|Home | About | Journals | Submit | Contact Us | Français|
Rationale: Airway mucous cell metaplasia and chronic inflammation are pathophysiological features that influence morbidity and mortality associated with asthma and other chronic pulmonary disorders. Elucidation of the molecular mechanisms regulating mucous metaplasia and hypersecretion provides the scientific basis for diagnostic and therapeutic opportunities to improve the care of chronic pulmonary diseases.
Objectives: To determine the role of the airway epithelial–specific transcription factor NK2 homeobox 1 (NKX2-1, also known as thyroid transcription factor-1 [TTF-1]) in mucous cell metaplasia and lung inflammation.
Methods: Expression of NKX2-1 in airway epithelial cells from patients with asthma was analyzed. NKX2-1+/− gene targeted or transgenic mice expressing NKX2-1 in conducting airway epithelial cells were sensitized to the aeroallergen ovalbumin. In vitro studies were used to identify mechanisms by which NKX2-1 regulates mucous cell metaplasia and inflammation.
Measurements and Main Results: NKX2-1 was suppressed in airway epithelial cells from patients with asthma. Reduced expression of NKX2-1 in heterozygous NKX2-1+/− gene targeted mice increased mucous metaplasia in the small airways after pulmonary sensitization to ovalbumin. Conversely, mucous cell metaplasia induced by aeroallergen was inhibited by expression of NKX2-1 in the respiratory epithelium in vivo. Genome-wide mRNA analysis of lung tissue from ovalbumin-treated mice demonstrated that NKX2-1 inhibited mRNAs associated with mucous metaplasia and Th2-regulated inflammation, including Spdef, Ccl17, and Il13. In vitro, NKX2-1 inhibited SPDEF, a critical regulator of airway mucous cell metaplasia, and the Th2 chemokine CCL26.
Conclusions: The present data demonstrate a novel function for NKX2-1 in a gene network regulating mucous cell metaplasia and allergic inflammation in the respiratory epithelium.
Elucidation of the cellular and molecular mechanisms by which respiratory epithelial cells interact with the innate immune system to modify lung inflammation are providing insights into the pathogenesis of common pulmonary disorders. The present work identifies the role of the respiratory epithelial-specific transcription factors regulating mucous metaplasia and lung inflammation.
We provide in vivo and in vitro evidence for the airway epithelial-specific transcription factor NK2 homeobox 1 (NKX2-1)/thyroid transcription factor-1 (TTF-1) role in the inhibition of aeroallergen-induced mucous metaplasia and lung inflammation. NKX2-1 inhibited Sam Pointed Domain Ets-like Factor (SPDEF), a transcription factor critical for mucous cell differentiation and mucous production in the lung. Although NKX2-1 is known to play an important role in lung morphogenesis before birth, the present work demonstrates its novel function as a regulator of allergen-induced mucous metaplasia and inflammation in the adult lung.
In the conducting airways, the epithelium is pseudostratified and consists primarily of basal, ciliated, secretory, and relatively lesser numbers of mucous cells (1). These cells form a physical barrier to the external environment and protect the airways surface through secretion of mucus that neutralizes noxious agents and traps inhaled particles, which are moved out of the airways via the action of the mucociliary escalator (2). The gel-like properties of this mucus are dictated by the mucin glycoproteins MUC5AC and MUC5B, derived from goblet cells and submucosal glands (3). Repeated exposure of the airways to toxicants and allergens, as well as viral infections, induces mucous cell metaplasia and increases the production of mucus (4). Overproduction of mucus impairs mucociliary clearance, causing airway obstruction that underlies the recurrent infections and ongoing inflammation associated with pulmonary morbidities in chronic airway diseases, including asthma, chronic obstructive pulmonary disease, and cystic fibrosis (4–6).
