Cardiac Potassium Channels: Physiological Insights for Targeted Therapy

The development of novel drugs specifically directed at the ion channels underlying particular features of cardiac action potential (AP) initiation, recovery, and refractoriness would contribute to an optimized approach to antiarrhythmic therapy that minimizes potential cardiac and extracardiac toxicity. Of these, K+ channels contribute numerous and diverse currents with specific actions on different phases in the time course of AP repolarization. These features and their site-specific distribution make particular K+ channel types attractive therapeutic targets for the development of pharmacological agents attempting antiarrhythmic therapy in conditions such as atrial fibrillation. However, progress in the development of such temporally and spatially selective antiarrhythmic drugs against particular ion channels has been relatively limited, particularly in view of our incomplete understanding of the complex physiological roles and interactions of the various ionic currents. This review summarizes the physiological properties of the main cardiac potassium channels and the way in which they modulate cardiac electrical activity and then critiques a number of available potential antiarrhythmic drugs directed at them.


Introduction
Orderly propagation of cardiac electrophysiological excitation and recovery depends on a normal sequence of cardiac action potential (AP) generation through its component myocytes. The depolarization and repolarization of AP is mediated by multiple, interacting, inward and outward currents mediated by different ion charge carriers dependent on the action of specific membrane ion channels ( Figure 1). The initial depolarization phase takes the form of a rapid upstroke and is mainly driven by inward Na þ current (I Na ) through voltage-gated sodium channels (Na v 1.5). The succeeding plateau phase is dominated by inward Ca 2þ current (I Ca ). The resulting entry of extracellular Ca 2þ induces release of sarcoplasmic reticular Ca 2þ stores, thereby activating excitation-contraction coupling. Repolarization, ultimately returning the membrane to the resting potential, is principally driven by outward current through voltage-gated K þ channels (K v ). 1 K þ channel activity is thus a principal determinant of AP duration (APD) as it limits the depolarization duration and therefore both the time course of the Ca 2þ -mediated contraction and the refractory period. There are numerous and diverse K þ channels types, each with particular kinetic and voltage-dependent properties. These result in numerous and diverse current contributions, each with specific roles at different phases of repolarization. Together these determine the relatively prolonged but finely tuned repolarization time course and the repolarization reserve following recovery of the resting membrane potential. The repolarization reserve refers to the partly overlapping function of these currents, namely, I Kr , I Ks , and I K1 , that gives a limited level of redundancy to the system. 2 The kinetics of repolarization varies greatly with cardiac region and species. This reflects variations in the occurrence and density of the different K þ channel subtypes. All these characteristics suggest that explorations of K þ channels may yield a useful group of pharmacological targets for arrhythmic conditions.

Potassium Channels
K þ channels represent the most functionally diverse cardiac ion channel type. [3][4][5][6] Together, they tightly regulate cardiac repolarization, thus ensuring stable and consistent AP signaling. The different K þ channel types have overlapping functions, 2,7 resulting in some degree of functional redundancy, 2 which in turn contributes to repolarization reserve. Table 1 summarizes their encoding genes with their chromosomal locations and the structural properties of their pore-forming aand accessory b-subunits. The a-subunit of different K þ channel types all possess a conserved pore-forming region allowing K þ movement across the plasma membrane down an electrochemical gradient possessing a selective permeability to K þ attributable to a specific structural motif. They may also exhibit gating mechanisms responsive to membrane depolarization and ligand-binding sites whose occupancy could alter channel conformation. Finally, individual monomeric a-subunits may assemble into functional dimers or tetramers due to the presence of one or more subunit-assembly domains. 