Garcinia cambogia Ameliorates Non-Alcoholic Fatty Liver Disease by Inhibiting Oxidative Stress-Mediated Steatosis and Apoptosis through NRF2-ARE Activation

Excessive free fatty acids (FFAs) causes reactive oxygen species (ROS) generation and non-alcoholic fatty liver disease (NAFLD) development. Garcinia cambogia (G. cambogia) is used as an anti-obesity supplement, and its protective potential against NAFLD has been investigated. This study aims to present the therapeutic effects of G. cambogia on NAFLD and reveal underlying mechanisms. High-fat diet (HFD)-fed mice were administered G. cambogia for eight weeks, and steatosis, apoptosis, and biochemical parameters were examined in vivo. FFA-induced HepG2 cells were treated with G. cambogia, and lipid accumulation, apoptosis, ROS level, and signal alterations were examined. The results showed that G. cambogia inhibited HFD-induced steatosis and apoptosis and abrogated abnormalities in serum chemistry. G. cambogia increased in NRF2 nuclear expression and activated antioxidant responsive element (ARE), causing induction of antioxidant gene expression. NRF2 activation inhibited FFA-induced ROS production, which suppressed lipogenic transcription factors, C/EBPα and PPARγ. Moreover, the ability of G. cambogia to inhibit ROS production suppressed apoptosis by normalizing the Bcl-2/BAX ratio and PARP cleavage. Lastly, these therapeutic effects of G. cambogia were due to hydroxycitric acid (HCA). These findings provide new insight into the mechanism by which G. cambogia regulates NAFLD progression.


Introduction
Obesity is a chronic disease resulting from an energy imbalance caused by hypercaloric food intake and insufficient energy expenditure [1] and is strongly associated with type 2 diabetes, hyperlipidemia, and non-alcoholic fatty liver disease (NAFLD) [2]. NAFLD is the most common chronic liver disease in western societies and is well recognized as the hepatic manifestation of metabolic disease [3]. NAFLD is characterized by excessive fat accumulation in hepatocytes and is strongly associated with obesity and energy imbalance [4]. If this pathological state is not properly treated, it can progress to non-alcoholic steatohepatitis (NASH) and continue to progress to cirrhosis and ultimately to hepatocellular carcinoma [5].
NAFLD is a various factorial disease associated with genetic, epigenetic, and environmental factors [6]. Although its pathogenesis is still not entirely clear, the "two-hit hypothesis" proposes an explanation for the development of NAFLD [7]. The first hit results from insulin resistance, which causes an imbalance between hepatic lipid influx and clearance [8]. The second hit involves the perturbation of redox homeostasis, causing reactive oxygen species (ROS) generation. Redox imbalance leads to hepatic mitochondrial dysfunction and inflammatory and oxidative processes [9]. An updated theory is Table 1. Body weight and biochemical markers of animal models (n = 7 per group).

Dosage Information
The dosages of G. cambogia were determined by consideration of the equivalent conversion of human dose [34] for a mouse model [35] and several studies, including an animal model [36,37]. The dosages of G. cambogia used for weight loss range from 1667 to 4668 mg (taken in divided doses) daily for 60 kg humans [34], and the G. cambogia dose for a mouse model based on a conversion of these dose regiment ranges from 307.5 to 947.1 mg/kg daily [35]. Considering dosages of other studies (200-250 mg/kg) [36,37], low dose (200 mg/kg) and high dose (400 mg/kg) were determined. In addition, the dosage of orlistat was determined similar to G. cambogia. The dosages of orlistat used for weight loss range from 180 to 360 mg (taken in divided doses) daily for 60 kg humans [38], and the orlistat dose for a mouse model based on a conversion of these dose regiments ranges from 32.2 to 64.4 mg/kg daily. Considering dosages of other studies (10 mg/kg) [39], orlistat dosage (20 mg/kg) was determined.

Histological Analysis
At the end of the experimental period, all animals were fasted overnight and anesthetized by CO 2 administration. After the mice were euthanized by cervical dislocation, liver, heart, spleen, lung, and kidney tissues were collected, weighed, rinsed, fixed with 4% paraformaldehyde, and embedded in paraffin. The tissues were cut into 4-µm-thick sections. The sections were stained with hematoxylin and eosin (H&E) for histological examination, and images were acquired using a microscope (Nikon Eclipse Ti, Nikon Instruments Inc., Tokyo, Japan). The scale bars in the images represent 50 µm. The degree of steatosis was quantified as described previously [40]. Briefly, macrovesicular steatosis (vacuoles displaced the nucleus to the side) and microvesicular steatosis (vacuoles not displaced around the nucleus) were both separately scored, and the severity was graded based on the percentage of the total area affected. The scores were: 0 (<5), 1 (5-33%), 2 (33-66%) and 3 (66%) [41]. Five fields were analyzed per animal, and an average score of each animal was considered for statistical analysis. In addition, the liver index was calculated as follows: Liver index = liver weight (g)/body weight (g) In vivo studies were performed in a blinded fashion at the histological analysis stage. For the quantification of tissue staining and pathological evaluation of animal specimens, the investigator was blinded to ensure an unbiased interpretation of the results.

