Prevention of free fatty acid–induced hepatic lipotoxicity by 18β-glycyrrhetinic acid through lysosomal and mitochondrial pathways

Authors

  • Xudong Wu,

    1. Jiangsu Center for Drug Screening, Jiangsu Center for Pharmacodynamic Research and Evaluation, China Pharmaceutical University, Nanjing, Jiangsu, People's Republic of China
    2. Department of Microbiology and Immunology, Virginia Commonwealth University, Richmond, VA
    Search for more papers by this author
  • Luyong Zhang,

    Corresponding author
    1. Jiangsu Center for Drug Screening, Jiangsu Center for Pharmacodynamic Research and Evaluation, China Pharmaceutical University, Nanjing, Jiangsu, People's Republic of China
    • China Pharmaceutical University, Nanjing, Jiangsu, 210009
    Search for more papers by this author
  • Emily Gurley,

    1. Department of Microbiology and Immunology, Virginia Commonwealth University, Richmond, VA
    Search for more papers by this author
  • Elaine Studer,

    1. Department of Microbiology and Immunology, Virginia Commonwealth University, Richmond, VA
    Search for more papers by this author
  • Jing Shang,

    1. Jiangsu Center for Drug Screening, Jiangsu Center for Pharmacodynamic Research and Evaluation, China Pharmaceutical University, Nanjing, Jiangsu, People's Republic of China
    Search for more papers by this author
  • Tao Wang,

    1. Jiangsu Center for Drug Screening, Jiangsu Center for Pharmacodynamic Research and Evaluation, China Pharmaceutical University, Nanjing, Jiangsu, People's Republic of China
    Search for more papers by this author
  • Cuifen Wang,

    1. Jiangsu Center for Drug Screening, Jiangsu Center for Pharmacodynamic Research and Evaluation, China Pharmaceutical University, Nanjing, Jiangsu, People's Republic of China
    Search for more papers by this author
  • Ming Yan,

    1. Jiangsu Center for Drug Screening, Jiangsu Center for Pharmacodynamic Research and Evaluation, China Pharmaceutical University, Nanjing, Jiangsu, People's Republic of China
    Search for more papers by this author
  • Zhenzhou Jiang,

    1. Jiangsu Center for Drug Screening, Jiangsu Center for Pharmacodynamic Research and Evaluation, China Pharmaceutical University, Nanjing, Jiangsu, People's Republic of China
    Search for more papers by this author
  • Phillip B. Hylemon,

    1. Department of Microbiology and Immunology, Virginia Commonwealth University, Richmond, VA
    2. Department of Internal Medicine/Gastrointestinal Division, Virginia Commonwealth University, Richmond, VA
    Search for more papers by this author
  • Arun J. Sanyal,

    1. Department of Internal Medicine/Gastrointestinal Division, Virginia Commonwealth University, Richmond, VA
    Search for more papers by this author
  • William M. Pandak Jr,

    1. Department of Internal Medicine/Gastrointestinal Division, Virginia Commonwealth University, Richmond, VA
    Search for more papers by this author
  • Huiping Zhou

    Corresponding author
    1. Jiangsu Center for Drug Screening, Jiangsu Center for Pharmacodynamic Research and Evaluation, China Pharmaceutical University, Nanjing, Jiangsu, People's Republic of China
    2. Department of Microbiology and Immunology, Virginia Commonwealth University, Richmond, VA
    3. Department of Internal Medicine/Gastrointestinal Division, Virginia Commonwealth University, Richmond, VA
    • Department of Microbiology & Immunology, Virginia Commonwealth University, P.O. Box 980678, Richmond, VA 23298-0678
    Search for more papers by this author
    • fax: 804-828-0676


  • Potential conflict of interest: Nothing to report.

Abstract

Nonalcoholic fatty liver disease (NAFLD) is the most common liver disease and affects millions of people worldwide. Despite the increasing prevalence of NAFLD, the exact molecular/cellular mechanisms remain obscure and effective therapeutic strategies are still limited. It is well-accepted that free fatty acid (FFA)-induced lipotoxicity plays a pivotal role in the pathogenesis of NAFLD. Inhibition of FFA-associated hepatic toxicity represents a potential therapeutic strategy. Glycyrrhizin (GL), the major bioactive component of licorice root extract, has a variety of pharmacological properties including anti-inflammatory, antioxidant, and immune-modulating activities. GL has been used to treat hepatitis to reduce liver inflammation and hepatic injury; however, the mechanism underlying the antihepatic injury property of GL is still poorly understood. In this report, we provide evidence that 18 β-glycyrrhetinic acid (GA), the biologically active metabolite of GL, prevented FFA-induced lipid accumulation and cell apoptosis in in vitro HepG2 (human liver cell line) NAFLD models. GA also prevented high fat diet (HFD)-induced hepatic lipotoxicity and liver injury in in vivo rat NAFLD models. GA was found to stabilize lysosomal membranes, inhibit cathepsin B expression and enzyme activity, inhibit mitochondrial cytochrome c release, and reduce FFA-induced oxidative stress. These characteristics may represent major cellular mechanisms, which account for its protective effects on FFA/HFD-induced hepatic lipotoxicity. Conclusion: GA significantly reduced FFA/HFD-induced hepatic lipotoxicity by stabilizing the integrity of lysosomes and mitochondria and inhibiting cathepsin B expression and enzyme activity. (HEPATOLOGY 2008.)

