Search

By interfering with the ability of airway epithelial cells to support repair processes, cigarette smoke could contribute to alterations of airway structures and functions that characterize chronic obstructive pulmonary disease (COPD). The current study assessed the ability of cigarette smoke extract (CSE) to alter human airway epithelial cell chemotaxis, proliferation, and contraction of three-dimensional collagen gels, a model of extracellular matrix remodeling. The volatile components contained in cigarette smoke, acetaldehyde and acrolein, were able to inhibit all three processes. Nonvolatile components contained within lyophilized CSE also inhibited chemotaxis but displayed no activity in the other two bioassays. CSE also inhibited the ability of airway epithelial cells to release transforming growth factor (TGF)-beta and fibronectin. Exogenous fibronectin was unable to restore epithelial cell contraction of collagen gels. Exogenous TGF-beta partially restored the ability of airway epithelial cells to contract collagen gels and to produce fibronectin. This supports a role for inhibition of TGF-beta release in mediating the inhibitory effects of cigarette smoke. Taken together, the results of the current study suggest that epithelial cells present in the airways of smokers may be altered in their ability to support repair responses, which may contribute to architectural disruptions present in the airways in COPD associated with cigarette smoking.

Proteolytic degradation of extracellular matrix is thought to play an important role both in emphysema and in tissue development and repair. Retinoic acid has been suggested to modify tissue injury, and in an animal model of emphysema may induce alveolar repair. Since cytokines can induce matrix metalloproteinase (MMP) production in fibroblasts and neutrophil elastase (NE) can activate MMPs, we hypothesized that retinoic acid could attenuate collagen degradation by modifying MMP production and activation. To evaluate this, human lung fibroblasts were cast into native type I collagen gels and floated in medium containing cytomix (TNF-alpha, IL-1beta, and IFN-gamma) alone or in combination with NE in the presence and absence of retinoic acid (1 microM). After 5 d, cytomix with elastase induced significant degradation of the collagen gels assessed by quantifying total hydroxyproline (41.6 +/- 1.6 microg versus 3.3 +/- 1.5 microg, P < 0.01). Retinoic acid significantly inhibited this degradation (23.3 +/- 1.5 microg versus 3.3 +/- 1.5 microg, P < 0.01). Gelatin zymography and Western blot revealed that MMP-1, MMP-3, and MMP-9 were induced by cytomix and that co-exposure to NE resulted in increased production of activated forms of these enzymes. Retinoic acid attenuated the induction and activation of MMP-1 and MMP-3. The current study, therefore, suggests that in addition to stimulating anabolic effects, retinoic acid may modulate proteolytic processes thought to contribute to tissue destruction in emphysema.

Contraction of three-dimensional collagen gels is a model of the contraction that characterizes normal healing and remodeling after injury. In the current study, we evaluated the hypothesis that a number of inflammatory factors, including tumor necrosis factor (TNF)-alpha, interleukin (IL)-1 beta, and interferon (IFN)-gamma, modulate this process by induction of prostaglandin (PG) E(2) and nitric oxide (NO) production and that these secondary mediators function in an autocrine or paracrine manner to modulate contraction. Human fetal lung fibroblasts (HFL) were cultured in type I collagen gels and floated in medium containing TNF-alpha, IL-1 beta, or IFN-gamma alone or in combination (cytomix). All cytokines inhibited the contraction significantly. The potency order was IL-1 beta, TNF-alpha, IFN-gamma. The cytomix was no more potent than was IL-1 beta alone. PGE(2) production was increased by TNF-alpha (5.0 versus 0.16 ng/ml, P < 0.01), IL-1 beta (5.3 versus 0.16 ng/ml, P < 0.01), and cytomix (5.9 versus 0.16 ng/ml, P < 0.01), and was completely inhibited by indomethacin. Indomethacin (P < 0.05) and L-NG-monomethyl arginine citrate (L-NMMA) (P < 0.05) alone both partially attenuated the inhibition of contraction caused by cytokines alone or by cytomix. Indomethacin and L-NMMA together attenuated inhibition more than either alone (P < 0.05). Exogenous PGE(2) and exogenous NO donors (DETA nononate and 5-amino-3-(4-morpholinyl)-1,2,3-oxadiazolium chloride) inhibited the contraction significantly. The protein kinase A inhibitor KT5270 and the protein kinase G inhibitor Rp-pCPT-cGMPS attenuated the inhibition induced by PGE(2) and NO, respectively. In summary, PGE(2) and NO appear to function in parallel as autocrine/paracrine mediators of cytokine-driven fibroblast inhibition of the contraction of collagen gels and may contribute to remodeling during repair and inflammation in lung disorders.

