Doornaert and V. Leblond contributed equally to this work. Address for reprint requests and other correspondence: C. Lafuma, INSERM U492 de Physiopathologie et The´rapeutique Respiratoires, Faculte´ de Me´decine, 8 rue du Geńe´ral. Sarrail, 94010 Cre´teil Cedex (E-mail: [email protected]). http://www.ajplung.org

ozone, and especially respirable particulate matter (PM 10, PM 2.5, and now PM 0.1; see Ref. 16). These particles are mainly derived from diesel combustion, which is particularly high in France because of the extensive development of diesel engines (30% of all cars in France in 2000; see Ref. 38). Diesel exhaust particles (DEPs) are composed of a carbonaceous core with adsorbed traces of heavy metals and a vast number of organic compounds, such as polyaromatic hydrocarbons (45). DEPs tend to form aggregates 0.1–0.5 ␮m in diameter, placing these particles within the respirable range (20). Airway epithelial cells are the primary target for air pollution and may play a key role in the pathophysiology of airway diseases. Recent studies have emphasized the role of DEPs in the development of an inflammatory response of bronchial epithelial cells. In vitro studies have shown that exposure of human bronchial epithelial (HBE) cells to DEPs significantly increased cell electrical resistance, decreased ciliary beat frequency (4), and induced release of interleukin (IL)-1␤, IL-8, and granulocyte-macrophage colony-stimulating factor (GM-CSF) inflammatory cytokines (4, 6, 7). However, although the proinflammatory impact of DEPs on bronchial epithelium is starting to be elucidated, little or no information about the other possible biological effects of DEPs is available at the present time. In particular, bronchial epithelial cells establish intercellular tight junctions and adhere to underlying basement membranes, which in turn may regulate bronchial epithelial cell shape, cell proliferation, and differentiation. In pulmonary diseases such as asthma, damage to bronchial epithelium is often associated with disruption of the underlying basement membrane and cell-cell interactions (34). On the basis of these data, it appeared important to determine whether DEPs may impair the interaction of bronchial epithelial cells with the matrix and cellular microenvironment. The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1040-0605/03 $5.00 Copyright © 2003 the American Physiological Society

L119

L120

DEP AND BRONCHIAL EPITHELIAL CELL-MATRIX INTERACTION

The present study investigates modulation of cellcell interactions between bronchial epithelial cells in response to DEP exposure, in terms of E-cadherin and CD44 (Cluster of Differentiation) adhesion molecule expression, since E-cadherin is responsible for homotopic adhesive interactions with E-cadherin on the surface of opposing cells (19), whereas hyaluronic acid receptor CD44 has a variety of functions, including participation in cell-cell adhesion by heterotopic and homotopic adhesions as well as cell-matrix interactions (11, 36). Cell-matrix interactions between bronchial epithelial cells and underlying basement membrane were also investigated by evaluating integrin expression in response to DEPs. Each transmembrane integrin is composed of ␣- and ␤-subunit heterodimers, and the combination of a particular ␤-subunit with a given ␣-subunit imparts a defined matrix ligand specificity to each integrin (8, 23). In addition, because integrins link the extracellular matrix (ECM) components to the cytoskeleton (CSK) via actin filaments (F-actin), and since the focal adhesion points mediated by these integrins are involved in cell shape, stiffness, and migration, actin CSK stiffness in response to DEP exposure was also investigated. Adhesive contacts may be broken by extracellular proteolytic matrix metalloproteinases (MMPs). MMPs degrade all components of the ECM and play key roles in normal physiological processes involving ECM remodeling, such as wound healing, angiogenesis, or development. More particularly, MMP-1 cleaves fibrillary collagen at a unique site in the triple helix of the protein, whereas MMP-2 and MMP-9 (gelatinases A and B, respectively) specifically degrade gelatin and type IV collagen, and to a lesser extent, laminin, entactin, fibronectin, fibrillin, and elastin. Our previous studies reported that these two gelatinases were expressed by HBE cells (9, 30, 46). MMP activity is also regulated by specific tissue inhibitors of matrix metalloproteinases (TIMPs; see Ref. 35), and our previous studies have demonstrated that the two tissue inhibitors (TIMP-1 and TIMP-2) were also expressed by HBE cells (47, 48). The balance between these MMPs and TIMPs present in these cells in response to DEP exposure was therefore examined. Overall, our results consistently showed that DEPs are able to alter bronchial epithelial cell-matrix interactions and epithelial cell-cell connections and ECM remodeling, in association with increased deadhesion cell capacity and a decreased rate of cell repair in response to mechanical injury. MATERIALS AND METHODS

Cell Cultures All parameters were investigated in cultures of a HBE cell line (16HBE14o-), allowing us to perform a large number of reproducible experiments by using methods as different as flow cytometry, magnetic twisting cytometry (MTC), or gel zymography. Some primary HBE cell cultures were also AJP-Lung Cell Mol Physiol • VOL

carried out to compare the data obtained with the 16HBE14o- cell line. Isolation and culture of primary HBE cells. HBE biopsies were obtained by fibroscopy in several patients investigated for lung cancer. Biopsies were taken away from the tumor. Pathological examination confirmed the presence of normal bronchial mucosa on each specimen. All the patients whose biopsies were used for collection of epithelial cells gave their informed written consent, and the experimental design was approved by a relevant ethical committee (Comite´ Consultatif pour la Protection des Personnes dans la Recherche Me´ dicale) in accordance with good clinical practice and French bioethical laws. Primary (p) HBE cells were cultured according to the modified method of Baeza-Squiban et al. (3). Two or three explants (0.5 ⫻ 0.5 mm) were placed on sterile 35-mm plastic dishes (Nunc, Napperville, IL) coated with a collagen G matrix (bovine type I and III collagens; Biochrom, Berlin, Germany). The explants were covered with 600 ␮l of culture medium and incubated for 24 h. Culture medium (2 ml) was then added to each dish. The culture medium consisted of serum-free DMEM-Ham’s F-12 (1:1; Life Technologies, Cergy-Pontoise, France) supplemented with 2% Ultroser G (UG; Life Technologies), 1% antibiotics (10,000 U/ml penicillin G sodium, 10,000 ␮g/ml streptomycin sulfate), 25 ␮g/ml amphotericin B (Life Technologies), and 2 mM glutamine (Life Technologies) at a final concentration. Explants were placed in a humidified incubator at 37°C in 5% CO2-95% air. The culture medium was changed every 2 days. Cells were cultured for 2 wk until confluence of p-HBE cells. The same explants were also successively transferred to new sterile coated plastic dishes, at 5- to 8-day intervals, to initiate new p-HBE cell cultures. HBE cell line culture. The 16HBE14o- cell line (a gift from Dr. D. C. Gruenert, University of Vermont, to B. Housset, Centre Hospitalo-Universitaire Henri Mondor, Cre´ teil, France) was cultured on collagen G- and matrix-coated dishes. Different dish types (Nunc) were used depending on the experiment: 48-well plates for cytotoxicity assays, 24-well plates for wound healing assays, 8 Lab-Tek wells for actin structure investigation and for proliferating cell localization during wound healing, 32-mm2 dishes for MTC, and 25-cm2 flasks for flow cytometry assays. The culture medium was the same as that described above. Experiments were performed at passages 25 to 40, and all cultures were incubated in a humidified incubator at 37°C in 5% CO2-95% air. The culture medium was changed every 2 days. Cell treatment by DEPs or carbon black. Carbon black, which represents the carbonaceous core of DEPs, was obtained from Sigma (L’Ile d’Abeau Cheˆ ne, France) and was used to test the involvement of the carboneous core in comparison with entire DEPs. Diesel PM SRM 1650 was purchased from the National Institute of Standards and Technology (Gaithersburg, MD). Stock solutions of particles were dispersed in 0.04% dipalmitoyl phosphatidylcholine (DPC; Sigma) in distilled water and then sonicated in ice for 5 min at maximum power (8 kilocycles). As previously described (6), DPC was used to mimic the lung surfactant lipid, and entire DEPs or carbon black were used at the final concentration of 1–100 ␮g/ml in the cell culture medium. Equivalent volumes of DPC alone were also used as DPC controls for each particle concentration. Subconfluent cultures of l-HBE cells were treated in the presence of particles for 24 h and compared with nontreated cells (controls) or DPC-treated cells. Concerning p-HBE cells, they were first incubated in culture medium for 24 h (control cells) and then treated with DPC 284 • JANUARY 2003 •