Nonalcoholic fatty liver disease (NAFLD) is the most common liver disease in the United States and worldwide, and has emerged as a major public health concern. NAFLD has two stages: a fatty liver and nonalcoholic steatohepatitis (NASH), the most severe form of NAFLD, and end-stage liver disease.1 Currently, the cellular mechanisms of NAFLD development and disease progression remain undefined and the therapeutic strategies are still limited. Numerous studies suggest that obesity, diabetes, and the metabolic syndrome are closely associated with the disease progression of NAFLD.2, 3

Free fatty acids (FFAs) are the important mediators of lipotoxicity. Circulating FFAs are the primary contributor to the liver triacylglycerol (TG) content, and plasma FFAs levels are correlated with disease severity in NAFLD patients.4 Although the mechanism by which FFAs induce lipoapoptosis is still not fully identified, most recent studies indicate that multiple mechanisms might be involved in FFA-induced hepatic lipotoxicity. FFAs not only induce c-jun N-terminal kinase (JNK)-dependent activation of apoptotic Bcl-2 proteins Bim and BCL-2-associated X protein, which trigger the mitochondrial apoptotic pathway,2 but also induce tumor necrosis factor (TNF)-α expression through a lysosomal pathway.5 Accumulation of intracellular FFAs results in lysosomal permeabilization and release of cathepsin B into the cytosol, which further promotes mitochondrial reactive oxygen species (ROS) production and release of cytochrome c.6, 7 Both lysosomal and mitochondrial apoptotic pathways play pivotal roles in TNF-α-induced hepatocyte apoptosis and liver injury.6, 8

Cathepsin B is the best characterized mammalian cysteine peptidase and is ubiquitously expressed.9 Intracellular cathepsin B is localized in the lysosomes. It has been shown that hepatic steatosis is associated with lysosomal permeabilization.10, 11 Disruption of the lysosomes and subsequent release of lysosomal proteases into cytosol has been implicated in cellular injury. Recent studies have demonstrated that inactivation of cathepsin B attenuates hepatocyte apoptosis and liver damage induced by TNF-α, cholestasis, and cold ischemia–warm reperfusion.10-12 These findings suggest that cathepsin B plays a prominent role in apoptosis and tissue injury. Cathepsin B inhibition may have a potential therapeutic effect in treating liver diseases.

Licorice is one of the most ancient medicinal plants; it has long been used as a conditioning and flavoring agent and in traditional Chinese medicine for the treatment of various inflammatory diseases.13 Glycyrrhizin (GL) is a major bioactive triterpene glycoside of licorice root extract and has a variety of pharmacological properties. GL is one of the leading natural compounds currently used clinically for treatment of chronic hepatitis C and human immunodeficiency virus infections.14, 15 Pharmacokinetic studies have shown that GL exhibits its pharmacological functions through its biologically active metabolite, 18β-glycyrrhetinic acid (GA), which is formed by presystemic hydrolysis (Fig. 1).15 Although it has been shown that GA has a protective effect on hepatic injury, the underlying mechanism by which GA improves liver biochemistry and histology remains unknown. Whether GA has a protective effect on FFA-induced lipotoxicity in liver has not been explored so far. The primary objective of the current study was to examine the effects of GA on FFA-induced lipotoxicity and the potential underlying mechanisms.

Figure 1.

Biotransformation of GL to GA.

Abbreviations

AO, acridine orange; DCF, 2,7-dichlorofluorescein; DCFH, 2,7-dichlorofluorescin; DCFH-DA, 2,7-dichlorofluorescein diacetate; FFA, free fatty acid; FITC, fluorescein isothiocyanate; GA, 18β-glycyrrhetinic acid; GL, glycyrrhizin; HFD, high fat diet; IC50, median inhibitory concentration; JC-1, 5,5′,6,6′-tetrachloro-1,1′,3,3′ tetraethyl benzimidazolylcarbocyanine iodide; JNK, c-jun N-terminal kinase; Me, methyl ester; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; PA, palmitate; PBS, phosphate-buffered saline; PI, propidium iodide; ROS, reactive oxygen species; SA, stearate; SEM, standard error of the mean; TG, triglyceride (or triacylglycerol); TNF, tumor necrosis factor

Materials and Methods

Cell Culture Treatment.

Human HepG2 cells were cultured in modified Eagle's medium with 10% fetal bovine serum. FFAs (oleate/palmitate [PA], 2:1) were mixed with FFA-free bovine serum albumin and the mixture was added to medium to a final concentration of 1 mM. GA was dissolved in dimethylsulfoxide.

Isolation and Culture of Primary Hepatocytes.

Primary hepatocytes were isolated from male Sprague-Dawley rats (250 to 300 g) and cultured as described.16

Animal Studies.

To examine the effect of high fat diet (HFD)-induced hepatic injury, male Sprague-Dawley rats (180 to 200 g) were randomly assigned to four groups (n = 8). The first control group was fed a normal diet; the three HFD groups were fed HFD containing 2% cholesterol and 10% lard for 4, 6, and 8 weeks, respectively. To examine the protective effect of GA on HFD-induced lipotoxicity in liver, rats were randomly assigned to the following four groups: (I) normal control; (II) HFD group; (III) HFD plus GA 25 mg/kg; and (IV) HFD plus GA 50 mg/kg. Rats in the normal control group (I) were fed a standard diet; the rats in the other three groups (II-IV) were fed HFD. The rats were gavaged with GA or control solution every day for 8 weeks. All rats were housed under identical conditions in an aseptic facility and given free access to water and food. The rats were weighed once a week to adjust drug intake. At the end of each time period, rats were fasted for 16 hours and blood samples were collected. Serum total cholesterol, free cholesterol, TG, low density lipoprotein-cholesterol, high density lipoprotein-cholesterol, alanine aminotransferase, and aspartate aminotransferase were measured using standard enzymatic techniques. All studies were approved by the Animal Study Committee of China Pharmaceutical University.

