Administration of Gas6 attenuates lung fibrosis via inhibition of the epithelial-mesenchymal transition and fibroblast activation

Abstract

The epithelial-mesenchymal transition (EMT) and fibroblast activation are major events in idiopathic pulmonary fibrosis pathogenesis. Here, we investigated whether growth arrest-specific protein 6 (Gas6) plays a protective role in lung fibrosis via suppression of the EMT and fibroblast activation. rGas6 administration inhibited the EMT in isolated mouse ATII cells 14 days post-BLM treatment based on morphologic cellular alterations, changes in mRNA and protein expression profiles of EMT markers, and induction of EMT-activating transcription factors. BLM-induced increases in gene expression of fibroblast activation-related markers and the invasive capacity of primary lung fibroblasts in primary lung fibroblasts were reversed by rGas6 administration. Furthermore, the hydroxyproline content and collagen accumulation in interstitial areas with damaged alveolar structures in lung tissue were reduced by rGas6 administration. Targeting Gas6/Axl signaling events with specific inhibitors of Axl (BGB324), COX-2 (NS-398), EP1/EP2 receptor (AH-6809), or PGD2 DP2 receptor (BAY-u3405) reversed the inhibitory effects of rGas6 on EMT and fibroblast activation. Finally, we confirmed the antifibrotic effects of Gas6 using Gas6^-/- mice. Therefore, Gas6/Axl signaling events play a potential role in inhibition of EMT process and fibroblast activation via COX-2-derived PGE₂ and PGD₂ production, ultimately preventing the development of pulmonary fibrosis.

Keywords: Axl; EMT; Fibroblast activation; Gas6; Pulmonary fibrosis.

Fig. 1

Inhibition of EMT and primary ATII cell invasion by rGas6 administration. Mice were intratracheally instilled with BLM (5 U/kg). Either rGas6 (50 μg/kg) or saline (Sal) was intraperitoneally administered 1 day before BLM treatment and once every 2 days thereafter. Mice were euthanized 14 days after BLM treatment. (a) Left: morphological changes in isolated ATII cells (Scale bars: 100 μm). Representative images are shown from three replicates per condition with cells pooled from two mice per replicate. Right: percentage of spindle shaped cells/high-power fields (HPF). (b) qRT-PCR of EMT markers in ATII cell samples. (c) Left: immunofluorescence staining for E-cadherin (green) and α-SMA (red). Right: quantification of proteins in ATII cells. Original magnification: 400 × . Scale bars: 20 μm. Imaging medium: Vectashield fluorescent mounting medium containing DAPI. (d) Immunoblot analysis of E-cadherin and N-cadherin in lung homogenates. Below: Densitometric analysis of each band normalized to that of β-actin. Values represent the means ± S.E.M. from three mice per group. (e) qRT-PCR of Snail1, Zeb1, and Twist1 in ATII cell samples. (f) Phase-contrast microscopy and quantification of invaded ATII cells. Scale bars: 100 µm. *P < 0.05, **P < 0.01, ***P < 0.001 compared with control or for BLM + Sal vs. BLM + rGas6. Data were obtained from three (c right, f below) or five replicates (b, e) per condition with cells pooled from two mice per replicate. Data are shown as the means ± S.E.M. (g) Immunofluorescence staining for E-cadherin (red), α-SMA (red), or S100A4 (green) in lung sections. Arrowheads indicate colocalization of E-cadherin in lung fibroblasts. Imaging medium: Vectashield fluorescence mounting medium containing DAPI. Scale bars: 20 μm. Representative images were obtained from three mice in each group. (h) Graph representing the number of S100A4/E-cadherin double-positive cells compared with the total S100A4-positive cell population in lung parenchyma. Mean of five HPFs/section ± S.E.M. from three mice in each group

Fig. 2

Inhibition of apoptosis in ATII cells by rGas6 administration. The experimental design was as described in Fig. 1. Mice were euthanized 14 days after BLM treatment. (a) Left: Representative TUNEL-stained and fixed ATII cells (original magnification: 400 ×). Positive staining depicted in green. Nuclei were observed by DAPI staining. Scale bars: 20 μm. Right: Quantitation of the number of TUNEL-positive cells (number/HPF) in the different groups. (b) The cell viability in primary ATII cells was measured by flow cytometry after annexin V-FITC/PI dual staining. Apoptotic cells were quantified as the sum of the percentages of cells in the early and late stages of apoptosis. **P < 0.01, ***P < 0.001 compared with control or for BLM + Sal vs. BLM + rGas6. Data were obtained from three replicates per condition with cells pooled from two mice per replicate (a right, b right). The data are shown as the means ± S.E.M. (c) Immunoblot analysis of Bax, Bcl-2, cleaved caspase-3, and cleaved PARP in lung homogenates. Below: Densitometric analysis of each band normalized to that of β-actin. (d) Representative confocal images of lung sections stained with an anti-SPC antibody (red), anti-cleaved caspase-3 antibody (green), and DAPI (blue) (left). Original magnification: × 400. Scale bars = 20 μm. Quantification of cleaved caspase-3 staining in SPC.⁺ ATII cells (right). The values represent the means ± S.E.M. of results from three mice from each group. **P < 0.01 compared with Sal control or for BLM + Sal vs. BLM + rGas6

