Animals and treatment
Fifty-two male Sprague-Dawley (SD) rats at 6 weeks of age and weighing 180-220 g were from the Experimental Animal Center of Guangdong Province (China), and housed in a specific pathogen free facility maintained at a cycle of 12 h light/dark and a constant temperature of 22°C~26°C, and a relative humidity of 65 ± 15%.
The rats were randomized and fed with normal chow diet (n = 14, 10% of calories derived from fat; D12450B) or high fat diet (HFD, n = 38, 60% of calories derived from fat; D12492) for 12 weeks to induce NAFLD with hyperlipidemia, as described previously [11, 12]. At the end of the 12-week induction, six rats from normal chow diet group or HFD group were randomly chosen and sacrificed. Their liver histopathology was analyzed for the development of NAFLD. Subsequently, the 32 HFD-fed rats were further randomized into four groups and continually fed with HFD (model group), with HFD and treated P.O with 30 mg/kg of ATO (Sigma-Aldrich, St. Louis, MO) (ATO group), with normal chow diet (dietary control, DC group), or with normal chow diet and treated with the same dose of ATO (combination of dietary control with ATO, DCA group) daily for 8 weeks, respectively. Rats fed with normal chow diet without exposure to HFD were used as the normal controls.
At the end of dietary control and ATO treatment, the rats were fasted for 12 hours and sacrificed. Their blood samples were collected from the abdominal aorta, and their sera were prepared by centrifugation, frozen, and stored at -20°C until analysis. Their livers were frozen in liquid nitrogen and stored at -80°C until gene expression analysis. A portion of the liver from individual rats was fixed overnight in 10% formalin for histological analysis. Two additional samples of liver tissues (150 mg) were stored at -80°C for Westernblot analysis assay and liver lipids measurement. The experimental protocols were approved by the Animal Care and Protection Committee of Sun Yat-Sen University.
Analyses of serum lipids and transaminases
The concentrations of serum total cholesterol (TC), triglyceride(TG), low density lipoprotein-cholesterol (LDL-C), high density lipoprotein-cholesterol (HDL-C), aspartate aminotransferase (AST), and alanine aminotransferase (ALT) were measured using the corresponding commercial enzyme kits (Biosino, Beijing, China) on an automatic biochemistry analyzer (Olympus AU600,Tokyo, Japan). The levels of serum free fatty acid (FFA) were assayed using a commercial kit, according to the manufacturer's instruction (R&D, Minneapolis, USA).
Liver triglyceride and cholesterol
Total lipids were extracted from 100 mg of liver tissues, according to the method of Bligh and Dyer . Briefly, total lipids in liver tissues were extracted with chloroform-methanol (2:1). After evaporation overnight, the extracted lipids were re-suspended in 10% triton and isopropanol, and quantified using a commercial enzyme assay kit (Biosino, Beijing, China) on an automatic biochemistry analyzer.
RNA extraction and quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)
The relative levels of sterol regulatory element binding protein 1c (SREBP-1c), fatty acid synthase (FAS), acetyl-CoA carboxylase (ACC), peroxisome proliferator-activated receptor alpha (PPARα), acyl-CoA oxidase (ACO), carnitine palmitoyltransferase-1 (CPT-1), and mRNA transcripts to control glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were determined by qRT-PCR. Briefly, total RNA was extracted from individual liver samples using Trizol reagent (Invitrogen, USA) and reversely transcribed into cDNA using the super-Transcript kit, according to the manufacturer's protocol (Genecopoeia, USA). Subsequently, the cDNA was used as the template for characterization of the relative levels of mRNA transcripts of individual target genes to control GADPH by qRT-PCR using SYBR Green fluorescence and the specific primers on an iCycler thermocycler (BioRad Hercules, CA). The sequences of the primers were synthesized, according to the previous studies[28, 29] and are shown in Table 4. The PCR reactions were performed in duplicate at 95°C for 10 min and subjected to 35 cycles of 95°C for 10 s, 60°C for 20 s, and 72°C for 15 s, followed by extension at 72°C for 10 min. The values of threshold cycle (Ct) were determined by automated threshold analysis using Opticon Monitor 3.1 software. The relative levels of each gene expression were determined by the 2-ΔΔCt method.
Western blot analysis of HMG-CoA reductase
The expression of hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase
in hepatic tissue was determined by Western blot assay. Briefly, frozen liver samples were homogenized in RIPA lysis buffer containing protease and phosphatase inhibitors. After centrifuged, the protein contents in the lysates were measured using a BCA protein assay kit (Beyotime, China). Individual lysates (30 ug) were separated by 10% SDS-PAGE and electrotransferred to polyvinylidene difluoride (PVDF) membranes. After being blocked with 5% non-fat milk in Tris-buffered saline with 0.1% Tween-20 (TBST, 25 mM Tris, pH 8.0, 137 mM NaCl, 2.7 mM KCl, and 0.1% Tween-20) at room temperature for 60 min, the membranes were incubated with anti- HMG-CoA reductase or anti-GAPDH antibodies (Santa Cruze Biotechnology, Santa Cruze, USA) overnight at 4°C. The bound antibodies were detected with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG antibodies (Santa Cruze Biotechnology) and visualized using the enhanced chemiluminescence (ECL, Pierce, Rockford, IL) system and exposed to X-ray film (Kodak, Japan).
Histological analysis of the liver
After eight weeks of intervention, the rats were sacrificed and their livers were fixed by 10% buffered formalin and embedded in paraffin. The liver sections (5 μm) were stained with hematoxylin and eosin (H&E), and the steatoic degrees of individual liver samples were examined and scored, according to the percentage of hepatocytes containing lipid droplets  by a pathologist in a blinded manner.
Values were expressed as mean ± SD. The difference among groups was analyzed by one-way ANOVA and Students t-test. The association among variants was analyzed by the least significant difference (LSD) test using the SPSS 13.0 software. The accepted level of significance was p < 0.05.