Histopathology Analysis.

For each rat, three specimens from different regions of the liver were collected and fixed in 4% paraformaldehyde in 0.1 M phosphate buffer at room temperature overnight. The regions of the specimens were standardized for all rats. The paraffin-embedded tissue sections (4 μm) were stained with hematoxylin and eosin according to standard techniques. The samples were examined blindly by a professional pathologist to evaluate the presence of ballooning, steatosis, inflammation, and fibrosis. Lesions were evaluated semiquantitatively on a four-point scale (1 = absent, 2 = mild, 3 = moderate, and 4 = intense) for each damage as described by Brunt et al.17 The extent of lesions in each rat was expressed as the average score of three separate specimens.

Electron Microscopy.

The HepG2 cells were collected and fixed with 4% glutaraldehyde in 0.1 M phosphate buffer, pH 7.2, at 4°C for 15 minutes. Cells were rinsed for 30 minutes in three changes of 0.1 M phosphate buffer, followed by treatment with 1% OsO4 in 0.1 M phosphate-buffer for 1 hour. After being washed with distilled water three times, the cells were dehydrated and embedded with epoxy resin. Ultrathin sections of 60 nm were prepared. Electron photomicrographs were taken of the ultrastructures of HepG2 cells under a transmission electron microscope.

Nile Red Staining.

Nile red staining was used to specifically stain the intracellular fat. The HepG2 cells were treated with 1 mM of FFAs together with GA (0 to 30 μM) for 24 hours. Cells were collected and incubated with Nile red (100 ng/mL) in PBS for 5 minutes. After being washed with PBS, the cells were resuspended in PBS and the fluorescence intensity was measured by flow cytometry at an excitation wavelength of 488 nm. Unstained cells were used to adjust instrument settings.18

Apoptosis Analysis.

HepG2 cells were treated with 1 mM FFAs together with various concentrations of GA or 20 μM of CA-074 methyl ester (Me) (a cathepsin B–specific inhibitor) for 24 hours. Cells were collected and washed with cold PBS, then resuspended in 400 μL of reaction solution (10 mM of 4-[2-hydroxyethyl]-1-piperazine ethanesulfonic acid, 140 mM of NaCl, 2.5 mM of CaCl2, pH 7.4) and incubated with Annexin V-fluorescein isothiocyanate (FITC) and propidium iodide (PI) staining solution according to the manufacturer's instructions. Cells stained with Annexin V-FITC and PI were further analyzed by two-color flow cytometry to quantify the apoptotic cells. Annexin V and PI emissions were detected in the FL1 and FL3 channels of a Cytinics FC 500 flow cytometer. At least 10,000 cells were analyzed in each sample.

Lysosomal Stability Analysis.

Lysosomal stability was assessed by using the acridine orange (AO)-uptake method.19 AO is a metachromatic fluorophore, which shows red fluorescence at high (lysosomal) concentrations and green fluorescence at low (nuclear and cytosolic) concentrations. Rupture of lysosomes was monitored as an increase in the number of cells with decreased AO uptake, as indicated by low red fluorescence. Cells were collected after treatment with 1 mM FFAs together with various concentrations of GA for 24 hours and were stained with AO (5 μg/mL) for 30 minutes at 37°C in the dark. The intensity of red fluorescence was measured by flow cytometry, using the FL3 channel. Cells with decreased red fluorescence were gated and their percentages were indicated.

Measurement of β-Galactosidase Activity.

To determine the effect of GA on lysosomal stability in liver, the free form and membrane-bound form of β-galactosidase activities were measured. Increase of free β-galactosidase activity indicates the lysosomal rupture. Liver tissues were homogenized in 0.3 M sucrose at 4°C and centrifuged at 35,000g for 20 minutes. The supernatant with free enzyme was collected. The pellets containing bound enzyme were suspended in 0.3 M sucrose containing 0.1% Triton X-100, and were incubated at 4°C for 24 hours. The supernatant containing bound enzyme was collected by centrifugation at 35,000g for 20 minutes.20 The β-galactosidase activity was determined based on the degradation of 4-methylumbelliferyl-β-D-galactopyranoside using a Tecan Safire2 Microplate Reader (excitation wave: 360 nm, emission wave: 450 nm). The enzyme activity was expressed as nmol/mg protein.

Western Blot Analysis.

The total cell lysates were prepared and processed as described.16 The expression levels of cathepsin B, cytochrome c, p-JNK, and total JNK were detected with specific antibodies. Both β-actin and α-tubulin were used as loading controls. The densities of immunoblot bands were analyzed with Image J computer software (National Institutes of Health).

Mitochondrial Transmembrane Potential.

