Clitorin ameliorates western diet-induced hepatic steatosis by regulating lipogenesis and fatty acid oxidation in vivo and in vitro

Nonalcoholic fatty liver disease (NAFLD) is usually correlated with metabolic diseases, such as obesity, insulin resistance, and hyperglycemia. Herein, we investigated the inhibitory effects and underlying governing mechanism of clitorin in a western diet (WD)-induced hepatic steatosis mouse model, and in oleic acid-stimulated HepG2 cells. Male C57BL/6 mice were fed a normal diet, WD, WD + 10 or 20 mg/kg orlistat, and WD + 10 or 20 mg/kg clitorin. HepG2 cells were treated with 1 mM oleic acid to induce lipid accumulation with or without clitorin. Clitorin significantly alleviated body weight gain and hepatic steatosis features (NAFLD activity score, micro-, and macro-vesicular steatosis) in WD-induced hepatic steatosis mice. Additionally, clitorin significantly decreased protein expressions of sterol regulatory element-binding protein 1 (SREBP1), peroxisome proliferator-activated receptor γ (PPARγ), and CCAAT/enhancer binding protein α (C/EBPα) in WD-induced hepatic steatosis mice. Moreover, clitorin significantly diminished the mRNA levels of SREBP1, acetyl-CoA carboxylase (ACC), fatty acid synthase (FAS), and hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) and enhanced the mRNA levels of peroxisome proliferator-activated receptor α (PPARα) and carnitine palmitoyltranserase-1 (CTP-1), as well as adenosine monophosphate-activated protein kinase (AMPK) in the liver of WD-induced hepatic steatosis mice and oleic acid-stimulated HepG2 cells. Overall, our findings demonstrated that clitorin can be a potentially efficacious candidate for NAFLD management.

levels. The significantly increased serum levels of ALT and AST by HFD feeding were effectively decreased in the 20 mg/kg orlistat-, and all clitorin-administered groups (Fig. 3A,B). To further confirm the increase in liver weight and liver index in the WD-induced obese mice model, we conducted hematoxylin and eosin (H&E) staining to evaluate the histological changes in the liver. We observed hepatocytic lipid vacuoles and hepatocyte ballooning in the WD group compared to those in the CON group; however, the orlistat-or clitorin-administered groups remarkably reversed the corresponding features in the liver of WD-fed obese mice (Fig. 3C). In addition, the significantly increased NAFLD activity score (NAS), micro-, and macro-vesicular steatosis were displayed in the WD group, whereas NAS and microvesicular steatosis were improved after orlistat or clitorin administration (Fig. 3D,E); notably, the macrovesicular steatosis was significantly reduced in the only clitorin 20 mg/kg-administered group (Fig. 3F). Furthermore, the marked increases of intrahepatic triglyceride and total cholesterol levels in the WD group were all lower in the orlistat-or clitorin-administered groups (Fig. 3G,H). Altogether, these data indicated that WD-induced hepatic steatosis mouse model was completely established and clitorin administration demonstrated a protective effect against hepatic steatosis in mice challenged with WD.
