Atrial Overexpression Of Angiotensin Converting Enzyme2Improves The Canine Rapid Atrial Pacing-induced Electrical Remodeling

Background:Atrial fibrillation (AF) is the most common clinical arrhythmia and isassociated with cardiovascular morbidity and excessive mortality. Previousstudies have demonstrated activation of the local renin angiotensin system(RAS), especially angiotensin II (Ang II) have been found to not only playan important role in atrial structural remodeling, but also have potenteffects on atrial ion channels and connexins, showing strong proarrhythmiceffects.Angiotensin-converting enzyme2(ACE2) is a monocarboxypeptidasethat metabolizes vasoconstrictive octopeptide Ang II into vasodilativeheptapeptide angiotensin-(1-7)[Ang-(1-7)], thereby functioning as anegative regulator of the renin-angiotensin system. Recent studies haveshown overexpression of ACE2could suppress Ang II–mediatedmyocardial hypertrophy and fibrosis, and prevent cardiac dysfunction.Recengtly, some researchers found that the product of ACE2, Ang-(1-7), might have some antiarrhythmogenic effects during myocardialischemia-reperfusion. Furthermore, another study suggested that Ang-(1-7)could effectively prevent the shortening of90%action potential duration(APD90) and improve ion channels remodeling induced by chronic rapidatrial pacing in dogs. However, the effects of ACE2/Ang-(1-7) on atrial ionchannels and connexins remodeling and their mechanisms are still notclear.Objective:Previous studies have shown that epicardial gene painting causeshomogeneous and complete transmural atrial gene transfer. Therefore, thepurpose of this study was to investigate whether atrial overexpression ofACE2by homogeneous transmural atrial gene transfer can help to reverseAF induced atrial electrical remodeling and their mechanisms in a canineatrial pacing model.Method:Twenty-eight mongrel dogs of either gender, weighing20to30Kg,were randomly divided into4groups: Sham-operated (Sham), control, genetherapy with Ad-EGFP (Ad-EGFP group) and gene therapy with Ad-ACE2(Ad-ACE2group)(n=7per subgroup). All dogs in the control, Ad-EGFPand Ad-ACE2group were paced at450beats per minute for a period of14days. The dogs in Sham group were instrumented without pacing. After2 weeks, all dogs underwent thoracotomy operation and an invasiveelectrophysiology (EP) study, then receveid epicardial gene painting. Onpostgene transfer day21, animals underwent electrophysiology study,histology, and molecular studies, as described follow:1) To identify the potential pathologic substrate underlying conductionabnormalities in rapid-pacing dogs, histologic studies were performed.Atrial tissue sections was stained with Hematoxylin and eosin (H-E) orPicrosirius Red staining by traditional methods.2) To study the effect of ACE2overexpression on other RAS components,we measured the expression levels of ACE2, Ang II and Ang-(1-7) byreal-time PCR, western blot, ELISA and immunohistochemistry.3) To further explore the effects of overexpression of ACE2on atrial ionchannel and connexins protein, they were detected by western blot,confocal immunohistochemistry and real time RT-PCR.Results:In addition to sham group, after5weeks atrial tachypacing, AERP atall BCLs at either site decreased significantly,(AERP350-AERP250)/100became significantly smaller, suggesting a reduction of rate adaptation ofAERP. After3weeks of gene transfer, AF became inducible in all controland Ad-EGFP dogs, the inducibility and duration of AF increased dramatically compared with the baseline and the Sham group (P<0.01),whereas the inducibility and duration of AF were found to be markedlylower in the Ad-ACE2group than those in the control and Ad-EGFP group.Serious pericardial inflammation, effusion, and hemorrhage were notobserved in any of the dogs. Atrial myocyte from sham dogs showed anormal composition of sarcomeres distributed throughout the cell, and theintra-cellular space also appeared normal. In contrast, atrial myocytes ofcontrol and Ad-EGFP dogs showed a loss of some contractile materials andabnormal sarcomeres. In addition, extensive interstitial fibrosis, evidencedby Picrosirius Red stain was found in these tissues. Thick layers of fibroustissue were observed in the endocardium and epicardium. In contrast, thesepathologic abnormalities of atrial tissues were attenuated in the Ad-ACE2group.A quantitative analysis of fibrosis showed that the percentage offibrosis in all atrial regions in the Ad-ACE2group was markedly lowerthan that in the Sham and Ad-EGFP group (5.8±2.4%vs.11.9±2.3%and14.3±3.4%at the right atrial appendage, p<0.001), and was comparablewith that in the Sham group (p=0.614).ACE2protein expression in the control and Ad-EGFP group wassignificantly decreased compared with that in the Sham subjects; comparedwith the later, it was increased two folds in the Ad-ACE2group. Similarly, ACE2gene expression showed the same statistical trend as its proteinexpression.The relative expression levels of Ang II and Ang-(1-7) in the atrialtissue were evaluated by semi-quantitative analysis ofimmunohistochemistry. In comparison with the Sham and Ad-ACE2group,the expression levels of Ang II were significantly higher in the Ad-EGFPand control group, but the expression levels of Ang-(1-7) was lower.Corresponding to that, the changing trend of both Ang II and Ang-(1-7) inatrial tissue detected by ELISA was similar to the results ofsemi-quantitative analysis of immunohistochemistry.Real time RT-PCR analysis showed that, compared withSham-operated dogs, both Cav1.2and Kv4.2mRNA abundance werelower in the myocardium of control and Ad-EGFP dogs (P<0.01), but wasup-regulated significantly in Ad-ACE2group. No significant differencewas found in terms of Kv4.3and KChiP2mRNA abundance among thefour groups.Compared with Sham, the protein expression levels of Cx40wassignificantly increased in the control and Ad-EGFP group, but wassignificantly increased in Ad-ACE2group when compared with the controland Ad-EGFP dogs. In contrast, the protein expression levels of Cx43werehigher in Ad-ACE2group that those in the control and Ad-EGFP dogs. NO significant differences were observed in the expression of Cx40and Cx43between Sham and Ad-ACE2group. Furthermore, the mRNA expressionlevels of Cx40and Cx43were significantly higher in the control andAd-EGFP group than those in Sham and Ad-ACE2group, as shown in. Weevaluated connexin localization using confocal immunohistochemistry. Inlongitudinally sectioned atrial myocytes from all groups, Cx43was locatedmainly at the intercalated discs, with little signal at the lateral sarcolemma;in contrast, Cx40was located predominantly at the lateral sarcolemma, butthere was still a certain amount of signal at the intercalated discs.Conclusion:The salient findings of this study are:(i) overexpression of ACE2leadto a significant reduction of the endogenous Ang II level and a significantincrease of the endogenous Ang-(1-7) level, thus shifting the RAS balancetowards the protective axis;(ii) overexpression of ACE2attenuated cardiacfibrosisour;(iii) the present study demonstrated that, for the first time,overexpression of ACE2could improve ion channels and connexinsremodeling induced by chronic rapid atrial pacing in dogsIn conclusion, our results demonstrate that overexpression of ACE2byhomogeneous transmural atrial gene transfer could decreased theinducibility and duration of AF, the mechanism of which are associatedwith shifting the RAS balance towards the protective axis, attenuateing cardiac fibrosis, and improving ion channels and connexins remodelinginduced by chronic rapid atrial pacing in dogs.