Astragaloside IV modulates TGF‐β1‐dependent epithelial‐mesenchymal transition in bleomycin‐induced pulmonary fibrosis

Abstract Epithelial‐mesenchymal transition (EMT) plays an important role in idiopathic pulmonary fibrosis (IPF). Astragaloside IV (ASV), a natural saponin from astragalus membranaceus, has shown anti‐fibrotic property in bleomycin (BLM)‐induced pulmonary fibrosis. The current study was undertaken to determine whether EMT was involved in the beneficial of ASV against BLM‐induced pulmonary fibrosis and to elucidate its potential mechanism. As expected, in BLM‐induced IPF, ASV exerted protective effects on pulmonary fibrosis and ASV significantly reversed BLM‐induced EMT. Intriguing, transforming growth factor‐β1 (TGF‐β1) was found to be up‐regulated, whereas Forkhead box O3a (FOXO3a) was hyperphosphorylated and less expressed. However, ASV treatment inhibited increased TGF‐β1 and activated FOXO3a in lung tissues. TGF‐β1 was administered to alveolar epithelial cells A549 to induce EMT in vitro. Meanwhile, stimulation with TGF‐β1‐activated phosphatidylinositol 3 kinase/protein kinase B (PI3K/Akt) pathway and induced FOXO3a hyperphosphorylated and down‐regulated. It was found that overexpression of FOXO3a leading to the suppression of TGF‐β1‐induced EMT. Moreover, ASV treatment, similar with the TGF‐β1 or PI3K/Akt inhibitor, reverted these cellular changes and inhibited EMT in A549 cells. Collectively, the results suggested that ASV significantly inhibited TGF‐β1/PI3K/Akt‐induced FOXO3a hyperphosphorylation and down‐regulation to reverse EMT during the progression of fibrosis.


| INTRODUCTION
Idiopathic pulmonary fibrosis (IPF) is a prototype of chronic, progressive and fibrotic lung disease. 1 Incidence of IPF has risen over time.
Although disease course is variable and somewhat unpredictable, the median survival time from diagnosis is 2-4 years. 2 IPF was typically characterized by excessive production and deposition of extracellular matrix (ECM), chronic inflammatory process, remodelling of abnormal lung tissue structure and the gradual decline of pulmonary function. 3 The current notion Myofibroblasts derived from epithelial cells by epithelial-mesenchymal transition (EMT) exhibit abnormal proliferation and ECM overproduction, resulting in the development of IPF. 4 And studies found approximately one-third of fibroblasts are of epithelial origin in pulmonary fibrosis. 5 When EMT was occurring in the lung, the E-cadherin in epithelial cells decreased and a-smooth muscle actin (a-SMA) mesenchymal cells increased. 6 EMT plays a critical role in the development of pulmonary fibrosis. The cytokine transforming growth factor-b (TGF-b) functions served as an important mediator of fibrogenesis. TGF-b1 induces the activation of fibroblasts to undergo a phenotypic transition to myofibroblasts, which are the effectors of the fibrotic state. 7 A plenty of works have identified that TGF-b1 is an important pro-fibrotic factor that has been shown to induce EMT in pulmonary fibrosis. 8 TGF-b1-induced EMT is mainly mediated by Smad-dependent or Smad-independent pathways. 9 Thus, anti-EMT pathway or the method of inhibiting of TGF-b1 signalling could provide a novel potential target for the treatment of IPF. anti-hypertensive, anti-asthma, and anti-fibrotic properties. 10 ASV was found to alleviate the progression of kidney fibrosis via inhibition of MAPK pathway and TGF-b/Smad signalling pathway. 11,12 Besides, ASV could protect against TGF-b1-induced EMT by inhibiting in peritoneal mesothelial cells by promoting smad 7 expression. 13 These all indicated the great potential of the anti-fibrosis and anti-EMT activity of ASV. Bleomycin (BLM), a drug widely used as an antineoplastic, causes a dose-dependent interstitial pulmonary fibrosis. 14 BLM-induced pulmonary fibrosis is the most commonly used animal model of IPF for studying disease pathogenesis and testing of novel pharmaceutical compounds. 15 Researchers have identified that ASV could attenuate BLM-induced pulmonary fibrosis by suppressing oxidative stress and inflammation. 16 In addition, a recent study found that ASV significantly inhibited BLM-induced ECM deposition. 17 However, whether the anti-pulmonary fibrosis of ASV was associated with its inhibition of EMT during pulmonary fibrosis is still unknown.