6,8-10 K þ channel a-subunits fall into 3 structural types based on subunit topology ( Figure 2). The first has 1 pore-forming region with 6 or 7 transmembrane regions (Figure 2A), the second has 1 pore-forming region and 2 transmembrane regions ( Figure 2B), and the third has 2 pore-forming and 4 transmembrane regions ( Figure 2C). 5,6,10 K þ channel b-subunits encompass many molecular groups, such as adenosine triphosphate (ATP)-binding cassette transport-related proteins (eg, sulfonylurea receptors) for inward rectifiers, cytoplasmic proteins (KChIP, KChAP, and K v b1-3), and single transmembrane spanning proteins (minK). 10 These b-subunits form complexes with the a-subunits and can modify the channel's functional properties. For example, K v b subunits can alter channel trafficking and the kinetics of current activation and inactivation when interacting with K v 1.5. 11 More specifically, K v b2.1 and K v b4.1 behave as chaperone proteins. 12 Furthermore, the N-terminus of K v b1.2 and K v b1.3 has an inactivation domain resembling the inactivation particle of the a-subunit, allowing it to modulate 2channel inactivation. 12-14 Figure 1. The ventricular action potential as a paradigm for cardiac electrophysiological activity. In the resting state, the voltage of the cell intracellular space is negative to the external environment. This reflects its higher K þ but lower Na þ and Ca 2þ concentrations and its lower membrane permeability to Na þ and Ca 2þ in comparison to K þ . K þ efflux from the cell is then controlled by the inward rectifier K þ channel (I K1 ). When excitation threshold is reached, a large Na þ influx (I Na ) into the cell through Na þ channels produces phase 0 depolarization. This is followed by activation of fast and slow transient outward K þ currents (I tof and I tos , respectively) mediating a K þ efflux driving a rapid phase 1 repolarization. There is also an activation of a depolarizing inward Ca 2þ current through L-type Ca 2þ channels (I Ca ), which initiates excitation contraction coupling. The reduced membrane K þ permeability due to I K1 rectification combined with I Ca maintains the action potential phase 2 plateau phase. Phase 3 repolarization is driven by K þ efflux through the rapid and slow delayed rectifier K þ channels (I Kr and I Ks , respectively), as well as I K1 . At the end of phase 3, the Na þ and Ca 2þ that have accumulated in the cells are removed by the Na þ , K þ pump, and the Na þ , Ca 2þ exchanger (NCX). The atrial action potential shows greater contributions to recovery from the ultrarapid delayed rectifier outward currents (I Kur ) and acetylcholine-activated inward rectifying K þ channel (I KACH ). Adapted with permission from Huang. 1 11q24 .

Cardiac Potassium Currents
Cardiac K þ channels vary in their permeability properties, membrane potential dependence, and their opening or closing activation and inactivation kinetics. The major currents are classified into the transient outward currents, delayed rectifier outward currents, and the inward rectifiers ( Figure 3). Advances in electrophysiological and molecular biology techniques have demonstrated additional currents that may fall outside this basic classification. Some brief notes on the major cardiac K þ currents, their role in the cardiac AP, and their functional importance follow.