Fig. 3

Inhibition of fibroblast activation by rGas6 administration. The experimental design was as described in Fig. 1. Mice were euthanized 14 (a, d-g) or 21 days (b) after BLM treatment. (a, d–f) Primary fibroblasts were isolated from murine lungs. (a) qRT-PCR of collagen type 1, fibronectin, and α-SMA in fibroblast samples. (b) Immunofluorescence staining for α-SMA (red) or S100A4 (green) was performed in lung sections. Arrowheads indicate colocalization of α-SMA in lung fibroblasts. Imaging medium: Vectashield fluorescence mounting medium containing DAPI. Scale bars: 20 μm. Representative images were obtained from three mice per group. (c) Graph representing the number of S100A4/α-SMA double-positive cells compared with the total S100A4-positive cell population in the lung parenchyma. Mean of five HPFs per section ± S.E.M. from three mice in each group. ***P < 0.001 compared with control or for BLM + Sal vs. BLM + rGas6. (d) Phase-contrast microscopy (left) and quantification of invaded fibroblasts (right) using Matrigel-coated Transwell plates. Scale bar: 100 µm. (e) qRT-PCR of Has2, CD44, MMP9, MMP12, and MMP14 in fibroblast samples. *P < 0.05, **P < 0.01, ***P < 0.001 compared with control or for BLM + Sal vs. BLM + rGas6. Data were obtained from five replicates per condition with cells pooled from two mice per replicate (a, d right, e). The data are shown as the means ± S.E.M. (f) Selected heatmaps showing differentially expressed genes encoding adhesion and ECM molecules in primary lung fibroblasts between the BLM + Sal and BLM + rGas6 groups. Red: increased expression; blue: decreased expression. Data were obtained from two replicates per condition with cells pooled from two mice per replicate. (g) Relative expression levels of selected genes from PCR array profiling (f). Log₂ fold-change values (ApoSQ-CAF CM vs. CAF CM, fold change > 1.5)

Fig. 4

Inhibition of lung fibrosis by rGas6 administration. The experimental design was as described in Fig. 1. Mice were euthanized on days 14 and 21 after BLM treatment. (a, b) Levels of the active form of the TGF-β1 and HGF proteins in BAL fluid were quantified by ELISAs. (c) qRT-PCR of collagen type1, fibronectin, and α-SMA in lung tissue samples. (d) Left: Immunoblot analysis of the indicated proteins in lung homogenates. Right: Densitometric analysis of each band normalized to that of β-actin. (e) Collagen deposition in the whole lung was determined by measuring the hydroxyproline content on day 21. (f) Lung sections were visualized with Masson’s trichrome staining on day 21. Representative results from five mice per group are shown (scale bar: 50 μm). (g) Ashcroft scoring of the lung sections. The values represent the means ± S.E.M. of results from three (d) or five mice (a–c, e, g) in each group. *P < 0.05, **P < 0.01, ***P < 0.001 compared with Sal control or for BLM + Sal vs. BLM + rGas6

Fig. 5

Axl activation and COX-2-derived PGE₂ and PGD₂ production induced by rGas6 administration. The experimental design was as described in Fig. 1. Mice were euthanized on day 14 after BLM treatment. (a) Left: Immunofluorescence staining for phospho-Axl (green), total Axl (red), phospho-Mer (red), and total Mer (green) in primary ATII cells. Images were captured at 400 × magnification. Right: Quantification of phospho-Axl, total Axl, phospho-Mer, and total Mer staining in ATII cells. Imaging medium: Vectashield fluorescence mounting medium containing DAPI. Scale bars: 20 μm. Data were obtained from three replicates per condition with cells pooled from two mice per replicate. (b) Left: Immunoblot analysis of total/phospho-Axl and total/phospho-Mer in lung tissue homogenates. Right: Densitometric analysis of each band normalized to that of β-actin. (c) Immunoblot analysis of total/phospho-Akt in lung tissue homogenates. Below: Densitometric analysis of each band normalized to that of total Akt. Data are from independent experiments with three mice per group (mean ± S.E.M.). (d, e) qRT-PCR of COX-2 and COX-1 in ATII cells and lung tissue samples. (f) Immunoblot analysis of COX-2 and COX-1 in lung tissue homogenates. Below: Densitometric analysis of each band normalized to that of β-actin. (d) Data were obtained from five replicates per condition with cells pooled from two mice per replicate. Data were obtained from independent experiments with five (e) or three (f) mice per group. (g, h) PGE₂ or PGD₂ levels in BAL fluid (BALF, n = 5 mice) and culture supernatants from ATII cells and alveolar macrophages (AM) were measured using an enzyme immunoassay. (h) Data were obtained from five replicates per condition with cells pooled from two mice per replicate. Values represent the means ± S.E.M. *P < 0.05, **P < 0.01, ***P < 0.001 compared with Sal control or for BLM + Sal vs. BLM + rGas6

