Kaempferol Protects Mice from D-GalN/LPS-induced Acute Liver Failure by Regulating the Autophagy Pathway

Kaempferol, a avonoid compound present in many edible plants, has been used in traditional medicine and has various biological functions. Acute liver failure (ALF) is a lethal clinical syndrome with severe liver function damage. There are currently no effective treatments for ALF except for liver transplantation. The aim of this study is to the mechanisms underlying the effect of in The ALF mouse model was established using D-galactosamine (D-GalN, 700 mg/kg)/lipopolysaccharide (LPS, 10 µg/kg). Two hours before the administration of D-GalN/LPS, different group of mice were pretreated according different doses of kaempferol, 6 hours after injection of D-GalN/LPS, and then killed. The survival rate, liver function and inammatory cytokine levels were assessed. It was determined whether kaempferol pretreatment protected hepatocytes from ALF induced by D-GalN/LPS via autophagy pathway in vivo and in vitro. by restraining the activity of pyruvate carboxylase and glucose-6 phosphatase, increase hepatic glucose metabolism and insulin resistance in obese mice induced by diet In our study, we revealed that pretreatment with kaempferol at different doses had different functional effects on D-GalN/LPS-induced ALF. Our results showed that 5 mg/kg of kaempferol pretreatment signicantly protected against liver injury induced by D-GalN/LPS in mice, while kaempferol at a high dose decreased the survival rate and induced more severe injury. As a result, when using kaempferol to treat ALF, it is important to pay attention to the different effects caused by different doses and choose the appropriate dose for treatment.

Acute liver failure (ALF) is a fatal hepatic disease associated with rapid loss of liver function resulting in multiorgan dysfunction, encephalopathy and coagulopathy in patients. This critical illness causes high mortality and morbidity. When it occurs, the survival of patients only depends on emergency liver transplantation [1]. Therefore, an effective drug to treat acute liver failure is urgently needed.
Kaempferol is a avonoid that mainly extracted from the root of Kaempferia galanga L., it is widely present in various natural plants, fruits, vegetables, beverages and teas [2,3]. Kaempferol possesses several pharmacological activities, including cardioprotective, neuroprotective, antioxidative, antidiabetic and anticarcinogenic properties [4]. A growing number of studies suggest that kaempferol can reduce the risk of developing various cardiovascular diseases, diabetes, cancer, and so on [5][6][7]. Studies have found that kaempferol is related to the treatment of many diseases. For instance, in a time-and concentrationdependent manner, kaempferol can decrease HeLa cervical cancer cell viability and induce apoptosis by remarkably restrain the PI3K/AKT pathway [8]. Studies also demonstrated that kaempferol distinctly inhibited bladder cancer EJ cells growth by inducing S cell cycle arrest and apoptosis and increased expression level of phosphorylated status of p53 by regulating mitochondria-mediated apoptotic signaling pathways [9].
Kaempferol are traditionally recognized as potential protective effects on liver injury [10][11][12] and was shown to be effective for anti-in ammatory properties in liver cells [13]. In addition, kaempferol has been demonstrated to signi cantly alleviate acute liver injury (ALI), in ammation and early hepatocyte apoptosis caused by propacetamol [14]. Our previous study showed that kaempferol pretreatment could alleviate liver damage in D-galactosamine (D-GalN)/lipopolysaccharide (LPS) -induced mice [15]. However, the underlying therapeutic mechanisms of kaempferol in ALF are still unclear.
Autophagy is an intracellular catabolic pathway with highly conserved where biomolecules and organelles can be degraded by the lysosomes. Autophagy is crucial for maintaining homeostasis of cell and replenish many types of substances for cell survival under stressful conditions [16]. Autophagy is closely associated with liver disease. Previous reports have revealed that autophagy may suppress the growth of tumors in chronic liver disease, and impaired autophagy can lead to a signi cant increase of glycolysis in liver cancer [17]. In addition, autophagy can protect against the accumulation of fat in hepatocytes during nonalcoholic fatty liver disease (NAFLD) [18]. In our recent study, we explored that autophagy activation protected mice from ALF by inhibiting glycogen synthase kinase 3β (GSK3β) activity [19]. In current therapeutic research, the problem of drug toxicity has always existed. There are few studies on the effects of different doses of kaempferol on ALF. Therefore, the toxicity of kaempferol requires further study.
Given the above information, in this study, A model of ALF induced by D-GalN/LPS-mouse is used and has been widely used to explore the underlying mechanisms involved in potential therapeutic drugs for clinical use in treating ALF [20,21], and the functional effects of different doses of kaempferol and related regulatory pathways were evaluated in the context of ALF. Our ndings demonstrated that a dose of kaempferol resulted in severe liver injury, whereas a low dose of kaempferol protected against ALF induced by D-GalN/LPS via regulating the autophagy pathway.

