Airway smooth muscle constriction leads to the development of compressive stress on bronchial epithelial cells. Normal human bronchial epithelial cells exposed to an apical-to-basal transcellular pressure difference equivalent to the computed stress in the airway during bronchoconstriction demonstrate enhanced phosphorylation of extracellular signal-regulated kinase (ERK). The response is pressure dependent and rapid, with phosphorylation increasing 14-fold in 30 min, and selective, since p38 and c-Jun NH2-terminal kinase phosphorylation remains unchanged after pressure application. Transcellular pressure also elicits a ninefold increase in expression of mRNA encoding heparin-binding epidermal growth factor-like growth factor (HB-EGF) after 1 h, followed by prominent immunostaining for pro-HB-EGF after 6 h. Inhibition of the ERK pathway with PD-98059 results in a dose-dependent reduction in pressure-induced HB-EGF gene expression. The magnitude of the HB-EGF response to transcellular pressure and tumor necrosis factor (TNF)-α (1 ng/ml) is similar, and the combined mechanical and inflammatory stimulus is more effective than either stimulus alone. These results demonstrate that compressive stress is a selective and potent activator of signal transduction and gene expression in bronchial epithelial cells.
- mechanical stress
in patients with asthma, airway wall thickening, subepithelial fibrosis, increased myocyte muscle mass, myofibroblast hyperplasia, and mucus metaplasia contribute to the process termed airway remodeling (reviewed in Ref. 9). It is widely believed that the inflammatory environment of the asthmatic airway is solely responsible for these changes (4). Bronchoconstriction, a characteristic of the asthmatic response, leads to airway wall buckling (10) and the development of compressive stresses in the airway wall (30). We have proposed that the mechanical stress that accompanies bronchoconstriction contributes to airway remodeling, in much the same way that mechanical stimuli lead to remodeling of musculoskeletal (12) and cardiovascular (23, 26) tissues. For example, transcellular pressure (compressive stress) applied to rat tracheal epithelial cells elicits increased expression of transforming growth factor-β, endothelin-1, and Egr-1 (24). Identical compressive stresses applied to human bronchial epithelial cells lead to increased collagen synthesis in cocultured fibroblasts (27).
We hypothesized that the mitogen-activated protein kinase (MAPK) family of signal transduction molecules was a key intermediate in the response of bronchial epithelial cells to compressive stress. Although MAPKs are known to be activated by a variety of mechanical stresses in a broad array of cell types (13, 15, 16, 18, 25), the response is frequently nonspecific (3, 8, 11, 14). Our results demonstrate that transcellular pressure elicits a unique and specific pattern of MAPK phosphorylation in human bronchial epithelial cells; there is rapid phosphorylation of extracellular signal-regulated kinase (ERK) but no change in the phosphorylation of the p38 or c-Jun NH2-terminal kinases (JNK, also known as stress-activated protein kinase). Furthermore, our data demonstrate that transcellular pressure increases heparin-binding epidermal growth factor-like growth factor (HB-EGF) expression and does so in an ERK-dependent manner. HB-EGF is a potent mitogen and chemotactic factor (22) with multiple potential roles in airway remodeling (7, 21,28). Finally, we show that transcellular pressure and tumor necrosis factor (TNF)-α induce similar increases in HB-EGF expression, both in an ERK-dependent fashion, and the combined stimulus is more potent than either alone. These results demonstrate that compressive stress is a discrete, selective, and potent stimulus for signaling and gene expression in the bronchial epithelium.
Normal human bronchial epithelial (NHBE) cells were obtained from Clonetics-BioWhittaker (San Diego, CA) and cultured at an air-liquid interface, as previously described (27). In brief,passage 2 cells were expanded on tissue culture-treated plastic (5% CO2, 37°C) in bronchial epithelial growth medium (BEGM; Clonetics) supplemented with BSA (1.5 μg/ml) and retinoic acid (50 nM). Passage 3 cells were then plated on uncoated nucleopore membranes (25 mm diameter, 0.4 μm pore size, Transwell Clear; Costar, Cambridge, MA) at 100,000 cells/well. The cells were fed as previously detailed (31) with a 1:1 mixture of BEGM and DMEM (Life Technologies, Frederick, MD) applied both apically and basally until confluent and then basally after an air-liquid interface was established. The cells were maintained until a uniform, differentiated cell population with prominent cilia, as confirmed by scanning electron microscopy (data not shown), and mucus-secreting capabilities was present. Approximately 16 h before each experiment, the culture medium was changed to a minimal medium containing a 1:1 mixture of bronchial epithelial basal medium (Clonetics) and DMEM supplemented with only insulin (5.7 μg/ml), transferrin (5 μg/ml), penicillin (100 U/ml), and streptomycin (100 μg/ml).
