Silymarin was purchased from Sigma Chemicals Co. (St. Louis, MO, USA). The MCD diet was obtained from Research Diets, Inc. (New Brunswick, NJ, USA).

Male Otsuka Long-Evans Tokushima Fatty (OLETF) rats, an established animal model of obese type 2 diabetes (8), were used (Otsuka Pharmaceutical, Tokushima, Japan). OLETF rats were reported to show obesity and hyperinsulinemia from 8 weeks of age, and demonstrated the hepatic accumulation of fat in an age-dependent manner (17,18). As the control animals, Long-Evans Tokushima Otsuka (LETO) rats, which originated from the same colony as the OLETF rats by selective mating and did not develop diabetes, were used. Both OLETF and LETO rats were 4-weeks-old and were housed in a room under controlled temperature (23°C), humidity, and lighting (12-h artificial light and dark cycle). Animals were given free access to the standard laboratory rat chow and tap water. At 24 weeks of age, rats were divided into experimental groups and fed for 8 weeks. OLETF rats were fed on one of three different diets as follows: the standard laboratory rat chow (OLETF/vehicle, n=10), the MCD diet (OLETF/MCD, n=10), and the MCD diet mixed with silymarin (OLETF/MCD+silymarin, silymarin content 0.5% w/w, n=10). LETO rats were continued to be fed on the standard laboratory rat chow (LETO/vehicle, n=10). Although all rats were allowed unrestricted access to water, OLETF and LETO rats, fed on the standard chow were pair-fed with either the MCD or with MCD and silymarin. The body weight and food intake in each group of rats were recorded every week. All animal procedures were performed in accordance with the guidelines set by the Institutional Animal Care and Use Committee at Inha University School of Medicine. After 8 weeks of feeding on experimental diets, rats were sacrificed.

Sections of liver tissue specimens, fixed in 10% formalin and embedded in paraffin wax, were stained with H&E and Masson’s trichrome for histological examination. A blinded investigator (J.M.K.) evaluated the slides for fatty change, inflammation existence of hepatocyte ballooning and fibrosis as described in previous studies with minor modifications (19–22). The degree of steatosis was scored as the percentage of hepatocytes containing macrovesicular fat (grade 0, no steatosis; grade 1, <25%; grade 2, 26–50%; grade 3, 51–75%; grade 4, 76–100%). Inflammation was histologically quantified by counting inflammatory foci in 20 consecutive high-power fields (40x objective) (average histological grade, grade 0, no foci; grade 1, <2 foci per high-power field; grade 2, ≥2 foci per high-power field). The individual scores of steatosis, inflammation and hepatocyte ballooning were added to produce an overall score, namely NAFLD activity score (NAS) as previously suggested (21). Fibrosis scores were as follows: 1, pericellular and perivenular fibrosis; 2, focal bridging fibrosis; 3, bridging fibrosis with lobular distortion; and 4, cirrhosis.

Sections of liver tissue specimens were immunostained with mouse anti-human α-smooth muscle actin (α-SMA) (Dako, Carpinteria, CA, USA). Detection of the primary antibody was carried out by immunoperoxidase technique using the ABC kit (Vector Laboratories). Peroxidase activity was identified by reaction with diaminobenzidine tetrahydrochloride substrate (DAB). Data are represented as the number of α-SMA positive cells present in thirty 40x fields (1.3 mm², approximately 3,000 hepatocytes).

It has been previously reported that HSCs could be isolated from the injured liver using a conventional density gradient centrifugation method (23). In this study, HSCs were isolated from the livers of each experimental group by in situ perfusion using collagenase and pronase (24). The viability and purity of HSC preparations were consistently found to be >95% as accessed via trypan blue (Gibco-BRL, Grand Island, NY, USA) exclusion and autofluorescence, respectively (25,26).

Total-RNA was extracted from either frozen whole liver or isolated HSCs using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. RNA samples were quantified by spectrophotometry. The RNA integrity was assessed using agarose gel electrophoresis and ethidium bromide staining. The RNA samples were then diluted in RNase-free water and stored at −70°C until use.