Mucous hyperproduction is regulated by a number of signaling pathways, including EGF, Notch, and IL-4 receptors, that influence transcription of genes associated with mucous cell metaplasia in respiratory epithelial cells (4, 7). In mouse models, airway mucous metaplasia was dependent on the expression of Sam Pointed Domain Ets-like Factor (SPDEF) that was necessary and sufficient for mucous metaplasia in vivo (8, 9). Likewise, expression of SPDEF was induced by exposure of mice to aeroallergens or IL-13, where it was associated with mucous cell metaplasia and the loss of NKX2-1 (NK2 homeobox 1; also known as TTF-1, thyroid transcription factor-1) (9). NKX2-1 plays a central role in the regulation of lung morphogenesis before birth and is required for differentiation and gene regulation of diverse subsets of respiratory epithelial cells (10–12). In the normal human lung, NKX2-1 is selectively expressed in subsets of epithelial cells lining conducting airways and in alveolar type II cells in the lung periphery (13). After aeroallergen exposure in the mouse, epithelial cells (Clara cells) lining the bronchiolar epithelium undergo mucous metaplasia in association with increased expression of genes associated with the synthesis and packaging of pulmonary mucins and the loss of expression of genes expressed in nongoblet secretory cells.
Because SPDEF and NKX2-1 were not coexpressed in epithelial cells after pulmonary allergen sensitization in mouse (9) (see Figure E1 in the online supplement), we tested whether NKX2-1 played an important inhibitory role in the regulation of mucous cell metaplasia. NKX2-1 inhibited both mucous cell metaplasia and Th2-mediated inflammation, indicating its important role in the regulation of respiratory epithelial cell differentiation and innate immune responses in the adult lung.
Full methodological details are available in the online supplement.
Lung samples from anonymous patients were obtained at autopsy from S. H. Randell (University of North Carolina at Chapel Hill, NC) and A. Günther (University of Vienna, Austria, and University of Giessen Lung Center, Germany). Bronchial brushings and biopsies from anonymous patients with asthma and healthy control subjects were obtained by fiberoptic bronchoscopy. All of human specimens were obtained in accordance with institutional guidelines for use of human tissue for research purposes.
NKX2-1 heterozygous mice (+/−) were kindly provided by S. Kimura (National Institutes of Health, Bethesda, MD) (11). Transgenic mice bearing the transgene (tetO)7CMV-Flag-NKX2-1-IRES-EGFP were generated and mated with Scgb1a1-rtTA (line 2) transgenic mice (14).
Mice were sensitized to ovalbumin by sequential systemic and pulmonary administration using a protocol previously described (9). Expression of Flag-tagged NKX2-1 in the Scgb1a1-rtTA/(tetO)7CMV-Flag-NKX2-1-IRES-EGFP mice was induced at 6 to 8 weeks of age by provision of doxycycline 72 hours before mice were sensitized by intratracheal aspiration of ovalbumin. Single transgenic littermates that received the same doxycycline and ovalbumin sensitization were used as controls. Doxycycline was continued until the time of death.
Immunoblot assays were performed as previously described (15). Staining was performed using 5-μm paraffin-embedded lung sections as previously described (9). Morphometric analysis was performed on lung sections from control and NKX2-1+/− mice (n = 3–4 per group).
Total RNA was isolated using standard procedures. Quantitative reverse transcriptase–polymerase chain reaction (qRT-PCR) was performed on a 7300 real-time PCR system (Applied Biosystems, Foster City, CA) with the TaqMan probes. RNA microarray analysis was performed on whole lung RNA in the CCHMC Affymetrix Core using standard procedures. For functional classification and pathway analysis, gene ontology analysis was performed using the publically available web-based tool David (Database for Annotation, Visualization, and Integrated Discovery) (16).
Human lung epithelial carcinoma A549 cells were infected with either control virus or NKX2-1–expressing virus. Five days after infection, cells were harvested for qRT-PCR assay for SPDEF mRNA and for immunoblot assay for NKX2-1 and actin. In separate experiments, control virus or NKX2-1–expressing virus-infected A549 cells were treated with IL-13 (catalog # 213-IL-005; R&D Systems, Minneapolis, MN) at a final concentration of 10 ng/ml. Twenty-four hours after treatment, cells were harvested for qRT-PCR assay for CCL26 mRNA.
