Epithelial cell alpha3beta1 integrin links beta-catenin and Smad signaling to promote myofibroblast formation and pulmonary fibrosis

Abstract

Pulmonary fibrosis, in particular idiopathic pulmonary fibrosis (IPF), results from aberrant wound healing and scarification. One population of fibroblasts involved in the fibrotic process is thought to originate from lung epithelial cells via epithelial-mesenchymal transition (EMT). Indeed, alveolar epithelial cells (AECs) undergo EMT in vivo during experimental fibrosis and ex vivo in response to TGF-beta1. As the ECM critically regulates AEC responses to TGF-beta1, we explored the role of the prominent epithelial integrin alpha3beta1 in experimental fibrosis by generating mice with lung epithelial cell-specific loss of alpha3 integrin expression. These mice had a normal acute response to bleomycin injury, but they exhibited markedly decreased accumulation of lung myofibroblasts and type I collagen and did not progress to fibrosis. Signaling through beta-catenin has been implicated in EMT; we found that in primary AECs, alpha3 integrin was required for beta-catenin phosphorylation at tyrosine residue 654 (Y654), formation of the pY654-beta-catenin/pSmad2 complex, and initiation of EMT, both in vitro and in vivo during the fibrotic phase following bleomycin injury. Finally, analysis of lung tissue from IPF patients revealed the presence of pY654-beta-catenin/pSmad2 complexes and showed accumulation of pY654-beta-catenin in myofibroblasts. These findings demonstrate epithelial integrin-dependent profibrotic crosstalk between beta-catenin and Smad signaling and support the hypothesis that EMT is an important contributor to pathologic fibrosis.

Figure 1. Lung epithelial cell–specific loss of α3 integrin in FASC mice.

(A) In triple transgenic FASC mice, rtTA is expressed in lung epithelial cells using the human SPC promoter. In the presence of doxycycline (dox), rtTA is an active transcriptional factor leading to expression of Cre recombinase and removal of the floxed exon 3 of the α3 integrin gene, resulting in lung epithelial cell–specific loss of α3 integrin. (B) PCR using primers that encompass the floxed region of the floxed α3 integrin gene. DNA from lungs of FASC mice fed doxycycline revealed a 1-kb band consistent with the recombined floxed α3 integrin and a 3.5-kb band corresponding to non-recombined floxed α3 integrin in non-epithelial cells of the lung. Littermate control mice and FASC mice not on doxycycline only demonstrated the 3.5-kb band. Murine embryonic fibroblasts (MEFs) derived from a floxed α3 integrin (α3cko) mouse and treated in vitro with adenovirus expressing Cre (AdCre) was used as a positive control and only exhibited the 1-kb band. (C) Immunoblot demonstrating normal levels of α3 integrin in kidney lysate of a FASC mouse compared with littermate controls (Ctrl) lacking 1 of the transgenes but an approximately 50% reduction of α3 integrin in whole lung lysate and an approximately 80% reduction of α3 integrin in lysates of isolated AECs. (D and E) Immunostaining of isolated AECs demonstrates that about 60%–90% of FASC AECs (E) lack expression of α3 integrin compared with littermate control AECs (D).

Figure 2. Baseline phenotypes of FASC lungs.

Littermate control (A, C, and E) and FASC (B, D, and F) lung sections. (A and B) Lung sections (original magnification, ×20) stained by H&E demonstrated similar alveolar architecture. (C and D) Lung sections (original magnification, ×60) immunostained for E-cadherin (E-cad, green) and pro-SPC (red) demonstrated a similar E-cadherin staining pattern and increased numbers of pro-SPC–positive cells in FASC lungs. (E and F) Lung sections (original magnification, ×60) stained with trichrome demonstrated increased diffuse staining within the alveolar basement membranes of FASC mice. (G) Immunoblot showed decreased expression of α3 integrin, increased expression of pro-SPC, a clear increase in collagen IV (col IV), similar levels of laminin 5, and a slight increase in collagen I (col I) in FASC lung lysate compared with littermate control lung lysate.

