Ionic Mechanisms of Arrhythmogenesis

Abstract

The understanding of ionic mechanisms underlying cardiac rhythm disturbances (arrhythmias) is an issue of significance in the medical science community. Several advances in molecular, cellular and optical techniques in the past few decades have substantially increased our knowledge of ionic mechanisms that are thought to underlie the arrhythmias. The application of these techniques in the study of ion channel biophysics and regulatory properties has provided a wealth of information, with some important therapeutic implications for dealing with the disease. In this review we briefly consider the cellular and tissue manifestations of a number of cardiac rhythm disturbances, while focusing on our current understanding of the ionic current mechanisms that have been implicated in such rhythm disturbances.

1. Introduction

An abnormal rhythm of the heart (arrhythmia) is a significant cause of concern in the medical community, and in its most severe manifestation, i.e., fibrillation, the abnormality has devastating consequences. Ventricular fibrillation (VF) is the most common cause of >250,000 sudden cardiac deaths each year, and atrial fibrillation (AF) affects ∼2 million people in the United States alone, increasing their risk for stroke and heart failure [1, 2]. In the past few decades, advances in a variety of techniques, including molecular and cellular electrophysiology and optical mapping of electrical excitation have led to a substantial increase in our knowledge of ionic mechanisms underlying cardiac arrhythmias. We have come to better understand that these abnormalities in rhythm may occur as a result of inherited genetic disorders [3], acquired cardiac disease conditions such as heart failure [4], or may be due to side-effects of drugs [5]. Regardless of the underlying cause, certain unifying principles have been found to underlie the initiation and maintenance of arrhythmias, which can be broadly classified as occurring at the cell and tissue levels (Fig. 1). At the cellular level, arrhythmogenesis can occur as a result of abnormal oscillations (automaticity), and early or delayed depolarizations. i.e. EADs and DADs respectively [6]. At the tissue level, vortex shedding and reflection can play a role in the initiation of arrhythmogenesis, whereas reentry in the form of spiral waves or rotors is now thought to play an important role in sustaining arrhythmias including AF and VF [7]. This review briefly describes these two levels (cellular and tissue) by summarizing the main ionic current determinants that are thought to underlie the various arrhythmia mechanisms.

Fig. 1. Mechanisms of Arrhythmias.

The various cellular and tissue mechanisms of arrhythmias are depicted.

2. Automaticity

The primary pacemaker of the heart, the sinoatrial node (SAN), generates spontaneous action potentials (normal automaticity), and in doing so is able to suppress the latent or subsidiary pacemakers such as the AV node, as well as the Purkinje fibers [6]. The ionic mechanisms governing the SAN automaticity have been studied in substantial details, and it is thought that this oscillatory nature is governed by multiple ionic mechanisms, which are shown in Fig. 2A [8]. The diastolic depolarization is thought to result from the inward currents generated by the hyperpolarization-activated funny current, I_f, the Na⁺-Ca²⁺ exchanger current, I_NaCa, the T-type Ca²⁺ current, I_CaT, and the concomitant decay of the outward repolarizing K⁺ currents (the rapid and slow delayed rectifiers I_Kr and I_Ks, as well as the transient outward current, I_to) [8]. The exact contribution of I_f versus the so-called Ca²⁺ clock in driving pacemaking activity via I_NaCa though remains unresolved [9]. The pacemaking activity can be enhanced due to sympathetic regulation (Fig. 2B), causing a faster diastolic depolarization, and sinus tachycardia. This occurs mainly due to adrenergic effects on multiple ion channels, which includes gating shifts in the activation of I_f (to more depolarized voltages), and the well-known effects of enhancing Ca²⁺ cycling, as well as the densities of the L-type Ca²⁺ current, I_CaL, I_Ks, and the Na⁺-K⁺ pump current, I_NaK [8]. Similar ionic mechanisms (albeit with different precise contribution of various ion channels) contribute to the other pacemaking sites in the heart [10]. However, abnormal automaticity can also arise from normally well polarized atrial and ventricular cells, and is often seen in the setting of ischemia, infarction or stretch [11, 12]. This phenomenon was elegantly demonstrated in single isolated rabbit Purkinje cells, which were coupled to a ventricular cell “model” wherein the resting potential (V_rest), and the coupling resistance could be varied in a controlled manner [13]. At a V_rest of -50 mV, and a moderate coupling resistance of 400 MΩ, which mimicked inducing injury currents, spontaneous oscillations, or “abnormal automaticity” could be induced in normally quiescent Purkinje cells (Fig. 2C). A similar abnormal automaticity has been observed in depolarized ventricular cells when the inward rectifier K⁺ current, I_K1, was suppressed by means of adenovirus [14]. These abnormal automaticities could underlie ventricular tachycardia seen in the setting of ischemia/infarction [15]. Further, abnormal automaticity originating in the pulmonary vein region is now thought to underlie the initiation of AF, predominantly in the paroxysmal form [16].