Serum Analysis
At the end of the experimental period, blood samples were collected from the abdominal aorta and centrifuged at 6000 rpm for 20 min to separate the serum. The levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), triglycerides, and total cholesterol were analyzed using a DRI-CHEM 7000i (Fujifilm, Tokyo, Japan). The malondialdehyde (MDA) content was determined using an OxiSelect™ TBARS Assay Kit (MDA Quantitation) (Cell Biolabs Inc., San Diego, CA, USA) following the manufacturer's protocol. The levels of MDA were measured at 532 nm using a microplate reader (Tecan Group Ltd., Männedorf, Switzerland).

Cell Culture
Human hepatoblastoma HepG2 cells were obtained from the American Type Culture Collection (Cat. No. ATCC HB-8065, Manassas, VA, USA) and grown in DMEM supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 100 IU/mL penicillin, and 100 µg/mL streptomycin at 37 • C in a humidified incubator under 95% air and 5% CO 2 . Cells were seeded on different types of plates and serum-starved for 24 h in a serum-free medium before experimental treatment. G. cambogia extract (dissolved in DW) and other reagents (dissolved in DMSO to a final DMSO concentration in the medium of ≤0.2%) were added for the indicated time periods. HepG2 cells were used at passages 4-7.

Free Fatty Acids (FFAs) Solution Preparation
The FFAs solution was prepared by previously described methods [42]. Briefly, the FFA-containing stock solution was prepared by conjugation of oleate (0.67 mM) and palmitate (0.33 mM) with 1% FFA-free BSA in a serum-free medium (final FFA concentration, 1 mM). After 24 h of incubation with FFA-containing medium and experimental treatments, the extent of steatosis, apoptosis, ROS levels, and protein expression levels was evaluated.

MTT Assay
Cell viability was assessed using the MTT assay. HepG2 cells were seeded on 96-well plates and incubated in a serum-free medium. After 24 h, the G. cambogia, other reagents, and 100 µg/mL digitonin (used as a positive control for cytotoxicity [43]) were added for 24 h. After the reaction was terminated, 5 mg/mL of MTT reagent was added, and the Antioxidants 2021, 10, 1226 5 of 20 cells were incubated at 37 • C under humidified 5% CO 2 . After 3 h, dimethyl sulfoxide was added to dissolve any formed crystals. The absorbance was measured at 565 nm using a microplate reader (Tecan Group Ltd.). All results are expressed as fold changes relative to the control (no stimulation) value.

Oil Red O Staining
Lipid droplets in cells were examined using oil red O staining after the reaction was terminated. HepG2 cells were fixed using 4% formaldehyde for 1 h and washed with 60% isopropanol. Then, the fixed cells were stained with filtered oil red O solution in 60% isopropanol for 10 min. After the oil red O solution was removed, the stained cells were washed with distilled water. Images of lipid droplets in the stained cells were acquired using an Olympus IX71 digital microscope (Olympus, Tokyo, Japan). The stained lipid droplets were dissolved in 100% isopropanol and quantified by measuring the absorbance at 520 nm using a microplate reader (Tecan Group Ltd.).

Quantitative Real-Time PCR (qPCR)
Total RNAs in cells and tissues were isolated using TRIzol™ (Invitrogen) according to the manufacturer's instructions. Reverse transcription was performed using AccuPower CycleScript RT Premix (Bioneer, Daejeon, Korea) and 1 µg RNA in a thermocycler. Realtime PCR amplification was performed using 96-well optical plates, with TOPreal™ qPCR 2X PreMIX (SYBR Green with low ROX) (Enzynomics, Daejeon, Korea), 400 nM each of forward and reverse primer and 1 µL of cDNA and H 2 O in a 20 µL reaction volume. Real-time fluorescence of PCR products was detected using CFX96 Real-Time Detection System (Bio-Rad, Hercules, CA, USA) at following thermocycling conditions: 1 cycle of 95 • C for 3 min; 39 cycles of 95 • C for 10 s, and 60 • C for 30 s; 1 cycle of 95 • C for 10 s and 65 • C for 5 s followed by 0.5 • C increments at 5 s/step back to 95 • C. Only primer pairs leading to the synthesis of a single fragment with the appropriate size were used in this study. Primer sets used for this study were listed in Table S1. The relative gene expression was calculated using a 2 −∆∆ct method, which was normalized by β-actin expression level.

