| BackgroudIt has been recognized for years that leukocytes and platelets play an important role in the pathogenesis of atherosclerosis. Erythrocytes, however, have been traditionally deemed as an innocent bystander in the process of atherosclerosis until 1983 when Torkhovskaia et al. first reported an association between cholesterol contents of the erythrocyte membrane and the plasma cholesterol level in patients with coronary artery disease. Later it was found that phagocytosis of erythrocytes may contribute to the formation of foam cells and erythrocyte ghosts may accumulate membrane lipid hydroperoxides and sensitize low-density lipoprotein (LDL) to free radical-mediated oxidative modification. Recent studies further revealed that erythrocyte membrane originating from intraplaque hemorrhage may form a source of free cholesterol in the necrotic core and a driving force in the progression of atherosclerosis. Moreover, plaques with intraplaque hemorrhage were subject to a new plaque hemorrhage which may not only stimulate the progression of atherosclerosis but also promote the transition from a stable to an unstable lesion.Therefore, erythrocytes have emerged as a new culprit contributing to plaque growth and destabilization. However, the exact mechanisms of erythrocyte-induced atherosclerosis remain obscure. In the present study, we established a rabbit model of intraplaque hemorrhage by a high cholesterol diet feeding, balloon-induced endothelial injury and intraplaque erythrocyte injection and tested the hypothesis that erythrocytes may induce plaque vulnerability by accumulating lipids and augmenting inflammation in a dose-dependent waywithin the plaques.Objective(1) To develop an animal model of intraplaque hemorrhage that is mimic to human beings'.(2) To evaluate the qualitative and quantitative relations between intraplaque hemorrhage and plaque rupture.(3) To elucidate the mechanism of vulnerability of intraplaque hemorrhage. Methods and matetials1. Animal modelTwenty male New Zealand white rabbits aged 3-4 months were obtained from the Animal Center of Shandong Agriculture Science Academy. The procedures conform to the Guide for the Care and Use of Laboratory Animals published by the Chinese National Institutes of Health. The rabbit model of intraplaque hemorrhage was established by modification of previous reported methods. After a high cholesterol diet (containing 1% cholesterol and 2% fat) for one week, all rabbits underwent balloon-induced endothelial injury in the abdominal aorta and the high cholesterol diet feeding was continued until the end of week 10 when the diet changed to a purified rabbit chow with the purpose to normalize the serum lipid level. At the end of week 18, all rabbits were randomly divided into two groups: group A and group B with 10 rabbits in each group and intravascular ultrasound (FVUS) imaging and adventitial injection were performed in both groups. After exposure of the abdominal aorta under general anesthesia in each rabbit, three distinct plaques with similar thickness was identified with IVUS and the positions of these plaques were marked by iliopsoas stitch and their distance to the bifurcation of the iliac artery. Under the guidance of IVUS, 50μL washed autologous erythrocytes and 50μL normal saline (NS) were injected from the adventitia into one of the pre-selected 3 plaques, respectively, while the third plaque received no injection and was used as a control in group A. The same procedure was repeated in group B but the injected dose of erythrocytes and saline was doubled to 100μL. At the end of week 24, IVUS imaging was conducted again to measure the changes of plaque morphological parameters and the abdominal aorta was harvested. Plaques injected with erythrocytes (RBC plaques) or NS (NS plaques) and plaques without injection (blank plaques) were kept separately for histopathological, immunohistochemical and molecular biological studies.2.Biochemical studieBlood was drawn from rabbits fasting overnight to measure total serum cholesterol (TC) at the beginning of the study, the end of week 10 on termination of high cholesterol diet and the end of week 18 before adventitial injection. TC concentration was determined enzymatically as previously described.3. IVUS studiesIVUS imaging was