Published ahead of print on February 10, 2005, doi:10.1165/rcmb.2004-0266OC
© 2005 American Thoracic Society DOI: 10.1165/rcmb.2004-0266OC Bronchial Epithelial Compression Regulates Epidermal Growth Factor Receptor Family Ligand Expression in an Autocrine MannerBrigham and Women's Hospital, Department of Pulmonary and Critical Care Medicine; and Harvard School of Public Health, Physiology Program, Boston, Massachusetts Correspondence and requests for reprints should be addressed to Eric K. Chu, Brigham and Women's Hospital, Department of Pulmonary and Critical Care Medicine, 75 Francis Street, Boston, MA, 02115. E-mail: echu{at}partners.org
The epidermal growth factor receptor (EGFR), an important signaling pathway in airway biology, is stimulated by compressive stress applied to human airway epithelial cells. Although the EGFR ligand, heparin-binding epidermal growth factorlike growth factor (HB-EGF), is known to be released as a result of this stimulation, whether compressive stress enhances expression of other EGFR ligands, and the duration of mechanical compression required to initiate this response, is not known. Human airway epithelial cells were exposed to compressive stress, and expression of four EGFR ligands was examined by quantitative PCR. Cells were exposed to: (1) continuous compressive stress over 8 h, (2) compression with and without EGFR inhibitor (AG1478), or (3) time-limited compression (3.75, 7.5, 15, 30, and 60 min). Compressive stress produced a sustained upregulation of the EGFR ligands HB-EGF, epiregulin, and amphiregulin, but not transforming growth factor- . Inhibition with AG1478 demonstrated that expression of HB-EGF, epiregulin, and amphiregulin is dependent on the signaling via the EGFR. Immunostaining for epiregulin protein demonstrated increased expression with compression and attenuation with EGFR inhibition. The response of all three EGFR ligands persisted long after the mechanical stimulus was removed. Taken together, these data suggest the possibility of a mechanically activated EGFR autocrine feedback loop involving selected EGFR ligands.
Key Words: epidermal growth factor receptor autocrine signaling, mechanical stress epiregulin heparin-binding epidermal growth factorlike growth factor
The epidermal growth factor receptor (EGFR) is implicated in multiple processes in airway biology. In asthma, its functions include contributing to goblet cell hyperplasia, epithelial repair, and airway wall remodeling (14). Increased expression of EGFR has been demonstrated in the bronchial mucosa of individuals with asthma (5), and in a separate study epithelial EGFR expression was found to correlate with the severity of asthma and subepithelial reticular membrane thickening (6). After mechanical compression of airway epithelial cells, a model for asthmatic bronchoconstriction, signals are transduced via the EGFR. Signaling of mechanical compression occurs, in part, through the interaction of the EGFR ligand heparin-binding epidermal growth factorlike growth factor (HB-EGF) with the ErbB1 epidermal growth factor receptor (7, 8). Interestingly, compressive stress both requires the interaction of HB-EGF with its receptor and increases HB-EGF mRNA transcript levels (8), suggesting a feedback mechanism to enhance downstream responses. Given our previous findings that HB-EGF interacts with EGFR in mechanically stimulated cells, we postulated that mechanically induced EGFR stimulation may similarly drive increased expression of other EGFR autocrine ligands (913). The data reported herein demonstrate that mechanical stimulation of airway epithelial cells in culture elicits increased expression of a broad but selective subset of EGFR ligand family genes, that both baseline and mechanically induced expression of these genes depends on the activity of the EGFR, and that enhanced expression of each gene persists after the mechanical stimulus is removed. This integrated response has the potential to amplify, perpetuate, and diversify autocrine and paracrine responses to compressive stress.
