Biology and Biophysics of Cardiac Ion Channels

From numerous detailed single-cell, voltage-clamp studies, it has become obvious that a very specific heterogeneity in ionic current properties exists, depending on whether the ion channel target is in a ventricular cell (either epicardial or endocardial), a Purkinje cell, or an atrial cell. This has led to the rational development of drugs that target atrial-specific ion channels. For example, since I_Kur, the ultra-rapid delayed rectifier potassium (K⁺) current (encoded by the Kv1.5 gene), is thought to occur only in atrial cells, I_Kur blockers have been used to convert atrial fibrillation (AF). Unlike I_Kr blockers, I_Kur blockers should delay atrial repolarization (a class III effect) without affecting ventricular repolarization. At this time, these agents remain investigational; on further study, many have been found to have a multiple-ion channel–blocking effect. An additional concern is that even complete I_Kur blockade would not lead to sufficient prolongation of atrial repolarization to terminate re-entrant excitation.

Newer antiarrhythmic class III agents have been developed for their marked reverse-use dependence (e.g., nifekalant), for selective I_Ks blockade (e.g., HMR1556), and for blockade of I_to, a major repolarizing K⁺ current that is found in both atrial and ventricular cells (e.g., tedisamil). In addition, several amiodarone derivatives have been developed to produce agents with a similar ion channel–blockade profile as that of amiodarone but with fewer side effects. An example is dronedarone, an amiodarone derivative with no iodine. Like amiodarone, dronedarone blocks multiple K⁺ currents to prolong ventricular action potential duration (APD).

A drug that blocks inward plateau currents would be considered an anti–class III drug in that it should shorten APD. An antianginal drug, ranolazine, is being investigated for its antiarrhythmic properties, since it strongly inhibits the late I_Na current with little or no effect on peak I_Na.⁷ Thus ranolazine would shorten APD rather than reduce excitability in cardiac tissues. More interestingly, this drug, by blocking sodium (Na⁺) influx into the cardiac cell, would be expected to reduce intracellular Na⁺ and therefore would indirectly affect intracellular Ca²⁺ and all its sequelae (see below).

One form of re-entrant excitation is anisotropic re-entry, which was first considered in atrial samples.⁸ In this type of re-entry, conduction in the longitudinal direction is faster than that in the transverse direction. Spach and his colleagues suggested that a component of conduction slowing in the transverse direction in these cardiac tissues was caused by a change in gap junctional conductance. Thus slowing of conduction was brought about by the uncoupling of cells at the level of the gap junctional proteins, for example, connexins 43 and 40 (Cx43 and Cx40). While experiments have clearly shown that gap junctional uncouplers (e.g., heptanol) are, in fact, arrhythmogenic, a preferential effect appears to occur on transverse conduction velocities in various animal models. Thus a corollary would be that in re-entrant circuits in highly remodeled substrates such as those affected by cell uncoupling caused by ischemia, one would expect drugs that enhance gap junctional conductance to be antiarrhythmic. Rotigaptide is now under study as a gap junctional coupler. In experimental arrhythmia models, this drug improves conduction and abolishes lines of block that perpetuate circuits and therefore has been deemed antiarrhythmic. This experimental drug is thought to affect coupling by preventing dephosphorylation of Cx43, a ventricular gap junctional protein, or by maintaining phosphorylation of the protein.

Factors Contributing to Cardiac Remodeling

Gene-Based Arrhythmias

Major advances have occurred in our understanding of the genetic basis of several forms of inherited arrhythmia syndromes. With these advancements has come the development of gene-specific therapies that depend on the genes involved in the syndrome.