performed following a standard procedure with a 3.2F catheter containing a single rotating element transducer of 40 MHz connected to an IVUS system (Galaxy, Boston Scientific Corporation, USA). IVUS was accomplished at the end of week 18 to identify 3 most prominent aortic plaques with similar thickness and in-between distance of at least 20mm, and guide adventitial injection into the plaques. IVUS was repeated at the end of week 24 to measure the external elastic membrane area (EEMA) and the lumen area (LA) from 10 equidistant cross sectional views along the abdominal aorta and the values measured were averaged. The percentage of plaque burden (PB%) was calculated as: PB% = (EEMA - LA) / EEMA.4. Histopathological and immunohistochemical analysisAbdominal aortic segments were fixed in 4% formaldehyde with some segments embedded in paraffin and cut into 5 μm thick sections for hematoxylin and eosin, Movat pentachrome, Perls (for iron display) or immunohistochemical staining. Cryosections cut into 6μm thickness were used for oil red O staining. Immunohistochemical staining was performed using standard techniques as described previously. In brief, endogenous peroxidase activity was inhibited by incubation with 3% hydrogen peroxide. Sections were blocked with 5% goat serum in PBS and incubated overnight at 4°C with primary antibodies. After PBS wash, the sections were incubated with secondary antibody at 37°C for 30 minutes. Avidin-biotin-peroxidase complex (Vectastain ABC KIT, VECTOR, USA) was also used in detecting erythrocyte membrane. Visualization of a positive reaction was developed with a peroxidase substrate solution containing 0.02%H2O2 and 0.1% 3, 3'-diaminobenzidine tetrahydrochloride (ZSBIO, China) in PBS to display the reaction product with a brown color, and the sections were then counterstained with hematoxylin. Incubation with PBS instead of the primary antibody was used as a negative control.The primary antibodies used in this study included monoclonal antibodies against rabbit macrophages (RAM11, Lab Vision Neomakers, USA) which were diluted to 1:400 to identify macrophages, and isolectin B4 from Bandeiraea simplicifolia which was conjugated with biotin and diluted to 10μg/mL (GSL-B4, Vector, USA) to identify erythrocyte membrane.Histopathological slides were analyzed using a computer-assisted morphometric analysis system (Image-Pro Plus 5.0). The fibrous cap thickness was measured at 10 equidistant points around the cap in each slice, three slices per section were measured, and the values were averaged. The area of positive staining of lipids, iron, erythrocyte membrane and macrophages were measured and expressed as a percentage of the staining area divided by the plaque area in at least 10 high power fields (×400). Plaque rupture was defined as thrombosis overlying fissured plaques or fibrous caps buried inside a plaque.5.Quantitative real-time reverse transcriptase polymerase chain reaction (RT-PCR)RNA from RBC, NS or blank plaques was isolated using Trizol reagent (Invitrogen, USA). Total RNA was quantified by spectrophotometry and reverse-transcribed with M-MLV Reverse Transcriptase System (Promega, Madison, USA) using the oligo (dT) (16) primers. The mRNA sequences of the investigated genes were obtained from GenBank and quantitative real-time RT-PCR was performed using LightCycler (Roche Applied Science, USA) following the manufacturer's instructions. TaqMan probes were used to detect mRNA expressions. The transcript amount of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was quantified as an internal RNA control. Quantitative values were obtained from the threshold cycle value (Ct), which is the point where a significant increase of fluorescence is first detected. Experiments were performed in triplicate for each data point. The primers and probes used were: GAPDH, Sense: GAA CGG GAA ACT CAC TGG CAT; Antisense: CCT TCT TGA TGT CGT CAT ACT TAGC; probe, CTC CAG GCG GCA GGT CAG GTC CAC, GenBank accession: No. AB231852, annealing temperature: 60°C; MCP-1, Sense: TCA TAG CAG TCG CCT TCA GC; Antisense: GTG TGT TCT TGG GTT GTG GAA TA; probe, CTT GCC CAG CCA GAT GCC GTG AAT, GenBank accession: No. M57440, annealing temperature: 58°C.6. Data AnalysisData were analyzed with SPSS 11.5 for Windows software. All values