Hepatoprotective effect of ultrasonicated ginseng berry extract on a rat mild bile duct ligation model.

BACKGROUND: The Panax ginseng berry extract (GBE) is well known to have an antidiabetic effect. The aim of this study is to evaluate and investigate the protective effect of ultrasonication-processed P. ginseng berry extract (UGBE) compared with GBE on liver fibrosis induced by mild bile duct ligation (MBDL) model in rats. After ultrasonication process, the composition ratio of ginsenoside in GBE was changed. The component ratio of ginsenosides Rh1, Rh4, Rg2, Rg3, Rk1, Rk3, and F4 in the extract was elevated.
METHODS: In this study, the protective effect of the newly developed UGBE was evaluated on hepatotoxicity and neuronal damage in MBDL model. Silymarin (150 mg/kg) was used for positive control. UGBE (100 mg/kg, 250 mg/kg, 500 mg/kg), GBE (250 mg/kg), and silymarin (150 mg/kg) were orally administered for 6 weeks after MBDL surgery.
RESULTS: The MBDL surgery induced severe hepatotoxicity that leads to liver inflammation in rats. Also, the serum ammonia level was increased by MBDL surgery. However, the liver dysfunction of MBDL surgery-operated rats was attenuated by UGBE treatment via myeloid differentiation factor 88-dependent Toll-like receptor 4 signaling pathways.
CONCLUSION: UGBE has a protective effect on liver fibrosis induced by MBDL in rats through inhibition of the TLR4 signaling pathway in liver.

Introduction

Liver fibrosis is resulted from the excessive production and progressive accumulation of extracellular matrix protein, which results from chronic liver damage [1], [2]. It can be caused by multiple reasons such as drugs, disease (hepatitis), autoimmune response, and toxicants [3], [4]. Chronic liver damage results in the dysfunction of liver cell, which causes the accumulation of ammonia [5]. If theses conditions persist, it can develop into liver cirrhosis, hepatic encephalopathy (HE), and hepatocellular carcinoma [6], [7]. Currently, lactulose is mainly used to reduce HE-related symptoms [8]. Lactulose is nonabsorbable disaccharide that acidifies gut lumen through biodegradation and reduces ammonia production from gut bacteria [9]. Lactulose is a strong laxative that frequently induces diarrhea as a side effect [10]. Therefore, it is necessary to study a new agent that restores liver function fundamentally with lower side effect.

The pathophysiological conditions are highly related to the inflammation [11]. Several studies have investigated the links between inflammatory liver injury and liver fibrosis [12], [13]. As a result, pathogens such as lipopolysaccharide have been proven to play an important role in this progress of inflammation [14]. Lactulose is a well-known treatment option for HE condition.

Most pathogens are recognized by specific receptor such as toll-like receptors on the basis of particular molecular patterns [15]. Especially, toll-like receptor 4 (TLR4) plays an important role in the proinflammatory response by producing proinflammatory cytokines by upregulating nuclear factor–kappa B (NF-κB) nuclear translocation [16]. This TLR4-related inflammatory response is mediated by specific adapter protein, myeloid differentiation primary response gene 88 (Myd88) [17]. Lipopolysaccharide, an endotoxin found in the outer membranes of gram-negative bacteria, is the main ligand of TLR4 that elicits an innate immune response [18]. Bile duct ligation (BDL) surgery often leads to bacterial translocation of intestinal endotoxin that initiates the Myd88-dependent TLR4 signaling pathway and inflammatory responses and oxidative stress in biliary obstruction condition [19].

Since hepatic inflammation and liver fibrosis are very common diseases in humans, various animal experiment models have been established in the past decades, which can lead to inflammatory liver injury and acute liver failure [20], [21]. One of the well-known models for liver inflammation is the BDL model. The ligation of common bile ducts can induce severe hepatic damage due to the excessive storage of bile acid in the liver [22]. Under BDL surgery, most animals experience severe liver injury such as liver fibrosis with morphological changes. BDL surgery is a widely accepted and used model that induces chronic liver disease in rats [23]. Sometimes, mild BDL (MBDL) surgery is applicable for long-term survival of rats and less-severity in liver damage [24], [25].

Panax Ginseng Meyer (P. ginseng) is one of the most widely used medicinal herbs that has a long medicinal history in East Asia [26]. It is well known to have beneficial effects on obesity, cardiac- and liver-associated diseases [27], [28], [29]. The representative compounds that contribute to these beneficial effects are ginsenosides [30], [31], [32]. Ginsenosides are class of natural product steroid glycosides and triterpene saponins [33]. Ginsenosides are dispersed in the roots, berries, and leafs of P. ginseng [34]. Ginsenosides Rb1, Rg1, Rc, Re, and Rd are the most common compounds identified among approximately 40 known ginsenosides [35], [36].