TUNEL Staining
A Click-iT™ Plus TUNEL Assay for in situ apoptosis detection (Invitrogen, Carlsbad, CA, USA) was used for TUNEL staining, which was performed using sections from liver tissues. As positive controls, sections were pre-treated with DNase I (30 U). Images were acquired using a confocal laser microscope (K1-Fluo, Nanoscope Systems, Daejeon, Korea) using K1-image software (Nanoscope Systems, Daejeon, Korea). The number of TUNEL-positive cells was quantified using ImageJ software (Version 1.53c, NIH, Bethesda, MD, USA).

Apoptosis Detection Assay
Apoptosis-mediated cell death was analyzed by annexin V (AV) and propidium iodide (PI) staining using a FITC Annexin V Apoptosis Detection Kit I (BD Pharmingen™, San Jose, CA, USA) according to the manufacturer's guidelines. Briefly, cells were collected by trypsinization and centrifuged at 5000 rpm for 3 min. The pellets were washed carefully with PBS and centrifuged. Then the cells were stained with AV-FITC/PI solution. The extent of AV-FITC and PI staining was measured using FACS Canto II and BD FACS Diva software. AV − PI − (viable), AV + PI − (early apoptotic), AV + PI + (late apoptotic), and AV − PI + (necrotic) cells were analyzed using FlowJo software (Version 10, FlowJo LLC, Ashland, OR, USA), and the data are summarized in one graph.

Caspase-3 Activity
The in vitro and in vivo caspase-3 activity assay was carried out in accordance with the instructions provided by the Caspase-3 Assay Kit (Abcam, MA, USA). Briefly, lysates of cells or liver tissues were collected, and reaction buffer with dithiothreitol (DTT, 10 mM) was added. Then, the caspase-3 substrate in the kit (DEVD-p-NA, 200 µM) was treated for 60 min at 37 • C. The absorbance was measured at 405 nm using a microplate reader (Tecan Group Ltd.). All results are expressed as fold changes relative to the control (no stimulation) value.

Determination of Intracellular ROS Level
HepG2 cells were seeded into black 96-well plates. Cells incubated with FFAs were cotreated with G. cambogia and N-acetyl-L-cysteine (NAC, 4 mM, positive control for antioxidant activity [44]). After the reaction was terminated, the medium was removed, and H 2 DCFDA (20 µM) diluted in serum-free medium was added to the cells and incubated for 30 min at 37 • C. The cells were washed and filled with PBS, and the fluorescence intensity (excitation = 485 nm; emission = 530 nm) was measured using a fluorescence reader (Infinite F200, Tecan Group Ltd.). All results are expressed as fold changes relative to the control (no stimulation) value.

Cell Fractionation
Cell fractionation was performed as described previously [45]. Briefly, HepG2 cells were treated with FFA and G. cambogia for 12 h. After the reaction was terminated, cells were centrifuged at 800× g for 10 min and homogenized in buffer 1 (250 mM sucrose, 50 mM Tris-HCl, pH 7.4, 5 mM MgCl 2 , 2 mM EDTA, 5 mM NaF, 1 mM PMSF, and 1 mM sodium orthovanadate) and kept on ice for 30 min. Then, the homogenate was mixed by vortexing for 15 s and centrifuged at 800× g for 15 min at 4 • C to obtain supernatant (S0) and pellet (P0). The pellet (P0) was mixed with buffer 1 by vortexing for 15 s and centrifuged at 500× g for 15 min at 4 • C. The obtained pellet was resuspended in buffer 2 (20 mM HEPES, pH 7.9, 1.5 mM MgCl 2 , 0.5 M NaCl, 20% glycerol, 1% Triton X-100, 2 mM EDTA, 5 mM NaF, 1 mM PMSF and 1 mM sodium orthovanadate) by vortexing for 15 s and kept on ice for 30 min. Then, the homogenate was placed on ice in a Vibra-Cell sonicator (Sonics, Newtown, CT, USA) for 8 × 5 s bursts with 10 s intervals. The homogenate was centrifuged at 9000× g for 30 min at 4 • C, and the supernatant was resuspended in 5× SDS loading buffer to generate the nuclear fraction. Western blot analysis was performed to identify nuclear fraction proteins. Lamin A/C was used as a loading control for the nucleus.

