Panax ginseng Meyer (P. ginseng) is one of the most widely used medicinal herbs with a long history in East Asia (Yun 2001). P. ginseng is well known to improve brain function, enhance immune system function and liver function, and have analgesic, anti-diabetic effects, and anti-oxidative effects (Choi 2008). Ginsenosides, compounds found in P. ginseng, contribute to these pharmacological effects (Ma et al. 2014). Ginsenosides comprise natural product steroid glycosides and triterpene saponins that belong to the steroidal family (Attele et al. 1999). They can be obtained from the roots, berries, and leaves of P. ginseng (Kim et al. 2012). Among the approximately 40 ginsenosides identified, ginsenoside Re is the most abundant ginsenoside, and Rb1, Rg1, Rc, Re, and Rd are the most common ones (Kim et al. 2009).

Due to the various pharmacological activities of P. ginseng, its roots are widely produced commercially as crude drugs or functional foods (Yang et al. 2016). Ginseng promotes the detoxification of toxic substances, protects against liver damage, and promotes the regeneration and recovery of the liver (Choi 2008). Ginsenoside Re is a saponin mainly isolated from ginseng berry extract (GBE). GBE has been shown to improve hyperglycemia and hyperlipidemia (Quan et al. 2012). In this study, however, we used ultrasonication-processed ginseng berry extract (UGBE). The composition ratio of P. ginseng berry extract changes after the ultrasonication process. While the major compound of GBE is ginsenoside Re, the main components of UGBE are ginsenosides Rk1, Rg2, Rg3, Rh1, and F4 after the ultrasonication process (Nam et al. 2018). Ginsenoside Rk1 is known to have anti-tumor activity in hepatocellular carcinoma cells and inhibits telomerase activity and cell growth (Kim et al. 2008; Ko et al. 2009). Ginsenoside Rg3 induces antioxidative and hepatoprotective effects by inhibiting heme oxygenase-1 (HO-1) expression and protects against kidney and liver damage associated with reactive oxygen species (ROS) (Lee et al. 2012a). The antiallergic action and anti-inflammatory activity of ginsenoside Rh1 have been linked to the inhibition of inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) protein expression (Park et al. 2004). Apoptosis-inducing effect of ginsenoside F4 is thought to have a role in its organ protection (Chen et al. 2013).

The pathophysiological mechanism of alcoholic liver disease (ALD) is not known yet, but several factors have been shown to influence alcohol-induced liver damage in ALD, including oxidative stress, toxic alcohol metabolites, fat accumulation, inflammation, and endotoxins (Kawaratani et al. 2013). As revealed through experiments using herbal extract, chronic ethanol-induced liver damage usually leads to secondary effects such as oxidative stress, fat accumulation, increase in the levels of endotoxins and cytokines, which increase the sensitivity of the liver to additional stress (Yoon et al. 2012). Chronic alcohol consumption can compromise the barrier function of the intestine and increase bacterial growth, and, in return, these proliferative bacteria increase plasma levels of endotoxin/lipopolysaccharide (LPS), thereby activating Toll-like receptor 4 (TLR4) signaling and pro-inflammatory cytokines such as tumor necrosis factor-α (TNF-α) (Zmijewski et al. 2014).

The major pathologic risk factor for alcohol-induced liver damage in ALD is oxidative stress. Alcohol-induced liver injury in an enteral alcohol-feeding model induces lipid peroxidation because of increased free radical generation and decreased levels of hepatic antioxidants, such as glutathione (Iimuro et al. 2000). In addition to glutathione, other liver antioxidants and enzymes, such as superoxide dismutase and catalase, can neutralize ROS. Increased production of ROS and lipid peroxidation can be associated with apoptotic cell death. The prevention of ethanol-induced apoptosis by inhibition of alcohol-induced oxidation emphasizes the importance of oxidative stress in the pathogenesis of ALD (Wu and Cederbaum 1999).

In the present study, we investigated the overall hepatoprotective effect of UGBE comparing it with silymarin and GBE in an ALD model in rats. To this end, the serum alanine transaminase (ALT), aspartate aminotransferase (AST), γ-glutamyl transferase (γ-GT), serum lipid parameters related to ALD, and pro-inflammatory cytokine levels were measured. Furthermore, the activity and expression levels of several enzymes and receptors related to oxidative stress and liver damage were evaluated.

