Cardiac conduction: a new multifactorial paradigm

Cardiac conduction is the phenomenon of sequential triggering of electrical action potentials and, thereby, mechanical contraction in the myocytes that make up cardiac muscle. While much has been learned about this phenomenon over the years, conduction abnormalities continue to underlie potentially lethal cardiac arrhythmias in a multitude of disease states. Further, it is also clear that conduction failure in many pathophysiological states results from multifactorial alterations to the myocardium. In the first part of this work, we demonstrate a novel role for the cardiac inward-rectifying potassium current (IK1) in determining conduction heterogeneities under normal conditions and the unmasking of regional heterogeneities in expression of the cardiac sodium channel (Nav1.5) when the cardiac sodium current (INa) is reduced. In the second part we demonstrate that pharmacological activation of repolarizing potassium currents - the ATP-sensitive potassium current (IKATP) and the slow delayed rectifier potassium current (IKs) - both slow conduction. However, these effects demonstrate different dependencies on INa availability depending on whether the potassium channel activated is inwardly-rectifying (such as Kir6.2 which carries IKATP) or voltage-gated (such as KvLQT1 which carries IKs). iv In the third part, we demonstrate a novel role for interstitial volume (VIS) in regulating anisotropic cardiac conduction and its dependence on gap junction coupling. These findings may help resolve the ongoing debate on the precise nature of the conduction velocity - gap junction relationship. Overall, this work advances the understanding of interrelationships between the determinants of cardiac conduction and their effect on conduction heterogeneities during normal as well as challenging conditions.

CARDIAC CONDUCTION: A NEW MULTIFACTORIAL PARADIGM by Rengasayee Veeraraghavan A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Bioengineering The University of Utah August 2011 Copyright © Rengasayee Veeraraghavan 2011 All Rights Reserved TheUniversityofUtahGraduateSchool STATEMENT OF DISSERTATION APPROVAL The dissertation of Rengasayee Veeraraghavan has been approved by the following supervisory committee members: , Chair Steven Poelzing 5/23/11 Date Approved , Member Rob S. MacLeod 5/16/11 Date Approved , Member John HB Bridge 5/13/11 Date Approved , Member Michael F Sheets 5/12/11 Date Approved , Member Edward Hsu 5/26/11 Date Approved and by the Department of Patrick Tresco Bioengineering and by Charles A. Wight, Dean of The Graduate School. , Chair of ABSTRACT Cardiac conduction is the phenomenon of sequential triggering of electrical action potentials and, thereby, mechanical contraction in the myocytes that make up cardiac muscle. While much has been learned about this phenomenon over the years, conduction abnormalities continue to underlie potentially lethal cardiac arrhythmias in a multitude of disease states. Further, it is also clear that conduction failure in many pathophysiological states results from multifactorial alterations to the myocardium. In the first part of this work, we demonstrate a novel role for the cardiac inward-rectifying potassium current (IK1) in determining conduction heterogeneities under normal conditions and the unmasking of regional heterogeneities in expression of the cardiac sodium channel (Nav1.5) when the cardiac sodium current (INa) is reduced. In the second part we demonstrate that pharmacological activation of repolarizing potassium currents - the ATP-sensitive potassium current (IKATP) and the slow delayed rectifier potassium current (IKs) - both slow conduction. However, these effects demonstrate different dependencies on INa availability depending on whether the potassium channel activated is inwardly-rectifying (such as Kir6.2 which carries IKATP) or voltage-gated (such as KvLQT1 which carries IKs). In the third part, we demonstrate a novel role for interstitial volume (VIS) in regulating anisotropic cardiac conduction and its dependence on gap junction coupling. These findings may help resolve the ongoing debate on the precise nature of the conduction velocity - gap junction relationship. Overall, this work advances the understanding of interrelationships between the determinants of cardiac conduction and their effect on conduction heterogeneities during normal as well as challenging conditions. iv To every mind that ever pursued a rational inquiry. TABLE OF CONTENTS ABSTRACT…. ................................................................................................... iii LIST OF FIGURES…......................................................................................... viii LIST OF ABBREVIATIONS AND ACRONYMS….............................................. x ACKNOWLEDGEMENTS…. ............................................................................. xii 1. INTRODUCTION.... ...................................................................................... 1 Background…. ........................................................................................ 2 Structural basis of function…. ................................................................. 3 The study of conduction…....................................................................... 5 Tissue excitability…. ............................................................................... 6 Intercellular coupling…............................................................................ 10 Conduction: A new multifactorial understanding….................................. 19 References…. ......................................................................................... 21 2. MECHANISMS UNDERLYING INCREASED RIGHT VENTRICULAR CONDUCTION SENSITIVITY TO FLECAINIDE CHALLENGE…. ..................................................................... 30 Introduction….......................................................................................... Methods…............................................................................................... Results…................................................................................................. Discussion…. .......................................................................................... Conclusions…. ........................................................................................ Limitations…. .......................................................................................... References…. ......................................................................................... 31 32 36 47 53 53 55 3. POTASSIUM CHANNEL ACTIVATORS DIFFERENTIALLY MODULATE THE EFFECT OF SODIUM CHANNEL BLOCKADE ON CARDIAC CONDUCTION….......................................................................................... 59 Introduction….......................................................................................... 60 Methods…............................................................................................... 61 Results…................................................................................................. Discussion…. .......................................................................................... Conclusions…. ........................................................................................ Limitations…. .......................................................................................... References…. ......................................................................................... 64 75 80 81 82 4. INTERSTITIAL VOLUME MODULATES THE CONDUCTION VELOCITY - GAP JUNCTION RELATIONSHIP …....................................................................................... 85 Introduction….......................................................................................... 86 Methods…............................................................................................... 87 Results…................................................................................................. 92 Discussion…. ..........................................................................................108 Conclusions…. ........................................................................................112 Limitations…. ..........................................................................................113 References…. .........................................................................................114 Appendix…..............................................................................................118 5. SUMMARY AND FUTURE DIRECTIONS….................................................121 Tissue excitability…. ...............................................................................123 Intercellular coupling…............................................................................127 Cardiac conduction: Revising our understanding…. ...............................129 Scope and limitations of our findings…. ..................................................129 The study of cardiac conduction: Beyond reductionism….......................129 Clinical perspectives…. ...........................................................................132 References…. .........................................................................................133 vii LIST OF FIGURES Figure ............................................................Page 2.1 Regional Nav1.5 expression…. ..................................................................... 37 2.2 Conduction velocity: representative data…. .................................................. 39 2.3 Conduction velocity: summary data…. .......................................................... 40 2.4 Regional Kir2.1 expression…......................................................................... 42 2.5 Conduction velocity: effects of IK1, INa blockade…. ........................................ 44 2.6 Conduction dependence on INa blockade…................................................... 45 3.1 Electrocardiograms: no di-4-ANEPPS.... ....................................................... 65 3.2 Electrocardiograms: 15 µM di-4-ANEPPS.... ................................................. 68 3.3 Conduction velocity.... ................................................................................... 71 3.4 Conduction dependence on INa blockade: effect of pinacidil…. ..................... 73 3.5 Conduction dependence on INa blockade: effect of R-L3…. .......................... 74 4.1 Expression of Cx43, pCx43…. .............................................................................. 93 4.2 Histology........................................................................................................ 95 4.3 Tissue water content.... ................................................................................. 98 4.4 Electrocardiograms and optical action potentials…. ...................................... 99 4.5 Conduction velocity: representative data…. .................................................. 101 4.6 Conduction velocity: summary data…. .......................................................... 103 4.7 Conduction velocity: dependence on Gj, VIS…. ............................................. 105 4.8 Arrhythmia incidence…. ................................................................................ 107 4.S1 Photomicrographs of isolated myocytes…. ................................................. 118 4.S2 Myocyte size: effects of albumin, mannitol…............................................... 119 4.S3 Representative frozen sections…................................................................ 120 5.1 Components of the action potential upstroke…. ............................................. 124 ix LIST OF ABBREVIATIONS AND ACRONYMS INa - Cardiac sodium current IK1 - Inward-rectifier potassium current IKr - Rapid delayed rectifier potassium current IKs - Slow delayed rectifier potassium current IKATP - ATP-sensitive potassium current BrS - Brugada syndrome SCD - Sudden cardiac death HF - Heart failure RV - Right ventricle LV - Left ventricle BCL - Basic cycle length Gj - Gap junction Cx43 - Connexin43 p Cx43 - Connexin43 phosphorylated at Ser368 H&E - Hematoxylin and eosin IS - Interstitial space VIS - Interstitial volume WW - Wet weight DW - Dry weight APD - Action potential duration (measured to full repolarization) APD30 - Action potential duration (measured to 30% repolarization) θ - Conduction velocity θT - Transverse conduction velocity θL - Longitudinal conduction velocity ARθ - Anisotropic ratio of conduction (defined as θL/θT) VT - Ventricular tachycardia BDM/DAM - Butanedione monoxime/Diacetyl monoxime xi ACKNOWLEDGEMENTS I would like to express my gratitude to the following people for their contributions to my personal and scientific development and who made this dissertation possible. First and foremost, thanks are owed to my advisor Dr. Steven Poelzing, for all the things I have learned from him scientific and otherwise, and for all the guidance and support. I have learned much from Dr. Poelzing about how to pursue scientific inquiry in the real world and about the importance of experiment design, of scientific rigor and identifying the right hypotheses to pursue. And Dr. Poelzing has also taught me a great deal about the importance and the craft of scientific communication. I am deeply grateful to my dissertation committee members for their scientific guidance. Dr. Rob MacLeod taught me a great deal about cardiac electrophysiology, particularly the importance of understanding how structure determines function. I have learned as much about teaching as I have science from the time spent in Dr. MacLeod's classes. I am also thankful to Dr. John Bridge, from whom I have learned much about theoretical and quantitative approaches to understanding electrophysiology, particularly at very fine spatio-temporal scales. I count the many discussions I have had with Dr. Bridge on topics ranging from the resting membrane potential to world history among the highlights of my graduate career. I owe thanks to Dr. Michael Sheets for the all the times he has patiently answered my questions about sodium channels and sodium channel modulating drugs. His insights on these topics have been critical not only in helping me design my experiments but also in the development of this dissertation. I must also thank Dr. Edward Hsu for always asking interesting questions about my research that revealed potential new lines of inquiry. His inputs on cardiac fiber structure have at once helped me understand the limitations of measurements made from the epicardial surface and to extending the mechanistic findings from such measurements to the threedimensional myocardium. I would be remiss were I not to acknowledge the vital importance of the wonderful training environment afforded by the Nora Eccles Harrison Cardiovascular Research and Training Institute (NEH-CVRTI). Dr. Kenneth Spitzer, Director of CVRTI, was always available for discussion and encouraged and supported me throughout my time at CVRTI as a research assistant. I would also like to thank Dr. Phil Ershler, Jayne Davis, Jerry Jenkins, Dennis King, Wilson Lobaina and Bruce Steadman whose technical expertise and assistance made it possible for me to conduct my research. xiii I owe a debt of gratitude to my fellow trainees in the trenches Prez Radwański, Anders Peter Larsen, Ryan Rigby and Katie Scuito for the many, no-holds-barred scientific discussions with my colleagues which were invaluable in helping me winnow out many, many red herrings. Thanks are also owed to Michael Heidinger for his help with experiments but far more importantly, for introducing this beer-drinking heathen to the pleasures of wine. But above all, I am grateful for their enduring friendships which made every day at work a little lighter and more rewarding. Chapter 2 of this dissertation was published in the March 2008 issue of Cardiovascular Research with Dr. Steven Poelzing as my co-author. Its reproduction in this dissertation is in accordance with the publication rights policies of Cardiovascular Research (Oxford Journals). xiv CHAPTER 1 INTRODUCTION 2 Background The heart is a pump; its primary function is mechanical. Its structural and functional unit is the cardiac myocyte, a brick shaped cell capable of contracting along its long axis. The heart's pump function is achieved through the contraction of its nearly eight billion cardiac myocytes in a specific sequence. Such synchronization of myocyte contraction is achieved through electrical signaling. Cardiac conduction is the process by which these electrical signals, known as action potentials (APs), are propagated from cell to cell, through the heart. The primary impetus for the study of cardiac conduction comes from the health consequences of aberrant conduction. Such aberrant conduction can precipitate disturbances in cardiac rhythm, termed arrhythmias and the consequent dyssynchrony of contraction can lead to potentially lethal compromise of the heart's pump function. Indeed, conduction disturbances underlie arrhythmias in a variety of different pathophysiological states and ventricular arrhythmias account for nearly 80%1 of the over 450,000 cases of sudden cardiac death (SCD) that occur annually2 in the U.S. alone. The prevention of such arrhythmias continues to pose a challenge in the absence of a clear understanding of the mechanisms of conduction failure. Therefore, there is an urgent need to understand the interrelationship between cardiac conduction and its various physiological determinants in both health and in disease. 3 Structural basis of function Any attempt to understand conduction must begin with an examination of the structural basis of the phenomenon. Macroscopically, the mammalian heart is a four chambered organ composed of two atria and two ventricles. Blood flow between these chambers occurs in series with the right atrium and ventricle handling pulmonary circulation and the left atrium and ventricle handling systemic circulation. At the tissue level, the heart is made up of cardiac muscle, a type of involuntary muscle distinct from smooth muscle. The structural unit of cardiac muscle is the cardiac myocyte, an electrically excitable, mechanically contractile cell. Each individual cardiac myocyte is a brick shaped cell, roughly six times as long as it is wide and packed with contractile proteins organized in bundles oriented along its long axis, enabling lengthwise contraction.3 These myocytes are coupled end-to-end, mechanically by adherens junctions and desmosomes (cell-cell junctions composed of cytoskeletal protein complexes) and electrically by gap junctions, forming fiber-like structures. In ventricular muscle, fibers of myocytes are organized in parallel producing an anisotropic layer structure. These layers are then arranged one on top of the other, with each layer slightly rotated in orientation relative to the previous to form the ventricular myocardium. The result is a gradual change in fiber orientation across the wall of the heart, called rotational anisotropy. As individual myocytes contract in sequence, their complex structural organization produces a macroscopic motion that enables the heart to function as a pump. Thus, the functional 4 properties of the heart derive directly from the structural organization of the myocardium. At the cellular scale, the electrical activity of myocytes results from asymmetric distribution of ions across its highly resistive cell membrane, called the sarcolemma. Whereas the intracellular concentration of potassium far exceeds its extracellular concentration, sodium is far more abundant outside the cell than it is inside. The net effect of the various ionic gradients is to render the interior of a resting cardiac myocyte negative relative to its exterior, thus giving rise to a resting membrane potential. However, the membrane is studded with ion channels which permit ionic current across the membrane, allowing for transient changes in the transmembrane potential called action potentials. These electrical action potentials actuate the contraction and relaxation of myocytes via Ca2+-based signaling. Rapid, synchronized triggering of action potentials in myocytes is achieved by means of electrical current flowing between myocytes in a process termed cardiac conduction. Each beat begins with action potentials arising spontaneously in the so-called pacemaker cells of the sinoatrial node. Electrical current flows from excited cells to resting cells electrically coupled to them, triggering action potentials in those cells. Excitation originating in the sinoatrial node is thus communicated via the atrioventricular node to the noncontractile, specialized conduction pathways of the His-Purkinje system and finally to the ventricular myocardium via numerous Purkinje-myocyte junctions. Excitation then propagates within the myocardium from myocyte to 5 myocyte via gap junctions, triggering synchronous contraction. The remainder of this text will focus mainly on this phenomenon of ventricular conduction. The study of conduction Historically, the phenomenon of cardiac conduction was first identified by Engelmann in strips of frog atrial muscle.4;5 Since then, significant advances have been made in understanding the biophysical mechanisms underlying cardiac conduction. Systematic study of any phenomenon demands that it be parameterized to enable the quantitative assessment of function. The primary metric of conduction at the tissue level is conduction velocity (θ), first measured by Lewis and colleagues in canine myocardium in 1914-15.6 Further, the aforementioned anisotropic structure of the ventricular myocardium confers direction dependence on conduction. This is quantified as anisotropy of conduction, defined as the ratio of velocity along the fastest direction of propagation to velocity along the slowest.7;8 A cellular metric relevant to the study of conduction is action potential upstroke velocity (dV/dtmax) - which has been used as a metric of excitability7 and even as a surrogate measure of conduction in isolated cells. In the subsequent sections, these metrics will be used to describe the phenomenon of ventricular conduction and its dependence on various anatomical and physiological factors. For the purposes of mechanistic inquiry, the process of cardiac conduction can be thought of as occurring in two phases: the generation of an 6 AP within a myocyte either spontaneously or in response to excitatory stimulus and the propagation of electrical impulses from cell to cell. Tissue excitability describes the properties of cardiac tissue concerning how its cells respond to a given electrical stimulus. Intercellular coupling is the process by which electrical excitation is communicated from one cell to another and is thought to occur by means of electrical current flow via gap junctions.9 Tissue excitability Excitability can be defined either in terms of the amount of excitatory current required to trigger an AP in a myocyte or in terms of the probability that an AP will be triggered by an excitatory pulse of a given strength. In neurons, the depolarization of membrane potential by the entry of sodium ions into cells was suggested as early as 1936 by Fenn and Cobb10 and later demonstrated by Hodgkin, Huxley and Katz.11-19 Hodgkin and Horowicz then extended this concept to muscle action potentials.20 Through these early studies, it was recognized that the sodium conductance of the cell membrane is very low except during the depolarizing phase of the action potential during which it increases rapidly before dropping off again. These and other findings prompted Hodgkin and Huxley to suggest that there may exist specialized "pores" or channels through which Na+ permeates the cell membrane.15 Beginning with Narahashi and colleagues' 1964 report on the sodium current blocking properties of tetrodotoxin,21 pharmacological blockers were used to further elucidate properties of the sodium current. These led up to Neher and 7 