Serpina3c Deciency Induced Necroptosis Promotes Non-Alcoholic Steatohepatitis Through β-Catenin/Foxo1/TLR4 Signaling

Background: Hepatocyte death and liver in ﬂ ammation have been recognized as central characteristics of nonalcoholic steatohepatitis (NASH); however, the underlying molecular mechanism remains elusive. The aim of this study is to determine the precise role of serpina3c in the progression of NASH. Methods: Male Apoe -/- /serpina3c -/- double knockout (DKO) and Apoe -/- mice were fed a high-fat diet (HFD) for 12 weeks to induce NASH. Several markers of steatosis and in ﬂ ammation were evaluated. In vitro cell models induced by palmitic acid (PA) treatment were used to evaluate the benecial effect of serpina3c on necroptosis and the underlying molecular mechanism. Results: Compared with Apoe -/- mice, DKO mice exhibited a signi ﬁ cantly exacerbated NASH phenotype that included hepatic steatosis, inammation, ﬁ brosis and liver damage, and increased hepatic triglyceride contents. We also indicated that the expression of the receptor-interacting protein 3 (RIP3) and phosphorylated mixed lineage kinase domain-like (MLKL) was increased in DKO mice. Our results found that serpina3c knockdown promoted necroptosis and lipid droplet formation under conditions of lipotoxicity in vitro. However, these phenomena were reversed by the overexpression of serpina3c. Mechanistically, downregulation of serpina3c expression promoted Foxo1 and β-catenin expression, and Foxo1 and β-catenin colocalized in the nucleus under conditions of lipotoxicity, consequently upregulating the expression of Toll-like receptor4 (TLR4). However, disruption of the Foxo1-β/catenin by Foxo1 and β-catenin inhibitors decreased TLR4 expression and ameliorated hepatic necroptosis in vitro. Conclusion: Serpina3c plays a protective role against the progression of NASH by inhibiting necroptosis. Serpina3c, expression. signicantly steatosis, brosis, exacerbated liver injury, manifested as hepatocyte ballooning and increased serum ALT and AST levels. In addition, the results showed that serpina3c plays a protective role against necroptosis. The possible underlying molecular mechanism is that serpina3c, as a Wnt inhibitor, suppresses necroptosis via β-catenin/Foxo1/TLR4 signaling. Taken together, our data revealed the importance of serpina3c in NASH and the underlying mechanism, which sheds light on potential strategies for using serpin in the treatment of NAFLD.


Introduction
Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease and comprises simple steatosis to more severe forms of disease, such as nonalcoholic steatohepatitis (NASH) [1]. It is now widely recognized that the mechanism underlying the pathogenesis of NASH is excessive lipid accumulation in hepatocytes that causes hepatocyte lipotoxicity and induces cell death. Hepatocyte death is accompanied by the massive recruitment of monocytes into the liver, and activation of macrophages leads to the release of many in ammatory factors and the occurrence of chronic in ammation [2]. Accumulating studies suggest that hepatic cell death plays critical roles in the promotion of liver fibrosis and progression of disease.
Multiple forms of cell death, including apoptosis, pyroptosis and necroptosis, are associated with steatohepatitis [3]. Receptor-interacting protein 3 (RIP3) is a molecular switch of cell necroptosis; the recruitment of RIP3 leads to the recruitment and phosphorylation of mixed lineage kinase domain-like protein (MLKL), eventually leading to perforated membrane structures that can eventually result in the rupture of cell membranes [4]. An early study showed that RIP3 expression is increased in the livers of patients with chronic liver disease and that liver necroptosis is increased in high-fat diet-induced experimental models of NASH [5]. Necroptosis can be regulated by Toll-like receptor 4 (TLR4). TLR4 can bind to TRIF receptor-domain-containing adaptor inducing interferon-β (IFN-β)) and then with receptorinteracting protein kinase 3 (RIPK3) to induce necroptosis [6].
Serpina3c, a serine proteinase inhibitor (serpin), is mainly expressed in the liver and adipose tissue.