The mitochondrial transmembrane potential (ΔΨ) was measured by using the fluorescent probe 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine iodide (JC-1), which is able to selectively enter the mitochondria and exists in a monomeric form emitting at 527 nm with excitation at 490 nm. When JC-1 forms aggregates, the emission is shifted to 590 nm. When the mitochondrial membrane is depolarized, the aggregates decrease and monomers increase. HepG2 cells were treated with 1 mM of FFAs together with various concentrations of GA for 24 hours. Cells were collected and stained with JC-1 (5 μg/mL) at 37°C in the dark for 15 minutes. The fluorescence of FL-1 for JC-1 monomers and FL-2 for JC-1 aggregates was measured by flow cytometry.21

Measurement of Cathepsin B Activity.

Cathepsin B (2 μg/mL) isolated from human liver (from Sigma Cat# C8571) was incubated with 200 μM of cathepsin B–specific substrate Z-Arg-Arg-7-amido-4-methylcourmarin hydrochloride together with various concentrations of GA or CA-074 in 60 μL of assay buffer (100 mM 2-(N-morpholino)ethanesulfonic acid, pH 5.5, 20 mM dithiothreitol) at 37°C for 30 minutes. The level of fluorescence, generated through the hydrolysis of Z-Arg-Arg-7-amido-4-methylcoumarin hydrochloride by cathepsin B only, was measured on an excitation wavelength of 355 nm and an emission wavelength of 460 nm.22 To measure the cathepsin B activity in HepG2 cells and rat liver tissue, the whole-cell lysates and liver homogenates were centrifuged at 15,000g for 10 minutes at 4°C. The supernatant containing cathepsin B was collected. The cathepsin B activity was assayed as described above.

Cathepsin B Immunofluorescence.

Animals were perfused intracardially initially with 0.9% saline and then with 4.0% paraformaldehyde (pH 7.4). After the liver capsule was carefully dissected away, the segments of tissue were postfixed in 4% paraformaldehyde for 2 hours and then in 30% sucrose overnight.22 Optimal cutting temperature (OCT)-embedded segments were sectioned (3 μm) and mounted onto slides at −30°C. The sections were blocked in blocking buffer (5% goat serum, 5% glycerol, 0.004% sodium azide) at 37°C for 30 minutes, then incubated for 2 hours with a rabbit polyclonal anticathepsin B antibody (1:200). After the sections were washed with PBS three times, FITC-conjugated goat anti-rabbit antibody was added and sections were incubated at 37°C for 45 minutes. Images were collected with a fluorescence microscope.

Measurement of Intracellular ROS.

Intracellular ROS was measured using the 2,7-dichlorofluorescein diacetate (DCFH-DA) method.23 DCFH-DA diffuses through the cell membrane and is enzymatically hydrolyzed by intracellular esterases to DCFH, which can be rapidly oxidized to highly fluorescent DCF in the presence of ROS. After cells were incubated with 1 mM FFAs together with GA for 24 hours, DCFH-DA was added to a final concentration of 5 mM and cells were incubated at 37°C in darkness for 30 minutes. DCF fluorescence intensity was detected with an emission wavelength of 530 nm and an excitation wavelength of 485 nm, using flow cytometry.

Statistical Analysis.

All results were expressed as mean ± standard error of the mean. One-way analysis of variance and Student t test were used to analyze the differences between different treatments. Statistics were performed using GraphPad Pro. P < 0.05 was considered statistically significant.

Results

Effect of GA on FFA-Induced Lipid Accumulation in HepG2 Cells.

It is generally believed that fatty liver results from an imbalance between the hepatic uptake of FFAs, TG synthesis, and excretion. Recently, Gomez-Lechon et al.24 validated an in vitro model of steatosis using HepG2 cells. HepG2 cells loaded with an FFA mixture containing oleate/PA (2:1 ratio) mimics benign chronic steatosis. FFA-overloaded HepG2 cells (1 mM FFAs) reached similar levels of maximal intracellular lipid accumulation as found in human liver with steatosis. Therefore, we used 1 mM FFAs (oleate/PA, 2:1) in the present study.

First, we examined whether GA was able to prevent FFA-induced lipid accumulation in HepG2 cells using Nile red staining. GA significantly reduced FFA-induced lipid accumulation (Fig. 2). CA-074 Me exhibited a stronger effect.

Figure 2.

Inhibition of FFA-induced lipid accumulation by GA in HepG2 cells. Human HepG2 cells were treated with 1 mM FFA together with various concentrations of GA (0 to 30 μM) or CA-074 Me (20 μM) for 24 hours. Intracellular lipids were stained with Nile red and measured with flow cytometry as described in Materials and Methods. The black curve represents normal control without FFA treatment. The red curve represents the cells treated only with FFA, while the lime, purple, yellow, and blue curves represent the cells treated with FFA together with 20 μM of CA-074 Me, or 3, 10, and 30 μM of GA, respectively. Flow cytometry diagram is representative of three independent experiments.

Inhibition of FFA-Induced Apoptosis by GA in HepG2 Cells.

We initially confirmed FFA-induced apoptosis in HepG2 cells. We treated the cells with 1 mM FFAs for 4, 8, 12, and 24 hours, and detected apoptotic cells by Annexin V-FITC/PI staining. As expected, FFA induced apoptosis, in a time-dependent fashion (data not shown). Then, we examined whether GA was able to inhibit FFA-induced apoptosis in HepG2 cells. The simultaneous treatment of cells with FFAs and GA for 24 hours dose-dependently inhibited FFA-induced apoptosis (Fig. 3A). CA-074 Me (20 μM), also partially attenuated FFA-induced apoptosis. We further used electron transmission microscopy to confirm the characteristic of cell apoptosis, such as disruption of nuclear integrity and lysosomal structure, chromatin margination and condensation, and lipid accumulation induced by 1 mM FFAs (Fig. 3B-b), and the protective effect of GA. GA significantly inhibited FFA-induced lipid accumulation and apoptosis (Fig. 3B-c,d). In rat primary hepatocytes, GA showed the similar protection against PA and stearate (SA)-induced apoptosis (Supplementary Fig. S1)

Figure 3.