Sackmann's landmark 1976 paper demonstrating current recordings from single ion channels22 and eventually, the identification of the cardiac sodium current (INa) - carried by the cardiac isoform of the voltage-gated sodium channel (Nav1.5)23 and primarily responsible for depolarization in cardiac myocytes. Subsequent studies using whole heart electrode mapping techniques24;25 and computer simulations26;27 extended these insights from cellular electrophysiology to the tissue and organ levels. Overall, it is now understood that sodium channel availability is a key determinant of excitability.27 Further, since sodium channel recovery from inactivation is voltage-dependent, the resting membrane potential (RMP) is also a determinant of excitability.28;29 In turn, K+ channels, through their influence on RMP, function as secondary determinants of Na+ channel availability. Sodium channels and excitability The magnitude of the sodium current is perhaps the most important determinant of excitability. Indeed, pharmacological blockade of sodium channels has been demonstrated to slow conduction and increase arrhythmia propensity in both animal models24;30 and humans.31-34 Furthermore, loss-offunction mutations in SCN5a, the gene that encodes the principal (α) subunit of the cardiac sodium channel (Nav1.5), lead to reduced INa and have been linked with several pathophysiologic conditions such as the Brugada syndrome, progressive cardiac conduction disease (inherited Lenègre's 8 disease), arrhythmogenic right ventricular cardiomyopathy (ARVC), and dilated cardiomyopathy.35;36 While pathological changes in INa secondary to mutations in Nav1.5 are well recognized, other mechanisms are being brought to light by recent research. Mutations in regulatory subunits of the sodium channel such as βsubunits (SCN1b through SCN4b) can produce phenotypes similar to sodium channel loss of function.37-40 The sodium current can also be affected by alterations of cytoskeletal proteins that are involved in targeting channels to the membrane (such as ankyrin-G, tubulin)41;42 or form macromolecular complexes with them (plakophilin-2).43 Indeed, mutations in ankyrin-G have been linked to a form of the Brugada syndrome.41;44 It has even been suggested that the localization of cardiac sodium channels to the intercalated disks may modulate INa in an intercellular coupling-dependent manner.45 Calcium channels and excitability The L-type calcium current, which is the primary calcium current in cardiac myocytes, is another inward current and activates when the action potential upstroke reaches about -25 mV. Therefore, it plays a greater role in sustaining the plateau of the cardiac action potential than in depolarization itself. However, in silico results suggest that the L-type Ca2+ current may play a role in depolarization under conditions where sodium channel availability is drastically reduced.27 9 Potassium channels and excitability Various potassium currents account for the majority of the outward current during the action potential and are responsible for repolarizing the membrane.46 Most potassium currents activate after the action potential upstroke and, therefore, are not directly involved in depolarization. However, changes in potassium currents can modulate a) the resting membrane potential, b) the action potential duration, and thereby, the diastolic interval. Thus, potassium currents determine resting potential between action potentials and the time available at that potential for sodium channels to recover from inactivation, thereby modulating sodium channel availability. Although most of the outward potassium current flows during the repolarizing phase of the action potential, the inward rectifier potassium current (IK1) is active during rest and early depolarization, albeit with a much smaller conductance compared to the sodium current. However, it does represent an opposing force to depolarization before the sodium current is fully activated; therefore, modulating IK1 could modulate excitability and conduction velocity. Indeed IK1 heterogeneities have been demonstrated to determine heterogeneities in excitability during ventricular fibrillation.47 An experimental investigation of IK1 heterogeneities and their role in determining heterogeneities in normal ventricular conduction will be presented in Chapter 2. Specifically, Chapter 2 details experimental results demonstrating that regional θ heterogeneities are determined by regional IK1 heterogeneities; however, the dependence of θ heterogeneities on 10 heterogeneous sodium channel expression is unmasked under conditions where INa is functionally reduced. Extending this line of reasoning, it is also possible that pharmacologically augmenting other potassium currents so as to increase outward current during depolarization may reduce excitability and slow conduction. Indeed, our group has demonstrated that pharmacological activation of the rapid delayed rectifier potassium current (IKr) can slow cardiac conduction and exacerbate the effects of pharmacological INa blockade.48 An investigation of pharmacological activation of the ATP-sensitive K+ current (IKATP) and the slow delayed rectifier potassium current (IKs) for possible effects on conduction and the mechanism underlying such effects will be detailed in Chapter 3. Specifically, experimental results presented in Chapter 3 demonstrate conduction slowing during pharmacological activation of IKATP and IKs. Importantly, IKATP activation preferentially affects conduction when Na+ channel availability is not reduced, whereas IKs activation may preferentially impact conduction when Na+ channel availability is reduced. Intercellular coupling The communication of activation from one cardiac myocyte to another involves the transmission of electric current between them. An excited myocyte is at a more positive potential relative to a resting myocyte which creates an electrical potential gradient between the two. This gradient results 11 in the flow of electric current between the cells through low resistance pathways. As a result, cardiac muscle behaves like an electrical syncytium. And, as early as 1952, Silvio Weidmann demonstrated that Purkinje fibers could be treated as a continuous core conductor and electrical propagation through such media described using cable theory.49 Cable theory Originally developed in 1855-56 by William Thomson (aka Lord Kelvin) to model the transmission of electrical signals through transatlantic telegraph cables, cable theory or the continuous core conductor model describes the decay in space and time of an electrical impulse transmitted through a long cylindrical conductor jacketed in a resistive material.50;51 This treatment was extended first to describe action potential propagation through nerve fibers by various investigators and later to cardiac Purkinje fibers by Silvio Weidmann.49 Briefly, for a fiber of radius a with intracellular resistance ri per unit length and extracellular resistance re per unit length bounded by a membrane with resistance rm per unit length and capacitance cm per unit length, the transmembrane potential vm is described by: rm ∂ 2 v m ∂v − c m rm m − v m = 0 2 ri + re ∂x ∂t By extending this treatment, it has been demonstrated that the velocity of conduction θ is related to ri and re by: 12 θ =k 1 ri + re where k is a constant representing membrane properties.52 In the heart, ri and re are determined by various anatomical structures that compose the electric current path inside, outside and between myocytes. These will be discussed individually in subsequent sections. Gap junctions and cardiac conduction While the cable theory description of conduction appeared consistent with the then-prevalent notion of cytoplasmic continuity between cardiac myocytes, electron microscopy revealed myocytes to be fully bounded by membrane.53 Therefore, the physical nature of electrical connectivity between cells remained a mystery until the 1960s when Lloyd Barr demonstrated the existence of low resistance pathways between cardiac myocytes.54;55 These were later labeled gap junctions. Since their identification, gap junctions have been the subject of intense study. We now know that, in the adult myocardium, functional gap junction channels are located primarily at the intercalated disks at the ends of myocytes, and are composed of two hexamers of proteins from the connexin family.56 While Weidmann and others recognized the significant contribution of these channels to axial resistance, the junctional conductance was still deemed high enough for cardiac conduction to be a continuous process.57;58 It should be noted that the majority of experiments up to the late 1960s had 13 been performed in cardiac muscle bundles and Purkinje fibers that structurally resemble a cable and also that the measurements of conduction were largely made on a macroscopic scale. Therefore, it is not surprising that continuous cable theory continued to fit well the available data on cardiac conduction. Anisotropic nature of conduction As described previously, cardiac tissue is anisotropic in structure due to the lengthwise organization of structures within the elongated cardiac myocytes, the end-to-end coupling between myocytes and the parallel organization of fibers in the myocardium. These structural factors suggest that the conduction of electrical impulses through the myocardium should be anisotropic. Indeed, during the late 1950s, the anisotropic nature of cardiac conduction began to be recognized. Woodbury and Crill, followed by others, demonstrated direction-dependent spatial decay of current injected into the myocardium, with the greatest decay along the fiber axis and the lowest perpendicular to it.59 These findings were followed by theoretical studies extending the continuous cable theory to cardiac conduction in 2D and 3D models of tissue which incorporated anisotropic resistivities for the intra- and extracellular spaces.60-62 These models were eventually developed into the ‘bidomain model' and suggested that cardiac conduction velocity should be anisotropic. By treating cardiac muscle as a continuous but anisotropic medium, Clerc demonstrated the inverse square relationship, predicted by 14 cable theory, between velocity and axial resistance (Ra) for conduction both parallel and perpendicular to the fiber axis.63 Experimental estimations of conduction velocity made using microelectrode recordings from multiple sites along different directions in the ventricular myocardium of various mammals agreed with these findings and even linked anisotropic conduction to fiber orientation.64;65 Transition from syncytium to discontinuous model While all the afore-mentioned findings appear consistent with the cable theory and with cardiac tissue being a continuous medium, measurements made during the late 70s, at greater resolution, began to reveal a more complicated picture. Specifically, microscopic measurements in atrial and ventricular myocardium made by Spach et al. revealed the maximal action potential upstroke velocity (dV/dtmax) and the time constant of the action potential foot (τfoot) to be inversely and directly related to conduction velocity, respectively.7;66 However, if the myocardium were to behave as a continuous but anisotropic medium, as is assumed by continuous cable theory, the time course of depolarization should be independent of the velocity of propagation and dV/dtmax should be constant between different directions of propagation.67 The afore-mentioned results are therefore contrary to cable theory predictions, whereas the previously discussed results obtained in cable-like Purkinje fibers were in agreement with cable theory. Analysis of the observed relationship between Ra and dV/dtmax pointed at conduction being 15 discontinuous as a result of high axial resistance encountered at the gap junctions. Indeed, direct measurements of gap junction resistance have revealed that the resistance encountered at the intercalated disk is roughly equal to the total axial intracellular resistance along a myocyte's entire length.9 This also suggests that conduction is slower and more discontinuous transverse to the fiber axis than longitudinally due to the greater number of gap junctions encountered per unit length. Indeed these inferences have been borne out by further experimental observations.9 For a more detailed discussion of the discontinuous nature of conduction, the reader is referred to a review by Spach and colleagues.67 Based on the inverse relationship between conduction velocity and axial resistance, it can be inferred that the components of the axial resistance are the important determinants of anisotropic conduction. These include the resistance encountered at each intercalated disk (i.e., functional expression and conductance of gap junctions), the number of gap junctions encountered per unit length and tissue architecture which includes myocyte geometry, extracellular volume and structural arrangement of myocytes.52 Over the years, each one of these factors has been examined for its influence on anisotropic conduction under normal and pathophysiological conditions. Gap junctions Given their key role in electrically coupling myocytes, it comes as no surprise that gap junctions are a key determinant of anisotropic conduction. 16 Indeed, conduction slowing by pharmacological gap junction uncoupling is well established.9;30;68 Gap junctions are located at the ends of myocytes69 which themselves are elongated in shape and organized end-to-end, and they constitute the primary source of delay during microscopic conduction.70;71 Therefore, the effect of alterations in gap junction coupling on conduction is proportional to the number of gap junctions encountered per unit length of the conduction path. As a result, changes in coupling preferentially impact transverse conduction over longitudinal and thereby alter anisotropy of conduction.7;72;73 In contrast, the impact of genetic alterations in connexin43 expression has been unclear. Results in transgenic mice heterozygous for a null mutation in connexin43 have varied between significant conduction slowing relative to wild type mice74;75 and no measurable difference in conduction velocity.76-82 The question of pathophysiological gap junction remodeling is even more complex. Gap junction remodeling in heart failure has been implicated in aberrant conduction and arrhythmogenesis;83 however, in a pacing induced canine heart failure model, gap junction remodeling was reported to precede conduction abnormalities.84 Likewise, the role of gap junction closure in mediating phase 1B arrhythmias in acute ischemia remains to be elucidated.85 A factor that confounds the impact of pathophysiological gap junction remodeling is that it often consists of downregulation as well as lateralization of connexins.86 Likewise, the effects of cellular redistribution of connexins during normal ventricular growth hypertrophy on conduction are 17 obscured by concomitant changes in myocyte geometry.87 Further, connexin life cycle and function are dynamically regulated by various biochemical pathways including the phosphorylation / dephosphorylation of multiple Cterminal residues.88-91 As with the cardiac Na+ channel, it is becoming increasingly apparent that gap junctions are part of a macromolecular complex and that their function is modulated by other proteins that associate with them. Indeed, it has been demonstrated that mechanical junctions, i.e., adherens junctions and desmosomes, must form at the intercalated disc before gap junctions can be formed and that alterations in junctional/cytoskeletal proteins can alter gap junction function.92;93 Further, recent studies have demonstrated alteration of gap junction expression / function secondary to changes in proteins involved in connexin trafficking.94 Consequently, the conduction velocity - gap junction relationship remains a topic of active research and will be discussed in greater detail in Chapter 4. Myocyte geometry Cardiac myocytes are elongated cells organized end-to-end with gap junctions, which form the electrical connections between myocytes, localized at the ends of myocytes. Consequently the contribution of cytoplasmic resistance to total axial resistance of myocardial tissue is direction-dependent and alterations in myocyte geometry affect conduction in a directiondependent manner and alter anisotropy.7;73 However, alterations in myocyte 18 geometry also alter the number of gap junctions per unit length of the conduction path, which in turn is an important determinant of conduction.9;70;71 While some in silico studies have suggested a positive correlation between cell size and conduction velocity,95;96 other studies have demonstrated a negative correlation between myocyte diameter and conduction velocity.97 Indeed, a recent simulation study has suggested that myocyte shape may be as important as myocyte size, if not more so, with respect to conduction velocity.98 Changes in myocyte geometry occur in a variety of conditions, both normal (such as growth87) and pathophysiologic (such as hypertrophic cardiomyopathy99). However, these conditions often affect connexin expression and localization as well, confounding the effects on conduction. Alternately, myocyte geometry can also change in the acute time scale in response to ischemia,100 osmolarity changes101 etc. The impact of such changes on anisotropic conduction is an area of ongoing research and will be discussed in greater detail in Chapter 4. Interstitial volume If one considers that myocytes are electrically coupled by gap junctions, it becomes clear that the current flowing through gap junctions requires a return path that would involve the interstitial space, i.e., the space between myocytes. As a result, changes in interstitial resistance, resulting from changes in its volume and/or resistivity, alter the total axial resistance of 19 myocardial tissue and thereby affect conduction in a direction-dependent manner and alter anisotropy.52;102 Cable theory-based models suggest that as interstitial volume increases, the interstitial resistance would decrease and consequently, conduction velocity should increase as well. 52;102 However, other in silico studies suggested the possibility of non-gap junction dependent mechanisms of intercellular coupling.103-105 Should such mechanisms indeed be operative in the heart, they could markedly alter the relationship between conduction velocity and VIS. We have investigated experimentally the effects of modulating VIS on conduction velocity and its dependence on gap junction coupling. The findings from these studies and the possible mechanisms underlying them will be discussed in Chapter 4. Specifically, we demonstrate an inverse relationship between VIS and θT and that myocardial edema is associated with slowed conduction, increased ARθ, increased conduction dependence on gap junctions and increased arrhythmia propensity. Conduction: A new multifactorial understanding As research continues into the mechanisms underlying cardiac conduction in both health and in disease, it is increasingly clear that the various determinants of conduction are interrelated. With respect to excitability, conduction velocity and its dependence on the sodium current (INa) are modulated by potassium currents such as IK1. Likewise, conduction dependence on gap junction coupling is modulated by changes in myocyte 20 geometry and interstitial volume. Taken together, these results suggest that changes in one determinant of conduction can unmask its dependence on another and the full complement of these determinants must be considered in order to determine the safety of conduction under any given condition. 21 References 1. Mehra, R. Global public health problem of sudden cardiac death. J Electrocardiol. 2007;40:S118-22. 2. Zheng, ZJ, Croft, JB, Giles, WH, Mensah, GA. Sudden cardiac death in the United States, 1989 to 1998. Circulation. 2001;104:2158-63. 3. Campbell, SE, Gerdes, AM, Smith, TD. Comparison of regional differences in cardiac myocyte dimensions in rats, hamsters, and guinea pigs. Anat Rec. 1987;219:53-9. 4. Engelmann, Th. Wilh. Ueber die leitung der erregung im herzmuskel. Pflügers Archiv European Journal of Physiology. 1875;11(1), 465-480. 5. Lewis, T and Rothschild, M. A. The excitatory process in the dog's heart. Part II. The ventricles. Philosophical Transactions of the Royal Society of London. Series B, Containing Papers of a Biological Character. 1915;206:181-226. 6. Lewis, T, Meakins, J, and White, P. D. The excitatory process in the dog's heart. Part I. The auricles. Philosophical Transactions of the Royal Society of London. Series B, Containing Papers of a Biological Character. 1914;205:375420. 7. Spach, MS, Miller, WT 3rd, Geselowitz, DB, Barr, RC, Kootsey, JM, Johnson, EA. The discontinuous nature of propagation in normal canine cardiac muscle. Evidence for recurrent discontinuities of intracellular resistance that affect the membrane currents. Circ Res. 1981;48:39-54. 8. Spach, MS. Anisotropy of cardiac tissue: a major determinant of conduction? J Cardiovasc Electrophysiol. 1999;10:887-90. 9. Rohr, S. Role of gap junctions in the propagation of the cardiac action potential. Cardiovasc Res. 2004;62:309-22. 10. Fenn, WO, Cobb, DM. Electrolyte changes in muscle during activity. Am J Physiol. 1936;115:345-56. 11. Hodgkin, AL, Katz, B. The effect of sodium ions on the electrical activity of giant axon of the squid. J Physiol. 1949;108:37-77. 12. Hodgkin, AL. Conduction of the nervous impulse: some recent experiments. Br Med Bull. 1950;6:322-5. 13. Hodgkin, AL, Huxley, AF. The components of membrane conductance in the giant axon of Loligo. J Physiol. 1952;116:473-96. 14. Hodgkin, AL, Huxley, AF. Propagation of electrical signals along giant nerve 22 fibers. Proc R Soc Lond B Biol Sci. 1952;140:177-83. 15. Hodgkin, AL, Huxley, AF. A quantitative description of membrane current and its application to conduction and excitation in nerve. J Physiol. 1952;117:500-44. 16. Hodgkin, AL, Huxley, AF. The dual effect of membrane potential on sodium conductance in the giant axon of Loligo. J Physiol. 1952;116:497-506. 17. Hodgkin, AL, Huxley, AF. Currents carried by sodium and potassium ions through the membrane of the giant axon of Loligo. J Physiol. 1952;116:449-72. 18. Hodgkin, AL, Huxley, AF, Katz, B. Measurement of current-voltage relations in the membrane of the giant axon of Loligo. J Physiol. 1952;116:424-48. 19. Hodgkin, AL, Huxley, AF. Movement of sodium and potassium ions during nervous activity. Cold Spring Harb Symp Quant Biol. 1952;17:43-52. 20. Hodgkin, AL, Horowicz, P. Movements of Na and K in single muscle fibres. J Physiol. 1959;145:405-32. 21. Narahashi, T, Moore, JW, Scott, WR. Tetrodotoxin blockage of sodium conductance increase in lobster giant axons. J Gen Physiol. 1964;47:965-74. 22. Neher, E, Sakmann, B. Single-channel currents recorded from membrane of denervated frog muscle fibres. Nature. 1976;260:799-802. 23. Fozzard, HA. Cardiac sodium and calcium channels: a history of excitatory currents. Cardiovasc Res. 2002;55:1-8. 24. Buchanan, JW Jr, Saito, T, Gettes, LS. The effects of antiarrhythmic drugs, stimulation frequency, and potassium-induced resting membrane potential changes on conduction velocity and dV/dtmax in guinea pig myocardium. Circ Res. 1985;56:696-703. 25. Cranefield, PF, Wit, AL, Hoffman, BF. Conduction of the cardiac impulse. 3. Characteristics of very slow conduction. J Gen Physiol. 1972;59:227-46. 26. Lieberman, M, Kootsey, JM, Johnson, EA, Sawanobori, T. Slow conduction in cardiac muscle. A biophysical model. Biophys J. 1973;13:37-55. 27. Shaw, RM, Rudy, Y. Ionic mechanisms of propagation in cardiac tissue. Roles of the sodium and L-type calcium currents during reduced excitability and decreased gap junction coupling. Circ Res. 1997;81:727-41. 28. Weidmann, S. The effect of the cardiac membrane potential on the rapid availability of the sodium-carrying system. J Physiol. 1955;127:213-24. 29. Kagiyama, Y, Hill, JL, Gettes, LS. Interaction of acidosis and increased 23 extracellular potassium on action potential characteristics and conduction in guinea pig ventricular muscle. Circ Res. 1982;51:614-23. 