Materials

GBE and UGBE were supplied by Prof. Sung Kwon Ko, Semyung University (Jecheon, Korea, Patent KR101688002B1). Four-year-old Korean ginseng berries were collected at Eumseong (Korea) on August 20, 2010. The specimens were stored at the Oriental Medical Food Research Laboratory, Semyung University. Ginsenoside standards, pure ethanol, and silymarin were purchased from Chromadex (Irvine, CA, USA), Sigma-Aldrich Co., LLC. (St. Louis, MO, USA). Dulbecco’s phosphate-buffered saline (DPBS) was purchased from Welgene, Inc. (Seoul, Korea). The alanine transaminase (ALT), aspartate aminotransferase (AST), gamma-glutamyl transferase (γ-GT), glutathione peroxidase (GPx), catalase (CAT), superoxide dismutase (SOD) activity kits and TNF-α enzyme-linked immunosorbent assay (ELISA) kit were purchased from Biovision, Inc. (San Francisco, CA, USA). The TLR4 ELISA kit and heme oxygenase-1 (HO-1) ELISA kit were purchased from CUSABIO (Wuhan, CN), Enzo Life Sciences, Inc. (NY, USA). Other essential materials were purchased from Sigma-Aldrich Co., LLC.

Preparation of UGBE

P. ginseng berries grown for 4 years were dried, and 2,000 mL of ethyl alcohol was added per 200 g of dried P. ginseng berries. The reflux was extracted twice, filtrated, and concentrated by vacuum evaporation to produce GBE. GBE was processed in an ultrasonicator (KODO, Hwaseong, Korea) run at 600 W at 100˚C for 10 hours. The residual solution was concentrated by vacuum evaporation and freeze-dried to obtain UGBE, a brownish extract. For further analysis of UGBE and GBE, 2 g of each extract was extracted three times with 50 mL of n-butanol using an ultrasonicator (KODO, Hwaseong, Korea), and then the supernatant was removed. The residue was treated three times in 50 mL of n-butanol again. The n-butanol fraction remaining in the ultrasonicator was filtered and concentrated by a vacuum evaporator.

Animal model

Male Sprague-Dawley (S.D.) rats (specific pathogen-free; body weight, 200-250 g) were used in the experiments (Samtako Bio, Osan, Korea). Animals were group-housed with direct bedding and were provided with filtered tap water and a normal laboratory diet from Samtako bio. The rats were housed in pathogen-free cages at 24-25℃ and 70-75% humidity. The rats were starved for 24 hours before the experiments, unrestricted water was provided during this period. All procedures of animal experiments were permitted by the Institutional Animal Care and Use Committee (IACUC) of Chung-Ang University, Seoul, Korea (CAU IACUC-20140031).

Experimental design

Sixty-three S.D. rats were randomly divided into seven experimental groups. The rats assigned to each experimental group received ethanol or experimental solution as follows: control group (10 mL/kg of saline), ALD group (5 g/kg of ethanol), silymarin group (5 g/kg of ethanol + 150 mg/kg of silymarin), GBE group (5 g/kg of ethanol + 250 mg/kg GBE), and UGBE groups (5 g/kg of ethanol + 100, 250 or 500 mg/kg of UGBE). Alcohol exposure for ALD induction was performed by oral administration of 40% ethanol at 5 g/kg. The control group received a daily intragastric administration of 10 mL/kg of saline. Experimental solutions such as silymarin, GBE, and UGBE were orally delivered 30 minutes after alcohol administration. Each substance was orally administered once a day for 6 weeks. For all experimental animals, food consumption, water intake, and body weight were measured daily between 12:00 pm and 2:00 p.m. Twenty-four hours after the last administration, all rats were euthanized by cervical dislocation, and the blood samples, liver tissues, and epididymal fat pad were excised immediately. Blood was collected through the hepatic vein, and the weights of the epididymal fat pad and liver were recorded. Blood samples were coagulated in serum separator tubes (Becton, Dickinson and Company, NJ, US) at room temperature and centrifuged (15,000 × g, 4℃, 20 minutes). Centrifuged serum samples were used for analysis. The part of the liver was fixed with 10% formalin for hematoxylin-eosin assays. The remaining liver tissues were washed with PBS buffer and flash frozen at −80℃ immediately for tissue assays.

