Twenty eight young male New Zealand White rabbits aged 5 weeks were obtained from Hanil Animal Laboratory Co. (Wanju, Korea). Rabbits were individually housed in metal cages in an air-conditioned room (22±1℃). Rabbits were fed Purina® rabbit diet (Agribrands Purina Korea, Korea) for 2 weeks for adaptation. Rabbits had free access to tap water. After two weeks, all rabbits were isolated and fed hypercholesterol diet (HD: cholesterol 0.5% w/w; Sigma Chemical, St. Louis, MO, U.S.A.). The diet was prepared as pellet and continued for 4 months for provoking atherosclerosis. After provoking, 7 animals were euthanized for lesion evaluation and remaining 21 animals were randomly divided into 3 groups: (1) 7 on ND, (2) 7 on HD, and (3) 7 on HD plus Genistein diet (Genistein 1.65 mg/kg of rabbits; Oncogene, Fremont, CA, U.S.A.). Rabbits were fed the respective diets for another 4 and half months. At the end of 4 and half months all animals were sacrificed for evaluation of the aorta (Table 1).

The aortas were removed and both proximal and distal 1 cm of the aorta were freezed by liquid nitrogen and stored at -80℃ for molecular study. Others were fixed in 10% buffered formalin solution. Two sections of 0.5 cm-length each from the proximal, middle and distal portion (total 3 sections) were embedded in paraffin (Paraplast, Oxford, St. Louis, MO, U.S.A.) for light microscopy and immunohistochemistry. Sections taken from each artery were stained with Verhoeff-Van Gieson staining (18) and evaluated the development of atherosclerotic lesion by image analyzer using Visus Image Analysis System (Image and Microscopy Technology, Korea). In short, we measured the round area consist of boundary line along the intimal elastic lamina, and then we subtracted the luminal area from each area. We considered it was a lesion area. The lesion size was calculated from nine sections (three sections from each block), and expressed as mean±SD.

Sections taken from each artery were evaluated for RAM 11 (DAKO Corporation, Carpinteria, CA, U.S.A.) and HHF-35 (DAKO) expression using an each monoclonal antibodies. Simultaneous staining of slides from all groups has performed. Four µ-thick paraffin sections from the arteries were incubated with 3% hydrogen peroxide for five minutes, then with 0.5% casein-Tris buffer for thirty minutes for blocking nonspecific binding. The slides were incubated with anti-RAM 11 and HHF-35 for two hours. After incubation with species appropriate biotinylated secondary antibodies (Vector Laboratories, Vectorstain ABC kit; Vector Laboratories, Burlingame, CA, U.S.A.) for 30 min, washed and incubated with streptavidin conjugated horseradish peroxidase (HRP) for 30 min. HRP visualization was carried out using 3,3'-diaminobenzidine tetrahydrochloride (ScyTek, Logan, Utah, U.S.A.) as substrate. Sections were counterstained with Light Green solution. Slides in which primary antibodies were omitted served as negative controls for each antibody used in this study.

The brown-stained areas by anti-RAM 11 and anti-HHF-35 antibodies were regarded as macrophages and smooth muscle cell, respectively, and measured by the image analyzer.

Each tissue were homogenized in ice-cold buffer of 50 mM Tris-HCl, pH 7.4/1 mM EDTA containing antipain (10 µg/mL), leupeptin (10 µg/mL), and phenylmethylsulfonyl fluoride (100 µg/mL), and centrifuged at 4℃, 12,000 rpm for 30 min and supernatant collected. Protein content in each sample was determined by a BioRad protein assay. Twenty µg of protein in each were removed and mixed with 6×sodium dodecyl sulfate (SDS) sample buffer (4×Tris-HCl/SDS pH 6.8, glycerol (20%), SDS (1%), bromophenol blue (1.2%), 3 M Urea). The sample mixture was boiled at 100℃ for 5 min and loaded onto a 6% polyacrylamide mini-gel for SDS polyacrylamide gel electrophoresis (SDS-PAGE). Proteins separated on the minigel were transferred onto a nitrocellulose membrane with a Bio-Rad transfer system at 80 V for 1 hr. The membrane was blocked with 5% nonfat dry milk in PBS at room temperature for 1 hr and incubated at 4℃ overnight with mouse monoclonal antibody against rabbit matrix metalloproteinase-3 (Oncogene, U.S.A.) diluted at 1:1,000 in blocking buffer. After washing with PBS containing 0.1% Tween-20, the membrane was incubated with a horseradish peroxidase-conjugated anti-rabbit IgG secondary antibody at room temperature for 1 hr. The membrane was washed with PBS, developed by an enhanced chemiluminescence (Amersham Pharmacia Biotech, Buckinhamshire, U.K.) western blot analysis system and exposed to Kodak XAR radiography film.

Statistical analysis was performed using Students' t-test to evaluate difference between the sample of interest and its respective control.

The lesions were developed mildly during the provocation period (average: 0.269 mm²), however, no aortas revealed any gross or microscopic atherosclerotic lesion in the control group. On light microscopic examination, the lesions consisted of intimal hyperplastic lesion, which mainly composed of lipidladen macrophages (foam cell) and smooth muscle cells. There were some variation of the lesions, but any fibrotic, calcific, or thromobotic plaque lesion was not demonstrated. No morphologic differences were demonstrated among experimental groups. All cholesterol-fed animals (ND, HD+G and HD group) had a lesion on the aortic endothelium (Fig. 1).

The lesion is progressed by continuous hyperlipidemic diet (average: 10.06 mm²), which is prevented partly by genistein (average: 0.997 mm²). Normal diet decreased the lesion size (average: 0.228 mm²).

The immunohistochemistry for RAM-11 revealed diffuse positive reaction in foamy cells and histiocytes in the lesion (Fig. 2). The characteristic of the reaction was not different among three experimental groups. The reactions with HHF-35 antibody showed in the smooth muscle cells within the lesion (Fig. 3). However, the staining intensities were slightly weaker than RAM-11, although it revealed similar reaction among the experimental groups. The average areas of positive reaction to each antibody were measured by the image analyzer, and the ratio of RAM-11 to HHF-35 area was measured (Fig. 4). There was no difference of statistical significance between the experimental groups (p>0.05).

ND and HD+G groups show similar bands of intensity on Western blotting with MMP-3. However, HD group revealed stronger expression of MMP-3 (Fig. 5).