Connexins and Atrial Fibrillation

Abstract

HomeCirculationVol. 125, No. 2Connexins and Atrial Fibrillation Free AccessEditorialPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessEditorialPDF/EPUBConnexins and Atrial FibrillationFilling in the Gaps Takeshi Kato, MD, PhD, Yu-ki Iwasaki, MD, PhD and Stanley Nattel, MD Takeshi KatoTakeshi Kato From the Department of Medicine and Research Center, Montreal Heart Institute and Université de Montréal, Montreal, Quebec, Canada. *Drs Kato and Iwasaki contributed equally to this work. Search for more papers by this author , Yu-ki IwasakiYu-ki Iwasaki From the Department of Medicine and Research Center, Montreal Heart Institute and Université de Montréal, Montreal, Quebec, Canada. *Drs Kato and Iwasaki contributed equally to this work. Search for more papers by this author and Stanley NattelStanley Nattel From the Department of Medicine and Research Center, Montreal Heart Institute and Université de Montréal, Montreal, Quebec, Canada. Search for more papers by this author Originally published8 Dec 2011https://doi.org/10.1161/CIRCULATIONAHA.111.075432Circulation. 2012;125:203–206Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: January 1, 2011: Previous Version 1 Atrial fibrillation (AF) is an extremely common arrhythmia with important consequences and presently suboptimal therapeutic options.1 A great deal of research has been performed to understand the detailed mechanisms of AF, with the hope that a better appreciation of the fundamental determinants of arrhythmogenesis will lead to the development of novel, more successful treatment possibilities.2Article see p 216One aspect of AF pathophysiology that has elicited great interest has been changes in gap junction/connexin physiology. An important role for altered gap junction function and the potential importance of gap junctions as a therapeutic target for AF were first emphasized by Spach and Starmer over 15 years ago.3 Since that time, our understanding of gap junction physiology, biophysics, and molecular biology has increased enormously,4,5 as have studies on their role in AF pathogenesis and management.2Gap junctions contain transmembrane ion-channel proteins called connexins (Figure,panel A). Connexons (containing 6 connexin molecules each) in the gap junctions of adjacent cardiomyocytes line up and attach, transferring ions or molecules <1 kDa freely between cells, coupling them electrically (Figure, panel B). Cardiac tissues express a variety of connexins, including connexin (Cx)40, Cx43, Cx45, Cx30.2/31.9, and Cx37,5 with all but Cx37 present in cardiomyocytes and the most important connexins in atrial tissue being Cx40 and Cx43. There is evidence that posttranslational modification, particularly connexin phosphorylation, is important in governing connexin localization and function.6Download figureDownload PowerPointFigure. A, Connexons contain 6 connexins each, arranged in a circle. Connexins in 1 cell connect with connexins in adjacent cells across the gap junction, connecting the cells electrically. B, Schematic of cell-to-cell connections in normal heart. C, Atrial fibrillation (AF) can be associated with a change in the number of connexins between cells. D, AF can also be associated with a change in connexin distribution, particularly lateralization of connexins.Given the central importance of connexins in cell-to-cell coupling, changes in connexin expression and distribution would be expected to have profound influences on cardiac conduction,2,5 a key determinant of reentry mechanisms that maintain AF.2 Furthermore, connexins can also affect refractoriness heterogeneity, also an important determinant of reentry,2 because tight cell coupling tends to smooth out variations in action potential duration. Thus, it is logical to consider the possibility that changes in connexins might be important in AF pathophysiology.An important role for connexins in AF is strongly supported by genetic studies indicating that connexin gene variants are associated with AF,7–10 although some of the results