An essential role for fibronectin extra type III domain A in pulmonary fibrosis

Abstract

Rationale: Tissue fibrosis is considered a dysregulated wound-healing response. Fibronectin containing extra type III domain A (EDA) is implicated in the regulation of wound healing. EDA-containing fibronectin is deposited during wound repair, and its presence precedes that of collagen.

Objectives: To investigate the role of EDA-containing fibronectin in lung fibrogenesis.

Methods: Primary lung fibroblasts from patients with idiopathic pulmonary fibrosis or from patients undergoing resection for lung cancer were assessed for EDA-containing fibronectin and alpha-smooth muscle actin (alpha-SMA) expression. Mice lacking the EDA domain of fibronectin and their wild-type littermates were challenged with the bleomycin model of lung fibrosis. Primary lung fibroblasts from these mice were assayed in vitro to determine the contribution of EDA-containing fibronectin to fibroblast phenotypes.

Measurements and main results: Idiopathic pulmonary fibrosis lung fibroblasts produced markedly more EDA-containing fibronectin and alpha-SMA than control fibroblasts. EDA-null mice failed to develop significant fibrosis 21 days after bleomycin challenge, whereas wild-type controls developed the expected increase in total lung collagen. Histologic analysis of EDA-null lungs after bleomycin showed less collagen and fewer alpha-SMA-expressing myofibroblasts compared with that observed in wild-type mice. Failure to develop lung fibrosis in EDA-null mice correlated with diminished activation of latent transforming growth factor (TGF)-beta and decreased lung fibroblast responsiveness to active TGF-beta in vitro.

Conclusions: The data show that EDA-containing fibronectin is essential for the fibrotic resolution of lung injury through TGF-beta activation and responsiveness, and suggest that EDA-containing fibronectin plays a critical role in tissue fibrogenesis.

Figure 1.

Extra type III domain A (EDA) cellular fibronectin (cFN) regulation in fibroblasts derived from patients with usual interstitial pneumonia (UIP). (A) EDA cFN and α-smooth muscle actin (α-SMA) expression are markedly increased by Western blot in lung fibroblasts from patients with idiopathic pulmonary fibrosis compared with control lung fibroblasts. The membrane was stripped and reprobed for β-tubulin as a loading control. The blot is representative of two similar experiments of separately derived lysates from each patient. (B) Relative expression of EDA-containing mRNA to total fibronectin transcripts in UIP (n = 8) and normal (n = 8) lung fibroblasts as assessed by quantitative reverse transcriptase–polymerase chain reaction. Data are presented as relative EDA inclusion into fibronectin normalized to total fibronectin expression. *P < 0.05.

Figure 2.

Extra type III domain A–null (EDA^−/−) mice fail to develop significant lung fibrosis after bleomycin-induced lung injury. (A) Lung tissue from EDA^−/− and wild-type (WT) controls 21 days after bleomycin injury was assessed for total collagen content by hydroxyproline assay. Data are representative of four independently performed experiments. *P < 0.005. (B) Lung sections of WT (top panels) and EDA^−/− (bottom panels) mice 21 d after intratracheal bleomycin stained with trichrome to evaluate collagen deposition and anti–α-smooth muscle actin (α-SMA) to identify myofibroblasts. WT mice displayed nests of α-SMA–positive cells corresponding to areas of collagen deposition (blue on trichrome stain), whereas EDA^−/− mice demonstrate minimal α-SMA or collagen staining. In EDA^−/− mice, α-SMA expression could be observed in airway (arrowhead) and vascular (arrow) smooth muscle cells. Three mice in each group were examined, with similar findings. PBS = phosphate-buffered saline.

Figure 3.

Persistent inflammation in the lungs of extra type III domain A–null (EDA^−/−) mice after intratracheal bleomycin challenge. (A) Bronchoalveolar lavage (BAL) fluid was collected on the indicated days and total cells counted. Results are expressed as mean number of cells (×10⁴). A minimum of seven mice were tested in each group at each time point. *P = 0.009. Open circles represent wild-type (WT) mice, solid triangles represent EDA^−/− mice. (B) Differential cell counting in BAL fluid isolated as in (A). Results are expressed as percentage of total cellular infiltrate. PMN = polymorphonuclear leukocytes. Open circles represent WT mice, solid triangles represent EDA^−/− mice. EDA cellular fibronectin is dispensable for transforming growth factor (TGF)-β1 production, but necessary for activation. (C) Twenty-four-hour conditioned media from WT and EDA^−/− lung fibroblasts were assayed for TGF-β1 by ELISA. No difference in 24-hour TGF-β1 production was observed. (D) Activation of TGF-β is impaired in EDA^−/− fibroblasts. WT and EDA^−/− fibroblasts were cocultured overnight with mink lung epithelial cells expressing a portion of the plasminogen activator inhibitor-1 promoter fused to luciferase. Reporter constructs cultured alone were a negative control. EDA^−/− cells possessed significantly less TGF-β activity than WT cells. *P = 0.0023.