Inhibition of FFA-induced apoptosis by GA in HepG2 cells. (A) Human HepG2 cells were treated with 1 mM FFAs together with various concentrations of GA (0 to 30 μM) or CA-074 Me (20 μM) for 24 hours, and stained with Annexin V-FITC and PI. The apoptotic and necrotic cells were detected by flow cytometry as described in Materials and Methods. The normal control, early apoptotic, late apoptotic, and necrotic cells were present in the lower left, lower right, upper right, and upper left quadrant, respectively. The percentage of cells in each quadrant is indicated. The flow cytometry diagram is representative of three independent experiments: (a) untreated cells; (b) cells treated with 1 mM FFAs; (c) cells treated with 1 mM FFAs and 3 μM GA; (d) cells treated with 1 mM FFAs and 10 μM GA; (e) cells treated with 1 mM FFAs and 30 μM GA; (f) cells were treated with 1 mM of FFAs and 20 μM CA-074 Me. (B) Representative transmission electron micrograph of ultrathin section of HepG2 cells treated with FFAs and GA. (a) normal HepG2 cell; (b) cells treated with 1 mM FFAs for 24 hours; (c) cells treated with 1 mM of FFAs and 3 μM GA; and (d) cells treated with 1 mM of FFAs and 10 μM GA. N, nucleus; M, mitochondria; L, lipid droplet.

Effect of HFD on Lipid Accumulation and Cathepsin B Expression in Rat Liver.

HFD-induced rat NAFLD models have been found to be useful models for human NAFLD.25 The hepatic histology of NAFLD can vary from isolated hepatic steatosis alone without histological evidence of hepatocellular damage to steatohepatitis with associated inflammation and hepatocyte damage. We fed the rats with HFD for 0, 4, 6, and 8 weeks. The histological analysis demonstrated a time-dependent induction of hepatic steatosis in these rats (Fig. 4A). After feeding with HFD for 8 weeks, the characteristics of NASH developed (Fig. 4A-d). It has been demonstrated that cathepsin B is an important mediator of FFA-induced hepatic injury.5, 6, 10, 11 To further confirm if HFD-induced lipotoxicity was correlated to the enzyme activity of cathepsin B in cytosol, we measured cytosolic cathepsin B enzyme activities in rat livers. Cathepsin B enzyme activities were significantly increased by HFD feeding, which was time-dependent (Fig. 4B). Analysis of the correlation of cathepsin B activity and the histopathological scores demonstrated that cathepsin B activity correlated positively with the extent of hepatic injury (P < 0.01) (data not shown). To determine if the increase of cytosolic cathepsin B activities was due to release of cathepsin B from lysosomes, we detected the protein levels of cathepsin B in cytosol by western blot analysis. The catalytically active fragments (p30 and p27) of cathepsin B in the cytosol were significantly increased after HFD feeding (Fig. 4C).

Figure 4.

Effect of HFD on induction of steatosis and cathepsin B expression in rats. Rats were fed with HFD for 0, 4, 6, or 8 weeks. (A) Photomicrographs of hematoxylin and eosin stained liver sections representing each of the four time points, respectively. (a) normal diet control group; (b) fed with HFD for 2 weeks; (c) fed with HFD for 4 weeks; and (d) fed with HFD for 8 weeks. (B) Effect of HFD on cathepsin B enzyme activity in rat liver. The enzyme activity of cathepsin B in rat liver cytosol was measured as described in Materials and Methods. Statistical significance relative to control group: *P < 0.05, **P < 0.01, ***P < 0.001. (C) Effect of HFD on cathepsin B expression. Representative immunoblots against cathepsin B and α-tubulin from the rat liver homogenates of each time point. α-tubulin was used as a protein loading control.

Prevention of Lipid Accumulation and Lipotoxicity by GA in HFD-Induced Rat NAFLD Models.

To examine whether GA has a protective effect on HFD-induced lipotoxicity in vivo, we gavaged rats with GA at a low dose (25 mg/kg) or high dose (50 mg/kg) for 8 weeks. GA had no significant effect on HFD-induced increase of serum TG, total cholesterol, and low density lipoprotein-cholesterol levels, but inhibited HFD-induced decrease of high density lipoprotein-cholesterol at high dose (Fig. 5A). GA inhibited HFD-induced increase of serum alanine aminotransferase and aspartate aminotransferase levels (Fig. 5B). Histological analysis indicated that HFD fed rats displayed significant steatosis, acinar and portal inflammation, infiltration of macrophages and lymphocytes, and lipid droplet accumulation (Fig. 5C-b). GA treatment significantly improved the hepatic histology. HFD-induced hepatic injury was decreased (Fig. 5C-c,d).

Figure 5.

