Isorhamnetin Inhibits Liver Fibrosis by Reducing Autophagy and Inhibiting Extracellular Matrix Formation via the TGF-β1/Smad3 and TGF-β1/p38 MAPK Pathways

Objective Liver fibrosis is a consequence of wound-healing responses to chronic liver insult and may progress to liver cirrhosis if not controlled. This study investigated the protection against liver fibrosis by isorhamnetin. Methods Mouse models of hepatic fibrosis were established by intraperitoneal injection of carbon tetrachloride (CCl4) or bile duct ligation (BDL). Isorhamnetin 10 or 30 mg/kg was administered by gavage 5 days per week for 8 weeks in the CCl4 model and for 2 weeks in the BDL model. Protein and mRNA expressions were assayed by western blotting, immunohistochemistry, and quantitative real-time polymerase chain reaction. Results Isorhamnetin significantly inhibited liver fibrosis in both models, inhibiting hepatic stellate cell (HSC) activation, extracellular matrix (ECM) deposition, and autophagy. The effects were associated with downregulation of transforming growth factor β1 (TGF-β1) mediation of Smad3 and p38 mitogen-activated protein kinase (MAPK) signaling pathways. Conclusion Isorhamnetin protected against liver fibrosis by reducing ECM formation and autophagy via inhibition of TGF-β1-mediated Smad3 and p38 MAPK signaling pathways.


Introduction
Liver fibrosis is associated with long-term chronic liver diseases caused by viral infection, metabolic disorders, drugs, cholestatic diseases, alcohol abuse, or immune attack [1]. Sustained liver insult leads to progressive fibrosis and ultimately to cirrhosis and hepatocellular carcinoma. Liver fibrosis represents a significant global health burden [2]. Antiviral drugs or immunosuppressive agents may prevent or attenuate liver fibrosis in a subset of cases. However, effective treatments are not available for many patients, leading to end-stage liver cirrhosis and poor prognosis [3,4]. The lack of preventive and therapeutic interventions has created an urgent need for effective antifibrotic agents.
Liver fibrosis is characterized by overproduction of collagen type I (Col-1) and deposition of extracellular matrix (ECM) in the subendothelial space [5,6]. Hepatic stellate cells (HSCs) are the main source of ECM. They are quiescent under physiological conditions and store hepatic retinoid and vitamin A [7]. Activation of HSCs by autocrine or paracrine stimuli promotes development of a myofibroblast-like phenotype, with loss of lipid droplets, and expression αsmooth muscle actin (α-SMA) [8]. Activated HSCs are responsive to both fibrogenic and proliferative stimuli that promote the proliferation of fibrogenic cells and the production of ECM. The activity of matrix metalloproteinase (MMP) and tissue inhibitor of matrix metalloproteinase (TIMP) enzymes, which regulate ECM formation and removal, is also disturbed in liver fibrosis [9]. The complex changes during liver fibrosis result in distortion and disruption of hepatic tissue structure and function [10]. Inhibition of HSC activation and proliferation and maintaining balanced ECM production and degradation are key potential liver fibrosis treatment targets.
Autophagy helps to maintain normal liver homeostasis by removal of defective organelles and components. Autophagy initiates with the formation of autophagosomes, which are vacuoles bounded by a double membrane that fuses with lysosomes and delivers the defective components, which are then digested [17]. It is well-known that autophagy is increased with HSC activation in mouse fibrosis models and in human fibrotic liver tissue. Blocking autophagy in HSCs inhibits HSC activation and alleviates liver fibrosis [18,19]. Thus, interventions involving autophagy in HSCs may be an attractive therapeutic target for hepatic fibrosis.
Isorhamnetin (IH) is a flavonol aglycone isolated from the plant Hippophae rhamnoides L. that has anti-inflammatory, antioxidant, and antitumor activity [20,21]. IH has hepatoprotective effects by inhibiting hepatocyte autophagy and apoptosis [20]. Zheng et al. reported that IH protected against pulmonary fibrosis induced by bleomycin in mouse models by inhibiting epithelial-mesenchymal transition and endoplasmic reticulum stress [22]. Recently, Yang et al. reported that IH attenuated carbon tetrachloride-(CCl 4 -) induced liver fibrosis by downregulating TGF-β/Smad signaling and reducing oxidative stress [23]. However, whether IH could attenuate bile duct ligation-(BDL-) induced hepatic fibrosis is unclear, and the involvement of autophagy has not been investigated.
Based on these, this study is aimed at investigating the role of IH on CCl 4 -or BDL-induced mouse models of hepatic fibrosis, mainly focus on the involvement of the TGF-β1/Smad3 and TGF-β1/p38 MAPK pathway signaling and autophagy. 2.2.1. Animal Experiment. A preliminary experiment was performed before the formal experiment to show the safety of IH treatment and sham operation on the liver. Mice were randomly allocated in four groups (5 animals per group): a control group that accepted no treatment, a vehicle group that was injected intraperitoneally with olive oil three times a week, a sham operation group with laparotomy only, or an IH group gavaged with 30 mg/kg IH in saline five times a week. The mice were treated for 8 weeks and sacrificed under anesthesia. Serum and hepatic tissues were obtained for biochemical assays and histopathology.