Fig. 2. Automaticity.

(A.) The normal SAN pacemaker activity and the main underlying ionic currents are shown [8]. (B.) Enhanced automaticity in the SAN pacemaker during adrenergic stimulation, and underlying ionic determinants are shown [8]. (C.) Shows coupling of Purkinje cell to a ventricular cell model via a resistance of 400 MΩ. The ventricular cell is clamped at V_rest of -50 mV, which induces abnormal automaticity in the Purkinje cell [13].

3.EAD and DAD

Initiation

Afterdepolarizations (EADs and DADs) are abnormal transmembrane electrical activities that are always associated with a preceding cardiac action potential (atrial or ventricular) [6]. A typical normal ventricular action potential is generated due to the activation/inactivation of inward or depolarizing currents such the Na⁺ current, I_Na, and I_CaL, and a number of repolarizing K⁺ currents (I_to, I_Kr, I_Ks, I_K1) [17]. Further, a number of electrogenic exchanger (I_NaCa) and pump (I_NaK) currents also contribute, and some of these are shown in Fig. 3A [17]. An EAD occurs, when the repolarization phase is prolonged either in acquired disease conditions such as heart failure [4] or inherited Long QT syndromes (LQT1, LQT2, LQT3) [3]. The prolongation of repolarization can occur as a result of a downregulation in various K⁺ currents, or sustained opening of inward currents such as the late Na⁺ current [3,4]. Other conditions that can contribute to a prolongation of the action potential and lead to EADs, include electrolyte abnormalities such as hypokalemia, acidosis, slow heart rate, catecholamines, or drugs that block K⁺ currents [6]. The prolonged repolarization at depolarized potentials leads to a reactivation of I_CaL, which then turns on and off, leading to small membrane depolarizations, also known as EADs [18]. Fig. 3B shows simulation results from modeling of LQT2 [19]. When the ventricular cell that incorporates a reduced I_Kr density (LQT2) is continuously paced at 500 ms basic cycle length (BCL), the action potential is prolonged, but repolarizes normally (pre-pause). However, when an action potential is evoked after a long pause of 1500 ms, the repolarization is abnormally prolonged, and displays EADs. The underlying Ca²⁺ transient is seen to be normal, but the EADs are driven by repeated re-activation of I_CaL [19]. EADs are thought to play an important role in the genesis of Torsade de Pointes (TdP) [6]. A DAD occurs after the action potential is fully repolarized [20], and its genesis is generally attributed to the occurrence of Ca²⁺ overload in a cardiac cell [21], which may occur in heart failure [4], due to digitalis toxicity [20], or even due to mutations in Ca²⁺ cycling proteins such as ryanodine, which lead to catecholaminergic polymorphic tachycardia (CPVT) [22]. Ca²⁺ overload causes calcium waves, and can generate a transient inward current (I_ti), which depolarizes and causes DADs [6]. It has been proposed that I_ti is generated either by the activation of an electrogenic I_NaCa [23], a nonspecific cationic current [23], or the calcium-activated chloride current [24]. A computer simulation of a DAD in a ventricular cell is shown in Fig. 3C; in this case Ca²⁺ overload due to a calsequestrin mutation is mimicked [25]. After the last paced beat, a spontaneous Ca²⁺ release leads to a DAD, which is also converted to a spontaneous action potential or triggered activity; this is because in addition to a larger inward I_NaCa, the I_K1 current density is also downregulated in this simulation [25]. Thus prolonged repolarization and I_CaL reactivation is key for EADs, and abnormal Ca²⁺ cycling, I_NaCa, and I_K1 are seen to be important ionic determinants of a DAD.

