After birth, the respiratory tract adapts to recurrent exposures to pathogens, allergens, and toxicants by inducing the complex innate and acquired immune systems required for pulmonary homeostasis. In this study, we show that Foxa2, expressed selectively in the respiratory epithelium, plays a critical role in regulating genetic programs influencing Th2 cell-mediated pulmonary inflammation. Deletion of the Foxa2 gene, encoding a winged helix/forkhead box transcription factor that is selectively expressed in respiratory epithelial cells, caused spontaneous pulmonary eosinophilic inflammation and goblet cell metaplasia. Loss of Foxa2 induced the recruitment and activation of myeloid dendritic cells and Th2 cells in the lung, causing increased production of Th2 cytokines and chemokines. Loss of Foxa2-induced expression of genes regulating Th2 cell-mediated inflammation and goblet cell differentiation, including IL-13, IL-4, eotaxins, thymus and activation-regulated chemokine, Il33, Ccl20, and SAM pointed domain-containing Ets transcription factor. Pulmonary inflammation and goblet cell differentiation were abrogated by treatment of neonatal Foxa2∆/∆ mice with mAb against IL-4Rα subunit. The respiratory epithelium plays a central role in the regulation of Th2-mediated inflammation and innate immunity in the developing lung in a process regulated by Foxa2.
After birth, the respiratory tract is recurrently exposed to pathogens, allergens, and toxicants. A multilayered innate and acquired host defense system develops to maintain pulmonary homeostasis in the face of these challenges. Th2-dominated inflammation and mucus hyperproduction are common features of acute and chronic pulmonary disorders, including asthma. Normally, inflammatory responses are precisely balanced to achieve mucociliary clearance and removal of pathogens. Various signaling and transcriptional programs influence allergic lung inflammation and mucus hyperproduction, including the IL-4R/STAT6 (1), TLR/NF-κB (2), and epidermal growth factor receptor/MAPK (3, 4) pathways. Allergen-induced inflammation is dominated by Th2 cell activation and production of IL-13, IL-4, IL-5, eotaxins, and other mediators that influence innate immune responses, Ab production, goblet cell metaplasia, airway remodeling, and hyperreactivity (5). IL-4 binds to two distinct receptor complexes, whereas IL-13 only binds to one of these complexes. Specifically, IL-4 binds to the IL-4Rα–chain, the functional receptor subunit of both the type I receptor, which is a heterodimer of IL-4Rα and the γc-chain, and the type II receptor, which is a heterodimer of IL-4Rα and IL-13Rα1. IL-13 does not bind to IL-4Rα directly but binds to IL-13Rα1 and can only activate the type II receptor. As the common subunit of receptor complexes for both IL-4 and IL-13, IL-4Rα is required for mediating signal transduction of these two cytokines (1, 6, 7). Targeting IL-4Rα function by antagonists or its expression by antisense oligonucleotides suppressed airway Th2 inflammation, hyperresponsiveness, and mucus production after allergen exposure in vivo (8, 9), indicating the pivotal role of IL-4Rα in mediating Th2 response. There is increasing evidence that respiratory epithelial cells lining conducting airways play important roles in modulation of inflammatory responses to allergens, pathogens, and injurious agents (10, 11). Innate immune responses induced by epithelial cells in response to pathogen are crucial for dendritic cells (DCs) to initiate and maintain allergic Th2 cell responses in experimental asthma (12, 13). Recent studies, from this laboratory and others, demonstrated that expression of Foxa2, a member of the forkhead box family of transcription factors expressed in the respiratory epithelium, is inhibited during allergen or IL-13–induced goblet cell metaplasia and Th2 inflammation (14, 15). Deletion of Foxa2 in respiratory epithelial cells in the mouse impaired surfactant production and lung maturation before birth and caused spontaneous inflammation and goblet cell metaplasia after birth, indicating a requirement for Foxa2 in normal postnatal pulmonary homeostasis (16). Goblet cell metaplasia induced by allergens or IL-13 was dependent on Stat6 in a process associated with the induction of SAM pointed domain-containing Ets transcription factor (Spdef) and Foxa3, indicating that these transcription factors interact in a network influencing airway epithelial differentiation (14, 17, 18). Recent studies demonstrated that goblet cells induced by pulmonary allergens are derived from the differentiation of resident Clara cells that serve as progenitor cells in the bronchiolar epithelium (18, 19). In initial studies, in which Foxa2 was deleted in the respiratory epithelium prior to birth, mucous cell metaplasia, alveolar remodeling, and inflammation were observed (14). The characteristics and mechanisms underlying lung inflammation caused by loss of Foxa2 in airway epithelial cells are unclear at present.