The siRNAs targeting NKX2-1 #1 (ID: 14152), NKX2-1 #2 (ID: 224731), and negative control siRNA (cat# 4390843) were purchased from Ambion/Applied Biosystems (Austin, TX). Transfection of human lung papillary adenocarcinoma NCI-H441 cells with siRNAs was performed according to a protocol using Lipofectamine RNAiMAX (catalog # 13778–150; Invitrogen, Carlsbad, CA). Forty-eight hours after transfection, total RNA was extracted and qRT-PCR assays for NKX2-1, SPDEF, CCL26, and MUC5AC were performed.
Statistical differences were determined using Student t test (two-tailed and unpaired). The difference between two groups was considered significant when the P value was less than 0.05 for all tests of human, mouse, and in vitro experiments.
Decreased expression of NKX2-1 was associated with Th2-mediated inflammation after aeroallergen exposure in the mouse (9) (Figure E1). We determined whether mucous cell metaplasia accompanying human asthma was associated with loss of NKX2-1. NKX2-1 was readily detected in the nuclei of subsets of epithelial cells in conducting airways and in alveolar type II epithelial cells in normal human lung tissue consistent with previous findings (13) (Figure 1A, top left). NKX2-1 staining was not detected in mucous cells in autopsy tissue from patients with asthma, wherein staining of NKX2-1 and SPDEF was mutually exclusive (Figure 1A). In bronchial brushings obtained by bronchoscopy from patients with asthma, NKX2-1 mRNA was significantly decreased compared with that in healthy control subjects (Figure 1B and Table 1).
To determine whether reduced NKX2-1 expression induced mucous cell metaplasia, adult NKX2-1 gene-deleted mice were studied. NKX2-1−/− die at birth related to respiratory failure associated with lack of respiratory epithelial differentiation and peripheral lung formation, whereas NKX2-1+/− mice survive normally to adulthood (11). NKX2-1 mRNA in lungs of the NKX2-1+/− mice was significantly reduced, consistent with haploinsufficiency at the locus (17) (Figure E2). Mucous cells, detected by Alcian blue, SPDEF, and MUC5AC staining, were sporadically observed in the airways of the NKX2-1+/− but not in lungs from NKX2-1+/+ mice at 5 to 6 weeks of age (Figure 2A). NKX2-1+/− mice and littermate control mice were sensitized by pulmonary administration of ovalbumin (Figure 2B). In lungs of NKX2-1+/− mice, MUC5AC and SPDEF expression extended into small conducting airways that normally lack mucous cells after aeroallergen exposure (Figure 2B, right panels, arrows). Morphometric analysis demonstrated that the extent of mucous cell metaplasia induced by ovalbumin sensitization was significantly increased in bronchioles of NKX2-1+/− mice compared with littermate control mice (Figure 2C).
To determine whether NKX2-1 negatively regulates mucous cell metaplasia, transgenic mice were produced in which NKX2-1 was expressed under control of the tet-operator (tetO)7CMV-Flag-NKX2-1-IRES-EGFP. These mice were bred to Scgb1a1-rtTA transgenic mice (line 2) that express the reverse tetracycline transactivator (rtTA) in nonciliated Clara cells, the predominant secretory cell lining the conducting airways of the mouse (Figure 3A) (14). Flag-NKX2-1 mRNA and immunostaining were induced when doxycycline was administered in the food (Figures 3B and 3C, top two panels). Scgb1a1-rtTA/(tetO)7CMV-Flag-NKX2-1-IRES-EGFP and single littermate control mice were sensitized to ovalbumin to test whether increased expression of NKX2-1 in airway epithelial cells influenced allergen-induced mucous metaplasia. In control mice, endogenous NKX2-1 was absent and SPDEF was induced in mucous cells by ovalbumin sensitization (Figure 3C, left panels, arrow; Figure E1). In contrast, when NKX2-1 was induced in respiratory epithelial cells after treatment with doxycycline, mucous cell metaplasia, indicated by Alcian blue, SPDEF, and MUC5AC staining, was markedly inhibited or absent (Figure 3C, right panels; Figure 3D).