Protective effects of GA on HFD-induced liver lipotoxicity in rats. Rats were fed with HFD and GA for 8 weeks as described in Materials and Methods. (A) Effect of GA on serum lipid levels of the rats treated with HFD. Each column represents the mean ± standard error of the mean (SEM), n = 6. Statistical significance relative to vehicle control: ##P < 0.01, ###P < 0.001; Statistical significance relative to HFD group: *P < 0.05. (B) Effect of GA on liver enzymes. Each column represents the mean ± SEM, n = 6. Statistical significance relative to vehicle control: ###P < 0.001; Statistical significance relative to HFD group: **P < 0.01, ***P < 0.001. (C) Photomicrographs of hematoxylin and eosin stained liver sections representing each of the four groups respectively. (a) normal diet control group; (b) HFD group; (c) HFD + GA 25 mg/kg group; (d) HFD+GA 50 mg/kg group. TG, triglycerides; TC, total cholesterol; HDL-C, high density lipoprotein-cholesterol; LDL-C, low density lipoprotein-cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase.

Effect of GA on Lysosomal Integrity in In Vitro and In Vivo NAFLD Models.

To identify the potential cellular mechanisms of the protective effect of GA on FFA/HFD-induced hepatic injury, we first examined the effect of GA on lysosomal stability using in vitro FFA-induced HepG2 models. By using the AO-uptake method, we were able to determine that FFAs treatment significantly induced lysosomal rupture as indicated by an increased number of “pale cells” (33.34%) (Fig. 6A-b). Simultaneous treatment with GA dose-dependently inhibited FFA-induced lysosomal rupture (Fig. 6A-c-e). GA exhibited better a protective effect than CA-074 Me on FFA-induced lysosomal rupture (Fig. 6A-f).

Figure 6.

Effect of GA on the lysosomal membrane stability. (A) Effect of GA on lysosomal membrane stability in HepG2 cells. Human HepG2 cells were incubated with 1 mM FFAs in the absence or presence of GA or CA-074 Me for 12 hours and then stained with AO. The fluorescence intensity was measured by flow cytometry as described in Materials and Methods. (a) normal HepG2 cells; (b) cells treated with 1 mM FFAs; (c-e) cells treated with 1 mM FFAs in the presence of 3, 10, and 30 μM of GA, respectively; (f) Cells treated with 1 mM FFAs and 20 μM of CA-074 Me. (B) Effect of GA on β-galactosidase activity in rat livers. The cytosol fraction and membrane fraction of liver tissue were separated and the β-galactosidase activity was measured as described in Materials and Methods. The enzyme activity in the cytosol fraction represents the enzymes released from lysosomes (free), the enzyme activity in the membrane fraction represents the enzymes retained in lysosomes (bound). Each column represents the mean ± standard error of the mean (SEM), n = 6. Statistical significance relative to vehicle control: #P < 0.05, ##P < 0.01; Statistical significance relative to FFA group: *P < 0.05.

To further ascertain the protective effect of GA, we examined the effect of GA on lysosomal integrity using in vivo rat NAFLD models by measuring free and bound β-galactosidase activities in rat liver. Free β-galactosidase activities were significantly increased, but the bound β-galactosidase activities were decreased after 8 weeks of HFD feeding in rat livers (Fig. 6B). Treatment with GA (50 mg/kg) significantly inhibited HFD-induced increase of free β-galactosidase activities and decrease of bound β-galactosidase activities, indicating that GA had a protective effect on lysosomal stability.

Effect of GA on Mitochondrial Stability in In Vitro and In Vivo NAFLD Models.

To determine if the mitochondrial pathway was also involved in the GA-mediated protective effect on lipotoxicity, we examined the effect of GA on mitochondrial stability by measuring membrane potential and cytosolic cytochrome c contents in HepG2 NAFLD models. Treatment with 1 mM FFAs for 24 hours significantly induced mitochondrial membrane depolarization as indicated by decrease of JC-1 aggregates and increase of JC-1 monomers (Fig. 7A-b). Treatment with GA or CA-074 Me for 24 hours, the FFA-induced mitochondrial membrane depolarization was dose-dependently inhibited (Fig. 7A-c-f).

Figure 7.

Effect of GA on mitochondrial membrane stability. (A) Human HepG2 cells were treated with 1 mM FFAs in the absence or presence of GA or CA-074 Me for 12 hours, and then stained with JC-1. The mitochondrial membrane potential was measured by flow cytometry as described in Materials and Methods. (a) Normal HepG2 cells; (b) cells treated with 1 mM FFAs; (c-e) cells treated with 1 mM FFAs in the presence of 3, 10, or 30 M of GA respectively; (f) cells treated with 1 mM FFAs in the presence 20 mM of CA-074 Me. Flow cytometry diagram is representative of three independent experiments. (B) Effect of GA on FFA-induced cytochrome c release in HepG2 cells. Cells were treated with 1 mM FFAs in the absence or presence of various concentrations of GA (0, 3, 10, and 30 μM) or 20 μM of CA-074 Me for 24 hours. The protein levels of cytosolic cytochrome c were detected by western blot analysis as described in Materials and Methods. α-Tubulin was used as a protein-loading control. Immunoblots are representative of three independent experiments. (C) Effect of GA on HFD-induced cytochrome c release in rat livers. Rats were treated with HFD or HFD + GA (25 mg/kg or 50 mg/kg) for 8 weeks. At the end of the experiment, the liver tissue was collected and homogenized in lysis buffer. The protein levels of cytosolic cytochrome c were detected by western blot analysis as described in Materials and Methods. α-Tubulin was used as a protein-loading control. Immunoblots are representative of four different experiments.