Serpins, including serpina3k and serpina4, have been reported to be Wnt signaling inhibitors [7,8]. Our previous research showed that serpina3c plays a protective role in metabolic diseases [9]. The absence of serpina3c expression promotes the onset of atherosclerosis and pancreatic dysfunction. In addition, Choi Y et al reported that serpina3c is a critical factor involved in adipogenesis, and inhibition of serpina3c may result in bene cial effects in the treatment of obesity [10]. The liver is the main metabolic organ of the human body. NASH is a kind of metabolic disease and the hepatic manifestation of metabolic syndrome (MetS). Given the role of serpina3c in metabolic syndrome, we focused our analysis on the pathogenic role of serpina3c in NASH.
In this study, we aimed to investigate the role of serpina3c in NASH using a high-fat, Apoe -/mouse model, which is an appropriate model of NASH associated with MetS. Apoe -/mice develop spontaneous hypercholesterolemia and atherosclerosis and are widely used as a mouse model to study fatty liver disease [11]. This study rst indicated that the absence of serpina3c expression signi cantly increased HFD-induced liver damage, steatosis, fibrosis and macrophage infiltration and activation. Here, we also found that serpina3c negatively regulated hepatocyte necroptosis. Furthermore, we established the involvement of β-catenin/Foxo1/TLR4 signaling in serpina3c-induced necroptosis in hepatocytes.

Animals
Male Apoe -/mice, and Apoe -/-/serpina3c -/-(DKO) mice were used in this study. As our previous study [9], the generation of the serpina3c-knockout mice was commissioned by Beijing Biocytogen (Beijing Biocytogen Co., Ltd). Eight-week-old male DKO mice and their Apoe -/littermates (n=10~12/group) were fed a HFD (10% fat, 2% cholesterol, and 0.5% sodium cholate) for 12 weeks. The mice were maintained at room temperature with a 12-hour light/dark cycle and given free access to food and water.

Liver Function Examination
Levels of ALT, AST, and triglycerides, cholesterol was determined using the commerial kit (Jiancheng Bioengineering Institute, China) according to the manufacturer's instructions.

Liver histopathology
For histopathology, hepatic sections (8 μm) were used, and images were captured by bright-eld microscopy (Olympus BX51 microscope). Three liver sections, spaced at 40 μm in tervals, were stained per mouse, and three elds of view were captured per section, resulting in nine elds of view per mouse.
Cell culture and treatment AML12 hepatocytes (CRL-2254, ATCC) were cultured in DMEM/F12 with 10% FBS, supplemented with insulin, transferrin, selenium, dexamethasone and pen/strep. AML12 hepatocytes were exposed to 200 μM palmitic acid for up to 24 h to model lipotoxicity. (PA, P0500, Sigma) complexed to BSA or BSA alone as control for up to 24 h to model lipotoxicity. All lentiviruses were purchased from GenePharma (Shanghai, China): serpina3c-knockdown lentivirus LV-Mus serpina3c shRNA (LV-KD), serpina3coverexpression lentivirus LV-Mus serpina3c (LV-OV) and control group (LV-NC Immuno uorescence AML12 cells were treated with 0.3% Triton (Triton X-100, Dilution with PBS) for 15 min, blocked with 5% BSA at 4°C and then incubated overnight with primary anti-Foxo1 and anti-β-catenin antibodies at 4°C. Subsequently, the specimens were incubated with the corresponding rhodamine-or FITC-conjugated secondary antibody at 37°C for 1 hour, and the nuclei were stained with DAPI for 5 minutes at room temperature. Digital images were acquired with an Olympus FV3000 confocal laser scanning microscope (Tokyo, Japan) Quantitative RT-PCR (qRT-PCR) Analysis After being extracted using TRIzol reagent, total mRNA from the mouse liver and AML12 cells was collected and depurated using DNase. cDNAs were then obtained using a Transcript or First Strand cDNA synthesis kit (TaKaRa Bio, Japan). Quantitative real-time PCR was carried out on a StepOne Plus

Statistical Analysis
All data are expressed as the means ± SD from at least three independent experiments. Comparisons between groups were evaluated using a two-tailed Student's t test or one-way ANOVA with multiple comparisons. All statistical analyses were performed using SPSS, version 16.0. P values <0.05 were considered to be statistically signi cant.