Clitorin regulated adipogenesis, lipogenesis, and fatty acid oxidation in the liver of WD-induced hepatic steatosis mice. To investigate the mechanisms of clitorin on suppressing hepatic steatosis in the WD-induced hepatic steatosis mice, adipogenic and lipogenic transcriptional expression profiles were examined. In the liver, the protein expression levels of SREBP1, as well as PPARγ and C/EBPα, were higher in the WD group; however, this was rescued by orlistat and clitorin administration (Fig. 4A). qRT-PCR analysis revealed that the upregulated SREBP1, PPARγ, and C/EBPα protein expression levels in the livers of WD mice coincided with increases in SREBP1, PPARγ, and C/EBPα mRNA levels. Notably, the marked upregulation of the mRNA levels of SREBP1 and PPARγ was strongly suppressed in the liver of orlistat-or clitorin-administered mice com- www.nature.com/scientificreports/ pared to the corresponding expression profiles in the WD group (Fig. 4B,C). The mRNA level of C/EBPα was significantly down-regulated in the 10 mg/kg orlistat-or 20 mg/kg clitorin-administered mice; herein, 10 mg/kg orlistat-administered group had stronger inhibitory effect than the 20 mg/kg orlistat-administered group in C/ EBPα protein expression (Fig. 4A), and this tendency was also observed in the mRNA level of C/EBPα (Fig. 4D). We next assessed the impact of clitorin on the mRNA levels of lipogenesis (LXR, ACC , FAS, and HMGCR ) and fatty acid oxidation genes (CPT-1, PPARα, and AMPK), and it was revealed that the mRNA levels of LXR and HMGCR were significantly inhibited in the all orlistat-or clitorin-administered groups (Fig. 4E). In addition, the mRNA levels of ACC and FAS were significantly repressed in the 20 mg/kg orlistat-or clitorin-administered groups, showing higher reduction in the clitorin-administered group than those in the orlistat-administered group (Fig. 4E). In addition, CPT-1 mRNA level was significantly up-regulated in the 20 mg/kg orlistat-or clitorin-administered mice. PPARα and AMPK mRNA levels, which was eliminated in the WD group, were significantly augmented by 20 mg/kg orlistat-or all dose of clitorin-administered groups; additionally, these mRNA levels were higher after the clitorin administration than those in the orlistat-administered group (Fig. 4E). These data supported the concept that clitorin alleviates hepatic steatosis by regulating lipogenesis and fatty acid oxidation genes in the liver of WD-induced hepatic steatosis mice. A peak arose before clitorin was manghaslin, and two peaks arose after clitorin were rutin and nicotiflorin, consecutively. HPLC: high performance liquid chromatography.  The values are represented as the mean ± SD (n = 6 per group). # P < 0.05, ## P < 0.01, and ### P < 0.001 vs. CON group; *P < 0.05, **P < 0.01, and ***P < 0.001 vs. WD group; significance was determined using two-way ANOVA followed by a Bonferroni post hoc test, and one-way ANOVA followed by Dunnett's post hoc test. WD western diet, FER food efficiency ratio. www.nature.com/scientificreports/ ## P < 0.01 and ### P < 0.001 vs. CON group; *P < 0.05, **P < 0.01, and ***P < 0.001 vs. WD group; significance was determined using one-way ANOVA followed by Dunnett's post hoc test. WD western diet, ALT alanine aminotransferase, AST aspartate aminotransferase, H&E hematoxylin and eosin. www.nature.com/scientificreports/ Figure 4. Effect of clitorin on adipogenic, lipogenic, and fatty acid oxidation-related genes in the liver of WD-induced hepatic steatosis mouse model. The protein levels of (A) SREBP1, PPARγ, and C/EBPα were determined using western blot analysis. The cropped gel images are shown for clarity. Densitometric analysis was performed using ImageJ ver. 1.50i (https:// imagej. nih. gov/ ij/). The mRNA levels of (B) SREBP1, (C) PPARγ, (D) C/EBPα, and (E) LXR, ACC , FAS, HMGCR, CPT-1, PPARα, and AMPK were determined using qRT-PCR analysis. The values are represented as the mean ± SD (n = 6 per group). # P < 0.05, ## P < 0.01, and ### P < 0.001 vs. CON group; *P < 0.05, **P < 0.01, and ***P < 0.001 vs. WD group; significance was determined using one-way ANOVA followed by Dunnett's post hoc test. WD western diet, SREBP1 sterol regulatory element binding protein 1, PPARγ peroxisome proliferator activated receptor γ, C/EBPα CCAAT/enhancer binding protein α, LXR liver X receptor, ACC acetyl-CoA carboxylase, FAS fatty acid synthase, HMGCR 3-Hydroxy-3-Methylglutaryl-CoA Reductase, CPT-1 Carnitine palmitoyltransferase-1, PPARα peroxisome proliferator activated receptor α, AMPK adenosine monophosphate-activated protein kinase. www.nature.com/scientificreports/ www.nature.com/scientificreports/ in the oleic acid-stimulated HepG2 cells (Fig. 5D,E). Furthermore, the mRNA levels SREBP1, FAS, ACC , and HMGCR were higher in the oleic acid-treated group than those in the non-treated group; however, this was strongly reversed by 200 µM clitorin treatment (Fig. 5F). Moreover, CPT-1, PPARα, and AMPK mRNA levels, which was significantly down-regulated in the WD group, were effectively reversed by 200 µM clitorin treatment (Fig. 5F). These data indicated that clitorin attenuates oleic acid-induced hepatic lipid accumulation via control of the lipogenesis and the fatty acid oxidation.

