TP (purity >98% by high-performance liquid chromatography analysis) was supplied by the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). Silymarin was purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany) and consisted primarily of silibinin (>30%), isosilibinin, silicristin and silidianin. TP was dissolved in 0.05% dimethyl sulfoxide saline. Silymarin was suspended in 0.5% sodium carboxymethyl cellulose (CMC-Na) distilled water solution. Other chemicals and reagents used were of analytical grade.

A total of 60 male Wistar rats (weight, 220±20 g; 6-8 weeks) were obtained from the Experimental Animal Center of Guangzhou University of Chinese Medicine (Guangzhou, China). All the experiments were performed in a specific-pathogen free laboratory at a controlled temperature of 25±2°C and relative humidity of 35–75% on a 12 h light/dark cycle. Animals had free access to rat chow and tap water. All experimental protocols were in compliance with the guidelines of the Committee for Animal Care and Use of Guangzhou University of Chinese Medicine, and were approved by the ethical committee of Guangzhou University of Chinese Medicine.

Rats were randomly divided (n=12/group) into five experimental groups. Normal control (NC) and triptolide (TP) groups received oral administration of normal saline (1 ml/100 g), whereas silymarin-treated groups were orally administrated with silymarin (50, 100 and 200 mg/kg) dissolved in 0.5% CMC-Na distilled water. The doses of silymarin employed in the present study were based on our prior trial and previous study (21). Following treatment with normal saline or silymarin for 7 consecutive days, all groups, excluding the NC group, were given TP (2 mg/kg) by intraperitoneal injection to establish the liver injury animal model. At the end of the experimental period (day 8), all rats were fasted for 12 h. On the following day, the animals were anesthetized with an intraperitoneal injection of 10% chloral hydrate. Blood samples were collected from the abdominal aorta, and rats were subsequently sacrificed by spinal dislocation and liver tissue was collected from the left lobe.

Blood samples were placed at room temperature for 1 h. Following centrifugation at 1,000 × g for 10 min at room temperature, the serum was separated for biochemical estimations. The serum levels of ALT, AST, alkaline phosphatase (ALP), total cholesterol (TC) and γ-glutamyl transferase (GGT) were measured using a Hitachi-7180 type biochemical analyzer following the standard methods.

The liver tissues were chopped into 1 cm³ sections and divided into two groups. One was prepared to make 1:9 (w/v) homogenates with cold saline or phosphate buffer saline (PBS) using a tissue homogenizer (IKA T 18 Basic). The homogenate was centrifuged at 1,000 × g for 10 min at 4°C to obtain the supernatant samples for subsequent determinations. The other group of liver tissues were fixed in 10% formalin at room temperature for at least 24 h for further observation of tissue sections.

The supernatant of normal saline homogenates was used to measure the activities of the important antioxidant enzymes superoxide dismutase (SOD), glutathione peroxidase (GSH-Px), glutathione S-transferase (GST) and catalase (CAT), as well as glutathione (GSH) and malonaldehyde (MDA) levels in the liver, according to the manufacturer's instructions [total SOD assay kit (hydroxylamine method), GSH-PX assay kit (colorimetric method), GST assay kit (colorimetric method), GSH assay kit (colorimetric method), CAT assay kit (ultraviolet) and MDA assay kit (TBA method); Nanjing Jiancheng Bioengineering Institute, Nanjing, China]

The levels of inflammatory cytokines tumor necrosis factor (TNF)-α, interleukin (IL)-6, IL-10, IL-1β, and cytochrome c (Cyt C) levels were measured in supernatants using specific ELISA kits [E-30644 (IL-6), E-30418 (IL-1β), E-30633 (TNF-α), E-30649 (IL-10) and E-30273 (Cyt C); Beijing Chenglin Biotechnology, Beijing, China] according to the manufacturer's instructions.

