TGF‐β/Smad and JAK/STAT pathways are involved in the anti‐fibrotic effects of propylene glycol alginate sodium sulphate on hepatic fibrosis

Abstract Liver fibrosis, a consequence of unhealthy modern lifestyles, has a growing impact on human health, particularly in developed countries. Here, we have explored the anti‐fibrotic effects of propylene glycol alginate sodium sulphate (PSS), a natural extract from brown algae, in fibrotic mice and cell models. Thus, we established bile duct ligature and carbon tetrachloride mouse models and LX‐2 cell models with or without PSS treatment. Liver pathological sections and the relevant indicators in serum and liver tissues were examined. PSS prevented hepatic injury and fibrosis to a significant extent, and induced up‐regulation of matrix metalloproteinase‐2 and down‐regulation of tissue inhibitor of metalloproteinase‐1 through suppressing the transforming growth factor β1 (TGF‐β1)/Smad pathway. PSS additionally exerted an anti‐autophagy effect through suppressing the Janus kinase (JAK) 2/transducer and activator of transcription 3 (STAT3) pathway. In conclusion, PSS prevents hepatic fibrosis by suppressing inflammation, promoting extracellular matrix (ECM) decomposition and inactivating hepatic stellate cells through mechanisms involving the TGF‐β1/Smad2/3 and JAK2/STAT3 pathways in vivo and in vitro.


| INTRODUC TI ON
Liver fibrosis or scarring, a damage-induced reaction to heal wounds by encapsulating the injury, is a significant cause of morbidity worldwide and contributes to 45% mortality in developed countries. 1,2 Following chronic injury, the extracellular matrix (ECM) accumulates, mainly due to activation of hepatic stellate cells (HSC). 3,4 HSCs, normally quiescent cells that store vitamin A, can be proliferated in response to liver injury.
This process goes along with release of the transforming growth factor β1 (TGF-β1) from activated Kupffer cells and is characterized by the appearance of smooth muscle α-actin (α-SMA). 1,3,5,6 Collagen type I (Col-1), a crucial constituent of the fibril-forming matrix, increases gradually in a background of HSC activation and presents a relatively late event in hepatic fibrosis. 1,7 Matrix metalloproteinase-2 (MMP-2), a basement membrane protease and type IV collagenase, degrades the extracellular matrix and acts against fibrosis. Conversely, binding of tissue inhibitors of metalloproteinase-1 (TIMP-1) to interstitial collagenases suppresses degradation of the accumulating matrix and prevents clearance of HSCs. 8 Smad2 and Smad3 proteins are phosphorylated with release of TGF-β1 and, in turn, promoting its activity and HSC activation. 1 Propylene glycol alginate sodium sulphate (PSS) is a heparinoid drug initially identified in brown algae by a Chinese scientist. 9,10 Studies have reported protective effects of PSS on concanavalin A and ischaemia reperfusion-induced liver injury models that are attributed to its anti-inflammatory ability and activity in reducing blood viscosity. 11,12 However, no literature has studied the efficiency of PSS in liver fibrosis. Interestingly, low-molecular-weight heparin is reported to exert anti-fibrotic effects, 13 bringing forward the hypothesis that PSS could prevent hepatic fibrosis in mice.
Autophagy is a well-known dynamic cellular process that plays an important role in fibrosis. 14 Activation of HSCs has been shown to facilitate autophagic flux, 15,16 and involvement of specific molecular pathways, including Janus kinase (JAK) 2/signal transducer and activator of transcription 3 (STAT3), was in regulation of autophagy as reported. 17,18 Furthermore, several studies have investigated the roles of STAT3 in fibrosis, both in vitro and in vivo. 19,20 Based on the findings to date, JAK2/STAT3 pathway has been highlighted as a potential anti-fibrosis target in clinical therapy.
Bile duct ligature (BDL) and carbon tetrachloride (CCl 4 ) mice models and activated LX-2 cells models are widely established for the assessment of liver fibrosis and cirrhosis. 4,21 As few effective treatments exist for hepatic cirrhosis, end stage of fibrosis or precancerous stages of hepatocellular carcinoma, more research focus is required on earlier fibrosis. 22 In this study, the above fibrotic models were employed to investigate the effects of PSS following chronic administration and the underlying mechanisms.