Preparation and biochemical assays of liver and serum samples

Blood samples were obtained from the rat inferior vena cava after euthanization. Whole blood samples were stored in SST II Plus plastic serum tubes (Becton, Dickinson, and Company, US). The stored blood samples were centrifuged at 10,000 × g at 4℃ for 20 min. Serum samples obtained via centrifugation were immediately flash frozen at −80℃ to be used in AST and ALT level assays. All methods in this experiment were performed according to the manufacturer’s recommendations. Red blood cells were removed from the liver by perfusion with pH 7.4 PBS solution through the portal vein. The portal vein was cannulated with a 23-gauge IV catheter (Korean vaccine, KO), and the abdominal inferior vena cava was immediately removed. The liver was then removed from the body and washed with saline. The liver samples thus obtained were immediately flash frozen at −80℃ for anti-oxidative effect analysis and inflammation assays (anti-oxidative effect; SOD, GPx, CAT, and HO-1, inflammation; TNF-α, a receptor; TLR-4). The procedures for all of the above analyses were performed according to the manufacturer’s instructions.

Protein assay

The protein concentrations of the serum samples and liver homogenate supernatants were measured spectrophotometrically using the Bio-Rad assay (Bio-Rad Chemical Division, Richmond, CA, US).

Assessment of liver function and lipid profiles

The serum ALT and AST activities, indicators of liver function, were measured as suggested by International Federation of Clinical Chemistry (IFCC) methods using Beckman-Coulter reagents (Beckman-Coulter, Ireland) (Sung et al. 2007). Serum λ-GT levels, another indicator of liver function, were measured by a kit according to the manufacturer’s instructions. Total cholesterol, LDL, and HDL levels were measured to determine the lipid profiles according to the manufacturer’s instructions using a kit.

Measurement of hepatic antioxidant enzyme activities

A colorimetric assay was used to measure the antioxidant enzyme activities of SOD, GPx, and CAT. Before the experiment, perfusion was performed with DPBS to remove all of the remaining red blood cells from the liver tissues before freezing at −80℃. Following the assay protocol provided by each antioxidant enzyme kit, the liver tissues (40-100 mg) were homogenized in a buffer or ice-cold 0.1 M Tris/HCl, pH 7.4 containing 0.5% Triton X-100, 5 mM β-ME, 0.1 mg/mL PMSF. After centrifugation at 21,000 × g at 4℃ for 10 minutes, the supernatants were collected and used in the assays. All other procedures were performed according to the manufacturer’s instructions.

Measurement of serum TNF-α levels and hepatic HO-1 expression

ELISA was used to measure the TNF-α and hepatic HO-1 protein levels in the blood. Serum samples for the TNF-α analysis were directly analyzed. The red blood cells were removed by perfusing liver tissue with DPBS and flash-frozen in liquid nitrogen. Fully frozen tissues were ground. An appropriate volume of reagent (0.1 mM PMSF, 1 μg/mL aprotinin, 1 μg/mL leupeptin, and 1 μg/mL pepstatin) for HO-1 extraction was added to the tissue powder when homogenized. After centrifugation at 21,000 × g at 4℃ for 10 minutes, the supernatants were collected and used for analysis. All other procedures were performed according to the manufacturer’s instructions.

Measurement of TLR4 protein expression

ELISA was used to measure the hepatic protein expression of TLR4. About 100 mg of liver tissue was washed and homogenized with 1× PBS and kept at 20℃ overnight. A freeze-thaw cycle was performed twice to destroy the cell membranes. After centrifugation at 5000 × g at 2-8℃ for 5 minutes, the supernatant was removed and analyzed immediately. All other procedures were performed according to the manufacturer’s instructions.

Immunohistochemistry

Red blood cells were removed from the liver by perfusion with pH 7.4 PBS solution through the portal vein. Then, the portal vein was cannulated with a 23-gauge IV catheter (Korean vaccine, KO), and the abdominal inferior vena cava was immediately removed. The liver sample was then removed, washed with saline, and immersed in 10% formalin at room temperature for 2 weeks for immunohistochemistry. The immersed liver samples were embedded in paraffin and cut into 5-μm-thick sections by a microtome. A secondary antibody (Dako Real^TM EnVision^TM Detection System Rabbit/Mouse; diluted 1:200) was used to observe TLR4 expression in immunohistochemistry assays. After development with diaminobenzidine, tissue sections were placed on poly-lysine gelatinized glass slides and dehydrated through graded ethanol solutions before the placement of coverslips. The stained tissues were observed on a Leica DMR 6000 microscope, and the images were taken at 20× with a Leica DM 480 camera (Wetzlar, Germany).