Cell Culture Normal human bronchial epithelial (HBEC) cells were obtained from Clonetics-Bio Whittaker (San Diego, CA) and cultured at an airliquid interface, as previously described (14). In brief, passage 2 cells were expanded on tissue culturetreated plastic, in an atmosphere of 5% CO2 in air at 37°C in bronchial epithelial growth medium (BEGM; Clonetics) supplemented with bovine serum albumin (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 (15) with a 1:1 mixture of BEGM and Dulbecco's modified Eagle's medium (Media Technologies, Herndon, VA) supplemented with BEGM SingleQuot (final concentration contains < 5 ng/ml of epidermal growth factor; Clonetics) applied both apically and basally until confluent and then basally after an airliquid interface was established. The cells were maintained until a uniform, differentiated cell population with prominent cilia and mucus-secreting capabilities was present. All cells were fed 18 h before each experiment.
Experimental Apparatus
Protocols
(1) . Continuous compressive stress. EGFR ligand mRNA expression was examined over an 8-h period after the onset of compressive stress. Cell lysates for RNA purification were collected at 30- to 60-min intervals up to 4 h, and again at 8 h during continuous application of compressive stress. Transcript levels for HB-EGF, epiregulin (EREG), amphiregulin (AREG), and transforming growth factor (TGF)- were examined by real-time polymerase chain reaction (PCR). EREG protein expression was also examined in uncompressed cells and in cells exposed to a continuous compressive stress. A compressive stress of 8 h duration was applied in the presence or absence of AG1478 (Calbiochem, San Diego, CA), an inhibitor of the EGFR tyrosine kinase. In cells treated with the inhibitor, AG1478 was added to the media 1 h before compression at a saturating concentration (300 nM). (2) . Inhibitor studies. The time of peak expression of the family of EGFR ligands (2 h) was determined from the continuous compressive stress experiment above. Gene expression responses to compressive stress were then examined in the presence or absence of AG1478 (Calbiochem), a specific inhibitor of the EGFR tyrosine kinase, at a saturating concentration (300 nM). Four groups were examined: (1) control (without compressive stress), (2) inhibitor control (AG1478 applied for 3 h), (3) compressive stress (2 h of compressive stress), and (4) compression and inhibitor (1 h pretreatment with AG1478, followed by 2 h of compressive stress in the presence of AG1478). Transcript levels for HB-EGF, EREG, and AREG was examined by real-time PCR.
(3) . Time-limited compression.
A. . HB-EGF expression over 8 h with time-limited compression.
B. . EGFR ligand expression at 2 hours with time-limited compression.
Reverse Transcription and Real-Time PCR
Immunofluorescence Membrane-bound epiregulin was immunolocalized using a specific monoclonal mouse IgG1 anti-human epiregulin antibody (MAB1425; R&D Systems, Minneapolis, MN) or a nonspecific human IgG1 control antibody (MAB002; R&D Systems). HBEC cells were fixed in 3% paraformaldehyde in PBS for 20 min, then washed with PBS. Fixed cells were then blocked in 10% goat serum for 45 min at room temperature, and then incubated with the primary antibody (15 µg/ml in PBS with 2% BSA) overnight at 4°C. The cells were then washed, followed by incubation (1.75 h) at room temperature with a goat anti-mouse IgG antibody (Alexa Fluor, 3.33 ng/ml; Molecular Probes, Eugene, OR) in 2% BSA and 5% goat serum. Samples were then washed, counterstained with propidium iodide for 5 min, and washed again. Nuclei were visualized using propidium iodide. Immunostaining was visualized using fluorescence microscopy (Eclipse TE2000-U; Nikon, Tokyo, Japan).
Statistics
Compressive Stress Increases Transcript Levels for a Panel of EGFR Ligands Continuous compressive stress applied over an 8-h period led to an increase in HB-EGF gene expression beginning at 30 min, peaking 5-fold above baseline at 2 h after the onset of compression (Figure 2A, P < 0.0001, ANOVA); these results replicated our previous findings (8). The upregulated gene expression was sustained beyond 4 h before returning to baseline at 8 h. EREG and AREG demonstrated similar sustained upregulated gene expression peaking at 4.2- and 2.3-fold at 2 h before returning to baseline at 8 h after the onset of compression (Figures 2B and 2C, P = 0.03 and P = 0.003, respectively, by ANOVA). In contrast, TGF- gene expression (Figure 2D) was not significantly altered by compressive stress.