www.ajplung.org


solvent, DEPs, or carbon black for the next 24 h (treated cells) after changing the culture medium. Cytotoxicity Assays DEP, carbon black, or DPC cytotoxicity was evaluated using Neutral Red vital staining by estimating the cell viability measured as the uptake of the dye. Nontreated l-HBE cells or cells treated with DPC or particles were washed two times in PBS and were incubated for 1 h in the presence of 10 ␮g/ml Neutral Red (Sigma). After removing the culture medium and three washes with PBS, Neutral Red associated with living cultured cells was extracted with 200 ␮l/well 1% acetic acid in 50% ethanol, at ⫺20°C, and absorbance was measured at 520 nm with a microplate photometer (SoftMax). Cytotoxicity measurements were carried out after 24, 48, and 72 h of treatment. DMSO (5%; Sigma) cytotoxicity was measured as a positive control, and collagen G matrix alone was investigated as Neutral Red background staining. DEP, carbon black, or DPC cytotoxicity after 24 h of treatment was also evaluated using the methylthiazoletetrazolium (MTT) assay. Briefly, exponentially growing cell culture cells were incubated for 90 min in the presence of 500 ␮g/ml MTT (Sigma). After removing the culture medium, reduced MTT product associated with viable cells was extracted with 400 ␮l/well isopropanol containing 8% 1 N HCl, and absorbance was measured at 450 nm with a microplate photometer. Cytotoxicity measurements were carried out after 24 h of treatment. DMSO (3 and 4%) cytotoxicity was measured as a positive control. Evaluation of Particle Phagocytosis or Cell Contact Phagocytosis or cell contact of DEPs and carbon black was investigated by optical microscopy and flow cytometry. Optical microscopy. Treated l-HBE cells were dissociated with enzyme-free cell dissociation buffer (131510–14; GIBCO-BRL, Cergy-Pontoise, France). For each particle concentration, phagocytic cells were counted using a hemocytometer (Malassez device) with a optical microscope (Zeiss). Flow cytometry. DEP, carbon black phagocytosis, or cell contact was investigated by flow cytometry (FACS LSR; Becton-Dickinson) after cell dissociation with enzyme-free cell dissociation buffer. The following two parameters were measured: 1) the side scatter (SSC) parameter, which assesses the more granular appearance of epithelial cells in response to phagocytosed or membrane-associated carbon black and DEPs and 2) the natural fluorescence of cells, which was enhanced only in response to phagocytosed and cell membrane-associated DEPs because of the autofluorescence of polyaromatic hydrocarbons adsorbed on the DEP surface (530 nm specific emission wave length for 488 nm specific excitation wave length). This property was used previously by other authors to investigate the polycyclic aromatic hydrocarbons associated with air pollutants (24, 42). F-actin Staining with Fluorescent Phallotoxin and Confocal Microscopy Control or treated l-HBE cells were rinsed two times with warm CSK buffer (in mM: 50 NaCl, 300 sucrose, 3 MgCl2, and 10 PIPES, pH 6.8), which maintains CSK integrity, as previously described (10). Cells were then fixed with CSK buffer supplemented with 1% glutaraldehyde (Sigma) for 15 min. Glutaraldehyde was removed, and samples were rinsed two times for 2 min with warm CSK buffer. Cells were permeabilized for 3 min with CSK buffer supplemented with saponin (25 ␮g/ml; Sigma) and were then rinsed three times AJP-Lung Cell Mol Physiol • VOL

L121

with CSK buffer. Rhodaminated phalloidin (1.5 ␮M; Sigma) was dissolved in CSK buffer and added to each sample for 30 min in darkness and in a humid chamber at room temperature. Coverslips were rinsed two times for 5 min with CSK buffer, and the last rinse was performed with ddH2O. Coverslips were mounted on slides with the cell side facing downward in Vectashield mounting medium (Vector Laboratories). Samples were stored overnight at 4°C before examination by laser confocal microscopy using an LSM 410 inverted microscope (Zeiss), comprising two internal heliumneon lasers and one external argon ion laser. Image processing was performed using LSM 410 software. Cell fields were randomly selected, brought into focus using a ⫻40/1.25 or 63/1.25 numeric aperture Plan Neofluor objective under transmitted light bright-field conditions, and briefly examined. Cross-sectional images were recorded under confocal conditions and used to establish a plane of focus above the glass surface. Optical sections were recorded every 1 ␮m to reveal intracellular fluorescence, and image reconstitution was performed. CSK and Actin CSK Stiffness Evaluation by MTC Control or treated l-HBE cells were incubated in UG-free medium supplemented with 1% BSA (Sigma) for 30 min at 37°C to block nonspecific binding. Carboxyl ferromagnetic beads (4.0–4.5 ␮m diameter; Spherotec) were coated with arginine-glycine-aspartic acid (RGD) peptide according to the manufacturer’s procedure (Telios Pharmaceuticals). Coated beads were incubated in serum-free medium supplemented with 1% BSA for 30 min before use. Beads were then incubated with the cells (30 ␮g/dish) for 30 min at 37°C in 5% CO2-95% air, and unbound beads were washed away with serum-free medium supplemented with 1% BSA. Dishes were then placed in the magnetocytometer. A brief 1,500-G magnetic pulse was first applied to magnetize all surfacebound beads in a unique horizontal direction, and a magnetic torque was then generated by applying an orthogonal uniform magnetic field (42 G). Associated changes in angular strain of the beads were measured by an on-line magnetometer and subsequently transformed to analog data with an acquisition system (Acqknowledge III; BIOPAC). Stiffness was defined as the ratio of stress to angular strain (43). This ratio measures the cell capacity to resist a local deformation, corresponding to CSK stiffness (Pa). After a first measurement of CSK stiffness, cytochalasin D (cyto D; Sigma), an agent that depolymerizes actin microfilaments, was added (10 ␮g/ml) for 25 min, and CSK stiffness was measured again. The difference of CSK stiffness before and after cyto D was calculated (Pa) and was representative of actin CSK stiffness or cell tone. Quantification of HBE Cell Adhesion Molecule Level by Flow Cytometry Assays Expression of cell surface adhesion molecules. All washings were performed in PBS supplemented with 0.01% NaN3 (Sigma). All centrifugations (500 g) were performed for 10 min at 4°C. l-HBE or p-HBE cells were dissociated by enzyme-free cell dissociation buffer (GIBCO-BRL). The cell suspension was centrifuged and incubated for 30 min at 4°C in PBS supplemented with 0.01% NaN3 and 10% AB human serum (AbCys, Paris, France) to avoid nonspecific binding to Fc fragments. Cells (105 to 2 ⫻ 105) were then incubated with saturating concentrations of either adhesion molecule-specific antibodies, purified mouse or rat isotype-matched controls (IgG1, IgG2a, or IgG2b; Immunotech, Marseille, France), or preimmune goat serum (Life Technologies) for 30 min at 284 • JANUARY 2003 •

www.ajplung.org

L122


4°C. We used monoclonal antibodies for ␣L, ␣M, ␣x, ␣1, ␣2, ␣3, ␣4, ␣5, ␣6, ␣v, ␤1, ␤2, ␤3, CD44, E-cadherin antigens (Immunotech), or for ␤4 (Novocastra Laboratories, Newcastle, UK) and polyclonal antibodies for ␤5- and ␤6-antigens (Santa Cruz Biotechnology). After incubation with specific antibodies or isotype-matched controls, cells were washed two times with PBS supplemented with 0.01% NaN3 and labeled with 50 ␮l appropriate secondary antibody for 30 min [either FITC-conjugated F(ab⬘)2 rabbit anti-mouse immunoglobulins (DakoPatts, Glostrup, Denmark), FITC-conjugated F(ab⬘)2 goat anti-rat immunoglobulins, or FITC-conjugated F(ab⬘)2 rabbit anti-goat immunoglobulins (Jackson Immunoresearch Laboratories)]. These secondary antibodies were diluted in PBS supplemented with 0.01% NaN3 and 10% FCS at dilutions of 1:20 for the first antibody and 1:50 for the other two antibodies. Cells were washed two times, fixed in 200 ␮l CellFix (Becton-Dickinson, San Jose, CA) solution containing 1% formaldehyde and 0.1% NaN3, and stored at 4°C in the dark until analysis. Each mean adhesion molecule-specific fluorescence intensity (MFI) was determined for 104 viable cells selected from total cells using an LSR flow cytometer (BD Bio Science) and was corrected for corresponding isotype control MFI. To determine whether the action of DEP on adhesion molecule expression was partly phagocytosis dependent, adhesion molecule expression was also assessed in cells treated by cyto D (10 ␮g/ml) for 24 h in the presence of DEP (100 ␮g/ml). Quantification of the adhesion molecule intracellular pool. For each experimental condition, cell surface adhesion molecules were immunostained as described above. Cells were then fixed and permeabilized with IntraPrep Permeabilization reagent according to the manufacturer’s recommendations (Immunotech). Briefly, cells were fixed for 15 min at room temperature with 5.6% formaldehyde, washed with PBS, centrifuged for 5 min at 500 g, and gently resuspended with a saponin solution kit for 5 min. After cell permeabilization, intracellular adhesion molecules were then stained. Cells were washed two times, fixed in 200 ␮l CellFix solution, and stored at 4°C in the dark for ⬍24 h. To quantify both membrane and intracellular adhesion molecules (total pool), the following four immunostaining combinations were performed: 1) cell membrane-specific adhesion molecule immunostaining followed by isotype-matched control immunostaining after cell permeabilization; 2) cell membrane isotype-matched control immunostaining followed by specific adhesion molecule immunostaining after cell permeabilization; 3) specific adhesion molecule immunostaining, before and after cell permeabilization; and 4) isotype-matched control immunostaining, before and after cell permeabilization. The intracellular pool of each adhesion molecule was calculated by substracting the cell-surface adhesion molecule pool from the total pool. Gel Zymography Culture supernatants were centrifuged, and samples were frozen at ⫺20°C until use. Collected media were resolved by 8% SDS-PAGE containing 1 mg/ml copolymerized porcine skin gelatin according to the method of Heussen and Dowdle (21), excluding any reducing agents or boiling procedures. After electrophoresis, the gel was washed for 30 min in 2.5% Triton X-100 at room temperature to remove SDS. The gel was then incubated overnight at 37°C in reaction buffer (100 mM Tris䡠HCl and 10 mM CaCl2, pH 7.4). After being stained with Coomassie Brilliant blue G-250 (ICN Biomedicals, Aurora, OH), gelatin-degrading enzymes (MMP-2 and MMP-9) were identified as clear zones of lysis against a blue background. Molecular AJP-Lung Cell Mol Physiol • VOL