Chronic obstructive pulmonary disease (COPD) is a condition in which continuous bronchodilation may have clinical advantages. This study evaluated salmeterol, a beta-agonist bronchodilator with a duration of action substantially longer than that of short-acting beta-agonists, compared with ipratropium, an anticholinergic bronchodilator, and placebo in patients with COPD. Four hundred and five patients with COPD received either salmeterol 42 microg twice daily, ipratropium bromide 36 microg four times daily, or placebo for 12 wk in this randomized, double-blind, parallel-group study. Patients were stratified on the basis of bronchodilator response to albuterol (> 12% and > 200-ml improvement) and were randomized within each stratum. Bronchodilator response was measured over 12 h four times during the treatment period. Salmeterol provided similar maximal bronchodilatation to ipratropium but had a longer duration of action and a more constant bronchodilatory effect with no evidence of bronchodilator tolerance. Both active treatments were well tolerated. Salmeterol was an effective bronchodilator with a consistent effect over this 12-wk study in patients with COPD, including those "unresponsive" to albuterol. The long duration of action of salmeterol offers the advantage of twice daily dosing compared with the required four times a day dosing with ipratropium.

Contraction of type I collagen gels is an in vitro model of tissue remodeling. In addition to fibroblasts, some epithelial cells can mediate this process. We therefore hypothesized that alveolar epithelial cells might contract extracellular matrices and have the potential to directly participate in the remodeling of the lung after alveolar injury. A549 cells were plated on top of collagen gels, and the gels were floated in culture medium. A549 cells contracted the gels in a time- and cell density-dependent manner. A549 cells, as well as human bronchial epithelial cells (HBEC) and rat alveolar epithelial cells (RalvEC) contracted collagen gels more when they were plated on top of the gel than when they were embedded inside, in contrast to human fetal lung fibroblast (HFL1), which contracted more when cast inside. The amount of hydroxyproline in the collagen gels remained unchanged throughout the contraction. Anti-beta(1) integrin antibody inhibited A549 cell-mediated contraction. Transforming growth factor beta augmented the contraction by A549 cells as well as that by HBEC and HFL1. Prostaglandin E(2) inhibited the contraction by HFL1 but did not affect the contraction by A549 cells, HBEC, or RalvEC. Cytomix (a mixture of tumor necrosis factor-alpha, interleukin-1beta, and interferon-gamma) inhibited the contraction by HFL1 but strongly enhanced the contraction by A549 cells. Cytomix also caused a morphologic change of A549 cells from a polygonal to a spindle shape. Immunocytochemistry showed that cytomix induced alpha-tubulin expression in A549 cells, whereas cytokeratin, vimentin, smooth muscle actin, beta(1) integrin, and paxillin expressions were not changed. This study thus demonstrates that alveolar epithelial cells can cause contraction of extracellular matrices and that this process is modulated by exogenous mediators, which also modify the microtubular system. Such an activity might contribute to alveolar remodeling after injury.

COPD is characterized by chronic inflammation and injury of both the airways and the parenchymal structures of the lung. These processes are associated with ongoing repair. Whether repair leads to restoration of normal tissue architecture or to altered tissue structure with loss of function depends on complex interrelationships of a variety of interacting mediators. The possibility that repair processes can be modulated by exogenous agents raises the possibility that therapeutic strategies aimed at repair can be effective. Such strategies offer tremendous promise both for slowing the relentlessly progressive natural history which most often characterizes COPD and, possibly, for restoring lung function. Rennard SI. Inflammation and repair processes in chronic obstructive pulmonary disease.