were expressed as mean±SD. Serum cholesterol changes over time were assessed by one way ANOVA analysis. The incidence of plaque rupture between 2 groups was compared by Chi-Square test. Measurements of different kinds of plaques in the same group were analyzed with ANOVA for randomized block design, while the different measurements between groups were analyzed using unpaired t-test. PB% at week 18 and 24 was compared by paired t-test. P value < 0.05 was considered statistically significant. Results1. Serum LipidsTC in all rabbits was 1.25 ± 0.21 mmol/L at the beginning of this experiment, and increased remarkably to 30.66 ± 12.82 mmol/L after a high cholesterol diet for 10 weeks (P=0.000). After a normal diet for 8 weeks, however, TC dramatically dropped to 1.24 ± 0.81 mmol/L which was equivalent to the baseline level (P>0.05). After randomization, the values of TC in group A and group B was similar (1.04 ± 0.76 vs. 1.32 ± 0.92 mmol/L, P>0.05).2. IVUS studiesAt the end of week 24, the value of PB% exhibited no significant difference among 3 kinds of plaques in either group A or group B. Moreover, the value of PB% was not significantly different between group A and group B in the same type of plaques (30.51±7.09% vs. 29.23±4.11% in RBC plaques, 34.14±9.49% vs.33.15±7.52% in NS plaques, and 33.25±6.06% vs. 31.25±4.12% in blank plaques, all P>0.05). However, PB% was reduced at the end of week 24 compared with that at the end of week 18 in both groups (30.23±7.12% vs. 39.55±9.38% in group A, and 33.11±8.10% vs. 41.65±9.13% in group B, all P<0.05, Fig.1A).3. Histopathological and Immunohistochemical analysis Hematoxylin and eosin staining showed that the fibrous cap was thinner inthe RBC plaques than in the NS or blank plaques in both groups (P<0.05), but the difference between the NS and the blank plaques was not significant. Fibrous caps were thinner in group B than in group A in the RBC plaques (10.93±2.12μm vs.l9.25±3.46μm, P<0.05) but the difference between the 2 groups in the NS or blank plaques was not significant (43.56±8.62μm vs.45.33±9.72μm in NS plaque, and 47.00μm±15.45μm vs. 51.00μm±10.74μm in blank plaques, P>0.05, Fig.1 B).The incidence of rupture plaque in group A tended to be lower than that in group B in 3 types of plaques (33.3% vs. 66.7% in RBC plaques, 10.0% vs. 22.2% in NS plaques, 10.0% vs. 11.1% in blank plaques, all P>0.05), but the difference between the 2 groups was not significant. Similarly, the incidence of rupture plaque appeared higher in the RBC plaques than that in the NS or blank groups, but again these differences were insignificant in each group(Fig1.C)(Fig.3 G,H).Movat pentachrome staining demonstrated that few cholesterol crystals and some foam cells were visible in the NS or blank plaques. In contrast, there were a large amount of cholesterol crystals or lipids and foam cells accumulated in the RBC plaques (Fig.2 E, F). Positive GSL-B4 and Perls staining was evident in the RBC plaques in both groups, but more extensive in group B than in group A (0.13±0.04 vs. 0.10±0.02 for GSL-B4 staining, and 0.10±0.02 vs. 0.08±0.01 for Perls staining, all P< 0.05), indicating that there were more erythrocytes accumulated in the RBC plaques in group B (Fig.2 A, B, C, D). On the other hand, there was no positive GSL-B4 and Perls staining detectable in the NS or blank plaques in either of the 2 groups.The lipid contents in the RBC plaques were significantly higher than those in the NS or blank plaques in both groups (P<0.05), but this difference between the NS and the blank plaques in both groups was not significant (P> 0.05) .The lipid contents in the RBC plaques were significantly lower in group A than those in group B (0.41±0.08 vs. 0.53±0.08, P< 0.05), but the differences between the two groups in terms of lipid contents in the NS (0.29±0.06 vs. 0.27±0.05, P>0.05) or blank (0.30±0.07 vs. 0.28±0.06, P> 0.05) plaques were insignificant (Fig.1 D) (Fig.3 A, B, C).Macrophage staining was more extensive in the RBC plaques than in the NS or blank plaques in both groups (P< 0.05), while the difference between the NS and blank plaques was not significant (P> 0.05). The macrophage staining in the RBC plaques was significantly higher in group B than that in group A (0.31 ±0.045 vs. 0.20±0.036, P< 0.05), but the differences