Cells from a single donor were used for all experiments except the comparison of transcellular pressure with an inflammatory stimulus (as detailed below), for which NHBE cells from a second donor were used.
To expose cells to transcellular pressure, silicon plugs with an access port for pressure application were press fit in the top of each transwell, creating a sealed pressure chamber over the apical surface of the NHBE cells (24). The basal surface and medium were left exposed to atmospheric pressure. Each plug was connected, in parallel, to a 5% CO2 (balance room air) pressure cylinder via a humidified chamber maintained at 37°C. All experiments were carried out by pressurizing the chamber first and then applying the transcellular pressure by opening a valve just upstream from the transwell plugs.
To ascertain whether compressive and osmotic stress elicit similar responses, cells were exposed to hyperosmotic conditions using minimal medium supplemented with NaCl. Concentrated solutions (10×) were pH and temperature equilibrated before the experiment and were applied (250 μl added to 2.25 ml) to the basal medium. Isosmotic medium was added to a subset of control wells in an identical manner. For comparison with an inflammatory stimulus, NHBE cells were exposed to TNF-α (1 ng/ml; R&D Systems, Minneapolis, MN) for 1 h.
MAPK phosphorylation was evaluated with commercially available kits that include antibodies for each MAPK protein and the dual-phosphorylated form of the protein (Cell Signaling Technology, Beverly, MA). Cells were lysed in 100 μl of SDS sample buffer consisting of 62.5 mM Tris · HCl, 2% wt/vol SDS, 10% glycerol, 50 mM dithiothreitol, and 0.1% wt/vol bromphenol blue. Cell lysates were boiled for 5 min, microcentrifuged for 5 min at 10,000g, and loaded in 10% SDS-Tris · HCl precast gels (Bio-Rad, Hercules, CA). A positive control for each phosphospecific protein was included on all gels. Gels were transferred to polyvinylidene difluoride membranes (Immobilon-P; Millipore, Bedford, MA), blocked with 5% milk, and incubated overnight (4°C) in the recommended primary antibody dilution buffer for each antibody (1:1,000). Bands were visualized with horseradish peroxidase-conjugated anti-rabbit IgG (1:2,000) and enhanced chemiluminescense. Blots were quantitated by standard densitometry and normalized using the nonphosphorylated protein levels.
Total RNA was purified from cell lysates with a commercially available kit (Rneasy; Qiagen, Valencia, CA). Equal amounts of RNA (10 μg/lane) were separated in 1.2% agarose gels and transferred to nylon membranes (S&S Nytran; Schleicher & Schuell, Keene, NH). Blots were hybridized with a 640-bp probe for HB-EGF and a 1,100-bp probe for the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Clontech, Palo Alto, CA). Probes were radiolabeled using a random primed DNA labeling kit (Roche, Mannheim, Germany), and blots were hybridized for 1 h at 68°C using ExpressHyb solution (Clontech) and then exposed to film (X-omat AR; Kodak, Rochester, NY). Blots were quantitated by standard densitometry and normalized with the housekeeping gene GAPDH.
Immunostaining for membrane-bound pro-HB-EGF was carried out with a polyclonal antibody raised in chicken that recognizes the cytoplasmic domain of the human HB-EGF precursor (generously provided by Rosalyn Adam, Children's Hospital, Boston, MA). NHBE cells were fixed in 3% paraformaldehyde and ice-cold methanol for 5 min each, blocked in 10% horse serum, and incubated with the primary antibody (1:100 dilution) overnight at 4°C. The cells were then incubated for 2 h at 4°C in donkey anti-chicken IgG (1:500) and for 30 min at room temperature in a 1:1,000 dilution of peroxidase-conjugated streptavidin (both from Jackson Immunoresearch Laboratories, West Grove, PA). Immunostaining was visualized with diaminobenzidine (Sigma, St. Louis, MO) counterstained with methyl green.
Data are presented as means ± SE and were compared where noted by unpaired Student's t-test. P values <0.05 were considered significant. ANOVA was used when multiple comparisons were undertaken.