In this study, we demonstrate that respiratory epithelial cell-specific deletion of Foxa2 caused spontaneous Th2 cytokine/chemokine-mediated inflammation and goblet cell metaplasia. The hypothesis, that the inflammatory response was mediated by the spontaneous activation of IL-4R signaling, was tested using a mAb to block IL-4R signaling in neonatal mice. Deletion of Foxa2 induced expression of a network of genes influencing or associated with Th2-cell mediated inflammation and DC recruitment and activation, indicating the role of respiratory epithelial cell in programming inflammatory responses in the developing lung in a process regulated by Foxa2 and requiring IL-4Rα signaling.
Materials and Methods
Transgenic mice and animal husbandry
Animals were maintained according to protocols approved by the Institutional Animal Care and Use Committee at Cincinnati Children’s Hospital Research Foundation (Cincinnati, OH). Mice were housed in a pathogen-free barrier facility in humidity- and temperature-controlled rooms on a 12:12 h light/dark cycle and were allowed food and water ad libitum. SFTPC-rtTA/tetO7CMV-Cre/Foxa2LoxP/LoxP compound transgenic mice were generated as previously described (14) and used to permanently delete Foxa2 in the fetal lung. Pregnant dams were maintained on doxycycline food from embryonic day (E) 6.5 to E12.5 to delete Foxa2 from respiratory epithelial cells during fetal development, producing SFTPC/Foxa2∆/∆ mice.
Conditional expression of Foxa2 in respiratory epithelial cells was achieved by producing tetO7-Foxa2-IRES-EGFP transgenic mice that were then mated to Scgb1a1-rtta (line II) mice (20). Full-length rat Foxa2 coding sequence together with the 5′ untranslated region and 3′ untranslated region was isolated from pRc/CMV-rFoxa2 (14) at EcoRI sites and cloning into the BamHI site of the pOtet7-IRES-EGFP vector (21) (the latter provided by Dr. Kenneth Campbell, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH) via blunt-end ligation. Transgenes were identified by PCR using the primer set: 5′-AGC AAA GAC CCC AAC GAG AAG C-3′ and 5′-CAA ACA ACA GAT GGC TGG CAA C-3′.
Histology and immunohistochemistry
Immunohistochemistry was performed on 5-μm lung sections using rabbit anti-Foxa2 (1:2000–1:8000), guinea pig anti-Spdef (1:4000) generated in this laboratory, goat anti-Foxa3 (1:100; SC-5361, Santa Cruz Biotechnology, Santa Cruz, CA), mouse mAb against Muc5ac (1:500, ab3649, Abcam, Cambridge, MA) (18). Sections were processed with Ag retrieval using heating in citrate buffer. Anti-mouse IL-4Rα mAb (lot number 9382-25B, 11/12/03, AS/CH) was kindly provided by Amgen (Thousand Oaks, CA).
Neonatal SFTPC/Foxa2∆/∆ pups and control littermates were injected i.p. on postnatal day (PN) 2 and PN9 with either anti–IL-4Rα Ab (50 μg/g body weight) or equal volume of sterile saline. On PN15, mice were anesthetized and the lungs lavaged five times with 0.3 ml saline, and the cells were counted from bronchoalveolar lavage fluid (BALF). Cytospins of BALF cells (5 × 104) were stained with a Diff-Quik kit (catalog number 24606, Polysciences, Warrington, PA). Cells (400–500) were counted to determine numbers and percentage of macrophages, eosinophils, neutrophils, and lymphocytes. Lung tissue from a second group of mice was fixed with 4% paraformaldehyde and processed into paraffin blocks for immunohistochemistry.