30. Rohr, S, Kucera, JP, Kleber, AG. Slow conduction in cardiac tissue, I: effects of a reduction of excitability versus a reduction of electrical coupling on microconduction. Circ Res. 1998;83:781-94. 31. Heath, BM, Cui, Y, Worton, S, Lawton, B, Ward, G, Ballini, E, Doe, CP, Ellis, C, Patel, BA, McMahon, NC. Translation of flecainide- and mexiletine-induced cardiac sodium channel inhibition and ventricular conduction slowing from nonclinical models to clinical. J Pharmacol Toxicol Methods. 2010. 32. Terrenoire, C, Simhaee, D, Kass, RS. Role of sodium channels in propagation in heart muscle: how subtle genetic alterations result in major arrhythmic disorders. J Cardiovasc Electrophysiol. 2007;18:900-5. 33. Pratt, CM. The Cardiac Arrhythmia Suppression Trial. Introduction: the aftermath of the CAST--a reconsideration of traditional concepts. Am J Cardiol. 1990;65:1B-2B. 34. Pratt, CM, Moye, LA. The Cardiac Arrhythmia Suppression Trial: background, interim results and implications. Am J Cardiol. 1990;65:20B-29B. 35. Amin, AS, Asghari-Roodsari, A, Tan, HL. Cardiac sodium channelopathies. Pflugers Arch. 2010;460:223-37. 36. Chen, PS, Priori, SG. The Brugada syndrome. J Am Coll Cardiol. 2008;51:1176-80. 37. Abriel, H. Cardiac sodium channel Na(v)1.5 and interacting proteins: physiology and pathophysiology. J Mol Cell Cardiol. 2010;48:2-11. 38. Hakim, P, Brice, N, Thresher, R, Lawrence, J, Zhang, Y, Jackson, AP, Grace, AA, Huang, CL. Scn3b knockout mice exhibit abnormal sino-atrial and cardiac conduction properties. Acta Physiol (Oxf). 2010;198:47-59. 39. Remme, CA, Scicluna, BP, Verkerk, AO, Amin, AS, van Brunschot, S, Beekman, L, Deneer, VH, Chevalier, C, Oyama, F, Miyazaki, H, Nukina, N, Wilders, R, Escande, D, Houlgatte, R, Wilde, AA, Tan, HL, Veldkamp, MW, de Bakker, JM, Bezzina, CR. Genetically determined differences in sodium current characteristics modulate conduction disease severity in mice with cardiac sodium channelopathy. Circ Res. 2009;104:1283-92. 40. Tan, BH, Pundi, KN, Van Norstrand, DW, Valdivia, CR, Tester, DJ, MedeirosDomingo, A, Makielski, JC, Ackerman, MJ. Sudden infant death syndromeassociated mutations in the sodium channel beta subunits. Heart Rhythm. 2010;7:771-8. 24 41. Mohler, PJ, Bennett, V. Ankyrin-based cardiac arrhythmias: a new class of channelopathies due to loss of cellular targeting. Curr Opin Cardiol. 2005;20:18993. 42. Casini, S, Tan, HL, Demirayak, I, Remme, CA, Amin, AS, Scicluna, BP, Chatyan, H, Ruijter, JM, Bezzina, CR, van Ginneken, AC, Veldkamp, MW. Tubulin polymerization modifies cardiac sodium channel expression and gating. Cardiovasc Res. 2010;85:691-700. 43. Sato, PY, Musa, H, Coombs, W, Guerrero-Serna, G, Patino, GA, Taffet, SM, Isom, LL, Delmar, M. Loss of plakophilin-2 expression leads to decreased sodium current and slower conduction velocity in cultured cardiac myocytes. Circ Res. 2009;105:523-6. 44. Hashemi, SM, Hund, TJ, Mohler, PJ. Cardiac ankyrins in health and disease. J Mol Cell Cardiol. 2009;47:203-9. 45. Kucera, JP, Rohr, S, Rudy, Y. Localization of sodium channels in intercalated disks modulates cardiac conduction. Circ Res. 2002;91:1176-82. 46. Snyders, DJ. Structure and function of cardiac potassium channels. Cardiovasc Res. 1999;42:377-90. 47. Warren, M, Guha, PK, Berenfeld, O, Zaitsev, A, Anumonwo, JM, Dhamoon, AS, Bagwe, S, Taffet, SM, Jalife, J. Blockade of the inward rectifying potassium current terminates ventricular fibrillation in the guinea pig heart. J Cardiovasc Electrophysiol. 2003;14:621-31. 48. Larsen, AP, Olesen, SP, Grunnet, M, Poelzing, S. Pharmacological activation of IKr impairs conduction in guinea pig hearts. J Cardiovasc Electrophysiol. 2010;21:923-9. 49. Weidmann, S. The electrical constants of Purkinje fibres. J Physiol. 1952;118:348-60. 50. Thomson, William. On peristaltic induction of electric currents. Proceedings of the Royal Society of London 8, 121-132. 1856. 51. Thomson, William. On the theory of the electric telegraph. Proceedings of the Royal Society of London 7, 382-399. 1854. 52. Spach, MS, Heidlage, JF, Barr, RC, Dolber, PC. Cell size and communication: role in structural and electrical development and remodeling of the heart. Heart Rhythm. 2004;1:500-15. 53. Sjostrand, FS, Andersson, E. Electron microscopy of the intercalated discs of cardiac muscle tissue. Experientia. 1954;10:369-70. 25 54. Barr, L, Dewey, MM, Berger, W. Propagation of action potentials and the structure of the nexus in cardiac muscle. J Gen Physiol. 1965;48:797-823. 55. Dewey, MM, Barr, L. Intercellular connection between smooth muscle cells: the nexus. Science. 1962;137:670-2. 56. Sohl, G, Willecke, K. Gap junctions and the connexin protein family. Cardiovasc Res. 2004;62:228-32. 57. Weidmann, S. The diffusion of radiopotassium across intercalated disks of mammalian cardiac muscle. J Physiol. 1966;187:323-42. 58. Weidmann, S. Electrical constants of trabecular muscle from mammalian heart. J Physiol. 1970;210:1041-54. 59. Crill, WE, Rumery, RE, Woodbury, JW. Effects of membrane current on transmembrane potentials of cultured chick embryo heart cells. Am J Physiol. 1959;197:733-5. 60. Jack, J. J. B., Noble, D., and Tsien, R. W. Electric Current Flow in Excitable Cells. Clarendon Press. 75. Oxford, Clarendon Press. 61. Peskoff, A. Electric potential in cylindrical syncytia and muscle fibers. Bull Math Biol. 1979;41:183-92. 62. Peskoff, A. Electric potential in three-dimensional electrically syncytial tissues. Bull Math Biol. 1979;41:163-81. 63. Clerc, L. Directional differences of impulse spread in trabecular muscle from mammalian heart. J Physiol. 1976;255:335-46. 64. Draper, MH, Mya-Tu, M. A comparison of the conduction velocity in cardiac tissues of various mammals. Q J Exp Physiol Cogn Med Sci. 1959;44:91-109. 65. Sano, T, Takayama, N, Shimamoto, T. Directional difference of conduction velocity in the cardiac ventricular syncytium studied by microelectrodes. Circ Res. 1959;7:262-7. 66. Spach, MS, Miller, WT 3rd, Dolber, PC, Kootsey, JM, Sommer, JR, Mosher, CE Jr. The functional role of structural complexities in the propagation of depolarization in the atrium of the dog. Cardiac conduction disturbances due to discontinuities of effective axial resistivity. Circ Res. 1982;50:175-91. 67. Spach, MS. Transition from a continuous to discontinuous understanding of cardiac conduction. Circ Res. 2003;92:125-6. 68. Kojodjojo, P, Kanagaratnam, P, Segal, OR, Hussain, W, Peters, NS. The effects of carbenoxolone on human myocardial conduction: a tool to investigate 26 the role of gap junctional uncoupling in human arrhythmogenesis. J Am Coll Cardiol. 2006;48:1242-9. 69. Hoyt, RH, Cohen, ML, Saffitz, JE. Distribution and three-dimensional structure of intercellular junctions in canine myocardium. Circ Res. 1989;64:563-74. 70. Kucera, JP, Kleber, AG, Rohr, S. Slow conduction in cardiac tissue: insights from optical mapping at the cellular level. J Electrocardiol. 2001;34 Suppl:57-64. 71. Rohr, S, Salzberg, BM. Discontinuities in action potential propagation along chains of single ventricular myocytes in culture: multiple site optical recording of transmembrane voltage (MSORTV) suggests propagation delays at the junctional sites between cells. Biol. Bull. Mar. Biol. Lab. 1992;183:342-43. 72. Rohr, S, Salzberg, BM. Multiple site optical recording of transmembrane voltage (MSORTV) in patterned growth heart cell cultures: assessing electrical behavior, with microsecond resolution, on a cellular and subcellular scale. Biophys J. 1994;67:1301-15. 73. Saffitz, JE, Kanter, HL, Green, KG, Tolley, TK, Beyer, EC. Tissue-specific determinants of anisotropic conduction velocity in canine atrial and ventricular myocardium. Circ Res. 1994;74:1065-70. 74. Eloff, BC, Lerner, DL, Yamada, KA, Schuessler, RB, Saffitz, JE, Rosenbaum, DS. High resolution optical mapping reveals conduction slowing in connexin43 deficient mice. Cardiovasc Res. 2001;51:681-90. 75. Guerrero, PA, Schuessler, RB, Davis, LM, Beyer, EC, Johnson, CM, Yamada, KA, Saffitz, JE. Slow ventricular conduction in mice heterozygous for a connexin43 null mutation. J Clin Invest. 1997;99:1991-8. 76. Morley, GE, Vaidya, D, Samie, FH, Lo, C, Delmar, M, Jalife, J. Characterization of conduction in the ventricles of normal and heterozygous Cx43 knockout mice using optical mapping. J Cardiovasc Electrophysiol. 1999;10:1361-75. 77. Vaidya, D, Tamaddon, HS, Lo, CW, Taffet, SM, Delmar, M, Morley, GE, Jalife, J. Null mutation of connexin43 causes slow propagation of ventricular activation in the late stages of mouse embryonic development. Circ Res. 2001;88:1196-202. 78. Thomas, SP, Kucera, JP, Bircher-Lehmann, L, Rudy, Y, Saffitz, JE, Kleber, AG. Impulse propagation in synthetic strands of neonatal cardiac myocytes with genetically reduced levels of connexin43. Circ Res. 2003;92:1209-16. 79. van Rijen, HV, Eckardt, D, Degen, J, Theis, M, Ott, T, Willecke, K, Jongsma, HJ, Opthof, T, de Bakker, JM. Slow conduction and enhanced anisotropy increase the propensity for ventricular tachyarrhythmias in adult mice with 27 induced deletion of connexin43. Circulation. 2004;109:1048-55. 80. Beauchamp, P, Choby, C, Desplantez, T, de Peyer, K, Green, K, Yamada, KA, Weingart, R, Saffitz, JE, Kleber, AG. Electrical propagation in synthetic ventricular myocyte strands from germline connexin43 knockout mice. Circ Res. 2004;95:170-8. 81. Stein, M, van Veen, TA, Remme, CA, Boulaksil, M, Noorman, M, van Stuijvenberg, L, van der Nagel, R, Bezzina, CR, Hauer, RN, de Bakker, JM, van Rijen, HV. Combined reduction of intercellular coupling and membrane excitability differentially affects transverse and longitudinal cardiac conduction. Cardiovasc Res. 2009;83:52-60. 82. Danik, SB, Liu, F, Zhang, J, Suk, HJ, Morley, GE, Fishman, GI, Gutstein, DE. Modulation of cardiac gap junction expression and arrhythmic susceptibility. Circ Res. 2004;95:1035-41. 83. Poelzing, S, Rosenbaum, DS. Altered connexin43 expression produces arrhythmia substrate in heart failure. Am J Physiol Heart Circ Physiol. 2004;287:H1762-70. 84. Akar, FG, Nass, RD, Hahn, S, Cingolani, E, Shah, M, Hesketh, GG, DiSilvestre, D, Tunin, RS, Kass, DA, Tomaselli, GF. Dynamic changes in conduction velocity and gap junction properties during development of pacinginduced heart failure. Am J Physiol Heart Circ Physiol. 2007;293:H1223-30. 85. Cascio, WE, Yang, H, Muller-Borer, BJ, Johnson, TA. Ischemia-induced arrhythmia: the role of connexins, gap junctions, and attendant changes in impulse propagation. J Electrocardiol. 2005;38:55-9. 86. Severs, NJ, Bruce, AF, Dupont, E, Rothery, S. Remodelling of gap junctions and connexin expression in diseased myocardium. Cardiovasc Res. 2008;80:919. 87. Spach, MS, Heidlage, JF, Dolber, PC, Barr, RC. Changes in anisotropic conduction caused by remodeling cell size and the cellular distribution of gap junctions and Na(+) channels. J Electrocardiol. 2001;34 Suppl:69-76. 88. Solan, JL, Lampe, PD. Connexin43 phosphorylation: structural changes and biological effects. Biochem J. 2009;419:261-72. 89. Solan, JL, Lampe, PD. Key connexin 43 phosphorylation events regulate the gap junction life cycle. J Membr Biol. 2007;217:35-41. 90. Procida, K, Jorgensen, L, Schmitt, N, Delmar, M, Taffet, SM, HolsteinRathlou, NH, Nielsen, MS, Braunstein, TH. Phosphorylation of connexin43 on serine 306 regulates electrical coupling. Heart Rhythm. 2009;6:1632-8. 28 91. Axelsen, LN, Stahlhut, M, Mohammed, S, Larsen, BD, Nielsen, MS, HolsteinRathlou, NH, Andersen, S, Jensen, ON, Hennan, JK, Kjolbye, AL. Identification of ischemia-regulated phosphorylation sites in connexin43: A possible target for the antiarrhythmic peptide analogue rotigaptide (ZP123). J Mol Cell Cardiol. 2006;40:790-8. 92. Oxford, EM, Musa, H, Maass, K, Coombs, W, Taffet, SM, Delmar, M. Connexin43 remodeling caused by inhibition of plakophilin-2 expression in cardiac cells. Circ Res. 2007;101:703-11. 93. Geisler, SB, Green, KJ, Isom, LL, Meshinchi, S, Martens, JR, Delmar, M, Russell, MW. Ordered assembly of the adhesive and electrochemical connections within newly formed intercalated disks in primary cultures of adult rat cardiomyocytes. J Biomed Biotechnol. 2010;2010:624719. 94. Smyth, JW, Hong, TT, Gao, D, Vogan, JM, Jensen, BC, Fong, TS, Simpson, PC, Stainier, DY, Chi, NC, Shaw, RM. Limited forward trafficking of connexin 43 reduces cell-cell coupling in stressed human and mouse myocardium. J Clin Invest. 2010;120:266-79. 95. Spach, MS, Heidlage, JF, Dolber, PC, Barr, RC. Electrophysiological effects of remodeling cardiac gap junctions and cell size: experimental and model studies of normal cardiac growth. Circ Res. 2000;86:302-11. 96. Ghaly, HA, Boyle, PM, Vigmond, EJ, Shimoni, Y, Nygren, A. Simulations of reduced conduction reserve in the diabetic rat heart: response to uncoupling and reduced excitability. Ann Biomed Eng. 2010;38:1415-25. 97. McIntyre, H, Fry, CH. Abnormal action potential conduction in isolated human hypertrophied left ventricular myocardium. J Cardiovasc Electrophysiol. 1997;8:887-94. 98. Seidel, T, Salameh, A, Dhein, S. A simulation study of cellular hypertrophy and connexin lateralization in cardiac tissue. Biophys J. 2010 ;99:2821-30. 99. Vetter, C, Zweifel, M, Zuppinger, C, Carrel, T, Martin, D, Haefliger, JA, Delacretaz, E. Connexin 43 expression in human hypertrophied heart due to pressure and volume overload. Physiol Res. 2010;59:35-42. 100. Tranum-Jensen, J, Janse, MJ, Fiolet, WT, Krieger, WJ, D'Alnoncourt, CN, Durrer, D. Tissue osmolality, cell swelling, and reperfusion in acute regional myocardial ischemia in the isolated porcine heart. Circ Res. 1981;49:364-81. 101. De Mello, WC. Cell swelling, impulse conduction, and cardiac arrhythmias in the failing heart. Opposite effects of angiotensin II and angiotensin (1-7) on cell volume regulation. Mol Cell Biochem. 2009; 330:211-7. 102. Plonsey, R. and Barr, R. C. Bioelectricity: A Quantitative Approach. 2000. 29 New York, Kluwer Academic/Plenum Publishers. 103. Sperelakis, N. An electric field mechanism for transmission of excitation between myocardial cells. Circ Res. 2002;91:985-7. 104. Sperelakis, N, McConnell, K. Electric field interactions between closely abutting excitable cells. IEEE Eng Med Biol Mag. 2002;21:77-89. 105. Copene, ED, Keener, JP. Ephaptic coupling of cardiac cells through the junctional electric potential. J Math Biol. 2008;57:265-84. CHAPTER 2 MECHANISMS UNDERLYING INCREASED RIGHT VENTRICULAR CONDUCTION SENSITIVITY TO FLECAINIDE CHALLENGE 31 Introduction Cardiac conduction velocity is determined by various factors including tissue excitability, gap junctional coupling and cell geometry. Decreased tissue excitability secondary to inhibition of the cardiac sodium current (INa) has been associated with slow and unsafe conduction and arrhythmogenesis under various conditions1;2 such as Brugada syndrome (BrS),3-5 acute myocardial ischemia and during treatment with class I anti-arrhythmic drugs. In some cases, BrS has been linked to loss of function mutations in SCN5A,5;6 the gene encoding for the pore-forming alpha subunit of the cardiac sodium channel (Nav1.5). The disease is characterized by electrophysiological manifestations in the right precordial leads and tachyarrhythmias that preferentially originate in the right ventricle (RV).4;7;8 While regional conduction heterogeneities in diseases such as BrS have been previously demonstrated, the mechanisms underlying those heterogeneities remain unknown. Furthermore, clinical unmasking of BrS via sodium channel blockade9 suggests that the underlying mechanism of interventricular depolarization heterogeneities is related to sodium channel availability. Interventricular electrophysiological heterogeneities in the normal myocardium has been well established,10-13 and it has been previously demonstrated that Nav1.5 is reduced in RV compared to the left ventricle (LV) in sheep.14 However, the functional consequences of interventricular Nav1.5 heterogeneities remain unknown. Therefore, we hypothesized that the mechanism of interventricular electrophysiological heterogeneity during loss of 32 sodium channel function is caused by heterogeneous distribution of cardiac sodium channels leading to preferential conduction slowing and decreased depolarization reserve in RV. In this study, we demonstrate that the RV exhibits greater conduction dependence on sodium channel blockade as a result of reduced Nav1.5 expression in RV compared to LV in guinea pig. Importantly, we also demonstrate that under normal conditions, IK1 heterogeneities mask Nav1.5 heterogeneities during conduction. Methods The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). Western immunoblotting Guinea pigs were anesthetized (30 mg/kg pentobarbital sodium [Nembutal] IP), and their hearts were rapidly excised. The ventricles were divided into 12 parts - 4 each (anterior base, posterior base, anterior apex and posterior apex) from the left and right ventricles and the septum. The samples were then snap-frozen, cut into small pieces and homogenized by sonication into a whole cell homogenate. Western immunoblotting was performed following a previously described procedure.15 Briefly, the protein content in the whole cell homogenates was assessed through a BCA assay and equal amounts of proteins from the 33 samples were electrophoresed on 4-12% Bis-Tris gels (BioRad, Hercules CA). The proteins were then transferred on to a nitrocellulose membrane. After treatment with a 5% casein solution to block nonspecific binding of antibodies, the membrane was treated with primary antibody (Rabbit Anti-Nav1.5 (n=4) / AntiKir2.1 polyclonal antibody (n=3) (Chemicon, Temecula CA) / goat polyclonal GAPDH antibody (abcam, Cambridge MA)) followed by goat anti-rabbit HRP conjugated secondary antibody (JacksonImmuno, West Grove PA). The membrane was then treated with enzymatic chemiluminescence (ECL) reagents and the bands were visualized on autoradiography film. Protein expression in the samples was quantified based on the size and density of the bands. Nav1.5 and Kir2.1 expressions were normalized to the regional GAPDH expression for interanimal comparison. Optical mapping Guinea pigs were anesthetized (30 mg/kg pentobarbital sodium [Nembutal] IP). Their hearts were rapidly excised and perfused as Langendorff preparations (perfusion pressure, 55 mm Hg) with oxygenated (100% O2) Tyrode's solution at 36.5 °C containing (mMol/l) CaCl2 2, NaCl 140, KCl 4.5, dextrose 5, MgCl2 0.7, HEPES 10 (pH 7.41). The right and left atria were excised to avoid competitive stimulation from the atria. Hearts were stained with the voltage-sensitive dye di-4-ANEPPS (15 µmol/L) by direct coronary perfusion for 10 minutes. 34 Conduction velocity (θ) was quantified by a previously described optical voltage mapping system.16 Specifically, we used two SciMedia MiCam02 HS CCD cameras (SciMedia, Irvine CA) in a tandem lens configuration capable of resolving membrane potential changes as small as 2 mV with 1 ms temporal resolution from 90 x 60 sites simultaneously. Following staining with the voltagesensitive dye di-4-ANEPPS, the preparation was excited by three- 60 LED light sources (RL5-A9018, Superbrightleds, St. Louis, MO) fitted with 510 ± 5 nm filters (Chroma, Rockingham, VT) and a 50 mm aspheric lens (Edmund Optics, Barrington, NJ). Fluoresced light passed through a 150 mm achromatic (BK7/Flint, Ealing, Rocklin, CA) lens, a 50 mm aspheric B270 crown glass lens (Edmund Optics, Barrington, NJ), a 35 mm planoconvex BK7 lens (Edmund Optics, Barrington, NJ), and a 610 nm LP filter (Newport, Irvine, CA) before it was incident on the CCD array. The interpixel resolution was 0.184 mm in the xdirection (90 pixels) and 0.199 mm in the y-direction (60 pixels). Optical action potential measurements Motion was reduced by perfusion of 7.5 mM 2,3-butanedione monoxime (BDM). Hearts were stimulated with a unipolar silver wire placed on the anterior epicardial surface close to the equatorial plane of the ventricle being mapped at 1.5 times the stimulation threshold with a basic cycle length (BCL) of 300 ms unless otherwise specified. Activation time was defined as the time of the maximum first derivative of the action potential as described previously.17 35 Conduction velocity measurements Activation times were fitted to a parabolic surface as previously described.18 The gradient at each point was assigned a conduction velocity vector. The averaged conduction velocity vectors in the slow and fast axis of propagation (±15°) are reported as they reflect transverse and longitudinal propagation.19 Interventions The class IC sodium channel blocker flecainide was varied from 0 to 2 µM to determine conduction dependence on sodium channel availability. A 10µM dose of BaCl2 was used to test the effect of partial IK1 blockade on conduction velocity (θ). To study the effects of hypokalemia, we reduced the concentration of extracellular potassium in the perfusate from 4.5 to 3mM. Statistical analysis Statistical analysis of the data was performed using a 2-tailed Student's ttest for paired and unpaired data or a single factor ANOVA. A p<0.05 was considered statistically significant. All values are reported as mean ± standard error unless otherwise noted. 36 Results Regional heterogeneities of Nav1.5 Nav1.5 expression was quantified using Western immunoblotting in homogenized tissue samples (n=4). Figure 2.1A shows representative bands corresponding to Nav1.5 protein expression from 12 regions of the heart. Visually, bands from the apical and basal regions of the anterior LV and the basal region of the posterior LV are denser than bands in the RV and septum, indicating greater Nav1.5 expression in these regions. Bands from the RV and septum appear to be similar, suggesting that the Nav1.5 expression in these regions may also be comparable. Figure 2.1B summarizes data from the 4 hearts tested, where the radii of the black edged circles represent normalized Nav1.5 expression and the dashed circles represent the standard error. Overall, there were no significant differences in Nav1.5 expression between the RV and the septum. Nav1.5 expression was significantly greater in the LV compared to RV by 18.2%. In particular, anterior RV basal myocardium expressed significantly less Nav1.5 (18.8%) than the anterior basal region of the LV. Lastly, Nav1.5 expression in the posterior RV base was significantly higher than RV anterior base by 10.5%. Regional conduction velocity heterogeneities in the normal heart Since protein quantity does not necessarily correlate to protein function, we measured conduction velocity (θ) in both ventricles precisely in regions with the lowest and greatest Nav1.5 expression in order to determine the effect of 37 Figure 2.1: Regional Nav1.5 expression. A) Representative Nav1.5 bands from the 12 different regions of the heart that were tested. B) Normalized Nav1.5 expression in the ventricles and the septum. The expression in each region is denoted by the radius and grayscale of the corresponding circle. Nav1.5 expression in the LV was 18.2% greater than that in the RV (*, p<0.05) while no significant difference was observed between the RV and the septum (p=ns). Also the anterior basal RV expressed less Nav1.5 than the posterior basal RV (**, p<0.05) and the anterior basal LV (‡, p<0.05). RV = Right Ventricle, LV = Left Ventricle, S = Septum, A = Anterior, P = Posterior. 