| Drugs and reagents
Propylene glycol alginate sodium sulphate was purchased from Dalian Tianyu Pharmaceuticals Co., Ltd and disposed in saline to obtain different drug doses of 12.5, 25 and 50 mg/kg, and stored at 4°C. 12

| Animal experiments
All animal experiments were conducted following the institutional guidelines and approved by the Animal Care and Use Committee of Shanghai Tongji University, China. Healthy male C57 mice (20-22 g) 6 weeks of age, provided by Shanghai Laboratory Animal Co., Ltd, were maintained in tidy cages with free access to food and water, and equilibrated for a week before experimental enrolment.
Following the experiment, mice were killed by CO 2 inhalation.
Blood samples were collected into 20 µL heparin (1000 U/mL), and serum samples were obtained by centrifugation for 10 minutes at 4500 g and 4°C. Liver samples were excised and processed after mice were killed, and stored at −80°C.

| Cell culture
Human primary HSC cell line, LX-2, was purchased from Cell Bank of Type Culture Collection of the Chinese Academy of Sciences and maintained in Dulbecco's modified Eagle Medium (DMEM) containing penicillin, streptomycin and foetal bovine serum as described. 4 HSC cells have been shown to be a primary effector of hepatic fibrosis, and in our study, LX-2 cells were treated with 10 ng/mL TGF-β1 (PeproTech) to be activated after 24 hours. 4,23

| Biochemical measurements and enzyme-linked immunosorbent assay (ELISA)
ALT and AST activities in serum were determined according to standard protocols using the test kits and an automated chemistry analyser (Olympus AU1000). Serum α-SMA and Col-1 were detected with commercial ELISA kits (R&D Systems).

| Histological analysis
Liver specimens were washed with saline, followed by fixation in 10% formalin and embedding in paraffin. Specimens 5 µm thick were subjected to haematoxylin and eosin (HE), Sirius Red and Mason's trichrome staining to evaluate hepatic fibrosis under a light microscope. The degree of liver fibrosis was examined by a specialist blinded to sample information according to strict criteria. Fibrosis scores and related stages were as follows: 0, no fibrosis; 1, perisinusoidal or periportal fibrosis; 2, perisinusoidal and portal/periportal fibrosis; 3, bridging fibrosis; and 4, cirrhosis. 24

| Immunohistochemistry
Immunostaining of liver sections embedded in paraffin was performed with primary antibodies against TGF-β1, Col-1, α-SMA, MMP-2, TIMP-1, Beclin-1, p62, p-smad2, p-smad3 and p-STAT3 as reported. 25 Sections were subsequently incubated with the corresponding secondary antibodies. Bound antibodies were observed under a light microscope and images obtained with a digital camera (Leica Wetzlar) after incubation with a peroxidase substrate (DAB) kit (Vector).

| Reverse transcription (RT)-PCR
Total RNA was obtained with TRIzol reagent (Takara) using standard protocols. After purity determination, single-stranded cDNA was synthesized via reverse transcription using a ThermoScript RT-PCR system (Invitrogen). Related gene expression was examined with a ViiA™ 7 Real-Time PCR System (Applied Biosystems).
All the detections were performed in duplicate, and experiments were repeated at least three times. The primers used are listed in Table 1.

| Detection of cell viability and the TGF-β/ Smad and JAK/STAT pathways in vitro
We detected the viable cells before and after the activation of TGF-

| Immunofluorescence (IF)
LX-2 cells were cultured, and the IF staining against p-Smad2 and p-Stat3 was carried based on a recent study. 27

| Statistical analysis
Results were presented as means ± standard deviation (SD) of at least three repeated independent experiments. Statistical analysis was conducted with the Student t test. P-values < .05 were considered significant.

| Protection of liver against the effects of BDL and CCl 4 by PSS
Sham operation or oil injection led to no significant differences in HE staining when compared with the NC group, and it was consistent with a recent research. 21 The results of HE staining in the experimental models revealed considerably greater histological changes in the BDL group relative to the sham group. We observed a marked reduction in these changes after  Figure 1A). Injection of CCl 4 for 12 weeks caused significant damage to liver, compared with the oil vehicle group, which was markedly prevented with simultaneous PSS treatment ( Figure 1B). Next, we examined the expression of IL-6, an important inflammatory factor in liver injury, via Western blot ( Figure 1C) Figure 1D). Notably, IL-6 expression increased in the BDL and CCl 4 groups was inhibited by PSS. To further confirm the protective function of PSS, ALT and AST activities in BDL operation and repeat CCl 4 injection groups were examined.
Notably, the significantly elevated activities of ALT and AST observed in both fibrosis models were reduced in the presence of PSS ( Figure 1E).