Cardiac sodium channel (SCN5a) mutations are involved in at least four genetic disorders: long QT syndrome type 3 (LQT3), some forms of Brugada syndrome (BrS), progressive conduction disease (CCD), and sick sinus syndrome. The last three disorders are considered to be caused by loss of function of the Na⁺ channel. For LQT3 patients (a gain of function from the destabilization of Na⁺ channel inactivation leading to an enhanced late I_Na), a rational approach to therapy has been to use long-acting Na⁺ channel–blocking drugs (e.g., class I, mexiletine) to reduce Na⁺ influx during the plateau. This would reduce the pathologically prolonged APD and long QT. Disease-associated genes here include SCN5a (the α-subunit of the cardiac Na⁺ channel), CAV3 (caveolin 3 protein), and SCN4B (a major accessory β-subunit essential for the proper functioning of the cardiac Na⁺ channel).

For loss of function of the Na⁺ channel, no specific Na⁺ channel agonists would be useful in restoring excitability to the mutant channel cardiac cells. While trafficking defective mutants have been identified in Na⁺ channels and are associated with BrS, they can be “rescued” using mexilitene. However, Na⁺ channel–rescuing agents are not practical, since, at this time, most candidates also have Na⁺ channel–blocking effects. Some have suggested that by blocking K currents (e.g., I_to), the altered plateau of action potentials in BrS can be overcome, restoring normal electrical function and preventing the initiation of arrhythmia in BrS patients. To this end, quinidine (class I) and tedisamil, a new agent, have been tested. Interestingly, isoproterenol has also been reported to be antiarrhythmic (see www.BrugadaDrugs.org).

Calcium channel mutations are involved in at least two genetic disorders, Timothy syndrome/LQT (a gain of function caused by the enhanced late I_CaL), and some forms of BrS combined with short QT syndrome (SQT; loss in function, APD shortening). Gene products involved here are CACNA1C (Timothy syndrome/LQT), and CACNA1C, and CACANB2 (in SQT/BrS). Rational therapy would be Ca²⁺ channel blockers (class IV) for patients with Timothy syndrome/LQT and Ca²⁺channel activators for patients with short QT/BrS.

Multiple K⁺ channel mutations underlie both LQT (LQT1, LQT2, LQT5, and LQT6) and SQT syndromes (SQTS1, SQTS2, and SQTS3). LQTS variants that are linked to K⁺ channel mutations are dominant-negative or trafficking defects. Despite differing biophysical mechanisms, these mutants all result in a loss in K⁺ channel function at the cell membrane. Genes associated with these variants are KCNQ1, KCNH2, KCNE1, KCNE2, and KCNJ2. Thus loss of K⁺ repolarizing currents leads to pathologic APD prolongation, which, under the appropriate sympathetic stimulation, can lead to early after-depolarizations (EADs) and torsades de pointes (TdP). On the one hand, β-blockers (class II) are useful in some forms of LQT (e.g., LQT1) since they ameliorate the effects of the sympathetic-induced triggers. On the other hand, for the second most common form of LQT, LQT2, β-blocker therapy is not always useful. Here the goal of therapy should be to counter the loss of function in I_Kr and thus correct the potentially malignant long APD. In this case, some have proposed that by increasing plasma K⁺ concentrations (K⁺ supplementation), one would enhance I_Kr conductance and, in so doing, counter the mutant channel loss in function. Recent experimentation has also suggested that a class of drugs could rescue mutant I_kr channels to restore APD values. More commonly, K⁺ channel activators such as nicorandil have been proposed as an appropriate therapy.

Characterized by AF, ventricular fibrillation (VF), or both, SQT syndrome reflects the opposite of long QT syndrome in that it results from K⁺ channel mutations leading to a gain in function. Three forms of this syndrome have been described. SQTS1 (gain in function of KCNH2[I_Kr]), SQTS2 (gain in function of KCNQ1[I_Ks]) and SQTS3 (gain in function of KCNJ2 [Kir2.1, major protein of I_K1]). Specific K⁺ channel blockers would be useful to normalize the patient’s QT intervals to protect from the initiation of lethal ventricular arrhythmias. Probably, hydroquinidine (class I), which blocks multiple K⁺ currents, would be the most successful.