38 interventricular Nav1.5 expression heterogeneities. Conduction was quantified during pacing from the anterior epicardial surface of either the RV or LV. Upstrokes of epicardial action potentials from equally spaced sites along the longitudinal direction demonstrate uniform conduction velocity as evidenced by equal spacing between upstrokes in both the left and right ventricle (Figure 2.2A). However, faster RV transverse conduction relative to LV is evidenced by reduced temporal spacing of action potential upstrokes in the RV. Representative isochrone maps of activation from the two ventricles demonstrate elliptical spread of activation, consistent with the anisotropic spread of activation characteristic of cardiac tissue (Figure 2.2B). Also, the isochrone lines are more closely spaced perpendicular to the fiber direction, which indicates slower impulse propagation in the transverse direction. The latest time of activation in the RV, as illustrated in Figure 2.2B, is earlier than that in the LV, and transverse isochrone spacing in the RV is greater than the LV, suggesting transverse RV θ is greater than that in the LV. For all experiments over all cycle lengths tested, transverse conduction velocity (θT) was significantly greater in the RV compared to the LV (n=4, Figures 2.2A, 2.2B and 2.3A) by 33.09±1.38%. However, longitudinal conduction velocity (θL) was similar between the RV and LV (Figures 2.2A and 2.3A) over all cycle lengths tested. During control perfusion, we observed no change in θ over 60 minutes (data not shown). All other experiments were conducted within this time frame. The anisotropic ratios for all experiments are summarized in Figure 2.3B. Anisotropy was significantly lower in the RV than in the LV at all cycle lengths 39 Figure 2.2: Conduction velocity: representative data. A) Action potentials from 3 equally spaced sites along both longitudinal and transverse directions of propagation are shown to demonstrate conduction delay. Representative isochrone maps of activation in the right and left ventricles. B) At baseline, RV θT is greater than LV θT. C) Partial IK1 blockade by BaCl2 significantly increases θ in both ventricles. D) INa blockade by 0.5µM flecainide decreases conduction velocity in both ventricles. RV θT remains significantly greater than LV θT. 40 Figure 2.3: Conduction velocity: summary data. A) θL and θT in the right and left ventricles during control conditions for different pacing cycle lengths. RV θT was significantly greater than LV θT for all cycle lengths (*, p<0.05). B) Anisotropic ratio did not change significantly in either ventricle over the range of cycle lengths tested relative to baseline values (p=ns). Interestingly, the RV had a significantly lower anisotropic ratio than the LV at all cycle lengths tested (*, p<0.05). 41 (n=4), which is consistent with the more elliptical pattern of propagation observed in the LV (Figure 2.2B). Since there were no measurable differences in θL between the RV and LV at baseline and the anisotropic ratio did not change with any experimental intervention, θT was chosen as the parameter for assessing conduction heterogeneities. Importantly, greater RV θT during control conditions is inconsistent with the lower Nav1.5 expression observed in the RV. These data suggest a possible role for an ion channel other than Nav1.5, also heterogeneously distributed between the ventricles, in determining θ heterogeneities under control conditions. The effect of IK1 heterogeneities on conduction Representative Western immunoblots in Figure 2.4A indicate reduced density of expression of Kir2.1, the principal ventricular pore forming subunit of the inward-rectifying potassium channel,20 in the RV relative to the LV. GAPDH expression was not significantly different between the RV and LV as illustrated in representative bands (Figure 2.4A). Therefore GAPDH was used as an additional control. For all experiments, Kir2.1 protein density was lower by 12.0±1.5% (n=3) (Figure 2.4B) in the RV compared to LV. Further, it has been previously shown that peak IK1 current in the guinea pig RV is lower than in the LV. 21 In order to determine the influence of IK1 on conduction, 10 µM BaCl222 was perfused. Representative isochrone maps of activation in the presence of BaCl2 are presented in Figure 2.2C. The latest time of activation is decreased in both 42 Figure 2.4: Regional Kir2.1 expression. A) A representative Western blot showing Kir2.1 expression in the RV and LV is shown along with corresponding GAPDH control bands. B) Normalized Kir2.1 expression in the RV (white bar) was 12.0±1.5% (*,p<0.05) lower relative to the LV (black bar). 43 ventricles relative to control while the spacing between isochrones is increased, demonstrating an increase in θ. Over all experiments, BaCl2 increased θT from 26.1±3.3 to 32.0±2.4 cm/s in the RV and from 21.9±2.2 to 27.5±2.9 cm/s in the LV (n=4, Figure 2.5). Effect of INa availability on conduction INa was reduced by perfusing flecainide (0.5µM), a relatively specific blocker of the cardiac sodium channel (Nav1.5). Representative activation isochrone maps in the presence of flecainide (Figure 2.2D) demonstrate that the latest time of activation is increased and the spacing between isochrones decreased in both ventricles relative to control indicating conduction slowing. Overall, flecainide reduced θT from 26.1±3.3 to 16.9±1.4 cm/s in the RV and from 21.9±2.2 to 15.5±1.9 cm/s in the LV (n=4, Figure 2.5). While flecainide (0.5µM) caused conduction slowing in both ventricles, the fractional decrease in θT was greater in the RV than in the LV. In order to assess interventricular heterogeneities in conduction dependence on sodium channel availability, different concentrations of flecainide (0.25, 0.50, 1.00, 1.50 and 2.00 µM) were perfused. A representative plot of θT versus the concentrations of flecainide is presented in Figure 2.6A. In both ventricles, θT decreased with increasing concentrations of flecainide until eventually pacing was no longer possible. The decrease of θT with increasing flecainide concentrations was fit to a straight line (R2 = 0.91 ± 0.03 for all experiments). Conduction dependence on sodium channel availability was 44 Figure 2.5: Conduction velocity: effects of IK1, INa blockade. Blockade of IK1 by BaCl2 caused a significant increase in θT in both ventricles relative to control (†, p<0.05). RV θT was significantly greater than LV θT both in the presence and absence of BaCl2 (*, p<0.05) (averaged over both flecainide and BaCl2 groups). Flecainide (0.5 µM) caused θT to significantly decrease in both ventricles (†, p<0.05). However, BaCl2 did not have a significant effect on θT in the presence of 0.5 µM flecainide (p=ns). The difference between RV and LV θT did not attain statistical significance in the presence of flecainide. 45 Figure 2.6: Conduction dependence on INa blockade. Representative plots of θT versus flecainide concentration are shown in panels A (control conditions), B (BaCl2) and C (hypokalemia). In all cases, θT decreased in both ventricles with increasing flecainide concentration. The slopes (S) of linear fits are indicated in cm/s/µM flecainide. D) Conduction dependence on sodium channel availability. Under control conditions, the slope was significantly greater in the RV compared to LV (*, p<0.05). During IK1 blockade, the slope was significantly greater in the LV relative to LV control (†, p<0.05) but remained unchanged in the RV (p=ns). While the difference in slope during BaCl2 between RV and LV was significantly reduced from control, the RV conduction dependence on sodium channel availability remained significantly different from LV (*, p<0.05). During hypokalemia, the slope was significantly reduced in both ventricles in comparison to control, (†, p<0.05) but the RV slope still remained significantly higher than the LV slope (*, p<0.05). 46 approximated by the absolute slope of the linear fit of θT versus flecainide concentrations in each ventricle. Importantly, the dependence of θT on sodium channel blockade was significantly greater in the RV than in the LV (n=3, Figure 2.6D). Effect of IK1 blockade on conduction While IK1 blockade by BaCl2 caused a significant increase in θT in the absence of flecainide, it did not significantly increase θT in the presence of flecainide (n=3, Figure 2.5). This suggests that IK1 heterogeneities may be a significant determinant of conduction heterogeneities under control conditions. However, the role of IK1 may be diminished when sodium channel availability is reduced. Therefore, we studied the role partial IK1 blockade plays in modulating conduction dependence on sodium channel availability. θT was measured in the two ventricles while varying flecainide concentration in the presence of 10µM BaCl2. A representative plot of θT versus flecainide concentrations in the presence of BaCl2 is presented in Figure 2.6B. As demonstrated previously, θT was greater in the RV relative to the LV before flecainide was perfused, and θT decreased in both ventricles with increasing concentrations of flecainide. Interestingly the dependence of θT on flecainide increased significantly in the LV but did not change in the RV when compared to control. However, the dependence of θT on flecainide remained significantly greater in the RV than in the LV, although the difference was reduced in magnitude when compared to control conditions. For all experiments, BaCl2 significantly increased the slope of 47 θT versus flecainide concentration in the LV (n=3) but did not significantly change conduction dependence on sodium channel availability in the RV (Figure 2.6D). Again, the difference of conduction dependence on sodium channel availability between the ventricles was significantly reduced by BaCl2 when compared to control conditions. These data suggest that IK1 heterogeneities underlie interventricular conduction heterogeneities under control conditions, but the interventricular heterogeneity in Nav1.5 expression becomes functionally relevant under challenging conditions where INa is functionally reduced. Effect of hypokalemia on conduction Changes in extracellular potassium concentration ([K+]o) affect IK1 as well as the resting membrane potential and thus have an important effect on conduction. Hypokalemia (3mM [K+]o) by itself significantly reduced θT in both ventricles (n=3). When varying concentrations of flecainide were applied during hypokalemia, θT decreased in both ventricles (n=3) as seen from the representative plot in Figure 2.6C. Interestingly, the dependence of θT on flecainide decreased in both ventricles during hypokalemia but RV dependence remained significantly greater than LV. (n=3, Figure 2.6D). Discussion Conduction slowing secondary to decreased INa has been well established as a pro-arrhythmic factor;1;2 however the precise role of conduction 48 heterogeneities in arrhythmogenesis remains incompletely understood. The aim of the study was to identify ion channel heterogeneities related to phase 0 depolarization and to determine their effects on conduction. While the role of the transient-outward potassium current (Ito) in arrhythmogenesis with reference to increased repolarization gradients has been well established,23 the Ito peak occurs about 20ms after initial depolarization.24;25 It is therefore unlikely that Ito plays a significant role in determining heterogeneities related to phase 0 depolarization. Therefore, guinea pig, which does not functionally express Ito,26 was chosen as the animal model for this study. We hypothesized that the mechanism of interventricular electrophysiological heterogeneity during loss of sodium channel function is based on heterogeneous distribution of cardiac sodium channels leading to preferential conduction slowing and decreased depolarization reserve in RV. Indeed, Nav1.5 expression in the RV is reduced relative to the LV, consistent with previous findings in sheep.14 While INa availability is a well established determinant of conduction velocity,2 the observed Nav1.5 expression heterogeneity does not correlate with conduction patterns. Heterogeneous conduction in the normal myocardium Baseline values for θL and θT are consistent with previously reported results in guinea pig.27 In particular, RV θT is significantly greater than LV θT which is also in agreement with previous results.28 While the effects of ion channel heterogeneities on θ are not expected to be directionally dependent, θL 49 was not significantly different between the ventricles. One possible explanation is that the large variation of θL (as measured by the standard deviation) may mask conduction heterogeneities. θT, on the other hand, had a smaller variability, allowing relatively more precise measurement of changes in θ. No intervention changed the anisotropic ratio, consistent with the idea that ion channel heterogeneities do not affect θ in a directional dependent manner.29 Therefore, the significantly different RV and LV conduction anisotropy may be due to greater measurement sensitivity of θT than θL. Therefore, θT was used as the more sensitive measure of conduction changes for this study. IK1 heterogeneities underlie conduction heterogeneities in the normal heart As mentioned previously, interventricular θT heterogeneity does not agree with Nav1.5 expression patterns. We hypothesized that this apparently paradoxical behavior may result from heterogeneities in IK1. The magnitude of the IK1 current is small in comparison to that of the sodium current, but IK1 opposes depolarization particularly before sodium channels fully activate. Therefore, we hypothesized that reducing IK1 opposition to INa mediated depolarization may increase θ. A smaller peak IK1 current and lower Kir2.1 RNA levels in the RV than in the LV have been previously reported in guinea pig,21 which are consistent with reduced RV Kir2.1 protein expression levels reported in this study. However, this is inconsistent with other studies that have been unable to demonstrate 50 difference in Kir2.1 RNA levels between ventricles.30 The discrepancy may be due to difference between animals, experimental technique, or that RNA quantity does not necessarily correlate to protein expression. Regions with reduced Kir2.1 demonstrated the greatest θT. Furthermore, when IK1 was blocked with BaCl2, θT significantly increased. Importantly, LV θT during BaCl2 increased to similar values as RV θT under control conditions. This suggests that the faster conduction in the RV under control conditions may in part be explained on the basis of the reduced IK1 in the RV relative to the LV. Effect of INa availability on conduction: normal vs. disease states While IK1 heterogeneities may explain baseline θ heterogeneities, they do not fully explain ventricular specific responses to sodium channel blockade. Therefore, one would expect that interventricular differences in conduction dependence on sodium channel availability reflect regional Nav1.5 heterogeneities. Flecainide significantly decreased θT in both ventricles,31 but the fractional decrease in θT was greater in the RV relative to the LV. In order to determine if there was any difference between the ventricles in conduction dependence on sodium channel availability, varying concentrations of flecainide were applied. Shaw and Rudy2 previously demonstrated in a mathematical model that θ and sodium conductance (gNa) were curvilinearly related. Our data were well fit by a line and the absolute value of the slope of this line was used as a measure of conduction dependence on sodium channel availability. The absence of the curvilinear relationship in this study may be due to the use of flecainide 51 concentration steps that were not sufficiently fine, particularly near the point of conduction failure. RV θT exhibited significantly greater dependence on flecainide concentration than LV θT, and the results were independent of the order of flecainide doses. This is consistent with the reduced expression of Nav1.5 in the RV relative to the LV. These results suggest that interventricular IK1 heterogeneities mask Nav1.5 heterogeneities under control conditions but fail to do so during sodium channel blockade. IK1 blockade unmasks effect of Nav1.5 heterogeneities on conduction Blockade of IK1 by 10µM BaCl2, which caused θT to increase significantly under control conditions, did not cause a significant change in θT in either ventricle in the presence of 0.5µM flecainide. This indicates that the dependence of interventricular conduction heterogeneities shifts from IK1 heterogeneities under control conditions to INa heterogeneities when INa is reduced. Therefore, the roles played by INa and IK1 in conduction are likely synergistic, not summative. In order to test this hypothesis, we applied different concentrations of flecainide to the heart in the presence of BaCl2. While flecainide decreased θT in both ventricles, IK1 blockade attenuated the chamber-specific differences in conduction dependence on sodium channel availability. This further supports the hypothesis that IK1 heterogeneities can compensate for INa heterogeneities during impulse propagation under normal conditions. 52 Interestingly, IK1 blockade increased θT dependence on flecainide in the LV but did not cause a significant change in the RV. This suggests that the mechanism underlying the interventricular heterogeneity in conduction dependence on sodium channel availability may be two-fold. While the larger IK1 current in the LV depresses control LV θ, the greater Nav1.5 expression may act to lower the dependence of LV conduction on sodium channel blockade. These effects may act synergistically to attenuate dependence of LV θT on sodium channel blockade. Since the LV expresses more IK1 and INa than the RV, further reducing IK1 unmasks LV conduction dependence on sodium channel availability, suggesting a greater "depolarization reserve" in the LV. The RV, on the other hand, may function with a reduced depolarization reserve. Therefore, additional IK1 reductions do not significantly affect conduction dependence on sodium channel availability when INa is functionally reduced. In other words, conduction dependence on sodium channel availability under conditions of reduced IK1 most likely reflects Nav1.5 expression heterogeneities. Synergistic conduction slowing by hypokalemia and sodium channel blockade Apart from specific blockade by compounds such as BaCl2, IK1 is also affected by changes in extracellular potassium concentration ([K+]o). However, changes in [K+]o also affect the resting membrane potential and consequently the availability of sodium channels. Thus, changes in [K+]o have a significant bearing 53 on conduction.32 Further, some anecdotal reports have linked hypokalemia to an increased risk of VT/VF in Brugada patients,33;34 suggesting that hypokalemia may exacerbate conduction abnormalities. Therefore, we studied the effect of hypokalemia (3mM [K+]o) on the dependence of conduction on sodium availability. Hypokalemia significantly decreased θT in both ventricles and caused a significant decrease in the dependence of θT on flecainide in both ventricles. While the depression of conduction dependence on sodium channel availability may appear anti-arrhythmic, this depression may be insufficient to counteract the proarrhythmic effects of conduction slowing. Therefore, hypokalemia in the RV may result in reduced safety of conduction at lower concentrations of flecainide when compared to control. Conclusions Interventricular IK1 heterogeneities underlie the conduction heterogeneities observed under control conditions. However, under conditions where the sodium current is reduced, the heterogeneity in sodium channel expression plays a greater role in determining conduction heterogeneities. Importantly, INa heterogeneities could exacerbate regional susceptibility to arrhythmias in disease states where INa may be functionally reduced. Limitations Although BaCl2 is a potent IK1 blocker, it has some cross-reactivity with the L-type calcium channel.35 However, the 10µmol/l used in this study is orders of 54 magnitude smaller than concentrations reported to significantly affect L-type calcium channels.36 Similarly, while flecainide is relatively specific in blocking INa, it has been shown to inhibit Ito and a maintained outward potassium current (IK) in rat.37 Guinea pigs do not functionally express Ito and IK block is significant at higher flecainide concentrations than used in this study. BDM has been shown to cause no significant change in conduction velocity in the guinea pig heart.38 In other animals, BDM is associated with a small depression of conduction velocity at doses significantly greater than the one used in this study.39;40 Therefore, it is unlikely that BDM significantly affected the principal findings of this study. This study does not address the effects of heterogeneities in cell size,41 cell geometry, wall thickness or connexin distribution on conduction. While these effects may be significant, particularly with respect to differences in conduction anisotropy between the ventricles, they are unlikely to affect the principle finding of this study that interventricular conduction differences are in part dependent on IK1 and INa heterogeneities. 55 References 1. Shaw, RM, Rudy, Y. The vulnerable window for unidirectional block in cardiac tissue: characterization and dependence on membrane excitability and intercellular coupling. J Cardiovasc Electrophysiol. 1995;6:115-31. 2. Shaw, RM, Rudy, Y. Ionic mechanisms of propagation in cardiac tissue. Roles of the sodium and L-type calcium currents during reduced excitability and decreased gap junction coupling. Circ Res. 1997;81:727-41. 3. Yan, GX, Antzelevitch, C. Cellular basis for the electrocardiographic J wave. Circulation. 1996;93:372-9. 4. Tukkie, R, Sogaard, P, Vleugels, J, de Groot, IK, Wilde, AA, Tan, HL. Delay in right ventricular activation contributes to Brugada syndrome. Circulation. 2004;109:1272-7. 5. Chen, Q, Kirsch, GE, Zhang, D, Brugada, R, Brugada, J, Brugada, P, Potenza, D, Moya, A, Borggrefe, M, Breithardt, G, Ortiz-Lopez, R, Wang, Z, Antzelevitch, C, O'Brien, RE, Schulze-Bahr, E, Keating, MT, Towbin, JA, Wang, Q. Genetic basis and molecular mechanism for idiopathic ventricular fibrillation. Nature. 1998;392:293-6. 6. Coronel, R, Berecki, G, Opthof, T. Why the Brugada syndrome is not yet a disease: syndromes, diseases, and genetic causality. Cardiovasc Res. 2006;72:361-3. 7. Chinushi, M, Washizuka, T, Chinushi, Y, Higuchi, K, Toida, T, Aizawa, Y. Induction of ventricular fibrillation in Brugada syndrome by site-specific right ventricular premature depolarization. Pacing Clin Electrophysiol. 2002;25:1649-51. 8. Kasanuki, H, Ohnishi, S, Ohtuka, M, Matsuda, N, Nirei, T, Isogai, R, Shoda, M, Toyoshima, Y, Hosoda, S. Idiopathic ventricular fibrillation induced with vagal activity in patients without obvious heart disease. Circulation. 1997;95:2277-85. 9. Brugada, R, Brugada, J, Antzelevitch, C, Kirsch, GE, Potenza, D, Towbin, JA, Brugada, P. Sodium channel blockers identify risk for sudden death in patients with ST-segment elevation and right bundle branch block but structurally normal hearts. Circulation. 2000;101:510-5. 10. Laurita, KR, Girouard, SD, Rosenbaum, DS. Modulation of ventricular repolarization by a premature stimulus. Role of epicardial dispersion of repolarization kinetics demonstrated by optical mapping of the intact guinea pig heart. Circ Res. 1996;79:493-503. 11. Ghanem, RN, Burnes, JE, Waldo, AL, Rudy, Y. Imaging dispersion of 56 myocardial repolarization, II: noninvasive reconstruction of epicardial measures. Circulation. 2001;104:1306-12. 12. Volders, PG, Sipido, KR, Carmeliet, E, Spatjens, RL, Wellens, HJ, Vos, MA. Repolarizing K+ currents ITO1 and IKs are larger in right than left canine ventricular midmyocardium. Circulation. 1999;99:206-10. 13. Di Diego, JM, Sun, ZQ, Antzelevitch, C. I(to) and action potential notch are smaller in left vs. right canine ventricular epicardium. Am J Physiol. 1996;271: H548-61. 14. Fahmi, AI, Patel, M, Stevens, EB, Fowden, AL, John, JE 3rd, Lee, K, Pinnock, R, Morgan, K, Jackson, AP, Vandenberg, JI. The sodium channel betasubunit SCN3b modulates the kinetics of SCN5a and is expressed heterogeneously in sheep heart. J Physiol. 2001;537:693-700. 15. Poelzing, S, Akar, FG, Baron, E, Rosenbaum, DS. Heterogeneous connexin43 expression produces electrophysiological heterogeneities across ventricular wall. Am J Physiol Heart Circ Physiol. 2004;286:H2001-9. 16. Poelzing, S, Veeraraghavan, R. Heterogeneous ventricular chamber response to hypokalemia and inward rectifier potassium channel blockade underlies bifurcated T wave in guinea pig. Am J Physiol Heart Circ Physiol. 2007;292:H3043-51. 17. Girouard, SD, Laurita, KR, Rosenbaum, DS. Unique properties of cardiac action potentials recorded with voltage-sensitive dyes. J Cardiovasc Electrophysiol. 