masses of gelatinolytic bands were estimated using prestained molecular mass markers (Amersham, Buckinghamshire, UK). Activities in gel slabs can be quantified using semiautomated image analysis (Scion image), which evaluates both the surface and intensity of lysis bands after gel scanning. ELISA assays The levels of MMP-1, TIMP-1, and TIMP-2 released in cell culture media were measured by human MMP-1, TIMP-1, and TIMP-2 ELISA kits (Amersham), according to the manufacturer’s recommendations. Cell Deadhesion After reaching confluency, l-HBE cells were dissociated with enzyme-free dissociation buffer, and the number of detached cells was counted with an hemocytometer (Malassez device) every 5 min for 40 min. In Vitro Wound Healing Assay The in vitro wound healing assay was carried out as previously described (32, 37). A linear wound was made by gently scratching a subconfluent culture of l-HBE cells with a pipette tip followed by extensive rinsing with culture medium to remove all cellular debris. The area of denuded surface was quantified for 24 h. The plate was placed on an inverted microscope (Zeiss, Rueil-Malmaison, France), and an image was obtained by using a charge-coupled device camera (CCD-IRIS, Sony) connected to the microscope. The image was subsequently captured by an image-analyzing frame-grabber (Video enhancer FA320; FUTEK) and analyzed by image analysis software (Scion Image 1.3b). Wound closure was evaluated as the average surface covered by the cells after 3.5, 7, and 24 h of incubation and expressed as a percentage of exposed surface compared with the covered surface in the DPC control. Assessment of Cell Proliferation by Bromodeoxyuridine Incorporation Localization of cell proliferation. l-HBE cells were plated on Lab-Tek 8 wells (Nunc) dedicated for immunofluorescence assay. At subconfluence, cells were either treated or not treated by particles or DPC for 24 h. A linear wound was then made as described above. Cultures were incubated for another 7 h at 37°C in humidified incubators under 5% CO2 in air, and 5-bromo-2⬘-deoxyuridine (BrDU) incorporation and detection were performed as previously described (17) using a BrDU Labeling and Detection Kit according to the manufacturer’s recommendations (Boehringer-Mannheim). Samples were stored at 4°C overnight, and localization of cell proliferation was analyzed by laser confocal microscopy using an LSM 410 inverted microscope. Quantification of cell proliferation by flow cytometry. Subconfluent l-HBE cells were incubated for 30 min at 37°C with the diluted (1:1,000) kit solution containing BrDU. Cells were then dissociated by enzyme-free cell dissociation buffer and were fixed and labeled as described above. Cells were stored at 4°C in the dark until analysis. BrDU specific fluorescence was quantified using an LSR flow cytometer. Statistical Analysis Statistical analysis was performed using nonparametric tests (Statview 5.0; SAS Institute). The Kruskal-Wallis test was used for multiple comparisons. Further comparisons were performed using Dunn’s test. Significance was considered for P ⬍ 0.05. The significance of flow cytometry data was 284 • JANUARY 2003 •

www.ajplung.org


L123

assessed by the Kolmogorov-Smirnov test, using CellQuest 3.1 software (Becton-Dickinson). RESULTS

Cytotoxicity of DEPs DEPs, carbon black, and DPC cytotoxicity was evaluated on the basis of l-HBE cell staining by Neutral Red dye after 24, 48, and 72 h of treatment. A net significant cytotoxicity was found for the highest concentration of DEPs (100 ␮g/ml) and was time dependent since 31 ⫾ 4 and 83 ⫾ 8% of cells had died at 48 and 72 h, respectively. Lower levels of cytotoxicity were observed with carbon black (100 ␮g/ml) or equivalent DPC exposure for 48 and 72 h. Indeed, significant toxicity was observed in 20 ⫾ 4 and 36 ⫾ 5% of cells in response to 72 h of exposure to carbon black and DPC, respectively. The reliability of the method was attested by the high level of cytotoxicity (98%) induced by 5% DMSO at all times and by the lack of staining interference with collagen G matrix alone. No cytotoxicity was found after 24 h of treatment, whatever the treatment and the dose. This absence of any cytotoxicity was corroborated after 24 h of treatment, using MTT assay (data not shown). Because our results for longer culture times showed a time- and dose-dependent cytotoxicity induced by DEP exposure, as previously described (6, 40), subsequent investigations of DEP effects were evaluated on l-HBE cells cultured for 24 h. DEPs Are in Contact with the Cell Membrane and May Be Phagocytosed by HBE Cells The number of l-HBE cells that were in contact with DEPs and carbon black particles at different concentrations (5, 20, and 100 ␮g/ml) and/or that have phagocytosed these particles was approximately evaluated by optical microscopy by counting the cells with an hemocytometer (Malassez device). Carbon black and DEP cell phagocytosis and/or cell contact was a dosedependent process, and DEPs appeared to be more intensely in contact with the cell membrane and more phagocytosed than carbon black (Table 1). Table 1. Evaluation by optical microscopy of cell percentage in association with DEP or carbon black particles Apparent Phagocytic Cells, % Concentration, ␮g/ml

Carbon black

DEPs

5 20 100

10.6 ⫾ 4.3 20.0 ⫾ 7.0 37.4 ⫾ 10.0

12.7 ⫾ 4.6 39.1 ⫾ 8.8* 57.6 ⫾ 13.0*

Data are expressed as mean ⫾ SE; n ⫽ 4 experiments. After exposure of 1-human bronchial epithelial (HBE) cells to diesel exhaust particles (DEPs) or carbon black for 24 h, cells were dissociated with enzyme-free dissociation buffer. Cells remaining in contact with particles and cells having phagocytosed them were counted with an hemocytometer. A significant dose-dependent increase in the percentage of cells with associated particles was observed in response to DEP exposure (* P ⬍ 0.01) compared with carbon black exposure. AJP-Lung Cell Mol Physiol • VOL

Fig. 1. Evaluation by flow cytometry of diesel exhaust particle (DEP) phagocytosis and l-human bronchial epithelium (HBE) cell contact. Subconfluent l-HBE cells were exposed to carbon black (CB) or DEPs at a concentration of 5, 20, or 100 ␮g/ml or with equivalent volumes of dipalmitoyl phoshatidylcholine (DPC) for 24 h. After treatment, cells were dissociated in free-enzyme dissociation buffer and fixed in CellFix. A: side scatter (SSC) parameter corresponding to DPC and particle exposure (100 ␮g/ml). B: amplification of l-HBE autofluorescence in response to particle exposure (100 ␮g/ml).