Results
Serpina3c deficiency promoted liver injury and hepatic steatosis in HFD-fed Apoe -/mice In our previous study, the results showed that body weight, plasma total cholesterol levels, triglyceride levels, and plasma glucose levels were not signi cantly different in between HFD-fed Apoe -/and DKO mice [14]. However, circulating levels of alanine aminotransferase (ALT) ( Figure 1A) and aspartate aminotransferase (AST) ( Figure 1B) were signi cantly increased in DKO mice compared with Apoe -/control mice. The data indicate that serpina3c de ciency promoted liver injury. Among the markers of hyperlipidemia, liver triglycerides have been associated with NASH [15]. We found higher hepatic triglyceride levels in DKO mice than in Apoe -/control mice ( Figure 1C), which supports a role of serpina3c in hepatic steatosis. Consistent with this conclusion, H&E staining and Oil Red O staining of liver sections revealed that the livers of DKO mice exhibited signi cant hepatocyte ballooning and steatosis ( Figure  1D). The severity of NASH was assessed using the NASH Activity Score (NAS), an unweighted sum of scores for steatosis, lobular in ammation, and hepatocyte ballooning. Compared to Apoe -/mice, DKO mice exhibited increased NAS scores ( Figure 1E). To test the corresponding molecular mechanisms that accounted for the increased liver triglyceride content and lipid accumulation in DKO mice, we performed RT-PCR and analyzed the expression of liver lipid metabolism-associated genes. As expected, deficiency of serpina3c expression increased the liver expression of SREBP1c and CD36, although CD36 expression was not statistically signi cant. Serpina3c deficiency did not affect the liver expression of ACC1, FASN, or ACOX ( Figure 1F). Therefore, these results illustrated that serpina3c played an essential role in liver injury and hepatic steatosis during NASH development.
Serpina3c deficiency promoted hepatic in ammation response in HFD-fed Apoe -/mice Liver inflammation is one of the most important characteristics of NASH [16]. An increase in the inflammatory milieu in the liver is a hallmark of NASH. During the progression of liver injury, activated hepatic macrophages represent a major source of in ammatory mediators, including cytokines and chemokines, which are stimuli that sustain HSC activation and brogenesis. Thus, we performed immunohistochemistry for the macrophage marker CD68, and we observed that macrophage in ltration was signi cantly increased in DKO mice compared to Apoe -/mice ( Figure 2A). Given the effects of adhesion molecules on inflammatory cell infiltration, we assessed the expression of the adhesion molecule ICAM-1, and our results revealed signi cantly increased expression of the adhesion molecule ICAM-1 in DKO mice compared with Apoe -/mice ( Figure 2B). Moreover, consistent with macrophage in ltration in the liver of DKO mice, compared with Apoe -/mice, serpina3c-knockout mice exhibited significantly increased hepatic production of proinflammatory cytokines, including TNF-a, monocyte chemoattractant protein-1 (MCP-1), and IL-18, and reduced expression of arginase-1 (Arg-1) and IL-1β ( Figure 2C). As the JNK/NF-ĸB signaling pathway plays a pivotal role in the pathogenesis of steatohepatitis and the cytokines IL-18, TNF-α, and MCP-1 are associated with the activation of JNK/NF-ĸB signaling in steatohepatitis [17], the effects of serpina3c on NF-ĸB activation were investigated. JNK and p65 were extensively activated in DKO mice compared to Apoe -/mice, as shown by the increased phosphorylation of JNK and p65 by western blotting ( Figure 2D). These data suggested that serpina3c is involved in the hepatic recruitment of inflammatory cells in HFD-induced NASH, mainly by the activation of JNK/NF-ĸB signaling and the infiltration of macrophages.
Serpina3c deficiency promoted hepatic brosis in HFD-fed Apoe -/mice It is well documented that brosis is a salient feature of steatohepatitis. Transforming growth factor-beta1 (TGF-β1) is a cell factor that has been found to be involved in tissue brosis, and it plays an important role in NASH [18]. To investigate whether serpina3c is involved in hepatic fibrosis, hepatic mRNA levels of transforming growth factor-beta1 (TGF-β1) were measured by real-time PCR. The hepatic TGF-β1 mRNA levels were higher in DKO mice ( Figure 3A). Collagen deposition was markedly increased in the liver sections of DKO mice compared with those of Apoe -/mice, as shown by Sirius Red staining ( Figure 3B). In addition, immunohistochemistry staining also revealed increased α-SMA expression in DKO mice ( Figure 3C). These data suggested that serpina3c is required for the development of hepatic brosis.
