A total of 48 ApoE^−/− mice [strain C57BL/6J] in the present study were introduced from the Jackson laboratory in the United States by the Peking University Medical Laboratory Animal Center. The mice were purchased as ApoE−/−. All mice [male; age, 8 weeks; weight, 18±2 g; license no. SCXK (Beijing) 2011–0012] were maintained in the Medical Animal Experimental Center of The First Affiliated Hospital of Xinjiang Medical University (Urumqi, China). They were fed normal diet at room temperature (relative humidity: 40–70%, 12 h/d light dark cycle). The present study was approved by the Animal Ethics Committee of The First Affiliated Hospital of Xinjiang Medical University.

Tianxiangdan Granule is composed of four types of Chinese medicine: Rhodiola root, Dalbergia wood, Ziziphora clinopodioides and Salvia miltiorrhiza. The drug was prepared by the Traditional Chinese Medicine Hospital Affiliated to Xinjiang Medical University (batch no. ZO4000820). Tianxiangdan Granule obtained a national patent in 2009 (patent no. 200910210063.9) (12). Atorvastatin calcium tablets (batch no. H20051408) were purchased from Pfizer Inc. (New York, NY, USA). Cholesterol was obtained from Amresco, LLC (Solon, OH, USA). Propylthiouracil tablets were purchased from Riemser Pharma GmbH (Greifswald, Germany). TNF-α (cat. no. 106958008) and IL-1β (cat. no. 92572000) ELISA kits were purchased from eBioscience; Thermo Fisher Scientific, Inc. (Waltham, MA, USA). Radioimmunoprecipitation assay (RIPA) buffer, Pierce Phosphatase Inhibitor Mini Tablets, Halt Protease Inhibitor Cocktail and bicinchoninic acid (BCA) Protein Assay kit were obtained from Thermo Fisher Scientific, Inc. Triglyceride (TG) assay kit (cat. no. A110-2), total cholesterol (TC) assay kit (cat. no. A111-2), low-density lipoprotein cholesterol (LDL-C) assay kit (cat. no. A113-2) and high-density lipoprotein cholesterol (HDL-C) assay kit (cat. no. A112-2) were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). The following primary antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA): Phosphorylated (p)-p38 MAPK (cat. no. 4631S), p38 MAPK (cat. no. 9212S), p-NF-κB p65 (cat. no. 3031S), NF-κB p65 (cat. no. 8242S) and GAPDH (cat. no. 5174S). Alkaline phosphatase-conjugated secondary antibody (Anti-Rabbit) (cat. no. WP0007) and BCIP/NBT Chromogenic Substrate were purchased from Invitrogen (Thermo Fisher Scientific, Inc.). Polyvinylidene fluoride (PVDF) membrane was obtained from Roche Diagnostics (Basel, Switzerland).

Of the 48 8-week-old ApoE^−/− mice, 12 were assigned to the normal control group, in which mice were provided ordinary diets, and were maintained at room temperature (20–24°C) and 40–70% relative humidity. The remaining 36 ApoE^−/− mice were assigned to the modeling group, in which mice were fed a high-fat diet and were maintained in an artificial climate box, in order to stimulate the climate and eating habit characteristics of Xinjiang. Every morning, ApoE^−/− mice in the modeling group were placed in the artificial climate box at 10:00 a.m. and were taken out at 09:00 p.m. when they were placed back in the room temperature environment. The temperature of the artificial climate box was set at 6±2°C, relative humidity was controlled between 25 and 32.8%, and the light-dark cycle was 12 h/day. The purpose of this method was to establish the Huizhuo Tanzu type AS model (13,14). After 12 weeks of feeding, 2 ApoE^−/− mice from the normal control group and 6 ApoE^−/− mice from the modeling group were randomly selected and sacrificed, after which, aortic roots were obtained and observed under a light microscope following hematoxylin and eosin (H&E) staining. In samples from the modeling group, an increased number of foam cells, cholesterol crystals and atheromatous plaques were observed, and lumen stenosis was clearly demonstrated. In addition, the remaining ApoE^−/− mice in the modeling group appeared lethargic, their participation in activities was decreased, food and water intake was reduced, their fur appeared lackluster, weight growth was not obvious, and they had sticky stools. These signs confirmed that the Huizhuo Tanzu type AS model had been successfully established. Furthermore, the remaining 30 ApoE^−/− mice were randomly divided into three groups: Model group, Tianxiangdan group [2.7 g·(kg·d)⁻¹], and atorvastatin group [6.1 mg (kg·d)⁻¹]; n=10 mice/group. The aforementioned doses were selected by equivalent conversion, based on the clinical routine dose used to treat human adults: Mouse experimental dose [g (kg·d)⁻¹]=9.1 × human adult clinical dose [g·(kg·d)⁻¹]. To obtain the experimental doses, drugs were dissolved in distilled water and administered by gavage once a day (15). Doses were adjusted according to body weight each week. Mice in the normal control and model groups were administered an oral gavage of distilled water (0.2 ml·d⁻¹). All mice were given free access to food and water, and their activities were not restricted. Mice in the normal control group were placed at room temperature (20–24°C) and 40–70% relative humidity, and were given a normal diet. Mice in the model, Tianxiangdan and atorvastatin groups were administered a high-fat diet and continued to undergo intermittent maintenance in the artificial climate box. The high-fat diet comprised: Basal diet, 72.5%; lard, 10%; sucrose, 10%; pig bile salt, 0.2%; cholesterol, 2%; propylthiouracil, 0.12%; egg yolk, 5%; and 21 Super-Vita, 0.1%. The temperature of the artificial climate box was set at 6±2°C, and relative humidity was controlled between 25 and 32.8%.