between the two groups in terms of macrophage staining in the NS (0.06±0.02 vs. 0.05±0.01, P> 0.05) or blank (0.06±0.02 vs. 0.05±0.02, P> 0.05, Fig.1 E) plaques were insignificant (Fig.3 D, E, F).4. Quantitative Real-time RT-PCRThe relative mRNA expression of MCP-1 was higher in the RBC plaque than that in the NS and blank plaques in both groups (P< 0.001), while such a difference between the NS and blank plaques became insignificant (P> 0.05). The mRNA expression of MCP-1 was higher in the RBC plaques in group B than that in group A (26.42±5.90 vs.20.43±7.71, P<0.05), while the differences between group A and group B with regard to mRNA expression of MCP-1 in the NS (1.32±0.73 and 1.02±0.64, P> 0.05) or blank (1.41±0.93 and 1.73±0.79, P> 0.05) plaques were not significant (Fig1 F). Conclusions(1) With injection of erythrocytes into established atherosclerotical plaque, an animal model of intraplaque hemorrhage that is mimic to human beings' was built successfully.(2) Intraplaque hemorrhage induces vulnerability of atherosclerotic plaques in a dose-dependent way.(3) Intraplaque hemorrhage can lead to accumulation of lipid contents, augmentation of inflammation and thin fribrous cap of plaque, which may be the main mechanisms of erythrocyte-induced plaque vunerability. BackgroudAccumulating evidence demonstrated that erythrocyte may represent a potent atherogenic stimulus and may increase the risk of plaque destabilization. The cholesterol content of erythrocyte membrane had been found to contribute to the progress of atherosclerosis since 1983. Later Kolodgie and Takaya further proved that the lipid derived from erythrocyte was associated with large necrotic cores of atherosclerotic plaques with intraplaque hemorrhage. On the other hand, peroxidation plays an important role in erythrocyte-induced atherosclerotic progress. Vila reported erythrocyte ghosts may accumulate membrane lipid hydroperoxides and sensitize low-density lipoprotein (LDL) to free radical-mediated oxidative modification. Hemoglobin-derived heme has been demonstrated to act as a catalyst for the oxidation of LDL. The free heme can also lead to a dose-dependent peroxidation of unsaturated lipids with the presence of H2O2 and microsomal membrane fractions.And heme oxygenase-1 (HO-1), as the first rate-limiting enzyme of heme catabolism, has been noticed for its remarkable anti-oxidative and anti-inflammatory properties recently. Furthermore, its antiatherogenic properties have been confirmed by several experiments. These properties depend not only on elimination of heme and other oxidants but also on its catabolic production —biliverdin IXα, carbon monoxide and bilirubin IX. In situations such as intraplaque hemorrhage, heme and other oxidants derived from erythrocyte would accumulate in the plaque and trigger the peroxidization and inflammation, and HO-1 would be induced in the plaque at the same time. However, little was known about the effect of HO-1 in the situation, and whether the erythrocyte-induced atherosclerotic instability would be affected by HO-1 overexpression remain unknown as well. Objective(1) To further elucidate the mechanism of erythrocyte induced-atherosclerotic plaque progress.(2) To investigate the protective effect of HO-1 for intraplaque hemorrhage plaques and the relationship between HO-1 and plaque vulnerability.(3) To investigate the the mechanisms involved in the protective effect ofHO-1. Methods and matirials1. Animals model58 New Zealand white rabbits three to four months old were obtained from the animal center of Shandong Agriculture Science Academy. The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the Chinese National Institutes of Health.The rabbit model of intraplaque hemorrhage was based on the model previously reported. Fed a high-cholesterol diet (containing 1 % cholesterol and 2% pork lard) one week later, sixty New Zealand white rabbits received balloon-induced injury in the abdominal aorta. The rabbits were continually fed high cholesterol diet for another 9 weeks. After that, they were fed purified rabbit chow. At 18 weeks, after the exposure of rabbit abdominal aorta under general anesthesia (injection of 30 mg/kg sodium pentobarbital intravenously through the ear vein), three similar plaques with similar morphous and