Materials and Methods
We then established the following two animal models to study the effects of IH on liver fibrosis.
Mice in the CCl 4 -induced fibrosis model were given with 10% CCl 4 (1 mL/kg, diluted in olive oil; i.p.) three times a week for 8 weeks. IH in saline was given by gavage at either 10 or 30 mg/kg five days per week for 8 weeks. Mice were randomly allocated in four groups (8 animals per group): a vehicle control group injected intraperitoneally with olive oil, a model group intraperitoneal injection with CCl 4 , intraperitoneal injection of CCl 4 and 10 mg/kg IH, or intraperitoneal injection of CCl 4 and 30 mg/kg IH.
Mice in the BDL model were randomly allocated in four groups (eight animals each group): a sham control group, BDL group, BDL+10 mg/kg IH group, and BDL+30 mg/kg IH group. Surgery was performed as previously described [14]. IH was administrated 24 h after the surgery. Mice were given 10 or 30 mg/kg IH by gavage for 5 days per week for 2 weeks.

Histopathology.
Fresh liver tissues were fixed with 4% paraformaldehyde for 2 days at room temperature, embedded in paraffin, and then sectioned at 5 μm. The thin sections were stained with hematoxylin and eosin (H&E) or Masson's trichrome following standard methods [13]. The tissues were viewed under light microscopy to evaluate liver injury and extent of collagen fibers.

Statistical
Analysis. Statistical differences were evaluated using analysis of variance followed by Tukey's correction for multiple comparisons. Data were analyzed with SPSS version 20.0 (Chicago, IL, USA). A value of p < 0 05 was considered statistically significant.

IH and Sham Operation Had No
Harmful Effects on the Liver. Serum ALT and AST and liver hydroxyproline in normal controls, sham-operated, vehicle-treated, and IHtreated mice were not significantly different (Figure 1(a)). H&E and Masson staining did not find any obvious pathological changes in the four groups ( Figure 1(b)). The results indicated that IH, vehicle, and sham operation had no harmful effects on the liver.

IH Attenuated Liver Fibrosis Induced by CCl 4 and BDL
in Mouse. ALT and AST, two markers of liver injury, increased significantly following CCl 4 injection and BDL surgery (Figure 2(a)). IH decreased liver enzymes in a dosedependent manner, which indicated that IH protected the liver from chronic injury. Hydroxyproline, a constituent of liver collagen, also increased significantly in CCl 4 -and BDLtreated mice. As shown in Figure 2(a), IH dose-dependently decreased liver hydroxyproline levels. Histological evaluation (Figure 2(b)) was performed by H&E and Masson staining. H&E-stained tissue revealed that inflammatory infiltration, hepatocyte swelling and necrosis, damage of the liver lobules, and the formation of fibrous septa were prominent in both CCl 4 -and BDL-treated mice. Bile duct proliferation was also evident in mice with BDL-induced liver fibrosis. The histological changes observed in both fibrosis models were significantly attenuated by IH. The extent of the improvement increased with the IH dose. Masson staining demonstrated extensive collagen deposition in both fibrosis models. Collagen deposition was significantly reduced by IH. The results confirmed the protective effects of IH against liver fibrosis in both mouse models.