Because of diverse physiological effects of P. ginseng, the root and leaf part of P. ginseng is widely used as crude drugs or functional foods [37], [38]. Many biochemical and medicinal studies have been conducted for scientific explanation of the efficacy of ginsenoside. By contrast, no systematic studies have been conducted on ginsenosides of different parts, including P. ginseng berry [39]. P. ginseng berry, which is mainly used for cultivation of P. ginseng, is discarded after removal of seeds. However, recent studies showed that P. ginseng berry has antidiabetic and antiobesity effects that may be derived from its major compound, ginsenoside Re [30].

In this study, the hepatoprotective effect of the ultrasonication-processed P. ginseng berry extract (UGBE) on MBDL-induced inflammatory liver injury and liver fibrosis was investigated. UGBE was made by new ultrasonication procedure from P. ginseng berry extract (GBE). The GBE was ultrasonicated to produce less-polar active ginsenosides in a time-efficient and cost-effective way. Liver inflammation, oxidative stress, and liver dysfunction were measured followed by MBDL surgery in rats.

Materials and methods

Materials

The aspartate aminotransferase (AST), alanine transaminase (ALT), glutathione peroxidase (GPx), catalase (CAT), superoxide dismutase (SOD) activity, nitric oxide (NO) assay, and tumor necrosis factor-alpha (TNF-α) enzyme-linked immunosorbent assay (ELISA) kits were purchased from BioVision, Inc. (San Francisco, US). The inducible nitric oxide synthase (iNOS) and p65 nuclear factor–kappa B (NF-κB) ELISA kit were purchased from CUSABIO (Wuhan, CN). The heme oxygenase-1 (HO-1) ELISA kit was purchased from Enzo Life Sciences, Inc. (New York, USA). The anti-TLR4 antibody, anti-Myd88 antibody, and ammonia assay kit were purchased from Abcam (Cambridge, UK). The TLR4 ELISA kit was purchased from MyBioSource (San Diego, US). The GBE and UGBE were kindly provided by Prof. S.K.K., Semyung University (Jecheon, KO). Ginsenoside standards were purchased from ChromaDex (Irvine, US). Dulbecco's phosphate buffered saline (PBS) was purchased from Welgene, Inc. (Seoul, KO). Other essential materials were purchased from Sigma-Aldrich Co., LLC (St. Louis, US).

Preparation of UGBE

Four-year grown P. ginseng berry was dried and added with 2000 mL ethyl alcohol per 200 g of dried P. ginseng berry product. The GBE was produced from P. ginseng berry-ethyl alcohol mixture after 2-time reflux extraction and filtration followed by being concentrated by vacuum evaporation. Then, GBE was put in an ultrasonicator (KODO, Hwaseong, KO) with an oscillation and vibration of 600 W at 100˚C and treated for 10 hrs. The remaining solutions were concentrated by vacuum evaporation and freeze-dried to obtain a brownish extract (UGBE).

For further analysis of UGBE and GBE, 2 g of each extract was extracted with 50 mL diethyl ether three times by using an ultrasonicator (KODO, Hwaseong, KO) followed by removal of supernatant. The residue was treated with 50 mL n-butanol three ties again. n-Butanol fraction that built up in the ultrasonicator was filtered and concentrated by a vacuum evaporator.

Analysis of ginsenosides in UGBE

The total ginsenoside content and ginsenoside composition of UGBE and GBE were analyzed using a Waters 1525 binary high-performance liquid chromatography (HPLC) system (Milford, US). The separation of UGBE was performed on an Eurospher analytical column (100-5 C18, 250 mm × 3.0 mm, 5 μm, Knauer, Berlin, DE) by gradient elution at room temperature. The eluent was a mixture of A (acetonitrile for HPLC) and B (distilled water for HPLC). The process of elution was carried out according to the following conditions: 0 mins, 17% of A; 25 mins, 25% of A; 50 mins, 40% of A; 105 mins, 60% of A; 110 mins, 100% of A. The flow rate was 0.8 mL/min, the injection volume was 20 μL, and the chromatograms were obtained using a Ultraviolet-visible Waters 2478 Dual λ Absorbance Detector (Milford, USA) at 203 nm.

MBDL model

Since complete ligation of all bile ducts leads to death in a few weeks, MBDL surgery was performed on rats to evaluate the long-term effect of UGBE [40]. Rats underwent mild bile duct ligation surgery, in which only 3 out of 5 bile ducts were ligated after midline ventral incision, under diethyl-ether anesthesia. The ligature was placed between three upper bile ducts (proximal) and two bottom bile ducts (distal) portion of the rest of bile duct. Nonabsorbable surgical suture materials were used for ligations. In sham-operation group, a midline incision was performed without MBDL surgery. Midline suture was performed with same materials used in MBDL surgeries. All animals recovered from operation with antibiotic drugs (ointment) and warm condition for 1 wk. The MBDL surgery is summarized in Fig. 1.