Transient outward K þ (I to1 ) currents. When first described, the transient outward currents (I to ) were attributed to 2 distinct channels, one blocked by 4-aminopyridine (4-AP) and unaffected by extracellular Ca 2þ (I to1 ) and the other not blocked by 4-AP but sensitive to Ca 2þ (I to2 ). 6 I to drives the initial rapid repolarization phase of the AP. Regions with shorter APDs, such as the epicardium, right ventricle, and septum, have higher I to expression. It was later discovered that I to2 is a Clrather than a K þ current. 15 Further characterizations subdivided I to1 into fast (I tof ) and slow (I tos ) currents ( Figure 3). I tof predominates in the atria, whereas both I tof and I tos occur in the ventricles. 16 While I tos requires longer recovery times, its classification as "slow" is relative only to I tof . Thus, both I tof and I tos channels activate and inactivate rapidly in comparison to the corresponding processes in other K þ channels. 15 Due to differences in the biophysical properties of I tof and I tos , the existence of molecular heterogeneity between these 2 channels has been previously suggested. 15 Ultrarapid delayed rectifier currents (I Kur ). In addition to I to , the ultrarapid delayed rectified K þ current (I Kur ) plays a role in the initial rapid phase 1 AP repolarization. I Kur activates rapidly in under 10 milliseconds at voltages in the plateau range and deactivates slowly over the course of the AP. [17][18][19] I Kur is the predominant delayed rectifier current for the atria and thus results in the shorter APD seen in the atria compared to the ventricles. 10,16,17,19 Where I Kur is present, its channels are not evenly distributed over the myocyte surface but instead found at high densities in the intercalated disk. 6 This pattern of distribution is often disrupted after cardiac ischemic damage. 10 The selective presence of I Kur in the atria makes it an interesting target for atria selective therapy, whereby inhibition of I Kur would prolong the APD in the atria but not the ventricles. 4 Rapid delayed rectifier K þ currents (I Kr ). The voltage-gated rapid delayed rectifier outward K þ current (I Kr ) is critical to phase 3 repolarization. It shows a relatively rapid activation with depolarization. However, its inactivation rate is around 10 times faster than its activation rate due to voltage-dependent C-type inactivation. This renders it relatively nonconducting in phases 1 and 2 of the cardiac AP. [20][21][22][23] Thus, although termed a delayed rectifier current, it also shows an inward rectification property at positive potentials. 22,24 However, with the end of phases 1 and 2, as the membrane potential becomes negative to 0 mV, I Kr becomes activated once again, but the deactivation during this phase is much slower. This results in a large outward K þ efflux during phase 3 repolarization. 2,10 I Kr is found in both human atria and ventricles but is differentially expressed with higher levels in the left atrium and ventricular endocardium. 16 Slowly activating delayed rectifier K þ current (IKs). Cardiac repolarization is also influenced by a third, slowly activating delayed rectifier K þ current (I Ks ). I Ks slowly activates at potentials positive to À20 mV. Unlike I Kr , I Ks barely inactivates 25,26 and consequently accumulates over phase 2 repolarization, significantly influencing phase 3 repolarization. 2 This feature of I Ks is particularly important during atrial and ventricular APs of long duration. It is also involved in APD shortening during physiological increases in heart rate. An increase in heart rate thus reduces the time required for I Ks inactivation. In consequence, more I Ks accumulates, leading to a steeper drop in the A, The 6-transmembrane 1-pore-region voltage-dependent K þ channel (K v ) a-subunits mediating I Kur , I to , I Ks , I Kr , and I f . B, The 2-transmembrane 1-pore-region inward rectifying K þ channel (K ir ) a-subunits mediating I K1 , I KATP , and I KAch . C, The 4-transmembrane 2-pore-region K þ channel (K 2P ) mediating "leak" K þ currents. The arrows indicate the location of the pore-forming region(s). HCN indicates hyperpolarization-activated cyclic nucleotide-gated channel; I f , inward rectifier mixed Na þ and K þ "funny" current; I K1 , inward rectifier K þ current; I KACH , acetylcholine-activated inward rectifier K þ current; I KATP , ATP-sensitive K þ current; I Kr , rapid component of the delayed rectifier K þ current; I Ks , slow component of the delayed rectifier K þ current; I Kur , ultrarapid component of the delayed rectifier K þ current; I to , transient outward K þ current. Reprinted with permission from Giudicessi and Ackerman. Macmillan Publishers Ltd, copyright 2012. 