IH Inhibited Massive Macrophage Recruitment in the
Liver. F4/80 is a marker of the liver macrophages (Kupffer cells). We determined the expression of F4/80 to analyze the effect of IH treatment on macrophage infiltration. The mRNA expression of F4/80 was significantly elevated in the fibrotic livers induced by CCl 4 and BDL when compared to the normal livers determined by qPCR; however, IH markedly reduced the expression of F4/80 in liver tissues (Figure 3(a)). Immunohistochemistry of F4/80 showed that spindle-shaped F4/80-positive cells increased significantly in the hepatic fibrosis groups compared with the control groups. However, the elevation was obviously inhibited by IH treatment (Figure 3(b)). The results indicated that IH could inhibit chronic CCl 4 -and cholestasis-induced massive macrophage recruitment in liver tissues.

IH Inhibited HSC Activation and ECM Formation. α-
SMA and PPAR-γ are considered as markers of HSC activation and quiescence, respectively [24][25][26]. mRNA and protein levels of α-SMA were significantly upregulated, whereas the PPAR-γ expression was markedly downregulated in mice exposed to chronic CCl 4 and BDL compared with controls. IH dose-dependently reduced the α-SMA expression and increased the PPAR-γ expression in liver tissues. Collagen (especially types I and III) is the main component of ECM in liver tissues. The qPCR, western blotting, and immunohistochemistry results showed that the Col-1 expression in the liver was obviously elevated in both fibrosis model mice compared with controls, whereas IH significantly   reduced the collagen expression in liver tissues (Figures 4(a)-4(c)). MMP-2 has been shown to be involved in suppressing the collagen expression, and TIMP1 overexpression has been associated with inhibiting ECM clearance [27,28]. As shown in Figures 4(a) and 4(b), the MMP-2 expression was significantly decreased, while the expression of TIMP1, an MMP inhibitor, was increased in both fibrosis models. As shown by qPCR and western blotting, both mRNA and protein expressions were affected in the fibrosis models.
The results are consistent with the activation of HSC and overproduction and impaired degradation of ECM in the fibrosis models. IH decreased the expression of TIMP1 and increased the MMP-2 expression at both mRNA and protein levels in a dose-dependent manner (Figures 4(a)-4(c)). The results showed that IH inhibited HSC activation and maintained the balance of ECM production and degradation in both fibrosis models.

IH Reduced Autophagy in Both Liver Fibrosis Models.
Beclin-1 and LC3 expressions are associated with autophagosome formation and considered autophagy markers. mRNA and protein levels of beclin-1 and LC3 were significantly elevated in both fibrosis models compared with control mice (Figures 5(a) and 5(b)); however, IH prevented their increase in a dose-dependent way. The results of beclin-1 immunohistochemical staining were consistent with the results of western blotting ( Figure 5(c)). The results indicated that IH inhibited autophagy in both liver fibrosis mouse models.    3.6. IH Downregulated TGF-β1-Activated Smad3 and p38 MAPK Signaling Pathways. The expression of TGF-β1, the most potent fibrogenic cytokine in the liver, was prominently increased in both fibrosis models, but IH treatment significantly decreased both mRNA and protein levels of TGF-β1 in liver tissues (Figures 6(b)-6(d)). The changes of TGF-β1 in serum were consistent with those in liver tissues (Figure 6(a)). We then focused on p-Smad3 and p-p38 MAPK, which are the downstream signaling molecules of TGF-β1. As shown by western blot assays and immunohistochemistry (Figures 6(c) and 6(d)), the expression of p-Smad3 and p-p38 MAPK proteins was upregulated in liver fibrosis tissues compared with controls, while both were dose-dependently attenuated by IH. The results demonstrated that IH improved liver fibrosis through downregulating TGF-β1 mediation of Smad3 and p38 MAPK signaling.