After being fixed in 10% formalin for 24 h, the liver tissue was dehydrated in graded ethanol (50–100%), cleared in xylene and embedded in paraffin wax. The paraffin sections (5 µm thickness) were examined for pathological changes in the liver tissue using a light microscope (BK-DM320; Chongqing Optec Instrument Co., Ltd., Chongqing, China) and photographed at ×200 magnification following hematoxylin and eosin (H&E) staining for 5–10 min at room temperature.

A terminal deoxynucleotidyl-transferase-mediated dUTP nick end labeling (TUNEL) apoptosis detection kit (Nanjing Lufei Biotechnology Co., Ltd., Nanjing, China) was used to assess the apoptosis rate of liver cells. Paraffin sections (5 µm) were incubated with the TUNEL reaction mixture at 37°C for 60 min, then incubated with converter-POD in a humidified chamber for 30 min at 37°C followed by incubation with DAB substrate (5 µl 20X DAB, 1 µl 30% H₂O₂ and 94 µ l PBS) for 10 min at 20°C. The sections were washed and mounted in PBS. A total of 15 fields (×200 magnification) for each section were randomly selected to quantify the positive cells and define the apoptosis indexes using ImageJ software (version 1.50; National Institutes of Health, Bethesda, MD, USA).

Immunohistochemical analyses were performed to determine the expression of Bcl-2-associated X (Bax) and Bcl-2 in the liver. Paraffin sections (5 µm thick) of the liver tissue were initially processed by dehydration. Following retrieval by citrate buffer (pH=6) microwave antigen retrieval for 15 min, 3% H₂O₂ was added to inactivate endogenous enzymes for another 15 min. Then, 5% normal goat serum (ab7481; Abcam, Cambridge, UK) diluted with PBS was used to block non-specific binding for 10 min at room temperature. Subsequently, sections were washed with 0.01 mol/l PBS following incubation of sections with 50 µl rabbit monoclonal antibodies for Bcl-2 (1:100, ab59348) and Bax (1:100; ab53154) (both from Abcam) at 4°C overnight. The secondary antibody was goat anti-rabbit IgG H&L (HRP) (1:1,000, ab6721; Abcam) for 30 min at 37°C. The slides were counterstained with hematoxylin (Modified Mayer's, ab220365; Abcam) for 2–3 min at room temperature. Finally, the sections were reacted with a staining solution containing 0.03% 3,3′-diaminobenzidine (Nanjing Lufei Biotechnology Co., Ltd.,) for 5–10 min at room temperature. The sections were observed with an Olympus IX71 fluorescence microscope (Olympus Corporation, Tokyo, Japan) at ×200 magnification. A total of 15 fields in each section were randomly selected to quantify positive cells by ImageJ software (version 1.50; National Institutes of Health).

Western blot analysis was performed to examine the expression of c-caspase-3, which stimulates cell apoptosis, and TNF-α induced phosphorylation of p38 and JNK in the inflammatory pathway. Liver tissue was homogenized with proteinase inhibitor on ice to obtain total protein extract according to the protocol of the T-PER Tissue Protein Extraction reagent (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Protein concentration was determined following the protocol of a BCA protein assay kit (G2026; Wuhan Goodbio Technology Co., Ltd., Wuhan, China). Extracts containing equal amounts of protein (50 µg) were separated by 10% SDS-PAGE and transferred to polyvinylidene fluoride membranes, which were blocked at room temperature for 1 h in 0.5% TBST (50 mmol/l Tris-HCl, pH 7.6; 150 mmol/l NaCl; 0.1% Tween-20) containing 5% non-fat milk and subsequently incubated with primary rabbit anti-rat polyclonal antibodies against c-caspase-3 (1:1,000, 9661; Cell Signaling Technology, Inc., Danvers, MA, USA) and β-actin (1:1,000, GB13001-1; Wuhan Goodbio Technology Co., Ltd.), and rabbit anti-rat monoclonal antibodies against p-p38 (1:1,000, 4511) and p-JNK (1:1,000, 4671) (both from Cell Signaling Technology, Inc.) overnight at 4°C. β-actin was used as an internal control. All primary antibodies were purchased from Cell Signaling Technology, Inc. Subsequently, the membranes were washed with TBST for 15 min and incubated with secondary antibodies (anti-rabbit IgG, HRP-linked antibody, 1:3,000; Cell Signaling Technology, Inc.) labeled with horseradish peroxidase for 30 min at room temperature, followed by washing with TBST for 15 min. The membranes were then developed using an electrochemiluminescence kit (G2014; Wuhan Goodbio Technology Co., Ltd.) and the densities of the immunoreactive bands were analyzed using ImageJ software (version 1.50; National Institutes of Health).