3 repolarization rate. 27,28 Blocking I Ks results in an APD prolongation at increased heart rates. 21,28 Inhibition of I Ks will increase the vulnerable window for reactivation of voltagegated Ca 2þ channels, thereby increasing the risk of arrhythmic trigger events. 2 I Ks is found in all cardiac cell types, but its expression is significantly reduced in the mid-myocardial wall; this accounts for the long APD seen in this region. 16 Inward rectifier K þ current (I K1 ) . The inward rectifier K þ current (I K1 ) functions over a narrow membrane potential range. Its rectifying property results in a marked reduction in I K1 conductance at positive, depolarized, membrane potentials and an increase in I K1 at negative membrane potentials, with the effect of stabilizing the membrane resting potential close to the K þ equilibrium potential (E K ). 10 The channel mediating I K1 does not show voltage-dependent gating and does not possess a voltage sensor. Nevertheless, I K1 modulation associated with movement of Mg 2þ and polyamines results in an indirect sensitivity to voltage. [29][30][31][32] Between phase 0 and phase 2 of the AP, the membrane potential is more positive than À20 mV, and at this potential, there is no conductance of I K1 as the channel is inhibited by Mg 2þ and polyamines. The resulting marked inward rectification property limits the outward current at these positive potentials. This in turn minimizes the inward depolarizing current, which confers energetic efficiency for AP generation as it minimizes changes to ionic gradients that would need to be restored. 16 As the potential returns to more negative values (typically around À40 mV), the inhibition by Mg 2þ and polyamines is reversed. I K1 conductance then resumes and this contributes to phase 3 cardiac repolarization. 31 I K1 occurs in both atria and ventricles and is thereby involved in setting their resting membrane potentials. Channels conducting I K1 are expressed in greater density in the ventricles, making the ventricles less susceptible to pacemaker influence. 16 Acetylcholine-activated K þ current (I KACH ). The inwardly rectifying acetylcholine (ACh)-activated K þ current (I KACH ) is regulated by G proteins rather than voltage gating. Cardiac parasympathetic nerve endings release ACh, thereby activating M2 muscarinic receptors. This reduces the depolarizing effect of the pacemaker current (I f ), reducing firing rates of pacemaker cells and in turn reducing heart rate. 6 Acetylcholine also opens muscarinic-sensitive I KACH channels allowing the inward rectification of K þ . The inward rectifying current shortens the AP and hyperpolarizes the membrane potential. 16 Membrane hyperpolarization reduces the rate at which the sinoatrial and atrioventricular (AV) nodes drive pacemaker depolarization in addition to reducing AV conduction velocity. 6,33 I KACH is thought to be specific to the atria, 2 but there has been a suggestion that it may exist both in the atria and ventricle, 16 but with densities 6 times greater in the atria than the ventricles. 34 ATP-activated K þ current (I KATP ). The ATP-activated K þ current (I KATP ) occurs at both the sarcolemmal (sarc-K ATP ) and mitochondrial inner membrane (mito-K ATP ) of cardiomyocytes. The sarc-K ATP channels are highly expressed in cardiomyocytes and are composed of K ir 6.2 and SUR2A subunits. There may also be contributions from K ir 6.1 and SUR1. 35 In contrast, although the subunits of mito-K ATP channels have been difficult to identify due to the challenge of isolating pure In some instances, this may vary. With the inward rectifying K þ channels, the name refers to the unusual characteristic whereby net potassium flow is into the cell at potentials lower than the reversal potential where channel conductance is high. As the channel potential becomes more positive, channel conductance decreases. Net ion flow direction reverses at the reversal potential, meaning that net potassium flow is outward at potentials more positive than this. Therefore, at depolarized potentials, potassium loss from the cell is low as conductance through the channel is low.