Fig. 3. Afterdepolarizations.

(A.) The normal ventricular action potential and underlying ionic currents are depicted [17]. (B.) Representative simulation of an action potential in LQT2 at a basic cycle length of 500 ms (pre-pause AP), and an EAD occurring after a pause of 1500 ms is shown. Also shown is the underlying Ca²⁺ transient as well as the repetitive reactivation of I_CaL [19]. (C.) Representative simulation of a DAD and a spontaneous action potential during Ca²⁺ overload, and the underlying Ca²⁺ transient, ionic currents (I_CaL, I_NaCa, I_Na), are shown [25].

Propagation into Tissue

In order to initiate arrhythmias, propagation of an EAD/DAD into tissue is necessary. An important factor that controls the spread of excitation via an EAD/DAD is intercellular coupling. A decrease in cell-cell coupling is commonly observed in diseases such as heart failure and ischemia (which gives rise to EADs and DADs), and has also been shown to be associated with slower impulse propagation [17]. Slower impulse propagation can lead to reentry (see next section). In experimental and theoretical studies of coupled cardiac cells and tissue, it was shown that strong intercellular coupling suppressed EAD/DAD formation, whereas in the presence of moderate uncoupling, these afterdepolarizations could propagate into the tissue; in the case of severe uncoupling however, the impulse propagation failed [26-28].

4. Spiral Waves: Ionic Bases of Initiation and Maintenance

Initiation

The concept of reentry has been around for almost a century, and the cardiac muscle was shown to sustain circulating excitatory wavefronts by early pioneering investigators, such as Mayer, Mines and Lewis; this topic is briefly reviewed in a recent article [7]. Beginning in the 1970s, the concept of leading circle driving reentry was put forward [29], whereas others, using both theoretical concepts, and experiments using voltage sensitive dyes in cardiac tissue postulated that spiral waves were driving functional reentry [30,31]. However, the precise mechanisms of fibrillation, i.e. small number of rotors, or multiple random wavelets, remain unresolved [32]. Despite this, the concept of spirals as underlying drivers of fibrillation has found support in both experiments and simulations [7], and recently in the clinic [33]. A spiral wave is a form of functional reentry in which the curved wavefront and wavetail meet each other, at a singularity point (white asterisk, Fig. 4A) [34], and the tissue at the center is not refractory, unlike the leading circle theory [7]. The wavefront represents an area of depolarized cells as the cardiac impulse travels, and the wavetail is made up of cells that are returning to rest (repolarization). A stationary spiral spins around a circular trajectory forming the “core” (Fig. 4A); however, the core can take various complex shapes, depending upon the tissue excitability. The concept of wavelength (λ) is defined as the distance between the wavefront and wavetail, and is measured by the product of refractory period (RP) and the conduction velocity (CV), i.e. λ = RP × CV, and is applicable for a planar wavefront of cardiac impulse propagation [17]. In contrast, there is no “fixed” wavelength for spiral waves, but the distance between the wavefront and the wavetail varies, increasing as a function of the distance from the singularity point (Fig. 4A) [7]. This is because electrotonic gradients established between cells at the core and cells in the near vicinity shorten significantly the action potentials of cells near the core. Fig. 4B shows a snapshot of a simulated rotor generated using a computer ionic model of a 2D sheet of atrial cells in persistent AF [35]. The top panel shows membrane voltage distribution, and the bottom panel shows a plot of the product of the variables “h.j”, which represent the fast (h) and slow inactivation variables (j) of the Na⁺ current, I_Na. These variables represent the availability of I_Na, and when “h.j” is 0.0, no I_Na is available, and the tissue in unexcitable (the white area in the bottom panel of Fig. 4B), whereas