Acacia mearnsiid De Wild bark was obtained from Guangxi Wuming Tannin Technology Co., Ltd. (Nanning, China). Lipoprotein-free calf serum, AICAR, 1,1-dimethylbiguanide hydrochloride (metformin) and protease inhibitor cocktail were obtained from Sigma (St. Louis, MO, USA). HALT phosphatase inhibitor and a bicinchoninic acid (BCA) protein assay kit were obtained from Peking Luqiao Bio-tech Co. (Beijing, China). Dulbecco’s modified Eagle’s media (DMEM), penicillin-streptomycin-glutamine, trypsin and fetal bovine serum (FBS) were obtained from Qi’ao Bio-tech Co. (Shanghai, China). Radiochemicals (¹⁴C-acetate at 56 mCi/mmol, ¹⁴C-mevalonate at 65 mCi/mmol and ¹⁴C-hydroxymethylglutaryl-CoA at 55 mCi/mmol) were obtained from American Radiolabeled Chemicals (St. Louis, MO, USA).

Extracts of Acacia mearnsiid De Wild were obtained via aqueous acetone (1:1) immersion of bark drillings (10 kg) for seven days. The acetone was evaporated and the aqueous solution was freeze dried, yielding a yellow-brown powder (554 g) which was partitioned into four 80-g samples between a butan-2-ol-water-hexane (5:4:1) mixture in a 20-tube, 100 cm³ under phase, Craig countercurrent assembly (Shanghai Chemical Machinery Plant Co., Ltd. Shanghai, China). Qualitative paper chromatographic analysis produced five main fractions. Subsequent column chromatography of fraction 3 (52 mg) on Sephadex LH-20 (5×170 cm column, flow rate 20 cm³/30 min) yielded 12 fractions. Sub-fraction Fr-3–5 (1020 mg) was further purified into five main fractions via semi-preparation chromatography. The most active fraction, Fr-3-5-2 (Rf 0.34, 43 mg), underwent further purification. Acacia mearnsiid De Wild bark tannins contained 93% RBF by high performance liquid chromatography, and contained M⁺, 700.2518 (theoretical molecular mass of C₃₉H₄₀O₁₂, 700.2520), in the peracetate mass spectrum (Thermo Fisher Scientific Inc., Waltham, MA, USA). Purified RBF was stored at 4°C.

Male Sprague-Dawley rats (weight, 200±25 g; Dossy Biological Technology Co., Ltd., Chengdu, China) were housed in groups of five under standard environmental conditions (23±10°C, 55±5% humidity and a 12 h light/dark cycle) and had free access to water and a standard diet, which consisted of AIN-93G, (Beijing Lianlixin Biological Technology Co., Ltd.) containing 200 g/kg casein lactic, 3 g/kg L-cystine, 397 g/kg corn starch, 132 g/kg maltoes dextrin, 100 g/kg sucrose, 50 g/kg cellulose, 70 g/kg soybean oil, 35 g/kg mineral mix, 10 g/kg vitamin mix and 2.5 g/kg choline bitartrate. All animal experimentation was performed in strict accordance with the recommendations set out in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Animal Care and Use Committee of the Sichuan University (Chengdu, China; permit no. 24). All surgery was performed under sodium pentobarbital anesthesia and all efforts were made to minimize suffering.

The forty rats were randomly assigned to four groups: A negative control group (NC group); a positive control group, fed a high-fat diet (HF group); a low dose RBF group (60 mg/kg/day); and a high dose RBF group (120 mg/kg/day). The rats of the NC group were fed with a standard diet and orally treated with 0.3% Tween-80 solution (10 ml/kg). The rats of the HF and RBF groups were fed with a high fat diet, comprising 7.5% pork oil (Chengdu Zhiwei Food Co., Ltd., Chengdu, China), 1% cholesterol (Sigma Chemical Co., St. Louis, MO, USA), 10% egg yolk powder (Chengdu Aimeiai Food Co., Ltd., Chengdu, China), 0.3% chocolate (COFCO Le Conte Food Co., Ltd., Shenzhen, China), 0.2% propylthiouracil (Sigma Chemical Co., St. Louis, MO, USA) and 81% standard diet, and gavaged with 0.3% Tween-80 solution (Sigma Chemical Co., St. Louis, MO, USA) or RBF solution, respectively, daily for six weeks.

McA-RH-7777 rat hepatoma cell lines, obtained from Shanghai Maisha Bioscience Co. (Shanghai, China), were cultured in DMEM supplemented with 100 U/ml penicillin and 100 lg/ml streptomycin (Sigma-Aldrich, St. Louis, MO, USA) containing 10% FBS. The cells were incubated at 37°C in a 5% CO₂ atmosphere.