Animals and treatments
Male wild-type mice (C57BL/6, 8-12 weeks old) were purchased from Capital Medical University (CMU), keep them in the CMU animal facility under conditions free of speci c pathogens, and subjected to humane care guidelines as required by the CMU Animal Care Committee. The animal experiment protocol

Serum aminotransferase activity
Blood samples were collected from the mice at 6 h after D-GalN/LPS administration. Serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST), which are markers of hepatic damage, were measured using a multiparametric analyzer (AU 5400, Olympus, Japan) according to the manufacturer's protocol.

Histopathological analysis
According to a standard protocol, liver samples were xed in formalin and embedded in para n wax, and sections were stained using hematoxylin and eosin (H&E) for histopathological evaluation, which carried out under light microscopy.
Quantitative reverse-transcription polymerase chain reaction (qRT-PCR) Total RNA was isolated from liver tissue using TRIzol reagent following the protocol of manufacturer.

Western blotting
Liver tissue samples or cells were lysed in radioimmunoprecipitation assay (RIPA) buffer containing phosphatase and protease inhibitors. A total of 20 µg of protein in each sample was separated by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis at 80 V for 30 minutes and 120 V for 1 h, and then use the Bio-Rad blotting transfer system to electrotransfer to PVDF membrane (Bio-Rad, Hercules, CA). Monoclonal rabbit antibodies against β-actin, p-JNK, p-ERK, p-p38, LC3B, ATG7, and p62 were appropriately diluted (1:1,000) in 10 ml of blocking buffer at 4 ℃ overnight. After washing the membranes with Tris-buffered saline with Tween-20 (TBST), the appropriate horseradish peroxidaseconjugated secondary antibody (1:2000) in 10 ml of blocking buffer was added and incubated for 1 h at room temperature. Then the membranes were washed 3 times with TBST for 30 min. Next, An enhanced chemiluminescence commercial kit (Thermo Fisher Scienti c, Inc., Rockford, IL, USA) was used for detection of the proteins. Image J software (National Institutes of Health, New York, NY, USA) was used for quanti cation of the western blotting results.
Atg7 small interfering RNA treatment in vivo siRNA (3 mg/kg; Jima, Suzhou) and an Entranster ™ in vivo transfection reagent (Engreen Biosystem Co, Beijing) were used to knock out Atg7 by hydrodynamic tail vein injection in mice. Scrambled siRNA (3 mg/kg) was used as a control. These steps were performed in accordance with the procedures of manufacturer.

Isolation of primary mouse hepatocytes
Hanks' solution containing collagenase was used to perfusion of mouse livers when the mice were 7 weeks old, and as mentioned earlier, live hepatocytes are separated by Percoll isocratic centrifugation [22].

Starvation-induced autophagy in vitro.
The most robust condition for inducing autophagy is starvation. Primary hepatocytes were transfected with the GFP-LC3 plasmid for 12 h, and then incubated in Earle's balanced salt solution for different times to starve their amino acids. The percentage of cells were calculate with GFP-LC3 puncta in the different treatment groups. GFP-positive cells were regared as cells that showed bright, punctate staining. About 50 cells were counted, and the experiment was repeated at least three times.

Statistical analysis
Results from three independent experiments were presented as the mean ± SD. Differences between two groups were analyzed using unpaired Student's t-tests by using GraphPad Prism 7 Software. P of values < 0.05 were considered statistically signi cant.

