慢性心房颤动电重构和结构重构机制的研究

【作者】 杨贵荣；

【导师】 张薇；

【作者基本信息】 山东大学 ， 心血管内科， 2004， 博士

【摘要】 研究背景 心房颤动（简称房颤）是最常见的具有临床意义的心律失常,发病率随年龄增长而增加。随着我国人口老龄化,房颤的防治作为一个难题更加突出。房颤的治疗效果有赖于对其发生和维持机制的深入了解,而合适的房颤动物模型是研究其发生机制的基础。因此,建立病理生理基础与临床情况相似的房颤动物模型对于其发生机制和防治的研究具有重要意义。房颤具有自身延续性,在这一发展过程中心房肌细胞结构、心房肌间质和心房血流动力学发生了改变,称为房颤心房结构重构。人们已经认识到心房结构重构在房颤的发生和维持中起重要作用,但有关心房颤时心房结构重构的分子机制和影响因素仍不明确。 正常情况下,维持心脏结构和功能完整的心肌细胞外基质（extracellular matrix,ECM）处于合成与降解的动态平衡之中。基质金属蛋白酶（matrix metalloproteinase,MMPs）和组织型基质金属蛋白酶抑制因子（tissue inhibitor of metalloproteinase,TIMPs）是影响心脏细胞外基质蛋白降解的主要因素。对衰竭心脏的研究发现,MMPs活性升高导致进行性心室扩大。基础研究表明心房组织中亦存在MMPs/TIMPs系统,但慢性房颤时MMPs/TIMPs系统激活是否与心房扩大有关未见相关报道。 临床研究表明,房颤时心房肌局部肾素—血管紧张素系统（renin angiotensin system,RAS）激活是房颤患者心房肌纤维化的重要机制。但在慢性房颤模型中RAS系统激活与结构重构的关系仍不清楚。心衰心肌纤维化时,胶原链断裂并被缺少连接的胶原所取代,这种ECM的重构是由MMPs活性调节,由此推测RAS系统激活导致的细胞外基质纤维堆积和MMPs/TIMPs系统介导的细胞外基质降解之间可能存在联系。 整合素（integrin）是粘附分子的一种,其中的整合素β₁亚组主要影响细胞与胶原、纤维连接蛋白和层粘连蛋白的粘附,为心肌细胞的相互联系和排列提供更多还原