Data are presented as the mean ± standard deviation. One-way analysis of variance followed by Fisher's least significant difference tests was performed to statistically analyze the data. SPSS software (version 16; SPSS, Inc., Chicago, IL, USA) was used for statistical analysis. P<0.05 was considered to indicate a statistically significant difference.

The levels of ALT, AST, ALP, TC and GGT were estimated in serum samples as sensitive biomarkers of liver function. These enzymes (ALT, AST, ALP and GGT) were enzymes derived from the injured liver tissue cells. As presented in Table I, the levels of these serum biochemical parameters in the TP group were significantly higher compared with the NC group (P<0.01), which was indicative of liver dysfunction. However, the levels of these indicators in the three silymarin-pretreated groups markedly reduced to normal levels in a dose-dependent manner, indicating that silymarin may effectively ameliorate TP-induced liver injury.

Hepatic histological changes directly depict the degree of liver injury. To morphologically characterize the potential pathological alternations in hepatocytes, routine H&E staining was performed. As demonstrated in Fig. 2A, a clear and regular lobular structure was observed, where hepatic cords remained legible and radiated from the central vein in the NC group. Compared with the NC group, hepatic parenchymal necrosis and inflammatory cell infiltration was apparent, and disordered hepatic cords accompanied by extensive vacuolation congestion was also observed in the TP group (Fig. 2B), which was consistent with the assay results for biochemical parameters in the serum, indicating the successful establishment of the hepatic injury model. Furthermore, hepatocytes were markedly swollen, with condensed karyons and loosened cytoplasts. However, silymarin pretreatment of three tested doses appeared to alleviate TP-induced liver injury in a dose-dependent manner, as evidenced by reduced inflammation infiltration, intact tissue structure, and reduced necrotic and apoptotic mass (Fig. 2C-E), compared with the TP group. Cell apoptosis and inflammatory infiltration was particularly limited when subjected to silymarin pretreatment at the doses of 100 and 200 mg/kg.

As demonstrated in Fig. 3A-E, TP administration markedly reduced the activities of the antioxidant enzymes SOD, GSH-Px, CAT and GST and GSH levels by 40.61, 28.18, 45.10, 35.97 and 79.81%, respectively, compared with the NC group (P<0.05). However, pretreatment with silymarin, particularly at middle and high doses, significantly upregulated the activities compared with the TP group (P<0.05). Furthermore, silymarin significantly elevated the CAT and GST activities to levels that surpassed those in the NC group (Fig. 3A and E). Therefore, silymarin exhibited excellent effects in mitigating TP-induced systemic antioxidant enzyme exhaustion.

The hepatic MDA level was determined to evaluate the degree of lipid peroxidation (Fig. 3F). MDA content increased significantly in the liver of the rats in the TP group compared with the NC group (P<0.01). However, silymarin significantly suppressed this excessive accumulation of MDA compared with the TP group (P<0.01), and no significant difference was observed between the NC and high-dose silymarin group. Therefore, silymarin markedly suppressed TP-induced lipid peroxidation, which was consistent with the improvement in antioxidant enzymatic activities.