| PSS prevents hepatic fibrosis induced by BDL and CCl 4
To evaluate liver fibrosis, we examined deposition of Col-1 with

| PSS up-regulates MMP-2 and down-regulates TIMP-1 in hepatic fibrosis
Next, the expression patterns of MMP-2 and TIMP-1 were examined in our fibrosis models. MMP-2 protein expression was significantly decreased after the BDL operation or CCl 4 injection, which was enhanced upon PSS treatment. Conversely, high TIMP-1 levels observed in the BDL and CCl 4 models were down-regulated upon PSS injection, indicating alleviation of fibrosis ( Figure 3A). Moreover, immunostaining with anti-MMP-2 and TIMP-1 antibodies confirmed the expression patterns of MMP-2 and TIMP-1 in tissue samples from PSS-treated mice ( Figure 3B). MMP-2 and TIMP-1 mRNA expression were consistent with those of the corresponding proteins ( Figure 3C).

| PSS suppresses the TGF-β1/Smad pathway in hepatic fibrosis
Phosphorylated Smad2 (p-Smad2) and phosphorylated Smad3 (p-Smad3) generated by TGF-β1 lead to the activation of hepatic fibrosis. 1 Accordingly, we evaluated p-Smad2, p-Smad3 and TGF-β1 expression to determine the mechanisms underlying the antifibrotic effects of PSS. In the BDL and CCl 4 model groups, TGF-β1, p-Smad2 and p-Smad3 levels were remarkably increased, which were suppressed following PSS treatment ( Figure 4A). We further examined these factors in liver tissues to confirm Western blot findings ( Figure 4B). As expected, PSS effectively reduced the expression levels of TGF-β1, p-Smad2 and p-Smad3 ( Figure 4B). Notably, the p-Smad2 and p-Smad3 in the liver tissues of BDL and CCl 4 group accumulated in the nuclei when compared with the sham or oil group. But this tendency could be mitigated with PSS treatment.
Consistently, mRNA expression of TGF-β1 was decreased upon treatment with PSS after BDL operation or repeated CCl 4 injection ( Figure 4C).

| PSS inhibits the JAK2/STAT3 pathway to exert anti-autophagic effects in hepatic fibrosis
In view of the finding that autophagic death plays a significant role in progression of hepatic fibrosis, we examined the potential impact of the JAK2/STAT3 pathway on regulation of cellular autophagy. Suppression of JAK2/STAT3 pathway has been showed to ameliorate liver fibrosis in rat in a recent study. 28 JAK2 is a novel regulator of TGF-β, and TGF-β/JAK2/STAT3 signalling is regarded as a non-canonical TGF-β signalling in fibrosis. [29][30][31] Protein expression of JAK2 and total STAT3 was not significantly different among the groups, while BDL-operated and CCl 4 -treated mice showed enhanced expression of phosphorylated STAT3 ( Figure 5A). STAT3 has been shown to be a major regulator of autophagy in some recent studies. 32,33 Beclin-1, a key regulator of autophagy, tended to show a significant increase in both BDL and CCl 4 models, which was markedly suppressed by PSS. Conversely, p62, inhibited in BDL-operated and CCl 4 -treated groups, was promoted upon synchronous PSS injection. Significant immunohistochemical staining of p-STAT3 in liver tissues was observed in hepatic fibrosis model mice, and in particular, the expression of p-STAT3 was highlighted in the nuclei. Meanwhile, PSS administration had a marked suppression effect ( Figure 5B). The results of Beclin-1 and p62 staining and RT-PCR were consistent with Western blot data ( Figure 5C).

| PSS suppressed the activation of HSCs downregulated the activation of TGF-β/Smad and JAK/ STAT pathways in vitro
To further reveal the inhibitory effect of PSS on hepatic fibrosis, we then used LX-2 cells to perform in vitro studies. As shown in Figure 6A, PSS showed no significant cytotoxicity in normal HSCs.
However, the viability of TGF-β1-activated LX-2 cells was significantly suppressed with PSS administration in a dose-dependent way in this study, and the IC50 of PSS was 5.8 mg/mL according to the data ( Figure 6B). Therefore, PSS of 5.