Discussion
Hepatic fibrosis, or liver scarring, is a response to liver injury. Sustained insult may lead to progression of liver fibrosis that produces serious, irreversible liver disease, cirrhosis, and hepatic carcinoma [1,5,6]. Therapeutic strategies for liver fibrosis are still limited. Studies showed that blocking the    TGF-β1 pathway could be a potential strategy for liver fibrosis [13][14][15][16]. TβR antagonists could effectively inhibit the TGF-β pathway and showed the antifibrotic potency both in vitro and in vivo [29,30]. However, the on-target toxicity of TβR antagonists, especially cardiac injuries, should be highly concerned. The concentration must be carefully controlled to reduce the toxicity, which limited the development and application of TβR antagonists [31][32][33]. Therefore, discovering an effective and safe therapeutic agent for liver fibrosis is urgently needed. Our previous study showed that IH treatment is hepatoprotective [20]. Recent studies also showed the protective role of IH in cardiovascular diseases [34], lung injury [35], and renal damage [36]. These studies demonstrated that IH treatment is safe and protective for multiple organ function in animal models. Remarkably, IH has previously been reported to show antifibrosis activity in lung and cardiac muscle [22,37]. IH was also found to reduce the extent of CCl 4 -induced hepatic fibrosis by reducing oxidative stress [23]. Therefore, IH might be a promising candidate for hepatic fibrosis treatment.
In this study, we investigated the antifibrotic effects of IH in both CCl 4 -and BDL-induced mouse models of liver fibrosis. Intraperitoneal injection of CCl 4 is a reliable method for the establishment of stable fibrotic models. The pathological appearance is similar to that of human hepatic fibrosis caused by hepatitis B infection [38]. BDL-induced fibrosis is similar to injury caused by cholestatic obstruction [39]. Our results showed that IH at both 10 and 30 mg/kg effectively protected hepatic function and improved liver pathology changes in both fibrosis models. The therapeutic mechanism of IH in liver fibrosis involves downregulating the TGF-β1/Smad3 and TGF-β1/p38 MAPK pathways. Therefore, the present study indicated that IH could be an attractive agent for the clinical therapy of liver fibrosis.
ALT and AST are two liver enzymes that reflect hepatocyte integrity. Liver enzymes leak into the bloodstream by liver injury and increase their levels in serum, which reflects hepatocyte injury [40]. The elevated ALT and AST levels were recovered following IH treatment in both fibrotic models, which suggested that IH protected hepatocytes from chronic injury. In parallel, IH treatment alleviated inflammation infiltration in liver sections. Our previous study also showed that IH inhibited hepatocyte apoptosis and autophagy induced by inflammation, thereby improving liver function [20]. It indicated that the hepatoprotective role of IH could be attributed to the anti-inflammation effect [41,42].
The importance of HSC activation in hepatic fibrosis is well known. Chronic damage of hepatic tissue leads to activation of resting HSCs and transdifferentiation into myofibroblasts. The myofibroblasts express α-SMA, lose their lipid droplets, and contribute to the formation of ECM [8]. The progressive accumulation of ECM in the Disse space produces mechanical irritation that continues to drive HSC activation [24], and the binding and accumulation of growth factors within the ECM further enhance HSC proliferation [24,43]. Therefore, decreased matrix deposition may attenuate liver fibrosis. ECM production is modulated by TIMPs and MMPs. MMPs degrade ECM proteins, while TIMPs inhibit the proteolytic activity of MMPs. The balance between MMP and TIMP activity is disturbed in liver fibrosis [28,44]. Shifting the MMP-TIMP balance facilitates the deposition of ECM. In this study, changes in α-SMA and Col-1 expressions revealed that HSC activation and ECM accumulation were reduced in liver tissues treated with IH. We also examined the expression of PPAR-γ, a quiescent HSC marker, in liver tissues. Studies have shown that PPAR-γ is involved in maintaining the quiescent HSC phenotype and acts as a negative regulator in HSC transdifferentiation [25,26]. The induction of PPAR-γ by IH reflects the alleviation of liver fibrosis and also plays a great role in maintaining quiescence of HSCs [45,46]. Additionally, the study confirmed, as would be expected, that the balanced expression of MMP-2 and TIMP1 was changed in both fibrosis models, but it was recovered by IH treatment. The results indicate that IH attenuated liver fibrosis by inhibiting HSC activation and regulating ECM balance.
Immune cells, especially liver macrophages (Kupffer cells), also play a pivotal role in hepatic fibrosis. In response to liver injury, massive Kupffer cells recruit to the liver, and produce inflammatory cytokines (such as TGF-β1), which perpetuate inflammation and mediate HSC activation [47,48]. Our result indicated that IH could effectively inhibit inflammation and regulate Kupffer cell recruitment in the chronic damaged liver, which contributed to the alleviation of liver fibrosis caused by chronic CCl 4 and cholestasis exposure [48].