Fig. 1

The graphical summary of mild bile duct ligation model. The mild bile duct ligation model was performed to induce hepatic encephalopathy in rats instead of bile duct ligation model. Rats underwent ligation surgery, in which only 3 out of 5 bile ducts were ligated after midline ventral incision for long-term survival.

Experimental design

Seventy-two male Sprague-Dawley rats (200-250 g) were purchased from Samtako Bio Korea (Osan, Korea). The animals were group-housed in cages with wire-net floors in a room controlled for temperature (24-25°C) and humidity (70-75%) and were fed a normal laboratory diet from Samtako Bio Korea. All animals were fasted for 24 hrs before surgery and sacrifice but were allowed free access to tap water. Animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Chung-Ang University, in accordance with the guide for the Care and Use of Laboratory Animals in Seoul, Korea (CAUIACUC-2016-00095).

All rats were randomly divided into the following eight groups and fed an appropriate reagent or extract for 6 wks (n = 9); (a) control group, 2 mL/kg (b.w.) of normal saline (p.o.); (b) sham-operation group, 2 mL/kg (b.w.) of normal saline (p.o.); (c) MBDL group, 2 mL/kg (b.w.) of normal saline (p.o.); (d) GBE group, 250 mg/kg (b.w.) of GBE (p.o.); (e) silymarin group, 250 mg/kg (b.w.) of silymarin (p.o.); (f) UGBE1 group, 100 mg/kg (b.w.) of UGBE (p.o.); (g) UGBE 2 group, 250 mg/kg (b.w.) of UGBE (p.o.); (h) UGBE3 group, 500 mg/kg (b.w.) of UGBE (p.o.). All rats underwent MBDL surgery except control and sham-operation groups. Twenty-four hours after the last administration, all rats were sacrificed by cervical dislocation. The blood and liver were removed immediately. All collected samples were properly handled for further assays.

Preparation and biochemical assays of liver and serum samples

After being sacrificed, whole blood samples (4–5 mL) were collected from rat inferior vena cava. Whole blood samples were collected in SST II Plus plastic serum tube (Becton, Dickinson, and Company, US). Collected samples were centrifuged at 10,000 g, 4°C for 20 mins. After being centrifuged, serum samples were obtained and immediately flash-frozen at −80°C for further assays (AST, ALT, and ammonia level).

Liver samples were perfused with PBS (pH 7.4) solution through the portal vein to remove any red blood cells. The portal vein was cannulated with a 23-gauge I.V. catheter (Korean vaccine, KO), and the abdominal inferior vena cava was cut immediately. After perfusion, liver samples were removed and washed in saline. Collected samples were flash-frozen immediately at −80°C for further assays (antioxidative effect; SOD, GPx, CAT, GSH and HO-1, lipid peroxidation; iNOS and NO, inflammation; TNF-α, a receptor; TLR4 and MyD88). All further assay procedures were carried out in accordance with the manufacturer's instruction.

Measurements of serum biochemical parameters

Serum samples were diluted for optimal reactions before being measured. The serum ALT and AST activities and ammonia level were measured by the optimal kits in accordance with the manufacture's instruction. The serum TNF-α protein expression was measured by ELISA.

Measurements of hepatic biochemical parameters.

Hepatic samples were diluted for optimal reactions before being measured. The enzymatic activities of SOD, GPx, CAT and GSH, and NO level were measured by colorimetric assays. The activity of hepatic iNOS, HO-1, Nf-κB, MyD88, and TLR4 were measured by an ELISA kit. All assay procedures were progressed according to the manufacturer's instructions.

Immunohistochemistry

Rat livers for immunohistochemistry were perfused in the same way as described previously. After being removed, liver samples were immersed for 2 weeks at room temperature. After immersing, liver samples were embedded in paraffin and were cut in 5-μm-thick cross sections using microtome. For staining of the expression of TLR4 and MyD88, TLR4 antibody (1:100) and MyD88 (1:100) antibodies were used; the secondary antibody was a Dako Real™ EnVision™ Detection System Rabbit/Mouse (1:200). After development with diaminobenzidine, sections were mounted on poly-lysine gelatinized glass slides and dehydrated through graded ethanol solutions before coverslipping.

Stained tissues were observed in Leica DMR 6000 microscope, and images were captured with a Leica DM 480 camera (Wetzlar, Germany). Images presented were photographed at 20 × 10 magnifications.