mitochondrial membrane fractions, ROMK2 pore-forming subunits and SUR2 regulatory subunits have been suggested to contribute. 36,37 Both channels are controlled by ATP and are thus directly responsive to the cell's metabolic status, thereby influencing cell membrane potential. 6 I KATP is inhibited by physiological intracellular ATP levels, but this reverses with ATP depletion. Thus, under normal energetic circumstances, there is limited I KATP current. However, under both physiological and pathological conditions that reduce ATP, there is increased I KATP current that is essential for adaptation to stress. For example, compared to wild-type controls, mice lacking K ir 6.2-containing K ATP channels perform less well in acute treadmill exercise testing. 38 The increased I KATP has a cardioprotective role in ischemia by shortening the cardiac AP, thus limiting calcium influx into the cytosol. [39][40][41] Specifically, studies have suggested that mito-K ATP rather than sarc-K ATP channel opening has an energy-modulating property that confers cardioprotection in ischemic hearts. 42,43 In some situations, the I KATP -mediated APD shortening and corresponding heterogeneities in repolarization can create a substrate for cardiac reentry arrhythmia. In other situations, K ATP channel openers have been described to have antiarrhythmic effects, [44][45][46][47][48] and evidence suggests that activation and block of K ATP can be pro-or antiarrhythmic depending on the arrhythmogenic mechanism in different animal models. 49 For example, selective sarcolemma K ATP channel blockers, such as HMR 1883, confer antiarrhythmic effects in the short term, 50 although this could be metabolically disadvantageous in the long term due to the abolished adaptive response to stresses. Finally, it is important to note that the channel involved in the conductance of I KATP is also thought to be involved in the regulation of smooth muscle tone and insulin secretion in pancreatic b-cells. 6 Other K þ channel family: Ca 2þ -activated K þ current (I KCa ), 2-pore domain K þ current (I K2p ), and hyperpolarization-activated cyclic nucleotide-gated channels. Recently, several further currents have been characterized. The Ca 2þ -activated K þ current, also known as the small conductance Ca 2þ -activated K þ (SK) current (I KCa ), and the 2-pore domain K þ current (I K2p ) have attracted considerable physiological and pharmacological interest. I KCa was initially thought to not exist in the human heart. 51 However, subsequent studies demonstrated the presence of I KCa , with a higher density in the atria than the ventricle. Various subtypes of Ca 2þ -activated K þ channels exist in different tissues; the channel subtype conducting the cardiac I KCa is the SK channel. 51,52 In neuronal cell, SK channels that are involved in modulating the tonic firing frequency and activation of these channels cause membrane hyperpolarization, thus limiting neuronal AP firing frequency. 51 In contrast, cardiac SK channels and consequently I KCa are involved in late AP repolarization, controlling the resting membrane potential in human atria. 52 I KCa appears to not play physiologically significant roles in the ventricle. 52 I KCa is accordingly of particular pharmacological interest for atrial fibrillation (AF) therapy. Thus, I KCa occurs during late repolarization, when the atrial AP is susceptible to irregular or abnormal excitation such as that resulting from early afterdepolarizations (EADs). 51 I K2p contributes to the background current, the resting membrane potential, and cellular excitability. The channel involved in the conductance of this current has no voltage dependence, but its activity is modulated by lipids, particularly fatty acids, pH, drugs, particularly local and inhalation anesthetics, and membrane stretch. 53,54 These mediators act upon the channel via secondary messenger phosphorylation. 55 I K2p is a background current that persists through all phases of the cardiac AP. It thus stabilizes the membrane potential toward E m . I K2p may also prevent the occurrence of EADs, and it may be involved in fine-tuning of Na þ channel availability for phase 0 depolarization. 56 The current has been found to occur selectively in the atria and AV node, thereby making it a target for drug development. [57][58][59] Although not entirely new but only recently well characterized, the hyperpolarization-activated cyclic nucleotide-gated (HCN) channel is instrumental in conducting the inward funny current (I f ) in the heart. The channel is activated by the hyperpolarization of the membrane and is additionally stimulated by intracellular cyclic nucleotides. 60,61 The generation of I f is attributable to the inward permeability of both Na þ and K þ and occurs at threshold close to the resting membrane potential. 62 Although the HCN channel under physiological circumstances conducts both Na þ and K þ , the primary sequence of the HCN pore region suggests that it is primarily related to a selective potassium channel. 63 In certain pathological conditions such as AF and myocardial infarction, I f is increased unusually outside the pacemaker cells, leading to increased propensity to arrhythmia. Thus, targeting the I f in such pathological conditions has proven to be therapeutically advantageous. 64