Loss in function of connexin proteins (GJAS, Cx40) and the I_f protein (HCN4) have both been linked to AF and sinus node dysfunction, respectively. At this time, no agents are used to activate the function of these channels for antiarrhythmic control.

Intracellular Calcium and Targets

Under normal conditions (see Chapter 2), during systole, Ca²⁺ is released from the sarcoplasmic reticulum (SR) through a channel known as the ryanodine receptor (RyR). The important property of RyR protein is its open probability that is increased by the elevation of cytoplasmic Ca²⁺ concentration [Ca²⁺]_i. Thus Ca²⁺ entry into the cell via the L-type Ca²⁺ current produces a small increase of Ca²⁺, which leads to an opening of the RyR and the release of a much greater amount of Ca²⁺ from the SR. This process is known as calcium-induced calcium release (CICR).

When a cell is “overloaded” with calcium, Ca²⁺ leaks out of the SR and waves of CICR propagate along the cell. It appears that Ca²⁺ waves occur when the SR Ca²⁺ content is elevated above a threshold value.^10,11 Some of the Ca²⁺ in the wave is pumped out of the cell by the electrogenic Na⁺-Ca²⁺ exchange (NCX). The resulting current depolarizes the membrane and can initiate an action potential (Figure 11-2).

FIGURE 11-2 Schematic diagram of the genesis of delayed after-depolarizations (DADs) and possible therapeutic strategies. The upper part of the diagram shows the factors leading to a triggered action potential. Ca²⁺ waves can be produced either by an increase of sarcoplasmic reticulum (SR) Ca²⁺ content or an increase in the opening of the ryanodine receptor (RyR). The increased SR content can be produced by either increased Ca²⁺ influx or decreased Ca²⁺ efflux across the cell membrane. The Ca²⁺ wave activates the Na⁺-Ca²⁺ exchange (NCX) and the resulting inward current will produce a DAD, which, if above threshold, will result in an action potential. The sites of action of various therapeutic maneuvers are shown below. Ca²⁺ antagonists will decrease Ca²⁺ entry into the cell. β-blockers will decrease phosphorylation of the L-type Ca²⁺ channel, thereby decreasing Ca²⁺ entry, and will decrease phosphorylation of phospholamban, thereby decreasing SR Ca activity and SR Ca²⁺ content. Drugs that decrease RyR opening will increase the threshold SR Ca²⁺ content required to produce a wave. Blockers of NCX will decrease the depolarizing current produced by a wave. Finally, local anesthetics will decrease the probability that a given DAD will activate an action potential.

Drugs that decrease Na⁺-K⁺ pump to increase Ca²⁺ to produce a positive inotropic effect are available. However, as is known, digitalis-type compounds also increase Ca²⁺ to such an extent as to cause triggered arrhythmias, presumably by causing spontaneous SR Ca²⁺ release, which then initiates a Ca²⁺ wave. A desirable antiarrhythmic agent would be one that modulates Ca²⁺ so that Ca²⁺ does not increase Ca²⁺-dependent currents to cause depolarization and action potentials of cells. If the spontaneous Ca²⁺ releases are targeted, then the initiators of Ca²⁺ waves, delayed after-depolarizations (DADs), and thus triggering beats could be reduced.

The arrhythmias mentioned above result when SR Ca²⁺ content is increased above the threshold level at which waves are produced. Recent work has suggested that a decrease of threshold may also produce waves. One example relates to arrhythmias seen in heart failure, where the involvement of DADs in some ventricular arrhythmias has been shown.^12,13 However, studies on heart failure have found that the SR Ca²⁺ content is actually decreased, which suggests that the threshold for Ca²⁺ release may be lower such that Ca²⁺ waves occur at a lower SR Ca²⁺ content. This may be a consequence of the increased leakiness of the RyR during diastole such that Ca²⁺ efflux is increased at a given SR Ca²⁺ content. The exact molecular mechanisms responsible for this are still being debated, but it may be associated with the increased phosphorylation of the RyR caused by protein kinase A or CaMKII.^14–16