a value of 1.0 means that the tissue is fully available for excitation. All other values between 1.0 and 0.0 represent the “excitable tissue or gap”, where I_Na is available [7]. A spiral wave is initiated when cardiac impulse propagation is blocked, in part due to tissue heterogeneities in excitability, repolarization or conduction velocity, or even dynamical properties such as action potential alternans, and is analogous to the formation of eddies, when a flow of water waves is broken when it interacts with an obstacle, and rooted in the concept of critical curvature [7]. The concept is explained in the schema shown in Fig. 4C. An artificial linear obstacle is etched into the ventricular muscle sheet (red line), and the curvature of the wavefront about to proceed is shown (Fig. 4C). The white curve bounds the area adjacent to the obstacle, which needs to propagate laterally in order to circumnavigate the obstacle. The radius of this area R is comparable to the width of the wavefront, which determines whether that wavefront (i.e., the source) will be able to excite the tissue ahead of it (i.e., the sink). The ability of the source (or propagating wavefront) to provide enough current in order to elicit a response in the area ahead (the sink) is what allows the impulse to propagate safely (“safety factor”). On the other hand, if the source is unable to provide the necessary current, the safety factor is compromised, and the resulting “source-sink mismatch” can lead to conduction blocks [17]. As shown in the top panel of Fig. 4C (condition I), when the excitability of the tissue is normal and a wave is initiated by a point stimulus near the lower right margin of the obstacle (asterix), the wavefront proceeds to circumnavigate the obstacle without detaching and activates the entire sheet [7]. Here, R is larger than the minimum radius of excitation R_Cr necessary to sustain normal propagation. In the lower panel of Fig. 4C, when the excitability of the tissue is diminished (condition II), R < R_Cr, and the wavefront now detaches from the obstacle, curls and begins to rotate around its broken tip, generating a spiral [7]. These concepts were illustrated in the combined experimental-theoretical studies of Cabo et al., where excitability was manipulated by blocking I_Na [36]. This scenario can exist in pathophysiological conditions such as ischemia or an infarct in the ventricles, or atrial remodeling due to persistent AF, where both I_Na density and excitability are reduced and numerous obstacles in the form of patchy fibrotic tissue exist, thereby reducing the safety factor for impulse propagation (resulting in slow conduction), and set the stage for initiation of the spirals that maintain tachycardia or fibrillation [7]. The ionic mechanisms underlying the safety factor have been investigated in a series of elegant studies [17,37]. The results showed that when excitability was reduced by gradual reduction in the availability of I_Na, the safety factor was also reduced monotonically, and at a certain level, when the safety factor was <1, the impulse propagation ceased. In contrast, as the intercellular coupling due to gap junctions was reduced, although the CV was also reduced, the safety factor was in fact increased, and very slow propagation could be sustained [17,37]. This also has implications for maintenance of spiral waves (see next section). In this regard, an interesting and important development in recent years has been the realization that proteins/ion channels thought to act independently such as the Na⁺ channel, are in fact interacting with other proteins including gap junctions, in a macromolecular complex [38]. For example, it was shown that loss of Cx43 expression also results in a reduction in the density of I_Na [39]. Similarly, an overexpression of SCN5A that underlies I_Na in turn has been shown to influence the density of the inward rectifier, I_K1, and vice versa [40]. Thus, our concepts regarding a safety factor for propagation will have to be revisited to take into account these newly described ion-channel interactions.