Effects of kaempferol on D-GalN/LPS-induced ALF
First, we examined the effects of different doses of kaempferol on ALF induced by D-GalN/LPS in mice. Different doses of kaempferol (2.5 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg and 40 mg/kg) were intraperitoneally administrated. As shown in the survival analysis (Fig. 1a), at 5 h after D-GalN/LPS injection, mice in the 20 mg/kg kaempferol group began to die, their survival rate was 0% (0 of 20 mice). In addition, the survival rates of the 2.5 mg/kg kaempferol, 10 mg/kg kaempferol and D-GalN/LPS groups were 25% (5 of 20 mice), 30% (6 of 20 mice) and 20% (4 of 20 mice), respectively, at 24 h. In contrast, the survival rate of the 5 mg/kg kaempferol group stabilized at 80% (16 of 20 mice) at 24 h. Pretreatment with 20 mg/kg and 40 mg/kg kaempferol exhibited higher serum ALT (sALT) and serum AST (sAST) levels than group of D-GalN/LPS, but the rest of groups exhibited lower levels of sALT and sAST, and especially the 5 mg/kg kaempferol group exhibited the lowest levels (Fig. 1c). Consistent with the ALT and AST activities, the liver histopathology in the 20 mg/kg kaempferol group showed increased hepatocyte injury, similar to that of the D-GalN/LPS group, but mice in the groups of 2.5 mg/kg and 5 mg/kg kaempferol showed reduced hepatocyte injury, and the 5 mg/kg kaempferol group exhibited the lowest amount of injury (Fig. 1b). Moreover, we also explored the protective effects of kaempferol (5 mg/kg) pretreatment on liver injury in mice at 2 h, 4 h, 6 h and 8 h after D-GalN/LPS administration.
The results showed that D-GalN/LPS-induced serum ALT and AST levels signi cantly decreased by 5 mg/kg kaempferol at 4 h, 6 h and 8 h (Fig. 1d). These results demonstrate that high doses of kaempferol can induce more severe injury, while pretreatment with low doses of kaempferol signi cantly increase the survival rate and protect against ALF induced by D-GalN/LPS. Effects of kaempferol on liver in ammation in mice with ALF induced by D-GalN/LPS Because our previous study suggested that the liver in ammatory response plays an important role in ALF [23], we sought to explore whether different effects on liver in ammation were induced by kaempferol at different doses in D-GalN/LPS-induced ALF mice. As shown in Fig. 2a, compared with the D-GalN/LPS group, the mice in pretreatment groups with 2.5 mg/kg and 5 mg/kg kaempferol showed lower gene expression levels of TNF-α, interleukin (IL)-6, IL-1β, IL-12p40, C-X-C motif chemokine ligand (CXCL)-10 and CXCL-2 in the liver, and the 5 mg/kg kaempferol group exhibited the lowest levels. In contrast, pretreatment with 10 mg/kg and 20 mg/kg kaempferol resulted in notably increased genes expression levels of these cytokines. Additionally, the pretreatment groups of 2.5 mg/kg, 5 mg/kg and 10 mg/kg kaempferol showed higher gene expression levels of IL-10 than the group of D-GalN/LPS, among which the 5 mg/kg kaempferol group exhibited the highest expression. Pretreatment with kaempferol at 10 mg/kg and 20 mg/kg decreased the genes expression level of IL-10. Furthermore, we further con rmed whether the MAPK signaling pathway was affected by different doses of kaempferol. The protein expression levels of p-JNK, p-ERK and p-p38 were determined using Western blotting, and the groups of 2.5 mg/kg, 5 mg/kg and 10 mg/kg kaempferol exhibited reduced levels of these proteins, with the 5 mg/kg kaempferol group exhibiting the lowest levels. In contrast, pretreatment with 10 mg/kg and 20 mg/kg kaempferol resulted in increased levels (Fig. 2b). Therefore, these results con rmed that pretreatment with high doses of kaempferol can increase the hepatic in ammatory response and that pretreatment with low dose of kaempferol can signi cantly decrease the hepatic in ammatory response in ALF induced by D-GalN/LPS.