Release of mitochondrial cytochrome c into cytosol is another indicator of perturbation of mitochondrial membrane stability. We further detected cytochrome c protein levels in cytosol using western blot analysis. FFAs significantly increased cytosolic cytochrome c level, which was dose-dependently inhibited by GA in HepG2 cells (Fig. 7B). We further confirmed the effect of GA on PA-induced and SA-induced cytochrome c release in rat primary hepatocytes. GA markedly inhibited PA and SA-induced increase of cytosolic cytochrome c (Supplementary Fig. S2). Similarly, in in vivo rat NAFLD models, treatment with GA inhibited HFD-induced cytochrome c release from mitochondria (Fig. 7C).

Effects of GA on Cathepsin B Activity in In Vitro and In Vivo NAFLD Models.

It has been shown that cathepsin B plays a critical role in FFA-induced lipotoxicity.5, 6, 10, 11 To further explore the underlying mechanisms by which GA exerts its protective effect on lipotoxicity, we examined the effect of GA on cathepsin B activity and protein expression using both in vitro and in vivo NAFLD models. In an in vitro assay system, both GA and CA-074 dose-dependently inhibited cathepsin B enzyme activity (Fig. 8A). The median inhibitory concentration (IC50) for GA was 12.5 μM; the IC50 for CA-074 was 36.3 nM.

Figure 8.

Inhibition of cathepsin B activity and expression by GA in HepG2 cells. (A) Inhibition of cathepsin B activity by GA in in vitro cell-free assay. Cathepsin B isolated from human liver was incubated with cathepsin B specific substrate in the presence of various concentrations of GA or CA-074. The enzyme activity was determined as described in Materials and Methods. IC50 for GA is 12.5 μM; IC50 for CA-074 is 36.2 nM. (B) Inhibition of FFA-induced cathepsin B activity by GA in HepG2 cells. Cells were treated with 1 mM FFAs in the presence of various concentrations of GA (0, 3, 10, and 30 μM) or 20 μM of CA-074 Me for 24 hours; the cathepsin B activity in the cytosolic fraction was measured as described in Materials and Methods. Data represent the average of three independent experiments. Statistical significance relative to vehicle control: #P < 0.05; Statistical significance relative to FFA group: *P < 0.05. (C) Effect of GA on cathepsin B expression in HepG2 cells. Cells were treated as described above. The protein levels of cathepsin B in the cytosol were detected by western blot analysis using specific antibody to cathepsin B as described in Materials and Methods. α-Tubulin was used as protein loading control. Immunoblots are representative of three different experiments.

In HepG2 cells, FFAs not only increased cathepsin B enzyme activity, but also increased protein levels in cytosol (Fig. 8B and C). Increase of both active fragments of cathepsin B (p30 and p27) in the cytosol after treatment with 1 mM FFAs for 24 hours provided further evidence that FFAs induced cathepsin B release from lysosome to cytosol. Treatment with GA (10 and 30 μM) or CA-074 Me (20 μM) significantly inhibited FFA-induced cathepsin B expression and enzyme activity. GA also inhibited PA-induced and SA-induced cathepsin B expression in rat primary hepatocytes (Supplementary Fig. S3). In contrast, both GA and CA-074 Me had no effect on cathepsin B activity in untreated HepG2 cells (data not shown).

Similar to the observation in in vitro HepG2 models, treatment with GA also dose-dependently inhibited HFD-induced increase of cathepsin B expression and enzyme activities in rat NAFLD models. Treatment with 50 mg/kg GA not only almost attenuated HFD-induced cathepsin B expression, but also markedly inhibited the enzyme activity (Fig. 9).

Figure 9.

Inhibition of HFD-induced cathepsin B expression in rat liver. Rats were fed with HFD and GA for 8 weeks as described in Materials and Methods. (A) Effect of GA on HFD-induced expression of cathepsin B in liver. Representative immunoblots against cathepsin B and α-tubulin from the rat liver homogenate of each group. α-tubulin was used as a protein loading control. (B) Immunofluorescent staining of cathepsin B in liver tissue. (a) Control group; (b) HFD group; (c) HFD+GA 25 mg/kg; (d) HFD+GA 50 mg/kg; (e) negative control for immunofluorescent staining. Expression of cathepsin B was detected using specific antibody against cathepsin B as described in Materials and Methods. (C) Effect of GA on HFD-induced enzyme activity of cathepsin B in liver. The cathepsin B activity of liver homogenate was measured as described in Materials and Methods. Statistical significance relative to vehicle control: ###P < 0.001; Statistical significance relative to HFD group: *P < 0.05, ***P < 0.001.

The Effect of GA on Intercellular ROS.

It has been shown that progression of hepatic steatosis into steatohepatitis can be promoted by oxidative stress.26 Intracellular ROS plays an important role in lipid peroxidation, inflammatory cytokine induction, and mitochondrial dysfunction.1 It also has been reported that GA has anti-inflammatory and antioxidant activities13, 27, 28 and FFAs induce oxidative stress in liver. To examine whether GA was able to reduce FFA-induced oxidative stress, we measured the ROS in HepG2 cells. FFA-induced increase of ROS was significantly inhibited by GA and the effect was dose-dependent (Fig. 10); however, CA-074 Me had no significant effect.

Figure 10.