size of each rabbit were selected under the guidance of intravascular unltrasound (IVUS) (Galaxy, Boston Scientific Corporation, USA). Then 50μL washed autologous erythrocytes and equal volume of normal saline were injected into two of the plaques respectively, the remainding one served as blank controls. The positions of these plaques were marked by iliopsoas stitch and their distances to the bifurcate of iliac artery were recorded also. All rabbits were randomly divided into two groups (hemin group, n=29; control group, n=29). Those in henin group were intraperitoneally injected with hemin (25mg kg-1 body weighs, 4 times Per week) (Sigma, USA) and those in control group were treated with NS for 6 weeks, as previous reported. At 24 weeks, after being examined the morphous and size by IVUS again, the abdominal aortas were harvested, plaques treated with erythrocytes (RBC plaques), NS (NS plaques) or nothing (blank plaques) were kept separately for histological staining or molecular biology studies.2.Plasma Lipid and erythrocyte membrane cholesterolAt the beginning, 10 weeks (the end of high cholesterol diet) and 18 weeks (the time of injecting erythrocyte), blood was collected from rabbit fasted overnight to detect total plasma cholesterol (TC), triglyceride (TG), LDL and high-density lipoprotein (HDL). TC, TG, LDL and HDL concentrations were determined enzymatically by using cholesterol esterase and cholesterol oxidase. Erythrocyte membrane cholesterol was measured at the beginning, 10 weeks and 18 weeks of the experiment according to the modified procedure of Dodge and Macchia . Briefly, after removal from plasma and platelets by centrifugation, the erythrocytes were dispersed and washed three times with NS. Then double volume double distilled water was added and mixed with erythrocytes. Thereafter, the tubes were left overnight at 4°C to complete hemolysis. Lipids were extracted with the methods as Macchia described. Finally, membrane cholesterol content was determined enzymatically.3. Oxidant and antioxidant status of erythrocyteThe oxidant and antioxidant status of erythrocyte were analyzed for superoxide dismutase (SOD) activity and malondialdehyde (MDA) by the thiobarbituric acid reaction. At the beginning, 10 weeks and 18 weeks of experiment the MDA and SOD activity of erythrocyte were measured using MDA kit and SOD kit following the manufacturer's instructions (Nanjing Jiancheng Bioengineering Institute, Nanjing,China).4. IVUS examinationIVUS studies were performed before and after plaque injection using a 3.2F catheter containing a single rotating element transducer of 40 MHz connected to an IVUS system (Galaxy, Boston Scientific Corporation, USA). The image frame frequency is 32f/sec, the depth resolution and lateral resolution is 150μm and 360μm respectively. After reaching the aortic arch, the catheter was withdrawn using a motorized pullback device at a constant speed of 0.5mm/s. Each IVUS study was carried out according to a standard procedure.At the 18 weeks of experiment, three similar plaques of the same aorta were selected to inject with erythrocyte, NS or nothing under the guidance of IVUS. The process of plaque injection is shown in Fig.1. IVUS was performed at week 18 and week 24 to measure the external elastic membrane area (EEMA) and the lumen area (LA) from 10 equidistant cross sectional views along the abdominal aorta and the values measured were averaged. The percentage of plaque burden (PB%) was calculated as: PB% = (EEMA - LA) / EEMA5. Histological and Immunohistochemical stainingAbdominal aorta segments were fixed in 4% formaldehyde. A part of them were embedded in paraffin, and 5μm sections were stained with hematoxylin and eosin, Movat pentachrome, Perls (to determined iron) or for immunohistochemical analysis. Cryosections (6μm) were stained with oil red O to determine the lipid content.Immunohistochemical staining was performed by using the standard technique as described previously. Briefly, endogenous peroxidase activity was inhibited by incubation with 3% hydrogen peroxide. After blocking sections with 5% (v/v) goat serum in PBS, sections were incubated overnight at 4°C with the primary antibodies. After PBS wash, the sections were incubated with secondary antibody at 37°C for 30 minutes. Visualization