Pulmonary OVA sensitization
Mice were sensitized to OVA by systemic and pulmonary administration using protocols previously described (18). Expression of Foxa2 in the Scgb1a1-rtTA/tetO7-Foxa2-IRES-EGFP (Scgb1a1/Foxa2-IE) mice was induced at 6 wk of age by provision of doxycycline 24 h before mice were intranasally sensitized with OVA (Supplemental Fig. 8). The control mice were the single transgenic littermates (Scgb1a1-rtTA) that received the same doxycycline and OVA treatments. Doxycycline was continued until the time of sacrifice.
mRNA microarray analysis
http://bioinf.wehi.edu.au/affylmGUI) from R/Bioconductor package (www.bioconductor.org). Differentially expressed genes were selected with the threshold of p value of <0.01, fold change ≥1.5, and a minimum of two present calls in three samples with relatively higher expression. Gene ontology (GO) analysis was performed using the web-based tool David (Database for annotation, visualization, and integrated discovery). Overrepresented pathways were identified by comparison of the overlap of differentially expressed genes identified postdeletion of FOXA2 and all genes in MOE430 mouse genome. Gene sets associated with known pathways and disease states were identified from the Kyoto Encyclopedia of Genes and Genomes (www.genome.ad.jp/kegg/), GenMAPP (www.genmapp.org/), and GEArrays (www.sobiosciences.com/microarrays.php/). A pathway was considered to be overrepresented when a probability p value was ≤0.01 and gene hits ≥5. Potential protein/protein or protein/DNA interactions were identified using Ingenuity Pathway Analysis (Ingenuity Systems, Redwood City, CA). Ingenuity Pathway Analysis software maps the differentially expressed genes identified from the microarray experiment onto the interactome according to Ingenuity Pathway Knowledge Base, a large curated database of published literature findings related to mammalian biology. Genetic networks preferentially enriched in the sets of mRNAs were generated based on their connectivity. Statistical scores were calculated to rank the resulting networks and pathways using Fisher’s right-tailed exact test. The score indicates the degree of relevance of a network to the input gene set, taking into account the number of network-eligible genes and the size of the network.
Cytokine and quantitative RT-PCR assays
To determine cytokine levels, BALF was collected as previously described (22n of 6–8 mice of each genotype (Foxa2∆/∆ Foxa2 (Mm00839704_mH), Il4 (Mm00445259_m1), Il5 (Mm99999063_m1), Il13 (Mm99999190_m1), Ccl11 (Mm00441239_g1), Ccl24 (Mm00444701_m1), Ccl17 (Mm01244826_g1), Il33 (Mm01195784_m1), Il6 (Mm99999064_m1), Ccl20 (Mm01268753_g1), and Ifn-γ (Mm00801778_m1) and normalized to endogenous 18s rRNA for control (probe part number 4352930E). qRT-PCR was performed with an n of 3 mice for each genotype at PN11 and PN15.
Lung cell suspensions were incubated with anti-CD16/32 (clone 2.4G2) for 30 min and then staining reactions were performed at 4°C. Myeloid DCs (mDCs; CD11c+, CD11b+, Gr1neg, and CD317neg) and plasmacytoid DCs (pDCs; CD11clow, CD11bneg, Gr1+, and CD317+
Plasmid and luciferase reporter assay
The control plasmid used in luciferase reporter assay expressing enhanced GFP (EGFP) was made by isolation of the IRES-EGFP fragment from pIRES2-EGFP (catalog number 6029-1, Clontech, Mountain View, CA) by BamHI and NotI sites and subcloned into the BamHI site of the p3XFLAG-Myc-CMV plasmid (catalog number E6401, Sigma-Aldrich, St. Louis, MO) via blunt-end ligation. The Foxa2 expression plasmid was made by cloning rat Foxa2Ccl17-pGL3 plasmid was cloned by amplification of the 2-kb promoter region from C57/B6 mouse genomic DNA using primers: 5′-GGA CCT GAA ATA GTC AGC ATCC-3′ and 5′-CTG AGG TGA AGG TCT TCA TGGG-3′ and cloned into the pGL3-basic plasmid (catalog number E1751, Promega, Madison, WI). The luciferase assay was performed by transfection of MLE-15 cells with the mCcl17-pGL3
Statistical differences in cell counts and concentrations of proteins assessed by ELISA were determined using Student t test (two-tailed and unpaired). The statistic method applied to analyze qRT-PCR was the Mood’s Median test. Difference between two groups was considered significant when the p value was <0.05 for all tests.