A total of 12 weeks following administration of the first gavage, ApoE^−/− mice in all groups were fasted. After 12 h the following operations were conducted: i) Mice were anesthetized by 0.1 ml/kg intraperitoneal injection of anaesthetic composed of 2 ml diazepam, 2 ml ketamin, 1 ml atropine and 5 ml 0.9% sodium chloride, and blood samples were collected from the eyeballs. The samples were placed in microcentrifuge tubes and were centrifuged at 3,000 × g for 10 min at 4°C; the serum obtained by separation was maintained at −80°C for the detection of serological markers. Repeated freezing and thawing was avoided. ii) Mice were anesthetized by 0.1 ml/kg intraperitoneal injection of anesthetic (injection of diazepam, ketamine and atropine), then were sacrificed by cervical dislocation. Thoraces and abdomens were opened; arterial tissues from the aortic root to the branches between the iliac and renal artery were obtained and rinsed in normal saline. A length of ~0.5 cm from the aortic root was obtained, and fixed in 10% formalin for H&E staining and plaque area (PA) analysis. Residual aortic tissues were quickly placed in microcentrifuge tubes, labeled, and placed into liquid nitrogen and quickly frozen. The samples were then transferred into an ultra-low temperature freezer at −80°C, and preserved for detection by western blotting.

Aorta tissues were fixed at 56°C for 12 h with 10% formaldehyde and dehydrated, prior to conventional paraffin embedding and dyeing with hematoxylin and eosin. From the aortic root, uniform slices were obtained (~5 µm); 40 consecutive slices were obtained from each sample. One slice was obtained from every five slices; therefore, a total of eight slices were obtained from each sample. After H&E staining, aortic morphological alterations were observed under a light microscope; measurement and analysis of relevant pathological morphological indices were conducted using the multifunctional image analysis software of Image-Pro Plus (version 6.0; Media Cybernetics, Inc., Rockville, MD, USA). Blood vessel lumen area (LA) and PA measurements were obtained, and the PA/LA ratio was calculated. The mean values of PA/LA were calculated for statistical analysis.

Prior to detection, serum samples were removed from the ultra low temperature freezer and were placed in a refrigerator at 4°C. Subsequently, TG, TC, HDL-C and LDL-C serum levels were detected according to the manufacturer's protocols an automatic biochemical analyzer (cat. no. BS200; Mindray Medical International Co., Ltd., Shenzhen, China).

Prior to detection, serum samples were removed from the ultra low temperature freezer and were placed in a refrigerator at 4°C. Subsequently, the serum levels of proinflammatory cytokines IL-1β and TNF-α were detected using an ELISA kit, according to the manufacturer's protocol. Absorbance values were determined using an enzymatic microplate reader at a wavelength of 450 nm. A standard curve was generated, with concentration as the horizontal coordinate and standard absorbance value as the vertical coordinate. The concentrations of IL-1β and TNF-α in the samples were calculated using the standard curve.

Aortic tissue protein was extracted using RIPA with a protease inhibitor and phosphatase inhibitor, and protein concentration was determined using a BCA protein assay kit. Protein samples (20 µg) were separated by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis at 80 V and were transferred onto a PVDF membrane by electroblotting. The membrane was blocked in TBS-Tween 20 (TBST) supplemented with 5% w/v bovine serum albumin (Beijing Raghan Biological Technology Co. Ltd., Beijing, China) for 1 h at room temperature. Subsequently, the membrane was washed three times with TBST, and was incubated with the following primary antibodies: GAPDH (cat. no. 5174S), NF-κB p65 (cat. no. 8242S), p-NF-κB p65 (cat. no. 3031S), p38 MAPK (cat. no. 9212S) and p-p38 MAPK (cat. no. 4631S) (all antibodies: 1:1,000 of dilution, Cell Signaling Technology, Inc.), at 4°C overnight with agitation. After washing the membrane with TBST three times, the alkaline phosphatase-conjugated secondary antibody (cat. no. WP0007) was added (1:1 of dilution; Thermo Fisher Scientific, Inc.), and the membrane was further incubated at 37°C for 1 h. The membrane was then washed with TBST and underwent visualization in the dark with the BCIP/NBT Chromogenic Substrate. The blots were scanned and analyzed using Quantity One v4.62 software (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

Statistical analysis was conducted using statistical software SPSS 17.0 (SPSS Inc., Chicago, IL, USA). Data were expressed as the mean ± standard deviation of three independent experiments. Intergroup comparison was performed using one-way analysis of variance, and the comparison between two groups was performed using the Least Significant Difference method. P<0.05 was considered to indicate a statistically significant difference.