Flow cytometry assays were also performed for quantitative determination of DEP phagocytosis and/or DEP cell contact. The DPC solution used to disperse the particles had no effect on the SSC parameter compared with control l-HBE cells. DEPs or carbon black, at the concentration of 5 ␮g/ml, did not affect SSC, probably because of the limit of sensitivity of flow cytometry. However, a significant net shift of SSC was induced in response to 100 ␮g/ml DEP exposure (Fig. 1A), whereas a less marked shift was observed in response to 20 ␮g/ml DEP exposure (data not shown). Also, only DEP cell phagocytosis or DEP cell contact was able to induce a dose-dependent increase in l-HBE cell autofluorescence evaluated by flow cytometry (Fig. 1B). 284 • JANUARY 2003 •

www.ajplung.org

L124


DEPs Are Engulfed by F-actin Localization and distribution of F-actin staining were observed by confocal microscopy by performing optical cell sections every 1 ␮m from the apical to basal

cell surface and after image reconstruction (Fig. 2). In the middle section of the confluent control and DPCtreated l-HBE cells (Fig. 2, A and B, respectively), intense and dense actin staining was observed at the cell periphery, whereas weak and scattered staining was observed in the cytosol. In carbon black-exposed l-HBE cells, internalized particles were clearly observed in the absence of any fluorescence (Fig. 2Cr), whereas little or no organized F-actin staining was observed around these internalized particles on adjacent serial sections (Fig. 2Cl). In DEP-exposed l-HBE cells, phagocytosed particles and particles located close to the cell membrane were observed in the absence of fluorescence (Fig. 2Dr), whereas colocalized and wellorganized F-actin staining was observed on adjacent serial sections and generally engulfed these particles (Fig. 2Dl). Figure 2E presents a higher-power view of F-actin engulfing DEPs. DEP Exposure Induces a Dose-Dependent Decrease in Actin CSK Stiffness Results for CSK stiffness and actin CSK stiffness (corresponding to the difference between CSK stiffness before and after cyto D) are shown in Fig. 3. The presence of DPC in the culture medium did not affect CSK stiffness or actin CSK stiffness (data not shown) of l-HBE cells. In contrast, DEP cell exposure induced a similar significant decrease in CSK stiffness, i.e., 24.4, 21, and 20% for the three levels of DEP concentration used (5, 20, and 100 ␮g/ml) and a dose-dependent decrease in actin CSK stiffness (34.4, 50, and 55%, respectively). No change in CSK stiffness or actin CSK stiffness was detected in response to DEP exposure at the lower concentration of 1 ␮g/ml (data not shown), confirming the dose-dependent response. Effects induced by carbon black exposure were usually associated with no change or a nonsignificant decrease in CSK stiffness and actin CSK stiffness (data not shown). DEPs Induce a Decrease in ␣3 (CD49)- and ␤1 (CD29)-Integrin Subunits and CD44 Receptor Expression at the l-HBE Cell Surface In the absence of DEPs, our results showed a similar pattern of adhesion molecule expression both by pFig. 2. Effect of DEP exposure on F-actin CSK. Subconfluent l-HBE cells plated on coverslips were exposed to 20 ␮g/ml CB or DEPs or with an equivalent volume of DPC for 24 h. Cells were then fixed by 1% glutaraldehyde for 15 min, and 1.5 ␮M rhodamine-labeled phalloidin was added after cell permeabilization with saponin. The coverslips were mounted and analyzed by confocal microscopy. Fluorescent F-actin staining of control cells (A) and DPC-treated cells (B) appears dense at the cell periphery but weak and scattered in the cytosol. Some phagocytosed CB particles (Cr) can be observed in the absence of fluorescence, whereas parallel fluorescent F-actin staining (Cl) appears similar to that obtained for control or DPC-treated cells. Phagocytosed DEPs and DEPs in contact with the cell membrane (Dr) can be seen in the absence of fluorescence, whereas parallel fluorescent F-actin staining (Dl) shows engulfment of phagocytosed DEPs and extension around DEP close to the cell membrane. A higher magnification (E) shows engulfment of several DEPs by F-actin. Bar: 10 ␮m.

AJP-Lung Cell Mol Physiol • VOL

284 • JANUARY 2003 •

www.ajplung.org


L125

HBE and l-HBE cells, since both culture types mainly express ␣3-integrin, ␤1-integrin, and the hyaluronic acid receptor CD44, a nonintegrin matrix adhesion molecule that also participates in maintenance of cellcell interactions. In contrast, the ␣2-, ␣5-, ␣6-, and ␣V-integrin subunits and E-cadherin appeared to be expressed less intensely by both HBE cell cultures, with no or barely detectable levels of ␤2-, ␤3-, ␣1-, ␣4-, ␣L-, ␣M-, ␣X-subunits (data not shown). However, because p-HBE cells are highly autofluorescent and to avoid any resulting artifact, flow cytometry of adhesion molecule expression in response to DEP exposure was only investigated in l-HBE cells. Expression of adhesion molecules at the l-HBE cell surface was never modified by DPC. At the concentrations of 20 and 100 ␮g/ml, DEP exposure induced a reproducible and significant dose-dependent decrease in the expression of the three major adhesion molecules (␣3- and ␤1-integrin subunits and CD44 receptor). Figure 4 shows the shift related to ␣3-, ␤1-, and CD44 immunofluorescence induced by the highest dose of DEPs. The apparent increase in staining with the isotype control antibodies did not reflect an increase in isotype control antibody binding but was only because of the poly-aromatic hydrocarbons-induced autofluorescence of l-HBE cells treated with DEPs. No effect of DEP exposure was detected at the concentrations of 5 and 1 ␮g/ml (data not shown). The effects induced by carbon black exposure were usually associated with no change or a nonsignificant decrease in ␣3- and ␤1integrin subunits and CD44 receptor (data not shown). DEPs Induce a Decrease in Intracellular Distribution of ␣3- and ␤1-Integrin Subunits and CD44 Expression

Fig. 3. Inhibitory effect of DEP exposure on cytoskeleton (CSK) and actin CSK stiffness. Subconfluent l-HBE cells were exposed to 1, 5, 20, or 100 ␮g/ml CB or DEPs or equivalent volumes of DPC for 24 h. Arginine-glycine-aspartic acid-coated ferromagnetic beads were then added to the cells. Cells in dishes were placed in the magnetic twisting cytometer. A brief 1,500-G magnetic pulse was applied to magnetize all surface-bound beads in a single horizontal direction, and a magnetic torque was then generated by applying an orthogonal homogenous magnetic field (42 G). Associated changes in angular strain of the beads were measured by the on-line magnetocytometer. CSK stiffness was defined as the ratio of shear stress to angular strain. Actin CSK stiffness was the difference between CSK stiffness before and after cytochalasin D (cyto D; 25 min treatment). Results were expressed as mean stiffness (Pa) ⫾ SE of 6 values for each experiment (n ⫽ 3). CSK and actin CSK stiffness differed in response to DEP exposure (* P ⬍ 0.05 and ** P ⬍ 0.001). A, B, and C: 5, 20, and 100 ␮g/ml of DEP exposure, respectively. AJP-Lung Cell Mol Physiol • VOL

To test whether cell internalization or intracytoplasmic retention of the ␣3, ␤1, and CD44 receptors was a possible mechanism to explain decreased expression of these molecules at the cell surface, double cell membrane and internal pool immunostaining of both integrins and CD44 receptor were performed. This double immunostaining showed the presence of a relatively large internal pool (⬃20%) comprising ␣3- and ␤1-subunits and a smaller internal pool (8%) for CD44 (Table 2). The results both confirmed the dose-dependent decrease in ␣3, ␤1, and CD44 expression on l-HBE cell surface in response to 20 and 100 ␮g/ml DEP exposure and clearly demonstrated a similar decrease in their intracellular counterparts. These results therefore supported that decreased expression of integrins and CD44 on the cell surface was not because of the increased cellular internalization of these molecules but was rather because of their global decreased expression by cells in response to DEP exposure. DEP Phagocytosis Is Not Required to Induce a Reduction of ␣3-, ␤1, and CD44 Expression at the HBE Cell Surface To assess whether or not DEP phagocytosis is required to induce a decrease in ␣3, ␤1, and CD44 expression at the cell surface, l-HBE cells were exposed for 284 • JANUARY 2003 •