Conditional deletion of Foxa2 caused Th2 cell-mediated pulmonary inflammation
To delete Foxa2 expression in respiratory epithelium, dams of SFTPC-rtTA/tetO7-CMV-Cre/Foxa2LoxP/LoxP transgenic embryos were treated with doxycycline from E6.5 to E12.5. Approximately 30% of the mutant mice died in the postnatal period as previously reported (14, 16). Surviving mice develop spontaneous pulmonary eosinophilic inflammation, goblet cell metaplasia, and airspace enlargement in the first 2 wk of life (Fig. 1A). On PN15, pulmonary inflammation was seen histologically and supported by quantitation of inflammatory cells in BALF. Numbers of eosinophils, lymphocytes, and macrophages were increased in BALF from Foxa2∆/∆ mice (Fig. 1B). Deletion of Foxa2 was associated with increased expression of Th2 cytokines and chemokines, including IL-13, IL-4, IL-5, and Ccl17 (thymus and activation-regulated chemokine [Tarc]) as measured in BALF (Fig. 1C). CD4+ T cells that were recruited into the lung were further analyzed by flow cytometry, revealing a significant increase of the median fluorescent intensity of IL-13 and IL-17a and abundance of Th2 and Th17 cell populations in the lungs of Foxa2∆/∆ mice. Expression mRNA of IL-6, a cytokine required for Th17 cell differentiation (23, 24), was significantly induced. In contrast, Ifn-γ mRNA was not altered (Supplemental Fig. 1). Taken together, Foxa2∆/∆ mice develop spontaneous pulmonary inflammation that is typical of enhanced Th2 cytokine/chemokine activity during the postnatal period of development.
Inhibition of IL-4R–mediated signaling inhibited eosinophilic inflammation and goblet cell differentiation induced by Foxa2 deletion
A key pathway that regulates allergic inflammation and tissue remodeling in the airways involves the Th2 cytokines IL-4 and IL-13 and the activation of the IL-4Rα subunit (IL-4Rα) (25). IL-4 is primarily involved in promoting the differentiation and proliferation of Th2 cells and the synthesis of IgE, whereas IL-13 has a critical role in mediating airway hyperresponsiveness (AHR), goblet cell metaplasia, and mucus hypersecretion, the elements that are most closely linked to clinical manifestations of asthma. Blocking IL-4R signaling pathway with neutralizing Ab against IL-4Rα or mutagenesis of IL-4Rα was sufficient to inhibit responses to both IL-4 and IL-13 cytokines in vivo and in vitro (26–28). Specifically, loss of IL-4Rα in Clara cells inhibited mucus production after OVA intrapulmonary exposure in adult mice (29), and loss of IL-4Rα significantly decreased spontaneously induced mucous cells and eosinophilia in neonatal mice (PN10) (30). To test whether pulmonary inflammation and goblet cell differentiation caused by conditional deletion of Foxa2 in the airway epithelium was dependent upon IL-4R–mediated signaling, Foxa2∆/∆ mice were treated with IL-4Rα mAb at PN2 and PN9. Eosinophilic inflammation was significantly inhibited by anti–IL-4Rα Ab, as shown by the reduction of inflammatory cells and eosinophils in BALF (Fig. 2A). Consistent with inhibition of Th2 inflammation, goblet cell metaplasia and mucus hyperproduction in Foxa2∆/∆ mice were inhibited by the IL-4Rα Ab. Goblet cell markers, including Alcian blue, and Spdef, Foxa3, and Muc5ac staining were also inhibited by neutralizing Ab against IL-4Rα (Fig. 2B), demonstrating that the goblet cell metaplasia was dependent upon activated IL-4Rα signaling caused by deletion of Foxa2.