Fig. 4. Spiral Waves, and their Initiation.

(A.) Schematic of the spiral wave: Electrotonic effects of the core decrease conduction velocity (arrows), action potential duration (representative examples shown from positions 1, 2 and 3) and wavelength (the distance from the wave front (black line) to the wave tail (dashed line) [7]. As CV decreases and wavefront curvature becomes more pronounced, a point singularity occurs as the wave front and tail meet “*” [34]. (B.) Computer simulation of spiral wave. Top panel: snapshot of the transmembrane voltage distribution during simulated reentry in chronic AF conditions in a 2D sheet incorporating human atrial ionic math models [35]. Bottom panel: snapshot of inactivation variables of sodium current, “h.j” during reentry. (C.) Schematic of spiral wave initiation. As a wave progresses along an obstacle (red line) the curvature of the wave at the edge of the obstacle will determine if the wave detaches. If the curvature of the wavefront (R) at the edge of the obstacle is greater than the critical curvature for detachment (R_Cr) the wave remains attached; if R is less than R_Cr, when excitability is reduced (due to reduced I_Na, increased I_K1, or enhance fibrosis) the wave will detach from the obstacle and initiate reentry [7].

Maintenance

Earlier studies in guinea pig hearts suggested an important role of I_K1 in VF, which was terminated by BaCl₂ at low concentrations [41]. The chamber-specific, left-to-right atrial spatiotemporal gradient of AF frequencies in the presence of acetylcholine was shown to be attributable to different densities of another inward rectifier K⁺ channel, the acetylcholine-activated K⁺ current, I_KACh [42]. The density of I_K1 was also found to be upregulated in atrial myocytes isolated from chronic AF patients [2]. When simulations were conducted to examine the consequences of I_K1 increase, the spirals were found to be faster, and the core area was smaller, thus stabilizing fibrillation [35]. The simulations suggested, that in addition to causing a shorter action potential duration (APD), I_K1 accelerated spirals by increasing the availability of I_Na; this was possible due to the hyperpolarization of the resting membrane potential (V_rest) by I_K1. In transgenic mice where I_K1 was overexpressed, it was possible to induce sustained spirals (> 1 hour) in isolated hearts, which were extremely fast (∼45-60 Hz), compared to wild-type hearts, where spirals were transient, and slower (∼20-25 Hz) [43]. In Fig. 5A, the activation map of a spiral observed experimentally in a TG mouse with I_K1 overexpression (right) versus a WT heart (left) is shown [43]. It can be seen that the spiral completes one rotation much earlier in the TG mouse [43]. To study the role of the two main repolarizing K⁺ currents in the human ventricle, i.e. the fast and the slow delayed rectifier K⁺ currents, viz. I_Kr and I_Ks [44], we utilized monolayers of confluent, electrically coupled neonatal rat ventricular myocytes (NRVMs) [45,46]. The density of either I_Ks or I_Kr was increased by adenoviral overexpression of either KvLQT1-minK, or hERG, respectively in NRVM monolayers. We found that I_Ks overexpression did not increase spiral frequency (∼5-13 Hz in both cases), but over time, caused an increasing number of wavebreaks to occur, as shown in Fig 5B by the representative phase maps in a control monolayer, and a monolayer in which I_Ks was increased [45]. Experiments in isolated HEK cells and suggested that the increased wavebreaks was because of the phenomenon of post-repolarization refractoriness, occurring due to residual outward I_Ks current, following the action potential [45]. In contrast to I_Ks, a larger density of I_Kr caused the spiral to accelerate significantly compared to control (∼21 Hz in the former, versus ∼9 Hz in the latter case) [46]. Both experiments and simulations showed that the faster spiral was due to APD shortening, as well as transient V_rest hyperpolarization, which again would indirectly affect spiral frequency by modifying I_Na availability [46]. Ionic currents affecting intracellular Ca²⁺ cycling ([Ca²⁺]_i), such as L-type Ca²⁺ current, I_CaL, also influence the APD, and are thus likely to influence the spiral wave dynamics. In a previous study, verapamil (a partial blocker of I_CaL), reduced the excitation frequency of VF, the core meander was increased, and VF was converted to ventricular tachycardia (VT) [47]. However, these results must be interpreted with caution, since at the concentrations of verapamil used in this study, verapamil blocks both I_CaL, as well as I_Kr [48]. The role of [Ca²⁺]_i in sustaining VF/spirals remains controversial. One report suggested that [Ca²⁺]i -action potential dissociation during VF was a consequence, not a cause of wavebreaks in VF, and that no spontaneous voltage-independent [Ca²⁺]_i waves could be seen [49]; however, this finding remains controversial [50]. The Na⁺ current, I_Na is a key determinant of excitability and the upstroke of the cardiac action potentials [17]. In presence of TTX, which blocks I_Na, the excitation frequency of the spiral wave was reduced, and the core meander was increased [51]. The role of other ionic currents such as I_to, the atrial-specific delayed rectifier I_Kur and I_CaT in maintaining spirals is less clear. Recent experiments have also shed light on the important role of intercellular coupling in the maintenance of spiral waves. In the presence of Carbenoxolone, a relatively selective inhibitor of intercellular coupling, spiral wave reentry was stabilized, and the incidence of pacing-induced sustained ventricular tachycardia (>30 seconds) was substantially increased in rabbit hearts (85+4%), compared to controls (38+4%) [52].

Fig. 5. Maintenance of Spiral Waves.

(A.) Spirals in wild-type (WT) mouse hearts and transgenic (TG) mice where I_K1 was overexpressed [43]. (B.) Spirals two-dimensional layers of rat neonatal ventricular myocytes in control, and where I_Ks was overexpressed via adenovirus [45]. (C.) Spirals in two-dimensional layers of rat neonatal ventricular myocytes in control, and where I_Kr was overexpressed via adenovirus [46].