Application of transcellular pressure (30 cmH2O) to differentiated bronchial epithelial cells increased ERK phosphorylation, with no measurable change in JNK or p38 phosphorylation (Fig. 1 A). The increase in ERK phosphorylation was noted as early as 5 min after the onset of transcellular pressure (Fig.2 A) and continued to increase over 30 min, reaching a peak 15-fold increase in ERK phosphorylation (15.4 ± 5.5, P < 0.05, p42 and p44 combined, Fig. 1 B). The increase in ERK phosphorylation at 30 min was pressure dependent, with significant increases in phosphorylation at pressures >20 cmH2O (P < 0.05, Fig.1 C). Although the magnitude of ERK phosphorylation was attenuated at 1 h, levels increased over the remainder of the experiment (Fig. 1 B).
To confirm that the p38 and JNK pathways were responsive in our cell preparation, we exposed cells for 30 min to a hypertonic stimulus (230 mosmol/l), a known activator of MAPK signaling (8). This stimulus resulted in increased phosphorylation of all three MAPK pathways (Fig. 2 B). Comparison of the positive controls between the transcellular pressure and osmotic stress experiments demonstrated that sufficient exposures were taken for chemiluminescent detection of increased phosphorylation in both sets of experiments (Fig. 2). Furthermore, when longer exposures were carried out for the p38 and JNK blots (Fig. 1 A), no increase in phosphorylation was observed for either pathway at any time under transcellular pressure (data not shown).
Transcellular pressure will drive fluid flow down the pressure gradient. As the pressure above the cells is increased, and the pressure below the cells remains constant, water will be driven across the basolateral cell membrane of apical cells, potentially resulting in cell volume loss. Because NHBE cells responded to a hyperosmotic shock (Fig. 2 B), which reduces cell volume by drawing water out of cells across the semipermeable cell membrane, we reasoned that transcellular pressure and osmotic stress may stimulate cells via similar mechanisms. To directly evaluate this hypothesis, we compared cells exposed to 30 cmH2O transcellular pressure with cells exposed to an osmotic stimulus equivalent to this pressure, derived from the relationship where ΔP is the change in pressure, Π is osmotic pressure,R is the universal gas constant, T is the temperature (310°K), and ΔC is the change in concentration of the bathing medium. From this relationship the concentration change needed to elicit a change in osmotic pressure equivalent to 30 cmH2O is 1.15 mosmol/l. To eliminate any artifact associated with the application of osmotic stress, we applied an isotonic control solution in an identical manner. Both the isotonic control and the osmotic equivalent of the transcellular pressure failed to elicit an increase in ERK phosphorylation (Fig. 3). An osmotic stimulus 200 times greater than the osmotic equivalent of 30 cmH2O elicited a 5-fold increase in ERK phosphorylation (5.4 ± 1.2, P < 0.05 vs. control), which was significantly less than the 14-fold induction after transcellular pressure (14.9 ± 2.9, P < 0.05, pressure vs. 200× osmotic).
Application of transcellular pressure (30 cmH2O) increased expression of the HB-EGF gene, as measured by steady-state mRNA levels. The response peaked at approximately ninefold induction 1 h after pressure application (9.1 ± 0.4, P < 0.001; Fig.4). The abundance of message was reduced at subsequent time points, falling to approximately a threefold induction after 4 h. However, HB-EGF expression remained significantly elevated at the conclusion of the 8 h (3.7 ± 0.6, P < 0.05) of continuous transcellular pressure and continued to be expressed at elevated levels 16 h after the cessation of pressure (2.7 ± 0.7-fold, P < 0.05, data not shown).
The enhanced HB-EGF message was followed by increased expression of pro-HB-EGF, the transmembrane precursor of mature HB-EGF, after 6 h of transcellular pressure (30 cmH2O; Fig.5). The increased immunostaining was localized to the apical surface of occasional cells and to the lateral cell boundaries of the confluent cells.
To determine whether ERK phosphorylation is a necessary component of the pressure-induced HB-EGF gene expression, we used PD-98059, a specific inhibitor of the ERK pathway that prevents phosphorylation and activation of mitogen/ERK1/2, the enzymes responsible for activation of ERK1/2 (6). As shown in Fig.6 A, PD-98059 inhibited the ERK phosphorylation response to a 30-min transcellular pressure stimulus in a dose-dependent manner, reducing the pressure-induced phosphorylation by 55% at a concentration of 10 μM and by 97% at 50 μM (P < 0.05). PD-98059 treatment likewise inhibited the upregulation of HB-EGF in response to transcellular pressure at 1 h in a dose-dependent fashion (Fig. 6 B), reducing the pressure-induced increase in HB-EGF mRNA by 78% at a concentration of 10 μM (P < 0.05) and by 100% at 50 μM (P < 0.05).