Discussion
Immoderate exposure to a high-fat diet has been determined as a key attribute in an increasing number of NAFLD patients, which is demonstrated by the fact that the prevalence of NAFLD in obese patients is reported to be up to 90% 18 ; consequently, a high-fat diet is widely used to construct NAFLD animal models 19 . In the present study, clitorin administration significantly reduced the body weight gain in the WD-induced obese mice ( Fig. 2A,B). To evaluate whether the hepatic steatosis was induced in this model, we first investigated the liver weight and index (mg/body weight), and the data showed that WD feeding caused the remarkable increases in the corresponding parameters compared to those in the CON group; on the other hand, clitorin administration significantly reduced the liver weight and index in the WD-induced obese mice (Fig. 2G,H). It has been reported that orlistat effectively alleviates steatosis and may serve as a viable treatment option for NAFLD 20 , therefore, it was used as a positive control. Next, we confirmed that clitorin administration significantly rescued the markedly increased serum levels of ALT and AST in WD-fed obese mice (Fig. 3A,B). Even though NAFLD typically is characterized by mild increases of serum ALT and AST, the normal ALT and AST levels can be observed in up to 50% of NAFLD patients 21 . In addition, previous clinical review study suggested that relying on liver enzyme abnormalities is unhelpful in the diagnosis of NAFLD and examination of disease severity 22 . For assessment of hepatic steatosis, the NAS, which includes steatosis, lobular inflammation, and ballooning, has been used in numerous clinical trials and cross-sectional studies 23 . Therefore, we conducted the H&E staining to further track whether the hepatic histological changes were induced in this in vivo model. We found that the NAS, micro-, and macro-vesicular steatosis were all higher in the WD-fed mice than the normal diet-fed mice; moreover, the marked elevations in the hepatic triglyceride and total cholesterol contents were displayed in the WD group, indicating that the WD-induced hepatic steatosis mouse model was completely established. Furthermore, clitorin administration significantly recovered the NAS and microvesicular steatosis (Fig. 3D,E) as well as lipid profiling (Fig. 3G,H) compared to those in the WD group. Notably, our data revealed that clitorin had stronger effect in suppressing macrovesicular steatosis than the orlistat (Fig. 3F). The hepatic effect of PPARγ appears to be steatogenic; hepatocyte-specific PPARγ knockout mice showed a remarkable decrease in the number of hepatic lipid vacuoles, as well as downregulation of de novo lipogenesis activators 24 . Conversely, PPARγ overexpression in the liver induced by HFD feeding leads to lipid accumulation, which is the initiation step in the development of NAFLD 7 . CCAAT/enhancer binding proteins (C/EBPs), including C/EBPα and SREBP1, are also considered key regulators of adipogenesis. SREBP1 plays an important role in the regulation of de novo lipogenesis in the liver 25 . SREBP1c levels are enhanced in the fatty livers of obese, insulin-resistant, and hyperinsulinemic ob/ob mice 10 . In addition, SREBP1c expression is also elevated in patients with NAFLD; additionally, in concordance with its lipogenic role, hepatic triglyceride levels are higher in SREBP1c-overexpressing transgenic mice 26 . Thus, SREBP1, PPARγ, and C/EBPα are crucial transcription factors that upregulate the expression of genes modulating fat accumulation in the liver. In this study, the hepatic mRNA and protein levels of SREBP1, PPARγ, and C/EBPα were significantly reduced after clitorin administration in a dose-dependent manner (Fig. 4A-D). However, both protein and mRNA levels of C/EBPα were effectively abolished in the 10 mg/kg orlistat-administered mice than those in the 20 mg/kg orlistat-administered mice. Additionally, in the orlistat-administrated groups, the mRNA levels of SREBP1 and PPARγ were not consistent with the protein expressions; 10 mg/kg orlistat-administered group had higher inhibitory effects in the mRNA levels of SREBP1 and PPARγ than those in the 20 mg/kg orlistat-administered group. This probably seems to have relatively limitation in our study, with only 6 mice in each group examined in both protein and mRNA expression; therefore, further studies (e.g. increase 'n' per a group) would be examined to prove these differences.