The effect of serpina3c on the necroptosis in vivo and in vitro Multiple forms of cell death, including apoptosis, pyroptosis and necroptosis, are associated with NAFLD. The liver pathology results showed that hepatocyte death was appreciably increased in DKO mice. To further determine which cell death pathway is involved in hepatocellular death due to serpina3c deficiency, the protein expression pattern in the liver was analyzed by western blotting. Compared with Apoe -/mice, DKO mice exhibited a dramatic increase in the expression of RIP3 and p-MLKL ( Figure 4A), whereas the expression of cleaved-caspase 3, bcl-2 and GSDMD were unchanged between the two groups. The data indicated that necroptosis was the main mechanism of cell death in serpina3c-de cient cells. To further explore the effect of serpina3c on necroptosis in NAFLD, we treated AML12 cells with palmitic acid (PA) to establish an in vitro model of NAFLD after lentivirus transfection to overexpress and knock down serpina3c. First, AML12 cells were treated with PA (0, 100, 200, 300, 400 μM) for 24 h. We investigated the impact of PA on necroptosis in AML12 cells. As shown in Figure 4B, the hepatic TG content was markedly increased in AML12 cells treated with different concentrations of PA for 24 h in a concentration-dependent manner. Moreover, AML12 hepatocyte staining with propidium iodine (PI) and trypan blue showed dose-dependent increases in the size of the PI-positive and trypan blue-positive populations after treatment with PA ( Figure 4C and 4D), indicating an increase in necroptosis. Next, we also examined the expression of the necrosome protein complex. The expression of RIP3 and p-MLKL was increased in a concentration-dependent manner ( Figure 4E). As shown above, the hepatic TG content and necroptosis of AML-12 cells started to signi cantly increase in response to 200 μM PA; thus, PA at a concentration of 200 μM was used to treat the cells. AML12 cells were transfected with lentivirus to overexpress (LV-OV) or knockdown serpina3c (LV-KD), and negative control (LV-NC) lentivirus was used as a control. We found that LV-OV signi cantly increased the expression of serpina3c and LV-KD decreased the expression of serpina3c in AML12 cells ( Figure 4F). We found that the protein expression of RIP3 and phosphorylated MLKL was strongly upregulated after the knockdown of serpina3c in AML12 cells treated with PA and obviously downregulated by the overexpression of serpina3c ( Figure 4G). To determine whether necroptosis could be attenuated via a necroptosis inhibitor, we treated AML12 cells with necrosulfonamide (Nec) for 2 h before stimulation with PA. After 24 h, we measured changes in necroptosis. The results showed that the protein expression of RIP3 and phosphorylated MLKL induced by serpina3c knockdown in AML12 cells was attenuated by the necroptosis inhibitor ( Figure 4H). Knockdown of serpina3c resulted in a marked increase in the TG contents, as shown by Oil red O staining, and overexpression of serpina3c reduced the hepatic TG contents ( Figure 4I). Moreover, necroptosis inhibitors also decreased the hepatic TG contents induced by serpina3c knockdown. Next, we found that knockdown of serpina3c increased the size of the PI-positive and trypan blue-positive populations, overexpression of serpina3c signi cantly inhibited necroptosis, and necroptosis inhibitors also reversed the necroptosis induced by serpina3c knockdown. (Figure 4J and 4K). Collectively, these various results thus showed that the specific role of serpina3c in PA-induced hepatocyte necroptosis is involved in the pathogenesis of NASH, which in turn can be rescued by Nec.