Immunofluorescence
HepG2 cells were seeded with coverslips and placed in 24-well plates. Then, HepG2 cells were treated with FFA and G. cambogia for 12 h. After the reaction was completed, cells were fixed with 4% formaldehyde and permeabilized with 0.25% Triton X-100 in PBS and 1% BSA for 5 min. Then cells were blocked with 5% goat serum in PBS. After 1 h, cells were incubated with NRF2 primary antibodies for 16 h at 4 • C and subsequently treated with secondary antibodies in PBS with 3% BSA for 2.5 h at room temperature. Then cell nuclei were stained by 4 ,6-diamidino-2-phenylindole (DAPI), and observed under a confocal laser microscope (K1-Fluo, Nanoscope systems, Daejeon, Korea) using K1-image software (Nanoscope systems, Daejeon, Korea). The intensity of nuclear NRF2 was measured by ImageJ software (Version 1.53c, NIH, Bethesda, MD, USA).

Dual-Luciferase Assay
HepG2 cells were co-transfected with a DNA mixture containing the pGL4.37[luc2P/ ARE/Hygro] reporter (Promega, Madison, WI, USA), renilla luciferase control reporter (pRL-TK, Thermo Fisher Scientific, Waltham, MA, USA), and 1 µL of Lipofectamine 2000. After 4 h, cells were treated with FFA and G. cambogia for 12 h. After cells were harvested using passive lysis buffer, ARE promoter activity was measured with the Dual-Luciferase Reporter Assay System (Promega) using a Glomax 20/20 luminometer reader (Promega). Specific promoter activity was expressed as the relative activity ratio of firefly luciferase to Renilla luciferase.

Identification of Hydroxycitric Acid in G. cambogia Using Liquid Chromatography-High Resolution Mass Spectrometry
For the identification of hydroxycitric acid in G. cambogia, the content of hydroxycitric acid was analyzed using liquid chromatography-high resolution mass spectrometry (LC-HRMS). The G. cambogia powder was dissolved at a concentration of 1 µg/mL in distilled water. After vortexing for 1 min, the solution was centrifuged at 12,000 rpm for 7 min. The supernatant was transferred to an LC vial for LC-HRMS analysis. The LC-HRMS system consisted of a CTC HTS PAL auto-sampler (LEAP Technologies, Carrboro, NC, USA), LC-20AD pumps (Shimadzu Corporation, Columbia, MD, USA), CBM-20A controller (Shimadzu Corporation), and a hybrid quadrupole time-of-flight mass spectrometer 5600 with an electrospray ionization source (Sciex, Foster City, CA, USA). The hybrid quadrupole time-of-flight mass spectrometer 5600, equipped with an electrospray ionization source, was operated in the negative ion mode. The mass transition of parent-to-parent ion was used for hydroxycitric acid (m/z 207.0→207.0). Water containing 0.1% formic acid (A) and acetonitrile containing 0.1% formic acid (B) was used as mobile phase. The flow rate was 0.4 mL/min. Error ppm was applied for the accuracy of measurement. Data processing was performed using the Analyst ® TF 1.6 and MultiQuant™ software (Version 3.0.3, Sciex).

Statistical Analyses
All data were expressed as mean ± S.D. of 4-7 independent experiments mentioned in the respective figure legends or graphs. Statistical analyses were performed with GraphPad Prism software (Version 9, San Diego, CA, USA). The normality of the data distributions was tested using the Shapiro-Wilk test. A two-sided, unpaired Student's t-test was used to analyze the difference between two groups of data. The data in Figures 1F and 2E were analyzed using Welch's t-test for unequal variances. Differences across three or more groups were tested via one-way analysis of variance (ANOVA), followed by a post hoc analysis with Bonferroni's test if F achieved statistical significance (p < 0.05) and there was no significant variance in homogeneity with the Barlett's test. The correlation results were determined by Pearson's rank test. Differences with p < 0.05 were considered statistically significant.