Expression of EGFR Ligands Depends on Signaling through the EGFR To examine the role of the EGFR in mediating the effects of mechanical compression, EGFR ligand gene expression was examined at the time of peak fold change (2 h) in the presence of an EGFR inhibitor. Treatment of cells with AG1478 (a specific inhibitor of the epidermal growth factor receptor tyrosine kinase) altered both baseline and compression-induced expression of all three monitored genes (Figure 3). Baseline constitutive expression of HB-EGF, EREG, and AREG were all significantly diminished by AG1478, demonstrating that transcripts for each gene are present under control conditions, and constitutive transcript levels are under the influence of basal EGFR activity (P = 0.0014, P = 0.0019, P < 0.0001, respectively, by Fisher's PLSD). Similarly, incubation with AG1478 before and simultaneous with compressive stress significantly reduced gene expression of HB-EGF, EREG, and AREG (P < 0.0001, P < 0.0001, P < 0.0001, respectively by Fisher's PLSD).
Gene Expression Is Dependent on Duration of Compressive Stress Variable durations of compressive stress (3.75, 7.5, 15, 30, and 60 min) were applied to differentiated bronchial epithelial cells and expression of HB-EGF was evaluated by real-time PCR over an 8-h period after the initiation of compression. Decreasing the duration of compressive stress attenuated the peak increase in HB-EGF fold change (Figure 4). However, the temporal pattern of expression appeared similar for all durations, with the peak upregulation occurring between 1 and 2 h.
Therefore, we investigated in more detail the effect of various durations of compressive stress on EGFR ligand expression at a single time point after the onset of compressive stress (2 h). Variable durations of compressive stress (3.75, 7.5, 15, 30, and 60 min) were applied to differentiated bronchial epithelial cells and expression of HB-EGF, AREG, and EREG was determined at 120 min after the initiation of compression. There was a duration dependent increase in fold change of HB-EGF, EREG, and AREG mRNA in response to the duration of compression applied (Figure 5, P < 0.0001, P = 0.005, P < 0.0001, respectively, by ANOVA). The effect of compression on gene expression at 2 h reached significance by post hoc analysis at a duration of 7.5 min for AREG, at 15 min for EREG, and at 30 min for HB-EGF.
Epiregulin Protein Expression Is Induced by Compressive Stress Signaling via the EGFR Epiregulin protein expression was examined by immunofluorescence after 8 h of compressive stress, 8 h of time-matched control without compression, or 8 h of compressive stress after preincubation with an EGFR inhibitor (Figure 6). Immunofluorescence showed specific staining for epiregulin (indicated in green, left panels) compared with nonspecific antibody controls (right panels). Compression resulted in an increase in EREG-immunopositive cells compared with control. The compression induced expression of EREG was partially inhibited by EGFR blocking AG1478 resulting in less prominent EREG staining.