【Abstract】 Background Atrial fibrillation （AF） is one of the most common sustained tachyarrhythmia in clinical practice. The incidence increased with aging. Atrial fibrillation is not only accompanied by stroke, but also an independent risk factor to predict the morbidity. However, current treatment modalities for AF are far from satisfactory. The underling mechanism of AF remains unclear, which make the pharmacologic therepy is still challenges. It is important to establish an atrial fibrillation model to investigate the mechanism underling AF. It is well established that AF cause atrial structural remodeling leading to maintenance of AF. The molecular basis for the development of atrial structural remodeling of atrial fibrillation is still a matter of debate. Its understanding, however, could have an important therapeutic impact.Extracellular matrix （ECM） contributes to the maintenance of cardiac geometry and the structural alignment of adjoining myocytes. Matrix metalloproteinase （MMPs） and tissue inhibitor of metalloproteinase （TIMPs） are responsible for extracellular collagen degradation and synthesis. MMPs activation can cause left ventricular dilation in end stage CHF. Basic reseach find MMPs/TIMPs are also presented in atrial tissue. Whether there is possible relationship between MMPs/TIMPs activation and atrial enlargement of chronic AF remains unclear.It is well known that local renin-angiotensin system （RAS） is involved in atrial myocardial fibrosis in AF patients. But there are still in debating on MMPs activation influencing atrial structural remodeling in chronic AF model. MMPs regulates ECM remodeling, so there are maybe relationship between ECM accumulation attributing to RAS activation and degradation attributing to MMPs/TIMPs.Integrin β₁ is one of the adhesion molecular, which regulates cell-cell and cell-matrix interactions. Its expression can maintain myocardial structural integrity. The role of integrin β₁ in atrial structural remodeling is deserved to investigate. It will help us to understand the interaction of myocytes and ECM in atrial structural remodeling of AF.Objective（1） To establish a rapid atrial pacing induced atrial fibrillation model;（2） To elucidate the characteristic of unltrastructure, cardiac extracellular matrix, atrial geometry and function of atrial tissue when atrial fibrillation is persistent;（3） To investigate the role of MMPs/TIMPs , integrin P 1 and renin angiotensin system in atrial structural remodeling and their relationships;（4） To find the activators and influnce factors of MMPs/TIMPs and RAS;（5） To compare the difference between left and right atrial remodeling. Methods（1） Establish a canine chronic atrial fibrillation model by rapid atrial pacing.（2） Echocardiography was performed to determined the changes of biatrial dimension before and after 8 weeks pacing. Mitral and tricuspid regurgitation were also viewed by colour Doppler flow image during pacing.（3） Left ventricular end diastolic pressure, right ventricular end diastolic pressure, right atrial pressure were assessed by cardie catheterization before and after 8 weeks rapid atrial pacing.（4） Angiotensin II concentration in atrial tissue was determined by radio-immunity.（5） The content of calcium was assessed by spectrocomparator in atrial myocardium.（6） HE stain and trichrome stain were preformed to observe myocyte structure and determine the content of collagen in atrial myocardium.（7） Ultrastructural changes of left and right atrial tissue both atrial fibrillation and control were observed by transmission electron microscopy.（8） The messenger ribonucleic acid（mRNA） level of angiotensin converting enzyme（ACE）, integrin P ? matrix metalloproteinases-9（MMP-9） and tissue inhibitor of matrix metalloproteinase-l（TIMP-l） were measured by reverse transcription polymerase chain reaction（RT-PCR） and normalized to the mRNA level of P actin.（9） Immunohistochemistry was performed to determine the protein expression and distribution of MMP-1, MMP-9 and TBVIP-1 in myocytes.（10） Statistical analysis. All continuous data were expressed as means± 1 standard deviation, SPSS 10.0 for windows was used to analyze all data. Comparisons betweengroups were performed with unpaired Student’s t-te&t Comparisons of continuous variables among multiple groups were performed by single- factor ANOVA. Pearson’s correlation coefficient （r） was used to determind the relation between groups. Mann-Whitney test was used to non-parameter. A value of P<0.05 was considered to be statistically significant. Results（1） AF model. Eleven anesthetized mongrel dogs underwent insertion of transvenous lead at the right atrial appendage that was continuously paced at 370-430 beat per minute for 8 weeks. Among those dogs, one had a sudden death after 12 hours pacing, one showed heart failure sympom after 4 weeks pacing and echocardiography showed thrombosis in right atrium, and pacemaker stop generating pulse in another dog at the end of study. Before rapid atrial pacing, AF was not induced in any dogs by programmed stimuli and burst pacing. After 8 weeks of continuous rapid atrial pacing, AF was occurred spontenously in 2 dogs （25%）, sustained AF was induced in 4 dogs by programed stimuli （50%）, and in 2 dogs by burst pacing （25%）. It was shown that chronic atrial fibrillation model was successfully established.（2） None of dogs showed valvular regurgitation before rapid atrial pacing. Different extent of mitral and tricuspid regurgitation was viewed in all rapid atrial pacing （RAP） dogs at the end of study. In contrast, mild mitral regurgitation was occurred in one control dog. Compared with control and before pacing, left atrial dimension and right atrial area enlargement was documented in RAP dogs. There were not significant changes of atrial dimension in control dogs.