Human neutrophil elastase (HNE), an enzyme secreted by activated neutrophils, can bind to and degrade extracellular matrix including human lung elastin. This protease is believed to play an important role in several destructive processes including pulmonary emphysema. In this study, we hypothesized that an alveolar macrophage (AM) product or products may interact with neutrophil elastase (NE) and modulate its binding to elastin. Elastase binding to elastase was evaluated by a modified elastase functional assay using a synthetic substrate. Supernatants from cultured AM inhibited elastase binding to elastin at a dose-dependent manner without inhibiting functional elastase activity. The AM products had a heterogeneous molecular weight ranging from 440,000 to 54,000. The activity was heat-stable, but was lost after ultracentrifugation. After lipid fractionation, neither the aqueous nor the lipid fractions contained activity, suggesting that the factor may be a lipid complex. Culture supernatants from smokers' AM released significantly higher amounts of the factor than nonsmokers. In addition, high-molecular-weight elastase was present in bronchoalveolar lavage fluid (BALF) obtained from patients with pneumonia. Most of the in vivo high-molecular-weight elastase was lost after lipid extraction. In conclusion, macrophages release a factor or factors, probably lipid, which can interact with NE and inhibit its binding to human lung elastin without inhibiting elastase activity. This macrophage-derived factor may play a role in protecting the lung from NE by partitioning elastase into the airspace and thus protecting the interstitial connective tissue matrix from elastase degradation.

In the present study, we tested the hypothesis that neutrophil elastase (NE) might mediate remodeling of extracellular matrix by affecting fibroblast-mediated contraction of three-dimensional collagen gels. Human lung fibroblasts were cast into type I collagen gels containing NE. After gelation, the gels were released into medium and the area was measured by image analyzer. NE augmented gel contraction (p < 0.001). This was not due to cell proliferation or to degradation to soluble collagen fragments because the amounts of DNA and hydroxyproline were not altered. alpha1-Protease inhibitor and the synthetic inhibitor of NE, L-680,833, when added in sufficient amount to inhibit free elastase activity, blocked the contraction induced by NE. Furthermore, neutrophil granulocytes (PMN) in coculture, as well as conditioned media from PMN, resulted in an increased contractility (p < 0.001 for both). Bronchoalveolar lavage fluid (BALF) from patients with increased PMN in their lower respiratory tract and free elastase activity had augmentive activity for gel contraction which could be partially blocked by the inhibitors. We conclude that NE augments fibroblast-mediated contraction of collagen gels. The findings support the notion that products secreted by PMN in inflammatory disorders may lead to rearrangement of extracellular matrix and could subsequently lead to tissue dysfunction.

Migration of epithelial cells is an important feature of wound healing. Components of extracellular matrix stimulate migration of other cells, and we hypothesized that basement membrane components, laminin and type IV collagen, stimulate migration of bovine bronchial epithelial cells (BBEC). BBEC cultured for 3 days were used in migration assays using the blindwell chamber technique. BBEC migrated in a concentration-dependent manner to laminin, type IV collagen, and fibronectin. "Checkerboard" analysis demonstrated that the migration was directional for each of the components studied. We also evaluated the stimulatory activity of fibronectin fragments. A fragment containing the RGDS peptide was stimulatory of migration while other fragments were much less so, suggesting a role for RGDS-sensitive, integrin-mediated mechanisms. In order to evaluate whether bound or soluble matrix components were required to direct cell migration, we performed haptotaxis assays with precoated filters in the migration chambers. Fibronectin-precoated filters demonstrated significant stimulation of migration, suggesting that some of the migration in our original assays with fibronectin present in the lower chambers could be termed haptotactic. laminin- and type IV collagen-precoated filters were less active. When insulin was used as a chemoattractant, the fibronectin-precoated filters were more facilitatory of migration than the other matrix coatings. In summary, BBEC can chemotactically migrate to extracellular matrix components. Thus, the composition of the provisional matrix that forms at sites of epithelial injury may play an important role in the repair processes that occur after injury to the bronchial epithelium.

Regulation of airway repair after injury is poorly understood but is thought to be important in the development of airway diseases such as chronic bronchitis and asthma. There is evidence that fibronectin (Fn), an extracellular matrix glycoprotein, has a role in repair processes. In addition, transforming growth factor-beta (TGF-beta) is also likely involved in would healing and is known to influence extracellular matrix constituents in other cell systems. We postulated that TGF-beta may effect airway repair by modulating Fn production from airway epithelial cells. To examine this hypothesis, we studied the effect of TGF-beta 1 on Fn production by bovine bronchial epithelial cells in culture. Fn, released into the media of cultures exposed to TGF-beta 1, increased in a dose- and time-responsive fashion. Fn in the cell layer also increased in response to TGF-beta 1. De novo protein synthesis was demonstrated by an increase in [35S]methionine incorporation into Fn immunoprecipitated from media of TGF-beta-treated cultures. TGF-beta 1 also induced an increase in expression of Fn mRNA from cultured bronchial epithelial cells, suggesting that TGF-beta modulates Fn production of these cells, at least in part, through modulation of Fn gene expression. These data support a role for TGF-beta in airway repair through modulation of Fn production by airway epithelium.