Mice in the normal control group were lively, irritable and restless, had a quick response, bit each other frequently, ate a lot of food and drank a lot of water, gained significant amounts of weight, had clean and shiny fur, and had dry granular stools. Mice in the model group became lethargic, their participation in activities significantly decreased, their food and water intake decreased, their fur was lackluster, weight growth was not obvious, and they had sticky stools. Characteristics of the mice in the Tianxiangdan and atorvastatin groups were better compared with those in the model group, their fur was relatively clean and shiny, and their water and food consumption, and weight gain, was larger than in the model group, and their stools were granular but slightly sticky (Table I).

In the normal control group, no obvious smooth muscle cell hyperplasia was detected, slight hyperplasia of the intima was observed, and no cholesterol crystal formation or obvious stenosis was detected in the aortic lumen (Fig. 1A). In the model group, numerous atherosclerotic plaque lesions appeared and the thickness of the aortic tube wall was obviously uneven. In addition, endothelial lesions had a larger PA and a thin fibrous cap. Large amounts of foam cells and cholesterol crystals were also detected, lipid cores were larger, and the majority of plaques did not closely adhere to the tube wall and some were detaching. Furthermore, calcification occurred locally, and considerable inflammatory cell infiltration was detected in the adventitia (Fig. 1B). In the Tianxiangdan group, atherosclerotic lesions were mainly fibrous plaque lesions, aortic tube wall thickness was uneven, the intima was markedly thicker, and fibrous plaques covered with a fibrous cap were clearly observed. The PA was markedly smaller than that in the model group. All plaques closely adhered to the aortic wall, and no detachment was observed. In addition, fibrous caps were thicker, lipid cores were small, and a small amount of inflammatory cell infiltration was detected in the adventitia (Fig. 1C). Pathological alterations were less severe than in the model group and plaque stability was improved, thus suggesting that Tianxiangdan increased plaque stability. In the atorvastatin group, the majority of alterations were detected in the atherosclerotic fibrous plaque lesions, the thickness of the aortic wall was uneven, the intima appeared markedly thicker, and fibrous plaques covered by a fibrous cap were observed. The PA was markedly smaller compared with in the model group, and the plaques did not closely adhere to the aortic wall; however, detachment was not observed. Lipid cores were not obvious, and adventitial inflammatory cell infiltration was rare (Fig. 1D). Pathological alterations were less severe than in the model group, thus indicating that atorvastatin could increase plaque stability.

The PA/LA ratio was compared among the groups; PA and LA were measured using Image-Pro Plus software (version 6.0). Compared with the normal control group, the aortic PA/LA ratio was significantly increased in the model group (P<0.01). Compared with in the model group, aortic PA/LA ratio was significantly decreased in the Tianxiangdan and atorvastatin groups (P<0.01). There was no statistically significant difference in aortic PA/LA ratio between mice in the Tianxiangdan and atorvastatin groups (P>0.05; Fig. 2).

The results indicated that compared with in the normal control group, TC, TG, LDL-C and HDL-C content was significantly higher in the model group (P<0.05). Compared with in the model group, TC and LDL-C serum content was significantly lower in the atorvastatin and Tianxiangdan groups (P<0.01). There was no statistically significant difference in serum lipid levels between mice in the Tianxiangdan and atorvastatin groups (P>0.05; Table II).

Compared with in the normal control group, serum IL-1β and TNF-α levels were significantly higher in the model group (P<0.01). Compared with in the model group, the serum levels of IL-1β and TNF-α were significantly lower in the Tianxiangdan and atorvastatin groups (P<0.01). There was no statistically significant difference in serum IL-1β and TNF-α levels between mice in the Tianxiangdan and atorvastatin groups (P>0.05; Table III).

Compared with the in normal control group, the protein expression levels of p-NF-κB p65 and p-p38 MAPK were significantly increased in the aortic tissue of the model group (P<0.01). Compared with in the model group, the protein expression levels of p-NF-κB p65 and p38 MAPK were significantly decreased in the Tianxiangdan and atorvastatin groups (P<0.01). There was no statistically significant difference in the protein expression levels of p-NF-κB p65 and p-p38 MAPK between mice in the Tianxiangdan and atorvastatin groups (P>0.05; Figs. 3 and 4).