Hematoxylin and eosin and cresyl violet staining

Hematoxylin and eosin (H&E) staining was performed for detecting liver damage. Paraffin-embedded liver samples were cut in 5 μm thick using a microtome in the same way as described previously. Samples were then stained with H&E for detection of liver damage.

Stained tissues were observed with a Leica DMR 6000 microscope, and images were captured with a Leica DM 480 camera (Wetzlar, Germany). Low-magnification images presented were photographed at 20 × 10, and high-magnification images presented were photographed at 40 × 10.

Statistical analysis

The data are expressed as mean ± S.E.M. Statistical comparisons between each experimental group of all data were analyzed by the Student's t-test and one-way analysis of variance test. Differences were considered significant at error probabilities smaller than 0.05.

Results

HPLC analysis of UGBE

Table 1 shows the composition ratio of the GBE and UGBE analyzed by the HPLC systems [41]. The ginsenosides Rg2, Rg3, Rh1, Rh4, Rk1, Rk3, and F4 were significantly increased after 10-hr ultrasonication at 100˚C. Furthermore, ginsenosides Rg3 and Rk1 were newly produced in UGBE, which were not identified in GBE. HPLC chromatograms also proved these changes (Fig. 2) [42]. Fig. 3 shows the major component change between GBE and UGBE. Ginsenosides which were increased by ultrasonification had less polar residues than ginsenoside Re which was decreased by ultrasonification.

Table 1

Composition ratio of GBE and UGBE (%, w/w)

Ginsenoside	GBE	UGBE
Rb1	0.758 ± 0.179	0.072 ± 0.052#
Rb2	0.594 ± 0.114	0
Rd	1.534 ± 0.182	0.025 ± 0.007#
Re	11.169 ± 0.158	0.288 ± 0.037#
Rf	0.330 ± 0.115	0
Rg1	0.567 ± 0.013	0.033 ± 0.004*
Rg2	0.801 ± 0.215	2.278 ± 0.368*
20S-Rg3	0	0.432 ± 0.063*
20R-Rg3	0	0.400 ± 0.059*
Rg6	0.044 ± 0.026	0.445 ± 0.063*
Rh1	0.629 ± 0.095	1.350 ± 0.208*
Rh4	0	0.083 ± 0.011*
Rk1	0	0.2071 ± 0.030*
Rk3	0	0.039 ± 0.005*
F1	0.193 ± 0.149	0.035 ± 0.017
F4	0.191 ± 0.026	1.210 ± 0.137*

Ginsenoside Rb1, Rd, Re, Rg1, Rg2, RG3, RG6, Rh1, Rh4, Rk1, Rk3, F1, and F4 were detected in UGBE. Data represent mean ± S.E.M.

*p < 0.05 increased composition ratio of UGBE compared with same ginsenoside in GBE.

#p < 0.05 decreased composition ratio of UGBE compared to same ginsenosides in GBE.

Fig. 2

High-performance liquid chromatography (HPLC) chromatograms of Panax ginseng berry extract and UGBE. (A) HPLC chromatogram of P. ginseng berry extract and (B) HPLC chromatogram of UGBE. The chromatograms were obtained using UV/Vis Waters 2478 Dual λ Absorbance Detector at 203 nm. a: ginsenoside Re, b: ginsenoside Rg2, c: ginsenoside Rh1, d: ginsenoside F4, e: ginsenoside Rk3 f: ginsenoside Rh4, g: ginsenoside Rg3, h: ginsenoside Rk1.

Fig. 3

Major component changes in Panax ginseng berry extract after ultrasonication. The ginsenoside Re was decreased by ultrasonication. Ginsenosides Rg3, Rg2, Rk1, F4, Rk3, Rh4, and Rh1 were increased by ultrasonication.

Effect of UGBE on liver dysfunction

AST and ALT are representative indicators of liver dysfunction. In this experiment, serum AST and ALT levels were measured (Table 2). The result shows that MBDL surgery significantly increased the activities of serum AST and ALT. However, administration of UGBE reduced the activities of serum AST and ALT in a dose-dependent manner. The treatment of low dosage of UGBE (100 mg/kg) showed better hepatoprotective effect than the treatment of silymarin (150 mg/kg). Also, serum ammonia level was measured as a result of liver dysfunction. The serum ammonia level in the experimental group showed similar tendencies to those of AST and ALT.