www.ajplung.org

L126


CD44 adhesion molecules, by 18, 26, and 39%, respectively; 2) cyto D treatment in the presence of DEP did not prevent the decreased expression of ␤1, ␣3, and CD44 adhesion molecules; and 3) the decreased expression of ␤1, ␣3, and CD44 adhesion molecule in the presence of cyto D (39, 47, and 37%, respectively) was therefore clearly higher than the decrease observed in the absence of cyto D (29, 35, and 22%, respectively). DEP Exposure Induces a Specific Decrease in MMP-1 Protein Expression with No Change in TIMP-1 and TIMP-2 Expression by p-HBE and l-HBE Cells To assess whether DEP exposure was associated with a change in ECM remodeling, the balance of MMPs/TIMPs was investigated. MMP-2, MMP-9, MMP-1, TIMP-1, and TIMP-2 protein expression was investigated in p-HBE and l-HBE cell supernatants. Gelatin gel zymography results clearly showed that MMP-2 and MMP-9 gelatinase protein expression, released for 24 h in p-HBE and l-HBE cell supernatants, in proactivated and activated forms (72 and 68 kDa for MMP-2 and 92 and 88 kDa for MMP-9), was unchanged in response to DPC, DEP, or carbon black exposure, regardless of the dose and regardless of the type of cell culture. Figure 6 shows gel zymography results obtained for the highest dose of particles. Concomitant TIMP-1 and TIMP-2 expression evaluated by ELISA assays also remained constant (Fig. 7). In contrast, exposure to the highest DEP dose (100 ␮g/ml) induced a significant net decrease in MMP-1 protein expression and a discrete nonsignificant decrease at the lower DEP dose (20 ␮g/ml), whereas DPC or carbon black exposure did not alter this expression (Fig. 7). However, this MMP-1 decrease was only observed in p-HBE cells. Fig. 4. Inhibitory effect of DEP exposure on CD29 (␤1), CD49 (␣3), and CD44 cell surface expression (A, B, and C, respectively). Subconfluent l-HBE cells were treated with 1, 5, 20, and 100 ␮g/ml CB or DEPs or with equivalent volumes of DPC for 24 h. Cells were then dissociated by enzyme-free dissociation buffer, incubated with adhesion molecule-specific antibodies or isotype-matched controls, and treated as described in MATERIALS AND METHODS. Data obtained by flow cytometry were expressed as mean fluorescence intensity (MFI)/ cell, based on 104 cells. Data are representative of 3 independent experiments and show the reduction of each adhesion molecule in response to 100 ␮g/ml DEP exposure (P ⬍ 0.001).

24 h to DEPs (100 ␮g/ml) in the presence of cyto D (10 ␮g/ml). As reported above, cyto D is an agent that disrupts actin microfilament polymerization, which is necessary for particle phagocytosis (1), since the presence of cyto D for 24 h actively inhibited DEP phagocytosis (50–70%) but inhibited carbon black phagocytosis to a lesser extent (⬃10%). The results summarized in Fig. 5 show that 1) cyto D treatment alone induced a reproducible and significant increase in the expression of ␤1, ␣3, and AJP-Lung Cell Mol Physiol • VOL

Table 2. Inhibitory effect of DEP exposure (100 ␮g/ml) on 1-HBE cell membrane and intracellular expression of CD29 (␤1), CD49 (␣3), and CD44 Adhesion Molecules

Untreated Cells

DPC-treated Cells

DEP-treated Cells (100 ␮g/ml)

534 190

539 219

368* 126*

805 71

851 121

578* 83*

310 73

303 76

164* 51*

CD29 m i CD44 m i CD49c m i

After exposure of 1-HBE cells to DEPs or carbon black or equivalent volumes of dipalmitoyl phosphatidyl choline (DPC) for 24 h, cells were studied as described in MATERIALS AND METHODS. Adhesion molecule expression was evaluated by flow cytometry, and data were expressed as mean fluorescence intensity/cell, taking into account 104 cells. Expression of both intracellular (i) and cell membrane (m) pools of each adhesion molecule differed in response to DEP exposure compared with DPC treatment or with controls (* P ⬍ 0.001). 284 • JANUARY 2003 •

www.ajplung.org


L127

sented in Fig. 8 indicate that DEP exposure induced a dose-dependent amplification of cell detachment at the early times, since, after 5 min of incubation, 4% of l-HBE cells were still detached in response to 5 ␮g/ml DEP, whereas only 0.5% was detached in response to an equivalent dose of DPC. After 10 min of incubation, 60% of cells were detached, whereas only 22% of cells were detached in the presence of DPC alone. For a concentration of 20 ␮g/ml DEPs, the percentage of detached cells was 3% at 5 min and 82% at 10 min vs. 0.8 and 32%, respectively in the presence of DPC alone. Previous exposure to carbon black induced no change or a nonsignificant increase in the percentage of detached cells (data not shown). DEP Exposure Reduces l-HBE Cell Wound Repair Capacity Kinetic studies of cell wound repair capacity were performed 3.5, 7, and 24 h after a linear mechanical wound performed on l-HBE cell confluent cultures. Wound closure, expressed as the average percentage of surface exposed compared with the surface covered in DPC controls, was not significantly changed in response to carbon black exposure. In contrast, DEPs inhibited wound closure in a dose-dependent fashion (Fig. 9). At 24 h, the wound was totally closed regardless of the type of treatment. Alteration of l-HBE Cell Repair Capacity Induced by DEP Exposure Is Partially Due to Decreased Cell Proliferation

Fig. 5. Maintenance of decrease in CD29 (␤1), CD49 (␣3), and CD44 expression induced by DEP exposure in response to concommitant cyto D treatment (A, B, and C, respectively). Subconfluent l-HBE cells were treated for 24 h with either 100 ␮g/ml DEPs or CB or DPC alone, with 10 ␮g/ml cytochalasin D alone, or with particles in the presence of cytochalasin D. Cells were then dissociated with enzyme-free dissociation buffer, incubated with adhesion molecule-specific monoclonal antibodies or isotypematched controls, and were then treated as described in MATERIALS AND METHODS. Data obtained by flow cytometry were expressed as MFI/cell, taking into account 104 cells, and are representative of 3 independent experiments. Expression of each adhesion molecule at the cell surface significantly differed in response to cytochalasin treatment alone and to DEP exposure with or without concomitant cytochalasin D treatment (** P ⬍ 0.001).

DEP Exposure Is Associated with an Increase in l-HBE Cell Deadhesion Capacity HBE cells were left untreated or were treated with DPC, carbon black, or DEP for 24 h and then incubated for 40 min in enzyme-free cell dissociation buffer. Kinetic evaluation of cell detachment was performed every 5 min for 40 min. The results preAJP-Lung Cell Mol Physiol • VOL

At 7 h postwound, l-HBE cell proliferation was localized by immunofluorescence (BrDU incorporation and detection) and was similar at the edge of the wound and any other site of the cell culture (data not shown). To more precisely quantify a possible discrete change in HBE cell proliferation in response to DEP exposure, BrDU-specific fluorescence was analyzed by flow cytometry on subconfluent l-HBE cells. A weak and nonsignificant increase in cell proliferation was observed in response to DPC volume used to disperse DEPs at a concentration ⱖ20 ␮g/ml. Com-

Fig. 6. Effect of DEP exposure on matrix metalloproteinase (MMP)-2 and MMP-9 protein expression (gel zymography). l-HBE cells were exposed to 5, 20, or 100 ␮g/ml DEPs or CB or equivalent volumes of DPC for 24 h. Collected media were resolved by 8% SDS-PAGE in the presence of 1 mg/ml porcine skin gelatin as described in MATERIALS AND METHODS. Molecular weights of gelatinolytic bands were estimated using prestained molecular mass markers. MMP-2 and MMP-9 protein expression appeared unchanged in response to 100 ␮g/ml DEP or CB or DPC exposure. 284 • JANUARY 2003 •

www.ajplung.org

L128


Fig. 8. Stimulatory effect of DEP exposure on cell deadhesion capacity. Subconfluent l-HBE cells were exposed to 5 or 20 ␮g/ml DEPs or CB or to equivalent volumes of DPC for 24 h. After exposure, cells were incubated in enzyme-free dissociation buffer, and the number of detached cells was evaluated every 5 min with an hemocytometer (Malassez device). Data are cell detachment kinetics representative of 3 independent experiments carried out in the presence of 5 or 20 ␮g/ml DEPs. The percentage of detached cells differed in response to DEP exposure at the early times.

molecule expression, associated with impaired wound repair capacities and increased cell deadhesion potential. These results therefore suggest that L-HBE cells exposed to DEP pollution might lose their capacity to interact with the environmental ECM, thereby accentuating their shedding susceptibility.