Figure 3. Preserved acute lung injury response in FASC mice.

(A and B) Five days after intratracheal bleomycin injury, littermate control (A) and FASC (B) lung sections (original magnification, ×20) were stained with H&E and demonstrated increased inflammation in FASC mice. (C) Cell counts from BAL of littermate control and FASC mice 5 days after intratracheal saline or bleomycin injury. FASC mice had an increased number of cells compared with littermate controls (n = 4–6 per group). (D) Lung permeability determined by extravasation of intravascular ¹²⁵I-albumin into the lungs and expressed as EVP%. FASC and littermate control mice demonstrated similar permeability 5 days after bleomycin injury (n = 4 per group). (E) Lung compliance (μl/cm H₂O) was determined from anesthetized and paralyzed ventilated mice. FASC and littermate control mice demonstrated a decrease in compliance 5 days after bleomycin injury. There was a trend toward less compliance in FASC mice after bleomycin injury compared with littermate control mice after bleomycin injury (P = 0.09; n = 4 per group). (F) Total protein concentration (mg/ml) from BAL 5 days after intratracheal saline or bleomycin injury. FASC and littermate control mice demonstrated a similar increase in BAL protein after bleomycin injury (n = 4–6 per group). (G) Excess lung water (determined as described in Methods) increased similarly in FASC and littermate control mice 5 days after bleomycin injury (n = 4 per group).

Figure 4. FASC mice have impaired myofibroblast accumulation and type I collagen response to bleomycin injury.

(A and B) Lung sections (original magnification, ×20) from littermate control (A; Ctrl Bleo) and FASC (B; FASC Bleo) mice 17 days after intratracheal bleomycin injury immunostained for α-SMA. (C) Quantification of α-SMA–positive myofibroblasts revealed that FASC mice developed fewer α-SMA–positive myofibroblasts compared with controls. (D and E) Lung sections (original magnification, ×20) from littermate control mice lacking at least 1 of the 3 transgenes (D) and FASC mice (E) were stained for type I collagen 21 days after intratracheal injection with bleomycin. (F) Type I collagen content from whole lung lysate from littermate control and FASC mice 21 days after intratracheal injection with saline was analyzed by immunoblot. (G) Hydroxyproline assay from entire left lung of FASC or littermate control mice 21 days after intratracheal injection of saline or bleomycin. Uninjured FASC mice had increased levels of hydroxyproline and a blunted increase in hydroxyproline after bleomycin injury (n = 7–16 per group). (H) Relative densitometry of immunoblots for type I collagen from lung lysate of FASC or littermate control mice 21 days after intratracheal saline or bleomycin injection. FASC mice had significantly less type I collagen after bleomycin injury compared with control mice (n = 3 per group).

Figure 5. EMT develops in vivo following intratracheal injection of bleomycin.

(A) In triple transgenic mice, rtTA is expressed in lung epithelial cells using the human SPC promoter. In the presence of doxycycline (dox), rtTA is an active transcriptional factor leading to expression of Cre recombinase and the removal of the floxed portion of the ZEG allele, resulting in lung epithelial cell–specific expression of GFP. (B and C) Density plots obtained during cell sorting for GFP-positive cells from whole lung single-cell suspensions prepared from littermate control (B) and triple transgenic (C; ZEG/SPC-rtTA/Cre) mice 17 days after intratracheal bleomycin injury. Percentages of GFP-positive cells are indicated. (D) Seventeen days after intratracheal saline or bleomycin injection, GFP-positive cells were sorted from whole lung single-cell suspensions of ZEG/SPC-rtTA/tetO-Cre mice. Immunoblot demonstrates de novo expression of α-SMA and downregulation of pro-SPC in epithelium-derived cells of bleomycin-injured mice. (E–G) Seventeen days after intratracheal saline or bleomycin injury, GFP-positive cells were sorted as described above and immunostained for GFP and mesenchymal markers vimentin (E), α-SMA (F), and procollagen I (G) and demonstrated expression of mesenchymal markers in epithelium-derived cells in bleomycin-injured mice, but none in saline-treated mice. The percentages of GFP-positive cells staining for mesenchymal markers are indicated. Original magnification, ×60.