An excessive inflammatory response is associated with oxidative stress (22). In the present study, the levels of important inflammatory cytokines, TNF-α, IL-1β, IL-6 and IL-10, in the liver were determined. The levels of these inflammatory mediators were significantly higher in the TP group compared with the NC group (Fig. 4), indicating that TP induced severe inflammatory response in the liver. Promisingly, pretreatment with silymarin significantly inhibited the production of these proinflammatory factors in a dose-dependent manner compared with the TP group (P<0.01). These results indicate that silymarin may attenuate TP-induced liver injury by suppressing the inflammatory response.

To illustrate the underlying mechanism of silymarin inhibiting inflammatory responses, the protein expression of key downstream proteins in the TNF-α pathway, and the apoptosis-associated protein c-caspase-3, was measured (Fig. 5A). Expression levels of p-p38 in the TP group were 3.5 times higher compared with the NC group (P<0.01). However, the overexpression of p-p38 induced by TP was reduced by 20.04, 33.11 and 48.44% in silymarin treated groups in a dose-dependent manner (Fig. 5B). A similar inhibitory effect of silymarin was also observed in the expression of p-JNK. The level of p-JNK in TP group was three times higher compared with the NC group (P<0.01). In silymarin pretreated groups, the level of p-JNK decreased by 8.82, 14.93 and 61.46% in low, middle and high-dose groups, respectively, compared with the TP group (Fig. 5C). Therefore, the results indicated that silymarin, particularly at 100 and 200 mg/kg, significantly inhibited the phosphorylation of inflammatory mediators that are induced by TNF-α. In addition, the protein expression of c-caspase-3, a protein associated with apoptosis, was also determined by western blot analysis. Following TP injection, the level of c-caspase-3 in the TP group was 2.24 times higher compared with the NC group (P<0.01). In silymarin-treated groups, the expression level of c-caspase-3 was significantly decreased by 10.55, 22.93 and 29.73% in silymarin low, middle and high-dose groups, respectively, compared with the TP group (P<0.01; Fig. 5D). These results indicate that silymarin may also exert antiapoptotic effects in the liver of TP-treated rats.

The apoptosis rates of different treatment groups were detected using TUNEL staining and quantified by ImageJ software, and representative photographs were taken under a light microscope. As demonstrated in Fig. 6, TP administration resulted in numerous TUNEL-positive hepatocytes and a significantly higher apoptotic index compared with the NC group (Table II), indicating the induction of severe hepatic apoptosis within 24 h of treatment with TP. However, fewer TUNEL-positive hepatocytes and significantly decreased apoptotic indexes (P<0.01) were observed in livers pretreated with silymarin compared with the TP group, indicating that silymarin pretreatment may protect the liver against TP-induced hepatic injury. Notably, high-dose silymarin pretreatment exhibited a superior effect in inhibiting the cell apoptosis. These results indicated that silymarin may effectively ameliorate the hepatocyte apoptosis induced by TP.

Representative photographs of different treatment groups were taken under a light microscope. The positive rates were calculated by ImageJ software based on 15 images of each group. No significant alterations in the expression of the antiapoptotic protein Bcl-2 were observed between NC, TP and silymarin treatment groups (Fig. 7 and Table II). By contrast, the expression of the proapoptotic protein Bax was significantly higher in the TP group compared with the NC group (P<0.01; Fig. 8 and Table II). However, silymarin significantly inhibited the expression of Bax and, therefore, the ratio of Bax/Bcl-2 was also lowered, compared with the TP group in a dose-dependent manner (P<0.01; Fig. 8 and Table II). As described above for western blot results, the changes in the levels of c-caspase-3 exhibited a similar pattern to Bax expression (Fig. 5). Furthermore, Cyt C levels in supernatants were measured by ELISA, and levels were significantly increased in the TP group compared with the NC group (P<0.01; Table II). However, silymarin pretreatment was observed to progressively reduce hepatic Cyt C levels in a dose-dependent manner compared with the TP group (P<0.01). These results are consistent with those for TUNEL staining and c-caspase-3 protein expression, indicating that an antiapoptotic effect may contribute to the protective action of silymarin in the liver of TP-treated rats.