Fig. 7. Effect of DEP exposure on MMP-1 (A), tissue inhibitor of matrix metalloproteinase (TIMP)-1 (B), and TIMP-2 (C) protein expression. l-HBE cells were exposed to 5, 20, or 100 ␮g/ml DEPs or CB or equivalent volumes of DPC for 24 h. MMP-1, TIMP-1, and TIMP-2 protein levels were investigated in collected media by ELISA according to the manufacturer’s recommendations (Amersham). MMP-1 level alone significantly differed in response to the highest dose of DEPs (100 ␮g/ml). * P ⬍ 0.05.

pared with an equivalent amount of DPC, DEP exposure induced a dose-dependent decrease in cell proliferation (Fig. 10) that was decreased by 26 and 35% in response to DEP exposure at the concentrations of 20 and 100 ␮g/ml, respectively. The effects of carbon black were not reproducible and were associated with no significant effect on cell proliferation (data not shown). DISCUSSION

The main objective of the present study was to determine whether DEP exposure could alter the balanced interaction between HBE cells and their matrix microenvironment. Our data clearly demonstrate that DEP exposure induces a dose-dependent reduction of l-HBE cell actin CSK stiffness and matrix adhesion AJP-Lung Cell Mol Physiol • VOL

Fig. 9. Inhibitory effect of DEP exposure on cell repair capacity after in vitro mechanical wound healing. l-HBE cells were seeded on 24-well plates. After confluence, the cells were exposed to 1, 5, 20, or 100 ␮g/ml DEPs or CB or equivalent volumes of DPC for 24 h. Cell monolayers were then scratched with a pipette tip and extensively rinsed with medium to remove all cellular debris. Wound closure was observed at 3.5, 7, and 24 h, as described in MATERIALS AND METHODS. Wound closure was measured as the average surface covered by the cells (mm2). Results are expressed as the mean percentage of exposed surface compared with the covered surface in the presence of DPC. Data illustrate 1 representative experiment (sextuplicates) among 3 independent experiments. The rate of wound healing in response to DEP exposure significantly differed compared with CB. 284 • JANUARY 2003 •

www.ajplung.org


L129

DEP Exposure Induces Alteration of Cell-Matrix and Cell-Cell Interactions

Fig. 10. Inhibitory effect of DEP exposure on cell proliferation. Subconfluent l-HBE cells were exposed to 5, 20, or 100 ␮g/ml DEPs or CB or equivalent volumes of DPC for 24 h. Cells were then dissociated by enzyme-free dissociation buffer, and 5-bromo-2⬘-deoxyuridine (BrDU) incorporation was performed according to the manufacturer’s recommendations (Boehringer). BrDU specific fluorescence was quantified using a FACscan flow cytometer. Data show 1 representative experiment among 3 independent experiments. Results were expressed as the percentage of cells incorporating BrDU. Cell proliferation differed in response to DEP exposure (* P ⬍ 0.05 and ** P ⬍ 0.001).

DEPs Are in Contact with the Cell Membrane and May Be Phagocytosed by l-HBE Cells First, to more clearly define our cell culture model, we analyzed contacts between l-HBE cells and DEPs by using three different approaches. Our optical microscopy analysis (Table 1), SSC determination by flow cytometry (Fig. 1A), and PAH-induced autofluorescence determination by flow cytometry (Fig. 1B) showed that DEPs were in contact with the cell membrane and/or phagocytosed by l-HBE cells in a dosedependent fashion. However, neither of the three approaches was able to clearly distinguish between cell-ingested or cell-bound particles; evaluation of phagocytosis was therefore only approximate. The obvious more intense phagocytosis of DEPs compared with carbon black may be because of their different size. The particle size of Sigma carbon black was slightly greater than that of DEPs, and it has been proposed that small particles are more readily phagocytosed by epithelial cells than larger particles (12, 13). Previous studies (22) have also reported phagocytosis of DEPs by cat bronchial epithelial cells in vivo. DEP phagocytosis is shown by the pattern of distribution of F-actin fluorescent phallotoxin staining in l-HBE-cells (Fig. 2). In control cells, cortical actin filaments are mainly organized into stress fibers involved in adhesion plaques at the basal surface and in circumferential belts, whereas scattered cytosolic actin represents small and slightly polymerized actin. In l-HBE cells exposed to DEPs, organized F-actin often engulfed the phagocytosed DEPs and also clearly extended as lamellipodia around the DEPs close to the cell membrane. This process appeared to be much more intense for DEPs than for carbon black particles and may be because of the presence of various organic compounds and metals adsorbed on the DEP surface. AJP-Lung Cell Mol Physiol • VOL

In the present study, we used MTC to evaluate possible changes in HBE cell CSK stiffness and more particularly changes in actin CSK stiffness in response to DEP exposure. MTC is a relatively new technique used to measure CSK stiffness by application of controlled mechanical stress applied directly to cell-surface integrins, using RGD-coated microbeads (43). This type of CSK stiffness measurement is directly proportional to the strength of cell-matrix interactions (29, 44), and the difference between stiffness before and after brief exposure (25 min) to cyto D reflects actin CSK stiffness. Our results demonstrated that CSK stiffness of l-HBE cells was significantly reduced by 20% in response to DEP exposure and that actin CSK stiffness was reduced in a dose-dependent fashion by 35, 50, and 55% for 5, 20, and 100 ␮g/ml DEPs, respectively (Fig. 3). Compared with the variable effect of carbon black, these results suggest that DEPs are specifically able to induce loss of CSK stiffness. Also, DEPs are specifically able to induce a loss of interactive cell-ECM properties, since flow cytometry quantification of the expression of all integrins at the l-HBE cell surface showed a significant dose-dependent decrease between 20 and 40% in major ␣3- and ␤1-integrin subunits and CD44 receptor in response to DEP exposure (Fig. 4). This decrease in both ␣3- and ␤1-integrin subunits strongly suggests a concomitant decrease in ␣3␤1-integrin heterodimer. The ␣3␤1-integrins and CD44 hyaluronic acid receptor have already been described in HBE cells (5, 33), and the marked decrease in these two major molecules in response to DEP exposure indicates that they play a key role in the loss of l-HBE cell-matrix interactions. Together, these data strongly suggest that the DEP-induced decrease in actin CSK stiffness may be at least partly because of the DEP-induced loss of adhesion molecules, in addition to the different F-actin distribution pattern. There is now increasing evidence that integrins regulate CSK assembly via a signaling pathway. More particularly, in hepatic stellate cells, Kojima et al. (25) observed that cell surface integrin binding to interstitial collagen induced a change in their CSK via a signal transduction system. Nevertheless, some studies have shown that the CSK itself may actively regulate integrin expression and binding conformation as well their mobility in the membrane (2). Moreover, because ␣3␤1-integrins and CD44 receptor are also known to be involved in cell-cell interactions (28, 41), their downregulation induced by DEP exposure may result in loss of cell-cell interconnections, predisposing to cell detachment. Our results confirm this predisposition, since the percentage of detached l-HBE cells in response to dissociation buffer was higher in the presence of DEPs (Fig. 8). Finally, we propose that this higher deadhesion capacity probably reflects an alteration of structural links between CSK actin and the ECM via ␣3␤1integrin and CD44 downregulation. Reduced expres284 • JANUARY 2003 •

www.ajplung.org

L130


sion of ␣3␤1-integrins induced by puromycin aminonucleoside was also recently proposed to contribute to glomerular epithelial cell detachment from the glomerular basement membrane (26). Decrease in ␤1, ␣3, and CD44 Expression by l-HBE Cells in Response to DEP Exposure Is Not Due to Intracellular Internalization or Cytoplasmic Sequestration The DEP-induced reduction of ␣3␤1-integrin and CD44 expression at the l-HBE cell surface raises the question of the possible intracellular internalization of these matrix adhesion molecules. Our data demonstrated that the intracellular pool related to each adhesion molecule exhibited a similar reduction to that observed for the cell membrane pool (Table 2). These data support that DEP exposure actually induces a negative global modulation of ␣3␤1-integrin and CD44 receptor expression without specific modification of intracellular internalization or cytoplasmic sequestration of cell adhesion molecules. Decrease in ␤1, ␣3, and CD44 Expression by l-HBE Cells in Response to DEP Exposure Is Not Due to DEP Phagocytosis We also raised the question about the possible involvement of the actin-mediated DEP phagocytosis process in the decrease in ␣3, ␤1, and CD44 expression. Curiously, blockade of actin-mediated DEP phagocytosis by treatment with 10 ␮g/ml cyto D for 24 h did not prevent the decrease in ␣3, ␤1, and CD44 expression by l-HBE cells after DEP exposure (Fig. 5). It is interesting to note that cyto D treatment alone induced a significant increase by ⬃20% in ␣3, ␤1, and CD44 adhesion molecule expression so that the specific inhibitory effect of DEP appeared to be largely amplified in the presence of cyto D. Overall, these data support the hypothesis that 1) the loss of actin CSK stiffness might result from a decrease in integrin expression, but the latter would not be secondary to loss of actin CSK stiffness; and 2) the negative modulation of ␣3, ␤1, and CD44 expression induced by DEPs might be because of mechanisms other than phagocytosis. Recent studies by Boland et al. (7) showed that cyto D, used at the same concentration (10 ␮g/ml) and for the same incubation time (24 h), was able to reduce DEP-induced GM-CSF secretion by l-HBE cells. DEP phagocytosis therefore appears to be a prerequisite for induction of cytokine release but not for negative modulation of ␣3, ␤1, and CD44 expression, suggesting that the respective transduction signaling pathway underlying transcriptional regulation may be somewhat different. DEP Exposure Is Associated with a Reduction of the HBE Cell Repair Process In agreement with our postulate that DEP exposure can alter the link between actin CSK and the ECM via a reduction in adhesion molecule expression, we also observed that, at 3.5 and 7 h, in vitro wound closure was dose dependent and decreased in the presence of AJP-Lung Cell Mol Physiol • VOL