Garcinia cambogia Attenuates NALFD in HFD-Induced Mice by Reducing Hepatic Steatosis
To investigate the effect of G. cambogia on NAFLD in HFD-fed mice, we administered G. cambogia and orlistat (positive control for the anti-obesity effect) for 8 weeks with HFD feeding. As shown in Table 1, G. cambogia significantly reduced weight gain resulting from HFD feeding, indicating an anti-obesity effect. In addition, administration of G. cambogia improved the abnormal serum levels of measured parameters, such as alanine aminotransferase (ALT) and aspartate aminotransferase (AST), markers of liver function, and triglycerides and total cholesterol, and markers of lipid metabolism, indicating that G. cambogia can regulate HFD-induced abnormal metabolic phenotype. Furthermore, G. cambogia inhibited the formation of hepatic lipid droplets and increased liver index, an indicator of NAFLD [46], resulting from HFD feeding ( Figure 1A). There were no signs of obvious pathological changes in major organs such as the heart, spleen, lung, and kidney, implying no other organ toxicity ( Figure S1). To examine the underlying mechanism, we used the HepG2 cell line for an in vitro study. Compared to digitonin (100 µg/mL, positive control for cytotoxicity), G. cambogia (20-80 µg/mL) did not show any cytotoxic effects for 24 h in HepG2 cells ( Figure 1B). In addition, G. cambogia significantly reduced free fatty acid (1 mM FFA, composed of oleate and palmitate)-induced lipid accumulation in HepG2 cells for 24 and 48 h, similar to EGCG (50 µM) as a known substance to have an anti-steatosis effect ( Figure 1C and Figure S2A) [47].
Since hepatic lipid accumulation is regulated by lipogenesis-related molecules, we examined the levels of C/EBPα and PPARγ, two major transcription factors of lipogenesis, in FFA-treated HepG2 cells. As shown in Figure 1D, FFA treatment increased C/EBPα and PPARγ expression indicating de novo lipogenesis. However, G. cambogia significantly decreased FFA-induced alterations in the C/EBPα and PPARγ expression. Then we measured the effect of G. cambogia on the lipogenic gene transcription activated by C/EBPα and PPARγ. FFA treatment increased the levels of fatty acid synthase (FASN), fatty acid-binding protein 4 (FABP4), and stearoyl-CoA desaturase (SCD) transcripts, and G. cambogia significantly reduced the transcription of these genes ( Figure 1E). As in the results of in vitro experiments, G. cambogia significantly reduced the protein levels of C/EBPα and PPARγ ( Figure 1F) and the transcript levels of Fasn, Fabp4, and Scd ( Figure 1G) in liver tissues of HFD-fed mice. As shown in Figure 1H, significant positive correlations were observed between Fasn transcript and C/EBPα and PPARγ proteins, indicating that increased C/EBPα and PPARγ by HFD were inhibited by G. cambogia, which regulated lipogenic genes such as Fasn. These results provide critical insight into the inhibitory effect of G. cambogia on NAFLD via suppressing hepatic steatosis.

Garcinia cambogia Suppressed FFA-Induced Apoptosis in the Liver
Excessive FFAs in the liver by consuming HFD cause apoptosis and triglyceride accumulation [48]. In addition, hepatocellular apoptosis plays a major role in the early mechanism underlying the progression of liver diseases, including NAFLD [49]. Thus, we examined whether G. cambogia regulates FFA-induced apoptosis. As shown in Figure 2A, G. cambogia administration significantly decreased the rate of HFD-induced apoptosis, as measured by TUNEL in liver tissues. In addition, FFA treatment reduced HepG2 cell viability; however, treatment with either G. cambogia (20-80 µg/mL) or EGCG (50 µM) improved cell viability, which had been reduced by FFA for 24 and 48 h ( Figure 2B and Figure S2B). Moreover, G. cambogia and EGCG (positive control of anti-apoptosis effect [50]) inhibited FFA-induced cell apoptosis, indicating that the improvement of cell viability by G. cambogia was due, at least in part, to suppression of apoptosis ( Figure 2C).  Since apoptosis is regulated by proapoptotic and antiapoptotic molecules such as Bcl-2 family members related to mitochondrial dysfunction and subsequent caspase-3 activation and cleavage of PARP [51][52][53], we examined the effect of G. cambogia on the alteration of apoptosis-related molecules in FFA-treated HepG2 cells. FFA treatment significantly decreased the expression of Bcl2 apoptosis regulator (Bcl-2) and increased the expression of Bcl-2-associated X protein (BAX), thereby suppressing the Bcl-2/BAX ratio, and increased the levels of caspase-3 activity and cleaved PARP indicating the induction of apoptosis ( Figure S3A and Figure 2D). These results were similar in experiments observed in the liver tissues of HFD-fed mice ( Figure S3B and Figure 2E). In both in vitro and in vivo experiments, G. cambogia (20-80 µg/mL) treatment increased the Bcl-2/Bax ratio and inhibited caspase-3 activation and PARP cleavage, indicating that G. cambogia suppresses apoptosis-related apoptosis pathways. Taken together, these results suggest that G. cambogia alleviates HFD-and FFA-induced apoptosis to inhibit the progression of NAFLD.