TGF-β1 is a fibrogenic cytokine secreted by both liver macrophages (Kupffer cells) and activated HSCs [11], and it promotes transdifferentiation and proliferation of HSCs. It targets the α1 and α2 procollagen type I genes and TIMP-1 and TIMP-2 expressions [11,[49][50][51][52]. The TGF-β1 expression mediates intracellular signaling through the Smad and p38 MAPK pathways [53]. Binding of TGF-β to the TGF-β receptor (TβR) type II transphosphorylates TβR type I, which subsequently phosphorylates downstream pathways of Smad [11,54]. The phosphorylated Smad protein then combines with Smad4, and the complex translocates into the nucleus where it regulates the transcription of genes coding collagens, α-SMA, and TIMP1 [12,55,56]. Suppression of Smad3 in HSCs is known to inhibit the Col-1 expression and reduce liver fibrosis [12,57,58]. In addition, other signal transducers such as the p38 MAPK pathway can also be medicated by activation of TβR. Previous studies have found that p38 MAPK signaling independently, and additively along with Smad, regulates collagen transcription [59][60][61]. p38 MAPK signaling also increases the stability of the Col-1 mRNA expression [61]. The evidence supports fibrogenic activity by Smad3 and p38 MAPK in TGF-β1-induced hepatic fibrosis. In this study, we explored the effect of IH on TGF-β1/Smad3 and TGF-β1/p38 MAPK pathways. The results found that IH blocked the increase of the TGF-β1 expression during liver injury. Accordingly, the downstream molecules, p-Smad3 and p-p38, were downregulated with IH treatment. Therefore, IH treatment efficiently blocked the TGF-β1 pathway, which was involved in the antifibrosis activity of IH.
Autophagy maintains liver homeostasis by degrading and recycling aggregated proteins and damaged organelles [17], but inappropriate activation of autophagy has been associated with various diseases, including liver fibrosis [17,62]. Autophagy provides energy through digestion of intracellular lipids to support HSC activation and promote fibrosis [18,63]. Knockout of the autophagy-related Atg5 gene, or treatment with the autophagy inhibitor 3-methyladenine, reduces HSC activation [19,64]. Treating HSCs with 3methyladenine also inhibits HSC proliferation [62]. Autophagy thus appears to be required for both HSC activation and proliferation and may be a novel target for the therapy of liver fibrosis. Our previous work also shows that inhibiting autophagy plays an important role in attenuating liver fibrosis [13][14][15][16]. Lu et al. reported that IH protected the liver against acute concanavalin A-induced injury by inhibiting autophagy [20]. In our study, we found that IH inhibited autophagy in both liver fibrosis models. It is reported that TGF-β1 can upregulate autophagy [65] and inhibiting TGF-β1/Smad3 signaling can suppress autophagy [13,66]. Inhibition of the TGF-β1/Smad3 pathway by IH was confirmed in this study. Therefore, we concluded that IH reduced autophagy and alleviated liver fibrosis by downregulating the TGF-β1/Smad3 signaling pathway.
In summary, IH alleviated liver fibrosis induced by CCl 4 and BDL in mice (Figure 7). It inhibited the production of TGF-β1 secreted by Kupffer cells and activated HSCs. A lack of TGF-β1 downregulated downstream signaling by the Smad3 and p38 MAPK signaling pathways, contributing to the inhibition of HSC activation and ECM production. Autophagy was inhibited by downregulation of the TGF-β1/Smad3 signaling pathway, and the resulting energy deficit decreased HSC activation.
The study demonstrates the potential protective role of IH against both noncholestatic and cholestatic fibrosis. This is the first study to show that inhibiting autophagy in HSCs is involved in the antifibrotic mechanism of IH. It also highlights the key role of regulating autophagy and TGF-β1 pathways in the treatment of hepatic fibrosis. However, other mechanisms of action and the safety of IH for clinical applications remain to be investigated.

Conclusions
The study confirmed the protective effects of IH against liver fibrosis in mice. The mechanism of the antifibrotic effects of IH involves inhibition of TGF-β1-mediated Smad3 and p38 MAPK signaling pathways, thereby reducing autophagy and ECM formation.  Figure 7: Mechanisms of IH protection against liver fibrosis. IH inhibited production of TGF-β1, thereby downregulating the TGF-β1/Smad3 and TGF-β1/p38 MAPK pathways. Decreased TGF-β1 reduced the activation of downstream Smad3 and p38 MAPK signaling, which inhibited the transcription of α-SMA, collagen, and TIMP1 genes to suppress ECM production. Autophagy is reduced because of the inhibition of the TGF-β1/Smad3 signaling pathway. Complex, IH attenuated liver fibrosis in mice by inhibiting autophagy and ECM formation. Downregulation of the TGF-β1-mediated Smad3 and p38 MAPK pathways was involved.