Mechanical stress induces lung fibrosis by epithelial-mesenchymal transition

Abstract

Objectives: Many mechanically ventilated patients with acute respiratory distress syndrome develop pulmonary fibrosis. Stresses induced by mechanical ventilation may explain the development of fibrosis by a number of mechanisms (e.g., damage the alveolar epithelium, biotrauma). The objective of this study was t test the hypothesis that mechanical ventilation plays an important role in the pathogenesis of lung fibrosis.

Methods: C57BL/6 mice were randomized into four groups: healthy controls; hydrochloric acid aspiration alone; vehicle control solution followed 24 hrs later by mechanical ventilation (peak inspiratory pressure 22 cm H(2)O and positive end-expiratory pressure 2 cm H(2)O for 2 hrs); and acid aspiration followed 24 hrs later by mechanical ventilation. The animals were monitored for up to 15 days after acid aspiration. To explore the direct effects of mechanical stress on lung fibrotic formation, human lung epithelial cells (BEAS-2B) were exposed to mechanical stretch for up to 48 hrs.

Measurement and main results: Impaired lung mechanics after mechanical ventilation was associated with increased lung hydroxyproline content, and increased expression of transforming growth factor-β, β-catenin, and mesenchymal markers (α-smooth muscle actin and vimentin) at both the gene and protein levels. Expression of epithelial markers including cytokeratin-8, E-cadherin, and prosurfactant protein B decreased. Lung histology demonstrated fibrosis formation and potential epithelia-mesenchymal transition. In vitro direct mechanical stretch of BEAS-2B cells resulted in similar fibrotic and epithelia-mesenchymal transition formation.

Conclusions: Mechanical stress induces lung fibrosis, and epithelia-mesenchymal transition may play an important role in mediating the ventilator-induced lung fibrosis.

Figure 1

Effects of HCl aspiration and/or mechanical ventilation (MV) on lung compliance and tissue content of hydroxyproline over time. Ctrl = healthy control mice receiving anesthesia but not HCl or MV; E_lung = lung elastance. * p<0.05 vs Ctrl; † p<0.05 vs HCl at identical conditions, respectively.

Figure 2

Representative high-power (x40) lung histological features (A) and lung injury score (B). Hematoxylin-eosin (H&E)-stained, formalin-fixed, paraffin-embedded lung tissues from mice, subjected to HCl aspiration alone, high pressure mechanical ventilation alone (MV), and HCL + MV for 15 days. Mice that received anesthesia alone served as controls (Ctrl). Mice in the experimental conditions exhibited evidence of extensive lung injury with interstitial and alveolar edema, hemorrhage, and inflammatory cell infiltration compared with control conditions. n = 9 mice/group. * p<0.05 vs Ctrl; † p<0.05 vs HCl at identical conditions, respectively.

Figure 3

Pulmonary fibrosis after challenge with HCl and mechanical ventilation. Mice were challenged with intratracheal HCl, high pressure mechanical ventilation (MV) or HCl and MV (HCl + MV). After 15 days, lung histologic analysis was performed with Masson’s trichrome staining (A, low-power x40 and B, high-power x60) to localize collagen fibers (blue) or α-SMA [(red) C, high-power x60] to identify myofibroblasts. Mice receiving anesthesia alone served as controls (D, low-power at upper panel and high-power at lower panel). Lung fibrosis score (E). n = 9/group. * p<0.05 vs Ctrl; † p<0.05 vs HCl and MV alone.

Figure 4

mRNA expression of TGF-β and β-Catenin after challenge with HCl and mechanical ventilation (MV). Data are reported as relative densitometry of TGF-β or β-catenin over β-actin in bar graphs. * p<0.05 vs Control (Ctrl); † p<0.05 vs HCl or MV alone.

Figure 5

EMT markers in lung tissue of mice after challenge with HCl and mechanical ventilation (MV). A. mRNA expression of the epithelial markers cytokeratin-8, E-Cadherin and SP-B, and the mesemchymal marker α-SMA and Vimentin in lung homogenates at indicated time points. B. Protein expression of the EMT markers in lung homogenates. Data are reported as relative densitometry of the EMT markers over β-actin in bar graphs. * p<0.05 vs Control (Ctrl); † p<0.05 vs HCl; ‡ p<0.05 vs MV.

Figure 6

Confocal microscopic analysis of colocalization of cytokeratin-8 (red fluorescence) with α-SMA-positive myofibroblasts (green) in lung tissue at day 15 after intratracheal instillation of HCl followed 24 h later by high pressure mechanical ventilation (A). Z stack demonstrates colocalization of red and green signals by yellow color corresponding to those in panel A, indicative of EMT (B). α-SMA-positive myofibroblasts without cytokeratin-8 colocalization is present in some area of the lung tissue (C).

Figure 7

Fibrocyte counts in control mice (Ctrl), and mice 15 days after receiving intratracheal instillation of HCl, mechanical ventilation (MV) or HCl+MV. There are no statistical differences between any groups.

Figure 8

Expression of fibrotic mediators and EMT markers after mechanical stretch of human lung epithelial cells. BEAS-2B cells were subjected to cyclic stretch for 24 and 48h at a frequency of 30 cycles/min and 30% elongation. A. Representative mRNA expression of TGF-β and β-Catenin in BEAS-2B cells in response to mechanical stretch. Average relative densitometry of TGF-β or β-catenin over β-actin is reported in bar graphs. B. mRNA expression of the epithelial markers cytokeratin-8, E-Cadherin and SP-B, and the mesemchymal marker α-SMA and Vimentin in the BEAS-2B cell lysates after mechanical stretch. C. Protein expression of the EMT markers in lung homogenates. Data are reported as relative densitometry of the EMT markers normalized for β-actin. * p<0.05 vs non-stretch control cells (Ctrl) at identical conditions.

Figure 9

Immunofluorescent staining for epithelial and mesenchymal markers in human lung epithelial cells after mechanical stretch. BEAS-2B cells were subjected to cyclic mechanical stretch for 48h at a frequency of 30 cycles/min and a 30% elongation. A. Cells receiving no mechanical stretch served as control (Ctrl). B. Exposure of the cells to mechanical stretch results in a reduction and redistribution of the epithelial marker E-cadherin from intercellular junction areas into the cytoplasm, compared to control. Mesenchymal marker F-actin is enhanced after stretch. Immunofluorescent staining for α-SMA is not detected in the cells under basal conditions but is noticeable after stretch (n = 5 experiments).