The effect of serpina3c on β-catenin and Foxo1 expression and nuclear transfer We next sought to dissect the molecular mechanisms underlying the regulation of necroptosis by serpina3c. Several studies have reported that serpin, as a Wnt inhibitor, downregulated the levels of βcatenin [19]. The Wnt/β-catenin pathway has been proven to play a key role in the development of necroptosis [20]. To investigate the potential effects of serpina3c on the regulation of β-catenin signaling, the total protein and nuclear protein levels of β-catenin were measured. The total protein levels of βcatenin were signi cantly increased in DKO mice versus Apoe -/mice ( Figure 5A). Moreover, the total protein and nuclear protein levels of β-catenin were signi cantly upregulated in serpina3c-knockdown cells, and these effects were reversed by serpina3c overexpression (Figure 5B and 5C). The regulation of forkhead box o1 (Foxo1) by β-catenin is particularly important in reducing liver necroptosis [21]. β-catenin binds to T-cell factor (TCF) to activate profibrotic genes and binds to Foxo1 to promote necroptosis under oxidative stress. Perhaps not surprisingly, as a result of the upregulation of β-catenin expression, the total protein levels of Foxo1 were increased and Foxo1 phosphorylation was decreased in DKO mice versus Apoe -/mice in vivo and in LV-KD-transfected cells versus LV-NC-transfected cells in vitro; however, serpina3c overexpression reversed these effects in vitro ( Figure 5A and 5B). Moreover, the nuclear protein levels of Foxo1 were signi cantly upregulated in serpina3c-knockdown cells, and this effect was reversed by serpina3c overexpression ( Figure 5C). These data suggest that the absence of Serpina3c expression promotes Foxo1 and β-catenin expression in the nucleus. We then asked whether there is putative crosstalk between Foxo1 and β-catenin signaling in serpina3c-knockdown AML12 cells treated with PA. Immuno uorescence staining revealed increased nuclear Foxo1 and β-catenin expression in siRNA-serpina3c-transfected AML12 cells. Strikingly, both Foxo1 and β-catenin were colocalized in the nucleus ( Figure 5D). However, the immuno uorescence staining of Foxo1 and β-catenin was decreased after AML12 cells were pretreated with recombinant serpina3c protein ( Figure 5D). TLR4, a death receptor on the plasma membrane, is a downstream target gene of Foxo1. Studies have shown that increasing Foxo1 activity promotes innate TLR4-mediated inflammatory responses and tissue injury [22,23]. We then analyzed the levels of TLR4 in vivo and in vitro. Our results showed increased TLR4 protein expression when serpina3c expression was downregulated, but TLR4 protein expression was reduced by serpina3c overexpression in vitro ( Figure 5A and 5B). Taken together, these results indicated that serpina3c, as a WNT inhibitor, prevented β-catenin activity and Foxo1/TLR4 expression.

Serpina3c inhibits necroptosis via β-catenin/Foxo1/TLR4 signaling pathway
To further investigate the role of β-catenin/Foxo1 in the regulation of necroptosis by serpina3c in vitro, we treated AML12 cells with β-catenin and Foxo1 antagonists for 2 h before stimulation with PA. After 24 h, we observed changes in the activation of necroptosis. The results showed that compared with PA alone in the LV-KD group, inhibition of β-catenin with a β-catenin antagonist suppressed the protein expression of Foxo1, TLR4, RIP3 and p-MLKL in the LV-KD group ( Figure 6A). This result suggested that serpina3c inhibited Foxo1/TLR4 expression through β-catenin inhibition. Then, we used a Foxo1 antagonist and found that PA alone in the LV-KD group, the Foxo1 antagonist suppressed protein expression of TLR4, RIP3 and p-MLKL in the LV-KD group ( Figure 6B). Moreover, the number of the PI-positive and trypan bluepositive populations were signi cantly smaller in the LV-KD group treated with the β-catenin or Foxo1 antagonist than in the LV-KD group treated with PA alone ( Figure 6C, 6D). Oil red O staining showed that the β-catenin or Foxo1 antagonist reversed the increased hepatic TG content induced by serpina3c downregulation ( Figure 6E). Overall, serpina3c inhibited necroptosis via β-catenin/Foxo1/TLR4 signaling in vitro.

Discussion
In this study, we rst con rmed herein that serpina3c plays an important role in the progression of NASH.
Most importantly, the results demonstrated that serpina3c de ciency signi cantly promoted NASH pathogenesis, including hepatic steatosis, in ammation and liver brosis, as well as noticeably exacerbated liver injury, which manifested as hepatocyte ballooning and increased serum ALT and AST levels. In addition, the results showed that serpina3c plays a protective role against necroptosis. The possible underlying molecular mechanism is that serpina3c, as a Wnt inhibitor, suppresses necroptosis via β-catenin/Foxo1/TLR4 signaling. Taken together, our data revealed the importance of serpina3c in NASH and the underlying mechanism, which sheds light on potential strategies for using serpin in the treatment of NAFLD.