Loss of Foxa2 in respiratory epithelium enhances mDC recruitment and activation
Th2-cell mediated inflammation is dependent on activation and migration of pulmonary DCs to the respiratory epithelium (31). As professional APCs, DCs are uniquely capable of fully activating naive T cells. In the lung, two distinct phenotypes of DCs have been identified that have markedly different effects on T cell function. mDCs (defined as CD11c+, CD11b+, Gr1neg, and CD317neg cells) promote robust T cell activation and efficiently promote allergen-induced AHR (32). In contrast, pDCs (defined as CD11clow, CD11bneg, Gr1+, and CD317+ cells) promote the development of Foxp3+ regulatory T cells, preventing allergen-induced AHR (33). Thus, to test the hypothesis that excessive Th2 cell activation observed in mice lacking Foxa2 in the respiratory epithelium was preceded by dysregulation in pulmonary DC recruitment or activity, we assessed mDC and pDC recruitment and activation at PN8, PN11, and PN15 by flow cytometry. Comparing the frequencies of DC subsets in control and Foxa2∆/∆ mice revealed a significant increase in the frequency of both mDCs and pDCs found in the lung of Foxa2∆/∆ mice (Fig. 3A). Although both subsets of DCs were found with greater frequency in the lungs of Foxa2∆/∆ mice, there was a substantial increase in the mDC/pDC ratio in Foxa2∆/∆ mice, particularly at PN15, suggesting an environment better suited to T cell activation in Foxa2∆/∆ mice. Comparing the expression of DC activation-associated costimulatory molecule expression on mDCs from control and Foxa2∆/∆ mice demonstrated significantly elevated frequencies of mDCs expressing B7-DC (Fig. 3B), B7-H1 (Fig. 3C), and CD86 (Fig. 3D). mDCs expressing CD80 were also increased in Foxa2∆/∆ mice, although changes were not statistically significant. MHC class II was not different from control and Foxa2∆/∆ mice (Supplemental Fig. 2). Interestingly, although pDCs were also generally more activated in Foxa2∆/∆ mice than in control mice, by PN15, a time when there was robust activation of T cell cytokine production, the levels of pDC activation had decreased and were indistinguishable from controls (Fig. 3B–D). Collectively, these data demonstrate that concomitant with the increased T cell activation, the recruitment and activation of pulmonary mDCs was increased in Foxa2∆/∆ mice. Expression of Th2 cytokine mRNAs, including Il4 and Il13, was significantly increased in the lungs of Foxa2∆/∆ mice at PN11 and PN15 (Fig. 4B, Supplemental Fig. 3). Mechanisms underlying the recruitment and activation of DCs in the lung are not well established at present. The IL-7–like cytokine thymic stromal lymphopoietin, produced by epithelial cells, is known to influence DCs expressing OX40 ligand, causing differentiation of naive CD4+ T cells to Th2 cells that produce IL-4, IL-13, and TNF but not IL-10 (34, 35). Similarly, CCL20 is a chemokine produced in epithelial cells and capable of inducing DC migration via interaction with CCR6 expressed on the immature DCs (36, 37). Ccl20, but not Tslp mRNA expression, was induced in the lungs of Foxa2∆/∆ mice at PN15 (Supplemental Fig. 4), suggesting a potential mechanism by which DCs are recruited and activated in the lungs of Foxa2∆/∆ mice. Although Tslp mRNA was not increased in whole lung at PN11 or PN15 (data not shown), it remains possible that time dependent or focal changes in its expression would not be detected in the current study design.