Hematoxylin and eosin (H&E) staining

H&E staining was performed to detect liver damage caused by ethanol. The liver sample was embedded in paraffin and cut into 5-μm-thick sections using a microtome. The prepared tissue samples were stained with H&E and observed under a Leica DMR 6000 microscope. The images were taken with a Leica DM 480 camera (Wetzlar, Germany). The images presented were taken at 20× magnification for low-magnification images and 40× magnification for high-magnification images.

Statistical analysis

The data are expressed as the mean ± S.D. of data collected in independent experiments. One-way ANOVA was used for statistical analyses, and p^-values less than 0.05 were considered to indicate statistically significant differences between the groups compared.

Liver and epididymal fat pad weights

Changes in body weight, food, and water intake during the study period did not show significant differences among the treatment groups (data not shown). We noted a small but not significant difference in liver weights, and there were no significant differences in the ratios of liver weight to body weight (Fig. 1A). However, both the total weight and relative weight of the epididymal fat pad showed a significant difference between the control group and ethanol treatment group. In the ethanol-treated group, the ratio was significantly increased, but in GBE and UGBE (500 mg/kg) groups, it was significantly lower compared with the Ethanol-treated group (Fig. 1B).

Figure 1. The weights of liver and epididymal fat pad samples relative to total body weight. (A) Liver weight relative to body weight (L/B). (B) Epididymal fat pad weight relative to body weight. Data are represented as the mean ± S.D (n = 9). *p < 0.05 compared to controls; #p < 0.05 compared to ethanol group. Control, control rats; Sham, sham-operated control rats; ethanol, rats treated with ethanol; Sil, ALD rats treated with silymarin (150 mg/kg); GBE, ALD rats treated with GBE (250 mg/kg); UGBE 100, ALD rats treated with UGBE (100 mg/kg); UGBE 250, ALD rats treated with UGBE (250 mg/kg); UGBE 500, ALD rats treated with UGBE (500 mg/kg).

UGBE affects the lipid profile in ethanol-induced liver injury model

To evaluate the effect of UGBE on lipid parameters, serum total cholesterol, LDL, and HDL levels were measured (Fig. 2). While the levels of LDL and total cholesterol were significantly elevated by ethanol consumption, and HDL levels were significantly reduced. However, LDL and total cholesterol levels in GBE and UGBE (250 mg/kg, 500 mg/kg) groups were significantly lower than those in the ethanol group. In addition, HDL levels were significantly higher in the GBE and UGBE (500 mg/kg) groups than those in the ethanol group. Overall, lipid parameters were improved in the UGBE high-dose group compared with the ethanol group in the rat ALD model.

Figure 2. Effect of UGBE on lipid parameters in ethanol-induced liver injury model. Control, control rats; Sham, sham-operated control rats; ethanol, rats treated with ethanol; Sil, ALD rats treated with silymarin (150 mg/kg); GBE, ALD rats treated with GBE (250 mg/kg); UGBE 100, ALD rats treated with UGBE (100 mg/kg); UGBE 250, ALD rats treated with UGBE (250 mg/kg); UGBE 500, ALD rats treated with UGBE (500 mg/kg). Data represent mean ± SD (n = 9). *p < 0.05 compared to controls; ^#p < 0.05 compared to ethanol group.

UGBE improved the liver dysfunction in the ethanol-induced liver injury model

The serum ALT, AST, and γ-GT levels were assessed to evaluate liver function and found to be significantly elevated in the ethanol group compared with the control group (Fig. 3). However, they were significantly lower in the silymarin group as well as in the UGBE group compared with the ethanol group. Especially, there was a significant difference in the γ-GT levels in the UGBE high-dose groups (250 mg/kg, 500 mg/kg). In addition, the UGBE 250 mg/kg group and the silymarin group showed similar results, but the liver function parameters in UGBE 500 mg/kg group were better than those in the silymarin group. These results suggest that UGBE ameliorates the deterioration in liver function and the liver damage induced by ethanol administration.