Neutrophils and mononuclear cells have been associated with the lower respiratory tract inflammation observed in both acute and chronic bronchitis. In order to transit into and remain within the airways, neutrophils and mononuclear cells would likely need to adhere to bronchial epithelium. To test this hypothesis, bovine bronchial epithelial cells (BBECs) were isolated and cultured on a round coverslip. After 7 to 10 days, 51Cr-labeled neutrophils and mononuclear cells were evaluated for their capacity to adhere to the BBEC monolayer. Both neutrophils and mononuclear cells readily bound to the BBEC monolayer (10.8 +/- 1.2% bound neutrophils; 40.5 +/- 2.8% bound mononuclear cells). Stimulation of the neutrophils and mononuclear cells with phorbol 12-myristate 13-acetate (PMA) increased the adherence (45.8 +/- 10.6% bound neutrophils, P less than 0.01 compared with unstimulated cells; 58.7 +/- 6.2% bound mononuclear cells, P less than 0.01 compared with unstimulated cells). Importantly, stimulating the BBEC monolayer with PMA, bacterial lipopolysaccharide, or a cigarette smoke extract for 4 to 72 h also increased the adherence of both cell types (P less than 0.01, all comparisons at 24 h). The adherence was not decreased by exposure of either the BBEC monolayer, the neutrophils, or the mononuclear cells to cycloheximide or to the anti-CD11/CD18 monoclonal antibody 60.3 (P greater than 0.05). However, exposure of the BBEC monolayer to trypsin before addition of the neutrophils significantly decreased adherence (P less than 0.05). Because neutrophils and mononuclear cells are thought to mediate cell cytotoxicity by adhering to the target cells, BBECs were labeled with 51Cr, and 51Cr release was measured as an index of cytotoxicity. There was a modest increase in 51Cr release by the addition of unstimulated neutrophils and mononuclear cells, and culturing the BBEC monolayer with PMA before the addition of the neutrophils or mononuclear cells resulted in a further modest enhancement of 51Cr release (P less than 0.05). Similar results were obtained using lactate dehydrogenase release as a measure of cytotoxicity. These results demonstrate that inflammatory cells can adhere to BBECs and may be capable of mediating cytotoxicity and adherence and cytotoxicity can be increased by stimulating BBECs.

Transforming growth factor-beta 1 (TGF-beta 1) has been shown to induce squamous differentiation of cultured airway epithelial cells. It has also been shown to increase expression of matrix proteins and integrin receptors in cell culture of these and other cells. However, it is unknown if TGF-beta 1 affects expression of genes encoding intercellular junctional proteins. Therefore, we have investigated the effect of TGF-beta 1 on the expression of proteins and mRNAs for desmoplakins (DPs) I and II, desmosomal plaque proteins. Fibronectin, known to be induced by TGF-beta 1 was used as a positive control and tubulin as a negative control. Twenty-four hours after TGF-beta 1 stimulation, DP I and II mRNA levels assessed by Northern blotting analysis had increased significantly (DP I mRNA, 1.8-fold, P less than 0.05; DP II mRNA, 2.4-fold, P less than 0.04), thereby indicating pretranslational regulation of DP expression. By comparison, mRNA for fibronectin increased 8.1-fold whereas mRNA for tubulin was unchanged. Immunofluorescence using the monoclonal anti-DP I and II antibodies revealed dramatic increased expression of punctate DP structures after exposure to TGF-beta 1. Immunoblot analyses with polyclonal anti-DP I antibodies showed increased levels of both DP I (250 kD) and DP II (215 kD), with the DP I increase being more pronounced (DP I, 2.5-fold; DP II, 1.4-fold at 48 h relative to controls), suggesting translational regulation by TGF-beta 1. This study therefore demonstrates the ability of TGF-beta 1 to alter cellular phenotype by altering expression of proteins involved in intercellular junctions.(ABSTRACT TRUNCATED AT 250 WORDS)