Effects of kaempferol on liver autophagy in mice with ALF induced by D-GalN/LPS
Given that our previous studies suggested that autophagy plays a signi cant role in ALF induced by D-GalN/LPS [19], we explored that whether different doses of kaempferol affected liver autophagy in D-GalN/LPS-induced ALF. Our data of qRT-PCR showed that (Fig. 3a), pretreatment with 2.5 mg/kg, 5 mg/kg, 10 mg/kg and 20 mg/kg kaempferol signi cantly promoted the gene expression of LC3 compared with D-GalN/LPS-induced ALF in mice. Moreover, pretreatment with 2.5 mg/kg, 5 mg/kg, and 20 mg/kg kaempferol also promoted the gene expression of Atg7, and the 5 mg/kg kaempferol group exhibited the highest expression. These data were consistent with western blot analyses. Pretreatment with 2.5 mg/kg, 5 mg/kg, 10 mg/kg and 20 mg/kg kaempferol resulted in increased protein expression of LC3, and the 5 mg/kg kaempferol group exhibited the highest expression. In addition, pretreatment with 2.5 mg/kg and 5 mg/kg kaempferol dramatically decreased the protein expression of p62, and 10 mg/kg and 20 mg/kg kaempferol signi cantly increased the protein expression of p62 (Fig. 3b). To further demonstrate that the effect of kaempferol (5 mg/kg) on autophagy ux, the fuse of lysosomes and autophagosomes was inhibited by CQ pretreatment. The CQ pretreatment did not further increased LC3II conversion and did not further degraded p62 compared to 5 mg/kg kaempferol and D-GalN/LPS in mice (Fig. 3c). These data suggest that kaempferol pretreatment maybe facilitates autophagosomes and inhibits autophagy ux in ALF mice induced by D-GalN/LPS.

Effects of kaempferol on starvation-induced autophagy in vitro
To further substantiate our results of the experiment in vitro, in starvation conditions, we evaluated the effects of kaempferol on primary hepatocytes. To observe the formation of autophagosomes, the GFP-LC3 plasmid was transfected into hepatocytes. Our data shown in Fig. 4a, compared to that of the control group, the GFP-LC3 signal was weak in the starvation group and high-dose kaempferol groups but was bright and punctate in the low-dose kaempferol groups. The protein expression levels of LC3, p62 and Atg7 were measured by western blot (Fig. 4b). The starvation and low-dose kaempferol groups showed the higher protein expression levels of LC3, and the high-dose kaempferol group showed lower expression levels compared with the control group. Furthermore, the low-dose kaempferol groups showed reduced protein expression level of p62.
The presence of autophagic ux was demonstrated by increased expression of LC3 and decreased expression of p62. Thus, our results suggest that pretreatment with low doses of kaempferol promotes induction of autophagic ux and that pretreatment with high doses of kaempferol restrains induction of autophagic ux in starvation-induced hepatocytes in vitro.

Kaempferol ameliorates injury in the livers of ALF mice through autophagy mechanisms.
Our previous studies con rmed that 5 mg/kg kaempferol had a signi cant effect on ALF induced by D-GalN/LPS. To further con rm whether pretreatment with 5 mg/kg kaempferol contributes to the induction of autophagy to protect against liver injury. Atg7 knockdown by siRNA was used in vivo. Our results revealed that 3-MA or Atg7 siRNA partially negated kaempferol-mediated hepatoprotection in ALF mice, in agreement with the levels of sALT and sAST in Fig. 5a, and the histology that relatively less wellpreserved liver architecture in Fig. 5b. Furthermore, we also measured the expression of LC3II and p62 when 3-MA pretreatment. The resulted showed that compared with mice treated with kaempferol (5 mg/kg) and D-GalN/LPS, the 3-MA pretreatment further decreased conversion of LC3II and further increased the degradation of p62 (Fig. 3c). These dates further indicated that pretreatment with 5 mg/kg kaempferol ameliorated liver injury by regulating autophagy in ALF mice.