Table 2

Effect of UGBE on serum AST, ALT, and ammonia levels

	AST (IU/L)	ALT (IU/L)	Ammonia(μM)
Control	89.07 ± 10.13	50 ± 4.66	706.45 ± 82.63
Sham	91.37 ± 15.37	53.47 ± 8.67	753.91 ± 90.14
MBDL	173.51 ± 30.34***	122.46 ± 21.17***	1738.39 ± 133.71**
Sil	120.05 ± 10.62#	66.58 ± 10.87##	1137.42 ± 201.56##
GBE	158.54 ± 51.19	109.75 ± 20.16	1613.13 ± 88.82
UGBE1	115.19 ± 11.38#	65.64 ± 18.76##	1160.01 ± 118.26##
UGBE2	106.07 ± 9.91##	60.35 ± 8.98##	805.94 ± 92.13###
UGBE3	90.52 ± 13.43###	54.21 ± 7.76###	741.26 ± 103.57###

Control: control rats; Sham: Sham Operation control rats; MBDL: MBDL rats; Sil: MBDL rats Treated With silymarin (150 Mg/kg); GBE: MBDL rats Treated With GBE (150 Mg/kg); UGBE1: MBDL rats Treated With UGBE (100 Mg/kg); UGBE2: MBDL rats Treated With UGBE (250 Mg/kg); UGBE3: MBDL rats Treated With UGBE (500 Mg/kg).

Data are expressed as means ± S.E.M. (n = 9).

***p < 0.005 Compared to control, #p < 0.05, ##p < 0.01, ###p < 0.001 compared with MBDL.

H&E staining results also supported the effect of UGBE on hepatotoxicity showing (Fig. 4). The control and sham group showed normal morphologies of liver. On the other hand, being compared with control and sham group, MBDL surgery group showed severe abnormal changes in hepatic lobules. In addition, the degeneration of hepatocyte, centrilobular necrosis, inflammatory cell infiltration, and inflammatory foci were frequently observed (Fig. 4C). GBE group also showed extensive histopathological changes such as hepatocytes degeneration, necrosis, and inflammatory cell infiltration (Fig. 4D). However, these histopathological changes were attenuated by the treatment of UGBE and silymarin. Especially, H&E staining results of UGBE2 and UGBE3 group showed normal morphology of hepatocyte (Figs. 4E–4H).

Fig. 4

Photomicrographs of liver section stained with hematoxylin and eosin. (A) Stained liver sections in control group, (B) sham-operation group, (C) mild bile duct ligation (MBDL) group, (D) silymarin treatment group, (E) Panax ginseng berry extract (GBE) treatment group, (F) UGBE1 treatment group, (G) UGBE2 treatment group, and (H) UGBE3 treatment group were observed and representative images of each section were captured at 20*10 magnifications. Control: control rats; Sham: sham operation control rats; MBDL: MBDL rats; Sil: MBDL rats treated with silymarin (150 mg/kg); GBE: MBDL rats treated with GBE (150 mg/kg); UGBE1: MBDL rats treated with UGBE (100 mg/kg); UGBE2: MBDL rats treated with UGBE (250 mg/kg); UGBE3: MBDL rats treated with UGBE (500 mg/kg).

Effect of UGBE on oxidative stress in liver

Oxidative stress can accelerate liver dysfunction and hepatotoxicity. The activities of representative antioxidant enzymes such as SOD, GPx, and CAT were significantly decreased by MBDL surgery (Table 3). The treatment of GBE did not recover the function of antioxidant enzymes, whereas the treatment of silymarin or UGBE did. Also, UGBE enhanced the protein expression of an inducible antioxidant enzymes against liver damage, HO-1, in dose-dependent manner. Even at low dosage (100 mg/kg), UGBE increased the antioxidant capacity of liver significantly. The treatment of GBE did not increase HO-1 levels against oxidative stress induced by MBDL surgery.

Table 3

Effect of UGBE on oxidative stress

	GPx (U/mg)	SOD (U/mg)	CAT (U/mg)	HO-1 (ng/mg)
Control	23.26 ± 6.29	79.16 ± 10.23	12.26 ± 1.02	51.05 ± 7.44
Sham	25.14 ± 5.14	76.51 ± 7.55	11.84 ± 1.45	53.38 ± 8.87
MBDL	11.39 ± 3.17 ***	31.13 ± 5.17***	3.07 ± 0.96***	143.61 ± 20.05*
Sil	17.67 ± 5.83#	60.02 ± 12.31##	7.55 ± 2.16##	168.40 ± 14.12
GBE	13.05 ± 4.41	35.27 ± 18.94	3.66 ± 1.08	137.12 ± 18.53
UGBE1	18.05 ± 5.03##	61.48 ± 9.32##	7.95 ± 2.46##	184.48 ± 30.67##
UGBE2	20.94 ± 5.87###	70.05 ± 11.38##	10.67 ± 1.21###	216.63 ± 22.94###
UGBE3	22.48 ± 6.02###	77.18 ± 12.68###	12.34 ± 1.54###	262.68 ± 28.11###

Control: control rats; Sham: Sham Operation control rats; MBDL: MBDL rats; Sil: MBDL rats Treated With silymarin (150 Mg/kg); GBE: MBDL rats Treated With GBE (150 Mg/kg); UGBE1: MBDL rats Treated With UGBE (100 Mg/kg); UGBE2: MBDL rats Treated With UGBE (250 Mg/kg); UGBE3: MBDL rats Treated With UGBE (500 Mg/kg).