Cardiac K þ Channel as Targets for Drug Development
Although there have been significant recent advances in the development and use of cardiological devices and procedures directed at arrhythmic conditions, antiarrhythmic drugs continue to be important whether by themselves or as adjunct therapy to such interventions. These include situations involving acute management of potentially fatal arrhythmic events, particularly where such procedures are contraindicated. Yet progress in antiarrhythmic drug development has been relatively limited. This likely reflects a lack of understanding of cardiac arrhythmic mechanisms. However, recent developments of our understanding of the role of the ion channels in normal AP generation have led to a specific interest into ion channels and their associated currents whose abnormal activity potentially leads to arrhythmia. This would encourage interest in the development of cardiac ion channel activator or blockers directed at modulating the cardiac AP or its refractory period. Introduction of drugs acting specifically on ion channels would optimize the efficacy of therapeutic actions on arrhythmogenic tendency, while minimizing problems arising from potential cardiac and extracardiac toxicity. K þ channels play a vital role in cardiac AP repolarization and thus naturally form potential targets for the development of ion channel-specific antiarrhythmic therapy, such as for AF. However, a limitation of this approach is that arrhythmic conditions, such as AF, are heterogeneous and the efficacy of targeting ion channels varies according to the cause and extent of the arrhythmia. This is complicated by the fact that in various physiological and pathological conditions, remodeling of K þ channel expression can occur, which can alter the AP and increase the risk of sudden cardiac death. 65 For example, AF is maintained and progressed partly due to electrical remodeling, mediating APD shortening. 66 Thus, in chronic AF, there is upregulation of I K1 , I Ks , and I K2P 3.1, which offsets the possible downregulation of I Kur and I to . [67][68][69] Nevertheless, the experimental evidence for the reduction in I Kur during remodeling is conflicting, as some reports suggest reduced I Kur density 70,71 and others suggest no change. 72,73 It has been suggested that receptor-activated I KACH (rI KACH ) mediates AF induced by vagal stimulation, while constitutive I KACH (c I KACH ) develops in the time course of AF remodeling. 67,74 In physiological cardiac hypertrophy, induced by chronic exercise, for example, there is an increase in I K density. 75 This contrasts with pathological cardiac hypertrophy caused by pressure overload where a reduced I K density is noted that was attributable to cellular hypertrophy rather than gene expression changes in I tof and I K1 . 76 In heart failure, AP prolongation is associated with downregulation of several genes, leading to reduced I tof , I Ks , I Kr , and I K1 . 65,77,78 Considering the changes in K þ channel expression in remodeling is clinically important as the sensitivity and efficacy of blocking these channels will change. Table 2 outlines selected drugs that have been experimentally proven to target different K þ channels, using either native cardiac myocytes or human cell line expression systems. Some of these drugs presently in clinical use have been primarily developed for other ion channels such as the Na þ or Ca 2þ cardiac ion channel but have corresponding effects on K þ channels. Several drugs have been proposed to be selective to specific K þ channels, such as A935142, XEN-D0103, and XEN-D0101. However, despite promising experimental findings, many of these drugs have not progressed to clinical use. This may be attributable to limitations associated with experimental studies. Expression systems can often produce offtarget effects or nonspecific interactions which may mask the true effect of these drugs. Additionally, expression systems may run the risk of either overexpressing or underexpressing the channel of interest. On the other hand, native cardiac myocytes, while more physiologically representative, may not provide the right platform for the study of specific targets. Additionally, acquisition of viable native cardiac myocytes from a minimally heterogeneous population remains a challenge, and it is widely accepted that channel functions can differ by gender and age. Consequently, while experimental studies may suggest potentially promising options to selectively target K þ channels, the translational capacity of such studies remains limited.