Although NKX2-1 inhibited the expression of SPDEF and MUC5AC, it prevented the loss of proteins triggered by allergen sensitization, including FOXA2 and cytochrome P450 reductase in Clara cells (Figure 4, top two panels; Figure E3). In contrast, FOXJ1, a transcription factor mediating cilia formation, and SOX2, a transcription factor expressed ubiquitously in the conducting airways, were not altered by the expression of NKX2-1. Ki-67, a cell proliferation marker, was not changed in bronchiolar epithelial cells of the NKX2-1–expressing transgenic mice (Figure 4, bottom three panels). Staining for CCSP, a Clara cell marker, was reduced in the bronchioles of NKX2-1–expressing transgenic mice at baseline (data not shown); thus, loss of CCSP was not a useful marker for mucous metaplasia in these mice.
mRNA microarray (Figure 5A) and qRT-PCR (Figures 5B and 5C, Figure E4) were used to identify mRNAs influenced by NKX2-1 after pulmonary ovalbumin sensitization. NKX2-1 inhibited mRNAs encoded by genes associated with goblet cell differentiation and mucous production, including Spdef, Muc5ac, Muc5b, Clca3/1, Agr2, and Tff2, consistent with histological data indicating inhibitory effects of NKX2-1 on mucous metaplasia.
To determine whether NKX2-1 inhibits SPDEF gene expression in a cell autonomous manner, human lung epithelial A549 cells that express SPDEF but not NKX2-1 were infected with a lentiviral vector containing NKX2-1. NKX2-1 significantly inhibited the expression of SPDEF mRNA in A549 cells (Figure 6A). Conversely, in H441 human lung epithelial cells that express both NKX2-1 and SPDEF, siRNA-mediated inhibition of NKX2-1 significantly induced the expression of SPDEF and MUC5AC mRNAs (Figure 6C), whereas siRNA-mediated inhibition of SPDEF did not alter NKX2-1 expression (data not shown). These results indicate that loss of NKX2-1 in lung epithelium promotes the expression of SPDEF and mucous-related genes, including MUC5AC.
A number of chemokines, cytokines, and innate immune molecules regulated by ovalbumin sensitization were also counter-regulated by NKX2-1. Ccl17, Ccl22, Il13, and Il4, but not Ifng mRNAs, were inhibited by expression of NKX2-1 after aeroallergen exposure (Figure 5, Figure E4). CCL17 (also known as TARC) is a Th2-attracting chemokine that is highly induced in the bronchial epithelium of patients with asthma. CCL17 mediates allergic airway inflammation and mucous cell metaplasia by activating the Th2 lymphocytes expressing IL-13 and IL-4 (18, 19). In human lung epithelial A549 cells, IL-13–mediated induction of the Th2 chemokine CCL26 (also known as eotaxin-3) (20) was significantly inhibited by NKX2-1 (Figure 6B). Conversely, the siRNA-mediated inhibition of NKX2-1 in H441 human lung epithelial cells significantly induced CCL26 expression (Figure 6C). CCL26 expression is increased in human bronchial epithelial cells from individuals with asthma (21). In our mouse models, Ccl26 expression was not altered by NKX2-1; however, the mouse Ccl26 gene is a pseudogene (22).