| Ethics statement
All of the animal procedures including housing, care and experimental protocols were approved by the Animal Care and Use Committee of Shandong University of Traditional Chinese Medicine. All procedures on the mice were performed in accordance with the guidelines from the National Institutes of Health.

| Pulmonary fibrosis model and treatment
Forty rats were randomly divided into 4 groups. Control group: the rats were given a single intratracheal instillation of 50 mL saline;

| Histologic analyses and Immunohistochemistry
The pulmonary tissues were prepared by being submerged in formaldehyde solution (10%) and embedded in paraffin. The

| Immunofluorescence
The previous procedures of immunofluorescence for lung tissues were the same as immunohistochemistry. After incubated with anti-

| Measurement of oxidative stress and inflammation in lung tissues
The right lung of the rats was homogenized in 50 mmol/L PBS containing 0.5% hexadecylammonium bromide and 5 mmol/L EDTA.
After the lung extracts were centrifuged at 12 000 g for 15 minutes at 4°C, the supernatants were acquired. For measurement of oxida-

| Western blotting
Total protein was extracted from frozen lung tissue or cells using  China) and the intensities of the signals were quantified using Ima-geJ software 6.0.

| Statistical analysis
Experimental results were expressed as mean AE standard deviation.
Differences were analysed by one-way ANOVA using SPASS 20.0 (SPSS, Inc., Chicago, IL, USA) and GraphPad Prism 5.0 (GraphPad F I G U R E 1 Astragaloside IV-treated mice are protected against bleomycin-induced pulmonary fibrosis. Body weight (A) and lung wet-to-dryweight ratio (B) of each rat subjected to saline or bleomycin at 30 mL/kg for 28 d with or without 20 mg/kg Astragaloside IV treatment (n = 5/group). Representative micrographs (1009) with hematoxylin and eosin staining and Masson's trichrome staining of paraffin lung sections (C). The expression level of COLIA1 mRNA detected by qPCR (n = 5/group). Representative Western blot of Type I collagen protein and densitometrically quantified data of Type I collagen/GAPDH expression ratio. *P < .05 vs control; # P < .05 vs BLM Software Inc., CA) was used to generate graphs. P < .05 was considered statistically significant.

| ASV alleviated BLM-induced pulmonary fibrosis
Following intratracheal spray of BLM, a significant weight loss and increase in the lung wet-to-dry weight ratio were observed. But, BLM-induced weight loss and the increase in the wet-to-dry weight ratio were reversed by ASV treatment (Figure 1A,B). H&E and Masson's trichrome staining showed pronounced diffuse fibrosis and lymphocyte infiltration and amounts of collagen were deposited in BLM group as compared with control group. Whereas, fibrosis showed apparent symptomatic relief in ASV treatment group compared with the BLM group ( Figure 1C). In addition, compared with controls, the expression of collagen I mRNA and protein was also markedly increased caused by BLM. And 20 mg/mg ASV alleviated collagen deposition significantly ( Figure 1D,E). These results suggest that ASV attenuated BLM-induced pulmonary fibrosis.

| ASV attenuated oxidative stress and inflammation in BLM-induced pulmonary fibrosis
In

| ASV suppressed BLM-induced EMT in pulmonary fibrosis in vivo
To investigate the occurrence of EMT during lung fibrosis, the expression levels of a-SMA and E-cadherin were analysed using