Garcinia cambogia Inhibits HFD-and FFA-Induced ROS Production via NRF2 Activation
Excessive accumulation of FFAs by energy imbalance in the liver produces ROS, and high levels of ROS can induce lipid accumulation via failure of lipid metabolism and liver damage via apoptosis [48]. Thus, we examined whether G. cambogia regulates FFA-induced ROS production. Intracellular ROS levels were increased by FFA in HepG2 cells, and both G. cambogia (20-80 µg/mL) and NAC (5 mM, positive control for the antioxidant effect [54]) significantly reduced FFA-induced ROS production for 24 and 48 h ( Figure 3A and Figure S2C). In addition, the serum level of MDA, as a marker of oxidative stress and antioxidant status [55,56], was increased in HFD-fed mice, but G. cambogia administration alleviated the increase in the MDA level, indicating G. cambogia-mediated antioxidant effect ( Figure 3B). Cellular ROS are regulated by endogenous antioxidant defense systems such as enzymatic antioxidants like heme oxygenase 1 (HO-1) and superoxide dismutase (SOD) [57]. In addition, NRF2, one of the major transcription factors of the antioxidant defense system, regulates HMOX1 and SOD transcription for ROS homeostasis [58]. Thus, we examined whether G. cambogia regulates NRF2 levels in HFD-fed mice and FFA-treated HepG2 cells. As a result of measuring NRF2 expression level in the livers of HFD-fed mice, we found that G. cambogia reduced HFD-induced inhibition of NRF2 expression ( Figure 3C). In addition, Hmox1 and Sod1 gene expression, which had been inhibited by HFD, were recovered by G. cambogia ( Figure 3D). Furthermore, significant negative correlations between MDA protein and Hmox1 and Sod1 transcripts were observed, indicating that increased MDA by HFD was inhibited by G. cambogia-induced alterations of the antioxidant defense system ( Figure 3E). Taken together, G. cambogia activates the intracellular antioxidant defense system by increasing the expressions of NRF2 and downstream genes to reduce ROS production. effect [54]) significantly reduced FFA-induced ROS production for 24 and 48 h (Figures 3  and S2C). In addition, the serum level of MDA, as a marker of oxidative stress and antiox dant status [55,56], was increased in HFD-fed mice, but G. cambogia administration allev ated the increase in the MDA level, indicating G. cambogia -mediated antioxidant effect (Fi ure 3B). Cellular ROS are regulated by endogenous antioxidant defense systems such as e zymatic antioxidants like heme oxygenase 1 (HO-1) and superoxide dismutase (SOD) [57 In addition, NRF2, one of the major transcription factors of the antioxidant defense sy tem, regulates HMOX1 and SOD transcription for ROS homeostasis [58]. Thus, we exam ined whether G. cambogia regulates NRF2 levels in HFD-fed mice and FFA-treated HepG cells. As a result of measuring NRF2 expression level in the livers of HFD-fed mice, w found that G. cambogia reduced HFD-induced inhibition of NRF2 expression ( Figure 3C In addition, Hmox1 and Sod1 gene expression, which had been inhibited by HFD, we recovered by G. cambogia ( Figure 3D). Furthermore, significant negative correlations b tween MDA protein and Hmox1 and Sod1 transcripts were observed, indicating th

Increased NRF2 Nuclear Expression and ARE Activation by Garcinia cambogia Regulates Antioxidant Gene Expression in FFA-Treated HepG2 Cells
NRF2 functions as a transcription factor in the nucleus by ARE promoter activation and subsequent modulation of antioxidant gene expression [59]. Thus, we examined whether G. cambogia affected the nuclear expression of NRF2 and ARE promoter activity and in FFAtreated HepG2 cells. Similar to the in vivo results, FFA significantly reduced the expression of NRF2, and G. cambogia restored FFA-induced inhibition of NRF2 expression ( Figure 4A). Moreover, G. cambogia increased the expression of NRF2 in the nucleus ( Figure 4B,C), which could influence the expression of downstream genes. In addition, G. cambogia increased FFA-induced reduction of ARE activity ( Figure 4D). As expected, G. cambogia restored FFA-induced reduction of HMOX1 and SOD1 gene expressions ( Figure 4E). Taken together, these results suggest that G. cambogia activates the NRF2-ARE signaling pathway to inhibit FFA-induced abnormal ROS production.
( Figure 4B,C), which could influence the expression of downstream genes. In addition cambogia increased FFA-induced reduction of ARE activity ( Figure 4D). As expected, cambogia restored FFA-induced reduction of HMOX1 and SOD1 gene expressions ( Fig  4E). Taken together, these results suggest that G. cambogia activates the NRF2-ARE sign ing pathway to inhibit FFA-induced abnormal ROS production.  Colocalized FITC (i.e., NRF2) and DAPI (i.e., nuclei) were quantified using ImageJ software, and relative intensities in the nuclear and cytoplasmic fractions were expressed as a histogram. Scale bars: 20 µm (n = 5 per group). (D) Effect of G. cambogia on ARE promoter activity in FFA-treated HepG2 cells. ARE-luc vector-transfected HepG2 cells were cotreated FFA and G. cambogia (20-80 µg/mL) for 12 h and detected ARE promoter activity (n = 6 per group). (E) Effect of G. cambogia (20-80 µg/mL) on the transcript levels of HMOX1 and SOD1 in FFA-treated HepG2 cells (n = 5 per group). ** p < 0.01 vs. Con, # p < 0.05 and ## p < 0.01 vs. FFA. Data are mean ± S.D.