NASH is the result of advanced NAFLD liver injury and contributes to liver-related morbidity and mortality.
Many kinds of serpin, including serpina1 and serpina4, are mainly expressed in the livers of humans. The level of serpina1 was markedly reduced in patients with NASH, suggesting that serpina1 contributes to the pathogenesis of NASH [24]. Another study also demonstrated that the serpina4 gene expression levels were signi cantly downregulated in the livers from obese patients with NAFLD. These studies illustrate that serpin is associated with NAFLD, but currently, there is a lack of further research. In this study, our results showed that serpina3c knockout worsened steatosis, suggesting that serpina3c may play an important protective role in NAFLD. Our results also showed that the gene expression of SREBP-1c and CD36 was increased in DKO mice fed HFD, although CD36 gene expression was not statistically signi cant. These data suggested that serpina3c de ciency led to the fatty acid disorder in liver. The contribution of necroptosis to the pathogenesis of NASH is the subject of great interest. It is now widely recognized that hepatic lipid accumulation and increased free fatty acid (FFA) levels cause lipotoxicity to hepatocytes and induce cell death [25]. In our study, hepatocyte nuclear marginalization and fragmentation were observed by HE staining in serpina3c-de cient cells. These histological changes indicate that under conditions of hepatic lipotoxicity, serpina3c de ciency causes hepatocellular death.
Regarding the mechanism of cell death in NASH, our results showed that the expression of apoptotic marker proteins was unchanged. In addition to apoptosis, recent study has shown a worsening of ballooning and brosis in clinical trials of NASH using pan-caspase inhibitors, raising the possibility of a switch to a necroptotic form of cell death [26]. Necroptosis relies on the kinase cascade, which activation of RIPK3 promotes the formation of the "necrosome", a complex containing RIPK1, RIPK3, and MLKL. Several studies have confirmed that canonical RIP3-MLKL signaling plays a critical role in the pathogenesis of NASH [27]. In this study, we showed that serpina3c knockout promoted the expression of RIPK3 and p-MLKL. In vitro, serpina3c inhibits PA-induced necroptosis, suggesting that serpina3c is able to inhibit necroptosis under conditions of lipotoxicity. However, our results also show that RIP3 and p-MLKL expression was not different in hepatocytes with reduced serpina3c expression without PA treatment. Thus, it is likely that necroptosis does not occur under physiological conditions, suggesting that Serpina3c de ciently increases the susceptibility of the liver to lipotoxicity and promotes cell necroptosis.
There is strong evidence that hepatocyte cell death drives in ammation and brosis in NASH [28]. The JNK/NF-κB signal is one of the important pathways associated with the in ammatory response. The activation of this pathway can lead to the expression of a large number of in ammatory cytokines and then induce a series of in ammatory responses. Multiple cytokines, including TNF-α, IL-6, and IL-8, are involved in the development of NASH [29]. Here, we proved that serpina3c deficiency greatly promoted activation of the JNK/NF-κB signaling pathway and increased the expression of the in ammatory cytokines IL-18, TNF-a, and MCP-1, which are downstream target genes of NF-ĸB. In addition, serpina3c deficiency contributes to the hepatic expression of the adhesion molecule ICAM-1, and indicate that serpina3c deficiency increased in ammatory cell in ltration in the liver, which is consistent with our previous reports and other reports about the anti-in ammatory effect of serpin in various metabolic diseases [30]. Fibrosis is an important prognostic marker in human NASH. In ammatory cells in ltrating the damaged area can release various cytokines to activate cells that secrete a large amount of ECM. In this study, serpina3c de ciency presented pro brotic activity due to the expression of the master regulator of brosis initiation, TGF-β, and collagenous ber deposition also increased. In summary, Serpina3c de ciency promotes the in ammatory response by activating the JNK/NF-κB signaling pathway, resulting in liver fibrosis.