Our data demonstrate a coordinated and selective increase in transcripts for three members of the EGFR ligand family in response to compressive stress; in contrast, this stimulus did not modify TGF- gene expression. These findings highlight the importance of the EGFR in mechanotransduction and suggest a mechanism by which compressive forces signaling through a single receptor can lead to multiple coordinated events relevant to airway remodeling. A key observation is that the increase in the level of EGFR ligand transcripts is dependent on signaling via their own receptor. Furthermore, we showed that these mechanically driven transcriptional events are associated with a mechanically driven increase in EREG protein expression that is also dependent on EGFR signaling. In addition, the increased transcript levels of all three ligands persist long after the removal of mechanical stress, suggesting either a physical mechanism that has considerable hysteresis or a biochemical cascade that once initiated can proceed without further stimulus. We have previously demonstrated that signaling via the EGFR is a key element in mechanotransduction in airway epithelium (7). Cell culture experiments demonstrate that compressive stress leads to intracellular signaling via the mitogen-activated protein (MAP) kinase pathway in an EGFR-dependent manner, and isolated lung studies show bronchoconstriction-mediated EGFR activation (7, 8). Compressive stress is thought to result in collapse of the intercellular space separating the lateral cell membranes of adjacent cells, leading to increased concentration of the EGFR ligand HB-EGF in the intercellular space, and subsequent signaling via the EGFR. Our inhibitor studies (Figure 3) show that blocking EGFR signaling attenuates both basal and compression-enhanced expression of EGFR ligand genes. These results suggest a feedback mechanism by which EGFR ligand gene expression is under control of signaling through the EGFR, and perturbations of EGFR activity (enhancement by mechanical stress, attenuation by AG1478) bring about like-directed changes in ligand expression. This idea is further supported by our data reported herein that demonstrate increased EREG protein expression in response to compression and the attenuation of the EREG response in the presence of EGFR blockade. Autocrine-positive feedback activity of the EGFR has been shown to be critical in cancer and developmental models (1821). Shvartsman and coworkers have demonstrated through mathematical modeling and the use of radiation stimuli on a cancer cell line the ability of the EGFR system to participate in positive feedback (13). In mechanically compressed cells, such a mechanism could explain the sustained upregulation in gene expression of all three ligands and provide a means for these EGFR ligands to not only initiate the signaling cascade, but propagate the response. Indeed, all three upregulated ligands identified here have been shown to have autocrine properties in various experimental systems (12, 21, 22). The physiological implication of these results is that even brief episodes of mechanical stress can contribute to the phenotype of the airway.
Despite the ubiquitous nature of the EGFR ligands and the central role the EGFR plays in lung development, cancer, and airway diseases (14), the specific functions of the individual EGFR ligands remain largely undefined (6). Triple EGFR ligand (EGF, AREG, TGF- The current study was limited to the examination of the response of four EGFR ligands in HBEC to mechanical stress, but other types of airway stress are also known to activate the EGFR system and other cell types contribute to the phenotype of the airway. For instance, oxidative stress has been shown to stimulate the release of HB-EGF from HBEC (28) and human lung fibroblasts (29). Zhang and colleagues have shown that under oxidative stress, HBEC release HB-EGF signaling mitogenic activity in human lung fibroblasts (28). Similarly, we have previously demonstrated the ability of compression stimulated HBEC to upregulate HB-EGF protein expression (8) and in co-culture systems to elicit a matrix remodeling response in human lung fibroblasts (14). Thus our study has implications to other stimuli and on other cell types, and may even suggest a possible mechanism by which these two stimuli, i.e., oxidative stress and mechanical stress, may contribute to airway remodeling in asthma. Our data demonstrate that the genes for several EGFR ligands are constitutively expressed at baseline and that they contribute to the sustained signal transduction which follows even brief periods of compressive stress via the EGFR (Figure 7). Taken together, these data suggest the presence of an autocrine or juxtacrine feedback loop involving the EGFR that may contribute to the perpetuation of the effects of brief durations of compressive stress on human bronchial epithelial cells.
Supported by Fellowship Grant from the Canadian Institutes of Health Research (E.K.C.), an NSF-IGERT fellowship (A.S.P.), a Parker B. Francis Fellowship in Pulmonary Research (D.J.T.), an NIH HL-33009, and an American Lung Association Research Grant. Conflict of Interest Statement: E.K.C. has no declared conflicts of interest; J.S.F. has no declared conflicts of interest; J.C. has no declared conflicts of interest; A.S.P. has no declared conflicts of interest; J.M.D. has no declared conflicts of interest; and D.J.T. has no declared conflicts of interest. Received in original form August 18, 2004 Received in final form February 2, 2005
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||