1996;7:1024-38. 18. Bayly, PV, KenKnight, BH, Rogers, JM, Hillsley, RE, Ideker, RE, Smith, WM. Estimation of conduction velocity vector fields from epicardial mapping data. IEEE Trans Biomed Eng. 1998;45:563-71. 19. Girouard, SD, Pastore, JM, Laurita, KR, Gregory, KW, Rosenbaum, DS. Optical mapping in a new guinea pig model of ventricular tachycardia reveals mechanisms for multiple wavelengths in a single reentrant circuit. Circulation. 1996;93:603-13. 20. Seemann, G, Sachse, FB, Weiss, DL, Ptacek, LJ, Tristani-Firouzi, M. Modeling of IK1 mutations in human left ventricular myocytes and tissue. Am J Physiol Heart Circ Physiol. 2007;292:H549-59. 21. Warren, M, Guha, PK, Berenfeld, O, Zaitsev, A, Anumonwo, JM, Dhamoon, AS, Bagwe, S, Taffet, SM, Jalife, J. Blockade of the inward rectifying potassium current terminates ventricular fibrillation in the guinea pig heart. J Cardiovasc Electrophysiol. 2003;14:621-31. 22. DiFrancesco, D, Ferroni, A, Visentin, S. Barium-induced blockade of the 57 inward rectifier in calf Purkinje fibres. Pflugers Arch. 1984 ;402:446-53. 23. Antzelevitch, C. Late potentials and the Brugada syndrome. J Am Coll Cardiol. 2002;39:1996-9. 24. Sun, X, Wang, HS. Role of the transient outward current (Ito) in shaping canine ventricular action potential--a dynamic clamp study. J Physiol. 2005;564:411-9. 25. Litovsky, SH, Antzelevitch, C. Transient outward current prominent in canine ventricular epicardium but not endocardium. Circ Res. 1988;62:116-26. 26. Findlay, I. Is there an A-type K+ current in guinea pig ventricular myocytes? Am J Physiol Heart Circ Physiol. 2003;284:H598-604. 27. Laurita, KR, Girouard, SD, Akar, FG, Rosenbaum, DS. Modulated dispersion explains changes in arrhythmia vulnerability during premature stimulation of the heart. Circulation. 1998;98:2774-80. 28. van Veen, TA, Stein, M, Royer, A, Le Quang, K, Charpentier, F, Colledge, WH, Huang, CL, Wilders, R, Grace, AA, Escande, D, de Bakker, JM, van Rijen, HV. Impaired impulse propagation in Scn5a-knockout mice: combined contribution of excitability, connexin expression, and tissue architecture in relation to aging. Circulation. 2005;112:1927-35. 29. Coromilas, J, Saltman, AE, Waldecker, B, Dillon, SM, Wit, AL. Electrophysiological effects of flecainide on anisotropic conduction and reentry in infarcted canine hearts. Circulation. 1995;91:2245-63. 30. Gaborit, N, Le Bouter, S, Szuts, V, Varro, A, Escande, D, Nattel, S, Demolombe, S. Regional and tissue specific transcript signatures of ion channel genes in the non-diseased human heart. J Physiol. 2007;582:675-93. 31. Brugada, J, Boersma, L, Kirchhof, C, Allessie, M. Proarrhythmic effects of flecainide. Experimental evidence for increased susceptibility to reentrant arrhythmias. Circulation. 1991;84:1808-18. 32. Buchanan, JW Jr, Saito, T, Gettes, LS. The effects of antiarrhythmic drugs, stimulation frequency, and potassium-induced resting membrane potential changes on conduction velocity and dV/dtmax in guinea pig myocardium. Circ Res. 1985;56:696-703. 33. Araki, T, Konno, T, Itoh, H, Ino, H, Shimizu, M. Brugada syndrome with ventricular tachycardia and fibrillation related to hypokalemia. Circ J. 2003;67:93-5. 34. Notarstefano, P, Pratola, C, Toselli, T, Ferrari, R. Atrial fibrillation and recurrent ventricular fibrillation during hypokalemia in Brugada syndrome. 58 Pacing Clin Electrophysiol. 2005;28:1350-3. 35. Ferreira, G, Yi, J, Rios, E, Shirokov, R. Ion-dependent inactivation of barium current through L-type calcium channels. J Gen Physiol. 1997;109:449-61. 36. Buljubasic, N, Rusch, NJ, Marijic, J, Kampine, JP, Bosnjak, ZJ. Effects of halothane and isoflurane on calcium and potassium channel currents in canine coronary arterial cells. Anesthesiology. 1992;76:990-8. 37. Slawsky, MT, Castle, NA. K+ channel blocking actions of flecainide compared with those of propafenone and quinidine in adult rat ventricular myocytes. J Pharmacol Exp Ther. 1994;269:66-74. 38. Liu, Y, Cabo, C, Salomonsz, R, Delmar, M, Davidenko, J, Jalife, J. Effects of diacetyl monoxime on the electrical properties of sheep and guinea pig ventricular muscle. Cardiovasc Res. 1993;27:1991-7. 39. Kettlewell, S, Walker, NL, Cobbe, SM, Burton, FL, Smith, GL. The electrophysiological and mechanical effects of 2,3-butane-dione monoxime and cytochalasin-D in the Langendorff perfused rabbit heart. Exp Physiol. 2004;89:163-72. 40. Baker, LC, Wolk, R, Choi, BR, Watkins, S, Plan, P, Shah, A, Salama, G. Effects of mechanical uncouplers, diacetyl monoxime, and cytochalasin-D on the electrophysiology of perfused mouse hearts. Am J Physiol Heart Circ Physiol. 2004;287:H1771-9. 41. Campbell, SE, Gerdes, AM, Smith, TD. Comparison of regional differences in cardiac myocyte dimensions in rats, hamsters, and guinea pigs. Anat Rec. 1987;219:53-9. CHAPTER 3 POTASSIUM CHANNEL ACTIVATORS DIFFERENTIALLY MODULATE THE EFFECT OF SODIUM CHANNEL BLOCKADE ON CARDIAC CONDUCTION 60 Introduction Loss of potassium channel function leads to a compromised repolarization reserve and contributes to the arrhythmogenic substrate in several pathophysiological states including the long QT syndromes (LQTS).1;2 The cause of reduced K+ current can vary between genetic mutations in ion channel proteins (inherited LQTS), pathophysiological remodeling, and K+ channel inhibition by drugs (acquired LQTS); however, it consistently leads to prolongation of action potential duration (APD) and increased dispersion of repolarization, which have been linked with potentially lethal cardiac arrhythmias.1;2 Recent studies have suggested augmenting the repolarization reserve using pharmacological K+ channel activators as a potential therapy for LQTS, primarily on the basis of experimental evidence that these drugs mitigate action potential duration (APD) prolongation and dispersion of repolarization.3-5 However, their effects on cardiac conduction remain incompletely understood. We previously demonstrated that pharmacological modulation of the inward rectifier K+ current (IK1)6 as well as the rapid delayed rectifier K+ current (IKr)7 can affect cardiac conduction in a cardiac sodium current (INa)-dependent manner. Further, modulation of IK1 only affected conduction when sodium channel availability was not reduced6 whereas modulation of IKr preferentially affected conduction when conduction was already compromised.7 Here we tested the hypothesis that pharmacological activation of the ATP-sensitive K+ current (IKATP), a K+ current carried by non-voltage-gated, inwardly rectifying channels7;8 will affect cardiac conduction and its dependence on sodium channel 61 blockade differently from activation of the slow delayed rectifier K+ current (IKs), a K+ current carried by voltage-gated K+ channels.7;8 We demonstrate that augmentation of both IKATP and IKs slows conduction; however, where IKATP activation only had this effect in the absence of sodium channel blockade, IKs activation slowed conduction regardless of sodium channel blockade. These data suggest the effects of K+ channel modulators on cardiac conduction vis-à-vis its dependence on sodium channel blockade may have important implications for their safety in a therapeutic context. Methods The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). Guinea pig Langendorff preparations Ventricles isolated from adult male guinea pig breeders (800-1000g) were perfused (at approximately 50 mm Hg) in Langendorff configuration with oxygenated Tyrode's solution (containing, in mMol/l, CaCl2 1.25, NaCl 140, KCl 4.5, dextrose 5.5, MgCl2 0.7, HEPES 10; pH 7.4) at 36.5 ± 0.5 °C as previously described.6;9 2,3-butanedione monoxime (7.5 mM; BDM) was added to the perfusate for optical mapping experiments only, and the anterior epicardium was mechanically pressed against the front wall of the perfusion chamber to further stabilize and flatten it. Hearts were stimulated via a unipolar silver electrode 62 placed on the anterior epicardial surface at the center of the mapping field at 1.5 times the stimulation threshold with a basic cycle length (BCL) of 300 ms unless otherwise specified. Electrocardiography (ECG) A volume-conducted bath ECG was obtained using a silver chloride anode located ~2 cm from lateral wall of the RV and a similar cathode located ~2 cm from the lateral wall of the LV. ECGs were recorded at 1 kHz and filtered to remove 60 Hz noise. Optical mapping Optical voltage mapping was performed using di-4-ANEPPS (15 µM) as a voltage indicator to quantify conduction velocity (θ) and anisotropy (ARθ; defined as the ratio of longitudinal θ (θL) to transverse θ (θT)) as previously described. 6;9 Briefly, the preparation was stained with di-4-ANEPPS by direct coronary perfusion for 10 minutes, then excited by three 60 LED light sources (RL5A9018, Superbrightleds, St. Louis, MO) fitted with 510±5 nm filters (Chroma, Rockingham, VT). Fluoresced light was passed through a 610 nm LP filter (Newport, Irvine, CA) before being recorded with a SciMedia MiCam02 HS CCD camera (SciMedia, Irvine CA) in a tandem lens configuration capable of resolving membrane potential changes as small as 2 mV with 1 ms temporal resolution from 90 x 60 sites simultaneously. The interpixel resolution was 0.184 mm in the x-direction (90 pixels) and 0.199 mm in the y-direction (60 pixels). 63 Activation time was defined as the time of the maximum first derivative of the action potential as described previously.10 A parabolic surface was fitted to the activation times as previously described.11 The gradient at each point was assigned a conduction velocity vector. The averaged conduction velocity vectors along the slow and fast axes of propagation (±15°) are reported as they reflect transverse and longitudinal propagation.12 Interventions The class IC sodium channel blocker flecainide was perfused in concentrations ranging from 0 to 1 µM to determine conduction dependence on sodium channel availability. Pinacidil (10 µM; Sigma Aldrich, St. Louis, MO) and R-L3 (10 µM; NeuroSearch A/S, Ballerup, Denmark) were applied to test the effects of increasing IKATP and IKs, respectively, on θ. In the dose response experiments, flecainide was applied in increasing doses, first by itself and then in the presence of a constant concentration of K+ channel activator. Statistical analysis Statistical analysis of the data was performed using a 2-tailed Student's ttest for paired and unpaired data or a single factor ANOVA. The Bonferroni correction was applied to adjust for multiple comparisons. A p<0.05 was considered statistically significant. All values are reported as mean ± standard error unless otherwise noted. 64 Results Controls Activating outward potassium currents is expected to shorten APD, which may alter cardiac conduction by altering sodium channel availability (aka refractoriness). To test this, we globally assessed cardiac conduction (QRS duration) and an index of refractoriness (QT interval). Representative traces of a volume conducted ECG demonstrates that QRS duration was prolonged during activation of the non-voltage-gated channels that carry the ATP-sensitive potassium current (IKATP) by pinacidil relative to control (Figure 3.1A). Interestingly, activation of the voltage-gated channels carrying the slow component of the delayed rectifier potassium current (IKs) by R-L3 did not prolong the QRS (Figure 3.1A). Summary data in Figure 3.1B (white bars) demonstrate that IKATP activation by pinacidil prolonged QRS duration relative to control (*), while IKs activation by R-L3 did not significantly change QRS duration. These data suggest that pinacidil slows cardiac conduction while R-L3 does not under normal conditions. Figure 3.1A also demonstrates that the QT interval, a global metric of action potential duration (APD) is shortened during perfusion of pinacidil and RL3 as expected. Overall, both agents shortened the QT interval relative to control (Figure 3.1C, white bars) despite differential effects on conduction. We observed no change in either QRS duration or QT interval over a period of 60 minutes under control conditions (data not shown); therefore, all other experiments were conducted within this time frame. 65 Figure 3.1: Electrocardiograms: no di-4-ANEPPS. A) Representative bath electrocardiograms recorded during control (top), pinacidil (middle) and R-L3 (bottom) in the absence of di-4-ANEPPS. B) QRS duration. Pinacidil (n=4) but not R-L3 (n=3) prolonged the QRS (*) relative to control (white bars). These measurements were then repeated in the presence of flecainide (black bars). Flecainide by itself prolonged the QRS relative to control and R-L3. Pinacidil + flecainide did not change the QRS relative to flecainide alone, but R-L3 + flecainide did (black bars). C) QT interval. Both pinacidil and R-L3 abbreviated the QT interval (white bars, *). Flecainide (black bars) by itself prolonged the QT interval relative to control (*) and pinacidil + flecainide (†, p<0.05) but not R-L3 + flecainide shortened the QT interval relative to flecainide alone (black bars). In panels B and C: * - p < 0.05 vs. control, † - p < 0.05 vs. flecainide. 66 67 Flecainide To elucidate the mechanism of preferential conduction slowing during pinacidil versus R-L3 perfusion, conduction velocity was first reduced by perfusion of 1 µM flecainide to partially block sodium channels. Flecainide alone significantly prolonged the QRS duration as well as the QT interval relative to control (Figure 3.1B, C, black bars, *). Interestingly, flecainide + pinacidil did not significantly alter QRS duration (Figure 3.1B, black bars), while it shortened the QT interval (Figure 3.1C, black bars) relative to flecainide alone. In contrast, flecainide + R-L3 prolonged the QRS duration relative to flecainide alone (Figure 3.1B, black bars, †). However, flecainide + R-L3 did not shorten the QT interval (Figure 3.1C, black bars) relative to flecainide alone. di-4-ANEPPS Optical mapping was performed with the voltage sensitive dye, di-4ANEPPS in order to directly quantify conduction velocity in the presence of pinacidil or R-L3. However, di-4-ANEPPS was previously demonstrated to prolong the QRS and unmask conduction slowing during IKr inhibition. 7 Indeed, over all experiments, di-4-ANEPPS perfusion prolonged the QRS (21 ± 3%, p < 0.05). Representative ECG traces in Figure 3.2A demonstrate QRS prolongation and QT interval shortening during IKATP activation by pinacidil in the presence of di-4-ANEPPS relative to di-4-ANEPPS alone. Summary data demonstrate that pinacidil prolonged the QRS (Figure 3.2B, white bars) and abbreviated the QT interval (Figure 3.2C, white bars) overall, in the presence of di-4-ANEPPS, as it 68 Figure 3.2: Electrocardiograms: 15 µM di-4-ANEPPS. A) Representative bath electrocardiograms recorded during control (top), pinacidil (middle) and R-L3 (bottom) recorded in the presence of di-4-ANEPPS. B) QRS duration. Both pinacidil (n=5) and R-L3 (n=4) prolonged the QRS (*) relative to control (white bars). With flecainide (black bars) by itself prolonged the QRS relative to control (i.e., di-4-ANEPPS alone). Pinacidil + flecainide did not alter the QRS relative to flecainide alone whereas R-L3 + flecainide prolonged the QRS (†) relative to flecainide alone (black bars). C) QT interval. Both pinacidil + di-4-ANEPPS and R-L3 + di-4-ANEPPS abbreviated the QT interval (*, white bars) relative to control (i.e., di-4-ANEPPS alone). With flecainide (black bars) by itself prolonged the QT interval relative to control (i.e., di-4-ANEPPS alone; *). Pinacidil + flecainide but not R-L3 + flecainide shortened the QT interval (†) relative to flecainide alone (black bars). In panels B and C: * - p < 0.05 vs. control, † - p < 0.05 vs. flecainide. 69 70 did in the absence of di-4-ANEPPS. Activation of IKs by R-L3 in the presence of di-4-ANEPPS had a similar effect, prolonging the QRS (Figures 3.2A, B, white bars) and shortening the QT interval (Figures 3.2A, C, white bars) relative to di-4ANEPPS alone. This is in contrast to the absence of di-4-ANEPPS, when R-L3 did not significantly alter QRS duration (Figures 3.1A, B-white bars). Flecainide, perfused in the presence of di-4-ANEPPS, again significantly prolonged both the QRS duration and the QT interval relative to di-4-ANEPPS alone (Figures 3.2B, C). As before, flecainide + pinacidil did not significantly alter QRS duration (Figure 3.2B, black bars) and shortened the QT interval (Figure 3.2C, black bars) relative to flecainide alone. Lastly, flecainide + R-L3 prolonged the QRS duration (Figure 3.2B, black bars) but did not alter the QT interval (Figure 3.2C, black bars) relative to flecainide alone. Conduction velocity: di-4-ANEPPS Longitudinal (θL) and transverse (θT) conduction velocities were subsequently quantified by optical mapping. Representative action potential (AP) upstrokes recorded during control conditions in Figure 3.3A demonstrate shorter delays between equally spaced sites along the longitudinal axis (top traces) relative to sites along the transverse axis (bottom traces). This produces the elliptical activation pattern characteristic of cardiac tissue seen in Figure 3.3B. During control conditions, we observed no change in θ over a period of 60 minutes (data not shown). All optical mapping experiments were conducted within this time frame. 71 Figure 3.3: Conduction velocity. A) Action potential upstrokes from equally spaced sites labeled A,B,C,D,E, and F in panel B. Representative activation isochrone maps: B) Control, C) Pinacidil, D) R-L3 and E) Flecainide. F) Longitudinal (θL) and transverse (θT) conduction velocity. Overall, pinacidil, R-L3 and flecainide decreased both θL and θT. 72 IKATP activation by pinacidil decreased the spacing between isochrones (Figure 3.3C), suggesting mild conduction slowing. Indeed, over all experiments, pinacidil decreased θL by 9.1 ± 2.6 % and θT by 10.0 ± 0.6 %, respectively (p<0.05; Figure 3.3F). Similarly, IKs activation by R-L3 also decreased isochrone spacing, suggesting slowed conduction (Figure 3.3D). Overall, R-L3 decreased θL by 11.7 ± 2.2 % and θT by 15.0 ± 3.8%, respectively (p<0.05; Figure 3.3F). No intervention altered anisotropy of conduction (ARθ; p=ns). Conduction velocity: di-4-ANEPPS + flecainide As reported above, pinacidil or R-L3 in the presence of flecainide produced diverse conduction changes. Conduction was assessed by optical mapping in the presence of different concentrations of flecainide (0, 0.5, 1 µM) to determine conduction dependence on INa during perfusion of pinacidil or R-L3. By itself, flecainide produced a linear decrease in both θL and θT with increasing dose (Figures 3.4A, 3.5A - dashed lines, R2 ≥ 0.95 for a linear fit). Also, flecainide did not alter ARθ relative to control (n = 9; p = ns). IKATP activation by pinacidil significantly decreased conduction in the absence (0 µM) but not in the presence of either 0.5 or 1 µM flecainide (Figure 3.4A - solid lines). Yet, θL and θT decreased linearly with flecainide dose in the presence of pinacidil (Figure 3.4A - solid lines, R2 ≥ 0.95) and ARθ was not different relative to control or flecainide alone (n = 4, p =ns). Consequently, pinacidil blunted conduction dependence on pharmacological INa inhibition 73 Figure 3.4: Conduction dependence on INa blockade: effect of pinacidil A) Plots of θL and θT versus flecainide concentration in the absence (dashed lines) and presence (solid lines) of pinacidil (n=5). Whereas pinacidil decreased both θL and θT by itself (0 µM flecanide), pinacidil + flecainide did not significantly alter either parameter relative to flecainide alone at any flecainide dose. * - p < 0.05 vs. 0 pinacidil. B) Conduction dependence on sodium channel blockade was quantified as the absolute slope of θ vs. flecainide dose. Pinacidil (blue bars) significantly decreased conduction dependence on sodium channel blockade (*) relative to control (white bars). 74 Figure 3.5: Conduction dependence on INa blockade: effect of R-L3 A) Plots of θL and θT versus flecainide concentration in the absence (dashed lines) and presence (solid lines) of R-L3 (n=4). R-L3 by itself (0 µM flecanide), decreased θL and θT (*). R-L3 + flecainide decreased both θL and θT (*) relative to flecainide alone for both 0.5 and1 µM doses of flecainide was present. * - p < 0.05 vs. 0 RL3. B) Conduction dependence on sodium channel blockade was quantified as the absolute slope of θ vs. flecainide dose. R-L3 (red bars) did not significantly alter conduction dependence on sodium channel blockade relative to control (white bars). 75 quantified as the absolute slope of θ versus flecainide dose (Figure 3.4B, p < 0.05) as demonstrated by the crossover of the two lines in Figure 3.4A. In contrast, R-L3 activation of IKs decreased θL and θT by itself and also in the presence of all flecainide doses (Figure 3.5A - solid lines). Similar to pinacidil, θL and θT decreased linearly with flecainide dose in the presence of RL3 (Figure 3.5A - solid lines, R2 ≥ 0.95) and ARθ was not altered relative tocontrol or flecainide (n = 5, p =ns). Consequently, R-L3 did not alter θL or θT dependence on pharmacological INa inhibition (Figure 3.5B, n=5, p = ns). Discussion It is well established that membrane excitability, in particular the amplitude of the cardiac sodium current (INa) is a key determinant of velocity (θ). It has long been recognized that decreasing INa slows conduction.13 We recently demonstrated that outward K+ currents such as the inward rectifier K+ current (IK1)6 and the rapid delayed rectifier (IKr)7 can modulate θ. Further, we demonstrated that modulation of IK1 and IKr altered conduction under different conditions. Therefore, we explored the hypothesis that non-voltage-gated and voltage-gated potassium channels heterogeneously affect conduction. Here we demonstrate that activation of non-voltage-gated IKATP8;14 slowed conduction by itself but flattened the response of conduction velocity to pharmacological INa inhibition. In contrast, voltage-gated IKs8 activation did not slow conduction by itself, and it did not affect the response of conduction velocity to pharmacological INa activation. 