LXRs are involved in hepatic lipogenesis via direct regulation of SREBP1c 27 , which positively modulates ACC expression 28 . ACC catalyzes a key rate-limiting step in fatty acid biosynthesis, and is also associated with the control of fatty acid oxidation by the synthesis of malonyl-CoA, an inhibitor of CPT-1 6 . Indeed, inhibition of the liver-specific isoform ACC1 in mice ameliorated hepatic triglyceride levels in mice by simultaneously suppressing fatty acid biosynthesis and augmenting fatty acid beta oxidation in the liver 6 . CPT-1 leads to beta-oxidation, as it allows fatty acids to reach the mitochondrial matrix 11 . CPT-1 is also linked to PPARα expression. Among the three PPAR isotypes, PPARα, PPARβ/δ, and PPARγ, PPARα is the most abundant isotype in hepatocytes and is related to numerous aspects of lipid metabolism 29 and high fatty acid oxidation rates 30 . Ineffective PPARα sensing leads to diminished energy burning, resulting in hepatic steatosis and steatohepatitis 31 ; thus, it is inferred that these genes can potentially prevent NAFLD. AMPK, a major energy sensor of the cell, downregulates ACC activity to suppress lipid biosynthesis 32 . In addition, AMPK regulates hepatic and adipose lipid metabolism by modulating lipogenesis, lipolysis, gluconeogenesis, and adipogenesis. AMPK inhibits de novo lipogenesis by downregulating PPARγ, C/EBPα, and SREBP1; furthermore, it promotes fatty acid oxidation by upregulating CPT-1a 33 . Our results showed that clitorin administration significantly decreased the mRNA levels of LXR, ACC , FAS, and HMGCR , which are lipogenic genes, and it also enhanced the mRNA levels of PPARα and CTP-1, fatty acid oxidation genes, as well as AMPK mRNA levels in the livers of WD-induced hepatic steatosis mice (Fig. 4E). Overall, clitorin had similar or excellent inhibitory effects compared to orlistat used as a positive control in the WD-induced hepatic steatosis mice. www.nature.com/scientificreports/ Oleic acid-stimulated HepG2 cells have been widely used to evaluate NAFLD in vitro [34][35][36][37][38][39] . Consistent with in vivo experiments, our results showed that clitorin treatment significantly diminished lipid accumulation by reducing the lipogenic genes (SREBP1, ACC , FAS, and HMGCR ) and enhancing the fatty acid oxidation genes (PPARα, CPT-1, and AMPK) in oleic acid-stimulated HepG2 cells (Fig. 5). Unexpectedly, there were significantly increases on the cell viability in 6.25-50 µM clitorin-treated cells. This issue should be considered in additional research to study the impact of clitorin in HepG2 cells, but we clearly demonstrated that 200 µM clitorin block the hepatic lipid accumulation by controlling the lipogenesis and fatty acid oxidation in oleic acid-stimulated HepG2 cells.