RNA (5 μg) were reverse-transcribed using the RNA PCR kit version 1.2 (Takara Bio, Inc., Japan) according to the manufacturer’s recommendations. Oligonucleotide primers and the TaqMan probe for α₁-procollagen, α-SMA, sterol regulatory element-binding protein-1c (SREBP-1c), tumor necrosis factor-α (TNF-α) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) internal control were obtained from Perkin-Elmer Applied Biosystems (Foster City, CA, USA), purchased as a ready-for-use form in Assays-on-Demand Gene Expression products. The TaqMan probe was labeled at the 5′ end with the reporter dye FAM and at the 3′ end with the quencher TAMRA. PCR was carried out in triplicate for each sample on a Bio-Rad iCycler Optical Module (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Each 20-μl reaction contained 10 μl of TaqMan Universal PCR Master Mix, 1 μl of Assays-on-Demand Gene Expression Assay Mix and 9 μl of cDNA diluted in RNase-free water. All reactions were carried out using the following cycling parameters: 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 1 min. mRNA fold changes in target genes relative to the endogenous GAPDH control were calculated as suggested in previous studies (27).

The cytosolic and nuclear fraction of the protein was extracted from liver tissue using a protein extraction kit (Bio-Rad Laboratories, Inc.) according to the manufacturer’s instruction. Briefly, 50 mg of liver tissue was added into the chilled the Dounce homogenizer with 0.75 ml of cytoplasmic protein extraction buffer (CPEP) and was broken up by stroking. After incubating Dounce homogenizer containing the homogenate on ice for 2 min, the supernatant was transferred using a pipette. The cell lysate was centrifuged at 1,000 x g for 10 min at 4°C. Upon completion of the centrifugation, the supernatant containing cytoplasmic protein was transferred and centrifuged, and the pellet containing nuclei was washed using CPEP. The protein concentration was determined with a Lowry protein assay (Bio-Rad Laboratories, Inc.).

Whole cell extracts were prepared from HSCs which were treated as described above by using Triton lysis buffer containing protease and phosphatase inhibitors as described else where (9). Total hepatic or whole cell extract protein was estimated using bovine serum albumin as a standard (DC protein assay; Bio-Rad Laboratories, Inc.).

Whole cell protein (50 μg) or cytosolic/nuclear protein (50 μg) was separated by 10% sodium dodecyl sulfatepolyacrylamide gel electrophoresis and the resolved proteins were transferred to a nitrocellulose membrane (Schleicher and Schuell, Middlesex, UK). The membrane was blocked with 5% skim milk in 10 mM Tris-HCl containing 150 mM NaCl and 0.5% Tween-20 [Tris-buffered saline (TBS)-T]. After washing with TBS-T, the membrane was then incubated with 1:1,000 dilution of specific primary antibodies against phospho-extracellular signal-related protein kinase (ERK) 1/2 for whole cell protein and nuclear factor erythroid 2-related factor 2 (Nrf2) (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) for the cytosolic/nuclear protein. The membrane was washed and then incubated with a horseradish peroxidase (HRP)-conjugated secondary antibody (New England Biolabs, Beverly, MA, USA) diluted 1:2,000. After washing with TBS-T, the membrane was developed using an enhanced chemiluminescence detection kit (Amersham, Piscataway, NJ, USA). Anti β-actin Ab (Cell Signaling Technology, Inc., Danvers, MA, USA) was used to verify the equal loading of the protein samples.

All results are expressed as means or means ± standard deviation of the mean (SD). Data were analyzed by nonparametric analysis (Kruskal-Wallis or Mann-Whitney U test) and P<0.05 was considered statistically significant. All calculations were performed with SPSS version 12.0 software for Windows (SPSS Inc., Chicago, IL, USA).

The initial weight of LETO rats, non-diabetic control animals for OLETF, was significantly lower as expected. The MCD diet-fed rats had lower final body weight compared to the standard chow-fed rat groups after 8 weeks despite the pair feeding as previously reported (8,28,29). However, the general condition of the MCD-fed rats remained healthy. Relative liver weight (liver weight/final body weight) of the MCD-fed rats was significantly higher compared to that of the standard chow-fed rats. Adding silymarin failed to reverse the increased relative liver weight (Table I).