Foxa2 regulates a genetic network associated with pulmonary Th2 inflammation and goblet cell differentiation
To investigate the role of Foxa2 and its downstream targets associated with the Th2 inflammation and goblet cell hyperplasia, RNAs were isolated from the lungs of Foxa2∆/∆ and control littermates at PN15. Lung cRNA was hybridized to the murine genome MOE430 chips and differential gene expression and GO analyzed. Deletion of Foxa2 significantly influenced the expression of 665 mRNAs. Among these, 516 out of 665 mRNAs were induced, and 148 were reduced in response to deletion of Foxa2 in the respiratory epithelium (Fig. 4A). The complete microarray dataset has been submitted to Gene Expression Omnibus under accession number GSE19204 (www.ncbi.nlm.nih.gov/geo/query/). GO analysis suggested that the induced genes are primarily enriched in “Immune/inflammatory response” (p value: 3.8E-13), indicating that Foxa2 plays a suppressive role in immune/inflammatory responses and mucus production in the postnatal lung; nearly 40% of induced genes encode cytokines, chemokines, and their receptors, and many are known to be associated with asthma or experimental allergen sensitization (Supplemental Table I). The overlap between the known asthma-related genes (genes causing, predisposing, or protecting from asthma) and those responding to the deletion of Foxa2 was highly statistically significant. Pathway analysis further indicated the “dendritic and Ag presenting,” “T cell and B cell activation,” “cytokine-cytokine receptor interaction,” “eicosanoid signaling,” and “JAK-STAT signaling pathway” were significantly induced. In contrast, “glutathione metabolism” (1.2E-04) was the most significantly overrepresented canonical pathway in mRNAs decreased postdeletion of Foxa2 (Supplemental Table II). Expression of mRNAs of major Th2 cytokines and chemokines in the whole lung was confirmed by qRT-PCR. Consistent with the severe eosinophilic inflammation and Th2 lymphocytes infiltration seen in Foxa2∆/∆ mice, Il4, Il5, and Il13 mRNAs in the whole lung were significantly increased, whereas expression of the Th1 cytokine Ifn-γ was not changed (Fig. 4B, Supplemental Fig. 1). Epithelial cells lining the respiratory and gastrointestinal tract play a pivotal role in initiation, regulation, and resolution of innate and adaptive immune response by expressing a wide range of immune response genes including costimulatory molecules, chemokines, cytokines, and PGs (38). Of particular interest, respiratory epithelium expresses IL-33, a cytokine that promotes Th2 cytokine production (39) and CCL17 (40, 41) that chemoattracts Th2 cells via interactions with CCR4 that is selectively expressed on the Th2 cells (42). Expression of Il33 and Ccl17 mRNAs was significantly induced in the Foxa2∆/∆ mice at PN15 (Fig. 4B). CCL17 (Tarc), a potent T cell chemoattractant induced in bronchiolar epithelial cells of asthmatics (40, 41), likely plays an important role in the pulmonary inflammation seen in the Foxa2∆/∆ mice. The effects of Foxa2 on Tarc gene expression were assessed in MLE-15 cells in vitro. Foxa2 inhibited the Tarc promoter by ∼40% (Supplemental Fig. 5). A schematic summarizing the effects of Foxa2 gene deletion on Th2-mediated processes in is provided by Fig. 6.
Conditional expression of Foxa2 in the respiratory epithelium inhibited allergen-induced goblet cell differentiation
A transgenic mouse in which Foxa2 was conditionally expressed in the respiratory epithelium was produced to test whether increased expression of Foxa2 was sufficient to inhibit allergen-induced inflammation or goblet cell differentiation (Supplemental Fig. 6). Double-transgenic mice Scgb1a1-rtTA/tetO7-Foxa2-IRES-EGFP (Scgb1a1/Foxa2-IE) were treated with doxycycline, and the induction of Foxa2 and EGFP expression was confirmed by immunohistochemistry and fluorescence microscopy (Fig. 5A, Supplemental Fig. 6). Goblet cells, although normally rare in the surface epithelium lining the conducting airways of adult mice, are usually present in young mice (30, 43). Expression of Foxa2 from PN4-PN18 by doxycycline treatment blocked the normal postnatal goblet cell differentiation (Supplemental Fig. 7). Conditional expression of Foxa2 in the respiratory epithelium in adult mice inhibited goblet cell differentiation (Fig. 5B, Supplemental Fig. 8). However, expression of Foxa2 in respiratory epithelial cells of the mature lung prior to OVA sensitization did not alter Th2 cytokine production or inflammation. Inflammatory cell counts (total or differential), as well as IL-4, IL-5, IL-13, IL-10, and IFN-γ concentrations, were similar in BALF from Foxa2-expressing and control mice after OVA exposure (Supplemental Fig. 9A, 9B). These finding are consistent with previous observations supporting the role of Foxa2 in regulating epithelial cell differentiation (14, 16). The present findings demonstrate that Foxa2 also plays an important role in modulating Th2 immunity in postnatal development, but does not directly control Th2-mediated inflammatory responses in the mature lung.