Figure 3. Effect of UGBE on AST, ALT, and γ-GT levels in ethanol-induced liver injury model. Control, control rats; Sham, sham-operated control rats; ethanol, rats treated with ethanol; Sil, ALD rats treated with silymarin (150 mg/kg); GBE, ALD rats treated with GBE (250 mg/kg); UGBE 100, ALD rats treated with UGBE (100 mg/kg); UGBE 250, ALD rats treated with UGBE (250 mg/kg); UGBE 500, ALD rats treated with UGBE (500 mg/kg). Data are represented as the mean ± SD (n = 9). *p < 0.05 compared to controls; ^#p < 0.05; ^##p < 0.01; ^###p < 0.005 compared to ethanol group.

The antioxidant effects of UGBE in ethanol-induced liver injury model

Activity tests for the hepatic SOD, GPx, and CAT were performed to examine the antioxidant effects of the treatments (Fig. 4). In the ethanol-treated group, all enzymatic activities tested were significantly reduced compared with the control group. In the silymarin group and the UGBE high-dose groups (250 mg/kg, 500 mg/kg), antioxidant enzyme activities were significantly higher than those in the ethanol group. However, in the GBE group, only SOD activity was significantly higher than that in the ethanol group. Therefore, UGBE shows antioxidative effects similar to those of silymarin and better than those of GBE in the ethanol-challenged liver.

Figure 4. The hepatic SOD, GPx, and CAT activities in different treatment groups. Control, control rats; Sham, sham-operated control rats; ethanol, Rats treated with ethanol; Sil, ALD rats treated with silymarin (150 mg/kg); GBE, ALD rats treated with GBE (250 mg/kg); UGBE 100, ALD rats treated with UGBE (100 mg/kg); UGBE 250, ALD rats treated with UGBE (250 mg/kg); UGBE 500, ALD rats treated with UGBE (500 mg/kg). Data are represented by the mean ± SD (n = 9). **p < 0.01; ***p < 0.005 compared to controls; ^#p < 0.05; ^##p < 0.01; ^###p < 0.005 compared to ethanol group.

Increases in serum levels of TNF-α and hepatic HO-1 in ethanol-induced liver injury model were inhibited by UGBE

Confirming the inflammatory phenotype, serum levels of TNF-α were significantly higher in the ethanol group compared with the control group (Fig. 5). However, in silymarin and the UGBE groups, TNF-α levels were significantly lower than those in the ethanol group. TNF-α levels decreased with increasing doses of UGBE, and the UGBE 500 mg/kg group exhibited TNF-α levels comparable to those in the silymarin group.

Figure 5. Serum TNF-α levels. Control, control rats; Sham, sham-operated control rats; ethanol, Rats treated with ethanol; Sil, ALD rats treated with silymarin (150 mg/kg); GBE, ALD rats treated with GBE (250 mg/kg); UGBE 100, ALD rats treated with UGBE (100 mg/kg); UGBE 250, ALD rats treated with UGBE (250 mg/kg); UGBE 500, ALD rats treated with UGBE (500 mg/kg). Data are represented as the mean ± SD (n = 9). ***p < 0.005 compared to controls; ^#p < 0.05; ^##p < 0.01; ^###p < 0.005 compared to ethanol group.

HO-1 levels in the ethanol group were higher than those in the control group, as expected (Fig. 6). In the silymarin and the GBE groups, the HO-1 levels were not significantly different from those in the ethanol group, but UGBE high-dose groups (250 mg/kg, 500 mg/kg) exhibited significantly higher HO-1 levels than those in the ethanol group. Also, as the UGBE dose increased, the HO-1 levels tended to increase. Overall, these results confirmed that UGBE effectively inhibits the inflammatory response to ethanol-induced hepatic injury, and it could be suggested that it contributes to the enhancement of antioxidative capacity by upregulating HO-1 protein levels.