An example of the occurrence of DADs in the absence of increased SR Ca²⁺ content is provided by catecholaminergic polymorphic ventricular tachycardia (CPVT). This arrhythmia is seen in patients during exercise or other stress. The similarity of the abnormalities in the electrocardiogram (ECG) to those observed in digitalis toxicity led to the suggestion of similarities in the underlying mechanisms. Genetic studies have shown that many patients with CPVT have a mutation in RyR or the intrasarcoplasmic protein calsequestrin. The current hypothesis is that the mutated protein causes an increased leak of Ca²⁺ from the SR. Thus Ca²⁺ waves and DADs occur at a lower SR Ca²⁺ content than in controls.¹⁷

Potential Therapies for Delayed After-Depolarization–Related Arrhythmias

In principle, as indicated in Figure 11-2, arrhythmias can be treated in several ways: (1) by preventing DAD, (2) by preventing DAD from producing an action potential, or (3) by both. The latter can potentially be achieved by Na⁺ channel blockers. A better solution, however, would be to remove the underlying DAD directly. Again, several potential approaches to this are possible. In the case of arrhythmias resulting from Ca²⁺ overload, it may be possible to remove the underlying “overload.” Local anesthetics reduce intracellular Na⁺ concentration as a consequence of decreasing Na⁺ entry; therefore, via NCX, this will decrease the Ca²⁺ load. β-Blockers (class II) are the mainstay of therapy for patients with CPVT, but even with this therapy, the recurrence rate is about 30%. β-Blockers decrease the cellular Ca²⁺ load by decreasing phosphorylation of the L-type Ca²⁺ channel and phospholamban, the latter leading to a decrease of SERCA2 activity and thus SR Ca²⁺ content. It would also be possible to modulate Ca²⁺ by affecting the membrane transports or channels involved in Ca²⁺ homeostasis. L-type Ca²⁺ channel pore blockers obviously decrease Ca²⁺ influx and, in so doing, would be expected to eventually reduce the SR load and [Ca²⁺]_i and diminish force. Thus, Ca²⁺ channel pore blockers will affect Ca²⁺ but at the expense of force generation. Alternatively, one might target the molecular mechanism involved in the inactivation of the Ca²⁺ channel proteins or the Ca²⁺-dependent processes known to affect the Ca²⁺ channel function (e.g., CaMKII) or the small proteins (e.g., Gem) that are known to affect Ca²⁺ channel subunit assembly.

An alternative approach would be to stop the Ca²⁺ wave from developing. One caution is required here. Although the Ca²⁺ efflux during the wave is proarrhythmogenic, it does have the useful effect of removing Ca²⁺ from the cell. Abolishing the wave may result in an increase of diastolic [Ca²⁺]_i and thus impair relaxation. Many drugs target the RyR. The local anesthetic tetracaine decreases RyR opening and thereby increases the SR threshold. In experimental studies, tetracaine was shown to abolish Ca²⁺ waves.¹⁸ Tetracaine is not used clinically for this purpose, since at concentrations at which it affects the RyR, it also blocks sarcolemmal Na⁺ channels. Very recent work has shown that flecainide suppresses CPVT arrhythmias both in humans and in a murine model.¹⁹ This appears to be caused by a combination of a direct effect to decrease RyR opening that decreases the occurrence of Ca²⁺ waves and an effect to inhibit Na⁺ channels.

Another compound is JTV519 (K201), which has been shown to decrease arrhythmias in animal models. JTV519 (K201) and its sister drug S107 appear to affect SR Ca²⁺ leak.^20–22 This drug is a 1,4 benzothiazepine and thus can also decrease the L-type Ca²⁺ current. However, in isolated SR vesicle systems, it has been shown to reduce Ca²⁺ leak by restoring the normal FKBP12.6 stabilization of the RyR complex and improving defective gating.^20,23,24

If the calsequestrin gene defect that occurs in some patients causes an increased free [Ca²⁺]_SR, the effect will be that the probability of opening of the RyR increases, which is positive inotropic and potentially arrhythmogenic. Restoration of the protein and its proper level by gene therapy would ameliorate this arrhythmogenic defect, but this being a useful alternative is still a distant possibility. Using a compound that would cause the existing calsequestrin to have a higher affinity for Ca²⁺ would be antiarrhythmic but may have possible negative inotropic effects.