Protease injury of the bronchial epithelium may play an important role in the pathogenesis of many airway diseases including asthma and chronic bronchitis. We hypothesized that neutrophil elastase can cause significant injury to the bronchial epithelium leading to detachment of bronchial epithelial cells. This detachment might be prevented if elastase is released into the airway lumen and the bronchial epithelium forms a barrier preventing access to the basal attachment sites. To assess this, the detachment of bronchial epithelial cells by elastase from extracellular matrix was measured. An increase in the resistance to detachment with time in culture was observed. The resistance to detachment was confirmed in bronchial epithelial cells, which were grown to electrically resistant monolayers on millipore filters and exposed to trypsin and elastase applied to both the apical and basal surfaces. Significantly greater detachment occurred when the proteases were applied at the basal surface versus the apical surface. Injury to the bronchial epithelium may enhance the proteolytic effect on the epithelium by disrupting epithelial barrier function. This was tested by exposing bronchial epithelial cell cultures to cigarette smoke extract before exposure to trypsin and elastase. The detachment of the bronchial epithelial cells exposed on the apical surface was increased greatly after smoke exposure. These data suggest that an intact bronchial epithelium can act as a barrier against proteolytic injury. Such a mechanism might protect the airway epithelium during intraluminal inflammation, and, if defective, might potentiate cigarette smoke-induced airway injury.

The proto-oncogenes are thought to play important roles in the regulation of cellular growth and differentiation. In order to evaluate the role of proto-oncogenes in the regulation of growth of bronchial epithelial cells, we studied steady-state levels of fos, jun, and myc transcripts in response to fetal calf serum, bovine pituitary extract, and insulin. Extensively quiescent populations of bovine bronchial epithelial cells in growth factor-free medium were stimulated to divide by each of these three additives. We observed rapid but transient increases of fos, jun, and myc expression in association with such growth stimulation. There were no changes in tubulin mRNA levels over the same time periods. Other "growth factors" (epidermal growth factor, hydrocortisone, epinephrine, triiodothyronine, and transferrin) were also studied and did not affect either cell growth or expression of fos, jun, or myc. We further examined the effect of transforming growth factor-beta1 (TGF-beta1) on the above stimulatory effects. TGF-beta1 consistently inhibited the growth induced by fetal calf serum, bovine pituitary extract, or insulin and, interestingly, reduced proto-oncogene myc mRNA level without altering that of fos and jun. In conclusion, proto-oncogenes fos, jun, and myc appear to play a role in the regulation of growth response in bovine bronchial epithelial cells. It is also possible that TGF-beta1 exerts its growth inhibitory effect, at least in part, through the processes that involve the regulation of proto-oncogene myc transcription in these cells.

Attachment of cells to extracellular matrix (ECM) plays an important role in the regulation of cell growth and differentiated function. We hypothesized that bronchial epithelial cells preferentially attach to ECM proteins and utilize specific receptors for ECM proteins. Bronchial epithelial cells were obtained from bovine lung by protease digestion. Both freshly isolated and cultured bronchial epithelial cells were plated onto plastic petri dishes coated with bovine serum albumin, type I collagen, type IV collagen, fibronectin, laminin, ECM synthesized by cultured bronchial epithelial cells, or uncoated. Freshly isolated cells demonstrated significant attachment to ECM but weak attachment to other matrix proteins. Cultured bronchial epithelial cells attached well to ECM; however, they had relatively increased attachment to type I collagen, type IV collagen, fibronectin, and laminin compared to freshly isolated cells. To determine whether the attachment of bronchial epithelial cells is arginine-glycine-aspartic acid (RGD)-mediated, an RGD-containing peptide known to block attachment mediated by many integrin receptors was added to the media (400 micrograms/ml). There was no inhibition of attachment of freshly isolated cells; however, there was significant but not complete inhibition of the attachment of the cultured cells to type IV collagen, laminin, and fibronectin, but not to type I collagen or ECM. Thus, freshly isolated bronchial epithelial cells readily adhere to ECM, and the attachment does not appear to be mediated by RGD-dependent receptors. Cultured bronchial epithelial cells demonstrate increased attachment to component proteins of ECM, and this attachment is, in part, to RGD-dependent receptors.