Furthermore, these activators and blockers often target more than 1 K þ channel species and thus are not entirely specific. 10,122 However, a large proportion of these drugs also typically target I Kr (known to be present in all cardiac regions) and as such do not constitute ideal candidates for targeted therapy. Nevertheless, mechanisms of cardiac arrhythmia are likely to be region dependent. Drugs that may be antiarrhythmic in some cardiac regions may potentially be pro-arrhythmic in others. Thus, the presence of atrial-specific K þ channels has provided focus on developing drugs that could specifically increase refractory periods, thus preventing atrial reentry arrhythmia, which is the most common mechanism for AF. 123 Of ion channels specific to the atrium that might offer specific therapeutic targets, the channel conducting I Kur tends to prolong repolarization and effective refractory period (ERP) without altering QT intervals. 124,125 The experimental drugs AVE0118 and XEN-D101 are thought to be I Kur selective blockers with both prolonging APD in atrial tissue from patients with permanent AF in common with the known I Kur blocker 4-AP.  However, a subsequent "first-in-human" study using the highly selective I Kur blocker MK-0448 (N-f6-[(1S)-1-(4-fluorophenyl)-2,2-di (pyridine-3-1) ethyl] pyridine2ylg methane sulfonamide) did not reveal any increase in atrial ERP. This led to the conclusion that selective blocking of I Kur may have limited clinical value. 130 I KACH channels are also atrium specific or at least predominantly occur in the atria and have minimal physiological function in the ventricle. 129 Opening of the I KACH channel will lead to shortening of atrial APD and thus increase the likelihood of AF. Therefore, blocking the opening of I KACH channels will prevent such shortening of APD with minimal effect on ventricular APD, in turn reducing the chances of AF. Several drugs block I KACH , but have limited specificity. Nevertheless, selective blocking of I KACH has been experimentally achieved using the compound NTC-801. The compound was found to have selective antifibrillatory properties, achieved by prolonging the atrial ERP. 131 Another potential atrial-specific therapeutic target of interest is the I KCa current conducted by SK channels. The selective presence of this current in the atria has recently led to several investigative drugs beings explored. NS8593 is a selective SK channel inhibitor demonstrating significant atrial antiarrhythmic effects in canine and equine experimental models. Experiments using human atrial cardiac myocytes from patients with normal sinus rhythm demonstrated reduction in K þ currents and prolongation in APD. No such changes were observed in intraventricular myocytes. 52

Conclusion
There is currently an incomplete understanding of the cellular physiological role of the various cardiac potassium currents and their interacting effects and how dysregulation of their function and expression can provide arrhythmogenic mechanisms. It is thought that the site-specific distribution of some K þ channels could allow targeted therapy to be more spatially selective. However, complex electrical remodeling events that occur in disease states may change channel expression levels to the extent that the selectivity of the drug is hindered, making even this potential therapeutic strategy challenging. Although targeting ion channels responsible for discrete parts of the cardiac AP to modulate the system towards a more physiological state has therapeutic appeal, there are inherent difficulties in developing successful drugs. This is because the ion channels targeted are functionally complex and are interdependent, thus adding a dynamic situation in which function and expression are altered depending on the cell environment. Furthermore, pathophysiological processes of arrhythmic disease may involve functional alterations in 1 or more ion channels. Such single or multiple ion channel functional abnormalities may therefore warrant corresponding use of a single or multichannel activator/blocker approach. However, this approach will only be possible if we are able to identify the specific pathophysiological process affecting individual patients (ie, is this arrhythmic disease related to a single or multichannel abnormality). Thus, although we may be able to develop single or multichannel activators/blockers, actual clinical use will be dependent on a detailed understanding of the exact arrhythmogenic mechanisms affecting individual patients, which thus far is limited. Presently, decisions to use single or multichannel activators/ blockers are largely dependent on resolution of clinical signs or the actual arrhythmia rather than a therapeutic approach targeting ion channel functional abnormality. Furthermore, the availability of truly specific ion channel activators/blockers is limited as these agents tend to have off-target actions with corresponding side effects, and this limits the clinical use of selective agents. Focusing on understanding the system at a cellular physiological level through further experimental and computational modeling is needed to enable development of novel insights at a pharmacological level.

Author Contributions
Kamalan Jeevaratnam contributed to conception and design, contributed to interpretation, drafted the manuscript, and critically revised the manuscript. Karan Raj Chadda contributed to interpretation, drafted the manuscript, and critically revised the manuscript. Christopher L-H. Huang contributed to design, contributed to interpretation, drafted the manuscript, and critically revised the manuscript. A. John Camm contributed to conception, contributed to interpretation, and critically revised manuscript. All authors gave final approval and agree to be accountable for all aspects of work ensuring integrity and accuracy

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: K.