of a positive reaction was developed with a peroxidase substrate solution containing 0.02% (wt/vol) H2O2 and 0.1% (wt/vol) 3, 3'-diaminobenzidine tetrahydrochloride (ZSBIO, Beijing, CHINA) in PBS to give the reaction product a brown color, and then the sections were counterstained with hematoxylin. A negative control, where the primary antibody was replaced with PBS was always included. The primary antibodies used are: monoclonal antibody against rabbit alveolar macrophages (RAM11) (Lab Vision Neomakers, USA)diluted 1:400 to identify macrophages, isolectin B4 from Bandeiraea simplicifolia conjugated with biotin (GSL-B4)(Vector, USA) diluted at 10μg/mL to identify erythrocyte membranes; mouse anti- rabbit HO-1 polyclonal antibody (OSA-111) (StressGen, Canada)diluted 1:100 to detect HO-1; mouse anti-nuclear transcription factor κB (NF-κB) P65 subunit monoclonal antibody(MAB3026)(Chemicon,USA) diluted 1:100 to detect NF-κB, which increases the transcription of cytokines and acute phase proteins; mouse anti-matrix metal proteinase -9(MMP-9) monoclonal antibody(sc-21733)(Santa Cruz) diluted 1:100 to detect MMP-9. Biotinylated secondary antibodies (zb-2020, zymed, USA), avidin-biotin-peroxidase complex (Vectastain ABC KIT, VECTOR, USA) were also used in detecting HO-1 following the standard procedure as described previously.Positive staining was counted and expressed as a mean percentage of the plaque area in at least ten high power fields (×400 magnification) by using Image pro plus 5.1 software. The Mean fibrous cap thickness, measured in 10 equidistant different sites per section, and the percentage lipid content of plaques were also counted. The plaque rupture was defined as the buried fibrous cap or thrombi overlying fissured plaques being found.6.Quantitative Real-time reverse transcriptase- polymerase chain reaction (RT-PCR)RNA from RBC plaques; NS plaques or control plaques was isolated using Trizol reagent (Invitrogen). Total RNA was quantified by spectrophotometry and was reverse-transcribed with M-MLV Reverse Transcriptase System (Promega, Madison, USA) using the oligo(dT) (16) primers. The mRNA sequences of the investigated genes were obtained from GenBank. Quantitative real-time RT-PCR was performed using LightCycler (Roche APPlied Science, USA) following the manufacturer's instructions. The SYBR Green I kit (TaKaRa Biotechnology, Dalian, China) was used for amplification and detection of the expression of HO-1 mRNA and TaqMan probes were used for detecting the expression of monocyte chemoattractant protein-1(MCP-1), vascular cell adhesion molecule-1(VCAM-1), matrix metalloproteinase 9 (MMP-9) and tissue inhibitor of metalloproteinase 1 (TIMP-1 )mRNA. Quantitative values were obtained from the threshold cycle value (Ct), which is the point where a significant increase of fluorescence is first detected. The transcript number of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was quantified as an internal control. Experiments were performed in triplicate for each data point. The data were analyzed with 2-△△CT method.The results of RT-PCR were confirmed by gel electrophoresis. The primers and probes used were listed in Table 1.7. Data AnalysisData were analyzed with SPSS 11.5 for Windows software. All values were expressed as mean±SD. Seruml, erythrocyte cholesterol, erythrocyte MDA and SOD were evaluated by one way ANOVA. The incidence of plaque rupture between 2 groups was compared by Chi-Square test. Measurements of different kinds of plaques in the same group were analyzed with ANOVA for randomized block design, while the different measurements between groups were analyzed using unpaired t-test. PB% at week 18 and week 24 was compared by paired Mest. P value < 0.05 was considered statistically significant. Results1. Plasma Lipids, erythrocyte membrane cholesterol and erythrocyte MDA, SODThe mean plasma TC, TG, LDL and HDL levels increased significantly after 10 weeks of the atherogenic diet (P>0.05), and then dramatically decreased to baseline level after 8 weeks of normal diet. However, after 10 weeks high cholesterol diet the erythrocyte membrane cholesterol increased and maintained at the high level though the plasma lipid decreased with the change of diet. The changes of erythrocyte MDA or SOD were in line with