Deletion of Foxa2 in respiratory epithelial cells caused spontaneous pulmonary inflammation associated with goblet cell metaplasia and enhanced expression of Th2 cell-associated cytokines, chemokines, and their receptors. Recruitment and activation of mDCs and Th2 lymphocytes were increased in Foxa2∆/∆ mice lungs and associated with increased Th2 cytokines (IL-4, IL-5, and IL-13) and Ccl17 (Tarc) production and Foxa2 inhibiting Tarc gene promoter activity in vitro. The inflammatory effects of Foxa2 deletion were inhibited by blocking IL-4R signaling in the developing mouse lung. Taken together, the respiratory epithelium, via Foxa2, plays a critical role in programming the innate immune system during postnatal development of the lung and serves to inhibit the development of Th2-dominated innate immunity.
Th2 cytokine-dominated inflammation postdeletion of Foxa2 in the lung
Allergic airway inflammation is characterized by recruitment and activation of Th2 lymphocytes and eosinophils, goblet cell metaplasia, and mucus hyperproduction in a process regulated by a number of cytokines and chemokines. A number of the components of the IL-4R signaling pathway (e.g., IL-4, IL-13, and IL-4Rα) were increased in Foxa2∆/∆ mice. IL-13 is produced primarily in Th2 cells and regulates many asthma-related processes, including mucus hyperproduction, eosinophil recruitment and survival, and airway hyperreactivity (44). IL-13 blockade abrogated many of the features of asthma (45), demonstrating that IL-13 is an important mediator in Th2 responses and asthma pathogenesis. The present observation, that inhibition of IL-4Rα signaling substantially inhibited pulmonary eosinophilic inflammation and goblet cell metaplasia postdeletion of Foxa2, supports the important role of Foxa2 in inhibiting development of Th2-mediated innate immunity that is dependent upon IL-4R signaling. Consistent with the observed Th2-mediated inflammation in Foxa2∆/∆ mice, a number of factors involved in T cell and eosinophil recruitment, including colony-stimulating factor receptors Csf2ra and Csf2rb (2- and 4-fold), Il1b (2-fold), tumor necrosis signaling factor-associated genes Tnfrsf1b, -4, -9, -13, -14, and -18, were induced (2–10-fold). Likewise, mRNAs encoding lymphocyte chemoattractants, Ccl13 (3.4-fold), Ccl22 (4.6-fold), Ccr9 (4-fold), Ccr5 (4-fold), and Ccr2 (2.6-fold), were induced, the latter being required for pulmonary DC accumulation in asthma (46). mRNAs associated with eosinophilic inflammation were induced in lungs of Foxa2∆/∆ mice, including Ccl11, Ccl24, Epx, Itgb2, Kng1, Lgals3, Ltc4s, Alox5, Arg1, Il1rn, and Rag1. Among these, eosinophil-specific chemokines Ccl11 and Ccl24 were increased 7- and 12-fold, respectively. Epx mRNA (eosinophil peroxidase), a heme-containing glycoprotein that may contribute to the pathogenesis of epithelial damage and bronchial hyperreactivity in human asthma (47), was markedly increased in Foxa2Δ/Δ mice. Eosinophil major basic protein (Prg2, 11-fold) and eosinophil-associated RNase (Ear11, 90-fold) mRNAs were also dramatically increased.
Th2-associated inflammatory mediators, including acidic chitinase (Chia), chitinase 3-like 1 (Chi3l1), chitinase 3-like 3 (Chi3l3), and chitinase 3-like 4 (Chi3l4) mRNAs, were increased in Foxa2∆/∆ mice. Chitinases and chitinase-like proteins are believed to play a key role in the innate immunity to parasites and other infectious agents and may play an important role in the pathogenesis of allergy and/or asthma and regulation of eosinophilia and eotaxin induction (48–50). Chia is induced via a Th2-specific, IL-13–mediated pathway in lung epithelial cells and macrophages and is increased in lungs of asthmatics (48). Chia stimulates chemokine production by pulmonary epithelial cells (50).