Figure 6. Hepatic HO-1 levels. Control, control rats; Sham, sham-operated control rats; ethanol, Rats treated with ethanol; Sil, ALD rats treated with silymarin (150 mg/kg); GBE, ALD rats treated with GBE (250 mg/kg); UGBE 100, ALD rats treated with UGBE (100 mg/kg); UGBE 250, ALD rats treated with UGBE (250 mg/kg); UGBE 500, ALD rats treated with UGBE (500 mg/kg). Data are represented as the mean ± SD (n = 9). *p < 0.05 compared to controls; ^#p < 0.05; ^##p < 0.01 compared to ethanol group.

UGBE down-regulated TLR4 protein expression in the ethanol-induced liver injury model

As expected, TLR4 expression levels in the ethanol group were significantly higher than those in the control group. Compared with the ethanol group, TLR4 levels were significantly lower only in UGBE 500 mg/kg group. In other experimental groups, TLR4 levels were lower than those in the ethanol group, but the difference was not statistically significant. As assessed by immunohistochemical analyses, TLR4 protein levels in hepatocytes and interface stem cells were significantly elevated upon ethanol administration but were almost undetectable after UGBE 500 mg/kg treatment (Fig. 7). Thus, it can be proposed that UGBE affects the TLR4 signaling pathway.

Figure 7. Effect of UGBE on TLR4 expression and liver injury in ethanol-induced liver injury model. TLR4 protein was shown in the figure with brown color. (A) TLR4 levels in hepatocytes in the ethanol group. (B) TLR4 levels in interface stem cells in the ethanol group. (C) TLR4 levels in hepatocytes in ethanol + UGBE 500 mg/kg group. (D) TLR4 levels in interface stem cells in ethanol + UGBE 500 mg/kg group. Control, control rats; Sham, sham-operated control rats; ethanol, Rats treated with ethanol; Sil, ALD rats treated with silymarin (150 mg/kg); GBE, ALD rats treated with GBE (250 mg/kg); UGBE 100, ALD rats treated with UGBE (100 mg/kg); UGBE 250, ALD rats treated with UGBE (250 mg/kg); UGBE 500, ALD rats treated with UGBE (500 mg/kg). Data are represented as the mean ± SD (n = 9). ***p < 0.005 compared to controls; ^###p < 0.005 compared to ethanol group.

Histological changes in the ethanol-induced liver injury model were ameliorated by UGBE

H&E staining results also supported the observed effect of UGBE treatment (Fig. 8). UGBE treatment with a dose of 500 mg/kg attenuated severe ethanol-induced abnormalities in hepatic lobules such as fat accumulation.

Figure 8. H&E staining in ethanol-induced liver injury model. (A) Control group. (B) Ethanol group, (C) ethanol + UGBE 500 mg/kg group. The red arrow indicated the fat accumulation in liver tissue. Control, control rats; ethanol, Rats treated with ethanol; UGBE 500, ALD rats treated with UGBE (500 mg/kg).

Due to its beneficial pharmacological effects, P. ginseng has been used as a crude drug or medicine (Wang et al. 2014). In particular, there have been several studies on ginseng berry and the positive effects of ginsenoside Re, a major component of ginseng berry, on hyperlipidemia and hyperglycemia (Quan et al. 2012). Recently, studies have been conducted to improve the effects of crude drugs by changing the composition of active ingredients through ultrasonication. Ultrasonication of the ginseng berry resulted in a significant increase in the composition ratios of ginsenosides Rk1, Rg2, Rg3, Rh1, and F4, confirming the HPLC results obtained from our previous study (Nam et al. 2018). Rh1 and Rg2 are known to affect the acidity and microbiota of the stomach in normal conditions (Peng et al. 2012). It has also been suggested that the components of UGBE intervene with bacterial pathways, such as fermentation and other enzymatic activities (Ryu et al. 2013). The major components of UGBE, i.e., Rg2, Rg3, Rh1, Rk1, and F4 are similar to those of red ginseng (Kim et al. 2011; Lee et al. 2012b), which protect against liver damage through various mechanisms (Igami et al. 2015).

In previous studies, ginsenoside Rk1 has been reported to induce apoptosis and inhibit hepatocellular carcinoma (Kim et al. 2008; Ko et al. 2009); ginsenoside Rg3 to improve chemosensitivity and to protect against ROS-induced tissue damage via HO-1 upregulation (Lee et al. 2012a); Rh1 to have anti-allergic and anti-inflammatory effects presumably through cell membrane stabilization and inhibition of iNOS and COX-2 expression (Park et al. 2004); and ginsenoside F4 to induce apoptosis in human lymphocytoma by regulating mitochondrial function (Chen et al. 2013).