| D ISCUSS I ON
In this study, we demonstrated the anti-fibrotic activity of PSS and examined the mechanisms underlying PSS-mediated hepatic fibrosis amelioration in our models (Figure 7). PSS suppressed inflammation and synthesis of ECM and inhibited activation of HSCs through a mechanism involving the TGF-β1/Smad2/3 pathway. Furthermore, JAK2/STAT3 pathway-related autophagy was also identified as a potential therapeutic target for hepatic fibrosis in our study.
Chronic liver injury leads to hepatic fibrosis, which usually progresses to cirrhosis, hepatocellular carcinoma and liver failure. 22 While significant progress has been made in clarifying the underlying mechanisms of hepatic fibrosis, no effective drugs or treatment methods are available at present, 36 highlighting the urgent medical requirement for novel anti-fibrosis strategies. PSS, an oral heparinoid with good anti-coagulative, hypotensive and anti-inflammation activities, has been used to treat patients with cerebrovascular, cardiovascular and other diseases in China for the past 30 years. 10 The low cost and high efficacy of this compound make it an ideal choice as a long-term therapeutic drug. 37 Here, we used the widely accepted BDL and CCl 4 mouse models and LX-2 cell models to verify the protective effects of PSS.
Inflammation is a critical regulator in hepatic fibrosis. 6 Hepatic injury triggers inflammatory activity and releases cytokines to synthesize ECM through activating HSCs. 38 And high level of TGF-β1 and IL-6 produced by Kupffer cells has been investigated to be important pro-inflammatory factors in HSC activation. 6  . Above experiments were all repeated three times. Data are presented as means ± SD. # P < .05 for TGF-β1-treated group vs normal group; & P < .05 for PSS group vs TGF-β1-treated group F I G U R E 7 Potential mechanism of action of PSS against hepatic fibrosis in our models. After BDL operation or CCl4 injection, Kupffer cells are activated to secrete IL-6 and HSCs were activated to produce TGF-β. Through the TGF-β1/Smad2/3 pathway, higher levels of Col I and α-SMA are produced that induce excessive ECM accumulation. Higher Beclin-1 and lower p62 levels are secreted through the JAK2/ STAT3 pathway to induce cellular autophagy. Up-regulation of TIMP-1 is associated with hepatic fibrosis, and the TIMP-1/MMP-2 balance is altered to regulate production and elimination of the extracellular matrix. PSS significantly suppresses the production of TGF-β1 and exerts anti-fibrosis effects Accumulating evidence has established that autophagy is a major participant in numerous liver diseases, such as autoimmune hepatitis, hepatic ischaemia reperfusion injury and hepatocellular carcinoma. 49 Although the autophagy of HSCs may moderate fibrogenesis in some studies, it saves energy for activated HSCs. 47 Increased autophagic flux during HSC activation was recently shown to be involved in the pathogenesis of hepatic fibrosis. 15,50 Beclin-1, together with autophagy-related genes (ATGs) and class III phosphatidylinositol 3-kinase (VPS34), forms a complex, which is necessary to the formation of autophagosome. 51 As one of the autophagy adaptors, p62 sequesters the autophagosome by selectively targeting cargo to the autophagosome membrane. 51 In our study, Beclin-1 displayed significantly higher expression in the fibrotic model groups and p62 was simultaneously inhibited. Interestingly, PSS administration led to suppression of autophagy conditions, clearly indicative of anti-autophagic activity.
We further demonstrated an important role of JAK2/STAT3 in fibrotic autophagy in our study. Both TGF-β1 and IL-6 have been shown to be regulators of JAK2/STAT3 pathway. 31 HIF1A promotes the expression of BNIP3 and its ligands, which can strength the inductions of autophagosome. 52 While the expression of total STAT3 and JAK2 proteins among these groups was not significantly different in our study, the p-STAT3 level was promoted in fibrotic mice and activated LX-2 cells, which was suppressed by PSS, as expected. These findings clearly demonstrate that PSS administration markedly inhibits the activation of the JAK2/STAT3 pathway, in turn, preventing autophagy, which may at least partially explain the mechanisms by which PSS exerts anti-fibrotic effects and inhibits HSC activation.
Our results collectively suggest that PSS prevents hepatic fibrosis by suppressing inflammation, promoting ECM decomposition and inactivating HSCs through the TGF-β1/Smad2/3 and JAK2/STAT3 pathways. Because of its multiple therapeutic effects, large number of sources and low cost, the natural extract PSS may effectively serve as a clinical supplement to treat hepatic fibrosis.

| CON CLUS IONS
Propylene glycol alginate sodium sulphate exerts anti-fibrotic activity in BDL and CCl 4 mouse models, and in vitro. TGF-β1/Smad2/3 and JAK2/STAT3 pathways are closely associated with the therapeutic effects of PSS, supporting its potential as an effective treatment agent for hepatic fibrosis.