Fig. 6

Inhibition of EMT and fibroblast activation via Gas6/Axl signaling events. Where indicated, the Axl inhibitor BGB324 (BGB, 5 mg/kg, i.o.), COX-2 inhibitor NS-398 (NS, 5 mg/kg, i.o.), EP1/EP2 inhibitor AH-6809 (AH, 5 mg/kg, i.p.), or DP2 inhibitor BAY-u3405 (BAY, 30 mg/kg, i.p.) was co-administered with rGas6 1 day before BLM treatment and then administered once/day (AH) or once every 2 days (BGB, NS, and BAY). Mice were euthanized 14 days following BLM treatment. (a, b) qRT-PCR of EMT markers and EMT-regulating transcription factors in primary ATII cells. (c, d) qRT-PCR of activated fibroblast markers and invasive myofibroblast-related molecules in primary lung fibroblasts. (e) Left: The cells were visualized by phase-contrast microscopy to analyze their invasive ability in Matrigel-coated Transwell assays. Scale bar: 100 µm. Right: The invaded fibroblasts were quantified by counting the number of cells adhering to the bottom surface of the upper chamber. *P < 0.05, **P < 0.01, ***P < 0.001 compared with BLM + Sal or for BLM + Gas6 vs. BLM + rGas6 + the inhibitor. Data were obtained from five replicates per condition with cells pooled from three mice per replicate (means ± S.E.M.)

Fig. 7

Effect of Gas6 deficiency on EMT and fibroblast activation. WT and GAS6^−/− mice were intratracheally instilled with BLM (5 U/kg). Mice were euthanized 14 days after BLM treatment. (a–c) qRT-PCR of EMT markers and EMT-regulating transcription factors in primary ATII cells (a, b) and lung tissue (c). (d) qRT-PCR of activated fibroblast markers and invasive myofibroblast phenotype-regulating molecules in primary fibroblasts. (e) Left: Immunoblot analysis of the indicated proteins in lung tissue. Right: Densitometric analysis of each band normalized to that of β-actin. (f, g) PGE₂ and PGD₂ levels in BAL fluid (BALF) and conditioned media of ATII cells and alveolar macrophages (AM) were measured using an enzyme immunoassay. *P < 0.05, **P < 0.01, ***P < 0.001 compared with Sal control or for WT + BLM vs. GAS6^−/− + BLM. Data were obtained from three (f and g middle, right) or five replicates (a, b, d) per condition with cells pooled from two mice per replicate (means ± S.E.M.). Values represent the means ± S.E.M. of results from three (e right, f and g left) or five mice (c) per group

Fig. 8

Effect of Gas6 deficiency on collagen deposition in lung fibrosis. WT and GAS6^−/− mice were intratracheally instilled with BLM (5 U/kg). Mice were euthanized 14 days after BLM treatment. (a) Collagen deposition in the whole lung was determined by measuring hydroxyproline content. (b) Lung sections were visualized with Masson’s trichrome staining on day 14. Representative results from three mice per group are shown (scale bar: 50 μm). (c) Ashcroft scoring of the lung sections. *P < 0.05, **P < 0.01, ***P < 0.001 compared with Sal control or for WT + BLM vs. GAS6^−/− + BLM. Values represent the means ± S.E.M. of results from three (c) or five mice (a) per group

Fig. 9

A schematic diagram summarizing the role of Gas6/Axl signaling events for the prevention of lung fibrosis. rGas6 inhibits EMT and apoptosis in ATII cells and concomitantly suppresses fibroblast activation, consequently preventing the development of BLM-induced lung fibrosis. This occurs through the activation of Axl signaling pathway, including COX-2-derived PGE₂ and PGD₂ production in ATII cells and alveolar macrophages