DEP (Fig. 9) but seemed to occur without any apparent signs of cell migration or cell spreading. In general, cell locomotion is a dynamic interplay between cell-ECM adhesion, extension of the leading edge of the cell, and retraction of the trailing edge. To move along the ECM, cells, and more particularly p-HBE cells, first adhere to the matrix by establishing primordial cell-ECM contacts distributed at the cell front heading in the direction of migration (30, 39). In contrast with p-HBE and type II alveolar pneumocyte repair models (37), no migrating or spreading cell morphology was identified at the leading edge of the wound during the repair process of our l-HBE cell repair model. In the latter, cell proliferation appeared to be the key factor during wound healing and was decreased in response to DEP exposure, as stated by quantitative BrDU incorporation data (Fig. 10). Taken together, these results, demonstrating a reduction of the cell repair process, decreased cell proliferation, and decreased ␣3␤1-integrin and CD44 receptor expression, are in accordance with previous studies indicating a possible relationship between the level of integrin or CD44 expression and cell functions, such as cell repair or cell proliferation. High expression of CD44 was found during in vitro repair of bronchial epithelial cells after a mechanical injury (31). CD44 can also stimulate cell proliferation (14), and CD44 upregulation has been demonstrated in areas of damaged bronchial epithelium both in normal and asthmatic subjects, strongly suggesting that CD44 may be directly or indirectly associated with repair processes occurring after damage (27). As for integrins, a new concept is emerging to suggest that integrins are multifunctional and may contribute not only to epithelial cell adhesion but also to regulation of cell growth and proliferation via a signaling pathway involving mitogen-activated protein kinase (18). More particularly, proliferation of human epithelial cells was significantly inhibited by a function-altering ␣3-integrin antibody (18). DEP Exposure Appears to Weaken Matrix Remodeling by Interstitial Collagenase MMP-1 Expressed by l-HBE Cells Our results concerning ECM remodeling by subconfluent l-HBE or p-HBE cells did not demonstrate any change in MMP-2 or MMP-9 protein expression (Fig. 6) or TIMP-1 and TIMP-2 levels (Fig. 7) in response to DEP exposure. This result emphasizes the persistence of a harmonious balance between matrix gelatinases and their tissue inhibitors. Because MMP-9 expressed by p-HBE cells during wound repair has been recently proposed to play a key role in remodeling primordial contacts via specific degradation of type IV collagen (9, 30), unchanged MMP-9 expression appears relevant to the absence of any migrating event associated with our cell model in response to DEP exposure. De-Bentzmann et al. (15) recently suggested that MMP-2 overactivation associated with a limited increase in TIMP-2 was responsible for inhibition of Pseudomonas aerugi284 • JANUARY 2003 •

www.ajplung.org


nosa wound closure in vitro. In contrast, our results suggest that DEP wound closure inhibition would not depend on MMP-2 activation, probably because of different proteolytic mechanisms. Interestingly, interstitial collagenase MMP-1 production by l-HBE cells appeared to be markedly decreased in response to the highest dose of DEPs (Fig. 7), supporting the hypothesis that DEP exposure may reduce interstitial matrix turnover. Conversely, but not in opposition to the present study, recent studies by our group have demonstrated that the addition of exogenous MMP-1 collagenase to cultured alveolar epithelial cells could enhance wound healing by promoting cell migration on type I collagen (37). In conclusion, the present study proposes that, besides inducing a well-known inflammatory reaction, DEP exposure of l-HBE cells could alter cell-matrix interactions and cell cohesion via a concomitant decrease in ␣3␤1-integrin subunits and CD44 adhesion molecule and a reduction of actin CSK stiffness. Alteration of cell-matrix interactions and cell cohesion results in a dose-dependent reduction of the wound closure capacity and enhancement of the cell deadhesion capacity. A reduction in wound closure also appears to result from reduced cell proliferation and probably from alteration of matrix remodeling via an imbalance between MMP-1 and TIMP-1 and -2 in favor of inhibitors. Overall, these results support the concept that DEP exposure tends to break the link between actin CSK and the ECM, suggesting an increased potential for cell detachment from the underlying basement membrane in vivo. Finally, the fact that there was nonsignificant or no effect of carbon black particle exposure on actin CSK stiffness, cell adhesion molecule expression, wound repair capacity, the cell deadhesion process, or cell proliferation supports a key role for polycyclic aromatic hydrocarbons adsorbed on the surface of DEP. We thank Dr. Bernard Maitre for providing human bronchial biopsies and Antoine Mary for assistance with magnetic twisting cytometry. This study was supported by Institut National de la Sante´ et de la Recherche Me´ dicale, Primequal-Predit grant from Ministe`re de l’Ame´ nagement du Territoire et de l’Environnement, and a grant from Socie´ te´ Total Fina Elf. REFERENCES 1. Allen LA and Aderem A. Mechanisms of phagocytosis. Curr Opin Immunol 8: 36–40, 1996. 2. Anderson SI, Hotchin NA, and Nash GB. Role of the cytoskeleton in rapid activation of CD11b/CD18 function and its subsequent downregulation in neutrophils. J Cell Sci 113: 2737–2745, 2000. 3. Baeza-Squiban A, Romet S, Moreau A, and Marano F. Progress in outgrowth culture from rabbit tracheal explants: balance between proliferation and maintenance of differentiated state in epithelial cells. In Vitro Cell Dev Biol 27A: 453–460, 1991. 4. Bayram H, Devalia JL, Sapsford RJ, Ohtoshi T, Miyabara Y, Sagai M, and Davies RJ. The effect of diesel exhaust particles on cell function and release of inflammatory mediators from human bronchial epithelial cells in vitro. Am J Respir Cell Mol Biol 18: 441–448, 1998. AJP-Lung Cell Mol Physiol • VOL