Airway epithelial cells provide barrier function, mucociliary clearance, and secrete diverse antimicrobials, mucins, chemokines, and cytokines that recruit and instruct inflammatory cells that serve to limit infection. Chronic epithelial cell injury and inflammation cause tissue remodeling and increased susceptibility to infection that together ultimately impair lung function in chronic airway diseases, including asthma, cystic fibrosis, and chronic obstructive pulmonary disease. Although transcriptional networks integrating mucous cell metaplasia and chemokine production by airway epithelial cells are presently poorly understood, the importance of the respiratory epithelium in the instruction of various aspects of innate immune system of the lung is highly relevant to the pathogenesis of common chronic lung diseases. Here, we observed that a transcription factor NKX2-1 in airway epithelial cells was suppressed in patients with asthma. In mouse models and in human epithelial cells, NKX2-1 inhibited genes associated with mucous metaplasia and Th2 inflammation.
To further understand the role of NKX2-1 in airway epithelial cells, we used Ingenuity Pathway Analysis and identified biological relationships of NKX2-1 among gene nodes within a network of corelated genes, the “network” being defined as a group of biologically related genes, proteins, or other molecules. Biological relationships were derived from the combined analyses of the present mRNA microarray data and previously published mRNA arrays obtained from transgenic mice in which Spdef, Foxa2, or NKX2-1 expression was selectively induced (9), deleted (23), or inhibited (12); from published and present experimental data that included mRNA, protein analysis, and immunohistochemistry; and from peer-reviewed publications in the Ingenuity knowledge base (Figure 7A). The network analyses revealed important interrelated cellular processes influenced by NKX2-1: (1) suppression of mucous cell metaplasia via inhibition of SPDEF-regulated mucous genes (e.g., Muc5ac, Clca3/1, Agr2, and Tff2); (2) inhibition of mRNAs associated with Th2-mediated inflammation (e.g., Ccl17, Ccl22, Ccr8, Cd28, Mmp12, Chia, and Icos); and (3) a complementary role of NKX2-1 with FOXA2 in the inhibition of mucous cell metaplasia and Th2-mediated inflammation.
mRNAs decreased by NKX2-1 during allergen exposure partially overlap with those induced by expression of SPDEF in airway epithelial cells (9) as shown schematically in Figure 7A. The finding that NKX2-1 inhibited allergen-induced expression of SPDEF provides a mechanism by which a number of genes associated with mucin production were coordinately regulated (Figures 5 and and7).7). The concept that NKX2-1 and SPDEF counter-regulate each other and associated target genes is consistent with previous studies in embryonic lungs from transgenic mice bearing phosphorylation sites mutant NKX2-1 allele (NKX2-1PM) (12), in which Spdef mRNA was markedly induced in lung before birth (Figure E5). In the present study, SPDEF induced the expression of mucins, including Muc5ac, Muc16, and a number of genes mediating various aspects of mucin biosynthesis and packaging, (e.g., glycotransferases, Agr2, and Clca3/1), which were inhibited by NKX2-1. Thus, reciprocal interactions between SPDEF and NKX2-1 likely serve as a switch to control the differentiation of airway epithelial cells to mucous cells.
Th2 chemokines and cytokines induced by allergen sensitization were also inhibited by NKX2-1 in vivo and in vitro, indicating that NKX2-1 suppresses some aspects of Th2-mediated responses in airway epithelial cells in a cell-autonomous manner. Although a number of mRNAs associated with Th2-mediated inflammation were inhibited by NKX2-1, expression of a number of mRNAs associated with innate immune mediated inflammation, including Il6, Saa3, Saa4, Orm1, and Orm2, were induced (Figures 5A and and7A).7A). Thus, the inhibitory effects of NKX2-1 on mucous cell metaplasia are likely mediated, at least in part, by inhibition of Th2-associated chemokines/cytokines and increased expression of a subset of mRNAs associated with acute inflammation.