We have shown that the processing of GBE by ultrasonication results in a change in the composition of ginsenosides and that the resultant composition may be more effective than ginsenoside ratios otherwise observed in red ginseng. However, UGBE has been studied only recently, and the effect of UGBE through these ginsenosides is unknown. Based on the increased ginsenoside components in UGBE, it has been thought that they would have a protective effect against liver damage. Therefore, in this study, we aimed to evaluate the protective effects of UGBE against liver injury. In addition to UGBE, we included silymarin, which is traditionally known for liver protection, and GBE without ultrasonication as positive controls and for comparison.

In the present study, liver injury was induced by oxidative stress and inflammatory reaction through chronic ethanol treatment, producing a rat model of ALD. The ALD model characteristically shows increases in serum levels of ALT, AST, and λ-GT due to hepatic tissue damage (Cha et al. 2013). We observed that ALT, AST, and λ-GT levels in the high-dose UGBE group were significantly lower than those in the ethanol group and comparable to those in the control group. Moreover, the ALT, AST, and λ-GT levels in UGBE 250 mg/kg group were similar to those in the silymarin group and lower than those in GBE 250 mg/kg group. In other words, UGBE showed hepatoprotective effects similar to those observed in the silymarin group and stronger than those observed in the GBE group of the same dose (250 mg/kg). The differences in the hepatoprotective effects of equal concentrations of GBE and UGBE can be attributed to the differences in the ratios of ginsenoside components introduced by the process of ultrasonication.

Cholesterol levels are not regulated properly in cases of impaired liver function induced by chronic alcohol administration, resulting in altered lipid parameters (Ramirez et al. 2013). Such compromised cholesterol homeostasis is observed in alcoholic fatty liver disease, but may also occur in non-alcoholic fatty liver disease (Wouters et al. 2008; Caballero et al. 2009). Among the lipid parameters, the levels of LDL and total cholesterol in the UGBE high-dose group were significantly lower than those in the ethanol group and were similar to those in the control group. HDL levels in the UGBE high-dose group were significantly higher than those in the ethanol group and were similar to those in the control group. The restoring effect of UGBE on the ethanol-induced deterioration of lipid parameters was stronger as the UGBE dose increased. The extent of improvement of lipid parameters in the UGBE high-dose group was similar to that in the GBE group. It means that the impact of GBE on lipid parameters remains consistent regardless of ultrasonication. It can be explained by the main component of GBE, ginsenoside Re. Ginsenoside Re was known to improve hyperglycemia and hyperlipidemia by changing the signaling pathway regarding metabolism in liver cells (Quan et al. 2012). It suggested that GBE may reduce the lipid parameter by another mechanism compared to UGBE, which can reduce the lipid parameter by ameliorating liver injury such as oxidative and inflammatory damage.

The ratio of epididymal fat pad weight to body weight is an indicator of fatty liver disease. This ratio was higher in the ethanol group but not in the UGBE group, which exhibited a ratio similar to that in the control group. There was no difference in the weight of the liver but only in the weight of the epididymal fat pad, possibly because the epididymal fat pad has a small mass compared to the liver, therefore the change in weight due to fat accumulation may be more easily reflected in the epididymal fat pad.

Oxidative stress plays an important role, especially in the initiation and progression of hepatic damage in inflammatory liver diseases such as ALD. An enzyme-dependent antioxidant system plays a crucial role in ROS-associated liver damage (Ai et al. 2013; Jang et al. 2017a; Lee et al. 2017). Chronic alcohol consumption causes an imbalance between ROS production and elimination, impairing antioxidant capacity in the liver. The overproduction of ROS should be countered by antioxidative enzymes such as SOD, CAT, and GPx. In the ALD model, the levels of these antioxidative enzymes were found to decrease upon ethanol administration. In the silymarin group and the UGBE high-dose group, the antioxidative activity was maintained, indicating that UGBE had antioxidative effects similar to those of silymarin. While it is not known whether UGBE directly removes ROS, this study shows that UGBE increases the activities of antioxidant enzymes and mitigates oxidative stress in ethanol-induced hepatotoxicity.