L131

5. Bloemen PG, van den Tweel MC, Henricks PA, Engels F, Wagenaar SS, Rutten AA, and Nijkamp FP. Expression and modulation of adhesion molecules on human bronchial epithelial cells. Am J Respir Cell Mol Biol 9: 586–593, 1993. 6. Boland S, Baeza-Squiban A, Fournier T, Houcine O, Gendron MC, Chevrier M, Jouvenot G, Coste A, Aubier M, and Marano F. Diesel exhaust particles are taken up by human airway epithelial cells in vitro and alter cytokine production. Am J Physiol Lung Cell Mol Physiol 276: L604–L613, 1999. 7. Boland S, Bonvallot V, Fournier T, Baeza-Squiban A, Aubier M, and Marano F. Mechanisms of GM-CSF increase by diesel exhaust particles in human airway epithelial cells. Am J Physiol Lung Cell Mol Physiol 278: L25–L32, 2000. 8. Buck C, Albelda S, Damjanovich L, Edelman J, Shih DT, and Solowska J. Immunohistochemical and molecular analysis of beta 1 and beta 3 integrins. Cell Differ Dev 32: 189–202, 1990. 9. Buisson AC, Zahm JM, Polette M, Pierrot D, Bellon G, Puchelle E, Birembaut P, and Tournier JM. Gelatinase B is involved in the in vitro wound repair of human respiratory epithelium. J Cell Physiol 166: 413–246, 1996. 10. Capco DG and Penman S. Mitotic architecture of the cell: the filament networks of the nucleus and cytoplasm. J Cell Biol 96: 896–906, 1983. 11. Carter WG and Wayner EA. Characterization of the class III collagen receptor, a phosphorylated, transmembrane glycoprotein expressed in nucleated human cells. J Biol Chem 263: 4193–4201, 1988. 12. Churg A. The uptake of mineral particles by pulmonary epithelial cells. Am J Respir Crit Care Med 154: 1124–1140, 1996. 13. Churg A, Stevens B, and Wright JL. Comparison of the uptake of fine and ultrafine TiO2 in a tracheal explant system. Am J Physiol Lung Cell Mol Physiol 274: L81–L86, 1998. 14. Cooper NL, Bardy P, Bacani J, Kuusk U, Dougherty GJ, Eaves CJ, and Emerman JT. Correlation of CD44 expression with proliferative activity of normal human breast epithelial cells in culture. Breast Cancer Res Treat 50: 143–153, 1998. 15. De Bentzmann S, Polette M, Zahm JM, Hinnrasky J, Kileztky C, Bajolet O, Klossek JM, Filloux A, Lazdunski A, and Puchelle E. Pseudomonas aeruginosa virulence factors delay airway epithelial wound repair by altering the actin cytoskeleton and inducing overactivation of epithelial matrix metalloproteinase-2. Lab Invest 80: 209–219, 2000. 16. Eggleston S, Hakman MP, Heyes CA, Irwin JG, Timmis RJ, and Williams ML. Trends in urban pollution in the United Kingdom during recent decades. Atmospher Environ 227: 29–39, 1992. 17. Garat C, Kheradmand F, Albertine KH, Folkesson HG, and Matthay MA. Soluble and insoluble fibronectin increases alveolar epithelial wound healing in vitro. Am J Physiol Lung Cell Mol Physiol 271: L844–L853, 1996. 18. Gonzales M, Haan K, Baker SE, Fitchmun M, Todorov I, Weitzman S, and Jones JC. A cell signal pathway involving laminin-5, alpha3beta1 integrin, and mitogen-activated protein kinase can regulate epithelial cell proliferation. Mol Biol Cell 10: 259–270, 1999. 19. Gumbiner B. Cadherins: a family of Ca2⫹-dependent adhesion molecules. Trends Biochem Sci 13: 75–76, 1988. 20. Health Effects Institute. Diesel Exhaust: A Critical Analysis of Emissions, Exposure and Health Effects. A Special Report of the Institute’s Diesel Working Group. Cambridge, UK: 1995, p. 294. 21. Heussen C and Dowdle EB. Electrophoretic analysis of plasminogen activators in polyacrylamide gels containing sodium dodecyl sulfate and copolymerized substrates. Anal Biochem 102: 196–202, 1980. 22. Hyde DM, Plopper CG, Weir AJ, Murnane RD, Warren DL, Last JA, and Pepelko WE. Peribronchiolar fibrosis in lungs of cats chronically exposed to diesel exhaust. Lab Invest 52: 195– 206, 1985. 23. Hynes RO. Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69: 11–25, 1992. 24. Kataoka H. Methods for the determination of mutagenic heterocyclic amines and their applications in environmental analysis. J Chromatogr A 774: 121–142, 1997. 284 • JANUARY 2003 •

www.ajplung.org

L132


25. Kojima N, Sato M, Imai K, Miura M, Matano Y, and Senoo H. Hepatic stellate cells (vitamin A-storing cells) change their cytoskeleton structure by extracellular matrix components through a signal transduction system. Histochem Cell Biol 110: 121–128, 1998. 26. Krishnamurti U, Zhou B, Fan W, Tsilibary E, Wayner E, Kim Y, Kashtan CI, and Michael A. Puromycin aminonucleoside suppresses integrin expression in cultured glomerul epithelial cells. J Am Soc Nephrol 12: 758–766, 2001. 27. Lackie PM, Baker JE, Gunthert U, and Holgate ST. Expression of CD44 isoforms is increased in the airway epithelium of asthmatic subjects. Am J Respir Cell Mol Biol 16: 14–22, 1997. 28. Larjava H, Peltonen J, Akiyama SK, Yamada SS, Gralnick HR, Uitto J, and Yamada KM. Novel function for beta 1 integrins in keratinocyte cell-cell interactions. J Cell Biol 110: 803–815, 1990. 29. Lee KM, Tsai KY, Wang N, and Ingber DE. Extracellular matrix and pulmonary hypertension: control of vascular smooth muscle cell contractility. Am J Physiol Heart Circ Physiol 274: H76–H82, 1998. 30. Legrand C, Gilles C, Zahm JM, Polette M, Buisson AC, Kaplan H, Birembaut P, and Tournier JM. Airway epithelial cell migration dynamics. MMP-9 role in cell-extracellular matrix remodeling. J Cell Biol 146: 517–529, 1999. 31. Leir SH, Baker JE, Holgate ST, and Lackie PM. Increased CD44 expression in human bronchial epithelial repair after damage or plating at low cell densities. Am J Physiol Lung Cell Mol Physiol 278: L1129–L1137, 2000. 32. Matthay MA, Folkesson HG, Campagna A, and Kheradmand F. Alveolar epithelial barrier and acute lung injury. New Horiz 1: 613–622, 1993. 33. Mette SA, Pilewski J, Buck CA, and Albelda SM. Distribution of integrin cell adhesion receptors on normal bronchial epithelial cells and lung cancer cells in vitro and in vivo. Am J Respir Cell Mol Biol 8: 562–572, 1993. 34. Montefort S, Herbert CA, Robinson C, and Holgate ST. The bronchial epithelium as a target for inflammatory attack in asthma. Clin Exp Allergy 22: 511–520, 1992. 35. Murphy G and Reynolds JJ. Extracellular matrix degradation. In: Connective Tissue and Tis Heritable Disorders. Molecular, Genetic, and Medical Aspects, edited by Royce PM and Steinmann B. New York: Wiley-Liss, 1993, p. 287–316. 36. Perschl A, Lesley J, English N, Trowbridge I, and Hyman R. Role of CD44 cytoplasmic domain in hyaluronan binding. Eur J Immunol 25: 495–501, 1995. 37. Planus E, Galiacy S, Matthay M, Laurent V, Gavrilovic J, Murphy G, Cle´rici C, Isabey D, Lafuma C, and d’Ortho MP.

AJP-Lung Cell Mol Physiol • VOL

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

Role of collagenase in mediating in vitro alveolar epithelial wound repair. J Cell Sci 112: 243–252, 1999. Socie`te´ Française de Sante´ Publique. La Pollution Atmosphe´ rique d’Origine Automobile et la Sante´ Publique. 1996, vol. 4, p. 4. Small JV, Anderson K, and Rottner K. Actin and the coordination of protrusion, attachment and retraction in cell crawling. Biosci Rep 16: 351–368, 1996. Steerenberg PA, Zonnenberg JA, Dormans JA, Joon PN, Wouters IM, van Bree L, Scheepers PT, and Van Loveren H. Diesel exhaust particles induced release of interleukin 6 and 8 by (primed) human bronchial epithelial cells (BEAS 2B) in vitro. Exp Lung Res 24: 85–100, 1998. Symington BE, Takada Y, and Carter WG. Interaction of integrins alpha 3 beta 1 and alpha 2 beta 1: potential role in keratinocyte intercellular adhesion. J Cell Biol 120: 523–535, 1993. Wada M, Kido H, Kishikawa N, Tou T, Tanaka M, Tsubokura J, Shironita M, Matsui M, Kuroda N, and Nakashima K. Assessment of air pollution in Nagasaki City: determination of polycyclic aromatic hydrocarbons and their nitrated derivatives, and some metals. Environ Pollut 115: 139– 147, 2001. Wang N, Butler JP, and Ingber DE. Mechanotransduction across the cell surface and through the cytoskeleton. Science 260: 1124–1127, 1993. Wang N and Ingber DE. Control of cytoskeletal mechanics by extracellular matrix, cell shape, and mechanical tension. Biophys J 66: 2181–2189, 1994. Westerholm RM, Jocob A, and Li H. Chemical and biological characterization of particulate, semivolatile and gaz-phase-associated compounds in diluted heavy-duty diesel exhaust: a comparaison of three different semivolatile-phase samplers. Environ Sci Technol 25: 332–338, 1991. Yao PM, Buhler JM, d’Ortho MP, Lebargy F, Delclaux C, Harf A, and Lafuma C. Expression of matrix metalloproteinase gelatinases A and B by cultured epithelial cells from human bronchial explants. J Biol Chem 271: 15580–15589, 1996. Yao PM, Delclaux C, d’Ortho MP, Maitre B, Harf A, and Lafuma C. Cell-matrix interactions modulate 92-kD gelatinase expression by human bronchial epithelial cells. Am J Respir Cell Mol Biol 18: 813–822, 1998. Yao PM, Maitre B, Delacourt C, Buhler JM, Harf A, and Lafuma C. Divergent regulation of 92-kDa gelatinase and TIMP-1 by HBECs in response to IL-1␤ and TNF-␣. Am J Physiol Lung Cell Mol Physiol 273: L866–L874, 1997.

284 • JANUARY 2003 •

www.ajplung.org