（3） There were no significant difference in left ventricular end diastolic pressure （LVEDP） and right atrial pressure between RAP and control dogs before pacing （LVEDP: 12.00±6.07 VS 9.00 + 5.47; right atrial pressure^.17 + 2.52 VS 4.95 + 3.24, P>0.05 respectively）. Compared with control dogs, LVEDP and right atrial pressure of RAP dogs elevated markedly after 8 weeks continuous rapid atrial pacing（ LVEDP: 14.87 + 2.79 VS 10.00+3.95; right atrial pressure: 7.09+ 1.13 VS 4.50+1.51, P <0.01, respectively）.（4） Compared with control, the content of calcium inleft and right atrial myocardium was elevated by 46.39% and 36.17% （ LA: 35.09+4.93 VS 23.97+4.50; RA: 37.68+10.45 VS 27.67 + 4.46, P < 0.001, 0.05, respectively）. Content ofmagnesium in left atrial tissue was decreased by 9.11% （61.69 + 6.73 VS 67.87 + 5.85 P <0.05）, whereas it had no statistical significance in right atrial tissue（56.45 +1.64 VS 58.42 + 5.15 P>0.05）.The content of calcium had no significant difference between left and right atrial myocardium in both control and RAP dogs.（5） Compared with control, the level of angiotensin II of both atrial myocardium was significantly increased in RAP dogs. The content of angiotensin II of left atrial myocardium is markedly increased than that of right atrium in control dogs. Whereas there was no significant difference between left and right atrial tissue in RAP dogs.（6） Histologic examination. Atrial myocardium sections from controls were normally structured cardiomyocytes, which were surrounded by a small amount of connective tissue. Sarcomeres were present throughout cytoplasm of cardiomycytes. Sections from RAP dogs showed severe myolysis. Early myocardial hypertrophy was observed in both atrial tissue from rapid atrial pacing dogs. Cardiomyocytes were disarrangement and connective tissue was accumulated. Some cardiomyocytes were surrounded by collagen tissue and myofilament was disrupted. Masson stain showed that collagen tissue was appropriate arranged among cardiomyocytes in control samples. Samples from RAP dogs showed, however, collagen tissue increased markedly, and disrupted in some area.（7） Electron microscopy. At the ultrastructural level, atrial myocytes from controls showed a highly organized sarcomeric structure with rows of uniformly sized mitochondria in between. Atrial granules were mainly confined to the perinuclear area. A typical distribution of heterochromatin in the form of clusters at the nuclear memrane was present in all cardiomyocyte nuclei. Intercalated disk were normally structured. Atrial myocytes from RAP dogs showed characteristic changes as below: ?Contractile material was depleted（myolysis）. The disapperance of sarcomeres was often limited to the vicinity of the nucleus but also frequently involved the entire cytosol, in which only fragments of sarcomeres were present. ?Huge amount of glycogen filled the myolytic space in almost all cells that underwent myolysis. in some area, even accumulated like" glycogen lake"? Typical changes in size and shape of mitochondria were seen in area delepted of sarcomeres. Many mitochodria had become elongated. （5） The heterochromatin was dispersed uniformly throughout the nucleoplasm. ? Intercalated disk were disrupted in some areas and indistinct. （7） the sarcoplasmic reticulum waspartially destroyed and showed edema. Although these abnomalities were seen in both the left and right atria, tissue from the left atrium displayed significantly more abnomalities than did the tissue from right atrium. ?Atrial granules were increased and scattered widely.（8） Results of RT-PCR. ?Compared with control, the mRNA level of ACE of left and right atrial myocardium from RAP dogs was significantly increased by 54.54% and 45%（ LA: 0.68+0.11 VS 0.44 + 0.19; RA: 0.29+0.08 VS 0.20 + 0.01, P <0.05）.The mRNA level of ACE of left atrial tissue was markedly increased compared with that of right atrium in controls and RAP dogs.（2）Compared with control, the mRNA level of integrin P , from RAP dogs was increased by 95.35% （0.84+0.33 VS 0.43±0.02, P<0.01） in left atrial tissue but had no significant changes in right atrial tissue （0.34+ 0.09 VS 0.35 + 0.02, P>0.05）. The mRNA level of integrin P , of left atrial tissue was markedly increased compared with that of right atrium in controls and RAP dogs. （3） Compared with control, the mRNA levels of MMP-9 in left and right atrial myocardium from RAP dogs were elevated significantly by 45.00% and 109.09% while TIMP-1 were elevated 46.67% and 71.43%, respectively. （MMP-9: LA 0.29+0.06 VS 0.20+ 0.03, RA 0.23 + 0.07 VS 0.11+0.009; TIMP-1: LA 0.22 + 0.02 VS 0.15 + 0.01, RA 0.12+0.02 VS 0.07+0.01, all P<0.01）.However, there were no significant changes of the ratio of MMP-9/TIMP-l.The mRNA level of MMP-9 and TIMP-l of left atrial tissue were markedly increased compared thoses of right atrium in control and RAP dogs.（9） Pearson correlation analysis. Left atrial dimension was positively correlated with the mRNA level of ACE, MMP-9, TIMP-1 and the concentration of calcium in atrial myocardium （P<0.01）. Right atrial area was only positively correlated with the mRNA level of ACE（P<0.05）. LVEDP was correlated with the mRNA level of MMP-9（P<0.05） while right atrial pressure was correlated positively with the mRNA level of TMP-l（P<0.01）.00) Immunohistochemistry Protein expression of MMP-1, MMP-9 from RAP atrial tissue were more significant than that from control tissue. Whereas protein expression of TIMP-1 from RAP atrial tissue was markedly less than that from controls. Conclusion（1） Chronic atrial fibrillation model was successfully established by rapid atrial更多还原