76 Pinacidil: effects on conduction By itself, 10 µM pinacidil, an IKATP agonist, significantly broadened QRS duration. This is in contrast to the study by Yang et al. which did not report a change in QRS duration up to 50 µM pinacidil.15 Important differences between the Yang et al. study and our study include the sites of pacing (atria and ventricle respectively), location of ECG electrodes (on the ventricle and bath, respectively), and temporal resolution of ECG quantification (Polaroid photos of the oscilloscope and 1 kHz digital data acquisition, respectively). Further, the QRS complex during atrial pacing is narrower relative to ventricular pacing, and as a result changes in QRS duration in the Yang et al. study may have been below the detection limit. Therefore, experimental differences likely explain these discrepant findings. Importantly, the mechanism of pinacidil mediated conduction slowing we observed under normal conditions may be due to pinacidil's effect of hyperpolarizing myocytes16 and increasing outward current during early depolarization. Hyperpolarization is a well-established mechanism for conduction slowing,17 where the amount of charge required to reach the activation threshold for INa is increased. When INa was inhibited by flecainide the QRS broadened. In the presence of pinacidil + flecainide, QRS did not significantly (p=0.52) prolong more than flecainide alone. Importantly, this is consistent with the lack of QRS change during pinacidil + flecainide in the Yang study, with the same caveats mentioned above. This result is also consistent with our previous finding that modulation of another inwardly-rectifying K+ current, IK1, impacted conduction under normal 77 conditions but not when INa was reduced.6 The question remains as to why pinacidil slowed conduction by itself but not in the presence of flecainide. One explanation is that resolution of the measurements may have been insufficient to detect a change in the pinacidil + flecainide groups. Alternatively, the more negative resting potential produced by pinacidil16 would decrease random openings of sodium channels during diastole,18 which may mitigate the impact of flecainide, a drug that preferentially binds to sodium channels in the open state.19 In order to directly quantify conduction, hearts were optically mapped with the voltage sensitive dye, di-4-ANEPPS. Pinacidil prolonged the QRS and slowed conduction in the presence of di-4-ANEPPS as it did in the absence of di4-ANEPPS, suggesting that the effects of pinacidil on conduction were not altered by di-4-ANEPPS. Furthermore, ARθ was not significantly different, which we interpret to mean that pinacidil principally affected sarcolemmal currents.17 Pinacidil + di-4-ANEPPS + flecainide did not change QRS duration, slow conduction or change ARθ relative to flecainide + di-4-ANEPPS. These data suggest that pinacidil has the same effect with and without di-4-ANEPPS. When flecainide concentration was increased in the presence of pinacidil + di-4-ANEPPS, we observed damping of conduction dependence on pharmacological INa inhibition. Specifically, the slope of the conduction velocityflecainide relationship was flattened by the addition of pinacidil. In our previous study, inhibiting IK1 (which is also carried by channels open during diastole) can increase the steepness of the conduction velocity-flecainide curve.6 Importantly, these data are consistent in that opening the non-voltage-gated potassium 78 channels flattens the conduction velocity-flecainide relationship, while inhibiting non-voltage-gated potassium channels steepens the same relationship. Taken together, the pinacidil data suggest that activation of an inwardlyrectifying potassium channel may lower resting membrane potential and slow conduction but without exacerbating conduction slowing by sodium channel blockers that bind in the open state. R-L3: effects on conduction In contrast to pinacidil, R-L3 activates the voltage-activated IKs. In addition to differences in voltage-dependent kinetics between IKATP and IKs, IKs activation by R-L3 has not been associated with changes in resting membrane potential.4;20 As might be expected, R-L3 perfusion alone did not significantly change QRS (p = 0.95). Thus, it was interesting to note that R-L3 + flecainide significantly broadened QRS more than flecainide alone, whereas we could not measure a difference between pinacidil + flecainide and pinacidil alone. This is consistent with our previous finding that activation of the voltage-activated IKr also prolonged the QRS only when conduction was already impaired.7 This is different from pinacidil which slowed conduction by itself, but did not cause further slowing in the presence of flecainide. These data suggest that there are mechanistic differences between the modulation of conduction by K+ currents like IKr and IKs which are carried by voltage-gated channels as opposed to K+ currents like IK1 and IKATP which are carried by inwardly-rectifying, non-voltage-gated channels. 79 It is interesting to note that flecainide prolongs time to dV/dtmax (i.e., increases latency) and slows conduction by delaying the time to maximal INa activation.19 Delaying time to peak INa activation may provide sufficient time for activated IKs to provide a significant current opposing depolarization by INa and thereby, slow conduction. In isolated guinea pig myocytes, R-L3 concentrations from 1 to 10 µM have been demonstrated to increase IKs between 750 and 1500%,20 which could represent a significant repolarizing current during depolarization. This hypothesis requires validation by single cell measurements. In the presence of di-4-ANEPPS, R-L3 prolonged the QRS whereas it did not do so by itself. This difference is likely related to the impairment of conduction by di-4-ANEPPS: it prolonged the QRS in the current study as well as in our previous studies.7;21 This effect of di-4-ANEPPS may be related to its previously suggested interaction with the sodium potassium ATPase.22 Indeed, sodiumpotassium ATPase inhibition by cardiac glycosides is known to increase intracellular sodium23 and to also broaden the QRS complex.24;25 Therefore, di-4ANEPPS's interaction with the sodium potassium ATPase could underlie the dye's effects on conduction; however, the specific mechanism warrants further study. Further, the results obtained with R-L3 are consistent with our previous findings that activation of another voltage-gated potassium current (IKr) slows conduction only when conduction is already impaired by di-4-ANEPPS or flecainide.7 Thus, the di-4-ANEPPS and flecainide data are consistent with the 80 hypothesis that voltage-gated potassium channel activation may slow cardiac conduction by opposing the delayed time to peak INa. Lastly, when flecainide doses were increased during perfusion of R-L3 + di-4-ANEPPS, conduction was slowed relative to flecainide + di-4-ANEPPS alone. As a result, the slope of the conduction velocity-flecainide relationship was not significantly altered by R-L3. This is in contrast with pinacidil which did not further decrease θ in the presence of flecainide and also decreased conduction dependence on INa blockade. These differences further suggest that R-L3 modulates conduction by a different mechanism from pinacidil. Conclusions In summary, we demonstrate that pharmacological activation of IKATP and IKs both decrease θ; however, they exhibit differential response to θ dependence on pharmacological inhibition of INa. Importantly, modulating K+ currents that do not display voltage-dependent kinetics may only affect conduction when Na+ channel availability is not reduced, whereas modulating K+ currents that display voltage-dependent kinetics may impact conduction when Na+ channel availability is reduced. These data suggest that pharmacological K+ channel activators must be evaluated for any possible effects on conduction velocity and safety of conduction before being used as antiarrhythmic therapy. 81 Limitations It is well established that most pharmacological agents have off-target effects which could interfere with the interpretation of this study. For example, both pinacidil and flecainide inhibit the transient outward current (Ito).26-28 Guinea pig does not functionally express Ito; therefore, these effects are ignored. R-L3 has been reported to block the L-type Ca2+ current (ICa,L) in isolated myocytes.20 ICa,L is normally active after the activation wavefront has passed a cell. Therefore, ICa,L should not affect conduction. Both R-L3 and flecainide have been shown to inhibit either IKr in mouse atrial tumor (AT-1) cells20 or maintained rat outward potassium current (IK), respectively. However, the interventions in this study either shortened or did not change QT. As a result, unexpected potassium channel inhibition does not likely affect the results or interpretation. BDM has not been associated with significant conduction changes in guinea pig heart.29 In other animals, BDM is associated with a small depression of conduction velocity at doses significantly greater than the one used in this study.30;31 Therefore, it is unlikely that BDM significantly affected the principal findings of this study. 82 References 1. Kannankeril, P, Roden, DM, Darbar, D. Drug-induced long QT syndrome. Pharmacol Rev. 2010;62:760-81. 2. Kaufman, ES. Mechanisms and clinical management of inherited channelopathies: long QT syndrome, Brugada syndrome, catecholaminergic polymorphic ventricular tachycardia, and short QT syndrome. Heart Rhythm. 2009;6:S51-5. 3. Radwanski, PB, Veeraraghavan, R, Poelzing, S. Cytosolic calcium accumulation and delayed repolarization associated with ventricular arrhythmias in a guinea pig model of Andersen-Tawil syndrome. Heart Rhythm. 2010. 4. Nissen, JD, Diness, JG, Diness, TG, Hansen, RS, Grunnet, M, Jespersen, T. Pharmacologically induced long QT type 2 can be rescued by activation of IKs with benzodiazepine R-L3 in isolated guinea pig cardiomyocytes. J Cardiovasc Pharmacol. 2009;54:169-77. 5. Grunnet, M, Hansen, RS, Olesen, SP. hERG1 channel activators: a new antiarrhythmic principle. Prog Biophys Mol Biol. 2008;98:347-62. 6. Veeraraghavan, R , Poelzing, S. Mechanisms Underlying Increased Right Ventricular Conduction Sensitivity to Flecainide Challenge. Cardiovasc Res. 2008;77:749-56. 7. Larsen, AP, Olesen, SP, Grunnet, M, Poelzing, S. Pharmacological activation of IKr impairs conduction in guinea pig hearts. J Cardiovasc Electrophysiol. 2010;21:923-9. 8. Snyders, DJ. Structure and function of cardiac potassium channels. Cardiovasc Res. 1999;42:377-90. 9. Poelzing, S, Veeraraghavan, R. Heterogeneous ventricular chamber response to hypokalemia and inward rectifier potassium channel blockade underlies bifurcated T wave in guinea pig. Am J Physiol Heart Circ Physiol. 2007;292:H3043-51. 10. Girouard, SD, Laurita, KR, Rosenbaum, DS. Unique properties of cardiac action potentials recorded with voltage-sensitive dyes. J Cardiovasc Electrophysiol. 1996;7:1024-38. 11. Bayly, PV, KenKnight, BH, Rogers, JM, Hillsley, RE, Ideker, RE, Smith, WM. Estimation of conduction velocity vector fields from epicardial mapping data. IEEE Trans Biomed Eng. 1998;45:563-71. 12. Girouard, SD, Pastore, JM, Laurita, KR, Gregory, KW, Rosenbaum, DS. 83 Optical mapping in a new guinea pig model of ventricular tachycardia reveals mechanisms for multiple wavelengths in a single reentrant circuit. Circulation. 1996;93:603-13. 13. Rohr, S, Kucera, JP, Kleber, AG. Slow conduction in cardiac tissue, I: effects of a reduction of excitability versus a reduction of electrical coupling on microconduction. Circ Res. 1998;83:781-94. 14. Flagg, TP, Nichols, C. Sarcolemmal KATP channels: what do we really know? J Mol Cell Cardiol. 2005;39:61-70. 15. Yang, Q, Padrini, R, Bova, S, Piovan, D, Magnolfi, G. Electrocardiographic interactions between pinacidil, a potassium channel opener and class I antiarrhythmic agents in guinea-pig isolated perfused heart. Br J Pharmacol. 1995;114:1745-9. 16. Baczko, I, Giles, WR, Light, PE. Pharmacological activation of plasmamembrane KATP channels reduces reoxygenation-induced Ca(2+) overload in cardiac myocytes via modulation of the diastolic membrane potential. Br J Pharmacol. 2004;141:1059-67. 17. Kleber, AG, Rudy, Y. Basic mechanisms of cardiac impulse propagation and associated arrhythmias. Physiol Rev. 2004;84:431-88. 18. Fozzard, HA, Hanck, DA. Structure and function of voltage-dependent sodium channels: comparison of brain II and cardiac isoforms. Physiol Rev. 1996;76:887-926. 19. Anno, T, Hondeghem, LM. Interactions of flecainide with guinea pig cardiac sodium channels. Importance of activation unblocking to the voltage dependence of recovery. Circ Res. 1990;66:789-803. 20. Salata, JJ, Jurkiewicz, NK, Wang, J, Evans, BE, Orme, HT, Sanguinetti, MC. A novel benzodiazepine that activates cardiac slow delayed rectifier K+ currents. Mol Pharmacol. 1998;54:220-30. 21. Larsen, AP, Sciuto, KJ, Moreno, AP, Poelzing, S. The Voltage-Sensitive Dye di-4-ANEPPS slows conduction velocity in isolated guinea pig hearts. In Review. 22. Fedosova, NU, Cornelius, F, Klodos, I. Fluorescent styryl dyes as probes for Na,K-ATPase reaction mechanism: significance of the charge of the hydrophilic moiety of RH dyes. Biochemistry. 1995;34:16806-14. 23. Demiryurek, AT, Demiryurek, S. Cardiotoxicity of digitalis glycosides: roles of autonomic pathways, autacoids and ion channels. Auton Autacoid Pharmacol. 2005;25:35-52. 84 24. Moe, GK, Mendez, R. The action of several cardiac glycosides on conduction velocity and ventricular excitability in the dog heart. Circulation. 1951;4:72934. 25. Swain, HH, Weidner, CL. A study of substances which alter intraventricular conduction in the isolated dog heart. J Pharmacol Exp Ther. 1957;120:13746. 26. Tseng, GN, Hoffman, BF. Actions of pinacidil on membrane currents in canine ventricular myocytes and their modulation by intracellular ATP and cAMP. Pflugers Arch. 1990;415:414-24. 27. Litovsky, SH, Antzelevitch, C. Transient outward current prominent in canine ventricular epicardium but not endocardium. Circ Res. 1988;62:116-26. 28. Slawsky, MT, Castle, NA. K+ channel blocking actions of flecainide compared with those of propafenone and quinidine in adult rat ventricular myocytes. J Pharmacol Exp Ther. 1994;269:66-74. 29. Liu, Y, Cabo, C, Salomonsz, R, Delmar, M, Davidenko, J, Jalife, J. Effects of diacetyl monoxime on the electrical properties of sheep and guinea pig ventricular muscle. Cardiovasc Res. 1993;27:1991-7. 30. Kettlewell, S, Walker, NL, Cobbe, SM, Burton, FL, Smith, GL. The electrophysiological and mechanical effects of 2,3-butane-dione monoxime and cytochalasin-D in the Langendorff perfused rabbit heart. Exp Physiol. 2004;89:163-72. 31. Baker, LC, Wolk, R, Choi, BR, Watkins, S, Plan, P, Shah, A, Salama, G. Effects of mechanical uncouplers, diacetyl monoxime, and cytochalasin-D on the electrophysiology of perfused mouse hearts. Am J Physiol Heart Circ Physiol. 2004;287:H1771-9. CHAPTER 4 INTERSTITIAL VOLUME MODULATES THE CONDUCTION VELOCITY- GAP JUNCTION RELATIONSHIP 86 Introduction The heart is an electrically actuated mechanical device. It pumps as a result of synchronous contraction of cardiac myocytes, achieved through the propagation of electrical signals from cell to cell. It is well recognized that intercellular transmission of these electrical signals occurs through gap junctions composed of connexin proteins.1 Indeed, it is well established that pharmacological Gj uncoupling slows cardiac conduction (θ).2 However, the impact of pathophysiological Gj remodeling is unclear. For instance, in heart failure, there is a lack of temporal coincidence between Gj remodeling and the development of arrhythmogenic conduction slowing.3 The impact of Gj remodeling has also been explored using transgenic mice heterozygous for a null mutation in connexin43 (Cx43), the principal ventricular gap junction (Gj) protein. This genotype is associated with a 50% reduction in Cx43. However, the results from these experiments have varied between no difference in conduction velocity from wild-type animals4-10 and significantly decreased conduction velocity.11;12 These apparently paradoxical findings have been attributed to experimental differences. However, the apparent disparity in results from genetic and pharmacological experiments suggests there may be other factor(s) that modulate the conduction velocity-gap junction (θ-Gj) relationship. We hypothesized that differences in perfusion may underlie the controversial results. We demonstrate in Langendorff-perfused guinea pig hearts that the perfusiondependent changes in interstitial volume (VIS) are a significant modulator of 87 conduction velocity, particularly transverse conduction (θT), as well as of the θ-Gj relationship. Methods The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). All animal study protocols were approved by Institutional Animal Care and Use Committee (IACUC) at the University of Utah. Guinea pig Langendorff preparations Adult male guinea pigs (800-1000g) were anaesthetized [30 mg/kg sodium pentobarbital (Nembutal) IP], their ventricles isolated and perfused (at 55 mm Hg) as Langendorff preparations with oxygenated Tyrode's solution (containing, in mMol/l, CaCl2 2, NaCl 140, KCl 4.5, dextrose 10, MgCl2 1, HEPES 10; pH 7.41) at 36.5 °C as previously described.13;14 In all experiments, 35 minutes of control Tyrode's solution was followed by 15 minutes of perfusion with Tyrode's solution containing either albumin (4 g/l) or mannitol (26.1 g/l). The respective measured perfusate osmolarities for control Tyrode's solution, Tyrode's solution with albumin (4 g/l) and Tyrode's solution with mannitol (26.1 g/l / 143.2 mOsm) were 290.7±1.8, 291.0±4.2 (p=ns vs. control) and 406.7±1.9 mOsm (p<0.05 vs. control). The electrocardiographic changes produced by both agents reached steady state within about 10 minutes for both agents (data not shown). The Cx43 88 uncoupler carbenoxolone was applied in 10 - 50 µM doses to assess the θ-Gj relationship. Guinea pig cardiomyocyte isolation Myocytes were isolated from excised, Langendorff-perfused hearts by enzymatic digestion. Hearts were perfused for 8 minutes with a calcium-free Tyrode's solution followed by an enzyme cocktail consisting of collagenase type 2 (Worthington Biochemical Corporation, Lakewood, NJ) and protease type XIV (Sigma Aldrich, St.Louis, MO) dissolved in Tyrode's solution containing 0.1mM calcium for 13 minutes. The enzyme solution was washed out for 5 minutes by perfusing 0.1mM calcium solution. After the washout, the hearts were removed from the cannula; the right and left ventricular free walls were sectioned out and then immersed in 0.1mM calcium solution. Each piece of tissue was cut into small bits and gently shaken with Tyrode's solution containing 2% albumin for 10 minutes to release the myocytes. Myocytes were moved to solutions of increasing calcium concentration (to reach a final concentration of 1mM in 3 sequential steps) with 6 minutes allowed for equilibration with each solution. Histology Computer morphometry was performed, as previously described,15;16 on H&E (hematoxylin & eosin) stained transmural slices of gluteraldehyde-fixed ventricular myocardium to quantify interstitial volume (VIS). A range of fixatives compared (ethanol, 18% formaldehyde, 27% formaldehyde and 8% 89 gluteraldehyde) based on the uniformity of H&E staining and degree of artifacts / tissue damage (data not shown) and gluteraldehyde was determined to yield the best results. Therefore, gluteraldehyde was used as the fixative in all experiments. A second set of slides was prepared from tissue samples that were frozen and cryosectioned. Briefly, the slides were scanned at 0.25 µm/pixel resolution using an Aperio ScanScope® XT system (Aperio Technologies Inc., Vista, CA). The interstitial space (IS) was segmented based on eosin staining by color deconvolution17 using the Aperio ImageScope™ software (version9, Aperio Technologies Inc., Vista, CA). Interstitial volume (VIS) was quantified as the fraction of total tissue area occupied by IS excluding blood and lymph vessels. As a metric of cell size the number of nuclei (segmented based on hematoxylin staining) per unit cellular area (areas with eosin staining) was quantified. In other words, the number of nuclei within the field of view was counted and normalized to the area within the field of view that was occupied by cells (determined based on eosin staining). All measurements were made from the subepicardium, defined as extending from the epicardium to a depth of 500 µm. Tissue water content Total tissue water content was measured by the ratio of the wet weight (WW; measured immediately following perfusion) to the dry weight (DW; measured after drying at 60°C for 24 hours), as previously described.18 90 Optical mapping Conduction velocity (θ) and anisotropy (ARθ) were quantified by optical voltage mapping using di-4-ANEPPS (15 µM) as a voltage indicator as previously described.13;14 Briefly, the preparation was stained with di-4-ANEPPS by direct coronary perfusion for 10 minutes, then excited by three- 60 LED light sources (RL5-A9018, Superbrightleds, St. Louis, MO) fitted with 510±5 nm filters (Chroma, Rockingham, VT). Fluoresced light was filtered using a 610 nm LP filter (Newport, Irvine, CA) before being recorded using a SciMedia MiCam02 HS CCD camera (SciMedia, Irvine CA) in a tandem lens configuration capable of resolving membrane potential changes as small as 2 mV with 1 ms temporal resolution from 90 x 60 sites simultaneously. Motion was reduced by perfusion of 7.5 mM 2,3-butanedione monoxime (BDM). The anterior epicardium was mechanically pressed against the front wall of the perfusion chamber to further stabilize and flatten it. Hearts were stimulated with a unipolar silver wire placed on the anterior epicardial surface at the center of the mapping field at 1.5 times the stimulation threshold with a basic cycle length (BCL) of 300 ms unless otherwise specified. Activation time was defined as the time of the maximum first derivative the action potential as described previously.19 The interpixel resolution was 0.184 mm in the x-direction (90 pixels) and 0.199 mm in the y-direction (60 pixels). A parabolic surface was fitted to the activation times as previously described.20 The gradient at each point was assigned a conduction velocity vector. The averaged conduction velocity vectors in the slow and fast axis of 91 propagation (±15°) are reported as they reflect transverse and longitudinal propagation.21 Western immunoblotting Western immunoblotting was performed as previously described14 to determine protein expression in the ventricles. Total Cx43 and pCx43 were respectively measured using a rabbit anti-Cx43 polyclonal antibody (Invitrogen, Carlsbad CA) and a rabbit anti-phospho-connexin43 (Ser368) polyclonal antibody (Cell Signaling Technology, Danvers MA) followed by the a goat antirabbit HRP-conjugated secondary antibody (JacksonImmuno, West Grove PA). The blots were then stripped and reprobed for actin (loading control) using a mouse anti-actin monoclonal antibody (Millipore, Temecula CA) followed by a goat anti-mouse HRP-conjugated secondary antibody (JacksonImmuno, West Grove PA). It should be noted that a separate dephosphorylated (P0) band for Cx43 is difficult to discern on the Cx43 Western blots given the high reactivity of the Cx43 antibody and the use of phosphatase inhibitors during sample preparation. However, the specificity of this antibody for Cx43 has been demonstrated previously.22 Statistical analysis Statistical analysis of the data was performed using a 2-tailed Student's ttest for paired and unpaired data or a single factor ANOVA. The Šidák correction was applied to adjust for multiple comparisons. Fisher's exact test was used to 92 test differences in nominal data. A p<0.05 was considered statistically significant. All values are reported as mean ± standard error unless otherwise noted. Results Mechanistic determinants of anisotropy In order to control for factors known to influence anisotropic conduction,1 we quantified total and phosphorylated Cx43 (pCx43) protein expression, myocyte size and interstitial volume (VIS). Representative Western immunoblots in Figure 4.1A and B demonstrate no difference in Cx43 or PCx43 band density during normal Tyrode (control) perfusion and after addition of albumin or mannitol to the control solution. The rightmost lane in Figure 4.1B is a positive control for p Cx43 dephosphorylation where the tissue sample was exposed to room temperature for 30 minutes. Over all experiments, there were no significant changes in either total Cx43 or pCx43 expression after perfusion of either albumin or mannitol relative to control (Figures 4.1 C, D). On the other hand, RV total Cx43 but not pCx43 bands were fainter relative to those from the LV. (Figures 4.1 A, B). Indeed, overall RV expression of Cx43 but not pCx43 was lower relative to LV (Figures 4.1C, D). Photomicrographs of isolated myocytes revealed that RV myocytes were significantly longer relative to LV but not different in width (see Appendix). Paired wide-field confocal images of a single myocyte exposed to the control, albumin or mannitol modified Tyrode's solution were used to produce a difference image by digital subtraction and overall, albumin did not significantly alter cell length or 93 Figure 4.1: Expression of Cx43, pCx43. Representative Western immunoblots of A) Cx43 and B) pCx43 from RV and LV with actin as loading control during control and after albumin and mannitol perfusion. A sample exposed to room temperature for 30 minutes is included in B as a positive control for dephosphorylation. Overall, neither albumin nor mannitol significantly altered expression of either C) Cx43 (n=6 per group) or D) pCx43 (n=6 per group). 