At the end of the treatment period, rats were kept for overnight fasting and then given the final gavage. Rats were weighed and anesthetized with 36 mg/kg sodium pentobarbital (Sigma-Aldrich) by intraperitoneal injection to collect blood samples from the aorta ventralis 1.5 h after the gavage. Following blood collection from rats sacrificed by anoxia, the livers were removed and weighed in order to calculate the liver index (liver weight/body weight). The plasma was separated by centrifugation at 1,000 × g for 15 min at 4°C. Plasma samples were stored at −80°C until required for analysis.

Serum total cholesterol (TC) and triglycerides (TG) were measured in a spectrophotometer (DU800, Beckman Coulter, Pasadena, CA, USA) using commercial kits (Jiancheng Institute of Biotechnology, Nanjing, China) according to the manufacturer’s instructions (CHODPAP and GPO-PAP methods). The optical density of the samples was measured on a spectrophotometer at a wavelength of 546 nm. Total levels TC and TG in the serum were calculated by dividing absorption of the tested sample by that of the standard sample and multiplying by the TG and TC content of the standard sample. Serum low density lipoprotein cholesterol (LDL-C) levels were evaluated according the manufacturer’s instructions (polyethylene sulfate precipitation method) to evaluate the difference between cholesterol in the supernatant and total cholesterol.

Following culture for 48 h in six-well plates, hepatoma cells were treated with ¹⁴C-acetate or ¹⁴C-mevalonate for 3 h. The RBF concentration (1, 2, 3, 4 or 5 mM) was included in incubations with ¹⁴C-mevalonate in order to inhibit the activity of HMG-CoA reductase. The medium was removed and cells were washed three times with phosphate-buffered saline (PBS), harvested by trypsin, resuspended in 20 mM Tris buffer (pH 7.4, containing 0.1% Triton X-100; Sigma-Aldrich) and lysed by a sonication three times for 5 sec (Shanghai Bilon Instruments Co., Shanghai, China) on a medium setting. Lipids were extracted into 5 ml chloroform/methanol (2:1). The medium was reduced to near dryness using Savant Speed Vac SC100 apparatus (Selby Anax, Adelaide, Australia). Dried extracts were resuspended in 50 μl chloroform/methanol and resolved by silica thin-layer chromatography (Whatman, Plc., Little Chalfont, UK) in petroleum ether/ethyl ether/acetic acid (60:40:1). Radiolabeled cholesterol was identified by comparison with standards visualized by iodine-vapor stain and quantified by electronic autoradiography (Packard Instant Imager, PerkinElmer Inc., Waltham, MA, USA).

Microsomal fractions were prepared from livers of the rats maintained on rat standard diet containing 5% cholestyramine for seven days. HMG-CoA reductase was solubilized from the microsomal fractions according to the method described by Heller and Shrewsbury (17) and purified through the second ammonium sulfate precipitation step described by Kleinsek et al (18). The enzyme preparation was stored at −80°C in 100 μl aliquots until required for analysis. Prior to use, the enzyme was activated at 37°C for 30 min. The assay was adapted from methods described in the study by Beg et al (19). In brief, the reaction mixture contained 100 μl of 0.14 M potassium phosphate buffer, pH 6.8; 0.18 M KCl; 3.5 mM EDTA, pH 7.0; 10 mM dithiothreitol; 0.1 mg/ml bovine serum albumin; 0.04 μCi of ¹⁴C-HMG-CoA; and 8 μg partially purified enzyme (specific activity 14 nmol/min/mg) with or without RBF. Following incubation for 5 min at 37°C, the reaction was initiated with 0.2 mM NADPH and terminated with 20 μl of 5 M HCl. Following an additional incubation for 30 min at 37°C to allow for complete lactonization of the product mevalonate, the mixture was passed through a column with an internal diameter of 0.5 cm and a length of 5 cm, which contained 100–200 mesh Bio-Rex, chloride form (Bio-Rad, Hercules, CA, USA). Any unreacted ¹⁴C-HMG-CoA was adsorbed onto this mesh. The product of ¹⁴C-mevalonolactone was eluted into scintillation vials with 3 ml distilled water. After the addition of 15 ml Aquasol II (Sigma-Aldrich), the radioactivity of the samples was measured.