that of erythrocyte membrane cholesterol. (Table .1)2. IVUS examinationAt week 24, there was no difference of PB% among RBC plaques, NS plaques or blank plaques (40.21±9.15, 39.11±8.07 and 41.21±12.32 respectively, all P>0.05). (P>0.05). However, the total PB% was lower at 24 weeks compared with 18 weeks (39.11±8.09 VS 42.5225±9.38, P<0.05). At 24 weeks, the total PB% in control group was higher than that in hemin group (45.75±11.23 VS 36.21±10.12, P<0.05).3 .Histology and Immunohistochemistry stainingHematoxylin and eosin staining showed that the fibrous cap was thinner in RBC plaques (12.93±6.12μm and 19.55±5.46μm) than in NS plaques (41.40±13.01μm and 42.77±11.34μm) or blank plaques (41.26±11.71μm and 44.12±6.07μm) in control and hemin groups (all P<0.05), while there was no difference between NS and blank plaques (all P>0.05). However, the difference of fibrous cap between hemin and control group was insignificant (all P>0.05).In control group, the incidences of RBC plaques were higher than that of NS or blank plaques (53.6% vs. 21.4% or 25.0%, all P<0.05). But in hemin group, there was no difference among three types of plaque (RBC plaque 26.9%, NS plaque 19.2% and blank plaque 19.2%, all P>0.05). The incidence of RBC plaque rupture was lower in hemin group than in control group (26.9% vs. 53.6%, PO.05). In NS plaques or blank plaques, the differences of incidence for plaque rupture between control group and hemin group were insignificant (21.4% and 19.2% in NS plaque; 25.0% and 19.2% in blank plaque, in control and hemin group respectively, all P>0.05)There was no positive GSL-B4 and Perls staining detectable in the NS or blank plaques in either of the 2 groups. The counts of GSL-B4 or Perls positive staining in RBC plaques were not different between hemin and control group (0.094±0.020 vs.0.099±0.016 for GSL-B4 staining; 0.075±0.011 vs. 0.066±0.019 for Perls staining, all P > 0.05).Movat'pentachrome staining demonstrated that few cholesterol crystals were visible in the NS or blank plaques. In contrast, there were a large amount of cholesterol crystals or lipids and foam cells accumulated in the RBC plaques.The lipid content of RBC plaques (0.54±0.06 and 0.58±0.07, in hemin and control group respectively) was remarkably higher than that of NS plaques (0.23±0.07 and 0.27±0.05, in hemin and control group respectively) or blank plaques (0.29±0.04 and 0.22±0.05, in hemin and control group respectively) in both groups (all PO.05), while the difference between the NS and blank plaque in both groups was insignificant (all P>0.05). The lipid content of three types of plaque was not different between hemin and control groups (P>0.05).In immunohistochemistry, the positive macrophage staining was higher in RBC plaques than in NS or blank plaques in both groups (PO.05). It was also higher in control group than in hemin group (P<0.05). The results of anti-NF-κB, anti-MMP-9 immunostaining were similar to that of macrophage staining except that anti-MMP-9 was not different between hemin and control group. The positive staining of HO-1 was higher in RBC plaques than in NS or blank plaques (P<0.05), and it was also higher in hemin group than in control group.4. Quantitative Real-time RT-PCRThe HO-1 mRNA expression was significant higher in RBC plaques than in NS or blank plaques, and was also higher in hemin group than in control group (all P<0.05).The relative mRNA expression of MCP-1, MMP-9 and VCAM-1 was higher in RBC plaque than in NS or blank plaque in both groups (all P<0.05). The relative mRNA expression of MCP-1 and VCAM-1 was lower in hemin group than in control group (all P<0.05). On the contrary, the TIMP-1 mRNA expression was lower in RBC plaque than in NS and blank plaque in both groups , and was higher in hemin group than in control group (all P<0.05). Conclusions(1) Erythrocytes enhance the instability of plaque by increasing cholesterol and oxidant status in the plaque, which result in lipid content enlargement, inflammatory cells recruitment and thinner fibrous cap.(2) HO-1 inhibites erythrocyte-induced atherosclerotic progress and enhances the stability of plaque.(3) HO-1 has anti-oxidative and anti-inflammatory properties in plaque, which may be the mechanisms involved in its protective effect on plaques with hemorrhage. |