Figure 6. Lung epithelial cell α3 integrin regulates EMT in vivo.

(A and B) Lung sections (original magnification, ×60) from a triple transgenic ZEG/SPC-rtTA/tetO-Cre/α3^WT/WT mouse 17 days after intratracheal bleomycin injury, immunostained for GFP (A) and α-SMA (B). Multiple cells costaining GFP and α-SMA are indicated by arrows. (C) Percentage of GFP-positive cells co-expressing α-SMA in ZEG-FASC (ZEG/α3^fl/fl/SPC-rtTA/tetO-Cre) and littermate control (ZEG/α3^fl/WT or α3^WT/WT/SPC-rtTA/tetO-Cre) mice treated with saline or bleomycin injection. Control mice demonstrated a marked increase in cells co-expressing α-SMA and GFP after bleomycin injury compared with FASC mice (n = 4 per group).

Figure 7. α3 integrin regulates association between β-catenin and pSmad2 ex vivo.

(A and B) Primary AECs from FASC (B) and littermate control (A) mice were cultured on Fn for 4 days, then stained for F-actin with phalloidin and counterstained with DAPI (original magnification, ×60). Control cells demonstrated actin stress fibers consistent with a mesenchymal morphology, while FASC AECs demonstrated cortical actin staining consistent with an epithelial morphology. (C) Primary AECs from FASC or littermate control mice lacking 1 of the 3 transgenes were analyzed by immunoblot immediately after isolation (day 0) and 4 days after culturing on Fn-coated surfaces (day 4). FASC AECs had a blunted expression of mesenchymal markers collagen I, α-SMA, and vimentin. (D) Primary AECs were infected with lentivirus expressing mycRI or GFP as a control. Immunoprecipita­tion of mycRI demonstrated coprecipitation of α3 integrin and E-cadherin. (E) Coimmunoprecipitation of β-catenin (β-cat) and pSmad2 was seen with AECs from α3^+/+ (control) but not AECs from FASC mice plated on Fn for 4 days to allow activation of TGF-β1 (top blot). TGF-β1–dependent tyrosine phosphorylation of β-catenin at Y654 and pY654–β-catenin/pSmad2 coprecipitation was only seen with AECs from α3^+/+ mice (bottom blot). (F) Primary AECs cultured on Fn for 4 days then stained for α-SMA and pY654–β-catenin showed nuclear accumulation of β-catenin (original magnification, ×40).

Figure 8. β-catenin and pSmad2 coimmunoprecipitation in murine lungs following bleomycin injury.

(A) Two weeks after intratracheal bleomycin or saline injection, FASC and littermate control lungs were lysed and analyzed by immunoblot and immunoprecipitation for β-catenin. Control mice injured with bleomycin demonstrated β-catenin/pSmad2 coimmunoprecipitation. (B and C) FASC (C) and littermate control (B) fresh frozen lung section (original magnification, ×60) 17 days after bleomycin injury immunostained for α-SMA (green) and pY–β-catenin (red). Numerous nuclei stained for pY–β-catenin within and around myofibroblast clusters in littermate control but not FASC mice.

Figure 9. pY654–β-catenin/pSmad2 complexes in IPF lungs.

(A) Normal human lung samples (N1–N4) and IPF lung samples (F1–F5) were lysed and analyzed by immunoblot and immunoprecipitation for β-catenin or pY654–β-catenin. All IPF samples demonstrated increased pSmad2 and pY–β-catenin. β-catenin and pY–β-catenin coimmunoprecipitated with pSmad2 in IPF samples, but not normal lung samples. (B and C) Fresh frozen normal (B) and IPF (C) lung sections (original magnification, ×20) were stained for pY–β-catenin (red) and α-SMA (green). Numerous nuclei stained for pY–β-catenin in IPF lung but not in normal lung. Myofibroblasts in IPF lung were frequently pY–β-catenin positive.