Would modulation of the Na/Ca exchanger activity be antiarrhythmic? NCX serves to maintain Ca²⁺ homeostasis and in several acquired diseases appears to be “upregulated.” Therefore reducing Na/Ca forward mode function would be expected to reduce the size of the current enhanced for any given change in Ca²⁺. Therefore there would be a decrease in the depolarization produced by the Ca²⁺ waves of the triggering beats. However the consequences of nonselective inhibitors of this transporter are complex. KB-R7943 is a well-characterized inhibitor but its specificity is open to question. A newer agent, SN-6, holds promise for reducing cell injury in the presence of ischemia but there are no data on its antiarrhythmic effects. There are also other agents such as SEA0400 that inhibit the exchanger.²⁵ A recent study using the XIP peptide in normal and failing cells in fact shows a positive inotropic effect of the peptide.²⁶ Indirect effects of such XIP effects would be expected to reduce Ca²⁺-dependent activation of I_ti and XIP-induced SR Ca²⁺ release would shorten the APD prolonged by aberrant Ca²⁺. Finally, at least in principle, it is clearly possible to partially inhibit NCX and thereby decrease the depolarizing current generated by a Ca²⁺ wave. The problem with this approach, however, is that it will increase [Ca²⁺]_i and lead to other undesirable consequences.

Would modulation of SR Ca²⁺ adenosine triphosphatase (ATPase; SERCA2) function be antiarrhythmic? Overexpression of SERCA would increase Ca²⁺ uptake into the SR at the expense of Ca²⁺ efflux via the Na⁺-Ca²⁺ exchanger. Therefore, decreased Na⁺-Ca²⁺ exchanger current would be evident immediately. In this way, on the one hand, cytosolic Ca²⁺ would (should) decrease and the SR fill, increasing the amplitude of the stimulated Ca²⁺ transient. On the other hand, the increase of SR Ca²⁺ content might make Ca²⁺ waves more likely. Gene transfer techniques to treat arrhythmias are still far from being used in practice, but some studies have offered proof-of-principle results. SERCA overexpression via gene transfer techniques has been shown to do just this, but no reports of an antiarrhythmic effect exist, although it decreases aftercontractions and also accelerates APDs.^27,28 In a more recent report, SERCA2a overexpression previous to ligation of the left anterior descending coronary artery greatly reduced episodes of ventricular tachycardia plus VF.²⁹ This effect was aligned with a decrease in Ca²⁺. Pharmacologic enhancement of SERCA pump activity is possible; however, it is not known if these agents are antiarrhythmic.

Stimulation of SERCA pump activity could be arrhythmic if it occurred in the presence of an RyR channel mutation that leads to spontaneous Ca²⁺ release. Such releases in the setting of enhanced SR filling would increase the likelihood of triggering a Ca²⁺ wave and thus a DAD. A combination of an agent that would stimulate the pump and one that would prevent spontaneous Ca²⁺ release should therefore be effective.

Future Directions

With sophisticated molecular techniques, certain arrhythmias may come to be identified as being dependent on intracellular Ca²⁺ for either their initiation or perpetuation. Further, the aberrant Ca²⁺-binding protein(s) will be identified, and using molecular approaches, the amino acid basis of the Ca²⁺-binding site will be identified. With this knowledge, new drugs that would correct the Ca²⁺-binding problem in a highly specific way and thus ameliorate the problem will certainly be developed.

Acknowledgment

Supported by grant HL67449 from the National Heart Lung and Blood Institute Bethesda, MD, and the British Heart Foundation, UK.