In this study, four flavonoids, including manghaslin, clitorin, rutin, and nicotiflorin, were identified in papaya leaf. Among the four flavonoids, previous study reported that a recommended dose of rutin is 250-500 mg twice per day 40,41 . Compare to rutin, 20 mg/kg clitorin, which equates to a 97 mg for a 60 kg person, seems not to be an excessive dose for daily intake. However, it is necessary to evaluate clitorin content in papaya plant to determine whether 97 mg clitorin (for human) can be taken through diet such as papaya plant or can be considered as a food supplement. Although there is evidence on clitorin content in freeze-dried papaya leaf juice 42 , it is not sufficient to figure out its quantity. Additionally, to decide the exact human dose, further study on the toxicity of clitorin must be needed. This study contributes new knowledge to the sparse literature on clitorin, which would help specific areas for future research including determination of clinical dose. Next, we are going to analyze the mechanisms of clitorin by inhibiting/silencing AMPK in vivo and in vitro model, to fully understand its action on regulation of lipolysis and lipogenesis in NAFLD. Overall, our results showed that clitorin alleviated hepatic steatosis by reducing both adipogenesis and lipogenesis, and enhancing fatty acid oxidation (Fig. 6). The present study is the first to report on the positive impact of clitorin on hepatic steatosis and our findings provide basic data, which lead to deeper understanding of the pharmacological effects of clitorin on the potential improvement of NAFLD. Experimental animal care protocols and treatment cycles. Six-week-old male C57BL/6J mice were procured from Daehan Biolink (Daejeon, Republic of Korea). The mice were maintained under conditions of controlled temperature (22 ± 2 °C) and humidity (55 ± 9%), with a 12-h light/dark cycle. After a week of adjustment, the mice were fed 45% WD for 7 weeks, except for the normal diet group (CON). After 7 weeks, the mice were randomly divided into five groups of six mice each: WD group, WD + treatment group with 10 or 20 mg/ kg orlistat as a positive control, and WD + treatment group with 10 or 20 mg/kg clitorin. Orlistat and clitorin were dissolved in 1:1:18 ratio of ethanol, cremophor, and distilled water and orally administered to the mice once daily for 4 weeks. Mice in the CON and WD groups were administered vehicle. The mice were allowed free access to water and food, and their body weight and food intake were measured every week. Food efficiency ratio (FER) was calculated by applying the equation: FER = (body weight gain (g)/food intake (g)) × 100. At the end of the experiment, mice were anesthetized with Zoletil 50 (20 mg/kg) by i.p. injection according to the manufacturer's instructions, and then the mice were euthanized by cervical dislocation. The livers of the mice were excised, cleaned with phosphate-buffered saline (PBS), weighed, and directly stored at − 80 °C. All experiments were performed under the Ethical Committee for Animal Care and the Use of Laboratory Animals, Sangji University (approval document no. 2017-22). Histological analysis. The liver tissues from the mice in each group were fixed in 10% formalin, embedded in paraffin, and cut into 8 µm sections. Certain sections were stained with H&E for histological examination. Stained liver sections were observed for the evaluation of hepatic steatosis using an Olympus SZX10 microscope (Olympus, Tokyo, Japan). NAFLD development was examined using NAS, which includes a numerical score for steatosis (0-3), hepatocyte ballooning (0-2), and lobular inflammation (0-3). Steatosis were determined at 200 × magnification and quantified as macrovesicular and microvesicular steatosis using ImageJ ver. 1.50i (https:// imagej. nih. gov/ ij/). Then, the percentage of steatotic cells was graded as follows: (1)   Quantitative reverse-transcription polymerase chain reaction (qRT-PCR) analysis. qRT-PCR analysis was performed as previously described 44 . Briefly, each liver tissue (50 mg) were homogenized, and total RNA was isolated using the 1 mL Easy-Blue® reagent according to the manufacturer's instructions (Intron Biotechnology; Seongnam, Republic of Korea). Total RNA (1 µg) was converted to cDNA using a high-capacity cDNA reverse transcription kit (Applied Biosystems; Foster City, CA, USA). For qPCR reactions, the 10 ng of cDNA was amplified and measured using SYBR® Master Mix (Applied Biosystems; Foster City, CA, USA). Gene expression was determined using the comparative threshold cycle method. GAPDH was used as an internal control. Sequences of mouse oligonucleotide primers are presented in Table 1.

Serum analysis.
Cell culture and treatment.  www.nature.com/scientificreports/ Oil red O staining of HepG2 cells. After stimulation with oleic acid, the cells were washed with PBS and fixed with 10% formaldehyde in PBS at 25 °C for 1 h. Cells were then washed thrice with distilled water and stained with Oil Red O working solution (3 mg/mL in 60% isopropanol) at 25 °C for 2 h. The cells were rinsed thrice with distilled water and photographed using an Olympus SZX10 microscope. Next, the Oil Red O dye was eluted with isopropanol to determine the intracellular lipid content and was measured using an Epoch® microvolume spectrophotometer at 520 nm.
Statistical analysis. Data are expressed as the mean ± standard deviation (SD) of triplicate experiments.
Statistically significant values were compared using ANOVA and Dunnett's post hoc test, and p-values < 0.05 were considered statistically significant. Statistical analysis was performed using SPSS statistical analysis software (version 19.0, IBM SPSS, Armonk, NY, USA).

Data availability
The datasets used and/or analyzed in this study are available from the corresponding authors on reasonable request.