Hydroxycitric Acid Contributed to Suppress FFA-Induced ROS Production, Lipid Accumulation, and Apoptosis
A recent study reported that hydroxycitric acid (HCA) reduced oleic acid-induced steatosis and oxidative stress in primary chicken hepatocytes [60]. In addition, it is known that G. cambogia contains from 20 to 60% hydroxycitric acid (HCA) as a main component [24]. The HCA content of G. cambogia used in this experiment was analyzed using liquid chromatography-high resolution mass spectrometry (LC-HRMS), and HCA content was 59.55 ± 2.66% ( Figure 5A). To verify whether the same action of G. cambogia as shown so far is actually the same in HCA, the effect of HCA (24 and 48 µg/mL) at concentrations equivalent to those of G. cambogia (40 and 80 µg/mL) on FFA-induced ROS production, lipid accumulation, and apoptosis was examined in FFA-treated HepG2 cells. HCA significantly inhibited FFA-induced ROS production and lipid accumulation ( Figure 5B,C). In addition, HCA attenuated FFA-induced reduction of cell viability and induction of apoptosis ( Figure 5D,E). Similar to the effect of G. cambogia, HCA ameliorated FFA-induced decrease of NRF2 expression and HMOX1 and SOD1 transcription, indicating an antioxidant effect of HCA ( Figure 5F,G). Furthermore, HCA regulated lipogenesis-and apoptosis-related signaling pathways. As shown in Figure 5H, HCA inhibited FFA-induced upregulation of C/EBPα and PPARγ expression, FAS, FABP4, and SCD transcription to suppress lipogenesis ( Figure 5H,I). In addition, HCA significantly altered the FFA-induced Bcl-2/BAX ratio, caspase-3 activity, and PARP cleavage to reduce apoptosis ( Figure 5J and Figure S3C). These results indicated that the underlying effects of G. cambogia are due, at least in part, to HCA.  Taken together, our results support the conclusion that NRF2 activation by G. cambogia and HCA contributed to the antioxidant effect also affects NAFLD progression through hepatic steatosis and apoptosis inhibition ( Figure 6). Taken together, our results support the conclusion that NRF2 activation by G. cambogia and HCA contributed to the antioxidant effect also affects NAFLD progression through hepatic steatosis and apoptosis inhibition ( Figure 6).

Discussion
To date, there are no effective pharmacological therapies for NAFLD; thus, therapeutic approaches to manage NAFLD by dietary supplementation, for example, silymarin [61], berberine [19], and resveratrol [62], are considered. Although G. cambogia, a natural product used as a weight-loss supplement in many countries, has been reported to have an anti-NAFLD effect [28], the underlying mechanism is poorly understood.
In the present study, we demonstrated for the first time that (1) G. cambogia suppressed hepatic steatosis and apoptosis in the liver of HFD-fed mice and in FFA-treated HepG2 cells; (2) G. cambogia inhibited HFD-and FFA-induced upregulation of C/EBPα and PPARγ expression, and FASN, FABP4, and SCD transcription to regulate lipogenesis; (3) G. cambogia reduced the HFD-and FFA-induced alterations of the Bcl-2/BAX ratio and PARP cleavage to regulate apoptosis; and (4) these normalizations were due to the anti-