Data are expressed as means ± S.E.M. (n = 9).

***p < 0.005 Compared to control, #p < 0.05, ##p < 0.01, ###p < 0.001 compared with MBDL.

Effect of UGBE on iNOS and NO in liver

Free radicals also affect the pathogenesis of liver damage which is related to lipid peroxidation. The protein expression of iNOS and production of NO were investigated in liver (Fig. 5). In MBDL group, the protein expression of iNOS and the production of NO were significantly increased more than four times those of control group. However, the treatment of UGBE significantly decreased the protein expression of iNOS and production of NO, whereas the treatment of GBE did not.

Fig. 5

Effect of UGBE on hepatic inducible nitric oxide synthase (iNOS) and NO levels. (A) Protein expression of iNOS in liver and (B) hepatic NO level. Control: control rats; Sham: sham operation control rats; MBDL: MBDL rats; Sil: MBDL rats treated with silymarin (150 mg/kg); GBE: MBDL rats treated with GBE (150 mg/kg); UGBE1: MBDL rats treated with UGBE (100 mg/kg); UGBE2: MBDL rats treated with UGBE (250 mg/kg); UGBE3: MBDL rats treated with UGBE (500 mg/kg). Data are expressed as means ± S.E.M. (n = 9). ***p < 0.005 compared to control, #p < 0.05, ##p < 0.01, ###p < 0.001 compared with MBDL.

Effect of UGBE on TLR4 signaling pathway

It is assumed that TLR4 signaling pathway might be involved in MBDL-induced liver dysfunction and hepatotoxicity. Based on this idea, the protein expression of TLR4 and its adaptable molecules, MyD88 were investigated (Figs. 6B and 6D). Our results showed that the treatment of UGBE significantly reduced MyD88-dependent TLR4 signaling pathway which increased by MBDL surgery. To further confirm the protein expression of TLR4 and MyD88, immunohistochemical analysis for TLR4 and MyD88 were conducted (Figs. 6A and 6C). The number of TLR4 and MyD88-positive cells in UGBE treatment group was decreased significantly compared with that in the MBDL group.

Fig. 6

Effect of UGBE on TLR4 signaling pathway. (B) and (D) are protein expression of TLR4 and MyD88 in liver, respectively. (A) and (C) are photomicrographs of TLR4 and MyD88 immunohistochemistry, respectively. Immunohistochemistry assays in stained liver sections were performed in liver and representative images were captured at 20*10 magnifications. Control: control rats; Sham: sham operation control rats; MBDL: MBDL rats; Sil: MBDL rats treated with silymarin (150 mg/kg); GBE: MBDL rats treated with GBE (150 mg/kg); UGBE1: MBDL rats treated with UGBE (100 mg/kg); UGBE2: MBDL rats treated with UGBE (250 mg/kg); UGBE3: MBDL rats treated with UGBE (500 mg/kg). a: control group, b: sham-operation group, c: MBDL group, d: silymarin treatment group, e: GBE treatment group, f: UGBE1 treatment group, g: UGBE2 treatment group and h: UGBE3 treatment group. Data are expressed as means ± S.E.M. (n = 9). ***p < 0.005 compared to control, #p < 0.05, ##p < 0.01, ###p < 0.001 compared with MBDL.

MyD88-dependent TLR4 signaling pathways activated the early phase of NF-κB, which produces inflammatory cytokines, including TNF-α. The expression of downstream signaling molecules of TLR4 signaling pathway, NF-κB, was reduced by UGBE treatment in a dose-dependent manner (Fig. 7A). Also, the protein expression of TNF-alpha was reduced by UGBE treatment (Fig. 7B). The treatment of GBE did not show the effect on upregulated TLR4 signaling pathway and overproduction of TNF-α induced by MBDL surgery.

Fig. 7

Effect of UGBE on nuclear factor–kappa B (NF-κB) and tumor necrosis factor-alpha (TNF-α). (A) Protein expression of total (p65) NF-κB. (B) Protein expression of TNF-α in liver. Control: control rats; Sham: sham operation control rats; MBDL: MBDL rats; Sil: MBDL rats treated with silymarin (150 mg/kg); GBE: MBDL rats treated with GBE (150 mg/kg); UGBE1: MBDL rats treated with UGBE (100 mg/kg); UGBE2: MBDL rats treated with UGBE (250 mg/kg); UGBE3: MBDL rats treated with UGBE (500 mg/kg). Data are expressed as means ± S.E.M. (n = 9). ***p < 0.005 compared to control, #p < 0.05, ##p < 0.01, ###p < 0.001 compared with MBDL.