The Wnt/β-catenin signaling pathway is involved in the regulation of various cellular actions, including apoptosis, necrosis and in ammation [31]. A study reported that serpina3k, which belongs to the same serpina branch as serpina3c, binds to low-density lipoprotein receptor-like protein 6 (LRP6) and blocks its Wnt ligand-induced dimerization with the Fz receptor, leading to the downstream inhibition of β-catenin [32]. Activation of β-catenin signaling promotes endothelial cell programmed necrosis under conditions of hyperlipidemia [33]. In addition, activated β-catenin can enter the nucleus and bind to Foxo1 to activate Foxo1 and increase hepatocyte necroptosis [21]. Our data show that serpina3c knockdown activated β-catenin/Foxo1 and promoted necroptosis under conditions of lipotoxicity. β-catenin and Foxo1 inhibitors were used to further verify that serpina3c inhibited necroptosis through β-catenin/Foxo1. These observations corroborated the hypothesis that serpina3c ameliorates hepatic necroptosis via βcatenin/Foxo1 signaling. Furthermore, the present study reported that the Foxo1 pathway altered TLR4 expression, which plays an important role in necroptosis [34]. TLR4 is a downstream target gene of the transcription factor Foxo1 and contributes to the regulation of in ammation and necroptosis. Our study showed that the absence of serpina3c increased TLR4 expression by β-catenin/Foxo1. Taken together, our ndings demonstrate that serpina3c, as a Wnt inhibitor that mediates β-catenin/Foxo1/TLR4 signaling, is a key regulator of hepatocyte necroptosis in NASH.
Remarkably, the present study highlights the importance of serpina3c as a key regulator of hepatocyte necroptosis in NASH. We demonstrated, for the rst time, that in HFD-fed Apoe -/mice, serpina3c de ciency exacerbates NASH histopathology. We also showed that serpina3c inhibits hepatocyte necroptosis, a key characteristic of NASH. Moreover, our study provided evidence that ceserpina3c, a Wnt inhibitor, prevents hepatocyte necroptosis via β-catenin/Foxo1/TLR4 signaling. The ndings of the present study show that serpina3c may provide potential therapeutic strategies for NASH.

Declarations
Data availability All authors have access to the entirety of the data underlying this manuscript. Access to the data can be granted at any time upon reasonable request. (ORO) staining of liver sections after HFD feeding for 12 weeks (n=5/group). Scale bars, 100μm. E, Scores for steatosis, hepatocyte ballooning, and lobular in ammation were determined in three H&Estained hepatic sections per mouse, and were summed to generate NAFLD activity scores (NAS). F, Quantitative RT-PCR-assisted detection of CD36, SREBP1c, ACC, FASN, and ACOX in liver (n=5/group). All data represent the mean ± SD. *p < 0.05, vs. Apoe-/-mice group.

Figure 3
Serpina3c deficiency promoted hepatic brosis in HFD-fed DKO mice A, Quantitative RT-PCR-assisted detection of TGF-β in liver (n=5/group). B and C, Representative Sirius Red staining and α-SMA+ staining of liver sections after HFD feeding for 12 weeks (n=5/group). Scale bars, 100μm. All data represent the mean ± SD. *p < 0.05, vs. Apoe-/-mice group.  The effect of serpina3c on β-catenin and Foxo1 expression and nuclear transfer A, western blot assay analysis of β-catenin and Foxo1 expression in liver. Quantification of the blots normalized to the level of β-actin expression. (n=3/group). All data represent the mean ± SD. *p < 0.05, vs. Apoe-/-mice group. B western blot assay analysis of total protein β-catenin, Foxo1 and TLR4 expression in PA-stimulated hepatocyte AML12. Quantification of the blots normalized to the level of β-actin expression. C, western blot assay analysis of total protein and nuclear β-catenin and Foxo1 expression in PA-stimulated hepatocyte AML12. Quantification of the blots normalized to the level of β-actin expression. D, Immunofluorescence staining for hepatocyte AML12 Foxo1 (red) and β-catenin (green) colocalization in Page 20/21 the nucleus after transfection knockdown serpina3c siRNA(siRNA-serpina3c) and control siRNA (siRNA-NC) in PA stimulation with or without recombinant serpina3c protein (R-serpina3c) pretreatment. Scale bars, 10 μm. All data are expressed as mean ± SD from three independent experiment. *P<0.05, **P<0.01, vs. LV-NC. #P<0.05, ##P<0.01, vs. LV-NC+PA.