TNF-α is known to be a major factor involved in the inflammatory mechanisms leading to liver damage in the ALD model and plays a crucial role in hepatotoxicity-mediated apoptosis (El-Beshbishy 2008). In the present study, the UGBE group exhibited TNF-α levels lower than those in the ethanol group, indicating that UGBE might be involved in anti-inflammatory mechanisms.

HO-1 is an enzyme that catalyzes the degradation of heme into ferrous iron and bilirubin (Yu et al. 2010). The role of LPS in ethanol-induced liver damage has been confirmed by several studies. LPS increases ROS production through the activation of Kupffer cells (Jang et al. 2017b). HO-1 protects cells from LPS-induced oxidative stress by improving microvascular perfusion of the liver (Roller et al. 2010). HO-1 is an antioxidant enzyme that plays an important role in cellular protection against inflammation through the regulation of intracellular homeostasis (Gomes et al. 2010; Paine et al. 2010; Yang et al. 2019). As shown previously, ethanol acts as an inducer of HO-1 (Liu et al. 2004) as a compensatory action. In this study, HO-1 levels were slightly elevated in the ethanol-treated group. In the UGBE high-dose group, the HO-1 levels were significantly higher than those in the ethanol-treated group. It means that UGBE high-dose group had more capability for anti-oxidative and anti-inflammatory effects by HO-1 to protect the liver than the ethanol-treated group. Therefore, it can be suggested that the hepatoprotective effect of UGBE is caused by a significant increase in HO-1 expression, which is thought to be induced by Rg3 (Lee et al. 2012a).

As mentioned earlier, LPS plays an important role in inducing liver damage in ALD, and LPS is a major ligand of TLR4. An earlier study demonstrated the crucial protective role of TLR4 against ALD (Hritz et al. 2008). The effect of TLR4 on the pathogenicity of ALD has been suggested to be associated with its function in the innate immune system (Inokuchi et al. 2011). Chronic alcohol consumption causes an imbalance of intestinal flora, resulting in changes in gut permeability and inducing bacterial overgrowth (Bishehsari et al. 2017). Such disturbance of the intestinal microbiome by ethanol is due to liver inflammation and injury as a result of increased LPS-mediated TLR4 upregulation (Szabo and Bala 2010). In this study, the expression levels of TLR4 were significantly elevated in the ethanol-treated group but significantly reduced upon UGBE (500 mg/kg) treatment, resulting in TLR4 levels similar to those in the control group. However, TLR4 is involved in two signaling pathways, and it is unclear which mechanism of action is affected by UGBE treatment. Nonetheless, it can be suggested that UGBE can provide liver protection by regulating TLR4 levels. These observations were also confirmed by immunohistochemistry.

In this study, we observed a protective effect of UGBE in a rat ALD model with ethanol-induced hepatic injury. The hepatoprotective effect of UGBE was stronger than that of silymarin, as judged by lipid parameters, and UGBE was more effective than GBE when applied at the same dose. Especially, UGBE induced higher levels of antioxidant activity compared with GBE, which may have a role in its enhanced hepatoprotective effect. Overall, UGBE is superior to GBE as measured by lipid parameters, liver function, inflammation index, and oxidative stress. This difference in efficacy seems to be due to the differences in the ratios of UGBE and GBE components. This hepatoprotective effect also appears to be suppressed by the expression of TLR4.

Although GBE (250 mg/kg) was selected as a positive control according to previous studies, a higher dose of GBE, 500 mg/kg for example, is expected to be evaluated in the future. In the present study, we used an ALD model, but the prevalence of the non-alcoholic fatty liver disease is also high. Although ALD and non-alcoholic fatty liver disease (NAFLD) are similar in pathology, infiltration of inflammatory cells is more common in ALD. However, since the antioxidant effects of UGBE may cause a significant hepatoprotective effect in NAFLD, studies on NAFLD are also expected in the future.

In conclusion, this study has shown that UGBE has a hepatoprotective effect in an ethanol-induced ALD model, and this effect is thought to be due to the changes in the ratios of ginsenoside components of GBE.

The authors declare that they have no conflict of interest.

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea and the Ministry of Education, Science and Technology (Grant number NRF-2019R1F1A1062070).