Deletion of Foxa2 from lung epithelium disrupted Th1-Th2 balance, inducing a large number of Th2 cytokines and associated signaling pathways. Il6 mRNA, a cytokine regulating Th1, Th2, and Th17 cell differentiation (51, 52), was increased 4.6-fold in Foxa2∆/∆ mice. Recent studies demonstrated that Foxa2 bound to the Il6 promoter and suppressed Il6 expression in liver (53). Because multiple Th1-associated genes are altered in the Foxa2∆/∆ mice, it is presently unclear which play the dominant roles in the observed phenotype. Of genes expressed in the respiratory epithelium, both Il33 and the chemokine Ccl17 were significantly induced postdeletion of Foxa2. IL-33 and CCL17 both promote Th2 cytokine responses (38, 39) and are expressed in respiratory epithelial cells. CCL17 attracts Th2 lymphocytes to mucosal sites (54) and is increased in the bronchial epithelium of patients with asthma (55). The present finding, that Tarc promoter activity in MLE-15 cells was inhibited by Foxa2 in vitro, provides a potential mechanism by which Foxa2 influences Th2 cell recruitment and activation in the lung.
Role of Foxa2 in the regulation of pulmonary DCs
A network of DCs is closely associated with the respiratory epithelium (56). Both mDCs and pDCs are present in the lung, but mDC subsets generally dominate the airway mucosa (57). In the current study, both mDCs and pDCs were detected in the lungs of normal neonatal mice, although mDCs were typically more abundant than pDCs. We previously demonstrated that the numbers of mDCs present typically exceed pDCs by 10–25-fold in the adult mouse lung (58). Thus, in neonatal mice, the lung appears to be a more tolerogenic environment, capable of promoting the development of T regulatory rather than activated T effector cells. In further support of this concept, we observed that although the level of mDC activation increased from PN8 to PN15 in control mice, the activation status of pulmonary pDCs peaked at PN11 and decreased thereafter, suggesting that between PN11 and PN15, there is a shift toward the activation of a population of pulmonary DCs better suited to promote T cell activation. Thus, in the absence of overt immunization, the neonatal lung may represent a generally suppressive environment, favoring the development of tolerance rather than overt immune responses. Collectively, these data demonstrate that, concomitant with the increased T cell activation observed in Foxa2∆/∆ mice, there is also greater recruitment and activation of pulmonary mDCs in the absence of Foxa2. Fig. 6 summarizes the present findings regarding mechanisms by which Foxa2 influences lung inflammation in the developing mouse.
Goblet cell differentiation caused by deletion of Foxa2 in the developing lung seen in the current study was associated with Th2 cell activation and the induction of Spdef and Foxa3 that appear to function in a transcriptional network in the respiratory epithelium (18). Consistent with recently published studies (59), we demonstrated that forced expression of Foxa2 in the adult mouse inhibited goblet cell differentiation, blocking Spdef and Foxa3, but did not influence allergen-induced inflammation or eosinophilic infiltration. These findings support the role of Foxa2 in the instruction of innate immunity during development rather than a direct role for Foxa2 in suppressing inflammation during OVA sensitization in the mature mouse.
We thank Ann Maher for preparation of the manuscript, Amgen for providing the IL-4Rα mAb, and Dr. Fred Finkelman for scientific input.
Disclosures The authors have no financial conflicts of interest.
This work was supported by the American Lung Association (to H.W.), National Institutes of Health Grants HL095580 and HL090156 (to J.A.W. and Y.X.), and National Heart, Lung, and Blood Institute Grant AR47363, which supported the flow cytometry core at Cincinnati Children’s Hospital Medical Center.
The sequences presented in this article have been submitted to the National Center for Biotechnology Information Gene Expression Omnibus (www.ncbi.nlm.nih.gov/geo/query/) under accession number GSE19204.
The online version of this article contains supplemental material.
Abbreviations used in this paper:
- airway hyperresponsiveness
- arbitrary units
- bronchoalveolar lavage fluid
- acidic chitinase
- dendritic cell
- embryonic day
- enhanced GFP
- gene ontology
- myeloid dendritic cell
- plasmacytoid dendritic cell
- postnatal day
- quantitative RT-PCR
- SAM pointed domain-containing Ets transcription factor
- thymus and activation-regulated chemokine.
- Received January 29, 2010.
- Accepted March 25, 2010.
- Copyright © 2010 by The American Association of Immunologists, Inc.