94 width in isolated myocyte experiments (see Appendix). In contrast, mannitol significantly decreased cell width by 12.1±0.9% and cell length by 1.1±0.4% (see Appendix). H&E stained transmural sections from 8% gluteraldehyde-fixed (Figure 4.2A) and frozen (see Appendix) tissue show the nuclei in blue (stained with hematoxylin), the intracellular compartment in pink (stained with eosin) and the interstitial space in white (no staining). In intact tissue, the number of nuclei per unit cellular area (the number of nuclei normalized to cellular area within the field of view) was quantified as a metric of cell size. Neither albumin nor mannitol significantly altered this parameter relative to control in either gluteraldehydefixed or frozen tissue (Figure 4.2B). Also, the number of nuclei per unit cellular area was not different between ventricles (p=ns). Furthermore, VIS (excluding lymph and capillary vessels) was quantified by morphometry from the H&E stained fixed (Figure 4.2A) and frozen (see Appendix) sections. Specifically, Figure 4.2C presents the percentage of total tissue area occupied by the interstitium, which is our metric of VIS. Albumin perfusion was associated with significantly reduced VIS (by 23±1%), relative to control. Surprisingly, mannitol had the opposite effect of increasing VIS by 61±6% (Figure 4.2C). Total tissue water content, as measured by ventricular wet to dry weight ratio (WW/DW), supports the histological VIS measurements. Specifically, albumin decreased WW/DW relative to control, while mannitol increased it 95 Figure 4.2: Histology: A) Representative scans of H&E stained transmural sections prepared from gluteraldehyde-fixed tissue from RV and LV during control and after albumin / mannitol perfusion. The left half of each image depicts the H&E stained tissue and the right half demonstrates segmentation based on eosin staining. B) Overall, neither albumin nor mannitol significantly altered the number of nuclei per unit cellular area (quantified as a metric of cell size). (n=6 per group) C) Albumin decreased VIS relative to control while mannitol increased it. (*, p < 0.05 vs. control; n=6 per group). 96 97 (Figure 4.3). Also, RV WW/DW was greater relative to LV under all conditions (Figure 4.3). Electrophysiology Volume-conducted ECGs were recorded from control Tyrode-perfused preparations over a 60-minute period, the same time frame over which other interventions were performed. Neither the QRS duration nor the QT interval was significantly altered over this period (data not shown). Representative ECGs in Figure 4.4A demonstrate changes in electrophysiology associated with albumin or mannitol perfusion. Specifically, QRS duration was decreased during albumin and prolonged during mannitol perfusion (Figure 4.4B). QT interval was not significantly altered in the albumin case whereas it was prolonged on the same order of magnitude as the QRS interval in the mannitol case (Figure 4.4B). Importantly, neither intervention was associated with any ST segment changes. Furthermore, neither agent altered action potential duration measured either to full repolarization (APD) or to 30% repolarization (APD30) (Figures 4.4C, D). Velocity (θ) and anisotropy (ARθ) of conduction Next, we quantified θ by optical mapping during perfusion of control, albumin or mannitol modified Tyrode's solution. Representative optical action potential upstrokes in Figure 4.5A demonstrate the expected shorter delay in the longitudinal relative to the transverse direction during control Tyrode perfusion. The representative activation isochrone maps (Figure 4.5B) evidence elliptical 98 Figure 4.3: Tissue water content. Total tissue water content was quantified as the ratio of wet weight (WW) to dry weight (DW). Albumin decreased WW/DW relative to control whereas mannitol increased it (*, p < 0.05 vs. control, n = 5 per group). RV WW/DW was greater relative to LV under all conditions (†, p < 0.05 vs. LV, n = 5 per group). 99 Figure 4.4: Electrocardiograms and optical action potentials. A) Representative bath electrocardiograms during control, albumin and mannitol perfusion. B) Overall, albumin decreased QRS duration relative to control whereas mannitol increased it. Whereas albumin did not significantly alter the QT interval, mannitol increased to about the same extent as the QRS. C) Representative optical action potentials sections under control conditions and after perfusion of albumin and mannitol. D) Overall, neither albumin nor mannitol significantly altered action potential duration (APD). 100 101 Figure 4.5: Conduction velocity: representative data A) Action potential upstrokes from equally spaced sites. Representative activation isochrone maps: B) Control: LV ARθ > RV ARθ., C) Albumin increased θ in both ventricles, D) Mannitol decreased θ in both ventricles. 102 spread of excitation, consistent with anisotropic conduction. During control conditions, RV transverse (θT; 21.6±1.3 vs. 17.2±0.8 cm/s, p<0.05) but not longitudinal conduction velocity (θL; 51.6±2.0 vs. 52.6±1.0 cm/s, p=ns) was greater relative to LV. Consequently, anisotropy of conduction (ARθ; defined as θL/θT) was lower in the RV relative to LV (2.4±0.1 vs. 3.1±0.1, p<0.05). These measurements were repeated in control Tyrode-perfused preparations over a 60 minute period, the same time frame over which other interventions were performed. No significant change was observed in θL, θT or ARθ over this period (data not shown). Albumin, which was associated with decreased VIS, increased isochrone spacing particularly along the transverse direction, suggesting increased θ (Figure 4.5C). Overall, albumin was associated with increased θT (Figure 4.6A) and lowered ARθ (Figure 4.6B) in both ventricles relative to control. Conversely, mannitol, which was associated with increased VIS, decreased isochrone spacing, also preferentially along the transverse direction, suggesting decreased θ (Figure 4.5D). Overall, mannitol was associated with decreased θT (Figure 4.6A) and increased ARθ in both ventricles relative to control (Figure 4.6B). Concomitantly, mannitol increased incidence of spontaneous ventricular tachycardias (VTs) compared to control (3/8 vs. 0/15, p<0.05). In all cases, the VTs persisted for at least a minute and were terminated by a single defibrillatory shock. Albumin however did not significantly affect VT incidence relative to control (0/7 vs. 0/15, p=ns). 103 Figure 4.6: Conduction velocity: summary data A) Changes in RV, LV θT and θL from control conditions. Albumin increased RV and LV θT (*, p<0.05 vs. control; n=4). Mannitol decreased RV θT and θL and LV θT (*, p<0.05 vs. control; n=4). B) RV and LV ARθ during control and after albumin / mannitol perfusion. Control: LV ARθ > RV ARθ by 24.7% (†, p<0.05) Albumin significantly reduced ARθ from control by 28.3 and 23.0% in the RV and LV respectively (*, p<0.05). LV ARθ > RV ARθ by 38.8% (†, p<0.05) Mannitol significantly increased RV ARθ from control by 19.4% (*, p<0.05). RV and LV ARθ were no longer significantly different. 104 The θ-Gj relationship We probed the effects of perfusate on the θ-Gj relationship by measuring θ and ARθ in the presence of the Gj uncoupler carbenoxolone.2 Under control conditions, there was no significant difference in either longitudinal (dashed lines) or transverse (solid lines) θ between 0, 10, and 13 µM carbenoxolone. At 50µM, carbenoxolone decreased RV θT and θL by 25±3 (p<0.05) and 6±3% (p = ns) respectively (Figure 4.6A, black lines), increasing ARθ from 2.0 to 2.5. Similarly, 50µM carbenoxolone reduced LV θT and θL by 19±8 (p<0.05) and 17±3% (p<0.05) respectively (Figure 4.6A, black lines). This increased LV ARθ from 2.5 to 2.7. During mannitol perfusion, 10µM carbenoxolone still did not significantly alter θ from mannitol alone. However, 13µM carbenoxolone now decreased RV θT and θL by 38±9 (p<0.05) and 16±10% (p = ns) relative to mannitol alone (Figure 4.7A, red lines). This increased RV ARθ from 3.1±0.2 to 3.5±0.5 (p<0.05). In the LV, 13µM carbenoxolone decreased θT and θL by 38±8 (p<0.05) and 7±2% (p=ns) respectively and increased LV ARθ from 3.5±0.1 to 4.0±0.3, relative to mannitol alone. The incidence of VTs during perfusion of 13 µM carbenoxolone and mannitol was also higher relative to 13 µM carbenoxolone alone (7/9 vs. 0/5, p<0.05; Figure 4.8). Further increasing carbenoxolone dose (15 µM) resulted in loss of capture and conduction failure which resulted in VT (5/5 vs. 0/5 with 15 µM carbenoxolone alone, p<0.05; Figure 4.8). Similarly, we measured θ and ARθ in the presence of varying doses of the mannitol. Interestingly, increasing doses of mannitol were associated with a 105 Figure 4.7: Conduction velocity: dependence on Gj, VIS. A) Plots of RV, LV θ vs. carbenoxolone dose in the absence (black lines) and presence (red lines) of mannitol. Dotted horizontal lines indicate control levels. By itself, carbenoxolone significantly reduced θT in both ventricles compared to control at 50 µM (*, p<0.05, n=3) but not at 10 (n=4) and 13 µM (n=3). In the presence of mannitol, however, 13 µM carbenoxolone was sufficient to significantly reduce θT (*, p < 0.05 vs. 0 carbenoxolone; n=4). Further increasing carbenoxolone dose in the presence of mannitol resulted in conduction failure. B) Plots of RV, LV θ vs. mannitol dose in the absence (black lines) and presence (red lines) of carbenoxolone (15 µM). Dashed horizontal lines indicate control levels. By itself, mannitol significantly reduced θT in both ventricles compared to control at 143.2 mOsm (*, p<0.05; n=4) but not at 14.3 and 45.3 mOsm. In the presence of carbenoxolone, however, 14.3 and 45.3 mOsm doses of mannitol significantly reduced θT in both ventricles (*, p < 0.05 vs. 15 µM carbenoxolone alone; n=5). 106 107 Figure 4.8: Arrhythmia incidence. Incidence of spontaneous ventricular tachycardias (VTs) during perfusion of varying doses of carbenoxolone in the absence (white bars) and presence (black bars) of mannitol (143.2 mOsm). Mannitol significantly increased spontaneous VT incidence relative to control (i.e., 0 mannitol + 0 carbenoxolone) in the presence of 0 - 15 µM carbenoxolone (*, p < 0.05 vs. control). Further, VT incidence was significantly greater in the presence of mannitol + carbenoxolone relative to carbenoxolone alone for 13 and 15 µM doses of carbenoxolone (†, p < 0.05 vs. corresponding dose of carbenoxolone without mannitol). 108 linear increase in perfusion pressure (R=0.93, p<0.05). Under control conditions, 14.3 and 45.3 mOsm mannitol did not significantly slow conduction but 143.2 mOsm mannitol did (Figure 4.7B, black lines). These measurements were then repeated in the presence of 15 µM carbenoxolone, which by itself had no significant effect on θ. Carbenoxolone (15µM) + 14.3 or 45.3 mOsm mannitol significantly slowed conduction relative to 15 µM carbenoxolone alone (Figure 4.7B, red lines). Also, 15 µM carbenoxolone + 14.3 mOsm mannitol did not significantly alter VT incidence relative to 15 µM carbenoxolone alone (1/6 vs. 0/6, p=ns). In contrast, spontaneous VT incidence was higher during perfusion of 15 µM carbenoxolone + 45.3 mOsm mannitol (4/6 vs. 0/6, p<0.05) relative to 15 µM carbenoxolone alone. Discussion Although the role of Gjs in cardiac conduction is widely acknowledged, the impact of pathophysiological Gj remodeling on conduction remains a point of intense debate.4-9;11;12 Discrepancies among experimental findings have largely been attributed to experimental differences; however they23 may reflect the complexity of the θ-Gj relationship, arising from the influence of myocardial tissue architecture. We examined the influence of VIS on the θ-Gj relationship, and demonstrate that increases in interstitial volume are correlated with decreases in conduction velocity, increases in anisotropy of conduction. Additionally, increased interstitial volume is associated with increased conduction sensitivity to gap junction blockade. 109 Perfusate effects on myocardium Interstitial volume (VIS). Albumin, a ~66 kDa globular protein, has been demonstrated to interact with the endothelial glycocalyx and reduce capillary hydraulic conductivity and filtration rate.18;24;25 It has been used as a colloidal additive to perfusates during postischemic reperfusion to improve vascular barrier function and promote water retention within the vasculature, thereby preventing interstitial edema. In our Langendorff-perfused ventricle preparations (without ischemia), it reduced VIS relative to control (Figure 4.3), consistent with previous studies.26;27 In contrast, mannitol is a small molecule (molecular weight 182.17) that raised perfusate osmolarity (see Methods) and should extravasate freely.28 It has been used as a nonmetabolisable impermeate added to perfusates during postischemic reperfusion to reduces cell edema (i.e., cell swelling) by retaining water outside myocytes.28 In our Langendorff-perfused ventricle preparations, mannitol perfusion was associated with increased VIS (Figure 4.3). This may reflect two factors: Firstly, mannitol is a crystalloid impermeate that promotes fluid retention outside myocytes (i.e., in both the interstitium and the vasculature). Secondly, in a Langendorff-perfused preparation like ours, the perfusate flow rate is held constant; therefore, changes in total circulating fluid volume and resulting physiologic response, which play a vital role in whole animals, are absent. The combination of these two factors may explain the increase in VIS, i.e., interstitial edema, observed during mannitol perfusion in our experiments. 110 Further, neither agent altered the QT interval beyond the level of changes observed in QRS duration or measurably changed either APD30 or APD (Figure 4.4). These data suggest that the observed effects of albumin and mannitol perfusion are not related to ischemia.29 Cell size. Isolated RV myocytes are longer and more anisotropic in shape relative to LV myocytes during control (see Appendix), which is consistent with previous results.30 This in turn would argue for greater RV ARθ31; however, RV ARθ was lower relative to LV under control conditions (Figure 4.6). Whereas albumin did not significantly affect cell size (Figure 4.2), it significantly altered ARθ (Figure 4.6). Mannitol on the other hand rendered isolated cells narrower, which should have increased θ and flattened the θ-Gj relationship.32;33 However, the opposite was observed. Mannitol decreased θ (Figures 4.5, 4.6) and steepened the θ-Gj relationship. Yet, mannitol increased ARθ as one might expect if it preferentially decreased myocyte width. Importantly, mannitol did not alter the number of nuclei per unit cellular area (an index of cell size) in intact hearts (Figure 4.3). Since albumin did not alter cell size, and mannitol did not affect the estimation of intact myocardial cell size, but ARθ changed in opposite directions with respect to both interventions, we conclude that cell size cannot fully explain the conduction changes produced by albumin and mannitol. Gap junctions. Lower RV Cx43 protein expression relative to LV (Figure 4.1) would argue for greater RV ARθ.34;35 However, the lack of measurable difference in p Cx43 might suggest similar ARθ between the ventricles. We 111 observed lower RV ARθ relative to LV (Figure 4.6), suggesting that Cx43 functional expression does not fully explain ventricular ARθ differences. It is possible that increased VIS causes Gj uncoupling due to increased mechanical stress.36 Likewise, decreasing VIS may facilitate Gj formation by bringing cells closer together.37 We measured no change in Cx43 or pCx43 protein levels following perfusion of either albumin or mannitol (Figure 4.1). A positive control revealed that we were capable of measuring changes in these parameters, but changes caused by perfusate may have still been below our detection level. Importantly, these data suggest that Cx43 remodeling cannot fully explain θ changes associated with VIS changes but do not preclude the possibility of Cx43 changes modulating these effects. The θ-VIS relationship Albumin perfusion was associated with decreased VIS, preferentially increased θT, and lowered ARθ in both ventricles. Mannitol perfusion, on the other hand, was associated with increased VIS and preferentially decreased θT in both ventricles, and increased RV ARθ. In short, we demonstrate an inverse θ- VIS relationship (decreasing VIS increased θ and vice versa; Figure 4.3). This is inconsistent with the direct θ-VIS proportionality demonstrated in rabbit papillary muscle38 and mathematical models.31 There are two plausible explanations for this. First, cardiac conduction may be dependent on alternative modes of intercellular coupling that are not exclusively dependent on gap junctions.39;40 Alternatively, the changes in VIS could have modulated other electrophysiologic 112 parameter(s) which alter θ and ARθ. For example, modulating stretch-activated channels,36 sodium channels and potassium channels14 has been shown to modulate conduction. However, altering ionic currents should have no effect on ARθ31 and yet, we report significant changes in ARθ with our interventions. Thus, VIS may play a mechanistic role in modulating the θ-Gj relationship. VIS modulates the θ-Gj relationship Experiments using the pharmacological Gj uncoupler carbenoxolone suggest that VIS modulates the θ-Gj relationship (Figure 4.7A), a notion suggested by previous in silico studies.41 Specifically, conduction slowing and spontaneous VTs were observed at lower doses of carbenoxolone in the presence of mannitol compared to control conditions. Likewise, 15 µM carbenoxolone, a dose that did not significantly affect conduction, increased conduction sensitivity to VIS changes (Figure 4.7B) - conduction slowing and spontaneous VTs occurred at lower doses of mannitol in the presence of carbenoxolone relative to control. These data suggest an interrelationship between VIS and Gj vis-à-vis conduction. Conclusions In summary, we demonstrate increased VIS is associated with slowed conduction, preferentially in the transverse direction and a steeper θ-Gj relationship. Therefore, regardless of the mechanism, this study offers a potential explanation for conflicting results concerning the θ-Gj relationship.4-9;11;12 113 Specifically, differences in the θ-Gj relationship literature are likely due to VIS differences associated with perfusion. Additionally, these results are important for diseases such as heart failure. For example, Cx43 remodeling in a pacing induced canine heart failure model precedes conduction changes.3 Since heart failure is associated with myocardial edema,42;43 it is possible that gap junction remodeling precedes interstitial tissue accumulation, and therefore the decrease in conduction velocity may not be apparent in this model until both Cx43 is functionally down-regulated and VIS is increased. However, this hypothesis requires validation. Limitations The lack of observable difference in p Cx43 protein levels does not necessarily mean that there is no difference, or that protein levels correlate directly with the functional Cx43 in channels. Further, the effect of mannitol needs to be further elucidated and separated from the effects of increased perfusion pressure. We have previously demonstrated that opening outward K+ currents slows conduction but in an isotropic manner. Thus, these data should be interpreted cautiously with the consideration that alternative mechanisms could contribute to the changes of conduction we report. Still, changes in anisotropy have not been associated with transsarcolemmal currents, and further studies are required to determine the mechanism by which changing VIS modulates ARθ and the θ-Gj relationship. 114 References 1. Kleber, AG, Rudy, Y. Basic mechanisms of cardiac impulse propagation and associated arrhythmias. Physiol Rev. 2004;84:431-88. 2. Rohr, S, Kucera, JP, Kleber, AG. Slow conduction in cardiac tissue, I: effects of a reduction of excitability versus a reduction of electrical coupling on microconduction. Circ Res. 1998;83:781-94. 3. Akar, FG, Nass, RD, Hahn, S, Cingolani, E, Shah, M, Hesketh, GG, DiSilvestre, D, Tunin, RS, Kass, DA, Tomaselli, GF. Dynamic changes in conduction velocity and gap junction properties during development of pacing-induced heart failure. Am J Physiol Heart Circ Physiol. 2007;293:H1223-30. 4. Morley, GE, Vaidya, D, Samie, FH, Lo, C, Delmar, M, Jalife, J. Characterization of conduction in the ventricles of normal and heterozygous Cx43 knockout mice using optical mapping. J Cardiovasc Electrophysiol. 1999;10:1361-75. 5. Vaidya, D, Tamaddon, HS, Lo, CW, Taffet, SM, Delmar, M, Morley, GE, Jalife, J. Null mutation of connexin43 causes slow propagation of ventricular activation in the late stages of mouse embryonic development. Circ Res. 2001;88:1196-202. 6. Thomas, SP, Kucera, JP, Bircher-Lehmann, L, Rudy, Y, Saffitz, JE, Kleber, AG. Impulse propagation in synthetic strands of neonatal cardiac myocytes with genetically reduced levels of connexin43. Circ Res. 2003;92:1209-16. 7. van Rijen, HV, Eckardt, D, Degen, J, Theis, M, Ott, T, Willecke, K, Jongsma, HJ, Opthof, T, de Bakker, JM. Slow conduction and enhanced anisotropy increase the propensity for ventricular tachyarrhythmias in adult mice with induced deletion of connexin43. Circulation. 2004;109:1048-55. 8. Beauchamp, P, Choby, C, Desplantez, T, de Peyer, K, Green, K, Yamada, KA, Weingart, R, Saffitz, JE, Kleber, AG. Electrical propagation in synthetic ventricular myocyte strands from germline connexin43 knockout mice. Circ Res. 2004;95:170-8. 9. Stein, M, van Veen, TA, Remme, CA, Boulaksil, M, Noorman, M, van Stuijvenberg, L, van der Nagel, R, Bezzina, CR, Hauer, RN, de Bakker, JM, van Rijen, HV. Combined reduction of intercellular coupling and membrane excitability differentially affects transverse and longitudinal cardiac conduction. Cardiovasc Res. 2009;83:52-60. 10. Danik, SB, Liu, F, Zhang, J, Suk, HJ, Morley, GE, Fishman, GI, Gutstein, DE. Modulation of cardiac gap junction expression and arrhythmic susceptibility. Circ Res. 2004;95:1035-41. 115 11. Eloff, BC, Lerner, DL, Yamada, KA, Schuessler, RB, Saffitz, JE, Rosenbaum, DS. High resolution optical mapping reveals conduction slowing in connexin43 deficient mice. Cardiovasc Res. 2001;51:681-90. 12. Guerrero, PA, Schuessler, RB, Davis, LM, Beyer, EC, Johnson, CM, Yamada, KA, Saffitz, JE. Slow ventricular conduction in mice heterozygous for a connexin43 null mutation. J Clin Invest. 1997;99:1991-8. 13. Poelzing, S, Veeraraghavan, R. Heterogeneous ventricular chamber response to hypokalemia and inward rectifier potassium channel blockade underlies bifurcated T wave in guinea pig. Am J Physiol Heart Circ Physiol. 2007;292:H3043-51. 14. Veeraraghavan, R, Poelzing, S. Mechanisms Underlying Increased Right Ventricular Conduction Sensitivity to Flecainide Challenge. Cardiovasc Res. 2008;77:749-56. 15. Teman, CJ, Wilson, AR, Perkins, SL, Hickman, K, Prchal, JT, Salama, ME. Quantification of fibrosis and osteosclerosis in myeloproliferative neoplasms: A computer-assisted image study. Leuk Res. 2010;34:871-876. 16. Drakos, SG, Kfoury, AG, Hammond, EH, Reid, BB, Revelo, MP, Rasmusson, BY, Whitehead, KJ, Salama, ME, Selzman, CH, Stehlik, J, Clayson, SE, Bristow, MR, Renlund, DG, Li, DY. Impact of mechanical unloading on microvasculature and associated central remodeling features of the failing human heart. J Am Coll Cardiol. 2010;56:382-91. 17. Ruifrok, AC, Johnston, DA. Quantification of histochemical staining by color deconvolution. Anal Quant Cytol Histol. 2001;23:291-9. 18. Jacob, M, Bruegger, D, Rehm, M, Stoeckelhuber, M, Welsch, U, Conzen, P, Becker, BF. The endothelial glycocalyx affords compatibility of Starling's principle and high cardiac interstitial albumin levels. Cardiovasc Res. 2007;73:575-86. 19. Girouard, SD, Laurita, KR, Rosenbaum, DS. Unique properties of cardiac action potentials recorded with voltage-sensitive dyes. J Cardiovasc Electrophysiol. 1996;7:1024-38. 20. Bayly, PV, KenKnight, BH, Rogers, JM, Hillsley, RE, Ideker, RE, Smith, WM. Estimation of conduction velocity vector fields from epicardial mapping data. IEEE Trans Biomed Eng. 1998;45:563-71. 21. Girouard, SD, Pastore, JM, Laurita, KR, Gregory, KW, Rosenbaum, DS. Optical mapping in a new guinea pig model of ventricular tachycardia reveals mechanisms for multiple wavelengths in a single reentrant circuit. Circulation. 1996;93:603-13. 116 22. Li, WE, Ochalski, PA, Hertzberg, EL, Nagy, JI. Immunorecognition, ultrastructure and phosphorylation status of astrocytic gap junctions and connexin43 in rat brain after cerebral focal ischaemia. Eur J Neurosci. 