FOXA2 is a member of the forkhead family of transcription factors that is normally expressed in both conducting airway and alveolar respiratory epithelial cells. Conditional deletion of Foxa2 in the developing respiratory epithelium caused severe eosinophilic lung inflammation, mucous metaplasia, and increased expression of mRNAs associated with mucous cell metaplasia and Th2-mediated inflammation, including Spdef, Muc5ac, Il13, Ccl17, and Ccl22 (23). Expression of FOXA2 in airway epithelial cells blocked allergen-induced mucous cell metaplasia and expression of SPDEF in adult mice (23, 24). In the present study, expression of NKX2-1 prevented the loss of FOXA2 and inhibited various mediators of Th2-mediated inflammation and mucous cell metaplasia after aeroallergen sensitization (Figure 4), supporting the concept that NKX2-1 and FOXA2 play complementary roles in the inhibition of mucous metaplasia and lung inflammation (Figure 7A).
Glucocorticoids are often used to treat patients with asthma; thus, the decreased expression of NKX2-1 in patients with asthma might be mediated, in part, by their treatment. Our patient cohort, however, included three patients not treated with glucocorticoids, and the decreased expression of NKX2-1 was also evident in these patients (data not shown). The numbers are small, and to conclusively address this larger numbers of patients would be needed to assess the potential role of glucocorticoid treatment in the expression of NKX2-1. Nevertheless, in the mouse experiments, NKX2-1 mRNA was decreased by allergen challenge in the absence of exogenous treatments (Figure E1). Furthermore, NKX2-1 was not reduced by treatment of dexamethasone, a corticosteroid, in H441 human lung epithelial cells (25). These results indicate that the expression of NKX2-1 is influenced by aeroallergen sensitization.
NKX2-1 inhibited aeroallergen-induced mucous cell metaplasia, in part, by inhibiting SPDEF and by maintaining expression of FOXA2. mRNA microarray analysis indicated that NKX2-1 and SPDEF act in an opposing manner within a gene network influencing both respiratory epithelial differentiation and the Th2 immunoregulatory axis. NKX2-1 inhibited allergen-induced Ccl17, Il13, CCL26, and other genes that play important roles in the pathogenesis of asthma and mucous metaplasia (Figure 7B). The present study demonstrates a novel function of NKX2-1 as an inhibitor of aeroallergen-induced airway mucous cell metaplasia in the adult lung and provides further support for the important role of the respiratory epithelium in the regulation of pulmonary inflammation.
The authors thank Drs. David Sammut, Benjamin Green, Nivenka P. Jayasekera, Patrick Dennison, and Christopher L. Grainge for performing bronchoscopies, and Scott H. Randell, Andreas Günther, Dennis K. Watson, Johannes C. M. van der Loo, Punam Malik, Kenneth Campbell, Shioko Kimura, Liqian Zhang, I-Ching Wang, Susan E. Wert, Laurent Plantier, Sheila M. Bell, Michael L. Mucenski, and Iris M. Fink-Baldauf for materials and technical assistance.
Supported by grants from the National Institutes of Health R01HL095580 (J.A.W.) and the University of Cincinnati Postdoctoral Fellow Research Program (Y.M.), and the Medical Research Council, London, UK (H.M.H., P.H.H., D.E.D., S.T.H.). H.M.H. is an MRC Clinician Scientist.
Author's contributions: Y.M., G.C., and J.A.W. designed research; Y.M., G.C., H.M.H., L.D., A.R.K., and P.H.H. performed research; Y.M., G.C., Y.X., H.M.H., and J.A.W. analyzed data; and –Y.M., G.C., Y.X., H.M.H., A.R.K., P.H.H., D.E.D., S.T.H., and J.A.W. contributed to writing of the manuscript.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.201101-0106OC on May 11, 2011
Author Disclosure: Y.M. received an institutional postdoctoral grant. G.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Y.X. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. H.M.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. L.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. A.R.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. P.H.H. was on the Advisory Board for Functional Therapeutics and received lecture fees from GlaxoSmithKline. He received institutional grant support from Schering Plough and Shionogi. D.E.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. S.T.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. J.A.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.