Migration of epithelial cells to cover areas of injury is thought to be important in the repair process following airway insult. Insulin is reported to be a growth factor for bronchial epithelial cells, and growth factors have been known to be chemotactic for many types of cells. Thus, we hypothesized that insulin may be a chemoattractant for bronchial epithelial cells. To evaluate this, we prepared bronchial epithelial cells and measured their chemotactic activity toward insulin. Bronchial epithelial cells were isolated by overnight digestion with bacterial protease, filtered through 100-microns nitex mesh, and then cultured at 1 x 10(6) cells/ml in tissue culture dishes in medium 199 supplemented with transferrin, insulin, epidermal growth factor, hydrocortisone, antibiotics, and 10% FCS for 3 d. The cultured cells were rinsed twice to remove supplements, trypsinized and resuspended at 1 x 10(6) cells/ml in medium 199 without supplements, and used as the cell source for chemotaxis. Chemotactic activity of bronchial epithelial cells was measured by the blindwell chamber technique using 8-microns Nuclepore filter membranes coated with 0.1% gelatin. The cells were added to the top wells in a 48-multiwell chamber with insulin in the bottom wells and incubated for 6 h at 37 degrees C, 5% CO2. Bronchial epithelial cells migrated in response to insulin in a dose-dependent manner up to an optimal dose of insulin, 100 micrograms/ml, and decreased at higher concentrations. The number of migrated cells per 10 high power fields was 33.7 +/- 1.9 at the optimum and 3.7 +/- 0.7 without insulin (P less than 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)

Macrophages are thought to participate in tissue repair following injury by releasing growth factors into the local environment. To evaluate whether pulmonary macrophages can mediate airway epithelial repair, we attempted to determine if pulmonary macrophages can stimulate growth of bovine bronchial epithelial cells in vitro. Bronchial epithelial cells isolated by protease digestion of the bovine bronchi were plated into tissue culture dishes with and without macrophage-conditioned medium. Bronchial epithelial cells cultured with macrophage-conditioned medium showed a significantly greater cell growth than those without macrophage-conditioned medium when assessed by direct enumeration of the cell numbers and by clonal growth assay. Stimulation of proliferation was confirmed by autoradiography using [3H]thymidine uptake into cell nuclei. Co-culture of pulmonary macrophages with bronchial epithelial cells also led to an increase in cell number. Immunohistochemical staining of the proliferating cells showed that these cells were positively stained by anti-keratin antibodies confirming that they were bronchial epithelial cells. Partial characterization of the activity in macrophage-conditioned medium showed that it was nondialyzable, pepsin- and acid-labile, and lipid-inextractable. Sephadex G-75 column fractionation indicated this activity existed in a high molecular fraction, thus suggesting a peptide. DEAE ion exchange chromatography revealed 3 peaks of stimulating activity. One peak resulted in a decrease in cell number, suggesting a possible inhibitory activity. The DEAE results thus suggest that macrophages may release several factors that can affect bronchial epithelial cell proliferation. In conclusion, pulmonary macrophages stimulate cell proliferation of bronchial epithelial cells in vitro. The stimulatory activity that may be heterogeneous appears to have the properties of a peptide.

The interaction between the epithelial cells and the subjacent mesenchymal cells in the airway is thought to play a major role during tissue repair after airway injury and lung morphogenesis. To evaluate this interaction, we cultured human lung fibroblasts, and bovine and human bronchial epithelial cells, and determined that bronchial epithelial cell-conditioned medium has a chemotactic activity for lung fibroblasts. This activity had the characteristics of protein: it was nondialyzable, heat-labile, pepsin-labile, acid-stable, and lipid-inextractable. Molecular sieve chromatography on Sephadex G-150 and affinity chromatography on gelatin-Sepharose revealed that there was one peak of chemotactic activity in high molecular weight range, which bound to gelatin, thus suggesting that the chemotactic factor might be fibronectin. Production and secretion of fibronectin into the culture media were demonstrated by biosynthetic incorporation of radioactive amino acid into fibronectin followed by immunoprecipitation on SDS-PAGE and autoradiography. Release into the culture medium was confirmed by ELISA. The identity of fibronectin as the chemotactic activity was confirmed by the addition of antifibronectin antibody to the conditioned medium, which inhibited chemotaxis in dose-dependent manner. Thus, bronchial epithelial cells produce fibronectin which can function as a chemotactic factor for lung fibroblasts. This production of fibronectin by bronchial epithelial cells may play an important role in regulating interaction between the bronchial epithelial cells that line the lumenal surface of the bronchial epithelial wall and the mesenchymal fibroblasts that underlie the bronchial epithelial basement membrane.