In cells from a second donor (Fig. 7), transcellular pressure induced a more modest, but still significant, increase in HB-EGF mRNA (2.5 ± 0.3, P < 0.01 vs. control). The magnitude of the response was statistically indistinguishable from the increase in HB-EGF induced by exposure to 1 ng/ml TNF-α (3.1 ± 0.2, P < 0.01 vs. control). Combined stimulation with TNF-α and transcellular pressure was more potent than either stimulus alone (4.1 ± 0.3-fold induction,P < 0.01 vs. control, P < 0.05 vs. TNF-α, and P < 0.01 vs. pressure). To determine if ERK phosphorylation represents a common pathway required for increased HB-EGF expression, we also applied mechanical and inflammatory stimuli to cells from this second donor in the presence or absence of PD-98059 (50 μM). Incubation with the ERK inhibitor eliminated both pressure and TNF-α-induced HB-EGF expression (Fig.7).
When the airway buckles under the imposed loading of smooth muscle constriction, as may occur in asthma, compressive stresses on the order of 30 cmH2O (30) are achieved. Our data show that creation of an equivalent compressive stress in vitro by application of a pressure difference across a layer of human bronchial epithelial cells leads to a selective and rapid increase in ERK phosphorylation and HB-EGF gene expression.
MAPKs are key points of convergence for multiple signaling pathways and regulate important downstream processes such as growth, differentiation, and apoptosis (6, 20). Although MAPKs are known to be responsive to many agonists, including growth factors, cytokines, hormones, oxidants, and environmental stress factors, they are also widely reported to be responsive to mechanical stress (3, 13, 16, 18, 25). Previous work indicates that all three MAPKs can be activated in response to physical stimuli such as shear stress (3, 11, 15), cyclic stretch (14), and hyperosmotic stress (8). Our results demonstrate a unique and specific pattern of MAPK phosphorylation in response to transcellular pressure, with only ERK phosphorylated. We interpret this finding to indicate that transcellular pressure initiates a series of biochemical events distinct from those resulting from previously investigated mechanical stresses.
The selective activation of ERK is a useful tool to probe the possible mechanisms by which bronchial epithelial cells are activated by transcellular pressure. Ressler et al. (24) have shown that substrate strain and hydrostatic pressure were not responsible for increased gene expression after application of transcellular pressure and speculated that cell shrinkage and volume regulation initiated by the pressure gradient across the cells were responsible. Two findings from our current work argue against this hypothesis. First, as in other cell types (8), application of hyperosmotic stress to induce human bronchial epithelial cell shrinkage activated all three MAPK pathways (Fig. 2 B), not a single one as in the transcellular pressure experiments (Fig. 2 A). Second, application of an osmotic stress equivalent to the transcellular pressure of 30 cmH2O did not stimulate ERK phosphorylation, and osmotic stress 200 times greater stimulated ERK less than half as much as transcellular pressure (Fig. 3). These findings conclusively demonstrate that transcellular pressure activates MAPK signaling in a manner distinct from that induced by cell shrinkage and volume regulation and indicate that NHBE cells exhibit a sensitive and specific response to transcellular pressure. Although we have ruled out volume regulation, and Ressler et al. (24) ruled out hydrostatic pressure and substrate strain as the mechanism by which bronchial epithelial cells sense transcellular pressure, a number of potential mechanisms remain, including shear stress in the intercellular spaces and deformation of the basal cell membranes into the underlying pores, with subsequent activation of signaling in the stretched cell membrane.
The biphasic response of ERK phosphorylation to transcellular pressure was an unexpected finding. In response to most mechanical and osmotic stimuli, MAPK activation is rapid and transient (8, 16,25). Although ERK phosphorylation was greatly attenuated at 1 h relative to 30 min, it did not return to baseline levels during the 8-h time course but increased steadily from 1 to 8 h (Fig. 1). It is possible that sustained pressure provides a long-term stimulus for ERK phosphorylation. Alternatively, increases in ERK phosphorylation after return to near baseline levels may represent a secondary response to pressure-induced protein synthesis, secretion, and autocrine or paracrine signaling. For instance, stimulation of HB-EGF synthesis and secretion in fibroblasts causes a delayed activation of JNK that can be inhibited with neutralizing antiserum to HB-EGF (17).