Discussion

Owing to its various physiological effects, P. ginseng has gained huge interest for use in medicinal applications [43], [44], [45], [46], [47]. Several studies have shown the antidiabetic and antiobesity effects of ginsenoside Re, which is the major component of the P. ginseng berry [30], [48], [49]. However, another protective effect of ginsenosides on HE is still not clear. The aim of this study is to produce UGBE containing a high concentration of active prosapogenins and to evaluate the protective effect of UGBE on MBDL-induced rat HE model.

The composition ratio of UGBE and GBE were analyzed by HPLC. After ultrasonication process, ginsenoside Rb1, Rb2, Rd, Re, Rf, Rg1, Rg2, Rg3, Rg6, Rh1, Rh4, Rk1, Rk3, F1, and F4 were identified (Table 1). Among these ginsenosides, the composition ratio of most less polar ginsenosides, such as ginsenoside Rh1, Rh4, Rg2, Rg3, Rk1, Rk3, and F4, were largely increased from that produced during red ginseng's process of manufacture, whereas ginsenoside Re was decreased [50]. Several studies have investigated the effect of these compounds on the liver antioxidative, anti-inflammatory, and anti-carcinogenic effects of ginsenoside Rg2 and have been reported [51], [52], [53], [54]. Ginsenoside Rk1 is known to have an anticancer effect in hepatocellular carcinoma cells [55], [56]. Ginsenoside Rg3 has antioxidative and hepatoprotective properties via the HO-1-related signaling pathway [57]. Ginsenoside Rh1 has been reported to exert an anti-inflammatory activity by inhibiting iNOS and cyclooxygenase-2 protein expression [58]. Ginsenoside F4 has shown an apoptosis-inducing effect that could be beneficial to organ protection [59], [60].

With the ultrasonication processing, active ginsenosides were produced more than with typical red ginseng manufacturing. The primary ginsenosides of red ginseng are known to have an inhibitory effect on liver injury from a variety of mechanisms [61], [62], [63]. The ginsenosides Rg2, Rh1, and F4 are commonly produced in the complicated manufacturing process of red ginseng and exist in a small amount in P. ginseng berry [50], [64]. The effect of the UGBE on liver is still in a veil. However, the potential of this ginsenoside on liver is expected to be useful.

Biliary obstruction or biliary drainage is a common pathophysiology of HE patients [65]. Once bile duct is ligated, the liver cannot excrete bile acid that could be strong toxicant itself in liver, which causes cholestatic liver disease and dysfunction [66]. Liver dysfunction often leads to overproduction of ammonia in liver [67], [68]. BDL surgery induces liver fibrosis and inflammation directly and definitely by an undercurrent of bile acid in liver, whereas PVL or PCA surgery remains a controversy in the induction of HE condition [69], [70]. The difference might be due to the uncertainty of ischemic condition and the way biochemical parameters are measured [71]. In this experiment, MBDL model was used instead of BDL surgery to induce liver fibrosis in rats. MBDL model has shown a longer survival than BDL rats, i.e., more than 3 months, allowing researchers to study the long-term effect on liver damage [40]. Although several studies reported the difference between BDL and MBDL models in the pattern of liver dysfunction, our MBDL induced liver fibrosis sufficiently without cirrhosis. The main differences are only the survival rate and rapidity of inflammatory liver injury [40].

In the present study, MBDL surgery significantly increased the levels of the representative indicator of liver dysfunction (serum ALT, AST, and ammonia) by MBDL surgery in rats. However, the long-term administration of UGBE recovered the liver function and serum ammonia level, dose-dependently (Table 2). H&E staining also supports these results (Fig. 4). Meanwhile, GBE did not affect the liver dysfunction induced by BDL surgery. The liver dysfunction by biliary obstruction is characterized by increased serum ammonia levels [72]. Ammonia level in serum is closely correlated with the severity of inflammatory liver injury [73]. At physiological pH, the majority (98%) of ammonia is found in ionic form (NH4+) with ∼2% arising in gas form (NH3). Both forms are capable of crossing cellular membranes. NH3 mainly pass through the cellular membrane through diffusion. Meanwhile, NH4+ cross the cellular membrane via K+ channels and cotransporters since NH4+ has very similar ionic properties (ionic radius and diffusion coefficient) to K+. The differences between GBE and UGBE on liver dysfunction may be due to the composition ratio of ginsenosides. In this study, we confirmed that the UGBE decreased serum ammonia level and restored liver function significantly in rat MBDL model. These result suggested that the UGBE has huge potential to be developed as a cure for HE.