| ASV inhibited TGF-b1-mediated hyperphosphorylation and inactivity of FOXO3a in pulmonary fibrosis
TGF-b1 plays a key role in the progression of pulmonary fibrosis.
We thus investigated whether the inhibitory effects of ASV on pulmonary fibrosis was involved in its regulation of TGF-b1 levels.
BLM instillation significantly increased TGF-b1 expression in lung tissue. As expected, we found that ASV treatment significantly alleviated BLM-induced upregulation of TGF-b1 both at mRNA and protein level ( Figure 4A,B). We subsequently tested whether ASV was involved in the downstream target genes of TGF-b1. FOXO3a was a critical downstream integrator of TGF-b1 in human lung fibroblasts and intriguing, we identified that BLM notably downregulated the expression of FOXO3a mRNA ( Figure 4C). Although ASV treatment didn't influence the transcription of FOXO3a, using immunohistochemistry, we found ASV significantly inhibited the decrease in FOXO3A protein caused by BLM ( Figure 4D). Changes of FOXO3A activity were assessed by measuring the phosphorylation status of Thr32 and Ser253 of FOXO3a ( Figure 4E). The levels of phosphorylated FOXO3a at Thr32 ( Figure 4F) and Ser253 ( Figure 4G) were significantly higher in BLM group as compared to the control and ASV group. However, co-treatment with BLM + ASV significantly inhibited the hyperphosphorylation of FOXO3a thus to inhibit the inactivity of FOXO3a caused by BLM (Figure 4H),

TGF-b1-induced EMT via PI3K/Akt
It has been widely investigated that TGF-b1 can induce and promote EMT. Similarly, we found that the TGF-b1 treatment induced disappearance of intercellular junction and spindle-like appearance were   Figure 6E). Furthermore, a gain of function analysis was performed to further validated TGF-b1induced EMT through the phosphorylation of PI3K/Akt and subsequent inactivity of FOXO3a. As shown in Figure 6F,G, the expression levels of FOXO3a mRNA and protein were restored by transfected with a pcDNA3.1 + FOXO3a. Moreover, western blot and immunofluorescence assay showed that the expression of E-cadherin was increased, whereas a-SMA was decreased in TGF-b1 + FOXO3a group as compared to that in the TGF-b1 + vector group.
The similar trends were observed after treatment with SB431542 or LY294002 ( Figure 6H,I). These data indicated that ASV suppressed TGF-b1/PI3K/Akt/FOXO3a-mediated EMT in pulmonary epithelial cell. The main histopathological features of IPF is aggregates of proliferating fibroblasts and myofibroblasts-fibroblastic foci-within a myxoid-appearing matrix, resulting in scarring and destruction of the lung architecture. And this fibrotic response is driven by abnormally activated alveolar epithelial cells. 23 The activated myofibroblasts, always considered transformed from epithelial cells and displayed a different phenotype as compared to normal lung fibroblasts, produce extracellular matrix proteins which contribute to the progression of IPF. 24 As expected, destroyed alveolar epithelial cells and a great accumulation of ECM were observed in BLM stimulated lungs when compared to the control and ASV-treated groups. In contrast, when treated with ASV, a natural saponin with a potential anti-fibrotic property, the above pathological changes were significantly attenuated. This was consistent with a previous study describing that ASV ameliorated BLM-induced ECM deposition and decreased the levels of hydroxyproline and type III collagen. 17 We also showed ASV decreased the level of type I collagen.
Inflammation and oxidative stress also play a pivotal role in pulmonary fibrosis. 25 Studies supported that anti-inflammation or antioxidant treatment strategies may achieved improvement in pulmonary fibrosis. 26 We found the levels of inflammatory cytokine TGF-b1 is an attractant for fibroblasts and monocytes/macrophages and stimulates these cells to synthesize the pro-inflammatory and fibrogenic cytokines. It is also an inducer of ECM production. 27 TGF-b1 was elevated after bleomycin treatment, as was consistent with previous investigations. TGF-b1 plays an important role in pulmonary fibrosis in that it mediates the differentiation of fibroblast. 28 The best-studied and central player in EMT is TGF-b1. As a result of the notion that EMT directly contributes to myofibroblast accumulation in the lung, 5,29 groups of studies tried to find efficient approaches to inhibit or reverse this progress mediated by TGF-b1. 30 In conclusion, our integrated approach demonstrates that ASV treatment suppressed TGF-b1/PI3K/Akt pathway to active FOXO3a, thus to prevents EMT in BLM-induced pulmonary fibrosis (Figure 7).