Discussion
To date, there are no effective pharmacological therapies for NAFLD; thus, therapeutic approaches to manage NAFLD by dietary supplementation, for example, silymarin [61], berberine [19], and resveratrol [62], are considered. Although G. cambogia, a natural product used as a weight-loss supplement in many countries, has been reported to have an anti-NAFLD effect [28], the underlying mechanism is poorly understood.
In the present study, we demonstrated for the first time that (1) G. cambogia suppressed hepatic steatosis and apoptosis in the liver of HFD-fed mice and in FFA-treated HepG2 cells; (2) G. cambogia inhibited HFD-and FFA-induced upregulation of C/EBPα and PPARγ expression, and FASN, FABP4, and SCD transcription to regulate lipogenesis; (3) G. cambogia reduced the HFD-and FFA-induced alterations of the Bcl-2/BAX ratio and PARP cleavage to regulate apoptosis; and (4) these normalizations were due to the antioxidant effect of G. cambogia and HCA through NRF2-ARE pathway activation. These findings are summarized in Figure 6.
The effect of G. cambogia and HCA, its main constituent, on lipogenesis has been reported in several studies in animal models [24,25,63]. It was recently published that the regulatory effect of HCA on lipid accumulation in cultured primary chicken hepatocytes reduces the acetyl-CoA supply, which is mainly achieved via inhibition of ATP-citrate lyase and acceleration of energy metabolism [63]. In addition, HCA reduced steatosis and oxidative stress by regulating the AMPK-mediated signaling pathway [60]. However, our study focused on the antioxidant effect of G. cambogia and HCA via NRF2-ARE activation, and it modulated lipogenesis and apoptosis to ameliorate NAFLD.
Reactive oxygen species (ROS) are generated in cells by several factors, such as stressor responses and environmental factors [64]. Excessive accumulation of free fatty acids (FFAs) due to western diets, including high-fat, is one of the important factors. ROS overproduction induces lipid peroxidation, mitochondrial dysfunction, and lipogenic protein activation to increase fat accumulation in the liver [12]. In addition, excessive ROS activates hepatic C/EBPα and PPARγ, key transcription factors that control many lipogenic genes, such as FASN, FABP4, and SCD, to promote NAFLD progression [65,66]. Our study firstly revealed that G. cambogia and HCA inhibited HFD-and FFA-induced C/EBPα and PPARγ activation by reducing intracellular ROS levels and leading to suppression of lipogenic gene transcription and hepatic steatosis (Figure 1, Figure 3A,B and Figure 5).
Increased ROS levels have been indicated to enhance the expression of apoptosisrelated proteins [48]. In addition, apoptosis is a prominent feature of NAFLD progression and liver disease [67]. Since ROS causes mitochondrial dysfunction and alterations in Bcl-2 family proteins such as Bcl-2 and BAX, which are major regulators of apoptotic processes and lead to collapse of the mitochondrial membrane potential [68], we examined the effect of G. cambogia on Bcl-2 and BAX expression and PARP cleavage. This study revealed that G. cambogia and HCA significantly attenuated the HFD-and FFA-induced decrease in the Bcl-2/BAX ratio and induction of PARP cleavage, indicating suppression of ROS-mediated apoptosis (Figures 2 and 5E,J).
G. cambogia contains many components, including alkaloids, flavonoids, phenolic compounds, saponins, tannins, carbohydrates, and proteins [24]. To date, few single compounds have been isolated from G. cambogia. HCA, garcinol, and guttiferone K isolated from the extract showed protective antioxidant effects against lipid and protein oxidation/glycation and endoplasmic reticulum (ER) stress [69][70][71]. In addition, advanced glycation end products (AGE) can be a potential biomarker of the NAFLD stage since AGE showed good criteria for dividing patients with minimal steatosis (BARD score 0-1) vs. moderate steatosis (BARD score 2-4) [72]. Furthermore, a high intake of fat and carbohydrates can induce ER stress, and dysfunction of ER develops metabolically-driven NAFLD progress [73,74]. Although this study revealed that HCA had therapeutic effects by suppressing lipid accumulation and apoptosis in liver cells, other components may be effective to NAFLD since recent clinical trials suggest that protein oxidation/glycation and ER stress are involved in liver fibrosis in NAFLD [75,76]. Thus, further studies are needed to investigate additional anti-NAFLD effects and the underlying mechanism of the G. cambogia and its components.
Finally, it has been reported that G. cambogia is associated with hepatotoxicity, including lipid peroxidation and oxidative stress genes [77]. However, this assertion is controversial because our study showed that G. cambogia administration suppressed the HFD-induced increases in ALT and AST levels (Table 1), and G. cambogia inhibited HFDand FFA-induced oxidative stress via upregulating antioxidative barriers. In addition, HCA, the main constituent, does not promote liver toxicity or inflammation [78].

Conclusions
In conclusion, this study demonstrated that G. cambogia inhibits HFD-induced NAFLD by suppressing hepatic steatosis and apoptosis through the regulation of lipogenesis-and apoptosis-related signaling pathways. These effects are due to the antioxidant effect of G. cambogia and HCA by activating the NRF2-ARE pathway. Our findings provide new insight into the mechanism by which G. cambogia suppresses NAFLD progression.