Inhibition of ROS by GA in HepG2 cells. Human HepG2 cells were treated with vehicle control (dimethylsulfoxide [DMSO]) or 1 mM FFAs together with GA (3 or 10 μM) or CA-074 Me (20 μM) for 24 hours. Cells were incubated with DCFH-DA (5 mM) for 30 minutes at room temperature in the dark. The fluorescence intensity of DCF was measured by flow cytometry. (A) Normal control, cells were treated with DMSO. (B) Positive control: cells were treated with H2O2 before incubation with DCFH-DA. (C) Cells were treated with 1 mM FFAs. (D,E) Cells were treated with 1 mM FFAs and 3 or 10 μM of GA. (F) cells were treated with 1 mM FFAs and 20 μM of CA-074 Me.

Discussion

FFAs are the key players in the pathogenesis of NAFLD. The elevated FFAs levels are closely correlated to the disease severity of NAFLD.4 Although it has been found that insulin resistance is a major risk factor for NAFLD, and a recent proof-of-concept study indicated that diet plus pioglitazone led to metabolic and histological improvement in NASH patients, more studies are needed to assess the long-term beneficial effect of insulin-sensitizing agents on the prevention of disease progression of NAFLD.29 Other significant factors leading to progressive liver injury remain to be identified and development of new therapeutic interventions is especially urgent.

An increasing amount of attention has been paid to the use of complementary and alternative medicine as a part of the treatment for liver diseases and the complications associated with current therapies. Licorice root has been used in Chinese medicine for thousands of years to reduce liver injury associated with a number of clinical disorders including chronic hepatitis C infection.13, 30 It also has been reported that GA is a potent inhibitor of bile acid-induced cytotoxicity in hepatocytes.28 Although several hypotheses have been proposed, such as inhibition of oxidative stress, anti-inflammatory activities, and immunomodulation, the precise cellular and molecular mechanisms accounting for its clinical benefits have not been identified. Whether GA has any beneficial effects on FFA-induced hepatic toxicity remains to be characterized.

In the present study, we provided the first evidence for the protective effect of GA on FFAs/HFD-induced lipotoxicity both in in vitro HepG2 and in in vivo NAFLD models. GA not only reduced FFA/HFD-induced lipid accumulation, but also significantly inhibited FFA/HFD-induced apoptosis.

The lysosomal compartment is responsible for the recycling of organelles and long-lived proteins. During the last decade, numerous studies have found that lysosomes are very vulnerable to various stress stimuli and release of lysosomal constituents into the cytosol may be an early or initiating event in apoptosis induced by oxidative stress, oxidized lipids and p53.7, 19 It is well accepted that lysosome-associated apoptosis is mediated by specific proteases; that is, cathepsin. Although there are about a dozen identified cathepsins in mammalian lysosomes, cathepsin B and D are the most prominent and stable under physiological conditions and are essential in hepatocyte apoptosis induced by bile-acids, TNF-α, and oxidative stress.26 It has been reported that inactivation of cathepsin B attenuates hepatic injuries induced by cholestasis, TNF-α, and cold ischemia–warm reperfusion.6, 10, 11 Redistribution of cathepsin B from lysosomes to the cytosol was also observed in liver biopsy specimens of NAFLD patients.5 Consistent with a previous report,5 we also found that FFAs significantly increased cytosolic cathepsin B protein levels and enzyme activities, which were inhibited by GA and the specific cathepsin inhibitor, CA-074 Me, in HepG2 cells. In HFD-fed rat NAFLD models, GA also significantly inhibited HFD-induced cathepsin B release and enzyme activities. Furthermore, GA not only prevented the FFA-induced release of cathepsin B into cytosol, but also directly inhibited cathepsin B enzyme activity with an IC50 of 12.5 μM.

Previous studies have demonstrated that FFAs promote hepatic lipotoxicity by stimulating TNF-α expression via a lysosomal pathway, activating mitogen-activated protein kinase cascades, especially JNK, and inducing oxidative stress.2, 5 In the current study, we found that GA inhibited FFA-induced expression of proinflammatory cytokines (TNF-α and interleukin-6) in macrophages (Supplementary Fig. S4). In addition, GA significantly inhibited FFA-induced JNK-2 activation both in HepG2 cells and primary rat hepatocytes (Supplementary Fig. S5). GA also inhibited PA-induced and SA-induced JNK activation in primary rat hepatocytes (Supplementary Fig. S6). These results suggest that GA exhibited its protective effect mainly by stabilizing lysosomal integrity and inhibiting inflammatory response and cathepsin B activity.

Recent studies indicate that released lysosomal enzymes promote mitochondrial ROS production and cytochrome c release. In the present study, we also observed that FFAs significantly induced cytochrome c release and enhanced mitochondrial ROS production, which were also inhibited by GA both in in vitro and in in vivo NAFLD models. However, the exact mechanisms by which FFAs induce lysosomal/mitochondrial dysfunction remain to be identified. Our preliminary studies suggest that activation of the endoplasmic reticulum stress and innate immune response may be involved in FFA-induced lipotoxicity. Whether GA is able to modulate endoplasmic reticulum stress and innate immunity remains to be identified and is the focus of our ongoing project.

In summary, stabilization of lysosomal membrane, inhibition of cathepsin B expression and enzyme activity, and reduction of mitochondrial cytochrome c release represent the major cellular mechanisms that might account for GA's protective effects on FFA-induced hepatic lipotoxicity.

Ancillary