regarding specific gene-variants have been contradictory.8–10 A wide range of studies has investigated changes in connexin expression and distribution in clinical and experimental AF models (Table).11–35 Alterations in both total connexin expression (Figure, panel C) and connexin distribution, particularly redistribution from cell-end gap junctions to lateral margins (Figure, panel D), have been described in AF. However, the results show wide variations, with opposing results even within the same models (Table), causing considerable uncertainty about the nature of connexin changes in AF. Transgenic animal models have the potential to provide clearer insights into the role of connexins in AF pathophysiology by revealing the effects of specifically engineered changes in connexins. Here, too, however, there have been contradictory results, with studies indicating a clear increase36 or no change37 in atrial tachyarrhythmia susceptibility with Cx40 knockout. Although Cx40 loss suppresses atrial conduction in knockout mice,36 studies in myocyte strands suggest that Cx40 loss can actually improve atrial conduction.38 Small-molecule drugs that enhance gap junction conductance have been developed as potential treatments for AF, with results showing improvement in some models (ischemia and mitral valve disease–related AF) but little or no change in other clinically relevant paradigms.39–41 Thus, the role of connexin abnormalities in AF and the potential value of modulating connexin function to treat AF remain quite unclear on the basis of the literature.Table. Results of Previous Studies on Connexins in AFSpeciesAF Type/Animal ModelCx40 ProteinCx43 proteinRemarksAuthorYearexp.heteroexp.heterodephos.1DogATP-induced persistent AFNANA↑NANAAblation suppressed AF, Cx43Elvan1119972GoatATP-induced persistent AF→↑→→NAVan der Velden1219983GoatATP-induced persistent AF↓↑→→↑Van der Velden1320004RabbitVolume overload 8 wk↓NA↓NANARotigaptide did not prevent AFHaugan1420065DogSterile pericarditis 4 d↓↑↓↑NATransmural Cx40/Cx43 gradientRyu1520076MouseTNF overexpression 8–16 wk↓↑→NANALateralization of Cx43Sawaya1620077DogCongestive heart failure 2 wk→NA→NA↑Lateralization of Cx43Burstein1720098RatAutoimmune myocarditis→NA↓NANAHayano1820109RatAldosterone infusion 8 wkNANA→→→Reil19201110PigATP-induced persistent AFNANA↓NA→Cx43 transfer prevented AFBikou20201111RabbitThyroxine injection 4 wk↓↑↑↑NALateralization of Cx43Xiao21201112RatElevated afterload 20 wk↓NANANANAKim22201113HumanPost-operative AF (CAD)↑→→→NACx40 heterogenous, even in SR groupDupont23200114HumanChronic AF >1 y↑NA→NANALateralization of Cx40/Cx43Polontchouk24200115HumanChronic AF >1 y↓↑↓→NALateralization of Cx40/Cx43Kostin25200216HumanChronic AF >5 mo↓↑→→→Phosphorylated Cx40↑Nao26200317HumanChronic AF >6 mo→↓↑→NANACx40 ↓ in AF with complex activationKanagaratnam27200418HumanLone AF and AF with MVD↑NA↑NANACx40/Cx43 unchanged in lone AF,Wetzel28200519HumanPersistent AF >3 mo↓↓→→NAWilhelm29200620HumanChronic AF >3 mo→→→→NATakeuchi30200621HumanChronic AF up to 6 mo↓NA↑NA↑Lateralization of Cx43Rucker-Martin31200622HumanChronic AF >1 y (valve disease)↑↑↑NANACx40/Cx43 lateralization.Dhein32200823HumanPost-operative AF (CAD)→NA→NANACx40/43 reduced in arrested-heart surgery; not beating-heart or in AFLi33200924HumanPersistent AF (MVD, CAD)→NA→NANAGirmatsion34200925HumanPermanent AF >3 mo→NA↑↑↓Lateralization of Cx43Adam352010Abbreviations: ATP, atrial tachypacing; CAD, coronary artery disease; dephos, dephosphorylation; exp, expression; hetero, heterogeneity; SR, sinus rhythm; MVD, mitral valve disease; NA, not available.In this issue of Circulation, Igarashi et al report the results of a novel approach to controlling connexin expression in a porcine model of AF.42 The authors used an epicardial gene-painting approach to transfer Cx40 or Cx43 via adenoviral vectors to right and left atrial tissues, comparing gene-transferred with sham-operated animals. AF was induced by 2-second bursts of 42-Hz, 7.5-V atrial pacing separated by 2-second intervals to observe rhythm. Tachypaced