Figure 4.

Increased mortality in extra type III domain A–null (EDA^−/−) mice after high-dose intratracheal bleomycin. By 29 days, 69.9% of wild-type (WT) mice but only 12.7% of EDA^−/− mice had survived (P < 0.01 by the log-rank test). Data are pooled from three experiments encompassing 28 EDA^−/− mice and 22 WT mice.

Figure 5.

Extra type III domain A (EDA) cellular fibronectin (cFN) is dispensable for early transforming growth factor (TGF)-β1 signaling. Lung fibroblasts from EDA^−/− and wild-type (WT) mice were serum starved, treated without (SF) or with TGF-β1 (2 ng/ml) for the indicated time points, and assayed for phosphorylated Smad2 (pSmad2) and total Smad2 by Western blot. EDA^−/− and WT fibroblasts equally demonstrate robust Smad2 phosphorylation by 1 hour after TGF-β1 treatment.

Figure 6.

Extra type III domain A (EDA)–containing fibronectin rescues EDA^−/− cell phenotypic deficiencies in transforming growth factor (TGF)-β1 responsiveness and activation. (A) Serum-starved EDA^−/− lung fibroblasts were cultured in the absence (−) or presence (+) of TGF-β1 (2 ng/ml) on tissue culture plastic (TCP) or EDA-containing cellular fibronectin (EDA cFN) for 24 hours. Cell lysates and RNA were collected for (A) α-smooth muscle actin (α-SMA) or (B) collagen I by Western blot (top panel) and semiquantitative reverse transcriptase–polymerase chain reaction (bottom panel). (C) EDA^−/− fibroblasts were cocultured overnight on TCP (−) or EDA-containing cFN (+) with mink lung epithelial cells expressing a portion of the plasminogen activator inhibitor-1 promoter fused to luciferase. The next day, lysates were assessed for luciferase activity. *P = 0.0069.

Figure 6.

Extra type III domain A (EDA)–containing fibronectin rescues EDA^−/− cell phenotypic deficiencies in transforming growth factor (TGF)-β1 responsiveness and activation. (A) Serum-starved EDA^−/− lung fibroblasts were cultured in the absence (−) or presence (+) of TGF-β1 (2 ng/ml) on tissue culture plastic (TCP) or EDA-containing cellular fibronectin (EDA cFN) for 24 hours. Cell lysates and RNA were collected for (A) α-smooth muscle actin (α-SMA) or (B) collagen I by Western blot (top panel) and semiquantitative reverse transcriptase–polymerase chain reaction (bottom panel). (C) EDA^−/− fibroblasts were cocultured overnight on TCP (−) or EDA-containing cFN (+) with mink lung epithelial cells expressing a portion of the plasminogen activator inhibitor-1 promoter fused to luciferase. The next day, lysates were assessed for luciferase activity. *P = 0.0069.

Figure 6.

Extra type III domain A (EDA)–containing fibronectin rescues EDA^−/− cell phenotypic deficiencies in transforming growth factor (TGF)-β1 responsiveness and activation. (A) Serum-starved EDA^−/− lung fibroblasts were cultured in the absence (−) or presence (+) of TGF-β1 (2 ng/ml) on tissue culture plastic (TCP) or EDA-containing cellular fibronectin (EDA cFN) for 24 hours. Cell lysates and RNA were collected for (A) α-smooth muscle actin (α-SMA) or (B) collagen I by Western blot (top panel) and semiquantitative reverse transcriptase–polymerase chain reaction (bottom panel). (C) EDA^−/− fibroblasts were cocultured overnight on TCP (−) or EDA-containing cFN (+) with mink lung epithelial cells expressing a portion of the plasminogen activator inhibitor-1 promoter fused to luciferase. The next day, lysates were assessed for luciferase activity. *P = 0.0069.