The liver-protective effect of silymarin has been attributed to its role as an antioxidant (30,31). Since Nrf2 is known to be crucial for several antioxidant responsive elements, resulting in relief of the oxidative burden of cells (32,33), nuclear translocation of Nrf2 was evaluated by western blotting of the cytoplasmic and nuclear protein. Cytoplasmic Nrf2 protein was increased in both MCD diet fed rats with or without silymarin. However, nuclear Nrf2 protein was markedly increased only in the livers of the OLETF/MCD+silymarin group (Fig. 1).

Histological analysis of NAFLD was performed as suggested by other studies (19–22). Fatty change, inflammation, and existence of hepatocyte ballooning degeneration was assessed and scored separately. Individual scores were added to produce the overall NAS. Feeding MCD diet to OLETF rats resulted in probable or definite NASH as expected. Concomitant administration of silymarin lowered NAS (Table II).

Histological evaluation of liver fibrosis was performed after H&E and Masson’s trichrome stain. Despite the significant fatty changes, insulin-resistant OLETF rats with ordinary diet did not demonstrate liver fibrosis. The MCD diet induced mild liver fibrosis confined to fibrosis stage 1. Concomitant administration of silymarin with MCD diet could not completely block the appearance of liver fibrosis (Fig. 2A). Effect of silymarin on expression of α₁-procollagen mRNA was assessed in the whole liver.

In accordance with the histological analysis, α₁-procollagen mRNA expression was not enhanced in OLETF rats without MCD diet when compared with LETO rats. Administration of the MCD diet significantly increased the expression of α₁-procollagen mRNA in the whole liver lysates, which decreased after silymarin administration (Fig. 2B).

HSC activation was evaluated in the liver by immunohistochemical analysis and assessment of α-SMA mRNA expression in the whole liver lysates. Feeding the OLETF rats with the MCD diet increased α-SMA positive cells in the liver and concomitant administration of silymarin with the MCD diet significantly decreased these cells (Fig. 3A). There was no significant difference in α-SMA positive cells between LETO and OLETF rats fed with standard diet. Expression of α-SMA mRNA in the liver lysates correlated with the result of the immunohistochemical study (Fig. 3B).

In order to investigate the direct effect of silymarin on HSCs of MCD fed OLETF rats, HSCs were isolated from the experimental animal. Expression of α-SMA mRNA on HSCs from OLETF rats without the MCD diet was compatible with that of the LETO rats. HSCs from MCD fed OLETF rats showed increased α-SMA mRNA levels which were significantly alleviated by adding silymarin (Fig. 3C). This result correlated with the data obtained from the evaluation of the whole liver α-SMA positive cells and liver lysates.

Expression of α₁-procollagen mRNA was increased in HSCs from MCD fed OLETF rats while no such change was noticed in both OLETF rats without MCD diet and LETO rats. Giving silymarin with the MCD diet significantly relieved this increase in α₁-procollagen mRNA expression on HSCs (Fig. 2C).

The role of silymarin on inflammatory cytokine TNF-α was investigated. Expression of TNF-α mRNA on the whole liver of MCD diet fed OLETF rats significantly increased and was further aggravated by the administration of the MCD diet. When silymarin was administered with the MCD diet, expression of TNF-α mRNA was markedly reduced in the liver (Fig. 4A). We examined whether a decrease in TNF-α expression in the liver was accompanied by reduced ERK activation in HSCs. Phosphorylated ERK1/2 protein expression was increased in HSCs isolated from MCD-fed OLETF rats. The effect was reversed by concomitant feeding of silymarin with the MCD diet (Fig. 4B).

The effect of silymarin on SREBP-1c mRNA expression, a transcriptional factor central to the regulation of lipid metabolism, was assessed in the whole liver. Expression of SREBP-1c mRNA was increased in the liver of OLETF rats compared to that of the LETO. Feeding with the MCD diet further increased the expression of SREBP-1c mRNA compared to that of OLETF rats with normal diet. However, concomitant administration of silymarin with the MCD diet could not reduce the expression of SREBP-1c on the transcriptional level in the whole liver tissue (Fig. 4C). There was no significant difference in SREBP-1c mRNA expression on HSCs among four experimental groups.