sham dogs were in continuous AF after 5.8±0.6 days, whereas both Cx40- and Cx43-transferred dogs showed significantly increased probability of sinus rhythm, with no significant efficacy differences between Cx40- and Cx43-treated dogs. Atrial conduction was slowed in tachypaced sham dogs and substantially improved by connexin transfer. Tachypaced sham dogs showed significant decreases in total and phosphorylated Cx43 expression, with no change in Cx40 expression. Cx43 transfer normalized Cx43 expression in tachypaced dogs without affecting Cx40 expression; Cx40 transfer modestly increased Cx40 expression (by ≈20%) and did not affect Cx43 expression.The results of Igarashi et al are exciting in showing that enhancing connexin expression can have substantial AF-suppressing activity. They also point to a clear pathophysiological role of connexin downregulation in this AF model. In addition, they suggest that gene therapy approaches to increase connexin function and improve AF-related conduction abnormalities may be useful for treating AF. Furthermore, these results are consistent with another recent publication that showed efficacy of adenoviral-mediated Cx43 gene transfer in suppressing AF in a similar porcine model.20 Taken together, the results of these 2 investigations provide novel and important data about the role of connexins in AF.The study of Igarashi et al has a number of limitations that must be considered. First, no scrambled connexin gene control group was used to control for adenoviral effects per se, and therefore potential nonspecific effects cannot be excluded. Second, the porcine model of Igarashi et al displayed different properties from other prior animal AF models in which atrial tachypacing was used: Conduction was markedly slowed and refractoriness was unaffected, whereas previous models in which tachypacing was used to induce AF showed striking refractoriness abbreviation with very limited conduction slowing.43 Finally, the results of Cx40 gene transfer are difficult to reconcile with the observed physiology. Tachypaced AF dogs showed no Cx40 downregulation; Cx40 gene transfer increased Cx40 overall expression by only ≈20% without affecting intercalated-disk Cx40 expression and failed to alter strong downregulation of total (by ≈60%) and phosphorylated (by ≈90%) Cx43. Nevertheless, Cx40 gene transfer substantially improved conduction and suppressed AF; it is difficult to understand how this happened.Despite rapidly increasing knowledge about basic mechanisms underlying AF,2 many crucial mechanistic elements related to clinical AF management remain unresolved.44 The study of Igarashi et al is an important advance with respect to both pathophysiology and therapeutics. From the mechanistic perspective, it provides some of the most solid evidence to date on the potential participation of connexin dysfunction in AF. However, because the investigators studied a very specific animal model, much more work needs to be done to determine the relevance of their findings to various clinical forms of AF. From the therapeutic point of view, the study points the way to further development of connexin gene transfer therapy to treat AF. Overall, these results highlight the importance of learning much more about the role of gap junctions and their regulation in AF pathophysiology and management.AcknowledgmentsThe authors thank France Thériault for excellent secretarial assistance.Sources of FundingThis work was supported by the Canadian Institutes of Health Research (MOP 44365), the Quebec Heart and Stroke Foundation, the Fondation Leducq (European–North American Atrial Fibrillation Research Alliance), and the MITACS Network of Centers of Excellence.DisclosuresNone.FootnotesThe opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.*Drs Kato and Iwasaki contributed equally to this work.Correspondence to Stanley Nattel, MD, Department of Medicine and Research Center, Montreal Heart Institute, Université de Montréal, 5000 Belanger St E, Montreal H1T1C8, Quebec, Canada. E-mail stanley.[email protected]org