For western blot analysis hepatoma cells were grown in 6-cm plates for 3–5 days with a daily medium change. Cells were treated with RBF or an AMP-kinase activator (AICAR or metformin, 1 mM final concentration) for 3 h in fresh media. Subsequent to the incubation period, cells were washed once with ice-cold PBS and removed with lysis buffer composed of, 0.25 M Tris buffer (pH 7.5), and protease and phosphatase inhibitor cocktails at standard concentrations. Lysates were centrifuged at 1,300 × g for 10 min at 4°C and the supernatant was stored at −80°C. Protein (30 μg, identified by BCA analysis) was fractionated by SDS-gel (8%) electrophoresis and transferred onto to a nitrocellulose membrane (Whatman; Sigma-Aldrich) The membrane was blocked with 0.05% Tween-20 and 5% non-fat milk. For immunodetection, a rabbit antibody against total AMP-kinase (polyclonal anti-AMPK α-pan, 1:2,000; Upstate Biotechnology Inc., Charlottesville, VA, USA) or phosphorylated AMP-kinase (anti-phospho-AMPKα, 1:500; Upstate Biotechnology Inc.) were used overnight at 4°C. Antibody binding was detected with a horseradish peroxidase (HRP)-conjugated polyclonal secondary goat anti-rabbit antibody (1:1,000; Chengdu Aoxin Biotechnology Co., Ltd., Chengdu, China) for 1 h at room temperature. The chemiluminescent image (Supersignal West Pico Chemiluminescent Substrate, Pierce Biotechnology, Rockford, IL, USA) was captured by autoradiography. Band densitometric analysis was conducted using Image J 1.416a (National Institues of Health, Bethesda, MD, USA). In order to determine the degree of HMG-CoA reductase phosphorylation, lysates were incubated for 1 h at 4°C with 20 μl mouse monoclonal antibody against phosphoserine/phosphothreonine/phosphotyrosine (51AB Biotech Co., Shanghai, China). Antibody conjugates were precipitated with 35 μl of protein G Plus-Agarose from Calbiochem (La Jolla, CA, USA) following Amersham Biosciences (Chalfont St. Giles, UK) immunoprecipitation instructions. Immunoprecipitated protein was released by boiling in a gel loading buffer, fractionated by SDS-gel electrophoresis and transferred to a nitrocellulose membrane. HMG-CoA reductase was detected and quantified via rabbit polyclonal HMG-CoA reductase antibody (Upstate Biotechnology Inc.), followed byincubation with an HRP-conjugated secondary goat anti-rabbit antibody, as described above.

All values are expressed as the mean ± standard deviation. The groups were compared using unpaired Student’s t-test. P<0.05 was considered to indicate a statistically significant difference.

Following a six week hypercholesterolemic diet, the serum TC in rats increased by 166%, from 52.55±2.53 to 140.05±6.77 mg/dl. Treatment with 120 mg/kg RBF significantly reduced cholesterol levels in the treated groups, with serum cholesterol being 25.9% lower than that in the positive control groups (Table I).

No changes in the serum levels of LDL cholesterol and HDL cholesterol were observed in the sham groups fed with a standard diet and treated with RBF, compared with the negative control group. In the treatment groups, 120 mg/kg RBF reduced mean serum LDL-C levels by 50.8%, from 79.93±6.59 to 38.72±4.11 mg/dl, whereas no significant changes were detected in HDL-C levels (51.81±3.28 to 47.01±2.91 mg/dl, P>0.05). RBF treatment therefore resulted in a decrease in the LDL/HDL ratio of 66% in hypercholesterolemic rats (Table I), indicating that RBF had a greater effect on LDL-C than HDL-C levels.

In a 3 h assay, addition of RBF to McARH7777 rat hepatoma cell cultures decreased the incorporation of ¹⁴C-acetate and ¹⁴C-mevalonate into cholesterol (Fig. 2). RBF-mediated inhibition of ¹⁴C-acetate incorporation into cholesterol was evident at 35 μg/ml and reached 72% at 100 μg/ml. By contrast, even at the highest concentration of RBF tested, inhibition of ¹⁴C-mevalonate incorporation did not exceed 15%. As mevalonate is the product of HMG-CoA reductase, these results suggest that RBF acts at or above the level of HMG-CoA reductase.

In order to detect the effect of RBF on the activity of HMG-CoA reductase, this rate-limiting enzyme in cholesterol biosynthesis was also investigated. As shown in Fig. 3, RBF inhibits HMG-CoA reductase activity in a dose-dependent manner and the concentration of RBF required for 50% inhibition of enzymatic activity was ~3 mM.

To investigate the mechanism underlying the inactivation of HMG-CoA by RBF, the phosphorylation of HMG-CoA reductase in hepatoma cells was measured with and without the addition of RBF. As shown in Fig. 4, phosphorylated HMG-CoA reductase increased by 47% and 227% after 3 and 6 h of RBF treatment, respectively. To further elucidate the pathways involved in RBF downregulation of HMG-CoA reductase activity, it was determined whether the activation by RBF of AMP-kinase via phosphorylation affected this process. As shown in Fig. 5, RBF administration tripled AMP-kinase phosphorylation, with the maximal effect at 10 μg/ml, indicating that its potency in terms of AMP-kinase activation may be on par with AICAR or metformin, which are known to activate AMP kinase by phosphorylation (21). It may be noted in Fig. 5B that AMP-kinase activation after RBF administration rapidly increased from 0–3 h, and then maintained a slight elevation over time (Fig. 5B).