The finding that ERK phosphorylation increased at 30 min and HB-EGF mRNA increased at 1 h is consistent with a role for ERK activation in the transcellular pressure-induced increase in HB-EGF expression. Our data on HB-EGF expression in cells exposed to transcellular pressure in the presence of PD-98059, which specifically binds to and inactivates the enzymes responsible for ERK phosphorylation (6), demonstrate that ERK activation is a necessary component in this response (Fig. 6). These results are consistent with findings in NIH/3T3 cells transformed by oncogenic Ras and Raf, where HB-EGF expression was dependent on ERK phosphorylation (17). We also demonstrate here that TNF-α-stimulated HB-EGF expression is ERK dependent (Fig. 7). These findings, in addition to the recent demonstration that vanadium-stimulated expression of HB-EGF in bronchial epithelial cells is ERK dependent (33), implicate ERK as a common pathway linking environmental, mechanical, and inflammatory stimuli to HB-EGF expression in the bronchial epithelium.
HB-EGF is likely to be a particularly important contributor to airway remodeling in asthmatic airways. HB-EGF in its soluble, mature form exhibits autocrine and paracrine signaling capabilities and is a potent mitogen (ED50 of 0.01–10 ng/ml) and chemotactic factor (ED50 of 0.1–10 ng/ml) for fibroblasts, smooth muscle cells, and epithelial cells (reviewed in Ref. 22). A recent report has demonstrated that HB-EGF is a dominant mitogen for fibroblasts in bronchial epithelial conditioned medium (33). HB-EGF is a ligand for the epidermal growth factor receptor (EGF-R), the activation of which is implicated in goblet cell hyperplasia (5), mucin production (28), and the epithelial response to toxic insult, inflammation, and wound healing (7). The EGF-R is present in the bronchial epithelium and submucosal glands (1, 2, 19) and is increased in asthmatic airways, exhibiting a positive correlation with subepithelial reticular membrane thickening (21) and mucin staining (29). HB-EGF is therefore likely to participate in several of the changes that characterize airway remodeling, including myofibroblast recruitment and hyperplasia, increased myocyte muscle mass, subepithelial fibrosis, and mucus metaplasia.
Transcellular pressure elicited an increase in HB-EGF mRNA statistically indistinguishable from that elicited by 1 ng/ml of TNF-α, a dose that provided a near-maximal upregulation of HB-EGF in vascular endothelial cells (3 ng/ml maximum; see Ref. 32). The simultaneous application of mechanical stress and TNF-α was more potent than either alone (Fig. 7). These results suggest that mechanically induced signaling of the bronchial epithelium may be of sufficient magnitude to be relevant in the context of the airway wall, where cells are exposed to a host of environmental and inflammatory insults, as well as a potentially dynamic mechanical environment. Furthermore, our results demonstrate that mechanical and inflammatory stimuli interact in vitro and raise the possibility that these stimuli combine to regulate gene expression in the asthmatic airway.
The airway epithelium is ideally situated to sense the mechanical stress that accompanies bronchoconstriction and to integrate mechanical stimuli with the various airborne and soluble factors that influence airway biology. Our in vitro approach is unique in that we show that differentiated bronchial epithelial cells, expressing many of the characteristics of the in vivo airway epithelium, respond to the application of a compressive stress that is matched to stresses calculated for the intact airway under smooth muscle constriction (30). The potent regulation by transcellular pressure of a discrete subset of the MAPK family of signal transduction pathways, linked to the downstream expression of HB-EGF, is a strong indication of the potential for mechanical forces to contribute to signaling and remodeling in the asthmatic airway.
We thank Jeffrey Fredberg, Roger Kamm, and Anna McVittie for helpful discussions of this work.
This work was supported by National Heart, Lung, and Blood Institute Grants HL-6407502, HL-07118, and HL-33009.
Address for reprint requests and other correspondence: J. M. Drazen, Pulmonary and Critical Care Division, Brigham and Women's Hospital, 75 Francis St., Boston, MA 02115 (E-mail:).
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.
- Copyright © 2002 the American Physiological Society