Discussion
Kaempferol is an ingredient in traditional Chinese herbs and is found in various vegetables and fruits, including tomatoes, citrus, grapefruit, onion, broccoli, cabbage and apples. Kaempferol, a avonoid, has been shown to possess a broad spectrum of pharmacological activities, such as antidiabetic, antioxidative, cardioprotective, angiogenic and anticancer activities [24,25]. A previous study revealed that kaempferol can suppress liver gluconeogenesis by restraining the activity of pyruvate carboxylase and glucose-6 phosphatase, increase hepatic glucose metabolism and insulin resistance in obese mice induced by diet [26]. In our study, we revealed that pretreatment with kaempferol at different doses had different functional effects on D-GalN/LPS-induced ALF. Our results showed that 5 mg/kg of kaempferol pretreatment signi cantly protected against liver injury induced by D-GalN/LPS in mice, while kaempferol at a high dose decreased the survival rate and induced more severe injury. As a result, when using kaempferol to treat ALF, it is important to pay attention to the different effects caused by different doses and choose the appropriate dose for treatment.
ALF have a high mortality rate with clinical symptoms including coagulopathy, hepatic dysfunction and abnormal liver biochemical values. In addition, ALF is closely related to the in ammatory response and is an injury process associated with in ammation-mediated hepatocellular carcinoma. There is currently no effective treatment for ALF. Our previous research proved that endoplasmic reticulum stress can reduce in ammation by regulating the immune mechanism in ALF [27]. This study showed that 10 mg/kg and 20 mg/kg of kaempferol pretreatment increased expression of proin ammatory cytokines, and the expression of proin ammatory cytokines were remarkably decreased in pretreatment with 2.5 mg/kg and 5 mg/kg of kaempferol in ALF induced by D-GalN/LPS. Therefore, these data suggested that high-dose kaempferol pretreatment can promote the hepatic in ammatory response, and low-dose kaempferol pretreatment can signi cantly suppress the hepatic in ammatory response in ALF mice induced by D-GalN/LPS.
Autophagy is a self-digestive process that can maintain cell homeostasis, supply a variety of substrates for cellular energy generation and ensure cell survival under stressful conditions. Autophagy have a signi cance for the processes of cell death regulation in speci c tissues, such as the liver and brain [28]. Autophagy and in ammation are closely related [29]. Our previous research found that PPARα activation alleviates the in ammatory response by promoting autophagy in ALF mice induced by D-GalN/LPS [30].
Moreover, our data con rmed that inhibition of GSK3β activity increased PPARα expression and decreased the in ammatory response by further increasing autophagy [27]. Our study demonstrated that different doses of kaempferol had differential in uences on the induction of autophagy and autophagosome formation in vivo and in vitro. The presence of autophagic ux demonstrated by increasing LC3 expression combined with decreased p62 expression [31]. Our date indicated that lowdose of kaempferol upregulated genes related with autophagy, increased LC3II conversion and p62 degradation; while high-dose kaempferol decreased LC3II conversion and p62 degradation, and increased autophagosome formation; the pretreatment with CQ did not signi cantly change the effect of 5 mg/kg of kaempferol on the expression of LC3 and p62 in ALF mice induced by D-GalN/LPS. We therefore concluded that in ammatory response was alleviated by pretreating kaempferol with a low dose to protect mice from ALF induced by D-GalN/LPS via autophagy pathway. However, we also believe that the effect of kaempferol on the autophagic ux at different concentrations is a very complicated physiological phenomenon and deserves further to be explored in ALF.

Conclusions
In this study, our study suggested that effects of kaempferol on ALF at different doses have different functional by regulating the autophagy pathway, and kaempferol at a low dose can signi cantly protect mice from liver injury in ALF. Therefore, we should select the optimal dose when kaempferol is clinically used for the effective strategy to treat ALF. For further preclinical studies of autophagy agonists, it is necessary to develop clinically applicable therapeutic strategies for ALF. Abbreviations ALF, acute liver failure; ALI, acute liver injury; D-GalN, D-galactosamine; LPS, lipopolysaccharide; NAFLD, nonalcoholic fatty liver disease; 3-MA, 3-Methyladenine CQ, chloroquine; ALT, alanine aminotransferase; AST, aspartate aminotransferase; H&E, hematoxylin and eosin; qRT-PCR, quantitative reverse-transcription polymerase chain reaction; RIPA, radioimmunoprecipitation assay; TBST, Tris-buffered saline with Tween-20; sALT, serum ALT; sAST, serum AST; IL, interleukin; CXCL, C-X-C motif chemokine ligand.