In the liver dysfunction, oxidative stress plays a crucial role in contributing to the initiation and progression of hepatic failure, especially in inflammatory liver disorders [74], [75], [76]. The enzyme-dependent antioxidative system plays important roles in reactive oxygen species (ROS)-induced liver damage [77]. The BDL surgery resulted in the imbalance of ROS production and the impairment of the antioxidant capacity of the liver [78]. Overproduction of ROS inactivates antioxidative enzymes such as SOD, CAT, and GPx in the BDL-induced inflammatory liver injury model in rats [79]. Recovery of these enzymes suggests that treatment of UGBE recovered liver dysfunction and reduced oxidative stress-induced by MBDL surgery effectively. Furthermore, HO-1 plays an important role in cytoprotection by decreasing the leukocytic response and ameliorating BD-induced liver damage [80]. HO-1 is an enzyme that catalyzes the degradation of heme into bilirubin, carbon monoxide, and free iron [81]. HO-1 expression increased dose-dependently according to UGBE treatment (Table 3). These results indicate that UGBE enhanced the hepatoprotective effect by increasing the HO-1 protein expression and is likely mediated by ginsenoside Rg2 [57]. The findings of the present study reveal that administration of UGBE protects antioxidative enzymes from oxidative damage (Table 3). Although it has not been identified that UGBE directly scavenges ROS, there is a strong likelihood that that UGBE increased the antioxidant capacity and ameliorated oxidative stress against MBDL-induced hepatotoxicity.

The expression of iNOS in hepatic ischemia-reperfusion injury aggravates the pathogenesis of chronic liver injury by increasing lipid peroxidation in hepatic macrophage [82], [83]. NO produced by iNOS dramatically reacts with ROS to form peroxynitrite (ONOOK) which is a toxic agent for liver [84]. In the present study, UGBE ameliorated MBDL-induced hepatotoxicity. The oral administration of UGBE significantly suppressed the protein expression of iNOS and the production of NO, whereas GBE was ineffective (Fig. 5). It is assumed that the hepatoprotective effect of UGBE is related not only with hepatocyte but also with macrophage in the liver.

Several studies have reported the correlation between BDL-induced liver injury and TLR4 through bacterial translocation [85], [86], [87]. Activation of TLR4 can signal through NF-κB nuclear location which results in the production of inflammatory cytokines [88]. There are two types of TLR4 signaling pathway, including MyD88-dependent and MyD88-independent (TRIF-related adaptor molecule; TRAM) signaling pathways [89]. MyD88 is an adaptor molecule that signals TLR4 signaling predominantly [90]. In our MBDL model, MyD88 signaled TLR4 signaling pathway, whereas MyD88-independent signaling pathway did not (data not shown). The UGBE treatment reduced the protein expression of TLR4 and MyD88 (Fig. 6). The NF-κB ELISA assay showed the same result with TLR4 and MyD88 (Fig. 6A). These results indicate that the protective effect of UGBE on BDL induced liver injury relies on TLR4 receptor signaling pathway.

TNF-α, representative proinflammatory cytokines-induced by TLR4 signaling pathway, is an important proinflammatory cytokine that triggers liver damage in the BDL-induced liver fibrosis model [91]. TNF-α also plays an important role in hepatotoxicity related to free-radical-mediated apoptosis [92]. Hepatic TNF-α levels were also reduced with UGBE treatment in a dose-dependent manner (Fig. 7B).

Based on the results of the present study, it can be summarized that MBDL-induced liver damage was ameliorated by oral administration of UGBE (Fig. 8). Oral administration of UGBE reduced hepatotoxicity in the liver, which induced by MBDL surgery. Also, the serum ammonia level was reduced by administration of UGBE. This protective effect is mediated by suppression of TLR4 protein expression in liver. Consequently, the UGBE showed great protective effect on MBDL-induced liver fibrosis and inflammatory liver injury model in rats and a potential to be developed as a new remedy for liver fibrosis.

Fig. 8

The graphical summary. The mild bile duct ligation (MBDL) surgery is induced severe hepatotoxicity which leads to liver inflammation in rats. Also, the serum ammonia level was increased by MBDL surgery. However, the liver dysfunction of MBDL surgery operated rats was attenuated by UGBE treatment via myeloid differentiation factor 88-dependent toll-like receptor 4 signaling pathways.

Conflicts of interest

All authors have no conflicts of interest to declare.

Source of funding

The authors declare no source of funding.

Others

Experiments were done by Y.N. in pharmacology laboratory in Chung-Ang University. The extract used in this article is kindly provided by S.K.K.. U.D.S. organized all procedure related to this article. This manuscript was not presented in any poster meeting as well as elsewhere.