Termination

Our previous experimental and theoretical studies have explored the mechanism(s) underlying the termination of sustained spiral waves [7], either by simulating drug block of K⁺ channels [35], or by a reduction in excitability by pathophysiological mechanisms such as hyperkalemia [53].

The results suggested that spiral waves can terminate either by collision with each other, or at the tissue boundary, whereby they are extinguished [35, 53]. A reduction in excitability [53] or an increased plateau phase of the action potential [35] resulted in an increase in the spiral tip or core meander, which in turn increased the possibility of the spiral waves to be pushed toward the boundaries and to self-terminate. Similar mechanisms have been proposed to explain the success of (i) defibrillation in terminating VF, where it has been suggested that shock results in the unpinning or destabilization of spiral waves [54], and (ii) termination of sustained tachycardia in the presence of rotigaptide, which enhances intercellular coupling [52].

5. Reflection

One of the earliest studies on reflection was conducted by Antzelevitch and colleagues, and is shown in Fig 6A [55]. Canine false tendons were placed in a three chamber solution, where the middle chamber was perfused with sucrose of high K⁺, to stop conduction. Transmembrane potential recordings were made in the proximal (P) and distal (D) chambers, and the sucrose gap was shunted by means of an extracellular resistance. At the right values of this resistance, the stimulus in the proximal segment elicited an action potential in this segment, and with an appropriate delay in the distal segment as well. Interestingly, once the proximal chamber action potential had recovered, the electrotonic currents flowing from the distal to the proximal direction could invoke an action potential in the P region as well (reflection); see Fig 6A [55]. More recently, a similar phenomenon was found in 2D tissue of NRVM monolayers, where proximal and distal rectangular patches were connected via a narrow isthmus (top left panel, Fig. 6B) [56]. When the proximal region was stimulated, and the propagation was plotted as a function of time across a narrow line through the isthmus and patches (bottom left panel, Fig. 6B), the impulse first propagated from the proximal side to the distal side, and then was reflected into the proximal side again. The time dependent action potentials can be seen in the right panel (Fig. 6B). In this monolayer, I_Na was overexpressed via adenovirus, which led to prolonged APD/EAD on the distal side; the isthmus provided a delay for the impulse propagation due to the source-sink mismatch, which allowed the proximal side to recover, and upon receiving electrotonic current from the isthmus, fire a reflected action potential [56]. Thus, reflection is a complex mechanism driven by delays and electrotonic current induced via source-sink mismatch, and is suggested to underlie either ventricular extrasystole or bigeminy [6].

Fig. 6. Reflection.

(A.) Shows transmembrane potentials in proximal (P) and distal (D) canine false tendons, with reflection giving rise to a second action potential in the P region [55]. (B.) The top left panel shows two rectangular patches of NRVM connected via a narrow isthmus. The bottom left panel shows time-space plot (Y axis, time, X axis-space) of impulse propagation (green) along the dotted line shown in the above panel. The right panel shows time plot of action potentials along the green line [56].

6. Other Mechanisms

There are other arrhythmias, which indeed can be classified as a subset of the mechanisms described above. One such phenomenon is phase 3 EADs [57], or phase-2 reentry [58]. In the case of the former, the EAD is thought to occur due to an abbreviated APD but a normal Ca²⁺ transient, and an inward current principally mediated via I_NaCa [57]. It has been proposed that phase-2 reentry is important in mediating J waves [6] and Brugada syndromes [58], and is caused via an I_to-mediated shortening of the action potential, which causes a re-excitaion of the epicardium, and subsequently leads to reentry [58]. Other mechanisms that predispose to arrhythmias include alternans (voltage or calcium), which is again due to insufficient recovery of either inward currents I_CaL or I_Na, or abnormal Ca²⁺ cycling mechanisms [59]. It is thought that alternans are precipitated at either high frequencies of pacing, or pathophysiological conditions such as ischemia, and give rise to reentrant tachycardia/VF [60]. Other forms of reentry have also been proposed, such as Figure of 8 reentry [61], or anisotropic reentry [62]. Anisotropic cellular coupling and microfibrosis has been proposed to underlie the latter [62], although further studies are necessary to examine their precise ionic determinants.