1998;10:2444-63. 23. Akar, FG, Tomaselli, GF. Conduction abnormalities in nonischemic dilated cardiomyopathy: basic mechanisms and arrhythmic consequences. Trends Cardiovasc Med. 2005;15: 259-64. 24. Stevens, AP, Hlady, V, Dull, RO. Fluorescence correlation spectroscopy can probe albumin dynamics inside lung endothelial glycocalyx. Am J Physiol Lung Cell Mol Physiol. 2007;293:L328-35. 25. van den Berg, BM, Vink, H, Spaan, JA. The endothelial glycocalyx protects against myocardial edema. Circ Res. 2003;92:592-4. 26. Qin, H, Kay, MW, Chattipakorn, N, Redden, DT, Ideker, RE, Rogers, JM. Effects of heart isolation, voltage-sensitive dye, and electromechanical uncoupling agents on ventricular fibrillation. Am J Physiol Heart Circ Physiol. 2003;284:H1818-26. 27. Arbel, ER, Prabhu, R, Ramesh, V, Pick, R, Glick, G. A perfused canine right bundle branch-septal model for electrophysiological studies. Am J Physiol. 1979;236:H379-84. 28. Jacob, M, Paul, O, Mehringer, L, Chappell, D, Rehm, M, Welsch, U, Kaczmarek, I, Conzen, P, Becker, BF. Albumin augmentation improves condition of guinea pig hearts after 4 hr of cold ischemia. Transplantation. 2009;87:956-65. 29. Aslanidi, OV, Clayton, RH, Lambert, JL, Holden, AV. Dynamical and cellular electrophysiological mechanisms of ECG changes during ischaemia. J Theor Biol. 2005;237: 369-81. 30. Campbell, SE, Gerdes, AM, Smith, TD. Comparison of regional differences in cardiac myocyte dimensions in rats, hamsters, and guinea pigs. Anat Rec. 1987;219:53-9. 31. Spach, MS, Heidlage, JF, Barr, RC, Dolber, PC. Cell size and communication: role in structural and electrical development and remodeling of the heart. Heart Rhythm. 2004;1:500-15. 32. McIntyre, H, Fry, CH. Abnormal action potential conduction in isolated human hypertrophied left ventricular myocardium. J Cardiovasc Electrophysiol. 1997;8:887-94. 33. Seidel, T, Salameh, A, Dhein, S. A simulation study of cellular hypertrophy and connexin lateralization in cardiac tissue. Biophys J. 2010 ;99:2821-30. 117 34. Gutstein, DE, Morley, GE, Tamaddon, H, Vaidya, D, Schneider, MD, Chen, J, Chien, KR, Stuhlmann, H, Fishman, GI. Conduction slowing and sudden arrhythmic death in mice with cardiac-restricted inactivation of connexin43. Circ Res. 2001;88:333-9. 35. Jongsma, HJ, Wilders, R. Gap junctions in cardiovascular disease. Circ Res. 2000;86:1193-7. 36. McNary, TG, Sohn, K, Taccardi, B, Sachse, FB. Experimental and computational studies of strain-conduction velocity relationships in cardiac tissue. Prog Biophys Mol Biol. 2008;97:383-400. 37. Bukauskas, FF, Elfgang, C, Willecke, K, Weingart, R. Biophysical properties of gap junction channels formed by mouse connexin40 in induced pairs of transfected human HeLa cells. Biophys J. 1995;68:2289-98. 38. Fleischhauer, J, Lehmann, L, Kleber, AG. Electrical resistances of interstitial and microvascular space as determinants of the extracellular electrical field and velocity of propagation in ventricular myocardium. Circulation. 1995;92:587-94. 39. Sperelakis, N. An electric field mechanism for transmission of excitation between myocardial cells. Circ Res. 2002;91:985-7. 40. Lin, J, Keener, JP. Modeling electrical activity of myocardial cells incorporating the effects of ephaptic coupling. Proc Natl Acad Sci U S A. 2010;107:20935-40. 41. Cabo, C, Boyden, PA. Extracellular space attenuates the effect of gap junctional remodeling on wave propagation: a computational study. Biophys J. 2009;96:3092-101. 42. Boyle, A, Maurer, MS, Sobotka, PA. Myocellular and interstitial edema and circulating volume expansion as a cause of morbidity and mortality in heart failure. J Card Fail. 2007;13:133-6. 43. Mehlhorn, U, Geissler, HJ, Laine, GA, Allen, SJ. Myocardial fluid balance. Eur J Cardiothorac Surg. 2001;20:1220-30. 118 Appendix Supplemental Figures Supplemental Figure 4.S1: Photomicrographs of isolated myocytes. RV myocytes were longer than those from LV (*, p < 0.05); however, they were not significantly different in width. (n=3) 119 Supplemental Figure 4.S2: Myocyte size: effects of albumin, mannitol A) Representative photomicrographs of isolated myocytes exposed to control Tyrode's solution (left) followed by solution containing albumin or mannitol (middle) and a subtraction of the two images (right). B) Albumin did not alter either myocyte length or width (n=4) while mannitol significantly decreased both dimensions (*, p < 0.05; n=8). 120 Supplemental Figure 4.S3: Representative frozen sections. Representative scans of H&E stained transmural sections prepared from frozen tissue from RV and LV during control and after albumin / mannitol perfusion. The left half of each image depicts the H&E stained tissue and the right half demonstrates segmentation based on eosin staining. CHAPTER 5 SUMMARY AND FUTURE DIRECTIONS 122 Over the 137 years since Engelmann's first demonstration of cardiac conduction,1 research has considerably deepened our understanding of the phenomenon, its underlying mechanisms and the various tissue properties that are its determinants. However, there continues to be an urgent need for mechanistic insights into normal and aberrant conduction given that ventricular arrhythmias resulting from aberrant conduction2 remain a major contributor to the over 450,000 cases of sudden cardiac death that occur in the U.S. every year.3 While much has been gleaned thus far from the reductionist approach of examining the determinants of conduction one at a time, it is becoming increasingly clear that conduction abnormalities in a wide range of pathophysiological states result from multifactorial remodeling.4-7 Therefore, it is imperative that the mechanisms of conduction, both normal and abnormal, be examined using a holistic, multifactorial approach. Specifically, we must: 1. Identify the various determinants of conduction, 2. Delineate their influence, individually and together, on functional descriptors of conduction (i.e., conduction velocity, anisotropy) under normal and challenging conditions, and, 3. Correlate these interrelationships with regional heterogeneities in the determinants of conduction to understand regional conduction heterogeneities. For the purposes of study, it is convenient to classify the determinants of conduction between those that impact excitability and those that impact intercellular coupling. 123 Tissue excitability In Chapter 2, we presented results demonstrating a novel role for the inward-rectifier K+ current (IK1) in modulating cardiac conduction. Specifically, we demonstrated that regional conduction heterogeneities correlated with heterogeneities in IK1 under normal conditions; however, their dependence on the cardiac sodium current (INa) was unmasked under conditions where INa was reduced. Although, the magnitude of IK1 is small in comparison to that of INa, IK1 opposes depolarization particularly before sodium channels fully activate. Therefore, our results at the organ level are consistent with the hypothesis that reducing IK1 opposition to INa-mediated depolarization may augment conduction velocity. However, the precise cellular mechanisms underlying this phenomenon bear further investigation. One possible way to elucidate the interplay between IK1 and INa during phase-0 depolarization is to analyze the rise time of the action potential upstroke (total rise time measured from resting potential to the action potential peak; RTT; Figure 5.1) and resolve it into two components - the passive component (RTP; from resting potential to INa activation threshold; Figure 5.1) and the active component (RTA; from INa activation threshold to the action potential peak; Figure 5.1). A candidate cellular mechanism for the observed influence of IK1 on conduction is that the amplitude of IK1 is a key determinant of passive rise time (RTP), and therefore of total rise time (RTT), under conditions where sodium channel availability is not reduced; however the dependence of RTT on INa will be unmasked by sodium channel blockade. This hypothesis can be tested by 124 Figure 5.1: Components of the action potential upstroke. A) Control. Total rise time (RTT) consists of two components: Passive rise time (RTP) measured from resting potential to the activation threshold of INa and active rise time (RTA) measured from the activation threshold of INa to the peak of the action potential. B) IK1 blockade will likely shorten RTP but not RTA. C) INa blockade will likely prolong RTA but not RTP. 125 126 quantifying RTT as well as the ratio of RTA to RTP during normal conditions as well as during pharmacological inhibition of IK1 and/or INa. If partial blockade of a K+ current active during phase-0 depolarization can augment conduction velocity, it stands to reason that pharmacological augmentation of K+ currents may slow conduction. This question takes on particular relevance in the light of recent studies that have suggested using pharmacological K+ channel activators as a potential therapy for LQTS.8;9 Indeed, such drugs have demonstrated promise in mitigating action potential duration (APD) prolongation and dispersion of repolarization in experimental studies.810 However, we demonstrated in Chapter 3 that pharmacological activation of the ATP-sensitive K+ current (IKATP) and the slow delayed rectifier K+ current (IKs) do indeed slow conduction. Further, we demonstrated that activation of IKATP and IKs have disparate effects on the dependence of conduction velocity on INa. It was previously demonstrated that similar levels of conduction slowing produced by different mechanisms can have diverse effects: conduction failed at higher velocities during INa blockade in comparison to gap junction uncoupling.11 Therefore, K+ channels need to be evaluated for their effects on conduction safety and arrhythmia propensity under normal conditions as well as during challenging conditions where INa is reduced. From a mechanistic standpoint, pharmacological K+ current augmentation would not only increase the opposition to depolarization during the passive phase of the action potential upstroke but also to alter the resting membrane potential (RMP). Cellular measurements of RMP in the presence of K+ channel activators 127 and in silico studies of their effects on conduction may help resolve the contributions of these two mechanisms to the observed changes in conduction velocity. Intercellular coupling While gap junctions have long been recognized as the electrical conduits between myocytes, the question of how much gap junction uncoupling is required to slow conduction remains a hotly debated issue.12-19 We demonstrated in Chapter 4 that the interstitial volume (VIS) not only modulates conduction but also the conduction velocity-gap junction relationship. Out results suggest that changes in myocardial fluid balance may be an important determinant of conduction, particularly in conditions where gap junction coupling is reduced. Therefore, the combination of myocardial edema and gap junction uncoupling must be investigated for its contribution to arrhythmia incidence in pathophysiological states associated with these phenomena such as heart failure.20;21 Further, we experimentally demonstrated an inverse relationship between conduction velocity and interstitial volume, whereas measurements in cable-like papillary muscle22 and cable theory-based computer models23 have demonstrated a direct proportionality between the two. This disparity in results, combined with the greater sensitivity of transverse conduction to interstitial volume changes in our experiments, suggests that there may be important mechanistic differences between transverse and longitudinal conduction. 128 Importantly, the deviation of transverse conduction from cable theory predictions may point to the involvement of non-gap junctional, alternative modes of intercellular coupling, which have been suggested by previous modeling studies.24;25 While further computer modeling studies may help shed light on the possible role of alternative coupling mechanisms in our experiments, direct experimental evidence for such mechanisms is lacking. The effect of varying the distance of apposition on the efficacy of intercellular coupling could potentially be tested using isolated myocytes in the current clamp configuration. Specifically, two myocytes, both current clamped would need to be connected using a pseudo gap junction composed of a resistor. One cell could then be stimulated and the transmission of the resulting action potential to the other cell examined while varying the distance of apposition between the myocytes. If, for a given resistance of the pseudo-gap junction, the efficacy of action potential transmission to the nonstimulated myocyte varied as a function of the distance of apposition, it would strongly suggest the involvement of a non-gap-junctional mode of coupling between the two cells. It will likely require the combination of such cellular experiments, computer modeling and further whole heart experiments in order to conclusively answer the question of whether non-gap junctional intercellular coupling mechanisms are indeed operative in the heart. 129 Cardiac conduction: Revising our understanding Overall, the work presented here demonstrates that the mechanisms underlying changes in conduction are multifactorial in nature, whether they impact excitability or intercellular coupling. By themselves, each of these specific mechanisms is relevant to predicting/assessing conduction disturbances and resultant arrhythmia risk in specific pathophysiological context(s). For instance, understanding how conduction dependence on sodium channels is unmasked when the sodium current is reduced will help understand the mechanism of conduction defects in pathophysiological states such as the Brugada syndrome - a disease linked with reduced INa. Likewise, understanding conditions under which a K+ channel activator may slow conduction would be vital to the safe use of such drugs to treat pathologies such as the long QT syndromes. Knowledge of conditions under which conduction may become sensitive to gap junction uncoupling would help prevent conduction disturbances in conditions such as heart failure and ischemia. Scope and limitations of our findings However, these findings hold broader significance for our overall understanding of cardiac conduction. Where previous studies relied on a fully reductionist approach, assessing the role of physiological determinants of conduction, one at a time, we demonstrate the importance of interrelationships between the determinants of conduction. 130 However, our studies assessed these mechanisms under experimental conditions which were considerably simplified and different compared to the complex physiological milieu that exists within an intact organism. • The Langendorff-perfused guinea pig ventricle preparations we used, while advantageous from an experimental standpoint, are limited by the absence of autonomic innervation, the hormonal control of cardiac function. Further, our experiments made use of nonworking hearts, a necessity for optical mapping. Also, we used a nonphysiological perfusate, Tyrode's solution which, unlike blood, does not contain proteins which have been demonstrated to regulate vascular barrier function,26-28 and epicardial pacing produces a nonphysiologic activation sequence. • Further limitations arise from the experimental techniques themselves: optical mapping is limited to measurements from tissue within less than a millimeter of the epicardium.29 Likewise, Western immunoblotting performed on whole cell lysates does not yield information on functional expression of a protein, its subcellular localization etc. These limitations notwithstanding, our findings still represent an advancement of the understanding cardiac conduction: rather than the absolute values we measure for various parameters, the value of such experiments lies in identifying how these parameters are altered under one condition relative to another, thereby providing insights into underlying mechanisms and helping to assess arrhythmia risk during a given pathophysiologic state. 131 The study of cardiac conduction: Beyond reductionism Extending beyond their individual value, the greatest value of our findings can be realized when they are taken together. All our findings point to the importance of interrelationships between determinants of conduction and suggest it may be essential to move beyond reductionism to fully understand mechanisms underlying conduction. Extrapolating from these findings, it is possible, even likely, that there exist higher order interdependencies involving more than two determinants of conduction. Similarly, we have studied propagation in two dimensions along the epicardium, where tissue architecture is more or less uniform. Moving forward, it is essential to extend the mechanisms identified here to understand conduction as it happens in complex, three-dimensional structural of myocardial tissue. Another dimension of complexity arises from the experimental conditions themselves: as discussed above, our studies were conducted under simplified experimental conditions where we altered physiological parameters, one or two at a time. However, pathophysiological conditions are often characterized by complex, multifactorial changes to the physiological milieu. Therefore, the next step in research would be to apply the knowledge of the mechanisms identified here to experimental models that more closely replicate a complex disease state in order to identify specific pathophysiological mechanisms and potential anti-arrhythmic therapies. 132 Clinical perspectives Identifying novel modulators of conduction and understanding interrelationships between determinants of conduction have the twin benefits of improving prediction of arrhythmia risk in patients and providing a wider array of targets for antiarrhythmic therapy. For instance, managing myocardial fluid balance may improve the efficacy of existing anti-arrhythmic therapies in patients with pathological gap junction remodeling. Likewise, pharmacological K+ modulators could be used in lieu of or in tandem with Na+ channel modulators to produce precise changes in tissue excitability. Such multifaceted therapeutic approaches may help significantly cut arrhythmia burden in patients, while minimizing the dose required of anti-arrhythmic drugs and thereby the side effects resulting from off-target drug binding. 133 References 1. Engelmann, Th. Wilh. Ueber die leitung der erregung im herzmuskel. Pflügers Archiv European Journal of Physiology. 1875:11(1), 465-480. 2. Mehra, R. Global public health problem of sudden cardiac death. J Electrocardiol. 2007;40:S118-22. 3. Zheng, ZJ, Croft, JB, Giles, WH, Mensah, GA. Sudden cardiac death in the United States, 1989 to 1998. Circulation. 2001;104:2158-63. 4. Shih, H, Lee, B, Lee, RJ, Boyle, AJ. The aging heart and post-infarction left ventricular remodeling. J Am Coll Cardiol. 2011;57:9-17. 5. Barry, SP, Townsend, PA. What causes a broken heart--molecular insights into heart failure. Int Rev Cell Mol Biol. 2010;284:113-79. 6. Wang, Y, Hill, JA. Electrophysiological remodeling in heart failure. J Mol Cell Cardiol. 2010;48:619-32. 7. Takamatsu, T. Arrhythmogenic substrates in myocardial infarct. Pathol Int. 2008;58:533-43. 8. Nissen, JD, Diness, JG, Diness, TG, Hansen, RS, Grunnet, M, Jespersen, T. Pharmacologically induced long QT type 2 can be rescued by activation of IKs with benzodiazepine R-L3 in isolated guinea pig cardiomyocytes. J Cardiovasc Pharmacol. 2009;54:169-77. 9. Diness, TG, Yeh, YH, Qi, XY, Chartier, D, Tsuji, Y, Hansen, RS, Olesen, SP, Grunnet, M, Nattel, S. Antiarrhythmic properties of a rapid delayed-rectifier current activator in rabbit models of acquired long QT syndrome. Cardiovasc Res. 2008;79:61-9. 10. Radwanski, PB, Veeraraghavan, R, Poelzing, S. Cytosolic calcium accumulation and delayed repolarization associated with ventricular arrhythmias in a guinea pig model of Andersen-Tawil syndrome. Heart Rhythm. 2010. 11. Rohr, S, Kucera, JP, Kleber, AG. Slow conduction in cardiac tissue, I: effects of a reduction of excitability versus a reduction of electrical coupling on microconduction. Circ Res. 1998;83:781-94. 12. Morley, GE, Vaidya, D, Samie, FH, Lo, C, Delmar, M, Jalife, J. Characterization of conduction in the ventricles of normal and heterozygous Cx43 knockout mice using optical mapping. J Cardiovasc Electrophysiol. 1999;10:1361-75. 13. Eloff, BC, Gilat, E, Wan, X, Rosenbaum, DS. Pharmacological modulation of 134 cardiac gap junctions to enhance cardiac conduction: evidence supporting a novel target for antiarrhythmic therapy. Circulation. 2003;108:3157-63. 14. van Rijen, HV, Eckardt, D, Degen, J, Theis, M, Ott, T, Willecke, K, Jongsma, HJ, Opthof, T, de Bakker, JM. Slow conduction and enhanced anisotropy increase the propensity for ventricular tachyarrhythmias in adult mice with induced deletion of connexin43. Circulation. 2004;109:1048-55. 15. Guerrero, PA, Schuessler, RB, Davis, LM, Beyer, EC, Johnson, CM, Yamada, KA, Saffitz, JE. Slow ventricular conduction in mice heterozygous for a connexin43 null mutation. J Clin Invest. 1997;99:1991-8. 16. Beauchamp, P, Choby, C, Desplantez, T, de Peyer, K, Green, K, Yamada, KA, Weingart, R, Saffitz, JE, Kleber, AG. Electrical propagation in synthetic ventricular myocyte strands from germline connexin43 knockout mice. Circ Res. 2004;95:170-8. 17. Vaidya, D, Tamaddon, HS, Lo, CW, Taffet, SM, Delmar, M, Morley, GE, Jalife, J. Null mutation of connexin43 causes slow propagation of ventricular activation in the late stages of mouse embryonic development. Circ Res. 2001;88:1196-202. 18. Thomas, SA, Schuessler, RB, Berul, CI, Beardslee, MA, Beyer, EC, Mendelsohn, ME, Saffitz, JE. Disparate effects of deficient expression of connexin43 on atrial and ventricular conduction: evidence for chamberspecific molecular determinants of conduction. Circulation. 1998;97:686-91. 19. Stein, M, van Veen, TA, Remme, CA, Boulaksil, M, Noorman, M, van Stuijvenberg, L, van der Nagel, R, Bezzina, CR, Hauer, RN, de Bakker, JM, van Rijen, HV. Combined reduction of intercellular coupling and membrane excitability differentially affects transverse and longitudinal cardiac conduction. Cardiovasc Res. 2009;83:52-60. 20. Boyle, A, Maurer, MS, Sobotka, PA. Myocellular and interstitial edema and circulating volume expansion as a cause of morbidity and mortality in heart failure. J Card Fail. 2007;13:133-6. 21. Mehlhorn, U, Geissler, HJ, Laine, GA, Allen, SJ. Myocardial fluid balance. Eur J Cardiothorac Surg. 2001;20:1220-30. 22. Fleischhauer, J, Lehmann, L, Kleber, AG. Electrical resistances of interstitial and microvascular space as determinants of the extracellular electrical field and velocity of propagation in ventricular myocardium. Circulation. 1995;92:587-94. 23. Spach, MS, Heidlage, JF, Barr, RC, Dolber, PC. Cell size and communication: role in structural and electrical development and remodeling of the heart. Heart Rhythm. 2004;1:500-15. 135 24. Sperelakis, N, McConnell, K. Electric field interactions between closely abutting excitable cells. IEEE Eng Med Biol Mag. 2002;21:77-89. 25. Lin, J, Keener, JP. Modeling electrical activity of myocardial cells incorporating the effects of ephaptic coupling. Proc Natl Acad Sci U S A. 2010;107:20935-40. 26. Jacob, M, Bruegger, D, Rehm, M, Stoeckelhuber, M, Welsch, U, Conzen, P, Becker, BF. The endothelial glycocalyx affords compatibility of Starling's principle and high cardiac interstitial albumin levels. Cardiovasc Res. 2007;73:575-86. 27. Arbel, ER, Prabhu, R, Ramesh, V, Pick, R, Glick, G. A perfused canine right bundle branch-septal model for electrophysiological studies. Am J Physiol. 1979;236:H379-84. 28. Jacob, M, Paul, O, Mehringer, L, Chappell, D, Rehm, M, Welsch, U, Kaczmarek, I, Conzen, P, Becker, BF. Albumin augmentation improves condition of guinea pig hearts after 4 hr of cold ischemia. Transplantation. 2009;87:956-65. 29. Byars, JL, Smith, WM, Ideker, RE, Fast, VG. Development of an optrode for intramural multisite optical recordings of Vm in the heart. J Cardiovasc Electrophysiol. 2003;14:1196-202.